Modification of structure and properties of AZO thin film by introducing H2 in sputtering atmosphere at low substrate temperature

Superlattices and Microstructures 64 (2013) 460–469

Contents lists available at ScienceDirect

Superlattices and Microstructures journal homepage: www.elsevier.com/locate/superlattices

Modification of structure and properties of AZO thin film by introducing H2 in sputtering atmosphere at low substrate temperature B.L. Zhu a,⇑, K. Li a, J. Wang a, J. Wu a, D.W. Zeng b, C.S. Xie b a Key Laboratory for Ferrous Metallurgy and Resources Utilization of Ministry of Education, Wuhan University of Science and Technology, Wuhan 430081, People’s Republic of China b Department of Materials Science and Engineering, Huazhong University of Science and Technology, Wuhan 430074, People’s Republic of China

a r t i c l e

i n f o

Article history: Received 27 August 2013 Accepted 8 October 2013 Available online 15 October 2013 Keywords: Al-doped ZnO Hydrogen doing Sputtering Resistivity Scattering Transmittance

a b s t r a c t Al-doped ZnO (AZO) thin films were prepared on soda-lime glass at 100 °C by RF magnetron sputtering with different H2 fluxes. The influences of H2 flux on structural, electrical, and optical properties were investigated by XRD, Hall Effect measurement, and transmittance spectra. The results show that hydrogen introduction significantly modifies both structure and properties of AZO films. As H2 flux increases, the increase of unit-cell volume of the films implies that hydrogen is incorporated into ZnO lattice; the obvious decrease of crystallite size indicates that the crystallinity of the films degrades. The resistivity of the films can be continuously decreased with increasing H2 flux, accompanying with increase of both carrier concentration and Hall mobility. The main factor of increasing carrier concentration and mobility is found to be related to hydrogen incorporation and effective substitution of Zn2+ sites by Al3+. The films deposited in Ar + H2 atmosphere show improved conductive stability in air due to the passivation of intercrystallite by hydrogen. The average transmittance in visible range of the films is hardly dependent on H2 flux. The Eg of the films increases with increasing H2 flux, and the blueshift values are close to the theoretical one according to the nonparabolic BM effect. Ó 2013 Elsevier Ltd. All rights reserved.

⇑ Corresponding author. Tel.: +86 27 68862718. E-mail address: [email protected] (B.L. Zhu). 0749-6036/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.spmi.2013.10.017

