You are on page 1of 9

Journal of Physics: Conference Series

PAPER • OPEN ACCESS

Characteristics of Cu-doped ZnO films prepared using magnetron co-


sputtering
To cite this article: A S Yusof and Z Hassan 2020 J. Phys.: Conf. Ser. 1535 012047

View the article online for updates and enhancements.

This content was downloaded from IP address 188.68.0.252 on 23/06/2020 at 14:28


ICoSeMT 2019 IOP Publishing
Journal of Physics: Conference Series 1535 (2020) 012047 doi:10.1088/1742-6596/1535/1/012047

Characteristics of Cu-doped ZnO films prepared using


magnetron co-sputtering

A S Yusof* and Z Hassan


Institute of Nano Optoelectronics Research and Technology (INOR), Universiti Sains
Malaysia, 11800 USM, Penang, Malaysia
*
sauffi@usm.my, zai@usm.my

Abstract. This work reports the fabrication of un-doped and cu-doped zinc oxide (ZnO) films
using magnetron co-sputtering technique. The fabricated films were deposited on glass substrates
followed by thermal annealing process in an ambient condition for 60 minutes at 400 °C. X-ray
diffraction (XRD), ultraviolet-visible (UV-Vis) spectrophotometry, field emission scanning
electron microscopy (FE-SEM) and Hall measurement with four-point Van der Pauw method
were used to study the crystal orientation, optical properties, surface structure and the resistivity
of the deposited films respectively. Polycrystalline films with hexagonal wurtzite structure were
observed in the CZO films. The redshift and bandgap narrowing were discussed in optical
analysis. The trend in bandgap narrowing was contributed by the hybridization of O 2p and Cu
3d band. Hall effect measurement showed that the resistance of deposited Cu-doped ZnO (CZO)
films will increase at excessive-high Cu doping level.

1. Introduction
ZnO is a well-established material and among the most significant II-VI semiconductor compounds.
The profound recognition in this particular material is motivated by the combination of several
advantages such as wide direct bandgap (3.37 eV) and large exciton binding energy (60 meV) which is
ideal for short-wavelength optoelectronic devices that can be operated at elevated temperature and
hostile environment [1]–[3]. ZnO also a promising candidate for transparent conductive oxide (TCO)
application and solar cell devices owning to its additional unique properties such as non-toxic,
luminesces in the visible and ultraviolet spectrum, and broad availability for a large high-quality bulk
single crystal.[4].
To meet the requirement of different application, the optoelectrical characteristics of the ZnO can be
manipulated to attain a preferable value by introducing an element impurities as a dopant into ZnO
lattice site such as aluminium (Al) [5], silver (Ag), copper (Cu) [6], and tin (Sn) [7]. Through doping,
the fundamental electrical and electronic characteristics of ZnO can be control which is the most
significant issue in the development of optoelectronic devices and active electronics. Between the
available dopants, Cu has identical electronic shell configuration and radius with Zn, making Cu a
suitable dopant in the modification of ZnO optoelectronic properties.
Usually, ZnO films which are widely used in TCOs application for optoelectronic devices are
prepared using physical vapor deposition (PVD) systems such as pulsed laser deposition, thermal
evaporator and reactive magnetron sputtering. Among those, magnetron sputtering offers a fast coating
rate, flexibility to sputter any materials such as metal, compound or alloy, high purity and uniformity
coating with high adhesion [8]. Throughout the years, many efforts have been done to integrate Cu into

Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution
of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Published under licence by IOP Publishing Ltd 1
ICoSeMT 2019 IOP Publishing
Journal of Physics: Conference Series 1535 (2020) 012047 doi:10.1088/1742-6596/1535/1/012047

the ZnO lattice site using a magnetron sputtering system. Thus, various configuration of the sputtering
target has been accomplished such as by using DC or RF reactive magnetron sputtering of Cu chips/foils
embedded on Zn metal target or ZnO ceramic target respectively [9], [10]. In this work, Cu doped ZnO
(CZO) thin films were deposited on the glass substrate by using a magnetron co-sputtering system of
Cu metal target and ZnO ceramic target. The co-sputtering method offers more homogeneous doping
content which could be regulated easily by varying the sputtering power.

