Fundamental Studies of SOFC Materials
Eric D. WachsmanUniversity of Florida - U.S. Department of Energy
High Temperature Electrochemistry CenterDepartment of Materials Science and Engineering
University of Florida
QuickTime™ and aGIF decompressor
are needed to see this picture. UF-DOE HiTEC
[email protected]://hitec.mse.ufl.edu
Determination of Electrochemical Performance and Thermo-Mechanical-Chemical Stability of SOFCs from Defect ModelingDOE SECA Contract No: DE-FC26-02NT41562DOE Project Manager: Travis Schultz•Advance the fundamental understanding of the continuum-level electrochemistry of oxide mixed ionic-electronic conductors.•Obtain fundamental constants required for implementing the continuum-level electrochemical model from experiment. •Extend the models to multilayer structures and incorporate microstructural effects.•Verify the models through experiment.•Develop a transient version of the continuum-level electrochemical model.•Obtain time constants for various transport processes from electrical impedance spectroscopy to examine the effects of transients on SOFC performance.•Develop and deliver software modules for incorporation of the continuum-level electrochemical model into SOFC failure analysis software used by NETL, PNNL, ORNL and the SECA industrial teams.
Electrocatalytically Active High Surface Area Cathodes for Low Temperature SOFCsDOE EE/FE Contract No: DE-FC26-03NT41959DOE Project Manager: Lane Wilson•Develop a fundamental understanding of heterogeneous electrocatalytic phenomena at the surface of ion conducting ceramics.•Fabricate high surface area SOFC cathodes with controlled microstructure and porosity.•Develop low ASR cathodes for low to intermediate temperature SOFCs.
UF - DOE High Temperature Electrochemistry CenterDOE Advanced Research, HiTEC Contract No: DE-AC05-76RL01830DOE Project Manager: Lane Wilson•Develop the fundamental understanding of ionic transport in, and electrocatalytic phenomena on the surface of, ion conducting materials, spanning the range from first-principles calculations and molecular dynamic simulations of ionic transport and gas-solid interactions to synthesis and characterization of novel ion conducting materials and electrocatalysts.
Highlights of:
Brouwer Region III2cOi
= ch
Brouwer Region Ice = 2cVO
20 15 10 5 0 5 108
6
4
2
0
VO••
h •
′ ′ O i
′ e
′ M A
Brouwer Region IIacD = 2cVO
Brouwer Region IIb
cD = ch
Intermediate-PO2 (model)
Low-PO2 (model) High-PO2 (model)
log
[ i ]
(m
-3)
DEFECT STRUCTURE DEFECT EQUILIBRIA
log PO2 (atm.)
Defect Concentration Dependence on:• Defect formation energy• Temperature• PO2 cV =
34
FUNDAMENTAL PROPERTIES
Kr
12PO2
− 14 +
A2
⎛ ⎝ ⎜
⎞ ⎠ ⎟
32
⎡
⎣
⎢ ⎢
⎤
⎦
⎥ ⎥
23
VO••
Defect Energetics and Mobility Based on:• Crystal structure• Cation radii• Cation polarizability• Cation oxidation state• Etc.
QuickTime™ and aTIFF (LZW) decompressor
are needed to see this picture.
