Individual Optical Multi-Gas Sensors as Next Generation Second-Order Unobtrusive and Continuous Operation Analytical Instruments

Individual Optical Multi-Gas Sensors as Next Generation Second-Order Unobtrusive and Continuous Operation Analytical Instruments | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Individual Optical Multi-Gas Sensors as Next Generation Second-Order Unobtrusive and Continuous Operation Analytical Instruments Radislav A. Potyrailo, Shiyao Shan, Baokai Cheng This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6234291/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 09 Jul, 2025 Read the published version in Microchimica Acta → Version 1 posted 10 You are reading this latest preprint version Abstract Simple, single-output (zero-order) gas sensors have established applications for industrial and residential safety. However, in detection of low levels of analyte gases in diverse chemical backgrounds, zero-order gas sensors often have poor performance, cancelling out their small size and low-cost advantages over traditional analytical instruments. Thus, gas sensors with one and two independent variables (first- and second-order sensors) are under development to meet contemporary gas monitoring needs for environmental, industrial, medical, and homeland security applications. Here, we show three designs of photonic nanostructured sensors as second-order analytical instruments. In all three designs, the first independent variable in sensor response is optical wavelength across the visible spectral range. The second independent variable is either (i) illumination angle, (ii) light polarization, or (iii) operation temperature of the sensor. These illustrative designs of second-order gas sensors bring mathematical understanding to boost the multi-gas resolution achieved by individual material-based sensors through the transducer, material design, appropriate excitation conditions, and data analytics. Results presented in this report should inspire new fundamental research and technological innovations in the areas of gas sensors, optics, materials science, nanofabrication, and data analytics. The strategic approach of expanding the dimensionality of sensor response described in this study will enable next generation sensors to perform on par with traditional orthogonal-output analytical instruments, but with a continuous real-time data stream and unobtrusive, energy-efficient designs. photonic nanostructure multi-gas sensing second-order individual sensors traditional analytical instruments Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 Figure 12 Figure 13 Figure 14 Introduction The knowledge of “what is in the air” drives mitigation of air pollutants—starting from the industrial revolution in 1800s [1], to detection of chemical agents by soldier “gas scouts” during the World War I [2], detection of leaks of cooking gases in residential households in 1960s [3], detection of outdoor pollutants in 1970s [4], and detection of per- and polyfluoroalkyl (PFAS) air pollutants today [5]. To analyze gaseous species (gases or volatiles), there are two broad categories of gas detectors: traditional analytical instruments and gas sensors. Traditional analytical instruments are designed with one, two, or more independent response variables as first-, second-, and higher-order instruments to provide accurate information about one or more analytes in chemically complex environments [6]. Examples of independent response variables are wavelength in spectroscopy, retention time in gas chromatography (GC), and mass-to-charge ratio in mass spectrometry (MS). These instruments require many materials and high energy to manufacture these instruments, and also demand costly maintenance, consumables, and power demands during their end-use [7, 8]. Gas sensors are designed as small-size, single-output (e.g., light intensity, resistance, etc.) devices, also known as zero-order devices [6, 9-12], to reversibly respond to a concentration change of an expected gas in a simple chemical environment. Gas sensors are designed either to monitor intrinsic properties of gases or to utilize a gas-sensing material [9]. Gas sensors have poor detection accuracy in complex chemical environments and experience sensor drift [9, 10, 12]. Contemporary designs of material-based gas sensors utilize signal-transduction principles based on electrical, acoustic, and optical energy [9, 13]. These sensors thrive to progress from proof-of-concept demonstrations to commercial implementations [14-18]. Optical material-based gas sensors consist of two broad categories: one that utilizes chemical effects of gas interactions with organic reagents and one that focused on the physical effects of gas interactions with nanostructured materials [13, 19-24]. Compared to electrical and acoustic types of sensors, optical material-based gas sensors have a broader range of design parameters, e.g., five properties of light such as intensity, wavelength, phase, polarization, and temporal property [25]. To expand gas sensor opportunities, the interdisciplinary research community is learning from mathematics behind zero-order sensor designs that originated from the 1960s [6, 9-12] how to overcome their limitations. Two directions are explored, such as arrays of zero-order sensors [26-28] and individual first-order (a.k.a. multivariable) sensors with diverse response variables (e.g., wavelength, frequency, temperature, etc.) [13, 29-38]. Arrays of zero-order sensors differentiate multiple volatiles [26, 39, 40] and have up to five dimensions (5-D) of sensor-array dispersion [41], but an independent drift of individual sensors in an array remains a persistent problem, preventing arrays from being broadly implemented by end-users [42-45]. First-order gas sensors showed self-correction of baseline drift [46-48], displayed 5-D sensor dispersion [49], and differentiated multiple volatiles with one sensor [13, 35, 50]. In this report, we show opportunities for photonic nanostructured sensors as second-order analytical instruments. In prior work, we noticed extraordinary multi-vapor differentiation using spectral responses of scales of iridescent Morpho butterflies [29]. Structurally colored Morpho butterfly scales are also known as natural metamaterials [51]. From experimental and theoretical studies of the multi-gas responses of Morpho scales and bio-inspired fabricated sensors, we developed design rules with physical (nanostructure geometry) and chemical (spatially controlled chemical functionalization) design criteria for first-order gas sensors for operation at room and high temperatures [29-31, 36, 37, 48, 52]. Here, we describe results of three designs of optical photonic nanostructured gas sensors that operate as second-order sensors. In all three designs, the first independent variable in sensor response is optical wavelength across the visible spectral range and the second independent variable is either an illumination angle, light polarization, or operation temperature of the sensor. These illustrative designs of second-order gas sensors show the opportunities that are unveiled by bringing mathematical understanding to boost the multi-gas resolution by individual material-based sensors. From zero- to first- and second-order individual multi-gas sensors At present, the interdisciplinary sensing communities rarely describe sensors with one or two independent response variables based on mathematical descriptions, such as first- and second-order sensors [10, 11]. Instead, such sensors are known as multivariable, intelligent, multiparameter, virtual multisensor systems, or virtual sensor arrays [13]. Mathematical descriptions of such sensor designs as first- and second-order sensors provide more insights on further improvements of gas sensors. Zero-order gas sensor designs have conflicting requirements for reversibility and selectivity because the interaction energy between the sensing material and the analyte should be low to provide a reversible sensor response, but this low interaction energy leads to poor selectivity [9]. Poor selectivity often cancels out size and cost advantages of sensors over traditional analytical instruments [8, 53-55]. The mathematics behind traditional analytical instruments inspire next-generation gas sensors. First- and second-order traditional analytical instruments are designed with one or two independent response variables. Such designs provide solutions to complex measurement problems in diverse applications. These instrument designs deliver huge societal benefits, recognized with three Nobel Prizes [56]. As shown in Fig. 1a , using a zero-order detector (with only a single output) allows detection of only a single analyte and requires a total selectivity for this analyte. Mathematically, a zero-order detector (sensor) cannot differentiate between gases. In detection of small analyte concentrations, the presence of interferences has more pronounced effects, degrading accuracy. First-order detectors have one independent variable in their response, as shown in Fig. 1b (e.g., wavelength in spectroscopic, retention time in GC, or mass-to-charge ratio in MS instruments). To differentiate between volatiles, it is critical to have different response patterns between different volatiles as a function of the chosen independent variable. For example, volatile 1 and volatile 2 have response patterns that are close to each other, while volatile 3 has a very different response pattern (see Fig. 1b ). Using a first-order detector, it is possible to reject interferences from the instrument response to volatiles. The response patterns from all expected interferences should be included in calibration (a.k.a. training). When a detector has two independent response variables (e.g., GCxGC, GCxMS), a second-order detector improves differentiation between volatiles in a two-dimensional (2-D) differentiation map (see Fig. 1c ) and does not require calibration for unknown interferences for analysis in complex samples [57, 58]. Such capability is known as the second-order advantage [6, 57, 58]. Second- and higher-order analytical instruments are must-have solutions for analysis of trace levels of analytes in cluttered environments (e.g., exhaled breath, battlefield air) [59-61], with multi-dimensional data analyzed by multivariate statistical data analysis techniques [62-65]. Inspired by mathematics behind traditional analytical instruments, next-generation gas sensors, based on first- and second-order detection principles, are being built by implementing four design aspects (four pillars), as shown in Fig. 2 . First , a physical transducer is designed with proper excitation conditions to operate with independent variables. Second , a sensing material is implemented with different responses to multiple volatiles under variable excitation conditions. Third , excitation type and range are designed to induce independent responses in the transducer / sensing material system. Fourth, data analytics provide not only multi-gas differentiation, quantification, and rejection against known (first-order sensors) and unknown (second-order sensors) interferences, but also self-correction against drift. These next-generation multi-gas sensors (see Fig. 3 ) are under development as Moonshot second-order detectors to compete against the exquisite performance of traditional analytical instruments. These multi-gas sensors are breaking the trade-off between size, weight, power, and cost (SWaP-C) versus performance of analytical instruments, facilitating the widespread application of these sensors to serve emerging societal needs. Experimental Nanostructured optical sensing elements In this work, two types of nanostructured optical sensing elements were utilized. A sensing element of type 1 was in the form of dried scales of Morpho butterflies, obtained from Butterfly Utopia (Brooklyn, NY, USA). A ~5 mm x 5 mm portion of a dried intact Morpho butterfly wing was utilized for vapor-response tests. A tree-like photonic nanostructure of iridescent tropical Morpho butterflies has an open-air architecture with ridges and lamella, allowing gas interactions with its regions [29, 30]. Lamella act as multilayer interferometric nano-reflectors and ridges act as a diffraction grating [66]. A sensing element of type 2 was in the form of a ~2 mm x 2 mm piece of wafer with nanofabricated bio-inspired nanostructures for operation at high temperatures. We developed design rules and fabricated such sensors for operation at room and high temperatures [31, 36, 37, 52]. For high temperature applications at ~300-380 o C, nanostructures were fabricated using conventional photolithography and chemical etching [36, 37, 52]. Nanostructure materials were inorganic and included catalytic metal nanoparticles and metal-oxide capping layers to promote reactions with gases of interest at high operation temperatures [37, 67]. Excitation and vapor testing of nanostructured optical sensing elements Sensing elements were illuminated with a white light source (a halogen lamp) in the reflection mode, with spectroscopic detection of the reflected light using a spectrograph [29-31, 37]. Measurements were done as the differential reflectance responses ΔR(λ) = R(λ)/R 0 (λ), where R(λ) is the spectrum of the sensing element upon exposure to the gaseous species (volatiles) of interest and R 0 (λ) is the spectrum of the sensing element upon exposure to a carrier gas (a blank) at a given condition (e.g., illumination angle, light polarization, operation temperature). Thus, the common features in the two spectra before and during gas exposure cancel out and the ΔR(λ) spectrum accentuates the subtle differences due to the gas response [21, 29-31, 37]. Two linear polarizations of white light were achieved using standard components. Model volatiles (vapors) such as water, ethanol, 1-methoxy-2-propanol, methyl ethyl ketone, acetonitrile, toluene, chloroform, tetrahydrofuran, dimethylformamide, and acetone as well as hydrogen gas were selected to demonstrate the capabilities of the second-order gas sensors. Different concentrations of volatiles were produced using built-in-house computer-controlled vapor-generation and mixing systems [31, 37]. Concentrations of majority of vapors were defined as P/P 0 , where P is the vapor partial pressure and P 0 is the saturated vapor pressure [29, 31]. Concentrations of water vapor were defined as percent relative humidity (%RH). Concentrations of H 2 gas were in percent by volume. Iridescent nanostructures of Morpho butterfly scales and reflectance spectra of a Morpho butterfly in dry air under parallel and perpendicular polarization states of white light are depicted in Fig. 4 . Examples of bio-inspired nanostructures and a reflectance spectrum are shown in Fig. 5. Multivariate data analytics Sensor data was analyzed using KaleidaGraph (Synergy Software, Reading, PA, USA) and PLS_Toolbox Software (Eigenvector Research, Wenatchee, WA, USA), operated with MATLAB (Mathworks, Natick, MA, USA). A multivariate data analysis technique such as Principal Components Analysis (PCA) was used for differentiation of volatiles as needed. PCA remains the most popular unsupervised tool for dimensionality reduction in diverse data sets ranging from traditional analytical instruments [62-65] to sensor arrays [27, 28, 68] and individual multivariable sensors [13, 50, 69, 70]. For statistics, we used several data points per concentration of each vapor and the blank or an added 5% noise [71] into the data for more conservative estimations. Results and Discussion Spectral and angular differentiation of volatiles A sensing element utilized in these experiments was a portion of a Morpho butterfly wing (Type 1 sensing element, see section Nanostructured Optical Sensing Elements ). The first independent variable in its response to vapors was illumination wavelength across the visible spectral range. The second independent variable was illumination angle of the sensing element. In earlier experiments with natural Morpho nanostructures