Supercritical extraction and local composition equation of state
Extraction equilibria of two liquids (1-methylnaphthalene and m-cresol) separately with two supercritical solvents (carbon dioxide and ethane) were determined in a semi-flow apparatus. For each system, measurements were performed at three temperatures, 308.2, 318.2 and 328.2 K, and pressures from 15 up to 240 bar. The experimental data correlated with the local composition embedded Soave equation of state. The local composition was determined for mixtures of square-well molecules of various sizes and energies by Monte Carlo simulation. The coordination numbers, internal energy, and pressure were also determined from the simulation. The calculations covered molecular diameter ratios up to two and a wide density range from a dilute gas to dense liquids for various interaction energies. A model was proposed to express the coordination numbers as a function of molecular diameters, characteristic energies, temperature, and density. From the expression of the coordination numbers, a model equation of state, internal energy, and Helmholtz free energy were obtained for squarewell fluids. Several local composition models were examined with the simulated data, leading to the development of a new density dependent local composition model. The proposed model was found to better describe the simulated results than the other previous models. The local composition model obtained from computer simulation of molecular fluid mixtures was embedded in the Soave, Patel-Teja, and Cubic Chain-of-Rotators equations of state. Representation of vapor-liquid equilibrium with the embedded equations was investigated at high pressures as well as low pressures for diverse polar mixture systems including aqueous, non-aqueous, hydrogen-bonded, or not hydrogen-bonded. Extensive comparison with data shows that equations of state with local composition mixing rules improve the representation of vapor-liquid equilibria of highly polar or non-ideal mixtures. The local composition embedded Soave equation is able to correlate the new extraction data of this work well.
Chao, Purdue University.
Off-Campus Purdue Users:
To access this dissertation, please log in to our