xvEPA
           United States
           Environmental Protection
           Agency
           Office of Air Quality
           Planning and Standards
           Research Triangle Park NC 27711
EPA-450/3-82-006a
March 1982
           Air
Fossil Fuel  Fired   _   Draft
Industrial Boilers-     EIS
Background Information
Volume 1:  Chapters 1-9

-------
                             EPA-450/3-82-006a
      Fossil Fuel Fired
     Industrial  Boilers-
 Background Information
Volume 1:  Chapters  1-9
    Emission Standards and Engineering Division
    U.S. ENVIRONMENTAL PROTECTION AGENCY
        Office of Air, Noise and Radiation
     Office of Air Quality Planning and Standards
    Research Triangle Park, North Carolina 27711

             March 1982

-------
This report has been reviewed by the Emission Standards and Engineering Division
of the Office of Air Quality Planning and Standards, EPA, and approved for publication.
Mention of trade names or commercial products is not intended to constitute endorsement
or recommendation for use.  Copies of this report are available through the Library
Services Office (MD-35), U.S. Environmental Protection Agency, Research Triangle
Park, N.C.  27711, or from National Technical Information Services, 5285 Port Royal
Road, Springfield, Virginia 22161.
                         For sale by Superintendent of Documents
                            U.S. Government Printing Office
                               Washington, DC 20402

-------
                              TABLE OF CONTENTS


                                                                      Page

1.0  OVERVIEW	    1-1

2.0  INTRODUCTION	    2-1

     2.1  Background and Authority Standards  	   2-1
     2.2  Selection of Categories  of Stationary Sources  	    2-5
     2.3  Procedure for Development of Standards of
            Performance	    2-7
     2.4  Consideration of Costs	2-9
     2.5  Consideration of Environmental  Impacts 	   2-10
     2.6  Impact on Existing Sources 	   2-11
     2.7  Revision of Standards of Performance 	   2-11

3.0  CHARACTERISTICS OF THE INDUSTRIAL BOILER SOURCE CATEGORY  ...    3-1

     3.1  GENERAL	    3-1

          3.1.1  Industrial Boiler Source Category 	   3-1
          3.1.2  Classification of Industrial  Boilers 	    3-1
          3.1.3  Population of Industrial Boilers 	    3-5
          3.1.4  Fuel Usage Patterns	3-15

     3.2  INDUSTRIAL BOILERS AND  THEIR EMISSIONS 	   3-19

          3.2.1  Uncontrolled Emissions Overview 	   3-19
          3.2.2  Coal-Fired Boilers 	    3-24
          3.2.3  Oil-Fired Boilers 	   3-39
          3.2.4  Natural Gas-Fired Boilers 	   3-49

     3.3  EMISSIONS UNDER CURRENT REGULATIONS 	    3-56

          3.3.1  Existing Regulations 	    3-56
          3.3.2  State Emission Limits 	   3-56

     3.4  REFERENCES	3-69

4.0  EMISSION CONTROL TECHNIQUES  	   4-1

     4.1  POST COMBUSTION CONTROL TECHNIQUES  FOR
            PARTICULATE MATTER 	   4-3

          4.1.1  Electrostatic Precipitators  	   4-3
          4.1.2  Fabric Filters	    4-14
          4.1.3  Wet Scrubbers	4-21

-------
          4.1.4  Multitube Cyclones 	    4-27
          4.1.5  Side Stream Separator	4-39
          4.1.6  Emission Data	4-42

     4.2  POST COMBUSTION TECHNIQUES FOR S02 CONTROL	4-60

          4.2.1  Sodium Scrubbing 	    4-65
          4.2.2  Double Alkali	4-70
          4.2.3  Lime and Limestone	    4-78
          4.2.4  Dry Scrubbing	4-85
          4.2.5  Emission Reduction Data	4-90

     4.3  COMBUSTION MODIFICATION TECHNIQUES FOR NITROGEN
            OXIDE CONTROL	    4-99

          4.3.1  Low Excess Air	    4-101
          4.3.2  Staged Combustion 	   4-106
          4.3.3  Flue Gas Recirculation	    4-112
          4.3.4  Low NO  Burners	4-114
          4.3.5  No Combustion Air Preheat or Reduced Air Preheat .  .  4-117
          4.3.6  Ammonia Injection 	   4-117
          4.3.7  NO  Emission Reduction Data	4-119
                   A
     4.4  POST COMBUSTION TECHNIQUES FOR N0y CONTROL	4-148
                                           A

          4.4.1  Selective Catalytic Reduction 	   4-149
          4.4.2  Wet Scrubbing	4-156
          4.4.3  Electron Beam Irradiation	4-157

     4.5  PRE-COMBUSTION TECHNIQUES FOR PM, NO. AND SO, CONTROL .  .   4-160
                                              A        £

          4.5.1  Naturally-Occurring Clean Fuels 	   4-161
          4.5.2  Physical Coal Cleaning	    4-163
          4.5.3  Oil Cleaning	    4-169
          4.5.4  Low-Btu Gasification 	    4-175
          4.5.5  Solvent Refined Coal	.  .  . .    4-181
          4.5.6  Oil/Water Emulsions 	   4-185

     4.6  COAL/ALKALI COMBUSTION TECHNIQUES FOR S02 CONTROL  ....    4-187

          4.6.1  Fluidized Bed Combustion for S0? and NO  Control .    4-188
          4.6.2  Coal/Limestone Pellets ......... 	    4-204

     4.7  REFERENCES	4-214

5.0  MODIFICATION AND RECONSTRUCTION 	   5-1

     5.1  SUMMARY OF MODIFICATION AND RECONSTRUCTION PROVISIONS  . .    5-1
                                     iv

-------
          5.1.1  Modification	    5-1
          5.1.2  Reconstruction	    5-2
     5.2  APPLICABILITY OF MODIFICATION  AND  RECONSTRUCTION
            PROVISIONS TO FOSSIL FUEL-FIRED  INDUSTRIAL  BOILERS  ...   5-3
          5.2.1  Modification	    5-3
          5.2.2  Reconstruction	    5-4
          5.2.3  Summary	5-5
     5.3  REFERENCES	5-6
6.0  MODEL BOILERS AND CONTROL ALTERNATIVES  	    6-1
     6.1  SELECTION OF STANDARD  BOILERS  	    6-3
          6.1.1  Capacities and  Fuel  Type	    6-3
          6.1.2  Standard Boiler Configurations  	    6-5
          6.1.3  Standard Boiler Specifications  	    6-8
     6.2  SELECTION OF CONTROL ALTERNATIVES  	    6-18
          6.2.1  Baseline Alternative 	    6-18
          6.2.2  Other Alternatives  	    6-21
     6.3  SUMMARY OF CONTROL SYSTEMS  AND MODEL BOILERS  	   6-21
     6.4  REFERENCES	6-26
7.0  ENVIRONMENTAL AND ENERGY IMPACTS 	  7-1
     7.1  AIR POLLUTION IMPACTS	7-1
          7.1.1  Primary Impacts	7-4
          7.1.2  Secondary Air Impacts 	   7-18
     7.2  LIQUID WASTE IMPACTS 	   7-21
          7.2.1  Effluent Quantities  and Characteristics 	   7-22
          7.2.2  Effluent Treatment  and  Disposal  	   7-24
          7.2.3  Applicable Regulations  	    7-25
     7.3  SOLID WASTE DISPOSAL IMPACTS 	   7-25
          7.3.1  Solids Waste Quantities and Characteristics ....   7-26
          7.3.2  Waste Treatment and  Disposal 	    7-31
          7.3.3  Waste Disposal  Regulations  	    7-33
     7.4  ENERGY IMPACT OF CONTROL TECHNOLOGIES  	    7-34

-------
     7.5  OTHER IMPACTS	    7-38
     7.6  OTHER ENVIRONMENTAL CONCERNS 	     7-38
          7.6.1  Long-Term Gains/Losses 	    7-38
          7.6.2  Environmental Impact of Delayed Standard 	    7-38
     7.7  REFERENCES	7-39
8.0  COSTS	8-1
     8.1  COSTING APPROACH 	   8-2
          8.1.1  Cost Bases	    8-6
          8.1.2  Capital Costs	     8-6
          8.1.3  Operating and Maintenance Costs 	   8-6
          8.1.4  Annualized Costs 	    8-16
     8.2  ANALYSIS OF COST IMPACTS	8-16
          8.2.1  Capital Costs	8-16
          8.2.2  Annualized Costs 	    8-29
          8.2.3  Retrofit Cost Impacts	8-40
     8.3  OTHER COST CONSIDERATIONS	    8-40
     8.4  REFERENCES	8-42
9.0  ECONOMIC IMPACT	9-1
     9.1  INDUSTRY ECONOMIC PROFILES  	   9-3
          9.1.1  Major Steam Users 	   9-3
          9.1.2  Selected Industries  	   9-8
     9.2  ECONOMIC IMPACT ANALYSIS 	   9-18
          9.2.1  Regulatory Options 	    9-18
          9.2.2  Major Steam Users 	   9-19
          9.2.3  Selected Industries  	   9-30
     9.3  REFERENCES	9-100
APPENDIX A - Evolution of the Background Information Document .  .  .    A-l
APPENDIX B - Index to Environmental Considerations 	   B-l
                                     vi

-------
APPENDIX C - Emission Test Data	    C-l
APPENDIX D - Emission Measurement and Monitoring Methods  	   D-l
APPENDIX E - Emerging Technology Model  Boiler Impact Analysis  ...    E-l
                                     vn

-------
                              LIST OF ILLUSTRATIONS
Figure                                                                Page
3-1       Categories of industrial boilers	3-3
3-2       Relative distribution by capacity of the three types
          of industrial boilers	3-10
3-3       Mass and energy balances for a 58.6 MW (200 X 106 Btu/hr)
          pulverized coal-fired boiler	3-28
3-4       Mass and energy balances for a 44 MW (150 X 106 Btu/hr)
          coal -fi red spreader stoker	3-30
3-5       Mass and energy balances for a 22 MW (75 X 106 Btu/hr)
          coal-fired chaingrate stoker	3-31
3-6       Mass and energy balances for an 8.8 MW (30 X 106 Btu/hr)
          underfeed stoker	3-33
3-7       Effect of excess oxygen on NO  emissions from coal-fired
          boilers	3-40
3-8       Mass and energy balances for a 44 MW (150 X 106 Btu/hr)
          residual oil-fired watertube boiler	3-43
3-9       Cutaway view of a four-pass Scotch firetube boiler	3-45
3-10      Effect of fuel oil carbon residue on base load particulate
          emi ssions	3-48
3-11      Effect of excess oxygen on NO  emissions from distillate
          and residual oil-fired  boilers	3-50
3-12      Effect of excess oxygen on NO  emissions from natural
          gas-f i red boil ers	3-55
4.1-1     Typical precipitator cross section	4-5
4.1-2     Relationship between collection efficiency and SCA for
          various coal sulfur contents	4-8
4.1-3     Precipitation rate versus particle resistivity	4-11
4.1-4     Precipitation rate versus coal sulfur percent	4-11
4.1-5     Variation of fly ash resistivity with temperature for
          coal of various sulfur contents	4-12
4.1-6     Variation of resistivity with sodium content for fly  ash
          from power plants burning western coals	4-12
4.1-7     Fly ash resistivity versus coal  sulfur content for several
          flue gas temperature bands	4-12
4.1-8     Measured fractional  efficiencies for a coldside ESP with
          operating parameters as indicated, installed on a
          pulverized coal-fired boiler burning low sulfur coal	4-13
                                     vm

-------
Figure
4.1-9     Isometric view of a  pulse-jet fabric  filter	4-15
4.1-10    Variable-throat venturi  scrubber	4-22
4.1-11    Venturi scrubber comparative  fractional efficiency curves...4-26
4.1-12    Aerodynamic cut diameter versus gas pressure drop with
          liquid-to-gas ratio  (L/G) as  a parameter	4-28
4.1-13    Schematic of a multiple  cyclone and detail of an
          individual  tube	4-30
4.1-14    Variation of a single  cyclone collection efficiency with
          gas velocity	4-32
4.1-15    Typical overall collection  efficiency of axial-entry
          cyclones	4-33
4.1-16    Mechanical  collector efficiency versus  boiler load
          (spreader stoker boilers)	4-36
4.1-17    Uncontrolled particulate emissions vs.  boiler load	4-37
4.1-18    Controlled  particulate emissions  vs.  boiler load	4-38
4.1-19    Side stream separator	4-40
4.1-20    Electrostatic precipitator  emission data	4-44
4.1-21    Fabric filter emission data	4-48
4.1-22    Emission data for wet  scrubbers	4-50
4.1-23    Single and dual mechanical  collector  emission data	4-53
4.1-24    Mechanical  collector emission data for  mass fed  stokers
          without fly ash reinjection	4-54
4.1-25    Side stream separator  emission data	4-56
4.2-1     Simplified  flow diagram  of  a  sodium scrubbing system	4-67
4.2-2     Simplified  flow diagram  for a sodium/lime  double-alkali
          process	4-71
4.2-3     S0« removal versus L/G ratio  for  the  EnviroTech/Gadsby
          Pilot Plant with a single stage  polysphere absorber	4-76
4.2-4     S02 removal versus scrubber effluent  pH for the
          EnviroTech/Gadsby Pilot  Plant with a  two-stage absorber	4-77
4.2-5     Process flow diagram for a  typical lime or limestone wet
          scrubbing system	4-80
4.2-6     Typical spray dryer/particulate  collection process flow
          di agram	4-86
4.2-7     Daily average S02 removal,  boiler load, slurry pH for
          the sodium scrubbing process  at  Location  1	4-93
4.2-8     Daily average S02 removal,  boiler load, and  slurry pH for
          the dual alkali scrubbing process at  Boiler No.  1,
          Location III	4-94
                                     ix

-------
Figure                                                                Page

4.2-9     Daily average S02 removal, boiler load, and slurry pH
          for scrubbing process at Boiler No. 3 Location III	4-95

4.2-10    Daily average S0« removal, boiler load, and slurry pH for
          lime slurry scruobing process at Location IV	4-96

4.2-11    Daily average SCL removal, boiler load, adipic acid
          concentration, and slurry pH for limestone system at
          Location IV	4-97

4.2-12    Daily average SCL removal, inlet SCL for lime spray system
          at Location VI	4-100

4.3-1     Schematics of two single burner units for packaged boilers
          showing location and control of combustion airflow vanes....4-103

4.3-2     Schematic of industrial boiler watertube boilers equipped
          with (A) OFA parts and (B) SFA parts	4-108

4.3-3     Schematics for FGR systems for industrial boilers	4-113

4.3-4     FGR test results on a 5.1 MW (17.5 X 106 Btu/hr) packaged
          watertube boiler	4-115

4.3-5     Continuous monitoring data for LEA/OFA combustion modi-
          fication on a pulverized coal-fired boiler (Month #1)	4-121

4.3-6     Continuous monitoring data for LEA/OFA combustion modi-
          fication on a pulverized coal fired boiler (Month #2)	4-122

4.3-7     Continuous monitoring data for LEA/OFA combustion modi-
          fication on a pulverized coal fired boiler (Month #3)	4-123

4.3-8     Continuous monitoring data for LEA/OFA combustion modi-
          fication on a pulverized coal fired boiler (Month #4)	4-124

4.3-9     Continuous monitoring data for LEA/OFA combustion modi-
          fication on a pulverized coal fired boiler (Month #5)	4-125

4.3-10    Continuous monitoring data for LEA/OFA combustion modi-
          fication on a pulverized coal fired boiler (Month #6)	4-126

4.3-11    Fuel NO  formation as a function of the coal oxygen to
          nitrogeft ratio and the coal nitrogen content	4-128

4.3-12    Continuous monitoring data for LEA combustion modification
          on a spreader stoker coal-fired boiler at Location II	4-129

4.3-13    Continuous monitoring data (8 hour average) for LEA com-
          bustion modification on a spreader stoker coal-fired
          boiler at Location III	4-130

4.3-14    NO  emissions vs. excess 0« - short-term data for coal-
          fifed spreader stokers, (unstaged combustion)	4-132

4.3-15    Short-term emission data for staged combustion in a
          spreader stoker boil er	4-134

-------
Figure
4.3-16    Short-term emission data for two mass fed  stokers.
          (unstaged and staged combustion)	4-135
4.3-17    Continuous monitoring data for LEA/SCA combustion modi-
          fication on a residual  oil-fired boiler at Location  IV	4-136
4.3-18    Predicted NO  emissions from residual  oil  units  as a
          function of Tuel  nitrogen content - LEA controls	4-139
4.3-19    Predicted NO  emissions from residual  oil-fired  boilers  vs.
          fuel  nitrogen - LEA and staged combustion  controls	4-140
4.3-20    Short-term emission data for distillate oil-fired boilers...4-142
4.3-21    Short-term data for a distillate oil-fired boiler without
          air preheat	4-143
4.3-22    Continuous NO  emission data for a small natural gas-fired
          boiler at Locltion V	4-145
4.3-23    Short-term emission data for natural  gas-fired boilers.
          (unstaged combustion)	4-146
4.3-24    Short-term emission data for a natural  gas-fired boiler	4-147
4.4-1     Typical industrial  boiler SCR system	4-150
4.4-2     Flow diagram of a simultaneous NOY SO  SCR system	4-152
                                           A   rt
4.4-3     Generalized flow diagram for wet NO /SO process	4-158
                                             A   rt
4.4-4     Process flow diagram for the Ebara-JAERI electron beam
          process	4-159
4.5-1     Physical coal cleaning unit operations employed  to achieve
          various levels of cleaning	4-165
4.5-2     Basic HDS process	4-172
4.5-3     Hydrogen consumption in desulfurization of residual  oil	4-174
4.5-4     Effect of metals content on catalyst consumption	4-176
4.5-5     Low-Btu gasification system process and pollution control
          modules	4-178
4.5-6     Flow diagram of the SRC-1 and SRC-II liquefaction
          processes	4-182
4.6-1     Typical industrial  FBC boiler	4-189
4.6-2     Projected desulfurization performance of FBC based  upon a
          model developed by Westinghouse	4-194
4.6-3     Ca/S molar feed required to maintain 90 percent sulfur
          removal in AFBC, as projected by the Westinghouse model	4-196
4.6-4     SOg reduction as a function of bed temperature	4-197
                                      xi

-------
Figure                                                                Page
4.6-5     Summary of data obtained at eight different AFBC
          test facilities	4-200
4.6-6     Composite NO  emissions diagram for FBC units operating
          wi thin normal ranges	4-202
6-1       Logic leading to selection of model boilers	6-2
7-1       Solid waste production  (fly ash and sludge)	7-30
8.2-1     Capital costs of control alternatives applied to HSC-fired
          model boilers	8-22
8.2-2     Capital costs of control alternatives applied to LSC-fired
          model boil ers	8-23
8.2-3     Capital costs of control alternatives applied to residual
          oil-fired model boilers	8-28
8.2-4     Annualized cost of control alternatives applied to
          HSC-fired model boilers	8-33
8.2-5     Annualized costs of control alternatives  applied to
          LSC-fired model boilers	8-34
8.2-6     Annualized costs of control alternatives  applied to
          residual oil-fired model boilers	8-38
9-1       Change  in product  price due to regulatory option	9-24
                                      xii

-------
                              LIST OF  TABLES

Table                                                                 Page
3-1       Manufacturers of Industrial  Firetube  and Watertube
          Boil ers	3-6
3-2       Manufacturers of Industrial  Size  Cast Iron  Boilers	3-9

3-3       Boiler Population Distribution by Heat-Transfer
          Configuration	3-9
3-4       Installed Capacity of U.S. Watertube  Industrial  Boilers
          by Unit Size and Fuel  Type	3-12

3-5       Installed Capacity of Industrial  Firetube Boilers by
          Size and Fuel Type	3-13

3-6       Installed Capacity of Industrial  Cast Iron  Boilers  by
          Fuel  Type	3-14

3-7       Regional  Fossil  Fuel  Consumption  by the Manufacturing
          Industries in 1976	3-16

3-8       Distribution of Industrial Boiler Fossil  Fuel  Consumption
          in 1974 by Industry and Fuel  Type	3-17

3-9       United States Primary Energy Consumption by Consuming
          Sector and Energy Source, 1974	3-18

3-10      Ultimate Analysis of Coal  Selected for the  Representative
          Boil ers	3-26
3-11      Uncontrolled Emission Factors for Coal-Fired Watertube
          Industrial  Boilers	3-35

3-12      Uncontrolled Emission Factors for Coal-Fired Underfeed
          Stoker Firetube Industrial Boilers	3-36

3-13      Particle Size Data for Particulate Emission from Typical
          Uncontrolled Coal-Fired Industrial  Boilers	3-38

3-14      Ultimate Analysis of Residual and Distillate Oil Selected
          for Representative Boil ers	3-42
3-15      Uncontrolled Emissions Factors for Oil-Fired Industrial
          Boilers	3-47

3-16      Ultimate Analysis of Natural  Gas  Selected for the
          Representative Boiler	3-51

3-17      Mass and Energy Balance for  a Natural  Gas-Fired  Firetube
          Boil er	3-52

3-18      Uncontrolled Emission Factors for Natural Gas-Fired
          Watertube and Firetube Industrial Boilers	3-54

3-19      Subpart D Emission Limits for Fossil  Fuel-Fired  Steam
          Generators	3-57
                                    xm

-------
                                                                      Page
3-20      State Participate Regulations	3-59
3-21      State S02 Regulations for Coal-Fired Boilers	3-62
3-22      State S02 Regulations for Oil- and Gas-Fired Boilers	3-65
3-23      State NOX Emission Limits that Differ from Subpart D	3-68
4-1       ITAR Report List	4-2
4.1-1     Baghouse Installations on Industrial Boilers in the U.S	4-17
4.1-2     Summary of Particulate Emission Data for ESPs on Oil-Fired
          Boil ers	4-46
4.1-3     Data on Particulate Scrubbing Systems Installed on Coal-
          Fired Utility Steam Generators	4-52
4.1-4     Fine Particulate Control Efficiency for Various PM
          Control Devi ces	4-59
4.1-5     Opacity Transmissometer Data	4-61
4.1-6     Opacity - EPA Reference Method 9	4-63
4.2-1     Performance Data for Operating Sodium Scrubbing Systems	4-68
4.2-2     Summary of Operating and Planned Industrial  Boiler
          Double Alkali Systems	4-74
4.2-3     Summary of Operating Lime and Limestone Systems for U.S.
          Industrial Boilers as of March 1978	4-81
4.2-4     Summary of Industrial Boiler Spray Drying Systems	4-88
4.2-5     Summary of Continuous S0? Emission Data at Five Industrial
          Boiler Wet FGD Systems...	4-92
4.3-1     Safe Operating Levels for LEA	4-105
4.3-2     Development Status of Staged Combustion Application to
          Industrial Boilers	4-109
4.4-1     Catalyst Design Variables for Various Catalyst Shapes	4-155
4.5-1     Physical Coal Cleaning Plants Categorized by States
          for 1976	4-167
4.5-2     Chemistry of Hydrodesulfurization Reactions in
          Petroleum Crude Oil	4-171
4.5-3     General Comparison and Relative Technical Status of the
          SRC-I and SRC-II Liquefaction Processes	4-183
4.6-1     Results of Particulate Emission Testing at the Georgetown
          FBC Unit	4-203
4.6-2     Comparison of Physical  Properties of Raw Coal and
          Fuel Pel let	4-205
                                     xiv

-------
Table
4.6-3     Model  Spreader Experiments	4-210
4.6-4     Emission Data Summary for Fuel  Pellet Demonstration	4-211
4.6-5     Sulfur Balance	4-211
6-1       Standard Boilers Selected for Evaluation	6-4
6-2       Representative Standard Boiler Capacities	6-6
6-3       Specifications for Natural  Gas-Fired Standard  Boilers	6-9
6-4       Specifications for Distillate Oil-Fired  Standard  Boilers	6-10
6-5       Specifications for Residual  Oil-Fired Standard Boilers	6-11
6-6       Specifications for High-Sulfur Coal-Fired Standard
          Boilers	6-12
6-7       Specifications for Low-Sulfur Coal-Fired Standard
          Boil ers	6-13
6-8       Ultimate Analyses of Fuels Selected for  the Model Boiler
          Anal ys i s	6-14
6-9       Typical  Excess Air Requirements for Industrial  Boilers	6-17
6-10      Control  Alternatives Selected for Evaluation	6-19
6-11      Coal-Fired Model Boilers	6-23
6-12      Oil- and Gas-Fired Model  Boilers	6-24
6-13      Abbreviations for Control Systems	6-25
7-1       Summary of Emission Factors for Coal-Fired Model
          Boil ers	7-2
7-2       Summary of Emission Factors for Oil- and Gas-Fired  Model
          Boil ers	7-3
7-3       Coal-Fired Model Boiler Annual  Emissions	7-5
7-4       Oil- and Gas-Fired Model  Boiler Annual  Emissions	7-6
7-5       Coal-Fired Model Boiler Annual  Emission  Reductions  over
          Basel i ne	7-8
7-6       Oil- and Gas-Fired Model  Boiler Annual  Emission
          Reducti ons over Basel i ne	7-9
7-7       Coal-Fired Model Boiler Percentage Emission Reductions
          over Basel i ne	7-10
7-8       Oil- and Gas-Fired Model  Boiler Percentage Emission
          Reductions over Baseline	7-11
7-9       Model  Boiler Dispersion Modeling Results	7-15
7-10      Secondary Air Pollution Impacts for Coal-Fired Model
          Bo il ers	7-19
                                      xv