to differentiate three vapors, diversity of spectral responses to vapors was noted to depend on the illumination angle [29, 37]. Here, the effect of illumination angle was explored with ten diverse model vapors such as water, ethanol, 1-methoxy-2-propanol, methyl ethyl ketone, acetonitrile, toluene, chloroform, tetrahydrofuran, dimethylformamide, and acetone at one level of ~0.5 P/P 0 . Vapor exposures were done at three illumination angles of 20 o , 40 o , and 60 o normal to the surface of the sensing element. Fig. 6 depicts spectral responses of this sensing element to ten vapors as a function of two independent variables such as the illumination wavelength across the portion of the visible spectral range where iridescence occurs (360–650 nm) and the illumination angle of a nanostructured sensing material. The visual diversity of differential reflectance spectra ΔR(λ) at different illumination angles can be summarized as three categories. First , these spectra had three general regions: 360–430 nm with an inflection point at ~400 nm, 430–490 nm with an inflection point at ~460–480 nm, and 490–650 nm with an inflection point at ~520 nm. Second , the relative intensity of the inflection point at 460–480 nm was increasing with illumination angle, while the relative intensity of the inflection point at ~520 nm was decreasing with illumination angle. Third , the overall spectral intensity was decreasing with illumination angle. For a quantitative interpretation, Fig. 7 illustrates PCA scores plots of the first several principal components (PCs) with the data analyzed with all three illumination angles. For conservative estimations of the number of useful PCs, we had five replicates of each spectrum with added 5% random noise [71]. Fig. 7a depicts a scores plot of PC1 vs PC2 without noticeable contributions of the added 5% noise. We have found that our data set with all three illumination angles had seven PCs that differentiated ten vapors, as shown in Fig. 7b-d, depicting scores plots of PC5 vs PC6 vs PC7 at their different projections. The first three PCs had contributions of PC1 = 54.61%, PC2 = 22.66%, and PC3 = 13.85%, capturing only 91.12% of the total spectral variance. Thus, the remaining variance of 8.88 % was available for the enhanced multi-vapor differentiation. To evaluate this remaining variance, the signal-to-noise ratio (SNR) of each PC was analyzed [13]. Fig. 8 depicts plots of estimated SNR for each PC when spectral measurements were performed at three individual angles and their combination. The number of PCs needed to describe the data with SNR > 3 is highlighted as green data points. The PCA models built based on individual illumination angles had four PCs with SNR > 3 for 20 o angle ( Fig. 8a ) and five PCs with SNR > 3 for 40 o and 60 o angles ( Fig. 8b, c ). The PCA model built based on the combination of three illumination angles had seven PCs with SNR > 3 ( Fig. 8d ). Thus, illumination of the Morpho nanostructure at multiple angles combined with the spectral analysis across the visible wavelengths achieved an extraordinary seven dimensions in the response dispersion of an individual sensor. A 5-D response dispersion was reported with a white light illumination of an iridescent Hypochrysops Polycletus butterfly at a single angle [49]. At this moment, we did not come across reports on response dispersion of gas sensor arrays higher than 5-D or individual multivariable gas sensors higher than 5-D [50]. A possibility for a 10-D dispersion using an individual bio-inspired optical sensor was demonstrated theoretically [13]. Spectral and polarization differentiation of volatiles Utilization of polarization readout in material-based optical gas sensors to improve their multi-gas differentiation is rare [25]. This approach requires system understanding of interrelations between sensor system components when improving sensor performance for multi-gas differentiation for emerging applications. The sensing element utilized in this section was a portion of a wing of a Morpho butterfly (Type 1 sensing element, see section Nanostructured Optical Sensing Elements ). The first independent variable in its response to vapors was the illumination wavelength across the visible spectral range. The second independent variable was two orthogonal linear polarization states of white light that were used to illuminate the sensing element. The two orthogonal linear polarization states were parallel and perpendicular polarizations. The parallel polarization is also known as P- (from the German parallel) and transverse-magnetic (TM) polarization. The perpendicular polarization is also known as S- (from the German senkrecht, perpendicular) and transverse-electric (TE) polarization [72]. These experiments were focused on the ability of this second-order sensor to differentiate between two model volatiles, acetonitrile and ethanol. Using only two model volatiles for initial demonstrations is common, as shown in other reports [73-80]. Polarization- and wavelength-dependent reflectivity changes of this sensing element were measured upon exposure to six concentrations of these two model vapors (0.1, 0.2, 0.3, 0.4, 0.6, and 0.8 P/P 0 ). At different wavelengths, diverse response patterns at two polarizations for two vapors at their six concentrations were observed, as depicted in Fig. 9 . This diversity can be summarized as three categories. First , the relative intensities of signal changes of the parallel and perpendicular polarizations were wavelength- and vapor-dependent. Second, diversity of vapor responses between parallel and perpendicular polarizations were substantially reduced at longer wavelengths (540 nm and 590 nm). Third, diverse dynamics of the responses of the parallel and perpendicular polarizations were also wavelength- and vapor-dependent. Previously with unpolarized light, diverse vapor-response dynamics of Morpho scales was observed [29, 30] due to physical adsorption and capillary condensation of different vapors in the nano-size domains of the Morpho scales. At low vapor concentrations, molecules adsorbed onto the nanostructure, while at higher concentrations, a capillary condensation of vapor molecules occurred starting at the smallest radii and spreading throughout the lamellae. Thus, abrupt dynamic changes in reflectance occurred mostly at shorter wavelengths, because the smaller spatial periodicity or size of the photonic substructure was associated with the shorter response wavelengths [29]. We purposefully did not include analysis of dynamic spectral vapor-response features that are clearly visible at the short illumination wavelengths with polarized light (see Fig. 9 at 410, 440, 480, and 520 nm). Numerous excellent reports demonstrate the ability of individual electrical, acoustic, and optical sensing elements to benefit from diverse dynamics of different vapors for their differentiation [80-84]. However, as noted earlier [11], such dynamic measurements that mimic natural olfaction [85, 86] require a well-controlled sample introduction, borrowing sample introduction principles with needed pumps, valves, and other mechanical components from gas chromatography and other gas introduction technologies [87, 88]. Such conventional sample-handlining complexity negates the intended simplicity of sensor system designs and applications. Fig. 10 depicts contour 2-D plots for two vapors with the first independent variable being wavelength and the second independent variable being polarization states of white light. For this illustrative visualization, the highest tested concentrations of vapors were utilized and contour plots were presented with eight contours of the DR signal at each polarization. At different wavelengths and polarizations, diversity of response patterns for two vapors were three-fold. First , the number and shapes of the contours were related to the vapor type and light polarization. At the parallel polarization, both vapors had only two contours but of different shape. At the perpendicular polarization, acetonitrile had four contours, while ethanol had three contours. Second, at shorter wavelengths of 410–440 nm, the diversity between vapors was more pronounced in the number and shapes of the contours. At the parallel polarization, both vapors had one contour but of different shape. At the perpendicular polarization, acetonitrile had two contours, while ethanol had one contour. Third, at longer wavelengths of 480–590 nm, the diversity between vapors was less pronounced—it was only a single but different shape contour for both vapors and their two polarizations. We also performed PCA analysis of responses of the sensor to all tested concentrations of vapors, using both polarizations in the PCA model. Fig. 11 depicts PCA scores plots of the first two and three PCs with the data analyzed with both polarizations. The first three PCs had contributions of PC1 = 88.40%, PC2 = 9.15%, and PC3 = 2.00%, capturing 99.55% of the total variance in the spectra. Some non-monotonic or abrupt changes in the data points in the scores plots were likely the effects of adsorption and condensation of different vapors into the nanostructured features of the sensing element, with an increase in concentrations of vapors ( Fig. 11a ). The 3-D plot shown in Fig. 11b attempts to show that in addition to two vapors in the data set, there were also two different polarization states that affected PCA scores plot by adding PC3. Spectral and temperature differentiation of volatiles Sensor operation at different temperatures is a known principle to increase the order of sensor response because chemical interactions of diverse volatiles with a sensing material can be substantially different over appropriately chosen temperature ranges [89]. Previously, individual first-order sensors based on variable temperature of operation of sensing materials were demonstrated on different transducers, such as semiconducting metal oxides on resistors [46, 90, 91], non-oxide semiconductors on field-effect transistors [92, 93], and polymers on acoustic resonant sensors [94-96]. Second-order sensors were also demonstrated with a temperature modulation of semiconducting metal oxide sensing materials, coupled with impedance readout [97, 98]. Importantly, multi-temperature operation is now a common practice in numerous types of commercially available chemiresistor sensors with semiconducting metal oxides [50]. The sensing element utilized in this experiment was a fabricated bio-inspired nanostructured sensing element (Type 2 sensing element, see section Nanostructured Optical Sensing Elements ). The first independent variable in its response to vapors was spectral response across the visible spectral range and its second independent variable was its operation temperature. As an illustrative example, we show results of initial experiments that were focused on the ability of this second-order sensor to differentiate between H 2 gas and H 2 O vapor in its future intended application for H 2 gas monitoring in the fuel stream of solid oxide fuel cells [52]. In these initial tests, the sensor was exposed to H 2 gas and H 2 O vapor at two temperatures, ~20°C (room temperature T1) and ~380 °C (T2). H 2 gas concentrations were 0.625, 1.25, 1.875, and 2.5%; H 2 O vapor concentrations were 4, 8, 12, and 16 % relative humidity (RH), relevant for gas monitoring in solid oxide fuel cell applications [52]. Fig. 12 illustrates dynamic responses of this second-order sensor to two volatiles at selected wavelengths and temperatures. At temperature T1 , the sensor responded only to H 2 O at certain wavelengths and did not respond to H 2 or H 2 O at other wavelengths. Response to H 2 O was always in one direction and linear with H 2 O vapor concentration. Response only to H 2 O and not to H 2 was due to the physical sorption of H 2 O vapor molecules to the nanostructure. The wavelength-dependence of this response was due to diversity of interactions of H 2 O vapor with different spatial regions of the nanostructure, related to corresponding spectral response differences [37, 52]. At temperature T2 , the sensor had a wavelength-dependent response to H 2 and H 2 O; it was increasing or decreasing as a function of H 2 and H 2 O concentrations. At some wavelengths, the response was barely noticeable. While at many wavelengths the sensor responded only to H 2 , at certain wavelengths, the small response to H 2 O was also noticed as an increase or decrease in sensor signal. Sorption of H 2 gas and H 2 O vapor molecules to the nanostructure occurred on different spatial regions of the nanostructure because of its design with spatially-distributed catalytic regions [37]. Such spatially diverse interactions with H 2 and H 2 O resulted in corresponding differences of spectral responses. Fig. 13 depicts contour 2-D plots of responses of the sensing element to H 2 gas and H 2 O vapor, with the first independent variable being wavelengths and the second independent variable being sensor temperature. For this illustrative visualization, the highest tested concentrations of volatiles were utilized, several wavelengths were selected (480, 560, 660, 700, and 965 nm) for the adequate two-volatile differentiation, and contour plots are presented with eight contours of the raw signal strength at each temperature. At different selected wavelengths and temperatures, diversity of response patterns for two volatiles was three-fold. First , the number and shapes of the contours was related to the vapor type and sensor temperature. Second, at T1, H 2 had no response (no contours), while H 2 O vapor had responses (contours) at all selected wavelengths. Third , at T2, H 2 had two contours of its response, while H 2 O vapor had no response. To visualize further responses of the sensor to all tested concentrations of H 2 gas and H 2 O vapor, we performed PCA using both temperatures and wavelengths selected in Fig. 12 . The PCA scores plot of the first two PCs is depicted in Fig. 14 . The first two PCs had contributions of PC1 = 73.40% and PC2 = 19.61%, capturing 93.01% of the total variance in the spectra. The third PC did not contribute to the two-volatile differentiation. Conclusions Zero-order gas sensors have their established applications as alarms for industrial and residential safety when “pollution levels are high and when the compound of interest swamps others,” as stated in Nature Comments [54]. In detection of low analyte levels in diverse chemical backgrounds, zero-order gas sensors often have poor performance that cancels out their small size and low-cost advantages over traditional analytical instruments [8, 53–55]. The effects of interferences make the data from zero-order sensors “essentially meaningless,” as stated in another Nature Comments [53]. As recently pointed out by the U.S. Environmental Protection Agency, zero-order sensors “have inherent limitations that are critical to understand before collecting and interpreting the data; many of these limitations require technological advances” [55]. Thus, first- and second-order gas sensors are under development by numerous academic and industrial research teams to provide advances to meet contemporary requirements of air quality monitoring. It was shown [18, 32, 97, 98] that advancements to bring sensing systems to operate as first- and second-order gas sensors can be achieved without making fundamental changes to the existing hardware, ensuring compatibility with existing manufacturing practices and tools. Results presented in this report should inspire new fundamental research and technological innovations in the areas of gas sensors, optics, materials science, nanofabrication, and data analytics. This strategic approach of expanding the dimensionality of sensor response has a huge societal impact by bringing the next generation sensors to perform on par with traditional orthogonal-output analytical instruments, but with a continuous real-time data stream and energy-efficient unobtrusive designs. Second-order sensors will break the trade-off between