High quality area-selective atomic layer deposition co using ammonia gas as a reactant

Article abstract of DOI:10.1149/1.3248002

Atomic layer deposition (ALD) Co was developed using bis(N, N′ -diisopropylacetamidinato)cobalt(II) as a precursor and NH₃ as a reactant, producing pure Co thin films with excellent conformality and nanoscale thickness controllability. In addition to NH₃, the Co films were also deposited by using H₂ gas as a reactant. Compared to ALD Co using H₂, the Co thin films deposited by NH₃ showed a higher film quality, a lower resistivity, and a higher density. The Co thermal ALD process was applied to area-selective ALD using an octadecyltrichlorosilane self-assembled monolayer as a blocking layer, which produced 3 μm wide Co line patterns without an etching process.

Full text of DOI:10.1149/1.3248002

D10
Journal of The Electrochemical Society, 157 ͑1͒ D10-D15 ͑2010͒
0
013-4651/2009/157͑1͒/D10/6/$28.00 © The Electrochemical Society
High Quality Area-Selective Atomic Layer Deposition Co
Using Ammonia Gas as a Reactant
a
b
b
a,
Han-Bo-Ram Lee, Woo-Hee Kim, Jeong Won Lee, Jae-Min Kim, *
d
b
b
c,d
Kwang Heo, In Chan Hwang, Yongjun Park, Seunghun Hong, and
a,
,z
Hyungjun Kim **
a
School of Electrical and Electronic Engineering, Yonsei University, Seoul 120-749, Korea
b
Department of Material Science and Engineering, Pohang University of Science and Technology, Pohang
7
90-784, Korea
c
d
Department of Physics and Astronomy and Interdisciplinary Program in Nano-Science and Technology,
Seoul National University, Seoul 151-747, Korea
Atomic layer deposition ͑ALD͒ Co was developed using bis͑N,NЈ-diisopropylacetamidinato͒cobalt͑II͒ as a precursor and NH as
3
a reactant, producing pure Co thin films with excellent conformality and nanoscale thickness controllability. In addition to NH3,
the Co films were also deposited by using H gas as a reactant. Compared to ALD Co using H , the Co thin films deposited by
2
2
NH3 showed a higher film quality, a lower resistivity, and a higher density. The Co thermal ALD process was applied to
area-selective ALD using an octadecyltrichlorosilane self-assembled monolayer as a blocking layer, which produced 3 ␮m wide
Co line patterns without an etching process.
©
2009 The Electrochemical Society. ͓DOI: 10.1149/1.3248002͔ All rights reserved.
Manuscript submitted December 12, 2008; revised manuscript received June 18, 2009. Published November 3, 2009.
Cobalt is one of the promising materials for practical applica-
tions as well as fundamental studies in nanotechnology. For ex-
ample, a giant magnetoresistance effect utilizing a Co/Ru multilayer
was a scientifically great finding, which produced a technological
such as lithography and etching. By SAM coating on a surface, the
surface properties can be changed into hydrophobic or hydrophilic
according to the end groups of the SAMs. Because ALD proceeds
through entire surface reactions, the deposition characteristics are
critically dependent on the surface properties of substrates. In many
cases, the nucleation of ALD is easy on hydrophilic OH-terminated
1
breakthrough for the development of magnetic recording media. In
addition to applications using the magnetic property of Co, e.g., in a
2
substrates ͑e.g., SiO ͒, while it is difficult on hydrophobic surfaces
nanomagnet, Co has been intensively investigated for nanoscale
2
3
,4
͑
e.g., a CH -terminated surface͒. Due to the differences in nucle-
semiconductor device fabrications such as nanocrystal memories
3
5,6
ation for SAM-modified surfaces toward the ALD reaction, AS-
ALD becomes possible. Several groups utilized AS-ALD to various
and contact materials. With device scaling, a highly conformal Co
thin-film deposition is required, especially for nanoscale contact
1
8,19
7
applications such as fuel cell and solar cell fabrications.
