Metal Oxide Sensors Research Articles

Characteristics of a Pt/NiO thin film-based ammonia gas sensor

The sensing characteristics of a Pt/NiO thin film-based resistor-type ammonia gas sensor are comprehensively studied and demonstrated. Experimentally, the studied Pt/NiO ammonia gas sensor exhibits improved performance, including a higher sensing response of ratio of 1278%, an extremely low detection limit of 10ppb NH3/air, and fast speeds, at an optimal operating temperature of 300°C. Based on the advantages indicated above and the benefits of its simple structure, relatively easy fabrication, and inherent p-type semiconductor properties, the studied device is promising for high-performance ammonia gas sensing and complementary metal oxide sensor (CMOS) array applications.

Detecting Gastric Contents by Electronic Nose Device Analysis of Exhaled Volatile Organic Compounds

Introduction: Though rare, aspiration events during endoscopy are associated with serious consequences. Furthermore, the presence of significant gastric contents can lead to aborted procedures, reducing endoscopy suite efficiency. However, there is no current system to determine if a patient has gastric contents prior to a sedated procedure, other than inquiries regarding timing of their last meal. We previously piloted the use of a first generation electronic nose device to detect the presence of gastric contents via exhaled volatile organic compounds (VOC's). Here we present the development of a testing program for gastric contents using an improved second generation electronic nose device. Methods: Patients provided 5-minute breath samples into a second generation Aeonose electronic nose (Aeonose, eNose Company, Zutphen, NL). Breath samples were collected from two groups: Group 1 was subjects fasting for gastrointestinal endoscopy; Group 2 was subjects who had eaten a meal 30 to 120 minutes before providing a sample. The Aeonose device analyzes a breath sample over 5-minutes. During this time the patient's breath is circulated over a metal-oxide sensor array. VOC's in the breath sample cause reduction-oxidation reactions in the sensor array. These reactions cause changes in the conductance of circuits that run through the sensor and in the temperature of the sensor array. The Aeonose device measures these changes over time to create a 3-D digital signature of each breath sample. These digital signatures are uploaded into an artificial neural network. This network performs pattern recognition analysis between breath samples from fasting and non-fasting subjects. Results: Breath samples were obtained from a total of 96 individuals; 50 fasting and 46 non-fasting. Optimal multivariate modeling developed a predictive model for the fasting state with performance characteristics of 80% sensitivity, 70% specificity, 75% accuracy and AUC=0.75 (Figure 1). The positive predictive value for fasting state was 74%. The tests negative predictive value was 76%. The Matthews Correlation Coefficient was 0.50.Figure: Detection of Gastric Contents via Electronic Nose Predictive Model Receiver Operating Curve.Conclusion: The second-generation Aeonose electronic-nose provides fair discrimination between a patient in a fasting state (no gastric contents) and a patient in a non-fasting state (present gastric contents). Given the lack of a validated tool to determine the presence of gastric contents, the significant harm gastric contents can cause in sedated procedures, and the low cost, non-invasive nature of electronic nose testing, this study suggests the Aeonose device can be used as a screening tool to identify patients with retained gastric contents prior to sedated procedures. The point of care nature of this testing would allow for an alteration in care to improve procedure safety and efficiency. Further work to refine the model and improve the tests efficiency is ongoing. A validation study is also in progress.

Developing a Screening Test for Barrettʼs Esophagus Using Electronic Nose Device Analysis of Exhaled Volatile Organic Compounds: 2017 Presidential Poster Award

