Thin La2Zr2O7 films made from a water-based solution

Article abstract of DOI:10.1016/j.jssc.2008.08.031

Thin films of La₂Zr₂O₇ (LZO) are highly regarded as possible buffer layers in the coated conductor configuration. This report describes a new synthesis for thin crystalline LZO films, based on a largely water-based solutio

Full text of DOI:10.1016/j.jssc.2008.08.031

ARTICLE IN PRESS
Journal of Solid State Chemistry 182 (2009) 37–42
Contents lists available at ScienceDirect
Journal of Solid State Chemistry
journal homepage: www.elsevier.com/locate/jssc
Thin La2Zr2O7 films made from a water-based solution
Ã
V. Cloet a, , J. Feys a, R. Hu¨hne b, S. Hoste a, I. Van Driessche a
a Ghent University, Krijgslaan 281–S3, 9000 Gent, Belgium
b Institute for Metallic Materials, IFW Dresden, Helmholtzstrasse 20, 01069 Dresden, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
Thin films of La2Zr2O7 (LZO) are highly regarded as possible buffer layers in the coated conductor
configuration. This report describes a new synthesis for thin crystalline LZO films, based on a largely
water-based solution, mainly containing metal acetates, acetic acid and an organic amine-base:
triethanolamine. Initially, a thin layer of amorphous material is deposited on the textured Ni-5 at%W
substrate by means of dip-coating. Only by careful control of the thermal treatment can the layer
be transformed into a crystalline layer. Important parameters in this respect are the heating rate and the
dwell time. The amorphous gel is analysed by HR-TGA/DTA and HR-TEM. The textured layers are
analysed by XRD, pole figures, RHEED, AFM and SEM.
Received 3 June 2008
Received in revised form
26 August 2008
Accepted 31 August 2008
Available online 9 October 2008
Keywords:
Coated conductor
Sol–gel
& 2008 Elsevier Inc. All rights reserved.
La2Zr2O7 films
Dip-coating
1. Introduction
production costs must be as low as possible. The technique
used for manufacturing these thin films will play a crucial
role in achieving this goal. Vacuum techniques, which are
expensive and difficult to up-scale, are therefore preferably
avoided. An alternative technique that is being explored
more often and applied with increasing success, is the chemical
solution deposition (CSD) technique [5]. This technique does not
require vacuum, has a higher deposition rate and uses widely
accessible starting materials. Moreover, the composition of the
solutions can be changed easily, which renders this method
versatile. One of the implementations that is comprised under the
general term CSD, is the sol–gel route. The starting sol(ution) can
be transformed into a gel by heating at low temperatures
(o100 1C) in ambient atmosphere. In the gel, all metal ions are
homogeneously dispersed and prevented from precipitation by
additives and pH adjustment [6].
So far, coated conductors possess the highest potential
to solve the problems of up-scaling high temperature super-
conductor (such as YBa2Cu3O7ꢀx
)
applications.
A
coated
conductor configuration addresses the two major problems of
HTS: their brittle ceramic nature and the need for textured
crystals. This configuration consists of a flexible metallic tape
coated with thin protective films—called buffer layers—and
finally coated with a thin film of YBCO. Ion beam assisted
deposition (IBAD) and rolling assisted biaxially textured sub-
strates (RABiTS) are the two most commonly used substrates, the
latter being most promising for long-length production [1].
Potential materials for use as buffer layers need three distinctive
qualities. First of all, the lattice parameter of the buffer layer must
be similar to, or commensurate with, those of both substrate and
YBCO. Only then can the buffer layers grow epitaxially on a
textured substrate. Secondly, the thermal expansion coefficients
of buffer layer and substrate must be comparable in order to avoid
cracking and delamination during film growth. Finally, the buffer
layer must act as an effective cation and/or oxygen diffusion
barrier to prevent chemical interaction between adjacent materi-
als. The most successful buffer layers reported so far are CeO2,
Y2O3-stabilised ZrO2 (YSZ), Y2O3, MgO and La2Zr2O7 (LZO) [2–4].
In order for this technology to be economically viable, the
This publication is concerned with the development and
use of a water-based sol–gel route for the production of epitaxial
LZO films on NiW-RABiTS tape. This method creates some
additional advantages over the usually organic solvent-
based CSD technique. Water is
a safe, widely accessible,
cheap and environmentally friendly solvent. Several coating
techniques (dip-coating, spin-coating, ink-jet printing, etc.) can
be applied to deposit the solution on the substrate [7,8]. This
publication explores dip-coating of a sol–gel solution for the
production of LZO thin films. Dip-coating allows the use of
substrates of different geometries and sizes, coats both sides
simultaneously and allows continuous processing in a reel-to-reel
system.
Ã
Corresponding author. Fax: +32 92 644 983.
E-mail address: veerle.cloet@ugent.be (V. Cloet).
0022-4596/$ - see front matter & 2008 Elsevier Inc. All rights reserved.
doi:10.1016/j.jssc.2008.08.031

