United States EPA-600/R-03-11 0
Environmental Protection October 2003
Research and
Development
Performance and Cost of Mercury and
Multipollutant Emission Control
Technology Applications on Electric
Utility Boilers
Prepared for
Office of Research and Development
Prepared by
National Risk Management
Research Laboratory
Research Triangle Park, NC 27711
-------
EPA-600/R-03/110
October 2003
Performance and Cost of Mercury and
Multipollutant Emission Control
Technology Applications on
Electric Utility Boilers
Prepared by:
James E. Staudt
Andover Technology Partners
112 Tucker Farm Road
North Andover, MA 01845
Wojciech Jozewicz
ARCADIS Geraghty & Miller
4915 Prospectus Drive, Suite F
Durham, NC 27713
EPA Contract No. 68-C-99-201, Work Assignment 4-028
EPA Project Officer: Ravi K. Srivastava
National Risk Management Research Laboratory
Research Triangle Park, NC 27711
U.S. Environmental Protection Agency
Office of Research and Development
Washington, DC 20460
-------
ABSTRACT
Under the Clean Air Act, as amended, the Environmental Protection Agency (EPA) has
determined that mercury emissions from coal-fired power plants should be regulated. Based on
this determination, EPA is to propose Maximum Achievable Control Technology Standards for
these emissions by December 2003. To aid in this regulatory effort, estimates of the performance
and cost of powdered activated carbon (PAC) injection-based mercury control technologies and
multipollutant control technologies that may be useful in controlling mercury emissions have
been developed. This report presents these estimates.
Estimates cost range for PAC injection, based on currently available data, is 0.03-3.096
mills/kWh. However, the higher costs are usually associated with the minority of plants using
Spray Dryer Absorbers and Electrostatic Precipitators (SDAs plus ESPs) or the small number
of plants using hot ESPs (ESPhs). Excluding the minority of plants using SDAs plus ESPs or
ESPhs, current cost estimates are from 0.03 to 1.903 mills/kWh. At the low end of these cost
ranges, 0.03 mills/kWh, it is assumed that no additional control technologies are needed, but
mercury monitoring will be necessary. In these cases, high mercury removal may be the result
of the type of particulate matter, nitrogen oxide, and sulfur dioxide control measures currently
employed, such as combinations of ESP, selective catalytic reduction (SCR), and wet flue gas
desulfurization (FGD) on bituminous coal-fired boilers.
Multipollutant control methods evaluated in this program that may provide cost effective
mercury control and control of other pollutants include Electro Catalytic Oxidation (ECO),
Advanced Dry FGD, and a coal beneficiation method. ECO and Advanced Dry FGD are flue gas
treatment methods and are estimated to have costs ranging from 3.28 to 12.33 mills/kWh over
a range of fuel types and conditions. A coal beneficiation method called K-Fuel was shown to
provide about 60% or greater reduction in mercury from Powder River Basin coal on a heating
value basis.
Based on this work, it is expected that future efforts in R&D are likely to focus on improved
understanding of mercury speciation across SCRs leading to beneficial effects of combinations
of SCR with wet FGD and developing sorbents that can improve performance and cost of
sorbent-based mercury control technologies. Multipollutant control technologies, which are more
costly than single-pollutant mercury control technologies but offer other environmental benefits,
will be another area for further development that could improve the cost of reducing emissions
from coal-fired power plants. Finally, removing mercury from the coal, along with other fuel
quality improvements, may prove to be a very cost effective approach for reducing emissions.
-------
Foreword
The U.S. Environmental Protection Agency (EPA) is charged by Congress with
protecting the Nation's land, air, and water resources. Under a mandate of national
environmental laws, the Agency strives to formulate and implement actions leading to a
compatible balance between human activities and the ability of natural systems to support
and nurture life. To meet this mandate, EPA's research program is providing data and
technical support for solving environmental problems today and building a science
knowledge base necessary to manage our ecological resources wisely, understand how
pollutants affect our health, and prevent or reduce environmental risks in the future.
The National Risk Management Research Laboratory (NRMRL) is the Agency's center
for investigation of technological and management approaches for preventing and
reducing risks from pollution that threaten human health and the environment. The focus
of the Laboratory's research program is on methods and their cost-effectiveness for
prevention and control of pollution to air, land, water, and subsurface resources; protection
of water quality in public water systems; remediation of contaminated sites, sediments,
and ground water; prevention and control of indoor air pollution; and restoration of
ecosystems. NRMRL collaborates with both public and private sector partners to foster
technologies that reduce the cost of compliance and to anticipate emerging problems.
NRMRL's research provides solutions to environmental problems by: developing and
promoting technologies that protect and improve the environment; advancing scientific and
engineering information to support regulatory and policy decisions; and providing the
technical support and information transfer to ensure implementation of environmental
regulations and strategies at the national, state, and community levels.
This publication has been produced as part of the Laboratory's strategic long-term
research plan. It is published and made available by EPA's Office of Research and
Development to assist the user community and to link researchers with their clients.
Hugh W. McKinnon, Director
National Risk Management Research Laboratory
EPA REVIEW NOTICE
This report has been peer and administratively reviewed by the U.S. Environmental
Protection Agency and approved for publication. Mention of trade names orcommercial
products does not constitute endorsement or recommendation for use.
This document is available to the public through the National Technical Information
Service, Springfield, Virginia 22161.
in
-------
ACKNOWLEDGMENTS
The authors would like to acknowledge the many contributors to this document, without whose
efforts this report would not be complete. In particular, we wish to acknowledge the technical
guidance and insights provided by Dr. Ravi Srivastava of EPA's National Risk Management
Research Laboratory, Office of Research and Development. The authors also appreciate many
helpful discussions with vendors of control technologies discussed in this document.
IV
-------
TABLE OF CONTENTS
Section Page
Abstract ii
Foreword iii
Acknowledgments iv
List of Figures vii
List of Tables viii
List of Acronyms and Abbreviations x
1.0 Introduction 1
2.0 Mercury Speciation and Capture 5
3.0 Mercury Control with Existing Technologies 7
3.1 Mercury Removal in PM Equipment 8
3.2 Impacts of NOX Controls on Mercury Speciation and Capture 9
3.2.1 SCR Impact on Mercury Speciation 9
3.2.2 Mercury Removal Though Combustion NOX Controls 15
3.3 Mercury Removal in SO2 Control Equipment 16
3.3.1 Mercury Removal in Wet FGD 16
3.3.2 Mercury Removal in SDA 17
3.4 Mercury Removal in Other Control Devices 18
3.5 Models of Mercury Removal by Existing Equipment 18
4.0 Emerging Control Technologies 21
4.1 PAC Injection-Based Technologies 21
4.1.1 Mercury Removal Models 22
4.1.2 Mercury Reduction by PAC Injection 23
4.2 Emerging Control Technologies 25
4.2.1 Electro Catalytic Oxidation (ECO 26
4.2.2 Advanced Dry FGD 32
4.2.3 K-Fuel 38
5.0 Technologies Currently under Development 43
5.1 Oxidation Technologies 43
5.2 Sorbent Technologies 44
v
-------
5.3 Other Technologies 46
6.0 Costs of Reducing Mercury Emissions 49
6.1 Mercury and Multipollutant Control Cost Models 49
6.2 Fuel Types, Plant Characteristics, and Model Plant Cases 53
6.3 Cost Model Results 57
6.3.1 High Sulfur Bituminous Coals (Model Plants 1-10, 26-28) 57
6.3.2 Low Sulfur Bituminous Coals 66
6.3.3 Low Sulfur Subbituminous Coals including Powder River
Basin Coals 71
6.4 Cost Impacts of Selected Variables 80
6.5 Summary of Mercury and Multipollutant Control Costs 86
7.0 Summary 89
8.0 References 91
Appendices
A Appendix A Description of Mercury and Multipollutant Control Performance
and Cost Model A-l
B Description of PAC Injection Algorithms Used in the Mercury and
Multipollutant Control Performance and Cost Model B-l
C Summary of Mercury Control Cases Analyzed C-l
D Results of Model Runs D-l
VI
-------
LIST OF FIGURES
Figure Paq<
1. Mercury oxidation without a catalyst as a function of residence time, gas
temperature, and HC1 content 10
2. Mercury oxidation across SCR catalysts and without SCR catalyst 11
3. Oxidation of mercury across C-l SCR catalyst in PRB-derived flue gas 12
4. Effect of flue gas exposure time on C-l SCR catalyst oxidation of elemental
mercury at 700 °F and space velocity of 1,450 hr"1 12
5. Location of ECO installation in a power plant 26
6. Detailed process flow diagram of ECO 27
7. ECO power consumption versus NOX 31
8. F.L. Smidth Airtech AirTech Gas-Solids Absorber (GSA) 33
9. Installation of an advanced dry FGD upstream of an existing ESP 33
10. Installation of an advanced dry FGD downstream of an existing ESP 34
11. Advanced dry scrubber SO2 removal performance 35
12 Advanced dry scrubber performance 35
13. Overall schematic of K-Fuel processing plant 39
14. K-Fuel thermal processing plant 39
15. SO2 and NOX emissions from test burns of K-Fuel and untreated fuels 40
16. Estimated Effect of K-Fuel Cost on Generation Cost 77
17. Cost of PAC Injection for 500 MW Coal Fired Boilers with existing
ESPc orFF 81
18. Cost of PAC Injection for 500 MW Coal Fired Boilers with Existing
ESPc—Effect of Medium Versus High Performance PAC 82
19. Effect of Capital Cost on 90 Percent Mercury Control with PAC on Boiler
with Existing ESPc and Retrofit of Downstream PJFF 83
20. Effect of Fertilizer Value on Cost of Emissions Control with ECO on a
500 MW Bituminous Coal Boiler 84
21. Effect of Power Value on Cost of Emissions Control with ECO on a
500 MW Bituminous Coal Boiler 85
22. Effect of Reagent Cost on Cost of Emissions Control with Advanced Dry
FGD on a 500 MW Boiler Firing Low Sulfur Bituminous Coal 85
vn
-------
LIST OF TABLES
Table Page
1. Average Mercury Capture by Existing Post-Combustion Control
Configurations Used for PC-Fired Boilers 7
2. Summary of Results from Full-Scale SCR Mercury Oxidation Tests 14
3a. Parameters Used for Equations 1 and 2 Which Estimate Mercury Removal
by Existing Equipment 19
3b. Parameters Used for Equations 1 and 2 Which Estimate Percent of
Remaining Mercury in Gas that Is Elemental Mercury 20
4. ECO Pollutant Removal Efficiencies 30
5. Estimated Cost of CFB-FGD System for a 500 MW Plant Burning PRB Coal . . 38
6. Comparison of Typical PRB Coal with K-Fuel 41
7. Fuels Used In Model Plant Analysis 54
8. Power Plant Characteristics 55
9. Mercury Control Technology Applications and Co-benefits 56
lOa. High Sulfur Coal, ESP plus FGD Without SCR Co-benefit (Model Plants 1,6) . 59
lOb. High Sulfur Coal, ESP plus FGD With SCR Co-benefit (Model Plants 1, 6) .... 60
1 la. High Sulfur Coal, ESPh plus FGD Without SCR Co-benefit (Model Plants 3,8) 61
1 Ib. High Sulfur Coal, ESPh plus FGD with SCR Co-benefit (Model Plants 3,8) ... 61
12. Advanced Dry FGD on High Sulfur Coal (Model Plants 4, 9), Sensitivity
to Capital Cost 62
13. ESP and ECO on High Sulfur Coal (Model Plants 5 and 10), Sensitivity
to Capital Cost 63
14. High Sulfur Coal, 100 MW SDA, and ESPc (Model Plant 26) 64
15. High Sulfur Coal, 100 MW SDA, and FF (Model Plant 27) 65
16. High Sulfur Coal, 100 MW ESPh, and FGD (Model Plant 28) 66
17. Low Sulfur Coal, ESPc, and No SO2 Controls (Model Plants 11 and 29) 67
18. Low Sulfur Coal, FF, and No SO2 Controls (Model Plants 12 and 30) 68
19. Low Sulfur Coal, ESPh, and No SO2 Controls (Model Plants 13 and 31) 69
20. ECO Installed After Paniculate Removal (Model Plants 14-16, 32-34),
Sensitivity to Capital Cost 70
21. Advanced Dry FGD (Model Plants 17-19, 35-37), Sensitivity to Capital Cost. . 71
Vlll
-------
22. Low Sulfur Subbituminous Coals, ESPc, and No SO2 Control (Model
Plants 20, 38) 73
23. Low Sulfur Subbituminous Coals, FF, and No SO2 Control (Model
Plants 21, 39) 74
24. Low Sulfur Subbituminous Coals, ESPh, and No SO2 Controls (Model
Plants 22, 40) 75
25. Low Sulfur Subbituminous Coals with ECO (Model Plants 23-25, 41-43)
Sensitivity to Capital Cost 76
26. Comparison of Estimated Mercury Emissions from PRB and K-Fuel Boilers
Equipped with Particulate Control and No Additional Mercury or SO2 Controls . 77
27. K-Fuel, ESPc and No SO2 Control (Model Plants 44, 47) 78
28. K-Fuel, FF, and No SO2 Control (Model Plants 45, 48) 79
29. K-Fuel, ESPh, and No SO2 Controls (Model Plants 46, 49) 80
30. Estimated Cost of Mercury Control—Current and Potential Cost Estimates .... 87
31. Estimated Costs of Multipollutant Controls 88
IX
-------
LIST OF ACRONYMS AND ABBREVIATIONS
ACRONYM DEFINITION
ADP acid dew point
ATS approach to saturation
CAAA Clean Air Act Amendments of 1990
CEMS continuous emission monitoring system
COHPAC compact hybrid particulate collector
CRF capital recovery factor
ECO electro catalytic oxidation
EPA U.S. Environmental Protection Agency
ESP electrostatic precipitator
ESPc cold-side electrostatic precipitator
ESPh hot-side electrostatic precipitator
FF fabric filter
FGD flue gas desulfurization
GSA gas suspension absorber
HC1 hydrogen chloride
Hg mercury
HgCl2 mercuric chloride
Hg° elemental mercury
Hg++ oxidized mercury
Hgp particle-bound mercury
HgT total mercury
HgO mercury oxide
ICR information collection request
IPM integrated planning model
kWh kilowatt hour
LOT loss of ignition
LNB low NOX burner
LSFO limestone forced oxidation
MACT maximum achievable control technology
MEL magnesium enhanced lime
MW megawatt
MWCs municipal waste combustors
NETL National Energy Technology Laboratory
-------
LIST OF ACRONYMS AND ABBREVIATIONS
(continued)
ACRONYM DEFINITION
NOX
OAR
OFA
OH
O&M
PAC
PC
PFF
PJFF
PM
PPPP
PRB
PS
RGFF
RAP
R&D
SC
SCR
SDA
SNCR
SO2
TCLP
UBC
WESP
oxides of nitrogen
EPA's Office of Air and Radiation
overfire air
Ontario Hydro
operation and maintenance
powdered activated carbon
pulverized coal
polishing fabric filter
pulse jet fabric filter
particulate matter
Pleasant Prairie Power Plant
Powder River basin
particle scrubber
reverse-gas fabric filter
rapid absorption process
research and development
spray cooling
selective catalytic reduction
spray dryer absorber
selective noncatalytic reduction
sulfur dioxide
toxicity characteristic leaching procedure
unburned carbon
wet electrostatic precipitator
XI
-------
1.0 INTRODUCTION
In the atmosphere, mercury exists in two forms: elemental mercury vapor (Hg°) and ionic mercury
(Hg++). Hg° can circulate in the atmosphere for up to one year and, consequently, can undergo
dispersion over regional and global scales. Hg++ in the atmosphere either is bound to airborne
particles or exists in gaseous form. This form of mercury is readily removed from the atmosphere
by wet and dry deposition. After deposition, mercury is commonly re-emitted back to the
atmosphere as either a gas or a constituent of particles and redeposited elsewhere. In this fashion,
mercury cycles in the environment.1
A number of human health and environmental impacts are associated with exposure to mercury.
Mercury is known to bio-accumulate in fish and animal tissue in its most toxic form,
methylmercury. Human exposure to methylmercury has been associated with serious neurological
and developmental effects. Adults exposed to methylmercury show symptoms of tremors, loss
of coordination, and memory and sensory difficulties. Offspring exposed during pregnancy show
atrophy of the brain with delayed mental development. The incidence and extent of such effects
depend on the level of exposure to methylmercury. Hg° is readily absorbed through lungs and,
being fat-soluble, is rapidly distributed throughout the body. Subsequently, it slowly oxidizes to
Hg++, which accumulates in the brain and can lead to tremors, memory disturbances, sensory loss,
and personality changes. Hg++ is absorbed through the digestive tract, accumulates in the kidneys,
and can lead to immune-mediated kidney toxicity. Adverse effects of mercury on fish, birds, and
mammals include reduced reproductive success, impaired growth, behavioral abnormalities, and
even death. Details of the risks associated with exposure to mercury are discussed in the
literature.1 A severe case of human exposure occurred in Minamata, Japan in the 1950s.2
Under the Clean Air Act as amended the Environmental Protection Agency (EPA) has determined
that mercury emissions from coal-fired power plants should be regulated.3 Based on this determi-
nation, EPA is to propose Maximum Achievable Control Technology (MACT) Standards for
these emissions by December 2003. To aid in this regulatory effort, this report has been prepared
as an update of a previous report (EPA 600/R-00-083)4 that presented preliminary estimates of
the performance and cost of promising mercury control technologies applicable to coal-fired
electric utility boilers. Although most of these technologies are based on injection of powdered
activated carbon (PAC) into boiler flue gas, additional technologies that offer promise in control
of mercury and other pollutants are also discussed in this report.
-------
The report layout is as follows. First, the general principles of mercury speciation and capture
are discussed. Second, mercury removal by existing equipment on coal-fired boilers is discussed.
Third, promising mercury control technologies for coal-fired electric utility boilers are identified,
and the performance characteristics of these technologies are estimated. These include
characterization of mercury removal performance possible as a function of various parameters.
Fourth, model plants representing the spectrum of retrofit possibilities are identified and a matrix
of cases to be studied is developed. Next, costs of controlling mercury emissions from these model
plants with the technologies of interest are examined. Finally, potential future improvements in
these costs are discussed. During discussion of cost and potential improvements, research and
development (R&D) areas are identified for near-term emphasis.
Two multipollutant air pollution control technologies were evaluated as well as a coal beneficiation
technology that offers pollution control advantages. The two air pollution control technologies
include electro catalytic oxidation (ECO) and advanced dry flue gas desulfurization (FGD). ECO
is aunique technology without any full-scale commercial experience on utility boilers. However,
commercial-scale demonstrations are being built, and this technology has been studied extensively
by the U. S. utility industry and the Department of Energy5. So, results are presented here for ECO
technology; however, they should be considered preliminary. Advanced Dry FGD has extensive
experience in waste incineration applications and some limited commercial experience on coal-fired
boilers. Moreover, advanced dry FGD is similar in many respects to spray dryer absorber (SDA)
technology—a well established sulfur dioxide (SO2) control technology used on utility boilers.
However, data regarding the control of mercury using this technology is limited, so the results
presented here may be preliminary with regard to mercury control but are expected to be more
reliable with regard to control of SO2 and particulate matter (PM) and with respect to cost. Coal
beneficiation through the K-Fuel process has been demonstrated on the pilot scale, and the first
commercial plant for coal beneficiation through the K-Fuel process is planned. In addition to
enhancing coal heating value for low-rank fuels, this process can reduce the content of mercury,
sulfur, and nitrogen in the coal, providing multipollutant benefits. The costs are estimated in terms
of increased fuel cost.
Use of sorbent injection technologies to control mercury emissions from electric power plants
would result in mercury-impregnated sorbent waste, which would need to be disposed of either
by itself or in mixture with flyash. One of the more commonly practiced solid waste disposal options
is landfilling. However, there is limited information available on the stability of mercury in ash
and sorbent residue. Therefore, itisunclear whether any potential exists for the release of mercury
back into the environment from landfilled mercury-impregnated solid waste. Further research
-------
is needed on ash and sorbent residue to evaluate mercury retention and the potential for release
back into the environment. Due to lack of information, this report does not address any potential
costs that may result if mercury has to be stabilized in sorbent waste.
-------
-------
2.0 MERCURY SPECIATION AND CAPTURE
Mercury is volatilized and converted to Hg° in the high temperature regions of combustion devices.
As the flue gas cools, Hg° is oxidized to Hg++. The rate of oxidization is dependent on the
temperature, flue gas composition and properties, and amount of flyash and any entrained sorbents.
In coal-fired combustors, where the concentrations of hydrogen chloride (HC1) are low, and where
equilibrium conditions are not achieved, Hg° may be oxidized to mercuric oxide (HgO), mercuric
sulfate (HgSO4), mercuric chloride (HgCl2), or some other mercury compound.6 The oxidization
of Hg° to HgCl2 and to other ionic forms of mercury is abetted by catalytic reactions on the surface
of flyash or sorbents and by other compounds that may be present in the flue gas. Applications
of nitrogen oxides (NOX) control technologies such as selective catalytic reduction (SCR) can
assist in oxidation of Hg°.
Hg°, HgCl2, and HgO are primarily in the vapor phase at flue gas cleaning temperatures. Therefore,
each of these forms of mercury can potentially be adsorbed onto porous solids such as flyash, PAC,
and other sorbents for subsequent collection in a PM control device. These mercury forms may
also be captured in carbon bed filters or other reactors containing appropriate sorbents.
Mercury removal with wet scrubbers also appears to be possible. HgCl2 is water-soluble and reacts
readily with alkali metal oxides in an acid-base reaction; therefore, conventional acid gas scrubbers
used for SO2 control can also effectively capture HgCl2. However, Hg° is insoluble in water and
must be adsorbed onto a sorbent or converted to a soluble form of mercury that can be collected
by wet scrubbing. HgO has low solubility and probably has to be collected by methods similar
to those used for Hg°. Therefore, the form of mercury that is most easily removed is HgCl2 and
this form of mercury is most readily formed when burning coals that are higher in chlorine content,
such as Eastern bituminous coals. Furthermore, as will be described in Sections 3.1 and 3.2, the
equipment on the boiler also plays an important role in determining mercury speciation. For this
reason coal type, coal chlorine content, and the boiler equipment all play a significant role in
determining the ease with which mercury can be removed from coal combustion flue gas streams.
The following sections will describe mercury removal technologies pertinent to coal-fired boilers.
For many technologies described in the following sections, the coal properties and the existing
equipment on the boiler will have an impact on the total mercury removal when that facility is
retrofitted with mercury removal equipment. This is because mercury speciation is important in
-------
determining the ease or difficulty of removing mercury from the exhaust gas. Additionally, fuel
and the equipment used in the facility play role in determining the mercury speciation.
-------
3.0 MERCURY CONTROL WITH EXISTING TECHNOLOGIES
Data derived from an EPA Information Collection Request (ICR) showed that mercury released
from coal combustion may be partly removed from the exhaust gases by existing equipment without
additional retrofit technology .7 This chapter discusses the mercury control achieved with existing
technologies utilized for control of PM, NOX, and SO2 emissions at electric utility coal-fired boilers.
Table 1 shows the average reduction in total mercury (HgT) emissions from ICR data for coal-boiler-
control classes that burn pulverized coal (PC). Plants that employ only post-combustion PM controls
display class average HgT emission reductions ranging from 1 to 90 percent. Units with fabric
filters (FFs) obtain the highest average levels of control. Decreasing average levels of control are
generally observed for units equipped with a cold-side electrostatic precipitator (ESPc), hot-side
ESP (ESPh), and particle scrubber (PS). For units equipped with dry scrubbers, the class average
HgT emission reductions ranged from 2 to 98 percent. The estimated class average reductions for
wet FGD scrubbers were similar and ranged from 10 to 98 percent.
Table 1. Average Mercury Capture, in Percent, by Existing Post-Combustion Control
Configurations Used for PC-Fired Boilers.8
Post-Combustion
Control Strategy
°M Control Only
°M Control and
SDA
°M Control and
Wet FGD System3
Post-Combustion
Emission Control
Device
Configuration
ESPc
ESPh
FF
PS
SDA+ESP
SDA+FF
SDA+FF+SCR
PS+FGD
ESPc+FGD
ESPh+FGD
FF+FGD
Coal Burned in PC-Fired Boiler Unit
Bituminous
36
14
90
not tested
not tested
98
98
12
81
46
98
Subbituminous
9
7
72
9
43
25
not tested
10
29
20
not tested
Lignite
1
not tested
not tested
not tested
not tested
2
not tested
not tested
48
not tested
not tested
' Estimated capture across both control devices
-------
For PC-fired boilers, the amount of Hg captured by a given control technology is greater for
bituminous coal than for either subbituminous coal or lignite. For example, the average capture
of Hg, based on Ontario Hydro (OH) inlet measurements in PC-fired plants equipped with an ESPc,
is 36 percent for bituminous coal, 9 percent for subbituminous coal, and 1 percent for lignite.
