SECA Core Technology Program: Materials Development at PNNL
J.W. Stevenson, Y.S. Chou, O.A. Marina, S.P. Simner, K.S. Weil, Z. Yang, and P. Singh Pacific Northwest National Laboratory Richland, WA 99352
SECA Core Technology Program Review Meeting Lakewood, CO, October 25, 2005
Outline
SOFC Materials Development Activities Overview / Technical Issues Being Addressed 5 Development Topics
Results & Discussion Summary Future Work
Acknowledgements
2
SECA CTP at PNNL: SOFC Materials Development “Technical Issues Addressed” “Technical Addressed”
Redox-, hydrocarbon-, and sulfur-tolerant anode materials Improved intermediate temperature cathode materials - (LSM, LSCF) Stable metallic interconnect materials -
Air
Performance evaluation (screen testing); Bulk/surface modification for enhanced performance
Also:
Stable, low resistance contact materials – electrode/interconnect interfaces Stable, thermally cyclable seals (glass-ceramic, reactive air brazes, compressive, compliant) Cost-effective fabrication techniques
Understanding/eliminating degradation mechanisms
3
Degradation Regions Addressed in Presentation
Interconnect/Cathode/Electrolyte (alloy oxidation, Cr poisoning, interfacial reactions, cathode instability) bipolar plate cathode electrolyte anode anode substrate
cell frame
Seal/Interconnect/Seal Interface (CTErelated stresses, interfacial reactions, seal volatility)
4
Accomplishments
Initiated joint study on Cr degradation effects on LSM cathodes Characterized behavior of LSM and LSCF cathodes Studied interconnect alloy behavior in SOFC environment Developed and tested protective coatings for alloys Initiated development of “refractory” glass-ceramic seals
5
Presentation Topics
Effects of Cr on Cathode Performance (Joint study w/ GE Energy and ANL) Degradation Mechanisms in Mixed-Conducting Cathode Materials Environmental Effects of Interconnect Oxidation Protective Coatings for Interconnects “Refractory” Glass-Ceramic Seals
6
Effects of Cr on Cathode Performance
Objective: To quantitatively assess, under realistic conditions, effects of Cr species on cathode performance
Determine under what conditions, if any, chromium transport has a detrimental effect on LSM-based cathodes Determine if the observed Cr transport is predominantly vapor phase, solid state or both Determine Cr compounds formed at cathode/electrolyte interface and cathode/interconnect interface regions Correlate Cr observed at interfaces vs. observed performance degradation (if any)
7
Effects of Cr on Cathode Performance
Approach: Collaboration with GE Energy & ANL Test Conditions
InDEC cells w/ LSM-YSZ cathodes
700, 800ºC; 1000 hours; realistic airflow conditions
Cr sources evaluated:
E-brite flow field (ANL, GE) Vapor phase delivered from upstream w/ Au flow field (PNNL) Transpiration experiments (PNNL)
Further discussion in next presentation (Atul Verma from GE Energy)
8
Presentation Topics
Effects of Cr on Cathode Performance (Joint study w/ GE Energy and ANL) Degradation Mechanisms in Mixed-Conducting Cathode Materials Environmental Effects of Interconnect Oxidation Protective Coatings for Interconnects “Refractory” Glass-Ceramic Seals
9
Degradation Mechanisms in Mixed-Conducting Mixed-Conducting Cathode Materials Experimental Approach
Long-term cell tests (750oC/0.7V) for La0.6Sr0.4Co0.2Fe0.8O3 (LSCF-6428) cathodes on anode-supported YSZ cells. Standard potentio-static/dynamic analysis (0.7 V, 750oC), and full cell impedance spectroscopy to separate ohmic and non-ohmic degradation. Processing and testing variables affecting degradation. Cathode calcination temperature – phase homogeneity. Constant voltage or OCV hold. Cathode-interconnect contact pastes Pre- and post-test analysis.
SEM, XPS, EDS, TEM, adhesion
10
LSM-20 vs. LSCF-6428 - Power Density and Long-Term Stability
LSM-20/SDC-20 composite cathode is relatively stable over >1000 hour period. LSCF offers significantly higher power densities but exhibits substantial degradation.
