Selective-area atomic layer deposition using poly(vinyl pyrrolidone) as a passivation layer

Article abstract of DOI:10.1149/1.3250936

Selective-area atomic layer deposition (ALD) was studied using poly(vinyl pyrrolidone) (PVP) films as growth-preventing mask layers. The PVP films were prepared by spin coating and patterned by UV lithography. The PVP films were tested in several ALD processes: iridium, platinum, ruthenium, Al ₂O₃, and ZrO₂. The deposition temperatures were 250-300°C. In general, the PVP film passivated the surface against the noble metal processes, but the oxide films grew on PVP. However, the oxide films did not grow through the PVP film on the substrate surface and, therefore, the films could still be patterned, though with more of a lift-off method rather than with pure selective-area ALD.

Full text of DOI:10.1149/1.3250936

K10
Journal of The Electrochemical Society, 157 ͑1͒ K10-K14 ͑2010͒
0
013-4651/2009/157͑1͒/K10/5/$28.00 © The Electrochemical Society
Selective-Area Atomic Layer Deposition
Using Poly(vinyl pyrrolidone) as a Passivation Layer
z
Elina Färm, Marianna Kemell, Eero Santala,* Mikko Ritala,** and
Markku Leskelä
Department of Chemistry, University of Helsinki, P. O. Box 55, FI-00014 Helsinki, Finland
Selective-area atomic layer deposition ͑ALD͒ was studied using poly͑vinyl pyrrolidone͒ ͑PVP͒ films as growth-preventing mask
layers. The PVP films were prepared by spin coating and patterned by UV lithography. The PVP films were tested in several ALD
processes: iridium, platinum, ruthenium, Al O , and ZrO . The deposition temperatures were 250–300°C. In general, the PVP film
2
3
2
passivated the surface against the noble metal processes, but the oxide films grew on PVP. However, the oxide films did not grow
through the PVP film on the substrate surface and, therefore, the films could still be patterned, though with more of a lift-off
method rather than with pure selective-area ALD.
©
2009 The Electrochemical Society. ͓DOI: 10.1149/1.3250936͔ All rights reserved.
Manuscript submitted June 22, 2009; revised manuscript received September 21, 2009. Published November 3, 2009.
1
-3
Atomic layer deposition ͑ALD͒ is a method to grow thin films
layer by layer through self-limiting surface reactions between alter-
nately supplied gaseous precursors. The film thickness and compo-
sition can be controlled with atomic layer accuracy in the growth
direction. With selective-area ALD, the film growth can also be
controlled on the surface. Selective-area ALD requires that desig-
nated areas of the surface are passivated or protected against ALD
growth, in which case the film is deposited only on the desired parts
of the surface.
small patterns are desired. Thin PMMA mask layers ͑about 100 nm
thick͒ can also be used at temperatures higher than 150°C, but the
sizes of the patterns that can be made are quite large, i.e., tens of
2
3
micrometers.
Experimental
PVP ͑Mw = 1,300,000͒ ͑from Alfa Aesar͒ films were used as
passivation layers for selective-area ALD. The PVP films were pre-
pared on silicon ͑100͒ surfaces from 1 wt % PVP-ethanol solution
by spin coating. After spin coating, the samples were baked in an
oven at 100°C for 60 min to ensure removal of the solvent. The PVP
film thicknesses were 75–120 nm, as determined by fitting the opti-
cal reflectance spectra measured by a Hitachi U-2000 spectropho-
tometer.
The PVP films were patterned by UV lithography through a me-
chanical mask. The mask had three sizes of holes, the diameters of
which were 500, 250, and 50 ␮m. Patterning was done by a UV
lamp that produced UV light at wavelengths of 264 and 185 nm ͑the
power of the lamp was 10 W͒. The PVP removal was accomplished
just by UV-induced decomposition in air; no further development
was done. The patterning time was typically 12 h. After the ALD
process, the PVP film was removed by ultrasonication in warm wa-
ter.
Selective-area ALD has commonly been studied by using pat-
4,5
terned self-assembled monolayers ͑SAMs͒ as growth-preventing
mask layers. SAMs can be formed spontaneously through adsorp-
tion on the solid surface from liquid and gas phases. The function of
the SAM is to passivate the surface against ALD growth so that the
film is deposited only on the areas without a SAM. Patterned SAMs
can be prepared directly by microcontact printing and from con-
tinuous SAMs by destroying the SAM from selected areas with
energetic ions, electrons, or photons with the aid of either mechani-
cal masks or focused beams or by scanning-probe-based
6
,7
8
lithography. Patterned SAMs have been used for selective-area
ALD of HfO , TiO , ZnO, ZrO , ruthenium, platinum, iridium, and
2
2
2
9
-19
cobalt.
