Significant roughness enhancement of fluorine-doped tin oxide films with low resistivity and high transparency by using HNO₃ addition

Abstract

Current development of fluorine-doped tin oxide (FTO) films for application in solar cells is limited by tradeoffs among surface roughness related light trapping, optical transparency and electrical conductivity. Proposed methods for improving the surface roughness are usually dependent upon thickness increment or sacrificing the conductivity. In this study, we report a simple method for preparing FTO films with a flower-like grain morphology and thus a higher than ever mean roughness of 51.4–73.3 nm or mean-root-square roughness of 70.5–91 nm, by inclusion of HNO₃ additive into the deposition system. The very rough films are 200–300 nm thick and also have maintained a high transparency of 80–85% in the visible region and a low sheet resistance of 14 Ω □⁻¹ and resistivity of 4.2 × 10⁻⁴ Ω cm, perhaps resulting from the effects of HNO₃ on heterogeneous nucleation and development of the (200) preferred orientation. The high roughness, combined with high transparency and low resistivity, would be beneficial to improving the light trapping of FTO films and thus the efficiency of solar cells.

---

1 Introduction

Among transparent conductive oxide (TCO) thin films, fluorine doped tin oxide (FTO) ones possess outstanding features of chemical inertness, stability to heat treatment and mechanical hardness, in addition to their excellent electrical and optical properties. As a result, FTO films have been applied in many fields, such as gas sensors,^1,2 solar cells,^3–5 flat panel displays⁶ and low emissivity windows.⁷ Each of these applications sets specific requirements for electrical, optical, morphological and structural properties.⁸

Solar cell is a device which converts the solar energy into useful electrical energy. The main application of FTO thin films in this field is as the front electrical contact, which allows the solar spectrum in the visible region to pass through. A high optical transmittance is required for an efficient transport of light into the active stack, and a good electrical conductivity for a low transport loss of the generated electrical energy. In general, an excellent transparency (more than 80%) in the visible range and low electrical conductivity (<10⁻³ Ω cm) are essential for the front FTO films.^9,10

In addition to transparency and conductivity, the light trapping ability of the film is another important factor to affect the efficiency of solar cells.¹¹ A light trapping structure contributes to increase the light path length by reflecting and scattering the incident light at different angles. Since a high surface roughness is beneficial to diffuse transmission or total internal reflections of lights and reduce the optical losses in the film,^12,13 roughness enhancement has been regarded as an effective method for enhancing light trapping^12,14–16 and thus energy conversion efficiency of solar cells.^17–22

For TCO thin films, some approaches have been used to enhance the surface roughness determined as either R_a, the mean roughness, or R_mrs, the mean root square roughness. The primary previous method for increasing roughness was increasing the film thickness.^13,18 However, increasing film thickness always decreased the transparency. Alternatively, roughness was increased by adding additives into the deposition system, including alcohols and chelating reagents.^3,13 However, the thickness effect was not eliminated. Recently, we reported a significant increase in roughness R_a from 13 to 60 nm, the highest ever reported, particularly at a low thickness of 300 nm by adding NH₃ to the deposition process.²³ However, the conductivity of the FTO film decreased noticeably, as evidenced by increases of sheet resistance from 14 to 35 Ω □⁻¹ and resistivity from 4.2 × 10⁻⁴ to 10.5 × 10⁻⁴ Ω cm. The loss of conductivity could have a negative effect on photovoltaic applications, and thus should be suppressed.

Although great efforts have been made, effective methods for improving the roughness of FTO thin films without sacrificing transparency and conductivity are still lacking. In this study, we report the addition of HNO₃ into the system of atmospheric pressure chemical vapor deposition (APCVD) for improving roughness. Corresponding to these roughness increases, the resistivity and the transmittance maintained at a high level. The roughness enhancement is illustrated to be related to the development of a flower-like grain morphology on the film surface.

