Flue Gas Desulfurization - FGD - Performance Capability

Flue Gas Desulfurization (FGD) Performance Capability
High Efficiency Design and Operating Options
12239985
12239985
Flue Gas Desulfurization (FGD)
Performance Capability
High Efficiency Design and Operating Options
1014171
Final Report, March 2008
EPRI Project Manager

C. Dene
12239985
DISCLAIMER OF WARRANTIES AND LIMITATION OF LIABILITIES
THIS DOCUMENT WAS PREPARED BY THE ORGANIZATION(S) NAMED BELOW AS AN ACCOUNT OF WORK
SPONSORED OR COSPONSORED BY THE ELECTRIC POWER RESEARCH INSTITUTE, INC. (EPRI). NEITHER EPRI,
ANY MEMBER OF EPRI, ANY COSPONSOR, THE ORGANIZATION(S) BELOW, NOR ANY PERSON ACTING
ON BEHALF OF ANY OF THEM:
(A) MAKES ANY WARRANTY OR REPRESENTATION WHATSOEVER, EXPRESS OR IMPLIED, (I) WITH RESPECT
TO THE USE OF ANY INFORMATION, APPARATUS, METHOD, PROCESS, OR SIMILAR ITEM DISCLOSED IN
THIS DOCUMENT, INCLUDING MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE, OR (II) THAT
SUCH USE DOES NOT INFRINGE ON OR INTERFERE WITH PRIVATELY OWNED RIGHTS, INCLUDING ANY
PARTY’S INTELLECTUAL PROPERTY, OR (III) THAT THIS DOCUMENT IS SUITABLE TO ANY PARTICULAR USER’S
CIRCUMSTANCE; OR
(B) ASSUMES RESPONSIBILITY FOR ANY DAMAGES OR OTHER LIABILITY WHATSOEVER (INCLUDING ANY
CONSEQUENTIAL DAMAGES, EVEN IF EPRI OR ANY EPRI REPRESENTATIVE HAS BEEN ADVISED OF THE
POSSIBILITY OF SUCH DAMAGES) RESULTING FROM YOUR SELECTION OR USE OF THIS DOCUMENT OR
ANY INFORMATION, APPARATUS, METHOD, PROCESS, OR SIMILAR ITEM DISCLOSED IN THIS DOCUMENT.
ORGANIZATION(S) THAT PREPARED THIS DOCUMENT
Washington Group International
NOTE
For further information about EPRI, call the EPRI Customer Assistance Center at 800.313.3774 or
e-mail askepri@epri.com.
Electric Power Research Institute, EPRI, and TOGETHER…SHAPING THE FUTURE OF ELECTRICITY are
registered service marks of the Electric Power Research Institute, Inc.
Copyright © 2008 Electric Power Research Institute, Inc. All rights reserved.
12239985
Citations
This report was prepared by
Washington Group International

A Subsidiary of the Washington Division of URS Corporation
7800 East Union Avenue, Suite 100
Denver, CO 80237
Principal Investigators
R. Keeth
W. Hoskins
P. Spath
This report describes research sponsored by the Electric Power Research Institute (EPRI).
The report is a corporate document that should be cited in the literature in the following manner:
F lue Gas Desulfurization (FGD) Performance Capability: High Efficiency Design and
Operating Options. EPRI, Palo Alto, CA: 2008. 1014171.
FGD Performance
12239985 Capability: High Efficiency Design and Operating Options iii
12239985
Report Summary
State-of-the-art flue gas desulfurization (FGD) system design innovations are allowing plants to achieve
SO2 reductions above 99% removal for high sulfur coals and above 98.5% for low sulfur coals.
Eventually, these developments will result in SO2 emission rates that approach the single digit ppm
level. The study summarizes the future FGD design alternatives under consideration for new and retrofit
FGD systems installed after 2010, including system modifications that that can achieve very high SO2
removal. The report also discusses the impacts that the increased demand for FGD systems are having
on the installed capital cost for these systems.
Background
More stringent regulatory requirements from state agencies and CO2 emissions control considerations
may force the U.S. coal-fired generating plants to install FGD systems with very high SO2 removal
capability. Future state and/or Federal regulatory limits on SO2 for existing plants as well as retrofit
CO2 removal process operating requirements could require FGD processes to reduce SO2 emission rates
to single digit ppm levels, especially if integrated gasification combined cycle (IGCC) plants become
the emissions benchmark for coal-fired power generation technologies. To meet these challenges for
near-zero SO2 emissions from pulverized coal plants, it is likely that FGD systems will have to incor-
porate design modifications, and utility FGD system operating practices will have to be modified
to achieve the lower utility SO2 emission rates. In any case, although utilities are already installing
FGD systems with performance guarantees of 98% SO2 removal, these performance levels are hard to
sustain after installation; and additional design upgrades may still be necessary to achieve continuous
single digit SO2 emission rates.
Objectives
■ To provide background on regulatory requirements that have resulted in a significant

increase in the number of FGD installations required over the next 5–10 years
■ To identify new units or FGD retrofits worldwide that have been designed for 98% to
99+% SO2 removal and will be installed by 2010
■ To describe technical advances that FGD system suppliers are incorporating into their
current commercial offerings to achieve lower SO2 outlet emission rates, including new
technology developments or upgrades/additions to existing process components
Approach
The project team compiled a list of FGD suppliers serving the U.S. utility industry and contacted many
of these suppliers to identify systems that have already been achieving 98% to 99% removal as well
as modifications that have been or could be incorporated to achieve outlet emissions that approach
single digit ppm levels. In addition to FGD processes, the report contains information on multi-pollutant
processes with the potential to achieve 99+% SO2 removal as well as an ability to remove NOX and
mercury and discusses how FGD process performance will influence the operation and costs of
post-combustion CO2 reduction systems installed downstream of pulverized coal fired boilers.
Results
In order to meet the 2010 operational deadlines required by the Clean Air Interstate Rule (CAIR), U.S.
utilities have been installing FGD systems at a very fast pace over the last five years. In the period from
2007 to 2010, about 180 FGD systems have been placed into operation or are under contract and
are projected to enter commercial service. The latest design modifications by some FGD suppliers have
FGD Performance
12239985 Capability: High Efficiency Design and Operating Options
led to high removal efficiencies without the need for either a major redesign of the absorber tower or
the need to have two absorbers operating in series. Despite this progress, utilities will need to consider
adopting a different operating philosophy in order to achieve 99+% SO2 removal on a multi-day
rolling average basis. They will need to place a greater emphasis on monitoring FGD performance by
augmenting operating staff with engineering staff and paying greater attention to instrumentation and
equipment maintenance.
Some of the latest process developments in the area of multi-pollutant control technologies may result
in cost-effective alternatives that achieve high SO2 removal in conjunction with removal of NOX and
mercury. As federal, state and local agencies demand additional reductions in SO2, NOX, and mercury
these integrated emissions control systems may have cost and operating advantages compared to the
installation of separate control systems for each pollutant. Integrated emissions control processes
continue to gain commercial experience, improve performance, and achieve higher reliability.
EPRI Perspective
This report focuses on high efficiency SO2 removal alternatives. Most vendor designs are sufficiently
robust enough to achieve single digit outlet emissions, but achieving this level of performance over
an extended time will probably require significant redundancy and added process oversight. Long-term
compliance with ultra low level emissions has not been demonstrated anywhere in the world. Meanwhile,
the looming possibility of regulations requiring fossil fuel-fired power plants to reduce CO2 emissions
make consideration of CO2 removal processes pertinent. The flue gas SO2 concentration from PC-fired
power plants needs to be in the low ppm range at the inlet to the CO2 removal system for these
technologies to operate cost-effectively. Thus, when used in power plant applications the operational
and chemistry requirements of CO2 removal processes provide another justification for FGD systems to
reduce SO2 emissions to single digit ppmv levels.
Keywords
Flue gas desulfurization
Emissions
NOX reduction
Clean Air
SO2
vi FGD Performance Capability: High Efficiency Design and Operating Options

12239985
Abstract
This EPRI report assesses the current state-of-the-design of high efficiency flue gas desulfurization
(FGD) control systems and the move toward systems capable of 99+% SO2 removal and/or SO2
emissions approaching the single digits. Utilities are responding to the emission limits and deadlines
of the Clean Air Interstate Rule (CAIR) but are also specifying FGD systems with increasingly higher
SO2 removal efficiencies as a result of local regulations, incentive to bank as many SO2 allowances as
possible prior to 2010, and/or anticipation of the future need to install CO2 removal systems. In
addition, environmental activists/groups are using the courts or public opinion to pressure public utility
commissions to require that utilities to document how PC plants can achieve emission levels similar to
integrated-gasification-combined cycle (IGCC) or natural gas combined-cycle (NGCC) plants.
This report is based on discussions with the major suppliers, information compiled from recently
completed EPRI studies, and research conducted during this study. It identifies the FGD system designs
of major suppliers and the latest modifications being made in order to achieve very low SO2 emissions.
Based on the information supplied from the FGD system suppliers, it is apparent that existing system
designs and operating conditions can achieve these 99+% reduction efficiencies for instantaneous
periods. However, long-term continuous operation is another matter. Modifications to allow this level of
performance on a continuous basis are discussed throughout the document. These design modifications
must overcome the emissions upsets that occur with current designs and result in intermittent tran-
sients of higher SO2 emission rates. These upsets are typically attributed to such causes as operating
transients, non-uniformity within the absorber tower, increased particulate emissions from performance
upsets in the particulate collection system, and instrumentation failures.
The report also reviews multi-pollutant emission control technology capabilities that can potentially
achieve very high levels of SO2 reduction as well as simultaneous removal of NOX and Hg. It discusses
the significant increases in costs of FGD systems that have occurred since about 2004, including the
factors that are causing the costs to rise so quickly. Finally, the report briefly discusses how CO2
removal will require very low inlet concentrations of SO2 in the flue gas to operate in a cost-effective
manner.
FGD Performance
12239985 Capability: High Efficiency Design and Operating Options vii
12239985
Contents
1 INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1
2 WHY INCREASE SO2 REMOVAL EFFICIENCY? . . . . . . . . . . . . . . . . . . . . . . . . . 2-1
3 FGD SYSTEMS DESIGNED FOR HIGH SO2 REMOVAL. . . . . . . . . . . . . . . . . . . . 3-1
4 APPLICABLE SO2 REMOVAL PROCESSES. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-1

Chiyoda CT-121 Jet Bubbling Reactor (JBR) . ................................................................ 4-1
Chiyoda Process Description................................................................................ 4-1
Recent CT-121 Process/Design Developments......................................................... 4-3
Mitsubishi/Advatech Double Contact Flow Scrubber (DCFS)........................................... 4-4
Mitsubishi (MHI) Process Description..................................................................... 4-4
Recent MHI Process/Design Developments............................................................. 4-4
Alstom Power SO2 Removal Processes.......................................................................... 4-6
Alstom Spray Tower Process Description................................................................ 4-6
Alstom Flowpac Process Description...................................................................... 4-7
Recent Flowpac and Spray Tower Process/Design Developments.............................. 4-8
Babcock & Wilcox – Wet Lime or Limestone Forced Oxidation ...................................... 4-9
B&W Process Description..................................................................................... 4-9
Recent B&W Process/Design Developments........................................................... 4-10
Babcock Power......................................................................................................... 4-11
Babcock Power Process Description...................................................................... 4-11
Recent Babcock Power Process/Design Developments............................................. 4-11
Siemens-Wheelabrator – Wet Limestone Forced Oxidation Process................................. 4-12
Hitachi America........................................................................................................ 4-12
Hitachi Process Description.................................................................................. 4-12
Recent Hitachi Process/Design Developments......................................................... 4-13
Alternate FGD Technologies for High SO2 Removal Efficiency......................................... 4-13
Dry Injection of Trona.......................................................................................... 4-13
Dynawave/Reverse Jet Scrubber........................................................................... 4-13
5 FGD SYSTEM COST ESCALATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-1

Background.............................................................................................................. 5-1
Increases in Material and Equipment Prices.................................................................. 5-3
Cost Considerations Related to Craft Labor Workforce................................................... 5-3
Adequacy of Engineering Staff................................................................................... 5-12
Increased Cost of FGD Systems and Impact of Higher % Removal................................... 5-12
6 MULTI-POLLUTANT TECHNOLOGIES. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-1

Powerspan Electrocatalytic Oxidation Process (ECO™ & ECO2™)................................... 6-1
Powerspan Process Description............................................................................. 6-1
CANSOLV™ Process................................................................................................. 6-2
Airborne Process....................................................................................................... 6-4
J-Power/EPDC ReACT Process..................................................................................... 6-6
ReACT Process Description................................................................................... 6-6
Recent ReACT Process/Design Developments......................................................... 6-8
Other Developing Multi-Pollutant Processes................................................................... 6-9
FGD Performance
12239985 Capability: High Efficiency Design and Operating Options ix
7 CO2 REMOVAL PROCESSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-1
Types of CO2 Removal Processes................................................................................. 7-1
Chemical Solvents............................................................................................... 7-1
CANSOLV................................................................................................................ 7-2
Physical Solvents................................................................................................. 7-2
Chilled Ammonia Scrubber.................................................................................. 7-3
Powerspan - ECO2™ Process................................................................................ 7-3
Cost Impacts of SO2 Concentration on CO2 Removal Processes....................................... 7-4
Introduction ....................................................................................................... 7-4
Incremental Costs of SO2 and CO2 Removal for Increasing FGD System Efficiencies... 7-4
8 SUMMARY AND CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-1
9 REFERENCES. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-1
FGD Performance Capability: High Efficiency Design and Operating Options

12239985
List of Figures
Figure 1‑1 Current and Projected SO2 Scrubber and NOX SCR Controls on PC Units ........... 1-2
Figure 3‑1 Incremental SO2 Control Cost for Increasing Removal Rates for 500 MW LSFO... 3-4
Figure 4‑1 Chiyoda CT-121 Schematic Flow Diagram (Round Vessel)................................. 4-2
Figure 4‑2 Chiyoda CT-121 Schematic Flow Diagram (Rectangular Vessel)......................... 4-2
Figure 4‑3 MHI/Advatech Schematic Flow Diagram Single Tower...................................... 4-5
Figure 4‑4 MHI/Advatech Schematic Flow Diagram Dual Tower........................................ 4-5
Figure 4‑5 Alstom Flowpac Process Flow Diagram............................................................ 4-8
Figure 4-6 Siemins-Wheelabrator Generic Wet Scrubbing Vessel....................................... 4-12
Figure 4-7 Hitachi FGD Absorber................................................................................... 4-14
Figure 5-1 Cost Index for Ready-Mix Concrete................................................................. 5-4

Figure 5-2 Cost Index for Structural Steel......................................................................... 5-5
Figure 5-3 Cost Index for Copper Wire.......................................................................... 5-6
Figure 5-4 Cost Index for Large Centrifugal Pumps........................................................... 5-7
Figure 5-5 Cost Index for Large Centrifugal Fans.............................................................. 5-8
Figure 5-6 Cost Index for Wet Scrubber.......................................................................... 5-9
Figure 5-7 Published Wet FGD System Actual Installed Costs . .......................................... 5-10
Figure 6-1 ECO™ Process Flow Diagram........................................................................ 6-1

Figure 6-2 CANSOLV Simplified Process Flow Diagram.................................................... 6-3
Figure 6-3 Airborne Simplified Process Flow Diagram....................................................... 6-4
Figure 6-4 ReACT Simplified Process Flow Diagram......................................................... 6-8
Figure 7‑1 PRB Coal – Incremental Cost of FGD System Plus CO2 Removal System............... 7-6
Figure 7‑2 Lignite - Incremental Cost of FGD System Plus CO2 Removal System.................... 7-7
Figure 7‑3 Appalachian Coal - Incremental Cost of FGD System Plus CO2
Removal System........................................................................................... 7-8
FGD Performance
12239985 Capability: High Efficiency Design and Operating Options xi
12239985
List of Tables
Table 3‑1 High Efficiency SO2 Reduction Systems Installed on U.S. Utility
Boilers by 2010........................................................................................... 3-2
Table 5-1 Percentage of Workers Age 50 and Older (as of 2006)................................... 5-11
Table 7‑1 Summary of SO2 Requirements for CO2 Removal Technologies........................... 7-1
FGD Performance
12239985 Capability: High Efficiency Design and Operating Options xiii
12239985
Introduction
1 INTRODUCTION
T his report is intended to provide EPRI’s Integrated Environmental Control Interest Group (IECIG)
and the future Integrated Environmental Control Target (IEC T75) with a summary of the methods
that may be used to achieve increases in FGD system performance. The objective of this report is to
provide an assessment of the design and operating procedure upgrades for wet FGD processes that
are planned by the major system suppliers. Developing emissions control technologies with high SO2
reduction efficiency capability are also described.
The report also identifies high efficiency FGD processes that have been installed and operated in the
past, as well as those systems under contract for installation within the next few years. The information
provided in this report is based on public domain literature and/or information obtained from supplier
interviews. The report also contains descriptions of the FGD designs offered by the major OEMs serv-
ing the U.S. utility market. In addition, the report discusses the innovations that are currently being
incorporated made by these suppliers to increase SO2 removal to levels above 99%.
The major suppliers serving the U.S. utility market for wet FGD systems include the following:
■ Siemens-Wheelabrator
■ Chiyoda
■ Babcock & Wilcox
■ Babcock Power
■ Alstom
■ Advatech/Mitsubishi
In 2006, these six companies had combined sales covering over 90% of the wet FGD market. Other
wet FGD system suppliers with installations currently operating in North America include Hitachi
America and Marsulex.
In order to meet the 2010 operational deadlines required by CAIR, U.S. utilities have been installing
FGD systems at a very fast pace over the last five years. In the period from 2007 to 2010, about 180 FGD
systems have been placed into operation, are under contract and are projected to enter commercial service.
Figure 1-1 reflects the FGD controlled capacity in terms of gigawatts of power generation, identifying
the number of FGD systems already installed, currently being installed, or projected to be installed by 2010
and beyond. This figure indicates that 111 units were committed between 2007 and 2010, an additional
69 units are projected to be on-line by 2010, and another 180 FGD installations are projected by 2020.
As part of the CAIR regulations, utilities are allowed to bank one allowance for each ton of SO2
removed for those units that are operating prior to 2010. Per the CAIR regulations, allowances generated
after 2010 will be devalued by one-half, requiring two allowances to emit one ton of SO2 removed.
As a result, power generators are trying to get FGD systems operating as soon as possible in order to
bank as many allowances as they can prior to 2010.
In response to the 2010 requirements of CAIR, local regulations, and/or the incentive to bank SO2
allowances 1-for-1, some utilities have been specifying FGD systems with SO2 performance guarantees
of 98% to 99% removal. This report identifies 15 FGD systems that have been or are currently being
designed and/or guaranteed to achieve 98% to 99% SO2 removal. These systems will be in operation
starting in 2008 and their ability to consistently achieve this level of SO2 removal on a long-term basis
will be demonstrated within the first couple of years of operation.
FGD Performance
12239985 Capability: High Efficiency Design and Operating Options 1-
Introduction
Figure 1‑1
Current and Projected SO2 Scrubber and NOX SCR Controls on PC Units (taken from the
Acid Rain and Related Programs, 2006 Progress Report, Environmental Protection
Agency, 2006, Figure 27)
The latest design modifications by some FGD suppliers have led to high removal efficiencies without the
need for either a major redesign of the absorber tower, or the need to have two absorbers operating
in series. These changes include:
■ Installations of wall baffles/rings

