Improving vapour–liquid-equilibria predictions for mixtures with non-associating polar components using sPC-SAFT extended with two dipolar terms

• Published in: Fluid phase equilibria 2011-06, Vol.305 (2), p.174-184
• Main Authors: de Villiers, A.J., Schwarz, C.E., Burger, A.J.
• Format: Article
• Language: English
• Subjects: Adjustable; Aldehydes; Chemistry; Esters; Ethers; Exact sciences and technology; General and physical chemistry; Inclusions; Interaction parameters; Liquid-vapor equilibria; Mathematical models; Phase equilibria; Regression; sPC-SAFT-GV; sPC-SAFT-JC; Vapour–liquid equilibria
• ISSN: 0378-3812; 1879-0224
• DOI: 10.1016/j.fluid.2011.03.025

► sPC-SAFT extended with dipolar terms of Jog and Chapman ( JC) and Gross and Vrabec ( GV). ► Number of polar segments in sPC-SAFT-GV included as adjustable parameter. ► New model parameters for selected compounds for these EOSs are determined. ► Accurate VLE prediction of polar/alkane systems without interaction parameters. ► Similar performance for JC and GV models with improvement compared to sPC-SAFT. A considerable improvement in VLE predictions is obtained for mixture with non-associating polar components when the sPC-SAFT equation of state (EOS) is extended with the dipolar terms of Jog and Chapman (JC) and Gross and Vrabec (GV). These extended models are termed sPC-SAFT-JC and sPC-SAFT-GV, respectively. New model parameters for selected ketones, aldehydes, esters and ethers are presented for these two EOSs, as determined by the inclusion of pure component data and binary vapour–liquid-equilibria (VLE) data in the objective function of the regression procedure. Predictions of the VLE of polar/alkane and polar/polar systems show that most of these systems can be represented accurately by both models without using any binary interaction parameters (BIPs). Little difference in the performance of the two models is observed. In sPC-SAFT-GV, n p (number of polar segments) is included in the regression routine as an adjustable pure component parameter instead of using a default value of 1. This enables the model to capture the contribution of the polar term more correctly, resulting in more accurate VLE predictions.