ORIGINAL PAPER

Effect of annealing atmosphere on structural, opticaland electrical properties of Al-doped Zn12xCdxO thin films

L. B. Duan • X. R. Zhao • J. M. Liu •

W. C. Geng • H. N. Sun • H. Y. Xie

Received: 5 September 2011 / Accepted: 29 February 2012 / Published online: 9 March 2012

� Springer Science+Business Media, LLC 2012

Abstract 1 at.% Al-doped Zn1-xCdxO (x = 0–8 at.%)

thin films were prepared on glass substrates by sol–gel

method. The codoping films retained the hexagonal wurtzite

structure of ZnO, and showed preferential c-axis orientation.

The effect of annealing ambient (in vacuum and nitrogen) on

the optical and electrical properties of (Cd,Al)-codoped ZnO

films were investigated using transmission spectra and

electrical measurements. The transmittances of the codop-

ing films were obviously degraded by vacuum annealing to

50–60 %, but enhanced to 70–80 % after nitrogen anneal-

ing. The carrier concentration and Hall mobility both

increased, and resistivity decreased with narrowing band

gap of Al-doped Zn1-xCdxO, below different critical con-

centrations x = 4 % (in vacuum) and x = 6 % (in nitrogen).

It is revealed that the conductivity is also improved by Cd

doping along with band gap modification. The variations in

optical and electrical properties are ascribed to both the

changes of the crystallinity and concentration of oxygen

vacancies under different ambient. In view of transmittance

and conductivity, nitrogen annealing might be a more

effective post-annealing way than vacuum annealing for our

(Cd,Al)-codoped ZnO films to meet the requirements of

transparent conducting oxide (TCO).

Keywords (Cd,Al)-codoped ZnO films � Transparent

conducting � Band gap modification � Crystallinity �Oxygen vacancies

1 Introduction

Transparent conducting oxide (TCO) films, which are

characterized by a unique combination of low electrical

resistivity and high optical transparency, have been widely

investigated for their practical applications as window

materials in flat panel display, solar cells and light emitting

diodes (LED) [1–3]. Tin-doped indium oxide (ITO) is a

traditional material for TCO applications, however, indium

is rare and its supply may be limited by the availability of

natural resources. Recent developments support that Al-

doped ZnO (AZO) is one of the most promising candidates

for replacing ITO, due to its relatively abundance, low

fabrication cost and thermal stability [4–7]. A number of

investigators examined the electrical and optical properties

of Al-doped ZnO films with a broad range of nominal Al-

doping, and critical Al contents with the minimum resis-

tivity were usually observed, depending on preparation

method and conditions [8–11]. Lu et al. [10] reported that

the optimal result for AZO films synthesized by magnetron

sputtering was obtained at Al content of 4 %. Nasr et al.

[11]. revealed that the minimum resistivity was gotten

around 1.5–2 % Al-doped ZnO films prepared by sol–gel

method, and above the critical concentration, the segre-

gated Al started to dominate the electrical transport.

In addition, the band gap of ZnO (Eg = 3.36 eV) can be

tuned by alloying with bivalent Mg or Cd. While Mg is

known to enhance the band gap, Cd substitution leads to

reduce in band gap, the resultant (Zn,Cd)O and (Zn,Mg)O

alloys have allowed band gap covering a wide range of

2.8–4.5 eV in practice [12–14]. A TCO with modified band

gap gives rise to many of its scientific and technical

applications, such as improving the efficiency of different

wavelength light emitting devices when used as a trans-

parent electrode, and the realization of heterojunction and

L. B. Duan (&) � X. R. Zhao � J. M. Liu �W. C. Geng � H. N. Sun � H. Y. Xie

Key Laboratory of Space Applied Physics and Chemistry,

Ministry of Education of China and School of Science,

Northwestern Polytechnical University, Xi’an 710072,

People’s Republic of China

e-mail: lbduan@nwpu.edu.cn

123

J Sol-Gel Sci Technol (2012) 62:344–350

DOI 10.1007/s10971-012-2731-9

superlattice structures [15–17]. Theoretically, due to the

isovalent ions of Mg2?, Cd2? and Zn2?, there is no con-

tribution of extra free charge carriers for AZO from the Mg

or Cd substitution. However, the incorporation of Mg or Cd

may enhance electron scattering and grain boundary barrier

effects and then destroy the conductivity [16, 18]. There-

fore, Mg or Cd doping content might affect both of the

optical and electrical properties of band gap engineered

AZO. Resistivity of Al-doped Zn1-xMgxO films was

always found to increase with increasing Mg composition

[15, 19], while the effect of narrowing band gap by Cd

doping on the electrical and optical properties of (Cd,Al)-

codoped ZnO films has been rarely reported.