-------
Table                                                                 Page
7-11      Secondary Air Pollution Impacts for Gas- and Oil-Fired
          Model Boilers	7-20
7-12      Water Pollution Impacts for the Sodium Throwaway System	7-23
7-13      Solid Waste Impacts from Coal-Fired Model  Boilers	7-28
7-14      Solid Waste Impacts from Oil-Fired Model Boilers	7-29
7-15      Model Boiler Energy Requirements	7-35
8.1-1     Coal-Fired Model Boilers	8-3
8.1-2     Oil- and Gas-Fired Model Boilers	8-4
8.1-3     Abbreviations for Control Methods	8-5
8.1-4     Summary of Sources of Costing Information	8-7
8.1-5     Emission Control System General Design Specifications	8-8
8.1-6     Capital Cost Components	8-11
8.1-7     Operating and Maintenance Cost Components	8-12
8.1-8     Load Factors and Utility and Unit Operating Costs	8-14
8.1-9     Fuel Prices	8-15
8.1-10    Annual ized Cost Components	8-17
8.2-1     Capital Costs of HSC Model Boilers	8-18
8.2-2     Capital Costs of LSC Model Boilers	8-19
8.2-3     Capital Costs of Oil- and Gas-Fired Model  Boilers	8-20
8.2-4     Annual ized Costs of HSC Model Boilers	8-30
8.2-5     Annualized Costs of LSC Model Boilers	8-31
8.2-6     Annualized Costs of Oil- and Gas-Fired Model Boilers	8-32
9-1       Fossil Fuel Consumption Characteristics of the Major
          Steam Users	9-5
9-2       Characteristics of the Major Steam Users	9-6
9-3       Industrial Production Growth Rate Projections	9-7
9-4       Base Case Air Emission Regulations	9-20
9-5       Regulatory Option 1	9-21
9-6       Regulatory Option V	9-22
9-7       1976 Steam Consumption Per Dollar Product	9-26
9-8       Percent of New Steam in Total Steam Consumption	9-27
                                    xvi

-------
Table
9-9       Nationwide Average  Annualized  Costs for New  Industrial
          Boilers by Industry -  1985	9-28
9-10      Nationwide Average  Annualized  Costs for New  Industrial
          Boilers by Industry -  1990	9-29
9-11      Change in Product Price From the  Base Case	9-32
9-12      Summary of Change in Product Cost and Return on Assets
          for Model  Plants in Selected Industries - 1990	9-33
9-13      Model  Firm and Plant Configuration: Beet Sugar Industry	9-38
9-14      Financial  Analysis: Beet Sugar Industry	9-40
9-15      1990 Boiler Costs:  Beet Sugar  Model Plant	9-41
9-16      Change in Product Cost:  Beet Sugar Model Plant	9-44
9-17      Change in Profit Margin Due to New Boiler Investment:
          Beet Sugar Model  Plant	9-45
9-18      Capital Availability:  Beet Sugar  Model Firm	9-46
9-19      Model  Firm and Plant Configuration: Fruit and Vegetable
          Canning Industry	9-50
9-20      Financial  Analysis: Fruit and  Vegetable Canning
          Industry	9-51
9-21      1990 Boiler Costs:  Fruit and Vegetable Canning Model
          Plant	9-52
9-22      Change in Product Cost:  Fruit  and Vegetable  Canning
          Model  Plant	9-53
9-23      Change in Profit Margin Due to New Boiler Investment:
          Fruit and Vegetable Canning Model  Plant	9-54
9-24      Capital Availability:  Fruit and Vegetable Canning Model
          Firm	9-55
9-25      Model  Firm and Plant Configuration: Rubber Reclaiming
          Industry	9-58
9-26      Financial  Analysis: Rubber Reclaiming Industry	9-59
9-27      1990 Boiler Costs:  Rubber Reclaiming Model Plant	9-60
9-28      Change in Product Cost: Rubber Reclaiming Model Plant	9-61
9-29      Change in Profit Margin Due to New Boiler Investment:
          Rubber Reclaiming Model  Plant	9-64
9-30      Capital Availability:  Rubber Reclaiming Model Firm	9-65
9-31      Model  Firm and Plant Configuration: Automobile Manu-
          facturing Industry	9-66
9-32      Financial  Analysis: Automobile Manufacturing Industry	9-67
                                    xvii

-------
Table                                                                 Page
9-33      1990 Boiler Costs: Automobile Manufacturing Model
          Plant	9-70
9-34      Change in Product Cost: Automobile Manufacturing
          Model PI ant	9-71
9-35      Change in Profit Margin Due to New Boiler Investment:
          Automobile Manufacturing Model Plant	9-72
9-36      Capital Availability: Automobile Manufacturing Model
          Firm	9-73
9-37      Model Firm and Plant Configuration Petroleum Refining
          Industry	9-74
9-38      Financial Analysis: Petroleum Refining Industry	9-76
9-39      1990 Boiler Costs: Petroleum Refining Model Plant	9-78
9-40      Change in Product Cost: Petroleum Refining Model Plant	9-79
9-41      Change in Profit Margin Due to New Boiler Investment:
          Petroleum Refining Model Plant	9-82
9-42      Capital Availability: Petroleum Refining Model Firm	9-83
9-43      Model Firm and Plant Configuration: Iron and Steel
          Manufacturing Industry	9-84
9-44      Financial Analysis: Iron and Steel Manufacturing Industry.. .9-85
9-45      1990 Boiler Costs: Iron and Steel Manufacturing Plant	9-86
9-46      Change in Product Cost: Iron and Steel Manufacturing
          Model Plant	9-87
9-47      Change in Profit Margin Due to New Boiler Investment:
          Iron and Steel Manufacturing Model Plant	9-90
9-48      Capital Availability: Iron and Steel Manufacturing
          Model Fi rm	9-91
9-49      Model Firm and Plant Configuration: Liquor Distilling
          I ndus try	9-92
9-50      Financial Analysis: Liquor Distilling Industry	9-93
9-51      1990 Boiler Costs: Liquor Distil ling Model Plant	9-95
9-52      Change in Product Cost: Liquor Distilling Model Plant	9-96
9-53      Change in Profit Margin Due to New Boiler Investment:
          Liquor Distilling Model Plant	9-97
9-54      Capital Availability: Liquor Distilling Model Firm	9-99
                                   xvm

-------
                               1.0  OVERVIEW

     This document was prepared to provide the public and industry with
background information on the industrial  boiler source category in
support of potential new source performance standards.  Fossil  fuels
discussed and analyzed include coal, oil, and natural gas.   Background
information for nonfossil fuel fired boilers (wood, solid waste, bagasse
and fossil/nonfossil mixtures) is included in a separate document
EPA 450/3-82-007.
     This document contains information on the use of industrial boilers
in different industries and an assessment of controlled and uncontrolled
emissions from different configurations of boilers firing fossil fuels.
Cost and environmental assessments for several  model  boiler configurations
to meet alternative control  levels are also presented.
                                    1-1

-------
                              2.0  INTRODUCTION

2.1  BACKGROUND AND AUTHORITY FOR STANDARDS
     Before standards of performance are proposed as  a  Federal  regulation,
air pollution control methods available to the affected Industry  and  the
associated costs of Installing and maintaining the control  equipment  are
examined 1n detail.  Various levels of control based  on different technolo-
gies and degrees of efficiency are expressed as control alternatives.   Each
of these alternatives is studied by EPA as a prospective basis  for a
standard.  The alternatives are investigated in terms of their  impacts  on
the economics and well-being of the industry, the impacts on  the  national
economy, and the impacts on the environment.  This document summarizes  the
information obtained through these studies so that interested persons will
be able to see the Information considered by EPA in the development of  the
proposed standard.
     Standards of performance for new stationary sources are  established
under Section 111 of the Clean A1r Act (42 U.S.C. 7411) as amended, herein-
after referred to as the Act.  Section 111 directs the  Administrator  to
establish standards of performance for any category of  new stationary source
of air pollution which ". . . causes, or contributes  significantly to air
pollution which may reasonably be anticipated to endanger public  health or
welfare."
     The Act requires that standards of performance for stationary sources
reflect "... the degree of emission reduction achievable which  (taking
Into consideration the cost of achieving such emission  reduction, and any
nonalr quality health and environmental impact and energy requirements) the
Administrator determines has been adequately demonstrated for that category
of sources."  The standards apply only to stationary sources, the construc-
tion or modification of which commences after regulations are proposed  by
publication 1n the Federal Register.
                                     2-1

-------
     The 1977 amendments to the Act altered or added numerous provisions
that apply to the process of establishing standards of performance.
     1.  EPA is required to list the categories of major stationary sources
that have not already been listed and  regulated under standards of perform-
ance.  Regulations must be promulgated for these new categories on the
following schedule:
     a.  25 percent of the listed categories by August 7, 1980.
     b.  75 percent of the listed categories by August 7, 1981.
     c.  100 percent of the listed categories by August 7, 1982.
A governor of a State may apply to the Administrator to add a category not
on the list or may apply to the Administrator to have a standard of perform-
ance revised.
     2.  EPA is required to review the standards of performance every
4 years and, if appropriate, revise them.
     3.  EPA is authorized to  promulgate  a standard based on design, equip-
ment, work practice, or operational procedures when a standard based on
emission levels is not feasible.
     4.  The term  "standards of performance" is redefined, and a new term
"technological system of continuous emission reduction" is defined. The new
definitions clarify that the control  system must be continuous and may
include a low- or  non-polluting process or operation.
     5.  The time  between the  proposal and promulgation of a standard under
Section 111 of the Act may be  extended to 6 months.
     Standards of  performance, by themselves, do not guarantee protection of
health or welfare  because they are not designed to achieve any specific air
quality levels.  Rather, they  are designed to reflect the degree of emission
limitation achievable through  application of the best adequately
demonstrated technological system of  continuous emission reduction, taking
into consideration the cost of achieving  such emission reduction, any
non-air-quality health and environmental  impacts, and energy requirements.
     Congress had  several reasons for including these requirements. First,
standards with a degree of uniformity are needed to avoid situations where
some States may attract industries by relaxing standards relative to other
                                      2-2

-------
States.  Second, stringent standards enhance the potential  for long-term
growth.  Third, stringent standards may help achieve long-term cost  savings
by avoiding the need for more expensive retrofitting when  pollution  ceilings
may be reduced in the future.  Fourth, certain types of standards  for  coal-
burning sources can adversely affect the coal  market by driving up the price
of low-sulfur coal or effectively excluding certain coals  from the reserve
base because their untreated pollution potentials are high.   Congress  does
not intend that new source performance standards contribute  to these
problems.
     Promulgation of standards of performance does not prevent State or
local agencies from adopting more stringent emission limitations for the
same sources.  States and local agencies if authorized by  State law  are free
under Section 116 of the Act to establish even more stringent emission
limits than those established under Section 111 or those necessary to  attain
or maintain the National Ambient Air Quality Standards (NAAQS) under
Section 110.  Thus, new sources may in some cases be subject to limitations
more stringent than standards of performance under section 111, and
prospective owners and operators of new sources should be  aware of this
possibility in planning for such facilities.
     A similar situation may arise when a major emitting facility  is to be
constructed in a geographic area that falls under the prevention of  signif-
icant deterioration of air quality provisions of Part C of the Act.  These
provisions require, among other things, that major emitting  facilities to be
constructed in such areas are to be subject to best available control
technology.  The term Best Available Control Technology (BACT), as defined
in the Act, means
     ... an emission limitation based on the maximum degree of reduction
     of each pollutant subject to regulation under this Act  emitted  from, or
     which results from, any major emitting facility, which  the permitting
     authority, on a case-by-case basis, taking into account energy,
     environmental, and economic impacts and other costs,  determines is
                                     2-3

-------
     achievable for such facility through application of production
     processes and available methods, systems, and techniques, including
     fuel cleaning or treatment or innovative fuel combustion techniques for
     control of each such pollutant.  In no event shall application of "best
     available control technology" result in emissions of any pollutants
     which will exceed the emissions allowed by any applicable standard
     established pursuant to Sections 111 or 112 of this Act.
     (Section 169(3))."
     Although standards of performance are normally structured in terms of
numerical emission limits where feasible, alternative approaches are some-
times necessary.  In some cases physical measurement of emissions from a new
source may be impractical or exorbitantly expensive.  Section lll(h)
provides that the Administrator may promulgate a design or equipment
standard in those cases where it is not feasible to prescribe or enforce a
standard of performance.  For example, emissions of hydrocarbons from
storage vessels for petroleum liquids are greatest during tank filling.  The
nature of the emissions, high concentrations for short periods during
filling and low concentrations for longer periods during storage, and the
configuration of storage tanks make direct emission measurement impractical.
Therefore, a more practical approach to standards of performance for storage
vessels has been equipment specification.
     In addition, Section lll(i) authorizes the Administrator to grant
waivers of compliance to permit a source to use innovative continuous
emission control technology.  In order to grant the waiver, the
Administrator must find:  (1) a substantial likelihood that the technology
will produce greater emission reductions than the standards require or an
equivalent reduction at lower economic energy or environmental cost; (2) the
proposed system has not been adequately demonstrated; (3) the technology
will not cause or contribute to an unreasonable risk to the public health,
welfare, or safety; (4) the governor of the State where the source is
located consents; and (5) the waiver will not prevent the attainment or
maintenance of any ambient standard.  A waiver may have conditions attached
to assure the source will not prevent attainment of any NAAQS.  Any such
                                     2-4

-------
condition will  have the force  of a  performance  standard.   Finally, waivers
have definite end dates and may be  terminated earlier if  the  conditions  are
not met or if the system fails to perform as expected.  In such  a case,  the
source may be given up to 3 years to meet the standards with  a mandatory
progress schedule.
2.2  SELECTION OF CATEGORIES OF STATIONARY SOURCES
     Section 111 of the Act directs the Administrator to  list categories of
stationary sources.  The Administrator "... shall  include a category of
sources in such list if in his judgment it causes,  or contributes signifi-
cantly to, air pollution which may  reasonably be  anticipated  to  endanger
public health or welfare."  Proposal and promulgation of  standards of
performance are to follow.
     Since passage of the Clean Air Amendments  of 1970, considerable atten-
tion has been given to the development of a system  for assigning priorities
to various source categories.   The  approach specifies areas of interest  by
considering the broad strategy of the Agency for  implementing the Clean  Air
Act.  Often, these "areas" are actually pollutants  emitted by stationary
sources.  Source categories that emit these pollutants are evaluated and
ranked by a process involving  such  factors as  (1) the level of emission
control (if any) already required by State regulations,  (2) estimated  levels
of control that might be required from standards  of performance  for the
source category, (3) projections of growth and  replacement of existing
facilities for the source category, and (4) the estimated incremental  amount
of air pollution that could be prevented in a  preselected future year  by
standards of performance for the source category.  Sources for which new
source performance standards were promulgated  or  under development  during
1977, or earlier, were selected on  these criteria.
     The Act amendments of August 1977 establish  specific criteria  to  be
used in determining priorities for  all major source categories  not  yet
listed by EPA.  These are (1)  the quantity of  air pollutant emissions  that
each such category will emit,  or will be designed to emit; (2)  the  extent to
which each such pollutant may  reasonably be anticipated to endanger public
                                     2-5

-------
health or welfare; and (3) the mobility and competitive nature of each such
category of sources and the consequent need for nationally applicable new
source standards of performance.
     The Administrator is to promulgate standards for these categories
according to the schedule referred to earlier.
     In some cases it may not be feasible immediately to develop a standard
for a source category with a high priority.  This might happen when a
program of research is needed to develop control techniques or because
techniques for sampling and measuring emissions may require refinement.  In
the developing of standards, differences in the time required to complete
the necessary investigation for different source categories must also be
considered.  For example, substantially more time may be necessary if
numerous pollutants must be investigated from a single source category.
Further, even late in the development process the schedule for completion of
a standard may change.  For example, inability to obtain emission data from
well-controlled sources in time to pursue the development process in a
systematic fashion may force a change in scheduling.  Nevertheless, priority
ranking is, and will continue to be, used to establish the order in which
projects are initiated and resources assigned.
     After the source category has been chosen, the types of facilities
within the source category to which the standard will apply must be
determined.  A source category may have several facilities that cause air
pollution, and emissions from some of these facilities may vary from
insignificant to very expensive to control.  Economic studies of the source
category and of applicable control technology may show that air pollution
control is better served by applying standards to the more severe pollution
sources.  For this reason, and because there is no adequately demonstrated
system for controlling emissions from certain facilities, standards often do
not apply to all facilities at a source. For the same reasons, the standards
may not apply to all air pollutants emitted.  Thus, although a source
category may be selected to be covered by a standard of performance, not all
pollutants or facilities within that source category may be covered by the
standards.
                                     2-6

-------
2.3  PROCEDURE FOR DEVELOPMENT OF STANDARDS OF PERFORMANCE
     Standards of performance must (1)  realistically reflect best demon-
strated control  practice;  (2) adequately consider the cost,  the non-air-
quality health and environmental  impacts, and the energy requirements  of
such control; (3) be applicable to existing sources that are modified  or
reconstructed as well  as new installations; and (4) meet these conditions
for all variations of operating conditions being considered  anywhere in the
country.
     The objective of a program for developing standards is  to identify the
best technological system of continuous emission reduction that has been
adequately demonstrated.  The standard-setting process involves three
principal phases of activity:  (1) information gathering, (2) analysis of
the information, and (3) development of the standard of performance.
     During the information-gathering phase, industries are  queried through
a telephone survey, letters of inquiry, and plant visits by  EPA representa-
tives.  Information is also gathered from many other sources, and a litera-
ture search is conducted.   From the knowledge acquired about the industry,
EPA selects certain plants at which emission tests are conducted to provide
reliable data that characterize the pollutant emissions from well-controlled
existing facilities.
     In the second phase of a project,  the information about the industry
and the pollutants emitted is used in analytical studies.  Hypothetical
"model plants" are defined to provide a common basis for analysis.  The
model plant definitions, national pollutant emission data, and existing
State regulations governing emissions from the source category are then used
in establishing "control alternatives."  These control alternatives are
essentially different levels of emission control.
     EPA conducts studies  to determine the impact of each control alterna-
tive on the economics of the industry and on the national economy, on  the
environment, and on energy consumption.  From several possibly applicable
alternatives, EPA selects  the single most plausible control  alternative  as
the basis for a standard of performance for the source category under study.
                                     2-7

-------
     In the third phase of a project, the selected control alternative is
translated into a standard of performance, which, in turn, is written in the
form of a Federal regulation.  The Federal regulation, when applied to newly
constructed plants, will limit emissions to the levels indicated in the
selected control alternative.
     As early as is practical in each standard-setting project, EPA repre-
sentatives discuss the possibilities of a standard and the form it might
take with members of the National Air Pollution Control Techniques Advisory
Committee.  Industry representatives and other interested parties also
participate in these meetings.
     The information acquired in the project is summarized in the background
information document (BID).  The BID, the standard, and a preamble
explaining the standard are widely circulated to the industry being
considered for control, environmental groups, other government agencies, and
offices within EPA.  Through this extensive review process, the points of
view of expert reviewers are taken into consideration as changes are made to
the documentation.
     A "proposal package" is assembled and sent through the offices of EPA
Assistant Administrators for concurrence before the proposed standard is
officially endorsed by the EPA Administrator.  After being approved by the
EPA Administrator, the preamble and the proposed regulation are published in
the Federal Register.
     As a part of the Federal Register announcement of the proposed
regulation, the public is invited to participate in the standard-setting
process.  EPA invites written comments on the proposal and also holds a
public hearing to discuss the proposed standard with interested parties. All
public comments are summarized and incorporated into a second volume of the
BID.  All information reviewed and generated in studies in support of the
standard of performance is available to the public in a "docket" on file in
Washington, D. C.
     Comments from the public are evaluated, and the standard of performance
may be altered in response to the comments.
                                     2-8

-------
     The significant comments and EPA's position on the issues raised are
included in the "preamble"  of a promulgation package,"  which also contains
the draft of the final  regulation.   The regulation is then subjected to
another round of review and refinement until it is approved by the EPA
Administrator.  After the Administrator signs the regulation, it is
published as a "final rule" in the Federal  Register.
2.4  CONSIDERATION OF COSTS
     Section 317 of the Act requires an economic impact assessment with
respect to any standard of performance established under Section 111 of the
Act.  The assessment is required to contain an analysis of:  (1)  the costs of
compliance with the regulation, including the extent  to which the cost of
compliance varies depending on the effective date of  the regulation and the
development of less expensive or more efficient methods of compliance;
(2) the potential inflationary or recessionary effects  of the regulation;
(3) the effects the regulation might have on small business with respect to
competition; (4) the effects of the regulation on consumer costs; and
(5) the effects of the regulation on energy use. Section 317 also requires
that the economic impact assessment be as extensive as  practicable.
     The economic impact of a proposed standard upon  an industry is usually
addressed both in absolute terms and in terms of the  control costs that
would be incurred as a result of compliance with typical, existing State
control regulations.  An incremental approach is necessary because both new
and existing plants would be required to comply with  State regulations in
the absence of a Federal standard of performance.  This approach requires a
detailed analysis of the economic impact from the cost differential that
would exist between a proposed standard of performance and the typical State
standard.
     Air pollutant emissions may result in additional costs for water
treatment and captured potential air pollutants may pose a solid waste
disposal problem.  The total environmental  impact of an emission source
must, therefore, be analyzed and the costs determined whenever possible.
     A thorough study of the profitability and price-setting mechanisms of
the industry is essential to the analysis so that an accurate estimate of
                                     2-9

-------
potential adverse economic impacts can be made for proposed standards.  It
is also essential to know the capital requirements for pollution control
systems already placed on plants so that the additional capital requirements
necessitated by these Federal standards can be placed in proper perspective.
Finally, it is necessary to assess the availability of capital to provide
the additional control equipment needed to meet the standards of
performance.
2.5  CONSIDERATION OF ENVIRONMENTAL IMPACTS
     Section 102(2)(C) of the National Environmental Policy Act (NEPA) of
1969 requires Federal agencies to prepare detailed environmental impact
statements on proposals for legislation and other major Federal actions
significantly affecting the quality of the human environment.  The objective
of NEPA is to build  into the decisionmaking process of Federal agencies a
careful consideration of all environmental aspects of proposed actions.
     In a number of  legal challenges to standards of performance for various
industries, the United States Court of Appeals for the District of Columbia
Circuit has held that environmental impact statements need not be prepared
by the Agency for proposed actions under Section 111 of the Clean Air Act.
Essentially, the Court of Appeals has determined that the best system of
emission reduction requires the Administrator to take into account counter-
productive environmental effects of a proposed standard, as well as economic
costs to the industry.  On this basis, therefore, the Court established a
narrow exemption from NEPA for EPA determination under Section 111.
     In addition to  these judicial determinations, the Energy Supply and
Environmental Coordination Act (ESECA) of 1974 (PL-93-319) specifically
exempted proposed actions under the Clean Air Act from NEPA requirements.
According to Section 7(c)(l), "No action taken under the Clean Air Act shall
be deemed a major Federal action significantly affecting the quality of the
human environment within the meaning of the National Environmental Policy
Act of 1969." (15 U.S.C. 793(c)(l)).
     Nevertheless, the Agency has concluded that the preparation of environ-
mental  impact statements could have beneficial effects on certain regulatory
                                    2-10

-------
actions.   Consequently,  although  not legally required  to  do  so  by  sec-
tion 102(2)(C)  of NEPA,  EPA has adopted a  policy requiring that environmen-
tal impact statements be prepared for various regulatory  actions,  including
standards of performance developed under Section 111 of the  Act.   This
voluntary preparation of environmental  impact statements, however,  in no way
legally subjects the Agency to NEPA requirements.
     To implement this policy, a  separate  section  in this document  is
devoted solely  to an analysis  of  the potential  environmental  impacts associ-
ated with the proposed standards.  Both adverse and beneficial  impacts  in
such areas as air and water pollution,  increased solid waste disposal,  and
increased energy consumption are  discussed.
2.6  IMPACT ON  EXISTING SOURCES
     Section 111 of the Act defines a new  source as ".  .  . any  stationary
source, the construction or modification of  which  is commenced  ..." after
the proposed standards are published.  An  existing source is redefined  as a
new source if "modified" or "reconstructed"  as  defined in amendments to the
general provisions of Subpart  A of 40 CFR  Part  60, which were promulgated in
the Federal Register on December  16, 1975  (40 FR 58416).
     Promulgation of a standard of performance  requires States  to  establish
standards of performance for existing sources in the same industry  under
Section 111 (d) of the Act if  the standard for  new sources limits  emissions
of a designated pollutant (i.e.,  a pollutant for which air quality  criteria
have not been issued under Section 108 or  which has not been listed as  a
hazardous pollutant under Section 112). If  a State does  not act,  EPA must
establish such  standards.   General provisions outlining procedures  for
control of existing sources under Section  lll(d) were  promulgated  on
November 17, 1975, as Subpart  B of 40 CFR  Part  60 (40  FR  53340).
2.7  REVISION OF STANDARDS OF  PERFORMANCE
     Congress was aware that the  level  of  air pollution control achievable
by any industry may improve with  technological  advances.  Accordingly,
Section 111 of the Act provides that the Administrator "... shall, at
least every 4 years, review and,  if appropriate, revise ..."  the
                                    2-11

-------
standards.  Revisions are made to assure that the standards continue to
reflect the best systems that become available in the future.  Such
revisions will not be retroactive, but will apply to stationary sources
constructed or modified after the proposal of the revised standards.
                                     2-12

