Nanocrystalline titanium dioxide electrodes are standard choice for the construction of Dye-sensitized Solar Cells (DSC). Nanocrystallinity, small number of defect sites, optimized electrical conduct between nanoparticles and sufficient active surface area are crucial factors, which determine the quality of the electrodes and, subsequently, the overall efficiency of the cell. The maximum efficiency of DSC ever reported was around 10.4 % and referred to cells based on a liquid electrolyte and on non-transparent Light-Scattering Electrodes (LSE) of 18 m thickness. For several years, we have made efforts to reach comparable cell efficiencies with DSC’s employing thinner transparent TiO2 electrodes by optimizing the synthesis and deposition procedures. The motive of these efforts is related with the prospect of utilizing such transparent cells as photovoltaic windows. LSEs do not offer this possibility but, instead, they entrap light and result in higher cell efficiencies. In order for Transparent Electrodes (TE) to reach performance of LSE, it is necessary that the former are deposited with optimized parameters: good quality nanocrystals without defects, compact enough films to allow sufficient electric contact between nanoparticles and sufficient mesoporosity to allow dye-sensitizer and electrolyte filling. We have achieved this goal by synthesizing titania through the sol-gel procedure in the presence of surfactant templates under ambient conditions. Spatially isolated self-organized entities seem to encourage formation of good quality nanosrystalline particles. For example, reverse micelles dispersed in an organic solvent, originally used to synthesize CdS nanocrystals proved excellent templates for TiO2 nanocrystals as well. However, reverse micelles are rather costly and less environmentally friendly systems. It has been found that simpler systems can offer similar results and more handy experimental procedures. Thus organic acid solvolysis of titanium isopropoxide in the presence of the non-ionic surfactant Triton X-100 (polyoxyethylene-(10) isooctylphenyl ether), can assist nanocrystalline anatase deposition with nanoparticle size of around 12 nm and active surface area around 110 m g. About 2 lm thick TE were made by subsequent layer deposition by dipping under ambient conditions. Organic acid solvolysis in the absence of water has the advantage of slow sol-gel evolution that allows surfactant assembling and shell formation around titania core. The choice of this route was rewarding. Indeed, as it has been previously reported, such electrodes gave DSCs with efficiency > 9 %, i.e., not far from the reported maximum. The DSC of ref. [4] was a non-optimized cell employing cis-bis(isothiocyanato)bis(2,2′-bipyridyl-4,4′-dicarboxylato)-ruthenium(II) [abbreviated RuL2(NCS)2] as dye sensitizer and I3 /I in acetonitrile as liquid redox electrolyte. When the same electrodes were used to construct quasi-solid state DSCs, where the redox couple I3 /I was incorporated into a sol-gel nanocomposite organic-inorganic gel, efficiency dropped but it was still at satisfactory levels of 5.3 %. This efficiency decrease was due to current decrease, which in turn was due to lower ionic mobility in the gel. Gel electrolytes based on nanocomposite organic-inorganic materials is our approach to ensure cell stability and avoidance of electrolyte leakage. In the present work we are testing an alternative method of titania deposition, in which we try to further increase cell efficiency by producing LSEs, using new organic templates or combined templating methods. It will be seen that indeed LSEs do increase cell efficiency by an important percentage, however, this increase should be appreciated by also taking into account the disadvantages related with the loss of transparency. The cell described in the following paragraphs is made by using a titania electrode that consists of two layers: a rough nanocrystalline titania film making a light-scattering layer and a compact nanocrystalline titania layer filling the voids and standing on the top of the lower layer. As a matter of fact, the deposition of the second layer did not produce a separate stratum and did not increase the thickness of the film but it was accommodated within the voids and the pores of the first layer. The assembly made a LSE. The efficiency of this cell climbed to 6.9 %, a value that is more than 20 % higher than that of the corresponding cell employing a single transparent layer. This efficiency is very high, if we take into account the fact that it concerns a solid-state device. Another interesting feature of this cell is the employment of a silicabased nanocomposite organic-inorganic gel as host for the I3 / C O M M U N IC A TI O N