3.1 Mercury Removal in PM Control Equipment
Approximately 77 percent of the coal-fired utility boilers currently operating in the United States
are equipped with only an ESP or an FF. Gaseous mercury (both Hg° and Hg++) can potentially
be adsorbed on fly ash and be collected in a downstream ESP or FF. The modern ESPs and FFs
that are now used on most coal-fired units achieve very high capture efficiencies for total PM.
As a consequence, these PM control devices are also effective in capturing PM-bound mercury
(Hgp) in the boiler flue gases.
The degree to which mercury can be adsorbed onto fly ash for subsequent capture in PM control
is dependent on the speciation of mercury, the flue gas concentration of fly ash, the properties
of fly ash and the temperature of the flue gas in the PM control device. It is currently believed
that mercury is primarily adsorbed onto the unburned carbon in fly ash. Approximately 80 percent
of the coal ash in PC-fired boilers is entrained with the flue gas as fly ash. PC-fired boilers with
low-NOx burners have higher levels of carbon in the fly ash with a correspondingly higher potential
for mercury adsorption. Cyclone and stoker boilers tend to have high levels of carbon in the fly
ash but have lower flue gas concentrations of fly ash than PC-fired boilers. Fly ash concentrations
in fluidized-bed combustors tend to be higher than those in PC-fired boilers. Also, the carbon
content of fluidized-bed combustor fly ash is generally higher than that of PC-boiler fly ash.
Gas-phase mercury in units equipped with an ESP can be adsorbed on the entrained fly ash upstream
of the ESP. The gas-phase mercury in units equipped with a FF can be adsorbed by entrained fly
ash or it can be adsorbed as the flue gas passes through the filter cake on the surface of the FF.
The degree to which gaseous mercury adsorbs on the filter cake typically depends on the speciation
of gaseous mercury in the flue gas; in general, gaseous Hg++ is easier to adsorb than gaseous Hg°.
The very intimate contact between the gas and collected PM (which can act as a sorbent for the
gas-phase mercury) that occurs in a FF significantly enhances the gas-phase mercury collection
efficiency of the FF over what is possible with an ESP. As indicated in Table 1, the ICR data showed
that, for both bituminous and subbituminous coals, mercury collection in boilers equipped only
with FFs is much higher than for boilers equipped only with ESPs. As will be shown later in this
-------
document, this effect also contributes to much more efficient collection of mercury when PAC
is injected for additional mercury control upstream of a FF as opposed to injection upstream of
an ESP. New hybrid ESP-FF technologies, such as the Combined Hybrid Particle Collector
(COHPAC), offer ways to cost-effectively retrofit ESP's with FF and realize this benefit. The
COHPAC approach also offers the benefit enabling segregation of injected PAC from much of
the collected fly ash.
ICR data reflected that plants which employ only post-combustion PM controls display average
Hg emission reductions ranging from 0 percent to 89 percent.8 The highest levels of control were
observed for units with FFs. Decreasing levels of control were shown for units with ESPs,
paniculate scrubbers, and mechanical collectors. The average mercury reduction for two PC-fired
units equipped with a FF baghouse and burning bituminous coal averaged 90 percent while two
similarly equipped units burning subbituminous coals displayed an average mercury reduction
of 72 percent. The average capture of Hg for PC-fired plants equipped with an ESPc was 3 5 percent
for bituminous coal, 3 percent for subbituminous coal, and near zero for lignite.
3.2 Impacts of NOX Controls on Mercury Speciation and Capture
Several NOX control technologies, including low NOX burners (LNBs), overfire air (OFA),
reburning, selective noncatalytic reduction (SNCR), and SCR, are employed at utility coal-fired
boilers to control NOX emissions. Of these control technologies, SCR has an impact on the
speciation of mercury in flue gas and, therefore, subsequent capture in wet FGD systems. Based
on recent data, combustion controls such as LNBs, OF A, and reburning may also have the potential
to increase mercury capture in flyash. The effects of SCR and combustion controls on mercury
capture are described in the following sections.
3.2.1 SCR Impact on Mercury Speciation
The speciation of mercury is known to have a significant impact on the ability of air pollution
control equipment to capture it. In particular, the oxidized form of mercury, mercuric chloride
(HgCl2), is highly water-soluble and is, therefore, easier to capture in wet FGD systems than the
elemental form of mercury which is not water-soluble. SCR catalysts can act to oxidize a significant
portion of the elemental mercury, which makes it easier to remove in downstream wet FGD.
-------
The results of studies have suggested that oxidation of elemental mercury by SCR catalyst may
be affected by9"12
• The space velocity of the catalyst
• The temperature of the reaction
• The concentration of ammonia
• The age of the catalyst
• The concentration of Cl in the gas stream
Tests on a laboratory combustor has shown that mercury oxidation without a catalyst was enhanced
with higher Cl concentration (higher HC1 at inlet) and that oxidation increased with residence
time and at lower temperatures, as shown in Figure 1 9 Reference 9 also describes the results of
laboratory tests of oxidation of mercury across different types of SCR catalysts. The results of
these tests, shown in Figure 2, demonstrated that the catalyst significantly increased the amount
of elemental mercury that oxidized to mercuric chloride.
1200
4§0
3 4
[a]
Figure 1. Mercury oxidation without a catalyst as a function of residence
time, gas temperature, and HCI content.9
10
-------
*»
I
•£ *
100
80
35-ce!Is
plats-type
22-calls
60
'i
X ^
Is "S
-------
5 70
I
C SO
•a
8
S so
5
I,o
1
1 30
™ 20
W
6 •
2000 4000 6000
• * yOO'f; OppfTtNH™
•O- 700 'f; 3CO ppm fdHj
•*—aOO'f. 0 ppm NH.J
800 ;F, 300 |jpm NH^
- TOO'-F. 0 P(>m HHy (ESP lni«)
10006 120CO
14000 16000 iaooo
Figure 3. Oxidation of mercury across C-1 SCR catalyst in PRB-
derived flue gas.10
100
90
BO
70
80
80
40
30
20
10
Space Velocity a 1450 h( *
800 1,000
"0-.,
1.500 3,500 3,000
T!m» (hr*>
- 0 ppm NH j
3,800 4,000
Figure 4. Effect of flue gas exposure time on C-1 SCR catalyst oxidation of
elemental mercury at 700 °F and space velocity of 1,450 hr1.10
12
-------
to boiler gases at a space velocity of 1450 hr l. When exposed to 3 00 ppm of ammonia, fresh catalyst
continued to oxidize 80 to 90 percent of the elemental mercury. However, after 4200 hours of
exposure, no oxidation was measured across the catalyst when ammonia was present, suggesting
that ammonia may play a role in suppressing mercury oxidation.
Oxidation of elemental mercury to mercuric chloride across an SCR catalyst, therefore, may be
a function of space velocity, temperature, ammonia concentration, and catalyst life. Other factors,
such as fly ash characteristics, are also believed to play a role.
Reference 11 describes the results of a program that evaluated mercury oxidation across full-scale
utility boiler SCR systems. A summary of the results of the tests conducted in 2001 is shown in
the first four entries in Table 2. Testing was performed at four coal-fired electric utility plants
having catalyst age ranging from around 2500 hours to about 8000 hours. One plant fired
subbituminous coal, and three other plants fired Eastern bituminous coal. The test results showed
high levels of mercury oxidation in two of the three plants firing eastern bituminous coal and
insignificant oxidation at the other two plants (one firing bituminous coal and the other
subbituminous). However, it should be noted that for both of the plants where little or no mercury
oxidation was measured (S1 and S3), over 85 percent of the mercury at the particle control device
inlet was already in the non-elemental form. For the one bituminous coal-fired plant with low
mercury oxidation (S3), over 50 percent of the mercury at the SCR inlet was already in the oxidized
form. At the plant firing subbituminous coal (SI), mercury oxidation was fairly low. But, due
to the high carbon in that plant's fly ash, it is believed that the elemental mercury was adsorbed
onto the ash, resulting in high particulate mercury levels. Finally, in contrast with the findings
of Reference 10, ammonia appeared to have little or no effect on mercury oxidation on these actual,
full-scale facilities.
Subsequent tests on sister units at those plants and at other plants are shown in the second four
entries in Table 2. All of the units fired bituminous coal and showed that mercury oxidation was
generally enhanced to high levels of oxidized mercury at the SCR outlet. In each case where a
scrubber was installed, the mercury removal was high. For the unit with an ESP and no scrubber,
mercury removal was not improved by the SCR.
13
-------
Table 2. Summary of Results from Full-Scale SCR Mercury Oxidation Tests.
13 a
Power Plant
51 : 650 MW
Dyclone, ESP
52: 1360MW0
A/all, ESP+FGD
MELd)
S3: 750 MW
Tangential, ESP
54: 704 MW0 Cy-
;lone, Lime Ven-
:uri Scrubber
384 MW0 Wall,
ESP+FGD
300 MW0
Tangential, ESP
1 360 MW0 Wall,
ESP+FGD (MEL)
Dyclone, Lime
i/enturi Scrubber
Catalyst
Type
Space Velocity
(hr1)
Honeycomb
1800
Plate
2125
Honeycomb
-3930
Honeycomb
2275
"corregated"
-3750?
Honeycomb
3800
Plate
2125
Honeycomb
2275
Age
8000 hr
3.5
months
1 ozone
season
1 ozone
season
2
months
2 seas.
2 layers
repl.
after 1st
seas.
2 ozone
seasons
2 ozone
seasons
Coal
Type
PRB
OH
bit.
PA bit.
blend
KY bit.
PA/
WV
bit.
KY/
WV
bit.
OH
bit.
KY bit.
Sulfur"
(%)
0.2
3.9
1.7
2.9
3.6
1
3.9
3.1
Clb
(ppm)
<60
1640
1150
360
470
1000
520
750
bypass
250
w/SCR
Flue Gas Hg++
Content
(SCR in/out)
(1 OH sample)
8%/18%(Unit2)
(2 OH samples)
48%/91%
no effect of alkali
injection (Unit 1)
(2 OH samples)
55%/65%
(2 OH samples)
35%/61%for2nd
coal in sister unit
(2 OH samples)
9%/80%
Oxidation to >80%
(+38% net)
Oxidation to >80%
(+21% net)
Oxidation to >80%
(+33% net)
Oxidation to 60%
(+20% net) (more
oxidation if 1 data
outlier not used)
Flue Gas Hg++ Content at
PM Inlet
withoutwith SCR
(1 OH sample each)
5%: 8%
(2 OH samples each)
73%: 97%
77%:67% (possible filter
effect due to reactive
ash)
2nd coal/sister unit not
tested
(2 OH samples each)
56%: 87%
Oxidation to 95%
(+15%net) (using data
from sister unit w/o SCR)
Oxidation to 89% (-0%
net) (using data from
sister unit w/o SCR)
Oxidation to >95% (did
not test w/o SCR)
Oxidation to >90%
(+39% net) (Cl in coal
changed between tests)
Total Hg Removal
across PM+FGD
withoutwith SCR
(1 OH sample each)
60%:65% (within
experimental error)
(2 OH samples each)
51%:88%
FGD removed 94% of
Hg++
(2 OH samples each)
16%: 13% (within
experimental error)
2nd coal/sister unit not
tested
(2 OH samples each)
46%:90%
Oxidation to >90%
(+40% net)
No effect; actually lower
Hg removal in ESP (-6%
vs 23%)
-85% Hg removal (did
not test w/o SCR)
Oxidation to >90%
(+47% net)
Effect of
NH3onHg
Oxidation
(SCR
in/out)
No effect
Not tested
small neg.
effect. Not
tested in
2nd coal/
sister unit
Small
negative
effect
Not tested
Not tested
Not tested
Not tested
a NH3, Cl, SO3 sampled at SCR outlet unless noted
b Based on Energy Environmental Research Center's analyses
c Gross MW
d MEL = magnesium-enhanced lime scrubber
-------
Bench scale testing strongly suggests that HC1 is an important exhaust gas constituent that is
necessary for providing the chlorine for oxidation of Hg° to HgCl across the SCR catalyst.12 This
important result provides a scientific base for explaining the differences observed between coals
in field-testing. Subbituminous coals tend to have lower chlorine levels and higher calcium in
the ash than bituminous coals. Hence, they would be expected to produce exhaust gas with lower
HC1 concentrations than bituminous coals.
It is acknowledged that, at this point in time, the understanding of the effects of SCR catalyst on
mercury oxidation is not complete. There is a great deal to learn with regard to the science of this
phenomenon. However, apparently significant mercury oxidation by SCR catalyst occurs with
bituminous coal, and oxidation is less certain with other coals. In this work, when evaluating this
effect, it is assumed that when bituminous coals are being used, 90 percent of the mercury after
the SCR is in the non-elemental form and is captured by a downstream wet FGD. It is also assumed
that the SCR catalyst has no effect on mercury oxidation when other coals are fired.
3.2.2 Mercury Removal Though Combustion NOX Controls
The staged introduction of fuel and combustion air is a common practice for reducing formation
of nitrogen oxides. This is often achieved within the burner in LNBs and also through the use
of OF A when deeper staging and greater NOX reduction than afforded by LNBs alone is desired.
Air staging reduces NOX formation by causing fuel-bound nitrogen to be released from the fuel
at high-temperature and fuel-rich conditions. The fuel subsequently burns out under lower-
temperature, oxygen-rich conditions to ensure high combustion efficiency with low formation
of nitrogen oxide (NO). In the case of reburning (or fuel staging), a secondary, fuel-rich combustion
zone is introduced after the initial combustion zone to reduce the NO that was formed in the initial
combustion zone to nitrogen (N2). A downstream burn-out zone—effectively an OF A zone after
the reburn zone—provides complete combustion of the reburning fuel under oxygen-rich conditions.
Because all of these staged combustion methods used for minimizing NOX formation result in
delayed combustion when compared with combustion methods that do not try to minimize NOX
formation (and therefore burn the fuel only with maximum efficiency in mind), they also tend
to reduce combustion efficiency and increase the amount of unburned fuel—in the form of unburned
carbon (UBC), also known as loss of ignition (LOI). The UBC ends up in the fly ash that is collected
in the PM control device. This carbon in the fly ash may act to adsorb Hg° and Hg++. Therefore,
existing combustion controls might be expected to enhance removal of mercury from the exhaust
gases by downstream PM collection devices.
15
-------
Combustion of bituminous and low-rank (subbituminous and lignite) coals have been tested at
pilot-scale under simulated air staging conditions.14"16 In that effort, it was found that mercury
removal efficiencies by the downstream ESP improved with air staging. Up to 90 percent mercury
removal was achieved with bituminous coals through air staging. With low-rank coals, air staging
improved mercury removal from about 20 percent removal (without air staging) to about 40 percent
removal (with air staging). These tests confirmed the expectation that combustion NOX controls
can improve the mercury capture by the PM control devices.
3.3 Mercury Removal in SO2 Control Equipment
Both wet and dry flue gas desulfurization technologies are being used in the United States to control
SO2 emissions from coal-fired boilers. SDA is being used at the maj ority of the plants employing
dry FGD technologies. Available data reflects that some mercury capture occurs in wet FGD and
SDA systems.
3.3.1 Mercury Removal in Wet FGD
More than 20 percent of coal-fired utility boiler capacity in the United States uses wet FGD systems
to control SO2 emissions. In such systems, a PM control device is installed upstream of the wet
FGD scrubber. The PM control device used in combination with a wet FGD scrubber may be a
PS, ESPc, ESPh, or a FF baghouse. Wet FGD systems remove gaseous SO2 from flue gas by
absorption. In wet scrubbers, gaseous species are mixed with a liquid in which they are soluble.
For SO2 absorption, gaseous SO2 is mixed with a caustic slurry, typically water and limestone
or water and lime.
Gaseous compounds of Hg++ are generally water-soluble and can absorb in the aqueous slurry
of a wet FGD system. However, gaseous Hg° is insoluble in water and therefore does not absorb
in such slurries. When gaseous compounds of Hg++ are absorbed in the liquid slurry of a wetFGD
system, the dissolved species are believed to react with dissolved sulfides from the flue gas, such
as H2S, to form mercuric sulfide (HgS); the HgS precipitates from the liquid solution as sludge.
In the absence of sufficient sulfides in the liquid solution, a competing reaction that reduces/converts
dissolved Hg++ to Hg° is believed to take place. When this conversion takes place, the newly formed
(insoluble) Hg° is transferred to the flue gas passing through the wetFGD system. The transferred
Hg° increases the concentration of Hg° in the flue gas passing through the wet FGD (since the
incoming Hg° is not absorbed), thereby resulting in a higher concentration of gaseous Hg° in the
16
-------
flue gas exiting the wet FGD compared to that entering. Transition metals in the slurry (originating
from the flue gas) are believed to play an active role in the conversion reaction since they can act
as catalysts and/or reactants for reducing oxidized species.
The capture of Hg in units equipped with wet FGD scrubbers is dependent on the relative amount
of Hg++ in the inlet flue gas and on the PM control technology used. As described in Reference
8, ICR data reflected that average Hg captures in wet FGD scrubbers ranged from 23 percent for
one PC-fired ESPh plus FGD unit burning subbituminous coal to 97 percent in a PC-fired FF plus
FGD unit burning bituminous coal. The high Hg capture in the FF plus FGD unit was attributed
to increased oxidization and capture of Hg in the FF followed by capture of any remaining Hg++
in the wet scrubber.
3.3.2 Mercury Removal in SDA
More than 10 percent of coal-fired utility boiler capacity in the United States uses SDA systems
to control SO2 emissions.8 An SDA system operates by the same principle as a wet FGD system
using a lime scrubbing agent, except that the flue gas is mixed with a fine mist of lime slurry instead
of abulk liquid (as in wet scrubbing). The SO2 is absorbed in the slurry and reacts with the hydrated
lime reagent to form solid calcium sulfite and calcium sulfate. The heat of the flue gas evaporates
the water in the mist leaving dry solid particles of calcium sulfite and calcium sulfate. Entrained
particles (unreacted sorbent particles, reaction products, and fly ash) are captured in the downstream
PM control device (either an ESP or FF).
The performance of SDA systems in controlling SO2 emissions is dependent on the difference
between the SDA outlet temperature and the corresponding flue gas water vapor saturation
temperature. SDA systems on coal-fired boilers typically operate about 20 °F (11 °C) above the
saturation temperature (i.e., a 11 °C approach to saturation temperature). The relatively low flue
gas temperatures afforded by SDA systems increase the potential for mercury capture. The caking,
or buildup, of moist fly ash deposits, which can plug the SDA reactor and coat downstream surfaces,
dictates the minimum flue gas temperatures which can be employed at the outlet of SDAs.
Hgp is readily captured in SDA systems. Both Hg° and Hg++ can potentially be adsorbed on fly
ash, calcium sulfite, or calcium sulfate particles in the SDA. They can also be adsorbed and captured
as the flue gas passes through the ESP or FF, whichever is used for PM control. In addition, gaseous
Hg++ may be absorbed in the slurry droplets and react with the calcium-based sorbents within the
droplets. Nearly all of the Hgp can be captured in the downstream PM control device. If the PM
17
-------
control device is a FF, there is the potential for additional capture of gaseous mercury as the flue
gas passes through the bag filter cake composed of fly ash and dried slurry particles.
As described in Reference 8, ICR data reflected that units equipped with lime spray dryer absorber
scrubbers (SDA/ESP or SDA/FF systems) exhibited average Hg captures ranging from 98 percent
for units burning bituminous coals to 3 percent for units burning subbituminous coal. The
predominance of Hg° in stack gas units that are fired with subbituminous coal and lignite resulted
from low levels of Hg° oxidization.
3.4 Mercury Removal in Other Control Devices
Some units use particulate scrubber systems, primarily venturi scrubbers, to control PM emissions.
Capture of Hg in these systems is limited to soluble Hg compounds such as HgCl2. PS systems
are typically poor collectors of fine PM, and capture of Hgp by such scrubbers may be poor if the
Hgp in the flue gas is associated with fine PM. Hg° is insoluble and will not typically be captured
by the scrubber. It is possible to capture Hg++ in the wet scrubbers, but the scrubber chemistry
and the manner in which the scrubber is operated will determine whether it is effectively removed,
or whether it is stripped, from the scrubbing liquor. Stripping can occur if the Hg++ is not adsorbed
on the particles or reacted chemically with liquid-phase reactants within the scrubber.
Mechanical collectors such as cyclones do a poor j ob of capturing fine PM, and, in general, mercury
capture in these control devices should be limited to the capture of Hgp associated with particles
larger than 10 |j,m.
3.5 Models of Mercury Removal by Existing Equipment
As noted in the preceding sections, there are a number of parameters that impact the mercury
removal by existing equipment. Chlorine is widely acknowledged as having a role in mercury
removal. SO2 is also expected to have a role as well. Fly ash characteristics and the temperature
of the exhaust gas leaving the air preheater exit have also demonstrated a strong influence on
mercury removal. Of course, the equipment type plays an important role as well.
Expressions approximating the effects of equipment type, coal chlorine content, and SO2 level
on mercury removal have been developed through statistical analysis of ICR data.18 Data used
by the Reference included equipment type, coal chlorine, and sulfur information but not some
18
-------
of the other parameters expected to influence mercury removal from existing equipment. Since
these expressions do not include other effects expected to be significant, such as ash characteristics
and gas temperature, they should only be used for approximations. These algorithms are
Algorithm 1 (ESPc):
/existing equipment = Ct x hi [(coal Cl, ppm)/(SO2, in Ib/MMBtu)] + C2 Eq. 1
Algorithm 2 (all other categories):
/existing equipment = Q X hi (coal Cl, ppm) + C2 Eq. 2
Where /existing equipmen, is the fraction of mercury removed by existing equipment. These same
algorithms were shown in Reference 18 to also provide a means to approximate the remaining
mercury that is in the elemental form. There are minimum and maximum allowable values that
set the allowable range for the results of Equations 1 and 2. Tables 3a and 3b show values for
Cj and C2 and minimum and maximum values to use in Equations 1 and 2 for estimating fraction
of mercury removed by existing equipment (Table 3a) and the fraction of remaining mercury that
is elemental (Table 3b).
Table 3a. Parameters Used for Equations 1 and 2 Which Estimate Mercury Removal by Existing
Equipment18
Existing Equipment
ESPc
ESPc + wet FGD
ESPh
ESPh + WET FGD
FBCa + FF
FF
FF + wet FGD
SDA + ESP
SDA + FF
Ci
0.1233
0.1157
0.0927
0.2845
0.1394
0.1816
0.1943
-0.1087
0.2854
C2
-0.3885
-0.1438
-0.4024
-1.3236
0.1127
-0.4287
-0.2385
0.6932
-1.1302
Minimum
(%)
0.0
24.0
0.0
4.0
66.0
40.0
79.0
5.0
0.0
Maximum
(%)
55.0
70.0
27.0
65.0
99.0
85.0
96.0
25.0
99.0
FBC = fluidized bed combustor
19
-------
Table 3b. Parameters Used for Equations 1 and 2 Which Estimate Percent of Remaining Mercury
in Gas that Is Elemental Mercury18
Existing Equipment
ESPc
ESPc + wet FGD
ESPh
ESPh + WET FGD
FBCa + FF
FF
FF + wet FGD
SDA + ESP
SDA + FF
Ci
-0.1283
-0.039
-0.1639
-0.0945
-0.1198
-0.1182
-0.426
-0.0355
-0.1125
C2
1.23
1.11
1.55
1.45
1.2
0.88
3.1
1.13
1.48
Minimum
(%)
12.0
81.0
34.0
80.0
44.0
30.0
45.0
91.0
64.0
Maximum
(%)
85.0
98.0
91.0
99.0
68.0
33.0
84.0
98.0
99.0
FBC = fluidized bed combustor
The correlations of Reference 18 should be used only for making approximate estimates.19 Although
the algorithm will provide reasonable estimates in most cases, Reference 19 showed that factors
not addressed by these algorithms, such as fly ash characteristics or gas temperature, can have
a significant effect on the mercury capture in existing facilities.
The capacity of PAC to adsorb mercury is large enough that it should not be limiting except at
temperatures of about 350 °F (177 °C) or more, which is greater than the gas temperature at the
exit of most air preheaters. So, with the possible exception of lignite coals, cooling usually has
little or no beneficial effect on mercury absorption by PAC. However, the ability of fly ash and
unburned carbon in the fly ash to absorb mercury is far less than that of PAC and may be enhanced
by cooling. Therefore, although spray cooling may enhance mercury adsorption by fly ash and
downstream capture in the ESP or FF, it i s not expected to enhance mercury capture by PAC except
in the case where lignite coals are burned. According to Reference 8, over 90 percent of the coals
burned in the United States are bituminous, subbituminous, or blends of bituminous and
subbituminous, which are not likely to use spray cooling with PAC injection. Therefore, spray
cooling, which was evaluated in Reference 4, is not evaluated here.
