SECA SOFC Program at GE Global Research

SECA SOFC Program at GE Global Research Matt Alinger and Seth Taylor GE Global Research Niskayuna, NY 8th Annual SECA Workshop and Peer Review Meeting San Antonio, TX August 7-9, 2007 SECA SOFC Program at GE Global Research Highlights • Performed SOFC performance sensitivity analysis on baseline IGFC system. Results indicate 50% HHV efficiency achievable by improving SOFC performance. SOFC requirements that yield 50% efficiency are extremely challenging, but not inherently impossible. Identified component performance requirements that exceed today’s capability. Evaluated cell manufacturing techniques, sintering and air plasma spray, for impact on cell performance and manufacturing cost. Determined, through the manufacturing down-select study, that economic feasibility of SOFC is primarily dependent upon improving long-term stability of cell performance over choice of manufacturing process. Demonstrated effectiveness of Co,Mn spinel coated interconnect with LSM cathodes at reducing degradation rates from ~100 to ~25mΩ-cm2/1000h. Demonstrated Co,Mn spinel coated interconnect with LSCF cathodes is effective at reducing degradation rates. Validated effectiveness of Co,Mn spinel interconnect coating at impeding Cr bulk diffusion. Identified ‘free’ silicon in interconnect alloy as likely contributor to high performance degradation. 2 SECA Coal Based Systems 8th SECA Annual Workshop August 2007 • • • • • • • SECA Coal Based System Program Overview Team: GE Global Research University of South Carolina Pacific Northwest National Laboratory Program Objective – Identify significant barriers to feasibility and to develop solutions to enable high performing, cost-effective solid oxide fuel cells (SOFCs). – Develop and optimize a design of a large-scale (>100 MW) integrated gasification fuel cell (IGFC) power plant incorporating a SOFC and a gas turbine (GT) in a hybrid system that will produce electrical power from coal. The system will be: – Highly efficient (>50% HHV), – Environmentally friendly (90% CO2separation), and – Cost-effective ($400/kW projected factory cost, exclusive of coal gasification andCO2separation subsystems). 3 SECA Coal Based Systems 8th SECA Annual Workshop August 2007 Presentation Outline • IGFC system analysis • IGFC technology gap analysis • Manufacturing down-select study • Degradation testing 4 SECA Coal Based Systems 8th SECA Annual Workshop August 2007 IGFC System Study 5 SECA Coal Based Systems 8th SECA Annual Workshop August 2007 IGFC System performance DOE Requirements End Date Fuel Cost (Power Blocks) Efficiency (Coal HHV) CO2 Isolated Validation Test (hours) Degradation (/1000h) Phase I Phase II Phase III FY2008 Fy2010 FY2015 Coal-Derived Hydrogen or Syngas $600/kW $400/kW $400/kW 40% 45% 50% 90% 90% 90% 1,500 1,500 >25,000 <2.0% <0.2% <4.0% 6 SECA Coal Based Systems 8th SECA Annual Workshop August 2007 IGFC System performance Coal Feed, HHV Total Gross Generated Power Total Parasitic Power Net System Power System Efficiency Power Summary, MW Baseline System with Baseline System 'Super' SOFC 1047.1 1047.1 542.5 592.9 71.9 69.7 470.6 523.2 44.9% 50.0% Pressurized System 1047.1 585.8 64.9 520.9 49.7% *Note that all cases shown include 90+% CO2 isolation, as required. • Baseline system (SOFC + HRSG/ST) – Efficiency of only 44.9% at these conditions. – Performance adequate for Phase I and Phase II • “Super” SOFC (SOFC + HRSG/ST) – Target achieved by increasing the SOFC performance requirements • Pressurized System (SOFC+ST+GT – 15atm) – Baseline stack - capable of achieving the 50% HHV efficiency target. 