Introduction The selective quantification of target gases in complex mixtures is an important part of numerous applications of chemical gas sensors. To this end, the data is usually evaluated with statistical models using lab calibration data. In most cases, the experimental design exposes the sensor to only a few fixed concentration levels, either one gas at a time or a limited number of pre-determined combinations [1]. This approach neglects masking effects and gas interactions as well as the risk of overfitting due to systematic influences. We present a calibration method based on randomized gas mixtures in which all calibration gases are present at each calibration point, with concentrations drawn from well-defined distributions reflecting the targeted application. Using indoor air quality monitoring as an example, we show that this method is superior to the classic “sequential” approach as well as flexible and easy to configure. Method The relevant gases for the measurement of indoor air quality can be divided into two groups: background (zero air, humidity and inorganic gases) and (very) volatile organic compounds ((V)VOC). Figure 1 illustrates the generation of randomized gas mixtures. The aim of the evaluation could be the quantification of TVOC (total VOC, here approximated by few gases as VOCsum). Alternatively, the (V)VOCs can be sub-divided into interferent and target VOCs with the aim of quantifying toxic or carcinogenic substances (here benzene and formaldehyde). As our current gas mixing system [2] is limited to six gases, two representatives were selected for inorganic background gases (hydrogen H2, carbon monoxide CO), interferent VOCs (acetone, toluene) and target VOCs (formaldehyde, benzene), respectively. The randomized gas mixtures were generated with a Python script and the distributions of the individual gases can be found in [3]. A total of 800 different mixtures were tested with each offered to the sensors for 20 min. For comparison with a sequential calibration strategy (ascending concentration values, one gas at a time), we also measured such a calibration profile. Each gas was offered at four different concentrations for 20 min each and this sequence was repeated three times at different humidity (Table 2). A total of eleven different sensors were tested, of which only the AS-MLV-P2 (ams AG) sensor is shown here. The sensor was operated in temperature cycles (TCO) [4]. For feature extraction mean and slope values were calculated for 120 equidistant ranges of each cycle. The best 20 features are selected using a principal component analysis to prevent overfitting. The quantification of the desired target value is carried out with partial least squares regression (PLSR). To compare the ability to quantify each individual gas in the given mixture, a performance value of performance = std(c(g))/RMSEP(g) - 1 is defined [3] where std(c(g)) is the standard deviation of the concentration distribution of gas g and RMSEP(g) is the root mean square error (RMSE) of the concentration prediction of the PLSR model for this gas. Results Figure 2 shows an example of a PLSR model for acetone trained with 100 randomized gas exposures (grey). The RMSE is calculated from 100 additional random gas exposures (dashed lines). The blue dots show the points predicted for 100 more gas exposures that were not used in training and validation. A good agreement of the test data set with the training data can be seen. To calculate the performance for the individual regression models (Table 2), the models are trained, validated and tested with a data set of 400 randomized gas mixtures. The highest performance (8.58) among individual gases is achieved for CO, i.e. the AS-MLV-P2 can quantify CO accurately in a variable background of all other gases. To compare randomized and sequential calibration (Table 3), three combinations are investigated: (1) training/validation randomized, testing randomized; (2) training/validation randomized, testing sequential; and (3) training/validation sequential, testing randomized. Validation is always performed with 6-fold cross-validation. For a fair comparison, randomized mixtures with all gases in the same concentration range as the sequential measurements were chosen. The best overall performance is achieved for randomized training with randomized testing with performance results well above one for five target gases whereas sequential training with randomized testing achieves the worst performance with a non-zero performance result only for CO. The randomized data is obviously more challenging to predict, which is expected due to the more realistic background. At the same time, this allows a more efficient training, as one data point for each gas is obtained from each gas mixture. Sequential calibration cannot provide a model for any target gas which is able to predict the more realistic random mixtures. The presented randomized calibration generates reliable models from a few gas exposures for a whole series of target gases and is, therefore, much more effective. For the calibration of mixtures, it is also much more efficient than combinatorial approaches. The presented method thus offers a promising approach for the successful transition of chemical sensors from the laboratory into the field. References [1] Wolfrum et al., Calibration transfer among sensor arrays designed for monitoring volatile organic compounds in indoor air quality, IEEE Sensors Journal, 6(2006), 1638–1643. doi:10.1109/JSEN.2006.884558. [2] Helwig et al., Gas mixing apparatus for automated gas sensor characterization, Measurement Science and Technology, 2014, 25(5), 055903–055903, doi:10.1088/0957-0233/25/5/055903. [3] Bastuck, dissertation, Saarland University and Linköping University (2019). doi:10.3384/diss.diva-159106. [4] Schütze, Sauerwald, Dynamic operation of semiconductor sensors, in: Semiconductor Gas Sensors, Woodhead Publishing, 2nd Edition, 2020, doi:10.1016/B978-0-08-102559-8.00012-4. Figure 1