1 0.8 0.6 0.4 0.2 0
La(Sr)MnO3-CeO2 composite La(Sr)Fe(Co)O3 750oC, 0.7V 200 sccm (1:1 H :N2 mix)
2
300 sccm air
(Tested in absence of Cr)
0
200
400
600
800
1000
11
Time (h)
LSCF-6428 – Long-Term Impedance
0.15
0h 49h 242h 481h
100 Hz
0.1
0.05
1 kHz 10 Hz
0
0
0.1
0.2
0.3
1 Hz 0.4
0.5
0.6
Zreal (Ω.cm2)
Initial degradation attributed to substantial increase in cathode polarization (>1Hz). After 50-100 hours degradation predominantly associated with increased ohmic resistance.
12
Possible Sources of Degradation – Non-Ohmic
Increased Cathode Polarization
Coarsening of cathode microstructure
- loss of active electrochemical reaction area and/or impeded gas flow - no evidence from standard SEM but may be too subtle to detect; FE-SEM required.
Phase segregation within cathode
- changes in defect chemistry, and electrochemical activity (e.g. Sr precipitation leading to reduced acceptor dopant concentration).
Interdiffusion at interfaces
- forming solid solutions with altered defect chemistry.
13
Possible Sources of Degradation - Ohmic
To attribute the increased ohmic resistance to a reduction in bulk electronic conductivity of an individual layer: 0.15 - LSCF - Anode 103 decrease
0.1
0h 49h 242h 481h
102 decrease 101 decrease
100 Hz
- SDC or YSZ
0.05
1 kHz 10 Hz
Delamination at interfaces
- Loss of contact area
0
0
0.1
0.2
0.3
1 Hz 0.4
0.5
0.6
Zreal (Ω.cm2)
Formation of resistive interfacial phases 5 nm reaction layer with conductivity 10-5 S/cm would account for the observed increase in ohmic resistance.
14
LSCF Degradation – SrZrO3 Formation?
Sr-Zr-O diffusion (spure ~10-6 S/cm at 750oC) layers are sometimes observed after sintering of the cathode, but the layers do not discernibly increase in thickness during testing, and may even disappear. YSZ SDC LSCF
Sr-Zr-O
15
LSCF Degradation – XPS (Sr Enrichment) (Sr
Detailed microscopy of pre- and post-tested samples is being conducted but to date indicates no discernible changes in microstructure to account for the observed degradation. However, initial XPS analysis of pre-and post-tested samples suggests increased Sr segregation at the cathode-current collector interface after cell operation. For the pre-tested sample For the post-tested sample
Sr / (La+Sr) ≈ 0.5 (should be 0.4) Sr / (La+Sr) ≈ 0.65 with regions up to 0.9
Some Sr enrichment during cathode fabrication Extensive Sr enrichment during cell operation
Pure SrO conductivity at 750oC ~5 x 10-5 S/cm (W.D. Copeland, J. Phys. Chem. Solids 29 [1968] 313) 15-20 nm layer would account for observed ohmic increase.
16
LSCF Degradation – Influence of Processing and Test Variables
Variables studied to date: Calcination temperature of cathode powder – higher calcination temperatures improve phase homogeneity but provide no benefit to long-term stability. Thickness of ceria barrier layer – thicker interlayers prevent (or reduce) SrZrO3 formation but have no effect on long-term degradation. Constant voltage (0.7V) or OCV – cell continues to degrade in non-operation mode. Cathode-interconnect contact – manipulating this interface appears to have the most significant effect on long-term stability.
17
LSCF Degradation at Zero Current
Testing at zero current (with intermittent I-V/impedance measurement) indicates cell degradation thermally induced degradation.
1.2 1 0.8 0.6 0.4 0.2 0 0 50 100 150 200 250 0.7V hold 750oC 300 350
18
OCV testing
● Cell subjected to 5 min hold at 0.7V every 24 hours
Time (h)
Cathode Current Collector
Initial data indicates that changing the contact paste between the LSCF cathode and gold mesh current collector has the most significant effect on cell stability.