■ Changes in spray patterns within the absorber
■ Installation of better spray nozzles to produce smaller droplet sizes, thereby improving
the liquid to gas contact
■ Increase the shear forces on the droplets to overcome the liquid-phase controlled reaction
between SO2 and limestone in the slurry
■ Installation of offset headers and modifying spray angles to further eliminate the potential
for gas sneakage and minimize slurry impingement on the walls of the absorber.
In a number of cases, it has also been necessary to increase liquid to gas (L/G) slurry recycle ratio by
20–30% and to use spare absorber circulation pumps and spray headers that might not be needed if a
lower reduction efficiency was acceptable. Other FGD absorber designs may not rely on an increase in the
L/G, but rather will adjust some other operating characteristic of their system such as increases in pH, addi-
tion of a second stage of scrubbing, increasing the froth contact depth, use of chemical additives, etc. More
details of these process design or operating adjustments will be provided in Section 4 of this document.
In order to achieve 99+% SO2 removal on a multi-day rolling average basis, utilities will need to consider
adopting a different operating philosophy. To achieve these single digit SO2 levels, the utilities will
need to place greater emphasis on monitoring FGD performance including augmenting operating staff
with engineering staff and paying greater attention to instrumentation and equipment maintenance.
1- FGD Performance Capability: High Efficiency Design and Operating Options
12239985
Introduction
In essence, U.S. utilities would need to adopt the Japanese FGD operating philosophy, which includes
hands-on involvement of operating engineers on a day to day basis, regular preventative maintenance
to instruments and operating components, and consistent vigilance by operating staff to ensure that
the desired process chemistry is achieved so that the FGD performance can be maintained.
FGD suppliers are constantly upgrading their system designs. Some of these upgrades have been
demonstrated in commercial installations overseas, and their R&D efforts continue to ensure that the
design modifications will be applicable to U.S. coal-fired utility boiler operating conditions. However,
some modifications in these Japanese and European plants will not obtain the same performance
benefits in the U.S. because of the sulfur content of U.S. coals, required longer operating periods
between planned shutdowns, or the limited demand for FGD byproducts. However, it is expected that
experience the OEMs derive from overseas innovations will be, and has been, used to improve U.S.
FGD system performance.
FGD Performance
12239985
Why Increase SO2 Removal Efficiency?
2 WHY INCREASE SO2 REMOVAL EFFICIENCY?
T he CAIR, mercury, and visibility regulations have been imposed on the U.S. utility industry by the
Environmental Protection Agency (EPA) within the last few years. Most power producers have
concluded that continuing public concerns regarding the overall impacts of coal-fired power generation
will result in even more stringent limits of NOX, SO2, mercury, and primary particulates in the future. In
addition, some state regulations have established more stringent SO2 limits than the federal emission
standards. Moreover, the global warming theory has the attention of the public as well as federal and
state regulators, leading some in the utility industry to believe that CO2 emission control for coal-fired
power plants is inevitable.
This report concentrates on the FGD design and operating modifications that may be necessary to
achieve very high SO2 removal efficiency from the flue gas generated by pulverized coal-fired power
plants (PC plants). It also addresses how the need for higher SO2 removal is being influenced by more
stringent state regulations, utilities desire to bank as many SO2 allowances as possible, and comparisons
in emissions performance between IGCC and PC-fired power plants.
Utilities that have decided that PC-fired plants are more viable to meet increasing load demand
because they are more cost-effective than IGCC designs may face regulatory requirements for new and
retrofit FGD systems that can reduce sulfur emission to levels achieved by IGCC. In addition, utilities
may be required to reduce the CO2 emitted by PC-fired power plants. Some CO2 removal processes
operate more efficiently and are more cost-effective when inlet SO2 concentrations are very low. All
of these factors are leading utilities toward FGD processes that can achieve SO2 emission levels that
approach the single digit level when firing all types of coal in PC plants.
FGD Performance
12239985
FGD Systems Designed for High SO2 Removal
3 FGD SYSTEMS DESIGNED FOR HIGH SO2 REMOVAL
W ithin the past five years, about 50,000 MW of existing PC-fired plants have been retrofit with
FGD systems. In addition, all new PC-fired plants installed within the past five years have been
equipped with high efficiency FGD systems. In most cases, the new and retrofit FGD systems were
designed to achieve at least 95% SO2 removal. In some cases, state regulations, consent decrees, and
other issues resulted in FGD systems that were specified to achieve SO2 removal efficiencies higher
than 97% at least for the FGD system performance guarantee.
Table 3-1 shows 14 FGD retrofits that are being designed for 98% or 99% SO2 removal. Of the 14
systems, 9 systems are being designed for 98% removal and 5 systems are being designed for 99%
removal. The plants designed for 99% removal are scheduled to be operating in late 2008 or early
2009. Based on the reported 2007 uncontrolled SO2 content of the coal burned, the SO2 emission from
the FGD systems designed for 99% removal will be less than 0.02 lb/MMBtu. This table also shows the
existing Trimble unit FGD system that was upgraded and designed for 98% SO2 removal in
2005. Further, it shows the Coleman units that were designed for 95% SO2 removal. Although
designed for 95% removal, the Coleman FGD system has been able to achieve more than 99%
SO2 removal.
Over the past few years, more stringent state regulations, interest in generating more SO2
allowances, and other factors have resulted in specification and purchase of FGD systems de-
signed for increasingly higher SO2 removal. Up to now, the cost of achieving 97% SO2 removal
has resulted in some redesign of the absorber internals. However, higher percentages of SO2
removal will require the use of higher L/G ratios in spray tower configurations, and can require
changes in process chemistry and operating philosophy as well. These factors result in higher
capital and operating costs associated with larger absorber slurry pumps and in some cases the
use of chemical additives.
Based on the information supplied from the FGD system suppliers, it is apparent that existing system
designs and operating conditions can achieve these 99+% reduction efficiencies for instantaneous
periods. However, long-term continuous operation is another issue. Modifications to allow this level of
performance on a continuous basis are discussed throughout the document. These design modifications
must overcome the emissions upsets that occur with current designs, resulting in intermittent transients of
higher SO2 emission rates. These upsets are typically attributed to:
■ Operating transients
■ Non-uniformity within the absorber tower
■ Increased particulate emissions from performance upsets in the particulate collection system
■ Instrumentation or control system failures, etc.
An EPRI study completed in 2006 identified the incremental cost of SO2 removal for one case based
on the use of a wet limestone with forced oxidation FGD system (LSFO) in a spray tower absorber
vessel. The system was installed on one 500 MW unit burning Appalachian coal (Pittsburgh # 8 coal
with HHV of about 13,400 Btu/lb and 2.1 wt. % sulfur). Figure 3-1 provides the incremental cost
of this LSFO case as the SO2 removal increases from 90% to 99%. From this figure, it is apparent that
the incremental cost increases moderately from 92% to 97%, and dramatically rises from 97% to 99%.
To achieve these removal rates, the L/G ratio must be increased for each percentage point increase in
SO2 removal in this scenario. In addition, for SO2 removal above 96%, dibasic acid can be added to
FGD Performance
3-
Table 3‑1
High Efficiency SO2 Reduction Systems Installed on U.S. Utility Boilers by 2010
Customer Plant or Supplier State Capacity Fuel SO2 Ave. Projected Projected Commercial
Power (MW) Removal Pre-scrubbed Outlet Outlet SO2 Operation
Station SO2 Content, SO2 Emissions, Date
lb/MMBtu Emission ppmv
(1st Qtr 2007) Rate lb/
MMBtu
Constellation Brandon Not selected MD Coal 98% Bituminous, 1.1 0.022 10 2010
FGD Performance
2 × 685
Energy Shores
12239985
1&2
OH Valley Kyger CT-121 OH 5 × 217 Coal 98% Bituminous, 1.6 0.032 15 2009
Elec. Creek 1-5
Reliant Keystone Wheelabrator PA 2 × 936 Coal 98% Bituminous, 3.0 0.06 25 2009
Energy 1&2
AEP John B&W WV 1 × 816 Coal 98% Bituminous, 1.3 0.026 <15 2009
Amos
Reliant Cheswick Wheelabrator PA 1 × 565 Coal 98% Bituminous, 2.4 0.048 20 2009
Energy
Alabama Gorgas Advatech AL 1167 [1] Coal 98% Bituminous, 2.1 0.042 20 2008
Power 8,9 & 10
LG&E Ghent 2, Babcock KY 3 × 556 Coal 98% Bituminous, 1.1 0.022 10 2008
Energy 3&4 Power
LG&E EW Babcock KY 738 [2] Coal 98% Bituminous, 2.7 0.054 25 2009
Energy Brown Power
1,2 & 3
Minnesota Boswell Hitachi MN 1 × 350 Coal 98% Subbituminous, 0.022 10 10/09
Power #3 1.1
Ameren Coffeen Hitachi IL 1 × 360 Coal 99% Bituminous, 1.4 0.014 <10 4/09
Capability: High Efficiency Design and Operating Options

Energy #1
Gen. Co
(Continued)
Ameren Coffeen Hitachi IL 1 × 590 Coal 99% Bituminous, 1.4 0.014 <10 3/09
Energy #2
Gen. Co
Ameren Duck Hitachi IL 1 × 444 Coal 99% Bit. & Sub., 0.004 <10 11/08
12239985
Energy Creek #1 0.41
Gen. Co
FGD Performance
Ameren UE Sioux #1 Hitachi MO 1 × 535 Coal 99% Bit. & Sub., 1.5 0.015 <10 10/08
Ameren UE Sioux #2 Hitachi MO 1 × 535 Coal 99% Bit. & Sub., 1.5 0.015 <10 12/08
LG&E Trimble Babcock KY 566 Coal 98% Bituminous, 5.5 0.11 50 2005
Energy [3] #1 Power Upgrade
Western KY Coleman Wheelabrator KY 3 × 160 Coal 95% [5] Bituminous, 4.5 0.045-0.09 20-50 02/06
Energy [4] [4]
TXU Oak Babcock TX 2 × 860 Coal 98% Lignite NA NA 2010
Grove #1 Power
&2
Alcoa Warrick Babcock IN 300 Coal 98% Bituminous, 1.4 0.014 <15 2008
#4 Power
LS Power Longleaf TBD GA 2 × 600 Coal 95% PRB, 1.1 0.065- 25-40 TBD
1&2 0.105
Toquop Toquop 1 TBD NV 1 × 750 Coal 97% Bit. & Sub. 0.06 25 TBD
FP&L Glades TBD FL TBD Coal 97% Bitum., 1.3 0.04 17 TBD
County
1 – Gorgas 8, 9, & 10 are 188, 190 and 789 MW , respectively.
2 – EW Brown 1, 2, & 3 are 113, 179, and 446 MW, respectively.
3 – Trimble met 98% removal guarantee. Two short-term tests – performance test showed 99.3% removal of 6.6 lb/MMBtu coal. Thus, outlet SO2 = 0.05 lb/MMBtu (20 ppmv).
Ninety days later, 2nd test showed 99.2% removal

4 – Coleman: 1 absorber for 3 units
5 – Coleman: Reports indicate SO2 removal “generally exceeds 98% with 99% removal being quite common”. At 98% removal of 4.5 lb/MMBtu, outlet SO2 = 0.09 lb
/MMBtu (40 ppmv); at 99% outlet SO2 is about 0.045 lb/MMBtu (20 ppmv).
3-
$/Incremental tonne SO2 Removed (Dec-2007) Incremental Cost of SO2 Control for LSFO, Pitts #8 Coal - SI Units
$3,000
Levelized Control Cost of LSFO System,
BASIS/DEFINITIONS:
1.Incremental Cost in $/tonne SO2 is Difference in Annual Cost divided by Annual
$2,500 Difference in SO2 Removed.
2. Levelized Costs Include TCR and All O&M Costs.
3. NOTE: SO 2 Removal is Based on Design/ Performance Conditions.
$2,000 4. 90% SO2 Removal is Base Case.
$1,500
$1,000
Low End of
$500 Range
$0
0.91 0.92 0.93 0.94 0.95 0.96 0.97 0.98 0.99 1
SO2 Removal, %
Figure 3‑1
Incremental SO2 Control Cost for Increasing Removal Rates for 500 MW LSFO
the circulating slurry to further enhance performance. The impacts of increasing the L/G ratio include:
need to purchase larger absorber circulation piping, more spray nozzles, larger and/or additional
absorber circulation pumps, and an increase in the flue gas-side pressure drop, and repositioning of
the spray header to accommodate the new spry pattern at the increased flow rates.
Earlier assessments identified other design modifications that might be considered to achieve continuous
low SO2 emission rates:
■ Addition of a polishing wet scrubber using highly reactive, clear liquor recycle solutions.
This might include sodium or ammonia scrubbing solutions. However, the sulfite/sulfate
salts produced by these systems might result in additional challenges for final disposal.
■ Installation of dry alkali injection and a polishing baghouse might be considered.
■ Improvements in performance of the FGD system components might be achieved by the
use of better materials of construction, careful attention to management of the process
chemistry, design for flue gas flow uniformity throughout the scrubbing vessel.
■ Increasing the gas/liquid contact period through alternate scrubber designs has already
demonstrated the capability to achieve single digit SO2 emission rates on low sulfur, and
in some cases high sulfur, coal fired boilers.
■ Additives like dibasic acid can enhance FGD performance in some cases.
■ Reagents with even higher reactivity could be considered, such as hydrogen peroxide.
When used as a polishing step in the FGD process, they might be economically feasible.
■ Upstream SO2 removal might also be considered to reduce the inlet concentration to the
FGD system, and thereby reduce the outlet emission rate by using the excess capacity
available in the scrubber. This might take the form of furnace sorbent injection for a
higher sulfur fuel to remove 30-40% of the SO2 in the optimum higher flue gas temperature
zone present in the boiler.
12239985
Apllicable SO2 Removal Processes
4 APPLICABLE SO2 REMOVAL PROCESSES
T his section contains brief descriptions of the basic absorber designs for the six FGD suppliers
that have a combined 90% share of the U.S. utility FGD market. In addition, this section contains
the developments and innovations that these suppliers are currently using or considering for future
installations to achieve FGD system performance exceeding 99% SO2 removal. Wet LSFO processes
currently dominate the U.S. utility market due primarily to the lower cost of the limestone reagent vs. lime
or other alkali options. This system also generates a solid product that can be landfilled or sold if a local
market exists for the gypsum product. In the process limestone reacts with SO2 gas to form calcium salts.
The chemical reactions are described by the following equations:
SO2 Reaction: CaCO3 (s) + SO2 (g) + 1/2 H2O → CaSO3 • 1/2 H2O + CO2
Sulfite Oxidation: CaSO3 • 1/2 H2O + 1/2 O2 + 3/2 H2O → CaSO4 • 2 H2O
The primary differences between the various suppliers are the absorber vessel designs.
Chiyoda CT-121 Jet Bubbling Reactor (JBR)
Chiyoda Process Description