B.L. Zhu et al. / Superlattices and Microstructures 64 (2013) 460–469

461

1. Introduction Transparent conducting oxides (TCO) thin films are widely used in flat panel displays, solar cells, touch screens, electrochromic windows, and low-emissivity windows [1,2]. Currently, indium tin oxide (ITO) is the most popular and practical used TCO materials in above applications due to its low resistivity and high transparency [2,3]. In recent years, Al-doped ZnO (AZO) thin films exhibit considerable electrical and optical properties with low material cost, non-toxicity and stability in hydrogen plasma, and thus they are regarded as an alternative material to ITO [3,4]. Many preparation and treatment methods have been carried out to obtain high performance AZO thin films, such as pulsed laser deposition [5], spray pyrolysis [6], magnetron sputtering [7], and post-annealing [8]. However, many studies indicate that high performance AZO thin films are obtained at relatively high substrate temperatures and/or postannealing temperatures (300–400 °C) [9–11], which are higher than required temperatures for above mentioned applications (<200 °C) [3]. As a result, it is necessary to improve the properties of AZO thin films at low temperature to be compatible with the fabrication of photoelectric device. In 2000, Van de Walle [12] had predicted that hydrogen occurs exclusively in the positive charge state and acts as a shallow donor in ZnO materials by density functional theory calculations. Experimentally, it has been confirmed that the hydrogen can be intentionally introduced into AZO thin films by H-plasma treatment [13], adding H2 in growth atmosphere [14], and post-annealing in H2 [15]. When H2 is introduced into deposition atmosphere during radio frequency (RF) magnetron sputtering, which is a widely used method to prepare AZO films, hydrogen can be doped into ZnO lattice by hydrogen radicals in plasma to improve the electrical conductivity of the prepared films at low substrate temperatures. However, effect of introduction of H2 on the structure and properties of magnetron sputtered AZO films is not understood completely. In this paper, AZO thin films were prepared by RF magnetron sputtering under various H2 fluxes in sputtering atmosphere at substrate temperature of 100 °C. Effects of hydrogen incorporation on the structural, electrical, and optical properties of AZO films were investigated. The aim is to obtain high performance AZO thin films and explain the behavior of hydrogen in AZO films. 2. Experimental procedures AZO thin films were prepared on soda-lime glass in a conventional RF magnetron sputtering system with a sintered AZO ceramic target (99.9% in purity, 2 wt.% Al2O3, and 60 mm in diameter) purchased from Wuhan Sinotarget Materials Co., Ltd. Before deposition, the glass substrates were cleaned in alcohol, acetone and distilled water by ultrasonic for 15 min successively. The vacuum chamber was evacuated to 1.0  103 Pa, after that H2 and Ar were introduced into chamber as sputtering gas. To investigate the influence of H2 flux on the structure and properties of AZO thin film, the H2 flux was varied in the range of 0–12 standard-state cubic centimeter per minute (sccm) and the Ar flux was fixed at 95 sccm by using mass flow controllers. The substrate temperature, sputtering power, sputtering pressure, and substrate-target distance were fixed at 100 °C, 150 W, 0.8 Pa, and 65 mm, respectively. To exclude the influence of thickness, all films had a thickness of about 515 nm by varying the deposition time. Interference microscope (SC57-6JA) was used to determine the thickness of AZO thin films. The crystalline structure of the films was identified by an X-ray diffraction (XRD; Bruker AXS D-8 Advance Model) using Cu Ka radiation. The electrical properties of the films were determined by Hall effects measurement using Van der Pauw method at room temperature. The transmittance spectra of the films were obtained from UV–visible spectrophotometer (UV-2102PC) in the wavelength range of 300–1000 nm. 3. Experimental results and discussion 3.1. Structural characterization Fig. 1 reveals the XRD patterns of AZO thin films deposited at various H2 fluxes. It is observed that AZO thin film deposited in pure Ar atmosphere has a relatively strong (1 0 0) diffraction peak and a

462

B.L. Zhu et al. / Superlattices and Microstructures 64 (2013) 460–469

(100)

weak (0 0 2) one. After introducing H2 into deposition atmosphere, the (1 0 0) diffraction peak is first enhanced then gradually weakened with increasing H2 flux. The weak (0 0 2) diffraction peak can be obviously observed when the H2 flux increases from 0 to 6 sccm, but it is hardly observed as H2 flux further increases to 8–12 sccm. As a whole, the relative intensity of (1 0 0) diffraction peak to (0 0 2) one is enhanced after introducing H2 into deposition atmosphere. Strong (1 0 0) peak suggests that the AZO thin films have the polycrystalline structure with a preferred orientation along a-axis perpendicular to the substrate surface. The appearance of (1 0 0) diffraction peak in AZO films deposited in pure Ar atmosphere can be ascribed to low substrate temperature and high deposition rate [16,17], in which atoms have no sufficient energy and the surface mobility to settle in stable position. After introducing H2 into deposition atmosphere, the relative intensity of (1 0 0) diffraction peak to (0 0 2) one is enhanced, as described above. Previous studies also reported that (0 0 2) peaks was weakened or vanished and (1 0 0), (1 0 1), and (1 1 0) peaks were enhanced or created simultaneously as H2 flux increased [18–20]. It is thus speculated that hydrogen may stabilize (1 0 0) such an unstable configuration although the (1 0 0) plane is thought less stable [20]. The angles of (1 0 0) and (0 0 2) diffraction peaks as a function of H2 flux are plotted in Fig. 2. It is observed that the angle of (1 0 0) diffraction peak [2h(100)] decreases from about 31.70–31.40° as H2 flux increases from 0 to 12 sccm. The angle of (0 0 2) diffraction peak [2h(002)] first decreases from about 34.5 to 34.1° as H2 flux increases from 0 to 0.5 sccm, and then it gradually increases to 34.4° as H2 flux further increases to 6 sccm. The lattice constants along a- and c-axis (a and c) of the AZO thin films were evaluated from the angles of (1 0 0) and (0 0 2) peaks, respectively. The change of lattice constant is opposite to that of the angle of the diffraction peak with H2 flux, that is a shows increased trend, but c first increases then decreases with H2 flux. It is noted that almost all a and c values of the films are larger than these of ZnO powder (0.3250 and 0.5207 nm), as shown in Fig. 2. According to lattice constants a and c, the unit-cell volumes of the films are calculated and are shown in Fig. 2. As H2 flux increasing from 0 to 6 sccm, it is found that the unit-cell volume keeps the increased trend although c does not linearly increase. Considering that H2 is introduced into deposition atmosphere in present study, the expansion of unit-cell volume could be related to the incorporated hydrogen. Most researchers thought that the incorporated hydrogen could be located at anti-bonding oxygen sites (ABO) or bond centered sites of O–Zn bonds (BC) with parallel or not parallel to the c-axis, which should be responsible for the increase of the lattice constant and expansion of the unit-cell volume [12,14,21]. Fig. 3 shows the full width at half maximum (FWHM) of (1 0 0) and (0 0 2) diffraction peaks as a function of H2 flux. The change of FWHM of (1 0 0) diffraction peak with H2 flux could be roughly regarded as two sections: about 0.45° with H2 flux from 0 to 2 sccm and about 0.75° with H2 flux from 4

