“Ligand-Enabled Reactivity and Selectivity in a Synthetically Versatile Aryl C–H Olefination” Wang, D.-H.; Engle, K. M.; Shi, B.-F. and Yu, J.-Q. Science 2010, 327, 315-319. DOI: 10.1126/science.1182512
“Highly Convergent Total Synthesis of (+)-Lithospermic Acid via a Late-Stage Intermolecular C–H Olefination” Wang, D.-H. and Yu, J.-Q. J. Am. Chem. Soc. 2011, 133, 5767-5769. DOI: 10.1021/ja2010225
The Mizoroki-Heck reaction is a widely-used method for cross-coupling of the C-X bond of aryl halides/pseudo-halides with the C-H bond of an olefin – the reaction is so popular that Richard Heck won a share of the 2010 Chemistry Nobel for palladium cross-coupling. An equivalent of H-X is produced as a byproduct, which is typically neutralized by the addition of stoichiometric base.
A direct coupling of two C-H bonds is an interesting and potentially green alternative to this reaction, since it could simplify the synthesis of the halide coupling partner (shorter synthesis = less waste) and improve the reaction’s atom economy. In this sort of reaction a directing group is typically needed so the functionalization occurs at only one of the many aryl C-H positions. Additionally, an oxidant is needed in the catalytic cycle because the metal catalyst, typically Pd(II), is reduced when it mediates the C-C coupling (this sort of mechanism has been proposed). Typically silver or copper salts are used in superstoichiometric amounts for this purpose. From a Green Chemistry perspective this is only a marginal improvement over the original Heck reaction – using several equivalents of a transition metal oxidant isn’t a real improvement over generating an equivalent of neutralized acid (like triethylammonium chloride).
Using oxygen as the terminal oxidant, however, is ideal – it is cheap and renewable, and forms water as a byproduct when it is reduced. Researchers in Jin-Quan Yu‘s lab at Scripps have been developing these sorts of reactions for the last few years, and have made some pretty amazing progress (although they are not the only ones – these reviews have most of the major players represented). Early last year, they reported the following C-H olefination reaction of phenylacetic acids catalyzed by palladium acetate, with an atmosphere of oxygen serving as the oxidant. The reaction temperature isn’t too brutal, which is somewhat uncommon for C-H activation chemistry, and I can’t imagine that the reaction is very water sensitive, since an equivalent is generated as a byproduct.
These original reaction conditions were great for symmetrical phenylacetic acids (where positional selectivity doesn’t matter), or unsymmetrical phenylacetic acids where one of the positions ortho to the directing group is more sterically congested than the other. Substrates with two different meta substituents, however, gave a mixture of regioisomers. To overcome this limitation they needed to do some catalyst screening. If they were choosing phosphine or amine ligands, there are plenty of different commercially available ligands to throw at their reaction, but there aren’t so many obvious options for replacing an acetate ligand. The authors made the smart choice to look at different N-protected amino acids – they’re commercially available with a variety of protecting groups, so there are lots of options. They could probably even look at dipeptides too, and had hundreds of different ligands to try. In any case, they found a few amino acid ligands that gave them good regioselectivity in cases where their original conditions didn’t.
They recently applied this chemistry in the key step of their synthesis of (+)-lithospermic acid, a compound with potent anti-HIV activity. Like their simpler substrates, a nearby carboxylic acid controls the regioselectivity of the reaction, giving coupling at one of the six possible aryl C-H sites. They took a pretty big hit on the yield in the final deprotection sequence, but the yield on the C-H olefination was 93%. Pretty cool!
I think this chemistry is amazing, and hopefully these sorts of reactions will find use outside of the labs in which they were developed. Apparently this will still take some work – at a seminar last year, Green Chemistry/Pharma guru Buzz Cue mentioned that while C-H activation methods are desired by process chemists, the temperatures are usually too high and the catalysts are usually too sensitive to be used on an industrial scale. So perhaps methods like this one, which are conducted under 100 degrees and use relatively cheap catalysts, will make this chemistry more attractive.