ARTICLE IN PRESS
38
V. Cloet et al. / Journal of Solid State Chemistry 182 (2009) 37–42
2. Method
epitaxial growth of the films were investigated using standard
X-ray diffraction. For acquiring –2 -scans, a Siemens (D5000)
was used. Pole figures, -scans and -scans were collected with a
y
y
f
o
2.1. Precursor solution
Bruker, AXS Discover. These measurements allow the determina-
tion of the full-width at half-maximum (FWHM) of the out-of-
plane and the in-plane distribution of the oxide layer. Further
analysis of the surface crystallinity was carried out by reflection
high energy electron diffraction (RHEED, Staib Instruments). This
technique allows analysis of the top 5 nm of the surface. An
The LZO solution was prepared by dissolving 1.6 g lanthanum
(III) acetate hydrate (La(CH3COO)3.1H2O, 99.9% Aldrich) in a
mixture of 18 mL distilled water and 6.3 mL glacial acetic acid
(CH3COOH, 99.8%, Aldrich) at 90 1C. An equimolar quantity of
zirconium (IV) hydroxide acetate hydrate (Zr(OH)3(CH3COO).H2O,
Aldrich) was added to this mixture, after all lanthanum acetate
was dissolved. Upon cooling the solution to room temperature,
triethanolamine (N(CH2CH2OH)3, 99+%, Acros) was added. Finally,
the aciditiy of the solution was decreased up to a value of 7 by
adding ammonium hydroxide (25wt%, Chemlab). This procedure
has been previously described in detail for the production of thin
YBa2Cu3O7-layers [9].
electron energy beam of 30 keV and a beam current of 50 mA was
applied under a grazing incident angle (0.5–1.51) to the sample.
The diffraction pattern was recorded using a CCD camera and
analysed with a computer program based on the kinematic theory
of electron scattering. The surface microstructure was analysed
with FEG–SEM (LEO 1530 Gemini ) and its roughness by atomic
force microscopy (AFM, molecular imaging, PicoPlus) in AC
tapping mode. For transmission electron microscopic (TEM)
analysis, a JEM 2200FS (JEOL) was used to obtain the images.
Copper grids with mesh 2000 were immersed in the precursor
solution and subsequently heated to the desired temperature
under forming gas atmosphere.
2.2. Substrate and coating
The Ni-5 at%W tape (evico GmbH, Germany) (80 mm thick) was
cut into strips of approximately 2.5 cm length. Prior to dip-
coating, the substrates needed cleaning to enhance wettability. In
a first step these substrates were thermally cleaned at 800 1C
(60 min) under Ar-5% H2 atmosphere. After this, the substrates
were chemically cleaned in different solvents. This cleaning
procedure is described elsewhere in more detail [10]. The
substrates were preserved in methanol to prevent degradation
of the wettability and were used as soon as possible.
3. Results and discussion
3.1. TGA–DTA
In order to obtain information about the decomposition
behaviour of the gels and to adjust the thermal treatment
accordingly, a TGA–DTA analysis of the gels was carried out. In
Fig. 1 the TGA–DTA spectrum until 1000 1C is shown with an insert
of a HR-TGA until 500 1C. At temperatures below 200 1C, two large
decomposition steps occur in rapid succession. These two steps
can be associated with the release of water, ammonia and acetic
acid. These products are the main constituents of the gel and are
most likely to evaporate quickly since their boiling point is lower
than 200 1C. Soon after, at temperatures ranging from 200–500 1C,
gases such as CO and CO2 escape from the layer [12,13]. As can be
seen from the thermogravimetric plot, there is no weight loss
above 4501C, only a small exothermic peak can be seen in the DTA
curve at 7801C. This corresponds most likely to the formation of
LZO oxide out of La2O3 and ZrO2. These findings are also confirmed
by the TGA–DTA plots of lanthanum acetate and zirconium
hydroxy acetate, in which it is clear that La2O3 and ZrO2 are being
formed at approximately 7501C (not displayed here).
The substrates were dip-coated in a cupboard with a laminar
flow classified as Class 10 according to the Federal Standard 209
and mounted in a Class 1000 clean room. The results presented
here, were obtained after exploration of several variables such as
dip-coat velocity, number of dip-coated layers, presence of seed
layers. One could conclude that higher dip-coating velocities lead
to thicker but less textured layers. In this research, in contrast to
previous research [11], the presence of seed layers, did not
improve the results. Therefore, the results of only one single layer,
are discussed in this paper. A solution with a total metal
concentration of 0.4 M was used for dip-coating. Dip-coating
was carried out at room temperature with a dip-coat velocity of
20 mm/min and the substrates were withdrawn from the solution
under an angle of 901. The substrates were immersed in the
solution for about 30 s before they were withdrawn. After
deposition of the thin film, the substrate was dried in horizontal
position at 60 1C for about 30 min. The colour change from opaque
to yellow during gelation can be attributed to a colour change of
the triethanolamine upon heating. A thermal treatment at high
temperatures was needed to convert the amorphous gel to a
crystalline textured film. The thin film was heated to 900 1C with a
heating rate of 5 1C/min. After 50 min dwell time, the temperature
was rapidly increased to 1050 1C and maintained constant for
50 min. This thermal treatment was carried out entirely under
forming gas (Ar-5%H2) atmosphere. The gas flow was kept
constant at a value of 0.4 L/min. The sample was cooled to room
temperature at a rate of 10 1C/min and under continuous flowing
of forming gas.
3.2. TEM
The transition from gel state to crystalline material is analysed
by high resolution TEM of LZO powder. In Fig. 2, an overview of
TEM micrographs of the gels at different temperatures is
displayed. The dip-coated copper grids were heated to tempera-
tures at which a weight loss was observed in the high resolution
TGA. In this way, we can follow the behaviour of the gel and the
transition into crystalline powder. Gels produced at low tempera-
tures (60 1C) are unstable and quickly damaged by the electron
beam, as the illuminated gel tends to blow up and burst open.
Discrete particles are absent in the early gel stages until 305 1C. At
360 1C, small particles (10–50 nm) can be distinguished in the gel.
At higher temperatures, larger particles form agglomerates. The
particles formed at 820 1C are crystalline and have an average size
of 30 nm. These findings are in good agreement with the
conclusions drawn from the TG–DT analysis. Elemental analysis
with electron energy-loss spectroscopy (EELS) will allow us to
investigate the nature of the particles in the gels at 360 1C.
2.3. Characterisation
The decomposition behaviour of the LZO gel was investigated
by high resolution thermogravimetric analysis (TGA, TA-Instru-
ments, Q500). An air flow on the sample of 60 mL/min was used.
The maximal heating rate was 20 1C/min. Differential thermal
analysis (DTA) was coupled to TGA to obtain information on
physical changes that do not involve mass loss. Crystallinity and