Another approach to selective-area ALD is to use unreactive and
thermally stable polymer films as growth-preventing mask layers.
Compared to SAMs, polymer films are fast and simple to prepare by
spin coating, and the polymer film can be easily removed after an
ALD process. Polymer films can work in selective-area ALD the
same way as SAMs, i.e., the film can passivate the surface against
ALD growth. Alternatively, the ALD film can grow on the polymer
film while the growth on the substrate surface can still be prevented
by controlling the thickness of the polymer film and the pulse and
purge times. Under these conditions, lift-off patterning may be fea-
sible provided that the film grown on the polymer does not encap-
sulate it too well. Polymeric films such as poly͑methyl methacrylate͒
Nanometer scale patterns were tested using PVP fibers as masks.
The fibers were electrospun from 7 wt % PVP-ethanol solution. The
solution was placed in a plastic syringe with a metallic needle. The
needle was connected to a high voltage supply, and the Si͑100͒
substrate was grounded. After electrospinning, the fibers were dried
at 120°C for 1 h.
The patterned PVP samples were tested in several ALD pro-
2
5
cesses. Iridium
was deposited from
Ir͑acac͒3
͑acac
= 2,4-pentanedione͒ and O2 at 250 and 300°C for 1000 cycles,
2
6
platinum was deposited from MeCp͑cyclopentadienyl͒PtMe3 and
2
7
O at 300°C for 700 cycles, and ruthenium was deposited from
2
2
0-23
24
͑
PMMA͒
and poly͑tert-butyl methyl acrylate͒ have been stud-
RuCp and air at 300°C for 1000 cycles. Al O was deposited at
2
2
3
ied in selective-area ALD as a mask layer against the growth of
250°C for 500 cycles from AlCl and H O and from trimethylalu-
3
2
20-24
28
TiO , Al O , Ir, Ru, and Pt.
minum ͑TMA͒ and H O. ZrO
was deposited from ZrCl4 and
2
2
3
2
2
In this study, selective-area ALD was studied using a poly͑vinyl
pyrrolidone͒ ͑PVP͒ polymer film as the growth-preventing mask
layer. The suitability of PVP as the mask layer was tested in several
noble metal and oxide processes. The PVP film was chosen as the
mask layer because of its higher melting point ͑300°C͒ compared to
PMMA ͑150°C͒. A higher melting point allows the use of the PVP
mask in a wider range of temperatures and, thus, the selection of
possible ALD processes increases. This is important especially when
H O at 250°C for 1000 cycles. The films were grown in an F120
2
reactor ͑ASM Microchemistry, Ltd., Finland͒ and a SUNALE reac-
tor ͑Picosun Oy, Finland͒. Nitrogen was used as a carrier and purg-
ing gas. The reactors were operated at pressures of about 10 mbar.
The patterned films were studied with a Hitachi S-4800 field-
emission-scanning electron microscope and an INCA 350 energy
dispersive X-ray ͑EDX͒ spectrometer. The film thicknesses were
calculated from the EDX results using a GMR electron-probe thin-
2
9
film microanalysis program and bulk densities for Ir, Pt, Ru, and
ZrO . The thicknesses of the Al O films were calculated using a
2
2
3
3
density of 2.8 g/cm because amorphous Al O films deposited by
ALD had lower densities than the crystalline Al O . Film thick-
2
3
*
Electrochemical Society Student Member.
3
0
*
* Electrochemical Society Active Member.
2
3
z
E-mail: elina.farm@helsinki.fi
nesses were measured from the dots that had a diameter of 500 ␮m.
Downloaded on 2014-11-09 to IP 132.174.255.116 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, 157 ͑1͒ K10-K14 ͑2010͒
K11
b)
b) 3
000
000
000
0
Ru dot
5
4
3
2
1
0000
0000
0000
0000
0000
0
PVP film after ALD process
Ir dot
below PVP film after ALD process
PVP film after ALD process
below PVP film after ALD process
Si
2
1
Si
C
Ru
O
Ir
C
O
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
0.0
0.5
1.0
1.5
2.0
2.5
3.0
Energy keV
Energy keV
Figure 2. ͑Color online͒ ͑a͒ SEM micrograph of a patterned ruthenium film
deposited at 300°C. The lighter areas are ruthenium and the darker areas are
silicon. ͑b͒ EDX spectra measured on a Ru dot, on the PVP film after the Ru
process, and on the Si surface after PVP removal.