size, weight, power, and cost versus instrument performance, facilitating pervasive applications of next-generation gas sensors. Declarations Acknowledgements R. A. Potyrailo dedicates this report to Otto Wolfbeis for his vision, influence, and inspiration for our interdisciplinary scientific community. R. A. Potyrailo is grateful for Otto Wolfbeis for fostering collaborations across various fields of optical sensors and for inspiring scientists worldwide through his innovative perspectives. Special thanks to librarians in the former USSR who provided photocopies of original papers by Otto Wolfbeis to R. A. Potyrailo. Special thanks to J. Sgambati for her thoughtful feedback and suggestions on the manuscript, which enhanced the clarity and readability of the described complex mathematical concepts for the interdisciplinary scientific community. Multi-gas sensors described here were encouraged and supported by V. Greanya at the Defense Advanced Research Projects Agency (DARPA) and S. Markovich and V. K. Venkataraman at U.S. Department of Energy (DOE). Sensing materials described in this report were inspired, characterized, and fabricated by creative scientists and engineers who coauthored original referenced contributions. Funding Declaration: Funding was from GE, DARPA Contract W911NF-10-C-0069 and DOE Contracts DE-FE0027918 and DE-FE0031653. The conclusions in this report should not be interpreted to represent any determination or policy of DARPA and DOE and do not necessarily reflect the position or the policy of the US Government. Mentions of commercial products do not imply endorsement. Author contributions Radislav A. Potyrailo: conceptualization, methodology, data collection, writing—review and editing; Shiyao Shan: visualization, review, and editing; Baokai Cheng: data curation and investigation. All authors have read and agreed to the published version of the manuscript. Data availability The data that support the plots within this paper and other findings of this study are available from the corresponding author upon reasonable request. Declarations Conflict of interest The authors declare no competing interests. References Des Voeux HA (1908) Discussion On Smoke Abatement. Brit Med J. 2(2487):549-54. Smart JK (2000) History of Chemical and Biological Detectors, Alarms, and Warning Systems. U S Army Soldier and Biological Chemical Command.https://www.hsdl.org/?view&did=1670. Ihokura K, Watson J. Stannic Oxide Gas Sensor: Principles and Applications. Boca Raton, FL: CRC Press; 1994. The Plain English Guide to the Clean Air Act. (2024) US EPA.https://www.epa.gov/clean-air-act-overview/plain-english-guide-clean-air-act. Li W-L, Kannan K (2024) Determination of legacy and emerging per-and polyfluoroalkyl substances (PFAS) in indoor and outdoor air. ACS ES&T Air. 1(9):1147-55. Booksh KS, Kowalski BR (1994) Theory of analytical chemistry. Anal Chem. 66:782A-91A. Grinberg N, Rodriguez S, editors. Ewing's Analytical Instrumentation Handbook. 4th ed.: CRC Press; 2019. Robinson JW, Frame EMS, Frame II GM. Instrumental analytical chemistry: an introduction. CRC Press; 2021. Janata J. Principles of Chemical Sensors. 2 ed. New York, NY: Springer; 2009. Taylor RF, Schultz JS, editors. Handbook of Chemical and Biological Sensors. Bristol, UK: IOP Publishing; 1996. Hierlemann A, Gutierrez-Osuna R (2008) Higher-Order Chemical Sensing. Chem Rev. 108:563-613. Johnson KJ, Knapp AC (2020) Design Theory for Chemical Detection. Naval Research Laboratory, Washington, DC. https://apps.dtic.mil/sti/citations/AD1111377. Potyrailo RA (2016) Multivariable sensors for ubiquitous monitoring of gases in the era of Internet of Things and Industrial Internet. Chem Rev. 116:11877–923. Wohltjen H. A Journey: From Sensor Ideas to Sensor Products. Plenary talk at the 11th International Meeting on Chemical Sensors, University of Brescia, Italy, July 16-19, 2006. Elsevier Science; 2006. Mayer F (2010) Innovation for commercial Sensors: From stepping into an opportunity Gap to continuous Innovation. Proc Eng. 5:1-4. Butterfield D. Market review of gas sensors for industrial applications. National Physical Laboratory, Middlesex, UK, NPL Report ENV40; 2021. Potyrailo RA. Enabling New Applications with Multivariable Chem/Bio Sensors: From Ideas to Products. Sensors Summit, San Diego, CA, December 10-12. 2018. General Electric Research Center: Gas sensing with dielectric excitation. (2021) Innovation Award 2021 Association for Sensor and Measurement Technology (AMA) https://www.ama-sensorik.de/en/science/ama-innovation-award/ama-innovation-award-2021/. Potyrailo RA (2006) Polymeric Sensor Materials: Toward an Alliance of Combinatorial and Rational Design Tools? Angew Chem Int Ed. 45:702-23. Potyrailo RA, Mirsky VM (2008) Combinatorial and High-Throughput Development of Sensing Materials: The First Ten Years. Chem Rev. 108:770-813. Potyrailo RA, Ding Z, Butts MD, Genovese SE, Deng T (2008) Selective Chemical Sensing Using Structurally Colored Core-Shell Colloidal Crystal Films. IEEE Sensors J. 8:815-22. doi: 10.1109/JSEN.2008.923191. Potyrailo RA, Naik RR (2013) Bionanomaterials and bioinspired nanostructures for selective vapor sensing. Annu Rev Mater Res. 43:307–34. doi: DOI: 10.1146/annurev-matsci-071312-121710. Musgrove AL, Cockerham A, Pazos JJ, Shahriar S, McMahon M, Touma JE, et al. (2024) Bio-inspired photonic and plasmonic systems for gas sensing: applications, fabrication, and analytical methods. J Opt Microsyst. 4(2):020902-. Ahmed MA, Hu D, Shi Y, Chen Y, Akhavan S, Yang Z (2025) Exploration of Biologically-Inspired Nanostructures: Review on the Sensing Potential and Technological Integration of the Morpho Butterfly Wing. Photonic Sens. 15(2):250227. Potyrailo RA, Hobbs SE, Hieftje GM (1998) Optical waveguide sensors in analytical chemistry: today's instrumentation, applications and future development trends. Fresenius' J Anal Chem. 362:349-73. Persaud K, Dodd G (1982) Analysis of discrimination mechanisms in the mammalian olfactory system using a model nose. Nature. 299:352-5. Shulaker MM, Hills G, Park RS, Howe RT, Saraswat K, Wong HSP, et al. (2017) Three-dimensional integration of nanotechnologies for computing and data storage on a single chip. Nature. 547(7661):74-8. doi: 10.1038/nature22994. Wang C, Chen Z, Chan CLJ, Wan Za, Ye W, Tang W, et al. (2024) Biomimetic olfactory chips based on large-scale monolithically integrated nanotube sensor arrays. Nat Electron. 7:157–67 Potyrailo RA, Ghiradella H, Vertiatchikh A, Dovidenko K, Cournoyer JR, Olson E (2007) Morpho butterfly wing scales demonstrate highly selective vapour response. Nat Photonics. 1:123-8. Potyrailo RA, Starkey T, Vukusic P, Ghiradella H, Vasudev M, Bunning T, et al. (2013) Discovery of the surface polarity gradient on iridescent Morpho butterfly scales reveals a mechanism of their selective vapor response. Proc Natl Acad Sci USA. 110:15567–72. Potyrailo RA, Bonam RK, Hartley JG, Starkey TA, Vukusic P, Vasudev M, et al. (2015) Towards outperforming conventional sensor arrays with fabricated individual photonic vapour sensors inspired by Morpho butterflies. Nat Commun. 6:7959. Potyrailo RA, Go S, Sexton D, Li X, Alkadi N, Kolmakov A, et al. (2020) Extraordinary performance of semiconducting metal oxide gas sensors using dielectric excitation. Nat Electron. 3:280–9. Potyrailo RA, Burns A, Surman C, Lee DJ, McGinniss E (2012) Multivariable passive RFID vapor sensors: roll-to-roll fabrication on a flexible substrate. Analyst. 137:2777-81. Nagraj N, Slocik JM, Phillips DM, Kelley-Loughnane N, Naik RR, Potyrailo RA (2013) Selective sensing of vapors of similar dielectric constants using peptide-capped gold nanoparticles on individual multivariable transducers. Analyst. 138:4334-9. Potyrailo RA (2017) Toward high value sensing: monolayer-protected metal nanoparticles in multivariable gas and vapor sensors. Chem Soc Rev. 46:5311-46. Potyrailo RA, Karker N, Carpenter MA, Minnick A (2018) Multivariable bio-inspired photonic sensors for non-condensable gases. J Opt. 20:024006. Potyrailo RA, Brewer J, Cheng B, Carpenter MA, Houlihan N, Kolmakov A (2020) Bio-inspired gas sensing: boosting performance with sensor optimization guided by “machine learning”. Faraday Discuss. 223:161-82. Potyrailo RA (2024) Multi-Gas Differentiation Using an Individual Optical Fiber Sensor with a Chemically Modified Cladding. ISOCS/IEEE International Symposium on Olfaction and Electronic Nose (ISOEN).4092. Arasaradnam RP, Krishnamoorthy A, Hull MA, Wheatstone P, Kvasnik F, Persaud KC (2025) The Development and Optimisation of a Urinary Volatile Organic Compound Analytical Platform Using Gas Sensor Arrays for the Detection of Colorectal Cancer. Sensors. 25(3):599. Sachan A, Castro M, Feller J-F (2025) Volatolomics for Anticipated Diagnosis of Cancers with Chemoresistive Vapour Sensors: A Review. Chemosensors. 13(1):15. Lonergan MC, Severin EJ, Doleman BJ, Beaber SA, Grubbs RH, Lewis NS (1996) Array-Based Vapor Sensing Using Chemically Sensitive, Carbon Black-Polymer Resistors. Chem Mater. 8:2298-312. Liu T, Wang Y, Wang H (2024) Open set Domain Adaptation for Electronic Nose Drift Compensation on Uncertain Category Data. IEEE Trans Instrum Meas. 73:2505514. Abideen ZU, Arifeen WU, Bandara YND (2024) Emerging trends in metal oxide-based electronic noses for healthcare applications: a review. Nanoscale.Advance Article. Tanveer S, Ahmad MI, Khan T (2024) Technological progression associated with monitoring and management of indoor air pollution and associated health risks: A comprehensive review. Environ Qual Manage.Early View. Han J, Li H, Cheng J, Ma X, Fu Y (2025) Advances in metal oxide semiconductor gas sensor arrays based on machine learning algorithms. J Mater Chem C. Schütze A, Sauerwald T. Dynamic operation of semiconductor sensors. Semiconductor Gas Sensors. Elsevier; 2020. p. 385-412. Potyrailo RA, St-Pierre R, Crowder J, Scherer B, Cheng B. Boosting stability of electronic multi-gas sensors. IEEE SENSORS, Dallas, TX, Oct 30 - Nov 2, Paper 2385; 2022. Potyrailo RA, Scherer B, Brewer J, Ruffalo R. Boosting stability of photonic multi-gas sensors. IEEE SENSORS, Dallas, TX, Oct 30 - Nov 2, Paper 2381; 2022. Piszter G, Kertész K, Bálint Z, Biró LP (2020) Stability and Selective Vapor Sensing of Structurally Colored Lepidopteran Wings Under Humid Conditions. Sensors. 20(11):3258. Potyrailo RA. Tutorial: Next generation gas sensors: surprising advantages over last-century designs drive societal impact. IEEE International Symposium on Olfaction and Electronic Nose (ISOEN), Grapevine, TX, May 12-15; 2024. Nicolet A, Zolla F (2009) Cloaking with curved spaces. Science. 323(5910):46-7. Potyrailo RA, Brewer J, Scherer B, Srivastava V, Nayeri M, Henderson C, et al. (2019) Multi-gas sensors for enhanced reliability of SOFC operation. ECS Transactions. 91:319-28. Austen K (2015) Pollution patrol. Nature. 517:136-8. Lewis A, Edwards P (2016) Validate personal air-pollution sensors. Nature. 535:29-31. Barkjohn KK, Clements A, Mocka C, Barrette C, Bittner A, Champion W, et al. (2024) Air Quality Sensor Experts Convene: Current Quality Assurance Considerations for Credible Data. ACS ES&T Air. 1(10):1203–14. The Nobel Foundation.https://www.nobelprize.org/prizes/lists/all-nobel-prizes/. Wu H-L, Wang T, Yu R-Q (2020) Recent advances in chemical multi-way calibration with second-order or higher-order advantages: Multilinear models, algorithms, related issues and applications. Trends Anal Chem. 130:115954. Alcaraz MR, Monago-Maraña O, Goicoechea HC, de la Peña AM. Recent applications of third-order/four-way and fourth-order/five-way data analysis. Data Handling in Science and Technology. Elsevier; 2024. p. 337-62. Bussey III RO. Uses of Portable Gas Chromatography Mass Spectrometers. Novel Aspects of Gas Chromatography and Chemometrics. IntechOpen; 2022. Smith ME, Westbrook E, Stastny AL, Streicher RP, Elliott MG (2024) Method development for on‑site monitoring of volatile organic compounds via portable TD‑GC-MS: evaluation of the analytical performances of HAPSITE® ER instrumentation and thermal desorption sampling media. J Environ Anal Chem. 104(17):5369-86. Lamy E, Roquencourt C, Zhou B, Salvator H, Moine P, Annane D, et al. (2024) Combination of real-time and hyphenated mass spectrometry for improved characterisation of exhaled breath biomarkers in clinical research. Anal Bioanal Chem. 416(22):4929-39. Gilbert N, Mewis RE, Sutcliffe OB (2020) Classification of fentanyl analogues through principal component analysis (PCA) and hierarchical clustering of GC–MS data. Forensic Chem. 21:100287. Lee LC, Jemain AA (2021) On overview of PCA application strategy in processing high dimensionality forensic data. Microchem J. 169:106608. Dombrowski A, Le D, Lurie IS (2025) GC-VUV spectroscopy of synthetic cannabinoid isomers, diastereomers and homologs: Increasing differentiation by derivative spectral processing. Forensic Chem. 42:100635. Lippitt W, Carlson NE, Arbet J, Fingerlin TE, Maier LA, Kechris K (2024) Limitations of clustering with PCA and correlated noise. J Stat Comput Simul.1-29. Vukusic P, Sambles JR, Lawrence CR, Wootton RJ (1999) Quantified interference and diffraction in single Morpho butterfly scales. Proc R Soc Lond B. 266:1403-11. Houlihan N, Karker N, Potyrailo RA, Carpenter M (2018) High sensitivity plasmonic sensing of hydrogen over a broad dynamic range using catalytic Au-CeO2 thin film nanocomposites. ACS Sens. 3:2684-92. Mei H, Peng J, Wang T, Zhou T, Zhao H, Zhang T, et al. (2024) Overcoming the Limits of Cross-Sensitivity: Pattern Recognition Methods for Chemiresistive Gas Sensor Array. Nanomicro Lett. 16(1):269. Meier M, Kittle JD, Yee XC (2022) Supervised dimension reduction for optical vapor sensing. RSC Adv. 12(16):9579-86. Potyrailo RA, Surman C, Nagraj NN, Burns A (2011) Materials and Transducers Toward Selective Wireless Gas Sensing. Chem Rev. 111:7315–54. Zellers ET, Park J, Hsu T, Groves WA (1998) Establishing a limit of recognition for a vapor sensor array. Anal Chem. 70:4191-201. Born M, Wolf E. Principles of optics: electromagnetic theory of propagation, interference and diffraction of light. 7 (expanded) 60th anniversary ed. Cambridge, UK: Cambridge University Press; 2019. Nagarajan V, Thayumanavan A (2018) CdFe 2 O 4 films for electroresistive detection of ethanol and formaldehyde vapors. Microchim Acta. 185:1-9. Kovalska E, Antonatos N, Luxa J, Sofer Z (2021) Edge-hydrogenated germanene by electrochemical decalcification-exfoliation of CaGe2: Germanene-enabled vapor sensor. ACS Nano. 15(10):16709-18. Zhang R, Wang Y, Li J, Zhao H, Wang Y, Zhou Y (2022) Mesoporous cellulose nanofibers-interlaced PEDOT: PSS hybrids for chemiresistive ammonia detection. Microchim Acta. 189(8):308. Ji H, Mi C, Yuan Z, Liu Y, Zhu H, Meng F (2022) Multicomponent gas detection method via dynamic temperature modulation measurements based on semiconductor gas sensor. IEEE Trans Ind Electron. 70(6):6395-404. Tong L, Wan J, Xiao K, Liu J, Ma J, Guo X, et al. (2023) Heterogeneous complementary field-effect transistors based on silicon and molybdenum disulfide. Nat Electron. 6(1):37-44. Singh S, Shin KY, Moon S, Kim SS, Kim HW (2024) Phase-engineered MoSe2/CeO2 composites for room-temperature gas sensing with a drastic discrimination of NH3 and TEA gases. ACS Sens. 9(8):3994-4006. Diab R, Boltaev G, Kaid MM, Fawad A, El-Kaderi HM, Al-Sayah MH, et al. (2025) Fabrication of heteroatom-doped graphene-porous organic polymer hybrid materials via femtosecond laser writing and their application in VOCs sensing. Sci Rep. 15(1):3682. Bhagat B, Paine S, Manna B, Choudhury A, Mukherjee K (2025) Kinetic Modelling and Machine Learning Approaches on the Conductance Transients of Zn0. 5Ni0. 5Fe2O4 Sensors for Addressing Cross‐sensitivity towards Ethanol and Acetone Vapors. Chem Eur J e202404111. Müller R, Lange E (1986) Multidimensional sensor for gas analysis. Sens Actuators. 9(1):39-48. Singh P, Yadava R (2011) Effect of film thickness and viscoelasticity on separability of vapour classes by wavelet and principal component analyses of polymer-coated surface acoustic wave sensor transients. Meas Sci Technol. 