The effects of ultraviolet exposure on the device characteristics of atomic layer deposited-ZnO:N thin film transistors

Article abstract of DOI:10.1149/1.3560191

We investigated the effects of ultraviolet (UV) light illumination on nitrogen-doped atomic layer deposited (ALD)-ZnO:N thin film transistors (TFTs). ALD ZnO:N thin films grown at 125C were used as active layers for back-gate TFT devices. As-fabricated ALD ZnO:N TFTs showed proper drain current modulation response to a gate voltage sweep with a 5.4 V threshold voltage and a clear pinch-off. However, the threshold voltage was significantly shifted in the negative direction by UV exposure due to an associated increase in carrier concentration, resulting in the loss of current modulation by gate voltage sweep. In addition, we observed a resistivity change in ALD ZnO:N thin films with time after UV exposure. The resistivity decreased by several orders of magnitude upon UV light exposure and recovered toward its original value after switching off the UV light. Accordingly, the transfer curves of TFT devices using a ZnO:N active layer also exhibited recovery characteristics. We formed a thin Al₂O₃ passivation layer on top of the TFT surface in order to suppress the recovery effect.

Full text of DOI:10.1149/1.3560191

J150
Journal of The Electrochemical Society, 158 (5) J150-J154 (2011)
0013-4651/2011/158(5)/J150/5/$28.00 The Electrochemical Society
C
V
The Effects of Ultraviolet Exposure on the Device
Characteristics of Atomic Layer Deposited-ZnO:N
Thin Film Transistors
Jae-Min Kim,a S. J. Lim,b Taewook Nam,a Doyoung Kim,a and Hyungjun Kima,
,z
*
aSchool of Electrical and Electronic Engineering, Yonsei University, Seoul, 120-749, Korea
bDepartment of Materials Science and Engineering, Pohang University of Science and Technology,
Pohang 790-784, Korea
We investigated the effects of ultraviolet (UV) light illumination on nitrogen-doped atomic layer deposited (ALD)-ZnO:N thin
film transistors (TFTs). ALD ZnO:N thin films grown at 125ꢀC were used as active layers for back-gate TFT devices. As-fabricated
ALD ZnO:N TFTs showed proper drain current modulation response to a gate voltage sweep with a 5.4 V threshold voltage and
a clear pinch-off. However, the threshold voltage was significantly shifted in the negative direction by UV exposure due to an
associated increase in carrier concentration, resulting in the loss of current modulation by gate voltage sweep. In addition, we
observed a resistivity change in ALD ZnO:N thin films with time after UV exposure. The resistivity decreased by several orders of
magnitude upon UV light exposure and recovered toward its original value after switching off the UV light. Accordingly, the transfer
curves of TFT devices using a ZnO:N active layer also exhibited recovery characteristics. We formed a thin Al2O3 passivation layer
on top of the TFT surface in order to suppress the recovery effect.
C
V
2011 The Electrochemical Society. [DOI: 10.1149/1.3560191] All rights reserved.
Manuscript submitted November 9, 2010; revised manuscript received January 21, 2011. Published March 15, 2011. This was
Paper 1790 presented at the Las Vegas, Nevada, Meeting of the Society, October 10–15, 2010.
Experimental
In recent years, the demand for transparent thin film transistors
(TFTs) has enormously increased for next generation transparent
displays such as transparent organic light emitting diodes, heads-up
displays, and smart windows.1 To meet this demand, oxide semicon-
ductors have been extensively studied to replace conventional Si-
based TFTs mainly due to their high transparency to visible light
and good electrical properties.2–4 Particularly, polycrystalline zinc
oxide (ZnO) thin films are considered as attractive candidates
because they possesses evident advantages over opaque Si-based
TFTs such as high saturation mobility, high on-off current ratio,
wide optical band gap, and low temperature processability.
