1

Shiguang Li, Travis Pyrzynski, Naomi Klinghoffer, Timothy Tamale, James Aderhold, S. James Zhou, and Howard Meyer, Gas Technology Institute (GTI)

Yong Ding and Ben Bikson, Air Liquide Advanced Separations (ALaS)Katherine Searcy and Andrew Sexton, Trimeric Corporation (Trimeric)

DOE Contract DE-FE0012829

Pilot Test of a Nanoporous, Super-Hydrophobic Membrane Contactor Process for Post-

Combustion CO2 Capture

CO2 Capture Technology Project Review MeetingAugust 13 - 17, 2018, Pittsburgh, PA

2

2

Project overview

Performance period: Oct. 1, 2013 – June 30, 2019 Total funding: $13.7MM (DOE: $10.6MM, Cost share: $3.1MM) Objectives:

Build a 0.5 MWe pilot-scale CO2 capture system and conduct tests on coal flue gas at the National Carbon Capture Center (NCCC)

Demonstrate a continuous, steady-state operation

Goal: achieve DOE’s goal of 90% CO2 capture rate with 95% CO2 purity at a cost of $40/tonne of CO2 captured by 2025

Team: Member Roles• Project management and planning• Process design and testing• Membrane and module development• Techno-Economic Analyses (TEA)

NCCC • Site host

ALaS

3

What is a membrane contactor? High surface area membrane device that facilitates mass transfer Gas on one side, liquid on other side

Membrane does not wet out in contact with liquid Separation mechanism: CO2 permeates through membrane,

reacts with the solvent; N2 does not react and has low solubility in solvent

4

Technical challenges of applying HFMC to existing coal-fired plants Performance – Overall mass transfer

resistance consists of three parts Minimize each resistance

Module design and durability – Long-

Make membrane surface super hydrophobic Improve membrane potting to provide good

seal between the liquid and gas sides

Fouling – Flue gas contaminants and/orparticulates may affect performance Determine required pretreatments

Scale-up and cost reduction Make larger diameter modules

1𝐾𝐾 =

1𝑘𝑘𝑔𝑔

+1𝑘𝑘𝑚𝑚

+𝐻𝐻𝑎𝑎𝑎𝑎𝑎𝑎𝑚𝑚𝐸𝐸 � 𝑘𝑘𝑙𝑙

Overall mass transfer coefficient K (cm/s)• In the gas phase, kg• In the membrane, km• In the liquid phase, kl

Hadim: non-dimensional Henry’s constant E: enhancement factor due to reaction

HFMC = Hollow fiber membrane contactor

term membrane wetting in contact with solvent

5

5

PEEK ( ) characteristics and advantages of PEEK HFMC

Exceptional thermal & mechanical resistances

Surface modified to be super hydrophobic

PolymerTensile

modulus(GPA)

Tensile strength

(MPa)

Max service temperature

(°C)TeflonTM 0.4-0.5 17-21 250

Polysulfone 2.6 70 160

PEEK 4 97 271

+Hydrophilic Hydrophobic

Ethanol

Hollow fibers w/ high CO2 flux and packing density

PEEK = Polyether Ether Ketone

PEEK HFMC advantages (compared to conventional absorbers) High packing density results in over

100x increase in mass transfer coefficient, and thus much smaller equipment size

Reduction in weight for over 30% Reduction in footprint due to versatile

modular layout Easy scaleup by adding membrane

modules Flexibility: commercial solvent aMDEA

being used; advanced solvents can be used for additional savings

Reduction in solvent degradation due to an indirect contact of flue gas contaminants and solvent

6

6

Module scaled to 8-inch by ALaS and tested at GTI with aMDEA solvent using air/CO2 mixed feed

Intrinsic CO2 permeance: 2,000 GPU Improved mass transfer coefficient of 2.0 (sec.)-1 obtained in lab

