Metal Oxide Semiconductor Sensors Research Articles

A comparative study of machine learning models for identifying noxious gases through thermal fingerprint measurements and MOS sensors

Recently, there has been a notable surge of interest in developing swift and precise gas detection and categorization tools through artificial intelligence. This work specifically focuses on designing a simple sensor array configuration and empirically compares various machine-learning models for identifying multiple gases. The study employs eight metal oxide semiconductor sensors, each operating independently based on the thermal fingerprint principle. These sensors are utilized to gather data on responses to different toxic gases, and a range of traditional machine learning classifiers are trained and assessed using this dataset. The classifiers undergo investigation across various train/test dataset split ratios and different principal component analysis dimensionalities. Average accuracy metrics are employed over the entire testing dataset to evaluate model performance. The results demonstrated the challenge of selectively detecting each gas type solely through conventional analyses with the gas response data of the proposed sensor array. Nevertheless, the utilization of machine learning models has exhibited promising outcomes in augmenting gas identification and classification. The findings reveal a consistent improvement in testing accuracy for the Gradient Boosting model, which achieves an outstanding accuracy rate, with instances of reaching 100% accuracy in the training dataset when three or more principal components are considered. This study highlights the potential of the proposed straightforward sensor array and presents effective recognition algorithms applicable to creating electronic noses for gas-sensing applications.

Distributed Embedded System for Air Quality Monitoring based on Long Range (LoRa) Technology

This study addresses the persistent challenge of urban air quality deterioration through the introduction of the Distributed Embedded System (DES). In response to limitations associated with conventional air quality measurement methods, the DES system offers a cost-effective and portable alternative for real-time monitoring. The study focuses on implementing a low-cost distributed system and developing effective communication protocols between sensor nodes. Equipped with Metal Oxide Semiconductor (MOS) sensors for pollutant gases, optical sensors for particulate matter, and meteorological sensors, the system strategically deploys multiple nodes within a 4 km range of urban areas. Real-time AQI and pollution severity have been measured for various locations. Implementation of Long Range (LoRa) communication technology allows the sensor nodes to efficiently transmit data to a central base station. Observations of received signal strength and signal quality indicate reliable and effective communication. To validate accuracy and reliability, acquired DES system data undergoes comparative analysis with data from a government-established meteorological station, revealing a strong correlation. This innovative approach presents a promising solution for widespread, continuous, and cost-effective real-time air quality monitoring in urban areas. The DES system addresses key challenges associated with air pollution, offering a portable and accessible tool that could revolutionize urban air quality management.

Atomically dispersed Ru on three-dimensionally-ordered macroporous In2O3 for highly sensitive detection of triethylamine and the freshness of seafood

As a volatile organic compound, triethylamine (TEA) is extensively employed in industry. Besides, seafood releases gases such as organic amines during the spoilage process, making the detection of TEA of great significance. In this work, three-dimensionally-ordered macroporous (3DOM) indium oxide (In2O3) sensing materials decorated with ruthenium (Ru) dopants were synthesized. The TEA-sensing properties of the Ru/In2O3 were significantly improved than those of the pure In2O3 and the results reported in the literature. Notably, the 0.5Ru/In2O3 sensor had a high response (∼1943) to 50 ppm TEA at 220 °C with good selectivity, long-term stability, and fast response time. In addition, the 0.5Ru/In2O3 sensor had a response value of 14 to a low concentration of 1 ppm TEA. The exceptional TEA-sensing performance was ascribed to the remarkable pore structure of the 3DOM material, facilitating the efficient adsorption and diffusion of TEA. The atomically dispersed Ru increased the amounts of active sites, resulting in enhanced adsorbed oxygen content, reduced activation energy, and improved sensing response. Consequently, incorporating relatively cheap Ru with the 3DOM structure presents a novel approach to enhance the TEA-sensitive performance of metal oxide semiconductor sensors.

