1

New Amine-Containing Membranes for CO2 Capture

Kai Chen, Yang Han, Witopo Salim, Zi Tong,

Dongzhu Wu and W.S. Winston Ho

William G. Lowrie Department of Chemical & Biomolecular Engineering

Department of Materials Science and Engineering

The Ohio State University, Columbus, Ohio, USA

The Carbon Management Technology Conference 2017 Houston, TX, July 17 – 20, 2017

2

Outline

• CO2 Capture from Flue Gas in Coal-

and/or Natural Gas-fired Power Plants

• CO2 Capture from <1% CO2

Concentration Sources, e.g.,

– Residual flue gas after the primary CO2 capture system

– Coal-mine gas streams

3

● Coal-fired power plants

40% of global CO2 emission

Remain as major energy supply

● Membranes for CO2 capture from flue gas

System compactness

Energy efficiency

Operational simplicity

Kinetic ability to overcome thermodynamic

solubility limitation

Introduction

4

Challenges and Goals

● High CO2 Permeance

Reduce Membrane Area

Thin Membrane Thickness Needed

● High CO2/N2 Selectivity

Provide High CO2 Purity

High Selective Membrane Material Needed

>700 GPU CO2 permeance and >140 CO2/N2

selectivity at 57oC To meet the DOE target < $40/tonne CO2 (2007 $)

𝑷𝑪𝑶𝟐=

𝑱𝑪𝑶𝟐

𝒑𝑪𝑶𝟐,𝒇  − 𝒑𝑪𝑶𝟐,𝒑

𝜶𝑪𝑶𝟐 𝑵𝟐 =𝑷𝑪𝑶𝟐

𝑷𝑵𝟐

5

Amine Polymer Layer Contains Mobile and Fixed Carriers: Facilitated Transport

CO2 CO2

Membrane

CO2+

CO2

CO2 CO2

Mobile

Carrier

Facilitated Transport

Feed Side Permeate Side

Non-Reacting

Gas: N2 N2

Physical Solution-Diffusion

Mobile

Carrier

CO2 Mobile

Carrier

CO2 Mobile

Carrier

6

Amine-Containing Polymer Membrane Structure

Non-woven fabric

Porous PES or PSf

Amine layer

Simplicity of Membrane for Low Cost

7

Amine Layer Composition

Fixed-site carrier:

Polyvinylamine (PVAm)

Mobile carrier:

Piperazine glycinate (PG)

O

NH2 O-

NHNH2+

NH2 NH2 NH2 NH2 NH2

Surfactant:

Sodium dodecyl sulfate (SDS)

Na+

CH3 O S

O

O

O-

8

High-Molecular-Weight PVAm

Synthesis

Polymer

Monomer

Conc.

(wt.%)

Initiator/Monomer

Weight Ratio

3 wt.%

Polymer

Solution

viscosity (cp)

Weight

Average MW

PVAm (a) 30 0.5/100 486 719,000

PVAm (b) 40 0.14/100 1,400 1,200,000

Lupamin® N/A N/A 50 340,000

• MW measured by Dynamic Light Scattering (DLS)

• Lupamin® contains ~66% sodium formate

9

a) High viscosity - Decrease coating solution concentration

- Allow for preparation of thinner membrane

- Improve adhesion to support – less defects

b) Low salt amount - Improve membrane stability

High-Molecular-Weight PVAm Synthesized

Viscosity:

3 wt.% commercial PVAm: 50 cp

3 wt.% high MW PVAm: 450 – 1950 cp

10

Composite Membrane Synthesized Selective Amine Polymer Layer on PES Support

Selective layer

Selective layer = 165 nm

PES support

11

Membrane Scale-up: Continuous Roll-to-Roll Fabrication Machine at OSU

Oven

Unwind

Roll

Coating

Knife

Rewind

Roll

12

𝒍 × 𝝆𝒅𝒓𝒚 = 𝟎. 𝟓 × 𝒄 × 𝝆𝒔𝒐𝒍 × 𝒍𝒈𝒂𝒑

Tuning Membrane Thickness

𝒍: dry membrane thickness (µm)

𝝆𝒅𝒓𝒚: density of dry membrane (g/cm3)

𝝆𝒔𝒐𝒍: density of coating solution (g/cm3)

c: total solid concentration of coating

solution (by weight)

