INCREASING CHEMICAL SELECTIVITY OF CARBON NANOTUBE-BASED SENSORS BY FLUCTUATION-ENHANCED SENSING

Nowadays gas detection in the ppm and sub-ppm domain is essential in terms of environmental protection as well as reducing sanitary risks. However, detecting systems to perform these measurements (e. g., gas chromatographs) are expensive and take up too much space, thus their use is not likely to become wide-spread. Small, cheap and easily mountable sensors, such as resistive sensors are more applicable for this purpose. But the main disadvantage of these sensors is the lack of chemical selectivity. Yet, a novel method called fluctuation-enhanced sensing (FES), which considers the sensor noise as the source of chemical information, can be used to improve selectivity. Since carbon nanotube (CNT)-based sensors are regarded as promising devices for FES measurements, we investigated whether stationary fluctuations in output signal (dc-resistance) of a CNT sensor could be used to increase chemical selectivity. In this work we prove that FES is applicable to increase selectivity of CNT sensors: air polluting gases (N2O, NH3 and H2S) and their mixtures can be distinguished. Furthermore, we also show that different concentrations of the same analyte can be differentiated and chemical selectivity can be extended into the sub-ppm region.