U.S. patent application number 12/349735 was filed with the patent office on 2010-02-25 for process for depositing an electrically conductive layer and assembly of the layer on a porous support substrate.
This patent application is currently assigned to PLANSEE SE. Invention is credited to Karl Kailer, Georg Kunschert, Stefan Schlichtherle, Georg Strauss.
Application Number | 20100047565 12/349735 |
Document ID | / |
Family ID | 38456742 |
Filed Date | 2010-02-25 |
United States Patent
Application |
20100047565 |
Kind Code |
A1 |
Kailer; Karl ; et
al. |
February 25, 2010 |
PROCESS FOR DEPOSITING AN ELECTRICALLY CONDUCTIVE LAYER AND
ASSEMBLY OF THE LAYER ON A POROUS SUPPORT SUBSTRATE
Abstract
A process for depositing an electrically conductive, preferably
perovskitic layer uses a pulsed sputtering process. The layer has a
low diffusivity for the elements in the iron group and is
especially suitable for use in solid oxide fuel cells (SOFC). An
assembly of the electrically conductive ceramic layer on a porous
support substrate is also provided.
Inventors: |
Kailer; Karl; (Breitenwang,
AT) ; Kunschert; Georg; (Pflach, AT) ;
Schlichtherle; Stefan; (Ehrwald, AT) ; Strauss;
Georg; (Munster, AT) |
Correspondence
Address: |
LERNER GREENBERG STEMER LLP
P O BOX 2480
HOLLYWOOD
FL
33022-2480
US
|
Assignee: |
PLANSEE SE
Reutte
AT
|
Family ID: |
38456742 |
Appl. No.: |
12/349735 |
Filed: |
January 7, 2009 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
PCT/AT2007/000338 |
Jul 5, 2007 |
|
|
|
12349735 |
|
|
|
|
Current U.S.
Class: |
428/336 ;
204/192.15 |
Current CPC
Class: |
C23C 14/3485 20130101;
C23C 14/0688 20130101; C23C 14/34 20130101; C23C 14/08 20130101;
Y02E 60/50 20130101; H01M 8/10 20130101; Y10T 428/24997 20150401;
C23C 14/088 20130101; H01M 4/9033 20130101; Y10T 428/265
20150115 |
Class at
Publication: |
428/336 ;
204/192.15 |
International
Class: |
B32B 15/04 20060101
B32B015/04; C23C 14/34 20060101 C23C014/34 |
Foreign Application Data
Date |
Code |
Application Number |
Jul 7, 2006 |
AT |
GM 534/2006 |
Claims
1. A process for depositing an electrically conductive ceramic
layer, the process comprising the following steps: producing the
layer by a pulsed sputtering process.
2. The process according to claim 1, wherein the layer has a
perovskitic structure.
3. The process according to claim 1, which further comprises using
an oxide-ceramic sputtering target in the pulsed sputtering
process.
4. The process according to claim 1, which further comprises using
a sputtering target in the pulsed sputtering process, and a
concentration of elements in the sputtering target differs at most
by 5% from a concentration of respective elements in the layer.
5. The process according to claim 1, which further comprises
depositing the layer at a frequency of a pulsed voltage of 1 to
1000 kHz.
6. The process according to claim 1, which further comprises
depositing the layer at a frequency of the pulsed voltage of 10 to
500 kHz.
7. The process according to claim 1, which further comprises
depositing the layer at a frequency of the pulsed voltage of 100 to
350 kHz.
8. The process according to claim 1, which further comprises
depositing the layer with a voltage root-mean-square value of +100
to -1000 V.
9. The process according to claim 1, which further comprises
depositing the layer with a voltage root-mean-square value of +100
to -500 V.
10. The process according to claim 1, which further comprises
depositing the layer with a mean power density of 1 to 30
W/cm.sup.2.
11. The process according to claim 1, which further comprises using
an inert gas as a process gas with a pressure of 1.times.10.sup.-4
to 9.times.10.sup.-2 mbar.
12. The process according to claim 11, which further comprises
using argon as the process gas.
13. The process according to claim 1, wherein the layer has a
structural formula ABO.sub.2, where A includes one or more elements
selected from the group consisting of La, Ba, Sr and Ca; and B
includes one or more elements selected from the group consisting of
Cr, Mg, Al, Mn, Fe, Co, Ni, Cu and Zn.
