Self-exchange intermolecular Ru III/II electron transfer, a process commonly referred to as "hole-hopping", is of great interest as it provides a means of charge transport across the surface of nanocrystalline (anatase) TiO 2 mesoporous thin films without the loss of free energy. This process was characterized by cyclic voltammetry and chronoabsorptometry for three homologous Ru diimine compounds of the general form [Ru(LL) 2 (dcbH 2 )](PF 6 ) 2 , where LL is 2,2′-bipyridine (bpy), 4,4′-dimethyl-2,2′-bipyridine (dmb), or 4,4′-di-tert-butyl-2,2′-bipyridine (dtb) and dcbH 2 is 2,2′-bipyridyl-4,4′-dicarboxylic acid. Apparent electron diffusion coefficients, D, abstracted from this data increased with dtb < bpy < dmb. Both techniques were consistent with this trend, despite differences in the magnitude of D between the two methods. Temperature dependent measurements revealed an activation barrier for electron selfexchange of 250 ± 50 meV that was within this error the same for all three diimine compounds, suggesting the total reorganization energy, λ, was also the same. Application of Marcus theory, with the assumption that the 900 ± 100 meV total reorganization energy for self-exchange electron transfer was independent of the Ru compound, revealed that the electronic coupling matrix element, H AB , followed the trend dtb (0.02 meV) < bpy (0.07 meV) < dmb (0.10 meV). The results indicate that insulating side groups placed on redox active molecules can be utilized to tune the electronic coupling and hence selfexchange rate constants without significantly altering the reorganization energy for electron transfer on TiO 2 surfaces.

The synthesis, electrochemistry, and photophysical characterization are reported for 11 tris(bidentate) cyclometalated ruthenium(II) compounds, [Ru(N^N)(C^N)]. The electrochemical and photophysical properties were varied by the addition of substituents on the 2,2'-bipyridine, N^N, and 2-phenylpyridine, C^N, ligands with different electron-donating and -withdrawing groups. The systematic tuning of these properties offered a tremendous opportunity to investigate the origin of the rapid excited-state decay for these cyclometalated compounds and to probe the accessibility of the dissociative, ligand-field (LF) states from the metal-to-ligand charge-transfer (MLCT) excited state. The photoluminescence quantum yield for [Ru(N^N)(C^N)] increased from 0.0001 to 0.002 as more electron-withdrawing substituents were added to C^N. An analogous substituent dependence was observed for the excited-state lifetimes, τ, which ranged from 3 to 40 ns in neat acetonitrile, significantly shorter than those for their [Ru(N^N)] analogues. The excited-state decay for [Ru(N^N)(C^N)] was accelerated because of an increased vibronic overlap between the ground- and excited-state wavefunctions rather than an increased electronic coupling as revealed by a comparison of the Franck-Condon factors. The radiative (k) and non-radiative (k) rate constants of excited-state decay were determined to be on the order of 10 and 10-10 s, respectively. For sets of [Ru(N^N)(C^N)] compounds functionalized with the same N^N ligand, k scaled with excited-state energy in accordance with the energy gap law. Furthermore, an Arrhenius analysis of τ for all of the compounds between 273 and 343 K was consistent with activated crossing into a single, fourth MLCT state under the conditions studied with preexponential factors on the order of 10-10 s and activation energies between 300 and 1000 cm. This result provides compelling evidence that LF states are not significantly populated near room temperature unlike many ruthenium(II) polypyridyl compounds. On the basis of the underlying photophysics presented here for [Ru(N^N)(C^N)], molecules of this type represent a robust class of compounds with built-in design features that should greatly enhance the molecular photostability necessary for photochemical and photoelectrochemical applications.

