Supercritical fluid techniques as preparative- and analytical tools in the “Analytical Grey Area” between volatiles and hydrophilic compounds
Chemistry Department, Royal Veterinary and Agricultural University
Thorvaldsensvej 40, DK-1871 Frederiksberg C, Denmark
Lipophilic-amphiphilic compounds with properties between volatiles and hydrophilic compounds often create problems in connection with their isolation and analytical determination, resulting in an “analytical grey area”. Supercritical fluid techniques (SFT) comprising supercritical fluid extraction (SFE) and supercritical fluid chromatography (SFC) seem to have the potential to cover this. The basis developed for use of SFE and SFC will be presented as will selected examples of these techniques used in evaluation of oil quality and for determination of lipophilic-amphiphilic compounds in products from cruciferous crops. It is found that SFT allows an efficient separation of the intact triacylglycerols in the oil. SFC has also been found to give relatively fast and simple procedures for determination of oil constituents as chlorophyll and its derivatives, carotenoids, tocopherols, phenolics and other oil constituents, which may be important for the oil quality. Thereby it gives a tool to determine the origin of the oil and improved possibilities of determination of the relations between oil constituents and physical-biochemical properties of the oil.
KEYWORDS Cruciferous crops, Oil quality, Supercritical Fluid Chromatography, Supercritical Fluid Extraction, Triacylglycerols
Supercritical fluid techniques (SFT) comprising supercritical fluid extractions (SFE) and supercritical fluid chromatography (SFC) have been known for many years but are yet not used as widespread standard techniques at universities or in the industry. However, it is expected that the use of these methods will expand in the years to come due to the growing number of applications and the continuing development of commercially available instruments (Bøwadt and Hawthorne, 1995; Chester, 1997; Anklam et al, 1998).
Figure 1. The phase diagram illustrating the supercritical region above the critical pressure (Pc) and temperature (Tc). CP = critical point, TP = triple point.
Supercritical fluid techniques use high density gases / fluids, which are formed when the pressure and temperature of the gas are higher than the critical point, where the density increases dramatically due to the high compressibility near this point (Figure 1). SFE and SFC mainly employ a mixture of a compressed gas, like carbon dioxide, and a modifier, like alcohols, as the solvent / mobile phase, of which the modifier acts to enhance the solvent strength of the fluid. Some of the advantages of the supercritical fluids are the lower viscosity and higher diffusivity and selectivity compared to liquids in addition to gentle working conditions.
In the present paper, both SFE and SFC are used as fast methods for quantitative and qualitative determination of triacylglycerols and other oil constituents in various oil seeds used in the industrial oil production, including rape seed with both high and low erucic acid content.
High and low erucic acid rape seed, sunflower and soya beans were obtained from a local market and crushed before extraction. Mature olive seeds were obtained from Spain, freeze-dried and ground before extraction.
All seeds were extracted by SFE using a Speed SFE with 10 mL extraction vessels (Applied Separations, Allentown, PA). Grounded seeds were extracted at 75oC and 60 MPa using pure carbon dioxide (99.7 %). The total extraction time was 20 to 30 min, depending on the oil content and extracted oils were collected in bottles and weighed (Buskov et al., 1997a, b). Before qualitative analysis by SFC, the oils were diluted in hexane to a concentration of approximately 3 mg/mL. The analyses were performed on a standard Gilson SF3 Supercritical Fluid Chromatography system. Separation conditions were as follows: Flow rate: 1.5 mL/min; Mobile phase: 96 % carbon dioxide (99.998 %) and 4 vol % (3.4 mol %) acetonitrile; column outlet pressure: 30 MPa; column temperature: 40oC. The separation column was a Spherisorb S3 ODS2 (150 x 4.6 mm, 3 μm particles) and UV-detection was performed at 205 nm (Buskov et al., 1999a).
Use of supercritical fluid extraction as method for determination of total lipid content in seeds seems to be a promising method compared to Soxhlet-extraction with e.g. hexane or ether. A quantitative extraction is often done in less than 30 min compared to 6 to 20 hours for a normal Soxhlet extraction. It has been found that the amount of co-extracted amphiphilic compounds like phospholipids is up to 100 times lower than found for oils extracted by Soxhlet where the phospholipid content may account for up to 1 – 2 percent of total oil extracted (Buskov et al., 1997b).
Analysis of triacylglycerols from oil seeds and other sources are traditionally performed by analysis of the corresponding fatty acid methyl esters using GC. However, valuable information is often lost in this process, i.e. information about the individual intact glycerol esters, occurring both as mono-, di- and mainly triacylglycerols. Direct chromatographic analysis of the intact plant oils using high-resolution methods like capillary or packed column supercritical fluid chromatography (pSFC) gives a much more detailed picture as shown in Figure 2, where SFC-chromatograms for selected oils are illustrated. One of the disadvantages by using pSFC is the UV-detection, which gives a low sensitivity due to the lack of good chromophores in the triacylglycerols. The presence of one or more double bonds in the fatty acid chains is required and therefore makes detection of saturated triacylglycerols difficult.
