Abstract
The hydrogenated TiO2 porous nanocrystalline film is modified with CdS quantum dots by successive ionic layer adsorption and reaction (SILAR) method to prepare the cosensitized TiO2 solar cells by CdS quantum dots and hydrogenation. The structure and topography of the composite photoanode film were confirmed by X-ray diffraction and scanning electron microscopy. With deposited CdS nanoparticles, UV absorption spectra of H:TiO2 photoanode film indicated a considerably enhanced absorption in the visible region. The cosensitized TiO2 solar cell by CdS quantum dots and hydrogenation presents much better photovoltaic properties than either CdS sensitized TiO2 solar cells or hydrogenated TiO2 solar cells, which displays enhanced photovoltaic performance with power conversion efficiency (η) of 1.99% ( mA cm−2, V, and FF = 0.49) under full one-sun illumination. The reason for the enhanced photovoltaic performance of the novel cosensitized solar cell is primarily explained by studying the Nyquist spectrums, IPCE spectra, dark current, and photovoltaic performances.
1. Introduction
Photoelectrochemical solar cells based on TiO2 nanocrystalline films sensitized with organic dyes have been studied intensely for the past 20 years as a potential promising low-cost alternative to traditional solid-state solar cells. Semiconductor nanocrystals with narrow band gap such as CdS, CdSe, PbS, and InP were also demonstrated as efficient sensitizers in the spectral range from the visible to mid infrared with advantages such as low-cost fabrication, good stability, multiple exciton generation, and the tunability of optical properties and electronic structure by changing the size of nanocrystals [1, 2]. The photoelectrochemical cells with semiconductor nanocrystals as sensitizers also were called quantum dot sensitized solar cells (QDSSCs) [1]. Among the common semiconductors, CdS is one of the most commonly used inorganic sensitizers [3] due to its optical properties and narrowed band gap adjusted by the particle size [4] and better UV stability [5] and it has reduced dark current [6]. However, the efficiencies of CdS sensitized TiO2 solar cells have still stayed low up to now. It has been extensively studied how to enhance their photovoltaic efficiency. One common effective method is cosensitization of more than one kind of quantum dots with different band gap. CdS and CdSe quantum dots cosensitized nanocrystalline TiO2 is one of the most common cosensitization structures with a power conversion efficiency of more than 3% [7, 8].
The other alternative method is to modify TiO2. In order to improve the photoelectric properties of TiO2 under sunlight, some metal or nonmetal impurities [9, 10] were added to generate donor or acceptor states in the band gap and to modulate energy band structure [11]. Recently, the hydrogenated TiO2 has attracted extensive attention. Chen et al. [12] demonstrated that the hydrogenated TiO2 nanocrystal enhanced solar absorption by introducing disorder in the surface layers of nanophase TiO2, and Wang et al. [13] reported hydrogen treatment was a simple and effective strategy to fundamentally improve the photoelectrochemical performance of TiO2 nanowires for water splitting. Our previous work reported the self-sensitized effect of hydrogenated TiO2 film which led to enhanced photovoltaic properties in the solar cell with hydrogenated TiO2 as photoanode without adding any dye [14].
If a photoanode was prepared by depositing CdS quantum dots on the surface of hydrogenated TiO2 film, the self-sensitized effect of hydrogenated TiO2 film would work together with quantum dots sensitization, which should also be a kind of cosensitization to be a promising method to enhance the photovoltaic properties of quantum dot sensitized solar cells. In this work, hydrogenated TiO2 films were fabricated on fluorine-doped tin oxide (FTO) glasses by screen printing and annealing under the specific temperature and time; then CdS quantum dots were attached to the surface of hydrogenated TiO2 by successive ionic layer adsorption and reaction (SILAR) method. Structural characterization, photoelectrochemical properties, and photovoltaic performances were investigated and discussed. The cosensitized TiO2 solar cells by CdS quantum dots and hydrogenation present much better photovoltaic properties than either CdS sensitized TiO2 solar cells or hydrogenated TiO2 solar cells. The reason for the enhanced photovoltaic performance of the novel cosensitized inorganic solar cell was also explained in detail.