22(2):025202. Luo Z, Weng Z, Shen Q, An S, He J, Fu B, et al. (2020) Vapor detection through dynamic process of molecule desorption from butterfly wings. Pure Appl Chem. 92(2):223-32. Brandt S, Pavlichenko I, Shneidman AV, Patel H, Tripp A, Wong TS, et al. (2023) Nonequilibrium sensing of volatile compounds using active and passive analyte delivery. Proc Natl Acad Sci USA. 120(31):e2303928120. Friedrich RW, Laurent G (2001) Dynamic optimization of odor representations by slow temporal patterning of mitral cell activity. Science. 291(5505):889-94. doi: 10.1126/science.291.5505.889. Ackels T, Erskine A, Dasgupta D, Marin AC, Warner T, Tootoonian S, et al. (2021) Fast odour dynamics are encoded in the olfactory system and guide behaviour. Nature. 593(7860):558-63. van den Broek J, Abegg S, Pratsinis SE, Güntner AT (2019) Highly selective detection of methanol over ethanol by a handheld gas sensor. Nat Commun. 10(1). doi: 10.1038/s41467-019-12223-4. Spencer TL, Clark A, Fonollosa J, Virot E, Hu DL (2021) Sniffing speeds up chemical detection by controlling air-flows near sensors. Nat Commun. 12(1):1232. Weimar U, Göpel W (1998) Chemical imaging: II. Trends in practical multiparameter sensor systems. Sens Actuators, B. 52:143–61. Sears WM, Colbow K, Consadori F (1989) Algorithms to improve the selectivity of thermally-cycled tin oxide gas sensors. Sens Actuators. 19(4):333-49. doi: 10.1016/0250-6874(89)87084-2. Semancik S, Cavicchi R (1998) Kinetically controlled chemical sensing using micromachined structures. Acc Chem Res. 31:279-87. Bur C, Reimann P, Andersson M, Schutze A, Spetz AL (2012) Increasing the selectivity of Pt-gate SiC field effect gas sensors by dynamic temperature modulation. IEEE Sens J. 12(6):1906-13. doi: 10.1109/JSEN.2011.2179645. Bur C, Bastuck M, Puglisi D, Schütze A, Lloyd Spetz A, Andersson M (2015) Discrimination and quantification of volatile organic compounds in the ppb-range with gas sensitive SiC-FETs using multivariate statistics. Sens Actuators, B. 214:225–33. doi: 10.1016/j.snb.2015.03.016. Singh H, Bhaskerraj V, Kumar J, Mittal U, Mishra M, Durani F, et al. (2014) SAW E-nose using single sensor with temperature modulation. IEEE 9th International Conference on Industrial and Information Systems (ICIIS).1-5. Zeng G, Wu C, Chang Y, Zhou C, Chen B, Zhang M, et al. (2019) Detection and Discrimination of Volatile Organic Compounds using a Single Film Bulk Acoustic Wave Resonator with Temperature Modulation as a Multiparameter Virtual Sensor Array. ACS Sens. 4(6):1524-33. Zhang J, Goel N, Bart S, Tadigadapa S. Identification of Volatile Organic Compound Mixtures with a Micromachined Quartz Resonator through Temperature Modulation with On-Chip Integrated Heater. 2024 IEEE International Symposium on Olfaction and Electronic Nose (ISOEN): IEEE; 2024. p. 1-4. Schütze A, Reimann G, Darsch S, Schüler M, Gillen T (2011) Electrical Impedance Spectroscopy Combined with Temperature Cycled Operation for Semiconductor Gas Sensors. Proceedings SENSOR 2011.40-5. Zampolli S, Elmi I, Bruschi P, Ria A, Magliocca F, Vitelli M, et al. (2025) An ASIC-based system-in-package MEMS gas sensor with impedance spectroscopy readout and AI-enabled identification capabilities. Sens Actuators, B. 424:136924. Additional Declarations No competing interests reported. Supplementary Files graphicalabstract.jpg Cite Share Download PDF Status: Published Journal Publication published 09 Jul, 2025 Read the published version in Microchimica Acta → Version 1 posted Editorial decision: Revision requested 03 May, 2025 Reviews received at journal 24 Apr, 2025 Reviews received at journal 23 Apr, 2025 Reviewers agreed at journal 15 Apr, 2025 Reviewers agreed at journal 11 Apr, 2025 Reviewers agreed at journal 31 Mar, 2025 Reviewers invited by journal 30 Mar, 2025 Editor assigned by journal 20 Mar, 2025 Submission checks completed at journal 20 Mar, 2025 First submitted to journal 15 Mar, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6234291","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":443775963,"identity":"2d225764-7e99-4e22-95d5-f459c3872559","order_by":0,"name":"Radislav A. Potyrailo","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAvklEQVRIiWNgGAWjYPACCRkG9gaiVTODtfAw8BwgTQsDD4NEApEazNn7j0lX1Fjw8Es+v/iAocYmmqAWy57DbJJnjknwSM7OKTZgOJaW20BIi8GNZDbJBjYJHoPbOWkSjA2HidXyD6jl5pn0H8RraWwDarnBfoyBKC1AvxhbNvYB/dKTwyyRQIxfzNkbH95s+FYnx89+/OGHDzU2RDiMgYFFAsLkMWBIIKQcqoX5A4TJ/oAYDaNgFIyCUTACAQB5xTlasyj16gAAAABJRU5ErkJggg==","orcid":"","institution":"GE Vernova Advanced Research","correspondingAuthor":true,"prefix":"","firstName":"Radislav","middleName":"A.","lastName":"Potyrailo","suffix":""},{"id":443775964,"identity":"06f5aab5-41ce-442d-b01f-fa6b68e53b6d","order_by":1,"name":"Shiyao Shan","email":"","orcid":"","institution":"GE Vernova Advanced Research","correspondingAuthor":false,"prefix":"","firstName":"Shiyao","middleName":"","lastName":"Shan","suffix":""},{"id":443775965,"identity":"6955edc0-d897-4f32-ab06-2a4478ad37e8","order_by":2,"name":"Baokai Cheng","email":"","orcid":"","institution":"GE Vernova Advanced Research","correspondingAuthor":false,"prefix":"","firstName":"Baokai","middleName":"","lastName":"Cheng","suffix":""}],"badges":[],"createdAt":"2025-03-15 18:08:11","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6234291/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6234291/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s00604-025-07294-8","type":"published","date":"2025-07-09T15:57:20+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":80904160,"identity":"3ccc2f6f-0801-4b4c-8f3c-1eac4be63361","added_by":"auto","created_at":"2025-04-18 14:14:24","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":84524,"visible":true,"origin":"","legend":"\u003cp\u003eMathematics of zero-, first-, and second-order gas detectors. (a) Zero-order detector measures a single parameter, allows detection of only a single gas or volatile analyte, requires total selectivity for this analyte, and has no ability to correct for interferences. (b) First-order detector with one independent variable in its response, differentiates between volatiles based on their response patterns, and rejects calibrated-for interferences. (c) Second-order detector with two independent variables in its response, differentiates between volatiles in a 2-D differentiation map, and makes calibration for numerous unknown interferences mathematically unnecessary.\u003c/p\u003e","description":"","filename":"fig1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6234291/v1/94ff608309267c28a192ae7f.jpg"},{"id":80903366,"identity":"d56085a3-6a72-4249-b28c-fff7cc49f969","added_by":"auto","created_at":"2025-04-18 13:58:26","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":107709,"visible":true,"origin":"","legend":"\u003cp\u003eThe four design aspects (four pillars) of the next generation multi-gas sensors based on first- and second-order detection principles.\u003c/p\u003e","description":"","filename":"fig2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6234291/v1/a0bbcb62055e1651c462435c.jpg"},{"id":80903346,"identity":"59adfd5b-fc84-468b-8a8b-a47a69883478","added_by":"auto","created_at":"2025-04-18 13:58:25","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":72643,"visible":true,"origin":"","legend":"\u003cp\u003eNext-generation multi-gas sensors as Moonshot second-order detectors to compete against exquisite performance of second-order traditional analytical instruments and to break through tradeoff between SWaP-C versus instrument performance.\u003c/p\u003e","description":"","filename":"fig3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6234291/v1/b6348542091006b496f5c463.jpg"},{"id":80903344,"identity":"bc4b97b7-0d3b-4019-8aa5-7d4685207154","added_by":"auto","created_at":"2025-04-18 13:58:24","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":125579,"visible":true,"origin":"","legend":"\u003cp\u003eNatural nanostructures of \u003cem\u003eMorpho\u003c/em\u003e butterflies for\u003cstrong\u003e \u003c/strong\u003esecond-order gas sensing. Examples of \u003cem\u003eMorpho\u003c/em\u003ebutterflies: (a) \u003cem\u003eMorpho\u003c/em\u003e sulkowskyi and (b) \u003cem\u003eMorpho\u003c/em\u003e rhetenor. (c) While-light illumination of a portion of a \u003cem\u003eMorpho\u003c/em\u003e butterfly wing. (d) Scanning electron microscopy image of \u003cem\u003eMorpho\u003c/em\u003e tree-like photonic nanostructure, showing an open-air architecture with vertical ridges and horizontal lamella, allowing gas interactions with its regions. (e) Reflectance spectra of a \u003cem\u003eMorpho\u003c/em\u003ebutterfly in dry air under parallel and perpendicular polarization states of white light.\u003c/p\u003e","description":"","filename":"fig4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6234291/v1/ee2e9b0c3642404145c4531a.jpg"},{"id":80903336,"identity":"aa46b73b-1828-4f9d-a1b9-143252e8e61d","added_by":"auto","created_at":"2025-04-18 13:58:24","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":101288,"visible":true,"origin":"","legend":"\u003cp\u003eFabricated bio-inspired nanostructures for\u003cstrong\u003e \u003c/strong\u003esecond-order gas sensing. (a, b) Scanning electron microscopy images of different geometries of nanostructures with catalytic nanoparticles. (c) Fabricated bio-inspired nanostructured optical sensing elements under ambient-light illumination. (d) Bio-inspired nanostructured optical sensing element positioned on a microheater for second-order chemical sensing. Also shown are fiber-optic probe and a gas-delivery tubing. (e) Reflectance spectrum of a bio-inspired nanostructured optical sensing element in dry air under white light illumination.\u003c/p\u003e","description":"","filename":"fig5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6234291/v1/1b3cb2dcd00d22429fb274c4.jpg"},{"id":80903639,"identity":"9fb96896-c3e1-45a2-984a-59c005e9c3aa","added_by":"auto","created_at":"2025-04-18 14:06:24","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":93082,"visible":true,"origin":"","legend":"\u003cp\u003eDifferential reflectance spectra ΔR (l, q) of the \u003cem\u003eMorpho\u003c/em\u003e nanostructure to ten vapors as a function of two independent variables, the illumination wavelength l across the visible spectral range and the illumination angle q of the sensing element. Ten vapors 1–10: water, ethanol, 1-methoxy-2-propanol, methyl ethyl ketone, acetonitrile, toluene, chloroform, tetrahydrofuran, dimethylformamide, and acetone. Zero is blank.\u003c/p\u003e","description":"","filename":"fig6.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6234291/v1/2f75a968a5438671cff3e4ac.jpg"},{"id":80903641,"identity":"795ac53d-2d0a-46de-ac8b-646a950d6cfe","added_by":"auto","created_at":"2025-04-18 14:06:25","extension":"jpg","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":66204,"visible":true,"origin":"","legend":"\u003cp\u003ePCA scores plots of the first several PCs with the data analyzed with all three illumination angles of 20\u003csup\u003eo\u003c/sup\u003e to 40\u003csup\u003eo\u003c/sup\u003e and to 60\u003csup\u003eo\u003c/sup\u003e. (a) Scores plot of PC1 vs PC2. (b-d) Scores plots of PC5 vs PC6 vs PC7 at their different projections, demonstrating that all ten vapors were well differentiated.\u0026nbsp; Ten vapors 1–10: water, ethanol, 1-methoxy-2-propanol, methyl ethyl ketone, acetonitrile, toluene, chloroform, tetrahydrofuran, dimethylformamide, and acetone. Zero is blank.\u003c/p\u003e","description":"","filename":"fig7.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6234291/v1/1a8bf9c078d58c71973aee93.jpg"},{"id":80903375,"identity":"c492d49e-f3b5-4472-9ebe-374bb418ee55","added_by":"auto","created_at":"2025-04-18 13:58:26","extension":"jpg","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":101630,"visible":true,"origin":"","legend":"\u003cp\u003eEstimated SNR of each PC in the PCA models when spectral measurements were performed at different angles. (a–c) Individual 20\u003csup\u003eo\u003c/sup\u003e, 40\u003csup\u003eo\u003c/sup\u003e, and 60\u003csup\u003eo\u003c/sup\u003e angles, respectively. (d) Combination of three illumination angles. \u0026nbsp;\u0026nbsp;\u0026nbsp;\u003c/p\u003e","description":"","filename":"fig8.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6234291/v1/03cbbed5b07b91af59575663.jpg"},{"id":80903637,"identity":"d4880f05-159f-4828-95be-a3677edafa44","added_by":"auto","created_at":"2025-04-18 14:06:24","extension":"jpg","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":154180,"visible":true,"origin":"","legend":"\u003cp\u003ePolarization- and wavelength-dependent reflectivity changes of the \u003cem\u003eMorpho\u003c/em\u003e structure upon exposure to acetonitrile and ethanol vapors.\u0026nbsp; (a, b) Dynamic response patterns to acetonitrile and ethanol vapors at six wavelengths, two polarizations of white light, and six concentrations of vapors (0.1, 0.2, 0.3, 0.4, 0.6, and 0.8 P/P\u003csub\u003e0\u003c/sub\u003e).\u003c/p\u003e","description":"","filename":"fig9.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6234291/v1/43b774622eb469c9c46bf92d.jpg"},{"id":80903331,"identity":"1bad4549-46ca-45fb-98b0-677ed29a1b0f","added_by":"auto","created_at":"2025-04-18 13:58:24","extension":"jpg","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":70893,"visible":true,"origin":"","legend":"\u003cp\u003eContour 2-D plots for (a) acetonitrile and (b) ethanol vapors as a function of two independent variables, the illumination wavelength and the polarization state of white light.\u003c/p\u003e","description":"","filename":"fig10.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6234291/v1/29111543069e26f40448bf44.jpg"},{"id":80903646,"identity":"285b7244-dc86-492f-823b-26a4ec473f67","added_by":"auto","created_at":"2025-04-18 14:06:26","extension":"jpg","order_by":11,"title":"Figure 11","display":"","copyAsset":false,"role":"figure","size":41711,"visible":true,"origin":"","legend":"\u003cp\u003ePCA scores plots with the data for acetonitrile and ethanol vapors with their tested concentrations analyzed with both polarizations. (a) PC1 vs PC2. (b) PC1 vs PC2 vs PC3.\u003c/p\u003e","description":"","filename":"fig11.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6234291/v1/c4f3b0a954146bb274970207.jpg"},{"id":80904390,"identity":"080b051a-a3f1-40a2-a325-58d24d6bfefe","added_by":"auto","created_at":"2025-04-18 14:22:24","extension":"jpg","order_by":12,"title":"Figure 12","display":"","copyAsset":false,"role":"figure","size":171181,"visible":true,"origin":"","legend":"\u003cp\u003eDynamic responses of the second-order sensor to two volatiles at selected wavelengths and temperatures. (a, b) temperatures of 20°C and 380°C. H\u003csub\u003e2\u003c/sub\u003e gas concentrations: 0.625, 1.25, 1.875, and 2.5%; H\u003csub\u003e2\u003c/sub\u003eO vapor concentrations: 4, 8, 12, and 16% RH. Illustrative wavelengths of sensor response at 450, 480, 520, 560, 660, 700, and 965 nm.\u003c/p\u003e","description":"","filename":"fig12.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6234291/v1/5f808175663423440f2a902e.jpg"},{"id":80903326,"identity":"81fa2d65-f11d-4225-91f0-87d9955a7382","added_by":"auto","created_at":"2025-04-18 13:58:24","extension":"jpg","order_by":13,"title":"Figure 13","display":"","copyAsset":false,"role":"figure","size":51967,"visible":true,"origin":"","legend":"\u003cp\u003eContour 2D plots for (a) H\u003csub\u003e2\u003c/sub\u003e gas and (b) H\u003csub\u003e2\u003c/sub\u003eO vapor as a function of two independent variables, the illumination wavelength and the sensor temperature. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e","description":"","filename":"fig13.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6234291/v1/ed40435b2abc138ee56e79c1.jpg"},{"id":80903640,"identity":"0050c6a9-bf80-4d9e-afe8-7f881d80f06a","added_by":"auto","created_at":"2025-04-18 14:06:25","extension":"jpg","order_by":14,"title":"Figure 14","display":"","copyAsset":false,"role":"figure","size":18779,"visible":true,"origin":"","legend":"\u003cp\u003ePCA scores plot with the data for H\u003csub\u003e2\u003c/sub\u003e gas and H\u003csub\u003e2\u003c/sub\u003eO vapor with their tested concentrations analyzed with both temperatures.