United States
Environmental Protection
Agency
EPA-600/R-05/034
March 2005
Multipollutant Emission
Control Technology Options
for Coal-fired Power Plants
-------
Foreword
The U. S. Environmental Protection Agency 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 is the agency's center for investigation of
technological and management approaches for reducing risks from threats to human health and the
environment. The focus of the laboratory's research program is on methods for the prevention and control
of pollution to air, land, water, and subsurface resources, protection of water quality in public water
systems; remediation of contaminated sites and groundwater; and prevention and control of indoor air
pollution. The goal of this research effort is to catalyze development and implementation of innovative,
cost-effective environmental technologies; develop scientific and engineering information needed by EPA
to support regulatory and policy decisions; and provide technical support and information transfer to
ensure effective implementation of environmental regulations and strategies.
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.
Sally Gutierrez, 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 or commercial 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.
-------
EPA-600/R-05/034
March 2005
Multipollutant Emission Control Technology Options
for Coal-fired Power Plants
Prepared by:
E. Stratos Tavoulareas
Energy Technologies Enterprises Corporation
1112 Towlston Road
McLean, VA 22105
and
Wojciech Jozewicz
ARCADIS G&M, Inc.
4915 Prospectus Drive, Suite F
Durham, NC 27713
EPA Contract No. 68-C-99-201, Work Assignment 6-13
EPA Project Officer: Ravi K. Srivastava
National Risk Management Research Laboratory
Research Triangle Park, NC 27711
Prepared for:
U.S. Environmental Protection Agency
Office of Research and Development
Washington, DC 20460
and
U.S. Environmental Protection Agency
Office of Air and Radiation
Washington, DC 20460
-------
Abstract
This report presents and analyzes various existing and novel control technologies designed to achieve
multi-emission [sulfur dioxide (SO2), nitrogen oxide (NOX), and mercury (Hg)] reductions. Summary
descriptions are included of 27 multi-emission control technologies that have reached a stage of
development beyond pilot scale. These can broadly be divided into: environmental control options (post-
combustion controls), advanced power generation options, and power plant upgrading and operating
options. For each evaluated technology, the report includes background information, applicability, status
of commercialization, any secondary environmental impacts of the technologies, identification of primary
process variables that impact performance relative to NOX, SO2, and Hg, as well as capital and operation
and maintenance costs.
More than half of the technology options listed are in the commercial or early commercial stage (15 out of
27). However, nearly all the technology options in the commercial stage are proven SO2 control
technologies, which also appear to remove Hg. Some technologies, such as Advanced Silicate and
Confined Zone Dispersion, have been tested in either the pilot or demonstration scale in the early phase of
the U.S. Department of Energy's Clean Coal Technology program, but have not been adopted by
industry. Some of these technologies could become more cost-effective as environmental requirements
evolve. Activated coke, electro-catalytic oxidation, EnviroScrub, and the combination of flue gas
desulfurization with LoTOx or selective catalytic reduction exhibit the potential to significantly control
(above 80 percent) all three pollutants (SO2, NOX, and Hg).
Although the report is limited to addressing technologies with a certain level of maturity, the authors
expect a rapid technological evolution in the development and commercialization of several multi-
emission control technologies not addressed in this report.
-------
Acknowledgments
ARCADIS 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 provided by
Dr. Ravi Srivastava of EPA's National Risk Management Research Laboratory, Office of Research and
Development, and many helpful discussions with Mr. Reynaldo Forte and Mr. Sikander Khan, both of
EPA's Clean Air Markets Division, Office of Air and Radiation. Helpful suggestions were received from
Mr. Kevin Culligan and Ms. Mary Jo Krolewski of EPA's Clean Air Markets Division.
in
-------
1.1 Intended Use and Organization of this Report..
1.2 Background
1.3 Human Health and Environmental Impacts
1.3.1 Fine Particles
1.3.2 Ground-Level Ozone
1.3.3 Acid Deposition
1.3.4 Visibility
1.3.5 Mercury
1.4 Multi-emission Control Options
1.5 References
Table of Contents
Abstract ii
Acknowledgments iii
Table of Contents iv
List of Tables vi
List of Figures vi
List of Acronyms vii
Conversion of Units ix
Metric Prefixes ix
Chapter 1 Introduction -1
-1
-2
-2
-2
-3
-3
-3
-4
-4
-5
Chapter 2 Characterization of U.S. Power Plants and Associated Emissions 2-1
2.1 SO2 Emissions from Coal-Fired Power Plants 2-2
2.2 NOX Emissions from Coal-Fired Power Plants 2-4
2.3 Mercury Emissions from Coal-Fired Power Plants 2-8
2.4 CO2 Emissions from Coal-Fired Power Plants 2-9
2.5 References 2-10
Chapter 3 Multi-emission Control Technologies and Options for Coal-fired Power Plants 3-1
3.1 Introduction 3-1
3.2 Environmental Control Options 3-3
3.2.1 SO2 and Mercury Control 3-3
3.2.1.1 Dry Scrubbers 3-3
3.2.1.1.1 Conventional Dry Scrubbers 3-3
3.2.1.1.2 Advanced Dry Scrubbers 3-8
3.2.1.2 Sorbent Injection 3-12
3.2.1.2.1 Activated Carbon with Particulate Controls 3-13
3.2.1.2.2 SO2Sorbents 3-19
3.2.1.2.3 Combined Mercury and SO2 Sorbents 3-24
3.2.1.3 Wet Scrubbers 3-27
3.2.1.3.1 Wet Scrubbers with Mercury Oxidation Processes 3-28
3.2.1.3.2 Wet Scrubbers with Wet ESP 3-31
3.2.1.3.3 Plasma-Enhanced ESP (PEESP) 3-34
3.2.1.4 MerCAP 3-36
3.2.2 SO2 and NOX Control 3-38
3.2.2.1 Electron Beam Process 3-38
3.2.2.2 ROFA-ROTAMIX (Mobotec) 3-42
3.2.2.3 SNOX 3-45
iv
-------
3.2.2.4 SOx-NOx-Rox Box (SNRB) 3-49
3.2.2.5 THERMALONOx and FLU-ACE Process 3-51
3.2.3 SO2, NOX, and Mercury Control 3-53
3.2.3.1 Activated Coke 3-53
3.2.3.2 Electro-Catalytic Oxidation (ECO) 3-57
3.2.3.3 SCR and Wet FGD 3-63
3.2.3.4 EnviroScrub Pahlman 3-67
3.2.3.5 LoTOx 3-69
3.2.3.6 K-Fuel 3-74
3.2.4 Advanced Power Generation Technology Options 3-78
3.2.4.1 Circulating Fluidized Bed Technology 3-78
3.2.4.2 Integrated Gasification Combined Cycle 3-82
3.2.4.3 Pressurized Fluidized Bed Combustion (PFBC) 3-86
3.2.4.4 Supercritical Pulverized Coal Plant 3-90
3.2.5 Power Plant Upgrading and Operating Options 3-94
3.2.5.1 Fuel Blending and Cofiring 3-94
3.2.5.2 Plant Efficiency Improvements 3-100
3.2.5.3 Power Plant Optimization 3-103
Chapter 4 Summary 4-1
-------
List of Tables
Table 2-1. Emissions in 2001 from Coal and Non-coal Units 2-2
Table 2-2. SO2 Emissions from Units with FGD Installed in 1990-1999 2-4
Table 2-3. 1999 NOX Emissions for Title IV Affected Units 2-6
Table 2-4. Mean Mercury Emission Reduction for Pulverized-coal-fired Boilers 2-8
Table 3-1. Average Mercury Reductions for Spray Dryer and ESP-FF per Coal Rank from ICR Data ...3-5
Table 3-2. Duct Injection of Lime Slurry 3-20
Table 3-3. Duct Injection of Dry Lime 3-21
Table 3-4. Duct Injection of Sodium 3-21
Table 3-5. Summary of MerCAP Field Sites 3-37
Table 3-5 Sintering Plants Utilizing Activated Coke 3-55
Table 3-6. Selected SCR Test Results 3-65
Table 3-7. Mercury Data for Various Coals 3-71
Table 3-8. CFB Units largerthan 200 MW 3-80
Table 3-9. Commercial-Size IGCC Power Plants 3-83
Table 3-10. PFBC Plants and Operating Experience 3-88
Table 3-11. Power Plants with Supercritical Design Parameters 3-91
Table 3-12. Equivalent Availability for Subcritical and Supercritical Power Plants 3-92
Table 3-14. Main O&M Impacts and Potential Mitigations 3-98
Table 4-1. Summary Descriptions of 27 Multi-emission Control Technologies for Coal-fired Units 4-3
List of Figures
Figure 2-1. Major sources of SO2 and NOX emissions in 2001 and of mercury emissions in 1998 2-1
Figure 2-2. Annual SO2 emissions for coal-fired units 2-3
Figure 2-3. Annual NOX emissions for coal-fired units 2-5
Figure 2-4. Annual CO2 emissions for coal-fired units 2-9
Figure 3-1. Mercury removal across spray dryer and ESP-FF from ICR data 3-5
Figure 3-2. Mercury removal versus activated carbon loading in a pilot scale ESP and baghouse 3-16
Figure 3-3. E-Beam process schematic 3-39
Figure 3-4. NOX emissions from Cape Fear Unit 5: comparison before and after ROFA 3-44
Figure 3-5. SNOX process schematic 3-46
Figure 3-6. Activated coke process flow diagram 3-54
Figure 3-7. ECO process flow diagram 3-59
Figure 3-8. ECO power consumption versus NOX 3-62
Figure 3-8. Schematic diagram of LoTOx™ system 3-70
Figure 3-9. NOX with LoTOx at MCO's 25 MW-thermal gas and coal fired boiler 3-72
Figure 3-10. K-FuelPMR technology 3-75
Figure 3-11. K-Fuel thermal separation unit 3-75
Figure 3-12. CFB process schematic 3-79
Figure 3-13. PFBC conceptual design 3-87
VI
-------
List of Acronyms
AC Activated carbon
ACI Activated carbon injection
AD VACATE Advanced Silicate
AFBC Atmospheric fluidized bed combustor
APCD Air pollution control device
BOP Balance of plant
Btu British thermal unit
CAAA Clean Air Act Amendments of 1990
CCT Clean Coal Technology
CDS Circulating Dry Scrubbing
CEDF Clean Environment Development Facility
CFB Circulating fluidized-bed
CFD Computerized Fluid Dynamics Modeling
CO2 Carbon dioxide
COHPAC Compact hybrid particulate collector
CZD Confined-zone Dispersion
DBA Dibasic Acid
DOE Department of Energy
ECO Electro-catalytic Oxidation
EdF Electricite de France
EIA Energy Information Administration
EOY End of year
EPA Environmental Protection Agency
EPRI Electric Power Research Institute
E-SOX ESP enhanced for SO2 removal
ESP Electrostatic precipitator
ESP-CS Cold-side electrostatic precipitator
ESP-HS Hot-side electrostatic precipitator
FF Fabric filter
FGD Flue Gas Desulfurization
GSA Gas Suspension Absorption
HALT Hydrate Addition at Low Temperature
HAP Hazardous air pollutant
Hg Mercury
HHV Higher heating value
HP Horse power
HRSG Heat recovery steam generator
HYPAS Hybrid pollution abatement system
ICR Information collection request
IDS In-duct scrubbing
IGCC Integrated gasification combined cycle
IPM Integrated planning model
kWh Kilowatt hour
KRW Kellogg Rust Westinghouse
Ib Pound-mass
vn
-------
LOI Loss-on-ignition
MM Million (106)
MSW Municipal solid waste
MW Megawatt (electric)
NAAQS National Ambient Air Quality Standards
NOX Nitrogen oxides
NSPS New source performance standards
NSR New source review
OFA Overfire Air
O&M Operation and maintenance
PAC Powdered activated carbon
PC Pulverized coal
PEESP Plasma-enhanced ESP
PFBC Pressurized fluidized-bed combustion
PM Particulate matter
PS Particulate scrubber
PSD Prevention of significant deterioration
RAP Rapid absorption process
RCFB Reflux circulating fluidized bed
ROFA Rotating opposed air
SCR Selective catalytic reduction
SDA Spray dryer absorber
SIP State implementation plan
SNCR Selective non-catalytic reduction
SO2 Sulfur dioxide
SRG SO2-rich gas
TAG Technical assessment guide
TVA Tennessee Valley Authority
TOXECON ACI combined with COHPAC
UNFCCC United Nations Framework Convention on Climate Change
WESP Wet electrostatic precipitator
WSA Wet gas sulfuric acid condenser
Vlll
-------
Conversion of Units
Multiply
POWER
Watts
Watts
Watts
ENERGY
Joules
Kilowatt-hours
(kWh)
Kilowatt-hours
Kilowatt-hours
Btu
MASS
Kilograms
Pounds
Ton (short or US)
Tonne (metric ton)
By To Obtain
1 Newton-meter/second or Joule/second
0.05692 Btu/minute
1.341 x 10"3 Horsepower
1 Watt-seconds or Newton-meters
3415 Btu
1.341 Horsepower-hours
3.60xl06 Joules
1,054 Joules
2.2046 Pounds
0.4536 Kilograms
2000 Pounds
1.1023 Ton (short or US)
Metric Prefixes
Prefix
Tera
Giga
Mega
Kilo
Hecto
Deka
Deci
Centi
Milli
Micro
Symbol
T
G
M
k
h
da
d
c
m
I-1
Value
12
10
109
106
103
102
10
lo-1
io-2
io-3
io-6
-------
-------
Chapter 1
Introduction
Recent changes in the structure of the electric utility industry, including the shift towards restructuring,
the growing demand for electricity generation, and environmental needs, are driving additional reductions
of multiple pollutants. Historically, industry has developed and implemented control technologies in
incremental steps to mitigate emissions of sulfur dioxide (SO2), oxides of nitrogen (NOX), particulate
matter (PM), and other pollutants, as driven by air pollution requirements. Control technologies that are
capable of simultaneously reducing emissions of multiple pollutants may offer the potential to achieve
this at lower cost and reduced footprint when compared to conventional controls.
This report presents and analyzes various existing and novel control technologies designed to achieve
multi-emission reductions. Having up-front knowledge of environmental performance, cost, and
limitations of multi-emission control technologies can help power companies select effective and less
expensive compliance strategies at individual plants, compared with compliance choices made when the
requirements are addressed individually.
1.1 Intended Use and Organization of this Report
The intended use for this report is to provide the current state-of-the art information on the multi-emission
control technologies and options that are available for coal-fired power plants with a capacity of 25
megawatts-electric (MW) or larger in the United States. For the purposes of this report, multi-emission
control technologies are those capable of simultaneously controlling emissions of at least two of these
three pollutants, NOX, SO2, and mercury, from electric utility sources. This evaluation includes
background information, applicability, status of commercialization, any secondary environmental impacts
of the technologies, identification of primary process variables that impact performance relative to NOX,
SO2, and mercury, as well as capital and operation and maintenance (O&M) costs, if available. In addition
to simultaneously controlling at least two of the SO2, NOX, and mercury emissions, some technologies
reviewed under this study promise to enable reductions of the carbon emissions that result from coal-
based electricity production. Such capability is viewed as an additional benefit and discussed for each
applicable technology, given that reducing carbon emissions would be advantageous to the fulfillment of
U.S. commitments under the United Nations Framework Convention on Climate Change (UNFCCC) and
would further the President's commitment to reduce greenhouse gas intensity in the United States by 18
percent over the next decade.1
The audience for this report is expected to comprise persons: (1) engaged in air pollution related research
and development (R&D) efforts, (2) responsible for developing and implementing emission control
strategies at sources, (3) and involved in developing air regulations, as well as (4) the interested public.
The report is organized into four chapters. The first chapter provides general background information on
air emissions from coal-fired power plants and multi-emission control technologies. The second chapter
characterizes U.S. power plants of 25 MW or greater and their respective air emissions. Chapter 3
presents evaluation summaries of 27 multi-emission control technologies. These summaries include
details on the technology, commercial readiness and industry experience, emission control performance,
future outlook, as well as capital and O&M costs. Chapter 4 is a summary of the report.
1-1
-------
1.2 Background
Electricity is critical to the well functioning of the residential, commercial, and industrial sectors in the
United States. More than 3,170 traditional electric utility plants and 2,110 non-utility power plants are
responsible for ensuring an adequate and reliable source of electricity to consumers in their service
territories.2 While electricity plays a critical role in sustaining the nation's economic growth, the
unintended by-products of electricity generation can have an undesirable effect on the environment and
public health. Most of these health impacts result from emissions produced through the combustion of
fossil fuels (coal, oil, and natural gas), which supply about 70 percent of the nation's requirements for
electricity generation.
As a result, the focus of recent regulatory actions has been to require power plants to reduce emissions of
NOX and SO2. The revisions of the National Ambient Air Quality Standards (NAAQS) for PM and ozone
may also limit power plant emissions. These revisions may require electric utility sources to adopt control
measures designed to reduce concentrations of fine PM in the atmosphere. Fine PM is PM at or below 2.5
micrometer in size (PM2 5). In addition, EPA has recently proposed a rule that calls for regulation of
mercury emissions from electric utility coal-fired plants. Concurrently, legislation has been proposed in
both the previous and current Congresses that would require simultaneous reductions of multiple
emissions, and the Administration's National Energy Policy3 recommends the establishment of
"mandatory reduction targets for emissions of three main pollutants: sulfur dioxide, nitrogen oxides, and
mercury."
This report focuses on control technologies that promise to simultaneously control more than one
pollutant from coal-burning power plants. The report focuses on emissions from coal-fired power plants
since these plants generate about 50 percent of the electricity used in the United States. The coal-burning
electric power industry is a major source of various air pollutant emissions including SO2, NOX, and
mercury. In addition, the combustion of fossil fuels contributes to CO2 emissions to the atmosphere.
1.3 Human Health and Environmental Impacts
Over the last 30 years, the reduction of emissions from the electric power sector has led to significant
human health and environmental benefits. Title IV of the Clean Air Act Amendments of 1990 (CAAA),
for example, has reduced acid deposition by approximately 25 percent and substantially reduced airborne
particulate matter.1 At the same time, the U.S. economy has expanded, and the amount of electricity
supplied to consumers has grown significantly. In spite of the progress made to date, the electric power
industry is still a significant source of SO2, NOX, and mercury emissions (see Chapter 2). These pollutants
continue to pose threats to human health and the environment as described below.
1.3.1 Fin e Particles
Emissions of SO2 and NOX lead to the formation of fine particles (particles less than 2.5 micrometers).
Most fine particles, namely sulfates and nitrates, are formed when SO2, NOX and volatile organic
compounds and ammonia react in the air. Fine particles can travel for long distances in the air before
Title IV of the Clean Air Act Amendments of 1990, Acid Deposition Control, sets a goal of reducing annual SD2 emissions by ID million tons below I9SD levels and a 2 million ton annual
reduction in NDx emissions by the year 2DDD from fossil fuel-fired power plants. The program affects existing utility units serving generators with an output capacity of greater than 25
megawatts and all new utility units. To achieve these reductions, the law required a two-phase tightening of the restrictions. Phase I was from 1995-1999 and affected 2G3 units at IID
mostly coal-burning electric utility plants located in 21 eastern and midwestern states. An additional IS2 units joined Phase I of the program as substitution or compensating units,
bringing the total of Phase I affected units to 445. Phase II, which began in the year 2DDD, tightened the annual emissions limits imposed on these large, higher emitting plants and also
set restrictions on smaller, cleaner plants fired by coal, oil, and gas, encompassing over 2.DDD units in all.
1-2
-------
being deposited far away from where the emissions took place. Fine particles pose serious threats to
public health because their size allows them to easily reach the most sensitive parts of the lungs. Scientific
studies have linked fine particles (alone or in combination with other air pollutants) with a series of
significant health problems, including premature death.
1.3.2 Ground-Level Ozone
Ozone is formed when emissions of NOX and volatile organic compounds (VOCs) react in the presence of
sunlight. Ground-level ozone is the primary component of smog and tends to be a problem over broad
regional areas, particularly in the eastern United States and some urban areas including Los Angeles and
Houston. Ozone can damage lung tissue, reduce lung function, and adversely sensitize the lungs to other
irritants. Long-term exposures to ozone can cause repeated inflammation of the lung, impairment of lung
defense mechanisms, and irreversible changes in lung structure, which can lead to premature aging of the
lungs or chronic respiratory illnesses such as emphysema and chronic bronchitis. Ozone can also
aggravate asthma, causing more frequent and severe asthma attacks.
1.3.3 A cid Deposition
Acid deposition results from the reaction of SO2 and NOX air-borne emissions with water, oxygen, and
oxidants to form various acidic compounds. It occurs both as wet deposition through acidic rain, fog, and
snow and as dry deposition through acidic gases and particles. Acid deposition lowers the pH of sensitive
lakes and streams to toxic levels. Lower pH can impact entire ecosystems through changes in the food
chain. Acid deposition causes release of aluminum from soil, which is also very toxic to fish. It can
directly affect forest ecosystems by damaging plant tissue involving leaching of foliar calcium2 in certain
plants. In other cases, multiple pollutants such as ozone, SO2, and NOX can combine to weaken trees and
make them vulnerable to other threats, such as pests, that cause mortality.
Atmospheric deposition is a rapidly growing anthropogenic source of biologically available nitrogen in
estuarine and near-coastal ecosystems. Depending on the location, from 10 to 80 percent of new nitrogen
inputs to coastal waters along the east and Gulf coasts of the United States are of atmospheric origin.4
This nitrogen contributes to eutrophication of these waters, which results in one or more undesirable
ecological impacts. These impacts include algae blooms, massive die-offs of estuarine and marine plants
and animals (including fish), loss of biological diversity, and degradation or loss of essential coastal
ecosystem habitat (such as sea grass beds). These ecological changes impact commercial and recreational
fisheries and reduce our ability to use and enjoy coastal ecosystems. Recent studies also link atmospheric
nitrogen deposition, coastal eutrophication, and harmful algal blooms with human health impacts.5"8
1.3.4 Visibility
Visibility impairment is another undesirable by-product of air pollution. Visibility is generally defined as
the degree to which the atmosphere is transparent to visible light; visibility impairment is a reduction in
visual range and atmospheric discoloration. Regional haze obscures the clarity, color, texture, and form of
what is seen. The same fine particles that adversely affect human health are the primary cause of regional
haze. Visibility is of more concern in national parks and wilderness areas.
Leaching of foliar calcium from the needles of red spruce, which reduces the cold tolerance of individual trees, has contributed to the decline of high-altitude red spruce forests
throughout eastern North America. Recent studies indicate that foliar calcium loss may occur with other species as well.
1-3
-------
7.5.5 Mercury
Mercury cycles in the environment as a result of natural and human activities. Most anthropogenic
mercury emissions to the air come from stationary combustion sources, including waste and fossil fuel
combustion. About 64 percent (48 tons) of the mercury in the coal burned in power generation plants in
1999 was emitted to the atmosphere.9 The percent contribution to total anthropogenic mercury emissions
from power generation is expected to become even greater as coal-fired generation increases and the final
mercury standards for municipal waste combustors and medical and hazardous waste incinerators are
fully implemented. Current releases are adding to the mercury that already exists in land, water, and air,
both naturally and as a result of previous human activities.
Mercury in combustors exists in elemental, ionic, or particulate forms. Elemental mercury can be widely
dispersed and transported for many miles, whereas ionic mercury, which is soluble in aqueous solutions,
and particulate mercury are typically deposited close to the emissions source. Once mercury is deposited
in lakes, rivers, and oceans, it is converted to methylmercury by aquatic organisms and bioaccumulates in
the food chain, resulting in high concentrations in predatory fish. In the United States, most human
exposure to mercury is the result of consumption offish contaminated with methylmercury. Adverse
effects of mercury on fish, birds and mammals include death, reduced reproductive success, impaired
growth and development, and behavioral abnormalities.
While EPA has already taken action to reduce mercury emissions from other major sources such as
municipal waste combustors, medical waste incinerators and hazardous waste combustors, coal-fired
power plants' mercury emissions into the air remain uncontrolled. The EPA took the first step towards
controlling mercury emissions from the power generation sector on December 14, 2000 by issuing a
finding that mercury emissions from the coal-fired power plants should be controlled. Based on this
determination, the agency proposed the "Utility Mercury Reductions Rule" in December 2003, which
calls for significant reductions of mercury from the utility coal-fired power plants.
1.4 Multi-emission Control Options
This report addresses multi-emission control technologies that have reached a stage of development
beyond pilot scale. It includes those technologies that integrate in-situ or post-combustion controls of at
least two of the SO2, NOX, and mercury pollutants, either in one process or a combination of coordinated
and complementary processes. In addition, the report includes new coal-fired electricity-generating
technologies, which are inherently more efficient than conventional coal-fired power plants or have the
potential to generate lower emission of air pollutants and CO2. The report also looks at power plant
improvements capable of reducing emissions and producing higher energy efficiencies.
It is envisioned that the report will assist the electric power industry in selecting emission control options
for power plants. Information on environmental performance, cost, commercialization barriers, and other
relevant factors on these technologies could enable power companies to choose less expensive air
pollution compliance approaches, compared to controlling individual emissions through a series of
controls in succession, each targeting a different pollutant. In certain cases, such as in a constrained plant
layout, multi-emission control technologies may represent the most practical method of providing the
necessary environmental benefits.
This report is limited to coal-fired power plant emission controls and power generation technologies and
does not include consideration of other power generating options such as systems based on renewable
energy sources. The authors do recognize that these latter options do offer tremendous potential in
reducing air pollution, and suggest they be addressed under separate efforts.
1-4
-------
1.5 References
1. Energy Information Agency. International Energy Outlook, U.S. Greenhouse Gas Intensity
Target. http://tonto.eia.doe.gov/FTPROOT/forecasting/0484(2003Vpdf. (accessed Dec 1, 2004).
2. Overview of the Electric Power Industry, Energy Information Agency, U.S. Department of
Energy: Washington, DC, Sept 2000 (1998 survey data).
3. Bush, G. W. Remarks by the President on National Energy Policy in Photo Opportunity with
Cabinet Members: 2001; ISBN 0-16-050814-2; Report of the National Energy Policy
Development Group, U.S. Government Printing Office: Washington, DC, May 16, 2001.
http://www.whitehouse.gov/energy/ (accessed Dec 1, 2004).
4. Valigura, R. A.; Alexander, R. B.; Castro, M. S.; Meyers, T. P.; Paerl, H. W.; Stacey, P. E.;
Turner, R. E. Nitrogen Loading in Coastal Water Bodies: An Atmospheric Perspective; American
Geophysical Union, 2000.
5. Paerl, H. W.; Whitall, D. R. Anthropogenically-derived Atmospheric Nitrogen Deposition,
Marine Eutrophication and Harmful Algal Bloom Expansion: Is There a Link? Ambio 1999, 28.
6. Epstein, P. R. Positive Incentives for an Energy Transition. In: Global Environment Facility
(GEE), Valuing the Global Environment: Actions and Investments for a 21st Century; Global
Environmental Facility: Washington, DC, 1998; 98-99.
7. Burkholder, J. M.; Glasgow, H. B., Jr. Pfesteria Piscicida and Other Pfesteria-like
Dinoflagellates: Behavior, Impacts and Environmental Controls. Limnol. Oceanogr. 1997, 42,
1052-1075.
8. Paerl, H. W. Coastal Eutrophication and Harmful Algal Blooms: Importance of Atmospheric
Deposition and Groundwater as "New" Nitrogen and Other Nutrient Sources. Limnol. Oceanogr.
1991,42, 1154-1165.
9. Kilgroe, J. D.; Sedman, C. B.; Srivastava, R. K.; Ryan, J. V.; Lee, C.W., Thorneloe, S. A. Control
of Mercury Emissions from Coal-Fired Electric Utility Boilers, Interim Report; EPA-600/R-01-
109 (NTIS PB2002-105701); National Risk Management Research Laboratory: Research
Triangle Park, NC, Dec 2001.
1-5
-------
1-6
-------
Chapter 2
Characterization of U.S. Power Plants and Associated Emissions
As a preview to discussing multi-emission control technologies, it is beneficial to characterize the
magnitude and sources of the emissions that drive the need for these technologies. Sulfur dioxide (SO2),
nitrogen oxides (NOX), and mercury are generated not only from electricity generation but also from a
number of sectors of the economy. Figure 2-1 presents the 2001 estimates of SO2 and NOX and 1998
estimates of mercury emissions from major economic sectors in the United States.1 During 2001, fuel
combustion-electric utilities contributed 69 percent of the total SO2 emitted and 22 percent of the NOX.
2001 SO9 Emissions
2001 NOY Emissions
18 %|
22%
55% \
69%
17%
1998 Mercury Emissions
5%
14% /
35%
42%
LEGEND
Other
stationary
combustion
Fuel
combustion-
electric utility
Miscellaneous
YV /Transportation
Industrial
processing
Figure 2-1. Major sources of SO2 and NOX emissions in 2001 and of mercury emissions in 1998.
2-1
-------
Since the majority of U.S. power plant capacity is regulated under Title IV of the CAAA,2'3 the data for
NOx, SC>2, and CC>2 collected under this title provide a good representation of these emissions. Title IV of
the Clean Air Act covers all boiler units with capacity greater than 25 MW, excluding a small percentage
of units classified as co-generating units. Title IV affected a total of 2,638 units in 2001, from which
1,089 are considered coal-fired units based on burning over 50 percent coal as the primary fuel. Emissions
data for Title IV affected units is available at EPA Clean Air Market Division's website.4
Table 2-1 summarizes the 2001 total emissions data for all Title IV affected units, apportioned by two
broad primary fossil fuel type categories (coal and non-coal). Non-coal units affected by Title IV include
those that burn liquid or gaseous fossil fuel (oil, diesel, natural gas, etc.) or other solid fuel, such as wood,
as their primary fuel. The data reflect the predominance of coal use by U.S. facilities. Of the totals in
2001, coal accounts for 95 percent of SO2 emissions, 90 percent of NOX emissions, and 86 percent of CO2
emissions. Note that data on mercury emissions is not collected under Title IV. Coal units also account
for 79 percent of the total heat input.
Table 2-1. Emissions in 2001 from Coal and Non-coal Units5
SO2 (Tons)
NOX (Tons)
CO2 (Tons)
Coal
10,052,279
4,235,104
2,049,776,430
Non-Coal
581,799
464,770
344,080,879
Total
10,634,077
4,699,874
2,393,857,309
However, in 1999, the EPA conducted an Information Collection Request (ICR)6 to obtain information on
mercury emissions from the U.S. coal-fired electric utility steam generating facilities. The ICR collected
general information and data on quantity of fuel consumed, quantity of mercury of that fuel, and mercury
speciation in flue gas before and after air pollution control devices upstream of the stack. This ICR
identified 1140 boilers of 25 MW or greater capacity including cogenerators that supplied more than one-
third of their potential electric output capacity to utility distribution systems for sale. The ICR effort
further identified the breakdown of these units as 979 pulverized coal-fired, 87 cyclone-fired, 42
fluidized-bed combustors, and 32 stoker-fired boilers. The information on mercury emissions presented in
this report is based on EPA's 1999 ICR.
2.1 SC>2 Emissions from Coal-Fired Power Plants
SO2 emissions from coal-fired power plants increased slightly between 1996 and 1997 by about 400,000
tons. Despite this emissions increase, a sharp decrease of about 2,650,000 tons or 21 percent occurred
between 1997 and 2001. Sulfur dioxide emissions from 1996 to 2001 are illustrated in Figure 2-2.:
2-2
-------
1996
1997
1998 1999 2000 2001
Figure 2-2. Annual SO2 emissions for coal-fired units.
Emissions of SO2 are highly dependent on the sulfur content in the coal burned and the emissions control
system employed. Emissions control systems are typically based on dry or wet flue gas desulfurization
(FGD). FGD technologies are based on the use of sorbents to scrub SO2 from the flue gas and are
conventionally classified as throwaway and regenerable depending on how the sorbent is treated, and
further into wet or dry FGDs. In wet processes, wet slurry waste or by-product is produced, and flue gas
leaving the absorber is saturated with moisture. In dry processes, dry waste material is produced and flue
gas leaving the absorber is not saturated.
SO2 emission rates for 60 units that had FGD technologies installed between 1990 and 1999 are presented
in Table 2-2.7"9 Wet limestone and wet lime FGDs were the two most installed control technologies for
these units. Emission rates varied widely among the different technologies and coal types.
2-3
-------
Table 2-2. SO2 Emissions from Units with FGD Installed in 1990-1999
FGD Type
Dry Lime
Dry Lime
Dry Lime
Dual Alkali
Dual Alkali
Sodium
Based
Sodium
Based
Wet Lime
Wet Lime
Wet Lime
Wet Lime
Wet Lime
Wet
Limestone
Wet
Limestone
Wet
Limestone
Wet
Limestone
Wet
Limestone
Wet
Limestone
Wet
Limestone
Wet
Limestone
Number of
Units3
2
4
8
2
1
8
1
24
4
2
4
8
50
9
5
3
1
5
3
23
Fuels Burnedb
Bit.
Lignite
Subbit.
Bit.
Lignite
Bit.
Subbit.
Bit.
Bit./petcoke
Bit./subbit.
Lignite
Subbit.
Bit.
Bit./petcoke
Bit./subbit.
Bit./subpetcoke
Bit/tires
Lignite
Lignite/subbit.
Subbit.
Average SO2
Emission
(Ib/MMBtu)
0.32
0.61
0.28
0.65
1.02
0.70
0.43
0.55
0.47
0.39
0.89
0.42
0.59
0.45
0.53
0.45
0.24
0.89
1.06
0.39
Maximum SO2
Emission
(Ib/MMBtu)
0.47
1.09
0.61
0.84
1.02
1.93
0.43
1.12
0.53
0.57
1.15
0.51
3.02
0.73
0.89
0.57
0.24
1.14
1.16
0.76
Minimum SC>2
Emission
(Ib/MMBtu)
0.17
0.40
0.09
0.46
1.02
0.05
0.43
0.08
0.37
0.20
0.35
0.14
0.05
0.05
0.22
0.22
0.24
0.56
0.87
0.08
Annual
SO2
Emission
(tons)
1,848
36,818
36,477
11,253
21,863
18,878
1,858
163,936
30,912
8,896
78,118
49,743
369,087
65,006
50,883
16,320
978
120,824
101,745
178,358
FGD units under construction in 1999, or known not to be operating, were not included in the unit counts.
This included 3 wet limestone, 2 dry lime FGD, and 1 dry limestone FGD.
Data from EPA's 2001 ICR; bit. = bituminous coal; subbit. = subbituminous coal; petcoke = petroleum coke.
2.2 NOx Emissions from Coal-Fired Power Plants
Annual NOX emissions for coal-fired units regulated under Title IV are shown in Figure 2-3.: These
emissions steadily declined from 1997 to 2001 by about 1,470,000 tons.
2-4
-------
UJ
1996
1997
1998
1999
2000
2001
Figure 2-3. Annual NOX emissions for coal-fired units.
NOX is formed during most combustion processes by one or more of three chemical mechanisms: (1)
"thermal" NOX resulting from oxidation of atmospheric molecular nitrogen, (2) "fuel" NOX resulting
from oxidation of chemically bound nitrogen in the fuel, and (3) "prompt" NOX resulting from reaction
between atmospheric molecular nitrogen and hydrocarbon radicals. In fuel-lean combustion of nitrogen-
free fuels, thermal NOX is the primary component of NOX emissions. Thermal NOX formation is quite
sensitive to temperature and can be controlled by appropriately controlling peak temperature in the
furnace. In fuel-lean combustion of fuels containing nitrogen, fuel NOX contributes significantly to total
NOX emissions, depending on the percentage of nitrogen in the fuel. Formation of fuel NOX depends on
the availability of oxygen to react with the nitrogen during coal devolatilization and the initial stages of
combustion. Under fuel-rich conditions, the formation of NOX may compete with the formation of
molecular nitrogen (N2) and may result in a reduction of NOX emissions. Prompt NOX contributes a
relatively minor fraction of total NOX emissions for both nitrogen-free and nitrogen-containing fuels.
The factors described above that dictate NOX formation (devolatilization of fuel-bound nitrogen, oxygen
concentration, and flame temperature) can all be adjusted by controlling the rate at which the fuel and air
mix. This allows staging the combustion process, such that an initial fuel-rich zone is followed by a
burnout zone that is high enough in oxygen to complete the combustion process, but low enough in
temperature to minimize thermal NOX production. Combustion modification NOX controls utilize this
combustion staging.
In general, NOX control technologies are categorized as being either primary control technologies or
secondary control technologies. Primary control technologies reduce the formation of NOX in the primary
combustion zone. In contrast, secondary control technologies destroy the NOX present in the flue gas from
the primary combustion zone. Primary control technologies being used in the United States are low NOX
burner (LNB) and overfire air (OFA).
The secondary NOX control technologies in use on U.S. coal-fired utility boilers include reburning,
selective non-catalytic reduction (SNCR), and selective catalytic reduction (SCR). More than 100 boilers
either have used, or will use, these technologies to achieve the desired NOX reductions. The NOX
2-5
-------
reductions achieved, or projected, at these applications range from 20 to more than 80 percent. Table 2-3
summarizes the 1999 NOX emissions and emission rates for units regulated under Title IV.10
Table 2-3. 1999 NOX Emissions for Title IV Affected Units
Boiler Type
(Coal Units Only)
Arch-fired
Cell burner
Circulating
fluidized bed
Cyclone
Dry bottom turbo-
fired
Dry bottom
vertically-fired
Number
of Units
9
2
3
1
1
29
2
4
1
1
1
1
1
84
2
3
2
5
2
3
21
NOX Control
Uncontrolled
Low NOX Burner Technology
Low NOX Burner Technology for
Cell Burners
Low NOX Burner Technology
with Overfire Air (Dry Bottom
Boilers Only)
Other
Uncontrolled
Other
Uncontrolled
Combustion Modification with
Fuel Reburning
Other
Overfire Air
Selective Catalytic Reduction
Selective Non-catalytic
Reduction
Uncontrolled
Low NOX Burner Technology
Low NOX Burner Technology
with Overfire Air (Dry Bottom
Boilers Only)
Overfire Air
Uncontrolled
Low NOX Burner Technology
Overfire Air
Uncontrolled
Average NOX
Emission
(Ib/MMBtu)
0.74
0.62
0.46
0.39
0.61
0.75
0.21
0.37
0.77
0.93
1.12
0.47
0.69
0.96
0.39
0.59
0.43
0.41
0.48
0.51
0.77
Maximum
NOx Emission
(Ib/MMBtu)
0.91
0.71
0.46
0.39
0.61
1.41
0.22
0.47
0.77
0.93
1.12
0.47
0.69
1.81
0.40
0.62
0.44
0.43
0.73
0.51
1.16
Minimum NOX
Emission
(Ib/MMBtu)
0.25
0.52
0.46
0.39
0.61
0.43
0.20
0.19
0.77
0.93
1.12
0.47
0.69
0.26
0.37
0.54
0.42
0.37
0.22
0.51
0.37
Annual NOX
Emission
(tons)
11,170
39,125
20,067
8,089
13,044
464,518
2,536
1,368
2,831
3,259
23,579
4,627
3,226
708,760
16,110
18,078
13,128
18,813
4,316
3,855
64,234
(continued)
2-6
-------
Table 2-3 (concluded)
Boiler Type
(Coal Units Only)
Dry bottom wall-
fired
Stoker (coal and
wood)
Tangentially-fired
Wet bottom turbo-
fired
Wet bottom
vertically-fired
Wet bottom wall-
fired boiler
Number
of Units
153
41
1
3
4
5
213
15
5
45
31
40
26
22
10
2
254
1
2
2
2
3
1
2
18
NOX Control
Low NOX Burner Technology
Low NOX Burner Technology
with Overfire Air (Dry Bottom
Boilers Only)
Low NOX Burner Technology
with Overfire Air & Selective
Catalytic Reduction
Other
Overfire Air
Selective Non-catalytic
Reduction
Uncontrolled
Uncontrolled
Combustion Modification with
Fuel Reburning
Low NOX Burner Technology
Low NOX Burner Technology
with Close-coupled and
Separated OFA
Low NOX Burner Technology
with Close-coupled OFA
Low NOX Burner Technology
with Separated OFA
Other
Overfire Air
Selective Non-catalytic
Reduction
Uncontrolled
Low NOX Burner Technology
Overfire Air
Uncontrolled
Uncontrolled
Low NOX Burner Technology
Other
Selective Non-catalytic
Reduction
Uncontrolled
Average NOX
Emission
(Ib/MMBtu)
0.43
0.45
0.17
0.36
0.46
0.36
0.57
0.40
0.42
0.34
0.36
0.37
0.38
0.41
0.34
0.39
0.43
0.53
0.46
0.93
0.90
0.37
0.45
0.79
0.73
Maximum
NOx Emission
(Ib/MMBtu)
0.81
0.71
0.17
0.37
0.51
0.40
1.11
0.57
0.44
0.45
0.63
0.67
0.71
0.72
0.49
0.47
0.86
0.53
0.47
1.13
0.93
0.47
0.45
0.82
0.86
Minimum NOX
Emission
(Ib/MMBtu)
0.18
0.30
0.17
0.36
0.37
0.30
0.20
0.32
0.37
0.13
0.16
0.17
0.18
0.23
0.23
0.30
0.24
0.53
0.45
0.73
0.86
0.32
0.45
0.76
0.43
Annual NOX
Emission
(tons)
848,965
182,191
1,641
16,083
25,737
4,003
539,907
2,008
6,893
202,120
146,046
181,774
104,590
163,214
70,431
2,795
800,505
6,242
5,606
15,496
456
23,852
10,351
12,759
80,549
2-7
-------
2.3 Mercury Emissions from Coal-Fired Power Plants
Coal-fired power plants represent a significant source of mercury emissions into the atmosphere. EPA has
estimated that a total of 48 tons of mercury was emitted from coal-fired power plants in the United States
in 1999.n Another analysis has estimated the total amount of mercury emissions from U.S. coal-fired
plants to be 45 tons in 1999. This weight consists of 18 tons of oxidized mercury, 26 tons of elementary
mercury (Hg°), and less than 1 ton of particulate-bound mercury.12 This latter analysis also estimates the
total amount of mercury in the fuel entering these power plants at 75 tons in 1999.
