An experimental methodology using UV-visible spectroscopy has been developed to study the effect of several operating variables on the adsorption of three different dyes (Direct Red 23, Fast Green SFC and ruthenium dye N719) onto titanium dioxide powders and also onto films of nanoparticulate titania particles which have been presintered onto TCO-coated glass substrates. The operating variables studied have included the concentration of the initial dye solution, the solvent used and sensitization time (24 h analysis) at room temperature. An initial adsorption study using four different grades of commercial titania powders including Degussa P25 and several paint grade titania powders, (named as Sample A, B, C and D) and two organic dyes (Direct
Red and Fast Green) using methanol as solvent was carried out. The equilibrium dye uptake data were fitted to model Langmuir and Freundlich isotherm equations and the experimental data were generally found to be well represented by both isotherm models. The sorption data were analyzed as well using first order and pseudo-second order kinetic models and the sorption kinetics were found to fit much more closely to a pseudo-second order kinetic model. The titania powders were characterized by XRD, SEM and TGA. Zero point charge (ZPC) and BET surface area data were also determined in order to correlate differences in the adsorption process with physical and chemical characteristic of the titania powders. TiO2 powder P25 from Degussa (Sample
A) exhibited the highest and fastest dye adsorption of the four powders tested.
Adsorption experiments were also carried out using titania films (opaque and
transparent) and the ruthenium dye -N719. The operating variables studied were initial dye concentration, the effect of the solvent used (ethanol and a mixture 50% in volume of acetonitrile/tert-butanol in this case) and contact time (24 h analysis). Langmuir and Freundlich models were applied to the experimental data. The kinetic adsorption data were analyzed by the application of two kinetics models (i.e. theoretical first and pseudo-second order models), to investigate the adsorption mechanism. The experimental data after a minimum of 8 h adsorption fitted well to the Langmuir model equation which predicts monolayer type adsorption. The adsorption data also fitted best to the pseudo-second order kinetic model. The sorption process onto titania films was found to be limited by the rate of dye-solvent diffusion through the nanoparticulate films during the first few hours of the adsorption process. Percolation analysis indicated that solvent and initial dye concentration have a very big impact on the rate of percolation and, because of this, on the dye adsorption rate as well. The influence of dye adsorption variables (e.g. dye concentration, the solvent used and the dyeing time) on the performance of dye-sensitised solar cell (DSC) devices has also been studied using both IV and IPCE data. These data have shown that a minimum of 5 h sensitization time using highly saturated dye solution is required to obtain DSC with 4% efficiency.