ALD-ZnO:N thin films were deposited using a home-made hot
wall-type ALD reactor with diethyl zinc (DEZ) (Epichem adduct
grade) as a precursor and diluted ammonium hydroxide (NH4OH)
solution (0.01%) as a single source for both oxygen and nitrogen
(via ammonia) doping. The growth temperature was maintained at
125ꢀC during the ALD process using a ceramic heater located under
the graphite substrate holder. The flow of Ar carrier gas with precur-
sor molecules was controlled by a needle valve to maintain an oper-
ating pressure of 160 mTorr during the DEZ exposure step. Argon
gas (99.99% purity) was used as a purging and a carrier gas, and the
flow was controlled by a mass flow controller (5 sccm for DEZ bub-
bler and 20 sccm for purging). A typical ALD process is composed
of four consecutive steps: DEZ exposure for 2 s, Ar purging for 8 s,
NH4OH exposure for 3 s, and Ar purging for 4 s. The thicknesses of
the films were routinely measured using an ellipsometer (Rudolph
auto ELII), and resistivity was measured using a four-point probe
with a source meter (Keithley 2400).
To investigate the UV exposure effects on the device properties,
we prepared inverted, staggered-type ALD-ZnO:N TFT devices
(with bottom gate and top contact) on an nþ Si substrate, also used
as the gate electrode. First, 100-nm-thick ALD-Al2O3 was deposited
as a gate insulator using trimethyl aluminum (TMA) and water
vapor at a growth temperature of 150ꢀC. Then, a 66-nm-thick
ZnO:N active layer was prepared using the ALD process mentioned
above. The active channel area was defined using a standard litho-
graphic process with AZ-5214 as a PR material, followed by wet
etching with a diluted HCl solution (HCl:H2O ¼ 1:40). The channel
width and length of the device were confined to be 40 and 20 ꢀm,
respectively. Finally, a 100-nm-thick Ti layer was deposited and
patterned via lift-off using magnetron sputtering for the source=
drain contact. The schematic configurations of the tilt and top view
of the TFT device are shown in Fig. 1.
However, since ZnO-TFTs inherently contain deep level defects
in the channel=dielectric interface which act as recombination cen-
ters, they could generate an unwanted photo-current and alter film
properties during light transmission and device operation, resulting
in severe problems.5–7 For example, Park et al. reported that simultane-
ous exposure of white light on ZnO-based TFTs underwent nega-
tive shift in threshold voltage under negative-bias stress condi-
tions.8 Besides, Bae et al. observed the generation of photocurrent
in ZnO-based TFTs in UV illuminated condition.9,10 Additionally,
Goldberger et al. reported that UV irradiation causes changes in the
electrical properties of ZnO field effect transistors (FETs) due to
the decrease in resistivity with increasing carrier concentration.11
However, detail electrochemical mechanisms as a result from UV
illumination on ZnO-based TFTs have rarely been discussed.
In this article, we investigated the changes in device properties
of ALD-ZnO:N TFTs with respect to UV light exposure. Especially,
we concentrate on the electrochemical mechanisms in occurrence
with UV illumination with ZnO surface analysis to support the
anticipated phenomena. We previously reported that TFT devices
with an atomic layer deposited (ALD)-, nitrogen-doped ZnO thin
film active layer show excellent characteristics of high ꢀsat (6.7
cm2=V s) and Ion=off (ꢁ108).12 A significant negative shift in the
threshold voltage (VTH) was observed after UV exposure, indicating
an increase in carrier concentration in the ZnO channel area. How-
ever, this change was fully recovered to its original state after
switching off the UV light when the device was exposed in air with-
out a surface passivation layer. In contrast, we adopted a thin Al2O3
passivation layer on the top surface of the TFT and observed that
the recovery was effectively suppressed.
After the device preparation, UV light was applied for 0, 5, and
30 min under vacuum conditions using a specially designed UV ex-
posure system maintained at 50 mTorr base pressure via a mechani-
cal pump. The wavelength and the illumination power of the UV
light used in this study were 368 nm and 0.1 mW=m2, respectively.
The characteristics of ZnO:N TFTs, including transfer and output
curves, during UV exposure were measured using a Keithley 4200
semiconductor parameter analyzer with three probes.
To investigate the resistivity changes in the ZnO:N thin films
post-UV exposure in ambient air, we prepared 66-nm-thick ALD-
ZnO:N films on glass (Corning 1737) substrates followed by
*
Electrochemical Society Active Member.
z E-mail: hyungjun@yonsei.ac.kr
Downloaded on 2015-06-24 to IP 157.182.150.22 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 158 (5) J150-J154 (2011)
J151
Figure 2. The changes in transfer curves for ALD-ZnO:N TFT as a function
of UV exposure time.
