No abstract

We report an assessment of the predictive and correlative capability of the SAFT coarse-grained force field as applied to mixtures of CO2 with n-decane and n-hexadecane. We obtain the pure and cross interactions parameters by matching simulations to experimental phase equilibrium behavior and transfer these parameters to predict shear viscosities. We apply both equilibrium (based on the Green-Kubo formulation) and non-equilibrium (based on the application of an external force to generate an explicit velocity field) algorithms. Single and two-site models are explored for CO2 and, while for volumetric properties both models provide good results, only the model that aligns with the molecular shape is found to be robust when describing highly asymmetric binary mixtures over wide ranges of temperature and pressure. While the models provide good quantitative predictions of viscosity, deviations amongst the algorithms and with experimental data are encountered for binary mixtures involving longer chain fluids, and in particular at high-pressure and low-temperature states.

A molecular modeling methodology is presented to analyze the wetting behavior of natural surfaces exhibiting roughness at the nanoscale. Using atomic force microscopy, the surface topology of a Ketton carbonate is measured with a nanometer resolution, and a mapped model is constructed with the aid of coarse-grained beads. A surrogate model is presented in which surfaces are represented by two-dimensional sinusoidal functions defined by both an amplitude and a wavelength. The wetting of the reconstructed surface by a fluid, obtained through equilibrium molecular dynamics simulations, is compared to that observed by the different realizations of the surrogate model. A least-squares fitting method is implemented to identify the apparent static contact angle, and the droplet curvature, relative to the effective plane of the solid surface. The apparent contact angle and curvature of the droplet are then used as wetting metrics. The nanoscale contact angle is seen to vary significantly with the surface roughness. In the particular case studied, a variation of over 65° is observed between the contact angle on a flat surface and on a highly spiked (Cassie–Baxter) limit. This work proposes a strategy for systematically studying the influence of nanoscale topography and, eventually, chemical heterogeneity on the wettability of surfaces.

This work is framed within AIChE's 10 th Industrial Fluid Properties Simulation Challenge, with the aim of assessing the capability of molecular simulation methods and force fields to accurately predict the pressure dependence of the shear viscosity of 2,2,4-trimethylhexane at 293.15 K (20 °C) at pressures up to 1 GPa. In our entry for the challenge, we employ coarsegrained intermolecular models parametrized via a top-down technique where an accurate equation of state is used to link the experimentally-observed macroscopic volumetric properties of fluids to the force-field parameters. The state-of-the-art version of the statistical associating fluid theory (SAFT) for potentials of variable range as reformulated in the Mie incarnation is employed here. The potentials are used as predicted by the theory, with no fitting to viscosity data. Viscosities are calculated by molecular dynamics (MD) employing two independent methods; an equilibrium-based procedure based on the analysis of the pressure fluctuations through a Green-Kubo formulation and a non-equilibrium method where periodic perturbations of the boundary conditions are employed to simulate experimental shear stress conditions. There is an indication that, at higher pressures, the model predicts a solid phase (freezing) which we believe to be an artefact of the simplified molecular geometry used in the modelling. A comparison (made after disclosure of the experimental data) show that the model consistently underpredicts the viscosity by about 30%, but follows the pressure dependency accurately.