Numerous deposition methods have been applied to

prepare highly conductive and transparent ZnO-based thin

films, including chemical vapor deposition (CVD),

molecular beam epitaxy (MBE), pulsed laser deposition

(PLD), magnetron sputtering, etc. [10, 11]. However, the

industrial production is limited due to the complex and

expensive vacuum technique. Furthermore, the preparation

of homogeneous and large-area films is also an upfront

challenge. The sol–gel method is a kind of cost-effective

process and is helpful to realize the preparation of large-

area homogeneous films [20]. More importantly, sol–gel

method has the distinct advantages in excellent composi-

tion control and the ability to achieve atomic scale mixing

of individual components [21]. In addition, post-annealing

treatment by various atmospheres, such as air, oxygen,

hydrogen, nitrogen, or in vacuum for as-prepared TCO

films is usually considered as an essential and effective

technique to improve the electrical and optical properties

[22]. The effects of atmospheres could be simply divided

into two classes, (a) Annealing the TCO films in air,

oxygen or nitrogen could improve the crystallinity and

transmittance, but also would degrade the electrical

properties due to the chemisorptions of O2 or N2, which

were trapped at grain boundary and/or formed AlOx or

AlNx; (b) Annealing in hydrogen environment was

reported to be significant in improving the electrical

conductivity of AZO films, and the reason was considered

to be both the production of additional oxygen vacancies

and desorption of the absorbed oxygen at the grain

boundaries. Treatment in vacuum, which was qualitatively

similar to the effect of annealing in hydrogen, was also

usually introduced to improve the conductivity of TCO by

enhancing oxygen vacancies [16, 22, 23]. In this work, we

prepared lightly (1 at.%) Al codoped Zn1-xCdxO (with

nominal Cd content x = 0–8 at.%) thin films by dip-

coating sol–gel method. Consequently, the effects of post-

annealing in two representative ambient (in vacuum and

nitrogen) on the structural, optical and electrical properties

of the band gap modified (Cd,Al)-codoped ZnO thin films

were investigated.

2 Experimental procedure

Analytical grade zinc acetate [Zn(CH3COO)2�2H2O] was

firstly dissolved in a 2-methoxyethanol (C3H8O2) and

monoethanolamine (MEA, C2H7NO) solution at room

temperature. The concentration of the sol was 0.75 mol/L

and the molar ratio of MEA to zinc acetate was kept at 1.0.

The solution was stirred at 60 �C for 1 h until it became

clear and homogeneous. Aluminum nitrate [Al(NO3)3�9H2O] and/or cadmium nitrate [Cd(NO3)3�4H2O] were

added into some of the previous solutions in an appropriate

ratio and then stirred vigorously at 60 �C for another 1 h.

The final solutions served for coating were aged for 36 h at

room temperature. The slice glass (Sail 7101, China) was

used as substrate for sol–gel dip-coating after being

cleaned in an ultrasonic bath for 15 min with HCl, ethanol

and distilled water, respectively. The glass sheets were

dipped into the aged solution for 1 min and then pulled out

at the velocity of 4 cm/min. The films were kept at 100 �C

in a drying box for 10 min and then preheated at 500 �C for

15 min. Such process was repeated for 10 times, and then

the 10-layer films with nominal Cd content x = 0–8 at.%

was firstly annealed at 550 �C in air for 1 h. However, the

resistivity of the air-annealed films is out of the measure-

ment range of our Hall effect system. To improve the

conductivity and/or transmittance, one group of 10-lay

films with nominal Cd content x = 0–8 at.% was further

post-annealed in vacuum (P * 10-2 Pa), while the other

group of 10-lay films in nitrogen at 550 �C for 1 h,

respectively.

