Cyclometalated gold(iii) complexes: noticeable differences between (N,C) and (P,C) ligands in migratory insertion

Abstract

Gold(III) complexes are garnering increasing interest for opto-electronic, therapeutic and catalytic applications. But so far, very little is known about the factors controlling their reactivity and the very influence of the ancillary ligand. This article reports the first comprehensive study on this topic. The reactivity of a cationic (N,C) gold(III) complex, namely 1A, towards ethylene has been thoroughly studied and compared with that of the related (P,C) complex 1C. A cationic gold(III) complex 5A resulting from double insertion of ethylene was selectively obtained. Complex 5A was found to be remarkably stable. It was trapped with chloride and fully characterized. In marked contrast to that observed with 1C, no β-H elimination or linear-to-branched rearrangement of the alkyl chain occurred with 1A. The energy profile for the reactions of 1A with ethylene has been comprehensively investigated computationally, and the influence of the ancillary ligand has been precisely delineated. Because nitrogen is a weaker donor than carbon (and phosphorus), the (N,C) ligand is very electronically dissymmetric, much more than the (P,C) ligand. This makes the two reactive sites at gold quite different, which noticeably influences the competition between migratory insertion and β-H elimination, and actually changes the outcome of the olefin insertion at gold. This study provides valuable insight into the influence of ancillary ligands on gold(III) reactivity, something critical to further develop Au(III) and Au(I)/Au(III) catalysis.