A nanotubular coating with both high transparency and healable superhydrophobic self-cleaning properties

Abstract

Transparent superhydrophobic coatings for the protection of material surfaces are becoming more and more important along with the rapid development of intelligent electronic displays. Here we present an easily fabricated, transparent and healable superhydrophobic self-cleaning coating. The polyaniline nanofiber network is coated with a silica shell by chemical vapor deposition of tetraethoxysilane catalyzed by ammonia. A porous silica nanotubular coating is obtained after calcination. By releasing perfluorooctyl acid packed in the pores of nanotubes at mild conditions, healable self-cleaning performance of the coating is achieved. The coating has high transparency for reading letters and pictures beneath the coated glass plate.

---

1. Introduction

The construction of superhydrophobic surfaces, based on roughness and low surface energy, is of great technological importance for surface self-cleaning performance, because water drops on these surfaces can roll off easily at a small tilt angle and remove dirt.^1–9 Owing to the increasing level of attention focused on transparent optical devices, there is an emerging need for transparent self-cleaning coatings enabling us to protect device surfaces from being contaminated. Many efforts have been made towards fabricating transparent superhydrophobic self-cleaning surfaces. One of these efforts is to achieve good transparency for superhydrophobic coating. Superhydrophobicity of a surface is based on the combination of nano–micro complex structure and low surface energy. The scale of micrometer structure is larger than that of visible light wavelength and the surface transparency is reduced owing to the scattering of propagated light. To increase light transmittance, coatings with porous nano–micro complex structure were fabricated and the thickness of shell was less than that of light wavelength.^10–16 These porous coatings improved the transparency by reducing the scattering of light. Another efforts is to achieve healable superhydrophobicity. Unlike coatings in nature, artificial surfaces can not generate special self-healing material in their body to restore its superhydrophobicity once their surfaces were destroyed.¹⁷ Mimicking self-healing behavior of natural surfaces lead to the development of many artificial healable superhydrophobic surfaces by the trigger of heating,^18–24 humidities,^25–28 mechanical force^29,30 and solution.³¹ A rough anodized alumina surface with a large number of nanopores for packing perfluorooctyl acid (PFA) could consecutively release PFA and heal the superhydrophobicity.¹⁸ Upon chemical vapor disposition of 1H,1H,2H,2H-perfluorooctyltriethoxysilane on the layer by layer assembled polymeric porous coatings with micro- and nanoscaled hierarchical structures, the self-healing superhydrophobicity of the coatings could be easily achieved in a humid ambient environment by transfering the loaded fluoroalkylsilane to the coatings surface.²⁷ By filling fluorinated silsesquioxane (FD-POSS) and fluorinated alkyl silane to coating directly, the self-healing superhydrophobicity of the coating was achieved. Polar groups of FD-POSS generated by air plasma treatment moved inside the coating, and fluorinated alkyl chains were exposed to the surface.²³ These efforts to self-healing superhydrophobicity enormously increased the life span of self-cleaning property of coatings. On the basis of the results reported, we fabricated the superhydrophobic coating with the combination of transparency and healable superhydrophobicity which has been seldom reported.

Here, we present a feasible way to prepare a porous silica nanotube-network coating with good transparency and healable superhydrophobicity. Polyaniline (PANI) nanofibers prepared by dilute chemical polymerization was coated on a glass slide.³² This PANI nanofiber network was coated with a silica shell by chemical vapor deposition (CVD) of tetraethoxysilane (TES) catalyzed by ammonia.¹¹ Porous silica nanotube network was formed and the coating became transparent after calcination. PFA was loaded by immersing the network coating in a solution of PFA in trichlorotrifluoroethane which was then removed by vacuum. When superhydrophobicity of the coating was destroyed, the healing of superhydrophobicity happened by migrating PFA to the surface at mild conditions.