14. The process according to claim 1, which further comprises
depositing the layer with a thickness of 0.1 to 5 .mu.m.
15. The process according to claim 1, which further comprises
depositing the layer with a density >99% of a theoretical
density.
16. The process according to claim 1, which further comprises
depositing the layer with an impurity content <0.5% by
weight.
17. The process according to claim 16, wherein the impurity content
is <0.1% by weight.
18. The process according to claim 1, which further comprises
depositing the layer on a component used in a solid oxide fuel cell
(SOFC).
19. The process according to claim 18, which further comprises
depositing the layer on a porous substrate.
20. An assembly, comprising: a porous support substrate having a
density of 40 to 70% of a theoretical density and a predominantly
open-pored structure and being formed of sintered grains of an
Fe-based alloy including 15 to 35% by weight Cr, 0.01 to 2% by
weight of one or more elements selected from the group consisting
of Ti, Zr, Hf, Mn, Y, Sc, rare earths, 0 to 10% by weight Mo and/or
Al, 0 to 5% by weight of one or more metals selected from the group
consisting of Ni, W, Nb, Ta, 0.1 to 1% by weight O, a remainder Fe
and impurities; and an electrically conductive ceramic PVD layer
with a thickness of 0.1 to 5 .mu.m deposited on said porous support
substrate.
Description
CROSS-REFERENCE TO RELATED APPLICATION
[0001] This is a continuing application, under 35 U.S.C. .sctn.
120, of copending International Application No. PCT/AT2007/000338,
filed Jul. 5, 2007, which designated the United States; this
application also claims the priority, under 35 U.S.C. .sctn. 119,
of Austrian Patent Application GM 534/2006, filed Jul. 7, 2006; the
prior applications are herewith incorporated by reference in their
entirety.
BACKGROUND OF THE INVENTION
Field of the Invention
[0002] The invention relates to a process for producing or
depositing an electrically conductive layer with a perovskitic
structure. The invention also relates to an assembly of the layer
on a porous support substrate.
[0003] Protective layers with a perovskitic structure are used, for
example, in high-temperature fuel cells (SOFC: Solid Oxide Fuel
Cell). They are operated at temperatures of approximately 650 to
900.degree. C., since it is only at those temperatures that the
thermodynamic conditions for efficient energy production prevail.
In the case of planar SOFC systems, individual electrochemical
cells made up of a cathode, a solid electrolyte and an anode are
stacked to form a stack and connected through the use of metallic
components, so-called interconnectors.
[0004] The interconnector also separates anode and cathode gas
spaces. A dense interconnector is just as important as a dense
electrolyte layer for ensuring a highly effective high-temperature
fuel cell. Suitable materials for interconnectors also have to have
a sufficient conductivity and resistance to the oxidizing
conditions on the air side and reducing conditions on the fuel gas
side. Those requirements are currently best met by lanthanum
chromite, chromium/iron alloys which are doped with yttrium oxide,
and chromium-rich ferrites.
[0005] In order to reduce the evaporation of chromium during use, a
conductive layer with a perovskitic structure is deposited on the
interconnector.
[0006] Depending on the nature and extent of doping, perovskites,
which are currently used for producing cathodes in SOFCs, have the
property of being mixed conductors, i.e. they conduct both
electrons and ions. Perovskites are thermodynamically stable even
at high oxygen partial pressures and, in order to improve the
contact-connection, are also applied to that side of the
interconnector which faces towards the cathode.
[0007] In addition to simply coating the surfaces with contact
pastes including perovskitic ceramic, processes such as APS
(Atmospheric Plasma Spraying), VPS (Vacuum Plasma Spraying), dip
coating or wet powder spraying have been evaluated, or are used on
a semi-industrial scale, for coating interconnectors. Those
spraying processes are also used to produce electrochemically
active cell layers in the high-temperature fuel cell. VPS and APS
are spraying processes which are carried out either under vacuum or
under atmospheric conditions. In those processes, powder is melted
in a plasma jet, and the powder grains immediately solidify in flat
form with a thickness of a few pm when they impact on the substrate
surface. A certain degree of residual porosity cannot be prevented
in that technique. In order to achieve gas-tight structures as far
as possible, thick layers with layer thicknesses of approximately
30 to 50 .mu.m are deposited. Large quantities of ceramic material
are therefore applied to the substrate surfaces, and that leads to
correspondingly high costs. In addition, the electron transmission
is reduced in the case of thick layers. The ever-present porosity
also reduces the electron conductivity.