Ebond = A
r m
EXTENSION OF MODEL TO THERMO-MECHANICAL PROPERTIES
− Br n
A, B, n and m are constants
Lattice constant, a, has linear relationship with cv
Therefore, r ~ a ~ cV
Thermal expansiona − a0
a0
= αΔT
Chemical expansiona − a0
a0
=θa0
cV
Thermo - chemical expansiona − a0
a0
= αΔT +θa0
cV
cV =34
Kr
12PO2
− 14 +
A2
⎛ ⎝ ⎜
⎞ ⎠ ⎟
32
⎡
⎣
⎢ ⎢
⎤
⎦
⎥ ⎥
23
0
5000
10000
15000
20000
25000
0 24 48 72 96 120 144
Heat up in air
Dry Argon
5% wet H2/N
2
50% wet H2/N
2
Wet H2
Dry H2
Cool down in H2
Decreasing pO2
Ther
mo-
Che
mic
al E
xpan
sion
Δl/l
o x 1
0-6
Time (hrs)
Thermo-Chemical Expansion of GDC
EXTENSION OF MODEL TO THERMO-MECHANICAL PROPERTIES
0 100
3 103
6 103
9 103
12 103
0 5 10 15 20 25-log( pO
2 ) /atm
Che
mic
al E
xpan
sion
Δl/l
o x 1
0-6
0 100
3 103
6 103
9 103
12 103
0 5 10 15 20 25Che
mic
al E
xpan
sion
Δl/l
o x 1
0-6
-log( pO2 ) /atm
Kr = 1072 m-9atm-0.5 [1]
A = 0 m-3
θ = 3.2 x 10-3 nm3 [2,3]
Undoped Ceria GDC
Δll0
= αΔTthermal{ +θ 3
4KR
12 PO2
−14 + 1
2A( )
32
⎛
⎝ ⎜
⎞
⎠ ⎟
23
chemical1 2 4 4 4 3 4 4 4
Kr = 1072 m-9atm-0.5 [1]
A = 2.5 x 1027 m-3
θ = 3.2 x 10-3 nm3 [2,3]
1. T. Kobayashi et al., Solid State Ionics, 126 (1999) 3492. D.J. Kim, J. Am. Ceram. Soc., 72 (1989) 14153. M. Mogensen et al., Solid State Ionics, 129 (2000) 63 4. K. Sasaki and J. Maier, Solid State Ionics, 134 (2000) 303
EXTENSION OF MODEL TO THERMO-MECHANICAL PROPERTIES
0 100
3 103
6 103
9 103
12 103
0 10 20 30 40 50-log( pO
2 ) /atm
Che
mic
al E
xpan
sion
Δl/l
o x 1
0-6
Kr = 1060 m-9atm-0.5 [4]
A = 4.4 x 1027 m-3
θ = 3.2 x 10-3 nm3 [2]
YSZ
QuickTime™ and aTIFF (LZW) decompressor
are needed to see this picture.
Ybond = 1r0
d 2Edr2
⎛
⎝ ⎜ ⎜
⎞
⎠ ⎟ ⎟ r=r0
Ebond = A
r m − Br n
A, B, n and m are constants
Lattice constant, a, has linear relationship with cv
Therefore, r ~ a ~ cV
1. D-J. Kim, J. Amer. Ceram. Soc. 72 (1989) 1415.2. M. Mogensen, N. Sammes, G. Tompsett, Solid State Ionics 129 (2000) 63
YY∗
≈ aa0
⎛
⎝ ⎜
⎞
⎠ ⎟ − δ+3( )
aa0
= θcV +1
Y ≈ Y∗ θcV +1( )− δ+3( )
δ is equivalent to:• n (if A is constant)• m (if B is constant)as oxygen vacancies are introduced
EXTENSION OF MODEL TO THERMO-MECHANICAL PROPERTIES
EXTENSION OF MODEL TO THERMO-MECHANICAL PROPERTIESExperimental Validation
PO 2= 0.22 atm
PO2=9.5x10-5 atm
PO2=1.8x10-17 atmPO
2=4.5x10-22 atmPO
2 =5.1x10 -25atm
A
B
C
D
E
F
PO 2= 0.22 atm
PO2=9.5x10-5 atm
PO2=1.8x10-17 atmPO
2=4.5x10-22 atmPO
2 =5.1x10 -25atm
A
B
C
D
E
F
PO 2= 0.22 atm
PO2=9.5x10-5 atm
PO2=1.8x10-17 atmPO
2=4.5x10-22 atmPO
2 =5.1x10 -25atm
A
B
C
D
E
F
P O2
Vacancies preserved by fast cooling
5oC/
min
Fast cool
800oC, 5 hr
Tem
p. o C
Time, hr
H2, H2/H2O,N2, Air
EXTENSION OF MODEL TO THERMO-MECHANICAL PROPERTIESExperimental Validation - Nanoindents and Microstructure
SEM image of surface after thermal etch. Average grain size ~12 µm.
