a castrogiovanni-acent-ices.pdf

A premier aerospace and defense company

A High Efficiency g yInertial CO2 Extraction System

(ICES)(ICES)

Dr. Anthony Castrogiovanni, Dr. Vladimir Balepin,Andrew Robertson, and Bon Calayag

Presented at the

NETL CO Capture Technology MeetingNETL CO2 Capture Technology MeetingPittsburgh, PAJuly 11, 2012

1


Company Backgrounds

• ATK is a leading aerospace & defense contractor• ACENT is a small business dedicated to applying expertise in aerospace and defense to clean energy challenges

• Founded in 2007, ACENT is developing

• ATK is a leading aerospace & defense contractor• ATK GASL in Ronkonkoma, NY operates the ATK Center for Energy and Aerospace Innovation

• Expertise and research interests include :• Aerospace propulsion Founded in 2007, ACENT is developing

technologies in CO2 capture, algal biomass, hydrogen vehicles, and enhanced oil recovery

• Aerospace propulsion• Carbon capture• Hydrogen fueled vehicles• Clean coal technologies• Oil recovery solutions• Oil recovery solutions

• ICES utilizes some methods developed under

2

a DOE SBIR with ACENT


Project Overview

Funding Summary:ARPA e: $ 2 693 KARPA-e: $ 2,693 KATK and ACENT Cost Share: $ 632 KNYSERDA (New York State) $ 200 K ( ) $TOTAL $ 3,525 K

Project Performance Dates:Phase 1: July 2010 – March 2011 (completed)Ph 2 J l 2011 J 2013 ( i )Phase 2: July 2011 – June 2013 (ongoing)

Project Participants:Project Participants:Alliant Techsystems (ATK)ACENT Laboratories LLC

3

WorleyParsons


Overall Project Objectives

• Demonstrate proof-of-concept of aero-thermodynamic CO condensation and separationCO2 condensation and separation

• Develop and benchmark analysis tools with experimental d t t bldata to enable:

• Scaling of demo system to power plant size

• Projection of economics in terms of COE and parasitic loads

• Provide basis for next-phase slip-stream testing with real flue gas

• Minimize flue gas pressurization requirements

• Maximize CO2 capture (>90% goal)

4

Maximize CO2 capture ( 90% goal)


ICES Technology Fundamentals

• Pulverized coal power plant flue gas contains ~16% CO2 in gaseous form at low pressure

• In ICES we compress flue gas to a moderate level andgas to a moderate level and use the low temperature created by supersonic expansion to freeze theexpansion to freeze the CO2 in the flow

• ICES uses turning induced in the flow to inertiallyseparate the solid particles from the gas stream

• We capture and collect the CO2 (as dry ice) and then process using a self-pressurization system exploiting power plant waste heat

5

system exploiting power plant waste heat


ICES on a P-T Diagram – Supersonic Expansion

12.01000

Isentropic Expansion of 16% CO2 in Nitrogen Relative to Phase Diagram of CO2

Liquid Phase

10.0100

Triple Point

Region of incipient

8.010

mbe

r

[psia]

pcondensation Gas Phase

4.0

6.0

0.1

1

Mach Num

Pressure [

Evolution of Partial Pressure of CO2 during Isentropic Expansion in Supersonic Nozzle(p0=30psia, T0=520ºR)

Solid Phase

2.00.01

(p0 30psia, T0 520 R)

0.00.001

0 100 200 300 400 500 600

T [R]

6

Temperature [R]


ICES Integration in PC Plant

18Flue Gas C l

Flue Gas Compressor

TBD Coolant

1719

22

Cooler Compressor

ICES Unit

21

22ICES Unit

Water20

21

7


ICES System Schematic

Compressor/HEX16

T 330K

18 19

22

HEX

Solid CO2

Capture DuctFluegas

Supersonic expansion

T=330KP=1.05 atm15% CO2

Subsonic diffusion

HEX

Cyclone separator (slip gas + CO2)

