Performance evaluation of titanium oxide deposited by electrophoresis in photoelectrodes of dye-sensitized solar cells

Nanoparticles of TiO2 have been the main semiconductor applied in Dye-sensitized solar cells (DSSCs). In this work, solar cells were developed from the electrophoretic deposition of anatase titanium dioxide (TiO2) films on conductive glasses. The electrophoretic deposition was perfomed with a constant voltage of 80V for 2 minutes and, after drying, the films were sintered at three different temperatures: 450 °C, 550 °C and 600 °C. In the assembly of the cells, the titanium dioxide films were sensitized by Ruthenizer 535-bisTBA dye (N719) and an electrolyte containing the iodide / tri-iodide redox pair and a commercial transparent platinum counter electrode were used. TiO2 films were characterized by scanning electron microscopy, dispersive energy spectroscopy and X-ray diffraction and photo-electrochemical techniques. The energy conversion rates were 0.2646% for the sintered film at 450 °C, 0.1209% for the sintered film at 550 °C and 0.1137% for the film sintered at 600 °C.


INTRODUCTION
The crystalline titanium dioxide (TiO 2 ), which is a semiconductor with energy gap in the ultraviolet region [1][2][3], owns excellent optical and electrical properties and to this day is vastly studied for numerous applications such as pigment [4], photocatalysis [5], gas sensor [6], sunscreen [7] and dye-sensitized solar cell (DSSC) [8] due to its unique optoelectronic features. Dye-sensitized solar cells are considered one of the most promising of the new generation of cells, due to the low cost of manufacture compared to silicon cells, and their development is in continuous progress [9]. These cells are conventionally build with a nanoporous layer of a semiconductor material of large gap, usually TiO 2 [7], ZnO [10,11] or SnO 2 [12], covered by a layer of photoexcited dye, generally the dyes with Ruthenium (II) poly pyridine complex. [13,14] The work electrode is connected to a counter electrode of catalytic material (Platinum, gold or activated coal) through an electrolyte with a iodide/triiodide redox pair (usually 3I -/I 3 -) which holds good stability and reversibility. [15,16] Highly porous layers of titanium dioxide can be produced by a number of techniques such as doctor blade [10], electrospinning [17], sputtering [18], micro-plasma oxidation [19] and electrophoretic deposition [8]. Because of the combinations of the high versatility of its use with different materials and costeffectiveness, the electrophoretic deposition (EPD) technique, with a wide range of novel applications in the processing of advanced ceramic materials and coatings, has recently gained increasing interest in academia and industrial sector. [20,21] In EPD, charged powder particles, dispersed or suspended in a liquid medium are attracted and deposited onto a conductive substrate of opposite charge under the application of a DC electric field. [22,21] In EPD, charged powder particles, dispersed or suspended in a liquid medium are attracted and deposited onto a conductive substrate of opposite charge under application of a DC electric field. [22,21] Given the above, the present work aims to evaluate the performance of solar cells sensitized by the dye N719, which use a TiO 2 film, deposited by the electrophoresis technique and treated with different sintering temperatures, as the photoanode. Measurement of the crystalline structure of the films by X-ray diffraction simultaneously with the calcination process. In addition: to characterize the semiconductor oxide films by means of Scanning Electron Microscopy (SEM); Dispersive Energy Spectroscopy (EDS) and current-voltage (J-V) density plots.

MATERIALS AND METHODS
All chemicals (reagent grade: Sigma-Aldrich, Dynamics and Vetec) were used as received, without further purification processes. All solutions were prepared with deionized water.

Preparation of the electrodes with TiO 2
The method for the preparation of the electrodes with commercial TiO 2 nanoparticles is known [8,[23][24][25]. Initially, a larger area of the FTO surface received the TiO 2 nanoparticles by the electrophoretic process. Following, from this area, it was limited an effective area, which, would be, afterwards, used on the fabrication of the studied devices.

Scanning Electron Microscopy (SEM) and Energy-dispersive x-ray spectroscopy (EDS)
The SEM and EDS measures were done by the scanning electron microscope (Inspect S50 -FEI) with an EDS detector connected. The measurement conditions were: working distance (WD) of 11.7 mm and accelerating voltage of 20 kV.

X-Ray Diffraction (XRD)
X-ray diffraction measures for phase identification were performed on a Bruker D8 Advance device, with CoKα1 (λ = 0.1789 nm) as the radiation source and using a voltage of 40kV, a current of 40mA and an angular variation of 20°-75°. A thermal chamber was attached to this equipment so that the measurements could be monitored simultaneously with the gradual increase in temperature.

