1. RelHy International Workshop on
High Temperature Electrolysis Limiting Factors
June 09-10, 2009
Karlsruhe, Germany
S. Elangovan, J. Hartvigsen,
Feng Zhao, Insoo Bay, and Dennis Larsen
Office of Naval Research Contract: N00014-08-C-0680
DOE subcontract through Idaho National Lab.

2. Energy, Environment & Economy
• Environment
– Climate Change
• GHG sources
– 8 tons CO2/kW-yr from coal or oil
– Leaky natural gas pipelines
– Ruminants
• Ozone hole - no, that’s a different topic
– Habitat Impacts One Thing Is Clear
• Drilling in Arctic National Wildlife Refuge
Energy Is The Key To Prosperity
• Wind turbines in Chesapeake Bay
– Air pollution
• Limited Resources
– Oil
• National security
– Gas
• Heating vs. power generation
• Transportation issues
– Renewables

3. Carbon Free Energy Source Options
• Renewable energy resources
– Large Scale Wind
• 800 GW at class 4+ US wind sites
– Small Hydro
• 45GW potential, 2000 sites
– Concentrator Photovoltaic
• Land area 12km2 /GW
– Biomass
• Ag/Forestry byproduct
• Carbon neutral cycle assuming production
and processing are carbon free
• Nuclear
– 25 new plants announced
– Increased output of existing units
• Note trend in figure since 1970
• Hydrogen production from electrolysis
• High temperature 29 tons/GW-hr
• Conventional 21 tons/GW-hr
http://www.eia.doe.gov/emeu/aer/pdf/pages/sec1_6.pdf

4. Storing Hydrogen With CO2
• Energy Sources
– Wind, Solar (PV & heat), Hydro, Nuclear
• Carbon Sources
– Metallurgical Reduction, Cement Kilns
– Fermentation, Digester gas
– Biomass gasifiers
– Fossil Power Systems
• Conversion Technology
– SOFC electrolyzer, steam+CO2=> syngas
• Products
– SNG, Fischer Tropsch liquids

5. Why Electrolytic Synfuels
• Electrolysis efficiency
• Energy storage density
• Fits in existing distribution infrastructure
• Synergy with intermittent energy sources
• No added CO2 emissions
• Reduced work of compression
• Compatible with existing vehicle fleet
– 20 to 50 year crossover

6. The New Alchemy: C ⇒ Au
• Turning carbon into gold
– Low value carbon
• CO2 -$55/ton (Norway C tax)
• Coal $20-100/ton
• Bitumen ~ $100/ton
– High value carbon
• Natural Gas $444/ton carbon
($7/decatherm)
• Crude Oil $888/ton carbon ($105/bbl)
• Refined fuel (pre-tax) ~$1000/ton carbon

7. Synthetic methane from electrolysis of CO2
Gas analysis in volume %
From From
Electrolysis Methanation
Unit
CH4 CO H2 CH4 CO H2
Test 1 .0 14.3 60.7 42.5 0 13.8
Test 2 .7 18.5 58.0 47.7 0 9.2
Test 3 .3 20.1 63.5 50.0 0 9.4
Test 4 .1 15.8 58.9 42.0 0 4.6
Test 5 .1 15.2 59.5 40.4 0 8.2
FEED TO ELECTROLYSIS IS CO2 + H2O + ELECTRIC ENERGY

8. FT- Liquid Products
• Ceramatec produced catalyst
– FeCuK composition
– 8mm La promoted alumina rings
– Automated in-situ reduction
profile using dewpoint controlled
temperature ramp
– Oil fraction
– Water fraction
235 C Reactor Operation

9. One Technology - Multiple Modes Of Operation
Solid Oxide Stack Module
NG
Biogas
Diesel Fuel Syngas
JP-8
Coal
Electricity CO2 & Steam
+ Electricity
Steam + Hydrogen
Electricity (High Purity)