Microwave-assisted hydrothermal synthesis and gas sensing properties of ZnSn(OH)6, ZnSnO3, and Zn2SnO4/SnO2 hierarchical nano-/hetero-structures

Although semiconducting metal oxide sensors present reasonable sensitivity, an improved lower detection limit and/or selectivity would allow broadening real-time monitoring applications. This work reports the growth mechanism and gas sensing performance of zinc tin oxide-based structures synthesised via a microwave-assisted hydrothermal route. The synthesised materials were characterised by X-ray diffraction (XRD), Raman and Fourier-transform infrared (FTIR) spectroscopy, scanning and scanning transmission electron microscopy (SEM and STEM), X-ray photoelectron spectroscopy (XPS), energy-dispersive X-ray spectroscopy (EDS), and nitrogen adsorption/desorption experiments. Gas sensor measurements showed that ZnSnO3 presents an outstanding lower detection limit to nitrogen dioxide (NO2), in which a 10-fold increase in electrical resistance is expected in the presence of 1 ppb NO2 at an operating temperature of 150 ˚C. Moreover, the Zn2SnO4/SnO2 heterostructure exhibited superior selectivity to NO2 relative to hydrogen (H2) and carbon monoxide (CO), exhibiting a sensor response ∼1500 times higher for the oxidising gas. Hence, it is demonstrated that nanostructures’ growth engineering can realise lower detection limits and ultra-selective high-performance gas sensor devices through a greater surface area and enhanced contact potential barriers.

New method for optimizing the detection of low-energy soft X-rays (200–300 eV) in a CMOS detector

Complementary metal-oxide semiconductor (CMOS) sensors have been widely used as soft X-ray detectors in several fields owing to their recent developments and unique advantages. The CMOS detector parameters have been extensively studied and evaluated. However, the methods for discriminating signals and dark noise have not been sufficiently studied, particularly for low-energy soft X-rays. Setting a threshold, which is widely used for discrimination, is not suitable for low-energy soft X-rays because of the crossing of the energy bands of dark noise and signals. In this study, we analyzed the charge correlation of the CMOS detector GSENSE2020BSI (G2020BSI) and proposed a new two-dimensional segmentation method to optimize the discrimination of signals according to the charge correlation. The effectiveness of the method used to discriminate low-energy soft X-rays on the G2020BSI detector was qualitatively evaluated. The optimal feature parameters used in the two-dimensional segmentation method were investigated for G2020BSI. However, the two-dimensional segmentation method is insensitive to feature parameters.

Polycrystalline hollow MOF derived Co3O4 semiconductor to achieve room-temperature ammonia detection in human exhaled breath

Functional gas-sensing devices, on their portable and non-invasive merits, have stimulated a new exploring in clinical detection of gaseous biomarkers for critical diseases such as hepatopathy and nephropathy. Metal oxide semiconductor (MOS) sensors owing to their high sensitivity and facile manufacture appears to be the most promising for integration into illness surveillance devices. However, the high operating temperature of MOS poses great challenge for their realistic application. Herein, MOF-derived Co3O4 sensor series are prepared by facile manipulation of the morphological structure and crystalline state of their maternal ZIF-67. It shows that polycrystalline hollow ZIF-67 (PH-ZIF-67) derived PH-Co3O4 exhibits the optimum response and selectivity to trace NH3 at room-temperature sensing. PH-Co3O4 has a low limit of detection and well moisture resistance, which further behaves a significant quantitative response to actual exhaled breath of patients with hepatic encephalopathy. Furthermore, the scientific link between performance enhancement and structural adjustment is revealed by XPS, EPR, etc. It is found that the construction of the polycrystalline structure increases the oxygen vacancies generated during calcination, and additionally, the hollow structure provides more chemisorbed oxygen on the extra inner surface, which coupled to enhance the number of reactive oxygen sites responsible for the response improvement.

Delafossite CuGaO2-Based Chemiresistive Sensor for Sensitive and Selective Detection of Dimethyl Disulfide.

