Relaxation spectra of molecular glass formers measured by tandem Fabry–Perot interferometry (TFPI) and by photon correlation spectroscopy (PCS) are revisited from well above Tm down to Tg. We scrutinize the claim that a generic relaxation stretching is found close to Tg, while varying stretching is established at high temperatures. The TFPI spectra observed in this work reveal no change of the stretching over a large temperature range, and a Cole–Davidson susceptibility yields stretching parameters within βCD = 0.39–0.80. The spectra display a high-frequency excess wing contribution, which prohibits a single power law description of the high-frequency flank of the main relaxation. Corresponding PCS decays measured close to Tg display no change in the stretching either. Yet, the PCS spectra are overall broader than the TFPI spectra. They exhibit some variation and are superiorly described by a Kohlrausch function with βK = 0.52–0.73. Three of the eight systems significantly deviate from the previously reported generic relaxation function. We do not find indications that the spectral width changes within a narrow temperature interval to bridge the different stretching monitored by the two techniques. In addition to other possible explanations, we consider a sharp transition of the dynamics in the ns range not covered by the two techniques so far. Comparing PCS spectra of weakly polar liquids with their dielectric spectra, we find counterexamples of the claim that they become identical; the PCS spectra are narrower in these cases. Still, in this limit, the dielectric spectra display identical spectral shapes.
Rössler et al. (Wed,) studied this question.