How-
fabrication.
ever, previous studies on AS-ALD have been mostly limited to ox-
Atomic layer deposition ͑ALD͒ is considered a promising tech-
nique for nanoscale film deposition due to its excellent conformality
and atomic thickness controllability. ALD is a suitable technique for
the fabrication of nanostructures. For example, Ru, Co, and Ni nano-
2
0,21
20,22,23
HfO2,
24
25
ides such as ZrO2,
ZnO, and TiO . For metal,
2
26 11,27 10
AS-ALD processes were reported for only Ru, Pt,
and Ir,
which were deposited using O as an oxidant. Overall, AS-ALD has
2
8
,9
been mostly explored in ALD processes using oxygen or H O as
structures were fabricated using ALD and nanotemplates. Even
with practical importance, however, the ALD of most transition-
2
reactants, and chloride or methoxide precursors and AS-ALD using
2
8
8
NH or H as a reactant have rarely been studied. Especially, AS-
3 2
metal films is difficult. For some noble metal systems including Ru,
1
0
11
ALD during PE-ALD has not been studied to our knowledge.
Recently, we have developed Co PE-ALD using bis͑cyclopenta-
dienyl͒ cobalt ͑CoCp ͒ as a metal precursor and NH plasma as a
Ir, and Pt, oxygen was used as a reactant to deposit pure metal
films through oxidative decomposition of the precursors. However,
2
3
the O -based thermal atomic layer deposition ͑T-ALD͒ cannot be
2
5
reactant for nanoscale contact applications. From the PE-ALD pro-
cess, pure Co films with a very low resistivity of 10 ␮⍀ cm were
obtained on both H-terminated Si͑001͒ and OH-terminated SiO2
substrates. However, high quality Co could not be deposited by
T-ALD using the same precursor. Previously, T-ALD using
applied to the deposition of most transition metals. Thus, the choice
of a proper reducing agent is important in metal ALD because high
quality metal deposition is usually achieved by a complete reduction
of the metal precursor during the reaction between the precursor and
the reducing agent. For example, hexafluoroacetylacetonate-based
precursors were used for metal ALD, with formalin as a reactant for
bis͑N,NЈ-diisopropylacetamidinato͒cobalt͑II͒ ͓Co͑iPr-AMD͒ ͔ and
2
2
9
1
2
13
H2 as a reactant was reported. The same group reported more
results on ALD Co, mainly focusing on the role of ALD Co as the
nucleation and adhesion layer for ALD Cu. In this study, we de-
ALD of Cu and Pd. The use of highly reactive plasma reactants
1
4
15
16
such as H and NH plasma was reported for Ti, Ta, and Ru
2
3
30
ALD. Compared to T-ALD, however, plasma-enhanced atomic layer
deposition ͑PE-ALD͒ has several disadvantages, such as the require-
ment of an additional apparatus for plasma generation and the nega-
tive effects of plasma on film properties, such as damage. Previ-
ously, a complicated process employing Ni oxide ALD using an
veloped NH -based T-ALD of Co using Co͑iPr-AMD͒ . Compared
3
2
to T-ALD Co using H , the T-ALD Co using NH showed a high
2
3
film density and a low resistivity. AS-ALD Co was achieved up to
1
000 ALD cycles from T-ALD using NH as well as H , and Co
3
2
nucleation on octadecyltrichlorosilane ͑OTS͒ was observed only
oxidant and sequential reduction by H plasma was reported for
2
17
above 1000 cycles due to the thermal degradation of OTS.
producing Ni thin films. However, the direct deposition of ALD
transition metals including Co and Ni has rarely been reported.
Meanwhile, area-selective atomic layer deposition ͑AS-ALD͒ us-
ing self-assembled monolayers ͑SAMs͒ is attracting great attention
due to its simple process, removing expensive patterning processes
Experimental
In this study, a commercial ALD chamber ͑Quros Plus 150͒ with
a loadlock chamber was used. This system has a double showerhead
system for good uniformity. Further information on chamber con-
8
figuration can be found in our previous report. Co͑iPr-AMD͒ was
2
contained in a stainless steel bubbler, and the temperature of the
bubbler was maintained at 65°C to obtain a suitable vapor pressure
for the ALD process. During the precursor exposure step, the pre-
*
Electrochemical Society Student Member.
