The 77 K emission spectra of 21 [Ru(L) 4bpy] ( m+ ) complexes for which the Ru/bpy metal-to-ligand-charge-transfer ( (3)MLCT) excited-state energies vary from 12 500 to 18 500 cm (-1) have vibronic contributions to their bandshapes that implicate excited-state distortions in low frequency ( lf; hnu lf < 1000 cm (-1)), largely metal-ligand vibrational modes which most likely result from configurational mixing between the (3)MLCT and a higher energy metal centered ( (3)LF) excited state. The amplitudes of the lf vibronic contributions are often comparable to, or sometimes greater than those of medium frequency ( mf; hnu mf > 1000 cm (-1)), largely bipyridine (bpy) vibrational modes, and for the [Ru(bpy) 3] (2+) and [Ru(NH 3) 4bpy] (2+) complexes they are consistent with previously reported resonance-Raman (rR) parameters. However, far smaller lf vibronic amplitudes in the rR parameters have been reported for [Os(bpy) 3 ] (2+), and this leads to a group frequency approach for interpreting the 77 K emission bandshapes of [Ru(L) 4bpy] ( m+ ) complexes with the vibronic contributions from mf vibrational modes referenced to the [Os(bpy) 3] (2+) rR parameters (OB3 model) and the envelope of lf vibronic components represented by a "progression" in an "equivalent" single vibrational mode ( lf1 model). The lf1 model is referenced to rR parameters reported for [Ru(NH 3) 4bpy] (2+). The observation of lf vibronic components indicates that the MLCT excited-state potential energy surfaces of Ru-bpy complexes are distorted by LF/MLCT excited-state/excited-state configurational mixing, but the emission spectra only probe the region near the (3)MLCT potential energy minimum, and the mixing can lead to larger distortions elsewhere with potential photochemical implications: (a) such distortions may labilize the (3)MLCT excited state; and (b) the lf vibrational modes may contribute to a temperature dependent pathway for nonradiative relaxation.

The lowest energy metal to ligand charge transfer (MLCT) absorption bands found in ambient solutions of [Ru(NH(3))(4)(Y-py)(2)](2+) and [Ru(L)(2)(bpy)(2)](+) complexes (Y-py a pyridine ligand and (L)(n) a substituted acetonylacetonate, halide, am(m)ine, etc.) consist of two partly resolved absorption envelopes, MLCT(lo) and MLCT(hi). The lower energy absorption envelope, MLCT(lo), in these spectra has the larger amplitude for the bis-(Y-py) complexes, but the smaller amplitude for the bis-bpy the complexes. Time-dependent density functional theory (TD-DFT) approaches have been used to model 14 bis-bpy, three bis-(Y-py), and three mono-bpy complexes. The modeling indicates that the lowest unoccupied molecular orbital (LUMO) of each bis-(Y-py) complex corresponds to the antisymmetric combination of individual Y-py acceptor orbitals and that the transition involving the highest occupied molecular orbital (HOMO) and LUMO (HOMO-->LUMO) is the dominant contribution to MLCT(lo) in this class of complexes. The LUMO of each bis-bpy complex that contains a C(2) symmetry axis also corresponds largely to the antisymmetric combination of individual ligand acceptor orbitals, while the LUMOs are more complex when there is no C(2) axis; furthermore, the energy difference between the HOMO-->LUMO and HOMO-->LUMO+1 transitions is too small (<1000 cm(-1)) to resolve in the spectra of the bis-bpy complexes in ambient solutions. Relatively weak MLCT(lo) absorption contributions are found for all of the [Ru(L)(2)(bpy)(2)](m+) complexes examined, but they are experimentally best defined in the spectra of the (L)(2) = X-acac complexes. TD-DFT modeling of the HOMO-->LUMO transition of [Ru(L)(4)bpy](m+) complexes indicates that it is too weak to be detected and occurs at significantly lower energy (about 3000-5000 cm(-1)) than the observed MLCT absorptions. Since the chemical properties of MLCT excited states are generally correlated with the HOMO and/or LUMO properties of the complexes, such very weak HOMO-->LUMO transitions can complicate the use of spectroscopic information in their assessment. As an example, it is observed that the correlation lines between the absorption energy maxima and the differences in ground state oxidation and reduction potentials (DeltaE(1/2)) have much smaller slopes for the bis-bpy than the mono-bpy complexes. However, the observed MLCT(lo) and the calculated HOMO-->LUMO transitions of bis-bpy complexes correlate very similarly with DeltaE(1/2) and this indicates that it is the low energy and small amplitude component of the lowest energy MLCT absorption band that is most appropriately correlated with excited state chemistry, not the absorption maximum as is often assumed.

The effects of ligand perdeuteration on the metal-to-ligand charge-transfer (MLCT) excited-state emission properties at 77 K are described for several [Ru(L)(4)bpy](2+) complexes in which the emission process is nominally [uIII,bpy-] --> [RuII,bpy]. The perdeuteration of the 2,2'-bipyridine (bpy) ligand is found to increase the zero-point energy differences between the ground states and MLCT excited states by amounts that vary from 0 +/- 10 to 70 +/- 10 cm(-1) depending on the ligands L. This indicates that there are some vibrational modes with smaller force constants in the excited states than in the ground states for most of these complexes. These blue shifts increase approximately as the energy difference between the excited and ground states decreases, but they are otherwise not strongly correlated with the number of bipyridine ligands in the complex. Careful comparisons of the [Ru(L)(4)(d(8)-bpy)](2+) and [Ru(L)(4)(h(8)-bpy](2+) emission spectra are used to resolve the very weak vibronic contributions of the C-H stretching modes as the composite contributions of the corresponding vibrational reorganizational energies. The largest of these, 25 +/- 10 cm(-1), is found for the complexes with L = py or bpy/2 and smaller when L = NH(3). Perdeuteration of the am(m)ine ligands (NH(3), en, or [14]aneN(4)) has no significant effect on the zero-point energy difference, and the contributions of the NH stretching vibrational modes to the emission band shape are too weak to resolve. Ligand perdeuteration does increase the excited-state lifetimes by a factor that is roughly proportional to the excited-state-ground-state energy difference, even though the CH and NH vibrational reorganizational energies are too small for nuclear tunneling involving these modes to dominate the relaxation process. It is proposed that metal-ligand skeletal vibrational modes and configurational mixing between metal-centered, bpy-ligand-centered, and MLCT excited states are important in determining the zero-point energy differences, while a large number of different combinations of relatively low-frequency vibrational modes must contribute to the nonradiative relaxation of the MLCT excited states.