Thermal cycling of glasses. III. Upper peaks

The response of glasses to three-step thermal cycles in which the glass is poorly stabilized prior to reheating has been examined both theoretically and experimentally. Theory predicts that, during the heating stage of these cycles, a type of peak (called an upper peak) occurs which is quite different from the type of peak (called a main peak) observed for well-stabilized glasses. The dependence of the upper peak temperature and height on the experimental variables defining the thermal cycles have been derived theoretically, and have been compared with experimental data for polystyrene. All of the features predicted theoretically have been verified experimentally. In particular, the variation of the upper peak temperature is insensitive to the structure (or nonlinearity) parameter x, and cannot therefore lead to its evaluation. On the other hand, it is shown that, given the value of x obtained, for example, by the peak-shift method, then an analysis of the upper peak height can yield information about the width of the distribution of retardation times.