Cent. Eur. J. Phys. • 6(3) • 2008 • 638-642
DOI: 10.2478/s11534-008-0032-2
Central European Journal of Physics
Microstructural study of MBE-grown ZnO film on
GaN/sapphire (0001) substrate
Research Article
Hua Li12∗ , Jianping Sang2 , Chang Liu2 , Hongbing Lu2 , Juncheng Cao1
1 State Key Laboratory of Functional Materials for Informatics, Shanghai Institute of Microsystem and Information Technology,
Chinese Academy of Sciences, 865 Changning Road, Shanghai 200050, China
2 Key Laboratory of Acoustic and Photonic Materials and Devices of Ministry of Education, Wuhan University, Wuhan 430072,
China
Received 22 October 2007; accepted 31 December 2007
Abstract:
Single crystalline ZnO film is grown on GaN/sapphire (0001) substrate by molecular beam epitaxy. Ga2 O3
is introduced into the ZnO/GaN heterostructure intentionally by oxygen-plasma pre-exposure on the GaN
surface prior to ZnO growth. The crystalline orientation and interfacial microstructure are characterized
by X-ray diffraction and transmission electron microscopy. X-ray diffraction analysis shows strong c-axis
preferred orientation of the ZnO film. Cross-sectional transmission electron microscope images reveal
that an additional phase is formed at the interface of ZnO/GaN. Through a comparison of diffraction patterns, we confirm that the interface layer is monoclinic Ga2 O3 and the main epitaxial relationship should be
(0001)Z nO //(001)Ga2 O3 //(0001)GaN and [2-1-10]Z nO //[010]Ga2 O3 //[2-1-10]GaN .
PACS (2008): 61.72.uj, 75.50.Pp, 78.30.Fs
Keywords:
zinc oxide • molecular beam epitaxy • transmission electron microscopy
© Versita Warsaw and Springer-Verlag Berlin Heidelberg.
1.
Introduction
ZnO has attracted great interest for its excellent optical,
electrical, catalytic, and gas sensing properties, and for
potential applications in various fields [1–6]. The most attractive properties of ZnO are its wide band gap of 3.3
eV at room temperature and high exciton binding energy
(60 meV) compared to those of ZnS (20 meV) and GaN
(25 meV) [7, 8]. Due to its wide band gap, ZnO can be
used for short wavelength light emitting devices such as
∗
E-mail: hli@mail.sim.ac.cn
light emitting diodes (LED) and laser diodes (LD), surface
acoustic wave (SAW) bandpass filters, optical waveguides,
and laser deflectors. Furthermore, experiments have approved that ZnO is one of the most pronounced candidates
for diluted magnetic semiconductors (DMS) [9]. In a word,
ZnO is a useful material for optoelectronics and other applications.
Recently, significant progress in ZnO crystal quality has
been made. High-quality ZnO films have been grown with
molecular beam epitaxy [10, 11], metal-organic chemical
vapor deposition [12–14], pulsed-laser deposition [15, 16],
and sputtering [17, 18]. By reason of the high electron
concentration in as-grown ZnO films, the fabrication of ptype ZnO, which represents the bottleneck of fabricating
638
Unauthenticated
Download Date | 10/1/16 6:50 PM
Hua Li, Jianping Sang, Chang Liu, Hongbing Lu, Juncheng Cao
ZnO-based light-emitting devices, is still unstable due to
the self-compensation effect. However, n-type ZnO and
p-type GaN can be easily realized. Because the lattice
mismatch between ZnO and GaN (1.8% mismatch) is much
smaller than that between ZnO and sapphire (18% mismatch) [19, 20], it is easy to realize the hybrid growth
of ZnO and GaN. Furthermore, the growth and process
techniques of GaN are quite mature. Therefore, the hybrid growth of n-ZnO and p-GaN for optoelectronic devices has been an idea receiving a great attention [21, 22].
To grow high crystalline quality ZnO/GaN films for optoelectronic devices, the microstructural study on ZnO/GaN
interface is an interesting and significant work.
In this article, we present the detailed hybrid growth of
ZnO and GaN on sapphire (0001) substrate by molecular beam epitaxy. Ga2 O3 is intentionally introduced
into the ZnO/GaN heterostructure by oxygen-plasma preexposure on GaN surface prior to ZnO growth. Studies
on the crystalline quality and microstructure of ZnO/GaN
hetero-interface are performed.
