Systematic differences in the relaxation stretching of polar molecular liquids probed by dielectric vs magnetic resonance and photon correlation spectroscopy

Abstract

Relaxation spectra of molecular glass formers devoid of secondary relaxation maxima, as measured by dielectric spectroscopy (DS), nuclear magnetic resonance (NMR) relaxometry, photon correlation spectroscopy (PCS), and Fabry-Perot interferometry, are quantitatively compared in terms of the Kohlrausch stretching parameter βK. For a reliable estimate of βK, the excess wing contribution has to be included in the spectral analysis. The relaxation stretching probed by PCS and NMR varies only weakly among the liquids (βK = 0.58 ± 0.06). It is similar to that found in DS, provided that the liquid is sufficiently nonpolar (relaxation strength Δϵ≲6). For larger strengths, larger βKDS (narrowed relaxation spectra) are found when compared to those reported from NMR and PCS. Frequencyerature superposition (FTS) holds for PCS and NMR. This is demonstrated by data scaling and, for the few glass formers for which results are available, by the equivalence of the susceptibilities χPCS″ωτ∝χNMR″τ∝χNMR″ω, i.e., measuring at a constant frequency is equivalent to measuring at a constant temperature or constant correlation time. In this context, a plot of the spin-lattice relaxation rate R1(T) as a function of the spin-spin relaxation rate R2(T) is suggested to reveal the stretching parameter without the need to perform frequency-dependent investigations. Dielectrically, we identify a trend of increasing deviations from FTS with increasing Δϵ. Depending on the technique and glass former, the relative relaxation strength of the excess wing varies, whereas its exponent appears to be method independent for a given substance. For polar liquids, we discuss possible reasons for the discrepancy between the results from PCS and NMR as compared to those from DS.