*
* Electrochemical Society Active Member.
z
E-mail: hyungjun@yonsei.ac.kr
Downloaded on 2015-04-02 to IP 169.230.243.252 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 D10-D15 ͑2010͒
D11
(
PR
a)
(b)
0.8
T-ALD Co/SiO /Si(001)
OTS solution
2
TH Co
TNH Co
0.6
0.4
0.2
0
SiO
2
Si(001)
(
d)
(c)
OTS
0
1
2
3
4
5
tr (s)
Figure 1. The schematic drawings of preparation sequences for prepatterned
OTS: ͑a͒ PR patterning on SiO substrate by photolithography, ͑b͒ dipping of
PR patterned substrate in OTS solution, ͑c͒ OTS formation only on PR-free
region, and ͑d͒ PR removal by acetone of the remaining patterned OTS.
Figure 2. The growth rate vs reactant exposure time of Co T-ALD using
Co͑iPr-AMD͒2 and two reactants, NH3 ͑᭢, noted as TNH Co͒ and H2 ͑b,
noted as TH Co͒, on SiO2 substrate.
2
sured from at least five different sites of a substrate and were aver-
aged. In our study, the OTS-coated substrate showed a contact angle
as high as 100°, indicating that the surface was highly hydrophobic,
while the uncoated substrate ͑with chemical oxide͒ showed a contact
cursor vapors were carried into the main chamber by a carrier gas
͑
50 sccm of Ar͒. For T-ALD Co, two reactants, H and NH gases,
2 3
were used, and their flows were maintained at 400 sccm. One ALD
cycle was composed of four steps consisting of precursor exposure
33
angle of less than 20°, agreeing with the previously reported result.
͑
t ͒, purging ͑t ͒, reactant exposure ͑t ͒, and purging, where the flow
s p r
of Ar purging gas was 50 sccm. t and t were 3 and 1 s, respec-
The deposition selectivity on patterned OTS substrates was esti-
mated by measuring the thickness difference between OTS-coated
and -uncoated regions using a profilometer. Current measurements
on AS-ALD Co on a prepatterned OTS surface were carried out
using HP 5270A.
s
p
tively, and t was systematically controlled from 1 to 5 s. Routinely,
r
two kinds of substrates, Si͑001͒ and SiO , were used for T-ALD Co.
2
The Si͑001͒ substrates were cleaned by dipping in a buffered oxide
etchant ͑6:1͒ for 10 s followed by deionized ͑DI͒ water rinsing and
N blowing, resulting in H-terminated Si. The SiO substrates were
2
2
Results and Discussion
cleaned in acetone, isopropyl alcohol, and DI water, followed by N2
blowing resulting in OH-terminated SiO2.
ALD of Co films using NH and H .— In an earlier report on
3
2
Co ALD using the Co͑iPr-AMD͒2 precursor and H , a very low
For AS-ALD, OTS-coated substrates were prepared. The Si͑001͒
substrate was cleaned by a piranha solution ͑H SO :H O = 4:1 by
2
growth rate was problematic for practical applications below T
s
2
4
2
2
2
9
=
350°C. Thus, we performed Co ALD at a relatively high Ts
volume ratio͒, resulting in a chemical oxide-coated Si substrate.
Subsequently, the Si substrate was dipped in a diluted OTS solution
= 350°C. The Co thin films were deposited by T-ALD using NH3
͑TNH͒ or H2 ͑TH͒ as a reactant, and the growth characteristics and
film properties were comparatively investigated. Figure 2 shows the
͑
toluene:OTS = 10 mL:25 ␮L͒ for 24 h in a glove box for OTS
coating. For the prepatterned OTS coating, the photolithography-
assisted process shown in Fig. 1 was employed. The photoresist
2
growth rates of TNH and TH Co films on SiO with increasing tr
from 1 to 5 s. Except for the change in reactants, other process
͑
1
PR͒ ͑AZ5214͒ patterning was carried out by photolithography ͑Fig.
a͒ followed by dipping in an OTS solution for 10 min ͑Fig. 1b͒,
conditions such as T , t , t , and gas flows of the T-ALD were the
s
r
s
resulting in OTS coating on the PR-free region ͑Fig. 1c͒. The coat-
ing was performed in a high humidity condition of about 80%, in
contrast to the case of blank wafers. Because the water vapors assist
in the ordering of OTS, we could coat OTS on the prepatterned
substrate for a short time, accompanied with similar hydrophobicity
same for both precursors. The growth rates are saturated at tr
Ͼ 1 s for both precursors, indicating the self-saturation ALD reac-
tions. However, the saturation growth rate of TH Co ͑0.43 Å/cycle͒
is larger than that of TNH Co ͑0.26 Å/cycle͒. Meanwhile, the resis-
tivity of TNH Co ͑ϳ50 ␮⍀ cm͒ is three to four times lower than
31,32
for the case of glove box coating.
