Membrane Reactor for

Hydrogen Production
Ashok Damle
Jim Acquaviva
Pall Corporation
November 17, 2008
This presentation does not contain any proprietary or confidential information
AIChE 2008 Annual Meeting, Philadelphia, PA
Photo courtesy of
Pall Corporation
2
Contributors & Acknowledgments
Pall Corporation
Scott Hopkins
Daniel Henkel
Rick Kleiner
Rajinder P. Singh
Hongbin Zhao
Keith Rekczis
Chuck Love
Kevin Stark
Colorado School of
Mines
J. Douglas Way
Oyvind Hatlevik
RTI International
Carrie Richardson
DOE (EERE)
Sara Dillich
3
Presentation Outline
Drivers for Hydrogen Production and CO
2
capture
Process intensification / Membrane reactor concept
Status of Pd-alloy composite membrane at Pall
Inorganic substrate development
Composite Pd-alloy membrane development
Membrane reactor model simulations
WGS Membrane reactor experimental studies
Palls capabilities and future activities
4
Hydrogen Economy and Production
Two major drivers for hydrogen production
Hydrogen as energy carrier Transportation,
Power/heat generation, and Chemical production
Pre-combustion CO
2
capture and hydrogen production
has potential to reduce GHG emissions
Hydrogen Production
Can be produced from multiple pathways natural
gas, coal, biomass and renewables
Near term hydrogen production from Natural Gas
Longer term hydrogen production from Coal and
renewable energy sources (biomass, solar, wind)
5
Conventional Hydrogen Production
Exhaust
Natural
Gas
Air
Syngas
Generator
WGS
PSA
Hydrogen
Product
800
o
C
400
o
C
Water
Residual Gas
Current State:
> 90 % of H2 is
produced from
NG by this
process
Very efficient
on large scale
Future State:
Combining hydrogen generation and separation (process
intensification) can potentially reduce capital and
operating cost of hydrogen production at various scales
6
WGS Membrane Reactor Process
WGS Membrane
Reactor
800
o
C
400
o
C
Hydrogen
Product
Water
Exhaust
Natural
Gas
Air
Syngas
Generator
Residual Gas
Increased conversion due to equilibrium shift
Compact system, smaller footprint
Simpler operation and lower operating/energy costs
Need compressor for high pressure hydrogen product
7
Membrane Reformer Process
Natural
Gas
Air
Membrane
Reformer
Hydrogen
Product
Water
Exhaust
CO
2
600
o
C
Steam
Residual
Gas
Compact unit, smaller footprint Lower capital cost
Milder conditions
Increased hydrogen yield Greater energy efficiency,
Less steam Lower cost of H
2
production
Need high temperature inorganic membrane for H
2
separation
Efficiency
improvement
through
process
intensification
8
1. Lee, D., Zhang, L., Oyama, S. T., Niu,
S., and R. F. Saraf, J. Membr. Sci., 231,
117(2004).
2. Kajiwara, M., Uemiya, S., Kojima, T., and
E. Kikuchi, Catal. Today, 56, 65(2000).
3. DeVos, R. M. and H. Verweij, Science,
279, 1710(1998).
4. Hassan, M. H., J. D. Way, P. M. Thoen,
and A. C. Dillon, J. Membr. Sci. , 104,
27(1995).
5. Polymer line from :
Robeson, L. M., J. Membr. Sci., 62,
165(1991).
6. Wu, J. C. S. et al., J. Membr. Sci., 77,
85(1993).
7. Hatlevik, ., Gade, S. K., Keeling, M. K.,
Thoen, P. M. and J. D.Way, "Palladium
and Palladium Alloy Membranes for
Hydrogen Separation and Production:
History, Fabrication Strategies, and
Current Performance," submitted to
Separation and Purification Technology,
Sept. 2008.
Why Palladium Membrane ?
1
10
100
1000
10
4
10
5
10
-9
10
-8
10
-7
10
-6
10
-5
0.0001
10
-9
10
-8
10
-7
10
-6
10
-5
0.0001
Polymeric Membrane Materials
Inorganic Membrane Materials
H
2
/
N
2
I
d
e
a
l

