Probing the reactivity of ruthenium indenyl complexes in P-C bond forming reactions




Derrah, Eric James

Journal Title

Journal ISSN

Volume Title



Asymmetric hydrophosphination, the addition of a P-H bond across a C-C double bond, is an attractive potential route to chiral phosphines, which have important applications in many other types of asymmetric catalysis. However, a highly active and stereoselective catalyst for this reaction has yet to be identified. The ruthenium indenyl complex [RuCl(η5-indenyl)(PPh3)2] (1) was investigated as a potential catalyst for hydrophosphination through an exploration of the steps involved in this process: substrate coordination, P-H bond activation, and P-C bond formation. Substitution of triphenylphosphine ligands at the metal centre of 1 by alkyl- and aryl-substituted secondary phosphines (PR2H: R = Cy (a), Pri (b), Et (c), Ph (d) or Tolp (e)) gave predominantly the monosubstituted secondary phosphine complexes [RuCl(η5-indenyl)(PR2H)(PPh3)] (3a-e). Hydride ([RuH(η5-indenyl)(PR2H)(PPh3)] (6a,d)) and cationic nitrile ([Ru(η5-indenyl)(NCR')(PR2H)(PPh3)][PF6] (7a,d: R' = CH=CH2; 8a-b,d: R = CH3)) derivatives of 3 were prepared and in all cases the potentially reactive P-H bond of the secondary phosphine ligand did not interfere with the chemical transformation. Deprotonation of the P-H bond of the bulky dialkylphosphine-substituted chloro complexes 3a-b with KOBut gave five-coordinate, planar terminal phosphido complexes [Ru(η5-indenyl)(PR2)(PPh3)] (10a-b) that contain a unique Ru-PR2 π-bond. The analogous phosphido complexes 10d-e, containing less bulky aryl substituents at phosphorus, were found to be unstable at room temperature and were observed only by low temperature 31P{1H} NMR spectroscopy. Phosphido complexes 10a-b were found to be highly P-basic, capable of deprotonating the C-H bond of acetonitrile (pKa = 24) to give the metallated acetonitrile complex [Ru(CH2CN)(η5-indenyl)(PR2H)(PPh3)] (9a-b), and to be very P-nucleophilic, reacting with iodomethane (MeI) to give a new P-C bond in [RuI(η5-indenyl)(PCy2Me)(PPh3)] (17a). As might be expected, the addition of donor ligands to low-coordinate 10a-b was found to disrupt the Ru-PR2 π-bond to give six-coordinate terminal phosphido complexes [Ru(η5-indenyl)(L)(PR2H)(PPh3)], with pyramidal, instead of planar, geometry at phosphorus. These additions are irreversible in the case of CO (19a-b) or PCy2H (21a), while pyridine (23a-b) or NCPh (24a-b) adducts were shown by 31P{1H} NMR spectroscopy to be in equilibrium with 10a-b and the uncoordinated ligand. The addition of known substrates for transition metal-mediated hydrophosphination, phenylacetylene and acrylonitrile, to 10a-b resulted in a [2+2] cycloaddition of the unsaturated C-C bond at the Ru-PR2 π-bond to give metallacyclic complexes [Ru(η5-indenyl)(κ2-PhC=CHPR2)(PPh3)] (27a-b) and [Ru(η5-indenyl)(κ2-NCCHCH2PR2)(PPh3)] (32a-b) respectively. Surprisingly the addition of simple non-activated olefins (i.e. ethylene, 1-hexene, or norbornene), which were not previously known to be active substrates for this reaction, also gave [2+2] cycloaddition products. These cycloaddition reactions were found to be 100% regioselective, and are also stereoselective in the case of substituted alkenes (>96%). Experimental evidence suggests that these P-C bond forming reactions proceed via a concerted [2+2] cycloaddition pathway.



Asymmetric, Complexes, Substrates, Phosphine