Synthesis and application of rhodium(I) Fischer carbene complexes

Abstract

In this study, novel rhodium(I) carbene complexes were synthesized and fully characterized via a carbene ligand transfer methodology from Group 6 Fischer carbene complex precursors and subsequent ligand modification. The classic Fischer route was followed towards the isolation of mono- and biscarbene complexes of the form [M(CO)5{C(OR)R'}] [M = Cr, W; R = Me, Et; R' = 2-furyl, 2-thienyl, ferrocenyl] (complexes 1 6, 9 and 10), and [M(CO)5?{C(OR)-R''-C(OR)}M(CO)5] [M = Cr; R = Me, Et; R'' = 2,2'-bithien-5,5'-diyl, 2,5-furadiyl, 1,1'-ferrocendiyl] (complexes 7, 8 and 11), respectively. Analogous aminocarbene complexes [M(CO)5{C(NH2)R'}] [M = Cr, W; R' = 2-thienyl, ferrocenyl) (complexes 12 and 13) were prepared by simple aminolyses of the alkoxycarbene complex precursors and all isolated products were characterized using NMR and FT-IR spectroscopic methods, the results of which were comparable with literature values. Transmetallation techniques were employed in an attempt to transfer the carbene ligands to a rhodium(I) metal center of the dimeric [Rh(cod)Cl]2 precursor to result in novel 2-furyl, 2-thienyl and ferrocenyl Fischer carbene complexes of rhodium(I). Only the ferrocenylcarbene complex 15 [Rh(cod)Cl{C(OEt)Fc) were found to be stable enough to isolate, as the heteroaryl (thienyl, furyl) substituted carbene ligands dissociated in solution, with resultant decomposition dimerization to form the corresponding alkene and starting [Rh(cod)Cl]2 complex, as indicated by NMR spectroscopy. The ferrocenylcarbene complex 15 was then employed as a precursor for the syntheses of all other rhodium(I) carbene complexes via cod ligand substitution and aminolysis reactions to isolate mono- and dicarbonylcarbene complexes 16 23, [Rh(LL)Cl{C(X)Fc}] [LL = cod, (CO)2, (CO, PPh3), (CO, PCy3), (CO, P(OPh3)), (CO, AsPh3); X = OEt,NHnPr], with variable ?-acceptor properties. Full characterization of the novel complexes were achieved by single crystal XRD and spectroscopic methods. From the FT-IR data collected, the donor ability of the electronic environment around the rhodium(I) center was found to correlate with the known electron-donor ability of the coligands in the order PCy3>PPh3,AsPh3>P(OPh)3. This trend was corroborated by cyclic voltammetric methods through which the electron-withdrawing effects of the coligands were studied, and it was confirmed that the cod ligand is the most electrondonating whilst the dicarbonyls were found to be the least donating in the series. In addition, the increased electron donation of the aminocarbene ligands compared to the ethoxycarbene ligands was found to significantly influence the redox potentials of the metal centre in the studied complexes. The isolated rhodium(I) Fischer carbene complexes 15 - 22 were screened as catalyst precursors for the hydroformylation of 1-octene. Good to excellent catalytic activities, with selectivity toward the formation of the linear nonanal, was observed. These results were found to be comparable to results reported for rhodium(I) N-heterocyclic carbene complexes. A mercury-drop test was done to exclude a heterogeneous catalytic mode of action. Finally, the stability of the catalyst precursor 15 (and 17) was probed by an NMR experiment carried out under hydroformylation conditions. The Rh-Ccarbene bond is retained, although the presumed catalytically active species, the rhodium carbene carbonyl hydride complex could not be identified.