The CT-121 FGD process was developed more than 20 years ago by the Chiyoda Corporation in
Japan. They have since licensed the technology to engineering and construction contractors throughout
the world. The CT-121 process utilizes the same reaction chemistry as the wet Limestone with Forced
Oxidation (LSFO) process.
The major difference between the CT-121 process and other LSFO FGD systems is the Jet Bubbling
Reactor that serves as the SO2 absorber, oxidizer, and reaction crystallizer in one vessel. Flue gas from
the boiler is cooled using a water quench and then is forced down through a series of sparger tubes
positioned in a tube sheet above the reaction tank and immersed in the limestone/gypsum slurry. The
gas flows down through the tubes and then bubbles upward through the absorber slurry to form a froth
layer. The froth provides a gas-liquid contact zone to promote SO2 dissolution into the reaction tank
slurry. The limestone slurry feed contains 20–30 wt. % solids. The cleaned gas passes through mist
eliminators and then on to the stack.
The solid content in the scrubbing liquid is maintained at 20–30% by wt. solids by drawing off a
small portion of the liquid to the dewatering system. Clarified liquid is recycled to the absorber
for reuse and the solid product can be used for landfill or possible sold as a gypsum by-product.
Figure 4-1 is a pictorial cutaway of the JBR. Single JBR vessels have been installed on units up
to 1000 MW.
The CT-121 process does not use large recycle pumps to feed slurry to the absorber vessel. This
eliminates the maintenance associated with large slurry pumps. However, the power saved by eliminat-
ing the pumps is offset by the power consumed by the flue gas I.D. fans. These fans must overcome
the higher pressure drop across the JBR vessel compared to an open spray tower. This added power
consumption is offset in part by the elimination of the large recycle slurry pump motors that are
required in a typical open spray tower application. The JBR is not as tall as an open spray tower,
although the tower diameter is much larger to accommodate the large reaction tank. The plan area of
the standard spray tower and recycle pump enclosure is similar to the plan area for a JBR. Figure 4-2
is a flow diagram of the CT-121 process using a rectangular absorber vessel that is typical for recent
Japanese installations.
FGD Performance
Figure 4‑1
Chiyoda CT-121 Schematic Flow Diagram (Round Vessel)
Figure 4‑2
Chiyoda CT-121 Schematic Flow Diagram (Rectangular Vessel)
12239985
Chiyoda has stated that the CT-121 process can achieve SO2 removal of more than 99%. They have
some limited experience on very high SO2 inlet concentrations (4000 ppm), and have demonstrated
that achieving 99+% removal on these units is relatively easy. Chiyoda has also provided guarantees
for 99+% removal on medium sulfur units, in particular the Shinko-Kobe Unit #2 with a design sulfur
content of 0.93% S on these two coal fired units (700 MW). At this facility, Chiyoda guarantees
greater than 99% SO2 reduction from an inlet flue gas stream containing 840 ppm SO2. The actual
condition for the performance test was 320 ppm, and the Chiyoda system attained 99.8% SO2 reduction
under those conditions. Chiyoda used their rectangular absorber design at this installation. These
high reduction efficiencies are achieved with a slight increase in pressure drop across the absorber
(added power consumption) and operation at a higher slurry pH.
Recent CT-121 Process/Design Developments

The following indicate some of the recent design modifications and/or enhancements for the
CT-121 Process:
■ The cross section of the tube sheet surface and the vessel has been reduced by 20–30%,
resulting in significant reductions in the plan area occupied by the JBR. The reduction in
the vessel cross section was accomplished by increasing the sparger tube diameter from
4” to 6”. This design upgrade is not offered universally by CT-121 licensees.
■ Pre-scrubbers included in earlier installations have been eliminated and replaced by
quenching sprays at the JBR inlet to both cool the flue gas and ensure that the tube sheet
top surface and the sparger tubes remain wetted during all operating conditions.
■ To reduce water use, the deck wash is done with slurry and filtrate from the dewatering
system to prevent scale formation. The inlet and outlet duct floors are also washed with
slurry to control solids buildup.
■ The recent installations outside the U.S. have used rectangular JBR vessels. The rectangular
absorber allows more shop fabrication and reduces field construction time. At one
Japanese installation, the absorber construction time was reduced from 75 to 40 days
by using prefabricated panel construction. The disadvantage of the rectangular shape is
that it requires heavier steel plate than the round design.
■ Japanese installations use lined carbon steel instead of fiberglass-reinforced plastic (FRP).
Some of the latest designs in the U.S. will use 2205 alloy, while others will use FRP
JBR vessels.
■ By increasing the operating pH of the slurry by 1.0, the latest CT-121 installations have
been able to reduce their power consumption by 15% by reducing the depth of the
sparging tube immersion in the slurry.
■ In order to achieve higher reduction efficiencies, the CT-121 absorber can be run with
a higher pressure drop by increasing the depth of the slurry through which the flue gas
must flow to reach the top of the froth layer.
■ Chiyoda noted that high dissolved chloride concentrations can significantly suppress
the pH in the absorber slurry, potentially reducing the SO2 removal. This effect is due to the
high ionic strength caused by the chloride concentration, which negatively impacts the
transport of hydrogen ions to the surface of the limestone particles. This H+ ion transport
is necessary for the limestone dissolution process. The CT-121 process is relatively
unaffected by the chloride concentration. In a typical spray tower design it would
be necessary to increase the dissolved alkalinity to maintain SO2 removal. This requires
more un-dissolved limestone to be present, increasing the limestone feed rate and potentially
the scaling potential of the slurry.
FGD Performance
Chiyoda has indicated that their project awards in the U.S. utility industry will represent 30% of
the installations scheduled to startup between 2005 and 2020 (based on independent data, Chiyoda
had about one-sixth of the U.S. market share in 2006).
Mitsubishi/Advatech Double Contact Flow Scrubber (DCFS)
Mitsubishi Heavy Industries (MHI) has been providing emissions control systems for power plants for
more than 40 years. The Double Contact Flow Scrubber (DCFS) design is currently offered by MHI
outside the U.S., and under the name Advatech in the North American markets. The DCFS is the
current state-of-the art for the MHI FGD technology and has been installed on power generation plants
for the last 30 years.
Mitsubishi (MHI) Process Description

In low sulfur applications flue gas enters the bottom of the absorber where the absorbent slurry
is sprayed upward through large diameter (1.5”) nozzles creating a dense liquid-filled area. Slurry is
supplied to a single spray header by multiple large recycle pumps. The flue gas contacts the slurry
as it is sprayed out of the nozzle (co-current), and as the liquid falls back down towards the nozzles
(counter-current). Therefore, the flue gas double contacts the limestone slurry.
The slurry solids content in the reaction tank is maintained at 30% by weight, promoting gypsum crystal
growth and reducing scaling potential on the internal surfaces of the vessel. The high level of suspended
solids also provides higher alkalinity in the scrubber slurry to enhance SO2 absorption and make the
system better able to handle changes in the inlet SO2 concentration. The gas flow rate is maintained
between 10–15 fps through the absorber vessel. MHI indicates that the absorber system footprint is
about 30% less than a conventional open spray tower. Availability for the 20+ installations to-date is
reported to be greater than 99%. Removal efficiencies in excess of 99% have been maintained in high
sulfur applications using a single DCFS module.
The scrubbed gas flows upward and out of the absorber while the slurry reaction products are either
recycled into the absorber or exit the absorber reaction tank with the slurry bleed stream that is trans-
ferred to the dewatering process. A process flow diagram for this single tower arrangement is provided
in Figure 4-3.
For larger units and those requiring higher removal efficiencies, as well as some higher sulfur fuel
applications, MHI can add a second tower section so that the double-contact between the slurry and
the flue gas occurs in two towers fed by a common reaction tank. The inlet tower is referred to as the
co-current tower, and the outlet tower as the counter-current tower. One header is used to feed both
spray levels. Figure 4-4 provides a diagram of this dual tower application. The DCFS absorber tower
is approximately 1/3 the height of a typical spray tower, reducing the structural steel and piping
requirements for the system.
The single tower design is used for lower sulfur fuels sized up to 700 MW and for applications
requiring up to 97% removal. For facilities that are larger, or that require higher removal efficiencies,
they offer the two-tower configuration that has demonstrated removal efficiency capabilities of
99+%. One DCFS installation on a 2% oil-fired boiler has consistently been able to reach 99.9%
SO2 reduction.
Recent MHI Process/Design Developments

The following indicate some of the recent design modifications and/or enhancements for the MHI
DCFS Process:
12239985
Figure 4‑3
MHI/Advatech Schematic Flow Diagram Single Tower
Figure 4‑4
MHI/Advatech Schematic Flow Diagram Dual Tower
FGD Performance
■ The DCFS process uses CaCO3 monitors to maintain the correct concentrations of alkali
in the recycle slurry.
■ Chemical oxygen demand (COD)/sulfite monitoring system measures and controls the
Oxidation Reduction Potential (ORP) of the slurry. This monitor maintains absorber slurry
chemistry reducing the potential for mercury reemission (typically caused by reaction of
dissolved ionic mercury with sulfite in the recycled slurry).
■ The ORP monitor also indicates the slurry conditions that could lead to aluminum fluoride
blinding of the limestone particles in the slurry, allowing the operators to take action to
stop the potential loss of reduction efficiency.
■ Another feature included in some recent installations is the addition of air eductors for
introduction of oxidation air into the reaction tank. Eductors are open pipes that are
connected to the recycle slurry feed lines. As the slurry moves past the eductor, the air is
drawn into and intimately mixed with the slurry. The eductors replace the air compressors
of previous designs, reducing maintenance requirements. The power consumption on the
recycle slurry pumps will increase when eductors are being used, partially offsetting the
elimination of the oxidation air compressor motors.
■ Medium sulfur fuel applications have now demonstrated continuous operation at greater
than 99% SO2 removal efficiency. This commercial experience may translate to high
efficiency commercial guarantees in the future. This level of performance can be achieved
using either the single or twin tower design depending upon other operating conditions
at the plant. The DCFS design has achieved 99.9% efficiency with a single tower applied
to a high inlet SO2 flue gas stream (2,670 ppm on 150 MW unit). The outlet concentra-
tion was measured at 3 ppm. This performance is stated to be possible due to the enhanced
liquid-to-gas contact and high reagent activity in the 30 wt.% slurry. Similar performance
was obtained at a second unit operating on a 2,600 ppm gas stream (99.6% SO2 reduction
with design guarantee at 98.3% on 100 MW unit).
The DCFS process has been ordered by a number of U.S. utility companies in the last 2 years. The TVA
units at Gorgas and Paradise will both use the two-tower DCFS configuration. The two-tower design
is reported to have lower pressure drop than the one tower design because the slurry in injected
co-current with the flue gas in the second tower providing push to the gas. The air eductor design has
been proposed for a unit modification in the southwestern U.S. and will be used for the Bull Run instal-
lation of IP&L (900 MW). The demonstrated capability to handle higher sulfur fuels and maintain high
removal efficiency will also be directly applicable to the U.S. power generation industry.
Alstom Power SO2 Removal Processes

Alstom Spray Tower Process Description
The open spray tower was Alstom’s first FGD product offering with the first unit starting up in 1975.
Alstom has sold spray towers for over 37,000 MW of electrical generating capacity at over
50 plants.
The open spray tower absorbers generally range in size from 6 to 20 meters in diameter (20 to 65
feet). These absorbers are constructed with various types of alloys, stainless steels, and mild steel with
various corrosion/erosion resistant linings depending on the specific application.
The absorbers generally have two to four installed spray levels. Each spray level is fed through an
individual riser by a dedicated recycle pump. Each spray level consists of tangential-inlet, hollow cone
spray nozzles comprised of nitride-bonded silicon carbide (SiC). This nozzle type provides the proper
sized droplets for the prescribed SO2 removal. In addition, the hollow cone SiC nozzles are erosion
12239985
resistant. Design, selection and placement of spray nozzles are of critical importance in high-efficiency
wet FGD systems. The spray nozzle configuration is determined by computer modeling in order to
achieve the desired gas to liquid contact and scrubbing efficiency.
One of the innovations added to the original spray tower design are wall rings. Wall rings are radial
baffles that projecting inward from the spray tower wall and redirect any flue gas traveling up the tower
wall back into the spray zone. These baffles have been shown to improve SO2 removal efficiency
by 2–5% or alternatively to allow 10–20% reduction in liquid-to-gas ratio and achieve the same SO2
removal as without baffles.
Alstom Flowpac Process Description

Alstom has recently developed a new scrubber design -- the Flowpac turbulent bed absorber. The
chemical reactions occurring in the Flowpac absorber are essentially identical to those in all other
LSFO systems. The primary difference between this process and standard open spray tower design is
the unique absorber vessel used in the Flowpac system.
An engineer from a Swedish power plant is credited with conceiving the original design of the Flowpac
absorber. The process was first proven in pilot tests. Based on the proof of concept pilot testing, Alstom
went to full-scale commercial system at a 340 MW oil-fired unit located at the Karlshamm power plant
in Sweden. This full scale unit achieves 98% SO2 removal continuously when firing a 3.5 weight %
sulfur oil. It is also reported to achieve 99.5% SO2 removal when acetic acid is added to the slurry and
the plant was burning 2.4% S fuel oil for a 3-month period.
The first Flowpac absorber consisted of an absorber tank with a circumferential sieve tray installed at the
top of the tank. Slurry flows through eight return pipes downward from the outside of the sieve trays to
the bottom of the absorber tank. In the Flowpac design, the flue gas enters the absorber beneath the
sieve tray. Recycled slurry pours over the sieve tray, and the flue gas bubbles up through the slurry forming
a highly turbulent froth layer. The froth layer provides a large surface area of reagent and flue gas. The
gas/liquid contact is similar in some respects to the froth layer produced in the JBR absorber.
The limestone slurry is added to the absorber tank above the sieve tray. As the slurry moves toward
the bottom of the reaction tank, it enters a central core where oxidation air is injected. As the air
rises to the top of the tank, it carries the slurry with it. The oxidation air acts as the lifting method
recycling the slurry from the bottom to the top of the reaction tank, thereby eliminating the need
for slurry recycle pumps. The oxidation air compressors are larger than those used in typical LSFO
spray tower reaction tanks, so the power saved by eliminating the recycle pumps is partially offset
by the larger motors required for the air compressors. The density differences between the sieve
tray slurry and the absorber tank forces the slurry to flow down the return pipes to the bottom of
the absorber tank where it is forced back up by the air and flue gas. Figure 4-5 provides a flow
diagram of the process.
The smaller size of the Flowpac absorber offers an advantage over the typical spray tower design. The
power consumption of the Flowpac absorber vs. an open spray tower will normally be lower than a
spray tower until about 500 MW or a sulfur content of greater than 1.5% sulfur. The spray tower will
typically be twice the height of the Flowpac absorber. Alstom generally uses 2205 alloy or rubber
lining for the absorber vessel. In the area near the internal quench the materials of construction are
usually Hastelloy clad carbon steel.
The pressure drop of the Flowpac absorber is typically in the range of 10–14” w.g., with two-thirds
attributed to the froth layer and one-third to the sieve tray. The gas velocity through the sieve tray
FGD Performance
To Other
Absorber(s)
SO2 Removal Area 10 Demister
Wash Water
To Other
Absorber(s) 4 5 Reheat Mix 8 To
Flowpac Chamber Chimney
3 Absorber
8
Flue Gas from Booster 1 2 Steam Coil 7 Cold
ESP / FF
ID Fans 13
Air Heaters Air Heater Air
From Process 9
26 12 15 16 Oxidation
Water Storage Air 11 Steam
Tank
To Other To Other
To Other Absorber(s) Absorber(s)
Absorber(s)
14 21
From Other From Limestone

Absorber(s) Slurry Tank
22 To Gypsum
Slurry Tank
Figure 4‑5
Alstom Flowpac Process Flow Diagram
is 1.6–1.8 m/sec (5.2–5.9 ft/sec), and the tray is made out of polypropylene. The minimum load
capability of the Flowpac design has been indicated to be 20%. The slurry bleed stream is typically
processed in two stages of hydroclones.
Alstom has recently begun to actively market the Flowpac absorber design around the world, and has
introduced the concept to the U.S. utility market. To-date, there are no Flowpac installations or sales in
the U.S. However, two have been sold overseas – one in Lithuania (3 × 150 MW) and one in Denmark
(1 × 150 MW). The units are scheduled for startup in 2008 and 2009, respectively.
Recent Flowpac and Spray Tower Process/Design Developments

