The crystal structure and quality of ZnO thin films were enhanced by high temperature vacuum annealing. 150 nm thick ZnO films were deposited on a-plane sapphire substrates by RF sputtering at 600°C and then annealed in situ at temperatures from 700°C to 900°C. Higher annealing temperatures produced smoother films with root mean square roughness reaching 0.3 nm at 850°C. Raman spectroscopy showed the A1(TO) mode at all annealing temperatures and the A1(LO) mode appeared above 800°C, indicating improved crystal quality. X-ray diffraction revealed narrower diffraction peaks and a relaxed lattice constant matching bulk ZnO at 900°C annealing, demonstrating high quality c-axis oriented epit

1. Superlattices and Microstructures 73 (2014) 268–274
Contents lists available at ScienceDirect
Superlattices and Microstructures
journal homepage: www.elsevier. com/locate/superlattices
Ultra-smooth and lattice relaxed ZnO thin films
A.N. Fouda a,⇑, El Shazly M. Duraia a,b,⇑, E.A. Eid c
a Physics Department, Faculty of Science, Suez Canal University, 41522 Ismailia, Egypt
b Texas State University-San Marcos, Department of Chemistry and Biochemistry, 601 University Dr., San Marcos, TX 78666, USA
c Department of Basic Science, Higher Technological Institute, 44629 10th of Ramadan City, Egypt
a r t i c l e i n f o
Article history:
Received 24 April 2014
Received in revised form 3 May 2014
Accepted 6 May 2014
Available online 5 June 2014
Keywords:
Zinc oxide thin films
Vacuum annealing
Lattice relaxation
Raman spectroscopy
a b s t r a c t
The crystal structure and quality of ZnO thin films were enhanced
by high temperature vacuum annealing. High quality ZnO thin
films have been grown on a-plane sapphire substrate by radio
frequency (rf) magnetron sputtering method at a substrate
temperature of 600 C. A remarkable improvement in the epilayer
quality were established by in situ high temperature annealing.
The film quality, smoothness, the in plane stress, and the degree
of epitaxy of the films have been evaluated. The crystalline quality
was investigated by X-ray diffraction (XRD), scanning electron
microscopy (SEM), atomic force microscopy (AFM) and Raman
spectroscopy analyses. An extremely smooth ZnO films were
achieved at higher annealing temperatures with root mean square
roughness of 0.3 nm. The transverse optical mode A1(TO) observed
in all the samples and the longitudinal optical mode A1(LO)
appeared only at higher annealing temperatures over 800 C in
the micro-Raman scattering measurements. The strain of c-axis
were relaxed and the lattice parameter was comparable to that
of bulk ZnO at high annealing temperature of 900 C.
2014 Elsevier Ltd. All rights reserved.
1. Introduction
Zinc oxide has a wide band gap of 3.37 eV at room temperature. It has a potentially useful
properties including chemical stability, non-toxic (environmental friendly), catalytic activity,
⇑ Corresponding authors. Present address: Texas State University-San Marcos, Department of Chemistry and Biochemistry,
601 University Dr., San Marcos, TX 78666, USA. Tel.: +1 512 557 1495 (E.S.M. Duraia).
E-mail addresses: duraia_physics@yahoo.com, ed24@txstate.edu (E.S.M. Duraia).
http://dx.doi.org/10.1016/j.spmi.2014.05.022
0749-6036/ 2014 Elsevier Ltd. All rights reserved.

2. A.N. Fouda et al. / Superlattices and Microstructures 73 (2014) 268–274 269
piezoelectricity, low cost in production, commercial availability of large area single crystal ZnO sub-strates,
and bio-compatibility [1,2]. ZnO has a wide range of applications like gas sensors, photo-cat-alysts,
transparent electrodes, dilute magnetic semiconductors (DMS), photo-detectors, microsensors,
window layer for solar cells, and active channel layer of transparent thin film transistor (TTFT) wave-guides,
surface acoustic wave devices, and ultra-violet/blue emission devices [3,4].
Light emitting diodes and lasing development have been focused on GaN with exciton binding
energy of 24 meV. However, ZnO is more efficient emitter and promising in optoelectronic applica-tions
than GaN [5]. ZnO has a large exciton binding energy of 59 meV and the major challenge for
ZnO in optoelectronic device applications is the production of good quality and wide area ZnO films
by low cost method [6]. Based on Bose–Einstein condensation, Zamfirescu et al. proposed a model
for ZnO microcavity structure with a promising exciton oscillating strength toward the fabrication
of exciton–polariton laser in ZnO [7].
