Effect of annealing temperature on PL spectrum and surface morphology of zinc oxide thin films

Applied Surface Science 270 (2013) 163–168

Contents lists available at SciVerse ScienceDirect

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

Effect of annealing temperature on PL spectrum and surface morphology of zinc oxide thin films A. Zendehnam a , M. Mirzaee c , S. Miri b,∗ a b c

Thin Film Laboratory, Physics Department, Science Faculty, Arak University, Arak 38156-8-8349, Iran Department of Physics, Gachsaran Branch, Islamic Azad University, Gachsaran, Iran Department of Physics, Faculty of science, Arak University, Arak 38156-8-8349, Iran

a r t i c l e

i n f o

Article history: Received 11 September 2012 Received in revised form 17 September 2012 Accepted 28 December 2012 Available online 5 January 2013 Keywords: ZnO Fractal analysis Morphology Annealing temperature PL AFM

a b s t r a c t Zinc oxide (ZnO) thin films were produced by thermal oxidation of Zn layers (200 nm thickness) which were coated on Si (1 0 0) substrate by DC magnetron sputtering. In order to study the effect of annealing temperature on photoluminescence (PL) properties and the surface morphology of the ZnO samples, the annealing temperature range of 500–700 ◦ C was employed. Scanning electron microscopy (SEM) and atomic force microscopy (AFM) for investigation of surface morphology of the ZnO samples were carried out. The surface statistical characteristics of these ZnO thin films are then evaluated against data which outcome from AFM. SEM and AFM results indicated that the annealing temperature produces larger grains and rough surfaces at higher temperatures. The results of PL spectra represent an increase in interstitial zinc with increasing annealing temperature. © 2013 Elsevier B.V. All rights reserved.

1. Introduction Transparent conducting oxides (TCO) have been recently investigated for their interesting optical, mechanical and electrical performance [1,2]. The lasting recorded high optical transparence in the visible domain, and low electrical resistivity led to numerous applications of these materials in the new generation of optoelectronic devices [3–5]. Zinc oxide (ZnO) represents, in this context, an important basic material for the construction of nanoscale structures. Recently, zinc oxide has attracted much attention within the scientific community as a ‘future material’. It has a direct band gap of 3.37 eV at room temperature and a high exciton binding energy of 60 meV [5], which is much larger than the thermal energy at room temperature and belongs to a member of hexagonal wurtzite class. Some of the optoelectronic applications of ZnO overlap with GaN (Eg = 3.4 eV at 300 K), which is widely used for the production of green, blue, ultraviolet and white light-emitting devices [6–9]. However, ZnO has some advantages over GaN, among which are the availability of fairly high-quality ZnO bulk single crystals and a large exciton binding energy and the ability to grown single crystal substrate. These interesting properties of ZnO thin films have been for use in many applications such as room temperature UV

∗ Corresponding author. Tel.: +98 861 2777400; fax: +98 861 2774031. E-mail address: sadegh [email protected] (S. Miri). 0169-4332/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.apsusc.2012.12.154

laser and short-wavelength optoelectronic device [10]. Furthermore, ZnO can be use in the photo-conductors [11], gas sensors [12], and solar cell applications [13]. The highly preferential orientation a long c-axis and wide band gap energy of films are useful characteristic in optical wave-guides [14], surface acoustic wave [15] and acousto-optic device [16]. The surface, which is the first interface of the material, has an important role in the interaction between the material and the environment [17]. Surface roughness has an enormous influence on many important physical phenomena such as mechanical contact, sealing, adhesion, wave scattering and friction [18]. Most surfaces in nature are rough, and this fact is a motivation that can be studied as a random process. In fact, a surface for a special application requires specified statistical properties. ZnO thin films have been synthesized by a variety of processes, including physical vapor deposition (PVD), chemical vapor deposition (CVD), pulsed laser deposition (PLD), molecular beam epitaxy (MBE), sputtering, sol–gel processing, etc. [19–24]. In this work we used DC magnetron sputtering for deposition of pure zinc, and method of metallic Zn thermal oxidation was employed to produce ZnO thin films. High packing density, uniform film thickness, suitable coating rate, very good adhesion of the film to substrate due to high energy of deposition particles are some of magnetron sputtering advantages. In this work the effect of the annealing temperature on the photoluminescence (PL) spectrum and the surface morphology of the ZnO layers deposited on the Si (1 0 0) substrate were investigated. To investigate surface morphology of these ZnO thin

