Understanding UV sensor performance in ZnO TFTs through the application of multivariate analysis

Abstract

Zinc oxide (ZnO) thin film transistors are well suited to UV sensing application because they absorb predominantly in the UV region due to the wide bandgap (Eg = 3.37 eV) and possess a large exciton binding energy (60 meV) with high radiation hardness. When operated as a transistor, many device performance parameters alter such as threshold Voltage, on-off current and channel mobility. As a result, it is to distinguish between changes in electrical performance induced by UV light and environmental effects that add noise to the sensor performance. In this work, the UV response of zinc oxide thin film transistors (ZnO TFTs) is examined using Taguchi Design of Experiment (DOE) method. By using this multivariate analysis approach, it is possible to reduce the number of calibration tests required for the sensor to accurately assess UV irradiation It is observed that different input conditions (UV power, exposure time, temperature, bias conditions) affect different TFT performance parameters more or less significantly. From the perspective of UV sensing, ON current in the saturation region appears to be the best performance parameter in a ZnO TFT for examining differences in UV exposure.