Synthesis and reactivity of molybdenum, iron, and palladium complexes with mono-, bi- and tridentate phosphorus and nitrogen donor ligands

Abstract

The first part of this work describes the synthesis of two classes of ligands. First, the concept of aminophosphine-based tridentate PNP pincer ligands was extended to include tridentate PNP ligands based onN-heterocyclic diamines such as 1,3,5-triazine and bidentate PN ligands based on 2-aminopyridine. In most cases these coumpounds react readily with electrophilic chlorophosphines and -phosphites in the presence of base. The respective chlorophosphines are commercially available, the chlorophosphites are easily derived from PCl3 and the corresponding diols. Employing this methodology even chiral derivatives can be obtained when e.g. dimethyl tartrate is used as diol.<br />The second class covers ligands prepared via a condensation reaction of 2,5-thiophenedicarboxaldehyde and 2-thiophenecarboxaldehyde with various amines. The second part of this work comprises the synthesis, characterisation and reactivity of molybdenum, iron, and palladium complexes of these ligands. In the case of molybdenum, octahedral tricarbonyl PNP complexes were obtained under mild conditions by reaction of Mo(CO)3(CH3CN)3 (prepared in situ by refluxing Mo(CO)6 in acetonitrile) with the respective pincer ligands.<br />These zerovalent complexes undergo oxidative addition with halogens leading to heptacoordinated monocationic complexes. Tetracarbonyl complexes with the bidentate PN ligands were obtained by heating equimolar amounts of Mo(CO)6 and the ligand in toluene. These complexes show enhanced reactivity towards iodine-which results in the cleavage of all carbonyl ligands-and undergo oxidative addition with allyl bromide. Attempts to obtain vinylidene compounds by reacting the complexes with phenylacetylene failed as well as the reaction with methyl iodide or crotyl bromide.<br />Upon reaction of the PNP pincer ligands with [Fe(H2O)6](BF4)2 dicationic octahedral tris(acetonitrile) complexes were obtained. Up to three of the acetonitrile ligands are labile and can be replaced by chelating nitrogen donor ligands or carbon monoxide. The catalytic activity of the complexes was tested in the coupling of aromatic aldehydeswith ethyl diazoacetate. It was found that the formation of 3-hydroxyacrylates is favoured in comparison to the beta-ketoester. In addition, a plausible mechanism for the catalytic cycle was proposed supported by mechanistic investigations based on DFT/B3LYP calculations. Starting from Pd(cod)Cl2 and PNP pincer ligands tetracoordinated, cationic palladium complexes with different counterions could be obtained. The 2,5-thiophenedicarboxaldehyde- based ligands proved to be too rigid to coordinate in a tridentate fashion. When reacted with Pd(cod)Cl2 two 2-thiophenecarboxaldehyde-based ligands were observed to coordinate only via the imine or amine nitrogen atom, forming neutral square planar complexes. The catalytic activity of these compounds in the Suzuki-Miyaura coupling of aryl bromides with phenyl boronic acid was investigated and was found to be comparable to related alpha- and beta-diimine systems. Qualitatively, it was found that complexes with imine ligands featuring sterically demanding substituents such as mesityl are clearly better catalysts.