The epitaxial growth of ZnO have been conducted by a variety of methods, such as pulsed laser
deposition PLD [8], laser assisted molecular beam epitaxy L-MBE [9], plasma assisted MBE [10],
MBE [11], chemical vapor deposition CVD [12], MOCVD [13] chemical spraying [14], sol gel [15],
ion beam deposition [16] and sputtering [17–21]. Among these, sputtering method can produce wide
area films with controlled composition economically. However, the film quality and roughness do not
match the requirements of optoelectronic devices. Epitaxial growth of ZnO films can be established on
different kind of substrates like ZnO [22], GaN [11], AlN [23], ScAlMgO4 [5] and Al2O3 [24]. Al2O3 is one
of the most extensively adopted substrates for the growth of ZnO epilayers because of its hexagonal
symmetry, low cost and commercial availability. Many researchers try to grow high quality ZnO films
on sapphire substrate. Lin et al. [25] succeeded in the growth of high quality ZnO films on c-plane sap-phire
substrate without a buffer layer. However, for epitaxially grown (0001) ZnO on (0001) Al2O3
defects are easily generated, because of the large lattice mismatch 18.4% which decreases to 0.08%
for (0001) ZnO grown on (11–20) Al2O3 [26].
In the present paper, high quality ZnO films were grown by rf magnetron sputtering on a-plane
sapphire substrate. The influence of high temperature vacuum annealing on the epilayer quality,
microstructure and crystalline perfection has been investigated. We try to find out the optimum
growth and heat treatment conditions.
2. Experimental
150 nm thick ZnO thin films were deposited by rf magnetron sputtering, using radio frequency
source (RF, 13.56 MHz). High purity ZnO target (99.999) with 5.1 diameter has been used. Growth
temperature of 600 C were established by SiC heater. The films deposited at a bias voltage of
350 V, working pressure of 5 104 Torr, background pressure of 2 106 Torr and radio frequency
power of 700 W. High purity Ar (99.999), O2 (99.9999) with Ar to O2 ratio of (Ar/O2 = 4/1) were used.
Ar and O2 flow were controlled by mass flow controller. Pre-sputtering was performed to remove con-taminations
on the target. Smooth a-plane sapphire substrate were cleaned using organic solvents,
rinsed in DI-water and baking in high vacuum at 750 C before deposition. The smoothness of the sub-strate
is shown in Fig. 2e. In situ high temperature vacuum annealing of (0001) ZnO films was carried
out for an hour at 700 C, 800 C, 850 C and 900 C respectively.
The film thickness were measured using the cross-section profile of scanning electron microscopy
(Model JEOL JSM-840). XRD measurements were performed by using Burker-D8 diffractometer with
Cu Ka radiation. In non-contact mode, AFM was used to study the roughness and surface morphology
of the deposited films. The characterizations were extended by Raman spectroscopy measurements at
room temperature. Micro-Raman spectroscopy system (Model Renishaw system 2000) with Ar+ laser
at wavelength of 488 nm with power of 100 mW has been used to investigate the local vibration
modes from ZnO epilayers grown on a-plane sapphire substrates.
3. Results and discussion
SEM has been used to investigate the film thickness. Fig. 1 depicts the side view of the film at dif-ferent
locations. The growth rate was 73 nm/h and it can be seen that, the thickness of the grown films

3. 270 A.N. Fouda et al. / Superlattices and Microstructures 73 (2014) 268–274
Fig. 1. Side view of SEM image for c-plane ZnO on a-plane sapphire substrate.
was around 150 nm. The sharpness of the interface between film and substrate indicate the unifor-mity,
smoothness and homogeneity of the grown films.
Fig. 2 shows AFM images for the films annealed at 700 C (Fig. 2a), 800 C (Fig. 2b), 850 C (Fig. 2c),
and 900 C (Fig. 2d), respectively. Small grains can be observed in little rough surface in Fig. 2a and b
for the vacuum annealed films at 700 C and 800 C. In addition, with increasing the annealing tem-perature,
coalescence process of grains is extended. Atomically clearly flat ZnO epilayers exhibiting
z-axis height less than 3 nm were conducted and RMS value decreases to 0.3 nm. The variation of
RMS values with annealing temperature is shown in Fig. 2f. RMS values regularly decrease with
increasing the annealing temperature up to 850 C and the increment at 900 C is related to the for-mation
of intermediate hexagonal pits. Small typically hexagonal pits were observed for the films
annealed at 900 C, which illustrate the backing of ZnO on a nearly lattice matched substrate [26].
