Laboratory data to develop industrial supercritical CO2 separations of biological substrates: Extraction of lipids from oilseeds

By Dr. José Manuel del Valle

Pontificia Universidad Católica de Chile, Chile

Wednesday August 29th, 2018 at 10:00 am (Mexico City, MX; UTC -05:00)

The extraction of valuable compounds from biological substrates using supercritical (sc) carbon dioxide (CO₂) as a replacement of conventional organic solvents has been an industrial reality for almost four decades. However, users have been reluctant in adopting scCO₂ technology because it is perceived as a noncompetitive alternative. This presentation seeks to disprove this misconception by addressing the economics of scCO₂ oil extraction from prepressed seeds. My four-part presentation will begin with a description of the fundamentals of the scCO₂ extraction of biological matrices. Next, I’ll describe the mathematical simulation of industrial scCO₂ plants for the extraction vegetable oils. This application can be used as a case study because of the availability of a mathematical model of the extraction process that can be applied for process simulation purposes. Pre-pressed oilseeds have an interconnected network of pores that are filled with part of the oil expelled from fractured cells during pressing. The shrinking-core model hypothesis applies to mass transfer in this substrate having as a parameter the effective diffusivity (D_e) within the pore network. D_e relates to a particle-size and scCO₂-condition-independent, and pretreatment dependent microstructural factor. The shrinking core model is partially predictive in that there are published dimensionless correlations for the film mass transfer and axial dispersion in packed beds operating with supercritical fluids, and for the solubility of vegetable oils in scCO₂. Thirdly, I’ll describe the economics of industrial scCO₂ extraction plants. This will include both capital and operational costs. Finally, I’ll exemplify the use of mathematical simulation and cost analysis for the optimization of several process parameters. Examples will include substrate particle size, superficial CO₂ velocity, and size and number of extraction vessels.