We have extended this work to [5+2] cycloisomerizations of ynamides to give 5,7-fused azacyclic products. This chemistry could be effected with high enantioselectivity using phosphoramidite-rhodium complexes. In both cases we have carried out extensive mechanistic studies to develop an understanding of selectivity and reaction pathway.
Our work on [5+2] cycloisomerization also exploited computation (collaboration with the Paton group) to understand how phosphoramidite ligands can effect highly enantioselective [5+2] cycloisomerizations of ynamides. This identified an unusual (for rhodium) reaction pathway via metallacyclopentene formation, and associated pathway and selectivity determining transition state that enabled us to design a new ligand exhibiting enhanced selectivity and reactivity. This simple para-fluorinated phosphoramidite enables enantioselective reactions to proceed in a matter of minutes, and also permitted the first examples of diastereoselective reactions of single enantiomer substrates that can proceed under catalyst stereocontrol, even when high levels of substrate stereocontrol are present in a mismatched setting. Crucial to this reaction is the binding of the p-fluoroarene arm of the phosphoramidite to the metal centre.
Our interests in palladium-catalyzed processes also led to the serendipitous discovery of an alkynyl epoxide to furan isomerization. Here, we discovered a reaction pathway that benefits from Pd(0) andPd(II) catalysis. The former converts the alkynyl epoxide to an allenylketone intermediate; the latter promotes cyclization of this compound to the furan. In the absence of either pre-catalyst, slower reaction was observed.
We are also interested in developing silicon-based transformations. One example is the formation of cyclic alkenyl silyl ethers (cyclic alkenylsiloxanes) by a conceptually simple Lindlar hydrogenation of the corresponding alkynylsilanes, which removes the need to expensive catalysts in other routes to these interesting structures. We have demonstrated that these make excellent substrates for polyene-forming Hiyama coupling reactions, and are applying this methodology in synthesis.
We have also worked extensively on the Tamao oxidation of arylsilanes to phenols – a variant of this methodology that has to date been missing from the literature. We found that hydrosilanes and alkoxysilanes could be oxidized to the phenol with hydrogen peroxide using catalytic, or even no, fluoride promoter. This again is rather different to many traditional Tamao oxidations, and derived from the higher electrophilicity of the sp2-substituted silicon atom. Finally, we developed a dihalomethylsilane reagent for Takai olefinations of aldehydes, which delivers benzyldimethylvinylsilanes suitable for Hiyama cross-coupling in high yield and stereoselectivity.
Our work on arylsilane oxidation provided a different opportunity to explore mechanism, where analysis of kinetic NMR data for the oxidation revealed a subtle interconversion of a number of species during the oxidation of hydrosilanes by basic hydrogen peroxide, including methoxysilanes, silanols, and (unreactive) disiloxanes. The catalytic role of fluoride was also tested, challenging the dogma that a) stoichiometric fluoride is required for this oxidation; and b) fluoride can be recycled from silane intermediates, despite the strength of the Si–F bond. Here, Swain-Lupton analysis (a refined version of the Hammett method) allowed us to probe the specific contributions of mesomeric and inductive effects on this reaction.