The exponential dielectric relaxation dynamics in a secondary alcohol’s supercooled liquid and glassy states

Abstract

To gain insight into a recent observation that the prominent, Debye-type relaxation process observed for a primary alcohol may not be the ␣-relaxation process associated with molecular diffusion of a liquid ͓Europhys. Lett. 40, 549 ͑1997͒, J. Chem. Phys. 107, 1086 ͑1997͔͒, the dielectric spectra of an uncrystallizable secondary alcohol, 5-methyl-2-hexanol, has been investigated by broadband spectroscopy. Measurements made over a temperature range from 110 to 298 K showed that three relaxation processes occur. Processes I and II have a non-Arrhenius variation of the relaxation rate with temperature, and process III an Arrhenius. Only process I, the slowest of the three, has a single relaxation rate, the other two, a broad distribution. The contribution to permittivity from process II was 0.8, i.e., ϳ3% of the static permittivity, and from process III, the fastest was 0.1, i.e., ϳ0.3%. It is argued that the mechanism of process I is the breaking followed by dipolar reorientation and reforming of the H-bonds in the intermolecularly H-bonded structure, and process II is that of the orientation of the other dipolar groups, such as the -OR group. Process III is the usual JohariGoldstein process. For 5-methyl-2-hexanol, the mode-coupling and another theory by Souletie and Bertrand ͓J. Phys. I 1, 1627 ͑1991͔͒ seem to agree with the relaxation rate of processes I and II, and predict temperatures for 10 Ϫ4 Hz relaxation rate, within a few degrees of that expected.

Localized relaxation in a glass and the minimum in its orientational polarization contribution

“…1 ͑curve 5͒ is for a temperature ͑151.7 K͒ above the T g , while curve 2 ͑125.5 K͒ is below the T g , which is ϳ148 K for 5-methyl-2-hexanol. Peaks in the ⑀Љ spectra at frequencies higher than the frequency of the ␣-relaxation peak in this temperature range have been observed before 32 and shown to be due to the Johari-Goldstein relaxation. The increasing ⑀Љ with decreasing frequency below 500 Hz for curve 5 in Fig.…”

“…1 and 2͒ cannot be due to the dc conductivity for two reasons: ͑i͒ the slope of the plot of log ⑀Љ against log f in the region of this ''wing'' is not unity; and ͑ii͒ this ''wing'' turns into the high-frequency part of the ␣-relaxation peak at higher temperatures, which has already been shown in Ref. 32, and therefore is not shown in Figs. 1 and 2 here.…”

“…A detailed analysis of the dc conductivity of 5-methyl-2-hexanol, and the characteristics of its ␣-and ␤-relaxation processes had previously been carried out in a separate study, which may be consulted for details. 32 A typical fit of Eq. ͑1͒ to the ⑀Ј and ⑀Љ spectra of the ␤-relaxation process is shown in Fig.…”

“…However, for an unbiased analysis of the data, we use the Havriliak-and Negamitype equations, 27 which fits the data to several relaxation processes by minimizing the deviation of the data from the fits. It was done in the manner described earlier, 32 …”

“…Dielectric relaxation features of 5-methyl-2-hexanol as a function of temperature have been reported before. 32 …”