- Email: [email protected]
Hydrophobic coating on glass surfaces via application of silicone oil and activated using a microwave atmospheric plasma jet
SCT-19633; No of Pages 5 Surface & Coatings Technology xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Surface & Coatings Technology journal homepage: www.elsevier.com/locate/surfcoat
Hydrophobic coating on glass surfaces via application of silicone oil and activated using a microwave atmospheric plasma jet Julie Anne S. Ting a,⁎, Leo Mendel D. Rosario a,b, Henry V. Lee Jr. a, Henry J. Ramos a, Roy B. Tumlos a,c a b c
Plasma Physics Laboratory, National Institute of Physics, University of the Philippines, Diliman, Quezon City 1101, Philippines Sciences Department, College of Arts, Sciences and Education, FEATI University, Sta. Cruz, Manila 1003, Philippines Department of Physical Sciences and Mathematics, College of Arts and Sciences, University of the Philippines, Manila 1000, Philippines
a r t i c l e
i n f o
Available online xxxx Keywords: Glass Silicone oil Microwave plasma Surface activation Hydrophobic coating
a b s t r a c t In this study, a hydrophobic coating on glass surfaces was fabricated by application of a silicone oil lubricant and activated using a microwave atmospheric plasma jet. Optimization of the treatment was done by variation of the working gas flow rates, input microwave power and plasma treatment time, based on contact angle measurements. In comparison with the untreated glass (37.6°), results show that at best discharge conditions of 600 W microwave power, 5/0.5 LPM Ar/N2 flow rate and 10 s treatment time, the plasma-treated glass obtained a water contact angle of 105.7°. Surface energy of the glass also decreased from 45.07 mN/m for the untreated to 27.97 mN/m after plasma treatment. Atomic force microscopy (AFM) and Fourier transform infrared (FTIR) spectroscopy results suggest that increased root-mean-square roughness and introduction of hydrophobic species may have been responsible for the hydrophobicity of the glass surface. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Hydrophobic glass surfaces are widely used in different fields such as in microbiology, optics and microelectronics [1–3]. These surfaces are characterized by low surface energy (γ b 35 mN/m), and high contact angles (θ N 90°) [3,4]. In order to fabricate hydrophobic surfaces, different approaches have been done including surface texturing [5], chemical treatment [6,7], solvent exchange method [8], spin coating technique [9], and plasma treatment [10–15]. Surface treatment of glass using atmospheric plasma has the advantage of modifying surface properties without alteration of the bulk material. In addition, it is fast, effective and relatively cheaper due to the removal of a vacuum system. Previous studies on atmospheric plasma treatment of glass surfaces employed silicone-based oils to form the hydrophobic coating. Silicones are composed of Si\O\Si, Si\O, and freely rotating methyl (CH3) or phenyl (C6H5) groups, rendering them hydrophobic [16]. Three approaches were suggested for hydrophobization of glass including: (i) surface roughening [13], (ii) hydroxylation [3,14], and (iii) plasma-oil interaction [12–15]. For these approaches, dielectric barrier discharge [12–15], plasma torch [17], and radio frequency atmospheric plasmas [3,17] are more frequently used. In terms of efficiency and cost effectiveness, microwave-induced atmospheric plasmas are one of the most attractive atmospheric plasma
⁎ Corresponding author. E-mail address: [email protected] (J.A.S. Ting).
systems due to its lower power and gas flow rate requirements. However, no study has been found implementing this type of atmospheric plasma system for the hydrophobization of glass. As such, the aim of this study is to fabricate hydrophobic coatings on glass surfaces via application of a silicone oil lubricant, and activated using a microwave atmospheric plasma jet. A set of discharge parameters including working gas flow rates, input microwave power and treatment time was optimized based on contact angle measurements. Aging time in air of the hydrophobic glass coating was also determined. 2. Methodology Glass slides (Sail Brand CAT. No. 7101) with dimension 76.2 ×25.4 × 1.0 mm were used in the experiments. A two-stage treatment process was done using the plasma device as discussed in the succeeding sections. 2.1. Microwave atmospheric plasma jet The microwave atmospheric plasma jet (MAPJ) device is composed of a 2.45 GHz magnetron with up to 3 kW continuous wave output, the tuning system and the tapered waveguide as shown in Fig. 1. The tuning system consists of the three-stub tuner and a sliding short. Gases are injected through the gas nozzle located at the top of the tapered waveguide. Microwaves generated by the magnetron propagate through the waveguide assembly and ignites the plasma in a hollow cylindrical quartz discharge tube 20 mm in diameter inside the tapered waveguide. To prevent melting of the discharge tube due to
http://dx.doi.org/10.1016/j.surfcoat.2014.08.008 0257-8972/© 2014 Elsevier B.V. All rights reserved.
Please cite this article as: J.A.S. Ting, et al., Surf. Coat. Technol. (2014), http://dx.doi.org/10.1016/j.surfcoat.2014.08.008
2
J.A.S. Ting et al. / Surface & Coatings Technology xxx (2014) xxx–xxx
using plasma treatment. Treatment was optimized using One-Factorat-a-Time method wherein a baseline set of values is set and each factor/parameter was successively varied while other parameters were kept constant. For this study, the baseline values were 600 W microwave power, 5/0.5 LPM Ar/N2 gas flow rates, and 5 s treatment time. The first parameter varied was input microwave power (300, 400, 500, 550, 600, 700, 800, 900 and 1000 W). Based on the highest average water contact angle (WCA), the best power was determined and used to vary the succeeding parameters which were N2 gas flow rate (0.5, 1, 3, 5 and 7 LPM), treatment time (1, 3, 5, 8, 10, 12 and 15 s) and Ar gas flow rate (4, 5, 6, 7 and 8 LPM). Based on the highest average WCA, the best set of input microwave power, treatment time, and Ar/N2 gas flow rates were determined. Aging time in air of the best sample was also investigated for a period of 50 days to determine the stability of the hydrophobic coating. 2.3. Characterizations Changes in the wettability of the glass samples were determined via water contact angle measurements using Dino-Lite Digital Microscope Premier and DinoCapture 2.0 software. For surface energy calculations, deionized water (DI) and ethylene glycol (EG) were used as polar test liquids, whereas diiodomethane (DM) was used as the non-polar liquid. Test liquids were dropped onto the sample surface using a 0.5 mL syringe. The average of ten measurements was considered as the contact angle of the sample. The chemical composition and surface roughness of the untreated and plasma-treated glass surfaces were analyzed using Fourier transform infrared spectroscopy (FTIR-ATR) and an NT-MDT Solver atomic force microscopy (AFM). AFM images were taken for a sample size of 50 × 50 μm2, and were corrected using Flatten Correction 2D option to remove the “tilt” effect. 2.4. Surface energy calculations
Fig. 1. (a) Schematic diagram and (b) actual image of the microwave atmospheric plasma jet.
