Abstract
The development of amine-cured epoxy resins for use as matrices for fibre reinforced composites has focused attention on the reactions of amines with epoxides. The aim of this thesis is to obtain qualitative and quantitative kinetic information for the reaction of mono-amines with model epoxides, so that the reactions taking place in commercial formulations may be better understood. The analytical methods used are radio high performance liquid chromatograph (HPLC) - a radiochemical tracer technique - and Fourier transform infra-red spectroscopy (FTIR). The radiochemical method involved the tritiation (labelling with the radioactive isotope of hydrogen) of the amine component of a reaction mixture. The fate of the labelled molecule in the reaction mixture could then be found by separating and detecting the labelled components using the radio HPLC technique. The amount of radioactivity (counts per second) is directly proportional to the concentration of the labelled component in a mixture, and hence the quantification of the data was very simple. The technique was employed to study the reactions of amines with phenyl glycidyl ether (PGE). The kinetic data could be processed using a classical kinetic treatment and by computer modelling methods. This enabled us to calculate the individual rate constants for the following scheme: RNH2+CH2-CHCH2OPh k1→RNHCH2CHCH2OPh (step1) RNHCH2CHCH2OPh + CH2-CHCH2OPh k2→ RN(CH2CHCH2oPh)2 (step2) The reactions were found to follow an autocatalytic course. Values of k1, k2 and the rate ratio k2/k1 were obtained for a number of different systems. In the case of the computer modelling analysis, the existence of an additional reaction step was observed, with a rate constant kn. The rate constant for this reaction was much smaller than k1 and k2. Using the rate constant obtained it was possible to construct a Hammett plot for the reactions of a series of substituted anilines with PGE, and an Arrhenius plot for the reaction between aniline and PGE. The effects of solvents and steric hindrance on amine-epoxide reactions are also explored. FTIR was employed as an analytical technique because it can be employed on industrial amine-cured epoxy resins, where the products become an insoluble, infusible mass. The changes in the FTIE spectrum of a reaction mixture were monitored as a function of time; quantification was achieved using a ratio method. It was found that the technique can provide quantitative information concerning the reactions of amines with PGE, but the accuracy was not as good as the radio HPLC technique. It is clear from the study that computer modelling techniques would be better suited to analysing the kinetic data of more complex commercial systems than the classical kinetic treatment, which requires the epoxide component to be present in large excess. Further research is required before the cure kinetics of commercial systems can be fully understood. This would include a study of the effects of viscosity and diffusion control on amine-epoxy reactions. Ultimately the work should be extended to commercially available epoxy formulations. This may require the development of new analytical techniques such as solid state NMR and solid state tritium NMR.