The following items indicate some of the recent design modifications and/or enhancements to the
Alstom Flowpac process:
■ Operating at higher than design alkali feed rates can lead to supersaturated conditions,
which are quickly followed by scale formation due to precipitation onto tank surfaces. Control
systems and instrumentation have improved dramatically over the last decade, reducing
these high pH excursions. Use of dual pH monitors with frequent calibration, feed forward
boiler signals, and proper liquid/gas (L/G) ratios has helped to reduce the scaling rates.
■ Use of conservative design parameters in the areas of oxidation air stoichiometry and
good air distribution has reduced the occurrence of oxidation deficiencies in the latest
designs. This produces a better gypsum product and limits the scaling potential of the
internal surfaces. Reducing the amount of internal surface by installing self-supporting
spray headers also helps to reduce scaling by reducing the sites for solids buildup.
■ Maintaining 18–20% gypsum solids in the recycle slurry provides more surface for crystal
growth in the slurry than is available on the internal surfaces. Nuclear density meters have
demonstrated their ability to tightly control suspended solids density in the recycle slurry.
■ The latest innovation to the Flowpac absorber is the conversion to a rectangular tank
configuration. The principles of air lift and the froth layer above the sieve tray for
gas/slurry contact are still used in the rectangular Flowpac absorber. Further, Alstom has
12239985
now incorporated an integral quench spray into the gas inlet of the absorber so that the
sieve tray will always remain wetted.
■ The Flowpac absorber can reduce the depth of the bubble layer above the sieve tray to
follow the plant load or changes in inlet SO2 concentration. This ability to change the
depth above the sieve tray also provides an inherent capability to achieve higher SO2
reduction efficiencies, with the 99+% capability already possible. This will come with a
higher absorber pressure drop.
■ Alstom has just completed a conceptual design for a 600 MW capacity Flowpac absorber
module. However, the current commercial offering is limited to a 300–400 MW module.
■ The central core of the Flowpac absorber has been reduced to decrease the air required
to move the slurry through the vessel.
Alstom is also considering upgrades to their standard spray tower design to further enhance its performance
capability:
■ 98% reduction has been the high efficiency that Alstom has guaranteed to-date. Some
of these operating Alstom spray tower FGD systems have achieved 99+% removal for
extended periods. Some plants can achieve less than 2–3 ppm without the addition of
any organic acid. The Navajo installation has maintained an outlet emission rate of 0.04
lbs/MMBtu, which is approximately equivalent to 17 ppm SO2.
■ Alstom is using their commercial spray tower design at these high performance installations,
incorporating wall rings and slurry spray patterns to eliminate the potential for flue gas sneakage
past the sprays. Trays or additives should not be required to achieve these high levels of
continuous performance in the future. To maintain constant 99+% reduction, it will probably
be necessary to add another spray level, and increase the L/G to the absorber by 20–25%.
■ There is currently no financial incentive to maintain these low outlet emission levels with
no exceedances. The low outlet emission rates now provide the operating utility with margin
to use for those periods that might exceed their regulated emission rate. The additional
tons removed can also be sold in the allowance markets.
■ Alstom also feels that their Flowpac absorber can maintain these low outlet emission rates.
The level of slurry maintained above the internal sieve tray will be increased to develop a
deeper froth layer, thereby increasing the absorber removal efficiency. Alstom is focusing
their Flowpac design on the rectangular absorber design from this point onward.
■ Alstom is also continuing to enhance the performance of their dry scrubbing system
through an on-going research program. This Flash Dryer Absorber (FDA) may at some
point be able to maintain 99+% removal capability in the future.
■ It was noted that if cost is not a limiting factor, then there is no theoretical limit to the outlet
SO2 concentration that can be maintained. A single absorber could achieve the single
digit outlet concentrations if there was sufficient redundancy in the system components.
One plant was sighted as an example where financial incentives for the operating staff
resulted in the plant achieving its emission targets on a continuous basis. At this facility
the plant staff operate conservatively to over-comply in the event that there is some upset
in the operation of the FGD system.
Babcock & Wilcox – Wet Lime or Limestone Forced Oxidation

B&W Process Description
B&W has supplied FGD systems to the U.S. utility industry for decades. Their original focus was on
the use of lime reagents in a tray tower design. Recent experience includes many installations of LSFO
FGD Performance
systems, again using a combination of trays and an open spray tower. B&W is also developed an
oxidation systems for lime FGD systems, allowing the process to produce a gypsum product instead of
a sulfite product that must be mixed with lime and flyash before disposal in a landfill. The typical B&W
absorber uses a perforated tray and two absorber spray levels. The two spray levels include one under
and one above the tray. Two multiple branch spray headers comprise each of the two levels (each
enters the absorber from opposite sides). Absorber recirculation pumps feed the spray headers.
The hot flue gas enters the absorbers and is cooled to saturation by absorber slurry sprayed from
the spray level under the tray, as well as the slurry falling from the tray. The saturated flue gas
flows through the absorber tray, which distributes the flue gas and slurry from the upper spray level
across the absorber cross-section. As the gas passes through the tray perforations the increased
gas velocity generates energetic bubbling thereby assuring liquid-gas contact. The flue gas then
passes through the upper absorber spray zone, where absorber slurry spraying countercurrent to
the flue gas flow completes sulfur dioxide removal. The gas then exits the vessel through a two
stage mist eliminator. The pH in the absorber liquid is controlled by the addition of fresh lime or
limestone slurry.
Recent B&W Process/Design Developments

The following items indicate some of the recent design modifications and/or enhancements to the
B&W process:
■ Tests conducted on a LSFO FGD system operating at a coal-fired power plant showed
that addition of a second perforated tray and operation at a higher L/G ratio increased
the SO2 removal from 90% to 98% at slightly lower stoichiometry. In this test, the L/G
ratio was increased from 85 to 107 gpm/kACFM. The addition of the second tray and
the higher L/G ratio increased the pressure drop from 3.6 in. H2O to about 9 in. H2O. It
should be noted that at the same L/G ratio the pressure drop of the single tray configuration
was 3.6 in. H2O and dual tray configuration was 5.3 in. H2O.
■ In a second test at the same installation, addition of the second perforated tray and
operation at a higher pH increased the SO2 removal from 90% to 98% (two trays total,
as in the first test discussed above). In this test, the pH in the first run was 5.6 and in the
second run it was 6.1.
■ From an operational standpoint, if a pump is lost reducing the L/G by 25%, then the
removal efficiency could drop by 8% or more. In addition, if the upstream particulate
collection system performance deteriorated, the ash contamination of the FGD slurry
could result in a drop in slurry pH and major reductions in FGD performance.
■ Magnesium enhanced lime (MEL) can also achieve the same SO2 removal efficiencies
at a much lower L/G ratios. Based on medium sulfur Appalachian coal, the B&W LSFO
system was able to achieve 98% removal at an L/G ratio of 130, whereas MEL was able
to achieve 98% removal at an L/G ratio of 40. When using the lime reagent, further
increases in the recycle rate could lead to a system capable of 99+% SO2 reduction.
There are 30 MEL installations in the U.S. ranging in size from 80 MW to 1400 MW. The new
MEL oxidized process is used at 7 of the units. This process allows for the SO2 reaction products to
be oxidized from calcium sulfite to calcium sulfate. Absorber slurry density is controlled by the addition of
the mist eliminator and filtrate water. The absorber is designed to operate at a slurry density equivalent
to 15 percent suspended solids. The liquid level of each absorber reaction tank is controlled by bleeding
gypsum slurry to the gypsum dewatering system.
4-10 FGD Performance Capability: High Efficiency Design and Operating Options
12239985
Babcock Power
Babcock Power Process Description
Babcock Power relies on an open spray tower design using bi-directional sprays (slurry sprays upward
and downward with respect to the flue gas flow). The tower has wall baffles to mitigate the chance of
internal flue gas bypass. The geometry of each spray level varies and the direction of the spray from
the spray nozzles is also designed to mitigate the chance of internal flue gas bypass. This LSFO process
has the same process chemistry as the other LSFO processes. Babcock Power markets the wet FGD
technology developed in part by the Fisia Babcock group in Germany. Babcock Power also licenses
the dry scrubbing technology under development by Austrian Energy and Environment group (AE&E),
called Turbosorb. The capabilities of this dry scrubbing technology are described later in this section.
Recent Babcock Power Process/Design Developments

The following are recent design modifications and/or enhancements to the Babcock Power wet FGD
process technology:
■ The absorber inlet is optimized to provide the best possible gas flow distribution – they
no longer use a tangential gas inlet and the gas velocity through the scrubber has been
increased by more than 20% to reduce the capital cost while improving the gas-liquid
contact efficiency. This is offset by an increase in the pressure drop across the absorber.
■ The system uses only a moderate solids density in the scrubbing slurry.
■ Babcock Power employs CFD modeling to optimize the design of their FGD absorbers.
This modeling resulted in changes to the location of and the number of nozzles per spray
levels in their current wet absorber designs. The nozzle design has also been optimized,
using dual-direction nozzles that can spray both co-current and counter-current to the
direction of the gas flow through the absorber vessel. These nozzles produce small
droplets that increase the number of droplet collisions, producing additional spray pattern
coverage. A staggered spray pattern is also used.
■ Nozzles are installed along the walls of the absorber vessel to spray countercurrent to the
gas flow and minimize the potential for gas sneakage (wall effects) past the spray pattern.
High spray flows along the absorber walls help to increase overall collection efficiency.
Wall rings are also used to reduce the potential for gas slippage past the spray zones.
■ Their installation list of existing and contracted high efficiency scrubber systems was
previously shown in Table 3-1.
■ Babcock Power also offers a dual loop design for very high efficiency or high sulfur
applications while reducing the overall power consumption.
Babcock Power stated that in order to achieve the 99+% reduction efficiency, it is likely that it could be
met by the addition of another spray level with its associated recycle pump. The L/G increase would
correspond to the increase in pumping capacity. No internal packing, trays or additives should be
necessary. Babcock is currently testing these design modifications to confirm single digit performance
capability, to enable them to offer this level of performance commercially in the future.
As was previously mentioned, Babcock Power also offers a dry circulating fluid bed FGD technology
through their agreement with AE&E. This system has been installed commercially on multiple coal-fired
boilers in Europe, and on one unit here in the U.S. located in New York State. These systems are
typically limited to 95% SO2 reduction, but short term demonstration tests have shown the process
capability to achieve 99% removal. However, these higher efficiencies come with a significant lime
reagent usage penalty.
FGD Performance
12239985 Capability: High Efficiency Design and Operating Options 4-11
Figure 4-6
Siemins-Wheelabrator Generic Wet Scrubbing Vessel
Siemens-Wheelabrator – Wet Limestone Forced Oxidation Process

The Siemens-Wheelabrator (Siemens) FGD absorber design comes in various configurations, ranging
from a typical open spray tower to the use of a sieve tray and spray headers. They have supplied
systems based on lime and limestone reagents, and provided recent installations with performance
guarantees up to 98% achieving outlet SO2 emission rates of 20–25 ppm. A diagram of their generic
wet scrubbing tower is provided in Figure 4-6.
Siemens did not provide any additional input regarding their future plans to meet the high efficiency
FGD market at this time.
Hitachi America
Hitachi Process Description
Hitachi America is currently installing FGD systems at multiple U.S. utility installations. Hitachi
relies on an open spray tower design. The tower has wall baffles to mitigate the chance of internal
flue gas bypass. The spray levels and spray nozzles are designed to minimize flue gas bypass
of the spray zone. This LSFO process utilizes the same process chemistry as the other LSFO
processes. Hitachi America markets the FGD technology developed in Japan and is supported
in the FGD design research by the Hitachi group in Japan. Some of their newest installations in
the U.S. are being designed and guaranteed to achieve less than 10 ppm SO2 emission rates is
some cases. Their design capability to achieve these performance guarantees was demonstrated
at their 1050 MW installation at Tachibanawan 2 in Japan. This installation has the following
design features:
12239985
■ Compact circular spray tower absorber design with high spray density, high slurry
solids concentrations, and high velocity mist eliminators installed in the horizontal
outlet duct
■ Decreased number of spray levels with increased number of spray nozzles, resulting in
more spray pattern overlap
■ Higher gas velocity through the absorber – to 16 fps
■ An online CaCO3 analyzer
■ Reduced flue gas inlet temperature with a forced recirculation gas-gas heater to provide
outlet gas reheat.
■ Tachibanawan 2 achieves 10 ppm SO2 at the stack from the 420 ppm inlet concentration.
The guarantee for the unit is less than 50 ppm at an inlet of 820 ppm. The unit began opera-
tion in 2000. A diagram of the Hitachi absorber design is provided in Figure 4-7.
Recent Hitachi Process/Design Developments

The following are recent design modifications and/or enhancements to the Hitachi FGD system:
■ A split tower design is under development to increase the absorber reduction efficiency.
The inlet gas passes up through one half of the absorber, contacting the slurry spray
countercurrent to the spray downward. The flue gas then flows flow over the internal wall
in the vessel and passes down through a second set of spray in the cocurrent direction
(with the direction of spray.
■ A pilot installation is testing the use of a low pressure drop venturi operating upstream of
the FGD absorber to capture Hg, SO3, HF and HCl. Demonstrating 99% SO2 reduction
on a high sulfur coal application. This will reduce the impact of the chloride buildup in
the scrubber solution that can inhibit SO2 reduction efficiency.
■ A 97% scrubber design should be able to achieve single digit SO2 outlet emissions by
turning on the spare spray level, equivalent to a 20+% increase in the recycle rate.
■ The Hitachi installation list of existing and contracted high efficiency scrubber systems
was previously shown in Table 3-1.
Alternate FGD Technologies for High SO2 Removal Efficiency

Other methods have been proposed to increase SO2 reduction efficiency across the FGD system. Some
of these options are discussed in the following material.
Dry Injection of Trona

O’Brien and Gere, using trona supplied by FMC and Solvay, offers the trona injection technology for
both SO2 and SO3 removal from coal-fired flue gas. Although these dry injection systems are not neces-
sarily capable of achieving the 99+% reduction efficiency targeted in this analysis, they could provide
some trim to the SO2 inlet concentrations entering the wet FGD systems, thereby reducing the load on
the scrubber and possibly its outlet emissions concentration. Reduction rates of 40–90% have been
demonstrated in demonstration tests, with the most economical reduction efficiency falling in a range
of 50–60%. Recent advancements include the use of milled trona down to 15 micron particle size to
improve reagent utilization rates, injection at higher gas temperatures in front of the air preheater may
provide the best performance.
Dynawave/Reverse Jet Scrubber

Monsanto EnviroChem (now MECS) offers this technology for electric utility applications. This technology
has seen most development and application for removal of SO2 from sulfur recovery units (SRUs)
operating in refineries, achieving 99.8% reductions. However, the inlet concentrations of SO2 for these
FGD Performance
Figure 4-7
Hitachi FGD Absorber
units are very high, and the technology has not been applied to a coal-fired boiler to-date. Pilot testing
has demonstrated the system capability to achieve less than 6 ppm SO2. The process relies on a reverse
jet scrubber design
SO2 is captured by contacting the flue gas stream with an alkaline reagent in a Froth Zone region of
intense turbulence with substantial back-mixing and a high rate of liquid surface renewal. This froth
zone is generated in an inlet duct to the absorber where slurry is injected into the gas pass counter-
current to the gas flow. These fountains provide intimate gas-liquid contact. The momentum of the
gas causes the liquid to spread out and form a standing wave at the point where the liquid and gas
momentums are balanced. Turbulent films of liquid occlude and envelop the particulate or gaseous
contaminants, while also providing the gas-liquid contacting necessary for gas quenching and acid
gas absorption.
Until 1987, this technology was confidential and applied only within DuPont. In 1987, MECS realized
the applicability of the Froth scrubbing technology for FGD applications, and entered into a licensing
agreement with DuPont. Since then, MECS has built more than 300 DynaWave® systems for various
industries around the world.
12239985
FGD System Cost Escalation
5 FGD SYSTEM COST ESCALATION
W orldwide demand for industrial facilities has resulted in substantial increases in the costs
of commodities, equipment and construction labor. These increases have led to significant
spikes in installed plant costs in both the industrial sector and in the power generation construction
markets. These same market conditions have contributed to significant cost increases for FGD
systems over the last 3–5 years above and beyond the increased costs associated with higher
levels of SO2 reduction performance. Another factor is the number of installations that must be
installed in a 5 year period to meet the regulatory requirements. This has overloaded many of
the major FGD system suppliers and their subsystem equipment suppliers, resulting in additional
“market-demand” escalation.
The following subsections discuss the worldwide market conditions associated with increased demand
for large capital projects. If further reductions in outlet SO2 emissions continue, then these modifications
for higher removal efficiencies (99%+) are likely to result in additional increases in wet FGD capital
costs for the U.S. power generation industry.
Background
U.S. utilities are installing FGD and SCR systems at a very fast pace to meet the emission requirements
of the Clean Air Interstate Rule (CAIR) and other visibility impact/Best Available Retrofit Technology
(BART) rulings by state agencies. Over the period from 2007 to 2010, about 180 FGD units are
projected to enter commercial service in the U.S. It is also projected that an additional 200,000
MW of coal-fired capacity will be retrofit with FGD systems by 2015 to comply with the second tier
requirements of the CAIR. Many U.S. utilities have already installed FGD systems to comply with the
requirements of the Clean Air Act and the new CAIR requirements. The timing of the new installation
on-line dates is driven in part by first tier requirements of CAIR, which require that most systems begin
operation by 2010, or the utilities will be faced with the purchase of SO2 allowances from a potentially
volatile allowance market.
As part of the CAIR regulations, SO2 allowances generated after 2010 will be worth half as much
as those generated before 2010 (an allowance generated in 2010 and after will be defined as ½ ton
instead of one ton). Then in 2015, an allowance will be defined as 1/3 ton instead of one ton.
Thus after 2015, a utility would need three times as many allowances to emit the same amount of
SO2 as it did in 2009. As a result, those FGD systems that begin operation before 2010 will be
generating SO2 allowances at a much faster pace, so there is a significant financial incentive to
be operational as soon as possible. This further aggravated the demand for FGD systems in a very
narrow timeframe.
CAIR was not the only driver for the FGD retrofits. Many coal-fired units must comply with the Clean Air
Act (through New Source Review), consent decrees, or the Clean Air Visibility Rule. Operators of these
units have or will have to commit to installing FGD systems that meet the regulatory requirements of
best available control technology (BACT) or best available retrofit technology (BART) depending upon
their specific site requirements.
The dramatic increase in the U.S. demand resulted in many FGD OEMs operating at their maximum
capacity given their current staffing levels. Staff expansions have occurred to meet the increasing
demand, but some, if not all, FGD suppliers have found it necessary to extend their delivery
schedules significantly because of their backlog. Due again to the demand, lead times for major
FGD Performance
FGD components have also increased, leading to additional pressure on the project schedule.
Items such as rubber-lined pumps, ball mills, and booster fans (or replacement ID fans) have further
extended the FGD system construction schedule. Over the past three years, the lead times for these
items from the major suppliers have quadrupled, doubled, and increased by 20%, respectively.
As an example, large rubber-lined pumps in early 2007 had lead times of about 120 weeks. In
some cases, OEMs are placing orders for components and commodities before signing contracts
just to get in the cue for deliveries to support upcoming projects, thereby minimizing construction
schedule impacts.
In addition to the U.S., a number of overseas countries have seen dramatic growth in their economies,
which has translated to rapid increases in the number of large capital projects (including coal-fired
power plants). The price of oil and gas has resulted in major investments in production facilities.
Mining projects have increased in number around the world due to the increase in commodity
prices. The U.S. infrastructure is aging, resulting in major expenditures on roads, bridges, etc. It is
projected that China will install $5 billion in FGD systems at its coal-fired generating stations within
the next 10 years. The industrial project demand in the U.S. combined with the demand in China
and India will continue to cause significant increases in material and equipment cost escalation in
the near-term.
Of the overseas countries, China has experienced the most substantial growth related to demand for
such items as:
■ Equipment, steel, concrete, and other bulk materials for large industrial, power generation,
environmental control, and infrastructure projects;
■ Building materials, electrical wiring, and concrete for commercial projects;
■ Wiring, plumbing, and concrete for residential housing; and
■ Lumber for residential housing and furniture exports.
The scale of construction of coal-fired electric generating stations provides a “picture” of the magnitude
of China’s growth. Currently, China is building the equivalent of two-500 MW coal-fired electric
generating units per week, which is comparable to building the capacity of the entire United Kingdom
power grid each year (McRae testimony, 2007). This level of power plant construction represents an
enormous demand for steel, rotating equipment, electric wiring, other electrical components, and
concrete. It also results in fierce competition for shop space at steel fabricators and equipment suppliers,
which is reflected in the extended lead times. Further, this situation translates to significant demand for
the raw materials to make the products provided by steel mills, equipment manufacturers, and ready-mix
concrete companies.
Added to all the above is the “mini-resurgence” in construction of new coal-fired units in the U.S. As
of May 2007, the National Energy Technology Laboratory (NETL) was tracking a total of about 150
coal-based units totaling about 90,000 MW. NETL data indicates that about 50,000 MW of coal-based
units are either under construction or planned for operation by 2014. The majority of the units currently
under construction are PC-fired and most are being installed at existing sites. Many of these sites were
originally approved for the addition of future units.
With public and political concerns regarding global warming, a growing number of the 150 coal-fired
units have been delayed or cancelled since the compilation of the May 2007 list. Although the final
number of U.S. power generation units actually installed could be well less than the NETL projections
due to the GHG emissions issue and public perception, worldwide demand will still continue to put
12239985
upward pressure on the cost of power plants and power plant subsystems. Further, U.S. regulatory
requirements and worldwide growth of coal-fired plants will keep the demand for FGD systems for utility
applications very strong for at least the next 5–10 years.
Increases in Material and Equipment Prices

