Three Gurus of Supercritical Fluid Chromatography
Is the ongoing resistance to SFC a simple misunderstanding and the legacy of past disappointments? Here, Isabelle François, Davy Guillarme, and Eric Lesellier highlight the positive aspects of this underused technique and urge sceptics to re-evaluate the potential of modern SFC systems.
Isabelle François, Davy Guillarme, Eric Lesellier |
What are the main advantages offered by SFC?
Isabelle François: SFC can be used in a wide range of industries. In laboratories where organic synthesis is carried out, the medium used for synthesis is very often an apolar solvent, which can be easily injected into an SFC system. The pharmaceutical industry is significantly benefitting in this respect from this technology. Achiral SFC provides an alternative method to RPLC to search for additional impurities present in the sample, whereas enantiomers are separated by chiral SFC. Chiral separations are the niche application for SFC. SFC is even more advantageous when separations are scaled up to preparative mode. Compared with normal-phase liquid chromatography (NPLC), preparative SFC uses less solvent, produces less waste, and requires less time for fraction evaporation. The result is a greener, more cost-effective solution.
In the food industry, SFC can be used for the analysis of nutrients, vitamins, lipids and so on. In nutriceutical applications, it can aid in the search for interesting compounds, certainly when coupled with time-of-flight mass spectrometry.
In the petrochemical industry, applying a pressure gradient allows the density of the mobile phase to be increased, resulting in enhanced solvent strength, which allows the implementation of flame-ionization detection (FID).
Davy Guillarme: SFC becomes a good alternative to NPLC when large amounts of toxic and expensive solvent can be substituted by a mixture of supercritical carbon dioxide and a polar co-solvent, such as methanol. It is superior to RPLC and NPLC for chiral separations; here, SFC has a high success rate, particularly with polysaccharide stationary phases. At the preparative scale, SFC has some advantages over other chromatographic techniques, since the main mobile phase component evaporates easily, leaving only the analyte and a small volume of polar co-solvent.
Today, SFC is mostly used in the pharmaceutical industry for the purification of enantiomers. However, based on its kinetic performance and orthogonal interaction mechanism compared to RPLC, I believe it will become more and more widely used, even for the analysis of achiral substances in pharmaceutical, food, and environmental fields.
Eric Lesellier: It is almost impossible to write an exhaustive list of the benefits of SFC systems. All of the classical separation problems are handled, with the exception of water-soluble compounds. It handles interactions through mobile/stationary phase couplings, it allows coupling of columns for higher theoretical plate number, and it provides the natural concentration of the mobile phase for fraction collection. Of all the benefits, I would highlight isomer separation. Such compounds differ only by subtle geometry, so their separation requires high efficiency and interaction specificity – requirements that are easily met by SFC.
What are the barriers to implementation of SFC?
Isabelle François: Since the introduction of SFC in the 1960s, popularity has seen highs and lows. From a user perspective, I think that resistance is a legacy of past negative experience. Major issues have included a lack of robustness and repeatability (particularly detrimental for implementation into QA/QC environments), and low sensitivity when using UV detection. Instrument builders have addressed these problems in the latest SFC equipment, and will undoubtedly continue to develop the technology, if interest is maintained.
The limited understanding of the fundamental theory of SFC is one of the barriers against uptake. Certainly, when adopting SFC, interaction mechanisms are not always easily understood – a fact that is even more challenging in chiral SFC, where method development is performed by trial-and-error.
Davy Guillarme: At the analytical scale, there are several ‘historical’ shortcomings of SFC. These issues, which really only apply to old-generation devices, include:
- Low UV detection sensitivity
- Low robustness and reliability
- Unacceptable quantitative performance (precision and accuracy) for rigorous method validation.
There is also a general lack of SFC education. This makes people reluctant to adopt it, even when the advantages are obvious. Recently, two important providers of chromatographic systems have entered the SFC market. Given the promotional capabilities of these providers, curiosity and knowledge about SFC will steadily grow.
