development of dense ceramic membranes for hydrogen … file3 pioneering science and technology doe...

A U.S. Department of EnergyOffice of Science LaboratoryOperated by The University of Chicago

Argonne National Laboratory

Office of ScienceU.S. Department of Energy

Development of Dense Ceramic Membranes for Hydrogen Separation*

U. (Balu) Balachandran

May 25, 2005

This presentation does not contain any proprietary or confidential information

*Work supported by U.S. DOE, Office of Fossil Energy, National Energy Technology Laboratory.

PDP-14

Overview

2

Pioneering Science andTechnology

DOE Program Review (May 25, 2005)

Timeline• This is ongoing DOE Lab work funded by Fossil Energy on an annual basis.• Started FY 1998.

PartnerNETL - National Energy Technology Laboratory

Barriers to “Separations and Other Cross-Cutting H2 Production”

(L) Hydrogen Transport Membrane (HTM) will be durable at high temp./pressure.

(M) HTM is stable in steam, CO, CO2, CH4, and H2S.

(O) Selectivity is infinite.(Q) HTMs give industrially significant flux.(S) Cermet HTM reduces membrane cost.

• Total Funding: $3350K (DOE)• FY 2004 Funding: $400K• FY2005 Funding: $550K

Budget

3



ObjectivesDevelop dense ceramic membranes for separating hydrogen from mixed gases at commercially significant fluxes under industrially relevant operating conditions. Product streams from coal gasification and methane reforming are of particular interest. Membrane must:

- have low cost- have high selectivity for H2

- give industrially significant flux - withstand high pressure and temperature- be chemically stable in presence of steam, CO, CO2,

CH4, H2S, etc.

4



Approach

Our three-pronged approach aims to develop:

• Pure mixed proton-electron conductors e.g., acceptor-doped cerates and zirconates

• Cermets (i.e., ceramic-metal composites) that contain mixed conductors and a metal that enhances the membrane’s ambipolar conductivity and flux

• Cermets composed of mechanically durable ceramic and a metal/alloy with high hydrogen permeability

5



Approach - Continued• Select/fabricate candidate materials based on fundamental

principles of defect chemistry and mass transport. • Measure hydrogen flux of candidate material versus temperature.• If flux is high, test short-term (≈100 h) chemical stability in gases

containing H2S, CO, CO2, H2O, etc.• If membrane seems chemically stable, measure mechanical

properties (fracture strength, creep).• Measure hydrogen flux at high pressures (NETL).• Optimize fabrication methods to reduce membrane thickness and

maximize flux.• Test long-term (≈1000 h) chemical stability in simulated coal

gasification atmospheres.• Under guidance from NETL program managers, transfer membrane

technology through industrial collaborations.

6



Advantages of Dense Cermet Membranes for Hydrogen Separation

• Selectivity is theoretically infinite.• Flux rates >200 cm3/min-cm2 (400 scfh/ft2) are attainable under

“real-world” pressure conditions. - flux >30 cm3/min-cm2 (60 scfh/ft2) already shown at ambient P

• Commercial, well-proven ceramic processing is used.- device will be economical

• Will accept high-temperature and high-pressure gas streams- suitable for FutureGen plants

• No pore plugging/closure• Tolerates steam, CO, CO2, CH4, and H2S • Supplies CO2 stream at high temperatures and pressures

- important for carbon sequestration• Can enhance equilibrium product conversion

7



ANL’s Approach to HTM Development(U.S. Patent 6,569,226, May 27, 2003)

FeedCO2COH2etc.

H+

H2

H2

H+

e-H /

e-H / ANL-2

Mixed Conductor With Metal

H+

H2

H2

H+

e-

e-

ANL-1

FeedCO2COH2etc.

• High Flux• High Selectivity• Good Mechanical Integrity

Structural CeramicWith Hydrogen Transport Metal

H2

H2

H

H ANL-3

FeedCO2COH2etc.

JH2=

AΦl

pH2

feed − pH2

sweep( )

Single-Phase Mixed Conductor

• Low σe-• Low Flux• Poor Mechanical Integrity

JH2=

RT4F2l

⋅σ amb ⋅lnpH2

II

pH2

I

H+

e-

H2

ANL-0

σ amb =σ

H + •σe −

σH + + σ

e −

FeedCO2COH2etc.

