Performance and Costs of Performance and Costs of

Mercury Control Technology Mercury Control Technology for Bituminous Coalsfor Bituminous Coals

NC DAQ Mercury and CO2 WorkshopNC DAQ Mercury and CO2 Workshop

April 20, 2004April 20, 2004

Raleigh, NCRaleigh, NC

Michael D. Durham, Ph.D., MBAADA-ES, Inc.

8100 SouthPark Way, Unit BLittleton, CO 80120

(303) 734-1727miked@adaes.com

Outline

• Mercury Emissions from Coal Fired Boilers

• Background on Control Technology

• Sorbent Injection for Controlling Hg Emissions

• Costs for Mercury Control

• Regulatory Parameters from a Control Device Perspective

PRS4003

Hg Removal with Existing Equipment

Controls BituminousPM Only

CS-ESP 46%HS-ESP 12%FF 83%PM Scrubber 14%

Dry FGDSDA + ESPSDA + FF 98%

Wet FGDCS-ESP+Wet FGD 81%HS-ESP+Wet FGD 55%FF+Wet FGD 96%

Subbituminous

16%13%72%0%

38%25%

35%33%

PRS4003

Existing Source MACT Limits

NOTE: Output-based standards are referenced to a baseline efficiency (35% for new units; 32% for existing units).

Subcategory Hg

(lb/TBtu)1

Bituminous-fired 2.0

Subbituminous-fired 5.8

PRS4003

Mercury Emissions with Average Capture for Bituminous (46%) and Subbituminous (16%) Coals

0

20

40

60

80

100

0 5 10 15 20

Mercury in Coal (lb/TBtu)

Cu

mu

lati

ve D

istr

ibu

tio

n (

%)

Other Bituminous

Subbituminous

PRS4003

Options Available for Reducing Mercury Emissions

1. Wet Flue Gas Desulfurization (FGD) Scrubbers.

2. Sorbent Injection.

Novel approaches are not considered viable as time from development to market is too long

PRS4003

Control of Mercury in Wet FGD Scrubbers

• Oxidized Mercury is water soluble and can be captured in wet scrubbers.– Some captured mercury gets re-emitted.

• Elemental mercury cannot be captured by scrubbers.

PRS4003

Mercury Removal in Wet Scrubbers for Bituminous Coals

Low correlation of existing data; difficult to predict the mercury removal that will be achieved in a WFGD

0

20

40

60

80

100

0 500 1000 1500 2000

Coal Chloride (ppm)

Me

rcu

ry R

em

ov

al (

%)

Eastern Bit

Western Bit

Pittsburgh #8

PRS4003

Enhancing Capture of Hg in Wet Scrubbers:Increase Amount of Oxidized Hg

CoalElectrostaticPrecipitator

Wet Scrubber

Oxidizing Chemicals

SCR for NOx

Oxidizing Catalysts

PRS4003

Status of Technologies for Oxidizing Mercury

• SCRs: – Documenting performance on full-scale installations.

– Better performance on bituminous than subbituminous coals.

– Possibility of aging effects.

– Possibility of interferences from other chemicals.

– Catalysts are being designed to reduce oxidation of SO3;

this may impact oxidation of Hg.

• Oxidizing Catalysts: – Pilot-scale testing under way.

• Oxidizing Chemicals:– Some very short-term full-scale tests.– Concerns with corrosion.

PRS4003

Sorbent InjectionSorbent InjectionMercury ControlMercury Control

TechnologyTechnology

PRS4003

Coal-Fired Boiler with Sorbent Injection

Sorbent Injection

Ash and Sorbent

ESP or FF

Hg CEM

PRS4003

PRS4003

Activated CarbonStorage and Feed System

PRS4003

Powdered Activated Carbon Injection System

PRS4003

ADA-ES Hg Control Program: Phase I

• Full-scale field testing of sorbent-based mercury control on coal-fired boilers.

• Primary funding from DOE National Energy Technology Laboratory (NETL).

• Cofunding provided by:– Southern Company;– We Energies;– PG&E NEG;– EPRI;– Ontario Power Generation;– TVA;– FirstEnergy;– Kennecott Energy; and– Arch Coal.

