During the past five years, we have developed in our laboratory a new type of solar cell that is based on a photoelectrochemical process. The light absorption is performed by a monolayer of dye (i.e., a Ruthenium complex) that is adsorbed chemically at the surface of a semiconductor (i.e., titanium oxide (TiO2)). When excited by a photon, the dye has the ability to transfer an electron to the semiconductor. The electric field that is inside the material allows extraction of the electron, and the positive charge is transferred from the dye to a redox mediator that is present in solution. A respectable photovoltaic efficiency (i.e., 10%) is obtained by the use of mesoporous, nanostructured films of anatase particles. We will show how the TiO2 electrode microstructure influences the photovoltaic response of the cell. More specifically, we will focus on how processing parameters such as precursor chemistry, temperature for hydrothermal growth, binder addition, and sintering conditions influence the film porosity, pore‐size distribution, light scattering, and electron percolation and consequently affect the solar‐cell efficiency.

A monolayer of a phosphonated triarylamine adsorbed on nanocrystalline TiO2, ZrO2, or Al2O3 film deposited on conducting glass displays reversible electrochemical and electrochromic behavior although the redox potential of the electroactive molecules (0.80 V vs NHE) lies in the forbidden band of the semiconducting or insulating oxides. The mechanism of charge transport was found to involve hole injection from the conducting support followed by lateral electron hopping within the monolayer. The apparent diffusion coefficient ranged from 2.8 × 10(-12) m(2) s(-1) in the neat 1-ethyl-2-methylimidazolium bis(trifluoromethylsulfonyl)imide (EtMeIm(+)Tf2N(-)) to 1.1 × 10(-11) m(2) s(-1) in acetonitrile + 2 M EtMeIm(+)Tf2N(-). A percolation threshold for electronic conductivity was found at a surface coverage corresponding to 50% of a full monolayer.

Solid-state dye-sensitized photovoltaic cells have been fabricated with TiO2 as the electron conductor and CuSCN as the hole conductor. These cells involve the nanoscale mixing of crystalline n-type and p-type semiconductors in films that are more than 100 times thicker than the individual n- and p-type domains. Charge transport and field distribution in this kind of material are as yet unexplored. We have used photocurrent and photovoltage transients, combined with variation in the layer thickness, to examine the limiting factors in charge transport and recombination. Charge transport (t 1/2 ≈ 200 μs) is found to be similar to that in dye-sensitized electrolyte cells. Recombination at V oc (t 1/2 ≈ 150 μs) is 10 times faster than in electrolyte cells, and recombination at short circuit (t 1/2 ≈ 450 μs) is 100 times faster. In the solid-state cells, the similarity of the charge transport and recombination rates results in a low fill factor, and photocurrent losses, both important limiting factors of the efficiency. A simple model is given, and suggestions are made for improvements in efficiency.

The onset wavelengths of the surface photovoltage (SPV) in dye-sensitized solar cells (DSSCs) with different mesoporous, wide-band gap electron conductor anode materials, viz., TiO 2 (anatase), Nb 2 O 5 (amorphous and crystalline), and SrTiO 3 , using the same Ru bis-bipyridyl dye for all experiments, are different. We find a clear dependence of these onset wavelengths on the conduction band edge energies (E CB ) of these oxides. This is manifested in a blue-shift for cells with Nb 2 O 5 and SrTiO 3 compared to those with TiO 2 . The E CB levels of Nb 2 O 5 and SrTiO 3 are known to be some 200-250 meV closer to the vacuum level than that of our anatase films, while there is no significant difference between the optical absorption spectra of the dye on the various films. We, therefore, suggest that the blue shift is due to electron injection from excited-state dye levels above the LUMO into Nb 2 O 5 and SrTiO 3 . Such injection comes about because, in contrast to what is the case for anatase, the LUMO of the adsorbed dye in the solution is below the E CB of these semiconductors, necessitating the involvement of higher vibrational and/or electronic levels of the dye, with the former being more likely than the latter. While for Nb 2 O 5 hot electron injection has been proposed earlier, on the basis of flash photolysis experiments, this is the first evidence for such ballistic electron-transfer involving SrTiO 3 , a material very similar to anatase but with a significantly smaller electron affinity. Additional features in the SPV spectra of SrTiO 3 and amorphous Nb 2 O 5 (but not in those of crystalline Nb 2 O 5 ) can be understood in terms of hole injection from the dye into the oxide via intraband gap surface states.

Solid-state dye-sensitized photovoltaic cells have been fabricated with TiO 2 as the electron conductor and CuSCN as the hole conductor. The cells show photocurrents of ≈8 mA/cm 2 , voltages of ∼600 mV, and energy efficiencies of ≈2% at 1 sun. The CuSCN was deposited into the pores of the nanoparticulate TiO 2 /dye film from dilute solution in propylsulfide. The degree of pore filling achieved is near 100% for TiO 2 films <2-µm thick and falls to ≈65% for films near 6 µm. The final drying step after the CuSCN deposition is shown to be critical; drying in vacuum or argon is required for photocurrents above 2 mA/cm 2 . The photocurrent IVs of these cells are fit to a single diode equation and the results are discussed and compared to those for equivalent photoelectrochemical cells, and similar solid cells composed of ZnO/dye/CuSCN.

Silver nanoparticle arrays placed on top of a high-refractive index substrate enhance the coupling of light into the substrate over a broad spectral range. We perform a systematic numerical and experimental study of the light incoupling by arrays of Ag nanoparticle arrays in order to achieve the best impedance matching between light propagating in air and in the substrate. We identify the parameters that determine the incoupling efficiency, including the effect of Fano resonances in the scattering, interparticle coupling, as well as resonance shifts due to variations in the near-field coupling to the substrate and spacer layer. The optimal configuration studied is a square array of 200 nm wide, 125 nm high spheroidal Ag particles, at a pitch of 450 nm on a 50 nm thick Si(3)N(4) spacer layer on a Si substrate. When integrated over the AM1.5 solar spectral range from 300 to 1100 nm, this particle array shows 50% enhanced incoupling compared to a bare Si wafer, 8% higher than a standard interference antireflection coating. Experimental data show that the enhancement occurs mostly in the spectral range near the Si band gap. This study opens new perspectives for antireflection coating applications in optical devices and for light management in Si solar cells.

Lithium insertion into nanotextured anatase, either pure or stabilized with zirconia, was studied using cyclic voltammetry. The voltammograms were sensitive to various morphologies of the tested electrodes. Special attention was paid to the zirconia-stabilized molecular sieve, PNNL-1 (Pacific Northwest National Laboratory), which was prepared by a surfactant-templated route. The voltammetric peak-to-peak splitting decreases monotonically with the electrode specific surface area. The charge-transfer rate constants decrease in the same series. PNNL-1 is more stable against prolonged heat-treatment, as compared to ordinary nanocrystalline anatase. The latter exhibits thermal particle growth and, presumably, partial phase transformation to rutile. PNNL-1 presents a promising model electrode material for Li-ion batteries, although its faradaic capacity is limited by the presence of a nonactive stabilizer (zirconia).

No abstract