High Pressure Ethanol Reformingfor Distributed H2 Production

S. Ahmed and S.H.D. Lee

Bio-Derived Liquids to Hydrogen Distributed Reforming Working Group

Kickoff Meeting October 24, 2006, Baltimore, Maryland

Background

� Compressed hydrogen is needed for storage and delivery – Compressed gas tanks at 5,000-10,000 psig – Metal hydride (150-450 psig)

� Pressurized reformate is needed for many purification and enrichment paths – Membrane separation

• Hydrogen permeation, e.g., Pd-membrane • By-product removal, e.g., COx (CO & CO2)

– Pressure Swing Adsorption (PSA)

� Gas compression is energy intensive

2Work sponsored by U.S. Department of Energy, Hydrogen, Fuel Cells and Infrastructure Technologies Program

40

The energy required to compress hydrogen to 6,000 psigcan be as high as 31% of the LHV of hydrogen

18.519.8

21.3

23.5

31.5

27.5

20

25

30

35

(%

)C

ompr

essi

on L

oss

/ LH

V

5-Stage Intercooled Compressor Compressor Efficiency: 70% Mechanical Efficiency: 97%

Electric Motor Efficiency: 90% FinalPressure: 6000 psi

Electricity Generation Efficiency = 40%

15

10

1 2 3 4 5 6 7 8 9 10

Initial Pressure of Hydrogen, atm

� Starting with a pressurized hydrogen / reformate stream lessens the energy required by the compressor

Note: Hydrogen from SMR-PSA processes may be available at 200 psig

3Work sponsored by U.S. Department of Energy, Hydrogen, Fuel Cells and Infrastructure Technologies Program

(SMR- Steam Methane Reforming; PSA – Pressure Swing Adsorption)

Steam reforming of liquid fuels generates a pressurized reformate with little energy penalty � Injecting liquid (ethanol + water) feeds into a high pressure reactor requires little energy � Hydrated ethanol is less expensive than fuel-grade ethanol

– Available upstream before water separation (distillation, adsorption, membrane)

6

5

4

� Steam reforming of ethanol at elevated pressure does not favor hydrogen yield, 3

however – Equilibrium predicts increasing 2

methane yields with increasing pressures 1

0 0 1000 2000 3000 4000 5000

H2

CO2 CO

CH4

T = 700°C, S/C = 3

Pressure, psia

Yiel

d, m

ol/m

ol o

f EtO

H

4Work sponsored by U.S. Department of Energy, Hydrogen, Fuel Cells and Infrastructure Technologies Program

The negative effect of pressure on hydrogen yield can be offset with higher temperature and steam-to-carbon ratio

46 S/C = 3, P = 2000 psia

35

H2

CO2

CH4

CO

T=700°C, P = 2000 psia

Yiel

d, m

ol/m

ol o

f EtO

H

Yiel

d, m

ol/m

ol o

f EtO

H

3

1

H2

CO2 CH4

CO 1

4

2 3

2

2 1

0 0 500 600 700 800 900 1000 1 2 3 4 5 6 7 8 9

Temperature, °C Steam-to-Carbon Molar Ratio

� High pressure-temperature combinations add to hardware cost � High steam-to-carbon ratio reduces overall process efficiency

5Work sponsored by U.S. Department of Energy, Hydrogen, Fuel Cells and Infrastructure Technologies Program

Simulated process efficiencies approach 70%at a steam-to-carbon ratio of 5

Burner

AirHydrogen

equilibrium Equilibrium Reactor �C2H5OH + xH2O(l)

Ethanol+Water

Exhaust

Membrane CO2, CO, H2, H2O(g), CH4, CnHm, … Separator

�Chemcad simulated process based on Heat – steam-reformer at equilibrium Exchanger

– hydrogen separation with membrane • 90% hydrogen recovery

– combustion of raffinate to generate heat 100

– heat exchange to reformer feeds 90 – exhaust at 200°C

80

�Efficiency decreases with increasing S/C 70

60 � Various alternative and more detailed

system solutions need to be evaluated 50

Ideal Reaction

Simulated Process (SR at 2000 psi)