X-ray diffraction (XRD) patterns were collected from

20� to 80� using PANalytical X’pert MPD PRO with Cu Ka

radiation. Optical properties such as transmittance were

measured by UV–VIS spectrophotometer (Hitachi UV–VIS

spectrophotometer U3010) in the wavelength of

300–800 nm. The thickness of the samples was carried out

by spectroscopic ellipsometer (Spec EI-2000-VIS). The

electrical properties such as resistivity, carrier concentra-

tion, and Hall mobility were detected by Hall effect mea-

surements in the Van der Pauw configuration using an

electrical transport property measurement system (Beijing

Jingcheng, China, ET-9000) at room temperature.

3 Results and discussion

Figure 1 displays the XRD patterns of 1 at.% Al-doped

Zn1-xCdxO (x = 0–8 % with an increment of 2 at.%) thin

films treated by vacuum and nitrogen annealing, respec-

tively. It is implied that all the films have a single phase

which can be identified as the hexagonal wurtzite structure

of ZnO (space group P63mc). No trace of other impurities

is found within the detection limit of instrument. All the

J Sol-Gel Sci Technol (2012) 62:344–350 345

123

films show an extremely pronounced (002) texture with

dominant peak 2h& 34.4�, indicating that the preferred

orientation is along the crystallographic c-axis and per-

pendicular to the substrate. According to the Vegard’s law,

since the ionic radius of Zn2? (0.60 A, coordination

number CN = 4) is smaller than that of Cd2? (0.78 A,

CN = 4), as showed in the inset of Fig. 1a, the lattice

constant c evaluated from the shift of the position of (002)

peak shows a roughly increasing evolution from x = 0 to

x = 8 %, indicating a statistical substitution of Cd2? for

Zn2? in their solid solution after vacuum annealing.

However, as displayed in the inset of Fig. 1b, the lattice

constant c of films annealed in nitrogen increases slightly

from x = 0 to x = 4 %, and then decreases with x. It

manifests the reduction of c-axis preferred orientation and

more internal planes are exhibited [24], which can be

revealed by the appearance of (100) and (101) peaks in

Fig. 1b.

Actually, all (Cd,Al)-codoped ZnO thin films were

originally processed under air annealing. The optical

transmission spectra exhibit a high transmittance (about

80–90 %) in visible region and a high absorption (near

100 %) in ultraviolet (UV) region (not shown here).

However, the resistivity is out of the measurement range of

our Hall effect system, this degradation of conductivity

might be due to the chemisorptions of O2 [23]. Therefore,

we post-annealed these films in vacuum (*5 9 10-2 Pa)

and nitrogen, respectively. The growth dynamics might

differ in the films post-annealed in vacuum and nitrogen. In

vacuum, the c-axis preferred orientation of air annealed

films is almost retained with the production of additional

oxygen vacancies and desorption of the absorbed oxygen at

the grain boundaries, and the electrical conductivity is

expected to be enhanced [22, 23]. In the nitrogen ambient,

the possible filling and/or replacement of oxygen by

nitrogen, which could improve the crystallinity and trans-

mittance of the films, and also postpone the growth of c-

axis orientation, and induce the growth in a,b orientations

[24].

Figure 2 displays the summarized electrical properties

of Al-doped Zn1-xCdxO films post-annealed in vacuum.

When x \ 4%, the resistivity decreases slightly with Cd

content until a platform, while the carrier concentration and

Hall mobility both increase. This is contrast to the results

of (Mg,Al)-codoped ZnO, in which the resistivity was

found to increase with Mg content [15, 19]. With higher Cd

doping, the Hall mobility drops rapidly, due to more

20 40 60 80

(101

)

Nitrogen

(b)

0 2 4 6 85.20

5.22

Lat

tice

co

nst

ant

c ( Å

)

Cd content x (%)

Inte

nsi

ty (

a.u

.)

