Determination of enantiomeric purity has become important in recent years due to increasing restrictions on the composition of pharmaceuticals whose efficacy is dependant upon a chiral moiety. The behaviour of the enantiomers of a chiral drug may show striking differences in terms of biological activity, potency, toxicity, transport mechanisms and routes of metabolism. Therefore analysts in the pharmaceutical industry now face the challenge of assessing the stereoisomeric composition of potential drugs during both the discovery and development stages. The characterisation of the drug for enantiomeric purity is required after synthesis and also during in vivo and in vitro studies for determining the fate of a chiral molecule and its metabolites. There are several methods commonly employed in the pharmaceutical industry for the determination of enantiomeric purity including circular dichroism and specific rotation, separation techniques such as liquid and gas chromatography and more recently, capillary electrophoresis.

None of these techniques are amenable to real-time analysis and while the separation methods can offer good analytical performance in terms of precision and accuracy, they are expensive techniques in terms of reagent consumption and cost of instrumentation, and in addition generate significant waste. In contrast, sensor-based analysis provides real-time analysis, low cost of instrumentation, amenability to automation, no expensive reagents and virtually no waste.

Our ultimate aim is to develop a sensor which can distinguish between enantiomers of the same molecule. This is a difficult task, as enantiomers usually exhibit very similar chemical properties, and in contrast to chromatography, sensor-based analysis involves determination of the target species in the sample matrix without separation. The power of chromatography arises from the dynamic interaction of analytes between the mobile phase and stationary phase, which, over the length of a typical column, can greatly magnify any small differences in relative affinity for the two phases which may exist between enantiomers of the same compound. In contrast, however, a sensor can only rely on a single partitioning or exchange event to generate the selectivity required. Hence development of successful materials for chiral sensors is extremely challenging in view of the difficulty in predicting the host-guest properties of new receptors from a theoretical point of view, as the overall observed selectivity is the sum of many subtle interactions which may, or may not be interdependent. 

We have investigated whether we could use a calixarene macrocycle as a template for producing materials which in principle satisfy the above criteria. Calixarenes have attracted appreciable interest due to their ability to act as host molecules for various substrates, a property which makes them attractive candidates for use in sensors.

Figure 1: Representations of the calixarene tetramer macrocycle showing the upper and lower rim.

Bearing this in mind, we synthesised three receptors with substituent groups which were both chiral and fluorescent attached to the lower rim of a calixarene macrocycle (Figure 1). Compound 1 is based on the attachment of the chiral fluorescent moiety, S-di-2-naphthylprolinol, to the four positions at the lower rim of the calix[4]arene molecule whereas compounds 2 and 3 involve the attachment of (R/S)-1-(9-anthryl)-2,2,2-trifluoroethanol to three of the lower rim positions in the tetramer, and four of the lower rim positions in a hexamer, respectively.

The efficiency of the quenching of a fluorescent species by a quenching species follows the Stern-Volmer relationship, if the fluorophore and quencher concentrations are in the appropriate range

where I₀ is the fluorescence intensity in the absence of the quencher; I is the fluorescence intensity in the presence of quencher at a concentration [Q], and K_sv is the Stern-Volmer constant which is a measure of quenching efficiency.

If a system obeys the Stern-Volmer equation, a plot of I₀/I versus [Q] will give a straight line with a slope of K_sv and an y-axis intercept of 1. K_sv is therefore a measure of quenching efficiency and a large value for this parameter equates to a sensitive response. Ideally therefore, the aim is to obtain a material which displays a large magnitude for K_sv with one enantiomer and zero for the other (no quenching effect) and subsequently to employ this material in an optical sensor.

Three fluorescent chiral calixarenes have been synthesised. The Stern-Volmer plots obtained demonstrate that the fluorescent tetra-S-di-2-naphthylprolinol calix[4]arene (1) is the most enantiomerically selective. The (R)-PEA was seen to have greater efficiency as a quencher, proving that the calixarene was more sensitive to the presence of (R)- than the (S)-PEA. Enantiomeric selectivity was also observed for (L) and (D) norephedrine which has a similar structural conformation to PEA. It is possible to determine the enantiomeric composition of samples of (R)- and (S)-PEA and (L)- and (D)-norephedrine to within a few % if the total concentration of both enantiomers is known. While the mechanism underlying the chiral recognition displayed by 1 is not fully understood at present, it would appear to arise from a combination of hydrogen bonding recognition sites located within a 3-d chirally restricted space defined by tight clustering of the four S-di-2-naphthylprolinol groups around the calixarene lower rim.

Figure 10: A possible explanation for the origin of the enantiomeric selectivity of 1 may lie in hydrogen bonding between the PEA amine group and the hydroxyl group of one of the S-di-2-naphthylprolinol groups of 1 which would be favoured by parallel alignment of the aryl groups of the PEA molecule and the dinaphthyl fluorophore. Simultaneous hydrogen bonding from the amine to the nearby carbonyl oxygen is also possible.