The air pollution control technologies now used on pulverized-coal-fired utility boilers exhibit average
levels of Hg control that range from 0 to 98 percent, as shown in Table 2-4.4 The best levels of control are
generally obtained by emission control systems that use fabric filters (FF). The amount of Hg captured by
a given control technology is better for bituminous coal than for either subbituminous coal or lignite.
The lower levels of Hg capture in plants firing subbituminous coal and lignite are attributed to low fly ash
carbon content and the higher relative amounts of Hg° in the flue gas from combustion of these fuels. The
average capture of Hg based on inlet measurements in pulverized coal (PC)-fired plants equipped with a
cold-side electrostatic precipitator (ESP) is 35 percent for bituminous coal, 3 percent for sub-bituminous
coal, and near zero for lignite.
Plants that employ only post-combustion PM controls display average Hg emission reductions ranging
from 0 to 89 percent. The highest levels of control were observed for units with FFs. Lower levels of
control were shown for units with ESPs and other controls.
Table 2-4. Mean Mercury Emission Reduction for Pulverized-coal-fired Boilers
Post-Combustion Emission
Controls
Used for PC Boiler
PM Control
Only
PM Control
and
Spray Dryer
Adsorber
PM Control
and
Wet FGD
System
ESP-CSb
ESP-HS
FF
PS
SDAandESP
SDA and FF
SDA, FF, and SCR
ESP-CS and FGD
ESP-HS and FGD
FF and FGD
Average Mercury Emission Reduction (%) a
Bituminous
Coal-fired
35
14
89
12
Not tested
98
97
81
45
97
Subbituminous
Coal-fired
3
12
73
0
50
23
Not tested
30
25
Not tested
Lignite
Fired
0
Not tested
Not tested
33
Not tested
17
Not tested
42
Not tested
Not tested
' Mean reduction from 3-run test averages for each PC boiler unit in ICR Phase III database.
1 Refer to Acronyms, page vii for definitions.
2-8
-------
Units equipped with lime spray dryer absorber scrubbers and with ESP or FF 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 results from low levels of Hg° oxidization.
The capture of Hg in units equipped with wet FGD scrubbers is dependent on the relative amount of Hg2+
in the inlet flue gas and on the PM control technology used. Average Hg captures in wet FGD scrubbers
ranged from 23 percent for one PC-fired hot-side ESP (ESP-HS) and FGD unit burning sub-bituminous
coal to 97 percent in a PC-fired FF and FGD unit burning bituminous coal. The high Hg capture in the FF
and FGD unit is attributed to increased oxidization and capture of Hg in the FF.
Mercury captures in PC-fired units equipped with spray dry scrubbers and wet limestone scrubbers appear
to provide similar levels of control on a percentage reduction basis. However, this observation is based on
a small number of short-term tests at a limited number of facilities. Additional testing will be required to
characterize the effects of fuel, combustion conditions, and air pollution control device (APCD)
conditions on the speciation and capture of Hg.
2.4 CO2 Emissions from Coal-Fired Power Plants
Carbon dioxide emissions from the combustion of fossil fuels are mainly driven by the source of energy
and its carbon content. Sources of energy include natural gas, diesel oil, biomass, and coal. The amount of
carbon in fuels varies significantly by fuel type, and coal contains the highest amount of carbon per unit
of useful energy, resulting in the highest rate of CO2 emissions per kilowatt-hour of electricity.
Title IV does not require control of CO2 emissions; it only requires that they be measured and reported.
Figure 2-4 summarizes emissions from these units for 1996 to 2001.4 Emissions of CO2 from all Title IV
affected units increased by 2.5 percent from 1996 to 1997. In 1998, CO2 emissions increased by 1.5
percent. The CO2 emissions remained the same between 1998 and 1999, but in 2000, CO2 emissions rose
by 1.9 percent. In 2001, CO2 emissions fell below the 1997 level.
2.15
2.13
1.95
1996 1997 1998 1999 2000 2001
Figure 2-4. Annual CO2 emissions for coal-fired units.
2-9
-------
2.5 References
1. National Air Quality and Emissions Trends Report. U.S. Environmental Protection Agency,
Office of Air Quality Planning and Standards, http://www.epa.gov/air/airtrends/aqtrnd03/.
(accessed Dec 1, 2004)
2. Acid Rain Program: Nitrogen Oxides Emission Reduction Program, Direct Final Rule. Fed.
Regist.l995,60(7l).
3. Acid Rain Program: Nitrogen Oxides Emission Reduction Program: Final Rule. Fed Regist.
1996, 61(245).
4. Emissions Scorecard for Acid Rain Program Units, Including SO2, NOX, CO2, and Heat Input,
2001; U.S. Environmental Protection Agency, Clean Air Markets Division: Washington, DC.
http://www.epa.gov/airmarkets/emissions/score01/index.html (accessed Dec 1, 2004).
5. National Totals ofSO2, NOX, CO2, and Heat Input for Coal Fired and Non-Coal Fired Title IV
Affected Units from 1996 through 2001. http://www.epa.gov/airmarkets/emissions/score01/
tablel.pdf (accessed Dec 1, 2004).
6. Electric Utility Steam Generating Unit Mercury Emissions Information Collection Effort; OMB
Control No. 2060-0396; U.S. Environmental Protection Agency, Office of Air Quality Planning
and Standards: Research Triangle Park, NC, April 2001.
7. Hg ICR Part II, Coal Analysis Results; U.S. Environmental Protection Agency, Emissions
Standards Division: Washington, DC, June 2001. http://utility.rti.org/backgroundinfo.cfrn
(accessed Dec 1 , 2004).
8. Electric Power Annual Report 1999, Volume 2; DOE/EIA-0348 (99)/2; U.S. Energy Information
Administration: Washington, DC, Oct 2000. http://www.eia.doe.gov/cneaf/electricity/epav2/
epav2.pdf (accessed Dec 1, 2004).
9. Annual Steam-Electric Plant Operation and Design Data; EIA-767 Form Data File; U.S. Energy
Information Administration: Washington, DC, 1999. http://www.eia.doe.gov/cneaf/electricity/
page/eia767.html (accessed Dec 1, 2004).
10. 7999 Emissions Scorecard; U.S. Environmental Protection Agency, Clean Air Markets Division:
Washington, DC, 1999; Appendix B, Table Bl.
http://www.epa.gov/airmarkets/emissions/score01/ (accessed Dec 1, 2004).
11. Kilgroe, J. Control of Mercury Emissions from Coal-Fired Utility Boilers. Presented at the
MACT Working Group Meeting, Washington, DC, Aug 1, 2001. http://www.epa.gov/ttn/
atw/combust/utiltox/utoxpg.html#MAC (accessed Dec 1, 2004).
12. Chu, P.; Behrens, G; Laudal, D. Estimating Total and Speciated Mercury Emissions from U.S.
Coal Fired Power Plants. Proceedings of the U.S. EPA-DOE-EPRI Combined Power Plant Air
Pollutant Control Symposium: The MEGA Symposium and The A&WMA Specialty Conference
on Mercury Emissions: Fate, Effects, and Control, Chicago, IL, Aug 20-23, 2001.
2-10
-------
Chapter 3
Multi-emission Control Technologies and
Options for Coal-fired Power Plants
3.1 Introduction
This section provides a summary description of the technologies identified as multi-emission control
technologies and which have reached a stage of development beyond pilot scale. The technologies
selected satisfy the following definition established for this report: Multi-emission control technologies
are defined as options which integrate pre-combustion, in-situ or post-combustion controls of at least two
of the SO 2, NOX>, and mercury pollutants, either in one process or a combination of coordinated and
complementary (synergistic) processes.
There were no constraints set as to the percentage reduction of each emission, such as: there should be a
significant reduction of more than 80 or 90 percent. As a result, options such as advanced power
generation technologies, power plant rehabilitation or upgrading, fuel switching or blending and power
plant optimization were included in the multi-emission control category. These technologies were
included to emphasize that there are multiple options a power company may use to control multiple
pollutants in addition to the post-combustion control options, which are the focus in most cases. For
example, utilizing a high plant efficiency technology, such as supercritical pulverized coal firing, is
beneficial for all emissions (compared to the conventional subcritical pulverized coal technology), even
though, by itself, it does not reduce emissions by more than 10-12 percent.
In addition to the above definition, two other criteria were applied in selecting the technologies described
in this report. These were: (1) there should be at least one installation in operation in a power plant
worldwide as of July 1, 2001, and (2) while it is acceptable for the technology to be used even in a slip-
stream (not the entire power plant), the size of the technology installation should be at least 5 MW or
equivalent.
Using the above definition and criteria, a literature search was done (using technical papers from
conferences, the internet, technical reports by organizations such as DOE, EPA, and EPRI, and contacting
vendors and utilities) and the technologies presented in this section were identified. The technologies
were grouped into:
• environmental control options (in-situ and post-combustion controls),
• advanced power generation options, and
• power plant upgrading and operating options.
Environmental controls include processes that control SO2-mercury, SO2-NOX, and SO2-NOx-mercury
emissions. Injection of activated carbon in front of electrostatic precipitators (ESPs) and fabric filters
(FFs) was also included, because of the significant role it may play in controlling mercury from existing
power plants. Activated carbon can also be combined with other technologies, such as sorbent injection,
to provide multiple emissions control.
Advanced power generation technologies include circulating fluidized bed combustion (CFB), integrated
gasification combined cycle (IGCC), pressurized fluidized bed combustion (PFBC) and supercritical
pulverized coal. In the category of plant upgrading and operating changes, the report includes: fuel
3-1
-------
blending or cofiring, plant upgrading and efficiency improvements, and optimization. Some of these
options require extensive plant modifications, such as boiler conversion from coal to natural gas or
repowering of an existing coal-fired power unit to combined cycle unit burning natural gas. Others, such
as optimization, are strictly plant-operating adjustments. Most of these options are very site-specific,
especially with regard to the potential emission reduction and the required costs. As such, this report
generically describes plant upgrading to raise the awareness about its potential and provides general
guidelines with regard to its potential. More site-specific analyses are needed (beyond the scope of this
document) to develop more accurate estimates.
A number of technologies did not satisfy the above definition and selection criteria. Many of these are
under investigation at the laboratory or pilot scale. For others, adequate information is not available.
Examples of such technologies include: Pioneer Technologies' Non-thermal Plasma Arc technology,
Consummator's plasma arc by-product recovery, Phoenix's retrofit slagging combustor, ISCA's C12
injection, and BioDeNOx, which uses bacteria to reduce NOX emissions in FGDs by adding FeEDTA and
ethanol. These, and other technologies, may emerge in a relatively short period of time depending on the
results of on-going laboratory and pilot tests, as well as the demand for multi-emission controls. Others
may never reach commercial stage either because of technical problems they will face or because of
unfavorable economics relative to competing options. Finally, there are technologies which were
demonstrated in the past and for which recent information is not available. These technologies, such as
LIMB, NOXSO, or Milliken Clean Coal, are not discussed in this report.
For each technology, the following information is provided wherever possible:
• brief technology description,
• commercial readiness and industry experience,
• emission control performance,
• O&M impacts,
• costs,
• issues associated with the technology, and
• references.
The description of each technology included in this section is brief, but references are provided if the
reader is interested in more detailed information. Performance and cost of selected technologies for
mercury and multipollutant control have been discussed in more detail elsewhere.1
A significant effort was made to obtain costs, both capital and O&M. Cost ranges were provided for
technologies that have reached maturity because significant information is available. If no adequate
information was available in the literature, cost estimates that are provided are based on the authors'
experience.
The costs reflect End of Year (EOY) 2000 US dollars, unless otherwise indicated. When the basis is
different from EOY2000, the costs are adjusted using the Chemical Engineering Annual Plant Index
(CEI). Considering that the cost data are based on literature reviews and other publicly available
information, it should be noted that the level of uncertainty is of the order of-30 to +80 percent, as noted
in EPRI's Technical Assessment Guide (EPRI TAG) when referring to the accuracy of estimates based on
development stage of the technology and the design and cost basis.
3-2
-------
In addition to the costs, an attempt is made in the report to provide also a size (in MW) to which the costs
correspond. No scale-up correlations are provided (costs vs plant size), because some of the technologies
may not follow the standard scale-up rules. Furthermore, there are other factors that may affect costs
equally or more than plant size. Some of these factors include coal composition and emission reduction
requirements.
References
1. Staudt, J.E. and Jozewicz, W., Performance and Cost of Mercury and Multipollutant Emission
Control Technology Applications on Electric Utility Boilers, EPA-600/R-025-2003, U.S.
Environmental Protection Agency, Research Triangle Park, NC, August 6, 2003.
3.2 Environmental Control Options
3.2.1 SO2 an d Mercury Control
3.2.1.1 Dry Scrubbers
Dry scrubbing technology is a flue gas desulfurization (FGD) method capable of reducing multiple
pollutants (specifically SO2 and mercury) and is typically used in low- to medium-sulfur coal-fired power
plants. "Dry" refers to the fact that the flue gas and the refuse (ash and solid reaction products) leaving the
scrubber are not saturated with water vapor as in wet FGD (wet scrubbers). The technology is suitable for
new and retrofit applications. Dry scrubbers have been installed on utility and industrial boilers, as well as
hazardous and municipal waste incinerators. In general, dry scrubbers fall into several categories
associated with the type of vessel or reactor used. The following sections describe the most common and
available technologies - conventional spray dryer and fluidized bed.
3.2.1.1.1 Conventional Dry Scrubbers
Conventional Dry FGD (Spray Dryer Absorbers) Summary
Status
SO2 Reduction (%)
NOX Reduction (%)
Hg Reduction (%)
Cost
Capital ($/kW)
Fixed O&M ($/kW-yr)
Variable O&M (mills/kWh)
Applicability
Issues
Commercial
90 to 95
Oto95
150-230 for a 300 MW plant
1.5-7
0.2-0.7
SC>2 and Hg control for low- to-medium-sulfur coals
Hg removal can vary significantly with coal type, operating
conditions, and other pollution control devices present in the
plant
Technology Description
In conventional spray dryer technology, a slurry of alkaline reagent (most often lime slurry or hydrated
lime) is atomized via rotary atomizers or, alternatively, pneumatic nozzles and injected into a vessel
where it reacts with the SO2 in the flue gas to produce calcium sulfate or sulfite products. The vessel must
be appropriately sized to allow sufficient residence time (-10 seconds) for droplet evaporation and SO2
capture to take place.
3-3
-------
In addition, spray dryers have been shown to reduce mercury in the flue gas.1"3 In this case, the more
complex chemistry and speciation forms of the mercury make it more difficult to fully understand the
processes and reactions that take place. However, fly ash, lime and activated carbon-based sorbents all
provide varying degrees of affinity to adsorb vapor-phase mercury present in the flue gas.
The relatively simple designs of the spray dryers include a cylindrical vessel with conical bottom. The
atomizer may be located either at the top or the bottom and spraying into the moving flue gas flow. Flue
gas exiting the spray dryer is directed to a downstream ESP or FF, where the dry material is then either
collected in its entirety for disposal or may be partially re-introduced into the absorber as part of the slurry
mixture to enhance the overall efficiency of process.
This technology has been used predominantly on low- and medium-sulfur coals, but can be applied to
plants burning higher-sulfur coal, too. The reason for this application flexibility is mainly economic;
spray dryers have lower costs and can achieve reasonable SO2 removal for low- and medium-sulfur coal.
For high-sulfur coal, the costs (both capital and O&M) increase substantially and cannot compete easily
with wet FGDs. In all cases, two factors need to be considered. The capacity of the existing particulate
control device (in retrofit situations) needs to be checked to make sure that the increase in particulate
loading can be accommodated (the quantity of sorbent is proportional to the SO2 concentration and the
desired reduction sought, therefore it increases with higher sulfur and reductions required). Also, the
existing ash handling system may not be able to handle the increased amount of ash and adjustments may
be required because of the properties of the ash (unreacted CaO combined with water and moisture
releases heat and may create handling problems).
The quantity of sorbent for a given application is typically referred to in terms of calcium-to-sulfur ratio
or stoichiometry and ranges from about 1.0 to >2.0 for SO2 removals of 75 to 95 percent. For 80 percent
SO2 removal, calcium-to-sulfur ratio is in the lower end of this range. Also, when a baghouse is used
downstream of the spray dryer, it reduces further the required sorbent (lower calcium-to-sulfur ratio), as it
provides more residence time (in some cases up to 10-15 minutes) for the unreacted sorbent to react
further.
Commercial Readiness and Industry Experience
Spray dryers are a commercial and well-established technology in the United States and abroad, with over
11,000 MW and 7,000 MW of installed capacity, respectively4 The technology was first used in the early
1980's, and has been deployed in bituminous, subbituminous, and lignite applications. Significant
experience was gained in the United States through extensive testing programs such as those conducted at
TVA's Shawnee Station, EPRI's Environmental Control Technology Center, Northern States Power's
Riverside Station, and B&W's Alliance Research Center.5
Emission Control Performance
Spray dryers are capable of very high SO2 reductions (up to 95 percent). Data from lEA's coal research
indicate SO2 reductions from 70 to 96 percent with a median value of 90 percent, comparable to that of
wet FGD technology. This performance reflects applications with coals less than 2 percent sulfur.4 Most
U.S. applications of dry scrubbing (spray dryer or baghouse) have been for western subbituminous coal
with sulfur contents of <1 percent.
Information and experience with mercury is less available than for SO2. However, a number of test
programs as well as the recent EPA's Mercury Information Collection Request (ICR) program have
yielded some insight into the potential mercury reductions in spray dryers.1"3 It is important to recognize
that the performance of the spray dryers is typically reported together with the associated particulate
control device (ESP or FF). In other words, mercury reductions are reported from the inlet to the spray
3-4
-------
dryer to the outlet of the ESP or FF. In a 1994 study,6 spray dryers were documented to capture mercury
in a wide but not fully understood range (6 - 96 percent) based on data for seven installations on coal-
fired power plants. Although at that time, mercury speciation was not measured, the amount of mercury
removal increased with coal chlorine content suggesting that spray dryers preferentially remove oxidized
mercury. The wide range in the reduction values indicates more research is needed to understand
completely the physical and chemical processes taking place in the control devices.
More recently, EPA's ICRhas documented mercury reductions from spray dryer-ESP-FF systems using
different rank coals (from bituminous to lignite). Figure 3-1 presents the results for spray dryer and ESP-
FF combinations from the ICRtest program. Total mercury reductions across the system are plotted
against chlorine content of the coal (an indicator of the likely fraction of elemental vs oxidized gas-phase
mercury in the flue gas). The large variation in the data is evident. Mercury reduction ranges from a low
of less than 10 to over 95 percent at the high end. Factors that impact overall performance are not yet
fully understood. Table 3-1 presents average reductions for spray dryer and ESP-FF system by coal rank
from the ICR data.
03
>
O
E
CD
01
D)
SDA/ESP
SDA/FF
10 100 1,000 10,000
Coal Chlorine Content, ppm dry
Figure 3-1. Mercury removal across spray dryer and ESP-FF from ICR data.2
Table 3-1. Average Mercury Reductions for Spray Dryer and ESP-FF per Coal Rank from ICR Data
Configuration
SDA-ESP
SDA-FF
Coal Rank
Bituminous
Not available
83%
Sub-Bituminous
53%
22%
Lignite
Not available
25%
3-5
-------
O&MImpacts
O&M impacts include: 1. additional pressure loss associated with the spray dryer vessel and associated
auxiliary power; 2. increased solids loading, which impacts the performance of the downstream
particulate collector (mainly ESP) and ash handling system, and may increase the corrosion of the
particulate collector and downstream ductwork; and 3. stack corrosion due to low approach-to-saturation
temperatures. The magnitude of these impacts is a function of the specific design and operating
conditions, and whether the ESP or FF is new or existing. Also, water usage is increased mainly for
conditioning of the ash.
Capital Costs
Spray dryers represent a significant capital cost addition to a power plant. Costs have been reported in the
150 to 230 $/kW range.7"9 For example in 1996 EPRI's TAG10 provided capital costs for a 300 MW unit
ranging from 154 to 232 $/kW and representing a variety of coals as well as geographical locations in the
United States As with other technologies, prices have tended to decrease as improvements to the
technology have occurred.
O&M Costs
Fixed O&M costs are reported in the 1.5-7 $/kW-yr range11"13 for a range of coal sulfur contents of about
0.4 to 1 percent.
Variable O&M are reported,11 in the range of about 0.2 to 0.7 mills/kWh, for the same sulfur content
range.
Issues Associated with Dry Scrubbing; Future Outlook
The efficiency of mercury removal by dry scrubbers is related to mercury speciation, as well as a number
of other factors. Additional information on mercury speciation and operating parameters in spray dryer-
equipped power plants is necessary to better understand and predict mercury reduction performance.
"Dedicated" mercury sorbents such as activated carbon should increase mercury capture potential.
References
1. Afonso, R. Assessment of Mercury Removal by Existing Air Pollution Control Devices in Full
Scale Power Plants. Proceedings of the U.S. EPA-DOE-EPRI Combined Power Plant Air
Pollutant Control Symposium: The MEGA Symposium and The A&WMA Specialty Conference
on Mercury Emissions: Fate, Effects, and Control, Chicago, IL, Aug 20-23, 2001.
2. Chu, P.; Behrens, G.; Laudal, D. Estimating Total and Speciated Mercury Emissions from US
Coal-fired Power Plants. Proceedings of the U.S. EPA-DOE-EPRI Combined Power Plant Air
Pollutant Control Symposium: The MEGA Symposium and The A&WMA Specialty Conference
on Mercury Emissions: Fate, Effects, and Control, Chicago, IL, Aug 20-23, 2001.
3. Weilert, C.V. Analysis of ICR Data for Mercury Removal from Wet and Dry FGD. Proceedings
of the U.S. EPA-DOE-EPRI Combined Power Plant Air Pollutant Control Symposium: The
MEGA Symposium and The A&WMA Specialty Conference on Mercury Emissions: Fate,
Effects, and Control, Chicago, IL, Aug 20-23, 2001.
4. Jozewicz, W.; Singer, C.; Srivastava, R. K.; Tsirigotis, P. E. Status of SO2 Scrubbing
Technologies. Proceedings of the U.S. EPA/DOE/EPRI Combined Utility Air Pollutant Control
Symposium: The MEGA Symposium, Atlanta, GA, Aug 16-20, 1999.
3-6
-------
5. Guidelines for P articulate Control for Advanced SO 2 Control Processes; EPRI TR-104594; Palo
Alto, CA, Dec 1994.
6. Gleiser, R.; Felsvang, K. Mercury Emission Reduction Using Activated Carbon with Spray Dryer
Flue Gas Desulrurization. Presented at the American Power Conference, Chicago, IL, April 1994.
7. Keeth, R. J; Ireland, P.; Radcliffe, P. Utility Response to Phase I and Phase II Acid Rain
Legislation - An Economic Analyses. Proceedings of the EPRI/DOE/EPA 1995 SO2 Control
Symposium, Miami, FL, 1995.
8. Krause, T.; Rosenquist, W. Environmental Control Technologies for Our Next Generation of
Coal Plants. Proceedings of the U.S. EPA-DOE-EPRI Combined Power Plant Air Pollutant
Control Symposium: The MEGA Symposium and The A&WMA Specialty Conference on
Mercury Emissions: Fate, Effects, and Control, Chicago, IL, Aug 20-23, 2001.
9. Martinelli, R. Dry Scrubber/fabric Filter Retrofit: the Hay den Emissions Control Project.
Proceedings of the US EPA/DOE/EPRI Combined Utility Air Pollutant Control Symposium: The
MEGA Symposium, Atlanta, GA, Aug 16-20, 1999.
10. Ramachadran, G. EPRI Technical Assessment Guide; EPRI Report No. P-6587-L; Palo Alto, CA,
1989; I (rev 6).
11. Weilert, C. V. Statistical Analyses of FGD System Operating Costs: Update for 1993-1997.
Proceedings of the US EPA/DOE/EPRI Combined Utility Air Pollutant Control Symposium: The
MEGA Symposium, Atlanta, GA, Aug 16-20, 1999.
12. Blythe, G. EPRI FGD Operating and Maintenance Survey. Proceedings of the US EPA/DOE/
EPRI Combined Utility Air Pollutant Control Symposium: The MEGA Symposium, Atlanta, GA,
Aug 16-20, 1999.
13. Felsvang, K.; Boscak, V.; Andersen, P. Mercury Reduction Control Options. Proceedings of the
U.S. EPA-DOE-EPRI Combined Power Plant Air Pollutant Control Symposium: The MEGA
Symposium and The A&WMA Specialty Conference on Mercury Emissions: Fate, Effects, and
Control, Chicago, IL, Aug 20-23, 2001.
3-7
-------
3.2.1.1.2 Advanced Dry Scrubbers
Advanced Dry FGD (CFB Absorbers) Summary
Status Pilot to Commercial
SO2 Reduction, % 90 - 98
NOX Reduction, %
Hg Reduction, % 0 to 95
Cost
Capital ($/kW) 50-150 (depending on the type of technology used)
Fixed O&M ($/kW-yr) <1 - 3 (based on 1- 2 percent of capital)
Variable O&M (mills/kWh) 0.2 - 0.7
. .. ..... SO2-Hg control for low to medium sulfur coals (same as
Applicability spray dryers)
Hg removal may vary significantly with coal type, operating
Issues conditions, and other pollution control devices present in the
plant (similar to spray dryers)
Technology Description
Advances in dry scrubbing technology have focused on the general concept of increasing gas-solids
mixing, hence reducing residence times in the absorber and allowing for more rapid evaporative cooling
(1-2 seconds vs about 10 seconds in the conventional spray dryer). From a configuration perspective,
these designs, for the most part, represent variations of Circulating Fluid Bed (CFB) technology with
differentiating design features specific to each vendor. As in conventional spray dryers, lime slurry or
hydrated lime are the typical sorbents used. A generic description is presented below, followed by brief
descriptions of specific technologies offered by the major vendors. They include Circulating Dry
Scrubbing (CDS) offered by LURGI, Gas Suspension Absorbers by FLS Miljo, Reflux Circulating
Fluidized Bed Absorbers (RCFB) by WULFF GmbH, and Rapid Absorption Process (RAP) by Beaumont
Environmental.
CFB Absorber Technology
The CFB absorber is a vertical reactor where a dense material bed of recycle products (ash and sorbent)
ensures high gas-solids contact and more rapid cooling. The flue gas flows upward through the bed, with
sorbent typically being sprayed as a slurry into the gas upstream of the bed. Product recycle is
accomplished through a dedicated cyclone integral to the system, or directly from the final particulate
collection device (ESP or FF).
CDS - LURGI13
This technology uses a circulating fluid bed to establish a zone of high particle density. This bed is
enhanced with activated carbon for mercury adsorption. Carbon utilization is enhanced due to the high
residence time in the bed, while fine particles tend to agglomerate through "collisions" in the bed,
facilitating their subsequent capture in a conventional ESP or FF. The technology is used with hydrated
lime injection for control of acid gases.
GSA - FLS Milio4'5
Gas Suspension Absorption (GSA) uses a cyclone (designed for about 90 percent particulate removal) to
recycle products into a dense bed which allows for rapid evaporative cooling as with the other
technologies. This, in turn allows for lower temperatures [less than 11 °C (20 °F) approach to saturation
3-S
-------
temperatures]. Flue gas from the boiler flows directly into the bottom of the GSA vessel. Simultaneously,
lime-slaked slurry is atomized into the reactor, flowing upward with the flue gas. The lime content of the
slurry is varied, depending on the SO2 removal objectives.
GRAF - WULFF GmbH6 7
The RCFB introduces an internal reflux within the circulating fluidized bed designed to increase the gas-
solid mixing and sorbent residence time. This reflux (recirculation) within the RCFB reactor is equivalent
to approximately 30-50 percent of the external product recirculation (from the downstream ESP or FF).
As with the other technologies, gas temperature is controlled via internal water injection and SO2
reduction via the amount of sorbent supply.
RAP - Beaumont8'9
The Rapid Absorption Process uses a flash drying reactor technology combined with an external mixing
chamber. In the flash drying process, recycle is accomplished through the final particulate control (ESP or
FF). This reduces total pressure drop through the system. Differently from circulating fluid beds, lime
slurry is introduced into in a recycle transfer bin where it is mixed with recycle products, and then
introduced into the reactor. Rapid cooling occurs as the products are introduced into the reactor, allowing
for close-to-saturation temperatures and lower residence times.
Commercial Readiness and Industry Experience
While not widely used in the U.S. at present, the four CFB-based technologies are commercial, with
installations in the U.S. and abroad. The RAP technology has undergone pilot-scale tests and has been
demonstrated at a facility in Ohio combined with a low temperature NOX removal system.
CDS - LURGI
The first commercial installation in the U.S. was deployed in 1995 at Black Hills Power and Light's Neil
Simpson Station, Unit 2 on Powder River Basin coal. This is an 80 MW, coal-fired unit. The CFB was
installed in a system arrangement with an ESP for particulate control. A 55 MW unit was installed in
1995 at the Roanoke Valley Energy Facility on eastern bituminous coal. The CFB reactor has been tested
for multi-emission control in pilot scale at the 321 MW coal-fired boiler at PSE&G's Mercer Station,
using activated carbon and lime for mercury and SO2 control, respectively. Testing was carried out daily
during a three-month period.
GSA - FLS Milio
FLS Miljo offers GSA commercially, with over 35 installations worldwide in operation since 1986. In the
U.S., GSA was demonstrated as part of DOE's Clean Coal Technology (CCT) program, at TVA's Center
for Emissions Research. This program evaluated GSA and ESP as well as GSA and FF system
configurations. More recently, a GSA retrofit was implemented to a 130 MW boiler at the Xiaolongtan
power station in China. Demonstration of GSA on Boiler #9 (500,000 lbs.hr of 1,275 psig, 950 °F steam)
at the City of Hamilton, OH showed SO2 control exceeding 90 percent.10
GRAF - WULFF GmbH
WULFF GmbH commercially offers the RCFB. The technology is in full commercial use in plants
ranging from 3 to 300 MW and firing various fuels for the simultaneous removal of SO2 and mercury.
Commercial Single module 660 MW system for a coal-fired plant is available.11
RAP - Beaumont
The technology is currently being demonstrated at the SRI's combustion test facility. In addition, a full-
scale demonstration program at the Medical College of Ohio is currently proceeding.
3-9
-------
SO? and Mercury Emission Control Performance
In general, the advanced scrubbers are capable of >90 percent SO2 reduction.12'13 With respect to mercury
control, less information is available, but high removal rates (such as in conventional spray dryers) have
been reported. Results will vary (as seen from the spray dryer data) depending on many operational
factors such as solids concentration, temperatures, type of sorbent, or downstream particulate control
device (ESP vs FF).
CDS - LURGI
Results from the Neil Simpson Station retrofit indicated around 95 percent SO2 reduction in the CFB and
ESP system, using coal with sulfur content varying from 0.2 percent to 1.2 percent. EPA-sponsored tests14
at the Roanoke Valley unit in September of 2000 indicated total mercury removal of 97 percent. At the
Mercer Station, results for the CFB without activated carbon injection, indicated 50 percent capture of the
mercury vapor. Mercury was reduced by 80 percent when the CFB was injected with iodine-impregnated
activated carbon (at 1000:1 AC-to-Hg ratio).
GSA - FLS Milio15
Results from the demonstration project at TVA under the CCT showed overall SO2 removal efficiency of
about 95 percent in the GSA and FF configuration at Ca to S ratio of 1.4, and about 90 percent in the
GSA and ESP configuration with the same Ca to S ratio. The results of mercury removal tests in the GSA
ranged from about 41.5 to 89.5 percent, without the use of activated carbon. The significant difference in
removal efficiencies was attributed to variations in the residual carbon in the fly ash and the chlorine
content of the coal.
GRAF - WULFF GmbH
SO2 removal efficiency in the RCFB is reported from several plants to be in the 85 to 99 percent range. At
the Dessau Heat and Power Station (operating on brown coal), SO2 reduction was 90 - 96 percent. At the
Strakonice plant in the Czech Republic, the desulfurization efficiency ranged between 85 and 95 percent.
At the Theiss 2000 Power Station in Austria, SO2 capture ranged from 90 to 99 percent. All three projects
were carried out during the late 1990s through 2000. According to WULFF GmbH, mercury reduction
using activated carbon in the RCFB can reach 98 percent. Mercury removal without activated carbon
injection can reach up to 80 percent.
RAP - Beaumont
The technology is currently being piloted at a combustion test facility. Initial results were as high as 95
percent removal, but formal testing and demonstration on various coals and varying conditions will be
part of a continuing program. According to Beaumont, at the Medical College of Ohio, the test program
will investigate SO2 reductions from 70 to 95 percent and sorbent utilization will be documented. Recycle
rates and other variables for various coals are planned.
O&MImpacts
O&M impacts are similar to those of conventional spray dryers. For example, increased pressure loss,
increased auxiliary power, increased solids loading, and its impact on the downstream particulate
collector (ESP or FF) and ash handling equipment, corrosion of the downstream equipment, ductwork and
stack due to low approach-to-saturation temperatures, and increased water requirements for ash
conditioning. Because of better drying capability, the RAP technology, with lower exit moisture content,
may minimize low temperature corrosion concerns. The magnitude of these impacts is a function of
specific designs and operating conditions and whether the ESP or FF is new or existing.
Capital Costs
Capital costs vary from a low of 50 $/kW (for a 150 MW plant) estimated by Beaumont, to about 150
$/kW as reported for GSA from the CCT program.16 Capital cost for Lurgi CDS were estimated to be 140
3-10
-------
$/kW.16 According to GRAF GmbH, the capital cost for the RCFB at the 275 MW Theiss 2000 plant in
Austria was 90 $/kW.
O&MCosts
Fixed O&M costs are not well documented but should fall in the general area of about 1-2 percent of
capital cost (EPRI TAG Guidelines). This would translate to a range of about 1-3 $/kW-yr.
Variable O&M costs are expected to be similar or lower than those for spray dryers, therefore about 0.2-
0.7 mill/kW-hr,10'12'17'18 corresponding to a sulfur content range of 0.4 - 1 percent.
Issues Associated with Advanced Dry Absorbers; Future Outlook
Similar to spray dryers, SO2 performance is well documented for the three CFB-based absorbers. Mercury
capture can potentially be >90 percent but is not well understood at present. Opportunity to add
"dedicated" mercury sorbent will increase overall mercury removal potential.
References
1. Helfritch, D. J.; Feldman, P. L.; Pass, S. J. A Circulating Fluid Bed Fine Particulate and Mercury
Control Concept. Proceedings of the EPRI-DOE-EPA Combined Utility Air Pollution Control
Symposium: The MEGA Symposium, Washington, DC, Aug 25-29, 1997.
2. Helfritch, D. J.; Feldman, P. L.; Waugh, E. The Pilot Scale Testing of a Circulating Fluid Bed for
Mercury Adsorption and Particle Agglomeration. Proceedings of the EPRI-DOE-EPA Combined
Utility Air Pollution Control Symposium: The MEGA Symposium, Atlanta, GA, Aug 16-20,
1999.
3. Lavely, L. L.; Toher, J.; Schild, V. First North American Circulating Dry Scrubber and
Precipitator Remove High Levels of SO2 and Particulate. Proceedings of the EPRI-DOE-EPA
Combined Utility Air Pollution Control Symposium: The MEGA Symposium, Washington, DC,
Aug 25-29, 1997.
4. Burnett, T. A.; Norwood, V. M; Puschaver, E. J.; Hsu, F. E.; Bhagat, B. M.; Marchant, S. K.;
Pukanic, G. W. 10- MW Demonstration of the AirPol Gas Suspension Absorption Flue Gas
Desulfurization Process. Proceedings of The 1993 SO2 Control Symposium, Boston, MA, Aug
24-27, 1993.
5. Felsvang, K.; Lund, C.; Kumar, S.; Bechoux, E.; Hoffman, D. Mercury Reduction Control
Options. Proceedings of the U.S. EPA-DOE-EPRI Combined Power Plant Air Pollutant Control
Symposium: The MEGA Symposium and The A&WMA Specialty Conference on Mercury
Emissions: Fate, Effects, and Control, Chicago, IL, Aug 20-23, 2001.
6. Graf, R.; Zimmer, S. Dry Flue Gas Scrubbing in Heat and Power Stations: Operating Experience
with Medium- and Large-Size Units. Proceedings of the Pittsburgh Coal Conference, Pittsburgh,
PA, Sept 11-15, 2000.
7. Graf, R.; Zimmer, S. High Efficiency Circulating Fluid Bed Scrubber. Proceedings of the U.S.
EPA-DOE-EPRI Combined Power Plant Air Pollutant Control Symposium: The MEGA
Symposium and The A&WMA Specialty Conference on Mercury Emissions: Fate, Effects, and
Control, Chicago, IL, Aug 20-23, 2001.
8. Gross, W. Advances in Semi-Dry Absorption for Multipollutant Control. Proceedings of the U.S.
EPA-DOE-EPRI Combined Power Plant Air Pollutant Control Symposium: The MEGA
3-11
-------
Symposium and the A&WMA Specialty Conference on Mercury Emissions: Fate, Effects, and
Control, Chicago, IL, Aug 20-23, 2001.
9. Lutwen, R. C.; Fitzpatrick, T.; Knott, D.; Goss, W. L.; Ferrell, R.; Suchak, N. Rapid Absorption
Process SO2 Reduction System LoTOx NOX Reduction System. Installation at: Medical College
of Ohio, Powergen, May 2001.
10. Kumar, K. S; Sisler, P. Airpol Gas Suspension Absorber System Project; Final Report; City of
Hamilton, Ohio, March 2003.
11. Graff, R. E.; Seitz, A.; Gao, X. F. Advanced, Large-capacity, Commercial Technology for
Multipollutant Control. Proceedings of the Combined Utility Air Pollutant Control Symposium:
The MEGA Symposium, Washington, DC, May 19-21, 2003.
12. Keeth, R.; Ireland, P.; Radcliffe, P. Economic Evaluation of 28 FGD Processes. Proceedings of
the!991 SO2 Control Symposium, Washington, DC, Dec 3-6, 1991.
13. Jozewicz, W.; Singer, C.; Srivastava, R. K.; Tsirigotis, P. E. Status of SO2 Scrubbing
Technologies. Proceedings of the US EPA/DOE/EPRI Combined Utility Air Pollutant Control
Symposium: The MEGA Symposium, Atlanta, GA, Aug 16-20, 1999.
14. ARCADIS Geraghty and Miller. Roanoke Valley Energy Facility Mercury Testing, Test Report;
Nov 6, 2000.
15. 10-MW Demonstration of Gas Suspension Absorption, Clean Coal Technology Demonstration
Program, Program Update; U.S. Department of Energy: Washington, DC, Oct 1997.
16. Guidelines for Particulate Control for Advanced SO 2 Control Processes; EPRI TR-104594; Palo
Alto, CA, Dec 1994.
17. Krause, T.; Rosenquist, W. Environmental Control Technologies for Our Next Generation of
Coal Plants. Proceedings of the U.S. EPA-DOE-EPRI Combined Power Plant Air Pollutant
Control Symposium: The MEGA Symposium and The A&WMA Specialty Conference on
Mercury Emissions: Fate, Effects, and Control, Chicago, IL, Aug 20-23, 2001.
