Qiu X.G. (Ed.) High Temperature Superconductors
Подождите немного. Документ загружается.
© Woodhead Publishing Limited, 2011
Preface
Almost 25 years have passed since the celebrated discovery of high temperature
superconductors (HTSCs) by G. Bednorz and K. A. Müller in 1986. HTSCs
opened a new era for research on superconductivity, from both fundamental
physics and materials science perspectives. In the eld of fundamental physics,
the research on HTSCs has greatly enriched our understanding of the physics in
strongly correlated electron systems, although many crucial issues remain to be
understood. In the materials arena, great achievements have been made over the
last two decades, both in single crystals and in thin lms. It is important to
acknowledge that the advances are not limited to the sample quality itself, but also
include new methods and techniques developed to grow HTSC thin lms. For
example, it is because of the research on HTSC thin lms that the fabrication
techniques for oxide thin lms have been able to be developed so fast; facilitating
the growth of oxide thin lms other than HTSC, such as the SrTiO
3
/LaAlO
3
system used in interfacial conductivity studies. As time has passed, the fervour
over HTSCs has gradually faded away. Now, therefore, is the right time to consider
what has been achieved and to summarize the information that will be of help to
students and colleagues working at the frontiers of research into oxide thin lms
in general and superconducting thin lms in particular. Therefore, we invited
researchers from all over the world, who have been working for many years on
different aspects of HTSC thin lms, to contribute their cutting edge knowledge
on the fabrication and physics of HTSC thin lms, reviewing the fabrication
techniques and physical properties, as well the applications, of HTSC thin lms.
The objective of this book is to serve as a handbook for those who are interested
in the growth of HTSC thin lms as well as their characterization and applications.
It aims to give newcomers such as students a fast track means of understanding
what has been achieved in the past and to provide an overview of the status of
current research on HTSC thin lms. The book will also be a valuable source of
references for their research. It is also aimed at the veterans in this eld; providing
a convenient and consolidated source of information, so avoiding the necessity to
search through the numerous references that have been published over the last two
decades.
This book is organized into three parts. The rst three chapters cover the
fundamentals of the materials and physical aspects of HTSCs. Chapter 1 overviews
xi
© Woodhead Publishing Limited, 2011
xii Preface
the different deposition techniques used for thin lm processes as well as the
characterization techniques. Chapters 2 and 3 summarize the transport and optical
properties of HTSC, focusing mainly on the results obtained on thin lms. Part II
covers the fabrication and characterization of different HTSC thin lms. As we
know, HTSCs can be categorized into hole-doped and electron-doped thin lms.
Depending on the material and properties of interest, different deposition
techniques such as magnetron sputtering and laser ablation are utilized. In
Chapters 4 to 7, deposition techniques for several representative hole-doped and
electron-doped HTSC thin lms are reviewed in detail. The characterization and
physical properties of those lms is also discussed. Part III of the book describes
the application of HTSC thin lms. After more than 20 years of development,
there has been much progress in HTSC applications, especially in the area of
weak signal detecting using thin lms. In Chapters 8 to 10, Josephson junctions,
SQUIDs and microwave lters based on HTSC thin lms are introduced and
discussed in detail.
At the time when we started to prepare this book, a new family of HTSCs, i.e.
iron pnictide superconductors, had just been discovered. It is anticipated that the
thin lm fabrication techniques developed for HTSC cuprates can also be applied
to the growth of iron pnictide superconductors. Of course, the thin lm fabrication
techniques described in this book will also be helpful for those working on oxide
thin lms other than superconductors.
Finally, we would like to thank Woodhead Publishing Limited who have made
it possible to have such a handbook published at a time when the superconductivity
community is wondering what to do next with HTSCs; we hope that readers will
nd this book of help to their researches.