\u0026nbsp;\u0026nbsp; \u0026nbsp;\u0026nbsp;\u003c/p\u003e","description":"","filename":"fig14.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6234291/v1/fdf0113f011cad08434ba4a0.jpg"},{"id":86699362,"identity":"3411de0b-a7b4-4c9e-9964-08b954c6d764","added_by":"auto","created_at":"2025-07-14 16:08:19","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2040197,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6234291/v1/9613f28d-e9dd-4dd4-826b-6a46e1a4f18f.pdf"},{"id":80903634,"identity":"a1deabd1-2e76-47aa-a867-4c0c26384d86","added_by":"auto","created_at":"2025-04-18 14:06:24","extension":"jpg","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":67812,"visible":true,"origin":"","legend":"","description":"","filename":"graphicalabstract.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6234291/v1/9d4a8608b677024a9ca43465.jpg"}],"financialInterests":"No competing interests reported.","formattedTitle":"Individual Optical Multi-Gas Sensors as Next Generation Second-Order Unobtrusive and Continuous Operation Analytical Instruments","fulltext":[{"header":"Introduction","content":"\u003cp\u003eThe knowledge of \u0026ldquo;what is in the air\u0026rdquo; drives mitigation of air pollutants\u0026mdash;starting from the industrial revolution in 1800s [1], to detection of chemical agents by soldier \u0026ldquo;gas scouts\u0026rdquo; during the World War I [2], detection of leaks of cooking gases in residential households in 1960s [3], detection of outdoor pollutants in 1970s [4], and detection of per- and polyfluoroalkyl (PFAS) air pollutants today [5]. To analyze gaseous species (gases or volatiles), there are two broad categories of gas detectors: traditional analytical instruments and gas sensors. \u003c/p\u003e\n\u003cp\u003e\u003cem\u003eTraditional analytical instruments\u003c/em\u003e are designed with one, two, or more independent response variables as first-, second-, and higher-order instruments to provide accurate information about one or more analytes in chemically complex environments [6]. Examples of independent response variables are wavelength in spectroscopy, retention time in gas chromatography (GC), and mass-to-charge ratio in mass spectrometry (MS). These instruments require many materials and high energy to manufacture these instruments, and also demand costly maintenance, consumables, and power demands during their end-use [7, 8]. \u003c/p\u003e\n\u003cp\u003e\u003cem\u003eGas sensors\u003c/em\u003e are designed as small-size, single-output (e.g., light intensity, resistance, etc.) devices, also known as zero-order devices [6, 9-12], to reversibly respond to a concentration change of an expected gas in a simple chemical environment. Gas sensors are designed either to monitor intrinsic properties of gases or to utilize a gas-sensing material [9]. Gas sensors have poor detection accuracy in complex chemical environments and experience sensor drift [9, 10, 12].\u003c/p\u003e\n\u003cp\u003eContemporary designs of material-based gas sensors utilize signal-transduction principles based on electrical, acoustic, and optical energy [9, 13]. These sensors thrive to progress from proof-of-concept demonstrations to commercial implementations [14-18]. Optical material-based gas sensors consist of two broad categories: one that utilizes chemical effects of gas interactions with organic reagents and one that focused on the physical effects of gas interactions with nanostructured materials [13, 19-24]. Compared to electrical and acoustic types of sensors, optical material-based gas sensors have a broader range of design parameters, e.g., five properties of light such as intensity, wavelength, phase, polarization, and temporal property [25]. \u003c/p\u003e\n\u003cp\u003eTo expand gas sensor opportunities, the interdisciplinary research community is learning from mathematics behind zero-order sensor designs that originated from the 1960s [6, 9-12] how to overcome their limitations. Two directions are explored, such as arrays of zero-order sensors [26-28] and individual first-order (a.k.a. multivariable) sensors with diverse response variables (e.g., wavelength, frequency, temperature, etc.) [13, 29-38]. Arrays of zero-order sensors differentiate multiple volatiles [26, 39, 40] and have up to five dimensions (5-D) of sensor-array dispersion [41], but an independent drift of individual sensors in an array remains a persistent problem, preventing arrays from being broadly implemented by end-users [42-45]. First-order gas sensors showed self-correction of baseline drift [46-48], displayed 5-D sensor dispersion [49], and differentiated multiple volatiles with one sensor [13, 35, 50]. \u003c/p\u003e\n\u003cp\u003eIn this report, we show opportunities for photonic nanostructured sensors as second-order analytical instruments. In prior work, we noticed extraordinary multi-vapor differentiation using spectral responses of scales of iridescent \u003cem\u003eMorpho\u003c/em\u003e butterflies [29]. Structurally colored \u003cem\u003eMorpho\u003c/em\u003e butterfly scales are also known as natural metamaterials [51]. From experimental and theoretical studies of the multi-gas responses of \u003cem\u003eMorpho\u003c/em\u003e scales and bio-inspired fabricated sensors, we developed design rules with physical (nanostructure geometry) and chemical (spatially controlled chemical functionalization) design criteria for first-order gas sensors for operation at room and high temperatures [29-31, 36, 37, 48, 52]. Here, we describe results of three designs of optical photonic nanostructured gas sensors that operate as second-order sensors. In all three designs, the first independent variable in sensor response is optical wavelength across the visible spectral range and the second independent variable is either an illumination angle, light polarization, or operation temperature of the sensor. These illustrative designs of second-order gas sensors show the opportunities that are unveiled by bringing mathematical understanding to boost the multi-gas resolution by individual material-based sensors. \u003c/p\u003e\n\n\u003cp\u003e\u003cstrong\u003eFrom zero- to first- and second-order individual multi-gas sensors \u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAt present, the interdisciplinary sensing communities rarely describe sensors with one or two independent response variables based on mathematical descriptions, such as first- and second-order sensors [10, 11]. Instead, such sensors are known as multivariable, intelligent, multiparameter, virtual multisensor systems, or virtual sensor arrays [13]. Mathematical descriptions of such sensor designs as first- and second-order sensors provide more insights on further improvements of gas sensors. \u003c/p\u003e\n\u003cp\u003eZero-order gas sensor designs have conflicting requirements for reversibility and selectivity because the interaction energy between the sensing material and the analyte should be low to provide a reversible sensor response, but this low interaction energy leads to poor selectivity [9]. Poor selectivity often cancels out size and cost advantages of sensors over traditional analytical instruments [8, 53-55]. \u003c/p\u003e\n\u003cp\u003eThe mathematics behind traditional analytical instruments inspire next-generation gas sensors. First- and second-order traditional analytical instruments are designed with one or two independent response variables. Such designs provide solutions to complex measurement problems in diverse applications. These instrument designs deliver huge societal benefits, recognized with three Nobel Prizes [56]. \u003c/p\u003e\n\u003cp\u003eAs shown in \u003cstrong\u003eFig. 1a\u003c/strong\u003e, using a zero-order detector (with only a single output) allows detection of only a single analyte and requires a total selectivity for this analyte. Mathematically, a zero-order detector (sensor) cannot differentiate between gases. In detection of small analyte concentrations, the presence of interferences has more pronounced effects, degrading accuracy. \u003c/p\u003e\n\u003cp\u003eFirst-order detectors have one independent variable in their response, as shown in \u003cstrong\u003eFig. 1b\u003c/strong\u003e (e.g., wavelength in spectroscopic, retention time in GC, or mass-to-charge ratio in MS instruments). To differentiate between volatiles, it is critical to have different response patterns between different volatiles as a function of the chosen independent variable. For example, volatile 1 and volatile 2 have response patterns that are close to each other, while volatile 3 has a very different response pattern (see \u003cstrong\u003eFig. 1b\u003c/strong\u003e). Using a first-order detector, it is possible to reject interferences from the instrument response to volatiles. The response patterns from all expected interferences should be included in calibration (a.k.a. training). \u003c/p\u003e\n\u003cp\u003eWhen a detector has two independent response variables (e.g., GCxGC, GCxMS), a second-order detector improves differentiation between volatiles in a two-dimensional (2-D) differentiation map (see \u003cstrong\u003eFig. 1c\u003c/strong\u003e) and does not require calibration for unknown interferences for analysis in complex samples [57, 58]. Such capability is known as the second-order advantage [6, 57, 58]. Second- and higher-order analytical instruments are must-have solutions for analysis of trace levels of analytes in cluttered environments (e.g., exhaled breath, battlefield air) [59-61], with multi-dimensional data analyzed by multivariate statistical data analysis techniques [62-65]. \u003c/p\u003e\n\u003cp\u003eInspired by mathematics behind traditional analytical instruments, next-generation gas sensors, based on first- and second-order detection principles, are being built by implementing four design aspects (four pillars), as shown in \u003cstrong\u003eFig. 2\u003c/strong\u003e. \u003cem\u003eFirst\u003c/em\u003e, a physical transducer is designed with proper excitation conditions to operate with independent variables. \u003cem\u003eSecond\u003c/em\u003e, a sensing material is implemented with different responses to multiple volatiles under variable excitation conditions. \u003cem\u003eThird\u003c/em\u003e, excitation type and range are designed to induce independent responses in the transducer / sensing material system. \u003cem\u003eFourth, \u003c/em\u003edata analytics provide not only multi-gas differentiation, quantification, and rejection against known (first-order sensors) and unknown (second-order sensors) interferences, but also self-correction against drift. \u003c/p\u003e\n\u003cp\u003eThese next-generation multi-gas sensors (see \u003cstrong\u003eFig. 3\u003c/strong\u003e) are under development as Moonshot second-order detectors to compete against the exquisite performance of traditional analytical instruments. These multi-gas sensors are breaking the trade-off between size, weight, power, and cost (SWaP-C) versus performance of analytical instruments, facilitating the widespread application of these sensors to serve emerging societal needs. \u003c/p\u003e\n"},{"header":"Experimental ","content":"\u003cp\u003e\u003cstrong\u003e\u003cem\u003eNanostructured optical sensing elements \u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn this work, two types of nanostructured optical sensing elements were utilized. A sensing element of type 1 was in the form of dried scales of \u003cem\u003eMorpho\u003c/em\u003e butterflies, obtained from Butterfly Utopia (Brooklyn, NY, USA). A ~5 mm x 5 mm portion of a dried intact \u003cem\u003eMorpho\u003c/em\u003e butterfly wing was utilized for vapor-response tests. A tree-like photonic nanostructure of iridescent tropical \u003cem\u003eMorpho\u003c/em\u003e butterflies has an open-air architecture with ridges and lamella, allowing gas interactions with its regions [29, 30]. Lamella act as multilayer interferometric nano-reflectors and ridges act as a diffraction grating [66]. A sensing element of type 2 was in the form of a ~2 mm x 2 mm piece of wafer with nanofabricated bio-inspired nanostructures for operation at high temperatures. We developed design rules and fabricated such sensors for operation at room and high temperatures [31, 36, 37, 52]. For high temperature applications at ~300-380\u003csup\u003eo\u003c/sup\u003eC, nanostructures were fabricated using conventional photolithography and chemical etching [36, 37, 52]. Nanostructure materials were inorganic and included catalytic metal nanoparticles and metal-oxide capping layers to promote reactions with gases of interest at high operation temperatures [37, 67]. \u003c/p\u003e\n\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eExcitation and vapor testing of nanostructured optical sensing elements\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSensing elements were illuminated with a white light source (a halogen lamp) in the reflection mode, with spectroscopic detection of the reflected light using a spectrograph [29-31, 37]. Measurements were done as the differential reflectance responses \u0026Delta;R(\u0026lambda;) = R(\u0026lambda;)/R\u003csub\u003e0\u003c/sub\u003e(\u0026lambda;), where R(\u0026lambda;) is the spectrum of the sensing element upon exposure to the gaseous species (volatiles) of interest and R\u003csub\u003e0\u003c/sub\u003e(\u0026lambda;) is the spectrum of the sensing element upon exposure to a carrier gas (a blank) at a given condition (e.g., illumination angle, light polarization, operation temperature). Thus, the common features in the two spectra before and during gas exposure cancel out and the \u0026Delta;R(\u0026lambda;) spectrum accentuates the subtle differences due to the gas response [21, 29-31, 37]. Two linear polarizations of white light were achieved using standard components. Model volatiles (vapors) such as water, ethanol, 1-methoxy-2-propanol, methyl ethyl ketone, acetonitrile, toluene, chloroform, tetrahydrofuran, dimethylformamide, and acetone as well as hydrogen gas were selected to demonstrate the capabilities of the second-order gas sensors. Different concentrations of volatiles were produced using built-in-house computer-controlled vapor-generation and mixing systems [31, 37]. Concentrations of majority of vapors were defined as P/P\u003csub\u003e0\u003c/sub\u003e, where P is the vapor partial pressure and P\u003csub\u003e0\u003c/sub\u003e is the saturated vapor pressure [29, 31]. Concentrations of water vapor were defined as percent relative humidity (%RH). Concentrations of H\u003csub\u003e2\u003c/sub\u003e gas were in percent by volume. Iridescent nanostructures of \u003cem\u003eMorpho\u003c/em\u003e butterfly scales and reflectance spectra of a \u003cem\u003eMorpho\u003c/em\u003e butterfly in dry air under parallel and perpendicular polarization states of white light are depicted in \u003cstrong\u003eFig. 4\u003c/strong\u003e. Examples of bio-inspired nanostructures and a reflectance spectrum are shown in \u003cstrong\u003eFig. 5.