10. SOFC - SOEC Differences
• Cells tested to date are virtually identical
– Same electrolyte, electrodes, pattern, etc.
• SOEC seals more challenging
– Higher back pressure on seals due to product
collection
– Low molecular weight stream vs. reformate
• Diffusion mechanism more active relative to
hydrodynamic
• Hydrogen permeation in metal icon destabilizes air
side scale

11. Repeat Unit Elements
Ferritic Stainless Steel Separator
Corrugated Ferritic Stainless Steel or High Ni alloy
50 µm
50 µm Cobaltite (current distribution layer)
air
electrode Manganite + Zirconia Composite
electrolyte Sc - ZrO2 (partially stabilized)
Ni + ceria cermet
H
electrode Ni (current distribution layer)
Corrugated Ni flow field on hydrogen side

12. Stack Components

13. SOEC Limiting Factors
• Degradation/Lifetime • Thermodynamics
– Oxygen bond layer stability – Operating Voltage
– Oxygen electrode – Steam Utilization
delamination – Co-electrolysis of CO2
– Electrolyte stability • High Temp Heat Duty
– Chromium migration
– 0-15% of energy input
– Seals
– Wind hydrogen feasible
– Interconnect scale growth &
resistance • Energy Cost
– Electrode microstructure – $50/MW-hr => $1.72/kg H2
• Electrode coarsening • Amortized CAPEX
• Cell Size & Stack Height – 24 month life => $0.46/kg H2
– Very large cells feasible
– Porous metal support

14. SOFC-SOEC Contrasts
in (Potential:Composion) Space
Vtn

15. Energy of fuel-cell vs. electrolysis mode
current density, A/cm2
0.4 0.3 0.2 0.1 0 -0.1 -0.2 -0.3
0.4
reaction
ohmic Stack ASR = 1.25,
net T = 927 C,
fuel cell electrolysis
yH2,i = 0.1,
yH2,o = 0.95
heat flux, W/cm
0.2
2
thermal neutral
voltage −ΔhR
Vtn =
0
2F
open-cell
(1.291 V at 1200 K)
potential |<- 100% Efficiency Range ->|
-0.2
-0.4 -0.6 -0.8 -1 -1.2 -1.4
operating voltage, V

16. Typical SOEC and SOFC Temperature Maps
SOFC ΔT > 90°C
Resistance doubling ~ 67 °C
Thermal expansion issues

17. System Efficiency vs. Steam+CO2 Utilization
INL Process Model shows little need to
operate with utilization much beyond 50%

18. Humidified CO2 Cell V-I Characteristic
CO2ELEC01:800°C
80F H2O, 45CC CO2, 8CC H2
1.60
1.40
y = -0.6907x + 0.8717
2
R = 0.9997
1.20
1.00
Voltage, Volts
0.80
0.60
No reduction in performance compared to cells with steam electrolysis
0.40
0.20
0.00
-0.80 -0.70 -0.60 -0.50 -0.40 -0.30 -0.20 -0.10 0.00 0.10 0.20 0.30
Current Density, A/cm^2

19. Electrolysis Cells at 20cm Scale

20. Half ILS 2x60 Cell Stack Module

21. Pre-ILS Module Test at Ceramatec
• “Half-ILS Module” test at Ceramatec
– Integrated Laboratory Scale (ILS) Demonstration
– 2x60 cell stacks, 10x10 cm cells
– Summer 2006, ~2000 hr operation
– Cells & Stacks same as full ILS modules 2007,2008
– Show performance scales with stack height
– Assess system issues with tall stacks
– Exercise component production capacity
– Probably most extensive post test examination
• Tested component examination
– ANL, MIT, UNLV, Ceramatec

22. Half ILS Initial Performance
– 3.8 kW
– 1,200 Liters/hr
– Electrical Efficiency = 96.4%
– System thermal distribution issues
– 2,000 hrs total operation
– 1,000 hrs on CO2/H2O
• Syngas production sufficient for 100
gallons of FT diesel