18. Weilert, C.V. Statistical Analyses of FGD System Operating Costs: Update for 1993-1997.
Proceedings of the US EPA/DOE/EPRI Combined Utility Air Pollutant Control Symposium: The
MEGA Symposium, Atlanta, GA, Aug 16-20, 1999.
3.2.1.2 Sorbent Injection
Sorbent injection processes refer to the use of sorbent materials, typically in a powder or slurry form,
which are injected into the flue gas upstream of a particulate control device. A key difference between
these processes and those described in the previous section (dry scrubbers) is that in the latter, a dedicated
reactor (the scrubber) is used to maximize the desired reactions, whereas in sorbent injection, the
reactions occur in the flue gas duct and the downstream particulate control device. Inherently,
performance of sorbent injection is directly related to the type of particulate control used with it (ESP or
FF), as these devices offer additional "opportunity" for the reactions to take place, through residence or
"contact" time.
3-12
-------
Sorbent injection technologies can be characterized by the type of sorbent they utilize and by the type of
pollutant they address. For example, activated carbon may be used for mercury control or calcium-based
sorbent may be used for SO2 control. Other differentiating features may include details on sorbent
preparation, injection approach, and integration with the particulate control device, etc.
Sorbent injection technologies received significant "attention" in the US in the 1980's, being developed
as lower-cost and lower efficiency alternatives for SO2 control. More recently, the technology has seen
renewed interest, development and demonstration activities, driven by EPA's December 2000
determination to control emissions of mercury from power plants.
The following sections describe several sorbent injection processes using different sorbents to achieve
mercury or SO2 reductions.
3.2.1.2.1 Activated Carbon with Particulate Controls
Activated Carbon with Particulate Controls Summary
Status
SC>2 Reduction
NOx Reduction
Hg Reduction
Cost
Capital ($/kW)
Fixed O&M ($/kW-yr)
Variable O&M (mills/kWh)
Applicability
Issues
Pilot to commercial
None
None
50 - 90 %
3 to >8 (for the carbon injection system only, low end of
range for FF and COHPAC, high end for ESP application)
1 -2
0.2-2
Retrofit and new units with ESP or FF
Not widely demonstrated at full scale on coal-fired plants,
ash salability, particulate control effect on performance,
impact of coal type
Technology Description
The technology involves the injection of an activated carbon (or alternative non-carbon sorbent) powder
into the flue gas duct, somewhere between the air preheater and the ESP or FF. This is typically in the
120-175 °C (250-350 °F) range. Vapor-phase mercury is adsorbed onto the activated carbon, which is
then collected in the ESP or FF. The mercury-activated carbon interaction continues to occur in the ESP
or FF, where, in fact, the majority of the adsorption takes place. The technology can be used in
conjunction with flue gas temperature control, usually accomplished through the injection of water
droplets (spray cooling) into the flue gas. This can be done to optimize the temperature at which the
activated carbon-mercury adsorption occurs. Lower approach-to-saturation temperatures favor the
process.
A variation of this technology was developed and patented by EPRI. Named TOXECON, it is based on
the combination of an ESP and a high air-to-cloth pulse-jet baghouse with sorbent injection technology
(ACI plus COHPAC). This approach applies to retrofit situations where an ESP already exists and
focuses on improving the efficiency of sorbent injection by providing high efficiency particulate
collection as well as a good "contact" scheme for the sorbent and mercury, such as in the FF.
3-13
-------
The most commonly studied sorbent for mercury control has been activated carbon. This material has
been successfully used as a sorbent in municipal and hazardous waste combustors.1 Activated carbon is
carbon that has been "treated" to reflect certain properties such as surface area, pore volume, and pore
size. Activated carbon can be manufactured from a variety of sources, including lignite, peat, coal, and
wood. More commonly, steam is used for activation, which requires carbonization at high temperatures in
an oxygen-lean environment. As some carbon atoms are vaporized, the desired highly porous activated
carbon is produced. Commercially, activated carbon is available in a range of particle sizes, as well as
other performance characteristics.2 Furthermore, special activated carbon products such as iodine and
sulfur impregnated are also available, and have been studied.3'4 Sorbents are often compared in terms of
"capacity" and "reactivity." These are a function of many parameters including surface area and porosity.
Reactivity refers to the initial rate of reaction. Capacity refers to its ability to adsorb mercury in terms of
mass of mercury captured per mass of sorbent.
Commercial Readiness and Industry Experience
To date, several pilot scale test programs and demonstrations have been conducted. Also, full-scale
demonstration programs were completed to evaluate the performance, cost, and impacts of activated
carbon injection. The program, sponsored by DOE, investigated several plant configurations, coals, and
sorbents. This experience, combined with the experience gained in the 1980's with sorbent injection for
SO2 control, provides a sound basis for designing and implementing the technology.
The following are some of the pilot and full-scale demonstrations that were conducted:
• Public Service of Colorado, Comanche Station - pilot scale ESP, FF, and COHPAC and TOXECON5
• PSE&G Hudson Station - pilot scale COHPAC and TOXECON6
• Alabama Power, Gaston Plant - full scale COHPAC and TOXECON7
• WEPCO, Pleasant Prairie - full scale ESP7
• PGE NEG, Salem Harbor Station - full scale ESP7
• PGE NEG, Brayton Point Station - full scale dual ESPs7
The other component of the technology - activated carbon - is a commercially available product widely
used in industry. Another source of carbon-based sorbent may be coal fly ash. Coal ash with varying
carbon levels has shown utility as a mercury sorbent.8 However, properties of unburned carbon that
contribute to mercury sorption are still not well understood. It is expected that fly ash-derived sorbents
may become, in some cases, a cost-effective alternative to activated carbon. In addition, sorbents for
mercury may also be produced from other inexpensive materials, such as corn-derived biomass and waste
tires.9
Emission Control Performance
Approximately 86 percent of the coal-fired utility boilers currently operating in the United States are
equipped with only an ESP or an FF, with ESP as the predominant PM emission control device. Gaseous
mercury (both Hg° and Hg2+) can potentially be adsorbed on fly ash and be collected in a downstream
ESP or FF. The modern ESPs or 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. 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
3-14
-------
to which gaseous mercury adsorbs on the filter cake typically depends on the speciation of gaseous
mercury in the flue gas; in general, gaseous Hg2+ 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.10
The Information Collection Request (ICR) data showed that, for both bituminous and subbituminous
coals, mercury collection in boilers equipped only with FFs was much higher than for boilers equipped
only with ESPs. ICR data reflected that plants which employ only post-combustion PM controls display
average Hg emission reductions ranging from 0 percent to 89 percent. The highest levels of control were
observed for units with FFs. Decreasing levels of control were shown for units with ESPs, particulate
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 a cold-side ESP was 35 percent for bituminous coal, 3 percent for
subbituminous coal, and near zero for lignite.11
This effect also contributes to much more efficient collection of mercury when powdered activated carbon
(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. The overall
performance of the activated carbon injection technology is a function of many factors, including sorbent
characteristics as well as plant configuration and operating conditions. Mercury reductions of 80 - 90
percent were documented at Alabama Power, Gaston Station for activated carbon with COHPAC.12'13
Reduction levels from 50 to 90 percent are possible for the range of technology configurations and
activated carbons available. Model predictions14 for various configurations, sorbent characteristics, and
operating conditions confirm this range of performance. If COHPAC is added to an existing ESP, in
addition to the mercury reduction, particulates (including PM2 5) are reduced, too.15
Laboratory, pilot scale, and modeling programs16"18 have indicated that the following parameters can
affect the ultimate performance of the technology:
• sorbent type and properties,
• gas-phase mercury species (Hg° or HgCl2),
• temperature,
• concentration of acid gases (HC1, SO2, NO, NO2) in the flue gas,
• overall residence time, and
• dispersion of the sorbent in the flue gas.
• Further, the overall performance of the technology is a function of the quantity of sorbent required to
achieve a desired result as can be observed below. Figure 3-2 contrasts pilot-scale ESP data (open
symbols) with baghouse data (closed symbols) for sorbent injection and the same eastern bituminous
coal burned at PSE&G's Hudson Station.19 As expected, given the better contact between sorbent and
gas in a FF, the data confirm a higher mercury capture for the same sorbent loading.
Recent review and modeling20 of full-scale data obtained during field tests at Gaston, Pleasant Prairie,
Brayton Point, and Salem Harbor power plants produced the following findings:
3-15
-------
PAC injection followed by a FF results in much lower injection concentrations being necessary for a
given level of mercury reduction than for PAC injection followed by a cold-side ESP.
Sorbent selection appears to have little effect on performance when PAC injection is followed by a
FF, but it appears to have a significant effect when PAC injection is followed by an ESP.
Loss-on-ignition (LOI) and temperature can have a significant effect on the mercury removal by
existing equipment.
In some cases PAC injection without a downstream FF may not be able to achieve mercury removal
rates of 90 percent or more regardless of PAC injection concentration.
5
o
E
a:
a>
80
60
90 -
zu
o I
* "
V
7
* o
o
*
V
o
2463
Activated Carbon Loading, Ib/MMcf
10
115-120 :'C (240-24? °F)
1400C(2SO°F)
Figure 3-2. Mercury removal versus activated carbon loading in a pilot scale ESP (open symbols) and
baghouse (closed symbols).19
O&M Impacts
The major potential O&M impacts associated with the activated carbon injection involve the following:
• increased particulate loading to the ESP or FF; this is not expected to be significant enough to affect
performance, especially for the case of the FF,
• increased carbon content and different particle size distribution which may affect ESP and FF
performance or operation factors such as entrainment, dust cake pressure loss,
• increased carbon content of fly ash may affect its salability,
• low flue gas temperatures (if spray cooling is used) affecting SO3 dew point and, thus, intensifying
corrosion and reducing bag life,
• auxiliary power for activated carbon injection, even though the amount is relatively small, and
• disposal and utilization of ash; testing at a number of DOE-funded demonstration projects indicates
that AC injection does not affect the leaching properties of the ash to require changes in its disposal.
However, if the ash is utilized or sold for construction applications, the increased concentration of
carbon may affect the ability of the utility to sell it.
3-16
-------
Capital Costs
Capital costs will vary significantly for applications with a FF or COHPAC vs those with an ESP. This is
because of the significantly higher activated carbon requirements (may be a factor of 10) for ESP
applications and the associated costs for sorbent storage. According to ADA-ES,7 the costs for the carbon
injection system will vary from about $2/kW for a FF and COHPAC configuration, to >$6/kW for ESP
applications. If spray cooling is used, the additional cost is estimated at $1 - 2/kW. COHPAC's capital
costs are estimated to be between 57 and 59 $/kW for mercury removal between 50 and 90 percent,
respectively.20
O&MCosts
Fixed O&M costs are estimated to be in the range of $l-2/kW-yr for a nominal plant size of 100 - 200
MW. Variable O&M will be driven by sorbent cost. Hence FF and COHPAC applications are estimated
at 0.15-0.2 mills/kWh, with ESP potentially up to 2 mills/kWh. Variable O&M costs for COHPAC are
estimated to be between 0.24 and 0.39 mills/kWh for mercury removal between 50 and 90 percent,
respectively.20
Issues Associated with Activated Carbon Injection; Future Outlook
At present, the technology is being demonstrated at full scale in several different applications. No
significant "setbacks" have been identified. Due to its low capital cost, ease of retrofit, and compatibility
with both ESP's and FF's, it is expected that carbon injection may become an efficient approach for
retrofit in power plant applications without a combination of NOX controls and scrubbers that may
provide significant mercury capture of their own.
If the technology were used on a wide scale basis, deployment of new kilns and furnaces would be
necessary to increase the production of activated carbon to meet the potential market for coal-fired
boilers. The current market for activated carbon is 250,000 tons per year. Once mercury regulations are
fully implemented, this could increase the demand.
References
1. Environmental Regulation and Technology Innovation: Controlling Mercury Emissions from
Coal-Fired Boilers; NESCAUM: Boston, MA, Sept 2000. http://www.nescaum.org/pdf/hg-exec-
summ.pdf (accessed Dec 1, 2004).
2. Brown, T. D.; Smith, D. N.; Hargis, R. A.; O'Dowd, W. J. Mercury Measurement and Its Control:
What We Know, Have Learned, and Need to Further Investigate. J. Air Waste Manage. Assoc.
1999, 1-97.
3. Hsi, H.; Rostam-Abadi, M.; Rood, M. J.; Chen, S. Mercury Removal from Simulated Flue Gases
with Illinois Coal-Derived Activated Carbon (ICDAC). Presented at the A&WMA 91st Annual
Meeting and Exhibition, San Diego, CA, June 14-18, 1998; Paper 98-RA79B.03.
4. Hsi, H.; Rostam-Abadi, M.; Rood, M. J.; Chen, S.; Chang, R. Preparation of Sulfur-Impregnated
Activated Carbon Filters (ACFs) for Removal of Mercury Vapor from Simulated Coal-
Combustion Flue Gases. Presented at the A&WMA 93rd Annual Meeting and Exhibition, Salt
Lake City, UT, June 19-22, 2000.
5. Waugh, E.; Jensen, B.; Lapatnick. L.; Gibbons, F.; Sjostrom, S.; Ruhl, J.; Slye, R.; Chang, R.
Mercury Control in Utility ESPs and Baghouses Through Dry Carbon-based Sorbent Injection-
3-17
-------
Pilot Scale Demonstration Flue Gas Using Sorbent Injection. Proceedings of the EPRI-DOE-EPA
Combined Utility Air Pollutant Control Symposium, Washington, DC, Aug 28, 1997.
6. Hunt, T.; Haythornthwaite, S.; Sjostrom, S.; Ebner, T.; Ruhl, J.; Slye, R; Smith, J.; Chang, R.;
Brown, T. Demonstration of Dry Carbon-Based Sorbent Injection for Mercury Control in Utility
ESPs and Baghouses. Proceedings of the EPRI-DOE-EPA Combined Utility Air Pollutant
Control Symposium, Washington, DC, Aug 28, 1997.
7. Durham, M. D.; Bustard, J.; Schlager, R.; Martin, C.; Johnson, S.; Renninger, S. Controlling
Mercury from Coal-Fired Utility Boilers: A Field Test. Environ. Manage. 2001, 27-33.
8. Butz, J.; Smith, J.; Grover, C.; Haythornthwaite, S.; Fox, M.; Hunt, T.; Chang, R.; Brown, T.D.
Coal Fly Ash as a Sorbent for Mercury. Proceedings of the Mercury in the Environment Specialty
Conference, Air & Waste Management Association, Minneapolis, MN, Sept 15-17, 1999.
9. Rostam-Abadi, M.; Chen, S.; Lizzio, A.; His, H.; Lehmann, C. M. B.; Rood, M. J.; Chang, R.;
Richardson, C.; Machalek, T.; Richardson, M. Development of Novel Sorbents for Mercury
Removal from Utility Flue Gas. Proceedings of the U.S. EPA-DOE-EPRI Combined Power Plant
Air Pollutant Control Symposium: The MEGA Symposium and The A&WMA Specialty
Conference on Mercury Emissions: Fate, Effects, and Control, Chicago, IL, Aug 20-23, 2001.
10. Studt, J. E.; Jozewicz, W. Performance and Cost of Mercury and Multipollutant Emission
Control Technology Applications on Electric Utility Boilers; EPA-600/R-03-110; Research
Triangle Park, NC, Oct 2003.
11. 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.
12. Bustard C. J.; Durham, M.; Lindsey, C.; Starns, T.; Baldrey, K.; Martin, C.; Sjostrom, S.; Slye,
R.; Renninger, S.; Monroe, L. Full-Scale Evaluation of Mercury Control with Sorbent Injection
and COHPAC at Alabama Power E. C. Gaston. Proceedings of the U.S. EPA-DOE-EPRI
Combined Power Plant Air Pollutant Control Symposium: The MEGA Symposium and The
A&WMA Specialty Conference on Mercury Emissions: Fate, Effects, and Control, Chicago, IL,
Aug 20-23, 2001.
13. Bustard, J.; Renninger, S.; Chang, R.; Miller, R.; Monroe, L.; Sjostrom, S. Results of Activated
Carbon Injection for Mercury Control Upstream of a COHPAC Fabric Filter. Proceedings of the
Combined Utility Air Pollutant Control Symposium: The MEGA Symposium, Washington, DC,
May 19-21,2003.
14. Carey, T.; Richardson, C.; Meserole, F. B.; Chang, R. Estimating Electric Utility Mercury
Control Costs Using Sorbent Injection. Proceedings of the EPRI-DOE-EPA Combined Utility Air
Pollutant Control Symposium, Atlanta, GA, Aug 16-20, 1999.
15. Miller, R.; Morris, W. J. Effective Use of COHPAC Technology as a Multi-Pollutant Control
Technology. Proceedings of the US EPA/DOE/EPRI Combined Utility Air Pollutant Control
Symposium: The MEGA Symposium, Chicago, IL, Aug 20-23, 2001.
16. Carey, T. R; Hargrove, O. W.; Richardson, C. F.; Chang, R.; Meserole, F. B. Factors Affecting
Mercury Control in Utility Flue Gas Using Activated Carbon. J. Air Waste Manage. Assoc. 1998,
48, 1166-1174.
3-18
-------
17. Miller, E. S.; Dunham, G. E.; Olson, E. S.; Brown, T. D. Mercury Sorbent Development for Coal-
Fired Boilers. Presented at the Conference on Air Quality, McLean, VA, Dec 1-4, 1998.
18. Afonso, R.; Sjostrom, S.; Chang, R.; Renninger, S. Results of Activated Carbon Injection
Upstream of ESP for Mercury Control. Proceedings of Combined Utility Air Pollutant Control
Symposium: The MEGA Symposium, Washington, DC, May 19-21, 2003.
19. Butz, J. R.; Chang, R.; Waugh, E. G.; Jensen, B. K.; Lapatnick, L. N. Use of Sorbents for Air
Toxics Control in a Pilot-scale COHPAC Baghouse. Presented that 92nd Annual Meeting and
Exhibition of the Air & Waste Management Association, St. Louis, MO, June 20-24, 1999.
20. Staudt, J. E.; Jozewicz, W.; Srivastava, R. K. Modeling Mercury Control with Powdered
Activated Carbon. Proceedings of the Combined Power Plant Air Pollutant Control Symposium:
The MEGA Symposium, Washington, DC, May 2003.
3.2.1.2.2 SO2 Sorbents
SO2 Sorbents Summary
Status
SO2 Reduction
NOx Reduction
Hg Reduction
Cost
Capital ($/kW)
Fixed O&M ($/kW-yr)
Variable O&M (mills/kWh)
Applicability
Issues
Pilot scale to pre-commercial demonstration
40 - 85 %
None
NA
25 - 120 for 500 MW plant
1.6-7
0.5-1
Units with ESP or FF for particulate control
Calcium-based compounds not used commercially in coal
fired plants. Waste disposal issue with sodium-based
compounds. Potential impacts on ESP or FF.
Technology Description
SO2 sorbent injection technologies include several variations based on the point of injection (furnace,
economizer, ductwork upstream of an ESP or FF) and the sorbent type (calcium vs sodium based). In all
cases, the major premise is that, by using the existing system as the "reactor," significant cost savings can
be achieved in situations where the SO2 reductions of 40 - 50 percent are appropriate. In this section, the
focus is on technologies involving the injection of a calcium- or sodium-based sorbent into the flue gas
duct, somewhere between the air preheater and the ESP or FF (commonly referred to as "dry duct
injection"). This is typically in the 120-175 °C (250-350 °F) range. SO2 reacts with the sorbent to produce
a mixture of sulfate and sulfite salts, which are then collected in the ESP or FF. The sorbent is either
injected as atomized slurry, or as a powder separate from the water injection. Low approach-to-saturation
temperatures enhance the ensuing reactions, therefore maximizing the SO2 reductions. Several different
approaches to the technology have been developed and are described briefly below.1
Duct Injection - Lime Slurry
The duct injection of lime slurry is a process similar to a conventional spray dryer. The main difference is
the elimination of the large reaction vessel by the direct spraying of lime slurry into the ductwork between
3-19
-------
the air heater and the participate control device. To offset the fact that the ductwork provides a much
shorter residence time than the spray dryer vessel, slurry is atomized into very fine droplets in the duct
injection process. The slurry droplets in such a fine mist dry out before reaching the particulate control
device or prior to coming into contact with the duct walls, as required for proper operation. Two-fluid (air
and slurry) atomizing spray nozzles are commonly used to provide the small droplets. Lime slurry
injection was shown to provide slightly higher SO2 removals than duct injection of dry hydrated lime.
Specialized processes and additives to improve the performance of the lime slurry injection process were
developed at several commercial enterprises. They include Bechtel's Confined-Zone Dispersion (CZD)
process, using pressure-hydrated dolomitic lime; General Electric's In-Duct Scrubbing (IDS) process, an
in-duct spray drying system based on a rotary atomizer; and EPA's E-SOX process, consisting of an array
of spray nozzles fitted into a cavity created by removing the internals from the first field of an ESP.
Duct Injection - Dry Lime with Humidification
The use of dry hydrated lime requires separate injection of water, upstream or downstream of the lime
injection point, to activate the lime. The water is finely atomized to enhance evaporation and to avoid
wetting the duct walls. As with lime slurry, a number of commercial processes were developed including:
• Dravo's HALT Process - the Hydrate Addition at Low Temperature process which included the
addition of NaOH to the water spray to enhance the lime-SO2 reaction
• CONSOL's Coolside Process - similar to the HALT
• EPA's AD VACATE Process - AD VACATE uses an advanced sorbent produced by reacting lime
and fly ash at elevated temperature
• EPRI's HYPAS - the Hybrid Pollution Abatement System injects hydrated lime and water between
an ESP and a downstream FF
Duct Injection - Sodium Compounds
The use of sodium-based processes not requiring water was developed primarily for water-scarce regions
such as in certain western states. Tests of various sodium compounds, including nahcolite, trona, soda
ash, sodium bicarbonate, and sodium sesquicarbonate were conducted at a number of locations. The
sodium sorbent is pulverized and injected into the flue gas through an array of nozzles located in the
ductwork.
Commercial Readiness and Industry Experience
These technologies were tested or demonstrated extensively during the mid 1980's but never gained wide
commercial acceptance. However, significant experience was gained regarding various sorbents as well as
several injection approaches. Tables 3-2, 3-3, and 3-4 identify some of the test or demonstration and
commercial programs.
Table 3-2. Duct Injection of Lime Slurry
Process
Confined-Zone Dispersion (CZD)
In-Duct Scrubbing (IDS)
E-SOx
Developer
Bechtel
General Electric
U.S. EPA
Test Site
Seward Station
Muskingum River Station
Burger Station
Reference
2
3
4
3-20
-------
Table 3-3. Duct Injection of Dry Lime
Process
Hydrate Addition at Low Temperature
(HALT)
Coolside
Advanced silicate (ADVACATE)
Hybrid Pollution Abatement System
(HYPAS)
Developer
Dravo Lime Company
Conoco
U.S. EPA
EPRI
Test Site
Toronto Station
Edgewater
Edgewater Station (sorbent)
TVA Shawnee (process)
EPRI High-Sulfur Test Center
Reference
5
6
7
8
Table 3-4. Duct Injection of Sodium
Utility or Operator
Public Service Company of Colorado
City of Colorado Springs
Public Service Company of Colorado
Wisconsin Electric Power Company
Plant and Location
Cameo Unit No. 1 Grand Junction, CO
Ray D. Nixon Unit No. 1 Fountain, CO
Cherokee Unit No. 4 Denver, CO
Port Washington Unit No. 3 Port Washington, Wl
Reference
9
10
11
12
Emission Control Performance
The results from the various programs indicated SO2 reductions in the range of 40 to about 85 percent,
depending on sorbent type and stoichiometry, amount of recycle, temperature, and plant configuration.
Little information is available on mercury reduction from these programs.
O&MImpacts
The major potential O&M impacts associated with SO2 sorbent injection involve the following:
• increased particulate loading to the ESP or FF,
• different particle size distribution which may affect ESP and FF performance because of entrainment
or dust cake pressure loss,
• impact of sodium injection reaction products or lowered gas temperatures on particulate resistivity,
mostly resulting in an improved ESP performance,
• low flue gas temperatures (if spray cooling is used) affecting SO3 dew point and, thus, intensifying
corrosion and reducing bag life,
• impact on plume visibility from formation of NO2 (for sodium-based sorbents),
• increased solid waste flow rate and associated auxiliary power; potentially, more operating problems
may be experienced in the ash handling system if the right precautions are not taken to address the
different properties of the ash when it includes unreacted sorbent,
• In case of sodium-based sorbents, disposal of solid waste needs to take into account the solubility of
sodium compounds in water and associated concerns with the handling of sodium-containing leachate
from the landfills, and
• increased water consumption mainly due to ash conditioning.
3-21
-------
Capital Costs
Capital costs for these processes vary over a wide range from $25 to 50/kW (FY 1990)13"15 to 70-120
$/kW (FY 1990) for a 500 MW plant. Because there is no commercial application of these technologies, it
is difficult to address the accuracy of these projections.
O&MCosts
O&M costs are not readily available. Fixed O&M is expected to be similar to or lower than spray dryers.
Therefore, the same range of $1.6-7/kW-yr is presented.16 Variable O&M costs are expected to be
somewhat higher than those for conventional spray dryers (lower reagent effectiveness), therefore in the
range of 0.5-1 mills/kWh for the same range of coal sulfur content of 0.4 to 1 percent.
Issues Associated with SO2 Sorbent Injection; Future Outlook
The technology did not gain commercial acceptance as part of the compliance strategies for Phase I and II
of the CAAA of 1990. In the future, broad compliance strategies and increased use of low sulfur coals
may make this technology an attractive option for smaller units, plant or system "bubble" strategies, or in
combination with activated carbon or other multipollutant sorbents to address mercury emissions.
References
1. Guidelines for Particulate Control for Advanced SO 2 Control Processes; EPRI TR-104594;
Electric Power Research Institute: Palo Alto, CA, Dec 1994.
2. Bechtel National, Inc. Desulfurization of Flue Gas by the Confined Zone Dispersion Process;
Final Report; Submitted to the U.S. Department of Energy, DOE Contract No. DE-AC22-
85PC81009, April 1988.
3. Samuel, E. A.; Murphy, K. R.; Demian, A. A 12-MWPilot Study ofln-Duct Scrubbing (IDS)
Using a Rotary Atomizer; Final Report; Submitted to the U.S. Department of Energy, DOE
Contract No. DE-AC22-85PC81010, Jan 1989.
4. Redinger, K. E.; Hovis, L.; Owens, F.; Chang, J. C. S.; Wilkinson, J. Results from the E-SOX
5-MW Pilot Demonstration. Proceedings of the 1990 SO2 Control Symposium, New Orleans, LA,
May 8-11, 1990.
5. Babu, M.; Forsythe, R. C.; College, J.; Herbert, R; Kanary, D.; Lee, K. 5-MW Toronto HALT
Pilot Plant; Final Report; Submitted to the U.S. Department of Energy, DOE Contract No.
DE-AC22-85PC81012, Dec 1988.
6. Statnick, R.; Kanary, D. A.; Yopon, H.; McCoy, D.; Withum, J.; Kudlac, G. Coolside Process
Demonstration at the Ohio Edison Company Edgewater Unit 4-Boiler 13. Proceedings of the
1990 SO2 Control Symposium, New Orleans, LA, May 8-11, 1990.
7. Jozewicz, W.; Chang, J. C. S.; Brna, T.; Sedman, C. B. Characterization of Advanced Sorbents
for Dry SO2 Control. React. Solids 1988, 6, 243.
8. Rhudy, R. G.; Carr, R. C. Pilot-Scale Evaluation of the HYPAS SO2 and Particulate Matter
Removal Process. Presented at the Sixth Annual Coal Preparation, Utilization, and Environmental
Control Contractors Conference, Pittsburgh, PA, Aug 1990.
9. Muzio, L.; Offen, G. R. Dry Sorbent Emission Control Technologies. J. Air Pollut. Control
Assoc. 1987, 37, 642.
3-22
-------
10. Ablin, D. W.; Hammond, J. J.; Watts, D. B.; Ostop, R. L.; Hooper, R. G. Full Scale
Demonstration of Dry Sodium Injection Flue Gas Desulfurization at City of Colorado Springs,
Ray D. Nixon Power Plant. Proceedings of the 1986 Joint Symposium on Dry SO2 and
Simultaneous SO2/NOX Control Technologies, Volume 2, EPA-600/9-86-029b (NTIS PB87-
120457), Air and Energy Engineering Research Laboratory: Research Triangle Park, NC, Oct
1986.
11. Fox, M. R.; Hunt, T.;G. Flue Gas Desulfurization Using Dry Sodium Injection. Proceedings of
the 1990 SO2 Control Symposium, New Orleans, LA, May 8-11, 1990.
12. Coughlin, T.; Schumacher, P.; Hooper, A. R. Injection of Dry Sodium Bicarbonate to Trim Sulfur
Dioxide Emissions. Proceedings of the 1990 SO2 Control Symposium, New Orleans, LA, May
8-11, 1990.
13. Bechtel Corporation. Confined Zone Dispersion Project: Final Technology Report, Clean Coal
Technology Demonstration Program, Program Update; U.S. Department of Energy: Washington,
DC, March 1999.
14. Frazier, W.; Ireland, P.; Brown, G.; Radcliffe, P. Site Specific Evaluation of Six Sorbent Injection
Processes. Proceedings of The FGD and Dry SO2 Control Symposium, St. Louis, MO, Oct 25-28,
1988.
15. Smith, P. V. Dry Sorbent Injection and Acid Rain Compliance. Presented at: Power-Gen '90,
Environmental Trends and Issues, Orlando, FL, Dec 4-6, 1990.
16. Weilert, C. V. Statistical Analyses of FGD System Operating Costs: Update for 1993-1997.
Proceedings of the US EPA/DOE/EPRI Combined Utility Air Pollutant Control Symposium: The
MEGA Symposium, Atlanta, GA, Aug 16-20, 1999.
17. Brusger, E.; Wayland, R. Clean Air Response: A Guidebook to Strategies; EPRI, GS-7105; Dec
1990.
18. Meserole, F.; Richardson, C. F.; Machalek, T.; Richardson, M.; Chang, R. Predicted Costs of
Mercury Control at Electric Utilities Using Sorbent Injection. Proceedings of the U.S.
EPA/DOE/EPRI Combined Utility Air Pollutant Control Symposium: The MEGA Symposium,
Chicago, IL, Aug 20-23, 2001.
19. Singer, C.; Ghorishi, B; Sedman, C. B. Lime Based Multi-Pollutant Sorbents. Proceedings of the
U.S. EPA/DOE/EPRI Combined Utility Air Pollutant Control Symposium: The MEGA
Symposium, Chicago, IL, Aug 20-23, 2001.
3-23
-------
3.2.1.2.3 Combined Mercury and SO2 Sorbents
Combined Mercury and SO2 Sorbents Summary
Status
SO2 Reduction, %
NOX Reduction
Hg Reduction, %
Cost
Capital ($/kW)
Fixed O&M ($/kW-yr)
Variable O&M (mills/kWh)
Applicability
Issues
Pilot scale to commercial
40-85
NA
Up to 90
28 - 58 for a nominal 500 MW plant
2.6-9
0.7-3
Units with ESP or FF for participate control
Not used commercially, potential impacts on ESP or FF
Technology Description
As described in Sections 3.2.1.2.1 and 3.2.1.2.2, the basic technologies for sorbent injection involve the
injection of a sorbent into the flue gas at temperatures in the range of 120-175 °C (250-350 °F) and before
the particulate collection device (ESP or FF). Several actual methods of injection, such as dry powder and
atomized slurry, have been developed to optimize the processes as a function of specific requirements and
conditions. The technologies for combined or multipollutant sorbent injection involve the same
approaches as described in the previous sections. The approaches may involve using sorbents together
(activated carbon plus hydrated lime) or single, multipollutant-capability sorbents. In this section, a brief
summary of combined and multipollutant sorbents is provided.
Commercial Readiness and Industry Experience
Data from the mercury ICR1"3 program demonstrated that dry scrubbers remove mercury and SO2,
suggesting the ability of calcium-based sorbents to enhance mercury capture in addition to SO2. Pilot test
programs have documented the performance of combined sorbents.4"7 Much laboratory activity has
focused on the development of novel and enhanced sorbents.8'9 Based on this experience, sorbent
injection technology for combined SO2-mercury reduction represents a viable, although not fully
quantified approach for multi-emission control.
Emission Control Performance
Calcium-based sorbents have been characterized most extensively.8'10 Co-injection of activated carbon
and calcium-based sorbents in air pollution control equipment has been known to increase the removal of
mercury from flue gas. This was first demonstrated in spray dryer absorber (SDA) systems on full-scale
coal-fired power plants.11 There has been pilot work on injection of calcium-based sorbents upstream of a
baghouse4 and into a specialized fluidized bed reactor.6 Combining activated carbon with hydrated lime
can reduce the amount of carbon required (for an equivalent mercury removal) by one-half to one-third.
Pilot tests of limestone furnace injection, followed by a cyclone separator, also showed good removal of
mercury from flue gas in a pilot-scale unit burning eastern bituminous coals.12
Laboratory investigations of calcium-based sorbents for mercury control8'10 have shed light on the
mechanisms involved that offer the potential for more efficient use of such sorbents across a range of
applications. Fly ash, hydrated lime, and AD VACATE (a trademarked, pressurized fly ash-lime mixture)
have been tested for mercury sorption in a fixed-bed reactor.10 All calcium-based sorbents captured HgCl2
from simulated flue gas at 100 °C (although less than commercial activated carbon). Addition of SO2 to
3-24
-------
the gas mixture decreased the sorption of HgCl2, suggesting that there is competition for the same alkaline
sites between the two species. In contrast, the calcium-based sorbents showed little or no removal of Hg°
in the absence of SO2. Addition of SO2 to the gas greatly enhanced the uptake of elemental mercury,
suggesting the possibility of some chemical reaction on the surface. Increased sorbent surface area and
internal pore volume also increased the capture of elemental mercury by calcium-based sorbents. More
recently, hydrated lime and silicates have been evaluated for mercury, NOX, and SO2 capture in bench-
scale tests. Oxidant-enhanced silicate sorbents indicated enhanced mercury capture. The practical
significance of these results is that it is possibly more effective to separate the injection of sorbents
dedicated to bulk acid gas removal (lower cost alkaline sorbents) from the higher porosity, oxidant-
enhanced sorbents for mercury control.8'9 Unfortunately, while these studies offer a great deal of new
understanding about the chemical and physical interactions between the flue gas, mercury, and sorbent,
the results cannot yet be directly translated to full-scale performance.
Pilot-scale testing at PSE&G Hudson Station4'7 with activated carbon, sodium, and calcium sorbents in a
COHPAC and TOXECON configuration has shown the ability to inject activated carbon simultaneously
with other sorbents. Activated carbon performance was enhanced when tested with hydrated lime. Similar
mercury capture of >80 percent was obtained with much lower levels of activated carbon (about a factor
of 4) when combined with hydrated lime. In other tests using sodium sesquicarbonate and sodium
bicarbonate injection, SO2 reduction was up to about 90 percent for the bicarbonate and about 20 percent
for sesquicarbonate. In other tests at PSE&G's Mercer Station,6 injection of activated carbon and hydrated
lime into a fluid bed reactor resulted in mercury removals of up to 80 percent with iodine-impregnated
carbon at a ratio of 1000:1 and about 67 percent for non-impregnated carbon at a ratio of 2000:1. SO2
removal with hydrated lime was about 70 percent with 100 percent sorbent utilization.
In summary, combined and multipollutant sorbent use with sorbent injection technology is neither used
nor quantified in full-scale applications. However, based on information from activated carbon and SO2
sorbent technologies, it is expected that combined performance will be similar to that of the individual
technologies, while the development of new sorbents may optimize the utilization and efficiency of future
technology applications.
O&M Impacts
The major potential O&M impacts associated with combined and multipollutant sorbent injection include
the following:
• increased particulate loading to the ESP or FF,
• impact of reaction products on particulate resistivity and associated ESP performance,
• low flue gas temperatures (if spray cooling is used) affecting SO3 dew point and, thus, intensifying
corrosion and reducing bag life,
• impact on plume visibility from formation of NO2 (for sodium-based sorbents),
• increased carbon content and different particle size distribution which may affect ESP and FF
performance (entrainment, dust cake pressure loss), and
• impact of increased carbon content on fly ash salability.
Capital Costs
No information on capital costs is available for multipollutant sorbent injection technologies. However, it
can be extrapolated that at one extreme the combination of activated carbon and SO2 sorbent injection
technologies would cost no more (likely less) than the sum of the two. In this case, the range of about
$28-58/kW would apply. In practice, economies would exist in design, construction, and installation of
such multipollutant control technologies.
3-25
-------
O&MCosts
O&M costs are not directly available. However, following the same rationale as above, the combined
technologies would have costs as follows:
• Fixed O&M in the range of $2.6 - 9/kWh
• Variable O&M in the range of 0.7 - 3 mills/kWh.
These are predicated on a coal sulfur range of 0.4 to 1 percent.
13
Issues associated with Combined Mercury and SO2Sorbent Injection; Future Outlook
The same issues apply as for the application of the individual activated carbon and SO2 sorbent injection
technologies.
References
1. Afonso, R. Assessment of Mercury Removal by Existing Air Pollution Control Devices in Full
Scale Power Plants. Proceedings of the U.S. EPA-DOE-EPRI Combined Power Plant Air
Pollutant Control Symposium: The MEGA Symposium and The A&WMA Specialty Conference
on Mercury Emissions: Fate, Effects, and Control, Chicago, IL, Aug 20-23, 2001.
2. Chu, P.; Behrens, G.; Laudal, D. Estimating Total and Speciated Mercury Emissions from US
Coal-fired Power Plants. Proceedings of the U.S. EPA-DOE-EPRI Combined Power Plant Air
Pollutant Control Symposium: The MEGA Symposium and The A&WMA Specialty Conference
on Mercury Emissions: Fate, Effects, and Control, Chicago, IL, Aug 20-23, 2001.
3. Weilert, C. Analysis of ICR Data for Mercury Removal from Wet and Dry FGD. Proceedings of
the U.S. EPA-DOE-EPRI Combined Power Plant Air Pollutant Control Symposium: The MEGA
Symposium and The A&WMA Specialty Conference on Mercury Emissions: Fate, Effects, and
Control, Chicago, IL, Aug 20-23, 2001.
4. Butz, J. R.; Chang, R.; Waugh, E. G.; Jensen, B. K.; Lapatnick, L. N. Use of Sorbents for Air
Toxics Control in a Pilot-scale COHPAC Baghouse. Presented at the 92nd Annual Meeting and
Exhibition of the Air & Waste Management Association, St. Louis, MO, June 20-24, 1999.
5. Helfritch, D. J.; Feldman, P. L.; Pass, S. J. A Circulating Fluid Bed Fine Particulate and Mercury
Control Concept. Proceedings of the EPRI-DOE-EPA Combined Utility Air Pollution Control
Symposium: The MEGA Symposium, Washington, DC, Aug 25-29, 1997.
6. Helfritch, D. J.; Feldman, P.; Waugh, E. The Pilot-scale Testing of a Circulating Fluid Bed for
Mercury Adsorption and Particle Agglomeration. Proceedings of the U.S. EPRI-DOE-EPA
Combined Utility Air Pollution Control Symposium: The MEGA Symposium, Atlanta, GA, Aug
16-20, 1999.
7. Waugh, E. G.; Jensen, B. K.; Lapatnick, L. N.; Gibbons, F. X.; Sjostrom, S.; Ruhl, J.; Slye, R.;
Chang, R. Mercury Control in Utility ESPs and Baghouses Through Dry Carbon-Based Sorbent
Injection - Pilot-scale Demonstration. Proceedings of the U.S. EPA-DOE-EPRI Combined Power
Plant Air Pollutant Control Symposium: The MEGA Symposium, Washington, DC, Aug 1997.
8. Ghorishi, B. S.; Singer, C.; Sedman, C. B. Preparation and Evaluation of Modified Lime and
Silica-Lime Sorbents for Mercury Vapor Emissions Control. Proceedings of the U.S. EPRI-DOE-
3-26
-------
EPA Combined Power Plant Air Pollutant Control Symposium: The MEGA Symposium, Atlanta,
GA, Aug 16-20, 1999.
9. Singer, C.; Ghorishi, B. S.; Sedman, C. B. Lime-based Multipollutant Sorbents. Proceedings of
the U.S. EPA-DOE-EPRI Combined Power Plant Air Pollutant Control Symposium: The MEGA
Symposium and The A&WMA Specialty Conference on Mercury Emissions: Fate, Effects, and
Control, Chicago, IL, Aug 20-23, 2001.
10. Ghorishi, S. B.; Sedman, C. B. Low Concentration Mercury Sorption Mechanisms and Control by
Calcium-based Sorbents: Application to Coal-fired Processes. J. Air Waste Manage. Assoc. 1998,
48, 1191-1198.
