U.S. patent application number 11/898065 was filed with the patent office on 2009-03-12 for processing of powders of a refractory metal based alloy for high densification.
This patent application is currently assigned to BLOOM ENERGY CORPORATION. Invention is credited to Sudhakara Sarma Sreedhara, Ranganathan Sundaresan.
Application Number | 20090068055 11/898065 |
Document ID | / |
Family ID | 40432052 |
Filed Date | 2009-03-12 |
United States Patent
Application |
20090068055 |
Kind Code |
A1 |
Sreedhara; Sudhakara Sarma ;
et al. |
March 12, 2009 |
Processing of powders of a refractory metal based alloy for high
densification
Abstract
A powder metallurgy method of making a chromium base alloy
includes blending a first powder comprising a chromium powder and a
second powder comprising at least one of titanium, titanium
hydride, zirconium or zirconium hydride, annealing the first powder
and the second powder in a reducing atmosphere after the step of
mixing, compacting a blend of the first and the second powders, and
sintering the compacted blend to form a chromium base alloy. The
chromium alloy may be used as an interconnect for a solid oxide
fuel cell, and includes least one of iron or nickel greater than
zero and equal to or less than 7 weight percent, yttria greater
than zero and equal to or less than 2 weight percent, at least one
of titanium or zirconium greater than zero and equal to or less
than 1 weight percent and at least 90 weight percent chromium.
Inventors: |
Sreedhara; Sudhakara Sarma;
(Hyderabad, IN) ; Sundaresan; Ranganathan;
(Hyderabad, IN) |
Correspondence
Address: |
FOLEY AND LARDNER LLP;SUITE 500
3000 K STREET NW
WASHINGTON
DC
20007
US
|
Assignee: |
BLOOM ENERGY CORPORATION
|
Family ID: |
40432052 |
Appl. No.: |
11/898065 |
Filed: |
September 7, 2007 |
Current U.S.
Class: |
420/428 ; 419/31;
419/32 |
Current CPC
Class: |
C22C 32/0089 20130101;
B22F 2998/00 20130101; B22F 3/10 20130101; B22F 1/0085 20130101;
B22F 3/02 20130101; B22F 1/0003 20130101; B22F 3/02 20130101; B22F
1/0014 20130101; B22F 3/1017 20130101; C22C 1/045 20130101; B22F
2998/00 20130101; B22F 1/0003 20130101; C22C 32/0026 20130101; C22C
27/06 20130101; B22F 2998/10 20130101; B22F 2998/10 20130101; B22F
2998/10 20130101 |
Class at
Publication: |
420/428 ; 419/31;
419/32 |
International
Class: |
C22C 27/06 20060101
C22C027/06; B22F 1/00 20060101 B22F001/00 |
Claims
1. A powder metallurgy method of making a chromium base alloy,
comprising: blending a first powder comprising a chromium powder
and a second powder comprising at least one of titanium, titanium
hydride, zirconium or zirconium hydride; annealing the first powder
and the second powder in a reducing atmosphere after the step of
mixing; compacting a blend of the first and the second powders; and
sintering the compacted blend to form a chromium base alloy.
2. The method of claim 1, wherein the step of annealing is
conducted before the step of compacting.
3. The method of claim 1, wherein the step of annealing is
conducted after the step of compacting.
4. The method of claim 1, wherein the step of annealing is
conducted before and after the step of compacting.
5. The method of claim 1, wherein the step of annealing is
conducted in a temperature range of about 800 to about 1200.degree.
C. in a hydrogen atmosphere.
6. The method of claim 1, wherein the first powder comprises
chromium powder having a bimodal distribution comprising a blend of
coarse and fine chromium powders.
7. The method of claim 1, wherein the blend contains 1 weight
percent or less of the second powder.
8. The method of claim 1, further comprising blending the first
powder and the second powder with at least one of iron powder,
nickel powder and yttria powder to form the blend.
9. The method of claim 8, wherein the annealing is conducted after
the step of blending the first powder and the second powder and
before the step of blending the first powder and the second powder
with at least one of iron powder, nickel powder and yttria
powder.
10. The method of claim 8, wherein the yttria powder has an average
particle diameter of less than one micron.
11. The method of claim 8, wherein the blend comprises at least one
of iron and nickel powder greater than zero and equal to or less
than 7 weight percent, yttria powder greater than zero and equal to
or less than 2 weight percent, at least one of titanium, titanium
hydride, zirconium or zirconium hydride powder greater than zero
and equal to or less than 1 weight percent and at least 90 weight
percent chromium powder.
