Investigation on thermal stability of Ta2O5, TiO2 and Al2O3 coatings for application at high temperature

Investigation on thermal stability of Ta2O5, TiO2 and Al2O3 coatings for application at high temperature

Applied Surface Science 285P (2013) 713–720 Contents lists available at ScienceDirect Applied Surface Science journal homepage: www.elsevier.com/loc...

3MB Sizes 97 Downloads 122 Views

Applied Surface Science 285P (2013) 713–720

Contents lists available at ScienceDirect

Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc

Investigation on thermal stability of Ta2 O5 , TiO2 and Al2 O3 coatings for application at high temperature Peng Shang a,b,∗ , Shengming Xiong a , Linghui Li a , Dong Tian a , Wanjun Ai a a b

Institute of Optics and Electronics, Chinese Academy of Sciences, Chengdu 610209, China Graduate School of the Chinese Academy of Sciences, Beijing 100039, China

a r t i c l e

i n f o

Article history: Received 29 April 2013 Received in revised form 22 August 2013 Accepted 24 August 2013 Available online 31 August 2013 Keywords: Thermal stability TiO2 /Ta2 O5 /Al2 O3 Coating Mismatch

a b s t r a c t In this paper, tantalum pentoxide (Ta2 O5 ), titanium dioxide (TiO2 ) and aluminum oxide (Al2 O3 ) coatings are deposited on silicon substrates by ion beam sputtering (IBS). The influences of the thermal exposure at high temperature in air on the surface morphology, roughness, and the structure were investigated. The results indicate that the chemical composition of the as-deposited TiO2 and Ta2 O5 coatings are apparently close to the stoichiometry ratios and both of them are amorphous structures. The peaks corresponding to anatase TiO2 appear at 400 ◦ C while the anatase-to-rutile transformation is not observed after 800 ◦ C and 1000 ◦ C bake. Ta2 O5 coating crystallizes at 800 ◦ C and 1000 ◦ C to form the hexagonal structure and orthorhombic structure, respectively. TiO2 and Al2 O3 single layers all develop catastrophic damage at 400 ◦ C in the form of noted spallation or blisters, whereas there is no visible damage for the Ta2 O5 coating even at 1000 ◦ C. To understand possible damage mechanisms, the thermal stress distributions through the thickness of Ta2 O5 and TiO2 coatings and the influence of the microstructure transformation are discussed. Finally, some possible approaches to improve the thermal stability are also proposed. © 2013 Elsevier B.V. All rights reserved.

1. 1 Introduction Because of the development of optical coatings which are deployed in more extreme environment, the stability of coating subjected to different thermal loading becomes an issue of crucial importance [1,2]. So thermal robustness must be taken into account when designing the coating requiring long term temperature stability up to 1000 ◦ C [3]. TiO2 and Ta2 O5 are the most commonly used high refractive index materials in the multilayer dielectric mirror due to their high transmittance, low optical loss and low self-emission at high temperature [4,5]. In addition, they also possess excellent chemical and thermal stability, and favorable mechanical properties which are of advantage to be used as protective coatings [5–7]. So more recently, TiO2 and Ta2 O5 coatings have received significant attention and have been applied to a wide range of scientific and technological applications [1,7–13]. Many techniques have been explored to deposit the Ta2 O5 and TiO2 coatings such as chemical vapor deposition (CVD), RF magnetron sputtering, ion beam assisted deposition and the sol–gel method [1,2,14–20]. Some literatures [19,21] reveal that Ta2 O5 and TiO2 coatings deposited by ion-beam sputtering possess good properties such as high refractive index, low absorption, less defects and typically amorphous microstructure. As the

∗ Corresponding author. Tel.: +86 028 85101036; fax: +86 028 85101036. E-mail address: [email protected] (P. Shang). 0169-4332/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.apsusc.2013.08.115