\u003c/strong\u003e \u003c/p\u003e\n\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eMultivariate data analytics \u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSensor data was analyzed using KaleidaGraph (Synergy Software, Reading, PA, USA) and PLS_Toolbox Software (Eigenvector Research, Wenatchee, WA, USA), operated with MATLAB (Mathworks, Natick, MA, USA). A multivariate data analysis technique such as Principal Components Analysis (PCA) was used for differentiation of volatiles as needed. PCA remains the most popular unsupervised tool for dimensionality reduction in diverse data sets ranging from traditional analytical instruments [62-65] to sensor arrays [27, 28, 68] and individual multivariable sensors [13, 50, 69, 70]. For statistics, we used several data points per concentration of each vapor and the blank or an added 5% noise [71] into the data for more conservative estimations. \u003c/p\u003e"},{"header":"Results and Discussion","content":"\u003cp\u003e\u003cstrong\u003e\u003cem\u003eSpectral and angular differentiation of volatiles\u0026nbsp;\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA sensing element utilized in these experiments was a portion of a \u003cem\u003eMorpho\u003c/em\u003e butterfly wing (Type 1 sensing element, see section \u003cem\u003eNanostructured Optical Sensing Elements\u003c/em\u003e). The first independent variable in its response to vapors was illumination wavelength across the visible spectral range. The second independent variable was illumination angle of the sensing element. In earlier experiments with natural \u003cem\u003eMorpho\u003c/em\u003e nanostructures to differentiate three vapors, diversity of spectral responses to vapors was noted to depend on the illumination angle [29, 37]. Here, the effect of illumination angle was explored with ten diverse model vapors such as water, ethanol, 1-methoxy-2-propanol, methyl ethyl ketone, acetonitrile, toluene, chloroform, tetrahydrofuran, dimethylformamide, and acetone at one level of ~0.5 P/P\u003csub\u003e0\u003c/sub\u003e. Vapor exposures were done at three illumination angles of 20\u003csup\u003eo\u003c/sup\u003e, 40\u003csup\u003eo\u003c/sup\u003e, and 60\u003csup\u003eo\u003c/sup\u003e normal to the surface of the sensing element.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFig. 6\u003c/strong\u003e depicts spectral responses of this sensing element to ten vapors as a function of two independent variables such as the illumination wavelength across the portion of the visible spectral range where iridescence occurs (360\u0026ndash;650 nm) and the illumination angle of a nanostructured sensing material. The visual diversity of differential reflectance spectra \u0026Delta;R(\u0026lambda;) at different illumination angles can be summarized as three categories. \u003cem\u003eFirst\u003c/em\u003e, these spectra had three general regions: 360\u0026ndash;430 nm with an inflection point at ~400 nm, 430\u0026ndash;490 nm with an inflection point at ~460\u0026ndash;480 nm, and 490\u0026ndash;650 nm with an inflection point at ~520 nm. \u003cem\u003eSecond\u003c/em\u003e, the relative intensity of the inflection point at 460\u0026ndash;480 nm was increasing with illumination angle, while the relative intensity of the inflection point at ~520 nm was decreasing with illumination angle. \u003cem\u003eThird\u003c/em\u003e, the overall spectral intensity was decreasing with illumination angle.\u003c/p\u003e\n\u003cp\u003eFor a quantitative interpretation, \u003cstrong\u003eFig. 7\u003c/strong\u003e illustrates PCA scores plots of the first several principal components (PCs) with the data analyzed with all three illumination angles. For conservative estimations of the number of useful PCs, we had five replicates of each spectrum with added 5% random noise [71]. \u003cstrong\u003eFig. 7a\u003c/strong\u003e depicts a scores plot of PC1 vs PC2 without noticeable contributions of the added 5% noise. We have found that our data set with all three illumination angles had seven PCs that differentiated ten vapors, as shown in \u003cstrong\u003eFig. 7b-d,\u003c/strong\u003e depicting scores plots of PC5 vs PC6 vs PC7 at their different projections. The first three PCs had contributions of PC1 = 54.61%, PC2 = 22.66%, and PC3 = 13.85%, capturing only 91.12% of the total spectral variance. Thus, the remaining variance of 8.88 % was available for the enhanced multi-vapor differentiation. To evaluate this remaining variance, the signal-to-noise ratio (SNR) of each PC was analyzed [13].\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFig. 8\u003c/strong\u003e depicts plots of estimated SNR for each PC when spectral measurements were performed at three individual angles and their combination. The number of PCs needed to describe the data with SNR \u0026gt; 3 is highlighted as green data points. The PCA models built based on individual illumination angles had four PCs with SNR \u0026gt; 3 for 20\u003csup\u003eo\u003c/sup\u003e angle (\u003cstrong\u003eFig. 8a\u003c/strong\u003e) and five PCs with SNR \u0026gt; 3 for 40\u003csup\u003eo\u003c/sup\u003e and 60\u003csup\u003eo\u003c/sup\u003e angles (\u003cstrong\u003eFig. 8b, c\u003c/strong\u003e). The PCA model built based on the combination of three illumination angles had seven PCs with SNR \u0026gt; 3 (\u003cstrong\u003eFig. 8d\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003eThus, illumination of the \u003cem\u003eMorpho\u003c/em\u003e nanostructure at multiple angles combined with the spectral analysis across the visible wavelengths achieved an extraordinary seven dimensions in the response dispersion of an individual sensor. A 5-D response dispersion was reported with a white light illumination of an iridescent \u003cem\u003eHypochrysops Polycletus\u003c/em\u003e butterfly at a single angle [49]. At this moment, we did not come across reports on response dispersion of gas sensor arrays higher than 5-D or individual multivariable gas sensors higher than 5-D [50]. A possibility for a 10-D dispersion using an individual bio-inspired optical sensor was demonstrated theoretically [13].\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eSpectral and polarization differentiation of volatiles\u0026nbsp;\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eUtilization of polarization readout in material-based optical gas sensors to improve their multi-gas differentiation is rare [25]. This approach requires system understanding of interrelations between sensor system components when improving sensor performance for multi-gas differentiation for emerging applications. The sensing element utilized in this section was a portion of a wing of a \u003cem\u003eMorpho\u003c/em\u003e butterfly (Type 1 sensing element, see section \u003cem\u003eNanostructured Optical Sensing Elements\u003c/em\u003e). The first independent variable in its response to vapors was the illumination wavelength across the visible spectral range. The second independent variable was two orthogonal linear polarization states of white light that were used to illuminate the sensing element. The two orthogonal linear polarization states were parallel and perpendicular polarizations. The parallel polarization is also known as P- (from the German parallel) and transverse-magnetic (TM) polarization. The perpendicular polarization is also known as S- (from the German senkrecht, perpendicular) and transverse-electric (TE) polarization [72].\u003c/p\u003e\n\u003cp\u003eThese experiments were focused on the ability of this second-order sensor to differentiate between two model volatiles, acetonitrile and ethanol. Using only two model volatiles for initial demonstrations is common, as shown in other reports [73-80]. Polarization- and wavelength-dependent reflectivity changes of this sensing element were measured upon exposure to six concentrations of these two model vapors (0.1, 0.2, 0.3, 0.4, 0.6, and 0.8 P/P\u003csub\u003e0\u003c/sub\u003e). At different wavelengths, diverse response patterns at two polarizations for two vapors at their six concentrations were observed, as depicted in \u003cstrong\u003eFig. 9\u003c/strong\u003e. This diversity can be summarized as three categories. \u003cem\u003eFirst\u003c/em\u003e, the relative intensities of signal changes of the parallel and perpendicular polarizations were wavelength- and vapor-dependent. \u003cem\u003eSecond,\u0026nbsp;\u003c/em\u003ediversity of vapor responses between parallel and perpendicular polarizations were substantially reduced at longer wavelengths (540 nm and 590 nm). \u003cem\u003eThird,\u0026nbsp;\u003c/em\u003ediverse dynamics of the responses of the parallel and perpendicular polarizations were also wavelength- and vapor-dependent.\u003c/p\u003e\n\u003cp\u003ePreviously with unpolarized light, diverse vapor-response dynamics of \u003cem\u003eMorpho\u003c/em\u003e scales was observed [29, 30] due to physical adsorption and capillary condensation of different vapors in the nano-size domains of the \u003cem\u003eMorpho\u003c/em\u003e scales. At low vapor concentrations, molecules adsorbed onto the nanostructure, while at higher concentrations, a capillary condensation of vapor molecules occurred starting at the smallest radii and spreading throughout the lamellae. Thus, abrupt dynamic changes in reflectance occurred mostly at shorter wavelengths, because the smaller spatial periodicity or size of the photonic substructure was associated with the shorter response wavelengths [29]. We purposefully did not include analysis of dynamic spectral vapor-response features that are clearly visible at the short illumination wavelengths with polarized light (see \u003cstrong\u003eFig. 9\u0026nbsp;\u003c/strong\u003eat 410, 440, 480, and 520 nm). Numerous excellent reports demonstrate the ability of individual electrical, acoustic, and optical sensing elements to benefit from diverse dynamics of different vapors for their differentiation [80-84]. However, as noted earlier [11], such dynamic measurements that mimic natural olfaction [85, 86] require a well-controlled sample introduction, borrowing sample introduction principles with needed pumps, valves, and other mechanical components from gas chromatography and other gas introduction technologies [87, 88]. Such conventional sample-handlining complexity negates the intended simplicity of sensor system designs and applications.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFig. 10\u003c/strong\u003e depicts contour 2-D plots for two vapors with the first independent variable being wavelength and the second independent variable being polarization states of white light. For this illustrative visualization, the highest tested concentrations of vapors were utilized and contour plots were presented with eight contours of the DR signal at each polarization. At different wavelengths and polarizations, diversity of response patterns for two vapors were three-fold. \u003cem\u003eFirst\u003c/em\u003e, the number and shapes of the contours were related to the vapor type and light polarization. At the parallel polarization, both vapors had only two contours but of different shape. At the perpendicular polarization, acetonitrile had four contours, while ethanol had three contours. \u003cem\u003eSecond,\u0026nbsp;\u003c/em\u003eat shorter wavelengths of 410\u0026ndash;440 nm, the diversity between vapors was more pronounced in the number and shapes of the contours. At the parallel polarization, both vapors had one contour but of different shape. At the perpendicular polarization, acetonitrile had two contours, while ethanol had one contour. \u003cem\u003eThird,\u0026nbsp;\u003c/em\u003eat longer wavelengths of 480\u0026ndash;590 nm, the diversity between vapors was less pronounced\u0026mdash;it was only a single but different shape contour for both vapors and their two polarizations.\u003c/p\u003e\n\u003cp\u003eWe also performed PCA analysis of responses of the sensor to all tested concentrations of vapors, using both polarizations in the PCA model. \u003cstrong\u003eFig. 11\u003c/strong\u003e depicts PCA scores plots of the first two and three PCs with the data analyzed with both polarizations. The first three PCs had contributions of PC1 = 88.40%, PC2 = 9.15%, and PC3 = 2.00%, capturing 99.55% of the total variance in the spectra. Some non-monotonic or abrupt changes in the data points in the scores plots were likely the effects of adsorption and condensation of different vapors into the nanostructured features of the sensing element, with an increase in concentrations of vapors (\u003cstrong\u003eFig. 11a\u003c/strong\u003e). The 3-D plot shown in \u003cstrong\u003eFig. 11b\u003c/strong\u003e attempts to show that in addition to two vapors in the data set, there were also two different polarization states that affected PCA scores plot by adding PC3.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eSpectral and temperature differentiation of volatiles\u0026nbsp;\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSensor operation at different temperatures is a known principle to increase the order of sensor response because chemical interactions of diverse volatiles with a sensing material can be substantially different over appropriately chosen temperature ranges [89]. Previously, individual first-order sensors based on variable temperature of operation of sensing materials were demonstrated on different transducers, such as semiconducting metal oxides on resistors [46, 90, 91], non-oxide semiconductors on field-effect transistors [92, 93], and polymers on acoustic resonant sensors [94-96]. Second-order sensors were also demonstrated with a temperature modulation of semiconducting metal oxide sensing materials, coupled with impedance readout [97, 98]. Importantly, multi-temperature operation is now a common practice in numerous types of commercially available chemiresistor sensors with semiconducting metal oxides [50].\u003c/p\u003e\n\u003cp\u003eThe sensing element utilized in this experiment was a fabricated bio-inspired nanostructured sensing element (Type 2 sensing element, see section \u003cem\u003eNanostructured Optical Sensing Elements\u003c/em\u003e). The first independent variable in its response to vapors was spectral response across the visible spectral range and its second independent variable was its operation temperature. As an illustrative example, we show results of initial experiments that were focused on the ability of this second-order sensor to differentiate between H\u003csub\u003e2\u003c/sub\u003e gas and H\u003csub\u003e2\u003c/sub\u003eO vapor in its future intended application for H\u003csub\u003e2\u003c/sub\u003e gas monitoring in the fuel stream of solid oxide fuel cells [52]. In these initial tests, the sensor was exposed to H\u003csub\u003e2\u003c/sub\u003e gas and H\u003csub\u003e2\u003c/sub\u003eO vapor at two temperatures, ~20\u0026deg;C (room temperature T1) and ~380 \u0026deg;C (T2). H\u003csub\u003e2\u0026nbsp;\u003c/sub\u003egas concentrations were 0.625, 1.25, 1.875, and 2.5%; H\u003csub\u003e2\u003c/sub\u003eO vapor concentrations were 4, 8, 12, and 16 % relative humidity (RH), relevant for gas monitoring in solid oxide fuel cell applications [52]. \u003cstrong\u003eFig. 12\u003c/strong\u003e illustrates dynamic responses of this second-order sensor to two volatiles at selected wavelengths and temperatures.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eAt temperature T1\u003c/em\u003e, the sensor responded only to H\u003csub\u003e2\u003c/sub\u003eO at certain wavelengths and did not respond to H\u003csub\u003e2\u003c/sub\u003e or H\u003csub\u003e2\u003c/sub\u003eO at other wavelengths. Response to H\u003csub\u003e2\u003c/sub\u003eO was always in one direction and linear with H\u003csub\u003e2\u003c/sub\u003eO vapor concentration. Response only to H\u003csub\u003e2\u003c/sub\u003eO and not to H\u003csub\u003e2\u003c/sub\u003e was due to the physical sorption of H\u003csub\u003e2\u003c/sub\u003eO vapor molecules to the nanostructure. The wavelength-dependence of this response was due to diversity of interactions of H\u003csub\u003e2\u003c/sub\u003eO vapor with different spatial regions of the nanostructure, related to corresponding spectral response differences [37, 52].\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eAt temperature T2\u003c/em\u003e, the sensor had a wavelength-dependent response to H\u003csub\u003e2\u003c/sub\u003e and H\u003csub\u003e2\u003c/sub\u003eO; it was increasing or decreasing as a function of H\u003csub\u003e2\u003c/sub\u003e and H\u003csub\u003e2\u003c/sub\u003eO concentrations. At some wavelengths, the response was barely noticeable. While at many wavelengths the sensor responded only to H\u003csub\u003e2\u003c/sub\u003e, at certain wavelengths, the small response to H\u003csub\u003e2\u003c/sub\u003eO was also noticed as an increase or decrease in sensor signal. Sorption of H\u003csub\u003e2\u003c/sub\u003e gas and H\u003csub\u003e2\u003c/sub\u003eO vapor molecules to the nanostructure occurred on different spatial regions of the nanostructure because of its design with spatially-distributed catalytic regions [37]. Such spatially diverse interactions with H\u003csub\u003e2\u003c/sub\u003e and H\u003csub\u003e2\u003c/sub\u003eO resulted in corresponding differences of spectral responses.