Nucleation controlled surface oxidation epitaxy of thermally grown NiO on (0 0 1) Ni for coated conductor applications assisted by Mo or Mn microalloying

Article abstract of DOI:10.1016/j.physc.2004.01.027

High temperature oxidation of cube textured (100)Ni and (100)Ni-0.1wt%(Mo or Mn) was carried out in a wide temperature regime between 900 and 1400 °C in air. Special attention was paid to avoid pre-oxidation at lower temperatures, thus favouring nucleation controlled surface oxidation epitaxy (SOE) as opposed to SOE achieved by competitive grain growth. Consequently, cube textured epitaxial NiO layers, under 1 μm in thickness, have been grown on pure Ni in a much wider working window than previously published. Furthermore, thin cube textured NiO layers have been grown on Ni-0.1%Mo tape in a still wider working window of ～200 °C between 1150 and 1350 °C. Preliminary results on Ni-0.1wt%Mn show a similar trend. An in-plane orientation of <5° and out of plane orientation approaching 2° was obtained. Single crystal like transparent NiO films were grown under optimal conditions.

Full text of DOI:10.1016/j.physc.2004.01.027

Physica C 405 (2004) 219–226
www.elsevier.com/locate/physc
Nucleation controlled surface oxidation epitaxy of
thermally grown NiO on (0 0 1) Ni for coated
conductor applications assisted by Mo or Mn microalloying
a,
b
*
A. Kursumovic , R. H u€ hne, R. Tomov, B. Holzapfel ,
B.A. Glowacki, J.E. Evetts
a
IRC in Superconductivity and Department of Materials Science, University of Cambridge, Cambridge CB2 3QZ, UK
b
IFW Dresden, Institute of Metallic Materials, POB 270016 Dresden, Germany
Received 27 May 2003; received in revised form 14 January 2004; accepted 27 January 2004
Abstract
High temperature oxidation of cube textured (1 0 0)Ni and (1 0 0)Ni–0.1wt%(Mo or Mn) was carried out in a wide
temperature regime between 900 and 1400 ꢀC in air. Special attention was paid to avoid pre-oxidation at lower tem-
peratures, thus favouring nucleation controlled surface oxidation epitaxy (SOE) as opposed to SOE achieved by
competitive grain growth. Consequently, cube textured epitaxial NiO layers, under 1 lm in thickness, have been grown
on pure Ni in a much wider working window than previously published. Furthermore, thin cube textured NiO layers
have been grown on Ni–0.1%Mo tape in a still wider working window of ꢀ200 ꢀC between 1150 and 1350 ꢀC. Pre-
liminary results on Ni–0.1wt%Mn show a similar trend. An in-plane orientation of <5ꢀ and out of plane orientation
approaching 2ꢀ was obtained. Single crystal like transparent NiO films were grown under optimal conditions.
ꢁ
2004 Elsevier B.V. All rights reserved.
PACS: 81.65.Mq; 68.55.J; 85.25.K
Keywords: Oxidation; Epitaxy; Coated conductors; Texture
1
. Introduction
ture coated superconductors [1]. The optimum
texturing conditions, such as annealing tempera-
ture and time for recrystallization, as well as pre-
vention of secondary recrystallization are closely
dependent on the Ni purity [2–4]. In this sense
addition of microalloying elements such as Mo
and Mn stabilises the structure and shifts the start
of secondary recrystallization towards higher
temperatures.
Cube textured Ni and Ni-based alloy tapes
prepared by heavy cold rolling and subsequent
annealing (RABiTS: rolling assisted biaxially tex-
tured substrate) are promising substrates for the
preparation of YBa Cu O
2
type high tempera-
3
7ꢁd
*
A buffer layer structure between the Ni tape and
the superconductor is found necessary to reduce
Corresponding author. Fax: +44-1223-334373.
E-mail address: ak237@cus.cam.ac.uk (A. Kursumovic).
