Dye-Sensitized Solar Cell Constructed with Titanium Mesh and 3-D Array of TiO 2 Nanotubes † Cyrus S. Rustomji, ‡,| Christine J. Frandsen, | Sungho Jin, §,| and Michael J. Tauber* ,‡,| Department of Chemistry and Biochemistry, Department of Mechanical and Aerospace Engineering, and Materials Science and Engineering Program, UniVersity of California, San Diego, 9500 Gilman DriVe, La Jolla, California 92093-0314 ReceiVed: March 13, 2010; ReVised Manuscript ReceiVed: May 16, 2010 We have designed and constructed dye sensitized solar cells based on new, 3-D configurations of TiO 2 nanotubes. The overall efficiency of our best cells is 5.0% under standard air mass 1.5 global (AM 1.5 G) solar conditions, and the incident photon-to-current efficiency exceeds 60% over a broad part of the visible spectrum. Unlike prior nanotube-based cells where tubes are grown vertically in a 2-D array, the anodes of the present cells consist of tubes that extend radially in a 3-D array from a grid of fine titanium wires. The nanotubes are tens of micrometers in length, and the radial nature of the anode allows the photon absorption path length to exceed the electron transport distance (nanotube length). The cells are front-illuminated and do not require a transparent conductive oxide substrate at either the anode or cathode. The use of 3-D configured nanotubes and low-resistance titanium metal substrates are expected to enhance the performance and simplify the construction of large area dye-sensitized solar cells. Introduction There is a worldwide research effort aimed at improving the efficiency of dye-sensitized solar cells (DSSCs) beyond the current 11.2% record. 1 The primary goal is to reach efficiencies that are comparable to traditional solid-state solar cells while avoiding the costs associated with high-purity materials. 2–5 The exploration of new architectures for the photoanodes of DSSCs is one focus of current research that is expected to yield significant improvements. 6 The photoanode of traditional DSSCs is composed of a mesoporous network of TiO 2 nanoparticles. Electron transport in this layer is known to be a relatively slow process in DSSCs, and is one cause of decreased efficiencies. 5,7 Nanowires or nanotubes are expected to display improved charge transport properties in comparison with nanoparticles. The first reports of DSSCs based on 2-D vertical arrays of nanowires were important milestones, 8,9 but efficiencies of these cells were less than 2%. Subsequently, several groups reported DSSC anodes consisting of oriented TiO 2 nanotubes. 10–12 Cells with TiO 2 nanotubes currently have the best efficiencies among all nanotube and nanowire DSSCs, with the highest reported numbers ranging from 6.9 to 7.6%. 13–17 Although these efficien- cies are lower than the best numbers reported for traditional nanoparticle-based cells, improvements with nanotube cells have occurred rapidly. Furthermore, modulated photocurrent and photovoltage spectroscopies provide important evidence that TiO 2 nanotubes exhibit an order of magnitude slower charge recombination rate and significantly greater diffusion lengths relative to nanoparticle-based cells. 18,19 The unique properties of nanotubes allow new architectures for DSSCs. Most importantly, the improved charge transport of nanotubes spurs the exploration of photoactive anodes that are considerably thicker than the typical ∼10 µm thickness of nanoparticle-based cells. A thicker photoanode is a major advantage because the weak absorption of the best dyes in the near-infrared region can then be counteracted with a long path length for photon absorption. Among the best nanotube cells reported to date are those with tubes ranging from 20-35 µm in length. 14–16 Another feature of TiO 2 nanotube cells is the metallic substrate that typically underlies a back-illuminated array. The metal improves electrical conduction by several orders of magnitude in comparison with a transparent conductive oxide (TCO) layer. 20,21 The relatively high resistivity of TCO requires that large-area cells have closely spaced interconnect or “buss” wires. 22 The use of highly conductive metallic substrates in combination with nanotubes could significantly improve the performance and simplify the construction of large- area and flexible DSSCs, in comparison with traditional designs with TCO layers. 23,24 Despite the clear advantages associated with nanotube-based DSSCs, the benefits of a design with a 3-D nanotube array have not been fully realized. We report here a mesh-based cell with a 3-D array of nanotubes that are oriented in all directions ranging from parallel to perpendicular to the incoming photons. The 5.0% efficiency we measure is several times higher than the 1.47% efficiency reported in a concurrent study employing a similar mesh and nanotube design. 24 Our cell is easily constructed and does not require a TCO layer at the anode or cathode. We quantitatively analyze dye loading and the incident photon-to-current efficiency (IPCE) of these cells for nanotubes with lengths ranging from 11 to 42 µm. The analysis reveals that tubes oriented in all directions contribute substantially to the performance of the cell and that further improvements in efficiency are likely. Experimental Methods All solvents were obtained from Fisher or Sigma-Aldrich and used as received unless otherwise noted. Mesh made of † Part of the “Michael R. Wasielewski Festschrift”. * Corresponding author. E-mail: mtauber@ucsd.edu. Tel: (858)-534- 7334. ‡ Department of Chemistry and Biochemistry. § Department of Mechanical and Aerospace Engineering. | Materials Science and Engineering Program. J. Phys. Chem. B 2010, 114, 14537–14543 14537 10.1021/jp102299g  2010 American Chemical Society Published on Web 06/07/2010