Category Archives: C-H functionalisation

Publication of the week, number 83, 26th June 2015

This week we have a a tandem C–H oxidation/cyclization/rearrangement of isochroman-derived allylic silylethers, promoted by DDQ and InCl3. This appeared in Nature communications (open access) and was submitted by Yong-Quiang Tu and colleagues from the Lanzhou University in China. Recognising that an interesting  strained [3.2.1] bicyclic structure appears in quite a few natural products they decided to see if there were any different ways to make this structure:

bicyc01

They proposed a C-H oxidation which would be followed by a cyclisation and a subsequent rearrangement:

bicyc02

So first the C-H oxidation. DDQ is the obvious choice, benzylic and all that. But in this case it gave a very high yield, 0%. Absolutely no reaction. So not to be defeated our intrepid authors started adding catalytic amounts of “additives”. So FeCl3 (1,0 equiv) and DDQ (1.2 equiv. caused decomposition. SnCl4 also initiated decomposition, but SnBr4 gave 33% yield after 8 hours. Eventually they ended up at indium trichloride, 0.1 equiv. DDQ, 2.0 equiv. and a base, 2,6-di-tbutyl pyridine, 5 equivalents. This brew gave the bicyclic compound in 81% yield after 12 hours.

The substrate scope is somewhat limited, as described, with R1 being alkyl, vinyl, acetylene. Methoxy, and bromine were described for R2. So a bit limited. So now thew proceeded to utilise this chemistry to synthesise (-)-przewalskine E and (-)-brussonol, both of which are closely related:

bicyc03

So working this backwards leads to the following intermediate and ultimately starting material:

bicyc04

This turned out to be a 15(16) step asymmetric synthesis, in, well I didn’t work it out but it’s around 1% overall. However there are a couple of notable steps. One is the chirality introduction which was achieved with a Sharpless asymmetric epoxidation/epoxide ring opening:

bicyc05

This, according to the authors is “a rare example of successful tetra-substituted olefin epoxidation“. The conditions; (-)-DET, Ti(i-PrO)4, t-BuO2H, 25 °C to 50 °C, CH2Cl2, 90% yield, 83% ee. Some more steps along the line we come across a Wittig reaction followed by a Simmons-Smith and hydrogenation to give the gem-dimethyl substituent.

The crucial C-H activation sequence proceeded in 82% yield. From the authors, “we nearly always obtained the desired product with a benzoxa[3.2.1]octane framework as a single diastereomer. In terms of the mechanism of this transformation, there are two possible pathways whereby intermediate may react with a benzylic oxocarbenium cation under oxidative conditions. One is a tandem cyclization/semipinacol rearrangement and the other is a tandem [3,3]-Cope rearrangement/aldol reaction. Our results suggest that the latter is more likely. “

So here we have a nicely written paper, lots of chemistry and an absolutely enormous supplemental information file. There is some 200 odd pages of this, NMR, experimentals, etc. Really lots of information. The way it should be. AND the chemistry was done on a scale between 0.1 and 2.5g. Note no comment from the authors, the work speaks for itself.

2,897 total views, no views today