Metallacyclic Phosphine Complexes and Intramolecular Csp3-H Bond Activation -Synthesis, Characterisation and Catalytic Application

The work presented in this thesis deals with metallacyclic late transition metal complexes, obtained by intramolecular Csp3-H activation, and the application of such compounds in catalysis. The new hemilabile P,O phosphines N-methyl-N-propionyl-2-(diphenylphosphino)benzylamine (DPPAM), o-(diphenylphosphino)benzoic acid ethyl ester (DPPES), o-(diphenylphosphino)benzoic acid methyl ether (DPPET) and ±-1-(2- diphenylphosphino)phenylethyl propionate (±-DPPMES) and the symmetrical diphosphorus donor ligands cis-1,3-bis[(diphenylphosphino)methyl]-cyclohexane (DPCY), cis-1,3-bis[(di-tertbutylphosphino)methyl]cyclohexane (DTBCY) and cis-1,3-bis[(di-isopropylphosphino)methyl]cyclohexane (DIPCY) have been synthesised, allowing the preparation of a range of new metallacyclic rhodium(III), iridium(III) and palladium(II) complexes.

DPPAM, DPPES and DPPET are all capable of intramolecular Csp3-H activation in iridium chemistry, while only the two former can cyclometallate in rhodium chemistry. It is clear that a transition metal from the third row, coordinating electron-releasing spectator ligands and is capable of P,O ring closure is the optimal candidate for insertion into a C-H bond in these ligands. A kinetic investigation of the coordination chemistry of the DPPAM ligand suggests that the oxidative addition of the Csp3-H bond takes place in the plane of the complex via a 14-electron, highly coordinatively unsaturated intermediate. For the ±-DPPMES ligand, which also is set-up for Csp3-Csp3 activation, Csp3-H activation is kinetically preferred, resulting in the first example of transition metal complex capable of breaking a tertiary C-H bond at room temperature. However, if the metal complex coordinating the ±-DPPMES ligand can not achieve a highly coordinatively unsaturated state competing Csp3-O activation occurs, probably as an SN2 reaction.

Of the three diphosphorus donor ligands DPCY, DTBCY and DIPCY, only DTBCY is capable of Csp3-H activation at an electrophilic palladium centre, thus coordinating in a tridentate PCP mer fashion. The other two are coordinated either as a monomeric cis-(P~P)PdII adduct or as a dimeric trans-((P~P)PdII)2 adduct. Density functional modelling of the free ligands suggests that they would have to build up various degrees of strain in order to arrange the candidate Csp3-H bond in the optimal side-on fashion in respect to the metal centre. In the case of the less sterically demanding DPCY and DIPCY ligands, the cyclohexane back bone of the ligand, rather than building up this strain, inverts from diequatorial to diaxial configuration, efficiently suppressing the possibility of C-H activation.

The synthesised complexes [Pd(DPCY)(CO2CF3)2] (15) and [Pd(DTBCY) CO2CF3] (16) are highly efficient as catalyst precursors in the olefination of aryl halides. The arylation of methylacrylate with iodobenzene mediated by complex 15 resulted in a turnover number of 1,176, 000 and a mean turnover frequency of 11, 760 h-1, thus 15 is the most efficient diphosphine-palladium(II) precatalyst reported. The key feature of complex 15 is the labile 8-membered chelate, facilitating de-chelation of the catalyst during the catalytic cycle. The catalytic activity of the tridentate PCsp3P complex 16 was shown to be lower than that of 15. However, 16 is very robust and can be recovered unchanged after catalysis. By comparing with the analogous previously reported PCsp2P PdII precatalyst, it is clear that by changing the central carbon donor atom from sp2 to sp3 hybridisation the catalytic efficiency is increased.