Home Search Collections Journals About Contact us My IOPscience

ZnO nanowire-embedded Schottky diode for effective UV detection by the barrier reduction

effect

This article has been downloaded from IOPscience. Please scroll down to see the full text article.

2010 Nanotechnology 21 115205

(http://iopscience.iop.org/0957-4484/21/11/115205)

View the table of contents for this issue, or go to the journal homepage for more

Download details:
IP Address: 165.132.28.124
The article was downloaded on 16/07/2010 at 06:09

Please note that terms and conditions apply.

IOP PUBLISHING NANOTECHNOLOGY
Nanotechnology 21 (2010) 115205 (5pp) doi:10.1088/0957-4484/21/11/115205

ZnO nanowire-embedded Schottky diode

for effective UV detection by the barrier
reduction effect
Joondong Kim1,6 , Ju-Hyung Yun1 , Chang Hyun Kim1,2 ,
Yun Chang Park3 , Ju Yeon Woo1 , Jeunghee Park2 , Jung-Ho Lee4 ,
Junsin Yi5 and Chang-Soo Han1,6
1
Nano-Mechanical Systems Research Division, Korea Institute of Machinery and Materials
(KIMM), Daejeon 305343, Korea
2
Department of Chemistry, Korea University, Jochiwon 339700, Korea
3
Measurement and Analysis Division, National Nanofab Center (NNFC), Daejeon 305806,
Korea
4
Department of Materials and Chemical Engineering, Hanyang University, Ansan 426791,
Korea
5
School of Information and Communication Engineering, Sungkyunkwan University,
Suwon 440746, Korea

E-mail: joonkim@kimm.re.kr (Joondong Kim) and cshan@kimm.re.kr (Chang-Soo Han)

Received 14 January 2010, in final form 2 February 2010

Published 22 February 2010
Online at stacks.iop.org/Nano/21/115205

Abstract
A zinc oxide nanowire (ZnO NW)-embedded Schottky diode was fabricated for UV detection.
Two types of devices were prepared. The ZnO NW was positioned onto asymmetric metal
electrodes (Al and Pt) for a Schottky device or symmetric metal electrodes (Al and Al) for an
ohmic device, respectively. The Schottky device provided a rectifying current flow and was
more sensitive to UV illumination than the ohmic device. The Schottky barrier plays an
important role for UV detection by a UV-induced barrier reduction effect.
The fabrication of the ZnO NW-embedded Schottky diode and the UV reaction mechanism
are discussed in light of the UV light-induced Schottky barrier reduction effect.
(Some figures in this article are in colour only in the electronic version)

Nanomaterials, such as carbon nanotubes (CNTs) and
nanowires (NWs), have emerged as potential one-dimensional
building blocks due to their unique features. Electric
conductive NWs have been utilized as functional microscopy
tips [1] and nanoscale interconnects [2, 3]. Recently,
researchers have attempted to address the substantial demand
of nanomaterials for cost-effective solar cells [4].
Photodetectors are among the most promising applications
of semiconducting NWs. Dangling bonds on a NW easily
trap holes, which induce a delay in recombination, thereby
causing an increase in the electron current. Furthermore,
wavelength-selective detection may be possible by combining
different types of NWs [5]. Due to its direct and wide energy
bandgap (3.37 eV) and large surface-to-volume ratio, a zinc
6 Authors to whom any correspondence should be addressed.
oxide (ZnO) NW is currently considered to be a promising
metal oxide material for UV detectors and light-emitting
diodes [6–9]. Although the use of ZnO NWs has advantages, it
is difficult to achieve p-type conductivity, which is an obstacle
to applications in diodes.
The Schottky barrier plays an important role in electronic
applications, such as gas sensors [10], solar cells [11]
and nanogenerators [12]. Although many researchers
have investigated the Schottky contact between CNTs and
metal [13–15], few studies have been conducted on ZnO
NW Schottky devices, and most used the same metal
electrodes [5, 16, 17]. With respect to device fabrication, it
is necessary to verify the Schottky formation between the ZnO
NW and the metal electrode for UV detection.
This study presents the systematic fabrication of a
ZnO NW-embedded Schottky diode using asymmetric metal

0957-4484/10/115205+05$30.00 1 © 2010 IOP Publishing Ltd Printed in the UK

Nanotechnology 21 (2010) 115205 J Kim et al

Figure 1. (a) A low-magnification TEM image. (b) A high-resolution TEM image. ZnO NW has a single-crystalline structure with a lattice
space of 0.258 nm. The inset is an FFT image. (c) Energy dispersive x-ray spectrum of ZnO NW.

