Thermodynamic and kinetic parameters for the supercritical extraction of biphenyl from a contaminated soil

Carbon dioxide supercritical extraction is applied to a soil artificially contaminated with biphenyl. We studied three aspects of the extraction process in order to control it. An “excess function equation of state” model is used to represent the thermodynamic properties of the binary biphenyl-carbo...

Full description

Saved in:
Bibliographic Details
Published inChemical engineering science Vol. 51; no. 15; pp. 3861 - 3873
Main Authors Barna, L., Blanchard, J.M., Rauzy, E., Berro, C., Moszkowicz, P.
Format Journal Article
LanguageEnglish
Published Elsevier Ltd 1996
Subjects
Online AccessGet full text

Cover

Loading…
More Information
Summary:Carbon dioxide supercritical extraction is applied to a soil artificially contaminated with biphenyl. We studied three aspects of the extraction process in order to control it. An “excess function equation of state” model is used to represent the thermodynamic properties of the binary biphenyl-carbon dioxide mixture. This model allows calculation of solid compound solubility and of liquid-gas equilibrium. The adsorption of biphenyl in soil in the presence of supercritical carbon dioxide is experimentally studied. In order to demonstrate the influence of the non-ideal character of the supercritical fluid on the amount adsorbed, we used a Bragg Williams lattice model. The adapted model allows adsorption calculation for a given set of temperature-pressure-fluid composition data. Both thermodynamic models are necessary to shape the extraction behaviour. In the soil, the pollutant is distributed between a precipitated phase and an adsorbed phase. Thus, the extraction model contains a dissolution kinetic term and a desorption kinetic term. Three parameters are unknown: the dissolution rate coefficient, the overall mass transfer coefficient between soil particles and the fluid and the fraction of precipitated biphenyl from the total quantity in soil. These parameters are estimated by fitting the model with experimental data.
ISSN:0009-2509
1873-4405
DOI:10.1016/0009-2509(96)00223-0