𝒍𝒈𝒂𝒑: gap setting of coating knife (µm)

13

• A thinner thickness with a higher web speed for a

given coating solution supply rate

Coating Technique Modification: Insert

After Modification Before Modification

Substrate

Motion

Unwind

Roll

Coating Knife Coating Knife

Insert

Substrate

Motion

Unwind

Roll

14

Coating knife

assembly

configuration

Polymer

concentration

(wt.%)

Viscosity of

Coating

Solution

(cp)

Web

speed

(ft/min)

Selective

amine layer

thickness

(nm)

Insert with

1-inch flow

channel height a

1.8 870

3 175

4 135

5 115

2.4 940 3.25 210

3.5 195

3 1180 3.5 185

Insert with

2-inch flow

channel height a

1.5 690

1 230

2 165

4 135

1.8 870 2 220

3 155

3.4 1430 1.6 210

2 178

Thin 14”-Wide Amine Coatings on Substrates

a 2.5 mil of flow channel opening

15

Membrane Element Fabrication

Spiral-Wound Membrane Element Element Rolling Machine

Membrane Module

Sweep

Inlet

Sweep

Outlet

Feed Inlet

Feed Outlet

16

Good Membrane Module Stability Obtained

Wu et al., JMS in press (2017)

0

100

200

300

400

500

600

700

0

200

400

600

800

1000

1200

1400

1600

0 50 100 150 200

CO

2/N

2 s

ele

cti

vit

y

CO

2 p

erm

ea

nc

e (

GP

U)

Time (hour)

Simulated Flue Gas

16% CO2 + 64% N2 + 17%H2O +

3% O2 and 3 ppm SO2

Change

CO2 con.

17

Process Proposed for CO2 Capture from Flue Gas in Coal-Fired Power Plants

Air

2200 psia

CompressorCoal

Combustor

Membrane

Module 1

Membrane

Module 2

Vacuum

Pump

Treated

Flue Gas

Coal Feed

Air Stream

With CO2

Flue Gas

6% CO2

0.2 atm

20% CO2

*

• Proposed membrane process does not require cryogenic distillation (compared to competition)

* Air Sweep first used by MTR

1.4% CO2

>95% CO2

Ramasubramanian et al., JMS, 421-422, 299 (2012)

Amine

PEO

18

SO2 Polishing & Membrane Process

• Absorption into 20 wt% NaOH Solution – Polishing step based on NETL baseline document

• Estimated to be ~ $4.3/tonne CO2 (in 2007 $, 6.5% COE increase)

– Non-plugging, low-differential-pressure, spray baffle scrubber

– High efficiencies (>95%)

Baghouse

Air

2200 psi

Compressor

Membrane Module 1

Membrane Module 2

Vacuum

pump

Treated flue gasFGD

Primary Air Fans

Induced Fan Draft

Forced Fan Draft

Steam to Turbine

Feed Water

Coal Feed Bottoms Ash

PulverizedCoal

Boiler

SO2 Polishing

19

Techno-Economic Analysis

• Basis: In 2007 dollar • DOE target: $40/tonne CO2 set for 2025

• Include membrane module installation cost

• Also include 20% process contingency

• Membrane Results at 57oC • Amine-containing membrane with >140 selectivity

key for stand-alone membrane process

• Membrane process does not require cryogenic

distillation

• PEO-containing membrane with 2000 GPU CO2

permeance and 20 selectivity

20

0

50

100

150

200

250

300

350

400

200 400 600 800 1000 1200 1400

CO

2/N

2 S

ele

cti

vit

y

CO2 Permeance (GPU)

Scale-up amine/PES

Spiral-wound Module: Scale-up Amine/PES

Sprial-wound Module: Scale-up Amine/ZY/PES

Plate-and-frame Module: Scale-up Amine/PES

Lab Amine/PES

Lab Amine/Lab PES

Lab Amine/ZY/PES

Scale-up Amine/ZY/PES

Scale-up membrane

(14-inch wide)

Prototype module

(2.25” dia. by 14” long)

Composite Membranes Containing Amine Cover Layer: Simulated Flue Gas at 57oC

$40/ tonne CO2

$42/ tonne CO2

21

Process Proposed for CO2 Capture from <1% CO2 Sources

•Proposed membrane process does not require cryogenic distillation (compared to competition)

