Solid Oxide Fuel Cells
The Power for the 21st Century
Meilin Liu
Center for Innovative Fuel Cell and Battery Technologies School of Materials Science and Engineering Georgia Institute of Technology Atlanta, GA 30332-0245
What is a Fuel Cell?
Electricity
Hydrogen Fossil Fuels (Gas, oil, coal)
Why Fuel Cells?
From: McDermott
TG FC
DG DG TG
FC
SSFC: Ship Service Fuel Cell Program; SSDG: Ship Service Diesel Generator
GTG: Gas Turbne Generator
Motives for Fuel Cells
• High efficiency • Low emissions (no SOx or NOx, half CO2) • Low operating costs & low maintenance • Quiet • Distributed power generation (avoid power outages) • Stricter transportation emission standards
Applications of Fuel Cells
Stationary power Hybrid propulsion Compact power systems Distributed power systems Portable power systems Microturbine combined cycles Naval power systems
Naval Applications
Driving Force
• US Department of Energy
$70 million for SOFC research 2002-2003 $140 million supporting hydrogen production and stationary PEM fuel cell technologies
•
FreedomCar initiative
To promote development of hydrogen as a primary fuel for automobiles Will spend $150 million on PEMFC technologies in 2002
• •
Private auto and energy companies European, Japanese, and Chinese governments
US DOE Performance Targets
• 2nd Generation Efficiency: 50-60% LHV Cost: $1,000-$1,500/kW Year: 2003 • 21st Century System Efficiency: 70-80% LHV Cost: $400/kW Year: 2015
Cost and Reliability Issues
• • • • • Moving parts (pumps, compressors) unable to perform for thousands of continuous hours Stationary fuel cells must operate continuously for 40,000 hours Automotive fuel cells require 5000-hour lifetime Catalyst costs (platinum, ruthenium, palladium) For automotive applications fueling infrastructure is an issue
Proton Exchange Membrane Fuel Cells
e’ Fuel in H2
H+
Load Oxidant in O2
H2
H 2O
Depleted fuel
Anode
Proton Exchange Membrane Cathode
Depleted oxidant
Proton Exchange Membrane (PEMFC)
• • • • • • • • • Specialty power (small stationary) Buses/Vehicles Gemini space program Dow Chemical (MI, AR) Electrochem (MA) Energy Partners (FL) H-Power (NJ, CA) UTC Technologies Corp.(CT) 1000 PEMFCs (each averaging 5 kW per unit) were put into service on an experimental basis
PEMFC
• • • Solid state (No liquid to circulate) Flexible membrane won’t leak or crack Produce high power density at temperatures below 100°C; allows fast start-ups and immediate response to changes in power demand
Fuels must be purified to avoid catalyst poisoning by CO and sulfur • Platinum accounts for 20% of total costs of a 50-kW PEMFC system for vehicles – 200 grams of Pt required based on a loading of 0.5 mg/cm2 in the MEA
•
PEMFC
• Goal is to bring system cost down to $2500 per vehicle; current cost is about $14,000 per fuel cell engine Stationary PEMFCs cost more than $3600 per KW installed; must bring system cost down to $1200-$1500 per kilowatt to be competitive
•
Direct Methanol (DMFC)
• • • No need for fuel reformer Operate at 120-190 F 40% efficiencies are expected
anode: CH3OH + H2O → CO2 + 6H+ + 2ecathode: 3/2O2 + 6H+ + 6e- → 3H2O overall: CH3OH + 3/2O2 → CO2 + 2H2O
Challenges & Opportunities
Catalysts insensitive to contaminants in the fuel such as CO; Novel membranes with minimal MeOH across-over
and high H+ conductivity at 120-180°C ;
Efficient catalysts that promote a high rate of oxygen reduction; Alternative catalysts less expensive than Pt/Ru to reduce the cost.
Solid Oxide (SOFC)
e’ Fuel in H2 H2
Hydrocarbon H O 2 Fuels
Oxidant in O2
H+ Anion conductor
H 2O O2
Electrolyte
Depleted fuel
Anode
(Ionic conductor) Cathode
Depleted oxidant
CO, H2+O2→CO2, H2O+2e-
4e- + O2 → 2O2-
SOFC Materials
• Electrolyte: YSZ, GDC, LSGM, SrCeO3 • Anode: Porous Ni-ZrO2 or Cu-YSZ cermet • Cathode: LSM/YSZ, LSCF/GDC, SSC
Why SOFC ?