12. The method of claim 1, wherein: the step of compacting
comprises cold compacting at least the first powder and the second
powder; and the step of sintering is conducted in an atmosphere
containing hydrogen at a temperature of about 1300 to about
1500.degree. C.
13. The method of claim 1, wherein the sintered alloy comprises an
interconnect for a solid oxide fuel cell, such that the
interconnect has a coefficient of thermal expansion from 30.degree.
C. to 1000.degree. C. of between about 11.times.10.sup.-6/.degree.
C. and about 13.times.10.sup.-6/.degree. C.
14. The method of claim 13, further comprising providing the
interconnect into a solid oxide fuel cell stack.
15. A powder metallurgy method of making a chromium base alloy,
comprising: blending a first powder comprising a chromium powder
and a second powder comprising yttria powder having an average
particle size of less than 1 micron; compacting a blend of the
first and the second powders; and sintering the compacted blend to
form a chromium base alloy.
16. The method of claim 15, further comprising: blending the first
powder and a third powder comprising at least one of titanium,
titanium hydride, zirconium or zirconium hydride before, after or
during the step of blending the first powder and the second powder;
and annealing at least the first powder and the third powder in a
reducing atmosphere after the step of blending the first powder and
the third powder.
17. A chromium alloy interconnect for a solid oxide fuel cell
comprising least one of iron or nickel greater than zero and equal
to or less than 7 weight percent, yttria greater than zero and
equal to or less than 2 weight percent, at least one of titanium or
zirconium greater than zero and equal to or less than 1 weight
percent and at least 90 weight percent chromium.
18. The alloy interconnect of claim 17, wherein the alloy comprises
2 to 5 weight percent iron and 0.25 to 2 weight percent nickel.
19. The alloy interconnect of claim 17, wherein the alloy comprises
3 to 7 weight percent of either iron or nickel.
20. The alloy interconnect of claim 17, wherein the interconnect is
located in a solid oxide fuel cell stack and the interconnect
comprises gas flow channels.
Description
FIELD OF INVENTION
[0001] The invention relates to a chromium alloy in general and to
an alloy for use as an interconnect in high temperature fuel cell
systems, such as solid oxide fuel cell (SOFC) systems and methods
of making thereof.
BACKGROUND OF THE INVENTION
[0002] One of the major constraints to producing cost effective
SOFC systems or stacks is the cost of functional interconnects. In
planar SOFC stacks, a planar or plate shaped interconnect is
located between adjacent SOFCs. The interconnect provides reactant
gas separation and containment, mechanical support to the cells,
and a low resistance path for electrical current between adjacent
SOFCs. Moreover, the reactant gas flow channels on either side of
the interconnect are designed to ensure distribution of the fuel
and the oxidant with minimal pressure drop in the overall SOFC
stack, especially in respect to the air flow channels of the
interconnect because of the relatively high air flow rates employed
to dispose of heat from the stack. In addition, each interconnect
within the stack should be resistant to deleterious reactions (such
as corrosion), free from interconnected porosity to ensure
separation of the reactant fuel gases on the one side and oxidant
on the other, and should possess a coefficient of thermal expansion
(CTE) compatible with those of the SOFC electrode and electrolyte
materials in order to minimize the effects of displacement caused
by differential thermal expansion.
[0003] Due to the high temperatures of operation of the SOFC
system, initially ceramic materials such as lanthanum chromite had
been used as interconnects. While these materials had provided some
amount of success, the cost had been prohibitive, and the
reliability in the various aspects of their function had been less
than adequate. While electronic conductivity in the interconnect
must be high for its function, most of the ceramic materials
employed possess high ionic conductivity. They are also prone to
chemistry changes during their life cycle due to loss of oxygen
ions in the reducing atmosphere of fuel, such as hydrogen. Many of
the alternatives considered among the ceramic materials also
possess unacceptable values of CTE.
[0004] Metallic materials, particularly wrought alloys, such as
nickel base superalloys and high temperature stainless steels,
considered for interconnects in SOFC systems can provide pore-free
structure. However, many of these fail on account of inadequate
strength at the temperatures of operation of the SOFC, typically
700 to 1000.degree. C., or by growth of an oxide layer that
constrains electrical conductivity. Chromium based alloys have been
used as interconnects with success since the oxide layer that grows
on the chromium alloys at the SOFC operating temperatures is
actually conducting. However, there are metallurgical limitations
to producing such alloys by any process involving casting and
subsequent metal working. These alloys are more amenable to
production by powder metallurgy (PM) processing. Conventional PM
processing comprising of consolidation by compaction of the powder,
typically in a die system, followed by a high temperature treatment
of sintering generates porous materials. In order to remove pores,
these alloys may be manufactured by a process of hot consolidation,
such as hot pressing, hot isostatic pressing of canned powder, hot
rolling of canned powder, etc. All such hot consolidation
processes, however, lead to slow production rates and result in
excessively high costs.