temperature increases, the first phase transition occurs at the temperature between 600 ◦ C and 800 ◦ C for Ta2 O5 and ∼300 ◦ C for TiO2 . And another phase transition can be observed at ∼1000 ◦ C for Ta2 O5 and ∼1100 ◦ C for TiO2 [22–24]. These behaviors have a great influence on the stability of structure and applications. However, the thermal stability of the coatings deposited by IBS is seldom reported although this deposition method can provide the higher density coatings using hard and durable optical materials [3]. In this paper, single TiO2 , Ta2 O5 and Al2 O3 layers are deposited on silicon substrates ( 30 × 3 mm) by ion beam sputtering (IBS). The influences of thermal exposure at high temperature on the coating composition, microstructure, surface morphologies and the root mean square values of roughness (RMS) are characterized systematically. The thermal robustness of Ta2 O5 , TiO2 and Al2 O3 coatings are conducted by experiments. To understand possible damage mechanisms, the thermal stress distributions through the thickness of Ta2 O5 and TiO2 coatings and the influence of the microstructure transformation are discussed. Finally, some possible approaches to improve the thermal stability are also proposed. 2. Experimental 2.1. Fabrication The single TiO2 , Ta2 O5 and Al2 O3 layers were deposited on silicon substrates using Veeco Ion Tech SPECTOR system. They are fabricated in an IBSD-1000 coater, equipped with a 160 mm ion

714

P. Shang et al. / Applied Surface Science 285P (2013) 713–720

Fig. 1. Schematic diagram of the ion beam sputtering deposition system IBSD-1000. Fig. 2. Flow chart of experimental procedure for bake experiments.

source for sputtering and a 120 mm ion source for assistance (see Fig. 1). High purity (99.999%) argon and oxygen are used as bombarded and reactive gas, respectively. The physical thickness is controlled by deposition time based on the stable deposition rate. The thicknesses of the coatings are all 500 nm in design and the error is about ± 2 nm. All the silicon substrates used in our experiments came from the same batch and had been polished at the same time. Before the substrates were placed in the chamber, they were subjected to a series of chemical cleaning. All samples are deposited with optimized process in our experiments. And the detailed process parameters are outlined in Table 1. Heat treatment on samples is performed in a furnace with air environment to the desired temperature (200 ◦ C, 400 ◦ C, 600 ◦ C, 800 ◦ C and 1000 ◦ C) for 1 h. They are then rapidly cooled to room temperature in air and the cooling rate is about 50 ◦ C/s. Fig. 2 presents the flow chart of experimental procedure. All samples are carefully packed in a ceramic vessel for avoiding contamination.

2.2. Characterization The evolution of surface morphology and RMS roughness for Ta2 O5 and TiO2 coatings as a result of thermal exposure to different temperatures are investigated by scanning electron microscope (SEM) (or Normarski microscopy) and optical profilometer made by BRUKER, respectively. X-ray photoelectron spectroscopy (XPS) is used to analyze the composition of the as-deposited coatings. The XPS patterns are recorded using Mg K␣ radiation at 12 KV and 10 mA. The crystalline structure is characterized by X-ray diffraction (XRD) measurement with Cu K␣ as the incident radiation in the  − 2 mode. The crystallite size L of TiO2 coating is determined using the following formula. L = 0.9/(ˇ cos )

(1)

Where , ˇ and  are the wavelength of X-ray (0.1541 nm), the full width of peak at half maximum intensity (FWHM) and the Bragg diffraction angle, respectively. The average residual stress in the coating is composed of thermal stress and intrinsic stress. The thermal stress originates from the different CTE between the coating and substrate. Combing the analytical model of Tsui and Clyne [25] with the well-known Stoney’s equation [26], the thermal stress of the film can be obtained as

Td (˛s − ˛f )dT

Eef f =

T

(2)

1 + 4(Eef /Ees )(h/H)

where Eef = Ef /(1 − vf ), Ees = Es /(1 − vs ), Ef , Es , h, H, ˛s , ˛f , vs , vf , Td and T are effective Young’s modulus of the film, effective Young’s modulus of the substrate, Young’s modulus of the film, Young’s modulus of the substrate, film thickness, substrate thickness, CTE of the substrate, CTE of the film, Poisson’s ratio of the substrate, Poisson’s ratio of the film, deposition temperature and the bake temperature, respectively. The physical and mechanical properties of the Ta2 O5 , TiO2 and silicon materials are given in Table 2. The values are taken from references [33–35]. It is worth noting that the CTEs of TiO2 and Ta2 O5 films in this study are taken as an average value over the temperature range between room temperature and high temperature. To analyze the thermal stress in the coating, it is assumed that all the materials are isotropic and linear thermoelastic, have perfect bonding between coating and substrate and plain biaxial stress. Thermal loading is applied by setting the reference temperature as the bake temperature (400 ◦ C) and uniform temperature as the room temperature (20 ◦ C). The thermal stress distributions through the thickness of Ta2 O5 and TiO2 coating are evaluated by COMSOL Multiphysics system, respectively.