-------
        3.0  CHARACTERISTICS OF THE INDUSTRIAL  BOILER SOURCE  CATEGORY

3.1  GENERAL
     In this section,  industrial boilers  are  described and  classified  by
type, fuel, and method of construction.   The  existing population  of
industrial  boilers is  characterized by design type  and capacity.   In
addition, fuel  usage patterns are discussed by  EPA  region,  and  boiler  usage.
Existing regulations applicable to industrial boilers are also  presented.
3.1.1 Industrial Boiler Source Category
     Industrial boilers are used in manufacturing,  processing,  mining,  and
refining industries to provide steam,  hot water,  and  electricity.  The
industrial  generation  of electricity is quite limited, however, with the
majority of industrial boiler fuel  consumption  dedicated to steam or hot
water production.  Between 10 and 15 percent  of industrial  boiler coal
consumption and 5 to 10 percent of industrial boiler  oil and  gas  consumption
is used for electricity generation.
     Industrial boilers cover a broad  range of  sizes, with  a  few  units  as
large as 200 MW (700 x 10  Btu/hr)  thermal  input.   Some units are
sufficiently small to  allow shop fabrication  and  shipment as  packaged
boilers.  When  used for heating, they  may have  a  steam pressure as low as
13.8 kPa (2 psi) and a temperature no  higher  than 375 K (215°F).   Extremely
large units, in contrast, may produce  as  much as  545  Mg/hr  (1,200,000  Ib/hr)
of steam at a pressure of 12400 kPa (1800 psi)  and  a  temperature  of 811 K
(1000°F).2
3.1.2  Classification  of Industrial Boilers
     Boilers can be classified by type, fuel, and method of construction.
Boiler types are identified by the heat transfer method (watertube, fire-
tube, or cast iron), the arrangement of the heat transfer  surfaces
(horizontal or  vertical, straight or bent tube),  and, in the  case of  coal,
the fuel feed system (pulverized or stoker).   Pulverized coal-, oil-,  and
gas-fired boilers can  be subclassified further  by burner configuration
                                     3-1

-------
(tangential, front wall or horizontally opposed).  Most industrial boilers
that are equipped with burners use the front wall configuration.  For the
purposes of this study, the burner-equipped boiler group will not be
subclassified according to burner configuration.
     Figure 3-1 illustrates a scheme for classifying boilers.  According to
this scheme boilers are first classified by the major distinguishing charac-
teristic -- heat transfer mechanism -- into watertube, firetube, or cast
iron design types.  Watertube boilers can be either field-erected or shop-
built packaged units, but essentially all firetube and cast iron units are
packaged boilers.  These boilers are further classified by fuel type.  The
fuel feed mechansim is an important characteristic affecting coal-fired
boiler emissions.  As shown on Figure 3-1, three types of stoker feeding
mechanisms are used, in addition to pulverized coal systems.  Differences in
emissions and their potential for control exist among these boiler types due
to the fuel fired and the'mechanism for introducing that fuel (see
Section 3.2 and Chapter 4).
     3.1.2.1  Industrial Boiler Design Types.  The three major boiler
designs — watertube, firetube, and cast iron -- are each manufactured to
meet specific application and site requirements.  Unit size, design steam
pressure and temperature all depend on the application.  Each of these
boiler types may burn coal, oil, or gas and are increasingly being designed
to burn more than one fuel type.  Each of the three major design types are
discussed in the following subsections.
     3.1.2.1.1  Watertube boilers.  Watertube boilers are used in a variety
of applications ranging from supplying large amounts of process steam to
providing space heat for industrial facilities.  As the name implies,
watertube boilers are designed to pass water through the inside of heat
transfer tubes while the outside of the tubes is heated by direct contact
with the hot combustion gases.  This process results in generation of
high-pressure, high-temperature steam.
     Watertube boilers are available, as packaged or field-erected units, in
capacities ranging from less than 2.9 to over 200 MW (10 to 700 x
10  Btu/hr) thermal  imput.   As discussed in Section 3.2, industrial boilers
                                     3-2

-------
                                                                               PULVERIZED
                                                                     COAL
                                   WATERTUBE (FIELD-ERECTED AND
CO
CO
                     INDUSTRIAL
                    BOILERS
                                              PACKAGED)
FIRETUBE (PACKAGED)
                                    CAST IRON  (PACKAGED)
                                                                                 STOKER
                                                                 RESIDUAL OIL
                            DISTILLATE OIL
                                                                 NATURAL  GAS
                                                                     COAL
                            RESIDUAL OIL
                                                                DISTILLATE OIL
                                                                NATURAL GAS
                                                                    COAL
                                                                RESIDUAL OIL
                                                                DISTILLATE OIL
                                                                NATURAL GAS
                                                         SPREADER
                                                         OVERFEED
                                                         UNDERFEED
                                      Figure 3-1.  Categories of industrial boilers.

-------
typically have a thermal efficiency of approximately 80 percent.   Hence,  the
steam capacity range of these watertube units is less than 3.6 Mg/hr
(8,000 Ib/hr) to over 254 Mg/hr (560,000 Ib/hr).  Packaged boilers are
generally smaller than the field-erected units.  Virtually all units with a
steam capacity of 4.5 to 45 Mg/hr (10,000 to 100,000 Ib/hr) are packaged.3
The maximum size of industrial packaged boilers is limited by transport-
ability.  Units up to 115 Mg/hr (250,000 Ib/hr) of steam can be transported
to operation sites by railroad, and larger ones can be moved by barge or
freighter.
     Industrial watertube boilers can burn coal, residual oil, distillate
oil, natural gas, liquefied petroleum gas, and other fossil and nonfossil
fuels.  The packaged units, however, usually use premium quality fossil
fuels — oil and natural gas — and do not offer the fuel flexibility that
                                  4
many buyers have recently desired.   More than half of the installed
capacity of coal-fired industrial boilers are field-erected watertube
       5
models,  which are available as stoker or pulverized coal-fired units. A
stoker is a conveying system that feeds coal into the furnace and also
provides a grate upon which the coal is burned.  In comparison, pulverized
coal-fired units operate by using suspension burning.  Coal pulverized to
the consistency of powder is pneumatically injected into the furnace.
Specific details of these boiler types are presented in Section 3.2.
     3.1.2.1.2  Firetube boilers.  Firetube boilers are used primarily for
heating systems, industrial process steam, and portable power boilers.
Essentially all firetube boilers are packaged units with some being portable
(or movable) rather than stationary.  In firetube boilers, the hot gas flows
through the tubes and the water being heated circulates outside of the
tubes.
     Firetube boilers are usually limited in size to 5.9 MW (20 x 10
Btu/hr) thermal input.   However, some firetube designs have been built with
heat input up to 14 MW (50 x 10  Btu/hr).  In general, firetube boilers
offer the benefit of quick response to moderate load changes.   Most
industrial firetube boilers currently available have tube arrangements that
classify them as either horizontal return tube (HRT), firebox, or Scotch.
                                     3-4

-------
These tube arrangements are described in Section 3.2.3.2.   Most of the
                                                           Q
installed capacity of firetube units is oil- and gas-fired.    Coal  may play
                                             3
a more important role in the future, however.
     Boilers which are portable or semiportable are also used in industrial
applications.  Small boilers on wheels are used in a variety of industries
for steam cleaning and pressing.  Larger firetube units with complete piping
and water supply systems may be mounted on trailers or flat beds and used
for temporary heat or wall drying at construction sites.  Portable or rental
boilers do not differ in design from other package boilers but are typically
                                 g
provided with a very short stack.
     3.1.2.1.3  Cast iron boilers.  In cast iron boilers,  the hot gas is
contained inside the tubes and the water being heated circulates outside the
tubes.  The units are constructed of cast iron rather than steel.  Cast iron
boilers are used to produce either low-pressure steam or hot water.
Generally, boiler capacity ranges from 0.001 to 2.9 MW (0.003 to 10 x 106
Btu/hr)10 thermal heat input with pressure ratings up to 690 kPa (100 psi)
for hot water units and 100 kPa (15 psi) for steam units.     Thus, cast iron
boilers are most commonly used in domestic or small commercial
   -,-  .u-    10
applications.
     3.1.2.2  Industrial Boiler Manufacturers.  Tables 3-1 and 3-2 list
manufacturers of industrial boilers.  Manufacturers of watertube and fire-
tube boilers are listed in Table 3-1.  Most manufacturers of industrial
watertube boilers make coal-fired units, but only a few of the firetube
boiler manufacturers make coal-fired units.  Establishments manufacturing
cast iron boilers are listed in Table 3-2.  According to the Hydronics
Institute, which is the trade association for cast iron boiler manufac-
turers, only five firms produce units large enough for industrial
             12
applications.
3.1.3  Population of Industrial Boilers
     The installed population of industrial boilers is summarized by design
type in Table 3-3.  Figure 3-2 illustrates the relative distribution by
capacity.  As shown on this graph, watertube boilers are available over a
larger size range than the other types.  Note on Table 3-3 that  nearly
                                    3-5

-------
Table 3-1.   MANUFACTURERS  OF INDUSTRIAL FIRETUBE AND WATERTUBE BOILERS
                                                                                     13
Fire tubes
Oil &
Manufacturer gas Coal
ABCO Industries X
Abilene. TX
American Hydrotherm Corp.
New York, NY
Babcock and Wilcox
North Canton. OH
Bettran Associates, Inc. X
Brooklyn, NY
Bethlehem Corp. X X
Easton, PA
Btgelow Co. X
New Haven, CT
Bryan Steam Corp.
Peru, IN
Burham Corp. X
Lancaster. PA
Clayton Manufacturing Co.
El Monte, CA
Cleaver Brooks X
Milwaukee, Wl
Combustion Engineering
Windsor. CT
Combustion Service I Equipment Co. X X
Plttsburg. PA
Delta Steel Boiler Industries X X
Chicago, IL
Oeltak
Minneapolis, MN
Durham-Bush X
West Hartford, CT
Watertubes
Upper size limit.'
Oil & MU (106 tttu/hr)
gas Coal thermal input
X 29
(100)
X 10
(35)
X X


X X
X 6
(21)

X 6
(21)
X 29
(100)
X X


X

        'Manufacturers of watertube boilers with no upper size limit listed produce units
         greater than 73 MU (250 x 10* Btu/hr) heat Input.
                                           3-6

-------
Table 3-1   Continued.   MANUFACTURERS OF INDUSTRIAL  FIRETUBE  AND WATERTUBE BOILERS13
Fire tubes
Oil (,
Manufxturer gas Coal
Eclipse Lookout Co. X
Chattanooga. TN
Foster-Wheeler
Livingston. NJ
Industrial Boiler X X
Thonasvllle. GA
Industrial Combustion, Inc. X
Milwaukee. MI
International Boiler
East Stroudsburg, PA
Johnston Boiler Co. X X
Ferrysburg, MI
E. Keeler
Wllllamsport. PA
Kewanee Boiler Corp. X X
Kewanee, IL
Kipper and Sons Engineers, Inc.
Seattle. WA
James Leffel Co. X X
Springfield, OH
Nebraska Boiler Co. X
Lincoln, NE
North American Manufacturing Co. X X
Cleveland. OH
Ocean Shore Boiler Works X
San Francisco, CA
Oswego Package Boiler X
Oswego, NY
Ray Burner Co. XX
San Francisco, CA
- — » • • 	 '

Water tubes
Upper size limit,"
Oil 4 MU (10* 3tu/hr)
gas Coal thermal input

X X


X X 73
(2SO)

X X

X X

X X 73
(250)





              "Manufacturers of watertube boilers with no upper size limit listed  produce units
               greater than 73 Mw (250 x 10* Btu/hr) heat input.
                                                 3-7

-------
Table 3-1  Concluded.  MANUFACTURERS OF  INDUSTRIAL  FIRETUBE  AND WATERTUBE BOILERS
                                                                                                 13
Fl retries
on i.
Manufacturer gas Coal
Riley Stoker
Worcester. MA
Seattle Boiler Works X
Seattle, WA
Sellers Engineering Co. X
Danville. KY
Steamaster Automatic Boiler Co. X
Los Angeles, CA
Struthers Wells
Utnfleld, KS
Superior Boiler Works, Inc. X
llutchlnson, KS
Thermo- Pak Boiler, Inc.
Memphis, TN
Trane Company X
Lacrosse, WI
Vapor Division of Brunswick Corp.
Chicago. IL
Henry Vogt Machine Company
Louisville. KY
Williams and Davis Bo Her and X
Welding Co., Inc.
Hutch ins. TX
York-Shipley X
York, PA
John Zink X
Tulsa, OK
Zurn Industries X X
Erie, PA
Water tubes
Upper size limit,'
Oil & MU (ID6 Btu/hr)
gas Coal them*) input
X X
X 14
(50)


X 29
(100)

X 6
(20)
X 73
(2SO)
X 4
(15)
X X 73
(250)


X 73
(250)
X X
              'Manufacturers of watertube boilers with no upper size limit listed  produce units
              greater than 73 MM (250 x 10* Btu/hr) heat Input.
                                                3-8

-------
TABLE 3-2.   MANUFACTURERS OF INDUSTRIAL SIZE CAST IRON BOILERS
                                                               12
                      Burnham Corporation
                        Lancaster, PA

                      Peerless Heater Company
                        Boyertown, PA

                      Slant/Fin Corporation
                        Greenvale, NY

                      H.B.  Smith, Incorporated
                        Westfield, MA

                      Weil-MeLain
                        Michigan City, IN
         TABLE 3-3.  BOILER POPULATION DISTRIBUTION BY
                     HEAT - TRANSFER CONFIGURATION7

Heat-
transfer
configuration
Watertube
Firetube
Cast iron
Boiler
Number of
Boilers
37,969
173,936
295,298
Population
Percent
of Total
7.5
34.3
58.2
Total Boiler
MW Thermal
Input
(106 Btu/hr)
638,665
(2.2 x 106)
219,360
(7.6 x 105)
52,570
(1.8 x 105)
Capacity
Percent
of Total
70-0
24.2
5.8
                               3-9

-------
U)
l-»
o
                       300,000
                    (1,000,000)
                 CO
                vo
                 o
                  a.
                  c
                  0)
                  o
                  <
                  a.
                       200,000
                      (680,000)
                       100,000
                      (340,000)
\\X\\\\\
                                                                                           CAST  IRON
                                                                                           FIRETUBE
                                                                                           WATERTUBE
                                     0-2.9     2.9-14.7     14.7-29.3   29.3-73.3    >73.3
                                    (0-10)     (10-50)     (50-100)     (100-250)    (>250)
                                          SIZE RANGE, MM thermal  input (106  Btu/hr)
                          Figure 3-2.  Relative distribution J)y  capacity of the three types
                                       of industrial  boilers.

-------
60 percent of the boilers are cast iron units,  but these  boilers  account  for
only 6 percent of the installed capacity.   Watertube boilers,  on  the  other
hand, represent 7 percent of the boilers by number,  but account for
70 percent of the installed capacity.   Figure 3-2  shows that the  largest
concentration of boiler capacity is in  the 2.9  to  14.7  MW range,  which
contains 26 percent of the installed capacity.   Units over 73.3 MW thermal
input account for 23 percent, and those less than  2.9 MW  thermal  input
account for 20 percent of the installed capacity.
     Table 3-4 gives the distribution of watertube boilers by  capacity  and
fuel.  About 25 percent of the installed capacity  is coal-fired,  32 percent
is oil-fired, and 43 percent is natural gas-fired.  This  distribution varies
with size.  In the smallest size range  (less than  2.9 MW  thermal  input),
only 7 percent of the capacity is coal-fired, whereas in  the largest  boiler
size group (above 73 MW thermal input), 30 percent of the installed capacity
is coal-fired.  Even in this large size group,  however, 47 percent of the
currently installed capacity is gas-fired.
     Table 3-5 presents the distribution of firetube units, which range in
size up to 14.7 MW (50 x 10  Btu/hr) thermal input.   Only 6 percent of  the
installed capacity is coal-fired; 43 percent is oil-fired and  51  percent  is
natural gas-fired.
     Cast iron boilers are the smallest of the  three boiler types, with a
maximum size of only 2.9 MW (10 x 10 Btu/hr) thermal input.   In  this group
12 percent of the installed capacity is coal-fired,  33  percent is oil-fired,
and 55 percent is natural gas-fired, as shown in Table  3-6.
     Information on the age distribution of existing industrial boilers was
estimated by PEDCo Environmental, Inc.  using sales data obtained  from the
American Boiler Manufacturers Association (ABMA) and the  Hydronics
Institute.    Based on these data, PEDCo stated that in 1978 about 20
percent of the watertube boiler capacity currently in place was less  than
7 years old. (The sales data from ABMA  were available for the  previous
7 years.)  About 25 percent of the current capacity of  firetube and  cast
iron boilers is less than 10 years old.  Further,  based on discussions  with
boiler manufacturers, PEDCo estimated that 27 percent of  the current sales
                                    3-11

-------
                           Table  3-4.   INSTALLED  CAPACITY OF U.S. WATERTUBE INDUSTRIAL BOILERS

                                        BY UNIT  SIZE  AND FUEL TYPE14



                                                 {MW thermal  Input (10& Btu/hr))
to
i
Fuel
Pulverized coal
Spreader -stoker coal
Under feed -stoker coal
Overfeed-stoker coal
Total coal
Residual oil
Distillate oil
Total oil
Natural gas
Total all fuels

0 to 2.9
(0 to 10)
0
(0)
70
(240)
680
(2,300)
85
(290)
835
(2,830)
3,960
(13,500)
2,560
(8,700)
6,520
(22,200)
4,475
(15,300)
11,830
(40,330)

2.9 to 14.7
(10 to 50)
0
(0)
4,650
(15,900)
14 , 105
(48,000)
3,470
(11,800)
22,225
(75,700)
48,190
(164,000)
8,280
(28,200)
i
56,470
(192,200)
57,900
(197,500)
136,595
(465,400)
Capacity by
14.7 to 29.3
(50 to 100)
0
(0)
6,175
(21,060)
17,265
(58,900)
4,455
(15,200)
27,895
(95,160)
35,640
(122,000)
4,295
(14,600)
39,935
(136,600)
53,585
(182,800)
121,415
(414,560)
unit size
29.3 to 73.3
(100 to 250)
19,895
(67,800)
20,295
(69,000)
7,080
(24,200)
3,555
(12,100)
50,825
(173,100)
44,790
(153,000)
6,370
(21,700)
51,160
(174,700)
63,320
(216,000)
165,305
(563,800)

>73.3
(>250)
40,180
(137,000)
11,010
(37,600)
5,230
(17,800)
3,510
(12,000)
59,930
•(204,400)
43,570
(148,600)
4,085
(13,900)
47,655
(162,500)
95,935
(327,200)
203,520
(694.100)

Totals
60,075
(204,800)
42,200
(143,800)
44,360
(151,200)
15,075
(51,390)
161,710
(551,190)
176,150
(601,100)
25,590
(87,100)
201,740
(688,200)
275,215
(938,800)
638,665
(2,178,190)

-------
TABLE 3-5.
INSTALLED CAPACITY OFQINDUSTRIAL FIRETUBE BOILERS
BY SIZE AND FUEL TYPEB
              [MW thermal input (10$ Btu/hr)]
                            Capacity by unit size
                      0 to 2.9   2.9 to 14.7
Fuel
Coal
Residual oil
Distillate oil
Natural gas
Total
(0 to 10)
5,650
(19,270)
35,280
(120,330)
17,770
(60,610)
59,120
(201,630)
117,820
(401,840)
(10 to 50)
7,780
(26,530)
25,860
(88,200)
15,770
(53,790)
52,130
(177,790)
101,540
(346,310)
Total
13,430
(45,800)
61,140
(208,530)
33,540
(114,400)
111,250
(379,420)
219,360
(748,150)
                             3-13

-------
        TABLE 3-6.  INSTALLED CAPACITY OF INDUSTRIAL
               CAST IRON BOILERS BY FUEL TYPE15

               [MW Thermal Input (106 Btu/hr)]
     Fuel                                Boiler Capacity
     Coal                                    6,330
                                           (21,590)

     Residual oil                           10,780
                                           (36,770)

     Distillate oil                          6,740
                                           (22,990)

     Natural gas                            28,720
                                           (97,950)


aAll cast iron boilers have a capacity less than
 4.0 MW thermal input (14 x 106 Btu/hr).
                                                       a
                               3-14

-------
of watertube and firetube boilers  and 50 percent of the  sales  of cast  iron
units are replacements for existing boilers  that are being  retired.
     Seasonal and time-of-day changes in energy demand result  in excess
capacity during nonpeak demand periods.   Section 3.1.4 discusses these
variations in capacity utilization by industry.  Since capacity utilization
varies widely in industry, the installed capacity by itself is not  an
adequate measure of fuel consumption.  This  information  must be obtained
from actual fuel usage data.   Summaries of industrial  fuel  consumption data
are presented in the next subsection.
3.1.4  Fuel Usage Patterns
     Table 3-7 gives fossil fuel  consumption by the manufacturing industries
in each EPA region.  Region V (including Ohio,  Michigan, Illinois,  and
Indiana) is a heavy consumer of fossil  fuels, using nearly  40  percent  of  the
national annual coal consumption  by manufacturers and about 20 percent of
the U.S. annual consumption of distillate oil and natural gas  by the
manufacturing industries.  Region  VI (including Texas and Louisiana)
consumes 40 percent of the natural gas.   In  general, the largest end use  of
each fuel in each region is for process steam,  accounting for  about
                                                  19
one-third of the fuel  used by industry nationwide.     Use of fuel as
feedstock accounts for less than  one-third of the manufacturing fuel use.
Industrial process heating in furnaces and space heating accounts for  most
                                             19
of the remaining industrial fuel  consumption.
     Table 3-8 lists the percent  of industrial  boiler fossil fuel consump-
tion by industry and fuel type for 1974.  The most energy intensive
industries—chemicals and paper—account for over one-third of the
industrial boiler fuel consumption.  The other industries listed (petroleum
refining, steel, aluminum, and food) account for much of the remaining
industrial boiler fuel consumption.  Other industries, using smaller  but
significant amounts of fuel in boilers, include textiles, lumber, rubber,
metal fabrication, and transportation.
     Table 3-9 relates the share  of fuel used in this country  by industry to
the quantities used by other sectors.  In addition to manufacturing,  the
                                    3-15

-------
TABLE 3-7.  REGIONAL FOSSIL FUEL CONSUMPTION BY THE MANUFACTURING
            INDUSTRIES IN 1976a
                     [Exajoules (1015 Btu)]
EPA
Region
I
II
III
IV
V
VI
VII
VIII
IX
X
Total
Residual
oil
0.19
(0.18)
0.21
(0.20)
0.28
(0.27)
0.38
(0.36)
0.26
(0.25)
0.09
(0.085)
0.023
(0.022)
0.029
(0.027)
0.05
(0.047)
0.051
(0.048)
1.6
(1.5)
Distillate
oil
0.033
(0.031)
0.077
(0.073)
0.12
(0.11)
0.14
(0.13)
0.12
(0.11)
0.039
(0.037)
0.026
(0.025)
0.0095
(0.009)
0.023
(0.022)
0.023
(0.022)
0.61
(0.58)
Coal
0.0071
(0.0067)
0.20
(0.19)
1.2
(1.1)
0.48
(0.45)
1.4
(1.3)
0.079
(0.075)
0.11
(0.1)
0.10
(0.095)
0.088
(0.083)
0.019
(0.018)
3.7
(3.5)
Natural
gas
0.081
(0.077)
0.18
(0.17)
0.51
(0.48)
0.69
(0.65)
1.4
(1.3)
2.7
(2.6)
0.32
(0.3)
0.14
(0.13)
0.48
(0.46)
0.19
(0.18)
6.7
(6.4)
Total
0.31
(0.29)
0.67
(0.63)
2.1
(2.0)
1.7
(1.6)
3.2
(3.0)
2.9
(2.8)
0.48
(0.45)
0.28
(0.26)
0.64
(0.61)
0.28
(0.27)
13
(12)
  3The  fuel  consumption  data  include  use  in  process  heaters
  and  as feedstock  as well as  in  boilers.   Reference  18.
                                3-16

-------
Table 3-8. DISTRIBUTION OF INDUSTRIAL BOILER FO
IN 1974 BY INDUSTRY AND FUEL TYPE zo/

(percent)
SSIL FUEL

CONSUMPTION

Fuel type
Industry
Chemicals
Paper
Steel and aluminum
Food
Petroleum refining
Other manufacturing
Coal, Oil,
4.5-5.6 1.9-2.4
2.9-3.6 6-7.4
2.6-3.3 0.68-0.85
1.1-1.3 1.3-1.6
0.07-0.09 1.2-1.5
3.5-4.3 6.1-8.5
Gas,
14-17
4-5
5.5-6.8
4.8-5.9
3.8-4.8
14-35
Total,
20-25
13-16
8.8-11
7.1-8.8
5.1-6.4
24-48
Total
15-18
17-22
46-75
100
                               3-17

-------
                              TABLE  3-9.
CO

l-»
oo
UNITED STATES PRIMARY ENERGY CONSUMPTION BY CONSUMING
SECTOR AND ENERGY SOURCE, 19742'
                                                    [ExajoulesdO15  Btufl


Consuming sector3
Household and
commercial
Industrial

Transportation

Electrical
generation
Total

Percent


Coal
0.31
(0.29)
4.44
(4.2)
0.01
(0.01)
9.15
(8.67)
13.9
(13.2)
18.0


Petroleum
6.75
(6.39)
6.38
(6.04)
18.6
(17.6)
3.64
(3.45)
35.4
(33.5)
45.9


Natural gas
7.51
(7..1)
11.75
(11.1)
0.70
(0.66)
3.51
(3.23)
23.5
(22.2)
30.4


Nuclear


— _

....

