The Effect of Binary Interaction Parameters and Induced Association on the Performance of the PC-SAFT Equation of State: Validation against a Benchmark Database of High-Quality Binary-System Data

The performance of the Perturbed Chain Statistical Associating Fluid Theory (PC-SAFT) equation of state (EoS) using classical mixing rules, induced-association schemes for mixtures exhibiting cross-association phenomena between two components that do not self-associate, and one regressed temperature independent binary interaction parameter (kij) per mixture is assessed using a high-quality reference database containing binary-system data for various properties. The binary mixtures that are included in the database are divided into nine categories (Binary Association Codes, BACs) according to their ability to be involved in a hydrogen bond. Grading of the EoS is done by applying an already published procedure with certain modifications that were also published recently. An important feature is the inclusion of a performance indicator, called the "Success Ratio" (SR), that quantifies the percentage of experimental data points which the model can qualitatively reproduce. The original marks of the PC-SAFT EoS when implemented with classical mixing rules, no induced association scheme, and a kij parameter set equal to zero are recalculated with the modified grading procedure and are directly compared with the new marks obtained in this work. The effect of including induced-association effects in PC-SAFT for mixtures that belong to BAC6, BAC7, and BAC8 (mixtures in which cross-association takes place without both components being self associating) is also investigated. The effect of regressing cross-association parameters vs regressing kij parameters for mixtures belonging to the BAC6 group (mixtures in which cross-association arises although the two components do not self-associate) is showcased. The final optimized PC-SAFT EoS utilizes classical mixing rules, induced-association effects (using regressed cross association parameters for some binary systems belonging to BAC6 and a predictive method for BAC7 and BAC8), and one temperature-independent kij parameter per binary mixture (except for binary systems belonging to BAC6 for which cross-association parameters were adjusted). The final average mark of the optimized EoS is 9.3/20.