EPA- 600/7- 90-018;
                                          September 1990
 ASSESSMENT OF CONTROL TECHNOLOGIES FOR REDUCING

           IONS OF S02 AND NOX FROM EX1

           COAL-FIRED UTILITY BOILERS
EMISSIONS OF S02 AND NOX FROM EXISTING
                  FINAL REPORT
                       by

        David M. White and Mehdi Maibod1
               Radian Corporation
              Post Office Box "13000
  Research Triangle Park, North Carolina  27709
           EPA Contract No. 68-02-4286
               Work Assignment 84


                 Project Officer

                  Norman Kaplan
      U. S. Environmental Protection Agency
 Air and Energy Engineering Research Laboratory
  Research Triangle Park, North Carolina  27711

This Project Was Conducted in Association With the
  National Acid Precipitation Assessment Program

                  Prepared for
      U. S. Environmental Protection Agency
       Office of Research and Development
             Washington, O.C. 20460

-------
                                 TECHNICAL REPORT DATA
                           (Please read Instructions on the reverse before completi"
 I. REPORT NO.
 EPA-600/7-90-018
                            2.
                                                              PB90-27357U
                                                          v
 4. TITLE AND SUBTITLE
 Assessment of Control Technologies for Reducing
  Emissions of SO2 and NOX from Existing Coal-fired
  Utility Boilers
 7. AUTHOR(S)

 David  M. White and Mehdi Maibodi
                                                        5. REPORT DATE
                                                         September 1990
                                                        6. PERFORMING ORGANIZATION CODE
                                                        8. PERFORMING ORGANIZATION REPORT NO
9. PERFORMING ORGANIZATION NAME AND ADDRESS
 Radian Corporation
 P. O.  Box 13000
 Research Triangle Park, North Carolina  27709
                                                        10. PROGRAM ELEMENT NO.
                                                        11. CONTRACT/GRANT NO.
                                                         68-02-4286,  Task  84
 12. SPONSORING AGENCY NAME AND ADDRESS
 EPA.  Office of Research  and Development
   Air and Energy Engineering Research Laboratory
   Research Triangle Park, North Carolina  27711
                                                        13. TYPE OF REPORT AND PERIOD COVERED
                                                        Task final; 1/87 - 12/89
                                                        14. SPONSORING AGENCY CODE
                                                        EPA/600/13
 is.SUPPLEMENTARY NOTES AEERL project officer is Norman Kaplan, Mail Drop 62, 919/541-
 f\ r- ^ rt
 2556.
        \
 16. ABSTRACT-Vr,,             .        ,n  . ,   . .            ,     .      .          .. _    :
          The report reviews available  information and  estimated costs on 15 emission
 control technology categories applicable to existing coal-fired electric utility boilers.
 The categories include passive controls such as least emission  dispatching, conven-
 tional processes, and emerging technologies still undergoing pilot scale and commer-
 cial demonstration. The status of each technology i.s-reviewed relative to four ele-
 ments;-'(l)£Description--how the technology works;-*(2^Applicability--its applicability
 to existing plants;J(3^Performance--the expected emissions reduction; and-j(4)^:Costs
 --the capital cost,  busbar cost, and cost per ton of SO2 and NOx removed. Costs
 are estimated  for new and retrofit applications for various boiler sizes, operating
 ^Characteristics, fuel qualities,  and boiler retrofit difficulties. Capital costs vary
 from($2/kW for overfire air to(j£2800/kW for integrated gasification combined cycle
 in 1988 dollars.^NOTE: A major objective  of the National Acid Precipitation Asses-
 sment Program-- is to evaluate alternative methods for reducing  SO2  and NOx emis-
 sions from combustion sources and to  identify options which appear  most promising
 from both  an emissions reduction and cost  standpoint.  Part of this overall  effort is
 to develop up-to-date generic assessments of commercial, near-commercial and
 emerging emission control technology  categories applicable to these utility boilers.)
                              KEY WORDS AND DOCUMENT ANALYSIS
                 DESCRIPTORS
                                           b.lDENTIFIERS/OPEN ENDED TERMS
                                                                    c. COSATl Field/Group
 Pollution
 Utilities
 Boilers
 Coal
 Combustion
 Sulfur Dioxide
                    Nitrogen Oxides
                    Cost Effectiveness
                    Gasification
                    Emission
Pollution Control
Stationary Sources
13 B

13A
21D
21B
07B
.14A
14H, 07A
14G
 8. DISTRIBUTION STATEMENT
 Release to Public
                                           19. SECURITY CLASS (ThisReport)
                                           Unclassified
                                                                    21. NO. OF PAGES
                                           20. SECURITY CLASS (Thispage)
                                           Unclassified
                                                                     22. PRICE
EPA Form 2220-1 (9-73)

-------
                      NOTICE

This document has been reviewed in accordance with
U.S. Environmental Protection Agency policy and
approved for publication.  Mention of trade names
or commercial products does not constitute endorse-
ment or recommendation for use.

-------
                                  ABSTRACT
     A major objective of the National Acid Precipitation Assessment Program
 is to evaluate alternative methods for reducing S(L and NO  emissions from
 combustion sources and to identify those options which appear most promising
 from both an emissions reduction and cost standpoint.   Part of this overall
 effort is to develop up-to-date generic assessments of commercial, near-
 commercial, and emerging emission control technologies applicable to existing
 coal-fired electric utility boilers.  This report reviews available information
 and estimated costs on 15 technology categories, including passive controls
 such as least emission dispatching, conventional processes, and emerging
 technologies still undergoing pilot scale and commerical  demonstration.
     The status of each technology is reviewed relative to the following four
 elements:
     t    Description—how does the technology work?
     •    Applicability—what is its applicability to  existing plants?
     •    Performance--what is the expected emissions  reduction?
     t    Cost—what is the capital cost, busbar cost,  and cost
          per ton of StL and NO  removed?
     Cost estimates are presented for new and retrofit  applications for  various
boiler sizes,  operating characteristics,  fuel  qualities,  and boiler retrofit
difficulty.   Capital  costs vary from $2 per kilowatt for  Overfire Air to $2,800
per kilowatt for Integrated Gasification  Combined Cycle in 1988 dollars.

-------
                                CONTENTS
Abstract	;... j.'-;. . iii
Figures	-....' :X .  vi
Tabl es	!.<£ . y.f.'..  i x
Abbrevi ations and Symbol s	y.'... ;•->..  xi

   1.    Introduction and Summary		  1-1
         References	  1-12

   2.    Commercial Technologies	  2-1

         2.1  Fuel  Switching and Blending	  2-2
         2.2  Least Emissions Dispatch	  2-15
         2.3  Physical Coal Cleaning	  2-18
         2.4  NO  Combustion Control Technology	  2-28
         2.5  Thfowaway Wet FGD	  2-36
         2.6  By-Product Recovery FGD Technologies	  2-47
         2.7  Spray Drying	  2-57
         References		'.		  2-68

   3.    Near-Commercial Technologies	  3-1

         3.1  Integrated Gasification Combined Cycle	  3-2
         3.2  Fluidized Bed Combustion		  3-13
         3.3  Post-Combustion NO  Control...	  3-25
         3.4  Furnace Sorbent Injection	  3-33
         3.5  Low-Temperature Sorbent  Injection	  3-44
         3.6  Reburning	  3-51
         References	  3-56

   4.     Emerging Technologies	.4-1

         4.1  Advanced Coal Cleaning	  4-2
         4.2  Advanced Post-Combustion S0?/N0  Processes.....	  4-8
         References...	T.	  4-16
Appendices
         A.  Summary of Control Costs Coal Switching
               and Blending	 A-l
         B.  Low NO  Combustion Control Technologies	 B-l
         C.  Lime/Limestone FGD	 C-l
         D.  Integrated Gasification Combined Cycle	•	 D-l
         E.  Atmospheric Fluidized Bed		 E-l
         F.  Lime Spray Drying	—	 F-l
         G.  Selective Catalytic Reduction	 G-l
         H.  Furnace Sorbent Injection.	 H-l
         I.  Natural Gas Reburn	 1-1
            Preceding, page Wank

-------
                                    FIGURES

Figure                                                                  Page
1-1     Capital Costs  - Constant  1988  Dollars	  1-9
1-2     Levelized Annual Cost  - Constant  1988  Dollars		  1-10
1-3     Unit  Cost - Constant 1988  Dollars	  1-11
2.1-1   Effect of Coal Rank on Furnace Sizing	  2-3
2.1-2   Coal  Supply Regions	  2-7
2.1-3   Coal  Switching - Capital Cost	  2-12
2.1-4   Coal  Switching - Levelized  Annual Cost	  2-12
2.1-5   Coal  Switching - Cost  Per Ton  of S02 Removed	  2-13
2.3-1   Coal  Washability Curves	  2-20
2.3-2   Level 4 Coal Preparation Plant	  2-23
2.4-1   NOX Combustion Controls - Capital Cost..	  2-34
2.4-2   NO  Combustion Controls - Levelized Annual Cost	  2-34
2.4-3   NOX Combustion Controls - Cost Per Ton of NOX Removed	  2-35
2.5-1   Lime/Limestone FGD System Flow Diagram	  2-37
2.5-2   Simplified Flow Diagram for a  Dual Alkali
        FGD System	  2-39
2.5-3   L/LS  FGD - Capital Cost	  2-45
2.5-4   L/LS  FGD - Levelized Annual Cost	  2-45
2.5-5   L/LS  FGD - Cost Per Ton of S02  Removed	  2-46
2.6-1   Wellman-Lord Process Schematic	;	  2-48
2.6-2   Magnesia Slurry Process Schematic	  2-50
2.7-1   Lime  Spray Drying Process Flow  Diagram	  2-58
2.7-2   Lime  Spray Drying - Capital Cost	  2-64
                                     VI i
                                    —i

-------
                                    FIGURES
                                  (Continued)
   jure                                                                 Page
2.7-3   Lime Spray Drying  - Annualized Cost	 2-65
2.7-4   Lime Spray Drying  - Cost Per Ton of S02 Removed	 2-66
3.1-1   Generalized Block  Flow Diagram of Combined Cycle Coal
        Gasification Power Generation	 3-3
3.1-2   IGCC - Capital Cost	 3-11
3.1-3   IGCC - Level ized Annual Cost	 3-11
3.2-1   Simplified AFBC Process Flow Diagram	 3-14
3.2-2   AFBC - Capital Cost	 3-23
3.2-3   AFBC - Level ized Annual Cost	 3-23
3.3-1   Possible SCR Configurations	 3-26
3.3-2   Selective Catalytic Reduction - Capital Cost..	 3-31
3.3-3.   Selective Catalytic Reduction - Levelized Annual Cost	 3-31
3.3-4   Selective Catalytic Reduction - Cost Per Ton of NOX Removed... 3-32
3.4-1.   Simplified Schematic of Furnace Sorbent Injection	 3-34
3.4-2   Peak Sorbent Reactivity as a Function Of Temperature	 3-34
3.4-3   S02 Removal as a Function of Calcine Surface Area	 3-37
3.4-4   S02 Removal as a Function of Ca/S Ratio and Residence Time	3-37
3.4-5   Increase in Solids Loading as a Function Of Coal Sulfur
        Content and Ca/S Ratio for a Typical 10% Ash Coal	 3-40
3.4-6   Furnace Sorbent Injection - Capital  Cost	 3-42
3.4-7   Furnace Sorbent Injection - Level ized Annual Cost	 3-42
3.4-8   Furnace Sorbent Injection - Cost Per Ton of S02 Removed	 3-43
3.5-1   SO, Removal as a Function of Normalized Stoichiometric Ratio
        (NSR)	 3-45
                                    vii

-------
                                   FIGURES
                                  (Continued)
Figure                                                                  Page
3.6-1   Natural Gas Reburning - Capital Cost	  3-54
3.6-2   Natural Gas Reburning - Levelized Annual Cost	  3-54
3.6-3   Natural Gas Reburning - Cost Per Ton of NOX Removed	  3-55
4.2-1   E-Beam/Ammonia Process Flow Diagram	  4-9
4.2-2   Simplified Flow Diagram for the Fluidized-Bed Copper
        Oxide Process	  4-11

-------
                                   TABLES
Table                                                                   Page
1-1     Control Technologies  Reviewed	  1-2
1-2     Bases for Cost  Estimates	  1-4
1-3     Summary of Cost Resul ts	  1-8
2.1-1   Technical Factors Affecting Coal Switching	  2-4
2.1-2   Typical Delivered Coal  Properties for Selected
        Supply Regions	  2-8
2.1-3   Factors Influencing the Decision to Switch or
        Scrub to Reduce Sulfur  Dioxide Emissions	  2-14
2.3-1   Breakdown of Coal Cleaning Plants by Region
        and Level, 1978	  2-24
2.3-2   Potential Reductions  in Sulfur Dioxide Emission Rates from
        Physical Coal Cleaning	  2-24
2.3-3   Economics of Conventional Physical Coal Cleaning	  2-27
2.4-1   Site-Specific Parameters Affecting NO  Emissions from
        Combustion Modification Controls	  2-30
2.5-1   Total Operating and Planned Throwaway FGD Capacity
        by U. S. Electric Utilities (as of December 1985)	  2-42
2.6-1   Wellman-Lord Utility  Installations in the U. S.....	  2-52
2.6-2   Cost Estimates for Wellman-Lord FGD..	  2-55
3.1-1   Available S0~ Emissions Data for Cool Water
        Demonstration Program	  3-6
3.1-2   Design Sulfur Emissions Information for EPRI IGCC
        Studies	  3-8
3.1-3   Available NO  Emissions Data for Cool Water
        Demonstration Program	  3-9
3.1-4   Summary of Available Emission and Cost Data for IGCC
        Facilities	  3-10

-------
                                   TABLES
                                 (Continued)
Table                                                                  Page
3.2-1   Estimated Operating Data for 200 MW AFBC	 3-17
3.2-2   Full-Scale Utility FBC Demonstrations	:	 3-18
3.2-3   Comparative Economics of New 500 MW Conventional and FBC
        Power Plants3.'	 3-22
3.4-1   SO, Capture as a Function of Ca/S Ratio, Quench Rate, and
        Sofbent	 3-36
4.1-1   Economics of Advanced Physical Coal Cleaning	 4-7
4.1-2   Economics of Chemical Coal  Cleaning	 4-7
4.2-1   Cost Estimates for Electron Beam	 4-14
4.2-2   Cost Estimates for Copper Oxide	 4-15
                                    i  Y I
                                      x

-------
ABBREVIATIONS
                          ABBREVIATIONS AND SYMBOLS
ADVACATE
AEERL
AFBC
AFDC
ASTM
AUSM
CCR
CCTF
CF
CS/B +$5

CS/B +$15

DM
DOE
DSD
EER
EPRI
ESP
FBC
FF
FPD
FSI
GRI
HALT
HRSG
IAPCS
IGCC
kW
L/LS FGD
LED
LNC
LNB
LTSI
LSD
MHI
MW
NAPAP
NESCAUM
NGR
NSPS
advanced silicate process
Air and Energy Engineering Research Laboratory
atmospheric fluidized bed combustion
allowance for funds during construction
American Society for Testing and Materials
advanced utility simulation model
Conoco Coal Research
coal cleaning test facility
capacity factor
coal switching and blending at $5 coal cost
   differential
coal switching and blending at $15 coal cost
   differential
dense media
(U.S.) Department of Energy
duct spray drying
Energy and Environmental Research Corporation
Electric Power Research Institute
electrostatic precipitator
fluidized bed combustion
new fabric filter
fuel price differential
furnace sorbent injection
Gas Research Institute
hydrate addition at low temperature
heat recovery steam generator
integrated air pollution control system
integrated gasification combined cycle
kilowatt
lime/limestone flue gas desulfurization
least emissions dispatch
low NOx combustion
low NOx burners
low-temperature sorbent injection
lime spray drying
Mitsubishi  Heavy Industries
megawatt
National Acid Precipitation Assessment Program
Northeast States for Coordinated Air Use Management
natural gas reburn
new source performance standards
                                     xi

-------
                                ABBREVIATIONS
                                  (Continued)
NSR                --  normalized sto1chiometr1c ratio
OFA                --  overfire air
PCC                --  physical coal cleaning
PM                 --  particulate matter
PFBC               --  pressurized fluldized bed combustion
SCA                --  specific collection area
SCR                --  selective catalytic reduction
S.G.               --  specific gravity
STAR               --  state acid rain
U.S. EPA           --  U.S. Environmental Protection Agency
CONVERSION TO METRIC VALUES

Btu                --  1,054.8 Joules
Btu/lb             --  2,325.445 Joules/Kg

                              "3'
ft2/1000 acfm      --  0.3048 m2/actual m3/sec
gallon             --  0.0038 m

pound (Ib)         --  0.45359 Kg

lb/106Btu          --  0.0043 Kg/Joules

ton                --  907.18 Kg
                                    xii

-------
                          INTRODUCTION AND SUMMARY
BACKGROUND AND PURPOSE

     One of the objectives of the National Acid Precipitation Assessment
Program is to evaluate the potential performance and cost of alternative
methods for reducing S02 and NO  emissions from combustion sources.  Part of
this overall effort is to develop up-to-date generic information on commer-
cial, near-commercial, and emerging emission control technologies applicable
to coal-fired electric utility boilers.  This report presents a review of
available information on the technologies shown on Table 1-1.  Because the
various acid rain regulatory proposals focus on reduction of S0? and NO  in
                                                               Cf       rt
the eastern half of the United States, the report focuses primarily on each
technology's potential for retrofit onto existing boilers in the eastern
U. S. burning medium and high sulfur coals.

ORGANIZATION

     The technology reviews are divided into three major sections covering
technologies which are commercial (Section 2),  near-commercial  (Section 3),
and emerging (Section 4).  These three classes  are respectively defined,  as
follows: technologies routinely used by U. S. electric utilities,
technologies undergoing large-scale demonstration by U.  S.  utilities or
commercially used in Japan or Europe,  and those still  undergoing laboratory
or pilot-scale testing.  Designation of a technology to  one of these three
classes is based on the technology's demonstrated status on low and high
sulfur coals.
     Within each major section,  the technologies are presented in the
following order:  passive controls, pre-combustion controls,  combustion
controls,  post-combustion controls, and combined systems.  The term "passive
controls"  refers to technologies which in many  cases require little or no

                                     1-1

-------
                  TABLE  1-1.  CONTROL TECHNOLOGIES REVIEWED
Section
Technology
                                                        Potential Emission
                                                          Reductions  (%)
SO,
NO.
Commercial :
2.1
2.2
2.3
2.4
2.4
2.5
2.5
2.5
2.6
2.7
Near Commercial :
3.1

3.2
3.3
3.4
3.5
3.6
Emerging;
4.1
4.2
4.2

Fuel Switching and Blending
Least Emissions Dispatch
Physical Coal Cleaning
Low NO Burners
OverfiPe Air
Lime/Limestone FGO
Additive Enhanced FGD
Dual Alkali FGD
By-Product Recovery FGD
Spray Drying

Integrated Gasification
Combined Cycle
Fluidized Bed Combustion
Selective Catalytic Reduction
Furnace Sorbent Injection
Low-Temperature Sorbent Injection
Reburning

Advanced Coal Cleaning
Electron Beam Irradiation
Copper Oxide FGD

50-80
0-90
20-50
0
0
90-95
90-95
90-95
90-95
70-90

90-95

80-90
0
50-70
50-70
15-20

45-60
80-95
90-95

0-10
0-40
0
30-50
15-30
0
0
0
0
0

90-95

>50
80-90
0
0
35-50

0
55-90
90-95
FGD - Flue Gas Desulfurization.
                                     1-2

-------
 capital  expenditure  (i.e., hardware), but which will require changes in a
 utility's  operating  methods.  The status of each technology is reviewed
 relative to the  following four elements:
      •     Description—how does the technology work?
      •     Applicability--what is its applicability to existing plants
           burning low and high-sulfur coals?
      •     Performance—what is the expected emissions reduction?
      •     Cost—what are the capital cost, busbar cost, and cost
           per ton of S0? and NO  removed?
                       £       A
 Because  of the importance of consistent treatment of each technology, a
 consistent set of economic procedures was used for most technologies to
 allow comparisons.  The methodology used for this purpose is discussed
 below.

 METHODOLOGY

     Because of the diversity of plant sizes and designs,  operating charac-
 teristics, fuel quality, and financing arrangements found throughout U.  S.
 utilities, it was necessary to define a uniform methodology for use in the
 report.  These procedures can be divided into two major categories:  boiler
 design and economic assumptions.   Base case, high,  and low values were
 selected for boiler design and economic parameters.  The range in values was
 evaluated  to present boiler conditions which may favor the selection of  one
 technology option over another.   Table 1-2 presents the range of boiler
 design and economic assumptions  selected.
     For the technologies addressed  in this report, order of magnitude cost
 estimates  are presented.  Cost estimates presented  in the text are based on
 a range of boiler and coal  parameters.   Cost of many technologies is very
 site-specific and varies significantly depending on the boiler and coal
 characteristics.
     The Integrated Air Pollution Control  Systems (IAPCS)  (1)  cost model,
which is currently being updated  to  include more technologies,  was used  to
develop the cost  estimates  for some  of the technologies in this report.
                                     1-3

-------
                     TABLE  1-2.   BASES  FOR COST  ESTIMATES3
Parameter Descriptions
Base Case
Value
High Case
Value
Low Case
Value
                     Boiler and Coal Characteristic Assumptions

Unit:
  Size in MM                        300            100              700
  Capacity factor, %                 50             10               90
  Specific collection area          300            200.              400

Coal Characteristics:
  Sulfur content, %                 2.0            1.0              4.0
  Switched fuel sulfur content, %   0.9            0.9              0.9
  Ash content, %                   10.0            5.0             15.0
  High heating value, Btu/lb     11,000          9,000           13,000


                              Economic Assumptions

Capital Cost Indirects:
  General facilities, %             10%
  Engineering, %                    10%
  Project contingencies, %          30%
  Process contingencies, %           0-10%, commercial technologies
                                    10-30%, developing technologies
Retrofit factor (for FGD or SCR)    1.3            1.5              1.0
Economic Life    ...                  20             15               30
Carrying Charge Factor              0.189          0.205            0.175
O&M Levelizing Factor               1.57           1.45             1.75
Operating Costs:
Fuel price differential,
Operating labor, $/hr
Steam, S/1000 Ib
Electricity, mills/kWh
Lime, $/ton
Limestone, $/ton
Organic acid, $/ton
Ammonia, $/ton
SCR catalyst, $/ton
Waste disposal, $/ton
Water, 5/1000 gal.
Natural gas, $/10° Btu
Sulfur, $/ton

$/ton 10 15 5
21.4
7.0
57.0
60.0
16.0
1,725.0
150.0
20,300.0
10.0
0.65
2.0
65.0
 It is EPA's policy to use metric units.  However,  for the convenience of
 the reader non-metric units are used in this report.   Conversion factors
 to metric units are given in the table of Abbreviations and Symbols.

                                     1-4

-------
 The cost/performance assumptions are the same as used under the National
 Acid  Precipitation Assessment Program site-specific retrofit cost study
 under which the costs of retrofitting SO- and NO  controls at 200 coal-fired
 utility power plants are being estimated (2).  The IAPCS cost model was used
 to develop cost estimates for the following control technologies.
     •    Coal switching and blending (CS/B),
     •    Furnace sorbent injection (FSI),
     t    Lime spray drying with reuse of the existing electrostatic
          precipitators (LSD+ESP),
     •    Lime spray drying with a new fabric filter (LSD+FF),
     •    Lime/limestone FGD (L/LS FGD),
     t    Natural gas reburn (NGR),
     t    Low NO  burner (LNB),
     •    Overfire air (OFA),
     t    Selective catalytic reduction (SCR),
     •    Integrated gasification combined cycle (IGCC), and
     t    Atmospheric fluidized bed combustion (AFBC).
For the other technologies addressed in this report,  costs are from
referenced publications.   These costs are not included in this section for
comparison,  since other cost model assumptions were used in generating costs
which may not be consistent with assumptions used 1n  the IAPCS cost
estimates.

Economic Assumptions

     Cost estimates are presented in 1988 dollars using both current and
constant dollar procedures.   The Electric Power Research Institute's (EPRI)
general  costing procedures were  used to incorporate inflation, cost of
capital, and levelization of future expenses (3).  The  cost of replacement
power or lost capacity while a plant is out-of-service  during retrofit is
not included in the analysis.  Downtime replacement power costs depend on
the duration of the downtime period and the cost differential  between the
purchased or replaced electricity and the cost of power generated by the
out-of-service unit.   For example, assuming a power cost differential of
10.0 mills/kWh for three  different downtime periods of  1, 3,  and 6 months
                                     1-5

-------
with  a capacity  factor of  50 percent, the following additional capital
 investments would  be required:

      Downtime  Period               Downtime Replacement Power Costs
         (Months)                               ($/kW)
           1                                      4
           3                                     11
           6                                     22
New coal-fired plant cost  of power would be approximately 60 mills/kWhr with
half  the cost being fixed  costs and half the cost being fuel and consumable
costs.   For post combustion technologies the downtime replacement power cost
1s less of a factor than for In-situ technologies.   Constant dollar
calculations are based on  standard return on investment (i.e., annuity)
calculations without consideration of tax incentives (e.g., accelerated
depreciation, investment tax credits) or allowance  for funds used during
construction (AFDC).  The  cost calculations include a state and federal
income tax rate of 38 percent.
      The costs presented in the appendices are in current 198B dollars and a
30-year book .life.  To approximate the total level1zed busbar cost of power
in constant dollars, divide the current dollar costs by 1.75.

SUMMARY OF RESULTS

     Table 1-3 and Figures 1-1 through 1-3 summarize for each technology the
range of cost estimates developed in Table 1-2 using the high and low case
values.  The most representative value, the base case,  is shown on the
figures for each technology as a mid-way point on the bar graphs.  This is
to show the technology sensitivity to variation in  boiler and coal
characteristics and that there is no single "winner" for all retrofit
applications.
     Only those costs which were developed using the IAPCS cost model were
presented in this section  for consistency.  Cost estimates for other
technologies which were obtained from other references  are presented in the
respective technology sections.
                                     1-6

-------
      Sensitivity case cost estimates developed using the IAPCS cost model
 are  also presented  in the appendices.  The major cost parameters were
 varied  for  the  sensitivity analysis.  Major cost parameters differ for the
 different technologies.  For example, FGD costs are very sensitive to
 retrofit factors, coal sulfur content, capacity factor, and boiler size,
 while coal  switching is mainly a function of fuel price differential and
 percent reduction required.  A list of sensitivity case parameters for
 different technologies is summarized in Table 1-3.
     Figures 1-1 through 1-3 present cost estimates for both low and high
 cases for capital,  levelized annual, and unit costs.  Costs as well as
 pollutant removal efficiencies vary for different technologies.  These two
 factors should  be balanced in choosing one technology over another and
 determining the cheapest technology for meeting acid gas removal
 requirements for a  given boiler and coal characteristics.
     In this study  both high and low sulfur coals are switched to a 0.9
 percent West Virginia bituminous coal.  Therefore, for high sulfur coal (low
 case) over 80 percent S02 removal is achieved,  while for low sulfur coal
 (high case) the removal  value is less than 10 percent.   Because of a very
 low removal efficiency due to switching from one low sulfur coal to another
 low sulfur coal with less than 10 percent SO* removal,  the unit cost (dollar
 per ton of S02 removed)  resulted in a very large number (the division
 denominator was a very small  value for tons of S02 removal).  The AFBC and
 IGCC costs presented are for new systems.  The costs for these two
 technologies are much higher than for other technologies presented in this
 report because pulverized coal  boiler costs (equivalent to AFBC and IGCC)
 are not included with the other technologies.   For FSI, it 1s assumed that
 70 percent S02 removal  can be achieved with humidification and that existing
 ESPs are adequate in size and can be reused.   Therefore the major cost items
 are sorbent preparation  and modification of the existing furnace for sorbent
 injection.
     SCR costs are much  greater than other NOX removal  technologies.   This
 is mainly due to the initial  as well as the replaced catalyst cost.
However, unlike other N0x removal technologies,  SCR can achieve more than 80
percent NO  removal.
                                     1-7

-------
                          TABLE  1-3.   SUMMARY OF  COST  RESULTS  - CONSTANT 1988 DOLLARS

Emission Reduction
Technology


Commercial
Fuel Switching and Blending
Lime/Limestone fGD
Lime Spray Drying with
reuse of existing ESP
Lime Spray Drying with
new fabric filter
Lou NO Burners
Overfire Air

Near Commercial
Add-on Controls:
Furnace Sorbent Injection
Natural Gas Reburn
Selective Catalytic Reduction

Advanced Combustion Systems:
Integrated Casi.f ication
Combined Cycle*
Atmospheric Fluldlzed
Bed Combustion*
Capital Costs
Percent
SO
2

2-80
90
76

86

Q
0



70
15
0


95

90

NO
X

0
0
0

0

50
25



0
60
80


60-70

20-50

MU - size in megawatts, XS - coal sulfur content, CF
FPD - fuel price differential.
RF - retrofit
factor.
Lou


20
120
70

UO

B
2



25
10
90


1,710

1,560


-------
      3000
      2800
                   Legend
                 =  Base case vclue
                                                                            ICCC
to
o
o
<
o
      2600
      2400
      2200
      2000
      1800
1600
      1400
      1200
      1000
      800
       600
       400
      200
             CS/8
         . 0
                                                            LSD+ESP
                  20     30     40      50      60      70



                     PERCENT NOX OR SO2 REMOVED
                                                                       L/LSFGO
              Figure  1-1. Capital Costs  - Constant  1988 Dollars
                                          1-9

-------
    700
    600
M
O
u


I
    200  —
    100  —
               Legend
               Base cose value
             \

            10
                  20
I


30
                              *O    50     60    70


                       PERCENT NOX OR S02  REMOVED
      Figure  1-2.  Levelized Annual  Cost - Constant  1988 Dollars
                                   1-10

-------
o
CO
o
o
H
Z
               10
                     20
 30     40     50     60     70     80

PERCENT NOX OR S02 REMOVED
                                                                   100
             Figure 1-3.  Unit Cost - Constant  1988 Dollars
                                 i-ll

-------
References

1    Maibodi, M.,  A. L. Blackard, and R. J. Page.  Integrated Air Pollution
     Control System.  Draft Report.  U. S. Environmental Protection Agency,
     Research Triangle Park, North Carolina.  February 1990.

2.    Emmel, T.  E.  and M. Maibodi.  Retrofit Costs for SO- and NO  Control
     Options at 200 Coal-Fired Plants.  Draft Report.  UT S. Environmental
     Protection Agency, Research Triangle Park,  North Carolina.  December 1989.

3.    TAG - Technical Asesssment Guide. (Volume 1).  EPRI Report P-4463-SR.
     Electric Power Research Institute, Palo Alto, California, 1986.
                                       1-12

-------
                                  SECTION 2

                           COMMERCIAL TECHNOLOGIES
INTRODUCTION

     A variety of commercial technologies are available for reducing SOp and
NO  emissions from existing power plants.  These technologies include both
  A
passive (primarily associated with technologies which generally require 1ittle
or no new equipment) and active (based on installation of new equipment)
controls.  For purposes of presentation, these controls are divided into four
main groupings and represent the following six technology areas:

     •  Passive Controls
        - Fuel Switching and Blending (Section 2.1)
        - Least Emissions Dispatch (Section 2.2)

     •  Pre-Combustion Controls
        - Physical Coal Cleaning (Section 2.3)

     •  Combustion Controls
        - Combustion Modifications (Section 2.4)

     t  Post-Combustion Controls
        - Throwaway Wet Flue Gas Desulfurization (FGD) (Section 2.5)
        - By-product Recovery FGD (Section 2.6)

In addition, a discussion of combining these technologies to further reduce
total  plant emissions is presented in Section 2.7.
                                        2-1

-------
 2.1   F.UEL SWITCHING AND BLENDING

 Description

 Switch Ing--
     Coal switching represents one of several passive (non-hardware) control
 methods for emissions reduction.  The sulfur content of coals used in the U. S.
 ranges from less than 0.5 percent up to roughly 6 percent.  Thus, one
 alternative for reducing S02 emissions from coal combustion 1s for power plants
 to switch from burning high-sulfur coal to lower-sulfur coals.
     Most coal-burning facilities are designed to burn coals within a
 specified range of heating value, ash content, and other physical and chemical
 properties.  Largely as a result of empirical experience gained over time,
 boiler manufacturers have developed specialized knowledge of how to design a
 boiler for a given coal.  As shown in Figure 2.1-1, significant variations in
 boiler size and configuration exist among boilers designed for coals wiIn-
 different fuel characteristics.  Switching an existing boiler to a different
 coal, an area in which boiler manufacturers have had relatively little
 experience, can result in mismatching plant capabilities and may require major
 modifications to the plant.  Primary areas of concern include coal  handling and
 pulverization, combustion kinetics, ash deposition on heat transfer surfaces,
 particulate collection, and ash handling.  Major coal characteristics and
 hardware systems which must be considered in evaluating coal switching are
 listed in Table 2.1-1.
     The impact of these factors on a given boiler is coal and boiler specific.
 In general, switching to a coal of the same rank (I.e., conversion from a
 high-sulfur bituminous to a low-sulfur bituminous) will cause fewer problems
 than conversion to a lower rank coal.  However, other coal properties may still
 restrict fuel switching in individual boilers.  For example, most low-sulfur
coals have high fusion temperature ashes which do not slag at normal  furnace
temperatures, effectively precluding their use in cyclone and wet-bottom
boilers.
     Power plant systems other than the boiler may also be affected.   Of
concern in evaluating fuel switching as an S02 control option is that low-
                                        2-2

-------
                                                      Legend

                                       W  •   Width
                                       D  =   Depth
                                       h  •   Nose / Top Burner Row Distance
                                       H  •   Furnace Height
 WxD
               1.08WX 1,080       1.18WK1.08D        1.2BWK 1,240
                                                                     1.29W X 1.28D
EASTERN
 MIDWESTERN
 BITUMINOUS
SUB-BITUMINOUS
WILCOX SEAM

        TEXAS LIGNITE
YEGUA-JACK80N      NORTHERN PLAINS
                     LIGNITE
   Figure 2.1-1.  Effect  of coal  rank on furnace  sizing  (1).
                                     2-3

-------
         TABLE 2.1-1.  TECHNICAL  FACTORS AFFECTING COAL SWITCHING
     Change In
Coal Characteristics

Lower heating value
Higher moisture
Higher ash
Lower sulfur
Higher ash fusion
Higher sodium
and iron content
in ash
Higher volatility
Harder grindability
 Resulting Potential
 Qperatino Problems

Insufficient coal
  handling capacity
Unable to achieve design
  steam output

Longer/cooler flames
Lower furnace exit
  temperature

Higher gas flow

Increased particulates
  in flue gas
                        Increased solid waste
Lower particulate
  collection efficiency
Incompatible with cyclone
  and wet-bottom furnaces

Increased slagging and
  foul ing
Changed heat transfer
  characteristics
                        Heating and potential fires
                          in coal handling equipment
Insufficient pulverizer
  capacity
  Possible Solutions

  Enlarge coal  handling
    equipment
  Derate capacity
  Derate capacity
Increase/modify boiler
  heat transfer surface
    area
  Increase fan capacity

  Increase soot blowing
  Modify boiler convective
    section heat transfer
    area
  Increase ESP plate area
  Modify ash handling and
    disposal systems

  Increase ESP plate area
  Install  flue gas
    conditioning

  Use different coal
  Increase soot  blowing
  Use ash  additives
  Accept higher  forced
    outage rates

  Change boiler  tube
    distribution in
    furnace and  convective
    sections
  Modify pulverizers,
    silos, and other  coal
    handling equipment

  Increase pulverizer
    capacity
  Derate boiler  capacity
                                        2-4

-------
 sulfur coals generally have higher ash resistivity.  As a result, ESP
 performance is  likely to be poorer than with high-sulfur coals.  This may be
 especially critical  in boilers equipped with older ESP's which are only
 marginally in compliance with existing particulate standards.

 Blendlng--
     Coal blending to reduce the average sulfur content of coals is another
 strategy for compliance with existing SO- emission limits.  Depending on
 plant-specific  requirements, "blending" can denote either the use of multiple
 coals by a single plant even though each boiler at the plant may burn a single
 coal or the use of multiple coals in a single boiler.  In the first case,
 several distinct coal piles are maintained for supplying individual boilers.  A
 central 'coal receiving facility and certain other common equipment may be used
 to handle all  coals used at a plant, but most of the equipment connecting the
 coal pile to the boiler is unit specific.  The major limitation on retrofit
 application of  this form of coal blending is the availability of space for
 additional coal  piles.
     The second form of blending involves the use of two or more coals in a
 single boiler.  If the individual coals have significantly different combustion
 or ash characteristics (e.g., reactivity, ash fusion temperature, or base/acid
 ratio), failure to adequately blend the coals into a uniform mixture can result
 In rapid fluctuations in fuel composition being fed to the boiler and cause
major plant operational  problems.  Although a number of power plants receive
 coals from different suppliers, in most cases coals purchased by an individual
 plant come from the same coal producing area and have similar combustion and
 ash characteristics.   In these cases, blending operations may be fairly simple
 and relatively  inexpensive.
     When significant differences in fuel quality exist or when continuous
monitoring requirements on plant emissions impose short averaging times,
 additional coal-handling equipment (stack-out equipment,  storage piles,  reclaim
hoppers, conveyors, and belt scales) must be added to the plant to assure
 adequate mixing of the individual coals.   Capital and operating costs of this
equipment can be significant, as evidenced by Detroit Edison's $350 million
                                        2-5

-------
coal blending facility at  its Monroe Station.  The facility has the capability
of blending up to  10,000 tons/hr of high-sulfur bituminous and low-sulfur
subbiluminous coals.
     Combustion characteristics of coal blends cannot be accurately estimated
as simple linear averages  of the individual coals used.  This is especially
true for ash slagging, fouling, and resistivity which are influenced by
eutectlcs formed from the  mineral matter in the Individual coals in the blend.
The only way to be sure a  given coal or coal blend can be satisfactorily burned
in a given boiler  is to conduct a test burn of several days to several weeks  in
the actual boiler.  Preliminary testing of candidate blends in a pilot-scale
combustor can be useful in better defining potential problems prior to full-
scale tests.

Applicability

     Low-sulfur coals (<1  percent sulfur) are available in several  regions of
the U. S.:  central and southern Appalachia, Powder River Basin, southern
Wyoming, and Colorado/Utah (Figure 2.1-2).   Low-sulfur coals from southern
Appalachia are expensive to mine and limited In quantity,  and are therefore
unlikely to be used as an  alternative to high-sulfur coal  in existing boilers.
Basic coal characteristics for the remaining low-sulfur coals and for high-
sulfur coals from northern Appalachia and the Illinois Basin are presented in
Table 2.1-2.  On a pounds of SO. per million Btu basis, low-sulfur coals
generally have one-quarter to one-half the sulfur content  of the high-sulfur
coals.
     Central Appalachia coals are used widely in the eastern U.  S.,  accounting
for virtually all  of the coal used 1n Virginia, North Carolina,  South Carolina,
southern West Virginia, eastern Kentucky, and Tennessee.   Significant amounts
of central Appalachian coal are also used in Ohio and Michigan to meet SIP
requirements for low-sulfur coal.   Estimated reserves of low-sulfur coal  in the
region are 23 billion tons.  Production in 1982 was 118 million  tons.   Coal
seams In the region are generally thin (2.5 to 7 feet thick).   Mining and
reclamation in the region  is difficult due to the area's mountainous topography
and result in high mining costs.  Coal  prices FOB mine over the  last few years
have ranged from $30-40/ton (2,3).

                                        2-6

-------
ro
 i
          1  Northern Appalachia
          2  Central Applachia
          3  Southern Applachia
          4  Illinois Basin
          5  Western Interior
          6  Gulf Lignite
 7  Plains Lignite
 8  Powder River Basin
 9  Southern Wyoming
10  Colorado/Utah
11  Arizona/New Mexico
12  Washington
       SOURCE Bwgy Ventures Analysis. Inc., COALCAST Quartorty Report Service, 1983.
                                                 Figure  2.1-2.   Coal supply  regions (1).