2 Experimental methods

2.1 Materials

Dimethyltin dichloride (DMTDC) and trifluoroacetic acid (TFAA) were used as Sn and F precursors, respectively. DMTDC (>98%) and TFAA (99.9%) were purchased from J&K Chemical Ltd. The additive of HNO₃ (65–68%) was purchased from Shanghai Ling Feng Chemical Reagent Co. All water used was deionized (DI) by Millipore system with resistance of 18 MΩ. Soda lime glass was supplied by Nippon Sheet Glass (NSG) Group. Before it was used as a substrate for FTO films, it was deposited with a layer of SiO₂ of 20 nm thick to impede the migration of alkali metal ions from the glass into the FTO layer.

2.2 Deposition of FTO thin films

FTO thin films were deposited by APCVD in a custom-built CVD reactor. DMTDC was vaporized in a bubbler heated at 145 °C and carried away by N₂ at a fixed flow rate of 0.98 L min⁻¹. For F doping, TFAA was dissolved in DI water at a concentration of 30 vol.%, and injected into a vaporizer heated at 200 °C. They were then carried away by N₂ at a fixed flow rate of 5 L min⁻¹. For morphology control, HNO₃ was also fed into this vaporizer by a pump at one of the feeding rates of 0.5, 1, 1.5, 1.8, and 2.25 mL min⁻¹. For Sn oxidation, in addition to the water mixed with other chemicals, O₂ was also supplied at a flow rate of 2.1 L min⁻¹ from a separate gas line. All chemicals and gases were mixed in a stainless steel pipe and picked up by the main carrier stream of N₂ at a flow rate of 4.5 L min⁻¹. This pipe was heated by heater tapes at 200 °C to avoid condensation of chemicals. For film deposition, the mixed chemicals were delivered to a glass substrate in the reactor in an approximately laminar flow mode. Prior to the delivery, the glass substrate was heated to 600 °C and stabilized at this temperature for at least 15 minutes. The deposition time was adjusted under all specific deposition conditions to control the thickness. When the deposition completed, the mixed gases were directed to an incinerator for decomposition of unused chemicals at 750 °C, and the sample was cooled to <150 °C before it was taken out from the reactor.

2.3 Characterization

Sheet resistance was measured with a SX1944 four-point probe meter with 0.5 mm tip radii and 1 mm tip space over a uniform area of coating. Film thickness was determined by W-VASE ellipse polarization at the wavelength from 193–1700 nm. The beam diameter was 1 mm. UV-VIS-NIR spectrophotometer (Varian, Cary 500) was used to investigate the optical properties. Transmission and reflection spectra were measured between 200–2500 nm in 1 nm steps. Transmission was measured with the coated side of the glass against the integrating sphere. The reflection from both sides of the glass was measured.

The surface morphology of FTO films was examined with a Hitachi High-Technology S-4800 field emission scanning electron microscope (FE-SEM) at an accelerating voltage of 5 kV. Atomic force microscopy (AFM) measurement was performed with Nanoscope IIIa Multimode AFM with the roughness analysis being conducted by the software of Nanoscope 5.30. The structures of FTO thin films were investigated by X-ray diffraction (XRD) at room temperature with a Rigaku D/max 2550 VB/PC apparatus using Cu Kα radiation (λ = 0.15406 nm) and a graphite monochromator, operated at 40 kV and 100 mA. Diffraction patterns were recorded in the angular range of 20–80° with a step width of 0.02° s⁻¹. An ION-TOF 5 Time of Flight Secondary Ion Mass Spectrometry (TOF-SIMS) instrument was used to obtain a compositional positive ion depth profile for several coatings. The analysis beam was Bi³⁺ and the sputter beam was 1 keV Cs⁺ with a beam current of 82+/−2 nA. The sputter beam was rastered over a 200 × 200 micron area, and the bismuth analysis beam was rastered over a 50 × 50 micron area at the centre of the sputtered region.

3 Results

3.1 Optical property

All the results are described for FTO thin films with the same thickness of about 300 nm unless otherwise noted. This is to minimize the possible effects of thickness on transmittance, grain size and surface roughness. At such a thickness, the film transmittance with or without HNO₃ addition was high in the visible region. Fig. 1 shows the typical results of the baseline sample for which no HNO₃ was included and those for which HNO₃ was added. All films with HNO₃ addition showed a transmittance higher than 80% in the visible region. The maximum transmittance was up to 90% at the wavelength of 550 nm. The transmittance varied with HNO₃ feeding rate over a small range. All these transmittance values included the absorption and reflection by the glass substrate as well. If the effect of the glass substrate was subtracted, the transmittance for the films themselves was higher than 90%.

