Performance Enhancement of Dye-Sensitized Solar Cells Using a Natural Sensitizer

Dye-sensitized solar cells (DSSCs) based on natural sensitizers have become a topic of significant research because of their urgency and importance in the energy conversion field and the following advantages: ease of fabrication, low-cost solar cell, and usage of nontoxic materials. In this study, the chlorophyll extracted from papaya leaves was used as a natural sensitizer. Dye molecules were adsorbed by TiO2 nanoparticle surfaces when submerged in the dye solution for 24 h. The concentration of the dye solution influences both the amount of dye loading and theDSSCperformance.The amount of adsorbed dyemolecules by TiO2 nanoparticle was calculated using a desorption method. As the concentration of dye solution was increased, the dye loading capacity and power conversion efficiency increased. Above 90mM dye solution concentration, however, the DSSC efficiency decreased because dye precipitated on the TiO2 nanostructure. These characteristics of DSSCs were analyzed under the irradiation of 100mW/cm . The best performance of DSSCs was obtained at 90mM dye solution, with the values of Voc, Jsc, FF, and efficiency of DSSCs being 0.561 V, 0.402mA/cm, 41.65%, and 0.094%, respectively.


Introduction
Performance of dye-sensitized solar cells (DSSCs) is influenced by semiconductors, electrolytes, transparent conductive oxide (TCO) substrates, counter electrode, and dye sensitizers [1][2][3][4][5][6]. Dye plays an important role in the performance of DSSCs [7][8][9]. When synthetic dye such as ruthenium complex is used as a sensitizer in a wide-bandgap semiconductor, the DSSC efficiency becomes over 10% [3]. Because of high costs, presence of heavy metals, and complex synthesis processes of synthetic dyes, natural dyes obtained from leaves, fruits, and plants are a cheaper option and they are nontoxic and completely biodegradable also [10].
Performance of DSSCs based natural sensitizer can be enhanced by using chlorophyll. When ethanol is used as a solvent, the efficiency of DSSCs improves to 36.11% compared to that achieved when distilled water is used [10]. The acidity of chlorophyll solution can be controlled to pH 3.5 by adding benzoic acid, which helped increase the efficiency of DSSCs from 0.07% to 0.28% [11]. The amount of dye molecules that can be absorbed by the semiconductor is also essential in improving the performance and the electron injection of DSSCs [12]. High electron injection from photoexcited sensitizers to the conduction band of semiconductors strongly influences the current density of DSSCs [1].
The amount of dye adsorbed by semiconductors can be studied by controlling the size of the semiconductor and modifying the morphological structure of the semiconductor [12][13][14][15][16]. The properties of the dye solution can also affect the amount of dye molecules adsorbed by semiconductors [17]. One of the solution's properties that can be easily controlled is the concentration of the solution. The dye solution concentration on the semiconductor is also important for study because of its influence in DSSC performance. In this work, the performance of DSSCs with natural chlorophyll as the sensitizer at various concentrations was studied. The performance of chlorophyll-based DSSCs was compared with the results obtained using N719-based DSSCs.

Material and Method
2.1. Chlorophyll Dye Synthesis. The chlorophyll dye extracted from 100 g of papaya leaves was added to 1 L of ethyl alcohol (96%, Merck, Germany) and heated at 70 ∘ C for 3 h. The separation of this solution was performed using Buchner funnel with single number 42 Whatman filter paper. The extracted dyes were then isolated from the ethanolic solution by rotary evaporation and got 8.5 g of crude chlorophyll. The crude chlorophyll was added to petroleum ether until a yellow color appears in the column chromatography. The separation process was followed by replacing the eluent with a solution of 10% diethyl ether in petroleum ether to produce 0.76 g (8.9%) chlorophyll (green), 0.035 g (0.41%) -carotene (yellow), 0.03 (0.35%) phycoerythrin (orange), and residues. A detailed description of the synthesis of chlorophyll dyes can be found elsewhere [11]. The chlorophyll was dissolved in ethyl alcohol to obtain solution of various concentrations (60, 70, 80, 90, and 100 mM). The synthetic dye used in this research was 0.03 mM N719 (Dyesol). The optical properties of the dyes were measured by UV-Vis spectroscopy (Lambda 25, Perkin Elmer). The functional groups in the dyes were characterized by Fourier transform infrared spectroscopy (FTIR, Shimadzu). Cyclic voltammetry (CV, Metrohm AG) was used to determine the reduction and oxidation potentials ( red and ox , resp.) of the dyes. The Pt wire, Pt plate, and Ag/AgCl were used as a counter, working, and reference electrodes, respectively. The potential applied in the CV process was from −2.0 V to +2.0 V with a scan rate of 100 mV/s.