Eric Lesellier: Working with a compressible fluid raises theoretical questions. Compressibility also occurs in GC, but because of a lack of solute–mobile phase interactions, there are no serious consequences. In HPLC, liquids are generally considered to be uncompressible. However, in SFC, compressibility means that fluid properties can vary – firstly along the column and secondly when changing other parameters. Some of these can be surprising. Due in part to that compressibility, the retention factor can vary when changing the flow rate or column dimensions, which can be disturbing for those used to working with LC. It also promotes different method development practices, isopycnic or not, which lead to straightforward approaches in the first case – far from the reality of SFC practice, but more rigorous in a ‘university sense’ for studying specific theoretical points.Solubility in supercritical fluids – and the methods of measuring it – can also be challenging. Though we understand the non-polar characteristics of carbon dioxide, the effects of 5 or 10 percent modifier on elution can be surprising. Moreover, we are far from modeling (and understanding) the relationships between solubility in organic solvents and carbon dioxide/modifier mixtures.
What are the analytical challenges?
Isabelle François: As identified earlier by Eric, density plays a crucial role in SFC – if it is not managed properly, solvent strength can differ between analyses, leading to shifting retention times. In SFC systems, pressure is maintained by the backpressure regulator (BPR), so excellent design is essential for maximal control. Changes in incoming carbon dioxide temperature also result in a shift in retention times, so repeatable cooling of the carbon dioxide pump heads is also crucial.
As with LC, good batch-to-batch column reproducibility is essential. Furthermore, when scaling up to preparative mode, the selectivity of the analytical sub-2µm or 5 µm particles and the preparative 5, 10 or 20 µm particles must be identical.
Injection volume flexibility was limited in older generations of SFC equipment, which indicates that designing the partial loop injector was a serious challenge – not surprising, if you consider the different states of the solvents used (injection solvent and mobile phase are under liquid and super (or sub) critical conditions, respectively) and the changes in pressure (aspiration of the sample is done in liquid mode, and when placing the loop off-line from the mobile phase flow, expansion occurs and carbon dioxide gas is present in the loop).
From a detection perspective, there are various options for MS hyphenation, but further development is always welcome. In UV detection, reduction of the noise level (linked to the difference in refractive indexes of the mobile phase constituents) needs to be optimized.
Davy Guillarme: In the beginning, SFC was dedicated to the analysis of lipophilic compounds using capillary columns, supercritical carbon dioxide and FID detectors (similar to GC). It has always been a reference standard for analyzing apolar substances, such as lipids and lipo-soluble vitamins, but was less appropriate for polar compounds. Today, this has been corrected to a large extent since SFC is now performed in the presence of a polar co-solvent, using a packed column and UV detection (similar to LC). However, because a polar stationary phase is often used in SFC, polar substances may not easily elute from the column without significant amounts of co-solvent.
Another challenge for SFC is the analysis of substances in biological matrices, which is clearly of prime importance for clinical, forensic, and toxicological applications. So far, only a limited number of such applications have been demonstrated.
Eric Lesellier: The utopia? Separation of proteins and highly polar compounds. However, capillary electrophoresis for protein, and hydrophilic interaction chromatography (HILIC) for very polar compound separation already do these jobs rather well.
From a practical point of view, the real challenge for SFC is method development; firstly, because of the many parameters acting on retention and separation, and secondly, because the stationary phase choice is huge, covering all those used in RPLC, NPLC, and HILIC. For RPLC, it is rather easy: choose a C18 phase, 35°C, 1 ml/min and then perform two or four water/methanol or water/acetonitrile gradients, at two pH values, at which point, dedicated software can calculate the optimal point. But in SFC, choosing the stationary phase (from pure silica to bonded C18), the modifier (from methanol to hexane), the modifier percentage (from 3 to 50 percent), the temperature (from 10 to 60°C), and the backpressure (from 8 to 40 MPa), in a system where flow rate, particle diameter and column dimensions act on fluid density – and therefore, retention – can get very complicated and daunting.
Any other potential pitfalls?
Isabelle François: Instrument manufacturers have made huge investments in countering the main pitfalls. And development has not only been devoted to instrumentation but also to column technology, for example, the introduction of typical SFC stationary phases in sub-2 µm particles.
The future will bring a focus on overall system performance (in terms of individual parts and their integration), and on pushing pressure and flow rate boundaries to fully exploit the features of using a supercritical fluid as the mobile phase (low viscosity and low resistance to mass transfer). Accessibility and flexibility must also be improved, both in terms of hardware and software.