NOT TO SCALE Current-voltage wires Gold mesh current collector
LSCF cathode
Contact paste between cathode and mesh (gold or LSCF)
Anode-electrolyte substrate
19
Effects of Cathode Current Collector
0.7 0.6 Au Mesh - Au Paste 0.5 0.4 0.3 Au Mesh - LSCF Paste (Mn,Co) Coated Crofer - LSCF Paste
750oC/0.7V
0
50
100
150
200
250
300
350
Time (h)
Contact Paste Au LSCF
Degradation (%/hour)
0 - 100 h 0.2 0.07 100 - 200 h 0.08 0.02 200+ h 0.04 <0.01
20
Cathode Current Collector – Effect on Stability
Au contact paste
0.1
J
LSCF contact paste
0.1
0h 146h
J
0h 146h
0.08 0.06 0.04 0.02 0 0.1
10 Hz
J JJ J J JJJJJ J JJ
J
0.08 0.06 0.04
1 Hz 10 Hz
J
J J 100 Hz J JJJJJJ J JJ J JJJJJJ J JJ J JJJ J J JJ J J JJ J JJJ JJ JJJ JJ 1 JJ 10 Hz J J J JJ JJ JJ 100 Hz J J J J J J 2.5 kHz JJ
J
J
JJ
JJ
1 Hz
JJ JJJ J JJ J
J
Hz
0.02 3
0.2
0.3
0.4
2
0.5
0.6
0 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45
J J J J J J J J J J JJ J J JJJ J JJ J JJ J J J J J J J J JJJJ J J JJJ J JJ J J JJ JJJJJJJJJ J J J J J JJ J kHz JJJJJJJJJJJJJJJJJ J J J J 100 Hz J 100 Hz J J JJ JJJ
Zreal (Ω.cm ) Similar ohmic resistance in both cases; for Au paste, non-ohmic degradation mechanism(s) correlated with frequencies < 100 Hz; for LSCF, with frequencies > 100 Hz
Zreal (Ω.cm2)
Phenomena are reproducible
21
Summary
LSCF-6428 offers higher power density than LSM but exhibits degradation over time.
Both ohmic and non-ohmic factors associated with observed degradation in performance Degradation mechanisms may include the formation of SrO at the cathode surface. Manipulation of the cathode-current collector interface results in reason(s) not yet established. significant changes in stability
22
Future Work
Complete Cr degradation study (with GE Energy and ANL) Complete investigation of LSCF degradation mechanisms, considering the effects of:
Materials - Sr content, A-site deficiency, acceptor dopant (Sr or Ca). Processing – cathode firing temperature, cathode and ceria thickness. Testing – temperature, air or O2 cathode gas, cathode-current collector configuration.
Evaluate dependence of performance of LSM-based cathodes on various composition/processing parameters Optimize LSM-based cathodes for higher stable performance
23
Presentation Topics
Effects of Cr on Cathode Performance (Joint study w/ GE Energy and ANL) Degradation Mechanisms in Mixed-Conducting Cathode Materials Environmental Effects on Interconnect Oxidation Protective Coatings for Interconnects “Refractory” Glass-Ceramic Seals
24
Environmental Effects on Interconnect Oxidation
Objective: To evaluate and understand oxidation and corrosion behavior of SOFC interconnect alloys under simulated stack operating conditions Approach:
Conduct oxidation tests of selected alloys under single and dual (oxidizing/reducing) environments representative of SOFC interconnection exposure conditions Examine alloy and oxide scale chemistry and structure using metallography, SEM, XRD, XPS, EDS etc
25
Potential Candidate Alloy Systems
Fe
Emphasis on “Chromia-forming” Ferritic Stainless Steels: •CTE match Austenitic •Conductive, stainless steels protective oxide Fe-Ni-base scale superalloys •low cost Ni-Fe-base •ease of superalloys fabrication
B Ni
FS S
Ferritic stainless steels
AS S
B Fe
BCC + F CC
SA
FCC
Cr-base alloys
BC C
C
BA r
Cr
Also: Co-base superalloys
Ni
Yang, Weil, Paxton, Stevenson, J. Electrochem. Soc., 150, A1188 (2003).