[0008] CVD processes on the basis of chloridic compounds and
reaction with water vapor at deposition temperatures of 1100 to
1000.degree. C. have also been tested. Layers with thicknesses of
20 to 50 .mu.m have been produced using that process. That process,
too, is costly and the high deposition temperatures are also
disadvantageous.
[0009] In order to deal more effectively with the problems which
have occurred in the past during contact-connection of the
interconnectors, new planar SOFC concepts have also been developed.
Most recently, those include the so-called MSC (Metal Supported
Cell) concept, in which electrochemical cells are applied directly
to porous support substrates. Those porous metallic support
substrates can be used both on the cathode side and on the anode
side.
[0010] MSC concepts on the anode side permit high power densities
to be achieved and constitute an inexpensive alternative.
[0011] Anodes which are formed of NiO--YSZ (nickel oxide-yttrium
stabilized zirconium oxide) and are in direct material contact with
the porous metallic support substrate, are used for the MSC
concept. Since those support substrates are conventionally formed
of an Fe-based alloy with a high Cr content, Ni diffuses from the
anode into the support substrate or Fe and Cr diffuse from the
support substrate into the NiO--YSZ anode when used at temperatures
of 650 to 900.degree. C. The result of that interdiffusion
phenomenon is that a diffusion zone of Fe--Cr--Ni is formed in the
contact area. In comparison with the support substrate, that
diffusion zone has a considerably higher coefficient of thermal
expansion. That can lead to spalling which would result in severe
degradation or total failure. The interdiffusion can be prevented
by perovskitic diffusion barrier layers. To date, that has been
done by depositing porous layers, since otherwise the support
substrate is no longer gas-permeable. Plasma-sprayed diffusion
barrier layers are also applied to the cathode side of
interconnectors. It is disadvantageous in that case that very thick
and therefore expensive layers are used due to the intrinsically
high porosity.
SUMMARY OF THE INVENTION
[0012] It is accordingly an object of the invention to provide a
process for depositing or producing an electrically conductive
layer and an assembly of the layer on a porous support substrate,
which overcome the hereinafore-mentioned disadvantages of the
heretofore-known processes and assemblies of this general type and
which initially permit an inexpensive production of thin, dense and
electrically conductive ceramic layers.
[0013] A further object of the invention is to provide a conductive
ceramic diffusion barrier layer applied to a porous support
substrate, which does not considerably impair the gas permeability
of the support substrate.
[0014] With the foregoing and other objects in view there is
provided, in accordance with the invention, a process for
depositing or producing an electrically conductive ceramic layer.
The process comprises producing the layer by a pulsed sputtering
process.
[0015] In this case, an electrically conductive ceramic layer,
preferably with a perovskitic structure, is deposited through the
use of a pulsed sputtering process. The process according to the
invention makes it possible to uniformly apply thin, dense and
functional ceramic layers to the surface of dense, but also porous,
substrate materials. The mass transfer through the applied layer is
restricted to defect mechanisms. Even at temperatures above
600.degree. C., the diffusivity of, for example, chromium in the
layer according to the invention is very low. This makes it
possible to use very thin layers with a thickness of preferably 0.1
to 5 .mu.m as the diffusion barrier layer. The barrier effect is
insufficient at a thickness of below 0.1 .mu.m. The layer tends to
spall at a thickness of above 5 .mu.m.
[0016] The use of thin layers increases the electron transmission
of the metallic interconnector/layer system.
[0017] If the layer according to the invention is applied to a
porous support substrate, a closed top layer is not formed. The
predominantly open-pored structure of the support substrate is
therefore retained. As a result, the coated support substrate has
good transport and contact-connection properties. The porous
support substrate preferably has a density of 40 to 70% of the
theoretical density and preferably is formed of sintered grains of
an Fe-based alloy including 15 to 35% by weight of Cr, 0.01 to 2%
by weight of one or more elements selected from the group
consisting of Ti, Zr, Hf, Mn, Y, Sc, rare earths, 0 to 10% by
weight of Mo and/or Al, 0 to 5% by weight of one or more metals
selected from the group consisting of Ni, W, Nb, Ta and 0.1 to 1%
by weight of O, a remainder of Fe and impurities.