2. Experimental
2.1. Preparation of Hydrogenated TiO2 Films and CdS Quantum Dots
Hydrogenated TiO2 (H:TiO2) films were prepared on fluorine-doped tin oxide (FTO) glasses with an area of 0.16 cm2 via the same processing as reported previously [14]. Then CdS quantum dots were attached to the surface of hydrogenated TiO2 by successive ionic layer adsorption and reaction (SILAR) method [15, 16]. In brief, 0.1 M Cd(NO3)2 in methanol was used as the cation source and 0.1 M Na2S in 1 : 1 methanol and water as the anion source. The H:TiO2 films with FTO substrates were successively dipped into the cation source and anion source for 5 min each. Following each dip, the films were rinsed for 1 min or more using pure ethanol to remove excess precursor, and the electrode was dried for 10 min before the next dipping. This dip cycle was repeated several times to obtain desirable CdS quantum dots on the surface of the H:TiO2 films as photoanodes. According to the preliminary experimented research on the effect of SILAR cycles for CdS quantum dots, an optimized SILAR cycle was determined as 9 cycles. For comparison, the H:TiO2 films without any quantum dots and TiO2 films with CdS quantum dots were also fabricated as photoanodes.
2.2. Assembly of Solar Cells
To assemble solar cells, the mixed solution of water and methanol with volume ratio of 3 : 7 consisting of 0.2 M KCl, 0.5 M Na2S, and 2.0 M S was used as polysulfide electrolyte. The photoanode, a platinized counter electrode, and the polysulfide electrolyte were sealed together with a hot-melt polymer film (Surlyn 1702-25, DuPont) to constitute a sandwich-like solar cell to measure photoelectrochemical properties. The active area of the cell is 0.16 cm2.
2.3. Testing Device and Characterization Method
The crystal structure of TiO2 films was characterized by an X-ray diffraction (XRD; X′ Pert Pro MPD) with Cu Kα1 radiation ( nm) and the film surface morphology was observed by SEM (INSPECT F, FEI Company). Ultraviolet-visible absorption spectra were measured by Shimadzu UV2100 Spectrophotometer. In order to measure photovoltaic performance of the solar cells, simulated sunlight (AM 1.5 G) was irradiated using a solar simulator with a Xe lamp (calibrated with a standard Si-based solar cell), and the current-voltage (-) curve was recorded by a CHI-660 electrochemical workstation. The photocurrent density (), open-circuit voltage (), fill factor (FF), and corresponding photoenergy conversion efficiency (η) were calculated from the obtained - curves. The incident photon to current conversion efficiencies (IPCE) spectra were obtained by Newport QE-PV-SI. Electrochemical impedance spectroscopy (EIS) was also measured at the CHI-660 electrochemical workstation with sinusoidal perturbations of 10 mV at frequencies from 0.01 Hz to 10 kHz with zero bias potential.
3. Results and Discussion
3.1. Characterization of TiO2 Films Cosensitized by CdS and Hydrogenation
X-ray diffraction patterns of the H:TiO2 film, CdS/TiO2 film, and CdS/H:TiO2 film are shown in Figure 1. TiO2 in all the films is in the anatase phase, the same as reported in our previous work [14]. With the successive ionic layer adsorption in 9 cycles, the peaks at 27.9° and 43.8° corresponding to (100) and (110) planes of CdS quantum dots, respectively, can be observed in the patterns of CdS sensitized TiO2 and H:TiO2 films as shown in Figures 1(c) and 1(d), respectively. No obvious difference is observed between the XRD patterns of the CdS sensitized TiO2 films with or without hydrogenation. Figure 2 shows SEM surface morphology of pure H:TiO2 film and CdS/H:TiO2 film. The H:TiO2 film is typical porous nanocrystals as in Figure 2(a), while the CdS/H:TiO2 film displays dense-packing CdS quantum dots with the grain size of about 25 nm according to Figure 2(b).