Fig. 1 Typical transmission spectra of FTO thin films deposited with different amounts of HNO₃.

At the wavelength higher than 1200 nm, the transmittance began to decrease to be lower than 80% and decreased rapidly with increasing wavelength. The addition of HNO₃ appeared to reduce such decrease. With increasing the feeding rate of HNO₃, it was particularly evident that the decrease was slower than the one without HNO₃.

3.2 Structure analysis

With TFAA as the fluorine precursor, the presence of fluorine was checked by TOF-SIMS in the sample without any HNO₃ addition (Fig. 2a). The sputter time is related to the film thickness. It was seen that the doping level was relatively small. In addition, F was doped in the whole film and varied slightly with thickness.

Fig. 2 The amount of F doping in the baseline film (a), XRD pattern (b) and the texture coefficients of the main diffraction planes (c) for FTO thin films prepared with different additions of HNO₃.

XRD patterns for the baseline sample and those with addition of HNO₃ are shown in Fig. 2b. It is illustrated that SnO₂ crystals were polycrystalline with a rutile tetragonal structure. Peaks for SnO or Sn phases were not detected, indicating that the films were fully oxidized. No appreciable peak shift could be observed, which was reasonable considering that the doping level was relatively small for this system and the anionic radii were rather similar. This result also verified the small amount of F doping measured by TOF-SIMS. The major XRD peaks are associated with diffractions of (200), (110), and (211) planes, and the minor diffractions of (101), (301), and (310) planes are also indexed. The diffraction intensities of these planes varied with HNO₃ feeding rate. For the baseline sample, the diffraction of (200) plane was dominant. With increasing HNO₃ addition, the relative intensity of (200) plane showed an apparent increase first and then a decrease, but that of (211) plane a decrease first and then a gradual increase, suggesting a possible change in preferred orientation of the deposited film.

The feature of the preferred orientation of a thin film can be quantified by the texture coefficient (TC) which is calculated from the XRD of (hkl) plane as follows:²⁴

where I is the measured intensity, I₀ the ASTM standard intensity of SnO₂ power, and n the number of (hkl) diffraction peaks. From the definition, it is obvious that the deviation of TC from unity is indicative of the presence of a preferred orientation of the film along the diffraction plane, implying an enhancement in the number of grains along the plane.

The TC values for the major diffraction planes are depicted in Fig. 2c. It is clearly shown that the TC for (200) increased to a high value with 1.5 mL min⁻¹ HNO₃ addition but then decreased with more addition. In contrast, the TC for (110) and particularly that for (211) declined first and then increased. Especially, with 2.25 mL min⁻¹ HNO₃ addition, the TCs for (110) and (211) were much larger than that for TC (200). This led to that (110) and (211) planes substituted for (200) plane as the preferred orientation with increasing HNO₃ addition.

3.3 Surface morphology

Fig. 3 shows the surface morphology of FTO thin films with HNO₃ addition. The surface morphology of the baseline film was relatively uniform and compact, and most grains appeared to be irregular in shape with some being prismatic or pyramidal with a size of generally 100–150 nm (Fig. 3a). Stripes were observed on some grains, indicating the presence of twins. In the case of 1 mL min⁻¹ HNO₃, the grains had no significant difference from the baseline film except for the observation of several big grains, as shown in Fig. 3b.

Fig. 3 SEM surface morphologies of FTO thin films of different thicknesses prepared at different feeding rates of HNO₃. (a) 300 nm, 0 mL min⁻¹; (b) 300 nm, 1 mL min⁻¹; (c and d) 200 nm, 1.5 mL min⁻¹; (e and f) 300 nm, 1.5 mL min⁻¹; (g and h) 500 nm, 1.5 mL min⁻¹.