ARTICLE IN PRESS
V. Cloet et al. / Journal of Solid State Chemistry 182 (2009) 37–42
39
8
100
90
80
70
60
100
HR-TGA: TA Instruments, Belgium
80
60
6
4
40
20
2
100 200 300 400 500
Temperature (°C)
0
-2
-4
-6
-8
TGA
DTA
200
400
600
Temperature (°C)
800
1000
Fig. 1. HR-TGA of LZO gel.
Fig. 2. TEM analysis of gel to solid phase transition.
˚
˚
3.3. XRD and RHEED
LZO has a cubic pyrochlore structure with a theoretical lattice
(10.77 A) and the one calculated from LZO bulk (10.79 A).
This lattice parameter results in a lattice mismatch of 0.5% and
1.8% with YBCO (resp. a- and b-axis). The lattice mismatch with
the Ni-substrate is slightly higher (7.6%) but still allows the
˚
parameter of 10.79 A, which is in good agreement with the lattice
parameter calculated from the (0 0 4) peak of the thin film
growth of epitaxial LZO. In Fig. 3 a y–2y scan of LZO–NiW is

ARTICLE IN PRESS
40
V. Cloet et al. / Journal of Solid State Chemistry 182 (2009) 37–42
Ni K
Ni (001)
2000
1500
1000
500
0
LZO (002)
1500
1000
500
0
(004)
Cu Kβ
40
(222)
(008)
28
30
32
34
10
20
30
50
60
70
80
2 Theta (°)
2 Theta (°)
300
250
200
150
100
50
30
40
50
60
70
Theta (°)
Fig. 3. Left:
y–2y
scan of LZO–NiW, right: rocking curve of (0 0 4) with FWHM ¼ 8.81.
phiscan (111) LZO
100
25
0
30405060708090100 120 140 160 180 200 220 240 260 280 300 320 340 360 380
Position [°PHI]
Fig. 4. Left: (2 2 2) pole figure of LZO, right:
f-scan of (2 2 2).

ARTICLE IN PRESS
V. Cloet et al. / Journal of Solid State Chemistry 182 (2009) 37–42
41
shown. As can be seen in this figure, only the (0 0 4) peak at
¼ 33.31 is present. LZO has only two (0 0 l) reflections that can
be observed in the range of 0–801 (Fig. 3b); the (0 0 4) orientation
at 2
2y
y
¼ 33.31 and the less intense (0 0 8) orientation at 2y ¼ 69.81.
This XRD spectrum dates from a previous sample and contains a
small (111) peak. From Fig. 3a we can conclude that an epitaxial
LZO layer is obtained as a result of the heat treatment described
above. All c-axes of the LZO crystals are aligned perpendicular
to the plane of the substrate. In this case, the (2 2 2) orientation
(at 2
y ¼ 28.591) has apparently been completely suppressed. The
Fig. 6. Left: AFM micrograph, right: SEM micrograph at magnification ¼ 50 K.
rocking curve of the (0 0 4) peak shows a FWHM of 8.81. To check
the in-plane misalignment of the thin film, (2 2 2) pole figures
(at 2
y ¼ 28.61) are taken. The FWHM of the peaks is 6.651 (Fig. 4).
For the epitaxial growth to be successful, the surface of the thin
film must be well textured too. This can only be significantly
measured with RHEED analysis. The highly energetic electrons
used in RHEED analysis, penetrate only the top 5 nm which
renders this technique surface sensitive. The pattern was indexed
˚
using a 10.6 A lattice parameter for LZO. The RHEED pattern
(Fig. 5) shows a well defined array of bright spots, which indicates
that the top layer of the buffer layer is perfectly textured and
crystalline. Therefore, we can conclude that the texture of the
Ni-substrate was translated into a biaxial texture of the LZO buffer
layer. The biaxial texture can in principle be passed on to the