0
921-4534/$ - see front matter ꢁ 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.physc.2004.01.027

220
A. Kursumovic et al. / Physica C 405 (2004) 219–226
Ni contamination and uncontrolled NiO growth
during superconductor deposition in an oxygen-
containing atmosphere. Epitaxial oxide films such
research [9] and SOE film growth for coated con-
ductor applications [11–14]. The main aim of this
work is to investigate the possibilities of preparing
thin cube textured NiO on textured Ni tapes by
precise control of oxidation parameters, primarily
by removing initial oxide layer and avoiding any
significant oxidation at lower temperatures, hence
assuring the NiO nucleation at temperatures of
optimal cube on cube NiO growth. The influence
of doping (microalloying) elements already used to
control Ni grain growth [3,4] is also investigated.
2
as CeO and MgO are usually deposited on Ni in
high vacuum as oxygen barriers as well as to
transfer the cube texture from the metal into the
superconductor [5,6]. Similar protective layers
were previously investigated as a method for
improving Ni oxidation resistance [7–9]. An
alternative approach is to form an epitaxial NiO
film on the Ni surface by thermal surface oxidation
epitaxy (SOE) [10–14].
The texture of oxide films grown on oriented
substrates results either from epitaxial film nucle-
ation at the substrate or from preferential growth
of grains with the desired texture. Surface prepa-
ration has an essential influence on the initial NiO
growth. On bare Ni exposed to air, there is fast
oxide formation even at room temperature where a
2. Experimental
Ni, Ni–0.1%Mo and Ni–0.1%Mn tapes of high
Ni purity (>99.97%) were prepared by cold rolling
to a deformation of ꢀ97% and a thickness of 80
lm. These tapes were recrystallized in vacuum
ꢁ
6
1
nm layer of oxide forms in less than a minute,
followed by slow logarithmic growth kinetics
from 1 to 10 nm) [15]. Although this initial oxide
(ꢀ10 mbar) or reducing atmosphere (ꢀ5% H
2
in
Ar) in a temperature regime from 600 to 1000 ꢀC
to get a sharp cube texture. Details of the Ni tape
preparation and texturing procedures are reported
elsewhere [3].
(
appears to be in epitaxial relationship with the
substrate, it goes through structural changes as the
oxide becomes thicker [15] and grains of mixed
orientation are formed from the initial oxide layer,
that appear to be mainly composed of (1 0 0) and
For oxidation, Ni and microalloyed Ni tapes
were cut into pieces 5–25 mm · 10 mm. The sam-
ples were ultrasonically cleaned in acetone and
methanol prior to oxidation. When necessary,
chemical and electrochemical surface preparation
as well as hydrogen reduction prior to oxidation
was carried out to remove the initial oxide layer
built up during earlier processing stages or during
storage [15,17].
(
1 1 1) oriented NiO grains on (0 0 1) Ni RABiTS
tapes [11,14]. During further oxidation, at high
temperatures, the cube on cube texture is eventu-
ally achieved through a competitive (1 0 0) growth
due to faster lateral growth of (1 0 0) NiO grains
[
11,12,14]. This competitive grain growth, as well
as grain boundary enhanced diffusion, create a
relatively rough NiO surface in thin NiO films. As
already reported, cube on cube preferred NiO
growth on pure Ni tapes was observed, for >15 lm
NiO films, in a narrow temperature range around
The oxidation was carried out in air or in pure
oxygen in the temperature range between 950 and
1400 ꢀC. A tubular furnace allowed the application
of a controlled atmosphere or purified air, as well
as fast switching between different atmospheres. In
order to avoid oxide growth at lower temperatures
and achieve NiO nucleation only at the growth
temperature, Ni tapes were placed in the furnace
1
200–1260 ꢀC in air and pure oxygen [14]. At these
high temperatures, the thermal oxidation of Ni
occurs by Ni volume diffusion from the metal–
oxide interface through the NiO film to the outer
NiO surface via Ni vacancy transport mechanism
2
under a protective (Ar) or reducing (5% H in Ar)
atmosphere followed immediately by a hot oxy-
gen–argon mixture or air.