-------
                                     TABLE 2.1.Z  TYPICAL DELIVERED COAL PROPERTIES FOR SELECTED SUPPLY REGIONS"
Property
Rank
Higher Heating range
Value, Btu/lb. average
Sulfur, X range
average
Moisture, X
Ash, X range
average
Crindabilltyb
Ash Fusion Temperature,
Fluidity, -F
Northern
Appalachia
Bituminous
11,000-13,000
12,138
0.7-5.0
2.5
5-10
7-15
12.8
50-100
2.100-2,800
Central
Appalachia
Bituminous
11,500-13,500
12,252
0.6-3.5
1.1
5-10
6-15
10.9
40-90
2,10Q-2,800
Illinois
Basin
Bituminous
10,000-12,000
11,114
1.0-3.5
2.8
5-15
8-15
10.5
50-70
1.900-2,400
Powder River
Basin
Subbi tumi nous
8,200-9.700
N/A
0.3-0.7
N/A
20-30
5-9
N/A
40-55
2,100-2,300
Southern
Wyoming
Subbi luminous
9,000-10,500
N/A
0.4-1.0
N/A
15-20
7-11
N/A
40-55
2,100-2,400
Colorado/
Utah .
Bituminous
10,000-12,500
11,130
0.3-0.8
0.5
7-20
6-13
9.3
45-50
2,200-2,600
*See Reference 1.
bHardgrove grindability index (AST* 0-409).

-------
     Western  low-sulfur  coals,  especially those from the Powder River Basin,
 are currently being  shipped to  utilities in the western, south-central, and
 midwestern U.  S.   Estimated reserves of low-sulfur coal exceed 140 billion
 tons.  Total  production  in 1982 was 157 million tons.  Coal seams range up to
 over 100 feet  thick  in the Powder River Basin where mining is by large pit-type
 surface mines.  Seams in southern Wyoming, Colorado, and Utah are generally
 thinner and are recovered by both surface and underground mines.  Coal prices
 FOB mine have  ranged from less  than $6/ton in the Powder River Basin up to
 $30/ton in Utah (1,3).
     Two major questions exist  regarding the future price of low-sulfur coals:
 the impact of  acid rain policies on 1) minemouth low-sulfur coal prices
 (especially in central Appalachia) and 2) transportation rates from the mine to
 the consumer.  Several studies  based on computer modeling have estimated that a
major acid rain control program could increase the demand for central
Appalachia low-sulfur coal by up to 50 million tons over otherwise projected
 levels and increase minemouth prices by as much as $15 per ton (4).   Minemouth
costs of low-sulfur western coals are not expected to increase as much due to
region's current excess mining capacity and larger reserves.   However, regard-
less of coal prices FOB mine, railroad transportation rates may increase the
delivered cost of low-sulfur coal and decrease the attractiveness of fuel
switching.  Other analysts argue that recent structural and technological
changes in the coal and railroad industries are not captured  in these modeling
efforts, and that price increases will be less than projected (1).   For
example, increases in labor productivity within the coal industry and the
declining demand for metallurgical coal  by the steel  industry have both worked
to keep the cost of coal  from central  Appalachia below forecast prices.

Performance

     In 1985 the 31 states adjoining or east of the Mississippi River contained
230,000 MW of coal-fired  capacity.  Of this total,  roughly 171,000 MW had S02
emissions in excess of 1.2 Ibs/million Btu.  Total  annual  SO^ emissions from
these units were 12.6 million tons.   An estimated 85,000 MW of this  total  are
believed to be technically capable of converting to low-sulfur coal.   The
remaining 86,000 MW include cyclone and wet-bottom furnaces unable to use low-

                                        2-9

-------
 sulfur coals due  to  high  ash-fusion  temperatures and minemouth plants with
 limited  transportation  alternatives.   If all of these units are converted to
 lower sulfur coals,  annual SO- emissions would decrease by roughly 4-7 million
 tons depending on the sulfur contents  of the coal fired in the units prior to
 being converted and  utility dispatch decisions.  The resulting increase  in the
demand for lower sulfur coals would  be roughly 125 million tons per year.
     The actual extent of fuel switching will partially depend on the nature of
 the SO* reduction program mandated.  Key factors impacting fuel switching
 benefits at a given  plant include the delivered price of low-sulfur coal, the
 sulfur and Btu content of the new fuel, the relative capital cost of FGD
 retrofit versus plant modifications  for fuel switching, plant-specific
 considerations such  as plant layout  and projected capacity factors, and the
 expected ease of implementing a major SO. control program throughout an
 integrated utility system.  Potential SO. reductions at a typical plant range
 from 50-80 percent.

Cost

     Fuel switching  and blending are generally assumed to have low capital
cost.  While in many instances the capital  cost for conversion between two
coals may be small,  this assumption  is not true in every instance.
Unfortunately,  many  analyses of SO- retrofit costs have Ignored the technical
limitations and capital costs associated with coal switching.
     The technical and economic feasibility of switching fuels in a given
boiler are specific  to the boiler design and location.  Where switching is
technically feasible, economics, will depend on the cost of coal delivered to
the plant, the capital and O&H costs associated with necessary plant
modifications,  the cost of replacement power associated with plant derating
when a different coal 1s used, and the comparative sulfur contents of the
existing and candidate coals.  In this study using the IAPCS cost model  (5),
estimated capital  costs are for particulate control  improvements and fuel price
differential  (FPD).  Particulate control  impacts were evaluated by considering
the altered particulate loading and particulate resistivity associated with the
replacement coal  as compared to original  conditions.  ESP upgrades were costed
                                       2-10

-------
by calculating additional plate area in the presence of SCL gas conditioning to
maintain current PM emission rates.  The premiums of $5 and $15 per ton were
used to span the range of fuel cost increased associated with low sulfur coal
demand.  Using the EPRI cost procedures (6), capital costs include
preproduction costs estimate as 25 percent of full capacity fuel cost for one
month.  This results in a higher capital costs for $15 FPD versus the $5 FPD.
The IAPCS costs do not include the cost impact of additional coal
receiving/storage/handling facilities (if needed) and the cost impact of boiler
derate due to pulverizer capacity and boiler fouling, slagging, and erosion.
The majority of the capital costs are due to coal inventory costs
(preproduction costs).
     The following table provides the range of values for capital cost,
annualized cost, and cost per ton of pollutant removed estimated using the
IAPCS cost model.  The lower capital and annualized costs are for a large unit
with high capacity factor and $5 FPD,  while the higher capital and annualized
costs are for a small unit with low capacity factor and $15 FPD.  Figures 2.1-3
through 2.1-5 show these costs as a function of boiler size and fuel price
differential.  Appendix A contains tables for estimating coal  switching costs
as a function of size, coal sulfur content, capacity factor and FPD.

                                                      Range
     Capital Cost ($/kW)                           19.1 to 43.1
     Annualized Cost (mills/kWh)                    5.4 to 17.2
     Cost Per ton of SO, Removed ($/ton)          213.7 to 20,354.1
                       e.
     The delivered cost of low-sulfur coals used in these calculations may not
include sufficient fuel premiums to account for rapid increases in low-sulfur
coal demand as a result of "acid rain" legislation.
     Table 2.1-3 summarizes key factors influencing the decision to switch coal
versus retrofit a wet scrubbing system or some other form of active emission
control technology.
                                       2-11

-------
                                                    Legend

                                            fuel Price Differentiol =  5 $/ton
                                            "uel Price Differential = 15 $/ton
                                            Sulfur Content      =  1 or 2 %
                                            Cooocity Factor     =  50 %
o
a
      18
        100
300
                             500
                                    MW
                     700
                                                 900
i.lOO
1.300
 Figure 2.1-3.  Coal Switching  - Capital Cost, Current 1988 Dollars

oc
111
Q.
V)
I




1 D -
15 -
14 -
13 -
12 -
1 1 -
10 -
9 -
8 -
7 -I
6 -
5 -
10


Legend
• Fuel Price Differentiol = 5 J/ton
+ Fuel Price Differentials 15 S/toi
Sulfur Content = 1 or 2 7,
Capacity Factor = 50 %


^^^^

1





0 300 500 700 900 1.100 1.300
MW
          Figure 2.1-4.  Coal Switching  -  Levelized Annual Cost,
                          Current  1988  Dollars
                                      2-12

-------
0.0
is -
18 -
16 -

14 -
13 -

12 -

1 1 -
10 -
9 -
8 J
7 -
6 -
5 -
4 -
3 -
1 -
T

~"~~ --—^^ .









i^^^

Legend

Stifir Content . 1%
• Fuel Price Differential = 5 $/ton
+ Fuel Price Differential = 15 $/ton

Stitur Content - 2%
o Fuel Price Differential = 5 $/ton
A Fuel Price Differential = 15 $/ton
Capacity Factor = 50 %







i i i i













1




100 300 500 700 1000 1300
                               MW
         Figure  2.1-5. Coal Switching - Cost Per Ton of, SO2
                      Removed, Current  1988 Dollars
                                2-13

-------
           TABLE 2.1-3.  FACTORS INFLUENCING THE DECISION TO SWITCH
                         OR SCRUB TO REDUCE SULFUR DIOXIDE EMISSIONS

Factor
Coal
Switching
Retrofit
Scrubbers
High-sulfur coal cost
Boiler type
High-sulfur coal supply
Low-sulfur coal transportation costs
Powerplant age
Difficulty of scrubber construction
Existing coal delivery facilities
Located In high-sulfur coal
      high*
   dry bottom
not restrictive
       low
       old
      high
   rail,  barge
     low*
    cyclone
  restrictive
     high
      new
      low
truck,  conveyor
producing state
Required SO. emission reduction
Capacity margin
System costs of derating
Load growth outlook
Utility financial condition
no
low
high
low
low
weak
yes
high
low
high
high
strong
Source:  Reference 1.
*This table 1s read in the following manner:   If the current cost of high
 sulfur coal delivered to the plant is high,  the utility is more likely to
 choose coal switching.  If the cost of the high sulfur coal is low,  the
 utility is more likely to choose a scrubber retrofit.
                                       2-14

-------
2.2   LEAST EMISSIONS DISPATCH

Description

     One of the key operational decisions facing a utility (or group of
interconnected utilities) on day-to-day basis is which plants to operate or
"dispatch" to satisfy the demand for electricity at any given time.  Presently,
utilities dispatch plants in accordance with state regulatory guidelines to
minimize the cost of generation, subject to physical constraints such as
individual plant availability, flexibility in adjusting to variations in
electrical demand, and plant maintenance requirements.  If environmental
regulations were changed to limit system-wide emissions of SO- or if emissions
were taxed to reflect the "cost" of emissions, the economics of plant dispatch
would be changed and a different dispatch pattern would potentially occur.
This alternate dispatch pattern has been referred to in various publications as
"least emissions dispatch" (LED) and "environmental dispatch".
     Several  different LED concepts have been described in the literature.   Key
differences in these concepts are the time period (temporary vs.  permanent) and
geographic area affected.  One proposal is to use LED to reduce annual
emissions over a large area.  In this approach,  maximum use is made of nuclear
and hydroelectric capacity and low-emitting fossil-fuel plants (natural  gas,
low-sulfur oil and coal, and plants with scrubbers for SCL reduction) in
preference to plants burning hi.gh-sulfur fuels.   Interregional transfers of
power are also used, including the importation of electricity from Canada.
     The other major LED approach would reduce SO- emissions only during
periods of adverse air quality in environmentally sensitive areas.   For
example, LED could be used during periods of air stagnation in the Adirondacks.
The objective of this approach is to reduce ambient concentrations of S02 in
selected areas as opposed to reducing total  emissions of all  plants in a large
region.  To be effective, this approach requires knowledge of source/receptor
relationships affecting pollutant transport and  transformation.   The only known
use of episodic LED has been in the Los Angeles  air basin where power plant
operations were controlled to reduce NOX emissions during periods of poor air
quality.  This effort is reported to have been a failure and was  discontinued
(7).

                                       2-15

-------
Applicability  and  Performance

     Because every utility and region of the U.S. is different, assessment of
the potential  and cost effectiveness of LED for reducing SO- is utility
specific.  For utilities operating both large, thermally efficient plants
equipped with  FGD systems and smaller, less efficient plants without pollution
control systems, least cost and least emissions dispatching may result in
operation of the same plants.  For utility systems which can import power from
other geographic regions or which depend on plants that are not equipped with
pollution control equipment for base-load power generation, LED may signifi-
cantly alter dispatch decisions.  Also, the utility's capacity factor will
influence whether LED can be effective.  Utilities with low capacity factors
may be able to alter dispatching significantly and thus appreciably reduce SO-
emissions.  Utilities with high capacity factors have limited opportunity
to shift generation between plants.  To the extent low-sulfur coal plants are
dispatched In  preference to high-sulfur coal plants, the impact of LED in coal
markets will be similar to coal switching.
     The greatest potential for LED is in a large utility or interconnected
utility pool having multiple plants, using a variety of fuels,  and having a
relatively low system-wide capacity factor.   Although the concept of LED has
been presented in a number of studies, very little analytical work has been
done to quantify its cost effectiveness or the magnitude of potential  emission
reductions.  Given the requirements of state regulatory agencies to minimize
the cost of electricity,  such analyses will  be required before LED can be
incorporated into or replace the existing dispatch process.

Cost

     LED does not involve capital  expenditures and only the differentials in
fuel  and variable O&M costs need to be considered in estimating the cost of
LED.   For example,  the cost differential
per kilowatt-hour from dispatching a plant burning $41/ton central Appalachia
coal  with 0.7% sulfur versus a $35/ton northern Appalachia coal with 2.4%
sulfur (assuming a 10,000 Btu/kWh  heating value and equal  O&M costs for both
                                       Z-16

-------
plants)  Is $0.0025  ($6 coal price differential divided by 24 million Btu per
ton).  This equates to a 30-year current 1988 dollar cost per ton of SO-
removed  of $334.
     Estimating the cost-effectiveness of LED for an entire utility network is
analytically similar to "bubbling" regional emissions.  The major analytical
difference with LED is that the focus is on altered dispatch to minimize SO-
rather than to minimize cost.  As with fuel switching, the cost of the first
ton of SO. reduction using LED is small.  As the number of altered dispatches
Increases, the incremental cost of LED can increase significantly.
     Two studies have been published on the cost-effectiveness of LED on a
regional basis.  These studies were conducted by the Northeast States for
Coordinated Air Use Management (NESCAUM) (7) and the Wisconsin Public Service
Commission (8) as part of EPA's State Acid Rain (STAR) program.
     The NESCAUM study examined SO- and NO  control alternatives for the region
containing New York, New Jersey, and the six New England states.  This
analysis, based on a regionalized version of EPA's Advanced Utility Simulation
Model (AUSM),  concluded that the region's utilities could reduce SO- emissions
by 35 percent (158,000 tons) and NO  emissions by 41 percent (51,900 tons)
                                   ^
through use of LED at an average current 1988 dollar cost of $4,893/ton.
     The Wisconsin study estimated that If 1) all power plants in the state
were centrally dispatched and 2) significant conservation efforts are
implemented (resulting in relatively low system capacity factors), LED could
reduce SO- emissions by roughly 15 percent (73,400 tons) at an average current
dollar cost of $494/ton of S02-  At higher system capacity factors,  the SO-
reduction potential of LED is reduced.
     The large difference in projected cost from these two studies indicates
the regional-  and utility-specific differences in LED effectiveness,  as well as
the significant difference in cost associated with various emission reduction
targets  (15 percent in Wisconsin versus 35 percent in New England).
                                       2-17

-------
2.3   PHYSICAL COAL CLEANING

Description

      Run-of-mine coal consists of  Individual particles containing varying
amounts of organic and  Inorganic matter.  The Inorganic materials are largely
sedimentary rock (e.g., shale, claystone, sandstone, and limestone) and pyritic
sulfur.  Physical coal  cleaning separates the combustible organic material
(i.e., coal) from the non-combustible impurities.  Although the primary
economic Impetus for most existing coal cleaning Is removal of ash, significant
benefits also accrue from the removal of sulfur.  Of the total coal consumed by
power plants, about 42  percent 1s cleaned using coal cleaning processes (9).
      For most medium- and high-sulfur bituminous coals, pyritic sulfur which
can potentially be removed by physical separation methods accounts for 30 to 70
percent of the total sulfur.  Most of the remaining sulfur is associated with
the organic structure of the coal and is not separable by physical means.   The
upper limit on sulfur removal depends on the physical liberation of pyrite from
the organic structure during coal crushing prior to cleaning.  Sulfur contents
of some Northern Appalachla coals (e.g., the Freeport and Kittaning seams of
Pennsylvania) can be reduced by 50-60 percent using currently available
cleaning methods, but removals of 30 percent or less are common for most coals.
     Most commercial techniques rely on differences In specific gravity to
accomplish separation.(10)  For example, clean bituminous coal has a specific
gravity (s.g.) of 1.2-1.3; shale, 2.0-2.7; and pyrite, 4.8-5.0.  Selective
adhesion, magnetism, and surface tension are also used.
      In cleaning processes based on specific gravity differences,  the raw coal
is separated into float and sink fractions by submerging it in a fluid with a
specific gravity between that of the coal  and the inorganics.  In  some
technologies, such as jigging and tabling, the differences in specific gravity
are combined with mechanical separation techniques to enhance the  recovery of a
clean coal  fraction.  However, the clean coal float will always contain some
impurities, and the rejected sink fraction will  contain some coal.  There are
two reasons for this.  First, inherent inefficiencies in separating processes
cause a fraction of heavy particles to be misplaced in the float instead of the
                                       2-18

-------
 sink, and vice versa for some light particles.  Second, the organic and
 inorganic matter cannot be totally separated because they frequently occur in
 the same particles.
     Yields and qualities of clean coal achievable for different specific
 gravities and particle top sizes can be estimated from laboratory washability
 data.  Figure 2.3-1 shows washability data for two coals.  The coal shown on
 the left is easily cleaned because, as shown on Curve A, the ash content of
 incremental amounts of coal recovered increases rapidly as the specific gravity
 of separation exceeds a given level, in this case roughly 1.5 (Point X).  As
 shown on Curve B, the cumulative ash content of the recovered coal at this
 specific gravity is less than 10 percent versus an original ash content of
 roughly 30 percent (as plotted on Curve B at 100 percent recovery).  The
 cumulative ash content of the sink material (Curve C) at s.g.=1.5 is roughly 65
 percent.  The right side of Figure 2.3-1 represents a difficult-to-clean coal
 for which there is no easily definable point of separation.  Commercial
 cleaning equipment is most efficient when separations are easily made.
     Because particles of different sizes behave differently when suspended in
 a fluid, technologies have been developed to clean each size fraction
 separately.  These technologies are based on three general  size classifi-
 cations:
     Fraction Classification            Typical Size Range
     Coarse                             greater than 3/8 inch
     Intermediate                       28 mesh (0.6 mm) to 3/8 inch
     Fine                               0 to 28 mesh (0.6 mm)
     Jigs and dense media (DM) vessels are the most common equipment used for
 coarse coal cleaning.  A jig feeds raw coal through a column of pulsating
water.  These pulsations suspend the lighter coal  particles at the top of the
jig while allowing the dense ash and pyrite particles to sink.  Dense media
 vessels use a suspension of fine magnetite particles (Fe304) and water to float
 all material below a certain s.g.  After separation, the refuse and clean coal
 are screened and rinsed with water to remove the magnetite.  Magnetite in the
 rinse water is reconcentrated using a magnetic separator.  Figure 2.3.1
Coarse coal cleaning accounts for one-half to two-thirds of all  coal cleaned in
 the U. S.
                                       2-19

-------
rv>
o
                 A- Easy-to-Clean Coal

      WEIGHT RECOVERY OF CLEAN COAL (%)
              SPECIFIC
              GRAVITY
          B-Difficult-to-Clean Coal

WEIGHT RECOVERY OF CLEAN COAL (%)
SPECIFIC
GRAVITY
         0   10  2O 30  40 50  60 70  80 90 100
                  ASH CONTENT (%)
                               1O 2O  3O 40  SO 60  70 80  90 10O
                                         ASH CONTENT (%)
          Curve A

     Incremental Ash Content
         In the Float
     Curve B                       Curve C

Cumulative Aeh In the Float      Cumulative Ash In the Sink
                              Near Gravity Material
                           Figure 2.3-1.  Coal washability curves. Reference 11.

-------
      Intermediate-sized particles are generally cleaned using concentrating
tables  and dense media cyclones.  A concentrating table is an inclined,
vibrating surface, equipped with diagonal riffles.  Raw coal slurry is fed to
an elevated corner of the table.  The riffles act as miniature jigs to stratify
the raw coal.  Lighter clean coal is carried by water over the riffles and off
the lower end of the table.  Impurities sink between the riffles and are
conveyed to the side of the table by the vibration.  DM cyclones use
centrifugal force to separate clean coal from refuse, much as gravity is used
in dense media vessels.  Because small particles settle slowly under gravity,
the much stronger centrifugal forces in a cyclone permit larger tonnages of
fine particles to be handled than in a DM vessel.  Intermediate coal cleaning
accounts for one-quarter to one-third of U.S. cleaned coal.
     Fine coal (less than 28 mesh) is difficult to handle because of problems
1n separating fine particles from the cleaning fluid.  The principal commercial
methods for cleaning fine coal  are froth flotation and fine coal  cyclones,
accounting for 5-10 percent of total production.  Froth flotation takes
advantage of the hydrophoblc (i.e.,  water repelling) tendencies of most coals,
versus ash which is hydrophilic (i.e., water adhering).  When air is bubbled up
through a flotation cell,  the small  coal particles attach to the air bubbles
and rise to the surface.
     Coal cleaning plants  are frequently classified based on the following
definitions (12):
     Level  1 - Crushing and particle sizing only.  Little or no cleaning
               takes place.
     Level  2 - Coarse (+3/8 inch) coal cleaning only.
     Level  3 - Coarse (+3/8 inch) and intermediate (3/8 inch to 28 mesh)
               coal  cleaning.  Material  less than 28 mesh is not  cleaned
               and is either incorporated with the product or discarded
               depending  on its quality.
     Level  4 - All  size fractions are cleaned,  but only a single  product
               consisting  of a  blend of cleaned fractions is sold.
     Level  5 - This level  involves the most rigorous coal  cleaning,  and
               produces two or  more  usable coal  products of different ash
               and sulfur  contents from one raw coal.

                                       2-21

-------
      This  numbering  Is generally  compatible with existing coal cleaning
 literature.  A  schematic  flow diagram for a Level 4 cleaning plant 1s shown in
 Figure 2.3-2.   The quantity of material processed as coarse, intermediate,
 and fine coal will depend upon the characteristics of the coal and the desired
 product quality.  The distribution of Level 2 through 4 cleaning plants by
 region and level  is  shown on Table 2.3-1.  Level 1 plants are omitted because
 of their minimal  effect on ash and sulfur levels of coal.  The only Level 5
 cleaning plant  in the U. S. is the Homer City cleaning plant In Pennsylvania.

 Applicability and Performance

     Most coal  cleaning plants in the U. S. were installed by coal companies to
 meet specific ash and sulfur specifications.  Approximately two-thirds of the
 coal in the eastern  U. S. is physically cleaned (12).  For the nation as a
 whole in 1980,  241 million out of 687 million tons of steam coal  were cleaned,
 or 35 percent.
     Table 2.3-2 provides average sulfur dioxide emission rates for major seams
 in the two major high-sulfur coal regions in the U. S.   Although not shown on
 the table, current coal cleaning plants (designed primarily to remove mineral
 matter) generally remove 20-30 percent of the sulfur 1n Illinois Basin coal  and
 from 10-40 percent of the sulfur from Northern Appalachia coal.  This level  of
 sulfur removal  is accomplished while recovering most of the heat content of the
 raw coal  (>95 percent), thus minimizing the dollar value of coal  lost during
cleaning.  Estimated emissions from state-of-the-art removal  are assumed to be
 representative  of cleaning in a Level 4 preparation plant.  In general,
 state-of-the-art coal cleaning is most effective for bituminous coals from
 North Appalachia  (especially those seams in the Allegheny Formation such as the
 Freeport and K1ttan1ng seams)  where sulfur reductions near 50 percent are
 possible and the Illinois Basin which contain significant amounts of
macroscopic pyrite.   Additional sulfur removal is possible,  but a substantial
portion of the heating value is lost in the reject material.
                                       2-22

-------









ro
i
u>








RAW
FEED
RRFAKPR

1
WET SCREEN AT
- 3/8 INCH
*
WET SCREEN AT
- 28 MESH
j
FROTH
FLOTATION
*









Raw Coarse
Coal

Intermediate Raw
Coal

CteanRne
Coal

fc| p 1 C

^L
Y
POND




JIG OR HEAVY
MEDIA WASHER

DENSE MEDIA CYCLONES
OR TABLES

VACUUM FILTER


RPPIJCP CM TPR








Clean Coarse MECHANICAL
Coal DEWATERING

Clean Intermedate MECHANICAL
Coal DEWATERING
^ 1 H ^^^^M»
S^\ THERMAL
^\J DRYER ~~
^

©CLEA
«b 4 * L. 4L. •!_ 1 »
Determine wrtdtner tne plant CO A
has open or closed water circuit .
O = Open
C a Closed
Figure 2.3-2.  Level 4 Coal  Preparation Plant
              Source:  Reference  1.

-------
              TABLE 2.3-1.  BREAKDOWN OF COAL CLEANING PLANTS
                              BY REGION AND LEVEL, I97B
Region
Northern Appalachia
Central Appalachia
Southern Appalachia
Midwest
Other

Total
Percent of Total
Total Plants3
128
239
25
53
19

464a

Level 2
35
47
6
23
--
.
Ill
25
Level 3
52
71
10
22
--
k
155°
35
Level 4
41
121
9
8
--
K
179°
40
Source: 1979 Keystone Coal Industry Manual. "Directory of Mechanical Coal
        Cleaning Plants."
uDoes not include 12 pneumatic preparation facilities.
 Only Includes plants in the four major regions.

  TABLE 2.3-2.  POTENTIAL REDUCTIONS IN SULFUR DIOXIDE EMISSION RATES FROM
                              PHYSICAL COAL CLEANING

Northern Appalachla
Pittsburgh
Sewickley
Freeport
Kittaning
Illinois Basin
Illinois 6/Kentucky 11
Illinois 5/Kentucky 9
i
Raw
Coal

5.7
6.2
3.9
4.8

6.5
6.5
State-of-
the-Art1

4.2
4.8
2.2
2.6

4.1
4.4
Maximum .
Achievable

3.3
3.8
1.3
1.7

3.6
3.9
-Based on 90% Btu recovery with 1 1/2" x
 Based on 14 mesh x 0 feed at s.g. = 1.4.
                                       x 0 feed.
                      Teed at s.g. = 1
Source:  Reference 14.
                                       2-24

-------
Cost

     Physical coal cleaning has many economic benefits besides sulfur removal,
although they can be difficult to quantify (15).  These include reduction in
coal transportation costs; reduced coal handling, pulverization, and storage
requirements; generally improved boiler performance; smaller ash handling and
waste disposal volumes; and reduced costs for post-combustion flue gas
desulfurization (FGO) to meet state and federal emission limits.  For utilities
having trouble meeting particulate emission limits, this lower ash content can
bring them into compliance and avoid expensive upgrading of ESPs.  While these
savings can be significant, they are hard to document quantitatively and vary
from plant to plant.  "Typical" cost benefits from coal cleaning have been
estimated at $2.82/ton, but can range from $0.54 to over $9/ton (10).  Cost
savings associated with use of coal cleaning to reduce FGD requirements are
discussed in Chapter 5.  Other benefits, such as improved boiler performance
and reduced coal handling costs, are very boiler- and site-specific and are not
included in the costs discussed below.
     The three major components of coal cleaning costs are the initial  capital
equipment,  operation and maintenance, and the lost heating value of discarded
coal.  The most important variables in determining these costs are the
characteristics of the raw coal and the desired level of cleaning (which in
turn determines the selection of cleaning technology and the quantity of Btu
rejected).
     Of the total  capital  costs, roughly one-third are associated with  the
cleaning equipment per se.  These costs (including equipment delivery and
installation) range from around $2,200 per ton of raw coal per hour for a jig
or DM vessel  processing coarse coal to nearly $22,000 per ton per hour for a
froth flotation cell for cleaning fine coal.   Another third is for structural
steel,  electrical, engineering and construction fees, and other costs that are
generally proportional to the size and complexity of the project design.  The
remaining third covers items such as refuse ponds, site preparation,  conveyors,
silos,  and  load out facilities for which costs are relatively independent of
the cleaning level.   On a total plant basis,  total capital cost estimates range
from around $11,000 per hourly ton of input capacity for a Level  2 plant to
                                       2-25

-------
near $43,000 per  hourly ton  for a Level 4 plant Incorporating froth flotation
for fine coal cleaning.
     Operating costs  increase substantially with higher levels of cleaning, as
well as with coal washability.  The operating cost for a coal circuit designed
for cleaning intermediate coal (3/8" x 28 mesh) may be double on a per ton
basis the cost of a circuit  for coarse coal (+ 3/8").  More complex plants
require more personnel and greater training in addition to power, supplies and
maintenance.  Magnetite costs can be very important in dense media plants.  The
heating value of  the  coal that is thrown away as refuse can be the highest cost
of coal cleaning, and 1s both a function of the level of cleaning and the coal
washability.
     As a result, the cost of sulfur removal using coal cleaning varies widely
depending on raw  coal quality and user specifications.  Table 2.3-3 shows
ranges of capital costs, O&M cost, and Btu loss for several Northern
Appalachian and Illinois Basin coals on a life-cycle basis (i.e., per ton of
cleaned coal).  Busbar costs and cost effectiveness are also shown.   These
values assume Level 2 and Level  4 cleaning of a raw coal  with a sulfur content
of 3.5 percent.    Btu recoveries were 90-96.5 percent, weight recoveries 73-90
percent, sulfur reductions 17-62 percent,  and raw coal cost $1.30/million Btu.
Although Level  2 cleaning has lower capital  and annualized costs than Level 4,
costs per ton of S02 removed Level 4 are frequently lower because of higher
sulfur removal  rates.  Cost per ton of S02 removed for lower sulfur coals
generally increase due to the decreased amount of removable pyrite in the raw
coal.
                                       2-26

-------
TABLE 2.3-3.   ECONOMICS OF CONVENTIONAL  PHYSICAL  COAL  CLEANING
   $/Ton of Clean Coal                    Current  1989  Dollars
   Capital                                  0.54  -   2.43
   0 & M                                   3.79  -   7.84
   Btu Loss                                1.50  -   6.50
   Total                                    5.83  -  16.77
   Busbar Cost  (mills/kwh)                  2.7   -   6.2
   $/ton  of S02                            325   -  595
                                2-27

-------
2.4 NOX COMBUSTION  CONTROL  TECHNOLOGY

     The production of NO   during combustion of coal (or other fuel) is
dependent on boiler design,  size, and operating parameters as well as fuel
composition.  By modifying  flame geometry and the rate of air/fuel mixing,
thermal oxidation of atmospheric and fuel nitrogen can be reduced.
Specifically, "staging" of  combustion to provide a fuel-rich (i.e., reducing)
zone during the initial, high-temperature stage of combustion will favor
formation of molecular N- instead of NO.  Additional (i.e., secondary) air is
added after the initial high-temperature reaction to complete combustion of
unburned fuel.  Secondary combustion air is added at a point where much of the
fuel-bound nitrogen has been reduced to N2 and flame temperatures are too low
for thermal NOX formation.
     Wall-fired wet-bottom  and cyclone units are large generators of NO  due to
their intense combustion conditions.  Tangentially-fired units with their
longer flames and cooler combustion temperature produce less NOX-  Because of
the wide variations in boiler characteristics and NO  levels, the applicabi-
lity of combustion modification technologies for reducing NO  is boiler
specific.  This section discusses the two major combustion modification
technologies for NO  control:  low NO  burners and overfire air.

Description
Low NO  Burners--
     Low NOV burners are designed to reduce NO  production (typically from wall
           A                                  A
fired boilers) by controlling air/fuel mixing.  This changes combustion
reactions in several ways.  First, substoichiometric amounts of air'are used
during initial combustion.  This results in fuel nitrogen forming molecular
nitrogen (N-) rather than NO.  Second, due to controlled addition and mixing
of secondary air, local zones of excess air that promote N0x formation are
minimized.  Third, Increasing combustion times and allowing for more heat
transfer from the flame results in a cooler flame and reduced production of
thermal NOX.
                                       2-28

-------
     The  controlled  air/fuel mixing  is achieved by using separate air registers
within  the burner.   Part of the air  enters with the fuel and another part
enters  through  an annulus  surrounding the central fuel/air stream.  The
remaining air is injected  through a  second, outer annulus.  Separating the air
and fuel delays fuel/air mixing, thus "staging" the combustion process and
creating the low NO  conditions described above.
     The same effect can be achieved on tangent1ally-f1red boilers, but a
different, somewhat more expensive burner design 1s required.  However, the
diagonal firing pattern and larger furnace dimensions of most tangentially-
fired boilers result in lower flame  temperatures and less NO  formation than
                                                            A
wall-fired boilers, thus the potential NO  reduction by low NO  burners is
                                         A             -       A •
reduced.  Because of this, overfire  air is the predominant commercial NO
control technique for tangentially-fired boilers.  Therefore, the subsequent
low NO  burner discussion will focus on wall-fired units only.
      A

Overfire Air--
     Another fundamental method of controlling flame stoichiometry, in addition
to (or  in place of) the localized combustion staging of low NO  burners,  is
                                                              A
combustion staging within the furnace volume using overfire air (OFA) ports
With this approach, 15 to 20 percent of the required combustion air is diverted
from active burners to OFA ports located above the top row of burners.  The
active burners operate fuel-rich,  providing a reduction in both thermal  and
fuel NO  generation.  The unburned fuel  escaping the fuel-rich flame zone burns
higher in the furnace where the diverted combustion air is mixed.   An overfire
air system includes the OFA ports (involving penetrations  in the  furnace  wall),
additional duct work, potentially a separate fan, and dampers/air flow
controls.

Applicability

     The applicability and effectiveness of low NOX burners and OFA, especially
for retrofit cases, are dependent on a number of site-specific parameters,  some
of which are listed in Table 2.4-1.   For example, NOX emissions increase  with
increasing coal  nitrogen and oxygen contents.
                                       2-29

-------
    TABLE 2.4-1.  SITE-SPECIFIC PARAMETERS AFFECTING NOV EMISSIONS FROM
                         COMBUSTION MODIFICATION CONTROLS
Furnace Characteristics             - Manufacturer
                                    - Size (width, depth, height)
                                    - Heat Input (Btu/hour)         ,
                                    - Heat Release Rate (Btu/hour-ft')
                                    - Firing Type (tangential, wall, cyclone)
                                    - Bottom Ash Removal (slag, dry)
                                    - Number, Spacing, and Size of Burners
                                    - Nose/Top Burner Row Distance
                                    - Gas Velocity
                                    - Furnace Exit Temperature/Attemperation
                                       Margins
                                    - Windbox Size
                                    - Excess Fan Capacity

Coal Characteristics                - Heating Value
                                    - Composition (N, 0, H.O, S, Ash, wt. %)
                                    - Ash Fusion Temperature
                                    - Ash Slagging/Fouling Indices
                                    - Volatility
                                    - Variability
Furnace design, size, and dimensions also affect NOX emissions, with low heat
release rate furnaces generally exhibiting lower NOX emissions.  Specific

reviews of the applicability of low NO  burners and overfire air are presented

below.
Low NOX Burners--
     Since the 1971 NSPS, low NO  burners have been installed on more than
40,000 HW of boiler capacity.  Most of these boilers have been new wall-fired

units with circular burners.  An additional  30,000 MW of pre-NSPS wall-fired

boilers are believed technically capable of being retrofit with low NOX
burners.  This 30,000 MW represents approximately 20 percent of nation-wide N0>

emissions from pre-NSPS utility boilers (16).
     Retrofit of low NO  burners to existing boilers is currently being

examined 1n several full-scale utility boilers.   Low NOX burners are generally
                                       2-30

-------
 larger  than conventional burners and require more precise control of fuel/air
 distribution.  Their performance depends partially on increasing the size of
 the combustion zone to accommodate longer flames while avoiding interaction
 with flames from other burners.  Because of this, low NO  burners are expected
 to be less effective when retrofit on relatively small furnaces.
     Low NOX burners are not currently available for dry-bottom cell-fired and
 roof-fired boilers, cyclone boilers, and wet-bottom boilers.  Cell burners
 consist of two or three burners in a single cluster and were designed
 to produce an intense, high-temperature flame.  Babcock & Wilcox is currently
 conducting a low NO  burner development program for cell burner applications
 for the Electric Power Research Institute (17).  Because of difficulties in
 applying low NO  burners to cyclone boilers, primary emphasis for reducing NO
               ^                                                             A
 emissions from these units is on reburning technology (see Section 3.7).
     In order to retrofit low NO  burners, the existing burners must be
 replaced.  In some cases, some of the water-wall tubes may have to be bent in
 order to install the larger low NO  burners.  Also,  low NO  burners will
                                  A-                       A
 generally produce a somewhat longer flame; thus, flame impingement on furnace
 walls and superheater tubes can be a problem for boilers with high heat release
 rates (typical of boilers built in the 1960's) (1).   If flame impingement is a
 problem, potential solutions include adjusting burner tilt, adjusting
 coal/primary air versus secondary air velocities, biased firing of burner rows,
 and relocating some superheater tubes.   Boilers with very small furnaces may
 have to be derated in order to prevent flame impingement at full load.
     Other factors to consider in retrofitting low N0¥ burners include fuel
                                                     A
 characteristics, furnace exit temperature, and fan capacity.  Fuel-rich
 operating conditions in the lower furnace region-associated with low NOX
 burners can increase the slagging tendency of the coal.   The generally longer
 flames of low NOX burners will tend to increase furnace exit and superheat/
 reheat tube temperatures.  Some low NO  burners operate with a higher pressure
                                      A
drop or may require slightly higher excess air levels at full load to ensure
good carbon burnout, thus increasing fan requirements.
                                       2-31

-------
     Another consideration in retrofitting low NO  burners is modifying the
windbox.  Modifications may include the addition of dampers and baffles for
better .control of combustion air flow to burner rows and combustion air
distribution to burners within a row.  Also, the windbox must be large enough
to accommodate the low NOX burners.  If the existing windbox requires
significant modifications to structural components, major re-piping, and/or
windbox replacement, retrofitting low NO  burners may not be feasible.

Overfire Air (OFA)--
     Overflre air has been installed on approximately 30,000 MW of new
tangentially-fired (T-fired) boiler capacity built since the 1971 NSPS.  In
addition an estimated 38,000 MW of pre-NSPS T-fired boilers are believed
technically suitable for OFA retrofit; these units account for an estimated 15
percent of nation-wide pre-NSPS NO  emissions (16).  OFA can also be used with
wall-fired boilers, but is less effective at controlling NOX emissions than
low-NO  burners.  As a result,  use of OFA is primarily applied to T-fired
boilers.
     With retrofit applications,  physical obstructions outside of the boiler
may restrict extension of the windbox and make installations of ductwork needed
to supply air to the OFA ports  difficult.  There must also be adequate distance
between the top burner row,  the OFA ports,  and the furnace exit for good carbon
burnout and maximum NO  reduction performance.   Extra fan capacity may also be
necessary.
     As in the case of low NOX  burners, increased furnace exit gas temperatures
and flame impingement are concerns.  To avoid these problems,  the distribution
of combustion air between the OFA ports and the burner registers must be
carefully controlled.  The Injection velocity required for good OFA mixing  with
the combustion gases and the additional pressure drop associated with the OFA
ductwork may require installation of a separate-OFA fan.
                                       2-32

-------
Performance

     In new boilers, low NO  burners can reduce uncontrolled NO  emissions by
                           "                                   «
40 to 60 percent.   Performance  in retrofit applications will depend on boiler-
specific considerations, but is generally expected to be somewhat poorer than
in a new boiler.
     Overfire air is generally  capable of a 15-30 percent NO  reduction, with
higher reductions possible in a few cases.  OFA can achieve the 1971 NSPS NO
emission limit for wall- and tangentially-fired utility boilers and 1978 NSPS
for NO  emission limit for tangentially-fired units.