Introduction: The prognosis for advanced esophageal adenocarcinoma (EAC) remains dismal. However, while persons with Barrett's esophagus (BE) (the only known precursor of EAC) undergo surveillance, over 90% of EACs are diagnosed outside of surveillance programs. This suggests current screening and surveillance programs are inadequate. Currently available BE screening modalities are expensive and intrusive. In an effort to develop a highly acceptable, less expensive screening method, we piloted the use of a non-invasive, highly portable electronic nose device to detect BE from breath samples. Here we present the development of a screening test for Barrett's esophagus using a second generation electronic nose device. Methods: Patients provided 5-minute breath samples into a second generation Aeonose electronic nose (Aeonose, eNose Company, Zutphen, NL). Breath samples were collected from two groups: Group 1 (Group BE) was subjects with untreated BE (≥ 1 cm of columnar mucosa from the gastroesophageal junction with histopathologic confirmation of intestinal metaplasia); Group 2 (Group no BE) were subjects undergoing EGD who did not have BE and had no history of BE. The Aeonose device analyzes a breath sample over 5-minutes. During this time the patient's breath is circulated over a metal-oxide sensor array. VOC's in the breath sample cause reduction-oxidation reactions in the sensor array. These reactions cause changes in the conductance of circuits that run through the sensor and in the temperature of the sensor array. The Aeonose device measures these changes over time to create a 3-D digital signature of each breath sample. These digital signatures are uploaded into an artificial neural network in to perform pattern recognition analysis between breath samples from Group BE versus Group no BE. Results: Breath samples were obtained from a total of 58 individuals; 27 in Group BE and 31 in Group No BE. Optimal multivariate modeling based on our VOC signatures demonstrated a predictive model for BE with performance characteristics of 89% sensitivity, 71% specificity, 79% accuracy, and AUC=0.81 (Figure 1). The positive predictive value was 0.73 and negative predictive value was 0.88. The Matthews Correlation Coefficient was 0.60. The recruitment rate was 99% for breath testing.FigureConclusion: The second-generation Aeonose electronic-nose provides good discrimination between persons with BE and those without BE. The high sensitivity, low cost, non-invasive nature of electronic nose testing, and high rate of patient usage suggest the Aeonose device can be used as a screening tool to identify patients with BE for further surveillance. Further work to refine the model and improve the tests efficiency is ongoing. A validation study is also in progress.

Improved sensing characteristics of methane over ZnO nano sheets upon implanting defects and foreign atoms substitution

Thanks to the growing interests of metal oxide sensors in environmental and industrial uses, this study presents the sensing mechanism of methane gas (CH4) on recently synthesized two-dimensional form of ZnO, ZnO nano sheets (ZnO-NS). The adsorption energy of CH4 on pristine ZnO-NS, calculated by means of van der Waals corrected first-principles calculations, is found to be insufficient restricting its application as an efficient nano sensor. However, the creation of (O/Zn) vacancies and the substitution of foreign dopants into ZnO-NS considerably intensify the binding energy of CH4. Through a comprehensive energetic analysis, it is observed that among all the substituents, boron (B), sulphur (S) and gallium (Ga) improves the binding of CH4 to 2.75, 6.1 and 7.5 times respectively than its values on pristine ZnO-NS. In addition to the CH4 binding energies falling ideally between physisorption and chemisorption range, a prominent variation in the electronic properties before and after CH4 exposure indicates the promise of substituted Zn-NS as a useful nano sensors.

Quantifying Neighborhood-Scale Spatial Variations of Ozone at Open Space and Urban Sites in Boulder, Colorado Using Low-Cost Sensor Technology.

Recent advances in air pollution sensors have led to a new wave of low-cost measurement systems that can be deployed in dense networks to capture small-scale spatio-temporal variations in ozone, a pollutant known to cause negative human health impacts. This study deployed a network of seven low-cost ozone metal oxide sensor systems (UPods) in both an open space and an urban location in Boulder, Colorado during June and July of 2015, to quantify ozone variations on spatial scales ranging from 12 m between UPods to 6.7 km between open space and urban measurement sites with a measurement uncertainty of ~5 ppb. The results showed spatial variability of ozone at both deployment sites, with the largest differences between UPod measurements occurring during the afternoons. The peak median hourly difference between UPods was 6 ppb at 1:00 p.m. at the open space site, and 11 ppb at 4:00 p.m. at the urban site. Overall, the urban ozone measurements were higher than in the open space measurements. This study evaluates the effectiveness of using low-cost sensors to capture microscale spatial and temporal variation of ozone; additionally, it highlights the importance of field calibrations and measurement uncertainty quantification when deploying low-cost sensors.

Response of a Zn₂TiO₄ Gas Sensor to Propanol at Room Temperature.

