Anodes for Direct Hydrocarbon Solid Anodes for Direct Hydrocarbon Solid Oxide Fuel Cells (SOFCOxide Fuel Cells (SOFC’’s) s)

Challenges in Challenges in –– materials selection and materials selection and depositiondeposition

VenkatesanVenkatesan V. KrishnanV. KrishnanDepartment of Chemical EngineeringDepartment of Chemical Engineering

IIT DelhiIIT Delhi

Barriers to the hydrogen economyBarriers to the hydrogen economy• How to make hydrogen (SR, electrolysis, Solar, Borohydride)?

• Loss of efficiency in making hydrogen

• Storage and Transportation?

• SOFC’s can work with any ‘fuel’

• Let us work in parallel with the Hydrogen advocates by using hydrocarbons directly with SOFC’s

• Do not forget – well to wheels efficiency!

DIRECT OXIDATION OF DIRECT OXIDATION OF HYDROCARBONS AND OXYGENATESHYDROCARBONS AND OXYGENATES

Reactions:Cathode: (x + y/4 - z/2) O2 + (4x + y- 2z) e (2x + y/2 - z) O2-

Anode: CxHyOz + (2x + y/2 - z) O2- (y/2) H2O + x CO2 + (4x + y - 2z) e

Overall: CxHyOz + (x + y/4 - z/2) O2 (y/2) H2O + x CO2

Natural Gas, Propane/ Butane (LPG), Naphtha, Diesel, Alcohols,Natural Gas, Propane/ Butane (LPG), Naphtha, Diesel, Alcohols,Syngas (from coal and biomass) Syngas (from coal and biomass)

WHAT HAPPENS IN THE ANODE IN SUCH CASES ? WHAT HAPPENS IN THE ANODE IN SUCH CASES ? -- Direct oxidation; Direct oxidation; CHCH44 + 4O+ 4O22-- COCO22 + 2H+ 2H22O + 4eO + 4e-- Internal reforming; Internal reforming; CHCH44 + H+ H22O O CO + 3HCO + 3H22

-- Syngas oxidation; Syngas oxidation; CO + HCO + H22 + O+ O22-- COCO22 + H+ H22OO-- Coke deposition (thermal, catalytic); Coke deposition (thermal, catalytic); CHCH44 CHCHxx + H+ H22

WHY DO WE NEED NEW ANODESWHY DO WE NEED NEW ANODES

•• Incompatibility of NiIncompatibility of Ni--YSZ anodes with any fuel other YSZ anodes with any fuel other than Hthan H22 –– reforming essential; reforming essential; coking with dry coking with dry hydrocarbonshydrocarbons Reference: Toebes et al.Reference: Toebes et al.

•• Poisoning of NiPoisoning of Ni--YSZ by YSZ by sulfur compoundssulfur compounds•• Poor Redox tolerancePoor Redox tolerance of Niof Ni--based anodesbased anodes•• YET, must satisfy the basic needs of SOFC anodes YET, must satisfy the basic needs of SOFC anodes ––

–– Electronic conductivityElectronic conductivity–– Ionic conductivityIonic conductivity–– Good catalytic activityGood catalytic activity–– Compatibility of Compatibility of CTECTE’’ss with that of electrolytewith that of electrolyte

EFFECT OF COKING ON NiEFFECT OF COKING ON Ni--YSZ ANODESYSZ ANODES20% CO/7% H2550°C. Ni particles

Coking due to decomposition and/or Bouduard Reaction

In H2, 800oC, 3 hrs In CH4, 800oC, 3 hrs

Toebes et al. (2002)Toebes et al. (2002)

ALTERNATIVE ANODES ?ALTERNATIVE ANODES ?Perovskites, Irvine et al.Perovskites, Irvine et al.BoukampBoukamp, et al. , et al.

AABBOO33

Lanthanides/ Alkali earth Metal ionsOxygen ionsTransition metal ions

Cu/CeOCu/CeO22/YSZ/YSZ Gorte et al. (2000)Gorte et al. (2000)

CeCe0.90.9GdGd0.10.1OO1.95 1.95 combined withcombined with

(La(La0.80.8SrSr0.20.2) (Cr) (Cr0.80.8MnMn0.20.2)O)O33--δδ

Barnett et al. (2002)Barnett et al. (2002)

Irvine et al. (2001)Irvine et al. (2001)

TitaniaTitania--Niobia, NbNiobia, Nb22TiOTiO77--xx

Irvine et. al. (2003)Irvine et. al. (2003) ––(La(La0.80.8SrSr0.20.2) (Cr) (Cr0.50.5MnMn0.50.5)O)O33--δδ

Why perovskites???Why perovskites???