next layer.
Fig. 7. Left: (111) pole figure of NiW/LZO/CeO2, right: (10 3) pole figure of NiW/
LZO/CeO2/YBCO.
of LZO/CeO2 have a FWHM of 9.11. Comparing this to the 6.651 of
Ni/LZO, we can conclude that the in-plane misorientation has
increased a bit, but is still acceptable. For the actual super-
conducting layer, a (10 3) pole figure was taken and showed a
FWHM of 8.71. This shows the texture has been transfered
successfully from the LZO layer up to the YBCO layer (Fig. 7).
3.4. SEM and AFM
The surface roughness, analysed by AFM, is given by the rms-
value. In Fig. 6 the AFM micrograph is shown. Over an area of
25 m
m2, the rms-value is 11.3 nm and the average height is
42.2 nm. Upon omitting the large impurtity, the rms-value is
further decreased to 8.66 nm and the average height is only
20.6 nm. These results give a better indication of the real surface
morphology. In the SEM micrograph, at a magnification of 50 K,
the small grains of LZO are visible. The overall surface of the LZO
film is smooth and homogeneous.
4. Conclusion
The 0.4 M solution used in this research is composed of water,
metal acetates, triethanolamine, acetic acid and ammonia and has
been successfully used to produce epitaxial layers on NiW
substrates. By applying a carefully chosen heat treatment, the
growth of (2 2 2) oriented LZO crystals could be suppressed.
The epitaxial films exhibit an out-of-plane FWHM of 8.81. The
in-plane orientation of the crystals can be linked to the FWHM of
the (2 2 2) orientation, which is 6.651 and sufficiently small to
deposit the next layer in the coated conductor configuration. The
surface of the LZO layer was found to be smooth with only small
inhomogeneities. TEM analysis of the gel state shows that until
360 1C the homogeneity of the gel is maintained and no
precipitates are formed. At 360 1C, however, small inhomogene-
ities (15–50 nm) are found in the gel matrix. The finally obtained
LZO solid grains have an average size of 30 nm and are found to be
crystalline from 820 1C onwards. Further elemental analysis
(electron energy-loss spectroscopy) of the inhomogeneities at
360 1C is under way to indicate whether these are small LZO nuclei
or carbon residues. The PLD deposition of CeO2 and YBCO showed
that the biaxial texture could be maintained throughout the
subsequent layers. A next step will be to deposit CeO2 and YBCO
by a similar sol–gel synthesis route as the LZO layer.
3.5. Thin CeO2 and YBa2Cu3O7 films
In the final coated conductor structure, a thin film of CeO2 and
YBa2Cu3O7 must be deposited on top of the LZO film. It is
therefore important to test the compatibility of this LZO film with
the following layers in reality. In this research, a 50 nm thick CeO2
and a 300 nm thick YBCO layer are deposited by pulsed laser
deposition (PLD) [14,15]. The peaks in the (111) pole figure
Acknowledgments
For realisation of XRD and pole figure analysis, the authors
would like to thank Olivier Janssens (Ghent University, Belgium).
For AFM analysis, the authors would like to thank Danny
Vandeput (Ghent University, Belgium). The authors would also
Fig. 5. RHEED pattern, view along the /10 0S and /110S.

