Rationale for Kinetic Heterogeneity of Ultrafast Light-Induced Electron Transfer from Ru(II) Complex Sensitizers to Nanocrystalline TiO₂

Abstract

The conversion of light into electricity in dye-sensitized solar cells (DSC) is initiated by the charge injection from an electronically excited dye molecule into the conduction band of a wide band gap oxide semiconductor. 1,2 Because of their successful use in such devices, Ru(II) polypyridyl complex dyes adsorbed on nanocrystalline titanium dioxide films have been regarded as model systems for the experimental study of the ultrafast dynamics of interfacial light-induced electron transfer (ET). Yet, the complex kinetic behavior observed for the charge injection process in this case has prevented the development of a satisfying kinetic model and has led to often contradicting conclusions. One of the most successful sensitizers used up to now for DSC is the N3 ruthenium(II) complex (Ru II (dcbpyH 2 ) 2 (NCS) 2 ), which is known to inject an electron into TiO 2 with practically unit quantum efficiency. 3 The kinetics of ET from N3 have been under study for the past decade. [2][3][4][5][6][7][8] In a widely referred to study, in particular, Benkö et al. reported the charge injection process to take place with a biphasic kinetics. 5 The first ultrafast component was estimated to have a rise time of 28 fs and the second multiexponential part to occur within the 1-50 ps time range. This behavior was rationalized in terms of a two-state mechanism, the fast and slow components being attributed to the injection from the singlet and triplet excited states of the ruthenium complex, respectively. Although several other studies confirm the presence of the slower component, its relative contribution ranges from <5 to 65%, and the time constants vary from 0.7 to 100 ps, with a marked nonexponential behavior and dependence upon experimental conditions. [4][5][6][7][8][9] This discrepancy actually questions the proposed interpretation. In other published works, the kinetic heterogeneity of the charge injection was attributed to dye molecules adsorbed on energetically diverse sites or in various spatial configurations at the surface of the nanocrystalline titania films. 9 Here, we show that the slow component of electron injection arises from sensitizer molecules that are loosely attached onto the surface or are present in an aggregated form. A monophasic ET with a rise time shorter than 20 fs is observed when the formation of aggregates is prevented and the sensitizer is adsorbed as a monolayer on the surface of TiO 2 nanocrystals.Kinetic measurements employed a commercially available N3 dye (Solaronix, Switzerland) and also home-synthesized N3 and its doubly deprotonated derivative (Bu 4 N) 2 [Ru(dcbpyH) 2 (NCS) 2 ] (N719). 9 Ruthenium complexes were adsorbed onto nanocrystalline titanium dioxide films from 0.3 mM N3 solutions in ethanol or 0.5 mM N719 solutions in acetonitrile/tert-butyl alcohol (1:1) solvent mixture for 12 h. 3,10 Dyed samples displayed typically an OD of 1.5 at λ ) 534 nm. They were covered with a film of the redoxinactive ionic liquid (1-ethyl-2-methylimidazoliumbis(trifluoromethylsulfonyl)imide). 11 The f...

Effect of Posttreatment of Titania Mesoscopic Films by TiCl4 in Solid-State Dye-Sensitized Solar Cells: A Time-Resolved Spectroscopy Study

“…19 A similar biphasic behavior was found in the case of the treated films, with a 60 ps component accounting for 20% of the total signal. This is also consistent with findings of Tiwana et al, who observed by terahertz measurements a biphasic charge injection with a 50−70 ps component and no noticeable difference in charge injection for treated and nontreated samples.…”

Ultrafast Terahertz Conductivity Dynamics in Mesoporous TiO2: Influence of Dye Sensitization and Surface Treatment in Solid-State Dye-Sensitized Solar Cells

“…Similar biphasic charge injection kinetics have been observed by other groups, who have suggested that they are caused by injection from the singlet and triplet excited states of the ruthenium complex, 38,39 injection from energetically or spatially diverse sites from the dye into TiO 2 , 40 or because of aggregated or loosely attached dye molecules on the TiO 2 surface. 41 We will explore the origins of this slow rise component for injection via Z907 dye in future work.…”

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

“…31 We note that in the studies reported herein the senisitizing solution was sonicated prior to senitization to break up any such dye aggregates. We further note we obtained similar injection data for a broad range of dye loadings, suggesting that any dye aggregation induced by high dye loadings did not significantly impact upon the observed data.…”

“…30 Similarly, dye aggregation has been suggested to reduce the efficiency of electron injection for some sensitizer dyes, thereby reducing device performance. 31,32 Of particular relevance to this paper, studies to optimise device performance have frequently employed electrolyte additives, such as t-butyl pyridine (tBP) and Li + , to enhance device performance. The influence of these additives on device performance has been generally attributed to their influence on the TiO 2 surface charge, and therefore conduction band (CB) edge.…”