Table 1 Deposition conditions of single layers by IBSD. Thin film

Target material

Chamber pressure (torr)

˚ Deposition rate (A/s)

Flow rate of Ar(O2 ) (sccm)

TiO2

Titanium dioxide

4.0 × 10−6

0.08



−6

Ta2 O5

Tantalum

4.0 × 10

0.15

8(25)

Al2 O3

Aluminum

4.0 × 10−6

0.14

8(25)

P. Shang et al. / Applied Surface Science 285P (2013) 713–720

715

Table 2 The CTE values for the coating materials and substrate material used in this study. Serial number

Properties

1 2 3

Poisson’s ratio E (GPa) CTE (10−6 /K)

Materials Si

TiO2

Ta2 O5

0.28 131 4.0147 (700 K) 4.342 (1100 K) 4.416 (1300 K)

0.27 230 8.5

0.27 136 4.4

The intrinsic stress is commonly thought to be due to phase transformation, various defects, lattice mismatch between the coatings and substrates, chemical reactions, etc. And the crystalline phase stress of TiO2 can be determined using the formula 2d sin  = 

(3)

Where d is interplanar distance value of the coating.  = E(d0 − d)/(2d0 )

(4)

Where d0 and d are interplanar distance values corresponding to various crystal planes of the standard and the measured XRD spectra, respectively E and v are Young’s modulus and Poisson’s ratio of the coating, respectively. 3. Result 3.1. Damage morphology characterization Figs. 3–5 show a sequence of photos for TiO2 , Ta2 O5 and Al2 O3 coating at different temperatures, respectively. We continue to bake the samples after they have been damaged, as we are interested in the evolution of the damage morphology with the increasing temperature. It can be obviously seen (Fig. 3b) that there are some micro-cracks on the surface of TiO2 coating at 400 ◦ C. After thermal exposure to 1000 ◦ C, the number of spalled pieces increases and the area of spallation enlarge (Fig. 3c). For comparing with TiO2 coating, we also study the surface morphology evolution for Ta2 O5 and Al2 O3 coatings at different temperature. As shown in Fig. 4b, we can know that, ∼15 micron diameter blisters appear in the Al2 O3 coating after ∼400 ◦ C bake. Additional the diameter of blister enlarges (about 50 ␮) and “rough patches” appear on the remaining coating plateaus after thermal exposure to 1000 ◦ C (Fig. 4c). Fig. 5a presents that the surface topology for the as-deposited Ta2 O5 coating looks smooth without any apparent feature such as pinhole-type defects. There are neither visible cracks nor blisters even at 400 ◦ C and 1000 ◦ C, despite the severe failure for TiO2 and Al2 O3 coatings. But it should be noted that Ta2 O5 coating did show notable crystallization and roughness at 1000 ◦ C. And the crystallites can be seen in the Fig. 5c with typical grain sizes between 200 nm and 1000 nm. So Ta2 O5 coating shows the significantly higher temperature stability compared to TiO2 and Al2 O3 . 3.2. XPS characterization The spectra of TiO2 and Ta2 O5 coatings show that the surfaces mainly contain C, Ti, O elements and C, Ta, O elements, respectively. And the C elements are both from the measurement equipment. The high resolution XPS spectra of the as-deposited TiO2 coating near Ti 2p (a) and O 1s (b) peaks are presented in Fig. 6. The O 1s peak at 529.9 eV is attributed to O (O Ti O). Because of the selforbital coupling effect, the binding energy of the Ti (4+) 2p3/2 has two peaks at 458.4 eV and 464.1 eV. Fig. 7 illustrates the Ta 4f and O

Fig. 3. Surface morphologies of TiO2 coating at different temperatures, (a) before bake; (b) after 400 ◦ C bake; c) After 1000 ◦ C bake.

716

P. Shang et al. / Applied Surface Science 285P (2013) 713–720

Fig. 4. Surface morphologies of Al2 O3 coating at different temperature, (a) before bake; (b) After 400 ◦ C bake; (c) after 1000 ◦ C bake; (d) the image magnified about four times.

