Gas Transport and Electrochemistry in Solid Oxide Fuel Cell Electrodes

Transcription

1 Gas Transport and Electrochemistry in Solid Oxide Fuel Cell Electrodes Wilson K. S. Chiu University of Connecticut

2 Overview of Research Activities Heat & Mass Transfer with Chemical Rxns: CVD Nanomaterials Photonics Fuel Cells 7 microns Ni-YSZ Anode P. Russell, Science 99:358, 003.

3 Outline Introduction Energy Conversion Fuel Cell Background & Motivation Imaging & Analysis at the Pore Level Pore Imaging Gas Transport Electrochemistry Electrode Microstructure and Grading Conclusions & Future Research Plans

4 Global Energy Usage

5 U.S. Energy Usage Energy Information Association Annual Energy Outlook, DOE/EIA Report # 0383, 007. C. Song, Catal. Today 115:

6 Energy Conversion Efficiency US Power Plants, 003, in quadrillion BTU C. Song, Catal. Today 115:

7 Energy Conversion Technologies For Fossil Fuels: SOA: Heat Engines Steam Cycles Gas Turbines Internal Combustion Fuel Cells? High Electrical Efficiency Fuel Flexibility Scalable

8 Outline Introduction Energy Conversion Fuel Cell Background & Motivation Imaging & Analysis at the Pore Level Pore Imaging Gas Transport Electrochemistry Electrode Microstructure and Grading Conclusions & Future Research Plans

9 Types of Fuel Cells Fuel Cells Electrolyte Operating Temperature Fuel Supply Applications Alkaline (AFC) circulating liquid or matrix Hydrogen or NH 3 -Cracker small units, up to automobiles Proton Exchange Membrane(PEM) immobilized, acidic Hydrogen or Converter small units, up to automobiles Phosphoric Acid (PAFC) concentrated acid gel Hydrogen or Converter Power Plants 50 to 00 kw Molten Carbonate (MCFC) Li/Na-carbonate melt Selected fuel or Converter Power Plants up to 1 MW Solid Oxide (SOFC) ceramic Zr/Yoxides All Fuels, direct feed Small to large Power Plants

10 Solid Oxide Fuel Cell Hydrogen (& CO) Fueled Internal Reforming from Steam Reforming Shift Reaction CH H O H + CO CO + H + O H CO

11 How, specifically, can we design fuel cell microstructure to optimize performance and durability? Design Input Electrochemistry Materials Design Fabricate Test Previous Generation SOFC Current Generation SOFC Next Generation? Atkinson et al., Nature Mat. 3:17-7, 004. La 0.3 Sr 0.7 TiO 3 current collector Porous YSZ-active region YSZ electrolyte Gorte et al., 006 Modeling Tool needed to Determine Optimal Microstructure.

12 Outline Introduction Energy Conversion Fuel Cell Background & Motivation Imaging & Analysis at the Pore Level Pore Imaging Gas Transport Electrochemistry Electrode Microstructure and Grading Conclusions & Future Research Plans

13 SOFC Pore Structure Imaging X-Ray Microtomography (XMT) Non-destructive tomography capable of reconstructing 3-D pore structure. X-rays penetrate sample, which is mounted on a rotation stage, and pass through a scintillation screen. The -D slice is captured by a camera. The sample is rotated through 180 degrees where each slice is captured at specified angle intervals and reconstructed into a 3-D volume of the pore structure.

14 SOFC Pore Structure Imaging cont d XMT Experiment details Xradia (Concord, CA) 8 kev copper source 181 projections at 300 sec per projection.6 µm field of view 50 nm resolution 3-D tomographic reconstruction

15 Imaging Processing and Modeling anode Digital image file Geometry Input file for Model ImageJ 5 µm Actual SEM image Model Image A.S. Joshi, W.K.S. Chiu, et al., 9th AIAA , 006.

16 Modeling Challenges Gas diffusion occurs at high temperature and through micron size pores. Continuum theory is no longer valid. Gas particles can get adsorbed on the solid material. Complex structure. SOLID 5 [µm] Chemical reactions among gas components need to be included for the case of internal reforming and at the TPB.

