Article ID Journal Published Year Pages File Type
232266 The Journal of Supercritical Fluids 2006 13 Pages PDF
Abstract

Predictive models that describe supercritical fluid extraction (SCFE) processes would be welcomed to support its industrial application, particularly for the supercritical carbon dioxide (SCCO2) extraction of vegetable oil from seeds subjected to common high-shear pretreatments. This work explores the application of microstructure-extractability relationships for modeling the SCFE of lipids from vegetable substrates. We measured the extraction kinetics of prepressed rapeseeds, olive husks, and flaked rosehip seeds, with SCCO2 at 313 K and 30 MPa, and simulated the extractions using a shrinking-core model. Model parameters included the oil solubility and film mass transfer coefficient from literature correlations, and an effective diffusivity (De) inside the porous particles. We determined that De could be calculated as D12 × F, where D12 is the diffusivity of oil in CO2, and F is a microstructural correction factor, estimated as the ratio between the final porosity (ɛp, from Hg porosimetry) and pore-network tortuosity (τ, from fractal-texture analysis of binary light-microscopy images) of the substrates. Simulations adjusted the experimental data reasonably well (5.4% < mean percent error < 15%). Additionally, best-fit estimates of De were obtained for literature data on SCCO2 extraction of lipids from prepressed and flaked seeds. Resulting values of F did not depend on particle size and spanned a narrow range – one order of magnitude (0.030–0.29) – as it would be expected when comparing similar systems. Although further work will be required to refine the relationship between τ and fractal parameters, or between τ and the hysteresis of Hg infiltration, this work demonstrates that is possible to develop predictive models for SCFE of solid substrates subjected to high-shear pretreatments.

Related Topics
Physical Sciences and Engineering Chemical Engineering Chemical Engineering (General)
Authors
, , , , ,