Carbon nanotube thin-film-transistors for gas identification

Single-walled carbon nanotube (CNT) based gas sensors have enormous potential in pollution monitoring in low concentration level because of its high sensitivity, fast response, and physical/chemical stability. However, the lack of selectivity has been a major drawback for its wide range employment. In this work, we fabricate thin film transistors (TFTs) using randomly distributed CNTs and investigate them for ammonia and nitrogen dioxide detection in air at low ppm concentrations. A sensing mechanism is proposed based on the interaction between gas molecules and different types of dwelling spots inside the channel area of a TFT. We present double exponential-convolution model to decipher sensor response as well as to explore its application in gas identification. In this context, the consistency in time constants is recognized, which is independent of gas concentration. More importantly, the time constants vary with respect to different gas types and TFTs. The uniqueness of time constants can work as identity verification for different sensing gases, which demonstrates that the sensor response is a distinctive behavior determined by the unique channel structure of each TFT. This work provides us a general strategy for gas identification in ppm level and a practical path to realize the advantages of CNT gas sensors in air quality detection as well as the industrial emission control.