In-Situ Diagnostic Methods for SOFC G. Schiller, K.A. Friedrich, M. Lang, P. Metzger, N. Wagner

Transcription

1 In-Situ Diagnostic Methods for SOFC G. Schiller, K.A. Friedrich, M. Lang, P. Metzger, N. Wagner German Aerospace Center (DLR), Institute of Technical Thermodynamics, Pfaffenwaldring 38-48, D Stuttgart, Germany International Symposium on Diagnostic Tools for Fuel Cell Technologies, Trondheim, Norway, June 23-24, 2009

2 Outline Introduction Electrochemical Impedance Spectroscopy on Stacks Spatially Resolved Measurements: Current density Voltage Impedance Temperature Gas Composition Optical Spectroscopy X-Ray Tomography Conclusion

3 Investigation of Degradation and Cell Failures Insufficient understanding of cell degradation and cell failures in SOFC Extensive experimental experience is not generally available which would allow accurate analysis and improvements Long term experiments are demanding and expensive Only few tools and diagnostic methods available for developers due to the restrictions of the elevated temperatures

4 Conventional Test Stand Diagnostics Conventional test stand diagnostics: provide important and essential information about fuel cell performance and behaviour: U(i) characteristics, OCV EIS on single cells Current interrupt methods Performance degradation with time U(t); i(t) Cell voltage distribution U stack = U 1 + U 2 + U 3. Pressure loss / Gas tighness test Gas utilization measurement Temperature distribution and control

5 Sophisticated (non-traditional) in-situ Diagnostics Electrochemical impedance spectroscopy on stacks Spatially resolved measuring techniques for current, voltage, temperature and gas composition Optical imaging Optical spectroscopy Acoustic emission detection X-ray tomography

6 Challenges for EIS for Stack Investigations Large areas (e.g. 100 cm 2 ) lead to high current and low impedances of about 1 mohm. Electrochemical processes appear at high frequencies (up to 100 khz) due to the high reaction rates at high temperatures. Stacks generally contain metallic components leading to high frequency disturbances. Contacting of all cells and sensing in specific cells does not account for the voltage distribution in the stack. The sensor wires are at high temperatures: an optimization of the measurement system is not possible during operation. Strong overlapping of electrode processes; evaluation with equivalent circuits can be inaccurate. For system with current > 40 A no commercial equipment available.

7 Mitigation of EIS Problems Reduction of the high frequency disturbances by optimization of the wiring of the electrical sensing of the SOFC stack. Variation of the operating conditions (gases, temperature) in order to determine the different impedances of the electrode processes Modeling of the spectra by an equivalent circuit. Development of advanced EIS equipment for high currents / high frequencies in corporation with instrument manufacturer (Zahner Elektrik GmbH).

8 Experimental Set-up for EIS Measurements of Stacks at DLR

9 Performance of the 5-Cell Short Stack at 750 C (5 H 2 +5 N 2 +3%H 2 O / 20 air (SLPM), 94 h) Cell 1-4 : 1.10 V Cell 5 : 1.05 V cell voltage U [V] 1,2 1,0 0,8 0,6 0,4 0,2 cell 5 (top) cell 4 cell 3 cell 2 cell 1 (bottom) 3,5V Pstack = 184 W FU = 37% Cell 5: 404 mw/cm² Cell 4: 476 mw/cm² Cell 3: 447 mw/cm² Cell 2: 472 mw/cm² Cell 1: 415 mw/cm² p power density p [mw/cm²] 0,0 CSZ05-DGF09-CT, 750 C 5H2+5N2+3%H2O / 20air (SLPM) 94h current density i [ma/cm²]

10 Nyquist Plot of one Cell of a 5-Cell Short Stack at Different Current Densities (750 C, 2.5 H N 2 / 20 air (SLPM), 142 h) Im Z [Ohmcm²] 1,25 1 0,75 0,5 0,25 0 ma/cm2 60 ma/cm2 120 ma/cm2 180 ma/cm2 240 ma/cm2 300 ma/cm2 360 ma/cm2 420 ma/cm2 Anode 80 Hz Cathode 6 Hz Gas Concentration 1 Hz 5-Zellen Short Stack [CSZ CT], cell 5 T=750 C, Zellfl. 84cm² 0-0,25 50mHz 0 0,25 0,5 0,75 1 1,25 1,5 1,75 2 2, khz cm cm 2-0,5 Re Z [Ohm*cm²]

11 Equivalent Circuit for the Fitting of the Impedance Spectra C dl (A) C N (A) C dl (C) R Z L R P (A) R N (A) R p (C) Anode Gas Concentration Ohmic Cathode Inductivity

12 Voltage Losses at one Cell of a 5-cell Short Stack at Different Current Densities (750 C, 2.5 H N 2 / 20 air (SLPM), 142 h) 1,2 CSZ CT, 750 C 2,5H2+2,5N2 / 21 Air (SLPM) cell 5, 142 h 100 cell voltage U [V] 1,0 0,8 0,6 0,4 U cell ΔU (Anode) ΔU (Cathode) ΔU (Gas Concentr.) ΔU (Ohm) U cell 700mV : 380 mw/cm 2 ΔU voltage loss [mv] 0,2 0,0 Contact resistances current density i [ma/cm²] 20 Polarisation resistances

13 Motivation Strong local variation of gas composition, temperature, current density Distribution of electrical and chemical potential dependent on local concentrations of reactants and products Reduced efficiency Temperature gradients Thermo mechanical stress Degradation of electrodes O 2 O 2 O 2 H 2 H 2 HO 2 HO 2