2.
Experimental details
The ZnO wafer used in the study had a thickness of 1
µm and was grown on a single crystalline GaN template.
The GaN template was grown on a sapphire (0001) substrate. Both ZnO and GaN layers were deposited by different molecular beam epitaxy (MBE) systems. The GaN
layer was grown on sapphire (0001) substrate by radio
frequency (RF) plasma-assisted MBE. Elemental Ga and
RF plasma-enhanced N2 were used as sources. After thermally cleaning the sapphire substrate surface at 900 o C,
it was nitrided under the N2 flux of 1.0 SCCM at 600
o
C for 1 hour. A 150 nm thick GaN buffer layer was
grown at a substrate temperature of 600o C. Then, a 20
nm thick GaN layer was grown on the GaN buffer layer
at 650o C followed by a thermal treatment at 750o C for
30 min. Next, the sample was transferred to another RF
oxygen plasma-assisted MBE system. RF oxygen plasma
and solid K-cell were used as oxygen and zinc sources,
respectively. Plasma oxygen irradiation at 500o C for 10
min was performed prior to the ZnO film growth to insert
Ga2 O3 at the interface of ZnO/GaN heterostructure intentionally. Next , a 1 µm thick ZnO layer was grown at 600
o
C. A typical beam equivalent pressure 2×10−4 Pa for Zn
and a flow rate 0.5 SCCM for oxygen gas were employed.
Figure 1.
current of 40 mA. Transmission electron microscope (TEM)
measurements were carried out using a JEM-2010FEF
with an accelerating voltage of 200 kV. Microstructural
properties of the ZnO/GaN hetero-interface were characterized using cross-sectional TEM. The specimens for
the cross-sectional TEM measurements were prepared by
cutting, polishing with boron carbide powder to a thickness of approximately 100 µm, then dimpling on one face
down to 20 µm and finally Ar+ ion milling down to electron transparency at a voltage of 4-6 keV with an incident
angle of 6o .
Figure 2.
After growth, crystalline quality and orientation of the
ZnO film were evaluated by D8 Advanced X-ray diffraction
(XRD) which employed a high-resolution diffractometer
with Cu Kα radiation, operating at a voltage of 40 kV and a
X-ray diffraction pattern of the as-grown ZnO film on
GaN/sapphire. The inset is the ZnO (0002) rocking curve.
Bright-field cross-sectional TEM image of the ZnO film on
GaN/sapphire (0001) substrate. The inset is the corresponding SADP with electron beam parallel to the [2-1-10]
zone axis of ZnO or GaN, in which 1 and 2 are GaN (0110) and (0001) reflections and a and b are ZnO (01-10)
and (0001) reflections.
639
Unauthenticated
Download Date | 10/1/16 6:50 PM
Microstructural study of MBE-grown ZnO film on GaN/sapphire (0001) substrate
3.
Results and discussion
Figure 4.
Cross-sectional high-resolution TEM image of ZnO/GaN
interface. The interface layer is indicated by a rectangle.
and [2-1-10]Z nO //[2-1-10]GaN .
Figure 3.
Cross-sectional high-resolution TEM image of ZnO/GaN
interface along [2-1-10] zone axis of ZnO. The rectangular region shows a different lattice structure from ZnO and
GaN.
Fig. 1 shows a typical XRD diffraction pattern of the asgrown ZnO film grown on GaN/sapphire substrate measured with 2θ-θ scan. Pronounced ZnO (0002) diffraction
peak located at 34.4o and ZnO (0004) diffraction peak located at around 73o are observed clearly from it. The ZnO
(0002) rocking curve is shown in the inset of Fig. 1 and the
corresponding full-width at half-maximum (FWHM) value
is 0.22o which indicates that the ZnO film has a strong
c-axis preferred orientation. The (0006) diffraction peak
corresponding to the sapphire (0001) substrates is also
clearly observed.
A bright-field cross-sectional TEM image of the ZnO film
on GaN/sapphire is shown in Fig. 2. It is clear to see
the top ZnO preferentially oriented film, GaN layer, GaN
buffer layer and sapphire substrate. The inset of Fig.