Applications like air quality, fire detection and detection of explosives require selective and quantitative measurements in an ever-changing background of interfering gases. One main issue hindering the successful implementation of gas sensors in real-world applications is the lack of appropriate calibration procedures for advanced gas sensor systems. This article presents a calibration scheme for gas sensors based on statistically distributed gas profiles with unique randomized gas mixtures. This enables a more realistic gas sensor calibration including masking effects and other gas interactions which are not considered in classical sequential calibration. The calibration scheme is tested with two different metal oxide semiconductor sensors in temperature-cycled operation using indoor air quality as an example use case. The results are compared to a classical calibration strategy with sequentially increasing gas concentrations. While a model trained with data from the sequential calibration performs poorly on the more realistic mixtures, our randomized calibration achieves significantly better results for the prediction of both sequential and randomized measurements for, for example, acetone, benzene and hydrogen. Its statistical nature makes it robust against overfitting and well suited for machine learning algorithms. Our novel method is a promising approach for the successful transfer of gas sensor systems from the laboratory into the field. Due to the generic approach using concentration distributions the resulting performance tests are versatile for various applications.

We present an equivalent circuit model for a titanium dioxide-based humidity sensor which enables discrimination of three separate contributions to the sensor impedance. The first contribution, the electronic conductance, consists of a temperature-dependent ohmic resistance. The second contribution arises from the ionic pathway, which forms depending on the relative humidity on the sensor surface. It is modeled by a constant-phase element (CPE) in parallel with an ohmic resistance. The third contribution is the capacitance of the double layer which forms at the blocking electrodes and is modeled by a second CPE in series to the first CPE. This model was fitted to experimental data between 1 mHz and 1 MHz recorded at different sensor temperatures (between room temperature and 320 ∘C) and different humidity levels. The electronic conductance becomes negligible at low sensor temperatures, whereas the double-layer capacitance becomes negligible at high sensor temperatures in the investigated frequency range. Both the contribution from the ionic pathway and from the double-layer capacitance strongly depend on the relative humidity and are, therefore, suitable sensor signals. The findings define the parameters for the development of a dedicated Fourier-based impedance spectroscope with much faster acquisition times, paving a way for impedance-based high-temperature humidity sensor systems.

In order to facilitate the widespread use of gas sensors, some challenges must still be overcome.Many of those are related to the reliable quantification of ultra-low concentrations of specific compounds in a background of other gases.This thesis focuses on three important items in the measurement chain: sensor material and operating modes, evaluation of the resulting data, and test gas generation for efficient sensor calibration.New operating modes and materials for gas-sensitive field-effect transistors have been investigated.Tungsten trioxide as gate oxide can improve the selectivity to hazardous volatile organic compounds like naphthalene even in a strong and variable ethanol background.The influence of gate bias and ultraviolet light has been studied with respect to the transport of oxygen anions on the sensor surface and was used to improve classification and quantification of different gases.DAV 3 E, an internationally recognized MATLAB-based toolbox for the evaluation of cyclic sensor data, has been developed and published as opensource.It provides a user-friendly graphical interface and specially tailored algorithms from multivariate statistics.The laboratory tests conducted during this project have been extended with an interlaboratory study and a field test, both yielding valuable insights for future, more complex sensor calibration.A novel, efficient calibration approach has been proposed and evaluated with ten different gas sensor systems.i styrning och data-utvärdering som tagits fram har både en jämförande undersökning av modellernas/metodernas prestanda av två oberoende laboratorier och fält-mätningar i en av de tilltänkta tillämpningarna genomförts.Bl.a.baserat på resultaten och insikterna från dessa övningar har ett helt nytt angreppssätt avseende robust och effektiv kalibrering och kvalitets-utvärdering av sensor-system utvecklats och utvärderats för tio olika sensor-system.

To fulfil today's requirements, gas sensors have to become more and more sensitive and selective. Temperature-cycled operation has long been used to enhance the sensitivity and selectivity of metal-oxide semiconductor gas sensors and, more recently, silicon-carbide-based, gas-sensitive field-effect transistors (SiC-FETs). In this work, we present a novel method to significantly enhance the effect of gate bias on a SiC-FET's response, giving rise to new possibilities for static and transient signal generation and, thus, increased sensitivity and selectivity. A tungsten trioxide (WO3) layer is deposited via pulsed laser deposition as an oxide layer beneath a porous iridium gate, and is doped with 0.1 AT% of lithium cations. Tests with ammonia as a well-characterized model gas show a relaxation effect with a time constant between 20 and 30 s after a gate bias step as well as significantly increased response and sensitivity at +/- 2V compared to 0V. We propose an electric field-mediated change in oxygen surface coverage as the cause of this novel effect.

