Self-cleaning SiO_x-embedded TiO₂/SiO₂ alternating multilayer

Abstract

A new method of inserting an eco-sustainable hydrophilic system into an anti-reflection (AR) system was developed and its applicability to photovoltaic systems was determined. We created a TiO₂ nanoparticle composite with a densely multi-layered AR system top layer to demonstrate its eco-sustainable, hydrophilic performance and to act as a low refractive index material, which is essential for AR. The hydrophilic system displayed the lowest reflectance and a remarkable contact angle reduction to 0 degrees under UV light irradiation. The annealed hydrophilic system (the eco-sustainable system) exhibited the highest photodecomposition activity of 100% within 90 min by effective separation of electron–hole pairs using SiO_x nanoparticles inside the TiO₂ layer as trap levels. The successful AR, hydrophilic, and eco-sustainable characteristics of the simplified system composed of only two materials, TiO₂ and SiO_x, are anticipated to contribute to the enhancement of photovoltaic system efficiency.

---

1. Introduction

In recent years, transition metal oxide materials (TMOM) have received considerable attention due to their remarkable optical properties in anti-reflection (AR) and photocatalytic applications. AR systems are used in photovoltaic devices to improve the absorption efficiency of solar irradiation. On the other hand, photocatalytic systems, which show great promise for environmental purification, are called eco-sustainable systems and have received considerable attention due to recent worldwide environmental problems. So far, AR and eco-sustainable studies appear to be contradictory because most studies have separately focused on the characterization and fabrication of either AR or eco-sustainable systems. Combining the two systems is expected to widen TMOM usage.

An AR system is an optical thin film used for increasing optical transmission rate of a substrate. It is a method of reducing reflection by using the interference effect of light generated by the refractive index differences between thin-film and substrate. Such systems have been widely used in liquid crystal displays, plasma display panels, light-emitting diodes, organic electroluminescence, including optical elements, optical filters, solar cells, and other applications.^1–4 Generally, using materials with a high (H), low (L), and middle (M) refractive index, a multilayer AR coating with high transmittance in a wide visible wavelength range can be realized. Three common methods are employed in AR coating design for a wide wavelength range: the first is W-coating (air/LHH/glass), the second is quarter-half-quarter coating (air/LHHM/glass), and the third is multilayer AR coating (air/LHH0.32L0.26H/glass).⁵ The key feature of these three methods is that a low refractive index material is used in the first layer to reduce reflection because a high refractive index material shows high reflectance.

Eco-sustainable photocatalytic systems have recently attracted considerable attention due to their various applications in environmental purification and dye sensitized photovoltaic solar cells. Among various metal oxide photocatalysts, titanium dioxide (TiO₂) has been one of the most promising materials in both fundamental studies and practical applications because of its high photoactivity, biological and chemical inertness, cost effectiveness, non-toxicity, and long-term stability against photo-corrosion and chemical corrosion.^6–8 Upon band-gap excitation of TiO₂, photo-induced electrons and positively charged holes can reduce and oxidize the species adsorbed on the surface of TiO₂ particles, respectively.^9–12 Metal-doped TiO₂ can reduce electron–hole pair recombination and increase hydroxyl radical concentration on the TiO₂ surface, resulting in increased photocatalytic activity of the TiO₂ thin film.^13,14 Although there has been a number of studies for enhancing photovoltaic device efficiency, there are chronic and environmental problems caused by dust and contaminants on device surfaces. Photovoltaic systems are usually located outdoors and are exposed to many pollutants, resulting in reduction of solar transmittance.¹⁵ To protect these systems from dust and contaminants, many researchers focus on self-cleaning effect, for which hydrophilicity is one of the solutions. The hydrophilic surface flattens water and this water cleans dust and contaminants on the surface when it is pulled down by gravity.¹⁶

In this study, we applied TiO₂ nanoparticle composite to a multi-layer AR thin film to actualize a novel eco-sustainable AR (E-AR) system and a hydrophilic AR (H-AR) system and tried to demonstrate their mechanism by investigating the refractive index, crystallinity, and distribution of SiO_x material in TiO₂ layers. Reflectance and transmittance were analyzed to evaluate the AR effect; the water contact angle and photodecomposition of methylene blue under black-light irradiation were measured to confirm the hydrophilic and eco-sustainable effects, respectively.