1s photoelectron spectra of Ta2 O5 coating. The Ta 4f signal (Fig. 7a) is fitted with two peaks located at 26.7 eV and 28.6 eV which represents for Ta 4f7/2 to Ta 4f5/2 spin orbital splitting. Deduced from the results of XPS spectra, the measured surface O/Ti, O/Ta ratios are 2.1:1 and 2.49:1, respectively. The first value, which is greater than 2, may due to the total measured oxygen such as TiO2 matrix, SiO2 native layer and humidity occupying the coating pores. And the last value approaches the standard stoichiometric ratio. 3.3. XRD characterization Figs. 8 and 9 present the XRD spectra of single layer TiO2 and Ta2 O5 coating at different temperature. The TiO2 coating generally possesses three crystal polymorphs in nature: amorphous, anatase and rutile. As the Fig. 8 shows, the as-deposited TiO2 coating is amorphous structure. With the temperature increasing, the peaks (such as (1 0 1), (0 0 4), (2 0 0)) corresponding to anatase phase appear at 400 ◦ C. However, the diffraction peaks of rutile phase do not appear after 1000 ◦ C which is not consistent with others reports [27–29] where there is a transformation of anatase–rutile

at temperature between 700 ◦ C and 900 ◦ C and TiO2 coating only exhibits single rutile phase at the temperature of over 900 ◦ C. This phenomenon may be a result of the difference of cooling rates. As the anatase-to-rutile transformation is known to occur slowly, anatase TiO2 is favored to be formed under high cooling rate while rutile is favored to be formed under near- equilibrium solidification conditions [30,31]. Fig. 9 displays that the as-deposited Ta2 O5 coating is amorphous and after thermal exposure to 800 ◦ C, the structure of Ta2 O5 coating transforms from amorphous to hexagonal (␦–Ta2 O5 ) phase. The orthorhombic (L–Ta2 O5 ) phase of Ta2 O5 appears when the bake temperature increases to 1000 ◦ C. Table 3 shows the interplanar distance, crystallite size and stress  corresponding to different crystalline planes of anatase TiO2 at 400 ◦ C, 800 ◦ C and 1000 ◦ C. The intensity of anatase peaks such as (2 0 0), (1 0 1) and (0 0 4) increases at 400 ◦ C and 800 ◦ C indicating the growth of average grain-size. But after 1000 ◦ C bake, the crystalline sizes of (2 0 0) and (1 0 1) crystal planes decrease. In addition, as the temperature increases, some diffraction peaks of anatase TiO2 (such as (2 0 0), (0 0 4) and (1 0 1)) move to the right and the lattice space decreases. So the tensile stress

P. Shang et al. / Applied Surface Science 285P (2013) 713–720

717

presents at 800 ◦ C and 1000 ◦ C while it is compressive stress at 400 ◦ C. 3.4. RMS characterization The evolution of surface roughness at different temperature is shown in Fig. 10. The RMSs of TiO2 are 2.4903 nm, 1.9487 nm and 2.9064 nm at the temperature of 400 ◦ C, 800 ◦ C and 1000 ◦ C, respectively. It goes through a local minimum value at about 800 ◦ C. And the RMS increase at 400 ◦ C and 800 ◦ C can be attributed to the phase change and crystal growth during the phase transformation course. The surface roughness of Ta2 O5 coating decreases after 400 ◦ C due to the increasing of the diffusion of particles. But when the temperature increases to 800 ◦ C and 1000 ◦ C, Ta2 O5 coating is transformed from amorphous to crystalline and the roughness increases significantly. 4. Discussion

Fig. 5. Surface morphologies of Ta2 O5 coating at different temperatures, (a) before bake; (b) after 400 ◦ C bake; (c) After 1000 ◦ C bake.