7 SECA Coal Based Systems 8th SECA Annual Workshop August 2007 IGFC technology gap analysis • Coal – DOE Minimum Requirement (high-rank bituminous coal - Pittsburgh No. 8) – Lower-rank coals result in a lower system efficiency – Factor to be considered • Gasifier – Oxygen from ASU for gasification is significant efficiency driver – Assumption of ~10% improvement over current requires advancement in gasifier design / slurry mixing – High technology risk • Syngas Coolers – Conventional RSC produce saturated steam: Texit = ~650oF – System in analysis - RSC generates superheated steam: Texit = 850oF – Modification not major gap but, Higher Top = materials change / cost challenge – Moderate technology risk • High Temperature CO Shift – Current shift reactors operate with excess steam (avoid C-containing byproducts) – Analysis assume no byproducts produced despite steam/carbon near equilibrium – Capability requires major advances in catalyst or change to new shift methods – High technology risk 8 SECA Coal Based Systems 8th SECA Annual Workshop August 2007 IGFC technology gap analysis • SOFC – Majority of gap between current technology and 50% efficient IGFC systems – >0.5 W/cm2 required for economically viable IGFC systems – Cell voltage and fuel utilization requirements extremely challenging – Methods of controlling degradation at T >800oC must be developed – Achieving high UF in large stack of 100+ cells is a major engineering challenge – Risk of achieving SOFC performance targets extremely high • SOFC Recycle – IGFC design ~50% recycle of the SOFC air – Recycle fraction huge driver on efficiency (reduce fresh air flow requirement) – Blowers for 800+oC do not exist at present and will need development – Largely reliability and cost challenge as opposed to a technology challenge – Reliability and cost risks significant 9 SECA Coal Based Systems 8th SECA Annual Workshop August 2007 Manufacturing down-select 10 SECA Coal Based Systems 8th SECA Annual Workshop August 2007 Manufacturing down-select Process • Detailed Data Gathered by Entire SOFC Team • SOFC Team Risk Sensing Sessions Input • Independent Team Review • Scorecard • Greatest Concern • 4 Categories on Sinter vs. Deposition • Risks of Technology Elements • Independent Assessment 11 SECA Coal Based Systems 8th SECA Annual Workshop August 2007 Technical Team Review Conclusions • No meaningful difference in perceived success between cell manufacturing technologies • Material cost and degradation solution are keys to success • Viability of technology elements are greater challenge than manufacturing process 12 SECA Coal Based Systems 8th SECA Annual Workshop August 2007 SOFC degradation 13 SECA Coal Based Systems 8th SECA Annual Workshop August 2007 SOFC degradation - materials focus Mechanism Air Interconnect – GE-13L • Chromia scale resistance (Cr2O3) • Insulating scale growth (SiO2) Air Interconnect IC coating Cathode Bondpaste Cathode Electrolyte Anode Support Cathode bondpaste - LSC / coating interface Ohmic resistance • SrCrO4 layer resistance • ‘other’ reaction phases Ohmic resistance • Mechanical delamination Active area reduction Cathode - LSCF • Chromia poisoning • Coarsening Cathode / electrolyte interface • Zirconate resistive layer • Cathode poisoning • Mechanical delamination Leakage driven degradation • Cathode reduction • Anode oxidation • Combustion thermal Protective Coating – MnCo spinel • Coating resistance (MnCo spinel) Type Ohmic resistance Ohmic resistance Ohmic resistance Electrochem. activity red. Electrochem. activity red. Anode Bondpaste Ohmic resistance Electrochem. activity red. Active area reduction Fuel Interconnect Active area reduction Active area reduction Active area reduction Using a ‘fixed’ materials set: Focus on cathode side, high-impact degradation mechanisms 14 SECA Coal Based Systems 8th SECA Annual Workshop August 2007 Ceramic Test Vehicle – The Browaller* Idealized test fixture (2”x2” active