Figure 1. ͑Color online͒ ͑a͒ SEM micrograph of a patterned iridium film
deposited at 300°C. The lighter areas are iridium and the darker areas are
silicon. ͑b͒ EDX spectra measured on an Ir dot, on the PVP film after the Ir
process, and on the Si surface after PVP removal.
of these films was tens of hours. With 100 nm thick films prepared
from 1 wt % solution, patterning could be done in 12 h. During this
time, most of the PVP could be removed from the surface. Atomic
force microscopy ͑AFM͒ studies showed that after patterning, the
patterned surface had a few scattered particles. Sizes of these par-
ticles were about 5 nm in height and less than 100 nm in width.
Otherwise, the patterned area seemed to be clear of PVP. The par-
ticles may be traces of PVP or impurities from the PVP solution that
cannot be destroyed by UV. The thicknesses of the patterned noble
metal films were somewhat lower than the thicknesses of reference
films grown on bare Si͑100͒ substrates ͑see below͒ in the same ALD
process. Nevertheless, the results were repeatable.
Figure 1a shows a scanning electron microscopy ͑SEM͒ micro-
graph of a patterned iridium film grown at 300°C. In the figure, the
PVP film is removed, and the lighter areas are iridium and the darker
areas are silicon. The thickness of the patterned iridium film was 17
nm, and the thickness of the reference film on Si͑100͒ was 21 nm.
PVP film residues probably retard the iridium growth.
Results
PVP mask layers for selective-area ALD were prepared by spin
coating and patterned by a simple UV-induced decomposition pro-
cedure by illuminating through a mechanical mask that had holes
with diameters of 500, 250, and 50 ␮m. The patterning time was 12
h because after UV illumination, no development solution was ap-
plied. With a more powerful lamp, the patterning time could surely
be shortened. After ALD, the PVP film was removed by water im-
mersion. This patterning procedure allows testing of the patterned
PVP films in selective-area ALD. In case of successful selectivity
results, a higher accuracy can be looked for straightforwardly by
improving the patterning method. Nanometer scale patterning was,
however, tested using a PVP fiber as a mask, as discussed below.
The PVP film passivated the surface against the ALD growth of
all the tested noble metals. EDX measurements in Fig. 1b, 2b, and
3
b show that the noble metals were grown only on patterned areas of
the sample, i.e., areas from where the PVP had been removed by UV
lithography, but no growth occurred on PVP. As expected, the sur-
face below the PVP film also stayed without a noble metal, as veri-
fied by EDX measurements after the removal of the PVP film.
The PVP film prevented the ALD growth so well that even pos-
sible traces of PVP on patterned areas inhibited or retarded the
growth of noble metals. When the PVP film was made from 2 wt %
solution, the film thickness was about 200 nm. The patterning time
Ruthenium films were deposited at 300°C for 1000 cycles. Fig-
ure 2a shows an SEM micrograph of a patterned ruthenium film.
The thickness of the ruthenium film on the patterned areas was 20
nm, and the thickness of the reference film on silicon was 30 nm.
Like in iridium, the PVP film probably retarded the film growth, but
the growth rate of Ru on the PVP patterned samples was repeatable.
ALD noble metal films are typically deposited on a freshly de-
2
6,27
posited oxide seed layer to ensure good nucleation.
The absence
Downloaded on 2014-11-09 to IP 132.174.255.116 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

K12
Journal of The Electrochemical Society, 157 ͑1͒ K10-K14 ͑2010͒
b) 3
000
000
000
0
Pt dot
below PVP film after ALD process
PVP film after ALD process
Si
2
1
C
Pt
O
Al
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
Energy keV
Figure 3. ͑Color online͒ ͑a͒ SEM micrograph of a patterned platinum film
deposited at 300°C. Al O seed layer was deposited before platinum. The
2
3
lighter areas are platinum and the darker areas are silicon. ͑b͒ EDX spectra
measured on a Pt dot, on the PVP film after the Pt process, and on the Si
surface after PVP removal.
of this seed layer may also trouble the ruthenium growth especially
on the patterned areas where possible traces of PVP may also disturb
the growth. Similar results were noticed when a PMMA film was
2
3
used as a mask layer to grow patterned ruthenium films at 300°C.