164

A. Zendehnam et al. / Applied Surface Science 270 (2013) 163–168

films, atomic force microscopy (AFM) (Park Scientific Instrument Auto Probe model CP) was employed using force constant mode. In this paper, we study the annealing process as a stochastic process. Surface roughness, roughness exponents, correlation length have been measured.

The height–height correlation function can be expressed as [25]:





H(md)

N

1 N−m 2 (h(i + m, j) − h(i, j)) N−m i=1

=

m = 0, 1, 2, 3 . . . N − 1

(3)

2. Experimental details Sputtering of the layers of Zn at room temperature (RT) on the silicon substrate were performed in Ar ambient (purity 99.999%) and a 99.9% pure metal zinc target with 12.5 cm diameter and 3 mm thickness was used for sputtering. A vacuum system (Hind High Vacuum, H.H.V, 12 MSPT) with base pressure of 10−6 mbar was used. Si (1 0 0), n type wafer was cut to required dimensions (1 cm × 1 cm) and 1 mm thickness and were used as substrate for deposition of Zn thin films. Before deposition, silicon substrates were cleaned in the heated acetone ultrasonic bath for 2–3 min. A shutter was placed between the zinc target and silicon substrate for two reasons, firstly to control period of coating, secondly the shutter was removed only when the line spectra belonging to Zn, Ar (atoms, ions) was checked, and no band spectra due to contamination was detected. After many runs of discharge and sputtering, the optimum distance of 12 cm between Zn target and silicon substrate was found and this distance during this work was kept constant. The thickness of these samples was about 200 nm which were measured using a digital vibrating quartz crystal thickness monitor. Layers of zinc after thermal oxidation in air ambient at 400 ◦ C were employed to produce ZnO thin films. In order to study the effect of annealing temperature, the samples were heated in a quartz tube furnace at various temperatures. The samples were held at each temperature for 10 min and then furnace was cooled to room temperature. The rate of the cooling was slow enough to avoid the possibility of any type of stress and strain in the thin films. PL spectra were taken at room temperature under 340 nm xenon lamps (with Fabry Perot filter) as the excitation source (Stellnet model Epp-200). To investigate the surface morphology of these ZnO thin films scanning electron microscopy (SEM) (Hitachi S4160), and AFM (Park Scientific Instrument Auto Probe model CP) were also performed using the force constant mode and digitized into 256 × 256 pixels. A commercial standard pyramidal Si3N4 tip was used. All AFM images were acquired in ambient air. For a better comparison of the effects of different interfaces, we kept all other experimental parameters unchanged. To analyses the AFM images, the topographic image data were converted into ASCII data. AFM images of samples indicated changes in surface behavior of the film. 3. Statistical analysis A random rough surface can be described mathematically as h(i, j), where h is the height of the surface relative to the reference level, that reference level defined by a mean surface height and r(i, j) is the position on the surface. We assume that the distance between two adjacent discrete positions is d, and the number of surface points is N. The average surface height is the arithmetic average of surface heights. Analytically it can be expressed, for a digitized surface, as:

  h

N

=

1 N h(i, j) N i=1

(1)

Root mean square (RMS) roughness is one of the most important parameters for characterizing a rough surface. Analytically, it can be estimated as: wN =

  N  1 N

i=1

  2

h(i, j)) − h

N

1/2 (2)

For self-affine surfaces, the dynamic scaling hypothesis suggests that the height–height correlation function H(r) has the scaling form of:



H(r) =

(r)2˛

for

r  l0

2w2

for

r  l0

(4)

where  = w1/˛ /l0 is the local slope, and ˛ is called the roughness exponent (0 < ˛ < 1) which describes how wiggly the surface is. The quantity l0 is the lateral correlation length, within which the surface heights of any two points are correlated. The roughness exponent ˛ is related to the fractal dimension Df of the random surface by Df = E + 1 − ˛ with 0 < ˛ < 1, where E + 1 is the dimension of the embedded space (E = 1 for a profile; E = 2 for a plan). A larger value of ˛ corresponds to a locally smooth surface structure while a smaller value of ˛ corresponds to a more jagged local surface morphology [25,26]. The height distribution function provides a complete specification of the random variable h(r) at the position r. Although different rough surfaces may have different height distributions, the most generally used height distribution is the Gaussian height distribution. The statistical analysis of AFM data was done using the height distribution histograms. The height asymmetry is described by the statistical parameters such as surface skewness and kurtosis. Unlike RMS roughness, skewness is dimensional. Skewness is a measure of the symmetry of a distribution about a mean surface level. Kurtosis is also a dimensionless quantity. It is a measure of the sharpness of the height distribution function. These two parameters represent the shape of the surface height distribution and can be estimated as [25]: Rsk = Rku =

N

1 Nw3N 1

 4

N h

i,j=1

 

(h(i, j) − h

N i,j=1

N

)

 

(h(i, j) − h

N

3

)

4

(5)

(6)

N

For comparison of these parameters for all samples, we used central limit theorem. This theorem mean of a sufficiently large number of independent random variables, each with finite mean and variance, will approximately have a normal distribution [27]. 4. Result and discussion 4.1. Photoluminescence properties In order to study the effects of annealing temperature on PL spectra and the surface morphology of the ZnO thin films, three samples were deposited under similar conditions (Ar pressure, substrate temperature, oxidation temperature, period of heating and deposition rate). All of them were heated at 400 ◦ C for oxidation of zinc film then they were annealed at different temperatures (500, 600 and 700 ◦ C), and they are named as b1 , b2 and b3 respectively. It is well known that the defects in ZnO thin films which may occur include oxygen vacancy (VO ), zinc vacancy (VZn ), interstitial zinc (Zni ), interstitial oxygen (Oi ) and anti-site oxygen (OZn ) depend upon the method of coating and important parameters such as temperature of annealing, deposition rate, atmosphere of heating, etc.

A. Zendehnam et al. / Applied Surface Science 270 (2013) 163–168

165

Fig. 1. Energy levels of the defects in ZnO thin films.

Fig. 1 presents the energy levels of the usual defects in ZnO thin films [28]. Fig. 2 shows the room temperature PL spectra in the wavelength range of near-band-edge emission (NBE), of the ZnO thin films grown on Si substrate and oxide at 400 ◦ C and then annealed at 500, 600 and 700 ◦ C temperatures. A near-band-edge emission (NBE) appears around 380 nm in the UV spectral region. The NBE peak can be attributed to free-exciton recombination. The intensity of NBE emissions remains constant, which probability means the crystal quality of the ZnO films at these temperature ranges were constant. The NBE peak shifted from small wavelength (3.27 eV) for sample which annealed at 500 ◦ C to high wavelength (3.21 eV) for the sample annealed at 600 ◦ C temperature. Energy gap shifts toward higher wavelengths, probability is due to the departure of oxygen from layer and increase the conductivity of the layer at high annealing temperatures. At low temperature of annealing (500 ◦ C) there are other peaks present which are related to deep-level emission (DLE) and the usual ZnO defects such as oxygen and zinc vacancies. Fig. 3 shows PL spectra of the ZnO samples annealed at 500, 600 and 700 ◦ C temperatures. The DLE peak position was relatively constant for all samples. The peak at 435 nm also observed in all of the samples, intensity of this peak remained constant at 500 and 600 ◦ C and increased at 700 ◦ C temperature. This peak is concerned to interstitial zinc which increased with the annealing temperature. The intensity of another peaks (DLE) was constant for all of the samples. There were some reports that the defect of ZnO thin films improved by annealing at high temperatures (400–700 ◦ C) [29].

