Separation of enantiomers

Enantiomers are traditionally separated by crystallization of diastereoisomeric salts or by diastereoisomeric derivatization followed by crystallization. This technique is somewhat limited if there is no acidic or basic nature or no point of derivatization. In addition, it can be a laborious process to optimize the crystallization process. 

More than 30 years ago, Bill Pirkle, the recognized inventor of modern chiral HPLC, realized that it may be possible to effect a chromatographic separation of enantiomers by use of chiral selectors (or ligands) bound to a silica matrix 3 3,3 4]. There has been a phenomenal amount of development in chiral stationary phases over subsequent years but, a relatively small number of 

Although they are extremely useful analytically, the protein based stationary phases 3-6,371 have found little application in preparative HPLC because they suffer from low loading capacity, due primarily to the low number of active sites. The natural macrocylic molecules Cyclodextrin 3 8,3 91 and antibiotics such as Vancomycin 3 10] have shown some promise. Synthetic chiral crown ethers 311 are particularly useful for the separation of chiral primary amines. 

More recently, polymeric tartaric derivatives 3121 covalently bound to silica are proving to be useful in preparative applications due to enhanced physical and chemical stability. However, the most extensively used media by far are based upon 

There have been several useful reviews written on chiral chromatography over recent years but one of the most comprehensive was written by Levin and Abu-Lafi in 1993 3 l4l 

Enantiomers are molecules that are non superimposable on their mirror images. Two mirror-image forms exist, the chemical and physical properties of which are identical, except for the sense of rotation of the plane of vibration of linear polarized light. They cannot be separated by any of the separation methods described so far. 

Enantiomers can be separated by chromatography, provided that the system used is asymmetric, i.e. chiral. This can be achieved by various means  

In all the above cases, diastereomeric complexes are formed between the sample molecules and the asymmetric species in the chromatographic system and these will migrate with different velocities through the column. As an example, amino 

Practical High-Performance Liquid Chromatography, Fifth edition Veronika R. Meyer 

Enantiomers can be separated by traditional chromatographic methods, provided they have been previously derivatized with a chiral compound to produce diaste-reomers. This method of indirect separation of enantiomers is explained in Section 22.5. 

In aU the above cases, diastereomeric complexes are formed between the sample molecules and the asymmetric species in the chromatographic system and these will migrate with different velocities through the column. As an example, amino acid-copper compounds give diastereomeric complexes of well-known structures— two copper bonding sites may be occupied by the sample molecules (Fig. 21.1). Amino acid-copper compounds may be bonded to silica, as shown in Fig. 21.1, or they may be added to the mobile phase. 

Enantiomers can be separated by traditional chromatographic methods, provided they have been previously derivatized with a chiral compound to 

Practical High-Performance liquid Chromatography, Fourth edition Veronika R. Meyer 2004 John Wiley Sons, Ltd ISBN 0-470-09377-3 (Hardback) 0-470-09378-1 (Paperback) 

There are two general approaches for the separation of enantiomers [1-4,28-32]. The direct method is based on the formation of transient diastereomer association complexes with a chiral selector immobilized in the stationary phase, or added to the mobile phase. The former approach requires the use of special stationary phases (section 10.4) while the later uses conventional stationary phases with special additives included in the mobile phase (section 10.5). When preparative applications are contemplated the use of immobilized chiral selectors is the more common approach. Method selection also depends on the choice of the separation mode. Table 10.2. While chiral stationary phases are the only choice for gas chromatography [16,28,33-38], and are used almost exclusively for supercritical fluid chromatography [39-43] and capillary electrochromatography [44-47], they also dominate the practice of liquid chromatography 

Common methods for separating enantiomers by formation of transient diastereomer complexes 

Gas Chromatography Stationary phase Amino Acid Derivatives Metal Chelates Cyclodextrin Derivatives 

Liquid Chromatography Stationary phase Amino Acid Derivatives Low-Mass Synthetic Selectors Poly(saccharide) Derivatives Cyclodextrin Derivatives Glycopeptides Metal Chelates Proteins Helical Polymers 

Thin-Layer Chromatography Stationary phase Metal Chelates Poly(saccharide) Derivatives 

The classical method for separating enantiomers is to form diastereomeric compounds using a stoichiometric amount of a resolving agent. This method was described in Section 2.1.8. In this section, we discuss methods of resolution based on physical separations, including chromatography with chiral packing materials and capillary electrophoresis. 

One important type of chiral packing material is derivatized polysaccharides, which provide a chiral lattice, but separation is improved by the addition of structural features that enhance selectivity. One group of compounds includes aroyl esters and carbamates, which are called Chiralcels (also spelled Chiracel) two of the most important examples are the 4-methylbenzoyl ester, called Chiralcel OJ, and the 3,5-dimethylphenyl carbamate, called Chiralcel OD. There is a related series of materials derived from amylose rather than cellulose, which have the trade name Chiralpak. 