11. Laudal, D. L. Mercury Speciation Sampling at Cooperative and United Power Associations, Coal
Creek Station; Final Report; Energy & Environmental Research Center: University of North
Dakota, Jan 1999.
12. Amrhein, G. T.; Holmes, M. J.; Bailey, R. T.; Kudlac, G. A.; Downs, W.; Madden, D. A.
Advanced Emissions Control Development Program, Phase III; Approved Final Report; DOE
Contract DE-FC22-94PC94251, July 1999.
13. Weilert, C. V. Statistical Analyses of FGD System Operating Costs: Update for 1993-1997.
Proceedings of the U.S. EPA/DOE/EPRI Combined Utility Air Pollutant Control Symposium:
The MEGA Symposium, Atlanta, GA, Aug 16-20, 1999.
3.2.1.3 Wet Scrubbers
Wet desulfurization refers to the most widely used SO2 control technology worldwide (approximately 220
GW of installed capacity) and in the United States (approximately 84 GW of installed capacity),1
commonly known as wet scrubbers or wet FGD. Wet scrubbers have been shown to be efficient devices
in capturing oxidized mercury in the flue gas.1"7 This fact has triggered a number of developments geared
towards understanding and promoting the oxidation of elemental mercury in the flue gas of wet-scrubber-
equipped plants as a means of maximizing the mercury capture within the wet scrubber. These efforts
have focused mostly on catalyst-enhanced oxidation and reagent injection approaches for mercury
oxidation. In addition, and as a result of developments in wet precipitator technology (WESP) and the
compatibility of WESPs with wet scrubbers, the wet scrubber-WESP combination represents another
system approach to combined SO2 and mercury capture. Because these various processes are predicated
on well-known, conventional and widely used wet scrubber technology, the basic wet scrubber
technology description is presented in this section to avoid repetition. The "add-on" technology
components are discussed in their respective sections.
Technology Description - Wet Scrubber
The most commonly used wet scrubber technology uses a wet limestone process to remove SO2 from the
flue gas with in-situ forced oxidation to produce a gypsum-grade by-product. This is typically
accomplished in a vertical vessel with flue gas cooling and reaction with limestone slurry to produce a
mix of calcium sulfite and sulfate. Through controlled oxidation of the reaction products, a salable by-
product in the form of commercial grade gypsum (hydrated calcium sulfate) may be produced. The
intimate contact between gas and liquid is ensured through different design approaches, usually involving
several counter flow spray levels and mass transfer "trays" to optimize gas-liquid interactions. The
technology has evolved over the years through "mechanical" improvements, which have included better
gas and liquid distribution within the scrubber, droplet size and size distribution, as well as "chemistry"
improvements such as the addition of organic acids [adipic acid or dibasic organic acid (DBA) used most
often], which not only improve overall SO2 capture but also help the settling characteristics of the waste
3-27
-------
products. Several commercial variations of the technology exist based on reagent type, vessel design, etc.
Because oxidized vapor-phase mercury is water soluble (whereas, elemental mercury is not), the intense
gas-liquid mass transfer in wet scrubbers is potentially an excellent mercury control mechanism.
3.2.1.3.1 Wet Scrubbers with Mercury Oxidation Processes
Wet FGD with Mercury Oxidation Processes Summary
Status
SO2 Reduction, %
NOX Reduction
Hg Reduction, %
Cost
Capital ($/kW)
Fixed O&M ($/kW-yr)
Variable O&M (mills/kWh)
Applicability
Issues
Under development, Pilot-scale testing
95
NA
>80
160 - 275 for a nominal 400 MW plant
1.2-14 (scrubber only)
0.1 -1
Wet Scrubber Plants
Full scale demonstration underway, insufficient information
at present
The two major areas of development underway in the area of mercury oxidation in the flue gas, upstream
of wet scrubbers involve catalytic oxidation5'7 and oxidation resulting from reagent injection.6'8 One other
approach injects chlorine into the flue gas to form HgCl2 upstream of the wet scrubber.
Catalytic Oxidation
This approach involves the deployment of a catalyst in the flue gas (similar to the use of SCR for NOX
control) to oxidize elemental mercury. While catalyst development and testing are at the laboratory scale,
full-scale application would likely involve a conventional support structure (honeycomb) placed between
the particulate control device and the wet scrubber. A number of catalyst materials have been investigated
at several test sites including carbon, palladium, iron, and high carbon fly ash8 with varying degrees of
success. Full-scale application of the technology envisions the catalyst to be exposed to flue gas in an area
of low velocity [~1.5 m/s (5 ft/s)], possibly downstream of the last field of an ESP.5'9
Reagent-based Oxidation
This technology involves the introduction of dedicated reagents into the flue gas or the scrubber itself. In
both cases, the objective is to promote the conversion of elemental mercury to an oxidized form (most
often HgCl2). The flue gas injection approach is expected to promote the conversion of HC1 to C12 in the
flue gas, thereby providing a pathway for the formation of HgCl2 The direct scrubber injection approach
involves the addition of small amounts of a proprietary reagent into the scrubber recirculation system.6
Details are not publicly available at present.
Commercial Readiness and Industry Experience
Catalytic Oxidation
At present, the technology is at laboratory- and pilot-scale development. Laboratory- and pilot-scale tests
funded by the DOE and EPRI have identified several catalyst materials successful in oxidizing elemental
mercury. Further testing of these catalysts has focused on two issues associated with the catalytic
oxidation process: (1) catalyst life and (2) the applicability of the process for the U.S. electric utility
industry.5'7 Results to date suggest that larger scale testing is warranted at this time.
3-28
-------
Reagent-based Oxidation
Reagent tests have been conducted for wet scrubber enhancement at B&W's Clean Environment
Development Facility (CEDF). In the course of several tests, a reagent was found which allowed for over
80 percent mercury removal while having no negative effects on scrubber operation.6 The technology was
later demonstrated at the 55 MW Endicott Station (limestone forced oxidation system). Further tests of
the technology at Zimmer Station (magnesium enhanced lime) demonstrated average Hg removal across
wet FGD of 51 percent.10
Emission Control Performance
Catalytic Oxidation
In laboratory- and pilot-scale tests, several of the catalysts tested exhibited levels of elemental mercury
oxidation in the 70 to 95 percent range. In particular, palladium, carbon, and high-carbon fly ash-based
catalysts exhibited high levels of oxidation. Further, tests to address catalyst longevity, while preliminary,
indicate that the palladium catalyst, and three of the five catalysts tested retained better than 70 percent
oxidation of the inlet elemental mercury at the end of the 5-month test period.5 Palladium catalyst showed
little deactivation after approximately 4,000 hours of operation.
Reagent-based Oxidation
Results from two series of testing indicated that high levels of mercury removal (up to 86 percent) were
repeatedly achieved with small amounts of proprietary reagents with no adverse effects on scrubber
operation or SO2 removal. This is in comparison to baseline (no reagent) removal of mercury across the
scrubber of about 72 percent.10
O&M Impacts
There is not sufficient information at present to assess the impacts of the oxidation process itself. In the
case of catalytic oxidation, the presence of the catalyst could have a minimum impact on flue gas pressure
loss. The wet FGD adds additional pressure drop and increases the auxiliary power consumption by up to
1.5-2.5 percent.
Capital Costs
Wet scrubber: $160-273/kW;n as low as $103/kW reported12
Catalytic oxidation: estimated for 400 MW plant to be about $5/kW5
Reagent Injection: NA
O&M Costs
Wet scrubber
• Fixed O&M: $1.2-14/kW-yr13 (for a range of coal sulfur from 0.4 to about 3.5 percent)
• Variable O&M: 0.1-1 mill/kWh13 (for a range of coal sulfur from 0.4 to about 3.5 percent)
Catalytic oxidation: estimated for 400 MW plant to be about $3.5/kW-yr.5 This is total O&M and includes
the catalyst cost, which is treated as an O&M expense, projected as catalyst replacement every 3 years.
Issues Associated with the Technology; Future Outlook
Catalytic Oxidation
Further pilot-scale testing is needed for periods of 1 to 3 years to more clearly understand catalyst life, as
several catalyst candidates appear to warrant such testing. Further testing of the potential for oxidation of
other flue gas elements, including SO2 and NO, should be carried out. Parallel research to characterize the
stability and fate of mercury in the FGD sludge or gypsum is ongoing.
3-29
-------
Reagent-based Oxidation
Potential for re-emission of elemental mercury from the scrubber's tank, and means to prevent it, are
being investigated.14
References
1. Jozewicz, W.; Smouse, S. Review of Advanced FGD Technologies and Associated Co-Benefits.
Presented at the 21st Annual International Pittsburgh Coal Conference, Osaka, Japan, Sept 13-17,
2004.
2. Bielawski, G. T.; How Low Can We Go? Proceedings of the U.S. EPA-DOE-EPRI Combined
Power Plant Air Pollutant Control Symposium: The MEGA Symposium and The A&WMA
Specialty Conference on Mercury Emissions: Fate, Effects, and Control, Chicago, IL, Aug 20-23,
2001.
3. Fahlke, J.; Bursik, A. Impact of the State-of-the-Art of Flue Gas Cleaning on Mercury Species
Emissions from Coal-Fired Steam Generators. Water, Air, SoilPollut. 1995, 80, 209-215.
4. Licata, A.; Fey, W. Advance Technology to Control Mercury Emissions. Proceedings of the U.S.
EPA-DOE-EPRI Combined Power Plant Air Pollutant Control Symposium: The MEGA
Symposium and The A&WMA Specialty Conference on Mercury Emissions: Fate, Effects, and
Control, Chicago, IL, Aug 20-23, 2001.
5. Blythe, G. M.; Richardson, C. F.; Rhudy, R. G. Catalytic Oxidation of Mercury in Flue Gas for
Enhanced Removal in Wet FGD Systems. Proceedings of the U.S. EPA-DOE-EPRI Combined
Power Plant Air Pollutant Control Symposium: The MEGA Symposium and The A&WMA
Specialty Conference on Mercury Emissions: Fate, Effects, and Control, Chicago, IL, Aug 20-23,
2001.
6. Milobowski, M. G. Wet FGD Enhanced Mercury Control for Coal-Fired Utility Boilers.
Proceedings of the U.S. EPA-DOE-EPRI Combined Power Plant Air Pollutant Control
Symposium: The MEGA Symposium and The A&WMA Specialty Conference on Mercury
Emissions: Fate, Effects, and Control, Chicago, IL, Aug 20-23, 2001.
7. Richardson, C. F. Enhanced Control of Mercury by Wet FGD Systems. Proceedings of the U.S.
EPA/DOE/EPRI Combined Utility Air Pollutant Control Symposium: The MEGA Symposium,
Atlanta, GA, Aug 16-20, 1999.
8. Nolan, P. S. Development of Mercury Emissions Control Technologies for the Power Industry.
Proceedings of the U.S. EPA/DOE/EPRI Combined Utility Air Pollutant Control Symposium:
The MEGA Symposium, Atlanta, GA, Aug 16-20, 1999.
9. Blythe, G.; Richardson, C. F; Rhudy, R. G. Catalytic Oxidation of Mercury for Enhanced
Removal of wet FGD Systems. Proceedings of the U.S. EPA/DOE/EPRI Combined Utility Air
Pollutant Control Symposium: The MEGA Symposium, Chicago, IL, Aug 20-23, 2001.
10. McDermott Technology, Inc. Full-Scale Testing of Enhanced Mercury Control Technologies for
Wet FGD Systems - Final Report; Submitted to the Ohio Coal Development Office, May 2003.
11. Keeth, R. J. Utility Response to Phase I and Phase II Acid Rain Legislation - an Economic
Analysis. Proceedings of the EPRI/DOE/EPA 1995 SO2 Control Symposium, Miami, FL, March
28-31, 1995.
3-30
-------
12. Miller, M; Webster, L.; Rader, P.; Bussell, C. Centralia WFGD System, State-of-the-art SO2
Compliance. Proceedings of the U.S. EPA-DOE-EPRI Combined Power Plant Air Pollutant
Control Symposium: The MEGA Symposium and The A&WMA Specialty Conference on
Mercury Emissions: Fate, Effects, and Control, Chicago, IL, Aug 20-23, 2001.
13. Weilert, C. V. Statistical Analyses of FGD System Operating Costs: Update for 1993-1997.
Proceedings of the U.S. EPA/DOE/EPRI Combined Utility Air Pollutant Control Symposium:
The MEGA Symposium, Atlanta, GA, Aug 16-20, 1999.
14. Chang, J.; Ghoroshi, B. Why Flue Gas Elementary Mercury Concentration Increases Across a
Wet Scrubber. Proceedings of the Combined Utility Air Pollutant Control Symposium: The
MEGA Symposium, Washington, DC, May 19-21, 2003.
3.2.1.3.2 Wet Scrubbers with Wet ESP
Wet FGD with Wet ESP Summary
„. . Commercially available, being tested in power plant
applications
SO2 Reduction, % 99
NOX Reduction NA
Hg Reduction, % Up to 80
Cost
Capital ($/kW) 10 - 25 (WESP only) for a nominal 400 MW plant
Fixed O&M ($/kW-yr) 0.5 - 1.5
Variable O&M (mills/kWh)
Applicability Integration with wet scrubbers, retrofit dry ESPs, new units
. Few applications in power industry, potentially expensive
es alloys required
Technology Description
Wet ESP
Wet Electrostatic Precipitators (WESPs), like dry ESPs, operate in a three-step process: charging of the
entering particles, collection of the particles on the surface of an oppositely charged surface, and finally,
cleaning the collection surface. Both technologies employ separate charging and collection systems.
However, unlike dry ESPs, in the WESP the collecting surface is cleaned with water as opposed to
mechanically. As a result, the two technologies differ in the nature of particles that can be removed, the
overall efficiency of removal, and the design and maintenance parameters.1 While dry ESPs are typically
limited to power levels of 100-500 watts per 0.47 m3/s (1,000 cfm), WESPs can handle power levels as
high as 2,000 watts per 0.47 m3/s (1,000 cfm). As a result, WESPs can handle a wide variety of pollutants
and flue gas conditions and are highly efficient on submicrometer particles and acid mist. WESPs have
also been found to be most efficient in treating flue gases with high moisture content or sticky particulate
matter. As a result of the wet cleansing of the collection system, particulate matter does not accumulate
on the ESP collection electrodes, therefore mitigating particle re-entrainment.1'2
WESPs can be configured for vertical or horizontal gas flows in tubular or plate designs. Tubular designs
offer smaller footprints and, in general, are more efficient than the plate type.
3-31
-------
WESP Integration with Wet Scrubber
WESPs are compatible and easily integrated into a system design with a wet scrubber.3 In fact, integration
of the WESP within the wet scrubber is a design option with many synergisms and attractive features,4
• compact footprint, considering that the scrubber and the WESP could share a common casing,
• ability to integrate the handling of the wash water and solids from the WESP with scrubber slurry,
avoiding the need for separate tank and blowdown system, and
• ability to collect the fine sulfuric acid mist that typically escapes the scrubber due to its very small
droplet size.
Commercial Readiness and Industry Experience
WESPs have been used for almost a century as standard technology in abating the submicrometer particle
SO3 mist in sulfuric acid plants.5 It was until recently, however, a relatively unknown technology to the
electric power industry. An up-flow tubular design WESP has been retrofit at Northern States Power
Company's Sherco Station in a wet scrubber-WESP configuration. In addition, a horizontal flow, plate
WESP system was recently installed at Potomac Electric Power Company's Dickerson Generating
Station, converting an existing dry ESP to hybrid operation by replacing the third field of the existing
ESP to wet operation.1
In 1986, the first commercial WESP application on a U.S. power plant took place when AES Deepwater,
a 155 MW cogeneration plant firing petroleum coke as the primary fuel, was equipped with a WESP. The
other air pollution control equipment included a dry ESP and a wet scrubber. The WESP was installed
mainly for removing relatively high levels of sulfur trioxide. With the WESP in operation, the plume
opacity at the plant is generally 10 percent or less.6 Two more power plant applications are underway
presently: (1) 2.35 m3/s (5,000 cfm) slipstream at the Bruce Mansfield Station and (2) a plate type WESP
for integration with Powerspan's ECO technology to be demonstrated at First Energy's R.E. Burger plant.
The WESP at the Mansfield Station is achieving greater than 95 percent removal of SO3 and PM2 5 and
stack flow with near-zero opacity.7'8
Emission Control Performance
When integrated with upstream technology, including wet scrubbers, multiple pollutants can be removed
by WESPs. A hazardous waste facility fit with a two-stage tubular WESP following a scrubber achieved
99.9 percent removal of acid gases, dioxins, furans, PM25, and metals. It achieved 78 percent removal of
mercury.1'2 At a mining operation, a combined scrubber and WESP system achieved an SO2 removal of
99 percent.1'2
Tests at the Sherco Station (WESP retrofit to the outlet section of the wet scrubber) allowed the scrubber
to maintain a 70 percent SO2 reduction, while keeping particulate emissions at 0.01 Ib/MMBtu and
opacity under 10 percent. Full conversion of all the plant's scrubber modules with WESPs is now
underway.
In pilot scale tests at Southern Research Institute,9 a plate type WESP yielded the following removal
results:
• SO2: 10 - 25 percent
• SO3: ~ 65 percent
• PM: 90 - 99 percent
• Hg: ~ 30 percent
3-32
-------
O&MImpacts
When WESP is integrated with a wet FGD, there are no significant O&M impacts other than auxiliary
power requirements; in this case, the wastewater from the WESP is handled together with the FGD
wastes. Auxiliary power is higher than dry ESPs by a factor as WESP operating power could reach 2,000
Watts per 1,000 cfm, while dry ESP power ranges form 100 to 500 watts per 1,000 cfm.10'11
Capital Costs
According to vendors,12 capital costs are estimated in the range of $10 - 25/kW depending on unit size,
flue gas conditions, and overall design configuration. If special alloys are necessary, due to acid gas
concentrations, the costs could be considerably higher. Non-metal materials are being investigated with
the potential for cost reduction. Note that these costs refer to the WESP only, not the SO2 scrubber.
O&M Costs
According to vendors, there in no appreciable increase in O&M costs. O&M for WESP is estimated to be
less than half when compared to a dry ESP. A general estimate of about $l/kW-yr was provided for fixed
O&M cost including the increased power consumption.
Issues Associated with the Technology; Future Outlook
Several conditions determine the efficiency of a WESP system and should be considered in each specific
design. These include air distribution, sparking, and corona current suppression. Integration with wet
scrubbers can offer significant advantages to wet scrubber operations, specifically in sulfuric acid mist
control. Material performance will be key to overall cost of the technology. Expensive alloys may reduce
market appeal.
References
1. Altaian, R. Wet Electrostatic Precipitation Demonstrating Promise for Fine Particulate Control:
Croll-Reynolds Clean Air Technologies. Power Eng. Jan 2001.
2. Altaian, R. Wet Electrostatic Precipitation Demonstrating Multiple Pollutant Control in Industrial
Applications Holds Promise for Coal-Fired Utility Emission Reduction of Acid Mist, PM2s, and
Mercury. Proceedings of the U.S. EPA-DOE-EPRI Combined Power Plant Air Pollutant Control
Symposium: The MEGA Symposium and The A&WMA Specialty Conference on Mercury
Emissions: Fate, Effects, and Control, Chicago, IL, Aug 20-23, 2001.
3. Buckley, W.P.; Altshuler, B. Sulfuric Acid Mist Generation in Utility Boiler Flue Gas. Presented
at the ICAC Forum'03, Nashville, TN, Oct 2003.
4. Bielawski, G. T. How Low Can We Go? Proceedings of the U.S. EPA-DOE-EPRI Combined
Power Plant Air Pollutant Control Symposium: The MEGA Symposium and The A&WMA
Specialty Conference on Mercury Emissions: Fate, Effects, and Control, Chicago, IL, Aug 20-23,
2001.
5. Staehle, R. C.; Triscori, R. J.; Ross, G.; Kumar, S.; Cothon, R. Wet Electrostatic Precipitators for
High Efficiency Control of Fine Particulates and Sulfuric Acid Mist. Presented at the ICAC
Forum'03, Nashville, TN, Oct 2003.
6. Kumar, K. S.; Mansour, A. Wet ESP for Controlling Sulfuric Acid Plume Following an SCR
System. Presented at the ICAC Forum 2002, Houston, TX, Feb 12-13, 2002.
7. Srivastava, R.; Miller, A.; Erickson, C.; Jambhekar, R. Emission of Sulfur Trioxide from Coal-
Fired Power Plants. Presented at PowerGen 2002, Orlando, FL, Dec 2002.
3-33
-------
8. Reynolds, J.; Buckley, W.; Ray, I. Advances in Wet Electrostatic Precipitation Technology at
Fossil Fuel Power Plants for Multipollutant Control. Presented at the ICAC Forum'03, Nashville,
TN, Oct2003.
9. Monroe, L. S. Testing of a Combined Dry and Wet Electrostatic Precipitator for Control of Fine
Particle Emissions from a Coal-Fired Boiler. Proceedings of the U.S. EPA/DOE/EPRI Combined
Power Plant Air Pollution Control Symposium: The MEGA Symposium, Washington, DC, Aug
25-29, 1997.
10. Altaian, R.; Offen, G.; Buckley, W.; Isaac, R. Wet Electrostatic Precipitation/Demonstrating
Promise for Fine Particulate Control - Part 1. Power Eng. Magazine Jan 2001.
11. Altaian, R.; Buckley, W.; Isaac, R. Wet Electrostatic Precipitation/Demonstrating Multiple
Pollutant Control in Industrial Applications Holds Promise for Coal-fired Utility Emission
Reduction of Mist, PM25 and Mercury. Proceedings of the U.S. EPA/DOE/EPRI Combined
Utility Air Pollutant Control Symposium: The MEGA Symposium, Chicago, IL, Aug 20-23,
2001.
12. Croll-Reynolds Clean Air Technologies. Air Pollution Control Systems; Brochure 5M798; 1998.
3.2.1.3.3 Plasma-Enhanced ESP (PEESP)
PEESP Summary
Status Bench scale
SO2 Reduction, % (>90 with FGD)
NOX Reduction, % NA
Hg Reduction, % Up to 90
Cost
Capital ($/kW) < $5/kW for retrofit of Wet ESP
Fixed O&M ($/kW-yr) NA
Variable O&M (mills/kWh) NA
Applicability Both new and retrofit applications
. Early stage of development; demonstration and further
assessment of the technology is needed
Technology Description
Wet Electrostatic Precipitators (Wet ESP) have demonstrated that they can remove multiple pollutants.
For example, pilot scale dry-wet ESP (addition of a wet ESP field in a dry ESP without PEESP) funded
by the Electric Power Research Institute (EPRI)1 in a 1995 demonstrated the following removals: 95
percent particulate, 20 percent SO2, 35 percent hydrogen chloride, 45 percent hydrogen fluoride, and 50
percent oxidized mercury. In 2001, a wet ESP pilot unit funded by Croll-Reynolds and First Energy
demonstrated greater than 90 percent removal of PM25 and SO3 mist in a two-field electrical
configuration2 with the ESP after an FGD system. In addition, 40 percent removal of elemental mercury
and greater than 70 percent removal of particulate and oxidized mercury were achieved in a single
electrical field configuration.
3-34
-------
The ability of Wet ESP to remove elemental mercury can be enhanced further through Plasma-Enhanced
ESP (PEESP) technology. PEESP oxidizes vapor phase elemental mercury into oxidized form and then
removes it within the Wet ESP process. This technology involves injection of a reagent gas mixture,
through a corona discharge needle that is attached to the central electrode within an electrostatic field.
Injection 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 inject the
reagent gas.
Commercial Readiness and Industry Experience
The technology has been tested at bench scale. Scale-up to pilot scale size of 5,000 actual cubic feet per
minute (ACFM) is planned and testing will commence later in 2004.
Emission Control Performance
The supplier projects that up to 90 percent total mercury removal can be achieved at pilot and full scale.
Bench scale testing3 demonstrated mercury removal efficiencies up to 83 percent.
O&MImpacts
O&M impacts are not known, mainly because of the lack of information, but are expected to be minimal
since the PEESP technology is a passive device retrofitted within a wet ESP. There are no significant
impact expected on pressure drop and performance of upstream equipment.
Costs
Capital costs for Wet ESPs are estimated at $20-$35 per kW depending upon the difficulty of the
installation at a site. The cost to retrofit the PEESP technology within a wet ESP is less than $5 per kW,
as it requires only modification of the central discharge electrode.
O&M costs are not available presently because of the technology is at an early stage of development.
Issues Associated with the Technology; Future Outlook
There has been limited mercury removal experience with Wet ESPs. What has been reported shows some
oxidation of elemental mercury and capture of particulate and oxidized forms of mercury similar to that
for PM2 5.4 The PEESP technology seeks to enhance oxidation of elemental mercury to improve total
mercury removal. The technology is still at an early development stage (bench scale towards pilot plant)
and requires further demonstration and techno-economic assessment to develop a more comprehensive
picture of its cost-effectiveness. Additionally, further investigation is needed regarding the water
chemistry within the wet ESP to keep the oxidized mercury from being reduced back to elemental
mercury.
References
1. Altaian, R.; Buckley, W.; Ray, I. Multi-Pollutant Control with Dry-Wet Hybrid ESP Technology.
Proceedings of the Combined Utility Air Pollutant Control Symposium: The MEGA Symposium,
Washington, DC, May 19-21, 2003.
2. Reynolds, J. Membrane-Based Wet ESP Technology; First Quarterly Report dated 12/31/02;
funded by DOE's NETL office.
3-35
-------
3. Montgomery, J. L.; Battelson, D. M.; Whitworth, C. G.; Ray, I.; Buckley, W.; Reynolds, J.;
Altman, R. Plasma-Enhanced Electrostatic Precipitation Technology for Control of Elemental
Mercury. Presented at Power-Gen 2002, Orlando, FL, Dec 2002.
4. Wet Electrostatic Precipitation Demonstrating Promise for Fine Particulate Control. Power Eng.
Magazine Jan 2001.
5. 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 Offgas. Proceedings of the Combined Power Plant Air Pollutant Control Symposium:
The MEGA Symposium, Washington DC, May 19-22, 2003.
6. Battleson, D. Latest Developments of the Plasma-Enhanced ESP for Mercury Removal in Offgas.
Proceedings of the Combined Utility Air Pollutant Control Symposium: The MEGA Symposium,
Washington, DC, May 19-21, 2003.
3.2.1.4 MerCAP
MerCAP Process Summary
Status Early pilot plant testing
SO2 Reduction, % Existing scrubber
NOX Reduction, % NA
Hg Reduction >80% for 10 ft long plates spaced 0.5 inches apart
Cost
Capital ($/kW) 18.8 $/kWfor a 250 MW plant
Fixed O&M ($/kW-yr) 0.46 - 0.87 mills/kWh
Variable O&M (mills/kWh) 0.19 mills/kWh
Applicability New power plants and retrofits
Issues Did not perform well in unscrubbed gas
Mercury Control via Adsorption Process (MerCAP) concept is to place fixed sorbent-coated structures
into a flue gas stream at temperature below 400 °F to adsorb mercury and then periodically regenerate
them and recover the captured mercury. Slip-stream field tests of this concept were conducted at four
different sites with the in-situ probes with various spacing and gold coatings (different thickness and
support plates). The in-situ probes were designed and fabricated to allow testing of full-length (10-foot
long) plates. Table 3-5 summarizes the test conditions and results at the four sites. The sites included two
PRB units, one equipped with an ESP (probe located downstream of the ESP) and one with a wet
particulate scrubber (WPS) (probe located upstream of the WPS in full dust loading) and two North
Dakota lignite units, one equipped with an ESP (probe located downstream of ESP) and one with a spray
dryer-baghouse (probe located downstream of spray dryer-baghouse).1'2
3-36
-------
Table 3-5. Summary of MerCAP Field Sites
Site
ID
S2
S9
L1
L2
Plant
Pleasant
Prairie
Laskin
Coal
Creek
Stanton
Coal Type
Subbituminous
Subbituminous
ND Lignite
ND Lignite
Fuel Composition
S (%)
0.5
0.4
1
0.06
Cl
(ppm)
14
<50
20
30
Test
Location
ESP
Outlet
WPS Inlet
ESP
Outlet
FF Outlet
HCIat
Test
Location
(est. ppm)
< 10
0.7
<10
<10
SO2at
Test
Location
(ppm)
280-340
200-400
1000
100
Initial
Mercury
removal,
%
0-24
20
10
55-89
Two long-term slip-stream tests are planned for a period August 2004 through January 2005. At Great
River Energy's Stanton Station, which burns North Dakota lignite, sorbent structures will be retrofitted
into a single compartment in the Unit 10 baghouse enabling reaction with a 6 MW equivalence of flue
gas. At Southern Company Services' Plant Yates, which burns Eastern bituminous coal, gold-coated
plates will be configured as a mist eliminator located downstream of a 1 MW pilot wet absorber, which
receives flue gas from Unit I.1
Additional tests will evaluate the ability to thermally regenerate the gold-coated plates. The results of this
study will provide data required for assessing the feasibility and estimating the costs of a full-scale
MerCAP process for flue gas mercury removal. The study will provide information about optimal
operating conditions for different flue gas conditions, the effectiveness of sorbent regeneration, and the
ability of the gold sorbent to hold up to flue gas over an extended period. In addition, if successful, the
novel approach of incorporating MerCAP structures in existing baghouse compartments will demonstrate
a cost-effective means for achieving mercury control using existing baghouse technologies.
Commercial Readiness and Industry Experience
The technology has been tested at pilot scale.
Emission Control Performance
Tests indicate that given the right flue gas conditions, MerCAP with gold coated plates around 10 ft long
and spaced 0.5 inches apart can remove more than 80 percent of mercury. In the tests conducted to date in
flue gas derived from low rank fuels, this performance was achieved downstream of a spray dryer-
baghouse. Similar results have recently been obtained in other tests conducted downstream of a wet
absorber in bituminous-derived flue gas. Developer claims that higher removals (>90 percent) for the
short term should be achievable by increasing plate length, decreasing plate spacing, or reducing gas
velocities.2
O&M Impacts
A factor that may reduce cost is the use of sorbent material that is cheaper or has higher mercury
capacities than gold. Development work is needed to evaluate these options to reduce costs as well as to
establish sorbent capacity, regeneration frequency, and sorbent life expectancy for the range of power
plant configurations and coals burned. Projected additional pressure drop is less than 2" of H2O for plate-
to-plate spacing of 1-inch and a gas velocity of 60 ft/s.
Costs
The capital costs for a MerCAP system with 90 percent mercury control, 3-month regeneration, and 100
percent redundancy is estimated at $4.7 million for a 250 MW unit (this corresponds to $18.8/kW). Of the
3-37
-------
capital cost, $2.3 million is the gold media and its substrate. For 1-year regeneration and lower mercury
capacity (10 percent compared to 15 percent of the weight of gold), the costs are $14.9 million for a 250
MW unit (this corresponds to $59/kW) with $12.6 million in gold media and substrate cost.2
O&M costs are estimated by a preliminary engineering economic study to range from 0.46 to 0.87
mills/kWh and 0.19 mills/kWh for fixed and variable component, respectively. The amount and cost of
the sorbent media is the major cost component for a MerCAP installation.2
Issues Associated with the Technology; Future Outlook
MerCAP with gold as the sorbent surface did not perform well in non-scrubbed flue gas. Mercury
removal effectiveness, although low in non-scrubbed gases, also did not appear to degrade over time in
flue gas. It is uncertain at this point whether the low effectiveness is due to specific flue gas components,
which reduced the gold capacity for mercury in the gases tested, or due to an effect of temperature, or to a
combination of both. Sorbent materials other than gold are also being tested as alternate coatings. Some of
the materials may be more effective or offer a cost-effective alternative to gold for specific flue gas
conditions.2
References
1. Control Technology Evaluation of MerCAP™ for Power Plant Mercury Control.
http://www.netl.doe.gov/coal/E&WR/mercury/control-tech/mercap.html (accessed Dec 1, 2004).
2. Sjostrom, S.; Chang, R.; Strohfus, M.; Johnson, D.; Hagley, T.; Ebner, T.; Slye, R.; Richardson,
C.; Belba, V. Development and Demonstration of Mercury Control by Adsorption Process
(MerCAP™). Presented at the MEGA Symposium, Washington, DC, May 2003.
3.2.2 SO2 and NOX Control
3.2.2.1 Electron Beam Process
Electron Beam Process Summary
Status
SO2 Reduction, %
NOX Reduction, %
Hg Reduction
Cost
Capital ($/kW)
Fixed O&M ($/kW-yr)
Variable O&M (mills/kWh)
Applicability
Issues
Early commercialization stage
>95
Up to 90
NA
180-250 for 100-300 MW
NA
NA
New power plants and retrofits
Demonstration is required. High auxiliary power requirement
is the main barrier.
Technology Description
The electron beam (E-beam) process is capable of removing simultaneously SO2 and NOX and involves
cooling of the flue gas, injection of ammonia, and then irradiation by high-energy electrons.1"5 As Figure
3-3 shows, the E-beam process equipment is placed after the ESP. The first component is an evaporative
spray cooler, where the flue gas is cooled to 60-66 °C (140-150 °F). The spray cooler is operated with a
3-38
-------
dry bottom (all the water injected into the flue gas is evaporated). In addition, gaseous ammonia is
injected into the flue gas either before or after the spray cooler.
The main component of the E-beam process is a chamber where the flue gas is irradiated by a beam of
high-energy electrons, while water is added to counteract the temperature rise. The irradiation also
generates hydroxyl radicals and oxygen atoms, which oxidize the SO2 and NOX. These oxidized species
mix with water in the flue gas to form sulfuric acid and nitric acid, which are neutralized by the ammonia.
The by-products of the E-beam process are solid ammonium sulfate and ammonium sulfate-nitrate, which
are collected downstream of the E-beam chamber by an ESP or a baghouse, and can be used as fertilizer
after processing into a granular product. The by-product particles are small and sticky and pose some
problems to both ESPs and FFs. A combination of ESP-FF has been proposed as more effective. In
addition, use of inert materials has been considered to make it easier to clean the bags of the baghouse.
BOILER
FLYASH
COLLECTOR
AIRHEATER
a
FAN
FLYASH
HANDLING
1
FAN
STACK
/?=*
FAN
w
BY-PRODUCT
COLLECTOR
Figure 3-3. E-Beam process schematic.
Commercial Readiness and Industry Experience
The technology is in an early commercialization stage with a number of demonstration plants in
operation, but no operating commercial applications. Ebara International Corp. is actively promoting this
technology, but others are preparing to offer related systems (primarily different methods for producing
the ionization energy). The main reason for limited acceptance of the technology is that it is more
expensive (relative to other competing options such as FGDs) as an SO2 removal process only. Future
market acceptance of this technology will depend on whether it is cheaper than the combination of
individual SO2 and NOX controls for the levels of SO2 and NOX emissions required. Another major factor
3-39
-------
will be the price that can be obtained for the fertilizer by-product. Should it be possible to sell the
fertilizer by-product, this revenue stream would possibly help partially offset O&M costs.
Pilot-scale testing was conducted at Indianapolis Power & Light's Stout station in an 8 MW slipstream in
1986. Similarly, the process was tested in 1992 at Chubu Electric Power Company's Nishi-Nagoya plant
on a slipstream (12,000 Nm3/h) of the 200 MW boiler firing Australian coal. The emission reductions
achieved were: 92 percent SO2 and 60 percent NOX. After the successful operation of this plant, Chubu
decided to install the E-beam process in the Nishi-Nagoya 1 unit, which burns high-sulfur residual oil.
Also, the E-beam process was demonstrated at the Chengdou Power Plant in Sichuan Province, China.
The plant is 90- MW and has been operating since July 1997. SO2 removals have been in the mid-80
percent range and NOX removal was between 15 and 20 percent using low-to-high-sulfur bituminous coal
(600 - 2500 ppm inlet SO2).6
Emission Control Performance
The E-beam process is capable of achieving SO2 removals of 95 percent or greater and NOX removals of
about 90 percent. High SO2 removals require a minimal E-beam dose, generally much lower than the E-
beam dose necessary for NOX removal. Once the minimum E-beam dose is achieved, the primary factors
affecting SO2 removal are flue gas temperature and ammonia stoichiometry. The E-beam dosage required
for 90 percent SO2 removal is a minimum of 1.0 mrad.
The removal of NOX depends primarily on the E-beam dosage; temperature and SO2 concentration are of
secondary importance. Higher NOX removals require higher radiation dosages. A dose of about 0.3-0.6
mrad is required to achieve 50 percent NOX removal, and 90 percent NOX removal requires at least 2.7
mrad according to the data obtained to date.7'8 Higher NOX removals are obtained at higher temperatures,
contrary to SO2 removal. Higher SO2 concentrations also improve NOX removal, making the process
better suited for high-sulfur applications. Furthermore, additional particulates are removed in the new
ESP or FF.
O&M Impacts
The energy requirement for E-beam depends greatly on the NOX reduction being sought; when significant
NOX reduction is not required, the auxiliary power for the E-beam process may range from 2 to 3 percent
of the total plant output. The Chengdou demonstration plant consumed about 2 percent of the plant energy
for an SO2 removal of 80 percent and NOX removal of 10 percent. When NOX reduction is above 60
percent, the auxiliary power may reach 5 percent.
Particulate collection by either ESP or baghouse can be problematic if appropriate steps are not taken to
overcome the stickiness of the particulates.
Capital Costs
Ebara projects capital costs for a 100-300 MW plant in the $180-250/kW range.1
O&M Costs
The main O&M costs are related to the auxiliary power requirements and the ammonia injection. Early
studies project that the total O&M costs are 11-15 mills/kWh resulting in levelized costs (including cost
of capital) of 16 to 22 mills/kWh.1
These costs do not include capital and O&M costs associated with by-product treatment, such as
granulation, or revenue from the sale of the by-products. A concept has been proposed for the fertilizer
production companies to provide the ammonia needed to the power plant and receive the "upgraded"
solid nitrogen granular fertilizer (net credit of $1 million per year or 3-4 mills/kWh). Ebara claims that
3-40
-------
this concept would result in significant savings, but detailed cost estimates have not been developed and it
has not been applied anywhere.
Issues Associated with the Technology; Future Outlook
The technology has yet to reach commercial stage and there is uncertainty regarding its cost and long-
term reliability. Also, the process currently uses a significant amount of electricity, up to 5 percent of the
plant output depending on the required NOX removal.
References
1. Hirano, S.; Aoki, S.; Izutsu, M.; Yuki, Y. Simultaneous SO2, SO3, and NOX Removal by
Commercial Application of the EBARA Process. Proceedings of the EPRI-DOE-EPA Combined
Utility Air Pollution Control Symposium: The MEGA Symposium, Atlanta, GA, Aug 16-20,
1999.
2. Chmielewski, A. G.; Dobrowolski, A.; Iller, E.; Tyminski, B.; Zakrzewska-Trznadel, G.; Zimek,
Z.; Licki, J. Development and Application Experience with Technology of SO2 and NOX
Removal from Flue Gas by Electron Beam Irradiation. Proceedings of the EPRI-DOE-EPA
Combined Utility Air Pollution Control Symposium: The MEGA Symposium, Atlanta, GA, Aug
16-20, 1999.
3. Saku, S. Electron Beam Flue Gas Treatment System. Presented at the Asia-Pacific Workshop on
Water and Air Treatment by Advanced Oxidation Technologies, Ibaraki, Japan, Dec 1998.
4. Maezawa, A.; lizuka, Y. Electron Beam Flue Gas Treatment Process Technology. Presented at
the International Congress of Acid Snow and Rain, Niigata, Japan, Oct 1997.
5. Izutsu, M.; Okabe, R. Final Development to Commercialize EBARA Process. Presented at the
Second Symposium on Non-thermal Plasma Technology for Pollution Control, Brazil, Oct 1997.
6. Technology Assessment of Clean Coal Technologies for China. Electric Power Production;
World Bank ESMAP and AESEG Technical Paper Oil, May 2001.
7. FGD Handbook. International Energy Agency Coal Research, London, 1994.
http://www.caer.uky.edu/iea/ieacr65.shtml (accessed Dec 1, 2004)
8. Frank, N. W.; Hirano, S. Combined SO2/NOX Removal by Electron-beam Processing. Presented
at the 4th Symposium on Integrated Environmental Control, Washington, DC, March 2-4, 1988.