Xianggang Qiu
1
2
3
4
5
6
7
8
9
10
1
2
3
4
5
6
7
8
9
20
1
2
3
4
5
6
7
8
9
30
1
2
3
4
5
6
7
8
9
40
1
2
43X
© Woodhead Publishing Limited, 2011
3
1
Deposition technologies, growth and
properties of high-T
c
films
R. WÖRDENWEBER, Jülich Research Centre GmbH, Germany
Abstract: Due to the extreme demands of high-temperature superconductor
(high-T
c
HTS) materials – especially the extremely complex structure, small
coherence length and large anisotropy – existing deposition technologies have
been greatly improved and novel technologies have been developed in the last
decade. This article presents an overview on the deposition and deposition
techniques of high-T
c
material. The major technological approaches are
discussed and compared in the light of the demands of the different
superconductor and substrate materials. Nucleation and film growth of ceramic
superconducting material are sketched, and consequences of the heteroepitaxial
growth are discussed. Finally, a classification of the state of the art and the pros
and cons of the different technologies are given.
Key words: high-temperature superconductor thin film deposition, ceramic
films, epitaxial growth of oxide films, critical thickness.
1.1 Introduction
Since the discovery of the first high-temperature superconductor (HTS) (Bednorz
and Müller, 1986), a significant effort has been put into the research and realisation
of textured and epitaxial HTS films. This effort is motivated largely by the potential
applications of thin HTS films in a number of cryoelectronic devices and by the
possibility of using epitaxial single or multilayer HTS films to study new physical
properties of these unique 2D layered materials. The technology of devices and
integrated circuits fabricated from low-temperature superconductors (LTS) capable
of operating at or near the temperature of liquid helium (4.2K) is by now well
established. However, the future application of HTS in cryoelectronic devices which
are capable of operating at temperatures between 20–77K will strongly depend on
the development of reproducible deposition technologies for high-quality single and
multilayer HTS thin films. Nevertheless, a number of technical applications of HTS
thin films (mainly passive (linear) and active (nonlinear) devices or circuits) already
have a firm implementation base for the next few years, while others represent
tentative ‘dreams’ for a more distant future. Typical application areas include:
metrology and electronic instrumentation, radioastronomy and environmental
spectroscopy (atmosphere and space), neurology and medical diagnostics, electronic
warfare (radar, electronic countermeasures (ECM), magnetic anomaly detection);
non-destructive materials evaluation (NDE/NDT), geological and environmental
prospecting, telecommunication, ultrafast digital signal and data processing.
4 High-temperature superconductors
1
2
3
4
5
6
7
8
9
10
1
2
3
4
5
6
7
8
9
20
1
2
3
4
5
6
7
8
9
30
1
2
3
4
5
6
7
8
9
40
1
2
43X
© Woodhead Publishing Limited, 2011
In this chapter the deposition of HTS thin films and film systems is reviewed.
Starting with a sketch of the requirements of HTS material and substrates, with
brief introductions to the most commonly used deposition techniques, a number
of important issues and problems related to a reproducible fabrication of high
quality HTS thin film devices for technical applications and basic research are
addressed.
1.1.1 HTS materials and requirements
The most commonly examined HTS material with superconducting transition
temperatures T
c
above the temperature of liquid nitrogen are ReBa
2
Cu
3
O
7–
δ
(Re = Y or rare earth element, typical transition temperature T
c
= 90–95K),
Bi
2
Sr
2
Ca
1
Cu
2
O
8
(T
c
= 90K), Bi
2
Sr
2
Ca
2
Cu
3
O
10
(T
c
= 120K), Tl
2
Ba
2
Ca
1
Cu
2
O
8
(T
c
= 110K), Tl
2
Ba
2
Ca
2
Cu
3
O
10
(T
c
= 127K), HgBa
2
Ca
1
Cu
2
O
8
(T
c
= 134K), MgB
2
(T
c
= 39K), and, lately, the so-called ferropnictide superconductors (e.g.
Sm
0.95
La
0.05
O
0.85
F
0.15
FeAs with T
c
= 57.3K) (Zheng Wie et al., 2008). For HTS
cryoelectronic devices, thin films are mainly grown from YBa
2
Cu
3
O
7–
δ
(YBCO)
material. Reasons for this choice are, among others, the phase stability, high
crystalline quality, high flux pinning level, low surface resistance, the possibility
of using a single deposition step with in-situ oxidation, and the absence of
poisonous components that are present in Tl, Hg, or As containing HTS films.