20 μ m
1 µ m
Nanoindents
Size: ~0.6 µm
Depth: ~125 nm
•Effect of crystallographic orientation on elastic modulus and hardness evaluated statistically by applying many indents on grains of known orientation.
•In-plane anisotropy can be measured by changing the indent orientation.
EXTENSION OF MODEL TO THERMO-MECHANICAL PROPERTIESExperimental Validation - Nanoindents and Microstructure
SEM image of surface after thermal etch. Average grain size ~12 µm.
• 100 indents were applied on the sample, which covered 100 µm X 100 µm ( ~25 different grains)
Modulus: 218.35±11.12 GPa
Hardness: 9.00±0.73 Gpa
• The small variations imply that ceria is elastically isotropic.
20 μ m
1 µ m
Nanoindents
Size: ~0.6 µm
Depth: ~125 nm
experiment
Y,Effect of Oxygen Vacancy Population on Elastic Modulus of
Ceria(measured in air)
Y,
experiment
model
Y( PO2) ≈ Y∗ θcV (PO2
) +1( )− δ+3( )
Effect of Oxygen Vacancy Population on Elastic Modulus of
Ceria(measured in air)
Y,
experimentmodel
Effect of Oxygen Vacancy Population on Elastic Modulus of
Ceria(measured in air)
Y( PO2) ≈ Y∗ θcV (PO2
) +1( )− δ+3( )
Y,
experiment
Effect of Oxygen Vacancy Population on Elastic Modulus of
Gadolinia-Doped Ceria (GDC)
(measured in air)
Y,
experiment model
Y( PO2) ≈ Y∗ θcV (PO2
) +1( )− δ+3( )
Effect of Oxygen Vacancy Population on Elastic Modulus of
Gadolinia-Doped Ceria (GDC)
(measured in air)
Y,
experiment model
Effect of Oxygen Vacancy Population on Elastic Modulus of
Gadolinia-Doped Ceria (GDC)
Y( PO2) ≈ Y∗ θcV (PO2
) +1( )− δ+3( )
(measured in air)
Y,
experiment
Effect of Oxygen Vacancy Population on Elastic Modulus of
Yttria-Stabilized Zirconia (YSZ)
(measured in air)
Y,
experimentmodel
Y( PO2) ≈ Y∗ θcV (PO2
) +1( )− δ+3( )
(measured in air)
Effect of Oxygen Vacancy Population on Elastic Modulus of
Yttria-Stabilized Zirconia (YSZ)
Higher temperature and higher current will shift decrease in modulus to higher PO2
Y,Effect of Oxygen Vacancy Population on Elastic Modulus of
Ceria, GDC,YSZ
Y( PO2) ≈ Y∗ θcV (PO2
) +1( )− δ+3( )
Extend to Include Microstructural Effects:
Y(p) = Yp=0(1−p)r where p is porosity and r ≈ 2 for porous ceramics [1]
Y(PO2, p) = Y*(θcV(PO2)+1)-(δ+3)(1−p)r
1. A. S. Wagh, R. B. Poepel, J. P. Singh, J. Mat. Sci., 26 (1991) 3862.
EXTENSION OF MODEL TO THERMO-MECHANICAL PROPERTIESY,
Y(PO2) ≈ Y ∗ θ 3
4 KR
12 PO2
− 14 + 1
2 A( )32( )
23
chemical1 2 4 4 4 3 4 4 4
+1⎛
⎝
⎜ ⎜ ⎜
⎞
⎠
⎟ ⎟ ⎟
− δ +3( )
0 100
3 103
6 103
9 103
12 103
0 5 10 15 20 25-log( pO
2 ) /atm
Che
mic
al E
xpan
sion
Δl/l
o x 1
0-6
Δll0
= αΔTthermal{ + θ 3
4 KR
12 PO2
− 14 + 1
2 A( )32( )
23
chemical1 2 4 4 4 3 4 4 4
SAME!
