Preparation of superhydrophobic coating with excellent abrasion resistance and durability using nanofibrillated cellulose

Abstract

Nanofibrillated cellulose (NFC) is a renewable and environmentally friendly material. Its large ratio of length to diameter and its pliability are very helpful for forming a three-dimensional network structure that could contribute to suitable roughness for superhydrophobic coatings. At present, superhydrophobic coatings have limited applications, mainly due to their poor abrasion resistance and durability. Therefore, in this study, a semi-translucent NFC superhydrophobic coating was prepared by spraying NFC ethanol suspension onto a wood surface that had been previously covered with a commercial spray paint as adhesive, followed by modification via phase chemical vapor deposition (CVD) to reduce surface free energy. The results showed that in addition to good superhydrophobic and self-cleaning properties, the coating also possessed excellent abrasion resistance and durability and was able to resist sandpaper abrasion, knife-scratch, finger-wipe, long-time impregnation in water, UV radiation, and low temperatures, demonstrating good potential for development and application.

---

Introduction

The properties of being waterproof,¹ self-cleaning,² and anti-fogging³ and possessing drag reduction⁴ could be implemented in superhydrophobic surfaces which were initially found in nature such as the surface of lotus leaves⁵ and cicada wings.⁶ In addition, materials developed with these properties could be used in daily life, industry and agriculture.⁷ In recent years, superhydrophobic surfaces that can be built onto the surfaces of materials like metals,⁸ glasses,⁹ textiles,¹⁰ aerogels¹¹ and paper¹² have become a hot topic for research. Such surfaces having a water contact angle (WCA) of larger than 150° and a small contact angle hysteresis (that is, a small slide angle (SA)) are described as superhydrophobic surfaces.¹³

Wood as a kind of biomass material is widely used in many fields, but is easily swelling and shrinking due to its hydrophilic properties and lots of hydroxyl on the surface, also often appears microbial decay leading to reduction of service life. So superhydrophobicity could be a kind of effective measures to protect wood.

To get a superhydrophobic surface, it is essential to have reasonable surface roughness and low surface free energy.¹⁴ Currently, a common way to change a hydrophilic surface into a superhydrophobic one is to use a superhydrophobic coating. In such coatings, various inorganic nanoparticles including SiO₂,¹⁵ Al₂O₃,¹⁶ CuO,¹⁷ CaCO₃¹⁸ and TiO₂ ¹⁹ are usually employed to build reasonable roughness and siloxanes or fluoropolymers are typically used as low surface energy materials.²⁰ At present, some superhydrophobic coatings that have been extensively reported are relatively easy to implement, and some progress has been made in preparing them successfully to yield such attributes as ultrahigh WCA,²¹ high transparency²² or multi-functionality,²³ but some questions still exist, hampering further application. Among them, abrasion resistance and durability remain key points that require solutions.^24,25

In addition to abrasion resistance and durability, minimizing the environmental impact of superhydrophobic coatings is a problem to which the use of natural cellulose may be the answer. Natural cellulose is widely abundant, mainly existing in plants.²⁶ The cellulose content of cotton is greater than 90%, while that of wood is about 50%.²⁷ Nanocellulose could be cleaved from purified cellulose by means of chemical hydrolysis or mechanically, and can be further divided into nanocrystalline cellulose (NCC) or nanofibrillated cellulose (NFC) according to various processes.²⁸ NFC, which has a large length-to-diameter ratio and large specific surface area,²⁹ can easily form a network structure by mutual entanglement and help build a rough structure that serves as a base for a superhydrophobic coating. Nanocellulose-based superhydrophobic coatings are a very promising high-value use of nanocellulose.

Although there has been a good deal of research on the superhydrophobic modification of cellulose-based materials such as nanocellulose aerogels,³⁰ cotton textile³¹ and paper,³² it was rare to use only NFC as the main structure material in superhydrophobic coatings. To our best knowledge, only Mertaniemi et al.³³ have prepared NFC microparticles using for superhydrophobic surfaces which mainly imitated the surface microstructure of lotus leaves, but haven't studied other properties.

