Selective wet-chemical etching of the barrier layer during formation of porous anodic aluminum oxide template

Article abstract of DOI:10.1149/1.3207011

A porous anodic aluminum oxide (AAO) template with a TiN bottom electrode was fabricated on a SiO₂/Si substrate. The anodizing process ended up stably with the formation of TiO2 pillars penetrating the barrier layer, and the TiO₂ pil

Full text of DOI:10.1149/1.3207011

Journal of The Electrochemical Society, 156 ͑11͒ K181-K185 ͑2009͒
K181
0013-4651/2009/156͑11͒/K181/5/$25.00 © The Electrochemical Society
Selective Wet-Chemical Etching of the Barrier Layer during
Formation of Porous Anodic Aluminum Oxide Template
Sang-Hyun Park,a Satbyul Kim,a Do-Joong Lee,a Seongjin Yun,b
,z
Zheong Gou Khim,b and Ki-Bum Kima,
*
aDepartment of Materials Science and Engineering, and bDepartment of Physics and Astronomy,
Nano Systems Institute-National Core Research Program, Seoul National University,
Seoul 151-742, Korea
A porous anodic aluminum oxide ͑AAO͒ template with a TiN bottom electrode was fabricated on a SiO2/Si substrate. The
anodizing process ended up stably with the formation of TiO2 pillars penetrating the barrier layer, and the TiO2 pillars were
selectively and uniformly removed using standard clean-1 solution ͑NH4OH:H2O2:H2O = 1:1:5͒. Ru nanowires were fabricated in
the AAO nanochannels with an aspect ratio of 10:1 by an atomic layer deposition process using Ru͑EtCp͒ as a precursor and O2
2
as a reducing agent. The contact between Ru nanowires and a TiN bottom electrode was confirmed by a transmission electron
microscope. The current–voltage characteristic of each Ru nanowire through the TiN bottom electrode was resolved by a conduc-
tive atomic force microscope, which showed the uniform ohmic behavior of the contact with a resistance of as low as
2.7 Ϯ 0.3 k⍀ per single Ru nanowire. The stable wet-chemical route to the selective and uniform removal of the barrier layer of
the AAO template was applied to devices that require a good electrical connection to nanowires with high aspect ratios, such as
sensors, batteries, or optoelectronic devices.
© 2009 The Electrochemical Society. ͓DOI: 10.1149/1.3207011͔ All rights reserved.
Manuscript submitted May 7, 2009; revised manuscript received July 9, 2009. Published September 2, 2009.
An array of vertically aligned nanowires has attracted consider-
able interest for device applications that require either a large sur-
face area or the unique properties of nanowires, such as sensors,
batteries, field emission devices, and optoelectronic devices.1-4
While there have been various attempts to fabricate these structures,
the use of an anodized aluminum oxide ͑AAO͒ as a template is one
of the most intensively investigated processes.5-9 In this respect, it is
not surprising that many researchers have attempted to develop a
process of directly integrating the AAO template on a substrate,
namely, fabricating an AAO template starting from a deposited Al
thin film instead of using Al foil.10-17 However, there are various
challenges in accomplishing this task, including the surface rough-
ness of an Al film,18 the pattern of the template,18,19 and the stress
caused by volume expansion.17,18 In particular, the removal of a
bottom barrier layer formed under the nanochannel is one of the
major issues to resolve.10-15 The fact that the thickness of the barrier
layer is approximately half that of the surrounding wall20 highlights
the need for the development of a method to remove this bottom
barrier layer selectively without destroying the entire template struc-
ture.
