Interfacial Electron Transfer Dynamics for [Ru(bpy)₂((4,4′-PO₃H₂)₂bpy)]²⁺ Sensitized TiO₂ in a Dye-Sensitized Photoelectrosynthesis Cell: Factors Influencing Efficiency and Dynamics

Abstract

Nanosecond laser flash photolysis and photocurrent measurements have been used to investigate use of [(Ru(bpy)₂(4,4′-(PO₃H₂)₂bpy)]²⁺ attached to TiO₂ nanoparticle films, TiO₂−Ru^II, in a dye-sensitized photoelectrosynthesis cell (DSPEC) configuration for H₂ production. In these experiments, laser flash excitation of TiO₂−Ru^II and rapid injection lead to TiO₂(e⁻)−Ru^III with subsequent TiO₂(e⁻)−Ru^III → TiO₂−Ru^II back electron transfer monitored on the nsec time scale with and without added triethanolamine (TEOA) and deprotonated ethylenediaminetetraacetic tetra-anion (EDTA⁴⁻) as irreversible electron transfer donors. With added TEOA or EDTA⁴⁻, a competition exists between back electron transfer and scavenger oxidation with the latter leading to H₂ production in the photoelectrosynthesis cell. Reduction of TiO₂(e⁻)−Ru^III by both TEOA and EDTA⁴⁻ occurs with <i>k</i>_D ∼ 10⁶ M⁻¹ s⁻¹. EDTA⁴⁻ is a more efficient scavenger by a factor of ∼3 because of a more favorable partition equilibrium between the film and the external solution. Its increased scavenger efficiency appears in incident photon-to-current conversion efficiency (IPCE) measurements, in electron collection efficiencies (η_coll), and in photocurrent measurements with H₂ production. Evaluation of electron collection efficiencies by transient current measurements gave η_coll ∼ 24% for TEOA and ∼ 70% for EDTA⁴⁻. The dynamics of back electron transfer are minimized, and collection efficiencies, photocurrents, and hydrogen production are maximized by application of a positive applied bias consistent with the results of I−V measurements. A pH dependent plateau is reached at ∼0 V at pH = 4.5 (EDTA⁴⁻) and at ∼ −0.4 V at pH 6.7 (TEOA). The difference is qualitatively consistent with the influence of pH on electron population in trap states below the conduction band and the role they play in back electron transfer. The excitation dependence of IPCE measurements matches the spectrum of TiO₂−Ru^II with IPCE values ∼3 times higher for EDTA⁴⁻ than for TEOA as noted above. Absorbed photon-to-current efficiency (APCE) values are light-intensity dependent because of the effect of multiple injection events and the influence of increasing trap site electron densities on back electron transfer. The key to efficient H₂ production is minimizing back electron transfer. Application of a sufficiently positive potential relative to <i>E</i>_CB for TiO₂ accelerates loss of electrons from the film in competition with back electron transfer allowing for H₂ production with efficiencies approaching 14.7% under steady-state irradiation.