The cleanest, most efficient & versatile system for chemical to electrical energy conversion
50% electrical, 85% overall
Advantages of SOFCs
Efficient – 40-60% efficiency in individual electric systems and up to 80% in hybrid systems Environmental – Reduces global warming and air pollution Fuel Flexible – Uses available liquid fuels such as gasoline and diesel as well as natural gas and propane Fuel Extending – Extends the use of variety of fossil fuels, including vast domestic coal reserves
Modular: as the basic building block for multiple applications
Distributive Energy – Brings electricity to remote locations where no transmission exists Domestic Security – Reduces dependence on foreign oil
SECA
Applications
Stationary – SOFCs will efficiently provide clean, economical electricity either in urban settings, or in remote locations for homes, hospitals, farms, businesses, or recreation facilities. Transportation – SOFCs are able to work with all standard transportation fuels to provide auxiliary power (AP) for trucks and other vehicles. Military – Fuel cells are attractive for military use because they represent quiet, clean, uninterruptible energy that can be delivered at the point of power application. SECA
Honeywell SOFC System Concept
20 IN.
36 IN.
15 IN.
McDermott's
2-kW Technology Demonstration Unit
Delphi/Battelle’SOFC APU
SOFC system as an AUXILIARY POWER UNIT (APU)
Markets: stationary, passenger automobiles, trucks, recreational vehicles, military
12V Alternator 12V-24V trailer converter Refrigeration Unit CAB LIGHTS CB RADIO AM/FM HEATER FRIDGE SATELLITE Refrigeration 12V Alternator Battery Isolator
Refrigeration Unit Lift-gate Battery Battery
Low Voltage Disconnect
Voltage trailer lights, ABS,.. conditioner for
Delphi Gasoline APU for passenger automobiles
Diesel Truck APU
Stationary applications Natural gas or diesel power generators
Delphi Next Generation APU System
Proof of Concept (PoC) APU
Mass reduction: PoC: 200 + kg Next Generation: 50 kg
Next Generation APU Mock-up
Size reduction: PoC: 152 Liters Next Generation: 50 Liters
SOFC Obstacles
• Electrolyte conductive only at high temp • Expensive alloys needed to house the cell • High temperature causes thermal stresses in ceramic structures • Too expensive: $4000-6000/kW
Challenge: Cost Reduction
• Materials selection: inexpensive materials • Fabrication processes: simple & cost-effective
Lower Operating Temp → Material Cost
• Advantages
– Inexpensive metallic components may be used for interconnect, heat exchanges, and other components – Greater system reliability & longer optional life – Potential for mobile applications
• Challenges
– Conductive electrolytes – Catalytically active electrodes – Macro- & meso-porous electrodes/interfaces
GT – FC/BT
Cost-Effective Fabrication→Manufacturing Cost • Fabrication Techniques
– Screen Printing – Dry Pressing – Co-Extrusion
• Advantages
– Simple, inexpensive, reproducible
• Challenges
– How to retain competitive performance
GT – FC/BT
SECA Development: Progressive Applications
2005
2015
• $800/kW • Prototype ($Unit) 3 - 10 kW
2010
• $400/kW • Commercial
• Vision 21 Power Plants 75% efficient plants • Propulsion <$200?/kW
Solid Oxide Fuel Cells Fabricated by Screen-Printing & Dry-Pressing
Characteristics of GDC Powder by GNP Large surface area
Compositional homogeneity
4µm
b
Loose agglomerates Foam-like structure Fill density 0.059 g/cm3 120th of theoretical value Easy to densify 92% at 1250oC/5 hrs 95% at 1350oC/5 hrs
1µm
Dry Pressing of GNP Powder
The thickness of the loose powder is about 120 times that of the dense film 2400 µm 2.4 mm ~ 0.1” 20 µm
Dense Film
Loose Powder
The thinnest: 8 µm
Microstructures of Dry-Pressed Films
10µm 30µm
GDC film
cathode
electrolyte
Substrate
anode
~8 µm
~15 µm
Cross-Sectional View of a Single Cell
A single cell
cathode electrolyte
Porous SSC and 10 v%SDC Cathode
anode 30µm
Dense SDC
2µm
Porous Ni-SDC Anode
2µm
2µm
Changrong Xia, Fanglin Chen and Meilin Liu, Electrochemical and Solid State letters, 4(5) A52-A54 (2001).