2θ (deg.)

x=8%

x=6%

x=4%

x=2%

x=0

(100

)

0 2 4 6 8

5.21

5.22

Lat

tice

co

nst

ant

c (Å

)

Cd content x (%)

Inte

nsi

ty (

a.u

.) x=8%

x=6%

x=4%

x=2%

x=0Vacuum

(a)

(002

)

Fig. 1 XRD patterns of 1 at.% Al-doped Zn1-xCdxO (x = 0–8 %

with an increment of 2 at.%) thin films treated by vacuum (a) and

nitrogen (b) annealing. Inset: the lattice constant c evaluated from the

shift of the position of (002) peak

0

1

2

2

3

4

5

0 2 4 6 815

20

25

30

(b)

(c)

(a)

Res

isti

vity

(10

-2 c

m)

Hal

l mo

bili

ty (

cm2v-1

s-1)

Car

rier

co

nce

ntr

atio

n (

1019

cm-3)

Cd content x (at.%)

Fig. 2 Resistivity (a), carrier concentration (b), and Hall mobility

(c) of 1 at.% Al-doped Zn1-xCdxO films post-annealed in vacuum

346 J Sol-Gel Sci Technol (2012) 62:344–350

123

defects and/or residual strain might be introduced, i.e. the

electron scattering was enhanced [25]. Therefore, the

doping of Cd in band gap modified (Cd,Al)-codoped ZnO

films also has an optimal content for conductivity, similar

to the cases of Al-doping in ZnO [10, 11].

Transmittance spectra of vacuum treated (Cd,Al)-cod-

oped ZnO films in the wavelength range of 300–800 nm

are shown in Fig. 3a. The transmittances in visible region

of vacuum treated films are obviously decreased to

50–60 %, due to more oxygen vacancies are introduced

and the degradation of crystallization after vacuum pro-

cessing [22, 23]. The optical energy gap Eg for the direct

electron transition can be determined using the following

equation:

ahm ¼ Cðhm� EgÞ1=2 ð1Þ

where C is a constant, a is the absorption coefficient and mis the photon frequency [26]. As shown in Fig. 3b, the

fundamental absorption, which corresponds to the electron

excitation from valance band to conduction band, is usually

used to calculate the value of band gap using Eq. 1 by

plotting (ahm)2 as a function of the photo energy and by

extrapolating the linear region to the energy axis. The

determined values are shown in the inset of Fig. 3b. The

linear variation of band gap by Cd doping in ZAO further

confirms that the Cd is doped into the matrix, which is

consistent with our result of XRD study. As a result, Cd

works effectively on band gap engineering, irrespective of

the existence of Al codoping.

As we know, the traditional ITO films commonly have

high transmittance in visible region (at least 90 %) [27–

29], and the transmittance and resistivity of inexpensive

AZO are nearly comparable to those obtained from ITO

films [10, 11, 15]. Obviously, the low transmittance

(50–60 %) of our vacuum treated (Cd,Al)-codoped ZnO

films is far from fulfilling the minimum requirement for

TCO, although the band gap and resistivity could be turned

down by Cd doping. The transmittance and conductivity of

our TCO films seem irreconcilable. As we mentioned above,

oxygen and nitrogen annealing have been proved to be good

at improving the crystallinity and transmittance [22], and the

optical transmission spectra of our air-annealed films exhibit

high transmittances (about 80–90 %), but bad conductivity.

Alternatively, the air annealed films were further post-

annealed in nitrogen. Transmittance spectra of nitrogen

treated Al-doped Zn1-xCdxO films are shown in Fig. 4a, the

transmittances in visible region of nitrogen treated films are

distinctly increased to 70–80 %, comparing with the films

annealed in vacuum (50–60 %), which are more close to

those of ITO films. The determined Eg after nitrogen

annealing is shown in the inset of Fig. 4b. The Eg of Al-

doped Zn1-xCdxO films after nitrogen annealing decreases

slowly below x = 4 %, whereas shows a fast drop when

x = 6 %, and then increases at x = 8 %. The irregular var-

iation illustrates a probable compositional non-uniformity at

higher Cd doping content after nitrogen annealing [16, 22,

25].

Figure 5 displays the summarized electrical properties

of Al-doped Zn1-xCdxO films post-annealed in nitrogen.