Fig. 2. PL spectra of ZnO thin films that oxide at 400 ◦ C and annealed at 500, 600 and 700 ◦ C temperature in 360–420 nm ranges.

Fig. 3. Effect of annealing temperature on PL spectrum in the range of 400–600 nm wavelengths.

However at higher annealing temperature ZnO films grown on Si substrates usually show degradation of luminescence property [30,31], although the melting point of ZnO (1975 ◦ C) is much higher than the annealing temperature. For this reason in this work we did not annealed the samples at very high temperatures. When one takes the effect of substrate into consideration, this discrepancy can be understood. During annealing, not only the recrystallization but also the interdiffusion between ZnO film and Si substrate will be accelerated. Therefore the improvement of crystal quality is only observed within a restricted temperature region. At very high annealing temperatures a thin SiO2 interfacial layer between ZnO and Si substrate can be formed which may change during and after annealing and this may cause defect of the produced samples. 4.2. Surface morphology Fig. 4 presents the SEM micrographs of the surface topography of ZnO thin films with large and densely packed grains grown on Si substrate and annealed at different temperatures. For sample b1 , the image shows a flat and uniform surface and for b2 and b3 samples which annealed at high temperatures leads to increase of the particle size on surfaces, the average particle size were 95, 128 and 169 nm respectively. These results are consistent with previous work suggesting that a high annealing temperature enhances the interaction among the seed particles and causes the formation of bigger ZnO seeds [32]. This fact indicates that the size of the seed particles is one of the key factors that influence the formation of ZnO thin films with a large grain size, and grain growth continues with rise of temperature. Formation of oxide whiskers could be observed widely on many pure metals during oxidation at a moderately high temperature and low oxygen partial pressure especially with incorporation of water vapour [33]. In this study, ZnO nano whiskers with a length of 1–2 ␮m long were formed on a dense oxide base film by oxidation in air (relative humidity was in the range of 50–60%). Since the temperatures of 500 and 600 ◦ C is close to the melting point temperature of the Zn (420 ◦ C). These could promote the formation and growth of whisker oxide, hence these temperatures are appropriate for formation the whiskers in ZnO nano layers [34]. In some other reports [35,36], these whiskers on the ZnO surface have been observed when low temperature of annealing (350–550 ◦ C) is used, specially employing thermal oxidation method produced more nanowires and whiskers. Three-dimensional AFM images of these samples are shown in Fig. 5(a). The scanning area is 5 ␮m × 5 ␮m. It is clear that the surface of the film is covered with dense and uniform nanocone ZnO

166

A. Zendehnam et al. / Applied Surface Science 270 (2013) 163–168

Fig. 4. SEM images of ZnO thin films annealed at 500, 600 and 700 ◦ C temperatures for b1 , b2 and b3 samples respectively.

grains. These ZnO nanocones are all grown perpendicular to the substrate surface and the formation of these nanocones structure mainly results from annealing treatment. The AFM images indicate that, by altering the annealing temperature, the grain size and the roughness change. Therefore, it seems that by choosing a suitable annealing temperature, the grain size and the roughness may be controlled. Two-dimensional images of specimens that were obtained by using software Suffer, shows the increase in particle size with increasing annealing temperature (Fig. 5(b)) Given the statistical parameters introduced in previous section, it is possible to obtain some quantitative information about the effect of annealing temperature on surface topography of ZnO thin films. For obtaining the mean height, RMS roughness, and height-height correlation function of samples, Eqs. (1)–(3) are used respectively. The height–height correlation function H(r) is calculated along the x direction. Fig. 6 shows the height–height

correlation function as a function of position l for sample b1 . The slope of each curve at small scales yields the roughness exponent (2˛) of the corresponding surface. The saturation limit in this curve is the correlation length which is the lateral size of the grains. For b1 sample which annealed at 500 ◦ C, the RMS roughness is low (59 nm) and the correlation length is maximum value in comparison with the other samples (185 nm). When the annealing temperature is increased to 700 ◦ C, the value of the RMS roughness increased (to 171 nm) and the correlation length decrease to 146 nm. Fig. 7 shows the height–height correlation function for all samples. The values of the statistical parameters summarized in Table 1 for b1 , b2 and b3 samples. For many device fabrications such as gas and light sensors, an increase of the active surface is required. The active surface

Fig. 5. Two and three-dimensional AFM images of b1 , b2 and b3 samples.