In general, the improvement of the superconducting properties (e.g. increase
of T
c
and energy gap
Δ
E) is accompanied by an enhancement of (i) the number of
elements, (ii) structural complexity, and (iii) anisotropy (Fig. 1.1).
1.1 The improvement of the superconducting properties (e.g. increase
of T
c
and energy gap) is achieved via an enhancement of complexity
(structural and stoichiometric) and anisotropy of the superconductor.
Deposition technologies, growth and properties of high-T
c
films 5
1
2
3
4
5
6
7
8
9
10
1
2
3
4
5
6
7
8
9
20
1
2
3
4
5
6
7
8
9
30
1
2
3
4
5
6
7
8
9
40
1
2
43X
© Woodhead Publishing Limited, 2011
First, the quasi two-dimensional (2D) nature of the oxide superconductors
related to their layered structure leads to a large anisotropy in nearly all parameters.
Anisotropy factors of
γ
=
ξ
ab
/
ξ
c
=
λ
c
/
λ
ab
≈ 6–8 in YBCO,
γ
≈ 55–122 in Bi
compounds, and
γ
≈ 100–150 in Tl compounds are reported for the anisotropy
between the crystallographic c and a-b direction (Farrell et al., 1988, 1989, 1990;
Okuda et al., 1991). Furthermore, extremely small coherence lengths are
determined for the HTS. For Bi
2212
values of
ξ
c
= 0.02–0.04 nm and
ξ
ab
= 2–2.5 nm,
for YBCO
ξ
c
= 0.3–0.5 nm and
ξ
ab
= 2–3 nm are measured, which, furthermore,
strongly depend on sample quality and oxygen content (Ossandon et al., 1992).
Therefore, local variations in the sample properties on the scale of
ξ
will
automatically result in a spatial variation of the superconducting properties. Due
to the extremely small coherence length along the crystallographic c-axis, only
the current directions along the a–b direction (i.e., along the CuO-planes) can be
used in most applications. Since even misalignments in the a–b plane lead to grain
boundaries that strongly affect the critical properties (see Fig. 1.2), a perfect
biaxial c-axis orientated epitaxial film growth including the avoidance of anti-
phase boundaries is necessary for most applications.
Second, the superconducting properties strongly depend upon variations in
stoichiometry, especially in the oxygen concentration. Due to the complex
crystallographic structure and the small coherence length, extreme requirements
are imposed upon the uniformity and stability of the deposition process.
Finally, in a number of applications HTS films are grown on less ideal
carriers, e.g. single crystals with large lattice misfit, or, in the case of coated
conductors, tapes with crystalline buffer layers. In these cases, polycrystalline or
1.2 Critical current density of YBCO bicrystals as a function of
misorientation angle at 4.2K (Dimos et al., 1990; Hilgenkamp et al., 1998).
Open symbols represent J
c
values of epitaxial films on single crystals.
6 High-temperature superconductors
1
2
3
4
5
6
7
8
9
10
1
2
3
4
5
6
7
8
9
20
1
2
3
4
5
6
7
8
9
30
1
2
3
4
5
6
7
8
9
40
1
2
43X
© Woodhead Publishing Limited, 2011
textured HTS films are obtained in which the intergranular properties (i.e., grain
boundaries) clearly dominate the superconducting properties like the transport
current density J
c
or microwave surface resistances R
s
. The effect of the grain
boundaries upon the J
c
is demonstrated in Fig. 1.2. On the one hand, J
c
decreases
with increasing misorientation angle. On the other hand, at large angles the grain
boundary imposes a Josephson-type behaviour that can be utilised for the
engineering of Josephson contacts, e.g. by deposition of HTS thin films on
bicrystalline substrates with perfectly defined grain boundaries (Divin et al., 2008).
In conclusion, the extraordinary properties of HTS material – the large
anisotropy, small coherence length and strong dependence of the superconducting
parameters on local modifications – set extremely high requirements for the
deposition of HTS thin films:
• perfect stoichiometry and structure
• perfect biaxial epitaxial orientation
• avoidance of defects that hamper the superconducting properties and
• avoidance of grain boundaries.