In this study, three different compositions of ZnO and TiO2 powders were cold compressed and then heated at 1250 °C for five hours. The samples were ground to powder form. The powders were mixed with 5 wt % of polyvinyl butyral (PVB) as binder and 1.5 wt % carbon black and ethylene-glyco-lmono-butyl-ether as a solvent to form screen-printed pastes. The prepared pastes were screen printed on the top of alumina substrates containing arrays of three copper electrodes. The three fabricated sensors were tested to detect propanol at room temperature at two different concentration ranges. The first concentration range was from 500 to 3000 ppm while the second concentration range was from 2500 to 5000 ppm, with testing taking place in steps of 500 ppm. The response of the sensors was found to increase monotonically in response to the increment in the propanol concentration. The surface morphology and chemical composition of the prepared samples were characterized by Scanning Electron Microscopy (SEM) and X-Ray Diffraction (XRD). The sensors displayed good sensitivity to propanol vapors at room temperature. Operation under room-temperature conditions make these sensors novel, as other metal oxide sensors operate only at high temperature.

An Ultrasensitive Organic Semiconductor NO2 Sensor Based on Crystalline TIPS‐Pentacene Films

Organic semiconductor gas sensor is one of the promising candidates of room temperature operated gas sensors with high selectivity. However, for a long time the performance of organic semiconductor sensors, especially for the detection of oxidizing gases, is far behind that of the traditional metal oxide gas sensors. Although intensive attempts have been made to address the problem, the performance and the understanding of the sensing mechanism are still far from sufficient. Herein, an ultrasensitive organic semiconductor NO2 sensor based on 6,13-bis(triisopropylsilylethynyl)-pentacene (TIPS-petacene) is reported. The device achieves a sensitivity over 1000%/ppm and fast response/recovery, together with a low limit of detection (LOD) of 20 ppb, all of which reach the level of metal oxide sensors. After a comprehensive analysis on the morphology and electrical properties of the organic films, it is revealed that the ultrahigh performance is largely related to the film charge transport ability, which was less concerned in the studies previously. And the combination of efficient charge transport and low original charge carrier concentration is demonstrated to be an effective access to obtain high performance organic semiconductor gas sensors.

TiO2 Nanorods Decorated with Pd Nanoparticles for Enhanced Liquefied Petroleum Gas Sensing Performance.

Development of highly sensitive and selective semiconductor-based metal oxide sensor devices to detect toxic, explosive, flammable, and pollutant gases is still a challenging research topic. In the present work, we systematically enhanced the liquefied petroleum gas (LPG) sensing performance of chemical bath deposited TiO2 nanorods by decorating Pd nanoparticle catalyst. Surface morphology with elemental mapping, crystal structure, composition and oxidation states, and surface area measurements of pristine TiO2 and Pd:TiO2 nanorods was examined by high resolution transmission electron microscopy with energy-dispersive X-ray spectroscopy, X-ray diffraction, X-ray photoelectron spectroscopy, and nitrogen adsorption-desorption characterization techniques. LPG sensing performance of pristine TiO2 and Pd:TiO2 nanorods was investigated in different LPG concentration and operating temperature ranges. The LPG response of 21% for pristine TiO2 nanorods is enhanced to 49% after Pd catalyst decoration with reasonably fast response and recovery times. Further, the sensor exhibited long-term stability, which could be due to the strong metal support (Pd:TiO2) interaction and catalytic properties offered by the Pd nanoparticle catalyst. The work described herein demonstrates a general and scalable approach that provides a promising route for rational design of variety of sensor devices for LPG detection.

Evaluation of an electronic nose for odorant and process monitoring of alkaline-stabilized biosolids production

Electronic noses have been widely used in the food industry to monitor process performance and quality control, but use in wastewater and biosolids treatment has not been fully explored. Therefore, we examined the feasibility of an electronic nose to discriminate between treatment conditions of alkaline stabilized biosolids and compared its performance with quantitative analysis of key odorants. Seven lime treatments (0–30% w/w) were prepared and the resultant off-gas was monitored by GC-MS and by an electronic nose equipped with ten metal oxide sensors. A pattern recognition model was created using linear discriminant analysis (LDA) and principal component analysis (PCA) of the electronic nose data. In general, LDA performed better than PCA. LDA showed clear discrimination when single tests were evaluated, but when the full data set was included, discrimination between treatments was reduced. Frequency of accurate recognition was tested by three algorithms with Euclidan and Mahalanobis performing at 81% accuracy and discriminant function analysis at 70%. Concentrations of target compounds by GC-MS were in agreement with those reported in literature and helped to elucidate the behavior of the pattern recognition via comparison of individual sensor responses to different biosolids treatment conditions. Results indicated that the electronic nose can discriminate between lime percentages, thus providing the opportunity to create classes of under-dosed and over-dosed relative to regulatory requirements. Full scale application will require careful evaluation to maintain accuracy under variable process and environmental conditions.

