Pilot Test Results from a PolarisTM

Membrane 1 MWe CO2 Capture System

Richard Baker, Carlos Casillas, Ken Chan, Brice Freeman,

Don Fulton, Pingjiao Hao, Jenny He, Jurgen Kaschemekat,

Jay Kniep, Jennifer Ly, Tim Merkel, Vincent Nguyen, Zhen Sun,

Xuezhen Wang, Xiaotong Wei, and Steve White

Carbon Management Technology Conference

Houston, TX

Wednesday, July 19, 2017

MTR Introduction

2

• Privately-held, 60 employees mostly located in

Newark, California

• Started in 1982 funded by U.S. SBIR grants

• Commercial products in petrochemical, natural gas

and refinery industries

• Provides complete turn-key solutions with over 200

membrane systems installed worldwide

• 15 person R&D group with expertise in membrane

materials/formation and process design

Background: MTR CO2 Capture Process

3

Benefits of selective recycle:

• Increases CO2 concentration going to the capture step, and

• Reduces the fractional CO2 removal required by the capture step

20% 10%

2%

8%

65%

U.S. Patents 7,964,020 and 8,025,715

(1) Membrane

performance

studied at NCCC

(2) Impact of

CO2 recycle

studied at B&W

MTR CO2 Capture Development Timeline

Feasibility study (DE-NT43085) • Sweep concept proposed

• Polaris membrane conceived

APS Red Hawk NGCC Demo • First Polaris flue gas test

• 250 lb/d CO2 used for algae farm

APS Cholla 0.05 MWe Demo (DE-NT5312) • First Polaris coal flue gas test

• 1 TPD CO2 captured (50 kWe)

NCCC 1 MWe Demo (DE-FE0005795) • 11,000 hours of 1 TPD system operation

• 1,500 hours of 1 MWe (20 TPD) system operation

2006 2008 2010 2012 2014 2016 2018 2020

TRL6 TRL7 TRL8 TRL5 TRL4

Hybrid Capture (DE-FE13118) • Membrane-solvent hybrids with UT-Austin

Low Pressure Mega Module (DE-FE0007553) • Design and build a 500 m2 optimized module

Future 10 MWe Large Pilot

TRL3

B&W Integrated Test

5

Pros and Cons of a Membrane

Post-Combustion Capture Process

Benefits:

• No hazardous chemical handling, emissions, or disposal issues

• Not affected by oxygen, SOx or NOx

• Water use lower than other technologies (recovers H2O from flue gas)

• No steam use → no modifications to existing boiler/turbines

• Near instantaneous response; high turndown possible

• Modular and compact; relatively simple installation/operation

• Particularly well-suited for partial capture (40-60%)

Challenges:

• Very low partial pressure driving force favors high permeance membranes

• Unknown impact of real flue gas on membrane-module lifetime

• Module pressure drop and flow distribution issues

6

PolarisTM Membranes Continue to Improve

0

10

20

30

40

50

60

70

0 1,000 2,000 3,000

CO2/N2

selectivity

CO2 permeance (gpu)

PolarisTM

1st generation

(commercial scale)

Commercial

CO2 membranes

Polaris

2nd generation

(pilot scale)

Polaris

advanced

(lab scale)

• In addition to lowering

costs, these

improvements are

important to shrink the

size of the capture

system

1 gpu = 10-6 cm3(STP)/(cm2 s cmHg)= 3.35 x 10-10 mol/(m2 s Pa)

7

0.05 MWe (1 TPD) Testing at NCCC

• Field laboratory that allows lifetime evaluation (>11,000 hours cumulative), and validation testing of new membranes

• Many lessons learned (i.e., ammonium sulfate deposition) applied to scale up

• System is testing vacuum and air sweep membrane steps

• Sized to capture 1 ton CO2/day using commercial sized module

• System was in operation from Spring 2012 through Summer 2015

8

MTR system

• System showed stable performance

over 1,500 hours of operation,

including demonstration of

prototype low-pressure drop plate-

and-frame sweep modules

1 MWe (20 TPD) System at NCCC

• System installation completed Aug 2014

• Operation during January – June 2015

• Decommissioned in July 2015

9

1 MWe System Shows

Stable Performance

• System operated in

slipstream mode

(no recycle to

boiler) with varying

ambient conditions

(sub-freezing in

January to >95°F in

June)