Dimethyl disulfide (DMDS) is a common odor pollutant with an extremely low olfactory threshold. Highly sensitive and selective detection of DMDS in ambient humid air background, by metal oxide semiconductor (MOS) sensors, is highly desirable to address the increased public concern for health risk. However, it has still been a critical challenge up to now. Herein, p-type delafossite CuGaO2 has been proposed as a promising DMDS sensing material owing to its striking hydrophobicity (revealed by water contact angle measurement) and excellent partial catalytic oxidation properties (indicated by mass spectroscopy). The present CuGaO2 sensor shows a selective DMDS response, with satisfied humidity resistance performance and long-term stability at a relatively low operation temperature of 140 °C. An ultrahigh response of 100 to 10 ppm DMDS and a low limit of detection of 3.3 ppb could be achieved via a pulsed temperature modulation strategy. A smart sensing system based on a CuGaO2 sensor has been developed, which could precisely monitor DMDS vapor in ambient humid air, even with the presence of multiple interfering gases, demonstrating the practical application capability of MOS sensors for environmental odor monitoring.

MOF-derived CdS@c/h-In2O3 hollow hexagonal prism with surface-grown nanofibers for ppb-level formaldehyde detection: an insight into synergistic enhancement of endogenous phase-homojunction and heterojunction

Detection of ppb-level formaldehyde at low temperatures is a great pursuit for metal oxide semiconductor sensors. To realize this goal, metal organic framework (MOF) derived hollow hexagonal prism with surface-grown nanofibers were successfully prepared and modified using quantum dots, affording a 3D/0D heterojunction with excellent sensing performance. The CdS@c/h-In2O3-2 hollow hexagonal prism exhibits outstanding sensing properties (107.37 for 100 ppm), ideal selectivity, and low detection limit (160 ppb) at 160 °C. Moreover, the possible formation mechanism of the hollow hexagonal prism and endogenous phase-homojunction are proposed. Further, an insight into the synergistic enhancement of endogenous phase-homojunction and heterojunction is presented, including the gas diffusion, density functional theory (DFT) calculations, and resistance-controlled mechanism. The distinguished HCHO-sensing performance can be attributed to MOF-derived endogenous phase-homojunction, 3D/0D heterostructure, hollow morphology, and confined internal space, which ultimately synergize to strengthen the adsorption and electron transfer behavior of oxygen and HCHO molecules. Overall, this work provides new insights into the feasible strategy and theoretical viewpoint concerning high-performance HCHO sensors.

Ambient Monitoring Portable Sensor Node for Robot-Based Applications.

The leakage of gases and chemical vapors is a common accident in laboratory processes that requires a rapid response to avoid harmful effects if humans and instruments are exposed to this leakage. In this paper, the performance of a portable sensor node designed for integration with mobile and stationary robots used to transport chemical samples in automated laboratories was tested and evaluated. The sensor node has four main layers for executing several functions, such as power management, control and data preprocessing, sensing gases and environmental parameters, and communication and data transmission. The responses of three metal oxide semiconductor sensors, BME680, ENS160, and SGP41, integrated into the sensing layer have been recorded for various volumes of selected chemicals and volatile organic compounds, including ammonia, pentane, tetrahydrofuran, butanol, phenol, xylene, benzene, ethanol, methanol, acetone, toluene, and isopropanol. For mobile applications, the sensor node was attached to a sample holder on a mobile robot (ASTI ProBOT L). In addition, the sensor nodes were positioned close to automation systems, including stationary robots. The experimental results revealed that the tested sensors have a different response to the tested volumes and can be used efficiently for hazardous gas leakage detection and monitoring.

Noninvasive Diabetes Detection through Human Breath Using TinyML-Powered E-Nose.

