HF-assisted one-step synthesis of pompon-like/chip-like FeSe₂ particles and their superamphiphobic/antireflective property

Abstract

Hydrofluoric acid (HF)-assisted one-step synthesis of pompon-like/chip-like FeSe₂ particles by a solvothermal method has been studied for the first time in this paper. By adjusting the dosage of HF used, FeSe₂ particles with pompon-like and chip-like morphologies were obtained. The reaction mechanism was presented based on the experimental phenomena, and the roles played by HF in the synthesis were clearly explained, proving that HF had a significant influence on controlling the morphologies of the FeSe₂ particles. By altering the iron sources without changing any other conditions (similar experimental results were observed). We further demonstrated that this method was universally applicable. The wettability and light-trapping effects of the pompon-like/chip-like particles were also investigated, respectively. After modification with heptadecafluorodecyltrimethoxy-silane (HTMS), the pompon-like particles exhibited excellent superhydrophobic/superoleophobic (superamphiphobic) properties with water/oil contact angle of about 156.0°/154.6° and water/oil sliding angle of about 2.0°/5.0°. Ultralow reflectance of the samples (lower than 3%) in the entire wavelength range of 300–1800 nm was also observed in our experiments. Utilizing HF to control the morphologies of FeSe₂ particles is an innovative attempt and is expected to show promising potential in controlling the morphologies of other transition metal chalcogenides.

---

Introduction

In recent years, transition metal chalcogenides have sparked worldwide research interest owing to their tunable optical, electrical and magnetic properties.^1–8 Among these chalcogenides, iron selenide (FeSe₂), as a p-type chalcogenides semiconductor, has promising and prospective applications in solar energy fields on the account of its narrow band gap energy (E_g = 1.0 eV) and high optical absorption coefficient.^1,9–11 There are several ways to synthesize FeSe₂ such as hydrothermal synthesis,¹² solvothermal synthesis,^1,9,13 soft selenization of iron films,^14,15 electrodeposition method,¹¹ etc. FeSe₂ synthesized by different routes has diverse morphologies and sizes, which play important roles in their physical properties.^9,11,12 Although some unique morphological FeSe₂ compounds have been reported with reasonable physical properties, it is far from enough for researchers to understand and utilize them. Therefore, we need to seek new methods to synthesize FeSe₂ with various morphologies to yield better physical performances.

Hydrofluoric acid (HF) has been proven to be effective in changing the morphology of ZnO by Yang et al. with little dosage¹⁶ but the applicability of HF in controlling the morphology of FeSe₂ is still unknown. Inspired by this view, we introduced HF in the reaction and successfully synthesized pompon-like FeSe₂ particles by a one-step method, i.e. heating the mixture of ferrous chloride (FeCl₂·4H₂O), selenium powder (Se), oleylamine (OLA) and moderate HF at 200 °C. Then, we increased the amount of HF without changing the other experimental conditions and found that the shape of FeSe₂ started to turn from pompon-like particles to chip-like slices. When an excess HF was added, particles finally transformed into slices. Based on these morphological changes, we proposed a reaction mechanism and legitimately explained the roles that HF played in the synthesis, proving that HF had a considerable impact on controlling the morphologies of FeSe₂ particles. Then, the same experimental phenomena were observed when replacing FeCl₂·4H₂O with ferrous sulfate (FeSO₄·7H₂O), further implying that this method possesses wide applicability.

