Heterosegmented Perturbed-Chain Statistical Associating Fluid Theory as a Robust and Accurate Tool for Modeling of Various Alkanes. 1. Pure Fluids

Heterosegmented molecular models based on statistical associating fluid theory (SAFT) seem to be very promising and robust tools for modeling thermodynamic properties of fluid mixtures. They differ from conventional SAFT-based methodologies as they take into account varying sizes and interactions of segments constituting chain molecules. Those different types of segments can be assigned to functional groups, and hence, the group contribution (GC) method is incorporated in a straightforward manner into the SAFT approach. In this contribution, we applied a heterosegmented version of perturbed-chain SAFT (hs-PC-SAFT) for modeling thermodynamic behavior of a great variety of pure saturated hydrocarbons, including n-alkanes, branched alkanes, and alkyl-monosubstituted cyclohexanes and cyclopentanes. All the investigated compounds were assumed to be composed of 11 distinct functional groups defined within the GC model. The properties under consideration were saturated liquid and vapor density, vapor pressure, enthalpy of vaporization, surface tension, isobaric heat capacity, speed of sound, the Joule-Thomson coefficient, the Joule-Thomson inversion curve, and the second virial coefficient. The respective model parameters (segments number m(i), segment diameter sigma(i)) for each defined functional group i as well as self- and/or cross-interaction parameters (u(ij)/k(B)) for each pair of groups i-j were determined by fitting experimental saturated liquid densities and vapor pressures of some selected alkanes over a wide range of temperature. Then, the optimized parameters were used to predict the properties of other compounds, and finally, the resulting predictions were compared to those obtained by using different similar methods described in the literature. It was shown that an overall predictive capacity of the hs-PC-SAFT approach is comparable to other similar methods based on variable-range SAFT (SAFT-VR) and superior over the conventional homosegmented SAFT models involving groups contributions.