There have been various attempts to selectively remove the bot-
tom barrier layer of AAO including voltage reduction,21 insertion of
a noble metal underlayer,10 plasma-assisted dry etching,11,12,22 oxide
breakdown by applying overpotentials,13 or application of a reverse
bias.10,14 However, none of these approaches appears to be success-
ful because each of these processes demonstrates some limitations,
such as the continuous diameter reduction in the nanochannels, tem-
plate delamination caused by extensive gas evolution, and the diffi-
culty in applying to high aspect ratio channels. Recently, Oh and
Thompson15 reported a two-step method for the selective removal of
the barrier layer using a W underlayer. They reported that the anod-
ization of an Al film deposited on W resulted in the formation of
WO3 pillars, which are insoluble under anodizing conditions. In
addition, they found that these WO3 pillars could be etched selec-
tively with respect to Al2O3 using a neutral solution based on Pour-
baix’s calculation of solubility.23 This approach is apparently the
most successful way of removing the barrier layer. However, they
did not measure the contact resistance directly and only demon-
strated that they could successfully deposit metal nanowires using
electrochemistry. This paper reports a wet-chemical route for suc-
cessfully removing the bottom barrier layer using TiN as a bottom
electrode, which is quite similar to that reported by Oh and Thomp-
son. The difference is that our approach to finding the candidate
material system is based on the kinetic view-point. Moreover, me-
tallic Ru nanowires were deposited inside the nanochannels by
atomic layer deposition ͑ALD͒, and the electrical properties of the
contact between Ru nanowires and TiN bottom electrode were mea-
sured by conductive atomic force microscope ͑C-AFM͒ in relation
to the degree of barrier layer removal.
Experimental
Both TiN ͑100 nm͒ and Al ͑1 ␮m͒ were deposited successively
in situ on a thermally grown SiO2 100 nm layer using a dc magne-
tron sputter system. The base pressure was approximately 5
ϫ 10−6 Torr, the working pressure was 5 mTorr, and the sputtering
power was 150 W. For the reactive sputtering of TiN, the N2 ͑0.6
sccm͒ supply was increased slowly after Ar ͑6 sccm͒ plasma ignition
at 5 mTorr. When the N2/Ar supply had stabilized, the working
pressure increased slightly to 5.2 mTorr. The resistivity of TiN was
approximately 200–250 ␮⍀ cm. After deposition of the TiN layer,
the N2 supply was turned off while the Ar ͑6 sccm͒ plasma was
maintained for the successive deposition of Al.
Anodization was carried out using 0.3 M oxalic acid as an elec-
trolyte at 10°C with a graphite counter electrode in a two-step
method.24 An electric potential ͑40 V͒ was supplied using a Keithley
2440 sourcemeter ͑Keithley Instruments, Inc.͒. After an initial anod-
ization for 450 s, AAO was removed with a mixture of 6 wt %
phosphoric acid and 1.8 wt % chromic acid at 50°C for 1 h without
stirring. The remaining Al was then anodized until the anodizing
current ͑ϳ6 mA cm−2͒ had decreased to a saturated level ͑approxi-
mately 0.5–1 mA cm−2͒, and the anodization was stopped 200 s
after the current drop point. Finally, a pore widening process was
conducted using a mixture of 6 wt % phosphoric acid and 1.8 wt %
chromic acid at 30°C for 5 min without stirring. The length and the
diameter of the channel were approximately 500 and 50 nm, respec-
tively.
The TiO2 pillars penetrating through the barrier layer under the
AAO channels were selectively removed using standard clean-1
͑SC1͒ solution. A fresh SC1 solution ͑NH4OH:H2O2:H2O = 1:1:5͒
was prepared. The temperature was maintained at 30°C; the sample
was immersed for the desired time without stirring and was rinsed
sufficiently with deionized water. The morphology of the samples
was observed by a scanning electron microscope ͑SEM͒ ͑JSM-
7401F, JEOL͒.