H2

ANL Membrane Compositions

8



Membrane Matrix Metal

ANL-0 BCY ----

ANL-1a BCY Ni

ANL-1c TZ-8Y Ni

ANL-2a BCY Pd

ANL-2b CMO Pd/Ag(23 wt.%)

ANL-3a Al2O3 Pd

ANL-3b BaTiO3 Pd/Ag(23 wt.%)

ANL-3c Al2O3 Nb

ANL-3d Al2O3 Pd/Ag(23 wt.%)

ANL-3e TZ-3Y Pd

ANL-3f TZ-8Y Pd

ANL-3g CaZrO3 Pd

ANL-4a Cu Nb

Notes:BCY = BaCe0.8Y0.2O3-∂CMO = Ce1-xMxO2-∂ (M: Gd, Y)TZ-3Y = ZrO2 (3 mol.% Y2O3)TZ-8Y = ZrO2 (8 mol.% Y2O3)

ANL Membranes (HTMs)ANL-0: Mixed conductor (e.g., BCY)ANL-1: Mixed conductor + metal with low H2

permeability (e.g., Ni)ANL-2: Mixed conductor + metal with high H2

permeability (e.g., Pd, alloys, etc.)ANL-3: Structural ceramic (e.g., Al2O3) +

metal with high H2 permeability (e.g., Pd, alloys, etc.)

ANL-4: Ductile metal + metal with high H2permeability (e.g., Pd, alloys, etc.)

Schematic of Experimental Setup

9



Sweep Gas (Ar/N2)

GCGlass or Gold Seal Alumina Tube

To Hood

N2

Furnace

Alumina TubeGas Chromatography

Unknown H2 Concentration

To Hood

Feed Gas

Alumina Tube

To Vacuum Pump

Membrane

To Hood

10



Comparison of Cermet Membranes

0.050.100.15600700/Temperature (°C) H

0.05

0.10

0.15

0.20

0.25

0.30

600 700 800 900

Feed Gas: 4% H2 in Ar @ Ambient Pressure

Membrane Thickness = 0.50 mm

ANL-1ANL-2ANL-3

0.16

0.24

0.32

0.40

0.48

0.56

H2 F

lux

(cm

3 /min

-cm

2 )

Temperature (°C)

H2 F

lux

(scf

h/ft2 )

• ANL-3 HTMs give highest H2 flux of ANL membranes.

11



H2 Flux (ANL-3a) vs. TemperatureThickness ≈ 40 µm, Feed Gas: 100% H2 at ambient pressure

H2

Flux

(scf

h/ft2 )

H2

Flux

(cm

3 /min

-cm

2 )

Temperature (°C)

0.0

5.0

10.0

15.0

20.0

25.0

300 400 500 600 700 800 900 10000.0

8.0

16.0

24.0

32.0

40.0

48.0

• ANL-3a gives highest flux of self-supported HTMs.

12



H2 Flux vs. ∆pH21/2

∆pH21/ 2 ≡ pH2 ( feed) − pH2 (sweep) (atm1/ 2)

• Linear dependence of H2 flux on ∆pH2

1/2

shows flux is limited by bulk diffusion.

• Extrapolation shows stand-alone ANL-3e HTM should yield H2flux >200 scfh/ft2 at 900°C with feed gas at ≈300 psi.

0

20

40

60

80

100

120

0

50

100

150

200

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0

H2 F

lux

(cm

3 /min

-cm

2 )

H2 F

lux

(scf

h/ft2 )

900°C

600°C

ANL-3e (Thickness - 22 µm)

=AΦl

pH2

feed − pH2

sweep⎛ ⎝ ⎜ ⎞

⎠ ⎟ H2 Flow Rate

13



H2 Flux at High Feed Gas PressuresANL-3a (Thickness ≈0.8 mm)

0.0

1.0

2.0

3.0

4.0

5.0

0 50 100 150 200 250 300 350 400

465°C635°C765°C900°C

0.0

2.0

4.0

6.0

8.0H

2Fl

ux (c

m3 /m

in-c

m2 )

H2

Flux

(scf

h/ft2 )

pH2-Feed (psia)

• Measured flux values agree with extrapolated values.