The kinetics of the oxidation was monitored by
mass and thickness increase during successive
oxidations. The mass was measured by a micro-
balance with ±10 lg sensitivity. The thickness was
[16] resulting in a flat NiO surface as the film be-
come thicker.
Although the oxidation of pure Ni to form NiO
has been extensively studied, it is still attracting
attention in the case of fundamental oxidation

A. Kursumovic et al. / Physica C 405 (2004) 219–226
221
o
calculated from the measured mass increase and
was compared with results from cross section
microscopy. The texture analysis was carried out
by X-ray diffraction techniques, i.e. XRD h–2h
scans, rocking curves and pole figures. Optical
microscopy, atomic force microscopy (AFM) and
scanning electron microscopy (SEM) were used to
monitor microstructure development during the
NiO growth.
temperature ( C)
1
8
400 1300 1200 1100
1000
-
1
0
oxidation in air
-10
1
1
0
vol.
-
9
10
diff.
NiO/Ni
vol. diff.)
(
g
.
b
.
NiO/Ni-0.1% Mo
vol. diff.
diff.
0.8
-
11
0
3
. Results
g.b. diff.
-
10
10
0.6
0.7
3
.1. Kinetics
-
1
-1
T
(1000K )
The NiO growth kinetics on pure cube textured
Fig. 1. Arrhenius plots of parabolic oxidation constants for
(––) Ni with the contribution from grain boundary diffusion
omitted [7,14,18], and (m) Ni–0.1%Mo oxidation in air. The
Ni tapes has been investigated previously and re-
ported in detail [14,18]. The oxidation kinetics of
Ni–0.1%Mo, relevant for controlling the NiO
thickness, is established in this work. The NiO
solid line is from earlier work [18]. k
according to mass and thickness increase of the growing NiO/
p g
and k are related
ꢁ
2
Ni film are related: 1 mg · cm of the mass increase per unit
area increase corresponds to roughly 7 lm thickness increase
[14,18].
growth in all cases was found to obey a parabolic
2
rate law for both thickness increase: X ¼ 2k
p
t and
2
mass per unit area increase: (Dm=A Þ ¼ 2k t, where
g
k
and kg are the corresponding parabolic rate
3.2. Texture and microstructure
p
constants. The nature of this kinetics is that the
NiO growth is diffusion limited across the growing
NiO layer. The oxidation kinetics data for the Ni
and Ni–0.1%Mo are summarised in Fig. 1. The
NiO on Ni growth kinetics from this and earlier
work [14,18] closely follows the tracer diffusion
data [7]. It can be seen that NiO grows faster on
Ni–0.1%Mo than on pure Ni tape, which is be-
lieved to be due to the creation of extra vacancies
in NiO when Ni is substituted with Mo. The in-
crease in the oxidation rate occurring at lower
temperatures ( K 1100 ꢀC), seen as deviation from
the volume diffusion line in Fig. 1, is caused by
NiO grain boundary diffusion [7,14]. However,
although the grain boundary diffusion is much
faster than the volume diffusion, NiO grain
boundaries have an effective width of only ꢀ1 nm
Typical XRD scans for ꢀ1 lm thick NiO film
grown on cube textured Ni–0.1%Mo in air at
temperatures between 950 and 1350 ꢀC are shown
in Fig. 2a. At all temperatures a strong (0 0 2) NiO
peak is visible besides the (0 0 2) Ni substrate peak.
At lower temperatures (950 ꢀC) a significant
amount of (1 1 1) and (2 2 0) oriented NiO grains is
present that vanishes for higher oxidation tem-
peratures. At temperatures above 1350 ꢀC sec-
ondary recrystallization of the substrate starts,
leading to a degradation of the NiO texture (not
shown). The full width of half maximum (FWHM)
of the out-of-plane orientation was determined to
be from 5ꢀ to 8ꢀ which is of the same value as for
Ni-tape itself. Furthermore, pole figure measure-
ments showed that the thin NiO films are highly
textured with cube on cube relation to the sub-
strate (Fig. 2b). Preliminary experiments indicate
(Fig. 3) that thin NiO layer grown on Ni–0.1%Mn
tape can be highly textured, as well, when grown
at those high temperatures. After further oxida-
tion towards thicker NiO films the other (hkl)
peaks disappear almost completely followed by
[
7] hence the total contribution of the grain
boundary diffusion depends on the NiO grain size,
which is increasing in time and temperature
[
7,14,18]. Thus grain boundary diffusion makes
significant contribution to the average thickness or
mass increase at lower temperatures and shorter
times.