plasma heating, two sets of gas ports were created: (a) a central port for the central gas, and (b) eight peripheral gas ports positioned uniformly around the central port for the “shroud” gas. The shroud gas acts as a barrier that protects the wall of the discharge tube from the plasma. This results to a stable plasma that is confined at the center of the discharge tube. In addition, the shroud gas also serves as a secondary gas that partially mixes with the working gas and undergoes ionization as well. A 5.9 kW recirculating chiller serves as the cooling system. 2.2. Plasma treatment Glass slide samples were positioned 3 cm below the discharge tube. Initial plasma treatment was done at 600 W input microwave power and 5/0.5 LPM Ar/N2 gas flow rate, for 10 s treatment time to increase its surface energy and remove any surface contaminations. It is noted that for all treatments, argon (Ar) was injected as the central gas, with nitrogen (N2) serving as a shroud gas. Afterwards, the samples were subjected to further plasma treatment. The hydrophobic coating was formed by application of one drop of silicone oil (3-in-one Professional Silicone Spray Lubricant No. 10041) using a 0.5 mL syringe on the glass surface, then subsequent activation
The Young's equation [Eq. (1)] describes the relation between the interfacial energies and the equilibrium contact angle (θ) formed by a given liquid drop (L) on top of a solid surface (S). The surface energies at the liquid–vapor, solid–vapor, and the solid–liquid interfaces are designated as γLV, γSV, and γSL, respectively. Since interactions with vapor are very low, γLV, is often considered as the surface energy of the liquid (γL), whereas, γSV is the surface energy of the solid surface (γS) [18,19]. In order to calculate the solid surface free energy γS, an estimate of γSL is done through Eq. (1). γLV cos θ ¼ γSV −γSL
ð1Þ
Good, Van Oss and Chaudhury developed an equation expressing the total surface energy γTOT as the sum of the Lifshitz–van der Waals apolar component γLW and Lewis acid–base polar component γAB, as shown in Eq. (2). γLW considers the contributions from van der Waals forces originating from dipole–dipole (Keesom force), and dipole-induced dipole (Debye and London) interactions. On the other hand, γAB takes into account all acid–base interactions such as electron donor–electron acceptor interactions [19–21]. γTOT ¼ γ
LW
þγ
AB
ð2Þ
The acid–base polar component γAB resulting from electron donor (γ−) and electron acceptor (γ+) interactions is given by Eq. (3). Using Eqs. (2) and (3), the surface energy for a solid can be expressed as in Eq. (4) γ
AB
¼2
qffiffiffiffiffiffiffiffiffiffiffiffiffi γþ γ−
Please cite this article as: J.A.S. Ting, et al., Surf. Coat. Technol. (2014), http://dx.doi.org/10.1016/j.surfcoat.2014.08.008
ð3Þ
J.A.S. Ting et al. / Surface & Coatings Technology xxx (2014) xxx–xxx
qffiffiffiffiffiffiffiffiffiffiffiffiffi LW − γS ¼ γS þ 2 γþ S γS
3
ð4Þ
The interfacial energy between the solid and liquid can then be defined by Eq. (5). Finally, by combining Eq. (5) with the Young's equation in Eq. (1), the relation between the measured contact angle, the solid surface and the liquid surface free energy components is expressed as in Eq. (6). qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffi þ LW − γ SL ¼ γS þ γL −2 γLW γþ γ− S γL þ S γL S γL þ
ð5Þ
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffi þ LW − γLW γ− γþ γ L ð cosθ þ 1Þ ¼ 2 S γL þ S γL S γL þ
ð6Þ
+ Calculation of the three surface free energy components γLW s , γs and of the solid is done using three independent linear equations, taking into account the measured contact angles of one nonpolar and two polar − AB liquids with known surface components (γL, γ+ L , γL , γL ) [18,19].
γ− s
Fig. 3. Effect of varying nitrogen gas flow rate on water contact angles of plasma-treated glass [microwave power = 600 W, treatment time = 5 s, Ar gas flow rate = 5 LPM].
The increase in surface energy leads to a corresponding decrease in water contact angles (WCA) [14,22]. After the first plasma treatment, WCA of the glass samples significantly decreased from 37.6° ± 1.6° for the untreated to 5.0° ± 1.3° for the plasma-treated, rendering the surface superhydrophilic. WCA values of treated glass surfaces with hydrophobic coatings are depicted in Figs. 2–5. The effect of input microwave power on WCA is shown in Fig. 2. Below 600 W, relatively lower WCAs were obtained in the range 41°–65°. Increasing the power to 600 W leads to a hydrophobic glass surface, with minimal changes in WCA for higher input power. Therefore, 600 W is considered as the best input power. Results were attributed to the increased energy provided by the microwave power for the generation of more energetic plasma species. These reactive species may be responsible for the decomposition of silicone oil into CHx-containing groups that could form the hydrophobic coating, thereby preventing further reaction of the water molecules with the polar bonds of the glass surface [12,14,15]. In Fig. 3, the effect of varying N2 shroud gas flow rate is shown. In this study, for a fixed Ar gas flow rate of 5 LPM, the addition of a shroud gas effectively isolates the central Ar gas flow from the discharge tube to prevent overheating. In addition, WCA is increased by 34.3°. This result may be a consequence of the confining effect of the N2 shroud gas, allowing more Ar species to reach and consequently react with the
silicone oil on the sample surface. With varying N2 gas flow rate from 0.5 LPM to 7 LPM, no significant changes in WCA were observed. Hence, the best N2 flow rate is 0.5 LPM with a WCA of 96.9°. A gradual increase in WCA was observed for longer treatment times as shown in Fig. 4. A minimum of 5 s is required to obtain hydrophobic glass surface with WCA greater than 100°. This result may indicate that a hydrophobic coating has already been generated on the glass surface due to the decomposition of silicone oil. In addition, the best treatment time is 10 s for which the highest measured WCA of 115.4° was obtained. Full decomposition of the oil may have occurred as evidenced by minimal changes in WCA for longer treatments. Finally, variation of Ar gas flow rate is given in Fig. 5. A minimum flow rate of 4 LPM was used due to overheating and breaking of glass samples at much lower flow rates of 1–3 LPM. At 4 LPM, a relatively lower WCA of 88.4° was measured. Increasing the gas flow rate to 5 LPM resulted to a higher WCA (105.7°), with minimal changes when further increased until 8 LPM (101.4°). These results were attributed to the density of Ar species ionized at the given amount of energy for the activation of silicone oil on the glass surfaces. Hence, 5 LPM is considered as the best Ar gas flow rate. In summary, the best discharge conditions found were 600 W input microwave power, 5/0.5 LPM Ar/N2 gas flow rate, and 10 s treatment time, in which an average WCA of 105.7° ± 4.2° was obtained. Initial plasma treatment improved the hydrophilicity of the glass surface, in preparation for the succeeding treatment with silicone oil. Comparison of the WCA values showed that less hydrophobicity, as indicated by a relatively lower WCA (98.8° ± 2.2°) was measured for surfaces that did not undergo the first treatment. Generation of a hydrophobic surface can then be attributed not only to increased surface roughness and surface energy, but also to the formation of a hydrophobic coating as will be shown in the following sections.