Volatile organic compounds (VOCs) in exhaled human breath serve as pivotal biomarkers for disease identification and medical diagnostics. In the context of diabetes mellitus, the noninvasive detection of acetone, a primary biomarker using electronic noses (e-noses), has gained significant attention. However, employing e-noses requires pre-trained algorithms for precise diabetes detection, often requiring a computer with a programming environment to classify newly acquired data. This study focuses on the development of an embedded system integrating Tiny Machine Learning (TinyML) and an e-nose equipped with Metal Oxide Semiconductor (MOS) sensors for real-time diabetes detection. The study encompassed 44 individuals, comprising 22 healthy individuals and 22 diagnosed with various types of diabetes mellitus. Test results highlight the XGBoost Machine Learning algorithm's achievement of 95% detection accuracy. Additionally, the integration of deep learning algorithms, particularly deep neural networks (DNNs) and one-dimensional convolutional neural network (1D-CNN), yielded a detection efficacy of 94.44%. These outcomes underscore the potency of combining e-noses with TinyML in embedded systems, offering a noninvasive approach for diabetes mellitus detection.

Predicting the quality attributes related to geographical growing regions in red-fleshed kiwifruit by data fusion of electronic nose and computer vision systems

The ability of a data fusion system composed of a computer vision system (CVS) and an electronic nose (e-nose) was evaluated to predict key physiochemical attributes and distinguish red-fleshed kiwifruit produced in three distinct regions in northern Iran. Color and morphological features from whole and middle-cut kiwifruits, along with the maximum responses of the 13 metal oxide semiconductor (MOS) sensors of an e-nose system, were used as inputs to the data fusion system. Principal component analysis (PCA) revealed that the first two principal components (PCs) extracted from the e-nose features could effectively differentiate kiwifruit samples from different regions. The PCA-SVM algorithm achieved a 93.33% classification rate for kiwifruits from three regions based on data from individual e-nose and CVS. Data fusion increased the classification rate of the SVM model to 100% and improved the performance of Support Vector Regression (SVR) for predicting physiochemical indices of kiwifruits compared to individual systems. The data fusion-based PCA-SVR models achieved validation R2 values ranging from 90.17% for the Brix-Acid Ratio (BAR) to 98.57% for pH prediction. These results demonstrate the high potential of fusing artificial visual and olfactory systems for quality monitoring and identifying the geographical growing regions of kiwifruits.

THE FEASIBILITY OF USING METAL OXIDE SEMICONDUCTOR (MQ2) IN CREATING A GAS LEAK DETECTOR

Gas leaks and the danger they bring with them remain a primary concern for many despite the advances in current technology. The challenges people face in preventing gas leakage in their environment are the knowledge about gas leaks and the expensive tools and countermeasures they need to purchase in return for their safety. Thus, the study utilized experimental research wherein the main objective of the study was to create a gas leak detector using a Metal Oxide Semiconductor (MQ2) as the primary component and interfaced with Arduino UNO to activate the alarm system of the device. The study's results proved the feasibility of the Metal Oxide Semiconductor sensor as the main component in making a gas leak detector in terms of the distance the device can detect gas leakage, detecting up to 2 meters. Second, the amount of time it takes for the sensor to detect a gas leak, with averages of 4.62 seconds for 0.5 meters and 9.39 seconds for 1 meter. The researchers determined that the gas leak detector can detect gas leaks up to 2 meters with an interval of half a meter, as all the trials that were done resulted in a passing rate.

Pattern recognition system for rapid detection of gases using microfluidic olfaction detector: A case study using methane and ethane

Two significant shortcomings of metal oxide semiconductor (MOS) sensors are their low selectivity and prolonged recovery time after exposure to target gases. Microfluidic olfaction detectors (MFOD), having an embedded MOS at the distal end of a microchannel, improve MOS selectivity but further prolong the recovery time due to the delay in the gas movement as the result of diffusion through the microchannel. This paper aims to reduce the sampling time of a single MFOD. Such reduction is achieved here by developing a pattern recognition system involving: (i) an ensemble of four heterogeneous base machine learning models, (ii) a Long-Short Term Memory Neural Network (LSTM) to construct a portion of the transient response, (iii) features from an additional fast response sensor, and (iv) detachment of MOS from the microchannel for accelerated recovery. Using data collected from methane and ethane as a case study, the results of classification and regression demonstrate the need for only 17.5% of the previously used response time without any loss in accuracy. This reduces the total testing cycle of an MFOD by 60%, further enhancing the ability of selective MFODs by enabling faster detection of gases.

