Discrimination of VOCs molecules via extracting concealed features from a temperature-modulated p-type NiO sensor

Poor selectivity of metal oxide semiconductor (MOS) gas sensors (toward volatile organic compounds, VOCs) poses a significant challenge for their applications in the emerging areas of personal health and air quality monitoring. Extensive efforts have been devoted to improving the selectivity of gas sensors via extracting features from their electrical response signals. Alternative to the conventional strategy of enlarging the number of sensor arrays, analyzing the transient signal of a temperature modulated gas sensor provides an efficient approach to extract molecule features. Despite p-type MOS outperforms n-type counterpart in terms of (photo)catalytic properties, further exploration on thermal modulation of p-type MOS sensor has been scarcely reported. In this work, p-type NiO nanoparticles with grain size of 17.4 ± 4.0 nm have been synthesized with the assistance of bacterial cellulose (BC) scaffold. Transient response characteristics of NiO sensor (modulated by a staircase waveform) toward 5 kinds of VOCs have been investigated. The removals of irrelevant electrical signals, particularly induced by large temperature coefficient of resistance (TCR) of p-NiO, allows us to extract the intrinsic features of tested VOCs molecules by discrete wavelet transform (DWT). Successful classification and recognition of tested VOCs molecules, including three kinds of benzene series (benzene, toluene and chlorobenzene), have been achieved by typically non-selective p-type NiO sensor with a low sensitivity. Our work highlights that eliminating the irrelevant thermally modulated electric signals is essential for expanding the recognition capability of a single MOS sensor (toward VOCs molecules), and sheds light on the exploring future smart gas molecule recognition chips.