•• Ionic and electronic conductivity can be tailoredIonic and electronic conductivity can be tailored

•• Good Hydrocarbon oxidation catalytic activity has been Good Hydrocarbon oxidation catalytic activity has been reported.reported.

•• No cokingNo coking

•• Good thermal and redox stabilityGood thermal and redox stability

•• And yet, concern remains And yet, concern remains –– do we have adequate do we have adequate electronic conductivity under reducing conditions?electronic conductivity under reducing conditions?

Review of work on alternate anodes Review of work on alternate anodes -- Gorte et al.Gorte et al.• Anodes based on Copper – conductor; Ceria – electrocatalyst

– Cu and CeO2 deposited on pre-formed porous anode substrates– Intermediate temperature SOFC’s

• Considerable data on butane, methane, diesel, propane –– Anode stable over long periods of time– Direct electrochemical Oxidation observed in button cells– Workable with fuel with existing sulfur levels

• Technology transfer to Franklin Fuel Cells, PA, US– Demonstration level, with diesel, gasoline and ethanol

• Performance tends to be lowered due to Copper sintering; enhancement necessary by carbon deposition

• Little/ no data on redox tolerance (deposition technique may be advantageous, use of Vol % < 20%)

Gorte et al., contGorte et al., cont’’dd

0.0

0.3

0.6

0.9

1.2

1.5

0.0 0.2 0.4 0.6 0.80.00

0.04

0.08

0.12

0.16

0.20

V P

I, A/cm2

A B

C

C

0.0

0.3

0.6

0.9

1.2

1.5

0.0 0.2 0.4 0.6 0.80.00

0.04

0.08

0.12

0.16

0.20

V P

I, A/cm2

A B

C

C

1E-03

1E-01

1E+01

1E+03

1E+05

0 5 10 15 20 25 30

Volume Percent Cu, %

Effe

ctiv

e Co

nduc

tivity

, S/c

m

Overall ‘connectivity’ of Cu is good, leading to about 4300 S/cm (from 4-point probe)

Compare with Ni-YSZ made by solid state methods

Performance, Conductivity with Cu contentPerformance, Conductivity with Cu content• Carbon enhancement

occurs even at 25% Vol Cu; constant, independent of Cu

• But sufficient conductivity is attained with less than 10% volume

Volume fraction of Cu

σ, C

o nd u

c tiv

ity, S

/cm

Max

. P D

, W/c

m2(w) carbon

Conductivity(w/o) carbon

Performance

Overall conductivity not the problem; the issue is proper ‘connectivity’ at the 3-phase boundary

0

1000

2000

3000

4000

5000

0.00 0.05 0.10 0.15 0.20 0.25 0.300.00

0.05

0.10

0.15

0.20

0.25

Irvine et al. (2003)Irvine et al. (2003) (La(La0.750.75 SrSr0.250.25 CrCr0.50.5 MnMn0.50.5 OO33--δδ), ), LSCMLSCM

Conductivity of perovskite at 850oC- in air, 28 S/cm- in reducing atmosphere, 1.12 S/cm

Max. P. D. (W/cm2) = 0.47, 900oCMax. P. D. in Methane = 0.3 W/cm2

(at 950oC)

Irvine et al., NiobiaIrvine et al., Niobia--Titania rutile system (2001)Titania rutile system (2001)

Reduction

•• Max. conductivity under reduction, 300 S/cmMax. conductivity under reduction, 300 S/cm•• Good catalytic activity towards methane Good catalytic activity towards methane

OxidationOxidation

Barnett et al. (2002) Barnett et al. (2002) –– LSCMLSCM--GDC composite GDC composite

• Usage of LSCM – high electronic conductivity in reducing conditions, as well

• Compatible with GDC electrolytes

Small qty of Ni +Small qty of Ni +CeCe0.90.9GdGd0.10.1OO1.951.95 –– ionic conductionionic conduction

(La(La0.80.8SrSr0.20.2) (Cr) (Cr0.80.8MnMn0.20.2)O)O33--δδ −− electronic conduction electronic conduction

Compared to NiCompared to Ni--GDC, performance in CGDC, performance in C33HH88 is significantly higheris significantly higher

Focus of research at Chemical Eng. Dept., IITFocus of research at Chemical Eng. Dept., IIT--DD•• Nature of work essentially fundamental Nature of work essentially fundamental -- ‘‘button cellsbutton cells’’•• Aim is to contribute to existing knowledge Aim is to contribute to existing knowledge ‘‘poolpool’’, via publications , via publications

and patents by addressing problems involving Electrochemistry, and patents by addressing problems involving Electrochemistry, Materials and CatalysisMaterials and Catalysis