Cost
     Low NO  burners and overfire air are relatively inexpensive control
           "
methods compared to post-combustion N0x control techniques.  NO  reduction
performance is the major uncertainty for a retrofit application.
     The following table presents the IAPCS estimated capital costs, annualized
costs and cost per ton of pollutant removed for OFA and LNB.  The lower capital
and annualized costs are for a large unit with a high capacity factor and the
higher capital and annualized costs are for a small unit with a low capacity
factor.  Figures 2.4-1 through 2.4.3 show these costs as a function of boiler
size for overfire air and low NO  burners.

     Basis for Range                                  Range
     Overfire Air -
       Capital Cost ($/kW)                          1.3 to 6.2
       Annualized Cost (mills/kWh)                  0.1 to 0.5
       Cost Per ton of NO  Removed ($/ton)         49.7 to 1,621.2
                         A
     Low NO  Burners -
       Capital Cost ($/kW)                          5.5 to 25.5
       Annualized Cost (mills/kWh)                  0.2 to  2.1
       Cost Per ton of N0y Removed ($/ton)         79.9 to 2,388.8
                         ^
Appendix B contains tables for estimating the cost of combustion controls as a
function of boiler size,  capacity factor and NO  reduction.
                                       2-33

-------
oc
in
a.
OT
                                                       Legend

                                                  •    Overfire Air
                                                  +    Lo* Nox Burners
                                                  Copocity Foctor  = 50%
       100
                   300
                                   MW
                                           700
                                                       1000
                                                                   1300
          Figure 2.4-1.  NOx  Combustion Controls  - Capital  Cost,
                          Current  1988 Dollars
1.2

1.1 -

  1 -

0.9 -

0.8 -

0.7 -

0.6 -

0.5 -

0.4 -

0.3

0.2 H

0.1
                                                       Legend
                                                  •    Overfire Air
                                                  •*•    Low Nox Burners
                                                  Capacity Factor  =  50%
       100
                   300
                               500
                                           700
                                                       1000
                                                               1300
                                    MW
           Figure  2.4-2.  NOx Combustion  Controls  -  Levellzed
                            Annual Cost,  Current  1988  Dollars
                                     2-34

-------
If
fc
                      Overfire Air - 10% NOx Reduction .
                      Overfire Air - 30% NOx Reduction
                      Low NOx Burners  - 40% NOx Reduction
                      Low NOx Burners  - 55% NOx Reduction
                        Copocity Factor- =  50%
        TOO
300
500
                                             700
                                     1000
1300
                                    MW
        Figure 2.4-3.  NOx Combustion  Controls  - Cost Per Ton
                         of  NOx Removed,  Current 1988 Dollars
                                       2-35

-------
 2.5   THROWAWAY  WET  FGD

      Throwaway  wet  FGD  technologies are the most widely used method of post-
 combustion  SO.  control  in  the U. S.  This section discusses the two major
 processes used  in the U. S.:  lime/limestone scrubbing and dual alkali   The
 use of additives to enhance lime/limestone performance is also discussed.
      Additional processes  are in use 1n Japan and Europe and have been
 pilot-scale tested  in the  U. S.  These processes Include Chiyoda Thorough-
 bred  121 and Dowa from  Japan and Saarberg-Holter from West Germany.  For
 information on  these processes refer to Reference 18.

 Description

 Lime/Limestone  Scrubbing--
      Lime/limestone scrubbing is the most common post-combustion S0? control
 technique currently applied to utility boilers.  A process flow diagram is
 shown in Figure 2.5-1.  There are four fundamental  steps to the process:
 preparation of  the Hme or limestone slurry, contacting the flue gas and
 slurry, reaction of the lime or limestone with S0_,  and removal of solid  waste.
     Reagent preparation involves crushing limestone or slaking lime to produce
 a slurry of fine hydrated lime or limestone particles.  Although many existing
 systems use lime,  most  future systems are expected to use limestone due to its
 lower cost.   Flyash from alkaline western coals has  also been used in
conjunction with lime and limestone.  The slurry is  fed to the scrubbing  loop
which consists of an absorber (usually a spray tower) and reaction tank;  the
 scrubber system for most utility plants consists of  multiple absorber/reaction
tank modules.   SO. 1s removed from the flue gas (which has already been cleaned
of flyash by an ESP or  fabric filter)  In the absorber while the reaction  tank
provides time for limestone dissolution and reaction with the dissolved S0?.
The product of this reaction is a mixture of crystalline calcium sulfate  and
calcium sulfite.  Small  droplets of slurry which become entrained in the  flue
gas are captured by a mist eliminator.   Reacted solids are removed and
dewatered with the clear liquor being returned to the process.   The solid waste
is either ponded or landfilled.
                                       2-36

-------
                            Stack Gas
       Flue Gas
  Organic
Acid Addition
 (Optional)
 Reagent
Preparation
 Lime or
Limestone
                             Absorber
Slurry
          1
             Reaction
              Tank
                                                   Slurry
                                          Clear Liquor
  Solids
Preparation
                                                              T
                                            Waste
                                            Disposal
   Figure 2.5-1.  Lime / limestone  FGD system flow  diagram.
                                2-37

-------
     During the  initial  applications of lime/limestone FGO systems a variety of
chemistry problems were  encountered, often with the result being rock-like
gypsum  (calcium  sulfate) scale formation on equipment.  These problems have
been overcome through experience and better understanding of the complex
chemical interactions.   Today, most lime/limestone FGD problems involve
adjusting the chemistry  to optimize raw material utilization and SO* removal.
However, materials selection continues to be a problem.  To reduce impacts on
boiler operation, scrubber systems are generally designed with an extra
absorber module which can be operated when any of the other modules are down
for maintenance or repair.
     One technique for Improving the performance of limestone FGD systems is
the use of organic acid  additives, such as adipic and other dibasic acids.
This improves SO. removal and enhances other aspects of the process,  but does
not greatly increase operating costs.  As a result of EPA Air and Energy
Engineering Research Laboratory (EPA-AEERL) sponsorship of development efforts,
this technique has been  successfully used in retrofit applications at several
utility Installations.  Use of organic acids allows existing FGD systems to
improve system reliability and SO- removal without affecting other aspects of
the process or requiring significant additional S0» control equipment.

Dual Alkali  Scrubbing--
     Dual  alkali scrubbing was developed to avoid the problems of erosion,
scaling, and solids deposition found with lime/limestone FGD.   As the name
implies, two alkalis are used:  a clear solution (generally sodium sulfite)  in
the absorber followed by addition of lime in the reaction tank to regenerate
the spent solution for recycle to. the absorber.  The resulting calcium
sulfite/sulfate sludge is then dewatered and landfilled.   Makeup sodium
(typically soda ash) Is added to the regenerated solution to replace  residual
sodium lost in the filter cake.
     Several variations of the dual alkali process have been investigated
(e.g.,  using ammonia and potassium as the first-stage alkali).   However, the
use of sodium is the most advanced and is the only process in  commercial use in
the U.  S.   A simplified diagram of the sodium/lime process is  shown in
Figure 2.5-2.
                                       2-38

-------
                                                                                  Lime silo
Flue gas
by-pass
                                                      Scrubbed gas to
                                                      upper stack breechlngs
                                                      Clean gas to lower
                                                      stack breechlngs
CA(OH)2
Flue gas
from
electrostatic
preclpltator
              Soda Ash
                     CaO
                            Water
                        Figure 2.5-2.   Simplified flow diagram for a dual akali FGD system
                                           Source:  Reference.  1.
                    Solid to on-slte
                        landfill

-------
     Sodium-based, dual alkali systems are more suitable for high-sulfur than
 for  low-sulfur  (less than  1.5 percent) coal applications.  With low-
 sulfur coals, a greater percentage of the sulfHe is oxidized to sulfate.  This
 lowers the available alkalinity in the scrubber and, thus, lowers S02 removal
 capacity.

 Additive-Enhanced FGD--
     Several additives have been added to lime/limestone FGD systems to improve
 SO.  removal, sorbent utilization, and/or system operation.  These include:
 adiplc acid, mixtures of dibasic organic acids (DBA), magnesium, and
 thiosulfate.  The decision to use a specific additive depends upon system
 chemistry and performance.    In general, except for the additive feed system,
 there are no major hardware differences compared to a conventional lime/
 limestone system.
     Organic acids such as adipic or DBA are currently the most widely used
 additives.  These acids buffer the gas/liquid contacting interface at a pH  of 3
 to 5, thus improving the driving force for S02 removal.  .For example, addition
 of only 2000 ppm of adipic acid can more than double the normal dissolved
 alkalinity of the absorbing limestone slurry, and thus can compensate for
 inadequate gas/liquid contact in the absorber.  Organic  acids buffering also
 allows operation at lower pH than limestone-only systems,  without loss of SO.
 removal capability.  This results in higher limestone utilization which can,  in
 some cases,  reduce mist eliminator scaling problems.
     Magnesium-promoted FGD involves addition of magnesium-containing (i.e.
dolomitic) lime or magnesium sulfate (MgSO.)  to either a lime or limestone
 system.  Magnesium salts are more soluble than the corresponding calcium salts,
 providing higher liquid-phase alkalinity and increased absorption of SO..
     Sodium thiosulfate (Na.S.O.)  addition has been used to reduce scaling  in
 both lime and limestone systems and to improve sludge dewatering.   Sodium
thiosulfate is also a very active inhibitor of sulflte oxidation;  as little as
 100 ppm S203 can virtually halt oxidation of calcium sulfite to calcium
 sulfate.   However, consumption of this additive can be high due to degradation
                                       2-40

-------
 of  thiosulfate  to  other  species,  such  as trithionate, which do not  inhibit
 oxidation.  This approach was commerciany demonstrated in 1985 at  Seminole
 Electric  Cooperative's plant near Palatka, Florida.

 Applicability

      Lime, limestone, sodium carbonate, dual alkali, and additive-enhanced FGD
 are commercially available and have been widely applied to utility  boilers.
 The total capacity of operating and contracted throwaway FGD systems as of
 December  1985 exceeds 68,000 MW as shown in Table 2.5-1.  There is  currently an
 estimated 4,000 MW of scrubber capacity using additives (19).  From a process
 viewpoint, lime/limestone FGD is  applicable to all boilers.  Dual alkali
 systems are best suited to boilers burning high-sulfur and high-chlorine coals.
      Depending upon the arrangement of the boiler and associated equipment, a
 scrubber retrofit may be simple or difficult.  Difficult retrofits often
 require either long runs or complex arrangements of ductwork, which can be very
 expensive and time consuming to install.  In addition to the absorber and
 reaction tank in which SO* removal actually occurs, additional  space is also
 required for sorbent receiving and storage, slurry preparation, dewatering
 of reacted sorbent, and solid waste disposal.  Replacement or relining of the
 stack may also be required to eliminate corrosion problems caused by the wet
 flue gas.  Modification of the boiler may also be required to balance and
 control pressure changes caused by retrofitting a scrubber and additional
 fans.  The high capital cost of the equipment combined with retrofit
 difficulties makes it very expensive to apply lime/limestone FGD on plants
 operating at low capacity factors or with little remaining plant life.
      In the U. S.,  additive enhancement is currently used solely as a retrofit
method for upgrading the performance of existing FGD systems.  The minimal
 changes in system design required for addition of additives make retrofit
 relatively easy.  Future systems may be designed with additives to reduce
capital and operating costs and to boost FGD removal  efficencies well above
90%.
                                       2-41

-------
       TABLE 2.5-1.  TOTAL OPERATING AND PLANNED THROWAWAY FGD CAPACITY
                     BY U. S. ELECTRIC UTILITIES (as of December 1985)
                                           Units
                    Capacity Factor
Limestone

  In operation
  Under construction
  Contracts awarded

Lime
 60
  9
_8
 77
 26,008
 17,113
  1.505
 36,977
  In operation
  Under construction
  Contracts awarded
Sodium Carbonate

  In operation
  Under construction
  Contracts awarded
Dual Alkali

  In operation
  Under construction
  Contracts awarded
Total
  6
  1
 _2
  9
150
                          17,113
                               0
                           2.036
                          19,149
  1,505
    550
  1.100
  3,155
  1,963
    265
	0
  2.228

 68,072
Source: Reference 19.
                                       2-42

-------
Performance

     Performance of FGD systems is measured with respect to both SO- removal
efficiency and system availability.  In general, both lime/limestone and
dual alkali are capable of up to 95 percent SCL removal with careful design and
operation.  Removals of 90 percent are more common.  As discussed earlier,
additives such as adipic acid frequently double the dissolved alkalinity of
limestone slurries.  This improvement in alkalinity can reduce SO- emissions
from existing scrubbers by a factor of two or more while at the same time
increasing unit availability.
     In general, lime/limestone scrubbers tend to be more chemically complex
in plants using high-sulfur coals versus low-sulfur coals, and thus availabi-
lity is lower.  Addition of spare absorber modules can improve overall system
availability by providing backup units and allowing more time for maintenance
and repair of each module.  However, new scrubber design and the use of
additives reduce the need for module sparing in the future.

Cost

     The range of estimated costs using the IAPCS cost model for both new and
retrofit lime/limestone FGD systems applied to coal-fired utility boilers are
presented below.  Depending on local sorbent availability, waste disposal
options, and other site-specific factors, these costs may vary somewhat,  but
will generally be within the range shown.  The lower capital and annualized
costs are for large units with high capacity factors, low retrofit factors and
low sulfur coal  content, while the higher capital  and annualized costs are for
small units with low capacity factors,  high retrofit factors and high sulfur
coal content.  Appendix C contains tables of costs as a function of boiler
size, capacity factor,  coal  sulfur and retrofit difficulty.

  Basis for Range                                  Range
  Capital Cost ($/kW)                             125 to 628.5
  Annualized Cost (mills/kWh)                     11 to  88.6
  Cost Per ton of S02  Removed ($/ton)           508.3 to 10,618.6

     These costs represent conventional  design practices for new installations
relative to the size of absorbers and  installation of backup systems to assure
                                    2-43

-------
 compliance with New  Source  Performance Standard (NSPS) reliability
 requirements.  If these design practices are altered to comply with less
 stringent reliability requirements, larger—but fewer—absorbers might be
 installed.  Analysis of cost  savings at several plants using EPA's IAPCS
 computerized FGD costing model found this altered design philosophy could
 reduce capital cost  by 16-32  percent and total annual costs by 9-12 percent
 relative to conventional designs.  Under the Department of Energy Clean Coal
 program, the wet limestone  Mitsubishi FGD process was selected for commerical
 demonstration.  The  program will demonstrate the technical and economic
 feasibility of an advanced, single, 600 MW absorber system to obtain 90-95% SOp
 removal on high sulfur (>3  %  S)  Indiana/Illinois coals.
     The costs and benefits from additive enhancement depends in large part on
 the performance of the existing  FGD system.  For a limestone FGD unit operating
 at 90 percent SCL removal,  but at high liquid-to-gas absorber ratios, use of
 organic acids can significantly  reduce operating (i.e., slurry pumping) costs
 while maintaining the same  SO. removal; EPA's IAPCS model  estimates a total
 cost savings of roughly six percent at these conditions.   For a unit needing to
 increase its SO- removal  efficiency,  the economic advantages of adding organic
 acids may be significantly  higher.  In general, the capital costs of converting
 an existing limestone FGD system to use these additives are minimal and quickly
 pay for themselves if improved system reliability is achieved.
     Engineering estimates  of capital and fixed O&M costs  for dual  alkali
 systems are generally lower than for lime/limestone systems (1).   However,
 total costs (including variable O&M)  are similar due to the high  cost of lime.
 Enhancement of the dual  alkali process to allow use of limestone  rather than
 lime for regenerating the scrubber solution could significantly reduce the  cost
 of the dual  alkali process.  Detailed calculations of dual  alkali costs are not
 included in this report.
     The costs for a specific FGD system are a function of many variables
 including unit size,  coal  sulfur level, percent SO. removal,  and  load factor.
 As previously mentioned,  the remaining llfespan of a plant is also important.
 An operator does not want to make large capital Investments in  a  plant that has
 a short remaining lifetime.  Also, smaller plants (less than  100  MW)  are much
more expensive per kW to retrofit with scrubbers than larger plants.
 Figure 2.5-3 through 2.5-5 present costs as a function of  plant size and coal
 sulfur content.
                                    2-44

-------
O
O.
to
C
UJ
QL
(A
                                                          Legend
                                                         •    1% s
                                                         +    2% S
                                                         o    3% S
                                                         &    4% S
                                                      Capacity Factor = 50%
                                                   Limestone Cost = $l7.3/ton
                                                      Retrofit Factor = 1.0
         100
300
500
                                    MW
700
900
1,100
1,300
     Figure 2.5-3.  L/LS  FGD -  Capital Cost, Current  1988  Dollars
                                                          Legend
                                      +
                                      o
                                                             1% s
                                                             2% S
                                                             3% S
                                                             4% S
                                                     Copacity Factor = 50%
                                                   Limestone Cost = $17.3/ton
                                                      Retrofit Factor = 1.0
       12
         100
300
500
                                        700
          900
          1,100.
          1.300
                                    MW
            Figure 2.5-4.  L/LS  FGD -  Levelized Annual  Cost,
                             Current  1988 Dollars
                                        2-45

-------
0.5
   100
             300
                                                  Legend

                                                  •   1% s
                                                  +  • 2% S
                                                  O   3% S
                                                  a   4% 5
                                              Copacity Factor = 50%
                                            Limestone Cost  = $17.3/ton
                                              Retrofit Factor = 1.0
1.100
1,300
  Figure  2.5-5.  L/LS FGD - Cost Per  Ton  of S02  Removed,
                  Current  1988 Dollars
                                2-46

-------
 2.6   BY-PRODUCT RECOVERY  FGD  TECHNOLOGIES

 Description

      Recovery  of sulfur from  SO-  as a useful product rather than discarding  it
 is beneficial  with  respect to resource conservation, and also because of the
 potential  improvement  in  process  economics.  Numerous processes have been
 proposed  in an  effort  to  achieve  these goals.  These processes vary widely in
 sorbent used,  product  made, and stage of development.  Currently, three types
 of processes have attained commercial status.

      t     Gypsum recovery - Conventional FGD system with forced oxidation
           and  additional  process  features to produce salable gypsum.
      t     Wellman-Lord -  Reaction of sodium sulflte with S02 to form sodium
           bisulfite followed by steam stripping to recover sulfuric acid or
           elemental sulfur.
      t     Magnesia  slurry - Reaction of MgO/Mg(OH)- with S02 to form magnesia
           sulfite which is then separated, dewatered, dried, and calcined to
           recover sulfuric acid.

 Gypsum Recovery--
      In general, wet lime/limestone FGD requires several process enhancements
 to supply  consistently pure gypsum.  These enhancements include forced
 oxidation  to produce the required gypsum content in the slurry solids,  particle
 agglomeration,  and washing of gypsum filter cake.  The gypsum slurry is then
 dewatered  using  conventional equipment,  resulting in a cake that is greater
 than  80 weight  percent solids (20, 21).   Approximately 2.7 tons of gypsum is
 produced per ton of S02 removed from the flue gas (22).  Washing of product
 gypsum, especially with high-chloride coals, generates a liquid waste stream.

Wellman-Lord--
     The main features of the Wellman-Lord process (Figure 2.6-1) are the use
 of a  sodium-based sorbent, regeneration of the sorbent solution, and conversion
of S02 to  sulfuric acid or elemental sulfur.  The scrubbing portion
                                       2-47

-------
     Doubte-Loop
      Absorber
oo
                                                                                                               S(X Rich Stream
                                                                                                               lo Elementate or
                                                                                                                SuHuric Acid
                                                                                                             Low
                                                                                                       •   Pressure
                                                                                                            Steam
                                                  Sodium SulfalB
                                                  Solids Purge
                                    Figure 2.6-1. Wellman-Lord Process Schematic

-------
of the process  involves circulating sodium sulfite solution, typically in a
plate-type absorber, to absorb SO..  Sodium bisulfite is the primary reaction
product, though some oxidation to sodium sulfate will occur depending on the
flue gas oxygen content.  This sulfate is purged as a solid, using a
crystallizer followed by drying.  The sodium bisulfite solution from the
absorber is thermally decomposed in a steam-heated evaporator to regenerate
sodium sulfite.  Solids formed due to evaporation are redissolved in a separate
vessel; at this point make-up soda ash (Na^CO,) is added as necessary.
Overhead vapor from the evaporator is an SOg-rich stream, suitable for
conversion to elemental sulfur or sulfuric acid using commercial technology.  A
purge stream requiring treatment and disposal is produced in order to keep
thiosulfate levels (i.e., dissolved solids) below a threshold.

Magnesia Slurry--
     In the magnesia slurry process (Figure 2.6-2), an aqueous slurry of
regenerable magnesium hydroxide and magnesium sulfite is used to absorb SO.
from flue gas.  Reaction products formed in the absorber include several
magnesium sulfite hydrates and magnesium bisulfite.  A portion of the reacted
slurry from the main scrubbing loop is sent to a thickener for partial
dewatering.  A centrifuge is then used to recover a wet cake of magnesium
sulflte/sulfate crystals.  The crystals are sent to a rotary or fluidized bed
dryer while the liquor is returned to the circulation loop.   Following the
solids drying step, decomposition of the magnesium sulfite/sulfate compounds is
carried out in a high-temperature calciner.  This liberates  a concentrated SO.
gas stream that can be routed to the sulfur (or sulfuric acid) production unit.
The calcined solids are regenerated as MgO for recycle to the absorber loop.

Applicability

Gypsum Recovery--
     Gypsum recovery technology can be used almost anywhere  ordinary lime/
limestone scrubbing is appropriate.   It is only necessary to add forced
oxidation,  product solids washing (to assure adequate gypsum purity), and
                                       2-49

-------
                                             Rehealar
    SO, Absorber
FhjeGas—»
                                                                                                     To Slack
                                                                    Coke
                                                                                Catelner
                             Figure 2. 6-2.  Magnesia Slurry Process Schematic (23)

-------
shipping facilities.  The primary commercial use of this technology is to
produce gypsum for wall board or as a cement retarder.  Approximately 20 forced
oxidation units are operating in the U. S.  However, only two plants have plans
to recover gypsum for sale.  In Japan, FGD gypsum is routinely substituted for
natural gypsum, and practically all the lime/limestone FGD installations in
Europe successfully market recovered gypsum.  The main motivation for
by-product utilization overseas is lack of space and high cost for waste
disposal.
     Annual  consumption of raw gypsum in the U. S. was approximately 20 million
tons as of 1984.  The wallboard industry consumes about 70 percent of this
amount, with the balance used in the manufacture of cement and plaster, and in
agriculture (22).    Wallboard manufacturers historically considered only purity
and grain-size; however, for recovered FGD gypsum other factors must be
considered.   For example, as soluble salt concentration increases the gypsum
calcination temperature decreases, which may result in weakening of the bond
between the paper coating and gypsum core.

Wellman-Lord--
     The Wellman-Lord process has been retrofit on coal-fired utility boilers
in both the United States and Japan.   Important features of the technology
include SO- removal  levels in excess of 90 percent over a wide range of inlet
SO2 concentrations (24) and minimal scaling and plugging.  However,  as with
other regenerate FGD processes, the Wellman-Lord system is mechanically
and chemically complex.  Also,  handling the concentrated SO^ product stream as
elemental  sulfur or sulfuric acid entails additional complex processing,  and
advantageous economics require an available market for the end product.
Another limiting factor for utility applications may be the disposal  of the
sodium sulfate solids, although the waste volumes are considerably less than
for throwaway FGD systems.
     The Wellman-Lord system is commercially available and as of 1985 has been
applied to the four U. S. utility plants listed in Table 2.6-1 (25).
                                       2-51

-------
       TABLE 2.6-1.  WELLMAN-LORD UTILITY INSTALLATIONS IN THE U.S.  (24)
Utility           Plant Name (MW)      %S Fuel     SC>2 End Product     Start-up
N. Indiana
Public Ser.

Public Ser.
Co. of N. M.

Oelmarva
Power & Lt.

Public Ser.
Co. of N. M.
Mitchell (115)         3.0
San Juan (700)         0.8

Delaware City          7.0
(180)                (coke)
San Juan (1100)        0.8
 Elemental S


 Elemental S

Sulfuric acid



Sulfuric acid
1976


1978

1979



1982
                                       2-52

-------
These applications represent both low sulfur (0.8 percent) and high sulfur (7
percent) coals.   In early 1986, It was reported that one German and one
Italian utility had ordered construction of Wellman-Lord units; the latter
being a 10 MW demonstration plant (26).  In Japan there are two utility
Installations, one at Chubu Electric's Nishinagoya station (620 MW) and the
other at the Niigata station (380 MW) of Tohoku Electric (1).
     Retrofits are possible, but there must be adequate land available near the
boiler and It must be possible to modify the existing ductwork to route the
flue gas to and from the FGD system.  Many of the Industrial  sites where the
Wellman-Lord system is employed are existing sulfurlc add or sulfur plants.

Magnesia Slurry--
     The magnesia slurry process has been demonstrated on both oil- and
coal-fired utility systems.  Retrofit is possible and the process has good
turn-down capability.   The technology is well developed; in the mid-1970's two
demonstration units having capacities of 90 and 150 MW operated for
2 years (27).  Three commercial units totaling 724 MW were brought into service
by Philadelphia Electric Company in 1983 (1).  MgSO, is shipped off-site to
existing sulfuric acid plants and the regenerated MgO is returned to the power
plant site.

Performance

     S02 removal  performance of gypsum by-product recovery plants is about the
same as conventional  lime/limestone FGD systems.   However,  system operation
must be more carefully controlled in order to avoid excessive unreacted reagent
in the product solids.   The 2250 MW Martin Lake plant has operated since late
1984 at better than 90 percent SO- removal, with greater than 95 percent
oxidation and 90 percent utilization of limestone reagent.   The plant burns a
lignite with a sulfur content of 1-4 percent.  The highly oxidized sludge
clarifies quickly, but careful  solids handling is needed to avoid line
pluggage.   Also,  close pH control  is necessary to maintain a  sufficiently high
oxidation level.
                                       2-53

-------
     The  160 MW Muscatlne, Iowa plant has successfully operated a forced
oxidation system at 94 percent sulfur removal with 3.5 percent sulfur coal
(20).  After initial start-up in early 1984, the operability of the system has
averaged better than 95 percent, and reliability above 99 percent.  There have
been some problems with scaling of gypsum on the scrubber internal components
due to the high oxidation level.  The largest problem to date with the produced
gypsum has been excess unreacted limestone remaining in the solids.
     Commercial Wellman-Lord systems have demonstrated removal of SO- in excess
of 90 percent.  High SCL removal is achievable due to the high alkalinity of
the sodium-based sorbent solution.  In addition, the process can handle widely
varying inlet SCL concentrations, and scaling or solids-plugging do not present
a problem.  However, process complexity has resulted in greater maintenance
requirements than for conventional lime/limestone FGD operation.

Cost

     The costs for producing salable gypsum include many components,  but are
partially offset by disposal  cost savings.  In general, the additional
operating costs for gypsum production roughly balance those for disposal   The
key. variable then becomes whether the market: and price for the product can
justify the additional  capital costs of increased washing capacity, forced
oxidation, or other processing.   Another factor affecting this option in the
U.S.,  was a tax depletion allowance of 14 percent (as of 1984) granted to
gypsum mining companies (22).  This tax credit reduces the actual  cost of mined
gypsum, making it more difficult for FGO gypsum to be competitive.
     Table 2.6-2 presents estimated capital  costs,  operating costs, and cost
effectiveness for a Wellman-Lord FGD system as applied to a coal-fired utility
boiler using EPRI guidelines  (28).  The Wellman-Lord process is more  costly
than conventional lime/ limestone FGO systems.  However, the capital  cost
differential of the two systems  decreases with lower sulfur coals.  Operating
costs are considerably higher for the Wellman-Lord process,  as might  be
                                       2-54

-------
               TABLE 2.6-2.  COST ESTIMATES FOR WELLMAN-LORD  FGD
                               (Current 1988 $)

Technology
New 500-MU Basel oad Power
Capital Cost, S/kW
M1lls/kWh, Constant $
Current $
$/Ton, Constant $
Current $
% SO-
Removed
Plant:
90
90
90
90
90
3.5%

259 -
11.7 -
20.0 -
443 -
763 -
S

290
12.1
21.0
461
795
2%

204 -
9.4 -
16.1 -
624 -
1,070 -
S

229
9.7
16.7
647
1,21
Retrofit 500-MW Baseload Power Plant1:

  Capital Cost, $/kW            90             335  - 375            265 - 296
Mills/kWh, Constant $
Current $
$/Ton, Constant $
Current $
Retrofit 250-MW Intermediate
Capital Cost, $/kW
Mills/kWh, Constant $
Current $
$/Ton, Constant $
Current $
90
90
90
90
Load Power
90
90
90
90
90
12.8
22.1
484
832
Plant2:
538
31.5
44.1
1,197
1,679
- 13.3
- 23.2
- 506
- 870

- 602
- 33.8
- 47.5
- 1,289
- 1,811
10.2 -
17.7 -
680 -
1,169 -

424 -
25.7 -
35.9 -
1,712 -
2,398 -
10.6
18.6
709
1,270

475
27.6
38.7
1,839
2,579
Retrofit difficulty of 1.3 and capacity  factor  of 65.

Retrofit difficulty of 1.6 and capacity  factor  of 35.
                                       2-55

-------
expected in view of the greater mechanical  and chemical complexity.   Retrofit
costs are also substantially higher relative to new Installation of equivalent
or greater capacity lime/limestone systems.  In these cost comparisons,  the
market potential for recovered by-product at a particular-location is not
expllclty included.
     Extremely limited experience, to date, with the magnesia slurry process on
a commercial  scale would make specific cost comparisons unreliable.   However,
it is generally recognized that capital  and O&M costs of the magnesia process
are higher than conventional lime/limestone FGD.
                                       2-56

-------
 2.7   SPRAY  DRYING

 Description

      Spray  drying  is an established technology for desulfurization of flue gas
 from  high-  and low-sulfur coals.  The first installation of utility spray dryer
 systems occurred in the late  1970's.  Thirteen systems, representing nearly
 4,300 MW of generating capacity, are currently in service, and an additional
 five  systems representing over 2,800 MW of generating capacity are in design or
 construction phases (29).  All of these full-scale systems are based on coals
 containing  less than two percent sulfur.  Evaluation of spray dryer systems for
 utility boilers burning greater than three percent sulfur coal was initiated in
 1985  using  two 1 MW, pilot-scale spray dryers at TVA's Shawnee Test Facility.
     A typical spray dryer system is shown in Figure 2.7-1.  Lime is the
 reagent in  17 of the utility  systems mentioned above, while one (the first
 spray dryer system purchased) uses a sodium-based reagent.  Only lime-
 based systems are described here.
      In a lime-based system,  quicklime is slaked to form calcium hydroxide,  and
 the resulting slurry is combined with additional  process makeup water and (in
 most  systems) recycled solids.  This combined slurry is then atomized and mixed
 with the flue gas at air heater exit conditions in the spray dryer vessel.
 Slurry atomization is generally accomplished with a rotary atomizer.   Two-fluid
 air atomizing nozzles are used in some systems.  Several simultaneous reactions
 occur immediately following atomization:
     •    Water in the slurry droplet evaporates;
     •    The evaporating water cools and humidifies the flue gas; and

     •    Flue gas SO- reacts with the lime,  producing calcium sulfite
          and calcium sulfate solids.
The amount of water fed to the spray dryer is carefully controlled to avoid
complete saturation of the flue gas.  However, a  close approach to
                                       2-57

-------
                                                          Head Tank
    Lime
    Feed
     Bin
                 Flue Gas from Unit 7
                             Dilution
                      [ritlon    Water
                        ker
in
CD
  Slaking
   Water
             Recycle
             Storage
               Silo
                                   Fabric Fitter
                             lime Slurry
                           Storage Trough
Mix Tank
 Atomizer
Feed Tank
                                                                                                             I.D. Fan


\ /
c
V.

1
Recycle Solids


                                                      Y
                                                     Solids to
                                                     Disposal
                                                                                                                         Stack
                         Figure  2.7-1.  Lime  Spray Drying Process  Flow  Diagram (30).

-------
 saturation temperature  (20°-35°F)  is needed to achieve high SO- removal and
 lime utilization.  Sorbent feed rate, degree of slurry atomization (i.e.,
 droplet  size),  flue gas residence  time in the spray dryer vessel, and the
 approach to flue gas saturation temperature are carefully controlled to achieve
 acceptable S02  removal performance while avoiding problems with drop out or
 caking of incompletely dried solids in the spray dryer vessel.  The calcium
 content  in the  slurry is limited to a maximum of about 35 percent solids.
 Typical  flue gas residence time in the spray dryer vessel is 8-12 seconds.
     The nearly-saturated flue gas leaves the spray dryer vessel and enters the
 particulate collection device which is either a fabric filter or an
 electrostatic precipitator (ESP).  Most utility experience to date has been
 with fabric filtration.  In the particulate collection device, the original
 coal flyash and the dried sorbent  are removed from the flue gas.  Additional
 SO- removal occurs across the particulate control device.  Fabric filters
 contribute as much as 20-30 percent to overall system S02 removal (30).  Less
 information is  available on SOL removal across an ESP, but it may be as high as
 25 percent at low approach-to-saturation temperatures (20°F) (31).
     In most systems, a portion of the collected solids are recycled to the
 slurry mix tank.  Solids recycle offers several  advantages over once-through
 lime operation, including improved lime utilization, improved rotary atomizer
 operation,  faster droplet drying, and reduced solid waste (32).
     Lime spray drying has several  perceived advantages over wet lime/limestone
 scrubbing.   First, the need to recirculate large quantities of slurry are
avoided.  L1quid-to-gas volume ratios for spray dryer systems tend to be 10 to
50 times smaller than what is typical  for lime/limestone scrubbing.   Second,
wastes from the spray drying process are produced as dry solids, rather than as
a wet sludge.   However, moving and storing relatively large quantities of dry
solids can also be a troublesome operation.   Third, fewer corrosion problems
occur with Hme spray drying since the system is dry rather than wet.
                                       2-59

-------
Applicability

     Although spray drying 1s generally considered a commercially established
technology for low-sulfur coal utility applications, data with high-sulfur
coals are limited.  In high-sulfur coal applications, heat balance
considerations will limit SO- removal by limiting the amount of lime and water
that can be added as well as the amount of sorbent recycle.  Specifically, the
differential between the spray dryer inlet temperature and the approach to flue
gas saturation temperature at the spray dryer outlet will determine the amount
of water that can be evaporated by the flue gas.  If too much water is used,
the flue gas will cool to below saturation temperature and result in scaling
and corrosion problems.  If too little water is used, poor slurry atomization
and low lime utilization will occur.  Additives (specifically chlorides) have
been successfully used in short-term tests to Increase sorbent utilization at
high SO- removal rates.
     Spray dryers can be used in retrofit applications 1f there 1s sufficient
land near the boiler and If ductwork can be cost-effectively run from the air
heater outlet to the spray dryer inlet and from the spray dryer outlet to the
particulate control device.
     The type of existing particulate control device may influence the
applicability of spray drying as a retrofit FGO technique,  particularly for
high-sulfur coal applications.  Fabric filters are well  suited for spray dryer
retrofits,  but few existing utility boilers are so equipped (particularly in
high-sulfur coal service).   Existing ESPs have several limitations,  including:
     •    Less contribution to system SCL removal  than a fabric filter;
     •    Greater sensitivity to increased particulate loading caused by the
          sorbent (addition of a spray dryer will  require greater particulate
          removal efficiencies to maintain emission rates at pre-spray drying
          levels); and
     •    Potential corrosion problems.
However, these Impacts are poorly quantified and more testing is needed.
                                       2-60

-------
      For  low-sulfur coal applications where high recycle rates are achievable
 (30), adequate S(L removal performance may result even with relatively low SO-
 removal in the particulate control device.  Also, most existing ESPs designed
 for low-sulfur coal have high specific collection area (SCA) values of 400
  2
 ft /thousand acfm and greater.  With high initial SCA values, the beneficial
 effects of gas volume shrinkage and gas conditioning which results from cooling
 and humidification in the spray dryer may more than offset the effects of
 increased mass loadings.
     Many existing ESPs in high-sulfur coal applications have SCA values of 200
 or less.  At these SCA values, it is unlikely that existing emission rates will
 be maintained after a spray dryer is installed upstream.  Pilot studies
 conducted by TVA (31) indicate that an ESP with an SCA of 300 which currently
 meets existing particulate emission regulations by a comfortable margin should
 be able to maintain compliance after a spray dryer is added.  If the ESP cannot
 be adequately upgraded, installation of a replacement baghouse would be
 required.
     Other issues for spray dryer retrofits include increased corrosion of
 particulate control devices due to operation at lower gas temperatures and
 higher moisture content, and the need to upgrade existing particulate collector
 solids handling equipment to deal  with increased loadings and alkalinity caused
 by the spray dryer.  Replacement of wet sluice ash handling systems with a dry
 system would also be required to preheat plugging of the sluice lines.