[0018] PVD processes are not used for coating interconnectors with
ceramic materials since, to date, it has been assumed that the
difficult process management of reactive PVD processes makes them
unsuitable for depositing conductive ceramic layers, in particular
dielectric layers with a perovskitic structure, since it is not
possible to achieve a stoichiometric deposition of the complex
layer material and the required layer properties, such as high
density and good electrical conductivity.
[0019] In the case of sputtering, the layer is made up of
electrically neutral particles which have different energies due to
the process. The coating material is therefore opposite the
substrate as a target. During sputtering, the target is bombarded
with positive ions. A voltage is applied between the substrate
(holder) and the target, and therefore positive ions are
accelerated towards the target where they eject atoms or molecules
which then settle on the substrate as neutral particles,
uninfluenced by the outer field, and form the thin functional
layer. In order to produce the positive ions required for the
sputtering process, use is made of the independent glow discharge
principle. The space between the anode and the cathode is evacuated
and then filled with the process gas. A voltage is applied between
the anode and the cathode. This results in the formation of
substantially three regions: a cathode dark space, a quasi-neutral
transition zone and a positive column. The procedures between the
cathode and the anode can therefore be summarized into the
following processes: electron bombardment ionization of gas atoms,
ion-induced electron emission at the cathode, electron-induced
secondary emission at the anode and ion bombardment sputtering.
[0020] The use of pulsed sputtering plasmas makes it possible to
use significantly higher target currents and arc currents. The
higher current intensities mean that considerably higher coating
rates can be achieved.
[0021] The use of oxide-ceramic sputtering targets makes it
possible to carry out the process in non-reactive fashion, and this
achieves high process stability and reduces the technical
complexity. The content of multiply charged particles and the
kinetic energy of the particles can be increased due to the
relatively high degree of plasma excitation, and this makes it
possible to provide a coating without a substrate bias. This
results in improved layer properties such as, for example, a higher
layer density, improved adhesion, higher electrical conductivity
and improved chemical resistance.
[0022] The concentration of the elements in the sputtering target
differs at most by 5% from the concentration of the respective
element in the layer. The process according to the invention
therefore permits ceramic, preferably perovskitic layers to be
stoichiometrically deposited, even on unheated substrates. The
layer surfaces produced are smooth and chemically stable.
[0023] In this case, the layer is deposited at a frequency of the
pulsed voltage of 1 to 1000 kHz. At below 1 kHz or during DC
operation, no stable coating processes can be carried out in the
case of the deposition of dielectric materials. Instead, electrical
flashovers and arcing occur.
[0024] At above 1000 kHz, the technical outlay for the voltage
supply and the number of adapting units required for controlling
the process impedance become too great for an economic coating. The
process becomes particularly economical when a frequency of 10 to
500 kHz, preferably 100 to 350 kHz, is chosen. In this case, the
voltage root-mean-square value is +100 to -1000 V, preferably +100
to -500 V. The voltage root-mean-square value of an AC voltage is
understood to mean the DC voltage value which corresponds to the
same heat effect. The ratio of peak amplitude to root-mean-square
value is referred to as the crest factor and is, for example, 1.41
for sinusoidals. At values below the stated limit value, the
particle energies are too low for the desired sputtering process.
At values above the stated limit value, the high particle energies
can result in undesirable effects such as, for example, sputtering
away of the evaporating layer, electrical flashovers due to high
field strengths, implantation in the substrate, undesirable
increase in temperature, etc., which prevent layers with a
perovskitic structure from being deposited.
[0025] It has proven to be advantageous to deposit the layer with a
mean power density of 1 to 30 W/cm.sup.2. At below 1 W/cm.sup.2,
the coating rate is too slow and the coating duration is therefore
too long for industrial implementation. At above 30 W/cm.sup.2, the
input of energy at the target is too great, and this results in the
thermal destruction of the perovskitic target material.
[0026] The process gas used is an inert gas, preferably argon, with
a pressure of 1.times.10.sup.-4 to 9.times.10.sup.-2 mbar. At below
1.times.10.sup.-4 mbar, the sputtering process cannot be ignited.
At above 9.times.10.sup.-2 mbar, the free path lengths of the
sputtered layer particles become too small as a result of too many
impact processes. This reduces the kinetic energy of the sputtered
layer particles, which means that the desirable layer properties
cannot be achieved.