CO2 capture testingGPU= Gas Permeation Unit, 1 GPU = 3.348 x 10-10 mol/m2/s/Pa

7

0.5 MWe pilot plant designed, constructed and installed at the NCCC

Plant constructed

Plant installed at the NCCC12 m (L) x 7.5 m (W) x 3.5 m (H)

GTI HFMC system(0.5 MWe)

NCCC PSTU system(0.5 MWe)

8

8

BlowerFilter

NCCC’s PC4

PEEK HFMC 0.5 MWe plant

T (oC) P (psig)40-60 1-10

Membrane absorber

Process description

ether ketone

SteamSteam

T (oC) P (psig)~130 1-50

Flash desorber

CO2(50 psig)

CO2(20 psig)aMDEA solvent

9

9

Initial tests with 4 modules and flue gas at NCCC indicates DOE’s technical target can be achieved

CO2 removal rate:

CO2 purity: > 98.6% CO2

0

200

400

600

800

0

20

40

60

80

100

0:00 1:12 2:24

Flue

gas

flow

rate

(lb/

h)

Perc

ent (

%)

Time (hour : minute)

Flue gas flow rate

CO2 removal rate

Inlet CO2 concentration (dry basis, vol.)

Outlet CO2 concentration (dry basis, vol.)

90% removal line

10

Issues observed: 1) water vapor capillary condensation in PEEK pores, 2) concentration polarization

0

100

200

300

0

10

20

30

40

50

60

70

80

90

100

0 5 10 15 20 25 30 35 40

Flow

rate

(lb/

h)

CO

2re

mov

al o

r con

cent

ratio

n (%

)

Duration (hours)

08/11 08/12 08/13 08/14 08/15 08/16

CO2 inlet concentration

CO2 removal rate

CO2 outlet concentration

Flue gas flow rate

Example of liquid side concentration polarization issue

Concentration polarization issue: higher CO2concentration in the fluid boundary layer (next to the fiber) relative to the bulk flow stream

11

11

Concentration polarization issue was resolved by decreasing aMDEA concentration

The concentration change (from 50 wt.% to 35 wt.%) moves in a good direction for reducing both liquid side concentration polarization and membrane wetting

40

44

48

52

56

20 30 40 50 60 70

Surf

ace

Tens

ion,

dy

nes/

cm

Temperature, C

30 wt%35 wt%40 wt%50 wt%

0

5

10

15

20 30 40 50 60 70

Visc

osity

, cP

Temperature, C

40 wt%

35 wt%

30 wt%

50 wt% (5.4 wt% CO2 loaded)

50 wt% (0 wt% CO2 loaded)

0

5

10

15

20 30 40 50 60 70

CO

2Sa

tura

tion

Load

ing,

wt%

Temperature, C

30 wt%

40 wt%50 wt%

35 wt%

12

Issues resolved, steady state performance achieved for a single module during 224-h continuous testing

20

40

60

80

100

120

140

160

180

0102030405060708090

100

0 50 100 150 200 250

Flue

gas

flow

(lb/

h)

Perc

ent (

%)

Run time (hours)

inlet CO2 conc. dry vol. %

outlet CO2 conc. dry vol. %

CO2 removal rate

Flue gas flow rate

13

13

Examples of parametric testing results

0.0

0.2

0.4

0.6

0.8

1.0

1.2

0%

20%

40%

60%

80%

100%

70 80 90 100 110 120

Flux

(no

rmal

ized)

CO

2 re

mov

al r

ate

Solvent Temperature (°F)

removal rateflux

0

0.2

0.4

0.6

0.8

1

1.2

0%

20%

40%

60%

80%

100%

110 120 130 140 150

Flux

(no

rmal

ized)

CO

2re

mov

al r

ate

Flue gas temperature (°F)

removal rateflux

0.0

0.2

0.4

0.6

0.8

1.0

1.2

80% 85% 90% 95% 100%

Flux

(nor

mal

ized

)

CO2 removal rate

Parameters investigated: Flue gas temperature Solvent temperature CO2 capture rate Flue gas feed pressure Solvent flow velocity