With more HNO₃ addition the surface morphology changed significantly. In the case of 1.5 mL min⁻¹ addition, flower-like grains appeared and dispersed on the surface of the film, while the background grains had no significant difference from the baseline sample except for smaller grain sizes (Fig. 3c–h). The size of the flower-like grains was around 300 nm in the lateral dimension, but the population of such grains appeared to decrease significantly in the films prepared with a thickness of 500 nm (Fig. 3e and f), compared with those prepared with a thickness of 200 nm (Fig. 3c and d) or 300 nm (Fig. 3e and f). The variation of the background grain size is attributed to the thickness change.

The surface morphology was also characterized by AFM measurements. Fig. 4 shows the AFM images for the films prepared with the same thickness of 300 nm but with different HNO₃ additions. From these images, the surface of the film prepared with 1.5 mL min⁻¹ HNO₃ addition illustrates the presence of abundant high peaks which correspond to the flower-like grains observed in SEM images. Such film had a R_a roughness as high as 51.4 nm (R_mrs: 70.5 nm) (Table 1). All other film surfaces showed the absence or only a few high peaks, leading to a much smaller roughness.

Fig. 4 AFM images of FTO thin films of 300 nm thick prepared at different feeding rates of HNO₃ (mL min⁻¹). (a) 0, (b) 0.5, (c) 1, (d) 1.5, (e) 1.8, and (f) 2.25.

Table 1 Roughness and electrical property data for FTO films prepared with the addition of HNO₃. Note that the resistivity values are calculated from the sheet resistance multiplied the film thickness

Feeding rate (mL min^−1)	Film thickness (nm)	Sheet resistance (Ω □^−1)	Resistivity × 10^−4 (Ω cm)	Ra (nm)	Rmrs (nm)
0	300	14	4.2	13	17
1	300	14	4.2	7.5	10.4
1.5	300	14	4.2	51.4	70.5
1.8	300	21	6.3	18.8	27.2
2.25	300	45	13.5	16.6	20.2
1.5	100	170	17	18.5	25.7
1.5	200	43	8.6	73.3	91.0
1.5	500	10	5	24.0	36.6

---

In order to investigate the growth process of the flower-like grains, the AFM surface morphologies of the films prepared with 1.5 mL min⁻¹ HNO₃ addition but with different thicknesses were examined (Fig. 5). It is revealed that the flower-like grains formed in the early stage of the deposition process. With increasing thickness, the size of such grains increased, but the number of such grains decreased beyond the thickness of 300 nm (Fig. 4d). Such a morphological variation generated a R_a roughness up to 73.3 nm (R_mrs: 91 nm) on the film of 200 nm thick but a R_a roughness of only 24 nm on the film of 500 nm (Table 1).

Fig. 5 AFM images of FTO thin films prepared at 1.5 mL min⁻¹ of HNO₃ with a thickness of 100 (a), 200 (b), and 500 nm (c). See Fig. 4d for the film prepared under the same condition with a thickness of 300 nm.

3.4 Electrical property

Without F doping, the sheet resistance was very high, which could be up to hundreds of ohm per square, and the resistivity up to 10⁻² Ω cm. But, after F was doped, the sheet resistance and resistivity decreased sharply. The results of sheet resistance R_s of all films are included in Table 1. At a thickness of 300 nm, R_s was similar to that of the baseline sample with the addition of up to 1.5 mL min⁻¹ HNO₃ (∼4.2 × 10⁻⁴ Ω cm). But more additions led to an increase of resistivity. With the same addition, the film thickness also influenced R_s; the thicker the film, the smaller the R_s.

4 Discussion

In the present study, the highest ever reported roughness was observed (R_a: 51.4 and 73.3 nm, and R_mrs: 70.5 and 91 nm) (Fig. 6a and b). The correlation of the high surface roughness with grain morphology is straightforward. That is, the high roughness values were provided mainly by the flower-like grains. Such grains were large-sized, extended out from fine background grains, and thus introduced big ups and downs on the film surface.

Fig. 6 Variations of roughness with film thickness (a and b) and resistivity (c and d).