Promoting quality in low-cost gas sensor devices for real-world applications

Recent advancements in metal oxide semiconductor sensors enable system integrators to make sensor devices without the traditional complications of operating barebone sensor components. Anyone, so it seems, can put together multi-sensor-systems by combining an integrated sensor subsystem with a small piece of digital infrastructure, resulting in low-cost sensor systems or sensor networks for all kinds of applications. With low energy consumption, the deployment of close meshed sensor networks is becoming a reality, promising high density data for big data models. However, data quality is not necessarily a feature of such devices, since the device output signals are heavily processed and the insight into the actual operating technique is black-boxed by the respective sensor manufacturer. High volume production of a sensor model requires widely applicable output information, yet this mainstreaming negatively impacts quality control efforts on the system level as well as niche applications that require specialized operation modes. This article exemplarily examines the measurement chain of typical metal oxide semiconductor sensor applications and deduces requirements for a technically sound advancement of the subject for research and commercial purposes. Equivalent considerations can be made for any low-cost sensor principle and their respective challenges. The conclusion urges all stakeholders participating in the development and marketing of sensor devices to advance a scientifically valid state of knowledge and to educate their customers and even the general public accordingly.

Optical Biosensing of Polarized Light

Interactions between liquid crystal molecules and target analytes open up various biosensing applications for quick screening and point-of-care applications. In this review, we categorized biosensors by type, depending on the liquid crystal mesophase, and considered several applications for the detection of biomolecules, point-of-care diagnostics and environmental monitoring. We also discuss interactions between polarized light and target pathogens dispersed in biological fluids, which result in the change of the polarization state. An array of the Stokes parameters can be compared with the pattern, and a proper pathogen can be manifested. We suggest that a combination of a micropolarizer array and a complementary metal oxide semiconductor sensor is an optimal setup for the detection of pathogens. Herein, we discuss the working principles of liquid crystal biosensors and their fabrication principles. In addition, relevant theoretical and practical issues related to liquid crystal biosensors are outlined. In general, this review gives an in-depth survey of the research on liquid crystal-based sensors, making it easier for researchers to locate their niche and make contributions to this subject from multiple viewpoints.

Gas sensor-based machine learning approaches for characterizing tarragon aroma and essential oil under various drying conditions

Aroma is one of the most significant quality traits for many pharmaceutical plants and their products as it indicates the quality of the raw material. The aroma may be lowered to imperceptible levels, altered, or damaged during the processing of herbs, such as drying or fermentation. Using an electronic nose (e-nose) is one of the most efficient approaches for identifying and evaluating the aroma of essential oils (EO). In this study, tarragon was dried in a hybrid dryer designed explicitly for frying the herbs at four temperatures (40, 50, 60, and 70 °C) and three air velocities (1, 1.5, and 2 m/s). After extracting its EO, the purity of the tarragon EO was assessed by an e-nose comprising nine metal oxide semiconductor (MOS) sensors. The highest EO levels (0.359) was obtained upon drying at 40 °C. By increasing the temperature from 40° to 70°C, the EO was declined, and its lowest level (0.26) was assessed at 70 °C. multivariate data analysis and artificial neural networks modeling were also employed to quantify and classify the obtained EOs based on the output of the sensor. Multivariate discrimination analysis (MDA) and Quadratic discriminant analysis (QDA) offered a 100 % accuracy in classifying 12 groups of EO.