Related materials can be prepared in which the polysaccharides are linked to a silica support by covalently bound tether groups. For example, silica derivatized by 3-aminopropyl groups can be linked to polysaccharides using diisocyanates. These materials seem to adopt organized structural patterns on the surface, and this factor is believed to contribute to their resolving power. The precise structural basis of the chiral recognition and discrimination of derivatized polysaccharides has not been elucidated, but it appears that in addition to polar interactions, tt-tt stacking is important for aromatic compounds.  

Other types of CSPs, known as brush type, have been constructed synthetically. A chiral structure, usually an amide, is linked to silica by a tether molecule. This approach has the potential for design of the chiral recognition elements. The ability to synthetically manipulate the structures also permits investigation of the role of specific structural elements in chiral selectivity. Several synthetic CSPs were developed by W. H. Pirkle and co-workers at the University of Illinois. An important example is the 3,5-dinitrobenzoyl (3,5-DNB) derivative of 7 -phenylglycine, which is attached to silica by aminopropyl tethers (CSP 2). The 3,5-DNB derivatives of several other amino acids (e.g., CSP 4) and diamines have also been explored.  

Several variations of these CSPs have been developed, such as the phosphonate ester CSP 30 and the tetrahydrophenanthryl amide CSP 33. These compounds are used in pharmaceutical studies. The former CSP is a good resolving agent for the (3-adrenergic blocker class of compounds, such as propanolol, whereas the latter is a good CSP for separation of NSAIDs, such as naproxen and ibuprofen.  

The asymmetric synthesis approach for obtaining optically active thiahelicenes is still in its infancy and so far this approach has led to unsatisfactory optical purity. Therefore, while waiting for innovative approaches related to asymmetric synthesis, researchers involved in this field make extensive use of optical resolution of enantiomers, which, up to the present time, represents the most practical way to obtain optically pure thiahehcenes. 

---

AGP columns have wide appHcation for the direct separation of enantiomers of many different classes of dmgs, amines, acids, and nonprotolytic compounds (18,23). Acidic dmgs resolved include ibuprofen [15687-27-17, C 2H g02, ketoprofen [22071 -15 ] and naproxen [22204-53-17,... 

Separation of enantiomers by physical or chemical methods requires the use of a chiral material, reagent, or catalyst. Both natural materials, such as polysaccharides and proteins, and solids that have been synthetically modified to incorporate chiral structures have been developed for use in separation of enantiomers by HPLC. The use of a chiral stationary phase makes the interactions between the two enantiomers with the adsorbent nonidentical and thus establishes a different rate of elution through the column. The interactions typically include hydrogen bonding, dipolar interactions, and n-n interactions. These attractive interactions may be disturbed by steric repulsions, and frequently the basis of enantioselectivity is a better steric fit for one of the two enantiomers. ... 

The potential for use of chiral natural materials such as cellulose for separation of enantiomers has long been recognized, but development of efficient materials occurred relatively recently. Several acylated derivatives of cellulose are effective chiral stationary phases. Benzoate esters and aryl carbamates are particularly useful. These materials are commercially available on a silica support and imder the trademark Chiralcel. Figure 2.4 shows the resolution of y-phenyl-y-butyrolactone with the use of acetylated cellulose as the adsorbent material. 

Another means of resolution depends on the difference in rates of reaction of two enantiomers with a chiral reagent. The transition-state energies for reaction of each enantiomer with one enantiomer of a chiral reagent will be different. This is because the transition states and intermediates (f -substrate... f -reactant) and (5-substrate... R-reactant) are diastereomeric. Kinetic resolution is the term used to describe the separation of enantiomers based on different reaction rates with an enantiomerically pure reagent. 

Scheme 2.5. Chiral Stationary Phases for HPLC Separation of Enantiomers...

An interesting and practical example of the use of thermodynamic analysis is to explain and predict certain features that arise in the application of chromatography to chiral separations. The separation of enantiomers is achieved by making one or both phases chirally active so that different enantiomers will interact slightly differently with the one or both phases. In practice, it is usual to make the stationary phase comprise one specific isomer so that it offers specific selectivity to one enantiomer of the chiral solute pair. The basis of the selectivity is thought to be spatial, in that one enantiomer can approach the stationary phase closer than the other. If there is no chiral selectivity in the stationary phase, both enantiomers (being chemically identical) will coelute and will provide identical log(Vr ) against 1/T curve. If, however, one... 

It is seen from equation (22) that there will, indeed, be a temperature at which the separation ratio of the two solutes will be independent of the solvent composition. The temperature is determined by the relative values of the standard free enthalpies of the two solutes between each solvent and the stationary phase, together with their standard free entropies. If the separation ratio is very large, there will be a considerable difference between the respective standard enthalpies and entropies of the two solutes. As a consequence, the temperature at which the separation ratio becomes independent of solvent composition may well be outside the practical chromatography range. However, if the solutes are similar in nature and are eluted with relatively small separation ratios (for example in the separation of enantiomers) then the standard enthalpies and entropies will be comparable, and the temperature/solvent-composition independence is likely be in a range that can be experimentally observed. 