 Performance

     All existing and currently planned lime spray dryer FGD systems are
designed for coals with less than two percent sulfur; design S02 removals are
61 to 90 percent (29).  Most of the 13 operating units readily meet their
design S02 removal  levels.   Less experience is available for high-sulfur coal
applications.  Long- term utility test results are not yet available.
Short-term spray dryer tests combined with fabric filtration have been
conducted  on high-sulfur coals at Argonne National  Laboratory,  General Motors'
Buick Steam Plant,  and Northern States Power Company's Riverside Station.
These tests demonstrated over 90 percent SO, removal  on coals up to 4.2 and 3.2
                       -» .                  £          '
percent sulfur with and without the use of additives, respectively (33).   In

                                       2-61

-------
 pilot-scale  testing  using  a  lime  spray dryer/baghouse in high-sulfur coal
 applications, TVA demonstrated  that greater than 80 percent removal is
 feasible.  Lime  utilization  at  these  SO- removal rates was 60-65 percent.
     Pilot-scale tests  for a  high-sulfur spray dryer/ESP system conducted by
 Wheelabrator A1r Pollution Control and the U.S. Department of Energy achieved
 SOp removal levels as high as 90  percent with a 2.7 percent sulfur coal  (34).
 Lime utilization was only  about 30 percent; however, these tests did not employ
 recycle and the approach to  flue  gas  saturation was not as close as is possible
 to maximize lime utilization.   The spray dryer inlet temperature for these
 tests was rather high at 400°F.
     Tests conducted by TVA on  a  10 MW spray dryer/ESP pilot unit at Shawnee
 with a 3.5 to 4 percent sulfur  coal and a 300°F inlet flue gas temperature
 achieved overall SO- removal  levels of 90 percent (35).  Most of these tests
 were conducted at a  18°F approach to  adiabatic saturation at the spray dryer
 outlet.
     The data discussed above indicate that high levels of S02 removal
 (90 percent) are possible  for new units that are designed to operate at these
 levels.  However, for retrofit  applications where the existing ESP will be
 reused, SO. removal may be limited by ESP performance.   EPRI also is sponsoring
 research on lime spray drying at  Its High Sulfur Test Center.   Pilot-scale
 tests on a 4 MW' spray dryer with  fabric filter have achieved 90% sulfur dioxide
 removal on coals with 3.5-4% sulfur and a calclum-to-sulfur ratio of 1.3 to
 1.6.
     The potential  for combined SO./NO  reduction also  exists  by addition of
 roughly 10 weight percent  sodium hydroxide to the lime.   The NaOH both reacts
with SCL and extends particle reactivity as well  as serves as  a catalyst to
 reduce NO  emissions.  In  pilot-scale tests with a three percent sulfur coal
 and a lime-to-sulfur ratio of 1.3, SO- and NO  reductions of 95 and 55 percent,
 respectively, were achieved.   However, the process oxidizes a  portion of the NO
to NO-, resulting in increased plume opacity and discoloration (36).   Ammonia
 injection has been demonstrated to reduce the NO. plume.
                                       2-62

-------
Cost
     Due to the sensitivity of spray dryer performance to plant design and
operating variables, capital and operating costs for retrofit spray dryer
applications are expected to be more site specific than wet lime/limestone FGD.
Key factors in evaluating spray dryer costs at a given site include air heater
outlet temperature, length of ducting between the air heater outlet and spray
dryer inlet, percent sulfur in the coal, and whether the existing particulate
control device (typically an ESP) can be upgraded to handle the increased
particulate loading resulting from spray dryer operation.  The expected SO-
reduction across a baghouse versus an ESP is also important..
     A critical consideration in spray dryer economics will be whether to use
the existing ESP or install a new baghouse.  Installation of a new baghouse can
increase retrofit capital costs by 30-50 percent and annualized costs by 25-40
percent above the cost of a spray dryer system in which the existing ESP can be
used.   Because of the uncertainty in spray dryer performance and cost, two cost
cases are examined.  One case assumes that the existing ESP can be upgraded and
continues in service.   The other assumes that a new baghouse will  be installed.
     The following table presents the range of values estimated using IAPCS
cost model  for capital  cost, annualized cost, and cost per ton of pollutant
removed.  The lower costs are for a large unit with a low retrofit factor
burning low sulfur coal.  The higher costs are for a small  unit with a high
retrofit factor burning high sulfur coal.  Figures 2.7-2 through 2.7-4 show
these costs as a function of boiler size, coal sulfur content, and particulate
control option.
                                       2-63

-------
 o
 a.
JC
u
o
a
                                               Lime Cost = $63.3/ton
                                               ESP-SCA = 400 fr/lOOO ACFM
        100
        100
                   300
                             500
                                       700
                                   MW

                          (a)  Existing ESP
                                                 900
                                                           i. 1 DO
                       1,300
           1% S   •

           2% S


           3% ^


           4% S

Capacity Factor = 50/5

Lime Cost = $63.3/lon

ESP-SCA = 400 ft2/1000 ACFM
                  300
                             500
                                  MW
                                       700
                                                 900
            1.100
1.300
                          (b)  New  Baghouse


             Figure  2.7-2.  Lime  Spray Drying - Capital  Cost,

                             Current 1988 Dollars

                                     2-64

-------
tc
\u
OL
QC
UJ
0.
                                               Capacity Factor = 50%
                                               Lime Cost = $63.3/ton
                             ESP-SCA = 400 reViooo ACFM
        100
        TOO
300
500
                                   MW
700
900
1100
1300
                          (a)  Existing ESP
                                        1% S


                                        2% S


                                        3% S


                                        4% S

                            Capacity Factor = 50%

                            Lime Cost = $63.3/ton

                            ESP-SCA = 400 ft2/!000 ACFM
300
500
                                   MW
700
900
1100
                                                                     1300
                          (b)  New  Baghouse


          Figure  2.7-3.  Lime Spray Drying - Annualized Cost,

                          Current  1988  Dollars

                                   2-65

-------
CT3
OC
*•«
l. <0
03
Q.O
 -
        Capacity Factor = 50%
        Lime Cost = $63.3/ton
        ESP-SCA = 400 ftVlOOO ACFM
        100
        100
                  300
                            500
                                  MW
                                     •700
                                                900
                                                         1100
                                                                   1300
                          (a)  Existing ESP
                   1% s
                   2% S
                   3% 5
                   4% 5
        Capacity Factor = 50%
        Lime Cost = $63.3/ton
        ESP-SCA = 400 ftVlOOO ACFM
                  300
                            500
700
900
                                                         1100
                                                                   1300
                                 MW
                         (b)  New  Baghouse
Figure 2.7-4.  Lime Spray Drying - Cost Per  Ton  of S02 Removed,
                 Current  1988 Dollars
                                 2-66

-------
                                                      Range
     Existing ESP -
       Capital Cost ($/kW)                         64.7 to   535.4
       Annualized Cost (mills/kWh)                  6.3 to    65.7
       Cost Per ton of SC>2 Removed ($/ton)        441.5 to 8,084.6

     New Baghouse -
       Capital Cost ($/kW)                        149.2 to 639.2
       Annualized Cost (mills/kWh)                 10.2 to 75.7
       Cost Per ton of S02 Removed ($/ton)        522.2 to 8,632.4

     Appendix F contains tables of costs as a function of boiler size, coal
sulfur,  capacity factor, and retrofit factor.
                                       2-67

-------
References

1.   Miller, M. J.  SO- and NO   Retrofit Technologies Handbook.  CS-4277-SR,
     Electric Power Research  Institute, Palo Alto, California, 1985, 366 pp.

2.   Policy Analysis Department. Steam Electric Plant Factors, 1987 Edition.
     The National Coal Association, Washington, D.C., 1987.

3.   Energy Ventures Analysis,  Inc.  Coal Markets and Utilities' Compliance
     Decisions.  Palo Alto, CA.  EPRI Report P-5444, September 1987.

4.   Tennican, M. L., R. E. Wayland, and D. M. Weinstein.  Agenda of Critical
     Issues:  Coal Price and Availability.  EA-3750, Electric Power Research
     Institute, Palo Alto, California, 1984, 41 pp.

5.   Palmisano, P. J. and B. A.  Laseke, User's Manual for the Integrated Air
     Pollution Control System Design and Cost Estimating Model, Version II,
     Volume I, EPA-600/8-86-031a (NTIS PB87-127767), September 1986.

7.   E. H. Pechan and Associates, Inc. and PEI Associates, Inc.  Acid
     Deposition Control Techniques for the New England States.  Northeast
     States for Coordinated Air Use Management, Boston,  Massachusetts, 1986.

8.   Wisconsin Public Service Commission.  Utility Sulfur Dioxide Cleanup -
     Cost and Capability.  Madison, Wisconsin, 1986, 116 pp.

9.   Bechtel National, Inc.  Advanced Physical Fine Coal Cleaning,
     DOE/PC/81205-T6-Vol. 1 (DE89003925), September 1988, p.  1-1.

10.  Kilgroe,  J.  D.  and J. Strauss.  Coal Cleaning Options for SO*  Emission
     Reduction.  Power Magazine Conference on Acid Rain, Washington, D.C.,
     1985, 53 pp.

11.  McCandless,  L.  C. and R.  B. Shaver.  Assessment of Coal  Cleaning
     Technology,  First Annual  Report.  EPA-600/7-78-150 (NTIS PB287-091),  July
     1978.

12.. Onursal,  A.  B.,  J. Buroff, and J. Strauss.  Evaluation of Conventional  and
     Advanced Coal  Cleaning Techniques.  EPA-600/7-86-017 (NTIS PB87-104535),
     September 1986.

13.  Hucko, R.  E.,  H.  B.  Galal, and P. S. Jacobsen,  Status of DOE Sponsored
     Advanced Coal  Cleaning Processes.  DOE/PETC/TR-89/3 (DE89008772),
     March 1989.  pp.  1, 2.

14.  Cavallaro,  J.  A., M.  T.  Johnston, and A.  W.  Deurbrouck.   Sulfur Reduction
     Potential  of the Coals of the United States.   EPA-600/2-76-091  (NTIS
     PB252-965),  April 1976,  323 pp.

15.  Holt, E.  C.   An  Engineering/Economic Analysis of Coal  Preparation Plant
     Operation and Cost.   EPA-600/7-78-124:: (NTIS PB285-251), U.  S.  Department
     of Energy and U.  S.  Environmental Protection Agency, Washington,  D.  C.,
     1978, 296 pp.

                                       2-68

-------
 16.   Thompson,  R.  E.  and M. W. McElroy.  Guidelines for Retrofit  Low NO
      Combustion Control.   In:  Proceedings: 1985 Symposium on Stationary
      Combustion N0v Control, Vol.  1,  EPA-600/9-86-021a (NTIS PB86-22
      5042), July 1986,  p.  27-1.

 17.   Douglas, J.   Retrofit Strategies for NOW Control.  EPRI Journal, 12(2):
      26-31, 1987.                           x

 18.   Keeth, R.  J., et at.  Economic Evaluation of FGD Systems  (Volume 1).
      CS3342, Electric Power Research  Institute, Palo Alto, California, 1983,
      400 pp.

 19.   Melia, M.  T., R. S. McKibben, F. M. Jones and J. L. Kelly.   Trends in
      Commercial  Applications of FGD.  In:  Proceedings: Tenth  Symposium on Flue
      Gas Desulfurization, Atlanta, Georgia, November 1986, Vol. 1,
      EPA-600/9-87-004a  (NTIS PB87-166609), February 1987, p. 2-23.

 20.   Liegois, W. A. and D. A. Wicks.  Gypsum By-product FGD System.
      In:  Ninth  Symposium on Flue Gas Desulfurization, Cincinnati, Ohio,
      June 1985,  Volume  2, EPA-600/9-85-033b (NTIS PB 86-138658),  December 1985,
      p. 716.                            .

 21.   Mzyk, D. and J. Zmuda.  By-product Gypsum Production at a 2300 MW Power
      Plant.  In:  Ninth Symposium on Flue Gas Desulfurization, Cincinnati,
      Ohio, June  1985, Volume 2, EPA-600/9-85-033b (NTIS PB 86-138658),
      December 1985, p. 730.

 22.   Ellison, W., and L. M. Luckevich.  FGD Waste: Long-Term Liability or
      Short-Term Asset?  Power, 128(6)-.79-83,1984.

 23.   Ottmers, D. M., Jr., et al.  Evaluation of Regenerate Flue Gas
      Desulfurization Processes.  RP 535-1, Electric Power Research Institute,
      Palo Alto, California, 1976.

 24.   Hudak, C. E. and J. M. Burke.  Sulfur Oxides Control  Technology Series:
      Flue Gas Desulfurization, Well man-Lord,  Summary Report.
      EPA-625/8-79-001, U.S. Environmental Protection Agency, 1979.

 25.   Behrens, G. P., et al.  The Evaluation and Status of Flue Gas
      Desulfurization Systems.   CS-3322,  Electric Power Research Institute,
      Palo Alto, California, 1984,  702 pp.

 26.   Ellison, W., et al.  West Germany Meets Strict Emission Codes by
     Advancing FGD.  Power, 130(2):29-33, 1986.

27.   Erdman,  D. A.  Mag-Ox Scrubbing Experience at the Coal-Fired Dickerson
     Station of Potomac Electric Power Company.   In:  Proceedings:
      Symposium on Flue Gas  Desulfurization,  Atlanta, November 1974,
     Volume II, EPA-650/2-74-126b (NTIS PB 242573),  December 1974, p. 729.

28.  Shattuck,  et al.  Retrofit FGD Cost  Estimating Guidelines.   Prepared by
      Steams-Catalytic Corporation for the Electric Power Research Institute
      (EPRI),  Palo Alto,  California, EPRI Report CS-3696,  October 1984.

                                       2-69

-------
29.  Palazzolo, M. A. and M. A. Baviello.  Status of Dry SO, Control Systems:
     Fall 1983, EPA-600/7-84-086  (NTIS PB 84-232503),  U. Sf Environmental
     Protection Agency, Research.  Triangle Park, North Carolina, August 1984.

30.  Blythe, G. M., et al.  Field Evaluation of a Utility Spray Dryer System.
     EPRI Report CS-3954, Electric Power Research Institute, Palo Alto,
     California, 1985, 272 pp.

31.  Robards, R.,  et al.  Spray Dryer/ESP Testing for Utility Retrofit
     Applications on High-Sulfur  Coal.  In: Proceedings of the American Power
     Conference, TVA/OP/ED4T86/8, Chicago, Illinois, 1986.

32.  Blythe, G. M., et al.  Evaluation of a 2.5-MW Spray Dryer/Fabric Filter
     SO- Removal System.  EPRI Report CS-3953, Electric Power Research
     Institute, Palo Alto, California, 1985,  312 pp.

33.  Donnelly, J.  R. and K. S. Felsvang.   Spray Dryer Absorption Applications
     for High Sulfur Coal.  In: Proceedings of the Third Annual Pittsburgh Coal
     Conference, Pittsburgh, Pennsylvania, 1986, 974 pp.

34.  Yen, J. T., et al.  Performance of a Spray Dryer/ESP Flue Gas Cleanup
     System During Testing at the Pittsburgh  Energy Technology Center.  In:
     Ninth Symposium on Flue Gas Desulfurization,  Cincinnati, Ohio, June 1985,
     Volume 2, EPA-600/9-85-033b  (NTIS PB 86-138658), December 1985, p. 670.

35.  Brown,  C. A.  et al.  Results From the TVA 10-MW Spray Dryer/ESP
     Evaluation.  In:  Proceedings: First  Combined FGD and Dry SO, Control
     Symposium, EPA-600/9-89-036a (NTIS PB 89-172159),  March 198$, p. 3-1.

36.  Donnelly, J.  R.,  S. K. Hansen,  M. T.  Quach, and P.  S. Farber.  Industrial
     Spray Dryer FGD Experience.  In: Proceedings of the 1986 ASME Industrial
     Power Conference, Pittsburgh, Pennsylvania, 1986,  pp. 71-76.
                                       2-70

-------
                                  SECTION 3

                        NEAR-COMMERCIAL TECHNOLOGIES
INTRODUCTION

     This section discusses six technologies which are in the process of being
demonstrated by U. S. utilities or which have been commercially applied to
power plants overseas.  The first two technologies, integrated gasification
combined-cycle (Section 3.1) and fluidized bed combustion (Section 3.2), are
integrated processes that can increase the electric generating capacity of a
plant while minimizing SO- and NO  emissions.  Because of their design and high
capital cost, these two repowering technologies will  be used by utilities that
have a need to increase or maintain base load electric generating capacity as
well as to reduce emissions.  When compared with the cost of a new plant
(including emission control systems), addition of base load generating capacity
through repowering of existing capacity can be very economical.
     Post-combustion NO  control technology (Section 3.3), has been
                       ^
commercially applied to boilers burning low-sulfur fuels in Europe and Japan,
but has little or no long-term operating experience with medium- and
high-sulfur coals.  Because of specific process concerns, it is categorized in
this report as a near-commercial technology until its use with higher sulfur
fuels has been demonstrated.
     The final three technologies,  furnace and duct (also known as hot-side and
cool-side) sorbent injection (Sections 3.4 and 3.5) and reburning (Section 3.6)
have limited commercial application.   These technologies have.limited or
planned large-scale demonstration experience in the U. S. and should be
commercially available within the next few years.
                                        3-1

-------
3.1   INTEGRATED GASIFICATION  COMBINED CYCLE

Description

      Integrated gasification  combined cycle  (IGCC) is an alternative to
conventional coal-fired electric power generation with post-combustion emission
controls.  Because of  its overall design, emissions of sulfur and nitrogen
oxides and participates from  IGCC facilities are projected to be significantly
lower than from existing technologies.  Figure 3.1-1 is a generalized block
flow diagram of an IGCC facility designed to 1) convert coal (via partial
oxidation and gasification) into a fuel gas, 2) remove reduced sulfur species
(primarily hydrogen sulfide)  from the fuel gas, 3) use the clean fuel gas to
produce electricity in a combined-cycle application, and 4) treat and/or
recover byproducts from the waste streams generated.  Existing technology for
removal of sulfur species from fuel gas requires gas cooling followed by
scrubbing with any of  several proprietary chemical processes.  Cooling the gas
imposes both additional capital cost and a significant penalty on process
thermal efficiency.  Advanced processes for removal of H^S from hot fuel gas
(1,000 to 1,200°F) using adsorption by metal oxides are under development with
funding from DOE.  Development of hot gas cleanup technology will simplify IGCC
process complexity and reduce costs.
     A number of coal gasification processes have been examined for use in
combined-cycle applications.  The most successful demonstration of IGCC
technology in the U.  S. is the Cool Water Coal  Gasification Demonstration
Project.   The Cool Water facility,  located in Daggett,  California,  is based on
use of Texaco gasification technology with cold gas cleanup.  The facility is
designed to gasify 1000 tons of coal per day and produce approximately 100 MM
of net electrical power.  The facility underwent successful startup in 1984 to
become the first successful domestic operation combining coal gasification and
combined cycle equipment.   The description of Cool Water presented below is
representative of the general  concept of IGCC facilities.
                                        3-2

-------
OJ
f
CO
      Air	Ik-
Air Separation
               Oxygen
          Coal
          Coal
       Preparation
                                                          Steam
                       Steam
•»
-p»
                                            Saturated Steam
              Gasification
Acid Gas
Removal
                        Condensate    Ash to
                       to Treatment   Disposal
                                                     I
                                                                           Flue Gases
                                    Sulfur
                                  Recovery
                            i
                                                          Heat Recovery
                                                        Steam Generation
                                                                    Hot Turbine
                                                                    Exhaust Gas
                                               Clean
                                              Fuel Gas
Combustion
 Turbines
                                                                               Air
                                                    Sulfur
                                                 By - Product
                                                                                Superheated
                                                                                   Steam
                                                                                                   Steam
                                                                                                   Turbine
                                                                                  Electric
                                                                                   Power
                          Figure  3.1-1. Generalized  block flow diagram  of combined
                                          cycle  coal gasification power generation

-------
     The Texaco gasifier  is  a  pressurized, downflow, entrained reactor in which
pulverized coal is  fed  as  a  coal/water slurry.  High purity oxygen
(95+ percent)  is also introduced  into the top of the gasifier.  The coal
undergoes partia> oxidation  at high temperature and pressure to produce a raw
fuel gas composed primarily  of hydrogen, carbon monoxide, and carbon dioxide.
The raw fuel gas also contains reduced sulfur and nitrogen species, water
vapor, entrained molten slag,  and soot particles; no condensible hydrocarbons
are found in the gas.
     The raw fuel gas 1s cooled in radiant and convective coolers, producing
saturated steam for use In the combined-cycle power generation system.  The
fuel gas is initially treated  in a particulate scrubber and a series of heat
exchangers and coolers.  Cooled fuel gas is then sent to the sulfur removal
unit (the Selexol process  is used at Cool Water) with a design sulfur removal
efficiency of 97 percent.  Removed sulfur compounds are subsequently converted
Into elemental sulfur 1n a Claus sulfur recovery unit equipped with a SCOT tail
gas treating unit.
     The clean fuel gas is then reheated and saturated with moisture before
being combusted in a gas turbine to produce electricity.  Moisture addition is
used to reduce NO  formation during combustion.   Exhaust gas from the gas
turbine is sent through a heat recovery steam generator (HRS6) to produce
steam.  This steam plus that from the fuel  gas heat exchangers and coolers is
sent to the steam turbine to produce additional  electricity.   Cooled flue gas
from the HRSG is discharged through a stack to the atmosphere.
     Auxiliary processes included in an IGCC facility include 1)  coal  storage
and preparation, 2) oxygen production (if pure or nearly pure oxygen is used as
the gasifier reactant) and compression, 3)  wastewater treatment,  and
4) other processes normally present in a coal-fired electric generating plant
(e.g., cooling towers and ash handling/disposal).

Applicability

     A number of coal  gasification technologies  are commercially  available,  and
oil- and gas-fired combined-cycle power plants are currently in commercial use.(l)
                                        3-4

-------
With the recent success of the Cool Water Demonstration Program and
continued research/demonstration activities, IGCC has become a rapidly
emerging alternative for electricity generation at new plants and for
repowering existing plants.  Potential economic advantages include
high modularity which will allow IGCC plants to be constructed in relatively
small increments and ability to use lower priced (higher sulfur) coals.
Retrofit to an existing coal-fired power plant will require construction of
new gasification and combustion turbine facilities.  However, depending on
site-specific factors, it may be possible to incorporate some existing
equipment into the IGCC plant (e.g., steam turbines, boiler feed-water and
cooling systems, coal handling, and electric generator and auxiliaries).

Performance

     The electric power industry has a growing interest in IGCC technology
because of a number of environmental advantages versus conventional
coal-fired power plants, including 1) lower SO-, NO , and particulate
emissions, 2) lower water consumption, and 3) lower land requirements.
Sulfur dioxide emissions from an IGCC facility primarily arise from two
sources:  combustion flue gas and tail gases from the sulfur recovery unit.
Very stringent control of sulfur emissions from an IGCC plant is attainable
even for gasification of high-sulfur coals.
     The sulfur removal unit in an IGCC is designed to remove sulfur species
(primarily hydrogen sulfide) from the fuel gas prior to combustion, rather
than removal of SO- following combustion.  Sulfur control  efficiencies
greater than 95 percent are achievable with a properly designed sulfur
recovery unit.  Sulfur species remaining in the fuel gas are converted to
sulfur oxides in the gas turbine combustor.
     Sulfur emission test results from Cool Water are summarized in
Table 3.1-1.  These results indicate that S02 emissions from IGCC can be
reduced to levels significantly below current utility NSPS requirements.
                                     3-5

-------
                           TABLE  3.1-1.   AVAILABLE  SO   EMISSIONS  DATA  FOR  COOL  WATER DEMONSTRATION  PROGRAH

Combined -Cycle Combustion Flue Cases
Sulfur
Coal Gasified Removal (SO ppmvd") (Ib SO /hr> (Ib SO./ 10 Btu )
Lou sulfurC NR NR NR 0.032
Lou sulfur0 971 NR NR 0.034
Lou sulfur' NR fl. 59 33. l" NR
Incinerator Stack
(SO ppavd8) (Ib SO /hr) (Reference)
NR NR 2
NR NR It
54 3.2 7
NR = Not reported.



 Parts per million by volume,  dry basis.
 Pounds SO  per million Btu of coal.



 Sulfur content not reported.
 0.46 ppnvd of H,so, mist also reported.
 2.7 Ib/hr of H SO  mist also reported.
               2  4

-------
Although the Cool Water test data shown are for gasification of a low-sulfur
coal,  similar results are expected with medium- and high-sulfur coals.
     The Electric Power Research Institute (EPRI) has funded a number of
studies relating to  IGCC technologies (3,5,6,8,10).  As part of these
studies, estimates of the performance and cost of IGCC facilities gasifying
high-sulfur Illinois #6 coal have been developed.  Table 3.1-2 summarizes
the sulfur control levels incorporated into these designs.  Even higher
levels of sulfur removal (97-99 percent) are achievable at small increases
in overall capital cost.
     Nitrogen oxide emissions from IGCC facilities are primarily associated
with the gas turbine combustion gases.  Test data from Cool Water,
summarized in Table 3.1-3, indicate NO  emissions equal to one-tenth the
current utility NSPS for coal-fired boilers are achievable.  As discussed
previously, the Cool Water facility injects steam into the fuel gas prior to
combustion in the gas turbine to reduce NO  emissions.

Cost

     Cost and emission data for several  IGCC power plants are summarized in
Table 3.1-4.   Additional cost data compiled by EPRI (9) provide capital  cost
estimates of $1,204-1,576 per kW (in December 1984 $) for a 500 MW plant
using bituminous coal.  Key cost variables were the gasification technology
(Texaco, Shell,  KRW, and British Gas/Lurgi),  method of gas cooling (none of
these include hot-gas cleanup), and turbine design (current or advanced).
     The following table presents the range of IAPCS values for capital  and
annualized costs.  The lower capital  and annual1zed costs are for a large
unit with a high capacity factor and high heat rate, while the higher
capital and annualized costs are for a small  unit with a low capacity factor
and low heat  rate.  Figures 3.1-2 and 3.1-3 show these costs as a function
of boiler size and heat rate (cost bases differ from the costs presented in
Table 3.1-4).   Reported heat rate values (9)  range from 7,130 Btu/kWh for
Texaco fuel cell combined cycle to 10,250 Btu/kWh for Texaco gasification
combined cycle burning bituminous coal.   For this study two cases were
considered for cost comparison, 8,000 and 10,000 Btu/kWh heat
                                     3-7

-------
    TABLE 3.1-2.  DESIGN SULFUR EMISSIONS INFORMATION FOR EPRI IGCC STUDIES


Gasification
Technology
Texaco
British Gas Corp/
Lurgi (slagging)
Shelld
Kellogg Rust f
Westinghouse
Feed

Type
111. #6

111. #6
111. #6

111. #6
Coal
Sulfur
Content
3.4%

3.4%
3.4%

3.4%

Overall
Sulfur Removed
95%

94.4%
90%

95%

Sulfur Dioxide
Emissions, lb/10 Btua
0.28

0.34
0.56

0.28
aBtu of coal feed.
 Reference 3.

d
cReference 5.
 Reference 6.
ePlant was designed to meet NSPS (i.e., 90% control).
 Reference 8.
                                        3-8

-------
          TABLE 3.1-3.   AVAILABLE NOV EMISSIONS  DATA  FOR  COOL WATER
                            DEMONSTRATION  PROGRAM

Reference
2
4
7d
NR = Not reported.
aPer million Btu of
Parts per million
and 15% 02.
lb/106 Btua
0.058
0.059
NR

coal .
by volume, dry basis;
ppmvd
23
NR
22.8/20.8


normalized to
lb/hrc
61
NR
61.2/57.2


ISO humidity (60%)
Calculated as N02.

 Results shown are for two test  periods.
                                     3-9

-------
               TABLE  3.1-4.   SUMMARY  OF  REFERENCE  INFORMATION OH EMISSION AND COST DATA FOR IGCC FACILITIES
Plant SO SO Emissions
Facility Size, MU Removal Ib/MBtu
Texaco (3)
Radiant 600 95X 0.2B
Radiant & Corwectlve 589 95X 0.28
Total Quench 571 95X 0.28
Shell (a) 1.122 90X 0.56
Kellogg Rust 547 95X 0.28
Uestlnghouse (8)
Capital Annual
NO Emissions Requirement O&M Cost. Cost
X
Ib/NBtu Millions S/kV SI. 000 Basis

0.04 . 820 1,360 24,800 1/83 S
01.3 860 1,460 26,100 1/83 S
0.04 740 1,300 22,700 1/83 S
NR 1,420 1,260 43,300 1981 S
<75C 805 1,470 26.800 1/83 S
Reflects cost of entire plant.
Based on 100X capacity and combination of Illinois 06 coal.
Units « ppnvd, parts per million volume (dry basis).

-------

                                                  Heat Rate  = 8,000 Btu/kWh
                                                  Heat Rote  =10,000 Btu/kWh
                                                  Capacity Factor  = 50%
         100
                                                    1.300
C0
        Figure  3.1-2.  IGCC -  Capital Cost,  Current 1988 Dollars
     210
                                                 Heot Rate  = 8.000 Btu/kWh
                                                 Heat Rate  = 10,000 Btu/kWh
                                                 Capacity Factor = 50%
        100
300
500
700
900
1,100
1,300
                                      MW
               Figure 3.1-3.  IGCC - Levelized  Annual Cost,
                                Current  1988 Dollars
                                       3-11

-------
 rate,  with direct capital  costs for both cases assumed to be the same.   This
 is not to be confused with the more advanced gasification process but is
 simply to present costs for a range of heat rates.reported.   The total
 capital  cost for the low heat rate value is considerably lower due to the
 lower  preproduction and inventory costs.

                                                    Range
   Capital  Cost  ($/kW)                           1,344  to  2,919
   Annualized  Cost  (mills/kWh)                    80.3  to  206.0

      Appendix 0 contains  tables  of costs  as  a  function of boiler size,  capacity
 factor,  and boiler heat rate.
      Because  IGCC  is  an integrated process producing  electricity as  well  as
 reducing  SO,  and NO  emissions,  it is  difficult  to  separate  costs associated
           £        A
 with  emission reduction processes.   However, for one  facility the capital
.cost  for  acid gas  removal  and  sulfur recovery  were  calculated to be
 approximately 6 percent of the total plant capital  investment..  Additional
 cost  information may  be found  in references  3, 6, and 8.
                                    3-12

-------
3.2  FLUIDIZED BED COMBUSTION

Description

     Fluidized bed combustion  (FBC) is an integrated technology for reducing
SOg and NOX emissions during combustion of coal.  As with IGCC, FBC can be
installed at new plants or used to repower existing plants.  FBC
technologies based on operation at atmospheric and pressurized conditions
have been developed.  Atmospheric FBC (AFBC) systems are similar to a
conventional boiler in that the furnace operates at near atmospheric
pressure and depends upon heat transfer of a working fluid (i.e., water) to
recover the heat of combustion.  Pressurized FBC operates at pressures
greater than atmospheric pressure and recovers energy through both heat
transfer to a working fluid and use of the pressurized gas to power a gas
turbine.  The pressurized systems offer the potential for smaller equipment
sizes (and thus simpler retrofits), lower capital cost, and higher
efficiency.

Atmospheric FBC --
     Figure 3.2-1 presents a simplified AFBC process flow diagram.   Coal and
limestone are fed into a bed of hot particles (1400-1600°F) fluidized by
upflowing air.  SOg formed during combustion reacts with the calcined lime-
stone to form calcium sulfate, thus reducing SCL emissions and avoiding the
need for post-combustion controls.  The relatively low combustion
temperature limits NO  formation, reduces ash fusion problems,  and optimizes
sulfur capture.  One of the major advantages of AFBC technology is its
ability to burn a wide range of fuels, including those with low heat and
                                                            %
high sulfur contents.
     There are two major AFBC types:   the bubbling or dense bed,  and the
circulating or dilute bed.  In the bubbling bed system, coal  and  limestone
are continuously fed into the boiler from over or under the bed.   The bed
materials consisting of unreacted, calcined, and sulfated limestone, coal,
and ash are suspended by the combustion air blowing upwards through the air
distributor plate.  Bed material  is drained from the bed to maintain the
                                    3-13

-------
          Convection
                    i
            Pass

Coal  Limestone
             Freeboard «
            Splash Zone «

                  Bed  «
 Transport Air
                         Fluldlzlng Air
    Forced Draft Air
                                                            Flue Gas
                                                           Cyclone
                                             Recycle
                                                        Distributor
                                                          Plate
                                                        Plenum
                                       Waste
Waste
       | Compressor |
 Figure  3.2-1.  Simplified  AFBC  process flow diagram.
                               3-14

-------
desired bed depth.  Some bed material is entrained in the upflowing flue gas
and escapes the combustion chamber.  Approximately 80 to 90 percent of this
fly ash is collected in the cyclone and is then either discarded or
reinjected into the bed.  Reinjection of ash is useful in improving
combustion efficiency and limestone utilization.  In general, combustion
efficiency increases with longer freeboard residence times and greater ash
recycle rates.  Fly ash not collected in the cyclone is removed from the
flue gas by an ESP or fabric filter.
     Circulating fluidized bed is a more recent development in AFBC
technology.  The two major differences between circulating and bubbling
AFBC's are:  the size of the limestone particles fed to the system, and the
velocity of the fluidizing air stream.  Limestone feed to a bubbling bed is
generally less than 0.1 inches in size,  whereas circulating beds use much
finer limestone particles, generally less than 0.01 inches.  The bubbling
bed also incorporates relatively low superficial air velocities, ranging
from 4 to 12 ft/sec.  This creates a relatively stable fluidized bed of
solid particles with a well-defined upper surface.  Circulating beds have
superficial velocities ranging from 20 to 40 ft/sec.   As a result,  a
physically well-defined bed is not formed; instead, solid particles (coal,
limestone, ash, sulfated limestone, etc.) are entrained in the transport
air/combustion gases.  These solids are then separated from the combustion
gases by a cyclone or other separating device and circulated back into the
combustion region, along with fresh coal and limestone.  A portion  of the
collected solids are continuously removed from the system to maintain
material balances.  Circulating beds are characterized by very high
recirculated solids flow rates, up to three orders of magnitude higher than
the combined coal/limestone feed rate (11).
     Several advantages of circulating bed over bubbling bed design have
been identified by FBC unit vendors (due to limited test and commercial
operating data, however, not all  of the advantages have been fully
demonstrated):
     t    Higher combustion efficiency,  exceeding 99 percent;
     •    Greater limestone utilization, due to high recycle of unreacted
          sorbent and small limestone feed size (greater than 85 percent SO-
                                    3-15

-------
           removal  efficiency is  projected  with  a  Ca/S  ratio of about  1.5,
           with  the potential  for greater than 95  percent  SO- removal
           efficiency);

      t     Lower NO  emissions  because  of staged combustion  (less  than
           100 ppm  NO  are  projected);
                     A
      t     Potentially  fewer  corrosion  and  erosion problems,  since the  heat
           transfer surface is  less  likely  to be located in  the primary
           combustion zone;
      •     Minimal  excess air requirements,  since the high velocities promote
           good  mixing  and  combustion efficiency;
      •     Less  dependence  on limestone type, since reactivity  is  improved
           with  the fine particle  sizes;  and
      •     Reduced  solid waste  generation rates, because of  lower  limestone
           requirements.

Potential  disadvantages are:

      •     Increased  capital costs;
      •     Greater  energy losses due to high pressure drops;
      •     Increased  combustor  height compared to a bubbling  bed;
      t     Uncertainty regarding the hot cyclone's ability to adequately
           separate solids and  gases and to resist erosion and  corrosion;
      •     Difficulty in scaling up for large units (over approximately
           200 MW),  limiting application generally to smaller boiler sizes;
      •     Less ability to use  existing boiler equipment in retrofit
           applications; and
      •     Erosion of components subjected to impingement of high
           velocity particles.

     Table 3.2-1 presents the estimated operating data for a 200 MW AFBC.
Table 3.2-2 summarizes full-scale utility AFBC demonstration facilities

                                    3-16

-------
   planned for startup between  1986 and the early 1990's.  During retrofit of
   Northern States Power's Black Dog Unit No. 2, generating capacity was
   increased from 85 MW up to 125 MW by increasing operating pressures and
   modifying the turbine-generator as part of the overall retrofit program.
   Estimated cost was $58 million ($460/kW).  Retrofitting AFBC on the four
   units at Wisconsin Electric's Oak Creek Station is expected to cost an
   estimated $380 million ($760/kW).  The projected output includes recovering
   130 MW of generating capacity lost due to coal changes and mechanical
   limitations on the existing  boilers (12).  The recovery of generating
   capacity lost during previous plant derates or expansion in capacity due to
   overall plant upgrading is one of the key economic benefits from repowering
   an existing plant with FBC.

              TABLE 3.2-1.  ESTIMATED OPERATING DATA FOR 200 MW AFBC
      Performance at Full Load
         Steam Cycle Heat Rate, Btu/kWh                      7724
         Boiler Efficiency, percent                          87.5
         Gross Plant Heat Rate, Btu/kWh                      8823
         Auxiliary Power, MW                                 14.9
         Net Plant Heat Rate, Btu/kWh                        9500
         Turbine Generator Gross Output, MW                  209.6
         Turbine Generator Net Output, MW                    194.7

         Consumption/Production @ 65% CF
         Coal, 103 TPY                                       521
         Limestone 9 Ca/S = 2.5, 103 TPY                     173
         Waste Solid, 103 TPY                                229

Source:   Reference 13.
                                       3-17

-------
             TABLE 3.2-2.  FULL-SCALE UTILITY AFBC DEMONSTRATIONS

Utility
Colorado Ute
Montana-Dakota
Station
Nucla
Heskett
New/
Retrofit
New
Retrofit
Bed Type
Circulating
Bubbling
Size
(MW)
110
85
Start-Up
1987
1987
 Utilities
Northern States
Power
Tennessee Valley
Black Dog

Shawnee
Retrofit

New
Bubbling

Bubbling
125

160
1986

1988
Pressurized FBC --
     Pressurized fluidized bed combustion (PFBC) is similar to AFBC with the
exception that combustion occurs under pressure.  By operating at pressure, it
is possible to reduce the size of the combustion chamber and to develop a
combined-cycle or turbocharged boiler capable of operation at higher efficien-
cies than atmospheric systems.  The turbocharged boiler approach recovers most
of the heat from the boiler through a conventional  steam cycle, leaving only
sufficient energy in the gas to drive a gas turbine to pressurize the
combustion air. .The combined cycle system extracts most of the system's energy
through a gas turbine followed by a heat recovery steam generator and steam
turbine.  As with IGCC (Section 3.1), hot gas cleanup technology is a critical
need in combined cycle PFBC development.
     The smaller size of a PFBC boiler may be especially attractive in retrofit
applications where space is limited.  However, PFBC 1s not as well  developed as
                                       3-18

-------
AFBC.  To date, work on PFBC has been limited mainly to component and
pilot-scale testing at facilities in the U. S., England, and Sweden.
Commercial PFBC systems have been ordered by two European utilities.  There is
currently one utility boiler size PFBC unit (75 MW) under construction in the
United States (Tidd) (14) with another one planned for 1995 (Philip Sporn).
Initial utility PFBC demonstration projects are not expected to be in operation
until the early 1990's with earliest commerical availability projected in the
1995-2000 time frame (15).

Applicability

     For new plants, AFBC can be applied to nearly any plant site because of
the inherently wide design ranges for fuel and operating conditions.  Single
unit AFBC design is presently limited in size to about 300 MW.   At these sizes,
unit design can be modularized to reduce construction cost.  Construction times
of 2-3 years are possible, thus allowing the utility to schedule construction
in response to load growth or other requirements.
     Retrofitting AFBC at an existing plant requires either 1)  installation of
a totally new boiler to supply steam to an existing turbine or  2) replacement
of the lower furnace section of the boiler while continuing to  use portions of
the convective heat transfer sections of the existing boiler.   Major design
questions in making this decision are the layout,  condition, and operating
limits (e.g.,  pressure/temperature) of the existing boiler and  particulate
control system and the availability of additional  space if an entire new boiler
is installed.
     Current AFBC designs are limited to heat release rates of
                  2
1.0 million Btu/ft .  Therefore,  an existing boiler with a heat release rate of
                  2
1.5 million Btu/ft  would either require an AFBC approximately  50 percent
larger than the existing boiler plan area or have  to be derated to two-thirds
of its present capacity (16).  Based on this, retrofit of AFBC  will  likely be
                                                                   2
limited to boilers with heat release rates below 1.5 million Btu/ft .   Because
                                       3-19

-------
most cyclone-fired and wet-bottom pulverized coal boilers have higher heat
releases, these units are not likely candidates for AFBC retrofit.  Bubbling
beds are preferred for retrofit applications because they require less space
and can use more of the existing plant equipment.

Performance

     Extensive testing of AFBC has been performed in numerous vendor pilot
plants, full-scale Industrial AFBC's, the EPRI/Babcock & Wilcox test facility
in Alliance, Ohio (rated at 20,000 Ib/hr of steam), the 20 MW Tennessee Valley
Authority pilot plant at Shawnee Station, and the 15 MW Northern States Power
French Island Pilot Plant.  The EPRI/B&U facility has unique scaling features
for commercial units, such as a high free board zone, to simulate utility
boiler residence times and temperatures.   Test results have shown overall
combustion efficiency of 98 percent and sulfur capture of 90 percent with a
calcium-to-sulfur ratio of 2 to 1.  NOV emission levels achievable with AFBC
                                    6
are estimated at less than 0.4 lb/10  Btu.
     The 20 MW TVA pilot plant has been used to evaluate a number of areas
requiring technological developments related to size scale-up and long-term
operation of commercial AFBC systems.  Objectives of the 20 MW pilot test,
cosponsored by TVA and EPRI, included:
     •    Demonstrate acceptable,  long-term operation and performance
          (efficiency, reliability,  emissions control,  etc.);
     t    Develop and demonstrate automatic control and load following
          capability;
     •    Develop design correlations for confident scaling to commercial size
          boilers;
     0    Demonstrate the environmental control capability of the unit as a
          basis for assessing the environmental acceptability of AFBC on a
          commercial  scale;
     0    Develop the capability to specify and design critical  auxiliary
          equipment,  especially the coal, limestone,  and recycle feed systems;
                                       3-20

-------
     •    Test and evaluate materials of construction for the boiler and its
          auxiliaries; and
     •    Train engineers and operators for future units.
     Performance results were favorable.  Combustion efficiency, sulfur
capture, and NO  emissions have been well within expectations.  The coal feed
systems were the largest problem.  The underbed feed system demonstrated
high combustion and sulfur control efficiencies, but required a more complex
feed system design and operated less reliably.  The overbed feed system
demonstrated high equipment reliability but less acceptable process
efficiencies, especially when the coal contained significant (15-20 percent)
fines.  Remedies for these problems were investigated prior to the final design
of the 160 MW TVA demonstration project.
     Flyash recycle is another key design factor for sulfur capture and
combustion performance optimization.  As previously mentioned, combustion
efficiency and limestone utilization increase with greater recycle.  Other
methods for improving limestone utilization include reinjection of sulfated
limestone after pulverizing and hydration.