(VDS). Further, the output curves showed an almost linear increase
with VDS, instead of the normal saturation behavior.
Since VTH is intimately related to the carrier concentration in the
channel layer, the negative VTH shift may be related to an increase
in electron concentration in the active layer. Thus, resistivity meas-
urements were carried out for the UV-exposed sample, indicating
that the resistivity of the as-deposited ZnO:N film with an initial re-
sistivity of 1500 X cm significantly decreased to 5 X cm after UV
exposure for 30 min. According to previous reports,10,13–16 the
increase in electron concentration in ZnO as a result of UV exposure
can be explained by the generation of oxygen vacancies denoted by
Figure 1. (Color online) The schematic drawing of an inverted staggered
type TFT using ALD-ZnO:N as an active layer.
patterning of 100-nm-thick Ti electrodes using sputtering with a
stainless steel hard mask. After sample preparation, we exposed the
ZnO:N films to UV light for 30 min with the same wavelength and
power as above under vacuum conditions. Then, we measured the
resistivities of the ZnO:N thin films with increasing post-UV expo-
sure time up to 168 h (1 week) after breaking the vacuum. Addition-
ally, X-ray photoemission spectroscopy (XPS) was conducted to
analyze the changes in the chemical bonding states of oxygen
molecules on the surfaces of the ZnO:N films as a function of UV
exposure time without surface cleaning via Ar sputtering.
For experiments on the recovery of ALD ZnO:N TFT character-
istics, UV was applied for 30 min to the devices with and without an
Al2O3 passivation layer, and the transfer curves were measured im-
mediately after UV exposure and after 168 h in air. The passivation
layer, i.e., a 30-nm-thick ALD-Al2O3 thin film, was deposited at
Ts ¼ 150ꢀC on top of the TFT surface layer.
Results and Discussion
The effects of UV exposure on ZnO:N TFT devices.— As we
have previously reported, undoped ALD-ZnO thin films prepared at
Ts ¼ 150ꢀC have a low resistivity of several X cm due to a large n-
type carrier concentration greater than 1017 cm–3, resulting in non-
operative TFT devices.12 However, the resistivity increased to more
than 1500 X cm after nitrogen doping due to a decrease in carrier
concentration using 0.01% NH4OH as a reactant. As a result, the de-
vice exhibited a clear pinch-off and current saturation with good de-
vice characteristics of high saturation mobility (6.7 cm2=Vs) and
on=off current ratio (ꢁ108).
Here, we investigated the changes in device properties of ALD-
ZnO:N TFT with respect to UV exposure. The transfer and output
curves as a function of exposure time are shown in Figs. 2 and 3,
respectively. The device with no UV exposure was well modulated
by a gate voltage sweep and exhibits hard saturation, evidenced by
the flatness of the slope of each curve, indicating that the entire
thickness of the ZnO channel layer can be depleted. The threshold
voltage (VTH) and saturation current (ISat) were 5.4 V and 21 ꢀA,
respectively, at VDS ¼ 30 V for VG ¼ 30 V. Furthermore, the device
operated in enhancement mode, exhibiting normally-off characteris-
tic. However, the transfer curves after UV exposure for 5 min
showed significant negative shifts in VTH from 5.4 to ꢂ10.5 V. Af-
ter a UV exposure of 30 min, the device could not be turned off
even with a ꢂ35 V gate voltage at a 5 V source=drain voltage
Figure 3. The changes in output curves for ALD-ZnO:N TFT as a function
of UV exposure time for (a) 0 min, (b) 5 min, and (c) 30 min.
Downloaded on 2015-06-24 to IP 157.182.150.22 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

J152
Journal of The Electrochemical Society, 158 (5) J150-J154 (2011)
V0O0 . Incident UV light, which possesses higher energy than the
bandgap energy of ZnO, can generate electron-hole pairs due to
band-to-band excitation. The generated holes are consumed by the
desorption of oxygen ions in the ZnO lattice near the surface, result-
ing in the creation of V0O0 through the following reactions
et al. observed that dissociatively adsorbed water molecules (OH–)
are preferable adsorbents to physically adsorbed water molecules
(H2O).10 Also, Feng et al. showed the effect of the increase in OH–
concentration on the defect sites under UV illumination when sur-
rounded by water molecules.23 Likewise, we observed the existence
of hydroxyl groups on the surfaces of ALD-ZnO:N films using XPS.
