Synergistic Integration of Frequency-Dependent Impedance and Machine Learning in Semiconductor Metal Oxide-Based Breath Sensors for High-Performance Gas Discrimination.

In this research, we propose a humidity compensation method using a deep learning network for minimizing the effect of humidity on the semiconductor metal oxide (SMO) gas sensors. The SMO gas sensors were fabricated by depositing In <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> O <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</inf> on a suspended microheater platform, and the gas tests were conducted under various humidity conditions using NO <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> gas. The gas sensing data and humidity data were simultaneously used as input data of the deep learning network for real-time humidity compensation. Through the proposed method, we successfully improved the accuracy of predicting the concentration of NO <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> gas in different humidity conditions.

Semiconductor metal oxide gas sensors: A review

Electronic noses or array sensors are very popular in the last decades because of their ability to avoid the cross-sensitivity issue in semiconductor metal oxide (SMO) gas sensors. The identification and discrimination of toxic gases have a significant role in industrial applications. This work encompasses the classification of carbon monoxide (CO) and methane (CH4) toxic gases using a gas sensor array. Classification algorithm based on artificial neural network (ANN) with one hidden layer is used for identifying the gas type from the gas mixture. This metal oxide gas sensor array is built with six SMO gas sensors, which are sensitive to several types of gases. The ANN model ensures a training accuracy of 94.57% and a validation accuracy of 93.33%. For practical applications, the gas concentration is randomly assigned in the training stage. Neural network-based classification algorithm provides better performance in identifying the type of gas.

Sensitive detection of acetone and hydrogen sulfide levels in exhaled human breath, serving as breath markers for some diseases such as diabetes and halitosis, may offer useful information for early diagnosis of these diseases. Exhaled breath analyzers using semiconductor metal oxide (SMO) gas sensors have attracted much attention because they offer low cost fabrication, miniaturization, and integration into portable devices for noninvasive medical diagnosis. However, SMO gas sensors often display cross sensitivity to interfering species. Therefore, selective real-time detection of specific disease markers is a major challenge that must be overcome to ensure reliable breath analysis. In this work, we report on highly sensitive and selective acetone and hydrogen sulfide detection achieved by sensitizing electrospun SnO2 nanofibers with reduced graphene oxide (RGO) nanosheets. SnO2 nanofibers mixed with a small amount (0.01 wt %) of RGO nanosheets exhibited sensitive response to hydrogen sulfide (Rair/Rgas = 34 at 5 ppm) at 200 °C, whereas sensitive acetone detection (Rair/Rgas = 10 at 5 ppm) was achieved by increasing the RGO loading to 5 wt % and raising the operation temperature to 350 °C. The detection limit of these sensors is predicted to be as low as 1 ppm for hydrogen sulfide and 100 ppb for acetone, respectively. These concentrations are much lower than in the exhaled breath of healthy people. This demonstrates that optimization of the RGO loading and the operation temperature of RGO-SnO2 nanocomposite gas sensors enables highly sensitive and selective detection of breath markers for the diagnosis of diabetes and halitosis.

