The New Face of SFC: Why Analysts Should Take Another Look

August 1, 2015
Caroline West

LCGC North America

LCGC North America, LCGC North America-08-01-2015, Volume 33, Issue 8
Pages: 522–523

What exactly is SFC? Although you may have heard and read much about it recently, it is still rather vague to many a chromatographer.


I would like to tell you about supercritical fluid chromatography (SFC). Although you may have heard and read much about it recently, it is still rather vague to many a chromatographer. What exactly is SFC? Because the mobile phase used is a mixture of a gas (carbon dioxide) and a liquid (cosolvent, most often methanol), there are two ways to look at SFC:

• We could say that SFC is like gas chromatography (GC), but with a more viscous fluid. The GC approach, with open-tubular columns, was the primary approach practiced in the early development of the method, but is not very much in use nowadays. Comparing SFC to GC in terms of applications also makes little sense except for two particular cases: when one wishes to carry out direct analysis of nonvolatile species without derivatization (taking advantage of a denser fluid with significant eluting strength) or when one wishes to transfer an analysis of volatile species to preparative scale, given that preparative SFC is clearly easier than preparative GC.

• Or we could say that SFC is like high performance liquid chromatography (HPLC), but with a less viscous fluid. Packed-column SFC, carried out with HPLC-like columns, is one area that is really developing today and, in terms of application areas, this is where SFC can express its full potential.

Apart from the columns, current SFC systems also resemble ultrahigh-pressure liquid chromatography (UHPLC) systems in that they use pumps, automatic injectors, column ovens, and detectors based on UHPLC systems. Only a few changes are necessary to adapt to the particular features of compressible fluids: The carbon dioxide pump must be cooled at a low temperature to pump carbon dioxide as a liquid, so that flow rate is accurately measured before mixing with the cosolvent; UV detectors need a pressure-resistant cell (usually up to 400 bar); and a back-pressure regulator is required (after UV detection, and before or after splitting the flow toward evaporating light scattering detection [ELSD] or mass spectrometry [MS] detection) to maintain the fluid under high pressure (typically 100–200 bar). Thus modern SFC systems look pretty much like modern UHPLC systems, except that the tower of modules is taller because of the additional boxes for the special pump and back-pressure regulator.

SFC Is Simply About Using Carbon Dioxide in the Mobile Phase 

Now, what is a supercritical fluid? Surely, you have seen this famous figure before (Figure 1). It is a state diagram for pure carbon dioxide. It shows that there is an equilibrium line between the liquid and the gas, and that, at some point (called the critical point), there is no longer a phase transition between the two, and there is the supercritical fluid.


But actually, SFC chromatographers do not care for this figure anymore. We have travelled a long, long road since 1962. At that time, SFC was about using a supercritical fluid as a mobile phase. But today, more than 50 years later, SFC is simply about using carbon dioxide in the mobile phase, whatever the exact state of it. (I should point out that the very term supercritical fluid chromatography is unfit, most of the time, but that is another debate.) The question of mobile-phase composition is actually evolving fast nowadays because we tend to see application papers where the concentration of carbon dioxide is not the major one anymore, with gradient methods moving from nearly 100% carbon dioxide to 100% cosolvent (1) to achieve the elution of hydrophobic and hydrophilic species in the same run, thereby breaking all boundaries between SFC, enhanced-fluidity liquid chromatography (EFLC), and HPLC.




Why do we like carbon dioxide? Because it has a number of benefits. First, it is not flammable, nor explosive (obviously, because it is used in fire extinguishers). It has a low toxicity, which is good for the chromatographer’s health. It is largely available at a low cost, which is good for the chromatographer’s wallet, especially when one needs to move to preparative scale. And it has a low viscosity, which is good for fast and efficient chromatography.

It was shown a long while ago that introducing some carbon dioxide in the liquid mobile phase is beneficial to efficiency, because the van Deemter curve is displaced to the right, meaning that high efficiency is reached at a higher flow rate than in classical HPLC. This notion was revisited recently, with sub-2-µm particle stationary phases (2), showing that, although the optimum efficiency is not much improved when compared to UHPLC, at least it is obtained at a much faster rate. But what is most interesting is that this high efficiency at high flow rates can be obtained with moderate pressures, thanks to low fluid viscosity. Then you may conclude that the major benefit in using carbon dioxide in your liquid mobile phase is to achieve high efficiency at a high flow rate, without the need for ultrahigh pressure.