ARTICLE IN PRESS
42
V. Cloet et al. / Journal of Solid State Chemistry 182 (2009) 37–42
like to thank Sebastian Engel (IFW, Dresden, Germany) for SEM
analysis. For receiving Ni-W tapes, the authors would like to thank
Trithor GmBH. One of us (V.C.) wish to thank the Institute for the
Promotion of Innovation through Science and Technology in
Flanders (IWT-Vlaanderen) for funding this research.
[5] M.S. Bhuiyan, M. Paranthaman, K. Salama, Superconductor Sci. Technol. 19
(2006) R1–R21.
[6] J. Livage, M. Henry, C. Sanchez, Prog. Solid State Chem. 18 (1988)
259–341.
[7] V. Cloet, M.C. Cordero-Cabrera, Adv. Sci. Technol. 47 (2006) 153–158.
[8] S. Morlens, L. Ortega, B. Rousseau, S. Phok, J.L. Deschanvre, P. Chaudouet, et al.,
Mater. Sci. Eng. B 104 (2003) 185–191.
[9] B. Schoofs, T. Mouganie, B.A. Glowacki, V. Cloet, S. Hoste, I. Van Driessche,
J. Sol–Gel Sci. Technol. 41 (2007) 113–122.
References
[10] G. Penneman, I. Van Driessche, E. Bruneel, S. Hoste, Euro Ceram. Viii, Pts 1–3
264–268 (2004) 501–504.
[11] X.B. Zhu, L. Chen, S.M. Liu, W.H. Song, Y.P. Sun, K. Shi, et al., Phys.
C Superconductivity 415 (2004) 57–61.
[12] L. Fabbrini, I. Rossetti, L. Forni, Appl. Catal. B Environ. 56 (2005) 221–227.
[13] G. Hussein, J. Therm. Anal. Calorimetry 42 (1994) 1091–1102.
[14] S. Engel, K. Knoth, R. Huhne, L. Schultz, B. Holzapfel, Superconductor Sci.
Technol. 18 (2005) 1385–1390.
[1] J. Eickemeyer, D. Selbmann, R. Opitz, H. Wendrock, E. Maher, U. Miller, et al.,
Phys. C Superconductivity 372–376 (2002) 814–817.
[2] T.G. Chirayil, M. Paranthaman, D.B. Beach, D.F. Lee, A. Goyal, R.K. Williams,
et al., Phys. C Superconductivity 336 (2000) 63–69.
[3] K. Knoth, R. Huhne, S. Oswald, L. Schultz, B. Holzapfel, Acta Materialia 55
(2007) 517–529.
[4] M.S. Bhuiyan, M. Paranthaman, K. Salama, Superconductor Sci. Technol. 19
(2006) R1–R21.
[15] M.A. Arranz, B. Holzapfel, N. Reger, J. Eickemeyer, L. Schultz, Phys.
C Superconductivity 366 (2002) 109–116.