H2 flux: 12 sccm

Intensity (arb. units)

(002)

H2 flux: 10 sccm H2 flux: 8 sccm H2 flux: 6 sccm H2 flux: 4 sccm H2 flux: 2 sccm H2 flux: 1.5 sccm H2 flux: 1 sccm H2 flux: 0.5 sccm H2 flux: 0 sccm

25

30

35

40

45

2θ (o) Fig. 1. XRD patterns of AZO thin films deposited with different H2 fluxes at 100 °C.

463

31.60 31.50 31.40 0.3250 nm

34.80 34.60

0.5207 nm

34.40 34.20

34.00-1 1.47x10

c (nm)

0.330 0.329 0.328 0.327 0.326 0.325 0.530 0.525 0.520 0.515 0.510

31.70

a (nm)

Unit-cell volume (nm3) o 2θ(002) (o) 2θ(100) ( )

B.L. Zhu et al. / Superlattices and Microstructures 64 (2013) 460–469

-1

1.44x10

-1

1.41x10

0

1

2

3

4

5

6

7

8

9

10 11 12

H2 flux (sccm)

FWHM of (002) peak (o)

0.90

22 20 18 16 14 12 10

0.75 0.60 0.45 0.30 1.65

10

1.50

8

1.35

6

1.20

4

1.05

2 0

0.90 0

1

2

3

4

5

6

7

8

9

10 11 12

Crystallite size Crystallite size along a-axis (nm) along c-axis (nm)

FWHM of (100) peak (o)

Fig. 2. The position of (1 0 0) and (0 0 2) diffraction peaks, lattice constant a and c as well as the unit-cell volume of AZO films as a function of H2 flux.

H2 flux (sccm) Fig. 3. The FWHM of (1 0 0) and (0 0 2) diffraction peaks as well as crystallite size along a- and c-axis of AZO films as a function of H2 flux.

to 12 sccm. The FWHM of (0 0 2) first increases from about 1.2° to 1.6° as H2 flux from 0 to 1.5 sccm, and then it decreases to about 1.1° as H2 flux further increasing to 6 sccm. According to Scherrer equation, the crystallite sizes along a- and c-axis can be evaluated, and they are shown in Fig. 3. It is found that the changed trend of crystallite size with H2 flux is opposite to that of FWHM with H2 flux. The crystallite size along a-axis is about 18 and 10 nm for H2 flux range of 0–2 sccm and 4–12 sccm, respectively. For the crystallite size along c-axis, it first decreases from about 6.8 to 5.3 nm as H2 flux from 0 to 1.5 sccm, and then it increases to about 7.2 nm as H2 flux further increasing to 6 sccm. Obviously, the crystallite size along c-axis should be much small due to fact that (0 0 2) diffraction peak is hardly observed for AZO films deposited at H2 flux of 8–12 sccm. As H2 flux increases from 0 to 0.5 sccm, it is worth noting that crystallite sizes along a- and c-axis of the films increase and thus the crystallinity of AZO films should be slightly improved. The general changed trend of crystallite size (or FWHM) with H2 flux indicates that the crystallinity of AZO films is greatly degraded with increasing H2 flux to 4 sccm. Usually, it is thought that weak bound and loose adatom could be removed by the hydrogen addition and thus the crystallinity of AZO thin films is enhanced with increasing H2 flux [14,21–24]. Nevertheless, excess addition of hydrogen maybe lead to loose of structure of resulting films [23] or adsorption of H2 (both chemically and physically) on the growing surface to hinder further growth of the ZnO crystallites [24], and thus the crystallinity of AZO film is deteriorated.

