Control of Mercury Emissions

by Injecting Powdered Activated Carbon (PAC)

Michael D. Durham, Ph.D., MBA

ADA Environmental Solutions

8100 SouthPark Way B-2

Littleton, CO 80120

303 734-1727

Presentation to

Utility MACT Working Group

May 13, 2002

EPA, RTP, NC

Outline

ADA-ES DOE/NETL Hg Control Program

Background on PAC Injection Technology

Results from PAC with an ESP

Results from PAC with a FF

Conclusions and Future Plans

ADA-ES Hg Control Program

Full-scale field testing of sorbent-based mercury control on non-scrubbed coal-fired boilers

Primary funding from DOE National Energy Technology Laboratory (NETL)

Cofunding provided by:– Southern Company – Wisconsin Electric– PG&E NEG– EPRI– Ontario Power Generation– TVA– First Energy– Kennecott Energy– Arch Coal

Project Overview

Perform first full-scale evaluations of mercury control on

coal-fired boilers (up to 150 MW equivalent).

Evaluate effectiveness of sorbent-based Hg control

(activated carbon).

Test several different power plant configurations.

Document all costs associated with Hg control.

DOE/NETL Test Sites

Test Site Coal Particulate TestControl Dates

Alabama Power Bituminous HS ESP Spring

Gaston COHPAC FF 2001

Wisconsin Electric PRB Cold Side ESP Fall

Pleasant Prairie 2001

PG&E NEG Bituminous Cold Side ESP Summer

Brayton Point 2002

PG&E NEG Bituminous Cold Side ESP Fall

Salem Harbor 2002

Coal-Fired Boiler with Sorbent Injection and Spray Cooling

Sorbent Injection

Spray Cooling

H2O

Air

Ash and Sorbent

ESP or FF

Hg CEM

Semi-Continuous Mercury Analyzer

Chilled Impingers

Flue GasCVAA

Mass Flow Controller

Gold Trap

Waste

Heater

Micro controller with Display

Dry Air

Sampling Time Required

Vapor-Phase Mercury

Concentration (ug/m3)

Minimum Sample Time (min)

Noise level (ug/m3)

5 1 0.1

2.5 2 0.05

1 5 0.02

0.5 10 0.01

Comparison of OH and S-CEM*, Long Term Tests (10 lbs/MMacf)

Run Number Run 1 Run 2 Run 3 AverageDate 11/12/2001 11/13/2001 11/13/2001

S-CEM* OH S-CEM* OH S-CEM* OH S-CEM* OH

Inlet (micrograms/dncm) 13.5 15 13.7 18.3 14.3 19.1 13.8 17.4Outlet (micrograms/dncm) 4.8 4.0 5.1 5.0 5.4 4.7 5.1 4.7Removal Efficiency (%) 64.4% 73.4% 62.8% 72.8% 64.0% 75.3% 63.7% 72.9%

* S-CEM measures only gas phase mercury, average calculated over same time as OH tests

Capture of Vapor Phase Hg by Solid Sorbents

Mass Transfer Limits (getting the Hg to the sorbent)

– Removal increases with particle concentration

– Produces percentage removal independent of concentration

– Particle control device (FF vs ESP) is a critical parameter

Sorbent Capacity to hold Hg depends upon:

– Sorbent characteristics

– Temperature

– Mercury concentration

– Concentrations of SO3 and other contaminants

Equilibrium Adsorption Capacities at 250°FUpstream and Downstream of SO3 Injection

0

1000

2000

3000

4000

5000

6000

7000

8000

9000

10000

FGD Carbon Norit Insul Carbon P4 Ash

Eq

uili

bri

um

Ad

sorp

tio

n C

apac

ity

at

50m g

/Nm

3 (µ

g H

g/g

)Upstream ofSO3

Downstream ofSO3

WEPCO Pleasant Prairie

Testing completed fall of 2001

PRB coal

ESP only

Spray cooling

SO3 conditioning system

Activated Carbon Storage and Feed System

ESP Configuration, PPPP

ESP 2-4

Spray Cooling

Carbon Injection

Powdered Activated Carbon Injection System

Baseline Hg Measurements (mg/dscm)