According to the experiment results above, the different thermal robustness of TiO2 and Ta2 O5 coatings is closely related to the phase constitution, crystal structure, the internal stress and mechanical properties. Firstly, As the Fig. 8 shows the transformation from amorphous to anatase occurs at 400 ◦ C. Because the anatase TiO2 is the denser phase at 3.84 g cm−3 as compared to the amorphous phase at 2.30 g cm−3 , the phase transition is accompanied by a significant change in the structure size, which may lead to swell, lattice mismatch and localized fracture in TiO2 coating. In addition, as seen in Table 3, the lattice mismatching results in higher compress intrinsic stress in the TiO2 coating at 400 ◦ C which can lead to the damage of the coating. As the temperature increases from 400 ◦ C to 800 ◦ C, the crystallite grain grows up and the coating is contracted, the intrinsic stress is translated from compress to tensile stress. At the same time, heat treatment can also increase the density of coating and drive away the water in the hole which is also in favor of the increasing of tensile stress further so that more cracks are induced in the coating. Last but not least, the different mechanical properties between the coatings and substrate which can induce the higher stress at high temperature are also very important factors to influence the reliability of the coating for its various function applications [3,32]. The thermal stress distributions through the thickness of Ta2 O5 and TiO2 coating-substrate systems at high temperatures are evaluated and compared by finite element method (FEM). (Fig. 11). From Fig. 12a, we can know that the maximum tensile shear stress, denoted by  xz , is observed at the interface of the coating edge and the stress reversal from tensile at the interface to compressive can be also observed. Although the maximum shear stress  xz shows a decreasing trend at the distances of −4 h, −6 h and −8 h away from the edge, the largest tensile shear stress at the interface of the coating edge can contribute either shear or mixed modes of failure such as spallation. The general distribution of the shear stress in TiO2 and Ta2 O5 coating is similar, but Ta2 O5 coating has less stress value as

Table 3 The calculated crystallite size and stress in TiO2 films at 400 ◦ C, 800 ◦ C and 1000 ◦ C. temperature ◦

400 C

800 ◦ C

1000 ◦ C

Crystal plane 004 101 200 004 101 200 004 101 200

Location,◦

Lattice space A˚

Crystallite size, nm

37.717 25.200 47.970 37.922 25.358 48.127 37.905 25.334 48.102

2.3831 3.5320 0.1895 2.3706 3.5104 0.1890 2.3720 3.5136 1.890

39.731 51.286 38.259 47.200 60.1078 40.013 47.450 56.7588 36.986

Average grain-size, nm 43.092

49.107

47.065

Stress, GPa −1.02646 −1.46106 −0.77422 1.197379 1.158686 0.533499 1.013896 0.768815 0.324647

718

P. Shang et al. / Applied Surface Science 285P (2013) 713–720

458.4ev

a

b

529.9ev

464.1ev

456

458

460

462

464

466

528

Binding energy (eV)

532

536

540

Binding energy (eV)

Fig. 6. XPS spectra of TiO2 film after deposition, (a) Ti 2p and (b) O 1s.

Fig. 7. XPS spectra of Ta2 O5 film after deposition, (a) Ta 2p and (b) O 1s.

(002)

(1112)

(002)

(110)

(201) (101)

(1110)

(200)

(100)

(001) (001)

Intensity (counts)

(211)

(211)

(105) (105)

(200)

(103) (004) (112)

673.15 K

axial stress ( z ) is noticed at the interface of the coating edge. As the distance away from the coating edge increases, stress reversal from compressive to tensile near the edge of the interface is also observed. From the discussion above, it can be seen that there are remarkable stress concentrations and large compressive stresses near the edge of the interface for all specimens that may cause the spallation and the buckling of coating. This area is a very critical region for cracks and flaking. In addition, TiO2 coating exhibits high thermal stress such as radial stress  x , axial stress  y and shear stress  xz , as compared to Ta2 O5 coating at high temperature.

(211)

(105) (200)

(103) (004) (112)

(101) 1073.15 K

(101)

Intensity

counts

1273.15 K

(200)

(103) (004) (112)

(101)

compared to TiO2 coating at the same distance from the system edge. The radial stress ( x ) distributions through the thickness of Ta2 O5 and TiO2 coatings are shown in Fig. 13a and b, respectively. It can be found that the tensile stress decreases abruptly and changes to compressive stress at the interface between the coating and substrate. As the distance increases from the edge to the center, such as −2 h, −4 h and 6 h, the average radial stress value through the coating and substrate increases. The large radial stresses in the TiO2 coating may cause the formation of surface cracks. Similar radial stress distribution was found, but there is an eight times decrease for Ta2 O5 coating when compared to TiO2 . From Fig. 14 we can also observe that through the thickness of TiO2 coating, the largest

24

27

1073.15 K

673.15 K

As-deposited

As-deposited 21

1273.15 K

30

33

36

39 42 2θ (º )

45

48

51

Fig. 8. XRD spectra of TiO2 film at different temperatures.