1.1.2 Substrate requirements
The choice of the substrate material is of vital importance for the performance of
a superconducting device. For example microwave applications require substrates
with low microwave losses (typically tan
δ
< 10
–5
), whereas coated conductors are
based on metallic tapes. As a consequence, large lattice misfits or even non-
crystalline carriers with complex crystalline buffer systems have to be taken into
account (see Fig. 1.3). Therefore, the impact of the substrate has to be taken into
account for the development of a reliable deposition technology for high-quality
HTS thin-films. The basic requirements for the substrates (including buffer layer)
can be summarised as follows:
• reasonable crystallographic lattice match between HTS film and substrate,
• similar thermal expansion coefficient of HTS and substrate,
• no chemical interaction at the interface between HTS and substrate, and
• surface properties (e.g. polished surface), stability, robustness and size of the
substrate suitable for film deposition and application.
Depending upon the application of the films, specific requirements are imposed
upon the substrate. For a number of applications twinned substrates should be
avoided, or the available size of the substrate and/or its cost are of importance. For
example, for microwave applications the dielectric properties, surface quality and
available size of the substrate are relevant. Generally we can distinguish between
different classes of substrates:
• ‘perfectly compatible’ substrates onto which HTS material can be deposited
without buffer layer,
Deposition technologies, growth and properties of high-T
c
films 7
1
2
3
4
5
6
7
8
9
10
1
2
3
4
5
6
7
8
9
20
1
2
3
4
5
6
7
8
9
30
1
2
3
4
5
6
7
8
9
40
1
2
43X
© Woodhead Publishing Limited, 2011
• ‘less compatible’ substrates that require an optional buffer layer, and
• ‘non-compatible’ substrate materials, which have to be covered with an
epitaxial buffer layer prior to the deposition of the HTS film due to large
lattice mismatch and/or chemical interaction between substrate and HTS
material or due to missing in-plane orientation (e.g. for deposition of biaxial
oriented YBCO on polycrystalline substrates), respectively.
1.3 Substrate material for microwave applications and coated
conductors. (a) For microwave applications substrate material with low
losses (tan
δ
< 10
–5
) are required. Ideal candidates (e.g. MgO, Al
2
O
3
)
suffer from large lattice misfit (a
film
– a
substrate
)/a
film
, where ‘a’ represents
the in-plane lattice parameter. (b) Coated conductors are based on
metallic tapes (e.g. hastelloy, NiCroFer, Ni-alloy). The crystalline
structure is generated in the oxide buffer system (e.g. La
2
Zr
2
O
7
plus
CeO
2
or Nb-doped titanates Sr
1 – x
(Ca,Ba)
x
Ti
1 – y
– Nb
y
O
3
).
8 High-temperature superconductors
1
2
3
4
5
6
7
8
9
10
1
2
3
4
5
6
7
8
9
20
1
2
3
4
5
6
7
8
9
30
1
2
3
4
5
6
7
8
9
40
1
2
43X
© Woodhead Publishing Limited, 2011
Deposition on to compatible substrates is generally easier. Typical candidates
are LaAlO
3
, SrTiO
3
and MgO for YBCO deposition. Nevertheless, buffer layers
can be added, e.g. a CeO
2
buffer layer on LaAlO
3
substrates reduce the probability
of a-axis growth. Important properties of different material that represent
suitable substrate materials as well as buffer or interlayer are summarised in
Table 1.1.