1. D-J. Kim, J. Amer. Ceram. Soc. 72 (1989) 1415.2. M. Mogensen, N. Sammes, G. Tompsett, Solid State Ionics 129 (2000) 63
KR = Equilibrium constant for VO•• formation
A = Dopant concentration
θ = Empirical constant = 3.2 x 10-3 nm3 [1,2]
FUNDAMENTAL QUANTITATIVE DEFECT CONSTANTS
UF-DOE HiTEC
Thermodynamics of Oxides•Computational and experimental thermodynamicsof SOFC materials.
Calculated Zr-O phase diagram
Equilibrium constant for VO•• formation
OOX = VO
•• + 2e’ + 1/2O2
KR = [VO•• ] n2 PO2
0.5/[OOX]
KR
FUNDAMENTAL QUANTITATIVE DEFECT CONSTANTS
UF-DOE HiTEC
Computational Materials Thrust•Large-scale molecular dynamics simulations to elucidate the effects ionic radius and polarizability of on ionic conductivity, the structure of vacancy clusters, and the mechanisms of oxygen transport.•First principles, electronic structure simulations. Calculation of defect formation energy in oxides from first principles and thermodynamics. Study of oxygen reactions at surfaces and interfaces.
Ni-GDC GDC LSCF
QuickTime™ and aTIFF (LZW) decompressor
are needed to see this picture.
Ebond = A
r m − Br n
A, B, n and m are constants
θ = ?
FUNDAMENTAL QUANTITATIVE DEFECT CONSTANTS
Ni-GDC GDC LSCF
Ab-initio calculation of ZrO2 grain boundary and comparison with Z-contrast TEM image
Defect formation energiesas a function of PO2
QuickTime™ and aTIFF (LZW) decompressor
are needed to see this picture.
VO••
VO•
VOx
Computational Materials Thrust•Large-scale molecular dynamics simulations to elucidate the effects ionic radius and polarizability of on ionic conductivity, the structure of vacancy clusters, and the mechanisms of oxygen transport.•First principles, electronic structure simulations. Calculation of defect formation energy in oxides from first principles and thermodynamics. Study of oxygen reactions at surfaces and interfaces.
UF-DOE HiTEC
θ = ?
log
[ i ]
(m
-3)
log PO2 (atm.)
High-PO2 (model)
FUNDAMENTAL PROPERTIES
Brouwer Region Ice = 2cVO
Brouwer Region IIacD = 2cVO
Brouwer Region IIbcD = ch
Brouwer Region III3cVA
+ 3cVB = ch
′ ′ ′ V A
′ e
VO••
′ M A
h •
Intermediate-PO2 (model)
Low-PO2 (model)
′ ′ ′ V B
20 15 10 5 0 515
10
5
0
Same approach being applied to perovskites• Defect equilibria already developed
DEFECT EQUILIBRIADEFECT STRUCTURE
LaFeO3: Rhombohedrally distorted perovskite
• Structural optimization in progress
UF-DOE HiTEC
Calculated Lattice Constants
a= 3.7822 Å c= 3.6493 Å
Cutoff Energy : 500 eVExchange-Correlation approximation: LDAK-POINT spacing: 2X2X2
CATHODE DEVELOPMENTOptimize Microstructure for:
• Activation Polarization– Electrocatalytic Activity
• Increase specific catalytic activity• Increase TPB• Dispersed catalyst
• Ohmic Polarization– Electronic vs. Ionic Transport
• Electronic conduction path• Ionic conduction path
• Concentration Polarization– Gas transport
• Graded porosity • Gas vs. solid state transport
OO× ↔ VO
•• + 2 ′ e + 12
O2
• Temperature programmed reaction (TPR) – Ramp temperature in reacting gas mixture to determine catalytic activity and
selectivity• Temperature programmed desorption (TPD)
– Ramp temperature in He to determine adsorbed species
CATHODE DEVELOPMENT - Electrocatalytic Activity
0
10
20
30
40
50
60
0 100 200 300 400 500 600 700 800
Con
cent
ratio
n (p
pm)
Temperature ( oC)
O2
NO
La2CuO
4 TPD
500 ppm NO and 1% O
2 adsorption
Lattice-O
Surface-O
NO used as probe molecule:
NO = 1/2N2 + 1/2O2vs.
