Modern chemistry's grand challenge is to significantly improve catalysts for water splitting. Further progress requires detailed spectroscopic and computational characterization of catalytic mechanisms. We analyzed one of the most studied homogeneous single-site Ru catalysts, [Ru(II)(bpy)(tpy)H2O](2+) (where bpy = 2,2'-bipyridine, tpy = 2,2';6',2″-terpyridine). Our results reveal that the [Ru(V)(bpy)(tpy)═O](3+) intermediate, reportedly detected in catalytic mixtures as a rate-limiting intermediate in water activation, is not present as such. Using a combination of electron paramagnetic resonance (EPR) and X-ray absorption spectroscopy, we demonstrate that 95% of the Ru complex in the catalytic steady state is of the form [Ru(IV)(bpy)(tpy)═O](2+). [Ru(V)(bpy)(tpy)═O](3+) was not observed, and according to density functional theory (DFT) analysis, it might be thermodynamically inaccessible at our experimental conditions. A reaction product with unique EPR spectrum was detected in reaction mixtures at about 5% and assigned to Ru(III)-peroxo species with (-OOH or -OO- ligands). We also analyzed the [Ru(II)(bpy)(tpy)Cl](+) catalyst precursor and confirmed that this molecule is not a catalyst and its oxidation past Ru(III) state is impeded by a lack of proton-coupled electron transfer. Ru-Cl exchange with water is required to form active catalysts with the Ru-H2O fragment. [Ru(II)(bpy)(tpy)H2O](2+) is the simplest representative of a larger class of water oxidation catalysts with neutral, nitrogen containing heterocycles. We expect this class of catalysts to work mechanistically in a similar fashion via [Ru(IV)(bpy)(tpy)═O](2+) intermediate unless more electronegative (oxygen containing) ligands are introduced in the Ru coordination sphere, allowing the formation of more oxidized Ru(V) intermediate.

Catalytic O(2) evolution with cis,cis-[(bpy)(2)(H(2)O)Ru(III)ORu(III)(OH(2))(bpy)(2)](4+) (bpy is 2,2-bipyridine), the so-called blue dimer, the first designed water oxidation catalyst, was monitored by UV-vis, EPR, and X-ray absorption spectroscopy (XAS) with ms time resolution. Two processes were identified, one of which occurs on a time scale of 100 ms to a few seconds and results in oxidation of the catalyst with the formation of an intermediate, here termed [3,4]'. A slower process occurring on the time scale of minutes results in the decay of this intermediate and O(2) evolution. Spectroscopic data suggest that within the fast process there is a short-lived transient intermediate, which is a precursor of [3,4]'. When excess oxidant was used, a highly oxidized form of the blue dimer [4,5] was spectroscopically resolved within the time frame of the fast process. Its structure and electronic state were confirmed by EPR and XAS. As reported earlier, the [3,4]' intermediate likely results from reaction of [4,5] with water. While it is generated under strongly oxidizing conditions, it does not display oxidation of the Ru centers past [3,4] according to EPR and XAS. EXAFS analysis demonstrates a considerably modified ligand environment in [3,4]'. Raman measurements confirmed the presence of the O-O fragment by detecting a new vibration band in [3,4]' that undergoes a 46 cm(-1) shift to lower energy upon (16)O/(18)O exchange. Under the conditions of the experiment at pH 1, the [3,4]' intermediate is the catalytic steady state form of the blue dimer catalyst, suggesting that its oxidation is the rate-limiting step.

L(2,3)-edge X-ray absorption spectroscopy (XAS) has demonstrated unique capabilities for the analysis of the electronic structure of di-Ru complexes such as the blue dimer cis,cis-[Ru(III)(2)O(H(2)O)(2)(bpy)(4)](4+) water oxidation catalyst. Spectra of the blue dimer and the monomeric [Ru(NH(3))(6)](3+) model complex show considerably different splitting of the Ru L(2,3) absorption edge, which reflects changes in the relative energies of the Ru 4d orbitals caused by hybridization with a bridging ligand and spin-orbit coupling effects. To aid the interpretation of spectroscopic data, we developed a new approach, which computes L(2,3)-edges XAS spectra as dipole transitions between molecular spinors of 4d transition metal complexes. This allows for careful inclusion of the spin-orbit coupling effects and the hybridization of the Ru 4d and ligand orbitals. The obtained theoretical Ru L(2,3)-edge spectra are in close agreement with experiment. Critically, existing single-electron methods (FEFF, FDMNES) broadly used to simulate XAS could not reproduce the experimental Ru L-edge spectra for the [Ru(NH(3))(6)](3+) model complex nor for the blue dimer, while charge transfer multiplet (CTM) calculations were not applicable due to the complexity and low symmetry of the blue dimer water oxidation catalyst. We demonstrated that L-edge spectroscopy is informative for analysis of bridging metal complexes. The developed computational approach enhances L-edge spectroscopy as a tool for analysis of the electronic structures of complexes, materials, catalysts, and reactive intermediates with 4d transition metals.