Dye-sensitized bromide oxidation was investigated using a series of four ruthenium polypyridyl photocatalysts anchored to SnO/TiO core/shell mesoporous thin films through 2,2'-bipyridine-4,4'-diphosphonic acid anchoring groups. The ground- and excited-state reduction potentials were tuned over 500 mV by the introduction of electron withdrawing groups in the 4 and 4' positions of the ancillary bipyridine ligands. Upon light excitation of the surface-bound photocatalysts, excited-state electron injection yielded an oxidized photocatalyst that was regenerated through bromide oxidation. High injection quantum yields (Φ) and regeneration quantum yields (Φ) were essential to obtain efficient bromide oxidation yet required a photocatalyst that is both a potent photoreductant and a strong oxidant after excited-state injection. The four photocatalysts utilized in this manuscript ranged from unity Φ (1.0) and minimal Φ (0.037) to minimal Φ (0.09) and unity Φ (1.0). The photocatalyst that displayed the highest overall dye-sensitized photoelectrosynthesis cell performances exhibited near unity Φ (0.99), while a significant Φ was still preserved (0.59). Thus, these results highlighted the delicate interplay between the ground- and excited-state reduction potentials of photocatalysts for dye-sensitized hydrobromic acid splitting.

Lateral self-exchange electron transfer across oxide surfaces is important to many solar energy capture and conversion schemes. Substituent effects on lateral RuIII/II self-exchange electron transfer were studied using a series of RuII polypyridyl compounds of the type [Ru­(R2bpy)2(P)]2+, where P is 2,2′-bipyridyl-4,4′-diphosphonic acid and R2bpy was a 4,4′-substituted-2,2′-bipyridine with six different R groups: −OCH3, −C­(CH3)3, −CH3, −H, −Br, and −CF3. These functional groups were chosen mainly for their electron-withdrawing or -donating ability. Chronoabsorptometry was used to probe the apparent diffusion coefficient, D CA, that was proportional to the self-exchange rate constants, and these values were found to be between 2.8 × 10–11 and 7.9 × 10–9 cm2/s. The measured D CA values showed no correlation with the electron-withdrawing or -donating ability of the functional groups, but were instead correlated with the steric size of the substituents that also influenced the saturation surface coverage and thus the intermolecular distance. With some assumptions to estimate the intermolecular distance, the self-exchange rates were found to possess an exponential dependence with the distance, β = 1.2 ± 0.2 Å–1. Independent tests of the were carried out by varying the surface coverages from which β = 1.18 ± 0.09 Å–1 was found. The results indicate that the substituent’s steric size is the dominant factor that controls lateral RuIII/II self-exchange electron-transfer rates at these interfaces.

Three chromophores of the general form [Ru(bpy′) 2 (4,4′-(PO 3 H 2 ) 2 -2,2′bipyridine)] 2+ , where bpy′ is 4,4′-(C(CH 3 ) 3 ) 2 -2,2′-bipyridine (Ru(dtb) 2 P); 4,4′-(CH 3 O) 2 -2,2′-bipyridine (Ru(OMe) 2 P), and 2,2′-bipyridine (RuP) were anchored to mesoporous thin films of TiO 2 nanocrystallites at saturation surface coverages to investigate lateral self-exchange Ru III/II intermolecular hole hopping in 0.1 M LiClO 4 / CH 3 CN electrolytes. Hole hopping was initiated by a potential step 500 mV positive of the E 1/2 (Ru III/II ) potential or by pulsed laser (532 nm, 8 ns fwhm) excitation and monitored by visible absorption chronoabsorptometry and time-resolved absorption anisotropy measurements, respectively. The hole hopping rate constant k R extracted from the potential step data revealed self-exchange rate constants that followed the trend: TiO 2 |Ru(OMe) 2 P (k et = 1.4 × 10 6 s −1 ) > TiO 2 | RuP (7.1 × 10 5 s −1 ) > TiO 2 |Ru(dtb) 2 P (6.5 × 10 4 s −1 ). Analysis of the anisotropy data with Monte Carlo simulations provided hole hopping rate constants for TiO 2 |RuP and TiO 2 |Ru(dtb) 2 P that were within experimental error the same as that measured with the potential step. The hole hopping rate constants were found to trend with the TiO 2 (e − )|Ru III → TiO 2 |Ru II charge recombination rate constants. The atomic layer deposition of an ∼10 Å layer of Al 2 O 3 on top of the dye-sensitized films was found to prevent hole hopping by both initiation methods even though the chromophore surface coverage exceeded the percolation threshold and excited-state injection was efficient. The dramatic hole hopping turnoff was attributed to a larger outer-sphere reorganization energy for self-exchange due to the restricted access of electrolyte to the redox active chromophores. The implications of these findings for solar energy conversion applications are discussed.