In order to quantify the contributing factors to the increases in FGD system costs, comparisons of
escalation were made for two dates: January 2004 and Mid-2007. January 2004 was selected as the
approximate beginning of the period when the costs of commodities (bulk materials) and equipment
began to escalate at rates much higher than in the years prior to 2004.
U.S. Producer Price Indexes from the Bureau of Labor Statistics (BLS) provide quantitative comparisons
of cost escalation for examples of bulk materials and equipment used in FGD systems (Producer Price
Index = PPI). Figure 5-1 through Figure 5-5 show cost indexes for representative items used in FGD
systems: concrete, structural steel, electrical wiring, large centrifugal pumps, and large centrifugal fans.
Although January 2004 to mid-2007 is the period when FGD system costs have increased substan-
tially, the graphs cover the period from 1996 to 2007 to give historical context to periods before and
after the significant worldwide growth of large capital projects began to occur. These graphs show
the cost indexes and the corresponding annual compound escalation rates for the period from 1996
through 2003 and the period from 2004 to mid-2007.
Figure 5-6 provides escalation rates for wet scrubbers based on the Vatavuk index. This cost index
reflects the influence of the items shown in Figures 5-1 through 5-5, as well as many other compo-
nents included in FGD systems. The escalation trend for the FGD system is similar to the trends reflected
in the five examples except that the FGD system cost index begins to climb about two years earlier than
the indexes shown in Figures 5-1 through 5-5.
The Vatavuk Index curve does not appear to take into account the actual costs incurred by U.S. utilities
in the current market. Figure 5-7 provides a summary of reported costs for FGD systems over the last
few years. This curve makes it very clear that other market pressures are acting on the U.S. market, driving
the costs well beyond typical inflationary impacts, which appears to be due to the market-demand
factor previously discussed.
Cost Considerations Related to Craft Labor Workforce
The installation of the FGD systems not only includes the capital cost of commodities and equipment,
but also the costs associated with engineering staff and construction labor. The systems must be
designed, transported to the site, and installed within the time frames noted above. This requires
that there are sufficient experienced engineering personnel to design the systems and sufficient
experienced craft labor workers to construct the FGD systems and other capital projects in
the U.S.
The discussion in this section identifies the labor issues related to the near simultaneous construction
effort of multiple projects that will be required over the next 3 years. The sufficiency of the craft
labor workforce to meet the staffing of concurrent or overlapping construction projects is very
important. However, it is as important that the craft workforce has sufficient experience and skills
to produce quality workmanship and meet the productivity levels designated for the projects. An
overriding situation is the impending retirements of experienced craft workers and the associated
knowledge “drain”.
FGD Performance
5-
Cost Index for Ready-Mix Concrete
Producer Price Index (Source: Bureau of Labor Statistics)
(Curve from Jan 1996 to Jan 2003 Not Shown)
220.0
Compound Annual Esc. from 2004 to 2007.5 = 8.8%

200.0
FGD Performance
12239985
180.0
160.0
140.0
120.0 Compound Annual Esc. from 1996 to 2003 = 2.0%
Ready Mix Concrete (1/1982 = 100)

100.0
May-02 Dec-02 Jun-03 Jan-04 Aug-04 Feb-05 Sep-05 Mar-06 Oct-06 Apr-07 Nov-07
Month-Year
Figure 5-1

Cost Index for Ready-Mix Concrete
Cost Index for Structural Steel
12239985
220.0
FGD Performance
200.0
180.0
160.0
140.0
Structural Steel (Jan 1982 = 100)

100.0

Month-Year
Figure 5-2
Cost Index for Structural Steel
5-
5-
Cost Index for Copper Wire & Cable
400.0
350.0 Compound Annual Esc. from 2004 to 2007.5 = 25.5%
FGD Performance
12239985
300.0
250.0
200.0
150.0
Copper Wire & Cable (Dec 1986 = 100)

Compound Annual Esc. from 1996 to 2003 = -2.2%
100.0
Month-Year
Figure 5-3

Cost Index for Copper Wire
Cost Index for Industrial Pumps
12239985
220.0
FGD Performance
200.0
180.0
140.0
Pump Index (Jan 1982 = 100)

120.0
100.0

Month-Year
Figure 5-4
Cost Index for Large Centrifugal Pumps
5-
5-
Cost Index for Industrial Fans
220.0
200.0 Compound Annual Esc. from 2004 to 2007.5 = 4.5%
FGD Performance
12239985
180.0
160.0
Fan Index (Dec 1983 = 100)

120.0
100.0
Month-Year
Figure 5-5

Cost Index for Large Centrifugal Fans
Cost Index for Wet Scrubbers
Vatavuk Index (Source: Chemical Engineering)
220.0
12239985
210.0
FGD Performance
200.0
190.0
180.0 Compound Esc. From 2004 to 2007 = 9.8% per yr
170.0
160.0
150.0
140.0
130.0 Compound Esc. From 1997 to 2003 = 1.6% per yr
Wet Scrubbers (1994 = 100)

120.0
110.0
100.0
1996 1998 2000 2002 2004 2006 2008

Year
Figure 5-6
Cost Index for Wet Scrubber
5-
Published Wet FGD System Actual Installed Costs

Figure 5-7
12239985
The material below is a compilation of information provided by construction executives and publications
from the Construction Labor Research Council. It describes the evolution of the craft labor workforce
over the last 30 years, as well as the impacts of the multitude of projects and the impending retirement
of a large percentage of the craft workers.
■ It has been predicted that there will be shortages of skilled workers in selected regions of
the U.S. There are already shortages of certain union trades in parts of the country where
there are multiple large projects within 100–200 mile proximity,
■ With regard to CAIR, the highest potential for shortages are predicted be the boilermaker
and pipefitter trades in Illinois, Indiana, Missouri, and Ohio,
■ The high price of oil has contributed to a significant growth in planning and project
awards in the oil, gas, and chemical industries. A number of these projects will occur
within the same timeframe as FGD retrofits and compete for non-union and union crafts,
■ The average age of today’s construction workers is 47,
■ Over the next seven years, the number of workers age 55 and over will increase four
times the overall growth rate of the rest of the labor force,
■ The percentage of workers 50 and older by craft is shown in Table 5-1.
Table 5-1
Percentage of Workers Age 50 and Older (as of 2006)
Craft Job % Age 50 and Older
Boilermakers 28
Carpenters 16
Cement Masons 14
Equipment Operators 21
Electricians 17
Ironworkers 16
Laborers 15
Millwrights 12
Painters 16
Pipefitters 22
Sheet Metal Workers 13
■ The rate of growth of people entering the construction labor workforce through 2015 is
projected to be the same as it was from 1995 to 2005 – 1.1% per year. This is the slowest
rate of growth since the 1960s,
■ Projections from a survey of 50 large utility and chemical companies indicate that in the
Southeast U.S. overall demand for craft labor will increase by 25% in 2007 and 2008.
Over the same period, demand for ironworkers, pipefitters, insulators, and painters will
increase by more than 40%.
The large demand for craft workers is contributing to the increase in the labor cost associated with heavy
construction projects (including FGD retrofits). Prior to 1999, craft workers’ year-to-year wage increases
were in the range of 2.5% to 3.5%. Since 1999, wage increases have stayed steadily above 4%.
In 2006, the nationwide wage increases for open-shop crafts increased 4.6%. In addition, wage and
benefit settlements in 2007 have resulted in an annual increase in construction labor costs of 4.7%.
FGD Performance
Adequacy of Engineering Staff
It is important that FGD projects are executed by firms with experienced engineering staff. The knowledge
of experienced personnel will improve the quality of the final product and reduce the chances of schedule
overrun. As previously indicated, about 180 FGD units have been awarded and/or will need to be
completed within the 2007 to 2010 timeframe,
■ If it is assumed that there would be two units per plant, then there would be 90 projects
to be completed within the 2007 to 2010 timeframe,
■ Five major engineering companies have received about 75% of the FGD project awards.
This means that the five companies would have to complete about 65 projects,
■ If it is assumed that the awards are equally distributed, then each engineering company
would have 12 FGD projects that must be designed and be operating by 2010,
■ In addition to FGD, about 120 SCR units (60 projects) have been or will be awarded and
must be designed and operating by 2009,
■ If the same values for number of projects, engineering companies, and market share are
applied to SCR, then around 45 SCR projects have or will have to be designed by the
5 companies,
■ Assuming equal market share as above, then each engineering company would have 9
SCR projects to be completed within the 2007 to 2009 timeframe,
■ Using the assumptions above, the combination for each company would be 12 FGD and
9 SCR projects all of which will have to be completed by 2010. When combined with
other major capital projects, engineering firms are and will remain challenged to execute
all this work within the required time constraints,
The average age of engineers as a whole is 51 and about half will retire by 2015. It is projected that
there will be a 10% shortfall of engineers by the year 2012.
The limited supply of qualified engineers has resulted in increased hiring competition between
engineering companies, higher engineering personnel turnover, and increased salaries. This situation
adds to the cost of engineering industrial facilities or FGD retrofits, other emission control system retrofits,
and new coal-based power plants.
Increased Cost of FGD Systems and Impact of Higher % Removal

In the last three years, the costs of FGD systems have increased substantially due to the expanded
demand in the U.S. and overseas. The previous discussion on escalation of commodities and equipment,
craft labor, and engineering show how these increases have influenced the significant increase in cost
escalation of large capital projects including FGD retrofits.
Table 3-1 shows eight systems that will be installed by late 2008 and are designed to achieve SO2
emissions of 10 ppmv or less. These units will incorporate the latest improvements into the absorber
designs and operating parameters. It is not clear whether the SO2 emissions stated in Table 3-1 are
for the performance test or for longer-term performance. However, most FGD installation contracts
to-date have performance guarantee requirements that are met by successful completion of the
performance test to demonstrate FGD system removal capability. Long term operation is not considered
except from the perspective of operating reliability. To compete with IGCC SO2 emissions, PC
plant FGD systems will need to achieve an emission of 10 ppmv SO2 or less on a multi-day rolling
average basis.
12239985
The 2010 CAIR compliance date and the timing of other regulatory compliance legislation will result in
U.S. FGD market sales that peak in 2010 or 2011. Unless local or other regulations require PC plants
to achieve even lower SO2 emissions after 2010, there appears to be a very limited window during
which sales will support expenditure of R&D funds needed to achieve very high FGD efficiency. In the
absence of even more stringent SO2 emissions regulations or pressures from public utility commissions:
■ Suppliers will probably be reluctant to invest in R&D programs to decrease SO2 emissions
to the range of 10 ppmv knowing that their U.S. FGD sales will decline significantly after
2010 or 2011.
■ Unless mandated by additional regulations, utilities may be reluctant to pay a premium
for FGD systems capable of consistently achieving SO2 emissions in the range of 10 ppmv.
As indicated above, utilities are already faced with significant increases in the capital
costs of FGD systems.
■ Projections indicate cost escalation of commodities and equipment for FGD systems will
be high for at least the next few years.
■ In addition to higher capital costs, consistently achieving very low SO2 emissions will
require that utilities incur higher fixed and variable operating costs. It will be necessary
for utilities to change their operating philosophy and to augment operating staff with
full-time engineers whose mission will be consistent vigilance on FGD chemistry and
operating conditions. The higher efficiency requirements will result in higher variable
operating costs for chemicals, waste disposal and power consumption due to increased
recycle rates and absorber pressure drops.
FGD Performance
12239985
Multi-Pollutant Technologies
6 MULTI-POLLUTANT TECHNOLOGIES
T he following material provides descriptions of multi-pollutant processes that remove SO2, NOX, and
mercury (Hg). These processes have the potential to remove high levels of SO2 and consequently
are included in this report. Some of the processes have commercial-scale experience overseas, but
most are only at the demonstration stage. These processes are of interest because of the benefit of
being able to remove several pollutants, but it may be difficult for utilities to justify selection of one
of these processes until they gain more extensive commercial status.
Powerspan Electrocatalytic Oxidation Process (ECO™ & ECO2™)