20
-------
4.0 EMERGING CONTROL TECHNOLOGIES
Based on published literature,1'4'19"30 control technologies using inj ection of P AC into the flue gas
appear to hold promise for reducing mercury emissions from utility boilers. These technologies
have been applied successfully on municipal waste combustors (MWCs). Despite differences
between MWCs and utility boilers (e.g., mercury concentration and speciation in the flue gas),
full-scale and pilot-scale tests indicate that these technologies may be able to provide significant
mercury removal from the flue gas of coal-fired utility boilers. Accordingly, this evaluation focused
on the characterization of performance and costs of PAC injection-based technologies.
Other technologies have shown promise for control of mercury and other pollutants such as SO2,
NOX, and PM. These multipollutant control technologies may offer cost-effective mercury control
when considering the combined control of mercury with control of other pollutants. Multipollutant
control technologies evaluated in this effort include ECO, advanced dry FGD, and a coal treatment
technology known as K-Fuel.
This section begins with the description of PAC injection-based control technologies that can
be retrofitted to existing boilers for control of mercury emissions, PAC injection estimates for
these technologies, multipollutant control technologies that control mercury and other pollutants,
and model plants used in this work. Subsequently, control technology applications on model plants
used to develop cost estimates are discussed. Finally, this section discusses ECO, advanced dry
FGD, and K-Fuel.
4.1 PAC Injection-Based Technologies
Injection of PAC for mercury emissions control has been developed and tested at the full scale
on coal-fired utility boilers. Test programs have been performed on a utility boiler firing
subbituminous coal with a downstream cold-side electrostatic precipitator (ESPc), on utility boilers
firing bituminous coal with a downstream ESPc, and on a utility boiler firing bituminous coal
with a Compact Hybrid Particle Collector (COHPAC) arrangement (upstream ESPh with
downstream baghouse after the air preheater). 19'25~29 Using the data from these test programs and
from pilot-scale testing, performance models were developed for PAC injection based mercury
control applications.19'30 These models are in a form where they can be updated as new information
is developed on these applications and for other boiler applications.
21
-------
4.1.1 Mercury Removal Models
If/equipmentlS equal to the fraction of mercury removed from the boiler gases by a specific piece
of equipment, then (1 -/quipment) equals the fraction of mercury remaining in the gases after that
specific equipment. The fraction of mercury remaining after n pieces of equipment is equal to
IA J equipment I/ \ 7/equipment 2/ \ J equipments/ ... ^A J equipment n/J 4*
Therefore, the total mercury removal fraction,/^!, is
/Total 1 IA1 /equipment I/ V ^ ~J equipment 2/ V1 /equipments/ • • • V1 /equipment n/J •'-'I' ^
If one of the pieces of equipment is PAC injection, then the total mercury removal fraction is
./Total IA ./equipment I/ \ ~Jequipment 2/ \ ./equipment 3/ ... ^ A J PAC injection/ • • •
x (I _ / . Y| Ea 5
V* J equipment n/J J-^M- ^
where/PACmjection is the fraction of mercury removed by PAC injection.
If PAC injection is simply added to existing equipment and the removal effects of the existing
equipment are combined into one term, then we can represent Equation 5 as
J Total IA J existing equipment/ \ J PAC injection/J
and, solving for/PACinjection
* PAC injection ~ *• ~ IA1 "/Total)' I1 "/existing equipment)] Eq. O
where/existing equipment is the removal fraction associated with the existing equipment and may be
approximated by Equations 1 and 2 in Section 3.5 of this document if the removal by existing
equipment is not known. Given a total mercury reduction requirement and knowing the reduction
by existing equipment, it is possible to determine how much additional reduction is necessary
from PAC injection.
In this research, data from full-scale and pilot-scale tests of mercury reduction were used to
formulate models for mercury reduction from existing equipment and from PAC inj ection. Full-scale
22
-------
data for mercury removal by existing equipment are available from the ICR data. Full-scale testing
results of mercury reduction from PAC injection are available from the Department of Energy's
field testing programs at Southern Company's Gaston Plant, Wisconsin Electric Power Company's
Pleasant Prairie Power Plant (PPPP), and at PG&E Corp. National Generating Group's Brayton
Point and Salem Harbor Plants.19
4.1.2 Mercury Reduction by PAC Injection
Reference 4 has algorithms developed from pilot-scale data for mercury reduction on boilers
equipped with PAC injection. The following model improvements, discussed in Reference 19,
have been made:
1. The algorithms of Reference 4 were developed from pilot-scale tests and characterize total
mercury reduction from both PAC injection and existing equipment as a function of PAC
injection concentration. When using the algorithms of Reference 4, it is necessary to have
a different PAC injection algorithm for each type of equipment configuration, including
upstream and downstream equipment. These PAC inj ection algorithms may have to be updated
as new information regarding mercury control from existing equipment becomes available.
As described in Reference 19, the mercury reduction from PAC injection was isolated from
that of the other equipment in Equation 5. Therefore, as more information on reduction of
mercury from equipment other than PAC injection is developed, it should not be necessary
to perform new regressions on the PAC inj ection models. Also, using Equation 5, it will also
be possible to assess the fate of mercury in equipment that is either upstream or downstream
of the PAC injection system.
2. The algorithms of Reference 4 are of a form in which it is possible to approach 100 percent
mercury removal by inj ection of very high concentrations of PAC. As demonstrated at a full-
scale demonstration at the Pleasant Prairie Power Plant, under some circumstances the mercury
reduction by PAC inj ection can be limited to something well below 100 percent.28 Therefore,
the algorithm for mercury reduction from PAC inj ection was modified as described in Reference
19 and in the following paragraphs to permit an upper limit to mercury removal that may be
less than 100 percent.
3. Because the algorithms of Reference 18 for mercury reduction from existing equipment are
based on the full-scale ICR data, it is desirable to use them to characterize mercury reduction
from existing equipment. However, it is not possible to integrate the algorithms of Reference
23
-------
4 into the approach used in Reference 18. By treating the mercury reduction from PAC inj ection
independently from mercury reduction from other equipment, it is possible to use the algorithms
of Reference 18 to characterize mercury reduction from existing equipment.
Because mercury reduction by PAC injection may be limited to a value well below 100 percent,
as identified in the second point, the equation that is used in Reference 4 to characterize the
relationship between mercury reduction and PAC injection
% reduction = • - 100 x/fromPACinjection = 100-[A/(M+B)C] Eq. 7
where M is the mass injection rate of PAC (in Ib/MMacf) so that
M={[A/(100-')](1/C)}-B Eq. 8
was modified in Reference 19 to be
M={[A/((100xD)-')](1/C)}-B Eq. 9
where D is the fraction of mercury reduction that is asymptotically approached.
A set of constants A, B, C, and D are specified for a given existing plant configuration, coal type
(bituminous or subbituminous), PAC sorbenttype, and retrofit configuration (PAC alone or PAC
plus retrofit fabric filter). These constants are based upon full-scale data where available and based
upon pilot-scale data or input from experts in this technology where full-scale data are not available.
Reference 19 showed that, for systems with FFs, all of the PAC-based sorbents appeared to offer
similar performance in terms of PAC injection concentration (in Ib/MMacf) necessary for a given
mercury reduction. On the other hand, for units with ESPs and without a fabric filter, PAC selection
did have a significant effect on performance.
The constants A, B, C, and D are determined based upon a PAC injection matching key, which
is a five-digit number composed of the following elements:
• Existing Particle Control Equipment (10000, 20000, 30000 for FF, ESPc, ESPh,
respectively)
• The SO2 control technology, if any (0, 1000, or 2000, or 3000, for none, FGD, SD, or
advanced dry FGD, respectively)
24
-------
• The fuel type: (100 for Bituminous or 200 for Subbituminous)
• Whether an additional FF will be retrofit (0 or 10 for none or FF, respectively)
• The PAC capacity (1 for high, 2 for medium, and 3 for low—In Reference 19, PAC
selection appeared to make a difference for facilities with ESPs but not for facilities
with FFs)
For example, ESPc SD Bituminous FF med (Matching Key # 22112) indicates a Bituminous coal-
fired boiler currently equipped with a spray dryer absorber and an ESPc that will retrofit PAC
injection (medium capacity) and a fabric filter. The list of constants used for Equation 9 in this
work is shown in the Appendix 2. The matching key above will be used to determine the set of
constants used for Equation 9 to estimate the PAC injection concentration.
Equation 9 is used to determine PAC injection rate is the following manner. If/existingequipment is
greater than or equal to_/jotal, then no additional mercury removal is necessary, and addition of
PAC inj ection or any other technology to remove mercury is unnecessary. However, if/existingequipment
is less than/Total, additional mercury removal is necessary through retrofit of another technology,
in this case PAC injection. Using Equation 6, it is possible to determine the amount of mercury
reduction that must be performed by injection of PAC. Keeping in mind from Equation 7 that
• •= 100 x /g.om PAC injections
Equation 9 is then used to determine the injection concentration (M) of PAC (in Ibs/MMacf). M
is then multiplied by the total gas flow rate to determine the inj ection rate of PAC (in Ibs per hour).
4.2 Emerging Control Technologies
Certain emerging technologies appear to offer significant potential for the combined reduction
of mercury and SO2, or of mercury, NOX, and SO2. The technologies of interest that are evaluated
in this report include
• Electro catalytic oxidation (ECO)
• Advanced dry FGD
• K-Fuel
25
-------
These technologies generally have limited commercial experience on coal-fired utility boilers,
but experience with the technologies in other applications or experience with related technology
may be extensive. In the case of ECO, for example, this is a unique technology with limited
experience beyond the development and demonstration efforts currently underway in the United
States. Some elements of the ECO process—ammonia scrubbing and wet ESPs—are well-
established technologies, however. Wet ESP's, while not used widely in the utility industry, have
a large experience base in other industries. Advanced dry FGD has experience on other applications
and is also closely related to spray dryer absorber technology, which has extensive experience
on coal-fired utility boilers. K-Fuel is an approach for beneficiating western fuels, especially Powder
River Basin coals. K-Fuel removes moisture (increasing heating value), nitrogen (reducing NOX),
sulfur (reducing SO2 emissions), and mercury (reducing mercury emissions). K-Fuel has been
tested on utility boilers and a commercial production plant is under construction. Therefore, each
of these technologies appears to have promise, may be used in the near term, and can be analyzed
with some confidence. However, because of the limited experience with these technologies in
coal utility applications, the results of the analysis shown here should be considered preliminary.
,31-34
4.2.1 Electro Catalytic Oxidation (ECO)
ECO technology has been developed by Powerspan. It is expected that the ECO system would
be installed downstream of the existing ESP or FF, as shown in Figure 5.
I1 L1 . I
Ul&t-lllL-iL
f=m
Figure 5. Location of ECO installation in a power plant.31
26
-------
This technology consists of
• A dielectric barrier discharge reactor that induces oxidation of pollutants
• A two-loop ammonia scrubber tower that removes SO2 and water-soluble oxidized forms
of the pollutants
• A wet ESP that removes acid mist and fine particles
• A co-product (saleable fertilizer by-product) processing and mercury removal system
that removes mercury with carbon filters and crystallizes the ammonium nitrate and
ammonium sulfate fertilizers from the scrubber tower.
In the dielectric barrier discharge reactor, the following happens:
• NO gas forms nitrogen dioxide (NO2) gas and nitric acid (HNO3) aerosol mist.
• SO2 gas forms sulfur trioxide (SO3), leading to the formation of sulfuric acid (H2SO4)
aerosol mist.
• Hg° vapor forms HgO particles.
Figure 6 shows a detailed process flow diagram for an ECO application.
|
|
j
|
CffSUfe
tat
Figure 6. Detailed process flow diagram of ECO.33
27
-------
The water-soluble forms of the oxidized pollutants—HNO3 and H2SO4—are removed in an absorber
tower that is equipped with a wet ESP. The absorber tower is a two-stage process with an absorption
stage (at the top of the tower) for absorbing the pollutants and a concentrating stage at the bottom.
An aqueous ammonia solution is used as the scrubbing agent to absorb SO2 and convert the absorbed
SO2, nitric acid and sulfuric acid to ammonium nitrate and ammonium sulfate, respectively. NO2
reacts with ammonium sulfite [(NH4)2SO3], which forms in the scrubbing liquor, to form ammonium
sulfate [(NH4)2SO4] andN2. With 90 percent NOX reduction, about 40 percent of the NOX becomes
nitrate and about 50 percent becomes NO2 and ultimately N2. Thus, some of the NOX is ultimately
converted to ammonium nitrate and the remainder is ultimately converted to N2. All of the SO2
removed is ultimately removed in the form of ammonium sulfate. Mercury collected in the absorber
tower water is removed from the liquid stream by an activated carbon filter. Acid mist and fine
particles removed in the wet ESP drain into the absorber tower and are removed in the liquid
discharge of the lower loop. Due to chemistry considerations of the absorber tower, the ECO process
works best if the SO2 to NOX ratio in the flue gas is equal to 3 or more on a molar basis.
The consumables and the by-products of the ECO process include
Consumables
• Electric power for the barrier discharge reactor, pumps and blowers. The power for
the discharge reactor is related to the NOX reduction desired.33
• Heat for the by-product crystallizer.33
• Ammonia reagent, which can be estimated as roughly two moles of ammonia per mole
of SO2 removed. Additional amine is provided by other chemicals discussed below.
• Make-up water for the ab sorption tower—about 1 gal/min per MW—no special quality
specification.33
• Carbon filters for mercury removal from the liquid discharge of the absorber.33
• Additional, proprietary chemicals that provide the balance of the amine for the conversion
of NOX to ammonium nitrate and SO2 to ammonium sulfate. These are estimated at
around $150/ton of NOX removed and $15/ton of SO2 removed.33
By Products
• Ammonium nitrate and ammonium sulfate crystals that can be sold as fertilizer.
Typically, for 90 percent reduction for every mole of inlet NOX, 0.40 moles of ammonium
nitrate are produced, and one mole of ammonium sulfate is produced for every mole
of SO2 reduced.33
• Mercury captured on the activated carbon (a waste to be disposed of) at a cost of about
$1000/lb of mercury captured.33
28
-------
• A small amount of coal fly ash that was not captured by the ESP is filtered out of the
liquid stream to the fertilizer crystallizer.33
• Water vapor.33
Experience
The ECO technology has been demonstrated in the laboratory and on a 2000 scfm pilot at First
Energy's Burger Plant. The pilot has been operating since March 2002. A 50 MW commercial
demonstration system is currently being built at Burger Station. The commercial demonstration
unit at Burger Plant is designed to handle 110,000 scfm of gas flow.34
Capital Cost
AmerenUE, Sargent & Lundy, Wheelabrator, The Andersons, and Powerspan performed a detailed
cost estimate of an ECO unit at AmerenUE's Sioux plant. The capital cost of an ECO system for
this 510 MW installation was estimated at $ 114,500,000, inclusive of process equipment, general
facilities, owner's costs, and contingencies. This also included the fertilizer plant and balance
of plant modifications.35 It is the only comprehensive, full-scale cost analysis that has been made
available publicly. Therefore, a cost of $200/kW is a reasonable estimate to use. Reference 33
confirmed this estimate.
Operating Cost
Variable operating cost is the cost of power and other consumables. Ammonia consumption is
determined by the molar ratio described above under Consumables. Specialty chemical costs are
estimated at $ 150/ton of NOX removed and about $ 15/ton of SO2 removed, based upon information
from Reference 33. Reference 33 also provided information on power consumption requirements
for both the ECO reactor and auxiliaries, described in the next section. Carbon filter replacement
costs and the costs of disposal of used carbon filters are estimated at $ 1000/lb of mercury removed.
Fertilizer value, which produces a revenue stream that offsets a portion of the cost, is approximated
at $110/ton of fertilizer produced.33 The ammonium sulfate and ammonium nitrate fertilizer are
widely traded commodity chemicals and their value will depend largely on market conditions at
the time and transport costs.
Fixed Operating Costs include an estimated 1.5 percent of process capital per year plus 3 operators
and one maintenance person per shift.33 The manpower needs are not expected to be a significant
function of unit size.33
29
-------
Pollutant Removal Efficiencies and Power Consumption
Because ECO is a true multipollutant process that can remove NOX, SO2, mercury, and fine particles,
the effectiveness of the process can depend upon several variables, and there may be some
interrelationships. However, atypical coal system would be designed for the removal efficiencies
shown in Table 4.33
Table 4. ECO Pollutant Removal Efficiencies33'35
Pollutant
S02
NOX
Hg
Fine Particles
Removal Rate
98%-99%
90%
80%-90%
95%
Conditions
Any inlet condition
Up to 250 ppm or up to about 0.450 lb/MMBtua
Any inlet condition
(Outlet to less than 0.004 Ibm/MMBtu)
1 At higher NOX levels, 90% reduction is achievable at higher ECO reactor power levels.
The ECO system tends to operate best when the molar SO2 to NOX ratio is 3.0 or greater.33 Although
ECO can operate at lower ratios, the reliability of both performance and cost predictions may be
somewhat less.
The power for the dielectric barrier discharge reactor is largely determined by both the amount
of NO oxidation needed and the gas flow. To increase the amount of NOX removed by the ECO
process, it is necessary to increase reactor power. So, for a given percent of NOX reduction, the
reactor power is roughly proportional to the NOX mass flow. Therefore, to achieve a low outlet
NOX level while minimizing power demand, it is best to start with a low NOX level from the boiler.
As a result, one would typically use an ECO system in combination with lowNOx burners or other
devices to minimize NOX into the ECO reactor. Based upon information from Reference 33, this
work assumes that reactor power (in watts/scfm) is equal to the lesser of 20 watts/scfm or
58.22»(NOX) - 6.2431 (see Figure 7), where NOX is measured in Ib/MMBtu. Reactor power could
potentially be higher than 20 watts/scfm; however, this would likely be unattractive when compared
to reducing NOX by other means such as low NOX burners.
Other power demands include fan power to overcome about 9 inches of water total pressure drop
(calculated as actual volume flow times pressure drop with an assumed fan efficiency of 65 percent)
and another estimated 0.75 percent of plant output for auxiliary loads for the absorber and fertilizer
30
-------
plants. Although the fertilizer auxiliary costs are more closely related to the SO2 being removed,
a single percentage is used for simplicity.33
25 T
£ 20 -I
o
I
TJ c 1 R
E| ^
35 -«»
? 10
o
Q.
5 H
0 J
0.1
= 58.22x- 6.2431
0.2 0.3 0.4
NQV (Ib/MMBTUI
0.5
Figure 7. ECO power consumption versus NOX.
33
Calculation Methods
A worksheet titled ECO Cost and Performance was developed. Inputs for other calculations were
largely taken from CUECost worksheets and the Plant Configuration Inputs and the Constants_CC
worksheets. Inlet gas conditions are taken as the outlet of the air preheater except for particle
loading, which is reduced due to the ESP or FF. Relationships for mass balances were developed
from the information provided above regarding consumables, and relationships for power
consumption were developed from information discussed regarding power consumption.
As noted earlier, the reliability of the algorithms used for predicting ECO system performance
and cost may be somewhat reduced for molar SO2 to NOX ratios below 3.0. Therefore, the worksheet
includes warnings when the inlet SO2 to NOX ratio is below this value. Low SO2 to NOX does
not mean that the output is incorrect or that ECO cannot be used. It simply means that the
calculations are somewhat less reliable, and it would be advisable to consult with the technology
supplier for more information.
31
-------
4.2.2 Advanced Dry FGD36'3941'47
Dry FGD reacts hydrated lime slurry with exhaust gases in an absorber vessel to capture SO2, SO3,
HC1, and mercury that may be present in the exhaust gases. The exhaust gases are maintained
above saturation temperature to avoid condensation. The by-product of dry scrubbing can usually
be landfilled safely or, in some cases, reutilized for another purpose.
Advanced dry scrubbing is intended to improve performance over conventional spray dryer
absorbers. Advanced dry scrubbing utilizes a fluidized bed or a flash dryer for the reactor with
recycle loops to enhance lime utilization. Figure 8 shows a diagram of the FLS AirTech Gas-Solid
Absorber installed upstream of an ESP. Many systems use venturi mixers to introduce the sorbent
slurry. The high solids loading in the reactor provides high gas-solids interaction and high solids-
solids interaction. The high solids-solids interaction allows reactive particle surface to continuously
be exposed as the particles impact one another. The solids-solids interaction also causes some
agglomeration of fine particles to form larger particles that are easier to capture in downstream
equipment. Because the reactions occur more efficiently in the reactor and because the particles
are larger, Advanced dry scrubbing does not require a downstream fabric filter for the SO2 removal
reaction to be completed. When a cyclone is used for recycle to produce a lower outlet particle
concentration, a smaller downstream particle collection device is needed. In this case, the technology
can often be installed upstream of the existing ESP, as shown in Figure 9. Without the cyclone,
it is likely that the existing ESP will require some modifications to improve collection efficiency,
or a new particle collection device may be necessary. However, in many cases it is envisioned
that the absorber will be installed downstream of the existing ESP so that fly ash does not get
contaminated and can continue to be sold. In this case, an additional particle removal device will
be necessary, as shown in Figure 10.
Some fly ash is beneficial to mercury capture in the absorber, especially for subbituminous coal.36
Reference 36 also indicated that 98 percent mercury capture is possible for bituminous coal. The
fly ash also helps to improve the qualities of the solid by-product, making it more suitable for
use as fill or for sale of cement-like products.
32
-------
jj"* SasaBsssswsBwsJ.I
|_|
N |
Figure 8. F.L. Smidth AirTech Gas-Solids Absorber (GSA).
37
Existing ESP
or Fabric
Filter
Absorber/recycle unit —
Not all systems use cyclones. For
some systems, recycle is taken from
downstream particle control de-rice
Figure 9. Installation of an advanced dry FGD upstream of an existing ESP.
33
-------
Absorber/recycle unit
Not all systems use
cyclones. For some
systems, recycle is taken
from downstream particle
control device
Figure 10. Installation of an advanced dry FGD downstream of an existing ESP.
There are a number of companies that supply different versions of advanced dry FGD technology:
Lurgi Lentj es, F.L. Smidth AirTech' s Gas Suspension Absorber (GS A), RJM Beaumont's Rapid
Absorption Process (RAP), and WULFF. The advanced dry scrubbing systems all use hydrated
lime slurry as the principal reagent to remove SO2 and can remove mercury as well. Figure 11,
which is the published performance of three different advanced dry scrubber systems,40'43'46 shows
that the SO2 removal efficiency is a strong function of stoichiometric ratio, approach-to-saturation
(ATS, expressed in degrees Fahrenheit), and chlorine content. Data is shown for chlorine contents
of 0.04 percent (400 ppm) and 0.12 percent (1200 ppm). Stoichiometric ratio is defined as moles
Ca/molesSO2. As shown, ATS, fuel chlorine content and lime stoichiometry determine the SO2
removal efficiency. For any given ATS, chlorine level, and stoichiometric ratio, all technologies
offer similar SO2 removal. For example, note that the GSA 18 deg and 1200 ppm Cl line, if
continued, would closely follow the RAP 18 degree line. Since all versions of advanced dry FGD
offered by the different suppliers rely on the same chemistry, similar performance would be expected
for a specific set of conditions.
Although advanced dry FGD has not been as widely used as SD A technology, there is significant
operating data on it for pulverized coal-fired applications. Moreover, the chemistry of SO2 capture
is very similar to that of SDA technology. For the analysis of this report, algorithms were developed
to predict performance. To calculate lime feed necessary for a particular level of SO2 reduction,
Figure 11 was re-plotted as stoichiometry versus SO2 reduction, and curves were fit based on ATS
and chlorine content of the coal. The results are shown in Figure 12.
34
-------
g 85
§
D_
esi
o
m
_____
0.9
-GSA8 deg,400 ppm Cl
GSA18 deg,400 pprn Cl
-GSA18 deg,1200ppmCI
•••••™*:•'•=•'•'WULFF - Theiss
" °" RAP-18deg
" ;» RAP 30 deg
1.1 1.2 1.3 1-4 1.5 1.6
Fresh Lime Stoichiometry (moles Ca/Moles SO2)
1.7
Figure 11. Advanced dry scrubber SO2 removal performance.
5 w
o *
o o
c
,1 «t
1.7
1.6
1.5
1.4 -
1.3 -
1.2 -
1.1 -
1.0 -
0.9
<> 8 deg,4UOppm Cl
D 18 deg. 400 ppm CI
A 18deg,1200 ppm Cl
x 30 deg, 1200 ppm Cl
* 30 deg 400 pprn Ci
-Poly. (3 deg, 400 pprn C|
•=--- Poly. (18 deg, 400 pprn Cl)
— Poly. (18deg, 1200 ppm Cl)
—Poly. (30 deg, 1200 pprn Cl)
:-Poly, (30 deq 400 pprn Cl)
y = 4.2332x2 - 4.9916x +2.6369
\
.7683
y = 22.333?? - 37.012x +16.329
60% 65% 70% 75% 80% 85% 90% 95% 100%
SO, Removal
Figure 12. Advanced dry scrubber performance.