In this study, we built a different structure from that of the lotus leaf surface, also focusing on improving the abrasion resistance and durability of the superhydrophobic coating. A NFC superhydrophobic coating with a stable three-dimensional network structure could be attained by a simple method of spraying NCF ethanol suspension quickly onto wood surfaces pre-primed with adhesive and then dried at room temperature or in an oven, followed by low-surface energy modification via phase chemical vapor deposition (CVD). The design process was shown in Fig. 1. In order to get a NFC coating with the desired roughness, it was decided to choose a highly volatile solvent whose rapid evaporation would allow the NFC to form a rough surface to avoid forming a film. First, the NFC water suspension was replaced with an anhydrous ethanol into NFC ethanol suspension in order to get the suitable roughness due to rapid evaporation of solvent. However, if the NFC ethanol suspension were directly sprayed onto wood surfaces, the NFC coating's adhesion would be too low to resist abrasion. To enhance the abrasion resistance of the coating, we chose to fix the NFC coating to the wood surfaces with adhesive. Thus, the substrate surfaces were first treated with adhesive before being sprayed with the NFC ethanol suspension. Dried at room temperature or in an oven, the wood surfaces would show a layer of NFC coating with three-dimensional network structures, and then the fluoride group was grafted onto the NFC coating surface by CVD.

Fig. 1 Sketch of the procedure to prepare superhydrophobic NFC coatings on wood substrate.

The prepared NFC coating showed not only excellent superhydrophobic properties, but also had outstanding stability and was able to maintain its superhydrophobic properties after tests by sandpaper abrasion, knife-scratch, finger-wipe, long-time impregnation in water, UV radiation and long-time staying at low temperature.

Experimental

Materials

Nanofibrillated cellulose (NFC) (water suspension, 2.95% solid content), whose microstructure is shown in Fig. 2(a) and the diameter was about 3–15 nm,³³ was purchased from the University of Maine, Process Development Center (Orono, ME, USA). A kind of commercial Krylon K09116000 COVERMAXX Spray Paint (Gloss Crystal Acrylic) was used as the adhesive between the NFC and the substrate surface. 1H,1H,2H,2H-Perfluorooctyltrichlorosilane (CF₃(CF₂)₅(CH₂)₂SiCl₃, 97%) and anhydrous ethanol (200 proof) were purchased from Fisher Scientific, USA. All chemical reagents were directly used without further processing. Small pieces of wood of dimensions 4.5 cm × 2.5 cm × 1.5 cm were first cleaned using deionized water, then ultrasonic cleaning was performed in anhydrous ethanol for 10 min, and the pieces were finally dried for 5 hours under 103 °C.

Fig. 2 (a) TEM image of NFC; (b) self-made slide angle (SA) measuring station; top morphology of (c) pristine wood, (d) spray paint layer and (e) NFC coating; (f) cross-section topography of NFC coating with adhesive.

Preparation of NFC superhydrophobic coating

Replacement. 2.95 wt% NFC water suspension was centrifuged at 10 000 rad min⁻¹ through a high-speed centrifuge. Then the upper liquid was discarded and the lower sediment was immersed in the same weight of fresh anhydrous ethanol as the discarded upper liquid and mixed again by continuous stirring for 1 h to form new NFC ethanol suspension. This process represented one cycle. A total of three cycles were conducted to get a highly pure NFC ethanol suspension. Then the prepared NFC ethanol suspension was diluted to a 1.2 wt% concentration with fresh anhydrous ethanol.

Spraying. First, the adhesive was sprayed onto the wood surfaces with about 0.1–0.2 g spraying quantity. And after 3–5 min, the 1.2 wt% NFC ethanol suspension, which of about 0.3–0.5 g was used on each wood surface, was sprayed using an airbrush (Uxcell Mini 0.5 K3 HVLP Gravity Feed Paint Spray Gun Airbrush) onto the previously sprayed adhesive surface. The distance from the spray nozzle to the wood surface was between 45–60 cm, and the air pressure of the compression pump was controlled between 0.2–0.4 MPa. Though the NFC became floccules in anhydrous ethanol, this had no effect on the spraying at 1.2 wt% concentration.

Drying. The wood pieces covered with adhesive and NFC were dried at room temperature for above 3 h or at 90 °C in an oven for above 10 min to completely volatilize the alcohol to sufficiently cure the adhesive.

CVD. The prepared samples were put into a 500 ml bottle with a 30 ml bottle loaded with 0.5 g 1H,1H,2H,2H-perfluorooctyltrichlorosilane. The large bottle was then sealed with a lid and placed in an oven for over 4 h at 80 °C. After that, the samples were taken out from the bottle and dried at 90 °C for more than 30 min to remove unreacted 1H,1H,2H,2H-perfluorooctyltrichlorosilane and byproducts of the CVD process.

