This study describes the development of gas-sensitive resistors (chemiresistors) as sensors for the detection of nitroaromatic vapours. 2,4-DNT was chosen as the example of the nitroaromatic, and a vapour generator was designed and built to deliver a vapour stream, in the required low ppmv concentration range, to a sensor port housing the
chemiresistor.
The response of chromium titanium oxide (CTO) chemiresistors towards 2,4-DNT vapours was observed as an increase of the films resistance. The response was found to be reversible and proportional to the square root of 2,4-DNT concentration. When a tungsten oxide (W03) chemiresistor was exposed to vapours of 2,4-DNT, the response was also observed as an increase of film resistance. However, this response was found to be independent of 2,4-DNT concentration in the range studied, and it is suggested that a competitive adsorption reaction exists between oxygen and 2,4-DNT. Chemiresistors
based on conducting polymers were constructed using platinum microband electrodes.
Polyaniline and poly(3-methylthiophene) exhibited a small decrease of film resistance in the presence of 2,4-DNT. These responses were enhanced by the incorporation of heteropolyanions (HPAs) into the film. The data indicated that the HPAs acted as charge transfer agents, transferring charge from the 1t-conjugated backbone of the polymer to the
nitroaromatic.
In the second part of this project, the redox switching characteristics of polyaniline (Pani) films, deposited onto platinum microband electrode arrays, were studied using voltammetry and chronoamperometry. The size of the electrode contact area was found to shift the voltammetric oxidation peak to more positive potentials, and influence the charge associated with the reduction process. The time required for the oxidation process to complete after a potential step was also influenced. Furthermore, reducing the contact area between the electrode and the polymer film resulted in diminishing the number of sites available for oxidation of the polymer at the electrode surface, which in tum modified the percolation threshold. The distance that the conducting-insulating interface was required to propagate was also increased. The position of the working electrode within the array was observed to modify the redox switching characteristics of the Pani
film. Electrodes at the centre of the array gave rise to faster switching times and lower oxidation peak potentials than electrodes at the edge of the array. It is suggested that the conducting-insulating interface propagated in two opposing directions, which halved the propagation distance. The propagation rate was examined by potentiometric monitoring
of electrodes adjacent to the working electrode, and was found to be 2000 μm s·1
.