54

57

20

25

30

35

40 45 O 2θ ( )

50

55

60

Fig. 9. XRD spectra of Ta2 O5 film at different temperatures.

65

P. Shang et al. / Applied Surface Science 285P (2013) 713–720

3.5

1600

TiO2

3.0

Analycial(Si//Ta2O5)

1400

Ta2O5 2.5

Comsol Multiphysic system(Si//Ta2O5) Analycial(Si//TiO2)

1200

2.0 1.5 1.0 0.5 0.0 0

200

400

600

800

1000

Thermal Stress /MPa

RMS roughness (nm)

719

Comsol Multiphysic system (Si//TiO2)

1000 800 600 400 200

Baked dwell temperature (ºC) 0 Fig. 10. Surface roughness evolution for TiO2 , Ta2 O5 coatings.

-200 200

400

600

800

1000

1200

Temperature /K Fig. 11. Thermal stress variation as a function of temperature.

Fig. 12. Shear stress ( xz ) distribution through the thickness of TiO2 (a) and Ta2 O5 (b) coatings.

Fig. 13. Radial stress ( x ) distribution through the thickness of TiO2 (a) and Ta2 O5 (b) coatings.

1400

720

P. Shang et al. / Applied Surface Science 285P (2013) 713–720

with the increasing temperatures. So Ta2 O5 coating shows the significantly higher temperature stability compared to TiO2 and Al2 O3 coatings. Their different thermal robustness is closely related to the crystal structure, the internal stress, phase constitution and mechanical properties. So in order to get high-temperature stable optical coating, it is necessary to select the high temperature stable material (such as Ta2 O5 ) or phase (such as rutile TiO2 ). In addition, it is also an effective way to improve the structure stability and minimize the CTE mismatch of substrate-coatings system by doping or intermediate composite layer. Acknowledgments The authors would like to thank Junmu Zhu of Si Chuan University for both the use of XRD and his insight and tireless work. References

Fig. 14. Axis stress ( z ) distribution through the thickness of TiO2 (a) and Ta2 O5 (b) coatings.

5. Conclusion TiO2 , Al2 O3 and Ta2 O5 coatings are deposited by ion-beam sputtering deposition. The XPS and XRD results indicate that chemical composition of the as-deposited TiO2 and Ta2 O5 coatings are close to the stoichiometry ratios and both of them have amorphous structures. After thermal exposure to 400 ◦ C, the structure of TiO2 coating transforms from amorphous to anatase phase. And the average grain-size of anatase phase has the maximum value at 800 ◦ C. Ta2 O5 coating crystallizes at 800 ◦ C and 1000 ◦ C to form the hexagonal (␦–Ta2 O5 ) structure and orthorhombic (L–Ta2 O5 ) structure, respectively. The RMS of TiO2 coating goes through a minimum at 800 ◦ C while the roughness of Ta2 O5 coating decreases after 400 ◦ C bake and then increases with the temperature. In addition, we note that TiO2 and Al2 O3 single layers all develop catastrophic damage at 400 ◦ C in the form of noted spallation or blisters, whereas there is no visible damage for the Ta2 O5 coating even at 1000 ◦ C. And the damage morphology continues to evolve