2. Experimental details
ZnO and CZO films were deposited by using a magnetron co-sputtering technique. A ceramic ZnO
target and Cu metal target were simultaneously sputtered by radio frequency (RF) and direct current
(DC) power supply respectively. The films were prepared on the glass substrates and went through a
thermal annealing process in an ambient environment for 60 minutes at 400 °C.
The glass substrates were cleaned using standard cleaning procedures which using a cleaning solvent
combined with an ultrasonic agitation system. The selected substrate was immersed in an ultrasonic bath
of deionized water, acetone and isopropanol sequentially each for 15 minutes at 50°C. Then, the
substrates were immersed in the deionized water to rinse any acetone or isopropanol leftover. The drying
process was done by blowing the substrates using nitrogen gas.
Before the deposition process initiated, the pressure within the sputtering chamber was evacuated
down to approximately 2 x 10-5 Pa. Then, the argon gas at the rate of 12 sccm flowed into the sputtering
chamber. After achieved the desired working pressure at around 5 x 10-3 Pa, the pre-sputtering process
was initiated for 10 minutes with a closed shutter to remove any contaminations or native oxides on the
surface of the targets. RF power source was fixed at 100 W and the DC power was varied at 8 W, 10 W
and 12 W to attain a different value of the deposited Cu percentage. The thickness of the deposited films
was maintained at 300 nm that can be monitored by Inficon SQM-160 Rate/Thickness Monitor
embedded in the sputtering system.
PANalytical X’PERT PRO X-ray diffraction (XRD) and Agilent Technology carry series ultraviolet-
visible (UV-Vis) spectrometer system was used to investigate the crystalline structure and the optical
properties from 300 to 1500 nm wavelength of the deposited films respectively. Field emission scanning
electron microscopy (FE-SEM, FEI Nova NanoSEM 450) was used to observe the surface morphology
of the deposited films. The elemental composition of the deposited films was determined using Oxford
Instrument energy dispersive X-ray (EDX). Lastly, Hall measurement with a four-point Van der Pauw
method was used to investigate the electrical properties of the deposited films.

3. Results and discussion


Effect of different DC sputtering power of Cu target on the surface structure of the ZnO and CZO films
were examined by utilizing FE-SEM, as displayed in Figure 1. The percentage value of Cu that exist
within the ZnO structure for each sample were measured based on the EDX results as shown in Figure
2 by a ratio of;

𝑁𝐶𝑢
× 100% (1)
𝑁𝑡𝑜𝑡

where 𝑁𝐶𝑢 are the number of Cu atom and 𝑁𝑡𝑜𝑡 are the total number of atoms. Un-doped ZnO films
comprised of compact polygonal grain. However, CZO films consist of prolonged grain mostly parallel
to the glass surface. Deposited films were observed to be compact and consistent. The granular grain
morphology was dispersed uniformly without any pinholes and dislocation, showing that high-quality
films have been deposited.
Figure 3 and Table 1 show the XRD pattern and (0002) diffraction peaks parameters of CZO films
deposited at different DC power. ZnO films show only one diffraction peak at 2θ = 34.325° which
belongs to ZnO (0002). This suggests that the film belonged to a hexagonal wurtzite structure with
preferential orientation along the c-axis normal to the substrate surface. When Cu introduced within the

2
ICoSeMT 2019 IOP Publishing
Journal of Physics: Conference Series 1535 (2020) 012047 doi:10.1088/1742-6596/1535/1/012047

structure deposited at 8 W (Cu DC sputtering power), an additional peak was observed which belong to
(101̅0) plane. There were no secondary phases detected within the sensitivity of the instrument. Thus,
indicate that most of the Cu atom that exhibits in the lattice site of ZnO crystallites act as interstitials
that share Zn and oxygen atoms [11].

Figure 1. FE-SEM images of CZO films fabricated using


magnetron co-sputtering technique at different DC power. (a)
0 W (ZnO), (b) 8 W, (c) 10W, (d) 12.

However, when the DC sputtering power was raised to 10 W and 12 W, metallic Cu (111) peak begin
to arise. Instead exist as interstitials, Cu-doping atoms can be energetically favourable to coalesce
simultaneously and become a residual cluster of Cu metallic [12]. These phenomena also can be seen
from calculated stress of the films shown in Table 1, where the stress of the film was lower when
deposited at 10 W compared to films deposited at 8 W. As observed in Figure 3, there was no peak
observed for ZnO (0002) orientation when the Cu sputtering power was set to 12 W. In combination
with high temperature (annealing) and heavy doping level, the crystal orientation which was parallel
and perpendicular to the glass surface will prone to coalesce together [13], resulting in the reduction of
ZnO (0002) peak.
From Table 1, the lattice constant of the deposited films was calculated using Miller-Bravais indices
for hexagonal lattice structure.
−1
2
4(ℎ2 + 𝑘 2 + ℎ𝑘) 𝑙 2
𝑑ℎ𝑘𝑙 =[ + 2] (2)
3𝑎2 𝑐

where 𝑎 and 𝑐 are the crystal plane lattice constant and 𝑑 is the crystalline plane spacing of indices (ℎ𝑘𝑙)
which is calculated from the Bragg’s formula. The calculate c of the CZO films were smaller than ZnO
films (0.5222 nm). This suggests that Cu2+ ions emerge as major substitutional for Zn2+ ions since Cu2+
ions has smaller ionic radii compare to Cu+ and Zn2+ which were at 0.060 nm each [14].