O2 = O2
Set up for 18O2 probe moleculewith 16O oxide catalyst:
18O2 = 18O2 - phys-adsorbed
18O2 = 18O16O - scrambled product
18O2 = 16O2 - lattice oxygen
Kinetics of scrambled productformation indicative of chargetransfer reaction and surface exchange coefficient, ko
CATHODE DEVELOPMENT - Electrocatalytic Activity
0
0.002
0.004
0.006
0.008
0.010
0.012
0 200 400 600 800TEMPERATURE (°C)
La2CuO
4-δ
LaFeO3-δ
LaCrO3-δ
LaMnO3-δ
LaCoO3-δLaNiO
3-δ
TPR of NO over partially reduced LaBO3-δ
Effect of B-site transition metal on catalytic activity:• Cations with partially filled d-orbitals (Co, Ni) more active
TPR of NO over La1-xSrxCo1-y(Ru/Fe)yO3-δ
• Ru most active
CATHODE DEVELOPMENT - Electrocatalytic Activity
• Potential programmed reaction (PPR) – Ramp voltage in reacting gas mixture to determine catalytic activity and selectivity
• Potential programmed desorption (PPD) – Ramp voltage in He to determine adsorbed species
Electrodes
Current/Voltage
Capillary to Mass Spec
UF-DOE HiTEC
YSZ
CATHODE DEVELOPMENT - Electrocatalytic Activity
16O2
18O2
Electrodes
Current/Voltage
Capillary to Mass Spec
UF-DOE HiTEC
YSZ
CATHODE DEVELOPMENT - Electrocatalytic Activity
Include:• Electrode structure• Current-voltage behavior: io~ko, k = f(V)
16O2
18O2
OO× ↔ VO
•• + 2 ′ e + 12
O2
18O2
16O2
18O2
16O2
KR = ko/ko
18O16O16O2
I (A)
C. Xia, Y. Zhang, M. Liu, Appl. Phys. Lett., Vol. 82, No. 6, 10 February 2003
(LSM-YSZ)(LSM-GDC-LSCF)
(LSM-GDC)
Benchmark
CATHODE DEVELOPMENT - Electronic vs. Ionic Transport
Ag-ESBsAg-ESBvm
CATHODE DEVELOPMENT - Electronic vs. Ionic Transport
C. Xia, Y. Zhang, M. Liu, Appl. Phys. Lett., Vol. 82, No. 6, 10 February 2003
Relative size of ionic/electronic conducting phase• Same volume fractions• ESB particle size reduction
20
30
40
50
60
70
80
90
100
0 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500
Particle Size (nm)
Cou
nt (%
of h
ighe
st)
Unsieved Sieved Vibratory-milled(7 day)
VMagglomerate
VM Sieved
CATHODE DEVELOPMENT - Electronic vs. Ionic Transport
0.04
0.05
0.06
0.07
0.08
0.09
0.1
0 20 40 60 80 100
Time (h)
ASR
( Ωcm
2 )Stability @ 650˚C (100h) in air
Ag-ESBs
Ag-ESBvm
ASR = 4.10x10-04t + 6.30x10-02
ASR = 1.91x10-05t + 4.78x10-02
CATHODE DEVELOPMENT - Electronic vs. Ionic Transport
Smaller ESB particle size increases stability
However, ASR still increases with time under current due to electromigration of Ag
New Cathode Materials - Pyrochlores
K. S.Lee, J. Solid state Chem. 131 (1997), 405
• High Electrical Conductivity ~103 Scm-1
• Metallic (increases with decreasing temperature)
QuickTime™ and aTIFF (Uncompressed) decompressor
are needed to see this picture.