Fig. 5a shows the XPS spectra of the O 1s binding energy region for
ZnO:N films after exposure by UV light for 0, 5, and 30 min. It
should be noted that the XPS measurements were conducted 50 h af-
ter the samples were removed from the UV exposure chamber. All
of the three XPS spectra contain two peaks at 530.5 and 532.3 eV,
corresponding to O-Zn and O-H bonding, respectively. Even in the
unexposed sample, a peak for O-H bonding is observed. This is a
general observation for ZnO films deposited using various techni-
ques including sputtering, pulsed laser deposition (PLD), spray py-
rolysis, and chemical vapor deposition (CVD).10,24–27 However, we
observed that the relative intensity between O-H and O-Zn bonding
(O-H=O-Zn) was increased for the film exposed to UV light for a
longer time, as shown in Fig. 5b. The longer the sample was
exposed to UV light, the larger was the number of photo-generated
oxygen vacancies created by the desorption mechanism, which, as
mentioned previously, provides a higher density of adsorption sites
for water molecules on ZnO surfaces.
Since the UV exposure was conducted under vacuum, there is lit-
tle possibility that the adsorption of existing water molecules occurs
at oxygen vacancy sites during the exposure step. Meanwhile, if the
sample was exposed to water in air ambient after extracting the sam-
ple from the UV chamber, it would cause a chemisorption process
of oxygen vacancies with water molecules, leading to the formation
of hydroxyl bonding on the ZnO surface. Hence, it is reasonable to
consider that the increased adsorption sites on a UV-illuminated
ZnO surface enhance the reaction rate for the recovery of oxygen
vacancies through an adsorption mechanism, resulting in an addi-
tional increase in the O-H level compared to those of samples not
exposed to UV. This result provides a strong basis for the recovery
behavior observed in ZnO:N films.
ZnO þ 2 hm ! eꢂ þ hþ
[1]
[2]
[3]
O
2ꢂ þ hþ ! Oꢂ
O
ꢂ þ hþ ! 1=2O2ðgÞ þ VO00 þ 2eꢂ
Consequently, desorption of an oxygen ion creates free carriers in
the form of 2eꢂ, which results in significant VTH decreases in ZnO
TFTs. Also, it is well known that the holes in ZnO-based semicon-
ductors exhibit strong localization at oxygen 2p levels or in an upper
edge of the valence band due to the highly electronegative nature of
oxygen.17–20 Thus, it is worth mentioning that photo-generated
holes under UV illumination tend to be localized to regions rich in
oxygen, exhibiting more electron trapping in the channel layer. Con-
sequently, a more negative gate voltage is required to deplete elec-
tron carriers from the channel layer in order to turn off the TFT de-
vice.14 This explanation is in agreement with our TFT characteristic
measurement results. Thus, ZnO:N TFTs after UV exposure for 30
min cannot operate properly due to the generation of excess
electrons.
Post-exposure recovery effects.— Next, the effects of air expo-
sure on the electrical properties of ALD-ZnO:N thin films after turn-
ing off the UV light were investigated. Figs. 4 shows the resistivity
change in the 30 min UV-exposed ALD-ZnO:N film with post-UV
exposure time in air. For the initial 30 h, the resistivity remained at
a relatively constant value of 5 X cm and then increased gradually,
reaching 600 X cm after 168 h (1 week), illustrating recovery to-
ward its original value before UV exposure. This recovery behavior
of the film resistivity can be explained by the surface adsorption of
oxygen molecules from an oxygen source, such as water vapor in
ambient air, which reduces the oxygen vacancies in ZnO films and
decreases carrier concentration. As previously mentioned, holes
generated by UV exposure under vacuum condition are consumed
by desorption of oxygen ions, resulting in the creation of V0O0 in the
ZnO surface. However, when the UV-treated samples are exposed
to air, the generated V0O0 act as adsorption sites for oxygen sources,
leading to the following adsorption reaction21
The recovery of electrical properties in ZnO:N thin films due to
air exposure produces significant effects on TFT device properties.
