SECA Core Program–Recent Development of Modeling

Activities at PNNL

SECA Core ProgramSECA Core Program––Recent Development of Modeling Recent Development of Modeling

Activities at PNNLActivities at PNNL

MA KhaleelEmail: moe.khaleel@pnl.gov Phone: (509) 375-2438

KP Recknagle, X Sun, BJ Koeppel, EV Stephens, BN Nguyen, KI Johnson, VN Korolev, JS Vetrano* and P Singh

Pacific Northwest National LaboratoryRichland, WA 99352

Travis Shultz, Wayne Surdoval, Don Collins National Energy Technology Laboratory

SECA Core Technology Program Peer ReviewPhiladelphia, PA

September 12-14, 2006

2

R&D Objectives & ApproachR&D Objectives & ApproachR&D Objectives & Approach

Objective: Develop integrated modeling tools to: Evaluate the tightly coupled multi-physical phenomena in SOFCsAid SOFC manufacturers with materials development Allow SOFC manufacturers to numerically test changes in stack design to meet DOE technical targets

Approach: Finite element-based analysis tools:Mentat-FC: Easy-to-use pre- and post-processor to construct a complete analytical model from generic geometry or templatesSOFC-MP: A multi-physics solver that quickly computes the coupled flow-thermal-electrochemical response for multi-cell SOFC stacksTargeted evaluation tools for eminent engineering challenges:

Interface and coating durabilityReliable sealingOn-cell reformation for thermal managementScale upTime dependent material degradation

3

Technical Issues AddressedTechnical Issues AddressedTechnical Issues Addressed

Provide a flexible, multi-physics SOFC stack design tool capable of importing SOFC manufacturer designs (planar & tubular) for electrochemical/thermal/structural analyses.Provide a coarse design methodology by which SOFC manufacturers can develop a stack design.Provide analysis tools for evaluating the effects of on-cell steam-methane reformation in stacks for optimal thermal management.Address the challenging reliability issues of glass-ceramic seals, interfaces, and scales due to steady operation and thermal cycling of stacks.Experimentally gather necessary physical and mechanical property data to support model development.

4

AccomplishmentsAccomplishmentsAccomplishmentsCommercially Available SOFC Stack Design Tool: PNNL and MSC-Software combined efforts to develop and release a user-friendly electrochemical-thermal-structural stack design software package. Design tool capability includes import of planar and non-planar SOFC stack designsTools and Methods for Optimization of Internal Reforming: Methodology developed in which control of the reaction rate via custom anode and/or control of the percentage of reformation to occur on-cell is used to optimize the thermal management of a proposed stack designGlass-Ceramic Sealant Damage Model: Developed a viscoelastic continuum damage mechanics model based on experimental characterization of G18 glass-ceramic to evaluate sealants in SOFC stack models to prevent failure/delaminationInterconnect Creep and Degradation: Developed a modeling methodology to predict interconnect integrity at different stages of oxide scale growth. Examined influence of interconnect creep on the possible stack geometry change and stress redistribution Experiments Provide Critical Properties: Testing has provided fundamental material properties enabling model development

5

PublicationsPublicationsPublications

SECA PresentationsKhaleel MA, KP Recknagle, JS Vetrano, X Sun, BJ Koeppel, KI Johnson, V Korolev, BN Nguyen, AM Tartakovsky, and P Singh, "SECA Core Program-Recent Development of Modeling Activities at PNNL." SECA Core Technology Peer Review, Lakewood, CO, October 25-26, 2005.