222
A. Kursumovic et al. / Physica C 405 (2004) 219–226
Fig. 2. (a) h–2h scans for SOE NiO on (0 0 2) Ni–0.1%Mo grown in air at 950, 1150 and 1350 ꢀC for about 50, 9 and 2 s respectively,
corresponding to thickness of ꢀ1 lm. (b) A (1 1 1) NiO pole figure of ꢀ1 lm thick NiO film on (0 0 2) Ni–0.1%Mo grown in air at 1250
ꢀC for ꢀ4 s, typical for the wide temperature window between 1150 and 1350 ꢀC.
5
1
1
1
1
1
0
0
0
0
0
to hours at 950 ꢀC). Each data point represents at
least four different Ni or microalloyed Ni samples.
In these experiments with nucleation controlled
NiO epitaxy, even thin (ꢀ1 lm) NiO films on pure
Ni can be grown with a good, R > 100, cube on
cube (0 0 2) orientation a temperature regime from
4
3
2
1
1220 to 1320 ꢀC, while excellent texture with
R > 1000 was found from 1250 to 1260 ꢀC. This is
much wider range than found earlier for grain
growth controlled texturing [14], as compared in
Fig. 5. The addition of a small amount of Mo
or Mn leads to a further broadening of the opti-
mum temperature window to 1250–1350 ꢀC, with
the maximum around 1350 ꢀC for microalloyed
Ni–0.1%Mo tapes.
3
5
40
45
50
55
60
65
o
2
θ ( )
Fig. 3. XRD h–2h scans of (0 0 2) Ni–0.1%Mn tape after 10 s
oxidation at 1250 ꢀC in air.
improved texture for NiO films on both pure Ni
tapes and microalloyed Ni–0.1%Mo or Ni–
.1%Mn tapes. Under optimised conditions NiO
films were grown with an out-of-plane FWHM
Transparent and almost structurally perfect
NiO layers, Fig. 6, were formed in the optimal
temperature window of 1250–1300 ꢀC for pure Ni
and 1250–1350 ꢀC for microalloyed Ni. AFM
micrographs of optimally grown NiO, on pure Ni
and Ni–0.1%Mo tapes, showed a flat NiO surface,
comparable to the original metal surface. How-
ever, such NiO quality is often locally degraded
with up to ꢀ50 nm high ridges sometimes found at
the Ni grain boundaries as shown in Fig. 7a. The
SEM pictures (Fig. 7b) indicate that polycrystal-
line NiO formed at Ni surface or at Ni grain
boundaries appeared as ridges and protrusions.
These defects gradually disappear as the oxidation
0
decreasing to about 3ꢀ and an in-plane FWHM of
5
0
ꢀ on pure Ni and of 2ꢀ and 7ꢀ respectively for Ni–
.1%Mo tape (Fig. 4).
A working window for the cube textured
growth of NiO on Ni and Ni–0.1%Mo tapes was
established from the extensive X-ray data, sum-
marised in Fig. 5 where the ratio of maximum
intensities R ¼ Ið200Þ/[ IðhklÞ ꢁ Ið200Þ] is plot-
ted against the oxidation temperature for samples
oxidized to a thickness of about 1–5 lm NiO (i.e.
oxidation times ranging from seconds at 1350 ꢀC
P

A. Kursumovic et al. / Physica C 405 (2004) 219–226
223
Fig. 4. (a) XRD h–2h scans for optimally grown SOE NIO on pure (0 0 2) Ni at 1250 ꢀC and on (0 0 2) Ni–0.1%Mo at 1350 ꢀC both
prepared in air; (b) the FWHM of the rocking curve for these optimally grown NiO on Ni is ꢀ3ꢀ while for on Ni–0.1%Mo is 2ꢀ; and (c)
(1 1 1) pole figures.
progressed. Nevertheless, preliminary results on
Ni–0.1%Mn tapes showed rather uniform NiO
growth with much fewer of those defects present.