•Ni-base superalloys also of interest
SA
26
Dual Atmosphere Study: Experimental Approach: Approach
T/C Fuel Air Air T/C
Materials studied:
FeSS
x x x x x x x x x x
NiBA
Seal
{ {
E-brite-27%Cr Crofer22-22%Cr AISI430-17%Cr Haynes 230-22%Cr Hastelloy S-17%Cr Haynes 242-9%Cr Ag, Ni
Air
Air
Variables: Alloy composition Isothermal vs. cycling Hydrogen & Simulated reformate
27
Stainless steel
Oxidation of Alloys in Dual (Air/Hydrogen) Environment
Previous PNNL results: Air-side oxidation of FSS is function of Cr content, thermal history, environment Effects of dual atmosphere on oxidation (iron incorporation into oxide scale; local attack via formation of iron oxide nodules) increases with decreasing Cr content:
Ebrite (25% Cr): no significant effect Crofer22 APU (22% Cr) and 430 (17% Cr) both significantly affected Fe2O3
430 Isothermal: 800oC, 300h
The presence of moisture in air, thermal cycling, and higher temperatures further accelerate the anomalous air side oxidation Little difference observed on fuel side Ni and Ni-base alloys did not exhibit significant dual atmosphere effect
Related work at Allegheny Tech., Univ. of Pitts., ARC; also MCFC (FuelCell Energy)
A-A
Fe
Cr Mn
b
Crofer22 APU Thermal cycling: 800oC, 3x100h
28
Extension to Carbon-containing Carbon-containing environments
Motivation: SOFCs will operate on partially and fully reformed hydrocarbons and syngas containing various levels of CHx, CO and CO2. There is a need to understand effects of C species in fuel gas on alloy oxidation/corrosion behavior:
Accelerated metal loss (e.g., pitting) has been reported in complex gas atmospheres (Birks, Meier, Natesan, Pettit and others) Carburization and sensitization of stainless steels have been reported via molecular and chemical transport of gaseous (CO,CO2, CHx) species. Ni clad FSS experienced extensive carburization (carbide formation) due to C diffusion through Ni cladding (Jian)
Question: Will thermally grown oxide scales provide effective protection and diffusion barrier at 600-800ºC?
29
Preliminary results
Crofer22 APU; 0.5mm thick; 1000 hours; 800ºC Air vs. “Fuel”
“Fuel” composition:
Temp. (oC) 800 (calc.) 800 (exp.)* H2 72.1 74.2 CO 12.4 9.8 CH4 0.2 0 CO2 2.5 6.2 H2O 12.8 9.7
*Outlet composition analyzed with GC
Oxide scale chemistry and structure similar to previous Cfree tests (air vs. moist hydrogen) Some evidence of highly localized carbon diffusion into the metal
30
Airside
Fe
Preliminary results
Fuel side
Fe
Cr
Cr Mn Mn O O C
Fe
Cr Mn O Al Mn
Cr
Fe
31
Presentation Topics
Effects of Cr on Cathode Performance (Joint study w/ GE and ANNL) Degradation Mechanisms in Mixed-Conducting Cathode Materials Environmental Effects of Interconnect Oxidation Protective Coatings for Interconnects “Refractory” Glass-Ceramic Seals
32
Protective Coatings for Interconnects
Objective: To develop protective coatings for SOFC interconnect alloys which will reduce oxidation kinetics and, if necessary, mitigate Cr volatility Approach:
Synthesize and characterize coating materials Fabricate coatings onto interconnect alloy coupons Evaluate structure and performance of coatings and alloy substrate by SEM, XRD, resistance measurements, etc.