T, °C
BRO7BRO7-ESB(56/44)
BRO7vm-ESB(44/56) BRO7vm-ESB
(56/44)
CATHODE DEVELOPMENT - Electronic vs. Ionic Transport
C. Xia, Y. Zhang, M. Liu, Appl. Phys. Lett., Vol. 82, No. 6, 10 February 2003
0.1
1
10
100BRO7s-ESBvm
BRO7vm-ESBvm
BRO7vm-ESBs
CATHODE DEVELOPMENT - Electronic vs. Ionic Transport
BRO7s
ESBvm
ESBs
BRO7vm
( ) ⎟⎟⎠
⎞⎜⎜⎝
⎛⋅
ε⋅
τ⋅−−= JLDq
kTPPPPc
c
c
c
4airOatmosatmosinterface eelectrolyt-cathodeO 22exp@,
JLDq
kTPPa
a
a
a ⋅ε
⋅τ
⋅−=2fuelHinterface eelectrolyt-anodeH 22 @,
JLDq
kTPPa
a
a
a ⋅ε
⋅τ
⋅+=2O@fuelHinterface eelectrolyt-anodeOH 22 ,
ANODE
tortuosity
gas diffusivity porosity
J.-W. Kim, A. Virkar, K.-Z. Fung, K. Mehta and S. Singhal, J. Electrochem. Soc., 146 (1999) 69-78S. Chan, K. Khor, Z. Xia, J. Power Sources, 93 (2001) 130
CATHODE
EXTEND MODEL TO INCLUDE MICROSTRUCTURAL EFFECTS
21
2O
2
0 1P
ccc
kTqK
V
eVJ −
=⎟⎠⎞
⎜⎝⎛ η
=exp
DEFECT CONCENTRATION
( ) ⎥⎦⎤
⎢⎣⎡
⎟⎠⎞
⎜⎝⎛ ηα−−−⎟
⎠⎞
⎜⎝⎛ αη= 10 kT
qkTqJJ expexp
ACTIVATION OVERPOTENTIAL
⎟⎟⎠
⎞⎜⎜⎝
⎛Φ−Φ
η−−−
⎟⎟⎠
⎞⎜⎜⎝
⎛Φ−Φ
η−−−
⋅
⎟⎟⎠
⎞⎜⎜⎝
⎛Φ−Φ
η−−
⎟⎟⎠
⎞⎜⎜⎝
⎛Φ−Φ
η−+=−Φ−Φ
extthVe
VAeV
extthVe
VAeV
V
B
extthVe
extthVeV
V
V
V
Bthext
uu
zcuc
uu
zcuc
qzTk
uu
uuz
cc
qzTk
L
L
1
1
1
1
00
lnln
POTENTIAL
•Electrochemical model (with pore diffusion incorporated) matches “Virkar”* model, but with less fitting parameters, (3 vs. 10)
•Fitting parameters: τa/Da (effective tortuosity anode), τc/Dc (effective tortuosity cathode) and io (exchange current density).
YSZ @ 800 °CLSM cathode (air)Ni-YSZ anode (H2/H2O)
*J.-W. Kim, A. Virkar, K.-Z. Fung, K. Mehta and S. Singhal, J. Electrochem. Soc., 146 (1999) 69-78
0
0.2
0.4
0.6
0.8
1
1.2
0 1 2 3 4 5 6
Vol
tage
(V)
Current Density (A/cm2)
τa/Da = τc/Dc = 104 s/m2
io = 0.1 A/cm2
YSZ @ 800 °CLSM cathode (air)Ni-YSZ anode (H2/H2O)
EXTEND MODEL TO INCLUDE MICROSTRUCTURAL EFFECTS
0
0.2
0.4
0.6
0.8
1
0 0.5 1 1.5 2 2.5 3
Vol
tage
(V)
Current Density (A/cm2)
YSZ @ 800 °CLSM cathode (air)Ni-YSZ anode (H2/H2O)
0
0.2
0.4
0.6
0.8
1
1.2
0 1 2 3 4 5 6
Vol
tage
(V)
Current Density (A/cm2)
GDC @ 800 °CLSM cathode (air)Ni-YSZ anode (H2/H2O)
τa/Da = τc/Dc = 104 s/m2
io = 0.1 A/cm2
τa/Da = τc/Dc = 104 s/m2
io = 0.1 A/cm2
I-V & power density curves can also be generated for mixed conducting materials:
•Electrolytes such as GDC (above)-Shows reduction in OCP and current density due to low ti at 800°C
•Cathodes such as LSF and LSCF (near future)
EXTEND MODEL TO INCLUDE MICROSTRUCTURAL EFFECTS
0
0.2
0.4
0.6
0.8
1
0
0.1
0.2
0.3
0.4
0.5
0.6
0 0.5 1 1.5 2 2.5 3
Vol
tage
(V)
Pow
er Density (W
/cm2)
Current Density (A/cm2)
0
0.2
0.4
0.6
0.8
1
0
0.4
0.8
1.2
1.6
0 5 10 15 20
