Premium
Adsorption equilibria of fatty acids between methanol/water and reversed‐phase chromatographic adsorbents
Author(s) -
Ib ez Gonz lez M. J.,
Medina A. Robles,
Esteban Cerd n L.,
Camacho P ez B.,
Gim nez Gim nez A.,
Grima E. Molina
Publication year - 2001
Publication title -
journal of the american oil chemists' society
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.512
H-Index - 117
eISSN - 1558-9331
pISSN - 0003-021X
DOI - 10.1007/s11746-001-0257-4
Subject(s) - chemistry , adsorption , methanol , chromatography , acetic acid , solvent , freundlich equation , phaeodactylum tricornutum , fatty acid , polyunsaturated fatty acid , eicosapentaenoic acid , organic chemistry , botany , biology , diatom
The adsorption equilibria are discussed for fatty acids 16∶1n−7, 16∶2n−4, and 20∶5n−3 (eicosapentaenoic acid, EPA). These fatty acids are major components of a polyunsaturated fatty acid concentrate from the microalga Phaeodactylum tricornutum . The solvents used were methanol/water (1% acetic acid) mixtures of different compositions, and the adsorbents used were chromatographic reversed‐phases octylsilyl C 8 , octadecylsilyl C 18 , and dodecylsilyl C 22 of different particle and pore sizes. The kinetic studies showed that equilibrium was attained instantaneously, suggesting an absence of mass transfer limitations. The equilibrium data were fitted by the Freundlich isotherm. The separation efficiency of EPA from 16∶1n−7 and 16∶2n−4 in all the adsorbent‐solvent systems was compared in terms of the separation factors α EPA/16:2n‐4 = K EPA / K 16:2n‐4 and α 16:1n‐7/EPA = K 16:1n‐7 / K EPA , where K i is the fatty acid distribution ratio between the stationary and the liquid phases. The EPA separation from 16∶1n−7 and 16∶2n−4 by liquid chromatography could be predicted using the Craig model for the various solvent‐adsorbent combinations. The best adsorbents for purifying EPA were: C 18 , PEP, 8 μm, 100 Å, and C 22 , 10 μm, 100 Å, and the best solvent was methanol/water (1% acetic acid) 75∶25, w/w.