*
Electrochemical Society Active Member.
z E-mail: kibum@snu.ac.kr
Ru was deposited using a thermal ALD process ͑GENI-MP1000,
Downloaded on 2014-06-26 to IP 137.99.31.134 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

K182
Journal of The Electrochemical Society, 156 ͑11͒ K181-K185 ͑2009͒
ASM-Genitech, Inc.͒ with diethylcyclopentadienyl ruthenium
͓Ru͑EtCp͒ ͔ as a precursor and O2 as a reactant to remove the
2
cyclopentadienyl ligands from the Ru precursor. The deposition tem-
perature was 300°C; 1000 ALD cycles ͑consisting of precursor puls-
ing, purging, reactant pulsing, and purging for 5, 5, 10, and 5 s,
respectively, and the nominal growth rate was approximately
0.048 nm cycle−1͒ were performed. The resistivity of the Ru film
was measured as 15 ␮⍀ cm. The Ru film deposited on top of the
AAO template was dry etched using O2 ͑90 sccm͒ and Cl2 ͑10
sccm͒ at a working pressure of 50 mTorr to physically isolate each
Ru nanowire. After dry etching at 200 W for 10 min with an etch
rate of 4 nm/min−1, the AAO was etched slightly to approximately
7 nm with a mixture of 6 wt % phosphoric acid and 1.8 wt %
chromic acid at 30°C to ensure reliable contact of the atomic force
microscope ͑AFM͒ tip to the Ru nanowire during the resistance
measurements. The Ru nanowires grown in an AAO template were
observed by a transmission electron microscope ͑TEM͒ ͑JEM-
3000F, JEOL͒.
The resistance of Ru nanowires was measured by AFM ͑SPA-
400, Seiko Instruments͒. A 25 nm Pt/Ir-coated conducting tip ͑PPP-
ContPt, Nanosensors, Inc.͒ was used for the resistance measure-
ments in contact mode while the surface topography was observed
in tapping mode. To supply the electric potential, a function genera-
tor ͑HP33120A, Agilent Technologies͒ was connected to the nonan-
odized part of the samples with silver paste. The voltage ranges
were Ϫ1 to 1 V, Ϫ200 to 200 mV, Ϫ20 to 20 mV, and Ϫ2 to 2 mV
for the samples etched for 3, 4, 5, and 6 min, respectively. The
current through the conducting AFM tip was measured using a Kei-
thley 6485 picoammeter ͑Keithley Instruments, Inc.͒ The scanned
area was 1 ϫ 1 ␮m, where the distance between each nanowire
was approximately 100 nm ͑center to center͒.
Results and Discussion
TiN is one of the most commonly used materials in the semicon-
ductor industry in areas such as electrodes and as a diffusion barrier
for Al metallization.25-27 An anodizing Al/TiN bilayer was reported
to form amorphous TiO2 pillar arrays penetrating the AAO barrier
layer.28-31 Therefore, to remove this TiO2 and expose the underlying
TiN, it is important to determine a suitable etchant for the selective
removal of TiO2 from TiN and AAO. Initially, an attempt was made
to determine if it can be etched selectively based on the solubility
calculation. The dissolution reaction of TiO2 can be written as
follows23
Figure 1. ͑a͒ Effect of pH on the dissolution of TiO2 and Al2O3 in aqueous
solutions at 25°C ͑Ref. 23͒. ͓͑b͒ and ͑d͔͒ Plan-view and cross-sectional-view
SEM images of the TiN film before etching and ͓͑c͒ and ͑e͔͒ after etching in
SC1 etchant at 30°C for 5 min.
drastically at a lowered temperature of 30°C while it etches under a
normal etching condition of 90°C.33 For example, when sputter-
deposited TiN ͑100 nm͒ was immersed in a SC1 solution at 30°C,
the film thickness was not reduced even after etching for 5 min ͑Fig.
1b-e͒. Instead, the grain boundaries of the TiN columns, which are
easily oxidized to form TiO2,25 were etched selectively to form a
rough surface. This shows that SC1 etchant dissolves TiO2 more
rapidly than TiN. Etch selectivity of TiO2 vs TiN͑ϳ6:1͒ and
AAO͑ϳ40:1͒ was obtained despite the negative prediction based on
the solubility calculation.