Location Particle Bound

Oxidized, Hg2+

Elemental, Hg0

Total, Hg

Inlet ’99 0.16 2.29 6.21 8.65

Inlet ‘01 1.84 2.34 11.39 15.55

Mercury Trends Week 1

0

5

10

9/24 9/24 9/25 9/25 9/26 9/26 9/27 9/27 9/28 9/28 9/29

Hg

(µ

g/N

m3)

0

10

20

30

40

lb/M

macf

So

rbe

nt

Inlet Outlet Feeder

200

300

400

500

600

700

9/24 9/24 9/25 9/25 9/26 9/26 9/27 9/27 9/28 9/28 9/29

Lo

ad

(M

W)

250260

270280

290300

310320

9/24 9/25 9/26 9/27 9/28 9/29

Tem

pera

ture

(F

) Inlet Hg Probe Temp Outlet Hg Probe Temp

Response Time for PAC Injection on an ESP

0

4

8

12

16

6:00 10:00 14:00 18:00 22:00

Tota

l H

g (

µg

/dN

m3 )

0

10

20

30

40

lbs/

Mm

acf

So

rben

t

ESP Inlet

ESP Outlet

Carbon Injection Performance on a PRB Coal with an ESP

0

10

20

30

40

50

60

70

80

0 10 20 30 40 50

Injection Concentration (lb/MMacf)

% H

g R

em

ov

al

FGD

FGD humid

FGD(g)

FGL

FGD, no SO3

FGD(g), no SO3

Insul

Long Term Tests

Ontario Hydro

Long Term Trend Data

0

5

10

15

20

10/29 11/5 11/12

Hg

(u

g/d

Nm3

)

0

5

10

15

20

Inj.

Co

nc

. (l

b/M

Ma

cf)ESP Outlet Total ESP Inlet Total Inj. Conc.

020406080

100

10/29/01 11/5/01 11/12/01

Hg

Re

m.

(%)

300

500

700

10/29/01 11/5/01 11/12/01

Lo

ad

(M

W)

Speciated Mercury Measured by Ontario Hydro Method (10 lbs/MMacf)

PARTICULATE ELEMENTAL OXIDIZED TOTAL

Baseline ESP Inlet 1.97 12.22 2.51 16.71 ESP Outlet 0.01 9.80 6.01 15.82

Removal Efficiency 99.5% 19.8% -139.3 5.3%

(microgram/dncm)

PAC Injection ESP Inlet 0.98 14.73 1.73 17.44 ESP Outlet 0.00 4.27 0.44 4.71

Removal Efficiency 100.0% 71.0% 74.5% 73.0%

Alabama Power E.C. Gaston

Alabama Power Company E.C. Gaston Electric Generating Plant Unit 3, Wilsonville, AL

270 MW Firing a Variety of Low-Sulfur, Washed Eastern Bituminous Coals

Particulate Collection System– Hot-side ESP, SCA = 274 ft2/1000

acfm– COHPAC baghouse supplied by

Hamon Research-Cottrell

Wet Ash Disposal to Pond

Site Test Configuration with EPRI TOXECON at Alabama Power Plant Gaston

Sorbent Injection

COHPAC

Fly Ash (2%) + PAC

Coal

Fly Ash (98%)

ElectrostaticPrecipitator

S-CEM Duct Traverse

0

2

4

6

8

10

12

14

16

4/28/01 12:00 4/29/01 0:00 4/29/01 12:00 4/30/01 0:00

Hg

(u

g/N

m3)

Inlet Total

Inlet Elemental

Change Point

Example of S-CEM Data

0

5

10

15

20

3/12/2001 3/13/2001 3/14/2001 3/15/2001

Hg

(g

/Nm

3 )

0

1

2

3

4

5

Inj.

Co

nc

. (lb

/MM

ac

f)

Total Inlet

Total Outlet

Injection Conc.