A. Zendehnam et al. / Applied Surface Science 270 (2013) 163–168

Fig. 6. Log–log plot of the height–height correlation function versus l for sample annealed at 500 ◦ C temperature.

167

Fig. 8. Height distribution histograms from AFM analysis for all samples.

high kurtosis (larger than 3.0) values are favorable for tribological applications (e.g. low friction bearings) [37]. ZnO thin films are potential materials suitable for tribological applications Table 2 shows the values of these parameters for the produced samples. All of the samples seem appropriate in these terrains. 5. Conclusion

Fig. 7. Height–height correlation functions H(r) versus position l for all of the samples.

Table 1 Statistical parameters for all samples. Df

l0 (nm)

hN (nm)

wN (nm)

Samples

2.37 2.22 2.53

185 ± 10 166 ± 10 146 ± 10

300 ± 1 178 ± 1 152 ± 1

59 ± 1 53 ± 1 171 ± 1

b1 b2 b3

is proportional to roughness parameters such as standard deviation (which indicates vertical grain size), roughness exponent and correlation length (which indicates lateral grain size) [18]. So the annealing temperature is one parameter for controlling the surface activity. The height asymmetry is described by the surface statistical parameters such as skewness and kurtosis. These parameters are calculated by Eqs. (5) and (6), respectively. Fig. 8 indicates the height distribution histograms for samples that acquire from AFM analysis, which is normalized using the central limit theorem. Table 2 shows the values of these parameters for the produced samples. Film surfaces with positive skewness (larger than 0.2) and

Table 2 The values of the skewness and kurtosis. Rku

Rsk

Samples

3.69 4.8 43.12

0.38 0.7 6.15

b1 b2 b3

ZnO thin films were produced by thermal oxidation of Zn layers which were coated on Si (1 0 0) substrate by DC magnetron sputtering. SEM, AFM and PL analysis have been used for characterization of the thin films morphologies and photoluminescence properties. Increase of the annealing temperature leads to larger grain size and stronger peak intensity related to interstitial zinc. The surface roughness was great and the correlation length was low for high annealing temperature (700 ◦ C), smooth and uniform ZnO surface observed in low annealing temperatures (500, 600 ◦ C) and so the correlation length at 500 and 600 ◦ C temperatures were 185 and 166 nm respectively. At low annealing temperatures some whiskers were observed on ZnO surface, but at higher temperatures they were disappeared, and at these temperatures suitable values for skewness and kurtosis also were obtained. The fractal dimensions were estimated by applying the height–height correlation function method. The fractal dimension (Df ) corresponds to changes in the surface morphology which occur due to altering the annealing temperature. Acknowledgment The authors would like to thank and appreciate the financial assistance rendered by Arak University and Iran Nano-Technology Initiative. References [1] D.R. Kammler, T.O. Mason, K.R. Poeppelmeier, Bulk phase relations, conductivity, and transparency in novel bixbyite transparent conducting oxide solution in the cadmium–indium–tin oxide system, Journal of the American Ceramic Society 84 (2001) 1004. [2] S.M. Park, T. Ikegami, K. Ebihara, Investigation of transparent conductive oxide Al-doped ZnO films produced by pulsed laser deposition, Japanese Journal of Applied Physics 44 (2005) 8027. [3] S.H. Jeong, B.N. Park, D.G. Yoo, J.H. Boo, Al–ZnO thin films as transparent conductive oxides: synthesis, characterization, and application tests, Journal of the Korean Physical Society 50 (2007) 622. [4] M. Sibinski, M. Jakubowska, K. Znadek, M. Sloma, B. Guzowski, Carbon nanotube transparent conductive layers for solar cells applications, Optica Applicata XII (2011) 375–381.