14



H2 Flux vs. Inverse HTM ThicknessH

2Fl

ux (c

m3 /m

in-c

m2 )

1/Thickness (cm-1)

H2

Flux

(scf

h/ft2 )

Feed Gas: 80% H2/He at Ambient PressureANL-3e @ 900°C

0.0

2.0

4.0

6.0

8.0

10.0

12.0

14.0

16.0

0 100 200 300 400 5000.0

5.0

10.0

15.0

20.0

25.0

30.0(-22 µm)

(-210 µm)

• Flux data indicate that reducing HTM thickness should increase H2 flux.

Chemical Stability of ANL-3 Membranes

15



0.0

1.0

2.0

3.0

4.0

5.0

0.0

2.0

4.0

6.0

8.0

0 1 2 3 4 5

900°CThickness - 0.20 mm

H2

Flux

(cm

3 /min

-cm

2 )

H2

Flux

(scf

h/ft2 )

Time (h)H

2Fl

ux (c

m3 /m

in-c

m2 )

0.0

1.0

2.0

3.0

4.0

5.0

0.0

2.0

4.0

6.0

8.0

0 20 40 60 80 100

900°CThickness - 0.20 mm

H2

Flux

(scf

h/ft2 )

Time (h)

Feed Gas: 61.3% H2, 8.2% CH4, 11.5% CO, 9.0% CO2, 10% He (Ambient Pressure)

(A) (B)

• ANL-3 membranes are chemically stable in CO, CO2, CH4 and H2S.

Feed Gas: Same as (A) + 100 ppm H2S

Chemical Stability of ANL-3 Membrane

16



H2

Flux

(cm

3 /min

-cm

2 )

H2

Flux

(scf

h/ft2 )

Time (h)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

0 100 200 300 400 500 600 700 8000.0

0.8

1.6

2.4

3.2

4.0

4.8

5.6

ANL-3e (Thickness ≈ 0.20 mm)Feed Gas: 400 ppm H2S, 73% H2, Balance He at ambient pressure

• ANL-3 HTM appears to be stable in 400 ppm H2S.

17



H2 Flux Values of Conventionally Sintered and Hot-Pressed ANL-3a Membranes

H2

Flux

(cm

3 /min

-cm

2 )

Feed Gas: 80% H2/Balance N2 at ambient pressureThickness ≈0.40 mm

Time (h)

0.0

2.0

4.0

6.0

8.0

10.0

0.0

4.0

8.0

12.0

16.0

600 700 800 900

Hot-PressedSintered

H2

Flux

(scf

h/ft2 )

• Flux depends on microstructure.

• Hot-pressing gives higher flux.

18



Microstructures of Conventionally Sintered and Hot-Pressed ANL-3a Membranes

Conventionally Sintered1500°C/10 h/air

Hot-Pressed1250°C/25 min/N2

• Hot-pressing gives much finer microstructure due to shorter processing time at lower temperature.

19



Improvements in H2 Flux (2004 vs. 2005)Feed: 100% H2 at ambient pressure

ANL-3e Thin Film on Porous Support

• Thin film HTM gives significantly higher flux.

Time (h)

H2

Flux

(scf

h/ft2

)

H2

Flux

(cm

3 /min

-cm

2 )5.0

10.0

15.0

20.0

25.0

30.0

35.0

400 500 600 700 800 900 1000

16.0

24.0

32.0

40.0

48.0

56.0

64.02005Thin film on porous support

Thickness - 15 µm

2004Self-supported diskThickness - 22 µm

20



H2 Flux (ANL-3e) vs. ∆pH21/2

0

50

100

150

200

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0

Thin film/porous support-900°CThin film/porous support-500°CSelf-supported disk-900°CSelf-supported disk-600°C

0

50

100

150

200

250

300

350

400

2005

2004

∆pH21/2 ≡ pH2(feed) − pH2(sweep) (atm1/2)

H2

Flux

(cm

3 /min

-cm

2 )

H2

Flux

(scf

h/ft2 )

• Flux >400 scfh/ft2 can be achieved at feed pH2 ≈300 psi.