2. Experimental

Materials

Aniline (Aldrich) was distilled under vacuum before used. Ammonium persulfate (APS, Aldrich), tetraethoxysilane (TES, Shanghai Chemical Regent Co.), ammonia solution (Aladdin, 25–28%), perfluorooctyl acid (PFA, Shanghai Chemical Regent Co.), 1H,1H,2H,2H-perfluorooctyl-triethoxysilane (J&k), rhodamine 6G (J&k), sylgard 184 (Dow corning corporation), perchloric acid (HClO₄, Shanghai Chemical Regent Co.) and trichlorotrifluoroethane (Shanghai Chemical Regent Co.) were used as received without further purification.

Preparation of PANI nanofiber network coating

Aniline was dissolved in a HClO₄ solution (1 M). The solution was slowly poured to the solution of APS dissolved in a HClO₄ solution (1 M) in a beaker. The molar concentration of aniline to the whole solution was 8 mM, and the molar ratio of aniline to APS was controlled at 2 : 1. The reaction was kept at room temperature without any disturbance for 24 h and the dark-green precipitate was purified with deionized water until the filtrate was colorless. The suspension of the dark-green precipitate in about 20 ml deionized water was coated on glass slides by spin coating at the speed of 1000 rpm for 30 s, and the nanofiber network coating was obtained after water evaporation.

Preparation of nanotubular coating

PANI nanofiber network coating was placed in a closed desiccator together with TES and aqueous ammonia solution (2 ml separately). Chemical vapor deposition of TES was kept at room temperature for 24 h. After the silica deposited network coating was calcinated at 600 °C for 2 h in air and the heating rate is 2 °C min⁻¹, the coating became transparent and colorless from the initial dark-green.

Preparation of PFA-filled nanotubular coating

Silica nanotubular coating was immersed into a solution of PFA in trichlorotrifluoroethane (0.01 M) for 10 minutes at room temperature. Then the coating was taken out and trichlorotrifluoroethane was removed under vacuum. The nanotubular coating loaded with PFA was obtained after three filling.

Preparation of artificial sweat

Sodium chloride (20 g), ammonia chloride (17.5 g), urea (5 g), lactic acid (15 g) and acetic acid (2.5 g) were added into deionized water (1000 ml) and then sodium hydroxide was added until pH value reached to 4.7.

Preparation of striped PDMS stamp

The sylgard 184 was poured in a mould with stripes on the bottom. After under vacuum for 10 minutes to remove air bubble, the mould was put into oven at 80 °C for 2 h. PDMS stamp which cross-sectional area is 1 cm² was detached from the mould. PDMS stamp with artificial sweat was pressed on the tested surface with a leverage (100 g) for 10 s.

Plasma treatment

The failure of superhydrophobility of the coating was obtained using a plasma instrument (DienerElectronic, Germany) for 10 s with power of 98 W. The plasma treated coating was heated at 70 °C for 5 h, and the contact angle was then measured again after healing.

Characterization

The surface wettability was characterized using a DSA-100 optical contact-angle meter at room temperature (Kruss Co., Ltd., Germany). The volume of water drop was 5 μl and the CA value was obtained automatically using the Laplace–Young fitting algorithm. Average CA values were calculated by measuring five different positions, and images were obtained by using a digital camera (Sony, Ltd., Japan). The coating surface morphology was evaluated on a SU 8020 field emission scanning electron microscope (FE-SEM, Japan) at 1.5 kV and a FEI Tecnai G2 F30 transmission electronic microscopy. All photos were obtained by a Canon camera. Fluorescence images were collected using an Olympus BX51 microscope equipped with a 100 W mercury lamp. All fluorescence pictures were taken under identical lamp illumination and charge-coupled device (CCD) exposure conditions: exposure time 1.13 s, ISO 400. Transparency were measured by ultraviolet visible absorption spectrometer (UV 8100C, Lab Tech). Surface element component was analyzed by X-ray photoelectron spectroscopy (ESCALAB 250Xi multifunctional spectrometer, Thermo Fisher) using Al Kα radiation and the binding energies were referenced to the C 1s line at 284.8 eV from adventitious carbon.