35
-------
At this time, some suppliers of this technology limit the gas-flows per reactor to an equivalent
of around a 125-150 MW plant.38 On the other hand, Lurgi has supplied single-reactor systems
as large as 250 MW and expects to be able to supply single reactor systems as large as about 400
MW.39 WULFF has supplied single-reactor systems as large as 300 MW.40 As a result, in some
(but not all) cases it may be necessary to have more absorber towers than for conventional spray
dryers. For this reason, a design of a RAP system for a 500 MW boiler may require as many as
four reactors and a 250 MW system may require as many as two reactors.41 Some other suppliers
may not require as many reactors.
Limited experience has shown high mercury removal rates. Testing of mercury control from the
Roanoke Valley Energy Facility showed that the Lurgi-designed system provided in excess of
95 percent mercury removal from the 55 MW Unit #2 while firing eastern bituminous coal.42 For
subbituminous coals, where there is lower chlorine content and a greater proportion of elemental
mercury versus total mercury as compared to bituminous coals, lower capture efficiencies may
be expected with normal lime sorbent. However, activated carbon or other specialized sorbents
may offer the potential to capture more mercury. RJM-Beaumont Environmental Systems offers
oxidized calcium sorbent that they claim is effective in removal of SO2 and mercury.43
The consumables and the by-products of the ECO process include
Consumables
• Electric power for the booster fan, pumps, blowers, and reagent preparation
• Lime - normally delivered as dry lime and hydrated to hydrated lime
• Water - for producing hydrated lime and for gas cooling
• Costs associated with downstream particle removal (energy, fabric filters, etc.)
The algorithms of Figure 19 are used to calculate lime feed necessary for a particular level of SO2
reduction.
By Products
• Calcium sulfate, calcium sulfite, a small amount of unreacted sorbent, ash and other
solids collected in the downstream particle collection device. Based upon review of
process designs from different suppliers, the approximate amount of this is 2.5-2.8 Ibs/lb
of sorbent inj ected. Most often the by-products are landfilled, but they can occasionally
be re-used.
• Water vapor that goes up the stack
36
-------
Experience
This technology has been widely used on incineration systems and to a lesser degree on coal-fired
boilers. Lurgi has supplied the technology to a 250 MW plant in Puerto Rico, to the 55 MW
Roanoke Valley facility in North Carolina, to the Black Hills Power & Light (Gillette, WY) plant,
and to nine coal-fired plants in Europe.44 WULFF has supplied the technology for the 300 MW
coal- and oil-fired Theiss plant in Austria. There is significant commercial experience on coal
as a result of these installations.40
Capital Cost
Information from technology providers suggests that the installed cost of a reactor and all material
handling equipment excluding any downstream particle control device (ESP, FF, or cyclone) is
roughly $35/kW for 250 MW or 500 MW systems.41'45 It may then be necessary to add the cost
of the equipment and installation of a FF or an ESP, associated material handling equipment,
ductwork, fans, and other equipment, which could cost up to an additional $60/kW. Total cost
of process capital, therefore, is in the range of $ 100/kW. Additional costs associated with the proj ect,
such as engineering and construction management, allowance for funds during construction,
contingency, and general facilities would increase the total cost to above $ 100/kW. This is consistent
with Reference 46, which estimates the cost of an F.L. Smidth AirTech Gas Suspension Absorber
system at $149/kW (1990 dollars) for a 300 MW plant. It is unclear from Reference 46 whether
or not this cost includes the downstream ESP, but it does include the cost of the cyclone separator
in the recycle system. Table 5 provides a cost estimate for a Lurgi Lentjes CFB-FGD system at
a 500 MW plant and shows a $ 150/kW cost estimate. Therefore, a cost of about $ 150/kW appears
to be consistent among several sources and will be used here.
Operating Costs
The principal variable operating costs are for lime reagent, water, power, and solid disposal. The
preceding discussion described how lime consumption is determined. Water consumption (city
water is considered adequate quality—demineralized water is not required) is determined by the
slurry concentration. Booster fan power for the reactor and cyclone is estimated from gas flow
rate (actual cubic feet per minute) and pressure drop assuming a blower efficiency of 65 percent,
and material handling power is estimated at 0.1144 hp/lb/h of lime feed.
37
-------
Table 5. Estimated Cost of CFB-FGD System for a 500 MW Plant Burning PRB Coal
Subsystem
Reagent Feed Syatem
SO2 Removal System
Particulate Collector
Flue Gas System
Waste Handling and Recycle System
General Support Equipment
Miscellaneous Equipment
Total Process Capital (TPC)
General Facilities (A, 5% of TPC)
Engineering and Construction Management (B)
Project Contingency (15% of TPC+A+B))
Total Plant Cost (TC)
Allowance for Funds (3.2% of TC)
Owner's Cost (5% of TC)
Total Plant Investment (TPI)
Inventory Capital (spare, 1 % of TPI)
Royalties
Total Capital Requirement
Cost, U.S. $
5,000,000
7,000,000
16,000,000
7,000,000
7,000,000
4,000,000
6,000,000
52,000,000
2,600,000
5,200,000
8,970,000
68,770,000
2,201,000
3,439,000
74,410,000
744,000
0
75,154,000
$/kW
10.0
14.0
32.0
14.0
14.0
8.0
12.0
104
5.2
10.4
17.9
137.5
4.4
7.0
148.9
1.5
0
150
4.2.3 K-Fuel
K-Fuel is a beneficiated coal that is derived from western coal. The resulting fuel is lower in ash,
higher in Btu value, and produces lower pollutant emissions than untreated western subbituminous
or PRB coals. K-Fuel uses a pre-combustion process that improves the quality of the coal—
including removing the mercury, moisture, ash, sulfur, and some of the fuel NOX precursors—before
the coal is burned at the power plant. Because these constituents are removed prior to combustion,
the need for post-combustion controls may be reduced.
The K-Fuel Coal Beneficiation Process484950
The K-Fuel process employs both mechanical and thermal means to increase the quality of the
coal by removing moisture, rock, sulfur, mercury, and other heavy metals. To begin the process,
coal is delivered to the K-Fuel processing plant from the mine. The coal enters the first stage
separator, developed using conventional coal cleaning technology, where it i s crushed and screened
38
-------
to remove the large rock and rock material. The processed coal is then passed on to an intermediate
storage facility prior to being sent to the next stage in the process, as shown in Figure 13.
As-Mired ^
Coal
Hi
Dik
i
ling
•foil
L
K-Fuel
First Stage
S eparafcr
Upgiaded .^
Coal
K-Fuel
Tli£321ial
Piocess
tlpgiaded ^
High-Btu
Fuel
Sul
Miu
Cap
i
K-F
S ecoi*
Sepa
for,
srals
tare
I
uel
Stage
latoz
Low
Fuel
Power
Hart
Figure 13. Overall schematic of K-Fuel processing plant.48
From the intermediate storage facility the coal is sent via a distribution system to the K-Fuel thermal
processing unit (Figure 14). This process employs Lurgi Mark IV vessels under high pressure
and temperature to place thermal stress on the coal. The coal passes through pressure locks into
the processors, and then steam at 460 °F (238 °C) and 485 psi is inj ected into the processors. While
the coal is maintained at these conditions, the mineral inclusions are fractured under the thermal
stress, removing both the included rock (which contains some mercury) and sulfur-forming pyrites.
The inherent moisture of the coal is liberated as well.
Coal
\.
¥•
Waste water
to tfeateieni
V
" K-Puel Has™ Praduct
Qualified
Disposal
\ Site
Figure 14. K-Fuel thermal processing plant.4
39
-------
After it has been treated in the main processor, the processed coal is discharged into a second
pressurized lock, which is then sealed off from the primary reactor. After sealing, the processor
pressure is vented into a water condenser to return the processor to atmospheric pressure, and
to flash cool the coal to approximately 200 °F (93 °C). The coal is then discharged onto a belt
and further cooled by convection and indirect cooling. After cooling, the coal is sent on to a second
stage separator for additional screening to remove sulfur- and mercury-containing mineral matter
which has been liberated by the thermal process. Mercury that is released from the coal during
thermal treatment is captured in a carbon filter. The carbon filter can then be disposed.
Pollutant Reduction Performance
Test burns of coal treated with the K-Fuel Process showed reductions in NOX and SO2 emissions.
Figure 15 shows SO2 and NOX emissions for two coals that were tested—with and withoutK-Fuel
treatment—and emissions from an eastern bituminous coal.48
0.8
0,7
0.6
0.5
0.4
0.3
0.2
0.1
0
QDry Fork
OK -Fuel from Dry Fork
EjFort Union
jljK-Fuel from Fort Union
• Lone Mountain
ib/MBtuSO-
NO,
Figure 15. SO2 and NOX emissions from test burns of K-Fuel and untreated
fuels.
Because the K-Fuel process simultaneously reduces the mercury content of the coal and increases
its Btu value, there is a significant overall mercury emission reduction on a heating value basis.
Table 6 below compares the fuel analysis of a typical PRB coal used for the analysis of this report
40
-------
and a similar analysis of K-Fuel that was assembled from information provided by KFx
Corporation.49 The table shows that mercury content of the K-Fuel derived from Cordero Mine
is about 43 percent lower on a mass basis and about 60 percent lower on a Btu basis than the PRB
coal used in this comparison. This is consistent with the roughly 65 to 70 percent mercury reduction
reported in Reference 48 for the K-Fuel process when applied to different PRB coals.
Table 6. Comparison of Typical PRB Coal with K-Fuel
Contents
Coal Type
PRB
K-Fuel
Proximate Analysis (ASTM, as received)
Volatile Matter (wt%)
rixed Carbon (wt%)
30.79
32.41
100.00
40.20
45.50
99.62
Ultimate Analysis (ASTM, as received)
Moisture (wt%)
Carbon (wt%)
Hydrogen (wt%)
Mitrogen (wt%)
Chlorine (wt%)
Sulfur (wt%)
<\sh (wt%)
Oxygen (wt%)
Total (wt%)
Mercury (mg/kg)
Modified Mott Spooner HHV (Btu/lb)
30.40
47.85
3.40
0.62
0.03
0.48
6.40
10.82
100.00
0.07
8304.
7.50
66.70
4.80
1.00
0.03
0.38
6.42
13.20
100.03
0.04
11,718.
Coal Ash Analysis (ASTM as received)
SiO2 (wt%)
M2O3 (wt%)
riO2 (wt%)
re2O3 (wt%)
CaO (wt%)
MgO (wt%)
Ma2O (wt%)
<20 (wt%)
°205 (wt%)
SO3 (wt%)
Other Unaccounted for (wt%)
Total (wi%)
31.60
15.30
1.10
4.60
22.80
4.70
1.30
0.40
0.80
16.60
0.80
100.00
28.40
17.30
1.60
6.00
23.50
4.00
1.40
0.27
2.43
13.63
1.47
100.00
41
-------
Experience
The K-Fuel Process has been tested at the pilot stage, and construction of a commercial plant is
planned for this year.48'50 Lurgi is providing much of the process reactor technology. The first
commercial plant is to be built at the Black Thunder mine in Wright, WY. Construction is currently
scheduled to start in mid-2003, with completion expected in the first half of 2004. The plant is
expected to produce more than 700,000 tons per year of K-Fuel Plus. This plant is intended to
not only prove the commercial value of the technology but to also provide a basis for optimizing
the technology, and thus, it is designed for the possibility of future expansion on site.48'50
Technology Cost
For the purpose of this program, the cost of the K-Fuel is going to be assessed on the basis of
incremental fuel cost. K-Fuel may be utilized on any facility where it can be economically
transported to the site.50 According to Reference 50, some interior eastern bituminous fuel-fired
boilers might consider K-Fuel over PRB because K-Fuel will avoid retrofits that may be necessary
to avoid derates that might otherwise occur when switching from bituminous coal to PRB coal.
Reference 50 discusses possible delivered prices ranging from $20 to $32 per ton of coal. At this
price range and a heating value of about 11,700 Btu/lb, the K-Fuel has an estimated price of about
$1.20-$! .36/MMBtu, which makes it competitively priced on a Btu basis with many bituminous
coals and somewhat more expensive than PRB coals. $20/ton would equate to only about
$0.85/MMBtu, which makes it cost competitive with (and in some cases less expensive than) PRB
coals according to the fuel cost information in Reference 50. There is a chance that some retrofit
costs might be incurred by switching from bituminous or PRB coals to K-Fuel, but these capital
costs will be relatively small and will be dominated by the fuel costs in most cases. As a result,
the cost analysis performed here will focus on incremental fuel cost as the major cost of using
K-Fuel. Incremental costs of control may be incurred if mercury removal technologies, such as
PAC injection, are retrofit to reduce mercury emissions further.
42
-------
5.0 TECHNOLOGIES CURRENTLY UNDER DEVELOPMENT
There are several multipollutant control technologies that are currently under development or nearing
commercial status in the United States. These, and other technologies, may emerge in a relatively
short time depending on the results of on-going tests as well as the demand for mercury and
multipollutant controls. Others may never reach commercial stage either because of technical
problems or because of unfavorable economics relative to competing technologies.
For the purpose of this report, technologies currently under development were divided into three
groups: employing oxidation (other than SCR), utilizing sorbents, and other technologies.
5.1 Oxidation Technologies
Oxidation technologies aim at oxidizing mercury in order to facilitate its subsequent removal in
wet FGD or wet ESP. Examples of these technologies include LoTOx, PEESP, and ESP
modification discussed below.
LoTOx
LoTOx is a gas phase low-temperature oxidation system which involves injection of ozone in
the flue gas upstream of a wet FGD to oxidize NOX to higher oxides of nitrogen such as N2O5,
and mercury to HgO. Subsequently, these compounds are removed in a wet FGD because they
are water soluble. The LoTOx system consists of an integrated oxygen/ozone generation unit
complete with ozone injection system either into the LoTOx reactor or directly into the exhaust
duct prior to the wet FGD (if sufficient residence time can be provided). Ozone is produced in-situ
and on demand by passing oxygen through a conventional industrial ozone generation system.
It is produced in response to the amount of NOX present in the flue gas generated by the combustion
or process source. Theoretically, there is the potential for oxidation of SO2 to SO3; however, as
proven in field testing, the reaction rates are very low compared to the predominant NOX and Hg
reactions.
PEESP
Plasma-Enhanced Electrostatic Precipitators (PEESP) technology offers the potential to enhance
the ability of wet ESP to remove elemental mercury. PEESP oxidizes vapor phase elemental mercury
into oxidized form and then removes it within the wet ESP process. This technology involves
inj ection of a reagent gas mixture, through a corona discharge needle that is attached to the central
43
-------
electrode within an electrostatic field. Inj ection into the area surrounding the sharp discharge point
results in generation of hydroxyl radicals, ozone, and other reactive compounds. These react with
elemental mercury vapor to form oxidized mercury particles. These negatively charged particles
are attracted to the positively charged collecting electrode where they are collected. The mercuric
oxide particles and other absorbed pollutants are removed during the wash-down cycle of the wet
ESP. PEESP can be incorporated in an existing wet ESP by modifying the central electrode to
inj ect the reagent gas. Greater than 80 percent total mercury removal is proj ected at pilot and full
scale; bench scale testing demonstrated mercury removal of up to 83 percent.
ESP Modification for Mercury Oxidation
In this arrangement, a catalyst, most likely in honeycomb form, is inserted into the flue gas path
upstream of the FGD system (last section of the ESP). This placement provides for low velocity
and a relatively particulate-free flue gas. As a result, a close-pitched catalyst can be used.
Downstream of the catalyst, the oxidized mercury is scrubbed in the FGD absorber, and co-
precipitates with the calcium sulfite or gypsum byproduct. Preliminary cost estimates show that
a catalytic process, if installed upstream of a wet FGD system, should allow plants to achieve 90
percent overall mercury control at a cost that is 5 0 percent less than by inj ecting activated carbon.
However, the cost of the process will depend largely on the catalyst life and required catalyst
volume. So far, only catalysts produced at a laboratory scale have been tested.
5.2 Sorbent Technologies
Sorbent technologies utilize improved dry sorbents to accomplish mercury removal in systems
without wet FGD. Additionally, improvements in sorbents may help reduce the cost of controlling
mercury. Novel sorbents discussed here include Amended Silicate, MerCAP, and Pahlmanite.
Amended Silicate
Amended silicate sorbents use a commodity substrate material impregnated with a chemical additive
that binds mercury to the surface of the particles. The sorbents have been prepared in a number
of formulations tailored to provide economic mercury recovery in multiple applications. In packed-
bed tests these sorbents showed mercury removal of several times that of activated carbon. Pilot
plant tests demonstrated 70-96 percent mercury capture at injection rates of 1.6-9.1 Ib/MMacf,
respectively. Out of the total removal, the inj ected amended silicate removed 40 percent or more
of the mercury in the first one second of contact time in the pilot tests. Good stability of captured
mercury on the sorbent was observed. When samples of sorbent mixed with fly ash were collected
44
-------
from the pilot fabric filter hopper and subjected to Toxicity Characteristic Leaching Procedure
(TCLP) tests; the leachate mercury levels were below the limit of detection. Preliminary process
design and analysis indicate that amended silicates can be cost competitive with activated carbon
for mercury control.
Mercury Control Adsorption Process
Mercury Control Adsorption Process (MerCAP) deploys a mercury adsorbing sorbent-coated
cartridge placed in the flue gas duct at temperatures of 400 °F (204 °C) or less. Mercury is removed
from the flue gas as it flows past the rigid structure. Once the cartridge is saturated with mercury,
it can then be removed or regenerated in-situ. Sorbent materials that have been considered for
MerCAP configuration include activated carbon or metals which can amalgamate with mercury,
such as gold and silver. Pilot scale experiments indicated that a MerCAP with gold coated plates
approximately 10 ft long and spaced 0.5 inches apart was placed downstream of the spray
dryer/fabric filter could remove more than 80 percent of mercury from a low-rank fuel flue gas.
In preliminary tests, MerCAP did not perform as well in nonscrubbed flue gas.
Pahlman Process
The Pahlman Process is a dry sorbent system comprised of two discrete steps. One step involves
capturing target pollutants such as NOX, SOX, mercury, and particulates using Pahlmanite dry
mineral sorbent compounds. The other step involves the regeneration of the spent or partially spent
sorbent compounds for reuse and the separation and isolation of useful by-products such as nitrates
and sulfates for use in fertilizers and industrial chemicals. The Pahlmanite sorbents are low-density
oxides of manganese (MnO2) in the form of fine black powder. The sorbent is inj ected in a reactor,
which operates at temperature between ambient and 320 °F (160 °C). The technology is in pilot
scale stage; a trailer-mounted pilot plant is available which has been tested at a number of power
plants using flue gas slipstreams. Testing indicated above 99 percent SO2,93.6-96.6 percentNOX,
and up to 67 percent mercury reduction.
ROFA and ROTAMIX
ROFA (rotating opposed fire air) is a combustion control technology that employs staged
combustion to reduce NOX emissions. In the staged combustion, fuel is initially burned in a fuel-rich
zone for low NOX generation. The fuel-rich zone is followed by an OF A system that induces rapid
mixing of burn-out air with furnace gases in the upper furnace region using highly-turbulent, rotating
flowto complete fuel combustion while minimizingNOX formation. When sorbent orNOx-reducing
chemicals (such as ammonia or urea) are added to the OFA ports, the technology is referred to
as ROTAMIX. TheNOx-reducing chemicals reduce NOX through selective non-catalytic reduction
45
-------
reactions. The sorbent may be added for adsorption of SO2 or mercury. Full-scale tests at Cape
Fear Unit 5, firing eastern bituminous coal, showed that, in addition to the NOX removal provided
by the staged combustion and the baseline mercury removal from the existing ESP, up to 89 percent
removal of mercury, 64 percent removal of SO2 and 4 percent removal of NOX was achieved by
limestone injection. Addition of Trona produced results of up to 67 percent removal of mercury,
69 percent removal of SO2 and 11 percent removal of NOX. Slagging of the superheater by sorbent
and ash was found to be a problem during the tests. However, it is believed that this problem might
be avoided in the future through selection of inj ection locations at a lower temperature where ash
softening is not as great.
5.3 Other Technologies
This group of technologies includes hybrid processes that accomplish mercury removal in a modified
baghouse (Promoted MB Felt), by a combined dry and wet process (multipollutant control process),
or in a regenerable sorbent bed (activated coke). These processes are presented below.
Promoted MB Felt
This process is centered on a proprietary, low-pressure, mercury-capturing filter fabric (Promoted
MB Felt) incorporated into a pulse-jet baghouse. The fabric is designed in a way that allows for
mercury capture to be segregated from particulate control, thereby avoiding fly ash contamination.
Bench-scale experiments at 185 °C demonstrated approximately 75 mg Hg/(g filter medium) were
captured. Fligh mercury capture was later confirmed in a pilot plant over the course of seven weeks.
Multipollutant Control Process
This process involves sequential injection of dry sorbent and liquid oxidant to accomplish SO2
removal and oxidation/removal of NOX, and Hg in a multipollutant control reactor. The reactor
consists of three sequential tubular sections followed by a fabric filter. The first vertical section
is a humidifier, which is followed by two vertical sorption sections. Dry sorbent inj ection is located
approximately one third from the top of the first sorption section. Liquid oxidant is dispersed into
the flue gas at the top of the same section. Preliminary results from a 1 MW slipstream of a 300
MW low sulfur coal-fired boiler indicated SO2, NOX, and mercury removals across the reactor
of up to 92, 80, and 68 percent, respectively. The extent of oxidation and subsequent removal
of NOX and mercury was a function of the type and amount of oxidizer used.
46
-------
The Activated Coke Process
This process involves three steps: adsorption, desorption, and (optional) by-product recovery.
In the first step (adsorption), flue gas passes through a bed of activated coke slowly moving
downwards in a two-stage adsorber. The activated coke consists of carbon with large porous inner
surface area. In the first stage, sulfur dioxide is removed by adsorption into the activated coke,
where it forms sulfuric acid or ammonium hydrogen sulfate [NH4HSO4]. Mercury can also be
removed by adsorption on the coke at a rate of up to 1.7 mg/g of activated coke at a temperature
below 180 °C and condensation in the middle of desorber where the coke is about to be heated
for regeneration. Therefore, mercury can be removed by extracting the coke in the middle of
desorber. One method being considered for removing mercury is the use of a selenium filter, which
absorbs the mercury from the flue gas and forms HgSe, a chemically stable compound. The selenium
filter i s expected to have 98 percent Hg collection efficiency during the filter life (usually 4-5 years).
47
-------
48
-------
6.0 COSTS OF REDUCING MERCURY EMISSIONS
The costs of controlling any pollutant are composed of capital costs associated with installation
of the equipment and the operating costs associated with operating the equipment. In this report,
these costs are estimated and assessed on a mills/kWh (or $/MWh) basis. In this effort, costs are
determined on a constant dollar basis—that is to say thatthe costs are represented in 2003 dollars
and the effects of general inflation are, therefore, normalized. We also assume that the escalation
of operating costs equals the general inflation rate. Therefore, inflation is assumed to offset
escalation so that the levelization factor for operating costs is equal to 1.0.
The approach to assess costs included the use of EPA's CUECost model and additional worksheets
that are specific to the technologies of interest. EPA's CUECost model was used to estimate flue
gas conditions, establish basic economic parameters, and perform cost and performance calculations
for those technologies already integrated into CUECost. This approach was used to first estimate
the costs for model plants under specific conditions. Then, the cost impacts of some selected
variables are determined.
6.1 Mercury and Multipollutant Control Cost Models
For this work, mercury removal is from existing air pollution control equipment as well as from
additional equipment such as PAC injection. These models were described in Sections 3.5 and
4.1, and the PAC inj ection rate algorithm constants for each equipment configuration are provided
in the Appendices.
Costs are comprised of capital and operating costs. These costs are assessed to develop a total
annual cost of pollution control expressed in mills/kWh or $/MWh.a The total installed capital
cost is annualized to produce an annual charge. This is done by multiplying the total installed capital
charge by a capital recovery factor (CRF). The CRF is a function of variables such as proj ect life,
cost of capital, tax rate, depreciation methods, and others. In this analysis, a CRF of 0.133 (or
13.3 percent) was chosen to be consistent with Reference 4. The annualized capital charge is then
divided by the total power output of the plant for the year to determine the annual capital cost
contribution to electric cost in mills/kWh (or $/Mwh).
aCosts expressed in mills/kWh and $/MWh are numerically equal.
