Thermal cycling of glasses. II. Experimental evaluation of the structure (or nonlinearity) parameter x

The structure parameter x (also known as the nonlinearity parameter), which in the KAHR model determines the relative contributions of temperature and structure to the retardation times of polymers in the glass transition region, is evaluated experimentally for a narrow molecular weight fraction of atactic polystyrene by the peak-shift method. This method is based on the dependence on heating rate and annealing time of the endothermal peak temperatures T_p, obtained by differential scanning calorimetry during the heating stage of well-defined three-step thermal cycles. The results obtained from the shifts of T_p, with both heating rate and the amount of isothermal recovery during the annealing stage before reheating, are consistent with the theoretical predictions of the KAHR model. The appropriate analysis of the data for a polystyrene fraction of molecular weight 30,100 leads to a value of x = 0.46 ± 0.02. Furthermore, this value of x is constant for a wide range of annealing times and heating rates and for two different annealing temperatures, in full agreement with the theoretical model. The advantages of the peak-shift method over that used by most other workers, the curve-fitting method, are critically discussed.