3
ICoSeMT 2019 IOP Publishing
Journal of Physics: Conference Series 1535 (2020) 012047 doi:10.1088/1742-6596/1535/1/012047

(a) Element At% (b) Element At%


O 47.72 O 49.52
Zn 52.28 Cu 7.25
Zn 43.23

(c) Element At% (d) Element At%


O 52.67 O 50.10
Cu 13.84 Cu 16.00
Zn 33.49 Zn 33.90

Figure 2. EDX measurement of the deposited films at different DC power. (a)


ZnO, (b) 8 W, (c) 10W, (d) 12 W.

Figure 3. XRD scan of the deposited films under varied DC


sputtering power.

4
ICoSeMT 2019 IOP Publishing
Journal of Physics: Conference Series 1535 (2020) 012047 doi:10.1088/1742-6596/1535/1/012047

Table 1. (0002) diffraction peaks parameters of CZO films deposited


at different DC power.
DC power (W) 0, (ZnO) 8 10 12
2θ (°) 34.325 34.617 34.515 -
𝒅 (nm) 0.2611 0.2590 0.2597 -
FHWM (o) 0.3231 0.4232 0.481 -
D (nm) 26.885 20.542 18.068 -
c 0.5222 0. 5180 0.5194 -
Stress -0.7610 1.1191 0.4924 -

Table 2 revealed the electrical properties of CZO films deposited at different DC power target,
determined using Hall effect measurement with four-point Van der Pauw method. The failure to measure
the resistivity of ZnO films may be attributed to the high resistance of the films and the measurement
limitation of the system. For CZO films deposited at 8 W, it was measured that the films resistivity was
higher compared to films deposited at 10 W (Cu sputtering power). This was because the state of Cu in
films was mostly exhibited as Cu2+ which has 3d9 4s0 2p6 valence band configuration and may therefore
can trap electron within the Cu 3d hole state [15].
When the DC sputtering power of Cu target was increased to 10 W, the ratio of Cu2+ phase compared
to Cu metallic phases were reduced due to segregation of Cu occur in the grain boundaries or interstices
site of the ZnO structure as predicted in XRD results where Cu metallic (111) phases start to appear,
causing a fall in resistance. For higher sputtering power of Cu (12 W), Cu atom acts as a discontinuous
carrier scattering, causing its mobility to decrease and resistance to increase [16].

Table 2. Electrical properties of CZO films deposited at different DC power.


DC power Resistivity, Ρ Mobility
Concentration (Cm-3)
(W) (Ω-cm) (Cm2/V-S)
0, (ZnO) - - -
8 1.78 X 103 16.95 2.07 X 1014
10 1.30 X 103 19.34 2.48 X 1014
12 5.43 X 104 4.89 2.35 X 1014

Figure 4 (a) and Figure 4 (b) show the transmittance spectra and the Tauc plot of CZO films
determined by using a UV-VIS spectrophotometer. Un-doped ZnO films showed good optical
transmittance with over 80% in the visible and infrared spectrum. For CZO films, the transmittance of
the films was decreased monotonously with increasing the sputtering power of Cu. This phenomenon
can be ascribed to free carrier absorption of photons [17].
Free carrier absorption is a photoinduced absorption process where it reduced the transmitted light
intensity without creating any extra electron-hole pair. Hence, the incident photons were absorbed via
either free electron or holes in the conduction and valence band structure of the material. Consequently,
it will induce an intra-band absorption (transition between states in the same band) or inter-band
absorption (transition between states in a different band) [18]. By increasing the sputtering power of Cu,
the generated Cu dopant will increase the concentration of free carriers within the deposited films as
shown in EDX data at Figure 2.
The injected carrier will fill the conduction band causing the band structure to be lifted up to higher
energies level in the same band structure. Therefore, the absorption of the films was enhanced
consequently. The optical bandgap of the films was calculated by extrapolating the straight line of the
Tauc relation to the point of (𝛼ℎ𝑣)2 = 0 [19]. It is showed that the redshift (ΔEg = 0.45 eV) and the
narrowing of the optical band gap occurred as Cu was doped into the ZnO structure. This was the result

5
ICoSeMT 2019 IOP Publishing
Journal of Physics: Conference Series 1535 (2020) 012047 doi:10.1088/1742-6596/1535/1/012047

of the energy level of Cu 3d band and O 2p band which were almost aligned together, and which led to
the exchange interaction between these bands [20], [21].

Figure 4. (a) Transmittance spectra CZO films deposited at different DC


power, (b) Calculated Tauc plot of the CZO films.