A cheap and third-age-friendly home device for monitoring indoor air quality

Nowadays, thanks to the hardware costs reduction everyone at home owns at least one hi-technology devices to support life quality. This paper proposes a new methodology to analyse the indoor air quality with a cheap and third-age-dedicated device. A new prototype, called Home Pollution Embedded System (HOPES), was developed to give simple and understandable information, also comprehensible for people with cognitive problems or that are not familiar with new technologies. The device gathers pollutants data and displays to the user different air pollutants concentrations, from toxics gasses up to explosives. Moreover, an overall air quality index has been elaborated and is displayed by HOPES with lights and numerical information. HOPES is an Internet of things device that works in real time and that can be connected to the web and to a geographic information system platform to add spatial information of each pollutant. The hardware architecture employs a set of gas semiconductor sensors and an IR particulate matter sensor. Experimental results demonstrated that HOPES has an accurate sensor response and is suitable to be used with the proposed indices. In fact, despite of the poor-selective metal oxide sensors, the results highlighted how it is possible to get useful air quality information with a cheap device. However, the system needs more tests to validate the sensors array with different substances.

Selective detection of individual gases and CO/H2 mixture at low concentrations in air by single semiconductor metal oxide sensors working in dynamic temperature mode

Highly selective detection of various individual gases (CO, H2, CH4, C3H8, NO, NO2, H2S, SO2) at low concentrations (0.01–667ppm) in air by a single SnO2-based metal oxide sensor (MOX-sensor) is presented. The sensor operates in dynamic temperature mode combined with a number of adaptive signal processing algorithms. Artificial neural networks were proven to be more effective among the other adaptive algorithms implemented in this study. Identification of individual gases by a single sensor, averaged over all the gases and concentrations, resulted in only 13.2% false recognitions. Most of the failures were attributed to NO2 detection in 0.01–0.1ppm concentrations range.The ability of a single sensor to identify gas mixtures in a complex background was tested on the example of CO+H2 mixture in air, which simulates smoldering in the early stages of fire. The algorithm showed the ability to identify and quantify CO+H2 mixture with less than 10% error rate, even in constant presence of background gas (NO2 1.4ppm). Chemical modification of SnO2, increasing sensor response and sensitivity to individual components of the mixture, was proven to be beneficial for improvement of identification and quantification of gas mixture. Significant improvement in quantification accuracy (decrease in relative error from 7 to 2.5%) was achieved by utilizing a 3 sensor array in combination with an adaptive data processing algorithm, compared to the use of a single sensor alone. The prominent negative effect of humidity (Rh 30%, 25°C) on the performance of adaptive algorithms, sensor signal processing, system selectivity, and gas mixture identification is demonstrated.

CeO2/ionic liquid hybrid materials with enhanced humidity performance

CeO2/ionic liquid (IL) hybrid material has been successfully prepared by a facile ionothermal route under relatively mild temperature. When using it as sensing material for humidity sensor, the CeO2/IL hybrid material exhibits ultrafast response (4s) and recovery (<1s) performance. This result ranks it among the fastest ever reported values for humidity detection based on metal oxide sensor.

Low-cost mobile air pollution monitoring in urban environments: a pilot study in Lubbock, Texas

ABSTRACTThe complex nature of air pollution in urban areas prevents traditional monitoring techniques from obtaining measurements representative of true human exposure. The current study assessed the capability of low-cost mobile monitors to acquire useful data in a city without a monitoring network in place (Lubbock, Texas) using a bicycle platform. The monitoring campaign resulted in 30 days of data along a 13.4 km fixed concentric route. Due to high sensitivities to airflow, the apparent wind velocity was accounted for throughout the route. The data were also normalized into percentiles in order to visualize spatial patterns.The highest estimated pollution levels were located near frequently busy intersections and roads; however, sensor issues resulted in lower confidence. Additional research is needed concerning the appropriate use of low-cost metal oxide sensors for citizen science applications, as measurements can be misleading if the user is unaware of sensors specifications. The simultaneous use of several low-cost mobile platforms, rather than a single platform, as well as the use of high-end cases, are recommended to create a more robust spatial analysis. The issues addressed from this research are important to understand for accurate and beneficial application of low-cost gaseous monitors for citizen science.