3. Results and discussion

Fig. 1 illustrates the surface morphology and wettability of the samples characterized by SEM, TEM and contact-angle meter. Silica nanoparticles formed by CVD of TES were arranged closely on the surface of PANI nanofibers (Fig. 1A–D), and the diameter of the silica coated PANI fiber was about 160 nm. After calcination at 600 °C, PANI fibers were removed by thermal decomposition and porous silica nanotubes were fabricated with a decreased diameter of about 130 nm (Fig. 1G and H). On the surface of nanotubes, largely distributed nanopores (diameter of from 10 to 20 nm) and closely arranged nanoprotrusions (scale of about 20 nm) were found. Nanopores were formed during the course of PANI decomposition and nanoprotrusions were from initial CVD of TES. After immersing the coating into solution of PFA in trichlorotrifluoroethane which was then removed under vacuum, the surface of silica nanotubes were modified with PFA and nanopores were filled with PFA (Fig. 1E and F).^18,26 A water drop deposited on the fluorizated coating formed a static water contact angle of 173 ± 1° (Fig. 1E). Owing to extremely low surface adhesion force, water drops immediately rolled off from the surface under a little tilted angle. Superhydrophobicity of the coating developed is based on a combination of low surface energy and roughness. Unlike the micro- and nanostructure on lotus leaves surface, air space of network and nanoprotrusions on the surface of nanotubes together lead to a decrease of contact area between a water drop and the surface and make the water drop stay on the surface in Cassie state.

Fig. 1 Morphology of porous network surface and wettability. (A and B) Scanning electron microscope (SEM) images of PANI fibers. (C and D) SEM images of silica coated PANI fibers. (E and F) SEM images of porous silica nanotube network, the inset shows a static water contact angle. (G) Transmission electron microscope (TEM) of silica nanotubes with nanopores and nanoprotrusions. (H) SEM images of cross section of porous silica nanotube network. (I) Schematic of the coating preparation.

PANI template was removed after calcination, and the coating became transparent and colorless from the initial dark-green. From the element composition of the surfaces on the coating of PANI, silica coated PANI and nanotubular coating, we found that N 1s core level spectrum of the PANI shows a peak at a binding energy of about 402.5 eV attributable to C–N species, and Si 2p core level spectrum of the SiO₂ shows a peak at a binding energy of about 103.6 eV attributable to Si–O species and N 1s peak disappeared after the calcination of silica coated PANI coating. The change in element on these surfaces is consistent with the preparation process.

The thickness of nanotube shell was well below the wavelength of visible light which decreased light scatters and lead to a good light transparency. The transmittance of the coating was more than 87% and reduced by less than 5% as compared to that of pristine glass slide, 91%, for 500 nm wavelengths (Fig. 2A). Compared to nano–micro complex structure like lotus leaves surface, such nanotubular structure is beneficial for reducing the scattering of light and increasing the light transmittance of the coating. The high transparency is of importance for reading letters and pictures beneath the coated glass plate (Fig. 2B).

Fig. 2 Light transmittance of the superhydrophobic surface. (A) Ultraviolet-visible transmittance spectra of a superhydrophobic surface compared to pristine glass. (B) Photograph of a drop of dyed water deposited on a superhydrophobic glass slide and a pristine glass, the inset shows the photograph of silica coated PANI fiber coating. The two slides were placed on a piece of labeled paper. Water drops were deposited on the surface of the two slides and rhodamine 6G was added to the water drop on the uncoated slides for observation.