33
Oxidation of Alloy Interconnects
ASR (ohm.cm2)
Long-term issues: Increased scale resistance Scale spallation Localized metal loss Cr volatility Approaches to solution: Bulk alloy modification, cladding, surface modification (including coatings)
0.015 0.012 0.009
Crofer22 APU
0.006 0.003 0 0 200 400 600 800 1000 1200 1400 1600 1800 Time (hours)
800oC, air Pt paste contact
A-A
Fe
4000 h, 800ºC w/ thermal cycles
b
Cr Mn
34
Protection Layers on Interconnects
Conductive oxide layer: Acts as a mass transport barrier to both Cr3+ outward and O2- inward diffusion Reduce subscale growth rate Minimize scale resistance Prevent Cr release from metal substrate Conductive oxide
Cr2O3
Chromia subscale: Slow growth via O2inward diffusion;
Chromia forming alloy
Requirements: High electrical conductivity, stability over SOFC temperature range, appropriate thermal expansion, low/no Cr volatility Candidate Oxides: Perovskites: LSM, LSCr, LSCo, LSCF, LSF Spinels: (Mn,Co)3O4
35
(Mn,Co)3O4 Spinel
• Extensive solid solution between Mn3O4 and Co3O4 • Wide range of compositions available for consideration
•Related work:
• Larring & Norby, J. Electrochem. Soc., 147, 2000 •LBNL
Mn1.5Co1.5O4
E. Aukrust and A. Muan, J. Am. Ceram. Soc., 46, 511, 1963
36
Electrical Conductivity of (Mn,Co)3O4 Spinels
σ Mn
2
1.5Co1.5O4
=
3~ 4
Electrical Conductivity vs. Temperature
10 σ MnCr2O4 = 10
70 60 50 40 30 20 10 0 0
σ Cr O
2
1.00E+02
3
Electrical Conductivity (S/cm)
1.00E+01
1.00E+00
Mn1.5Co1.5O4
800ºC
1.00E-01
1.00E-02
Conductivity (s)
1.00E-03
1.00E-04
MnCr2O4 Sasamoto, et al. MnCo2O4 NiMn2O4 MnCo2O4 Measured at PNNL Mn1.5Co1.5O4 Mn2CoO4
Mn1Co2O4 Mn2Co1O4 Mn2.5Co0.5O4
1 2 3 Co/Mn 4 5
1.00E-05 0 200 400 T(ºC) 600 800 1000
Mn0.5Co2.5O4
37
HT XRD of Mn1.5Co1.5O4: from RT to 600oC, 5oC/min
600oC
500oC 400oC
300oC 200oC
100oC RT
28 33 38 2θ
38
43
48
Thermal Expansion
Structurally stable up to 1,200oC Good CTE matching to the FSS substrate:
1.5 1.5 4
CTEMn Co O = 11.5 × 10−6 K −1,20 − 800o C
Mn1.5Co1.5O4
2.0 1.8 1.6 1.4 1.2 δ L/L% 1.0 0.8 0.6 0.4 0.2 0.0 0 200 400 600 Temperature ( C)
39
o
MnCo2O4 Mn2CoO4 Mn0.5Co2.5O4 Mn2.5Co0.5O4
800
1000
1200
Thermogravimetric Analysis
Minimal change in oxygen stoichiometry of spinel with temperature
0.20 0.00 -0.20 -0.40 -0.60 -0.80 -1.00 -1.20 0 200 400 600 800 1000 1200 1400 Temperature (ºC)
% Change in Mass
LSCF-6428 Mn1.5Co1.5O4
Measured in air
40
Thermal Growth of Mn1.5Co1.5O4 Spinel Protection Layer
Approach
Synthesis of Mn1.5Co1.5O4 Powder
Solid state reaction or combustion synthesis
Preparation of slurry and application of coating
Screen printing, dip coating, brush application
Heat treatment in reducing environment
4Mn1.5Co1.5O4 ⇒ 6Co + 6 MnO + 5O2 ↑
Oxidation and reactionsintering in oxidizing environment
6Co + 6 MnO + 5O2 ↑⇒ 4 Mn1.5Co1.5O4
Reaction sintering of Co, MnO mixture provides increased coating density at 800ºC
41
Study of Interfacial Resistance
ASRcathode / int erconnect = Φ (scale, contacts, reactions )
I V
Interconnect Screen-printed cathode
Cathode Material
Sintered, dense pellet Contact paste
V 6.5 PSI Load 500mA.cm-2
1.0 ’’
ca th od e
I
2.0 mm
Screen printed, sintered cathode, with same composition as a real cathode, as well as thickness, density and surface finish. Sintered perovskite pellet with same composition as cathode, but higher density. ASR @ 800°C = ~1.0 mΩ-cm2