Wettability is a very important physical property of surface materials. Materials with water/oil contact angle (CA) larger than 150° and sliding angle (SA) less than 10° are usually regarded as superhydrophobic/superoleophobic (superamphiphobic) materials.¹⁷ Two major properties are required for fine superamphiphobic performance, namely, surface roughness and low surface energy.^18,19 Therefore, there are two routes to obtain superamphiphobic materials: (1) directly building rough structures by low-surface-energy materials^20,21 and (2) constructing rough structures by other materials and modifying these structures with low-surface-energy chemicals.^16,22–26 Among these two methods, the latter is more important because of its ability to switch the non-superamphiphobic materials to superamphiphobic materials. This method was also adopted in our experiments by modifying the pompon-like FeSe₂ particles with heptadecafluorodecyltrimethoxy-silane (HTMS). After modification, the pompon-like FeSe₂ particles possessed excellent superamphiphobic property with water/oil CA of about 156.0°/154.6° and water/oil SA of about 2.0°/5.0°; moreover, it was also found that the superhydrophobic property can be effectively maintained in a wide pH scope (1–14). These properties endowed the particles with an effective self-cleaning function and corrosion resistance. Finally, the light-trapping effect of these particles was studied and ultralow reflectance (lower than 3%) was observed in the wavelength range of 300–1800 nm, guaranteeing that the FeSe₂ particles yielded sufficient light absorption.

Our approach is a creative exploration in controlling the morphology of FeSe₂ and may be used in the synthesis of other transition metal chalcogenides. Moreover, the low price and low dosage of HF make this method economical and efficient for extensive FeSe₂ synthesis in the future.

Experimental

Materials

Selenium powder (Se, AR) was purchased from Shanghai Meixing Chemical Reagent Co., China. Oleylamine (OLA, 70%) was purchased from Sigma-Aldrich (USA), and heptadecafluorodecyltrimethoxy-silane (HTMS) was purchased from Dow Corning Co (USA). All the other reagents, including ferrous sulfate (FeSO₄·7H₂O, AR), ferrous chloride (FeCl₂·4H₂O, AR), hydrofluoric acid (HF, 40 wt.%, AR), absolute alcohol (AR), hexahydrobenzene (AR), toluene (AR), ethylene glycol (AR), diiodomethane (AR) and formamide (AR), were obtained from Sinopharm Chemical Reagent Co., Ltd. Cover glasses (22 mm × 22 mm × 0.15 mm) were obtained from Jiangsu Feizhou Plastic Product Co., China.

Synthesis of FeSe₂ particles

Solution 0 was prepared by stirring a mixture of OLA (5 mL), FeCl₂·4H₂O (0.0318 g) and Se (0.0505 g), which were held in a vial for 30 min. The preparations of solutions 1–4 were almost the same as that of solution 0, except that different amounts of HF (10 μL, 50 μL, 100 μL, and 200 μL, respectively) were added into the mixed solution after 30 min stirring with additional stirring for 2 min. All these solutions were heated at 200 °C for 1 h with continuous stirring. After the heat treatment, the solutions were cooled to room temperature naturally. Black precipitates were obtained by centrifugation at 8500 rpm for 5 min, and were refined by washing them with a solution of alcohol/hexahydrobenzene (4/1 vol) for 5 times. Finally, the purified particles were dried in vacuum overnight for further characterization. For a better description, these particles were named as samples 0–4 corresponding to solutions 0–4, respectively.

Surface modification

One-third of every sample was mixed with absolute alcohol (500 μL) to form the rheumy ‘coating’; these ‘coatings’ were then uniformly smeared on cover glasses (22 mm × 22 mm × 0.15 mm) and dried at room temperature. HTMS (150 μL) and toluene (3 mL) were added into a beaker followed by a 1 min ultrasonic treatment to make the solution uniform. The beaker and all the samples were placed in a sealed case, which was then heated in a drying oven at 90 °C for 4 h. After heat treatment, the case was opened in a fume cupboard and the samples were naturally cooled to room temperature.