The formation of similar hexagonal pits was also reported by Wang et al. for ZnO films grown by
MBE [27]. The obtained small roughness is so much smaller than the values reported for sputtered
ZnO films [18–21] and even comparable to high quality films grown by MBE [11] and PLD [8]. To
our knowledge, this is one from the best result reported for sputtered ZnO films on sapphire
substrates. Such enhancement was attributed to the fact that thermal annealing enhances the mass
transport and coalescence of adjacent grains [28] besides selecting the optimum preparation
conditions on a smooth a-plane sapphire substrate as shown in Fig. 2e.
Micro-Raman measurements have been carried out to investigate the nature of crystalline quality.
Wurtzite ZnO belongs to the C6v symmetry group. According to the well-known selection rules, there
are the Raman active phonon modes E2(low), E2(high), A1(TO), A1(LO), E1(TO), and E1(LO). The E2
modes and the A1(LO) mode are expected to be observed in un polarized Raman spectra [29]. Fig. 3
represents the room temperature Raman spectrum of ZnO films at different annealing temperatures
and Optical surface images for the spots of incidence. From the figure, one can see that; The transverse
optical mode A1(TO) appears at 380 cm1 for all the samples. While the longitudinal optical mode
A1(LO) at 576 cm1 appears only at higher annealing temperatures over 800 C. The disappearance
of E1(LO) in all samples indicates that our samples have little defects such as Zn interstitials, oxygen
vacancies, and the free carrier lack [30,31]. The other peaks at 417, 650 and 750 cm1 are attributed to
the Al2O3 substrate [32,33]. The peaks are slightly asymmetric and broadening, this is due to the quan-tum
confinement of the phonons. Therefore, the contribution to the Raman spectrum will not only
from the phonons of Brillouin zone center but also from the other phonons that is confined due to
the nanoscale effect [34]. The low intensity E2 at 430 cm1 indicates the small stress due to the
difference in the thermal coefficients between sapphire and zinc oxide [35].
h–2h scan for ZnO thin films annealed at 700 C, 800 C, 850 C and 900 C are shown in Fig. 4a. The
dominant diffraction peak in all scans is the (0002) peak beside the substrate (11–20) Al2O3 peak.
(0002) ZnO peak arises from diffraction from basal ZnO planes with c-axis perpendicular to the sub-strate.
The sharpness and symmetric nature of the peak confirm the quality of the epilayers. FWHM

4. A.N. Fouda et al. / Superlattices and Microstructures 73 (2014) 268–274 271
2.5
2.0
1.5
1.0
0.5
decreases with increasing the annealing temperature and for the films annealed at 900 C, (0002) ZnO
peak with FWHM of 0.0672 was observed in Fig. 4a. No reflections corresponding to other planes
were observed except (0002) ZnO which is very symmetric. This indicates the absence of any other
phases. The sharpness of ZnO (0002) reflection shows a good c-axis orientation perpendicular to
a-plane sapphire substrate without the well-known rotation by 30 which ZnO films exhibited on
c-plane sapphire substrate [25,26]. x scan for the films annealed at 900 C is shown in Fig. 4b. The
value of FWHM for rocking curve scan has a very small value of 0.0026, implying the ordering and
little tilt in c-plane.
Fig. 5 shows the lattice constant obtained from ZnO (0002) reflection in comparison with bulk ZnO
lattice constant. A well identified lattice relaxation by high temperature annealing were confirmed.
This implies that the film was relaxed by increasing the annealing temperature and the lattice con-stant
is comparable to bulk ZnO [4]. High temperature annealing enhances the quality and orientation
of the grown ZnO epilayers.
700 750 800 850 900
0.0
RMS VALUE (nm)
ANNEALING
TEMPERATURE (°C)
a b
c d
e
1nm
0nm
f
Fig. 2. AFM images (2 lm 2 lm) and cross-sectional height profiles of atomically smooth ZnO films grown on a-plane
sapphire substrates which annealed at (a) 700 C, (b) 800 C, (c) 850 C and (d) 900 C. (e) AFM for the used smooth a-plane
sapphire substrate. (f) Graphical representation of annealing temperature dependence of RMS values for ZnO epilayers.

5. 272 A.N. Fouda et al. / Superlattices and Microstructures 73 (2014) 268–274
TA= 850οC
TA= 800οC
400 600 800 1000 1200
900
10 μm
850°
800°
700
TA= 900οC
TA= 700οC
INTENSITY (arb. units)
WAVE NUMBER (cm-1)
Fig. 3. Raman spectra of ZnO films on a-plane sapphire substrates measured at room temperature under various annealing
conditions and optical images for the measuring spot.