Membrane

Module 1

Membrane

Module 2

Vacuum

Pump 1

Vacuum Pump 2

<1% CO2

Feed Gas Treated Gas

Compressor

152 bar

>95% CO2

<0.1% CO2

0.2 atm

0.2 atm

Location of Proposed Technology in

Coal-fired Power Plant

22

Pulverized

Coal

Boiler

Coal

Feed Bottom Ash

Feed Water

Steam to

Turbine

Primary

Air Fans

Secondary

Air Fans

Primary

CO2

Capture

System

Baghouse

Induced

Draft Fans

FGD

Vacuum

Pump 2

Membrane

Module 1

Membrane

Module 2

Vacuum

Pump 1 Compressor

152 bar

<1% CO2

Feed Gas <0.1% CO2

Treated Gas

>95% CO2

Performance Improves as CO2 Conc. Reduces

100

300

500

700

900

500

750

1000

1250

1500

0 5 10 15 20 25

CO

2/N

2 s

ele

cti

vit

y

CO

2 p

erm

ea

nc

e (

GP

U)

Feed CO2 Concentration (%) 23

24

High-Level Techno-Economic Calculations

• Calculated Cost Results • 32.1 tonne/h of CO2 captured from 1% CO2 source

• $107.8 million bare equipment cost Membrane 34%, blowers and vacuum pumps 62%, others 4%

• 1.76 ¢/kWh (1.24 ¢/kWh capital cost, 0.22 ¢/kWh fixed cost,

0.26 ¢/kWh variable cost, and 0.04 ¢/kWh T&S cost) COE = 8.09 ¢/kWh for 550 MW supercritical pulverized coal power plant

• $302/tonne capture cost ($17.6/MWh × 550 MW/(32.1 tonne/h))

• 21.8% Increase in COE (1.76/8.09 = 21.8%)

• Basis: Membrane Results at 57oC • 982 GPU & 211 Selectivity for 1% CO2 concentration feed gas

• 806 GPU & 173 Selectivity for 20% CO2 concentration feed gas

• Include Membrane Module Installation Cost and 20% Process

Contingency

• In 2011 dollar: NETL Case 12 of Updated Costs (June 2011

Basis) for Selected Bituminous Baseline Cases

25

Effect of CO2 Permeance on Techno-economic Analysis Results

18

20

22

24

26

28

250

270

290

310

330

350

700 900 1100 1300 1500

CO

E i

nc

reas

e (

%)

Cap

ture

co

st

($/t

on

ne)

CO2 permeance (GPU)

CO2/N2 selectivity = 200

982 GPU

26

Summary • CO2 Capture from Flue Gas – High-Molecular-Weight PVAm Membrane Synthesized

– Composite Membranes Synthesized in Lab

+ ~800 GPU with >200 Selectivity at 57°C

– Membrane Scaled up Successfully

– Scale-up Membrane Promising for Meeting DOE Cost

Target of $40/tonne CO2 (in 2007 dollar) for 2025

• CO2 Capture from <1% CO2 Conc. Sources – Membrane showed 982 GPU with 211 CO2/N2 selectivity

obtained at 57oC using 1% CO2 concentration feed gas

+ 806 GPU with 173 selectivity obtained using 20% CO2

conc. feed gas due to carrier saturation phenomenon

– Performance improves as CO2 conc. reduces

– High-level techno-economic analysis conducted + Capture cost of ~$302/tonne CO2 (in 2011 $)

+ 21.8% increase in COE

27

Acknowledgments

Norman N. Li Richard Song – Nanoporous Membrane Support

José Figueroa – Strong Efforts & Helpful Inputs for CO2-Sel. Membranes

Yuanxin Chen Kartik Ramasubramanian Varun Vakharia Lin Zhao – CO2-Selective Membranes

BASF in Wyandotte, MI – Free PES Samples

Purolite Corporation, Bala Cynwyd, PA – Free Ion-Exchange Resin Samples

28

• U.S. Department of Energy - DE-FE0007632 and DE-FE0026919

• Ohio Development Services Agency - OER-CDO-D-15-09 for Continuous Membrane

Fabrication Machine

Acknowledgments for Financial Support

29

Decreasing Emissions Preserves Environment