Characterization

The microstructures of the FeSe₂ particles were demonstrated by the scanning electron microscopy (SEM) images, which were obtained on a field emission scanning electron microscope (JSM-6700F, 5.0 kV). Transmission electron microscopy (TEM) images and high-resolution transmission electron microscopy (HRTEM) images, which were used to analyze the nanostructure and crystallization direction of the FeSe₂ ‘petal’, were obtained on a high-resolution transmission electron microscope (JEM-2011). X-ray diffraction patterns (2θ ranges from 20° to 70°) obtained using an X-ray diffractometer (TTR-III) were used to analyze the phase, purity and crystallization of the samples. Contact Angle Tester SL200B (Solon Tech (Shanghai) Co., Ltd.) and its relevant software (CAST 2.0) were used to test the CAs and SAs of the samples with 4.0 μL water/oil droplets on the surfaces of the coated glasses. To investigate the corrosion resistance of the pompon-like particles, water droplets with different pH values were used. All the values of CAs and SAs in this paper are averages, which were obtained by measuring five times in different regions of the surface. Diffuse reflectance spectra, which could characterize the light-trapping effect of the particles, were obtained using SolidSpec-3700 UV-VIS-NIR Spectrophotometer (SHIMADZU).

Results and discussion

Crystallization and phase

Fig. 1 presents the XRD patterns of samples 0–4 in the 20–70° 2θ range. It is observed that all the peaks in the patterns correspond well with the standard diffraction peaks (PDF#65-1455), indicating that all these samples are orthorhombic iron selenide. Moreover, without impurity peaks (such as FeSe, FeCl₂ or Se) being observed, the XRD patterns demonstrate the high purity and crystallization of the particles.

Fig. 1 XRD patterns of FeSe₂ particles (samples 0–4).

Morphologies and growth mechanism

Fig. 2(a) and (b) exhibit the SEM images of sample 0 and sample 1, respectively, and the insets in the bottom left corners are the corresponding magnified images. It is observed that sample 0 shows a bush-like lumpish structure with branch diameters ranging from 60 nm to 110 nm and lengths ranging from 200 nm to 400 nm, while sample 1 shows a regular pompon-like spherical structure with an average sphere diameter of 1.56 μm. On comparing sample 0 with sample 1, it is found that on one hand, the dispersive ‘pompons’ are obviously smaller than the bush-like chunks, while on the other hand, the rod-like ‘petals’ of sample 1, whose lengths range from 164 nm to 251 nm and widths range from 119 nm to 234 nm, are shorter and wider than those of sample 0. Considering that the other experimental conditions are unchanged except for the use of HF, it is reasonable to speculate that HF plays a significant role in the formation of FeSe₂ particles. To clearly understand this influence mechanism, we altered the HF dosage without changing any other conditions and observed the morphological changes of FeSe₂ to investigate the relationship between them.

Fig. 2 (a)–(e) are the SEM images of samples 0–4, respectively; insets in the bottom left corners of (a) and (b) are the corresponding magnified images, and insets in the top right corners of (a)–(e) are the corresponding water CA images, where all the droplets are 4.0 μL with pH = 7, (f) is the water CA and SA variation diagram of different samples, (g) is the TEM image of FeSe₂ ‘petal’ stripped from sample 1, (h) is the corresponding HRTEM image of (g).

Fig. 2(c)–(e) show the morphological variations of FeSe₂ particles when different amounts of HF are used: (c)–(e) correspond to samples 2–4, where the HF dosages are 50 μL, 100 μL and 200 μL, respectively. In Fig. 2(b), regularly dispersive ‘pompons’ are the main existing forms without any morphologies being observed but in Fig. 2(c), when 50 μL HF is used, chips begin to appear. On increasing the amount of HF to 100 μL, the number of ‘pompons’ decrease obviously in contrast with sample 2, and chips occupy most part of FeSe₂. Finally, when HF is increased to 200 μL, stacked chips become the dominant morphology. Combining these morphological variations with the results obtained from Fig. 2(a) and (b), it is fairly reasonable to believe that HF plays an important role in the synthesis of FeSe₂ particles. Fig. 2(g) and (h) are the TEM and corresponding HRTEM images of FeSe₂ ‘petal’ stripped from sample 1, respectively. The ‘petal’ in (g) shows the same size as that observed in (b). Clear lattice points are observed in the HRTEM image, intimating excellent crystallization of the FeSe₂ particles, which corresponds well with the results obtained from the abovementioned XRD pattern. Two kinds of mutually perpendicular fringes can be observed, where the plane intervals of different fringes are 3.60 Å and 2.46 Å. Combining with the values provided by powder diffraction file (PDF#65-1455), it is reasonable to consider that the two fringes correspond to (001) (3.58 Å) and (120) (2.48 Å) lattice planes, respectively. To further explain these results, we propose the following suppositions.