The low fluid viscosity has a number of other consequences. For instance, it means that you can use very long columns (up to 2 m), with the same stationary phase (improving efficiency) or different stationary phases to take advantage of complementary selectivities.

Detection No Longer a Problem

At this point, you might be wondering: If SFC is so brilliant, why has it taken such a long time to emerge? Well, let’s face it, there were some problems with the early instrumentation, and not all instruments were equally good. A significant problem was poor UV detection because of high baseline noise. But now that manufacturers have taken it in their hands to produce good instrumentation, this problem is much less significant. Indeed, with modern instrumentation, UV baseline noise is now much reduced, although still not exactly equivalent to that found with modern UHPLC systems.

But is UV baseline noise really a problem? And why not just use MS instead? Now that benchtop single-quadrupole mass spectrometers are available at a reasonable price, and it has been shown that SFC–MS performance is comparable to that of LC–MS, or even better sometimes, I tend to ask: What’s the problem with detection in SFC? Because it is no longer a problem.

Huge Choice of Stationary Phases

Another interesting feature with SFC is that there is a huge choice of stationary phases. Basically, any column that is commercialized for HPLC can be used in SFC. There are reversed-phase-type columns like C18, normal-phase-type columns like silica, or hydrophilic-interaction chromatography (HILIC) phases. Additionally, there are new stationary phases that were designed specifically for SFC, such as the famous 2-ethylpyridine-bonded silica. This is all making the task of choosing a column a very difficult one for an inexperienced chromatographer, but during the past 10 years, in our laboratory we have developed a classification of stationary phases dedicated to SFC use (Figure 2). It is based on linear solvation energy methodology (3,4). It is intended to facilitate column selection, but above all I believe that is has greatly helped in better understanding the fundamentals of retention and selectivity in SFC. It currently comprises about 90 different stationary phases. We update it whenever new interesting columns are marketed, and improve it to better reflect current use of the technique.



And given that SFC uses the same columns as HPLC, all the progress that is made with HPLC packed columns is also beneficial to SFC. Obviously, SFC is moving toward sub-2-µm particles and superficially porous particles. But the column manufacturers who are taking an interest in SFC also improve the column packing process and design new stationary phases especially dedicated to SFC use. There is already a large choice of selectivities available.

Advantages Over HPLC 

Comparing SFC to normal-phase LC, very similar separations can usually be obtained, apart from the fact that they are much faster with SFC. Comparing SFC to reversed-phase UHPLC, we can often obtain excellent and extremely fast separations in both modes. Even small polar molecules that are usually analyzed in the HILIC mode are often well retained and separated with SFC, with the advantage of much faster equilibration times that significantly reduce the time required to develop a new method. The solvent strength of the fluid can be easily manipulated by changing pressure and temperature, but most importantly the mobile-phase composition can be changed easily (with the help of additives); therefore the application area for carbon dioxide-based mobile phases can be extended up to a point that is currently only poorly explored: that of biomolecules.

Current developments in SFC are very exciting and the knowledge that remains to be acquired is still huge. SFC is an iceberg: What we currently know is only a very small part of it, and a large unexplored area remains.


(1)    K. Taguchi, E. Fukusaki, and T. Bamba, J. Chromatogr. A.1362, (2)    A. Grand-Guillaume Perrenoud, J.-L. Veuthey, and D. Guillarme, J. Chromatogr. A.1266, 158-167 (2012). doi:10.1016/j.chroma.2012.10.005
(3)    C. West and E. Lesellier, J. Chromatogr. A.1191, 21–39 (2008).doi:10.1016/j.chroma.2008.02.108
(4)    C. West, S. Khater, M. Khalikova, and E. Lesellier, The Column10,



Caroline West is a professor of analytical chemistry at the University of Orléans (France), in the Institute of Organic and Analytical Chemistry and a member of LCGC’s Editorial Advisory Board. Please send comments about this editorial to