REVIEW ARTICLES Physical Chemical Principles of Photovoltaic Conversion with Nanoparticulate, Mesoporous Dye-Sensitized Solar Cells Juan Bisquert,* ,² David Cahen,* ,‡ Gary Hodes,* ,‡ Sven Ru 1 hle, ‡ and Arie Zaban* ,§ Departament de Cie ` ncies Experimentals, UniVersity Jaume 1, 12080 Castello ´ , Spain, Department of Materials & Interfaces, Weizmann Institute of Science, RehoVot 76100, Israel, and Department of Chemistry, Bar Ilan UniVersity, Ramat Gan 52900, Israel ReceiVed: July 4, 2003; In Final Form: December 12, 2003 We review the status of the understanding of dye-sensitized solar cells (DSSC), emphasizing clear physical models with predictive power, and discuss them in terms of the chemical and electrical potential distributions in the device. Before doing so, we place the DSSC in the overall picture of photovoltaic energy converters, reiterating the fundamental common basis of all photovoltaic systems as well as their most important differences. 1. Introduction Solar energy is one of the most promising future energy resources. The direct conversion of sunlight into electric power by solar cells is of particular interest because it has many advantages over most presently used electrical power generation methods. Electricity is produced without the exhaust of green- house gases and without nuclear waste byproducts. The dye- sensitized solar cell (DSSC) appears to have significant potential as a low-cost alternative to conventional p-n junction solar cells. A DSSC consists of a nanocrystalline, mesoporous network of a wide band gap semiconductor (usually TiO 2 ), which is covered with a monolayer of dye molecules (usually a Ru dye). The semiconductor is deposited onto a transparent conductive oxide (TCO) electrode, through which the cell is illuminated. The TiO 2 pores are filled with a redox electrolyte (I - /I 3 - ) that acts as a conductor and that is electrically connected to a platinum electrode. Upon illumination, electrons are injected from the photoexcited dye into the semiconductor and move toward the TCO substrate, while the electrolyte reduces the oxidized dye and transports the positive charges to the Pt electrode. Such systems can reach solar to electric conversion efficiencies of about 10% 1 but are still not produced on a large scale mainly because of technical problems such as sealing. At present, p-n junction solar cells are the most efficient light-to-electric power conversion devices, and they are produced in much larger quantities than any other types of solar cell. In ap-n junction solar cell, the difference in the work function between the p and n material leads to a spatial variation of the band energies (reflected in the “bending” of the conduction and valence bands 2 ), which is thought to be the main origin of the photovoltaic response. Because of the dominant position of this type of cell, possible alternatives have not attracted very much commercial attention. From a fundamental scientific point of view, most alternatives to the single- or multicrystalline Si cells have often been described in terms of the models that are valid for the latter cell types (i.e., a “diode principle” according to which charge dissociation and charge collection in photovoltaic devices is determined by a built-in electrostatic field). However, this approach should be scrutinized carefully, at least for DSSCs with their nanocrystalline mesoporous electrodes and for most types of organic solar cells. The photochemical model used to describe photosynthesis is also relevant to the description of DSSC operation. 3,4 In a photochemical converter, 3 light selectively excites the light- absorbing molecules, which constitute part of the converter, and causes a transition of the electronic carriers from a lower, ground level to a higher-lying electronic level. The system can now be viewed as being in a combination of a ground and an excited electronic state. Quasi-chemical potentials of the electrons will be associated with the system in the ground and excited states (by analogy to quasi-Fermi levels or, for brevity, Fermi levels 5 ), and their difference determines the amount of useful work (or free energy) that can be obtained as a result of light absorption by such a system. These systems are most often heterogeneous, with different phases microscopically mixed. With the advent of DSSCs and plastic solar cells, some of which are much closer to the photochemical converter than to the photoelectric diode, it became interesting to look for the common denominator of these two seemingly distinct classes of converters to identify their common basic physical features. Recent work has produced useful ideas in this sense, albeit using extremely idealized models. 6-8 Concerning DSSCs, these are questions that have been the subject of discussion and some controversy. 9-13 The analysis of general principles is scientifically interesting and useful for understanding new kinds of solar cells, even though empirical optimization played a major role in the development of most of today’s best cells. Still, there is room for models, with predictive power, that can describe the devices. * Corresponding authors. E-mail: david.cahen@weizmann.ac.il. ² University Jaume 1. ‡ Weizmann Institute of Science. § Bar Ilan University. 8106 J. Phys. Chem. B 2004, 108, 8106-8118 10.1021/jp0359283 CCC: $27.50 © 2004 American Chemical Society Published on Web 05/25/2004