49
-------
Operating costs are estimated by determining the sum of the annual cost of consumables—reagents,
power, water, etc. that contribute to variable operation and maintenance (O&M) and the annual
cost of additional operators, maintenance, or parts (that contribute to fixed O&M) and dividing
that sum by the total power output of the plant for the year to determine the operating cost
contribution to electric cost in mills/kWh (or $/MWh).
The total annual cost of pollution control (in mills/kWh) is determined by adding all of the cost
components—annual capital cost, annual variable O&M, and annual fixed O&M.
It i s important to note that the costs of the multipollutant control technologies are likely to be greater
than those for mercury control only. However, these technologies are providing additional pollution
control benefits over and above mercury control.
In the tables that follow, estimates of capital and total annual cost for mercury and multipollutant
control technology applications on model plants are shown. These estimates were determined
with the cost model discussed. It is noted that these estimates are based on currently available
data.
While developing the cost estimates for the model plant applications, the following specifications
were used with the cost model.
1. Mercury concentration in the coal was taken to be 0.10 mg/kg for eastern bituminous
coal and 0.07 mg/kg for subbituminous coal. These concentrations are in the range
of concentration reported for utility boilers in Reference 72.
2. PAC injection rate correlations (see Section 4.1, Appendices and Reference 19)
generally reflect that PAC inj ection requirements increase nonlinearly with an increase
in mercury removal efficiency. To characterize the impact of this behavior, model plant
cost estimates were obtained for mercury removal efficiencies of 50, 60, 70, 80, and
90 percent wherever possible. In some cases existing equipment provided in excess
of 50 percent removal and PAC inj ection was not needed to achieve the specified level
of reduction. For PAC injection with a downstream ESP, 90 percent reduction may
not be possible with subbituminous coals without retrofit of a downstream pulse jet
fabric filter (PJFF). For bituminous coal fired boilers with an ESP, 90 percent removal
may not be cost effective by PAC injection alone when compared to PAC injection
and retrofit of a downstream PJFF to achieve 90 percent mercury removal.
50
-------
3. Spray cooling was not used in any of these model runs because PAC has sufficiently
high capacity for most temperatures of interest—air preheater (APH) exit temperature
under 350 °F— that any temperature effect is expected to be small. Moreover, spray
cooling may have adverse effects on high-sulfur fuel boilers [due to acid dew point
(ADP) effects] and PRB fuel boilers (due to cement-like properties of the ash).
However, at lignite coal-fired plants, which are not evaluated here, spray cooling might
be used to improve mercury removal. No data are currently available for recycling
of sorbent in technology applications utilizing PAC inj ection and PJFF. Accordingly,
no sorbent recycle was used.
4. Wet FGD performance for mercury control is determined by Equation 2 if no SCR
exists or 90 percent removal if the boiler fires bituminous coal and is equipped with
an SCR. No oxidation (or co-benefit) by SCR is assumed for subbituminous coals.
If PAC is added to provide additional reduction of mercury, then PAC is added upstream
of the ESP or FF.
5. In each of the model plant cost determinations, a plant capacity factor of 65 percent
was used.
6. The cost of PAC was taken to be $l,000/ton of carbon.
7. In some cases, it is assumed that PAC is added upstream of the existing particulate
control equipment. In others, particularly for high removal rates, it is assumed that
a downstream FF is added. In the case of spray dryer absorbers, it is assumed thatPAC
is added upstream of the spray dryer, and a fabric filter may be added between the
upstream PAC injection point and the downstream spray dryer. This is because the
removal of HC1 by the spray dryer will adversely affect the ability of PAC to achieve
reasonable removal rates. This will require a larger fabric filter than if the fabric filter
were installed downstream of the existing paniculate control device because, in the
upstream arrangement, the fabric filter would need to be sized to capture all of the
fly ash as well as the injected PAC.
8. The multipollutant control technologies evaluated—ECO, advanced dry FGD, and
K-Fuel—were designed to provide other benefits (e.g., reduction of NOX or SO2
emissions, improvement in fuel heating value) besides mercury reduction. Therefore,
the higher cost of these multipollutant control approaches over control methods
51
-------
developed solely for mercury control should be considered with special attention to
the greater environ-mental and other benefits associated with multipollutant control
approaches. For ECO, it was assumed that 98 percent SO2 removal and 90 percent
NOX removal were provided in addition to 85 percent mercury reduction. For advanced
dry FGD, it was assumed that 90 percent SO2 removal was achieved in addition to
about 95 percent mercury reduction. For K Fuel, the reduction of emissions of other
pollutants, especially SO2 and NOX, will depend upon how the K-Fuel compares to
the base fuel. For example, the K-Fuel has 44 percent lower sulfur content on a heating
value basis than the base DOE PRB. Therefore, the environmental benefits of these
methods—over and above mercury reduction—can be very significant.
9. Costs include capital and operating costs associated with the retrofit and the expected
cost of a continuous emission monitoring system (CEMS). The cost of existing
technology is not included in the cost estimate. However, the effect of existing tech-
nology is included in the total mercury removal performance. In some cases, existing
technology will provide adequate mercury removal and no additional mercury removal
technology is required. In these cases, a small cost associated with the expected cost
of the mercury CEMS will be shown.
10. In this analysis it was assumed that the percent mercury removal possible from
additional controls was not affected by the mercury removal from existing controls.
While it is possible that there may be some interaction, this is not expected to be a
significant effect for the cases evaluated here.
11. In all of the cases evaluated here, the cost calculations assumed that all collected fly
ash is currently sold. Therefore, calculations for PAC inj ection configurations in which
fly ash and PAC are collected together include incremental costs to landfill fly ash
at a cost of $30/ton. In many cases these costs will not be incremental because fly ash
may currently be landfilled orbecause fly ash may not be rendered completely unaccep-
table for re-use. The large maj ority of plants currently landfill their flyash,30 and PAC
inj ection would increase disposal costs only in proportion to PAC usage for them. Also,
in situations where flyash is currently sold, fly ash contaminated with some used PAC
might still be beneficially reused depending upon the amount of PAC added, the
properties of the fly ash, and the intended use of the sold ash. According to ASTM
Standard C618-03, coal fly ash with carbon contents as high as 6 percent may be
acceptable as a concrete additive.73 There are other criteria that may determine
52
-------
acceptability of the fly ash as an additive to a buyer. However, the presence of small
amounts of carbon in the fly ash will not necessarily render it unacceptable for beneficial
re-use. Disposal costs can be a significant portion of total costs. As will be shown in
a sensitivity analysis, elimination, or at least a reduction, of disposal costs can
significantly improve economics of PAC inj ection from what is presented in this section.
The ash content of the coal (on a heating value basis) largely determines the magnitude
of this cost impact. So, coals with higher or lower ash contents will have higher or
lower ash disposal costs, respectively. Based upon the characteristics of the fuels used
in this program, the cost of landfilling fly ash at $30/ton is estimated to be about 0.37
mills/kWh for the low sulfur bituminous cases with PAC inj ection. Similarly, for the
subbituminous and high-sulfur bituminous cases with PAC injection, the cost of
landfilling fly ash at $30/ton is estimated to be about 1.01 mills/kWh and about 0.93
mills/kWh, respectively. Because these costs, which may not apply in some cases, are
included as incremental costs in the results shown whenever a PJFF is not added to
segregate the fly ash from the injected PAC, the cost estimates shown here should be
regarded as likely to overestimate the cost in this respect.
6.2 Fuel Types, Plant Characteristics, and Model Plant Cases
Four different fuel types were evaluated for estimating costs of mercury control options presented
in the following chapter:
1. A high sulfur bituminous coal
2. A low sulfur bituminous coal
3. APRS Coal
4. A special, subbituminous fuel from K-Fuel.
The first three fuels are taken from Reference 4, where the fuel information was developed by
NETL of the Department of Energy. The properties of these fuels are identified in Table 7 below.
53
-------
Table 7. Fuels Used In Model Plant Analysis
Contents
Coal Type
High Sulfur
Bituminous
Low Sulfur
Bituminous
PRB
Subbituminous
K-Fuel
Proximate Analysis (ASTM, as received)
Volatile Matter (wt%)
Fixed Carbon (wt%)
40.40
47.50
100.00
44.00
50.00
100.00
30.79
32.41
100.00
40.20
45.50
99.62
Ultimate Analysis (ASTM, as received)
Moisture (wt%)
Carbon (wt%)
Hydrogen (wt%)
Nitrogen (wt%)
Chlorine (wt%)
Sulfur (wt%)
Ash (wt%)
Oxygen (wt%)
Total (wt%)
Mercury (mg/kg)
Modified Mott Spooner HHV (Btu/lb)
3.10
69.82
5.00
1.26
0.12
3.00
9.00
8.70
100.00
0.10
12,676.
2.20
78.48
5.50
1.30
0.12
0.60
3.80
8.00
100.00
0.10
14,175.
30.40
47.85
3.40
0.62
0.03
0.48
6.40
10.82
100.00
0.07
8304.
7.50
66.70
4.80
1.00
0.03
0.38
6.42
13.20
100.03
0.04
11,718.
Coal Ash Analysis (ASTM, as received)
Si02 (wt%)
AI203 (wt%)
TiO2 (wt%)
Fe2O3 (wt%)
CaO (wt%)
MgO (wt%)
Na2O (wt%)
K2O (wt%)
P205 (wt%)
S03 (wt%)
Other Unaccounted for (wt%)
Total fwt%')
29.00
17.00
0.74
36.00
6.50
0.83
0.20
1.20
0.22
7.30
1.01
100.00
51.00
30.00
1.50
5.60
4.20
0.76
1.40
0.40
1.80
2.60
0.74
100.00
31.60
15.30
1.10
4.60
22.80
4.70
1.30
0.40
0.80
16.60
0.80
100.00
28.40
17.30
1.60
6.00
23.50
4.00
1.40
0.27
2.43
13.63
1.47
100.00
Costs for installing and operating the mercury control technologies described in previous sections
are estimated with model plants. Approximately 75 percent of the existing coal-fired utility boilers
in the United States are equipped with ESPs for the control ofPM.8 The remaining boilers employ
FFs, particulate scrubbers, or other equipment for control of PM. Additionally, units firing medium-
to-high sulfur coals may use FGD technologies to meet their SO2 control requirements. Generally,
54
-------
larger units firing high sulfur coals employ wetFGD, and smaller units firing medium sulfur coals
use spray dryers. While developing the model plants, these PM and SO2 control possibilities were
taken into account.
Several model plants, with various flue gas cleaning equipment configurations and firing either
bituminous or subbituminous coal, were used in this work. Table 7 exhibits fuels used in model
plants. Power plant characteristics are given in Table 8, and model plants are shown in Table 9.
Note that boiler sizes of 100 and 975 MW used in this work were selected to approximately span
the range of existing boiler sizes and to be consistent with the size of the model plants used in
previous work.4 It was also envisioned that use of SCR can enhance oxidation of mercury in flue
gas and result in the "co-benefit" of increased mercury removal in wet FGD. Since SCR is a capital-
intensive technology, generally its use is cost-effective on larger boiler sizes. Accordingly, in this
work, the mercury co-benefit resulting from SCR use was evaluated for model plants utilizing
large (975 MW) boilers and wet FGD.
Table 8. Power Plant Characteristics
Characteristic
vlW Equivalent of Flue Gas to Control System
\let Plant Heat Rate
Dlant Capacity Factor
Total Air Downstream of Economizer
\\r Heater Leakage
\\r Heater Outlet Gas Temperature
nlet Air Temperature
Ambient Absolute Pressure
^ressure after Air Heater
Moisture in Air
Units
MW
Btu/kWh
%
%
%
°F
°F
in. of Hg
in. of H2O
Ib/lb dry air
Value
100, 300, 500, 975
10,500
65
120
12
300
80
29.4
-12
0.013
\sh Split
Fly Ash
Bottom Ash
Seismic Zone
%
%
integer
80
20
1
55
-------
Table 9. Mercury Control Technology Applications and Co-benefits
Model
Plant
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
Size
(MW)
975
975
975
975
975
300
300
300
300
300
975
975
975
975
975
975
975
975
975
975
975
975
975
975
975
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
975
975
975
100
100
100
Coal
Type3
Bit
Bit
Bit
Bit
Bit
Bit
Bit
Bit
Bit
Bit
Bit
Bit
Bit
Bit
Bit
Bit
Bit
Bit
Bit
Subbit
Subbit
Subbit
Subbit
Subbit
Subbit
Bit
Bit
Bit
Bit
Bit
Bit
Bit
Bit
Bit
Bit
Bit
Bit
Subbit
Subbit
Subbit
Subbit
Subbit
Subbit
K-Fuel
K-Fuel
K-Fuel
K-Fuel
K-Fuel
K-Fuel
%S
3
3
3
3
3
3
3
3
3
3
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.5
0.5
0.5
0.5
0.5
0.5
3
3
3
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.5
0.5
0.5
0.5
0.5
0.5
0.4
0.4
0.4
0.4
0.4
04
Existing Controls
ESP+FGD
FF+FGD
ESPh+FGD
ESP
ESP
ESP+FGD
FF+FGD
ESPh+FGD
ESP
ESP
ESP
FF
ESPh
ESP
FF
ESPh
ESP
FF
ESPh
ESP
FF
ESPh
ESP
FF
ESPh
SD+ESP
SD+FF
ESPh+FGD
ESP
FF
ESPh
ESP
FF
ESPh
ESP
FF
ESPh
ESP
FF
ESPh
ESP
FF
ESPh
ESP
FF
ESPh
ESP
FF
ESPh
Additional Controls
PAC, PAC +PJFF
PAC, PAC +PJFF
PAC, PAC +PJFF
Adv Dry FGD
ECO
PAC, PAC +PJFF
PAC, PAC +PJFF
PAC, PAC +PJFF
Adv Dry FGD
ECO
PAC, PAC +PJFF
PAC, PAC +PJFF
PAC +PJFF
ECO
ECO
ECO
Adv Dry FGD
Adv Dry FGD
Adv Dry FGD
PAC, PAC +PJFF
PAC, PAC +PJFF
PAC +PJFF
ECO
ECO
ECO
PAC, PAC +PJFF
PAC, PAC +PJFF
PAC +PJFF
PAC, PAC +PJFF
PAC, PAC +PJFF
PAC +PJFF
ECO
ECO
ECO
Adv Dry FGD
Adv Dry FGD
Adv Dry FGD
PAC, PAC +PJFF
PAC, PAC +PJFF
PAC +PJFF
ECO
ECO
ECO
PAC, PAC +PJFF
PAC, PAC +PJFF
PAC +PJFF
PAC, PAC +PJFF
PAC, PAC +PJFF
PAC +PJFF
Co-benefit
Case(s) with
SCR
SCR
SCR
SCR
SCR
SCR
' Bit = bituminous coal; Subbit = subbituminous coal
56
-------
6.3 Cost Model Results
Three costs are typically shown in the following tables. One is the capital cost expressed in $/kW.
This is the one-time capital charge for the equipment, installation, start-up, and such. Total
annualized cost (Total Cost in the tables) and the Variable Cost, both expressed in mills/kWh,
are also shown. Total Cost includes the annualized capital cost as well as annual variable and fixed
operating cost. The Variable Cost is only the portion of the Total Cost that is attributable to variable
operating and maintenance costs. In the following tables and discussion, the Model Plant numbers
are noted so that more detailed cost information can be reviewed in the model plant tables of
Appendix 4.
For those applications, particularly PAC inj ection, in which multiple mercury control levels were
evaluated (i.e., 50 through 90 percent), several columns are shown in the tables presented. Each
column indicates the estimated capital cost ($/kW) and total annual cost (mills/kWh) for the plant
size indicated and whether or not a PJFF was added. In some cases, existing controls provided
adequate mercury reduction, so the only items added were mercury emissions monitoring equipment
(no PAC injection or PJFF). Estimated mercury emissions on mg/kWh basis are also indicated
in the tables. This estimated mercury emissions number may be useful for estimating total mercury
mass emissions for particular configurations.
Only one level of mercury removal is shown for ECO and advanced dry FGD. Multiple columns
and costs are estimated for sensitivity to different variables, such as capital cost.
6.3.1 High Sulfur Bituminous Coals (Model Plants 1-10, 26-28)
For all of the model plants in which high sulfur bituminous coal is fired, it was assumed that the
boiler was equipped with an FGD technology that might include limestone forced oxidation (LSFO),
SDA, ECO, or advanced dry FGD. In the case of boilers equipped with wet FGD (LSFO), it was
assumed any necessary additional mercury control was performed through injection of PAC
upstream of the existing particulate removal device or between the existing particulate removal
device and a new downstream PJFF. The co-benefit of SCR was also evaluated assuming that
the combination of SCR with LSFO would result in the greater of 90 percent mercury removal
or the amount of mercury removal from the existing particle removal equipment and LSFO.
57
-------
For high sulfur units in which the co-benefit of SCR with LSFO (or wet FGD) is evaluated, the
size range chosen was 3 00 MW to 975 MW. The reason 3 00 MW rather than 100 MW was selected
as the low end of the size range is that SCR plus LSFO is a capital-intensive approach for combined
SO2 and NOX control. Therefore, it would be unlikely to be selected over other approaches for
SO2 and NOX control on a unit as small as 100 MW. For ECO and advanced dry FGD, the same
size range was evaluated for consistency with the LSFO cases. On the other hand, SDA was not
assessed for large, high-sulfur fuel boilers because it would rarely be economical for SO2 removal
on such large boilers when compared to LSFO.
High Sulfur Coal ESP plus FGD (Model Plants 1, 6)
As shown in Tables 1 Oa and 1 Ob, existing equipment (ESPc and wet FGD) are expected to provide
68 percent mercury removal under the conditions of Model plant #1. Mercury removal by PAC
injection is necessary for higher mercury removal. Under these conditions, to achieve 90 percent
mercury removal, a PJFF downstream of the ESP and PAC inj ection will permit more economical
removal through PAC injection. With the SCR, PAC injection is not expected to be necessary
for achieving over 90 percent removal of mercury, but PAC inj ection may be necessary for higher
than 90 percent mercury removal.
High Sulfur Coal FF plus FGD (Model Plants 2, 7)
When a facility is equipped with a fabric filter and an FGD system, it is expected that no additional
mercury removal will be necessary because 96 percent mercury removal is expected from existing
equipment. In this case SCR co-benefit is not significant because mercury removal is expected
to be high already.
58
-------
Table 10a. High Sulfur Coal, ESP plus FGD Without SCR Co-benefit (Model Plants 1, 6)a
Parameter
Hg Reduction of Existing Equipment (%)
Hg Reduction by PAC (%)
Total Outlet Hg (mg/MWh)
Specified Hg Reduction (%)
50
67.7
none
12.2
60
67.7
none
12.2
70
67.7
7.3
11.3
80
67.7
38.2
7.5
90
67.7
69.1
3.8
375 MW and No PJFFb
Retrofit PJFF?b
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Costb (mills/kWh)
no
$0.094
0.000
0.003
no
$0.094
0.000
0.003
no
$1.601
1.195
1.242
no
$2.437
1.447
1.520
no
$4.304
2.175
2.303
J75 MW with PJFF
Retrofit PJFF?
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Cost (mills/kWh)
no
$0.094
0.000
0.003
no
$0.094
0.000
0.003
yes
$36.216
0.215
1.122
yes
$36.322
0.234
1.144
yes
$36.538
0.278
1.195
300 MW and No PJFFb
Retrofit PJFF?b
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Costb (mills/kWh)
no
$0.126
0.000
0.004
no
$0.126
0.000
0.004
no
$2.370
1.195
1.265
no
$3.600
1.447
1.554
no
$6.330
2.175
2.363
300 MW with PJFF
Retrofit PJFF?
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Cost (mills/kWh)
no
$0.126
0.000
0.004
no
$0.126
0.000
0.004
yes
$45.989
0.215
1.352
yes
$45.147
0.234
1.376
yes
$46.467
0.278
1.430
a This table shows Total Cost and the portion of Total Cost that is variable (Variable Cost), both in mills/KWh.
For a more comprehensive breakdown of costs, please see the model plant tables in the appendices.
b The calculations performed to generate the results in this table assumed that all collected fly ash is currently
sold, which is the situation with the most conservative assumption. Therefore, these calculations include
costs to landfill fly ash with an impact to total cost of around 0.93 mills/kWh. In many cases these costs will
not apply because either ash may currently be landfilled or it may not be rendered completely unacceptable
for re-use.
59
-------
Table 10b. High Sulfur Coal, ESP plus FGD With SCR Co-benefit (Model Plants 1, 6)a
Parameter
Hg Reduction of Existing Equipment (%)
Hg Reduction by PAC (%)
Total Outlet Hg (mg/MWh)
Specified Hg Reduction (%)
50
90.
none
3.8
60
90.
none
3.8
70
90.
none
3.8
80
90.
none
3.8
90
90.
none
3.8
975 MW
Retrofit PJFF?
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Cost (mills/kWh)
no
$0.094
0.000
0.003
no
$0.094
0.000
0.003
no
$0.094
0.000
0.003
no
$0.094
0.000
0.003
no
$0.094
0.000
0.003
300 MW
Retrofit PJFF?
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Cost Cmills/kWh')
no
$0.126
0.000
0.004
no
$0.126
0.000
0.004
no
$0.126
0.000
0.004
no
$0.126
0.000
0.004
no
$0.126
0.000
0.004
This table shows Total Cost and the portion of Total Cost that is variable (Variable Cost),
For a more comprehensive breakdown of costs, please see the model plant tables in the
both in mills/KWh.
appendices.
High Sulfur Coal ESPh plus FGD (Model Plants 3, 8)
If an ESPh is used in combination with FGD, it is assumed that a low temperature PJFF will follow
the ESP and air preheater. Tables 1 la and 1 Ib show the results of cost estimates. As shown, the
co-benefit of SCR has substantial cost impacts because, without SCR, PAC must be added with
a downstream PJFF.
60
-------
Table 11a. High Sulfur Coal, ESPh plus FGD Without SCR Co-benefit (Model Plants 3, 8)a
Parameter
Hg Reduction of Existing Equipment (%)
Hg Reduction by PAC (%)
Total Outlet Hg (mg/MWh)
Specified Hg Reduction (%)
50
65.0
none
13.2
60
65.0
none
13.2
70
65.0
14.3
11.3
80
65.0
42.9
7.5
90
65.0
71.4
3.8
J75MW
Retrofit PJFF?
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Cost (mills/kWh)
no
$0.094
0.000
0.003
no
$0.094
0.000
0.003
yes
$36.236
0.218
1.126
yes
$36.345
0.239
1.149
yes
$36.566
0.284
1.201
300 MW
Retrofit PJFF?
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Cost (mills/kWh)
no
$0.126
0.000
0.004
no
$0.126
0.000
0.004
yes
$46.018
0.218
1.357
yes
$46.180
0.239
1.382
yes
$46.508
0.284
1.437
This table shows Total Cost and the portion of Total Cost that is variable (Variable Cost), both in mills/KWh.
For a more comprehensive breakdown of costs, please see the model plant tables in the appendices.
Table 11 b. High Sulfur Coal, ESPh plus FGD with SCR Co-benefit (Model Plants 3, 8)a
Parameter
Hg Reduction of Existing Equipment (%)
Hg Reduction by PAC (%)
Total Outlet Hg (mg/MWh)
Specified Hg Reduction (%)
50
90.
none
3.8
60
90.
none
3.8
70
90.
none
3.8
80
90.
none
3.8
90
90.
none
3.8
975 MW
Retrofit PJFF?
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Cost (mills/kWh)
no
$0.094
0.000
0.003
no
$0.094
0.000
0.003
no
$0.094
0.000
0.003
no
$0.094
0.000
0.003
no
$0.094
0.000
0.003
300 MW
Retrofit PJFF?
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Cost (mills/kWh)
no
$0.126
0.000
0.004
no
$0.126
0.000
0.004
no
$0.126
0.000
0.004
no
$0.126
0.000
0.004
no
$0.126
0.000
0.004
This table shows Total Cost and the portion of Total Cost that is variable (Variable Cost),
For a more comprehensive breakdown of costs, please see the model plant tables in the
both in mills/KWh.
appendices.
61
-------
ESP plus AdvDry FGD (Model Plants 4, 9)
Because of the high levels of mercury reduction that are assumed to occur with advanced dry FGD
on bituminous coals (over 95 percent mercury removal was measured at the Roanoke facility),42
additional mercury removal by PAC injection is not necessary. As will be shown later, this
technology is more cost effective on low sulfur coals than on high sulfur coals as shown here.
It should also be kept in mind that this technology, while more costly than PAC inj ection, provides
SO2 removal. An SO2 removal of 90 percent was assumed for each of these cases.
Due to the limited experience with this technology on utility plants in the United Ststes, capital
cost estimates may be uncertain. Sensitivity of the total cost with respect to capital cost is shown
in Table 12. The base estimated cost for the size range is shown in bold on Table 12. Sensitivity
analysis in a later section of this report and the appendices also show cost sensitivity with respect
to reagent price.