(11-20) Al2O3)
(0002) ZnO
T = 900°C
T =850°C
T = 800°C
T = 700°C
33 34 35 36 37 38 39
2θ (deg.)
INTENSITY (arb.units)
-1.0 -0.5 0.0 0.5 1.0
Δω (deg.)
(0002) ZnO
FWHM=0.0026°
INTENSITY (arb.units)
(a)
(b)
Fig. 4. (a) Shows h–2h scan of (0001) ZnO on (11–20) Al2O3 substrate, ZnO (0002) reflection is observed. (b) Shows x-scan
rocking curve recorded for the (0002) line with a width of only 0.0026.

6. A.N. Fouda et al. / Superlattices and Microstructures 73 (2014) 268–274 273
c - BULK (ZnO)
700 800 900
5.217
5.214
5.211
5.208
5.205
c - AXSIS STRAIN (%)
c - LATTICE CONSTANT (Å)
ANNEALING TEMPERATURE (C°)
0.18
0.12
0.06
0.00
Fig. 5. Relation between c-lattice parameter and strain for ZnO thin films under various annealing conditions.
700 750 800 850 900
5x109
4x109
3x109
2x109
1x109
0
IN PLANE STRESS (dyn/cm2)
ANNEALING TEMPERATURE (C°)
Fig. 6. The in plane stress of ZnO epilayers on a-plane sapphire substrates under different annealing temperatures.
The residual stress in the c-plane of annealed ZnO films can be calculated based on biaxial strain
model for hexagonal lattice [36]
r ¼ ð233 1010Þ
cfilm cbulk
cbulk
dyne=cm2
Here r is the in plane stress, cfilm is the lattice constant obtained from the (0002) reflection in the XRD
profile, cbulk is the corresponding bulk value [37,38]. The calculated in-plane stress values for thermally
annealed ZnO films are shown in Fig. 6. The value of stress in our samples is two order of magnitude
lower than the values reported by some other groups [36]. Generally, intrinsic stress and thermal
stress can be found in ZnO films. Due to the small lattice mismatch in our films, the intrinsic stress
induced by lattice mismatch is small. We believe that the resultant stress is mainly due to thermal
stress which is attributed to the difference in the thermal expansion coefficients of the substrate
and film [35]. High temperature annealing reduce the in plane stress in ZnO films and promote grain
growth.

7. 274 A.N. Fouda et al. / Superlattices and Microstructures 73 (2014) 268–274
4. Conclusions
The effect of vacuum annealing on the relaxation of ZnO films was illustrated. ZnO epilayer had a
smooth surface comparable to layer by layer films grown by MBE and PLD. As determined by XRD
measurements, the sharpness of ZnO (0002) reflection shows a good c-axis orientation perpendicular
to a-plane sapphire with a small FWHM. Our result suggests few threading dislocations which is
related to x-rocking curve of ZnO (0002) reflection and high temperature annealing improves the
ordering of the films. A well identified lattice relaxation by high temperature annealing were con-firmed.
This implies that the film was relaxed by increasing the annealing temperature and the lattice
constant is comparable to bulk ZnO. Micro-Raman analysis confirm the small stress in the films.
References
[1] D.C. Look, J.W. Hemsky, J.R. Sizelove, Phys. Rev. Lett. 82 (1999) 2552.
[2] I. Vallsa, M. Cantu, Energy Environ. Sci. 2 (2009) 19–34.
[3] S. Krishnamoorthy, A.A. Iliadis, A. Inumpudi, S. Choopun, R.D. Vispute, T. Venkatesan, Solid State Electron. 46 (2002) 1631.
[4] C. Jagadish, S. Pearton, Zinc Oxide Bulk, Thin Films and Nanostructures, Elsevier, 2006.
[5] A. Tsukazaki, A. Ohtomo, T. Onuma, M. Ohtan, T. Makino, M. Sumiya, et al, Nat. Mater. 4 (2005) 42.
[6] W. Shiao, C. Chi, S. Chin, C. Huang, T. Tang, Y. Lu, et al, J. Appl. Phys. 99 (2006) 054301.
[7] M. Zamfirescu, A. Kavokin, B. Gil, G. Malpuech, M. Kaliteevski, Phys. Rev. B 65 (2002) 161205(R).
[8] A. Tsukazaki, A. Ohtomo, M. Kawasaki, T. Makino, C. Chia, Y. Segawa, H. Koinuma, Appl. Phys. Lett. 83 (2004) 2784.
[9] Y. Liu, S. Yang, G. Wei, J. Pan, Y. Yuan, C. Cheng, J. Mater. Sci. Technol. 29 (12) (2013) 1134–1138.