42
¾’’
Pr in
te d
ASR Test: 6 months, 125 thermal cycles
25
ASR (mohm.cm )
•Spinel-coated Crofer22 APU, LSCM contact paste, LSF “cathode” •Thermally cycled from RT to 800ºC in air
20
2
15
10
5
0 0 25 50 75 100 125 Cycle Number
43
i=500 mA.cm2
3
Crofer22 APU
4
8
7
protection layer
Contact layer LSF cathode
LSF substrate
5
6
LSF cathode Contact layer
Crofer22 APU
protection layer
1
SEM/EDS Analysis
2
i=500 mA.cm2
44
Effective Cr and O Barrier
No Cr penetration after 6 months of test and 125 cycles 7-7 5-5
Fe
Fe
Cr Mn Co Mn
Cr Co
45
Improved Surface Stability
No spallation was observed at areas with protection layer Significant spallation of Mn-Cr spinel was found on surfaces without protection layers 1-1 6-6
46
Fabrication and performance of Mn1.5Co1.5O4 Protection Layers on Other FSSs
(Mn,Co)3O4 protection layers applicable to other ferritic stainless steels as well Stability and performance depend on their ferritic substrate composition
45 40 Area Specific Resistance (mOhm.cm )
2
Fe Cr Co Si
35 30 25 20 15 10 5 0 0 50 100 150 200 250 Time (hour) 300 350 400 450 E-brite AISI430
LSCM
LSCM electrical contact, 800oC, air
Spinel
AISI430
Fe Cr Co Si
LSCM
Spinel
E-brite
47
Summary
Initiated dual atmosphere studies with simulated reformate; observed Fe enrichment of fuel side scale. In 1000 hour test on Crofer22 APU at 800ºC, no iron oxide nodules observed on air side, and no pitting corrosion observed on fuel side Spinel protection layers successfully fabricated onto several different FSS; substantial reduction of subscale growth and Cr transport observed.
48
Future Work
Interconnects:
Continue to investigate impact of dual atmosphere (air/reformate) tests on alloy oxidation/corrosion behavior Optimize composition, structure, and thickness of spinel coatings Evaluate alternative approaches for coating fabrication
49
Presentation Topics
Effects of Cr on Cathode Performance (Joint study w/ GE and ANNL) Degradation Mechanisms in Mixed-Conducting Cathode Materials Environmental Effects of Interconnect Oxidation Protective Coatings for Interconnects “Refractory” Glass-Ceramic Seals
50
“Refractory” Glass-Ceramic Seals “Refractory” Glass-Ceramic
Objectives
To develop and evaluate new “refractory” sealing glasses with higher sealing temperatures than typical SOFC sealing glasses to minimize seal reactivity and increase seal stability Evaluation includes sealing temperature, crystallization behavior, CTE, chemical compatibility, leak testing, and thermal cycle stability Standardized tests allow for meaningful comparison between different sealing materials (PNNL or other SECA participants)
51
Why “refractory” glass (glass“refractory” (glassceramics) for sealing?
•State-of-the-art sealing at ~850ºC: G18 (US Pat.: 6430966)
BaO CaO Al2O3 SiO2 B2O3
35 15 5 35 10
Glasses (including residual glass in glass-ceramics) are more reactive near or above their melting point. Residual glass likely to have high boria content. Glasses of higher sealing temperatures (≥ 950oC) may exhibit lower interfacial reactivity and better thermal stability (e.g., lower mobility, volatility) during longterm operation at 700-800oC.
