A dielectric test of the validity of the Adam–Gibbs equation out-of-equilibrium: Polymers vs. small molecules

The α-relaxation time, both at equilibrium and out-of-equilibrium, has been investigated by means of dielectric relaxation in a series of fragile glass formers including polymers and several small molecules. The dielectric experiments combine both, thermally stimulated depolarization (accessible time range 10–106 s) and standard broad-band dielectric relaxation measurements (frequency range 10−2–107 Hz). The analysis of both set of data was based on a stretched exponential response. The experimental temperature dependence so derived for the relaxation times out-of-equilibrium has been found to give account not only of global polarization experiments but also for those thermally stimulated depolarization experiments involving partial polarization. Moreover, the so obtained activation energies in the glassy range have been quantitatively compared with the predictions of the Adam–Gibbs equation. For the small molecule glass formers a good agreement was found, but systematic discrepancies are apparent for polymers. By studying the influence of physical aging on this comparison, we found that after significant annealing also the polymers tends to agree with the predictions of the Adam–Gibbs equation. All these results would support the validity of the Adam–Gibbs equation for describing the temperature dependence of the time scale of the α-relaxation also out-of-equilibrium. However, they would also imply a noticeable qualitative difference between polymers and small molecule systems concerning the configurational entropy behavior. Whereas a significant temperature variation of the configurational entropy would occur for non-annealed polymers some 10 K below the glass transition, this temperature dependence would be much weaker for small molecules and well annealed polymers.