area) – Simulate real SOFC operating conditions – Known ‘boundary conditions’ – High performance (<300 mΩcm2) – High utilization (80% UF) – Monitor fuel and air gases – Interchangeable interconnect – Gold – Ferritic stainless steel Provide confidence and accuracy in degradation measurement 15 SECA Coal Based Systems 8th SECA Annual Workshop August 2007 *after Ken Browall, ret 6/1/07 Browaller Test I,II 800oC, Low Utilization, 64% H2 1.1 2.0 voltage, V Sintered Supercell LSCF cathode LSC Bond paste Au mesh CC power density, W/cm2 1.0 0.9 1.5 1.0 0.8 0.7 0.5 ASR Data Cell I 173 mΩ-cm2 @ 0.7V Cell II 142 mΩ-cm2 @ 0.7V 0.6 0.0 0.5 1.0 1.5 2.0 0.0 2.5 current density, A/cm2 Extrapolated points: Lost cathode current lead Excellent cell performance – equal to buttons Fully sealed – no leakage, cracking 16 SECA Coal Based Systems 8th SECA Annual Workshop August 2007 Electrochemical testing -Button Cells 800oC Galvanostatic LSCF Cathode LSC Bond paste Interconnect – Ferritic SS ‘6-gun’ 17 SECA Coal Based Systems 8th SECA Annual Workshop August 2007 Button cells - Coated Vs Bare GE-13L ferritic stainless steel interconnect (Co,Mn)3O4 spinel coating *Data from I-V curves at 0.7V and 800oC (Mn,Co)3O4 spinel coated samples exhibit lower degradation rate with respect to bare. 18 SECA Coal Based Systems 8th SECA Annual Workshop August 2007 Button cells - Coated Vs Bare GE-13L ferritic stainless steel interconnect (Co,Mn)3O4 spinel coating *Data from I-V curves at 0.7V and 800oC (Mn,Co)3O4 spinel coated samples exhibit lower degradation rate with respect to bare. 19 SECA Coal Based Systems 8th SECA Annual Workshop August 2007 (Mn,Co)3O4 coating - Cr barrier IC 80 Atom % La Fe Mn Co Co1.5Mn 3O4 (Mn,Co)1.5O4 100 Fe Mn Co La Cr LSC 40 2.5 2.0 60 1.5 1.0 20 0.5 0 0 10 20 30 40 50 Microns 60 70 80 90 0 100 (Oxygen, Sr and other IC elements not shown for plot simplification) No measurable Cr found in the LSC Bond Paste after 886h at 800oC 20 SECA Coal Based Systems 8th SECA Annual Workshop August 2007 Atom % Cr Button cells - Coated E-Brite Vs GE-13L (Mn,Co)3O4 spinel coating 1 MnCoO Coated perf plate GE-13L power density, W/cm2 0.8 0.6 0.4 E-Brite 0.2 0 3 2 1 1 3 0 200 400 600 800 1000 time, h GE-13L exhibits higher performance over E-Brite 21 SECA Coal Based Systems 8th SECA Annual Workshop August 2007 Mn,Co Spinel Coated E-Brite Mn1.5Co1.5O4 Mn1.5Co1.5O4 E-Brite Cr Oxide Scale E-Brite Cathode BP, Spinel Coating on E-Brite shown after 900C 24h Button Cell Test after 886h at 800 C - 1 A/cm2 22 SECA Coal Based Systems 8th SECA Annual Workshop August 2007 Mn,Co Spinel Coated E-Brite 80 Ebrite 70 Atom % Fe Mn O 60 50 40 30 20 0.5 10 0 -5 0 5 10 15 20 0 25 Co1.5Mn1.5O4 (Mn,Co)3 O4 2.0 O by diff Fe Prebond Mn Prebond Si Prebond Si 3000 h 2.5 1.0 Interface Location of Prebond and 3000 h Sample Using Si Signal microns Increased Si at E-Brite interface after 3000 h test. 23 SECA Coal Based Systems 8th SECA Annual Workshop August 2007 Atom % Si 1.5 E-Brite – Cr2O3 / SiO2 SiO2 chromia grain boundary chromia grain interior Cr Cr2O3 Ebrite 1000 2000 3000 4000 5000 6000 7000 Intensity Si Energy (eV) Significant SiO2 concentrations are observed at Cr2O3 interface on E-Brite. Si content: E-Brite ~0.2wt% GE-13L <0.1wt% 24 SECA Coal Based Systems 8th SECA Annual Workshop August 2007 Ohmic losses I (off-line testing) -Contact resistance testing Electrochemical cell configuration Air Interconnect (GE-13L) Coating (Co1.5Mn1.5O4) Cathode Pre-bond (LSC) Cathode Bond Paste (LSC) Cathode (LSCF) Barrier Layer (GDC) Electrolyte (8YSZ) Anode (Ni,YSZ) Anode bond paste (NiO) Fuel interconnect (GE-13L) Equivalent circuit Interconnect/Coating interface B C RIC RCB C IC C CB C Ca B Ca Coating/Bond paste interface Cathode RCa RΩ Total ohmic due to bulk material resistivity Direct