1.24
(1.17)
1.24
(1.17)
1.6
Hydropower
and
geothermal
„

__

•v»

3.19
(3.02)
3.19
(3.02)
4.1


Total
14.6
(13.8)
22.6
(21.4)
19.3
(18.3)
20.7
(19.6)
77.2
(73.1)
100.0


Percent
18.9

29.3

25.0

26.8

100.0


            Excludes use of electrical  energy  by  household, commercial, industrial,
             and transportation sectors
             Includes all manufacturing sectors.

-------
industrial  sector includes the mining,  agricultural,  and construction
industries.
3.2  INDUSTRIAL BOILERS AND THEIR EMISSIONS
     An overview of industrial boiler emissions is presented in  this
section.  Representative boilers for each major class of boilers are
selected and mass and energy balances are presented.   These mass balances
are based on emission factors for each  fuel type.   Finally, in conjunction
with the mass balances and discussion of boiler types, factors affecting
emissions for each boiler type are discussed.
     This section begins with a qualitative discussion of uncontrolled
industrial  boiler emissions (Section 3.2.1).   Following this overview are
individual  subsections dealing with emissions  from various types of boilers.
For purposes of this analysis, these subsections are  arranged by fuel type
with subsections for coal, oil, and natural gas (3.2.2, 3.2.3, and 3.2.4
respectively)..  At the conclusion of each subsection, emission factors  (on  a
ng/J or lb/10  Btu heat input basis) are presented.   These emission factors
are used to quantify uncontrolled emissions throughout the remainder of this
report.
     Because cast iron boilers are small, less than 2.9 MW (10 x 10 Btu/hr)
thermal imput, and most new boilers will be watertube or firetube types,
cast iron boilers are not discussed in  this section.   In addition, fugitive
emissions,  which result from processes  such as the transfer and  storage of
coal and oil supplies, and the preparation of  the coal (grinding and
pulverizing), are not considered.
3.2.1  Uncontrolled Emissions Overview
     Emissions from industrial boilers  include particulate matter (PM),
sulfur oxides (SO ), nitrogen oxides (NO ), and lesser amounts of carbon
                 A                      A
monoxide (CO), hydrocarbons (HC), and trace elements.  In the following
subsections, sources of these pollutants are noted, and factors  affecting
their emission rates are discussed qualitatively.
     3.2.1.1  Particulate Matter (PM) Emissions.  Uncontrolled  PM emissions
from coal-fired boilers include the ash in the fuel as well as  unburned
                                    3-19

-------
carbon resulting from incomplete combustion.  Emission factors for PM are
normally expressed as a function of fuel ash content for coal-fired boilers
(see Section 3.2.2).  Coal ash may either settle out in the boiler (bottom
ash) or be carried out with the flue gas (fly ash).  The distribution of ash
between the bottom and flyash fractions directly affects the PM emissions
    22
rate   and is a function of the following:

       -  Boiler firing method -- The type of firing is perhaps the most
          important factor in determining ash distribution.  For example,
          stoker-fired units emit less fly ash than dry bottom, pulverized-
          coal-fired boilers.
       -  Wet or dry bottom furnace -- Furnaces which are designed to
          generate a dry bottom ash entrain PM from the bottom ash hopper
          into the flue gas stream more easily than do boilers whose bottom
          ash is in the molten state.
Boiler load also affects PM emissions from coal-fired boilers.  In general,
decreasing load tends to reduce PM emissions, however, the magnitude of the
reduction varies considerably depending on boiler type, fuel, and boiler
operation.
     For oil-fired boilers, carbon residue, a measure of the heaviest and
                                    2*3
least volatile components in the oil, is the most important fuel property
influencing PM emissions of size greater than 10 ym.  The PM emitted by
distillate oil-fired boilers is primarily carbonaceous particles resulting
from the partial combustion of the fuel.  PM emissions from distillate
oil-fired boilers do not correlate with the ash or sulfur content of the
fuel.24  Unlike the emissions from coal-fired boilers, the PM emissions from
distillate oil-fired boilers do not necessarily vary with boiler load in a
              25
general trend.
     Residual oil-fired boiler PM emissions result from ash in the fuel as
well as incomplete combustion of the fuel.  Test data reported in AP-42
shows PM emissions from residual oil-fired units vary with the sulfur
content of the fuel.26  Thus emission factors for PM, which are presented
later, are expressed as a function of sulfur content for residual oil-fired
                                    3-20

-------
units.   PM emissions from residual  oil-fired boilers  are influenced  by
boiler load and tend to decrease with decreasing load.
     The PM emission factors for industrial  boilers firing natural gas are
very low because natural gas has little or no ash content and combustion  is
more complete than with other fuels.
     Soot blowing is another source of PM emissions in  coal-  and  residual
oil-fired boilers.  Steam soot blowing is used intermittently in  industrial
boilers to dislodge ash from heat transfer surfaces in  the boiler furnace,
convection section, and economizer/preheater.  On small  boilers with single
soot blowers, soot blowing may only take place for a  few seconds  once a
shift.  Large industrial boilers may have numerous soot blowers  installed
and operated in a cycle which may approach "continuous"  soot  blowing.   The
incremental PM emissions associated with soot blowing are not reflected in
AP-42 emission factors and boiler owners and equipment  vendors disagree as
to whether or how emissions resulting from soot blowing are accounted for in
                                                              27
the design of industrial boiler PM emission control equipment.    Test data
reported in Chapter 4 show varied impacts of soot blowing on  controlled
emissions depending on the type of emission control technique employed.  In
general, soot blowing appears to have only a small effect on  opacity with
                                                  28
opacity increases of less than 5 percentage points.
     3.2.1.2  Sulfur Oxide Emissions.  SC^ emissions  are generated in
industrial boilers due to oxidation of sulfur contained in fuels.  SO
                                                                     X
emissions from industrial boilers are predominantly  in  the form  of S09; SO.,
                                                                     £    u
emissions account for only 1 or 2 percent of the total  SO  emissions.
                                                         /\
Uncontrolled emissions of S0« depend directly on the  sulfur content  of the
fuel.  The type of firing mechanism does not affect  SO, emissions, but
                                 29
variations in fuel properties do.    Therefore, a different emission factor,
that is primarily a function of the amount of sulfur  in the fuel, is used
for each fuel type.  This factor is essentially constant for  all  boiler
types firing the same fuel.
     The emission factor in AP-42 for coal-fired units assumes that less
than 5 percent of the sulfur in the coal is emitted with the  particulate
                                    3-21

-------
matter or retained by the bottom ash.  The amount of sulfur retained by the
fly ash and bottom ash appears to be a function of ash composition, related
to the alkalinity of the ash.    Combustion of highly alkaline Western
subbituminous coals can result in 20 percent of the sulfur in the coal being
emitted with the fly ash or retained in the bottom ash.31  Thus, the SCL
emission factor for coal is based on an average of emissions from many coal
types, and variations from this average will occur.
     3.2.1.3  Nitrogen Oxide Emissions.  Oxides of nitrogen (including NO
and NOp) formed in combustion processes are due either to thermal fixation
of atmospheric nitrogen in the combustion air, resulting in formation of
thermal NO , or to the conversion of chemically bound nitrogen in the fuel,
          A
resulting in formation of fuel NO .  For natural gas and distillate oil
                                 A
firing, nearly all NO  emissions are thermal NO .  With residual oil and
                     A                         X
coal fuels, the contribution from fuel NO  can be significant and even
            32
predominant.
     Experimental measurements of thermal NO  formation have shown that NO
                                            A                             X
concentration is exponentially dependent on temperature and also propor-
tional to the N9 concentration, the residence time, and the square root of
                              33
09 concentration at the flame.    Thus, the formation of thermal NO  is
 L-                                                                 A
affected by four factors: (1) nitrogen concentration, (2) oxygen concentra-
tion, (3) peak temperature, and (4) time of exposure at peak temperature.
The emission trends due to changes in these factors are fairly consistent
for all types of boilers; an increase in flame temperature, oxygen
availability, and/or residence time at high temperatures leads to an
increase in NO  production regardless of the boiler type.
              A
     As mentioned previously, fuel NO  is of importance for residual oil and
                                     A
coal firing.  It can account for 50 percent of the total NO  emissions in
                                                      34
residual oil firing and for 80 percent in coal firing.    The percent
conversion of fuel nitrogen to NO , however, varies greatly.  Anywhere from
                                 A
20 to 90 percent of nitrogen in oil is converted to NOV while the percentage
                                                                 35
of nitrogen in coal converted to NO  ranges from 5 to 60 percent.
                                   A
Furthermore, test data indicate that the percent of fuel nitrogen conversion
decreases as the fuel nitrogen content increases.  An average conversion of
                                     3-22

-------
46 percent was found for residual  oil, and nearly 100 percent for distillate
    oc
oil.    For coal-fired units the fuel  nitrogen conversion varies  depending
on the combustion conditions present with the particular boiler and fuel
(see Section 4.3).
     3.2.1.4  Carbon Monoxide Emissions.   The rate of CO emissions  from
boilers depends on the efficiency of the  combustion of the fuel.   By
controlling the combustion process carefully, CO emissions can be minimized.
The effects of combustion modifications for purpose of NO  control  on
                                                         X
uncontrolled CO and HC emissions are discussed in Section 4.3 of  this
document.
     3.2.1.5  Hydrocarbon Emissions.  The rate of HC emission from boilers
also depends on the combustion efficiency.  Hydrocarbon emissions are
minimized by use of proper combustion practices.  Fuel  type also  affects  HC
emissions.  Liquid and gaseous fuels have better mixing and firing charac-
teristics than solid fuels, accounting in part for the lower hydrocarbon
emissions for oil and natural gas-fired units than for comparable coal
units.
     3.2.1.6  Trace Element Emissions. Trace elements are found  in fossil
fuels, especially in coal and residual oil.  Smaller concentrations of  trace
elements are also found in distillate oil, but virtually none are found in
natural gas.
     Trace elements can be classified according to the way they are emitted
during the combustion process: (1) distributed between bottom ash and fly
ash, (2) concentrated in fly ash, especially the fine particulate in the
flue gas; or (3) as vapors.  Trace elements that do not vaporize  during fuel
combustion are emitted in about equal  concentration in bottom ash and fly
ash particles.  Those with lower boiling  points, which vaporize during
combustion, become concentrated in fly ash and are carried out by the flue
gas.  Some trace elements, such as mercury, are emitted through the stack
into the atmosphere as vapors.  Others, such as arsenic, cadmium, copper,
lead, tin, and zinc, condense on fly ash  particles and are emitted with them
                    oc
into the atmosphere.
                                    3-23

-------
     The quantity of trace elements actually emitted depends on three
factors:
       •  combustion temperature,
       •  fuel analysis and feed mechanism, and
       •  characteristics of the flue gas.
     The fuel analysis determines the quantity of trace elements present.
The combustion temperature determines the degree of volatilization for
specific trace elements, and the fuel feed mechanism influences the
partitioning of non-combustible substances between bottom ash and fly ash.
The temperature of the flue gas affects the relative amounts of volatile
trace elements which are emitted condensed on fly ash particles compared to
being emitted as vapors.
3.2.2  Coal-fired Boilers
     The different types of coal-fired boilers are described in this section
and uncontrolled emission factors for each boiler type are discussed.  Mass
and energy balances are presented for representative boilers.  The mass and
energy balances were developed from the combustion and flue gas information
                           37
developed by Devitt, et al.    The combustion information considered
includes fuel rates, excess air percentages, and fuel analysis.
     The fuel input rate was computed from the specified heat input rate and
fuel heating value, while the theoretical (no excess air) combustion air
requirement per unit mass of fuel burned was calculated from the fuel
analysis.  The actual mass rate of combustion air supplied to the boiler was
then determined from the fuel input rate and a specified excess air
percentage.  Pollutant emission rates were computed from the emission
factors, the heat input rate, and the fuel ash and/or sulfur content, as
appropriate.  The bottom ash discharge rate was taken to be the difference
between the ash input rate (from the fuel analysis and fuel input rate) and
the fly ash emission rate.  The flue gas rate was then calculated by
subtracting all the mass emission and discharge rates from the sum of fuel
and total combustion air mass input rates.
                                    3-24

-------
     For all  of the coal-fired units,  a high sulfur,  high-ash  coal  was  used
as the fuel  to develop the mass balances.   The ultimate  analysis  of this
coal, along  with a representative low-sulfur, low-ash coal,  is presented  in
Table 3-10.
     Similarly, energy balances were calculated using boiler heat input
rates and typical efficiencies provided by Devitt,  et al.    Boiler energy
losses which include the flue gas losses and boiler radiative  and convective
losses were  combined and treated as one "loss" term.
     All coal-fired industrial boilers have common  characteristics.   Coal
storage and  handling are necessary at the boiler site to ensure that an
adequate supply of fuel is on hand and that the fuel  is  ready  for
combustion.   For pulverized coal-fired units, this  involves  crushing and
grinding the coal to the proper consistency, and for stoker  units it
involves crushing and screening the coal to acceptable size.  Coal-fired
units require ignition with either oil or gas, and  many  are  designed to fire
oil or natural gas as a backup fuel.
     Excess  air is necessary for proper combustion, but  too  much  can be
detrimental  to the performance of the combustion system.  The  detrimental
effects of too much combustion air include:
       •  Reducing combustion temperatures and retarding the combustion
          rate;
       •  Reducing thermal efficiency, thus requiring more fuel for a given
          steam output; and
       •  Increasing gas velocities in the furnace  causing transport of fuel
          particles out of the furnace before complete combustion.
The effects  of too much combustion air on uncontrolled PM emissions are most
significant  if it is injected as undergrate air.  Increasing undergrate air
directly affects the upward furnace gas velocities  and increases  fuel and
particle entrainment.
     As mentioned earlier, cast iron industrial boiler emissions  are not
examined because of the small size and minimal installed capacity of cast
iron units.   The emissions from the remaining two types of coal-fired
industrial boilers, watertube and firetube, are discussed in the  following
                                    3-25

-------
CO
I
ro
                     Table 3-10.   ULTIMATE  ANALYSIS OF CjpAL  SELECTED FOR THE
                                  REPRESENTATIVE BOILERS38
                                            Composition,  percent by weight                    Heating
                       	   value,
                                                                                               kJ/kg
             Fuel         Water    Carbon   Hydrogen  Nitrogen   Oxygen    Sulfur      Ash     (Btu/lb)
              High-
              sulfur,
              high-
              ash
              coal
8.79     64.80
4.43     1.30     6.56      3.54     10.58
 27,447
(11,800)
              How-
              sulfur,  20.80
              low-
              ash
              coal
         57.60
3.20     1.20    11.20      0.60      5.40
 22,330
 (9,600)

-------
subsections.  Following these two subsections,  emission factors for coal-
fired boilers are presented.
     3.2.2.1  Vlatertube Boilers.   A watertube boiler is one in which the hot
combustion gases contact the outside of the heat transfer tubes, while the
boiler water and steam are contained within the tubes.   The tubes are
interconnected to common water channels and to  a steam outlet or outlets.
     Watertube boilers can generate high-pressure,  high-temperature steam,
up to 12,000 kPa (1740 psi) and 810 K (1000°F), and are available in many
sizes (see Table 3-4).  The tubes are of relatively small  diameter, 5 cm
(2.0 inch), providing rapid heat transfer, good response to steam demands,
                    39
and high efficiency.
     There are two main types of coal-fired watertube boilers:   pulverized
coal and stoker-fired.  Industrial size pulverized  coal units range from
29.3 MW to over 200 MW (100 to 700 x 106 Btu/hr) heat input,14 and burn the
coal in suspension.  A stoker is a conveying system that serves both to feed
the coal into the furnace and to provide a grate upon which the coal  is
burned.  Since feed rates by stoker units are limited,  stokers are generally
used on units rated at less than 117 MW (400 x  106  Btu/hr)  heat input.14
The three main types of stoker furnaces are spreader, overfeed or chain-
grate, and underfeed.
     Pulverized Coal-Fired
     In pulverized coal-fired boilers the fuel  is pulverized to the
consistency of light powder and pneumatically injected through the burners
into the furnace.  Combustion begins at the burners and continues into the
furnace volume.  Wet-bottom furnaces are designed to operate at high
temperatures and therefore keep the ash in the  molten state until it
collects in the bottom ash hopper.  Dry-bottom furnaces, on the other hand,
operate at lower combustion temperatures; consequently, the bottom ash
remains in the solid state.  Wet-bottom units are not expected to be
                                    40
manufactured and sold in the future.    Figure  3-3  illustrates a typical
58.6 MW (200 x 10  Btu/hr) heat input, dry-bottom pulverized coal unit with
the corresponding mass and energy balances.  Thirty percent excess air and a
                                   3-27

-------
     KEY
                                        RADIATIVE. CONVECTIVE,  AND STACK LOSSES
                                        10.5
                                        (35.8 i 10" Btu/hr)
            MASS  FLOW STREAM
            ENERGY FLOW STREAK
STEAM
OUTPUT	
'8.1 MW
(164.2 » 106 Btu/hr)
                                         HC,  I kg/hr
                                            (2 Ib/hr)
                                         NO  69 kg/hr
                                           1 (152 Ib/hr)
                                         SO  517 kq/hr
                                            (1UO Ib/hr)
                      FLUE GAS
                      97018 kg/hr
                      (213439 Ib/hr)
FLYASH  651 kg/hr
     (1437 Ib/hr)
                                                           7700 kg/hr
                                                           (16940 Ib/hr)
                                                         FUEL INPUT"
                                                           S3. 6 MW
                                                          (200 x 10"  3tu/hr
                                      I
                                  BOTTOM
                                  ASK
                                  164 kg/hr
                                  (360 Ib/hr)
      COMBUSTION AIR
      90715 kg/hr
      (199.572 Ib/hr)
      ?Flue qas  is  defined here as major components only  (H20,  C0?,  0-, N?).
      °Fuel is high  sulfur eastern coal.
        3-3.   Mass and  energy  balances  for a,58.6 MW  (200 x 10°  Btu/hr)
                pulverized  coal-fired boiler.
                                         3-28

-------
boiler efficiency of 82.1 percent were used to compute the mass and energy
         42
balances.
     Spreader Stoker
     The spreader stoker combines suspension burning and a thin,  fast-
burning fuel bed on a grate.  The modern spreader stoker, as shown in
Figure 3-4, consists of feeder units (arranged to distribute fuel  over the
grate area), a grate (which may be stationary or moving), forced-draft
systems for both undergrate and overgrate air, and combustion controls to
coordinate air and fuel supply.
     Some spreader stokers use a fly ash reinjection system, where the fly
ash removed in a downstream control  device is reinjected into the  boiler.
This technique tends to increase carbon utilization and boiler efficiency
                    43
(up to 2-3 percent).    However, it  also increases corrosion and  slagging in
the boiler and increases uncontrolled PM emissions.  Fly ash reinjection  was
quite popular 10 years ago, but recent boiler designs have increased carbon
utilization to the point where the advantages no longer outweigh  the
                           40
disadvantages in new units.
     Traveling-grate spreader stokers are generally installed with one large
plenum or air chamber under the entire grate surface.  Overfire air systems
are useful in promoting good combustion and reducing the formation of  smoke,
especially for lower loading rates.   The spreader stoker boiler shown  in
Figure 3-4 has a capacity of 44 MW (150 x 106 Btu/hr) heat input.   The mass
and energy balance is based on an overall boiler thermal efficiency of
81 percent and 50 percent excess air, including 5 percent overfire air.
     Overfeed (Chaingrate) Stoker
     Overfeed stokers are generally  equipped with chain or moving  grates.
In addition, they have refractory arches or overfire air jets to  improve
combustion.  This type of stoker is  now usually designed for forced draft
operation; natural draft designs are gradually becoming obsolete.   Chain-
      st
      14
grate stokers are generally less than 73.3 MW (250 x 106  Btu/hr)  heat
input.
     As shown in Figure 3-5, coal  is fed from a hopper onto a  moving grate
and enters the furnace after passing under an adjustable grate that
                                    3-29

-------
                   'AOIAT1VE, CDKYECTIVE. MO STACX LOSSES: e « m
                      ;

                        (U Ib/hr)
                                                    3S.6 Ml
                                             	• (121.3 • 10« Btu/hr)
                                           .^1 STEAM OUTliT
                                                     ?UE. INPUT:  5771 kg/hr
                                                             (12.696 Ib/hr)


                                                       FUEL INPUT :b

                                                  !     (ISO i 10s Btu/hr)
                                                     OYERFIRE AIR:
                                                         3881 kg/hr
                                                             Ib/hrI
                                                        UNOERF1RE AW:
                                                            '3.74Z kg/hr
                                                            (162.232 iD/hr)
                                 BOnON ASH:
                                     210 ka/hr
                                     («1 Ib/hr)
              MASS FLOu STREAM
                  ' FLOW STREAM
?Flue gas is defined  in  this  figure as major components (C02t H,,0t N2, 02).
 Fuel is high sulfur  eastern  coal.
Figure 3-4.    Mass  and energy balances  for  44  MW  (150  x  10   Btu/hr)
                   coal-fired spreader  stoker.4b
                                     3-30

-------
                                                        .  RADIATIVE. CONVECTIVE. AND STACK LOSSES --
                                                        T  4.4 Hi   .
                                                        I  (15.1 « «>5
CO
    NO. - II  kg/hr
         (24  Ib/hr)

    SO, - 193 kg/hr
         (425 Ib/hr)
FLUE
41078 kg/hr
(90371 Ib/hr)
                                   SUAM -
                             HC --
                              C
                                                                   ' '»/•"•
                                                                   (3 Ib/hr)
                                                                               KEY:
                                           	»- MASS FLOW STREAM

                                           	>- ENERGY FLOW STREAM
                                                              CO -  3 kp/hr
                                                                   (6 Ib/hr)

                                                              FLYASH -  76 kg/hr
                                                                     (167 Ib/hr)
                                                                                      OVERFIRf-AIR
                                                                                         NOZZLES
                                                                                      1936 kg/hr
                                                                                     (4259 Ib/hr)
          ,7 6 m                AIR-CONTROL DAMPERS

          (59.9 x 106 Btu/hr)           /        . /  .
BOTTOH ASH
229 kg/hr
(S04 Ib/hr)
                                                                 pm urn
                                               RETURN
                                                BEND
                                                                             ?H HTG HTCI DTD:
                                                                                       FUEL INPUT'-'—

                                                                                       (75* lO6 Btu/hr)
                                                                                   COAL
                                                                                  HOPPER         K
                                                                                       FUEL  INPUT —
                                                                                       2806  kg/hr
                                                                                       (6348 Ib/hr)
                    DRAG
                    PLATE
UNDERFIRE AIR - 36.783 kg/hr
             (80.924 Ib/hr)
STOKER
CHAIN
 DRIVE
SPROCKET
                                                                                HYDRAULIC
                                                                                  DHIVC
                                  ?Flue gas is defined in this figure as major components (C02> H20, N2. 02).
                                  5Fuel is high sulfur eastern coal.
                                              Figure  3-5.  Mass and  energy  balances  for  a  22  MW                47
                                                               (75  x  106 Btu/hr) coal-fired  chaingrate  stoker.