Figure 6 shows the shift in transfer curves of ZnO:N TFT with
1=2O2ðgÞ þ V0O0 þ 2eꢂ ! OX
[4]
As a result, the adsorption of the oxygen to the V0O0 defect sites
destroys the free carrier 2eꢂ, resulting in an increase in the resistiv-
ity of ZnO film, as shown in Fig. 4.
The defective site, V0O0 , is known to be a favorable adsorption site
for water molecules (H2O) and hydroxyl groups (OH–).11,22 Sun
Figure 5. (a) The XPS spectra of the O 1s peak of ALD-ZnO:N thin films
50 h after UV exposure for 0, 5, and 30 min. (b) The peak area ratio (O-H=
O-Zn) of each sample.
Figure 4. The resistivity change in 30 min UV-exposed ALD-ZnO:N thin
film with increasing post-UV exposure time.
Downloaded on 2015-06-24 to IP 157.182.150.22 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 158 (5) J150-J154 (2011)
J153
Table I. The time dependence of device property change for
ALD-ZnO:N TFT (a) without and (b) with an Al2O3 passivation
layer after 30 min of UV exposure.
The time after
UV exposure
Before UV Just after UV
exposure exposure
After 3 h After 168 h
(a) UV exposure for 30 min on ZnO:N TFT without passivation layer
VTH (V)
S (V=dec)
ꢀ
IOFF (pA)
ION=OFF
5.4
0.5
1.4
1.12
1.9 ꢃ 107
—
—
1.1
—
—
–21.1
0.9
1.2
5.4
0.5
1.4
1.06
7.3 ꢃ 106
sat (cm2=Vs)
161
1.0 ꢃ 106
(b) UV exposure for 30 min on ZnO:N TFT with passivation layer
VTH (V)
S (V=dec)
ꢀ
IOFF (pA)
ION=OFF
5
0.95
1.2
1.7
1.8 ꢃ 107
—
—
1.5
—
—
—
—
1.4
—
—
ꢂ2
0.5
1.6
98
9.3 ꢃ 105
Figure 6. The time dependence of transfer curve change for ALD-ZnO:N
TFT after 30 min of UV exposure.
sat (cm2=Vs)
respect to post-UV exposure times of 0, 3, 168 h in ambient air. As
mentioned previously, UV exposure for 30 min causes a significant
negative VTH shift, resulting in improper operation of the device.
However, we observed that this negatively shifted transfer curve
slowly recovers to its original state (before UV exposure) with time
after deactivating the UV light. Three hours after exposure, the de-
vice began to be modulated properly by a gate voltage sweep with
VTH ¼ ꢂ21.1 V. Moreover, the transfer curve returned to its initial
state 168 h after exposure. It should be noted that the VTH obtained
at 168 h after UV exposure has the same value with that before UV
exposure as to 5.4 V.
For comparison, we prepared ZnO:N TFT devices passivated
with 30-nm-thick ALD-Al2O3, and we subsequently exposed them
to UV for the same time and conditions as those in the control sam-
ple. Then, we measured the changes in the transfer curves of the de-
vice with respect to post-UV exposure time in order to investigate
the effects of the passivation layer on the recovery characteristics.
The specific purpose of the Al2O3 passivation layer is to block the
ZnO channel layer from air exposure. Experimental results are sum-
marized in Fig. 7. The device operated well with a gate bias accord-
ing to the transfer curve for the TFT before UV exposure with
VTH ¼ 5.0 V. However, modulation by the gate bias was not
observed within the sweeping range after UV exposure for 30 min
due to a large negative VTH shift, the same behavior as was seen
with the unpassivated ZnO:N TFTs.