In this work, we use a gas sensor system consisting of a commercially available gas sensor in temperature cycled operation. It is trained with an extensive gas profile for detection and quantification of hazardous volatile organic compounds (VOC) in the ppb range independent of a varying background of other, less harmful VOCs and inorganic interfering gases like humidity or hydrogen. This training was then validated using a different gas mixture generation apparatus at an independent lab providing analytical methods as reference. While the varying background impedes selective detection of benzene and naphthalene at the low concentrations supplied, both formaldehyde and total VOC can well be quantified, after calibration transfer, by models trained with data from one system and evaluated with data from the other system. The lowest achievable root mean squared errors of prediction were 49 ppb for formaldehyde (in a concentration range of 20-200 ppb) and 150 mu g/m(3) (in a concentration range of 25-450 mu g/m(3)) for total VOC. The latter uncertainty improves to 13 mu g/m(3) with a more confined model range of 220-320 mu g/m(3). The data from the second lab indicate an interfering gas which cannot be detected analytically but strongly influences the sensor signal. This demonstrates the need to take into account all sensor relevant gases, like, e.g., hydrogen and carbon monoxide, in analytical reference measurements.

We present DAV3E, a MATLAB toolbox for feature extraction from, and evaluation of, cyclic sensor data. These kind of data arise from many real-world applications like gas sensors in temperature cycled operation or condition monitoring of hydraulic machines. DAV3E enables interactive shape-describing feature extraction from such datasets, which is lacking in current machine learning tools, with subsequent methods to build validated statistical models for the prediction of unknown data. It also provides more sophisticated methods like model hierarchies, exhaustive parameter search, and automatic data fusion, which can all be accessed in the same graphical user interface for a streamlined and efficient workflow, or via command line for more advanced users. New features and visualization methods can be added with minimal MATLAB knowledge through the plug-in system. We describe ideas and concepts implemented in the software, as well as the currently existing modules, and demonstrate its capabilities for one synthetic and two real datasets. An executable version of DAV3E can be found at http://www.lmt.uni-saarland.de/dave (last access: 14 September 2018). The source code is available on request.

Static and dynamic responses of a silicon carbide field-effect transistor gas sensor have been investigated at two different gate biases in several test gases. Especially the dynamic effects are gas dependent and can be used for gas identification. The addition of ultraviolet light reduces internal electrical relaxation effects, but also introduces new, temperature-dependent effects.

Data from a silicon carbide based field-effect transistor were recorded over a period of nine days in a ventilated school room. For enhanced sensitivity and selectivity especially to formaldehyde, porous iridium on pulsed laser deposited tungsten trioxide was used as sensitive layer, in combination with temperature cycled operation and subsequent multivariate data processing techniques. The sensor signal was compared to reference measurements for formaldehyde concentration, CO2 concentration, temperature, and relative humidity. The results show a distinct pattern for the reference formaldehyde concentration, arising from the day/night cycle. Taking this into account, the projections of both principal component analysis and partial least squares regression lead to almost the same result concerning correlation to the reference. The sensor shows cross-sensitivity to an unidentified component of human activity, presumably breath, and, possibly, to other compounds appearing together with formaldehyde in indoor air. Nevertheless, the sensor is able to detect and partially quantify formaldehyde below 40 ppb with a correlation to the reference of 0.48 and negligible interference from ambient temperature or relative humidity.

Data from a silicon carbide based field-effect transistor were recorded over a period of nine days in a ventilated school room. For enhanced sensitivity and selectivity especially to formaldehyde, ...

DAV³E (Data Analysis and Verification/Visualization/Validation Environment) is an object-oriented MATLAB toolbox developed to facilitate the process of building a statistical model out of data generated by cyclically driven sensors or cyclic (industrial) processes.Evaluation of such data usually involves several steps of preprocessing and dimensionality reduction until a good classification or regression model can be built.However, it is rarely known which combination of algorithms and parameters will produce the best result due to its strong dependence on data structure.DAV³E allows data exploration, including visualizations at each step, with an interactive, modular GUI.When a good workflow has been found for the data structure at hand, it can be exported to a command line script for batch processing of big data collections.

