Chirality, derived from the Greek word kheir, "hand", is a symmetry property of an object or molecule. Whether something is "chiral" is determined by a single condition: that it is not super-imposable on its mirror image. There are many chiral objects in everyday life like gloves, shoes, screws or spiral staircases. Twiners or "snail shells" can be found in macroscopic nature as well, their chiral orientation being a consequence of the chirality of their molecular building blocks.
Often, we come across a situation that is not racemic (equal mixture of images and mirror images), but dominated by a specific chiral orientation. For example, snail shells are predominantly dextral (the coiling of the shell is right-handed). The chemistry of life selects for specific versions of chiral compounds (single enantiomers), such as choosing to use only left-handed forms of amino acids and right-handed forms of sugars.
The ability of biological molecules to discriminate between these enantiomers is vital for living systems; biology uses chirality to create function. This is a good reason for chemists, especially in the pharmaceutical industry, to aim for enantiopure products. There are thousands of chiral forms of biologically active molecules in the body responsible for controlling health. Drug molecules with opposite chiral orientation can have different effects in the body.
However, for the majority of known chiral compounds the two enantiomeric forms either both show the same biological activity (usually to differing degrees), or one form is active and the other inactive.
A tragic example of the different effects of the two enantiomeric forms is the drug thalidomide, prescribed to pregnant women in the 1960s. The left-handed molecule had the desired sedative effect, but the right-handed molecule induced fetal malformation.
A positive outcome of this tragic occurrence was the development and establishment of strict industry regulations governing chiral purity for active pharmaceutical ingredients (APIs). Currently, over 50% of the top-selling drugs are in a single enantiomeric form in order to minimise any potential side effects.
There are several possibilities to produce an enantiopure product. A chiral synthesis certainly is the most elegant way, but it has its drawbacks.
It is often a complex solution, and additional synthesis steps and suitable catalysts are needed. Scientists are working hard to find better and cheaper chiral auxiliaries, and good progress is being made.
The importance of the topic is emphasised by the 2001 Nobel Prize in Chemistry, which was given to leading scientists in the field of "asymmetric catalysis" for their work on chirally catalysed reactions leading to enantio-pure products.
It is problematic to find suitable catalysts that can be used not only on the lab scale but also on an industrial scale. Metal and enzyme catalysts are being used, but purification, costs, toxicity and suitability for scale-up are a great concern. Currently, scientists are working on new concepts using small organic molecules as efficient catalysts in the preparation of single enantiomers.
An alternative to chiral synthesis is the separation of a racemic mixture from an achiral synthesis by crystallisation or chromatography.
The advantage of chiral crystallisation is that it may be repeated on a large scale fairly easily, although this approach is not applicable in all cases. The disadvantage is that the crystallisation process itself can be lengthy. Depending on the molecule, the mother liquor needs to be enriched by the desired enantiomer or seeded by enantiopure crystals of this target enantiomer. Often, the yield is not as high as that produced with a chromatographic resolution.
In recent years, substantial progress has been made in the field of production scale chromatography, in the development of new chiral stationary phases (CSPs) and in the development and implementation of technology. For example, simulated moving bed (SMB) and supercritical fluid chromatography (SFC).
In chromatography, the two enantiomers are washed through a packed column by an eluent and are retained differently due to different affinity to the column packing, in this case the CSP. At the end of the column, the enantiomers can be collected in different fractions.
Most CSPs consist of chiral agents that are derivatised and immobilised on the surface of a support (silica gel mostly). They serve as the in situ chiral discriminators during the chromatographic process.
The demands on a CSP are high: the CSP should show a suitable enantio-selectivity for the target compound to be resolved, as well as a high loading capacity, to make the separation feasible at larger scale. Additionally, it should be robust, chemically inert and thermally stable.
The quality of the CSPs is being continuously improved and there is an increasing choice of better CSPs available on the market every year, also in bulk quantities. But still CSPs are cost-intensive and prices are several times higher than for non-chiral phases.
An achiral synthesis followed by chiral chromatography is a good alternative to a chiral synthesis. Chroma-tography in general is a flexible and mild separation technique, and is especially well suited to sensitive products.
CSPs offer several advantages: the solutes are unmodified and the separations can be rapid. The development time is substantially shorter than for a chiral synthesis, but CSPs are expensive and the solvent consumption is rather high, depending on the possible sample loading. The separation system (eluent mixture and CSP) has to be specifically selected for the particular separation at hand.
