Abstract
Highly ordered mesoporous TiO2 modified by CdS, CdSe, and ZnS quantum dots (QDs) was fabricated by successive ionic layer adsorption and reaction (SILAR) method. The quantity of material deposition seems to be affected not only by the employed deposition method but also and mainly by the nature of the underlying layer. The CdS, CdSe, and ZnS QDs modification expands the photoresponse range of mesoporous TiO2 from ultraviolet region to visible range, as confirmed by UV-Vis spectrum. Optimized anode electrodes led to solar cells producing high current densities. Pt, CuS, PbS, and Cu2S have been used as electrocatalysts on counter electrodes. The maximum solar conversion efficiency reached in this work was 1.52% and was obtained by using Pt electrocatalyst. CuS, PbS, and Cu2S gave high currents and this was in line with the low charge transfer resistances recorded in their case.
1. Introduction
As an alternative to dye molecules, semiconductor quantum dots (QDs) like CdS, CdSe [1], PbS [2], InAs [3], InP [4], and others [5] as well as extremely thin inorganic absorber layers [6, 7] have been used. QDs are very attractive because of their size-dependent optical band gap, the possibility to design hierarchical multilayer absorber structures, and the potential to use them for multiexciton generation from a single photon. One potential method for improving the performance of quantum dots solar cells (QDSSCs) is by constructing desired energy band structures using multiple QDs. Niitsoo and coworkers have firstly demonstrated that a desired cascade structure can be formed by sequential deposition of CdS and CdSe layers onto the TiO2 nanoparticle films [8]. Recently, Lee et al. have also reported a self-assembled TiO2/CdS/CdSe structure that exhibited a significant enhancement in the photocurrent response [9, 10]. In addition, nanostructured CuS, PbS, Cu2S, and Pt have been used as electrocatalysts on the counter electrode. Alternative catalysts have been proposed by several researchers [9–12]. Metal sulfides are considered a good choice. However, their deposition on plain FTO electrodes does not always produce materials with sufficiently high specific surface or with structural stability.
In this paper, we studied the effects of comodification by CdS, CdSe, and ZnS QDs on the photovoltaic response of mesoporous TiO2 based QDSSC. The mesoporous TiO2 were treated by SILAR of CdS, CdSe, and ZnS QDs and were used as photoanodes in QDSSC. We demonstrated that the comodified mesoporous TiO2 possesses superior photovoltaic response compared to the single QD sensitized devices. Pt, CuS, PbS, and Cu2S have been used as electrocatalysts on counter electrodes. The final TiO2/CdS/CdSe/ZnS photoanode leads to high efficiency QDSSCs.
2. Experiment
2.1. Materials
Cd(CH3COO)2·2H2O (99%), Cu(NO3)2, Na2S, Zn(NO3)2, Se powder, S powder, Na2SO3, and Brass foil were obtained from Merck. TiO2 paste was obtained from Dyesol, Australia.
2.2. To Prepare TiO2 Films
The TiO2 thin films were fabricated by silk-screen printing with commercial TiO2 paste. Their sizes ranged from 10 to 20 nm. Two layers of film with thickness of 8 μm were measured by microscope. Then, the TiO2 film was heated at 400°C for 5 min and 500°C for 30 min. Afterwards, the film was dipped in 40 mmol TiCl4 solution for 30 min at 70°C and heated at 500°C for 30 min.
The specific surface area of the mesoporous TiO2 was investigated by using the N2 adsorption and desorption isotherms before and after the calcination. The surface area is 120.6 m2 g−1 (measured by BET devices). This result indicates that the synthesized material has wider mesoporous structure.