On the lower temperature side, outside the re-
gime of optically transparent NiO, a non-trans-
parent NiO structure of small NiO grains is
formed with a rather rough surface. On the high
temperature side, however, secondary recrystalli-
zation of the Ni tape results in a degraded cube
texture even if the epitaxial relationship and opti-
cal transparency of NiO is maintained.
4. Discussion
Cube on cube (0 0 1)Æ1 0 0æk (0 0 1)Æ1 0 0æ
Ni
NiO
growth is the most probable arrangement for
these materials because this epitaxial relation

224
A. Kursumovic et al. / Physica C 405 (2004) 219–226
Ni [14]
4
3
2
1
10
10
10
10
10
Ni [this work]
Ni-0.1% Mo [this work]
0
Fig. 6. Optical micrograph of an SOE NiO/Ni sample oxidized
for 10 s at 1250 ꢀC in air. Note that NiO is fully transparent
here.
9
00
1000 1100 1200 1300 1400
o
temperature ( C)
P
Fig. 5. The intensity ratio Ið200Þ=½ IðhklÞ ꢁ Ið200Þꢂ of XRD
peaks for nucleation controlled SOE of thin NiO films grown
on pure (
either immediately by nucleation at the substrate
or from preferential grain growth as a slow diffu-
sion controlled process. Accordingly it is found
here that nucleation control, achieved by sup-
pressing NiO formation at lower temperatures, can
drastically improve the epitaxial growth of NiO,
on pure and microalloyed (0 0 1) Ni. Under such
conditions, even for the thinnest NiO films, large
area transparent thin single-crystal-like NiO film
ꢃ
) Ni and (m) microalloyed Ni tapes indicating the
effective working window. Additionally the results for thick
NiO layers ðꢄÞ grown on pure Ni tapes by preferred grain
growth epitaxy are shown as previously published [14].
corresponds to the smallest strain energy between
Ni and NiO [19]. However, the huge difference in
the lattice parameter (0.3522 nm for Ni and 0.4177
nm for NiO) makes direct epitaxial growth very
difficult. The texture of the SOE films is promoted
has been grown, in a temperature region around
the Ið200Þ/[ IðhklÞ ꢁ Ið200Þ] peaks in Fig. 5.
P
Fig. 7. (a) AFM picture of defect rich regions on 2.5 lm thick NiO film grown on (0 0 1) Ni for 120 s at 1250 ꢀC in air, with the inset
showing section analysis; and (b) SEM picture of a similar area with the inset showing polycrystalline NiO grains in a defective place.

A. Kursumovic et al. / Physica C 405 (2004) 219–226
225
For grain growth controlled NiO texturing, the
high intensity ratio versus temperature shows a
narrow peak which develops gradually in time, 2 h
at ꢀ1250 ꢀC in [14]. However, in the case of
nucleation controlled texture (this work) the high
intensity ratio occurs immediately as the oxidation
starts (Fig. 5) and it happens in much wider tem-
perature regime. In this latter case the NiO and Ni
grain size correspond to each other, Fig. 6. This is
in contrast to textured NiO layer grown by com-
petitive grain growth where the oxide is non-
transparent and its grain size is much smaller than
the substrate grain size [14,18]. However, there are
only a few reports [20] of sporadic single-crystal-
like NiO on Ni growth in the literature.