For preparative scale separations, the available techniques are the batch elution chromatography by liquid chromatography (LC) and by super-critical fluid chromatography (SFC) as well as the continuous simulated moving bed (SMB) process, which is based on LC.
Batch elution chromatography is discontinuous and uses repetitive injections of a feed mixture. This can be done using various column set-ups (usually a single column) or injection sequences (e.g. staggered injections), which can be automated. Traditionally, these processes are run using LC, but SFC is on the rise.
For SFC the same process set-up is used, but the liquid is replaced by a supercritical fluid. (Above a certain temperature and pressure, a fluid is called supercritical and has properties between a liquid and a gas.)
The density of the supercritical fluid corresponds to the solvating power. This property can be used to tune the separation. As the pressure in the system is increased, the supercritical fluid density increases, and its solvating power increases. Therefore, as the density of the supercritical fluid mobile phase is increased, components retained in the column can be made to elute. This is similar to using a solvent gradient in High Performance Liquid Chromatography (HPLC).
SFC is approximately three to five times faster than HPLC and its high resolution makes it ideal for the purification and separation of difficult samples. SFC plants available on the market are getting better and bigger. The main eluent, CO2, is sustainable and cheap and the product can be easily removed from the solvent. The productivity is high and the SFC technology is increasingly competitive with the continuous simulated moving bed (SMB) process.
The simulated moving bed is a continuous chromatographic binary separation process. From the technical point of view, it is more challenging than a batch chromatography set-up, but this continuous process has a very high potential for scale-up. The solvent consumption is several times lower and productivity is much higher compared with a batch LC process. However, the SMB process competes with the fast and "green" SFC batch process. The drawback of SMB is the technical complexity and its limitation on a binary separation.
The technology has its origin in the petrochemical industry, where today several million tonnes of product are produced each year using mainly zeolites as stationary phases. It is also extensively used in the sugar industry for the production of several mono- and oligosaccharides.1
In the past 15 years, the SMB process has been adopted for fine chemical and pharmaceutical applications as suitable chiral stationary phases became available on the market. Today more than 10 full-scale production units and numerous pilot-scale systems are operated on a routine basis in almost every major pharmaceutical company. Many top drugs, including Zoloft, Keppra, Cipralex/Lexapro and Xyzal exist only as a result of this new technology.
Within Carbogen Amcis, two SMB plants (see fig. 1) are used for binary, mainly chiral separations on a scale of 100-4000g racemate/day. Column sets of eight columns with 4.8cm I.D. and 2.5cm I.D. are available and are packed to a standard length of 10cm. The inlet and outlet flow rates can be up to 250ml/min, whereas the flow rate in the column loop can be up to 500ml/min.
The pressure in the plant depends on the chosen flow rates and the viscosity of the eluent mixture. The optimal operating pressure is below 50 bar.
The SMB principle is based on a circular counter-current flow of a liquid phase (eluent) and a solid phase (CSP) (see fig. 2). The racemate to be separated is fed continuously at the feed point to the unit. The two enantiomers have a different affinity to the applied solid phase (CSP). In the counter-current movement, the less retained enantiomer is carried with the liquid to the raffinate outlet, where it can be collected. The more retained enantiomer is carried with the solid to the extract outlet.
The velocities of the solid and the liquid phase are the process parameters that have to be adjusted for each separation task to make the counter-current principle work. In an SMB, the CSP volume is used very efficiently and, compared with the batch LC process, a higher loading of product can be realised.
Technically, it is problematic to move physically the solid without damaging the packing properties of the CSP. A real counter-current process, described above, therefore seems impossible to realise. Nevertheless, a chromatographic set-up that technically approximates the desired counter-current effect can be implemented. It is called simulated moving bed because the movement of the solid bed (packed HPLC column bed) is not a physical but a simulated movement.
An SMB plant consists of usually six to eight packed HPLC columns (fixed bed), which connect to form a closed loop. The plant has two inlet (feed and fresh eluent) and two outlet ports (extract and raffinate). For example, figure 2 shows a possible set-up for a SMB plant as it is used within Carbogen Amcis, whereas other set-ups are also common in industry.
The flow direction of the circular liquid flow is defined by a pump, which is integrated in the column loop. To simulate a circular solid movement in the opposite direction, all four inlet and outlet ports are periodically and simultaneously shifted by one column position in the direction of the fluid flow. Each time the ports are shifted one position further, the packing volume of one HPLC column is moved with respect to the inlet and outlet positions in the opposite direction. In this way, a true counter-current is approximated.