Comparison in the time region 0–1.2 ps of model and experimental absorptions of liquid and rotator phases in the far infrared

Abstract

The Brot–Larkin and Wyllie–Larkin models of permanent-dipolar librational and relaxational molecular motion in the liquid and plastic crystal (rotator) phases are tested in the time domain by comparing the theoretical rotational velocity correlation functions with the Fourier transform of the optical absorption coefficient per unit length [α(ω)]. The discord between model and observation is particularly acute at short times. The Brot–Larkin function does not have the necessary zero slope of the correlation function at t→ 0 because of its use of an effectively infinite intermolecular mean square torque. The Wyllie–Larkin treatment also results in an excessive apparent means square torque, although having the correct short-time zero slope. These features are reflected in the fact that the theoretical curves are too broad in the frequency domain, i.e., do not correctly predict the sharp high frequency fall off in α(ω). This has the further consequence that the model rotational velocity correlation functions do not show the pronounced short time oscillations of the experimental curves. Qualitative improvements are suggested.