| Abstract: | The response of glass-forming systems to isobaric three-step thermal cycles involving cooling, isothermal annealing, and subsequent reheating has been investigated comprehensively using numerous combinations of the experimental and material parameters. The latter include the retardation spectrum or response function and the parameter x which determines the relative contributions of temperature and structure to the retardation times. The results show that, on heating, multiparameter systems can display three kinds of peak in the expansion coefficient α or the heat capacity Cp originating from the interactions of the elementary retardation processes with the thermal history of the glass. The conditions under which these peaks occur, their shifts with x, and the experimental variables have been investigated in detail. In particular, it has been shown that for a thoroughly stabilized glass reheated rapidly, the temperature at which the main peak occurs is strongly dependent on the experimental variables and on x, and that this dependence can lead to an estimate of the value of x. For poorly stabilized glasses reheated slowly, on the other hand, the main peak apparently vanishes and allows an upper peak to appear, which is small and insensitive to x and the experimental variables. Intermediate situations in which the main and the upper peaks occur separately give rise to a range of effects which may become quite complicated as the two peaks approach each other. The occurrence of such multiple peaks, which has usually been overlooked or misinterpreted in the past, is discussed and explained in detail and is compared with some experimental observations reported in the literature. |
