Replacing the conventional TiO2
nanocrystalline (nc) electrode with organic/inorganic hybrid nanostructures will undoubtedly improve overall performance of dye-sensitized solar cells (DSSC). Inorganic nc-ZnO is a promising candidate for such hybrid structures, due to its unique properties, such as high conductivity, wide bandgap (3.2 eV), and high excitonic binding energy (60 mV) [1
]. In addition, the conduction band edge position of ZnO (-4.3 eV) is similar to that of TiO2
(-4.5 eV). Furthermore, ZnO can easily be electrochemically deposited (ECD) at low temperature, and hybrid nanostructures with ZnO can also be co-deposited with dyes such as Eosin-Y (EoY). It has been reported that the dye molecules are strongly attached to the ZnO matrix, filling the voids by means of sulfonic or carboxyl groups [2
]. In this article, we report on a study of different hybrids of ZnO nanostructures obtained by changing the dye in the deposition electrolyte. While different aspects of pure ZnO have been reported extensively by the other authors [3
], this article focuses on the growth of various ZnO porous structures with dyes and their use in energy conversion.
The ECD of nc-ZnO was carried out in a three-electrode cell consisting of a cathode [substrate, indium oxide (ITO)-coated glass with a ~10 Ω/□ sheet resistance], a Pt counter electrode, and an Ag/AgCl reference (+222 mV vs normal hydrogen electrode). Three ECD baths were prepared, each solution containing a mixture of 0.1 M KCl (Merck) and 5 × 10-3
(Merck). The first electrolyte bath contained 1 × 10-4
M of EoY (Sigma Aldrich, Spain), while the other baths had two different concentrations of tetrasulfonated copper phthalocyanines (TS-CuPc) (1 × 10-4
M and 3 × 10-5
M). The optimum EoY concentration for ZnO/EoY was derived during a previous study of EoY concentrations [5
]. The electrolytes were purged with O2
at a volume flow rate of 200 mL/min, while stirring by means of a magnetic bar stirrer to facilitate the oxygen diffusion. Before the deposition process, the ITO glass was ultrasonically cleaned in acetone and subsequently with ethanol for 15 min, and then rinsed with deionized water. The total deposition time for ZnO/Ts-CuPc (approx. 1 μm thick), was 600-800 s, while for ZnO/EoY films, it was 200-500 s. The potentiostatic deposition was carried out by applying a potential of -0.9 V to the substrate (1.17 cm2
), using an Autolab potentiostat. The bath temperature was set at 70°C controlled by a thermostat. After deposition, the ZnO/EoY films were immersed in a dilute aqueous NaOH solution (pH 10.5) for 40 min to desorb the loaded EoY molecules, while the ZnO/TS-CuPc films were immersed for 24 h. The films were dried in air for 1 h at 150°C. Desorbed thin films were re-sensitised with the relevant dye concentrations for characterization.
The morphology of the ZnO/organic hybrid films was studied using a JEOL-JSM6300 scanning electron microscope (SEM) operating at 10 kV. The structural characterization was carried out by high-resolution X-ray diffraction (XRD) using a Rigaku Ultima IV diffractometer in θ-2θ mode with a copper anticathode (CuKα, 1.54 Å). Optical transmittance measurements were performed by means of an Ocean Optics DT-MINI-2-GS deuterium-halogen lamp in association with a 500-mm spectrometer coupled to a backthinned CCD detector optimized for the UV-Vis range. To determine the surface topography, an atomic force microscope (AFM) Multimode Veeco was used, where the scanning was carried out in tapping mode using a silicon cantilever. The scanning frequency was set at 0.5 Hz, and the image size was 5 × 5 μm2. The photoelectrochemical study was performed in a conventional three-electrode arrangement in a glass cell, consisting of the deposited thin film as the working electrode, illuminated from the glass/ITO side, a Pt counter electrode, and an Ag/AgCl reference electrode in 0.1 M KCl electrolyte. The photocurrent was measured using a potentiostat/galvanostat and recorded. The illumination time of the electrode was controlled using an automatic mechanical shutter, with an adjusted illumination time of 10 s, for which a controller box had been designed. The shutter required approximately 10 ms to reach a completely open (or closed) position. All the measurements were performed at 0.05 V bias where the dark current was negligible.