Figure 2. (a) An SEM image of the ZnO NW connection between Pt and Al electrodes. (b) A schematic diagram of the ZnO NW-embedded
device. (c) The energy band diagram of ZnO and Pt. The high work function Pt metal forms a potential barrier for electrons. (d) The energy
band diagram of ZnO and Al. The low work function Al metal forms an ohmic contact.

electrodes. It was found that the Schottky device provides
higher UV responsivity than an ohmic device due to the
existence of a potential barrier between the metal and the
semiconductor. The fabrication of the ZnO NW-embedded
Schottky diode and the UV reaction mechanism are discussed
in light of the UV light-induced barrier reduction effect.
For ZnO NW growth, Au catalytic nanoparticles
(HAuCl4 · 3H2 O, Sigma) were deposited on an Si substrate. A
powder-type Zn source (99.998%, Aldrich) was placed inside
a quartz tube reactor. The growth temperature was controlled
to be 500 ◦ C under an Ar gas atmosphere.
An ultrasonication process was performed for 30 min to
detach the NWs from the substrate. They were then centrifuged
at 5000 rpm for 100 min to remove residuals. The purified
ZnO NW solution was dropped onto metal electrodes under
ac bias. Two types of devices were prepared. The ZnO NW
was positioned onto asymmetric metal electrodes (Al and Pt)
for a Schottky device or symmetric metal electrodes (Al and
Al) for an ohmic device, respectively. Rapid thermal process
(RTP-3000, SNTEK, Korea) was subsequently performed for
2 min under a nitrogen atmosphere at 500 ◦ C to improve the
adhesion between the NW and the metal electrode contacts.
Field emission transmission electron microscopy (FEI
Tecnai F30 Super-Twin) was used to investigate the ZnO
NW. Selected-area electron diffraction and fast Fourier
transformations (FFTs, GATAN) revealed a crystalline
structure and the growth direction of the ZnO NW. Energy
dispersive x-ray spectroscopy (EDAX Genesis) was used to
investigate the NW composition. The ZnO NW was placed
on a carbon-coated copper mesh grid for TEM and EDS
analyses. The NW position on the metal electrodes was
observed via field emission scanning electron microscopy
(FEI Sirion). The electrical characteristics of the ZnO NW-
embedded devices were obtained using a probe station with an
instrument (Keithley 2400).
Figure 1(a) shows a low-magnification TEM image. A
high-resolution TEM (HRTEM) image of a ZnO NW projected
in the [11̄0] zone axis is presented in figure 1(b). The space

2
Nanotechnology 21 (2010) 115205
Figure 3. (a) I –V characteristics of a Schottky device (Pt–ZnO
NW–Al) and an ohmic device (Al–ZnO NW–Al). (b) Semi-log plots.
The ohmic device showed symmetric I –V characteristics. The
Schottky device provided a diode performance with a rectifying ratio
of 371 and an ideality factor of 1.2.
between the adjacent planes was measured to be 0.258 nm,
corresponding to the (002) lattice plane of a hexagonal ZnO
structure [18]. The inset of the figure is an FFT image, which
confirms the single-crystalline structure. The growth direction
is [001] perpendicular to the (001) basal plane of the hexagonal
system. The energy dispersive x-ray spectrum clearly shows
Zn and O signals from the NW, as shown in figure 1(c). The
C Kα (0.28 keV) and Cu Kα (8.04 keV) peaks originate from
the carbon-film-coated copper mesh TEM grid.
A solution containing the prepared ZnO NWs (0.5 μl)
was dropped onto the patterned electrodes under an ac electric
bias of 5 Vp−p at 1 MHz. Figure 2(a) shows the ZnO NW
connection between Pt and Al electrodes through a 700 nm
space gap. A schematic view of the ZnO NW-embedded device
is presented in figure 2(b). ZnO has a work function () of
4.3 eV and is known to be a natural n-type semiconductor due
to the oxygen vacancy, which leads to a shallow donor level [6].
3
J Kim et al
Figure 2(c) is the energy band diagram of ZnO and Pt metal.
The Pt work function (pt = 5.3 eV) is higher than that of ZnO
(ZnO = 4.1 eV). When a contact is formed, electrons flow
from ZnO to Pt metal until the Fermi levels align, resulting in
band bending. In contrast, Al has a lower work function than
ZnO (Al = 4.06 eV < ZnO ), which has no barrier to the
electron flow and provides an ohmic contact, as depicted in
figure 2(d).
The Schottky device (Al–ZnO NW–Pt) clearly provided a
rectifying current flow, as shown in figure 3(a). Under forward
bias, the current flow is enhanced due to the reduction of
the potential barrier. In order to evaluate the performance of
the Schottky device comparatively, an ohmic device (Al–ZnO
NW–Al) was also investigated. The latter device showed a
linear current flow under both forward and reverse biases. This
implies that the design of metal contacts for ZnO NW is an
effective approach to form a Schottky diode, although some
reports presented Schottky formation by using homogeneous
metal contacts [5, 12, 18]. The semi-log plots are presented
in figure 3(b). While the ohmic device showed the symmetric
I –V characteristics, the Schottky device provided the diode
performance with a rectifying ratio of 371. Under thermionic
emission theory, the I –V characteristics are given by the
following relations:
J = Js (eqv/nkT − 1) or J = Js (eqv/nkT )
for V  kT /q (1)
 