measurement of interface contribution for ASR breakdown 25 SECA Coal Based Systems 8th SECA Annual Workshop August 2007 Ohmic losses V1 (off-line testing) -Contact resistance testing I V3 V4 V2 BP Rcont I 100 ΔASR, mΩ-cm2 GE-13L (Co,Mn)3O4 LSC (Co,Mn)3O4 GE-13L 80 60 40 20 0 -20 0 200 400 600 800 1000 1200 time, h Direct measurement of interface contribution for ASR breakdown 26 SECA Coal Based Systems 8th SECA Annual Workshop August 2007 Bond Paste Conductivity 5 4 Gold LSC Gold 900 800 ASR, mΩ-cm2 700 600 temperature, C 3 2 1 0 500 400 300 200 100 o 0 100 200 300 400 500 600 0 time, h 27 Stable bond paste (no increase in resistance over time) SECA Coal Based Systems 8th SECA Annual Workshop August 2007 Bond Paste Conductivity Activation Energies 100 Gold LSC Gold ASR (mohm.cm2) E a ~ 0.1 eV 10 y = 0.2718e 1.2811x R2 = 0.9977 y = 1.0463e 0.2837x R2 = 0.8348 E a ~ 0.02 eV 1 0 0.5 1 1.5 2 2.5 3 3.5 1000/T(K) Literature values for LSC at 800 C and pO2=1 atm: Bulk electrical conductivity ~ 1585 S/cm Activation energy ~ 0.015 eV in pure O2 – (Excellent agreement) Bond paste resistivity negligible 28 SECA Coal Based Systems 8th SECA Annual Workshop August 2007 Co,Mn coated Vs Bare (Mn,Co)3O4 – spinel coating 35 30 GE-13L (Co,Mn)3O4 LSC (Co,Mn)3O4 GE-13L ΔASR, mΩ-cm2 25 20 15 10 5 0 -5 0 200 400 (MnCo) O , LSC 3 4 LSC PNNL - 441SS, (MnCo)3O4, LSM 600 800 1000 1200 time, h Coating effective for LSC samples Commercial 441SS very promising 29 SECA Coal Based Systems 8th SECA Annual Workshop August 2007 (Mn,Co)3O4 Stability Study – As-Received 350 [b9420.raw] (Mn,Co)3O4 as-rec JCS=70> (Mn,Co)3O4 - (Mn,Co)3O4 080-1540> CoCo2O4 - Cobalt Oxide 300 250 (Mn,Co)3O4 – 2 phase spinel Mn1.5Co1.5O4 (tetrag) + MnCo2O4 (cubic) Intensity(Counts) 200 150 100 50 0 20 30 40 50 60 70 80 Two-Theta (deg) Sample shows spinel and tetragonal phases present initially. 30 SECA Coal Based Systems 8th SECA Annual Workshop August 2007 (Mn,Co)3O4 Stability Study – 500h, 800oC 600 [b9483.raw] (Mn,Co)3O4 800/500h Mn - MnCo2O4 [Wt%=53.9 (1.1)] (Mn - (Mn,Co)3O4 [Wt%=46.1 (1.2)] 500 (Mn,Co)3O4 – 2 phase spinel Mn1.5Co1.5O4 (tetrag) + MnCo2O4 (cubic) Intensity(Counts) 400 300 200 100 0 20 30 40 50 60 70 80 Two-Theta (deg) Sample shows both spinel and tetragonal present in near 50-50 wt% mixture. 31 SECA Coal Based Systems 8th SECA Annual Workshop August 2007 • Performed SOFC performance sensitivity analysis on baseline IGFC system – 50% HHV efficiency achievable by improving SOFC performance – SOFC requirements for 50% efficiency are challenging, but not impossible • Identified component performance requirements exceeding current capability • Evaluated cell manufacturing techniques, sintering and air plasma spray, for impact on cell performance and manufacturing cost – Determined economic feasibility of SOFC primarily dependent on improving long-term stability of cell performance over choice of manufacturing process • Demonstrated Co,Mn spinel coated interconnect with LSCF cathodes is effective at reducing degradation rates • Validated effectiveness of Co,Mn spinel interconnect coating at impeding Cr bulk diffusion • (Mn,Co)3O4 coating effective at reducing degradation rate in LSCF SOFCs • Free silicon in interconnect alloy results in detrimental SiO2 at IC/Cr2O3 interface 32 SECA Coal Based Systems 8th SECA Annual Workshop August 2007 Summary Acknowledgements • Travis Shultz, Wayne Surdoval, Lane Wilson of DOE/NETL • GE SOFC Team • The material presented was prepared with the support of the U.S. Department of Energy, under Award No. DE-FC26-05NT42614. However, any opinions, findings, conclusions, or recommendations expressed herein are those of the authors and do not necessarily reflect the views of the DOE. 33 SECA Coal Based Systems 8th SECA Annual Workshop August 2007