Table 12. Advanced Dry FGD on High Sulfur Coal (Model Plants 4, 9), Sensitivity to Capital
Cost3
Parameter
Hg Reduction of Existing Equipment (%)
Hg Reduction of Advanced Dry FGDC(%)
Total Hg Removal (%)
Total Outlet Hg (mg/MWh)
Capital Cost Category0
-20%
29.4
95.0
96.5
1.327
-10%
29.4
95.0
96.5
1.327
Projected
29.4
95.0
96.5
1.327
+10%
29.4
95.0
96.5
1.327
+20%
29.4
95.0
96.5
1.327
975 MW
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Cost (mills/kWh)
$115.46
5.323
7.940
$129.80
5.323
8.265
$144.23
5.323
8.592
$158.65
5.323
8.919
$173.07
5.323
9.246
300 MW
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Cost Cmills/kWh')
$127.47
5.323
8.212
$143.30
5.323
8.571
$159.23
5.323
8.932
$175.15
5.323
9.293
$191.07
5.323
9.654
This table shows Total Cost and the portion of Total Cost that is variable (Variable Cost), both in mills/KWh.
For a more comprehensive breakdown of costs, please see the model plant tables in the appendices.
Capital Cost (115.28-173.07 $/kW for a 975 MW plant and 127.47-191.07 $/kW for a 300 MW plant) is the
variable.
Advanced dry FGD is a technology developed primarily for SO2 removal; however, it can also provide
mercury control for bituminous coal.
62
-------
ESP plus ECO (Model Plants 5 and 10)
ECO is an emerging technology that appears to be capable of high pollution reduction. Because
it is an emerging technology, the sensitivity of total cost to capital cost was assessed and is shown
in Table 13. The economics rely, in part, on the revenue from fertilizer product sales and on the
value of power. Sensitivity analyses for the effect of fertilizer product value and for power cost
are shown in a later section of this report and in the appendices. For high sulfur coals, a high amount
of fertilizer product is possible, which makes the economics of the process appear more favorable
than for lower sulfur applications.
Table 13. ESP and ECO on High Sulfur Coal (Model Plants 5 and 10), Sensitivity to Capital
Cost3
Parameter
Hg Reduction of Existing Equipment (%)
ECO Hg Reduction FGDC(%)
Total Hg Reduction (%)
Total Outlet Hg (mg/MWh)
Capital Cost Category0
-20%
29.4
85.0
89.4
3.98
-10%
29.4
85.0
89.4
3.98
Projected
29.4
85.0
89.4
3.98
+10%
29.4
85.0
89.4
3.98
+20%
29.4
85.0
89.4
3.98
)75MW
Capital Cost ($/kW)
Variable Costd (mills/kWh)
Total Cost (mills/kWh)
$150.28
-0.820
3.276
$169.05
-0.820
3.764
$187.83
-0.820
4.252
$206.60
-0.820
4.740
$225.38
-0.820
5.228
300 MW
Capital Cost ($/kW)
Variable Costd (mills/kWh)
Total Cost (mills/kWh)
$190.23
-0.820
4.741
$214.00
-0.820
5.359
$237.76
-0.820
5.977
$261 .53
-0.820
6.595
$285.29
-0.820
7.212
3 This table shows Total Cost and the portion of Total Cost that is variable (Variable Cost), both in mills/KWh.
For a more comprehensive breakdown of costs, please see the model plant tables in the appendices.
b Capital Cost (150.28-225.38 $/kWfor a 975 MW plant and 190.23-285.29 $/kWfor a 300 MW plant) is the
variable.
c ECO is a technology developed primarily for NOX and SO2 removal; however, it can also provide mercury
control.
d Variable Cost includes a credit for fertilizer by-product sale. Negative numbers imply a net credit. Variable
Cost also includes the cost of power for the barrier discharge reactor—estimated at about 4.8% of plant
output for this case.
63
-------
100 MW SDA and ESPc (Model Plant 26)
For high sulfur fuels, a SDA with a downstream ESP is not expected to be very effective for mercury
removal. Therefore, most of the mercury removal must be performed by additional P AC inj ection.
In this case, a PJFF may be installed upstream of the SDA and must be sized for collection of
the full ash loading plus the PAC inj ection. Alternatively, a smaller polishing PJFF may be installed
downstream of the existing ESP. As will be shown later, a SDA with a downstream PJFF is expected
to achieve relatively high mercury removal and will only require moderate PAC inj ection to achieve
the additional reduction necessary for 90 percent removal. Therefore, it may be more economical
to install a polishing PJFF downstream oftheESP (COFIPAC conversion) and still inject the PAC
upstream of the SDA than to install a PJFF sized for collection of the full ash loading plus the
PAC injection. However, because data is not available on mercury removal from an SDA with
a downstream COFIPAC, it is uncertain if an SDA with a downstream COHPAC arrangement
would be as effective in removing mercury as an SDA with downstream PJFF.
Table 14. High Sulfur Coal, 100 MW SDA, and ESPc (Model Plant 26)a
Parameter
Hg Reduction of Existing Equipment (%)
Hg Reduction by PAC (%)
Total Outlet Hg (mg/MWh)
Specified Hg Reduction (%)
50
5.0
47.4
16.8
60
5.0
57.9
13.5
70
5.0
68.4
10.1
80
5.0
78.9
6.7
90
5.0
89.5
3.4
lOOMWnoPJFF
Retrofit PJFF?
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Cost (mills/kWh)
no
$6.014
0.659
0.838
no
$7.235
0.877
1.092
no
$8.996
1.226
1.493
no
$11.818
1.861
2.211
no
$17.266
3.309
3.821
100 MW and Full Size PJFF
Retrofit PJFF?
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Cost (mills/kWh)
yes
$110.342
0.242
2.907
yes
$110.891
0.255
2.934
yes
$111.094
0.275
2.960
yes
$111.413
0.308
3.002
yes
$112.065
0.383
3.096
100 MW with PJFF {COHPAC conversion)
Retrofit PJFF?
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Cost Cmills/kWh')
yes
$57.612
0.243
1.657
yes
$57.970
0.257
1.680
yes
$58.174
0.277
1.706
yes
$58.495
0.310
1.749
yes
$59.149
0.385
1.843
This table shows Total Cost and the portion of Total Cost that is variable (Variable Cost), both in mills/KWh.
For a more comprehensive breakdown of costs, please see the model plant tables in the appendices.
64
-------
100 MW SDA and FF (Model Plant 27)
As shown in Table 15, a SDA with a downstream fabric filter is expected to provide high mercury
removal, approaching 90 percent. A small amount of PAC might be added upstream of the SDA
to provide some more mercury reduction at a relatively low cost.
Table 15. High Sulfur Coal, 100 MW SDA, and FF (Model Plant 27)a
Parameter
Hg Reduction of Existing Equipment (%)
Hg Reduction by PAC (%)
Total Outlet Hg (mg/MWh)
Specified Hg Reduction (%)
50
89.3
none
4.0
60
89.3
none
4.0
70
89.3
none
4.0
80
89.3
none
4.0
90
89.3
6.3
3.8
100 MW, no PJFF
Retrofit PJFF?
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Cost Cmills/kWh')
no
$0.165
0.000
0.005
no
$0.165
0.000
0.005
no
$0.165
0.000
0.005
no
$0.165
0.000
0.005
no
$3.388
0.270
0.370
This table shows Total Cost and the portion of Total Cost that is variable (Variable Cost), both in mills/KWh.
For a more comprehensive breakdown of costs, please see the model plant tables in the appendices.
65
-------
100 MW ESPh and FGD (Model Plant 28)
In this case, the ESPh followed by a wet FGD (LSFO) will provide about 65 percent mercury
removal, and additional mercury removal will, therefore, be necessary to achieve 90 percent. As
shown in Table 16, it is more economical for 90 percent mercury removal to install a polishing
fabric filter after the air preheater and upstream of the flue gas desulfurization than to only inj ect
PAC without the fabric filter. But for 70 percent removal, it may be most economical to inject
PAC in the ductwork between the air preheater and the FGD.
Table 16. High Sulfur Coal, 100 MW ESPh, and FGD (Model Plant 28)a
Parameter
Hg Reduction of Existing Equipment (%)
Hg Reduction by PAC (%)
Total Outlet Hg (mg/MWh)
Specified Hg Reduction (%)
50
89.3
none
4.0
60
89.3
none
4.0
70
89.3
none
4.0
80
89.3
none
4.0
90
89.3
6.3
3.8
1 00 MW without PJFF
Retrofit PJFF?
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Cost (mills/kWh)
no
$0.165
0.000
0.005
no
$0.165
0.000
0.005
no
$3.751
1.236
1.347
no
$5.595
1.510
1.675
no
$9.657
2.287
2.573
1 00 MW with PJFF
Retrofit PJFF?
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Cost Cmills/kWh')
no
$0.165
0.000
0.005
no
$0.165
0.000
0.005
yes
$57.533
0.217
1.627
yes
$57.767
0.237
1.654
yes
$58.241
0.282
1.714
This table shows Total Cost and the portion of Total Cost that is variable (Variable Cost), both in mills/KWh.
For a more comprehensive breakdown of costs, please see the model plant tables in the appendices.
6.3.2 Low Sulfur Bituminous Coals
These coals are represented by Model Plants 11-19 and 29-37 in Table 9 and have a 0.6 wt%
sulphur content as shown in Table 7. The existing controls on these plants are either a cold- or
hot-side ESP or FF. Plants with ESPc or FF would need additional controls that consist of PAC,
PAC with a PJFF, ECO, or advanced dry FGD. Plants with an ESPh would need PAC with a PJFF,
ECO, orFF.
66
-------
With ESPc and No SO2 controls (Model Plants 11 and 29)
For these cases, PAC injection is expected to be necessary for mercury reduction in excess of 50
percent. As shown in Table 17, addition of a PJFF for the 975 MW plant case improves overall
economics for removal in excess of 70 percent. However, for a smaller 100 MW plant, the addition
of a polishing PJFF is more economical only for the 90 percent mercury removal case.
Table 17. Low Sulfur Coal, ESPc, and No SO2 Controls (Model Plants 11 and 29)a
Parameter
Hg Reduction of Existing Equipment (%)
Hg Reduction by PAC (%)
Total Outlet Hg (mg/MWh)
Specified Hg Reduction (%)
50
50.6
none
16.601
60
50.6
19.0
13.452
70
50.6
39.2
10.089
80
50.6
59.5
6.726
90
50.6
79.7
3.363
J75 MW and No PJFFb
Retrofit PJFF?b
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Costb (mills/kWh)
no
$0.094
0.000
0.003
no
$1.855
0.709
0.764
no
$2.467
0.901
0.974
no
$3.490
1.277
1.381
no
$5.711
2.282
2.451
975 MW with PJFF
Retrofit PJFF?
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Cost (mills/kWh)
no
$0.094
0.000
0.003
yes
$36.248
0.220
1.128
yes
$36.324
0.234
1.144
yes
$36.445
0.258
1.171
yes
$36.690
0.311
1.233
lOOMWandlMoPJFF"
Retrofit PJFF?b
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Costb (mills/kWh)
no
$0.165
0.000
0.005
no
$3.791
0.709
0.827
no
$5.271
0.901
1.057
no
$7.430
1.277
1.497
no
$12.057
2.282
2.639
1 00 MW with PJFF
Retrofit PJFF?
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Cost Cmills/kWh')
no
$0.165
0.000
0.005
yes
$57.563
0.220
1.631
yes
$57.729
0.234
1.650
yes
$57.989
0.258
1.682
yes
$58.518
0.311
1.751
This table shows Total Cost and the portion of Total Cost that is variable (Variable Cost), both in
mills/KWh. Fora more comprehensive breakdown of costs, please seethe model plant tables in
the appendices.
The calculations performed to generate the results in this table assumed that all collected fly ash is
currently sold, which is the situation with the most conservative assumption. Therefore, these
calculations include costs to landfill fly ash with an impact to total cost of around 0.93 mills/kWh. In
many cases these costs will not apply because either ash may currently be landfilled or it may not
be rendered completely unacceptable for re-use.
67
-------
With FF and No SO2 controls (Model Plants 12 and 30)
Due to the high mercury removal expected from existing equipment in these cases, PAC inj ection
is only expected to be necessary for mercury reduction in excess of 85 percent. As shown in Table
18, installation of a PJFF is expected to be economically beneficial for neither the 975 nor the
100 MW plant cases.
Table 18. Low Sulfur Coal, FF, and No SO2 Controls (Model Plants 12 and 30)a
Parameter
Hg Reduction of Existing Equipment (%)
Hg Reduction by PAC (%)
Total Outlet Hg (mg/MWh)
Specified Hg Reduction (%)
50
85.0
none
5.0
60
85.0
none
5.0
70
85.0
none
5.0
80
85.0
none
5.0
90
85.0
33.3
3.4
975 MW and No PJFFb
Retrofit PJFF?b
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Costb (mills/kWh)
975 MW with PJFF
Retrofit PJFF?
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Cost (mills/kWh)
no
$0.094
0.000
0.003
no
$0.094
0.000
0.003
no
$0.094
0.000
0.003
no
$0.094
0.000
0.003
no
$0.094
0.000
0.003
no
$0.094
0.000
0.003
no
$0.094
0.000
0.003
no
$0.094
0.000
0.003
no
$0.821
0.458
0.482
yes
$36.299
0.229
1.139
10QMWandNoPJFFb
Retrofit PJFF?b
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Costb (mills/kWh)
no
$0.165
0.000
0.005
no
$0.165
0.000
0.005
no
$0.165
0.000
0.005
no
$0.165
0.000
0.005
no
$1.752
0.458
0.510
1 00 MW with PJFF
Retrofit PJFF?
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Cost Cmills/kWh')
no
$0.165
0.000
0.005
no
$0.165
0.000
0.005
no
$0.165
0.000
0.005
no
$0.165
0.000
0.005
yes
$57.674
0.229
1.644
This table shows Total Cost and the portion of Total Cost that is variable (Variable Cost), both in mills/KWh.
For a more comprehensive breakdown of costs, please see the model plant tables in the appendices.
The calculations performed to generate the results in this table assumed that all collected fly ash is currently
sold, which is the situation with the most conservative assumption. Therefore, these calculations include
costs to landfill fly ash with an impact to total cost of around 0.93 mills/kWh. In many cases these costs will
not apply because either ash may currently be landfilled or it may not be rendered completely unacceptable
for re-use.
68
-------
With ESPh and No SO2 Controls (Model Plants 13 and 31)
Due to the low mercury removal possible from existing equipment in these cases, PAC injection
is expected to be necessary for all of the conditions, and a polishing PJFF must be added because
PAC inj ection would normally be added downstream of the ESPh and air preheater. Table 19 shows
the economics of this type of installation for 975 MW and 100 MW plants.
Table 19. Low Sulfur Coal, ESPh, and No SO2 Controls (Model Plants 13 and 31)a
Parameter
Hg Reduction of Existing Equipment (%)
Hg Reduction by PAC (%)
Total Outlet Hg (mg/MWh)
Specified Hg Reduction (%)
50
25.5
32.9
16.8
60
25.5
46.3
13.5
70
25.5
59.7
10.1
80
25.5
73.2
6.7
90
25.5
86.6
3.4
975 MW
Retrofit PJFF?
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Cost (mills/kWh)
yes
$36.162
0.229
1.135
yes
$36.360
0.241
1.152
yes
$36.447
0.258
1.172
yes
$36.584
0.287
1.205
yes
$36.865
0.353
1.280
IOOMW
Retrofit PJFF?
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Cost (mills/kWh)
yes
$57.458
0.229
1.638
yes
$57.805
0.241
1.659
yes
$57.994
0.258
1.682
yes
$58.290
0.287
1.720
yes
$58.893
0.353
1.804
This table shows Total Cost and the portion of Total Cost that is variable (Variable Cost), both in mills/KWh.
For a more comprehensive breakdown of costs, please see the model plant tables in the appendices.
69
-------
ECO Installed after Particulate Removal (Model Plants 14-16, 32-34)
Since the most common location for installation of ECO is likely to be after an ESPc and because
the economics are similar regardless of the existing PM removal equipment, only the results of
analysis for ECO installed after an ESPc are shown in Table 20. ECO is a capital-intensive
technology, and there remains some uncertainty regarding its actual capital cost because it is an
emerging technology. Therefore, the costwas evaluated for a range of capital costs plus or minus
20 percent of the projected cost for the size unit. The sensitivity analysis of Section 6.4 shows
estimated ECO economics while varying the fertilizer product value and the power cost for a 500
MW plant. Additional information is available in the Model Runs in the appendices. For a given
coal, unit size, and cost of consumables, the economics of ECO are estimated to be roughly the
same regardless of the type of upstream paniculate control device.
Table 20. ECO Installed After Particulate Removal (Model Plants 14-16,32-34), Sensitivity to
Capital Cost3
Parameter
Hg Reduction of Existing Equipment (%)
ECO Hg Reduction FGDC(%)
Total Hg Reduction (%)
Total Outlet Hg (mg/MWh)
Capital Cost Category0
-20%
50.6
85.0
92.6
2.49
-10%
50.6
85.0
92.6
2.49
Projected
50.6
85.0
92.6
2.49
+10%
50.6
85.0
92.6
2.49
+20%
50.6
85.0
92.6
2.49
975 MW
Capital Cost ($/kW)
Variable Costd (mills/kWh)
Total Cost (mills/kWh)
$150.28
1.243
5.340
$169.05
1.243
5.828
$187.83
1.243
6.316
$206.60
1.243
6.804
$225.38
1.243
7.292
100MW
Capital Cost ($/kW)
Variable Costd (mills/kWh)
Total Cost Cmills/kWh')
$236.99
1.243
9.521
$266.59
1.243
10.021
$296.20
1.243
10.790
$325.80
1.243
1 1 .560
$355.41
1.243
12.329
This table shows Total Cost and the portion of Total Cost that is variable (Variable Cost), both in mills/KWh.
For a more comprehensive breakdown of costs, please see the model plant tables in the appendices.
Capital Cost (150.28-225.38 $/kWfor a 975 MW plant and 236.99-355.41 $/kWfor a 100 MW plant) is the
variable.
ECO is a technology developed primarily for NOX and SO2 removal; however, it can also provide mercury
control. Data shown is for retrofit on a boiler with existing ESPc. For boilers with existing ESPh or FF, costs
would be very similar, but outlet mercury would differ somewhat from what is shown here. See appendices
for details
Variable Cost includes a credit for fertilizer by-product sale. Negative numbers imply a net credit. Variable
Cost also includes the cost of power for the barrier discharge reactor—estimated at about 4.8% of plant
output for this case.
70
-------
Advanced Dry FGD (Model Plants 17-19, 35-37)
Advanced Dry FGD is another technology that is somewhat capital intensive and controls SO2
and mercury. Like ECO, the economics are a strong function of capital cost and sulfur level and
less affected by the type of upstream parti culate control technology. Shown in Table 21 are the
results of analysis for advanced dry FGD downstream of an existing ESPc.
Table 21. Advanced Dry FGD (Model Plants 17-19, 35-37), Sensitivity to Capital Cost3
Parameter
Hg Reduction of Existing Equipment (%)
Hg Reduction of Advanced Dry FGDC(%)
Total Hg Removal (%)
Total Outlet Hg (mg/MWh)
Capital Cost Category0
-20%
50.6
95.0
97.5
0.830
-10%
50.6
95.0
97.5
0.830
Projected
50.6
95.0
97.5
0.830
+10%
50.6
95.0
97.5
0.830
+20%
50.6
95.0
97.5
0.830
975 MW
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Cost (mills/kWh)
$115.46
1.071
3.688
$129.80
1.071
4.013
$144.23
1.071
4.340
$158.65
1.071
4.667
$173.07
1.071
4.994
IOOMW
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Cost Cmills/kWh')
$162.17
1.071
4.747
$182.31
1.071
5.203
$202.57
1.071
5.662
$222.82
1.071
6.122
$243.08
1.071
6.581
a This table shows Total Cost and the portion of Total Cost that is variable (Variable Cost), both in mills/KWh.
For a more comprehensive breakdown of costs, please see the model plant tables in the appendices.
b Capital Cost (115.28-173.07 $/kWfor a 975 MW plant and 162.17-243.08 $/kWfor a 100 MW plant) is the
variable.
c Advanced dry FGD is a technology developed primarily for SO2 removal; however, it can also provide
mercury control for bituminous coal. Data shown is for retrofit on a boiler with existing ESPc. For boilers with
existing ESPh or FF, costs would be very similar, but outlet mercury may differ somewhat from what is
shown here. See appendices for details.
6.3.3 Low Sulfur Subbituminous Coals Including Powder River Basin Coals
Mercury removal with existing equipment is typically lower for subbituminous coals than for
bituminous coals. As a result, mercury reduction is more dependent on PAC injection for high
levels of mercury removal. In the case of boilers currently equipped with ESPs, it may not be
possible to achieve 80 or 90 percent reduction without addition of a downstream PJFF as indicated
in Model Plants 20-22 and 38-40. ECO is examined on Model Plants 23-25 and 41-43. The
effectiveness and cost of K-Fuel as a mercury control technology is examined in Model Plants
44-49. The K-fuel process is primarily designed to improve fuel-heating value, but it also provides
some SO2 reduction, may provide NOX reduction, and provides mercury reduction.
71
-------
Boilers with Particulate Control and No SO2 Control (Model Plants 20-22, 38^40)
As shown in Table 22 for the cases without a downstream PJFF, estimates for 80 or 90 percent
mercury reduction show high costs due to high predicted inj ection rates. It is recognized that, despite
the high injection rates, the specified Hg reduction may not be achievable without addition of
a PJFF after the ESP. However, it should be noted that the algorithms used for P AC inj ection here
(Equation 9 and the associated constants for this case) were developed from test results at the
Pleasant Prairie Power Plant, which had a coal chlorine content of only 15 ppm, which is lower
than typically expected for this type of fuel.19 So, it is possible that other PRB fueled boilers may
be easier to control with PAC than those shown here.
For mercury reduction from boilers firing subbituminous coals and equipped with a downstream
FF, PAC injection is necessary for greater than about 60 percent mercury reduction. Addition of
a downstream PJFF provides the benefit of much lower waste disposal costs because fly ash is
not contaminated. Therefore, although the cost of sorbent is similar regardless of whether or not
a PJFF is added, the additional cost of waste disposal roughly compensates for the cost of the PJFF
for the 975 MW case as shown in Table 23.
In the case where an ESPh is currently installed, it is necessary to install a downstream PJFF for
mercury removal by PAC injection. As shown in Table 24, this can generally be performed at
a cost of below 2 mills/kWh.
PRB Coals with ECO (Model Plants 23-25, 41^43)
Since the most common installation for an ECO is likely to be after an ESPc and the economics
are similar regardless of existing particle removal equipment, only the ECO after an ESPc is shown
in Table 25. ECO is a capital-intensive technology, and there remains some uncertainty regarding
its capital costbecause it is an emerging technology. Therefore, the cost was evaluated for a range
of capital costs within 20 percent of the projected cost for the size unit. The model runs in the
appendices show estimated ECO economics for various fertilizer product value and power value
for 975 MW and 100 MW plants equipped withESPc, FF, or ESPh. The lower NOX level associated
with PRB coals helps to reduce the power demand of the ECO barrier discharge reactor compared
to the power demand of the ECO barrier discharge reactor when firing bituminous coals. However,
because of the low fertilizer product revenue (due to low NOX and SO2 levels for PRB coals),
economics of ECO on this application are estimated to be less favorable than for the bituminous
coals that have higher NOX and SO2 levels.
72
-------
Table 22. Low Sulfur Subbituminous Coals, ESPc, and No SO2 Control (Model Plants 20, 38)a
Parameter
Hg Reduction of Existing Equipment (%)
Desired Hg Reduction by PAC (%)
<\ctual Hg Reduction by PAC (%) without
3JFFb
Total Actual Hg Reduction without PJFFb
Specified Hg Reduction (%)
50
29.7
28.9
28.9
50.0
60
29.7
43.1
43.1
60.0
70
29.7
57.3
57.3
70.0
80
29.7
71.5
69.3
78.5
90
29.7
85.8
69.3
78.5
975 MW with ESPc and No PJFFC
Retrofit PJFF?C
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Costc (mills/kWh)
Total Outlet Hg (mg/MWh)
no
$0.401
1.027
1.039
20.1
no
$1.238
1.181
1.218
16.1
no
$3.232
1.811
1.907
12.1
no
$27.744
20.102
20.924
8.7
no
$27.744
20.102
20.924
8.7
975 MW with ESPc and PJFF
Retrofit PJFF?