Fig. 2. Effect of varying input microwave power on water contact angles of plasma-treated glass [N2 gas flow rate\0.5 LPM, treatment time = 5 s, Ar gas flow rate = 5 LPM].
Fig. 4. Effect of varying treatment time on water contact angles of plasma-treated glass [microwave power = 600 W, N2 gas flow rate\0.5 LPM, Ar gas flow rate = 5 LPM].
3. Results 3.1. Variation of discharge conditions
Please cite this article as: J.A.S. Ting, et al., Surf. Coat. Technol. (2014), http://dx.doi.org/10.1016/j.surfcoat.2014.08.008
4
J.A.S. Ting et al. / Surface & Coatings Technology xxx (2014) xxx–xxx
Fig. 5. Effect of varying Ar gas flow rate on water contact angles of plasma-treated glass [microwave power = 600 W, N2 gas flow rate = 0.5 LPM, treatment time = 10 s]. Fig. 6. FTIR spectra of the untreated and plasma-treated glass surfaces.
3.2. Surface roughness and chemical composition 3.4. Stability of hydrophobic coating Based from AFM results, the untreated glass has a root-mean-square roughness (RRMS) value of 11 nm. Surface bombardment during the first plasma treatment led to almost 4 times increase in surface roughness with an RRMS value of 43 nm. After the activation of the silicone oil, the surface exhibited a relatively lower RRMS value of 36 nm. Nevertheless, the treated surface remained significantly rougher compared with the untreated (Table 1). From Fig. 6, the FTIR spectrum of the untreated glass surface shows peaks centered at 767.7 cm−1, 918.1 cm− 1 and 2918.3 cm−1, corresponding to Si\O\Si, Si\O− and CH2 bonds [23]. Similar peaks were found after the initial plasma treatment, in exception for the disappearance of the CH2 peak verifying the removal of hydrocarbon surface contaminations. On the other hand, FTIR spectrum obtained for the treated sample after the second treatment shows the addition of new peaks including Si\O\Si (1008.7 cm−1), Si-CH3 (1259.5 cm−1), CH2 and CH3 (2854.6 cm−1, 2927.9 cm−1, and 2962.7 cm−1). Apparent decrease in the intensity of Si\O− peak and increase in the intensities of Si\O\Si peaks may be caused by changes in surface functional groups from Si\O− to Si\O\Si bonding [23]. On the other hand, appearance of Si-CH3, CH2 and CH3 peaks are attributed to the decomposition of the silicone oil upon exposure to plasma. These hydrophobic radicals can react with the glass to form a hydrophobic coating [12,15].
Stability of the hydrophobic coating on glass surfaces was determined by aging tests in air for a period of 50 days. As shown in Fig. 7, minimal changes in WCA were observed from day 1 (96.9°). Overall, analysis of the measured WCA indicates that the hydrophobic coating of the treated glass remains stable until day 50 (96.3°).
4. Discussion
The calculated surface energy of the untreated and plasma-treated glass surfaces is shown in Table 2. An increase in the total surface energy (γTOT) of the glass from 45.07 mN/m to 46.43 mN/m was observed after the first plasma treatment due to an increase in van der Waals contribuAB tions (γLW S ) and a decrease in acid–base interactions (γ ). After the AB second treatment, a decrease in both γLW and γ components led to a S considerable decrease in the calculated γTOT (27.97 mN/m). The low total surface energy (b30 mN/m) verifies the hydrophobic property of the plasma-treated glass. Furthermore, this result may be attributed to changes in functional groups as observed in the FTIR results, or a re-orientation of the functional groups from the surface to the bulk material [23].
During the first treatment, exposure to plasma cleans glass surfaces by removal of surface contaminants, which consequently leads to rougher surfaces as verified by the large increase in RRMS [14,24]. In addition, an activated surface may have been formed due to the breaking of the Si\O bonds of the glass to form chemically active dangling bonds. According to a previous study by Wang and Xe [14], initial plasma treatment generates an active layer of hydroxyls on the glass surface. The hydrophilic nature of the hydroxyl groups combined with an increase in surface roughness improves the hydrophilicity of the glass. However, FTIR spectrum after the first treatment shows no peaks corresponding to hydroxyl bonds. Therefore, an observed decrease in contact angle (5.0° ± 1.3°) and an increase in surface energy (46.43 mN/m) can be attributed to the large increase in surface RRMS roughness (43 nm). According to Wenzel's theory, surface roughness improves the hydrophilicity of surfaces with a contact angle less than 90°, and in the same way enhances the hydrophobicity of surfaces with a contact angle greater than 90° [5,25]. When silicone oil applied to the treated glass surface was exposed to the second stage of treatment, the plasma decomposed the ethoxy bonds of the oil to produce hydrophobic CH2, and CH3 groups [3]. These hydrophobic groups shield the glass by covering wateradsorbing sites. As a consequence, any hydrogen bonding sites competing with van der Waals interactions decreases [4]. Moreover, anchor sites may have also been provided for the hydrophobic groups by reaction with the activated glass surface to form oxane (Si\O\Si) bonds [3,13,14]. Influences of surface roughening and formation of Si\O\Si bonds on the generation and adhesion of the hydrophobic coatings on glass surfaces were found similar to previous results by Ward et al. [13].