•Initial CTE: ~12.5x10-6/oC •Aged CTE (1000hrs/750oC) ~11.1x10-6/oC due to formation of mono-celsian phase •Reaction with alloy to form BaCrO4 (αa=16.5, αb=33.8, αc=20.4)
52
Experimental Approach
glass formulation: oxide mixing glass melting @ 1500oC/0.5h glass casting glass annealing @ 600oC/6h thermal properties measurement Tg, Ts, CTE
glass crushing grinding & milling
die-press sintering
thermal properties measurement Tg, Ts, CTE 1”x1” coupon sealing @ To(C), hr
thermal stability test
thermal properties & microstructure characterization
tape casting
RT leak test
thermal cycle & stability test
RT leak test
postmortem analysis
53
“Refractory” Glass-Ceramic Seals “Refractory” Glass-Ceramic
Compositional Approaches:
Minimize B2O3 content
Glass former, also acts as flux, so reduction of B2O3 increases glass refractoriness Evaluate compositional effects on CTE, reactivity
Alkaline earth substitution (MgO, CaO, SrO, BaO)
Evaluate compositional effects on sealing temp, CTE, reactivity
54
Approach 1: minimize B2O3 content Thermal properties
properties of as-cast glass 820 800 degree in C
CTE (ppm/C) 12 11.8 11.6 11.4 11.2 11 10.8 0 1 2 3 4 B2O3% 5 6 7 8 thermal properties TEC(glsss) TEC(crystallized)
780 760 740 720 700 0 1 2 3 4 5 6 7 8 9 10 B2O3 %
#ID YS1 YS61 YS7 YS8 YS9 YS11 YSO1 SrO 42.5 42.5 42.5 42.5 42.5 40.0 42.5 CaO Al2O3 10 4 10 4 10 4 10 4 10 4 10 10 4 0
Tg Ts
Y2O3 6 6 6 6 6 6 6
G18: Tg – 630ºC Ts – 685ºC
B2O3 7.5 6.0 5.5 4.5 3.5 7.5 7.5
SiO2 30.0 31.5 32.0 33.0 34.0
note clear glass clear glass clear glass with very light skin clear glass with very light skin 1/3 opaque
32.5 clear glass 34.0 clear glass
55
Dilatometric behavior
crystallized G18 (750c/4h) 1st heating 1.20E-02 1.00E-02 linear expansion
YS1 att 1000C/1h 051005 1.20E-02 1.00E-02 linear expansion 8.00E-03 6.00E-03 4.00E-03 2.00E-03 0.00E+00
8.00E-03 6.00E-03 4.00E-03 2.00E-03 0.00E+00 0 100 200 300 400 500 600 700 800 900 temperature (C)
0
100 200 300 400 500 600 700 800 900 temp(C)
G18, 850oC/1h, 750oC/4h
YS1, 1000oC/1h, 800oC/4h
56
RT leak test of sealed coupons
Metal squares (1”x1”) and YSZ-bilayer (0.75”x0.75”) are sealed with glass at ~5 psi
All tests remained hermetic through 10 thermal cycles; indication of acceptable CTE match and minimal reactivity; larger and longer tests required
glass# YS8 YS8 YS8 YS1 YS1 YS1 YS1 YSO1 YSO1 YSO1 YSO1 YSO1 YSO1 YSO1 YSO1
metal Crofer-ox Crofer-ox Crofer-ox Crofer-ox Crofer-ox Crofer-ox Crofer-ox Crofer-ox Crofer-ox Crofer-ox Crofer-ox Crofer-AR Crofer-AR Crofer-AR Crofer-AR
sealing 1050/1 950/1 1100/1 850/1 900/1 950/1 1000/1 900/1 950/1 1000/1 1050/1 900/1 950/1 1000/1 1050/1
as-sealed hermetic hermetic hermetic hermetic hermetic hermetic hermetic hermetic hermetic hermetic hermetic hermetic hermetic hermetic hermetic
10 T.C./air hermetic hermetic hermetic hermetic hermetic hermetic
10 T.C./red. hermetic hermetic hermetic hermetic
hermetic hermetic hermetic hermetic hermetic hermetic hermetic hermetic
57
Interfacial reactions (glass/YSZ)
G18
YS8 YSZ
BaZrO3 YSZ
G18, 850oC/1h+750oC/4h
YS8, 1050oC/1h+800oC/4h
58
Interfacial reactions(glass/crofer22APU)
crofer
3 1 2 1 4
2
crofer
3
G18m, 850oC/1h+800oC/4h
ponit # 1 2 3 O Al Si Ca 39.11 1.77 2.21 0.82 10.97 12.5 48.91 1.29
YS8, 1050oC/1h+800oC/4h
Al Si Ti 1.83 1.56 0.33 2.32 0.86 0.59 1.73 0.75 1.31 Cr Mn 35.82 1.40 27.59 7.9 22.93 22.96 Fe Sr 20.91 0.73 34.72 0.38 76.48 73.24