In platinum, the PVP residues are even more detrimental than in
ruthenium. Without the seed layer, the PVP film passivated the sur-
face against the platinum growth, but the film did not grow on the
patterned areas either. Whereas a ruthenium film could be grown
without a seed layer, an Al O seed layer was needed to deposit a
2
3
platinum film on the patterned areas. Platinum growth on patterned
areas was achieved when 10 cycles of Al O was deposited before
2
3
platinum ͑Fig. 3͒. Al O film also grows on PVP, enabling a subse-
2
3
quent platinum growth ͑Fig. 3b͒. The Al O3 film did not grow
2
through the PVP film to the substrate surface, however, and the PVP
film, together with the Pt and Al O films on top, could be removed
2
3
after the ALD process. The thickness of the platinum film on the
patterned area was 24 nm.
The PVP film acts differently with the tested ALD oxide pro-
cesses than with the noble metal processes. EDX measurements in
Fig. 4b and d and 5b show that the Al O3 and ZrO films were
grown on patterned areas as well as on the PVP film. However, the
oxide films did not grow on the silicon surface below the PVP film.
This means that the PVP film and the films deposited onto it can be
removed after the ALD processes, and the result is an ALD film
patterned by a lift-off process. The PVP residues did not inhibit the
Figure 4. ͑Color online͒ ͑a͒ SEM micrograph of a patterned Al O film from
2
3
AlCl3 and H O deposited at 250°C. The lighter areas are Al O and the
2
2
3
2
2
darker areas are silicon. ͑b͒ EDX spectra measured on an Al O dot, on the
2
3
PVP film after the Al O process, and on the Si surface after PVP removal.
2
3
͑
c͒ SEM micrograph of a patterned Al O film from TMA and H O depos-
2
3
2
ited at 250°C. ͑d͒ EDX spectra measured on an Al O dot, on the PVP film
2
3
after the Al O process and on the Si surface after PVP removal.
2
3
Downloaded on 2014-11-09 to IP 132.174.255.116 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, 157 ͑1͒ K10-K14 ͑2010͒
K13
difficulties in removing the polymer films after the ALD process,
selective-area ALD is most conveniently achieved with PVP as the
mask layer and AlCl as an aluminum precursor.
3
ZrO was deposited from ZrCl and H O. The ALD film grew on
2
4
2
PVP and probably also inside the PVP film, judging from the time-
consuming removal of the PVP film. The PVP film could, however,
be removed after the ALD process, and the ZrO did not grow below
2
the PVP film ͑Fig. 5b͒. The film removal took hours in hot water.
The thickness of the patterned ZrO film was 80 nm.
2
PVP mask layers could be used up to its melting point of 300°C
but not above that. When tested with Ru ALD at 350°C, the original
patterns in the PVP mask changed, and the film was grown ran-
2
3
domly to the sample. In our earlier study, the PMMA mask layer
could be used at higher temperatures than its melting point ͑150°C͒
when the sizes of patterns were tens of micrometers and when the
2
3
mask layer was about 100 nm thick. But nanometer scale patterns
disappeared at temperatures above 150°C, though in that case, a
thicker layer of PMMA mask with a different molecular weight was
used. Apparently, the molecular weight and the thickness of the
polymer film and possibly the sizes of patterns influence the stability
of the mask layer at different temperatures.
When the PVP film was patterned through a mechanical mask by
UV illumination, the sizes of the patterned dots were tens of mi-
crometers. To see if nanometer scale patterns are possible to achieve
using a PVP mask, electrospun PVP fibers were tested as mask
layers as well. The iridium film was deposited on the sample at
b)
2
5000
0000
5000
0000
ZrO2 dot
PVP film after ALD process
Below PVP film after ALD process
2
1
1
Si
O
2
50°C by 1000 cycles, and the fibers were removed after the ALD
Zr
process by ultrasonication in warm water.
5
000
C
Figure 6a shows an SEM micrograph of a patterned iridium film.
The iridium film was grown only on areas without the PVP, and the
fibers prevented the growth. The width of the area without iridium is
0
2
00 nm. This indicates that nanometer scale patterns are possible to
0
.5
1.0
1.5
2.0
2.5
3.0
reach using PVP film as a passivation layer. When spin-coated PVP
films were used as passivation layers, the iridium film thickness on
the patterned samples was less than the Ir film thickness on the
reference Si͑100͒. But when the PVP fiber was used for passivation,
the iridium film thickness was the same on the patterned sample and
on the reference Si͑100͒. This is probably due to a smaller amount
of iridium growth disturbing PVP in these samples. Also, different
from the samples made with the spin-coated PVP films, in the
samples patterned with the fibers, the growth areas were never cov-
ered by PVP and thus could stay free from PVP residues.
