Unconventional pincer ligands and their transition metal complexes

Abstract

This thesis describes investigations into unconventional pincer systems incorporating non-classical central donor groups. Direct reaction of 1,3-bis(diphenylphosphinomethyl)-2,3-dihydro-1H-1,3,2-benzodiazaborole (dppBH) with metal complexes and subsequent ligand manipulations have provided novel ruthenium and, for the first time, osmium boryl pincer complexes. The boryl complexes thus prepared have been observed to undergo various subsequent reactions involving the chloride and triphenylphosphine co-ligands to give new complexes. Further reactions of dppBH with complexes dichlorotris(triphenylphosphine)ruthenium(II) and -osmium(II) resulted in the first examples of sigma-borane pincer complexes, which represent intercepted intermediates in the B-H activation process, affording rare structural data for an osmium sigma-borane complex. The B-H bond of the ruthenium complex could be readily cleaved upon ligand manipulation, such as substitution of triphenylphosphine with the pi-acidic carbonyl and isocyanide ligands, which gave the disubstituted boryl complexes via spontaneous loss of HCl. Attempts to isolate a 1,8-diaminonaphthalene-based borane analogue of dppBH proved unsuccessful, though the compound 1,3 bis(diphenylphosphinomethyl)-2,3-dihydroperimidine (PhDHP) was obtained as a side-product in one case. Further investigations into this compound resulted in the development of a convenient one-pot synthesis, which could be extended to the cyclohexyl analogue 1,3 bis(dicyclohexylphosphinomethyl)-2,3-dihydroperimidine (CyDHP). These compounds were observed to react with various platinum group metal complexes to give novel N-heterocyclic carbene (NHC) pincer complexes via double geminal C-H activation of the central methylene group, providing the first examples of perimidine-based NHC inclusion as the central equatorial group of a pincer system. Reactions of dihydroperimidines with group 8 complexes dichlorotris(triphenylphosphine)ruthenium(II) and -osmium(II) have thus provided the novel pincer NHC complexes. However, reactions of PhDHP with dichlorotris(triphenylphosphine)ruthenium and CyDHP with less electron-rich ruthenium precursors instead gave asymmetric PNP coordinated complexes, in which C-H activation had not occurred, though this could be induced thermally in one case. Reactions of PhDHP and CyDHP with chlorotris(triphenylphosphine)rhodium(I) gave 16-electron chloro rhodium(I) NHC pincer complexes, and their reactivity was investigated via co-ligand manipulations and a preliminary catalytic study. The latter revealed that, while this complex was not particularly efficient for most of the reactions investigated, in some cases simple modifications of the co-ligands could substantially improve catalytic activity. Reactions of the pro-ligands with iridium precursors instead favoured the formation of coordinatively saturated complexes. The reaction of CyDHP with chlorobis(cyclooctene)iridium(I) dimer resulted in a dihydrido chloro iridium(III) NHC pincer complex, while reaction with chlorocarbonylbis(triphenylphosphine)iridium(I) instead afforded a sigma perimidinyl hydrido complex, resulting from oxidative addition of only one C-H bond to the metal centre. This was subsequently shown to readily form an NHC complex upon hydride abstraction. It became apparent that, upon reaction of the dihydroperimidine pro-ligands with metal complexes, carbene formation occurs more readily for electron-rich systems, otherwise resulting in either single or no C-H activation. These observations have provided some insight into the mechanism of NHC formation via chelate-assisted C-H activation of these precursors.