In this work, we exposed an MIS capacitor with porous platinum as gate material to different concentrations of CO and NH3. Its capacitance and typical reaction products (water, CO2 and NO) were monitored at high and low oxygen concentration and different gate bias voltages. We found that the gate bias influences the switch-point of the binary CO response usually seen when either changing the temperature at constant gas concentrations or the CO/O2 ratio at constant temperature. For NH3, the sensor response as well as product reaction rates increase with bias voltages up to 6 V. A capacitance overshoot is observed when switching on or off either gas at low gate bias, suggesting increasing oxygen surface coverage with decreasing gate bias.

Temperature cycled operation and multivariate statistics have been used to compare the selectivity of two gate (i.e. sensitive) materials for gas-sensitive, silicon carbide based field effect transistors towards naphthalene and ethanol in different mixtures of the two substances. Both gates have a silicon dioxide (SiO2) insulation layer and a porous iridium (Ir) electrode. One of it has also a dense tungsten trioxide (WO3) interlayer between Ir and SiO2. Both static and transient characteristics play an important role and can contribute to improve the sensitivity and selectivity of the gas sensor.The Ir/SiO2 is strongly influenced by changes in ethanol concentration, and is, thus, able to quantify ethanol in a range between 0 and 5 ppm with a precision of 500 ppb, independently of the naphthalene concentrations applied in this investigation. On the other hand, this sensitivity to ethanol reduces its selectivity towards naphthalene, whereas Ir/WO3/SiO2 shows an almost binary response to ethanol. Hence, the latter has a better selectivity towards naphthalene and can quantify legally relevant concentrations down to 5 ppb with a precision of 2.5 ppb, independently of a changing ethanol background between 0 and 5 ppm. (C) 2016 Elsevier B.V. All rights reserved.

Gas sensitive metal/metal-oxide field effect transistors based on silicon carbide were used to study the sensor response to benzene (C6H6) at the low parts per billion (ppb) concentration range. A combination of iridium and tungsten trioxide was used to develop the sensing layer. High sensitivity to 10 ppb C6H6 was demonstrated during several repeated measurements at a constant temperature from 180 to 300 °C. The sensor performance were studied also as a function of the electrical operating point of the device, i.e., linear, onset of saturation, and saturation mode. Measurements performed in saturation mode gave a sensor response up to 52 % higher than those performed in linear mode.

With the increased interest in development of cheap, simple means for indoor air quality monitoring, and specifically in relation to certain well-known pollutant substances with adverse health effects even at very low concentrations, such as different Volatile Organic Compounds (VOCs), this contribution aims at providing an overview of the development status of the silicon carbide field effect transistor (SiC FET) based sensor platform for ppb level detection of VOCs. Optimizing the transducer design, the gas-sensitive material(s) composition, structure and processing, its mode of operation - applying temperature cycled operation in conjunction with multivariate data evaluation - and long-term performance it has been possible to demonstrate promising results regarding the sensor technology's ability to achieve both single-digit ppb sensitivity towards e.g. naphthalene as well as selective detection of individual substances in a mixture of different VOCs.

This study is a short overview on air pollution, its definition, history, and adverse effects on health, environment, and economy, with a particular emphasis on indoor air quality. We demonstrate that gas sensors based on silicon carbide field effect transistor (SiC-FET) technology are suitable for indoor air quality applications due to their high sensitivity, long-term stability, and robustness. Enhancement of selectivity can be reached through the optimization of the sensing layer (gate contact) in combination with smart operation and multivariate statistical methods for data analysis. Formaldehyde, naphthalene, and benzene, belonging to the class of hazardous volatile organic compounds (VOCs), are our target gases at the low parts per billion concentration level.

DAV3E, eine vollständig objektorientierte MATLAB-Toolbox zur Auswertung zyklischer Sensordaten, wird vorgestellt. Es können beliebig viele Daten importiert und individuelle Vorschriften zur Vorverarbeitung und Merkmalsextraktion definiert werden. Aus den Merkmalen kann anschließend mit vordefinierten Dimensionsreduktionsund Klassifizierungsalgorithmen ein Modell erzeugt werden, das in der Lage ist, neue, unbekannte Daten zu klassifizieren. Der graphische Aufsatz für die Toolbox leitet den Anwender durch die Auswertung und stellt umfangreiche Funktionen zur Visualisierung zur Verfügung, so dass neue Daten einfach untersucht werden können, um ein optimales Ergebnis zu erreichen. Im Falle einer bereits vorhandenen Auswertestrategie kann die kommandozeilenbasierte Version der Toolbox genutzt werden, um die Auswertung zu automatisieren, oder verschiedene, neue Methoden schnell zu validieren.