Prof Henry Snaith FRS

Abstract:

Hybrid composites of semiconducting polymers and metal oxides are promising combinations for solar cells. However, forming a well-controlled nanostructure with bicontinuous interpenetrating networks throughout the photoactive film is difficult to achieve. Pre-structured "mesoporous" metal oxide electrodes can act as a well-defined template for latter polymer infiltration. However, the long range infiltration of polymer chains into contorted porous channels has appeared to elude the scientific community, limiting the advancement of this technology. Here we present a structural and electronic characterisation of poly(3-hexylthiophene) (P3HT) infiltrated into mesoporous dye-sensitized TiO 2. Through a combination of techniques we achieve uniform pore filling of P3HT up to depths of over 4 μm, but the volumetric fraction of the pores filled with polymer is less than 24%. Despite this low pore-filling, exceptionally efficient charge collection is demonstrated, illustrating that pore filling is not the critical issue for mesoporous hybrid solar cells. © 2011 The Royal Society of Chemistry.

Abstract:

Dye-sensitized solar cells (DSSC) are a realistic option for converting light to electrical energy. Hybrid architectures offer a vast materials library for device optimization, including a variety of metal oxides, organic and inorganic sensitizers, molecular, polymeric and electrolytic hole-transporter materials. In order to further improve the efficiency of solid-state dye-sensitized solar cells, recent attention has focused on using light absorbing polymers such as poly(3-hexylthiophene) (P3HT), to replace the more commonly used "transparent" 2,2′,7,7′-tetrakis-(N,N-di-p- methoxyphenyl-amine)9,9′spiro-bifluorene (spiro-OMeTAD), in order to enhance the light absorption within thin films. As is the case with spiro-OMeTAD based solid-state DSSC, the P3HT-based devices improve significantly with the addition of lithium bis(trifluoromethylsulfonyl)imide salts (Li-TFSI), although the precise role of these additives has not yet been clarified in solid-state DSCs. Here, we present a thorough study on the effect of Li-TFSI in P3HT based solid-state DSSC incorporating an indolene-based organic sensitizer termed D102. Employing ultrafast transient absorption and cw-emission spectroscopy together with electronic measurements, we demonstrate a fine tuning of the energetic landscape of the active cell components by the local Coulomb field induced by the ions. This increases the charge transfer nature of the excited state on the dye, significantly accelerating electron injection into the TiO2. We demonstrate that this ionic influence on the excited state energy is the primary reason for enhanced charge generation with the addition of ionic additives. The deepening of the relative position of the TiO2 conduction band, which has previously been thought to be the cause for enhanced charge generation in dye sensitized solar cells with the addition of lithium salts, appears to be of minor importance in this system. The cascade of photophysical events that occurs within the operating device when ions are incorporated in the dye-sensitized solar cells is described. It is demonstrated that the ionic influence on the excited state energy is the primary reason for enhanced charge generation and devices performance. Copyright © 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.

Abstract:

High-performance dye-sensitized solar cells are usually fabricated using nanostructured TiO(2) as a thin-film electron-collecting material. However, alternative metal-oxides are currently being explored that may offer advantages through ease of processing, higher electron mobility, or interface band energetics. We present here a comparative study of electron mobility and injection dynamics in thin films of TiO(2), ZnO, and SnO(2) nanoparticles sensitized with Z907 ruthenium dye. Using time-resolved terahertz photoconductivity measurements, we show that, for ZnO and SnO(2) nanoporous films, electron injection from the sensitizer has substantial slow components lasting over tens to hundreds of picoseconds, while for TiO(2), the process is predominantly concluded within a few picoseconds. These results correlate well with the overall electron injection efficiencies we determine from photovoltaic cells fabricated from identical nanoporous films, suggesting that such slow components limit the overall photocurrent generated by the solar cell. We conclude that these injection dynamics are not substantially influenced by bulk energy level offsets but rather by the local environment of the dye-nanoparticle interface that is governed by dye binding modes and densities of states available for injection, both of which may vary from site to site. In addition, we have extracted the electron mobility in the three nanoporous metal-oxide films at early time after excitation from terahertz conductivity measurements and compared these with the time-averaged, long-range mobility determined for devices based on identical films. Comparison with established values for single-crystal Hall mobilities of the three materials shows that, while electron mobility values for nanoporous TiO(2) films are approaching theoretical maximum values, both early time, short distance and interparticle electron mobility in nanoporous ZnO or SnO(2) films offer considerable scope for improvement.