Topical ReportsVetrano J.S., Y.-S. Chou, G.J. Grant, B.J. Koeppel, B.N. Nguyen and M.A. Khaleel, “Mechanical Testing of Glass Seals for Solid Oxide Fuel Cells.” PNNL-15463.Koeppel BJ, BN Nguyen, JS Vetrano, and MA Khaleel, “Experimental Characterization, Model Development, and Numerical Analysis of Glass-Ceramic Sealant Relaxation Under Thermal-Mechanical Loading.” PNNL-15659.Recknagle KP, ST Yokuda, DT Jarboe, MA Khaleel, “Analysis of Percent On-Cell Reformation of Methane in SOFC Stacks: Thermal, Electrical, and Stress Analysis.” PNNL-15787.Sun X, WN Liu, P Singh, and MA Khaleel, “Effects of Oxide Thickness on Scale and Interface Stresses under Isothermal Cooling and Micro-Indentation.” PNNL-15794.

Conference Papers & PresentationsVetrano JS, Y-S Chou, BJ Koeppel, BN Nguyen and MA Khaleel, “Modeling and Measurement of Material Behavior in Solid Oxide Fuel Cells,” Fuel Cell Seminar, November 15, 2005, Palm Springs, CA. PNNL-46383.Khaleel MA, X Sun and A Tartakovsky, "Multi-Component-Based Reliability Design for SOFC- A Coarse Design Methodology," Fuel Cell Seminar, November 15, 2005, Palm Springs, CA. PNNL-SA- 47472.Recknagle KP, KI Johnson, V Korolev, DT Jarboe, MA Khaleel, and P Singh, “Electrochemistry and On-Cell Reformation Modeling for Solid Oxide Fuel Cell Stacks,” 30th International Conference & Exposition Advanced Ceramics, January 24, 2006, Cocoa Beach, FL.Koeppel BJ, JS Vetrano, BN Nguyen, X Sun, and MA Khaleel, “Mechanical Property Characterizations and Performance Modeling of SOFC Seals,” 30th International Conference & Exposition Advanced Ceramics, January 24, 2006, Cocoa Beach, FL.Nguyen BN, BJ Koeppel, JS Vetrano, and MA Khaleel, “On the Nonlinear Behavior of a Glass-Ceramic Seal and Its Application in Planar SOFC Systems,” 4th International Conference on Fuel Cell Science, Engineering, andTechnology, June 19-21, 2006, Irvine, CA.

Journal ArticlesNguyen BN, BJ Koeppel, S Ahzi, MA Khaleel, and P Singh. 2006. “Crack Growth in Solid Oxide Fuel Cell Materials: From Discrete to Continuum Damage Modeling.” J Am Ceram Soc 89(4):1135-1368.

6

Teaming & CollaborationsTeaming & CollaborationsTeaming & Collaborations

IndustryModeling and Software Training

GEDelphiAcumentricsSiemensFCE

University & National LabsModeling

U of Illinois, ChicagoGeorgia Tech

MaterialsORNLCarnegie Mellon UniversityPenn State

Software TrainingU of Connecticut

ResultsResultsResults

SOFC Analysis OverviewSOFC-MP/Mentat-FC

On-Cell Reformation & Thermal ManagementSeal Property & Time Dependent Behavior

Interconnect-Coating Interface & Interconnect CreepExperimental Property Measurement

8

SOFC Analysis OverviewSOFC Analysis OverviewSOFC Analysis Overview

Developed tools to build/analyze SOFC cells and stacks

Mentat-FC: GUI to build models from templates, CAD files, or FEA meshesSOFC-MP: Coupled thermal, flow, and electrochemistry solverMSC.Marc: Structural finite element analysis using SOFC-MP temperatures

9

SOFC-MP/Mentat-FCSOFCSOFC--MP/MentatMP/Mentat--FCFCMentat-FC GUI

Guides user through entire analysisBuilds geometry from CAD files, FEA meshes, or templates (planar co-, counter-, cross-flow)SOFC operating parameters (I-V, fuel/oxidant inputs, polarizations)Exterior thermal boundary conditionsMaterial properties databaseHas tubular capability

SOFC-MP SolverFinite element basedGeneric fuel and oxidants (CEA)Efficient reduced order dimensional analyses for electrochemistry and gas flowsContact algorithms treatincompatible meshes

