Mathematical simulation of heat and mass transfer during controlled depressurization of supercritical CO2 in extraction vesselsRevista : Journal of Supercritical Fluids
Volumen : 122
Páginas : 43-51
Tipo de publicación : ISI Ir a publicación
Even after almost forty years of industrial application, companies are still reluctant to use supercritical (sc) CO2 as a solvent for extractions due to the perceived high production costs. Literature on the matter suggests that, because extraction at high pressure needs to be done in batches, using multiple extraction vessels with simulated-countercurrent flow could reduce operational costs. However, the more extraction vessels used, the less time there is to recondition them in order to have a semi-continuous operation; and if the reconditioning, and particularly the depressurization, is done too fast, the vessel could become brittle and permanently damaged. With the goal of optimizing the depressurization process in mind, numerical simulation of temperature and mass was carried considering a 1-dm(3) vessel filled with a packed bed made with model materials using a mass flow function that depends on the chocked mass flux and a valve opening area. The correlation Nu = 0.0777 Da(-0.373) Ra-0.397 was obtained for convective heat transfer at the vessel wall, and temperature, pressure, and vented mass flow were simulated with about 20% improvement in predictions in comparison to our previous correlation. To explore the use of the model for practical purposes, it was used to simulate depressurization processes with volumes up to 1 m(3) and with different initial conditions and vessel geometries so as to have a first approach on the effect of these parameters on the depressurization time. Simulated depressurization times reached a maximum value of 54.5 min for depressurizations of a 1-m(3) extraction vessel starting at 60 degrees C and 70 MPa, which are very plausible extraction conditions. This model can be used to determine optimal reconditioning time in industrial plants for cost minimization.