Response Time of PAC Injection with a Fabric Filter

0

10

20

4:00 8:00 12:00 16:00 20:00 0:00

Hg

(u

g/N

m3)

3% O

2

0

1

2

3

Inj.

Co

nc.

(lb

/MM

acf)

Total Outlet Total Inlet Inj. Conc.

Mercury Removal vs. Injection Rate

0

10

20

30

40

50

60

70

80

90

100

0 1 2 3 4 5

Injection Concentration (lb/MMacf)

% H

g R

emo

val

Pressure Drop Increase from PAC Injection

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

0 0.5 1 1.5 2 2.5 3 3.5 4

Injection Concentration (lb/MMacf)

Avg

. pul

ses/

bag/

hr

Mercury Removal vs. Injection Rate

0

10

20

30

40

50

60

70

80

90

100

0 1 2 3 4 5

Injection Concentration (lb/MMacf)

% H

g R

emo

val

PAC Rate Limit Due to Pressure Drop

5-Day Continuous Injection

0

5

10

15

20

25

4/22 4/23 4/24 4/25 4/26 4/27

Hg

(µ

g/N

m3 )

Total Inlet

Total Outlet

Ontario Hydro

0

50

100

150

200

250

300

4/22 4/23 4/24 4/25 4/26 4/27

Bo

iler

Lo

ad (

MW

)

0

2

4

6

8

10

12

Inj.

Co

nc

. (l

b/M

Mac

f)

Load

Sorbent Injection Concentration

Average Mercury Removal Long-Term Tests Gaston, Ontario Hydro

(microgram/dncm)PARTICULATE OXIDIZED ELEMENTAL TOTAL

Baseline COHPAC Inlet 0.09 9.54 5.97 15.60 COHPAC Outlet 0.01 11.19 3.34 14.54

Removal Efficiency 89.1% -17.3% 44.1% 6.8%

PAC Injection COHPAC Inlet 0.23 6.37 4.59 11.19 COHPAC Outlet 0.12 0.91 0.03 1.05

Removal Efficiency 45.6% 85.7% 99.3% 90.6%

Comparison of Sorbent Costs for a Fabric Filter and an ESP

020406080

100

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6

Sorbent Costs (US$/MWh)

% H

g R

emo

val

ESP

COHPAC FF

Conclusions (PAC General) PAC injection can effectively capture elemental and oxidized

mercury from both bituminous and subbituminous coals

Additional field tests and long-term demonstrations are necessary to continue to mature the technology

Fabric filters provide better contact between the sorbent and mercury than ESPs resulting in higher removal levels at lower sorbent costs

New COHPAC FF’s will have to be designed to handle higher loadings of PAC to insure high (>90%) mercury removal

Conventional FF’s should not require any modifications for PAC

Conclusions (Response to Concentration Variations)

Response times to changes in inlet concentrations:– Feedback data from outlet CEMs—tens of minutes

– Impact of changes in injection rate: tens of minutes to hours

Long averaging times will be required to recover from upsets

Injection at somewhat higher rates will make the technology more capable to handle inlet fluctuations

PAC injection lends itself to the use of feed rate parameters as a definition of Maximum Achievable Control Technology

Future Plans

Short-term testing at additional sites–PG&E Brayton Point (Bituminous coal, large ESP) 6/ 2002

–PG&E Salem Harbor (Bituminous coal, SNCR, large ESP) 9/2002

–* TBD (PRB coal, small ESP) 3/2003

–* Southern Company (Bituminous coal, small ESP) 8/ 2003

Long-term testing–*Alabama Power (Bituminous coal, COHPAC FF) 2002-2003

–*CCPI Program (PRB Coal, COHPAC FF) 2004-2006

–*CCPI Program (Bituminous Coal, COHPAC FF) 2004-2006

* Proposed

For More Information www.adaes.com

www.adaes.com/mercury.htm

– Link to other mercury related web sites

– Publications/reports

www.adaes.com/MercuryPublic.htm

– Public information on DOE/NETL Mercury Control Program

www.netl.doe.gov/products/environment/index.html DOE/NETL Website