C-C bond activation enabled by dyotropic rearrangement of Pd(IV) species

Abstract

The weak carbon-metal bond combined with the kinetic inertness of the carbon-carbon bond renders metal-catalysed C-C bond activation to be highly challenging. Most of the reported C-C bond activation methodologies involve strain-releasing cleavage of small rings to compensate for unfavourable kinetic and thermodynamic penalties associated with C-C bond cleavage. Here we report that the 1,2-positional interchange of vicinal C-C and C-Pd(IV) bonds (dyotropic rearrangement) can be realized in a stereospecific manner under mild conditions, giving access to quaternary carbon-palladium bonds. An enantioselective synthesis of medicinally relevant fluorinated cyclopentanes, featuring this rearrangement as a key step, has been developed. We anticipate that implementing a Pd-based dyotropic rearrangement in reaction design could provide a new tool in the development of Pd-catalysed transformations.

Figure 1. Dyotropic rearrangement involving metals.

a, Dyotropic rearrangement of 1-hetero-1-alkenyl higher order organocuprates. The reaction is stoichiometric in Cu using C–Cu bond as a stationary scaffold. b, Schematic presentation of proposed 1,2-aryl(alkyl)/Pd dyotropic rearrangement. The Pd species is one of the migrating group involving C–C bond as a stationary phase. The reaction could be catalytic in metal if catalytically active [Pd] species could be regenerated at the end of the catalytic cycle. c, β-Carbon elimination/syn-carbopalladation sequence. The β-carbon elimination happens nevertheless seldomly within a non-strained acyclic framework. d, Reaction design. The proposed reaction sequence involves an enantioselective carbopalladation (step 1), Pd-walking (step 2), oxidation of Pd(II) to Pd(IV) species (step 3), the 1,2-aryl(alkyl)/Pd(IV) positional interchange (step 4) and finally, the C–F bond forming reductive elimination process (step 5). Generation of α-carbonyl Pd(IV) species D is thought to be a driving force for the programmed dyotropic rearrangement. A quaternary C–Pd bond (D) is generated after the dyotropic rearrangement. Ligand was omitted for the sake of clarity.

Figure 2. Side products and mechanistic implication.

a, The isolated 1,3-fluoroarylation products 6s, 6y, 6z. The formation of these products not only materialize the presence of intermediate C, but also indicate that the C–F bond forming reductive elimination from C is a competitive process to the dyotropic rearrangement. b, Possible reaction pathways. Path a, reductive elimination of complex C leading to the side product 6s; Path b, ring-opening of the phenonium ion F by fluoride could also produce the side product 6s, but not the rearranged product 2s. The phenonium ion F could potentially be generated via intramolecular nucleophilic displacement of the Pd(IV) species, which is a good nucleofuge, by the neighboring 4-methoxyphenyl group; Path c, dyotropic rearrangement pathway leading to the product 2s. Release of steric interactions and the formation of a thermodynamically more stable α-carbonyl Pd(IV) species could be the driving force of the rearrangement. Ligand was omitted for the sake of clarity.

Figure 3. Structure characterization of key reaction intermediates.

a, Synthesis, isolation and structural characterization of the Pd(II) intermediates (±)-7a and (±)-7b, precursors of the dyotropic rearrangement. b, Detection of the Pd(IV) intermediates (±)-8a and/or (±)-9a. C–F bond forming reductive elimination of (±)-8a affords (±)-6l, while that of (±)-9a produce (±)-2l. c, Control experiments on the oxidation of Pd-complex 7a with SET oxidants. Only β-hydride elimination product (±)-5l was isolated indicating that Pd(III) species might not be involved in the conversion of 7a to 2l. SET = Single electron transfer. Ligand was omitted for the sake of clarity.