Characterization

The surface microstructure of the NFC coating was investigated via a Zeiss Auriga SEM/FIB crossbeam workstation (Germany) operated at an accelerating voltage of 3 kV. The chemical composition of the NFC coating before and after modification was analyzed via Fourier transform infrared spectroscopy (FTIR, Perkin Elmer, USA) at a scanning range of 4000–600 cm⁻¹, scanning 16 times.

A Drop Shape Analysis System (EasyDrop, Germany KRUSS) was used to measure water contact angle (WCA) at ambient temperature. Slide angle (SA) was measured by a self-made SA measuring station expressly fabricated for the purpose as shown in Fig. 2(b). A protractor with a hole of 2 mm diameter in the center of its baseline was vertically fixed by adhesives to a strip of aluminum alloy whose section center also had a hole of 5 mm diameter; the two holes overlapped with each other. A smooth panel was horizontally fixed to two intersectional sides of a “T” iron frame with an adhesive plaster. Inserted into the overlapped holes, the remaining side served as a rotational axis, allowing the panel to be rotated 180° along the axis, so the rotational angle, equal to the SA, could be directly read when water droplets placed onto the smooth panel surface rolled down. Before each measurement, the smooth panel was adjusted into level surface using a level measurement instrument. Each measurement of WCA and SA was conducted with a water droplet of 4 μL, selecting randomly five different points each time and getting average value.

Self-cleaning test. Some dust was randomly placed on the surface of the treated coating placed on a horizontal plane with an approximately 5° inclination angle. Water droplets were then dropped onto the coating surface from the upper part above the coating.

Sandpaper abrasion test. Similarly to a previously reported method,³⁴ the sample with superhydrophobic coating was placed against sandpaper (1500 mesh) and weighted 200 g of weight. The sample was moved in a straight manner 4 inch (about 10 cm), followed by being horizontally wiggled 90° and moved another 4 inch (about 10 cm). This process was denoted as one cycle.

Knife-scratch test. The sample was put onto a horizontal plane. The scratches covering the coating were made by a knife along a line intersecting into about 45° with two adjacent sides of the substrate. Other scratches covering the coating were conducted along the direction perpendicular to previous ones.

Finger-wipe test. One hand kept samples stationary and a finger of the other one pressed onto the surface of the samples and then moved backwards. That was conducted three times in different areas.

Long-time impregnation test. Each wood sample was completely immersed in water for 5 h and pressed with heavy stuff to prevent them from floating, then being taken out and dried at 103 °C for 3 h.

UV radiation test. A Blak-Ray® XX-15M UV bench lamp was used with a light source (power: 15 W; wavelength: 254 nm) and the samples which were 10 cm from light source were continuously exposed under the UV for 72 h.

Low temperature test. The samples were stayed at −30 °C for 10 hours in a low-temperature refrigerator and then taken out and their wettability was tested within 30 s.

Results and discussion

Surface morphology of NFC coating

The surface morphology of the pristine wood is shown in Fig. 2(c). The surface showed many large grooves formed from broken vessels. As shown in Fig. 2(d), the surface was smooth after being treated with spray paint because the large grooves were filled in. When further treated with NFC ethanol suspension, the surface topography of the coating, shown in Fig. 2(e), had a three-dimensional network structure and irregular protuberances. The first reason for that appearance was uneven spraying. As the NFC ethanol suspension came out the nozzle, NFC fog droplets formed due to shear force. The protuberances formed in these areas where fog droplets were located. The other reason was the result of ethanol volatilization. The distance between NFC and NFC became closer and closer and eventually the NFC aggregated in the process, promoting further formation of the protuberances and creating the micro-structure roughness. As shown by the local amplification of the NFC coating, there were many little convex in the protuberances. Thus, the protuberances and the little convex on them formed the micro–nano scale roughness needed for superhydrophobic properties. This three-dimensional network was somewhat similar to the surface microstructure of paper which was composed of winding cellulose. But being larger size than NFC, cellulose had more difficulty in forming nano-structures. The prepared NFC coating was more conducive to superhydrophobic modification than cellulose paper. Fig. 2(f) showed different areas of the cross-section of the sample treated with NFC coating. The wood and NFC were contacted by the spray paint as adhesive and there was a part of the NFC integrating into the adhesive. This NFC and adhesive interacting area resulted from the compatibility between the spray paint and ethanol and could give the coating strong adhesion.

