The next paper in this series is by Prof. Andrew Myers. I followed his work closely during my PhD out of sheer interest; Myers’ research, while focused on total synthesis, is very broad, and he and his coworkers have had many important discoveries and achievements over the years. Off the top of my head, some of the major ones include his work on the tunicamycin and dynemycin classes of antibiotics, which also led to important discoveries regarding a class of cyclizations now known as the Myers-Saito reaction (which is a variant of the Bergman cyclization). Andy Myers also published interesting work on the use of pseudoephedrine as a chiral auxiliary in organic synthesis, and the use of silacyclobutane enolates in aldol chemistry (these are particularly interesting when you notice the amazing rate enhancements due to the effects of the strain in the 4-membered ring). Myers was also the first to publish a variant of the Heck reaction in which you can use benzoic acids as one of the coupling partners; in essence, decarboxylative palladation. This is now an active area of research and decarboxylative coupling reactions are being studied by several research groups, in particular that of Lucas Gooßen. Andrew Myers’ lab also carried out some very nice, complex synthetic work on the development of new classes of tetracycline antibiotics, which ended up getting spun off into a company, Tetraphase Pharmaceuticals. This is an important area of research, as the number of antibiotic-resistant bacteria is growing every day; without effective antibiotics, it would be very difficult to prevent infections, and a simple open wound can end up being fatal. On a side note, Myers was sued by his former PhD student over the royalties stemming from this work, but I’m not sure what happened to the case.
Some of Andy Myers’ early papers are gems of physical organic chemistry, and this one is particularly interesting. It details the synthesis and characterization of 1,6-didehydroannulene, which had been a challenge for physical organic chemists for over 40 years. The parent compound C10H10 (or annulene) should be aromatic as per Huckel’s rule, but it is not due to angular and steric strain. 1,6-didehydroannulene is also a 10π system, but the geometry is planar and so the compound should display aromaticity.
The synthesis of the precursor for 1 is not trivial, but it uses some very interesting reactions. It is mentioned that the final ring closure could not be accomplished with standard methods involving metal acetylides, and so a variant of the Takai olefination had to be employed. The synthesis also involves a Sonogashira coupling and “Wittig reaction of [a] ylide with (trimethylsilyl)propionaldehyde”; this type of Wittig reaction is better known as a Peterson olefination. The late-stage oxidation of an alcohol to an aldehyde is done with the Dess-Martin periodinane, and the final cyclization, as mentioned above, is a variant of the Takai Olefination carried out with chromium doped with a small amount of nickel, conditions reminiscent of the Nozaki-Hiyama-Kishi reaction. What’s interesting is the final statement in the paragraph describing the synthesis: “Due to the extreme sensitivity of 6 [the precursor to 1] toward adventitious decomposition when neat, this product was typically handled in solution in the presence of a free radical inhibitor”. Since this compound (6) was isolated by flash column chromatography, I’m guessing that the column was probably done on a small scale with deuterated solvents (!), since the characterization (coming up) was done by NMR.
1,6-didehydroannulene (1) was generated in an NMR tube in CD2Cl2/THF-d8 solution at -90 ℃, using triflic anhydride and triethylamine. At temperatures above -75 ℃, 1 slowly cyclized to naphthalene, and deuterium incorporation was observed at the indicated carbon-centered radicals.
The NMR spectra (13C (insert) and 1H) are shown below.
The downfield shifts of the signals in both spectra is evidence for an aromatic species, due to a diamagnetic ring current. The 13C spectrum is suspiciously clean, however; if stoichiometric triflic anhydride was used to generate 1, then the 13C peaks for the CF3 group should appear in that range, as reported here. Yet the 13C spectrum in the paper (above) does not have those signals! Odd indeed…
In any case, Myers and Finney were able to measure the kinetic parameters for the cyclization by NMR. In spite of a fairly high activation energy of 16 kcal/mol, the cycloaromatization to naphthalene is fairly quick (25 min at -51 ℃). The quantification of these parameters is important due to this same type of cycloaromatization mechanism being operative in the enediyne class of antibiotics.
In any case, this is a very nice piece of work in pure physical organic chemistry. A lot of work in physical organic chemistry, including what I did for my PhD, concerns the isolation or characterization of unstable species or reaction intermediates, and this falls squarely in that category.