Presumably, at the beginning of the reaction, FeSe₂ seed crystals are formed as the original existing form in solution. Then, these seed crystals experience two synchronous processes: (1) growth of the seed crystals and (2) self-assembly among different seed crystals. Process (1) makes these seed crystals grow into nanorods and process (2) makes these nanorods to form bush-like structures. It is noteworthy that without HF in solution, process (2) is dominant when compared with process (1), which can be confirmed in Fig. 2(a), where a bush-like lumpish structure formed through process (2) is the main morphology and rod-like ‘petals’ formed through process (1) is a subordinate morphology. When HF is added, the seed crystals will adsorb some F⁻ ions on their surfaces, which will restrain both the growth and the self-assembly of the seed crystals. Affected by this suppression, the size of sample 1 is smaller than that of sample 0. Moreover, on the basis of the fact that the ‘petals’ of sample 1 are shorter and wider when compared with those of sample 0, we infer that radial growth is restrained more than transverse growth. When the dosage of HF is increased, restraint effects are enhanced; moreover, the restraint to process (1) is weaker than process (2), thus finally this restraint yields two results, i.e. the seed crystals will grow alone rather than aggregate together and the seed crystals will grow into ‘chips’ due to radial-growth restrain. Therefore, when 200 μL HF is used, there are no ‘pompons’ in sample 4 except for the stacked ‘chips’.

Fig. 3 simply demonstrates the growth mechanism of FeSe₂ particles discussed above.

Fig. 3 The schematic illustration of the morphological changes with dosages of HF.

For further checking our results, we replaced FeCl₂·4H₂O by FeSO₄·7H₂O without changing other conditions (molar ratio of Fe to Se is 4 : 1), except that the temperature increased to 220 °C. Fig. 4 exhibits the morphological variations of FeSe₂ particles synthesized by FeSO₄·7H₂O; Fig. 4(a)–(d) are the FeSe₂ particles synthesized with 0 μL, 10 μL, 50 μL and 200 μL HF, respectively. In Fig. 4, basically, the same phenomena as that in Fig. 3 are observed, i.e. moderate HF (10 μL) helps FeSe₂ particles to form more regular pompon-like particles and excess HF (50 μL or 200 μL) makes the structure change from ‘pompons’ to ‘chips’. An interesting thing is that the ‘pompons’ formed by FeSO₄·7H₂O (average sphere diameter is 670 nm) are smaller than those formed by FeCl₂·4H₂O by comparing Fig. 3(a) with Fig. 4(b). This is presumably caused by different iron sources, which are also influential in the formation of FeSe₂ particles.¹ Discrepancy between Fig. 2(a) and 4(a) also reflects this presumption to a certain extent.

Fig. 4 SEM images of FeSe₂ particles synthesized by FeSO₄·7H₂O with different amounts of HF: (a) 0 μL, (b) 10 μL, (c) 50 μL, (d) 200 μL.

So far, from the review of the experimental results provided above, it is reasonable to conclude that adjusting the amount of HF in the reaction can control the morphologies of FeSe₂.

Wettability

Wettability is a very important property of materials, which is closely related to their morphologies and has several applications, for example, in surface protection. Therefore, the wettability of FeSe₂ particles was investigated in this paper.

