Mass diffusion of organic fluids : a molecular dynamics perspective

Abstract

The Optimized Potential for Liquid Simulation All-Atom (OPLS-AA) forcefield, a wellestablished interaction potential, within the MD framework was used to determine self-and mutual diffusivity of several fluids, in the high-pressure and near-and supercritical regimes. The test fluids were benzene, heptane and carbon dioxide. The Green-Kubo method of calculating mass diffusivity yielded considerably overestimated values of self-diffusivity; better results were obtained with use of the Stokes-Einstein method. In addition, a number of relatively significant problems with the default OPLS-AA parameterization were identified. There exist ranges of pressure and temperatures, in which the results for the selected pure fluids show good agreement with experiment, and thus reparameterization of OPLS-AA may not be required. However, the results for mixtures all show significant deviation from experiment. Based on the observed results, the current predictive power of OPLS-AA for mass diffusivity of the fluids in question is not satisfactory. We are therefore convinced that a systematic reparameterization of OPLS-AA both in terms of the intermolecular (van der Waals, electrostatic) and intramolecular (bonding) interactions is needed to increase its accuracy. In addition, OPLS-AA may not be generally appropriate for the gaseous state and therefore near-critical systems may benefit from using a different forcefield.

Liquid–Liquid Flow at Nanoscale: Slip and Hydrodynamic Boundary Conditions

“…The GK relation eq leads to the following expression of the slip length of water on a rigid heptane wall: Despite its simplicity, this approach gives the basic ingredients to understand the origin of slip , and a more systematic and accurate derivation leads to very similar results. , A symmetrical formula is obtained for the slip length of heptane on a rigid water wall. On average, if the box size is big enough, so that ρ w ϵ w 2 σ w = ρ h ϵ h 2 σ h , we finally obtain In this work, the diffusion coefficient of heptane is D h = 3.7 × 10 –9 m 2 ·s –1 and the diffusion coefficient of water is D w = 2.97 × 10 –9 m 2 ·s –1 . These are bulk values taken in order to propose an interpretation of the difference between the solid–liquid slip length and the liquid–liquid slip length by modeling the different effects.…”

“…13,14 A symmetrical formula is obtained for the slip length of heptane on a rigid water wall. On average, if the box size is big enough, so that ρ w ϵ w 2 σ w = ρ h ϵ h 2 σ h , we finally obtain (14) In this work, the diffusion coefficient of heptane 65 is D h = 3.7 × 10 −9 m 2 •s −1 and the diffusion coefficient of water 55 is D w = 2.97 × 10 −9 m 2 •s −1 . These are bulk values taken in order to propose an interpretation of the difference between the solid−liquid slip length and the liquid−liquid slip length by modeling the different effects.…”

“…In this work, the diffusion coefficient of heptane65 is D h = 3.7 × 10 −9 m 2 •s −1 and the diffusion coefficient of water55 is D w = 2.97 × 10 −9 m 2 •s −1 . These are bulk values taken in order to propose an interpretation of the difference between the solid−liquid slip length and the liquid−liquid slip length by modeling the different effects.…”

Diffusivity and hydrophobic hydration of hydrocarbons in supercritical CO2 and aqueous brine

“…We speculate that the discrepancy between simulation results and experimental data could be due to inaccuracy in internal degrees of freedom of molecules, which could occur for isotherms at higher pressures. 43 Unlike the case of selfdiffusivity of molecules, the mixture of hydrocarbon and CO 2 would create a more complex system considering the number of degrees of freedom compared to the water-CO 2 mixed system. Fig.…”