Performance of SOFCs
2.0
-2
Power density, Wcm
1.8 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0
SDC (30 µm), Xia, dry pressing YSZ (5-10 µ m), Tsai YSZ (8 µ m), YDC interlayer, Ghosh YSZ (10 µ m), Kim YSZ (9 µ m), Visco
LBL Utah NW ANL
Dry pressing
GT
400
500
600
700
800
Temperature, °C
Significance of Interfacial Resistances
Im Z, Ωcm
0.4 0.2 0.0 0.0 0.2 0.4 0.6 0.8 1.0 1.2
2
2
a 550 C
o
500 C
o
600 C
1.4 1.6 1.8 2.0
o
10
Re Z, Ω cm
Resistance, Ω cm2
8
Interfacial Resistance
6
4
Performance is determined by Rp at low temperatures!
Electrolyte, 30µm
2
0 400 450 500 550 600
Temperature, °C
Modeling and Design of Porous Mixed-Conducting Electrodes
Supported by
National Science Foundation
GT – FC/BT
SOFC: Key Components
FUEL: H2 or Propane
e-
anode electrolyte cathode
• H 2 +O X → H 2O + VO • + 2e′ O
• VO •
Load
1 O 2 + VO•• + 2 e ′ → O X O 2
OXIDANT: air or O2
e-
Active sites for Electrochemical reactions
Metallic Electrode: TPB
2 e′( electrode ) + V (
•• O electrolyt e
1 X ) + O2 ( gas ) → OO 2
O2 (gas)
Metallic Electrode
2eElectrolyte Vo..
Reaction Rate
Active sites for Electrochemical reactions
MIEC Electrode: Solid/Gas Interface
1 .. X O2 ( gas ) + {VO + 2 e′}( MIEC ) → OO 2
2eMIEC Vo Electrolyte Vo..
1 • X ′( electrode ) + VO • ( electrolyt e ) + O2 ( gas ) → OO 2e 2
O2
Reaction Rate
Elementary Steps
2 e′( electrode ) + VO•• ( electrolyt e ) + 1 X O2 ( gas ) → OO 2
Porous MIEC Electrode
Ionization of Oad Adsorption O2 Diffusion
Electrolyte
eO2Ionic and Electronic Transport
Oad Surface diffusion
Electrolyte
Modeling of Porous MIEC Electrodes
• In the Solid MIEC
uk J k = − zk F [(1 − p )ck ](∇µ k + zk F∇φ ) τs
• Through the Pores of MIEC
N O2 u O2 ∇ µ O + v ( pcO ) = − 2 2 τg
• At the MIEC/O2 Interface & TPBs
~ * α ∆ µ p c e J V = J 0 ,V V exp a − O2 RT p * cV O2 ~ α ∆ µ e exp c RT 1 2
Distribution of Reaction Rate
Useful Thickness
CC Porous MIEC El. CC Porous MIEC El.
Jv
TPB1
TPB1
MIEC/O2 MIEC/O2 TPB2
Functionally Graded Electrode
Current collector Layer
Intermediate Layer Catalytic layer Electrolyte
In-situ potential-dependent FTIR Emission Spectroscopy
To understand elementary steps involved in electrode reactions in SOFCs, such as adsorption, dissociation, charge transfer, and mass transfer; To provide surface structural details under conditions for actual fuel cell operation; and To rationalize the pd-FTIRES spectra correlated to electrochemical data with the types of the intermediate species found at the functional interfaces. GT – FC/BT
Experimental Arrangements for
Investigations into SOFC Reactions Using in-situ FTIR-ES, IS, and MS/GC
Process Control System
FTIR Accessory
Mass Spectrometer
Gas Chromatograph
Argon
Fuel/O2
Mass Flow Controllers Drier Oxygen Sensor
To Vent
To Vent
Impedance Spectroscopy Electroanalytical measurements
GT – FC/BT
Optical Configuration
for In-Situ pd-FTIR Emission Spectroscopy
Top: Oxygen Reduction
X 1/ 2O2 +VO•• + 2e− →OO
(1/ 2O2 + 2e− →O2− )
Bottom: Oxygen Evolution
O2− − 2e− →1/ 2O2
FTIR Spectrometer (Liquid-N2-Cooled MCT Detector) PO2
O2−
Thermocouple
Impedance Spect. DC Overpotential PO2
Heating Cartridge
GT – FC/BT
In-Situ Pd-FTIRES Spectra
wavenumber / cm
4000 8 3500 3000 2500 2000
-1
- After local baseline correction
w a v e n u m b e r / c m -1
1200 1400 1300 1100 1000 900
1500
1000
500
6
0 .0
4
In air
(η: 0 to 320 mV)
-0 .1
2
(a )
0
25
20
0
15
In 1%O2
(η: 0 to 760 mV)
-3
10
(b )
5
0
25
0
20
15
In N2
(η: 0 to 690 mV)
− O2
-3
3500 3000 2500 2000 1500 1000 500
10
(c )
5
0 4000
1400
1300
1200
1100
-1
1000
900
wavenumber / cm-1
w avenum ber / cm
Evolution of Oxygen
8
-Im{Z} (Ω⋅cm )
6
- SSC/SDC/SSC, in 1% O at 500°C
2
SSC/SDC/SSC, in 1%O2, at 550 C with DC bias, 0~1.1V with 0.2V increment
4
2
o
7 6 5 4
550 C, with DC bias
nt me re inc V 0. 2
930
o
2
0 0 2 4 6 8
2
10
12
Re{Z} (Ω⋅cm )
∆E/E0(%)
3 2 1 0 -1 -2 -3 -4 -5 4000 3500 3000
0~
,w 1V 1.