First, we tested the water CAs and SAs of samples 0–4 after modifying with HTMS. Insets in the top right corners of Fig. 2(a)–(e) show the corresponding water CA images of the five samples. CA and SA variations are shown in Fig. 2(f), and the snapshots of the water droplets rolling off on the slant surfaces of samples 0–1 are exhibited in Fig. 5. All the water droplets here are 4.0 μL with pH = 7, and all the measured values are averages, which were tested at five different points on the same sample. From these data, it is observed that both sample 0 and sample 1 are superhydrophobic, and between them the latter one has a larger CA (156.0°) and lesser SA (2.0°) than the former (153.2° and 3.6°, respectively), implying that the FeSe₂ particles synthesized with moderate HF exhibit better superhydrophobicity. However, when more HF is added (50 μL, 100 μL, 200 μL), the CA/SA increases/decreases obviously, indicating that the excess HF destroys the superhydrophobic property of FeSe₂ particles.

Fig. 5 Snapshots of water droplets (4.0 μL, pH = 7) rolling off on the slant surfaces of different samples: (a₁)–(a₄) are those for sample 0, (b₁)–(b₄) are those for sample 1. Subscript (1)–(4) stand for different shooting times: (1) 0 s, (2) 0.042 s, (3) 0.084 s, (4) 0.126 s.

These CA and SA variations can be ascribed to the morphological changes of the FeSe₂ particles. As we described earlier, when moderate amounts of HF are used, FeSe₂ particles change from bush-like chunks to regular pompon-like particles, which possess better micro–nano structures, and are therefore more beneficial to the superhydrophobic property.^27–30 However, when more HF is added, the transformation from ‘pompons’ to ‘chips’ occurs, resulting in a micro–nano structure transforming into a micro structure and reducing the surface roughness of the particles. According to the liquid contact theory, this alteration makes the liquid-surface contact model change from the Cassie model to the Wenzel model. Numerous studies have shown that the former model is much better than the latter one with regard to its superhydrophobic performance, and therefore, the CA/SA shows a decreasing/increasing trend with an increase in the HF amount.

Second, we investigated the role played by the superhydrophobic property of particles in corrosion resistance by altering the pH values of the water droplets, and then comparing the changes in CA and SA. Sample 1 was chosen as the experimental object because of its excellent superhydrophobic property. As shown in the results in Fig. 6, it is observed that although the acidity or alkalinity of the water droplets is altered, the water CAs/SAs of sample 1 are almost unchanged, except for some slight floatation that can be considered normal. This is a very important property for our pompon-like particles, because it ensures their superhydrophobicity in a large pH range, which makes the sample capable of coping with various extreme and difficult conditions for their corrosion resistance.

Fig. 6 Water CAs and SAs variations of sample 1 for different pH values.

Third, the superoleophobic property of the particles was investigated, where sample 1 was also chosen as the experimental object. Three kinds of oils, including ethylene glycol, diiodomethane and formamide, were used in this test for investigating the universality of the superoleophobic property. Fig. 7 exhibits different oil CAs and SAs of sample 1, and the insets are the corresponding oil CA images. From this figure, it is found that the FeSe₂ particles can always maintain oil CAs larger than 150.0° and SAs less than 10° for these three types of oil droplets, implying that the modified particles are sufficiently superoleophobic to some extent. Comparing these data with the water CA and SA (pH = 7) of sample 1, it is found that the superoleophobic property is slightly worse than the superhydrophobic property. This difference is mainly caused by the lower surface energy of the oil in contrast with water.

Fig. 7 Oil CAs and SAs of sample 1 with different kinds of oils.