2. Results and discussion

We prepared a traditional multi-layer AR (T-AR) system of SiO₂/TiO₂/SiO₂/TiO₂ and a hydrophilic AR (H-AR) system, which introduces a TiO₂ nano-sphere (L-TiO₂) layer on the T-AR system (Fig. 1a). The TiO₂ nano-sphere was developed using glancing angle deposition and its refractive index was reduced to be less than 1.500 at 550 nm, similar to SiO₂ thin film, as shown in Fig. 1a. A low refractive index nano-sphere TiO₂ (L-TiO₂) layer with a thickness of 10 nm was fixed and the optimization process was carried out using Essential Macleod thin-film design software. The refractive indices of the TiO₂, L-TiO₂, and SiO₂ layers of the H-AR system were found to be 2.365, 1.498, and 1.487 at a reference wavelength of 550 nm, respectively (Fig. 1b). The refractive index of L-TiO₂ was conspicuously reduced and made nearly equal to that of SiO₂. This shows the possibility of simultaneous manifestation of AR and photocatalytic activity with only one system by substituting L-TiO₂ for SiO₂. The reflectance values of the T-AR and H-AR systems are shown in Fig. 1c. Average reflectance in the visible light range (400–700 nm) is reduced to 0.90% and 0.35% for the T-AR and H-AR systems, respectively. The average transmittance of these systems, considering the backside surface effect of the substrate,¹⁷ is over 95%. Fig. 1d shows the change in transient water contact angle in the T-AR system, L-TiO₂ thin film, and H-AR system under UV light irradiation. The T-AR system shows a slight loss of 5° and the L-TiO₂ system drops the angle to 20° in 80 min. However, in the H-AR system, there is an extraordinary reduction to 0° in only 40 min. The L-TiO₂ adopted system shows a remarkable decrease in contact angle value. This can be explained by metastable OH groups produced by the interaction of holes and H₂O molecules under UV light irradiation. These OH radicals generated from the photoinduced reaction are thermodynamically less stable and increase TiO₂ surface energy, resulting in hydrophilic modification.¹⁸ The best result is obtained from the H-AR system, which has a SiO₂ layer under L-TiO₂ nanoparticles. This result corresponds with chemical changes in the surface structure and an increase in hydroxyl groups at the interfaces due to enhanced acidity of the Si–O–Ti bonds by UV light irradiation.¹⁹

Fig. 1 (a) The schematic diagram and the cross-sectional STEM-HAADF images of the H-AR systems. (b) The refractive indices of the TiO₂, L-TiO₂, and SiO₂ thin-films in the visible range. (c) The reflectance of the T-AR system and H-AR system. (d) Photo-induced reduction of the water contact angles of the T-AR system, L-TiO₂ thin film, and H-AR system.

Fujishima et al.²⁰ studied the hydrophilic conversion of TiO₂ surface and demonstrated that the hydrophilic conversion phenomenon can be explained by photocatalytic oxidation and the metastable state. The metastable state is caused by an increase in the number of OH groups on the TiO₂ surface, which has high surface energy. A highly hydrophilic conversion was completed even if stains remained on the surface and it retained that condition for a day or two under ambient conditions without additional light irradiation. After several days, it returned to the initial hydrophobic condition. They predicted the hydrophilic and hydrophobic conversions of the H-AR system. The E-AR system can be obtained under various conditions such as under light irradiation and dark condition. The rate of conversion between H-AR and E-AR depends on the number of OH groups. The E-AR system may exhibit faster hydrophilic conversion and a slower hydrophobic conversion rate.

