John Jensvold
Generon IGS, Inc., Pittsburg, CA
www.rti.org 3/27/2009
Project Overview
• DOE/NETL Cooperative Agreement # DE-NT0005313
– DOE Project Manager: José Figueroa
– RTI Project Manager: Lora Toy
• Period of Performance
– October 1, 2008 – September 30, 2010
• Project Team
– RTI (Prime – Technology developer, membrane evaluation, management)
– Arkema Inc. (Polymer synthesis and development)
– Generon IGS, Inc. (Membrane module development and fabrication)
– ARCADIS, Inc. (Test skid operation)
– U.S. EPA (Test facility provider) 3
www.rti.org 3/27/2009
CO2 Capture Membrane Process for Power
Plant Flue Gas (CCM) Project
Overall Project Objective
Develop an advanced polymeric membrane-based process that can be cost-effectively and reliably
retrofitted into existing pulverized coal (PC)-fired power plants to capture ≥90% CO2 from plant’s
flue gas at 50-60 °C with <35% increase in Cost of Electricity.
www.rti.org 3/27/2009
Membrane Approach
High pressure Low pressure
p p
1 2
N2 Advantages
CO2
• Passive separation
SO
2
H 2O – Inherently energy-efficient
www.rti.org 3/27/2009
Challenges of CO2 Capture Application with
Membrane Process
Example Single-Stage Membrane Process
(~500-MW plant; ~800,000 acfm flue gas;
CO2 permeance ~ 100 GPU; CO2/N2 selectivity ~ 35)
• Low CO2 concentration (~13-15%)
in flue gas Blower/
Compressor
Membrane
separation
Coal pfeed unit 90% CO2
• Low flue-gas pressure (~15 psia) Combustion
unit
Flue gas
Cooling
(~3 bars) removal To
stack
13% CO2
Air (O2 + N2)
in N2
• Large flue gas volumes (~15 psia;
55-60 °C;
>800,000 acfm) Water/ Ppermeate
www.rti.org 3/27/2009
Project Approach
Three Parallel Efforts Roadmap of Technical Approach
• Membrane materials development
– Novel fluorinated polymers
– Next-generation, high-flux polycarbonate
– Flux / Selectivity
– Chemical resistance
• Membrane module design and
development
– Hollow fiber dimensions
– Module gas flow distribution
– Module sealing
– Module fabrication
• Process design and development
– Process variables
– Process design options
– Process integration
– Technoeconomic analysis
7
www.rti.org 3/27/2009
Specific Objectives and Task Structure
Specific Objectives Project Tasks
• Develop two or three new chemistries/structures of
fluorinated polymer membrane materials that have • Task 1. Synthesize Novel Polymers /
Prepare Membrane Films
– high CO2 permeance [300-3,000 GPU targeted]
– high CO2/N2 selectivity [30-50 selectivity targeted] • Task 2. Characterize Permeation Properties
– excellent chemical stability to moisture, SO2, and of Membrane Films
NOx
• Task 3. Produce and Characterize Membrane
• Identify and develop power- and cost-effective CO2 Hollow Fibers
capture membrane process design/integration
strategies with refined membrane CO2 permeance • Task 4. Make and Characterize Prototype
Hollow-Fiber Membrane Modules
and selectivity targets.
• Develop and fabricate improved membrane hollow • Task 5. Demonstrate Membrane Modules in
fibers and membrane module designs designed to Field Test
handle large flue-gas flow rates and high CO2 • Task 6. Perform Process Design / Technical
permeate flows and Economic Analysis
• Demonstrate CO2 capture membrane • Task 7. Manage Project / Prepare Reports
performance/reliability in field test with real coal-
fired process flue gas.
1 GPU = 1 × 10-6 cm3(STP)/(cm2·s·cmHg) 8
www.rti.org 3/27/2009
Membrane Materials Development
Arkema
www.rti.org 3/27/2009
Membrane Materials Development
Arkema
Approach #1 Approach #2
• Copolymerize bulky hexafluoropropylene • If crystallinity reduction not sufficient, then
(HFP) monomer into VDF backbone. copolymerize new non-fluorinated
– Bulky –CF3 disrupts crystallization, monomers (e.g., vinyl esters, acrylic
reducing crystallinity (down to <2%) esters, etc.) with VDF to improve CO2
– Should increase gas transport in transport.
membrane
F F F
H Approach #3
C C C C
H F F CF3 • Design composites (co-continuous
VDF HFP phases) of low crystallinity and chemically
modified PVDF with other compatible but
H F H F F F H F not miscible polymers (i.e., PEO) to
C C C C C C C C
further enhance CO2 transport without
compromising chemical resistance of
H F H F F CF3 H F PVDF.