168

A. Zendehnam et al. / Applied Surface Science 270 (2013) 163–168

[5] M. Chen, Z.L. Pei, C. Sun, L.S. Wen, X. Wang, Surface characterization of transparent conductive oxide Al-doped ZnO films, Journal of Crystal Growth 220 (2000) 254. [6] D. Zhang, J. Zhang, Q. Wu, Ultraviolet emission of ZnO nano-polycrystalline films by modified successive ionic layer adsorption and reaction technique, Journal of Sol–Gel Science and Technology 54 (2010) 165. [7] U. Ozgur, Y.I. Alivov, C. Liu, A. Teke, M.A. Reshchikov, S. Dogan, V. Avrutin, S.J. Cho, H. Morkoc, A comprehensive review of ZnO materials and devices, Journal of Applied Physics 98 (2005) 041301. [8] L. Xu, L. Shi, X. Li, Preparation of nanocone ZnO thin film and its aging effect of photoluminescence, Applied Surface Science 255 (2009) 5957. [9] C.T. Lee, Fabrication methods and luminescent properties of ZnO materials for light-emitting diodes, Materials 3 (2010) 2218. [10] R.S. Yadav, R.K. Srivastava, M. Kumar, A.C. Pandey, S.G. Prakash, Hydrothermal synthesis and optoelectronic properties of spike-like ZnO nanostructures, Optoelectronics and Advanced Materials-Rapid Communications 4 (2010) 689. [11] J.H. He, Y.H. Lin, M.E. McConney, V.V. Tsukruk, Z.L. Wang, Enhancing UV photoconductivity of ZnO nanobelt by polyacrylonitrile functionalization, Journal of Applied Physics 102 (2007) 084303. [12] D.Y. Kim, J.Y. Son, Horizontal ZnO nanowires for gas sensor application: Aldoping effect on sensitivity, Solid State Letters 12 (2009) J109. [13] G. Ganguly, D.E. Carlson, S.S. Hegedus, D. Ryan, R.G. Gordon, D. Pang, R.C. Reedy, Improved fill factors in amorphous silicon solar cells on zinc oxide by insertion of a germanium layer to block impurity incorporation, Applied Physics Letters 85 (2004) 479. [14] J.C. Johnson, H. Yan, P. Yang, R.J. Saykally, Single nanowire waveguides and lasers, Journal of Physical Chemistry B 107 (2003) 8816. [15] Q.J. Wang, C. Pflugl, W.F. Andress, D. Ham, F.O. Capasso, Gigahertz surface acoustic wave generation on ZnO thin films deposited by radio frequency magnetron sputtering on III–V semiconductor substrates, Journal of Vacuum Science and Technology B 26 (2008) 1071. [16] U. Keller, K.D. Li, B.T. Khuri-Yakub, D.M. Bloom, High-frequency acousto-optic mode locker for picosecond pulse generation, Optics Letters 15 (1990) 45. [17] M. Vahabi, G.R. Jafari, N. Mansour, R. Karimzadeh, J. Zamiranvari, Stochastic features of rough surfaces: analysis of laser-induced silicon surface modification, Journal of Statistical Mechanics 03 (2008) 1742–5468. [18] G.R. Jafari, M.R. Rahimi Tabar, A. Iraji zad, G. Kavei, Etched glass surfaces, atomic force microscopy and stochastic analysis, Physica A 375 (2007) 239–246. [19] A. George, P. Kumari, N. Soin, S.S. Roy, Microstructure and field emission characteristics of ZnO nanoneedles grown by physical vapor deposition, J.A. McLaughlin, Materials Chemistry and Physics 123 (2010) 634. [20] M.D. Barankin, E. Gonzalez II, A.M. Ladwig, R.F. Hicks, Plasma-enhanced chemical vapor deposition of zinc oxide at atmospheric pressure and low temperature, Solar Energy Materials & Solar Cells 91 (2007) 924. [21] S.H. Bae, S.Y. Lee, B.J. Jin, S. Im, Variation of structural, electrical, and optical properties of Zn1–xMgxO thin films, Applied Surface Science 154–155 (2000) 458–461. [22] Y.S. Jung, W.K. Choi, Spatially resolved investigations of the emission around 3.31 eV (A-line) from ZnO nanocrystals, Journal of Applied Physics 99 (2006) 013502.