3-41
-------
3.2.2.2 ROFA-ROTAMIX (Mobotec)
ROFA-ROTAMIX Summary
Status
SO2 Reduction, %
NOX Reduction, %
Hg Reduction, %
Cost
Capital ($/kW)
Fixed O&M ($/kW-yr)
Variable O&M (mills/kWh)
Applicability
Issues
ROFA and ROTAMIX commercially available for NOX
reduction; ROTAMIX demonstrated on a 150 MW coal-fired
unit
69 with Trona, 64 with limestone
40 - 60 with ROFA;
up to 80 with ROTAMIX
Additional removal with Trona (11) or limestone (4)
67 with Trona, 89 with limestone
ROFA: $20-25/kW for 100 and 200 MW plants
ROTAMIX: $25-30/kW for units in the 100 - 200 MW range
0.5-1.0
Depends on the use of ammonia or SO2 control absorbent
Existing plants, especially older units less than 300 MW
ROFA has been demonstrated up to 175 MW scale. Also,
the potential increases in boiler slagging and fouling and
erosion of boiler tube surfaces with the injection of sorbents
for SO2 and Hg control have not been resolved.
Technology Description
Mobotec offers three NOX control technologies: (1) the Rotating Opposed Fire Air (ROFA) system, (2)
the ROTAMIX, and (3) in-duct SCR that may be used in conjunction with the previous two systems. This
report includes description of the ROFA and ROTAMIX. According to the vendor, these two
technologies are generally designed as one system; therefore, they are considered a multi-emission control
option. The in-duct SCR is an independent NOX control option.
ROFA System
The ROFA system incorporates asymmetrically placed ROFA air nozzles (in concept similar to overfire
air), which contribute to mixing of the flue gas in the furnace by introducing turbulence. One or more
levels of nozzles are placed asymmetrically in opposite walls, usually higher than those used in
conventional overfire air systems. The ROFA system is designed for 20-40 percent of the total airflow
and uses a booster fan to provide the required pressure and achieve adequate air penetration into the
furnace.1
ROTAMIX
ROTAMIX is used in conjunction with ROFA and involves injection of various chemicals or additional
fuel (for example, reburn gas) to remove NOX, SO2, and heavy metals (including mercury). In addition to
the ROFA ports, lances are provided to inject the chemicals. So far, Mobotec has used urea and ammonia
for NOX reduction, CaCO3 for SO2 control, and is considering adsorbents for mercury control.
Computational fluid dynamic (CFD) modeling is used to take into account site-specific considerations,
determine the optimal design characteristics for each boiler, and determine the placement of the ROFA
boxes.
3-42
-------
Commercial Readiness and Industry Experience
The ROFA-ROTAMIX system is considered commercial. There are 18 installations in the U.S. and 17 in
Sweden (mainly stoker-fired, fluidized-bed, and dry-bottom boilers burning coal, heavy fuel oil, wood
wastes, and municipal waste). In the U.S., ROFA-ROTAMIX has been recently installed at Carolina
Power & Light's (CP&L) Cape Fear 5 and 6, T-fired boilers burning pulverized eastern bituminous coal.
Cape Fear 5 is a single furnace boiler generating 154 MW. Cape Fear 6 is a twin furnace generating 172
MW.2 In Sweden, the combination of ROFA-ROTAMIX has been installed in a 78 MW coal-fired stoker
boiler (Jordberga plant of Danisco Sugar Ltd.). In the Jordberga plant, a 90 percent reduction of SO2 was
achieved using CaCO3.
Emission Control Performance
The 17 ROFA retrofits in Sweden (mainly stoker-fired, fluidized-bed, and dry-bottom boilers burning
coal, heavy fuel oil, wood wastes, and municipal waste) achieved an average of 50 percent NOX reduction
with ROFA, while maintaining CO emissions and reducing excess O2.
As Figure 3-4 shows, CP&L's Cape Fear 5, a T-fired boiler burning pulverized eastern bituminous coal
and generating 154 MW of power, achieved 55 percent NOX reduction at full load and up to 69 percent at
low loads. Also, the level of excess oxygen was reduced from 5.5 to 3.7 percent at full load resulting in
higher boiler and plant efficiency. This reduction of the excess oxygen did not adversely affect CO
emissions. They were also reduced from 50-90 ppm to less than 20 ppm. Adverse impacts included a
small increase in unburned carbon in the fly ash or LOI (from 3 to 5 percent), approximately 11 °C
(20 °F) superheat and reheat temperature reduction and increased auxiliary power due to the high-volume
ROFA fan [820 kW (1100 HP)].
The combination of ROFA-ROTAMIX at Jordberga (78 MW coal-fired stoker boiler in Sweden)
achieved 40 percent NOX reduction and 90 percent SO2 reduction.1 The same performance (especially
with regard to the SO2 reduction) has not been confirmed in pulverized coal boilers, which operate at
significantly higher temperatures than stokers.
The combination of ROFA-ROTAMIX has been installed in CP&L's 154 MW coal-fired Unit 5 at the
Cape Fear Generating Station for the multipollutant technology evaluation program. The program
included injection of CaCO3 to determine the effect on SO2 and Hg reduction and injection of tronato
determine the effect on NOX, SO2, HC1, and Hg reduction. SO2 reductions of 69 percent were achieved
with trona and 64 percent with CaCO3. Mercury reductions of 89 percent were achieved with limestone
and 67 percent with Trona.3"6
Key factors affecting the performance of the ROFA and ROTAMIX systems are:
• height between top burner and furnace outlet which determines whether the ROFA system can be
accommodated and there is adequate residence time for complete combustion,
• furnace dimensions, which will impact the degree of penetration of the air into the furnace, and
• temperature at the top of the boiler.
3-43
-------
0.8 n
-i—•
GO
c
o
'w
E
LJJ
X
O
0.5
0.3
0.2
0.1
After ROFA
40
60
80
100
120
140
160
180
Gross Electrical Power, MWe
Figure 3-4. NOX emissions from Cape Fear Unit 5: comparison before and after ROFA.
O&MImpacts
The limited experience with ROFA systems in the U.S. indicates that there are positive and negative
O&M impacts. On the positive side, combustion efficiency and CO emissions have improved. Also,
excess oxygen has been reduced resulting in higher boiler and plant efficiency. On the negative side,
reduction in superheat and reheat temperatures have been experienced. At Cape Fear, outlet steam
temperatures declined by 11 °C (20 °F) (an equivalent of 0.40 percent reduction in plant efficiency).
However, the O&M impacts are very site-specific (usually depend on the boiler design, operating
condition, and fuel characteristics), and one site is not adequate to generalize the potential impacts of
ROFA for coal-fired boilers.
Impact that needs to be taken into account is the increased auxiliary power due to the high-volume ROFA
fan, estimated (by the vendor) to be approximately 0.3 percent of the plant output. A 150- MW boiler
requires a fan with a 450-750 kW (600-1,000 HP) motor.
There will be a similar increase in auxiliary power due to ROFA-ROTAMIX, plus there will be the use of
chemicals such as ammonia, urea, or CaCO3.
Costs
Typical costs for ROFA average around $20 - 25/kW for boilers between 100 and 200 MW.2 The retrofits
take approximately 8 months from order to final installation. A 2-week outage is expected for installation
of the system.
Capital costs of ROFA-ROTAMIX range between $25 and 30/kW for boilers between 100 and 200 MW.
The installation costs ($/kW) are projected to be lower for larger boilers.
3-44
-------
Issues Associated with the Technology; Future Outlook
The performance of ROFA in large boilers may degrade, as the boiler dimensions increase and either the
air penetration will be less efficient or a more powerful fan would be required to achieve the same
efficiency. CFD modeling completed for boiler sizes up to 550 MW indicate satisfactory performance.
Presently, ROFA has been demonstrated up to the 172 MW scale. Slagging of the superheater by sorbent
(CaCO3 and Trona) and ash was found to be a problem during the Unit 5 Cape Fear Generating Station
tests. Also, the effect of long-term injection of such sorbents on boiler tube erosion is not known.
References
1. Moberg, G.; Blid, J. O.; Wallin, S.; Fareid, T.; Ralston, J. Combined DeNOx/DeSOx and
Additional NOX Reduction by Cleaning Flue Gas Condensate from Ammonia. Presented at the
PowerGen International Conference, New Orleans, LA, Nov 30 - Dec 2, 1999.
2. MobotecUSA - Installations, http://www.mobotecusa.com/projects/installations.htm (accessed
Dec 1, 2004).
3. Haddad, E.; Shilling, M. Installation of Combined Combustion and Post Combustion NOX
Control Technologies at Cape Fear Station and the Future Application of the System for
Mitigation of Other Pollutants. Proceedings of the US EPA/DOE/EPRI Combined Utility Air
Pollutant Control Symposium: The MEGA Symposium, Chicago, IL, Aug 20-23, 2001.
4. Ralston, J.; Haddad, E.; Moberg, G. Full-scale Evaluation of Various Sorbents to Reduce Multi-
Pollutants: NOX, SO2 and Mercury. Proceedings of the Combined Utility Air Pollutant Control
Symposium: The MEGA Symposium, Washington, DC, May 19-21, 2003.
5. Ralston, J.; Haddad, E.; Green, G.; Castagnero, S.; Full-scale Evaluation of Multipollutant
Reduction Technology. Proceedings of the ICAC Forum'03, Nashville, TN, Oct 2003.
6. Haddad, E.; Ralston, J.; Green, G.; Castagnero, S. Full-scale Evaluation of Multipollutant
Reduction Technology. Proceedings of the MEGA Symposium, Washington, DC, May 2003.
3.2.2.3 SNOX
TM
SNOX Summary
Status
SO2 Reduction, %
NOX Reduction, %
Hg Reduction, %
Cost
Capital ($/kW)
Fixed O&M ($/kW-yr)
Variable O&M (mills/kWh)
Applicability
Issues
Early commercial stage; two commercial plants in operation
in Europe
Above 90
Above 90
Zero
270 - 315 for 300 - 500 MW plant
NA
NA
Both new and retrofit applications
Cost-effectiveness
3-45
-------
Technology Description
As shown in Figure 3-5, the SNOX process involves removing participates from the flue gas that leaves
the boiler in a high-efficiency fabric filter (baghouse) to minimize the cleaning frequency of the sulfuric
acid catalyst in the downstream SO2 converter.1"3 The ash-free gas is reheated to 725-752 °F (385-400 °C)
mainly in a gas to gas heat exchanger, and NOX is reduced with small quantities of ammonia in the first of
two catalytic reactors where the NOX is converted to nitrogen and water vapor. If the heat available in the
gas stream is not adequate, supplemental heating maybe used (see preheat burner in front of the baghouse
or the SO3 reactor). The SO2 is oxidized to SO3 in a second catalytic converter (SO3 Reactor). Then, the
gas passes through a novel glass-tube condenser (WSA: Wet gas Sulfuric Acid condenser) that allows the
SO3 to hydrolyze to concentrated sulfuric acid.
Because the SO2 catalyst follows the NOX catalyst, any unreacted ammonia (slip) is oxidized in the SO2
catalyst to nitrogen and water vapor. According to the supplier, downstream fouling by ammonia
compounds is eliminated, permitting operation at higher than normal stoichiometry. These higher
stoichiometries allow smaller catalyst volumes and higher reduction efficiencies.
Miles
Boiler number 2
Existing
Facilities
A
Ammonia Tank
Acid Storage
Tank 2K-U73C7
Figure 3-5. SNOX process schematic.
Commercial Readiness and Industry Experience
The technology is in an early commercial stage with five demonstration projects completed in the early -
and mid-1990s.4"7 The first commercial plant started operation in 1999. Initially, the technology was
tested at a pilot plant of Elsam (Danish utility) in a flue gas stream of 10,000 Nm3/h. From 1991 to 1993,
the following demonstrations took place:
3-46
-------
• Gela (Italy) of Enichem SpA; Gas: 100,000 Nm3/h with 600-4,000 ppm SO2 and 500 ppm NOX in the
inlet
• Vodskov (Denmark) of ELSAM; Gas: 1,000,000 Nm3/h (approximately 305 MW) with 200-2,000
ppm SO2 and 550 ppm NOX in the inlet
• Niles (USA) of Ohio Edison; Gas: 132,000 Nm3/h with 2,000 ppm SO2 and 500-700 ppm NOX in the
inlet
• Vresova (Czech Republic) of Sokolovska Uhelna A.G.; Gas: 54,000 Nm3/h with 2.4-3.6 H2S and 200
ppm NOX in the inlet
• Kawasaki plant (Japan) of Asahi Chemical Industry Ltd; Gas: 50,000 Nm3/h with 5.7 percent SO2 and
100 ppm NOX in the inlet
The U.S. demonstration was conducted at Ohio Edison's Niles Station in Niles, OH in a 35- MW
equivalent slipstream of flue gas from the 108- MW Unit No. 2 boiler, which burned a 3.4 percent sulfur
Ohio coal.
In addition to the Vodskov plant, which has been operating for several years, a new commercial size plant
was built and started operation in 1999 at AGIP Petroli SAP's Gela plant in Italy. It treats 1,000,000
Nm3/h of flue gas from a petroleum-coke-firing facility having 2,380 - 4,600 ppm SO2 and 330 ppm NOX
in the inlet of the SNOX process.3 The Gela plant is designed for 94 percent SO2 removal and 90 percent
NOX removal.
Emission Control Performance
SO2 and NOX removal efficiencies above 90 percent have been achieved. At Ohio Edison's Niles
demonstration, SO2 removal was approximately 94 percent with inlet concentrations averaging about
2,000 ppm SO2 and NOX removal was 95 percent with inlet NOX in the 500 - 700 ppm range.4 In addition,
particulates are expected to be reduced, relative to the original system with ESP.
SO2 removal is controlled by the efficiency of the SO2-to-SO3 oxidation, which occurs as the flue gas
passes through the oxidation catalyst beds. Control is accomplished through the space velocity and bed
temperature. Space velocity governs the amount of catalyst necessary at design flue gas flow conditions,
and the gas and bed temperatures have to be high enough to activate the SO2 oxidation reaction.
Particulates are very low, usually below 1 mg/Nm3. At Niles, air toxics were tested and showed high
capture efficiency of some trace elements in the baghouse. A significant portion of the boron and almost
all of the mercury escaped to the stack. Selenium and cadmium were effectively captured in the acid
drain, as were organic compounds.
O&MImpacts
While no systematic assessment of O&M impacts has been carried out, it is expected that the following
impacts would need to be taken into account:
• Fuel (most likely natural gas) requirements for the heaters
• Increased auxiliary power which is going to affect the plant efficiency and heat rate
• Safety issues associated with ammonia storage and feed system
3-47
-------
Costs
Costs for retrofitting a 500 MW plant with the SNOX process are projected to be 305 $/kW (EOY 1995
dollars) resulting in 6.1 mills/kWh levelized costs3 (over a 15-year period). Using the CEI (Chemical
Engineering Annual Plant Index), the adjusted cost in EOY 2000 dollars is 315 $/kW. Presently, the
supplier projects that the SNOX process could be retrofitted on a 300 MW power plant in the US for
approximately 270 $/kW.
Issues Associated with the Technology; Future Outlook
The main issue associated with the technology is its cost-effectiveness in reducing SO2 and NOX relative
to competing technologies.
References
1. Clean Coal Technology Demonstration Program: Update 2000; U.S. DOE, National Energy
Technology Laboratory: Pittsburgh, PA, April 2001; 5-74.
2. SNOxFlue Gas Cleaning Demonstration Project: A DOE Assessment; U.S. DOE/NETL-
2000/1125: Pittsburgh, PA, June 2000.
3. Bruno, L.; Ricci, R. Application of SNOX technology at the Gela power plant in Sicily. Presented
at the PowerGen 2000 Europe, Helsinki, Finland, June 20-22, 2000.
4. Final Report for Niles Demonstration, Vol. IT. Project Performance and Economics; DE-FC22-
90PC8955; U.S. DOE: Washington, DC, July 1996.
5. Final Report for Niles Demonstration, Vol. I: Public Design; DOE/PC/8955-T21 (NTIS:
DE96050312); U.S. DOE: Washington, DC, July 1996.
6. A Study of Toxic Emissions from a Coal-fired Power Plant Utilizing the SNOX Innovative Clean
Coal Technology Demonstration, Vols. 1 and 2; DOE/PC/93251-T3; U.S. DOE: Washington,
DC, July 1994.
7. Andeasen, J.; Laursen, J. K.; Bendixen, O. R. SNOX Plant in Full-scale Operation. Modern Power
Systems Magazine, April 1992.
3 Other assumptions: credit was assumed for sale of sulfuric acid at 25 $/ton and for heat recovery from the SNOX
process for use in the boiler at 2.00 $/MMBtu
3-48
-------
3.2.2.4 SOx-NOx-Rox Box (SNRB™)
SNRB Summary
Status Pilot stage
SO2 Reduction, % 80 - 90
NOX Reduction, % 90
Hg Reduction, % Zero
Cost
Capital ($/kW) 253 for 150 MW retrofit
Fixed O&M ($/kW-yr) NA
Variable O&M (mills/kWh) NA
Applicability Both new and retrofit applications
Issues Requires demonstration
Technology Description
The SNRB process combines the removal of SO2, NOX, and participates in one unit - a high-temperature
baghouse.1'2 It operates in the 425-455 °C (800-850 °F) temperature range and is placed before the air
heater. SO2 removal is accomplished using either calcium- or sodium-based sorbent, sodium carbonate
(Na2CO3), and sodium bicarbonate (NaHCO3), injected into the flue gas upstream of the baghouse. NOX
removal is accomplished by injecting ammonia (NH3) to selectively reduce NOX in the presence of a
selective catalytic reduction (SCR) catalyst, which is placed on the high-temperature ceramic filter bags
of the baghouse. Also, the baghouse removes particulates, its primary design function.
Commercial Readiness and Industry Experience
The SNRB technology requires further demonstration on a 50-100 MW scale. It has been demonstrated
successfully on a 5- MW slipstream at Ohio Edison's Burger No. 5 (a 156- MW boiler burning high-
sulfur bituminous coal) boiler.
Emission Control Performance
Four different sorbents were tested at the Burger pilot project including hydrated lime, sugar-hydrated
lime, lignosulfonate-hydrated lime, and sodium bicarbonate. The results achieved can be summarized as
follows:1'2
• SO2 removal efficiency of 80 percent with commercial-grade hydrated lime at a calcium-to-sulfur
(Ca-to-S) molar ratio of 2.0 and temperature of 425-455 °C (800-850 °F),
• SO2 removal efficiency of 90 percent with sugar-hydrated and lignosulfonate-hydrated lime at a Ca-
to-S ratio of 2.0 and temperature of 425-455 °C (800 - 850 °F),
• SO2 removal efficiency of 80 percent with sodium bicarbonate at a sodium-to-sulfur (Na2-to-S) molar
ratio of 1.0 and temperature of 220 °C (425 °F),
• SO2 emissions were reduced to less than 1.2 Ib/MMBtu, 80-85% SO2 reduction, with 3-4 percent
sulfur coal with a Ca-to-S molar ratio as low as 1.5 and Na2-to-S ratio of 1.0,
• Injection of calcium-based sorbents directly upstream of the baghouse at 440-480 °C (825-900 °F)
resulted in higher overall SO2 removal than injection farther upstream at temperatures up to 650 °C
(1,200 °F),
• NOX reduction of 90 percent was achieved with an NH3-to-NOx ratio of 0.9 and temperature of
425-455 °C (800-850 °F), and
3-49
-------
• Also, 99.9 percent participate removal has been demonstrated.
In summary, 80 - 90 percent SO2 removal, 90 percent NOX removal, and similar levels of control for
some air toxics were demonstrated. However, a demonstration on a larger scale (50 - 100 MW) is needed
to assess the performance and economics of this technology.
O&M Impacts
The only potential O&M impact relates to effects on the FFs (mainly life expectancy). The SNRB filters
are ceramic, incorporating a catalyst for NOX control, and their reliability is affected by both mechanical
and thermal stresses due to cycling (changing operating conditions) and cleanup of the bags. To address
these concerns, a 3,800-hour durability test of three fabric filters was completed2 at the Filter Fabric
Development Test Facility in Colorado Springs, CO, in December 1992. No signs of failure were
observed. All of the demonstration tests were conducted using 3M Company Nextel ceramic fiber filter
bags or Owens Corning Fiberglas S-Glass filter bags. No excessive wear or failures occurred in over
2,000 hours of elevated-temperature operation.
In most cases, SNRB is expected to be placed before the air heater, where the flue gas temperature is
adequate [425-455 °C (800-850 °F)]. In case, the temperature is lower, flue gas heating may be required
and the consumption of natural gas or oil needs to be taken into account, as an additional O&M impact.
Costs
The supplier estimates that capital costs for a 150- MW retrofit would be $253/kW (constant US$ 1994),
assuming 3.5 percent sulfur coal, baseline NOX emissions of 1.2 Ib/MMBtu, 65 percent capacity factor,
and 85 percent SO2 and 90 percent NOX removal. Levelized cost over 15 years (constant U.S.$, 1994) is
estimated to be 12.1 mills/kWh or equivalent to $553/ton of SO2 plus NOX removed.
Issues Associated with the Technology; Future Outlook
Considering that the technology has not been demonstrated, there are some uncertainties associated with
both the emission control potential and its economics. Since the Burger pilot program in the mid-1990s,
the supplier has not pursued further development of the technology because there was no demand for
multi-emission control technologies. With the pending multipollutant control regulation, demand may
increase for technologies such as SNRB.
References
1. Clean Coal Technology Demonstration Program: Update 2000; US DOE, National Energy
Technology Laboratory: Pittsburgh, PA, April 2001; 9-13.
2. Evans, A. P.; Kudlac, G. A.; Wolkinson, J. M.; Chang, R. SOx-NOx-Rox Box - High
Temperature Baghouse Performance. Presented at the 10th Particulate Control Symposium,
Washington, DC, April 5-8, 1993.
3-50
-------
3.2.2.5 THERMALONOx and FLU-ACE Process
THERMALONOx and FLU-ACE Process Summary
Status
SO2 Reduction, %
NOX Reduction, %
Hg Reduction, %
Cost
Capital ($/kW)
Fixed O&M ($/kW-yr)
Variable O&M (mills/kWh)
Applicability
Issues
In demonstration
Up to 95
Up to 90
NA
THERMALONOx: $35/kW for 500 MW; FLU-ACE: Not
available
NA
1.43*
Applicable to both new plants and retrofits
Need demonstration to confirm performance and reliability
on large scale, as well as economics.
*Assumed to be only for the THERMALNOx process
Technology Description
The THERMALONOx and FLU-ACE process is a two-step process.1'2 The first step is the gas phase
chemical reaction converting nitrogen monoxide (NO) into nitrogen dioxide (NO2). This is accomplished
by direct phosphorus (P4) liquid injection into the flue gas, which causes the release of ozone (O3) that
oxidizes NO into NO2. This NO to NO2 conversion takes place in a specially designed reactor normally
placed upstream of a wet scrubbing process.
The prevailing chemical reactions are:
P4 + O2 -^ P4O + O
P4On + O2 -^ P4On+1 + O, where n is 1 through 9
2NO + (O+O3) -» 2NO2 + O2
2 NO2 + H2O -» NO2 + NO3 + H2
In the second step the NO2 is removed, as a water-soluble gas in the wet FGD process. In addition, as in
the phosphoric acid manufacturing process, the solid P4Oi0 particles that are formed in the duct from the
first step, upon entering the wet FGD process are instantly hydrated and hydrolyzed to form the dibasic
phosphate (H2PO4~) and the mono-basic phosphate (HPO42~) anions.
This second step could be accomplished by a wet FGD process (especially if it already exists) or the FLU-
ACE process, a condensing reactive scrubber, which is also offered by Thermal Energy. The supplier
projects 90 percent nitrogen dioxide removal efficiency by wet FGD and 98 percent by FLU-ACE. If the
FGD cannot achieve the required NOX reduction, a FLU-ACE reactor can be added to boost the NOX
reduction above 90 percent and remove other pollutants such as mercury, VOCs, and fine particulates.
The by-products of the THERMALONOx process are non-toxic nitrogen and phosphate compounds used
as fertilizers and animal food additives.
The P4 liquid is maintained at a constant temperature of 60 °C (140 °F) in the P4 injection tank and is
pumped continuously through a steam-traced closed piping circuit connecting the P4 injection tank to the
P4 atomization nozzle injection location in the flue gas duct. The P4 injection rate is individually
controlled to each nozzle via corresponding P4 liquid flow control valves and atomizing steam pressure
3-51
-------
control valves to maintain optimum P4 droplet size, physical dispersion, and chemical reaction in the duct.
Atomization is done using steam, compressed air, or an aqueous emulsion of phosphorus.
When THERMALONOx is used, the existing wet FGD process must absorb the additional nitrogen and
phosphorus compounds generated. The system must also deal with the additional nitrogen and
phosphorus species in solution. These compounds must be controlled to maintain proper scrubber
operation to optimize NO2 and SO2 absorption and to control and maintain the by-product quality,
whether it is for gypsum production or landfill.
Commercial Readiness and Industry Experience
The technology is in the early demonstration stage. There were plans to install it at American Electric
Power's Conesville plant (a 375 MW unit), but the project was suspended on September 27, 2001, during
initial start-up. Tests were expected to last for 8-12 weeks with the target NOX reduction of 75 percent.
In parallel with the Conesville demonstration, Thermal Energy is carrying out an R&D program to test the
THERMALONOx process with an existing FLU-ACE scrubber near Ottawa, Canada.
Emission Control Performance
Thermal Energy projects 80-90 percent NOX removal for THERMALONOx with conventional FGD.1 A
combination of THERMALONOx and FLU-ACE is projected to achieve 98 percent NOX removal. In
both cases, SO2 removal is projected to be in the 90 - 95 percent range.
Key parameters affecting the THERMALONOx process are the flue gas temperature and oxygen
concentration, which affect the reaction rate for converting NO to NO2, and therefore the reaction time
required to achieve the desirable conversions (usually 99 percent). For a typical coal-fired power plant
with flue gas at 140-160 °C (280-320 °F) and 4-5 percent excess oxygen, the required reaction time is
about 1 second. The NOX to P4 ratio is projected by the supplier to be in the 3.0 - 4.0 range.
O&MImpacts
Potential O&M impacts are associated with the ability to control the spontaneous reaction of the
phosphorus and flue gas to produce ozone. Also, if there is no market for the wastes, the ability to dispose
of phosphate compounds needs to be assessed.
Costs
Thermal Energy projects the cost of THERMALONOx for a 500 MW coal-fired plant to be
approximately $35/kW.:
Variable O&M are projected to be 1.43 mills/kWh.1
Issues Associated with the Technology; Future Outlook
The main issues and uncertainties associated with the technology are:
• Confirmation of pilot scale results and projected economics in utility scale demonstration and
subsequently commercial projects.
• The ability to control the spontaneous reaction of the phosphorus and ability of the flue gas to
produce ozone in large scale needs to be demonstrated.
• Disposal of phosphate compounds needs to be assessed including the impact of phosphorus injection
on by-product properties and economics.
3-52
-------
References
1. Hinke, T. V. Utility NOX Control with Elemental Phosphorus. Proceedings of the U.S. EPA-
DOE-EPRI Combined Power Plant Air Pollutant Control Symposium: The MEGA Symposium
and The A&WMA Specialty Conference on Mercury Emissions: Fate, Effects, and Control,
Chicago, IL, Aug 20-23, 2001.
2. Leading Edge Environmental Compliance & Energy Conservation Solutions; Thermal Energy
International, Inc., Nepean, Ontario, Canada: http://www.thermalenergy.com/, (accessed Dec 1,
2004).
3. Hinke, T. Utility NOX Control with Elemental Phosphorus. Proceedings of the U.S.
EPA/DOE/EPRI Combined Utility Air Pollutant Control Symposium: The MEGA Symposium,
Chicago, IL, Aug 20-23, 2001.
3.2.3 SO2, NOX, and Mercury Control
3.2.3.1 Activated Coke
Activated Coke Summary
Status Commercial
SO2 Reduction, % 90 - 98
NOX Reduction, % 15 - 80 (depends on SO2 reduction)
Hg Reduction, % 90 - 99
Cost
Capital ($/kW) 120 - 200 for 300 - 1,000 MW
Fixed O&M ($/kW-yr) 8-13
Variable O&M (mills/kWh) 1.3-2.1
Applicability Both new and retrofit applications
. Demonstration of the combined SO2-NOx-Hg control is
needed in the United States
Technology Description
The activated coke process,1"3 shown in Figure 3-6, involves three steps: (1) adsorption, (2) desorption,
and (3) (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 (NFLIiSC^); the latter in case
ammonia is injected in addition to the presence of activated coke. These compounds are maintained in the
coke inner surface at temperatures ranging from 100 to 180 °C (212 to 356 °F). The adsorber acts also as
a particulate control device reducing particulates below 30 mg/Nm3 (0.0124 grain/scf) when the inlet is
kept below 200 mg/Nm3 (0.0827 grain/scf).
In the second stage of the adsorption process, the activated coke acts as a catalyst in the decomposition of
NOX to nitrogen and water after injection of ammonia in the activated coke bed. The chemical reaction
occurs in the 100-180 °C (212-356 °F) temperature range.
3-53
-------
Activated coke is a carbonaceous material produced by steam activation (at approximately 900 °C). It has
high mechanical strength against abrasion and crushing. Its surface area is 150-250 m2/g, less than the
conventional activated carbon but much higher than the metallurgical coal.
As the activated coke is loaded with sulfuric acid, its adsorption capacity declines. To regenerate the
activated coke, it is conveyed by a bucket elevator to a desorber. In the desorber the sulfuric acid or
ammonium hydrogen sulfate (NF^HSC^) is decomposed to nitrogen, sulfur dioxide, and water. The
regenerator is heated to 380-500 °C (716- 932 °F) temperature range through an external air heater,
which consumes either oil or natural gas. Utilization of flue gas for this purpose is possible to reduce fuel
requirements and associated costs, but has not been implemented. After cooling, the activated coke passes
through a vibrating screen to eliminate smaller particles (fines) and then it is recycled back into the
adsorber.
Basic process flow chart
Activated char
Chemical reaction for
sulfuric acid manufacture
(sulfur dioxide) (oxygen) (anhydrous sulfuric acid)
©SOs+HaO => HaSCk
(anhydrous sulfuric acid) (water) (suKunc acid)
Concentrated
Sulfuric acid
manufacturing facility
S02 gas
•>•
/
— rl
. Off gas
Sulfuric acid
Dust
Fuel
Air heating furnace
Figure 3-6. Activated coke process flow diagram7
SO2-rich gas can be reduced to H2S in a reduction column and then elemental sulfur can be produced in a
Claus unit (typical process for sulfur production). Alternatively, sulfuric acid can be produced.
Mercury can be removed also by adsorption. Once adsorbed on the coke, mercury must be collected in a
form suitable for disposal or recovery. Mercury is absorbed on the coke up to 1.7 mg/g-coke at the
temperature below 180 °C. Due to the temperature profile within the regenerator, mercury-rich coke can
be found in the middle section of the regenerator. The mercury concentration in this area increases over
time and removal of the mercury-rich coke is required once every few years (1-3 years depending on the
percentage sulfur and mercury in the coal). The coke removed, representing approximately 1 percent of
3-54
-------
the total coke in the system, is sent to special facilities, which recover the mercury and sell it for
commercial applications.
The mercury recovery facilities use various methods one of them being the use of a selenium filter, which
absorbs the mercury from the flue gas and forms HgSe, a chemically stable compound. The selenium
filter technology is commercial and can achieve up to 98 percent Hg collection efficiency during the filter
life (usually 4-5 years). Once spent, the selenium filter has to be disposed of in a hazardous waste facility.
Other methods of mercury removal or disposal are also considered, such as SO2-rich gas (SRG) off-gas,
sulfuric acid plant off-gas, and SRG scrubber waste water.
Commercial Readiness and Industry Experience
The process is commercially available in Japan and Germany. It was originally developed by Deutsche
Montan Technologie (formerly Bergbau-Forschung GmbH) and demonstrated at a 158,000 m3/s (93,000
scfm) plant, the Kellerman generating station of STEAG GmbH. Mitsui Mining Co., Ltd. of Japan
obtained a license from Deutsche Montan Technologie and tested it in a pilot facility from 1981 to 1983.4
Installations of the Mitsui activated coke process in Japan5'6 and Germany include the following (designed
for both SO2 and NOX control unless otherwise indicated):
• 31,775 Nm3/h at Mitsui's power generating station (1984)
• 236,000 Nm3/h at Idemitsu Kosan's refinery on a residual fluidized bed catalytic cracking process
(1987)
• 451,000 and 659,000 Nm3/h boilers at EVO GmbH's Arzberg power station in Germany (1987)
• 323,000 Nm3/h at Hoechst AG's power station in Frankfurt, Germany (1989)
• 10,000 Nm3/h at Electric Power Development Corp.'s (EPDC) Wakamatsu power station (1990)
• 1,157,000 Nm3/h (350 MW) AFBC boiler at the Electric Power Development Corp. Takehara power
station (1995), designed only for NOX reduction and achieved above 80 percent NOX reduction
Sumitomo of Japan has also developed its own activated coke technology and provided a commercial
system for the Electric Power Development Corp. Isogo station, a 600 MW power plant burning low-
sulfur coal (flue gas flow rate of 1,806,000 Nm3/h). The activated coke process is designed for 95 percent
SO2 removal and 30% NOX reduction.
Sumitomo also has utilized this technology in sintering plants of steel-making industry for SO2, NOX and
dioxin control. The following Table 3-5 provides the size and start-up year for sintering plants utilizing
Sumitomo's process.
Table 3-5 Sintering Plants Utilizing Activated Coke
Customer
A
B
C
D
E
F
Emission Removed
SO2
SO2, NOX, Dust
Dioxin, Dust
SO2, NOX, Dust
SO2, NOX, Dust, Dioxin
SO3, NOX, Dust
Gas Volume
[Nm3/h]
900,000
1,300,000
1,552,000
1,300,000
1,350,000
1,650,000
Start up
1987
1999
2003
2003
2004
2004
3-55
-------
Emission Control Performance
SO2 control efficiency has ranged from 90 to 98 percent and the NOX control efficiency from 15 to 80
percent.1'5'6 NOX reduction is higher for lower SO2 concentrations at the inlet of the adsorber. For
example, 1000 ppm SO2 concentration may result in NOX reduction in the 15-40 percent range, while
200 ppm SO2 concentration will raise NOX reduction to above 70 percent. NOX reduction is also affected
by the amount of ammonia injected (NH3-to-NOx ratio typically in 0.5-1.0 range results in above 70
percent NOX reduction), the oxygen concentration of the flue gas (lower excess O2 results in lower NOX
reduction), and the inlet gas temperature.
Based on pilot-scale tests carried out by Sumitomo and measurements by EPDC at its Matsushima power
station, 90 - 99 percent mercury removal is projected. These tests resulted in >99 percent mercury
reduction at operating temperatures of 150 to 180 °C. Mercury reduction data from the remaining
installations are not available, because these systems were not designed with mercury control in mind, but
rather for SO2 and NOX control; hence, mercury was removed, but no data were kept regarding mercury
capture efficiencies.
O&MImpacts
Due to the high heat capacity of the activated coke system, it takes longer than a conventional plant to
reach the operating temperature of the de- NOX process, which means that either NOX emissions will be
higher during start up or the ramp-up rate of the plant will be limited. Other O&M impacts are not known,
mainly because of the lack of information and experience in the United States. Required auxiliary power
is estimated to be 0.70 percent of the gross output, compared to 1.0-2.5 percent for a wet FGD.
Capital Costs
According to the suppliers, capital costs (including by-product equipment) are projected to be in the $120
- 200/kW range for a unit 300 - 1,000 MW in capacity, gas temperature 150-165 °C (300-330 °F), and
gas inlet conditions: SOX 500 - 1,000 ppm, NOX 170 - 260 ppm and dust 140 - 200 mg/Nm3. The capital
costs include all equipment (including mercury control and acid production) fully installed and
operational.
O&M Costs
Fixed costs are expected to be in the $8-13/kW-yr range. Operating costs (including electricity, active
coke, NH3, and fuel) are projected to be 1.3-2.1 mills/kWh (assumptions: activated coke: $833 per ton;
ammonia: $223 per ton; electricity: 35 mills/kWh; cost of labor: 50,000 $/year-person). As the sulfur
content in the coal increases above the 2 percent level, the O&M costs may increase substantially mainly
due to the increase in activated coke costs.
Issues Associated with the Technology; Future Outlook
During start-up, it takes longer to bring up the temperature in the De-NOx system; so, NOX reduction in
cycling units may suffer during start-up unless they are designed to utilize an external heat source to
preheat the De-NOx reactor.
References
1. Electric Power Development Corp. Isogo Thermal Power Station/Outline of Facilities; Brochure
describing the Isogo power station, 2000.
2. Mitsui Mining Co., Ltd. Dry Type Desulfurization and Denitrification Technology for Coal-fired
Power Plants; Brochure, May 2000.
3-56
-------
3. 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. Presented at the
1999 International Ash Utilization Symposium in Lexington, KY, Oct 18-20, 1999.
4. Ogaki, M.; Fujitsu, H.; Komatsubara, Y.; Ida, S. Catalytic Activity of Coke Activated with
Sulfuric Acid for the Reduction of Nitric Oxide. Fuel 1983, 62, 325.
5. Tsuji, K.; Shiraishi, I. Combined Desulfurization, Denitrification, and Reduction of Air Toxics
Using Activated Coke: 1. Activity of Activated Coke. Fuel 1997, 75, 549.
6. Tsuji, K.; Shiraishi, I. Combined Desulfurization, Denitrification and Reduction of Air Toxics
Using Activated Coke: 2. Process Applications and Performance of Activated Coke. Fuel 1997,
76, 555.
7. Watanabe, T.; Murayama, H.; Saito, I.; Morimoto, K.; Shibata, T. Coal-fired Power Stations and
Gas Clean up Processes using Activated Coke for the 21st Century - Multi-Pollutant Control
Technology. Presented at the 28th International Conference on Coal Utilization & Fuel Systems,
Clearwater, FL, March 9-13, 2003.
3.2.3.2 Electro-Catalytic Oxidation (ECO)
ECO Summary
Status
SO2 Reduction, %
NOX Reduction, %
Hg Reduction, %
CO2 Change, %
Cost
Capital ($/kW)
Fixed O&M ($/kW-yr)
Variable O&M (mills/kWh)
Applicability
Issues
In demonstration
98
90
90
5 increase
200 for 500 MW plant
6.83
1.5 projected
Both new and retrofit applications
50 MW demonstration in progress
Technology Description
The ECO process, shown in Figure 3-7,1 treats flue gas in three steps to achieve multipollutant removal.
First, a majority of the ash in the flue gas stream is removed in a conventional dry ESP. Following the
ESP, a barrier discharge reactor oxidizes the gaseous pollutants to higher oxides. For example, nitric
oxide is reacted to form nitric acid, sulfur dioxide is converted to sulfuric acid, and mercury is oxidized to
mercuric oxide. Products of the oxidation process are then captured in a wet electrostatic precipitator
(WESP) that also collects fine particulate matter. Liquid effluent from the WESP may be treated to
remove collected ash and then delivered to a system to produce concentrated sulfuric and nitric acids for
sale. The ECO system is designed for retrofit into the last fields of an existing ESP. If the ESP does not
have adequate space to fit the ECO system, some or all components could be built downstream of the
ESP. In the latter case, the downtime of the plant is reduced, but additional space (footprint) is needed.
3-57
-------
The sulfuric and nitric acids produced and captured in the WESP effluent can be made into salable by-
products such as concentrated acids, gypsum, or fertilizer.
Oxidation of gaseous pollutants in the barrier discharge reactor is the key component of the ECO process.
Oxidation is accomplished through generation of a non-thermal discharge or plasma. In a dielectric
barrier discharge, energetic electrons are produced throughout the reactor without heating the gas stream
to high temperatures, requiring considerably less energy than plasma discharges. Dielectric barrier
discharges can be operated over a wide range of temperatures and pressures and have been widely used
for commercial ozone (O3) generation.2"5
To form a barrier discharge, a dielectric insulating material is placed between two discharge electrodes.
Typically, the material (glass or ceramic) has a high dielectric strength and high dielectric constant and
covers one of the two electrodes. High voltage applied to the electrodes causes the gas in the gap to break
down. Presence of the dielectric barrier prevents this breakdown from forming an arc with its resulting
energy consumption. Instead, breakdown is in an array of thin filament current pulses, or
"microdischarges." They are well distributed spatially over the discharge gap. Typical duration of a
microdischarge is of the order of a few nanoseconds, and electron energies range from 1 to 10 electron
volts.
The electron energies formed in the microdischarge are ideal for generating gas-phase radicals, such as
hydroxyl (OH) and atomic oxygen (O) through collision of electrons with water and oxygen molecules
present in the flue gas stream:
O2 + e -> O + O + e
H2O + e -> OH + H + e
O + H2O -> 2OH
In a flue gas stream, these radicals simultaneously oxidize NOX, SO2, and Hg to form nitric acid (HNO3)
and NO2, sulfuric acid (H2SO4), and mercuric oxide (HgO), respectively. The above reactions leading to
radical formation and the subsequent oxidation reactions can be made to occur at a low temperature
65-150 °C (150-300 °F).