2.3. To Prepare TiO2/CdS/CdSe/ZnS Films
The highly ordered TiO2 was sequentially sensitized with CdS, CdSe, and ZnS QDs by SILAR method. First, the TiO2 film was dipped in 0.5 mol/L Cd(CH3COO)2-ethanol solution for 5 min, rinsed with ethanol, dipped for 5 min in 0.5 mol/L Na2S-methanol solution, and then rinsed with methanol. The two-step dipping procedure corresponded to one SILAR cycle and the incorporated amount of CdS QDs was increased by repeating the assembly cycles for a total of three cycles. For the subsequent SILAR process of CdSe QDs, aqueous Se solution was prepared by mixing Se powder and Na2SO3 in 50 mL pure water after adding 1 mol/L NaOH at 70°C for 7 h. The TiO2/CdS samples were dipped into 0.5 mol/L Cd(CH3COO)2-ethanol solution for 5 min at room temperature, rinsed with ethanol, dipped in aqueous Se solution for 5 min at 50°C, and rinsed with pure water. The two-step dipping procedure corresponds to one SILAR cycle. Repeating the SILAR cycle increases the amount of CdSe QDs (a total of four cycles). The SILAR method was also used to deposit the ZnS passivation layer. The TiO2/CdS/CdSe samples were coated with ZnS by alternately dipping the samples in 0.1 mol/L Zn(NO3)2 and 0.1 mol/L Na2S-solutions for 5 min/dip, rinsing with pure water between dips (a total of two cycles). Finally, it was heated in a vacuum environment with different temperatures to avoid oxidation (see Figure 1). The thickness of TiO2/CdS/CdSe/ZnS photoanodes were measured by Microscopic. The results of the average thickness of every layer of CdS, CdSe, and ZnS are 40 nm, 43.3 nm, 40 nm, respectively.
2.4. Construction of the Counter Electrodes
PbS films were deposited on fluorine doped tin oxide (FTO) conductive glass electrode by cyclic voltammetry (CV) from the solution of Pb(NO3)2 1.5 mM and Na2S2O3 1.5 mM. CV experiments were carried out at various potential scan rates in a potential range 0.0 to –1.0 V versus Ag/AgCl/KCl electrode, pH from 2.4 to 2.7, and ambient temperature. Pt films were fabricated by silk-screen printing with commercial Pt paste. Then, the Pt films were heated at 450°C for 30 min. CuS was also deposited on FTO electrodes by a SILAR procedure, by modifying the method presented in [13]. Precursor solutions contained 0.5 mol/L Cu(NO3)2 in methanol and 1 mol/dm3 Na2S·9H2O in a 1 : 1 water : methanol mixture. An FTO electrode was immersed for 5 min in the metal salt solution, copiously washed with triple-distilled water, dried in an air stream, immersed for 5 min in the Na2S·9H2O solution, and finally washed and dried again. This sequence again corresponds to one SILAR cycle. 10 SILAR cycles were performed. Finally, the electrode with deposited CuS film was first dried and then it was put for 5 min in an oven at 100°C. The counter electrode was a Cu2S film fabricated on brass foil. Brass foil was immersed into 37% HCl at 70°C for 5 min, rinsed with water, and dried in air. After that, the etched brass foil was dipped into 1 mol/L S and 1 mol/L Na2S aqueous solution, resulting in a black Cu2S layer forming on the foil [14].
2.5. Fabrication of QDSSCs
The polysulfide electrolyte used in this work was prepared freshly by dissolving 0.5 M Na2S, 0.2 M S, and 0.2 M KCl in Milli-Q ultrapure water/methanol (7 : 3 by volume). The CdS/CdSe/ZnS cosensitized TiO2 photoanode and Pt counter electrode were assembled into a sandwich cell by heating with a Surlyn. The electrolyte was filled from a hole made on the counter electrode, which was later sealed by thermal adhesive film and a cover glass. The active area of QDSSC was 0.38 cm2.
2.6. Characterizations and Measurements
The morphology of the prepared samples was observed using field-emission scanning electron microscopy (FE-SEM, S4800). The crystal structure was analyzed with an X-ray diffractometer (Philips, PANalytical X’pert, CuKα radiation). The absorption properties of the nanotube array samples were investigated using a diffuse reflectance UV-Vis spectrometer (JASCO V-670). Photocurrent voltage measurements were performed on a Keithley 2400 sourceMeter using a simulated AM 1.5 sunlight with an output power of 100 mW/cm2 produced by a solar simulator (Solarena, Sweden).