[23] V. Kapustianyk, B. Turko, A. Kostruba, Z. Sofiani, B. Derkowska, S. DabosSeignon, B. Barwin ski, Yu. Eliyashevskyi, B. Sahraoui, Influence of size effect and sputtering conditions on the crystallinity and optical properties of ZnO thin films, Optics Communication 269 (2007) 346. [24] H. Abdullah, S. Selmani, M.N. Norazia, P.S. Menon, S. Shaari, C.F. Dee, ZnO:Sn deposition by sol–gel method: effect of annealing on the structural, morphology and optical properties, Sains Malaysiana 40 (2011) 245. [25] Y. Zhao, G.C. Wang, T.M. Lu, Characterization of Amorphous and Crystalling Rough Surface: Principles and Applications, Academic Press, Troy, New York, 2001. [26] D. Raoufi, A. Kiasatpour, H.R. Fallah, A.S.H. Rozatian, Surface characterization and microstructure of ITO thin films at different annealing temperatures, Applied Surface Science 253 (2007) 9085–9090. [27] A.M. Mood, F.A. Graybill, D.C. Boes, Introduction to Theory of Statistics, McGraw-Hill, New York, 1974, p. 577. [28] S. Zhao, Y. Zhou, K. Zhao, Z. Liu, P. Han, S. Wang, W. Xiang, Z. Chen, H. Lu, B. Cheng, G. Yang, Violet luminescence emitted from Ag-nanocluster doped ZnO thin films grown on fused quartz substrates by pulsed laser deposition, Physica B 373 (2006) 154. [29] B.K. Choi, D.H. Chang, Y.S. Yoon, S.J. Kang, Laser-assisted sol-gel growth and characteristics of ZnO thin films, Journal of Materials Science – Materials in Electronics 17 (2006) 1011–1015. [30] M. Jung, J. Lee, S. Park, H. Kim, J. Chang, Oxygen rich p-type ZnO thin films using wet chemical route with enhanced carrier concentration by temperaturedependent tuning of acceptor defects, Journal of Crystal Growth 283 (2005) 384. [31] F.K. Shan, G.X. Liu, Annealing effects of ZnO thin films deposited on Si (1 0 0) by using pulsed laser deposition, Journal of the Korean Physical Society 54 (2009) 916. [32] S.Y. Hu, Y.C. Lee, J.W. Lee, J.C. Huang, J.L. Shen, W. Water, The structural and optical properties of ZnO/Si thin films by RTA treatments, Applied Surface Science 254 (2008) 1578. [33] Y.G. Wang, S.P. Lau, H.W. Lee, S.F. Yu, B.K. Tay, X.H. Zhang, H.H. Hng, Photoluminescence study of ZnO films prepared by thermal oxidation of Zn metallic films in air, Journal of Applied Physics 94 (2003) 354. [34] G.H. Lee, Relationship between crystal structure and photoluminescence properties of ZnO films formed by oxidation of metallic Zn, Electronic Materials Letters 6 (2010) 155–159. [35] K. Kongjai, S. Choopun, N. Hongsith, A. Gardchareon, Zinc oxide whiskers by thermal oxidation method, Chiang Mai Journal of Science 38 (2011) 39–46. [36] G. Zou, W. Chen, R. Liu, Z. Xu, Graphene/ZnO nanowire/graphene vertical structure based fast-response ultraviolet photodetector, Materials Letters 61 (2007) 4305. [37] G.P. Daniel, V.B. Justinvictor, P.B. Nair, K. Joy, P. Koshy, P.V. Thomas, Effect of annealing temperature on the structural and optical properties of ZnO thin films prepared by RF magnetron sputtering, Physica B 405 (2010) 1782–1786.