During the two-step anodization of an Al ͑1 ␮m͒/TiN ͑100 nm͒
film deposited on a SiO2 ͑100 nm͒/Si substrate, the anodizing cur-
rent decreased after complete aluminum consumption and when the
TiN began to be anodized. To ensure the complete formation of
TiO2 pillars, the anodizing process was maintained for a further 200
s after decreasing the anodizing current. The samples were then
etched with the SC1 solution for various times ͑2–6 min͒ at 30°C.
Figure 2 shows cross-sectional SEM images of AAO with a TiN
underlayer, with and without SC1 etching ͑Fig. 2a-c͒, as well as the
corresponding plan-view SEM images observed after stripping of
the AAO with a mixture of 6 wt % phosphoric acid and 1.8 wt %
chromic acid at 50°C ͑Fig. 2d-f͒. Figure 2a clearly shows that the
bottom of each AAO channel was blocked by the barrier layer, while
Fig. 2b and c reveals a dimple structure under each channel, indi-
cating the successful and selective removal of TiO2. The corre-
sponding plan-view SEM micrographs also show that the TiO2 pil-
TiO2 + 2H+ = TiO2+ + H2O
͓1͔
and
TiO2 + H2O = HTiO−3 + H+
͓2͔
Therefore, the solubility of certain species from hydrated tita-
nium dioxide can be expressed with respect to the pH of the solution
as follows23
log͓TiO2+͔ = − 1.18 − 2 pH
͓3͔
and
log͑HTiO−3͒ = − 18.00 + pH
͓4͔
A solubility diagram can be drawn based on these calculations
and with information on Al2O3 dissolution,23 as shown in Fig. 1a.
The diagram shows that the solubility of TiO2 is always lower than
that of Al2O3 at 25°C and at all pH values, indicating that it is
impossible to dissolve TiO2 selectively from Al2O3 based on this
thermodynamic guideline. However, the etch rate is a kinetic pro-
cess. The fact that one material has a higher solubility than another
does not guarantee a faster etch rate. Therefore, it is more important
to determine the etch rate of TiO2 vs Al2O3 and TiN under a given
etching condition. While surveying various etchants for TiO2 and
Al2O3, the conventional SC1 solution ͑NH4OH:H2O2:H2O
= 1:1:5͒ leaves aluminum hydroxide species on the surface32 while
it etches TiO2.33 The etch rate of TiN using SC1 etchant is reduced
Downloaded on 2014-06-26 to IP 137.99.31.134 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, 156 ͑11͒ K181-K185 ͑2009͒
K183
lars ͑white dots in Fig. 2d͒ were removed gradually with increasing
SC1 etching time. During etching of the TiO2 pillar, the diameter of
the AAO channel remained almost constant, indicating that the SC1
etchant does not etch Al2O3 under these conditions.
To check the successful removal of TiO2 in more detail, Ru
nanowires were deposited in the AAO channels by an ALD
process.8,9,12 Figure 3a and b shows the cross-sectional TEM images
of the interface area between the Ru nanowires and the underlying
TiN with and without TiO2 etching. While a barrier layer exists
between the Ru nanowires and the TiN layer in the samples without
TiO2 etching ͑Fig. 3a͒, the samples with TiO2 etching show Ru
nanowires in direct contact with the TiN layer ͑Fig. 3b͒. The bottle-
neck shape of the Ru nanowires near the contact point reflects the
shape of TiO2 before etching. Therefore, the electrical properties of
contact between Ru and TiN were determined by measuring each
nanowire directly using c-AFM with highly conducting Pt/Ir-coated
probe tips, as shown in Fig. 4a. A constant electric potential was
applied to the TiN layer via the remaining Al at the tip of a sample
that was not immersed in the electrolyte during the anodizing pro-
cess, and the current level through the Ru nanowire was measured.
