Band edge engineering of composite photoanodes for dye-sensitized solar cells Venkata Manthina a, b , Alexander G. Agrios a, b, * a Department of Civil & Environmental Engineering, University of Connecticut, 261 Glenbrook Rd Unit 3037, Storrs, CT 06269, USA b Center for Clean Energy Engineering, University of Connecticut, 44 Weaver Rd, Storrs, CT 06269, USA A R T I C L E I N F O Article history: Received 13 February 2015 Received in revised form 14 April 2015 Accepted 15 April 2015 Available online 16 April 2015 Keywords: zinc oxide titanium dioxide doping charge transfer nanocomposites A B S T R A C T As dye-sensitized solar cells (DSSCs) transition from iodide/triiodide-based electrolytes to organome- tallic complex redox couples with higher rates of recombination with electrons in the semiconductor, there is a need for semiconductor nanostructures that can rapidly transport electrons out of the device while maintaining high surface areas for the semiconductor/dye/electrolyte interface. A previously reported composite, with TiO 2 nanoparticles coating ZnO nanorods, met these criteria but suffered from a barrier to electron transfer from the TiO 2 to the ZnO. Here, the band edge positions of the TiO 2 and ZnO have been shifted by doping with Zr 4+ and Co 2+ , respectively, to arrive at the desired energetic alignment. The materials were characterized using diffuse-reflectance spectroscopy and a three-electrode measurement of the open circuit photovoltage under bandgap excitation (OCV). The OCV measurement indicated that the doping moved the conduction band minimum of ZnO to a more positive potential than that of the TiO 2 , enabling electron transfer from dye-sensitized TiO 2 nanoparticles to the underlying ZnO nanorods for efficient charge collection. However, DSSC devices fabricated with the composite nanostructures did not show improved performance. This paper details a methodology for producing and measuring band-edge shifts along with the benefits and limitations thereof. ã 2015 Elsevier Ltd. All rights reserved. 1. Introduction One-dimensional nanostructured metal oxides are promising materials for applications such as dye sensitized solar cells [1,2] (DSSCs) and water splitting [3] where rapid electron transport and high interfacial area are needed. This becomes especially impor- tant as DSSCs move away from iodide/triiodide-based electrolytes and toward redox couples, such as cobalt(II/III) [4] or ferrocene/ ferrocenium (Fc/Fc + ), [5] that have faster rates of recombination with electrons in the photoanode, necessitating fast electron transport out of the photoanode for efficient charge collection. ZnO has been found particularly useful in DSSCs [1,6–9] since it can be grown in 1-D morphologies using facile methods and has other favorable material properties such as a high electron mobility and appropriate band edge positions. The major drawbacks of ZnO nanorods in DSSCs, as compared to the standard TiO 2 nanoparticle film [10], are a reduced surface area and reactions with carboxylic dyes resulting in partial dissolution of the surface. Some ZnO–TiO 2 and ZnO–ZnO composite nano- structures have been proposed to address one or both of these problems [11–15], but in our own composite of ZnO nanorods coated with TiO 2 nanoaparticles, we reported evidence of an energy barrier preventing electron injection from TiO 2 nano- particles to ZnO nanorods, hindering charge collection [16]. For an efficient core-shell photoanode, the conduction band minimum (CBM) potential of the shell material must be more negative than that of the core material (but more positive than the dye LUMO level). Achieving this condition requires raising the CBM energy of the TiO 2 and/or lowering the CBM energy of the ZnO, as depicted in Fig. 1. A number of transition metal dopants have been incorporated into metal oxide semiconductors to change their optical and photoelectrochemical properties [17–20]. In particular, previous reports have shown that Co 2+ can substitute for Zn 2+ atoms in ZnO, causing the lattice cell to contract due to the slightly smaller Shannon crystal radius of Co 2+ (0.72 Å) compared to Zn 2+ (0.74 Å) in the wurtzite structure, resulting in a smaller bandgap [21–23]. In TiO 2 , there are multiple reports that Zr 4+ substitution of some Ti 4+ atoms results in an increased bandgap [24–27]. If the O 2p orbitals remain largely unchanged, keeping the valence band maximum (VBM) potential constant, the altered bandgap corresponds mainly to a CBM shift. We have doped the ZnO with cobalt to lower its CBM energy and doped TiO 2 with zirconium to raise its CBM energy. In this report * Corresponding author. Tel.: +1 860 486 1350. E-mail address: agrios@engr.uconn.edu (A.G. Agrios). http://dx.doi.org/10.1016/j.electacta.2015.04.080 0013-4686/ ã 2015 Elsevier Ltd. All rights reserved. Electrochimica Acta 169 (2015) 416–423 Contents lists available at ScienceDirect Electrochimica Acta journal homepa ge: www.elsev ier.com/locate/electacta