Fig. 6a and b illustrates the variation of roughness with film thickness for typical previous and present FTO films. One of the unique features of the present roughness is that it was not only the highest among those ever reported but also it was observed for the films with a thickness of as low as 200 or 300 nm. This is different from the previous relatively high roughness within the range of 40–55 nm for the films (including commercial ones) with a thickness of >700 nm.^{13,15,18,25,26} The consequence is that the present very rough films had a transmittance value of 85–90% at the wavelength of 550 nm, which is much higher than those for previous thick films (<80%). Thus, improving the surface roughness was not in tradeoff with transmittance in the present study.

The further unique feature of the present roughness is that as a low amount of HNO₃ was used as the additive, the sheet resistance of the rough film remained at a low level. For example, the very rough film of 300 nm thick had a resistivity of 4.2 × 10⁻⁴ Ω cm, one of the lowest resistivity ever reported for FTO films (Fig. 6c and d). This observation indicates that although the surface morphology and roughness were changed, the carrier concentration and mobility which are responsible for electrical conductivity did not change very much.

The unique features of the present rough film are apparently related to the use of HNO₃ as the additive. Deposition of FTO thin films requires the presence of tin, fluorine and oxygen sources. It is the complex interplay between these components in the reactor that controls the film formation and deposition. The microstructures of thin films are determined in part by surface diffusion and nucleation processes on the growth interface. This in turn is affected by substrate temperature, reactor pressure, and gas-phase composition.²⁷ Additives, when they are introduced in different amounts, modify the reaction chemistry and flow rate, which leads to the changes of film formation, growth rate, surface morphology and roughness.

The addition of HNO₃ might have induced a special nucleation and crystal growth processes, which did not taken place with other additives before. HNO₃ is well known being a strong oxidizing acid, especially at the high temperature of 600 °C. In the presence of HCl and/or HF from DMTDC and TFAA, respectively, HNO₃ could have even caused corrosion of the substrate glass used for FTO film deposition. During nucleation, the heterogeneous nucleation was much more common than homogeneous nucleation. An uneven or corroded surface could certainly promote heterogeneous nucleation. The other factor to consider is that, in the deposition process, the organic impurities or precursors could be oxidized to lead to the formation of hydroxyl or carboxyl groups, which could cause the substrate surface more hydrophilic.^28,29 Since the nucleation was determined by the nucleation barrier which was proportional to the surface area, when the surface became hydrophilic, the surface area decreased, resulting in the decrease of nucleation barrier. In this case, the heterogeneous nucleation was also much faster than homogeneous nucleation.

It may be the heterogeneous nucleation that governed the formation of flower-like grains and their further growth leading to a rough surface. Direct evidence comes from the observation of the formation of a heterogeneous and uneven surface structure at a thickness of as low as 100 nm (Fig. 5a). As the film thickness increased to 500 nm, such a surface structure tended to diminish, resulting in an even surface (Fig. 3g and 5c). Apparently, because the heterogeneous structure and flower-like grains formed at the early stage of film growth, a rough surface developed when the thickness of the film was still very thin in the present deposition. In contrast, in most previous studies, high roughness resulted from grain growth and thus was usually seen on the surfaces of very thick films.

The present film with a rough surface had a low resistivity. This may be related to the development of (200) preferred orientation during the process of film growth with the addition of HNO₃ of up to 1.5 mL min⁻¹ (Fig. 2). A clear beneficial effect of (200) preferred orientation on conductivity was also observed in some previous studies.^8,15,30–33 The origin of preferred orientations can be explained by the periodic band chains (PBC) theory.^23,34 According to this theory, the different faces of SnO₂ can be classified into flat (F), stepped (S) and kinked (K) faces. For cassiterite SnO₂, F{101} and K{110} faces are of great interest. F{101} and K{110} faces are polar, which consist of only tin atoms or oxygen atoms, while F{110} faces are less polar because they are comprised of both tin and oxygen atoms. For F{101} faces, there are two possible preferred orientations, namely, (110) and (001). Since the (001) orientation was not observed by XRD, the (110) preferred orientation formed by the crystalline pyramidal F(101)-faces. This may explain the major (110) preferred orientation observed in the baseline sample (Fig. 2).

