APPLIED PHYSICS LETTERS 89, 063123 共2006兲
Growth of RuO2 nanorods in reactive sputtering
Yu-Tsun Lin, Chun-Yu Chen, Chang-Po Hsiung, Kai-Wen Cheng, and Jon-Yiew Gana兲
Department of Materials Science and Engineering, National Tsing Hua University, Hsinchu City, Taiwan
30043, Republic of China
共Received 16 April 2006; accepted 29 June 2006; published online 11 August 2006兲
The synthesis of RuO2 nanorods with reactive sputtering was demonstrated in this work. The
synthesis process is very much like the metal organic chemical vapor deposition, except that RuO3
generated with reactive sputtering under high oxygen-to-argon flow ratio 共⬎5 SCCM/ 15 SCCM兲
共SCCM denotes cubic centimeter per minute at STP兲 and high substrate temperature 共⬎300 ° C兲 is
used in place of the metal organic precursor. RuO2 nanorods tend to grow steadily with constant
aspect ratio 共⬃27兲 and the field-emission characteristics appear very sensitive to their spatial
distribution. © 2006 American Institute of Physics. 关DOI: 10.1063/1.2269180兴
Over the last few years, there has been increasing interest to synthesize materials into one-dimensional 共1D兲 nanostructures such as nanotubes, nanowires, and nanorods.1,2
The peculiar structure and size of 1D nanostructures often
render materials with enhanced physical and chemical properties that are useful in various fields of application such as
light emission, field emission, electrochemical sensing, and
catalysis.3–5 Among the metal oxides, 1D synthesis of ruthenium dioxide RuO2 has recently caught attention. RuO2 is a
rutile-type tetragonal oxide that exhibits the metal-like electrical conductivity 共⬃40 ␮⍀ cm兲 and excellent thermal and
chemical stability. It is one of a few available electrode materials used in various electrochemical processes and ferroelectric thin film capacitors.6–9 Owing to its oxide nature and
high electrical conductivity, there also has been a consistent
interest in 1D RuO2 that represents a potential field emitter
and possibly an excellent nanoelectrode for electrochemical
sensing and catalysis.10–12
Recently, Hsieh et al. have demonstrated a direct deposition of RuO2 nanorods using the metal organic chemical
vapor deposition 共MOCVD兲.12 These nanorods are faceted
crystals and their formation has been attributed to the twodimensional growth of nuclei.1 Such a growth mode often
requires the participating reactants to have sufficient surface
mobility so that they can move easily to the ledges for incorporation rather than forming another nucleus on growing surfaces. This is particularly easy to achieve with MOCVD
wherein the metal organic precursor is introduced as a stable
vapor species at relatively low temperature, and therefore
possesses appreciable surface mobility.
It is interesting to note that, besides the metal organic
precursor, Ru hyperoxides such as RuO3 and RuO4 are also
low-temperature vapor species; consequently, they present
another potential precursor for RuO2 nanorod synthesis. Here
we report the synthesis of RuO2 nanorods using the reactive
sputtering to generate RuO3. The synthesis was found sensitive to the substrate temperature and detailed structure of the
growth tip. A dramatic dependence of field emission on the
morphology of RuO2 nanorods was also demonstrated. The
technology is of practical interest and has been extended for
the fabrication of more complicated 1D heterostructures.13
a兲
Author to whom correspondence should be addressed; FAX: 886-35722366; electronic mail: [email protected]
Experimentally, the growth of RuO2 was performed in a
radio-frequency magnetron sputtering system. The sputtering
target was a 2 in. Ru metal 共99.9% pure兲 and 4 in. thermally
oxidized Si wafers having 200 nm thick SiO2 were used as
substrates. Reactive sputtering was carried out in various gas
mixtures of Ar and O2 and substrate temperature, but at the
same working pressure 共10−2 torr兲 and rf power 共20 W兲.
Chemical analysis of the deposited RuO2 was performed
with the x-ray photoelectron spectroscopy 共XPS兲 共PHI 1600兲
using aluminum anode. Morphology of RuO2 nanorods was
examined with a field-emission scanning electron microscope 共JEOL-JSM 6500F兲. The detailed crystal structure of
RuO2 nanorods was analyzed with a transmission electron
microscope 共JEM-2010兲. The field-emission property was
measured in a vacuum chamber of 1 ⫻ 10−7 torr, using a
movable stainless-steel probe with diameter of 1 mm as anode and a 100 ␮m spacing between the anode and emitting
surface.
There is some indication that RuO3 may be formed in
reactive sputtering under high oxygen flow. Figure 1 is the
Ru 3d XPS spectra of two samples deposited under two different oxygen-to-argon flow ratios 共O2 / Ar= 3 / 17 and
5 / 15 SCCM/ SCCM兲 共SCCM denotes cubic centimeter per
minute of STP兲 but at the same substrate temperature
共450 ° C兲. For low O2 / Ar flow ratio 共3 SCCM/ 17 SCCM兲,
the deconvolution of Ru 3d spectra suggests a significant
amount of Ru metal 共44%兲 besides RuO2 共56%兲. In contrast,
RuO3 of 18% and none of Ru metal was detected from the
FIG. 1. XPS analysis of RuO2 deposited with reactive sputtering under the
oxygen-to-argon flow ratios of 共a兲 O2 / Ar= 3 / 17 and 共b兲 O2 / Ar
= 5 / 15 共SCCM/ SCCM兲.