-------
regulates the thickness of the fuel bed.  Combustion is completed by the
time the coal reaches the far end of the grate, and the remaining ash is
discharged into the ashpit.
     Figure 3-5 shows a mass and energy balance for a typical 22 MW
(75 x 10  Btu/hr) heat input chaingrate stoker. The calculations are based
on 80 percent efficiency and 50 percent excess air, with overfire air
     nting for 5 perce
     Underfeed Stoker
accounting for 5 percent of the total combustion air.
     Various types of underfeed stokers are used in industrial boiler
applications.  They vary depending on whether the coal is fed horizontally
or by gravity, whether the ash is discharged from the end or the sides, and
how many retorts, or channels through which the coal is fed, are contained
in the boilers.  Underfeed stokers can burn a wide range of coals, including
caking coals and anthracite.
     In the side-discharge, horizontal underfeed stoker shown in Figure 3-6,
coal is fed intermittently to the fuel bed by a ram.  In very small units,
the coal is fed continuously by a screw.  The coal moves in a retort, and
air is supplied through tuyeres on each side and through openings in the
side grates.  Single or double retort units are generally less than 73 MW
(250 x 106 Btu/hr) heat input.
     Overfire air is commonly used with underfeed stokers to provide some
combustion air and turbulence in the flame zone directly above the active
fuel bed.  The air is provided by a separate overfire-air fan and is
injected through small nozzles in the furnace walls.
     An efficiency of 78 percent, a heat input of 8.8 MW (30 x 106 Btu/hr),
and 50 percent excess air was used for the material and energy balances on
            49
this boiler.    Overfire air accounted for 5 percent of the total combustion
air.
     3.2.2.2  Firetube Boilers.  In firetube boilers, the products of
'-nmbustion flow through tubes that are surrounded by water.  These units
range in size from 0.1 to 5.9 MW (3.0 to 20 x 10  Btu/hr) thermal input and
are used primarily for heating systems, to produce industrial process steam,
and as portable power boilers.
                                     3-32

-------
                                            HC:
                                            NO
                                                      2 kg/hr
                                                     (4 Ib/hr)
         KEY:
                 MASS FLOW STREAM

                 ENERGY FLOW STREAM
                                                      5 kg/hr  ,,
                                                  " (11 Ib/hr)
                                                 FLUE GAS: 3
                                                   16401 kg/hr
                                                 (36082 Ib/hr)
L J
 CO:     -   .
    (2 Ib/hr)

,30,:  77 kg/hr
    (170 Ib/hr)

 FLYASH:
      30 kn/hr
     (67 Ib/hr)
                                        r
                                                                STACK
                    OYERFIRE
                         AIR-
                                        773 kg/hr
                                       (1700 Ib/hr)
FUEL
INPUT
1155 kg/hr
(2540 Ib/hr)
                                                        BOTTOM
                                                        ASH
                                                         92 kg/hr
                                                       (203 Ib/hr
                                                         <\ \
                                   PUSHER'BLOCKS

                                           AIR CHAMBER
                        STEAM:  7.2 MU
                               (24.5 8tu/hr)
                                                  UNOERFIRE AIR:
                                                  14685 kg/hr
                                                  (32308 Ib/hr)
                            RADIATIVE. CMVECTIYE. AND STACK	—
                                                LOSSES:  1.9 MW
                           FUEL INPUT:    ,       (6.6 x 106 Btu/hr)
                                8.8 MU    I
                        (30 x 106 Btu/hr)
   ?Flue qas is defined here as major components only (H20,  C02> 02, NZ).
    Fuel is high sulfur eastern coal.
Figure 3-6.   Mass and ..energy  balances  for  an  8.8 MW
                 (30 x  10° Btu/hr) underfeed stoker.™
                                      3-33

-------
     Firetube boilers are generally used where steam/hot water demand can be
maintained relatively constant because they are susceptible to structural
failure when subjected to large variations in steam demand.    Over
90 percent of the installed capacity of firetube boilers is oil- or
gas-fired.
     Six different types of firetube boiler configurations are commonly
used.  They include horizontal return tubular (HRT), Scotch marine,
vertical, locomotive, short firebox, and compact boilers.  The three
configurations used most are the HRT, Scotch marine, and vertical  units,
while the most common firing mechanism is the underfeed stoker.
     The feed and burner types differ between coal-fired and oil-fired units
for each of the boilers.  Scotch marine and HRT boilers are fired by all
types of fossil fuel, but firing with coal requires increased maintenance to
                              52
overcome scaling and slagging.    Since the majority of firetube boilers
burn oil or gas, this type of boiler will be discussed further in
Section 3.2.3 on oil-fired boiler emissions.
     3.2.2.3  Emission factors for coal-fired boilers.  Table 3-11 presents
emission factors for the various types of pulverized and stoker coal-fired
watertube industrial boilers discussed previously.  These factors were taken
                                                                          ?fi
from the U.S. EPA's compilation of Air Pollution Emission Factors (AP-42),
except for the NO  emissions from underfeed and overfeed stokers which came
                 )\
from Reference 53, and the trace element emission factors, which were taken
from Reference 54.  The factors in Table 3-11 will be used throughout this
study to represent uncontrolled emissions from industrial watertube boilers.
     AP-42 lists no emission factors for coal-fired firetube boilers.
Statistically reliable data on emissions from firetube boilers are not
available, as only limited testing has been performed.  The factors
presented in Table 3-12 are based on two tests on a 3.2 MW (11 x 10  Btu/hr)
heat input, underfeed stoker of unspecified tube configuration.  These small
units have higher CO emissions than watertube boilers but approximately the
same NO  and particulate emissions as underfeed watertube stokers.
       A
     As noted earlier, uncontrolled PM emissions from coal-fired boilers
depend primarily on fuel ash content, firing mechanism, and boiler load.
                                     3-34

-------
                                    TABLE 3-11.
UNCONTROLLED EMISSION FACTORS FOR COAL-FIRED a
       WATERTUBE  INDUSTRIAL BOILERS
            [ng/J  (lb/106  Btu)]
CAi
01
Pollutant
Partlculate Matter (PM)b
Sulfur Dioxide (S02)c
Nitrogen Oxides (N0x)
Carbon Monoxide (CO)
Hydrocarbons (HC)
Arsenic
Lead
Cadmium
Underfeed Stokers
2.93 - 29t3 MM
(10-100 x 106 Btu/hr)
91.2 A
(0.212 A)
693 S
(1.61 S)
150
(0.349)
182
(0.424)
54.6
(0.127)
3.4
(0.0079)
3.7
(0.0086)
0.20
(0.00046)
Chalngrate Stokers
2.93 - 29.3 MW
(10-100 x 106 Btu/hr)
91.2 A
(0.212 A)
693 S
(1.61 S)
140
(0.326)
182
(0.424)
54.6
(0.127)
3.4
(0.0079)-
3.7
(0.0086)
0.20
(0.00046)
Spreader Stokers
2.93 - 73.3 MW
(10-250 x 10° Btu/hr)
237 Ad
(0.551 A)
693 S
(1.61.S)
274
(0.636)
36.5
(0.0848)
18.2
(0.0424)
3.4
(0.0079)
3.7
(0.0086)
0.20
(0.00046)
Pulverized Coal
29.3 MW
( 100 x 106 Btu/hr)
292 A
(0.678 A)
693 S
(1.61 S)
274
(0.763)
18.2
(0.0424)
5.46
(0.0127)
3.4
(0.0079)
3.7
(0.0086)
0.20
(0.0004G)
            'Sources - Reference  26 for PH, SO-, NO , CO, and  IIC except.  Reference 62 for NO  from underfeed and
             chalngrate stokers,  and Reference  78 fOr arsenic, lead, and cadmium.           x

             Reference 26 expresses emissions on a Ib/ton fuel burned basis.  A conversion factor of 27,477 kj/kg
             (11,000 Btu/lb)  was  used to convert factors to a  heat Input basis.  Emission factors should  be
             adjusted for fuels with heating values different  from this value.

             A  Is  the weight  percent ash In coal.

            CS  Is  the weight  percent sulfur In  the coal.

             Flyash relnjectlon Increases PH emissions by 1.54.

-------
TABLE 3-12.  UNCONTROLLED EMISSION FACTORS FOR COAL-FIRED,,
             UNDERFEED STOKER FIRETUBE INDUSTRIAL BOILERS0*
                                  Emission factors,
         Pollutant                ng/J (lb/106 Btu)


     Participates3                  50A  (0.12A)

     Sulfur oxidesb                 692S (1.61S)

     Nitrogen oxides                177  (0.41)

     Carbon monoxide                261  (0.61)

     Hydrocarbons                       c


aA is the ash content of the fuel in weight percentage.
 this emission factor is based on a coal with a heat
 content of 27,447 kJ/kg.   It must be adjusted for coals
 with different heat content.

 S is the sulfur content of the  fuel in weight percentage.

cNo factor available.
                             3-36

-------
Stokers generally have lower PM emissions  than  pulverized  coal-fired  units
because the coal  is burned on a bed,  which leads  to  less entrainment  of  PM
than suspension burning.   PM emission rates for spreader stokers  are  higher
than they are for the other two stoker types because partial  burning  of  the
fuel in spreader stokers  occurs while it is still  in suspension.  The type
of coal being fired has a uniform effect on PM  emissions for  all  types of
coal-fired boilers.  Firing coals with higher ash  content  results in  higher
PM emissions.  Ash fusion temperature also has  an  indirect effect on  PM
emissions.   Coals with high ash fusion temperatures are generally  fired in
dry-bottom units and emit higher levels of PM.
     PM emissions from coal-fired industrial boilers also  depend  on the
boiler load.  Limited test data indicate that mass emissions  of PM  on a  heat
                                                  55
input basis tend to decrease with decreasing load.    The  data are  scattered
and the rate of change of PM emissions varies from one boiler to  another so
that a general  correlation is not possible.  However, for  each boiler firing
type the general  trend of decreasing  PM emissions  with decreasing load
exists.
     The variation of particle size as a function  of boiler type  is shown  in
Table 3-13.  As can be seen, spreader and chaingrate stokers  emit coarser
particles (mass median diameters of 59 to 88 pm)  than do underfeed  stokers
and pulverized coal-fired units (mass median diameter of about 17 pro).
     Sulfur oxide emissions, as mentioned earlier, are directly proportional
                                                              OC
to the sulfur content of the fuel.  Emission factors from  AP-42   which  are
used in this study neglect differences in emissions  due to differences in
ash partitioning and sodium content in the fuel.   As noted earlier, they
assume that about 95 percent of the fuel sulfur is emitted as gaseous SCL
and SO-, with the remaining 5 percent adsorbed  on the fly  ash or  bottom ash.
Individual fuel characteristics will  result in  deviations  from these  values.
     According to Table 3-11, nitrogen oxides emission rates  are  lowest for
chaingrate and underfeed stokers, at 140 and 150  ng/J (0.33 and
0.35 lb/10  Btu), respectively.  Both spreader  stokers and pulverized units
emit almost twice as much NO  as chaingrate stokers.  Underfeed  and
                            /\
chaingrate stokers have very large fireboxes and  consequently lower
                                     3-37

-------
TABLE 3-13.  PARTICLE SIZE DATA FOR PARTICULATE EMISSIONS FROM
                                                                Cfi C7 CQ
             TYPICAL UNCONTROLLED COAL-FIRED INDUSTRIAL BOILERS   *'
                                 Particle mass median
            Firing method             diameter,
           Pulverized                    17
           Spreader stoker               59
           Chaingrate stoker             88
           Underfeed stoker              16
                                 3-38

-------
volumetric and surface heat release rates.   The  lower heat  release  rates
                                                                       59
reduce peak temperatures and,  hence, contribute  to  lower  NO  emissions.
                                                           A
In addition, the partial staged combustion  that  naturally occurs  in all
stokers due to combustion on fuel  beds  contributes  to reduced NO  emissions
                                        60                      ^
relative to pulverized coal-fired  units.
     Figure 3-7 shows how excess oxygen levels affect NO  emissions for the
                                                       A
various coal-fired boiler types discussed.   More information  on this subject
is presented in Section 4.3.7  of this report.  For  the typical pulverized
coal-fired unit discussed earlier, the  30 percent excess  air  being  fired
translates to approximately 5  percent excess oxygen on the  figure.
Similarly, for the three types of  stokers,  all with 50 percent excess air,
this translates to approximately 7 percent  excess oxygen.  Thus,  as can be
seen in Figure 3-7, an increase in excess oxygen from 5 to  6  percent for
pulverized coal-fired units leads  to roughly a 20 ng/J increase in  NO
                                                                     J\
emissions, while an increase in excess  oxygen from  7 to 8 percent for the
stokers leads to roughly a 25  ng/J increase in NO  emissions.
                                                 A
     Reducing boiler load tends to decrease combustion intensity  which in
turn tends to decrease NO  emissions.  However,  load reduction is typically
                         A
accompanied by an increase in  excess oxygen which may offset  the  decrease  in
NO  emissions.
  A
     Carbon monoxide and hydrocarbon emissions are  dependent  on combustion
efficiency.  Generally their emission rate, defined as mass emissions per
unit of energy input in ng/J,  decreases with increasing boiler size.  For
example, HC emission rates are lowest for the pulverized  coal units at
5 ng/J (0.01 lb/10  Btu); they are over three times higher  for spreader
stokers, and 10 times higher for underfeed  and overfeed stokers as  a result
of increasingly less efficient combustion.
3.2.3  Oil-Fired Boilers
     The different types of oil-fired boilers are discussed in this section
and uncontrolled emission factors  for each  boiler type are  discussed.  Mass
and energy balances are presented  for representative boilers. The  mass and
energy balances were developed for the oil-fired boilers  in the  same manner
as described in Section 3.2.2 for  the coal-fired boilers.  Representative
                                    3-39

-------
400
300
200
100
         PULVERIZED COAL BOILERS
                                                          UNDERFEED
                                                          AND TRAVELING
                                                          GRATE STOKERS
         I     I     I      I     I     I     .1     I      I     I     I     I
0123456789

                        EXCESS OXYGEN, percent
                                                       10    11    12
        Figure  3-7,  Effect of  excess  oxvqen  on NOX emission from
                      coal-fired boilers^
                                      3-40

-------
residual and distillate oil fuels were chosen to perform the  mass  balance.
The ultimate analyses of these fuels are summarized in Table  3-14.
     Oil storage and preparation is necessary at the boiler site to  ensure
that an adequate supply of fuel is on hand and ready for combustion.  For
distillate oil this may not require more than providing storage, but for
residual oil the fuel is usually heated to keep the viscosity low  enough for
pumping and proper atomization.  Oil-fired units are usually  ignited with
the primary fuel being fired, but may use natural  gas for ignition.
     Oil-fired industrial boilers can be classified into two  main  cate-
gories; watertube and firetube.  Each of these boiler types is  discussed in
the following subsections.  Following these subsections a separate
subsection on emission factors is presented.
     3.2.3.1  Watertube Boilers.   Since the general  characteristics of
watertube boilers are discussed in Section 3.2.2.1, only the  specifics of
oil-fired units are discussed in this subsection.   Oil-fired  watertube
boilers are generally less than 73.3 MW (250 x 106 Btu/hr)  heat input.
These boilers are subclassified according to the configuration  of  the heat
transfer tubes.  Straight watertube boilers are no longer manufactured,
having been completely supplemented by firetube boilers in the  smaller sizes
and bent watertube boilers in the larger sizes.  However, a large  number of
straight tube boilers are still in operation.  Both types of  boilers may
fire either residual or distillate oil and may be further classified
according to how the fuel is atomized (steam, air, or mechanical).
     Figure 3-8 illustrates a typical bent tube, oil-fired watertube boiler
and gives the corresponding mass and energy balance.  Mass and energy
balances are shown for a 44 MW (150 x 10  Btu/hr) heat input  residual
oil-fired unit, operating at 85 percent efficiency and 15 percent  excess
air.  Such a unit would typically include an economizer, which preheats  the
feedwater, and an air preheater, which heats the combustion air.
     3.2.3.2  Firetube Boilers.   Oil-fired firetube boilers  are
subclassified in the same manner as coal-fired firetube boilers.   These  six
subclassifications are horizontal return tubular (HRT), Scotch marine,
                                     3-41

-------
                         TABLE 3-14.  ULTIMATE ANALYSES OF RESIDUAL AND DISTILLATE OIL

                                            SELECTED FOR REPRESENTATIVE BOILERS38
CO
I
ro

Composition
percent by weight
Fuel Water Carbon
Residual 0.08 86.62
oil
Distillate 0.05 87.17
oil
Hydrogen
10.20
12.28
Ni trogen
0.3
0.05
Oxygen
Trace
Trace
Heating
value,
kJ/kg
Sulfur Ash (Btu/lb)
3.00 0.10 43,043
(18,500)
0.50 Trace 45,346
(19,500)

-------
                                 RADIATIVE, CONVECTIVE,  AND
                                 STACK LOSSES:        i
                                 6.6 MW  ,
                                 (23 x 10° Btu/hr)
      SO :   220 kg/hr
        x  (483 Ib/hr)
FLUE GAS ?' 61773 kq/hr
         (135901 Ib/hr)
    STEAM:      -*-
    37.4 MW  ,
    (127 x  10b   Reheat...
      Btu/hr)  Superheater
FUEL INPUT:
3680 kg/hr
(8096 Ib/hr)
,   Attemperator
b      !i
    Secondary
   Superheater
FUEL INPUT:
44 MW
          .b
               Burners'
(150 x 10° Btu/hr) j,
               KEY:
                             •MASS FLOW STREAM
                     	*• ENERGY FLOW STREAM
                                   CO  2 kg/hr
                                     (5 Ib/hr)

                                   HC  1 kg/hr
                                     (2 Ib/hr)
                                                                NO:  27 kg/hr
                                                                  x  (60 Ib/hr)

                                                                FLYASH:   16 ko/hr
                                                                        (35 Ib/hr)|
                                            COMBUSTION  AIR
                                            58357 kg/hr
                                            (128,386 Ib/hr
 ?Flue gas 1s defined In this  figure as  major components (02,  N2, C02, H20).
  Fuel is 3 percent  sulfur residual fuel  (Table 3-14).
           Figure  3-8.   Mass and energy  balances  for  a 44 MW
                           (150 x  106  Btu/hr) residual oil-fired
                           watertube boiler.
                                        3-43

-------
vertical, locomotive, short firebox, and compact boilers.  The HRT,  Scotch
marine, and vertical units are discussed below.
     Horizontal Return Tubular
     In a HRT boiler the firetubes are horizontal.  The fuel  firing
mechanism is at one end, and the products of combustion are recirculated or
"returned" to make two, three, or four passes through the tubes within a
water medium.  The boiler is encased with brick, and the furnace is  set on
rollers or suspended on hangers to allow for expansion and contraction.
     Scotch Marine
     Scotch marine boilers consist of a water-cooled furnace and firetubes.
The boiler and the furnace are housed in one continuous containment  shell;
fuel is burned in the lower half of the unit. The combustion gases first
pass through the furnace tube, heating the bottom of the water basin,  and
then pass through the firetubes, heating the water in the basin.  Scotch
marine boilers are available as two-, three-, or four-pass units.
     Scotch marine boilers are complete, compact, portable, packaged units.
Figure 3-9 shows a mass and energy balance for a typical distillate
oil-fired Scotch marine firetube boiler.  The calculations are based on a
4.4 MW (15 x 10  Btu/hr) heat input unit, operating at 15 percent excess air
                                                                  £O
(equivalent to 3 percent excess oxygen) and 80 percent efficiency.
     Vertical
     Vertical boilers are single-pass units with firetubes arranged
vertically up from the water-cooled furnace and may be either exposed-tube
or submerged-tube type.  These complete furnace and boiler units are small
and portable, requiring less space than comparable HRT or Scotch marine
boilers.
     Only the fuel feed mechanism and burners differ between the oil-fired
and the coal-fired types for each of these boilers.  Oil-fired burners for
firetube boilers employ the same atomization techniques as for watertube
units. Likewise, the same general effects on emission factors are noted for
both boiler types.  Firetube boilers may fire either residual or distillate
fuel oils.
                                    3-44

-------
                                ARADIATIVE, CONVECTIVE, AND
                                "STACK LOSSES:
                                10.9 MW (3 x 106 Btu/hr)
                                 I
                   SOX:
                   3.5 kg/hr
                   (7.7 Ib/hr)

                   FLUE GAS: a
                   6634 kg/hr
                   (14595 Ib/hr)
AGO A
  0.3
  (0.7
  NOv
AND HC:
 kg/hr
  Ib/hr)

   1.1 kg/hr
   (2.4 lb/hr)
AFLYASH:  0.1 kg/hr
          (0.2 Ib/hr)
 COMBUSTION AIR
 6290  kg/hr
 (13838  Ib/hr)

    BLOWER
FUEL INPUT:
349 kg/hr
(768 lb/hr)
FUEL INPUT:-'
4.4 MW
(15 x 106 Btu/hr)
        BURNER
                                                                                     --"-STEAM:
                                                                                          3.5 MW
                                                                                          (12 x 106  Btu/hr)
                                                                   MASS FLOW STREAM

                                                                   ENERGY FLOW STREAM
  Flue gas in  this figure means  major components  (00, N0, C00,  H00).
                                                      £.   i    2    d.
  Fuel is distillate oil.
                   Figure 3-9.  Cutaway view of  a  four-pass Scotch firetube  boiler.
                                                                                       64

-------
     3.2.3.3 Emission factors for oil-fired boilers. The emission factors
for oil-fired boilers are summarized in Table 3-15.   These factors are
typical of firetube as well as watertube oil-fired boilers in the given size
ranges.
     As noted earlier, participate emissions from residual oil-fired boilers
are expressed as a function of the sulfur content of the fuel.  Assuming a
fuel sulfur content of 3 percent (see Table 3-14), particulate emissions
from residual oil-fired units are greater by roughly a factor of 15 than for
distillate oil-fired boilers.  This is due, in part, to the lower carbon
residue content of distillate oil.
     Figure 3-10 shows particulate emissions as a function of fuel oil
carbon residue and illustrates the range of values that have been measured.
As can be seen from Figure 3-10, industrial boilers firing oil containing
little or no carbon residue emit from 5 to 20 ng/J of particulate matter.
     The type of atomizati'on has been shown to affect the amount of
particulate matter emitted.  One study on residual oil-fired units has  shown
that a mechanically atomized unit produces 20 times the particulate matter
that a comparable air-atomized unit does (9.9 ng/J to 186.2 ng/J), and  that
steam atomization produces roughly three times as much particulate matter as
air atomization (25.2 ng/J to 9.0 ng/J).    The same study indicates that
mechanical atomization is the most common method of atomization, thus
accounting for the fact that the emission factor in Table 3-15, which
represents the current mix of installed atomization techniques, is
96.0 ng/J, a factor close to the high end of the range.  This is based  on a
fuel sulfur content of 3 percent.
     As noted in Table 3-15, NO  emissions from oil-fired boilers are
                               A
subject to a wide variety of influences which can interact to affect
emission rates.  In general, boilers firing residual oil emit more NO  than
                                                                     A
equivalent boilers firing distillate oil.  Furthermore, the range of NO
                                                                       A
emissions is wider for boilers firing residual oil.  Both these trends  are
accounted for by the larger amount and higher variability of fuel nitrogen
in residual  oil.
                                    3-46

-------
         TABLE 3-15.   UNCONTROLLED  EMISSIONS  FACTORS  FOR
                        OIL-FIRED INDUSTRIAL BOILERS  d
                        [ng/J  (lb/106  Btu)]
                                                Fuel Type
Pollutant
Partlculate Matter (PM)b
Sulfur Dioxide (S02)b
Residual
29.1 S + 8.72
(0.0675 S + 0.0203)
456 S
(1.06 S)
Distillate
6.30
(0.0146)
447 S
(1.04 S)
Nitrogen Oxides (NOX)                   a                         a

Carbon Monoxide (CO)                  14.5                       15.8
                                     (0.0338)                  (0.0366)

Hydrocarbons (HC)                     2.90                       3.15
                                     (0.00675)                 (0.00732)

Lead (Pb)c                            0.065                    No data
                                     (0.00015)

aNO  emissions are strongly dependent on boiler  type,  fuel nitrogen
 level, amount of air preheat, and excess air.   Reference 66 gives the
 following ranges:

          Residual  - Firetube           111    to  170 ng/J   fi
                                        (0.257   to  0.395 lb/10b Btu)
                      Watertube         87.5    to  362 ng/J   K
                       (w/o air preheat)(0.203   to  0.841 lb/10° Btu)
                      Watertube         66.8    to  188 ng/J   fi
                       (w/air preheat)  (0.155   to  0.437 lb/10b Btu)

         Distillate - Firetube          96.5    to  107 ng/J   K
                       (w/air preheat)  (0.224   to  0.249 lb/10° Btu)
                      Watertube         44.7    to  59.5 ng/J  ,
                       (w/o air preheat)(0.104   to  0.138 lb/10° Btu)
                      Watertube         69.0    to  102 ng/J   K
                       (w/a1r preheat)  (0.160   to  0.237 lb/10b Btu)

 S 1s the sulfur content of the fuel  in weight  percent.  The emission
 factor for residual oil represents the current  mix of installed atomization
 techniques.
cBased on one test only (Reference 67).

dSources - Reference 68 for PM, S02,  CO, &  HC
           Reference 66 for NO .
           Reference 67 for Pb.

