Theoretical studies on [Ru(bpy)2(N∧N)]2+ [N∧N = hydrazone and azine]:: Ground- and excited-state geometries, electronic structures, absorptions, and phosphorescence mechanisms

The ground- and excited-state geometries, electronic structures, absorptions, and emissions of two ruthenium(II) complexes Ru(bpy)(2)(N boolean AND N) [bpy = 2,2'-bipyridine, N boolean AND N = hydrazone (1) and azine (2)] were investigated theoretically. Their ground and the excited state geometries were fully optimized at the B3LYP/MP2/LANL2DZ and UB3LYP/UMP2/LANL2DZ levels, respectively, and the calculated geometries are consistent with the X-ray results. At the TD-DFT level with the PCM model, the absorptions and phosphorescence properties of 1 and 2 were calculated on the basis of the optimized ground- and excited-state geometries, respectively. The calculated lowest-lying absorptions of 1 (512 nm) and 2 (598 nm) are attributed to a {[d(x2)-(y2)(Ru) + d(xy)(Ru) + pi(N boolean AND N)] -> ([pi*(bpy)]} transition with MLCT/LLCT transition characters and a {[d(z2)(Ru) + d(xy)(Ru)] -> [pi*(N boolean AND N)]} transition with dominant MLCT transition character, respectively. The calculated phosphorescence of 1 (638 nm) and 2 (731 nm) can be described as originating from a (3){[d(x2)-(y2)(Ru) + d(xy)(Ru) + pi(N boolean AND N)] [pi*(bpy)]} excited state with (MLCT)-M-3/(LLCT)-L-3 character and a 3{[d(z2)(Ru) + d(x2)-(y2)(Ru)] [pi*(N boolean AND N]} excited state with dominantly (MLCT)-M-3 character, respectively. The calculated results showed that the modulation of the lowest (MLCT)-M-3 excited state of this kind of Ru complexes can be achieved by changing the N boolean AND N ligand from hydrazone to azine. Moreover, the fact that 2 displays phosphorescence but 1 does not can be interpreted by the different properties of the (MLCT)-M-3 excited state: the (MLCT)-M-3 excited state of 2 is more than 60% occupied, whereas that of I is less than 20% (k(r1) < k(r2)). The lowest-lying excited state of 1 is localized on the bpy ligand, whereas that of 2 lies on the N boolean AND N ligand, and the nonradiative decay pathways of I are easier than those of 2 (k(nr1) > k(nr2)) ((C) Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2008).