Executive Summary : | Semiconducting metal oxides (SMO) are widely used sensing materials due to their ability to detect a wide range of gases. These materials, such as ZnO, SnO2, TiO2, WO3, NiO, and V2O5, work on the variation of surface electrical conductance in a gaseous environment. Medium to wide bandgap oxide semiconductors are ideal for gas sensing due to their intrinsic donors/acceptors in the bandgap. However, bulk SMOs have low selectivity among gases and high-power consumption due to high operating temperatures. To address this issue, researchers propose designing nanostructure-based chemical sensors with ultra-small device footprints. The proposed sensor array can detect multiple analytes uniquely using multivariate data analysis tools, such as Principal component analysis (PCA). This can detect gases like volatile organic compounds, humidity, ammonia, NOx, and even biomolecules like glucose and lactose. This sensor array could be used for various applications, including monitoring body vitals, detecting air pollutants, process control in industries, space, and defense.
The proposed method involves growing oxide 2D layers through chemical synthesis and a scalable RF magnetron sputtering method. The large data set is screened using machine learning methods to evaluate the best oxide candidates and their sensor characterization. The final outcomes are processed using traditional PCA algorithms integrated with machine learning techniques, including neural networks, to select the desired gas for the end use application. |
Co-PI: | Dr. Joy Mitra, Indian Institute of Science Education and Research (IISER) Thiruvananthapuram, Kerala-695551, Dr. Nongmaithem Sadananda Singh, Indian Institute of Science Education and Research (IISER) Thiruvananthapuram, Kerala-695551, Dr. Sheetal Shashikant Dharmatti, Indian Institute of Science Education and Research (IISER) Thiruvananthapuram, Kerala-695551 |