Post-processing of electrical output, species, thermal distribution, deformations, and stresses

10

On-Cell Reforming & Thermal Management:Optimization of Percent On-Cell ReformationOnOn--Cell Reforming & Thermal Management:Cell Reforming & Thermal Management:Optimization of Percent OnOptimization of Percent On--Cell ReformationCell Reformation

40% OCR

80% OCR

0% OCR

FuelIN

Air IN

% OCR ∆T Tmax Anode S1max Seal S1max

0 85 781 25.8 11.340 45 768 13.8 8.750 47 768 14.7 8.860 54 769 16.3 9.280 69 773 19.0 10.2

Principal Stress Max

At 0% OCR temperature reached maximum at air exit

40% OCR heat load was decreased with cooling at fuel inflow

80% OCR heat load was minimum but air temperature was high

10x10 counter-flowCell Temperature = 750°C All Cases

0

10

20

30

40

50

60

70

80

90

100

0 20 40 60 80

Percent On-Cell Reforming

500

550

600

650

700

750

800

850

900

Inflow

Tem

per

ature

, °C

Temperature Difference, °C

Inflow Air Temperature, °C

SUMMARY:To maintain 750°C, inflow temperature increased to offset decreased heat load due to reforming (all configurations and sizes)Seal stresses, anode stresses, and ∆T had minimums at 40-50% OCR for this configuration and cell size

11

Seal Property & Time Dependent Behavior:Glass-Ceramic Seals

Seal Property & Time Dependent Behavior:Seal Property & Time Dependent Behavior:GlassGlass--Ceramic SealsCeramic Seals

CharacteristicSeal Assembly

Interface

0.0

0.5

1.0

1.5

2.0

2.5

0 20 40 60 80 100 120 140 160 180 200Time (min)

Raw

Cre

ep E

xten

sion

(mm

)

G800-20 Raw Creep ExtG800-21 Raw Creep ExtG800-22 Raw Creep ExtG750-23 Raw Creep ExtG700-33 Raw Creep ExtG750-36 Raw Creep ExtG750-43 Raw Creep ExtG700-47 Raw Creep ExtG700-51 Raw Creep Ext

TorsionTension

0

10

20

30

40

50

60

0 200 400 600 800 1000Temperature (C)

Stre

ss (M

Pa)

Normal StrengthShear Strength

Glass-CeramicInterface

0

20

40

60

80

100

120

0.000 0.002 0.004 0.006 0.008Strain (mm/mm)

Stre

ss (M

Pa)

4hr-25C4hr-600C4hr-700C4hr-750C4hr-800C

2. Test the weakerinterface strength

1. Test the glass-ceramic strength and creep behavior

CreepBending

3. Obtain coefficient of thermal expansion (CTE) and elastic modulus

12

Seal Property & Time Dependent Behavior:Creep of Glass-Ceramics

Seal Property & Time Dependent Behavior:Seal Property & Time Dependent Behavior:Creep of GlassCreep of Glass--CeramicsCeramics

Evaluated short-term creep behavior of G18

Glassy matrix above its Tg at cell operating temperaturesCreep rate increased with applied stress and temperature

Estimated G18 viscosityAssume Maxwell material model

Derived viscosity is very high (5-65e6 MPa-s) but still important for assessing steady state seal stresses in the stack

)1(o DEE −=

σσ

ε

0.E+00

1.E+07

2.E+07

3.E+07

4.E+07

5.E+07

6.E+07

7.E+07

8.E+07

0 10 20 30 40 50 60Applied Stress (MPa)

Visc

osity

(MPa

-s)

800C Viscosity750C Viscosity700C Viscosity

0.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

0 50 100 150 200Time (min)

Cre

ep S

train

800C 30MPa800C 20MPa800C 10MPa

13

Seal Property & Time Dependent Behavior:Glass-Ceramic Constitutive Model

Seal Property & Time Dependent Behavior:Seal Property & Time Dependent Behavior:GlassGlass--Ceramic Constitutive ModelCeramic Constitutive Model

