Polyvinyl pyrrolidone assisted low temperature synthesis of ZnO nanocones and its linear and nonlinear optical studies

Materials Research Bulletin 49 (2014) 132–137

Contents lists available at ScienceDirect

Materials Research Bulletin journal homepage: www.elsevier.com/locate/matresbu

Polyvinyl pyrrolidone assisted low temperature synthesis of ZnO nanocones and its linear and nonlinear optical studies M.K. Kavitha a, Honey John a,*, Pramod Gopinath b a b

Department of Chemistry, Indian Institute of Space Science and Technology, Valiamala, Thiruvananthapuram 695547, Kerala, India Department of Physics, Indian Institute of Space Science and Technology, Valiamala, Thiruvananthapuram 695547, Kerala, India

A R T I C L E I N F O

A B S T R A C T

Article history: Received 22 May 2013 Received in revised form 19 August 2013 Accepted 24 August 2013 Available online 31 August 2013

ZnO is synthesized by low temperature solution precipitation technique using polyvinyl pyrrolidone (PVP) as capping agent and at varying pH of the precipitating solution. The as-synthesized ZnO has a cone shape with rough surface, due to the chemisorption of PVP. The results show that PVP has a significant role in the nucleation and growth of ZnO at low temperature and a pH of 11 is required for the formation of the ZnO phase. When it is calcined at 550 8C, the morphology is modified to smooth nanocones. As the ZnO is calcined, optical bandgap is decreased and altered the nonlinear absorption properties of nano ZnO. The optical limiting property of ZnO colloids is investigated using Z-scan technique. Enhanced optical limiting is observed for ZnO calcined at 550 8C. ß 2013 Elsevier Ltd. All rights reserved.

Keywords: A. Semiconductor A. Nanostructures B. Chemical synthesis D. Optical properties

1. Introduction Recently numerous researchers have been focusing on new synthetic methods to control shape, size and crystallinity of semiconductor nanomaterials, which are important in tailoring their physical and chemical properties. Among these nanomaterials ZnO, n-type semiconductor material with a wide and direct band gap of 3.37 eV and a large exciton binding energy of 60 meV [1] at room temperature, has been identified as a good optoelectronic material used in optical limiting devices [2], solar cells [3], light emitting diodes [4], photocatalyst [5] and gas sensors [6]. The tremendous interest in ZnO is provoked by its multifunctional character, which can be varied by controlling the morphology and crystallinity. Recent advances in chemical synthesis have enabled the achievement of ZnO nanocrystals (NCs) with an increased degree of structural complexity and shape. For example, there are reports on the synthesis of ZnO NCs with different morphologies like rods [7], wires [8], belts [9], sheets [10], cubes [11], etc. Several interesting ZnO nanostructures can be synthesized by chemical vapour deposition [12] and thermal evaporation [13]. However, complex process control and sophisticated equipment are needed for the vapour methods. For large scale use, simple and cost-effective synthesis approaches are required. The facile solution method is simple and effective way to prepare

* Corresponding author. Tel.: +91 471 2568536; fax: +91 471 2568541. E-mail addresses: [email protected], [email protected] (H. John). 0025-5408/$ – see front matter ß 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.materresbull.2013.08.056

large scale and highly crystalline ZnO nanostructures at low growth temperature. The optical properties of wide bandgap materials are currently the subject of tremendous interest because of the industrial demand for optoelectronic devices that could be operated at short wavelength. Also there is an increased demand for high nonlinear optical (NLO) materials, which can be incorporated in optoelectronic device with relatively high damage threshold. ZnO have attractive nonlinear optical properties, which make them good candidate for NLO based devices. Because of the high damage threshold, ZnO find its application as an optical limiter. Optical limiters exhibit a decrease in transmittance with increase in incident intensity and this nonlinear absorption (NLA) is referred as reverse saturable absorption (RSA). Herein, we present a simple low temperature wet chemical approach to synthesize ZnO nanostructures in presence of Polyvinyl pyrrolidone (PVP) at ambient pressure. The influence of reaction conditions on the size and shape of ZnO is discussed. The as-obtained ZnO nanopowder has cone shape with rough surface and when it is calcined at 550 8C and the morphology is modified to smooth nanocones. The optical behaviour of ZnO is substantially changed because the optical bandgap is changed upon calcination. We compared the optical limiting behaviour of all the ZnO nanostructures by evaluating the nonlinear absorption coefficient (b) and saturation intensity (Isat) by Z-scan measurement. The optical limiting efficiency of ZnO nanocolloid increases with increase in calcination temperature.