 Reference 68 expresses emissions on a lb/1000  gal. fuel burned basis.
 Conversion factors of 43,043 kJ/kg (18,500 Btu/lb) and  45,345  kJ/kg
 (19,500 Btu/lb) were used to convert factors to a  heat  input basis  for
 residual and distillate oil respectively.   Densities of 0.96 kg/x,
 (8.0 Ib/gal) and 0.84 kg/a (7.0 Ib/gal) were also  used.
                                   3-47

-------
 CO

D
 O
 i—I

 JD



 •-3

 CD
 00
 LU
 err
 —I
 rD
 o
 
-------
     Air-to-fuel  ratios (typically expressed  in  terms  of  excess oxygen)  tend
to affect NOX emissions from residual  oil-fired  units  more  than distillate
oil-fired units,  as shown in Figure 3-11.   In all  cases,  increased excess
oxygen leads to higher NO  emissions.   However,  the  rate  of increase with
                         A
increased oxygen  levels is larger for  the  heavier, residual  oils.  Section
4.3.7 discusses the variation in NO emissions with  respect to excess air
                                   A
levels in more detail.
     The effect of load variation on NO emissions is  similar to the effect
                                       A
on coal-fired boilers.  Available data indicate  both increases and decreases
in NO  emissions  at reduced loads depending on whether or not excess air is
                                    70
held constant during load reduction.
3.2.4  Natural Gas-Fired Boilers
     Natural gas-fired industrial boilers  are classified  in two main
categories, watertube and firetube. A mass and  energy balance, developed  in
the same manner as for the coal- and oil-fired boilers is given for a
representative natural gas-fired industrial boiler.   The  ultimate analysis
of natural gas selected to perform the mass balance  is presented in
Table 3-16. Uncontrolled emission factors  for these  boilers are presented  at
the conclusion of this section.
     3.2.4.1  Watertube Boilers.  Units firing natural gas  alone are
generally similar in design, but physically smaller, to those units firing
only oil for the  same output.  These units are generally smaller than
73.3 MW (250 x 106 Btu/hr) heat input.71
     3.2.4.2  Firetube Boilers.   Natural gas-fired firetube boilers are
similar to oil-fired firetube boilers  except  for the burner. As with
watertube boilers, natural gas does not have  to  be atomized to be fired in
the firetube units, and only occasionally  is  a spray gun mechanism  designed
for a natural gas burner.  These units are generally designed for less  than
7.3 MW (25 x 106  Btu/hr) heat input.72
     Figure 3-9,  introduced earlier, shows a  firetube boiler that can  be
fired with oil or natural gas, and a mass  and energy balance for  oil  firing.
Table 3-17 presents a comparable mass  and  energy balance for the  same unit
firing gas.  These calculations are based  on  a boiler with  a heat input of
                                     3-49

-------
     300
    (0.70)
VO
 o
 -5


 C

  *
 l/l
 1   200
 £  (0.47)
    100
    (0.23)
                                                   PS 300  (HIGH NITROGEN
                                                    NO. 5 OIL)
                                 J_
                                 5678

                                EXCESS OXYGEN, percent
                                   10   11
12
        Figure 3_n
Effect of excess  oxygen on  NOX emissions from
distillate and  residual oil-fired boilers61
                                       3-50

-------
                TABLE 3-16.  ULTIMATE ANALYSIS OF NATURAL GAS  SELECTED  FOR  THE  REPRESENTATIVE  BOILER38

Fuel
Natural 1
gas
Composition
percent by weight
Water Carbon Hydrogen Nitrogen Oxygen Sulfur Ash
0.02 69.26 22.67 8.05 Trace Trace 0
Heating
value,
- kJ/kg
(Btu/lb)
50,707
(21,800)
01

-------
        TABLE 3-17.   MASS  AND  ENERGY BALANCE FOR A NATURAL GAS-FIRED
                     FIRETUBE  BOILER3
  ENERGY BALANCE
      la
          Fuel  input:
      Out
          Steam output:
          Corrective,
          radiative,
          and stack losses:
4.4 MW (15 x 106 Btu/hr)
3.5 MW (12 x 106 Btu/hr)
0.9 MW (3 x 10^ Btu/hr)
  MASS BALANCE
      In
          Fuel in:
          Combustion air:
312  ka/hr (688 Ib/hr)
5741   kq/hr  (12660  Ib/hr)
      Out
          Stack Components:   Flyash:
                    Major Components:
           0.07 kg/hr (0.15 Ib/hr)
           0.005 kg/hr (0.01  Ib/hr)
           0.80 kg/hr (1.8 Ib/hr)
           0.12 kq/hr (0.28 Ib/hr)
           0.02 kq/hr (0.05 Ib/hr)
           6054 kg/hr (13,350 Ib/hr)
?Refer to Figure 3-9  for locations of each mass and energy flow.
 Major components are CO* O. N,  HO.
                                  3-52

-------
4.4 MW (15 x 10  Btu/hr), 15 percent excess  air,  and  an  80  percent
efficiency.
     3.2.4.3.  Emission Factors for Natural  Gas-Fired Boilers.  Typical
emission factors for natural gas-fired industrial  boilers  (watertube  and
firetube) are summarized on Table 3-18.   As  noted in  the table, NO
                                                                 A
emissions show wide variability depending on a variety of  factors.   In  the
case of natural gas-fired boilers, the major factors  are those associated
with the formation of thermal NO , since fuel  NO   is  usually  negligible in
                                A              A
natural gas (the nitrogen percentage shown in  Table 3-16 is free  nitrogen
and does not contribute to fuel NO  emissions).  As explained earlier,  peak
                                  A
flame temperatures and excess air are the major influences  on thermal NO
                                                                        A
formation.  Since air preheaters increase peak flame  temperatures, use  of
this device may increase NO  emissions up to twofold  (see  Table 3-18) for
                           A
watertube boilers.  Another influence on flame temperatures is boiler size.
Larger boilers tend to operate at higher flame temperatures than  smaller
ones, increasing NO  emissions.     The type of boiler (watertube or fire-
                   A
tube) does not appear to have a large effect on NO  emissions provided  other
                                                  A
factors are held constant.  The limited data available indicate no
difference between NO  emissions of firetube boilers  and those of small
                     X                              yg
single burner watertube boilers without air preheat.
     Excess air variations affect NO  emissions for natural gas-fired
                                    A
industrial boilers as shown in Figure 3-12.   Typical  natural  gas-fired
boilers operate at 15 percent excess air or 3 percent oxygen. The  effect of
excess air on watertube boilers with air preheaters is most significant,
with roughly a 20 percent increase in NO  emissions per 1  percent increase
                                        A
in excess oxygen.   Load variations appear to affect NO  emissions  from
                                                       A
gas-fired boilers more uniformly than they affect emissions from coal-  or
oil-fired boilers.  Reduction in NO  emissions occurs with load  reduction
                                   A
for natural gas-fired boilers, with the most significant effect being on
boilers with air preheat.
     SO  emissions from natural gas-fired boilers are very low due to the
       A
fact that natural gas generally contains less than 0.1 percent sulfur.
Sulfur-containing mercaptan, however, is added to natural   gas for detection
                                     3-53

-------
    TABLE 3-18.   UNCONTROLLED EMISSION FACTORS FOR NATURAL GAS-FIRED
                 WATERTUBE AND FIRETUBE INDUSTRIAL BOILERSa>C

                          [ng/J (lb/106 Btu)]
         Pollutant                                 Emission Factor


     Particulate matter (PM)                         2.05 - 6.58
                                                 (0.00508 - 0.0153)

     Sulfur dioxide (S0~)c                              0.262
                       L                              (0.000610)

     Nitrogen oxides (NO )
                        ^

     Carbon monoxide (CO)                                7.44
                                                       (0.0173)

     Hydrocarbons (HC)                                   1.31
                                                       (0.00305)


Reference 74 used for all factors except NO .  Units converted using a-
 heating value.of 50707 kJ/kg (21,800 Btu/lb? and density of 0.722 kg/nT
 (0.0451 lb/ftj).

 NO  emissions strongly dependent on boiler type, air preheat and
 excess air levels.  Reference 75 gives the following ranges:

     Firetube         - 28.6 - 55.1 ng/J (0.066 - 0.128 lb/106 Btu)
     Watertube w/o                                           fi
       air preheater  - 30.1 - 97.9 ng/J (0.070 - 0.228 lb/10° Btu)
     Watertube w/                                             fi
       air preheat    - 49.0 - 190.1 ng/J (0.114 - 0.444 lb/10° Btu)

cAssumes pipeline quality natural gas.
                                   3-54

-------
    150
   (0.35)
2   100
K  (0.23)
     50
   (0.12)
                                         .WATERTUBE BOILERS WITH
                                         'AIR PREHEATER
                                                                   NATERTUEE
                                                                   BOILERS
                                                                   WITHOUT
                                                                   AIR
                                                                   PREHEATER
                                    F1RETUBE BOILERS
I	I
I     I
                                        I
I
I     I
                                                             I      I
                  23456789
                                EXCESS OXYGEN,  percent
               10    11    12    13
        Figure  3-12.  Effect  of excess oxygen on  NO  emissions from
                        natural  gas-fired boilers.fil
                                        3-55

-------
purposes, leading to small amounts of SO  emissions along with the  fuel
                                        A
sulfur available.
3.3  EMISSIONS 'UNDER CURRENT REGULATIONS
     As previously discussed (Section 3.2), industrial  boilers emit a  number
of pollutants, including sulfur oxides (SO ), nitrogen  oxides  (NO ),
                                          A                      A
particulate matter (PM), and lesser amounts of carbon monoxide (CO),
hydrocarbons, and trace elements.  Of these pollutants, however,  only  NO  ,
                                                                       A
SO   and PM are directly subject to emission limitations under existing
  A
State or Federal regulations.  This section discusses existing regulations
(both State and Federal) to which industrial boilers are subject.
3.3.1  Existing Regulations
     3.3.1.1  Subpart D Emission Limits.  New fossil fuel-fired industrial
boilers with capacities greater than 73.3 MW (250 x 10   Btu/hr) are subject
to 40 CFR 60, Subpart D which limits NO , SO   and PM emissions.  Most  states
                                       A    A
have been delegated the authority to administer New Source Performance
Standards (NSPS) and therefore have incorporated the provisions of  Subpart D
as part of their State implementation plan (SIP) for new units with
capacities greater than 73.3 MW.
     The currently applicable NSPS mass emission limits are presented  in
Table 3-19.  Subpart D for fossil fuel-fired steam generators  excludes
facilities using lignite coals from the NO  standard.
                                          A
     Subpart D also specifies an opacity limit for all  boilers subject to
its provisions.  The opacity standard limits visible emissions to 20 percent
opacity except for one six-minute average per hour which may be up  to
27 percent.
3.3.2  State Emission Limits
     New boilers with capacities less than 73 MW are subject to State
emission limits for NO , SO  and PM.  Particulate emissions are typically
                      A    A
limited by both an opacity or visible emission limit and a mass emission
limit.  There is limited State regulation of NO    Only Illinois has a limit
                                               A
for CO emissions (no greater than 200 ppm at 50 percent excess air).
                                    3-56

-------
          TABLE 3-19.  SUBPART D EMISSION LIMITS FOR FOSSIL
                     FUEL-FIRED STEAM GENERATORS

                    [>73.3 MW (250 x 106 Btu/hr)]

Fuel
Coal
Fuel Oil
Natural gas
PM
ng/J (lb/106 Btu)
43 (0.10)
43 (0.10)
43 (0.10)
S0x
ng/J (lb/106 Btu)
520 (1.2)
340 (0.80)

N0xa
ng/J (lb/106 Btu)
300 (0.70)
130 (0.30)
86 (0.20)
Excluding  lignite.
                                   3-57

-------
     SIPs reflect local conditions and needs.  As a result,  industrial
boiler emission limits vary considerably from state to state.   In  addition,
State emission limits usually reflect the fuel mix in a particular state.
States that depend on natural gas or fuel oils for their energy needs
typically have more stringent PM and SCL emission limits than  coal  burning
states.  Mid-western states, with large reserves of high-sulfur, high  ash
coal, tend to have relatively lenient PM and S02 emission limits.
     The type of PM regulation used by the majority of states  is a sliding
scale emission limit across a capacity range, which becomes  more stringent
as the capacity of the units increases.  In states with this type  of PM
emission limit, the variable emission limit usually begins above the 2.9 MW
(10 x 10  Btu/hr) capacity.  For units with capacities of less than 2.9 MW
(10 x 10  Btu/hr), the PM emission limit is, in most cases,  258 ng/J
(0.60 lb/10  Btu).  The emission limit then becomes progressively  more
stringent to 73.3 MW (250 x 10  Btu/hr) capacity, at which point the
Subpart D requirements generally are more stringent than the SIP limits for
new boilers.  Therefore, for units with capacities above 73.3  MW
(250 x 106 Btu/hr), the PM emission limit is 43 ng/J (0.1 lb/106 Btu). A
summary of the SIP particulate emission limits is presented in Table 3-20.
     Unlike SIP PM emission limits, S02 limits do vary with  the type of fuel
fired.  There is usually a limit for coal-fired boilers and a  separate limit
for oil- or gas-fired boilers.  Within each fuel category, there is usually
a single SOp emission limit that applies across the capacity range and is
based on either SO,, mass emission limits or fuel sulfur content.  Tables
3-21 and 3-22 list the allowable SIP S02 emissions for new and existing
coal-fired boilers and for oil and natural gas-fired boilers,  respectively.
     NO  emissions from new boilers with capacities above 73.3 MW  (250 x
  fi    *
10  Btu/hr), as already mentioned, are subject to Subpart D.  EPA  has
delegated authority for implementation of this standard to most states, and
it is now part of each SIP.  Few states have any limitation  on NO   emissions
                                                                A
from new boilers smaller than those subject to Subpart D.  Table 3-23
summarizes the data for states with NO  emission limits different  from
                                      A
Subpart D.
                                     3-58

-------
                                    Table 3-20.   STATE PARTICULATE  REGULATIONS  (lb/106 Btu)77
              State
          Existing
           New
CO
01
           Alabama


           Alaska
           Arizona
           Arkansas


           California
E = 3.109II"0-589 (class 1).
  E = 3.109H"0'589 (class 2)
0.1 gr/scf
Ambient concentrations may
  not exceed  75 \ig/m3 above
  background
10 Ib/hr or 0.1 gr/scf generally;
  however,  each county has
  separate  regulations
E = 1.38lf°-44 (class 1),
  E = 3.109H"0'589 (class  2)
0.1 gr/scf
E = 1.02lf°-769
Ambient concentrations
  may not exceed 75 \ig/n3
  above background
10 Ib/hr or 0.1 gr/scf generally;
  however, each county has
  separate regulations
Colorado
Connecticut
Delaware
District of Columbia
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas

E = 0.5H""-"
0.2
0.3
E = 0.17455lf0'23522
Best available technology
or 0.1 lb/106 Dtu
E = l.ll.f°-202
None for coal and oil
logE = -0.23 logH - 2.0111
E = 5.18lf°'715
E = 0.87ir°-16
0.8 maximum
E = 1.026H"0'233
-fi — . » . . i i • 	 .j. /info r>j — /L.»\
E=0.5H'u-<°
0.1
0.3
E = 0.17455H"0'23522
Best available technology
or 0.1 lb/106 Btu
E = 1.5B4H"0-5
None for coal and oil
logE = -0.23 logll - 2.0111
E=5.1BH-°'715
E = 0.87H"0'16
0.6 maximum
E = 1.026H"0'233
(continued)
         E = emission  limit

-------
                                                           Table 3-20.   (continued)
CO
CTl
O
State
Louisiana
Maine
Maryland





Massachusetts
Michigan
Existing
0.6
E = l.OSlf0-256
Residual oil IK10
Residual oil 10<|l<50
Residual oil 5CXIK200
Residua) oil 200>H
Solid fuel H<200
Solid fuel ll>200
0.15
Pulverized coal < 100, 000



0.03 gr/scfd
0.025 gr/scfd
0.02 gr/scfd
0.02 gr/scfd
0.05 gr/scfd
0.03 gr/scfd

Ib steam/hr,
New
0.6
E = 1.08lf0'256
Residual oil IK10
Residual oil 10H
Solid fuel IK200
Solid fuel 10200
0.1
Pulverized coal < 100, 000



0.03 gr/scfd
0.025 gr/scfd
0.02 gr/scfd
0.01 gr/scfd
0.05 gr/scfd
0.03 gr/scfd

Ib steam/hr,
           Minnesota
           Mississippi
           Missouri

           Montana
           Nebraska
           Nevada
  3 Ib/1,000 Ib  flue gas;
  MOO,000 Ib steam/hr:
  y=6.75H-°'176
  y = lb/1,000 Ib  flue gas
Stoker coal
  0-100,000 Ib steam/hr,
  0.65 lb/1,000  Ib flue gas;
  100,000-300,000  Ib steam/hr,
  0.65-0.45 lb/1,000 Ib flue  gas
0.4
E = 0.89611
,-0.174
logE = -0.23299  logH
  +2. 1454
           °-159
  = 0.865l
E = 1.026H
          -0.233
                                          3 lb/1,000 Ib flue  gas;
                                          MOO,000 Ib steam/hr:
                                          y=6.75H-°'176
                                          y = lb/1,000 Ib flue  gas
                                        Stoker coal
                                          0-100,000 Ib steam/hr,
                                          0.65 lb/1,000 Ib flue gas;
                                          100,000-300,000 Ib  steam/hr,
                                          0.65-0.45 lb/1,000  Ib flue gas
0.4
E = 0.89611
,-0.174
E = 1.02H
        ,-0.2131
                                        logE = -0.3382 logfl
                                          +2.1454
                                        E = 1.026lf°-233
                                        E = 1.026H"0'233
                                        E = 1.02H-0'2131
              E =  emission  limit (lb/106 Btu);  H =  heat  input (TO6 Btu/hr)
                                                  (continued)

-------
                                      Table 3-20.  (continued)

State
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
a\
Lvot
Rhode Island
South Carolina
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
Existing
E = 0.865H"0'159
E = 2.38lf0'598
E = 0.96135lf°-23471
E = 1.02M"0'29
E = 1.09H-°-2594
E = 0.811.r°-131
a
E = LOW'0-259
0.2 gr/scf
0.4 for <50 x 10s Btu/hr and
E = 3.6H~°'56 for >50 x 10« Btu/hr
0.2
0.6
E = 1.09H'0-2549
0.3
E = 1.58H~°'S
E = 1.58H"0'5
E = 0.8425H"0'2314
0.2 gr/scf
E = 0.706II"0-317
E = 0.87lf0-16
E = 0.896ir°-174

E =
E =
E =
E =
E =
E =

E =
0.1
0.4
E
0.2
0.6
E =
0.3
E =
E =
E =
0.1
E =
E =
E =
New
0.985H-0'215
2^38.r°-598
0.96135H-0'23471
1.02.r°-219
L09|,-0.2594
0.811H'0-131
a
1.09tf0'259
gr/scf
for <50 x 10" Btu/hr and
= 3.6H"0'56 for >50 x 10° Btu/hr


216,,-0.5566

1.58H'0-5
1.58H-0'5
0.8425H-0'2314
gr/scf
0.706ir°-314
0.87H-0'16
0. 896ir°- 174
dFor coal and residual oil, the emission limit varies between 0.1  and 0.4 lb/10  Btu  depending on capacity
 and boiler location.  For distillate oil and natural gas, the limit is 0.02 lb/10" Btu.

E = emission limit (lb/106 Btu); H = heat input (106 Btu/hr).

-------
                                          Table 3-21.
               State
                                              STATE S02  REGULATIONS FOR COAL-FIRED
                                              BOILERS  (lb/106 Btu)77
                                            Existing
                                                                       New
CO
a\
PO
Alabama
Alaska
Arizona
Arkansas

California

Colorado
Connecticut
Delaware
District or Columbia
Florida

Georgia
4.0
500 ppm
1.0
<0.2 ppn ambient (assuming no
  control)
200 Ib/hr, 0.2X by volume,
  1,000 pom, or 0.5X S
<500 ppm
0.55
IX S
0.5X S
Latest available control technology
  or 6.17 for solid fuel
2.5X S <200 x 10* Btu/hr and 3X for
  >200 x 106 Btu/hr
            E - emission limit  (lb/106  Btu);  H =  heat  input (106  Btu/hr)
            % denotes  maximum level sulfur  that can be burned
1.8
500 ppm
0.8
<0.2 ppm ambient  (assuming no
  control)
200 Ib/hr, 0.2X by volume,
  1,000 ppm, or 0.5X S
<500 ppm
0.55
IX S
0.5X S
Latest available  control  technology
  or 6.17 for solid fuel
1.2X S
Hawaii
Idaho
Illinois
Indiana


Iowa
Kansas
<2X S
<1.0X S
6.0
6.0 for IK23,
E = 17. OH"0*33 for 233,000
5.0
No regulation below
250 x 10* Btu/hr

-------
                                            Table 3-21.  icontinuea;
CT>
CO

State
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Existing
E = 9.464lf0'3740
Meet ambient regulations
(assuming no control)
<2.5X S
Low sulfur
0.55
0.2
2% S
2.4
2,000 ppm
<1
2.5% S
0.7
1.5
2.0
No regulations
2.8
2.3
3.0
5.70
Meet ambient regulations
IX S <150 x 10° Otu/hr,
1.6 > 150 x 10* Btu/hr

E = 9.46H~°-
Meet ambient
(assuming
<2.5X S
Low sulfur
0.28
2.4
1.5X S
2.4
500 ppm
<1
2.5X S
0.7
1.5
1.5
New
3740
regulations
no control)












No regulations
2.8
1.6
3.0
5.70
1.2
IX S <150 x
1.6 >150 x





10° Btu/hr,
10* Btu/hr
        E  =  emission  limit (lb/106);  H = heat input (106  Btu/hr)
        %S denotes  maximum fuel  sulfur that can be burned.
                                                                               (continued)

-------
                                               Table 3-21.  (continued)
CO
I
State
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
Existing
3.0 for <50 x 10e Btu/hr and
E » 5. Iff0' 14 for >50 x 106 Btu/hr
$0.55
3.5
3.0
4.0
3.0
IX S
2% S
2.64
2,000 ppm
3.2
1.5% S
No regulations for <250 x 10" Btu/hr
New
1.0 for <50 x 10e Btu/hr and
E = 17H~°'14 for >50 x 10s Btu/hr
SO. 55
2.3
3.0
1.6
3.0
IX S
2X S
1.06
1,000 ppm
1.6
1.5% S
No regulations for <250 x 10s Btu/hr
            E = emission limit  (lb/106 Btu); H = heat input (TO6 Btu/hr)
            %S denotes maximum  fuel sulfur that can be burned.

-------
                                Table  3-22.   STATE S00 REGULATIONS  FOR OIL- AND GAS-FIRED BOILERS
                                                (lb/106  Btu)77
                 State
                                              Existing
                                                                         New
cn
en
           Alabama
           Alaska
           Arizona
           Arkansas
California

Colorado
Connecticut
Delaware
District of Columbia
Florida
Georgia

Hawaii
Idaho

Illinois

Indiana
4.0
500 ppm
2.2
<0.2 ppm ambient
  (assuming no control)
200 Ib/hr,  0.2% by volume,  1,000 ppm,
  or 0.5% S
500 ppm
0.55
1.0% S
0.5% S
Latest technology or 2.75
2.5% for <100 x 106 Btu/hr  and
  3.0% for  >100 x 106 Btu/hr
No regulations
1.75% S for residual oil  and
  0.5% S for distillate oil
1.0 for residual oil and
  0.3 for distillate oil
6.0 for <23 x 106 Btu/hr,
  E = 17.OH"0'33 for
  233,000 x 10' Btu/hr
1.8
500 ppm
0.8
<0.2 ppm ambient
  (assuming no control)
200 Ib/hr, 0.2% by volume,
  1,000 ppm, or 0.5% S
500 ppm
0.55
0.3% S
0.5% S
Latest technology or 2.75
2.5% for <100 x 10« Btu/hr and
  3.0% for >100 x 106 Btu/hr
No regulations
1.75% S for residual oil and
  0.5% S for distillate oil
1.0 for residual oil and
  0.3 for distillate oil
6.0 for <23 x 10e Btu/hr,
  E = 17.OH"0'33 for
  233,000 x 108 Btu/hr
(continued)
           E  =  emission limit
           %S denotes maximum
                        (lb/10°  Btu);  H =  heat  input (106 Btu/hr)
                        fuel  sulfur that can be burned.

-------
                                              Table 3-22.  (continued)
State
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
oo Mississippi
en Missouri
Montana

Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Existing
2.5
No regulations
E = 5.6484lf°-354
2,000 ppm
<2.5% S
Low sulfur
0.55
2.2
1.75
2.4
2,000 ppm
1.0 for oil and
5 gr/100 ft3 for gas
2.5
0.7
1.5
2.0
0.34
2.8
2.3
3.0
3.2
New
2.5
No regulations
E = 5.6484U"0'354
2,000 ppm
<2.5% S
Low sulfur
0.28
1.7
1.75
2.4
500 ppm
1.0 for oil and
5 gr/100 ft3 for gas
2.5
0.7
1.5
1.5
0.34
2.8
1.6
3.0
1.0
E = emission limit  (lb/106 Btu); H =  heat  input  (106  Btu/hr)
% S denotes maximum fuel sulfur that  can be  burned.
(continued)

-------
                                                               Table  3-22.    (continued)
              State
                                               Existing
                                                                            New
CO
at
Oklahoma
Oregon

Pennsylvania

Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin

Wyoming
0.8 for oil  and 0.2 for gas
1.75% S for  residual oil and
  0.5% S for distillate oil
3.0 for <50  x 10* Btu/hr and
  E = 5. Ill"0'14 for >50 x 109 fltu/hr
SO. 55
3.5
3.0
4.0
440 ppm
1.5% S
2% S
2.64
2,000 ppm
3.1
1.0% S for residual oil and
  0.7% S for distillate oil
No regulations
  for <250 x 10* Btu/hr
0.8 for oil and 0.2 for gas
1.75% S for residual oil
  and 0.5% S  for distillate  oil
1.0 for <50 x 106 Btu/hr
  and E =  1.7H~°'14 for >50  x 109 Btu/hr
$0.55
2.3
3.0
1.6
440 ppm
1.5% S
2% S
1.06
1,000 ppm
1.6
1.0% S for residual oil  and
  0.7% S for distillate  oil
No regulations
  for <250 x 10* Btu/hr
         E =  emission limit  (lb/106 Btu); H  = heat input  (106  Btu/hr)
       %S denotes maximum fuel  sulfur  that can be burned.

-------
           TABLE 3-23.  STATE NOX EMISSION LIMITS THAT DIFFER
                        FROM SUBPART D?7
NO emission limit,
ng/J (lb/106 Btu)
State
Florida
Oklahoma
Wyoming
New Mexico
California9
Coal
300
300
300
194
225
(0.
(0.
(0.
(o.
ppm
7)
7)
7)
45)

130
130
130
130
225
Oil
(0.
(0.
(0.
(0.
ppm

3)
3)
3)
3)

Gas
86 (0.
86 (0.
86 (0.
86 (0.
125 ppm

2)
2)
2)
2)

Capacity
(106 Btu/hr)
>50
>50
An
>50
All
California regulations vary with each control region; values given (at 3
percent Op) are typical of emission limits for facilities other
than those covered by Subpart D.
                                    3-68

-------
3.4  REFERENCES

1.   Devitt, T.  (PEDCo Environmental, Inc.)  Population  and  Characteristics
     of Industrial/Commercial  Boilers in the  U.S.   (Prepared for
     U.S.  Environmental Protection Agency.)   Research  Triangle  Park,  N.C.
     Publication No.  EPA-600/7-79-178a.   August 1979.  pp. 32-33.