In contrast, the deposition of HTS material onto a number of technically
interesting substrate materials (e.g. Si, Al
2
O
3
, or metal tapes for HTS coated
conductors) requires a previous coating with an adequate buffer layer that enables
epitaxial growth (by reducing the lattice mismatch between substrate and HTS
material or, even providing the crystalline structure for epitaxy in the case of
coated conductors) and provides a sufficient barrier against chemical interdiffusion
between substrate and HTS material. As an example, one of the most interesting
substrate candidates for microwave applications is (1102) oriented Al
2
O
3
(r-cut
sapphire), which possesses high crystalline perfection, mechanical strength, and
low dielectric permittivity (
ε
≈10) and losses (tan
δ
(77K,10GHz) ≈ 10
–7
–10
–8
(Braginsky et al., 1987). However, the complex crystalline structure of sapphire
which is comprised of consecutive planes of oxygen and aluminium hexagons
with each third site vacant, provides two planes of rectangular (m or (1010) plane)
or pseudo-rectangular (r or (1102) plane) surface structure. Both planes possess a
rather poor lattice match to the rectangular basal plane of the c-oriented YBCO
structure. Due to smaller mismatch, r-cut sapphire is preferentially used for the
deposition of YBCO films. The lattice mismatch is accompanied by chemical
interaction between YBCO and Al
2
O
3
taking place at elevated deposition
temperatures (Gao et al., 1992, Davidenko et al., 1992). This yields substantial
diffusion of Al into the HTS film and formation of an uncontrolled BaAl
2
O
4
interfacial layer.
Two different approaches have been considered for the choice of buffer layers.
In one approach, material is chosen which is similar to YBCO with respect to
chemical and structural properties. One of the few promising candidates for such
a buffer is the semiconducting perovskite PrBa
2
Cu
3
O
7
(PBCO) (Gao et al., 1992;
Jia et al., 1990). PBCO layers are also good candidates for YBCO multilayers
(see, e.g. Triscone and Fischer, 1997). The specific resistance can be improved by
partial substitution of Cu by Ga without loss of chemical and structural
compatibility. Although the Al diffusion into the YBCO film is blocked by the
PBCO buffer layer, YBCO films on PBCO/sapphire exhibit higher R
s
and lower
J
c
values than observed for YBCO films of reasonable quality on compatible
substrate materials (Hammond and Matthaei, 1993).
In the second approach, oxides with a cubic structure and lattice parameters
comparable to the diagonal of both r-plane sapphire and (001) plane orthorhombic
YBCO are chosen. Among a large number of candidates MgO, YSZ (ZrO
2
stabilised with ~9 mol % of Y
2
O
3
), and CeO
2
are the most attractive candidates.
Whereas the lattice parameter of MgO is closer to that of sapphire, YSZ and CeO
2
1
2
3
4
5
6
7
8
9
10
1
2
3
4
5
6
7
8
9
20
1
2
3
4
5
6
7
8
9
30
1
2
3
4
5
6
7
8
9
40
1
2
43X
© Woodhead Publishing Limited, 2011
9
Table 1.1 Properties of crystalline substrate materials suitable for the preparation of YBCO thin films (metallic tapes for coated conductors
are not listed)
Attainable dielectric properties Crystallinity
Substrate
materials
ε
tan
δ
at
10GHz, 77K
Thermal
exp. coeff.
(10
–6
/°C)
Melting
temp. (°C)
Available
substrate
size (mm)
Lattice
misfit to
YBCO (%)
Twinning Chemical
stability
Mechanical
strength
Remarks
SrTiO
3
300 ~2 × 10
–2
9.4 2080 30–50 1.4
YSZ 27–33 > 6 × 10
–4
11.4 2550 Ø100 6
NdAIO
3
2070
MgO 9.6–10 6.2 × 10
–6
14 2825 > 30 9 No Good Poor Hygroscopic
α
–Al
2
O
3
(sapphire)
9.4–11.6
anisotropic
10
–8
9.4 2049 Ø100 6–11 (r cut)
18 (m cut)
No Poor Good Requires
buffering
YSZ buffered
r-cut sapphire
27–33 (for
YSZ)
> 6 × 10
–4
(for YSZ)
2550 (for
YSZ)
6 No Good Good
CeO
2
buffered
r-cut sapphire
21.2 (for
CeO
2
)
2600 (for
CeO
2
)
Ø100 0.7 No Good Good
Y
2
O
3
14 2400 Ø30 3 No Expected to
be good
Fair
YAIO
3
16 10
–5
1875 Ø30 3.5 No Fair Fair
LaAIO
3
20.5–27 7.6 × 10
–6
to
3 × 10
–4
10–13 2100 Ø100 2 Yes Good Fair
LaGaO
3
25 9 1715 Ø40 2 Yes Good Fair
NdGaO
3
23 4 × 10
–4
9–11 1670 Ø50 0.04 No Good Fair
PrGaO
3
24 7–8 1680 Ø10 0.3 Yes Good
Note: the lattice misfit is normalised to the YBCO lattice for c-axis orientation (a
film
–a
substrate
)/a
film
, the thermal expansion coefficient of c-axis
oriented YBCO is ~(1–1.3) × 10
–5
/K, YSZ refers to yttrium (~9 mol % of Y
2
O
3
) stabilised ZrO
2
. Other candidates, which are not listed here, are,
e.g., CaNdAlO
4
or XAlO
3
(with X = SrLa, Y, Gd, Eu, Sm, Nd or Pr).