Chemical and/or physical sensors are employed at domestic and industrial sites with the aim to perform a variety of functions, such as prediction, forecasting, prognostics, remaining useful life estimation, anomaly detection, and trend analysis. The sensing information contain the analog signals originating from gas-generated signals, pressure, shear, strain, torsion, temperature, humidity, illumination, electromagnetic radiation, sound, and vibration. Chemical gas sensors can be categorized mainly as semiconductor metal oxide (SMO), electrochemical, or photo-ionization detection (PID) sensors. SMO gas sensors are based on adsorption and desorption of gas molecules onto/from semiconductor metal oxides and however, suffer from the low selectivity to diverse gas species, i.e., the mutually interacting features.Here, we report a high-performance machine learning strategy for gas detection/discrimination against harmful gas combination and adopt the synergistic integration of supervised and unsupervised learning by exploiting a SMO gas sensor array. A 4 SMO sensor array was constructed for detecting carbon monoxide and ethyl alcohol (C2H5OH) mixtures using 15 different gas combinations. Gas detection/discrimination performance was probed in terms of different numbers of gas sensor integrated as an array and machine learning algorithms. Unsupervised K-Means clustering was successfully applied to the rational identification of the similarity features of targeted gases among 4 different groups, which are listed as matrix gas, two single-component gases, and one two-gas by employing only unlabeled voltage-based gas sensing information. Detailed classification was performed through a multitude of supervised algorithms, i.e., 2-layer artificial neural networks (ANNs), 4-layer deep neural networks (DNNs), 1-dimensional convolutional neural networks (1D CNNs), and 2-dimensional convolutional neural networks (2D CNNs). The numerical-based DNNs and image-based CNNs are evaluated to be excellent approaches for gas detection/classification, as corroborated by the highest accuracy and lowest loss parameters.Through the systematic investigation on the influence of the number of sensors on the arrayed gas sensor system, the applicability of machine learning methodology to an arrayed SMO gas sensor system is justified by the four unique features, i.e., i) a data augmentation methodology, ii) machine learning approach of combining K-means clustering and neural networks, and iii) a systematic approach to optimized sensor combinations, potentially recommending the minimized sensor systems based on chemical gas sensors against harmful gas species. Even two SMO sensor combinations are shown to be highly effective in performing gas discrimination against harmful gas species assisted through either numeric-based DNNs or image-based 1D CNNs, overcoming the simple clustering proposed through the unsupervised K-means clustering.

IntroductionThe enormous market demands for gas sensors to detect and monitor volatile organic compounds (VOCs) and toxic gases in environmental protection, medical treatment and safety requirements of industry and daily life have motivated enormous awareness in the fabrication of high performance gas sensing materials. Generally, VOCs are considered as low boiling point carbon-comprising materials that can evaporate simply at room temperature and contribute in atmospheric photochemical reactions, such as photochemistry smog and ozone formation. Besides, controlling and detection of these VOCs, such as methanol, ethanol, propanol and acetone is significant because they are considered as widespread concerns for common pollutants in indoor/outdoor air quality and for testing alcohol levels of drivers [1]. In comparison to other VOCS, propanol is less toxic, nonetheless its usage for numerous daily applications, for example hand sanitizer, cosmetics, may be imagined to be polluting the breathing air. Besides, Iso-propanol was presented to behave as an aesthetic and central nervous system sedative, leading in symptoms that can frighten the mental ability of the individual [2].Yet, the fabrication of the anticipated gas sensor with superior sensitivity, low detection limit, and exceptional long-term stability is still a significant research issue. In addition, the high operating temperatures and poor selectivity often limits the application and commercialization of semiconductor metal oxide (SMO) gas sensors. Thus, this work is directed to improve the selective, sensitivity, stability and to reduce the operating temperatures of the metal oxides based sensors, by exploiting SnO2 hollowspheres loaded with various NiO nanoparticles contents and SnO2/NiO loaded with various Au nanoparticles contents synthesized using a simple hydrothermal method. Our findings showed improved stability upon loading SnO2 hollowspheres with various NiO contents. Amongst the various NiO contents, the 0.01 wt.% NiO loaded SnO2 demonstrated higher response and sensitivity towards propanol (C3H7OH) in dry air and in the presence of 40 and 60 % RH at an operating temperature of 150 °C. While the as-fabricated SnO2/NiO/Au (2.5 wt.%) based sensor exhibited a response that is more than 2 times higher in comparison to that of SnO2/NiO (0.01wt. %) and lower operating temperature 75 °C towards ethanol (C2H5OH) and C3H7OH. The long-term stability analyses established that both the fabricated SnO2/NiO (0.01 wt.%) and SnO2/NiO/Au (2.5 wt. %) sensors were very stable towards C3H7OH and C2H5OH after a month in the presence of 40% RH. These findings showed that the current sensors can be employed for detecting C3H7OH and C2H5OH in a vastly sensitive and selective way with insignificant interference from ambient humidity.MethodPorous hollowspheres were synthesized according to the following procedure reported in ref. [3]. While the NiO loaded SnO2 hollowspheres were fabricated using the as-prepared hollowspheres as templates. Gold (Au) loaded NiO/SnO2 hollowspheres were prepared using NiO/SnO2 hollowspheres as template as reported elsewhere.The gas sensing properties of the Au-loaded SnO2/NiO were measured using the KSGAS6S gas testing system (KENOSISTEC, Italy). Sensing preparations and testing were performed according to the procedure in ref [3, 4].Results and ConclusionsTo investigate the operational temperature influence on the gas sensing characteristics, the sensing layers were analysed at various operating temperatures (75 to 225 °C). The variations in the gas sensing resistance (Ra) of SnO2/NiO loaded with various NiO contents and SnO2/NiO loaded with various Au contents as a function of operating temperature. The Ra of the all sensing materials decreased with an increase in operating temperature, except that that of low Ni loading. A decrease could be when the operational temperature increases more electrons from valence band of the sensing layer will jump to conduction band and accordingly more electrons are available for carrying current, resulting in a decrease of Ra and this is accordance with the previous studies. While the increase in Ra values for SnO2/NiO can be justified by the formation of nanoscale p/n junctions between the NiO and SnO2 phases.We further explored the stability of the SnO2/NiO (0.01 wt.%)-and SnO2/NiO (0.01 wt.%)/Au best performing based sensors in terms of repeatability and long-term stability measurements in dry air and in the presence of relative humidity (RH). The results exhibited the repeatability behavior of the SnO2/NiO (0.01 wt.%)-based sensor for eight successive cycles at 40 ppm C3H7OH for a month in the presence of 40% RH. The fresh sensor in day one and that tested after 30 days displayed a flawless repeatability of eight successive cycles towards 40 ppm C3H7OH. After 7 to 30 days, only minimal drift on the repeatability performance was witnessed, while the response increased by 8%, validating that our sensor is very stable. Same behavior was observed for the SnO2/NiO (0.01 wt.%)/Au(2.5 wt.%) based sensor. Thus, these findings corroborate that the current sensors are very stable when either exposed to dry air or in real conditions after long exposure to C3H7OH and C2H5OH.