17 A Comparison of Modeling Approaches Molecular Dynamics Monte Carlo SIMULATION COST Dusty Gas Model Lattice Boltzmann Method continuum slip transition free molecular KNUDSEN NUMBER

18 Lattice Boltzmann Method (LBM) Historically derived from the lattice gas cellular automata LBM is a numerical approximation to the Boltzmann equation Multiple species Complex geometry Parallel algorithm SOLID FLUID Wall interactions Non-continuum regime M.E. McCracken M. E. and J. Abraham, Phys. Rev. E, 71:

19 Basic LBM Algorithm: Stream and Collide new time level species old time-level 1 e 4 1 e 3 1 e n σ, n+ 1 i, r + e i = n σ, n i, r Ω σ, n i, r 1 e 5 1 e 0 1 e 1 direction discrete velocity velocity distribution function BGK collision term 1 1 e 6 e 7 1 e 8

20 Multi-Component Gas Transport in a SOFC Anode Digitized Pore Structure φ = 0.5 Z (lattice units) H O mole fraction Z (lattice units) H mole fraction Z (lattice units) X (lattice units) N mole fraction Gas Channel X (lattice units) TPB Gas Channel X (lattice units) TPB 0 A.S. Joshi, W.K.S. Chiu, et al., J. Power Sources, accepted, 006. A.S. Joshi, W.K.S. Chiu, et al., 4 nd Power Sources Conference, pp , 006.

21 LBM Validation Concentration polarization [ V ] Zhao & Virkar (005) Chan et al. (001) Present LB model Non-dimensional current density ( i* )

22 LBM Validation cont d. Knudsen number ( Kn ) Mole fraction ratio ( X* ) Pe = 0.01 Pe = 0.08 Pe = 0.16 Pe = 0.3 Pe = Diffusivity ratio ( D* ) LBM: data points Dusty Gas: solid lines A.S. Joshi, W.K.S. Chiu, et al., ASME IMECE , 006.

23 3D LBM Analysis of a SOFC Anode Z Electrolyte (YSZ) Y H X H O Pore space Anode (Ni + YSZ) Pore Structure (Ψ,<r>,<r >), Properties, η conc, η ohmic H mole fraction Surface representing the interface between the solid region and pores. Pore phase

24 Electrochemical Reaction Kinetics in a SOFC Anode Global Reaction: g g H + O H O + e electrolyte The Gauckler Reaction Mechanism: H k1, k 1 ( g) + S H S H O( g) O S + H H H k, k + S H k 3, k 3 S OH O S k 4, k 4 O S + O S OH O S + x O k 5, k 5 S OH S + A. Bieberle and L.J. Gauckler, Solid State Ionics. 146: 3-41, 00. S + O S S H S k 6, k S O S + VO + eni e - Electrolyte Ni H O S Anode k + / # : S: X S: V. ȯ : O o x : H O(g) e - (1) Reaction rate Surface vacancy Surface species Oxygen vacancies Oxygen ion H (g) () () OH S O S O - (4) () e - O - H S (5) (3) (1) Dissociation/Adsorption/ Desorption () Surface Diffusion (3) Triple Phase Boundary (4) O - Diffusion (5) Ion Transfer into Anode O S Triple Phase Boundary (TPB)

25 Electrochemical Model Rate Equations Traditional analysis requires selection of single rate limiting step and equilibration Langmuir-Hinshelwood type rate equation Approach breaks down when reaction mechanisms contain steps that occur at similar rates For proposed mechanism requires solution to set of (4) coupled nonlinear, stiff differential-equations θ t θ H HO ( k1θ ) k θ θ θ θ θ θ θ θ θ V 1 k H 3 H O k 3 OH V k5 H O V k 5 OH H = ( kθv k θh O + k θoh k θh OθV k θohθ 4 H ) 5 5 ( k3θh θo k 3θOHθV k4θh OθO k θoh k θh OθV k θohθ 4 H ) 5 5 ( k3θhθo k 3θOHθV k4θh OθO k OH k V k 4θ O) 6θ 6θ = + + t θoh t = + + θo = t = θ + θ + θ + θ + θ V H OH HO O θ i : Species Surface Coverage k k o F = k exp β η RT 6 6 ( β ) o F = k 1 exp η RT 6 6 ( θ θ ) i = F N N k k r r J n = o A i F 6 V 6 O