2 is the corresponding selected area diffraction pattern
(SADP) taken from the ZnO/GaN interface, in which 1
and 2 are GaN (01-10) and (0001) reflections, and a and
b are ZnO (01-10) and (0001) reflections. The electron
beam parallel to the [2-1-10] zone axis of ZnO or GaN.
The ZnO thin film is a highly c-axis oriented layer from the
analysis of the SADP. We can conclude the perfect main
epitaxial relationship of the sample: (0001)Z nO //(0001)GaN
A cross-sectional high-resolution TEM study is carried
out to determine the microstructure of ZnO/GaN heterointerface. Fig. 3 shows a cross-sectional high-resolution
TEM image of the heterostructure along the [2-1-10] zone
axis. It is clear to see the interfaces of ZnO/GaN and
GaN/GaN buffer layer, which are indicated by black arrows. Lattice mismatch between ZnO and GaN results
in a high density of misfit dislocation at the interface. It
is worthy to pay more attention to the interfacial area
between ZnO and GaN indicated by a rectangular box
in Fig. 3. Along the interface, the atoms’ positions are
highly displaced, suggesting the existence of a high strain
in the film. We can also see that the interfacial area
shows a different lattice structure from ZnO and GaN, i.e.,
atoms of the first five monolayers in the interface layer
rotate an angle off the [0001]-axis of GaN. In our former
research [23], we conjectured that the first five monolayers
may form another phase (such as monoclinic Ga2 O3 ) or
just be the lattice faults. To clarify this, we take a further
high-resolution TEM study. Fig. 4 shows another crosssectional high-resolution TEM image of the interface of
ZnO/GaN. About 20 monolayers show a different lattice
structure from ZnO and GaN, which are indicated by a
rectangle in Fig. 4. Thanks to the thicker interface layer
(compared to the five monolayers in Fig. 3), we can carry
out a simulated experiment to confirm the phase of the
interface layer as follows.
In order to confirm the phase of the interface layer, we
carry out a diffraction pattern analysis. Fig. 5(a) is a digital diffraction pattern obtained by fast-Fourier-transform
(FFT) of corresponding interface layer of Fig. 4. On the
640
Unauthenticated
Download Date | 10/1/16 6:50 PM
Hua Li, Jianping Sang, Chang Liu, Hongbing Lu, Juncheng Cao
Figure 5.
a) Digital diffraction pattern obtained by fast-Fourier-transform of corresponding interface layer of Fig. 4. (b) Schematic diagram of the
calculated diffraction pattern of the monoclinic Ga2 O3 with the [010]-zone axis.
other hand, we calculated the electron diffraction pattern
based on monoclinic Ga2 O3 (space group is A2/m) structure. Fig. 5(b) shows the simulated diffraction pattern
of the monoclinic Ga2 O3 with the [010]-zone axis. Arabic numerals 1, 2 and 3 in Fig. 5(a) can be proved to
be (001), (201) and (200) reflections of monoclinic Ga2 O3
by comparing the FFT image Fig. 5(a) with the calculated diffraction pattern Fig. 5(b). Thus, we conclude that
the interface layer which shows a different lattice structure should be monoclinic Ga2 O3 . Integrated with our
previous results, we could deduce the main epitaxial relationship of the structure: (0001)Z nO //(001)Ga2 O3 //(0001)GaN
and [2-1-10]Z nO //[010]Ga2 O3 //[2-1-10]GaN . The formation of
monoclinic Ga2 O3 at the interface layer is attributed to
the oxygen-plasma pre-exposure on GaN prior to ZnO
growth and other unintentional oxidations. The insertion
of Ga2 O3 may have negative effects on the luminescence
of devices and result in degradation of the crystal quality of ZnO films compared to ZnO films grown with Znexposure technology [24, 25]. However, monoclinic Ga2 O3
interface layer has a center of symmetry which leads to
the realization of crystal polarity control, i.e. an O-polar
(anion-polar) ZnO film can be grown on Ga-polar (cationpolar) GaN by inserting a Ga2 O3 layer at the interface
[25]. Therefore, this work which demonstrates a closer
look of monoclinic Ga2 O3 is helpful for us to understand
the important role of Ga2 O3 in ZnO/GaN heterostructure.
4.