After the PR was removed by
that of TH Co ͑ϳ200 ␮⍀ cm͒ at t = 5 s.
r
acetone, OTS patterns were formed on the SiO substrate ͑Fig. 1d͒.
The conformality of T-ALD Co was estimated by depositing the
films inside nanosize holes. Figure 3a shows the scanning electron
microscopy ͑SEM͒ image of a hole pattern with a 5:1 aspect ratio
͑450 nm depth and 90 nm diameter͒ after 1000 cycles of TNH Co at
tr = 1 s. The Co film thicknesses at the top, side wall, and bottom of
the pattern are almost the same as 24 nm, resulting in a 100% step
coverage. Similarly, the TH Co also produces highly conformal Co
films inside the holes, as shown in Fig. 3b. The excellent conformal-
ity of both T-ALD processes indicates a saturated surface reaction of
ALD, which is consistent with growth rate saturation, as shown in
Fig. 2. The SEM images show different surface morphologies of Co
films according to the reactants. The TNH Co shows a much larger
grain size than TH Co.
2
After the preparation of OTS-coated substrates, Co T-ALD was per-
formed at 350°C of growth temperature ͑T ͒ and t = 1 s.
s
r
Field-emission-scanning electron microscopy ͑FESEM͒ was
used for analyses of morphology, and the chemical composition
analysis was performed by energy dispersive spectroscopy ͑EDS͒
and X-ray photoelectron spectroscopy ͑XPS͒. The density and sur-
face morphology of the Co films were analyzed by synchrotron ra-
diation X-ray reflectivity ͑XRR͒ ͑Pohang Acceleration Laboratory,
3C2 beam line͒ and atomic force microscopy ͑AFM͒, respectively.
The resistivity of Co films was measured by a four-point probe. The
surface hydrophobicity was measured by a static contact angle go-
niometer ͑DSA-10, Kruss Co.͒. For this, 4 ␮L of water droplets
was dropped on a horizontal substrate with an automatic dispensing
system, and their contact angles were measured from charge-
coupled device camera images. The static contact angles were mea-
The chemical compositions of TNH and TH Co films deposited
at t = 5 s were analyzed by XPS. For TNH Co, narrow XPS peaks
r
are observed in the Co core level at 778.3 and 793.3 eV of binding
Downloaded on 2015-04-02 to IP 169.230.243.252 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

D12
Journal of The Electrochemical Society, 157 ͑1͒ D10-D15 ͑2010͒
energy, which correspond to 2p and 2p peaks of metallic Co
3
/2
1/2
(
a) TNH Co/5:1 via
34
͑
Fig. 4a͒. The XPS peak intensities related to impurities such as N,
O, and C are very weak, as shown in Fig. 4b. In contrast, the Co
peaks of the TH Co film shown in Fig. 4c are very broad compared
to Fig. 4a. Each of the Co 2p3 and 2p peaks is deconvoluted to
/2
1/2
three peaks denoted by solid, dashed, and dotted lines. The deconvo-
luted Co 2p3 peaks are observed at 778.3 ͑solid line͒, 780.3
/2
͑
dashed line͒, and 786.3 eV ͑dotted line͒ of binding energy. These
peaks correspond to Co–Co bonding, Co–O bonding, and a shake-up
satellite peak, respectively, indicating oxidation of the film. It was
reported that the shake-up satellite peaks are observed together with
3
5,36
the Co–O bonding peak according to oxidation states.
Indeed,
Fig. 4d shows a relatively large O peak. The O XPS peak is mainly
observed at 529.7 eV with a small shoulder peak at 531.3 eV, where
the high intensity peak corresponds to O–Co bonding, and the small
3
5,36
100 nm
peak corresponds to hydroxyl.
