J. of Supercritical Fluids 102 (2015) 80–91 Contents lists available at ScienceDirect The Journal of Supercritical Fluids j o ur na l ho me page: www.elsevier.com/locate/supflu Equilibrium partition of rapeseed oil between supercritical CO 2 and prepressed rapeseed Freddy A. Urrego a , Gonzalo A. Nú˜ nez b , Yazmin D. Donaire a , José M. del Valle a,c,∗ a Department of Chemical and Bioprocess Engineering, Pontificia Universidad Católica (UC) de Chile, Avda. Vicu˜ na Mackenna, 4860 Santiago, Macul, Chile b Department of Chemical and Environmental Engineering, Universidad Técnica Federico Santa María, Avda. Vicu˜ na Mackenna, 3939 Santiago, San Joaquín, Chile c ASIS-UC Interdisciplinary Research Program on Tasty and Healthy Foods, UC, Santiago, Chile a r t i c l e i n f o Article history: Received 15 December 2014 Received in revised form 22 March 2015 Accepted 3 April 2015 Available online 14 April 2015 Keywords: Carbon dioxide Equilibrium partition Sorption isotherm/isobar curve Rapeseed oil Prepressed rapeseed Supercritical a b s t r a c t The equilibrium partition of a vegetable extract between supercritical (SC) CO 2 (fluid phase) and a veg- etable substrate (solid phase) is commonly best-fitted in the mathematical simulation of extraction curves. The aim of this study was to develop and apply an experimental methodology to measure the equi- librium partition of rapeseed oil between SC CO 2 and prepressed rapeseed. We measured and modelled sorption isotherm/isobar curves using a methodology that intersperses extraction (to reduce oil content) and equilibration (by recirculation of the SC CO 2 phase) steps, with oil being sampled in each step. Oil desorption was a two-stage process; when rapeseed contains more than ∼70–80 g kg -1 oil/substrate the SC CO 2 phase gets saturated with oil (C f = C sat ), whereas when it contains less oil this is bound to the solid matrix in such a way that C f dips below C sat . Increasing pressure from 22 to 28 MPa at constant temperature (40 ◦ C) increases C f . The effect of the increase in pressure is less pronounced when SC CO 2 density is kept constant (857.1 kg/m 3 ) by simultaneously increasing temperature from 40 to 50 ◦ C. The heat of desorption of the oil is ∼100 kJ/mol which suggests it is bound to the prepressed seed by van der Waals interaction forces. However experimental values in this region are uncertain due to propagation of uncertainties to estimate C s . © 2015 Elsevier B.V. All rights reserved. 1. Introduction Industry requires reliable mathematical descriptions of Super- Critical (SC) Fluid (SCF) Extraction (SCFE) to facilitate process design, scale-up, and costing. SCFs, particularly carbon dioxide (CO 2 ) above its corresponding critical temperature and pressure, have been successfully applied to extract high-value-added com- pounds for the food, cosmetic, and pharmaceutical industries [1]. Reliable mathematical descriptions of SCFE of vegetable substrates depend critically on the determination of equilibrium and mass transfer parameters, including those characterizing solute parti- tion between SC CO 2 and the pretreated substrate. Mass transfer parameters include an axial dispersion coefficient (D L ) describing solute migration in the SCF phase moving along the packed bed, a film mass transfer coefficient (k f ) describing transport of the solute ∗ Corresponding author at: Pontificia Universidad Católica (UC) de Chile, Depart- ment of Chemical and Bioprocess Engineering, Avda. Vicu ˜ na Mackenna, 4860 Santiago, Macul, Chile. Tel.: +56 2 23544418; fax: +56 2 23545803. E-mail address: delvalle@ing.puc.cl (J.M. del Valle). from solid particles to the SCF phase through a thin layer of SCF stationed next to the particles, and an internal mass transfer coef- ficient (e.g., effective diffusivity, D e ) describing the transport of solute within solid particles under the influence of solute gradients imposed by the removal of solute at the surface of the particles [2,3]. Equilibrium parameters correspond to those of a sorption isotherm/isobar equation relating solute concentration in the SC CO 2 phase (C f ) with solute concentration in the solid (pretreated vegetable substrate) phase (C s ) [4]. Parameters D L and k f can be estimated using dimensionless correlations for packed beds oper- ating with SCFs [2,3]. On the other hand, D e and the parameter(s) of a simple sorption isotherm/isobar equation (e.g., a constant par- tition coefficient K) are usually estimated by best-fitting integral extraction curves (plots of cumulative yield, e.g., grams of solute per kilogram of substrate, versus specific solvent consumption, e.g., kilograms of CO 2 per kilogram of substrate) obtained in laboratory or pilot-plant units. Although it may be possible for a mathemat- ical model that does not fit reality to describe integral extraction plots when model parameters are best-fitted to experimental data, the best-fitted model will probably fail in simulating extractions at larger scales, including industrial extractions [4]. http://dx.doi.org/10.1016/j.supflu.2015.04.004 0896-8446/© 2015 Elsevier B.V. All rights reserved.