[10] D.M. Bagnall, Y.F. Chen, Z. Zhu, T. Yao, Appl. Phys. Lett. 70 (1997) 2230.
[11] H. Ko, Y. Chen, S. Hong, T. Yao, J. Cryst. Growth 209 (2000) 816–821.
[12] E.S.M. Duraia, G.W. Beall, Z.A. Mansurov, T.A. Shabanova, A. Hannora, Eurasian Chem.-Technol. J. 15 (2013) 19–24.
[13] S.I. Inamdar, K.Y. Rajpure, J. Alloys Compd. 595 (2014) 55–59.
[14] G. Srinivasan, J. Kumar, Cryst. Res. Technol. 41 (9) (2006) 893–896.
[15] Y. Kokubun, H. Kikura, S. Nakaakagomi, Jpn. J. Appl. Phys. 42 (2003) L 904–L 906.
[16] L. Bahadur, T.N. Rao, J.P. Pandey, Semicond. Sci. Technol. 9 (1994) 275.
[17] Y. Liang, X. Deng, J. Alloys Compd. 569 (2013) 144–149.
[18] T. Koyama, A.N. Fouda, N. Shibata, S.F. Chichibu, J. Appl. Phys. 102 (2007) 073505.
[19] T. Nishiyama, K. Takezawa, Y. Nakazawa, T. Oyanagi, K. Oishi, S. Nakamura, et al, Jpn. J. Appl. Phys. 53 (2014) 02BC02.
[20] L. Cui, Hua, Y. Zhang, G. Gen Wang, F. Xu Yang, Xu Ping, et al, Appl. Surf. Sci. 258 (2012) 2479–2485.
[21] V.V. Khomyak, M.M. Slyotov, I.I. Shtepliuk, G.V. Lashkarev, P.D. Marianchuk, V.V. Kosolovskiy, O.M. Slyotov, J. Phys. Chem.
Solids 74 (2013) 291–297.
[22] H. Kato, M. Sano, K. Miyamoto, T. Yao, Jpn. J. Appl. Phys. 42 (2003) L1002–L1005.
[23] U. Ozgur, Y.L. Alivov, C. Liu, A. Teke, S. Dogan, V. Avrutin, S.J. Cho, H. Morkoc, J. Appl. Phys. 98 (2005) 41301.
[24] T.E. Murphy, S. Walavalkar, J.D. Phillips, Appl. Phys. Lett. 85 (2004) 26.
[25] M. Lin, Chun Tsung Su, Hong Chang Yan, Ming Yau Chern, Jpn. J. Appl. Phys. 44 (31) (2005) L 995–L 997.
[26] P. Fons, K. Iwata, S. Niki, A. Yamada, K. Matsubara, M. Watanabe, J. Cryst. Growth 209 (2000) 532–536.
[27] X. Wang, Y. Tomita, O.H. Roh, M. Ohsugi, S.B. Che, Y. Ishitani, A. Yoshikawa, Appl. Phys. Lett. 86 (2005) 11921.
[28] S. Nitta, M. Kariya, T. Kashima, S. Yamaguchi, H. Amano, I. Akasaki, Appl. Surf. Sci. 159 (2000) 421–426.
[29] M. Tu, Y. Sua, C. Ma, J. Appl. Phys. 100 (2006) 053705.
[30] S.H. Jeong, J.K. Kim, B.T. Lee, J. Phys. D: Appl. Phys. 36 (2003) 2017–2020.
[31] S.B. Yahia, S. Ben Yahia, L. Znaidi, A. Kanaev, J.P. Petitet, Spectrochim. Acta Part A 71 (2008) 1234–1238.
[32] T. Wermelinger, C. Borgia, C. Solenthaler, R. Spolenak, Acta Mater. 55 (2007) 4657–4665.
[33] W. Jia, W.M. Yen, J. Raman Spectrosc. 20 (1989) 785.
[34] Y. Du, M.s. Zhang, Appl. Phys. A 76 (2003) 171–176.
[35] H. Kato, M. Sano, K. Miyamoto, T. Yao, Jpn. J. Appl. Phys. 42 (2003) L1002.
[36] P. Mondal, D. Das, Appl. Surf. Sci. 286 (2013) 397–404.
[37] J.G. Lu, Z. Ye, Y.J. Zeng, L.P. Zhu, L. Wang, J. Yuan, B.H. Zhao, Q.L. Liang, J. Appl. Phys. 100 (2006) 073714.
[38] Y.E. Jeong, S. Park, Curr. Appl. Phys. 14 (2014) 30–33.