Flue gas from FGD T=300K

P=1.03 atm T=300KP=2.0 atm

T=400KP=1 atm<1.5% CO2

X

X

X X

T=298KP 150 t

21

CO accumulation

P=150 atm>99.5% CO2

CO2 accumulation & self pressurization system

8

X

Water Drain20


ICES Economic Impact

• ICES operating costs are driven by flue gas pre-compressionP f t P /P• Pressure recovery factor = P22/P19

• Low CapEx/OpEx combined with low power consumption result in a

j t d t f l t i it i fprojected cost of electricity increase for CO2 capture just over 1/3 that of the amine process

• Compression to 2,250 psi from low grade waste heat (constant volume heat addition to solid). Cost is limited t CAPEX + t dito CAPEX + energy to move media.

Metric ICES AmineCOE % increase 35% 81%Parasitic Load 12.5% 21.5%

9

Cost per ton of CO2 avoided US$ 27 US$ 68


Energy Consumption

Process Minimum Energy [kJ/kg CO2]

ICES[kJ/kg CO2]

Amine[kJ/kg CO2]

S ti 175 683*Separation -175 -683* -CO2 Compression -247 ~68** -Total -422 -751 -1,506Total 422 751 1,506

* Pre-compression of flue gas to 2 bar (absolute)** + Approximately 760 kJ/kg of low grade waste heat used to compress CO2from solid phase to 2 250psiafrom solid phase to 2,250psia

Fixed volume at ambient

~760 kJ/kg latent sensible heat

Supercritical CO2

pressure partially filled

with solid CO2 at 200 ºF

Initial Volume % filled with solid

Pressure at 70ºF [psia]

2

-200 ºF filled with solid 70 F [psia]

60% 3,00065% 6,000

10

Compression energy is nearly “economically free” but it is not “thermodynamically free” i.e. this energy would otherwise be wasted


ICES Plant Integration and Footprint

An ICES system sized for 545MW-equivalent flue gas contains twelve 60” ICES units (flue gas compression not shown)

L= 183 ftL= 183 ftW= 60 ftH = 70ft

ICES is projected to have a significantly smaller footprint and complexity compared to competing CO2 capture technologies and hence significantly

11

compared to competing CO2 capture technologies and hence significantly lower capital and maintenance costs


Key Advantages of ICES over other options

• No moving parts (after start)

N h i l / dditi th bl di• No chemicals/additives or other consumable media

• No refrigeration expense – low temperatures from supersonic expansionexpansion

• Inexpensive construction (concrete, sheet metal)

• Small footprint• Small footprint

• ICES units in test are equivalent to 0.3-0.6 MW slip stream

• The latest unit (0 3 MW) is 24” x 24” x 3”• The latest unit (0.3 MW) is 24 x 24 x 3

• Small size enables distributed deployment for other process applications in the petroleum and chemical industriespp p

• Availability of “cold sink” in solid CO2 accumulated

12


ICES Development Challenges

• Development of optimized supersonic contour to maximize particle size/migration and minimize pressuremaximize particle size/migration and minimize pressure losses

• Minimization of “slip gas” that is removed with solid CO2

• CO2 purity unknowns - other flue gas impurities thatCO2 purity unknowns - other flue gas impurities that condense will be removed with the CO2

S lid CO t/ lf i ti• Solid CO2 management/self pressurization

• This really is rocket science….but once the design is y gcomplete, it is easy and inexpensive to build and operate

13


Project Status – Phase 1

• Phase 1 and early Phase 2 efforts focused on an axisymmetric system with swirl

ICES test bench

14


Axisymmetric System Results

Phase 1 data showed good CO2 condensation and apparent, but erratic migration due to unsteady and separated flowerratic migration due to unsteady and separated flow

Outer (glass) wall)

CenterbodySolid CO2

15


We recently changed to a 2D version of ICES

• Better aerodynamic performance (lowerperformance (lower losses)

