Transition metal complexes with high fluorous phase affinities for use and reuse in catalysis

A rhodium catalyst generated from ClRh[P((CH2)2(CF 2)n-1CF3)3]3 (n = 6, 8) was used to catalyze the hydrosilylation of ketones and enones. The catalyst was recovered and reused for a minimum of three cycles. Catalyst recovery was demonstrated through several representative protocols. One, the rhodium complex was dissolved in perfluoromethylcyclohexane, and cyclohexenone, PhMe2SiH, and toluene were added. The resulting biphasic mixture was heated at 60°C for 10 h and then cooled to room temperature. The top toluene layer containing the product was extracted. The rhodium catalyst remained in the bottom perfluoromethylcyclohexane (fluorous) layer and was recycled. Two, the above reaction was repeated with hexanes instead of toluene. The resulting monophase was heated to 60°C. The reaction was completed in 1 h cooling to -30°C gave a biphasic mixture. The product was extracted and the catalyst recycled as above.;In the third recovery protocol, ClRh[P((CH2)2(CF 2)5CF3)3]3 was combined with PhMe2SiH, cyclohexanone and dibutylether. The mixture was heated to 55°C to give the product C6H11OSiPhMe 2. Upon cooling, the catalyst rest states precipitated, giving colorless product-containing solutions (98%). The precipitated catalysts can be reused at least three times. When this sequence is conducted in the presence of Teflon RTM tape (thickness/width 0.0075/12 mm), the catalysts precipitate onto the tape, and can be thermally re-released for further reaction cycles. The catalyst precursor ClRh[P((CH2)2(CF2) 5CF3)3]3 can also be initially coated on the tape, which allows small quantities to be delivered by length as opposed to mass. Qualitative rate measurements show an induction period for the first cycle, excellent retention of activity for the second and third cycles, and significant loss of activity in the fourth (ascribed to deactivation). 19F NMR data show total ponytail leaching corresponding to 11.4% of the phosphine ligands over three cycles. Experiments in the presence of mercury show that colloidal or metallic rhodium are not responsible for catalysis. These data constitute (to our knowledge) the first catalyst recycling protocols involving common laboratory TeflonRTM tape, and there is every reason to expect that they can be extended to other fluorous catalysts and/or forms of TeflonRTM.