Self-diffusion of liquids is among the simplest forms of diffusion and characterizes molecules’ random translational motion within a system. Self-diffusion of liquids is of interest in various fields, including chemical engineering, liquid science, material physics, and theoretical chemistry. In this study, the authors compiled 547 self-diffusion coefficients into a large original database through a combination of a literature search and additional measurements performed using the pulsed-field gradient nuclear-magnetic-resonance method. Our database embraces extremely low (6.26 × 10−9 m2/s) to high (6.79 × 10−13 m2/s) values of self-diffusion coefficients of chemically diverse liquids, including various alcohols, amines, carbonyl compounds, ethers, hydrocarbons, nitriles, nitro and organosulfur compounds, and silanes. For molecular dynamics simulations, cubic-shaped simulation cells were prepared for each liquid, and periodic boundary conditions were employed. Following relaxation and equilibration of the simulation systems, sampling simulations were performed to calculate diffusion coefficients from the mean-square displacement of molecules. This study revealed that our all-atom molecular dynamics simulations had excellent predictive performance, yielding determination coefficient (R2) of 0.934 and mean absolute error of 0.203 between the observed and calculated logarithmic self-diffusion coefficients.

All-atom molecular dynamics simulation: Self-diffusion coefficients of diverse pure liquids

ACS Spring Meeting 2021

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.