• Stable

performance,

reaching up to 90%

capture

• System goes from

cold start to steady

state in ~15

minutes Run time (hours)

CO2

capture

rate

(%)

Figure data from NCCC campaign PO3 (May to July 2015)

PO-3 run time (hours)

CO2

capture

rate

(%)

New Modules Demonstrate Improved

Pressure Drop Performance

10

0

0.5

1

1.5

2

2.5

3

3.5

4

900 1,000 1,100 1,200 1,300 1,400 1,500 1,600

Sweep-sidepressure dop

(psi)

Sweep flowrate (lb/h)

Spiral with

flue gas

Plate-and-frame

with flue gas

Plate-and-frame

lab data

• Field data is consistent

with lab results, and

confirms much lower air

sweep pressure drop in

new modules

• At full scale, the

difference in pressure

drop amounts to

savings of about 10

MWe

B&W Studies of CO2 Recycle Impact

on Boiler Performance

Phase I – CFD modeling

• B&W modeled 2 boiler configurations (radiant boiler firing bituminous coal

and SWUP firing PRB coal) and 2 sweep recycle cases (constant

secondary air flow and constant stoichiometry)

• Main conclusion of modeling study: secondary air laden with CO2 appears

feasible as a retrofit in either of the boiler configurations examined if

oxygen mass flow to boiler is fixed

Phase II – Pilot testing

• B&W’s SBS-II 1.8 MWth pilot boiler operated with CO2-laden combustion air

• Two coals evaluated: a western sub-bituminous coal and a highly volatile

bituminous coal

• O2 content of windbox air varied from 21% to 16% through CO2 dilution

• Monitored flame stability, length, and shape; unburned combustibles in fly

ash, and furnace exit gas temperature

• Radiant furnace and convective pass heat absorptions were measured

• Boiler efficiencies for air and sweep firing were determined

11 11

B&W Boiler Study Highlights

12 12

• Stable and attached flames with air (21%

O2) and CO2-enriched air (16-18% O2)

• CO2-enriched flame was less luminous

than air-fired case

• Lower furnace heat absorption but higher

convection pass/air heater heat transfer for

CO2-enriched operation relative to air

• For bituminous coal, 30% lower NOx

emissions with CO2-enriched air

• No burner modifications necessary

• Net reduction in plant efficiency of ~0.75%

at 18% O2

Flame image from combustion

of PRB coal with air (21% O2)

Flame image from combustion of

PRB coal with CO2–enriched (18% O2)

13

B&W Pilot Boiler / Membrane Schematic

MTR Skid During Transport and

Installation at B&W – May 2016

14 14

Skid arriving at B&W

Installation of 2nd floor

MTR Skids at B&W’s

SBS-II Research Facility

15 15

Main skid and smaller low-pressure

drop sweep module anchored to

foundation

Main skid Sweep module

Sample Results from B&W

Integrated Tests

16

20% 10%

2%

8%

65%

2 – 6.7%

7.3 – 17.4%

14.3 – 20.6%

6.2 – 10.8%

Oxygen at

Boiler Windbox:

17.0 – 21%

Membrane-Boiler Integrated Test Plan

• 5 weeks of testing on natural gas, Powder River Basin

(PRB) Coal, and Eastern Bituminius Coal

• 90% capture achieved and a variety of partial capture

conditions

• CO2 content of flue gas increased as expected in

simulations

• Boiler flame was stable allowing a full battery of stream

conditions and boiler efficiency measurements to be

conducted (analysis is ongoing)

1 MWe System Sweep Flow Rate Parametric

Results from Integrated Tests

17

50

60

70

80

90

100

0 500 1,000 1,500 2,000 2,500

CO2

capture rate

(%)

Sweep flow rate (lb/h)

Feed = 30 psia

0

20

40

60

80

100

0 500 1,000 1,500 2,000 2,500

Sweep-step

CO2 capture

rate

(%)

Sweep flow rate (lb/h)

