B.R.S.M. Help! I'm trapped in a molecule factory!


Quality, not Quantity

Here’s another poser for you: what do the two molecules below have in common? Hint: in contrast to last week’s mechanistic question, this is more to do with their history.


The answer is... a few things. Firstly, they were both proposed incorrectly as structures for two very well known chemicals. On the left is the originally assigned structure for Meldrum’s acid, a useful reagent for acylations, generation of ketenes and Knoevenagel reactions. On the right is Dewar benzene, one of a number of different structures considered by James Dewar (of Dewar flask fame) for benzene.[1] When I look at these compounds, I’m shocked by the speed with which chemistry has moved forwards over the past hundred years. For example, although Kekulé proposed the correct structure of benzene sometime in the 1860’s it wasn’t actually confirmed until 1929 when Catherine Lonsdale, first female Fellow of the Royal Society, solved its structure using X-ray diffraction. That’s right: the structure of benzene wasn’t confirmed until four years after Sir Robert Robinson proposed the correct structure of morphine. And it only took another 30 years for the entire 64 kDa structure of haemoglobin to be solved! In fact, the size and complexity of molecules readily analysed with the kind of equipment found in any university chemistry department has leapt forward since 50 years ago. But we digress.

Another thing that these two molecules have in common is that they were both named after Scottish chemists. A letter in the RSC magazine Chemistry World by Hamish Kidd a few years stated:

“…2008 is the anniversary of the discovery of the only chemical to be named after a Scotsman - Meldrum's acid. Andrew Norman Meldrum prepared the acid by the condensation of malonic acid with acetone in acetic anhydride containing a small amount of sulfuric acid…”

But clearly this isn’t true. If you need a third example, a number of people on Twitter reminded me of the MacMillan imidazolidinone organocatalysts, which are sold under that name by Aldrich.


Finally, both compounds actually have rather interesting physical properties that their simple chemical structures belie. Meldrum’s acid possesses ludicrously acidic CH bonds – the main reason for its initially misassigned structure – and Dewar benzene is actually far more stable than first expected, with a half life of a couple of days at room temperature.

Although the existence of his eponymous molecule was considered by Dewar in 1865, it wasn’t actually synthesised until almost a century later by E. E. van Tamelen, who was surprised by its stability. It’s now known that the compound itself is metastable, i.e. it’s thermodynamically unstable (as it’d love to rearrange into benzene), but can exist due to a large kinetic barrier for change.[2] You see, thermal electrocyclic opening of four membered rings normally occurs via a conrotatory process. However, this means that (for cis Dewar benzene) the product is going to be cis,cis,trans-cyclohexatriene - a molecule even higher in energy than the starting material - and the reaction has a pretty large enthalpic barrier. Perhaps unsurprisingly, the π bond rotation then required to convert this intermediate into benzene is actually believed to be fairly facile (i.e. 1-2 kcal/mol). Alternatively, a “forbidden” disrotatory process would allow direct conversion of Dewar benzene to benzene without any high energy intermediate, but as you'd expect this also carries a large (symmetry imposed) enthalpic barrier.[3] The question is, which barrier is bigger?

Although for a long time the ring opening of this unusual compound was given as an example of a symmetry forbidden process that nevertheless occurred, recent computational efforts have in fact suggested that a symmetry allowed conrotatory opening to cis,cis,trans-cyclohexatriene followed by double bond rotation might actually represent a lower pathway.[4] It's actually pretty hard to tell which occurs. Regardless, whichever pathway the conversion to benzene takes it meets a large barrier, allowing the molecule to exist in its own little well on the reaction coordinate.


From J. Am. Chem. Soc., 1996, 118, 7381-7385

The anomalous acidity of Meldrum’s acid (aqueous pKa 4.97) compared to that of malonic acid (pKa ~13) was also a long standing mystery, and was only explained convincingly in 2004 using high level computational chemistry (see J. Org. Chem., 2004, 69, 4306 if you want the details). The currently accepted explanation (as I understand it) for this phenomenon is the fact that the the conformation of the molecule allows for extremely efficient overlap between one of the CH bonds and the C=O antibonding orbital, which has the effect of significantly weakening it, even in the ground state.


So, in conclusion, there may not be many chemicals named after Scottish chemists, but at least they're all quite interesting!



1. It’s important to note that Dewar actually agreed with Kekulé’s proposed structure; this and other isomers were found in his notebooks and mentioned in his lectures, but he never put them forward as the correct ones. Unhelpfully, a German chemist H. Wickelhaus popularised this structure and the community came to associate it with Dewar, and hence many now believe that he misassigned Benzene thus. It's quite interesting to note how the valence isomers of benzene, many of which were incorrect proposed structures from the 19th century, have turned out to be popular synthetic targets in their own right. In fact, all of the ones reasonably expected to exist have been made:


2. A more classic example of a metastable substance is diamond. At room temperature and pressure, diamond is actually thermodynamically higher in energy than its (much cheaper) allotrope graphite, but can still exist for long periods of time due to the fact that there’s no easy pathway for the two to interconvert. In order for diamond to be converted into graphite it’s necessary to break up and reorder the entire lattice, which requires a lot of energy and presents a huge kinetic barrier. Diamonds are essentially forever.

3. Take another look at the Woodward-Hoffmann rules and draw (or look up) a state correlation diagram to see why. The Woodward-Hoffmann rules are one Woodward Wednesday we’re not seeing any time soon. I’m thinking maybe reserpine or steroids in a couple of weeks.

4. I am not a computational chemist, but that’s my understanding of this paper.



Comments (2) Trackbacks (1)
  1. I knew McMillan when he was a postdoc. Since then he died his red hair dark, flattened his ‘r’s and the intonation and cut back on the self-deprecating jokes. For all purposes he could be from Midwest now

  2. Hey. I am still waiting on some stuff to arrive in Germany, but when it does I’ll post a bit more on the acidity of MA. I would go so far as to say it is still up for debate. The JOC paper is on reactive orbital theory, but other ideas that have not been disproved are still out there. Got to be my favourite theoretical debate!

Leave a comment