The molecular structure of the
Co͑iPr-AMD͒ is composed of Co atoms directly bonded to four
2
2
9
neighboring N atoms without oxygen. Because there is no oxygen
atom in the precursor and reactant, the oxygen contamination for TH
Co is probably due to postdeposition oxidation. The synchrotron
radiation XRR analysis indicates that the film density of TH Co is
significantly lower than that of TNH Co ͑62 vs 79%͒. Due to the low
film density, the oxygen molecules could easily penetrate the TH Co
thin films forming a Co–O bond, and the oxidation is exacerbated by
the small size grains of TH Co films. Meanwhile, the nitrogen levels
are very low in both of the TNH and TH Co films. Thus, the use of
(
b) TH Co/5:1 via
NH as a reactant does not significantly increase the level of nitro-
3
gen contamination in the film, indicating that the NH molecules
3
effectively cleave the Co–N bonds of the Co͑iPr-AMD͒ forming a
2
pure Co film. Similarly, the deposition of pure Ru by ALD using
3
7
NH as a reactant was reported previously.
3
AS-ALD using SAM.— Co T-ALD processes using different re-
1
00 nm
actants, H and NH , were applied to AS-ALD. For comparison, we
2
3
prepared two substrates, OTS-coated and -uncoated substrates, and
T-ALD Co was carried out simultaneously on the two samples for
000 cycles. For TNH Co, SEM and EDS analyses show that the Co
Figure 3. The FESEM images of ͑a͒ TNH Co and ͑b͒ TH Co films on
nanosized hole pattern ͑450 nm depth and 90 nm diameter͒.
1
film was not deposited onto the OTS-coated surface ͑Fig. 5a, EDS
(
a) TNH Co/SiO /Si(001)
(b) TNH Co/SiO /Si(001)
2
2
Co core level
N core level
394 392
O core level
4
04
402
400
398
396
5
26
528
530
532
534
536
C core level
280
282
284
286
288
290
292
805 800 795 790 785 780 775 770
Binding Energy (eV)
Binding Energy (eV)
Figure 4. The XPS spectra of ͓͑a͒ and ͑b͔͒
TNH Co and ͓͑c͒ and ͑d͔͒ TH Co in Co
core level and impurity content ͑N, O, and
C͒ core levels, respectively.
(
c) TH Co/SiO /Si(001)
(d) TH Co/SiO /Si(001)
2
2
Raw
Co-Co
Shake-up
Sum
Co-O
N core level
394 392
4
04 402
400
398 396
532
O core level
5
26
528
530
534
536
C core level
805 800 795 790 785 780 775 770
280
282
284
286
288
290
292
Binding Energy (eV)
Binding Energy (eV)
Downloaded on 2015-04-02 to IP 169.230.243.252 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 D10-D15 ͑2010͒
D13
(
a) TNH Co/OTS coated substrate
(b) TNH Co/uncoated substrate
100 nm
100 nm
Figure 5. The FESEM images of ͓͑a͒ and
͑
b͔͒ TNH Co films and ͓͑c͒ and ͑d͔͒ TH
Co films on OTS-uncoated and -coated
substrates, respectively.
(
c) TH Co/OTS coated substrate
(d) TH Co/uncoated substrate
100 nm
100 nm
data not shown͒, while 24 nm thick Co films were deposited onto
the uncoated substrate ͑Fig. 5b͒. Figure 5c and d shows the SEM
images of TH Co deposited onto the OTS-coated and -uncoated
substrates, respectively, indicating that Co films are observed only
on the uncoated substrate similar to the TNH Co case. These obser-
vations indicate that the OTS effectively blocks the Co film deposi-
tion during the Co TH-ALD, irrespective of the reactant.
Figure 7a shows the thickness profile of TNH Co on the pat-
terned OTS substrate for 2000 cycles. The thickness difference be-
tween the highest and lowest points was 32 nm, which is signifi-
cantly smaller that the expected thickness ͑48 nm͒. In contrast to
1000 cycles, both of the I-V characteristics on the same and different
Co lines show an ohmic behavior ͑Fig. 7b͒. The current detection
between different Co lines indicates that each Co line is not electri-
cally isolated. We observed 48 nm of Co on the OTS-free region and
16 nm of Co on the OTS-coated regions in SEM analysis, consistent
with the profilometer and I-V results. Therefore, we infer that the Co
was deposited even onto the OTS-coated region, reducing the thick-
ness difference to 32 nm, as shown in Fig. 7a.