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFig. 13\u003c/strong\u003e depicts contour 2-D plots of responses of the sensing element to H\u003csub\u003e2\u003c/sub\u003e gas and H\u003csub\u003e2\u003c/sub\u003eO vapor, with the first independent variable being wavelengths and the second independent variable being sensor temperature. For this illustrative visualization, the highest tested concentrations of volatiles were utilized, several wavelengths were selected (480, 560, 660, 700, and 965 nm) for the adequate two-volatile differentiation, and contour plots are presented with eight contours of the raw signal strength at each temperature. At different selected wavelengths and temperatures, diversity of response patterns for two volatiles was three-fold. \u003cem\u003eFirst\u003c/em\u003e, the number and shapes of the contours was related to the vapor type and sensor temperature. \u003cem\u003eSecond,\u003c/em\u003e at T1, H\u003csub\u003e2\u003c/sub\u003e had no response (no contours), while H\u003csub\u003e2\u003c/sub\u003eO vapor had responses (contours) at all selected wavelengths. \u003cem\u003eThird\u003c/em\u003e, at T2, H\u003csub\u003e2\u003c/sub\u003e had two contours of its response, while H\u003csub\u003e2\u003c/sub\u003eO vapor had no response.\u003c/p\u003e\n\u003cp\u003eTo visualize further responses of the sensor to all tested concentrations of H\u003csub\u003e2\u003c/sub\u003e gas and H\u003csub\u003e2\u003c/sub\u003eO vapor, we performed PCA using both temperatures and wavelengths selected in \u003cstrong\u003eFig. 12\u003c/strong\u003e. The PCA scores plot of the first two PCs is depicted in \u003cstrong\u003eFig. 14\u003c/strong\u003e. The first two PCs had contributions of PC1 = 73.40% and PC2 = 19.61%, capturing 93.01% of the total variance in the spectra. The third PC did not contribute to the two-volatile differentiation.\u003c/p\u003e"},{"header":"Conclusions","content":"\u003cp\u003eZero-order gas sensors have their established applications as alarms for industrial and residential safety when \u0026ldquo;pollution levels are high and when the compound of interest swamps others,\u0026rdquo; as stated in \u003cem\u003eNature Comments\u003c/em\u003e [54]. In detection of low analyte levels in diverse chemical backgrounds, zero-order gas sensors often have poor performance that cancels out their small size and low-cost advantages over traditional analytical instruments [8, 53\u0026ndash;55]. The effects of interferences make the data from zero-order sensors \u0026ldquo;essentially meaningless,\u0026rdquo; as stated in another \u003cem\u003eNature Comments\u003c/em\u003e [53]. As recently pointed out by the U.S. Environmental Protection Agency, zero-order sensors \u0026ldquo;have inherent limitations that are critical to understand before collecting and interpreting the data; many of these limitations require technological advances\u0026rdquo; [55].\u003c/p\u003e \u003cp\u003eThus, first- and second-order gas sensors are under development by numerous academic and industrial research teams to provide advances to meet contemporary requirements of air quality monitoring. It was shown [18, 32, 97, 98] that advancements to bring sensing systems to operate as first- and second-order gas sensors can be achieved without making fundamental changes to the existing hardware, ensuring compatibility with existing manufacturing practices and tools.\u003c/p\u003e \u003cp\u003eResults presented in this report should inspire new fundamental research and technological innovations in the areas of gas sensors, optics, materials science, nanofabrication, and data analytics. This strategic approach of expanding the dimensionality of sensor response has a huge societal impact by bringing the next generation sensors to perform on par with traditional orthogonal-output analytical instruments, but with a continuous real-time data stream and energy-efficient unobtrusive designs. Second-order sensors will break the trade-off between size, weight, power, and cost versus instrument performance, facilitating pervasive applications of next-generation gas sensors.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eR. A. Potyrailo dedicates this report to Otto Wolfbeis for his vision, influence, and inspiration for our interdisciplinary scientific community. R. A. Potyrailo is grateful for Otto Wolfbeis for fostering collaborations across various fields of optical sensors and for inspiring scientists worldwide through his innovative perspectives.\u003c/em\u003e\u0026nbsp; \u0026nbsp;Special thanks to librarians in the former USSR who provided photocopies of original papers by Otto Wolfbeis to R. A. Potyrailo. Special thanks to J. Sgambati for her thoughtful feedback and suggestions on the manuscript, which enhanced the clarity and readability of the described complex mathematical concepts for the interdisciplinary scientific community. Multi-gas sensors described here were encouraged and supported by V. Greanya at the Defense Advanced Research Projects Agency (DARPA) and S. Markovich and V. K. Venkataraman at U.S. Department of Energy (DOE). Sensing materials described in this report were inspired, characterized, and fabricated by creative scientists and engineers who coauthored original referenced contributions.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding Declaration:\u0026nbsp;\u003c/strong\u003eFunding was from GE, DARPA Contract W911NF-10-C-0069 and DOE Contracts DE-FE0027918 and DE-FE0031653. The conclusions in this report should not be interpreted to represent any determination or policy of DARPA and DOE and do not necessarily reflect the position or the policy of the US Government. Mentions of commercial products do not imply endorsement.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u0026nbsp;\u003c/strong\u003eRadislav A. Potyrailo: conceptualization, methodology, data collection, writing\u0026mdash;review and editing; Shiyao Shan: visualization, review, and editing; Baokai Cheng: data curation and investigation. \u0026nbsp;All authors have read and agreed to the published version of the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability\u0026nbsp;\u003c/strong\u003eThe data that support the plots within this paper and other findings of this study are available from the corresponding author upon reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDeclarations Conflict of interest\u0026nbsp;\u003c/strong\u003eThe authors declare no competing interests. \u0026nbsp;\u0026nbsp;\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eDes Voeux HA (1908) Discussion On Smoke Abatement. Brit Med J. 2(2487):549-54. \u003c/li\u003e\n\u003cli\u003eSmart JK (2000) History of Chemical and Biological Detectors, Alarms, and Warning Systems. U S Army Soldier and Biological Chemical Command.https://www.hsdl.org/?view\u0026amp;did=1670. \u003c/li\u003e\n\u003cli\u003eIhokura K, Watson J. Stannic Oxide Gas Sensor: Principles and Applications. Boca Raton, FL: CRC Press; 1994.\u003c/li\u003e\n\u003cli\u003eThe Plain English Guide to the Clean Air Act. (2024) US EPA.https://www.epa.gov/clean-air-act-overview/plain-english-guide-clean-air-act. \u003c/li\u003e\n\u003cli\u003eLi W-L, Kannan K (2024) Determination of legacy and emerging per-and polyfluoroalkyl substances (PFAS) in indoor and outdoor air. ACS ES\u0026amp;T Air. 1(9):1147-55. \u003c/li\u003e\n\u003cli\u003eBooksh KS, Kowalski BR (1994) Theory of analytical chemistry. Anal Chem. 66:782A-91A. \u003c/li\u003e\n\u003cli\u003eGrinberg N, Rodriguez S, editors. Ewing\u0026apos;s Analytical Instrumentation Handbook. 4th ed.: CRC Press; 2019.\u003c/li\u003e\n\u003cli\u003eRobinson JW, Frame EMS, Frame II GM. Instrumental analytical chemistry: an introduction. CRC Press; 2021.\u003c/li\u003e\n\u003cli\u003eJanata J. Principles of Chemical Sensors. 2 ed. New York, NY: Springer; 2009.\u003c/li\u003e\n\u003cli\u003eTaylor RF, Schultz JS, editors. Handbook of Chemical and Biological Sensors. Bristol, UK: IOP Publishing; 1996.\u003c/li\u003e\n\u003cli\u003eHierlemann A, Gutierrez-Osuna R (2008) Higher-Order Chemical Sensing. Chem Rev. 108:563-613. \u003c/li\u003e\n\u003cli\u003eJohnson KJ, Knapp AC (2020) Design Theory for Chemical Detection. Naval Research Laboratory, Washington, DC. https://apps.dtic.mil/sti/citations/AD1111377. \u003c/li\u003e\n\u003cli\u003ePotyrailo RA (2016) Multivariable sensors for ubiquitous monitoring of gases in the era of Internet of Things and Industrial Internet. Chem Rev. 116:11877\u0026ndash;923. \u003c/li\u003e\n\u003cli\u003eWohltjen H. A Journey: From Sensor Ideas to Sensor Products. Plenary talk at the 11th International Meeting on Chemical Sensors, University of Brescia, Italy, July 16-19, 2006. Elsevier Science; 2006.\u003c/li\u003e\n\u003cli\u003eMayer F (2010) Innovation for commercial Sensors: From stepping into an opportunity Gap to continuous Innovation. Proc Eng. 5:1-4.\u003c/li\u003e\n\u003cli\u003eButterfield D. Market review of gas sensors for industrial applications. National Physical Laboratory, Middlesex, UK, NPL Report ENV40; 2021. \u003c/li\u003e\n\u003cli\u003ePotyrailo RA. Enabling New Applications with Multivariable Chem/Bio Sensors: From Ideas to Products. Sensors Summit, San Diego, CA, December 10-12. 2018.\u003c/li\u003e\n\u003cli\u003eGeneral Electric Research Center: Gas sensing with dielectric excitation. (2021) Innovation Award 2021 Association for Sensor and Measurement Technology (AMA) https://www.ama-sensorik.de/en/science/ama-innovation-award/ama-innovation-award-2021/. \u003c/li\u003e\n\u003cli\u003ePotyrailo RA (2006) Polymeric Sensor Materials: Toward an Alliance of Combinatorial and Rational Design Tools? Angew Chem Int Ed. 45:702-23. \u003c/li\u003e\n\u003cli\u003ePotyrailo RA, Mirsky VM (2008) Combinatorial and High-Throughput Development of Sensing Materials: The First Ten Years. Chem Rev. 108:770-813. \u003c/li\u003e\n\u003cli\u003ePotyrailo RA, Ding Z, Butts MD, Genovese SE, Deng T (2008) Selective Chemical Sensing Using Structurally Colored Core-Shell Colloidal Crystal Films. IEEE Sensors J. 8:815-22. doi: 10.1109/JSEN.2008.923191.\u003c/li\u003e\n\u003cli\u003ePotyrailo RA, Naik RR (2013) Bionanomaterials and bioinspired nanostructures for selective vapor sensing. Annu Rev Mater Res. 43:307\u0026ndash;34. doi: DOI: 10.1146/annurev-matsci-071312-121710.\u003c/li\u003e\n\u003cli\u003eMusgrove AL, Cockerham A, Pazos JJ, Shahriar S, McMahon M, Touma JE, et al. (2024) Bio-inspired photonic and plasmonic systems for gas sensing: applications, fabrication, and analytical methods. J Opt Microsyst. 4(2):020902-. \u003c/li\u003e\n\u003cli\u003eAhmed MA, Hu D, Shi Y, Chen Y, Akhavan S, Yang Z (2025) Exploration of Biologically-Inspired Nanostructures: Review on the Sensing Potential and Technological Integration of the Morpho Butterfly Wing. Photonic Sens. 15(2):250227. \u003c/li\u003e\n\u003cli\u003ePotyrailo RA, Hobbs SE, Hieftje GM (1998) Optical waveguide sensors in analytical chemistry: today\u0026apos;s instrumentation, applications and future development trends. Fresenius\u0026apos; J Anal Chem. 362:349-73. \u003c/li\u003e\n\u003cli\u003ePersaud K, Dodd G (1982) Analysis of discrimination mechanisms in the mammalian olfactory system using a model nose. Nature. 299:352-5. \u003c/li\u003e\n\u003cli\u003eShulaker MM, Hills G, Park RS, Howe RT, Saraswat K, Wong HSP, et al. (2017) Three-dimensional integration of nanotechnologies for computing and data storage on a single chip. Nature. 547(7661):74-8. doi: 10.1038/nature22994.\u003c/li\u003e\n\u003cli\u003eWang C, Chen Z, Chan CLJ, Wan Za, Ye W, Tang W, et al. (2024) Biomimetic olfactory chips based on large-scale monolithically integrated nanotube sensor arrays. Nat Electron. 7:157\u0026ndash;67 \u003c/li\u003e\n\u003cli\u003ePotyrailo RA, Ghiradella H, Vertiatchikh A, Dovidenko K, Cournoyer JR, Olson E (2007) Morpho butterfly wing scales demonstrate highly selective vapour response. Nat Photonics. 1:123-8. \u003c/li\u003e\n\u003cli\u003ePotyrailo RA, Starkey T, Vukusic P, Ghiradella H, Vasudev M, Bunning T, et al. (2013) Discovery of the surface polarity gradient on iridescent Morpho butterfly scales reveals a mechanism of their selective vapor response. Proc Natl Acad Sci USA. 110:15567\u0026ndash;72. \u003c/li\u003e\n\u003cli\u003ePotyrailo RA, Bonam RK, Hartley JG, Starkey TA, Vukusic P, Vasudev M, et al. (2015) Towards outperforming conventional sensor arrays with fabricated individual photonic vapour sensors inspired by \u003cem\u003eMorpho \u003c/em\u003ebutterflies. Nat Commun. 6:7959. \u003c/li\u003e\n\u003cli\u003ePotyrailo RA, Go S, Sexton D, Li X, Alkadi N, Kolmakov A, et al. (2020) Extraordinary performance of semiconducting metal oxide gas sensors using dielectric excitation. Nat Electron. 3:280\u0026ndash;9. \u003c/li\u003e\n\u003cli\u003ePotyrailo RA, Burns A, Surman C, Lee DJ, McGinniss E (2012) Multivariable passive RFID vapor sensors: roll-to-roll fabrication on a flexible substrate. Analyst. 137:2777-81. \u003c/li\u003e\n\u003cli\u003eNagraj N, Slocik JM, Phillips DM, Kelley-Loughnane N, Naik RR, Potyrailo RA (2013) Selective sensing of vapors of similar dielectric constants using peptide-capped gold nanoparticles on individual multivariable transducers. Analyst. 138:4334-9. \u003c/li\u003e\n\u003cli\u003ePotyrailo RA (2017) Toward high value sensing: monolayer-protected metal nanoparticles in multivariable gas and vapor sensors. Chem Soc Rev. 46:5311-46. \u003c/li\u003e\n\u003cli\u003ePotyrailo RA, Karker N, Carpenter MA, Minnick A (2018) Multivariable bio-inspired photonic sensors for non-condensable gases. J Opt. 20:024006. \u003c/li\u003e\n\u003cli\u003ePotyrailo RA, Brewer J, Cheng B, Carpenter MA, Houlihan N, Kolmakov A (2020) Bio-inspired gas sensing: boosting performance with sensor optimization guided by \u0026ldquo;machine learning\u0026rdquo;. Faraday Discuss. 223:161-82. \u003c/li\u003e\n\u003cli\u003ePotyrailo RA (2024) Multi-Gas Differentiation Using an Individual Optical Fiber Sensor with a Chemically Modified Cladding. ISOCS/IEEE International Symposium on Olfaction and Electronic Nose (ISOEN).4092. \u003c/li\u003e\n\u003cli\u003eArasaradnam RP, Krishnamoorthy A, Hull MA, Wheatstone P, Kvasnik F, Persaud KC (2025) The Development and Optimisation of a Urinary Volatile Organic Compound Analytical Platform Using Gas Sensor Arrays for the Detection of Colorectal Cancer. Sensors. 25(3):599. \u003c/li\u003e\n\u003cli\u003eSachan A, Castro M, Feller J-F (2025) Volatolomics for Anticipated Diagnosis of Cancers with Chemoresistive Vapour Sensors: A Review. Chemosensors. 13(1):15. \u003c/li\u003e\n\u003cli\u003eLonergan MC, Severin EJ, Doleman BJ, Beaber SA, Grubbs RH, Lewis NS (1996) Array-Based Vapor Sensing Using Chemically Sensitive, Carbon Black-Polymer Resistors. Chem Mater. 8:2298-312. \u003c/li\u003e\n\u003cli\u003eLiu T, Wang Y, Wang H (2024) Open set Domain Adaptation for Electronic Nose Drift Compensation on Uncertain Category Data. IEEE Trans Instrum Meas. 73:2505514. \u003c/li\u003e\n\u003cli\u003eAbideen ZU, Arifeen WU, Bandara YND (2024) Emerging trends in metal oxide-based electronic noses for healthcare applications: a review. Nanoscale.Advance Article. \u003c/li\u003e\n\u003cli\u003eTanveer S, Ahmad MI, Khan T (2024) Technological progression associated with monitoring and management of indoor air pollution and associated health risks: A comprehensive review. Environ Qual Manage.Early View. \u003c/li\u003e\n\u003cli\u003eHan J, Li H, Cheng J, Ma X, Fu Y (2025) Advances in metal oxide semiconductor gas sensor arrays based on machine learning algorithms. J Mater Chem C. \u003c/li\u003e\n\u003cli\u003eSch\u0026uuml;tze A, Sauerwald T. Dynamic operation of semiconductor sensors. Semiconductor Gas Sensors. Elsevier; 2020. p. 385-412.