26 Validation of Reaction Kinetics Function H OH Vacancy Bessler (4) 1.0E E-06 HO O Bessler (4) Surface Coverage Surface Coverage 1.0E E Overpotential [V] 1.0E Overpotential [V] g k1, k 1 g k, k k3, k 3 H + S H S HO + S HO S O S OH S + S k4, k 4 k5, k 5 o x k6, k 6.. o HO S+ O S OH S HO S+ S OH S+ H S O + S O S + V + e W.G. Bessler, Solid State Ionics 176: , 005. K. N. Grew, W. K. S. Chiu, et al., ASME IMECE , 006. NI

27 Electrochemistry & Gas Transport in a SOFC Anode (a) (b) (c) Reifsnider et. al. 005 TPB at Ni-YSZ interfaces (d) H H, H O, & N Concentrations Specified Solid Wall Traditional TPB New method allows electrochemically Flux Interface active boundaries to be placed on (See Fig. 1d): obstacles in an ad hock manner as TPB dictates in SEM (See Fig. 1c) H Flux Impermeable Obstacle H O Flux Open Domain Solid Wall Y X Solid Wall L y / H x / L H Mole Fraction y / H x / L K. N. Grew, W. K. S. Chiu, et al., ASME IMECE , H Mole Fraction η act k +/-# θ i

28 Outline Introduction Energy Conversion Fuel Cell Background & Motivation Imaging & Analysis at the Pore Level Pore Imaging Gas Transport Electrochemistry Electrode Microstructure and Grading Conclusions & Future Research Plans

29 Electrode Microstructure Structure can be described by three parameters: channel anode electrolyte cathode Ψ,<r>, <r > TPB 0 x d model domain Ψ Structural Parameter (ε/τ) Porosity: 0 ε 1 Tortuosity: τ 1 ~10X as important upon performance as other parameters in conditions studied. <r> Average pore radius <r > Pore radius distribution Pore-Level Analysis Properties, η act, η ohmic, η conc

30 Effect of Grading Electrode Microstructure Performance evaluated with alternative anode designs Two fuel streams Diluted hydrogen - Conditions near cell outlet 10% H 3% HO balance inert gas Partially reformed methane with internal reforming- Approx. cell inlet 10% H 49% CH 4 30% H O 6% CO 5% CO Four microstructure cases {1} - Ψ is a constant high value of 0.5 {} - Ψ is a constant low value of {3} - Ψ is high at the TPB and low at the gas supply channel 0.5 -> 0.05 {4} - Ψ is low at the TPB and high at the gas supply channel > 0.5 Ψ 0.05 {3} {1} {} {4} Electrolyte Cathode

31 Model Validation by Experiments Concentric Alumina Tubes Furnace Cell Potential (V) Cell Potential Power Density Cell operating temperature 1173 K 113 K 1073 K Power Density (W/cm ) Current Density (ma/cm ) 0.0 SOFC Electric Leads E. S. Greene, W. K. S. Chiu, A. A. Burke and M. G. Medeiros, 4 nd Power Sources Conference, pp , 006.

32 Effect of Electrode Grading on Performance Diluted Hydrogen Feed With Internal Reforming E. S. Greene, W. K. S. Chiu and M. G. Medeiros, J. Power Sources 161:5-31, 006.

33 SOFC Performance Correlations Curve Fits Da ~ d 1.79 R = Da ~ Ψ R = Displays prevalently kinetically limited characteristics (Da<1) Da INCREASING Ψ τ diffusion Da = = τ reaction Ψ R D EC H d C H Exponent of power law fits display larger sensitivity of Da to d than Ψ d [mm] E.S. Greene, M.G. Medeiros and W.K.S. Chiu, J. Fuel Cell Sci. Tech. : , 005.

34 Conclusions SOFC is promising energy conversion device. Performance strongly dependent on electrode microstructure. Presented modeling/experimental approach to analyze and design SOFC electrodes. Optimized fuel cell electrodes can provide a durable high efficiency energy conversion technology for our society.

35 Acknowledgements Sponsors Office of Naval Research Army Research Office University of Connecticut - Institute of Materials Science Collaborators Naval Undersea Warfare Center Adaptive Materials NexTech Materials Xradia Advanced Photon Source, Argonne National Lab

36 Acknowledgements Fuel Cell Group Aldo Peracchio, Visiting Scholar Abhijit Joshi, Post-Doctoral Researcher Graduate Students Srinath S. Chakravarthy Eric S. Greene Kyle N. Grew John R. Izzo Andrew C. Lysaght Undergraduate & High School Students