Cost

     Analysis of system economies for new plants have generally found FBC to be
cost competitive with pulverized coal firing combined with limestone FGD.  Cost
estimates by EPRI for a conventional boiler, AFBC, and PFBC are presented in
Table 3.2-3 for a new plant.
     The following table presents the range of values estimated using IAPCS for
capital and annualized costs of new AFBC plants.  The lower capital cost is for
a large unit with a high capacity factor,  while the higher capital  cost is for
a small unit with a low capacity factor.  Figures 3.2-2 and 3.2-3 show these
costs as a function of boiler size and heat rate.  The reason for lower costs
in the lower heat rate value cases is due to the lower preproduction and
inventory costs.
                                        Range
     Capital Cost ($/kW)              1,132 to 2,382
     Annualized Cost (mills/kWh)       77.0 to 182.4

                                       3-21

-------
     TABLE 3.2-3.  COMPARATIVE ECONOMICS OF NEW 500 MW CONVENTIONAL AND
                      FBC POWER PLANTS3  (1988 dollars)


                        Conventional
                         Boiler w/   .          Turbocharged   Combined-Cycle
                          LS FGD      AFBC        PFBCD            PFBC


Capital Cost, S/kW        1,233        1,184      1,247           1,394


Heat Rate, Btu/kWh       10,060       10,000   .   9,703           8,980


Total Cost, Mills/kWh
   Constant $             37.3         23.0       36.9            39.4
   Current $              64.2         62.4       63.5            67.8


aBased on a 500 MW boiler with 3.5 percent sulfur coal at $1.55/million Btu and
 capacity factor of 65 percent.

bScaled from 250 MW based on a scaling  factor of 0.8; circulating bed
 design.

Source:  Reference 9.
                                       3-22

-------
03
0.0
                                                  Heot Rate =  9,000 Btu/kWh
                                                  Heat Rate = 11,000 Btu/kWh
                                                   Capacity Factor = 50%
         100
                         1.300
        Figure 3.2-2.  AFBC  - Capital Cost, Current  1988 Dollars
x
*
ec
UJ
Q.
                                                  Heot Rote  =  9,000 Btu/kWh
                                                  Heot Rote  = 11.000 Btu/kWh
                                                  Copocity Factor = 50%
      '90
         100
I     I     T     T
    900       1.100
1.300
    Figure 3.2-3.   AFBC  - Levelized Annual  Cost, Current  1988  Dollars
                                     3-23

-------
     Appendix E contains tables of costs as a function of boiler size,
capacity factor, and boiler heat rate.   Detailed costs are not available at
this time for PFBC,  and therefore these costs are not presented.
                                   3-24

-------
3.3   POST-COMBUSTION NOx CONTROL

Description

Selective Catalytic Reduction--
      Selective catalytic reduction  (SCR) is a flue gas treatment process for
removal of nitrogen oxides (NOX).   In this process, gaseous ammonia diluted
with  either air or steam is injected into the flue gas upstream of the SCR
catalyst.  The ammonia/flue gas mixture then enters the catalyst where the
ammonia reduces NO  to elemental nitrogen and water (17).

               4NO + 4NH3 + 02 	> 4N2 + 6H20

               2N02 + 4NH3 + 02 	> 3N2 + 6H20

Because the NO  composition of the flue gas from combustion sources is
primarily NO, the stoichiometric NH,/NO  mole ratio is approximately 1:1.
                                   A
An operating temperature of 570-750 F is required to catalyze NO  reduction.
                                                                ^
Below this temperature range,  the reaction between ammonia and NO  slows
significantly.  Above this temperature range,  ammonia is oxidized to NO  and
the catalyst can be damaged thermally.
     There are two possible SCR configurations.   The first design is a
hot-side or high-dust SCR where the SCR system is located between the
economizer and air preheater.   The second design is a cold-side or low-dust
SCR where the SCR is located typically downstream of the particulate control
device and possibly downstream of the FGD system.  Figure 3.3-1 presents
both SCR configurations.   An economizer bypass can be incorporated into the
system to control low catalyst temperature excursions.
     SCR catalyst formulations are typically based on oxides of titanium and
vanadium.   Other active metals such as platinum, palladium, and copper can
also be used.  In addition to  reducing NO  to N?, the catalyst also promotes
                                         A     £
the oxidation of SO- to S03 which in turn reacts with NH3 and H20 to form
ammonium bisulfate (NH^HSO^)  as the flue gas cools.  Ammonium bisulfate
deactivates the SCR catalyst by coating active catalyst sites and the air
                                    3-25

-------
                     Cold Sioo S«>tem
                        Ash
                     Hot Siao System
Figure 3.3-1.   Possible  SCR Configurations.
                     3-26

-------
 preheater elements reducing heat transfer efficiency and increasing metal
 corrosion.
     Catalyst elements can be produced in two general forms:  composed
 solely of catalyst material (homogeneous) or composed of either a ceramic or
metallic substrate that is coated with catalyst material (heterogeneous).
Each type of catalyst element has distinct advantages.  For example, the
homogeneous element will not lose activity even if the outer layer is lost
due to erosion by flyash; however, eventually the physical strength can be
affected.  To overcome catalyst degradation, the inlet face of the catalyst
can be protected with a hardened inert material.  Coated catalyst elements
also have the potential for erosion.
     The coated element will not be structurally weakened, but catalytic
activity can be lost.  Metallic substrates are more resistant to breakage
and cause less pressure drop than a ceramic substrate due to thinner walls.
Alternatively, the ceramic material has better catalyst bonding properties,
is acid resistant, and lightweight.
     Honeycomb, plate, and pipe catalyst supports are preferred for use with
coal-fired applications.  A number of tests were performed on honeycomb and
plate catalysts (18).  The honeycomb shape is the most common for reasons of
strength and ease of handling.   The honeycomb shape consists of a square
block with parallel  channels passing through it.  The block is typically
6-20 inches on a side and up to 40 inches in length.   The channels can be
either square, hexagonal,  or triangular in shape.   Plugging by particulates
is typically not a major problem when a vertical downflow reactor
arrangement and regular soot blowing are used.

Selective Non-Catalytic Reduction--
     Selective non-catalytic reduction (SNR) is an alternate approach for
post-combustion reduction of NO  to N-.  The primary  SNR technologies
involve injection of ammonia or urea (decomposes to ammonia)  into the
convective section of the boiler at 1600-2000°F; no catalyst is used.  At
temperatures above and below this range,  ammonia is oxidized to NOX or exits
as unreacted NH,, respectively.   In general, because  of the short residence
time at these temperatures in the convective section  and the difficulty of
thorough ammonia/flue gas mixing,  NO  reductions are  lower (40-80 percent)
                 \.
                 £
                                    3-27

-------
 and  ammonia  emissions  are  higher than with SCR.  Because a catalyst Is not
 required,  SNR  costs  are  significantly lower than for SCR.  However, because
 of the  lower performance level and limited experience with SNR on coal-fired
 boilers, the rest of this  section is limited to SCR technology.

 Applicability

     Since the SCR process requires flue gas at a specific temperature,  it
 is simpler to  apply  it to new boilers than to existing boilers.  In new
 boilers, the SCR system  is normally placed between the economizer and air
 preheater as economizer exit temperatures of utility boilers are typically
 in the  range of optimum SCR operation.   Retrofit installations will require
 flow modifications and additional ductwork to divert and return flue gas to
 the existing flue gas handling system.   Alternate arrangements include
 installation of the  reactor vessel  downstream of a hot-side participate
 control system or, by combining with flue gas reheat,  downstream of the
 existing emission controls for SO. and participates.  This last approach may
 be the most  suitable for many retrofits due to boiler limitations.
     SCR is  considered a commercial  technology; however, it has not been
widely applied in the U. S.  More than 30,000 megawatts of coal-fired
 utility boiler SCR systems have been operated in Japan and West Germany.
 Three coal-fired pilot SCR demonstration projects have been conducted in the
 U. S.  The only full-scale U.S. application to a utility boiler is  a gas-
 and oil-fired demonstration unit at Southern California Edison's Huntingdon
 Station.  There are  also three industrial boilers and  a number of gas
 turbines and process heaters using SCR for NO  control, all in California.
     Host of these applications involve very low-sulfur fuels; this
 applicability of SCR technology to high-sulfur fuels has not been
commercially demonstrated to date.   A primary concern  is the impact of high
SO* levels  on catalyst and air preheater plugging and  corrosion due to
formation of ammonium bisulfate.   Air preheater plugging is most severe  in
units which  fire high-sulfur fuels  and  remove particulates upstream of the
NO  reactor.   When particulates are removed downstream, plugging problems
are significantly reduced.   It is believed that the particulates produce a
sandblasting effect that cleans the  air preheater elements and that some  of

                                    3-28

-------
the NH4HS04 condenses on the flyash particles rather than the air preheater.
Air preheater plugging can also be reduced by limiting NH- slip to about 3
to 5 ppm  (NH3 slip  is the reacted NH, exiting the SCR reactor) and
redesigning the air heater.  However, reducing NH3 slip will require
increasing catalyst volume if a specific NOX removal rate is to be
maintained.  Off-line water washing of the air preheater may also be
required  to minimize pressure drop; soot blowers are generally ineffective
at removing NH^HSO^ deposits from the air heater on-line.
     NH.  from an SCR reactor does not impair FGD SO, removal performance.
However,  in some cases, the wastewater may require treatment to remove
nitrogen  (nitrate) compounds.  An unresolved concern is the potential for
NH, slippage or NH.HSO. to affect the performance of a downstream
particulate control device (e.g., by increasing pressure drop across a
downstream fabric filter or ESP).

Performance
     SCR is capable of 80 to 90 percent NO  reduction.  The primary
variables which determine the amount of NO  removed are the amounts of
                                          ^
ammonia and catalyst used; increasing ammonia and catalyst yields higher
removals.  However, increasing the amount of ammonia injected also increases
ammonia slip.  For this reason, the NH3 slip limit set for the initial NOX
concentration is the key catalyst sizing parameter, rather than the required
NOX reduction.
     Catalyst bed design is usually specified in terms of a space velocity
or area velocity.  The space velocity is the flue gas volumetric flow rate
divided by the catalyst bulk volume, while area velocity is flue gas
volumetric flow rate divided by the active catalyst area.  High NO
reductions require low space (or area) velocities and, hence, large
quantities of catalyst.  Low velocities are also required to control the
amount of NH, slip.  To reduce NH, slip to 3-5 ppm levels, space velocities
            o                  i j
in the range of 2,000-2,500 hr   are generally required.  Furthermore, NH3
slip increases dramatically at NH3:NOX injection ratios over 0.8 for high
space velocities and 0.9 for low space velocities.  This constraint
contributes in holding SCR to a practical NO  removal limit of about 80

                                    3-29

-------
   percent.  The exact relationships between velocity and NOV reduction and NH,
                                                            A          .       O
   emissions, however, are specific to the proprietary catalysts used by the
   various SCR manufacturers.

   Cost

        The capital cost of an SCR system is a function of several factors
   including NO  reduction level, flue gas flow rate, fuel type and sulfur
   content, and retrofit/new installation (19).  Operating costs are primarily
   a function of catalyst life.
        The following table presents the range of values for capital cost,
   annualized cost, and cost per ton of pollutant removed for hot side SCR
   systems using IAPCS.  The lower costs are for a large unit with a low
   retrofit factor and the higher costs are for a small  unit with a high
   retrofit factor.  Figures 3.3-2 through 3.3-4 show these costs as a function
   of boiler size and catalyst life.

                                                      Ranoe
       Capital Cost ($/kW)                         85.4 to 240.6
       Annualized Cost (mills/kWh)                  6.1 to  39.4
       Cost Per ton of NOX Removed ($/ton)        1,856 to 12,334.1

     Estimates of catalyst life vary from 1-5 years.   For the purpose of this
analysis, 1 and 3 year catalyst life is assumed.   Appendix G contains tables of
costs as a function of boiler size, retrofit difficulty,  capacity factor,  and
catalyst life.
                                       3-30

-------
o
Q.
z

JC
oc
Ul
Q.
      150
      140
     130 -i
     120 -
                                                         Legend

                                                    Catalyst Life = 1 yr or 3 yrs
                                                    Capacity Factor = 50%
                                                    Catalyst Cost = $23.342/ton
     110 -
     100 -
        100
                   300
                                1.100
                                1.300
        Figure 3.3-2.  Selective Catalytic Reduction - Capital Cost
                          Current 1988 Dollars
      19
      18 -

      17 -


      16 -

      15 -

      14 -

      13 -

      12—
                                                         Legend
                                                    •   Catalyst Life = lyr
                                                    +   Catalyst Life = 3yr
                                                    Capacity Factor =  50/5
                                                    Catalyst Cost = $23,342/ton
        100
                   300
500
                                    MW
700
900
1.100
1.300
         Figure  3.3-3.  Selective Catalytic Reduction -  Levelized
                           Annual  Cost, Current  1988 Dollars
                                     3-31

-------

Q.O
5.4
5.2 -
 5 -
4.8 -
4.6 -
4.4 -
4.2 -
 4 -
3.8 -
3.6 -
3.4
3.2 -
 3 -
2.8 -
2.6 -
2.4 -
        100
                                                          Legend
                                                    •    Cotolyst Life = 1yr
                                                    •*•    Catalyst Life = 3yr
                                                    Capacity Factor = 50% .
                                                    Cotolyst Cost = S23,342/ton
             300
500
  700
MW
900
1.100
1.300
       Figure 3.3-4.  Selective Catalytic Reduction - Cost  Per  Ton
                         of NOx Removed, Current  1988  Dollars
                                        3-32

-------
3.4  FURNACE SORBENT  INJECTION

Description

     Furnace sorbent  injection (FSI, or LIMB) reduces S02 emissions by
injecting powdered limestone or hydrated lime into the upper furnace of a
coal-fired boiler.  These sorbents decompose at high temperature to form
calcium oxide which reacts with SO. to form calcium sulfate (CaS03) (20).
Limestones ranging from calcitic (mainly CaC03, very low in MgC03) to dolomitic
(containing roughly equal amounts of Ca and Mg) have been tested.  Atmospheric
and pressure-hydrated limes have also been used.  The extent of S02 removal
that can be achieved depends on the flue gas composition, temperature, and
quench rate at the point of sorbent injection; sorbent composition and surface
area; and the calcium-to-sulfur ratio (Ca/S) and degree of mixing between the
sorbent and S02 in the flue gas (21).
     The resulting calcium sulfate and unreacted sorbent are collected with the
flyash in a baghouse or electrostatic precipitator (ESP).  Collected sorbent
and flyash can be either recycled to the furnace to increase sorbent
utilization or discharged to the solid waste disposal  facilities.  Figure 3.4-1
is a generalized diagram of the FSI process.
     The advanced silicate (ADVACATE) process (22) can be added to FSI (or
stand alone process) for better SO- removal efficiency.   Preliminary pilot
tests on ADVACATE system have resulted in 90+ percent  S02 removal with a
calcium to sulfur ratio near 1.0.   In this process cyclones may be used to
knock out coarse dust upstream of the ESPs.  Part of the high calcium flyash in
the front section of the ESPs,  silicate,  is slurried under temperature and
pressure.  Silica in the fly ash dissolves and reacts  with calcium to form high
surface area silicates.  The slurry is mixed with the  remaining high calcium
fly ash to form a damp powder,  which is then injected  into the ductwork.

Performance

     The most important variable affecting the calcination and sulfation
reactions is the gas temperature at the point of sorbent injection.  As shown
in Figure 3.4-2,  a peak in S02  removal occurs for most sorbents in the range

                                       3-33

-------
Sorbent
    Figure  3.4-1.  Simplified  schematic  of furnace  sorbent injection.
              100
         35
         I
         2
         8"
               80 -
60 -
40 -
               20 -
                            Genstar Pressure
                          Hydrated Oolomitlc Lima
                                  (1)
                                   Longview Lime
                                        (2)
Marianna
Limestone
  (2)
                 St. Genevieve
                  Limestone
                     (2)
                                                  Ca/S in 0
                 1400
               1800
                                           2200
                                                        2600
                          Maximum Injection Temperature, F

          Figure  3.4-2.  Peak  sorbent reactivity as a  function
                          of temperature (2 3).
                                       3-34

-------
between 1800°F and 2200°F.  In most boilers, these temperatures are found in
the upper furnace.  Because limited residence time is available in this zone to
allow thorough mixing of the sorbent and combustion gases, rapid distribution
of the sorbent over the full boiler cross-section is essential.  At
temperatures higher than 2200°F, sorbent deactivation (due to sintering of the
sorbent surface and instability of the reaction products) lowers S02 removal.
At temperatures below 1800°F, reaction rates for both calcination and sulfation
are significantly reduced.
     The sorbent quench rate (i.e., the rate of change in the flue gas
temperature after the point of sorbent injection) also affects SO- removal,
with SO. removal decreasing as quench rate increases (24).  As quench rates
increase,  sorbent residence time in the 1800-2200°F window is correspondingly
reduced.  Expected SO. capture as a function of quench rate,  Ca/S ratio, and
sorbent are shown in Table 3.4-1.  This factor is especially significant for
cyclone-fired boilers for which quench rates in the upper furnace can approach
1500°F/sec.
     Sorbent characteristics such as chemical composition and surface area also
affect the sorbent injection rate needed to achieve good SO.  removal.  The
surface area of several  sorbents evaluated in a pilot-scale study are shown in
Figure 3.4-3.  Pressure-hydrated dolomitic lime was the most  effective for
removing S02; limestone was the least effective.   Techniques  that have been
studied for improving sorbent surface area include thermal pretreatment and use
of chemical promoters.  Although these techniques have increased measured
surface area, pilot-scale combustion tests have not conclusively demonstrated
their effectiveness in improving SO- removal.
     Sorbent residence time and distribution in the flue gas  also impact S02
removal.  As shown in Figure 3.4-4, longer residence times give the sorbent and
SO. more time to react and increase S02 removal.   However, for retrofit
applications, the residence time of the boiler is already established and
cannot be  readily varied.   Optimal  location, number,  and type of injection
nozzles can maximize mixing of the flue gas and sorbent.
                                       3-35

-------
     TABLE 3.4-1. SO, CAPTURE AS A FUNCTION OF Ca/S RATIO,  QUENCH RATE, AND
                    '               SORBENT
                                         	 Ca/S
Quench rate = 900

  limestone                   15-16          26-29          35-43          42-56
  hydrate                     19-20          35-40          53-56          69-70
  CPHC                        27-33          48-54          68-71          88-94

Quench rate = 700

  limestone                   17-19          29-31          38-44          45-57
  hydrate                     24-26          40-46          57-62          73-76
  CPH                         33-38          54-62          75-79          92^95

Quench rate =500

  limestone                   18-22          31-33          41-45          48-58
  hydrate                     28-32          45-52          61-68          77-82
  CPH                         40-43          60-70          81-89           95+

Quench rate = 300

  limestone                   19-25          32-36          44-46          51-59
  hydrate                     33-38          49-58          66-74          82-88
  CPH                         46-48          67-78          88-95           95+
aSO- capture is expressed as a percentage.

 Quench rate is expressed as °F per second  for the sulfation "window"  of  2200
 to 1600°F.

CCPH = caldtic pressure hydrate.
°F
                                       3-36

-------
     o
     CO
          100
          80 -
          60 -
          40
          20 -
                   Pressure-Hydrated
       Calcium      Dolomitic Lime
       Hydroxide
Marianna
Limestone
                                St. Genevieve
                                 Limestone
                         10         20         30
                         Calcine surface area, m2/g
                                           40
FIGURE 3.4-3.  S02 removal as a function of calcine surface area (23).
                                           Residence Time
                                             1.54 sec
         0123456
FIGURE 3.4-4.   S02 removal  as a function of Ca/S ratio and residence time
               using limestone (23).
                                  3-37

-------
      Three  commercial-scale utility projects in the U. S. are demonstrating
 furnace  sorbent  injection on eastern bituminous coal-fired boilers.  The
 first  project, sponsored by EPRI, U. S. EPA, and Richmond Power and Light,
 involves a  61 MM tangentially-fired boiler, Whitewater Valley No. 2 at
 Richmond Power and Light.  The objective of this project is to evaluate
 different combinations of sorbents and coals (with varying sulfur contents),
 to assess process performance variables and determine retrofit control
 costs.  Testing began in 1987 with sorbents including hydrated lime,
 limestone,  dolomite hydrated lime, and "promoted" lime sorbent (25).
 Preliminary results of short-term tests indicate that sulfur dioxide
 emissions can be reduced by 25-40% at calcium to sulfur ration of 2.  A flue
 gas humidification system was designed and installed to counteract
 degradation of ESP performance resulting from use of sorbent injection.
     The second demonstration is a 105 MW wall-fired unit,  Edgewater No. 4,
 owned  and operated by Ohio Edison.  This has been funded by the U. S. EPA,
 DOE, and the Ohio coal Development Office.  Testing that began in 1987
 showed that the technology can exceed its sulfur dioxide control  goal of 50%
 at a calcium to sulfur ratio of 2 (26).  The demonstration  did encounter
 problems with downstream particulate control because of the high resistivity
 ash produced during sorbent injection.   A full-scale flue gas humidification
 system was added in 1988 to reduce ash resistivity and increase overall
 sulfur dioxide removal.
     The U. S. EPA and Combustion Engineering have begun the third
 demonstration—a program to demonstrate furnace sorbent injection on a
 tangential, coal-fired utility boiler at Virginia Power's Yorktown Unit No.
 2.  The overall objective is to achieve a significant reduction
 (approximately 50%) in sulfur dioxide while minimizing any  negative effects
 on boiler performance during long-term operation of an integrated sorbent
 injection system and a low-NO  firing system (27).   For moderate calcium to
 sulfur ratios (2-3:1), S02 reductions of 40-53  percent were obtained using
 calcium hydroxide and of 25-42 percent  were obtained using  limestone (28).
     Two Canadian utilities (Saskatchewan Power and Ontario Hydro) also are
 conducting full-scale furnace sorbent injection tests on lignite and
medium-sulfur bituminous coals,  respectively.   Using a limestone slurry, the
                                    3-38

-------
 Ontario  Hydro demonstration  system was  able to remove  up  to  70% of  the
 sulfur dioxide  at  a  calcium  to  sulfur ratio of 2.5  to  3.
      Baltimore  Gas and  Electric, Atlantic  Electric, the Electric  Power
 Research  Institute,  and  Babcock and Wilcox have co-funded a  program to
 evaluate  the potential  of dry sorbent injection technology on
 cyclone-equipped units  (29).  Upper-furnace sorbent injection  for sulfur
 dioxide  capture has  been examined using a  6 million Btu/hr cyclone-equipped
 boiler simulator.  The major operating  problems which  resulted was  a
 significant increase in  opacity from 10 to 60 percent.

 Applicability

      FSI  can be applied  to both new and retrofit applications.  However,.
 under existing  utility NSPS, application to new boilers will be limited to
 units burning low-sulfur coals  where 70 percent SO- removal  is acceptable.
 Retrofit  of FSI technology into an existing boiler will require installation
 of additional equipment  as well  as modifications to the boiler and  particu-
 late control device.  Additional equipment includes sorbent  preparation,
 storage,  handling, and injection systems;  participate  control modifications;
 and a new or expanded ash handling system.
     FSI  can increase the particulate loading in the boiler  and particulate
 control  device by two- to three-fold,  depending on the Ca/S  ratio and the
 coal sulfur and ash  content.  The increase in particulate  loading as a
 function of Ca/S ratio and coal  sulfur content for a typical 10 percent ash
 coal .is shown in Figure 3.4-5.
     Fouling and slagging of boiler tubes may become excessive due to
 increased solids loading and, particularly for acidic coal ashes,  reduced
 ash fusion temperatures.  These deposits reduce water wall heat transfer,
 and as a result, the superheater and reheater steam temperature.   Increased
dust loads may also  increase erosion.   Sootblower upgrading  by increasing
the number of blowers, blowing  pressure, and frequency should reduce the
 severity of. heat transfer problems.   Finned-tube economizers probably will
need to be replaced with larger bare-tube economizers with an in-line
arrangement.  In some cases,  derating  of the boiler may be necessary even
after these modifications.
                                    3-39

-------
       a?
       d>
       '•6
       w
       •g
       "5
       CO
             500
             400 -
             300-
             200 -
             100 -
                                            CaJS
                      0.5
                            I
                            1.0
2.0
           3.0
                       4.0
                              Coal Sulfur Content, %
   FIGURE  3.4-5.   Increase in  solids  loading as  a  function of coal  sulfur
                   content  and  Ca/S  ratio  for a typical  10% ash coal  (17).

     FSI may also  affect performance  of the  existing  particulate control
device  (generally  an  ESP)  due  to  the  increased solids loading,  the  high
alkalinity of the  sorbent,  and low  S03 level  in  the flue  gas  (30).
Pilot-scale tests  indicate that flue  gas  humidification can return  ESP
performance to near pre-sorbent Injection  levels for  most ESP's  (26).   For
baghouses, additional bag  area, more  frequent cleaning, and additional
induced draft fan  capacity may be needed  to  handle the  increased particulate
loading and pressure drop.  Also, the capacity of the ash handling  system
may need to be increased to handle  the extra solids.   Because  of the  large
amount of unreacted lime present in the solids, conversion  of  wet ash
handling systems to dry handling will also be necessary.
     On the positive side,  the  lower  S02/S03 content  of the flue gas  and
increased alkali In the flue gas particulate may improve  boiler  efficiency
by allowing lower boiler exit  temperatures without cold-end acid corrosion.
                                    3-40

-------
   Cost
        Estimated total capital costs for retrofitting FSI on an existing power
   plant are much lower than for conventional FGD technologies (31).  Depending
   on the design of the boiler and existing participate control system, total
   capital costs range from $35-50/kW if the existing participate control
   system can be used without significant upgrading.  If the existing ESP must
   be replaced by a new fabric filter (e.g., due to space constraints that
   limit installation of additional ESP plate area), total capital costs can
   approach $150/kW.
        Because of FSI's relatively low captial  cost, total busbar costs are
   very sensitive to operating costs associated with sorbent purchase and
   disposal.  Major factors affecting these costs are the Ca/S ratio required
   to achieve a given SO- reduction, coal sulfur content, and unit costs for
   sorbent and waste disposal.  Also, because of the relatively low capital
   cost of FSI, the cost per ton of S02 removed is relatively uniform over a
   wide range of percent removals and plant sizes.
        The following table presents the range of values for capital cost,
   annualized cost, and cost per ton of pollutant removed using IAPCS for FSI
   with humidification and upgrading of the existing ESPs (SCA = 300).   A
   sulfur dioxide removal  efficiency of 70% was  assumed based on the use of
   humidification to improve both ESP performance as well as S02 capture.  The
   lower costs are for a large unit burning low sulfur coal and the higher
   costs are for a small  unit burning high sulfur coal.   Figures 3.4-6  through
   3.4-8 show these costs as a function of boiler size and coal  sulfur  content.

                                                      Range
       Capital Cost ($/kW)                         23.6 to 81.4
       Annualized Cost (mills/kWh)                  3.6 to 21.8
       Cost Per ton of S02 Removed ($/ton)         516.1 to 4,995

     Appendix  H contains tables of costs as a function of boiler size,  capacity
factor, coal  sulfur and SO- reduction.
                                       3-41

-------
x
oc
Ul
a.
                                                     Capacity Factor = 50%

                                                     502 Removal = 70% with

                                                         Humidificotion
        100
300
500
700
900
1.100
1,300
                                   MW
         Figure  3.4-6.  Furnace  Sorbent Injection - Capital  Cost
                          Current  1988  Dollars
                                                             1% S

                                                             2% S.

                                                         o   3% 5

                                                             V7. 5

                                                    Copacity Factor = 50%

                                                    S02 Removal = 70% with

                                                         Humidification
        100
                  300
                             500
                     700
                     900
                    1,100
                    1.300
                                    MW
   Figure  3.4-7.  Furnace Sorbent  Injection  - Levelized Annual Cost

                   Current  1988 Dollars
                                      3-42

-------
!-w
«3
0.0
                                                  Copocity Factor = 50%

                                                  S02 Removal = 70% with
                                                      Humldificotion
        100
300
500
700
                                   MW
900
1,300
         Figure 3.4-8.  Furnace Sorbent Injection  -  Cost  Per Ton

                        of  S02 Removed, Current  1988 Dollars
                                    3-43

-------
3.5   LOW-TEMPERATURE SORBENT  INJECTION

Description

      Low-temperature sorbent  injection (LTSI) involves injection of dry or
s lurried sorbents into the duct area between the air preheater and participate
control device.  Several different process concepts have been proposed.  These
include injecting dry sodium  compounds, powdered lime hydrates in conjunction
with water or steam, and lime slurries.  Injection locations range from
immediately following the air preheater to the inlet of the particulate control
device.  Potential advantages of these processes include simplicity and low
capital cost of sorbent preparation and injection equipment, and adaptability
to retrofit applications.

Sodium-Based Sorbents--
      In sodium-based systems, SO. is captured in a gas-solid reaction.  Key to
this reaction is the thermal decomposition of the sorbent to form a porous,
high surface area matrix of reactive sodium carbonate (NaCO)..  The optimum
injection temperature is 275-350°F, physically just downstream of the air
heater.  Because of its ability to "popcorn" during heating, nahcolite (a
naturally occurring mineral containing 70-90 percent sodium bicarbonate--
NaHC03) has very good reactivity and sulfur reduction performance.  Trona (a
naturally occurring form of sodium sesquicarbonate--Na2C03'NaHC03'2H20) is more
densely packed and provides less surface area for reaction with S02> but is
easier to handle.  In general, 20-30 percent of SO. removal occurs in the duct
while 70-80 percent of S0« removal occurs in the particulate control system.
For this reason, a baghouse which allows considerable contact between the flue
gas and sorbent is preferred.
     Sulfur dioxide removal . performance with nahcolite and trona during
full-scale tests by the Colorado Springs (Colorado) Department of Utilities is
shown in Figure 3.5-1.  At SO. removals of up to 80 percent, nahcolite
utilization approached 100 percent.  NOX reductions of up to 23 percent were
also obtained during these tests (32).   Earlier tests with soda ash (NaOH) were
relatively ineffective.
                                       3-44

-------
a

u
a

CVJ
a
u
u
a
u
a.
                               ALL  LOAD CONDITIONS

                               COLORADO COAL
                               O  SODIUM SeSOUlCARBONATE

                               •  SODIUM BICARBONATE
         0,0
       Figure 3.5-1.   S02 removal as a function of normalized


                     stoichiometric ratio (NSR). (32)
                               3-45

-------
Calcium-Based Sorbents--
     Both dry and slurried calcium-based  sorbents are being investigated.  For
both sorbents, flue gas cooling to within 20-30°F of adiabatic saturation
is critical  (approximately 125°F  in a typical coal-fired boiler system).  To
achieve these temperatures with dry sorbents, water and/or steam is  injected
into the flue gas downstream of the air preheater.  In pilot-scale tests at
these temperatures, SO. removal rates exceeding 75 percent have been achieved
at Ca/S ratios of 2 using conventionally hydrated lime (33,34).  Depending on
the sorbent  and flue gas temperature, 70-90 percent of SO* removal occurs
during the Initial few seconds while the sorbent Is suspended In the flue gas.
The remaining 10-30 percent occurs in the particulate control  device.  The
quantitative advantages of a baghouse versus an ESP for SO. control are not
well defined.
     Significant similarities in process chemistry exist between conventional
spray drying and low-temperature sorbent injection.   For slurried sorbents, the
major differences are the available residence time and the requirement for
rapid sorbent drying to prevent caking of wet sorbent on duct walls.
Conventional spray dryers have a reactor vessel residence time of 5-12 seconds,
whereas duct spray drying must accomplish particle drying and SO. reaction in
two seconds or less.  As a result, proper choice of atomizer design, slurry
concentration, and slurry injection rate is essential.
     The concentration of sorbent in the slurry is a major variable in removing
SO. from the flue gas.  In conventional  lime-based spray dryers,  the sol Ids
content of the slurry ranges from 10-40 percent by weight (lime slurries cannot
be readily pumped at solids concentrations above 35 percent).   To avoid caking
on duct walls, the moisture content of the dried droplet should be less than 20
percent.  Although high solids concentration slurry will  dry more rapidly and
have less impingement on duct surfaces,  by drying faster the slurry quickly
loses its reactivity and has less time to remove SO. from the  flue gas.
Testing will be required to determine the optimum slurry concentration and
injection rate for a given SO. concentration and duct geometry.
                                       3-46

-------
     Two basic atomizer designs  are available:  two-fluid and rotary.  The
two-fluid atomizer can project a narrow cone of spray, thus reducing
impingement of slurry on  the duct walls.  However, an air compressor is needed
and will be a significant energy consumer.  Also, early experience with
two-fluid systems in conventional spray dryers encountered problems with
atomizer erosion and plugging.
     Rotary atomizers offer reliability, thorough mixing, and flexibility in
varying slurry feed rates without altering the droplet size distribution.
Rotary atomizers are the predominant atomization method used in conventional
spray drying.  However, rotary atomizers offer less control over the spray
pattern and may be more prone to duct wall caking problems.
     By using a dry sorbent, caking problems are reduced, but flue gas
humidification is required to achieve reasonable sorbent utilization.  Dry
injection has two potential mechanical advantages over slurry injection.
First, the sorbent material is handled in a dry form rather than as a slurry,
thus avoiding the need for a slurry mixing and handling system.  Second, by
separately feeding water and sorbent, operation of the system can be better
controlled than with a single feed system.  This is especially significant with
high-sulfur coals where the quantities of calcium hydroxide required for
significant S02 removal (>50 percent) are large.  In such a case, the quantity
of water required to maintain slurry rheology could result in excessive cooling
of the flue gas.   This flexibility in sorbent/water injection also can
facilitate system adjustments required by fluctuations in boiler load.

Performance

     Several  in-duct sorbent injection processes are being developed.  The
U. S.  Department of Energy (DOE)  Flue Gas Cleanup Program is sponsoring
development of low-cost emissions control  technology that can be installed on
existing coal-fired power plants  if acid rain legislation is enacted.  The
emphasis of this program is on in-duct sulfur capture.  In 1985, three projects
were undertaken to test duct injection technology on slipstreams from operating
coal-fired power plants.
     One of the processes tested  in the DOE program was the Hydrate Addition at
Low Temperature (HALT) process.   In this process,  calcium hydrate particles

                                       3-47

-------
 are  Injected  Into  the  flue gas  stream before the gas  Is cooled by
 humidification with  a  fine spray of water.  Sulfur dioxide removal efficiencies
 ranged  from 42 to  52%  (35).  Higher SOp reductions of 60-75% have been obtained
 with corresponding 23  to 28% sorbent utilization, although wall wetting and
 deposition problems  were encountered (36).  Another process being evaluated in
 the DOE program  is the Bechtel  Confined Zone Dispersion (CZD) process which
 captures sulfur  dioxide by spraying a finely atomized slurry of hydrated lime
 into the duct of a utility boiler between the air heater and the ESP.  Two
 reactive lime reagents were used for a pilot-scale test.  Sulfur dioxide
 removal efficiencies greater than 50% were achieved (Ca to S ratio of 1.5:1)
 using either pressure  hydrated  dolomitic lime or calcltlc lime.  A five-month,
 large-scale test was carried out when the CZD process was retrofitted onto one
 of two parallel  flue ducts on a 140 MW unit.  Wall wetting and deposition were
 generally a serious  problem.
     The General Electric in-duct scrubbing (IDS) process also has been tested
 at the pilot plant scale under  DOE sponsorship.  This type of scrubbing is
 accomplished by  injecting a slaked lime slurry with a rotary atomizer directly
 into existing ductwork.  Sulfur dioxide removal takes place in three areas: the
 slurry injection zone, the evaporation zone, and the drift zone.   Test results
 indicated a lime utilization of 35% calculated for a calcium to sulfur ratio of
 1.1 at an S02 removal efficiency of 39%.
     Another type of Injection  process being sponsored by'EPA is E-SOX>   This
 process appears to be especially appealing for retrofit applications since it
makes use of an existing ESP modified for injecting a lime slurry into the
 first field of the ESP.  The front end of an existing ESP is converted to a
 spray chamber where the gaseous sulfur dioxide comes Into contact with lime
 slurry droplets.  The remaining ESP fields are modified with cold pipe
 prechargers to bring the ESP particulate emissions back to current levels.
     The Ohio Coal  Development Office,  U.  S. EPA, Babcock & Wilcox,  and  Ohio
Edison are jointly sponsoring an evaluation of this process at a 5 MW pilot
plant;  a 10-month field pilot test program at Ohio Edison's R.  E.  Burger
Station was scheduled to begin  in early 1989 (37).   Testing will  focus on
demonstrating a minimum 50% sulfur dioxide removal  rate, maintaining acceptable
ESP performance with a reduced collector area (to make room for the  spray
chamber) and increased particulate loading,  and showing that solid waste can be
disposed of safely and economically.
                                       3-48

-------
      Another  variation  of  sorbent  injection  is a combination of furnace and
 duct  injection.   Tampella's  LIFAC  process, combines  in-furnace injection with a
 subsequent  vertical  humidification reactor for activating the unreacted calcium
 oxide with  water,  thus  increasing  limestone  utilization.  The process can
 achieve  an  S02 removal  efficiency  of 80-85%  with commercial grade pulverized
 limestone at  calcium to sulfur ratios of 2.0 or less.  Two full-scale
 applications  of this technology exist in Finland, and pilot-plant tests reveal
 that  the process  is  effective with coals of  sulfur contents from 0.5 to 4.3%.
      The Advanced  Silicate (ADVACATE) process is an  in-duct dry injection
 process  that  uses  a  highly reactive calcium  silicate hydrate sorbent.  The
 large surface area and  high  moisture retention capability of the sorbent makes
 it very  reactive to  sulfur dioxide and more  effective than hydrated lime for
 dry injection.  The  sorbent  can be produced  by mixing a calcium-containing
 material (e.g., lime) and a  silica-containing material (e.g., fly ash) in water
 at elevated temperature  and  pressure.  The calcium silicate hydrate remains a
 free  flowing  dry material with a moisture content up to 35% (37).   Pilot and
 bench scale testing  have achieved greater than 80% SO- removal  at a calcium to
 sulfur ratio  of 2.5  (38).  The unreacted lime and fly ash collected in the
 first particulate control section  is hydrated to produce the calcium silicates.
 The ADVACATE  process can be  used without furnace injection of limestone by
 adding lime directly to the  hydrator.  Preliminary pilot testing of the
 ADVACATE process has shown a combined furnace/duct SO* removal  efficiency
 greater than  90% with a calcium to sulfur ratio near 1.0.

 Applicability

      Reflective of its development status,  a number of unresolved  issues remain
 regarding retrofit of low-temperature sorbent injection:-flue gas  temperature
 and velocity  limitations and their significance in boiler operations (e.g.,
 load  fluctuation); performance on boilers firing high-sulfur coals;  method of
water injection and limitations on flue gas approach to saturation temperature;
 sorbent distribution and residence time; sorbent utilization rates at various
 stoichiometrlc ratios,  injection temperatures,  and residence times;  choice of
 and impact on the exiting particulate control system; potential  for sorbent
 fallout and caking in the duct and particulate control  systems;  and disposal  of
 solid wastes  containing high levels of unreacted lime.   Because  of
                                       3-49

-------
 similarities,  much  of  the data  collected  from earlier  furnace  sorbent  injection
 testing  should be directly  transferable.
      Equipment for  LTSI  can  potentially be  installed without major alterations.
 To achieve  satisfactory  contact  between the  sorbent and flue gas, the  sorbent
 injection point  should be located at a point where the gas flow  is reasonably
 uniform.  Assuming  a typical  flue gas velocity of 50 feet per  second at full
 load, 50 to  100  feet of  straight duct downstream from the point  of injection
 will  be required for drying  slurried sorbents.  An estimated 75  percent of
 existing coal-fired power plants burning medium- and high-sulfur coals have
 adequate duct  lengths to meet this requirement (39).