The E-AR system was prepared by annealing the H-AR system at 650 °C for 30 min (Fig. 2a). During heat treatment, the thin film becomes dense due to the reduction of voids, resulting in a decrease in optical thickness. Therefore, in this study, we prepared dense films through the evaporation process;²¹ thus, layer structure change was minimized during the annealing process. A preceding study on a dense TiO₂ thin-film has shown that changes in optical property due to modification of the thin film phase are minimized. The transmittance spectra of the H-AR and E-AR systems show that they have similar transmittance rates in the visible range (Fig. 2b and c). Slight changes in optical transmittance result from small refractive index differences due to film densification. The E-AR and H-AR systems have average transmittance values of about 95.76% and 95.95% in the range of 400–700 nm, respectively. Thus, we can expect their reflectance to be below 1.00% in the visible range.

Fig. 2 (a) The schematic diagram of post-annealed H-AR (E-AR) systems, HR-TEM images, and the EDS analysis of the L-TiO₂ top layer. (b and c) The transmittance of the E-AR system, H-AR system, and quartz glass. (d) The X-ray diffraction patterns of the L-TiO₂ top layer of the E-AR system and L-TiO₂ thin-film annealed at 650 °C.

There is a sudden decrease in transmittance below 400 nm, meaning that UV light is absorbed by the AR coating. This UV absorption is favorable for photocatalytic performance under black light irradiation but unfavorable for solar cell current (Fig. 2b). Although the AR coating has the disadvantage of UV absorption, it has the more attractive benefit of visible light transmittance.

Yu et al. investigated the effect of nanocone arrays as an antireflective coating on solar cell performance. Although the film coated with nanocone arrays shows a decreased transmittance below 350 nm, it exhibits a higher transmittance than uncoated film in the range of 400–1800 nm, which enhances device efficiency.¹⁶

Fig. 2a shows a schematic diagram of the E-AR system. The chemical composition of particles was determined by energy dispersive spectroscopy (EDS) (Table 1). The EDS spectrum indicated that the amorphous particles have Si–O bonding. According to the EDS results, Si–O bonding particles have a stoichiometry of SiO₂ during the annealing process. Since SiO₂ particles are smaller than TiO₂ particles, diffusion of SiO₂ particles (or Ti and Si in the oxide matrix) occurs throughout the TiO₂ particles and is activated by thermal energy.²²

Table 1 The EDS results of the E-AR system

Layers	O	Si	Ti
1	60.71	5.67	33.62
2	62.63	37.37
3	54.98	5.37	39.65
4	54.51	45.49
5	55.73	6.68	37.59

---

Fig. 2d shows the X-ray diffraction patterns of the L-TiO₂ top layer of the E-AR system and annealed L-TiO₂ thin-film deposited on quartz glass to compare the differences with respect to the existence of a SiO₂ layer under the L-TiO₂ layer. The crystallite sizes of anatase TiO₂ in the E-AR system and the annealed L-TiO₂ thin film are 12.5 and 18.7 nm, respectively. The SiO₂ under-layer seemed to suppress single crystallization of the anatase phase. Consequently, due to the SiO₂ under-layer, the anatase phase peak of the E-AR system revealed a smaller crystallite size in comparison to the annealed L-TiO₂ thin film.

Fig. 3 shows the surface morphology of the AR system. The surface of the H-AR system (before heat treatment) was composed of spherical TiO₂ particles, while the surface of the E-AR (after heat treatment) consisted of non-spherical particles as a result of phase transformation to anatase with embedding of SiO_x and agglomeration of particles.

Fig. 3 FE-SEM images of the AR system before (H-AR) and after (E-AR) heat treatment.