Natural surface can release certain self-repairing substance to maintain its superhydrophobicity, just as lotus leaf releasing wax-like substance. As for artificial material surface, superhydrophobicity will be destroyed permanently when fluorine element lose during the course of usage. It is crucial for an artificial surface to enable the releasing of low surface energy substance to maintain its superhydrophobicity. As shown in Fig. 3A, for the coating loading with PFA, it lost superhydrophobicity after treated with O₂ plasma. And its static water contact angle decreased from 170° to less than 10° (Movie S1†). After the surface was heated at 70 °C, its superhydrophobicity restored gradually and the WCA was 167° after 5 h. The surface can remain its superhydrophobicity after 10 plasma-etching/healing cycles (Fig. 3B). The healable superhydrophobicity of the surface can be confirmed by the free sliding of water drop on it (Movie S2†). At 30 °C, the surface treated with O₂ plasma can restore its superhydrophobicity after 48 h (Fig. S2†).¹⁸ For other hydrophobic reagents, such as 1H,1H,2H,2H-perfluorooctyl-triethoxysilane (POTS), the coating can also achieve healable superhydrophobicity (Fig. S3†).

Fig. 3 (A) Static water contact angles of a plasma treated surface with the healing time at 70 °C. (B) Static water contact angles after 10 plasma-etching/healing cycles. (C) Schematic of plasma treatment and healing of the superhydrophobic coating.

It is evident that PFA loaded in the pores of nanotubes play a major role in the course of healable superhydrophobicity. The O₂ plasma treatment might decompose the PFA on the surface of nanotubes and lead the surface to transfer from superhydrophobicity to superhydrophilicity. But the plasma treating time of 10 s is not enough long to decompose all the PFA in pores and PFA packed in pores migrates gradually to the nanotubes surface when being heated.¹⁸ This can also be confirmed by surface element component analysis in Fig. 4. For the surface loaded with PFA, fluorine element concentration decreased drastically when treated by O₂ plasma. Its atomic percent decreased from 33.4% to 13.3% and the surface behaved superhydrophilicity. However, when the surface was heated at 70 °C for 5 h, its fluorine element concentration increased from 13.3% to 31.8% and the surface restored its superhydrophobicity. The fluorine element concentration even remained up to 24.5% after the surface undergoing 10 plasma-etching/healing cycles. PFA migrated from pores to the surface of nanotubes enabled the surface to heal its superhydrophobicity when the loss of fluorine element happened.

Fig. 4 X-ray photoelectron spectra of (1) superhydrophobic coating, (2) after plasma treatment, (3) one plasma-etching/healing cycle, (4) five plasma-etching/healing cycles and (5) ten plasma-etching/healing cycles.

As the plasma-etching/healing cycles continued to increased, the WCA of the coating decreased evidently and the surface lost its superhydrophobicity with a WCA of 133° after 13 plasma-etching/healing cycles (Fig. S4A†). This is because fluorine compounds loaded in the coating could not provide enough fluorine element on the surface with the consecutive loss of fluorine element. This was confirmed by the evident decrease of fluorine element on the surface of the coating after 10 plasma-etching/healing cycles (Fig. S4B†).

The healable superhydrophocity is important to material surface and it gives a surface the ability to healable self-cleaning. It is of importance for a surface in the field of transparency display protection. The healable self-cleaning surface remains good repellency to frequently used liquids such as milk, juice and coffee after 10 plasma-etching/healing cycles. Static contact angles of these liquids were all more than 150° as well as that of the original surface (Fig. 5A). This indicates that the nanotubular coating have good healable self-cleaning performance to these liquid contaminants.

Fig. 5 (A) Photos of liquids drops on the surface. (B) Photos of artificial sweat and its adhesion on the surface. A striped polydimethylsiloxane stamp adhering a solution of artificial sweat and rhodamine 6G was pressed on the nanotube network surface with 100 grams of force for 10 s. (C) Fluorescence image of solution of rhodamine 6G in artificial sweat adhering on the surface after the press of a striped polydimethylsiloxane stamp. (D) The self-cleaning property of clearing up dusts evaluated by water-droplet impact on the superhydrophobic coatings.