Then, hybridization of Cu 3d band and O 2p band occurred causing the fundamental direct band gap
of CZO thin films to be narrowed, as observed in Figure 4 (b). The redshift in the bandgap also has been
theoretically explained using the s–d and p–d exchange interactions by utilizing the second-order
perturbation theory by Bylsma et al. (1986) [22]. The redshift that occurs within the structure also
confirms the homogenous substitution of Cu ions in the lattice site of ZnO films [23]. The value of the
optical band gap un-doped ZnO films calculated in this work was at 3.30 eV, which is comparable with
the literature value (3.37 eV) [24].

4. Conclusion
Magnetron co-sputtering technique was used to deposit ZnO and CZO thin films using Cu and ZnO
target. The preferential of ZnO (0002) orientation was enhanced when CZO films were deposited at 8
W compared to at 12 W. Cu doping can reduce the lattice constant c due to incorporation of Cu2+ ions
in the lattice site of ZnO structure. The secondary phase of Cu metallic (111) was observed when the
CZO films were deposited at 10 W and 12 W Cu sputtering power. The intensity of the optical
transmittance decreases in CZO thin films and causes optical bandgap reductions. Since Cu 3d orbital
is shallower than Zn 3d orbital, by occupying Zn site with Cu atom in the ZnO films, it will cause a
strong d-p coupling between Cu and O atoms; moving O 2p up and narrowing of band gap of the films.
The electronic properties of the sample have been measured which showed high resistivity at 1.30 x 103
Ω cm to 5.43 x 104 Ω cm.

Acknowledgments
Support from Universiti Sains Malaysia and Malaysia Ministry of Education (MOE) under LRGS (Wide
Band Gap Semiconductor), Project No: 203/CINOR/6720013 is gratefully acknowledged.

References
[1] Pearton S J, Norton D P, Ip K, Heo Y W and Steiner T 2003 Superlattices Microstruct. 34 3–32
[2] Minami T 2005 Semicond. Sci. Technol. 20
[3] Zu P, Tang Z K, Wong G K L, Kawasaki M, Ohtomo A, Koinuma H and Segaw Y Solid State
Commun. 103 459–463
[4] Wisz G, Virt I, Sagan P, Potera P and Yavorskyi R 2017 Nanoscale Res. Lett. 12 0–6
[5] Guillén C and Herrero J 2010 Vacuum 84 924–929

6
ICoSeMT 2019 IOP Publishing
Journal of Physics: Conference Series 1535 (2020) 012047 doi:10.1088/1742-6596/1535/1/012047

[6] Chen Y, Xu X L, Zhang G H, Xue H and Ma S Y 2009 Phys. B Condens. Matter 404 3645–3649
[7] Zhao J, Ni J, Zhao X and Xiong Y 2011 J. Wuhan Univ. Technol. Mater. Sci. Ed 26 388–392
[8] George J 1992 Preparation of Thin Films (England: Taylor & Francis)
[9] Peng X, Xu J, Zang H, Wang B and Wang Z 2008 J. Lumin. 128 297–300
[10] Xue H, Chen Y, Xu X L, Zhang G H, Zhang H and Ma S Y 2009 Phys. E Low-Dimensional Syst.
Nanostruct. 41 788–791
[11] Lee J B, Lee H J, Seo S H and Park J S 2001 Thin Solid Films 398 641–646
[12] Yusof A S, Hassan Z and Zainal N 2018 Mater. Res. Bull. 97 314–318
[13] Lee Y S, Eom T W, Thuy V T T and Lee Y 2010 J. Korean Phys. Soc. 57 264
[14] Shannon R D 1976 Acta Crystallogr. Sect. A 32 751–767
[15] Wang X B, Li D M, Zeng F and Pan F 2005 J. Phys. D. Appl. Phys. 38 4104–4108
[16] Sivaramakrishnan K and Alford T L 2009 Appl. Phys. Lett. 94 5
[17] Shukla R K, Srivastava A, Kumar N, Pandey A and Pandey M 2016 Mater. Sci. Pol. 34 354–361
[18] Rudiger M, Greulich J, Richter A and Hermle M 2013 IEEE Trans. Electron. Devices 60 2156–
2163
[19] Chung S M, Shin J H, Lee J M, Ryu M K, Cheong W S, Park S H, Hwang CS and Cho K I 2011
J. Nanosci. Nanotechnol. 11 782–786
[20] Klein A 2013 J. Am. Ceram. Soc. 96 331–345
[21] Cattin L, Bernède, J C and Morsli M 2013 Phys. Status Solidi Appl. Mater. Sci. 210 1047–1061
[22] Sajjad M, Ullah I, Khan M I, Khan J, Khan M Y and Qureshi M T 2018 Results Phys. 9 1301–
1309
[23] Chakraborti D, Narayan J and Prater J T 2007 Appl. Phys. Lett. 90 6
[24] Srikant V and Clarke D R 1998 J. Appl. Phys. 83 5447–5451

You might also like