Abstract
Investigations of blue phosphorescent organic light emitting diodes (OLEDs) based on [Ir(2-(2,4-difluorophenyl)pyridine)2(picolinate)] (FIrPic) have pointed to the cleavage of the picolinate as a possible reason for device instability. We reproduced the loss of picolinate and acetylacetonate ancillary ligands in solution by the addition of Bronsted or Lewis acids. When hydrochloric acid is added to a solution of a [Ir((CN)-N-boolean AND)(2)((XO)-O-boolean AND)] complex ((CN)-N-boolean AND = 2-phenylpyridine (ppy) or 2-(2,4-difluorophenyl)pyridine (diFppy) and (XO)-O-boolean AND = picolinate (pic) or acetylacetonate (acac)), the cleavage of the ancillary ligand results in the direct formation of the chloro-bridged iridium(III) dimer [{Ir((CN)-N-boolean AND)(2)(mu-Cl)}(2)]. When triflic acid or boron trifluoride are used, a source of chloride (here tetrabutylammonium chloride) is added to obtain the same chloro-bridged iridium(III) dimer. Then, we advantageously used this degradation reaction for the efficient synthesis of tris-heteroleptic cyclometalated iridium(III) complexes [Ir((CN1)-N-boolean AND)((CN2)-N-boolean AND)(L)], a family of cyclometalated complexes otherwise challenging to prepare. We used an iridium(I) complex, [{Ir(COD)(mu-Cl)}(2)], and a stoichiometric amount of two different (CN)-N-boolean AND ligands ((CN1)-N-boolean AND = ppy; (CN2)-N-boolean AND = diFppy) as starting materials for the swift preparation of the chloro-bridged iridium(III) dimers. After reacting the mixture with acetylacetonate and subsequent purification, the tris-heteroleptic complex [Ir(ppy)(diFppy)(acac)] could be isolated with good yield from the crude containing as well the bis-heteroleptic complexes [Ir(ppy)(2)(acac)] and [Ir(diFppy)(2)(acac)]. Reaction of the tris-heteroleptic acac complex with hydrochloric acid gives pure heteroleptic chloro-bridged iridium dimer [{Ir(ppy)(diFppy)(mu-Cl)}(2)], which can be used as starting material for the preparation of a new tris-heteroleptic iridium(III) complex based on these two (CN)-N-boolean AND ligands. Finally, we use DFT/LR-TDDFT to rationalize the impact of the two different (CN)-N-boolean AND ligands on the observed photophysical and electrochemical properties.