The dynamics of electron-transfer processes between bis(tetrabutylammonium) cis-bis(thiocyanato)bis(2,2′bypiridine-4,4′-dicarboxylato)ruthenium(II) (called N719) and nanostructured ZnO films have been investigated by femtosecond and nanosecond spectroscopy. The incident photon to current conversion efficiency (IPCE) for these dye-sensitized electrodes was 36% in the maximum of 530 nm, corresponding to a quantum efficiency of 80%. The highest IPCE values were obtained when the electrodes were prepared under conditions where formation of dye aggregates in the pores of the nanostructured films is avoided. For such films, the electron injection time was in the subpicosecond regime (<300 fs), which is comparable to the N719-TiO 2 system. The back electron-transfer kinetics between conduction band electrons and oxidized dye molecules were biexponential with time constants of 300 ns and 2.6 µs. Variation of the light intensity did not affect the time constants, but only their relative weights. The kinetics of back electron transfer in the N719-ZnO and N719-TiO 2 systems were found to be identical.

No abstract

The anchoring of the ruthenium dye {(C 4 H 9 ) 4 N}[Ru(Htcterpy)(NCS) 3 ] (with tcterpy ) 4,4′,4′′-tricarboxy-2,2′:6′,2′′-terpyridine), the so-called black dye, onto nanocrystalline TiO 2 films has been characterized by UV-vis and FT-IR spectroscopies. FT-IR spectroscopy data suggest that dye molecules are bound to the surface by a bidentate binuclear coordination mode. The interfacial electron-transfer (ET) dynamics has been investigated by femtosecond pump-probe transient absorption spectroscopy and nanosecond laser flash photolysis. The electron-injection process from the dye excited state into the TiO 2 conduction band is biexponential with a fast component (200 ( 50 fs) and a slow component (20 ps). These two components can be attributed to the electron injection from the initially formed and the relaxed dye excited states, respectively. Nanosecond kinetic data suggest the existence of two distinguishable regimes (I and II) for the rates of reactions between injected electrons and oxidized dye molecules or oxidized redox species (D + or I 2 •-). The frontier between these two regimes is defined by the number of injected electrons per particle (N e ), which was determined to be about 1. The present kinetic study was undertaken within regime I (N e > 1). Under these conditions, the back-electron-transfer kinetics is comparable to that in systems with other ruthenium complexes adsorbed onto TiO 2 . The reduction of oxidized dye molecules by iodide results in the formation of I 2 •on a very fast time scale (<20 ns). Within regime I, the decay of I 2 •occurs in ∼100 ns via reaction with injected electrons (I 2 •-+ ef 2I -). In regime II (N e e 1), which corresponds to the normal operating conditions of dye-sensitized solar cells, the decay of I 2 •is very slow and likely occurs via the dismutation reaction (2I 2 •f I -+ I 3 -). Our results predict that, under high light intensity (N e > 1), the quantum efficiency losses in dye-sensitized solar cells will be important because of the dramatic acceleration of the reaction between I 2 •and injected electrons. Mechanisms for the ET reactions involving injected electrons are proposed. The relevance of the present kinetic studies for dye-sensitized nanocrystalline solar cells is discussed.

No abstract

No abstract

The use of 4.2 nm gold nanoparticles wrapped in an adsorbates shell and embedded in a TiO2 metal oxide matrix gives the opportunity to investigate ultrafast electron-electron scattering dynamics in combination with electronic surface phenomena via the surface plasmon lifetimes. These gold nanoparticles (NPs) exhibit a large nonclassical broadening of the surface plasmon band, which is attributed to a chemical interface damping. The acceleration of the loss of surface plasmon phase coherence indicates that the energy and the momentum of the collective electrons can be dissipated into electronic affinity levels of adsorbates. As a result of the preparation process, gold NPs are wrapped in a shell of sulfate compounds that gives rise to a large density of interfacial molecules confined between Au and TiO2, as revealed by Fourier-transform-infrared spectroscopy. A detailed analysis of the transient absorption spectra obtained by broadband femtosecond transient absorption spectroscopy allows separating electron-electron and electron-phonon interaction. Internal thermalization times (electron-electron scattering) are determined by probing the decay of nascent nonthermal electrons (NNEs) and the build-up of the Fermi-Dirac electron distribution, giving time constants of 540 to 760 fs at 0.42 and 0.34 eV from the Fermi level, respectively. Comparison with literature data reveals that lifetimes of NNEs measured for these small gold NPs are more than four times longer than for silver NPs with similar sizes. The surprisingly long internal thermalization time is attributed to an additional decay mechanism (besides the classical e-e scattering) for the energy loss of NNEs, identified as the ultrafast chemical interface scattering process. NNEs experience an inelastic resonant scattering process into unoccupied electronic states of adsorbates, that directly act as an efficient heat bath, via the excitation of molecular vibrational modes. The two-temperature model is no longer valid for this system because of (i) the temporal overlap between the internal and external thermalization process is very important; (ii) a part of the photonic energy is directly transferred toward the adsorbates (not among "cold" conduction band electrons). These findings have important consequence for femtochemistry on metal surfaces since they show that reactions can be initiated by nascent nonthermal electrons (as photoexcited, out of a Fermi-Dirac distribution) besides of the hot electron gas.

No abstract

We present a method to evaluate the parameters defining the efficiency of luminescent solar concentrators (LSCs). The light harvesting and self-absorption properties of thin film LSCs on glass substrates are determined by optical spectroscopy and the resulting optical efficiency is consistent with the directly measured photon flux gain.