3. Results and Discussion
Shown in Figures 2(a) and 2(b) are the FESEM images of TiO2/CdS/CdSe/ZnS photoanode film. Figure 2(a) shows highly uniform porous morphology with the average inner diameter of nanostructure around 60 nm. For photovoltaic applications, the structure of QDs adsorbed TiO2 should meet at less two criteria. First, the QDs should be uniformly deposited onto the TiO2 surface without aggregation, so that the area of TiO2/QDs can be maximized. Second, a moderate amount the QDs should be deposited so that TiO2 is not blocked.
Figure 2(c) is a cross-sectional image showing that the QDs are well deposited onto the TiO2 with an average thickness of about 12 μm by the microscope. Figure 2(b) is the energy dispersive X ray spectroscopy of the TiO2/CdS/CdSe/ZnS film. It shows that the Ti and O peaks are from the TiO2 film; and Cd, Se, Zn, and S peaks, clearly visible in the EDS spectrum, are from the QDs. The Si is from the FTO and C is from the solvent organic. That shows that the QDs are well deposited onto the TiO2.
The structure of the TiO2/QDs photoelectrodes for photovoltaic applications, shown in Figure 3(a), is studied by the XRD patterns. It reveals that the TiO2 has an anatase structure with a strong (101) peak located at 25.4°, which indicates that the TiO2 films are well crystallized and grow along the (101) direction (JCPDS Card number 21-1272). Three peaks can be observed at 26.4°, 44°, and 51.6°, which can be indexed to (111), (220), and (331) of cubic CdS (JCPDS Card number 41-1049) and CdSe (JCPDS Card number 75-5681), respectively. Two peaks can be observed at 48° and 54.6°, which can be indexed to (220) and (331) of cubic ZnS, respectively. It demonstrates that the QDs have been crystallized onto the TiO2 film. Figure 3(b) is the Raman spectrum of the TiO2/QDs photoelectrodes. It shows that an anatase structure of the TiO2 films has five oscillation modes corresponding wave numbers at 143, 201, 395, 515, and 636 cm−1. In addition, two peaks can be observed at 201, 395, and 515 cm−1, which can be indexed to the cubic structure of CdS and CdSe. The results of the Raman are likely similar to the results of the XRD. The optical performance of the QDs coated TiO2 film is characterized by absorbance. Figure 3(c) shows the UV-Vis absorption spectra of the sensitized electrodes measured after each cycle of SILAR. As expected, the absorbance increased with the deposition cycles of CdS, CdSe, and ZnS. However, only absorption spectra with SILAR cycles of the electrode TiO2/CdS(3)/CdSe(3)/ZnS(2) show the best photovoltaic performance as discussed in the following section. In short-wavelength region (380–550 nm), the increase of absorbance is due to the fact that more CdS was loaded on TiO2 film and the coabsorption of CdS, CdSe, and ZnS. In long-wavelength region (550–629 nm), the deposition of higher amounts of CdSe and ZnS on TiO2/CdS electrode results in the increase of absorbance. Moreover, the increasing successive deposition cycles also trigger a red shift of absorption spectrum which is due to a slight loss of quantum confinement effect [15]. The evaluated sizes of CdS and CdSe are consistent with the sizes measured from the FE-SEM images. The effect of deposition cycles of CdS, CdSe, and ZnS can be clearly seen on the energy band gap values of CdS/CdSe/ZnS cosensitized TiO2 films. The estimated band gaps vary from 1.97 eV to 2.7 eV, which are higher than the values reported for CdS and CdSe in bulk (2.25 eV and 1.7 eV resp. [16]), indicating that the sizes of CdS, CdSe, and ZnS on TiO2 films are still within the scale of QDs. The diameter of QDs was calculated from 2 nm to 6 nm by (1). A higher absorption is thus obtained because the absorption spectrum of ZnS complements those of the CdSe and CdS QDs. Furthermore, ZnS acts as a passivation layer to protect the CdS and CdSe QDs from photocorrosion [17]. Consider the following equation by Yu et al. [18] group: The XRD patterns were used to characterize the crystal structure of the obtained products. As shown in Figure 4(a), it can be seen that the XRD pattern of the PbS counter electrode is in conformity with cubic ( Å). The observed peaks could be assigned to diffraction from the (111), (200), (220), (311), and (222) faces and there is no characteristic peak for other impurities. This indicates that pure crystalline PbS was formed via the cyclic voltammetry process. Figure 4(b) illustrates the XRD pattern of the synthesized Cu2S after 1 h by chemical bath deposition (CBD) method. The peaks of corresponding crystal planes were indexed in the figure, matching to the hexagonal phase chalcocite β-Cu2S (JCPDS card number 46-1195, Å, Å). Figure 4(c) illustrates that the XRD pattern of the Pt films were fabricated by silk-screen printing with commercial Pt paste. The peaks of corresponding crystal planes were indexed in the figure, matching to the hexagonal phase. As shown in Figure 4(d), it can be seen that the XRD pattern of the CuS counter electrode is in conformity with the hexagonal phase. It is in agreement with the reported data for CuS (JCPDS Card. number 79-2321).
