William W. Christie

William W. Christie


Abstract: Some years ago, there was great interest in the application of supercritical fluid chromatography (SFC) to the analysis of lipids. However, the promise of the technique was never matched by achievements and the interest appears now to have waned.

Supercritical Fluid Chromatography - Principles

A substance such as carbon dioxide can exist in solid, liquid and gaseous phases under various combinations of temperature and pressure, and there is a point, the critical temperature, where the liquid and vapour have the same density. If a phase diagram is drawn, there is a region above this point in which no phase change occurs and the substance acts as a supercritical fluid. In this physical state, carbon dioxide is a good solvent for many organic substances; it also has valuable properties as a mobile phase in chromatography as its viscosity is similar to that of a gas, permitting high flow rates and rapid equilibration with the stationary phase.

Commercial supercritical fluid chromatography (SFC) equipment making use of carbon dioxide as a mobile phase has been utilised for lipid separations in several laboratories and many publications have ensued. In addition, a number of review articles have appeared of which the most recent is by King [1]. At this point, I must confess that I do not have access to SFC myself and I have no practical experience of the technique. These notes were therefore compiled simply from my reading of the literature and from occasional discussions with those working with SFC. I do not claim that my opinions are definitive (I would not make such claims in any circumstance). Like all practising scientists, I try to keep up to date with new developments, as sooner or later it may be necessary to debate the merits of new equipment in deciding whether to recommend purchase.

Technology of SFC

In essence, much of the technology of SFC is similar to that for gas chromatography (GC), except for a need for syringe pumps of high quality and facilities for density and pressure programming (and for pressure reduction in the detector). Capillary columns of fused silica coated with cross-linked chemically bonded stationary phases, that have done so much for GC resolution, are equally valuable in SFC applications. Also, much work is being described in which packed columns developed for high-performance liquid chromatography (HPLC) are being used with SFC. A variation on the technique uses carbon dioxide and other solvents just below the critical temperature and is then termed ‘subcritical fluid chromatography’.

However, the real advantage in instrumental terms of SFC with carbon dioxide as the mobile phase is that it can permit a flame ionisation detector to be used, with all the benefits in terms of ease of use, robustness, linearity and sensitivity that we have come to expect of this device in GC applications.

In practice, SFC operates at low to moderate temperatures and seems most suited to the analysis of lipids of high molecular weight such as triacylglycerols. Much of the published work is in this area. The technique is therefore an alternative to high-temperature GC and to HPLC. Some superb separations of triacylglycerols by high temperature GC have now been described, and some resolution by degree of unsaturation as well as by chain-length has been achieved. On the other hand, the technology is near the limits of what may be possible without causing thermal "cracking" of the stationary phase or of the sample and quantification presents major difficulties. HPLC also affords excellent resolution, but often with relatively long elution times while quantitative detection remains a problem.

Evolution of SFC

Obviously there are opportunities for SFC in the separation and analysis of lipids. The first paper on the subject of SFC and lipids appears to be one by Chester [2] in which separations of paraffin wax, free fatty acids, mono-, di- and triacylglycerols, and detergents such as Triton X-100 are described. A capillary column coated with a non-polar stationary phase was utilised at temperatures between 60 and 120șC, certainly a range where no thermal damage to lipids would be anticipated. Pressure programming was used when the lipid components spanned a wide range of molecular weights. In this paper, no quantitative data are given, but superficially the chromatograms of triacylglycerols resemble those obtained two decades earlier by packed column GC, for example, although this is not necessarily a disadvantage.

Since then many more papers have appeared, notably from the laboratories of Sandra in Belgium and Kallio in Finland. With triacylglycerols, it is now evident that good quantification can be expected with flame-ionization detection and that the resolution is essentially on a carbon number basis, i.e. the separation is affected only by the combined chain-lengths of the component fatty acids and not by degree of unsaturation. The rapidity of the technique together with the advantages of quantification have lead to the use of SFC in this application in industry in quality control applications. This information has been gleaned from discussions with scientists in industry, as it is not evident from publications on the topic. The nature of the results achieved does not appear to have encouraged research laboratories to make the required investment, however.

The great range of polarities of natural lipids means that carbon dioxide alone as a mobile phase has severe limitations. To overcome this difficulty, analysts have added organic solvents to carbon dioxide in the mobile phase to modify its properties. Unfortunately, the major advantage of SFC in that it can make use of a flame ionisation detector is then negated. While detectors designed for HPLC can be adapted for the purpose, none has ideal properties where lipids are concerned. As an example, molecular fractions of triacylglycerols have been resolved by SFC on micro-packed silver ion columns with a mobile phase consisting of carbon dioxide containing isopropanol and acetonitrile [3].

SFC has also been applied to the analysis of phospholipids after conversion to diacylglycerol derivatives, Archaebacterial lipids, glycosphingolipids and fatty acids (this list is not exhaustive), but again there are no advantages over alternative techniques that are immediately apparent.

SFC has been successfully interfaced with mass spectrometry permitting separation, identification and quantification of triacylglycerols, for example (c.f. [4]), but it is again obvious that problems remain.


Many years ago when I first wrote on this subject, my conclusion was that SFC had promised much but had yet to deliver. Regretfully, I have not changed my view. It is my impression that most lipid analysts have come to a similar view since I only note one or two publications each year on the subject in the literature survey pages of this site.

Of course the use of supercritical carbon dioxide and other fluids in the extraction of lipids and other natural products as opposed to chromatography is quite another matter.


  1. King, J.W. Supercritical fluid chromatography (SFC) - global perspective and applications in lipid technology. In: Advances in Lipid Methodology - Five, pp. 301-366 (ed. R.O. Adlof, Oily Press, Bridgwater) (2003).
  2. Chester, T.L. Capillary supercritical-fluid chromatography with flame ionisation detection: reduction of detection artifacts and extension of detectable molecular weights. J. Chromatogr. A, 299, 424-431 (1984).
  3. Demirbuker, M. and Blomberg, L.G. Group separation of triacylglycerols on micropacked argentation columns using supercritical media as mobile phases. J. Chromatogr. Sci., 28, 67-72 (1990).
  4. Kallio, H., Vauhkonen, T. and Linko, R.R. Thin-layer silver ion chromatography and supercritical fluid chromatography of Baltic herring (Clupea harrengus membras) triacylglycerols. J. Agric. Food Chem., 39, 1573-1577 (1991).

This article has been updated appreciably from one by the author that first appeared in Lipid Technology, 2, 107-109 (1990) (by kind permission of P.J. Barnes & Associates (The Oily Press Ltd)), who retain the copyright to the original article.

W. Christie

James Hutton Institute (and Mylnefield Lipid Analysis), Invergowrie, Dundee (DD2 5DA), Scotland.

Author Updated: July 20th, 2011 Credits/disclaimer © AOCS