The detection of partial discharge through the analysis of SF6 gas components in gas-insulated switchgears is important for the diagnosis and assessment of the operational state of power equipment. A gas sensor based on anatase TiO2 is used to detect SF6 decomposed gases. In this paper, first principles density functional theory calculations are adopted for the theoretical investigation of the process of anatase (1 0 1) surface adsorbing SO2, SOF2 and SO2F2, which are the main components of SF6 decomposed gases. Simulation results show that the surface of anatase (1 0 1) is more sensitive and selective to SO2 than to SOF2 and SO2F2. This theoretical finding is corroborated by the sensing experiment using a TiO2 nanotube arrays gas sensor. The interpretation of the sensing mechanism is based on the simulation results. The study concludes that, under the same temperature and concentration, the response R% of the sensor to the three gases depends on the impurity state introduced by adsorbed gas molecules. (C) 2013 Elsevier B.V. All rights reserved.