P2AR.3 - Virtual Gas Sensor Array by Cyclic Optical Activation: Optimization of Activation Profile by Machine Learning
- Event
- 17th International Meeting on Chemical Sensors - IMCS 2018
2018-07-15 - 2018-07-19
Vienna, Austria - Chapter
- P-2 - Sensor Arrays
- Author(s)
- G. Njio, T. Wagner - Paderborn University, Department of Chemistry, Paderborn (Germany)
- Pages
- 905 - 906
- DOI
- 10.5162/IMCS2018/P2AR.3
- ISBN
- 978-3-9816876-9-9
- Price
- free
Abstract
Optical activation of semiconducting metal oxide based gas sensors offers new ways of improving the performance of these cost-efficient and highly sensitive devices. Exposure to light in the UV/VIS range - as opposed to high temperature operation - is a low energy alternative to speed-up reaction kinetics and to balance adsorption-desorption equilibrium. Yet no study explicitly utilizes optical activation to compensate for the inherent lack in selectivity of the semiconducting sensors. In the following we present a new approach for generating a virtual gas sensor array by cyclic optical activation and a method to optimize activation profiles by utilizing supervised learning algorithms. The presentation is focused on nanostructured indium oxide (In2O3). However, the described method can be applied to various oxides used for semiconducting sensing as e.g. tin oxide or tungsten oxide. Different activation parameters like shape of the intensity curve, the intensity itself as well as the duration of the light exposure are evaluated by a neural network. Results are used to implement a signal stabilization algorithm for humidity and ozone sensing utilizing In2O3.