(a)
(b)
3.2. Absorption of TiO2 Films Cosensitized by CdS and Hydrogenation
The UV-visible absorption spectra of H:TiO2 film, CdS/TiO2 film, and CdS/H:TiO2 film are showed in Figure 3. It has been demonstrated that hydrogenation treatment can improve light absorption of TiO2 due to the generated dangling bonds and disordered surface layers on the surface of TiO2 nanophase [12]. Compared with the H:TiO2 film with a band gap of about 3.10 eV, the CdS/TiO2 film has a much stronger absorption in the range of 300~800 nm in that CdS has a much smaller band gap (2.25 eV in bulk) than H:TiO2 [17]. When CdS quantum dots are deposited on the surface of the H:TiO2 films, the obtained CdS/H:TiO2 film combines the advantages of both hydrogenation treatment and CdS quantum dots and thus reveals the strongest absorption as shown in Figure 3.
3.3. Impedance Spectra of the Solar Cells Based on Cosensitization of CdS and Hydrogenation
In order to analyze the internal electron transport process of solar cells, the electrical impedance spectra (EIS) for the sensitized solar cells based on CdS/H:TiO2, CdS/TiO2, and H:TiO2 are shown in Figure 4(a), and Figure 4(b) shows the relevant equivalent-circuit model. Similar to a typical DSSC system [18, 19], and CPE1 represent the electron transfer resistance and interfacial capacitance at the interface between counter electrode and electrolyte, respectively. and CPE2 are electron transport resistance and interfacial capacitance at the interface between electrolyte and photoanode, respectively. is ohmic series connection resistance of the whole cell. is the electrolyte Nernst diffusion impedance. Theoretically, the solar cell impedance spectroscopy has three semicircles representing high-, middle-, and low-frequency features, associated with and CPE1, and CPE2, and , respectively [20]. In Figure 4(a), the largest semicircle at middle-frequency almost covers the other two, which is attributed to carrier transportation and recombination at photoanode/electrolyte interfaces. The fitted values of are 18 Ω, 32 Ω, and 56 Ω for the solar cells with photoanode of CdS/H:TiO2, CdS/TiO2, and H:TiO2, respectively. According to our previous work, hydrogenation treatment of TiO2 can reduce because the band gap is narrowed down and meanwhile the oxygen vacancy density increases. The loading of CdS nanoparticles on pure TiO2 also decreases of the CdS/TiO2/electrolyte interfaces significantly. For the CdS/H:TiO2 photoanode, both of hydrogenation self-sensitization and CdS sensitization work together and result in the lowest . The lower reflects quicker electrons transport at the interface between electrolyte and photoanode, which implies that smaller probability of interface recombination occurs at the photoanode/electrolyte interface.
(a)
(b)
3.4. Photovoltaic Performances of the Solar Cells Based on Cosensitization of CdS and Hydrogenation
The incident photon to current conversion efficiencies (IPCE) spectra of the solar cells with different photoanodes are shown in Figure 5. The IPCE can be expressed as [21] where is the light harvesting efficiency, is the electron injection yield, and is the charge collection efficiency. It can be seen that the profile of IPCE plot corresponds well with the UV-vis absorption spectra in Figure 3. The IPCE of TiO2 nanocrystal film without any sensitizer is low in the UV region and negligible in the visible region. The H:TiO2 solar cell exhibits stronger IPCE. The hydrogenated TiO2 increases oxygen vacancy density and has much more trap states near the conduction band, leading to the enhancement of the absorption as reported in our previous work [14]. The absorption enhancement implies the increase of the light harvesting efficiency, which increases the IPCE value according to the above formula. Significantly enhanced IPCE can be observed in the CdS sensitized solar cells, which can be attributed to the fact that CdS quantum dots have much higher absorption coefficients of 105 to 106 M−1 cm−1 above the band gap energy [22]. The CdS/H:TiO2 solar cell exhibits the strongest IPCE due to the cosensitization of CdS and self-sensitization of hydrogenated TiO2. The maximum IPCE of the CdS/H:TiO2 solar cell can reach 41.7% using a polysulfide electrolyte, suggesting that the cosensitization of CdS and hydrogenation has the potential to enhance the photoelectric properties.