464

B.L. Zhu et al. / Superlattices and Microstructures 64 (2013) 460–469

3.2. Electrical properties The results of Hall measurements revealed that deposited AZO thin films exhibited n-type conduction. The H2 flux dependence of carrier concentration (n), Hall mobility (l) and resistivity (q) are illustrated in Fig. 4. It is found that AZO thin film deposited in pure Ar exhibits poor conductive properties with carrier concentration of 1.60  1019 cm3, Hall mobility of 1.63 cm2/(V s) and resistivity of 2.4  101 X cm. As H2 flux increases to 2 sccm, the carrier concentration and mobility of the films steeply increase to 8.25  1019 cm3 and 5.64 cm2/(V s), respectively, and therefore the resistivity decreases to 1.34  102 X cm. With further increasing H2 flux to 12 sccm, the carrier concentration and mobility slowly increase to 2.04  1020 cm3 and 7.57 cm2/(V s), respectively, and thus the resistivity decreases to 4.05  103 X cm. The changed trend of carrier concentration, Hall mobility and resistivity with H2 flux is similar to the previous reports [21,23,24]. However, the obtained minimum resistivity and corresponding H2 flux are comparatively different due to the different deposition parameters. As discussed above, hydrogen can be incorporated into ZnO lattice to act as shallow donor [12,18], which should result in the increase of the carrier concentration in AZO thin films after introducing H2 into deposition atmosphere. Furthermore, more amount of hydrogen are incorporated into AZO with increasing H2 flux, thus the carrier concentration shows increased trend with H2 flux. The Hall mobility is related to different scattering mechanisms, mainly including inter-crystallite scattering, neutral impurity scattering, ionized impurity scattering, and lattice vibration scattering [25–27]. The lattice vibration scattering could be negligible at room temperature. The inter-crystallite scattering is related to mean free path of carrier (L), which can be calculated by the following equation [28,29]:

L¼

 2 1=3 3p h 2pe2 qn2=3

ð1Þ

where h and e are the Planck constant and the electron charge, respectively; q and n are the resistivity and carrier concentration of films, respectively. The calculated L (0.08–0.91 nm) is far smaller than the crystallite size of the films from XRD analysis, indicating that the inter-crystallite scattering could be neglected [28,29]. Thus the mobility of the films (l) can be obtained by the following equation:

1

l

¼

1

lI

þ

1

ð2Þ

lN

where lI and lN are mobilities due to the ionized impurity scattering and neutral impurity scattering, respectively. The lI could be given by the following equations [26,30–32]: 3

lI ¼

3ðer e0 Þ2 h n 1 np Z 2 m2 e3 nI F ii ðnd Þ

ð3Þ

100

1021

10 8

1020

ρ -3

n

10

10-4

1019 2

4

6

4 2

μ 0

6

μ [cm2/(V.s)]

10-2

n (cm-3)

ρ (Ω. Ω.cm)

10-1

8

10

0

12

H2 flux (sccm) Fig. 4. The carrier concentration (n), Hall mobility (l), and resistivity (q) of AZO films as a function of H2 flux.

465

B.L. Zhu et al. / Superlattices and Microstructures 64 (2013) 460–469

F np ii



    4nnp nnp 5nnp nd lnð1 þ nd Þ  ¼ 1þ 1  2nnp 1  nd 8 1 þ nd 16 

nd ¼

3p 2

1=3

ð4Þ

er e0 h2 n1=3 ð5Þ

m e2 nnp ¼ 1  m0 =m " 

m ¼

m0

#1=2 2  2=3 h 3 2=3 1 þ 2anp n 4m0 p

ð6Þ

In above equations, er is the relative permittivity and is taken as 8.3 for ZnO, and e0 is the vacuum permittivity; n and nI are carrier concentration and ionized impurity concentration, respectively; Z is *  the charge state of the ionized impurity; F np ii is the screening function; m0 and m are the electron effective masses at the conduction band edge and in the conduction band, respectively, and m0 is taken as 0.28me for ZnO; anp is non-parabolicity parameter, and it is taken as 1.04 eV1. In the calculation, Z is taken as 1 considering that the origin of carrier is Al replacing Zn and interstitial hydrogen, and thus nI is equal to n. Once the lI is determined, the lN can be obtained from Eq. (2). Furthermore, the neutral impurity concentration (nN) can be determined by the following equation [28,33,34]:

nN ¼

2p3 m e3 5e0 er h

3

ð7Þ

lN

In Fig. 5, lI, lN, nI, and nN are shown as a function of H2 flux. It is found that the lI decreases but the lN increases with increasing H2 flux. Correspondingly, the nI increases but the nN decreases with H2 flux. In AZO films, the main origin of the neutral impurity may be Al2O3 due to the fact that enthalpy of formation for Al2O3 is much smaller than that for ZnO and bonding strength of Al–O is stronger than that of Zn–O [35]. Thus, the mobility of the films increases with H2 flux can be attributed to the decrease of Al2O3 amount in the films. On the other hand, AZO films show low carrier concentration at low substrate temperature is related to that Al cannot effectively replace Zn. With increasing H2 flux, the decrease of Al2O3 amount implies that the active hydrogen species could improve the effective substitution of Al atoms for Zn atoms, as suggested by some researchers [23,36,37]. The substitutional Al atoms also contribute the increase of the carrier concentration with H2 flux in addition to the incorporation of hydrogen. Fig. 6 shows the carrier concentration, Hall mobility, and resistivity as a function of exposure time in air for AZO films deposited at different H2 fluxes. The inserted numerical values are ratio of the carrier concentration, Hall mobility or resistivity after exposure in air for 120 days to initial one. Basically, Hall mobility decreases, but the carrier concentration slightly decreases or almost keep constant with

9 7 5 4 3 2 1 0

5.0x10

21

4.5x10 20

2.0x10

nI nN

20

1.5x10

20

1.0x10

21

4.0x10

21

3.5x10

21

3.0x10

21

2.5x10

nN (cm-3)

6

μI μN

nI (cm-3)

μ [cm2/(V.s)] N

8

21

20

2.5x10

240 220 200 180 160 140 120 100 80 60 40 20

μI [cm2/(V.s)]

10

21

19

5.0x10

2.0x10

21

1.5x10 0.0 0

2

4

6

8

10

21

1.0x10

12

H2 flux (sccm) Fig. 5. The mobility resulting from ionized (lI) and neutral (lN) impurity scatterings as well as the concentration of ionized (nI) and neutral (nN) impurities as a function of H2 flux.

466

B.L. Zhu et al. / Superlattices and Microstructures 64 (2013) 460–469

increasing exposure time in air. Obviously, change of Hall mobility is more remarkable than that of carrier concentration with exposure time in air. Thus, the increase of the resistivity can be mainly attributed to the decrease of the Hall mobility, as observed by other researchers [38]. On the other hand, the decreased degree of carrier concentration and Hall mobility for AZO films deposited in Ar is obviously larger that for AZO films deposited in Ar + H2, and thus the resistivity of the films deposited in Ar is evidently increased after exposure in air. Usually, the crystallinity of deposited films is poor and more defects exist in the inter-crystallite boundary when substrate temperature is lower, which results in absorption of oxygen in air easily. The absorbed oxygen not only decreases the carrier concentration by trapping free carriers, but also decreases the Hall mobility by increasing barrier height between crystallites, thus the resistivity increases with exposure time in the air. After introducing H2 into sputtering atmosphere, inter-crystallite boundary is passivated by hydrogen [14,30,38,39], which keeps the films from the adsorption of oxygen species. Therefore, conductive stability of AZO films in air is improved. 3.3. Optical properties Fig. 7 shows the optical transmittance spectra of AZO thin film deposited at H2 flux of 0–12 sccm. The fluctuation of spectra in the visible range is caused by the light interference on the interfaces of film-air and film-substrate, which indicates that film thickness is uniform. All films exhibit a high average transmittance (>85%) in the visible regions (400–800 nm), as shown in Fig. 8. It seems that the influence of H2 flux on the average transmittance of the films is not obvious. As discussed above, the crystallite size of the films decreases and its crystallinity degrades with increasing H2 flux to 4 sccm, which should decrease film transmittance [40]. However, it is thought that the introduction of H2 results in the passivation of the crystallite boundary and intrinsic defect of the films [18,21,41]. This is beneficial to the decrease of light scattering and increase of the film transmittance. Thus, the higher transmittance is obtained at all H2 fluxes. The inset of Fig. 7 shows the absorption edge region of transmittance spectra, and it can be clearly observed that the absorption edge shifts to the shorter wavelength with increasing H2 flux. The value of the optical energy gap (Eg) can be determined from (a h m)2 / (hm  Eg) relationship in which a is the absorption coefficient and hm is the photon energy. The absorption coefficient (a) can be calculated from the formula [42]:

n (cm-3)