The presence of a dielectric barrier allows for several possible electrode configurations, including coaxial
cylinders, cylindrical electrodes with plates, and parallel plate electrodes. Different reactor designs have
little effect on overall conversion efficiency. This allows for spacing that reduces the potential for
plugging of the reactor and results in a minimal pressure drop across the reactor. Aerosols formed by the
oxidation reactions, including HgO, HNO3, and H2SO4, exit the barrier discharge reactor in the flue gas
stream. At this point the gas enters a condensing WESP where collection of the aerosols, fine PM, and
other air toxic compounds is accomplished.
The by-products of the ECO process are raw sulfur and nitric and sulfuric acids, which can be used in the
industry for fertilizer and gypsum production. Of course, the extent to which these by-products would be
actually used depends on economics (supply and demand of competing products) in the local market
(around the power plant).
Commercial Readiness and Industry Experience
The technology is in the demonstration stage. It was originally tested at laboratory scale (1 and 100 scfm).
Then, it was tested at pilot scale at First Energy's R.E. Burger No. 5 unit6'7 (a 156 MW unit), where a
slipstream of equivalent to approximately 1 MW was used to assess the performance of the ECO process.
The 1 MW pilot scale tests have been carried out since early 2002.
3-58
-------
Presently, Powerspan has installed and begun testing a commercial demonstration unit at First Energy's
R.E. Burger plant.8 The technology is being tested at a slipstream (110,000 scfm or 7.4 percent of the total
flue gas) equivalent to 50 MW. The unit burns a variety of fuels including Ohio coal with 2-4 percent
sulfur.9
Emission Control Performance
At First Energy's R.E.Burger No. 5 unit (1 MW slipstream), the technology achieved the following
emission reductions:9
• 90 percent NOX reduction,
• 98 percent SO2 reduction, and
• 90 percent mercury reduction.
Conventional
Electrostatic
Precipitator
Gas
Converter
Wet
Condensing
Precipitator
From
Boiler
ash
H2U
°2
S02'
NOX
Hg«
including other
trace elements
PM2.5 ash
02
S02
NOX
Hg'
Y7 Y/
1 rower bupply
90% of inn"
ash Station Provided
1
|
PM2.5 ash o /
H20 T\
2 A —
(75% of)S02 Y \
(25%of)NOx I/
HgO Y\
H2S04 A^
HN J3
High Voltage
Cooling
Spray
Injection
\-p/
\
\
^
/~
'
^
H2S04 (dilute)
HN03 (dilute)
Scrubbed SO2
PM 2.5 ash
HgO
Station Provided
Cooling Water
Surplus or
Makeup -:
Water
DIutEAjid
and Di^)k/ed Minerate
Station-Provided
Service Steam <
Miner a
Removal
Figure 3-7. ECO process flow diagram.
It should be pointed out that these results were achieved with 337 ppm NOX in the inlet of the ECO
system, approximately 40 percent higher than a similar installation with low-NOx burners.
3-59
-------
The early results from 50 MW demonstration (by Powerspan Corp.) achieved the following:
• Greater than 90 percent NOX conversion in the reactor, total reduction of 65 percent, hindered by
absorber performance
• Greater than 98 percent SO2 reduction, often >99 percent
• Mercury measuring equipment being installed
O&MImpacts
The main O&M impact of the ECO process is the auxiliary power consumption. Powerspan estimates that
approximately 3 percent of the plant's output is needed for the ECO process to reduce inlet NOX of 0.30
Ib/MMBtu down to 0.05 lb/MMBtu.8
Also, cooling water is required for the WESP and is assumed to be provided by the cooling water system
of the plant. Finally, heating for the acid recovery and ash drying is required and is expected to be
provided by the plant's auxiliary system. While these do not present any technical challenge, they will
have some impact on the plant mass and energy balances, as well as the plant's overall efficiency. At this
point, there is not enough information to assess the magnitude of this impact, but it is not expected to be
substantial.
The list of consumables and the by-products of the ECO process is given below.10
Consumables:
• Electric power for the barrier discharge reactor, pumps and blowers. The power for the discharge
reactor is related to the NOX reduction desired.
• Heat for the by-product crystallizer.
• Ammonia reagent - this 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 absorption tower - about 1 gpm per MW - no special quality specification.
• Carbon filters for mercury removal from the liquid discharge of the absorber.
• 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.
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 for
every mole of SO2 reduced, one mole of ammonium sulfate is produced.
• Mercury captured on the activated carbon (a waste to be disposed of) at a cost of about $1000/lb of
mercury captured.
• 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.
• Water vapor.
Costs
The cost of the demonstration project at First Energy's R.E. Burger (equivalent 50 MW scale) was
$18 million.8 This is equivalent to $200/kW for 500 MW unit. AmerenUE, Sargent & Lundy,
3-60
-------
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 $120,400,000, inclusive of process equipment, general facilities, owner's costs, and contingencies. This
also included the fertilizer plant and balance of plant modifications. It is the only comprehensive, full-
scale cost analysis that has been made publicly available. Therefore, a cost of $200/kW is a reasonable
estimate to use.10'11
Variable operating cost for ECO is the cost of power and other consumables such as ammonia and
specialty chemicals. Ammonia consumption is determined by the molar ratio and specialty chemical costs
are estimated at $150/ton of NOX removed and about $15/ton of SO2 removed. 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. 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 three
operators and one maintenance person per shift. The manpower needs are not expected to be a significant
function of unit size.10
The power for the dielectric barrier discharge reactor is largely determined by 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 low NOX burners or other devices to minimize NOX
into the ECO reactor. Based upon available information, reactor power (in W/scfm) can be assumed to be
equal to the lesser of 20 W/scfm or 58.22«(NOX) - 6.2431, where NOX is measured in Ib/MMBtu. This is
shown in Figure 3-8 below. 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.10
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 plants.10
Issues Associated with the Technology; Future Outlook
As the process is scaled up, the main uncertainties are whether it can achieve the performance (emission
reduction), which was achieved at smaller scale. Also, the costs and cost-effectiveness of the process is
uncertain and may be adversely impacted by relatively high auxiliary power requirements. Finally, some
uncertainty exists with regard to the scalability of the by-products both in terms of their suitability in
meeting market specifications and price. Technology remains in the demonstration phase; therefore,
performance and costs remain uncertain, especially for large-scale projects.
3-61
-------
c
o
•
I
•4-*
C -
Q) C
a "5
o
0>
I
Q_
25
20
15
10
0
0.1
y = 58.22x-6.2431
0.2 0.3 0.4
NOx(lb/MMBtu)
0.5
Figure 3-8. ECO power consumption versus NOX.
References
1. McLarnon, C. R.; Jones, M. D. Pilot Testing and Scale-up of a Multipollutant Control
Technology at FirstEnergy. Presented at the Power-Gen International Conference, Orlando, FL,
Nov 15, 2000.
2. Kogelschatz, U.; Eliasson, B.; Egli, W. Dielectric-Barrier Discharges Principle and Applications.
Proceedings of the International Conference on Phenomena in Ionized Gases, Toulouse, France,
July 1997.
3. Penetrante, B.; Schultheis, S. Non-Thermal Plasma Techniques for Pollution Control.
Proceedings of the NATO Advanced Research Workshop, Cambridge, England, Sept 1992.
4. McLarnon, C.; Penetrante, B. Effect of Gas Composition on the NOX Conversion Chemistry in
Plasma. SAE Technical Paper Series 982433, San Francisco, CA, Oct 1998.
5. McLarnon, C.; Penetrante, B. Effect of Reactor Design on the Plasma Treatment of NOX SAE
Technical Paper Series 982434, San Francisco, CA, Oct 1998.
6. McLarnon, C. R.; Jones, M. D. Electro-catalytic Oxidation Process for Multipollutant Control at
FirstEnergy's R.E. Burger Generating Station. Presented at the Electric Power 2000 Conference,
Cincinnati, OH, April 5, 2000.
7. McLarnon, C. R.; Horvath, M. L.; Boyle, P. H. Electro-catalytic Oxidation Technology Applied
to Mercury and Trace Elements Removal from Flue Gas. Presented at the Conference on Air
Quality II, McLean, VA, Sept 20, 2000.
8. Boyle, P.O., Operation of 50-MW ECO Commercial Demonstration Unit at FirstEnergy's R.E.
Burger Plant. Presented at the Electric Power 2004, Baltimore, MD, March, 2004.
3-62
-------
9. Boyle, P. D.; Steen, D.; Dovale, A. J. Commercial Demonstration of ECO Multipollutant Control
Technology. Presented at the ICAC Forum'03, Nashville, TN, Oct 2003.
10. Studt, J. E.; Jozewicz, W. Performance and Cost of Mercury and Multipollutant Emission
Control Technology Applications on Electric Utility Boilers; EPA-600/R-03-110; U.S. EPA:
Research Triangle Park, NC, Oct 2003.
11. McLarnon, C.; Steen, D. Combined SO2, NOX PM and Hg Removal from Coal-fired Boilers.
Proceedings of the Combined Utility Air Pollutant Control Symposium: The MEGA Symposium,
Washington, DC, May 19-21, 2003.
3.2.3.3 SCR and Wet FGD
SCR and Wet FGD Summary
Status
SO2 Reduction, %
NOX Reduction, %
Hg Reduction, %
Cost
Capital ($/kW)
Fixed O&M ($/kW-yr)
Variable O&M (mills/kWh)
Applicability
Issues
Commercial
95
90-95
40—90% depending on coal type
SCR: 50-140; Wet FGD: 160-275 for a nominal 400 MW
plant
SCR: 0.5-1.4; Wet FGD: 1.2-14
SCR: 0.75 - 4; Wet FGD: 0.1 - 1
Plants with SCR and wet scrubber technologies
Need additional confirmation of mercury oxidation levels in
the SCR
Technology Description
Selective Catalytic Reduction (SCR) - NOy Control
SCR technology1 reduces NOX through a catalytically enhanced reaction of NOX with ammonia, reducing
NOX to water and nitrogen. This reaction takes place on the surface of a catalyst, which is "housed" in a
"reactor" vessel. The reactor ensures that the flue gas is uniformly distributed over the catalyst as well as
determines the flue gas velocity. Typical catalyst materials are titanium-oxide and vanadium-oxide on a
"coated" substrate structure that may take forms such as plate or honeycomb. SCR system configuration
is generally referred to in accordance with the location of the SCR relative to the power plant:
• "high-dust" - SCR located between the economizer and the air preheater, upstream of the ESP
• "low-dust" - SCR located between a hot-side ESP and the air preheater
• "tail-end" - SCR located downstream of the air preheater, ESP, and FGD. This approach requires the
flue gas to be reheated prior to entering the SCR
Ammonia (anhydrous or aqueous) is injected into the flue gas upstream of the SCR reactor through a
nozzle grid designed to ensure its uniform distribution in the flue gas and then through the catalyst.
SCR-Mercury Control
The contribution of SCR technology to mercury reduction comes from the fact that SCRs have been
shown (also discussed in Section 3.2.1.3) to oxidize elemental mercury.2"5 Thus, there is synergism with
3-63
-------
wet scrubbers, which are effective in capturing oxidized mercury. In fact, the SCR-wet scrubber
combination is conceptually similar to the approaches discussed in the wet scrubber section (catalytic-
and reagent-based mercury oxidation upstream of a wet scrubber).
Wet Scrubber
As explained before in Section 3.2.1.3, the most commonly used wet scrubber technology uses a wet
limestone process with in-situ forced oxidation to remove SO2 from the flue gas while producing a
gypsum-grade by-product. This is accomplished typically in a vertical vessel with flue gas contacting and
reacting with limestone slurry to produce a mixture of calcium sulfite and sulfate. Through controlled
oxidation of the reaction products, a salable by-product in the form of commercial grade gypsum may be
produced. The intimate contact between gas and liquid is enhanced through different design approaches,
usually involving several counterflow spray levels and mass transfer "trays" to optimize gas-liquid
interactions. The technology has evolved over the years through "mechanical" improvements, which have
included better gas and liquid distribution within the scrubber, better controlled droplet size and size
distribution, as well as "chemistry" improvements such as the addition of organic acids which not only
improve overall SO2 capture but also help the settling characteristics of the waste products. Several
commercial variations of the technology exist based on reagent type, vessel design, etc.
The intense gas-liquid mass transfer, combined with the fact that oxidized, vapor-phase mercury is water
soluble (elemental mercury is not), is what allows wet scrubbers to be potentially an excellent means of
mercury control. Wet FGD is a mature technology that offers potential for a very effective mercury
control via minor process modifications.
Commercial Readiness and Industry Experience
Similar to wet FGD technology, SCR technology is widely used commercially worldwide. In Germany,
for example, essentially all coal-fired boilers are quipped with SCR technology combined with wet
scrubbers. Over 50,000 MW of capacity is deployed worldwide. In the United States the technology is
being deployed at a rapid pace at present.6 Therefore, both SCR (for NOX control) and wet scrubbers are
readily commercially available.
With respect to SCR performance on mercury oxidation, testing is on-going at pilot- and full-scale sites.2"5
Efforts by B&W2'4 and EPRI3 are examples of pilot-scale activities. Full-scale testing is taking place at
various sites as well.5'7
Emission Control Performance (Mercury Oxidation)
SCR systems are widely used in conjunction with wet FGD technology, and the combination has yielded
total mercury capture exceeding 80 percent.4'7 In tests conducted at Grosskraftwerk Mannheim AG in
Germany, mercury oxidation across the SCR increased HgCl2 content in the flue gas from 77 to 95
percent.5
Results from tests at B&W indicated that oxidized mercury increased from 50.9 to 93.4 percent in the
presence of the SCR catalyst at typical SCR temperatures, while at lower (air preheater outlet)
temperatures, oxidized mercury levels rose from 81.9 to 94.1 percent.2
Joint pilot- and full-scale tests3 by URS and EPRI showed a significant variation in results, suggesting
that the fundamental mechanisms of mercury oxidation across the SCR catalyst may not be fully
understood. Table 3-6, taken from Reference 3, summarizes some of the results. However, the mercury
oxidation is highly depended on the coal type and could range from 40 to 90 percent.
3-64
-------
Table 3-6. Selected SCR Test Results
Control Unit
Pilot SCR
Full scale SCR
Full scale SCR
NHs injection system
NH3 injection system
Control Unit Status
No ammonia
Normal
No ammonia
Normal
No ammonia
Inlet Mercury Oxidation
(% of total)
8-12
10-18
10-18
50-87
50-87
Outlet Mercury
Oxidation
(% of total)
2-10
4-7
50
67-85
70-90
O&MImpacts
No impacts are expected from the natural oxidation of elemental mercury across the SCR catalyst. Of
course, as mentioned in the section on wet FGD, the FGDs and SCRs themselves have O&M impacts
including increased auxiliary power and potential air preheater fouling due to ammonium bisulfate.
Capital Costs
No additional costs to the SCR-wet scrubber system are associated with the naturally occurring oxidation
of elemental mercury on the catalyst and the subsequent capture in the scrubber. SCR costs are highly
dependent on site-specific factors and range from about $50 to 140/kW.8 Capital costs for wet scrubbers
were provided in Section 3.2.1.3 and were in the range of $160-275/kW.
O&M Costs
Fixed O&M:
SCR: $0.5-1.4/kW-yr (based on 1 percent of capital costs).1
Wet scrubber: $1.2-14/kW-yr (for a range of coal sulfur from 0.4 to about 3.5 percent).
Variable O&M:
SCR: 0.75-4 mills /kWh.8 Wide range based on catalyst rate of replacement, number of air preheater
washes.
Wet scrubber: 0.1-1 mill/kWh (for a range of coal sulfur from 0.4 to about 3.5 percent).
Issues Associated with the Technology; Future Outlook
Studying mercury oxidation has become a focus in efforts to enhance SCR, and in turn, wet FGD
technology. Further testing on the speciation of mercury must be conducted to gain a better understanding
of oxidation potential, oxidation vs catalyst age, as well as fundamental mechanisms. As SCR and wet
FGD technologies are increasingly combined for NOX and SO2 reduction, this will result in a significant
and inexpensive way to also control mercury emissions. Parallel research to characterize the stability and
fate of mercury in the FGD sludge or gypsum is ongoing.
References
1. Status Report on NOX Control Technologies and Cost Effectiveness for Utility Boilers;
NESCAUM Report, Boston, MA, June 1998.
2. Milobowski, M.G. Wet FGD Enhanced Mercury Control for Coal-Fired Utility Boilers. The
Babcock & Wilcox Company. Proceedings of the U.S. EPA-DOE-EPRI Combined Power Plant
3-65
-------
Air Pollutant Control Symposium: The MEGA Symposium and The A&WMA Specialty
Conference on Mercury Emissions: Fate, Effects, and Control, Chicago, IL, Aug 20-23, 2001.
3. Richardson, C.; Machalek, T.; Miller, S.; Dene, C.; Chang, R. Effects of NOX Control Processes
on Mercury Speciation in Utility Flue Gas. Proceedings of the U.S. EPA-DOE-EPRI Combined
Power Plant Air Pollutant Control Symposium: The MEGA Symposium and The A&WMA
Specialty Conference on Mercury Emissions: Fate, Effects, and Control, Chicago, IL, Aug 20-23,
2001.
4. Bielawski, G. T. How Low Can We Go? Proceedings of the U.S. EPA-DOE-EPRI Combined
Power Plant Air Pollutant Control Symposium: The MEGA Symposium and The A&WMA
Specialty Conference on Mercury Emissions: Fate, Effects, and Control, Chicago, IL, Aug 20-23,
2001.
5. Fahlke, J.; Bursik, A. Impact of the State-of-the-Art of Flue Gas Cleaning on Mercury Species
Emissions from Coal-Fired Steam Generators. Water, Air, SoilPollut.1995, 80, 209-215.
6. Mcllvaine Company. U.S. Power Plants Will Commit $12 Billion for Air Pollution Control
Equipment in the Next Five Years, http://www.mcilvainecompany.com (accessed Dec 1, 2004).
7. Sloss, L. Mercury Emissions and Effects - the Role of Coal; IEAPER/19; IEA Perspectives, IEA
Coal Research: London, England, Aug 1995.
8. Retrofit NOX Controls for Coal-Fired Boilers - 2000 Update; EPRI: Palo Alto, CA, Dec 2000.
9. Bock, J.; Hocquel, M.; Hein, K. Mercury Oxidation Across SCR Catalysts of Flue Gas with
Varying HC1 Concentrations. Proceedings of the Combined Utility Air Pollutant Control
Symposium: The MEGA Symposium, Washington, DC, May 19-21, 2003.
10. Machalek, T.; Ramavajjala, M.; Richardson, M.; Richardson, C.; Dene, C.; Goeckner, B.;
Anderson, H.; Morris, E. Proceedings of the Combined Utility Air Pollutant Control Symposium:
The MEGA Symposium, Washington, DC, May 19-21, 2003.
11. Brickett, L.; Chu, P.; Lee, C.; Srivastava, R.; Laudal, D.; Thompson, J.; Wocken, C. Impact of
SCR on Mercury Speciation for Coal-fired Boilers. Proceedings of the Combined Utility Air
Pollutant Control Symposium: The MEGA Symposium, Washington, DC, May 19-21, 2003.
3-66
-------
3.2.3.4 EnviroScrub Pahlman
EnviroScrub Summary
Status Pilot scale
SO2 Reduction, % >99
NOX Reduction, % 93-97
Hg Reduction, % Up to 67
Cost
Capital ($/kW) 150 for 500 MW plant
Fixed O&M ($/kW-yr) NA
Variable O&M (mills/kWh) 1.45
Applicability Both new and retrofit applications
. Early stage of development; demonstration and further
assessment of the technology is needed.
Technology Description
EnviroScrub's Pahlman Process is a "closed-loop" dry sorbent system comprised of two discrete steps.
One step involves capturing the 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 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 injected in a reactor, which operates at temperature between ambient and 320 °F.
According to the supplier, different type of reactors are suitable including fluidized bed, baghouse,
transport and cyclone. SO2 and NOX react with the sorbent according to the following reactions:
SO2 + MnO2 -^ MnSO4
2NO2 + MnO2 -» Mn(NO3)2
Mercury vapor reacts with the Pahlmanite sorbent, which promotes oxidation to HgO followed by
sorption by MnO2.
Sorbent regeneration is a wet chemical process, which involves the following steps:3
1. Reacted ("loaded") sorbent is transported into a regeneration vessel where it comes in contact
with a hot, oxidizing, aqueous solution, which dissolves the nitrate and sulfate salts of
manganese. The chemical composition and operating conditions of the regeneration vessel are
controlled so that the dissolved manganese is precipitated to form fresh sorbent; also, the
remaining solid sorbent is re-oxidized and re-activated.
2. The slurry consisting of nitrates and sulfates is transported from the regeneration vessel to a
washing and filtration process, which separates the solids from the liquid fraction. The solids are
subsequently dried and returned to the reactor for reuse.
3. The liquid fraction is separated by membrane technology or other means into concentrated
solutions containing sulfate, nitrate and mercury compounds. These solutions may then be
3-67
-------
precipitated or evaporated to be converted to a solid form for either disposal or by-products.
Excess water is returned to the regeneration vessel.
4. Some of the sulfate-containing liquid from the washing and filtration system is sent to a chemical
regeneration process to produce base and oxidizer chemical which can be used in the Pahlmanite
sorbent regenerator.
Commercial Readiness and Industry Experience
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 (1,000 scfm):
• Ameren Energy's Hutsonville Power Station in Hutsonville, Illinois, which burns high-sulfur Eastern
Bituminous coal, without the use of an emissions control system for SO2 and NOX.
• Minnesota Power's Boswell Energy Center located in Cohasset, Minnesota, which consists of four
coal-fired steam boilers (total: 1000 MW) burning PRB (Powder River Basin) coal. The
demonstration of the EnviroScrub process was conducted on the exhaust gas stream of Unit 1, a 75
MW boiler, which currently operates without any form of back-end emission control systems for NOX
and SO2.
• Potlatch Corp. Northwest Paper Division's facility located in Brainerd, Minnesota. The demonstration
of the EnviroScrub process was conducted on boiler #1, a Zurn Industries 75,000 Ib/hr steam boiler,
operating exclusively on natural gas with NOX concentration output of approximately 133 ppm.4
Emission Control Performance
Results from Minnesota Power's Boswell testing indicate that the technology achieved above 99 percent
SO2 reduction and 94-97 percent NOX reduction.1 Also, the Hutsonville facility achieved over 99 percent
SO2 and 75 percent NOX reduction4 on 1750 ppm SO2 and 300 ppm NOX inlet.
According to the supplier, the technology also controls mercury and particulates (PM2 5). Testing at
Boswell indicated mercury removal up to 67 percent.4
O&MImpacts
O&M impacts are not known, mainly because of the lack of information. At least auxiliary power is
expected to increase, but no specific estimates are available.
Capital Costs
According to the supplier, capital costs for a 500 MW plant are projected to be in $150/kW.2
O&M Costs
Variable O&M costs are projected by the supplier to be approximately 1.45 mills/kWh.2
Issues Associated with the Technology; Future Outlook
The technology is still at an early development stage (pilot plant) and requires further demonstration and
techno-economic assessment to develop a more comprehensive picture of its cost-effectiveness. Also,
removal efficiencies of mercury and particulates need to be demonstrated.
References
1. Interpoll Laboratories Inc. Results of the Nov 8, 2001 Air Emission Monitoring on EnviroScrub
Technologies; Mobile Demonstration Pilot Scrubber Ducts at Minnesota Power's Boswell Energy
Center in Cohasset, MN; Submitted to Minnesota Power, Dec 18, 2001.
3-68
-------
2. Coal Plant Achieves 99% NOX and SOX Removal with Multi-pollutant Technology. Power Eng.
Magazine May 2002.
3. Pahlman™ Process Overview; Brochure; Enviroscrub.
http://www.enviroscrub.com/lit/ScrubBro.pdf (accessed Dec 1, 2004).
4. Modern Power Systems. Can Enviroscrub Clean Up in the Multi-pollutant Control Business?
Nov 2002.
3.2.3.5 LoTOx
LoTOx Summary
Status Early commercial (commercial in metal industry)
SO2 Reduction, % 95% due to the FGD which is part of the system
NOX Reduction, % 70-95%
Hg Reduction, % Up to 90%
CO2 Change, % NA (1-2.5% increase due to auxiliary power requirements)
Cost
Capital ($/kW) $35-70/kW (for LoTOx only; not including FGD)
Fixed O&M ($/kW-yr) NA
Variable O&M
(mills/kWh)
Applicability Both new and retrofit applications
Issues Further demonstration is needed at utility scale.
Technology Description
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.1"7
Subsequently these compounds are removed in a wet FGD, because they are water-soluble. As Figure
3-88 shows, the LoTOx system consists of an integrated ozone-from-oxygen generation unit complete
with ozone injection system into the LoTOx.
Ozone is injected into the 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, in response to the amount of NOX present in the flue gas
generated by the combustion or process source. Upon injection of the ozone in the flue gas (typically
below 300 °F), oxidation occurs according to the following simplified reactions:9"12
NO + O3 -» NO2 + O2
2NO2 + O3 -» N2O5 + O2
Hg° + O3 -> HgO + O2
N2O5 and HgO are water soluble. In the presence of water, N2O5 forms HNO3, which further reacts with
alkali compounds or alkaline earth metals to form corresponding nitrates:
2HNO3 + CaCO3 -> Ca(NO3)2 + CO2 + H2O
The selection of wet FGD type (lime, limestone or ammonia) does not impact the performance of the
LoTOx process because the solubility of N2O5 is significantly higher than that of SO2.
3-69
-------
Theoretically, there is the potential for oxidation of SO2to SO3; however, as proven in field testing, the
reaction rates are very low compared to the predominant NOX reactions.
Air Purifier Air Compressor
-0-—CTD
Induced Draft
Fan
TM
Stack
Figure 3-8. Schematic diagram of LoTOx system.
Commercial Readiness and Industry Experience
The technology has been demonstrated up to a scale of 25 MW, at boiler #4 of the Medical College of
Ohio (MCO) capable of burning high-sulfur Ohio coal.13 Commercial experience was also gained in the
steel industry prior to the MCO demonstration. For example: a 4,500 scfm metal pickling process and a
natural gas-fired boiler of 8,000 scfm are on-going applications. Also, LoTOx has been installed at a lead
recovery furnace (25,000 scfm) which is operating presently.14 LoTOx technology for NOX control is
being considered by the refining industry to be used with wet SO2 scrubbers and a number of commercial
installations are expected to be operating by 2005-07. Based on these developments, the supplier (BOC)
is prepared to provide the technology for commercial use with appropriate guarantees.
Emission Control Performance
Up to 95 percent NOX reduction has been achieved and projected for large-scale applications. As Figure
3.9 shows, NOX outlet levels of less than 5 ppm have been achieved. Similar NOX reductions have been
achieved in other facilities. For example, the 4,500 scfm metal pickling application achieved
approximately 84 percent NOX reduction from 1100 ppm inlet, while the natural gas-fired boiler of 8,000
scfm achieved nearly 98 percent NOX reduction (4 ppm outlet NOX emissions). Finally, LoTOx at the
lead recovery furnace averaged 80 to 95 percent NOX reduction.8'15
BOC normally recommends installation of LoTOx alone when inlet NOX is below 0.3 Ib/MMBtu. Above
0.3 Ib/MMBtu, LoTOx is recommended as part of an integrated control approach operating in series with
an alternate control process that is capable of moderate NOX removal. By combining the two
3-70
-------
technologies, users may be able to avoid installing an SCR system, which is expected to have higher
capital investment and operating costs.
Oxidized mercury species are removed in a forced oxidation wet FGD, which is downstream of the
LoTOx system, and removal rates depend primarily on the coal characteristics. As the Table 3-7 shows,
LoTOx is reported to have enhanced the mercury removal of FGD for all coals, especially subbituminous
coal and lignite.16
According to the supplier, LoTOx enhances the SO2 removal efficiency of the FGD by approximately
5 percent (depending on the FGD design) and has no impact on SO3 emissions.
O&MImpacts
O&M impacts are expected to be minimal since the LoTOx technology is injecting ozone in the flue gas.
According to the supplier, auxiliary power requirements for a 500 MW plant are projected to be
approximately 5.0 - 12.5 MWs or 1-2.5 percent of the gross power output.
Also, cooling water is required; which for a 500 MW plant is estimated at 4,500 gal/min and assumed
available at 70 °F.
Costs
Capital costs are estimated (by the supplier) to be in the $35 to 70 per kW range depending on inlet NOX
level and unit size.
The supplier estimates the fixed O&M costs to be approximately 2.5 percent of the capital costs. In the
electric utility industry, annualized costs are typically $1,200-2,200 per ton of NOX removed, inclusive of
mercury removal.
Issues Associated with the Technology; Future Outlook
The technology has yet to be demonstrated at a utility scale plant based on which the performance and
cost projections could be verified.
Table 3-7. Mercury Data for Various Coals8
Coal Type
Bituminous
Sub-Bituminous
Lignite
Typical Hg2+ as %
of Total Hg
70-85%
15-45%
10-30%
Hg Removal with
FGD Alone
76%
33%
19%
Hg removal with
LoTOx and FDG
94%
92%
91%
3-71
-------
Performance of LoTOx System at MCO
December 6,2001
Outlet NOx Setpoint = 5 PPM Changed to 10 PPM
PH
PH
50
40
30
10
40000
Flue Gas Temp = 179 - 234 F
Scrubber Temp = 75 -100 F
Inlet SOx = 74-178 ppm
Inlet CO = 150 - 300 ppm
20000 E
15000
10000
5000
Time
'60 Min Avg Inlet NOx PPM ^~60 Min Avg Outlet NOx PPM 60 Min Avg Flue Gas Flow ACFM
Figure 3-9. NOX with LoTOx at MCO's 25 MW-thermal gas and coal fired boiler.
13
References
1. Skelley, Authur P.; et al. Process for Removing NOX and SOX from Exhaust Gas. U.S. Patent
5,316,737, May 31, 1994.
2. Skelley, Authur P.; et al. Process for Removing NOX and SOX from Exhaust Gas. U.S. Patent
5,206,002, April, 27, 1993.
3. Skelley, Authur P.; et al. Low Temperature NOX/ SOX Removal. U.S. Patent 4,999,167, May 12,
1991.
4. Saxena, Neeraj; et al. Removal of NOX and SOX Emissions from Pickling Lines of Metal
Treatment. U.S. Patent 5,985,223, Nov 16, 1999.
5. LTO System Performance Test, Alta Dena Dairy; South Coast Air Quality Management District
Report, Feb 13, 1998.
6. Ferrell, R. Controlling NOX Emissions: A Cooler Alternative. Pollut. Eng. April 2000.
7. Fu, Y.; Diwekar, U. M.; Suchak, N.J. Optimization Framework for Modeling the Low
Temperature Oxidation Process for NOX Reduction. Adv. Env. Res. 2000, 3(4), 424-438.
3-72
-------
8. Jarvis, J. B.; Day, A.;T.; Suchak, N. J. LoTOx™ Process Flexibility and Multi-pollutant Control
Capability. Proceedings of the Combined Power Plant Air Pollutant Control MEGA Symposium,
Washington DC, May 19-22, 2003.
9. 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. Presented at Power-Gen 2001,
Las Vegas, NV, Dec 12, 2001.
10. Ferrell, R.; Suchak, N.; Hwang, S.C.; Tseng, J.; Kelton, R.; Goss, W.; Lutwen, R. A Report on
the Application of Low Temperature Oxidation for Control of NOX Emissions. Presented at
ICAC Forum 02, Houston, TX, Feb 13, 2002.
11. Anderson, M. H. A Low Temperature Oxidation System for the Control of NOX Emissions Using
Ozone Injection. Presented at the ICAC Forum'98, Durham, NC, March 1998.
12. Ferrell, R.; Anderson, M.; Suchak, N.; Kelton, R.; Cuddeback, B.; Hwang, S. C. LoTOx System
Demonstration for Coal-Fired Combustor. Presented at ICAC Forum'OO, Arlington, VA, March
2000.
13. Goss, W. L. Multi-pollutant Control System Installation at Medical College of Ohio; Technical
Transfer Paper; June 28, 2002.
14. Ferrell, R.; Suchak, N.; Kelton, R.; Hine, H.; Nadkarmi, S. Application of Low Temperature
Oxidation for NOX Emissions Control on a Lead Recovery Furnace. Presented at the Annual
A&WMA Meeting and Exhibition, San Diego, CA, June 2003.
15. Ferrell, R.; Anderson, M. H.; Hwang, S. C.; Suchak, N. J.; Kelton, R.; Tseng, J. T. Applications
and Economics of Low Temperature Oxidation NOX Control. Presented at Power-Gen 2000,
Orlando, FL, Nov 14, 2000.
16. Ellison, W. Chemical Process Design Alternatives to Gain Simultaneous NOX Removal in
Scrubbers. Presented at the ICAC Forum'03, Nashville, TN, Oct 2003.
3-73
-------
3.2.3.6 K-Fuel
K-Fuel Summary
Status
SO2 Reduction, %
NOX Reduction, %
Hg Reduction, %
Cost
Capital ($/kW)
Fixed O&M ($/kW-yr)
Variable O&M
(mills/kWh)
Applicability
Issues
Components of technology commercial
Up to 30% (relative to raw PRB coal or lignite)
Up to 45% (relative to raw PRB coal or lignite)
Up to 70% (relative to raw PRB coal or lignite)
Indicative prices: $20 per ton for K-Fuel vs $6.50 ton PRB, both FOB
in Wyoming
NA
NA
NA
The supplier is focusing on boilers burning subbituminous and lignite
because the process is effective for coals with high ash and
moisture. K-Fuel could apply to boilers burning bituminous coal too,
but significant boiler modifications may be required as is the case of
switching from bituminous to PRB.
A plant-specific evaluation is required to assess the cost
effectiveness of a switch to K-fuel, especially when the overall
benefit of the emission reductions is taken into consideration.
Technology Description1
K-Fuel is a beneficiated coal that is derived from PRB coal or lignite 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 burning the coal at the plant,
the need for post-combustion controls may therefore be reduced. Technology may be applicable to
bituminous coal. However, the supplier has focused exclusively on PRB and lignite applications, because
the K-Fuel is a moisture and ash reduction process, which PRB and lignites are high in.
The K-Fuel pre-combustion multi-pollutant reduction technology (PMR) is a two-step process, illustrated
in Figure 3-10.
1. Physical Separation: A gravity separation process - either wet or dry - is used to remove ash
along with other pollutants (sulfur and mercury). This involves crushing and screening to remove
the larger particles. Also, high-energy magnetic separation can be used, especially for coals with
a higher percentage of fine particles. The processed coal is then passed on to an intermediary
storage facility prior to being sent to the next step of the process. Wastes from physical cleaning
are returned to the coal mine.
2. Thermal Processing, which follows, employs Lurgi Mark IV vessels operating under high
temperature and pressure (460 °F and 485 psi). Mineral matter of the coal under thermal stress
fractures in these vessels liberating moisture, as well as sulfur and mercury. Figure 3-11 shows
the main components of the thermal processing unit.
The water and mercury vapors are condensed, and mercury is captured in a carbon-bed adsorption reactor.
Water is recycled in the process. Mercury-laced activated carbon is disposed in atoxic landfill.
3-74
-------
Physics
Separator
R=
^r
_-3lV
3' Conle -r.
J
L—3
TT-n-al
\Va'.er.
hg VELO-
I
V=rcLV
Enlri= nT=nl
h qh B'.J Oplin^^d
L—
M=-cu^-
DiSL-=S£
Figure 3-10. K-Fuel PMR technology.
Physically
Upgraded Coal
Surge
Hoppers
K-Fuet
Thermal
Processor
H, Water/Hq _ ^
1 Vapor *" ^
j ^ Process
Steam
Condenser
Water/Hg
41 ^ U-
Hg Removal
_H Process ^
Water
(recycle)
Processed
Fuel
Figure 3-11. K-Fuel thermal separation unit.
3-75
-------
Commercial Readiness and Industry Experience
The components of the PMR process are proven commercially. Physical cleaning is a well-proven
technology and thermal separation is used worldwide, with Lurgi having approximately 160 thermal
separation units in operation. In the U.S., thermal processing has been used since 1984.2 The first
commercial plant, 700,000 tons per year, is expected to be completed by the middle of 2004.
The K-Fuel has been tested in smaller scale facilities and a commercial plant, including Southern
Research Institute (SRI) and at American Electric Power's (AEP) Clifty Creek Station in Indiana.
Emission Control Performance
According to KFx, the developer of K-Fuel, physical separation studies on a number of low rank coals,
exhibited ash reduction in the 10-30 percent range, 10-36 percent sulfur reduction, and 28-66 percent
mercury reduction. Thermal separation adds more mercury reduction; for example, testing by Rio Tinto
Technical Services in Perth, Australia showed a 40 percent reduction3 in fuel mercury due to thermal
separation alone. Total mercury reduction at the Energy & Environmental Research Laboratory of the
University of North Dakota4 and the Western Research Institute5 were reportedly 66-67 percent. Also,
NOX reduction up to 46 percent has been experienced in tests at Southern Research Institute6 and 40
percent at AEP's Clifty Creek Station. All emission reductions are relative to similar type raw coal (PRB
or lignite).
Finally, reduced moisture in the coal is expected to improve boiler efficiency. For a typical Wyoming
coal, moisture is reduced from 20-30 percent to approximately 6 percent. Such a change may improve
boiler efficiency by up to 2.0 percent.
O&M Impacts
Most O&M impacts are expected to be positive including reduced auxiliary power and O&M costs
associated with the coal handling equipment, pulverizers, particulate collection, and ash handling
equipment. Also, less ash reduces transportation and disposal costs. Finally, more consistent coal quality
with lower ash would have a positive impact on plant reliability. However, there is inadequate
information to quantify such impacts.
While the supplier is targeting mainly boilers burning PRB coal or lignites for this technology, K-Fuel
could be used in boilers designed for bituminous coal, too. However, in the case of bituminous coal, the
O&M impacts could be significant including lower steam temperatures, higher unburned carbon, slagging
and potential de-rating (reduced output). Such impacts could be eliminated through boiler modifications,
but some investment would be required.
Costs
It is envisioned that the coal produced by the PMR process will be sold to the power plants rather than the
power plants having to set-up their own PMR process facility; if so, the price of the processed coal is the
important parameter. Presently, there is no adequate information to estimate the price of the processed
coal, but it would need to reflect the costs of selective mining, capital and operating costs of all
components (physical and thermal separations and carbon bed absorber for mercury removal) and
disposal costs of mercury and other wastes. Also, the price will have to be competitive with other coal; in
other words, the price for K-Fuel may be based on the price of competing coal on heating value basis plus
credits for environmental benefits (lower SO2, NOX and Hg reductions). Preliminary estimates suggest
that K-Fuel may be available in the $20-25 per ton range FOB Wyoming. Presently, this compares to raw
Wyoming coal of $6.50 per ton FOB Wyoming. However, the K-Fuel may also compete against Eastern
bituminous coal.
3-76
-------
Issues Associated with the Technology; Future Outlook
The technology is suitable mainly for western coals. Also, not adequate information is available to assess
its cost-effectiveness. With the operation of the Black Thunder plant by mid-2004 more information is
expected to become available.
References
1. Black & Veatch. Effective Mercury Reduction Strategy for Western Coal/K-Fuel Technology,
Report to the U.S. EPA, March 2003.
2. Ness, M. Improving Power Plant Performance and Reducing Emissions through Coal
Beneficiation. Presented at the Energy Generation Conference, Bismarck, ND, Jan, 2002.
3. Davies, M. Technical Report- RevisedK-FuelProcess; Rio Tinto Technical Services: Perth,
Nov 30, 2000.
4. Gunderson, J. R. Combustion testing of K-Fuel for Heartland Fuels; Energy & Environmental
Research Center, Nov, 1993; 59.
5. Vesperman, K. Utility SO2 Compliance with a Long-range Forecast. Presented at the Low-rank
Fuels Symposium, St. Louis, MO, May 10-13, 1993.
6. Monroe, L. S. Testing K-Fuel Synthetic Coal In The Southern Research and Southern Company
Combustion Research Facility; Southern Research Institute, Aug 1998.
7. Janssen, K. E. Lignite-to-gas Plant Reveals Numerous Innovations. Power May/June 1997,
79-82.
8. Alderman, J. K. Western Coal Preparation - Meeting the Demands for Clean Energy. Coal Age
March 2003, 707(3), 24-30.