2.   Steam, Its Generation and Use, 38th Edition.   New York, Babcock  &
     Wilcox, 1975.   p. 25-1.

3.   Schwieger, B.   Industrial Boilers - What's Happening Today.   Power.
     121(2):S.1-S.24.  February 1977.

4.   Reference 3, p.  S.2.

5.   Reference 1, pp. 17, 19,  21, 24, 25.

6.   Reference 1, pp. A-12.

7.   Schwieger, B.   Industrial Boilers - What's Happening Today,  Part II.
     Power.  122(2):S.19.  February 1978.

8.   Reference 1, p.  19.

9.   Telecon.  Wells, R.M., Radian Corporation, with Axtman, W.H., American
     Boiler Manufacturers Association.  May 23, 1980.   Information about
     portable boilers.

10.  Reference 1, p.  A-24.

11.  Telecon.  Herther, M., Acurex Corporation, with Ross, R.,  Hydronics
     Institute.  January 18,  1980.  Minimum/maximum pressure ratings  for
     cast iron boilers.

12.  Telecon.  Herther, M., Acurex Corporation, with Woodworth, J.,
     Hydronics Insitute.  October 16, 1979.   Information about cast iron
     boiler industry.

13.  Survey of Domestic, Commercial and Industrial Heating Equipment and
     Fuel  Usage.  (Prepared for U.S. Environmental Protection Agency.)  EPA
     Contract No. 68-02-0241.   June 1972.  pp. 1-11.

14.  Reference 1, p.  17.

15.  Reference 1, p.  21.

16.  Reference 1, p.  23.

17.  Reference 1, p.  34.
                                     3-69

-------
18.  Letter from Baum, N.P., Energy and Environmental Analysis, to
     Bunyard, F., EPA:EAB.  June 21, 1979.  Industrial boiler economic data.

19.  End Use Energy Consumption Data Base:  Series 1 Tables.  (Prepared for
     U.S. Department of Energy.)  Washington, D.C.  Publication
     No. PB-281-817.  June  1978.  p. 53.

20.  Letter from Baum, N.P., Energy and Environmental Analysis, to Herther,
     M., Acurex Corporation.  December 27, 1979.  Industrial boiler fuel
     consumption.

21.  Dupree, W.G. and J.S.  Corsentino.  United States Energy Through the
     Year 2000.  (Prepared  for  U.S. Department of the Interior.)
     Washington, D.C.  Publication No. PB-250-600.  December 1975.  p. 28.

22.  Ray, S.S. and F.G. Parker.   (Tennessee Valley Authority.)  Character-
     ization of Ash from  Coal-Fired Power  Plants.  (Prepared for
     U.S. Environmental Protection Agency.)  Washington, D.C.  Publication
     No. EPA-600/7-77-010.  January 1977.  pp. 5-6.

23.  Offen, G.R., et al.   (Acurex Corporation) Control of Particulate Matter
     from Oil Burners and Boilers.  (Prepared by U.S. Environmental
     Protection Agency.)   Research Triangle Park, N.C.  Publication No.
     EPA-450/3-76-005.  April 1976.  pp. 3-24 - 3-25.

24.  Reference 1, p. 71.

25.  Reference 23, p. 2-32.

26.  U.S. Environmental Protection Agency.  Compilation of Air Pollutant
     Emission Factors, Third Edition.  Research Triangle Park, N.C.
     Publication No. AP-42.  August 1977.  p. 1.1-2,  1.1-3.

27.  Letter and attachments from Marx, W.B., Council  of Industrial Boiler
     Owners, to Sedman, C.B., EPArlSB.  April 28, 1980.  18 p.  Information
     collected about soot blowing from eight users and two vendors.

28.  Draft Memo from Jennings,  M., Radian  Corporation, to Sedman, C.,
     EPA:ISB.  February 2,  1982.  Review of opacity data.  Table 5.

29.  Cato, G.A., et al.   (KVB,  Inc.)  Field Testing:  Application of
     Combustion Modifications to Control Pollutant Emissions from Industrial
     Boilers, Phase II.   (Prepared for U.S. Environmental Protection
     Agency.)  Research Triangle Park, N.C.  Publication No. EPA-600/2-
     76-086a.  April 1976.  p.  220.

30.  Gronhovd, G.H., et al.  (Grand Forks  Energy Research Laboratory.) Some
     Studies on Stack Emissions from Lignite-Fired Powerplants.  (Presented
     at  the 1973 Lignite  Symposium.  Grand Forks.  May 9-10, 1973.)  p. 9.
                                      3-70

-------
31.  Maloney, K.L., et al.   (KVB, Inc.)  Low-Sulfur Western Coal  Use in
     Existing Small and Intermediate Size Boilers.   (Prepared for
     U.S. Environmental Protection Agency.)  Research Triangle Park, N.C.
     Publication No. EPA-600/7-78-153a.  July 1978.  p.  23.

32.  Reference 1, p. 63.

33.  Lim, K.J., et al.  (Acurex Corporation.)  Technology Assessment Report
     for Industrial Boiler Applications: NO  Combustion  Modification.
     (Prepared for U.S. Environmental Protection Agency.)  Research Triangle
     Park, N.C.  Publication No. EPA-600/7-79-178f.  December 1979.  p. 2-3.

34.  Pershing, D.W., et al.   Influence of Design Variables on the Production
     of Thermal and Fuel NO from Residual Oil and Coal Combustion.   In:
     Air:  Control of NO  and SO  Emissions.  New York,  American  Institute
     of Chemical Engineers,  1975.  p. 24.

35.  Reference 33, p. 2-5.

36.  Reference 2, p. 5-19.

37.  Reference 1, pp. 93-102.

38.  Reference 2, p. 5-1 to 5-22.

39.  Reference 2, p. 25-10.

40.  Telecon.  Blackwell, C., Acurex Corporation, with Marx, W.B.,  Council
     of Industrial Boiler Owners.  January 28, 1980.  Information about
     industrial boilers.

41.  U.S. Environmental Protection Agency.  Electric Utility Steam
     Generating Units - Background Information for Proposed Particulate
     Matter Emission Standards.   Research Triangle Park, N.C.  Publication
     No. EPA-450/2-78-006a.   July 1978.  p. 3-8.

42.  Reference 1, p. 101.

43.  Reference 2, p. 11-3.

44.  Reference 1, p. 100.

45.  Reference 1, p. A-ll.

46.  Reference 1, p. 96.

47.  Reference 1, p. A-9.

48.  Reference 1, p. A-7.
                                     3-71

-------
49.  Reference 1, p. 95.

50.  Reference 1, pp. A-12, A-16.

51.  Reference 1, p. 19.

52.  Reference 1, p. A-18.

53.  Reference 33, p. 3-4.

54.  Cato, G.A., et alI.   (KVB,  Inc.)  Field Testing:  Application of
     Combustion Modifications to Control Pollutant Emissions from Industrial
     Boilers, Phase I.   (Prepared for U.S. Environmental Protection Agency.)
     Research Triangle  Park, N.C.   Publication No. EPA-650/2-74- 078a.
     October 1974.  p.  14.

55.  Reference 31, pp.  124, 140, 165, 181, 330.

56.  Roeck, D.R. and R.  Dennis.  (6CA Corporation.)  Technology Assessment
     Report for  Industrial Boiler Applications:  Particulate Collection.
     (Prepared for U.S.  Environmental Protection Agency.)  Research Triangle
     Park, N.C.  Publication No. EPA-600/7-79-178h.  December 1979.  p. 22.

57.  Smith, W.S. and C.W.  Gruber.   Atmospheric Emissions from Coal
     Combustion:  An Inventory  Guide.   (Prepared for U.S. Department of
     Health, Education  and Welfare.)  Washington, D.C.  Publication
     No. AP-24.  April  1966.  pp. 57, 59-60.

58.  Midwest Research Institute.  Particulate Pollutant System Study,
     Volume II - Fine Particle  Emissions.  (Prepared for U.S. Environmental
     Protection Agency.)   Durham, N.C.   Publication No. APTD-0744.  August
     1,  1971.  pp. 67-73.

59.  Reference 29, p. 60.

60.  Giammar, R.D. and  R.B. Engdahl.  (Battelle Columbus Laboratories.)
     Technical,  Economic  and Environmental Aspects of Industrial Stoker-
     Fired Boilers.  (Presented at  the  71st Annual Meeting of the Air
     Pollution Control  Association.  Houston.  June 25-30, 1978.)  p. 8.

61.  Reference 33, pp.  2-1 to 2-80.

62.  Reference 1, p. 97.

63.  Reference 1, p. 93.

64.  Reference 1, p. A-19.

65.  Reference 23, p. 3-25.
                                     3-72

-------
66.  Reference 33, p.  2-33.

67.  Leavitt, C., et al.   Environmental  Assessment of Coal- and Oil-Firing
     in a Controlled Industrial  Boiler;  Volume III.  Comprehensive
     Assessment and Appendices.   (Prepared for U.S. Environmental  Protection
     Agency.)  Research Triangle Park, N.C.   EPA Contract No.  68-02-2613,
     Task 8.  August 1978.   p.  6-22.

68.  Reference 26, p.  1.3-2.

69.  McGarry, F.J. and C.J.  Gregory.   A Comparison of the Size Distribution
     of Particulates Emitted from Air, Mechanical, and Steam Atomized
     Oil-Fired Burners.  JAPCA.   22(8):636-639.   August 1972.

70.  Reference 29, p.  34-41.

71.  Reference 1, p. 17.

72.  Reference 1, p. 19.

73.  Reference 1, p. 94.

74.  Reference 26, p.  1.4-2.

75.  Reference 33, p.  2-64.

76.  Reference 33, p.  2-63.

77.  Werner, A.S., et al.   (GCA Corporation.)  Screening Study to  Obtain
     Information Necessary for the Development of Standards of Performance
     for Oil-Fired and Natural  Gas-Fired Boilers.  (Prepared for
     U.S. Environmental Protection Agency.)   Research Triangle Park, N.C.
     EPA Contract No.  68-02-1316.  September 1976.  p. 96-111.

78.  Reference 67, p.  5-25.
                                     3-73

-------
                      4.0  EMISSION CONTROL TECHNIQUES

     Uncontrolled emissions from industrial boilers were identified in
Chapter 3.  Emission control  techniques potentially applicable to industrial
boiler: sources are described in this chapter.  These descriptions include
discussions of the design of each control  technique, its status of develop-
ment, and its applicability to industrial  boilers.  Also discussed are
factors which affect the performance of the control techniques, including
design parameters, operating conditions, and fuel  quality.   Emission data
taken by approved EPA test methods to verify control technique performance
is presented and discussed when available.  Additional information on these
tests is presented in Appendix C.
     Portions of the control  technology discussions contained in this
chapter were excerpted from a series of Individual Technology Assessment
Reports (ITAR's) prepared to assess the application of specific control
techniques to industrial boilers.  The ITAR's describe each technology in
more detail than that presented in this chapter.   Emissions test and system
performance data reported in this chapter include both data reported in  the
ITAR's and data gathered subsequent to their preparation.  Sources for these
data are specifically referenced in Appendix C.  The reader desiring
additional information on any of the technologies discussed in this chapter
is referred to the series of ITAR's listed in Table 4-1 and the other
references listed at the end of this chapter and in Appendix C.
     Control techniques discussed in this chapter are those meeting one  of
the following criteria:
       •  Currently used on industrial boilers or large pilot-scale
          installations;
       •  Currently applied in the utility or foreign sectors; technology
          transferability is indicated;
       •  Rapidly developing and likely to be commercially available in
          the next several years.
                                     4-1

-------
                     TABLE 4-1.   ITAR REPORT LIST
                   Report
  • Report No.
Technology Assessment Report for  Industrial Boiler
  Applications:  Oil Cleaning
Technology Assessment Report for  Industrial Boiler
  Applications:  Coal Cleaning and  Low Sulfur Coal
Technology Assessment Report for  Industrial Boiler
  Applications:  Synthetic  Fuels
Technology Assessment Report for  Industrial Boiler
  Applications:  Fluidized-Bed Combustion
Technology Assessment Report for  Industrial Boiler
  Applications:  NO  Combustion Modification
                   ^
Techno!oay Assessment Report for  Industrial Boiler
  Applications:  NO  Flue  Gas Treatment
                   ^
Technology Assessment Report for  Industrial Boiler
  Applications:  Particulate Collection
Techno!oay Assessment Report for  Industrial Boiler
  Applications:  Flue Gas  Desulfurization
EPA-600/7-79-178b
EPA-600/7-79-178C
EPA-600/7-79-178d
EPA-600/7-79-178e
EPA-600/7-79-178f
EPA-600/7-79-178g
EPA-600/7-79-178h
EPA-600/7-79-178i
                                   4-2

-------
     This chapter is organized into six sections.  The first three sections
discuss post-combustion controls for particulate and sulfur dioxide (SCL),
and combustion modification for nitrogen oxides (NO ) control, respectively.
                                                   A
Section 4.4 discusses post-combustion controls for NO .   Pre- combustion
                                                     A
control techniques for particulate, NO , and SOp are discussed in
Section 4.5, while fluidized bed combustion and other techniques involving
combustion of a coal/alkali fuel mixture to control SO^  emissions are
presented in Section 4.6.
4.1  POST-COMBUSTION CONTROL TECHNIQUES FOR PARTICULATE  MATTER
     The post combustion control of particulate matter emissions from
industrial boilers can be accomplished by using one or more of the following
particulate control devices:

       •  electrostatic precipitators,
       •  fabric filters,
       •  wet scrubbers,
       •  side stream separators, or
       •  multitube cyclones (single and dual  mechanical
          collectors)
These control devices are discussed separately in Sections 4.1.1 through
4.1.5.  Test data documenting the performance of each of these control
devices applied to industrial boilers are presented and  discussed in
Section 4.1.6.
4.1.1  Electrostatic Precipitators
     The collection mechanism, factors affecting performance, status of
development, and applicability of electrostatic precipitators (ESPs) to
industrial boilers is discussed in this section.
                                     4-3

-------
     4.1.1.1  Process Description
     4.1.1.1.1  System.  Participate collection in an electrostatic precipi-
tator occurs in three steps: suspended particles are given an electrical
charge; the charged particles migrate to a collecting electrode of opposite
polarity while subjected to a diverging electric field; and the collected
particulate matter is dislodged from the collecting electrodes.
     Charging of the particles to be collected is usually caused by ions
produced in a high voltage d-c corona.  The electric fields and the corona
necessary for particle charging are provided by high voltage transformers
and rectifiers.  Removal of the collected particul ate matter is accomplished
mechanically by rapping or vibrating the collecting electrodes.
Figure 4.1-1 shows a typical cross-sectional view of an ESP.
     4.1.1.1.2  Development status.  Electrostatic precipitator technology
is commercially developed and dates back to the early 1900's.  The first
successful application was made in 1907 when acid mist was collected at a
sulfuric acid plant.  ESPs have been used to control particulate emissions
                                                                      2
from coal-fired industrial and utility boilers since the early 1920's.
They are also the most commonly used collectors on utility oil-fired
boilers.
     4.1.1.1.3  Applicability to industrial boilers.  Electrostatic precipi-
tation technology is applicable to a variety of types and sizes of
                                                                       3
industrial boilers.  ESPs treating flue gas flow rates as low as 8500 m /hr
                          4
(5000 acfm) are available.   Because of their modular design, ESPs can be
expanded to treat flue gas from even the largest industrial boilers.  ESPs
have been installed on utility boilers with flue gas flow rates as high as
10,000,000 m /hr.  Application of an ESP to an industrial boiler should have
no adverse effect upon boiler operation.  The fuel quality and its effect on
particle characteristics is especially important and is discussed in detail
in the next section.
     4.1.1.2  Factors Affecting Performance.  ESP collection efficiency is
affected by a wide variety of factors related to the design of the ESP and
the type of particles collected.  Two factors have been specifically related
to the overall collection efficiency through the Deutsch-Anderson equation:
                                     4-4

-------
           BUS DUCT
  INSULATOR
 COMPARTMENT
RAPPER INSULATOR
   HIGH vb~LTAGE~SYSTEM
    SUPPORT INSULATOR
       COLLECTING SURFACE
             RAPPER
                 DISCHARGE ELECTRODE
                       RAPPER
       COLLECTING
         SURFACE
                             DISCHARGE
                             ELECTRODE
                      GAS PASSAGE
 TRANSFORMER
   RECTIFIER


    GAS
DISTRIBUTION
   DEVICE
                         Figure  4.1-1.  Typical preclpitator  cross section.

-------
          n  =  1 - exp (-WA/V)

where:
          n  =  removal efficiency
          W  =  migration velocity
          A  =  plate area
          V  =  volumetric flow rate

As indicated by this equation, ESP efficiency increases with increasing
plate area relative to the gas flow rate and with increasing migration
velocity.
     Field data has indicated that the Deutsch-Anderson equation
overpredicts collection efficiency.  To account for the observed particle
                                                                   5
collection efficiencies, White proposes the empirical relationship,

          n  =  1 - exp [-(Wk A/V)°'5]

as a more accurate predictor of efficiencies.  The exponent of 0.5 is
applicable for ESPs applied to coal-fired boilers.  The term W.  is a measure
of the effective migration velocity determined from experimental measure-
ments.
     The following discussion reviews the major factors which influence ESP
performance.  For purposes of this discussion, the factors are grouped into
two categories:  (1) ESP design factors and (2) particle characteristic
factors.
     Design Factors.  The specific collection area (SCA) is defined as the
ratio of the total plate area to  the gas flow rate and is usually expressed
             23      2
in terms of m /(m /s)[ft /1000 acfm].  SCA is an important design and
operating parameter.  For a given application, collection efficiency
improves as SCA increases.  But for a given gas flow, the ESP also becomes
larger and consequently more expensive as the SCA is increased.   Thus,
correct sizing of an ESP is important to both the performance and economics
                   21
of its application.
                                     4-6

-------
     Typical  relationships between precipitator collection efficiency and
SCA are shown in Figure 4.1-2 for coal  fly ash.  The separate lines for
different coal sulfur contents reflect  the dependence of fly ash
resistivity,  and hence collection efficiency, on the coal sulfur content.
Tests of low sulfur coals, for example, indicate that these variables may
cause Figure  4.1-2 to underestimate the SCA needed for a 0.5 percent sulfur
coal by 40 to 50 percent.    Practical  values of SCA range from 328 to
2630 m2/1000  m3/min (100 to 800 ft2/1000 acfm) for most field applications.6
     The actual  collection area during  ESP operation depends on the flue gas
flow rate which, for a particular boiler, is dependent on boiler load.  The
operating SCA increases as boiler load  decreases, provided all  ESP fields
remain charged.   Thus, the ESP must be  designed to have the desired SCA at
maximum boiler load where the flue gas  flow is the highest.
     The configuration and type of electrodes used in an ESP directly
influence ESP performance.  The electrode plate spacing, height, and length
all influence the electrostatic forces  exerted on the flue gas  particles and
thus influence the collection efficiency.  Proper design of the ESP
electrodes assures adequate residence time to allow the particles to migrate
to a collection electrode.
     Another key design variable is proper determination of the rapping
cycle.  If the cycle is too short, material that collects on the plates will
not be compacted enough to settle to the bottom of the precipitation chamber
and will be re-entrained.  This re-entrainment can be minimized by proper
design of collecting electrodes and rappers, minimizing rapping and rapping
only a small  section of the total precipitator plate area at a  time.  If the
time between rapping is too long, however, the material on the  collecting
plates will become too thick and collection efficiency will be  reduced.  In
addition, the rapping cycles must account for the differences in the amount
of particulate matter collected in different ESP sections.  ESPs typically
use multiple sections in series.  The section which treats the flue gas
first will collect more particles than  subsequent sections.  The rapping
cycles must be adjusted to insure each  section is rapped only when the
                                     4-7

-------
       99.9
       99.0
     c

     o

     0>
     Q.



     O
     2
     LU

     O

     u.
     LL ,
          .o
     o
     LJJ
     _J
     _J
     O
     o
80.0




70.0


60.0
                     Sulfur  Contents
                       5080          10160        15240
                       (100)          (200)         (300)

                             SCA, nrr/(m3/s) (ft2/1000 cfm)
                                                        20320
                                                         (400)
 Figure 4.1-2.   Relationship  between collection efficiency and  SCA for
                 various coal  sulfur contents. 14


a0.5 percent sulfur values based on limited data.
                                    4-b

-------
collected material  is the proper thickness.   This necessitates more frequent
cleaning cycles for the sections treating the raw flue gas.
     Gas flow distribution also has a strong impact on ESP efficiency.   Poor
flow distribution between the collecting electrodes results  in differing gas
flow rates between each plate and therefore  differing efficiencies  for  each
section of the ESP.  In addition, high velocities in the vicinity of hoppers
and collecting electrodes can result in re-entrainment of collected dust.
Another distribution consideration is the avoidance of flue  gas flow through
certain areas of the ESP.  The construction  of an electrostatic precipitator
is such that nonelectrified regions exist in the top of the  precipitator
where the electrical distribution, plate support and rapper  systems are
located.  Similarly, portions of the collection hopper and the bottom of the
electrode system contain nonelectrified regions.  Particulate-laden gas
streams flowing through these regions will  not be subjected  to collection
                                                                 20
forces and will tend to pass through the precipitator uncollected.    Gas
flow distribution problems can be corrected  by proper inlet  design, such as
adding straighteners, splitters, vanes, and  diffusion plates to the duct
work before the ESP and by internal baffles  and flow restrictors.
     The voltage applied to the ESP electrodes is also an important factor
affecting performance.  Proper voltage assures an adequate corona for
                                                             15
charging the particles while minimizing problems of sparking.     The use of
automatic power supply control is desirable  in many industrial boiler
applications because of the varying fly ash  and flue gas properties brought
on by varying boiler loads and fuel properties.  Automatic controls allow
the ESP to respond more effectively to these changes by reducing sparking
                 230
and current loss.
     Particle Characteristic Factors.  The suitability of particulate
collection by electrostatic precipitation depends on the resistivity of the
particles.  Particulates with resistivities  in the range of 10  to  10
ohm-cm have been shown by experience to be the most suitable for
electrostatic precipitation.  Particles with lower resistivities will give
up their charge too easily and will be re-entrained in the gas
       Ifi IT ip in
stream.10>1/>1>    Particles with higher resistivities will coat the
                                     4-9

-------
collecting plates and will be hard to dislodge, thereby reducing the ability
of the electrode to further collect particles.  The resistivity of a given
particle will vary with temperature and moisture.  Typical variations in
precipitation rate (which determines collection efficiency) with particle
resistivity and coal sulfur content are shown in Figures 4.1-3 and 4.1-4.
     The dependence of fly ash resistivity on fuel characteristics is very
important when considering the application of ESPs to industrial boilers.
The most notable fuel properties which affect the resistivity of the fly ash
are the sulfur and alkali (primarily sodium) content.  As shown in
Figure 4.1-5, resistivity is altered favorable (reduced) with an increase in
the sulfur content.  As shown by Figure 4.1-6, an increase in the sodium
content of the ash also tends to reduce the resistivity of the fly ash.
Resistivity varies with temperature as well as fuel sulfur content as shown
by Figure 4.1-7.  The typical "cold side" ESP is located downstream of the
air preheater, where the  temperatures range from 380 to 448 K (240° to
350°F).  A "hot side" ESP, on the other hand, is located upstream of the
boiler air preheater, where temperatures  range from 563 to 698 K (550 to
800°F).
     Particle size distribution directly  affects the precipitation rate
parameter.  Fractional collection efficiency data for ESPs applied to
oil-fired boilers demonstrated ESP collection efficiencies of at least
99 percent for fine and coarse particles  (less than 1 ym and greater than
10 ym), while collection  efficiency dropped off from 99 to 89 percent for
sizes between 1 and 10 microns.    In general, the fractional efficiency
drop off for coal-fired boilers has been  shown to occur in the range from
0.1 to 1 ym.    Figure 4.1-8 presents test results which show this for an
ESP applied to a pulverized coal-fired boiler.    For this particular
installation, fractional  efficiencies vary from 99 to 90 percent for
particles in the range of 2 to 0.1 ym.
     Boiler load affects  the particle size distribution for coal- and
oil-fired boilers with reduced loads generally corresponding to reduced
               IP 10
particle sizes.  '    Changing boiler load, and its effect on particle size
                                     4-10

-------
                  Ul
                    0.6
                    0.5
                    0.4
                    0.3
                    0.2
                 CC 0.1
                 CL
                                                         15.2
                                                         '.2.2
                cc

             9.! §
                                                             O
                      10*
                                  ,10
sll
  10'"        10

RESISTIVITY, ohm-cm
         10
                                                        12
  Figure 4.1-3.   Precipitation rate versus  particle  resistivity.