Figure 4b shows a topographical image of the Ru nanowire array,
which was obtained by a tapping mode of the AFM. The number of
peaks was well matched with the theoretical density of the Ru nano-
wires ͑115 nanowires ␮m−2͒, which indicates that the image shows
an array of Ru nanowires whose tips are exposed as a result of the
slight etching of the top part of AAO. The current mapping image
for the array of Ru nanowires at a constant potential was obtained,
where the Z scale reflects the current level ͑Fig. 4c͒. The distribution
of current peaks in Fig. 4c is well matched with the position of the
Ru nanowires in Fig. 4b, indicating that the measured current was
obtained from each Ru nanowire.
Figure 2. Cross-sectional-view and plan-view SEM images of the ͓͑a͒ and
͑d͔͒ as-prepared sample, ͓͑b͒ and ͑e͔͒ sample etched for 4 min, and ͓͑c͒ and
͑f͔͒ sample etched for 6 min in an SC1 solution at 30°C ͑scale bar
= 200 nm͒.
The optimum etching time was obtained from the current map-
ping images of the samples with various etching times ranging from
Figure 3. Cross-sectional-view TEM image of Ru nanowires deposited in an
AAO template with TiN underlayer ͑a͒ without selective etching of TiO2 and
͑b͒ with selective etching of TiO2 for 5 min ͑scale bar = 100 nm͒.
Figure 4. ͑Color online͒ ͑a͒ Schematic
diagram showing the resistance measure-
ment with conducting AFM, ͑b͒ typical to-
pographic image of the measured sample
͑tapping mode AFM͒, and ͑c͒ current
mapping image of Ru nanowires grown in
AAO template when the barrier layer was
etched for 6 min and the electric potential
͑2 mV͒ was applied to the TiN electrode
͓area is not the same as ͑b͔͒.
Downloaded on 2014-06-26 to IP 137.99.31.134 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

K184
Journal of The Electrochemical Society, 156 ͑11͒ K181-K185 ͑2009͒
Figure 5. ͑Color online͒ ͓͑a͒–͑d͔͒ Current mapping images of Ru nanowires
when the barrier layers were etched at various times. The etching times were
͑a͒ 3, ͑b͒ 4, ͑c͒ 5, and ͑d͒ 6 min. The images were chosen to show similar
average currents ͑0.53, 0.52, 0.85, and 0.60 ␮A, respectively͒ despite the
different applied electric potentials: ͑a͒ V = 500, ͑b͒ 200, ͑c͒ 10, and ͑d͒ 1
mV.
Figure 6. ͑Color online͒ ͑a͒ I-V plots of the measured Ru nanowires in a
sample whose barrier layer was etched for 6 min and ͑b͒ the change in the
average resistances at various etching times.
3 to 6 min ͑Fig. 5a-d͒. While the images were chosen to show a
similar average current per each nanowire ͑0.5–1.0 ␮A͒, the ap-
plied potential decreases abruptly from 500 to 1 mV with increasing
etching time. These results show that the resistance decreases dras-
tically with etching time. The current mapping image of a sample,
whose barrier layer was etched for 6 min showed a similar density
of Ru nanowires to the theoretical value even at the lowest applied
potential ͑1 mV͒. This suggests that it is close to the optimum etch-
ing time of the barrier layer. Therefore, the current–voltage ͑I-V͒
relationship of Ru nanowires with the TiN electrode was investi-
gated in the sample etched for 6 min.