[0005] Chromium based alloys provide advantages in their
application as interconnects because it is possible, by judicious
alloying, to realize a CTE in these alloys compatible with the
other components of SOFC. The oxide layer that forms under high
temperature oxidation conditions prevalent in the SOFC can be self
limiting, and further the oxide is electrically conducting. In view
of these factors, chromium based alloys are among the most
attractive systems for use in SOFC.
[0006] In view of the favorable cost consideration, it is desirable
to incorporate a processing sequence for the material that is less
complex and less expensive than the hot consolidation processes for
such a chromium alloy. However, the simple, comparatively
inexpensive PM processing of cold compaction in a die followed by
high temperature sintering has the limitation of leading to a
material that is porous in nature and therefore generally
unacceptable for use as an interconnect.
SUMMARY
[0007] One embodiment of the invention provides a powder metallurgy
method of making a chromium base alloy, comprising blending a first
powder comprising a chromium powder and a second powder comprising
at least one of titanium, titanium hydride, zirconium or zirconium
hydride, annealing the first powder and the second powder in a
reducing atmosphere after the step of mixing, compacting a blend of
the first and the second powders, and sintering the compacted blend
to form a chromium base alloy.
[0008] Another embodiment of the invention provides a chromium
alloy interconnect for a solid oxide fuel cell, comprising least
one of iron or nickel greater than zero and equal to or less than 7
weight percent, yttria greater than zero and equal to or less than
2 weight percent, at least one of titanium or zirconium greater
than zero and equal to or less than 1 weight percent and at least
90 weight percent chromium.
BRIEF DESCRIPTION OF THE DRAWINGS
[0009] FIG. 1 is a micrograph of the morphology of a ready-to-press
powder blend of the alloy Cr--Fe--Ni--Ti-nanoY.sub.2O.sub.3 made
according to the embodiments of the invention. The bimodal
distribution of the chromium powder is visible in the figure.
[0010] FIG. 2 is a plot of percent green density versus temperature
which shows the effect of annealing the powder with the addition of
cobalt or titanium on its compressibility.
[0011] FIG. 3 is a micrograph which shows the nano yttria powder
used in the powder blend made according to the embodiments of the
invention.
[0012] FIG. 4 shows a particle of base chromium powder with other
finer particles embedded in it (identified as nickel,
Y.sub.2O.sub.3, and iron) made according to the embodiments of the
invention.
[0013] FIG. 5(a) shows a scanning electron micrograph of blended
powder containing nano Y.sub.2O.sub.3 while FIG. 5(b) shows an EDS
pattern for Y in the scan showing uniform distribution of
Y.sub.2O.sub.3.
[0014] FIG. 6 is a plot of mass change percent versus temperature
which illustrates a comparison of the oxidation resistance of four
alloys measured by thermogravimetry.
[0015] FIG. 7 shows the effect of lubricant addition on the
compressibility of the powder made according to the embodiments of
the invention.
[0016] FIG. 8 shows the dilatogram of the sintered alloy made
according to the embodiments of the invention indicating the CTE at
different stages.
DETAILED DESCRIPTION OF THE EMBODIMENTS
[0017] One or more embodiments of the invention provide
ready-to-press powders of a chromium base alloy that can result in
high compacted density, thus enabling high sintered density with
minimum interconnected porosity in the sintered component.
Additional embodiments of the invention provide a high oxidation
resistant chromium based alloy such that the alloy may be applied
in the high temperature environment in SOFC. Preferably, the
chromium base alloy of the embodiments of the invention has high
compacted density and high oxidation resistance.
[0018] The high compacted density in the powder blend may be
achieved by selecting a desired particle size distribution in the
base chromium powder by blending of different powder size
fractions. Furthermore, the hardness of the base chromium powder is
reduced by including an element or compound, such as titanium,
zirconium and/or their hydride that has a two-way reaction with
hydrogen at elevated temperatures in the powder blend followed by a
thermochemical treatment by annealing in a hydrogen atmosphere.
[0019] In another embodiment, the alloy's oxidation resistance is
improved by incorporation of yttrium oxide (also referred to as
yttria) in the alloy. More specifically, the oxidation resistance
is enhanced by the inclusion of nano sized yttria powder rather
than yttria powders in the micron size range.