Energy keV
Figure 5. ͑Color online͒ ͑a͒ SEM micrograph of a patterned ZrO2 film
deposited at 250°C. The lighter areas are ZrO2 and the darker areas are
silicon. ͑b͒ EDX spectra measured on a ZrO dot, on the PVP film after the
ZrO process, and on the Si surface after PVP removal.
2
2
growth of the oxides on the patterned areas, where the film thick-
nesses were the same as those on the nonpatterned reference
samples.
Conclusion
Selective-area ALD was studied using PVP films as mask layers.
Mask layers were prepared by spin coating and patterned through a
mechanical mask by UV lithography. PVP films were tested in iri-
dium, ruthenium, platinum, Al O , and ZrO2 processes at
Al O films were deposited at 250°C from two precursor com-
2
3
2
3
binations: AlCl and H O and TMA and H O. In our earlier study,
3
2
2
a PMMA film was tested as a mask layer for Al O . After the Al O
2
3
2
3
2
3
process with AlCl and H O as precursors, the PMMA film was very
3
2
2
50–300°C. The PVP film acted differently with the tested noble
metal and oxide processes. The PVP film passivated the surface
against iridium, ruthenium, and platinum but Al O and ZrO films
grew on PVP. The ALD films, however, did not grow through the
PVP films to the silicon surface below the PVP film, and the PVP
film was removed after the ALD processes. The result is, in these
cases, a lift-off-patterned ALD film. Although the patterned areas in
this study were of micrometer scale, preliminary experiments with
electrospun nanofibers as mask layers indicate that nanometer scale
patterns are possible as well.
difficult to remove because the ALD film grew on PMMA and pos-
sibly also inside the PMMA film. The PMMA film could be re-
23
2
3
2
moved only by mechanical scribing. When the PMMA mask is
replaced with PVP, Al O is easier to deposit selectively. An Al O
2
3
2
3
film also grew on the PVP film but probably not inside the film and,
thus, the PVP film was easy to remove with water after the ALD
process. The film thickness measured from the patterned area was 39
nm.
When Al O was deposited from TMA and H O, the PVP film
2
3
2
was more difficult to remove after the ALD process than when AlCl3
was used as a precursor. The polymer film could still be removed by
dissolving into water, but it was time-consuming similar to the re-
Acknowledgment
2
3
moval of the PMMA film after the same ALD process. The differ-
The financial support from ASM Microchemistry and Academy
of Finland ͑project no. 115601͒ is gratefully acknowledged. The
authors gratefully acknowledge Dr. Esa Puukilainen for AFM mea-
surements.
ence between the AlCl - and TMA-based processes may arise from
3
a better solubility of TMA into PVP, which leads to more film
31
growth inside PVP. In summary, the PVP film seems to be a better
mask layer as compared to PMMA when Al O is deposited from
2
3
AlCl and H O. When Al O is deposited from TMA and H O, both
PMMA and PVP can be used as mask layers, but because of the
University of Helsinki assisted in meeting the publication costs of this
article.
3
2
2
3
2
Downloaded on 2014-11-09 to IP 132.174.255.116 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

K14
Journal of The Electrochemical Society, 157 ͑1͒ K10-K14 ͑2010͒
References
1
2
. M. Leskelä and M. Ritala, Thin Solid Films, 409, 138 ͑2002͒.
. M. Ritala and M. Leskelä, in Handbook of Thin Film Materials, H. S. Nalwa,
Editor, Academic, San Diego, CA ͑2002͒.
3
. M. Ritala and J. Niinistö, in Chemical Vapor Deposition: Precursors, Processes
and Application, A. C. Jones and M. L. Hitchman, Editors, pp. 158–206, Royal
Society of Chemistry, Cambridge, UK ͑2009͒.
4
5
. A. Ulman, Chem. Rev. (Washington, D.C.), 96, 1533 ͑1996͒.
. A. Ulman, An Introduction to Ultrathin Organic Films, From Langmuir-Blodgett to
Self-Assembly, Academic, San Diego, CA ͑1991͒.