By scanning the same region with five different potentials, the
I-V relation was plotted for each nanowire, and the resistance was
obtained from a linear fit of the current data. Figure 6a shows the
I-V plots of each nanowire representing the lowest and highest re-
sistances out of several tens of nanowires. The linear relationship
between the current and the voltage indicates ohmic contact while
the total resistance varied in the range from 1.98 to 3.52 k⍀. Figure
6b summarizes the resistance measurement for the samples at vari-
ous etching times. Although the I-V relationship in the samples with
partially removed barrier layers did not show ohmic behavior, the
resistance was obtained from the forced linear fit of the data points
for comparison. The resistance clearly decreased with increasing
etching time and became saturated at approximately 2.7 Ϯ 0.3 k⍀
after 6 min. This agrees well with the TiO2 removal observed by
SEM ͑Fig. 2͒. However, this value does not represent the actual
resistance between the Ru nanowires and the TiN underlayer be-
cause the measurements were made in a two-probe configuration. To
check the resistance of the measurement system, the contact resis-
tance between the conducting tip and the planar Ru film with a TiN
underlayer was measured to be approximately 3.4 Ϯ 3.0 k⍀. This
suggests that the actual contact resistance is below the detection
limit of this measurement system.
Conclusion
In summary, the porous AAO template with a TiN bottom elec-
trode was fabricated on a substrate. The SC1 etchant selectively
removed the TiO2 pillars formed at the end of the anodizing process
of Al deposited on top of the TiN layer. The successful etching of
the barrier layer was confirmed electrically by C-AFM with a con-
ducting probe tip to measure the I-V characteristics of the single Ru
nanowire deposited by ALD through the AAO nanochannel array.
The resistance distribution of each Ru nanowire indicated that the
TiO2 barrier layer could be removed uniformly. Clear ohmic behav-
ior was obtained with a contact resistance below a few kilo-ohms.
Overall, it is believed that the suggested process may lead to oppor-
tunities for the application of vertically aligned nanowires fabricated
using an AAO template.
Acknowledgments
This work was supported by the Nano-Systems Institute-National
Core Research Center program of Korea Science and Engineering
Foundation.
Seoul National University assisted in meeting the publication costs of this
article.
References
1. A. Modi, N. Koratkar, E. Lass, B. Wei, and P. M. Ajayan, Nature (London), 424,
171 ͑2003͒.
2. P. L. Taberna, S. Mitra, P. Poizot, P. Simon, and J. M. Tarascon, Nature Mater., 5,
567 ͑2006͒.
3. W. A. De Heer, A. Châtelain, and D. Ugarte, Science, 270, 1179 ͑1995͒.
4. T. Kuykendall, P. J. Pauzauskie, Y. Zhang, J. Goldberger, D. Sirbuly, J. Denlinger,
and P. Yang, Nature Mater., 3, 524 ͑2004͒.
5. T. Shimizu, T. Xie, J. Nishikawa, S. Shingubara, S. Senz, and U. Gösele, Adv.
Downloaded on 2014-06-26 to IP 137.99.31.134 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, 156 ͑11͒ K181-K185 ͑2009͒
K185
Mater. (Weinheim, Ger.), 19, 917 ͑2007͒.
Chem. Mater., 16, 682 ͑2004͒.
6. W. C. Yoo and J. K. Lee, Adv. Mater. (Weinheim, Ger.), 16, 1097 ͑2004͒.
7. M. S. Sander, M. J. Côté, W. Gu, B. M. Kile, and C. P. Tripp, Adv. Mater. (Wein-
heim, Ger.), 16, 2052 ͑2004͒.
8. J. Bachmann, J. Jing, M. Knez, S. Barth, H. Shen, S. Mathur, U. Gösele, and K.
Nielsch, J. Am. Chem. Soc., 129, 9554 ͑2007͒.
9. D. J. Lee, S. S. Yim, K. S. Kim, S. H. Kim, and K. B. Kim, Electrochem. Solid-
State Lett., 11, K61 ͑2008͒.
20. G. E. Thompson and G. C. Wood, in Treaties on Materials Science and Technology,
Vol. 23, J. C. Scully, Editor, p. 266, Academic, New York ͑1983͒.
