Effect of Multilayers CdS Nanocrystalline Thin Films on the Performance of Dye-Sensitized Solar Cells

,


Introduction
A few years ago, precisely in 2015, the researchers began to take an interest towards alternative materials to become the photoanode instead of ZnO and TiO 2 used in dye-sensitized solar cells (DSSCs) [1][2][3]. Te highest efciency was obtained from inexpensive materials, porous CdS as a photoelectrode only, and ease of manufacture of the DSSC device at about 1.25% [4]. In many studies CdS was deposited onto the surface of ZnO or TiO 2 as a second layer to enhance the efciency of DSSC [5][6][7][8][9][10]. DSSC based on nanostructures cadmium sulfde (CdS) as a photoanode (window layer) attracts a lot of attention as a result of various properties, including its ease of preparation methods [11][12][13][14], direct bandgap [15], good thermal stability [16], high electronic mobility [17], and piezoelectric properties [18]. Many researchers were produced CdS with diferent nanostructures such as porous [19], nanowire [20], nanorods [21], nanofower [22], and nanoribbons [23]. In addition, cadmium sulfde is a widely studied material for several applications such as solar cells [24], photocatalysis [25], and gas sensors [26]. However, DSSC is one of the most promising solutions in the feld of renewable energy instead of fossil fuels. In this study, for the frst time, porous and nanowall CdS thin flms have been used as a photoanode combination with methylene blue dye (active layer) to fabricate a DSSC device.

Synthesis of the DSSC Device
Te assembly of a DSSC was divided into three stages as follows: First, cadmium nitrate (4.3 mM) was dissolved in deionized water D.W (100 ml) as a source of Cd +2 . Ten, ammonium acetate (0.015 M) was added to the solution to control the synthesis process, and 1.5 ml) of ammonia solution was added to the aqueous solution to raise the pH to 11. Ten, thiourea (5 mM) to the mixture was added as a source of S −2 under vigorous stirrer at room temperature for 15 minutes. In the next step, a clean FTO/glass substrate was immersed vertically in the mixture and the temperature solution was raised to 65°C for 22 minutes. Porous and porous-nanowall CdS thin flms were formed into a substrate FTO/glass as a photoelectrode. Tis process is repeated three times to obtain three samples with diferent thicknesses: 1-layer CdS, 2 layers CdS, and 3 layers CdS. All samples were immersed in (0.003 g dissolved in 100 ml of D.W) methylene blue (MB) sensitizer for 24 hours.
Te second step was to prepare the electrolyte; 0.15 g of iodine was dissolved in 15 ml of ethylene glycol under stirring. Ten, 1 g of potassium iodine was added to increase  the conductivity. Ten, the counter electrode was prepared; 60 mg and 30 mg of carbon black and graphite powder were dissolved in 10 ml of ethylene glycol, respectively. Ten, 1 ml of the mixture was cast to the FTO/glass and then annealed at 300°C for 30 minutes. Te fnal step determines the area of an active layer of DSSC by using adhesive tape (0.28 cm 2 ). Figure 1 shows the whole steps of fabricated DSSC.

Results and Discussion
3.1. Morphology. Te morphologies of nanocrystalline CdS thin flms were investigated by the FE-SEM machine. Te most important observation that was diagnosed is the transfer of a surface shape of CdS from thin flm to porous and then porous nanowall arrays in the third layers, as shown in Figure 2. Table 1 shows that the thickness values of porous and porous nanowall-like CdS were calculated using the standard deviation formula (SD) as follows [27]: where samples are x 1 , x 2 , . . ., x n and n is the sample size. It is discernible that the standard deviation of the thickness of porous and porous-nanowall CdS thin flms decreased with increased layers. Figure 3 shows the adsorption of MB dye at the surface of samples in a unique and homogenous way to capture more energy from sunlight. Figure 4 shows the XRD patterns of 1 layer, 2 layers, and 3 layers of nanocrystalline CdS thin flms. Here, the FTO/glass patterns have been marked as (■). In 1layer crystalline plane (111) cubic/(002) hexagonal mixed phase of CdS observed at 2θ∼26.6°after deposition of 2 layer of CdS, whereas CdS(200)C and CdS(220)C/CdS(110)H, CdS(311)C appears corresponding to 2θ∼30°, 44°, and 52°, respectively. Te small peak of crystalline Sulfur dominated by the (511) plane was observed at 2θ∼33°. All peaks of various thicknesses were matched with a standard data card of CdS (JCPDS card no. 89-0019) [28]. It is difcult to distinguish between cubic and hexagonal phases, but disappearing planes (100) at 2θ∼24.9°and (101) at 2θ∼28.4°tend to the CdS cubic structure [29,30]. Te plane (111) difraction peak value shifts slightly towards the higher angle with increasing thickness.

