J. Phys. Chem. B 2009, 113, 2909–2912
2909
A Novel Self-Cleaning Coating with Silicon Carbide Nanowires Jun Jie Niu*,† and Jian Nong Wang‡ School of Materials Science and Engineering, Shanghai Jiao Tong UniVersity, 800 Dongchuan Road, Shanghai, 200240, P. R. China, and School of Materials Science and Engineering, Tongji UniVersity, 1239 Siping Road, Shanghai 200092, P.R. China ReceiVed: September 18, 2008; ReVised Manuscript ReceiVed: December 7, 2008
A novel self-cleaning glass was successfully achieved by coating macroscopical SiC nanowires (SiCNWs) in tetraethyl orthosilicate (TEOS) solution. The water contact angle (CA) was high, up to ∼160°, and the sliding angle was low, down to ∼5°, when SiCNWs were coated through 10 cycles, with a high roughness Ra of 1928.9 nm. High chemical stability was obtained even after immersing the sample in water for 14 days (336 h). The calculated data by using nano/micropillar composite structure model displayed a beneficial understanding on thehydrophobic property. The feasible coating on any substrate, high CA, and long lifetime make SiCNW a potential superhydrophobic material in various self-cleaning fields. 1. Introduction Superhydrophobicity is commonly obtained through the combination of high roughness and water-repelling chemical coating. Nano/microstructures improve the amount of air trapped within the pores.1 The hydrophobic chemical coating decreases the surface free energy.2 There have been several recent reports of superhydrophobic surfaces that combine hydrophobic chemical coatings with rough fractal surfaces characterized by either aligned nanostructures3,4 or random fractal geometry.5,6 Initiated by the “lotus leaf” effect in nature, various artificial superhydrophobic coatings with self-cleaning function have been being generated.7,8 In particular, several one-dimensional nanomaterials with hydrophobic properties are studied because of structures similar to the lotus leaf.9 Rosario et al. found that a rough surface morphology with Si nanowires (SiNWs) on a silicon substrate amplified the light-induced change in water contact angle of a photoresponsive surface.10 Verplanck et al. reported the reversible electrowetting of liquid droplets in air and oil environments on superhydrophobic SiNWs.2 Okamoto et al. investigated the diffusion and wetting at the interior surface of templatesynthesized silica nanotubes by using fluorescence microscopy techniques.11 Although the superhydrophobic coating with high contact angle (CA) was achieved by employing nanowires or nanotubes, the hydrophobic surface was normally fixed on the original growth substrate because of the limitation of quantity. It cannot be conveniently scaled up and coated on the needed surface. Furthermore, to date there are few reports of the mechanical stability of as-obtained superhydrophobic coating. The thermally and chemically durable hydrophobic-oleophobic coating with SiC particles has been developed by a spin-coating method;12 however, the promising SiCNWs superhydrophobic coating on glass has not been studied. In this paper, we report a novel, feasible, self-cleaning glass coating with superhydrophobic SiCNWs. Macroscopic SiCNWs in tetraethyl orthosilicate solution were conveniently coated. The roughness effect on CA was analyzed by coating different * To whom correspondence should be addressed. E-mail: jjniu@ sjtu.edu.cn. Tel.: +86-21-54743182. † Shanghai Jiao Tong University. ‡ Tongji University.
Figure 1. (a) TEM image of the SiCNWs; FESEM images of the SiCNWs coatings with cycles of 3 (b), 5 (c), and 10 (d), respectively.
cycles. A high roughness Ra of 1928.9 nm of 10-cycles coating contributed a high CA of ∼160° and a low sliding angle (SA) of ∼5°. A high CA of ∼145.3° with good chemical stability was observed by soaking the glass in water for 14 days. The composite model with nano/micropillar theory was tentatively used to analyze the superhydrophobic structures. 2. Experimental Details SiCNWs with large scales were directly synthesized by a vapor-solid reaction of carbon oxide and silicon powders in a simple alumina-ceramic tube furnace. ZnS powder was used as catalyst. The detailed process is described in our previous work.4,13 The 0.5 g SiCNWs sample was dropped in a mixture solution composed of TEOS ((CH3CH2O)4Si), ethanol, and hydrochloric acid with a volume ratio of 1:50:1. The mixture was agitated for 30 min by ultrasonic mixing. Subsequently, the SiCNWs sample was separately coated on a smooth glass with different cycles of 3, 5, and 10. The time interval was ∼10
10.1021/jp808322e CCC: $40.75 2009 American Chemical Society Published on Web 02/06/2009
2910 J. Phys. Chem. B, Vol. 113, No. 9, 2009
Figure 2. Roughness Ra of the SiCNWs coatings with different cycles and blank glass.