The surface characterizations of the E-AR and H-AR systems were obtained by X-ray photoelectron spectroscopy (XPS), as shown in Fig. 4a and b. The C1s photoelectron peak is shown at 284.60 eV, which may appear due to carbon pollution in the air or the presence of remaining organic components. Fig. 4a shows the expanded spectrum of the O1s photoelectron peak, which was fitted into three peaks by Lorentzian curves appearing at 532.5 eV, 531.6 eV, and 530.1 eV, and can be attributed to Si–O–Si, Si–O–Ti, and Ti–O–Ti components, respectively.²³ The XPS results show that more complicated oxygen coordination states appear in the E-AR system than in the H-AR system, and that more connections are present between titanium species and the silica matrix. Furthermore, the oxidation states of Ti in the L-TiO₂ top layer of the E-AR system for Ti³⁺2p (Ti³⁺2p_1/2 at 462.35 eV, Ti³⁺2p_3/2 at 456.51 eV) peaks and Ti⁴⁺2p (Ti⁴⁺2p_1/2 at 463.98 eV, Ti⁴⁺2p_3/2 at 458.21 eV) peaks were detected, as previously reported.²⁴ The XPS spectrum of the Si2p core electrons of the L-TiO₂ top layer clearly matches the EDS measurement for amorphous particles with Si–O bonding in Fig. 2d; the measured binding energy is 101.6 eV, which is a typical value for SiO_x species.²⁵ Moreover, SiO_x species were formed by Ti–O–Si bonding on the surface of the E-AR system through heat treatment. The formation process of SiO_x species can be explained as follows. Ti–O–Si bonding occurs by the electron beam effect due to the coexistence of TiO₂ and TiO_x. SiO₂ particles move into TiO₂ layers by the annealing effect and SiO₂ bonding particles interact with TiO_x bonding particles in the top layer by thermal energy. Because the binding energy of Ti–O bonding is stronger than that of Si–O bonding, a certain amount of Ti³⁺ is reduced during the thermal process, and a SiO_x species is formed. We can confirm this process by comparing O1s peaks and Ti2p peaks in Fig. 4a and b. In Fig. 4b, the spectrum of pure L-TiO₂ contains one peak appearing at 529.5 eV, which is typical for metal oxides and agrees with the O1s electron binding energy for TiO₂ molecules. In the XPS narrow scan spectrum, Ti³⁺2p (Ti³⁺2p_1/2 at 462.36 eV, Ti³⁺2p_3/2 at 457.02 eV) peaks and Ti⁴⁺2p (Ti⁴⁺2p_1/2 at 464.26 eV, Ti⁴⁺2p_3/2 at 458.73 eV) peaks were detected. The observation of the Ti³⁺ state on the TiO₂ surface is important because it can play a role as a metal or doped impurity that can capture the reaction electrons and remaining unpaired charges and support photocatalytic activity. The X-ray photoelectron spectrum of Si2p core electrons was not observed in the H-AR system.

Fig. 4 X-ray photoelectron spectra of the top layer of (a) E-AR system and (b) H-AR system. The PL spectra of (c) E-AR system and (d) annealed L-TiO₂ thin film.

Bard et al.²⁶ demonstrated the usefulness of the TiO₂/SiO₂ multilayer. TiO₂ has polyphases, such as anatase, rutile, and brookite phases, depending on annealing temperature. In the case of the TiO₂/SiO₂ layer system, the SiO₂ lattice locks the Ti–O species at the interface to prevent nucleation, which is necessary for phase transformation to rutile. As a result, the TiO₂/SiO₂ mixed layer shows more of the anatase phase than the TiO₂ single layer at the same annealing temperature. It is known that the anatase phase of TiO₂ has good catalytic performance. Accordingly, the TiO₂/SiO₂ layer is expected to show more effective photocatalytic performance than the TiO₂ single layer.

PL spectra of the E-AR system and annealed L-TiO₂ thin film at room temperature are shown in Fig. 4c and d. A peak at 385 nm corresponds to near-band-edge emission of TiO₂ and emissions at 405 nm, 421 nm, and 426 nm may be attributed to radiative recombination at the donor levels caused by Ti³⁺ ions.²⁷ The energy-level diagram in Fig. 4c shows emissions at 450 and 475 nm, which can be possibly related to two trap levels of the SiO_x-embedded TiO₂ particles.