Vol
tage
(V)
Pow
er Density (W
/cm2)
Current Density (A/cm2)
τa/Da = τc/Dc = 102 s/m2
io = 0.1 A/cm2τa/Da = τc/Dc = 104 s/m2
io = 0.1 A/cm2
Model shows decrease in effective tortuosity (τ/D) can dramatically increases power density
GDC @ 800 °CLSM cathode (air)Ni-YSZ anode (H2/H2O)
EXTEND MODEL TO INCLUDE MICROSTRUCTURAL EFFECTS
10-1
100
101
102
103
104
105
10-1 101 103 105 107
300 oC400 oC500 oC600 oC
700 oC800 oC900 oC
-Z''
/ ohm
s
Frequency / Hz
2.0 X 10-5 atm
Artifacts minimized by nulling
Oxygen diffusion through through porous cathode (5.9 s)
Ionic conductivity through electrolyte grain boundary
Ionic conductivity in the bulk electrolyte
DECONVOLUTION OF CATHODE MECHANISM
Dissociation and surface diffusion of O-species on LSM to TPB (0.18 s)
Oxygen exchange at TPB (0.0001 s)
LSM/YSZ
Step # Process identity “x” inR ∝ pO2
XEa (eV) τ (s)
at 800 °C1
2
3
4
5
Ionic diffusion through electrolyte bulk.
~0Independent
1.101.05d
-
Ionic diffusion across electrolyte grain boundary.
~0Independent
1.041.16d
-
Migration and incorporation of O2- from TPB into YSZ.
~0Independenta
0.971.10a
1.13b
8.5 X 10-5
1.6 X 10-5a
Dissociation and surface diffusion of O-species on LSM.
-0.15-0.268a,c
1.21.61a
1.69b
0.180.016a
Gas diffusion through porous electrode.
-1.1-1.02a,c
∼0∼0a
5.90.16a
a) X. J. Chen et al. / Journal of Power Sources 123 (2003) 17b) Jiang et al. / J. of Electrochemical Society 147 (2000) 3195
c) Kim et al. / Solid State Ionics, 143 (2001) 379d) Guo, Maier / J. of Electrochemical Society 148 (2001) E121
DECONVOLUTION OF CATHODE MECHANISM
0
500
1000
1500
2000
2500
0 500 1000 1500 2000 2500 3000
850 C950 C1100 C
Z' / Ω
450 C
850 °C 1100 °C950 °C
0.1
1
10
100
1000
104
0.1 1 10 100 1000 104 105 106 107
850 C950 C1100 C
Frequency / Hz
450 C
Microstructure/Impedance - LSCF Sintering Temperature Effect
5 µm 5 µm 5 µm
-Z”/Ω
-Z”/Ω
Highest temperature coarsens microstructure - multiple changes to impedancePowder supplied by NexTech
DECONVOLUTION OF CATHODE MECHANISM
z
50 nm
x UF-DOE HiTEC
y
LSM
YSZ
Focused Ion Beam•Enables 3-D analysis of electrode microstructure
- Particle-size, pore-size, & distribution- Triple-phase boundary density- Tortuosity
QUANTIFYING MICROSTRUCTURE
Measure grain to interface line distance
LSM-YSZ Grain Interface
Pore-YSZ Interface
Sample shown is screen-printed LSM sintered at 1350° C
LSM
YSZ13 um
Total Grain Interface: 8.1 umTotal Pore Interface: 4.9 um
UF-DOE HiTEC
QUANTIFYING MICROSTRUCTUREAREA FRACTION
150 nm150 nm
150 nm 150 nm
By combining consecutive line analysis, area density analysis is achieved.
