Abstract:
The interpretation and modelling of the dielectric response of thermosetting materials
during cure was the main focus of this study. The equivalence of complex permittivity
and complex impedance in terms of information content was outlined in a series of
case studies covering the separate effects of dipolar movements and charge migration
as well as the combined effect of the two polarisation mechanisms. Equivalent
electrical circuits were used in order to model the evolution of the complex impedance
during cure. A numerical method that can model consecutive spectra throughout the
cure was developed. The method is based on Genetic Algorithms and requires only
input from the modelling of the initial spectra.
Complex impedance spectra were collected during the cure of a commercial epoxy
resin formulation under isothermal and dynamic heating conditions. The spectra were
analysed and modelled. The modelling was successful over the whole frequency range
of the measurements (1 Hz – 1 MHz). The analysis of the estimated model parameters
showed that charge migration dominates the dielectric response in a wide frequency
range. In addition, the modelling algorithm also distinguished between the effects of
electrode polarisation and dipolar movements in the signal. A new equivalent circuit
was used in order to map the frequency regions where the each one of the three
phenomena that together comprise the dielectric signal can be monitored most
effectively.
A chemical cure kinetics model was developed for the studied system. A correlation
between the maximum point of the imaginary impedance spectrum and the reaction
conversion was established. A mathematical model, based on a simple linear
dependence of the dielectric signal on conversion and temperature, was built. The
model predictions agreed well with the experimental data.
The aim of simplifying the interpretation of the dielectric signals led to the
development of a new experimental technique. Temperature Modulated Dielectric
Analysis employs temperature modulations superimposed on an underlying thermal
profile in order to separate the influence on the signal of the temperature alone from
that of the cure reaction. The early study carried out here shows that such
measurements are feasible and reveals important issues for its further development.
