Seed oil supercritical extraction: Particle size distribution of the milled seeds and modeling

The paper is devoted to modeling seed oil supercritical extraction introducing the novelty of considering the granulometric distribution of the substrate to be extracted. Reference is done to a model which, as the others available in the literature, simplifies the milled seed particles as having a unique representative average dimension, the Sauter mean diameter. At first, the new developed model and the reference model were compared on reference test cases for the prediction of the kinetics of extraction. Then the new model was utilized with experimental granulometric distributions of milled grape seeds and with the relevant experimental extraction data and it was tested also for what concerns the calculation of the effective diffusivity of the oil into the seed matrix.The results obtained show that adopting the Sauter mean diameter instead of the real particle diameters is a consistent simplification where the particle granulometric distribution presents small variance (error of a few percentages). In any case the new model predicts slightly better that the reference one and significant differences occur when calculating the internal mass transfer coefficient.

Graphical abstractThe paper is devoted to modeling seed oil supercritical extraction introducing the novelty of considering the granulometric distribution of the substrate to be extracted. The results show that modeling adopting the Sauter mean diameter instead of the real particle diameters is a consistent simplification only where the particle granulometric distribution presents small variance (error of a few percentages). Ratio between extraction yields for test case granulometric distribution and the relevant Sauter mean diameter.Figure optionsDownload full-size imageDownload as PowerPoint slide