Percent of grain density per total area is current approach.
Example: 67% grain-to-interface density
Lg/Total: 64% Lg/Total: 71 %
Lg/Total: x % Lg/Total: x%
Lg/Total: x %
UF-DOE HiTEC
QUANTIFYING MICROSTRUCTUREAREA FRACTION
13 um
8 um
LSM
YSZTotal Grain Interface: 8.1 umTotal Pore Interface: 4.9 um
Measure grain and pore area
By analyzing over multiple evenly spaced slices, volume fraction can be determined.
Ex.: 89% Area GrainSample shown is
screen-printed LSM sintered at 1350° C
UF-DOE HiTEC
QUANTIFYING MICROSTRUCTUREVOLUME FRACTION
150 nm150 nm
150 nm 150 nm
By combining consecutive area analysis, a volume density analysis can be achieved.
Lg/Total: 71 % Lg/Total: x %
Lg/Total: x% Lg/Total: x %
UF-DOE HiTEC
QUANTIFYING MICROSTRUCTUREVOLUME FRACTION
Calculate triple phase boundary density
Triple-phase points
Sample shown is screen-printed LSM sintered at 1350° C
LSM
YSZ13 um
Triple Phase Points: 9
UF-DOE HiTEC
QUANTIFYING MICROSTRUCTURETRIPLE PHASE LINEDENSITY
150 nm150 nm
150 nm150 nm
By connecting all of the triple phase points, the interface lines can be determined in the sample.
LSM
YSZ13 um
Triple Phase Points: 9
UF-DOE HiTEC
QUANTIFYING MICROSTRUCTURETRIPLE PHASE LINEDENSITY
YSZ
LSM
LTPB
GDC
LSM LSCF
YSZ or GDC
ATPB
Butler-Volmer Equation: J = J0[exp(qαηact/kT) − exp(-q(1-α)ηact/kT)]
J0 = j0 x ATPB = LTPB x wTPB
LSM/YSZ wTPB ≈ Debye length
QUANTIFYING MICROSTRUCTURE
LSM/GDC, LSCF/YSZ, LSCF/GDC wTPB > Debye length
wTPB ≡ f(geometry, contact area, material property)
UF-DOE HiTEC
UF-DOE HiTEC
QuickTime™ and a decompressor
are needed to see this picture.
LSM (Nextech) on YSZ
Consecutive 50nm slices
QUANTIFYING MICROSTRUCTURE
Tortuosity
τ = zpath/zthickness
ACKNOWLEDGEMENTCollaborating Faculty:Dr. Kevin Jones - FIB/SEM CharacterizationDr.’s Susan Sinnott & Simon Philpott - Computational MaterialsDr. Fereshteh Ebrahimi - Mechanical PropertiesDr. Juan Nino - Novel Oxide Materials DevelopmentDr. Wolfgang Sigmund - Novel Synthesis & MicrostructuresDr. Hans Seifert - Materials ThermodynamicsDr. Xin Guo - Nano Ionics and Interfaces
Results by post-docs:Dr. Keith Duncan, Dr. Jiho Yoo & Dr. Heesung YoonResults by graduated students:Dr. Abhishek Jaiswall, Dr. Jun-Young Park, Dr. Jamie Rhodes, Dr. Sun-Ju Song, Dr. Keith Duncan, Terry Clites, Su-Ho Jung, Sai Boyapati, Naixiong JiangResults by current graduate students:Jeremiah Smith, Matthew Camaratta, Sean Bishop, Yanli Wang, Briggs White, Joshua Taylor, Vincenzo Esposito, Chiara Abate, Jin Soo Ahn, Aidhy Dilpuneet, Brian Blackburn, Chin-Tang Hu, Shobit Omar, Eric Armstrong, Martin VanAssche, Cynthia Chao, Eric Macam, Tak-keun Oh, Doh Won Jung, Dan Gostovic, Aijiie Chen, JianlinLi, Chris Woan, Guojing Zhang UF-DOE HiTEC