[0020] The alloy can be used as an interconnect in a SOFC system.
Alloying elements, such as iron and/or nickel may be added to the
alloy to provide a CTE which is compatible with that of the
zirconia, such as yttria or scandia stabilized zirconia or the SOFC
electrolyte.
[0021] The chromium and alloying element powder blend is amenable
to cold compaction in a die arrangement commonly employed in powder
metallurgy processing. Such compaction made with the application of
a compaction pressure in any suitable range, such as less than 800
MPa, such as about 700-750 MPa, leads to a green density of the
compact of over 90% theoretical. Green compacts made from the
powder blend may be sintered in the temperature range
1300-1500.degree. C. in an atmosphere containing dry hydrogen to
result in a sintered body that can be used as an interconnect or in
other suitable applications.
[0022] In a first embodiment, the starting chromium powder has a
bimodal distribution and comprises a blend of coarse and fine
chromium powders (i.e., a blend of two chromium powders, one having
a larger average size or diameter than the other one). This bimodal
powder particle size distribution in a powder mass improves
densification in cold compaction. In one example of the first
embodiment, two batches of chromium powder from two different
sources with different particle size characteristics as given in
Table 1 were mixed in different proportions and subjected to
compaction. The coarse powder is labeled "W" and the fine powder is
labeled "D" in Table 1. Furthermore, Table 1 provides the general
size ranges for each mesh size as well as the actual values used in
the example.
TABLE-US-00001 TABLE 1 Chromium powder size distribution
characteristics Fraction (mesh size) Coarse Powder "W" Fine Powder
"D" -100 + 150 0-5%, such as 3.7% -- -150 + 200 35-50%, such as
41.9% -- -200 + 325 40-50%, such as 44.5% 0-10%, such as 4.2% -325
0-10%, such as 9.2% 90-100%, such as 95.8%
[0023] An idealized size distribution in powders for maximum
density based on packing of particles into the voids between larger
particles was considered. Accordingly a proportion of approximately
a third each (30-35%) of particles in the size ranges 74-104
microns, 43-74 microns and <43 microns was realized by blending
the two powders W and D in selected proportion. As shown in FIG. 1,
the powder blend has an optimum distribution of the chromium
powder. The preferred distribution has 0-5% of chromium powder in
the size range 104-149 microns, 30-35% in the range 74-104 microns,
25-35% in the range 43-74 microns, and 30-40% of size less than 43
microns.
[0024] The effect on green density of the compact made with under a
load of 600 MPa when the two powders were blended in different
proportions is given in Table 2. The first column in Table 2 lists
the sample number, the second column lists the sample composition,
with the powder blend components in weight percent, and the last
column lists the green density as percent of theoretical.
TABLE-US-00002 TABLE 2 Effect of blending chromium powders in the
powder blend on the compact density Sample Green Density- Number
Composition % theoretical S90 (65W + 35D) Cr blend + 5Fe +
1Y.sub.2O.sub.3 + Poor compact 0.25 Lubricant integrity S91 (85W +
15D) Cr blend + 4Fe + 0.5Ni + 90.76 0.5Ti + 1Y.sub.2O.sub.3 + 0.25
Lubricant S92 (85W + 15D) Cr blend + 5Ni + 1Y.sub.2O.sub.3 + 90.86
0.25 Lubricant S93 (75W + 25D) Cr blend + 5Fe + 1Y.sub.2O.sub.3 +
91.2 0.25 Lubricant S94 (75W + 25D) Cr blend + 4Fe + 0.5Ti + 91.73
0.5Ni + 1Y.sub.2O.sub.3 + 0.25 Lubricant
[0025] As can be seen in Table 2, the two chromium powders, W and
D, were blended with powders selected from Fe, Ti, Ni and yttria
powders and a lubricant. Table 3 shows the size characteristics of
the iron, nickel and yttria powders used in the blends shown in
Table 2. The iron powder is classified by mesh size, while the
nickel and yttria powders are classified by average particle size
as measured by FSSS/XRD.
TABLE-US-00003 TABLE 3 Size characteristics of iron, nickel and
yttria powders Fraction (mesh size) Iron Nickel Yttria -80 + 100 5
max -- -- -100 + 150 5-10 -- -- -150 + 200 30-50 -- -- -200 + 325
10-25 -- -- -325 15-30 -- -- Average particle -- 2.5-3.5 .mu.m
15-25 nm size, FSSS/XRD
[0026] In a second embodiment of the invention, the effect of
annealing of the chromium powder on the hardness, and hence on the
compact density after green compaction was studied. The annealing
was carried out at 1000.degree. C. under hydrogen. It was observed
that pure chromium powder did not respond to the annealing and no
reduction in its hardness was observed. Experiments were carried
out on blends of chromium powder with 0.5% addition of powders of
cobalt, titanium metal and titanium hydride. Table 4 shows the
characteristics of the cobalt, titanium and titanium hydride
powders used in the blends. The chromium powder included the
bimodal blend of powders W and D from the first embodiment.