Table 1 AFM root-mean-square roughness (RRMS) values for the untreated and plasma-treated glass surfaces.
Table 2 Surface energy calculations (mN/m) for the untreated and plasma-treated glass surfaces.
3.3. Surface energy
Glass surface
RRMS roughness (nm)
Glass surface
γLW S
γAB
γTOT
Untreated First plasma treatment Second plasma treatment
11 43 36
Untreated First plasma treatment Second plasma treatment
38.53 44.74 27.34
6.54 1.69 0.63
45.07 46.43 27.97
Please cite this article as: J.A.S. Ting, et al., Surf. Coat. Technol. (2014), http://dx.doi.org/10.1016/j.surfcoat.2014.08.008
J.A.S. Ting et al. / Surface & Coatings Technology xxx (2014) xxx–xxx
5
References
Fig. 7. Effect of aging time on water contact angle of the best hydrophobic glass surface for a period of 50 days [microwave power = 600 W, N2 gas flow rate = 0.5 LPM, treatment time = 10 s, Ar gas flow rate = 5 LPM].
5. Conclusions Microwave atmospheric plasma jet was employed to generate hydrophobic coatings on glass surfaces via application of silicone oil. Initial plasma treatment increased surface roughness in preparation for the succeeding treatment. At best discharge conditions of 600 W microwave power, 5/0.5 LPM Ar/N2 gas flow rates and 10 s treatment time, a hydrophobic glass surface with a WCA of 105.7° and low total surface energy of 27.97 mN/m was generated. Analysis of the glass surfaces attributed its hydrophobic property to increased RRMS surface roughness, formation of Si\O\Si bonds and introduction of hydrophobic CH2 and CH3 bonds on the glass surface. Moreover, the hydrophobic coating of the plasma-treated glass was found to remain stable after 50 days.
[1] H. Tsuji, P. Sommani, Y. Hayashi, H. Kojima, H. Sato, Y. Gotoh, G. Takaoka, J. Ishikawa, Surf. Coat. Technol. 206 (2011) 900–904. [2] J. Philipavičius, I. Kazadojev, A. Behanskienė, A. Melninkaitis, V. Sirutkaitis, A. Kareiva, Mater. Sci. 14 (4) (2008). [3] D. Barankin, E. Gonzalez II, S.B. Habib, L. Gao, P.C. Guschl, R.F. Hicks, Langmuir 25 (2009) 2495–2500. [4] Y.Y. Yan, N. Gao, W. Barthlott, Adv. Colloid Interf. Sci. 169 (2011) 80–105. [5] S. Beckford, N. Langston, M. Zou, R. Wei, Appl. Surf. Sci. 257 (2011) 5688–5693. [6] S.H. Gao, L.H. Gao, K.S. Zhou, Appl. Surf. Sci. 257 (2011) 4945–4950. [7] S. Bhattacharya, A. Datta, J.M. Berg, S. Gangopadhyay, J. Microelectromech. Syst. 14 (3) (2005). [8] S. Bhattacharya, Y. Gao, V. Korampally, M.T. Othman, S.A. Grant, K. Gangopadhyay, S. Gangopadhyay, Appl. Surf. Sci. 253 (2007) 4220–4225. [9] C. Elif Cansoy, H. Yildirim Erbil, O. Akar, T. Akin, Colloids Surf. A Physicochem. Eng. Asp. 386 (2011) 116–124. [10] W. Huang, L. Jiang, C. Zhou, X. Wang, Tribol. Int. 52 (2012) 87–93. [11] R.A. Lawton, C.R. Price, A.F. Runge, W.J. Doherty III, S. Saavedra, Colloids Surf. A Physicochem. Eng. Asp. 253 (2005) 213–215. [12] Z. Fang, Y. Qiu, E. Kuffel, J. Phys. D. Appl. Phys. 37 (2004) 2261–2266. [13] L.J. Ward, J.P.S. Badyal, A.J. Goodwin, P.J. Merlin, Polymer 46 (2005) 3986–3991. [14] C. Wang, X. He, Appl. Surf. Sci. 252 (2006) 8348–8351. [15] Z. Fang, Y. Qiu, H. Wang, E. Kuffel, Plasma Sci. Technol. 9 (5) (2007). [16] P. Somasundaran, S.C. Mehta, P. Purohit, Adv. Colloid Interf. Sci. 128–130 (2006) 103–109. [17] K.S. Chen, S.C. Liao, S.H. Tsao, N. Inagaki, H.M. Wu, C.Y. Chou, W.Y. Chen, Surf. Coat. Technol. 228 (Supplement 1) (2013) s33–s36. [18] Q. Bénard, M. Foi, M. Grisel, Appl. Surf. Sci. 253 (2007) 4753–4758. [19] M. Żenkiewicz, J. Achiev. Mater. Manuf. Eng. 24 (1) (2007). [20] A. Carre, J. Adhes. Sci. Technol. 21 (2007) 961–981. [21] C.J. Van Oss, R.F. Giese, Clay Clay Miner. 43 (4) (1995) 474–477. [22] S.W. Choi, W.B. Choi, Y.H. Lee, B.K. Ju, J. Korean Phys. Soc. 38 (3) (2001) 207–209. [23] J. Abenojar, M.A. Martínez, N. Encinas, F. Velasco, Int. J. Adhes. Adhes. 44 (2013) 1–8. [24] J.A.S. Ting, L.M.D. Rosario, M.C.C. Lacdan, H.V. Lee Jr., J.C. De Vero, H.J. Ramos, R.B. Tumlos, Int. J. Adhes. Adhes. 40 (2013) 64–69. [25] K. Liu, Y. Tian, L. Jiang, Prog. Mater. Sci. 58 (2013) 503–564.
Acknowledgments The authors would like to thank IBF Electronics GmbH & Co. KG and the Department of Science and Technology (DOST) for the financial support provided in this research.