that activation energy for motion of Mo created
vacancies is mainly measured. At higher tempera-
tures thermally induced Ni vacancies eventually
dominate, increasing the DE value due to enthalpy
of vacancy creation. At temperatures below 1100
ꢀC grain boundary ‘‘short circuit’’ diffusion be-
comes significant decreasing DE. Yet, in optimally
formed thin ‘‘crystal clear’’ NiO films the Ni vol-
ume diffusion can be maintained at other temper-
atures as well [18], resulting in much slower NiO
growth at lower temperatures compared with grain
boundary assisted growth. The optimally grown
NiO film thus makes a stable template for the
further growth of other buffers and Y-123 depo-
sition as the rest of the coated conductor archi-
tecture. The surface of these films is very smooth
compared to the surface of SOE NiO established
by competitive grain growth [11–14] thus elimi-
nating the need for the polishing [21]. However, in
thin NiO films grown on Ni and Ni–0.1%Mo, high
ridges and protrusions were found on some grain
boundaries and on these grains respectively. It is
believed that the ridges are formed due to local
nucleation and initially faster growth of micro-
crystalline NiO. The growth rate of ꢅ1 lm sized
NiO grains (inset in Fig. 7b) is increased by en-
hanced Ni diffusion along the high angle NiO
grain boundaries even in this temperature regime
around 1250 ꢀC of predominantly volume diffu-
sion [14,18], while the low angle grain boundaries
are not easy diffusion paths and the oxide is
growing at a rate essentially governed by diffusion
of nickel through the oxide lattice [17]. These de-
fects disappear with time (while NiO film thickness
increases) due to the competitive grain growth of
epitaxial NiO grains [14,18]. Nevertheless pre-
liminary results show that microalloying with
0.1%Mn suppresses formation of polycrystalline
NiO grains. However, further work is necessary to
understand the nature of this single-crystal like
NiO nucleation and growth and the role of Mn.
Microalloying, of Ni with 0.1% Mo or 0.1%
Mn, is primarily used to stabilise Ni grain growth
and resist secondary recrystallization during Ni
texturing and subsequent coated conductor pro-
cessing. Onset of secondary recrystallization in
(0 0 1) Ni tapes is displaced to much higher tem-
peratures when oxidation occurs [18], and micro-
alloying displaces the onset even further by ꢀ100
ꢀ
C allowing high quality cube textured NiO
growth at higher temperatures as shown in Fig. 5.
At high temperatures the Ni volume (lattice)
diffusion in NiO dominates, therefore the para-
bolic rate constant is directly related to the
mobility of the diffusing species, i.e. to the self-
diffusion coefficient of Ni in bulk NiO, k
[14], since oxygen diffusion is much slower [16].
The Ni diffusion in pure NiO occurs via Ni
p
/ DNi
vacancies, resulting from the Ni1ꢁyO non-stoichi-
ꢁ
4
ꢁ3
ometry with y ¼ 10 –10 which is relatively low
[16]. NiO doping with elements of higher valence
than Ni results in dopant ions substituted on a
normal Ni site, requiring a corresponding num-
ber of vacant Ni sites in order to achieve elec-
trical neutrality [16]. Therefore, even a small
2þ
2þ
>
2þ
concentration (ꢀ0.1%) of Mo
ions in NiO lat-
tice can significantly increase the number of cation
vacancies over the equilibrium density of thermally
created Ni vacancies in pure NiO. This results in a
much increased diffusion rate [16]. Thus, in the
regime from 1100 to 1350 ꢀC, an Arrhenius plot
for volume diffusion of Ni in Mo doped NiO
indicates a lower apparent activation energy DE
than in the case of pure NiO [14,18], due to the fact
5. Conclusions
Cube on cube NiO was grown on pure and
microalloyed (0 0 2)Ni. Microalloying enabled the
growth of epitaxial NiO with a much wider

226
A. Kursumovic et al. / Physica C 405 (2004) 219–226
[
[
5] M. Paranthaman, D.F. Lee, A. Goyal, E.D. Specht, P.M.
working window compared to pure Ni tapes. A
single crystal like growth of NiO on cube textured
Ni, Ni–0.1%Mo and Ni–0.1%Mn tapes was
achieved even for rather thin ꢀ1 lm NiO films at
optimum texturing conditions. This is believed to
be due to NiO nucleation rather than to texturing
by preferred grain growth. This compact NiO
layer is a valuable first buffer layer for coated
conductors applications.
