Nucleophilic Addition to π-Allyl Gold(III) Complexes: Evidence for Direct and Undirect Paths

Abstract

π-Allyl complexes play a prominent role in organometallic chemistry and have attracted considerable attention, in particular the π-allyl Pd­(II) complexes which are key intermediates in the Tsuji–Trost allylic substitution reaction. Despite the huge interest in π-complexes of gold, π-allyl Au­(III) complexes were only authenticated very recently. Herein, we report the reactivity of (P,C)-cyclometalated Au­(III) π-allyl complexes toward β-diketo enolates. Behind an apparently trivial outcome, i.e. the formation of the corresponding allylation products, meticulous NMR studies combined with DFT calculations revealed a complex and rich mechanistic picture. Nucleophilic attack can occur at the central and terminal positions of the π-allyl as well as the metal itself. All paths are observed and are actually competitive, whereas addition to the terminal positions largely prevails for Pd­(II). Auracyclobutanes and π-alkene Au­(I) complexes were authenticated spectroscopically and crystallographically, and Au­(III) σ-allyl complexes were unambiguously characterized by multinuclear NMR spectroscopy. Nucleophilic additions to the central position of the π-allyl and to gold are reversible. Over time, the auracyclobutanes and the Au­(III) σ-allyl complexes evolve into the π-alkene Au­(I) complexes and release the C-allylation products. The relevance of auracyclobutanes in gold-mediated cyclopropanation was demonstrated by inducing C–C coupling with iodine. The molecular orbitals of the π-allyl Au­(III) complexes were analyzed in-depth, and the reaction profiles for the addition of β-diketo enolates were thoroughly studied by DFT. Special attention was devoted to the regioselectivity of the nucleophilic attack, but C–C coupling to give the allylation products was also considered to give a complete picture of the reaction progress.

Au(I)/Au(III) Catalytic Allylation Involving π-Allyl Au(III) Complexes

“…The asymmetric σ+ π-coordination encountered in complex B contrasts with the symmetric η 3 -coordination we found in P,C-cyclometalated Au(III) π-allyl complexes, 12,15 but it is reminiscent of that observed by Tilset et al in related N,Ccyclometalated complexes. 11 Thus, the bonding mode of the πallyl moiety strongly depends on the ancillary ligand and it actually conveys its electronic features.…”

“…Of note, detailed mechanistic investigations of the addition of β-diketo enolates to π-allyl (P,C)Au(III) complexes have shown the feasibility of the nucleophilic attack at the terminal and central positions of π-allyl as well as gold, but the last two paths proved reversible. 15 To support the existence and catalytic relevance of key πallyl Au(III) complexes II, stoichiometric reactions were performed (Scheme 6). First, π-alkene Au(I) complex A 28,39 was synthesized, isolated, and fully characterized (NMR, EA).…”

“…The LUMO + 1 and LUMO + 2 show larger contributions from the central carbon atom (12.6 and 16.5%, respectively) than from the terminal ones (<6%), but they are significantly higher in energy (≥1.3eV), 31 in contrast to the (P,C)Au(π-allyl) + complex. 15 Having established the existence of (P ∧ N)(π-allyl) Au(III) complexes, we sought to demonstrate their ability to react with indoles to give the corresponding C3-allylation products. To this end, an in situ and freshly generated sample of π-allyl Au(III) complex B was treated with N-phenyl indole 1a and K 3 PO 4 (1 equiv each) at −30 °C.…”

“… − In addition, we have shown the P,C-cyclometalated Au­(III) π-allyl complexes to display rich reactivity toward β-keto enolates. A nucleophilic attack can occur at the terminal and central positions of the allyl as well as gold itself to give ultimately the C–C coupling product . These achievements suggest that it may be possible to develop a catalytic transformation involving π-allyl Au­(III) complexes.…”

“…A nucleophilic attack can occur at the terminal and central positions of the allyl as well as gold itself to give ultimately the C−C coupling product. 15 These achievements suggest that it may be possible to develop a catalytic transformation involving π-allyl Au(III) complexes. This was the aim of this work.…”