The influence of the strength of intermolecular interactions on mass diffusive processes remains poorly understood for mixtures of associative liquids with dissolved gases. For contributing to a fundamental understanding of the interplay between liquid structures and mass diffusivities in such systems, dynamic light scattering, Raman spectroscopy, and molecular dynamics simulations were used in this work. As model systems, binary mixtures consisting of the gases hydrogen, helium, nitrogen, carbon monoxide, or carbon dioxide dissolved in ethanol, 1-hexanol, or 1-decanol were selected. Experiments and simulations were performed at macroscopic thermodynamic equilibrium close to infinite dilution of solute for temperatures between 303 and 423 K. The Fick diffusion coefficients and self-diffusivities of the gas solutes increase with increasing temperature, decreasing alkyl chain length of the 1-alcohols, and decreasing molar mass of the solutes except for helium and hydrogen showing the opposite behavior. The analysis of the liquid structure of the mixtures showed that the fraction of hydrogen-bonded alcohol molecules decreases with increasing alkyl chain length and temperature. From the obtained structure–property relationships, a new correlation was developed to predict mass diffusivities in binary mixtures consisting of n-alkanes or 1-alcohols with dissolved gases close to infinite dilution within 10% on average.

In the present study, dynamic light scattering (DLS) experiments and molecular dynamics (MD) simulations were used for the investigation of the molecular diffusion in binary mixtures of liquids with dissolved gases at macroscopic thermodynamic equilibrium. Model systems based on the n-alkane n-hexane or n-decane with dissolved hydrogen, helium, nitrogen, or carbon monoxide were studied at temperatures between 303 and 423 K and at gas mole fractions below 0.06. With DLS, the relaxation behavior of microscopic equilibrium fluctuations in concentration and temperature is analyzed to determine simultaneously mutual and thermal diffusivity in an absolute way. The present measurements document that even for mole gas fractions of 0.007 and Lewis numbers close to 1, reliable mutual diffusivities with an average expanded uncertainty ( k = 2) of 13% can be obtained. By use of suitable molecular models for the mixture components, the self-diffusion coefficient of the gases was determined by MD simulations with an averaged expanded uncertainty ( k = 2) of 7%. The DLS experiments showed that the thermal diffusivity of the studied systems is not affected by the dissolved gas and agrees with the reference data for the pure n-alkanes. In agreement with theory, mutual diffusivities and self-diffusivities were found to be equal mostly within combined uncertainties at conditions approaching infinite dilution of the gas. Our DLS and MD results, representing the first available data for the present systems, reveal distinctly larger mass diffusivities for mixtures containing hydrogen or helium compared to mixtures containing nitrogen or carbon monoxide. On the basis of the broad range of mass diffusivities of the studied gas-liquid systems covering about 2 orders of magnitude from about 10 to 10 m·s, effects of the solvent and solute properties on the temperature-dependent mass diffusivities are discussed. This contributed to the development of a simple semiempirical correlation for the mass diffusivity of the studied gases dissolved in n-alkanes of varying chain length at infinite dilution as a function of temperature. The generalized expression requiring only information on the kinematic viscosity and molar mass of the pure solvent as well as the molar mass and acentric factor of the solute represents the database from this work and further literature with an absolute average deviation of about 11%.

This study contributes to a fundamental understanding of how the liquid structure in a model system consisting of weakly associative n-hexane ( n-CH) and carbon dioxide (CO) influences the Fickian diffusion process. For this, the benefits of light scattering experiments and molecular dynamics (MD) simulations at macroscopic thermodynamic equilibrium were combined synergistically. Our reference Fickian diffusivities measured by dynamic light scattering (DLS) revealed an unusual trend with increasing CO mole fractions up to about 70 mol %, which agrees with our simulation results. The molecular impacts on the Fickian diffusion were analyzed by MD simulations, where kinetic contributions related to the Maxwell-Stefan (MS) diffusivity and structural contributions quantified by the thermodynamic factor were studied separately. Both the MS diffusivity and the thermodynamic factor indicate the deceleration of Fickian diffusion compared to an ideal mixture behavior. Computed radial distribution functions as well as a significant blue-shift of the CH stretching modes of n-CH identified by Raman spectroscopy show that the slowing down of the diffusion is caused by a structural organization in the binary mixtures over a broad concentration range in the form of self-associated n-CH and CO domains. These networks start to form close to the infinite dilution limits and seem to have their largest extent at a solute-solvent transition point at about 70 mol % CO. The current results not only improve the general understanding of mass diffusion in liquids but also serve to develop sound prediction models for Fick diffusivities.

This work contributes to the characterization of long linear and branched alkanes and alcohols via the determination of their thermophysical properties up to temperatures of 573.15 K. For this, experimental techniques including surface light scattering (SLS) and molecular dynamics (MD) simulations were used under equilibrium conditions to analyze the influences of chain length, branching, and hydroxylation on liquid density, liquid viscosity, and surface tension. For probing these effects, 12 pure model systems given by the linear alkanes n-dodecane, n-hexadecane, n-octacosane, n-triacontane, and n-tetracontane, the linear alcohols 1-dodecanol, 1-hexadecanol, and 1,12dodecanediol, the branched alkanes 2,2,4,4,6,8,8-heptamethylnonane (HMN) and 2,6,10,15,19,23-hexamethyltetracosane (squalane), and the branched alcohols 2-butyl-1-octanol and 2-hexyl-1-decanol were investigated at or close to saturation conditions at temperatures between 298.15 and 573.15 K. Based on the experimental results for the liquid densities, liquid viscosities, and surface tensions with average expanded uncertainties (k = 2) of 0.061, 2.1, and 2.6%, respectively, the performance of the three commonly employed force fields (FFs) TraPPE, MARTINI, and L-OPLS was assessed in MD simulations. To improve the simulation results for the bestperforming all-atom L-OPLS FF at larger temperatures, a modified version was suggested. This incorporates a temperature dependence for the energy parameters of the Lennard-Jones potential obtained by calibrating only against the experimental liquid density data of n-dodecane. By transferring this approach to all other systems studied, the modified L-OPLS FF shows now a distinctly better representation of the equilibrium and transport properties of the long alkanes and alcohols, especially at high temperatures.

It is demonstrated that thermal and mutual diffusivities of binary mixtures of n-octacosane (n-C28H58) with carbon monoxide (CO), hydrogen (H2), and water (H2O) are simultaneously accessible by dynamic light scattering (DLS). As the light-scattering signals originating from thermal and concentration fluctuations appear in similar time scales, different data evaluation strategies were tested to achieve minimum uncertainties in the resulting transport properties. To test the agreement of the respective theoretical model with the DLS signals in the regression, an improved multifit procedure is introduced. With the selected data evaluation strategy, uncertainties of 4 to 15% and 4 to 30% in the thermal and mutual diffusivities, respectively, could be obtained for the binary mixtures. The mutual diffusivities for the mixtures measured at temperatures ranging from 398 to 523 K and pressures of 5 to 30 bar at saturation conditions are in good agreement with molecular dynamics simulations and data from the literature.