3-77
-------
3.2.4 Advanced Power Generation Technology Options
3.2.4.1 Circulating Fluidized Bed Technology
Circulating Fluidized-bed Technology Summary
Status
SO2 Reduction, %
NOX Reduction, %
Hg Reduction, %
CO2 Change, %
Cost
Capital ($/kW)
Fixed O&M ($/kW-yr)
Variable O&M (mills/kWh)
Applicability
Issues
Commercial up to 250 MW
>95
30 - 70 depending on the coal [90 with SNCR (ammonia or
urea)]
NA
Negligible
850 - 1100 (for a new CFB plant In the 100 - 250 MW
range)
30-37
2.0-5.5
Mainly new power plants; also for replacement of old boilers
while utilizing the existing turbine and balance of plant
(BOP)
Successful scale-up to 400 - 600 MW while maintaining its
cost-effectiveness and emission performance
Note: Emission reduction is based on comparison of this technology to a similar size subcritical
pulverized coal boiler with low NOx burners, but without FGD.
Technology Description
Circulating Fluidized Bed (CFB) combustion technology is one of the two variations of Atmospheric
Fluidized Bed Combustion (AFBC), the other one being the bubbling AFBC. CFB has been the
predominant design used in coal-fired power applications, especially in large-scale plants above 100 MW.
Bubbling AFBC is used too, but the most recent plants are small, 10-50 MW, burning biomass and
municipal solid wastes. CFB technology is described in this section because of its applicability for large
power applications.
CFB boilers (see Figure 3-12) are very similar to conventional PC boilers in many respects. The majority
of boiler components are similar, and hence manufacturing of the furnace and the back-pass can be done
in existing manufacturing facilities. In addition, a CFB boiler utilizes the Rankine steam cycle with steam
temperatures and pressures similar to PC boilers. CFB boilers can be designed for either subcritical or
supercritical conditions. Most CFB boilers utilized so far are of the subcritical type mainly because the
technology has been utilized in sizes up to 250 MW where subcritical operation is more cost-effective. As
the technology is scaled up (above 400-500 MW), the supercritical design may be used depending on the
site-specific environmental requirements or cost of fuel.
3-78
-------
Figure 3-12. CFB process schematic.
The difference of CFB relative to PC boiler stems from the lower operating temperature and the injection
of limestone in the furnace to capture SO2 emissions. Typical maximum furnace temperature in a CFB
boiler are in the 1500-1600 °F (820-870 °C) range, while conventional PC boilers operate at 2200-2700
°F (1200-1500 °C). This low combustion temperature limits the formation of NOX and is the optimum
temperature range for in-situ capture of SO2 The injected limestone is converted to lime, a portion of
which reacts with SO2 to form calcium sulfate (CaSO4), a dry solid, which is removed in the particulate
collection equipment (either ESP or FF). A cyclone is located between the furnace and the convection
pass to capture unreacted lime and limestone present in the flue gases exiting the furnace. The solids
collected in the cyclone are recirculated back to the furnace to improve the overall limestone utilization.
Limestone injection can remove from 90 to >95 percent of the sulfur in the coal1 eliminating the need for
flue gas desulfurization (FGD) downstream of the boiler. CFBs have NOX emissions 60-70 percent lower
than conventional PC boilers with low NOX burners
CFB boilers can efficiently burn low reactivity and low-grade fuels, which may not be burned in
conventional PCs. Such fuels include anthracite, coal cleaning wastes, and industrial and municipal solid
wastes. High-ash fuels, such as lignite, are particularly suitable for CFB technology.
3-79
-------
Commercial Readiness and Industry Experience
Commercial applications of this technology exist in sizes up to 265 MW net (-300 MW gross) unit size,
as demonstrated by hundreds of boilers operating throughout the world (Australia, China, Czech
Republic, Finland, France, Germany, India, Japan, Poland, South Korea, Sweden, Thailand, and United
States). Vendors are now offering boiler sizes for plants well over 300 MW in sizes. Foster Wheeler has
signed a contract to supply a 460 MW boiler with super-critical steam conditions at the Lagisza plant in
Poland.3 This plant is scheduled to be started up by the end of 2005. In 1996, EPRI estimated that there
are approximately 300 CFB units (larger than 22 tons/hr of steam capacity each) in operation worldwide.
Since then (1996), the number of CFB operating units has increased further. Experience from these units
has confirmed performance and emissions targets, high reliability, and the ability to burn a variety of low
quality fuels.
The most recent CFB projects on a scale larger than 200 MW are shown in Table 3-8.
Table 3-8. CFB Units larger than 200 MW
Company,
Plant Name
EOF,
Gardanne
Turow Power
Co., Turow
Station
KEPCO,
Kangwon-do
AES,
Warrior Run
JEA,
Jacksonville
Sithe/Tractabel
Red Hills
Location
Provence,
France
Silesia,
Poland
Kangwon-do,
South Korea
Warrior Run,
Maryland, USA
Jacksonville,
Florida, USA
Red Hills,
Mississippi, USA
Supplier
Stein (Lurgi)
Foster
Wheeler
KHI and
ABB-CE
ABB-CE and
Lurgi
Foster
Wheeler
Alstom (Stein)
Net
MW
232
2x230
2x200
2x200
265
2x220
Start-
up
1996
1999
1998
1999
2002
2001
Fuel Type
Subbituminous Coal (30% ash/4% S)
Brown Coal (23% ash/44% H2O/0.6% S)
Korean Anthracite
Bituminous Coal
Bituminous Coal,
Petroleum Coke
Lignite
Emission Control Performance
CFBs have demonstrated that they can remove up to 90 - 95 percent1 of the sulfur in the coal without the
need for flue gas desulfurization (FGD) downstream of the boiler. SO2 removal is affected mainly by the
bed temperature and by the Ca-to-S molar ratio. The bed temperature is designed for optimum sulfur
capture [1500-1600 °F (820 - 870 °C)]. Ca-to-S molar ratio depends on the amount of CaO and MgO in
the ash, as well as their reactivity. A Ca-to-S molar ratio of 2.0 - 2.5 is typical for 90 - 95 percent SO2
removal.
NOX emissions are 30 - 70 percent lower than conventional PCs with low NOX burners;1 CFB boilers
have achieved consistently 0.12-0.16 Ibs/MMBtu NOX emissions, compared to pulverized coal plants
which range from 0.2 to 0.5 Ibs/MMBtu. Further NOX reduction can be achieved by installing an SNCR
(ammonia or urea) injection system in the CFB furnace. NOX emissions in the 100 ppm level without
ammonia or urea injection and below 20 ppm (0.027 Ibs/MMBtu) with ammonia or urea injection have
been demonstrated.1 NOX emissions are impacted by the bed temperature, the nitrogen and volatile matter
in the coal, and the bed stoichiometry. Stoichiometry, defined as a ratio of available air (weight) to
theoretical air needed for complete combustion, is affected by the excess air and air distribution across the
3-80
-------
bed. The amount of ammonia or urea being injected (in case such system is available) also impacts the
additional NOX removed as a result of such an injection.
It is not clear whether CFB controls mercury, too. CFB plant efficiency is similar to new subcritical
pulverized coal plants equipped with NOX and SO2 controls, typically 34-38 percent (higher heat value
basis). Therefore, no significant CO2 reduction is expected. Supercritical CFBs will have higher
efficiency (38-40 percent) and proportionally lower CO2 emissions.
O&MImpacts
None.
Capital Costs
Capital costs of large CFB plants (between 150 and 250 MW) range from $850 to 1100/kW, similar to or
slightly above the costs of pulverized coal plants with FGD, which range from $800 to 1000/kW.1
O&M Costs
O&M costs are projected to be $30-37/kW-yr for fixed O&M and 2.0-5.5 mills/kWh for variable O&M
costs depending on the O&M practices of the utility, labor costs, and cost of consumables (especially
sorbent).1
Issues Associated with the Technology; Future Outlook
The main barrier to widespread utilization of the technology is scale-up to larger sizes (present focus on
400 - 500 MW) while maintaining its cost-effectiveness and emission control performance. This barrier is
expected to be overcome in the next 2-3 years as indicated by a number of studies and demonstration
projects being planned.2 One such effort is led by Electricite de France (EdF), which sponsored a study to
develop a 600 MW CFB design. Participating in this study are Alstom and Foster Wheeler, two of the
leading CFB vendors.
Presently, CFBs are being built worldwide, especially for solid fuels difficult to burn in a pulverized coal
boiler such as anthracite, lignite, brown coal, and coal wastes. Also, industrial and municipal solid wastes,
petroleum coke, and other waste fuels are being burned in most cases as supplemental fuels. CFB
technology is expected to be used widely in the future, mainly in new power plant applications.
References
1. Tavoulareas, S. Financing Clean Coal Technologies; World Bank: Washington, DC, July 2001.
2. Dyr, R. A.; Hebb, J. L.; Darling, S. L. The JEA CFB Demonstration Project: An Update.
Presented at the PowerGen-Europe, Helsinki, Finland, June 20-22, 2000.
3. Foster Wheeler Website - News Release 3/7/2003. http://www.fwc.com/ (accessed Dec 1, 2004).
3-81
-------
3.2.4.2 Integrated Gasification Combined Cycle
Integrated Gasification Combined Cycle Summary
Status
SO2 Reduction, %
NOX Reduction, %
Hg Reduction, %
CO2 Change, %
Cost
Capital ($/kW)
Fixed O&M ($/kW-yr)
Variable O&M (mills/kWh)
Applicability
Issues
Entrained and moving beds: commercial up to 500 MW;
fluidized bed: in demonstration stage
Up to 99
80-90 (compared to PC with low NOx burners)
Needs characterization via measurements
10 to 20% potential reduction over a comparably-
sized PC plant with FGD
1200 - 1600 (for new plant of 400 MW size)
30-45
0.5-2.0
New power plants or retrofits utilizing existing plant steam
turbine-generators and balance of plant equipment
High costs are the main barrier to widespread utilization of
entrained and moving bed IGCC. Fluidized bed IGCC
requires demonstration.
Technology Description
The integrated gasification combined cycle (IGCC) involves gasification of coal with either oxygen or air,
with the resulting syngas (an abbreviation for synthetic gas) cooled, cleaned, and fired in a gas turbine.
The hot exhaust from the gas turbine passes to a heat recovery steam generator (HRSG) where it produces
steam that drives a steam turbine.1'2 Power is produced from both the gas and steam turbines; hence,
combined cycle (CC).
The gasification plant could be designed to operate under atmospheric or pressurized conditions.
Pressurized gasification is preferred to avoid large auxiliary power losses for compression of the syngas,
which is required in an atmospheric IGCC, before the gas enters the gas turbine. Pressurized gasification
also reduces the size of the gasifier. Most gasification processes currently in use or planned for IGCC
applications are oxygen blown; however, the Pinon Pine Plant in the U.S. uses an air-blown fluid bed
process [developed by Kellogg Rust Westinghouse (KRW)].
The electric output of an IGCC plant is mainly determined by the firing temperature of the gas turbine
[typically: 1100 °C (2000 °F) or 1260 °C (2300 °F) for standard designs offered commercially] and by the
frequency of the electricity produced. Typical net output for single-train IGCC plants is approximately
275 MW for 60-Hz markets and 400 MW for 50-Hz markets. Plant net efficiency is typically 40 - 44
percent on a higher heating value (HHV) basis. Utilization of the new gas turbines (G- and H-class)
would increase the overall plant efficiency to 46 - 48 percent (HHV basis) with an output of 400-450
MW (in 60-Hz markets) and 500 - 550 MW in 50-Hz markets.
The IGCC design varies, especially in the degree of integration. For example, some IGCC plants such as
the Buggenum (Netherlands) and Puertollano (Spain) are highly integrated designs with all the air for the
air separation unit being taken as a bleed from the gas turbine compressor. In contrast, the U.S. plants at
Tampa and Wabash River are less integrated, and the air separation units have their own separate air
compressors. The more highly integrated design has higher plant efficiency, but lower plant availability
and operating flexibility. Presently, the general consensus among IGCC plant designers is to have the air
3-82
-------
separation unit derive part of its air supply from the gas turbine compressor and the remainder from a
separate dedicated compressor.
Commercial Readiness and Industry Experience
In general, IGCC technology has been demonstrated up to 500 MW in size and offered commercially, but
higher costs (relative to the conventional PC with FGD) limit its widespread utilization. Presently, there
are six coal-based IGCC demonstration plants in operation (three in the U.S and three in Europe) with
three more IGCC plants utilizing refinery wastes in Italy. Table 3-9 shows these plants, as well as the first
IGCC demonstration, Cool Water IGCC in Barstow, CA, which operated from 1984 to 1989.
In addition, there are approximately 160 gasification facilities in 28 countries being utilized by the
petrochemical industry. These are mainly gasifiers using heavy oil, petroleum bottoms, or petroleum
coke. While this experience is not directly applicable to coal-fired IGCCs, it is clearly helpful because
some of the technical problems and operating issues have been addressed.
Table 3-9. Commercial-Size IGCC Power Plants
Project name and location
Cool Water, Mojave, CA, US
SEP-Demkolec, Buggenum,
The Netherlands
Wabash River, West Terre Haute, IN, US
Tampa Electric, FL, US
Sierra Pacific Pifion Pine, NV, US
ELCOGAS, Puertollano, Spain
Schwartze Pumpe, Schwartze Pumpe, Germany
ISAB Energy, Sicily, Italy
Sarlux, Sardinia, Italy
API Energia, Falconara, Italy
Gasification
Technology
Texaco
Shell
Destec
Texaco
KRW fluid bed
Krupp-Uhde Prenflo
Lurgi moving bed
Texaco
Texaco
Texaco
MW
(gross)
120
253
296
312
107
335
75
512
548
280
Startup Date
1984
1994
1995
1996
1998
1997
1999
2001
2000
2001
Note: Above gasifiers are of entrained type, unless otherwise indicated.
Samples of additional projects in the planning stage include:
• Vresova, Czech Republic: 385 MW lignite-firing; upgrading of existing facility and increase of plant
output;
• Port Arthur, TX: 6,985 tonnes/day (7,700 U.S. ton/day) petroleum coke gasification plant; and
• Citgo's refinery at Lake Charles in Louisiana: 670- MW IGCC utilizing petroleum coke.
Actual efficiencies of IGCC plants have been measured in the 38-43.2 percent range (HHV-basis), mainly
because their operation was designed in a conservative way to demonstrate the reliability of the
technology and its ability to reduce acid rain pollutants rather than to achieve the highest efficiency.
Utilization of parts of existing plants in IGCC repowering cases also contributed to keeping the efficiency
on the low side. For example, the Wabash River IGCC repowering project achieved 39.7 percent
efficiency (this was 20 percent higher than the plant's efficiency prior to repowering). Similarly, the
Tampa (Polk) IGCC plant achieved 38 percent3 while Puertollano IGCC plant achieved 43.2 percent
efficiency.4
3-83
-------
Three types of gasification processes are available: moving-bed, fluidized-bed, and entrained-bed, each of
them at a different stage of development.
Entrained-flow gasifiers operate at high temperatures (between 1300 and 1500 °C) where the ash is
melted and removed from the bottom of the gasifier as slag. This type of gasification is suitable for low-
ash coals (less then 10 - 15 percent ash) and is the most widely used IGCC process. The main suppliers
are: Texaco (Tampa and Cool Water plants in the U.S., and three refinery waste IGCC plants in Italy) and
Destec (Wabash plant in the U.S.) offering coal-water-slurry-fed processes, and Shell (Buggenum,
Netherlands), Krupp-Uhde (Puertollano, Spain), GSP (Schwarze Pumpe, Germany), and Mitsubishi
(Nakoso, Japan) offering a dry-coal-fed process.
In moving-bed dry ash gasifiers, steam is injected along with the oxygen to keep the coal ash well below
its ash fusion temperature. This type of gasifier is offered mainly by Lurgi and has been used in many
countries including China, the Czech Republic, Germany, South Africa, and the United States. Also, a
slagging version of the Lurgi gasifier has been developed, and a commercial-sized unit based on this
technology has been commissioned at Schwarze Pumpe in Germany.
Fluidized-bed gasification, which is more suitable for high-ash coals, is still in the demonstration stage. A
number of small atmospheric-pressure Winkler gasifiers have been built in Germany, India, and Turkey.
Also, the U-Gas gasifier developed by the Institute of Gas Technology (IGT) in the U.S. is utilized in the
Shanghai Coking and Chemical plant in China. A high-temperature Winkler gasifier has been developed
by Rheinbraun and used in Germany and Finland for methanol and ammonia manufacture. The same
process is expected to be used in the Vresova IGCC plant in the Czech Republic. Presently, the largest
plant is the 100- MW Piiion Pine in Reno, NV, utilizing the KRW fluid-bed process, but it is experiencing
operating problems and is still in an extended commissioned program.
Emission Control Performance
By removing the emission-forming constituents (sulfur and nitrogen species and particulates) prior to
combustion in the gas turbine, IGCC plants meet extremely stringent air emission standards. Sulfur
emissions can be almost completely eliminated; SO2 emissions are expected to be 0.03-0.1 Ib/MMBtu,
(40-115 mg/Nm3 at 6 percent O2). NOX emissions have been controlled to levels below 0.1 Ib/MMBtu
(125 mg/Nm3 at 6 percent O2) at two of the demonstrations using steam or nitrogen dilution in the
combustor and to half that level at two other sites operating at lower combustion turbine temperatures.
Recently, General Electric (GE) has claimed that they can meet a NOX level of 60 mg/Nm3 even with
their 2300 °F (1260 °C) series FA gas turbine. Typical CO2 emissions will be 12 - 15 percent lower than
a comparably-sized PC plant with FGD, with further reduction possible with the utilization of the higher
efficiency (G- and H-class) gas turbines. There are indications that IGCC plants equipped with carbon bed
adsorption systems may be able to achieve 90 percent mercury removal.5
Emissions are affected mainly by the requirements of the gas turbine to maintain low particulates and SO2
emissions. NOX emissions are affected by the design of the gas turbine combustor and the environmental
requirements of the plant site; most gas turbines utilize a dry low NOX combustor design and have
achieved NOX emissions below 25 ppm(v) with the most recent turbine designs targeting below 10
ppm(v).6 Assuming 25 ppm NOX emissions, this results in 80-90 percent reduction compared to
pulverized coal boilers with low NOX burners which experience 0.20-0.50 Ibs/MMBtu.
Limited data available (mainly from Wabash River, Polk, and the Louisiana IGCC plants) suggest that
mercury emissions at the stack are similar to pulverized coal-fired plants, ranging from 1.5 to 5 Ib/trillion
Btu. While the mass balance closure in these plants was not good, there is good evidence that mercury is
removed by the amine solvent, accumulates in the acid gas scrubbing loop, or is stripped from the amine
3-84
-------
solvent upon regeneration and partitions to the sulfur recovery unit. Some mercury, especially particulate-
phase and oxidized forms, may be removed in the wet particulate scrubber and discharged with
wastewater sludge. More testing is required to assess the level of mercury emissions from IGCC plants. It
is also important to find out the fate of the mercury in IGCC wastes.
By-products of IGCC include production of elemental sulfur (usually in a Claus process) or sulfuric acid,
as well as solid wastes in the form of inert slag, which can be disposed of or sold for a variety of
construction applications.
O&M Impacts
The presence of toxic gases (CO and H2S) between the gasifier and the gas turbine requires some
additional precautions. However, such precautions are common in other industries and can be addressed
by installing CO and H2S sensors and appropriate design of the control system to manage plant start-up
and especially transition from the start-up fuel to coal or coal-based syngas. Plant availability continues to
be an issue especially during early start-up of most IGCC plants, as they experience operating problems.
However, after the commissioning period, most IGCC plants have achieved high reliability (in the 75 -
95 percent range).
Capital Costs
While significant progress has been made in reducing the capital costs of IGCC plants, they remain more
expensive than conventional PC with FGD. The typical range of IGCC capital costs of a 400 MW plant is
$1200 - 1600/kW1 with the lower end of the range corresponding to IGCC plants with a minimum level of
integration and, therefore, lower efficiency and the high end for maximum integration. The capital costs
are projected to decline, as the industry gains more experience with IGCC technology and technological
developments are incorporated (especially in higher efficiency gas turbines and hot gas clean-up
technologies), but the timing and impact of these developments are uncertain.
O&M Costs
For a 400 MW IGCC, fixed O&M costs are projected to be in the $30-45/kW-yr range. Variable O&M
costs depend on whether some revenue can be obtained from by-product sales. Assuming no sales of by-
products, the variable O&M costs are projected to be in the 0.5 - 2.0 mills/kWh range.1
Issues Associated with the Technology; Future Outlook
For the entrained and moving bed IGCC processes, the main barrier to widespread utilization is the high
capital costs relative to competing technologies. However, it should be pointed out that IGCC has much
lower SO2, NOX, and CO2 emissions than the conventional (subcritical) PC with FGD; furthermore,
because they are more efficient, they have lower fuel costs. The fluidized-bed gasification process
requires further development and demonstration. It has yet to demonstrate adequate reliability on a large
scale (above 100 MW). The future of IGCC technology depends on the ability of the industry to reduce its
costs and the environmental requirements associated with SO2, NOX, mercury, and CO2 emissions.7
References
1. Tavoulareas, S. Financing Clean Coal Technologies; World Bank: Washington, DC, July 2001.
2. Technology Assessment of Clean Coal Technologies for China. Electric Power Production;
World Bank ESMAP and AESEG Technical Paper Oil, May 2001.
3. Ratafia-Brown, J. A.; Manfredo, L. M.; Hoffmann, J. W.; Rameza, M. An Environmental
Assessment of IGCC Power Systems. Presented at the 19th Annual Pittsburgh Coal Conference,
Pittsburgh, PA, Sept 2002.
3-85
-------
4. Study on Clean Coal IGCC Technology for China; TA No. 2792 PRC; Asian Development Bank,
Sept 1998.
5. Klett, M. G.; Rutkowski, M. D. The Cost of Mercury Removal in an IGCC Plant; U.S.
Department of Energy, National Energy Technology Laboratory: Pittsburgh, PA, Jan 2001.
6. First H System Gas Turbine Planned for Baglan. Modern Power Systems May 1999, 37-42.
7. Todd, D. IGCC Experience and Technology Improvements Spreading to Other Process/Power
Plants; General Electric, July 1999.
3.2.4.3 Pressurized Fluidized Bed Combustion (PFBC)
PFBC Summary
Status
SO2 Reduction, %
NOX Reduction, %
Hg Reduction, %
CO2 Change, %
Cost
Capital ($/kW)
Fixed O&M ($/kW-yr)
Variable O&M (mills/kWh)
Applicability
Issues
Commercially available up to 350 MW
Up to 98
30 - 70 depending on the coal [90 with SNCR (ammonia or
urea)]
NA
Reduction up to 12
1000 - 1300 (for new plants of 250 MW size)
40-70
2.0-4.0
Mostly new power plants- also applicable to retrofit
applications
Some reliability problems and high costs (relative to
conventional PC and FGD) limit its short-term acceptance.
Also, scaling-up the technology to larger size (above 350
MW) has experience problems which are not clear how they
will be overcome
Note: Emission reduction is based on comparison of this technology to a similar size subcritical
pulverized coal boiler with low NOX burners but without FGD.
Technology Description
Pressurized Fluidized-bed Combustion (PFBC) is similar to AFBC in that it utilizes the fluidized bed
technology, but the PFBC boiler operates under pressure (typically 1.2-1.6 MPa).1'2 Also, it is a combined
cycle plant (see Figure 3-13), as pressurized hot flue gas, after particulate removal, is expanded through a
gas turbine (GT) to drive the combustion air compressor and generate additional electric power.
The main advantages of PFBC are:
• pressurized conditions result in a more compact boiler and therefore offers the potential to lower
capital costs,
• utilization of the combined cycle concept increases the overall plant efficiency and will take
advantage of future technological progress in gas turbine technology, and
• PFBC is able to accomplish sulfur removal at somewhat lower sorbent (Ca-to-S ratio) than AFBC.
3-86
-------
As in the case of AFBC, there are two types of PFBCs, the bubbling and circulating. Also, an "advanced
PFBC" design has been developed with a pyrolyzer added upstream of the PFBC combustor. The fuel gas
generated in the pyrolyzer is burned with the flue gas from the main combustor in a topping cycle to raise
the turbine inlet gas temperature and increase the power output and efficiency of the turbine. The char
from the pyrolyzer is burned in the main combustor. Sorbent is added to both the pyrolyzer and the main
combustor. The successful development of hot gas cleanup (HGCU) technology using ceramic particle
filters and alkali vapor removal is crucial to protecting the gas turbine and the successful demonstration of
advanced PFBC.
Cyclone
Water
Steam
PFBC
Coal&
orbent
To Stack
VVV
Steam
Water
Figure 3-13. PFBC conceptual design.
Commercial Readiness and Industry Experience
The technology is offered commercially up to 350 MW3'4 but there are a number of technical issues still
unresolved (see Issues below). As shown in Table 3-10,1 six commercial-scale bubbling PFBC units (five
plants, one with two units) have been put into service around the world. Most of these boilers are
demonstration units, with financial support from government or international agencies, and all but one are
less than 100 MW. A 360- MW supercritical unit (based on the ABB technology) and a 250- MW
subcritical unit (based on the Hitachi technology) have been constructed in Japan at Karita for Kyushu
Electric Power Company (KyEPCO) and at Osaki for Chugoku Electric, respectively. Both were
commissioned in mid-2000. The operating experience obtained from these units will have a strong
influence on the future of commercial PFBC technology.
3-87
-------
Ishikawajima-Harima Heavy Industries (IHI), which recently obtained a worldwide license from ABB,
Mitsubishi Heavy Industries (MHI), and Hitachi Ltd., are the main suppliers of bubbling PFBC. Five of
the six operating PFBC units listed in Table 3-10 are based on ABB's P 200 PFBC module, designed for
about 80 MW. The sixth unit, an 85- MW module designed by MHI, started up in early 1996.
The Karita PFBC plant, a 360- MW PFBC, replaces an old, conventional coal-fired power plant. It
consists of one novel ABB P 800 module with a GT140P 75- MW gas turbine and a 290- MW steam
turbine. The boiler is designed for supercritical steam conditions of 24.1 MPa/565 °C/593 °C. A wide
range of coal qualities, from lignite to anthracite, will be used at this plant, and the fuel and sulfur sorbent
mixture will be fed as paste. The order for the plant was placed with ABB Carbon's licensee, IHI, which
has undertaken the engineering, manufacturing, erection, and commissioning of the plant. The GT140P
turbine was supplied by ABB STAL and the steam turbine was supplied by Toshiba.
Table 3-10. PFBC Plants and Operating Experience
Technology
Facility Name
Gross Output, MW
Units
Coal Type
Coal Sulfur, %
Coal Ash, %
Coal Feed
Sorbent Feed
Sorbent
NOX Control
Cyclones
Hot Filter
Steam Pressure, bar
Steam Temperature,
°C
ABB
Escatron,
Spain
79
1
Lignite
7
36
Dry
Dry
Limestone
None
9x2
1/9
Slipstream
90
510
ABB
Vartan, Sweden
135 plus
224 MW
2
Polish
Bituminous
0.65
15
Paste
Paste
Dolomite or
Limestone
SNCRandSCR
7x2
None
130
530
ABB
Tidd, USA
73
1
Ohio
Bituminous
4
10
Paste
Dry or Paste
Dolomite
None
7x2
1/7 Slipstream
90
496
ABB
Wakamatsu,
Japan
71
1
Australian
0.4
10
Paste
Paste
Limestone
SCR
7x1
Full Gas Flow
(part time)
102
593/593
MHI
Tomatoh-Atsuma,
Japan
85
1
Various
0.9
Not Available
Dry
Dry
Limestone
SCR
2x1
Full Gas Flow
166
566/538
Lurgi Lentjes Babcock (LLB, the Lurgi-Deutsche Babcock partnership) and Foster Wheeler (now
incorporating Ahlstrom Pyropower) have been developing circulating PFBC. To date, development has
been mainly on a pilot plant scale. However, a 137 MW circulating PFBC plant is under construction at
the Mclntosh plant of the City of Lakeland, FL.
PFBC technology is applicable for both retrofit (replacement of an existing boiler and utilization of the
remaining equipment, especially the steam turbine and balance of plant) and new applications. However,
most PFBC projects in the future are expected to be new applications.
3-S
-------
Emission Control Performance
SO2 removal up to 98 percent is possible, especially with high-sulfur coals. SO2 removal is affected by the
bed temperature and by the Ca-to-S molar ratio. A Ca-to-S molar ratio of approximately 1.5 is typical for
an SO2 removal of 90 - 95 percent. The excess sorbent is not calcined. While this increases the volume of
solid wastes compared to an AFBC with the same Ca-to-S molar ratio, PFBC requires a lower Ca-to-S
molar ratio, which counterbalances this disadvantage and results in about the same waste volume
compared to an AFBC. Furthermore, the absence of lime makes the solid wastes easier to handle.
NOX emissions are 30 - 70 percent lower than conventional PCs with low NOX burners. Further NOX
reduction can be achieved by installing an SNCR or SCR system. NOX emissions in the 100 ppm (0.13
Ibs/MMBtu) level without SNCR have been demonstrated. NOX emissions are impacted by the bed
temperature, the nitrogen and volatile matter in the coal, and the bed stoichiometry, which is affected by
the excess air and air distribution across the bed.
It is not clear whether PFBC controls mercury, too.
Typical PFBC design plant efficiency ranges from 38 to 46 percent (FIFfV basis) depending on the coal
type, coal feed method, and steam cycle conditions. Compared to new subcritical pulverized coal plants,
which typically have efficiency in the range of 34 to 38 percent, PFBC can potentially result (as these
efficiencies indicate) in up to 10 - 12 percent lower CO2 emissions.
O&MImpacts
Reliability problems have been experienced by the most recent demonstration plants, but eventually have
been resolved. For example, Vartan and Wakamatsu reached the 80-90 percent reliability level, but others
had lower reliability. The impact of small particulates and alkali on the gas turbine was one of the issues.
ABB is offering a "ruggedized" gas turbine design, which is expected to improve reliability. The
properties of the ash are critical in determining the potential impacts on the various PFBC plant
components.
Capital Costs
Typical costs of PFBC plants are expected to be in the $1000 - 1300/kW range.1
O&M Costs
O&M costs are projected to be: $40 - 70/kW-yr fixed O&M cost and 2.0 - 4.0 mills/kWh variable O&M
cost1 depending on the O&M practices of the utility, labor costs, and cost of consumables (especially
sorbent). The high estimates are more typical of the 80 MW PFBC, while the lower estimates are for the
350 MW design.
Issues Associated with the Technology; Future Outlook
The PFBC plants in operation experienced a number of problems but, for the most part, they have been
satisfactorily addressed. However, some uncertainty still remains in the following areas:
• Coal feed. Wet feed systems need proper size distribution and a better indicator of proper
consistency; dry feed systems need designs that address erosion in transport pipes at high pressure
and redundant systems for high availability.
• Gas turbine lifetime. High-cycle fatigue damage and erosion are on-going problems being addressed
by the suppliers.
• Gas filter performance. Development continues on thermal shock-resistant ceramic candle filters.
Development of this component is critical for achieving higher plant efficiencies and the wider
acceptance of the technology.
3-89
-------
• Coal and sorbent distribution in the bed. Fuel and sorbent distribution need to be optimized to achieve
maximum sorbent utilization and uniform bed temperatures.
• Cyclone liner life. Proper application of select materials or unlined austenitic stainless steel cyclones
may resolve this issue.
In conclusion, PFBC technology has been demonstrated and is commercially available up to a 350 MW
plant size, but higher than conventional technology costs limit its widespread utilization, at least in the
short term.
References
1. Tavoulareas, S. Financing Clean Coal Technologies; World Bank: Washington, DC, July 2001.
2. Technology Assessment of Clean Coal Technologies for China. Electric Power Production;
World Bank ESMAP and AESEG Technical Paper Oil, May 2001.
3. Jansson, S.; Anderson, J. Progress of ABB's PFBC Projects. Presented at the 15th FBC
Conference, Savannah, GA, May 1999.
4. Anderson, J.; Anderson, L. Technical and Commercial Trends in PFBC. Presented at the Power-
Gen Europe, Frankfurt, Germany, June 1999.
3.2.4.4 Supercritical Pulverized Coal Plant
Supercritical Pulverized Coal Plant Summary
Status
SO2 Reduction, %
NOX Reduction, %
Hg Reduction, %
CO2 Change, %
Cost
Capital ($/kW)
Fixed O&M ($/kW-yr)
Variable O&M (mills/kWh)
Applicability
Issues
Supercritical: commercial; ultra- supercritical: needs
demonstration
4-12
4-12
4-12
4-12 reduction
Supercritical: 825-1080; Ultra-supercritical: 1000-1150
(for 700 - 1000 MW plant size)
Supercritical: 25 - 32; Ultra-supercritical: 30 - 35
3.0-5.0
New power plants
Ultra-supercritical requires further demonstration
Note: Emission reduction is based on comparison of this technology to a similar size subcritical
pulverized coal boiler with low NOx burners, but without FGD.
Technology Description
Supercritical pulverized coal technology is similar to subcritical in terms of the conceptual plant design,
but may operate at higher steam temperatures and higher pressures.1 Subcritical plants operate below
approximately 18 MPa (2600 psi) maximum steam pressure, while supercritical plants are designed to
operate from 23 to 35 MPa (3200 to 5000 psi). Also, supercritical plants can be designed to operate at
3-90
-------
steam temperatures (superheat and reheat) up to 650 °C (1200 °F) compared to 538 °C (1000 °F), which
is the typical temperature in subcritical plants.
Typical pulverized plant designs used presently and their steam parameters are given below:
1. Subcritical: 16.7 MPa/538 °C/538 °C (2400 psi/1000 °F/1000 °F);
2. Supercritical: 24.2 MPa/538 °C/565 °C (3500 psi/1000 °F/1050 °F), which is about 4.0 percent
more efficient than the subcritical; and
3. Ultra-supercritical with double reheat: 31 MPa/600 °C/600 °C/600 °C (4500 psi/1100 °F/1100
°F/1100 °F), which is about 8 percent more efficient than the subcritical design; or 35 MPa/650
°C/650 °C/650 °C (3500 psi/1200 °F/1200 °F), which would produce an efficiency gain of about
11 percent relative to the subcritical unit. Both these options are under development.
The high steam pressures and temperatures require higher grade materials for the furnace water walls,
superheat and reheat sections of the boiler, headers, steam piping between boiler and turbine, and first
stages of the superheat and reheat sections of the turbine. Availability of materials such as l%Cr-!/2Mo,
2%Cr-lMo, and 9-12 percent Cr class steels (T91 and T23) are critical for designing and manufacturing
supercritical plants. In addition to materials, water chemistry (corrosion protection) and power plant
controls are some of the most important aspects of supercritical plant design and operation. However,
appropriate design and O&M practices have proven adequate to ensure reliable operation.
Commercial Readiness and Industry Experience
Supercritical plants are commercially available in many countries including China, Denmark, Germany,
Italy, Japan, Russia, South Korea, and the United States.1'2'3 The ultra-supercritical unit is in the
demonstration stage. Supercritical pulverized coal technology is commercial with approximately 462
units operating worldwide (see Table 3-11). There are at least 12 boiler manufacturers and 11 steam
turbine suppliers that offer supercritical plant components.
Table 3-11. Power Plants with Supercritical Design Parameters4
Country or Region
Japan
USA
West European Countries
East European Countries
Other Countries
Total
Number of Units
108
149
53
123
29
462
Total Units (%)
23.4
32.3
11.5
26.5
6.3
100.0
Capacity, MW
67,900
106,454
29,310
51,810
13,520
268,994
Total Capacity (%)
25.2
39.6
10.9
19.3
5.0
100.0
The U.S. has the largest number of supercritical plants in operation, but all these plants were installed in
the 1960s and 1970s. In recent years, no supercritical plants have been built in the U.S. This is mainly
attributable to the perception that supercritical plants have lower reliability and the relatively low costs of
fuels in the U.S., which make low-cost but also low efficiency options (in this case subcritical PC)
economically more attractive. Indeed, there were some reliability problems experienced by the first
supercritical plants built in the U.S., but, as Table 3-12 shows, the reliability (expressed as equivalent
availability) of these plants improved significantly to the same level of subcritical plants. Similar
experience is documented by VGB in Germany.4
3-91
-------
Table 3-12. Equivalent Availability for Subcritical and Supercritical Power Plants4
U.S. Supercritical Plants
Size Range ( MW)
Subcritical
Supercritical
300-399
76.5
64.4
400-499
77.4
74.6
500-599
76.3
73.8
600-799
78.5
74.2
>800
77.2
75.6
Germany's Supercritical Plants
Year
Subcritical
Supercritical
1988
84.2
80.2
1989
82.5
74.9
1990
84.1
84.2
1991
84.9
85.2
1992
84.5
87.1
1993
82.0
89.8
1994
83.8
83.0
1995
83.7
84.7
1996
86.6
79.5
1997
88.5
90.3
Japan is one of the countries leading the development of supercritical technology. Until the early 1990s,
most Japanese plants had steam conditions of 3500 psi/1000 °F/1050 °F (24.6 MPa/538 °C/566 °C), but
starting in 1993 the steam temperatures of new plants are in the ultra-supercritical range, approaching
1100°F(600°C).
Also, there are about 53 supercritical units in Europe, especially in Germany, Italy (mostly oil fired), and
Denmark. The most recent European units (most of them coal-fired with one oil-biomass unit) utilize
ultra-supercritical steam conditions.
Korea has built sixteen 500 MW supercritical units utilizing a standardized design for bituminous coal.5
Also, China has ten supercritical units in operation and 10 more in the planning stage.
Emission Control Performance
This technology is not an environmental control technology per se, but due to its higher efficiency it
results in lower emissions prior to utilizing environmental controls such as SCR and FGD. The
supercritical design improves plant efficiency by 4.0 percent relative to a similar Subcritical plant.
If the ultra-supercritical design is used with double reheat, the efficiency increases by about 8 percent
relative to a similar Subcritical plant. Further potential exists (projected to increase efficiency by 10 - 12
percent) utilizing higher steam conditions, such as 35 MPa/650 °C/650 °C/650 °C (3500 psi/1200 °F/1200
°F), but this technology remains to be demonstrated.
O&M Impacts
There are no significant O&M impacts relative to Subcritical PC technology. Supercritical PC requires
more attention to water chemistry (requirements are tighter than for Subcritical plants) and operation,
especially during start-up and load following. However, power plant controls can be designed to include
all the precautions and safeguards needed for safe and reliable operation.
Capital Costs
Many recent studies1'6 have estimated the costs of supercritical vs Subcritical PC technologies. The
general consensus is that the capital costs of a supercritical plant are equal to or up to 8 percent higher
than a similar size Subcritical plant. So, considering that the latter is projected to cost $800 - 1000/kW, the
3-92
-------
supercritical unit is projected to cost $825 - 1080/kW and the ultra- supercritical unit $1000 - 1150/kW
(in all cases, the PC is equipped with FGD, but not SCR).1'6
Is should be noted also that supercritical plants are usually competitive in large sizes, above 500 MW.
However, recently power plant suppliers have been developing smaller supercritical plants (in the 350 -
500 MW range) to address cases with high cost fuels and limited demand growth.1
O&MCosts
Fixed O&M costs range from $25 to 32/kW-yr for supercritical units and $30 to 35/kW-yr for ultra-
supercritical units. Variable O&M costs range from 3 to 5 mills/kWh.1
Issues Associated with the Technology; Future Outlook
The supercritical PC technology is commercially available, and there are no issues associated with it. As
more emphasis is placed on CO2 emission reduction, it provides a proven technology (therefore low-risk
option) to achieve significant plant efficiency improvement relative to conventional (subcritical)
pulverized coal plants. An ultra-supercritical unit requires further demonstration, mainly to prove the
reliability of new alloy materials.
References
1. Tavoulareas, S. Financing Clean Coal Technologies; World Bank: Washington, DC, July 2001.
2. Torrens, I. M.; Stenzel, W. C. Increasing the Efficiency of Coal-fired Power Generation/State of
the Technology: Reality and Perceptions; International Energy Agency/CIAB: London, England,
March 1997.
3. Slettehaugh, R. A.; Dittus, M. H.; Voss, R. A. Lignite Power Generation in North Dakota- Cost,
Performance, and Emissions of Viable Technologies. Presented at the PowerGen Conference,
Orlando, FL, Nov 14-16, 2000.
4. Supercritical Power Plants/Evaluation of Design Parameters (March 17, 1999). http://www.
worldbank.org/html/fpd/em/supercritical/ppt supercritical (accessed Dec 1, 2004).
5. Yoo, K. J.; Kim, M. H.; Moskal, J. J. Standardization in the Design and Construction of 500 MW
Coal-fired Power Plants in the Republic of Korea. Presented at the Powergen Americas, Orlando,
FL, Dec 1996.