TABLE-US-00004 TABLE 4 Size characteristics of powders blended with
chromium prior to annealing Fraction (mesh size) Titanium Titanium
hydride Cobalt +80 -- 0-10 -80 + 100 -- 5-15 -100 + 140 -- 15-20
-140 + 200 35-45 15-20 -- -200 + 325 40-50 15-20 -- -325 10-20
20-30 -- Average particle -- -- 1.5-2.5 .mu.m size, FSSS
[0027] Table 5 shows the effect of annealing of the blend of
chromium powder with these additional powder. It can be seen from
Table 5 that blending the chromium powder with the cobalt powder
and subjecting the blend to annealing at 1000.degree. C. in
hydrogen did not lead to any significant lowering of the hardness
of the blend. However, the blends of chromium powder with titanium
and titanium hydride showed a decrease in the hardness of the blend
after annealing in hydrogen at 1000.degree. C.
TABLE-US-00005 TABLE 5 Effect of annealing on hardness of chromium
powder Hardness on particle, HV Cr + 0.5Co Cr + 0.5Ti Cr +
0.5TiH.sub.2 Sl. No. Blended Annealed Blended Annealed Blended
Annealed 1 214 217 225 260 287 261 2 215 213 225 239 287 260 3 208
218 225 256 282 250 4 195 203 428 241 274 243 5 192 212 265 248 289
221 6 213 222 239 220 289 248 7 209 203 255 228 289 390 8 204 199
258 241 313 217 9 197 186 238 236 404 258 10 200 201 254 200 241
238 11 202 210 221 194 199 366 12 182 201 263 197 282 250 Average
203 206 258 230 284 267
[0028] Without wishing to be bound by a particular theory, the
reduction in hardness when chromium powder annealed in the presence
of Ti or TiH.sub.2 may be attributed to a lowering of the oxygen
content of the chromium powder during the hydrogen anneal.
Annealing under hydrogen both pure chromium and chromium blended
with cobalt powder did not show a reduction in the hardness, since
the oxide layer on chromium is not susceptible to reduction under
gaseous hydrogen. This is to be expected on the basis of the
extremely high stability of chromium oxide. However, in the
presence of Ti or TiH.sub.2, the condition of hydrogen can be
different, since titanium can absorb hydrogen to form a hydride
which can also release hydrogen at the annealing temperature, and
titanium hydride can dissociate and subsequently continue an
absorption-dissociation cycle. Such hydrogen release from cyclic
formation of hydride and dissociation would result in the
availability of nascent hydrogen in the vicinity of chromium. Such
hydrogen as dissociated would be extremely pure with little content
of moisture, providing an adequately low dew point (i.e. high
H.sub.2/H.sub.2O ratio) which enables some reduction of chromium
oxide, and thereby a lowering of the hardness. Similar effect of
nascent hydrogen can also be expected from the addition of Zr
and/or ZrH.sub.2.
[0029] Thus, a powder metallurgy method of making a chromium base
alloy of the second embodiment includes blending a first powder
comprising a chromium powder and a second powder comprising at
least one of titanium, titanium hydride, zirconium or zirconium
hydride, annealing the first powder and the second powder in a
reducing atmosphere after the step of mixing, compacting a blend of
the first and the second powders, and sintering the compacted blend
to form a chromium base alloy.
[0030] The annealing step may be conducted before and/or after the
step of compacting. Preferably, the powder blend is annealed in
hydrogen and then compacted. Alternatively, the powder blend may be
compacted and then annealed in hydrogen, or the blend may be
annealed in hydrogen before and after compacting.
[0031] The annealing may be conducted in a temperature range of
about 800 to about 1200.degree. C., such as about 1000.degree. C.
However, other suitable temperatures may be used. Any suitable
hydrogen atmosphere may be used, such as pure hydrogen, forming gas
(a mixture of hydrogen and nitrogen), a mixture of hydrogen and a
noble gas, such as argon, etc.
[0032] Preferably, the first powder is a chromium powder having a
bimodal distribution comprising a blend of coarse and fine chromium
powders, such as for examples powders D and W described in the
first embodiment. However, a single mode distribution chromium
powder may also be used.