Powerspan Process Description
Powerspan’s ECO™ process is designed for installation downstream of an existing particulate collection
device. The process utilizes a dielectric barrier discharge reactor to oxidize pollutants in the flue gas
stream, followed by a wet ammonia scrubber and wet ESP for removal of oxidized pollutants and
byproducts. The technology produces a saleable ammonium sulfate fertilizer byproduct. A process flow
diagram is shown in Figure 6-1.
Figure 6-1
ECO™ Process Flow Diagram
Flue gas flows from the existing ESP or fabric filter into an electrocatalytic oxidation (ECO) reactor,
where hydroxyl (OH) radicals and atomic oxygen (O) are formed by the reaction of high-energy electrons
with water and oxygen molecules. These radicals then react with pollutants in the flue gas, oxidizing
mercury (Hg) to mercuric oxide (HgO), and a portion of the SO2 and NOX in the flue gas to sulfuric
acid (H2SO4) and nitric acid (HNO3). Approximately 90% of NO in the flue gas is oxidized to NO2
and HNO3 at this stage. Less than 10% of SO2 in the gas is oxidized to form SO3, which ultimately
forms sulfuric acid as the flue gas passes through the wet scrubber.
FGD Performance
The oxidized flue gas is then sent to a double loop ammonia scrubber where the final products
are removed from the flue gas stream. In the upper loop of the scrubber, ammonia is added to form
ammonium sulfate by reaction with sulfuric acid. Partially oxidized products are oxidized by NO2 to
form ammonium sulfate and N2. The lower loop of the scrubber cools and saturates the flue gas concentrating
the byproducts for removal. Oxidation air is added to the lower loop to further oxidize sulfites.
The flue gas then enters a mist eliminator that removes entrained droplets from the gas prior to entering
a wet ESP (WESP) located in the top portion of the scrubber. The WESP collects acid aerosols, particulate
matter, and mercuric oxide, while also collecting and recycling excess ammonia as an aqueous solution.
The aqueous solution of ammonium sulfate byproduct exits the scrubber in a bleed stream that is treated
by filtration to remove ash and insoluble metal compounds. The stream is then passed through an
activated carbon absorption bed where mercury compounds are absorbed onto the bed. The purified
byproduct stream is sent to a crystallizer where well-defined sulfate crystals are formed by evaporation.
The crystals are dried and granulated to form a fertilizer byproduct that could be marketable depending
on quality and location. At sites where the fertilizer byproduct could be sold, landfill disposal costs
would be avoided.
The ECO™ technology has been pilot tested on a 1-MW equivalent slipstream of FirstEnergy’s R.E.
Burger Plant, located in southern Ohio. The pilot system operated for more than four years. Testing on
this system has successfully demonstrated an average NOX removal efficiency of 90% with an average
inlet level of 250 ppm, and SO2 removal efficiency of 98% with an average inlet concentration of
1320 ppm.
Following pilot testing, Powerspan built a 50 MW commercial demonstration unit at that same location.
Emission results from the commercial scale demonstration have shown similar performance to the pilot
testing, achieving 98% SO2 removal, 90% NOX removal, 85% Hg removal, and less than 0.01 lb/
MMBtu emissions of fine particulate. Higher SO2 reduction efficiencies appear to be possible given the
results of short term testing.
ECO2™ Process -- In conjunction with the ECO™ technology, Powerspan is developing a method
to absorb CO2 from power plant flue gases. Called ECO2™, the process utilizes aqua ammonia to
absorb CO2 from the flue gas in addition to the SO2, NOX, and Hg removed by the ECO™ technology.
The CO2 is later released for sequestration, while the ammonia is recycled back to the system to
minimize operating costs.
Powerspan and the DOE’s National Energy Technology Laboratory (NETL) are working under a three-year
cooperative research and development agreement (CRADA) to conduct tests on ECO2™. Initial results
indicate up to 90% CO2 removal associated with the injection of aqua ammonia. Current plans are
to eventually run a pilot test on a power plant near the end of the CRADA, at which point the process
design and cost estimates will be finalized. Preliminary cost estimates predict that ECO2™ will cost less
than half of the current CO2 removal technologies that are commercially in use.
CANSOLV™ Process
The CANSOLV system expands on the regenerable amine SO2 removal process by including CO2
removal in an integrated tower configuration. NOX and Hg removal capabilities, previously marketed,
have been discontinued from CANSOLV’s product offerings. CANSOLV is focusing on marketing the
SO2 and CO2 process configuration as described below.
Flue gas enters a pre-scrubber section of the tower downstream of the ESP or fabric filter where it is
quenched to saturation before entering the SO2 scrubbing section. SO2 removal is achieved in this
12239985
section using a regenerable di-amine reagent solution. In the CANSOLV process, one of the amine
products is very basic (higher pH), and remains a salt throughout the process (see reaction 1). The
nature of the absorbent, present as a salt in solution minimizes vapor phase reagent losses. The other
amine functionality acts as a buffer, reacting with hydrogen ions in solution to form ammonium salts
and promoting the dissolution of SO2 (reaction 2). If reaction 1 is the anion of a strong acid, it can
neutralize the absorbing nitrogen functionality if concentrations become significant. For this reason, a
slipstream of solution is sent to an electro dialysis purification unit to keep the level of these ‘heat stable
salts’ (HSS) below the critical value.
(1) R1R2N–R3NR4R5 + HX ↔ R1R2NH+–R3NH+R4R5 + X-
(2) R1R2NH+–R3NR4R5 + SO2 + H2O ↔ R1R2NH+–R3NH+R4R5 + HSO3-
The rich amine (right-hand-side of reaction 2) is steam regenerated, resulting in a lean amine stream
back to the process and a concentrated, saturated SO2 stream. SO2 can be dried and distributed as
liquid SO2, converted to sulfuric acid, reduced to sulfur, or converted to another sulfite or sulfate byproduct.
If no method is preferred due to site-specific conditions, CANSOLV recommends the sulfuric acid option
as it has a non-seasonal market and has low operating costs. SO2 removal of up to 99% (or down to 10
ppmv SO2) can be achieved from this process. A process flow is shown for this system in Figure 6-2.
Flue Gas
To Stack
Water
Wash
SO2 to Sulfuric
Acid Production
Electrolysis Salt
Removal
SO 2
Absorption
SO2
Stripping
Column Steam
Condensate
Pre -
scrubber SO2 Lean -Rich
Existing ESP /FF Exchanger
Figure 6-2
CANSOLV Simplified Process Flow Diagram
The CANSOLV System was initially developed as a SO2 scrubbing technology for industrial
gas streams. CANSOLV’s SO2 system is currently operational at seven different commercial instal-
lations, with two more scheduled for startup at the end of 2007. Of the nine installations, four
are in the U.S., two in Canada, and the others overseas. All of the installations operating or
under construction are either at smelters or petrochemical plants. The inlet SO2 concentrations at
these installations range from 800 ppmv to 11 volume % sulfur. The largest of these installations
removes ~97% SO2 from an FCCU unit flue gas stream equivalent to 175 MW. At an acid gas
installation, SO2 is reduced from 3,000 ppmv to 15 ppmv in the outlet flue gas (99.5 % removal).
Three of the installations have been operating since 2002 and four have been operating since
2005/2006.
FGD Performance
CANSOLV’s SO2/NOX and CO2 technologies were tested at pilot-scale at the Saskatchewan Power
Poplar River station in 2006. These tests showed positive results for the CO2 system, and less interesting
results for the NOX and Hg process. As a result, CANSOLV has currently discontinued NOX control
from its commercial offerings, but is actively pursuing CO2 control.
CANSOLV has installed commercial facilities for their SO2 removal process in Washington, California,
Delaware, Canada, Belgium and India. They are currently competing for a large-scale CO2 capture system in
Norway (decision to be made in 2008). To better address the U.S. market, CANSOLV has teamed with
Massey Energy and HG Engineering to form CANSOLV LLC to market their SO2 and CO2 process.
Airborne Process
The Airborne Process™ uses a sodium bicarbonate scrubbing technology combined with post-scrubbing
oxidants to remove sulfur dioxide (SO2), sulfur trioxide (SO3), nitrous oxide (NOX), and mercury (Hg)
from combustion flue gases, while producing a saleable fertilizer byproduct. Additionally, the process
removes other heavy metals and toxics release inventory substances, such as sulfuric acid (H2SO4),
hydrochloric acid (HCl), and hydrofluoric (HF) acid. A flow diagram of the process is provided in
Figure 6-3.
Airborne Clean Energy (ACE) has developed a process where the sodium sulfite byproduct can be
used to regenerate sodium bicarbonate reagent, while simultaneously producing an agricultural grade
fertilizer. ACE is offering this technology in combination with oxidants to achieve additional NOX and
Hg removal.
Sulfur Dioxide (SO2) and Sulfur Trioxide (SO3) are removed from the flue gas by dry injection of sodium
bicarbonate into the flue gas upstream of a wet sodium carbonate solution absorber. Reduction of NOX
occurs by dry injection of sodium bicarbonate reagent combined with downstream wet scrubbing. First,
NO is converted to NO2 in the presence of dry bicarbonate particles and SO2, and a portion of the
NO2 is removed in the downstream wet scrubber to form NaNO3. The removal of NOX in the process
is closely related to the simultaneous removal of SO2. If SO3 removal is not required, it may not be
necessary to install the dry injection system, depending upon the NOX removal requirements. The use
of a wet scrubbing system is sufficient for SO2 removal alone.
Oxidants ID Booster
To Stack
Fan
Scrubber
Boiler Existing
ESP/Baghouse Sodium Sol’n
Sodium Treatment and
Sulfate Tank Purification
Fly Ash Airborne

Ammonium Bicarbonate
Sodium Regeneration
(Ammonia and CO2
Bicarbonate Process
from flue gas)
Conditioner
Fertilizer
Granule
Production
(Optional)
Ammonium
Sulfate Byproduct
Figure 6-3
Airborne Simplified Process Flow Diagram
12239985
In the Airborne Process™, the absorber tower functions in two modes: (1) a polishing unit for any
remaining SO2 and (2) a transformation tower to convert the sodium sulfate/nitrate powder from the
injection process into a solution that will react with the remaining SO2. An oxidant is injected in a
secondary distinct zone in the absorber to remove additional NO and mercury. Airborne has demon-
strated in a 7 kW-sized unit where the oxidant addition had removed over 99% of NO and 99% of
Hg in elemental from in the flue gas.
Effluent from the wet scrubbing and oxidant zones is purified by a pH adjustment that precipitates
metals collected from the gas by the recycled slurry. A filter press removes the particulate and
precipitated metals from the treated slurry. The purified solution of sodium sulfate/nitrate with chlorides
is then sent to the regeneration process. The sodium sulfate/nitrate solution from the scrubber is sent
on to a preparation train that removes particulate, Hg and other impurities from the stream. Sodium
sulfate/nitrate brine/slurry from the preparation step enters a crystallization train where reaction with
ammonium bicarbonate forms ammonium sulfate and regenerated sodium bicarbonate.
The regeneration of sodium bicarbonate can also be achieved by an alternate reaction of ammonia
and carbon dioxide gases with the sodium sulfate/nitrate solution. Sodium bicarbonate is precipitated
out of solution, dried, and sized for re-injection in the scrubbing system.
The purified solution of ammonium sulfate/nitrate/chloride is pumped into a final crystallization process
in order to extract the ammonium sulfate. The ammonium nitrate/chloride solution is sold directly to
the fertilizer market. The chloride purge from the scrubbing process is converted into an ammonium
chloride fertilizer. The solid ammonium sulfate product can be processed for sale as a granular fertilizer,
or processed to form alternative fertilizer products. ACE has a proprietary granulation process that
significantly improves the mechanical properties of the fertilizer, producing a round uniform sphere of
high strength that dissolves quickly in the soil.
The ammonium bicarbonate reagent fed to the sodium sulfate reactor described in the regeneration process
is created by reaction of ammonia, carbon dioxide, and water to form ammonium bicarbonate.
Process reagent feeds include anhydrous ammonia, carbon dioxide, and makeup sodium sulfate. ACE
anticipates the use of the flue gas stream as the CO2 source for the Airborne Process™ in the future.
Sodium scrubbing has been in operation at PacifiCorp’s Jim Bridger Power Plant, Naughton Station in
Wyoming, and at Nevada Power’s Gardner Station. There are over 3,200 MW of sodium scrubbing
experience that has been in operation for over 25 years. Sodium scrubbing at these units is made
economical by use of a waste product sodium material from a manufacturing process in Wyoming.
The regeneration component of the Airborne Process™ was successfully demonstrated at Airborne’s
commercial sodium sulfate mine in Ormiston, Saskatchewan, Canada. The 18 ton/day demonstration
plant was initially built with the intention of upgrading sodium sulfate to commercial-grade sodium
bicarbonate for sale. The demonstration plant at Ormiston produced a 98%+ sodium bicarbonate
product and 99%+ ammonium sulfate from a sodium sulfate feed stream. Both sodium bicarbonate and
ammonium sulfate were sold from the pilot plant.
The first demonstration of the integrated technology was installed at LG&E’s Ghent Generating Station
located in Carroll County, Kentucky. As a part of the DOE’s Clean Coal Power Initiative, the technology
was installed on a 5 MW equivalent slipstream of Unit 2, and tested between January and July 2003.
The tests demonstrated effective operation of the system including the reagent regeneration step.
According to data provided by the vendor, the demonstration unit achieved 99.9% SO2 removal on
a consistent basis. NOX removal was approximately 40% during normal operation, and increased up
to 90% removal with the addition of an oxidant. Additional NOX removal could be obtained by the
FGD Performance
reduction of sodium bicarbonate particle size. Mercury removal during operation of the demonstration
facility peaked at 60–80% removal.
A commercial scale demonstration at the 300 MW Mustang Generating Station near Milan, NM was
halted when Peabody Energy, HPD LLC was unable to secure an air permit from the New Mexico
Government. ACE is installing their system at Saskatchewan Minerals and they are in negotiations with
other clients in North America and China.
J-Power/EPDC ReACT Process
ReACT Process Description

The ReACT process utilizes a moving bed of activated coke (AC) with ammonia injection to simultaneously
remove SO2, NOX, and mercury from the flue gas. Spent activated coke from the adsorption process is
regenerated and recycled back to the adsorber, while SO2 rich gas is sent to a byproduct recovery unit
and processed into a saleable sulfuric acid or gypsum byproduct. The ReACT system is currently being
marketed worldwide by J-POWER EnTech Inc. J-Power/EPDC has been a Japanese utility since 1952.
They are the largest wholesaler of power in Japan, with a total of 67 plants with a total capacity of
more than 16,000 MW. Of these, approximately half of their capacity is generated by coal-fired boilers.
They have one earlier ReACT installation at the Takehara plant that has been operating since 1995.
In the adsorption step of the process, ammonia is injected into the flue gas which then passes through
an adsorption tower containing a moving bed of ½” diameter activated coke pellets. The coke flows
downward through a series of modules installed side by side. The flue gas passes thorough the coke
bed horizontally and then exits the system. SO2 in the flue gas reacts with oxygen, water, and ammonia
to form various compounds that are adsorbed onto the surface of the activated coke. The ReACT process
removes SO2 by adsorption reaction with water vapor and oxygen to form sulfuric acid:
2SO2 + O2 + 2H2O → 2(H2SO4)ads
The sulfuric acid remains adsorbed on the surface of the activated coke. Some of the SO2 in the flue gas
reacts with ammonia to form ammonium sulfate compounds, which remain adsorbed on the activated coke:
(H2SO4)ads + NH3 → (NH4HSO4)ads
(NH4HSO4)ads + NH3 → [(NH4)2SO4]ads
NOX in the flue gas is catalytically reduced to nitrogen and water by reaction with ammonia in the coke
bed within the adsorbers and in the regeneration process:
4NO + 4NH3 + O2 → 4N2 + 6H2O
NOX is also removed by reducing compounds on the surface of the activated coke.
The flow rate through the adsorber modules is set so that the activated coke is saturated with SO2 when
it exits the adsorber modules. The saturated coke is then conveyed to regenerator vessels where the
activated coke is thermally regenerated by indirect heating to 400–500°C (750–930°F). The regenerated
coke passes through a vibrating screen to remove any degraded material (fines), which can be reused
and sold as a sorbent for mercury, dioxins and toxins removal for industrial applications or burned in
the boiler for fuel. Fresh activated coke is added to replace any fines removed in the screening process,
and the regenerated activated coke is then recycled back to the adsorber.
In the regenerator, ammonium sulfate in the activated coke is decomposed to nitrogen, sulfur dioxide,
and water. Sulfuric acid is also decomposed to sulfur dioxide and carbon dioxide by reaction
12239985
with carbon. The SO2-rich gas (50–60% SO2 by volume) produced in the regenerator is then sent
to a byproduct recovery unit where it is further processed to form sulfuric acid or gypsum using
conventional methods.
Mercury in the flue gas is removed by adsorption onto the surface of the activated coke particles in the
presence of SO2 regardless of its speciation (elemental or ionic) and chlorine content in the flue gas.
Adsorbed mercury remains on the surface of the activated coke during the regeneration process. A
portion of the activated coke is extracted from the process every few years, depending on the mercury
concentration in the coal. The extracted AC is treated offsite to remove the adsorbed mercury, which
can be reused and sold as elemental mercury for industrial applications.
It should be noted that activation of the AC increases with time because the surface area of the AC
increases during the recycling process. Additionally, more reducing compounds are generated on the
surface of the activated coke by ammonia treatment during the regeneration process.
It is possible to enhance NOX removal and maximize AC utilization by injecting ammonia into the
regenerator. Ammonia injection to the regenerator generates more reducing compounds on the surface
of the activated coke, which increases NOX removal efficiency in the process. This allows optimization of
the system performance and cost based on the trade-off between the cost of ammonia and activated
coke which both reduce NOX emissions. The optimum design and operating point is site-specific and
depends on the prices of ammonia and activated coke.
The ReACT system is installed downstream of an existing particulate control device; however, the
technology does remove some residual particulate from the flue gas. It should be noted that some activated
coke fines are generated as the AC is conveyed through the adsorber, and are released into the flue
gas offsetting the particulate removal by the ReACT system. Net particulate removal across the ReACT
adsorber is approximately 70%.
The activated coke process was initially developed in Germany and Japan by Bergbau Forschung (BF)
and licensed to Mitsui Mining Company. The Mitsui-BF process was developed to utilize Mitsui-manu-
factured coke to adsorb SO2 and NOX in a two-stage adsorber, with a subsequent AC regeneration
step. Mitsui, Electric Power Development Company (currently known as J-POWER), and Sumitomo
Heavy Industries developed a variation of this process, consisting of a single-stage adsorber with
ammonia injection prior to the adsorption stage. J-POWER has recently acquired the patents for the
Mitsui-BF process and established J-POWER EnTech Inc. to market ReACT worldwide.
The ReACT technology has been installed on 14 commercial units to date, including 4 coal-fired utility
boilers and other industrial flue gas applications. All commercial installations of the technology
are located abroad, with the majority located in Japan. The technology has been demonstrated
successfully on large utility boilers, with a 350 MW installation in operation at J-Power’s Takehara Unit
#2 since 1995, and a 600 MW installation in operation at Isogo #1 since 2002. J-POWER is currently
constructing another 600 MW unit at the same site, Isogo #2, scheduled for startup in July 2009.
Figure 6-4 contains a simplified process flow diagram for the ReACT process.
Commercial installations located in Japan and Germany operate at 90–99% SO2 removal, with SO2
inlet concentrations as high as 1300 ppm SO2. NOX removal at existing installations ranges from
50–80% on coal-fired units, but is highly dependant on coal sulfur content. SO2 in the flue gas reacts
readily with injected ammonia, limiting the amount of NOX removal that can occur at an injection rate
that limits ammonia slip. J-POWER EnTech reports ammonia injection into the regenerator is necessary
if more than 50% NOX removal is required under higher SO2 concentrations. Greater than 90%
mercury removal has been demonstrated at the 600 MW Isogo #1 installation.
FGD Performance
Ammonia Flue Gas

to Stack
Flue Gas from

Existing ESP/FF
Byproduct Sulfuric
Adsorption Processing Acid
Reactor Plant Byproduct
Regen.
Gas Natural
Make-up Regeneration
Furnace Gas
Activated Tower
Coke Silo
Coke Fines
(to sale/disposal)
Figure 6-4
ReACT Simplified Process Flow Diagram
At this point, the ReACT technology has not been demonstrated on high-sulfur coal applications. To
simulate U.S. flue gas conditions, J-POWER has performed slipstream testing at Isogo #1 with sulfur
content up to 2,000 ppm. Performance tests are being run at various conditions in order to optimize
the design, and are expected to be complete in March of 2006. J-POWER EnTech expects that there
will be some modifications to the system design to adapt to the flue gas from Eastern U.S. coals, and
plans to do slipstream testing in the U.S. in the future.
Recent ReACT Process/Design Developments