Then, the Co AS-ALD was carried out using NH3 at Ts
=
350°C and t = 1 s on a substrate with prepatterned OTS pre-
r
pared by the sequences, as shown in Fig. 1. The line pattern was
formed by conventional photolithography, composed of an alterna-
tive OTS-coated region ͑9 ␮m width͒ and -uncoated region ͑3 ␮m
width͒. Figure 6a is an AFM image showing the morphology of Co
films by TNH Co for 1000 cycles. The AFM image shows that the
width of the Co lines and the distance between the lines are 3 and
In previous reports on AS-ALD utilizing the SAM-blocking
layer, the loss of deposition selectivity above a certain number of
growth cycles is a common observation. For instance, in AS-ALD of
4
0
31
9
␮m, respectively, exactly replicating the inverse of the OTS pat-
Pt and HfO2 using OTS as a blocking layer at T = 285 and
s
tern. The thickness profile of AS-ALD Co shows that the thickness
difference between the highest point and the lowest point is 24 nm,
which is the same thickness on the uncoated blank substrate ͑Fig.
250°C, the selectivity of deposition was lost at longer process times
than 25600 and 18015 s, respectively. Indeed, the hydrophobicity of
the OTS-coated surface was reported to be thermally degraded be-
low 200°C within a few minutes in air through the cleavages of the
6
b͒. Thus, we infer that Co is only deposited onto the uncoated
4
1
region without nucleation delay. Also, no deposition occurred on the
OTS-coated region.
Si–C and C–C bondings. In vacuum or other ambient, however,
the degradation of OTS hydrophobicity is retarded to a higher tem-
perature because the amount of chemical entity cleaving hydrocar-
The quality of the AS-ALD Co film was estimated by current–
voltage ͑I-V͒ measurement. When the electrical probes were placed
on the same Co line with 100 ␮m of probe distance, a clear ohmic
behavior was observed with a linear increase in current and with an
increase in voltage, as Fig. 6c shows. The resistivity of the Co line
was estimated to be 935 ␮⍀ cm by assuming that the Co line is
rectangular ͑cross-sectional area = 24 nm ϫ 3 ␮m; length
4
2,43
bon chains of OTS such as water is very small.
There were
reports that the degradation temperatures of OTS become even up to
400 and 470°C in N2 ambient and ultrahigh vacuum,
4
2,43
respectively.
Therefore, the loss of deposition selectivity with an
increasing number of ALD cycles can be explained by the thermal
degradation of OTS. In our AS-ALD, the loss of deposition selec-
tivity was observed over 1000 cycles, namely, 6000 s of total time of
the ALD process. Compared with that of Pt and HfO2, the total
process time until the loss of selective deposition is relatively short
probably due to the higher growth temperature of 350°C.
=
100 ␮m͒. The resistivity of the Co line calculated from I-V is
higher than that of the TH-ALD Co film measured by a four-point
probe. The resistivity increase is attributed mainly to contact resis-
tance and spreading resistance caused by two-point probe measure-
ments. The size effect may not be significant because we compare
the resistivity of the line and film with the same thickness, and the
Conclusion
38,39
width of the line is in the micrometer scale.
In contrast, no
detectable current was measured on different Co lines, indicating
that each Co line is electrically isolated, which shows that Co film is
not formed on the OTS-coated region.
In summary, Co T-ALD using a Co precursor, Co͑iPr-AMD͒2,
and two reactants, H2 and NH , was investigated and applied to
3
AS-ALD with OTS SAM as a blocking layer. Metal thin-film depo-
Downloaded on 2015-04-02 to IP 169.230.243.252 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

D14
Journal of The Electrochemical Society, 157 ͑1͒ D10-D15 ͑2010͒
500
400
300
200
(
a) TNH Co/patterned OTS
(a) TNH Co/patterned OTS
Co
OTS
100
0
0
10
20
30
Distance (µm)
10 μm
0
0
0
.6
.4
.2
0
(
b) TNH Co/patterned OTS
5
4
3
2
1
00
00
00
00
00
0
(
b) TNH Co/patterned OTS
on same line
on different lines
-0.2
-
0.4
0.6
-
-
3
-2
-1
0
1
2
3
0
10
20
30
Voltage (V)
Distance (µm)
Figure 7. ͑a͒ The thickness profile of TNH Co deposited onto patterned
OTS-coated substrate for 2000 cycles and ͑b͒ I-V results of AS-ALD Co on
same lines ͑᭿͒ and different lines ͑᭺͒ ͑inset: Cartoons of probes on Co
lines͒.
