Reliability of the correlation allowing the kij to switch from an alpha function to another one in hydrogen-containing systems

• Published in: Fluid phase equilibria Vol. 338; pp. 23 - 29
• Main Authors: Jaubert, Jean-Noël, Qian, Junwei, Privat, Romain, Leibovici, Claude F.
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 25.01.2013; Elsevier
• Subjects: Alpha functions; Binary interaction parameters; Chemical and Process Engineering; Chemical Sciences; Cubic equation of state; Engineering Sciences; Huron–Vidal mixing rules; Hydrogen; or physical chemistry; Peng–Robinson; PPR78; Soave–Redlich–Kwong; Theoretical and; Van Laar excess Gibbs energy model; Cubic equation of state; Alpha functions; Soave–Redlich–Kwong; Hydrogen; Peng–Robinson; Huron–Vidal mixing rules; Van Laar excess Gibbs energy model; Binary interaction parameters; PPR78; Hydrogen Cubic equation of state Binary interaction parameters Alpha functions Peng-Robinson PPR78 Soave-Redlich-Kwong Van Laar excess Gibbs energy model Huron-Vidal mixing rules
• ISSN: 0378-3812; 1879-0224
• DOI: 10.1016/j.fluid.2012.10.016

Due to its very low critical temperature, the reduced temperature of hydrogen can easily overcome 10 in chemical processes. As a consequence, the many alpha functions available in the literature – the parameters of which are determined at low temperature – lead to totally different aH2(T) values in the temperature range of industrial interest. As a consequence, the kij value used in the classical VdW1f mixing rules to correlate the VLE data of hydrogen-containing systems strongly depends on the selected equation of state (EoS) and/or alpha function. For hydrogen-containing systems, the kij may indeed vary from a large negative value to a large positive value by only changing the alpha function. This is a major concern since commercial process simulators generally provide a unique kij value which is assumed to be appropriate regardless of the equation of state and/or the alpha function. It was thus a challenge to check whether the relationship developed by our research group in 2010 – and aimed at calculating the kij for a desired thermodynamic model knowing the corresponding value for another one – could predict such large changes. This study concluded that such a correlation performed well and could thus be useful to simulate chemical processes in which H2 is a key component.