The ReACT process is installed commercially at multiple sites in Japan. The process is a modified version
of older activated coke designs, and some of these modifications are listed below:
■ J-Power has evaluated U.S. coals and found that they can make an effective activated
coke using these fuels as feedstock to the coke process. They have used PRB and lignite,
but the best cokes are made using bituminous coal when trying to achieve higher NOX
reduction rates.
■ Slipstream testing of the ReACT technology demonstrated 94% SO2 reduction with an
inlet SO2 concentration of 5 lbs/MMBtu (2.5 MW pilot test).
■ NOX reduction efficiency has been shown to increase with higher inlet gas temperatures.
J-Power is targeting 70% reduction for future installations.
■ J-Power acquired all rights to the Mitsui activated coke process in 2005, with Mitsui still
acting as the supplier of the activated coke.
■ Recent ReACT system designs are moving toward lower pressure drop. At the Isogo
plant, the Unit 1 pressure drop was 14” w.g., while the Unit 2 installation is expected to
be 7–8” w.g. (Unit 2 is expected to be on-line in 2009).
■ The process has no SO3 slip.
■ The ReACT system consumes less than 1% of the water used in a wet scrubbing process.
Installation of the ReACT process equipment will incur similar retrofit issues to those that accompany
conventional FGD system installation at an existing plant. However, the additional equipment required
12239985
for processing of the sulfuric acid byproduct could pose a problem in highly congested plants. It should
be noted that the overall plant area requirement for the ReACT system is likely to be less than that
needed for a standard LSFO process when considering the need for limestone and gypsum storage on
the plant site. The technology does not require any stack modifications. The pressure drop added
by the ReACT components will require a similar pressure transient analysis of existing components in
the gas path, as well as the boiler.
The ReACT technology is highly sensitive to coal sulfur content. The system was initially designed for
low-sulfur (<1300 ppm) coals, and has not been optimized for operation with higher sulfur contents at
this point. For high sulfur concentrations, the amount of activated coke fed to the adsorber increases
substantially due to the need for more contact area to remove SO2 from the flue gas. This results in the
need for additional adsorber and regenerator modules, and supplementary solids conveying equipment.
Additionally, the amount of sulfuric acid byproduct increases with the amount of SO2 removed, resulting
in increased sulfuric acid production plant costs. It should also be noted that higher NOX removal in the
process was limited by the amount of SO2 present in the flue gas, because reactions with SO2 compete
with the NOX removal reactions that occur in the adsorber. However, recent findings indicate that NOX
removal reactions from reducing compounds on the surface of AC allow removal of >50% NOX for gas
streams with higher SO2 concentrations. Demonstrated SO2 removal efficiency has approached 99%
in some low sulfur coal commercial installations.
Other Developing Multi-Pollutant Processes

The following developing technologies are targeting 99% SO2 removal efficiency or higher: ISCA,
Lextran, Consummator, Green Coal, Envirolution, Flue-Ace, PEA, EnviroScrub, and the IPR system. All
systems are proceeding through initial stages of commercial development.
FGD Performance
12239985
Co2 Removal Processes
7 CO2 REMOVAL PROCESSES
Types of CO2 Removal Processes
A lthough the bulk of this report covers high efficiency SO2 removal, the looming possibility
of regulations requiring fossil fuel-fired power plants to reduce CO2 emissions is relevant to the
discussion. Some CO2 removal processes will directly impact the design and performance required
from any FGD system operating upstream due to the low inlet SO2 concentration sometimes required
by the CO2 reduction system. In many cases, sulfur dioxide needs to be in the low ppm range so that
the CO2 removal processes can be a cost-effective candidate for installation at a PC-fired power
plant. Table 7-1 provides a listing of CO2 removal technology options and their applicability for use
on flue gas streams generated by coal-fired boilers.
Table 7‑1
Summary of SO2 Requirements for CO2 Removal Technologies
Removal Removes Operating Technical Issues for Coal-Fired
Technology Conditions Boilers
Typical amine H2S & CO2 Low temperature & Trade off between higher FGD SO2
chemical solvents low pressure removal costs and higher Amine
capital and operating costs for
control of heat stable salts
CANSOLV SO2 & CO2 Low temperature & If there are sulfur compounds other
low pressure than SO2 then incineration
is required
Physical solvents H2S & CO2 Low temperature & Gas needs to be pressurized
high pressure
Chilled ammonia CO2 Low temperature & In the demonstration phase
low pressure
ECO2™ SO2, NOX, Low temperature & In the demonstration phase; CO2
Hg and CO2 low pressure removal may not be cost effective
without an ammonia-based
scrubbing for SO2 removal
Chemical Solvents
The chemical solvent systems discussed below employ chemical reactions to remove compounds
and purify a gas stream. The solvents are then thermally regenerated, driving off the dissolved
contaminant.
Several types of amine solvents have been used as absorbents in alkanolamine gas purification
systems, with the primary amines being monoethanolamine (MEA), diethanolamine (DEA) and
methyldiethanolamine (MDEA). Ideally, chemical solvents require low pollutant concentrations, low
feed gas pressures, and low temperatures. Additionally, amine based systems are primarily used for
acid gas removal (i.e., H2S and CO2) and not SO2 removal. SO2 in the feed gas stream will form
sulfates and these sulfates form heat stable salts. CO is another compound that will hydrolyze to
formates and form heat stable salts. The heat stable salts are a contaminant and if they build up they
can cause corrosion problems. Corrosion problems generally appear long before the heat stable
salts impact the amine solutions effectiveness. However, a build up of impurities can cause foaming
problems in the absorber. Overall, the net effect is a reduction in the amine capacity to remove CO2.
FGD Performance
The solvent can be regenerated to control the amount of heat stable salts. Regeneration can be
performed in one of the following ways:
■ Thermal reclaimer
■ Ion exchange
■ Electrodialysis
Currently, thermal reclaiming is the most common form of regeneration. Thermal reclaiming involves
vaporization of volatile compounds and thus is energy intensive. Ion exchange requires a low amount
of energy, but this option can potentially result in high water and chemical usage. Additionally, this
option only removes ionized degradation products. Electrodialysis uses a direct current along with
ion-selective membranes. This method does not remove non-ionized degradation products.
The thermal reclaimer option results in the highest amine loss while electrodialysis results in the lowest
rates of amine loss. Many refinery applications actually rely on a mobile regeneration unit that treats
the amine solution every 1 to 2 years. Some amine systems have their own reclaiming unit. Overall,
reclaiming/regenerating the amine solution via any method requires additional capital and operating
costs. There will be some amine loss as well as some solid waste disposal.
Of note is the fact that other flue gas components will result in amine degradation. For example,
oxygen will degrade the solvent. Amines can be directly oxidized to organic acids. In the presence
of an amine, oxygen can react with H2S to form various decomposition products as well as a variety of
acid compounds. Additionally, although amines are used to remove H2S and/or CO2 from a gas stream,
irreversible reactions of the amine with CO2 typically produce degradation products. These degradation
products can tie up solution capacity, increase foaming tendency, and even contribute to corrosion.
Generally, the degradation reactions involving CO2 are relatively slow provided that the regeneration
temperature stays below a recommended maximum of around 260°F (127��
°��
C) for most amines.
CANSOLV
CANSOLV is an amine-based system that was developed to remove SO2 from industrial gas streams
(see Section 6 for more details). Currently, there are seven operational systems with two of the systems
operating on refinery CO boiler flue gas. To date no systems have been built at a power generation
plant, but one is in the engineering stage for a coal-fired cogeneration plant in China. CANSOLV can
also be designed for SO2 removal only or for removal of both SO2 and CO2.
If sulfur compounds such as H2S exist in the gas stream treated by the CANSOLV system, then the
gas must be incinerated in order to oxidize the sulfur species to SO2. This will increase the capital and
operating cost of the system. Like other amine systems, this process operates at low temperatures
and low pressures. To maintain amine solution quality, the process uses a proprietary salt removal/
electrolysis unit, which processes a slipstream of the lean amine solvent to avoid the buildup of salts
that can lead to degradation and corrosion issues.
Physical Solvents
Physical solvent systems dissolve acid gases and other impurities into the solvent then these components
are removed by a reduction of pressure in a flash vessel. Generally, a heat source is not required to
strip out the unwanted components. However, if a high degree of removal is required then an
additional flash at vacuum conditions can be employed. Another possibility is the use of inert gas
stripping (typically nitrogen) to lower the impurity content in the solvent, or thermal regeneration can
be applied to the stripper.
12239985
Rectisol, Selexol, and Purisol are three examples of commercial physical solvent systems that are
available today for CO2 reduction from gas streams. Physical solvents have a high solubility for
SO2. In fact the SO2 solubility is orders of magnitude higher than H2S or CO2 (e.g. H2S is about
10 times as soluble as CO2, and SO2 is about 90 times as soluble as CO2). However, physical
solvents also require a high partial pressure of the compound that is being removed. This means that
the compound needs to be available in the gas at a high concentration or at a high pressure. Plus,
physical solvent systems typically operate at pressures greater than 150 psi and low temperatures.
The partial pressure of CO2 in power plant flue gas is low due to both a low concentration and near
atmospheric conditions. Thus the flue gas would need to be compressed and this is not economical
for these large gas streams.
Chilled Ammonia Scrubber

Alstom’s Chilled Ammonia Process is in the demonstration phase. The technology was developed
for use after a flue gas desulfurization system. The flue gas is cooled to 32–68��
°��
F (0–20°C). The flue
gas enters an absorber containing a slurry mixture of ammonium carbonate and ammonium bicarbonate.
The CO2 rich slurry is then regenerated at high pressure (i.e., 300–600 psi) and an increased temperature
(140–175°F; 60–80°C).
Currently a 5 MW demonstration plant will be built in Sweden at E. ON’s Karlshamm station. Alstom
also has two contracts with American Electric Power (AEP): 1) to build a 30 MW demonstration unit in
New Haven, West Virginia at the Mountaineer Plant, which is scheduled for start up in 2008, and 2)
to construct a commercial scale 200 MW unit in Oologah, Oklahoma at the Northeastern Station. The
chilled ammonia CO2 capture technology will also be demonstrated at We Energies Pleasant Prairie
plant in Wisconsin. All of these power plants have an existing wet limestone FGD system. The Chilled
Ammonia Process is expected to be compatible with the current amount of sulfur removal from each of
these plants and thus it is not anticipated that 99+% removal of SO2 will be required for cost-effective
CO2 removal when using this process. Results from the demonstration facilities will confirm the chilled
ammonia process capability to handle double digit SO2 inlet ppm concentrations that are typically
achieved by current FGD system designs.
Powerspan - ECO2™ Process

The ECO2™ process is an add-on to the Powerspan’s ECO™ technology (see Section 6 for
more details). This system utilizes an aqueous ammonia solution to absorb CO2 from the flue
gas. It is designed to operate downstream of existing emissions control systems for SO2, NOX,
and Hg. It is particularly advantageous for sites where ammonia-based scrubbing technology
is already in operation, such as those sites operating the Powerspan ECO™ technology for
multi-pollutant removal. The process solution is regenerated to release the captured CO2 and
recover the ammonia reagent. The vendor states that no losses of ammonia occur in the system,
and no secondary byproducts are produced. Development of the process started in 2004 based
on cooperative work with the DOE-NETL research laboratory. 90% CO2 reduction was demonstrated
in the lab.
British Petroleum has teamed with Powerspan to pilot test (1 MW equivalent slipstream) the ECO2™�
technology at a power plant in Ohio, which is expected to be operational in 2008. NRG and Power-
span are planning on demonstrating the technology at their 125 MW plant in Fort Bend, Texas and
using the CO2 for enhanced oil recovery. This plant is expected to be operating by 2012. The ECO
technology was designed as a multi-pollutant system, but it is possible to apply the CO2 removal as
a retrofit technology at any plant. However, it may not be as cost effective at sites that do not use an
ammonia-based FGD system.
FGD Performance
Cost Impacts of SO2 Concentration on CO2 Removal Processes

Introduction
The impact of SO 2 concentration in PC-fired power plant flue gas will have an impact on the
performance of amine-based CO2 removal processes, such as monoethanolamine (MEA). This subsection
provides an assessment of the cost impact on these systems at various inlet SO2 concentrations. A 10 ppmv
SO2 concentration was used in the most recent power plant cost and performance analysis by the
National Energy Technology Laboratory’s (NETL, 2007). The NETL analysis was based on Illinois #6
coal fuel (sulfur content ≈ 4 wt%). The SO2 concentration exiting an LSFO FGD system design for 98%
SO2 reduction would be approximately 38 ppmv; therefore, the design employed a polishing scrubber
to reduce the SO2 outlet concentration to 10 ppmv. The polishing scrubber uses a sodium hydroxide
(NaOH) solution as the source of alkali.
Incremental Costs of SO2 and CO2 Removal for Increasing FGD System
Efficiencies
The main reason for reducing the SO2 concentration to a very low level prior to the amine system is
to minimize the buildup of heat stable salts in the amine scrubbing loop. A thermal reclaimer is
commonly included with an amine scrubbing system to remove these heat stable salts. This regeneration
step requires energy, produces a waste product, and results in an increase in the amine degradation/
makeup rate. Ion exchange and electrodialysis are two other steps that can be used to regenerate
the amine solution as an alternative to the thermal reclaimer, but they are not used as frequently as
thermal reclaiming. These two regenerating steps require less energy and result in lower amine loss
rates. However, this configuration may have increased operating costs associated with water and/or
chemicals, and is typically more capital intensive.
An analysis was performed to examine the levelized cost for a PC power plant with a FGD step and
CO2 removal using an MEA system (levelized cost is presented as dollars per tonne of CO2). The
following three scenarios were examined:
■ PRB coal (0.48 wt% sulfur) with wet LSFO FGD and CO2 removal by amine scrubbing,
■ Lignite (1.0 wt% sulfur) with LSFO FGD and CO2 removal by amine scrubbing,
■ Appalachian coal (2.6 wt% sulfur) with LSFO FGD and CO2 removal by amine
scrubbing.
The plant size was assumed to be 500 MW, gross, excluding the power requirement for CO2 and SO2
removal systems.
The analysis examined the incremental cost for removing additional SO2 in increments from 90% to
99% for the LSFO process. This included the incremental capital and operating costs associated with
the FGD system at various control efficiencies, as well as the incremental operating costs for the amine
scrubbing system as they related to variations in the inlet SO2 concentrations. These are costs are
primarily related to the operation of the thermal reclaimer including variations in steam usage, amine
losses, and waste generation rates. All of these operating costs typically increase as the inlet SO2
concentration increases.
The correlations for the increase in MEA consumption and increased reclaimer duty were taken from
Simisky (1986). The cost for makeup of the MEA was assumed to be $2.11/lb. The steam cost was
assumed to be $3.86/1000 lb and the amine waste disposal cost was based on $0.19/gallon. The
results are shown in Figures 7-1 to 7-3, showing the SO2 inlet concentration impact on the incremental
levelized cost for both the FGD and CO2 removal systems. The capital and operating cost for CO2
12239985
capture and storage on a PC power plant would be an additional $50/tonne of CO2 removed, as
estimated by another publication.
For each option, it is clear that the lowest inlet SO2 concentration to the amine unit will result in the
lowest additional cost for CO2 removal. The cost for amine losses is the largest contributor to the
increase in levelized cost, followed by the additional energy cost associated with the steam requirement
for thermal regeneration. This analysis assumes that wet LSFO will be required to achieve SO2 reductions
above 95%, so it was used for all cases. In all cases the highest removal rates evaluated will result in
inlet SO2 concentrations above 10 ppm for the lignite and bituminous coal cases. The exception is the
PRB case where a removal efficiency of 99% will produce an outlet emission rate of less than 10 ppm.
At this point the ability to sustain 99% removal or higher for LSFO systems on a 30-day or longer-term
averaging period has not been proven except at facilities located in Japan.
If both the energy and amine losses can be reduced by about half without any additional operating
costs, then the incremental levelized cost shown in Figures 7-1 to 7-3 would decrease by about half.
Further investigation would be needed, but it may be more cost-effective to use either ion exchange or
electrodialysis instead of thermal reclaiming in future applications.
FGD Performance
PRB Coal – Incremental Cost of FGD System Plus CO2 Removal System
Figure 7‑1
12239985
Lignite - Incremental Cost of FGD System Plus CO2 Removal System

Figure 7‑2
FGD Performance
Appalachian Coal - Incremental Cost of FGD System Plus CO2 Removal System
Figure 7‑3
12239985
Summary and Conclusions
8 SUMMARY AND CONCLUSIONS
T his analysis has shown that the current FGD technologies available to the U.S. utility industry
are capable of achieving 99% SO2 reduction or higher with only limited modifications to their
absorber designs and operating conditions. McIlvaine and others have stated that the future retrofit
of CO2 reduction technology will require an order of magnitude reduction in the other coal-fired plant
pollutants. This future technical requirement and the pressure on the industry to install and operate
power generation plants with lower emissions to meet the demands of the U.S. public and governmental
agencies will force the FGD industry to supply ever higher SO2 reduction performance. But as McIlvaine
also noted in January 2008, achieving 99% SO2 reduction is not that expensive.
FGD Performance Enhancement