Fabrication of a humidity-resistant formaldehyde gas sensor through layering a molecular sieve on 3D ordered macroporous SnO2 decorated with Au nanoparticles

The purpose of this exploratory study was to determine whether liver dysfunction can be generally classified using a wearable electronic nose based on semiconductor metal oxide (MOx) gas sensors, and whether the extent of this dysfunction can be quantified. MOx gas sensors are attractive because of their simplicity, high sensitivity, low cost, and stability. A total of 30 participants were enrolled, 10 of them being healthy controls, 10 with compensated cirrhosis, and 10 with decompensated cirrhosis. We used three sensor modules with a total of nine different MOx layers to detect reducible, easily oxidizable, and highly oxidizable gases. The complex data analysis in the time and non-linear dynamics domains is based on the extraction of 10 features from the sensor time series of the extracted breathing gas measurement cycles. The sensitivity, specificity, and accuracy for distinguishing compensated and decompensated cirrhosis patients from healthy controls was 1.00. Patients with compensated and decompensated cirrhosis could be separated with a sensitivity of 0.90 (correctly classified decompensated cirrhosis), a specificity of 1.00 (correctly classified compensated cirrhosis), and an accuracy of 0.95. Our wearable, non-invasive system provides a promising tool to detect liver dysfunctions on a functional basis. Therefore, it could provide valuable support in preoperative examinations or for initial diagnosis by the general practitioner, as it provides non-invasive, rapid, and cost-effective analysis results.

