We review several features of the glass transition phenomenon: crossover, hysteresis, and multi-exponential relaxation. The crossover effect caused a seemingly reversed kinetics in the classical experiment by Lillie. The hysterisis-type behavior is explained by a significant decrease in viscosity with increasing temperature. The relaxation times in multi-exponential relaxation are shown to be controlled by the average fictive temperature, in agreement with the original proposition by Narayanaswamy. We succeeded in validating the thermorheologically complex (TC) model by comparing its performance against six sets of data for the NBS 710 glass obtained by four different authors. We failed to fit the seventh and eighth sets of data with the same set of parameters. This is at least partially due to the glass in question being found on measurement to differ in viscosity from the certified data for the actual NBS 710 glass.
glass transition, crossover, hysteresis, TC model validation.