Martin, X. Cui, J.E. Mathis, J. Feenstra, D.K. Christen,
D.M. Kroeger, Supercond. Sci. Technol. 12 (1999) 319.
6] R.I. Tomov, A. Kursumovic, M. Majoros, D.-J. Kang,
B.A. Glowacki, J.E. Evetts, Supercond. Sci. Technol. 15
(2002) 598.
[
[
[
7] A. Atkinson, R.I. Taylor, Philos. Mag. A 43 (1981) 979.
8] F. Czerwinski, J.A. Szpunar, Acta Mater. 46 (1998) 1403.
9] R. Haugsrud, Corros. Sci. 44 (2002) 1569.
[
[
10] A. Ginsbach, R. Schneider, H.W. Grueniger, G. Grabe,
IEEE Trans. Magn. 27 (1991) 1410.
11] K. Matsumoto, K. Osamura, I. Hirabayashi, Extended
Abstract of 2001, International Workshop on Supercon-
ductivity, Co-Sponsored by ISTEC and MRS, Honolulu,
Hawaii, USA, 24–27 June 2001, p. 39.
Acknowledgements
[
[
[
[
12] Z. Lockman, W. Goldacker, R.B. Nast, B. De Boer, J.L.
MacManus-Driscoll, IEEE Trans. Appl. Supercond. 11
This research was funded by the Engineering
and Physical Sciences Research Council (EPSRC).
The support of ABB is also gratefully acknowl-
edged.
(
2001) 3325.
13] D. Selbmann, J. Eickemeyer, H. Wandrock, C. Jimenez, S.
Donet, F. Weiss, U. Miller, O. Stadel, J. Phys. IV France
1
1 (2001) 239.
14] A. Kursumovic, R. Tomov, R. H u€ hne, B.A. Glowacki, J.E.
Evetts, A. Tuissi, E. Villa, B. Holzapfel, Physica C 385
(
2003) 337.
15] M.J. Graham, M.J. Cohen, Electrochem. Soc. 120 (1972)
79.
References
8
[
1] A. Goyal, D.P. Norton, J. Budai, M. Parantharam, E.D.
Specht, D.M. Kroeger, D.K. Christen, Q. He, B. Saffian,
F.A. List, D.F. Lee, P.M. Martin, C.E. Klubunde, E.
Hatfield, V.K. Sikka, Appl. Phys. Lett. 69 (1996) 1795.
2] W.B. Hutchinson, H.-E. Ekstr o€ m, Mater. Sci. Technol. 6
[16] P. Kofstad, Nonstoichiometry, Diffusion, and Electrical
Conductivity in Binary Metal Oxides, Wiley-Interscience,
New York, 1972.
[17] M.J. Graham, R.J. Hussey, M. Cohen, J. Electrochem.
Soc. 120 (1973) 1523.
[
[
(1990) 1130.
[18] A. Kursumovic, R. H u€ hne, R. Tomov, B. Holzapfel, B.A.
Glowacki, J.E. Evetts, Acta Mater. 51 (2003) 3759.
[19] B. Pieraggi, R.A. Rapp, Acta Metall. 36 (1988) 1281.
[20] D. Caplan, M.J. Graham, M. Cohen, J. Electrochem. Soc.
119 (1972) 1205.
3] B. De Boer, J. Eickemeyer, N. Reger, G.-R. Fernandez, J.
Richter, B. Holzapfel, L. Schultz, W. Prusseit, P. Berbe-
richt, Acta Mater. 49 (2001) 1421.
[
4] J. Eickemeyer, D. Selbmann, R. Opitz, B. de Boer, B.
Holzapfel, L. Schutz, U. Miller, Supercond. Sci. Technol.
[21] K. Matsumoto, I. Hirabayashi, K. Osamura, Physica C
378 (2002) 922.
14 (2001) 152.