10 High-temperature superconductors
1
2
3
4
5
6
7
8
9
10
1
2
3
4
5
6
7
8
9
20
1
2
3
4
5
6
7
8
9
30
1
2
3
4
5
6
7
8
9
40
1
2
43X
© Woodhead Publishing Limited, 2011
matches that of YBCO. CeO
2
has proven to be one of the most effective buffer
layers due to its favourable thin film growth characteristics, minimal chemical
interaction and good lattice match with most HTS materials. For instance, (001)
CeO
2
buffer layers reduce the lattice mismatch between (1102) sapphire and
(001) YBCO and provide a sufficient barrier against diffusion of Al into the
YBCO film (Zaitsev et al., 1997b). (001) CeO
2
itself has a high lattice mismatch
with respect to (1102) sapphire. This can cause insufficient structural perfection
of the CeO
2
and YBCO layers despite the very high structural perfection of
sapphire substrates. In general, the lattice mismatch between sapphire, buffer and
ceramic film leads to large stress induced strain in the film. As a consequence,
defects will develop and, finally, films that exceed a critical thickness will crack
(Zaitsev et al., 1997a; Hollmann et al., 2009). This will be discussed in section
1.3.2. Nevertheless, YBCO films with high quality with respect to morphology,
critical properties and microwave properties can be grown onto technical substrates
like sapphire as long as the critical thickness is not surpassed (Lahl and
Wördenweber, 2005).
Another example for the need of buffer layers is given by the HTS coated
conductors (see Fig. 1.3b). One of the main obstacles to the manufacture
of commercial lengths of YBCO wire or tapes has been the phenomenon of
weak links, i.e., grain boundaries formed by any type of misalignment of
neighbouring YBCO grains are known to form obstacles to current flow. Therefore
HTS deposition on metallic tapes require carefully aligning the grains, as
low-angle boundaries between superconducting YBCO grains allow more
current to flow. For instance, below a critical in-plane misalignment angle of 4°, the
critical current density approaches that of YBCO films grown on single crystals
(see Fig. 1.2).
The schematics of a typical architecture of a HTS coated conductor is shown in
Fig. 1.3b. Coated conductors are based on metallic tapes (e.g. hastelloy, NiCroFer,
Ni-alloy). The crystalline structure is generated in a complex oxide buffer system
(e.g. La
2
Zr
2
O
7
plus CeO
2
or Nb-doped titanate Sr
1–x
(Ca,Ba)
x
Ti
1–y
– Nb
y
O
3
)
onto which the HTS film is deposited. The HTS is normally protected by an
additional metallic shunt, typically an Au or Ag layer. Several methods have
been developed to obtain biaxially textured buffer layers suitable for high-
performance YBCO films. These are ion-beam-assisted deposition (IBAD), the
rolling-assisted biaxially textured substrate (RABiTS) process, and inclined
substrate deposition (ISD). For more details on these different deposition
technologies for HTS coated conductors one should refer to overview articles like
Paranthaman and Izumi, 2004.
1.2 Deposition techniques
There exist a large variety of techniques for growing HTS films and multilayers
ranging from physical vapour deposition (PVD) to all chemical deposition