[0033] The step of compacting preferably comprises cold compacting
at least the first powder and the second powder. However, other
compacting methods may also be used. If desired, a second hydrogen
annealing step (which can be referred to as a presintering step) at
a temperature of about 800 to about 1200.degree. C., such as about
1000.degree. C., may be conducted after the step of compacting. An
optional calibration or sizing step, such as a pressing step, may
be added before and/or after the sintering step. The step of
sintering is also preferably conducted in an atmosphere containing
hydrogen at a temperature of about 1300 to about 1500.degree. C.
However, other sintering temperatures may also be used.
[0034] While titanium and titanium hydride are described in the
examples illustrated in Table 5, other chromium alloying elements
which have a two way reaction with chromium, such as zirconium or
zirconium hydride, may also be used in addition to or instead of
titanium or its hydride. Preferably, the blend contains 1 weight
percent or less, such as about 0.5 weight percent of the second
powder, such as Ti, Zr or their hydride.
[0035] If desired, the chromium powder and the Ti, Zr or their
hydride powder may also be blended with at least one of iron
powder, nickel powder and/or yttria powder to form the blend. The
step of annealing in hydrogen may be conducted after the step of
blending the Cr with the Ti, Zr or their hydride powder to form a
first blend, and before the step of blending the first blend the
with at least one of iron powder, nickel powder and yttria powder.
In other words, the chromium and Ti, Zr or their hydride blend may
be annealed in hydrogen prior to adding the iron powder, nickel
powder and/or yttria powder to the blend. Alternatively, the iron
powder, nickel powder and/or yttria powder may be added to the
blend before the step of annealing in hydrogen.
[0036] The powder blend preferably includes at least one of iron
and nickel powder greater than zero and equal to or less than 7
weight percent, yttria powder greater than zero and equal to or
less than 2 weight percent, at least one of titanium, titanium
hydride, zirconium or zirconium hydride powder greater than zero
and equal to or less than 1 weight percent and at least 90 weight
percent chromium powder.
[0037] For example, the powder blend may include both iron and
nickel powder, with the iron powder comprising 2 to 5 weight
percent and the nickel powder comprising 0.25 to 2 weight percent,
such as for example 4 to 5 weight percent iron and 0.5 to 1 weight
percent nickel. Alternatively, the powder blend may contain 3 to 7
weight percent, such as 4 to 5 weight percent of either iron or
nickel powder. The powder blend may contain between 0.25 and 2
weight percent, such as between 0.5 and 1.5 weight percent, for
example 1 weight percent yttria. Preferably, the yttria powder has
an average particle diameter of less than 1 micron. The powder
blend may contain between 0.25 and 1 weight percent, such as 0.5 to
0.75 weight percent of at least one of titanium, titanium hydride,
zirconium or zirconium hydride powder. The blend preferably
contains between 90 and 96 weight percent, such as between 93 and
95 weigh percent, for example 94 weight percent chromium powder.
The powder blend may also contain other additives, such as a
lubricant (0.1 to 1 weight percent, such as about 0.25 to 0.5
weight percent). The lubricant is removed during subsequent
processing and is not included in the sintered alloy
composition.
[0038] Even if titanium hydride or zirconium hydride is used in the
powder blend, due to the subsequent hydrogen anneal, the final
sintered chromium alloy should have trace or no hydrogen content.
Thus, the sintered chromium alloy preferably contains at least one
of iron or nickel greater than zero and equal to or less than 7
weight percent, yttria greater than zero and equal to or less than
2 weight percent, at least one of titanium or zirconium greater
than zero and equal to or less than 1 weight percent and at least
90 weight percent chromium.
[0039] For example, the sintered alloy may include both iron and
nickel, with the iron comprising 2 to 5 weight percent and the
nickel comprising 0.25 to 2 weight percent, such as for example 4
to 5 weight percent iron and 0.5 to 1 weight percent nickel.
Alternatively, the alloy may contain 3 to 7 weight percent, such as
4 to 5 weight percent of either iron or nickel. The alloy may
contain between 0.25 and 2 weight percent, such as between 0.5 and
1.5 weight percent, for example 1 weight percent yttria. The alloy
may contain between 0.25 and 1 weight percent, such as 0.5 to 0.75
weight percent of at least one of titanium or zirconium. The alloy
preferably contains between 90 and 96 weight percent, such as
between 93 and 95 weigh percent, for example 94 weight percent
chromium. The shaped alloy interconnect includes gas flow channels
or grooves and may have a chromium oxide layer or coating on its
surface. The interconnect may be used in a solid oxide fuel cell
stack.