Feed = 30 psia

Influence on the Overall CO2 Capture Rate Influence on Sweep Step CO2

Capture Rate Efficiency

Integrated Boiler/Membrane Systems

Transition Response

18

0

3

6

9

12

15

18

21

0

1000

2000

3000

4000

5000

6000

7000

8:10 8:20 8:30 8:40 8:50 9:00 9:10 9:20 9:30 9:40 9:50 10:00

Gas

Co

nce

ntr

atio

n (

Vo

l. %

, dry

)

Gas

Flo

w R

ates

(lb

/hr)

11/3/2016

Fresh Air Intake Flow

Sweep Gas Flow

Convection Pass Flow

Windbox O2

Convection Pass O2

Convection Pass CO2

0

3

6

9

12

15

18

21

0

1000

2000

3000

4000

5000

6000

7000

12:00 12:07 12:14 12:21 12:28 12:36 12:43 12:50

Gas

Co

nce

ntr

atio

n (

Vo

l. %

, dry

)

Gas

Flo

w R

ates

(lb

/hr)

11/16/2016

Fresh Air Intake Flow

Sweep Gas Flow

Convection Pass Flow

Windbox O2

Convection Pass O2

Convection Pass CO2

Integrated Boiler/Membrane Systems

Transition Response to E-Stop

19

0

3

6

9

12

15

18

21

0

1000

2000

3000

4000

5000

6000

7000

10:45 10:55 11:05 11:15

Gas

Co

nce

ntr

atio

n (

Vo

l. %

, dry

)

Gas

Flo

w R

ates

(lb

/hr)

9/28/2016

Fresh Air Intake Flow

Sweep Gas Flow

Convection Pass Flow

Windbox O2

Convection Pass O2

Convection Pass CO2

B&W’s Analysis of CO2 Recycle Impact

on Boiler Operation

• Furnace heat absorption is

lower resulting

• “Furnace” refers to the radiant

heat transfer section of the

boiler upstream of the tube

banks in the convection pass.

• Convection pass heat

absorption is higher

• Convection pass outlet heat

flux is higher

• Air heater heat absorption is

higher

• Air heater flue gas outlet heat

flux is higher

• Total heat absorption is slightly

reduced

20

Coal 30P M1 & M2 Coal 27P M2 Only Bit 32P 07 Bit 30P 06

20-Oct-16 18-Oct-16 3-Nov-16 2-Nov-16

Test Duration (h:mm) 7:00 7:15 6:00 3:04

PRB PRB Bit. Bit.

Load (MW) 1.5 1.4 1.6 1.5

FEGT (°C) 1,179 1,259 1,175 1,254

Convection Pass Exit Temperature (°C) 397 380 408 388

Air Heater Exit Temperature (Flue Gas) (°C) 217 210 222 212

53% 0% 75% 0%

Furnace Absorption (MW) 0.52 0.66 0.40 0.77

Convection Pass Absorption (MW) 0.96 0.91 1.11 0.90

Convection Pass Outlet Heat Flux (MW) 0.50 0.43 0.54 0.44

Total Heat Absorption (MW) 1.62 1.68 1.65 1.77

Air Heater Absorption (MW) 0.19 0.16 0.18 0.14

Air Heater Outlet Heat Flux (Flue Gas) (MW) 0.31 0.27 0.36 0.30

Date

Fuel

Membrane Secondary Air Ratio

Test Name

Summary

• CO2 capture membrane performance continues to improve and

has been validated on the 0.05 MWe slipstream system with over

11,000 hours of runtime at NCCC

• 1 MWe small pilot operation at NCCC was completed in 2015.

Testing successfully demonstrated optimized modules (low Dp,

low cost) with over 1,500 hours of runtime

• 1 MWe small pilot was successfully integrated with the B&W

research boiler for five weeks of integrated testing with CO2

recycle to the boiler in late 2016

• The integrated membrane-boiler field test experimentally

validated simulated system performance

• Boiler flame was stable throughout parametric testing allowing a

full battery of stream conditions and boiler efficiency

measurements 21

Acknowledgements

22 22

• U.S. Department of Energy,

National Energy Technology Laboratory

– José Figueroa

– Mike Mosser

• Southern Company Services (NCCC)

– Tony Wu

• Babcock & Wilcox

– Hamid Farzan

– Jennifer Sivy

– Andrew Mackrory