M.K. Kavitha et al. / Materials Research Bulletin 49 (2014) 132–137

Spectrum 100 Spectrometer. Thermogravimetric Analysis (TGA) was performed with TA Q-50 apparatus at the heating rate of 10 8C/ min in nitrogen atmosphere.

2. Experimental 2.1. Materials and synthesis The starting material in this work are Zinc acetate dihydrate (Zn(Ac)22H2O, Merck), Polyvinyl pyrrolidone (PVP, Mol. Wt. 40,000, Sigma–Aldrich), Acetic acid (98%, Merck), De-ionised water and Sodium hydroxide (NaOH, Merck). Zinc acetate dihydrate is stirred with 0.05% PVP solution (in 1% acetic acid) for 24 h to form Zinc acetate-PVP complex. ZnO is precipitated from this complex using NaOH solution. The molar ratio of Zinc acetate and NaOH solution is 1:6. The stirring is continued for 12 h to complete the precipitation. The precipitate is filtered and dried at 100 8C. This method of synthesis is already reported from our group [14]. To study the effect of pH on the growth of ZnO at low temperature, the synthesis was repeated by varying the pH of the precipitating solution. A part of ZnO nanopowder dried at 100 8C is calcined at 550 8C, for studying the effect of calcination temperature on the morphology and optical properties of ZnO. The ZnO nanopowder obtained by heating at 100 8C and 550 8C are indicated as ZnO@100 and ZnO@550 respectively. 2.2. Characterization X-ray Diffraction (XRD) analysis of the obtained powder was performed with a Bruker AXS D8 Advance X-ray diffractometer using Cu Ka (l = 1.54 A˚). The powder XRD can be used for phase identification and to calculate the crystallite size. The crystallite size can be evaluated using Scherrer formula (Eq. (1)). D¼

0:9 l B cos u

133

3. Results and discussion 3.1. Low temperature growth of nano ZnO using PVP The Zn2+ precursor is dissolved in 1% acetic acid solution and subsequently 0.05% aqueous solution of PVP is added. The transparent solution is stirred for 24 h to form Zn2+/PVP complex. The pH of Zn2+/PVP solution is found to be 4. When NaOH solution is added dropwise, pH of the solution slowly increases and turbidity starts to appear at a pH value of 6.4. The addition of NaOH is continued till the pH of the final solution is raised to 11. The reaction mixture is stirred for 12 h at 30–35 8C. The TEM image shown in Fig. 1a is dominated by cone nanostructure with rough surface average size of 90 nm (base)  170 nm (height). Fig. 1b shows the High Resolution Transmission Electron Microscope image of as-formed ZnO. The fringe spacing of 0.28 nm agrees well with the spacing of (1 0 0) lattice planes of ZnO. The XRD patterns (Fig. 2) confirm that ZnO nanostructures formed are hexagonal wurtzite structure. The sharp peaks indicate that the product formed is well crystallized and oriented, further confirming the room temperature crystallization of ZnO by this method. The crystallite size evaluated by Scherer formula is found to be 24 nm. This reveals that cone shaped nanostructures are formed by the crystallization of nanocrystallites with size of 24 nm. 3.2. Effect of calcination on the morphology of as-synthesized ZnO

(1)

where D is the crystallite size, l is the wavelength of X-ray radiation, and B is the line width at the half maximum height. The morphology and crystallinity of the products were characterized by FEI Quanta High Resolution Transmission Electron Microscope (HRTEM). The UV–vis absorption spectra of ZnO dispersed in methanol was obtained on a Varian Cary Bio 100 UV Spectrophotometer at room temperature in the wavelength ranging from 210 nm to 600 nm. A diffused Reflectance Fourier Transform Infrared (FTIR) spectrum was recorded on Perkin-Elmer

When ZnO@100 is heated to 550 8C, the PVP chemisorbed on the surface of ZnO get decomposed and the surface of the nanostructures get smoothened and resulted in a cone shape (Fig. 3a). These cone shaped nanostructures are polydispersed and have an average size of 70 nm (base)  120 nm (height). The XRD pattern of ZnO@550 is shown Fig. 3b. The crystallite size is calculated from XRD pattern and it shows that there is no significant variation in crystallite size, when it is calcined from 100 8C to 550 8C. Thus the ZnO nanocones are polycrystalline with an average crystallite size of 24 nm.

Fig. 1. (a) TEM micrograph and (b) HRTEM micrograph of ZnO@100.

