Insights into\nTarget Gas–Oxygen Interactions\nin Highly Sensitive Gas Sensors Using Data-Driven Methods

In the gas-sensing mechanism of a metal-oxide-semiconductor\n(n-type)\ngas sensor, oxygen adsorption or desorption on the oxide surface leads\nto an increase or decrease in the resistance of the gas sensor system.\nAdditionally, oxygen can be adsorbed again at the location where initially\nadsorbed oxygen reacted with the target gas. Thus, the adsorption–desorption\nequilibrium of the reducing gas on the oxide surface is a significant\nfactor in determining the sensitivity and reaction rate. In particular,\nfor ultralow-concentration gas measurements, the relative concentration\nof oxygen was very high. To design an ultrasensitive gas sensor, not\nonly the reaction of the target gas but also the competing reaction\nbetween the target gas and oxygen must be considered. Although qualitative\ninvestigations of these complex relationships have been performed\naccording to the gas concentration and flow rate, reliable quantitative\nresults are limited. In this study, a quantitative approach was used\nto understand the correlation between oxygen and a target gas by applying\ndata analysis methods. We investigated the behavior of oxygen and\nthe target molecules depending on the gas concentration and flow rate\nusing the parts per billion level of the acetone gas sensor. Initial\nresponse data according to various detection conditions were processed\nusing principal component analysis and K-means clustering; as a result,\nfour types of reaction behaviors were inferred for 15 types of reaction\nconditions. Furthermore, the response time, depending on the detection\nconditions, can be distinguished using the suggested categorization.\nOur investigation suggests a possibility beyond simple optimization\nthrough the data analysis of the gas-sensing results.