UV-exposed TFT characteristics. Generally, ZnO exhibits strong
chemisorption with molecules in the surrounding environment,
especially oxygen-containing gaseous species, resulting in a change
in conductivity.28,29 Thus, this slow recovery behavior can be
explained by the thin passivation layer prohibiting the chemisorp-
tion of water molecules and hydroxyl groups as a source for oxygen
to recover oxygen vacancies in ZnO lattice. Similarly, Chang et al.
reported that SiO2=Si3N4 bilayer passivation on ZnO nanowires
showed superior FET performance relative to that of unpassivated
ZnO, which was attributed to the reduction in surface chemisorp-
tions processes at oxygen vacancy sites.30
Tables I(a) and I(b) summarize the device parameter changes
according to the recovery behavior of ZnO:N TFT as described in
Figs. 6 and 7, respectively. While the saturation mobility remained
almost constant at around ꢁ1.3 cm2=V s, regardless of post-UV ex-
posure time, the photo-induced off-current was greatly increased in
both cases. For example, the IOFF for the TFT without passivation
layer 3 h after UV exposure was 161 pA, 2 orders of magnitude
higher than that before UV exposure. It is well known that UV light
generates photo-induced carriers in ZnO, providing higher current
than that in the normal state.31–34 Thus, this result can be explained
by the withdrawal of electrons generated through UV illumination
to the back channel area of the ZnO film, contributing to the
increase in off-current.
However, in the post-UV exposure recovery regime, the result
was different from that of the controlled sample. Specifically, the
device with a passivation layer still did not exhibit source-drain
current modulation by a gate bias sweep even 3 h after exposure.
We only observed that the device showed the proper TFT transfer
characteristic 168 h after UV exposure with ꢂ2.0 V of VTH but
still did not fully recover to its original state. This result confirms
that an Al2O3 passivation layer effectively retarded the recovery of
Conclusion
In conclusion, we investigated the changes in ALD-ZnO:N TFT
characteristics with respect to UV exposure time and post-exposure
time. We have shown that the VTH of the device shifts in the nega-
tive direction with increasing UV exposure. Although the resistivity
of ZnO:N thin film after UV exposure for 30 min had a much lower
value than that before UV exposure, it tended to increase toward its
initial value. The XPS spectra showed an increase in the number of
hydroxyl groups on the ZnO:N film surface with increasing UV ex-
posure time, illustrating the recovery mechanism occurring in
ZnO:N films, which explains the resistivity increase with time after
UV exposure. Also, we compared the recovery behaviors of ZnO:N
TFTs with and without an Al2O3 passivation layer. Although the
transfer curve of the device completely recovered to its original state
with increasing post-exposure time, the use of a thin passivation
layer effectively suppressed this phenomenon.
Acknowledgments
This research was supported by the Future-based Technology
Development Program (Nano Fields) and Basic Research Program
through the National Research Foundation of Korea (NRF) funded
by the Ministry of Education, Science and Technology (No. 2010-
Figure 7. The time dependence of transfer curve change for ALD ZnO:N
TFT with a 30-nm-thick ALD-Al2O3 passivation layer after 30 min of UV
exposure.
Downloaded on 2015-06-24 to IP 157.182.150.22 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

J154
Journal of The Electrochemical Society, 158 (5) J150-J154 (2011)
16. H. Y. Hannes Kind, B. Messer, M. Law, and P. Yang, Adv. Mater., 14, 158 (2002).
17. J.-L. Zhao, X.-M. Li, A. Krtschil, A. Krost, W.-D. Yu, Y.-W. Zhang, Y.-F. Gu,
and X.-D. Gao, Appl. Phys. Lett., 90, 062118 (2007).
18. H. Yanagi, S.-i. Inoue, K. Ueda, H. Kawazoe, H. Hosono, and N. Hamada, J. Appl.
Phys., 88, 4159 (2000).
19. A. N. Banerjee and K. K. Chattopadhyay, Prog. Cryst. Growth Charact. Mater.,
50, 52 (2005).
20. U. Ozgur, Y. I. Alivov, C. Liu, A. Teke, M. A. Reshchikov, S. Dogan, V. Avrutin,
S. J. Cho, and H. Morkoc, J. Appl. Phys., 98, 041301 (2005).
21. S. E. Ahn, J. S. Lee, H. Kim, S. Kim, B. H. Kang, K. H. Kim, and G. T. Kim, Appl.
Phys. Lett., 84, 5022 (2004).
0020230 and 2010-0024066). Also, this work was supported by
the Yonsei University Research Fund of 2010 (No. 2009-1-0195 and
2009-1-0233).
Yonsei University assisted in meeting the publication costs of this article.