FTIR of the CNF coating before and after modification was shown in Fig. 3. The new absorption peak at 1107 cm⁻¹ was created by the stretching vibrations of the Si–O bonds which formed in the reaction between 1H,1H,2H,2H-perfluorooctyltrichlorosilane and hydroxyl on the NFC surface, which affirmed CVD was an effective method to make fluoride group be grafted onto NFC surface. The absorption peaks presenting at 1147 and 1239 cm⁻¹ were attributed to the stretching vibrations of the C–F bonds which supported low surface free energy for the NFC coating.

Fig. 3 FTIR spectra of the NFC coating before and after modification.

Wettability and self-cleaning properties

As shown in Video 1 in ESI,† the pristine wood was hydrophilic and the water droplets were adhered to the surface. Thus, the wood treated with the NFC coating presented excellent superhydrophobic properties and it was easy for the water droplets roll down. In addition, Fig. 4(a–c) demonstrated the wettability of the different sample surfaces after modification with 1H,1H,2H,2H-perfluorooctyltriethoxysilane. Dripped onto the surface of the pristine wood after modification as shown in Fig. 4(a), the blue water droplet (dyed blue with methylene) presented as two-thirds spherical, having only 143.2° WCA, highly adhesive and not superhydrophobic, showing the pristine wood surface could not supply enough roughness. When the wood was treated only with spray paint, as shown in Fig. 4(b), the water droplet presented as a hemisphere firmly adhered to the surface and the WCA was lower than that of the pristine wood and reduced to 103.5°, which resulted from lower surface roughness due to the smoother adhesive layer covering the grooves on the pristine wood surface. The wood treated with both adhesive and NFC showed outstanding superhydrophobic properties as seen in Fig. 4(c). Dripped onto the surface, the water droplet presented spherically and could easily roll, possessing 161° WCA and below 10° SA, attributable to the suitable roughness contributed by the three-dimensional network structure NFC coating. This showed that the superhydrophobic properties were mainly decided by surface roughness under the same low surface energy. The NFC coating not only had excellent superhydrophobic properties, but its superoleophobic (glycerol) properties were also very remarkable as seen in Fig. 4(d). Being similar to water droplets, oil drops could roll on the coating surface and the contact angle was 153°, slide angle below 15°.

Fig. 4 Status of water droplets in (a) pristine wood, (b) wood treated only with adhesive and (c) wood treated with both adhesive and NFC coating after modification; (d) status of oil droplets in the wood treated with NFC coating after modification; (e) transparency of the treated CNF coating covering a glass slide; (f) silver mirror phenomenon of the wood treated with CNF coating after modification; (g–i) self-cleaning test of the wood treated with NFC coating after modification.

To better investigate the transparency of the NFC superhydrophobic coating, a transparent glass slide was used as a substrate. As shown in Fig. 4(e), the coating showed semitransparent properties. This was because the adhesive was transparent and although NFC wasn't transparent, the NFC coating could not completely cover the adhesive layer, and light was able to go through the uncovered area. That also demonstrated that the NFC coating was of a three-dimensional network structure.

As shown in Fig. 4(f), when the treated wood was dipped into water at room temperature and viewed at a glancing angle, the wood surface covered with NFC coating showed a silver mirror phenomenon that was caused by light reflectance of the air at the interface between water and coating,³⁵ but the surface without NFC coating did not show this effect. This proved also that the NFC coating was superhydrophobic.

As shown in Video 2, ESI,† some dirt was placed on the untreated wood with NFC coating tilted an angle of 5°. Dripped onto the surface, water droplets were hemispherical, mixed with dust, and attached to the surface without rolling. With the increase of water droplets volume from continuing dripping, the droplets became oblate and spread along the slope due to gravity. Instead of being taken away by the droplets, the dust extended along the surface and made a larger contaminated area. However, the wood covered with NFC coating showed good self-cleaning properties. Dripped onto the surface tilted with the same angle, the water droplets could form a sphere and roll down along the slope, and dust on the surface attached to the surfaces of the rolling water droplets and was carried away, leaving a clean surface without a trace of dust. The curt process was as shown in Fig. 4(g–i).

