Dye-sensitised solar cells based on TiO₂ nanotubes have been extensively characterised using a range of steady-state, time- and frequency-resolved techniques. In most cases the results can be explained fully in terms of models previously developed for dye-sensitised solar cells based on nano-particulate TiO₂ layers.[1]
TiO₂ nanotubes were grown by anodisation of Ti foil in fluoride containing electrolytes. The use of a Ti substrate naturally leads to a crack-free layer of compact TiO₂ at the base of the nanotubes which completely eliminates shunting via the substrate, this leads to improved efficiency under low-light conditions and simplifies analysis of kinetic data.
The effect of changing parameters such as tube length, wall thickness and crystallinity have been investigated and correlated with the kinetic characteristics of the cells as well as with overall power conversion efficiencies. Perhaps surprisingly it is found that neither transport nor recombination properties are markedly affected on changing from a nano-particulate to a nano-tubular morphology.
The most striking difference between nanotube and nanoparticle based cells is the highly non-ideal variation of steady-state photovoltage with incident light intensity in the case of nanotube based cells. Attempts at controlled surface modification are being made in order to establish the origin of the non-ideal steady-state photovoltage response to variation in incident light intensity.


[1] L. M. Peter, J. Phys. Chem. C, 111, 6601-6612, 2007