S
e
p
a
r
a
t
i
o
n

F
a
c
t
o
r
H
2
Permeance (mol/m
2
.s.Pa)
CSM PdAu
(1)
(2)
(3)
(4)
(6)
CSM Pd
(7)
(7)
Graph courtesy of Prof. Doug Way and CSM group
9
Pd-alloy membrane development
Self supporting membrane structures
Need membrane of sufficient thickness for
structural integrity and strength e.g. tubes or flat
sheets > 25 m
Expensive, Niche applications small H
2
purifiers
Composite membrane structures
Thin films on substrates
Substrate provides structural integrity and strength
Deposition of thin Pd-alloy films by various
techniques ~ 1 5 m
Better seals for High T High P applications
Lower cost thin Pd layer, less membrane area
10
Components of a Composite Membrane
3) Pd alloy membrane
Functional layer provides for
gas separation
Critical features: thickness,
alloy composition, durability
and number of defects
1) Porous stainless steel
Provides mechanical support that
can withstand the operating
conditions of the process
Critical features: permeability,
weld configuration, mechanical,
thermal and chemical
compatibility
2) Diffusion barrier
Enables formation of functional
layer
Critical features: surface
properties, material, gas
permeability, number of defects
Pd-alloy membrane development at Pall
3
2
1
Excellent adhesion to zirconia layer, uniform thickness,
and surface contour following of Pd-alloy metal film
11
Ceramic / PSS composite substrate for Pd alloy membranes
PSS
Medium
Ceramic
Coating
Its all about the substrate
Porous stainless steel tube with
ZrO
2
ceramic coating: Extensive
development work done to
optimize the composite structure
and surface properties to enable
formation of a high quality Pd alloy
or other functional layer.
Pd-alloy membrane development at Pall
All welded design
No polymer seals, Higher temp. capabilities
Thermal expansion
Uniform thermal expansion with the housing
and module components
Cost
All metal design with welded fittings, allows
for direct welding to a tube sheet. This
eliminates the need for intricate sealing
mechanism and reduces overall module cost
12
Durability : Pall Gas/Gas separation supports have been
exposed to multiple thermal cycles with no detrimental
effects to the composite structure
Ceramic layer stable and maintains adhesion to metallic
substrate through thermal cycles
Composite tube with 310SC can be used up to 550
o
C in
pure H
2
and up to 400
o
C in air or inert gases
Characterization Data of Composite Support
First bubble in IPA is > 30 psi
Air permeability @ 1000 cc/min ~ 7 psi
Zirconia coating pore structure is 70 nm
Base metallic tube pore structure is 2 microns in
average
Pd-alloy membrane development at Pall
13
Typical Performance Data
H
2
flux 150-180 8CFH/ft
2
H
2
/Ar 8F 10,000 - 20,000
Under test conditions
Pure gas permeation
400
o
C
Feed 20 P8G
Permeate atmospheric
15 cm
2
active surface area
75 cm
2
active surface area
Pd-alloy membrane development at Pall
14
Membrane Durability in Thermal Cycling
Thermal Cycle: Air Air Argon Hydrogen Air
Temperature (C): 25 400 400 400 25
Pressure (psig): 0 20 20 20 0
15
Components of a Gas/Gas Separation Module
Pressure vessel
with fittings
Internal
hardware
Non-porous
end fitting
Porous
substrate
Weld
Membrane tube
sub-assembly*
* Pd alloy membrane not shown, typically on the OD of the tube
Welds
16
16
Membrane Reactor Model Simulations
Model Assumptions
Temperature and total pressure constant on both sides
Reaction kinetics faster than hydrogen permeation
Feed side is in dynamic equilibrium
Hydrogen flux determined by local driving force
Permeate Hydrogen Membrane
Fuel Gas
with Steam
Residual Gas
Fuel Reforming/WGS Catalyst
H
2
H
2
17
WGS Membrane Reactor Experimental Results
Demonstrated > 80% Net Recovery of Hydrogen with >80% CO conversion
20
30
40
50
60
70
80
90
100
0 10 20 30 40 50 60 70 80 90 100
Net Hydrogen Recovery, %
C
O

