Theoretical basis for identification and measurement of air contaminants using an array of sensors having partly overlapping selectivities

doi:10.1016/0250-6874(84)85019-2

Sensors and Actuators

Volume 6, Issue 4, December 1984, Pages 225-243

https://doi.org/10.1016/0250-6874(84)85019-2 Get rights and content

Abstract

The objective of this paper is to provide a theoretical basis for the selection and effective use of an array of chemical sensors for a particular application. The array is to be used in conjunction with a microprocessor in a small portable device capable of detecting, identifying and quantifying any of a large number N of hazardous compounds, either singly or in mixtures of up to A such compounds. A compact array of S different sensors is used, each of which can operate in M differently selective modes so as to yield P=MS parameters, or data channels. The minimum number of parameters required to identify the compounds solely on the basis of the presence or absence of significant signals in the various channels is determined by the inequality

For example, P ⩾ 18 or P ⩾ 22 for N = 100 and A = 3 or 4, respectively.

A computational approach is proposed to first eliminate all candidate compounds whose known response patterns do not overlap the channels exhibiting the strongest signals. The remaining candidate compounds are then quantified by solving simultaneous linear equations.

A determinant, which is used to solve the simultaneous equations, provides a measure of the selectivity and/or non-redundancy of the set of channels exhibiting the strongest signals. This determinant is also inversely proportional to the experimental error or to the required accuracy of the measurements. By computing the values of this determinant for various mixtures, the selectivities and/or non-redundancies of alternative sets of P parameters with respect to a given set of N hazardous compounds can be evaluated and compared.

References (19)

P.K. Clifford et al.
Characteristics of semiconductor gas sensors, I: Steady state gas response
Sensors and Actuators
(1983)
J.N. Driscoll
Fluorescence photometry of thin-layer chromatograms and electropherograms
J. Chromatog.
(1977)
I. Lundstrom
Hydrogen sensitive MOS structures, Part 1: Principles and application
Sensors and Actuators
(1981)
D.T. Tuma et al.
Portable gas detection: enhancement of semiconducor capabilities by microcomputer control and signal processing
D.T. Tuma et al.
Microcomputer control and information processing technology for semiconductor gas sensors
J.R. Stetter
Portable device for detecting and identifying hazardous vapors
J. R. Stetter et al., Selective monitoring of hazardous chemicals in emergency situations, Proc. JANNAF Safety and...
J.R. Stetter
Portable instrument for the detection and identification of air pollutants
Extended Abstracts of the 4th National Symposium on Recent Advances in Pollutant Monitoring of Ambient Air and Stationary Sources, U.S. Environmental Protection Agency, Raleigh, N.C.
(8101984)
J. R. Stetter, S. Zaromb and W. R. Penrose, Sensor array for toxic gas detection, U.S. Patent application Serial No....

There are more references available in the full text version of this article.

Cited by (101)