ith
1124 1236
OCV
0 ~ -1
. 1 V, w ith -0 .
2V in
crem
ent
1000 500
2500
2000
-1
1500
Wavenumber/cm
Proposed Reaction Mechanism for Oxygen Reduction
Rate-determining step (rds):
PO 2
O2(gas) Oe2− O LC 2− O LC
− = O2,ad + e′ → O2,ad
O2(gas) OO eMIEC Electrode Gas Mixture
O
2− • VO • O LC 2− 2− O LC O LC 2− 2− O LC O LC 2− VO•• O LC
2− VO•• O LC
2− 2− O LC O LC
VO••
O2-
VO••
O2-
VO••
SDC Electrolyte
MIEC Electrode
+
PO 2
e-
e-
+
O2(gas)
O2(gas)
GT – FC/BT
In-situ FTIRES
20
– Anodes in SOFCs
817
H2 as background and CH4 as sample After 5 min in CH4
2143 cm : CO (Adsorbed) 1712 and 1540 cm : Graphite
-1 -1 -1
CH4(gas)
1050
∆E/E0(%)
16 Ni-S DC
1070-800 cm : Metal (Ni or Cu) carbonato (CO3) complexes
An od
e
CH4(gas)
1712 1540
12
2143
Cu-SDC
Anode
1540 1072 807
8 CO2 4 4000
3500
3000
2500
2000
-1
1500
1000
500
Wavenumber/cm
GT – FC/BT
New Electrode Materials for Low-Temp SOFCs
Supported by
Department of Energy
Microstructures and Impedances of SSC-SDC10 Cathodes
fired at different temperatures 900oC 950oC 1000oC
2µm
0.2
2µm
2µm
Im Z, Ωcm
2
0.1
o
0.0 0.0 0.1 0.2
950 C 900oC
0.3 0.4
2
1000 C
0.5 0.6 0.7
o
Re, (Z -Rb), Ωcm
The Best Performance at 400-600°C
0.65 0.60 0.55
-2
Power density, Wcm
0.50 0.45 0.40 0.35 0.30 0.25 0.20 0.15 0.10 0.05 0.00 -0.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6
-2
0.6 W/cm2 at 600°C
W400 W450 W500 W550 W600
1.8
2.0
2.2
2.4
Current density, Acm
The Best Performance of LT-SOFCs
2.0
-2
Power density, Wcm
1.8 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0
GDC (30 µm), Xia, dry pressing SDC (30 µm), Xia, dry pressing YSZ (5-10 µ m), Tsai YSZ (8 µ m), YDC interlayer, Ghosh YSZ (10 µ m), Kim YSZ (9 µ m), Visco
LBL Utah NW ANL
GT GT
400
500
600
700
800
Temperature, °C
Hybrid Metal/Electrolyte Monolithic Solid Oxide Fuel Cells
Supported by
DARPA/DSO - Palm Power
(with J. Cochran, J. Lee, D. McDowell, T. Sanders)
Tubular SOFC
Low Power Density
Planar SOFC
End Plate Anode Electrolyte Cathode Bipolar Separator Repeating unit Anode
Difficulties in manifolding and sealing
SOFC: An Extruded Cell
Hybrid Metal/Electrolyte Monolithic Low Temperature SOFCs
FeNiCoMo LCM
_
Fuel Air Fuel Air
Load
Georgia Institute of Technology
+
Hybrid Extrusion - ”In617”/YSZ
Web = 180 µm
2.5 mm
Note that 40% linear shrinkage occurs on sintering. Webs will be 110 mm.