PRS4003

Removal of Mercury Species with PAC on Bituminous Coal

Bituminous with FFPARTICULATE OXIDIZED ELEMENTAL TOTAL

PAC Injection μg/m3 μg/m3 μg/m3 μg/m3

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%

PRS4003

Cost and Performance of Sorbent-Based Mercury Control

0

10

20

30

40

50

60

70

80

90

100

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0

Sorbent Costs (mills/kWh)

Mer

cury

Rem

ova

l (%

)

FF Bituminous

FF PRB (EPRI Pilot)

ESP Bituminous

ESP PRB

PRS4003

Effect of Flue Gas Characteristics

• The capacity of sorbents to capture mercury decreases at higher temperatures.

• Chlorine and other trace acid gases play a significant role in the performance of PAC.

PRS4003

Adsorption Capacity vs. Temperature

Equilibrium Adsorption Capacity - Darco FGD

0

500

1000

1500

2000

2500

3000

200 250 300 350 400 450

Temperature (F)

g H

gC

l2 /g

AC

)(

PRS4003

Hg Capture vs. Temperature (w/ACI)

0

10

20

30

40

50

60

70

80

90

100

0 5 10 15 20 25 30

Sorbent Injection Rate (lb/Macf)

Mer

cury

Rem

ova

l (%

)

ESP Bitum 300F

ESP - Bitum 350F

PRS4003

Equilibrium Adsorption Capacities at 250°F Upstream and Downstream of SO3 Injection

0

1000

2000

3000

4000

5000

6000

7000

8000

9000

10000

FGD Carbon Norit Insul Carbon P4 Ash

Eq

uil

ibri

um

Ad

so

rpti

on

Ca

pa

cit

y (

µg

Hg

/g A

C)

Upstream ofSO3

Downstream ofSO3

PRS4003

PAC Performance with ESPs: Effect of Trace Acid Gases

0

10

20

30

40

50

60

70

80

90

100

0 5 10 15 20 25 30

Sorbent Injection Rate (lb/MMacf)

Merc

ury

Rem

oval (%

)

ESP Low S Bit

ESP PRB

ESP Hi S Bit

PRS4003

Sorbent Injection Upstream of a Wet Scrubber

• Injection of AC and capture in ESP will provide an additional mechanism to reduce mercury emissions.

• Oxidation of mercury produced by carbon could enhance capture in FGD.

• Decreased mercury levels in scrubber could reduce potential for reemission of elemental mercury from scrubber.

• Two DOE/Industry full-scale field tests are scheduled:

– Georgia Power Yates; currently on-going, medium-sulfur bituminous coal; and

– AEP Conesville; Spring ’05, high-sulfur bituminous.

PRS4003

Ash Issues

• The mercury captured by PAC, LOI, and ash appears to be very stable and unlikely to reenter the environment.

• The presence of PAC will most likely prevent the sale of ash for use in concrete.

• Several developing technologies to address the problem:– Separation– Combustion– Chemical treatment– Non-carbon sorbents– Configuration solutions such as EPRI TOXECON™

PRS4003

TOXECON™ Configuration

TOXECON™ N

CoalElectrostaticPrecipitator

Sorbent Injection

PJFF

Fly Ash (99%) Fly Ash (1%) + PAC

PRS4003

Alabama Power E. C. Gaston Unit 3• 270 MW firing a variety of low-sulfur, washed eastern

bituminous coals.

• Particulate Collection:

– Hot-side ESP;SCA = 274 ft2/kacfm

– COHPAC™ baghouse

• Wet ash disposal to pond.

• Primary funding from DOE/NETL with cofunding provided by:

– Southern Company– Duke Energy– Ontario Power Generation– TVA– Kennecott Energy– We Energies

– EPRI – First Energy – Hamon Research-Cottrell – Arch Coal

PRS4003

Phase I Test Results

0102030405060708090

100

0 1 2 3 4 5Injection Concentration (lbs/MMacf)

% H

g R

emo

val

PRS4003

Year-Long TOXECONTM Test

• Conduct ~ 1 year demonstration of TOXECONTM (sorbent injection into COHPAC) for power plant mercury control.

• Determine design criteria and costs for new TOXECONTM systems.

• Determine balance-of-plant impacts.

