The kinetics of enthalpic relaxation are reviewed and applied to the ageing of a range of blends made from polyether imide and polyether ether ketone. DSC has been used to follow the development of enthalpic relaxation and a Williams-Watt stretched exponential equation relating the extent of relaxation, ϕ(t), to the ageing time t and an average relaxation time, t́a', has been used to quantify the ageing process. where β' is inversely related to the breadth of the relaxation spectrum such that 0<β>1.0. The relationship was modified to incorporate non-linearity in the relaxation behaviour. ϕ(t) was measured directly from the enthalpy change observed in the endotherms on heating aged specimens through the glass transition in the DSC. The PEI/PEEK blends were compatible over the full composition range in that they exhibited a single glass transition with a temperature that varied almost linearly with composition between those of the homopolymers. Enthalpic relaxation was found to be a useful technique for probing the molecular relaxations of polymer blends and confirming the degree of compatibility of the system. The β' values changed systematically with the blend composition between those of the homopolymers suggesting that the breadth of the relaxation spectra were similar in the blends to that in the homopolymers. Physical ageing was observed to embrittle the blends, and there was a close correlation between the extent of enthalpic ageing and the change in mechanical and impact behaviour. The yield stress increased and the elongation to break decreased progressively with ϕ(t) in addition to a reduction in impact strength. The model of enthalpic relaxation and the kinetic relationships, outlined above, have been used to determine the onset of the glass transition temperature and subsequent progress of enthalpic relaxation at fixed ageing temperatures, for direct comparison with the change in specific heat observed in DSC experiments. Good agreement was observed between experiment and calculated glass transitions and the effect of variables, such as activation enthalpies, pre-exponential factors, non-linear factors such as X and β' and fictive temperature on the observed glass transition temperatures and the temperature range over which the glass transition occurred determined. Modifications to the model for the enthalpic relaxation have been suggested.