0003-6951/2006/89共6兲/063123/3/$23.00
89, 063123-1
© 2006 American Institute of Physics
Downloaded 19 Dec 2010 to 140.114.66.106. Redistribution subject to AIP license or copyright; see http://apl.aip.org/about/rights_and_permissions
063123-2
Appl. Phys. Lett. 89, 063123 共2006兲
Lin et al.
FIG. 2. Morphology of RuO2 deposited with reactive
sputtering at the substrate temperatures of 共a兲 200, 共b兲
300, and 共c兲 450 ° C.
sample with high O2 / Ar flow ratio 共5 SCCM/ 15 SCCM兲. At
high oxygen flow, RuO3 is the dominant sputtered species,
instead of RuO2, possibly because it is a stable vapor species
and thus easier to desorb under the oxygen impingement.
Similar result has also been suggested by other groups based
on the chemical analysis and the drastic change of deposition
rate with the oxygen flow rate.14
In reactive sputtering, RuO2 nanorods began to appear
with O2 / Ar flow ratio of 5 SCCM/ 10 SCCM and grew better by raising O2 / Ar flow ratio up to 10 SCCM/ 10 SCCM.
Nevertheless, the synthesis was also found sensitive to the
substrate temperature, as illustrated in Fig. 2. In the figure,
RuO2 morphology varies drastically with the substrate temperature. At low substrate temperature 共200 ° C兲, RuO2 was
deposited as a relatively smooth film, then as a pile of short
rods at the substrate temperature of 300 ° C, and finally became the perfectly faceted nanorods at 450 ° C. Since high
surface mobility is required in the growth of faceted structure, the result evidently suggests a strong dependence of
surface mobility on temperature, which is legitimate because
surface motion of sputtered species is normally a thermally
activated process.
Figure 3 shows the lattice image of one RuO2 nanorod.
In the figure, the facets that formed the sidewalls of nanorod
were identified with the diffraction and lattice spacing to be
共010兲 and 共001兲, which are the planes of low surface energy.
This is expected by the growth model which suggests that,
without kinetic constraint, crystals tend to grow into the
shape having the surface energy minimized. In addition, the
figure also reveals that it is atomically sharp and smooth on
the sidewalls of the rod, while there are plenty of ledges and
steps clearly visible on the tip. Ledges and steps facilitate the
incorporation; consequently, the tip presents a higher growth
rate than the sidewalls.
Figure 4 is the cross-section views and the corresponding field-emission characteristics of RuO2 nanorods grown
with different lengths of time. As illustrated, the “layer thickness” or the apparent height of RuO2 nanorods appears to
increase steadily with time. We have roughly estimated the
length-to-width aspect ratio of nanorods whose length is
close to the apparent height shown in the micrographs. The
estimated aspect ratio is about 1000/ 37 共nm/ nm兲 for the
0.5 h growth, 2000/ 74 共nm/ nm兲 for the 2 h growth, and
3700/ 133 共nm/ nm兲 for the 3 h growth. The almost identical
aspect ratio 共⬃27兲 suggests that RuO2 nanorods remain their
shape during the growth, which is legitimate in order to keep
the faceted structure and therefore the minimal increase of
surface energy. It is also interesting to note that, by increasing the growth time, nanorods seem to become more sparsely
distributed when viewed at the apparent-height level. This
may happen as some nanorods are slanted too much off the
surface normal such that their growth is terminated by bumping into others, and hence leaves more room to others.
Figure 4共b兲 presents plots of current density as a function of electric field applied for nanorods grown with different times. The plots show a sharp turning point of electric
field 共threshold field Eth兲 below which thermionic emission
dominates and the current has negligible dependence on
electric field. Above the threshold field, field emission, however, starts to dominate and the current becomes a strong
function of electric field as described by Fowler-Nordheim
共FN兲 equation,15
J=
冉
冊 冉
冊
A␤2 2
B␾3/2
E exp −
,
␾
␤E
where A = 1.42⫻ 10−10 ⫻ exp共10.4/ ␾1/2兲 in units of
A V−2 eV, ␤ is the field enhancement factor, ␾ is the work
function 共⬃4.87 eV for RuO2兲,12 and B = 6.44⫻ 103 in units
FIG. 4. 共a兲 Cross-section views and 共b兲 field-emission characteristics of
RuO2 nanorods grown for 0.5, 2, and 3 h at the same O2 / Ar flow ratio of
FIG. 3. Lattice images around the growth tip of a RuO2 nanorod.
10/ 10 共SCCM/ SCCM兲 and substrate temperature of 450 ° C.