Portable electronic nose system with elastic architecture and fault tolerance based on edge computing, ensemble learning, and sensor swarm
2023, Sensors and Actuators B: Chemical
Citation Excerpt :
The electronic nose (E-nose) analyzes odor components by mimicking the working principles of mammals’ olfactory system [1–3], which has been highly successful in a wide range of commercial applications, such as food safety, disease diagnosis, and environmental monitoring [4–6].
The portable electronic nose (E-nose) systems are suffering from the limited computing ability of microcontrollers and can only adopt simple pattern recognition algorithms. The heavy coupling between the sensors greatly limits the anti-fault ability of the system. Herein, a novel ensemble learning framework based on independent artificial neural networks (ANN) is proposed in a portable E-nose system to recognize volatile organic compounds (VOCs). Edge computing is applied to complete data processing and analysis on an ARM Cortex-M3 based microcontroller in an E-nose system. Each sensing unit can recognize VOCs by training independent ANN models. Then an ensemble learning method further organizes the diverse ANN models into an onboard sensor swarm with a significantly enhanced performance. The accuracy of the type classification can reach 81.1%, which is improved by more than 20% compared with the best individual classifier. The R² score of concentration prediction can reach 84.1%, which is improved by more than 25% compared with the best regressor in the ensemble group. For both type identification and concentration prediction, the tolerance to the sensor faults of swarm-based E-nose is 10 and 18 times greater than that of the conventional array-based sensors, respectively. Our work has designed an elastic architecture based on sensor swarms, providing a novel avenue for the development of E-nose systems with high fault tolerance.
Biomass-derived carbon decorated by gold nanoparticles as efficient methanol electrochemical sensor
2022, Materials Science in Semiconductor Processing
Herein, we have demonstrated the successful combination of gold nanoparticles (AuNPs) and green carbon (GC) (biomass-derived) obtained from date seeds to detect methanol (MeOH) via a simple electrochemical approach. We have synthesized GC by pyrolysis method whereas AuNPs were deposited by the photochemical deposition technique. We have systematically characterized the Au@GC nanocomposite by X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), FESEM, high-resolution transmission electron microscopy (HR-TEM), and Raman spectroscopy. In comparison to a bare glassy carbon electrode (GCE) or GC fabricated GCE, the newly-fabricated Au@GC nanocomposite modified GCE (Au@GC/GCE) has shown better performance in MeOH sensing. The proposed sensor exhibited excellent sensitivity (0.0143 μAmM⁻¹), a lower detection limit (LOD = 0.39 mM at S/N = 3), and a wide linear dynamic range (LDR = 1–80 mM). This Au@GC/GCE sensor also showed good operational stability, reproducibility, and repeatability in MeOH detection. The research findings indicated the potential application of the Au@GC nanocomposite as an electroactive material for effective electrochemical sensing of environmental pollutants.
Target discrimination, concentration prediction, and status judgment of electronic nose system based on large-scale measurement and multi-task deep learning
2022, Sensors and Actuators B: Chemical
Pattern recognition is the core component of the electronic nose (E-nose). Traditional machine learning algorithms highly rely on the feature data selected manually for model training and testing. A complete experiment must be performed before the data can be further processed. To realize the automatic extraction of response features and simplify the model’s training and application process, a multi-task convolutional neural network (MTL-CNN) with a dual-block knowledge-sharing structure is designed to train a model for the E-nose system. This model can simultaneously perform three different classification tasks, for the purposes of target discrimination, concentration prediction, and state judgment. Only a few consecutive seconds of response data are needed to be input into the trained model to obtain various information about the E-nose. With the utilization of an unmanned gas-sensing test system, large-scale measurements of the E-nose can be carried out automatically. A baseline tracking algorithm (BTA) is designed based on the relative changes of short-term data, reducing the impact of long-term shifts. Over thousands of gas response processes and more than 10 million sensing data have participated in the training of the deep learning model. The 5-fold cross-validation method shows that the fully trained model has an outstanding generalization performance. After the baseline is tracked automatically, the accuracy of three tasks towards 12 kinds of volatile organic compounds (VOCs) is about 95% (type recognition: 95.2%, concentration prediction: 92.1%, status judgment: 97.3%) using only 4 s of sensing data during the response status of the E-nose. Our work shows the distinct advantages of combining “big data” and “deep learning” in the gas-sensing field and further proves that the employment of MTL-CNN can significantly improve the training and application efficiency of the E-nose.
Estimation of multicomponent organic solvent vapor mixture composition with electroconducting polymer chemiresistors
2016, Sensors and Actuators, B: Chemical
Citation Excerpt :
On the contrary, application of cross-selective sensors for quantitative analysis is covered in the literature to a much lesser extent. While the theoretical bases for this kind of analysis are well known, and related works date back to the eighties of the last century [36], practical application of the theory to the chemical sensor signals meets many obstacles, including but not limited to the technical complexity of obtaining large amount of calibration data, uncertainties in the calibration samples concentration and composition, poor reproducibility of sensor responses and long-term stability of sensor characteristics, and the influence of environmental factors upon the measurement results which is difficult to prevent or compensate for. Nevertheless, some research in this direction have been made [37–40].