M.K. Kavitha et al. / Materials Research Bulletin 49 (2014) 132–137

134

O

N

Zn2+

O

N

*

n

*

Zn2+ + 4OH

*

[Zn(OH)4]2-

*

n ZnO + H2O + 2OH

Scheme 1. Coordination of Zn2+ with PVP and possible reaction to form ZnO.

Fig. 2. XRD pattern of ZnO@100.

3.3. Growth mechanism Based on the TEM analysis and XRD studies, we are proposing the following growth mechanism. In this study PVP is dissolved in 1% acetic acid solution, in this polar medium PVP is partially polarized. Since the oxygen atom is more electronegative than nitrogen, the partial negative charge resides on oxygen atom and partial positive charge on nitrogen atom. The nitrogen atom locates inside the PVP molecule and has a saturated covalent bond, while the oxygen locates outside of the molecule, hence PVP shows more anionic properties. When the Zinc salt is mixed with PVP, the Zn2+ cation can directly attach to the negative charged end of PVP to form Zn2+/PVP complex. On this coordinated complex the first layer of crystal growth will be initiated [15]. When NaOH is added, OH will react with the coordination complex to form [Zn(OH)4]2, since the growth and the conversion of ZnO is reversible, with the increase in OH concentration, the equilibrium tends to proceed towards right as given in Scheme 1. To confirm the proposed growth mechanism of ZnO, we have done two control experiments. One in which the precipitation reaction is repeated without using PVP, while keeping all other reaction conditions same as ZnO@100. The XRD pattern of assynthesized ZnO is shown in Fig. 4 and the product contains both ZnO phase and Zinc hydroxide phase. The marked peaks correspond to ZnO and all other peaks correspond to Zinc hydroxide. This confirms that in the absence of PVP all hydroxide

phase cannot transform to ZnO. Zhang et al. [16,17] studied the effect of PVP on the nucleation of ZnO. According to their studies, PVP can influence the ZnO growth in two ways. Firstly, it may accelerate the dehydration reaction of Zinc hydroxide by consuming the resultant water via binding water due to the excellent adsorption ability of PVP. Thus it promotes the ZnO nucleation even at low temperature. Secondly the possible coordination between PVP and Zn2+ ions may initiate the ZnO growth. This coordination of as-formed ZnO with PVP may suppress the further aggregation of nanostructures. TGA curve of ZnO@100 without PVP, ZnO@100 and ZnO@550 is shown in Fig. 5. For ZnO@100 with PVP, a weight loss of about 2.5% is observed in the temperature range 130–540 8C, which is due to the decompositon of chemisorbed PVP on the surface of ZnO. No further weight loss and no thermal effect is observed after 540 8C, indicating the presence of pure ZnO phase. But in the TGA curve of ZnO@100 without PVP, there is small weight loss of about 1.8% due to desorption of physisorbed water molecules from 110 8C to 130 8C. After that a significant weight loss of about 13% is observed in the temperature range 130–300 8C, due to the condensation dehydration of surface hydroxyl groups [18]. For ZnO@550, there is no significant weight loss, since it is in pure ZnO phase without chemisorbed PVP. Thus TGA confirms the role of PVP in the transformation of Zinc hydroxide phase to Zinc oxide at low synthesis temperature. In order to study the effect of pH of the reaction medium on the growth of ZnO, we have repeated the reaction for ZnO@100 and extracted the product at different pH. The XRD pattern of ZnO@100 at different pH is shown in Fig. 6 and the marked peak correspond to zinc hydroxide phase. The XRD analysis shows that in all the pH below 11, the samples contain Zinc hyroxide phase. This shows

Fig. 3. (a) TEM micrograph and (b) XRD pattern of ZnO@550.

M.K. Kavitha et al. / Materials Research Bulletin 49 (2014) 132–137

135

Fig. 4. XRD pattern of ZnO@100 synthesized without using PVP.