Photoelectrochemical characterization of the solar cells.
For all the assembled DSSCs, electric measurements were made by two electrodes potentiostat using 100 mW/cm 2 light source of Led white-neutral. The electrical parameters and current-voltage (J SC -V OC ) density curves, in addition to the parasitic resistances, were obtained based on the circuit model proposed by VI-TALL et al. [26] in order to obtain the efficiencies of the cells. All data of the curves (J-V) are shown in Table 5.
All the measurements were performed to environment temperature on a dye-sensitized solar cell, using the N719 dye, supplied by Solaronix. The light source was a triple LED array driven by the output current of 400mA of the Autolab LED Driver. The output of the LED Driver is controlled by the DAC164 of the Autolab, directly from the software. All the measurements were carried out with the NOVA software. The cells were fixed to distance 13 cm of the light source.

Electrophoretic Deposition
All the films holding the nanostructures were deposited in a conductive glass (FTO-SOLEMS ® ) [27] with resistance varying between 50.0 -70.0 Ω/cm 2 . The substrates were previously cleaned using the following solvents: deionized water, acetone and isopropyl alcohol, in this sequence. The conductive substrates were kept immersed for 15 minutes on ultrasonic bath, for each solvent.
For the deposition of the TiO 2 film on the glass substrate with FTO, an electrophoresis source with a maximum voltage of 300 V and a maximum current of 700 mA of the Kasvi brand was used. A suspension was prepared containing 0.04 g of TiO 2 anatase from the Sigma-Aldrich® brand dispersed in a solution containing 10 ml of acetone and 30 ml of isopropyl alcohol. Then, a solution was prepared with 0.18 g of resublimated iodine and 50 ml of acetone. The suspension and solution were mixed and subjected to magnetic stirring and ultrasonic vibration for 20 minutes.
shows that the reaction of iodine with acetone has hydrogen ions as one of the products and, according to SOUZA et al. (2018) [15], the presence of H + ions increases the conductivity of the suspension.
The positive electrode used the conductive glass with FTO mentioned above and the counter electrode used a commercial platinum substrate with a distance of 3 cm between them. Electrophoretic deposition was carried out with a constant voltage of 80V for 2 minutes. Figure 1 shows the process illustration for the electrophoresis deposition. At the end of the deposition, drying was carried out at 100 °C. After this procedure, the TiO 2 films were sintered at three different temperatures: 450 °C, 550 °C and 600 °C, with a heating rise of 30 °C/ min. Therefore, the samples remained at the respective temperatures for a period of 30 minutes.
Several studies have shown that the sintering temperature for an optimal degree of crystallinity of the TiO2 film deposited on FTO is around 450 ° C. This optimal pattern of crystallinity is associated with greater efficiency of dye-sensitized solar cells. [28][29][30][31] The samples containing a TiO 2 anatase film sintered at different temperatures were called sample 450, sample 550 and sample 650 for the sintering temperatures of 450 °C, 550 °C and 600 °C respectively (Table  1).  Lastly, the deposited films by electrophoresis were used as photoanodes of dye-sensitized solar cells as detailed as follows.

Assemble of the dye sensitized solar cells
After thermal treatment, the samples with the TiO 2 nanostructures also were under ethanol solution (3x10 -4 mol L -1 ) of the "Ruthenizer 535-bisTBA" (N719) [32] dye for 24 hours. After, the films were washed in ethanol to extract the dye concentrated in the surface, due to an incomplete adsorption. After drying at room temperature, the electrodes were placed in touch through a commercial [32] redox pair electrolyte (iodide/triiodide) with an counter electrode [32] also commercial containing a thin platinum layer deposited on FTO. The commercial selant [32] was used to avoid the leaking of the electrolyte from the cell.

Morphological analysis with MEV/EDS
TiO 2 films were analyzed morphologically using the scanning electron microscopy technique.    The reduction in titanium percentages as the sintering temperature increases may indicate that the areas mapped in samples 550 and 600 have irregularities in the film. As a consequence,the thicknesses of the films in these regions are smaller, which makes it possible to identify elements of Sn and O present in the substrate.

X-ray diffraction analysis (XRD)
The XRD results for samples 450, 550 and 600 are shown in Figure 6. As the same materials were used for all samples, the peak intensities for the TiO 2 diffractogram are similar and observed in the crystallographic directions (101), (004), (200), (105), (211) and (204) which, according to HOSSAIN et al. [35], indicate the anatase phase. This measurement showed that the electrophoresis deposition method and the thermal treatment did not alter the crystalline structure of the material, maintaining its most active phase (anatase). This is due to the fact that the minimum temperature for the transition from anatase to rutile phase is between 700 °C -900 °C according to FAZLI et al. [36].
In order to avoid fusing the FTO substrate to the sample holder, it was necessary to deposit a TiO 2 film on a platinum substrate. Thus, in the XRD results ( Figure 6) peaks of tin oxide are not observed as it is normally found in films deposited on FTO. The platinum peaks identified with the (*) were obtained due to the substrate used. It is worth noting that only in the X-ray diffraction measurement a different substrate was used, in the other characterization techniques the substrate used was glass with FTO.     The size of the crystallites of the film increases slightly with the increase in the sintering temperature [37] causing a decrease in the surface area of the film [15].