Indicative extent of humic and fulvic acids in soils determined by electronic nose

The principal humic substances present in soils like Humic Acid (HA) and Fulvic Acid (FA) are plant fertility components many of which basically are aromatic in nature. Assaying these components is a challenge. The present study aims at an enquiry to delineate aromaticity of soil extracted HA and extracted FA by capturing their odor. Since these soil components are bound to soil particles, an effort has been put to find out whether they are more released in dry or wet soil conditions. The aromaticity determination was explored by Electronic nose (E-nose) equipment that had specific array of metal oxide sensors. The other objective of the study was to profile the aromaticity of HA and FA by their extraction from garden soil and its subsequent amendment. The series of experiments comprised of extraction of HA and FA and profiling of Norm Aroma Index (NAI) by E-nose in both wet and dry conditions of soil. The findings were that the yield of HA was more compared to FA; the NAI noted was high in wet condition of the soil compared to dry soil; HA showed more aromaticity compared to FA; the amended soil showed more aromaticity compared to non-amended normal soils. Among the amendment exercises, HA amended soil showed maximum aromaticity. E-nose employment could indicate extent of HA and FA in soils.

Effect of Unsaturated Sn Atoms on Gas-Sensing Property in Hydrogenated SnO2 Nanocrystals and Sensing Mechanism

Sensing reaction mechanism is crucial for enhancing the sensing performance of semiconductor-based sensing materials. Here we show a new strategy to enhancing sensing performance of SnO2 nanocrystals by increasing the density of unsaturated Sn atoms with dangling bonds at the SnO2 surface through hydrogenation. A concept of the surface unsaturated Sn atoms serving as active sites for the sensing reaction is proposed, and the sensing mechanism is described in detail at atomic and molecule level for the first time. Sensing properties of other metal oxide sensors and catalytic activity of other catalysts may be improved by using the hydrogenation strategy. The concept of the surface unsaturated metal atoms serving as active sites may be very useful for understanding the sensing and catalytic reaction mechanisms and designing advanced sensing sensors, catalysts and photoelectronic devices.

Differentiation of proteins and cancer cells using metal oxide and metal nanoparticles-quantum dots sensor array

A novel nanoparticles-quantum dots-based fluorescence sensor array has been developed for sensing proteins and cancer cells. In this study, six types of nanoparticles (NPs, including CuO, ZnO, Eu2O3, AuNPs, AgNPs, Au-Ag core-shell） conjugated with CdSe quantum dots (QDs) were used as sensing elements to create the multi-sensors of array. The fluorescence of quantum dots was efficiently quenched by nanoparticles. Fluorescence turn-on or further quenching could be observed due to the disruption of the nanoparticles-QDs interaction by proteins, affording distinct fluorescence response patterns. Linear discriminant analysis (LDA) was used to analyze the patterns. Moreover, protein quantification could be performed according to the fluorescence intensity on a specific or single sensing element. The linear dynamic ranges of the sensor array for proteins was in the range of 2–50μM, and the limits of detection (LODs) was below 2μM that varied for different proteins. Six cancer cells were also successfully differentiated by their fluorescence response patterns. This work indicates that the nanoparticles-QDs-based sensor array has a great potentials to be used in such fields as proteomics and medical diagnostics.

Enhanced sensing performance and sensing mechanism of hydrogenated NiO particles

Hydrogenation is successfully used to enhance the sensitivity of NiO-based gas sensors for the first time. The hydrogenated NiO sensors show significantly higher response to volatile-organic-compound vapors than the samples without hydrogenation. The enhanced sensing performance is attributed to an increase in the density of the unsaturated Ni atoms with dangling bonds on the surface resulted from the hydrogenation process. A concept of the unsaturated Ni atoms with dangling bonds serving as the active sites for the sensing reaction thus is proposed, and the sensing mechanism is described in detail at atomic and molecule level for the first time. The hydrogenation strategy may be used to improve sensing properties of other metal oxide sensors and catalytic activities of photocatalysts and other catalysts. The concept of the unsaturated metal atoms with dangling bonds serving as active sites can deepen understanding of the sensing and catalytic reaction mechanisms and afford guidance to design advanced gas sensing materials, catalysts and photoelectronic devices.

Room Temperature Sensing Achieved by GaAs Nanowires and oCVD Polymer Coating.