The time stability of the superhydrophobicity of sample 1 was also investigated in this paper, and the results are shown in Fig. 8. The water CAs and SAs were measured every month, and it is observed that the CAs and SAs slightly decrease and increase, respectively, as time progressed during the last year, except for some narrow floats at individual data points. This performance indicates that the superhydrophobicity of our particles decreases when they are subjected to long-term storage, which may be caused by the slight degeneration of the HTMS when it is exposed to air for a long time. However, for the worst superhydrophobicity (measured at the twelfth month), the water CA and SA are still in the superhydrophobic scope (water CA: 154.0°; water SA: 3.7°). Moreover, when comparing the worst water CA/SA with the best, it is found that the difference between them is very small. Therefore, it is still reasonable to believe that our superhydrophobic particles possess good time stability.

Fig. 8 Water CAs and SAs variations of sample 1 over one year.

Light-trapping effect

Finally, we investigated the light-trapping effect considering that FeSe₂ would be used in solar cells. The diffuse reflectance of different samples is shown in Fig. 9(a), where we can observe that all the samples possess low reflectance. Among these samples, the reflectance of samples 0–1 are significantly lower in contrast with the other samples, and the reflectance of samples 2–4 gradually rise with an increase in the amount of HF. The appearance of ‘chips’ makes the surface flat, which is beneficial for reflecting light; therefore, from sample 2 to sample 4, the reflectance increases in sequence due to the increased number of ‘chips’. When compared with the relatively flat structures of samples 2–4, the low reflectance of samples 0–1 can be ascribed to their rough structures (bush-like structure and pompon-like structure), which can result in the repeated reflection of incident light between different particles or between different ‘petals’ onto a single particle. Fig. 9(b) and (c) describe the two different reflection schematics.

Fig. 9 (a) Diffuse reflectance of different samples with wavelength ranging from 300 nm to 1800 nm. (b) and (c) show two different reflection schematics of FeSe₂.

Conclusions

In summary, we utilized HF to successfully synthesize pompon-like and chip-like FeSe₂ particles for the first time. We proposed a reaction mechanism of HF and demonstrated that HF had a significant impact on controlling the morphology of FeSe₂ particles. Moreover, we further proved that this method was suitable for other iron sources by replacing FeCl₂·4H₂O with FeSO₄·7H₂O. For physical performances, the wettability and light-trapping effect of the FeSe₂ particles were tested. After modification with HTMS, the pompon-like FeSe₂ particles were proven to be not only superhydrophobic but also superoleophobic. Excellent antireflection in the wavelength range of 300 nm to 1800 nm were also found on these pompon-like particles, guaranteeing the particles sufficient light absorptions in applications related to solar cells. Our approach is a creative attempt and will show promising potential for controlling the morphologies of other transition metal chalcogenides.

Acknowledgements

We sincerely thank the National Natural Science Foundation of China (no. 51272246) and Scientific and Technological Research Foundation of Anhui (no. 12010202035) for financial support.