Abrasion resistance and durability

Abrasion resistance and durability of superhydrophobic coating were closely related to the duration of its working life. In order to be closer to actual application, the samples were tested by the sandpaper abrasion, knife-scratch, finger-wipe, resistance to long-time impregnation, resistance to low temperature, and resistance to UV radiation.

Sandpaper abrasion. Sandpaper abrasion is currently a common method of evaluating the abrasion resistance of superhydrophobic coatings. As shown in Fig. 5(a and b), the sample was weighted with 200 g, placed against 1500 mesh sandpaper, and moved 4 inch (about 10 cm) along the ruler. Then the sample was rotated 90° horizontally and moved another 10 cm along the sandpaper. This was defined as a cycle, and total 6 cycles were conducted. As shown in Video 3a, ESI,† the NFC coating untreated with adhesive began to be abraded by the sandpaper after only one cycle, which showed the weak adhesion between the coating and the substrate without adhesive. As shown in Fig. 5(c), these with red cycles were damaged areas, but as shown in Video 3b, ESI,† the one treated with adhesive was not damaged even after 6 cycles. The reason for the powerful adhesion was that in the process of forming the coating, due to the rapid curing of the adhesive and evaporation of the anhydrous ethanol, the compatibility between the adhesive and the ethanol led to the complete blending between the part of the NFC close to the adhesive and the adhesive itself, which resulted in the strong connection and guaranteed abrasion resistance. The top CNF failed to blend with the adhesive and was able to form a three-dimensional network structure coating which allowed the roughness to support superhydrophobicity after modification. As shown in Fig. 5(e), during the process of 6 cycles, on the whole, the WCA of the coating with adhesive was decreasing, but still reaching 150.7° after 6 cycles. That benefited from the highly intensive three-dimensional network structure protuberances that were formed from the constriction and caking of the NFC in the drying process and could resist sandpaper abrasion to a certain extent. With increasing sandpaper abrasion cycles, adhesion strength of water droplets increased and the SA showed a downward trend. The microscopic structure after 6 cycles of sandpaper abrasion was shown in Fig. 5(d). The protuberances were rubbed down and the part of NFC that was in contact with the sandpaper fractured, the NFC layer becoming smoother and smoother. That widened the space between the grooves and increased the contact area between the water droplets and the coating surface, so adhesion strength increased.

Fig. 5 (a and b) Demonstration of sandpaper abrasion for one cycle; (c) surface status of the NFC coating without adhesive after one sandpaper abrasion cycle; (d) surface morphology of the NFC coating with adhesive after 6 sandpaper abrasion cycles; (e) influence of sandpaper abrasion cycles on WCA and SA of the NFC coating with adhesive; (f) surface status of the NFC coating without adhesive after knife-scratch; (g) wettability of the NFC coating with adhesive after knife-scratch.

Knife-scratch and finger-wipe. As shown in Video 4, ESI,† the samples were horizontally placed on a level plane, and a knife was used to scratch the coating as described in Characterization. As shown in Fig. 5(f), after knife-scratch, there were a part falling off in the NFC coating treated without adhesive, and the scratch traces were wide. But as shown in Fig. 5(g), this did not occur with the sample treated with adhesive. That was mainly attributed to the strong adhesion of the adhesive which prevented the faulty NFC layer from separating. Though the knife-scratch made the NFC layer fracture, the crack gap was tiny due to adhesion, and the water droplets were not able to fall into the gap. So the water droplets on the scratched NFC coating treated with spray paint were still spherical and easily able to roll, in contrast to the surfaces with no knife-scratch, in which the WCA and AS showed almost no change. The finger-wipe test was carried out as shown in Video 5, ESI.† The water droplets still remained spherical and rolled easily after the finger-wipe. The superhydrophobic properties of the coating remained good after both knife-scratch and finger-wipe, indicating that the spray paint as adhesive played a key role in the fixation between the NFC coating and the wood.

Resistance to long-time impregnation. If water droplets at above 30 °C were directly used to measure WCA and the SA of superhydrophobic coating, the WCA would decrease and the SA showed an increasing trend, which was thought to be the result of water molecule adsorption caused by the cooler coating surface.³⁶ But after respectively being dipped into water of room temperature (24 °C), 30 °C, 40 °C, 50 °C and 60 °C for 5 hours, followed by being taken out and dried under 103 °C for 3 hours, the WCA and SA were measured with water at room temperature after drying. The results were shown in Fig. 6(a). WCA and SA basically were not different from measurements taken before immersion. Water droplets could still form spheres and easily rolled down, retaining good superhydrophobic properties. That illustrated that the microstructure of the NFC coating had not been destroyed and could be restored after drying, demonstrating outstanding resistance to long-time impregnation.