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Costc (mills/kWh)
Total Outlet Hg (mg/MWh)
yes
$35.998
0.209
1.111
20.1
yes
$36.258
0.231
1.139
16.1
yes
$36.422
0.262
1.176
12.1
yes
$36.666
0.315
1.236
8.0
yes
$37.139
0.435
1.369
4.0
100 MW with ESPc and No PJFFC
Retrofit PJFF?C
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Costc (mills/kWh)
Total Outlet Hg (mg/MWh)
no
$0.840
1.027
1.052
20.1
no
$2.651
1.181
1.259
16.1
no
$6.887
1.811
2.015
12.1
no
$55.806
20.102
21.756
8.7
no
$55.806
20.102
21.756
8.7
100 MW with ESPc and PJFF
Retrofit PJFF?
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Costc (mills/kWh)
Total Outlet Ha fma/MWh')
yes
$57.102
0.209
1.608
20.1
yes
$57.585
0.231
1.643
16.1
yes
$57.939
0.262
1.685
12.1
yes
$58.466
0.315
1.753
8.0
yes
$59.479
0.435
1.903
4.0
3 This table shows Total Cost and the portion of Total Cost that is variable (Variable Cost), both in mills/KWh.
For a more comprehensive breakdown of costs, please see the model plant tables in the appendices.
b With PAC injection on subbituminous coals without a downstream fabric filter, Hg reduction at very high
levels is not expected to be possible. Additional PAC injection will not improve Hg reduction.
c The calculations performed to generate the results in this table assumed that all collected fly ash is currently
sold, which is the most conservative situation. Therefore, these calculations include costs to landfill fly ash
with an impact to total cost of around 1.01 mills/kWh. In many cases these costs will not apply because ash
may currently be landfilled or because ash may not be rendered completely unacceptable for re-use.
73
-------
Table 23. Low Sulfur Subbituminous Coals, FF, and No SO2 Control (Model Plants 21, 39)a
Parameter
Hg Reduction of Existing Equipment (%)
Hg Reduction by PAC (%)
Total Outlet Hg (mg/MWh)
Specified Hg Reduction (%)
50
60.7
none
15.8
60
60.7
none
15.8
70
60.7
23.6
12.1
80
60.7
49.1
8.0
90
60.7
74.5
4.0
975 MW with FF and No PJFFb
Retrofit PJFF?b
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Costb (mills/kWh)
no
$0.094
0.000
0.003
no
$0.094
0.000
0.003
no
$0.616
1.057
1.075
no
$0.842
1.097
1.122
no
$1.259
1.186
1.223
975 MW with FF and PJFF
Retrofit PJFF?
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Costc (mills/kWh)
no
$0.094
0.000
0.003
no
$0.094
0.000
0.003
yes
$36.094
0.203
1.106
yes
$36.320
0.243
1.153
yes
$37.737
0.332
1.254
100 MW with FF and No PJFF"
Retrofit PJFF?b
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Costc (mills/kWh)
no
$0.165
0.000
0.005
no
$0.165
0.000
0.005
no
$1.308
1.057
1.096
no
$1.799
1.097
1.150
no
$2.696
1.186
1.266
1 00 MW with FF and PJFF
Retrofit PJFF?
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Costc (mills/kWm
no
$0.165
0.000
0.005
no
$0.165
0.000
0.005
yes
$57.230
0.203
1.604
yes
$57.721
0.243
1.659
yes
$58.618
0.332
1.774
This table shows Total Cost and the portion of Total Cost that is variable (Variable Cost), both in mills/KWh.
For a more comprehensive breakdown of costs, please see the model plant tables in the appendices.
The calculations performed to generate the results in this table assumed that all collected fly ash is currently
sold, which is the most conservative situation. Therefore, these calculations include costs to landfill fly ash
with an impact to total cost of around 1.01 mills/kWh. In many cases these costs will not apply because ash
may currently be landfilled or because ash may not be rendered completely unacceptable for re-use.
74
-------
Table 24. Low Sulfur Subbituminous Coals, ESPh, and No SO2 Controls (Model Plants 22, 40)a
Parameter
Hg Reduction of Existing Equipment (%)
Hg Reduction by PAC (%)
Total Outlet Hg (mg/MWh)
Specified Hg Reduction (%)
50
12.6
42.8
20.1
60
12.6
54.2
16.1
70
12.6
65.7
12.1
80
12.6
77.1
8.0
90
12.6
88.6
4.0
J75 MW with ESPh and PJFF
Retrofit PJFF?
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Cost (mills/kWh)
yes
$36.119
0.230
1.135
yes
$36.381
0.254
1.166
yes
$36.550
0.289
1.206
yes
$36.806
0.348
1.273
yes
$37.305
0.482
1.421
100 MW with ESPh and PJFF
Retrofit PJFF?
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Costc Cmills/kWh')
tes
$57.365
0.230
1.637
yes
$57.852
0.254
1.674
yes
$58.217
0.289
1.720
yes
$58.766
0.348
1.795
yes
$59.834
0.482
1.960
This table shows Total Cost and the portion of Total Cost that is variable (Variable Cost), both in mills/KWh.
For a more comprehensive breakdown of costs, please see the model plant tables in the appendices.
75
-------
Table 25. Low Sulfur Subbituminous Coals with ECO (Model Plants 23-25, 41-43) Sensitivity
to Capital Cost3
Parameter
Hg Reduction of Existing Equipment (%)
ECO Hg Reduction FGDC(%)
Total Hg Reduction (%)
Total Outlet Hg (mg/MWh)
Capital Cost Category0
-20%
29.7
85.0
89.5
4.24
-10%
29.7
85.0
89.5
4.24
Projected
29.7
85.0
89.5
4.24
+10%
29.7
85.0
89.5
4.24
+20%
29.7
85.0
89.5
4.24
975 WIW with ESPc
Capital Cost ($/kW)
Variable Costd (mills/kWh)
Total Cost (mills/kWh)
$150.28
0.442
4.539
$169.05
0.442
5.027
$187.83
0.442
5.515
$206.60
0.442
6.003
$225.38
0.442
6.491
1 00 MW with ESPc
Capital Cost ($/kW)
Variable Costd (mills/kWh)
Total Cost Cmills/kWh')
$236.99
0.442
8.451
$266.59
0.442
9.220
$296.20
0.442
9.990
$325.80
0.442
10.759
$355.41
0.442
1 1 .529
This table shows Total Cost and the portion of Total Cost that is variable (Variable Cost), both in mills/KWh.
For a more comprehensive breakdown of costs, please see the model plant tables in the appendices.
Capital Cost (150.28-225.38 $/kWfor a 975 MW plant and 236.99-355.41 $/kWfor a 100 MW plant) is the
variable.
ECO is a technology developed primarily for NOX and SO2 removal; however, it can also provide mercury
control. Data shown is for retrofit on a boiler with existing ESPc. For boilers with existing ESPh or FF, costs
would be very similar, but outlet mercury would differ somewhat from what is shown here. See appendices
for details
Variable Cost includes a credit for fertilizer by-product sale. Negative numbers imply a net credit. Variable
Cost also includes the cost of power for the barrier discharge reactor—estimated at about 4.8% of plant
output for this case.
Beneficiated Western Coals—K-Fuel
Table 26 compares the estimated mercury emissions on a mg/MWh basis for the base PRB coal
versus the K-Fuel beneficiated coal. The K-Fuel has about 60 percent less mercury than PRB on
a unit of heating value basis. Mercury removal rates for the particulate removal equipment are
similar, with the K-Fuel estimated to have a slightly higher removal with an ESPc due to the slightly
lower sulfur level in the fuel. It is also important to note that K-Fuel provides a 44 percent reduction
in SO2 emissions from the base PRB fuel.
Figure 16 shows the economics of K-Fuel assuming that all of the incremental costs of using the
fuel are born in an increased fuel cost over PRB fuel.
76
-------
Table 26. Comparison of Estimated Mercury Emissions from PRB and K-Fuela Boilers Equipped
with Particulate Control and No Additional Mercury or SO2 Controls
Characteristic
°articulate Control System
Hg Reduction with Existing Equipment
(%)
Hg from Coal (mg/MWh)
Total Outlet Ha (ma/MWrO
PRB
ESPc
29.7
40.2
28.2
K-Fuel
ESPc
36.8
16.3
10.3
PRB
FF
60.7
40.2
15.8
K-Fuel
FF
60.7
16.3
6.4
PRB
ESPh
12.6
40.2
35.1
K-Fuel
ESPh
12.6
16.3
14.2
The K-fuel process is primarily designed to improve fuel-heating value, but it may provide NOX reduction,
provides some SO2 reduction, and provides Hg reduction.
"m
2.5 r
2.0 |
15
0.5
0.0
0.00 0.05 0.10 0.15 0.20
K-Fuel Premium over PRB, IIMMBtu
0.25
Figure 16. Estimated Effect of K-Fuel Cost on Generation Cost.
Model runs were performed for P AC inj ection on boilers equipped with K-Fuel. These are model
runs 44 through 49, and Tables 27-29 summarize the results of these model runs. The economics
of PAC injection for K-Fuel are similar to those for PAC injection for PRB, except that it may
be appropriate to add to these costs the incremental cost of the K-Fuel over PRB. The major
difference is that lower emissions rates are possible with K-Fuel than with PRB due to the lower
initial mercury level.
77
-------
Table 27. K-Fuela, ESPc and No SO2 Control (Model Plants 44, 47)b
Parameter
Hg Reduction of Existing Equipment (%)
Desired Hg Reduction by PAC (%)
<\ctual Hg Reduction by PAC (%) without
3JFFC
Total Actual Hg Reduction without PJFFC
Specified Hg Reduction (%)
50
36.8
20.9
20.9
50.0
60
36.8
36.7
36.7
60.0
70
36.8
52.5
52.5
70.0
80
36.8
68.3
68.3
80.0
90
36.8
84.2
69.3
80.6
975 MW with ESPc and No PJFF"
Retrofit PJFF?d
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Costd (mills/kWh)
Total Outlet Hg (mg/MWh)
no
$0.391
0.745
0.756
8.1
no
$0.673
0.785
0.805
6.5
no
$2.240
1.180
1.247
4.9
no
$14.462
8.237
8.666
3.3
no
$26.735
18.846
19.638
3.2
975 MW with ESPc and PJFF
Retrofit PJFF?
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Costd (mills/kWh)
Total Outlet Hg (mg/MWh)
yes
$35.922
0.190
1.089
8.1
yes
$36.177
0.209
1.115
6.5
yes
$36.333
0.238
1.148
4.9
yes
$36.565
0.285
1.203
3.3
yes
$37.009
0.393
1.323
1.6
100 MW with ESPc and No PJFF"
Retrofit PJFF?d
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Costd (mills/kWh)
Total Outlet Hg (mg/MWh)
no
$0.817
0.745
0.769
8.1
no
$1.431
0.785
0.828
6.5
no
$4.792
1.180
1.322
4.9
no
$29.800
8.237
9.120
3.3
no
$53.860
18.846
20.441
3.2
100 MW with ESPc and PJFF
Retrofit PJFF?
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Costd (mills/kWh)
Total Outlet Ha fma/MWh')
yes
$56.936
0.190
1.584
8.1
yes
$57.409
0.209
1.616
6.5
yes
$57.748
0.238
1.654
4.9
yes
$58.248
0.285
1.717
3.3
yes
$59.202
0.393
1.853
1.6
The K-fuel process is primarily designed to improve fuel-heating value, but it may provide NOX reduction,
provides some SO2 reduction, and provides Hg reduction.
This table shows Total Cost and the portion of Total Cost that is variable (Variable Cost), both in mills/KWh.
For a more comprehensive breakdown of costs, please see the model plant tables in the appendices.
With PAC injection on subbituminous coals without a downstream fabric filter, Hg reduction at very high
levels is not expected to be possible. Additional PAC injection will not improve Hg reduction.
The calculations performed to generate the results in this table assumed that all collected fly ash is currently
sold, which is the most conservative situation. Therefore, these calculations include costs to landfill fly ash
with an impact to total cost of around 1.01 mills/kWh. In many cases these costs will not apply because ash
may currently be landfilled or because ash may not be rendered completely unacceptable for re-use.
78
-------
Table 28. K-Fuela, FF, and No SO2 Control (Model Plants 45, 48)b
Parameter
Hg Reduction of Existing Equipment (%)
Hg Reduction by PAC (%)
Total Outlet Hg (mg/MWh)
Specified Hg Reduction (%)
50
60.7
none
6.4
60
60.7
none
6.4
70
60.7
23.6
4.9
80
60.7
49.1
3.3
90
60.7
74.5
1.6
375 MW with FF and No PJFF°
Retrofit PJFF?C
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Costc (mills/kWh)
no
$0.094
0.000
0.003
no
$0.094
0.000
0.003
no
$0.598
0.773
0.791
no
$0.817
0.811
0.835
no
$1.219
0.895
0.931
375 MW with FF and PJFF
Retrofit PJFF?
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Costc (mills/kWh)
no
$0.094
0.000
0.003
no
$0.094
0.000
0.003
yes
$36.076
0.193
1.096
yes
$36.295
0.230
1.140
yes
$36.697
0.315
1.236
100 MW with FF and No PJFFC
Retrofit PJFF?C
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Costd (mills/kWh)
no
$0.165
0.000
0.005
no
$0.165
0.000
0.005
no
$1 .269
0.773
0.811
no
$1.744
0.811
0.862
no
$2.611
0.895
0.973
1 00 MW with FF and PJFF
Retrofit PJFF?
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Costc (mills/kWm
no
$0.165
0.000
0.005
no
$0.165
0.000
0.005
yes
$57.191
0.193
1.593
yes
$57.666
0.230
1.645
yes
$58.533
0.315
1.755
a The K-fuel process is primarily designed to improve fuel-heating value, but it may provide NOX reduction,
provides some SO2 reduction, and provides Hg reduction.
b This table shows Total Cost and the portion of Total Cost that is variable (Variable Cost), both in mills/KWh.
For a more comprehensive breakdown of costs, please see the model plant tables in the appendices.
c The calculations performed to generate the results in this table assumed that all collected fly ash is currently
sold, which is the most conservative situation. Therefore, these calculations include costs to landfill fly ash
with an impact to total cost of around 1.01 mills/kWh. In many cases these costs will not apply because ash
may currently be landfilled or because ash may not be rendered completely unacceptable for re-use.
79
-------
Table 29. K-Fuela, ESPh, and No SO2 Controls (Model Plants 46, 49)b
Parameter
Hg Reduction of Existing Equipment (%)
Hg Reduction by PAC (%)
Total Outlet Hg (mg/MWh)
Specified Hg Reduction (%)
50
12.6
42.8
8.1
60
12.6
54.2
6.5
70
12.6
65.7
4.9
80
12.6
77.1
3.3
90
12.6
88.6
1.6
975 MW with ESPh and PJFF
Retrofit PJFF?
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Cost (mills/kWh)
yes
$36.096
0.219
1.123
yes
$36.353
0.242
1.153
yes
$36.517
0.275
1.191
yes
$36.764
0.331
1.254
yes
$37.246
0.457
1.395
100 MW with FF and PJFF
Retrofit PJFF?
Capital Cost ($/kW)
Variable Cost (mills/kWh)
Total Cost Cmills/kWh')
yes
$57.315
0.219
1.624
yes
$57.792
0.242
1.660
yes
$58.145
0.275
1.7.03
yes
$58.676
0.331
1.775
yes
$59.707
0.457
1.932
a The K-fuel process is primarily designed to improve fuel-heating value, but it may provide NOX reduction,
provides some SO2 reduction, and provides Hg reduction.
b This table shows Total Cost and the portion of Total Cost that is variable (Variable Cost), both in mills/KWh.
For a more comprehensive breakdown of costs, please see the model plant tables in the appendices.
6.4 Cost Impacts of Selected Variables
Sensitivity analysis was performed for certain variables of interest on specific model plants. These
included the effects of new sorbents, capital cost on PAC injection, fertilizer cost/power cost on
ECO, and lime cost on Advanced Dry FGD.
Effect of New Sorbents on Cost ofHg Control
In the future, cost of controlling mercury could be reduced by new sorbents, which could potentially
eliminate disposal costs for spent sorbent/fly ash mixtures. To assess the effects of elimination
of ash and sorbent disposal cost, Figure 17 was developed. Figure 17 shows the results of cost
estimates comparing mercury control cost for 500 MW plants with and without disposal costs.
For each configuration, the sorbent and fly ash are collected together in an existing FF or ESPc.
As shown, the elimination of disposal costs can be quite significant, especially for the PRB fuels.
Low sulfur bituminous coal, with a low ash level and high Btu content, has a lower component
of ash disposal cost of the total cost, but it is nevertheless quite significant at about 0.3 8 mills/kWh
(this is just for the fly ash; the sorbent contributes to additional disposal cost). The PRB fuel has
80
-------
0-0
—o— 500 MW. Bit w ith ESFt
— 6_
. .4,
,— A,
— Q
0 500 MW, PRB w ith ESFb and
no disposal cost
*— 500 MW. Bit w ith FF
, 500 MW, Bit w ith FF- no
disposal costs
50% 60% 70% 80% 90%
Total Hg Reduction
100%
Figure 17. Cost of PAC Injection for 500 MWCoal Fired Boilers with existing ESPc or FF.
a higher ash loading and lower Btu content, causing ash levels to be much higher and disposal
costs for contaminated ash much higher (about 1.0 mill/kWh for the ash alone). The exact savings
in disposal cost for any particular application will depend upon the actual ash content and Btu
value of the particular fuel being used. So, facilities with fuel having different ash content or Btu
value will see a different impact on ash disposal costs.
It is also notable that the impact of disposal costs, as a percent of total cost, is much greater for
facilities with FFs than for facilities that capture the PAC and fly ash in ESPs. This is because
the much lower PAC injection rate for FF-equipped facilities makes ash disposal cost a very large
fraction of the total cost. Figure 17 shows that if disposal costs can be avoided by use of cost-
effective sorbents other than PAC, or by other means, then the cost of controlling mercury can
be reduced by 75 to 80 percent in some cases (PAC injection upstream of existing FF).
There is extensive research ongoing in the field of improved sorbents for mercury control that
may be more efficient in capturing mercury and potentially at a lower cost than PAC.59'74'75 These
improvements in sorbents may contribute to overall reductions in sorbent consumption, contributing
to reduced costs for disposal, sorbents, and capital equipment. Improved sorbents could enable
81
-------
users to avoid installation of fabric filters and will also reduce the sorbent storage and material
handing equipment on site.
For example, Reference 19 showed that PAC usage for facilities that capture PAC sorbent in
downstream ESPs was much more affected by selection of the type of PAC sorbent than facilities
that captured the PAC sorbent in a FF. Reference 19 cited the results of several full-scale tests
using differentPAC materials available from manufacturers. Reference 19 developed three different
performance curves for each of the two cases with downstream ESPs (one case with bituminous
coal and the other case with a PRB subbituminous coal). The three different curves were
characterized as high, medium, and low performance. Figure 18 shows estimates of total cost for
PAC inj ection (inclusive of disposal costs) for the medium performance sorbents used in this study
versus higher performance sorbents tested in field trials. The estimates assume that the higher
performance PAC sorbents are available at the same price (the higher performance sorbents are
PAC with some different physical characteristics—no additional chemicals). As shown in Figure
SOOMWLSBit ESP red
o— 500 MW LS Bit. ESP high
, _&. . 500 MW PRB. ESP meet
- 500 MW PRB, ESP high
50%
60%
70% 80%
Total Hg
90%
100%
Figure 18. Cost of PAC Injection for 500 MWCoal Fired Boilers with Existing
ESPc—Effect of Medium Versus High Performance PAC.
82
-------
18, cost reduction of 20 percent is estimated for the bituminous coal fired facility when using the
higher performance PAC. Because the sorbents used in the calculations to generate Figure 18 are
conventional PAC sorbents, even greater cost improvements are likely if other improved sorbents
are considered.
Effect of Capital Cost for PAC Injection
For facilities with ESPcs that may retrofit PAC inj ection and a P JFF downstream of the ESP, capital
cost is a major contributor to control cost. As shown in Figure 19, for boilers firing either PRB
or low sulfur bituminous coal, the total cost of control is expected to vary from about 1.2 to about
1.8 mills/kWh over the range of expected total capital costs.
3.0
2,5
M
H
!^ 4 C
E1'5
+f
«4 ft
01-0
CJ
0,5
0,0
30
• LS Bit
PRB
35
40 45 50
55
80
Figure 19. Effect of Capital Cost on 90 Percent Mercury Control with PAC on Boiler with
Existing ESPc and Retrofit of Downstream PJFF.
Effect of Fertilizer Value and Power Cost on ECO
Figure 20 shows the estimated effect of the value of fertilizer product on the economics of ECO
for a 500 MW plant firing either low sulfur or high sulfur bituminous coal. For a high sulfur coal
83
-------
fired boiler, the impact of fertilizer value in reducing cost is greater because more fertilizer is
produced from a high sulfur coal boiler. Also, for the high sulfur coal boiler, the net cost is lower
due to the higher fertilizer revenues. Figure 21 shows the effects of the value of power needed
by the ECO on the cost of controlling with ECO on a 500 MW plant firing either low sulfur or
high sulfur bituminous coal. For a PRB coal fired boiler or a boiler with lower initial NOX, the
effect of power on total cost is less because power consumption is closely related to NOX levels.
8,0
7.0
8.0
j 5.0
I
U.o
O
o
2.0
1.0
0.0
80
3%
80 100 120 140
Fertilizer Value,
180
Figure 20. Effect of Fertilizer Value on Cost of Emissions Control with ECO on a
500 MW Bituminous Coal Boiler.
Effect of Lime Cost on Advanced Dry FGD
Figure 22 shows the effects of the cost of lime reagent, an important cost factor, on the total cost
of control for advanced dry FGD. As shown, there is roughly ! 0.19 mills/kWh effect, or a roughly
!4 percent effect on the total cost of control over the range evaluated.
84
-------
9-0
8.0
7.0
6.0
5-°
4-0
3-0
o
2.0
1.0
0.0
• 3%
10 15 20 25 30 35
Power
40
45
Figure 21. Effect of Power Value on Cost of Emissions Control with ECO on a
500 MW Bituminous Coal Boiler.
4.70
4M5
4.60
^ 4.55
3 4.50
£ 4.45
**»F
«
4.40
4.35
4.30
4.25
40
50 80 70
Price,
80
90
Figure 22. Effect of Reagent Cost on Cost of Emissions Control with Advanced
Dry FGD on a 500 MW Boiler Firing Low Sulfur Bituminous Coal.
85
-------
6.5 Summary of Mercury and Multipollutant Control Costs
Table 30 shows expected costs for control of mercury from coal-fired boilers. Listed are control
costs for at least 80 percent and up to 90 percent reduction. The following assumptions were used
in making the current and potential cost estimates.
• For situations where one approach seemed to be more attractive than another (such as
P AC inj ection alone versus PAC inj ection plus a PJFF), it was assumed that the facility
owner would normally select the more economically attractive approach.
• An approach that is not considered in the results of Table 27 is changing of fuels to
lower mercury content fuels (such as conversion from PRB to K-Fuel, as described
earlier). However, depending upon the incremental cost of these fuels relative to the
current fuel, they could be more cost effective in the application than the additional
controls shown.
• The Current Cost Estimates use PAC sorbent inj ection levels that have been measured
in field tests or in pilot tests using currently available PAC sorbents19'30 assume that
all cases used PAC and that all fly ash that comes in contact with used PAC must be
disposed of. As discussed earlier, the assumption that all fly ash currently is sold is a
most conservative assumption. For the majority of plants that currently landfill their
fly ash, the incremental costs of PAC injection are estimated to be from 0.37 mills/kWh
to about 1.0 mills/kWh less than shown in Table 30. Moreover, current research programs
offer the potential to reduce operating costs in one of two ways: (1) reduction of sorbent
costs by development of less expensive sorbents, or sorbents that are more efficient
in mercury capture than existing PAC sorbents or (2) reduction or elimination of disposal
costs by utilizing sorbent materials that can be beneficially reused in the same manner
as fly ash that may be captured with it. Sensitivity analysis showed that, if disposal costs
caused by PAC inj ection could be avoided, the cost of control could be reduced by 15
to 17 percent for facilities where sorbent is collected in downstream ESPc's and by
about 80 percent for facilities where sorbent is collected in downstream FFs. The reason
the impact is greater for FFs than for ESPs is because PAC injection rates tend to be
much lower for FFs, and the impact of ash disposal cost on total cost is therefore much
greater.