An approach to quantitative analysis of multicomponent gaseous mixtures with an array of cross-selective sensors is proposed, and experimentally tested with the mixtures of four organic solvent vapors (acetone, chloroform, isopropanol and toluene) in atmospheric air and conducting polymer based chemiresistive sensors with three different sensitive coatings – polyaniline, polypyrrole and poly-3-methyltiophene doped with phosphotungstic and perchloric acid ions. The proposed method employs simplified semi-parametric model for prediction of the sensor responses, with emphasis on direct use of the information contained in the reference calibration data. Analysis algorithm is based on the matching between predicted and observed responses with additional regularization of the optimal prediction search procedure. Potential possibility of utilization of the presented approach for quantitative analysis of selected components, albeit with limited accuracy, is demonstrated, with the best case relative error of about 30% achieved for the studied sample data. The ways to further improvement of the analysis quality are discussed.
Thermally modulated multi sensor arrays of SnO<inf>2</inf>/additive/electrode combinations for enhanced gas identification
2015, Sensors and Actuators, B: Chemical
Citation Excerpt :
Thermo-cycling procedure could take advantage by operating an array of sensors, whose temperature is varied dynamically over time, and may show a higher degree of gas identification capability than the single sensor alternative. Even though the sensor array configuration has been introduced with isothermal operation over several years as a method to counteract the cross-selectivity of gas sensors [33–35], there are only a few studies [18,36–38] reported on the thermo-cycling of gas sensor arrays. In this context, the present work aims to develop sensor arrays, each encompasses four different SnO2/additive sensitive layers and operates thermo-cyclically, which can yield four different CTPs, i.e. four different information sequences simultaneously concerning an exposed gas.
We report the development of a multi sensor array integrated with four different layers of tin oxide/additive combinations, which operates thermo-cyclically by triangular heater voltages, for specific gas identification. These multi sensor arrays could yield simultaneously four different patterns of dynamic conductance, which can be used for specific gas component analysis. Pure SnO₂, SnO₂/Sb₂O₃, SnO₂/Sb₂O₃/La₂O₃ and SnO₂/Sb₂O₃/CuO were used as sensing materials on a single chip with Au or Pt as electrode material to study the response behavior to propene, CO and methane at different concentrations and temperature rates/cycle time. The nature of the specific gas recognition patterns is highly determined by additive material, electrode metallization, temperature rate, type of gas and gas concentration. The specific gas behavior is discussed in connection with the reaction processes taking place at the pores of the sensing layer and at the triple phase boundary of layer, electrode and gas.
Solid-state gas sensors for breath analysis: A review
2014, Analytica Chimica Acta
Citation Excerpt :
The aggregation of chemical sensors into arrays was first introduced as a method to overcome the limited selectivity of solid-state gas sensors. When complemented by a multivariate data analysis technique, such as those used in chemometrics, a sensor array allows for the quantification and the identification of compounds with a performance well beyond that of a single device [120,121]. Gas sensor arrays have a very close similitude with the properties of the olfactory receptor neurons (ORN).
The analysis of volatile compounds is an efficient method to appraise information about the chemical composition of liquids and solids. This principle is applied to several practical applications, such as food analysis where many important features (e.g. freshness) can be directly inferred from the analysis of volatile compounds.
The same approach can also be applied to a human body where the volatile compounds, collected from the skin, the breath or in the headspace of fluids, might contain information that could be used to diagnose several kinds of diseases. In particular, breath is widely studied and many diseases can be potentially detected from breath analysis.
The most fascinating property of breath analysis is the non-invasiveness of the sample collection. Solid-state sensors are considered the natural complement to breath analysis, matching the non-invasiveness with typical sensor features such as low-cost, easiness of use, portability, and the integration with the information networks. Sensors based breath analysis is then expected to dramatically extend the diagnostic capabilities enabling the screening of large populations for the early diagnosis of pathologies.
In the last years there has been an increased attention to the development of sensors specifically aimed to this purpose. These investigations involve both specific sensors designed to detect individual compounds and non-specific sensors, operated in array configurations, aimed at clustering subjects according to their health conditions. In this paper, the recent significant applications of these sensors to breath analysis are reviewed and discussed.

View all citing articles on Scopus

View full text

Theoretical basis for identification and measurement of air contaminants using an array of sensors having partly overlapping selectivities

Abstract

Sensors and Actuators

J. Chromatog.

Sensors and Actuators

Portable gas detection: enhancement of semiconducor capabilities by microcomputer control and signal processing

Microcomputer control and information processing technology for semiconductor gas sensors

Portable device for detecting and identifying hazardous vapors

Portable instrument for the detection and identification of air pollutants

Extended Abstracts of the 4th National Symposium on Recent Advances in Pollutant Monitoring of Ambient Air and Stationary Sources, U.S. Environmental Protection Agency, Raleigh, N.C.