0
.6
.4
.2
0
(
c) TNH Co/patterned OTS
on same line
on different lines
0
0
synchrotron radiation XRR analysis was performed at the 3C2 X-ray
scattering beam line of the Pohang Acceleration Laboratory ͑PAL͒.
-
-
-
0.2
0.4
0.6
Yonsei University assisted in meeting the publication costs of this article.
References
1
2
. S. S. P. Parkin, N. More, and K. P. Roche, Phys. Rev. Lett., 64, 2304 ͑1990͒.
. Q. Xiao, T. Yang, A. Ursache and M. T. Tuominen, J. Appl. Phys., 103, 07C521
͑2008͒.
-3
-2
-1
0
1
2
3
3
4
. F. M. Yang, T. C. Chang, P. T. Liu, P. H. Yeh, Y. C. Yu, J. Y. Lin, S. M. Sze, and
J. C. Lou, Appl. Phys. Lett., 90, 132102 ͑2007͒.
. C.-J. Kim, S.-W. Ryu, Y.-K. Choi, J.-J. Chang, S. H. Bae, and B.-H. Sohn, Appl.
Voltage (V)
Phys. Lett., 93, 052106 ͑2008͒.
Figure 6. ͑a͒ The AFM image, ͑b͒ the thickness profile of TNH Co deposited
onto patterned OTS-coated substrate for 1000 cycles, and ͑c͒ I-V results of
AS-ALD Co on same lines ͑᭿͒ and different lines ͑᭺͒ ͑inset: Cartoons of
probes on Co lines͒.
5. H.-B.-R. Lee and H. Kim, Electrochem. Solid-State Lett., 9, G323 ͑2006͒.
6
7
8
. H.-B.-R. Lee, J. Y. Son, and H. Kim, Appl. Phys. Lett., 90, 213509 ͑2007͒.
. J. H. Lee and J. H. Cha, Microelectron. Eng., 71, 321 ͑2004͒.
. W.-H. Kim, S.-J. Park, J.-Y. Son, and H. Kim, Nanotechnology, 19, 045302
͑2008͒.
9
. M. Daub, M. Knez, U. Goesele and K. Nielsch, J. Appl. Phys., 101, 09J111 ͑2007͒.
1
0. E. Farm, M. Kernell, M. Ritala, and M. Leskel, Chem. Vap. Deposition, 12, 415
2006͒.
1. R. Chen and S. F. Bent, Adv. Mater., 18, 1086 ͑2006͒.
͑
sition by T-ALD using NH was reported, and the T-ALD Co from
3
1
NH showed superior film properties, lower resistivity, lower impu-
3
12. J. Huo, R. Solanki, and J. McAndrew, J. Mater. Res., 17, 2394 ͑2002͒.
13. J. W. Elam, A. Zinovev, C. Y. Han, H. H. Wang, U. Welp, J. N. Hryn, and M. J.
Pellin, Thin Solid Films, 515, 1664 ͑2006͒.
rity level, and high density compared to that from H . Co AS-ALD
2
was achieved by both Co T-ALD processes to 1000 cycles. Through
the Co AS-ALD, 3 ␮m wide Co line patterns were obtained on a
prepatterned OTS-coated substrate without an etching process.
1
4. H. Kim and S. M. Rossnagel, J. Vac. Sci. Technol. A, 20, 802 ͑2002͒.
1
5. H. Kim, C. Cabral, C. Lavoie, and S. M. Rossnagel, J. Vac. Sci. Technol. B, 20,
1
321 ͑2002͒.
1
6. O.-K. Kwon, S.-H. Kwon, H.-S. Park, and S.-W. Kang, J. Electrochem. Soc., 151,
C753 ͑2004͒.
Acknowledgments
1
1
1
7. J. Chae, H. S. Park, and S. W. Kang, Electrochem. Solid-State Lett., 5, C64 ͑2002͒.
8. X. Jiang, H. Huang, F. B. Prinz, and S. F. Bent, Chem. Mater., 20, 3897 ͑2008͒.
9. C. Bae, H. Yoo, S. Kim, K. Lee, J. Kim, M. M. Sung, and H. Shin, Chem. Mater.,
20, 756 ͑2008͒.