Lengths can be continuous
GT – FC/BT
Electrolyte honeycomb
Honeycomb fuel cell
Slurry in Slurry out
Drying Sintering
Slurry coating
Projections for Hybrid Metal/Ceria Monolithic SOFC
1. Operation Fuels -- Hydrogen, Natural Gas, Propane, Coal Gas, Methanol, Ethanol, and Reformed Gasoline and Diesel; potentially insensitive to contaminants such as H2S in reformed fuels. Power Density – 1 Watt/cc in Near Future, 5 Watts/cc in 3-4 Years Operating Temperature – 400-6000C Fuel Cell Size – 6 X 6 mm square by 11cm long. (This Is 4 Watts at 1 W/cc and 20 watts at 5 W/cc. (This Is Not Palm Power. It Is Finger Power) Materials – Samaria Doped Ceria (SDC) Solid Electrolyte, CoCr Doped Ni Metal Interconnect, Anode of Porous Ni-SDC, and a Cathode Layer, Consisting of Sm0.5Sr0.5CoO3 and 10wt.% SDC. Catalysts Will be Added as Needed Depending on Fuel. Fuel Cell Cost -- $500/kW in One Year, $50/kW in 3-4 Years
Georgia Institute of Technology
2. 3. 4. 5.
6.
Functionally Graded Electrodes on Honeycomb Cells
GT – FC/BT
Functionally Graded Electrodes
YSZ GDC LSM LSCF
12
1
2 3 4
2
3
4
5
Interfacial resistance (Ωcm )
2
8
4
5
0
650
700
750
o
800
T e m p e ra tu re ( C )
GT – FC/BT
Sample 5: LSM/GDC/LSCF
LSCF LSCF50+GDC50 LSM25+LSCF25+ GDC50 LSM50+GDC50 YSZ GT – FC/BT
Graded Composition
High Electronic Conductivity Interconnect Compatibility Inter-Mixed Layer Highly Catalytic Electrolyte Adhesion Matched Thermal Expansion
Electrolyte
The Interfacial Resistances About 10 Times Smaller
30
Resistance, Ω cm2
25 20 15 10 5 0 550 600 650
Functional graded cathode LSM based cathode YSZ (300 µ m thick)
700
750
800
Temperature, °C
Cathodes for Zirconia Fuel Cells
850 C
10
o o
750 C
o o
650 C
o o
550 C
o o
Cathode for zirconia
700ºC 150ºC
Interfacial resistance, Ωcm2
2
1
0.1
One Order of Magnitude
Graded cathode, Hart, JPS 106(2002)42 LSM-GDC cathode, Murray, SSI 143(2001)265 Graded cathode, cathode YDB cathode
0.85
0.90
0.95
1.00
1.05
1.10
1.15
1.20
1.25
Temperature, 1000/T
GT – FC/BT
Summary - Electrode Development
• Cathodes graded in composition show interfacial resistances about 10 times lower than that of a conventional LSM-YSZ cathode; The performances are dependent on the microstructures, and is improved by low-temperature sintering; Interfacial resistance of graded cathodes as low as 0.47 Ωcm2 was achieved at 750oC. However, it increased to 4.1 Ωcm2 at 600oC; and A new cathode showed much lower interfacial resistances than the graded cathodes, 0.30 Ωcm2 at 600oC, about 10 times better.
600 mW/cm2 at 600°C
• •
•
GT – FC/BT
Other Alternative Advanced Energy Technologies
Future Energy Technology
Renewable, Regenerative Fuel Cells using Solar Energy
Water
Water
Courtesy: Aerovironment
The Cleanest Alternative
Solar Energy
Photoelectrochemical
Cell
H2O
Fuel Cell
H2
Electricity
Concluding Remarks
• Solid Oxide Fuel Cells represent the cleanest, most efficient & versatile system for efficient use of fossil fuels. • Recent advances in SOFC R&D suggest that SOFC technology has the greatest potential to be the primary energy technology for the 21st century.
Challenges & Opportunities
Cathode development
First-principle calculation to predict best materials and structures Rational design of functionally graded electrodes and interfaces
Contaminant-Resistant Anodes and reforming catalysts
Sulfur resistant, Carbon-deposition resistant, Sulfur removal
Cost-effective fabrication processes to dramatically reduce the cost
$4,000→$400/KW
Research Team Members
• • • • • • • Bill Rauch Alan Burke Erik Koep Xinyu Lu Dr. Changrong Xia (MSE) Dr. Siwen Li (MSE) Dr. Jessica Bartling (ChE)
Acknowledgements
NSF DoE/NETL DARPA/DSO-Palm Power ONR Grant N00014-99-1-0353 Center for Innovative Fuel Cell and Battery Technologies, Georgia Tech