Figure 6 shows the illuminated and dark - characteristics of the solar cells with different photoanodes, and the photovoltaic performances of them are listed in Table 1. The self-sensitization effect of H:TiO2 photoanode with polysulfide electrolyte is very weak, exactly as our previous results of H:TiO2 photoanode with the I−/ electrolyte. The CdS/TiO2 QDSSC also displays the ordinary photovoltaic performance as others’ work with an energy conversion efficiency of 1.1%. Compared with the other two solar cells, the CdS/H:TiO2 solar cell, which is cosensitized by CdS and hydrogenation, exhibits enhanced photovoltaic characteristics with short circuit current density () of 6.26 mA/cm2, open-circuit voltage () of 0.65 V, fill factor (FF) of 0.49, and power conversion efficiency () of 1.99%.
(a)
(b)
The enhancement of can be explained by the equation: , where and are the electron injection current density and the recombination current density [23]. The stronger absorption at the UV-vis (as shown in Figure 3) and the higher IPCE (as shown in Figure 5) of CdS/H:TiO2 hold out the potential of increasing and consequently enhancing the photoelectric effect. On the other hand, the recombination current is derived from the interface recombination mostly caused by the reduction reaction with the electron in the conduction band of TiO2 and in the electrolyte, which would form the dark currents and reduce photovoltaic performance of solar cells. As shown in Figure 6(b), the dark current density of the solar cell with CdS/H:TiO2 photoanode is obviously lower than that of the CdS/TiO2 or H:TiO2 solar cell, which should be attributed to lower of the CdS/H:TiO2 solar cell than that of the other two solar cells. The enhanced and the suppressed dark current density of the CdS/H:TiO2 solar cell lead to the enhancement of the .
Under constant illumination, the solar cell reaches a photostationary situation, and corresponds to the increase of the quasi-Fermi level of the semiconductor () with respect to the dark value (), which equals the electrolyte redox energy (). can be determined by the following equation [24]:where is the positive elementary charge; is the thermal energy; is the electron concentration in conduction band of the semiconductor photoanode under illumination; is the electron concentration in the dark condition. Here, and can be characterized by IPCE and dark current density, respectively. The stronger IPCE as shown in Figure 5 implies the higher of the CdS/H:TiO2 solar cell than the H:TiO2 solar cell and the CdS/TiO2 solar cell, while the lower dark current density as shown in Figure 6(b) manifests the lower of the CdS/H:TiO2 solar cell than the other two solar cells. As a result, the higher and the lower lead to the improvement of for the CdS/H:TiO2 solar cell as shown in Figure 6(a).
4. Conclusions
In summary, a cosensitized TiO2 photoanode by CdS quantum dots sensitization and self-sensitization of hydrogenated TiO2 film was achieved by depositing CdS quantum dots on the surface of hydrogenated TiO2 film. By comparing solar cells with different photoanodes of H:TiO2, CdS/TiO2, and CdS/H:TiO2, the cosensitization effect by CdS and hydrogenation in the CdS/H:TiO2 solar cell enhanced photovoltaic performance with power conversion efficiency (η) of 1.99%, which was increased by more than 80% compared with CdS/TiO2 solar cells. The cosensitization effect combined the quantum dots sensitization and self-sensitization of hydrogenated TiO2 films and caused larger extension absorption in the visible light range, quicker electrons transport, smaller probability of interface recombination, and consequently better photovoltaic performance. This study will give some useful enlightenment to the development of novel inorganic low-cost solar cells.
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 the National Natural Science Foundation of China (51472043, 51272035, and 51272037), China-Japan International Cooperation Program Funds (nos. 2010DFA61410 and 2011DFA50530), Program for New Century Excellent Talents in University (no. NCET-12-0097), and the Fundamental Research Funds for the Central Universities (no. ZYGX2014J023).