During the use of transparent screen, the surface may be sometimes contaminated by sweat. Usual method to reduce sweat contamination is through surface modification with a thin layer of fluorine compound. The fluorine compound can prevent the surface from sweat contamination at the beginning, however the performance of surface will disappear gradually along with the loss of fluorine compound during the course of usage. To evaluate the healable sweat resistant performance of the surface developed, a striped polydimethylsiloxane (PDMS) stamp adhering a solution of artificial sweat and rhodamine 6G was pressed on the nanotubular surface with 100 g of force for 10 s (Fig. S5†). On the original superhydrophobic surface, no evident artificial sweat was left, which was confirmed by the photo of camera and fluorescence image (Fig. 5B1 and C1). After the surface was treated with plasma, it became superhydrophilic. Red artificial sweat left on the surface was found easily and red fluorescence stripes image of rhodamine 6G were clear (Fig. 5B2 and C2). After the surface underwent 10 plasma-etching/healing cycle, it still repelled artificial sweat very well (Fig. 5B3 and C3).

To the solid granules contaminants (dusts) deposited on the healable self-cleaning surface, it could be easily taken off by quickly rolling water drops (Fig. 5D1). After the surface was treated with O₂ plasma treatment, water drops spreaded out on the surface and dusts couldn't be taken off. After the surface undergoing 10 plasma-etching/healing cycles, its self-cleaning performance recovered. Therefore, the superhydrophobic nanotubular coating developed have longer self-cleaning service life span than that of surface without healable superhydrophobicity performance.

In outdoor environment, superhydrophobic surface need to withstand some destructive conditions. To evaluate mechanical resistance of the coating, sand abrasion test was performed. 20 g sand grains outflowed from a funnel and impinged the tilted 45-degree surface for about 20 s. Diameters of sand grains were from 100 to 200 μm and the falling height was 20 cm (Fig. 6A). No evident destruction was found on the surface after sand abrasion and the coating still kept its superhydrophobicity with the WCA of 155° (Fig. 6B–D and Movie S3†).

Fig. 6 Mechanical resistance evaluated by sand abrasion. (A) Photograph of sand grains impinged the tilted 45-degree coating from a height of 20 cm. (B) Water contact angle of the coating after 20 g of sand abrasion from 20 cm height. (C and D) SEM images of the superhydrophobic surface after sand abrasion.

4. Conclusions

A transparent healable superhydrophobic self-cleaning nanotubular coating is developed. The coating is made up of silica nanotube network filled with perfluorooctyl acid, which gives the coating good transparency and healable self-cleaning performance. The healable self-cleaning performance is achieved by healable superhydrophobicity. When superhydrophobicity of the surface is destroyed, PFA loaded in pores can migrate on the surface at mild conditions and the superhydrophobicity is restored. The healing of self-cleaning performance can be achieved after 10 plasma-etching/healing cycles. It is anticipated that the transparent and healable self-cleaning coating is possible to be applied to some transparent material surfaces.

Acknowledgements

This research project was financially supported by the National Natural Science Foundation of China (51403220, 21303233, 51335010), HUAWEI Science and Technology Fund and Youth Innovation Promotion Association CAS.