23. Half ILS module load history

24. Half-ILS Post-Test Observations
• Electrodes
– Oxygen electrode delamination for 2,000 hr test
• No delamination in short stacks tested for shorter periods (~300 hrs)
– Hydrogen electrode & current distribution layer in good condition
• Electrolyte
– No cracking
– Some cubic to tetragonal/mono-clinic transition noted
• Metal Interconnect Edge Corrosion
– Edge rail coating & elimination of silica in seal eliminated the corrosion seen
in early SOEC stacks
– Cr transport to air electrode bond layer
– Sr migration from air electrode/bond layer
• Gross changes in bond layer chemistry, phase assemblage, conductivity and performance
• Initial Performance Reproducible – short to tall stacks
• Unacceptably High Initial Degradation

25. 720 Cell Full-ILS System at INL
5.7 Nm3/hr - 17.5kW H2 Production

26. Full-ILS Module #3 Post Test Examination
Hydrogen electrode attached,
bond layer separated with icon
Oxygen electrode delamination

27. ILS Module #3 Post Test Examination
It appears a layer deposited
at electrode interface is causing
the delamination
Oxygen electrode and icon
contacting layer (bond layer)

28. ILS Module #3 Post Test Examination
EDS Indicates only ScSz at the
interface. The apparent deposition
layer is a layer of zirconia that
has spalled off. Our hypothesis is
that Mn diffusing in from the
manganite electrode introduces
enough electronic conductivity
to allow oxygen to evolve inside
the electrolyte, build up pressure
and split off a layer near the
electrode.

29. ILS Module #3 Post Test Examination
Electrode section in following EDS Maps
Manganite-Zirconia Composite
Manganite Electrode
Cobaltite (LSCo) icon
contacting layer (bond layer)

30. Co-Mn Inter-diffusion in Oxygen Electrode

31. SOFC Button Cell (early - mid ‘90s)
Fully stabilized custom electrolyte from undoped zirconia
LaSrMnO3; Ni-YSZ

32. SOFC Button Cell (2002)
YSZ electrolyte; Ni-ceria anode; modified manganite

33. SOFC Stack (1996)
YSZ electrolyte; Ni-ceria; Modified manganite;
ceramic interconnect (doped lanthanum chromite)

34. Prioritized Areas of Focus
• O2 Electrode Delamination – inherent to SOEC operation?
– Electronic conductivity due to Mn diffusion into electrolyte
• Barrier layer & alternative compositions
• O2 Electrode Bond Layer Stability
– Sintering-shrinkage, phase change, Cr tol, stoichiometry changes
• O2 Electrode Chromium Poisoning
– Solid state diffusion by interconnect contact
• CrMn & CuMn spinel coatings
– Evaporation/Condensation from interconnect, manifold, piping
• Electrolyte Stability
– Minimum dopant level (strength trade-off)
– How much degradation is allowable
• System BOP Reliability
• Seal Stability

35. Electrolyte Composition Effect

36. Fully Stabilized (FS-ScSZ) Cell Performance
30% Reduction in initial ASR
vs. PS-ScSZ

37. Initial O2 Electrode Microstructure

38. Initial O2 Electrode Microstructure

39. Initial H2 Electrode Microstructure

40. Initial H2 Electrode Microstructure

41. Hydrogen Electrode & Bond Layer
Adherent and conductive hydrogen electrode
Most areas of bond layer separated with flowfield

42. Oxygen Electrode and Bond Layer
Extensive delamination of standard manganite Perovskite electrodes
No delamination of new cobalt-ferrite Perovskite electrodes

43. O2 Electrode Comparison Stack

44. New O2 Electrode Improves Stack Stability

45. Metal Interconnect Contact Resistance

46. High Temperature Electrolysis
• Leverage decades of SOFC R&D
• Inputs
– e- (green electrons)
– steam => hydrogen
– co-electrolysis of H2O + CO2 => syngas
– heat input optional, depends on operating point
• Most efficiency means of hydrogen production
– e- to hydrogen
• η=100% at 1.285V
• η= 95% at 1.35V
• η=107% at 1.20V, (heat required)
• Hot O2 and steam byproduct
– Valuable for biomass gasification