Similar to the evolution of vacuum annealing films, the

resistivity decreases with Cd content until a platform at

x = 6 %, while the carrier concentration and Hall mobility

increase. Interestingly, irrespective of vacuum or nitrogen

annealing, the band gap and resistivity of films have almost

same decreasing evolution, and there both exists critical

points (4 % for vacuum annealing, 6 % for nitrogen

annealing) for conductivity.

In Al-doped ZnO films studied by Lu et al. [10] and

Nasr et al. [11], the initial increase in the band gap with Al

2.5 3.0 3.5

0 2 4 6 83.20

3.25

3.30

3.35

Ban

d g

ap (

eV)

Cd content y (at.%)

x=8% x=0

(αhν

)2 (a

.u.)

Photon energy (eV)

(b)

300 400 500 600 700 800

0

20

40

60

80(a)

Vacuum

Tra

nsm

itta

nce

(%

)

Wavelength (nm)

x=0

x=2%

x=4%

x=6%

x=8%

Fig. 3 a Transmission spectra of 1 at.% Al-doped Zn1-xCdxO films

treated by vacuum annealing. b Plot of square of the absorption

coefficient versus photon energy for the vacuum treated Al-doped

Zn1-xCdxO films. Inset: the optical band gap energies as a function of

Cd content

J Sol-Gel Sci Technol (2012) 62:344–350 347

123

doping (having higher carrier concentration) can be

explained by the band gap widening phenomenon descri-

bed by Burstein and Moss [30, 31]. The Fermi level in

degenerate semiconductors is above the conduction band

edge (due to partially-filled states in the conduction band),

optical excitations from valence band to the Fermi level

require an extra energy. It has been shown that this wid-

ening is a function of the carrier density according to the

formula:

DEg ¼h2

8m�c

3

p

� �2=3

n2=3c ð2Þ

where h is the Planck constant, m�c is the reduced effective

mass, and nc is the charge carrier concentration. At higher

doping concentration of Al, the dopant atoms tend to

segregate more and more at the grain boundaries, signifi-

cantly increasing impurity scattering. Therefore, the

subsequent shrinkage of the band gap can be due to the

electron impurity scattering as described by Hamberg et al.

[32] In our (Cd,Al)-codoped ZnO films, the dependence of

band gap on carrier concentration is contrast to that of Al-

doped ZnO [10, 11], which manifests that the underline

physical origin might be quite different.

Generally, there are several scattering mechanisms in

the films, such as lattice vibration scattering, grain

boundary scattering, ionized impurity scattering, neutral

impurity scattering, etc., depending on the range of carrier

concentration and the temperature of the films [10, 20]. In

the present study, all the measurements of the conductive

properties were collected at room temperature and the

effect of lattice scattering which was determined by the

variation of temperature could be ruled out. Within the

framework of potential barrier model at grain boundaries

[33], defects located at grain boundaries can act as carrier

traps and the trapped electrons set up a negative charge,

contributing to rise to a space charge region in the grains.

This will produce a potential barrier at the grain boundaries

and decrease the Hall mobility of carriers. Zhu et al. [20]

find that when the carrier concentration is low (approxi-

mately N B 5.0 9 1018 cm-3), grain boundary model is

valid. However, the barrier height will decrease with the

increase in carrier concentration, when the carrier

300 400 500 600 700 800

0

20

40

60

80

100(a)

Tra

nsm

itta

nce

(%

)

Wavelength (nm)

x=0

x=2%

x=4%

x=6%

x=8%Nitrogen

2.5 3.0 3.5

0 2 4 6 8

3.20

3.25

3.30

Ban

d g

ap (

eV)

Cd content y (at.%)

x=6%,8%,2%,4%,0

(αhνν

)2 (a

.u.)