 Cost

      Although  both  sodium and calcium compounds are can be used with LTSI,
 uncertainties  regarding the  availability and cost of sodium compounds  in the
 eastern U. S.  and the disposal of sodium wastes are expected to limit their
 use.  Therefore, cost estimates  in this section apply only to  lime-based
 systems.
      Preliminary estimates of direct capital cost for lime injection systems
 (assuming the  existing ESP can be used without significant modification) ranged
 from  $ll-16/kW for dry injection (33,40) to $27-32/kW for slurry injection
 systems (based on comparison with spray dryer costs).   Whether the capital cost
 differences between dry and  slurry injection systems (as well  as between
 furnace and low-temperature  injection) are real  or result from variations in
 costing procedures used by different estimators Is uncertain.  If significant
modifications  to the particulate control system are required, capital costs can
 increase several-fold.
      Sorbent utilization rates achievable with LTSI are also uncertain due to
 the lack of clear understanding of process chemistry and limited test data.
 Estimates of Ca/S ratios required to achieve 50 percent SO- removal  range from
0.75  to 2.5.  As a result, significant uncertainty exists regarding  operating
costs for LTSI.  Based on I) the uncertainties in capital and operating costs
for LTSI and 2) general similarities in hardware for LTSI and FSI,  it is likely
that the cost of the two basic approaches to SO* control  are similar.
Therefore,  the costs presented in Section 3.4 for FSI  are likely to  be
applicable to  LTSI.
                                       3-50

-------
3.6   REBURNING

Description

      Reburning  involves  bypassing  a portion of the fuel around the main
combustion zone and  injecting  it above the main burners to form a reducing zone
in which NO  is converted to reduced nitrogen compounds.  The overall process
is divided into three zones:

      t    Main Combustion Zone.  Approximately 80-85 percent of the total fuel
          to the boiler  is combusted in this zone.  The burners operate at
          air-fuel ratios less than or equal to those typical of a normal
          boiler (i.e.,  10 percent excess air).  No boiler modifications are
          required in this zone.
      t    Reburning  Zone.  The remaining 15-20 percent of the fuel is injected
          downstream of  the main combustion zone to create a fuel-rich
          reburning  zone.  Nitrogen oxides (NO ) produced in the main
          combustion zone react with hydrocarbon radicals formed by  partial
          oxidation  of the reburning fuel to form reduced nitrogen species such
          as NH3 and HCN.  A portion of these compounds are converted to N£.
          The degree of  NO  reduction depends on overall air-fuel ratio
          (typically 0.80 to 0.95), temperature, residence time,  and
          concentration  of NOX in the reburning zone.
     t    Burnout Zone.  Additional air is added in the cooler, upper furnace
          to create  overall fuel-lean conditions,  and ensure complete oxidation
          of the reburning fuel.  Reduced nitrogen species formed in the
          reburning  zone are converted to either NO or H*'

     Any fuel can be used as the reburning fuel.  However, U. S.  pilot studies
indicate that fuels with Tittle or no fuel  bound nitrogen, such as natural  gas,
can achieve greater NO  reductions.  These advantages are particularly
significant under operating conditions of lower NO  levels exiting the main
combustion zone (less than 200 ppm) and short residence times.   The volatility
                                       3-51

-------
 and  low  fuel nitrogen  content of  natural gas  are  the primary contributing
 factors.   SOg  reduction  also results  from  use of  sulfur-free natural gas for
 reburning.  The  potential exists  to further reduce SO- emissions by combining
 reburning  with furnace sorbent  injection.

 Applicability

     Because the main  combustion  zone of the  furnace operates at normal
 air-fuel ratios, gas reburning  1s applicable  to a wide range of wall,
 tangential, and  cyclone-fired boilers.  Boiler modifications for reburning
 involve  Installation of  additional fuel injection burners and air ports above
 the  top  or cyclone burner row.  Since the main combustion zone operates
 essentially under original design conditions, problems such as flame
 impingement and  poor carbon burnout are minimized.
     For retrofit applications, adequate space (and residence time) between the
 top  burner row and the furnace exit must be available for the additional levels
 of fuel  and air  injection.  If adequate space 1s not available, a loss of NO
 reduction performance  and/or boiler derating  at full load would likely be
 incurred.  Also, if natural gas is used for reburning,  a natural gas supply
 should already be available at or near the plant; otherwise, installation of a
 natural gas supply pipeline may cause retrofit costs to be unreasonable.

 Performance

     The development of reburning technology has occurred primarily in the U.S.
 and Japan,  with the largest demonstrations occurring in Japan.   In the U.  S.,
the Gas Research Institute (GRI) has sponsored bench-scale (25 KW) and
pilot-scale (3 MM) furnace tests at Energy and Environmental Research
Corporation (EER) (34).  Various fuels including natural  gas,  high-nitrogen
residual  oil,  and coal  have been evaluated as reburning fuels.   These tests
showed a NO  reduction capability of 50-75 percent when using 15-20 percent
natural gas as the reburning fuel.  Use of oil and coal as reburning fuels
resulted in less NO  reduction and increased carbon content 1n flyash.
     In the U.  S., three commercial  demonstration projects are currently
planned;  two of the projects use natural  gas as the reburn fuel  and one will

                                       3-52

-------
use coal as the reburn fuel.  These projects are planned for start-up in the
early 1990's.
     In Japan, boiler manufacturer Ishikawajima-Harima Heavy Industries (IHI)
has conducted pilot-scale tests as well as tests on two full-scale utility
boilers, a 600 MW and a 55 MM unit.  These tests were conducted using fuel oil
and pulverized coal as the reburning fuels, respectively.   The IHI pilot tests
showed reburning with pulverized coal could achieve 40-50 percent NO
reduction.   However, in the two full-scale field tests, NO  reductions were
less than 20-25 percent using oil or coal as the reburning fuel (41).
     Another Japanese manufacturer, Mitsubishi Heavy Industries (MHI), has
conducted tests on a 125 MW oil-fired utility boiler.  MHI has also over the
last 3 years conducted a full-scale demonstration on a 600 MW oil/coal dual
fuel-fired unit.  Low-nitrogen oil was used as the reburning fuel  in this
long-term demonstration, and achieved 50 percent NO  reduction (42).
                                                   n

Cost

     The range of cost estimates for natural gas reburning applied to
coal-fired utility boilers using the IAPCS cost model is presented below.   The
lower costs are for a large unit with a low fuel price differential (FPD).  The
higher costs are for a small unit with a high FPD.  Fuel price differential is
the cost per million Btu difference between the current coal and the reburn
fuel.   Figures 3.6-1 through 3.6-3 present capital cost, annualized cost,  and
cost per ton of pollutant removed as functions of plant size and FPD.  A NO
                                                                           ^
reduction of 60 percent was assumed.

                                                      Range
     Capital Cost ($/kW)                            6.8 to 34.2
     Annualized Cost (mills/kWh)                      1 to 16.7
     Cost Per ton of NOX Removed ($/ton)          384.7 to 6,418.8

     Appendix I contains tables of cost as a function of boiler size, capacity
factor, FPD and percent gas substitution.
                                       3-53

-------
 o
 a
                                                   15 % Gas  Substitution
                                                Fuel Price Oifferentiol =  1 $/MM8tu
                                                Fuel Price Differential =  2 $/MMBtu
                                                   25 % Gas  Substitution
                                                Fuel Price Differential =  1 $/MMBtu
                                                Fuel Price Differential =  2 $/MMBtu
                                                  Capacity Factor = 50%
         100
                    300
500
700
900
1.100
1,300
                                       MW
            Figure 3.6-1.  Natural Gas  Reburning -  Capital  Cost
                             Current  1988  Dollars
Jt
       10 -
       9 -
       8 -
       7 -
       6 -
                                                       Legend
                    15 % Gas  Substitution
                 Fuel Price Differential =  1. S/MMBtu
                 Fuel Price Differential =  2 $/MMBtu
                    25 % Gas  Substitution
                 Fuel Price Differentiol =  1 $/MMBtu
                 Fuel Price Differentiol =  2 $/MM9tu
                   Capacity Factor = 50%
       5 -
       4 -
         100
                    300
500
                                      MW
                                          700
                                                     900
                     1.100
                                                                          1.300
              Figure  3.6-2.  Natural Gas  Reburning -  Levelized
                                Annual Cost, Current  1988  Dollars
                                          3-54

-------
H.£ —
4 -
3.8 -
3.6 -1
3.4 -
3.2 -
^ 3 -
o e 2 8 -

a40 2 6 -*
£i 2-4 1
~ 2.2 -
2 -
1.8 -
1.6 -
1.4.-
1.2 -
1 -










— -— « 	 » 	 : 	
Legend
15 % Gas Substitution
• Fuel Price Differential = 1 $/MMBtu
o Fuel Price Differential = 2 $/MMBtu
25 % Gas Substitution
+ Fuel Price Differentiol = 1 $/MM8tu
A Fuel Price Differentiol = 2 $/MM8tu

Copocity Factor = 50%
ft •"•

" •- 	


h


i i i i i i I i i i i


















100 300 500 700 900 1.100 1,300
                      MW
Figure  3.6-3. Natural  Gas Reburning  -  Cost Per Ton
             of NOx Removed,  Current  1988 Dollars
                       3-55

-------
References
1.   Simbeck, 0. R., et al.  Coal Gasification Systems:  A Guide to Status,
     Applications., and Economics.  AP-3109, Electric Power Research Institute,
     Palo Alto, California,  1983, 424 pp.

2.   Rib, D. M. and D. R.  Plumley.  Experience at Cool Water with General
     Electric Combined Cycle Equipment.  Presented at the EPRI Coal
     Gasification and Synthetic Fuels for Power Generation Conference,
     San Francisco, California, 1985.

3.   Matchak, T. A., A. D. Rao, V. Ramanathan and M. T. Sander.  Cost and
     Performance for Commercial Applications of Texaco-Based Gasification
     Combined-Cycle Plants (Volume 1).  AP-3486, Electric Power Research
     Institute, Palo Alto, California, 1984, 112 pp.

4.   McCarthy, C. B. and W. N. Clark.  Integrated Gasification/Combined Cycle
     (IGCC) Electrical Power Production -- A Rapidly Emerging Energy
     Technology.  Presented at the EPRI Coal Gasification and Synthetic Fuels
     for Power Generation  Conference, San Francisco, California, 1985.

5.   delaMora, J. A., et al.  Evaluation of the British Gas Corporation/Lurgi
     Slagging Gasifier in  Gasification-Combined-Cycle Power Generation.
     AP-3980, Electric Power Research Institute, Palo Alto, California, 1985,
     196 pp.

6.   Hartman, J. J., T. A. Matchak, H. E. Sipe and M. Wu.  Shell-Based
     Gasification-Combined-Cycle Power Plant Evaluations.  AP-3129, Electric
     Power Research Institute, Palo Alto, California, 1983, 354 pp.

7.   Cool Uater Coal Gasification Program.  Environmental Surveillance and
     Monitoring/Worker and Safety Program Plan -- Special Air Emissions Report.
     Daggett, California,  1986.

8.   Dawkins, R. P.,.et al.  Cost and Performance of Kellogg Rust
     Westinghouse-Based Gasification-Combined Cycle Plants.  AP-4018,  Electric
     Power Research Institute, Palo Alto, California, 1985, 340 pp.

9..   TAG - Technical Assessment Guide (Volume 1).  EPRI P-4463-SR,  Electric
     Power Research Institute, Palo Alto, California, 1986.

10.  Pietruszkiewicz, J., et al.  An Evaluation of Integrated-Gasification-
     Combined-Cycle and Pulverized-Coal-Fired Steam Plants. AP-5950, Electric
     Power Research Institute, Palo Alto, California, 1988.

11.  Yerushalmi, J.  Circulating Fluidized Bed Boilers.  In:   Proceedings of
     the 88th National Meeting of the AIChE (Volume 1), 1980, pp. 490-521.

12.  WEPCO Plans Four-Unit, 500-MW AFBC Retrofit.  Electric Light and Power.
     January 1986, p. 18.
                                       3-56

-------
 13.  Update of Project Review Presentation by Bechtel Group,  Inc.  RP-1860-3,
     Electric Power Research Institute, Palo Alto, California, 1983.

 14.  U. S. Department of Energy, Report to Congress on the Relationship Between
     Projects Selected for the Clean Coal Technology Program  and the
     Recommendations of the Joint Report of the Special Envoys on Acid Rain.
     DOE/FE-0072, 1986.  28 pp.

 15.  Coal and Syn Fuels Technology.  DOE Announces Clean Coal Winners.  July
     28, 1986. p. 1.

 16.  Pope Engineers.  Conversion of Coal-Fired Boilers to AFBC, A Statistical
     Evaluation  (Draft),  pp. 14-18.

 17.  Miller, M. J.  S0? and NO  Retrofit Control Technologies Handbook.
     CS-4277-SR, Electric Powe? Research Institute, Palo Alto, California,
     1985, 366 pp.

 18.  Balling, L. and D. Hein. DeNO  Catalytic Converters for Various Types of
     Furnaces and Fuels, Development, Testing, Operation.  In: Proceedings:
     1989 Joint Symposium on Stationary Combustion NO  Control, Vol. 2,
     EPA-600/9-89-062b (NTIS PB89-220537), June 1989,xp. 7A-27.

 19.  Bauer, T. K. and R. G. Spendle.  Selective Catalytic Reduction for
     Coal-Fired Power Plants:  Feasibility and Economics.  CS-3603,
     Electric Power Research Institute, Palo Alto, California, 1984, 164 pp.

20.  McElroy, M. and G. Offen.   In-Furnace SO- Control  for Pulverized-Coal
     Boilers.  EPRI Journal, 10(7):53-55, 198S.

21.  Makansi, J.  Limestone Injection Achieves 50 Percent SO- Removal With
     Minimal Side Effects.   Power, 129(2):88-92, 1985.       *

22.  Chang, J. C. S. and C. B.  Sedman.  Scale-up Testing of the ADVACATE Damp
     Solids Injection Process.   In: Proceedings: First Combined FGD and Dry SO-
     Control Symposium, Vol. 3,  EPA-600/9-89-036c (NTIS PB 89-172175), March  *
     1989, p. 8-122.

23.  Makansi, J.  Understand System Effects When Evaluating Sorbent
     Injection.  Power, 129(6):35-39, 1985.

24.  Kokkinos, A., D.  C. Borio,  R. W. Koucky,  J. P. Clark,  C. Y.  Sun, and D. G.
     Lachapelle.  Boiler Design Criteria for Dry Sorbent SO  Control
     With Low - NO  Burners.  In:  Proceedings:  First Joint Symposium on Dry
     SO- and Simultaneous SO,/NOV Control Technologies, Volume 2,
     EPA-600/9-85-020b (NTIS^PB8§-232361), July 1985, p. 32-1.

25.  England, G. C., B. A.  Folsom, R. Payne,  T. M. Sommer,  M. W.  McElroy, P. J.
     Chappel, and I. A. Huffman.   Prototype Evaluation of Sorbent Injection
     with Humidification.  In:  Proceedings: First Combined FGD and Dry S0«
     Control Symposium, Vol. 1,  EPA-600/9-89-036a (NTIS PB89-172159), MarCh
     1989, p. 4-15.


                                       3-57

-------
 26.   Nolan,  P.  S.,  R.  V.  Hendriks,  and  N.  Kresovlch.  Operation  of  the
      LIMB/Humidifier Demonstration  at Edgewater.   In:  Proceedings:  First
      Combined  FGD  and  Dry SO,  Control Symposium,  Vol.  1,  EPA-600/9-89-036a
      (NTIS  PB89-172159),  Marth 1989, p.  4-1.

 27.   Goglnenl,  M.  R.,  J.  P.  Clark,  J. L. Marion,  R. W.  Koucky,  D.  K. Anderson,
      A. F.  Kwasnik, E. Gootzait, D. G.  Lachapelle, and  S.  L. Rakes.
      Development and Demonstration  of Sorbent  Injection for SO,  Control on
      Tangentially  Coal-Fired Boilers.   In: Proceedings:  First Combined  FGD  and
      Dry S07 Control Symposium, Vol. 1,  EPA-600/9-89-036a  (NTIS  PB89-172159),
      March  1989, p. 4-35.

 28.   Stuart-Sheppard,  I.  J.  and C.  J. Barnett.  Retrofit Applications for
      Control of SO- and NO  .   In: Proceedings  of  the  81st  Annual Meeting of
      APCA.  Dallas, Texas, June 1988.

 29.   Farzan, H., et al.   Pilot Evaluation  of Reburning  for Cyclone  Boiler NO
      Control.   In: Proceedings: 1989 Joint Symposium  on Stationary  Combustion
      NOV Control,  Vol. 1, EPA-600/9-89-062a (NTIS  PB89-220529),  June 1989,  p.
      3-?.  /

 30.   Dahlih, R. S., J. P. Gooch, and J.  D. Kilgroe.   Effects of  Furnace Sorbent
      Injection  on  Fly Ash Characteristics  and  Electrostatic Precipitator
      Performance.  In:  Proceedings:  First Joint  Symposium on Dry  SO,  and
      Simultaneous SO,/NOV Control Technologies, Volume  2,  EPA-600/9-85-020b
      (NTIS  PB 85-232361)7 July  1985, p.  29-1.

 31.   Lachapelle, D. G., N. Kaplan,  and J.  Chappell.   EPA's LIMB  Cost
      Model:  Development  and Comparative Case  Studies.  In:  Proceedings:
      First Joint Symposium on  Dry SO, and  Simultaneous  S0,/N0  Control
      Technologies, Volume 2, EPA-60079-85-020b (NTIS  PB 85-232361), July 1985,
      p. 35-1.

 32.  Ablin, D.W., et al.  Full Scale Demonstration of Dry  Sodium Injection
      Flue Gas Desulfurization  at City of Colorado  Springs  Ray D. Nixon Power
      Plant.  In:  Proceedings:  1986 Joint Symposium on Dry SO,  and
     Simultaneous S0,/N0tf Control Technologies, Volume 2,  EPA-600/9-86-029b
      (NTIS PB87-120457),"October 1986,  p. 48-1.

33.  Babu, M., et al.   Results of 1.0 Million Btu/hour Testing and Plans
      for a 5 MW Pilot HALT Program for SO- Control.   In:   Proceedings of the
     Third Annual Pittsburgh Coal Conference,  Pittsburgh,  Pennsylvania,  1986.

34.  Abrams, J. Z., R.  M. Sherwin, and G. H.  Dyer.  Partial FGD  by Confined
     Zone Dispersion of Pressure-Hydrated Lime.  In:   Proceedings of Coal
     Technology '85, Pittsburgh, Pennsylvania,  1985.   pp.  153-166.

35.  Drummond, C. J.,  et al.  Duct  Injection Technologies  for SO, Control.
      In: Proceedings:  First Combined FGD and Dry SO, Control  Symposium,  Vol. 3,
     EPA-600/9-89-036C (NTIS PB89-172175),  March 1989, p. 8-24.
                                       3-58

-------
36.  Tischer, R. and C. Drummond.  Duct Injection Technologies for SO. Control.
     In: Proceedings of the 15th Annual Biennial Low-Rank Fuels Symposium.
     DOE/METC-90/6109.  St. Paul, Minnesota, May 22-25, 1989.

37.  Hovis, L. S., et al.  E-SO  Pilot Evaluation.  In: Proceedings:  First
     Combined FGD and Dry SO, Control Symposium, Vol. 3, EPA-600/9-89-036c
    .(NTIS PB89-172175), MarCh 1989, p. 8-177.

38.  Jorgensen, C.f et al.  Pilot Plant Evaluation of Post-Comt>ustion LIMB SO-
     Capture.  In: Proceedings: First Combined FGD and Dry S09 .Control
     Symposium, Vol. 2, EPA-600/9-89-036b (NTIS PB89-172167;, (). 5-91.

39,  Shilling, N. Z.  In-Duct Application of Dry Flue Gas Desulfurization:  A
     Cost Effective Retrofit for Reduction of Sulfur Emissions.  In:
     Proceedings of the Second Annual Pittsburgh Coal Conference, Pittsburgh,
     Pennsylvania, 1985.  pp. 158-163.

40.  Yoon, H., P. A. Ring, and F. P. Burke.   Coolside S02 Abatement Technology:
     1 MW Field Tests.  In:  Proceedings of Coal Technology 1985, Pittsburgh,
     Pennsylvania, 1985.  pp. 129-152.

41.  Miyamae, S., et al.  Evaluation of In-Furnace NO  Reduction.  In:
     Proceedings:  1985 Symposium on Stationary Combustion NO  Control, Volume
     1, EPA-600/9-86-021a (NTIS PB 86-225042), July 1986,  p.  24-1.

42.  Murakami, N.  Application of the MACT In-Furnace NO  Removal Process
     Coupled with a Low NO  SGR Burner.  In:  Proceedings:  1985 Symposium on
     Stationary Combustion NO  Control, Volume 1,  EPA-600/9-86-021a
     (NTIS PB 86-225042),  Jul? 1986, p. 32-1.
                                       3-59

-------
                                  SECTION 4

                            EMERGING TECHNOLOGIES
INTRODUCTION

     This section reviews two areas of emerging technology which are currently
being researched and developed, but have not been demonstrated using full-scale
equipment.  Depending on their economics and pollution control performance,
however, it is possible that these technologies could be in commercial  use by
the mid- to late-1990's.  These technologies are:

     •  Pre-Combustion Controls
        - Advanced Coal Cleaning (Section 4.1)

     •  Post-Combustion Controls
        - Advanced FGD (Section 4.2)
                                        4-1

-------
4.1  ADVANCED COAL CLEANING

     Advanced coal cleaning is divided into two primary areas:  1) advanced
physical cleaning and 2} chemical cleaning.  Research is also underway in
biological cleaning processes, but is currently limited to bench-scale
experimentation.  Evaluation of several advanced physical cleaning technolo-
gies 1s being conducted through a joint DOE/EPRI program at the EPRI Coal
Cleaning Test Facility (CCTF) 1n Homer City, Pennsylvania.

Description

Advanced Physical Coal Cleaning--
     Advanced physical coal cleaning processes are being developed to increase
the removal of pyritlc sulfur and mineral matter from fine coal  (<28 mesh);
organic sulfur is not affected.  Although froth flotation and dense media
cyclones have been used commercially on fines between 28 and 100 mesh, these
techniques are currently ineffective at separating materials below 100 mesh
(referred to a "ultrafines").  Additionally, flotation is more effective at
removal of mineral matter than at removal of pyrite (1).
     Research and development in advanced cleaning processes is  concentrated in
four areas:  1)  dense media cyclones, 2)  froth flotation, 3) electrostatic
separation, and 4) selective coalescence (2).  Most of these processes are
being developed to clean coal fines produced during crushing at  a conventional
cleaning plant.   Currently, these fines are generally either discarded or
recombined with the coarser cleaned product without being cleaned.  Briefly,
these processes are:
     t    Dense media cyclones use heavy liquids such as fluorocarbon
          refrigerants to separate coal  from mineral  matter based on
          gravity differences.
     •    Froth flotation involves fine grinding and  multistage  flotation
          to liberate sulfur and mineral  matter from  coal based  on
          differences in hydrophobic properties.
                                        4-2

-------
     •    Electrostatic separation involves feeding dry, electrically charged
          pulverized coal onto the surface of a rotating drum.  Differences in
          the dielectric properties of clean coal versus pyrite and ash cause
          the impurities to separate.
     t    Small  coal particles will selectively coalesce (i.e., agglomerate)
          into larger particles in the presence of an appropriate medium.   The
          coalesced particles can then be separated from undesirable impurities
          that do not coalesce.  Research is underway on both the basic
          physical mechanisms that are involved in this phenomenon and the use
          of novel, non-aqueous media such as liquid carbon dioxide.

Chemical Coal Cleaning--
     Chemical cleaning processes are being developed to remove both pyritic and
organic sulfur from coal.  A variety of methods and chemical  reagents have been
used in chemical  cleaning of coal, including alkali displacement,
oxydesulfurization, and chlorinolysis.  The major research and development
efforts currently underway involve caustic leaching, microwave treatment, and
partial hydrogenation.
     •    TRW's  Gravimelt process is the primary caustic leach process under
          development.  Gravimelt exposes coal  to a mixture of molten
          potassium and sodium hydroxide at temperatures of 615-735°F.
          Sulfur reacts with these alkalis to form sulfides.   Residence time
          is generally 2 to 3 hours.   The cleaned coal  floats to the
          top of the molten caustic and liberated mineral  matter.  The clean
          coal  is then recovered by skimming and washed with  water to remove
          soluble contaminants and residual  caustic.  Additional ash removal
          is achieved using an acid wash.  Sulfur in the spent caustic is
          converted to sulfuric acid, calcium sulfate,  or elemental  sulfur
          in the regeneration section.  The regenerated caustic is
          reconcentrated and recycled (3).
     •    Microwave treatment is based on the tendency of water, caustic,
          sulfur, and other ash components to absorb microwave radiation more
          readily than the other coal constituents.  When exposed to microwave
                                        4-3

-------
          radiation for less than a minute in the presence of potassium
          hydroxide, sodium hydroxide, and water, a portion of the sulfur is
          converted to alkali sulfides.  The product coal is water washed to
          remove-the sulfides and subsequently acid washed to dissolve the
          converted ash.
     •    Sulfur removal can also be achieved by hydrogenation at mild
          liquefaction conditions.  Organic sulfur is converted to hydrogen
          sulfide which can in turn be converted to elemental sulfur or
          sulfuric acid.  After separation of pyrite and insoluble mineral
          matter, the final product is cooled to form a clean solid.

Biological Coal Cleaning--
     Several microorganisms have been shown in laboratory and bench-scale tests
to promote removal of organic sulfur and pyrite occurring on the surface of
coal particles (4,5).  Biological desulfurization is a low energy process, has
low operating costs, and does not reduce the heating value of the coal  product.
Because the microorganisms attack the coal particle surface, crushing enhances
removal rates.  Organic sulfur removals of 25-35 percent have been reported in
bench-scale tests with -60 mesh coals.  Biological  treatment could be
especially useful for coals containing very finely disseminated pyrite and
organic sulfur that is generally not removable by mechanical techniques.
     Potential problems are relatively long bioprocessing times (days to weeks)
and the production of acidic, corrosive leaching solutions.   However,
bioprocessing of stored coal may reduce the time-factor problem, if corrosion
problems can be overcome.   Additionally, thermophilic microbial action
(occurring at >50°C) can accelerate processing with some bacteria.
     Recent evidence shows that microorganisms can  render pyrite hydrophilic in
minutes, thereby making it amenable to separation from coal  by techniques such
as oil  agglomeration.  Development of such a treatment process would be
valuable in that pyrite separation could be accomplished in  minutes compared to
several days required to remove comparable amounts  of sulfur through bacterial
oxidation of pyrite to sulfate.   Still to be determined are  process parameters
and the biological mechanism.
                                        4-4

-------
     Because of the limited data available on the performance and economics of
biological treatment processes, no additional consideration of their potential
is presented in this report.

Applicability

     Advanced coal cleaning processes can be applied to both new and retrofit
installations.  However, since advanced physical coal cleaning cannot remove
organic sulfur, its primary benefit is on coals with a high percentage of
fine-grained pyritic sulfur and ash.  Chemical cleaning can remove both pyritic
and organic sulfur as well as mineral matter and is therefore applicable to a
wider range of coals.

Performance

     Pyrite in most coals generally accounts for 50-70 percent of the total
sulfur.  Assuming advanced physical cleaning can remove 90 percent of the
pyritic sulfur, it would result in a 45-60 percent reduction in total sulfur.
Chemical cleaning can remove over 95 percent of both pyritic and organic
sulfur, and up to 99 percent of the mineral  matter.   Neither approach reduces
the nitrogen content of coal and,  thus, they have no effect on NO  emissions.

Costs

     Advanced physical  and chemical cleaning processes are both more expensive
per ton of product than conventional cleaning methods.  They are therefore of
primary use in cleaning coal fines which cannot be handled by conventional
technologies or when additional sulfur removal is needed to meet more stringent
regulatory limits.
     Depending on the specific process, capital cost estimates for advanced
physical coal  cleaning equipment range from $22,000 to $82,000 per ton/hr of
input capacity; O&M cost estimates range from $2-11  per ton of coal  processed
(6).   The total capital cost of an integrated coal  cleaning plant incorporating
                                        4-5

-------
 an  advanced physical cleaning  process can exceed.$120,000 per ton/hr of input
 capacity.  Typical O&M expenses  for the total plant range from $4-8 per ton of
 raw coal.
     Table 4.1-1-reflects costs  for a 450 ton per hour (input) integrated plant
 in  which the raw coal is crushed to -3/4 Inch (7).  Conventional cleaning is
 used on 28 mesh x 3/4-inch material and advanced processes are used on material
 <28 mesh and on coarse middling  products.  Input to the advanced cleaning
 system equals 40-50 percent of the total plant feed (I.e., 180-225 tons per
 hour).  Total capital cost ranged from $103,000-130,000 per ton/hr input
 capacity.  O&M ranged from $5.20-6.82 per Input ton.  Coal sulfur content was
 assumed to be 3.5 percent.  Btu  and weight recoveries ranged from 83.5-94
 percent and 68-85 percent, respectively.  Sulfur levels (in Ib SO^/mlllion Btu)
 decreased by 34-78 percent.  Because the least capital cost system may have
 higher O&M or fuel loss costs  (and vice versa), the lowest cost for the overall
 system is greater than summation of individual cost components.
     The capital and busbar costs from Table 4.1-1 are significantly higher
 than for conventional physical cleaning (Section 2.3).  The cost effectiveness
 values are fairly similar, however, reflecting the higher level  of sulfur
 removal achievable.  The cost for cleaning a 2 percent sulfur coal  may be
 similar, but due to the lower level of pyritic sulfur, the cost per ton of. SO-
 removed may more than double.
     The cost of chemical coal cleaning technologies is uncertain because of
 the limited research results available (8).   Current dollar cost estimates for.
 the Gravimelt process presented  In Table 4.1-2 range from $54-87 per ton of
 cleaned coal (at 13,400 Btu/lb and 10 percent of the original  sulfur).   This
 estimate is based on a 10,000 ton/day plant costing $270-380 million and
 operating 330 days/year.   As shown in Table 4.1-2,  capital costs account for
 20-25 percent of the annual 1 zed product price, O&M expenses 70-75 percent, and
 fuel losses the remaining 5 percent or less.
     These cost estimates do not include potential  cost savings  resulting from
 improved boiler operations due to reduced solids loadings to the boiler and
emission control  systems.  These savings can significantly improve the
economics of coal cleaning, but are frequently specific to individual  boilers
and coals.
                                        4-6

-------
         TABLE 4.1-1.  ECONOMICS OF ADVANCED PHYSICAL COAL CLEANING
                             (3.5% Sulfur Coal)
$/Ton of Clean Coal
Capital
0 & M
Btu Loss
Total
Busbar Cost (mills/kWh)
$/ton of S0
  Constant
 2.70- 3.80
 6.30- 9.00
 2.10- 7.00
11.10-19.80
 4.9 - 7.15
  220-490
 Current
 5.20- 8.10
10.80-15.70
 3.60-12.30
19.60-36.10
 8.40-12.30
  410-870
              TABLE 4.1-2.  ECONOMICS OF CHEMICAL COAL CLEANING
$/Ton of Clean Coal
Capital
0 & M
Btu Loss
Total
Busbar Cost (mills/kWh)
$/ton of SO-
  Constant
 6.00- 8.40
21.70-37.90
 1.60- 2.20
29.30-48.50
   11 -18
  430-730
 Current
12.50-17.60
37.90-66.60
 2.70- 3.20
53.10-87.40
   19 - 32
  785-1300
                                        4-7

-------
4.2
ADVANCED POST-COMBUSTION S02/NOX PROCESSES
     Work 1s currently underway to develop improved methods for simultaneously
controlling S02 and NO  emissions.  Common elements in most of these processes
1s either reagent regeneration or production of a salable by-
product to reduce solid waste disposal volumes.  This section focuses primarily
on two processes which are in pilot-scale and proof-of-concept testing:
electron-beam irradiation (E-Beam) and copper oxide.  Other advanced processes
for simultaneous S02/NOX control which are less advanced include CONOSOX (based
on a potassium salt reagent) and Flakt Boliden (based on a sodium citrate
reagent).

Description

E-Beam--
     Electron-beam processes Involve the irradiation of flue gas treated with a
reactant such as ammonia or lime to remove both SO- and NO.    Factors
affecting S02 and NO  removal  by electron-beam irradiation include gas moisture
content, gas temperature,  oxygen content, reagent ratio, and electron dosage.
In addition, efficient electron penetration of the gas stream requires a unique
discharge pattern and other special design considerations.
       A simplified flow diagram of the E-beam/ammonia process is shown in
Figure 4.2-1.  Incoming flue gas is cooled and humidified to about
10 percent moisture content in a water quench tower.  The cooled flue gas is
injected with ammonia and  then passed through an electron beam reactor.  Oxygen
and water in the flue gas  are ionized by the electrons to form the radicals
[HO],  [0], and [H02] which react with S02 and NOX to form sulfuric acid (H2S04)
and nitric acid (HNO,).  These acids are neutralized by the injected ammonia to
form solid ammonium sulfate ((NH.J-SO.) and ammonium sulfate nitrate
((NH.^SO '2NH.N03).  Reaction time for formation of the sulfate and nitrate
salts  is less than one second.  Product solids are collected in the hopper
located below the reactor  or in a downstream particulate collector.
                                        4-8

-------
Ammonia
                       Quench Water
          t  t   t  t   t

          t  t   t  t   t
 Flue Gas
E-Gun
                                             1
             E - Beam
              Reactor
                Drain
Participate
 Collector
ID Booster
  Fan
                                                                                        Product Solids
                       Figure 4.2-1.  E-beam / ammonia  process  flow diagram (9).

-------
      In  another version  of  the  E-beam process, the water quench tower and
ammonia  injection  system are replaced with a lime-based spray dryer.  Reactions
are the  same manner  as above except that the products formed are calcium salts
(CaS04,  Ca(N03)2,  and CaSOj) instead of ammonium salts.

Copper Oxide--
     Copper oxide  (CuO)  FGD is  based on the reaction of copper oxide and SO- to
form copper sulfate  (CuSO.).  The CuSO,, and to a lesser extent CuO, catalyzes
the selective reduction  of NO   to N- in the presence of ammonia.  Optimum
reaction temperature is  650-850°F.  Spent CuSO. is sent to a second
vessel for regeneration  by a reducing gas.  The resulting concentrated S02
stream can be economically recovered as salable sulfur or sulfuric acid.
     Two c.opper oxide processes have been developed.   Shell Oil's version of
the process uses a set of specially designed, parallel-passage,  fixed-bed
reactors containing copper oxide bonded to an alumina substrate.  The other
approach, under development at  DOE's Pittsburgh Energy Technology Center, uses
a fluidized bed of copper-impregnated alumina spheres.
     A simplified diagram of the fluidized-bed process is shown in
Figure 4.2-2.  In this configuration, the absorber is located between the plant
economizer and air heater; ammonia is injected upstream of the absorber.  The
sulfated sorbent leaves  the absorber through overflow pipes and weirs,  and is
pneumatically transported with  pre-heated air either  to a solids heater or
directly to a moving-bed regenerator.  The need for a solids heater depends on
the regeneration gas being used.  In the regeneration vessel, the sulfated
sorbent contacts a reducing gas (CH4, H-, or a mixture of CO and H2) flowing
countercurrent to the sorbent.   The regenerated sorbent leaves the bottom of
the reactor and is pneumatically conveyed in heated air to the absorber.
During transport,  the copper formed during regeneration is oxidized to copper
oxide.

Other Advanced Concepts--
     Durlng 1985,  Argonne National  Laboratory conducted laboratory and
bench-scale tests  of two advanced SO^/NO  processes.   In laboratory tests
                                       4-10

-------
                             To Air
                            Preheater
                NH3
 From
Economizer
Flue
                  I
                                 Flue Gas
Gas
                                           Fluidized-Bed
                                           Absorber
                                 Spent
                                Sorbent
                                                 Solids Heater
                                               Grativating-Bed
                                                Regenerator
                                                  CH4
                                                             To
                                                            Sulfur
                                                           Recovery
FIGURE 4.2-2.   Simplified flow diagram  for the Fluidized-Bed  Copper Oxide
                Process (10).
using double-alkali  chemistry with  proprietary additives, Argonne achieved
greater than  70  percent nitrogen oxides  removal  and 90 percent  sulfur oxide
removal.  Other  laboratory work at  Argonne has shown 50-70  percent nitrogen
oxide removal  with greater than 90  percent S02 removal using  lime spray drying
combined with  additives; because this work applies to current FGD processes,  it
could be scaled-up fairly quickly.
                                         4-11

-------
ADD!icabllitv and Performance

E-Beam--
     E-beam processes are in the early stages of development and have not been
tested at full-scale.  Two major pilot projects are being conducted under the
U. S.  Department of Energy Flue Gas Clean-up Program:  one with Research-
Cottrell to evaluate the E-beam/lime spray dryer process, and the other with
Ebara International Corporation to evaluate E-beam/ammonia injection.
     The E-beam/lime spray dryer pilot-scale system was installed on a
slipstream (4,000 ACFM) from a 150 MW coal-fired boiler at the TVA Shawnee
Steam Plant in Paducah, Kentucky.  The pilot work began in early 1983 and was
completed in 1985.  During the test program, five parameters were varied.  The
inlet SO. concentration was observed to have the greatest impact on SO- and NO
        £                                                             t       A
removal.  At a high inlet SO- concentration (2,500 ppm) and an absorbed
electron dose of 1.5 Mrad (a Mrad equals 10 joules/g of flue gas), S02 removal
was greater than 82 percent and NO  removal was greater than 90 percent.  For a
low inlet SO- concentration (400 ppm), SO- removal was.greater than 95 percent
and NO  removal  was about 55 percent.  The increased NO  removal at high SO-
      A                                                AC
concentrations probably results from reaction of NO and NO- with H-SO. to form
HNOS04. (11)
     The E-beam/ammonia Injection pilot-scale test facility is located at
Indiana Power and Light Company's E. W. Stout plant.  Construction of the pilot
plant  began in 1983.  Testing completed in 1985 claims to have demonstrated 90
percent reduction in both SO- and NO .  Detailed test results are not yet
available.

Copper Oxide--
     The copper oxide process has been tested at both bench- and pilot-scale
levels.  Bench-scale tests with a fluidized-bed absorber using flue gas from a
natural gas combustor doped to 3,000 ppmv have shown 90 to 95 percent SO-
removal and over 97 percent NO  removal.   Pilot-scale fluidized-bed tests using
flue gas from a 500 Ib/hour combustor burning a 3 percent sulfur bituminous
coal achieved 90 percent removal  of both SO- and NO  (12).  In 1985,  DOE
                                           £       A
awarded a contract to scale this  process up to the proof-of-concept level
                                       4-12

-------
(i.e., 5 MW equivalent).  This pilot-scale facility will be constructed and
operated at Commonwealth Edison's Kincaid Station (Kincaid, IL).  The Shell
process has also been tested at the bench- and pilot-scale levels and has been
applied on a commercial 40 MW oil-fired boiler in Japan (13).
     The integrated CuO process involves three major reactor systems (the
absorber, the regenerator, and SO- recovery) and a pneumatic solids handling
system between the absorber and regenerator.  The relative complexity and space
requirements of this system plus the need for 650-850°F flue gas will limit the
potential for applying the CuO process to many existing boilers.  One possible
alternative for overcoming space constraints is to install the absorber
downstream of the existing environmental control systems and to reheat the flue
gas to the desired temperature.  However, operation at the lower end of the
desired temperature range (for economic reasons) will  reduce S09/N0  removal
                                                               w   A
rates.  Because the sulfation reaction is exothermic,  part of the reheat energy
may be recoverable from the cleaned flue gas.