1.02

1020

1.02 1.01 1.01 0.93

μ [cm2/(V.s)]

1019 10

0 sccm 1 sccm 12 sccm

8

0.5 sccm 6 sccm

0.91 0.94

6 4

0.92

2

0.75 0.76

100

ρ (Ω.cm)

1.41 1.32

10-1

1.03 1.04 1.07

10-2 10-3 0

30

60

90

120

Exposure time in air (day) Fig. 6. The carrier concentration (n), Hall mobility (l), and resistivity (q) as a function of exposure time in air for AZO films deposited at different H2 fluxes. The inserted numerical values are the ratio of the carrier concentration, Hall mobility, or resistivity after exposure in air for 120 days to the initial one.

B.L. Zhu et al. / Superlattices and Microstructures 64 (2013) 460–469

467

100

80

60

40

20

80 60

T (%)

T (%)

0 sccm 0.5 sccm 1 sccm 1.5 sccm 2 sccm 4 sccm 6 sccm 8 sccm 10 sccm 12 sccm

0 sccm 0.5 sccm 1 sccm 1.5 sccm 2 sccm 4 sccm 6 sccm 8 sccm 10 sccm 12 sccm

40 20 0 340

360

380

400

420

440

Wavelength (nm)

0 300

400

500

600

700

800

900

1000

Wavelength (nm) Fig. 7. Transmittance spectra of AZO thin film deposited at different H2 fluxes in the wavelength of 300–1000 nm. The insert shows the absorption edge region.

100

3.8 3.7 3.6

80 3.5 70

Eg (eV)

Average T (%)

90

3.4 60

3.3

50

3.2 0

2

4

6

8

10

12

H2 flux (sccm) Fig. 8. The average transmittance in visible range and optical band gap (Eg) of AZO films as a function of H2 flux.

T ¼ A expðatÞ

ð8Þ

where A is a constant and taken as 1; T and t are the transmittance and thickness of the thin films, respectively. In Fig. 8, the Eg is shown as a function of H2 flux, indicating that the Eg increases from about 3.34 to 3.60 eV as H2 flux increases from 0 to 12 sccm. This changed trend is consistent with that of carrier concentration with H2 flux, indicating that increase in Eg is due to Burstein–Moss (BM) effect [43]. In fact, the Fermi level moves into the conduction band or rises in the conduction band with increasing carrier concentration, and the filling of the conduction band by electrons will generally result in the valence electrons requiring extra energy to be excited by photons to higher-energy states in the conduction band. Hence, the Eg increases with carrier concentration. The broadening of Eg due to BM effect (DEBM) can be expressed by the following equation [15,25,27]:

DEBM ¼

2  2=3 h 3 n2=3 8m0 p

ð9Þ

The obtained DEBM as a function of carrier concentration is shown in Fig. 9, and is compared with experimental value (Eg  Eg0, in which Eg0 is the fundamental energy gap of bulk ZnO semiconductor and is taken as 3.28 eV [44]). It can be seen clearly that DEBM are far larger than the experimental values, especially at higher carrier concentration, because this calculation is based on the assumption of a

468

B.L. Zhu et al. / Superlattices and Microstructures 64 (2013) 460–469 0.5

BM shift Nonparabolic BM shift Expermental data

0.4

ΔΕ (eV)

Films deposited at H2 flux of 10-12 sccm

0.3

0.2

Films deposited at H2 flux of 0-8 sccm

0.1

0.0 0

1x1020

2x1020

3x1020

4x1020

-3

Carrier concentration (cm ) Fig. 9. The dependence of Eg shift of AZO films on carrier concentration, and the comparisons of experimental values and theoretical results according to BM effect and the nonparabolic BM effect.