6. Sormani, G. F.; Monti, M. A.; Maffei, E. Comparison Between Cost of Electricity Generated
from Coal by Ultra-supercritical Steam Plants and IGCC. Presented at the Powergen Americas,
Orlando, FL, Dec 1996.
3-93
-------
5.2.5 Power Plant Upgrading and Operating Options
3.2.5.1 Fuel Blending and Cofiring
Fuel Blending and Cofiring Summary
Status
SO2 Reduction
NOX Reduction
Hg Reduction
Cost
Capital ($/kW)
Fixed O&M ($/kW-yr)
Variable O&M (mills/kWh)
Applicability
Issues
Commercial except for fuel blending requiring on-line
continuous control of emissions
Proportional to sulfur content of fuels; up to 100% reduction
for switching from high- to low-sulfur coal or gas
Up to 20 - 30% due to switching from bituminous to
subbituminous coal or biomass cofiring. 50 - 80% due to
gas cofiring and reburn or switching from coal to gas
Proportional to mercury content of fuels, as well as other
fuel properties
Coal blending: 20 - 100; Fuel switching: Up to 200 - 300;
Cofiring: 30-200
Project-specific
Project-specific
Both new and retrofit applications
Development of on-line coal quality monitoring is needed to
enhance coal blending
Technology Description
The ability to alter the fuel characteristics has a significant impact on both the thermal performance of the
unit and the resulting emissions of SO2, NOX, Hg, and CO2. Table 3-14 lists the key fuel properties, which
affect emissions. Therefore, altering the characteristics of the fuel by blending or cofiring could be
viewed as a multi-emission control option.
While the feasibility and attractiveness of changing the fuel characteristics are very site-specific (affected
significantly by plant design and fuel properties), in general the following options are available while
maintaining the basic configuration and design of the plant:
• Blending with a fuel of the same type with different properties (blending coal of the same
classification such as two eastern bituminous coals) or different classifications (eastern bituminous
coal with subbituminous coal),
• Switching to a fuel of the same type; for example, switching from eastern bituminous to
subbituminous coal,
• Switching to a fuel of different type: for example, switching from coal to natural gas, and
• Cofiring coal with fuels such as gas, biomass, and municipal solid wastes.
Blending and cofiring generally constitute operating changes (even though originally some hardware
modifications may be necessary or required to mitigate O&M impacts), while switching to another fuel is
a permanent change (at least until the decision is reversed) and includes hardware modifications.
Blending coals is very common in the utility industry worldwide with the primary focus being control of
sulfur content.1"5 The plant operator either mixes the coals in the coal yard or loads the coal silos partly
with the one coal and then with the other with mixing occurring inside the silo. In some cases, blending
3-94
-------
has been accomplished by mixing coals from two different conveyors on to a common one. The ability to
monitor and control coal quality (beyond just sulfur) is very limited. Monitoring is based on periodic
sampling of the coal and rough estimation of the average quality characteristics (mainly sulfur content), as
well as SO2 measurements using continuous emission monitors. Blending could be used to affect
emissions other than SO2 (NOX and mercury). However, the exact coal quality resulting from such
blending cannot be predicted; only an approximation of the average is possible with present technologies.
More detailed fine-tuning of the coal analysis through blending needs on-line coal quality monitoring
technologies, which are still under development.
Switching to another coal (from bituminous to subbituminous) is a common practice in the industry
(again, mainly for sulfur control) and usually involves hardware changes. Such changes are very site-
specific, but common areas, which are impacted by coal switching, are:
• Upgrading the coal crushing equipment and the pulverizers to avoid potential de-rating of the unit due
to limited coal throughput into boilers,
• Modifying the heating surface of the boiler (superheater and reheat sections) to maintain the steam
outlet temperatures over the load range without exceeding operating limitations set by materials
(maximum metal temperatures or gas velocities to avoid excessive erosion),
• Replacing or modifying the soot blowing system to accommodate higher slagging or fouling,
• Incorporation of a flue gas conditioning system to modify the resistivity of the sub-bituminous, low-
sulfur coal ash and make it suitable for collection in an existing precipitator originally designed for
high-sulfur coal ash,
• Modification of the ash handling system to handle quick-setting characteristics of sub-bituminous
coal ash, and
• Modification of the coal storage and conveying systems to meet the special handling characteristics of
sub-bituminous coal.
On the positive side, coal switching may provide an opportunity to improve performance (plant
efficiency) or at least minimize the adverse impacts on efficiency. For example, modifications in the
economizer, air heater, and feed water heaters may increase plant efficiency or mitigate boiler efficiency
reduction due to higher moisture of the new coal.
Switching from coal to a different fuel type such as natural gas is also possible. Hardware modifications
may be required depending on the design of the existing plant. The types of changes required are as
follows:
• The plant needs to be connected to a natural gas pipeline, in case gas is not available on site,
• The burners need to be modified to accommodate natural gas,
• The boiler heating surfaces may need to be modified to maintain proper steam conditions, and
• Boiler backend modifications may be required, depending on whether the plant elects to dismantle the
existing pollution controls, bypass these controls and leave them in place, or continue to pass flue
gases through the non-operating controls.
Cofiring involves utilization of more than one fuel, which represents a relatively small percentage
(usually 5-15 percent) of the total heat input.6'7 Typically, fuels which have been cofired with coal are:
natural gas, biomass, and municipal solid wastes (MSW).8'9 Depending mainly on the physical properties
of the secondary fuel, it may be introduced in the boiler through a separate burner or injection port or
mixed with the coal. Solid fuels may be pulverized together with the coal, but this limits the amount of
3-95
-------
cofired fuel to about 5 percent. In most case, separate injection ports are added. Cyclones and fluidized
bed boilers are better cofiring candidates than pulverized coal boilers, even though the latter have
demonstrated adequate performance and reliability. Cofiring natural gas requires separate burners or ports
within the existing coal burners.10
Commercial Readiness and Industry Experience
Fuel switching and cofiring are commercial options that have been used extensively worldwide both for
new plants and retrofit applications. Coal blending is also a common practice. In cases, where the average
coal quality needs to be adjusted to an approximate level or below a set level, coal blending is a well-
demonstrated option. However, if on-line continuous monitoring and control of coal quality is needed,
presently there is no available technology except monitoring of sulfur content. On-line coal quality
monitoring technologies are under demonstration and have not achieved the level of reliability needed for
widespread application.
Emission Control Performance
Blending of coals or switching to another coal clearly impacts all emissions, even though the impact at
each plant is very site-specific and depends on the characteristics of the coals, the design of the power
plant, and the operating conditions. Switching from a high-sulfur (4-5 percent sulfur content) coal to a
low-sulfur (0.5 - 0.8 percent sulfur content) coal reduces the SO2 emissions up to 80 - 90 percent. The
impact of coal blending on SO2 would be lower than this level and proportional to the percent of the low-
sulfur coal being used. Switching from bituminous to subbituminous coal or blending also reduces NOX
emissions by up to 20-50 percent as demonstrated in a number of well-documented utility sites such as
Arapahoe 4 of Public Service of Colorado,5 Genoa 3 of Dairyland, and Gibson 3 of PSI Energy.2 Coal
switching and blending is expected to have an impact on mercury emissions as each coal contains a
different level of mercury. Also, changes in chlorine content resulting from coal blending may affect the
speciation of mercury. Additonally, the ash composition and unburned carbon content may play a
catalytic role affecting mercury speciation. Presently, no adequate data exist to predict such impact.
Finally, coal switching and blending are expected to affect plant efficiency and CO2 emissions mainly
because of different moisture content of the coals, but also because of potentially different auxiliary
power requirements. For illustration purposes, if the moisture content of the coal changes (as a result of
switching or blending) from 8 to 20 percent, the impact is an approximately 1.2 percent lower boiler
efficiency, which corresponds to approximately a 3 percent or higher increase of CO2.
Switching from coal to another fuel has more drastic consequences to the performance and emissions. The
most common fuel switch is from coal to natural gas provided that the economics are favorable. In this
case, SO2 and mercury emissions may be eliminated. NOX emissions could be reduced by up to 70 - 80
percent. Finally, CO2 is reduced by approximately 30 percent as a result of switching from coal to natural
gas fuel. Boiler efficiency for natural gas is usually 2-3 percentage points lower than bituminous coal.
Due to the high percentage of hydrogen in the natural gas, the CO2 emission per fuel input is 200
Ib/MMBtu for coal and 130 Ib/MMBtu for natural gas based strictly on fuel carbon conversion.
The impact of cofiring on emissions is very site specific and depends on the characteristics of the fuels
being cofired, the plant design, and the design of the cofiring system. Cofiring of gas in a coal-fired plant
reduces SO2 and mercury proportionally to its input (typically up to 30 percent), NOX up to 50 - 70
percent, and CO2 by up to 7 percent (assuming maximum heat input due to natural gas of 20 percent).11"14
The characteristics of biomass and MSW vary significantly, but in general they are expected to reduce
SO2 and CO2 proportionally to their inputs, and NOX by up to 20 percent. It should be noted that these
fuels are considered renewable energy sources, so any substitution of coal results in a proportional
reduction of CO2 emissions. However, some of the fuels contain a significant amount of moisture (wood
wastes and wet MSW may contain 40 - 60 percent moisture), which reduces the plant efficiency and may
counterbalance the benefit of using a renewable energy resource.
3-96
-------
O&M Impacts
Changes in the fuel characteristics may have significant impacts on unit performance and reliability.
Table 3-14 summarizes the main impacts of unfavorable changes in the values of the key fuel
characteristics, as well as potential mitigation options.
In general, load reduction (de-rating) may be necessary to avoid high steam and metal temperatures or to
accommodate lower steam temperatures at the steam turbine inlet. Also, load may be limited by the
capacity of the existing fuel handling system pulverizers, or other boiler-related systems. In most cases,
these impacts can be mitigated by modifying the boiler and other effected equipment. Other potential
impacts include:
• Changes in the unburned carbon. This can be mitigated by design enhancements of the pulverizers
(addition of dynamic classifiers) and the firing system (new burners).
• Increase or reduction of furnace slagging and fouling. In most cases, this can be mitigated through
increased soot blowing, but in some cases de-rating may be needed.
• Depending on the fuel characteristics, pulverizer wear and reliability, as well as corrosion and erosion
of boiler heating surfaces, may be affected. The impact of blending coals of different hardness on the
final size distribution and compositional distribution of the pulverized coal is still not fully
understood. Blending hard and soft coals may cause difficulties in pulverizing the blend, as the
grinding requirements may not always be proportional to the mix of the two coals. In most cases, the
characteristics of the poor coal dominate the blend.
• Very often fuel changes impact the performance of the ESP and may result in higher particulate
emissions and opacity. If the ESP size is conservative, it may not require any modifications.
However, many older plants represent tight designs and may need to be modified to meet particulate
emissions and opacity requirements.
• Reliability and operating flexibility may be impacted, usually adversely because biomass and MSW
are difficult fuels to handle. Also, some coals may be harder to grind (pulverize) or more corrosive.
Of course, natural gas is easier to handle and would be expected to improve plant reliability and
operating flexibility.
• Other impacts requiring consideration on a case-by-case basis include potential changes in the
required excess air (which impacts plant efficiency), CO emissions, and unit turndown rate (speed at
which the operator can change unit output).
3-97
-------
Table 3-14. Main O&M Impacts and Potential Mitigations
Coal Property
Coal Type
% Sulfur
% Ash
Ash fusion
temperature
% Nitrogen
Volatile Matter
% Moisture
% Mercury
% CaO or MgO
Grindability
Impact on Emissions
SO2
X
X
X
X
NOx
X
X
X
X
X
Hg
X
X
X
CO2
X
X
X
X
Potential Plant Equipment
and O&M Impacts
Potential de-rating, lower
reliability, and lower
efficiency
ESP performance and FGD
system performance
ESP performance, ash
handling system impacts,
and erosion of mills and
boiler surfaces
Furnace slagging and
backpass fouling
Potential impact on boiler
heating surface performance
Mill throughput, boiler
efficiency, primary air
system performance, and
increased combustion air
and flue gas volumes
Mill throughput
Potential Mitigation
De-rating may be avoided through
hardware modifications, especially
in the mills and boiler heating
surfaces
ESP upgrading or SO3 flue gas
conditioning and upgrading of the
FGD system or increased reagent
use
ESP upgrading, upgrading or more
frequent operation of the ash
handling system and more active
maintenance of mills and boiler
surfaces
Sootblowing and fuel additives
Design or operation of NOx controls
for the increased baseline
emissions
Design or operation of NOx controls
for the increased baseline
emissions and boiler heating
surface modification
Mill and primary air system
upgrading, design or operation of
pollution controls for the increased
baseline emissions, and operation
of draft fans at higher capacities
Depends on whether mercury
controls would be present
Increased reagent use, if SO2
controls are present
Mill upgrading
Costs
The costs, both capital and O&M, are very site specific. The following estimates are for illustration
purposes only.
Coal blending is the least expensive option with regard to costs, which may range from $20 to 100/kW
depending on the amount of cofired coal, its characteristics, and the design of the existing power plant.
Switching to another fuel is usually more expensive and may reach up to $200 - 300/kW. Cofiring capital
costs may range from $30 to 200/kW. More specific examples6 (End of Year 1991 U.S.$; however,
capital costs of power plant equipment have been reduced significantly since then and it is expected that
the same costs reflect FY2000 conditions):
• Wood cofiring: $ 104/kW,
• Refuse Derived Fuel cofiring: $128/kW, and
• Tire-derived fuel cofiring: $37/kW.
3-98
-------
Issues Associated with the Technology; Future Outlook
Fuel blending, switching, and cofiring are common industry practices and are expected to continue. In the
past, the main driving force has been economics, but they have a role to play in emission control too, as
emissions become commodities and emission control becomes an operating cost. The key technological
barrier, especially with regard to coal blending, is the absence of reliable and commercially available on-
line continuous coal quality monitors. However, if only a not-very-accurate control of emissions is
required, all options are practical and commercially available.
Depending on site-specific constraints, not all options may be available to all power plants. Such
constraints may include: lack of natural gas in the proximity of the plant, very high transportation and
processing costs for biomass and waste fuels, and constraining power plant design with tight fuel
specification requirements (cyclone and wet bottom boilers).
References
1. In Effects of Coal Quality on Power Plants; Proceedings of 5th International Conference, Kansas
City, MO, May 20-22, 1997; EPRI TR-109340, Nov 1997.
2. Guidelines for Evaluating the Impact of Powder River Basin (PRB) Coal Blends on Power Plant
Performance and Emissions;EPRl TR-106340; EPRI: Palo Alto, CA, March 1996.
3. Kambara, S.; Takarada, T.; Yamamoto, Y.; Kato, K. Relation Between Functional Forms of Coal
Nitrogen and Formation of NOX Precursors During Rapid Pyrolysis. Energy Fuels 1993, 7 (6),
1013-1020.
4. Technology Assessment for Blending Western and Eastern Coals for SO2 Compliance; EPRI
TR-105748; EPRI: Palo Alto, CA, May 1996.
5. Smith, R. A. Arapahoe 4/Powder River Basin Coal Test Burn; U.S. DOE report: Washington,
DC, April 1996.
6. Strategic Analysis of Biomass and Waste Fuels for Electric Power Generation; EPRI TR-102773;
EPRI: Palo Alto, CA, Dec 1993.
7. Costello, R. Biomass Cofiring Offers Cleaner Future for Coal Plants. Power Eng. 1999, 45.
8. Kouvo, P.; Korelin, T.; Savola, T. Mercury Emissions and Removal During Cofiring of Coal,
Wood and Wastes. Proceedings of the U.S. EPA-DOE-EPRI Combined Power Plant Air Pollutant
Control Symposium: The MEGA Symposium and The A&WMA Specialty Conference on
Mercury Emissions: Fate, Effects, and Control, Chicago, IL, Aug 20-23, 2001.
9. Davidson, R. Experience of Cofiring Waste with Coal, International Energy Agency/Coal
Research Center: London, England, Nov 1998.
10. Pratapas, J. M.; Stenzil, W.; Mazurek, J. The Natural Gas Cofiring/Reburning Module for
Pulverized Coal Fired Plant Evaluations and Planning. Presented at the American Power
Conference, Chicago, IL, April 9-11, 1996.
11. Morrison, G. Cofiring of Coal and Waste - an International Perspective. Presented at the
Powergen Conference, Orlando, FL, Dec 1996.
3-99
-------
12. Guidelines for Cofiring Refuse-derived Fuel in Electric Utility Boilers; EPRI RP-1861; EPRI,
March 1998.
13. Hecklinger, R. S.; Rehm, F. R. Coal Refuse-derived Fuel Cofiring Project, Milwaukee County,
Wisconsin; EPA/600/2-85/136 (NTIS PB86-135381); U.S. EPA, Hazardous Waste Engineering
Research Laboratory: Cincinnati, OH, Nov 1985.
14. Proceedings: Municipal Solid Waste as a Utility Fuel; EPRI CS-4900-SR: Palo Alto, CA, Nov
1986.
3.2.5.2 Plant Efficiency Improvements (Upgrading, Newer Configurations, Technologies)
Plant Efficiency Improvements Summary
Status
SO2 Reduction
NOX Reduction
Hg Reduction
CO2 Change
Cost
Capital ($/kW)
Fixed O&M ($/kW-yr)
Variable O&M (mills/kWh)
Applicability
Issues
Commercial
Heat rate improvement and upgrading: SO2 reduction
proportional to heat rate improvement; Repowering with
gas: close to 100% SC>2 reduction
Heat rate improvement and upgrading: NOx reduction
proportional to heat rate improvement; Repowering with gas:
up to 70 - 80% NOX reduction
Heat rate improvement and upgrading: Hg reduction
proportional to heat rate improvement; Repowering with
natural gas: 100% Hg reduction
Heat rate improvement: Up to 1.5% reduction at full load (up
to 5% at low loads); Upgrading: up to 8% reduction;
Repowering with natural gas: up to 50% reduction
Heat rate improvement: <20 - 40; Upgrading: 20 - 50;
Repowering with natural gas: 100 - 400
Project-specific
Project-specific
Retrofit applications
Impact of environmental regulations pertaining to increased
output of existing plants
Technology Description
Upgrading and efficiency improvement in a power plant are geared towards increasing the plant output or
achieving higher efficiency. In the latter case, the same output can be produced utilizing less fuel and, in
most cases, proportionally less emissions. Most options affecting plant output and efficiency are
interrelated, but for simplicity they will be described separately in this document, grouped into three
categories: (1) efficiency improvement, (2) plant upgrading and (3) power plant repowering. The third
option involves significant changes in the plant design, usually by adding a gas turbine and converting the
plant to a combined cycle.
Efficiency Improvements
Efficiency improvements can be achieved by both operating changes and minor hardware modifications,1
such as:
3-100
-------
• adjustment and tuning of the pulverizer to improve coal fineness for better combustion efficiency,
resulting in lower CO and unburned carbon,
• reduction in excess air; this is accomplished by closer monitoring of excess air (usually by installing a
CO monitor in the stack), combustion system tuning (adjustments of air registers and coal-air flow
balancing), and pulverizer tuning,
• reduction in air leakage into the boiler, air heater, and pollution control equipment,
• reduction in water and steam leaks in valves and piping of the balance of plant, and
• cleaning of heating surfaces such as the boiler water walls, superheat and reheat sections, as well as
the condenser and the feed water heaters to improve heat transfer.
Power Plant Upgrading
Usually upgrading is geared towards increased plant output,2'3 or increased plant remaining life, but
invariably it improves plant efficiency, too. Examples of measures included in this category are:
• pulverizer upgrades including a dynamic classifier and increased capacity of the exhaust fans,
• replacement of the economizer or the air heater, resulting in lower stack temperature and better plant
efficiency; this is usually a desirable option when the plant switches from high-sulfur to low-sulfur
coal (either by coal switching or blending) and can afford to lower the stack temperature without
concern about corrosion in the backend equipment due to condensation of sulfuric acid,
• replacement of steam turbine blades with new blades (usually improved three-dimensional profile)
taking advantage of technological improvements in blade design and materials; this applies to all
sections of the steam turbine (high, intermediate, and low pressure), but especially in the high
pressure section where the economics are more favorable and more technological improvements have
been made,4"6
• more efficient modifications of feed water heaters and addition of low-pressure preheating,7
• refurbishment of condenser with new and more efficient tubes (with regard to heat transfer), heating
surface cleaning systems, and design modifications to reduce air in-leakage,
• installation of variable-speed drives on the main rotating equipment (forced- and induced-draft fans,
boiler-feed pumps, and circulating water pumps) to reduce the plant auxiliary power consumption,
and
• replacement of outdated controls with a state-of-the-art digital control system to improve overall plant
operation and efficiency.
Power Plant Repowering
The general principle followed in repowering an existing coal-fired power plant is to add a gas turbine
and associated equipment and convert it to a combined cycle plant burning natural gas. There are a
number of alternative approaches to repowering (full repowering, topping or windbox repowering, and
parallel repowering),8"12 but they differ only by the degree to which the existing equipment is used.
Topping or windbox repowering involves use of the flue gas from a new gas turbine as hot combustion air
into the burners of the existing boiler and may increase plant output up to 40 percent. Parallel repowering8
utilizes the existing boiler as a heat recovery device for the flue gas of the gas turbine, but because the
resulting outlet steam temperatures are lower, a new heat recovery steam generator is needed to ensure
that the plant output is maintained. Full repowering utilizes mainly the existing steam turbine and balance
of plant along with a new gas turbine and associated equipment to form a combined cycle. Power output
of the plant may increase up to 200 percent as a result of full repowering.
3-101
-------
Commercial Readiness and Industry Experience
All these options are commercially available, and the industry (both the utilities and the boiler vendors)
has significant experience with such projects.
Emission Control Performance
In the case of heat rate improvement and upgrading, reduction of SO2, NOX, mercury, and CO2 is
proportional to the heat rate improvement achieved. There are some exceptions where the change may not
be exactly proportional, but it is not significantly different. An example of such cases is the reduction of
excess air, which benefits plant efficiency, but may benefit NOX emissions more than indicated by the
proportional change.
Heat rate improvements may result in up to 1.5 percent heat rate improvement at full load and up to 5
percent at low loads. Upgrading projects may achieve higher efficiency than heat rate improvements, but
they are very site-specific.
In the case of repowering, there is a change in fuel from coal to natural gas, which results in significant
changes in the emissions. In general, topping or windbox repowering improves plant heat rate by 5 - 10
percent. Full repowering may increase the plant efficiency of an old coal-fired power plant (in most cases:
28 - 34 percent efficient on an HHV-basis) to 45 - 50 percent. CO2 reduction is achieved due to better
heat rate if less fuel is burned and the fact that natural gas releases less CO2 per unit heat input than coal.
The end result from a repowering project may reach up to 50 percent CO2 reduction, with parallel
elimination of SO2 and mercury.
O&M Impacts
Upgrading and repowering projects do not have significant adverse O&M impacts. In fact, the changes in
plant design and operation are usually beneficial to both performance and reliability of the plant. There
are cases, however, with heat rate improvements where a trade-off between emissions and heat rate may
be involved. For example, an attempt to reduce NOX emissions as much as possible through aggressive
combustion staging (high overfire air flow rate) and low excess air, may affect heat rate adversely due to
lower steam temperatures, high CO emissions, and high unburned carbon.
Costs
Costs of heat rate improvements are typically low, below $20-40/kW. Upgrading and repowering costs
are very site-specific, but they are typically in the following ranges:
• $20 - 50/kW for upgrading and
• $100-400/kW for repowering.12
Issues Associated with the Technology; Future Outlook
Repowering applications depend greatly on the availability of a reliable source of gas supply at a
candidate plant location and cost differential between coal and gas. Considering that gas prices continue
to be significantly above those of coal for most locations in the U.S., the attractiveness of repowering
options is limited. Another reason is the significant reduction of prices of new natural gas combined cycle
power plants (ranging from $400 to 550/kW), which can achieve higher performance than repowered
plants. However, if emission requirements are imposed on existing power plants, the cost-effectiveness of
repowering will likely improve. The main issue associated with heat rate improvements and plant
upgrading relates to environmental requirements, which may apply in case the plant increases its output
even marginally.
3-102
-------
References
1. Heat Rate Improvement Guidelines for Existing Fossil Plants; EPRI CS-4554; EPRI: Palo Alto,
CA, May 1986.
2. MacDonald, J. Upgrading Older Power Plants to Improve Their Competitiveness. Gas Turbine
World March-April 2001.
3. Smith, D. J. Plant Upgrades Add Capacity at Less Cost. Power Eng. April 2001.
4. Daniels, D. Turbine Deposits Rob Megawatts, But You Can Catch the Thief. Power Magazine
March/April 1999.
5. Giovando, C. Explore Opportunities from Today's Steam Turbines. Power Magazine July/August
1998.
6. Chambers, A. Partial Rotor Rebuilds Save Time and Money. Power Eng. June 1998.
7. Brummel, H. Upgrades Give Power Plans New Lease of Life. Siemens Power J. Jan 1998.
8. Bauer, G.; Joyce, J. S. The Benefits of Parallel Repowering Existing Steam Turbines with Gas
Turbines. Presented at the CEA Electricity '96, Montreal, Canada, April 30, 1996.
9. Farber, M.A. Economic Evaluation of Plant Upgrading Investments/Case Studies; EPRI
EA-3890, Vol. 2; EPRI: Palo Alto, CA, Feb 1985.
10. Makansi, J. Rehabilitation. Power Magazine May/June 1999, Special Report.
11. Repowering as a Competitive Strategy. EPRIJ. Sept/Oct 1995.
12. Stoll, H. G.; Smith, R. W.; Tomlinson, L. O. Performance and Economic Considerations of
Repowering Steam Power Plants; GER-3644D; GE Industrial and Power Systems.
3.2.5.3 Power Plant Optimization
Power Plant Optimization Summary
Status Commercial
SO2 Reduction, % Up to 1.5 at full load and up to 5 at low load
NOX Reduction, % 5-35
Hg Reduction, % Up to 1.5 at full load and up to 5 at low load
Cost
Stand-alone or one-time: $25,000 - 50,000
Capital Advisory: $100,000 - 250,000
Closed loop: $150 - 300,000
Fixed O&M ($/yr) $10,000 - 20,000
Variable O&M (mills/kWh) None
Applicability Both new and retrofit applications
Issues None
3-103
-------
Technology Description
Power plant optimization generally involves the use of a software program that can determine the set of
operating conditions which optimize an "objective function" such as "maximize plant efficiency,"
"minimize NOX emissions," or "maximize plant efficiency while maintaining NOX, CO, and unburned
carbon below certain limits." The software resides in a separate personal computer, which is linked
directly to the existing Data Acquisition System (DAS) of the power plant to monitor plant performance.
With the data obtained from the DAS, it establishes a model of how the process works (usually by
employing neural network-based technology) and then identifies the optimum operating conditions.
The optimum advice could be implemented automatically (without any action from the plant operator)
through a direct link into the power plant control system ("closed-loop" system). Alternately, it could be
shown on a computer screen in the control room and be left to the power plant operator to decide whether
and how to implement it ("advisory" system). A third variation of the power plant optimization system is
"stand-alone" or "one-time" optimization, in which case the software is linked temporarily to the DAS
(usually a few weeks), optimum settings are identified, and the existing control system (mainly the
"control set-points") is adjusted to take advantage of the optimum settings. After the completion of the
stand-alone or one-time optimization, the software may be removed from the plant or left on site so that
optimization can be repeated in the future.
Commercial Readiness and Industry Experience
The technology is commercially available. Presently, there are more than 250 boilers with power plant
optimization systems in operation or planned to be installed in the U.S.1 More than 95 percent of these
boilers are coal-fired, but there are a few gas- and oil-fired boilers, too. When the first optimization
systems were installed (1993 - 95), most were stand-alone or one-time and advisory systems, because
utilities wanted to gain experience and confirm their operation and reliability before putting them in
automatic control. Presently, most of the optimization systems are of the "closed loop" type.
Optimization software is offered by a number of organizations such as: NeuCo (ProcessLink), Pegasus
Technologies Corp. (offering NeuSIGHT), Pavillion (Process Insights software), Praxis Engineers/GE
(PECOS), ULTRAMAx,2'3 and URS (formerly Radian Corp. offering GNOCIS)4 Also, Lehigh
University offers BoilerOP software,5 but this software is geared more for optimization through
parametric testing.
Emission Control Performance
Industry experience has demonstrated that optimization can achieve 5-35 percent NOX reduction, along
with heat rate improvement (up to 1.5 percent at full load and up to 4 percent at low loads). Boilers that
have more operating variables, have not been tuned recently, or have not been tuned with the objective to
reduce emissions, fall in the middle to the high end of the 5-35 percent range. Boilers that have been
recently tuned (following a low NOX burner retrofit or a scheduled outage) are expected to achieve NOX
and heat rate improvement in the low end of the above range. SO2 and mercury emissions are expected to
be reduced proportionally to the heat rate improvement.
O&M Impacts
Significant O&M impacts are not expected in conjunction with power plant optimization. However, the
right precautions need to be taken to avoid O&M impacts, especially long-term impacts which are
difficult to monitor and quantify. Examples of such impacts are potential waterwall corrosion, which may
be caused by aggressive combustion staging (high overfire air flow rate) and low excess air. Also, if the
optimization program shows preference in using one specific pulverizer more than others, then
precautions need to be taken so that pulverizer maintenance schedule is adjusted to avoid reliability
problems.
3-104
-------
Costs
Typical capital costs,1 including installation but not including the time contributed by utility staff, are:
• For stand-alone or one-time optimization: $25,000-50,000
• For advisory optimization: $100,000-250,000
• For closed loop optimization: $150,000-300,000
In most cases, O&M costs are limited to a small license agreement (up to $10-20,000 per year) with the
software supplier to provide technical support on an as-needed basis, as well as revised versions of the
software.
Issues Associated with the Technology; Future Outlook
Increased competition due to deregulation and emission trading is expected to provide an additional
impetus for more extensive use of optimization software. There are no major issues associated with their
application.
References
1. Retrofit NOX Controls for Coal-fired Utility Boilers-2000 Update Addendum; EPRI TR-102906;
EPRI: Palo Alto, CA, Dec 2000.
2. Payson, E. P.; Brown, R.; Earley, D.; Moreno, C. Obtaining Improved Boiler Efficiency and NOX
Using Advanced Empirical Optimization and Individual Burner Instrumentation on a Boiler
Operated in Load-Following Mode. Presented at EPRI's 12th Heat Rate Improvement Conference,
Dallas, TX, Jan 30-Feb 1, 2001.
3. Romero, C. E.; Levy, E. K.; Sarunac, N. Impacts of Combustion Optimization on Power Plant
Heat Rate. Presented at EPRI's 12th Heat Rate Improvement Conference, Dallas, TX, Jan 30-Feb
1,2001.
4. Warriner, G.; Logan, S.; Pascoe, S. Full-scale Implementation of Results for GNOCIS TM Plus.
Proceedings of the U.S. EPA-DOE-EPRI Combined Power Plant Air Pollutant Control
Symposium: The MEGA Symposium and The A&WMA Specialty Conference on Mercury
Emissions: Fate, Effects, and Control, Chicago, IL, Aug 20-23, 2001.
5. Sarunac, N.; Romero, C. E.; Levy, E. K.; Weston, W. C. Anatomy of a Successful Optimization
Project. Proceedings of the U.S. EPA-DOE-EPRI Combined Power Plant Air Pollutant Control
Symposium: The MEGA Symposium and The A&WMA Specialty Conference on Mercury
Emissions: Fate, Effects, and Control, Chicago, IL, Aug 20-23, 2001.
3-105
-------
Chapter 4
Summary
During 2001, fuel combustion-electric utilities contributed 69 percent of the total SO2 emitted, and 22
percent of the NOX emitted in the United States. As demand for electricity generation and the need to
address emissions of multiple pollutants grow, new emissions control technologies are emerging. In the
past, these technologies have mainly addressed one pollutant at a time in harmony with individual
environmental requirements. In light of the need to address emissions of multiple pollutants, the
development of cost-effective and easily retrofittable emissions control technologies capable of
simultaneously handling several pollutants is increasingly becoming necessary.
This report presents and analyzes various existing and novel control technologies designed to achieve
multi-emission reductions. It provides an evaluation of multi-emission control technologies and options
that are available for coal-fired power plants with a capacity of 25 MW or larger in the United States. The
report addresses technologies and options that are capable of simultaneously controlling at least two of
the following three pollutants: NOX, SO2, and Hg.
The selection of control technologies was limited to those for which at least one installation, no lower
than 5 MW or equivalent, was in operation in a power plant worldwide as of July 1, 2001. Advanced
power generation technologies, power plant rehabilitation-upgrading options, fuel switching or blending,
and power plant optimization were also included due to their ability to reduce multiple emissions.
The technology reviews are based on several sources of information including technology vendors,
technical papers, expert consultations, reports published by the Department of Energy, and trade
publications. The results of this review reveal the following:
• The number of technologies under development, demonstration, or already commercially available is
significant; most of them have been used already in power plants or industrial applications;
• Fifteen out of the 27 evaluated technologies are in commercial or early commercial stage;
• A couple of technologies (Activated Coke and E-BEAM), which are not commonly used in the
United States, are already in use in other countries;
• Six of the evaluated control technologies (Activated Coke, ECO, wet FGD with SCR, EnviroScrub,
LoTOx and K-Fuel) are capable of achieving reductions of SO2 NOX, and mercury; most of these
exhibit the potential to significantly control (above 80 percent) all three pollutants (SO2, NOX, and
Hg).;
• Some technologies, such as SNOX, SNRB, AD VACATE, and CZD, have been tested either in pilot or
demonstration scale in the early phase of the U.S. Department of Energy's Clean Coal Technology
(CCT) program, but have not been adopted by industry. Some of these technologies could become
more cost-effective as environmental requirements evolve.
• A number of technologies under early development (Pioneer Technologies' Non-thermal Plasma Arc
technology, Consummator's plasma arc by-product recovery, Phoenix's retrofit slagging combustor,
ISCA's C12 injection, and BioDeNOx), which were not evaluated in this report, may offer additional
effective emission control of multiple pollutants in the near future.
4-1
-------
For each evaluated technology, the report includes background information, applicability, status of
commercialization, any secondary environmental impacts of the technologies, identification of primary
process variables that impact performance relative to NOX, SO2, and mercury, as well as capital and
operation and maintenance (O&M) costs. Table 4-1 includes summary descriptions of 27 technologies
identified as multi-emission control technologies, which have reached a stage of development beyond
pilot scale. These technologies can broadly be divided into:
• Environmental control (post-combustion controls),
• Advanced power generation, and
• Power plant upgrading and operating options.
For the purposes of this report, environmental controls include those processes that control SO2-mercury,
SO2-NOX, and SO2-NOx-mercury emissions. Injection of activated carbon upstream of electrostatic
precipitators and bag filters was also included because of the significant role it may play in controlling
mercury from existing power plants.
Advanced power generation technologies include circulating fluidized bed combustion (CFB), integrated
gasification combined cycle (IGCC), pressurized fluidized bed combustion (PFBC), and supercritical
pulverized coal. Plant upgrading and operating changes include fuel blending and cofiring, plant
upgrading and efficiency improvements, and plant optimization.
Although the report is limited to addressing technologies with a certain level of maturity, the authors
expect a rapid technological evolution in the development and commercialization of several multi-
emission control technologies not necessarily addressed in this report.
4-2
-------
Table 4-1. Summary Descriptions of 27 Multi-emission Control Technologies for Coal-fired Units
Technology
Status3
Emissions Reductions
Applicability
Issues
SO2 and Mercury Control
Dry scrubbers
SO2 sorbents
Activated carbon with SO2
sorbent processes
Activated carbon with
participate controls
Wet FGD with mercury
oxidation processes
Wet FGD with wet ESP
Advanced dry FGD
PEESP
MerCAP
C
P/C
P/C
P/C
p
C/P
P/C
B/P
B/P
SO2: 90-98%; NOX: NAC;
Hg: 0- 95%
SO2: 40-85%; NOX: NA;
Hg: NA
SO2: 40-85%; NOX: NA;
Hg: Up to 90%
Hg: 50-90%
SO2: 95%; NOX: NA;
Hg: >80%
SO2: 99%; NOX: NA;
Hg: Up to 80%
SO2: 90-98%; NOX: NA;
Hg: 0-95%
SO2: >90% (with wet FGD);
NOX: NA; Hg: Up to 98%
SO2: NA: NOX: NA;
Hg: >80%
Low-to-medium sulfur coal
Units with ESP or FF for
particulate control
Units with ESP or FF for
particulate control
Retrofit and new units with ESP or
FF
Wet scrubber plants
Integration with wet scrubbers,
retrofit dry ESPs, new units
SO2-Hg control for low-to- medium
sulfur coal (same as spray dryers)
New and retrofit
New and retrofit; scrubbed flue
gas
Hg removal can vary significantly with coal type,
operating conditions
Calcium-based compounds not used
commercially in coal fired plants. Waste
disposal issue with sodium-based compounds.
Potential impacts on ESP or FF
Not used commercially, potential impacts on
ESPorFF
Not widely demonstrated at full scale, ash
salability, ESP and FF performance, impact of
coal type (mercury speciation)
Full scale demonstration underway, insufficient
information at present
Few applications in power industry, potentially
expensive alloys required
Hg removal may vary significantly with coal
type, operating conditions (similar to spray
dryers)
Early stage of development; demonstration and
further assessment of the technology is needed
Did not perform well in unscrubbed gas; >80%
for 10 ft long plates spaced 0.5 inches apart
SO2 and NO* Control
E-BEAM
ROFA-ROTAMIX
SNOx
SNRB
THERMALONOx with wet
FGD or FLU-ACE
C/D
C/D
C
P
D
SO2: >95%; NOX: Up to 90%;
Hg: NA
SO2: 90%; NOX: 40-60%;
Hg: 67% with Trona, 89% with
CaCO3
SO2: >90%; NOX: >90%;
Hg: 0%
SO2: 80-90%; NOX: 90%;
Hg: NA
SO2: Up to 95%; NOX: Up to
90%;
Hg: NA
New and retrofit
Existing plants
New and retrofit
New power plants and retrofits
New and retrofit
Demonstration is required. High costs and
auxiliary power requirements
Demonstration phase
Cost-effectiveness
Requires demonstration
In demonstration
(continued)
4-3
-------
(Table 4-1 concluded)
Technology
Status3
Emissions Reductions
Applicability
ISSIMS
SO2, A/Ox, and Mercury Control
Activated coke
Electrocatalytic oxidation
(ECO)
Wet FGD and SCR
EnviroScrub
LoTOx
K-Fuel
c
D
C
P
D/C
D/C
SO2: 90-98%; NOX: 15-80%;
Hg: 90-99%
SO2: 98%; NOX: 90%;
Hg: 90%
SO2: 95%; NOX: 90-95%;
Hg: 40-90% depending on coal
type
SO2: >99%; NOX: 93-97%;
Hg: Up to 67%
SO2: 95%; NOX: 70-95%;
Hg: Up to 90%
SO2: Up to 30%; NOX: Up to
45%;
Hg: Up to 70%
New and retrofit
New and retrofit
Plants with SCR and wet scrubber
technologies
New and retrofit
New and retrofit
Mostly boilers burning PRB or
lignite
Demonstration of the combined SO2-NOx-Hg
control is needed in the United States
Demonstration in progress
Need additional confirmation of mercury
oxidation levels in the SCR
Demonstration required
Demonstration required
Demonstration required
Mvanceti Power Generation Ontions
Circulating fluidized bed
combustion
Integrated gasification
combined cycle
Pressurized fluidized bed
combustion
Supercritical pulverized coal
C
C/D
C
C/D
SO2: >95%; NOX: 30-70%;
Hg: NA
SO2: 99%; NOX: 80-90%;
Hg: 90%
SO2: Up to 98%; NOX: 30-
70%;
Hg: NA
SO2: 4-12%; NOX: 4-12%;
Hg: 4-12%
Mainly new
New
Mostly new
New
Successful scale-up to 400-600 MW while
maintaining its cost-effectiveness and emission
performance
High costs. Fluidized bed IGCC requires
demonstration
Some reliability problems and high costs
(relative to conventional PC and FGD). Also,
scale-up issues.
Ultra-supercritical unit requires further
demonstration
Power Plant Unaratiina
Fuel blending and cofiring
Plant upgrading
Power plant optimization
C
C
C
SO2: Up to 100%; NOX: 20-
80%;
Hg: flex%
Depends on choice
SO2: 1.5-5%; NOX: 5-35%;
Hg: 1.5-5%
New and retrofit
Retrofit
New and retrofit
Development of on-line coal quality monitoring
is needed to enhance coal blending
Impact of environmental regulations pertaining
to increased output of existing plants
None
' Status: B = Bench scale; P = pilot stage; C = commercial; D = demonstration
' NA = no removal reported
4-4
-------
End of Document
------- |