Photon energy (eV)

(b)

Fig. 4 a Transmission spectra of 1 at.% Al-doped Zn1-xCdxO films

treated by nitrogen annealing. b Plot of square of the absorption

coefficient versus photon energy for the nitrogen treated Al-doped

Zn1-xCdxO films. Inset: the optical band gap energies as a function of

Cd content

2

4

6

8

1.2

1.6

2.0

2.4

0 2 4 6 85

10

15

(a)

(b)

(c)

Res

isti

vity

(10

-2Ω

cm

)H

all m

ob

ility

(cm

2 v-1s-1

)

Cd content x (at.%)

Car

rier

co

nce

ntr

atio

n (

1019

cm-3)

Fig. 5 Resistivity (a), carrier concentration (b), and Hall mobility

(c) of 1 at.% Al-doped Zn1-xCdxO films post-annealed in nitrogen

348 J Sol-Gel Sci Technol (2012) 62:344–350

123

concentration is high, the carrier tunneling current plays a

more important role in the transportation of electrons and

the grain boundary scattering even can be neglected. The

carrier concentrations of our (Cd,Al)-codoped ZnO films

are higher ([1.0 9 1019cm-3), so the grain boundary

scattering was tiny enough to be ignored. In this case,

ionized impurity and natural scattering should make main

contribution to the variation of the Hall mobility.

In thermal equilibrium, the electron concentration in

conduction band (n0) and hole concentration (p0) in valence

band has the relation,

n0p0 ¼ 42pk

h2

� �3

ðm�nm�pÞ3=2T3 exp � Eg

kT

� �ð3Þ

where h is the Planck constant, k is the Boltzmann constant,

m�n and m�p is the electron and hole effective mass,

respectively [34]. In our Al-doped Zn1-xCdxO thin films,

the n-type carriers are dominant, the carrier concentration

increases expectedly with narrowing energy band, which is

rather contrast the carrier concentration dependence of

band gap shift in Al-doped ZnO films reported by Lu et al.

[35]. Meanwhile, the resistivity (q) and mobility (l) can be

defined as

q ¼ 1

lne; l ¼ esn

m�nð4Þ

where sn is the electron average scattering time and e is the

electron charge [36]. In Figs. 2 and 5, below the critical Cd

content, films with higher Cd content possess higher

mobility and higher carrier concentration, while lower

resistivity. Actually, the effective mass of Cd is larger than

that of Zn, so the electron average scattering time of

(Cd,Al)-codoped ZnO might be shorter than that of AZO.

Therefore, the conductivity of AZO is also improved by

Cd codoping, along with Eg modification. Comparing with

vacuum annealing, in view of transmittance and conduc-

tivity, the nitrogen annealing might be a more appropriate

post-annealing way than vacuum annealing for our

(Cd,Al)-codoped ZnO films to meet the requirements of

TCO. Due to the improvement of crystallization, the

transmittance is enhanced with sacrificing a portion of

conductivity after nitrogen annealing, but the order of

magnitude of conductivity is same in the films post-

annealed in vacuum and nitrogen.

4 Summary

(Cd,Al)-codoped ZnO thin films were prepared on glass

substrates by sol–gel method. The codoping thin films

retained a single phase which can be identified as the

hexagonal wurtzite structure of ZnO, and showed

preferential c-axis orientation. The transmittances of these

films are obviously degraded by vacuum annealing to

50–60 %, but enhanced to 70–80 % after nitrogen

annealing. The carrier concentration and Hall mobility both

increase, and resistivity decreases with narrowing band gap

in Al-doped Zn1-xCdxO below different critical concen-

tration x = 4 % (in vacuum) and x = 6% (in nitrogen). It

is revealed that the conductivity of AZO is also improved

by Cd doping, which is originally introduced for Eg mod-

ification. The variations in optical and electrical properties

are attributed to both the changes of the crystallinity and

concentration of oxygen vacancies under different ambient.

In view of transmittance and conductivity, the nitrogen

annealing might be a more appropriate post-annealing way

than vacuum annealing for our (Cd,Al)-codoped ZnO films

to meet the requirements of TCO.

Acknowledgments This work is financially supported by NPU

Foundation for Fundamental Research (NPU-FFR-JC201017), Ph.D.

Programs Foundation of Ministry of Education of China (Grant No.

20106102120051) and National Natural Science Foundation of China

(Grant No. 50872112, 51172186).