Previous and current analyses have shown that the liquid to gas ratio of slurry feed to the flue gas is
directly related to the FGD system performance. These higher slurry feed rates are somewhat ineffective
in that redundant coverage is provided in the vessel. However, the added slurry feed rate also helps
to maintain better flue gas distribution throughout the absorber, while providing additional gas-liquid
contact surface and reducing the potential for gas sneakage. Further improvement can be obtained by
increasing the slurry pH and improving the gas distribution within the absorber vessel. Units installed
overseas have guaranteed and are achieving greater than 99% SO2 removal, in one case continuously
maintaining 99.8% SO2 reduction.
EPRI has previously investigated what would be necessary to achieve these 99+% FGD reduction
efficiencies. Options included the use of additives (DBA and formate), increased reagent feed rates,
increased recycle rates for absorption slurry, increased slurry alkalinity, the addition of a tray in
the absorber gas path, etc. To maintain these high removal efficiencies on a continuous basis,
the FGD system must also be designed for very high reliability, and the suppliers and EPRI have
identified those components that need specific attention to minimize periodic maintenance issues
that can reduce system performance. This would include the agitators and other components in
contact with the low pH slurry, instrumentation to allow constant monitoring of process chemistry,
mist eliminators that historically have experience fouling issues, and overall materials performance
in this harsh environment. Recent advances in FGD designs and system specifications have helped
to both improve system reliability to nearly 100% for the latest installations, as well as achieve very
high SO2 reduction rates.
Regulatory Drivers for Higher Efficiency

Over the period from 2007 to 2010, power generation companies projected the need for approximately
180 FGD units to meet the deadlines of CAIR. Many utilities have already contracted and are in the
processes of the engineering, design, and construction of these facilities. Some of these CAIR-driven
FGD systems are already operating well ahead of the regulated operating date of 2010.
As part of the CAIR regulations, utilities are allowed to bank one allowance for each extra ton of SO2
(beyond their regulated emission rate) removed prior to 2010. Allowances generated after 2010 will
be devalued by one-half, requiring two allowances to emit one ton of SO2 removed.
In response to the 2010 requirements of CAIR, local regulations, and/or the incentive to bank SO2
allowances 1-for-1, some utilities have been specifying FGD systems with SO2 removal guarantees of
98% to 99% removal. This report identified 15 FGD systems that have been or are being designed
and/or guaranteed to achieve 98% to 99% SO2 removal.
FGD Performance
FGD Design Upgrades

The latest design modifications by some FGD suppliers have achieved high removal efficiencies without
need for major redesign of the absorber tower, without the need for performance additives, or the need
to have two absorbers in series. These changes included the use of wall baffles or changes in spray pat-
terns within the absorber to eliminate the possibility of flue gas slippage through the absorber modules
without contacting the absorption slurry. In a number of cases, it has also been necessary to increase
L/G to the range of 180 to 200, sometimes requiring an extra absorber circulation pump(s) and spray
level. Other options would include addition of fresh limestone slurry feed directly to the last spray level
so that the highest slurry pH is present at the lowest SO2 concentration in the flue gas. Some plants may
also benefits from the use of buffering agents such as dibasic acid to improve FGD performance.
FGD Operating Philosophy

In order to achieve 99+% SO2 removal on a multi-day rolling average basis, utilities need to consider
adopting a different operating philosophy. In particular, utilities need to place greater emphasis on
monitoring FGD performance and possibly including engineering staff in the control room to augment
the plant operating staff. In essence, U.S. utilities need to adopt the Japanese FGD operational
philosophy, which includes hands-on involvement of engineers in the day-to-day operation of the FGD
system. This typically will require consistent vigilance to ensure maintenance of the desired process
chemistry, and thereby maintain the high level of FGD performance on an hourly basis.
Typical causes of temporary FGD system shutdowns or reductions in performance would include:
■ Loss of control of the process chemistry resulting in internal scaling or plugging of process
piping or the mist eliminator
■ Loss of a recycle pump
■ Failure of a damper
■ Failure of the dewatering system
■ Loss of the limestone feed
■ Loss of the oxidation air system or a recycle tank agitator.
Some of the newest FGD installations are maintaining nearly 100% reliability due to design improvements,
upgraded operating practices and advances in control and process monitoring instrumentation. The
use of additional installed spare components throughout the FGD system and its supporting subsystems
may be required to further improve on the reliability of the system when maintaining continuous ultra-high
SO2 reduction efficiency.

The dramatic increase in U.S. and overseas demand has resulted in FGD suppliers becoming booked well
into the future, and in some cases significantly extending delivery/construction schedules. In addition, FGD
suppliers’ delivery times are being impacted by extended lead times from their major equipment suppliers.
One example is large rubber-lined pumps, which as of early 2007 had lead times in the range of 110 to
120 weeks.
The increase in demand for FGD commodities/equipment combined with the demand for commodities/
equipment for other large capital projects in the U.S. and overseas has led to significant increases in
FGD system costs. Over the period from 2004 to mid-2007, escalation of materials and equipment,
craft labor, and engineering have resulted in increases in costs of wet FGD systems of 40% to 60%.
The larger pumps or an increase in the number provided, along with an increase in sparing for other
critical components to achieve 99+% SO2 removal, will further add to the significant increases that
have already occurred in the capital cost of FGD systems.
12239985
British to Metric Conversion Factors

To Convert British Multiply By To Obtain Metric
(SI=Systems Intern)
Ac Acre 0.405 ha Hectare
Acfm Actual Cubic feet per 0.02832 Am /min 3
Actual cubic meters/min
minute
Btu British Thermal Unit 0.252 kcal kilocalories
Btu British Thermal Unit 1055.1 J Joule
°F Deg. Fahrenheit – 32 0.5556 °C Degree Centigrade
Ft Feet 0.3048 m Meters
Ft2 square feet 0.0929 m2 square meters
Ft3 cubic feet 0.02832 m3 cubic meters
Ft/m feet per minute 0.00508 m/s meters per second
Ft /m
3
cubic feet per minute 0.000472 m /s3
cubic meters/second
Gal gallons (U.S.) 3.785 L Liters
Gpm gallons per minute 0.06308 L/s Liters per second
gpm/Kacfm gallons per minute 133.65 liters/ liters per actual cubic
thousand actual cubic Am3 meter
feet/min
Gr Grains 0.0648 g grams
Gr/ft 3
grains per cubic foot 2.2881 g/m 3
grams per cubic meter
Hp Horsepower 0.746 kW kilowatts
In. Inches 0.0254 m meters
in. w.g. inches water pressure 249.089 Pa pascals (newton/m2)
(gage)
Lb Pounds 0.4536 kg kilograms
Lb/ft 3
pounds per cubic foot 16.02 kg/m 3
kilograms/cubic meter
Lb/hr pounds per hour 0.126 g/s grams per second
lb/MMBtu Pounds per million BTU *Depends on mg/Nm 3
milligrams per normal
Fuel Type cubic meter
Mi Miles 1609 m meters
MMBtu/hr million Btu per hour 1,055 Mjoule/ million joules per hour
hr
Oz Ounces 28.3495 g grams
Psi pounds per square inch 6895 Pa Pascals (newton/m2)
Rpm revolutions per minute 0.1047 rad/s radians per second

scfm Std. (60°F) cubic feet/ 1.6077 nm /hr 3
normal cubic meters/hr
minute
ton short tons 0.9072 tonne metric tons
t/hr short tons per hour 0.252 kg/s kilograms per second
$/ton dollars per short ton 1.1023 $/tonne dollars per metric ton
FGD Performance
Developing Multi-Pollutant Control Technology Options

The Powerspan, CANSOLV, Airborne, and ReACT multi-pollutant processes offer the advantage of
simultaneous removal of SO2, NOX, and mercury. Separate processes are not required for each pollutant.
In addition, many multi-pollutant processes show promise in achieving high levels of SO2 removal in early
demonstration testing. However, the SO2 removal capabilities of the various multi-pollutant processes do
not yet have the commercial experience to replace wet FGD processes, or provide a guarantee that they
will outperform the current commercial technologies in terms of continuous high SO2 removal efficiency.
The Powerspan and Airborne have been or will be tested at the demonstration level and are moving
toward full scale commercial installations. The 50 MW Powerspan process demonstration plant has
been operating for many years at a high sulfur coal fired boiler in Ohio. This system has demonstrated
a consistent ability to remove 98+% of the inlet SO2. The Airborne process was tested at a 5 MW scale,
and the vendor reported 99.9% SO2 removal. It appears that commercial scale units will be installed
for both of these technologies in the near future, but no commercial units are under construction at this
time. The CANSOLV process has already installed commercial scale units, but not at coal-fired power
plants. The process has achieved 99% to 99.5% SO2 removal at petrochemical and smelter installations.
The ReACT process has full-scale power plant installations in Japan and Germany. The plants have
been able to achieve up to 99% SO2 removal, but only on low to medium sulfur coals. The SO2 removal
and operational capability of the process has not proven on high sulfur coals. A recent pilot demonstration
in Nevada has further demonstrated the process applicability to U.S. coals.
CO2 Reduction Impacts on FGD Design

The bulk of this report covers high efficiency SO2 removal. However, the looming possibility of regulations
requiring fossil fuel-fired power plants to reduce CO2 emissions makes consideration of the performance
requirements of the CO2 removal processes pertinent to this discussion. The flue gas SO2 concentration
will need to be in the low single-digit ppm range in order for the CO2 removal processes to be cost-effective
for coal-fired utility applications.
12239985
References
9 REFERENCES
1. Acid Rain and Related Programs, 2006 Progress Report, Environmental Protection Agency,
2006.
2. Incremental Cost of Selected Emission Control Systems for PC Plants. EPRI EP-P6623/
C3368, September 2006.
3. Testimony by Professor Gregory J. McRae of Massachusetts Institute of Technology on
Clean Coal Technology – Science, Technology, and Innovation, United States Senate
Committee on Commerce, Science, and Transportation, April 27, 2007.
4. Tracking New Coal-Fired Power Plants, Coal’s Resurgence in Electric Power Generation,
DOE, National Energy Technology Laboratory, May 1, 2007.
5. Bureau of Labor, Producer Price Indexes, Commodity Data, www.bls.gov/ppi/home.htm
(Materials and Equipment Indexes).
6. Telephone discussions with vice-president of non-union subsidiary of Washington Group
International, Rust Constructors, August 2005 and August 2006.
7. Telephone discussions with Washington Group International vice-president in charge of
union construction projects, August 2005 and August 2006.
8. Craft Labor Supply Outlook, 2005 to 2015, Construction Labor Research Council,
2005.
9. Team Effort Needed to Head off Crisis, Engineering News Record, May 29, 2006.
10. Construction Week, Business Roundtable Pushes Job Training in Gulf Coast Area, Engineering
News Record, August 7, 2006.
11. Action Needed to Head off Looming Labor Shortage, National Association of Manufacturers,
Plant Engineering, November 1, 2005.
12. The Impact of Shortages on FGD Prices, Coal Power, September/October, 2007.
13. Worker Shortages Spike Wage Hikes, Labor Cost Report, Engineering News Record,
September 17, 2007.
14. Engineering Talent Squeeze --“People Deficit” -- Likely to Cause Further Delay in Some Oil
& Gas Production Projects through 2010, CERA, October 4, 2007.
15. State of the Art Emissions Control Strategy. EPRI Report, 2006.
16. Simisky, P.L. “The Recovery of CO2 From Flue Gases”. American Institute of Chemical
Engineers National Spring Meeting. New Orleans, Louisiana. April, 1986.
17. Judd, Bill. Dow. Personnel correspondence. Calgary, Canada. November 2007.
18. National Energy Technology Laboratory. Cost and Performance Baseline for Fossil Energy
Plants. Volume 1: Bituminous Coal and Natural Gas to Electricity. DOE/NETL –2007-
1281. August 2007.
FGD Performance
12239985
12239985
Export Control Restrictions The Electric Power Research Institute (EPRI), with major
Access to and use of EPRI Intellectual Property is granted with the locations in Palo Alto, California; Charlotte, North Carolina; and
specific understanding and requirement that responsibility for ensur- Knoxville, Tennessee, was established in 1973 as an independent,
ing full compliance with all applicable U.S. and foreign export laws nonprofit center for public interest energy and environmental
and regulations is being undertaken by you and your company. This research. EPRI brings together members, participants, the Institute’s
includes an obligation to ensure that any individual receiving access scientists and engineers, and other leading experts to work
hereunder who is not a U.S. citizen or permanent U.S. resident is collaboratively on solutions to the challenges of electric power. These
permitted access under applicable U.S. and foreign export laws and solutions span nearly every area of electricity generation, delivery,
regulations. In the event you are uncertain whether you or your com- and use, including health, safety, and environment. EPRI’s members
pany may lawfully obtain access to this EPRI Intellectual Property, you represent over 90% of the electricity generated in the United States.
acknowledge that it is your obligation to consult with your company’s International participation represents nearly 15% of EPRI’s total
legal counsel to determine whether this access is lawful. Although research, development, and demonstration program.
EPRI may make available on a case-by-case basis an informal as-
Together...Shaping the Future of Electricity
sessment of the applicable U.S. export classification for specific EPRI
Intellectual Property, you and your company acknowledge that this
assessment is solely for informational purposes and not for reliance
purposes. You and your company acknowledge that it is still the ob-
ligation of you and your company to make your own assessment
of the applicable U.S. export classification and ensure compliance
accordingly. You and your company understand and acknowledge
your obligations to make a prompt report to EPRI and the appropriate
authorities regarding any access to or use of EPRI Intellectual Prop-
erty hereunder that may be in violation of applicable U.S. or foreign
export laws or regulations.
Program:
Integrated Environmental Controls (Hg, SO2, NOx, and Particulate)
© 2008 Electric Power Research Institute (EPRI), Inc. All rights reserved. Electric Power
Research Institute, EPRI, and TOGETHER...SHAPING THE FUTURE OF ELECTRICITY are
registered service marks of the Electric Power Research Institute, Inc.
Printed on recycled paper in the United States of America
1014171
Electric Power Research Institute

3420 Hillview Avenue, Palo Alto, California 94304-1338 • PO Box 10412, Palo Alto, California 94303-0813 USA
800.313.3774 • 650.855.2121 • askepri@epri.com • www.epri.com
12239985

Flue Gas Desulfurization - FGD - Performance Capability

Uploaded by

Copyright:

Available Formats

You might also like

Flue Gas Desulfurization - FGD - Performance Capability

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Flue Gas Desulfurization - FGD - Performance Capability

Uploaded by

Copyright:

Available Formats

Flue Gas Desulfurization (FGD) Performance Capability

High Efficiency Design and Operating Options

High Efficiency Design and Operating Options

Final Report, March 2008

EPRI Project Manager

ORGANIZATION(S) THAT PREPARED THIS DOCUMENT

Washington Group International

This report was prepared by

Washington Group International

■ To provide background on regulatory requirements that have resulted in a significant

vi FGD Performance Capability: High Efficiency Design and Operating Options

2 WHY INCREASE SO2 REMOVAL EFFICIENCY? . . . . . . . . . . . . . . . . . . . . . . . . . 2-1

3 FGD SYSTEMS DESIGNED FOR HIGH SO2 REMOVAL. . . . . . . . . . . . . . . . . . . . 3-1

4 APPLICABLE SO2 REMOVAL PROCESSES. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-1

5 FGD SYSTEM COST ESCALATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-1

6 MULTI-POLLUTANT TECHNOLOGIES. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-1

8 SUMMARY AND CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-1

 FGD Performance Capability: High Efficiency Design and Operating Options

Figure 5-1 Cost Index for Ready-Mix Concrete................................................................. 5-4

Figure 6-1 ECO™ Process Flow Diagram........................................................................ 6-1

■ Installations of wall baffles/rings

2 WHY INCREASE SO2 REMOVAL EFFICIENCY?

3 FGD SYSTEMS DESIGNED FOR HIGH SO2 REMOVAL

Capability: High Efficiency Design and Operating Options

Capability: High Efficiency Design and Operating Options

4 APPLICABLE SO2 REMOVAL PROCESSES

Chiyoda CT-121 Jet Bubbling Reactor (JBR)

Chiyoda Process Description

Recent CT-121 Process/Design Developments

Mitsubishi/Advatech Double Contact Flow Scrubber (DCFS)

Mitsubishi (MHI) Process Description

Recent MHI Process/Design Developments

Alstom Power SO2 Removal Processes

Alstom Flowpac Process Description

From Other From Limestone

Recent Flowpac and Spray Tower Process/Design Developments

Babcock & Wilcox – Wet Lime or Limestone Forced Oxidation

Recent B&W Process/Design Developments

Recent Babcock Power Process/Design Developments

Siemens-Wheelabrator – Wet Limestone Forced Oxidation Process

Recent Hitachi Process/Design Developments

Alternate FGD Technologies for High SO2 Removal Efficiency

Dry Injection of Trona

Dynawave/Reverse Jet Scrubber

5 FGD SYSTEM COST ESCALATION

Increases in Material and Equipment Prices

Cost Considerations Related to Craft Labor Workforce

Compound Annual Esc. from 2004 to 2007.5 = 8.8%

120.0 Compound Annual Esc. from 1996 to 2003 = 2.0%

Ready Mix Concrete (1/1982 = 100)

Capability: High Efficiency Design and Operating Options

120.0 Compound Annual Esc. from 1996 to 2003 = 1.0%

Structural Steel (Jan 1982 = 100)

Capability: High Efficiency Design and Operating Options

350.0 Compound Annual Esc. from 2004 to 2007.5 = 25.5%

Copper Wire & Cable (Dec 1986 = 100)

Capability: High Efficiency Design and Operating Options

160.0 Compound Annual Esc. from 1996 to 2003 = 2.2%

FGD Performance Capability: High Efficiency Design and Operating Options