Novel structures comprised of GaAs nanowire arrays conformally coated with conducting polymers (poly(3,4-ethylenedioxythiophene) (PEDOT) or poly(3,4-ethylenedioxythiophene-co-3-thiophene acetic acid) display both sensitivity and selectivity to a variety of volatile organic chemicals. A key feature is room temperature operation, so that neither a heater nor the power it would consume, is required. It is a distinct difference from traditional metal oxide sensors, which typically require elevated operational temperature. The GaAs nanowires are prepared directly via self-seeded metal-organic chemical deposition, and conducting polymers are deposited on GaAs nanowires using oxidative chemical vapor deposition (oCVD). The range of thickness for the oCVD layer is between 100 and 200 nm, which is controlled by changing the deposition time. X-ray diffraction analysis indicates an edge-on alignment of the crystalline structure of the PEDOT coating layer on GaAs nanowires. In addition, the positive correlation between the improvement of sensitivity and the increasing nanowire density is demonstrated. Furthermore, the effect of different oCVD coating materials is studied. The sensing mechanism is also discussed with studies considering both nanowire density and polymer types. Overall, the novel structure exhibits good sensitivity and selectivity in gas sensing, and provides a promising platform for future sensor design.

Non-invasive breath monitoring with eNose does not improve glucose diagnostics in critically ill patients in comparison to continuous glucose monitoring in blood

Continuous glucose monitoring (CGM) can be beneficial in critically ill patients. Current CGM devices rely on subcutaneous or blood plasma glucose measurements and consequently there is an increased risk of infections and the possibility of loss of blood with each measurement. A potential method to continuously and non-invasively measure blood glucose levels is using exhaled breath. A correlation between blood glucose levels and volatile organic compounds (VOCs) in the exhaled breath was already reported. VOCs can be analyzed continuously using a so-called electronic nose (eNose). We hypothesize that continuous exhaled breath analysis using an eNose can be used to accurately predict blood glucose levels in intubated, mechanically ventilated ICU-patients. Mechanically ventilated patients whose blood glucose concentration was monitored with a CGM device were eligible. An eNose with four metal oxide sensors was used to continuously measure changes in exhaled breath. After pre-processing the data, several regression models were trained, consisting of: (1) only eNose sensor values; (2) only the 1st and 2nd principal components (PC) of eNose values; (3) eNose sensor values and last known blood glucose value as random effect; (4) 1st and 2nd PC of eNose sensor values and CGM value of one minute ago as fixed effect; (5) CGM value of one minute ago as fixed effect. Model performance was measured using the R2 value, the akaike information criterion and the Clarke error grid. Twenty-three patients were included in the study and 1165 hours of measurements were collected. Performance was low in models 1, 2 and 3 with a mean R2 of 0.07 [95%-CI: 0.00–0.28], 0.10 [95%-CI: 0.00–0.40] and 0.30 [0.02–0.79], respectively. Performance in models 4 and 5 was better with a mean R2 of 0.77 [0.02–1.00]. Subsequently, eNose data in model 4 had no added value over using CGM only in model 5. Continuous exhaled breath analysis using this eNose cannot be used to accurately predict blood glucose levels in intubated, mechanically ventilated ICU-patients.

Concentration Estimator of Mixed VOC Gases Using Sensor Array With Neural Networks and Decision Tree Learning

This paper aims to estimate the concentrations of volatile organic compounds (VOCs), such as ethanol, acetone, formaldehyde, and toluene, in a quaternary mixture. In order to develop an electronic nose for practical applications, the concentrations of VOCs and their combinations in the mixture are randomly assigned. The sensor array consists of five metal oxide sensors, which are produced in a laboratory. The algorithm for the estimation of concentration is realized using a back-propagation neural network (BPNN) with two hidden layers and decision tree learning. First, the data set of the VOC mixture is divided into four subclasses based on the concentration using classification and regression trees. Second, every subclass is classified and regressed using a corresponding BPNN; furthermore, its four output nodes provide a continuous prediction of the concentration of each VOC in the mixture. A single BPNN with two hidden layers is also constructed and evaluated for the purpose of comparison. The maximum error in the concentration estimation of each VOC using the proposed method is approximately 2 ppm, and the accuracy is better than the result obtained using the single BPNN. Moreover, the relative error is less than 5% when the predicted concentration is higher than 20 ppm. This paper reports some aspects of the potential of using neural networks for quantitatively analyzing the concentrations of a VOC mixture.