Ruthenium(II) polypyridine compounds often luminesce, with ligand-dependant lifetimes. This is the third in a series of papers devoted to finding orbital-based luminescence indices (LI) for predicting luminescence lifetimes. The third attempt (LI3) was based upon a frontier molecular orbital (FMO) like estimate of the height of the barrier to conversion from an initial phosophorescent triplet metal-ligand charge transfer (3MLCT) state to a nonluminescent triplet metal-centered (3MC) state which decays nonradiatively. A linear correlation was found between LI3 and extracted empirical average 3MLCT →3MC transition state (TS) barrier heights (Eaves) which were not believed to be quantitative but which were believed to capture the trends in the true barrier heights correctly. As it is known that Eave is a large underestimate of the true 3MLCT →3MC TS barrier height in the case of the trisbipyridine ruthenium(II) cation { [Ru(bpy)3]2+ }, but accurate TS barrier heights are difficult to obtain experimentally, it was judged useful to verify the ideas used to derive the LI3 index in the present work by calculating the energetics of the gas-phase 3MLCT →3MC reaction for a series of ruthenium(II) tris bipyridine complexes using the same density functional and basis sets used in calculating LI3. Specifically, four closely-related bipyridine complexes { [Ru(N∧N)3]2+ with N∧N = bpy (6), 4,4’-dm-bpy (70), 4,4’-dph-bpy (73), and 4,4’-DTB-bpy (74) } were used for these calculations. In the process, we examined the gas-phase trans dissociation mechanism in greater detail than has been previously done and uncovered a two part mechanism. In the first part, the electron is transferred to a single ligand rather than symmetrically to all three ligands. It is the two Ru-N bonds to this ligand which are equally elongated in the transition state. The intrinsic reaction coordinate then continues down a ridge in hyperspace and bifurcates into one of two symmetry-equivalent 3MC structures with elongated trans bonds. Interestingly, no significant difference is found for the TS barriers for the four complexes treated here. Instead, LI3 is linearly correlated with the energy difference ΔE=E(3MLCT)−E(3MC), different from what was originally intended but still consistent with the FMO arguments underlying the derivation of LI3.
Read full abstract