(a)
(b)
(c)
(a)
(b)
(c)
(d)
A relative energy level of different components is shown in Figure 5(a). According to the data reported in the literature [16, 19], the band gap of TiO2 (3.2 eV) limits its absorption range below the wavelength of about 400 nm. CdSe has a higher conduction band (CB) edge than TiO2, which is favorable for electron injection. However, with a band gap of 1.7 eV, the absorption of bulk CdSe is also limited below approximately 760 nm. The conduction band of CdSe is slightly lower than that of TiO2, so the electrons would flow from CdSe to TiO2 [20]. In addition, we have coated two layers of ZnS QDs, which could be attributed to several reasons. First, as the absorption edge of ZnS is at about 345 nm, a higher absorption can be obtained due to the complement of the absorption spectrum of the ZnS with that of the CdSe and CdS QDs. Second, ZnS acts as a passivation layer to protect the CdS and CdSe QDs from photocorrosion. Thus, the photoexcited electrons can efficiently transfer into the conduction band of TiO2. Third, the outer ZnS layer can also be considered to be a potential barrier between the interface of QDs materials and the electrolyte. ZnS has a very wide band gap of 3.6 eV; it is much wider than that on the CdS and CdSe QDs. As a result, the leakage of electrons from the ZnS, CdSe, and CdS QDs into the electrolyte can be inhibited. As a result, an ideal model for the cosensitized TiO2 electrode is shown in Figure 5(b). After CdSe and ZnS QDs are sequentially deposited onto a TiO2/CdS film, A cascade type energy band structure is constructed for the cosensitized photoanode. The best electron transport path is from the conduction band of ZnS and finally reaching the conduction band of TiO2. Meanwhile, this stepwise structure is also favorable for the hole transport.
(a)
(b)
We prepared the photoanodes with many different layers of QDs. Firstly, we have prepared CdS or CdSe films. However, the results were of very low performance. Therefore, we decided covering with ZnS layer for the following reasons. Firstly, extend peak adsorption spectrum in the visible light region. Secondly, the ZnS layers which acted as the agent passivated the surface of QDs. Moreover, they protected the light corrosion. Thus, the conversion excited electrons through the conduction band of TiO2 better. Thirdly, ZnS layers separated the surfaces of the CdS and CdSe with electrolyte. The ZnS has a very wide band gap of about 3.6 eV, much larger than other CdS and CdSe QDs. As a result, electrons move from CdS, CdSe, and ZnS to the electrolyte can be inhibited. Figure 6(a) shows that the power conversion efficiencies of QDSSCs are increasing with the SILAR cycle number of CdS, CdSe, and ZnS at 3, 3, and 2, respectively. It is noted that lower power conversion efficiency was obtained for those cells with either less than 3 CdS and CdSe SILAR cycles or more than 3 CdS and CdSe SILAR cycles (Figure 6(b)). The TiO2/CdS(3)/CdSe(3)/ZnS device shows an open-circuit voltage () of 0.76 V, a short-circuit current density () of 4.79 mA/cm2, fill factor (FF) of 0.41, and an energy conversion efficiency of 1.52%. When the deposition cycles of CdS and CdSe increase, slight changes in and FF values were obtained. Remarkably, the decreases, which results in a substantial reduction of efficiency from 1.52% to 0.45% (Table 1). These results indicate that although better light absorption performance was obtained when more CdSe was loaded on TiO2/CdS, the excessive CdSe on TiO2/CdS films may lead to an increase of recombination in photoanodes. On the contrary, the increase of ZnS leads to the increasing generation of photoelectron and is helpful to collect excited electrons from ZnS, CdSe, and CdS to TiO2 film.
