New Procedure for Enhancing the Transferability of Statistical Associating Fluid Theory (SAFT) Molecular Parameters: The Role of Derivative Properties

Abstract

Here, we present a simple method for optimizing the fitting of molecular parameters involving vapor-liquid equilibria (VLE) and selected second-order thermodynamic properties and experimental data. The procedure is applied, as an example to the soft Statistical Associating Fluid Theory (soft-SAFT) equation of state. The method involves the introduction and testing of coupling factors ranging from 0 (only one selected derivative property) to 1 (only VLE), to change the weight of one set of properties over the other in the fitting procedure; this allows one to assess the role of derivative properties in the robustness of the parameters and molecular model. The technique is illustrated by calculating a large number of thermodynamic properties of different compounds: the n-alkanes, n-perfluoroalkanes, and 1-alkanols families, and water, as representative of different types of molecular interactions, and in a wide range of thermodynamic conditions. The most relevant thermodynamic properties to be included in the parameters fitting, the weight they should have in the regression procedure, and the influence of association are discussed in detail. It has been found that the use of just VLE data in the fitting procedure for regular nonassociating compounds, such as n-alkanes or perfluoroalkanes, is enough to provide good predictions of derivative properties. The use of derivative properties helped in identifying the best fitting strategy for the perfluoroalkanes family, as some limitations of the original parameters showed up when used for estimating the derivative properties. In contrast, association plays a major role in derivative properties of associating compounds such as alkanols and water. A coupling factor of 0.5 (equal weight of derivative property coupled with VLE data to fit the molecular parameters) is required to regress robust pure compound parameters that are able to simultaneously describe vapor pressures, phase densities, and derivative properties for 1-alkanols and water. The method presented here is robust, simple, and straightforward to implement; since it is not equation-dependent, it can be transferred to any other equation, providing robust molecular parameters for global equations.