Fig. 6 (a) Influence of UV radiation on WCA and SA of the NFC coating treated with adhesive; (b) influence of impregnation in different temperature water on WCA and SA of the NFC coating treated with adhesive; (c) wettability of the slide glass covered with and without the treated NFC coating after staying for 10 hours at −30 °C.

Resistance to UV radiation. In general, cellulose easily absorbs ultraviolet light and can cause aging degradation to a certain degree. Fig. 6(b) showed the influence of UV radiation time on the WCA and SA of the wood treated with NFC coating when the NFC coating was exposed to UV radiation. During the irradiation process of total 72 hours, the superhydrophobic properties remained good and the WCA and SA basically showed no change. The NFC coating kept an excellent resistance to UV. The reason was that C–F bonds in the coating had a great deal of energy and weren't broken by UV radiation.³⁷

Resistance to low temperature. If application outdoors, it is necessary to artificially mimic low temperature conditions on the coating surface. To be more convenient to investigate, a glass slide which was more sensitive to temperature change than wood was chosen as a substrate. So samples were kept at −30 °C in a low-temperature refrigerator for 10 hours and then taken out. As shown in Fig. 6(c), a layer of thin mist settled on the untreated glass slide surface, but not occurring on the treated glass slide surface where water droplets could still remain spherical and be able to roll. That suggested the treated NFC coating had good resistance to low temperature.

Conclusions

An environmentally friendly and robust semitransparent superhydrophobic coating was acquired through CVD modification for reducing surface free energy after a simple two-spraying process, which included initially spraying a commercial spray paint as adhesive onto substrate surfaces to support high-adhesion-stress and then spraying NFC ethanol suspension to build suitable roughness. The resulting NFC superhydrophobic coating showed good self-cleaning properties. In addition, the coating also exhibited outstanding abrasion resistance and durability and could retain superhydrophobic properties after sandpaper abrasion, finger-wipe, knife-scratch, long-time impregnation in water, UV radiation, and long-time staying at low temperature. So, in addition to being suitable for coating both wood and glass, the NFC superhydrophobic coating can potentially be used on many different substrate surfaces, such as plastic, floors and fabrics. It is particularly noteworthy that NFC is a renewable, green, nontoxic material. It is to be expected that the NFC superhydrophobic coating will find extensive potential application in food packaging, children's toys, and other categories of products that must conform to strict health and safety regulations.

Acknowledgements

This work was financially supported by the Special Fund for Forest Scientific Research in the Public Welfare (No. 201504603), Tennessee Experimental Station Project #TEN00422 and the Special Fund of Chinese Central Government for Basic Scientific Research Operations in Commonweal Research Institutes (No. CAFINT2014K02). We thank the Department of Food Science & Technology, the University of Tennessee for providing Drop Shape Analysis System.

