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) [email protected]
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.