5 6 7 8 9 10 Steam-to-C Ratio

Effic

ienc

y%

of L

HV

6Work sponsored by U.S. Department of Energy, Hydrogen, Fuel Cells and Infrastructure Technologies Program

A micro-reactor apparatus is being used to generate yield data

� Rated for 1,000 psi, 900°C � 63.5-mm (0.25-in) ID reactor tube � 4 wt% Rh/ La-Al2O3

� Powder, 150-250 μm � 0.35 g of catalyst � 20-mm long-catalyst bed

7Work sponsored by U.S. Department of Energy, Hydrogen, Fuel Cells and Infrastructure Technologies Program

Vaporizer tests indicate the ethanol starts decompositionabove 440°C

1

CO

CH4

C2H4

C2H6

1000 psig S/C=12

1000 psig S/C=20

500 psig S/C=20

Exit

T =

440°

C

Exit

T =

470°

C

� At 1000 psig, S/C = 12-20 0.9 – BP = ~310°C

Con

cent

ratio

n, %

(N2-

free

) 0.8 � At 500 psig, S/C = 12-20

0.7– BP = ~255°C 0.6

0.5

0.4

0.3

0.2

0.1

0 250 350 450 550

Time, min

8Work sponsored by U.S. Department of Energy, Hydrogen, Fuel Cells and Infrastructure Technologies Program

The hydrogen yield with the nickel catalyst decreasedwith time, indicating deactivation of the base metal catalyst

� The nickel catalyst deactivated with time – Bed pressure drop increased – Carbonaceous deposits were

observed in the reactor

� The reactor was re-packed with a rhodium catalyst – 4 wt% Rh / La-Al2O3

� The rhodium catalyst performed better – Higher hydrogen yield – Maintained activity

Yiel

d, m

ol/(m

ol E

tOH

)

T=650°C, P=1,000 psig, S/C=12, GHSV=84,000 /hr 4.5

4.0

3.5

3.0

2.5

2.0

1.5

1.0

0.5

0.0

Rh - H2

Ni - H2

Ni - CH4

Rh - CH4

0 100 200 300 400Time, min

9Work sponsored by U.S. Department of Energy, Hydrogen, Fuel Cells and Infrastructure Technologies Program

The nickel catalyst yielded a considerable fraction ofcondensable products

� The Ni catalyst yielded more undesirable hydrocarbons

– Mass balance accounted for 76% of the carbon

• Condensable products and carbon deposits could not be factored in

� The Rh catalyst yielded more hydrogen and COx, the desired product species

Catalyst Ni Rh Temperature, °C 650 650 Pressure, psig 1000 1000 GHSV, /hr 83,000 83,000 Steam-to-Carbon 12 12 Product Yield

H2, mol/(mol EtOH) 2.4 3.9 CO, mol/(mol EtOH) 0.2 0.1 CH4, mol/(mol EtOH) 0.5 0.6 CO2, mol/(mol EtOH) 0.8 1.4 C2H4, mol/(mol EtOH) 0.006 ND C2H6, mol/(mol EtOH) 0.029 ND

Carbon Balance, % 76 101

ND – not detected

10Work sponsored by U.S. Department of Energy, Hydrogen, Fuel Cells and Infrastructure Technologies Program

Experimental data show multiple hydrocarbon species inthe product stream

� With increasing pressure 5 – Hydrogen yield decreases – Methane yield increases 4 – Ethylene is hydrogenated to

produce ethane

Yiel

d, m

ol/(m

ol E

tOH

)

/()

C2H6

C2H4

COx

H2

CH4

/

0.00

0.02

0.04

0.06

0.08

0.10

Yiel

d, m

olm

ol E

tOH

T=650°C, S C=6, GHSV=15,000 /hr, Rh catalyst

3

2

1

0

0 500 1000Pressure, psig

11Work sponsored by U.S. Department of Energy, Hydrogen, Fuel Cells and Infrastructure Technologies Program

