Multiple Time-Scale Behaviour and Network Dynamics in Liquid Methanol

Abstract

Canonical ensemble molecular dynamics simulations of liquid methanol, modeled using a rigid-body, pair-additive potential, are used to compute static distributions and temporal correlations of tagged molecule potential energies as a means of characterising the liquid state dynamics. The static distribution of tagged molecule potential energies shows a clear multimodal structure with three distinct peaks, similar to those observed previously in water and liquid silica. The multimodality is shown to originate from electrostatic effects, but not from local, hydrogen-bond interactions. An interesting outcome of this study is the remarkable similarity in the tagged potential energy power spectra of methanol, water and silica, despite the differences in the underlying interactions and the dimensionality of the network. All three liquids show a distinct multiple time scale (MTS) regime with a 1/f dependence with a clear positive correlation between the scaling exponent alpha and the diffusivity. The low-frequency limit of the MTS regime is determined by the frequency of crossover to white noise behaviour which occurs at approximately 0.1 cm{-1} in the case of methanol under standard temperature and pressure conditions. The power spectral regime above 200 cm{-1} in all three systems is dominated by resonances due to localised vibrations, such as librations. The correlation between $α$ and the diffusivity in all three liquids appears to be related to the strength of the coupling between the localised motions and the larger length/time-scale network reorganizations. Thus the time scales associated with network reorganization dynamics appear to be qualitatively similar in these systems, despite the fact that water and silica both display diffusional anomalies but methanol does not.