IntroductionThe enormous market demands for gas sensors to detect and monitor volatile organic compounds (VOCs) and toxic gases in environmental protection, medical treatment and safety requirements of industry and daily life have motivated enormous awareness in the fabrication of high performance gas sensing materials. Generally, VOCs are considered as low boiling point carbon-comprising materials that can evaporate simply at room temperature and contribute in atmospheric photochemical reactions, such as photochemistry smog and ozone formation. Besides, controlling and detection of these VOCs, such as methanol, ethanol, propanol and acetone is significant because they are considered as widespread concerns for common pollutants in indoor/outdoor air quality and for testing alcohol levels of drivers [1]. In comparison to other VOCS, propanol is less toxic, nonetheless its usage for numerous daily applications, for example hand sanitizer, cosmetics, maybe imagined to be polluting the breathing air. Besides, Iso-propanol was presented to behave as an aesthetic and central nervous system sedative, leading in symptoms that can frighten the mental ability of the individual [2].Yet, the fabrication of the anticipated gas sensor with superior sensitivity, low detection limit, and exceptional long-term stability is still a significant research issue. In addition, the high operating temperatures and poor selectivity often limits the application and commercialization of semiconductor metal oxide (SMO) gas sensors. Thus, this work is directed to improve the selective, sensitivity, stability and reduce the operating temperatures of the metal oxides based sensors, by exploiting SnO2 hollowspheres loaded with various NiO nanoparticles contents and SnO2/NiO loaded with various Au nanoparticles contents synthesized using a simple hydrothermal method. Our findings showed improved stability upon loading SnO2 hollowspheres with various NiO contents. Amongst the various NiO contents, the 0.01 wt.% NiO loaded SnO2 demonstrated higher response and sensitivity towards propanol (C3H7OH) in dry air and in the presence of 40 and 60 % RH at an operating temperature of 150 °C. While the as-fabricated SnO2/NiO/Au (2.5 wt.%) based sensor exhibited a response that is more than 2 times higher in comparison to that of SnO2/NiO (0.01wt. %) and lower operating temperature 75 °C towards ethanol (C2H5OH) and C3H7OH. The long-term stability analyses established that both the fabricated SnO2/NiO (0.01 wt.%) and SnO2/NiO/Au (2.5 wt. %) sensors were very stable towards C3H7OH and C2H5OH after a month in the presence of 40% RH. These findings showed that the current sensors can be employed for detecting C3H7OH and C2H5OH in a vastly sensitive and selective way with insignificant interference from ambient humidity.MethodPorous hollowspheres were synthesized according to the following procedure reported in ref. [3]. While the NiO loaded SnO2 hollowspheres were fabricated using the as-prepared hollowspheres as templates. Gold (Au) loaded NiO/SnO2 hollowspheres were prepared using NiO/SnO2 hollowspheres as template. Detailed synthesis preparation can be found in ref. [4].Sensing preparations and testing of the sensors towards various reducing gases (e.g. CO, CH4, and NH3), volatile organic compounds (C2H5OH and C3H7OH), and an oxidizing gas (NO2) in a background of synthetic dry air were performed according to the procedure in ref [3,4].Results and ConclusionsTo investigate the operational temperature influence on the gas sensing characteristics, the sensing layers were analysed at various operating temperatures (75 to 225 °C). The variations in the gas sensing resistance (Ra) of the SnO2 loaded with various NiO contents and SnO2/NiO loaded with various Au contents as a function of operating temperature were presented. The Ra of the all sensing materials decreased with an increase in operating temperature, except that that of low Ni loading (see Fig. 1A). A decrease could be that when the operational temperature increases more electrons from valence band of the sensing layer will jump to conduction band and accordingly more electrons are available for carrying current, resulting in a decrease of Ra [4]. While the increase in Ra values for SnO2/NiO can be justified by the formation of nanoscale p/n junctions between the NiO and SnO2 phases.We further explored the stability of the SnO2/NiO (0.01 wt.%)-and SnO2/NiO (0.01 wt.%)/Au best performing based sensors in terms of repeatability and long-term stability measurements in dry air and in the presence of relative humidity (RH). The fresh sensor in day one and that tested after 30 days displayed a flawless repeatability of eight successive cycles towards 40 ppm C3H7OH (see Fig. 1B). Based on the findings, after 7 to 30 days, only minimal drift on the repeatability performance was witnessed, while the response increased by almost 8%, validating that our sensor is very stable. Same behavior was observed for the SnO2/NiO (0.01 wt.%)/Au(2.5 wt.%) based sensor [4]. Thus, these findings corroborated that the current sensors are very stable when either exposed to dry air or in real conditions after long exposure to C3H7OH and C2H5OH.