(a)
(b)
FF is determined from measurement of the IV curve and is defined as . FF depending on values, the junction quality (related with the series ), and the type of recombination in a solar cell. From Table 1, values change according to the film thickness from 0.29 to 0.76, corresponding to the change in FF from 0.26 to 0.41. Therefore, the FF is the low value because is low. On the other hand, values depend on the recombination process; they are particularly large; gives low open-circuit voltages. In addition, FF is effected by . The equations of can be calculated by Thongpron and coworkers [21] as follows: Two operating points are (, ) and (, ) on a single - curve. ; , are the photocurrent and the diode reverse saturation current. values are calculated from 55 to 158 mΩcm2−. This result indicate that fill factor will decrease when VOC increase.
Four main types of counter electrodes have been studied. Their synthesis is detailed in experiment and method. The FESEM images of the corresponding electrocatalytic films are shown in Figure 7 (inset). In the first case, PbS films were deposited on fluorine doped tin oxide (FTO) conductive glass electrode by cyclic voltammetry (CV) from the solution of Pb(NO3)2 1.5 mM and Na2S2O3 1.5 mM. CV experiments were carried out at various potential scan rates in a potential range 0.0 to −1.0 V versus Ag/AgCl/KCl electrode, pH from 2.40 to 2.70, and ambient temperature. CuS was also deposited on FTO electrodes by a SILAR procedure, by modifying the method presented in [13]. The electrode with deposited CuS film was first dried and then it was put for 5 min in an oven at 100°C. The counter electrode was a Cu2S film fabricated on brass foil. Brass foil was immersed into 37% HCl at 70°C for 5 min and then rinsed with water and dried in air. After that, the etched brass foil was dipped into 1 mol/L S and 1 mol/L Na2S aqueous solution, resulting in a black Cu2S layer forming on the foil [14]. Figures 4(a), 4(b), and 4(c) show the image of PbS, CuS, Cu2S films that present a rough nanostructure, which are suitable for counter electrodes. The similar Pt films were fabricated by silk-screen printing with commercial Pt paste. Then, the Pt films were heated at 450°C for 30 min. In the high magnification image of Figure 4(d), one can distinguish the big blocks of FTO covered with Pt nanoparticles [22]; Figure 4 and Table 2 show that the maximum efficiency reached in the present work, that is, 1.52%, was obtained with Pt on the counter electrode. The Pt electrocatalysts, that is, Cu2S, CuS, and PbS, gave higher current densities than Pt but lower than Pt. On the contrary, open-circuit voltage values were practically not affected by the electrocatalyst. The major problem encountered in the present work was with the value of the fill factor (FF). It remained below 0.42 and this limited the overall efficiency, even though, the current densities presently recorded were high. The search for a higher FF is an open question and has occupied many other researchers. It is believed that higher FFs will be obtained with even better electrocatalysts and more functional counter electrodes.
(a)
(b)
(c)
(d)
4. Conclusions
QDSSCs have been constructed and optimized by combining TiO2 with CdS, CdSe, and ZnS nanostructure on the anode electrode. PbS, CuS, Cu2S, and Pt were used as electrocatalysts on counter electrodes in combination with a polysulfide electrolyte. The maximum solar conversion efficiency of 1.52% was obtained with a Pt counter electrode. The most important finding of this work is the importance of the first nanostructure layer deposited on the mesoporous TiO2 film, which affected the quantity and the quality of the subsequent QDs layers and the ensuing cell efficiency. High current densities were obtained with all cells having optimized anode electrodes. Among them, the highest currents were obtained with Pt electrocatalysts.
Conflict of Interests
The authors declare that there is no conflict of interests regarding the publication of this paper.
Acknowledgments
This work was supported by Vietnam National University with the name of the project being B 2012-18-05TD, the University of Science of Ho Chi Minh City, and Dong Thap University.