References
1. J. F. Wager, Science, 300, 1245 (2003).
2. K. Nomura, H. Ohta, A. Takagi, T. Kamiya, M. Hirano, and H. Hosono, Nature,
432, 488 (2004).
22. R. Wang, N. Sakai, A. Fujishima, T. Watanabe, and K. Hashimoto, J. Phys. Chem.
B, 103, 2188 (1999).
23. X. Feng, L. Feng, M. Jin, J. Zhai, L. Jiang, and D. Zhu, J. Am. Chem. Soc., 126, 62
(2003).
3. K. P. Sang-Hee, H. Chi-Sun, R. Minki, Y. Shinhyuk, B. Chunwon, S. Jaeheon, L.
Jeong-Ik, L. Kimoon, O. Min Suk, and I. Seongil, Adv. Mater., 21, 678 (2009).
4. S. J. Lim, K. Soon-ju, K. Hyungjun, and P. Jin-Seong, Appl. Phys. Lett., 91,
183517 (2007).
24. H. Xu, Y. Liu, C. Xu, Y. Liu, C. Shao, and R. Mu, Appl. Phys. Lett., 88, 242502
(2006).
25. T. P. Smith, W. J. Mecouch, P. Q. Miraglia, A. M. Roskowski, P. J. Hartlieb, and
R. F. Davis, J. Cryst. Growth, 257, 255 (2003).
5. H. S. Bae and S. Im, J. Vac. Sci. Technol. B, 22, 1191 (2004).
6. R. J. Collins and D. G. Thomas, Phys. Rev., 112, 388 (1958).
7. R.-D. Sun, A. Nakajima, A. Fujishima, T. Watanabe, and K. Hashimoto, J. Phys.
Chem. B, 105, 1984 (2001).
26. L. Mar, P. Timbrell, and R. Lamb, Thin Solid Films, 223, 341 (1993).
27. S. Park, T. Ikegami, and K. Ebihara, Thin Solid Films, 513, 90 (2006).
28. M. Liu and H. K. Kim, Appl. Phys. Lett., 84, 173 (2004).
29. J. Lagowski, J. E. S. Sproles, and H. C. Gatos, J. Appl. Phys., 48, 3566 (1977).
30. P.-C. Chang, Z. Fan, C.-J. Chien, D. Stichtenoth, C. Ronning, and J. G. Lu, Appl.
Phys. Lett., 89, 133113 (2006).
8. J. S. Park, T. S. Kim, K. S. Son, J. S. Jung, K.-H Lee, J.-Y. Kwon, B. Koo, and S.
Lee, IEEE Electron Device Lett. 31, 440 (2010).
9. H. S. Bae and S. Im, Thin Solid Films, 469, 75 (2004).
10. H. S. Bae, C. M. Choi, J. Hoon Kim, and S. Im, J. Appl. Phys. 97, 076104 (2005).
11. J. Goldberger, D. J. Sirbuly, M. Law, and P. Yang, J. Phys. Chem. B, 109, 9
(2004).
31. H. S. Bae, M. H. Yoon, J. H. Kim, and S. Im, Appl. Phys. Lett., 83, 5313 (2003).
32. A. Solbrand, A. Henningsson, S. Sodergren, H. Lindstrom, A. Hagfeldt, and S.-E.
Lindquist, J. Phys. Chem. B, 103, 1078 (1999).
33. Y. Ushio, M. Miyayama, and H. Yanagida, Jpn. J. Appl. Phys., 33, 1136 (1994).
34. V. Subramanian, E. E. Wolf, and P. V. Kamat, J. Phys. Chem. B, 107, 7479
(2003).
12. S. J. Lim, J.-M. Kim, D. Kim, S. Kwon, J.-S. Park, and H. Kim, J. Electrochem.
Soc., 157, H214 (2010).
13. D. H. Zhang and D. E. Brodie, Thin Solid Films, 238, 95 (1994).
14. K. Takechi, M. Nakata, T. Eguchi, H. Yamaguchi, and S. Kaneko, Jpn. J. Appl.
Phys., 48, 010203 (2009).
15. R. L. Hoffman, B. J. Norris, and J. F. Wager, Appl. Phys. Lett., 82, 733 (2003).
Downloaded on 2015-06-24 to IP 157.182.150.22 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).