Fundamental Researches Of Separating EPA-EE And DHA-EE By Supercritical Fluid Chromatography

EPA (Eicosapentaenoic acid ) and DHA (Docosahexaeonoic acid) are two ω -3 fatty acids. They are vital for the retina of the human eyes and for the nervous system, and can reduce the risk of cardiovascular disease and inflammatory disease as well. However, the functions of EPA-EE and DHA-EE are different, so it is significant to separate them into their pure compound form. Among the separation techniques in the review, supercritical fluid chromatography(SFC) is regarded as a promising one. Though some investigations on separation of EPA-EE and DHA-EE by SFC have been made, all of them focused mainly on the process itself. Thus, it is necessary to make fundmental researches on supercritical fluid chromatography (such as adsorption equilibria, diffusion and mass transfer) systematically, making the separation process base on a reliable fundamental. To date, there is nearly no systematic research about the above issues, so it is of great value, in a sense of academe, to research the fundamental of separation EPA-EE and DHA-EE by SFC. Some basic researches have been conducted in this work. It includes the following aspects:Diffusion Measurement. The binary diffusion coefficients (D_1,2) of AA-EE, EPA-EE and DHA-EE in supercritical carbon dioxide were studied by Taylor capillary peak broadening (CPB) method at 308.15 K to 338.15 K under 8.42 MPa to 29.95 MPa. The D_1,2s were in the ranges of (5.54 to 13.47)×10^-5cmV, (5.54 to 13.80)× 10^-5 cm²Â·s^-1 and (5.40 to 12.80)× 10^-5cm²Â·s^-1 respectively. It decreased with the increasing density of carbon dioxide and temperature. The curves of D_1,2/T vs. density of carbon dioxide almost coincided with each other. The D_1,2s for the esters of the three fatty acids are greater than those of the fatty acids themselves. The Scheibel, Catchpole-King, and He-Yu-Su equations predicted the experimental data quite well with average deviations smaller than 2 %.Adsorption Equilibria Research. A static apparatus which could determinatethe adsorption equilibrium data on-line and cost fewer sample was set up. The adsorption equilibrium of EPA-EE and DHA-EE on C18-bonded silica gel from supercritical carbon dioxide was studied on this set-up. The temperature was in the range of 318.15 K to 338.15 K and the pressure was in the range of 9.98 MPa to 21.97 MPa. The results indicated that the loading of EPA-EE or DHA-EE decreased with increasing temperature and density, and the isotherms were fitted well by BET equations. Under the conditions in this work, the monomolecular saturated capacity of these two fatty acids were (0.039 to 0.048) mmol·g^-1 and (0.040 to 0.063) mmol·g^-1 respectively. The heat of these two fatty acids on C18-bonded silica gel decreased with greater loading and density until it leveled off (at about 10 kJ·mol^-1). The adsorption data of EPA-EE and DHA-EE on silica gel from supercritical carbon dioxide were measured by the method of elution by character point (ECP). The temperature was in the range of 319.15 K to 340.45 K and the density was in the range of 0.668 g·ml^-1 to 0.734 g·ml^-1. The results showed that the isotherms satisfied the Langmuir equation. The equilibrium constant K decreased with the increasing density and temperature. Under the conditions in this work, the monomolecular saturated capacity of EPA-EE and DHA-EE were in the ranges of (0.125 to 0.144) mmol·g^-1 and (0.113 to 0.131) mmol·g^-1 respectively. The monolayer saturated capacities per square meter for C18-bonded silica gel and silica gel were very close and were in the ranges of 2.8×10^-7 mol·m^-2 to 3.9×10^-7 mol·m^-2.Mass Transfer Research. The lump kinetic model was adopted to describe the chromatographic behavior of EPA-EE and DHA-EE in supercritical carbon dioxide. By fitting the experimental elution curves with the lump kinetic model, the effect of mobile phase flow rate, temperature and pressure on mass transfer kinetic was studied. The results indicated that the theoretical plate number (N) had a maximum value with the increasing flow rate and decreased with greater pressure and lower temperature. The lump mass transfer coefficient (k_f) increased with the increasing flow rate when the flow rate was small and then leveled off when the flow rate was great. Increasing pressure or decreasing temperature makes the theoretical plate number smaller. The contribution of axial dispersion to the theoretical plate height was in the range of 3.2% to 16.9% and decreased with greater pressure and lower temperature.SFC Separate EPA-EE and DHA-EE. An effective fraction collection method (solid trapping) for semi-preparative supercritical fluid chromatography was developed. The recovery exceeded 95 % when the pressure in the collection column was in the range of 4 to 6 MPa. By this method, chromatographic characteristics on a semi-preparative chromatographic column (250mm×10mm I.D.) under various loadings offish oil and flow rate of carbon dioxide were determined. It was found that EPA-EE and DHA-EE were well separated, but other esters of fatty acids presented in the feed could not be separated. By appropriately cutting the effluent, EPA-EE with 90 % content and DHA-EE with 75 % content were obtained. Based on the chromatographic curves, the effect of loading and flow rate of mobile phase on the performance of the column was studied. At EPA-EE content greater than 90 %, recovery of EPA-EE decreased remarkably as the loading increased. The relationship between CO₂ consumption and loading had a minimum consumption at loading of 4.076 ml·l^-1. The optimal loading and flow rate for preparation of 90 % EPA-EE were 7.13 ml·l^-1 and 3.928 g·min^-1 under 12.1 MPa and 55℃. Chromatographic conditions were scaled up directly on a preparative column (250mmx50mm I.D.) loading feed with fewer impurities. The results showed that the component of the feed affected the product greatly.RP-HPLC Separate EPA-EE and DHA-EE, EPA-EE and DHA-EE were separated by reverse-phase HPLC. The effect of ratio of methanol/water, flow rate and column temperature on separation were studied. The optimal ratio is methanol/water (v/v) = 88:12. The chromatographic conditions were scaled up on a preparative column. Economic analyses and comparisons for separation by HPLC and SFC were conducted.