21. G. Sauer, G. Brehm, S. Schneider, K. Nielsch, R. B. Wehrspohn, J. Choi, H.
Hofmeister, and U. Gösele, J. Appl. Phys., 91, 3243 ͑2002͒.
22. R. Hellin Rico, B. Du Bois, A. Witvrouw, C. Van Hoof, and J. P. Celis, J. Elec-
trochem. Soc., 154, K74 ͑2007͒.
23. M. Pourbaix, Atlas of Electrochemical Equilibria in Aqueous Solutions, National
10. O. Rabin, P. R. Herz, Y. M. Lin, A. I. Akinwande, S. B. Cronin, and M. S.
Association of Corrosion Engineers, Houston ͑1974͒.
Dresselhaus, Adv. Funct. Mater., 13, 631 ͑2003͒.
11. S. Shingubara, O. Okino, Y. Murakami, H. Sakaue, and T. Takahagi, J. Vac. Sci.
Technol. B, 19, 1901 ͑2001͒.
12. W. H. Kim, S. J. Park, J. Y. Son, and H. Kim, Nanotechnology, 19, 045302 ͑2008͒.
13. M. M. Crouse, A. E. Miller, D. T. Crouse and A. A. Ikram, J. Electrochem. Soc.,
152, D167 ͑2005͒.
24. H. Masuda and M. Satoh, Jpn. J. Appl. Phys., Part 2, 35, L126 ͑1996͒.
25. W. Sinke, G. P. A. Frijlink, and F. W. Saris, Appl. Phys. Lett., 47, 471 ͑1985͒.
26. N. Kumar, J. T. McGinn, K. Pourrezaei, B. Lee, and E. C. Douglas, J. Vac. Sci.
Technol. A, 6, 1602 ͑1988͒.
27. K. C. Park and K. B. Kim, J. Appl. Phys., 80, 5674 ͑1996͒.
28. P. L. Chen, C. T. Kuo, T. G. Tsai, B. W. Wu, C. C. Hsu, and F. M. Pan, Appl. Phys.
Lett., 82, 2796 ͑2003͒.
14. M. Tian, S. Xu, J. Wang, N. Kumar, E. Wertz, Q. Li, P. M. Campbell, M. H. W.
Chan, and T. E. Mallouk, Nano Lett., 5, 697 ͑2005͒.
29. S. Gwo, C. L. Yeh, P. F. Chen, Y. C. Chou, T. T. Chen, T. S. Chao, S. F. Hu, and
15. J. Oh and C. V. Thompson, Adv. Mater. (Weinheim, Ger.), 20, 1368 ͑2008͒.
16. S. H. Jung, S. H. Jeong, S. U. Kim, S. K. Hwang, P. S. Lee, K. H. Lee, J. H. Ko,
E. Bae, D. Kang, W. Park, et al., Small, 1, 553 ͑2005͒.
T. Y. Huang, Appl. Phys. Lett., 74, 1090 ͑1999͒.
30. I. Milošv, H. H. Strehblow, B. Navinšek, and M. Metikoš-Huković, Surf. Interface
Anal., 23, 529 ͑1995͒.
17. A. D. Franklin and D. B. Janes, Appl. Phys. Lett., 92, 013122 ͑2008͒.
18. S. H. Park, Y. W. Lee, J. K. Lee, and K. B. Kim, Electrochem. Solid-State Lett., 9,
D31 ͑2006͒.
31. S. Z. Chu, S. Inoue, K. Wada, S. Hishita, and K. Kurashima, Adv. Funct. Mater.,
15, 1343 ͑2005͒.
32. W. Kern, J. Electrochem. Soc., 137, 1887 ͑1990͒.
19. D. A. Brevnov, M. Barela, M. E. Piyasena, G. P. López, and P. B. Atanassov,
33. S. Verhaverbeke and J. W. Parker, Mater. Res. Soc. Symp. Proc, 466, 447 ͑1997͒.
Downloaded on 2014-06-26 to IP 137.99.31.134 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).