Figure 2. SFC-chromatograms of various seed oils extracted by SFE. Separation conditions as described in the experimental section. Note that the separation time for high erucic acid rape seed and for Brassica campestris are 40 min compared to 20 min for other seed oils. Labels: Palmitic acid (P), stearic acid (S), oleic acid (O), linoleic acid (L), linolenic acid (Ln), erucic acid (E).
All individual triacylglycerols are not fully separated but still the method provides sufficient information for identification of the oils. As shown, the chromatograms for the high erucic acid oils (high erucic acid rape seed and Brassica campestris) differs from the other oils. The dominating triacylglycerols in these oils are O-O-E, O-E-E and E-E-E (O = 18:1; E = 22:1) whereas, as expected, the soya bean and sunflower oils are dominated by triacylglycerols rich in linoleic acid, like L-L-L, L-L-Ln, P-L-L and O-L-L (P = 16:0, L = 18:2, Ln = 18:3). Double low rape seed oil and olive oil have a high content of oleic acid, which therefore dominates the triacylglycerols in these oils: O-L-Ln, O-L-L, O-O-Ln, O-O-L and O-O-O.
Figure 3. Preparative fractionation of the dominating triacylglycerols in oil from Crambe abyssinica. Fractionation of Crambe abyssinica oil (topmost chromatogram, approx. two mg oil injected) was performed at slightly different conditions compared to the analytical test of fraction 1 – 3 (lower chromatograms) giving slightly different retention times. Labels as described in Figure 2.
Identification of the individual triacylglycerols can be done using preparative SFC as illustrated in Figure 3, where the dominating triacylglycerols in Crambe abyssinica oil were fractionated and each fraction identified by GC-analysis after transformation of the triacylglycerols into the fatty acid methyl esters (FAME). Enough purified substance for GC-FAME were produced in two hours using an analytical column for fractionation (approx. 3 mg of each triacylglycerol).
In addition to analyses of individual triacylglycerols, the SFT are also efficient techniques in connection with determination of other oil constituents important for the oil quality. This is thus the case with methods of analyses for determination of chlorophylls (Buskov et al., 1999b), antioxidants comprising tocopherols (unpublished data), and other vitamins and phenolics (Sørensen et al., 1999).
The Danish Ministry of Food, Agriculture and Fishery, Strukturdirektoratet, and the Commission of European Union (Contract FAIR CT 95-0260 and FAIR CT 98-3778) is gratefully acknowledged for financial support to this work.
1. Anklam, E., Berg, H., Mathiasson, L., Sharman, M. and Ulberth, F., Supercritical Fluid Extraction (SFE) in Food Analysis: a Review. Food Additives and Contaminants, 1998, 15(6), 729-750.
2. Buskov, S., Decker, M., Mortensen, K. and Sørensen, H. Hurtig Kvalitativ Analyse af Vegetabilske- og Animalske Olier med Superkritisk Chromatografi (SFC). Faglig Årsberetning 1998. Pelsdyrerhvervets Forsøgs- & Rådgivningsvirksomhed, 1999a, 131-141. (In danish, with english abstract).
3. Buskov, S., Sørensen, H. and Sørensen, J.C., Fedt- og Oliebestemmelse med Superkritisk Væske Ekstraktion (SFE). Faglig Årsberetning 1996. Pelsdyrerhvervets Forsøgs- & Rådgivningsvirksomhed, 1997a, 159-167. (In danish).
4. Buskov, S., Sørensen, H., Sørensen, J.C. and Sørensen, S., Extraction of Oil from Plant Material with Supercritical Carbon Dioxide. Polish Journal of Food and Nutritional Sciences, 1997b, 6/47(3), 115-124.
5. Buskov, S., Sørensen, H. and Sørensen, S., Separation of Chlorophylls and Their Degradation Products using Packed Column Supercritical Fluid Chromatography (SFC), Journal of High Resolution Chromatography, 1999b, 22(6), (in press).
6. Bøwadt, S and Hawthorne, S.B., Supercritical Fluid Extraction in Environmental Analysis. Journal of Chromatography, 1995, 703, 549-571.
7. Chester, T.L. Chromatography from the Mobile-Phase Perspective. Analytical Chemistry, 1997, 69, 165A-169A.
8. Sørensen, H., Sørensen, S., Bjergegaard, C. and Michaelsen, S. Chromatography and Capillary Electrophoresis in Food Analysis, RCS Food Analysis Monographs, Royal Society of Chemistry, Cambridge, UK, 1999, 470pp.