The photocatalytic activity of the prepared samples was evaluated by measuring the photodegradation rates of methylene blue (MB) under UV irradiation. The photodegradation rate is calculated using the following equation: photodegradation rate = C/C₀, where C₀ is the initial concentration of MB and C is the MB concentration at a given time. MB degradation with UV irradiation time for the E-AR system, H-AR system, L-TiO₂ thin-film, annealed L-TiO₂ thin film, and MB self-photofading are depicted in Fig. 5. It is clearly demonstrated that the E-AR system exhibits the best MB photodegradation performance. The MB pollutant is almost completely decomposed after 90 min. However, only about 38% and 61% of MB dye is removed after 300 min by the H-AR system and annealed L-TiO₂ thin film, respectively. For the L-TiO₂ thin film, the increased photodegradation rate after heat treatment is attributed to the crystallization of TiO₂ to anatase phase.

Fig. 5 (a) The photocatalytic activities of samples evaluated by photodegradation of methylene blue in an aqueous solution. (b) The energy-level diagram of E-AR system showing the Fermi level of TiO₂ and trap levels of SiO_x-embedded TiO₂ contacts.

The enhanced photocatalytic activity of the E-AR system caused by SiO_x-embedded TiO₂ can be explained as follows. As shown in the schematic diagram of Fig. 5b, the conduction band (CB) and valence band (VB) edges of TiO₂ and SiO_x-embedded TiO₂ are located within the trap levels. Under UV irradiation, electron–hole pairs are initially generated in TiO₂ and SiO_x-embedded TiO₂. The generated electrons accumulate in the CB and are attracted to the trap levels due to functional differences, thus preventing e⁻–h⁺ recombination. The generated holes (h⁺) react with H₂O and OH⁻ groups adsorbed on the TiO₂ surface to produce hydroxyl radicals. The electrons accumulated in the trap levels of SiO_x-embedded TiO₂ reunite with holes that do not react with water. The PL spectra of the E-AR system and the annealed L-TiO₂ thin film in Fig. 4c and d suggest that there are two trap levels located at 2.76 eV (450 nm) and 2.61 eV (475 nm) above the VB of TiO₂. Consequently, the excellent photocatalytic activity of E-AR system is attributed to SiO_x-embedded TiO₂ nanoparticles produced during the heat treatment process.

3. Experimental

The samples were prepared on quartz substrates using a high-vacuum e-beam evaporation system (SHT-CT-800A-DA, Korea). The evaporation sources were periodically shuttered via computer controlled pneumatic shutters with 7.0 kV of electron gun voltage. The current was controlled at 220 mA for TiO₂ and 110 mA for SiO₂. The sources of TiO₂ and SiO₂ were 99.99% pure.

To prepare a denser film, the electron beam evaporator was evacuated to a base pressure of 6.0 × 10⁻⁶ torr; TiO₂ and SiO₂ films were deposited at 200 °C. The partial pressure of oxygen gas was 6.0 × 10⁻⁵ torr for TiO₂ film deposition and 5.0 × 10⁻⁵ torr for SiO₂ film deposition. Oxygen was introduced during deposition using a mass flow controller (MKS 1179A) to promote the growth of a more stoichiometric film at an oxygen partial pressure of 5.0 × 10⁻⁵ torr. In the case of insufficient oxygen, a stoichiometric film was made using electron beam radiation. The substrates were rotated at 15 rpm to obtain uniform films. TiO₂ and SiO₂ thin films were deposited at rates of 2.5 Å s⁻¹ and 7.0 Å s⁻¹, respectively.

We fabricated a TiO₂ top layer with a low-refractive index by applying a nano-sphere layer. The nano-sphere layer was prepared by glancing angle deposition. L-TiO₂ film was fabricated with incident vapor flux at an angle of 72° relative to the substrate with a deposition rate of 2.0 Å s⁻¹, and no substrate rotation.