Davy Guillarme: The biggest pitfalls are related to understanding the interaction mechanism in SFC. Firstly, it is important to understand that retention in SFC is mostly driven by H-bond capability rather than hydrophobicity of the molecule. Secondly, the relationship between retention and lipophilicity is poor in SFC and can result in some surprising behavior. Interestingly, SFC’s retention mechanism and the chemical nature of its stationary phases are similar to those employed in HILIC for polar compound analysis, but HILIC cannot be considered a green technology. Thirdly, the dissolution solvent must contain as low an amount of polar solvent (water or methanol) as possible, to maintain reasonable peak shape. Finally, in modern SFC, up to 30-40% MeOH can be added to the mobile phase. Under such conditions, the kinetic advantage of SFC (low viscosity) is no longer valid and performance becomes close to that of LC.
Eric Lesellier: To echo Davy, a misunderstanding of the fundamental theory and retention/separation behavior is probably the biggest potential pitfall. For instance, the effect of temperature increase is dependent on other analytical conditions: when working with 5 percent modifier and a backpressure of 10 MPa, a rise in temperature increases retention, which is typical SFC behavior. On the other hand, at 10 percent modifier and a backpressure of 15 MPa, an increase in temperature decreases retention – typical LC behavior. The transition percentage (from SFC to LC behavior) is dependent on the nature of the modifier.
Confusion also arises from the (large) amount of mobile phase that is adsorbed onto the stationary phase in SFC, as it can induce unexpected retention and separation effects. In fact, SFC (as with NPLC in the past) is subject to many subtle phenomena that can be underestimated when switching from RPLC in which the acidic and hydrophobic properties of water are the main drivers of retention.
Do separation scientists need to change their attitude/focus/scope, if they want to stay in tune with developments in their field?
Isabelle François: To dispel the negative connotations of the past, there needs to be a shift in mentality or a re-education. The advantages of SFC include:
- Much faster separations, with higher efficiency
- Faster column regeneration
- The ability to use smaller particles or longer columns
- Alternative, orthogonal approaches to standard RPLC and GC setups
- The green aspect, which is certainly a consideration at preparative scale
Despite the fact that many chromatographers have recognized the advantages of SFC, it has never really taken off. due to the historical lack of robustness, repeatability, and low sensitivity when using UV detection. Furthermore, the simple introduction of a new analytical instrument can be a little frightening for some…
Davy Guillarme: Of course. It is always important for scientists to take the time to evaluate new analytical approaches. Those who were disappointed by SFC in the past because of its lack of robustness or sensitivity compared with RPLC should give it a second chance using state-of-the-art instrumentation. The potential of SFC is real – it has been employed for a wider range of compounds than RPLC (except therapeutic biomolecules, for which it is clearly unsuitable). In an analytical laboratory, having at ones disposal a ratio of five HPLC systems to one SFC instrument could be of interest.
Eric Lesellier: On my thesis manuscript, 25 years ago, I used this quote:
“When I believe I understand something, I note the day, the hour, the longitude, the latitude. Then I take one step to the side and ask myself again what I have understood.” – Yves Simon (translated).
Isabelle François: SFC has a lot to offer. I would highly recommend analytical laboratories to get in touch with an instrument vendor to discuss applications and take SFC for a test drive. Think not of SFC as a panacea for all your analytical challenges, but rather as an additional tool alongside your RPLC system – it might bring you solutions faster, more efficiently and in a more sustainable way: a win, win, win.
Davy Guillarme: What about mass spectrometry? Just like RPLC, SFC can easily be coupled with MS with either electrospray ionization (ESI) or atmospheric-pressure chemical ionization (APCI) sources. In the past, when SFC was predominantly used for apolar compound analysis, APCI was used because it is a mass flow-dependent device and offers better sensitivity at high mobile phase flow rates. However, ESI has been adapted for a wider range of compounds and often provides better sensitivity than APCI, so providers have developed several SFC interfaces to simplify coupling. Using an appropriate interface, analyte precipitation during carbon dioxide decompression can be avoided, while sensitivity and robustness are drastically improved for routine use of SFC-ESI MS.
Eric Lesellier: The pioneers who invented SFC and demonstrated its potential – Terry Berger, Larry Taylor, Pat Sandra, David Pinckston – should be thanked for giving us the opportunity to work today with such powerful separation science. Now, we have the responsibility to show how this versatile and unified chromatographic method can be used to separate terpene isomers, polymers, enantiomers, sunscreens, anti-oxidants, sugars, lipids, pigments, and so many other compounds.
SFC is what I have been involved in for the last 25 years – and I will fight with the last of my strength to give it the position it deserves!