•• We look forward to collaborate with We look forward to collaborate with ‘‘developmental laboratoriesdevelopmental laboratories’’, , e.g., test our ideas on a larger scale e.g., test our ideas on a larger scale –– CGCRI, BHEL R&DCGCRI, BHEL R&D

What we are involved in What we are involved in --•• SOFC component fabrication SOFC component fabrication –– tape casting, sintering, painting, tape casting, sintering, painting,

impregnationimpregnation•• Characterization of components Characterization of components –– bibi--layers, catalysts layers, catalysts –– SEM, SA/ SEM, SA/

PSD, Porosity, XRD, Thermal AnalysisPSD, Porosity, XRD, Thermal Analysis•• Testing of Electrical ConductivityTesting of Electrical Conductivity•• Fuel Cell TestingFuel Cell Testing

FUEL CELL REACTOR AND OPERATIONFUEL CELL REACTOR AND OPERATION

Fuel

Air

Cathode Lead (Ag/ Pt)

Anode Lead, Au

Ag/ Pt mesh

LSM-YSZ Cathode

Anode

Ceramabond, adhesive

1/2” OD Alumina tube

Tape Casting (nonTape Casting (non--AqAq))

1.2m LPEGPlasticizer2.0gPV BB inder0.5m LO leic A cidD ispersant10m LEtO H20m LM E KSolvent15gY SZPow der

Q uantityC om ponentC om ponent function

1.2m LPEGPlasticizer2.0gPV BB inder0.5m LO leic A cidD ispersant10m LEtO H20m LM E KSolvent15gY SZPow der

Q uantityC om ponentC om ponent function

Characterization of tapeCharacterization of tapeSEM of Graphite – as received

SEM of sintered bilayer

Electrolyte

TG/DTA of Electrolyte Tape

13

13.5

14

14.5

15

15.5

16

16.5

17

17.5

0 200 400 600 800 1000 1200 1400 1600

Temp (C)

wei

ght

(mg

)

-40

-30

-20

-10

0

10

20

30

uV

tgaDTA

Less than 500oC

Anode Preparation by Perovskite DepositionAnode Preparation by Perovskite Deposition

This method is totally different from the mixing and solidThis method is totally different from the mixing and solid--statestatesynthesis techniques listed in most literature.synthesis techniques listed in most literature.Moving away from the 3Moving away from the 3--phase boundary limitation to a 2phase boundary limitation to a 2--phase boundary situationphase boundary situation

Impregnation & calcination

Final Anode Structure

Porous anode matrix

Three Phase Boundary

back

e-

e-

e- e-

e- e-

Two Phase BoundaryTwo Phase Boundary-- a new approacha new approachfuel

electrolyte

anode

back

SEM and XRD both confirm that the required phase is formed inside the pores. But conductivity is very low i.e., 10But conductivity is very low i.e., 10--2 2 S/cm.S/cm.

Impregnated La Impregnated La CrOxCrOx on preformed anodeson preformed anodes

Perovskites depositedLaCrO3La0.75Sr0.25CrO3

La0.9Sr0.1CrO3

StarchStarch

Morphology of biMorphology of bi--layerslayers

• Investigation of pore sizes –choice of pore-formers, like Graphite, Starch, PMMA to name a few

• Porosity• Pore size distributionWhich are dependent on –• Particle sizes• Particle size distribution• Particle ‘shapes’

Tailoring the porosity of anodeTailoring the porosity of anode

Removal of pore formers by 950oC, completely

05

101520253035404550

0 200 400 600 800 1000 1200 1400 1600

Temp C

% re

d in

wei

ght

-30

-25

-20

-15

-10

-5

0

5

10

uV

TGADTA

TGA/DTA of Anode, in air;TGA/DTA of Anode, in air;Pore former is 50% Graphite (Alfa Pore former is 50% Graphite (Alfa AesarAesar))

TGATGA--DTA of anode pellet withDTA of anode pellet with30% graphite in N30% graphite in N22 atmatm

-2

0

2

4

6

8

100 200 400 600 800 1000 1200 1400

Temp C%

wei

ght L

oss

-30

-25

-20

-15

-10

-5

0

5

uV

TGADTA

Removal of pore formers only begins at 1100oC

Sintering temperature is usually 1450Sintering temperature is usually 1450ooC up to 1550C up to 1550ooCCPore structure may have a strong dependence on temperaturePore structure may have a strong dependence on temperatureof removal of pore formerof removal of pore former

Work in ProgressWork in Progress

• Preparation and testing of – chromites, titanates, niobates and vanadates

• Impregnation of perovskites onto pre-formed anode substructure

• Conductivity data on oxides as electrocatalysts under reducing conditions – redox stability of oxides, as well

• Systematic investigation of pore and particle morphology

• Testing of anode supported cells on various fuels