PRS4003

Daily and Weekly Average Mercury

7/21/03 1:217/21/03 1:227/21/03 1:237/21/03 1:247/21/03 1:257/21/03 1:267/21/03 1:277/21/03 1:287/21/03 1:297/21/03 1:307/21/03 1:317/21/03 1:327/21/03 1:337/21/03 1:347/21/03 1:357/21/03 1:367/21/03 1:377/21/03 1:387/21/03 1:397/21/03 1:407/21/03 1:417/21/03 1:427/21/03 1:437/21/03 1:447/21/03 1:457/21/03 1:467/21/03 1:477/21/03 1:487/21/03 1:497/21/03 1:50

0

5

10

15

20

25

30

7/18/03 8/7/03 8/27/03 9/16/03 10/6/03 10/26/03 11/15/03

Hg

(u

g/c

m)

InletOutletInlet WeeklyOutlet Weekly

0

20

40

60

80

100

7/18/03 8/7/03 8/27/03 9/16/03 10/6/03 10/26/03 11/15/03

Rem

ov

al E

ffic

ien

cy (

%)

Daily

Weekly

PRS4003

Impact of Injection on Performance

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 (lbs/MMacf)

Ave

rag

e p

/b/h

PRS4003

Phase I Test Results

0102030405060708090

100

0 1 2 3 4 5Injection Concentration (lbs/MMacf)

% H

g R

emo

val

PRS4003

Low Load/Low Flow Test

• Baseline conditions limit injection concentration.

• Test plan changed to accommodate real-life conditions.

• Current air-to-cloth ratio of 8.0 ft/min is too high for TOXECONTM.

• Low load test conducted to simulate operation at air-to-cloth ratio of 6.0 ft/min– APC arranged for 72 hours of operation at low,

steady load.

PRS4003

Low Load Test: A/C = 6.0 ft/min7/21/03 1:217/21/03 1:227/21/03 1:237/21/03 1:247/21/03 1:257/21/03 1:267/21/03 1:277/21/03 1:287/21/03 1:297/21/03 1:307/21/03 1:317/21/03 1:327/21/03 1:337/21/03 1:347/21/03 1:357/21/03 1:367/21/03 1:377/21/03 1:387/21/03 1:397/21/03 1:407/21/03 1:417/21/03 1:427/21/03 1:437/21/03 1:447/21/03 1:457/21/03 1:467/21/03 1:477/21/03 1:487/21/03 1:497/21/03 1:507/21/03 1:517/21/03 1:527/21/03 1:537/21/03 1:547/21/03 1:557/21/03 1:567/21/03 1:577/21/03 1:587/21/03 1:597/21/03 2:007/21/03 2:017/21/03 2:027/21/03 2:037/21/03 2:04

0

5

10

15

20

25

30

11/5/03 11/6/03 11/7/03 11/8/03 11/9/03

Co

nc

en

tra

tio

n (

ug

/cm

)

Outlet 3%

Inlet 3%

0

20

40

60

80

100

11/5/03 11/6/03 11/7/03 11/8/03 11/9/03

Re

mo

va

l E

ffic

ien

cy

(%

)

0.0

1.0

2.0

3.0

4.0

5.0

Inj

Co

nc

(lb

s/M

Ma

cf)

HgRemovalCarbon

Switched Feeders

0.00

0.05

0.10

0.15

0.20

0.25

0.30

11/5/03 11/6/03 11/7/03 11/8/03 11/9/03

Lo

adin

g (

gr/

acf)

0

1

2

3

4

5

6

Pu

lses

/bag

/h

B Mass Loading

Boiler Load/1000

B Pulse Freq.

Frequency

Mass Loading

Boiler Load/1000

PRS4003

ACI Cost Estimates for Bituminous Coals

• Assumptions– 250 MW Plant; 80% Capacity Factor

• Capital and Operating Costs for ESP– 50-70% Hg Removal: PAC Injection @ 10 lb/Macf

– PAC Injection Equipment: $790,000

– Carbon costs: $2,562,000/yr

• Capital and Operating Costs for FF– Add COHPAC Fabric Filter at $50/kW: $12,500,000

– 80-90% Hg Removal: PAC Injection @ 3 lb/Macf

– PAC Injection Equipment: $790,000

– Carbon costs: $769,000/yr

PRS4003

Costs of Mercury Control Depend on Plant Size Not on Amount Removed

• Costs of mercury control are unrelated

to the amount of mercury captured

– Sorbent Injection Technology

– SCR/FGD

– Catalytic Oxidation

– Other Developing Technologies

0

20

40

60

80

100

0 5 10 15 20

Mercury in Coal (Hg/TBTU)

Hg

Co

ntr

ol

Co

sts

($/w

t)

PRS4003

Conclusions on ACI Performance

• AC injection can effectively capture elemental and oxidized mercury from bituminous coals.