References

1. H. Zhou, H. Wang, H. Niu, A. Gestos, X. Wang and T. Lin, Adv. Mater., 2012, 24, 2409–2412.
2. K. Tu, X. Wang, L. Kong, H. Chang and J. Liu, RSC Adv., 2016, 6, 701–707.
3. Y. Chen, Y. Zhang, L. Shi, J. Li, Y. Xin, T. Yang and Z. Guo, Appl. Phys. Lett., 2012, 101(3), 033701.
4. Y. Wang, X. Liu, H. Zhang and Z. Zhou, RSC Adv., 2015, 5, 18909–18914.
5. L. Jiang, Y. Zhao and J. Zha, Angew. Chem., 2004, 116, 4438–4441.
6. X. Yao, Y. Song and L. Jiang, Adv. Mater., 2011, 23, 719–734.
7. M. Yamanaka, K. Sada, M. Miyata, K. Hanabusad and K. Nakano, Chem. Commun., 2006, 2248–2250.
8. M. H. Kwon, H. S. Shin and C. N. Chu, Appl. Surf. Sci., 2014, 288, 222–228.
9. H. M. Kim, S. Sohn and J. S. Ahn, Surf. Coat. Technol., 2013, 228, S389–S392.
10. M. Zhang, D. Zang, J. Shi, Z. Gao, C. Wang and J. Li, RSC Adv., 2015, 5, 67780–67786.
11. J. Fu, S. Wang, C. He, Z. Lu, J. Huang and Z. Chen, Carbohydr. Polym., 2016, 147, 89–96.
12. X. Tang, S. Nan, T. Wang, Y. Chen, F. Yu, G. Zhang and M. Pei, RSC Adv., 2013, 3, 15571–15575.
13. J. Song and O. J. Rojas, Nord. Pulp Pap. Res. J., 2013, 28, 216–238.
14. J. Wang, Y. Zheng and A. Wang, Chem. Eng. J., 2012, 213, 1–7.
15. H. Chang, K. Tu, X. Wang and J. Liu, RSC Adv., 2015, 5, 30647–30653.
16. H. Ogihara, J. Xie and T. Saji, Colloids Surf., A, 2013, 434, 35–41.
17. J. Zeng, B. Wang, Y. Zhang, H. Zhu and Z. Guo, Chem. Lett., 2014, 43, 1566–1568.
18. H. Zhang, X. Zeng, Y. Gao, F. Shi, P. Zhang and J. F. Chen, Ind. Eng. Chem. Res., 2011, 50, 3089–3094.
19. C. Jin, Y. Jiang, T. Niu and J. Huang, J. Mater. Chem., 2012, 22, 12562–12567.
20. H. Teisala, M. Tuominen and J. Kuusipalo, Adv. Mater. Interfaces, 2014, 1.
21. L. Feng, S. Li, H. Li, J. Zhai, Y. Song, L. Jiang and D. Zhu, Angew. Chem., 2002, 114, 1269–1271.
22. W. Ma, Y. Higaki, H. Otsuka and A. Takahara, Chem. Commun., 2013, 49, 597–599.
23. N. Chen and Q. Pan, ACS Nano, 2013, 8, 6875–6883.
24. Q. Yu, Z. Zeng, W. Zhao, M. Li, X. Wu and Q. Xue, Colloids Surf., A, 2013, 427, 1–6.
25. S. Peng and B. Bhushan, J. Colloid Interface Sci., 2016, 461, 273–284.
26. W. Chen, H. Yu, Y. Liu, P. Chen, M. Zhang and Y. Hai, Carbohydr. Polym., 2011, 83, 1804–1811.
27. H. P. S. A. Khalil, A. H. Bhat and A. F. I. Yusra, Carbohydr. Polym., 2012, 87, 963–979.
28. P. Tingaut, T. Zimmermann and G. Sèbe, J. Mater. Chem., 2012, 22, 20105–20111.
29. C. Aulin, J. Netrval, L. Wågberg and T. Lindström, Soft Matter, 2010, 6, 3298–3305.
30. Q. Zheng, Z. Cai and S. Gong, J. Mater. Chem. A, 2014, 2, 3110–3118.
31. L. Wu, J. Zhang, B. Li and A. Wang, J. Mater. Chem. B, 2013, 1, 4756–4763.
32. B. Balu, A. D. Berry, D. W. Hessa and V. Breedveld, Lab Chip, 2009, 9, 3066–3075.
33. H. Mertaniemi, A. Laukkanen, J. E. Teirfolk, O. Ikkalaa and R. H. A. Ras, RSC Adv., 2012, 2, 2882–2886.
34. Y. Lu, S. Sathasivam, J. Song, C. R. Crick, C. J. Carmalt and I. P. Parkin, Science, 2015, 347, 1132–1135.
35. I. A. Larmour, S. E. J. Bell and G. C. Saunders, Angew. Chem., 2007, 119, 1740–1742.
36. Z. J. Yu, J. Yang, F. Wan, Q. Ge, L. L. Yang, Z. L. Ding, D. Q. Yang, E. Sacherc and T. T. Isimjand, J. Mater. Chem. A, 2014, 2, 10639–10646.
37. K. Chen, S. Zhou and L. Wu, Chem. Commun., 2014, 50, 11891–11894.

---

Footnote

† Electronic supplementary information (ESI) available: Videos about the samples in water droplets dripping test, self-cleaning test, sandpaper abrasion test, knife-scratch and finger-wipe test. See DOI: 10.1039/c6ra23447j

---