Carbon nanotube (CNT) reinforced polymeric nanocomposites (PNCs) show significant potentials for various sensing applications. In recent years, a large number of laboratory-level sensor systems based on electrical conductivity have been reported;​ however, only a few commercial sensors based on CNT-PNCs have been produced. For commercialization, a sensor system should be highly reliable, reproducible, low-cost, stable and easy to install. To develop these characteristics in CNT-PNC based sensor systems, elucidation of various properties of CNT nanocomposites is vital to determine the behaviour and nature of sensor systems.

In this study, the modelling and characterization of electromechanical properties of CNT-PNCs and the investigation of influencing factors, such as the base polymer matrix and the CNT network inside the polymer matrix, have been addressed. Most of the models in recent studies assume CNTs as straight fibres with a fixed size, and only few of them account for features such as waviness and size variations of CNTs. However, experiments show that during the manufacturing process of CNT-PNCs, nanotubes become wavy. They either get entangled or broken, and the CNT alignment in the polymer matrix can be impacted. In this thesis, a model has been developed that considers the behaviour of wavy nanotubes with varying nanotube lengths. The model also incorporates varying degrees of CNT alignment in the polymer matrix and determines its effect on the physical properties of CNT-PNCs.

Nanocomposite specimens were fabricated using three different techniques, i.e. compression moulding, injection moulding and spray painting. The effects of various fabrication methods on the alignment and dispersion of multi-walled CNTs were experimentally characterized through microscopy and electrochemical impedance spectroscopy (EIS) techniques and compared with modelling results. A process has been developed that uses EIS analysis as a reliable, albeit indirect, tool to accurately predict the CNT alignments without any of the disadvantages of expensive imaging techniques.

This study also presents an analytical model for determining the piezoresistive and piezoelectric properties exhibited by CNT-reinforced piezoelectric polymer matrix nanocomposites. It has been well established that CNT nanocomposites exhibit a piezoresistive property. Force, pressure or strain applied to the nanocomposites contributes to changes in the network between CNTs and alters the inter-CNT distances, thereby modifying electron conduction and varying the resistivity of the nanocomposites. By using piezoelectric polymers, such as polyvinylidene difluoride (PVDF), which can be converted in to piezoelectric beta phase through electrical polarization, a piezoelectric property can be imparted to CNT-PNCs. This unique behaviour of CNT/PVDF nanocomposites has been thoroughly examined through analytical modelling and experimental analysis.

The results of this study show a new class of nanocomposite materials fabricated using piezoelectric polymers admixed with CNTs that exhibit enhanced electromechanical properties and both piezoresistive and piezoelectric behaviour. These nanocomposites were highly sensitive and could be effectively utilized as in situ sensors to monitor applied force/stress or pressure. We have focused on the design and development of various sensor systems, such as a flexible strain sensor, tactile sensor and multi-axis machining dynamometer, fabricated in this study using these nanocomposites.

The present research work provides a fundamental understanding of novel CNT-PNCs that can be used as sensing material in broad applications in the civil, mechanical, aerospace, biomedical, electronics, microelectromechanical (MEMS) systems and petroleum engineering fields.