• There will be difference in site to site performance of ACI due to differences in coal, equipment, and flue gas characteristics.

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

• Long-term results are promising showing consistent Hg removal greater than 85%.

• New COHPAC™ fabric filters will have to be designed to handle higher loadings of PAC to insure high (>90%) mercury removal.

PRS4003

Commercial Status of Technology1. Equipment

• Similar equipment has been used successfully in the waste industry to inject AC into flue gas.• It has successfully been scaled up for full-scale utility applications.• Operating continuously for nearly a year at Gaston.• Three AC injections systems currently operating.

2. Supply of Activated Carbon and Other Sorbents• Sufficient supply available to meet several State regulations.• Additional production needed to meet Federal regulations.• Tremendous progress being made with improved sorbents.

3. Performance• Will vary with type of equipment (FF vs. ESP).• Will vary from site to site due to flue gas characteristics (temperature, acid gases).

PRS4003

Availability of Activated CarbonsCurrent excess capacity

of AC production in Tons/year

NORIT Americas: 22,500Other US Suppliers: 40,000Total US Excess Capacity 62,500

Donau (Germany) 130,000CarboChem (China) 60,000Total Import Excess Capacity 190,000

Total US and Import Excess Capacity 252,500

PRS4003

Number of 250 MW Plants that Can Be Treated by Currently Available AC (out of 1100 in US)

Excess Capacity ESPs FFTons/yr (50-70%) (70-90%)

US AC 62,000 30 99

Total US 252,000 120 400

Plus Imports• Manufacturers plan to increase production to meet market

demand, but only upon regulatory certainty.

PRS4003

Regulatory Parameters from a Control Device Perspective

1. Long term averaging2. Dual Limit

• Removal Efficiency• Emission Limit

3. Flexibility in Achieving Mercury Removal• Accounts for site by site variation in performance• Enhances cost effectiveness

4. Mechanism to Encourage Adoption

PRS4003

Long-Term Averaging Time will Allow Control Devices to Adapt to Variations in Coal and Operating Conditions

0

5

10

15

20

25

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

Hg

(µg/

Nm

3 )

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

Boi

ler

Load

(M

W)

0

2

4

6

8

10

12

Inj.

Con

c. (l

b/M

Mac

f)

Load

Sorbent Injection Concentration

PRS4003

Decisions on Mercury Control with Flexibility in Achieving ReductionsUtilities would have a significant economic incentive to put mercury control on units that are:

• Higher emitters

• Larger plants

Therefore, a flexible approach would result in the greatest reduction in total mercury emissions while minimizing costs

PRS4003

Flexibility Would Provide a Framework for Fleet-wide Decisions on Mercury Control

0

10

20

30

40

50

60

70

80

90

100

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0

Sorbent Costs (mills/kWh)

Mer

cury

Rem

ova

l (%

)

FF Bitum

FF PRB

ESP Bituminous

ESP PRB

PRS4003

Flexibility Would Help Address Plant by Plant Variations in Performance Guarantees

0

10

20

30

40

50

60

70

80

90

100

0 5 10 15 20 25 30

Sorbent Injection Rate (lb/Macf)

Mer

cury

Rem

ova

l (%

)

ESP Bitum

ESP PRB

PRS4003

Early Adoption of Technology Provides Increased Experience Base• To date, 8 full-scale field tests have been completed through

funding from DOE-NETL, EPRI, Utilities, APC Vendors, and Coal Companies.

• An additional 12 field tests are planned for the next 2-3 years.

• Economic incentives for early compliance are needed to offset risks with new technology.

• This will increase the operational data base (different fuels and equipment), decrease uncertainty, solidify guarantees.