Modeling of oil extraction with supercritical CO2 from Atlantic mackerel (Scomber scombrus) at different moisture contents

Water is present in all natural materials at varying concentrations and plays an important role in the supercritical (SC) CO2 extraction of lipids from different matrices. The objective of this paper was to study the effect of moisture on oil extractability from Atlantic mackerel by SC-CO2 and to apply a mathematical model. Atlantic mackerel fillets were freeze-dried to different moisture levels (62.2, 40.5, 21.4, 10.1 and 3.7%, w/w) prior to extraction at 35 degrees C, 34.5 MPa. The amount of water in the extract fractions collected every 30 min was determined using the Karl-Fischer titration method. With a decrease in the moisture content of fish muscle, total amount of extract increased while the amount of water extracted over 7 h decreased. Effluent concentration of oil in SC-CO2 was close to its solubility, 2.67 kg m(-3) CO2 at extraction conditions for 3.7% moisture level. Oil solubility decreased significantly at moisture levels greater than or equal to 40.5%. Effluent water concentration of 0.3 kgm(-3) CO2 at 62.2% moisture was substantially below the solubility level of pure water. A mathematical model that incorporated the negative interaction between oil and water in the supercritical phase simulated the experimental data in the solubility-controlled region. Better understanding of the complex interactions between SC-CO2-oil-water-solid matrix will lead to improved process design and optimization of the SC-CO2 extraction processes for oil from natural matrices. (C) 1998 Elsevier Science B.V.