Niu and Wang
Figure 5. Self-cleaning test with dust on SiCNWs coating and blank glass. Cross-section (a) and top-view (b) photos.
chemically modified by vaporized perfluoroalkysilane gas at a temperature less than 200 °C. A blank glass was also modified as a reference. The chemical stability was investigated by immersing the samples in pure water for 7 (168 h) and 14 (336 h) days at room temperature, respectively. Field emission scanning electric microscopy (FESEM, JSM 7401F) and transmission electron microscopy (TEM, JEM 2010) at an acceleration voltage of 200 kV were carried out to observe the surface morphology, respectively. The surface roughness Ra was determined by Veeco Dektak 6 M Stylus Surface Profiler with a scanning scope of 1 mm. Water contact angle was measured with 3 µL of deionized water by a contact angle system (OCA 20, Dataphysics, Germany). All the CA measurements were checked in ambient atmosphere at room temperature. Because of the slight deviation during the measurement, an average value of CA was used after measuring tens of spots on one surface. The error of calculated result was estimated to be within 0.5°. Figure 3. Water contact angles of blank glass (a), SiCNWs coatings with cycles of 3 (b), 5 (c), and 10 (d), respectively.
min for each cycle. The next coating was started after the former dried at room temperature. Then the obtained sample was
3. Results and Discussion The SiCNWs sample was composed of uniform nanowires with diameters of 20 ∼ 40 nm and numbers of nanoparticles with sizes of 200 ∼ 400 nm. A TEM image of the SiCNWs is shown in Figure 1a. FESEM morphologies of the SiCNWs-
Figure 4. Water contact angles of SiCNWs coating with cycles of 10. Panels a-f show the dynamic contact angle variation during a water drop rolling procedure.
Self-Cleaning Coating with SiCNWs
J. Phys. Chem. B, Vol. 113, No. 9, 2009 2911
Figure 6. Average CA of the SiCNWs coatings of raw (a) and immersed in water for 1 (b) and 2 (c) weeks.
Figure 7. Water contact angles of 10-cycles SiCNWs coating after immersing in water for 2 (a) and 3 (b) weeks.
coated glasses with different cycles are displayed in Figure 1b-d. As can be seen from the figure, plenty of SiCNWs and nanoparticles with various sizes were formed on the surface of glass. Rugged nanostructures produced various fractals with nano/micropores and interstices. In particular, the surface became more concave-convex when the thick coating was covered (Figure 1b-d). Roughness measurement confirmed the Ra was improved almost three cycles from 766.3 to 1928.9 nm with the increasing cycles from 3 to 10 (Figure 2). As a comparison, the roughness Ra of blank glass was only 25.2 nm. Water contact angles of the samples with different roughness are shown in Figure 3. The CA is gradually increased from 152.3, 153.6, to 160.0° when the cycles varied from 3, 5, to 10, respectively (Figure 3 b-d). The large CA more than 150° suggests an ideal superhydrophobic surface while the blank glass displays a low CA of 105.8° (Figure 3a). The self-cleaning surface needs not only a high CA, but a low sliding angle, so a water droplet will easily roll off and the dirt can be removed. As shown in Figure 4, a water droplet rolled off the tilted surface of ∼5° within a very short 0.20 s. The droplet remained spherical even during the rolling. Shown in Figure 5 are the cross-section (a) and top view (b) photos of the self-cleaning function with 10-cycles sample. It is clearly observed that the dust was completely removed and thus a clean surface was obtained by dropping water on the superhydrophobic SiCNWs coating (left). However, a dirty surface with dust remained on the blank glass (right). Therefore, the current superhydrophobic coating can be indeed used in many self-cleaning fields. As for the artificial superhydrophobic coating, the lifetime is a key factor in industrial application. As we know, there are
Figure 8. Water contact angle (cos θr) as a function of surface structure (b2/a2) by using the nano/micro-pillar composite structure model.