Evaluation of the applicability of a metal oxide semiconductor gas sensor for methane emissions from agriculture

This work investigated the potential of metal oxide semiconductor (MOS) gas sensors for environmental monitoring of methane. Calibrations were performed under controlled conditions in the lab, and under semi-controlled conditions in the field, using a modified head space chamber set-up. Concentrations up to ±300 ppm methane were tested. The relationship between sensor conductance and methane concentrations could be very well described using principles from adsorption theory. The adjustable parameters were background conductance G0, a sensitivity constant S and a non-ideality coefficient n, where n has a non-rational value between 0 and 1. Sensor behaviour was very different in dry air than in humid air, with the background conductance increasing approximately tenfold and sensitivity decreasing between 20 fold and 80 fold, while the non-ideality coefficient increased from ±0.4 to ±0.6. Nevertheless, at high methane concentrations comparable conductance values were recorded in dry and humid air. The standard deviation of predicted values was 1.6 μS.for the least well described dataset. Using the corresponding calibration curve, a detection limit of 11 ppm is calculated for humid ambient air. This values suggests that MOS sensor are adequately sensitive to be used for methane detection in an agricultural context.

An Aluminum-coated sCMOS Sensor for X-Ray Astronomy

In recent years, tremendous progress has been made on scientific Complementary Metal Oxide Semiconductor (sCMOS) sensors, making them a promising device for future space X-ray missions. We have customized a large-format sCMOS sensor, G1516BI, dedicated for X-ray applications. In this work, a 200 nm thick aluminum layer is successfully sputtered on the surface of this sensor. This Al-coated sensor, named EP4K, shows consistent performance with the uncoated version. The readout noise of the EP4K sensor is around 2.5 e− and the dark current is less than 0.01 e− pixel−1 s−1 at −30°C. The maximum frame rate is 20 Hz in the current design. The ratio of single pixel events of the sensor is 45.0%. The energy resolution can reach 153.2 eV at 4.51 keV and 174.2 eV at 5.90 keV at −30°C. The optical transmittance of the aluminum layer is approximately 10−8 to 10−10 for optical lights from 365 to 880 nm, corresponding to an effective aluminum thickness of around 140 to 160 nm. The good X-ray performance and low optical transmittance of this Al-coated sCMOS sensor make it a good choice for space X-ray missions. The Lobster Eye Imager for Astronomy, which has been working in orbit for about one year, is equipped with four EP4K sensors. Furthermore, 48 EP4K sensors are used on the Wide-field X-ray Telescope on the Einstein Probe satellite, which will be launched at the end of 2023.

Image Reconstruction and Investigation of Factors Affecting the Hue and Wavelength Relation Using Different Interpolation Algorithms with Raw Data from a CMOS Sensor

An image processing method was employed to obtain wavelength information using light irradiated during camera exposure. Physically, hue (H) and wavelength (W) are closely related. Once the H value is known through image pixel analysis, the wavelength can be obtained. In this paper, the H-W curve was investigated from 400 to 650 nm using raw image data with a complementary metal oxide semiconductor (CMOS) sensor technology. We reconstructed the H-W curve from raw image data based on a demosaicing method with 2 × 2 pixel images. To date, no study has reported on reconstructing the H-W curve using several different interpolation algorithms in the 400~650 nm wavelength region. In addition, several factors affecting the H-W curve with a raw digital image, such as exposure time, aperture, and international organization for standardization (ISO) settings, were investigated for the first time.

Preparation and enhanced acetone sensing property of flower-like Sn-doped Fe2O3

Metal oxide semiconductor (MOS) nanostructures are used widely in acetone sensors, but pure MOS sensors usually have poor selectivity and low sensitivity. In this study, various Sn-doped Fe2O3 (Sn-Fe2O3) microflowers were obtained using FeOOH microflowers as precursors using a liquid hydrolysis reaction followed by a calcination process. Various characterization techniques verified the morphologies and composition of the products. The microflower-like 17.0 wt% Sn-Fe2O3 possessed a specific surface area of 138.9 m2 g−1. Employed as a sensing material, the 17.0 wt% Sn-Fe2O3 microflowers exhibited a strong response of 107.7 for 100 ppm acetone with short response/recovery times of 8/12 s. Furthermore, the 17.0 wt% Sn-Fe2O3 microflower sensor displayed good selectivity and high stability for acetone vapor with a detection limit of 114 ppb. The in-situ Raman spectrum verified that the Sn-Fe2O3 microflowers improved the adsorption of the acetone molecules, resulting in enhanced sensing performance.