[0040] FIG. 2 shows the effect of annealing temperature and
compacted blend composition on the density of the compact made from
the powder blend. Specifically, FIG. 2 shows the compacted density
(green density as percent of theoretical) after the powder was
subjected to different annealing treatments. Chromium powder,
without any additives, did not reflect any improvement in compact
density after any annealing temperature in the range
700-1000.degree. C. For example, FIG. 2 shows that a Cr powder
compact without Ti or Zr annealed at 800.degree. C. and
1000.degree. C. exhibited a relatively low green density. Likewise,
a Cr and Ti compact annealed at only 650.degree. C. also exhibited
a relatively low green density. In contrast, significant
improvement in compact density was obtained by annealing the Cr and
Ti compact at above 800.degree. C., such as at 1000.degree. C.
Annealing a compact of Cr and Co at 1000.degree. C. also resulted
in an improved green density which was about 1 percent lower than
that of the Cr and Ti compact. Thus, annealing at 800.degree. C.
and above is preferred.
[0041] In a third embodiment of the invention, the chromium powder
is blended with nanosized yttria powder. Nanosized yttria includes
yttria powder having an average particle size or diameter of less
than 1 micron, such as 5 to 100 nm, for example 10 to 30 nm. FIG. 3
shows a scanning electron micrograph of yttria of size in the
nanometer size range. The average particle/crystallite size was
verified as about 23 nanometer by measurement of the FWHM (full
width at half maximum) width of x-ray diffracted peaks following
standard practice. The nano yttria powder was blended with
chromium-titanium powder annealed as described above along with
alloying additions of iron and nickel. The primary benefit from
nano yttria powder can be seen in the form of uniformity of blend.
FIG. 4 shows chromium powder with other finer particles embedded in
it uniformly. The embedded particles were identified as nickel,
Y.sub.2O.sub.3, and iron by EDAX on the selected spots following
standard practice. The uniformity of Y.sub.2O.sub.3 distribution in
the overall blend was evaluated by EDS analysis of the powder in a
scanning electron microscope. The distribution of the element Y on
the total area of powders shown in FIG. 5(a) was seen to be
extremely uniform in FIG. 5(b). Such uniformity in the blend is a
major advantage since the small percentage of Y.sub.2O.sub.3 added
would cover a large surface area of chromium powders without using
mechanical alloying. Such distribution can be achieved in a blend
containing coarser (larger than micron) sized powders by using
additional mechanical alloying steps which increase process cost
and complexity.
[0042] The yttria enhanced the high temperature strength and the
oxidation resistance of the alloy. The adhesion between scale and
alloy is markedly improved and this increases the alloy's
resistance to thermal cycling exposure. Furthermore, the actual
growth rate of the oxide can also be reduced. Such nano sized oxide
particles also provide preferential nucleation sites for oxidation
of chromium and reduce oxide growth rate. In addition, the yttria
also modifies the chromium oxide layer microstructure, and hence
modifies diffusion rates and stresses in the oxide layer.
[0043] The powder metallurgy method of making a chromium base alloy
according to the third embodiment includes blending a first powder
comprising a chromium powder and a second powder comprising yttria
powder having an average particle size of less than 1 micron,
compacting a blend of the first and the second powders, and
sintering the compacted blend to form a chromium base alloy. The
chromium powder, compacting conditions and sintering conditions are
described in the prior embodiments. The method may also optionally
include blending the first powder and a third powder comprising at
least one of titanium, titanium hydride, zirconium or zirconium
hydride before, after or during the step of blending the first
powder and the second powder, followed by annealing at least the
first powder and the third powder in a reducing atmosphere after
the step of blending the first powder and the third powder
according to the process described in the second embodiment.
[0044] The effect of yttria on the oxidation rate of the alloy is
illustrated in FIG. 6. This figure shows a comparison of the
oxidation resistance of the alloy measured by thermogravimetry.
Curves (1) and (2) in FIG. 6 pertain to alloys made by the process
described in the second and third embodiments of the invention and
having the following composition:
Cr-4Fe-0.5Ni-0.5Ti-1Y.sub.2O.sub.3. The alloys corresponding to
curves (1) and (2) have a different density because the alloy which
corresponds to curve (2) was repressed after sintering while the
alloy which corresponds to curve (1) was not repressed. Curve (3)
pertains to a fully dense commercial alloy Ducrolloy and curve (4)
pertains to a simpler powder metallurgy version of the alloy
Cr-5Fe-1Y.sub.2O.sub.3. This alloy (4) was made by compaction and
sintering of powder that did not incorporate the processing
refinement of the embodiments of the current invention. The weight
gain by oxidation of the four alloys are provided in Table 6 below.