Stress-strain response of G18 glass-ceramic shows strong rate dependence

Due to viscoelastic response of residual glassProcessing/microstructure variations noted

Viscoelastic damage model predictions compare reasonably well to experimental data

Failure strain prediction goodStress prediction possibly affected by heterogeneous structure

0

10

20

30

40

50

60

0.000 0.001 0.002 0.003 0.004 0.005 0.006 0.007 0.008Strain (mm/mm)

Stre

ss (M

Pa)

0.025 mm/min

0.05 mm/min0.5 mm/min

14

Seal Property & Time Dependent Behavior:Stack Modeling Results

Seal Property & Time Dependent Behavior:Seal Property & Time Dependent Behavior:Stack Modeling ResultsStack Modeling Results

Temperature (C) Elastic Model: Anode Maximum Principal Stress

(MPa)

Viscoelastic Model: Anode

Maximum Principal Stress

(MPa)

Change

Cycle 1 Operation

38.4 366.3%

Cycle 1 Shut-Down

65.6 62.74.4%

Cycle 2 Operation

40.2 400.5%

Cycle 2 Shut-Down

74.4 67.49.4%

Seal Failure After 2 Thermal CyclesSeal

Damage Distribution

Anode Principal Stress Distribution

15

Interconnect Life PredictionInterconnect Life PredictionInterconnect Life Prediction

Sources of interconnect stress studied:

Thermal stressGrowth stress

Predicted interfacial stress compared with interface strength

scale delaminationPredicted scale stress compared with critical buckling stress

scale spallation Residual stresses measurements in Cr2O3 films*

-2000

-1500

-1000

-500

0

500

0 0.1 0.2 0.3 0.4 0.5Distance from Bottom Surface of IC (mm)

Nor

mal

Str

ess

-S11

(MPa

)

Case 1 - 5micorn scale

scale/substrate interface

Predicted stress profile through thethickness of the IC/oxide

-2000

-1500

-1000

-500

0

500

0 0.1 0.2 0.3 0.4 0.5Distance from Bottom Surface of IC (mm)

Nor

mal

Str

ess

-S11

(MPa

)

Case 1 - 5micorn scale

scale/substrate interface

Predicted stress profile through thethickness of the IC/oxide

Predicted scale stressPredicted scale stress

Near the sample edge

800

1200

1600

2000

0 10 20 30Thickness of Scale (microns)

Nor

mal

str

ess

S11

(MPa

)

0

200

400

600

800

Shea

r str

ess

S13

(MPa

)S11 (in compression)S13

16

Interconnect Life PredictionInterconnect Life PredictionInterconnect Life PredictionUsing indentation test to quantify strength of oxide/interconnect interface:

Effect of scale growth on interfacial shear stressEffects of thermal stress on indentation stress

Conclusions:Shear stress on the interface caused by indenter contact is high enough to lead to delaminationShear stress dominates the fracture mode of the interfaceThe thinner the thickness of oxide scale is, the bigger the interfacial shear stress is, at the small indentation depthEffect of thermal stress on the interfacial shear stress may be neglected

Effect of Scale Thickness on S12

0

2000

4000

6000

8000

0 1 2 3 4 5

Indentation Depth (micron)

Shea

r Str

ess

(MPa

)

2 microns 5 microns15 microns30 microns

Scale Thickness: 5 microns

0

1000

2000

3000

4000

5000

6000

0 1 2 3 4 5Indentation Depth (micron)

Shea

r Stre

ss (M

Pa)

-25000

-20000

-15000

-10000

-5000

0

Nor

mal

Stre

ss (M

Pa)

S12 w/o thermal stressS12 w thermal stressS11 w/o thermal stressS11 w thermal stress

17

Interconnect Scale Growth Kinetics and Growth Stress Prediction

Interconnect Scale Growth Kinetics and Interconnect Scale Growth Kinetics and Growth Stress PredictionGrowth Stress Prediction