Since the first separation of enantiomers by SMB chromatography, described in 1992 [95], the technique has been shown to be a perfect alternative for preparative chiral resolutions [10, 21, 96, 97]. Although the initial investment in the instrumentation is quite high - and often prohibitive for small companies - the savings in solvent consumption and human power, as well as the increase in productivity, result in reduced production costs [21, 94, 98]. Therefore, the technique would be specially suitable when large-scale productions (>100 g) of pure enantiomers are needed. Despite the fact that SMB can produce enantiomers at very high enantiomeric excesses, it is sometimes convenient to couple it with another separation... 

Recently, two examples of the separation of enantiomers using CCC have been published (Fig. 1-2). The complete enantiomeric separation of commercial d,l-kynurenine (2) with bovine serum albumin (BSA) as a chiral selector in an aqueous-aqueous polymer phase system was achieved within 3.5 h [128]. Moreover, the chiral resolution of 100 mg of an estrogen receptor partial agonist (7-DMO, 3) was performed using a sulfated (3-cyclodextrin [129, 130], while previous attempts with unsubstituted cyclodextrin were not successful [124]. The same authors described the partial resolution of a glucose-6-phosphatase inhibitor (4) with a Whelk-0 derivative as chiral selector (5) [129]. 

Liquid-liquid extraction is a basic process already applied as a large-scale method. Usually, it does not require highly sophisticated devices, being very attractive for the preparative-scale separation of enantiomers. In this case, a chiral selector must be added to one of the liquid phases. This principle is common to some of the separation techniques described previously, such as CCC, CPC or supported-liquid membranes. In all of these, partition of the enantiomers of a mixture takes place thanks to their different affinity for the chiral additive in a given system of solvents. 

Among the existing separation techniques, some - due to their intrinsic characteristics - are more adapted than others to processing large amounts of material. Such processes, which already exist at industrial level, can be considered in order to perform an enantioselective separation. This is the case for techniques such as distillation and foam flotation, both of which constitute well-known techniques that can be adapted to the separation of enantiomers. The involvement of a chiral selector can be the clue which changes a nonstereoselective process into an enantioselective one. Clearly, this selector must be adapted to the characteristics and limitations of the process itself. 

Several chiral selectors have been used in the separation of enantiomers by distillation [198]. Among them, the bisalcohol 8 (Fig. 1-6) has permitted obtainment of the ketone (+)-9 with an enantiomeric excess of 95 %. This example shows the feasibility of the process even though, in this particular case, the price of the chiral selector might prohibit scale-up of the separation. 

A compromise among all the properties mentioned herein should be established, depending on the technique used and on the particular application. Preparative separation of enantiomers is still an open subject which requires further investigation in the search of new chiral selectors and techniques well adapted to large scale processes. 

S. G. Allenmark, Separation of enantiomers by protein-based chiral phases in A practical approach to chiral separations by liquid chromatogra.phy, G. Subramanian, VCH, Weinheim (1994) Chapter 7. 

Fig. 8-2. Schematic representation of the separation of enantiomers R and S using a supported chiral macrocyclic ligand host.

Pais L. S., Rodrigues A. E. (1998) Separation of Enantiomers by SMB Chromatography Strategies of Modeling and Proeess Performanee, Fundamentals of Adsorption 6 Proceedings of the Sixth International Conference of Fundamentals of Adsorption, E Meunier (ed.), Elsevier, Paris, p. 371-376. 

Linear case This case is met when the adsorption isotherm is considered linear, which means operation under diluted conditions. Taking into account the saturation capacities of the CSP, this behavior is usually met for concentrations around or below 1 g for separation of enantiomers. 

The separation of enantiomers is a very important topic to the pharmaceutical industry. It is well recognized that the biological activities and bioavailabilities of enantiomers often differ [1]. To further complicate matters, the pharmacokinetic profile of the racemate is often not just the sum of the profiles of the individual enantiomers. In many cases, one enantiomer has the desired pharmacological activity, whereas the other enantiomer may be responsible for undesirable side-effects. What often gets lost however is the fact that, in some cases, one enantiomer may be inert and, in many cases, both enantiomers may have therapeutic value, though not for the same disease state. It is also possible for one enantiomer to mediate the harmful effects of the other enantiomer. For instance, in the case of indacrinone, one enantiomer is a diuretic but causes uric acid retention, whereas the other enantiomer causes uric acid elimination. Thus, administration of a mixture of enantiomers, although not necessarily racemic, may have therapeutic value. 

For the separation of enantiomers, we are interested in 0 -0,. Substituting a = I/ KV, using the expression relating the apparent mobility of an analyte to its binding constant with a chiral additive... 

---