Abstract
The present study is a step towards solving the problem of predicting the properties of a mixture in the entire PTx range, without resource to experimental data. Vapor-liquid phase equilibria and liquid-liquid equilibria are shown to be closely related. A novel semi-predictive approach (SPA) is introduced. It involves estimation of the binary adjustable parameters for a certain member of a homologous series using the Quantitative Global Phase Diagram method, and their implementation for predicting the data of other homologs. This approach has been combined with the equations of state (EOSs) of Peng-Robinson, Trebble-Bishnoi-Salim, and a four-parameter EOS (C4EOS), as described by Polishuk et al. (Chem. Eng. Sci. 55 (2000) 5705) for predicting data of carbon dioxide-alkanols. It is demonstrated that different isomeric configurations of alkanols do not affect the reliability of SPA. The predicted results exhibit the behavior characteristic for many series of alkanols, however, they are not accurate for carbon dioxide-methanol because of weak chemical interactions present in this system. C4EOS is more reliable predicting the topology of phase behavior, the critical, and the bubble-point lines. However, it is less exact for predicting the gas phase composition. The available experimental and theoretical results for carbon dioxide-alkanols are reviewed. The predictions of SPA usually exhibit a significant advantage over other methods that do not predict but correlate the data.
Original language | English |
---|---|
Pages (from-to) | 6485-6510 |
Number of pages | 26 |
Journal | Chemical Engineering Science |
Volume | 56 |
Issue number | 23 |
DOIs | |
State | Published - 30 Nov 2001 |
Keywords
- Alkanols
- Carbon dioxide
- Equation of state
- Parameter identification
- Phase equilibria
- Supercritical fluid
ASJC Scopus subject areas
- General Chemistry
- General Chemical Engineering
- Industrial and Manufacturing Engineering