Notes and references

1. B. X. Yuan, W. L. Luan and S. T. Tu, Dalton Trans., 2012, 41, 772–776.
2. K. B. Tang, Y. T. Qian, J. H. Zeng and X. G. Yang, Adv. Mater., 2003, 15, 448–450.
3. J. Park, K. J. An, Y. S. Hwang, J. G. Park, H. J. Noh, J. Y. Kim, J. H. Park, N. M. Hwang and T. Hyeon, Nat. Mater., 2004, 3, 891–895.
4. M. Akhtar, J. Akhtar, M. A. Malik, F. Tuna, M. Helliwell and P. O'Brien, J. Mater. Chem., 2012, 22, 14970–14975.
5. T. Harada, J. Phys. Soc. Jpn., 1998, 67, 1352–1358.
6. S. Thanikilkarasan, T. Mahalingam, K. Sundaram, A. Kathalingam, Y. D. Kim and T. Kim, Vacuum, 2009, 83, 1066–1072.
7. A. Kathalingam, T. Mahalingam and C. Sanjeeviraja, Mater. Chem. Phys., 2007, 106, 215–221.
8. H. H. Chen, R. J. Zou, N. Wang, H. H. Chen, Z. Y. Zhang, Y. G. Sun, L. Yu, Q. W. Tian, Z. G. Chen and J. Q. Hu, J. Mater. Chem., 2011, 21, 3053–3059.
9. J. Xu, K. Jang, J. Lee, H. J. Kim, J. Jeong, J. G. Park and S. U. Son, Cryst. Growth Des., 2011, 11, 2707–2710.
10. H. J. Kwon, S. Thanikaikarasan, T. Mahalingam, K. H. Park, C. Sanjeeviraja and Y. D. Kim, J. Mater. Sci.: Mater. Electron., 2008, 19, 1086–1091.
11. T. Mahalingam, S. Thanikaikarasan, R. Chandramohan, M. Raja, C. Sanjeeviraja, J. H. Kim and Y. D. Kim, Mater. Chem. Phys., 2007, 106, 369–374.
12. A. P. Liu, X. Y. Chen, Z. J. Zhang, Y. Jiang and C. W. Shi, Solid State Commun., 2006, 138, 538–541.
13. J. Yang, G. H. Cheng, J. H. Zeng, S. H. Yu, X. M. Liu and Y. T. Qian, Chem. Mater., 2001, 13, 848–853.
14. N. Hamdadou, A. Khelil, M. Morsli and J. C. Bernade, Vacuum, 2005, 77, 151–156.
15. B. Ouertani, J. Ouerfelli, M. Saadoun, M. Zribi, M. Ben Rabha, B. Bessais and H. Ezzaouia, Thin Solid Films, 2006, 511, 457–462.
16. Z. Yang, X. L. Xu, M. G. Gong, L. Liu and Y. S. Liu, Chin. Phys. B, 2010, 19.
17. K. S. Liu and L. Jiang, Nanoscale, 2011, 3, 825–838.
18. A. Nakajima, K. Hashimoto, T. Watanabe, K. Takai, G. Yamauchi and A. Fujishima, Langmuir, 2000, 16, 7044–7047.
19. W. Ming, D. Wu, R. van Benthem and G. de With, Nano Lett., 2005, 5, 2298–2301.
20. Y. Zhang, S. Wei, F. Liu, Y. Du, S. Liu, Y. Ji, T. Yokoi, T. Tatsumi and F.-S. Xiao, Nano Today, 2009, 4, 135–142.
21. D. Nystrom, J. Lindqvist, E. Ostmark, A. Hult and E. Malmstrom, Chem. Commun., 2006, 3594–3596.
22. M. G. Gong, X. L. Xu, Z. Yang, Y. Y. Liu, H. F. Lv and L. Lv, Nanotechnology, 2009, 20.
23. Z. W. Wang, Q. Li, Z. X. She, F. N. Chen and L. Q. Li, J. Mater. Chem., 2012, 22, 4097–4105.
24. D. Z. Xu, M. Z. Wang, X. W. Ge, M. H. W. Lam and X. P. Ge, J. Mater. Chem., 2012, 22, 5784–5791.
25. M. Wang, C. Chen, J. P. Ma and J. Xu, J. Mater. Chem., 2011, 21, 6962–6967.
26. C. Chen, J. Xu, Q. H. Zhang, Y. F. Ma, L. P. Zhou and M. Wang, Chem. Commun., 2011, 47, 1336–1338.
27. Y. B. Zhang, Y. Chen, L. Shi, J. Li and Z. G. Guo, J. Mater. Chem., 2012, 22, 799–815.
28. L. Feng, S. H. Li, Y. S. Li, H. J. Li, L. J. Zhang, J. Zhai, Y. L. Song, B. Q. Liu, L. Jiang and D. B. Zhu, Adv. Mater., 2002, 14, 1857–1860.
29. Z. G. Guo, F. B. Han, L. B. Wang and W. M. Liu, Thin Solid Films, 2007, 515, 7190–7194.
30. Z. G. Guo, F. Zhou, J. C. Hao and W. M. Liu, J. Am. Chem. Soc., 2005, 127, 15670–15671.

---