∗∗ 2 qφB
Js = A T exp − (2)
kT
q ∂V q ∂V
n≡ = (3)
kT ∂(ln J ) kT 2.3 × ∂ log I
where Js , n , kT , A∗∗ and φB are the saturation current density,
ideality factor, thermal energy (eV), Richardson constant and
barrier height, respectively. The ideality factor of the ZnO NW
Schottky diode is given as 1.2, indicating a little problematic
recombination in the device [5, 19].
The barrier heights of metal–semiconductor systems are
determined by both the metal work function and the surface
states. A general expression of the barrier height is given in
equation (2), where surface states are neglected:
 
−kT J 1
φB = ln qv/kT ∗∗ 2 . (4)
q e A T
In ZnO, the trapping mechanism governs the photocon-
duction. The oxygen molecules are absorbed along the ZnO
NW surface and they capture electrons from the semiconduc-
tor [5]. Under UV illumination, the photon energy releases the
trapped electrons and also generates photo-induced electrons,
causing an enhancement of the current.
The fabricated ZnO NW-embedded Schottky and ohmic
devices were investigated for UV responses. A relatively low
UV light (24 μW cm−2 , 254 nm) was illuminated onto the
devices. The responsivity ( R = IUV /Ii ) is defined as the ratio
of the UV illuminating current ( IUV ) versus the initial current
( Ii ).
Figure 4(a) presents the Schottky device responses to UV
light with bias voltage modulation. Two different bias voltages
Nanotechnology 21 (2010) 115205 J Kim et al

Responsivity

Figure 4. (a) UV responses of the Schottky device. The responsivity ( R = IUV /Ii ) was significantly affected by modulation of the bias
voltage. (b) I –V measurements of the Schottky device in the dark and under UV illumination. The turn-on voltage was lowered by UV
illumination. (c) UV responses of an ohmic device. The decay time, defined as 37% of the maximum photocurrent, was measured from (d) the
ohmic device at 1 V bias, (e) the Schottky device at 0.4 V, and (f) the Schottky device at 1.0 V.

of 0.4 and 1 V were applied to the Schottky device. While the
Schottky device showed a low responsivity of approximately
7 with a 1 V bias, a significantly improved responsivity of
nearly 200 was obtained with a 0.4 V bias, which is close
to the diode turn-on voltage. Beyond a critical voltage, the
higher voltage bias reduces the potential barrier. This enhances
the electron flow, resulting in lower sensitivity to UV. This
indicates that the potential barrier is an important factor in the
UV reaction mechanism. It should be noted that the applied
voltage may greatly affect the UV detection performance in a
given device.
Figure 4(b) shows the I –V characteristics of the Schottky
device measured both in the dark and under UV illumination.
There is no significant transition from Schottky to ohmic-
like characteristics [12, 19, 20]. In the dark, the reference
turn-on current (20 nA) was reached at 0.467 V. Under UV
illumination, the turn-on voltage was reduced to 0.176 V. This
is attributed to the reduction of the UV light-induced potential
barrier, resulting in current enhancement at a fixed voltage
value.
The barrier height values were found to be 1.33 eV in the
dark condition, which is close to the difference between the
Pt work function [21] of 5.3 eV and the electron affinity of
ZnO of ( X = 4.1 eV) [22]. The Richardson constant value
was chosen as 32 A cm−1 K−2 . Under UV illumination, the
Schottky barrier height was reduced to 1.037 eV. The barrier
reduction by UV illumination is thus obtained as 0.293 eV.
To confirm the proposed mechanism of a UV light-
induced barrier reduction effect, the UV response of an ohmic
device was investigated as presented in figure 4(c).
As there is no barrier between the metal and semiconduc-
tor, the electrons freely move to the metal, even without UV
illumination. This leads to a reduced UV response from the
ohmic device.