Cost

     Because of the relative complexity of these processes, retrofit onto
existing plants will be more difficult than other post-combustion FGD
technologies.   The impact of this difficulty on retrofit costs has not been
quantified, however.  Cost estimates for lime-based electron beam technology
are presented in Table 4.2-1.  The costs shown for S02 control are the same as
for lime spray drying.  The incremental costs for the  system are included under
the costs for NOX control.  The combined costs for the total system are shown
in the right-hand columns.  Cost estimates for control of S02 and NOX using
copper oxide technology are presented in Table 4.2-2.   In both cases, retrofit
factors for lime/limestone wet FGD were assumed.
                                       4-13

-------
                                             TABLE 4.2-1.  COST ESTIMATES  FOR  ELECTRON  BEAM
Technology
xso2
Removed 3.5X S
X NO
2X s Removed
S02 &
NO 3.5X S
X
NOx Combined
zx s
New 500-HW Baseload Power Plant:





  Capital Cost, S/ku         70 -  90





  MIlts/kUH,  Current         70 -  90





  S/Ton, Current S           70-90







Retrofit 500-HU Baseload Power Plant:





  Capital Cost. S/kU         70 -  90





  Mills/kWh,  Current *       70-90





  S/Ton. Current $           70-90







Retrofit 250-NU Intermediate Load Power Plant:





  Capital Cost, SAW         70 -  90        339 - 403





  Mills/kwh.  Current S       70-90       30.2 - 35.3





  S/Ton. Current »           70-90      1.219 -1.609
 174 - 207        153 - 182





15.7 - 18.6      12.0 • 14.0





 677 - 804       880 - 1,081
                  297 - 352





                 25.2 - 29.2





                1.753 - 2,365
90





90





90
226 - 270
17.1 - 20.4
730 - 883
198 - 235
13.2 - 15.4
961 - 1.202
90
90
90
90





90





90
  93 - 109        284 -  315        248 -  275





  6.3 - 6.7      22.4 -  25.2     18.3 -  20.1





2.632 - 3.732    930 -  1,135    1,298 •  1,629














  122 - 143       369 -  408        323 -  357





  7.0 - 7.6      24.7 -  27.8     20.3 •  22.4





2.944 - 4,182   1,017 -1,260    1,431 -  1,819














  183 - 213       552 -  612        483 -  535





 12.9 - 14.3     44.5 -  49.4     38.1 -  41.8





5,396 • 7,854   1,760 -  2.329   2,608 -  3,500
   Retrofit factor of 1.0 and capacity factory of 65 percent.
   Retrofit factor of 1.3 and capacity factor of 65 percent.

-------
                                                     TABLE 4.2-2.  COST ESTIMATES FOR COPPER OXIDE
-e*

1-^
en
Technology Removed 3.5X S
X NO
2X S Removed NO
X
S02 &
3.5X S
NOx Combined
2% S
Neu 500-MU Baseloed Power Plant:



  Capital Cost. S/kU         70 - 90



  Mills/kUH, Current         70 - 90



  S/Ton. Current $           70-90





Retrofit 500-MU Baseload Power Plant:'



  Capital Cost, J/kU         70 - 90



  Hills/kUh. Current S       70-90



  S/Ton, Current $           70-90
                                                   187 - 203         155  -  167



                                                  17.7 - 20.5       13.0  -  14.7



                                                   760 - 883        949  •  1,143
                                                   235  - 255         192  - 207



                                                   18.8  - 21.9       14.0  - 15.8



                                                   813  - 951       1,028  - 1.236
        Retrofit  250-MU Intermediate  Load  Power Plant:



          Capital  Cost,  $/kW         70  -  90         360  - 391         290  -  315



          Mills/kUh.  Current $       70-90        31.3  - 35.0       24.6  -  27.1



          S/Ton.  Current $           70-90       1.269  -1.613      1.715  -  2,216
90



90



90
90



90



90
                                                                                90



                                                                                90



                                                                                90
   (10)



   (12)



760 - 1,143
   (10)



   (12)



813 - 1.236
 187 - 203       155 -  167



17.7 - 20.5     13.0 -  14.7



 760 - 883      949 -  1.143
 235 - 255       192 -  207



18.8 - 21.9     14.0 -  15.8



 813 - 951     1.028 -  1,236
               (10)          360 - 391       290 -  315



               (12)         31.3 - 35.0     24.6 -  27.1



           1.269 - 2.216   1.269 - 1.613   1.715 -  2.216
         Retrofit factor of 1.0 and capacity  factory  of 65  percent.
         Retrofit factor of  1.3  and capacity  factor  of  65  percent.

-------
 References

 1.    Cavallaro,  J.  A.,  R.  P.  Killmeyer,  A.  W.  Deurbrouck,  and  K.  Rhee.   An
      Overview of PETC's Chemical,  Physical,  and  Surface-Enhanced  Beneficiation
      Program.   Pittsburgh  Energy  Technology Center,  U.  S.  Department  of Energy,
      1986,  29  pp.

 2.    Engineering and  Economics  Research,  Inc., et  al.   Supplemental  Report  to
      Congress  on Emerging  Clean Coal  Technologies.   DOE/MC/22121-1.   U.  S.
      Department  of  Energy,  Washington, D.C.,  1985.

 3.    Reference 2.

 4.    Olson, G. J.,  F. B. Brickman,  and W. P.  Iverson.   Processing of  Coal with
      Microorganisms.  AP-4472,  Electric  Power  Research  Institute,  Palo  Alto,
      California,  1986,  48  pp.

 5.    Isbister, J. D., et al.  Companion  Processes  for Removal  of  Sulfur  and Ash
      from Coal.   In:  Proceedings  of  the Second Annual  Pittsburgh  Coal
      Conference,  Pittsburgh,  Pennsylvania,  1985, 809 pp.

 6.    Santhanam,  C.  J. and  V.  Vejins.  Impact of Advanced Coal  Benefication on
      Utilization of Coal Slurry Fuels.   In:  Proceedings of the Seventh
      International  Symposium  on Coal  Slurry  Fuels  Preparation  and  Utilization,
      New Orleans, Louisiana,  1985.

 7.    Boron, D. J.,  J. P. Hatoney,  and M. C. Albrecht.   Advanced Physical Coal
      Cleaning:  An  Evaluation of Five Select Processes.  In:   Proceedings of
      the Third Annual Pittsburgh Coal Conference,  Pittsburgh,  Pennsylvania,
      1986, 974 pp.

8.   Meyers, R. A., L. C. McClanathan, and W. D. Hart.  Development of the TRW
     Gravimelt Process.  In:  Proceedings of the Second Annual Pittsburgh Coal
     Conference, Pittsburgh,  Pennsylvania,  1985, 890 pp.

9.   Aul,  E.,  S. Margerum,  and  R.  Berry.  Industrial Boiler SO. Technology
     Update Report.  EPA-450/3-85-009 (NTIS PB85-197093), U.S.^Environmental
     Protection Agency, Research Triangle Park, North Carolina, 1984.

 10.  Drummond, C. J., J. T. Yen, J. I. Joubert, and J. A. Ratafia-Brown.  The
     Design of a Dry Regenerative  Fluidized-Bed Copper Oxide Process for the
     Removal of Sulfur Dioxide  and Nitrogen Oxides from Coal-Fired Boilers.
     Presented at the 78th Annual  Meeting and Exhibition of the Air Pollution
     Control Association,  1985,  38 pp.

11.  Gleason, R.  J. and D. J. Helfritch.  High-Efficiency NO   and  SO  Removal
     by Electron Beam. Chemical  Engineering Progress, 81(10)?33-38, i985.

12.  Reference 10.

13.  Dickerman, J. C.  and K. L.  Johnson.   Technology Assessment Report for
     Industrial Boiler Applications: Flue Gas Desulfurization
     EPA-600/7-79-1781 (NTIS PB  80-150873),   U.S.  Environmental Protection
     Agency, Research Triangle Park, North Carolina,  November 1979.
                                       4-16

-------
        APPENDIX A
 Summary of Control Costs
Coal Switching and Blending
             A-l

-------
COAL SWITCHING AI0 ILCNOINQ
CAPITAL  COT (1/kU)
ESP FUEL Mice SULHA CAPACITY
SIZE DIFFERENTIAL (X) FACTOR
(«) («
LARGE ESP
(400 SCA)


LAIGB ESP
(400 SCA)


LARGE ESP
(400 SCA)


SMALL ESP
(200 SCA)


SMALL ESP
(200 SCA)


SMALL ESP
(200 SCA)


1.0
3 2.0 30.0 */W
3.0
4.0
1.0
5 2.0 SO.O */W
3.0
4.0
1.0
5 2.0 70.0 S/kW
3.0
4.0
1.0
S 2.0 30.0 */W
3.0
4.0
1.0
5 2.0 SO.O */kW
3.0
4.0
1.0
I 2.0 70.0 S/tt
3.0
4.0
BOILER GENERATING CAPACITY (NW)
100 300 500 700 1000
30.7
30.7
30.7
30.7
30.7
30.7
30.7
30.7
30.7
30.7
30.7
30.7
31.8
31.8
31.8
31. 8
31.8
31.8
31.8
31.8
31.6
31.8
31.8
31.4.
23.2
23. 2
23.2
23.2
23.2
23.2
23.2
23.2
23.2
23.2
23.2
23.2
24.3
24.3
24.3
24.3
24.3
24.3
24.3
24.3
24.3
24.3
24.3
24.3
21.3
21.3
21.3
21.3
21.3
21.3
21.3
21.3
21.3
21.3
21.3
21.3
22.4
22.4
ZZ.4
22.4
22.4
22.4
22.4
22.4
22.4
22.4
22.4
22.4
20.4
20.4
20.4
20.4
20.
20.
20.
20.
20.
20.
20.
20.
21.5
21. S
21.5
21.5
21.3
21.5
21.5
21. S
21.5
21.5
21.5
21.5
19.6
19.6
19.6
19.6
19.6
19.6
19.6
19.6
19.6
19.6
19.6
19.6
20.7
20.7
20.7
20.7
20.7
20.7
20.7
20.7
20.7
20.7
20.7
20.7
1300
19.1
19.1
19.1
19.1
19.1
19.1
19.1
19.1
19.1
19.1
19.1
19.1
20.2
20.2
20.2
20.2
20.2
20.2
20.2
20.2
43.1
4J.1
43.1
43.1
                                                       A-2

-------
COAL SWITCHING  AND BLENDING
CAPITAL COST  (S/kW)
ESP FUEL PRICE SULFUR CAPACITY
SIZE DIFFERENTIAL (%) FACTOR
(») . (X)
LARGE ESP
(400 SCA)


LARGE ESP
(400 SCA)


LARGE ESP
(400 SCA)

ouBBBonB BBBBSVI
SMALL ESP
(ZOO SCA)


SMALL ESP
(200 SCA)


SMALL ESP
(200 SCA)


1.0
15 2.0 50.0 S/ktt
3.0
4.0
1.0
15 2.0 50.0 S/kU
3.0
4.0
1.0
15 2.0 70.0 1/kW
3.0
4.0
raBBBBBBBBOBBBBiiBB*BBMBBnBABBBBBBBBBSttBWBBi
1.0
IS 2.0 30.0 t/W
3.0
4.0
1.0
IS 2.0 50.0 S/kW
3.0
4.0
1.0
15 2.0 70.0 S/kU
3.0
4.0
BOILER GENERATING CAPACITY (HW)
100 300 500 700 1000
41.1
41.1
41.1
41.1
41.1
41. 1
41.1
41.1
41.1
41.1
41.1
41.1
nmssvnsBi
42.2
42.2
42.2
42.2
42.2
42.2
42.2
42.2
42.2
42.2
42.2
42.2
33.6
33.6
33.6
33.6
33.3
33.5
33.3
33.5
33.3
33.3
33.5
33.3
•Bscnuu
34.7
34.7
34.7
34.7
34.7
34.7
34.7
34.7
34.7
34.7
34.7
34.7
31.6
31.6
31.6
31.6
31.6
31.6
31.6
31.6
31.6
31.6
31.6
31.6
32.8
32.8
32.8
32.8
32.7
32.7
32.7
32.7
32.7
32.7
32.7
32.7
30.7
30.7
30.7
30.7
30.7
30.7
30.7
30.7
30.7
30.7
30.7
30.7
31.8
31.8
31.8
31.8
31.8
31.8
31.8
31.8
31.8
31.8
31.8
31.8
29.9
29.9
29.9
29.9
29.9
29.9
29.9
29.9
29.9
29.9
29.9
29.9
tBBBBBQBVBB
31.1
31.1
31.1
31.1
31.1
31.1
31.1
31.1
31.1
31.1
31.1
31.1
1300
29.5
29.5
29.5
29.5
29.5
29.5
29.5
29.5
29.5
29.5
29.5
29.5
aaanas
30.6
30.6
30.6
30.6
30.6
30.6
30.6
30.6
30.6
30.6
30.6
30.6
                                                       A-3

-------
COAL SWITCHING AND BLENDING
ANNUAL COST  (•Illl/Uh)
ESP FUEL PtlCB SUirm CAPACITY
SIZE DIFFERENTIAL (X) FACTM
(«> (W
LABCf f


-------
COAL SWITCHING AMD  BLEND IKO
ANNUAL COST  (•Ull/kWh)
ESP FUEL PRICE SULFUR CAPACITY BOILER GENERATING CAPACITY (MU)
SIZE DIFFERENTIAL (X) FACTOR
(<) (X) 100 300 500 700 1000
LARGE ESP
(400 SCA)


LARGE ESP
(400 SCA)


LARGE ESP
(400 SCA)


SMALL ESP
(200 SCA)


SMALL ESP
(200 SCA)


SMALL ESP
(200 SCA)


1.0
15 2.0
3.0
4.0
1.0
15 2.0
3.0
4.0
1.0
15 2.0
3.0
4.0
1.0
15 2.0
3.0
4.0
1.0
15 2.0
3.0
4.0
1.0
15 2.0
3.0
4.0
17.1
30.0 •UK/Ml 17.1
17.1
17.1
15.7
50.0 •UK/kUh 15.7
15.7
15.7
15.1
70.0 •UU/kUh 15.1
15.1
15.1
17.2
30.0 •
-------
COU. SWITCHING AMD BLENDING
$02 COST EFFECTIVENESS (t/TOH)
ESP FUEL PtICf SULFM CAPACITY
Size DIFFERENTIAL (I) FACTOR
(S) (X)
URGE ESP
(400 $CA)


UWGt ESP
(400 SOU


LARSE ESP.
(400 SCA)


SMALL ESP
(200 SCA)


SMALL ESP
(200 SCA)


SMALL ESP
(200 SCA)


1.0
5 2.0 30.0
3.0
4.0
1.0
S 2.0 SO.O
3.0
4.0
1.0
J 2.0 70. 0
3.0
4.0
1.0
S 2.0 30.0
3.0
4.0
1.0
3 2.0 50.0
3.0
4.0
1.0
I 2.0 70.0
3.0
4.0
BOILER GENERATING CAPACITY (NW)
100 300 500 TOO 1000
9650.8
S/TOH 916.8
481.2
326.3
8246.4
t/TOH 78S.3
412.2
279.1
7674
•/TON 729
182.7
259.4
970.8
I/TOI 929.4
487.9
330.8
8349.9
S/TOM 793.2
416.4
282.3
7736.4
I/TW 734.9
383.8
261. S
8147.4
774
406.3
273.4
7159.9
680.2
357
242
6736.9
640
333.9
227.7
8211.9
786.7
413
280
7244.3
668.2
361.2
244.9
6799.8
643.9
339.1
229.9
7797.3
740.7
388.8
263.6
6910
656.4
3U.6
233.6
6529.8
620.3
323.6
220.7
7931.9
753.3
395.5
268.1
6994.2
664.4
348.8
236.4
6592.5
626.3
328.7
222.9
7637.1
725. S
380.8
258.2
6797.9
645.8
339
229.8
6438.3
611.6
321.1
217.7
7771.7
738.3
387.5
262.7
6882.2
653.8
343.2
232.7
6501.1
617.6
324.2
219.8
7105.2
713
374.3
253.7
6706.4
637.1
334.4
226.7
6364
604.6
317.3
215.1
7640
725.8
381
258.3
6790.8
645.1
338.6
229.6
6426.8
610.3
320.3
217.3
1300
7428.7
703.7
370.4
251.1
66S3.3
632
331.8
224.9
6321.1
600.3
315.2
213.7
7543.3
718.5
377.2
255.7
6737.7
640
336
227.8
63B3.9
606.4
318.3
215. B
                                                        A-6

-------
COAL SWITCHING AND  BLENDING
S02 COST EFFECTIVENESS  (S/TON)
ESP FUE
SIZE OIF
LARGE ESP
(400 SCA)


LARGE ESP
(400 SCA)


LARGE ESP
(400 SCA)


SMALL ESP
(200 SCA)


SMALL ESP
(200 SCA)


SMALL ESP
(200 SCA)


L PRICE SULFUR
FERENT1AL (X)
<«).
1.
IS 2.
3.
4.
1.
15 2.
3.
4.
1.
15 2.
3.
4.
1.
15 2.
3.
4.
1.
15 2.
3.
4.
1.
15 2.
3.
4.
C
,0
,0
0
,0
0
,0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
APACITY
FACTOR
(X) 100
20220.6
30.0 «/TON 1920.9
1008.3
683.6
18510.4
50.0 S/TOV 1758.4
923
625.8
17777.6
70.0 S/TOM 1688.8
886.5
601
20354.1
30.0 S/TON 1933.5
1015
688.1
18593.9
50.0 S/TON 1766.3
927.2
628.6
17840
70.0 S/TON 1694.7
889.6
603.1
BOILER
300
18717.
1778.
933.
632.
17403.
1653.
667.
588.
16840.
1599.
839.
569.
18851.
1790.
940.
637.
17487.
1661.
87
591.
16903.
1605.
842.
571.
GE
4
1
4
8
3
2
a
3
3
7
8
3
9
8
1
3
6
2
2
2
2
7
9
4
NERATING
500
18367.4
1744.8
915.9
620.9
17153.5
1629.5
855.4
579.9
16633.3
1580.1
829.4
562.3
18501.9
1757.6
922.6
625.5
17237.7
1637.5
859.6
582.7
16696
1586
832.6
564.4
CAP AC M
700
18207.
1729.
907.
615.
17041.
1618.
849.
576.
16541.
1571.
824.
559.
18341.
1742.
914.
620.
17125.
1626.
as
578.
16604.
1577.
82
561.
rr
2
6
9
5
3
a
a
,1
a
4
9
2
8
4
6
1
6
8
*'
9
6
3
8
3
(MO
NX
18075.
1717.
901.
611.
16949.
1610.
845.
57
16467.
1564.
821.
556.
18210.
1729.
908.
615.
17034.
1618.
849.
575.
16530.
1570.
824.
558.
W
4
1
3
1
8
,1
2
"3
5
3
2
7
1
9
1
6
2
2
4
9
3
3
3
8
1300
17998.9
1709.8
897.5
608.5
16896. S
1605.1
842.6
571.2
16424.5
1560.2
819
555.2
18133.5
1722.6
904.2
613
16981.1
1613.1
846.8
574.1
16487.4
1566.2
822.2
557.4
                                                        A-7

-------
              APPENDIX B
Low NOX Combustion Control  Technologies
                   B-l

-------
NO* COMBUSTION CONTROL  TECHNOLOGIES:
OVHRFIRI All,  LOU Kb BLBNEI8
CAPITAL COST (S/kU>
FIRIMG
CONFIGURATION

TANGENTIAL
FIIING
(OvnriRE All)
TANCEITIAL
FIRING
(OVEIFIRC AIR)
TANGENTIAL
FIRING
(OVUFIIE AIR)
tMLL FIRING
(LOW NOB
BURNERS)
WALL FIIING
(Lay NOB
BURNERS)
UALL FIIING
(LOU NOx
HJBNIRS)
CAPACITY NOX
FACTOR REDUCTION
(I) «>
10
30 20 S/kW
50
10
50 20 »/kU
30
to
70 20 S/kv
30
20
30 40 S/kU
»
20
SO 40 S/kU
55
20
70 40 l/ky
51
BOILER GENERATING CAPACITY

100
.2
.2
.2
.
.
•
6.
6.
6.2
25.5
. 25.3
25.5
25.5
25.5
25.3
25.3
25.3
25.3

300
.2
.2
.2
.2
.2
.2
.2
.2
.2
13.2
13.2
13.2
13.2
13.2
13.2
13.2
13.2
13.2

500 700
2.4 .9
2.4 .9
2.4 .9
2.4 .9
2.4 .9
2.4 .9
2.4 .9
2.4 .9
2.4 .9
9.7 T.9
9.7 7.9
9.7 7.9
9.7 7.9
9.7 7.9
9.7 7.9
9.7 7.9
9.7 7.9
9.7 7.9
(NU)

1000
1.
1.
1. ~
1.
1.
1.
1.6
1.6
1.6
6.4
6.4
6.4
A.4
6.4
6.4
6.4
6.4
6.4


1300
1.
1.
1.
1.
1.
1.
1.3
1.3
1.3
5.5
3.3
5.5
S.S
5.5
5.5
5.5
5.5
5.5
                                                         B-2

-------
MOx COMBUSTION CONTROL  TECHNOLOGIES:
OVERFIRE AIR,  LOU NO* MINERS
ANNUAL COST (•itls/ktfl)
FIflINC CAPACITY NOx
CONFIGURATION FACTOR REDUCTION
(%) (X)
TANGENTIAL
FIRING 30
(OVERFIRE AIR)
TANGENTIAL
FIRING SO
(OVERFIRE AIR)
TANGENTIAL
FIRING 70
(OVERFIRE AIR)
VALL FIRING
(LOU NOx 30
BURNERS)
UALL FIRING
(LOU NOx SO
BURNERS)
UALL FIRING
(LOU NOx 70
BURNERS)
(X)
10
20
30
10
20
30
10
20
30
20
40
55
20
40
ss
20
40
SS
BOILER GENERATING CAPACITY (MU)
100
O.S
•UU/kUh O.S
O.S
0.3
•ilUAUh 0.3
0.3
0.2
nillt/kUh 0.2
0.2
2.1
Bllls/kUh 2.1
2.1
1-2
nllU/kWh 1.2
1.2
0.9
•ills/Ml 0.9
0.9
300
0.3
0.3
0.3
0.2
0.2
0.2
0.1
0.1
0.1
1.1
1.1
1.1
0.6
0.6
0.6
0.5
O.S
O.S
500
0.2
0.2
0.2
0.1
0.1
0.1
0.1
0.1
0.1
0.8
0.8
0.8
O.S
O.S
O.S
0.3
0.3
0.3
700
0.2
• 0.2
0.2
0.1
0.1
0-1
0.1
0.1
0.1
0.6
0.6
0.6
0.4
0.4
0.4
0.3
0.3
0.3
1000 1300
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.1
0.1
0.1
0.1
0.1
0.1
0
0
0
O.S 0.4
O.S 0.4
0.5 0.4
0.3 0.
0.3 0.
0.3 0.
0.2 0.
0.2 0.
0.2 0.2
                                                         B-3

-------
NOX COMBUSTION COMTROl  TfCNWLOOIES:
OVERFIRC All,  LOU NO* UJUIRS
NO* COST EFFECTIVENESS  (I/TON)
FIRING CAPACITY
CONFIGURATION FACTOR
(X)
TANGENTIAL
FIRING 30
(OVWItt AIR)
TANGENTIAL
MB INC SO
(OVERFIRE AIR)
TANGENTIAL
FIRING 70
{OVERFIRE AIR)
WALL FIRING
(LOU NO* 30
BURNERS)
MIL units
(LOU MX SO
BURNERS)
MALL FIRING
(LOU NO* 70
BUSKERS)
NOX
REDUCTION
(S)
10
20 I/TON
30
10
20 S/TON
30
10
20 I/TOM
30
20
40 I/TOM
IS
20
40 S/TON
55
20
40 S/TON
55
BOILER GENERATING CAPACITY (MU>

100
1621.2
810.6
540.4
972.7
466.4
324.2
694.8
347.4
231.6
2368.8
1194.4
868.6
1433.3
716.6
521.2
1023.8
511.9
372.3

300
838.8
419.4
279.6
503.3
251.6
167.6
359.5
179.7
119.8
1235.5
617.8
449. J
741.3
370.7
269.6
529.5
264. 8
192.6

500
617.3
308.7
205.6
370.4
185.2
123.5
264.6
132.3
68.2
909.4
454.7
330.7
545.7
272.6
196.4
389.8
194.9
141.7

700
504.7
252.4
168.2
302.8
151.4
100.9
216.3
108.2
72.1
743.1
371.6
270.2
445.9
222.9
162.1
318.3
159.2
115.6

1000
407.6
203.6
135.9
244.6
122.3
61 .5
174.7
87.4
56.2
600.1
300
218.2
360
180
130.9
257.2
126.6
93.5

1300
348
174
116
208.8
104.4
69.6
149.1
74.6
49.7
512.6
256.3
186.4
307.6
153.8
111.6
129.7
109.9
79.9
                                                          B-4

-------
    APPENDIX C





Lime/Limestone FGD
         C-l

-------
1/S-fCO
CAPITAL COST  (»/kH)
  SULFUI    CAPACITY   UTIOFIT
  CONTENT   . MCTOI     FACTOt
    (X)        «>     '  (X)
      BQIUI GmfUTItt CAMCITY (NO

100      300      100      TOO       1000
                                             1300
1.0 10
1.0 50
1.0 70
2.0 50
2.0 SO
2.0 70
3.0 30
3.C SO
3.0 70
4.0 30
4.0 SO
4.0 70
1.0
1.S
2.0
1.0
1.S
2.0
. 1.0
1.S
2.0
1.0
1.S
2.0
1.0
1.5
2.0
1.0
1.S
2.0
1.0
1.5
2.0
1.0
1.5
2.0
1.0
1.5
2.0
1.0
1.5
2.0
1.0
1.5
2.0
t.O
1.5
2.0
»/kU
t/ku
S/KU
»/kW
»/ku
s/ku
s/ku
»/Ui
t/kM
t/ku
i/fcy
•/ku
296.6
426.3
356
2M.a
426.5
556.2
297
426.7
556.4
316.4
455.2
594
316.6
453.4
594.2
316.7
435.5
SM.4
328.4
472.5
616.6
328.6
472.7
616.8
328.
472.
616.
334.
481.
628.1
335
481.7
628.3
339.2
481.9
428.5
190.5
271
351.5
190.6
271.1
351.6
190.7
271.2
351.7
196.8
279.8
362.7
196.9
279.9
362.8
197
280
362.9
204
289.9
375.8
204.1
290
376
204.2
290.1
376.1
209.8
298
386.2
209.9
298.1
386.3
210
298.2
386.4
158.2
223.6
289.1
158.3
223.7
289.2
158.4
223.8
289.2
164.8
232.8
300.9
164.9
232.9
301
164.9
233
301.1
175.2
247.8
320.3
175.1
247.9
320.6
175.4
248
320.7
180.9
253.6
330.3
181
255.7
330.4
181.1
255.8
330.5
140.3
198
255.4
140.6
198.1
255.5
140.7
198.1
255.6
146.3
206
265.7
146.4
206.1
265.8
146.4
206.1
265.8
155.1
218.4
281.7
155.1
218.5
281.8
155.2
218.5
281.9
160.1
223.4
290.6
160.2
225.4
290.7
160.3
225.5
290.7
136.7
192.2
247.7
136.7
192.3
247.8
136.8
192.3
247.8
144.5
203.3
262.1
144.6
203.4
262.2
144.7
203.5
262.3
150.1
211
271.9
150.2
211.1
272
150 J
211.2
272
154.1
217.5
280.1
154.9
217.6
280.2
155
217.6
280.1
125
175.3
225.5
125.1
175.3
225.6
125.1
175.4
225.6
132
185.2
238.3
132.1
185.2
238.4
132.2
185.3
238.4
137.3
192.7
247.9
137.6
192.8
248
137.7
192.9
248.1
143
200.2
257.5
143.1
200.3
257.5
143.2
200.4
257.6
                                                          C-2

-------
L/S-FGD
ANNUAL COST
  SULFUR    CAPACITT   RETROFIT
  CONTENT    FACTOR   -  FACTOR
    (X)        (X)
      BOILER GENERATING CAPACITY (NU)

100      300     500      700       1000
1300

1.0 30


1.0 50


1.0 70


2.0 30


2.0 50


2.0 70


3.0 30


3.0 50


3.0 70


4.0 30


4.0 50


4.0 70

1.0
1.5
2.0
1.0
1.5
2.0
1.0
1.5
2.0
1.0
1.5
2.0
1.0
1.5
2.0
1.0
1.5
2.0
1.0
1.5
2.0
1.0
1.5
2.0
1.0
1.5
2.0
1.0
1.5
2.0
1.0
1.5
2.0
1.0
1.5
2.0
45.6
•Ul«/Wh 61.4
77.1
30
•i lit/Ml 39.4
48.8
23.2
•Ult/kUh 29.9
36.7
49
•Ult/kUh 65.8
82.7
32.4
•Ult/kUh 42.3
52.6
25.2
•Ult/kUh 32.4
39.6
51.3
•Ult/kUh 68.8
86.3
34.2
•Ult/kUh 44.7
55.2
26.8
•Ult/kUh 34.3
41.8
53
•Ult/kUh 70.8
88.6
33.6
•Ult/kUh 46.3
57
28.1
•Ult/kUh 35.7
43.3
29
38.7
48.5
19.5
25.4
31.2
15.4
19.6
23.8
30.6
40.7
50.8
20.9
26.9
33
16.7
21
25.3
32.
42.
53.
22.
28.
34.
18
22.5
26.9
34
44.7
55.4
23.7
30.1
36.5
19.2
23.8
28.4
24.3
32.2
40.1
16.6
21.3
26.1
13.2
16.7
20.1
26
34.2
42.5
18
22.9
27.9
14.5
18.1
21.6
28.1
37
45.8
19.7
25
30.3
16
19.8
23.6
29.7
38.8
47.9
21
26.3
31.9
17.3
21.2
25.1
21.7
28.7
35.7
IS
19.2
23.4
12.1
15.1
18.1
23.3
30.6
37.8
^16.3
20.7
25
13.3
16.4
19.3
25.3
S3
40.7
17.9
22.5
27.1
14.7
18
21.3
26.8
34.8
42.7
19.2
24
28.7
16
19.3
22.7
20.9
27.7
34.4
14.3
18.5
22.6
11.7
14.6
17.5
22.8
30
37.1
16
20.3
24.6
13.1
16.1
19.2
24.4
31.8
39.2
17.3
21.8
26.2
14.3
17.3
20.6
25.9
33.5
41.1
18.6
23.2
27.7
15.5
18.7
22
19.4
25.
31.
13.
17.
20.
11
13.6
16.2
21.1
27.6
34
13
18.8
22.7
12.3
15.1
17.8
22.7
29.4
36.1
16.3
20.3
24.3
13.5
16.4
19.3
24.3
31.2
38.2
17.6
21.8
25.9
14.8
17.7
20.7
                                                           C-3

-------
l/S-MD
SOZ COST EFFECTIVIMSS (»/TOH)
  SUfl*    CAPACITY   HTKVIT
  CONTENT    FACTOi     FACTOB
    U)        (Z)
•QUO C£«UTKC CAPACITY (W)
                                                100
                                                         300
                                                                 300
                                                                          700
                                                                                    1000
                                                                                            1100
1.0 30
1.0 50
1.0 70
2.0 30
2.0 SO
2.0 70
1.0 30
3.0 SO
3.0 70
4.0 30
4.0 50
4.0 70
1.0
1.3
2.0
1.0
1.5
2.0
1.0
1.5
2.0
1.0
1.5
2.0
1.0
1.5
2.0
1.0
1.5
2.0
1.0
1.5
2.0
1.0
1.5
2.0
1.0
1.5
2.0
1.0
1.5
2.0
1.0
1.5
2.0
1.0
1.5
2.0
l/TOi
S/TON
S/TON
S/TON
S/TON
S/TON
S/TON
S/TON
S/TON
S/TON
S/TON
S/TON
6282.2
8450.5
10618.6
4125.1
5426
6726.9
3189.5
4118.7
5047.9
3372.6
4333
5693. 3
2230.1
2926.4
3622.7
1733
2232.3
2729.7
2357.1
3160.2
3963.3
1570
2051.9
2533.7
1229.1
1373.3
1917.3
1825.9
2438.8
3051.8
1229.8
1593.6
1961.3
965.9
1228.6
1491.3
3988.9
5334.1
6679.3
2984.6
1491.7
4298.8
2122.3
2698.8
1275.3
2109.2
2802.7
1496.2
1438
1854.1
2270.2
1148.6
1445.9
1743.1
1487.7
1966.6
2443.3
1025.5
1312.8
1600.1
826.3
1031.5
1236.7
1171.3
1539.9
1908.6
815.8
1037
1258.2
662.6
820.6
978.6
3340
4434
5328.1
2280.2
2936.7
3593.1
1824.5
2293.4
2762.3
1787.2
2356.4
2925.6
1237.3
1578.8
1920.3
1000.8
1244.8
1488.7
1292
1696.9
2101.9
903.4
1146.4
1389.4
736.4
910
1083.3
1023.6
1333.9
1648.2
723.7
911.1
1098.5
594.8
728.7
862.5
2993.2
3953.7
4914.2
2064.2
2640.5
3216.9
1665.4
2077.1
2488.7
1607
2106
2605.1
1125.2
1424.6
1724
918.3
1132.2
1346.1
1162
1315
1867.9
822.7
1034.5
1246.2
677
828.3
979.6
923.8
1196.5
1469.5
661.8
829.4
989
549.3
666.2
783.1
2885.3
3813.3
4741.6
1993.9
2550.8
3107.7
1611.9
2009.7
2407.5
1571.4
2063
2554.6
1101.3
1396.3
1691.2
099.6
1110.3
1321.2
1120
1459.4
1798.8
795.5
999.2
1202.8
656.5
801.9
947.4
890.9
1152.7
1414.3
640.6
797.7
954.7
533.3
645.5
757.7
2668.1
3308.2
4348.4
1859.6
2363.7
2867.8
1513.6
1873.6
2233.7
1435.6
1899.7
2343.9
1029.8
1296.3
1562.8
847.5
1037.9
1228.2
1042.5
1350.1
1657.7
747.6
932.2
1116.8
621.4
753.3
885.1
835.7
1074.9
1314
606.5
749.9
893.4
508.3
610.8
713.3
                                                         C-4

-------
              APPENDIX D





Integrated Gasification Combined Cycle
                   D-l

-------
ICCC-t/kU
  SULFUR     CAPACITY      HEAT                               BOILER GENERATING CAPACITY (HU)
  CONTENT     FACTOR       RATE
   (X)          (X)       (Btu/kUh)                   100      300      500      700       1000     1300
      1.0           50       8000        */kW       2,373    1,849    1,652    1,536    1,423    1,345
                           10000                   2,816    2,199    1,967    1,829    1,695    1,603
     1.0           70       8000        SAW       2,372    1,848    1,652    1,535     1,422    1.344
                           10000                   2,815    2,198    1,966    1,828     1,694    1.602
     2.0           50       8000        $/kW       2,409    1,872    1,672    1,553    1,437    1.358
                           10000                   2,857    2,226    1,989    1,848    1,712    1,618
     2.0           70       8000        S/kU       2,408    1,871     1,671     1,552     1,436     1.358
                           10000                   2,856    2,225     1,988     1,847     1,711     1,617
     3.0           50       8000        S/kU       2,438    1,891     1,687    1,566    1,449     1,369
                           10000                   2,890    2,248     2,007    1,864    1,725     1,631
     3.0           70       8000        S/kW       2,437    1,890    1,686    1,566     1,448     1,368
                           10000                   2,869    2,247    2,006    1,863     1,724     1,630
     4.0           50       8000        S/kV       2,463    1,908    1,701     1,578     1.459    1,379
                           10000                   2,919    2,267    2,022     1,878     1.737    1,642
     4.0           70       8000        S/kV       2462     1,907    1,700     1,577    1.459    1.378
                           10000                   2,918    2,266    2,021     1,877    1.736    1,641
                                                           0-2

-------
IGCC-mills/kUh
SULFUR
CONTENT
(%)
1.0
1.0
2.0
2.0
3.0
3.0
4.0
(.0
CAPACITY HEAT
FACTOR RATE
<%) (Btu/kUh)
50 8000 mills/kUh
10000
70 6000 mills/kUh
10000
50 BOOO miUs/Wh
10000
70 8000 mills/kWh
10000
50 8000 mills/kWh
10000
70 8000 millsAUh
10000
50 BOOO mills/kUh
10000
70 8000 mills/kWh
10000
100
171.3
201.4
130.7
154.4
172.9
203.4
131.8
155.6
174.1
204.8
132.7
156.4
175.2
206.0
133.4
157.3
BOILER
300
131.2
156.8
102.0
122.3
132.3
157.9
102.7
123.0
133.0
158.7
103.3
123.7
133.7
159.4
103.6
124.1
GENERATING
500
118.7
142.3
93.1
112.0
119.5
143.2
93.6
112.5
120.1
143.9
94.0
113.1
120.6
144.6
94.2
113.4
CAPACITY
700
111.7
134.2
88.0
106.2
112.4
134.9
88.4
106.8
112.9
135.6
88.7
107.1
113.2
136.0
88.9
107.5
(MW)
1000
105.2
126.7
83.3
100.8
105.7
127.2
83.7
101.2
106.1
127.8
83.1
101.5
106.4
128.1
84.0
101.7
1300
100.8
121.6
80.3
97.1
101.2
122.2
80.5
94.5
101.5
122.5
80.7
97.8
101.9
122.9
80.9
98.0
                                                        D-3

-------
       APPENDIX E





Atmospheric Fluidized Bed
            E-l

-------
AFBC-$/kU
  SULFUR    CAPACITY      HEAT                               BOILER GENERATING CAPACITY (MU)
  CONTENT    FACTOR        RATE
   (%)         (%)       (Btu/kUh)                    100      300      500      700       1000     1300
      1.0           50        9000        $/kU       2,018    1,554    1,386    1,289    1,195    1,132
                            11000                   2,346    1,8H    1,621    1,509    1,401    1,328
      1.0           70       9000        t/kU       2,018    1,554    1,386    1,289    1,195    1,132
                            11000                   2,346    1,814    1,621    1,509    1,401    1,328
      2.0           50        9000        t/kU       2,031    1,562    1,393    1,295    1,200    1,137
                            11000                   2,361    1,823    1,628    1,515    1,407    1,333
      2.0           70        9000        $/kU       2,031    1,562    1,393    1,295     1,200    1,137
                            11000                   2,361    1,823    1,628    1,515     1,407    1,333
      3.0           SO       9000        S/kU       2,042    1,568    1,398    1,299    1.205    1,141
                            11000                   2,372    1,830    1,634    1,521     1,412    1,338
      3.0            70       9000        $/kU       2,042    1,568    1,398    1,299    1.205     1,141
                            11000                   2,372    1,830    1,634    1,521     1,412     1,338
      4.0            50       9000        */kW       2,050    1,574    1,403    1,304     1,209    1,145
                            11000                   2,382    1,836    1,640    1,526     1,416    1,342
      4.0           70       9000        S/kU       2,050    1,574    1,403    1,304     1.209    1,145
                           11000                   2,382    1,836.    1,640    1,526     .1,416    1,342
                                                           E-2

-------
AFBC-mills/kUh
  SULFUR    CAPACITY      HEAT
  CONTENT    FACTOR       RATE
   (%)         (X)      (Btu/kUh)
                                         BOILER GENERATING CAPACITY  (HU>
                                100      300      500      700
                                                             1000      1300
      1.0
50       9000     mills/lcUh     148.9    118.8    108.7    102.9    97.3      93.8
        11000                   172.9    139.5    127.9    121.5    115.3     111.1
      1.0
70
 9000
11000
mills/kWh
116.4.   95.4
135.8    111.8
         87.7
         103.6
         83.5
         98.9
         79.6
         94.5
         77.0
         91.7
      2.0
50
 9000
11000
iniUs/kUh
151.7
176.2
121.3
142.5
110.0
130.7
105.2
124.3
99.8
118.1
96.1
113.9
      2.0
70       9000     mills/tcWh     119.0    97.3     90.0     85.8     81.9     79.3
        11000                   138.8    114.6    106.4    101.7    97.3     94.3
      3.0
50       9000     mills/kUh     154.2    123.6    113.2     107.5     102.0    98.4
        11000                   179.4    145.3    133.7     127.0     120.8    116.7
      3.0
70       9000     mills/kWh     121.5     99.6    92.2     88.0     84.0     81.6
        11000                   141.8     117.4    109.2    104.5    100.0    97.0
      4.0
50       9000     mills/kWh     156.8    126.0     115.5     109.7     104.3    100.6
        11000                   182.4    148.1     136.5     129.9     123.6    119.4
      4.0
70       9000     mills/kUh     123.7    101.9    94.3     90.1     86.3     83.7
        11000                   144.6    120.2    111.8    107.1    102.7    99.8
                                                           E-3

-------
   APPENDIX F





Lime Spray Drying
        F-l

-------
LIME SPRAY DRYING WITH  REUSE OF THE EXISTING ESPs
LSD+ESP (URGE ESP:400  SCA)
CAPITAL COST <$/kV)
SULFUR CAPACITY RETROFIT
CONTENT FACTOR FACTOR
(X) (X) (X)
1.0
1 30 1.5
2.0
1.0
1 50 1.5
2.0
1.0
1 70 1.5
2.0
1.0
2 50 1.5
2.0
1.0
2 50 1.5
2.0
1.0
2 70 1.5
2.0
1.0
3 30 1.5
2.0
1.0
3 50 1.5
2.0
1.0
3 70 1.5
2.0
1.0
4 30 1.5
2.0
1.0
4 50 1.5
2.0
1.0
4 70 1.5
2.0
BOILER GENERATING CAPACITY (MW)
100 300 500 700 1000

t/kU .


