There are exactly five regular polyhedrons that can be made, and as they were first discussed in detail by Plato, they’re sometimes known as the platonic solids. Now, you might not have heard of them under that name, but I’m pretty sure most of them are familiar to chemists. In order of increasing size, the series starts with the tetrahedron, the shape of stereogenic centres at carbon, and the source of asymmetry in life. In fact, Jacobus Henricus van 't Hoff won the very first Nobel prize in chemistry ever back in 1901 for being one of the first to notice this. Although the parent hydrocarbon has yet to be synthesised, a number of tetrahedrane derivatives have been reported. Next comes the cube, the corresponding hydrocarbon of which, cubane, was famously first synthesised by Eaton back in 1964. Third up is the octahedron, more of an inorganic chemist's shape and rather unlikely to ever exist with a carbon skeleton due to the crazy C-C bond angles required. Fourth, the dodecahedron, has actually been better studied by organic chemists than most people realise and, after much competition, Leo Paquette was the first to synthesise the corresponding hydrocarbon in 1983. Last and largest in the series is the icosahedron, which I can’t think of a way to link to chemistry, but we’re certainly unlikely to ever see a carbon based version as all the atoms in the skeleton need to have five bonds.
In this post and the next two I’m going to discuss three syntheses; those of the two platonic solids made to date (cubane and dodecahedrane), and that of the non-regular polyhedron pentaprismane, because it’s also pretty cool. Why do this? Well, when if you consider, say, dodecahedrane in a retrosynthetic sense then unless you lived through the era when these compounds were fashionable targets, have studied them, or are a bit of a genius then it's not obvious where to start so hopefully we can learn a bit of chemistry. I also enjoy a bit of chemical history and some of the methods used were pretty neat as we'll see.
First up is the smallest of the three, cubane. I covered Eaton’s landmark synthesis in the early days of this blog (that is to say, last year) so I thought I’d use this opportunity to write about the extremely short and rather clever synthesis by Pettit and coworkers in 1966. The route began with 2,5-dibromobenzoquinone, and the group planned to functionalise it by doing a Diels-Alder reaction with cyclobutadiene. On the dienophile side of things, this seems perfectly reasonable, as quinones have a long and illustrious history in such a role. The problem is the dienophile; cyclobutadiene is ludicrously unstable, dimerising at 35k and generally being impossible to work with. However, the group had managed to synthesise the somewhat more stable (i.e. isolable and distillable) cyclobutadieneiron tricarbonyl the previous year and had discovered that when this compound was treated with the strong oxidant cerium(iv) ammonium nitrate (CAN) it served as a stable replacement for the crazier parent compound. As best as I can tell, the group actually never reported a yield or any details for this first step, but looking at their other publications aqueous ethanol or aqueous acetone were probably what they used. As for the yield, we can only guess. Presumably it wasn't great, but I’ve heard it said that there are only two yields, 'enough' and 'not enough', and clearly the reaction gave the former. The group then followed up with a swift photochemical [2 + 2] in benzene, closing the cage structure in excellent yield. Next, this was shrunk to the correct size by a double Favorskii rearrangement using aqueous hydroxide. The resulting diacid could then be decarboxylated using Eaton’s three step procedure (acid chloride formation, perester formation and pyrolysis) to give the unnatural product in just 6 steps. Breezy.
Addenda and References
- If you want to get all technical about it, a polyhedron is regular if ‘all its faces are congruent convex regular polygons, none of its faces intersect except at their edges, and the same number of faces meet at each of its vertices’. Thanks Wikipedia! I found this old Stoltz group meeting presentation to be helpful when writing this.
- Original paper is J. Am. Chem. Soc., 1966, 88, 1328 - 132
- Not that you can just buy this. Oh No. The synthesis starts from cyclooctatetraene, uses chlorine gas, temperatures ranging from -30 to +200 ºC, and photolysis.