Fig. 5. TGA curves of ZnO@100 synthesized with and without PVP and ZnO@550.

that at lower pH complete conversion of Zinc hydroxide to ZnO does occur, because at lower pH concentration of OH is not sufficient for the formation of [Zn(OH)4]2, which is the nucleation center for the formation of ZnO. It confirms that at a pH of 11 and the presence of PVP is essential for the formation of pure ZnO at low temperature. In a typical ZnO wurtzite structure, number of alternating planes composed of tetrahedrally coordinated O2 and Zn2+ stacked along the c-axis. Such a crystal structure is composed of  basal polar oxygen plane ð0 0 0 1Þ; a top tetrahedron cornerexposed polar zinc plane (0 0 0 1), and low index facets consisting  of nonpolar ð1 0 1 0Þ planes parallel to c-axis [19]. The nonpolar   ð1 0 1 0Þ facets are more stable than polar (0 0 0 1) and ð0 0 0 1Þ facets because of their relatively lower surface energies. But for ZnO nanocones the high energy ð1 0 1 1Þ facets were exposed  instead of more stable ð1 0 1 0Þ. The formation of hexagonal cone shape of ZnO can be attributed to the different growth rate of the  various growth facets in the order (0 0 0 1) > ð1 0 1 1Þ > ð1 0 1 0Þ. It is well studied that faster the growth rate, quicker will be the disappearance of that plane [20]. Therefore the (0 0 0 1) plane tend to disappear in the growth process and result in tip shaped nanostructures [15]. Since the surface energy of ð1 0 1 1Þ facet is higher and it is more exposed, the crystal termination on that face is favoured by the absorption of capping agent [10]. Due to these chemisorptions of PVP, surface of conical structures become rough. When ZnO@100 is calcined at 550 8C, smooth nanocones are formed as shown in Fig. 3a. It may due to the complete decomposition of chemisorbed PVP. This cone shaped nanostructures have no crystallographically defined side planes, due to the  creation of step egde on the ð0 0 0 1Þ surface at high temperature. Joo et al. [21] reported that, by the formation of step edges, the surface energy of the hexagonal cones can be lowered by reducing the number of dangling bonds on the side surface of the cone.

is due to the formation of shallow levels inside the bandgap as a result of impurity atoms in the lattice or strong disturbances of local symmetry in the ZnO lattice [23,24]. For direct band gap semiconductor, the band gap can be calculated using Tauc equation (Eq. (2)) [25].

ahn ¼ Dðhn  Eg Þ1=2

where hn is the photon energy, Eg is the band gap of the material and D is a constant. Fig. 7b shows the plot of (ahn)2 vs. hn, extrapolating the linear portion of the plot to hn axis gives the bandgap. The bandgap calculated for ZnO@100 is 3.23 eV, which is lower than bandgap of bulk ZnO (3.37 eV). This difference in bandgap is due to the existence of valence band-donor transition [24]. The bandgap for ZnO@550 is found to be 3.16 eV. This decrease in bandgap upon calcination is due to the oxygen vacancies producing intrinsic defects [26]. 3.4.2. Nonlinear absorption and optical limiting behaviour In this present investigation, we have employed Open Aperture (OA) Z-scan technique to measure nonlinear absorption and optical limiting property of the colloids of nano ZnO. The Z-scan technique is a sensitive single beam technique for measuring the optical nonlinearities by moving the sample through the focus of a

3.4. Linear and nonlinear optical properties of nano ZnO 3.4.1. UV–vis absorption spectra The UV–vis absorption spectra of ZnO@100 and ZnO@550 are shown in Fig. 7a. It is clear from the spectra that for all ZnO samples there is a blue shift in the excitonic absorption edge compared to the bulk ZnO which has an absorption edge reported at 380 nm [22]. ZnO@100 shows more blue shift in the absorption edge (362 nm) when compared to ZnO@550 (372 nm). The absorption edge shift to higher wavelength with the increase in calcination temperature, indicating the presence of smaller bandgap for calcined ZnO. This red shift in band edge wavelength ZnO nanopowders synthesized at high temperature is reported which

(2)

Fig. 6. XRD pattern of ZnO@100 synthesized at different pH.

136

M.K. Kavitha et al. / Materials Research Bulletin 49 (2014) 132–137

Fig. 7. (a) UV–vis absorption spectra and (b) Tauc plot of ZnO@100 and ZnO@550.

Gaussian beam and detecting the far-field sample transmittance as a function of sample position (z). The 532 nm radiation from a 5 ns Q-switched Nd: YAG laser operating at 10 Hz is used as the source to induce nonlinearity in the sample. A lens of focal length 10.5 cm is used to focus the laser beam and the sample is mounted on a computer controlled translation stage and moved through the focal point along the z-axis. The transmitted beam energy and the reference beam energy are measured simultaneously by an energy ratio meter (RJ7600, Laser Probe Corp.) having two identical pyroelectric detector heads (Rjp735). The Z-scan system is initially calibrated using CS2 as a standard. ZnO colloid is taken in a 1 mm thick quartz cuvette and is placed at the focus of the lens and scanned on either side of it. The data are analyzed by using the procedure described by Sheik Bahae et al. [27] and the nonlinear absorption coefficient is obtained by fitting the experimental Zscan data with the theoretical plot. The open aperture curve exhibits a normalized transmittance valley, indicating the presence of Reverse Saturable Absorption