Please cite this article as: J.A.S. Ting, et al., Surf. Coat. Technol. (2014), http://dx.doi.org/10.1016/j.surfcoat.2014.08.008
Contents lists available at ScienceDirect
Surface & Coatings Technology journal homepage: www.elsevier.com/locate/surfcoat
Hydrophobic coating on glass surfaces via application of silicone oil and activated using a microwave atmospheric plasma jet Julie Anne S. Ting a,⁎, Leo Mendel D. Rosario a,b, Henry V. Lee Jr. a, Henry J. Ramos a, Roy B. Tumlos a,c a b c
Plasma Physics Laboratory, National Institute of Physics, University of the Philippines, Diliman, Quezon City 1101, Philippines Sciences Department, College of Arts, Sciences and Education, FEATI University, Sta. Cruz, Manila 1003, Philippines Department of Physical Sciences and Mathematics, College of Arts and Sciences, University of the Philippines, Manila 1000, Philippines
a r t i c l e
i n f o
Available online xxxx Keywords: Glass Silicone oil Microwave plasma Surface activation Hydrophobic coating
a b s t r a c t In this study, a hydrophobic coating on glass surfaces was fabricated by application of a silicone oil lubricant and activated using a microwave atmospheric plasma jet. Optimization of the treatment was done by variation of the working gas flow rates, input microwave power and plasma treatment time, based on contact angle measurements. In comparison with the untreated glass (37.6°), results show that at best discharge conditions of 600 W microwave power, 5/0.5 LPM Ar/N2 flow rate and 10 s treatment time, the plasma-treated glass obtained a water contact angle of 105.7°. Surface energy of the glass also decreased from 45.07 mN/m for the untreated to 27.97 mN/m after plasma treatment. Atomic force microscopy (AFM) and Fourier transform infrared (FTIR) spectroscopy results suggest that increased root-mean-square roughness and introduction of hydrophobic species may have been responsible for the hydrophobicity of the glass surface. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Hydrophobic glass surfaces are widely used in different fields such as in microbiology, optics and microelectronics [1–3]. These surfaces are characterized by low surface energy (γ b 35 mN/m), and high contact angles (θ N 90°) [3,4]. In order to fabricate hydrophobic surfaces, different approaches have been done including surface texturing [5], chemical treatment [6,7], solvent exchange method [8], spin coating technique [9], and plasma treatment [10–15]. Surface treatment of glass using atmospheric plasma has the advantage of modifying surface properties without alteration of the bulk material. In addition, it is fast, effective and relatively cheaper due to the removal of a vacuum system. Previous studies on atmospheric plasma treatment of glass surfaces employed silicone-based oils to form the hydrophobic coating. Silicones are composed of Si\O\Si, Si\O, and freely rotating methyl (CH3) or phenyl (C6H5) groups, rendering them hydrophobic [16]. Three approaches were suggested for hydrophobization of glass including: (i) surface roughening [13], (ii) hydroxylation [3,14], and (iii) plasma-oil interaction [12–15]. For these approaches, dielectric barrier discharge [12–15], plasma torch [17], and radio frequency atmospheric plasmas [3,17] are more frequently used. In terms of efficiency and cost effectiveness, microwave-induced atmospheric plasmas are one of the most attractive atmospheric plasma
⁎ Corresponding author. E-mail address: [email protected] (J.A.S. Ting).
systems due to its lower power and gas flow rate requirements. However, no study has been found implementing this type of atmospheric plasma system for the hydrophobization of glass. As such, the aim of this study is to fabricate hydrophobic coatings on glass surfaces via application of a silicone oil lubricant, and activated using a microwave atmospheric plasma jet. A set of discharge parameters including working gas flow rates, input microwave power and treatment time was optimized based on contact angle measurements. Aging time in air of the hydrophobic glass coating was also determined. 2. Methodology Glass slides (Sail Brand CAT. No. 7101) with dimension 76.2 ×25.4 × 1.0 mm were used in the experiments. A two-stage treatment process was done using the plasma device as discussed in the succeeding sections. 2.1. Microwave atmospheric plasma jet The microwave atmospheric plasma jet (MAPJ) device is composed of a 2.45 GHz magnetron with up to 3 kW continuous wave output, the tuning system and the tapered waveguide as shown in Fig. 1. The tuning system consists of the three-stub tuner and a sliding short. Gases are injected through the gas nozzle located at the top of the tapered waveguide. Microwaves generated by the magnetron propagate through the waveguide assembly and ignites the plasma in a hollow cylindrical quartz discharge tube 20 mm in diameter inside the tapered waveguide. To prevent melting of the discharge tube due to
http://dx.doi.org/10.1016/j.surfcoat.2014.08.008 0257-8972/© 2014 Elsevier B.V. All rights reserved.
Please cite this article as: J.A.S. Ting, et al., Surf. Coat. Technol. (2014), http://dx.doi.org/10.1016/j.surfcoat.2014.08.008
2
J.A.S. Ting et al. / Surface & Coatings Technology xxx (2014) xxx–xxx
using plasma treatment. Treatment was optimized using One-Factorat-a-Time method wherein a baseline set of values is set and each factor/parameter was successively varied while other parameters were kept constant. For this study, the baseline values were 600 W microwave power, 5/0.5 LPM Ar/N2 gas flow rates, and 5 s treatment time. The first parameter varied was input microwave power (300, 400, 500, 550, 600, 700, 800, 900 and 1000 W). Based on the highest average water contact angle (WCA), the best power was determined and used to vary the succeeding parameters which were N2 gas flow rate (0.5, 1, 3, 5 and 7 LPM), treatment time (1, 3, 5, 8, 10, 12 and 15 s) and Ar gas flow rate (4, 5, 6, 7 and 8 LPM). Based on the highest average WCA, the best set of input microwave power, treatment time, and Ar/N2 gas flow rates were determined. Aging time in air of the best sample was also investigated for a period of 50 days to determine the stability of the hydrophobic coating. 2.3. Characterizations Changes in the wettability of the glass samples were determined via water contact angle measurements using Dino-Lite Digital Microscope Premier and DinoCapture 2.0 software. For surface energy calculations, deionized water (DI) and ethylene glycol (EG) were used as polar test liquids, whereas diiodomethane (DM) was used as the non-polar liquid. Test liquids were dropped onto the sample surface using a 0.5 mL syringe. The average of ten measurements was considered as the contact angle of the sample. The chemical composition and surface roughness of the untreated and plasma-treated glass surfaces were analyzed using Fourier transform infrared spectroscopy (FTIR-ATR) and an NT-MDT Solver atomic force microscopy (AFM). AFM images were taken for a sample size of 50 × 50 μm2, and were corrected using Flatten Correction 2D option to remove the “tilt” effect. 2.4. Surface energy calculations
Fig. 1. (a) Schematic diagram and (b) actual image of the microwave atmospheric plasma jet.