Semiconducting metal oxide (SMO) gas sensors were designed, fabricated, and characterized in terms of their sensing capability and the thermo-mechanical behavior of the micro-hotplate. The sensors demonstrate high sensitivity at low concentrations of volatile organic compounds (VOCs) at a low power consumption of 10.5 mW. In addition, the sensors realize fast response and recovery times of 20 s and 2.3 min, respectively. To further improve the baseline stability and sensing response characteristics at low power consumption, a novel sensor is conceived of and proposed. Tantalum aluminum (TaAl) is used as a microheater, whereas Pt-doped SnO2 is used as a thin film sensing layer. Both layers were deposited on top of a porous silicon nitride membrane. In this paper, two designs are characterized by simulations and experimental measurements, and the results are comparatively reported. Simultaneously, the impact of a heat pulsing mode and rubber smartphone cases on the sensing performance of the gas sensor are highlighted.

During the past few decades, semiconductor metal oxide (SMO) gas sensors have become a prime technology in several domestic, commercial, and industrial gas sensing. The semiconductor properties of zinc oxide along with its dopant remain to be trapped fully in its application as gas sensor. With the advent of nanotechnology, miniaturization and high sensitivity happens to be a key issue in sensor fabrication. Most of the SMO gas sensors fabricated by nanotechnology process operate at high temperature. This paper gives a new insight to hydrogen gas sensor characteristics, by reducing the operating temperature of hydrogen (H 2 ) sensor, fabricated from the nano particle of manganese doped zinic oxide(ZnO), synthesized by chemical precipitation method.

Chemiresistive gas sensors based on electrospun semiconductor metal oxides: A review

The simultaneous operation of multiple different semiconducting metal oxide (MOX) gas sensors is demanding for the readout circuitry. The challenge results from the strongly varying signal intensities of the various sensor types to the target gas. While some sensors change their resistance only slightly, other types can react with a resistive change over a range of several decades. Therefore, a suitable readout circuit has to be able to capture all these resistive variations, requiring it to have a very large dynamic range. This work presents a compact embedded system that provides a full, high range input interface (readout and heater management) for MOX sensor operation. The system is modular and consists of a central mainboard that holds up to eight sensor-modules, each capable of supporting up to two MOX sensors, therefore supporting a total maximum of 16 different sensors. Its wide input range is archived using the resistance-to-time measurement method. The system is solely built with commercial off-the-shelf components and tested over a range spanning from 100 Ω to 5 GΩ (9.7 decades) with an average measurement error of 0.27% and a maximum error of 2.11%. The heater management uses a well-tested power-circuit and supports multiple modes of operation, hence enabling the system to be used in highly automated measurement applications. The experimental part of this work presents the results of an exemplary screening of 16 sensors, which was performed to evaluate the system’s performance.

The choice of suitable semiconducting metal oxide (MOX) gas sensors for the detection of a specific gas or gas mixture is time-consuming since the sensor’s sensitivity needs to be characterized at multiple temperatures to find its optimal operating conditions. To obtain reliable measurement results, it is very important that the power for the sensor’s integrated heater is stable, regulated and error-free (or error-tolerant). Especially the error-free requirement can be only be achieved if the power supply implements failure-avoiding and failure-detection methods. The biggest challenge is deriving multiple different voltages from a common supply in an efficient way while keeping the system as small and lightweight as possible. This work presents a reliable, compact, embedded system that addresses the power supply requirements for fully automated simultaneous sensor characterization for up to 16 sensors at multiple temperatures. The system implements efficient (avg. 83.3% efficiency) voltage conversion with low ripple output (&lt;32 mV) and supports static or temperature-cycled heating modes. Voltage and current of each channel are constantly monitored and regulated to guarantee reliable operation. To evaluate the proposed design, 16 sensors were screened. The results are shown in the experimental part of this work.

Research activities regarding semiconductor metal oxide gas sensors are booming all over the world. The importance of semiconductor metal oxide gas sensors has been generally recognized in different research communities, which have been actively promoting fundamental research and practical application of gas sensors. It is well known that semiconductor metal oxide gas sensors have been widely used in various fields, and they are becoming a key demand in modern high-tech society. Therefore, basic research and application technology of semiconductor metal oxide gas sensors have attracted wide attention. This chapter briefly introduces the development history and research progress of semiconductor gas sensors, and particular emphasis is put on introducing recent innovative research and edge-cutting techniques on semiconductor gas sensors.