Optical Properties.
Te thicknesses of the prepared samples were about 391 nm, 457 nm, and 912 nm for 1, 2, and 3 layers of CdS nanocrystalline thin flm. Figure 5(a) shows the absorption was increasing with CdS thickness because of an increase in crystallinity and the amount of material deposited on the FTO/glass substrate [31]. All the samples are blue-shifted when increasing the thickness of CdS which indicates the formation of particles in the Journal of Nanotechnology nanoscale regime. Te characteristic wavelength of absorbance exhibited by the MB dye corresponds to λ max � 668 nm (π − π * ). Tis result confrms that the sensitizer was adsorbed at the surface of the photoanode CdS, as shown in Figure 5(b). In addition, the absorption coefcient (α) of the prepared samples shows higher values greater than 10 4 . Tis indicates there is a large probability of the allowed direct transition, as shown in Figure 6(a). Te optical bandgap has been measured by Tauc's relation αhυ � A(hυ − Eg) m [32,33]. It has been observed to increase Eg (2.45, 2.37, and 2.32 eV) with a decrease in the thickness of flms [34] ( Figure 6(b)). Figure 7 illustrates the performance of the multilayer CdS nanocrystalline thin flms that have been studied. Te parameters fll factor (FF) and efciency (ἠ) are used to characterize solar cells [35,36].        (2)

J-V Characteristics.
Te performance of various thicknesses of CdS without a MB sensitizer has been studied, and the short circuit current density J SC increases with increasing the thickness, reaching 0.462 mA/cm 2 with a power conversion efciency of 0.018% for FTO/3 layer CdS/electrolyte/carbon/FTO device, as shown in Figure 7(a). Gadalla et al. [37] studied the infuence of increasing the thickness of CdS thin flms on the output photocurrent, and they found that an increment in the photocurrent with increased thickness. Other researchers found the same results by increasing the thickness which agrees with [38][39][40].
One-layer CdS/MB device shows the lowest J SC (0.35 mA/ cm 2 ), open circuit voltage V OC (270 mV), and the output effciency η (0.02%), respectively. When increasing the layers to three of CdS nanocrystalline thin flms, it has been found enhancement of J SC, V OC , and η reach the highest value due to increased harvesting of adsorption between CdS nanocrystalline thin flm and methylene blue dye (2.27 mA/cm 2 ), (594 mV), and (0.47%), respectively. It can be seen the efect of multilayers and MB sensitizer on the performance of solar cells, and the increment is about 96% (η) with the dye of a 3 layers device. Table 2 shows all the parameters of DSSC devices.

Conclusion
Nanostructures of cadmium sulfde were prepared via the chemical bath deposition method, yielding porous and wall morphologies. CdS was used as a window layer, and how complementary its aspects are to improving the layers of DSSC. As revealed by SEM images, as-deposited CdS flms possess porous surface morphology. It is difcult to distinguish between cubic and hexagonal phases, but disappearing planes prove the CdS cubic structure. By Tauc's relationship optical bandgap has been measured, and it has been observed to increase with a decrease in the thickness of flms. Te highest efciency has been recorded at about 0.47% of 3 layers porous CdS/MB device.

Data Availability
Te data that support the fndings of this study are available on request from the corresponding author. Te data are not publicly available due to privacy or ethical restrictions.