Surface Passivation of Nanoporous TiO 2 via Atomic Layer Deposition of ZrO 2 for Solid-State Dye-Sensitized Solar Cell Applications Tina C. Li, † Ma ´rcio S. Go ´es, ‡,§ Francisco Fabregat-Santiago,* ,‡ Juan Bisquert, ‡ Paulo R. Bueno, § Chaiya Prasittichai, † Joseph T. Hupp,* ,† and Tobin J. Marks † Department of Chemistry and the Argonne-Northwestern Solar Energy Research Center, Northwestern UniVersity, EVanston, Illinois 60208-3113, PhotoVoltaic and Optoelectronic DeVices Group, Departament de Fı ´sica, UniVersitat Jaume I, 12071 Castello ´, Spain, and Departamento de Fı ´sico-Quı ´mica, Instituto de Quı ´mica de Araraquara, UniVersidade Estadual Paulista, R. Prof. Francisco Degni s/n, 14800-900 Araraquara SP, Brazil ReceiVed: July 12, 2009; ReVised Manuscript ReceiVed: August 22, 2009 We report here the utilization of atomic layer deposition to passivate surface trap states in mesoporous TiO 2 nanoparticles for solid-state dye-sensitized solar cells based on 2,2′,7,7′-tetrakis(N,N-di-p-methoxyphenylamine)- 9,9′-spirobifluorene (spiro-OMeTAD). By depositing ZrO 2 films with angstrom-level precision, coating the mesoporous TiO 2 produces over a two-fold enhancement in short-circuit current density, as compared to a control device. Impedance spectroscopy measurements provide evidence that the ZrO 2 coating reduces recombination losses at the TiO 2 /spiro-OMeTAD interface and passivates localized surface states. Low- frequency negative capacitances, frequently observed in nanocomposite solar cells, have been associated with the surface-state mediated charge transfer from TiO 2 to the spiro-OMeTAD. Introduction Dye-sensitized solar cells (DSCs) based on mesoporous titania and liquid electrolytes have been presented as a promising renewable energy source, achieving power conversion efficien- cies greater than 11%. 1 However, leakage of the commonly used I - /I 3 - redox electrolyte, with consequent compromise of long- term cell stability, has prompted efforts to find an efficient hole conductor for all solid-state DSCs (Figure 1). While the performance of solid-state DSCs (ssDSCs) is usually much lower than that with liquid electrolytes, the efficiencies of devices based on the molecular organic semiconductor spiro- OMeTAD have recently advanced. 2 Nevertheless, device per- formance is still constrained by competing transport and recombination dynamics, and electron diffusion lengths are limited to a few micrometers. 3 Like their liquid counterparts, ssDSCs generally rely on a network of sintered nanocrystalline TiO 2 particles, sensitized with a monolayer of a ruthenium-containing dye. 4-6 Upon illumination, the sensitizer is excited by absorption of a photon and injects an electron into the conduction band of the semiconductor. Electron transport through the titanium oxide framework can be described by a trapping/detrapping model, where an electron moves from trap-to-trap until collection at the cell’s transparent conducting oxide anode. The hole conduc- tor transports the hole remaining in the oxidized dye to the cathode via a hopping mechanism. 7 The collection efficiency of injected electrons in DSCs is less than unity on account of charge recombination processes. For ssDSCs, the principle charge losses come from “interception”, i.e., deleterious recombination of the electrons in the TiO 2 and the holes in the spiro-OMeTAD, due to the close contact between the two phases. 5,8 In comparison to liquid-junction DSCs, this effect is even more acute by lower charge screening, due to the lower ion mobility in the solid-state phase. Strategies to reduce interfacial interception have involved decreasing the TiO 2 electrode thickness below 3 µm, adding ionic salts to the spiro-OMeTAD solution, 9 replacing the traditional “N719” ruthenium polypyridyl sensitizer with the hydrophobic analogue “Z907”, 10 and inserting insulating layers between the TiO 2 and hole transport medium to prevent back reaction of the electrons. 11 Previously, dip-coating methods were employed to coat tunneling layers onto the nanoporous TiO 2 electrode 12-14 (i.e., Al 2 O 3 , MgO, SiO 2 ); however, control of the film growth through this technique is problematic, and com- paratively thick films are required to ensure complete coverage. Here, we address the importance of the inorganic/organic * To whom correspondence should be addressed. E-mail: fabresan@ fca.uji.es and j-hupp@northwestern.edu. † Northwestern University. ‡ Universitat Jaume I. § Universidade Estadual Paulista. Figure 1. MM2 energy minimization of the solid-state hole conductor, spiro-OMeTAD, used to replace I - /I 3 - (hydrogens omitted for clarity; carbons are gray, nitrogens are blue, and oxygens are red). J. Phys. Chem. C 2009, 113, 18385–18390 18385 10.1021/jp906573w CCC: $40.75  2009 American Chemical Society Published on Web 09/17/2009