Photovoltaic properties of cells
Lastly, all the photovoltaic devices showed in Figures 7, were assembled and characterized following the procedures on section 2.2.6 and 2.2.4, respectively.
The graphs of the current-voltage density plots are shown in Figures 8, 9 and 10. The photovoltaic parameters obtained for the three cells were compared and are summarized in Table 5.    The Cells efficiencies with samples 450, 550 and 650 were calculated according to Equation 3. The parameters R S and R SH correspond to the resistances in series and in parallel, respectively, assigned to the circuit that models the photovoltaic cells. ( Where, Voc is the open circuit voltage, Jsc is short-circuit current, it is the fill factor and is the radiation power incident on the cell. The short circuit current density (J SC ) decreased with the increase in the sintering temperature of the film and the highest value was obtained by the cell that had the highest efficiency. The amount of electrons generated by the dye determines the current generated. Therefore, the greater the amount of dye adsorbed on the film, the greater the generation of electrons and consequently the current density will, also, be greater. According to NONO [38], the open circuit potential (V OC ) depends on the ratio between the amount of electrons injected into the semiconductor and the amount of electrons that undergo recombination on the electrode or electrolyte surface. Table 5 shows that for the highest sintering temperature, the lowest V OC value was obtained, which indicates that in this cell a greater load recombination may have occurred, resulting in a lower efficiency.
The fill factor (FF) decreases as the sintering temperature increases. One of the main factors that influence the value of the FF is the series resistance (R S ) inside the cell that will directly affect the electron transfer process. SOUZA et al. [15], SEQUEIRA [39] and NUNES [40], also observed in their experiments that solar cells that had higher Rs values, had lower FF values.
The energy conversion efficiencies obtained for the cells were 0.2664% for the cell containing sample 450, 0.1209% for the cell containing sample 550 and 0.1137% for the cell containing sample 600.
According to CHANG et al. [41], the efficiency of the cell is also affected by the uniformity of the film thickness, and the non-uniformity results in reduced efficiencies due to recombination of charges.
Other studies have also reported dye-sensitized solar cells with efficiencies of less than 1% such as that made by CHIANG, LEE and HSU [42] and AZIZI et al. [43]. CHIANG, LEE and HSU [42] assembled cells with titanium dioxide deposited by electrophoresis and doctor blade on ITO substrates and obtained efficiencies of 0.0473% and 0.1327%, respectively. AZIZI et al. [43] deposited a TiO 2 film by electrophoresis on an FTO substrate, which underwent heat treatment at 450 ° C. In assembling the cells, different natural dyes were used and the efficiencies obtained varied between 0.007% and 0.09%.
The low efficiency found in this work was caused by high values of resistance in series (Rs) with reduction in the conductivity of the cell. This problem can be reduced with the chemical treatment of the films with TiCl 4 . The electrical conductivity has been improved at the interfaces between particles in the TiO 2 coating and between conducting glass and the TiO 2 film through TiCl 4 treatment. [44][45][46][47][48][49][50]

CONCLUSIONS
The electrophoresis dispersion and deposition process proved to be a promising technique in the construction of porous films for application in DSSCs. The XRD analyzes of all samples indicated similarity in the crystallographic directions that correspond to the anatase phase of TiO 2 , confirming that the electrophoresis deposition method and the heat treatment did not alter the crystalline structure of the material. It was verified through the analysis of XRD data the influence of the sintering temperature on the size of the crystallites, which suggests that the size of the crystallites increases slightly with the increase of the sintering temperature. The result of this reduction in the surface area of the film may cause a lesser injection of electrons for the semiconductor oxide, due to the lower amount of dye adsorbed on the surface of this oxide.
The film sintered at 450 °C showed smaller crystallites, which contributed to increasing the efficiency of this cell.
SEM analysis revealed that all films had crystalline surfaces with the presence of pores. The film sintered at 450 °C showed greater porosity compared to the other two. The EDS spectra showed that all films contained only titanium and oxygen, confirming that the electrophoresis deposition technique was efficient.
There was a reduction in the mass proportions of titanium and an increase in the proportions of tin as the sintering temperature increased, indicating a possible variation in the thickness of the film.
From the parameters obtained from the current-voltage plots it was possible to calculate the efficiency of the cells. The cell with TiO 2 film sintered at 450 °C presented the highest efficiency value, 0.2664%. This greater efficiency is due to the fact that the film had larger surface area, which results in greater dye adsorption and greater electron generation. In order to optimize the photovoltaic capacity of solar cells sensitized by dyes with TiO 2 films deposited by electrophoresis, it is suggested to carry out analyzes