Chiral Separations

Column selection and method development for chiral separations requires experience and a careful approach, and supercritical fluid chromatography (SFC) has become a a viable option. Participants in this Technology Forum are Martin Volmer of Agilent Technologies, Zachary S. Breitbach and Daniel W. Armstrong of the University of Texas at Arlington, and Dave DePasquale of Waters Corporation.

What guides your selection of stationary phase and mobile phase when beginning a new chiral application or separation?

Volmer: There is no general rule that guides the selection of chiral stationary phase (CSP). It is virtually impossible to deduct the optimum stationary phase for a given analyte or analyte mix. Key is therefore the usage of an intelligent column screening approach. This is best done with a professional method development system. A fully automated column screening study, including screening of 4–8 columns, can be set up with minimum effort by using a method development wizard. This is commercially available as predefined system in liquid chromatography (LC) as well as in SFC.

Breitbach and Armstrong: With more than 100 commercially available CSPs on the market today, choosing the right column and the proper mobile phase for new method development requires judicious selection by experienced chromatographers. First, and foremost, the most important tool in successfully developing a new chiral separation is an understanding of the molecular make-up of available chiral selectors, what types of chiral compounds are routinely separated by each phase, the general retention mechanism of the phases, and what screening mobile phases are recommended by the manufacturer.

When selecting stationary phases for screening, it is often helpful to search previously reported enantiomeric separations of analytes that are structurally similar to your compound. Many of the main column manufacturers have databases of separations available on their websites. Ultimately, the functionality of the analyte dictates the choice of phase. Neutral analytes may be first screened on macrocyclic glycopeptide, cellulosic, amylosic, or aromatic derivatized cyclofructan phases. For ionizable compounds, the macrocyclic glycopeptide columns may be a good initial screen. For highly aromatic analytes, the aromatic derivatized cyclodextrins, aromatic derivatized cyclofructans, or π-complex based selectors are good options. For primary amine–containing analytes, the aliphatic derivatized cyclofructan columns are most successful. If the compound is dominated by hydrophobic moieties, cyclodextrin-based phases may be the best bet.

Often, a chiral separation can be obtained on more than one stationary phase. In these cases, you have the option of choosing the mobile phase that best fits your separation and detection needs. If hyphenation with mass spectrometry is required, it is best to choose a polar organic or reversed-phase separation. The best columns for this are macrocyclic glycopeptide, cyclodextrin, and cyclofructan based. If a preparative or SFC separation is needed, normal-phase separations are desirable. Columns commonly used for these applications include the cellulosic, amylosic, and cyclofructan-based chiral selectors.

Lastly, when developing a new chiral separation method, the robustness, efficiency, and reproducibly of the column and the separation must be considered.

DePasquale: It's a well known fact that it's hard to correlate structure and phases (stationary and mobile). As far as the mobile phase, methanol with some amine additive would be a good starting point.

It seems like selecting a chiral phase by trial and error is an inefficient, labor-intensive process. Do you see any new chiral stationary phases that may be more "universal"?

Volmer: In my opinion, intelligent, fully automated column switching devices, easy-to-set-up studies with the aid of a software wizard, and fast chromatographic separation techniques such as SFC or ultrahigh-pressure liquid chromatography (UHPLC) are more productive than selecting randomly columns by trial and error. However, there are a few stationary phases that have a higher probability to lead to a successful separation for chiral compounds. These columns should be used in a tier 1 approach. If that fails the next tier of stationary phases can be used.

Breitbach and Armstrong: The discovery of a universal chiral selector for all analytes is not likely to come anytime soon. However, it is possible to develop CSPs that are nearly “universal” for a particular set of analytes.

Perhaps the greatest advancement in the field of chiral separations in the past decade or so is the advent of cyclofructan-based chiral selectors. One of the cyclofructan chiral selectors, an aliphatic derivative, can be considered universal for primary amine–containing enantiomers. Recent studies have shown that this single phase can give a 93% success rate for the enantiomeric separation of primary amine–containing racemates. This novel chiral stationary phase routinely performs these separations in the polar organic, the normal phase, and in SFC solvent systems without the need for acidic aqueous mobile phases (as is needed for other chiral crown ethers). Furthermore, this CSP separates nearly any primary amine compound, such as those that contain heterocycles, have the stereogenic centers near or far from the primary amine, as well as aliphatic compounds that lack other functional groups. For these reasons, this novel phase can be considered a nearly “universal” chiral selector for primary amines.

DePasquale: We have noticed some new commercial CSPs bring different selectivity, but we have not noticed any CSP being overwhelmingly universal. It is also worth mentioning that some new CSP claims to target a specific group of compounds such as primary amine. The general applicability remains to be seen.

How is the adoption of SFC influencing the way that new chiral methods are being developed?

Volmer: SFC has great potential for the separation of chiral compounds that has not been fully exploited until now. The major advantages of SFC for chiral separations are unique selectivity, speed, and the fact that less toxic solvent and waste is generated. There has been a dramatic improvement in respect to sensitivity and robustness with the launch of a new analytical SFC system last year that allows now to quantify enantiomeric impurities at 0.05% of the main component. This in combination with 600 bar capability and mass-selective and evaporative light scattering detection will broaden the application range of SFC in the chiral world and beyond.

Breitbach and Armstrong: Certainly the greatest impact of developing more SFC chiral separations is the need for novel chiral selectors that behave optimally in the normal phase. Though some normal-phase separations may not be as reproducible as reversed-phase separations, their use has become increasingly significant in recent years as researchers aim more toward preparative and SFC-based chiral purification methods. For this reason, column producers are focusing on producing novel stationary phases that not only work well under SFC conditions, but also have high capacities. Columns that work well for SFC include some derivatized liner polysaccharides and aromatic-derivatized cyclofructans.

DePasquale: SFC offers faster separation than normal-phase HPLC, and there are commercially available parallel systems. Rather than pursuing a universal CSP, now users can afford to screen multiple columns and stationary phases within a reasonable timeframe.