Under the condition with inclusion of HNO₃, F-faces having formed in the early stage may continue to grow, leading to prominent (200) growth perhaps for the following reason. HNO₃ molecules have strong polarity, and could adsorb on polar F(101) planes. This could restrain the (110) plane growth, but give rise to the (200) plane growth, leading to the SnO₂ crystals faceted exclusively by K faces with (200) preferred orientation, as previously reported for the favorable effect of the presence of halogen-rich gases on the development of (200) preferred orientation.^8,31–33 But, with the addition of HNO₃, the concentration of tin per unit volume was decreased, which was unfavorable for (200) growth which requires a high tin concentration.³⁰ So excess HNO₃ addition would slow down (200) plane growth, and benefit the growth of other planes. In addition, (211) plane also became the preferred orientation as the addition of HNO₃ increased. The higher density of (211) direction was thought to be characterized by low growth rates and believed to be favoured during initial thin film formation.³⁵ With HNO₃ increasing, the concentration of precursors per unit volume decreased, which would lead to the decrease of growth rate. So (211) plane became the preferred orientation gradually.

5 Conclusions

FTO thin films were prepared with the addition of HNO₃. HNO₃ addition resulted in a specific flower-like grain morphology which contributed to the highest ever reported roughness (R_a: 51.4–73.3 nm, and R_mrs: 70.5–91 nm). Due to the effect of HNO₃ on heterogeneous nucleation, the high roughness was obtained for the films with a thin thickness of 200–300 nm, thus having maintained the transparency in the high range of 80–85% in the visible region and even improved over that for the baseline sample in the infrared region. Perhaps as a result the beneficial effect of HNO₃ on the development of (200) preferred orientation, the achievement of high roughness did not sacrifice the original low sheet resistance of 14 Ω □⁻¹ and resistivity of 4.2 × 10⁻⁴ Ω cm at the thickness of 300 nm. The high roughness, combined with high transmittance and low resistivity would improve the light trapping of FTO films and thus the efficiency of solar cells.

Acknowledgements

Financial supports from Pilkington Technology Management Limited, National Natural Science Foundation of China (project #: 51271077 and U1362104) and Shanghai Nanoscience and Nanotechnology Promotion Center (project #: 12nm0503300) are greatly acknowledged.