14 Measurement Setup for Segmented Cells 16 galvanically isolated segments Local and global i-v characteristics Local and global impedance measurements Local temperature measurements Local fuel concentrations Flexible design: substrate-, anode-, and electrolyte-supported cells Co- and counter-flow

15 Cell design and Testing Station GC measurement Assembly and contacts From a simple cell design with manually controlled features Flexible housing, impedance spectra with reduced interferences All cell concepts Improved contacting Reliable assembly Impedance measurement Temperature measurement

16 Schematic Lay-out of the Electrical Circuit of the Segmented Cell Configuration Internal cell resistances: Ri,j, current busbar equipotential line current busbar equipotential line Resistances of the wires contacting the anode: RLA,j Resistances of the wires contacting the cathode: RLK,j Only segments 1, 2, 3, 16 are illustrated

17 OCV Voltage Measurement for Determination of Humidity Voltage distribution at standard flow rates: 48.5% H 2, 48.5% N 2 + 3% H 2 O, 0.08 SlpM/cm² air fuel gas air U rev Nernst equation: U 0 rev RT zf ln p p H 2O O2 p H 2 Produced water: S4: 0.61%, S8: 0.72%, S12: 0.78%, S16: 3.30%

18 Variation of Load - Reformate p(i) 100 ma/cm² p(i) 200 ma/cm² p(i) 400 ma/cm² p(i) 435 ma/cm² fu 100 ma/cm² fu 200 ma/cm² fu 400 ma/cm² fu 435 ma/cm² 300,0 250, f u , ,0 power density p [mw/cm²] Power density mw/cm 2 200,0 150,0 100,0 50, ,0 45,0 30, ,0 fuel utilisation fu [%] Fuel utilisation (%) 0,0 Segment 9 Segment 10 Segment 11 Segment 12 Anode supported cell, LSCF cathode, 73,96 cm², gas concentrations (current density equivalent): 54.9% N 2, 16.7% H 2, 16.5% CO, 6,6% CH 4, 2.2% CO 2, 3.2% H 2 O (0.552 A/cm²), 0.02 SlpM/cm² air 0,0

19 0,3 Reformate: Changes of the Gas Composition at 0 ma/cm² H2 CO CH4 CO2 H2O 0,25 H 2 Concentration Gaskonzentration / % / % 0,2 0,15 0,1 0,05 CO H 2 O Metallic KS4X housing, anode in Metallischem substrate, Gehäuse; active area Substrat: Anodensubstrat, cm² Anode: aktive 542 Zellfläche:73,78 µm NiO/YSZ, cm²,a: Electrolyte: 542 µm NiO/YSZ, 14 µm YSZ E: 14 + µm YDC, YSZ + YDC, Cathode: 28 K: µm 28 µm LSCF LSCF, Kontaktierung: 30 µm LSP16+Pt3600, Operation Integral, Gasflüsse: conditions: 0, A/cm² A/cm² Stromdichteäquivalent - Anode = 5.52(54,9% N 2, 16,7% H 2, (54.9% 16,5% N CO, 6,6%CH 4, 2,2%CO 2, 3,2% H 2 0) // 0,08 SlpM/cm² Luft, 2, 16.7% H 2, 16.5% CO, 6.6% CH 4, 2.2% CO 2, 3.2% H 2 O 0.08 Nlpm/cm² Air, 800 C) 800 C, 0 ma/cm² CH 4 CO Segment 9 Segment 10 Segment 11 Segment 12

20 Alteration of the gas composition at 435 ma/cm² 0,3 H2 CO CH4 CO2 H2O 0,25 Concentration / % Gaskonzentration / % 0,2 0,15 0,1 CO H 2 H 2 O CO 2 0,05 CH Segment 9 Segment 10 Segment 11 Segment 12

21 Combined Experimental and Modeling Approach Objectives of the study: Better understanding of the local variations Identification of critical conditions Optimisation of cell components interconnector H 2 H 2 /CO CH 4 H 2 O CO 2 gas z anode electrolyte elyt elde cathode y O 2 /N 2 N 2 x interconnector Experiments on single segmented SOFC Electrochemical model of local distributions

22 Potential for Optical Spectroscopies a) In situ microscopy b) In situ Raman laser diagnostics Digital CCD camera Distance microscope (resolution1 µm) Quarz window Imaging spectrograph Heat & radiation shield Lenses/filter 15 cm Transparent flow field SOFC Pulsed Nd:YAG laser (532 nm, 10 ns) Raman spectroscopy Laser Doppler Anemometry (LDA) Particle Image Velocimetry (PIV) Fast-Fourier Infrared (FTIR) Coherent Anti-Stokes Raman Spectroscopy (CARS) Electronic Speckle Pattern Interferometry (ESPI) Open tube (5 mm)

23 Tomography Diagnosis of PEM Fuel Cells in-situ synchrotron radiography neutron tomography in-situ neutron radiography Investigation of water management under operating conditions

24 X-Ray Tomography (CT) Facility at DLR 3 dimensional non intrusive imaging of SOFC cassette X-Ray CT Facility v tome x L450 at DLR Stuttgart

25 Summary The operating conditions (elevated temperature) reduce significantly the possibilities for in-situ SOFC diagnostic methods. EIS will remain the main diagnostic probe of the state of SOFC. Non-traditional in-situ diagnostics methods can provide additional important information: Spatially resolved measurements to obtain local distribution of cell properties (current, voltage, impedance, gas composition, temperature) Combined analytical and modeling approach Large future potential for optical spectroscopies (e.g. Raman spectroscopy) and x-ray tomography.