plasma heating, two sets of gas ports were created: (a) a central port for the central gas, and (b) eight peripheral gas ports positioned uniformly around the central port for the “shroud” gas. The shroud gas acts as a barrier that protects the wall of the discharge tube from the plasma. This results to a stable plasma that is confined at the center of the discharge tube. In addition, the shroud gas also serves as a secondary gas that partially mixes with the working gas and undergoes ionization as well. A 5.9 kW recirculating chiller serves as the cooling system. 2.2. Plasma treatment Glass slide samples were positioned 3 cm below the discharge tube. Initial plasma treatment was done at 600 W input microwave power and 5/0.5 LPM Ar/N2 gas flow rate, for 10 s treatment time to increase its surface energy and remove any surface contaminations. It is noted that for all treatments, argon (Ar) was injected as the central gas, with nitrogen (N2) serving as a shroud gas. Afterwards, the samples were subjected to further plasma treatment. The hydrophobic coating was formed by application of one drop of silicone oil (3-in-one Professional Silicone Spray Lubricant No. 10041) using a 0.5 mL syringe on the glass surface, then subsequent activation
The Young's equation [Eq. (1)] describes the relation between the interfacial energies and the equilibrium contact angle (θ) formed by a given liquid drop (L) on top of a solid surface (S). The surface energies at the liquid–vapor, solid–vapor, and the solid–liquid interfaces are designated as γLV, γSV, and γSL, respectively. Since interactions with vapor are very low, γLV, is often considered as the surface energy of the liquid (γL), whereas, γSV is the surface energy of the solid surface (γS) [18,19]. In order to calculate the solid surface free energy γS, an estimate of γSL is done through Eq. (1). γLV cos θ ¼ γSV −γSL
ð1Þ
Good, Van Oss and Chaudhury developed an equation expressing the total surface energy γTOT as the sum of the Lifshitz–van der Waals apolar component γLW and Lewis acid–base polar component γAB, as shown in Eq. (2). γLW considers the contributions from van der Waals forces originating from dipole–dipole (Keesom force), and dipole-induced dipole (Debye and London) interactions. On the other hand, γAB takes into account all acid–base interactions such as electron donor–electron acceptor interactions [19–21]. γTOT ¼ γ
LW
þγ
AB
ð2Þ
The acid–base polar component γAB resulting from electron donor (γ−) and electron acceptor (γ+) interactions is given by Eq. (3). Using Eqs. (2) and (3), the surface energy for a solid can be expressed as in Eq. (4) γ
AB
¼2
qffiffiffiffiffiffiffiffiffiffiffiffiffi γþ γ−
Please cite this article as: J.A.S. Ting, et al., Surf. Coat. Technol. (2014), http://dx.doi.org/10.1016/j.surfcoat.2014.08.008
ð3Þ
J.A.S. Ting et al. / Surface & Coatings Technology xxx (2014) xxx–xxx
qffiffiffiffiffiffiffiffiffiffiffiffiffi LW − γS ¼ γS þ 2 γþ S γS
3
ð4Þ
The interfacial energy between the solid and liquid can then be defined by Eq. (5). Finally, by combining Eq. (5) with the Young's equation in Eq. (1), the relation between the measured contact angle, the solid surface and the liquid surface free energy components is expressed as in Eq. (6). qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffi þ LW − γ SL ¼ γS þ γL −2 γLW γþ γ− S γL þ S γL S γL þ
ð5Þ
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffi þ LW − γLW γ− γþ γ L ð cosθ þ 1Þ ¼ 2 S γL þ S γL S γL þ
ð6Þ
+ Calculation of the three surface free energy components γLW s , γs and of the solid is done using three independent linear equations, taking into account the measured contact angles of one nonpolar and two polar − AB liquids with known surface components (γL, γ+ L , γL , γL ) [18,19].
γ− s
Fig. 3. Effect of varying nitrogen gas flow rate on water contact angles of plasma-treated glass [microwave power = 600 W, treatment time = 5 s, Ar gas flow rate = 5 LPM].
The increase in surface energy leads to a corresponding decrease in water contact angles (WCA) [14,22]. After the first plasma treatment, WCA of the glass samples significantly decreased from 37.6° ± 1.6° for the untreated to 5.0° ± 1.3° for the plasma-treated, rendering the surface superhydrophilic. WCA values of treated glass surfaces with hydrophobic coatings are depicted in Figs. 2–5. The effect of input microwave power on WCA is shown in Fig. 2. Below 600 W, relatively lower WCAs were obtained in the range 41°–65°. Increasing the power to 600 W leads to a hydrophobic glass surface, with minimal changes in WCA for higher input power. Therefore, 600 W is considered as the best input power. Results were attributed to the increased energy provided by the microwave power for the generation of more energetic plasma species. These reactive species may be responsible for the decomposition of silicone oil into CHx-containing groups that could form the hydrophobic coating, thereby preventing further reaction of the water molecules with the polar bonds of the glass surface [12,14,15]. In Fig. 3, the effect of varying N2 shroud gas flow rate is shown. In this study, for a fixed Ar gas flow rate of 5 LPM, the addition of a shroud gas effectively isolates the central Ar gas flow from the discharge tube to prevent overheating. In addition, WCA is increased by 34.3°. This result may be a consequence of the confining effect of the N2 shroud gas, allowing more Ar species to reach and consequently react with the
silicone oil on the sample surface. With varying N2 gas flow rate from 0.5 LPM to 7 LPM, no significant changes in WCA were observed. Hence, the best N2 flow rate is 0.5 LPM with a WCA of 96.9°. A gradual increase in WCA was observed for longer treatment times as shown in Fig. 4. A minimum of 5 s is required to obtain hydrophobic glass surface with WCA greater than 100°. This result may indicate that a hydrophobic coating has already been generated on the glass surface due to the decomposition of silicone oil. In addition, the best treatment time is 10 s for which the highest measured WCA of 115.4° was obtained. Full decomposition of the oil may have occurred as evidenced by minimal changes in WCA for longer treatments. Finally, variation of Ar gas flow rate is given in Fig. 5. A minimum flow rate of 4 LPM was used due to overheating and breaking of glass samples at much lower flow rates of 1–3 LPM. At 4 LPM, a relatively lower WCA of 88.4° was measured. Increasing the gas flow rate to 5 LPM resulted to a higher WCA (105.7°), with minimal changes when further increased until 8 LPM (101.4°). These results were attributed to the density of Ar species ionized at the given amount of energy for the activation of silicone oil on the glass surfaces. Hence, 5 LPM is considered as the best Ar gas flow rate. In summary, the best discharge conditions found were 600 W input microwave power, 5/0.5 LPM Ar/N2 gas flow rate, and 10 s treatment time, in which an average WCA of 105.7° ± 4.2° was obtained. Initial plasma treatment improved the hydrophilicity of the glass surface, in preparation for the succeeding treatment with silicone oil. Comparison of the WCA values showed that less hydrophobicity, as indicated by a relatively lower WCA (98.8° ± 2.2°) was measured for surfaces that did not undergo the first treatment. Generation of a hydrophobic surface can then be attributed not only to increased surface roughness and surface energy, but also to the formation of a hydrophobic coating as will be shown in the following sections.