6. A. P. Quist, E. Pavlovic, and S. Oscarsson, Anal. Bioanal. Chem., 381, 591 ͑2005͒.
7. Y. Xia and G. M. Whitesides, Annu. Rev. Mater. Sci., 28, 153 ͑1998͒.
8. R. K. Smith, P. A. Lewis, and P. S. Weiss, Prog. Surf. Sci., 75, 1 ͑2004͒.
9. W. Lee, C. Chao, X. Jiang, J. Hwang, S. F. Bent, and F. B. Prinz, ECS Trans.,
1
6͑4͒, 173 ͑2008͒.
0. R. Chen, H. Kim, P. C. MacIntyre, and S. F. Bent, Appl. Phys. Lett., 84, 4017
2004͒.
1. M. H. Park, Y. J. Jang, H. M. Sung-Suh, and M. M. Sung, Langmuir, 20, 2257
2004͒.
1
͑
1
1
1
1
͑
2. M. Yan, Y. Koide, J. R. Babcock, P. R. Markworth, J. A. Belot, T. J. Marks, and R.
P. H. Chang, Appl. Phys. Lett., 79, 1709 ͑2001͒.
3. K. J. Park, J. M. Doub, T. Gougousi, and G. N. Parsons, Appl. Phys. Lett., 86,
0
51903 ͑2005͒.
b)
1
2000
0000
4. R. Chen and S. F. Bent, Adv. Mater., 18, 1086 ͑2006͒.
Ir area
15. X. Jiang, R. Chen, and S. F. Bent, Surf. Coat. Technol., 201, 8799 ͑2007͒.
16. E. Färm, M. Kemell, M. Ritala, and M. Leskelä, Chem. Vap. Deposition, 12, 415
͑2006͒.
below PVP fiber after ALD process
PVP fiber after ALD process
1
1
7. E. Färm, M. Kemell, M. Ritala, and M. Leskelä, Thin Solid Films, 517, 972
2008͒.
8000
6000
4000
2000
0
͑
C
18. H. Lee and H. Kim, ECS Trans., 16͑4͒, 219 ͑2008͒.
19. J. Liu, Y. Mao, E. Lan, D. R. Banatao, G. J. Forse, J. Lu, H. O. Blom, T. O. Yeates,
Si
Ir
B. Dunn, and J. P. Chang, J. Am. Chem. Soc., 130, 16908 ͑2008͒.
2
2
0. A. Sinha, D. W. Hess, and C. L. Henderson, J. Electrochem. Soc., 153, G465
2006͒.
1. A. Sinha, D. W. Hess, and C. L. Henderson, J. Vac. Sci. Technol. B, 24, 2523
͑2006͒.
͑
O
Ir
2
2
2. T. Hua, W. P. King, and C. L. Henderson, Microelectron. Eng., 85, 934 ͑2008͒.
3. E. Färm, M. Kemell, M. Ritala, and M. Leskelä, J. Phys. Chem. C, 112, 15791
͑2008͒.
2
2
2
2
4. A. Sinha, D. W. Hess, and C. L. Henderson, Electrochem. Solid-State Lett., 9,
G330 ͑2006͒.
5. T. Aaltonen, M. Ritala, V. Sammelselg, and M. Leskelä, J. Electrochem. Soc., 151,
G489 ͑2004͒.
0
.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
Energy keV
6. T. Aaltonen, M. Ritala, T. Sajavaara, J. Keinonen, and M. Leskelä, Chem. Mater.,
1
5, 1924 ͑2003͒.
7. T. Aaltonen, P. Alén, M. Ritala, and M. Leskelä, Chem. Vap. Deposition, 9, 45
2003͒.
Figure 6. ͑Color online͒ ͑a͒ SEM micrograph of a patterned iridium film
deposited at 250°C. Patterning of film was done using PVP fibers as a mask
layer. The lighter areas are iridium and the darker areas are silicon. ͑b͒ EDX
spectra measured on Ir area, on the PVP fiber after the Ir process, and on the
Si surface after PVP removal.
͑
2
2
8. M. Ritala and M. Leskelä, Appl. Surf. Sci., 75, 333 ͑1994͒.
9. R. A. Waldo, Microbeam Anal., 23, 310 ͑1988͒.
30. M. D. Groner, F. H. Fabreguette, J. W. Elam, and S. M. George, Chem. Mater., 16,
39 ͑2004͒.
1. C. A. Wilson, R. K. Grubbs, and S. M. George, Chem. Mater., 17, 5625 ͑2005͒.
6
3
Downloaded on 2014-11-09 to IP 132.174.255.116 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).