The TiO₂ phase changed from anatase to rutile in the annealing process. According to our previous study, anatase and rutile phases of TiO₂ thin film are formed at 400 and 800 °C, respectively. TiO₂ thin film is amorphous below 400 °C and undergoes morphological and optical changes such as an increase in grain size and a decrease in transmittance above 700 °C.²⁸ Wide variations in grain size lead to defects between the TiO₂ and SiO₂ interfaces and a decreased film transmittance results in reduced photovoltaic system efficiency. Therefore, we slowly increased the annealing temperature to 650 °C at a rate of 10 °C min⁻¹ to prevent the previously mentioned problems and maintained that temperature for 30 min under atmosphere. The sample was then allowed to cool naturally.

The optical constant of the thin films was evaluated by spectroscopic ellipsometry (Ellipsotech, EllI-SE-F). The hydrophilicity of the films surface was quantified from water contact angle measurements. Experiments were performed at room temperature in air using a goniometer equipped with a digital camera. Several water droplets (0.5 μL) were spread on the samples and water contact angles were measured under irradiation of four surrounding 20 W black-light (UVA) lamps (wavelength range: 315–400 nm). Each lamp was located at the side of a square box and they were 10 cm away from the sample. Optical transmittance was measured with a HP 8453 spectrophotometer. The crystal phase of samples was determined by X-ray diffraction (XRD, Philips PW3710) in the 2θ mode using monochromatic Cu Kα radiation at 30 kV and 20 mA. The θ range used in XRD measurements was from 20° to 80° in steps of 0.02° s⁻¹. To investigate nanostructure, high angle annular dark field (HAADF-STEM) imaging was applied for observing the L-TiO₂ layer using JEM-2200FS. To evaluate compositional variations in the L-TiO₂ layer, high resolution transmission electron microscopy (HR-TEM) and energy dispersive spectroscopy (EDS) were also applied using JEM-2200FS. The surface chemical state of elements was analyzed by X-ray photoelectron spectroscopy (XPS, KBSI, ESCALAB 250) with a monochromatic Al Kα X-ray source. The photocatalytic properties of samples were determined by measuring the photodecomposition of MB (C₁₆H₁₈N₃S–Cl–3H₂O) with an initial concentration of 10⁻⁵ mol L⁻¹ under irradiation of four surrounding 20 W-black-light (UVA) lamps (wavelength range: 315–400 nm). Photocatalytic degradation was monitored by measuring the absorption spectra of the solution at λ_max = 664 nm. Photoluminescence (PL) spectra were collected using a Cary Eclipse fluorescence spectrophotometer (Varian) with an excitation wavelength of 295 nm.

4. Conclusions

Hydrophilic, photocatalytic, and AR properties have been successfully realized in one facile system of TiO₂ and SiO₂ by controlling the morphology of top layer and heat treatment. The low refractive index nano-sphere TiO₂ (L-TiO₂) layer endows the multi-layer system with considerable optical properties and extraordinary hydrophilicity under UV light irradiation. Heat treatment at 650 °C leads to specific chemical bonding changes and shows effective separation of photogenerated electron and hole pairs by the formation of trap levels between conduction band and valence band. This functionalized system showed remarkable photocatalytic activity of 100% within 90 min and has high potential for improving photovoltaic system efficiency.

Acknowledgements

This research was financially supported by the Priority Research Centers Program (2009-0093818) and by the National Research Foundation of Korea (NRF) through the Basic Science Research Program (2012R1A1A4A01015466).