few reports on lifetime studies for superhydrophobic coatings. Here we tested the chemical stability of the as-obtained superhydrophobic SiCNWs coating. After immersing the coated glass in water for 7 days (168 h), the CA was only decreased to 151.8° (Figure 6b and Figure 7a). A high CA of 145.3° remained even after immersing for 14 days (Figure 6c and Figure 7b). It indicated that a slight decrease of ∼7° per week was observed. The weak reduction of superhydrophobicity is supposed to be related to Si-O-Si hydrolysis and organic molecules removal due to the wetting environment. The detailed analysis on the structure changes should be further studied. The high chemical stability can be attributed to the chemical reaction of TEOS. The (CH3CH2O)4Si will hydrolyze with H2O to generate silanol Si(OH)4. Furthermore, the shell of the SiCNWs was normally composed of silicon oxides, which will possibly possess OH groups.13 The generated OH groups on silanol will bind with the OH groups on the shell of SiCNWs. After the condensation reaction, the binding of SiCNWs coating to the underlying substrate is remarkably strengthened (Figures 6 and 7). Besides the soaking test, mechanical scraping experiments were also processed. The resulting coating was difficult to remove by scraping with filter paper. But with further sand paper, it could be destroyed; therefore, although the current coating cannot be exposed to extremely terrible circumstances, it can be used in many indoor hydrophobic fields such as air conditioning, nonwetting micro/nanoelectronics, drying machine, etc. Further mechanical tests are underway in following studies.
TABLE 1 x (b2/a2)
0
1
2
2.32
3
4
6
8
10
y (cos θr)
-0.274
-0.819
-0.919
-0.934
-0.955
-0.971
-0.985
-0.991
-0.994
2912 J. Phys. Chem. B, Vol. 113, No. 9, 2009
Niu and Wang
TABLE 2 x (H/a1)
0
1
5
10
13.3
15
20
25
y (cos θr)
-0.274
-0.396
-0.882
-1.491
-1.892
-2.099
-2.71
-3.315
The roughness effect is tentatively analyzed by using the nano/micropillar composite structure model suggested by Patankar et al.14 developed from the Cassie and Wenzel models.1 Although as-obtained SiCNWs were randomly dispersed with a porous sponge-type surface, a lot of which are not vertically aligned, in order to better describe the hydrophobic property, we here try to use the model that relies on well-defined “pillar” geometries to explain the current nano/microstructures. According to the model, the present SiCNWs coating can be divided into two class structures: the first class is composed of micropillars with size of a1 × a1 (height H, interspace b1); the second class is nanopillars connecting to the micropillars, with size of a2 × a2 (interspace b2). If the Cassie model is considered, the function can be described as
cos θr )
1 (1 + cos θ) - 1 [(b2 /a2) + 1]2
where θr is the apparent contact angle and θ is the intrinsic contact angle (here is the CA of blank glass). From this equation, it can be seen that contact angle is independent of pillar height. After considering the nature lotus leaf and other one-dimensional nanomaterials hydrophobic coating,15 we regard the onedimensional SiCNWs with smaller size as a second class of long nanopillars. Thus, the nanopillar dimension is equal to the diameter of smaller SiCNWs, which can be calculated from plenty of nanowires with TEM images. The spacing parameter between nanowires/nanoparticles can be estimated from FESEM images. Here the θ is valued as 105.9°, and we can obtain the data shown in Table 1. According to the size and interspace of SiCNWs observed from TEM and FESEM images, the average diameter of the nanowires is ∼30 nm (a2) and the interspace is ∼70 nm (b2). Thus a high CA of ∼159° is calculated (Figure 8). If the height of first class of micropillars is considered, Wenzel model was used as following:
(
cos θr ) 1 + A)
)
4A cos θ a1 /H
1 [(b1 /a1) + 1]2
The big nanoparticles are regarded as a first class of micropillars. Then a mean size of ∼300 nm and an interspace of ∼600 nm are desired (Figure 1). Thus, series of cos θr data are obtained with the varying H/a1. These are shown in Table 2. With respect to the roughness, the height H of nanoparticles is suggested to be ∼4000 nm. Therefore, the cos θr is calculated to be -1.892. It demonstrated that the current composite structure possess a good superhydrophobic property even with a small value ∼13.3 of H/a1. Although the calculated error is
unavoidable, the results displayed a beneficial understanding of the present superhydrophobic mechanism. Cassie and Baxter’s equation of cos θr ) f1 cos θ - f2 is also used to estimate the ratio of air-trapping action.1 The f1 and f2 are the area fractions of solid and vapor on surface, respectively, and f1 + f2 ) 1. Here the θr is 159.1° and the θ is 105.9°. After calculation, a very high value of f2 ) 0.91 is obtained. This confirms that the air-trapping action plays a key role on superhydrophobic SiCNWs coating. 4. Conclusions In summary, a novel superhydrophobic SiCNWs coating on glass with self-cleaning function was successfully synthesized. Large scales of SiCNWs were dissolved in TEOS solution and can be indeed conveniently coated on glass and other surfaces. A high roughness Ra of 1928.9 nm with thick SiCNWs induced a high CA closing to 160° and a low SA of ∼5°. A better chemical stability was obtained by immersing the coating in water for 14 days. Acknowledgment. This work was sponsored by Shanghai Educational Development Foundation (No. 2007CG14), ShanghaiApplied Materials Research and Development Fund (No. 06SA06), the National Natural Science Foundation of China (50871067), and the National 863 Project of 2007AA05Z128 from the Ministry of Science and Technology of China. We would like to thank Instrumental Analysis Center of Shanghai Jiao Tong University for their great help in measurements. References and Notes (1) Cassie, A. B. D.; Baxter, S. Trans. Faraday Soc. 1944, 40, 546. (2) Verplanck, N.; Galopin, E.; Camart, J. C.; Thomy, V.; Coffinier, Y.; Boukherroub, R. Nano Lett. 2007, 7, 813. (3) Lafuma, A.; Quere, D. Nat. Mater. 2003, 2, 457. (4) Niu, J. J.; Wang, J. N.; Xu, Q. F. Langmuir 2008, 24, 6918. (5) Niu, J. J.; Wang, J. N. Cryst. Growth Des. 2008, 8, 2793. (6) Onda, T.; Shibuichi, S.; Satoh, N.; Tsujii, K. Langmuir 1996, 12, 2125. (7) Lau, K. K. S.; Bico, J.; Teo, K. B. K.; Chhowalla, M.; Amaratunga, G. A. J.; Milne, W. I.; Mckinley, G. H.; Gleason, K. K. Nano Lett. 2003, 3, 1701. (8) Feng, L.; Li, S.; Li, H.; Zhai, J.; Song, Y.; Jiang, L.; Zhu, D. Angew. Chem., Int. Ed. 2002, 41, 1221. (9) Coffinier, Y.; Janel, S.; Addad, A.; Blossey, R.; Gengembre, L.; Payen, E.; Boukherroub, R. Langmuir 2007, 23, 1608. (10) Rosario, R.; Gust, D.; Garcia, A. A.; Hayes, M.; Taraci, J. L.; Clement, T.; Dailey, J. W.; Picraux, S. T. J. Phys. Chem. B 2004, 108, 12640. (11) Okamoto, K.; Shook, C. J.; Bivona, L.; Lee, S. B.; English, D. S. Nano Lett. 2004, 4, 233. (12) Uyanik, M.; Arpac, E.; Schmidt, H.; Akarsu, M.; Sayilkan, F.; Sayilkan, H. J. Appl. Polym. Sci. 2006, 100, 2386. (13) Niu, J. J.; Wang, J. N. J. Phys. Chem. B 2007, 111, 4368. (14) Patankar, N. A. Langmuir 2004, 20, 7097. (15) Feng, L.; Li, S. H.; Li, Y. S.; Li, H. J.; Zhang, L. J.; Song, Y. L.; Liu, B. Q.; Jiang, L.; Zhu, D. B. AdV. Mater. 2002, 14, 1857.
JP808322E