Downloaded 19 Dec 2010 to 140.114.66.106. Redistribution subject to AIP license or copyright; see http://apl.aip.org/about/rights_and_permissions
063123-3
Appl. Phys. Lett. 89, 063123 共2006兲
Lin et al.
of 共eV兲−3/2 V ␮m−1. The threshold field Eth and field enhancement factor ␤ derived with the FN equation have also
been included in the figure. As shown, Eth and ␤ tend to vary
oppositely with the growth time, but their product 共␤Eth兲 is
pretty much a constant 共1337± 33 V / ␮m兲 because of the
negligible field dependence of thermionic emission. Field
emission of RuO2 nanorods depends strongly on the geometry and chemistry of emitting surface. The former is characterized with ␤ whose value is a function of aspect ratio and
spatial distribution of field emitters.16 The variation ␤ is evidently caused by the change of spatial distribution with time
because the aspect ratio has been shown nearly a constant.
Besides the geometry effect, the surface-sensitive term of
field emission has also been characterized with 共B␾3/2 / ␤兲
shown in the FN equation.17 The value of 共B␾3/2 / ␤兲 derived
in this work 共214–281兲 is close to that of RuO2 nanorods as
prepared with MOCVD 共235–318兲, suggesting that RuO2 nanorods fabricated from both technologies appear with similar
surface chemistry. Finally, it is noted that a small increase of
␤ may drastically improve the current drive capability of
RuO2 nanorods. For example, the current of the 3-h-growth
sample has been increased by five orders of magnitude at the
threshold field of the 0.5-h-growth one 共5.3 V / ␮m兲.
To summarize, we have demonstrated the synthesis of
RuO2 nanorods with reactive sputtering in this work. The
synthesis process is very much like the MOCVD, except that
RuO3 generated with reactive sputtering has been used instead of the metal organic precursor. RuO2 nanorods appear
to grow steadily with constant aspect ratio and their spatial
distribution may drastically affect the field emission characteristics.
The authors would like to acknowledge the support from
Ministry of Education 共A-94-E-FA104-4兲 and National Science Council of Taiwan, Republic of China 共NSC95-2216-E007-007兲.
1
X. Xia, P. Yang, Y. Sun, Y. Wu, B. Mayers, B. Gates, Y. Yin, F. Kim, and
H. Yan, Adv. Mater. 共Weinheim, Ger.兲 15, 353 共2003兲.
2
Z. R. Dai, Z. W. Pan, and Z. L. Wang, Adv. Funct. Mater. 13, 9 共2003兲.
3
Y. Li, G. W. Meng, L. D. Zhang, and F. Phillipp, Appl. Phys. Lett. 76,
2011 共2000兲.
4
Y.-K. Tseng, C.-J. Huang, H.-M. Cheng, I.-N. Lin, K.-S. Liu, and I.-C.
Chen, Adv. Funct. Mater. 13, 811 共2003兲.
5
A. Kolmakov, D. O. Klenov, Y. Lilach, S. Stemmer., and M. Moskovits,
Nano Lett. 4, 667 共2005兲.
6
D. E. Stilwell and S. M. Park, J. Electrochem. Soc. 135, 2254 共1988兲.
7
J. Rolewicz, C. Comninellis, E. Plattner, and J. Hinden, Electrochim. Acta
33, 573 共1988兲.
8
L. Krusin-Elbaum, Thin Solid Films 169, 17 共1989兲.
9
L. Krusin-Elbaum and M. Wittmer, J. Electrochem. Soc. 135, 2610
共1988兲.
10
J. V. Ryan, A. D. Berry, M. L. Anderson, J. W. Long, R. M. Stroud, V. M.
Cepak, V. M. Browning, D. R. Rolison, and C. I. Merzbacher, Nature
共London兲 406, 169 共2000兲.
11
C. Ducati, D. H. Dawson, J. R. Saffell, and P. A. Midgley, Appl. Phys.
Lett. 85, 5385 共2004兲.
12
C. S. Hsieh, D. S. Tsai, R. S. Chen, and Y. S. Yuang, Appl. Phys. Lett. 85,
3860 共2004兲.
13
K. W. Cheng, Y. T. Lin, C. Y. Chen, C. P. Hsiung, J. Y. Gan, J. W. Yeh, C.
H. Hsieh, and L. J. Chou, Appl. Phys. Lett. 88, 043115 共2006兲.
14
Y. Kaga, Y. Abe, H. Yanagisawa, and K. Sasaki, Jpn. J. Appl. Phys., Part
1 37, 3457 共1998兲.
15
D. Temple, Mater. Sci. Eng., R. 24, 185 共1999兲.
16
V. Filip, D. Nicolaescu, M. Tanemura, and F. Okuyama, Ultramicroscopy
89, 39 共2001兲.
17
C. S. Hsieh, G. Wang, D. S. Tsai, R. S. Chen, and Y. S. Yang,
Nanotechnology 16, 1885 共2005兲.
Downloaded 19 Dec 2010 to 140.114.66.106. Redistribution subject to AIP license or copyright; see http://apl.aip.org/about/rights_and_permissions