Supercritical fluid extraction from dandelion seeds

Investigations of the process of supercritical fluid extraction (SFE) from dandelion were so far mainly focused on extractions from herb, leaves and root [1-3]. In the work presented here, the extraction process from seeds was investigated on a laboratory and semi-industrial scale equipment. The milled plant material with an average particle size of 0.41 ± 0.02 mm was used in all experiments. All SFEs were conducted with carbon-dioxide in supercritical state (SCCO2) as a solvent on pressure of 450 bar and temperature of 60°C, as these parameters were previously shown to be optimal for obtaining high-value extracts from dandelion seeds [4]. Experiments were conducted on laboratory (dia.=7.7 cm, vol=0.56 l, CO2 flow=11 kg/h) and semiindustrial (dia.=23.5 cm, vol.=397 l, CO2 flow=240 kg/h) scale equipment. On both scales several experiments were conducted with initial mass of the plant material (Ms) being the only variable parameter: laboratory scale Ms=26–104 g; semi-industrial scale Ms=2.2–8.8 kg. Fractions of extract were collected during the extractions in order to obtain data on the kinetics of the SFE processes. The results of the performed experiments are depicted in Figure 1. It can be seen that for both scales the rate of SFE, as well as the final yield, increases with increase of the mass of plant material, indicating that with the selected conditions, namely flowrate of CO2, the residence time was not long enough to allow saturation of the CO2 in the first part of the process, nor to enable extraction of all available extract. Similar trends on both scales can be perceived as a good basis for further investigations aiming at scale-up of the process of SFE from dandelion seeds. Experiments with larger packed bed height, which would enable higher residence times, should be conducted in order to achieve saturation at least in the first part of the SFE process. Mathematical models of the SFE process, supported by models for estimation of the kinetic parameters, should be applied to gain full insight into the phenomenology of the SFE process