A comparative study of the motions of methane molecules and argon atoms in liquid and high-pressure gas states by neutron scattering

Abstract

The microscopic motions in methane in the liquid state, T = 102 K, and in the high-pressure gas state at 1640 bar, T = 295 K, as well as in high-pressure argon at 1815 bar, T = 295 K, mixed with methane of partial pressure of 82 bar were investigated by scattering of slow neutrons of wavelengths 4.18 and 4.73 Å. The data were carefully corrected for instrumental effects and multiple scattering in sample and container walls. The corrected data were compared to a few simple models for the micro-dynamics: Langevin diffusion and argon-type motion for the centre-of-mass motion, Langevin diffusion and free rotation for the rotational motion. Translational and rotational motions were supposed to be uncoupled. The centre-of-mass motion described as the argon-type of motion was derived from existing experimental data on liquid argon by use of the law of corresponding states. It is found that the molecular motions in liquid methane are well described by the argon type of centre-of-mass motion combined with the Langevin type of rotational diffusion. In contrast, the Langevin-diffusion model for both translational and rotational motion fits the data reasonably well in high-pressure methane as well as in high-pressure argon. It is, however, fair to state that a model for rotational motion inbetween free rotation and Langevin rotational diffusion would give the best fit to the gas data. It is thus demonstrated that a clear distinction exists between the dynamics of the liquid and the high-pressure gas forms of argon and methane. The experiments on high-pressure argon and methane are the first ones to reveal a case in which Langevin diffusion describes the translational motions in a dense fluid to a very good approximation. A comparison of the values of the diffusion constants obtained from these neutron studies to values calculated by use of Enskog's theory as well as other data is carried through. The various values of the diffusion constants agree reasonably well. The gap between microscopic and macroscopic measurements of the diffusion constant is thereby overbridged.