DSSC Fabrication and
Testing. The DSSCs were assembled as follows: fluorine-doped tin oxide (FTO, Sigma-Aldrich) conductive glasses were used as the substrate. The semiconductor paste was prepared by dissolving 0.24 g TiO 2 nanopowder (21 nm, Sigma-Aldrich) into 4 mL ethyl alcohol (96%, Merck). The TiO 2 paste was applied to FTO-coated glass and flattened using the doctor blade method until the TiO 2 film becomes a homogenous layer. The semiconductor layer was maintained at 20 m and an area of 1 × 1 cm 2 . Then, the TiO 2 layer coated on the FTO substrate was sintered at 450 ∘ C for 2.5 h to enhance the bonding between the semiconductor and the FTO glass. After that, the photoelectrodes were immersed in the respective dye solution for 24 h at 30 ∘ C.
The counter electrode was conducted by a sputtering process. The catalytic platinum was deposited on the FTO glass in a vacuum tube at 9.5 × 10 −5 Torr. The platinum material target was connected to the negative terminal at a high voltage of 404 V and 125 mA, while the FTO glass was connected to the positive terminal. The electrolyte solution was synthesized by mixing 3.3 g sodium iodide (99.95%), 523.875 mg pure iodine (99.95%), 5.481 g heteropolyacid (HPA), and 30 mL acetonitrile. HPA added in the electrolyte act as an electron acceptor to prevent recombination and photoreduction of iodide (I − ) [18]. The electrolyte solution was injected into the assembled DSSCs at 30 m intervals and sealed by glass glue. The performance of DSSCs was determined by a solar

Dye Characterization.
The absorption spectra in the visible light spectrum (400-700 nm) of N719 and chlorophyll dyes are shown in Figure 1. The absorption peak of N719 can be seen at 515 nm and chlorophyll at 660 nm [11]. Variation in the chlorophyll dye solution concentration did not affect the position of the absorption peak spectra but affected the absorbance value at a wavelength of 660 nm. This corresponds with the Lambert-Beer law, according to which the absorbance of a solution is proportional to its concentration at the same path length of the light beam [19]. Figure 2 and Table 1 show the cyclic voltammetry (CV) test results of N719 and chlorophyll dyes. The electrochemical oxidation and reduction onset potentials ( ox and red , resp.) were used to calculate the energy levels of the highest occupied molecular orbital ( HOMO ) and the lowest unoccupied molecular orbital ( LUMO ) [11]. The onset potentials were determined from the intersection of tangents between the rising current and the baseline charging current of the CV curves. As shown in Table 1, the various concentrations have a minor effect on HOMO and LUMO values of chlorophyll dyes. Figure 3 shows the FTIR measurements of N719 and chlorophyll dyes. All dyes display peaks in both the 2500-3000 and 1600-1750 cm −1 region, corresponding to the presence of the -OH and C=O groups, respectively. However, the N719 dye produced more C=O stretching and -OH groups than chlorophyll dyes. In general, the concentration variation in chlorophyll dyes did not change the C=O stretching and -OH groups.    Figure 4 and Table 2 show the performance of DSSCs based on N719 and chlorophyll dyes under irradiation of 100 mW/cm 2 . The energy conversion efficiency of N719 dye-based DSSCs was higher than all chlorophyll dye-based DSSCs. The N719 dye-based DSSCs achieved an efficiency of 1.293%, while the highest efficiency of chlorophyll dye-based DSSCs was 0.094% at 90 mM.

Performance of DSSCs.
The amount of dye loading on the semiconductor surface was measured by the dye desorption method [12]. This process was followed by the immersion of the TiO 2 -dye electrode in an ethanol solution of 0.1 M NaOH. The dye desorption took place after 1 h, with the solution becoming pink for the N719 dye and green for the chlorophyll dye. At  the same time, the TiO 2 electrode turned colorless because it loses the dye on its surface [12,20]. Figure 5 shows the UV-Vis absorption spectra of the solutions measured to estimate the concentration of the adsorbed dye molecules. The concentration of the adsorbed dye was calculated by using the Lambert-Beer law: where is the intensity of the UV-Vis absorption spectra at the peak of N719 and chlorophyll dye in 515 and 660 nm, respectively, is the molar extinction coefficient of dye, is the dye molecular concentration, and is the path length of the light beam. The molar extinction coefficient of N719 determined to be 14,100 M −1 cm −1 at 515 nm [12,21] and for chlorophyll is 86,300 M −1 cm −1 at 660 nm [22].  Increasing the concentration of chlorophyll dyes increased the DSSC efficiency and was seen by the increase in the current density ( SC ). The rise in current density is associated with the amount of dye that can be adsorbed by the TiO 2 electrodes [1]. However, at concentrations above 90 mM, the dye loading value increased, while the efficiency of DSSCs decreased. This result can be explained by the precipitation of dyes on the TiO 2 electrodes. Because of this phenomenon, some dissolution of TiO 2 by the acidic carboxylic groups of the dye can occur. The resulting Ti + ions form insoluble complexes with chlorophyll dyes, causing precipitation of these complexes in the pores of the film. This gives rise to inactive dye molecules on the TiO 2 surfaces [20,23]. Based on this work, it has been found that the concentration of 90 mM chlorophyll dye has good interaction with the TiO 2 surfaces, due to the high dye loading value without producing the significant inactive dye precipitation effect.

Conclusions
Chlorophyll extracted from papaya leaves can be used as a dye sensitizer in DSSCs. However, the energy conversion efficiency is still lower than that of synthetic dye-based DSSCs. The N719 dye-based DSSCs achieved an efficiency of 1.293%, while the highest efficiency of chlorophyll dyebased DSSCs was 0.094% at 90 mM. The variation in the chlorophyll dye solution concentration did not change the HOMO-LUMO energy level and -COOH functionalities. Increasing the dye solution concentrations also increased the efficiency of DSSCs because of the increase in the amount of dye adsorbed by TiO 2 . However, at concentrations above 90 mM, the efficiency of DSSCs decreased because of the precipitation of dye on the TiO 2 nanostructure. This raises the number of inactive dye molecules on the surface of TiO 2 , preventing the electron injection processes in DSSCs.