The lower weight gain shown by the alloys (1) and (2) made by the
process of the embodiments of the present invention confirms a
significantly higher oxidation resistance compared to the
commercial Ducrolloy alloy (3) with density close to theoretical,
even though the alloys (1) and (2) contained some porosity. In
alloy (4), the larger porosity level presumably resulted in some
breakage of the sample in oxidation as indicated by a sudden
decrease in mass followed by increase due to further oxidation of
the remnant sample.
TABLE-US-00006 TABLE 6 Thermogravimetric readings on mass change on
heating Sample Total mass No. Sample Description change, % (1)
Cr--4Fe--0.5Ni--0.5Ti--1Y.sub.2O.sub.3 sample 2.1 as sintered (2)
Cr--4Fe--0.5Ni--0.5Ti--1Y.sub.2O.sub.3 sample 1.2 sintered and
repressed (3) Ducrolloy Cr--Fe--Y.sub.2O.sub.3 sample, full density
3.9 (4) 5Fe--1Y.sub.2O.sub.3 sample, low density Erratic
[0045] It was possible to further enhance the compact density by
suitable selection and addition of the lubricant. In the fourth
embodiment, the powder blends of the alloy made according to the
prior embodiments incorporated either one of the two commercially
available lubricants employed widely in powder compaction. The
selection of either one of the lubricants, Acrawax C or Kenolube,
as shown in FIG. 7, enables the attainment of the projected green
density by addition of as low a proportion of the lubricant as
0.25%. In conventional compaction, lubricant additions of about 0.5
to 0.75% are common. In FIG. 7, the first bar shows the density of
the compact of Kenolube with CrTi powder, the second bar shows the
density of the compact of Kenolube with CrTiH.sub.2 powder and the
third bar shows the density of the compact of Acrawax C with
CrTiH.sub.2 powder. The benefits of reducing the amount of
lubricant include ease of removal of the lubricant prior to
sintering, and the densification of the volume occupied by the
lubricant during sintering being reduced to the corresponding
extent. Of the two commercial lubricants, Acrawax C and Kenolube,
the latter appears to be slightly superior.
[0046] FIG. 8 shows the coefficient of thermal expansion of the
alloy of the embodiments of the invention as evaluated by
dilatometry. The values of CTE obtained in the different
temperature ranges are shown in Table 7.
TABLE-US-00007 TABLE 7 Coefficient of Thermal Expansion Sample
Coefficient of Number Sample Composition thermal expansion 1 Alloy
30-250.degree. C.: 10.11 .times. 10.sup.-6/ .degree. C.
Cr--4Fe--0.5Ni--0.5Ti--1Y.sub.2O.sub.3 30-500.degree. C.: 10.33
.times. 10.sup.-6/ .degree. C. 30-750.degree. C.: 10.85 .times.
10.sup.-6/ .degree. C. 30-1000.degree. C.: 12.07 .times. 10.sup.-6/
.degree. C. 2 YSZ (from literature, for 10 .times.
10.sup.-6/.degree. C. comparison) 3 LSM (from literature, for 11
.times. 10.sup.-6/.degree. C. comparison) 4 Ni-YSZ (from
literature, for 13 .times. 10.sup.-6/.degree. C. comparison)
[0047] The data for sample numbers 2, 3 and 4 were taken from Azra
Selimovic, Miriam Kemm, Tord Torisson, Mohsen Assadi, "Steady state
and transient thermal stress analysis in planar solid oxide fuel
cells", Journal of Power Sources 145 (2005) 463-469. Thus, the
sintered alloys of the embodiments of the invention have a
coefficient of thermal expansion from 30.degree. C. to 1000.degree.
C. of between about 1.times.10.sup.-6/IC and about
13.times.10.sup.-6/.degree. C. This value is similar to the values
of YSZ electrolytes, LSM cathode electrodes and Ni--YSZ anode
electrodes that are widely used in solid oxide fuel cells. Thus,
when the alloy is used as a SOFC interconnect, it is CTE matched to
the components of the SOFC.
[0048] The foregoing description of the invention has been
presented for purposes of illustration and description. It is not
intended to be exhaustive or to limit the invention to the precise
form disclosed, and modifications and variations are possible in
light of the above teachings or may be acquired from practice of
the invention. The description was chosen in order to explain the
principles of the invention and its practical application. It is
intended that the scope of the invention be defined by the claims
appended hereto, and their equivalents.
* * * * *