[0027] The ceramic layer preferably has a perovskitic structure
with the structural formula ABO.sub.3. The crystal structure in
this case is cubic, orthorhombic or tetragonal. A includes one or
more elements selected from the group consisting of La, Ba, Sr, Ca.
B includes one or more elements selected from the group consisting
of Cr, Mg, Al, Mn, Fe, Co, Ni, Cu and Zn. The layers preferably
have a density >99% of the theoretical density and an impurity
content <0.5% by weight, preferably <0.1% by weight.
[0028] Other features which are considered as characteristic for
the invention are set forth in the appended claims.
[0029] Although the invention is illustrated and described herein
as embodied in a process for depositing an electrically conductive
layer and an assembly of the layer on a porous support substrate,
it is nevertheless not intended to be limited to the details shown,
since various modifications and structural changes may be made
therein without departing from the spirit of the invention and
within the scope and range of equivalents of the claims.
[0030] The construction and method of operation of the invention,
however, together with additional objects and advantages thereof
will be best understood from the following description of specific
embodiments when read in connection with the accompanying
drawings.
BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWING
[0031] FIG. 1 is an illustration diagrammatically showing a
configuration of a perovskitic layer on a porous substrate; and
[0032] FIG. 2 is a group of photographs showing EPMA measurements
of a porous substrate with an LSC layer and an Ni layer deposited
thereon, after ageing for 1000 h at 850.degree. C.
DETAILED DESCRIPTION OF THE INVENTION
[0033] Reference will now be made in detail to the figures of the
drawings, with which examples are described.
EXAMPLE 1
[0034] A porous support substrate having a composition of 26% by
weight Cr, 0.5% by weight Y.sub.2O.sub.3, 2% by weight Mo, 0.3% by
weight Ti and 0.03% by weight Al and a remainder of Fe was coated
through the use of a pulsed, non-reactive DC process. This was done
using an Edwards sputter coater fitted with an LSM target
(La.sub.0.8Sr.sub.0.2Mn oxide) with a diameter of 72 mm. A
sputtering power of 400 W, a voltage of 149 V, a current of 2.01 A,
a frequency of 350 kHz (with a pulse duration of 1.1 .mu.s) and a
process pressure of 5*10.sup.-3 mbar were also set. This produced
LSM layers being 3 .mu.m thick and having the composition
La.sub.0.8Sr.sub.0.2Mn oxide (LSM).
[0035] FIG. 1 diagrammatically shows the configuration of the
deposited layers on the porous support substrate.
EXAMPLE 2
[0036] Porous and dense support substrates having a composition of
26% by weight Cr, 0.5% by weight Y.sub.2O.sub.3, 2% by weight Mo,
0.3% by weight Ti and 0.03% by weight Al and a remainder of Fe were
coated through the use of a pulsed, non-reactive DC process. This
was done using an Edwards sputter coater fitted with an LSC target
(La.sub.0.8Sr.sub.0.2Cr oxide) with a diameter of 72 mm. A
sputtering power of 400 W, a voltage of 149 V, a current of 2.01 A,
a frequency of 350 kHz (with a pulse duration of 1.1 .mu.s) and a
process pressure of 5*10.sup.-3 mbar were also set. This produced
LSC layers being 3 .mu.m thick and having the composition
La.sub.0.8Sr.sub.0.2CrO.sub.3.
[0037] The LSC layers were then coated with an APS nickel layer
being 50 .mu.m thick. This structure of an iron/chromium alloy
(porous and non-porous)--LSC layer (3 .mu.m)--APS nickel layer (50
.mu.m) was used to investigate the diffusion barrier effect of the
thin LSC layer with respect to nickel into iron or iron into
nickel. In this case, the structure was aged in air for 100 h at
850.degree. C. to 1000.degree. C. The diffusion properties were
documented using EPMA measurements, as seen in FIG. 2. The LSC
layer prevents diffusion of nickel into iron or iron into nickel
under the stated test conditions. The LSC layers deposited have a
high electrical conductivity (corresponding to the target used), a
high density >99.9%, a homogeneous layer structure and a smooth
surface with a mean roughness value which is the same as the mean
roughness value of the substrate. As a result of the process, no
foreign atom inclusions can be measured through the use of EPMA and
EDX.
* * * * *