S/kU


S/kW


S/kU


S/kU


SAW


S/kU


S/kU


S/kU


S/kU


S/kU


t/ku

209.8
301.8
393.7
210.8
302.8
394.7
211.8
303.7
395.6
231.6
333.7
435.8
233.5
335.6
437.7
235.2
337.3
439.4
255.2
368.4
481.6
257.8
371.0
484.2
260.3
373.5
486.7
279.6
404.3
529.0
282.9
407.6
532.3
286.0
410.7
535.4
106.1
152.5
198.9
106.8
153.2
199.6
107.5
153.9
200.3
117.6
169.2
220.9
118.9
170.6
222.2
120.2
171.8
223.5
130.2
187.6
245.1
132.1
189.5
247.0
133.9
191.4
248.8
143.2
206.7
270.2
145.7
209.2
272.7
148.1
211.5
275.0
84.8
122.1
159.4
85.4
122.7
160.0
86.0
123.3
160.6
94.2
135.7
177.3
95.3
136.9
178.4
96.5
138.0
179.6
104.5
150.8
197.1
106.2
152.5
198.8
107.9
154.2
200.5
115.3
166.5
217.7
117.5
168.7
220.0
119.7
170.9
222.1
75.5
108.9
142.3
76.1
109.5
142.9
76.6
110.0
143.4
84.0
121.2
158.4
85.1
122.3
159.5
86.2
123.4
160.6
93.4
134.9
176.4
95.0
136.5
178.0
96.6
138.1
179.6
103.1
149.1
195.1
105.3
151.2
197.2
107.3
153.3
199.3
68.5
98.9
129.4
69.0
99.5
130.0
69.6
100.0
130.5
76.3
110.3
144.2
77.3
111.3
145.3
78.4
112.4
146.3
85.0
122.9
160.8
86.5
124.4
162.3
88.0
125.9
163.8
94.0
136.0
178.1
96.0
138.0
180.1
98.0
140.0
182.1
1300
64.7
93.6
122.5
65.3
94.2
.123.1
65.8
94.7
123.6
72.2
104.4
136.6
73.2
105.4
137.7
74.2
106.4
138.7
80.5
116.4
152.4
82.0
117.9
153.9
87.5
123.5
159.5
89.1
129.0
168.9
95.1
135.0
175.0
97.3
137.3
177.2
                                                           F-2

-------
LINE SPRAY DRYING WITH  REUSE OF  THE EXISTING ESPs
LSD+ESP (LARGE ESP:400  SCA)
ANNUAL COST (mills/kWh)
SULFUR CAPACITY RETROFIT
CONTENT FACTOR FACTOR
 (%) (%)

1


1


1


2


2


2


3


3


3


4


4


4


30


50


70


30


50


70


30


50

•
70


30


50


70

1.0
1.5
2.0
1.0
1.5
2.0
1.0
1.5
2.0
1.0.
1.5
2.0
1.0
1.5
2.0
1.0
1.5
2.0
1.0
1.5
2.0
1-0
1.5
2.0
1.0
1.5
2.0
1.0
1.5
2.0
1.0
1.5
2.0
1.0
1.5
2.0

miUs/kUh


(irills/kUh


mills/kUh


mills/kWh


mills/kUh


mills/kUh


mills/kUh


mills/kUh


mills/kWh


mills/kUh


mills/kUh


mills/kUh

BOILER GENERATING CAPACITY (MU)
100 300 500 700 1000
30
39.9
49.8
19.8
25.7
31.6
15.5
19.8
24
32.8
43.8
54.8
22
28.6
35.2
17.7
22.4
27.1
35.8
48
60.1
24.3
31.6
38.9
19.9
25.1
30.3
38.8
52.3
65.7
26.7
34.7
42.8
22.1
27.9
33.7
15.7
20.7
25.7
10..7
13.7
16.7
8.7
10.9
13
17.4
22.9
28.5
12.2
15.6
18.9
10.4
12.8
15.1
19.2
25.4
31.5
13.8
17.5
21.2
12
14.7
17.3
21
27.9
34.7
15.4
19.5
23.6
13.7
16.7
19.6
12.8
16.8
20.8
8.9
11.3
13.7
7.4
9.1
10.8
14.2
18.7
23.2
10.2
12.9
15.6
8.9
10.8
12.7
15.8
20.8
25.7
11.7
14.7
17.7
10.5
12.6
14.7
17.4
22.9
28.4
13.2
16.5
19.8
12
14.4
16.8
11.5
15.1
18.7
8.1
10.3
12.4
6.8
8.3
9.9
12.9
16.9
20.9
9.4
11.8
14.2
8.3
10
11.7
14.3
18.8
23.3
10.8
13.5
16.2
9.8
11.7
13.6
15.8
20.8
25.7
12.2
15.2
18.1
11.3
13.5
15.6
10.7
13.9
17.2
7.6
9.5
11.5
6.4
7.8
9.2
11.9
15.6
19.2
8.8
11
13.2
7.9
9.4
11
13.3
17.4
21.5
10.2
12.6
15.1
9.4
11.1
12.9
14.7
19.2
23.8
11.5
14.2
17
10.9
12.8
14.7
1300
10.3
13.4
16.5
7.4
9.2
11.1
6.3
7.7
9
11.5
15
18.4
8.6
10.7
12.8
7.7
9.2
10.7
12.8
16.7
20.6
9.9
12.3
14.6
9.4
11
12.7
14.2
18.5
22.8
11.5
14.1
16.7
10.9
12.7
14.6
                                                            F-3

-------
LIME SPRAT DRYING  WITH  REUSE OF THE EXISTING ESPS
LSD+ESP (LARGE  ESP:400  SCA)
S02 COST EFFECTIVENESS  (S/TOH)
SULFUR CAPACITY RETROFIT
CONTENT FACTOR FACTOR
(X) (X) (X)

1 30


1 50


1 70


2 . 30

,•{• ' ••".•1"
2 50


2 70


3 30


3 50


3 70


4 30


4 50


4 70

1.0
1.5
" 2.0
1.0
1.5
2.0
1.0
1.5
2.0
1.0
1.5
2.0
• -.1.0
1.5
2.0
1.0
1.5
2.0
1.0
1.5
2.0
1.0
1.5
2.0
1.0
1.5
2.0
1.0
1.5
2.0
1.0
1.5
2.0
1.0
1.5
2.0

S/TON


S/TON


S/TON


S/TON


S/TON


S/TON


S/TON


S/TON


S/TON


S/TON


S/TON


S/TON

BOILER GENERATING CAPACITY (HU)
100 300 500 700 1000
4868.4
6476.4
8084.6
3208.9
4173.7
5138.6
2526.2
3215.4
3904.6
2661.6
3554.6
4447.6
1784.5
2320.3
2856.1
1436.6
1819.3
2202
1936.6
2596.7
3256.7
1315.9
1711.9
2107.9
1077.7
1360.5
1643.4
1577.5
2122.9
2668.3
1083.5
1410.7
1737.9
899.5
1133.2
1366.9
2549.9
3361.7
4173.5
1738.3
2225.3
2712.4
1419.4
1767.3
2115.2
1411.3
1863
2314.6
993.1
1264.1
1535
842.2
1035.7
1229.3
1038.1
1373
1707.8
748.3
949.3
1150.2
652.3
795.8
939.3
853.6
1131.1
1408.7
627.2
793.7
960.2
558.1
677.1
796.1
2074.5
2727
3379.5
1439.4
1830.9
2222.4
1196.1
1475.8
1755.4
1155
1518.4
1881.8
832
1050
1268.1
721.9
877.6
1033.4
854.1
1123.9
1393.7
632.8
794.7
956.6
566.1
681.7
797.3
705.4
929.4
1153.4
534.2
668.6
803
488.8
584.8
680.8
1872.7
2457
3041.3
1314.6
1665.2
2015.8
1104.3
1354.7
1605.1
1046.2
1371.7
1697.3
764.6
960
1155.3
672.3
811.8
951.3
775.8
1017.8
1259.8
584.3
729.5
874.7
530.4
634.1
737.8
642.4
843.5
1044.5
495.2
615.8
736.4
460.1
546.3
. 632.4
1730.5
2263.5
2796.6
1230.5
1550.3
1870.2
1045
1273.5
1501.9
969.1
1266.3
1563.5
718.9
897.2
1075.5
639.9
767.3
894.6
720.3
941.4
1162.4
551.2
683.9
816.5
506.9
601.7
696.4
597.5
781.4
965.2
468.4
578.7
689
441.1
519.9
598.6
1300
1667.8
2173.3
2678.8
1198.8
1502.1
1805.4
1026.7
1243.3
1460
934.7
1216.7
1498.6
701.2
870.3
1039.5
629.4
750.2
871
695.2
905
1114.8
538.1
664
789.9
507.9
597.9
687.8
577
751.6
926.2
466.9
571.6
676.4
441.5
516.4
591.2
                                                            F-4

-------
LINE SPRAY DRYING WITH A NEW BAGHOUSE
LSD+FF
CAPITAL COST  ($/kU)
SULFUR CAPACITY RETROFIT
CONTENT FACTOR FACTOR

-------
LIME SPRAT DRYING WITH  A  NEW BAGHOUSE
LSD+FF
ANNUAL COST (mills/kUh)
SULFUR CAPACITY RETROFIT
CONTENT FACTOR FACTOR
(*) (X) (X)
1.0
1 30 "l.S
2.0
1.0
1 50 1.5
2.0
1.0
1 70 1.5
2.0
vo
2 30 1.5
2.0
1.0
2 50 1.5
2.0
1.0
2 70 1.5
2.0
1.0
3 30 1.5
2.0
1.0
3 50 1.5
2.0
1.0
3 70 1.5
2.0
1-0
4 30 1.5
2.0
1.0
4 50 1.5
2.0
1.0
4 70 1.5
2.0
BOILER GENERATING CAPACITY (MU)
100 300 500 700 1000

mills/kUh


mills/kUh


mills/kWh


mills/kUh


mills/kUh


mills/kUh


mllls/kuh


mills/kWh


mills/kUh


mills/kWh


mills/kUh


mills/kUh

40.2
50.1
59.9
26.1
32.0
37.9
20.2
24.5
28.7
43.0
53.9
64.9
26.3
34.8
41.4
22.3
27.0
31.7
45.9
58.1
70.2
30.6
37.8
45.1
24.5
29.7
34.9
49.0
62.4
75.7
32.9
40.9
49.0
26.7
32.5
36.2
24.0
19.0
34.0
15.9
18.9
21.9
12.6
14.7
16.9
25.7
31.2
36.8
17.4
20.7
24.0
14.2
16.6
19.0
27.4
33.6
39.8
19.0
22.7
26.4
15.9
18.5
21.1
29.3
36.1
42.9
20.6
24.6
28.7
17.5
20.4
23.4
20.7
24.7
28.7
13.8
16.2
18.6
11.1
12.8
14.5
22.1
26.6
31.0
15.2
17.8
20.5
12.6
14.5
16.4
23.7
28.6
33.6
16.6
19.6
22.5
14.1
16.2
18.3
25.2
30.7
36.2
18.0
21.3
24.6
15.6
18.0
20.4
19.7
23.3
26.8
13.2
15.3
17.5
10.6
12.1
13.7
21.0
25.0
29.0
14.5
16.9
19.3
12.0
13.7
15.5
22.5
26.9
31.4
15.8
18.5
21.2
13.5
15.4
17.4
23.9
28.9
33.8
17.2
20.2
23.1
15.0
17.2
19.3
18.6
21.9
25.1
12.5
14.5
16.5
10.1
11.5
12.9
19.9
23.5
27.2
13.8
16.0
18.2
11.8
13.4
14.9
21.2
25.3
29.4
15.5
17.9
20.3
13.3
15.0
16.8
22.6
27.1
31.6
16.8
19.5
22.2
14.8
16.7
18.6
1300
18.5
21.6
24.7
12.6
14.4
16.3
10.2
11.5
12.8
19.8
23.2
26.7
13.8
15.9
17.9
11.6
13.1
14.6
21.1
25. 0
28.8
15.1
17.4
19.7
13.0
14.7
16.3
22.5
26.8
31.0
16.4
19.0
21.5
14.5
16.3
18.2
                                                            F-6

-------
LIME SPRAY DRYING  WITH A NEW BAGKOUSE
LSD+FF
S02 COST EFFECTIVENESS (S/TON)
SULFUR CAPACITY RETROFIT
CONTENT FACTOR FACTOR
(X) (X) (X)

1


1


1


2


2


2


3


3


3


4


4


4

1.0
30 1.5
2.0
1.0
50 1.5
2.0
1.0
70 1.5
2.0
1.0
30 1.5
2.0
1.0
50 1.5
2.0
1.0
70 1.5
2.0
1.0
30 1.5
2.0
1.0
50 1.5
2.0
1.0
70 1.5
2.0
1.0
30 1.5
2.0
1.0
50 1.5
2.0
1.0
70 1.5
2.0

S/TON


S/TON


S/TON


S/TON


S/TON


S/TON


S/TON


S/TON


S/TON


S/TON


S/TON


S/TON

BOILER GENERATING CAPACITY
100 300 500 700
5790.9
7211.6
8632.4
3757.0
4609.5
5462.0
2913.5
3522.3
4131.3
. 3094.5
3883.5
4672.4
2035.5
2508.9
2982.3
1607.6
1945.9
2284.1
2205.3
2788.5
3371.7
1467.4
1817.3
2167.2
1176.9
1426.9
1676.8
1763.8
2245.7
2727.5
1185.3
1474.5
1763.6
962.8
1169.3
1375.8
3460.8
4177.9
4895.0
2289.8
2720.1
3150.3
1816.3
2123.7
2431.0
1849.4
2248.4
2647.4
1252.7
1492.1
1731.4
1023.2
1194.2
1365.2
1318.1
1613.9
1909.7
910.4
1087.9
1265.4
761.2
888.0
1014.7
1054.2
1299.5
1544.7
740.3
887.4
1034.6
630.9
736.0
841.1
2981.6
3558.0
4134.3
1990.1
2335.9
2681.7
1593.5
1840.5
2087.5
1593.7
1914.6
2235.6
1092.7
1285.3
1477.9
904.3
1041.8
1179.4
1136.0
1374.3
1612.7
796.5
939.6
1082.6
676.6
778.8
880.9
908.8
1106.7
1304.6
649.4
768.2
886.9
563.5
648.3
733.1
2834.1
3350.2
3866.2
1898.2
2207.9
2517.5
1525.4
1746.6
1967.8
1513.0
1800.5
2088.1
1042.5
1215.0
1387.5
867.1
990.3
1113.5
1078.1
1291.8
1505.6
760.5
888.7
1017.0
649.9
741.5
833.1
861.5
1039.2
1216.8
620.0
726.6
833.2
541.7
617.8
693.9
(MU)
1000
2681.0
3151.8
3622.6
1807.3
2089.8
2372.3
1461.2
1663.0
1864.8
1431.3
1693.7
1956.2
993.8
1151.3
1308.8
850.4
962.9
1075.4
1019.1
1214.4
1409.6
742.0
859.2
976.3
637.4
721.0
804.7
814.5
976.9
1139.3
604.9
702.3
799.7
531.4
601.0
670.6
1300
2672.0
3118.5
3564.9
1807.9
2075.8
2343.7
1465.9
1657.3
1848.6
1425.0
1674.0
1922.9
993.4
1142.8
1292.2
834.6
941.3
1048.0
1013.6
1198.9
1384.2
724.4
835.6
946.8
626.0
705.4
784.8
809.3
963.5
1117.7
590.7
683.2
775.8
522.2
588.3
654.4
                                                            F-7

-------
         APPENDIX G





Selective Catalytic Reduction
              G-l

-------
POST-CWBUrriW NOB aMTROlS:
SELECTIVE CATALYTIC tEDUCTKM
CAPITAL COST (S/kU)
CATALYST     Ml     CAPACITY   1ET80MT
  lift    REDUCTION   f ACTOR      r ACTOR
 (TEARS)     (X)        (X)         (X)
      BOILER  GENERATINO CAPACITY (MU)

100      500      500      700       1000     1300
1.00
1 70 30 f.SO
2.00
1.00
1 80 30 1.50
2.00
1.00
1 70 50 1.SO
2.00
1.00
1 80 50 1.30
2.00
1.00
1 70 70 1.50
2.00
1.00
1 80 70 1.50
2.00
1.00
3 70 30 1.50
2.00
1.00
3 80 30 1.50
2.00
1.00
3 70 SO 1.50
2.00
1.00
3 80 SO 1.50
2.00
1.00
3 70 70 1.50
2.00
1.00
3 SO 70 1.50
2.00
1/ku
SAW
t/ut
s/KU
S/kW
s/kv
t/kw
t/ku
»/fcW
>/kw
*/ku
»/ku
133
176.7
220.5
143
191.8
240.}
133
176.8
220.5
143.1
191.8
240.6
133
176.8
220.6
143.1
191.9
240.6
133
176.7
220.5
143
191.8
240.5
133
176.8
220.5
143.1
191.8
240.6
133
176.8
220.6
143.1
191.9
240.6
103.2
133
162.8
111
144.7
178.4
103.2
133
162.8
111.1
144.8
178.5
103.2
133
162.8
111.1
144.8
178.5
103.2
133
162.8
111
144.7
178.4
103.2
133
16T8
111.1
144.8
178.5
103.2
133
162.8
111.1
144.8
178.5
91.3
118.5
143.5
98.9
129.6
160.3
91.5
118.5
145.5
99
129.6
160.3
91.6
118.6
145.6
99
129.7
160.4
91.5
118.5
145.5
98.9
129.6
160.3
91. S
118.3
145.5
98.9
129.6
160.3
91. S
118.6
141.6
99
129.7
160.4
88.6
114.4
140.3
95.9
125.3
154.7
88.7
114.5
140.3
95.9
125.3
154.7
88.7
114.5
140.3
96
125.4
154.8
88.6
114.4
140.3
95.9
12S.3
154.7
88.7
114.1
140.3
95.9
125.3
154.7
88.7
114.5
140.3
96
129.4
154.8
86.6
111.5
136.4
93.7
122.1
150.6
86.6
111.6
136.5
93.8
122.2
150.6
86.7
111.6
136.5
93.8
122.2
150.7
86.6
111.5
136.4
93.7
122.1
150.6
86.6
111.6
136.5
93.7
122.2
150.6
86.7
111.6
136.3
93.8
122.2
150.7
85. 4
109.8
134.2
92.4
120.3
148.2
85.4
109.9
134.3
92.5
120.4
148.3
85.3
109.9
134.3
92.5
120.4
148.3
85.4
109.8
134.2
92.4
120.3
148.2
85.4
109.8
134.3
92.5
120.4
148.3
85.5
109.9
134.3
92.S
120.4
148.3
                                                           G-2

-------
POST'OMUSTION MR CWTKXS:
SELECTtVf CATALYTIC REDUCTION
ANNUM. COST (•fllt/lrwri)
CATALYST NO* CAPACITY RETROFIT
LIFE REDUCTION FACTOR FACTO*
(TEAR!) (X) (X) (X)
1.00
1 70 30 1.50
2.00
1.00
1 80 30 1.50
2.00
1.00
1 70 50 1.50
2.00
1.00
1 60 50 1.50
2.00
1.00
1 70 70-1.50
2.00
1.00
1 80 70 1.50
2.00
1.00
3 70 30 1.50
2.00
1.00
3 80 30 1.50
2.00
1.00
3 70 50 1.50
2.00
1.00
3 80 50 1.50
2.00
+
1.00
3 70 70 1.50
2.00
1.00
3 80 70 1.50
2.00
•OILER GENERATING CAPACITY (MW)
100 300 500 700 1000

•fttlAUl


•UU/tMh


•ttU/kUi


•Uti/ktfi


•UU/Mft


•UU/Uft


•ills/Mi


•lUl/blil


•UU/Ml


•llU/Uh


•iUs/kuh


•Ult/Mft

29.4
33.4
37.5
30.4
34.9
39.4
17.9
20.3
22.8
18.3
21.2
23.9
13
14.7
16.5
13.4
15.4
17.3
18.4
22.3
26.6
19.4
23.9
28.5
11.3
13.7
U.2
11.9
14.6
17.4
8.3
10
11.8
8.7
10.7
12.6
26.1
28.8
31.6
26.9
30
33.1
15.9
17.6
19.2
16.4
18.3
20.2
11.6
12.7
13.9
11.9
13.3
14.6
15.1
17.9
20.7
13.9
19.1
22.2
9.3
,11
ri-7
9.8
11.7
13.6
6.9
8
9.2
7.2
8.6
9.9
25
27.5
30
25.8
28.6
31.3
15.3
16.8
18.3
15.7
17.4
19.2
11.1
12.2
13.3
11.5
12.7
13.9
14.1
16.6
19.1
14.8
17.7
20.5
8.7
10.2
11.7
9.2
10.9
12.6
6.4
7.5
8.6
6.8
8
9.2
24.7
27.1
29.5
25.4
28.2
30.9
15.1
16.5
17.9
15.5
17.2
18.8
11
12
13
11.3
12.5
13.6
13.7
16.1
18.5
14.5
17.2
19.9
8.5
9.9
11.4
9
10.6
12.2
6.)
7.3
8.3
6.6
7.8
9
24.5
26.8
29.1
25.2
27.9
30.5
14.9
16.3
17.7
15.4
17
18.6
10.9
11.9
12.9
11.2
12.3
13.5
13.6
13.9
18.2
14.3
16.9
19.6
8.4
9.8
11.2
8.8
10.4
12
6.
7.
8.
6.
7.
8.8
1300
24.3
26.6
28.9
25.1
27.6
30.2
14.9
16.2
17.6
13.3
16.9
18.4
10.8
11.8
12.8
11.1
12.3
13.4
13.4
13.7
17.9
14.1
16.7
19.3
8.3
9.7
11
8.7
10.3
11.9
6.1
7.1
8.1
6.5
7.6
8.7
                                                         G-3

-------
              MR CMTMX.S:
SCLECTIVI CATALYTIC KWCTIW
S02 COST EFFECTIVENESS (S/TOH)
CATALYST NOx CAPACITY MTtOFIT
LIFE •EDUCTKM FACTO* FACTM
(YtARS) (X) (X) (X)
1.00
t 70 30 1.50
2.00
1.00
1 80 30 1.50
2.00
1.00
t 70 SO 1.50
2.00
1.00
1 80 50 1.50
2.00
1.00
1 70 70 1.50
2.00
1.00
1 80 70 1.50
2.00
1.00
3 70 30 1.50
2.00
1.00
3 80 30 1.50
2.00
1.00
3 70 50 1.50
2.00
1.00
3 80 SO 1.50
2.00
1.00
3 70 70 1.50
2.00
1.00
3 80 70 1.50
2.00

(/TON
«.
(/TON
(/TON
(/TON
(/TON
(/TON
(/TON
(/TON
(/TON
(/TON
(/TON
100
9657.6
10996.2
12334.1
8735.4
10039.9
11344.2
5877.9
6681
7483.7
5320.6
6103.1
6885.8
4263.8
4837.4
S410.8
3862.3
4421.4
4980.4
6058.5
7396.9
8734.9
5586.1
6890.5
8195
3718.5
4521.5
5324.2
3431
4213.6
4996.3
2721.4
3295
3868.4
2512.7
3071.7
3630.8
(OILER XNEUTINO CAPACITY (MU)
300 500 700 1000
8576
9487.1
10398.1
7730.2
8631.8
9533.3
5228
5774.8
6321.4
4716.7
5257.7
5798.6
3799
4189.5
4579.9
3430.3
3816.7
4203.1
4977.1
5888.3
6799.4
4581.2
5482.8
6384.3
3068.7
3615.5
418i>1
2827.3
3368.3
3909.2
2296.7
2647.2
3037.6
2080.8
2467.2
2893.6
8230.3
9055.9
9881.6
7415.9
8236.9
9057.9
5020.4
5515.8
6011.2
4528
5020.6
5513.2
3650.6
4004.4
4358.3
3295.4
3647.2
3999.1
4631.3
5457
6282.6
4266.8
5087.8
5908.8
2861.1
3356.5
3851.9
2638.3
3131.1
3623.7
2108.2
2462
2815.9
1945.8
2297.6
2649.5
8119.6
8908.7
9697.8
7314.1
8100.6
8887.1
4954
5427.4
5900.8
4466.8
4938.7
5410.6
3603.1
3941.2
4279.4
3251.6
1588.7
3925.7
4520.7
5309.8
6098.9
4165.1
4951.6
5738
2794.6
3268.1
3741.5
2577.4
3049.3
3521.2
2060.7
2398.9
2737
1902.1
2239.1
2576.2
8056.6
8818.2
9579.8
7255.2
8015.7
8776.3
4916.1
S373.1
5830
4431.4
4887.8
5344.1
3576
3902.4
4228.8
3226.3
3552.3
3878.2
4457.7
5219.3
5980.9
4106.1
4866.7
5627.2
2756.8
3213.7
3670.7
2542
2998.3
3454.7
2033.6
2360
2686.4
1878.7
2202.7
2528.6
1300
8005.8
8752.7
9499.5
7208.8
7955.4
8702.1
4885.6
S333.8
3781.9
4403.6
4851.5
5299.5
3554.2
3874.3
4194.3
3206.4
3526.4
3846.3
4406.9
S153.8
5900.6
4059.7
4806.4
5553
2726.3
3174.4
3622.3
2514.1
2962.1
3410.1
2011.8
2331.9
2652
1856.8
2176.8
2496.8
                                                    G-4

-------
       APPENDIX H





Furnace Sorbent Injection
            H-l

-------
FUMACt KXKIT  INJICTION AND HIMIDIFICATION
Fll*f$f (UUtfif IS»:400 SCA>
CAPii*i cos: (S/U)
  fULFM    CAJIACITT      S02
  COMTEK!    FACTO!     «£t«VAL
    <*>        (t)        (X)
      •01LM GEMUATING CAPACITY  (Ml)

100     300      500      TOO      1000
                                                                                            1300

1 30


1 50


1 70


2 30


2 50


2 70


3 30


3 50


3 70


4 30


4 50


4 70

30
50
70
30
50
70
30
50
70
30
50
70
30
50
70
30
50
70
30
SO
70
30
50
70
30
30
70
SO
50
70
30
SO
70
50
SO
70

*/kU


*/kU


s/kw


t/kW


*/kU


l/fcu


S/kH


Vku


VKW


t/kv


S/ku


8/W

54
54.9
57.7
54
54.9
57.7
54
54.9
57.7
44.5
48.3
44.3
44.3
45.5
44.4
44.3
45.5
44.4
71.4
73
74.3
71.4
75
74.5
71.4
75
74.5
78.2
79.9
81.4
78.2
79.9
81.4
78.5
79.9
•1.4
34.8
33.2
35.7
34.8
33.2
33.7
54.8
33.3
33.7
41.5
42
42.4
41.5
42
42.4
41.5
42
42.4
41
48.9
49.7
a
48.9
49.7
48.1
48.9
49.7
34.4
59.4
54.3
54.4
53.4
54.5
54.4
39.4
54.5
28.4
28.7
29.1
28.4
28.8
29.1
28.4
28.8
29.1
34.9
39.4
33.9
34.9
39.4
31.9
34.9
35.4
53.9
57
57.7
38.4
57.1
57.8
38.4
57.1
57.8
3*4
41.7
42.4
43.3
41.7
42.4
43.3
41.8
42.4
45.4
24.4
24.9
27.2
24.4
24.9
27.2
24.4
24.9
27.2
52.4
32.9
35.4
32.3
32.9
55.4
32.3
52.9
55.4
54.3
54.9
33.3
54.3
34.9
33.1
54.5
34.9
53.9
58.9
59.7
40.4
59
59.7
40.4
59
59.7
40.4
24.4
24.9
23.1
24.4
24.9
25.1
24.4
24.9
25.1
27.8
28.2
28.4
27.8
28.2
28.4
27.8
28.2
28.4
52.4
52.9
. 55.4
32.4
52.9
33.4
52.4
32.9
55.4
37.3
37.9
584
57.5
38
584
57.5
38
38.4
23.4
23.8
24.1
25.4
23.8
24.1
23.4
23.8
24.1
24.8
27.2
27.4
24.8
27.2
27.4
24.9
27.2
27.4
31.7
32.3
32.7
31.8
32.3
32.7
31.8
52.3
32.8
38.4
39.5
59.9
58.7
59.5
59.9
38.7
39.3
39.9
                                                        H-2

-------
WRNACt SORgENT INJECTION AND (MODIFICATION
FSI«€SP (LARGE E»t400  SCA)
ANNUAL COST (•!(!•/UK)
  SULFUR    CAPACITY      S02
  CONTENT     FACTOR     REMOVAL
    (X)         (X)    "   <*>
      BOILER GENERATING CAP AC ITT (MO

100     300      SOD      700       1000
                                                                                              1100

1


1


1


2


2


2


3


3


3


4


4


4


30


SO


70


30


SO


70


30


SO


70


30


50


70

30
50
70
30
50
70
30
50
70
30
50
70
30
50
70
30
SO
70
30
50
70
30
50
70
30
50
70
30
SO
70
30
50
70
30
50
70

•ids/Mi


•1 lit/Mi


•UU/Mi


•Ult/kVh


•flU/KMi


•UU/Mi


•lUi/kWh


•Ult/Hh


•lilt/Mi


•Ills/Mi


•Uli/kWh


•tUs/Uh

12.1
12.2
12.4
.3
.4
.5
.6
.7
.8
1S.2
15.4
15.6
11
11.2
11.4
9.3
9.4
9.6
18.1
18.4
18.7
13.7
14
14.3
11.9
12.1
12.3
21
21.4
21.8
16.4
16.7
17.1
14.5
14.8
15.1
7.2
7.3
7.4
5.3
5.4
5.5
4.5
4.6
4.7
10.1
10.3
10.5
8
8.2
8.3
7.1
7.2
7.4
13
13.3
13.3
10.7
10.9
11.1
9.7
9.9
10.1
1S.9
16.2
16.6
13.3
13.6
13.9
12.3
12.5
12.8
6
6.1
6.2
4.6
4.7
4.8
4
4.1
4.2
8.9
9.1
9.3
7.3
7.4
7.6
6.6
6.7
6.9
11.4
11.6
11.9
9.7
9.9
10.1
9
9.2
'#4
14.1
14.4
14.7
12.3
12.5
12.8
11.9
11.8
12
5.6
5.7
5.8
4.4
4.4
4.S
3.8
3.9
4
8.5
8.6
8.8
7
7.2
7.3
6.4
6.5
6.7
10.9
11.1
11.3
9.4
9.6
9.8
8.8
9
9.2
13.6
13.9
14.2
12
12.
12.
11.
11.
11.
5.3
5.3
5.4
4.1
4.2
4.3
3.7
3.7
3.8
7.8
8
8.1
.6
.8
.9
.1
.2
.4
10.5
10.7
11
9.2
9.4
9.6
8.
8.

13.
13.
13.
11.7
12
12.3
11.1
11.4
11.6
5.1
5.1
5.2
4
4.1
4.2
3.6
3.7
3.7
7.6
7.8
7-9
6.3
6.6
6.8
6
6.2
6.3
10.4
10.6
10.8
9.1
9.3
9.5
8.5
8.7
8.9
13.3
13.6
13.9
11.8
12.1
12.3
11.1
11.4
11.7
                                                         H-3

-------
FUUUCt SOUU1  INJECTION AND MUMMIFICATION
FS!*€$>> (uutoe E»:4oo sou
$02 COST EFrecrivewn <«/TON>
SULFUR CAPACITY $02
CONTENT ttCTOt . «ENOVAl
(X) (X) (*)
1
1
1
2
2
2
3
5
3
4
'• 4
4
30
SO
70
30
SO
70
30
SO
70
30-
SO
70
30
SO
70
30
SO
70
30
SO
70
30
SO
70
30
SO
70
30
SO
70
30
SO
70
30
SO
70
30
SO
70
30
SO
70
30
SO
70
30
SO
70
I/TOI
I/TON
•/TON
I/TOH
•/TON
.(/TOM
S/TON
»/TCB
•/TON
S/TOH
•/TON
S/TOi
80ILEI CCNEIATING CAPACITY 
-------
    APPENDIX I





Natural Gas Reburn
         1-1

-------
NATUUL GAI UKJM
CAPITAL COST (S/tW>
FUEL MICE CAPACITY fUtl
0IFFCRENTIAL fACTM SUMTITUTCD
<»> . (X)

1 S/IMTU


1 S/IMTU

i S/IMTU
2 I/IMTU

2 S/IMTU


2 S/IMTU


3 S/IMTU


3 S/IMTU


3 i/wwrv


30 IS
a
3
SO 1S
29
70 IS
29
30 IS
29
S
SO IS
29
S
70 IS
29
S
30 IS
29
S
SO IS
29
S
70 IS
29 •
MILCH CEttRATIM CAPACITY *
S/ku 23.1
27.9
18.3
S/kU 23.1
27.9
18.3
S/kU 23.1
27.9
19.4
S/kH 26.9
34.2
19.5
S/kW 26.9
34.2
19.S
»/kW 26.9
S4.2

1S.7
16
11.4
13.7
16
13.7
16
19 A
I*»O
17.4
22.2
12.6
17.4
22.2
12.6
17.4
22.2
13.9
21.2
28.5
13.9
21.2
28.5
13.9
21.2
28.5

11.8
14.1
9.3
11.8
14.1
11.8
14.1
15. S
20.1
10.7
15.5
20.3
10.7
15.5
20.3
12
19.3
26.6
12
19.3
26.6
12
19.3
3i.6

10.7
13
8.4
10.7
13
10.7
13
14.S
19.3
9.7
U.S
19.3
9.7
U.S
19.3
10.9
18.2
29.3
10.9
18.2
29.5
10.9
18.2
29.9

9.8
12.1
7.5
9.8
12.1
9.8
12.1
13.5
18.3
8.7
13.5
18.3
8.7
13.9
18.3
10
17.3
24.6
10
17.3
24.6
10
17.3
24.6
1300

9.1
11.4
6.8
9.1
11.4
9.1
11.4
12.9
17.7
8.1
12.9
17.7
8.1
12.9
17.7
9.3
16.
23.
9.
16.
23.
9.
16.
23.
                                                        1-2

-------
NATURAL  GAS RCUU
ANNUAL COST (•UU/kttl)
FUEL PRICE CAPACITY FUEL
DIFFERENTIAL FACTOR SUBSTITUTED
 . (X)

1 */wenj


1 S/MSTU


1 S/MQTU


2 S/WUTU


2 S/WttTU


2 S/MttTU


3 S/IW8TU


3 S/IMTU


3 S/i««TU

5
30 15
25
5
50 15
25
5
70 15
25
5
30 15
23
5
50 15
25
5
70 15
23
5
30 15
25
5
50 15
25
5
70 15
25
BOILER GENERATING CAPACITY (MO
100 300 500 700 1000
1.
•< ill/Ml 3.
5.
1.
•llU/Mft 3.
5-*
1.4
•UK/Ml 3.3
5.2
3
•ItU/Wh 7.2
11.3
2.6
•ills/Mi 6.6
10.6
2.4
•UU/kWh 6.4
10.3
4.1
•UU/Mi 10.4
16.7
3.6
•UU/Uh 9.7
15.9
3.4
•tlls/bft 9.3
15.5
1.5
3.5
5.5
1.3
3.2
3.1
1.2
3.1
3
2.6
6.7
10.9
2.3
6.3
10.4
2.2
6.2
10.1
3.6
10
16.3
3.4
9.
13.
3.
9.
13.
1.3
3.3
5.3
1.2
3.1
5
1.1
3
4.9
2.4
6.6
10.7
2.2
6.2
10.3
2.1
6.1
10.1
3.5
9.8
16.1
3.3
9.4
13.5
3.2
9.2
15.2
1.2
3.2
5.2
1.1
3
5
1.1
3
4.9
2.3
6.3
10.6
2.2
6.2
10.2
2.1
6.1
10
3.4
9.7
16
3.2
9.3
13.4
3.1
9.2
15.2
1.2
3.2
5.1
1.1
3
4.9
1
2.9
4.8
2.2
6.4
10.3
2.1
6.1
10.2
2.1
6
10
3.3
9.6
15.9
3.2
9.3
15.4
3.1
9.1
13.2
1300
1.1
3.1
5.1
1
3
4.9
1
2.9
4.8
2.2
6.3
10.5
2.1
6.1
10.1
2
6
10
3.3
9.6
15.9
3.1
9.2
15.4
3.1
9.1
13.1
                                                         1-3

-------
BATUMI CM UKM
WK COST EFFECTIVOMI (t/T(M)
nti pticc CAPACITY net
OirrUCMTIAL PACTOt SUBSTITUTED
(X) («)
1 S/IMTU
1 S/IMTU
1 S/IMTU
2 S/IMTU
2 S/IMTU
2 S/IMTU
3 S/IMTU
3 S/IMTU
3 «/>wrj
5
30 15 S/TOH
25
5
50 11 S/TOH
25
5
70 19 S/TOH
25
5
30 15 S/TOH
25
5
50 15 S/TOH
25
S
70 15 S/TOH
25
5
30 15 S/TW
25
5
50 15 S/TOH
25
5
70 15 S/TW
25
80ILO GEJtEIATIK CAPACITY
100 300 500 700
747.
1510.
2274.
sav.
1329.
2069.
521.
1251.
itei.
1161.
2754.
4346.
991.
2534.
4077.
570
1333.8
2097.6
483
1223.3
1963.5
445.7
1175.9
1906
984.4
2577.1
4169.5
884.6
2428
3971.4
917.9 841.8
2440.1 2364.1
3962.4 3886.5
1576.4 1398.8
3997.2 3820.2
6418.8 6241.3
1392.7 1286.2
3739.1 3632.7
6085.8 5979.3
1314 1237.9
3628.4 3352.4
5942.7 5866.9
510.7
1274.3
2038.4
447.4
1187.7
1927.9
420.2
1150.4
1880.6
925
2517.7
4110.3
849
2392.4
3935.9
816.3
Z338.7
3861
1339.4
3760.8
6182.3
1250.6
3597.1
5943.7
1212.4
3327
5844.5
477.8
1241.6
2005.4
427.6
1167.9
1908.1
406.2
1136.3
1866.5
892.1
2404.8
4077.4
829.2
2372.6
3916.1
802.2
2324.6
3846.9
1306.5
3727.9
6149.3
1230.8
3577.4
5924
1198.3
3512.9
5827.4
(NO
1000
447.4
1211.2
1975.1
409.4
1149.6
1890
393.1
1123.3
1853.5
861.7
2454.4
4047
811
2354.4
3697.9
789.2
2311.6
3833.9
1276.1
3697.5
6119
1212.5
3559.2
5905.8
1185.3
3499.8
5814.4
1300
427.
1191.
1953.
397.
1137.
1878.1
384.7
1114.8
1845
842
2434.6
4027.3
799.1
2342.5
3886
780.7
2303.1
3825.5
1256.4
3677.8
6099.2
1200.7
3547.3
5893.9
1176.8
3491.4
5805.9
                                                     1-4

-------