Don’t we all have pet peeves, something that just annoys us every time we see or hear it? Probably so. Some of my pet peeves are centred around terminology generally. I am a stickler for using the correct words as descriptors – terminology is not always just ‘a case of semantics’. As Siegfried Farnon tells Tristan Farnon in the original TV series of ‘All Creatures Great and Small’, the English language can express the most subtle of differences in the human condition. So, why can’t chemists get the English language to be succinct when it comes to stereochemical terminology?
But let’s try. 1) Stereoisomers are isomers that have the same bond connectivity but different spatial arrangements of atoms. The definition has nothing to do with chirality. 2) A chiral object is one that cannot be superimposed on its mirror image. It is important to note – every object has a mirror image, i.e. can be held in front of a mirror and you can see it in the mirror. But if the object is chiral, its mirror image cannot be superimposed on the original object. 3) Thus, chiral objects come in pairs (the mirror images), referred to as enantiomers. 4) All other stereoisomers are referred to as diastereomers when being compared to one another. None of these four definitions has anything to do with four different groups being attached to a tetrahedral atom, i.e. a ‘chiral centre’. Yet, I still see these simple definitions getting mixed up routinely.
Consider the structures given above. The first is a chiral object with a classic chiral centre. But the allene is chiral with no chiral centres. The third example is a biphenyl chiral structure (an atropisomer), again with no chiral centres. Lastly, the ferrocene derivative is also chiral without a chiral centre.
So why do so many chemists think that enantiomers and diastereomers need to contain tetrahedral atoms with four different groups attached? I personally believe this is a historical artefact of our educational system. The first organic chemistry textbook (for which I used the 1st edition, I’m that old!) was by Morrison and Boyd (M&B). This book taught about stereochemistry, enantiomers, diastereomers, chirality, etc., all using examples containing a ‘chiral carbon’, i.e. a tetrahedral carbon with four different groups attached. Nothing was wrong in their definitions; it was simply that all the examples contained chiral centres. After this, nearly all subsequent organic textbooks, by numerous different authors, did the same. Thus, lo and behold - the explanations and examples in M&B and later books gave the impression that four different groups on a tetrahedral central atom were needed.
Thus, in the ‘Modern Physical Organic Chemistry’ textbook that I wrote with my friend and colleague, Professor Dennis Dougherty, we went into great length to discuss terminology, recommending chemists avoid using the term chiral centre when discussing stereoisomerism. We introduced alternative terms, such as stereogenic centre to replace chiral centre. A stereocenter (or stereogenic centre) is one in which switching two different groups generates a stereoisomer. For example, the first two structures above have stereocenters. Switch any two groups on the central tetrahedral carbon of the first structure gives the enantiomer and switch the methyl and hydrogen on either end carbon (stereocenters) of the allene gives the enantiomer. The definition of stereocenter has nothing to do with chirality or chiral centres, nor does it imply that enantiomers or diastereomers will result upon switching two different groups.
Additional definitions related to stereochemistry have been introduced into the literature. These terms describe stereochemical units, i.e. various structural features that can generate stereoisomerism. The most obvious is the classic four different groups attached to a tetrahedral atom, which is referred as a ‘point stereocenter’. A C2 axis, as in the atropisomer above, is referred to as an ‘axis of chirality’. Similarly, the C=C=C functionality of an allene leads to compounds that are considered axially chiral. The ferrocene structure above is referred to as being ‘planar chiral’. This arises by virtue of having a group, here the Fe atom, placed on one face of an enantiotopic π-system (also commonly referred to as a prochiral π-system). Complexation of the Fe to the other face produces the enantiomer. The justification for terms such as point, axial, and planar chirality is that such molecules do not need to have stereogenic centres but rather are stereogenic units. An allene, biphenyl, or enantiotopic planar π-systems are groups that can generate stereoisomerism, as do tetrahedral atoms.
Besides the general fascination that chemists (including myself) have with stereochemistry, why discuss all this in a supramolecular chemistry journal? To show the relevance, let’s branch into some supramolecular chirality. A catenane can generate stereoisomers by virtue of having directionality around the two interlocked rings (a mechanical bond), as shown in the cartoon with circles below. Thus, a catenane is another stereochemical unit. The interlocking can generate stereoisomers (note, if there is no directionality around the rings there will be no stereoisomers). A chemical embodiment using R-substituted 1,10-phenanthroline rings (thereby imparting directionality around each ring) is shown below the cartoon circles. To generate the enantiomer, all that is needed is to switch the crossing of the bonds. Using the terminology introduced above, I would call this structure planar chiral (although this can be debated) because different faces of the planar enantiotopic phenanthroline rings are proximal to different R-groups in the enantiomers.
But now, let’s ramp up the complexity! Shown further below are catenanes created by Jochen Niemeyer [Citation1]. The actual structures created were a mixture of cis and trans alkenes, but we’ll focus on the cis isomers here. There is no directionality to the rings, only chirality, and so the interlocking of the catenane rings is not imparting a stereochemical unit. However, the rings have either an R or S BINOL helical stereochemical unit. This is analogous to tartaric acid, which has two point stereochemical units, i.e. stereocenters, and can exist as R,R and S,S enantiomers, as well as an R,S diastereomer which is meso. The R,S diastereomer is not chiral because it has an improper rotation axis, either an S1 (mirror plane) or S2 (centre of inversion) depending upon the conformations one draws. Recall, if any structure has an improper rotation axis, it is not chiral. There are only three stereoisomers, instead of four, because the two stereocenters have the same four groups attached. Tartaric acid is a classic compound with which chemists teach stereochemical terminology, and thus is an excellent analogy to use for the catenanes we are about to consider.
To start, let’s recognise that compounds 1 and 2 are nonsuperimposable mirror images. Note that the mirror images have the ring crossings (circled in structure 1) switched in enantiomer 2. But what happens if we simply switch the ring crossing without switching the handedness of the BINOL units, generating structure 3? If the rings were directional, this would generate a diastereomer. But looking carefully, 1 and 3 are the same structure, just rotated by 180° along the axis shown below structure 1. This is just a rotation; the molecule does not have a C2 symmetry element along this axis. But, if we swap the handedness of one of the BINOL units in 1, we generate the diastereomer 4. Let’s see what happens if the crossings in 4 (again, circled for clarity of presentation) are switched to generate 5. Once again, this really represents is a 180° rotation along the axis shown, so these are the same molecule. Similarly, if we flip the handedness of the BINOL on the right and the left and change the crossings, we return the same structure as can be seen via the 180° rotation that generates structure 6. Analogous reasoning shows that compound 7 is also the same structure as 4, 5, and 6. This structure must be meso, meaning achiral and therefore possess an improper symmetry axis. And in fact, it does – a S2 axis. Importantly, we have three stereoisomers, instead of four, because the two axially chiral units are the same, just as above with tartaric acid.
Wasn’t this fun? Yeah, I think so! Imagine now the trans alkene isomers. This would generate a set of diastereomers to the three the structures given here. But lastly, consider what happens if one ring has a cis alkene and the other ring has a trans alkene. Because cis or trans alkenes still do not impart directionality around the rings, we simply would again generate an entirely new set of diastereomers.
Given our nomenclature clarifications here, and analogies between catenanes and classic structures used to teach stereochemistry (such as tartaric acid), hopefully Siegfried Farnon would be satisfied.
But here’s another pet peeve! Can one molecule be more chiral than another? I’ll give my answer in a future Eric’s Corner.
Disclosure statement
No potential conflict of interest was reported by the author(s).