Fig. 2. Effect of varying input microwave power on water contact angles of plasma-treated glass [N2 gas flow rate\0.5 LPM, treatment time = 5 s, Ar gas flow rate = 5 LPM].
Fig. 4. Effect of varying treatment time on water contact angles of plasma-treated glass [microwave power = 600 W, N2 gas flow rate\0.5 LPM, Ar gas flow rate = 5 LPM].
3. Results 3.1. Variation of discharge conditions
Please cite this article as: J.A.S. Ting, et al., Surf. Coat. Technol. (2014), http://dx.doi.org/10.1016/j.surfcoat.2014.08.008
4
J.A.S. Ting et al. / Surface & Coatings Technology xxx (2014) xxx–xxx
Fig. 5. Effect of varying Ar gas flow rate on water contact angles of plasma-treated glass [microwave power = 600 W, N2 gas flow rate = 0.5 LPM, treatment time = 10 s]. Fig. 6. FTIR spectra of the untreated and plasma-treated glass surfaces.
3.2. Surface roughness and chemical composition 3.4. Stability of hydrophobic coating Based from AFM results, the untreated glass has a root-mean-square roughness (RRMS) value of 11 nm. Surface bombardment during the first plasma treatment led to almost 4 times increase in surface roughness with an RRMS value of 43 nm. After the activation of the silicone oil, the surface exhibited a relatively lower RRMS value of 36 nm. Nevertheless, the treated surface remained significantly rougher compared with the untreated (Table 1). From Fig. 6, the FTIR spectrum of the untreated glass surface shows peaks centered at 767.7 cm−1, 918.1 cm− 1 and 2918.3 cm−1, corresponding to Si\O\Si, Si\O− and CH2 bonds [23]. Similar peaks were found after the initial plasma treatment, in exception for the disappearance of the CH2 peak verifying the removal of hydrocarbon surface contaminations. On the other hand, FTIR spectrum obtained for the treated sample after the second treatment shows the addition of new peaks including Si\O\Si (1008.7 cm−1), Si-CH3 (1259.5 cm−1), CH2 and CH3 (2854.6 cm−1, 2927.9 cm−1, and 2962.7 cm−1). Apparent decrease in the intensity of Si\O− peak and increase in the intensities of Si\O\Si peaks may be caused by changes in surface functional groups from Si\O− to Si\O\Si bonding [23]. On the other hand, appearance of Si-CH3, CH2 and CH3 peaks are attributed to the decomposition of the silicone oil upon exposure to plasma. These hydrophobic radicals can react with the glass to form a hydrophobic coating [12,15].
Stability of the hydrophobic coating on glass surfaces was determined by aging tests in air for a period of 50 days. As shown in Fig. 7, minimal changes in WCA were observed from day 1 (96.9°). Overall, analysis of the measured WCA indicates that the hydrophobic coating of the treated glass remains stable until day 50 (96.3°).
4. Discussion
The calculated surface energy of the untreated and plasma-treated glass surfaces is shown in Table 2. An increase in the total surface energy (γTOT) of the glass from 45.07 mN/m to 46.43 mN/m was observed after the first plasma treatment due to an increase in van der Waals contribuAB tions (γLW S ) and a decrease in acid–base interactions (γ ). After the AB second treatment, a decrease in both γLW and γ components led to a S considerable decrease in the calculated γTOT (27.97 mN/m). The low total surface energy (b30 mN/m) verifies the hydrophobic property of the plasma-treated glass. Furthermore, this result may be attributed to changes in functional groups as observed in the FTIR results, or a re-orientation of the functional groups from the surface to the bulk material [23].
During the first treatment, exposure to plasma cleans glass surfaces by removal of surface contaminants, which consequently leads to rougher surfaces as verified by the large increase in RRMS [14,24]. In addition, an activated surface may have been formed due to the breaking of the Si\O bonds of the glass to form chemically active dangling bonds. According to a previous study by Wang and Xe [14], initial plasma treatment generates an active layer of hydroxyls on the glass surface. The hydrophilic nature of the hydroxyl groups combined with an increase in surface roughness improves the hydrophilicity of the glass. However, FTIR spectrum after the first treatment shows no peaks corresponding to hydroxyl bonds. Therefore, an observed decrease in contact angle (5.0° ± 1.3°) and an increase in surface energy (46.43 mN/m) can be attributed to the large increase in surface RRMS roughness (43 nm). According to Wenzel's theory, surface roughness improves the hydrophilicity of surfaces with a contact angle less than 90°, and in the same way enhances the hydrophobicity of surfaces with a contact angle greater than 90° [5,25]. When silicone oil applied to the treated glass surface was exposed to the second stage of treatment, the plasma decomposed the ethoxy bonds of the oil to produce hydrophobic CH2, and CH3 groups [3]. These hydrophobic groups shield the glass by covering wateradsorbing sites. As a consequence, any hydrogen bonding sites competing with van der Waals interactions decreases [4]. Moreover, anchor sites may have also been provided for the hydrophobic groups by reaction with the activated glass surface to form oxane (Si\O\Si) bonds [3,13,14]. Influences of surface roughening and formation of Si\O\Si bonds on the generation and adhesion of the hydrophobic coatings on glass surfaces were found similar to previous results by Ward et al. [13].