Conclusions
In summary, 1 µm thick ZnO film is grown on oxygenplasma pre-exposed GaN/sapphire (0001) substrate by
MBE. X-ray diffraction results show a strong c-axis preferred orientation of the ZnO film. Interfaces of different layers could be observed clearly by bright-field
cross-sectional TEM and the main epitaxial relationship of ZnO and GaN is (0001)Z nO //(0001)GaN and [2-110]Z nO //[2-1-10]GaN . Cross-sectional high-resolution TEM
images reveal that the interface layer of ZnO/GaN shows
a different lattice structure from ZnO and GaN. By
comparing the FFT image with the calculated diffraction pattern, we confirm that the interface layer is
monoclinic Ga2 O3 and the main epitaxial relationship
should be (0001)Z nO //(001)Ga2 O3 //(0001)GaN and [2-110]Z nO //[010]Ga2 O3 //[2-1-10]GaN .
Acknowledgments
The authors would like to thank L. J. Zeng from Institute
of Physics, Chinese Academy of Science, Beijing, Dr. Y.
J. Han for stimulating discussions and continuing interest,
F. Ren, F. Mei and L. Zhang for excellent technical assistances. This work was supported by the NSFC for the
grants under Nos. 10345006, 10475063, and by HBSF
under No. 2004ABA079, as well as by NCET and SRF
for ROCS, Ministry of Education.
641
Unauthenticated
Download Date | 10/1/16 6:50 PM
Microstructural study of MBE-grown ZnO film on GaN/sapphire (0001) substrate
References
[1] Z.W. Pang, Z.R. Dai, Z.L. Wang, Science 291, 1947
(2001)
[2] H. Rensmo et al., J. Phys. Chem. B 101, 2598 (1997)
[3] H.J. Muhret et al., Adv. Mater. 12, 231 (2000)
[4] M.H. Huang et al., Science 292, 1897 (2001)
[5] C.H. Liu et al., Appl. Phys. Lett. 83, 3168 (2003)
[6] W. Göpel, J.W. Gardner, J. Hesse, Sensors Applications (Wiely-VCH, Weinheim, 2005)
[7] D.M. Bagnall et al., Appl. Phys. Lett. 70, 2230 (1997)
[8] D.C. Look, D.C. Reynolds, J.W. Hemsky, R.L. Jones,
J.R. Sizelove, Appl. Phys. Lett. 75, 811 (1999)
[9] H. Saeki, H. Tabata, T. Kawai, Solid State Commun.
120, 439 (2001)
[10] K. Nakahara et al., J. Crystal Growth 227-228, 923
(2001)
[11] A. El-Shaer, A. CHe Mofor, A. Bakin, M. Kreye, A.
Waag, Superlattices Microstruct. 38, 265 (2005)
[12] J. Dai et al., J. Crystal Growth 290, 426 (2006)
[13] T. Gruber, C. Kirchner, K. Thonke, R. Sauer, A. Waag,
Phys. Status Solidi A 192, 166 (2002)
[14] B.P. Zhang, K. Wakatsuki, N.T. Binh, N. Usami, Y.
Segawa, Thin Solid Films 449, 12 (2004)
[15] H.L. Porter, A.L. Cai, J.F. Muth, J. Narayan, Appl.
Phys. Lett. 86, 211918 (2005)
[16] F.K. Shan et al., Appl. Phys. Lett. 86, 221910 (2005)
[17] S.F. Chichibu, T. Yoshida, T. Onuma, H. Nakanishi, J.
Appl. Phys. 91, 874 (2002)
[18] D.K. Hwang et al., Appl. Phys. Lett. 86, 151917 (2005)
[19] B.M. Ataev, W.V. Lundin, V.V. Mamedov, A.M. Bagamadova, E.E. Zavarin, J. Phys.–Condens. Matter 13,
L211 (2001)
[20] T. Makino et al., Appl. Phys. Lett. 79, 1282 (2001)
[21] S.J. An et al., Appl. Phys. Lett. 84, 3612 (2004)
[22] Y.I. Alivov, J.E. Van Nostrand, D.C. Look, M.V. Chukichev, B.M. Ataev, Appl. Phys. Lett. 83, 2943 (2003)
[23] H. Li et al., Appl. Surf. Sci. 253, 8524 (2007)
[24] S. Lee, D.Y. Kim, Mat. Sci. Eng. B–Solid 137, 80
(2007)
[25] S.K. Hong et al., Phys. Rev. B 65, 115331 (2002)
642
Unauthenticated
Download Date | 10/1/16 6:50 PM