References

1. Chopra KL, Major S, Pandya DK (1983) Thin Solid Films 102:1

2. Dawar AL, Joshi JC (1984) J Mater Sci 19:1

3. Zhao L, Zhao XR, Liu JM, Zhang A, Wang DH, Wei BB (2009) J

Sol-Gel Sci Technol 53:475

4. Minami T (2000) MRS Bull 25:38

5. Jiang X, Wong FL, Fung MK, Lee ST (2003) Appl Phys Lett

83:1875

6. Li ZZ, Chen ZZ, Huang W, Chang SH, Ma XM (2011) Appl Surf

Sci 257:8486

7. Oliveira C, Rebouta L, Lacerda-Aroso T, Lanceros-Mendez S,

Viseu T, Tavares CJ, Tovar J, Ferdov S, Alves E (2009) Thin

Solid Films 517:6290

8. Mass J, Bhattacharya P, Katiyar RS (2003) Mater Sci Eng B

103:9

9. Lin SS, Huang JL, Sajgalik P (2004) Surf Coat Technol 185:254

10. Lu JG, Ye ZZ, Zeng YJ, Zhu LP, Wang L, Yuan J, Zhao BH,

Liang QL (2006) J Appl Phys 100:073714

11. Nasr B, Dasgupta S, Wang D, Mechau N, Kruk R, Hahn H (2010)

J Appl Phys 108:103721

12. Makino T, Segawa Y, Kawasaki M, Ohtomo A, Shiroki R,

Tamura K, Yasuda T, Koinuma H (2001) Appl Phys Lett 78:1237

13. Ye ZZ, Ma DW, He JH, Huang JY, Zhao BH, Luo XD, Xu ZY

(2003) J Cryst Growth 256:78

14. Ramakrishna Reddy KT, Prathap P, Revathi N, Reddy ASN,

Miles RW (2009) Thin Solid Films 518:1275

15. Matsubara K, Tampo H, Shibata H, Yamada A, Fons P, Iwata K,

Niki S (2004) Appl Phys Lett 85:1374

16. Ghosh R, Basak D (2009) Appl Surf Sci 255:7238

17. Yang C, Li XM, Gao XD, Cao X, Yang R, Li YZ (2011) Solid

State Commun 151:264

18. Ghosh M, Dilawar N, Bandyopadhyay AK, Raychaudhuri AK

(2009) J Appl Phys 106:084306

19. Cohen DC, Ruthe KC, Barnett SA (2004) J Appl Phys 96:459

20. Zhu MW, Gong J, Sun C, Xia JH, Jiang X (2008) J Appl Phys

104:073113

J Sol-Gel Sci Technol (2012) 62:344–350 349

123

21. Meher SR, Biju KP, Jain MK (2009) J Sol–Gel Sci Technol

52:228

22. Tong H, Deng ZH, Liu ZG, Huang CG, Huang JQ, Lan H, Wang

C, Cao YG (2011) Appl Surf Sci 257:4906

23. Major S, Banerjee A, Chopra KL (1984) Thin Solid Films 122:31

24. Sharma M, Mehra RM (2008) Appl Surf Sci 255:2527

25. Li G, Zhu XB, Tang XW, Song W, Yang ZR, Dai JM (2011) J

Alloys Comp 509:4816

26. Basu PK (1997) Theory of optical process in semiconductors.

Clarendon, Oxford

27. Hamberg I, Granqvist CG (1986) J Appl Phys 60:R123

28. Kerkache L, Layadi A, Mosser A (2009) J Alloys Comp 485:46

29. Ahn MH, Cho ES, Kwon SJ (2011) Appl Surf Sci 258:1242

30. Burstein E (1954) Phys Rev 93:632

31. Moss TS (1954) Proc Phys Soc Lond Sect B 67:775

32. Hamberg I, Granqvist CG, Berggren KF, Sernelius BE, Engstrom

L (1984) Phys Rev B 30:3240

33. Bruneaux J, Cachet H, Froment M, Messad A (1991) Thin Solid

Films 197:129

34. Liu EK, Zhu BS, Luo JS (1997) Semiconductor physics. National

Defence Industry Press, Beijing

35. Lu JG, Fujita S, Kawaharamura T, Nishinaka H, Kamada Y,

Ohshima T, Ye ZZ, Zeng YJ, Zhang YZ, Zhu LP, He HP, Zhao

BH (2007) J Appl Phys 101:083705

36. Wei W, Jin CM, Narayan J, Narayan RJ (2009) Solid State

Commun 149:1670

350 J Sol-Gel Sci Technol (2012) 62:344–350

123