(RSA) in the colloids. The theoretical plot for two photon absorption process fit well with the experimental data indicating the nonlinear absorption process to be third order. The corresponding net transmission is given by Tðz; S ¼ 1Þ ¼



1 pq0 ðz; 0Þ

pffiffiffiffi

 Z1

h i 2 ln 1 þ q0 ðz; 0Þet dt

(3)

1

where q0(z, 0) = bI0 Leff, b is the nonlinear absorption coefficient, I0 is the on-axis peak intensity at the focus (z = 0), Leff = [1  exp (al)]/a, is the effective thickness of the sample, a is the linear absorption coefficient, l is the thickness of sample cell. The value of b can be evaluated from the theoretical fitting of the Z-scan experimental data. When the crystallite size is reduced to the order of Bohr exciton radius aB, quantum confinement effect occurs and drastic change in optical properties are expected. For semiconductor nanocrystals the quantum confinement effect can be classified into two regimes,

Fig. 8. Normalized Z scan transmittance and corresponding optical limiting response curve of ZnO@100 and ZnO@550. The circles denote the experimental results. The solid line indicates the theoretical fitting curves.

M.K. Kavitha et al. / Materials Research Bulletin 49 (2014) 132–137

namely strong and weak confinement regimes, according to the ratio of nanocrystal radius R to aB [28]. Nonlinear optical properties in nanocrystals have also been investigated for both the confinement regimes. In strong confinement regime, the photoexcited electron and hole are individually confined. Nonlinearity in this regime is due to state-filling effect. The size dependent third order nonlinearity in the strong confinement regime has been studied, but the results are inconsistent. Hall et al. [29] has reported larger third order susceptibility for large size CdSxSe1x nanocrystals, but Roussignol et al. [30] have shown that larger susceptibility for smaller particles. In ZnO the exciton Bohr radius is 2 nm, and in the present study the average crystallite size of ZnO is about 24 nm. Hence it comes under weak confinement regime. In weak confinement regime, the Coulomb interaction between hole and electron yields an exciton, and it is confined. The nonlinearity occurs due to exciton–exciton interaction [31]. The far field normalized transmittance as a function of the distance for the ZnO nanopowder dispersed in water is shown in Fig. 8. All these colloids clearly exhibit a normalized transmittance valley, indicating the presence of a Reverse Saturable Absorption. For an input fluence of 14.13 J/cm2 and 65% linear transmittance, the experimentally obtained values of nonlinear absorption coeffient (b) for ZnO@100 and ZnO@550 are 0.8 cm/GW and 2.1 cm/GW respectively. Theoretical fit taking into account two photon absorption process agrees with the Z-scan experimental data and they exhibit an enhancement in two photon absorption coefficient (b) with the increase in calcination temperature. The saturation intensity is decreasing with increasing calcination temperature. This indicates the viability of ZnO calcined at higher temperature as optical limiter. Optical limiters are those materials which transmit light at low input fluence but its transmittance decrease at high inputs. A material with low saturation intensity is a good optical limiter. The plot of normalized transmittance to input intensities at each position is given in Fig. 8. The saturation intensity of ZnO@550 is found to be a low value (5 GW/cm2) compared to ZnO@100 (9 GW/cm2). Thus ZnO calcined at 550 8C exhibit an enhanced optical limiting property. 4. Conclusion Nano ZnO is synthesized by low temperature solution precipitation method using PVP as nucleating as well as capping agent. The as-synthesized ZnO has a cone shape with rough surface due to the presence of chemisorbed PVP. In this method PVP has a significant role in the nucleation and growth of ZnO and a pH of 11 is required for the formation of pure ZnO at lower temperature. When it is calcined at 550 8C, the morphology is modified to

137

smooth nanocones. Calcination of ZnO nanopowder significantly influences its optical properties. Optical bandgap is reduced and nonlinear absorption of ZnO is increased with calcination. Acknowledgements Authors wish to thank Dr. Reji Plilip, Raman Research Institute, Banglore for extending the laboratory facilities for Z-scan measurement. They also thank NIIST, Trivandrum, and IIT Madras for TEM and XRD studies.