single-parabolic conduction band with a fixed value of electron effective mass ðm0 Þ. Considering that the bottom of conduction band is filled by electrons for degenerated semiconductor and thus many electrons distribute away from the bottom of conduction band which shows nonparabolic nature, electron effective mass is varied with carrier concentration, that is the electron effective mass is obtained from Eq. (6). By replacing the m0 in Eq. (9) with m*, the broadening of Eg due to nonparabolic BM effect (DEnp-BM) is obtained, as shown in Fig. 9. It is found that DEnp-BM are close to the experimental values. It is also worth noting that DEnp-BM are slightly larger than the experiment values for the films deposited at H2 flux of 0–8 sccm, which can be explained by the many-body interaction effects [25,27]. However, the experiment values are found to be higher than DEnp-BM for the films deposited at H2 flux of 10–12 sccm. This may be related to much small crystallite size of the films deposited at H2 flux of 10–12 sccm. According to previous report, small-sized crystallite also results in the additional broadening of Eg of the films [45]. 4. Conclusions The structure and properties of AZO thin films can be modified by introducing H2 into sputtering atmosphere at substrate temperature of 100 °C. With increasing H2 flux, the unit-cell volume of AZO films increases and crystallinity of the film is degraded. The carrier concentration and mobility keep increased trend and thus resistivity continuously decreases to 4.05  103 X cm with increasing H2 flux. The optical transmittance in the visible range of all the film is above 85% and it is litter dependent on H2 flux. The hydrogen is incorporated into films to acts as shallow donor and also improves effective substitution of the Zn2+ sites by Al3+, which should be the main source of improvement of conductive properties of AZO films at low substrate temperature. The passivation effect by hydrogen is thought to contribute high conductive stability and transmittance for AZO films. The Eg of all films is found to be roughly described by nonparabolic BM effect although other effects also appear. Acknowledgments This work was supported by the National Nature Science Foundation of China (Grant No. 50902105), the National Basic Research Program of China (Grant Nos. 2009CB939702 and 2009CB939705), the Opening Project of State Key Laboratory of High Performance Ceramics and Superfine Microstructure (Grant No. SKL201110SIC), and the Key Laboratory of Optoelectronic Devices and Systems of Ministry of Education and Guangdong Province (Grant No. 201208).

B.L. Zhu et al. / Superlattices and Microstructures 64 (2013) 460–469

469

References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45]

