Othonas A. Moultos

Thermodynamic and Transport Properties of H2O + NaCl from Polarizable Force Fields

Molecular dynamics and Monte Carlo simulations were performed to obtain thermodynamic and transport properties of the binary H2O + NaCl system using the polarizable force fields of Kiss and Baranyai ( J. Chem. Phys. 2013 , 138 , 204507 and 2014 , 141 , 114501 ). In particular, liquid densities, electrolyte and crystal chemical potentials of NaCl, salt solubilities, mean ionic activity coefficients, vapor pressures, vapor-liquid interfacial tensions, and viscosities were obtained as functions of temperature, pressure, and salt concentration. We compared the performance of the polarizable force fields against fixed-point-charge (nonpolarizable) models. Most of the properties of interest are better represented by the polarizable models, which also remain physically realistic at elevated temperatures.

Molecular simulation of thermodynamic and transport properties for the H2O+NaCl system

Molecular dynamics and Monte Carlo simulations have been carried out to obtain thermodynamic and transport properties of the binary mixture H2O+NaCl at temperatures from T = 298 to 473 K. In particular, vapor pressures, liquid densities, viscosities, and vapor-liquid interfacial tensions have been obtained as functions of pressure and salt concentration. Several previously proposed fixed-point-charge models that include either Lennard-Jones (LJ) 12-6 or exponential-6 (Exp6) functional forms to describe non-Coulombic interactions were studied. In particular, for water we used the SPC and SPC/E (LJ) models in their rigid forms, a semiflexible version of the SPC/E (LJ) model, and the Errington-Panagiotopoulos Exp6 model; for NaCl, we used the Smith-Dang and Joung-Cheatham (LJ) parameterizations as well as the Tosi-Fumi (Exp6) model. While none of the model combinations are able to reproduce simultaneously all target properties, vapor pressures are well represented using the SPC plus Joung-Cheathem model combination, and all LJ models do well for the liquid density, with the semiflexible SPC/E plus Joung-Cheatham combination being the most accurate. For viscosities, the combination of rigid SPC/E plus Smith-Dang is the best alternative. For interfacial tensions, the combination of the semiflexible SPC/E plus Smith-Dang or Joung-Cheatham gives the best results. Inclusion of water flexibility improves the mixture densities and interfacial tensions, at the cost of larger deviations for the vapor pressures and viscosities. The Exp6 water plus Tosi-Fumi salt model combination was found to perform poorly for most of the properties of interest, in particular being unable to describe the experimental trend for the vapor pressure as a function of salt concentration.

Atomistic Molecular Dynamics Simulations of CO2 Diffusivity in H2O for a Wide Range of Temperatures and Pressures

Molecular dynamics simulations were employed for the calculation of diffusion coefficients of CO2 in H2O. Various combinations of existing force fields for H2O (SPC, SPC/E, and TIP4P/2005) and CO2 (EPM2 and TraPPE) were tested over a wide range of temperatures (283.15 K < T < 623.15 K) and pressures (0.1 MPa < P < 100.0 MPa). All force-field combinations qualitatively reproduce the trends of the experimental data; however, two specific combinations were found to be more accurate. In particular, at atmospheric pressure, the TIP4P/2005-EPM2 combination was found to perform better for temperatures lower than 323.15 K, while the SPC/E-TraPPE combination was found to perform better at higher temperatures. The pressure dependence of the diffusion coefficient of CO2 in H2O at constant temperature is shown to be negligible at temperatures lower than 473.15 K, in good agreement with experiments. As temperature increases, the pressure effect becomes substantial. The phenomenon is driven primarily by the higher compressibility of liquid H2O at near-critical conditions. Finally, a simple power-law-type phenomenological equation is proposed to correlate the simulation values; the proposed correlation should be useful for engineering calculations.

Atomistic molecular dynamics simulations of H2O diffusivity in liquid and supercritical CO2

Molecular dynamics simulations were employed for the calculation of diffusion coefficients of pure CO2 and of H2O in CO2 over a wide range of temperatures (298.15 K < T < 523.15 K) and pressures (5.0 MPa < P < 100.0 MPa), that are of interest to CO2 capture-and-sequestration processes. Various combinations of existing fixed-point-charge force-fields for H2O (TIP4P/2005 and Exponential-6) and CO2 (elementary physical model 2 [EPM2], transferable potentials for phase equilibria [TraPPE], and Exponential-6) were tested. All force-field combinations qualitatively reproduce the trends of the experimental data for infinitely diluted H2O in CO2; however, TIP4P/2005–EPM2, TIP4P/2005–TraPPE and Exponential-6–Exponential-6 were found to be the most consistent. Additionally, for H2O compositions ranging from infinite dilution to , the Maxwell–Stefan diffusion coefficient is shown to have a weak non-linear composition dependence.

Atomistic molecular dynamics simulations of carbon dioxide diffusivity in n-hexane, n-decane, n-hexadecane, cyclohexane and squalane

Atomistic molecular dynamics simulations were carried out to obtain the diffusion coefficients of CO2 in n-hexane, n-decane, n-hexadecane, cyclohexane, and squalane at temperatures up to 423.15 K and pressures up to 65 MPa. Three popular models were used for the representation of hydrocarbons: the united atom TraPPE (TraPPE-UA), the all-atom OPLS, and an optimized version of OPLS, namely, L-OPLS. All models qualitatively reproduce the pressure dependence of the diffusion coefficient of CO2 in hydrocarbons measured recently, and L-OPLS was found to be the most accurate. Specifically for n-alkanes, L-OPLS also reproduced the measured viscosities and densities much more accurately than the original OPLS and TraPPE-UA models, indicating that the optimization of the torsional potential is crucial for the accurate description of transport properties of long chain molecules. The three force fields predict different microscopic properties such as the mean square radius of gyration for the n-alkane molecules and pair correlation functions for the CO2-n-alkane interactions. CO2 diffusion coefficients in all hydrocarbons studied are shown to deviate significantly from the Stokes-Einstein behavior.

System-size corrections for self-diffusion coefficients calculated from molecular dynamics simulations: The case of CO2, n-alkanes, and poly(ethylene glycol) dimethyl ethers

Molecular dynamics simulations were carried out to study the self-diffusion coefficients of CO2, methane, propane, n-hexane, n-hexadecane, and various poly(ethylene glycol) dimethyl ethers (glymes in short, CH3O-(CH2CH2O)n-CH3 with n = 1, 2, 3, and 4, labeled as G1, G2, G3, and G4, respectively) at different conditions. Various system sizes were examined. The widely used Yeh and Hummer [J. Phys. Chem. B 108, 15873 (2004)] correction for the prediction of diffusion coefficient at the thermodynamic limit was applied and shown to be accurate in all cases compared to extrapolated values at infinite system size. The magnitude of correction, in all cases examined, is significant, with the smallest systems examined giving for some cases a self-diffusion coefficient approximately 15% lower than the infinite system-size extrapolated value. The results suggest that finite size corrections to computed self-diffusivities must be used in order to obtain accurate results.

Molecular Thermodynamic Models for CO2 and Mixtures: Recent Developments and Applications for Process Design

Abstract In an effort to accurately predict the thermodynamic and transport properties of CO 2 in mixtures with other components, typically found in the carbon capture and sequestration process, both molecular simulation and equations of state, based on statistical mechanics are used. This paper provides a highlight of the use of these techniques to predict equilibrium properties such as phase equilibria and densities of multicomponent mixtures and transport properties, such as the mutual self-diffusion coefficients of H 2 O and CO 2 .