Notes and references

1. H. Y. Guo, Q. Y. Li, H. P. Zhao, K. Zhou and X. Q. Feng, RSC Adv., 2015, 5, 66901.
2. B. N. Sahoo and B. Kandasubramanian, RSC Adv., 2014, 4, 22053.
3. A. Lafuma and D. Quéré, Nat. Mater., 2003, 2, 457.
4. Y. Wang, X. W. Liu, H. F. Zhang and Z. P. Zhou, RSC Adv., 2015, 5, 18909.
5. E. Celia, T. Darmanin, E. T. Givenchy, S. Amigoni and F. Guittard, J. Colloid Interface Sci., 2013, 402, 1.
6. W. F. Zhang, X. Lu, Z. Xin, C. L. Zhou and J. Liu, RSC Adv., 2015, 5, 55513.
7. X. J. Feng and L. Jiang, Adv. Mater., 2006, 18, 3063.
8. L. W. Zhou, P. Yu, Y. H. He, H. H. Xia, X. L. Guo and Y. B. Luo, RSC Adv., 2015, 5, 92947.
9. D. P. Wang, A. W. Zhao, L. Li, Q. Y. He, H. Y. Guo, H. H. Sun and Q. Gao, RSC Adv., 2015, 5, 96404.
10. X. Deng, L. Mammen, Y. F. Zhao, P. Lellig, K. Müllen, C. Li, H. Butt and D. Vollmer, Adv. Mater., 2011, 23, 2962.
11. X. Deng, L. Mammen, H. Butt and D. Vollmer, Science, 2012, 335, 67.
12. L. Y. Xu, D. D. Zhu, X. M. Lu and Q. H. Lu, J. Mater. Chem. A, 2015, 3, 3801.
13. N. Yokoi, K. Manabe, M. Tenjimbayashi and S. Shiratori, ACS Appl. Mater. Interfaces, 2015, 7, 4809.
14. L. G. Xu, Z. Geng, J. H. He and G. Zhou, ACS Appl. Mater. Interfaces, 2014, 6, 9029.
15. L. L. Xiao, J. S. Li, S. Mieszkin, A. D. Fino, A. S. Clare, M. E. Callow, J. A. Callow, M. Grunze, A. Rosenhahn and P. A. Levkin, ACS Appl. Mater. Interfaces, 2013, 5, 10074.
16. T. S. Wong, S. H. Kang, S. K. Y. Tang, E. J. Smythe, B. D. Hatton, A. Grinthal and J. Aizenberg, Nature, 2011, 477, 443.
17. C. Neinhuis, K. Koch and W. Barthlott, Planta, 2001, 213, 427.
18. X. L. Wang, X. J. Liu, F. Zhou and W. M. Liu, Chem. Commun., 2011, 47, 2324.
19. N. Puretskiy, G. Stoychev, A. Synytska and L. Ionov, Langmuir, 2012, 28, 3679.
20. Y. H. Liu, Z. L. Liu, Y. P. Liu, H. Y. Hu, Y. Li, P. X. Yan, B. Yu and F. Zhou, Small, 2015, 11, 426.
21. H. X. Wang, Y. H. Xue, J. Ding, L. F. Feng, X. G. Wang and T. Lin, Angew. Chem., 2011, 48, 11635 (Angew. Chem. Int., Ed., 2011, 50, 11433).
22. H. X. Wang, H. Zhou, A. Gestos, J. Fang, H. T. Niu, J. Ding and T. Lin, Soft Matter, 2013, 9, 277.
23. H. Zhou, H. X. Wang, H. T. Niu, A. Gestos and T. Lin, Adv. Funct. Mater., 2013, 23, 1664.
24. H. X. Wang, H. Zhou, A. Gestos, J. Fang and T. Lin, ACS Appl. Mater. Interfaces, 2013, 5, 10221.
25. D. D. Zhu, X. M. Lu and Q. H. Lu, Langmuir, 2014, 30, 4671.
26. Q. Z. Liu, X. L. Wang, B. Yu, F. Zhou and Q. J. Xue, Langmuir, 2012, 28, 5845.
27. Y. Li, L. Li and J. Q. Sun, Angew. Chem., 2010, 35, 6265 (Angew. Chem., Int. Ed., 2010, 49, 6129).
28. S. S. Chen, X. Li, Y. Li and J. Q. Sun, ACS Nano, 2015, 9, 4070.
29. S. R. White, N. R. Sottos, P. H. Geubelle, J. S. Moore, M. R. Kessler, S. R. Sriram, E. N. Brown and S. Viswanathan, Nature, 2000, 409, 794.
30. Y. H. Liu, Y. P. Liu, H. Y. Hu, Z. L Liu, X. W. Pei, B. Yu, P. X. Yan and F. Zhou, J. Phys. Chem. C, 2015, 119, 7109.
31. B. C. Li and J. P. Zhang, Carbon, 2015, 93, 648.
32. N. R. Chiou and A. J. Epatein, Adv. Mater., 2005, 17, 1679.

---

Footnote

† Electronic supplementary information (ESI) available: Fig. S1–S5 and Movie S1–S3. See DOI: 10.1039/c5ra26977f

---