[1] J.K. Yao, H.L. Huang, C. Xu, J.Y. Ma, H.B. He, J.D. Shao, Y.X. Jin, Y.A. Zhao, Z.X. Fan, F. Zhang, Z.Y. Wu, Surf. Eng. 25 (2009) 116–119. [2] T.P. Mollart, K.L. Lewis, Diam. Relat. Mater. 10 (2001) 536. [3] C. Dale, Ness, Nick Traggis, Proc. SPIE 7504 (2009) 750406–750407. [4] J.K. Yao, Z.X. Fan, Y.X. Jin, Y.A. Zhao, H.B. He, J.D. Zhao, Thin Solid Films 516 (2008) 1237–1241. [5] Wen-Jen Liu, Xing-Jian Guo, Chia-Hung Chien, Surf. Coat. Technol. 196 (2005) 69–75. [6] E. Atanassova, M. Kalitzova, G. Zollo, A. Paskaleva, A. Peeva, M. Georgieva, G. Vitali, Thin Solid Films 423 (2003) 191–199. [7] H. Gruger, Ch. Kunath, E. Kunath, S. Sorge, W. Pufe, T. Pechstein, Thin Solid Films 447–448 (2004) 509–515. [8] Zhinong Yu, Yu qiong Li, Fang Xia, Wei Xue, Surf. Coat. Technol. 204 (2009) 131–134. [9] B. Hunsche, M. Vergohl, A. Ritz, Thin Solid Films 502 (2006) 188–192. [10] S. Ezhilvalavan, T.Y. Tseng, J. Mater. Sci. 10 (1999) 9. [11] S. Boughaba, M.U. Islam, G.I. Sproule, M.J. Graham, Surf. Coat. Technol. 120–121 (1999) 757. [12] C. Chaneliere, J.L. Autran, R.A.B. Devine, B. Balland, Mater. Sci. Eng. R R22 (1998) 269–322. [13] H. Shinriki, T. Kisu, Sh.-I. Kimura, Y. Nishioka, Y. Kawamoto, K. Mukai, IEEE Trans. Electron Dev. 37 (1990) 1939–1947. [14] K. Jadeesh Kumar, N. Ravi Chandra, A. Subrahmanyam, Surf. Coat. Technol. 205 (2011) S261–S264. [15] A. Porpoati, S. Roitti, O. Sbaizero, J. Eur. Ceram. Soc. 23 (2003) 247–251. [16] W.J. Liu, X.J. Guo, C.H. Chen, Surf. Coat. Technol. 196 (2005) 69–75. [17] S. Yidirim, K. Ulutas, D. Deger, E.O. Zayim, I. Turhan, Vacuum 77 (2005) 329–335. [18] J.B. Wang, P. Yang, H.B. Su, G.C. Li, N. Huang, Surf. Coat. Technol. 206 (2011) 1024–1028. [19] Reese Puckett, Surf. Coat. Technol. 41 (1990) 259–267. [20] Takuji Maekawa, Ken Kurosaki, Takanori Tanaka, Shinsake Yamanaka, Surf. Coat. Technol. 202 (2008) 3067–3071. [21] Dale Ness, Darrel Pitrat, Chris Wood, Proc. SPIE 7660 (2010) 766025. [22] J.K. Yao, J.D. Shao, H.B. He, Z.X. Fan, Vacuum 81 (2007) 1023–1028. [23] T. Dimitrova, K. Arshak, E. Atanassova, Thin Solid Films 381 (2001) 31–38. [24] Cheng Xu, Qiling Xiao, Jianyong Ma, Yunxia Jin, Jianda Shao, Zhengxiu Fan, Appl. Surf. Sci. 254 (2008) 6554–6559. [25] Y.C. Tsui, T.W. Clyne, Thin Solid Films 306 (1997) 23–33. [26] G.G. Stoney, Proc. R. Soc. London Ser A82 (1909) 172–175. [27] M. Radecka, K. Zajrzewska, H. Czternastek, T. Stapinski, Appl. Surf. Sci. 65/66 (1993) 227. [28] Y.Q. Hou, D.M. Zhuang, G. Zhang, M. Zhao, M.S. Wu, Appl. Surf. Sci. 218 (2003) 97. [29] L.J. Meng, M. Andridchky, M.P. Dos Santos, Appl. Surf. Sci. 65 (66) (1993) 235. [30] Y. Li, T. Ishigaki, J. Cryst, Growth 242 (2002) 511. [31] M. Bozorgtabar, M. Iafarpour, Surf. Coat. Technol. 205 (2011) S229–S231. [32] V. Chawla, R. Jayaganthan, R. Chandra, Mater. Chem. Phys. 114 (2009) 290–294. [33] Z. Xu, D. Yan, D. Xiao, P. Yu, J. Zhu, Ceram. Int. 38 (2012) 981–985. [34] E. Cetinorgu, B. Baloukas, O. Zabeida, E. Jolanta, K. Sapieha, L. Martinu, Appl. Opt. 50 (2011) 3351–3359. [35] E.P. MEAGHER, Can. Mineral. 17 (1979) 77–85.