VDF-co-HFP
10
www.rti.org 3/27/2009
Next-Generation, High-Flux Polycarbonate (PC) Membrane
Generon
Relative CO2/N2
CO2 flux selectivity
Generon Standard
1 20-30
PC Membrane
Generon High-Flux
4.0-4.5 30-35
PC Membrane
1 GPU = 1 × 10-6 cm3(STP)/(cm2·s·cmHg)
www.rti.org 3/27/2009
Membrane Module Design/Development
• Two widely used module types for
industrial gas separation membranes
– Hollow fiber
– Spiral wound Generon membrane module for air separation
• Hollow-fiber module type selected
End Plate
– Lower module cost per membrane area Epoxy Tube Sheet
Enriched
Nitrogen
– Substantially higher membrane packing Support Core Product
density Gas
www.rti.org 3/27/2009
Module Fabrication
Generon Facilities
www.rti.org 3/27/2009
Process Design/Integration Parameters
To Be Considered
• Both membrane area and cost vastly affected by
membrane properties, process parameters, and process
scheme.
– Compression/Vacuum
– Multi-step design
– Multi-stage design
14
www.rti.org 3/27/2009
Example Multistage Membrane Process Scheme:
Preliminary Simulation Results
1 bar, 23% CO2
9
8
2.4% CO2 Flue-gas flow = 800,000 scfm
3 bars CO2/N2 selectivity = 50
5.7% CO2 B4 CO2 permeance = 1,000 GPU
2
RTI
1 bar B5
17% CO2 1 B6
B1
10
RTI
3 1.8% CO2
11 B2
7 5
B7 0.33 bar
52% CO2
3 bars
18% CO2 B3
B8
12
RTI
10 bars
• ~89% CO2 removal 52% CO2
• Total Membrane Area = 3.3 × 106 m2 4 81% CO2 purity
• Process cost analysis still to be done 6
1 bar 15
www.rti.org 3/27/2009
Single-Stage Membrane Process Simulation:
Effect of Pressure Ratio [Constant permeate pressure 15 psia]
40 80
3
Increasing
P ratio
5
35 10 70
15
Membrane area x 10^6 m2
CO2 purity, %
25 50 CO2 Perm = 100 GPU
N2 Perm = 2.86 GPU
20 40 Permeate pressure = 15 psia
Single Stage
15 30 Shell Side Feed
10 Increasing 20
P ratio
5 10
0 0
0 10 20 30 40 50 60 70 80 90 100
CO2 removal, %
www.rti.org 3/27/2009
Single-Stage Membrane Process Simulation:
Effect of CO2 Permeance [Constant selectivity]
40 40
100 GPU
35 Increasing
CO2 permeance 300 GPU Flue gas FR = 800,000 scfm
500 GPU CO2 mole fraction = 0.13
Membrane area x 10^6 m2
30 35
CO2/N2 selectivity = 35
1000 GPU Fp = 45 psia, Pp = 15 psia
25
CO2 purity, %
Single Stage
Shell Side Feed
20 30
Increasing
15 CO2 permeance
10 25
0 20
0 10 20 30 40 50 60 70 80 90 100
CO2 removal, %
www.rti.org 3/27/2009
Single-Stage Membrane Process Simulation:
Effect of CO2/N2 Selectivity [Changing CO2 Permeance]
40 40
Increasing
36 selectivity 38
Membrane area x 10^6 m2
CO2 purity, %
Fp = 45 psia, Pp = 15 psia
24 32
100 GPU/Sel 35 Single Stage
20 300 GPU/Sel 105 30 Shell Side Feed
500 GPU/Sel 175
16 28
1000 GPU/Sel 350
12 Increasing
26
selectivity
8 24
4 22
0 20
0 10 20 30 40 50 60 70 80 90 100
CO2 removal, %
Increasing selectivity 10 times reduces membrane area only from 25 × 106 to 19 × 106 m2 for 90% CO2 removal.
Increasing selectivity enriches the permeate CO2 purity, for example, by 5% with 10 times increase in selectivity
for 90% CO2 removal.
18
www.rti.org 3/27/2009
Process Design Analysis
22
• Estimate targets for membrane properties Increasing
20 fiber ID
– CO2 flux (permeance)
– CO2 selectivity 18
100 µm
– Fiber dimensions 16
125 µm
• Analyze pressure drop issues associated 14
0 0.2 0.4 0.6 0.8 1
with high flue-gas flows and high CO2 Fiber length
www.rti.org 3/27/2009
Membrane Permeation Testing
Pressure
• Membranes gauge
• Gases Permeate
20
www.rti.org 3/27/2009
Generon Loop-Cell Module:
Pure- vs. Mixed-Gas Selectivities
100
Generon loop-cell module-3
T = 25 C
Selectivity
2 2
SO /N
2 2
10
www.rti.org 3/27/2009
Field-Test of CO2 Capture Membrane Modules
www.rti.org 3/27/2009
Anticipated Challenges
23
www.rti.org 3/27/2009
Project Timeline
24
www.rti.org 3/27/2009
Path Forward
www.rti.org 3/27/2009
Thank you
Questions?
26
www.rti.org 3/27/2009