Table 1 AFM root-mean-square roughness (RRMS) values for the untreated and plasma-treated glass surfaces.
Table 2 Surface energy calculations (mN/m) for the untreated and plasma-treated glass surfaces.
3.3. Surface energy
Glass surface
RRMS roughness (nm)
Glass surface
γLW S
γAB
γTOT
Untreated First plasma treatment Second plasma treatment
11 43 36
Untreated First plasma treatment Second plasma treatment
38.53 44.74 27.34
6.54 1.69 0.63
45.07 46.43 27.97
Please cite this article as: J.A.S. Ting, et al., Surf. Coat. Technol. (2014), http://dx.doi.org/10.1016/j.surfcoat.2014.08.008
J.A.S. Ting et al. / Surface & Coatings Technology xxx (2014) xxx–xxx
5
References
Fig. 7. Effect of aging time on water contact angle of the best hydrophobic glass surface for a period of 50 days [microwave power = 600 W, N2 gas flow rate = 0.5 LPM, treatment time = 10 s, Ar gas flow rate = 5 LPM].
5. Conclusions Microwave atmospheric plasma jet was employed to generate hydrophobic coatings on glass surfaces via application of silicone oil. Initial plasma treatment increased surface roughness in preparation for the succeeding treatment. At best discharge conditions of 600 W microwave power, 5/0.5 LPM Ar/N2 gas flow rates and 10 s treatment time, a hydrophobic glass surface with a WCA of 105.7° and low total surface energy of 27.97 mN/m was generated. Analysis of the glass surfaces attributed its hydrophobic property to increased RRMS surface roughness, formation of Si\O\Si bonds and introduction of hydrophobic CH2 and CH3 bonds on the glass surface. Moreover, the hydrophobic coating of the plasma-treated glass was found to remain stable after 50 days.
[1] H. Tsuji, P. Sommani, Y. Hayashi, H. Kojima, H. Sato, Y. Gotoh, G. Takaoka, J. Ishikawa, Surf. Coat. Technol. 206 (2011) 900–904. [2] J. Philipavičius, I. Kazadojev, A. Behanskienė, A. Melninkaitis, V. Sirutkaitis, A. Kareiva, Mater. Sci. 14 (4) (2008). [3] D. Barankin, E. Gonzalez II, S.B. Habib, L. Gao, P.C. Guschl, R.F. Hicks, Langmuir 25 (2009) 2495–2500. [4] Y.Y. Yan, N. Gao, W. Barthlott, Adv. Colloid Interf. Sci. 169 (2011) 80–105. [5] S. Beckford, N. Langston, M. Zou, R. Wei, Appl. Surf. Sci. 257 (2011) 5688–5693. [6] S.H. Gao, L.H. Gao, K.S. Zhou, Appl. Surf. Sci. 257 (2011) 4945–4950. [7] S. Bhattacharya, A. Datta, J.M. Berg, S. Gangopadhyay, J. Microelectromech. Syst. 14 (3) (2005). [8] S. Bhattacharya, Y. Gao, V. Korampally, M.T. Othman, S.A. Grant, K. Gangopadhyay, S. Gangopadhyay, Appl. Surf. Sci. 253 (2007) 4220–4225. [9] C. Elif Cansoy, H. Yildirim Erbil, O. Akar, T. Akin, Colloids Surf. A Physicochem. Eng. Asp. 386 (2011) 116–124. [10] W. Huang, L. Jiang, C. Zhou, X. Wang, Tribol. Int. 52 (2012) 87–93. [11] R.A. Lawton, C.R. Price, A.F. Runge, W.J. Doherty III, S. Saavedra, Colloids Surf. A Physicochem. Eng. Asp. 253 (2005) 213–215. [12] Z. Fang, Y. Qiu, E. Kuffel, J. Phys. D. Appl. Phys. 37 (2004) 2261–2266. [13] L.J. Ward, J.P.S. Badyal, A.J. Goodwin, P.J. Merlin, Polymer 46 (2005) 3986–3991. [14] C. Wang, X. He, Appl. Surf. Sci. 252 (2006) 8348–8351. [15] Z. Fang, Y. Qiu, H. Wang, E. Kuffel, Plasma Sci. Technol. 9 (5) (2007). [16] P. Somasundaran, S.C. Mehta, P. Purohit, Adv. Colloid Interf. Sci. 128–130 (2006) 103–109. [17] K.S. Chen, S.C. Liao, S.H. Tsao, N. Inagaki, H.M. Wu, C.Y. Chou, W.Y. Chen, Surf. Coat. Technol. 228 (Supplement 1) (2013) s33–s36. [18] Q. Bénard, M. Foi, M. Grisel, Appl. Surf. Sci. 253 (2007) 4753–4758. [19] M. Żenkiewicz, J. Achiev. Mater. Manuf. Eng. 24 (1) (2007). [20] A. Carre, J. Adhes. Sci. Technol. 21 (2007) 961–981. [21] C.J. Van Oss, R.F. Giese, Clay Clay Miner. 43 (4) (1995) 474–477. [22] S.W. Choi, W.B. Choi, Y.H. Lee, B.K. Ju, J. Korean Phys. Soc. 38 (3) (2001) 207–209. [23] J. Abenojar, M.A. Martínez, N. Encinas, F. Velasco, Int. J. Adhes. Adhes. 44 (2013) 1–8. [24] J.A.S. Ting, L.M.D. Rosario, M.C.C. Lacdan, H.V. Lee Jr., J.C. De Vero, H.J. Ramos, R.B. Tumlos, Int. J. Adhes. Adhes. 40 (2013) 64–69. [25] K. Liu, Y. Tian, L. Jiang, Prog. Mater. Sci. 58 (2013) 503–564.
Acknowledgments The authors would like to thank IBF Electronics GmbH & Co. KG and the Department of Science and Technology (DOST) for the financial support provided in this research.
Please cite this article as: J.A.S. Ting, et al., Surf. Coat. Technol. (2014), http://dx.doi.org/10.1016/j.surfcoat.2014.08.008