Diffusion potential used to govern stress-diffusion interactionPredicting:

Oxidation kineticsDiffusion and concentration evolution of species Stress distribution in scale and substrate

[74]. Mikkelsen, L. and S. Linderoth,. Materials Science & Engineering A (Structural Materials: Properties, Microstructure and Processing), 2003. A361(1-2): p. 198-212

Oxidation kinetics profile (22% CrOxidation kinetics profile (22% Cr--Fe alloy) Fe alloy)

Stress distribution and evolution

0.01 0.02 0.040.03 0.05

18

Interconnect Creep and Stress Re-distribution

Interconnect Creep and Stress ReInterconnect Creep and Stress Re--distributiondistribution

Creep law used for IC

⎩⎨⎧

≤×≥

=−

−

C710for1065.1C725for72.1

o/4887008.511

o/2774005.5

TeTe

RT

RT

c σσ

ε&

⎩⎨⎧

≤×≥>

=−

−

C710for2.28C725for10

o/4887005.10

o3

TeT

RTc σε&

MPa100<σ

MPa100>σ

Time = 0.01 h Time = 70 h

IC

anode

seal

19

Experimental Support of ModelingExperimental Support of ModelingExperimental Support of Modeling

Providing input data to the material models and validating the models through experimental testing. In addition, contributing

to the materials database maintained by NETL.

Creep testing of G18 glass to measure long-term deformation behavior

Torsion testing of thin glass-seal analogs to determine failure stress

and fracture behavior

4-pt bend bar testing of G18 glass to capture constitutive and failure

behavior of the material

20

Experimental Property MeasurementExperimental Property MeasurementExperimental Property MeasurementSeal Property & Time –Dependent Behavior

Creep tests of the G-18 glass performed to support damage models in predicting the long-term stack performance

Interfacial Degradation and Failure Mechanisms

Investigating failure locations and mechanisms to incorporate into seal and coating development models

A

B

C

D

SEM cross-section of a torsion sample that failed at 800oC

0.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

0 50 100 150 200Time (min)

Cre

ep S

train

800C 30MPa800C 20MPa800C 10MPa

Comparison of a non-tested creep sample

and a sample tested at 800oC and loaded at 129lb. Forty-three

percent strain observed.

Creep strain results of samples tested at 800oC under three different loadings

Creep specimen

21

Experimental Property MeasurementsExperimental Property MeasurementsExperimental Property Measurements

Interconnect-Scale/Coating Interface PropertiesIndentation tests performed to support interface models in establishing a stress-based criteria for interfacial strength and fracture toughness

Scale

Crofer 22 substrate

SEM image of the cross-section of a crofer plate oxidized at 800oC for 12hr

Top view of the indent test. Spallation of the oxide observed under 150kg load

22

Activities for the Next 6 MonthsActivities for the Next 6 MonthsActivities for the Next 6 Months

Mentat-FC/SOFC-MP: Deploy at the 6 industry sites and provide training and support.Scale-up: Examine planar SOFC design limitations considering electrochemistry, flow, and structural Parametric studies on the effects of creep of various SOFC components on long-term behavior of multiple cell stacksContinue interconnect degradation and life prediction workDevelopment of structure-property relationships for glass-ceramic sealants; improvement via microstructural designEffect of high pressure on electrochemistry and stack performanceMeasurement of Oxide-Crofer and Spinel-Crofer interfacial properties

23

Looking Forward- Phase IILooking ForwardLooking Forward-- Phase IIPhase II

Design Limitations for Scale-up of SOFC StacksVirtual feasibility study on:

Stack EC performance and thermal managementStack structural reliability

Degradation modeling and life predictionCreep considerations on the structural stability of stacksSeal and seal interfacesInterconnect scale and coatingsCell and cell interfacesCurrent collector interfaces

System integrationStack thermal management and cell thermal profilesSystem integration, power electronics and control

Validation