Notes and references

1. X. G. Han, M. S. Jin, S. F. Xie, Q. Kuang, Z. Y. Jiang, Y. Q. Jiang, Z. X. Xie and L. S. Zheng, Angew. Chem., Int. Ed., 2009, 48, 9180–9183.
2. X. X. Xu, J. Zhuang and X. Wang, J. Am. Chem. Soc., 2008, 130, 12527–12535.
3. Y.-F. Wang, B.-X. Lei, Y.-F. Hou, W.-X. Zhao, C.-L. Liang, C.-Y. Su and D.-B. Kuang, Inorg. Chem., 2010, 49, 1679–1686.
4. Z. D. Li, Y. Zhou, T. Yu, J. G. Liu and Z. G. Zou, CrystEngComm, 2012, 14, 6462–6468.
5. H.-Y. Chen, D.-B. Kuang and C.-Y. Su, J. Mater. Chem., 2012, 22, 15475–15489.
6. G. S. Chae, H. S. Soh, W. H. Lee and J. G. Lee, J. Appl. Phys., 2001, 90, 411–415.
7. I. Hamberg and C. G. Granqvist, J. Appl. Phys., 1986, 60, R123–R160.
8. I. Volintiru, A. de Graaf, J. van Deelen and P. Poodt, Thin Solid Films, 2011, 519, 6258–6263.
9. X. Díez-Betriu, R. Jiménez-Rioboo, J. Sánchez-Marcos, E. Céspedes, A. Espinosa and A. de Andrés, J. Alloys Compd., 2012, 536, S445–S449.
10. K. Ellmer, Nat. Photonics, 2012, 6, 809–817.
11. A. Poruba, A. Fejfar, Z. Remeš, J. Špringer, M. Vaněček, J. Kočka, J. Meier, P. Torres and A. Shah, J. Appl. Phys., 2000, 88, 148–160.
12. A. M. B. van Mol, Y. Chae, A. H. McDaniel and M. D. Allendorf, Thin Solid Films, 2006, 502, 72–78.
13. H. M. Yates, P. Evans, D. W. Sheel, Z. Remes and M. Vanecek, Thin Solid Films, 2010, 519, 1334–1340.
14. D. Sheel, H. Yates, P. Evans, U. Dagkaldiran, A. Gordijn, F. Finger, Z. Remes and M. Vanecek, Thin Solid Films, 2009, 517, 3061–3065.
15. C. Agashe, J. Hüpkes, G. Schpe and M. Berginski, Sol. Energy Mater. Sol. Cells, 2009, 93, 1256–1262.
16. M. N. van den Donker, A. Gordijn, H. Stiebig, F. Finger, B. Rech, B. Stannowski, R. Bartl, E. A. G. Hamers, R. Schlatmann and G. J. Jongerden, Sol. Energy Mater. Sol. Cells, 2007, 91, 572–580.
17. T. Tohsopon, J. Hupkes, H. Siekmann, B. Rech, M. Schultheis and N. Sirikulrat, Thin Solid Films, 2008, 516, 4628–4632.
18. Q. Gao, H. Jiang, C. J. Li, Y. P. Ma, X. Li, Z. H. Ren, Y. Liu, C. L. Song and G. R. Han, J. Alloys Compd., 2013, 574, 427–431.
19. J. Springer, B. Rech, W. Reetz, J. Muller and M. Vanecek, Sol. Energy Mater. Sol. Cells, 2005, 85, 1–11.
20. J. Springer, A. Poruba and M. Vanecek, J. Appl. Phys., 2004, 96, 5329–5337.
21. C.-H. Tsai, S.-Y. Hsu, T.-W. Huang, Y.-T. Tsai, Y.-F. Chen, H. J. Yuan, H. Lun, C.-C. Wu, Y.-S. Chen, C.-W. Chen and C.-C. Li, Org. Electron., 2011, 12, 2003–2011.
22. J. Müller, B. Rech, J. Springer and M. Vanecek, Sol. Energy, 2004, 77, 917–930.
23. J. T. Wang, X. L. Shi, X. H. Zhong, J. N. Wang, L. Pyrah, K. D. Sanderson, P. M. Ramsey, M. Hirata and K. Tsuri, Sol. Energy Mater. Sol. Cells, 2015, 132, 578–588.
24. C. Barret and T. B. Massalki, in Structure of Metals, ed. C. Barret and T. B. Massalki, Pergamon, Oxford, 1980, pp. 204–205.
25. Z. Remes, M. Vanecek, H. M. Yates, P. Evans and D. W. Sheel, Thin Solid Films, 2009, 517, 6287–6289.
26. A. V. Moholkar, S. M. Pawar, K. Y. Rajpure, S. N. Almari, P. S. Patil and C. H. Bhosale, Sol. Energy Mater. Sol. Cells, 2008, 92, 1439–1444.
27. A. M. B. van Mol, M. H. J. M. de Croon, C. I. M. A. Spee and J. C. Schouten, J. Phys. IV France, 1999, 09, 165–172.
28. Y. Masuda, T. Ohji and K. Kato, ACS Appl. Mater. Interfaces, 2012, 4, 1666–1674.
29. H.-W. Tien, Y.-L. Huang, S.-Y. Yang, S.-T. Hsiao, W.-H. Liao, H.-M. Li, Y.-S. Wang, J.-Y. Wang and C.-C. Ma, J. Mater. Chem., 2012, 22, 2545–2552.
30. C. Agashe, M. Takwale, V. Bhide, S. Mahamuni and S. K. Kulkarni, J. Appl. Phys., 1991, 70, 7382–7386.
31. A. Smith, J. M. Laurent, D. S. Smith, J.-P. Bonnet and R. R. Clemente, Thin Solid Films, 1995, 266, 20–30.
32. R. Y. Korotkov, P. Ricou and A. J. E. Farran, Thin Solid Films, 2006, 502, 79–87.
33. G. Gordillo, L. C. Moreno, W. D. L. Cruz and P. Teheran, Thin Solid Films, 1994, 252, 61–66.
34. C.-Y. Kim and D.-H. Riu, Thin Solid Films, 2011, 519, 3081–3085.
35. N. Noor and I. P. Parkin, J. Mater. Chem. C, 2013, 1, 984–996.

---