Extensive Study of the Capabilities and Limitations of the CPA and sPC-SAFT Equations of State in Modeling a Wide Range of Acetic Acid Properties

• Published in: Industrial & engineering chemistry research 2018-04, Vol.57 (16), p.5690-5704
• Main Authors: Ribeiro, Rafael T. C. S, Alberton, André L, Paredes, Márcio L. L, Kontogeorgis, Georgios M, Liang, Xiaodong
• Format: Article
• Language: English
• ISSN: 0888-5885; 1520-5045
• DOI: 10.1021/acs.iecr.8b00148

The accurate description of many thermodynamic properties of acetic acid and its mixtures can be a challenge to model with standard thermodynamic models such as local-composition activity coefficient models and cubic equations of state. A possible solution is offered by association equations of state, e.g., those belonging to the statistical associating fluid theory (SAFT) family. While several researchers have studied the use of SAFT variants to model acetic acid properties (pure compound and mixtures), with few exceptions, those studies focused exclusively on phase equilibria including vapor pressure and density. Other important properties, such as the speed of sound, second virial coefficient, compressibility factor, enthalpy of vaporization, and isobaric heat capacity have not been considered yet. Few studies investigate which is the appropriate association scheme to be used for acetic acid. In this work, we compare the capabilities of two association models, cubic plus association (CPA) and simplified perturbed-chain SAFT (sPC-SAFT), to predict a wide range of properties of acetic acid (mixtures) including derivative properties. We evaluate the influence of including one or more of those properties in the parameter estimation procedure for sPC-SAFT, we compare the results obtained with various choices of the association schemes (one or two sites), and finally we evaluate the performance of CPA and sPC-SAFT on correlating the phase equilibria of the binary mixtures of acetic acid with water, hexane, or ethanol. It is concluded that both equations of state perform overall similarly, with the one-site scheme performing better overall, especially for some properties. The results of the evaluation also show that some properties are intercorrelated in the parameter estimation process, making it essentially impossible to obtain sets that can accurately describe all the properties of acetic acid.