This work was supported by the Korea Science and Engineering
Foundation ͑KOSEF͒ grant funded by the Korean government
MOST͒ ͑no. 2007-02864 and no. R01-2007-000-20143-0͒. The
͑
Downloaded on 2015-04-02 to IP 169.230.243.252 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 D10-D15 ͑2010͒
D15
2
0. R. Chen, H. Kim, P. C. McIntyre, and S. F. Bent, Appl. Phys. Lett., 84, 4017
2004͒.
32. T. M. Mayer, M. P. de Boer, N. D. Shinn, P. J. Clews, and T. A. Michalske, J. Vac.
Sci. Technol. B, 18, 2433 ͑2000͒.
͑
2
2
2
1. J. P. Lee and M. M. Sung, J. Am. Chem. Soc., 126, 28 ͑2004͒.
2. R. Chen, H. Kim, P. C. McIntyre, and S. F. Bent, Chem. Mater., 17, 536 ͑2005͒.
3. R. Chen, H. Kim, P. C. McIntyre, D. W. Porter, and S. F. Bent, Appl. Phys. Lett.,
33. D. K. Schwartz, Annu. Rev. Phys. Chem., 52, 107 ͑2001͒.
34. J. F. Moulder, W. F. Stickle, P. E. Sobol, and K. D. Bomben, Handbook of X-Ray
Photoelectron Spectroscopy, Perkin-Elmer, Eden Prairie, MN ͑1992͒.
35. M. Oku and K. Hirokawa, J. Electron Spectrosc. Relat. Phenom., 8, 475 ͑1976͒.
36. M. Burriel, G. Garcia, J. Santiso, A. Abrutis, Z. Saltyte, and A. Figueras, Chem.
Vap. Deposition, 11, 106 ͑2005͒.
8
6, 191910 ͑2005͒.
2
2
4. B. H. Lee and M. M. Sung, J. Nanosci. Nanotechnol., 7, 3758 ͑2007͒.
5. E. K. Seo, J. W. Lee, H. M. Sung-Suh, and M. M. Sung, Chem. Mater., 16, 1878
͑
2004͒.
6. K. J. Park, J. M. Doub, T. Gougousi, and G. N. Parsons, Appl. Phys. Lett., 86,
51903 ͑2005͒.
37. H. Li, D. B. Farmer, R. G. Gordon, Y. Lin, and J. Vlassak, J. Electrochem. Soc.,
2
154, D642 ͑2007͒.
0
38. W. Steinhögl, G. T. Schindler, G. Steinlesberger, and M. Engelhardt, Phys. Rev. B,
66, 075414 ͑2002͒.
2
2
7. R. Chen and S. F. Bent, Chem. Mater., 18, 3733 ͑2006͒.
8. A. Dube, M. Sharma, P. F. Ma, and J. R. Engstrom, Appl. Phys. Lett., 89, 164108
39. S. M. Rossnagel and T. S. Kuan, J. Vac. Sci. Technol. B, 22, 240 ͑2004͒.
40. X. Jiang and S. F. Bent, J. Electrochem. Soc., 154, D648 ͑2007͒.
41. Y. X. Zhuang, O. Hansen, T. Knieling, C. Wang, P. Rombach, W. Lang, W. Be-
necke, M. Kehlenbeck, and J. Koblitz, J. Micromech. Microeng., 16, 2259 ͑2006͒.
42. G. J. Kluth, M. M. Sung, and R. Maboudian, Langmuir, 13, 3775 ͑1997͒.
43. U. Srinivasan, M. R. Houston, R. T. Howe, and R. Maboudian, J. Microelectro-
mech. Syst., 7, 252 ͑1998͒.
͑
2006͒.
2
3
9. B. S. Lim, A. Rahtu, and R. G. Gordon, Nature Mater., 2, 749 ͑2003͒.
0. Z. Li, R. G. Gordon, D. B. Farmer, Y. Lin, and J. Vlassak, Electrochem. Solid-State
Lett., 8, G182 ͑2005͒.
3
1. J. Hong, D. W. Porter, R. Sreenivasan, P. C. McIntyre, and S. F. Bent, Langmuir,
2
3, 1160 ͑2007͒.
Downloaded on 2015-04-02 to IP 169.230.243.252 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).