References

1. J. K. Kim, T. Gessmann, E. F. Schubert, J. Q. Xi, H. Luo, J. Cho, C. Sone and Y. Park, Appl. Phys. Lett., 2006, 88, 013501.
2. N. Wang, J. Fang, X. Zhang, G. Wang, L. Wang, C. Liu, H. Zhao, Z. Chen, X. L. Chen, J. Sun and Y. Zhao, Sol. Energy Mater. Sol. Cells, 2014, 130, 420.
3. Y. Wang, H. Wang, X. Meng and R. Chen, Sol. Energy Mater. Sol. Cells, 2014, 130, 71.
4. T. Glaser, A. Ihring, W. Morgenroth, N. Seifert, S. Schroter and V. Baier, Microsyst. Technol., 2005, 11, 86.
5. H. A. Macleod, Thin-Film Optical Filters, Hilger, 2nd edn, 1986.
6. H. S. Kim, J. W. Lee, N. Yantara, P. P. Boix, S. A. Kulkarni, S. Mhaisalkar, M. Grätzel and N. G. Park, Nano Lett., 2013, 13, 2412.
7. L. Etgar, D. Yanover, R. K. Čapek, R. Vaxenburg, Z. Xue, B. Liu, M. K. Nazeeruddin, E. Lifshitz and M. Grätzel, Adv. Funct. Mater., 2013, 23, 2736.
8. L. Etgar, W. Zhang, S. Gabriel, S. G. Hickey, M. K. Nazeeruddin, A. Eychmüller, B. Liu and M. Grätzel, Adv. Mater., 2012, 24, 2202.
9. A. L. Linsebigler, G. Lu and J. T. Yates, Chem. Rev., 1995, 95, 735.
10. X. Pang, W. Chang, C. Chen, H. Ji, W. Ma and J. Zhao, J. Am. Chem. Soc., 2014, 136, 8714.
11. G. Fu, P. S. Vary and C. T. Lin, J. Phys. Chem. B, 2005, 109, 8889.
12. R. Comparelli, E. Fanizza, M. L. Curri, P. D. Cozzoli, G. Mascolo, R. Passino and A. Agostiano, Appl. Catal., B, 2005, 55, 81.
13. F. Pan, J. Zhang, W. Zhang, T. Wang and C. Cai, Appl. Phys. Lett., 2007, 90, 122114.
14. J. M. Jung, M. Wang, E. J. Kim, C. Park and S. H. Hahn, Appl. Catal., B, 2008, 84, 389.
15. S. Ghazi, A. Sayigh and K. Ip, Renewable Sustainable Energy Rev., 2014, 33, 742.
16. J. W. Leem, J. S. Yu, J. Heo, W. K. Park, J. H. Park, W. J. Cho and D. E. Kim, Sol. Energy Mater. Sol. Cells, 2014, 120, 555.
17. J. D. Rancourt, Optical Thin Films: Users Handbook, McGraw-Hill, New York, 1987.
18. S. Nobuyuki, F. Akira, W. Toshiya and H. Kazuhito, J. Phys. Chem. B, 2003, 107, 1028.
19. F. Razan, D. Ralf and B. Detlef, Langmuir, 2013, 29, 3730.
20. K. Hashimoto, H. Irie and A. Fujishima, Jpn. J. Appl. Phys., 2005, 44, 8269.
21. J. S. Kim, S. H. Hahn and E. J. Kim, J. Korean Phys. Soc., 2007, 51, 2014.
22. H. Sankur and W. Gunning, J. Appl. Phys., 1989, 66, 4747.
23. Z. L. Hua, J. L. Shi, L. X. Zhang, M. L. Ruan and J. N. Yan, Adv. Mater., 2002, 14, 830.
24. J. Jun, J. H. Shin and M. Dhayal, Appl. Surf. Sci., 2006, 252, 3871.
25. V. Levitcharsky, R. G. Saint-Jacques, Y. Q. Wang, L. Nikolova, R. Smirani and G. G. Ross, Surf. Coat. Technol., 2007, 201, 8547.
26. C. Anderson and A. J. Bard, J. Phys. Chem. B, 1997, 101, 2611.
27. T. Sekiya, S. Kamel and S. Kurita, J. Lumin., 2000, 87–89, 1140.
28. D. J. Kim, S. H. Hahn, S. H. Oh and E. J. Kim, Mater. Lett., 2002, 57, 355.

---