21



Progress in Membrane Development at ANLTemperature: 900°C, Feed Gas: 100% H2 at ambient pressure

Year

H2

Flux

(scf

h/ft2 )

H2

Flux

(cm

3 /min

-cm

2 )

• HTM development has shown steady progress.

Focused on reproducibility and chemical stability

0.0

5.0

10.0

15.0

20.0

25.0

30.0

35.0

0.0

8.0

16.0

24.0

32.0

40.0

48.0

56.0

64.0

1998 1999 2000 2001 2002 2003 2004 2005

22



FE Reviewers’ Comments• FY 1999: Increase flux to >1 cm3/min-cm2 (2 scfh/ft2).

- Increased flux of ANL-2a HTM to >2 cm3/min-cm2 (4 scfh/ft2).

• FY 2000-2001: Enhance performance of mixed conductors.- Studied doping and surface modifications to increase flux.- Studied chemical stability in CO2-containing atmospheres.

• FY 2002: Increase flux to >10 cm3/min-cm2 (20 scfh/ft2).- Increased flux of ANL-3a HTM to ≈20 cm3/min-cm2 (40 scfh/ft2)

at 900°C.

• FY 2003-2004: Study chemical stability of membrane.- Studied effects of CO, CO2, CH4, and H2S on HTM performance.- Showed ANL-3e is stable for >700 h in 400 ppm H2S at 900°C.

• FY 2005: Reduce operating temperature to 500°C.- Increased flux of ANL-3e to ≈20 cm3/min-cm2 @ 500°C.

23



Future Work• Test long-term chemical stability of selected HTMs in

atmospheres typical of coal gasifiers.• Evaluate HTM microstucture before and after long-term tests

in coal-gasifier-type atmospheres.• Continue fabricating/testing thinner HTMs to maximize

hydrogen flux.• Evaluate HTM mechanical properties (fracture strength,

creep) before and after exposure to hydrogen.• Continue developing new HTMs to reduce cost, increase

flux, and improve mechanical/chemical stability.• Transfer membrane technology through industrial

collaborations.

24



Summary• Developed dual-phase dense membranes that nongalvanically

separate hydrogen.

• Flux is proportional to the difference in the square-root of hydrogen partial pressure on two sides of membrane.

• Highest H2 flux (≈66 scfh/ft2 at 900°C and ≈42 scfh/ft2 at 500°C) was measured on ≈15-µm-thick HTM using feed of 1 atm H2.

• Flux >400 scfh/ft2 can be achieved with hydrogen partial pressure ≈300 psi in feed gas.

• Short-term measurements showed stable flux in feed streams that contained CO, CO2, CH4, H2O, and H2S.

• Flux was stable for ≈700 h in feed stream with 400 ppm H2S.

25



SupplementalSlides

26



R&D 100 Award in 2004

• Argonne’s HTM was recognized as one of 2004’s 100 “most significant technological developments.”

27



ContributorsANL

Steve DorrisTae Lee

Jack PiccioloLing Chen

Sun-Ju SongChendong Zuo

Mark Mika

NETLRichard Killmeyer

Mike Ciocco

28



Hydrogen Safety - Hazards

• The most significant hydrogen hazard associated with this project is the potential for fire and/or explosion resulting from the release of hydrogen into the ambient atmosphere.

• Release of hydrogen from the reactor is the most serious hazard, because hydrogen in the reactor is at high temperature and/or pressure.

• Hydrogen flame can be especially hazardous to personnel in vicinity, because it is invisible.

29



Hydrogen Safety - Hazard Mitigation• The approach to deal with a potential release of

hydrogen to the atmosphere is:- Structural

- Sensors detect hazardous H2 concentrations.- Vented gas cabinets shut down H2 flow if H2

concentration or H2 flow exceeds safe levels.- Reactor has secondary containment purged by N2.

- Procedural- Establish standard operating procedures.- Perform safety review of equipment/procedures.- Permit work only by authorized personnel.

development of dense ceramic membranes for hydrogen … file3 pioneering science and technology doe...

Documents