C
o
n
v
e
r
s
i
o
n
,

%
Methane
Reformate
Feed Gas (Dry)
H
2
- 75.2%
CO - 15.6%
CO
2
- 7.1%
CH
4
- 2.1%
Steam:CO = 1.2:1
Temperature - 375
o
C
Feed Pressure - 100, 150 psig
High Temp. Fe-Cr WGS Catalyst
(Sud-Chemie - Shiftmax 120)
Model Prediction
(Fast Kinetics)
Experimental Data
Equilibrium CO Conversion
at Feed Gas Conditions
100 psig
150 psig
Experiments conducted by Damle at RTI International Fuel Cell Seminar 2007
18
Predicted Methane Conversion increase
@ T 600 C, Steam:C::2:1, P-100, 250 psig
Effect of Pressure Greater H
2
recovery and yield
in spite of unfavorable equilibrium
19
Membrane Reactor Performance
@ T 600 C, Steam:C::2:1, P-100 psig
Recovery of sensible and combustion heat of Residual Gas
Net heat requirement analysis
20
Membrane Reactor Performance
@ T 600 C, Steam:C::2:1, P-250 psig
Effect of Pressure Greater H
2
partial pressure
Less Membrane Area requirement
21
Membrane Reactor Performance
@ T 600 C, Steam:C::3:1, P-100 psig
Higher Steam:C ratio Greater H
2
partial pressure
Less Membrane Area requirement
Relatively small energy penalty
22
Membrane Reactor Performance
@ T 550 C, Steam:C::2:1, P-100 psig
Lower Temperature Lower H
2
partial pressure less conversion
(strong effect) Low hydrogen recovery
23
Components of Economic Analysis
Economic Model
Membrane
Reactor Model
Energy Model
H2 Cost
($/gge)
P
r
o
c
e
s
s

D
a
t
a
Process
Model
C
o
m
p
r
e
s
s
i
o
n
C
o
s
t
Hydrogen Permeate Flow
S
u
r
f
a
c
e
A
r
e
a
Membrane
Pilot Plant
Performance
Measurements
Compressor
OPEX & CAPEX
Membrane
OPEX & CAPEX
PSA
OPEX & CAPEX
NG reformer
24
Potential Benefits of Membrane Reactor
Basis: 100 Kg/day (1650 SCFH)
Membrane module area ~ 10 ft
2
Membrane Module cost with DOE Target ~ $7,500
Cost of metal ~ $ 235 (October 2008 prices)
Capital cost reduction by replacing PSA/WGS ~ $ 40,000
DOE H
2
A Forecourt Model Capital cost portion ~ $3.06/kg H
2
Potential reduction in capital cost contribution ~ $0.36/kg H
2
Penalty for 50% additional compressor cost ~ $ 0.08/kg H
2
Additional benefits not yet quantified
Increased hydrogen yield
Reduced operational cost
Reduced energy costs
25
Future R&D Needs Reduce Cost
Capital Cost
High flux membrane to reduce required membrane area and pressure
vessel size
High efficiency modules to maximize use of membrane area
Commercial scale manufacturing process for Pd alloy membranes
Process integration to reduce balance of plant cost
Process intensification (ex: membrane reactors) to minimize catalyst
and hardware cost
Operating Cost
High separation factor membranes that maximize H
2
recovery
Process integration to minimize the energy penalty for CO
2
capture
Maintenance Cost
Durable palladium alloys that can tolerate severe process conditions,
abrupt startups & shutdowns, and contaminants in feed streams
26
Research Development
Scale-up of Metal Tube Technology
Commercialization
27
Questions ?