• Easier to fabricate d t t

Aluminum plenum chamber and throat

Aluminum plate reinforces plenum chamber

and test• No swirl vanes to

induce Clear acrylic sidewallsturbulence/wake

effects• Simpler capture duct

Clear acrylic sidewallsSupersonic flowpath

components made on 3D printer (ABS)

without swirlVacuum interface flange (steel)

16


Gen5 Test Article Design – ATK Installation

24” Vacuum Pipe

17


Test Data Comparison to CFD – Static Pressure

0.45

0.5

Test Data - Top SurfacesT t D t B tt S f

0.35

0.4

cham

ber)

Test Data -BottomSurfacesCFD Simulation - Top SurfacesCFD Simulation - Bottom SurfacesMach 1 9

0 25

0.3

tion

(Ps/

Pc Mach 1.9Mach 2.1Mach 2.75

0 15

0.2

0.25

ress

ure

Rat

0.1

0.15

Stat

ic P

r

Nucleation Begins

0

0.05

-5 0 5 10 15 20 25 30

18

X Position in Rig Coordinates


Laser Particle Imaging Diagnostic

Test Article

High Speed HD Laser Sheet Image cameraImage

19

Nd:YAG Laser


1, 10, and 100 Micron Particle Trajectories

1 μm1 μm

10 μm

100 μm100 μm

20

At 10 microns+ particles separate and coalesce allowing for a slender capture slot


Three modes in typical ICES test

CO2 migration toward lower wall evident

1) Air flow only

wall evident

2) 30 psi, 20% CO2

Optical and CO2 sampling results show condensation as e pected b t less than desired migration e ident Particle si e

3) 30 psi, 10% CO2

21

expected, but less than desired migration evident. Particle size does not appear large enough in these tests


Condensate Particle Size Control

• Classical nucleation theory provides basis for predicting critical condensate cluster size and subsequent growth rate:

• Both are strong function of the saturation ratio (S) = partial pressure of vapor/saturation pressure (pv/ps)

0 8

1

1.2

Diameter

• Maximum initial cluster size near S=1• Desirable to grow these clusters versus

0 2

0.4

0.6

0.8

s Critical Cluster D nucleating new ones at higher S

• Nozzle contour shape needs to be optimized for this purpose

‐0.4

‐0.2

0

0.2

Dim

ension

less • Need to further increase residence time

of flow in this critical region• Increasing scale toward power plant

i ill h l

22

0.900 0.950 1.000 1.050 1.100

Saturation Ratio

size will help


Current Plans

• Remaining portion of Phase 2

I ti t fl di ith lid CO ( lf t d) d th• Investigate flow seeding with solid CO2 (self generated) and other media to promote large particle formation (ongoing)

• Update contour to further optimize particle sizeUpdate contour to further optimize particle size

• Integration of capture duct to remove CO2

• Integration of diffuser to return flow to atmospheric pressure with• Integration of diffuser to return flow to atmospheric pressure with minimal losses

• “Phase 3”

• Ideal next step desired is a slip stream test, e.g. at the National Carbon Capture Center (NCCC)

23


Accomplishments to date

• Three ICES configurations have been developed and tested to datetested to date

• Demonstrated clean nozzle flow with low apparent losses (to be verified with later diffuser tests)

• Demonstrated supersonic condensation with someDemonstrated supersonic condensation with some migration

Pl i l t i ti l i t hi d i d• Plans in place to increase particle size to achieve desired migration performance

24


Acknowledgements

ARPA E:

Dr. Karma Sawyer, Dr. Scott Litzelman, Dr. Daniel Matuszak, Dr. Mark Hartneyy , , , y

NYSERDA:

Dr. Barry Liebowitzy

ATK & ACENT Labs:

Dr. Pat Sforza, Troy Custodio, Vincenzo Verrelli, Skip Day, Dean Feola, Ed Mihalik, Florin Girlea, Kirk Featherstone, Fred Gregory, Randy Voland

25


BACKUP

26


Schematic of Condensation Process

S = ps / pv , where pv is the partial pressure of the vapor and ps is the vapor saturation pressure at the temperature of the system.

27

a castrogiovanni-acent-ices.pdf

Documents