Trost’s Ligand in Asymmetric Hydrogenation of Ketones

In 1992, Trost and co-workers firstly reported their trans-1,2-diaminocyclohexane-derived ligand for Pd-catalyzed asymmetric allylic alkylation (AAA). Since then, the Trost’s ligand has been used in several AAA reactions in combination with different nucleophiles, such as primary alcohols, enolates, amines and phenols. Despite the excellent results obtained in these reactions, as far as we know no examples of successful asymmetric hydrogenations involving the Trost’s ligand have been reported. In 2002, Jessop et al. reported the in situ formation of homogenous Ru-complexes for the hydrogenation of carbon dioxide, but no conversion was achieved with Trost’s ligand. Similarly discouraging results were reported also by Bellefon et al. in Rh-catalyzed hydrogenation. The Trost’s Ligand is a PNNP-type ligand having similar geometrical features to the ligands successfully employed by Morris and Gao, with both Ru and Fe, in the asymmetric hydrogenation of ketones. Intrigued by this similarity and considering that the Trost’s ligand is now readily available and relatively cheap, we decided to test its catalytic properties in the asymmetric hydrogenations of ketones. A number of metal sources were combined with the Trost ligand and the in situ formed complexes were tested in the hydrogenation of acetophenone. Unfortunately, Ni, Co and Fe did not show appreciable activity. Nevertheless, interesting results were obtained with Ru. Different Ru sources were screened in order to maximize the enantioselectivity. The best performances were achieved so far with RuCl3 and RuCl2(PPh3)3, which (at 35 °C with 30 bar of H2) gave full conversion and a promising enantioselectivity (up to 73 % e.e.). Different solvents and various bases have been screened and with the optimized conditions a substrate scope has been performed. We are now testing our catalytic system with more challenging substrates like imines and unsaturated ketones.