R.G. Gordon, MRS Bull. 25 (2000) 52–57. E. Fortunato, D. Ginley, H. Hosono, D.C. Paine, MRS Bull. 32 (2007) 242–247. T. Minami, T. Miyata, Thin Solid Films 517 (2008) 1474–1477. J. Lee, D. Lim, K. Yang, W. Choi, J. Cryst. Growth 326 (2011) 50–57. H. Agura, A. Suzuki, T. Matsushita, T. Aoki, M. Okuda, Thin Solid Films 445 (2003) 263–267. B. Joseph, P.K. Manoj, V.K. Vaidyan, Ceram. Int. 32 (2006) 487–493. S. Cornelius, M. Vinnichenko, N. Shevchenko, A. Rogozin, A. Kolitsch, W. Moller, Appl. Phys. Lett. 94 (2009) 42103–42105. G.J. Fang, D.J. Li, B.L. Yao, Thin Solid Films 418 (2002) 156–162. T. Minami, H. Sato, H. Imamoto, S. Takata, Jpn. J. Appl. Phys. 31 (1992) L257–L260. J. Hong, H. Paik, H. Hwang, S. Lee, A.J. deMello, K. No, Phys. Status Solidi A 206 (2009) 697–703. R.T. Wen, L.S. Wang, X. Wang, G.H. Yue, Y.Z. Chen, D.L. Peng, J. Alloys Compd. 508 (2010) 370–374. C.G. Van de Walle, Phys. Rev. Lett. 85 (2000) 1012–1015. H.P. Chang, F.H. Wang, J.Y. Wu, C.Y. Kung, H.W. Liu, Thin Solid Films 518 (2010) 7445–7449. S.J. Tark, Y.W. Ok, M.G. Kang, H.J. Lim, W.M. Kim, D. Kim, J. Electroceram. 23 (2008) 548–553. B.Y. Oh, M.C. Jeong, D.S. Kim, W. Lee, J.M. Myoung, J. Cryst. Growth 281 (2005) 475–480. B.L. Zhu, S.J. Zhu, J. Wang, J. Wu, D.W. Zeng, C.S. Xie, Physica E 43 (2011) 1738–1745. V. Tvarozek, I. Novotny, P. Sutta, S. Flickyngerova, K. Schtereva, E. Vavrinsky, Thin Solid Films 515 (2007) 8756–8760. B.L. Zhu, J. Wang, S.J. Zhu, J. Wu, R. Wu, D.W. Zeng, C.S. Xie, Thin Solid Films 519 (2011) 3809–3815. Y.R. Park, Y.S. Kim, J. Mater. Res. 23 (2008) 1674–1681. Y.S. Kang, H.Y. Kim, J.Y. Lee, J. Electrochem. Soc. 147 (2000) 4625–4629. Y.R. Park, J. Kim, Y.S. Kim, Appl. Surf. Sci. 255 (2009) 9010–9014. W.F. Liu, G.T. Du, Y.F. Sun, J.M. Bian, Y. Cheng, T.P. Yang, Y.C. Chang, Y.B. Xu, Appl. Surf. Sci. 253 (2007) 2999–3003. X.T. Hao, F.R. Zhu, K.S. Ong, L.W. Tan, Semicond. Sci. Technol. 21 (2006) 48–54. S.H. Lee, T.S. Lee, K.S. Lee, B. Cheong, Y.D. Kim, W.M. Kim, J. Phys. D: Appl. Phys. 41 (2008) 95303–95309. J.G. Lu, Z.Z. Ye, Y.J. Zeng, L.P. Zhu, L. Wang, J. Yuan, B.H. Zhao, Q.L. Liang, J. Appl. Phys. 100 (2006) 73714–73724. K. Ellmer, R. Mientus, Thin Solid Films 516 (2008) 4620–4627. T. Han, F.Y. Meng, S. Zhang, X.M. Cheng, J.I. Oh, J. Appl. Phys. 110 (2011) 63724–63729. Y. Shigesato, D.C. Paine, Appl. Phys. Lett. 62 (1993) 1268–1270. T. Yamamoto, T. Sakemi, K. Awai, S. Shirakata, Thin Solid Films 451–452 (2004) 439–442. W.M. Kim, Y.H. Kim, J.S. Kim, J. Jeong, Y.J. Baik, J.K. Park, K.S. Lee, T.Y. Seong, J. Phys. D: Appl. Phys. 43 (2010) 365406– 365410. K. Ellmer, J. Phys. D: Appl. Phys. 34 (2001) 3097–3108. A.V. Singh, R.M. Mehra, A. Yoshida, A. Wakahara, J. Appl. Phys. 95 (2004) 3640–3643. R.B.H. Tahar, N.B.H. Tahar, J. Appl. Phys. 92 (2002) 4498–4501. Z.L. Pei, C. Sun, M.H. Tan, J.Q. Xiao, D.H. Guan, R.F. Huang, L.S. Wen, J. Appl. Phys. 90 (2001) 3432–3436. K.K. Kim, S. Niki, J.Y. Oh, J.O. Song, T.Y. Seong, S.J. Park, S. Fujita, S.W. Kim, J. Appl. Phys. 97 (2005) 66103–66105. M.L. Addonizio, A. Antonaia, G. Cantele, C. Privato, Thin Solid Films 349 (1999) 93–99. Y.F. Sun, W.F. Liu, Z.D. He, S.L. Liu, Vacuum 80 (2006) 981–985. B.Y. Oh, M.C. Jeong, J.M. Myoung, Appl. Surf. Sci. 253 (2007) 7157–7161. S.Y. Myong, K.S. Lim, Appl. Phys. Lett. 82 (2003) 3026–3028. J.M. Kim, P. Thiyagarajan, S.W. Rhee, Thin Solid Films 518 (2010) 5860–5865. P.F. Cai, J.B. You, X.W. Zhang, J.J. Dong, X.L. Yang, Z.G. Yin, N.F. Chen, J. Appl. Phys. 105 (2009) 83713–83718. J.K. Lee, H.M. Kim, S.H. Park, J.J. Kim, B.R. Rhee, J. Appl. Phys. 92 (2002) 5761–5765. E. Burstein, Phys. Rev. 93 (1954) 632–633. J.G. Lu, S. Fujita, T. Kawaharamura, H. Nishinaka, Y. Kamada, T. Ohshima, Z.Z. Ye, Y.J. Zeng, Y.Z. Zhang, L.P. Zhu, H.P. He, B.H. Zhao, J. Appl. Phys. 101 (2007) 83705–83711. J.C. Nie, J.Y. Yang, Y. Piao, H. Li, Y. Sun, Q.M. Xue, C.M. Xiong, R.F. Dou, Q.Y. Tu, Appl. Phys. Lett. 93 (2008) 173104–173106.