4
Nanotechnology 21 (2010) 115205
The UV response times of the ZnO NW devices were also
obtained. The decay time, which is defined as the time to reach
37% of the maximum photocurrent [17, 23], was measured
from the ohmic device at 1 V bias, the Schottky device at 0.4 V
and the Schottky device at 1.0 V. These results are presented in
figures 4(d), (e), and (f), respectively.
The ohmic device required a relatively long decay time of
19.886 s. The Schottky device (operating at 0.4 V) meanwhile
provided a significant decrease of the decay time of 0.58 s,
indicating that the existence of the Schottky barrier is important
for effective UV detection. With a bias of 1 V, the decay time of
the Schottky device increased to be 16.435 s. This is attributed
to the decrease of the potential barrier due to the high bias
voltage, resulting in an insensitive reaction.
In conclusion, we have fabricated a ZnO NW-embedded
Schottky device for UV detection. The potential barrier
between the metal and semiconductor plays an important role
in the UV responses. Due to the effect of reducing the UV
light-induced barrier, the bias voltage significantly affects the
level of UV responsivity. To confirm this mechanism, an ohmic
device was also investigated and was found to be less UV
responsive due to the absence of a potential barrier.
Acknowledgments
The authors acknowledge the financial support from the Center
for Nanoscale Mechatronics and Manufacturing (CNMM) of
the 21C Frontier Research Program and the National Research
Foundation (NRF, 2009-0082018 and 2009-0082527) of the
Ministry of Education, Science and Technology (MEST) in
Korea. J Park thanks the WCU (World Class University)
program (R31-10035) for their support.
References
[1] Kim J, Shin Y-H, Yun J-H, Han C-S, Hyun M and
Anderson W A 2008 Nanotechnology 19 485713
J Kim et al
[2] Kim J and Anderson W A 2006 Nano Lett. 6 1356
[3] Kim J, Park Y C, Shin D H, Lee E-S and Han C-S 2007 Appl.
Phys. Lett. 90 253103
[4] Tian B, Zheng X, Kempa T J, Fang Y, Yu N, Yu G,
Huang J and Lieber C M 2007 Nature 449 885
[5] Soci C, Zhang A, Xiang B, Dayeh S A, Aplin D P R, Park J,
Bao X Y, Lo Y H and Wang D 2007 Nano Lett. 7 1003
[6] Lu J G, Chang P and Fan Z 2006 Mater. Sci. Eng. R 52 49
[7] Law J B K and Thong J T L 2006 Appl. Phys. Lett. 88 133114
[8] Lin D, Wu H, Zhang W, Li H and Pan W 2009 Appl. Phys. Lett.
94 172103
[9] Chosh R, Dutta M and Basak D 2007 Appl. Phys. Lett.
91 073108
[10] Salehi A and Kalantari D J 2007 Sensors Actuators B 122 69
[11] Kim J, Han C-S, Park Y C and Anderson W A 2008 Appl.
Phys. Lett. 92 043501
[12] Liu J, Fei P, Song J, Wang X, Lao C, Tummala R and
Wang Z L 2008 Nano Lett. 8 328
[13] Manohar H M, Wong E W, Schlecht E, Hunt B D and
Siegel P H 2005 Nano Lett. 5 1469
[14] Lu C, An L, Fu Q, Liu J, Zhang H and Murduck J 2006 Appl.
Phys. Lett. 88 133501
[15] Yang M H, Teo K B K, Miline W I and Hasko D G 2005 Appl.
Phys. Lett. 87 253116
[16] Kind H, Yan H, Messer B, Law M and Yang P 2002 Adv. Mater.
14 158
[17] Zhou J, Gu Y, Hu Y, Mai W, Yeh P, Bao G, Sood A K,
Polla D and Wang Z L 2009 Appl. Phys. Lett. 94 191103
[18] JCPSD International Center for Diffraction Data 2001 JCPDS
File No. 79-0205
[19] Heo Y W, Tien L C, Norton D P, Pearton S J, Kang B S,
Ren F and LaRoche J R 2004 Appl. Phys. Lett. 85 3107
[20] Keem K, Kim H, Kim G-T, Lee J S, Min B, Cho K,
Sung M-Y and Kim S 2004 Appl. Phys. Lett. 84 4376
[21] Graaf H, Maedler C, Kehr M and Oekermann T 2009 J. Phys.
Chem. C 113 6910
[22] Dhananjay, Nagaraju J and Krupanidhi S B Physica B 391 344
[23] Fan Z, Chang P-C, Lu J G, Walter E C, Penner R M,
Lin C-H and Lee H P 2004 Appl. Phys. Lett. 85 6128