86
-------
Table 30. Estimated Cost of Mercury Control—Current and Potential Cost Estimates
Coal
Type
Bitt
Bit
Bit
Bit
Bit
Bit
Bit
Bit
Bit
Bit
PRB
PRB
PRB
S (%)
3
3
3
3
3
3
3
0.6
0.6
0.6
0.5
0.5
0.5
Existing Controls
ESP + FGD
SCR + ESP + FGD
FF + FGD
SCR + FF + FGD
ESPh + FGD
SDA + ESP
SDA + FF
ESP
FF
ESPh
ESP
FF
ESPh
Additional Controls
PAC + PJFF + GEMS
PAC + GEMS
GEMS
GEMS
GEMS
PAC + PJFF + GEMS
PAC + (PJFF) + CEMSa
PAC + GEMS
PAC + PJFF + GEMS
PAC + CEMSb
PAC + CEMSb
PAC + PJFF + GEMS
PAC + PJFF + GEMS
PAC + CEMSb
PAC + CEMSb
PAC + PJFF + GEMS
Cost Estimates of
Additional Controls
(mills/kWh)
1.144-1.430
0.03-0.04
0.03-0.04
0.03-0.04
1.149-1.437
2.211-3.096a
0.05-0.370
1.171-1.751
0.003-0.510
1 .205-1 .804
1 .236-1 .903
1.122-1.266
1 .273-1 .960
a For 80 percent control assumes no PJFF. For 90 percent control, assumes full-size PJFF (sized for full ash
loading and much more expensive than if sized for downstream of an ESP or FF) being necessary for 90
percent control.
b The calculations performed to generate the results in this table assumed that all collected fly ash is currently
sold, which is the most conservative situation. Therefore, these calculations include costs to landfill fly ash
with an impact to total cost of around 0.37 mills/kWh for the low sulfur bituminous coal and around 1.01
mills/kWh for the low sulfur bituminous coal. In many cases these costs will not apply because ash may
currently be landfilled or because ash may not be rendered completely unacceptable for re-use.
In this work, two multipollutant controls were also evaluated: ECO and advanced dry FGD. ECO
is an emerging technology, and advanced dry FGD is a technology that is being introduced to the
United States by several suppliers. There currently is no commercial experience with ECO at this
time. There is limited commercial experience with advanced dry FGD on coal-fired boilers in
the United States; however, there is more experience with the technology in Europe. Therefore,
the cost information presented in Table 31 should be regarded as preliminary, especially for ECO.
However, it should be kept in mind, when considering the higher costs associated with multi-
pollutant controls over controls that remove only mercury, that other environmental benefits, such
as SO2 or NOX control, are realized in addition to mercury reduction.
The universe of multipollutant controls is not limited to the technologies presented in this work.
Therefore, other technologies that could offer lower control costs may become available to users.
87
-------
Table 31. Estimated Costs of Multipollutant Controls
Coal
Type
Bit
Bit
Bit
Bit
PRB
PRB
S (%)
3
3
0.6
0.6
0.5
0.5
Existing Controls
ESP
ESP
ESP
ESP
ESP
ESP
Additional Controls
ECO + GEMS
Adv Dry FGD + GEMS
ECO + GEMS
Adv Dry FGD + GEMS
ECO + GEMS
Adv Drv FGDa + GEMS
Cost Estimates of
Additional Controls
(mills/kWh)
3.28-7.21
7.94-9.65
5.34-12.33
3.69-6.58
4.54-11.53
3.89-6.79a
Mercury control with advanced dry FGD on PRB fuels may be uncertain
K-Fuel, which is developed from low rank western fuels, was also evaluated. It was shown in
this work that K-Fuel offers roughly 60 percent, sometimes more, reduction of mercury in the
fuel (on a heating value basis). This is achieved by simultaneously reducing the mercury
concentration in the fuel while increasing the fuel's heating value. K-Fuel may also offer the facility
other environmental or operating benefits besides mercury reduction. Moreover, it was shown
that mercury reductions beyond the 60 percent provided by K-Fuel (versus PRB) can be achieved
with PAC inj ection. As the first commercial K-Fuel plant will be built soon, there is no commercial
experience with this technology at this time. Costs are evaluated based on cost premium for K-Fuel
over a facility's existing fuel. For example, if K-Fuel were available at an incremental cost of
$0.12/MMBtu over a base fuel, then the incremental cost would be approximately 1.26 mills/kWh.
-------
7.0 SUMMARY
Cost estimates of PAC inj ecti on-based mercury control technologies for coal-fired electric utility
boilers have been determined. These estimates include those based on currently available data
as well as proj ections for future applications of more effective sorbent. Estimates based on currently
available data range from 0.03-3.096 mills/kWh. However, the higher costs are usually associated
with the minority of plants using SDAs plus ESPs or the small number of plants using ESPhs.
Potential costs, developed assuming improvements in sorbent technology for mercury removal,
range from 0.03-1.69 mills/kWh excluding applications with SDAs plus ESPs or with ESPhs.
At the low end of these cost ranges, 0.03 mills/kWh, it is assumed that no additional control
technologies are needed but that mercury monitoring will be necessary. In these cases, high mercury
removal may be the result of the type of NOX and SO2 control measures currently employed.
The estimates based on currently available data may be conservative for the following reasons:
(1) They assume that prior to addition of controls all fly ash is sold, and after addition of controls,
any fly ash that is combined with spent PAC must be disposed of; (2) The estimates of PAC
injection rates for PRB-fired boilers with PAC collected by downstream ESP are based upon
experience at Pleasant Prairie Power Plant, which fires a fuel with a lower chlorine content coal
than is typical for a PRB fuel; (3) A 65 percent capacity factor is assumed for all cases. The first
assumption is conservative because most plants do not currently sell their fly ash; furthermore,
fly ash sales might be possible even with small amounts of PAC present. The impact of this
assumption was estimated to be in the range from 0.37 mills/kWh to about 1.0 mills/kWh,
depending upon fuel characteristics. Thus, for plants that do not currently sell their fly ash (most
plants), the actual incremental cost of control would be lower than what we have estimated.
Moreover, improved sorbents available in the future may eliminate, or at least mitigate, any impact
of sorbent injection on disposal costs. The second estimate is conservative because, at the PPPP
tests, mercury reduction by PAC injection was limited to around 70 percent. This is believed to
be due, at least in part, to PPPP's unusually low chlorine content—much lower than for most PRB
fuels. Therefore, other PRB fueled boilers with chlorine contents typical of a PRB coal, and
equipped with a downstream ESP, may be more effectively controlled than what was demonstrated
at PPPP. The final assumption is conservative particularly for larger plants that are likely to be
base loaded with higher capacity factors.
Results of sensitivity analyses conducted on total annual cost of mercury controls reflect that:
(1) Elimination of disposal costs could reduce costs by 80 percent for some cases (PAC injection
upstream of an existing FF) and by 17 percent in others (PAC injection upstream of an existing
89
-------
ESP); (2) Using a more effective PAC sorbent might reduce costs by 20 percent; (3) The cost of
a retrofit PJFF has a significant impact on control costs and, over the expected cost range, may
cause cost to vary from about 1.2 mills/kWh to about 1.8 mills/kWh; (4) Fertilizer product value
and the value of process power requirement have a very significant impact on the cost of ECO,
with fertilizer product value being most significant for high sulfur fuel applications; and (5) As
expected, lime reagent cost has a large impact on the total cost of using advanced dry FGD.
Based on this work, it is expected that future efforts in R&D are likely to focus on improved
understanding of both mercury speciation across SCRs and the beneficial effects of combinations
of SCR with wet FGD, and on developing sorbents that can improve performance and cost of
sorbent-based mercury control technologies. Multipollutant control technologies, which are more
costly than single-pollutant mercury control technologies but offer other environmental benefits,
will be another area for further development that could improve the cost of reducing emissions
from coal-fired power plants. Finally, removing mercury from the coal, along with other fuel quality
improvements, may prove to be a cost effective approach for reducing emissions.
90
-------
8.0 REFERENCES
1. Keating, M.H.; Mahaffey, K.R. Mercury Study Report to Congress, Office of Air Quality
Planning and Standards and Office of Research and Development, U.S. Environmental
Protection Agency, Research Triangle Park, NC, December 1997, EPA-452/R-97-003 (NTIS
PB98-124738). Also available atthe web site http://www.epa.gov/ttn/oarpg/t3rc.html (Accessed
October 2003).
2. Mishima, Akio. Bitter Sea: The Human Cost of Minamata Disease, Kosei Publishing Co.,
Tokyo, Japan, 1992.
3. Regulatory Finding on the Emissions of Hazardous Air Pollutants from Electric Utility Steam-
Generating Units; Federal Register 2000, 65 (245).
4. Srivastava, R.K.; Sedman, C.B.; Kilgroe, J.D. "Performance and cost of Mercury Emission
Control Technology Applications on Electric Utility Boilers," EPA-600/R-00-083, September
2000.
5. Boyle, P.; DoVale, A., "Commercial Readiness of Electro-Catalytic Oxidation (ECO)
Multipollutant Control Technology, paper available from Powerspan and Wheelabrator.
6. Brown, T.D.; Smith, D.N.; Hargis, R.A.; O'Dowd, WJ. "1999 Critical Review, Mercury
Measurement and Its Control: What We Know, Have Learned, and Need to Further Investigate,"
Journal of the Air & Waste Management Association, June 1999, pp. 1-97.
7. Electric Utility Steam Generating UnitMercury Emissions Information Collection Effort, OMB
Control No. 2060-0396, EPA, Office of Air Quality Planning and Standards, Research Triangle
Park, NC.
8. Control of Mercury Emissions From Coal-Fired Electric Utility Boilers: Interim Report
Including Errata Dated 3-31-02, EPA-600/R-01-109, April 2002.
9. Hocquel, M.; Unterberger, S.; Hein, K.; Bock, J. "Behavior of Mercury in Different Gas
Cleaning Stages", International Conference on Air Quality, Crystal City, VA, September 9-12,
2002.
10. Richardson, C.; Machalek, T.; Miller, S.; Dene, C; Chang, R. "Effect ofNOx Control Processes
on Mercury Speciation in Utility Flue Gas", International Conference on Air Quality, Crystal
City, VA, September 9-12, 2002.
11. Selective Catalytic Reduction Mercury Field Sampling Proj ect, EPA-600/R-02-096, December
2002.
12. Lee, C.; Srivastava, R.; Ghorishi, S.B.; Hastings, T.W.; Stevens, F.M. "Study of Speciation
of Mercury under Simulated SCR NOX Emission Control Conditions," AWMAPaper 03-A-41-
91
-------
AWMA, Presented at the Joint EPRI DOE EPA Combined Utility Air Pollution Control
Symposium, The Mega Symposium, Washington, D.C., May 19-22, 2003.
13. Personal Communication, e-mail, from Ravi Srivastava, U.S. EPA, March 13, 2003.
14. Lissianski, V.; Zamansky, V.; Maly, P.; Seeker, R. "Control of NOX and Mercury Emissions
During Coal Reburning", 95th Annual A&WMA Conference and Exhibition, Baltimore, MD,
June 23-27, 2002.
15. Lissianski, V.; Zamansky, V.; Maly, P.; Seeker, R. "Integration of Combustion Modifications
with Mercury Control," Air Quality III Conference, Washington, DC., September 11, 2002.
16. Lissianski, V.; Maly, P.; Ho,L.; Zamansky, V.; Glenn,E.; Seeker, R. "Multi-PollutantControl
for Boilers Firing Low-Rank Coals," Presented at the Joint EPRI DOE EPA Combined Utility
Air Pollution Control Symposium, The Mega Symposium, Washington, D.C., May 19-22,
2003.
17. Srivastava, R.K.; Jozewicz, W. Flue Gas Desulfurization: The State of the Art, J. Air & Waste
Manage. Assoc. 51(12), 1676-1688, December 2001.
18. An Assessment of Mercury Emissions from U. S. Coal Fired Power Plants, EPRI, Palo Alto,
CA, 2000. EPRI 1000608
19. Staudt, J.E.; Jozewicz, W.; Srivastava, R. "Modeling Mercury Control with Powdered Activated
Carbon", AWMAPaper 03-A-17-AWMA, Presented atthe Joint EPRI DOE EPA Combined
Utility Air Pollution Control Symposium, The Mega Symposium, May 19-22, 2003,
Washington, D.C.
20. Study of Hazardous Air Pollutant Emissions from Electric Utility Steam Generating Units
- Final Report to Congress, Volume 1, Office of Air Quality Planning and Standards, U.S.
Environmental Protection Agency, Research Triangle Park, NC, February 1998, EPA-453R-98-
004a (NTIS PB98-131774).
21. Waugh, E.G.; Jensen, B.K.; Lapatnick, L.N.; Gibbons, F.. Sjostrom, S. Ruhl, J.; Slye, R.; Chang,
R. "Mercury Control in Utility ESPs and Baghouses through Dry Carbon-Based Sorbent
Injection Pilot-Scale Demonstration," EPRI-DOE-EPA Combined Air Pollutant Control
Symposium, Particulates and Air Toxics, Volume 3, EPRI TR-108683-V3, Electric Power
Research Institute, Palo Alto, CA, August 1997, pp. 1-15.
22. Haythornthwaite, S.M.; Smith, J.; Anderson, G.; Hunt, T.; Fox, M.; Chang, R.; Brown, T.
"Pilot-Scale Carbon Injection for Mercury Control at Comanche Station," presented at the
A&WMA 92nd Annual Meeting & Exhibition, St. Louis, MO, June 1999.
23. Haythornthwaite, S.M.; Sjostrom, S.; Ebner, T.; Ruhl, J. Slye, R. Smith, J.; Hunt, T.; Chang,
R.; Brown, T.D. "Demonstration of Dry Carbon-Based Sorbent Injection for Mercury Control
in Utility ESPs and Baghouses," EPRI-DOE-EPA Combined Air Pollutant Control Symposium,
92
-------
Particulates and Air Toxics, Volume 3, EPRITR-108683 - V3, Electric Power Research Institute,
Palo Alto, CA, August 1997.
24. Redinger, K.E.; Evans, A.P.; Bailey, R.T.; Nolan, P.S. "Mercury Emissions Control inFGD
Systems," EPRI-DOE-EPA Combined Air Pollutant Control Symposium, Parti culates and
Air Toxics, Volume 3, EPRI TR-108683-V3, Electric Power Research Institute, Palo Alto,
CA, August 1997.
25. Durham, M.; Bustard, J.; Schlager, R.; Martin, C.; Johnson, S.; Renninger, S. "Field Test
Program to Develop Comprehensive Design, Operating Cost Data for Mercury Control Systems
on Non-Scrubbed Coal-Fired Boilers," AWMA 94th Annual Conference and Exhibition,
Orlando, FL, June 24-28, 2001.
26. Bustard, J.; Durham, M.; Lindsey, C.; Starns, T.; Baldrey, K.; Martin, C.; Schlager, R.;
Sjostrom, S.; Slye, R.; Renninger, S.; Monroe, L.; Miller, R.; Chang, R., "Full-ScaleEvaluation
of Mercury Control with Sorbent Injection and COHPAC at Alabama Power E.G., Gaston,"
DOE-EPRI-U.S. EP A-A&WMA Power Plant Air Pollutant Control Mega Symposium, Chicago,
IL, August 20-23, 2001.
27. "Gaston Demonstrates Substantial Mercury Removal with Sorbent Injection," Power
Engineering, vol. 106, no. 11, November 2002.
28. Starns, T.; Bustard, J.; Durham, M.; Lindsey, C.; Martin, C.; Schlager, R.; Donnelly, B.;
Sjostrom, S.; Harrington, P.; Haythornthwaite, S.; Johnson, R.; Morris, E.; Chang, R.;
Renninger, S., "Full-Scale Test of Mercury Control with Sorbent Injection and an ESP at
Wisconsin Electric's Pleasant Prairie Power Plant," AWMA 95th Annual Conference and
Exhibition, Baltimore, June 23-27, 2002.
29. DOE National Energy Technology Laboratory Mercury Field Evaluati on-PG&ENEG Sal em
Harbor Station-Unit 1, Project No. 00-7002-76-10, Field Evaluati on Summary Report, January
2003.
30. Memo from Michael Durham, ADA Environmental Solutions to Mary Jo Krolewski, U.S.
Environmental Protection Agency.
31. Powerspan Corporation Web site, jrtt]ii//www^ (Accessed October 2003).
32. Personal Communication, telephone conversation with Phil Boyle, Powerspan, March 14,
2003.
33. Personal Communication, e-mail and telephone conversation with Phil Boyle, Powerspan,
March 6, 2003.
34. Boyle, P.; DoVale, A. "Commercial Readiness of Electro-Catalytic Oxidation (ECO™)
Multipollutant Control Technology," paper available from Powerspan and Wheelabrator.
93
-------
35. DePriest, W.; Pope, F., "Economic Evaluation of a Multi-Pollutant Control Approach for
Ameren UE' s Sioux Plant," International Power Generation Conference, Orlando, December
2002.
36. Personal Communication, e-mail from Will Goss, Beaumont Environmental, February 6,2003.
37. F.L. SmidthAirtech Corporation web site, http://www.flsairtech.com (Accessed October 2003).
38. Personal Communication, telephone conversation with Will Goss, RJM-Beaumont, February
4,2003.
39. Personal Communication, telephone conversation with Milas Ballard of Lurgi, February 20,
2003.
40. Graf, R., "High-Efficiency Circulating Fluidized Bed Scrubber," DOE-EPRI-U.S. EPA-
A&WMA Power Plant Air Pollutant Control Mega Symposium, Chicago, IL, August 20-23,
2001.
41. Design for 250 MW and 500 MW plants provided by Beaumont Environmental Systems,
February 4, 2003.
42. "Roanoke Valley Energy Facility Mercury Testing," report prepared by Arcadis Geraghty &
Miller for Roanoke Valley Energy Facility, January 25, 2001.
43. RJM-Beaumont Next Generation Multi-Pollutant Technology, available athttg;//www.MMxom
(Accessed October 2003).
44. Lurgi Lentjes CFB Scrubber Reference List, March 7, 2003.
45. Personal Communication with Sam Kumar, F.L. Smidth AirTech, March 3, 2003.
46. " 10 MWe Demonstration of Gas Suspension Absorption." Clean Coal Technology Program:
Program Update 2000. DOE/FE-0437, April 2001, pp 5-22 to 5-25.
47. Circulating Fluidized Bed - Flue Gas Desulfurization Technology, Prepared for Lurgi-Lentj es
North America, March 2003.
48. Black & Veatch Corporation, "Effective Mercury Reduction Strategy for Western Coal -Report
to EPA - K-Fuel Technology," Report to U. S. Environmental Protection Agency, March 2003.
49. Personal Communication with Ted Venners, KFx Corporation, May 8, 2003.
50. Black & Veatch Corporation, "K-Fuel Market Value and Market Potential Preliminary
Evaluation," August 2002.
51. U.S. Patent No. 5,316,737, "Process for Removing NOX and SOX from Exhaust Gas," Arthur
P. Skelley, et al; May 31, 1994.
52. Goss, W.L.; Lutwen, R.C.; Ferrell, R.; Suchak, N.; Hwang, S.C. "A Report of the Startup of
a Multi-Pollutant Removal System for NOX, SOX, and Particulate Control Using a Low
Temperature Oxidation System on a 25 MW Coal Fired Boiler," Power-Gen 2001, Las Vegas,
NV, December 12, 2001.
94
-------
53. Ferrell, R.; Suchak,N.; Hwang, S.C.; Tseng, 1; Kelton, R.; Goss, W.; Lutwen, R. "AReport
on the Application of Low Temperature Oxidation for Control of NOX Emissions," presented
at ICAC Forum 02, Houston, TX, February 13, 2002.
54. Jarvis,J.B.;Day, A.T.; Suchak, N J. "LoTOxTM Process Flexibility and Multi-pollutant Control
Capability," Combined Power Plant Air Pollutant Control Mega Symposium, Washington
DC, May 19-22,2003.
55. Altman, R.; Buckley, W.; Ray, I. "Multi-Pollutant Control with Dry-Wet Hybrid ESP
Technology," Combined Power Plant Air Pollutant Control Mega Symposium, Washington
DC, May 19-22,2003.
56. Montgomery, J.L.; Battelson,D.M., Whitworth, C.G., Ray, I, Buckley, W., Reynolds, J., and
Altman, R. "Plasma-Enhanced Electrostatic Precipitation Technology for Control of Elemental
Mercury," International Power Generation Conference, Orlando, December 2002.
57. Montgomery, J.L., Battelson, D.M.; Whitworth, C.G.; Ray, I; Buckley, W.; Reynolds, J.;
Altman, R. "Latest Developments of the Plasma-Enhanced Electrostatic Precipitator for Mercury
Removal in Off-gas," Combined Power Plant Air Pollutant Control Mega Symposium,
Washington DC, May 19-22, 2003.
58. Blythe, G.M.; Richardson, M.F.; Lani, B.W.; Rhudy, RG "Pilot Testing of Oxidation Catalysts
for Enhanced Mercury Control by Wet FGD," Combined Power Plant Air Pollutant Control
Mega Symposium, Washington DC, May 19-22, 2003.
59. Butz, J.; Lovell, J.; Broderick, T.; Sidwell, R.; Turchi, C.; Kuhn, A., "Evaluation of Amended
Silicates™ Sorbent for Mercury Control," Combined Power Plant Air Pollutant Control Mega
Symposium, Washington DC, May 19-22, 2003.
60. Sjostrom, S.; Chang R., Apparatus and Method of the Removal of Contaminants in Gases,
U.S. Patent 5,948,143, 1999 and U.S. Patent 6,136,072, 2000.
61. Sjostrom, S.; Chang, R.; Strohfus, M.; Johnson, D.; Hagley, T.; Ebner, T.; Richardson, C.;
Belba, V., "Development and Demonstration of Mercury Control by Adsorption Processes
(MerC APTM)," Combined Power Plant Air Pollutant Control Mega Symposium, Washington
DC, May 19-22,2003.
62. "Coal Plant Achieves 99% NOX and SOX Removal with Multi-pollutant Technology," Power
Engineering Magazine, May 2002.
63. Enviroscrub brochure: "Pahlman™ Process Overview"
64. Haddad, E.; Ralston, J.; Green, G.; Castagnero, S. "Full-Scale Evaluation of a Multi-Pollutant
Reduction Technology: SO2, Hg, and NOX," Combined Power Plant Air Pollutant Control
Mega Symposium, Washington DC, May 19-22, 2003.
65. Gebert, R.; Rinschler, C.; Davis, D.; Leibacher, U.; Studer, P.; Eckert, W.; Swanson, W.;
Endrizzi, J.; Hrdlicka, T.; Miller, S. J.; Jones, M. L.; Zhuang, Y.; Collings, M. "Commercializa-
95
-------
tion of the Advanced Hybrid™ Filter Technology," Air Quality III Conference, Arlington,
VA, September 2002, SessionBS.
66. Durante, V.A.; Stark, S.; Gebert, R.; Xu, Z.; Bucher, R.; Keeney R.; Ghorishi, B. "ANovel
Technology to Immobilize Mercury from Flue Gases," Combined Power Plant Air Pollutant
Control Mega Symposium, Washington DC, May 19-22, 2003.
67. Gostomczyk, M.A.; Jozewicz, W. "Simultaneous Control of SO2,NOX, and Mercury Emissions
from Coal-fired Boilers," Combined Power Plant Air Pollutant Control Mega Symposium,
Washington DC, May 19-22, 2003.
68. Electric Power Development Corp. "Isogo Thermal Power Station/Outline of Facilities,"
company brochure describing the Isogo power station, 2000.
69. Mitsui Mining Co., Ltd. "Dry TypeDesulfurization andDenitrification Technology for Coal-
fired Power Plants," company brochure, May 2000.
70. Olson, D.G.; Tsuji, K.; Shiraishi, I. "The Reduction of Gas Phase Air Toxics from Combustion
and Incineration Sources Using the MET-Mitsui-BF Activated Coke Process," 1999
International Ash Utilization Symposium, Lexington, KY, October 18-20, 1999.
71. Keith, R.; Blag, R.; Burklin, B.; Kosmicki, B.; Rhodes, D.; Waddell, T. "Coal Utility
environmental Cost (CUECost) workbook User's Manual, Version 1.0," EPA-600/R-99-056
(NTIS PB99-151938), June 1999.
72. Srivastava, R.K. "Current and Emerging Mercury Control Technologies," Western Mercury
Workshop, April 21-23, 2003, Denver, CO.
73. Standard Specification for Coal Fly Ash and Raw or Calcined Natural Pozzolan for Use as
a Mineral Admixture in Concrete, ASTM Standard C618-03.
74. Dombrowski, K.D.; Machalek, T.; Richardson, C.; Chang, R.; Rostam-Abadi, M. "Recent
Developments in EPRI's Novel Mercury Sorbent Development Program," Combined Power
Plant Air Pollutant Control Mega Symposium, Washington DC, May 19-22, 2003.
75. Biermann, J.; Voogt, N.; Sorooshian, A.; Wendt, J.; Merritt, R.; Gale, T. "In-Flight Capture
of Elemental and Oxidized Mercury in Coal Combustion Flue Gases by a Novel Mineral Sorbent
Between320 and 1540 °F (160-840 °C)," Combined Power Plant Air Pollutant Control Mega
Symposium, Washington DC, May 19-22, 2003.
96
------- |