The experimental hydrogen yield approaches equilibriumpredictions at 1000 psi

� At a constant GHSV, equilibrium is approached at higher pressures

– Faster reaction rates

Yiel

d, m

ol/(m

ol E

tOH

)

S/C=6, GHSV=15,000 /hr, Rh catalyst 6

5

4

4.26

0.76

0.09

2.17

0.94

0.71

0.24

0.64

20 psi-Equil

20 psi-Exp

1000 psi-Equil

1000 psi-Exp

3

2

1

0 H2 CO2 CO CH4

12Work sponsored by U.S. Department of Energy, Hydrogen, Fuel Cells and Infrastructure Technologies Program

The product species can be explained through acombination of reactions

1. Ethanol Dehydrogenation : C2H5OH = CH3CHO + H2

2. Ethanol Dehydration : C2H5OH = C2H4 + H2O 3. Dissociative Ads. of Water : C2H5OH + H2O = CH3COOH + 2H2

4. Acetaldehyde Dissociation : CH3CHO = CO + CH4

5. Ethanation : C2H4 + H2 = C2H6

6. Dissociation of Acetic Acid : CH3COOH = CH4 + CO2

7. Methane Steam Reforming : CH4 + H2O = CO + 3H2

8. Water Gas Shift : CO + H2O = CO2 + H2

13Work sponsored by U.S. Department of Energy, Hydrogen, Fuel Cells and Infrastructure Technologies Program

FY07 work directed at yield and kinetics of pressurized reforming

� Experimental yields and kinetics are being determined � Using a micro-reactor with a Pd-membrane separator � Hydrogen extraction exacerbates coking

Case 1: EtOH=1 H2O=3

Case 2: EtOH=1 H2O=3

Case 3: EtOH=1 H2O=4

St. Reformer

S/C=1.5

S/C=1.5 580 psi, 700°C H2=1.4, CH4=1.1, H2O, COx, No Carbon

H2=1.4

St. Reformer + H2 Extraction

580 psi, 700°C H2, CH4, H2O, COx, Carbon Formed

H2=1.7

St. Reformer + H2 Extraction

S/C=2 580 psi, 700°C H2=1.2, CH4=0.72, H2O, COx No Carbon

14Work sponsored by U.S. Department of Energy, Hydrogen, Fuel Cells and Infrastructure Technologies Program

How worthwhile is the pressurized steamreforming of bio-liquids? � We are currently seeking an answer for ethanol

– Experimental yields and kinetics are being determined • Using a micro-reactor with a Pd-membrane separator

– Reactor models will help extract kinetic information – Supported by new generation of catalysts

• Improve durability and reduce cost

� System model and analysis – Simple membrane reactor concept – Alternative purification / enrichment options – Go / NoGo determination on pressurized reforming with ethanol

15Work sponsored by U.S. Department of Energy, Hydrogen, Fuel Cells and Infrastructure Technologies Program

Acknowledgments

� R.K. Ahluwalia � M. Ferrandon � T. Krause

This work was supported by the US Department of Energy’s Hydrogen, Fuel Cells and Infrastructure Technologies Program in the Office of Energy Efficiency and Renewable Energy.

This presentation has been created by UChicago Argonne, LLC, Operator of Argonne National Laboratory (“Argonne”). Argonne, a U.S. Department of Energy Office of Science laboratory, is operated under Contract No. DE-AC02-06CH11357. The U.S. Government retains for itself, and others acting on its behalf, a paid-up nonexclusive, irrevocable worldwide license in said article to reproduce, prepare derivative works, distribute copies to the public, and perform publicly and display publicly, by or on behalf of the Government.

16Work sponsored by U.S. Department of Energy, Hydrogen, Fuel Cells and Infrastructure Technologies Program