\u003c/li\u003e\n\u003cli\u003ePotyrailo RA, St-Pierre R, Crowder J, Scherer B, Cheng B. Boosting stability of electronic multi-gas sensors. IEEE SENSORS, Dallas, TX, Oct 30 - Nov 2, Paper 2385; 2022.\u003c/li\u003e\n\u003cli\u003ePotyrailo RA, Scherer B, Brewer J, Ruffalo R. Boosting stability of photonic multi-gas sensors. IEEE SENSORS, Dallas, TX, Oct 30 - Nov 2, Paper 2381; 2022.\u003c/li\u003e\n\u003cli\u003ePiszter G, Kert\u0026eacute;sz K, B\u0026aacute;lint Z, Bir\u0026oacute; LP (2020) Stability and Selective Vapor Sensing of Structurally Colored Lepidopteran Wings Under Humid Conditions. Sensors. 20(11):3258. \u003c/li\u003e\n\u003cli\u003ePotyrailo RA. Tutorial: Next generation gas sensors: surprising advantages over last-century designs drive societal impact. IEEE International Symposium on Olfaction and Electronic Nose (ISOEN), Grapevine, TX, May 12-15; 2024.\u003c/li\u003e\n\u003cli\u003eNicolet A, Zolla F (2009) Cloaking with curved spaces. Science. 323(5910):46-7. \u003c/li\u003e\n\u003cli\u003ePotyrailo RA, Brewer J, Scherer B, Srivastava V, Nayeri M, Henderson C, et al. (2019) Multi-gas sensors for enhanced reliability of SOFC operation. ECS Transactions. 91:319-28. \u003c/li\u003e\n\u003cli\u003eAusten K (2015) Pollution patrol. Nature. 517:136-8. \u003c/li\u003e\n\u003cli\u003eLewis A, Edwards P (2016) Validate personal air-pollution sensors. Nature. 535:29-31. \u003c/li\u003e\n\u003cli\u003eBarkjohn KK, Clements A, Mocka C, Barrette C, Bittner A, Champion W, et al. (2024) Air Quality Sensor Experts Convene: Current Quality Assurance Considerations for Credible Data. ACS ES\u0026amp;T Air. 1(10):1203\u0026ndash;14. \u003c/li\u003e\n\u003cli\u003eThe Nobel Foundation.https://www.nobelprize.org/prizes/lists/all-nobel-prizes/. \u003c/li\u003e\n\u003cli\u003eWu H-L, Wang T, Yu R-Q (2020) Recent advances in chemical multi-way calibration with second-order or higher-order advantages: Multilinear models, algorithms, related issues and applications. Trends Anal Chem. 130:115954. \u003c/li\u003e\n\u003cli\u003eAlcaraz MR, Monago-Mara\u0026ntilde;a O, Goicoechea HC, de la Pe\u0026ntilde;a AM. Recent applications of third-order/four-way and fourth-order/five-way data analysis. Data Handling in Science and Technology. Elsevier; 2024. p. 337-62.\u003c/li\u003e\n\u003cli\u003eBussey III RO. Uses of Portable Gas Chromatography Mass Spectrometers. Novel Aspects of Gas Chromatography and Chemometrics. IntechOpen; 2022.\u003c/li\u003e\n\u003cli\u003eSmith ME, Westbrook E, Stastny AL, Streicher RP, Elliott MG (2024) Method development for on‑site monitoring of volatile organic compounds via portable TD‑GC-MS: evaluation of the analytical performances of HAPSITE\u0026reg; ER instrumentation and thermal desorption sampling media. J Environ Anal Chem. 104(17):5369-86. \u003c/li\u003e\n\u003cli\u003eLamy E, Roquencourt C, Zhou B, Salvator H, Moine P, Annane D, et al. (2024) Combination of real-time and hyphenated mass spectrometry for improved characterisation of exhaled breath biomarkers in clinical research. Anal Bioanal Chem. 416(22):4929-39. \u003c/li\u003e\n\u003cli\u003eGilbert N, Mewis RE, Sutcliffe OB (2020) Classification of fentanyl analogues through principal component analysis (PCA) and hierarchical clustering of GC\u0026ndash;MS data. Forensic Chem. 21:100287. \u003c/li\u003e\n\u003cli\u003eLee LC, Jemain AA (2021) On overview of PCA application strategy in processing high dimensionality forensic data. Microchem J. 169:106608. \u003c/li\u003e\n\u003cli\u003eDombrowski A, Le D, Lurie IS (2025) GC-VUV spectroscopy of synthetic cannabinoid isomers, diastereomers and homologs: Increasing differentiation by derivative spectral processing. Forensic Chem. 42:100635. \u003c/li\u003e\n\u003cli\u003eLippitt W, Carlson NE, Arbet J, Fingerlin TE, Maier LA, Kechris K (2024) Limitations of clustering with PCA and correlated noise. J Stat Comput Simul.1-29. \u003c/li\u003e\n\u003cli\u003eVukusic P, Sambles JR, Lawrence CR, Wootton RJ (1999) Quantified interference and diffraction in single \u003cem\u003eMorpho \u003c/em\u003ebutterfly scales. Proc R Soc Lond B. 266:1403-11. \u003c/li\u003e\n\u003cli\u003eHoulihan N, Karker N, Potyrailo RA, Carpenter M (2018) High sensitivity plasmonic sensing of hydrogen over a broad dynamic range using catalytic Au-CeO2 thin film nanocomposites. ACS Sens. 3:2684-92. \u003c/li\u003e\n\u003cli\u003eMei H, Peng J, Wang T, Zhou T, Zhao H, Zhang T, et al. (2024) Overcoming the Limits of Cross-Sensitivity: Pattern Recognition Methods for Chemiresistive Gas Sensor Array. Nanomicro Lett. 16(1):269. \u003c/li\u003e\n\u003cli\u003eMeier M, Kittle JD, Yee XC (2022) Supervised dimension reduction for optical vapor sensing. RSC Adv. 12(16):9579-86. \u003c/li\u003e\n\u003cli\u003ePotyrailo RA, Surman C, Nagraj NN, Burns A (2011) Materials and Transducers Toward Selective Wireless Gas Sensing. Chem Rev. 111:7315\u0026ndash;54. \u003c/li\u003e\n\u003cli\u003eZellers ET, Park J, Hsu T, Groves WA (1998) Establishing a limit of recognition for a vapor sensor array. Anal Chem. 70:4191-201. \u003c/li\u003e\n\u003cli\u003eBorn M, Wolf E. Principles of optics: electromagnetic theory of propagation, interference and diffraction of light. 7 (expanded) 60th anniversary ed. Cambridge, UK: Cambridge University Press; 2019.\u003c/li\u003e\n\u003cli\u003eNagarajan V, Thayumanavan A (2018) CdFe 2 O 4 films for electroresistive detection of ethanol and formaldehyde vapors. Microchim Acta. 185:1-9. \u003c/li\u003e\n\u003cli\u003eKovalska E, Antonatos N, Luxa J, Sofer Z (2021) Edge-hydrogenated germanene by electrochemical decalcification-exfoliation of CaGe2: Germanene-enabled vapor sensor. ACS Nano. 15(10):16709-18. \u003c/li\u003e\n\u003cli\u003eZhang R, Wang Y, Li J, Zhao H, Wang Y, Zhou Y (2022) Mesoporous cellulose nanofibers-interlaced PEDOT: PSS hybrids for chemiresistive ammonia detection. Microchim Acta. 189(8):308. \u003c/li\u003e\n\u003cli\u003eJi H, Mi C, Yuan Z, Liu Y, Zhu H, Meng F (2022) Multicomponent gas detection method via dynamic temperature modulation measurements based on semiconductor gas sensor. IEEE Trans Ind Electron. 70(6):6395-404. \u003c/li\u003e\n\u003cli\u003eTong L, Wan J, Xiao K, Liu J, Ma J, Guo X, et al. (2023) Heterogeneous complementary field-effect transistors based on silicon and molybdenum disulfide. Nat Electron. 6(1):37-44. \u003c/li\u003e\n\u003cli\u003eSingh S, Shin KY, Moon S, Kim SS, Kim HW (2024) Phase-engineered MoSe2/CeO2 composites for room-temperature gas sensing with a drastic discrimination of NH3 and TEA gases. ACS Sens. 9(8):3994-4006. \u003c/li\u003e\n\u003cli\u003eDiab R, Boltaev G, Kaid MM, Fawad A, El-Kaderi HM, Al-Sayah MH, et al. (2025) Fabrication of heteroatom-doped graphene-porous organic polymer hybrid materials via femtosecond laser writing and their application in VOCs sensing. Sci Rep. 15(1):3682. \u003c/li\u003e\n\u003cli\u003eBhagat B, Paine S, Manna B, Choudhury A, Mukherjee K (2025) Kinetic Modelling and Machine Learning Approaches on the Conductance Transients of Zn0. 5Ni0. 5Fe2O4 Sensors for Addressing Cross‐sensitivity towards Ethanol and Acetone Vapors. Chem Eur J e202404111. \u003c/li\u003e\n\u003cli\u003eM\u0026uuml;ller R, Lange E (1986) Multidimensional sensor for gas analysis. Sens Actuators. 9(1):39-48. \u003c/li\u003e\n\u003cli\u003eSingh P, Yadava R (2011) Effect of film thickness and viscoelasticity on separability of vapour classes by wavelet and principal component analyses of polymer-coated surface acoustic wave sensor transients. Meas Sci Technol. 22(2):025202. \u003c/li\u003e\n\u003cli\u003eLuo Z, Weng Z, Shen Q, An S, He J, Fu B, et al. (2020) Vapor detection through dynamic process of molecule desorption from butterfly wings. Pure Appl Chem. 92(2):223-32. \u003c/li\u003e\n\u003cli\u003eBrandt S, Pavlichenko I, Shneidman AV, Patel H, Tripp A, Wong TS, et al. (2023) Nonequilibrium sensing of volatile compounds using active and passive analyte delivery. Proc Natl Acad Sci USA. 120(31):e2303928120. \u003c/li\u003e\n\u003cli\u003eFriedrich RW, Laurent G (2001) Dynamic optimization of odor representations by slow temporal patterning of mitral cell activity. Science. 291(5505):889-94. doi: 10.1126/science.291.5505.889.\u003c/li\u003e\n\u003cli\u003eAckels T, Erskine A, Dasgupta D, Marin AC, Warner T, Tootoonian S, et al. (2021) Fast odour dynamics are encoded in the olfactory system and guide behaviour. Nature. 593(7860):558-63. \u003c/li\u003e\n\u003cli\u003evan den Broek J, Abegg S, Pratsinis SE, G\u0026uuml;ntner AT (2019) Highly selective detection of methanol over ethanol by a handheld gas sensor. Nat Commun. 10(1). doi: 10.1038/s41467-019-12223-4.\u003c/li\u003e\n\u003cli\u003eSpencer TL, Clark A, Fonollosa J, Virot E, Hu DL (2021) Sniffing speeds up chemical detection by controlling air-flows near sensors. Nat Commun. 12(1):1232. \u003c/li\u003e\n\u003cli\u003eWeimar U, G\u0026ouml;pel W (1998) Chemical imaging: II. Trends in practical multiparameter sensor systems. Sens Actuators, B. 52:143\u0026ndash;61. \u003c/li\u003e\n\u003cli\u003eSears WM, Colbow K, Consadori F (1989) Algorithms to improve the selectivity of thermally-cycled tin oxide gas sensors. Sens Actuators. 19(4):333-49. doi: 10.1016/0250-6874(89)87084-2.\u003c/li\u003e\n\u003cli\u003eSemancik S, Cavicchi R (1998) Kinetically controlled chemical sensing using micromachined structures. Acc Chem Res. 31:279-87. \u003c/li\u003e\n\u003cli\u003eBur C, Reimann P, Andersson M, Schutze A, Spetz AL (2012) Increasing the selectivity of Pt-gate SiC field effect gas sensors by dynamic temperature modulation. IEEE Sens J. 12(6):1906-13. doi: 10.1109/JSEN.2011.2179645.\u003c/li\u003e\n\u003cli\u003eBur C, Bastuck M, Puglisi D, Sch\u0026uuml;tze A, Lloyd Spetz A, Andersson M (2015) Discrimination and quantification of volatile organic compounds in the ppb-range with gas sensitive SiC-FETs using multivariate statistics. Sens Actuators, B. 214:225\u0026ndash;33. doi: 10.1016/j.snb.2015.03.016.\u003c/li\u003e\n\u003cli\u003eSingh H, Bhaskerraj V, Kumar J, Mittal U, Mishra M, Durani F, et al. (2014) SAW E-nose using single sensor with temperature modulation. IEEE 9th International Conference on Industrial and Information Systems (ICIIS).1-5. \u003c/li\u003e\n\u003cli\u003eZeng G, Wu C, Chang Y, Zhou C, Chen B, Zhang M, et al. (2019) Detection and Discrimination of Volatile Organic Compounds using a Single Film Bulk Acoustic Wave Resonator with Temperature Modulation as a Multiparameter Virtual Sensor Array. ACS Sens. 4(6):1524-33. \u003c/li\u003e\n\u003cli\u003eZhang J, Goel N, Bart S, Tadigadapa S. Identification of Volatile Organic Compound Mixtures with a Micromachined Quartz Resonator through Temperature Modulation with On-Chip Integrated Heater. 2024 IEEE International Symposium on Olfaction and Electronic Nose (ISOEN): IEEE; 2024. p. 1-4.\u003c/li\u003e\n\u003cli\u003eSch\u0026uuml;tze A, Reimann G, Darsch S, Sch\u0026uuml;ler M, Gillen T (2011) Electrical Impedance Spectroscopy Combined with Temperature Cycled Operation for Semiconductor Gas Sensors. Proceedings SENSOR 2011.40-5. \u003c/li\u003e\n\u003cli\u003eZampolli S, Elmi I, Bruschi P, Ria A, Magliocca F, Vitelli M, et al. (2025) An ASIC-based system-in-package MEMS gas sensor with impedance spectroscopy readout and AI-enabled identification capabilities. Sens Actuators, B. 424:136924. \u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"microchimica-acta","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"miac","sideBox":"Learn more about [Microchimica Acta](https://link.springer.com/journal/604)","snPcode":"604","submissionUrl":"https://submission.springernature.com/new-submission/604/3","title":"Microchimica Acta","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"photonic nanostructure, multi-gas sensing, second-order individual sensors, traditional analytical instruments","lastPublishedDoi":"10.21203/rs.3.rs-6234291/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6234291/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Simple, single-output (zero-order) gas sensors have established applications for industrial and residential safety. However, in detection of low levels of analyte gases in diverse chemical backgrounds, zero-order gas sensors often have poor performance, cancelling out their small size and low-cost advantages over traditional analytical instruments. Thus, gas sensors with one and two independent variables (first- and second-order sensors) are under development to meet contemporary gas monitoring needs for environmental, industrial, medical, and homeland security applications. Here, we show three designs of photonic nanostructured sensors as second-order analytical instruments. In all three designs, the first independent variable in sensor response is optical wavelength across the visible spectral range. The second independent variable is either (i) illumination angle, (ii) light polarization, or (iii) operation temperature of the sensor. These illustrative designs of second-order gas sensors bring mathematical understanding to boost the multi-gas resolution achieved by individual material-based sensors through the transducer, material design, appropriate excitation conditions, and data analytics. Results presented in this report should inspire new fundamental research and technological innovations in the areas of gas sensors, optics, materials science, nanofabrication, and data analytics. The strategic approach of expanding the dimensionality of sensor response described in this study will enable next generation sensors to perform on par with traditional orthogonal-output analytical instruments, but with a continuous real-time data stream and unobtrusive, energy-efficient designs. ","manuscriptTitle":"Individual Optical Multi-Gas Sensors as Next Generation Second-Order Unobtrusive and Continuous Operation Analytical Instruments","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-04-18 13:58:18","doi":"10.21203/rs.3.rs-6234291/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-05-03T15:49:30+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-04-24T14:22:08+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-04-23T15:31:50+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"328999429886841326573204731731582711778","date":"2025-04-15T11:23:06+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"220067090124421256955818526109103297842","date":"2025-04-11T14:32:43+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"121862764179504978305419795612669137399","date":"2025-03-31T10:12:21+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-03-30T17:00:06+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-03-21T01:39:01+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-03-21T01:37:55+00:00","index":"","fulltext":""},{"type":"submitted","content":"Microchimica Acta","date":"2025-03-15T18:01:54+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"microchimica-acta","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"miac","sideBox":"Learn more about [Microchimica Acta](https://link.springer.com/journal/604)","snPcode":"604","submissionUrl":"https://submission.springernature.com/new-submission/604/3","title":"Microchimica Acta","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"2022cf0f-669f-4242-b5ad-2ae6c79365ab","owner":[],"postedDate":"April 18th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-07-14T16:00:39+00:00","versionOfRecord":{"articleIdentity":"rs-6234291","link":"https://doi.org/10.1007/s00604-025-07294-8","journal":{"identity":"microchimica-acta","isVorOnly":false,"title":"Microchimica Acta"},"publishedOn":"2025-07-09 15:57:20","publishedOnDateReadable":"July 9th, 2025"},"versionCreatedAt":"2025-04-18 13:58:18","video":"","vorDoi":"10.1007/s00604-025-07294-8","vorDoiUrl":"https://doi.org/10.1007/s00604-025-07294-8","workflowStages":[]},"version":"v1","identity":"rs-6234291","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6234291","identity":"rs-6234291","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}