Transient Absorption Spectroscopy of Ruthenium and Osmium Polypyridyl Complexes Adsorbed onto Nanocrystalline TiO₂ Photoelectrodes

Abstract

Transient absorption spectroscopy has been used to probe the electron injection dynamics of transition metal polypyridyl complexes adsorbed onto nanocrystalline TiO 2 photoelectrodes. Experiments were performed on photoelectrodes coated with Ru(H 2 L′) 2 (CN) 2 , Os(H 2 L′) 2 (CN) 2 , Ru(H 2 L′) 2 (NCS) 2 , or Os(H 2 L′) 2 (NCS) 2 , where H 2 L′ is 4,4′-dicarboxylic acid-2,2′-bipyridine, to study how the excited-state energetics and the nature of the metal center affect the injection kinetics. All of these complexes exhibited electron injection dynamics on both the femtosecond and picosecond time scales. The femtosecond components were instrument-limited (<200 fs), whereas the picosecond components ranged from 3.3 ( 0.3 ps to 14 ( 4 ps (electron injection rate constants k 2 ′ ) (7.1-30) × 10 10 s -1). The picosecond decay component became more rapid as the formal excited-state reduction potential of the complex became more negative. Variable excitation wavelength studies suggest that femtosecond injection is characteristic of the nonthermalized singlet metal-to-ligand chargetransfer ( 1 MLCT) excited state, whereas picosecond injection originates from the lowest-energy 3 MLCT excited state. On the basis of these assignments, the smaller relative amplitude of the picosecond component for the Ru sensitizers suggests that electron injection from nonthermalized excited states competes more effectively with 1 MLCT f 3 MLCT conversion for the Ru sensitizers than for the Os sensitizers.

Femtosecond Absorption Spectroscopy of Transition Metal Charge-Transfer Complexes

“…This in turn helped to shape our thinking about the mechanism of charge injection in photovoltaic devices employing this molecule as a sensitizer. 15,16 It seems likely that as these and related 17 computational methods continue to evolve, they will become an important tool in the interpretation of ultrafast spectroscopic data.…”

Parameters Influencing the Efficiency of Electron Injection in Dye-Sensitized Solar Cells

“…[20][21][22] However the influence of such additives upon the efficiency of electron injection in complete DSSCs, and their correlation with device performance, have received only limited attention to date. 7,8,24,32,33 This gap arises partly from the difficulty of measuring injection in complete devices using femtosecond transient absorption instruments. As a consequence, DSSC device optimisation studies to date typically have not focused on electron injection dynamics as being a significant factor determining device performance.…”

Comparison of Interfacial Electron Transfer through Carboxylate and Phosphonate Anchoring Groups†

“…8,[11][12][13][14][15][16] The incident photon-to-current conversion efficiency was found to be near unity in RuN3-sensitized TiO 2 solar cells, 2,10 and the high efficiency has been attributed to an ultrafast electron injection and a much slower charge recombination that happens on the microsecond to millisecond time scales. [17][18][19] The injection kinetics has been shown to be biphasic, consisting of a primary <100 fs component and slower components on a few to tens of picosecond time scales. [19][20][21][22][23][24][25][26][27][28][29][30][31][32][33] The ultrafast component has been attributed to injection from the unrelaxed singlet metal-to-ligand charge transfer ( 1 MLCT) state and the slower components to injection from the 3 MLCT states near the band edge.…”

“…[17][18][19] The injection kinetics has been shown to be biphasic, consisting of a primary <100 fs component and slower components on a few to tens of picosecond time scales. [19][20][21][22][23][24][25][26][27][28][29][30][31][32][33] The ultrafast component has been attributed to injection from the unrelaxed singlet metal-to-ligand charge transfer ( 1 MLCT) state and the slower components to injection from the 3 MLCT states near the band edge. 24,27,31,32,34,35 The ultrafast injection process from the unrelaxed excited-state competes with the ultrafast (∼75 fs) 31,32 intramolecular relaxation processes within the dense manifold of excited states.…”

“…Negligible oxidized peak amplitude was observed at <200 fs, indicating a lack of instantaneous injection component that was observed for RuN3 and derivatives on TiO 2 . 19,[23][24][25][26][27]31,32 This has been attributed to the insertion of CH 2 spacer between the bipyridine and the anchoring group, which reduces its electronic coupling with TiO 2 , slowing down electron injection rate. 37 The injection rate can be better quantified by comparing injection kinetics in these systems.…”