Single-walled carbon nanotube networks and related composite materials for gas sensing applications

In this thesis, the gas sensing properties of single-walled carbon nanotube (SWCNT) networks and SWCNT-Zeolite composite materials were investigated in a variety of environmental conditions. The aim of the project was to establish the effect that adsorbed water vapour had on the electrical properties of SWCNT networks, along with any subsequent impact on the NO2 sensing responses of SWCNT-based chemiresistors. Motivated by these investigations, the sensitivity of the SWCNT networks to water vapour was exploited to develop the water-assisted regeneration (WAR) method, enabling the improved recovery of the baseline sensing signal. Zeolites, known as molecular sieves due to their selective adsorption properties, were utilised in SWCNT-Zeolite composite sensing layers to reduce the cross-sensitivity of functionalised SWCNTs to water vapour. Functionalisation of the SWCNTs with a range of anionic, cationic and nonionic surfactants to aid solution processing was found to enhance the conductancehumidity effect, in some cases by a factor of 10. An interesting bi-directional switch in conductance change was observed when anionic (conductance decrease) vs cationic (conductance increase) were used. Under experimental conditions, fluctuations in atmospheric humidity levels were shown to alter the gas sensing characteristics of the SWCNT networks. Formed from interconnected metallic and semiconducting SWCNTs, the chemiresistive sensors demonstrated increased response magnitudes, adsorption rates and recovery rates at higher levels (A 50% RH) of relative humidity. Raman spectroscopy, UV-Vis-NIR spectroscopy, electron microscopy and electrical characterisation techniques were used in conjunction with gas sensing experiments to study changes in the properties of the sensing elements, helping to elucidate potential mechanisms. Extraction of key sensing parameters was facilitated by the application of a model for completely irreversible adsorption of NO2, whilst a model based on partially reversible desorption was found to best describe the sensing data