14

14

Continuous testing with 28 membrane modules performed during May-June 2018

Timeline• May 25-30 (0-133 h): testing with all 28 membrane modules (A-G clusters)• May 30-June 12 (133-430 h): testing with better performance clusters A, E, F

continued (clusters B, C, D, and G isolated during this period)

Integrated absorption/desorption worked properly during testing CO2 purity target met, with CO2 purity >99% during the long term testing

Solvent regeneration system reliable• Rich and lean solvent samples collected daily and the CO2 loadings analyzed

by NCCC’s lab indicate solvent regeneration worked as HYSYS predicted• Solvent analysis indicates solvent oxidation and thermal degradation was not

an issue during our continuous operation

Fresh solvent Used solventRatio of amine to activator (normalized) 1.00 1.04

Concentration of degradation products < 0.01 wt. % < 0.3 wt. %

Concentration of metals Below detection limit < 0.002 wt. %

15

15

Continuous testing with 28 membrane modules performed during May-June 2018 (Cont’d)

Membrane absorption: CO2 capture performance declined with time

Fault tree analysis (FTA) ongoing, two major issues identified

0

2

4

6

8

10

12

14

16

0 50 100 150 200 250 300 350 400 450

CO

2co

ncen

tratio

n (%

)

Duration (h)

Inlet CO2

Outlet CO2

Testing w/ 28 modules Testing w/ 12 modules

16

16

Issue 1: potential reproducibility of membrane module fabrication In order to evaluate the consistency of capture for the individual modules,

the temperature rise of the amine was measured for each module. The measured temperature rise varied from the expected value of ~22 °F down to 4 °F, indicating some modules were not functioning well

Approaches to resolve the issue ALaS to further improve membrane module fabrication GTI to conduct QA/QC tests (CO2 permeation and water flow ∆P tests) for

selecting membrane modules

05

1015202530

0 20 40 60 80 100 120Tem

pera

ture

rise

, F

Duration (hours)

Temperature rise for modules in A & B clusters, first 133 hoursA1 A2 A3 A4 B1 B2 B3 B4

17

17

Issue 2: potential partial blockage of hollow fibers

Clean cartridge Cartridge after May-June testing

Gas inlet Gas outlet Gas inlet Gas outlet

Gas inletGas inlet

Analysis indicates the white material is calcium sulfate (possibly from FGD system) or sodium sulfate (possibly from PSTU pre-scrubber system)

Approaches to resolve the issue Additional filtration before the membranes Add pre-scrubber as needed

18

Future plans

In this project

After this project

Further testingMarch 2019

Final TEAJune 2019

Complete FTA andresolve issues

November 2018

2008 2010 2012 20182014 20222020 2024 2026Year

Scal

e

2016 2028

PoroGenLab scale

19

19

Summary

Commercial 8-inch-diameter membrane modules with intrinsic CO2permeance of 2,000 GPU fabricated for pilot scale testing of the PEEK HFMC technology (preliminary TEA based on bench-scale field testing: PEEK HFMC costs 16% less than DOE Case 12)

0.5 MWe pilot plant designed, constructed, installed, and being tested at NCCC

Achieved steady state CO2 capture performance with single module during our 224-h continuous operation at NCCC

Continuous testing with 28 membrane modules did not match single module results

Fault tree analysis ongoing Some potential issues and approaches to resolve the issues identified Plan to resume testing after we resolve the issues

20

Acknowledgements

Financial and technical support

DOE NETL: Steven Mascaro, José Figueroa and Lynn Brickett Southern Company Services: National Carbon Capture Center

JIP Partners ALaS

Contract DE-FE0012829

21

Disclaimer

This presentation was prepared by Gas Technology Institute (GTI) as an account of work sponsored by an agency of the United States Government. Neither GTI, the United States Government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors herein do not necessarily state or reflect those of the United States Government or any agency thereof.