References [1] Q. Yu, W. Fu, C. Yu, H. Yang, R. Wei, M. Li, S. Liu, Y. Sui, Z. Liu, M. Yuan, G. Zou, G. Wang, C. Shao, Y. Liu, J. Phys. Chem. C 111 (2007) 17521–17526. [2] C.H.O.S.K. Min, G.J. Lee, Y.P. Lee, J. Korean Phys. Soc. 55 (2009) 1005–1008. [3] Q. Zhang, C.S. Dandeneau, X. Zhou, G. Cao, Adv. Mater. 21 (2009) 4087–4108. [4] Y.J. Kim, J.H. Lee, G.C. Yi, Appl. Phys. Lett. 95 (2009) 213101–213103. [5] L. Jiang, L. Gao, Mater. Chem. Phys. 91 (2005) 313–316. [6] H. Tang, M. Yan, X. Ma, H. Zhang, M. Wang, D. Yang, Sens. Actuators B Chem. 113 (2006) 324–328. [7] K.S. Ahn, S. Shet, T. Deutsch, C.S. Jiang, Y. Yan, M. Al-Jassim, J. Turner, J. Power Sources 176 (2008) 387–392. [8] A.M. Glushenkov, H. Zhang, J. Zou, G.Q. Lu, Y. Chen, J. Cryst. Growth 310 (2008) 3139–3143. [9] Z.L. Wang, Mater. Sci. Eng. R 64 (2009) 33–71. [10] J. Chang, E.R. Waclawik, CrystEngComm 14 (2012) 4041–4048. [11] X. Feng, W. Ji, Opt. Express 17 (2009) 13140–13150. [12] K. Hani, C. Guangyu, L. Oleg, H. Helge, P. Sanghoon, S. Alfons, C. Lee, J. Phys. D Appl. Phys. 42 (2009) 135304. [13] J. Khanderi, R.C. Hoffmann, A. Gurlo, J.J. Schneider, J. Mater. Chem. 19 (2009) 5039–5046. [14] P.C. Haripadmam, M.K. Kavitha, H. John, B. Krishnan, P. Gopinath, Appl. Phys. Lett. 101 (2012) 071103–071105. [15] P. Gu, X. Wang, T. Li, H. Meng, H. Yu, Z. Fan, J. Cryst. Growth 338 (2012) 162–165. [16] J. Zhang, H. Liu, Z. Wang, N. Ming, Z. Li, A.S. Biris, Adv. Funct. Mater. 17 (2007) 3897–3905. [17] J. Zhang, H. Liu, Z. Wang, N. Ming, J. Cryst. Growth 310 (2008) 2848–2853. [18] K. Morishige, S. Kittaka, T. Moriyasu, T. Morimoto, J. Chem. Soc. Faraday Trans. 1 76 (1980) 738–745. [19] L. Xu, Y. Guo, Q. Liao, J. Zhang, D. Xu, J. Phys. Chem. B 109 (2005) 13519–13522. [20] W.J. Li, E.W. Shi, W.Z. Zhong, Z.W. Yin, J. Cryst. Growth 203 (1999) 186–196. [21] J. Joo, S.G. Kwon, J.H. Yu, T. Hyeon, Adv. Mater. 17 (2005) 1873–1877. [22] X.C. Liu, E.W. Shi, Z.Z. Chen, H.W. Zhang, B. Xiao, L.X. Song, Appl. Phys. Lett. 88 (2006) 252503. [23] V. Ischenko, S. Polarz, D. Grote, V. Stavarache, K. Fink, M. Driess, Adv. Funct. Mater. 15 (2005) 1945–1954. [24] V. Srikant, D.R. Clarke, J. Appl. Phys. 83 (1998) 5447–5451. [25] J. Tauc, The Optical Properties of Solids, Academic Press, New York, 1966. [26] D. Pradhan, S.K. Mohapatra, S. Tymen, M. Misra, K.T. Leung, Mater. Express 1 (2011) 59–67. [27] M. Sheik-Bahae, A.A. Said, E.W. van Stryland, Opt. Lett. 14 (1989) 955–957. [28] Y. Kayanuma, Phys. Rev. B 38 (1988) 9797–9805. [29] D.W. Hall, N.F. Borrelli, J. Opt. Soc. Am. B 5 (1988) 1650–1654. [30] P. Roussignol, D. Ricard, C. Flytzanis, Appl. Phys. B Lasers Opt. 51 (1990) 437–442. [31] A. Nakamura, H. Yamada, T. Tokizaki, Phys. Rev. B 40 (1989) 8585–8588.