Metal oxide-based gas sensor array for VOCs determination in complex mixtures using machine learning.

Detection of volatile organic compounds (VOCs) from the breath is becoming a viable route for the early detection of diseases non-invasively. This paper presents a sensor array of 3 component metal oxides that give maximal cross-sensitivity and can successfully use machine learning methods to identify four distinct VOCs in a mixture. The metal oxide sensor array comprises NiO-Au (ohmic), CuO-Au (Schottky), and ZnO-Au (Schottky) sensors made by the DC reactive sputtering method and having a film thickness of 80-100 nm. The NiO and CuO films have ultrafine particle sizes of < 50 nm and rough surface texture, while ZnO films consist of nanoscale platelets. This array was subjected to various VOC concentrations, including ethanol, acetone, toluene, and chloroform, one by one and in a pair/mix of gases. Thus, the response values show severe interference and departure from commonly observed power law behavior. The dataset obtained from individual gases and their mixtures were analyzed using multiple machine learning algorithms, such as Random Forest (RF), K-Nearest Neighbor (KNN), Decision Tree, Linear Regression, Logistic Regression, Naive Bayes, Linear Discriminant Analysis, Artificial Neural Network, and Support Vector Machine. KNN and RF have shown more than 99% accuracy in classifying different varying chemicals in the gas mixtures. In regression analysis, KNN has delivered the best results with an R2 value of more than 0.99 and LOD of 0.012 ppm, 0.015 ppm, 0.014 ppm, and 0.025 ppm for predicting the concentrations of acetone, toluene, ethanol, and chloroform, respectively, in complex mixtures. Therefore, it is demonstrated that the array utilizing the provided algorithms can classify and predict the concentrations of the four gases simultaneously for disease diagnosis and treatment monitoring.

Frequency dependent impedance response analysis of nanocrystalline ZnO chemiresistors.

ZnO is a widely studied gas sensor material and is used in many commercial sensor devices. However, selectivity towards any particular gas remains an issue due to lack of complete knowledge of the gas sensing mechanism of oxide surfaces. In this paper, we have studied the frequency dependent gas sensor response of ZnO nanoparticles of a diameter of nearly 30 nm. A small rise of synthesis temperature from 85 °C to 95 °C in the solvothermal process, shows coarsening by joining and thereby distinct loss of grain boundaries as seen from transmission electron micrographs. This leads to a substantial reduction in impedance, Z (GΩ to MΩ), and rises in resonance frequencyfres(from 1 to 10 Hz) at room temperature. From temperature dependent studies it is observed that the grain boundaries show a Correlated Barrier Hopping mechanism of transport and the hopping range in the grain boundary region is typically 1 nm with a hopping energy of 153 meV. On the other hand, within the grain, it shows a change of transport type from low temperature tunneling to beyond 300 °C as polaron hopping. The presence of disorder (defects) as the hopping sites. The temperature dependence offresagrees with different predicted oxygen chemisorbed species between 200 °C to 400 °C. As opposed to the traditional DC response, the AC response in the imaginary part of (Z″) shows gas specific resonance frequencies for each gas, such as NO2, ethanol, and H2. Among the two reducing gases, ethanol and hydrogen; the former shows good dependence on concentration in Z″ whereas the latter shows a good response infresas well as capacitance. Thus, the results of frequency dependent response allow us to investigate greater details of the gas sensing mechanism in ZnO, which may be exploited for selective gas sensing.