ponit # O Cr Mn Fe Ba 1 37.42 34.51 1.49 13.6 6.50 2 26.23 16.38 60.17 3 24.86 1.28 23.7 4
59
Approach 2: Replacement of Ca with other alkaline earth (Mg, Ba, Sr)
Ba+2 (1.36Ǻ), Sr+2 (1.16Ǻ), Ca+2 (1.00Ǻ), Mg+2 (0.72Ǻ)
# ID YSP4 YSP5 YSP6 YSP7 SrO 45.0 45.0 45.0 52.5 MgO 7.5 0 0 0 CaO 0 7.5 0 0 BaCO3 0 0 7.5 0 Y2O3 6 6 6 6 B2O3 10.0 10.0 10.0 10.0 SiO2 31.5 31.5 31.5 31.5 note clear glass clear glass clear glass clear glass
glass # YSP4 YSP5 YSP6 YSP7
Tg 653 661 660 661
Ts TEC(glsss) TEC(crystallized) heat treatment 708 11.45 9.91 (RT-866C) 600/2,1000/1,800C/4h 713 11.84 11.51 (RT-873C) 600/2,1000/1,800C/4h 704 12.10 12.43 (30-860C) 600/2,1000/1,800C/4h 711 11.92 12.27 (30-928C) 600/2,1000/1,800C/4h
60
Thermal stability: effect of ageing on CTE
glass# G18 YS1 YS61 YS8 YSO1 YSP5 YSP6 YSP7 CTE (as-sintered) 12.50 (RT-570C) 11.61 (RT-946C) 11.69 (100-934C) 11.50 (RT-1000C) 11.73 (RT-870C) 11.51 (RT-873C) 12.43 (30-860C) 12.27 (30-928C) CTE (1000hr air aged) 11.10 (RT-850C) 11.33 (RT-961C) 10.89 (RT-940C) 10.84 (RT-986C) 11.51 (RT-962C) to be determined to be determined 12.43 (RT-1000C) ageing temp 750 900 900 900 900 900 900 800
61
Short-term ageing test in 30%H2O/70%H2 Short-term
glass # G18 G18 YS1 YS1 YS1 YS1 YSO1 YSO1 P5 P5 P6 P6 P7 P7 as-sealed hermetic hermetic hermetic hermetic hermetic hermetic hermetic hermetic hermetic hermetic hermetic hermetic hermetic hermetic 850C/200h/30%H2O leak hermetic hermetic hermetic hermetic hermetic hermetic hermetic hermetic hermetic hermetic leak hermetic hermetic 850C/500h/30%H2O leak hermetic hermetic hermetic hermetic hermetic hermetic hermetic hermetic hermetic hermetic hermetic
•Tests included 3 thermal cycles •Survival of seals attributed to minimal chemical interaction and good CTE match
62
Summary
“Refractory” sealing glasses were formulated and characterized. The effects of reducing B2O3 content, and substitution of various alkaline earths, were evaluated. Coupon tests showed hermetic sealing to Crofer22 APU for several glasses even after 10 thermal cycles in air and/or in reducing (70% H2/30% H2O) environment. No distinct interfacial reaction of “refractory” glasses with YSZ or Crofer22 APU was observed. Several refractory glasses retained hermetic sealing to FSS after 850oC/500 hrs ageing in air and reducing environments, indicating good CTE match and chemical stability with Crofer22 APU.
63
Future Work
Glass-ceramic seals:
Continue optimizing the refractory sealing glass formulations with respect to sealing temperature, CTE, crystallization rate Study the chemical compatibility of the refractory sealing glasses with interconnect alloys; testing extended to include coated alloys. To evaluate the thermal stability (1000 hrs) of the refractory glasses in simulated SOFC environments. Validation of results on larger samples (4”x4”); thermal cycle tests in dual environments.
64
Acknowledgements
The work summarized in this paper was funded under the U.S. Department of Energy’s Solid-State Energy Conversion Alliance (SECA) Core Technology Program. The authors wish to thank Wayne Surdoval, Lane Wilson, Travis Shultz, and Don Collins (NETL) for their helpful discussions regarding this work. Additional PNNL contributors: G. Xia, M. Anderson, G. Maupin, J. Coleman, S. Carlson, N. Saenz.
65