•5
•^
D
LU
t-

z
o
p
H
CL
U
LU
CC
a.


u./
0.6

0.5


0.4


0.3


0.2



0.1
0
i 1 1 1 1
-

RAMSDELL* y
CURVE 300°F /
V J -^
^^k. f ^^^^
"BARRETT REGRESSION / ^^.•••'
ANALYSIS CURVE >^-****^""
^^^ • • *2^^^
.••'X^*"^.
..•^•^'^TVA DATA 320°F
l^»
— —



-
1 1 1 1 1
'
18.3
VI
15.2 |
LU*
^B
^
12.2 cc
O

9.1 <
—
U
6 LU
CC


3
0
                      0    0.5    1.0   1.5    2.0    2.5   3.0

                              COAL SULFUR, percent


    Figure 4.1-4.   Precipitation rate  versus coal sulfur  percent.8
aRamsde11 number for sulfur percents above 2.5 have been
 disputed and are not Included here.
                                     4-11

-------
                TEMPERATURE,  c
        100         ?50        200
     10'
     10'
    E
    i
   to
   35
     10'
     10'
                     0.25% SULFUR COAL
0.6% SULFUR COAL


         Q.
                       \
                       2.1% SULFUR COAL,
                      3.6% SULFUR COAL
      200
             250
                  300
                        350
                              400
                                   450
                 TEMPERATURE, F
Figure 4.1-5.
                                                10"
                                      -10'
                                      e
                                              en
                                              35
                                              •Mo10
       Variation of  fly ash
       resistivity with
       temperature for coal of
       various sulfur
       contents.
       0.1  0.2 0.3  0.5 0.7 1   23   5  7 10
          SODIUM CONTENT AS N«-O. p.rc.nt

Figure 4.1-6.  Variation of  resistivity
               with sodium content for
                fly ash from  power plants
                burninn Western  coals.9
                            0.5
                 Figure 4.1-7,
                        1.0   1.5   2.0  2.5   3.0
                          COAL SULFUR. p*mnt

                        Fly  ash  resistivity  versus coal
                        sulfur content for several flue
                        gas  temperature bands.
                                       4-12

-------
I—«
CO
          c
          «
          o
          O
          z
          UJ
          u.
          u.
          UJ

          (O

          O
          U
          Ul
          o
          o
99.98



 99.9

 99.8

 99.5

   99

   98


   95


   90
               60
               0.05
                                                  MEASUREMENT METHOD'
                                                 A CASCADE IMPACTORS

                                                 o OPTICAL PAPTICLE  COUNTERS
                                                 + DIFFUSIONAL

                                                  PRECIPITATOR CHARACTERISTICS!
                                                   TEMPERATURE-320°F
                                                   SCA-340ft2/IOOOocfm
                                                   CURRENT DENSITY O.Ol5mA/ft2 -
                                                   EFFICIENCY-98.3%  .
             O.I
0.5
1.0
5.0
10.0
                                            PARTICLE  DIAMETER,/im
                   Figure 4.1-8.
                    Measured fractional efficiencies for  a cold-side  ESP with
                    operating parameters as  indicated,  installed on a pulverized
                    coal boiler burning low  sulfur

-------
distribution may subsequently affect the overall collection efficiency of an
ESP if not considered in the original design.
4.1.2  Fabric Filters
     The collection mechanisms, design and operating parameters, development
status, and applicability of fabric filters to industrial boilers are
discussed in this section.
     4.1.2.1  Process Description
     4.1.2.1.1  System.  A fabric filtration system (baghouse) consists of a
number of filtering elements (bags) along with a bag cleaning system
contained in a main shell structure with dust hoppers.  Particulate-laden
gases are passed through the bags so that the particles are retained on the
upstream side of the fabric, thus cleaning the gas.  Typically a baghouse is
divided into several compartments or sections.  In larger installations an
extra section is often provided to allow one compartment to be out of
service for cleaning at any given time.  A typical baghouse is shown in
Figure 4.1-9.
     The basic mechanisms available for filtration are inertial impaction,
diffusion, direct interception, and sieving.  The first three processes
prevail only briefly during the first few minutes of filtration with new or
recently cleaned fabric, while the sieving action of the dust layer
accumulating on the fabric surface soon predominates.  This is particularly
true at high, >1 g/m   (0.437 gr/dscf), dust  loadings.  The sieving
mechanism, in the case of coal fly ash filtration, leads to high efficiency
collection unless defects such as pinhole leaks or cracks appear in the
filter cake.23
     In fabric filtration both the collection efficiency and the pressure
drop across the bag surface increase as the dust layer on the bag builds up.
Since the system cannot continue to operate with an increasing pressure
drop, the bags are cleaned periodically.  The cleaning processes used in
coal-fired systems ordinarily consist of reverse-flow with bag collapse or
mechanical shaking.  These are sometimes used in combination with each
other.  Pulse-jet cleaning also has had considerable application while the
                                                                      24
reverse-jet concept (traveling blow ring) has not been widely applied.
                                     4-14

-------
I
(—»
on
                         Pyramidal or
                         Trough Hoppers
                                                                                           Access Plates
                                                                                             Solenoid Valves
                                                                                             Compressed Air
                                                                                             Manifold
                                                                                             Dirty Air Inlet
Bailie Plate
                                                                                          Access Door
                             Figure 4.1-9.   Isometric view of  a  pulse-jet fabric filter.
                                                                                           22

-------
     4.1.2.1.2  Development status.  Fabric filtration is a well-established
technology with early industrial process applications dating back to the
late 1800's.  However, application to boiler flue gas has been a relatively
recent development with the first successful installations designed in  the
late 1960's and early 1970's.  Data published in December of 1979 shows that
there were 104 industrial boilers at 61 locations either using, or planning
on using fabric filtration systems for particulate emission control. These
systems are summarized in Table 4.1-1.  These boilers have flue gas rates
ranging from 5940 to 1.5 x 106 m3/hr (3500 to 900,000 acfm).25
     4.1.2.1.3  Applicability to industrial boilers.  Fabric filtration
technology is applicable to various industrial boiler types as shown in
Table 4.1-1.  A possible limitation, however, is the application to oil-
fired boilers.  Large quantities of unburned carbon particles present in the
flue gas of oil-fired boilers could cause difficulty in cleaning the dust
                    ?fi
layer from the bags.    Only a  limited number of baghouses have been
installed on oil-fired boilers.  One installation initially experienced
difficulty with bag life and plugging  and was later retired when changes  in
local air pollution control  regulations forced a fuel switch to gas.
     During baghouse operation  it is essential that baghouse temperatures  be
maintained above the water and  acid dewpoints of the gas so that condensa-
tion will not occur on the compartment walls and filter surfaces.  In the
case of condensation on filter  surfaces, resultant plugging may restrict gas
flow and cause irreversible  bag damage.  This is most likely to occur during
transient operations such as startup, shutdown or fluctuating loads.
Bypassing or preheating the  baghouse prior to system startup, continuous gas
recirculation during brief shutdowns, and sufficient insulation (7.6 cm or
3 inches of mineral wool or  fiberglass) can prevent condensation
         97 ?fl
problems.   '°
     4.1.2.2  Factors Affecting Performance.  Several factors can affect the
performance of baghouse systems including air-to-cloth ratio, fuel
properties, baghouse temperature and filter fabric and weave.  These
factors, and their affect on baghouse performance are discussed separately.
                                    4-16

-------
TABLE 4.1-1.   BAGHOUSE INSTALLATIONS ON INDUSTRIAL BOILERS - U.  S.
                                                                  25

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.


Name/location
Ado!?h Coon Co.
Coiden, Colo.
Allied Chemical
Southpoinc, Ohio
Allied Chemical
Moundsville, W. Va.
Amalgamated Sugar Co.
Nampa , Idaho
Amalgamated Sugar Co.
Nanpa, Icaho
Amalgamated Sugar Co.
Nyssa, Oreg.
Amalgamated Sugar Co.
Nyssa, Oreg.
Amalgamated Sugar Co.
Twin Falls, Idaho
Amecek, Inc.
Moline, 111.
Ashland Chemical Co.
Peoria, 111.
Carborundum Co.
Niagara Falls, N.Y.
Case Western Reserve U.
Cleveland, Ohio
Caterpillar Tractor Co.
Decacur, 111.
Consolidated Rail Corp.
Alcoona, Pa.
Deleo-Remy-Div. CM
Anderson, Ind.
Denver Federal Cenccr
Denver, Colo.
E.I. DuPont Co.
Cooper R, S.C.
E.I. DuPont Co.
Martinsville, Va.
E.I. DuPont Co.
New Johnsonville, Tenn.
E.I. DuPont Co.
Parkcrsburg. V*.
E.I. DuPont Co.
Waynesboro, Va.
Energy Development Co.
Hanna , Wyo .
Formica Corp.
Evendale. Ohio
Hanoermill Paper Co.
Lockhaven, Pa.
Hanes Dye and Finishing
Uini ton-Sal em, N.C.
Harrison Radiator-
Division CM
Locttporc, N.Y.
Manu-
facturer
WF

WF

WF

wp

EB

WF

wp

WP

AAF

SH

CAR

FK

SH

WF

SH

ZU

wp

WP

SH

SH

wp
'test unit)
ICA

WF

ICA

DX

WP


CUjnint
n1. sm
RA, sa

RA, sa

RA, sa

RA

Sh

RA, sa

RA

RA

RA

P

RA

Tbd

POL

RA, sa

P

RA

RA, va

RA, va

P

P

RA, va

RA

RA, sa

RA

P

P


Soi ler
f i ri nn
me t hod
PC

PC

S

PC

PC

S

PC

1-PC
1-S
S

S

S

-

S

S

S

S

S

PC

S

S

PC

S

S

S

S

S


Size
<»*>.
31

(6)-12

(4)-32

28

29

21

13

21
each
9

16

9

-

33

OMB

(3)-9

9

20

45

(2)-29

(4)-50

76

5

3

53

(2)-13

30


A/C*
3.3/1

2.oq/i

2.89/1

2.4/1

2.5/1

3.56/1

2/1

2.5/1

4/1

4.4/1

2/1

Tbd

4.3/1

3.5/1

3/1

2.23/1

1.9/1

1.9/1

4.4/1

4.4/1

1.9/1

2.5/1

3.38/1

2/1

8.3/1

5/1


JC fn
1 !>0 . 'JOO

50,000

156,400

136,000

130.000

9 7., 000

57,000

100. 000
each
40,000

7o,ono

42,000

Tbd

150,000

108,000

24 ,000

174,000
*
90,000

203,000

130,000

221,000

340,000

24,000

42,000

150,000

61,000

139,000


St artup
iate
1976

1978

107P

1974

1975

1973

1975

1975

1974

1176

1967

Tbd

1976

1978

1976

1978

1977

1977

1975

1974

1977

1976

1978

1976

1975

1974


                                  4-17

-------
TABLE 4.1-1.  (Continued)

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

i9.

50.

*«'*n>«/ 1 ocat i on
Hiram UalVer 4 Sons
Peoria, 111.
Keener Rubbc: ~o.
Alliance, Ohio
Kerr Inouicries
Concord, N.C.
Kingiiey Air force Sase
Klanath Falls. Oreg.
uon( Like Lumber Co.
Spokane, U'ash.
Lubrizol Corp.
Painesville, Ohio
lonroe Reformatory
Monroe , Wash.
Pennsylvania Class Sand Corp
Union, Pa..
Republic Steel
-arren, Ohio
Simpson Timber Co.
SheUon, Wash.
Sorg Paoer Co.
Middlecown, Ohio
Uni royal, Inc.
Painesville, Ohio
Uni royal, Inc.
Mishawaka, Ind.
University of Illinois
Chicago, 111.
University of Iowa
Oakdale, Iowa
University of Minnesota
Minneapolis, Minn.
University of North Carolina
Chapel Hill. N.C.
University of Noere Dame
South Bend, Ind.
Utah-Idaho Sugar Co.
Moses Lake. Ua*h.
U.S. Navy
Hawthorne, Nev.
U.S. Steel Co.
Prove, Utat-
Uestinghouse Electric
Rich land. Wash.
•-•estvacp
Tyranne , Pa.
'-'i ceo Chcsical
!r»dford, Pa.
Manu-
facturer
Tbd

WF

ES
(test unit)
SH

MP

SH

ICA

ro

WF

SH

zu

SH

Tbd

DV

ES

CAR

W7

WF
(test unit)
EB

ICA

WF

MP

WF

WF

nism
Tbd

P

Var

P

P

P

Sh

P

RA, sa

POL

RA

P

Tbd

P

Tbd

RA

RA

P

Sh

RA

RA. si

RA

RA, sa

RA. sa

»oiler
f i rinc
n«e:hod
PC

Hdf

S

S

HF

OF

S

PC

PC

HF

PC

PC

PC

OF

-

S

.

S

S

S

PC i gac

S

S

1-S
1-PC
Size
60

100 hp

8

5

5

8

3

6

35

SI

10

9

22

8

-

20

(2)-6
each
1

22

21

(3)-90

7

20

(2)-l8

A/C*
Tbd

4.36/1

3-14/1

5/1

4.5/1

4.3/1

2.8/1

7/1

3.34/1

4.3/1

1.8/1

2.6/1

Tbd

6/1

Tbd

2/1

Tbd

7/1

2/1

1.7/1

3.2/1

2/1

3.26/1

3.17/1

, T
ac rm
270,000

5.500

35,000

24,000

24,000

35.000

11,000

40,000

2:5.000

230,000

45,000

42,000

100,000

35,000

Tbd

90,000

Tbd

3.500

98,000

96,000

900,000

32,000

135,000

1C5.300

Startup
date
1978

1977

1974

1976

1973

1974

1976

1972

1978

1976

1972

1976

1977

1976

Tbd

1976

1978

1972

1976

1976

1977

1976

1979

1978

           4-18

-------
TABLE 4.1-1.   (Continued)
u Cleaning Boiler Jlze
Name /location 'arnjrer siecna~ firing /yv)
51. General Motors Corp. SH - 7-S
Kecterlng t Norwood, Ohio
Three Riven, Mich.
Warren, Ohio
52. Scott Paper Co. - - 5-HF
Everett, Wash.
53. Federal Bureau of Prlons ES - S -
Fed. Correct. Institution
Anderson, W.Va.
54. Tennessee State Univ. CE RA 3-coal
Nashville, Tenn.
55. Georgetown Univ. ES - FBC
Washington, D.C.
56. CSA, West Heating Plant RC P 2-S
Washington, D.C.
57. West Point PeppereTI , Inc. BS - coal
Opelika, Ala.
58. U.S. Cypsum Co. - P 3-S
Plasterco Plant
Saltvll-le. Va.
59. AVTEX Fibers. Inc. EB - 5-coal
Front Royal, Va.
60. Michigan State Univ. RC RA 2-PC 2-60
61. 3-M Company 1CA RA 2-S 2-14
St. Paul, Minn.
*A/C as given is in ft/min. To convert to n/min, multiply by 0.30&8.
"To convert acfa to tnVhr, multiply by 1.699
Manufacturers: Symbols:
AAF - American Air Filter Co. Hdf
CAR - Carborundum Co. Pollution Control Div. HF
DV - DaVair Inc. OF
DX - Dust ex, Sub. Amer. Precision Ind. P
EB - Envirotech Corp. Buell Div. PC
ES - Enviro System Inc. Pol
FD - Fuller Co., Sub CA1X RA
FK - Flex-Kleen - Sub. R.C. RA, si
ICA - Industrial Clean Air Inc. RA , va -
ME - Menardi -Sou them Div., U.S. Filter Corp. S
MP - Mikropul Corp., Sub. U.S. .Filter Corp. Sh
SH - Standard Havens Inc. Sp
UF - Wheelabrator-Frye Inc. Tbd
WP - Joy Mfg. Co Western Precip. Div. Var
20 - Zurn Industries, Air Systems Div. FBC
CE - CE Air Preheater
RC - Research-Cottrell
BS - Banco Systems, Inc.
. ,-« , * Stanup
A/C* acfm d.re
1979



260,000 1979

2.6/1 16.000 1979


50,000

5/1 43,000

1979

-
.
41,500


600,000 "arch
1980
l.o/l 300,000 1980
2.2/1 70.000 1978




Hand-fired
Hogged fuel
Oil-fired
Pulse
Pulverized coal
Pulse, off-line
Reverse air
Reverse air, shake assist.
Reverse air, vibrator assist
Stoker-fired
Shaker
Special
To be determined
Various
Fluidized Bed Cora us t ion



            4-19

-------
     Air to Cloth Ratio.  The most important design and operating factor for
a baghouse is the air-to-cloth ratio (A/C).  This parameter relates the
volume of gas filtered (m /min or acfm) to the available filtering area
  22
(m  or ft ).  The A/C ratio is, in effect, the superficial  velocity of the
gas through the filtering media.  Air-to-cloth ratios typically range from
0.6 to 1.2 m/min (2 to 4 ft/min) for reverse-air cleaning systems and from
                                                                29
1.2 to 2.4 m/min (4 to 8 ft/min) for pulse-jet cleaning systems.     Emission
tests have shown that fabric filter collection efficiency generally improves
                                       on
as the air-to-cloth ratio is decreased.    Since the air-to-cloth ratio is
greatest at maximum flue gas flow (i.e., maximum boiler load), the fabric
filter must be designed to operate at the desired air-to-cloth ratio at
maximum boiler load.  Operation at lower boiler loads will  result in a lower
air-to-cloth ratio and a collection efficiency equal to or greater than that
at maximum boiler load (provided all fabric filter compartments are kept on
line during reduced load operation to maintain the same available cloth
area).
     Fuel Properties and Baghouse Temperature.  Variations in fuel
properties are not as critical in fabric filtration as they are with ESP
technology.  However, fuel sulfur content dictates the flue gas S0? content
and subsequent acid condensation temperature.  The baghouse temperature must
be maintained above the acid condensation point in order to reduce corrosion
of the baghouse internals and ductwork in addition to reducing bag wear and
destruction.  This is especially important during start-up and shut-down
operations when the temperature is most likely to fall below the acid
condensation temperature.  If acid condensation occurs after shutdown, the
acid mist moisture eventually evaporates and crystallization on the bag
filter may occur.  In this situation, the bag filter may become brittle and
                                                      28
subject to cracking when stress is once again applied.
     Bag Fabric and Weave.  In general, bag material is chosen to withstand
the specific flue gas environment expected to be encountered.   Mechanical
strength is also an important factor with respect to the mechanical demands
exerted on the fabric by the gas flow and cleaning system.   The bag material
used in coal-fired boiler applications is usually fiberglass with a coating
                                    4-20

-------
                                     24
of silicone, graphite, and/or teflon.     Teflon coated felt bags  are used in
some pulse jet systems.
     In general, although nonwoven fabrics (i.e.,  felt) are the most
efficient particle collectors, they are the most difficult to clean.
Texturized filament fabrics (i.e., teflon coated fiberglass)  represent  a
                                                         32
middle ground in cleanability, durability and efficiency.
     Most fabrics are efficient in collecting a wide range of sub-micron
particles.  Emission tests conducted on a 63,100 kg steam/hr (139,000 Ib
steam/hr) spreader stoker equipped with a reverse-air fabric filter
demonstrated that for particles in the 0.02 to 2 micron range, fabric filter
                                                      33
fractional efficiency did not fall below 99.9 percent.
4.1.3  Wet Scrubbers
     The collection mechanism, status of development, applicability to
industrial boilers, and factors which affect the performance of wet
scrubbers for particulate control  are discussed in this section.
     4.1.3.1  Process Description
     4.1.3.1.1  System.   A wet scrubber is a collection device which uses an
aqueous stream or slurry to remove particulates and/or gaseous pollutants.
When scrubbing is applied for control  of fly ash from combustion  processes,
the contactor used is usually one  of the following types:   gas-atomized
spray scrubbers such as  venturi and flooded disc scrubbers, fixed-bed
absorbers such as sieve  tray units, turbulent contact absorbers (TCA) or
                                                                   34
moving bed scrubbers and high pressure spray impingement scrubbers.
     There are three basic mechanisms involved with collecting particulate
in wet scrubbers.  These mechanisms include the interception, inertial
impaction and diffusion of particles on droplets.   The inertial impaction
and interception effects dominate  at large particle diameters, while the
diffusion effects dominate at small particle diameters.
     In a typical venturi scrubber, which is illustrated in Figure 4.1-10,
the primary collection mechanisms  are interception and impaction.  Gas
entering the venturi is smoothly accelerated in the converger until it
reaches a maximum velocity in the  throat.  This converts the static pressure
head to a kinetic energy head and  typically requires from 1.2 to  5 kPa  (5 to
                                   4-21

-------
                                         CLEAN GAS
                                            OUT
                        PARTICLE
                       LADEN GAS
                            IN
Scrubbing  liquor  inlet  (V
         Venturi  Throat
                                                           -Cyclone Separator &
                                                             Mist Elimination
                                           ASH
                Figure 4.1-10.  Variable-throat  venturi scrubber.
                                                               35
                                         4-22

-------
20 inches of water)  pressure drop.   Scrubbing liquid is atomized by the high
velocity gas stream to produce droplet particles which act as targets  for
interception and impaction type collection.
     In general, high interception  and impaction collection efficiencies
result from the high differential  velocity between the gas stream and  the
atomized droplets created in the throat.   Therefore, an increase in the
system pressure drop will result in an increased differential  velocity and
subsequent increase in efficiency.   Because system pressure drop is a
function of energy expenditure, the energy imparted to the gas stream  is a
measure of the systems efficiency.   The droplets are removed from the  gas
stream by centrifugal action in a  cyclone separator and mist elimination
        36
section.    Variable throat venturi scrubbers are generally the favored type
of scrubbers for particulate control since pressure drop can be maintained
at constant levels across a wide range of boiler loads.
     Sulfur content of the boiler fuel is important not as it affects
collection efficiency, but from a  corrosion standpoint.  Recirculation of a
low pH (pH less than 3) liquor has  resulted in corrosion problems in partic-
ulate scrubbers.  Low slurry pH results from the absorption of acidic
species (e.g., SCL, SO- and HC1) from the flue gas.  Consideration must
therefore be given to the construction materials used in the contactor.
Fiberglass reinforced polyester or rubber-lined steel are the most common
materials used.  These materials are also resistant to the errosive effects
of the slurries which must be handled in wet scrubbing systems.
     A common operating technique  used to prevent low pH conditions is the
addition of an alkali compound.  The addition of an alkali compound to the
wet particulate scrubber for pH control results in the recirculation of a
scrubbing slurry with sufficient dissolved alkalinity to absorb significant
amounts of S0« from the flue gas,  thus forming a combined particulate
matter/S02 removal system.  For example, if sodium carbonate (Na2C03)  is
used as the chemical for pH neutralization, the overall chemical reaction
that occurs is the following:

          Na2C03 + S02   -v   Na2S03 + C02                      (4.1.3-1)
                                   4-23

-------
If sufficient alkalinity is added to the scrubbing liquor, ther  nigh S02
removals can be achieved.  Flue gas desulfurization processes are described
in Section 4.2.
     4.1.3.1.2  Development status.  Particulate control  by wet  scrubbing  is
a well-established technology.  The use of wet scrubbers  in Great Britain
for cleaning boiler flue gases dates back to 1933.  However, this technology
has only been adapted within the last 10 to 20 years to control  fly ash
emissions from power boilers in the U. S.  Since the early 1960's, wet
scrubbing has been applied to fossil fuel-fired boilers in the U. S. for
combined particulate collection and S02 absorption.
     4.1.3.1.3  Applicability to industrial boilers.  Wet scrubbers are
applicable to both coal- and oil-fired industrial boilers.  The two major
considerations in their use are (1) fuel sulfur content and (2)  disposal of
a wet sludge versus a dry product as collected by ESPs, fabric filters, or
mechanical collectors.  The sulfur content of the fuel can impact the use  of
a wet scrubber in two ways: if no S02 removal  is desired  the use of a wet
scrubber on high sulfur fuel-fired units will  require that the scrubber be
constructed of a high quality corrosion-resistant material.  However, if SO,,
removal is required the wet scrubber can serve as the single control device
for both S02 and particulate, thus reducing the total cost of add-on
controls over a wet scrubber for S02 removal and a fabric filter or ESP for
particulate control.
     The chloride content of the coal is also important.   Chloride build-up
in the scrubbing liquor, resulting from absorption of chloride species
present in the flue gas, can result in low pH liquor with potential stress
corrosion of the scrubber vessel.
     4.1.3.2  Factors Affecting Performance.  Factors that affect scrubber
performance include:
       •  scrubber design
       •  liquid-to-gas ratio (L/G)
       •  gas velocity
       •  energy consumption
       •  particle size distribution
                                    4-24

-------
       •  particulate loading at the inlet to the scrubber
       •  construction materials
       •  collection of wetted particles by cyclones and mist eliminators
     Scrubber design has an important effect on the amount of particulate
matter that can  be removed from the gas stream.  Of the several  general
scrubber types (plate tower, packed tower and venturi) plate towers and
Venturis are the best choices for particulate removal.  Packed towers are
not generally well-suited to particulate removal.  Multiple plate towers are
effective in removing particulate matter over 1 micron in diameter, but
venturi scrubbers are more effective than plate towers on submicron
particles. Plate towers do not resist plugging and scaling as well  as
Venturis do, but the use of a mechanical collector to remove the bulk of the
fly ash particles upstream of the scrubber can help alleviate these
problems.  Plate towers are well equipped to handle the high liquid rates
and greater residence times that might be required for simultaneous SO^
control.
     Several features of venturi scrubbers make them a practical choice  for
particulate removal  by wet scrubbing:
       •  high particulate removal  capability,
       •  relatively low scaling potential, and
       •  easily controllable pressure drop.
Venturi scrubbers generally consume more electrical energy than plate
towers.
     Although the performance of a venturi scrubber depends directly on  both
the L/G and the  gas  velocity past the droplets, the gas phase pressure drop
                                                           38
is the major factor influencing particulate matter removal.    As shown  by
Figure 4.1-11, fractional removal efficiency increases with increasing gas
phase pressure drop and subsequent increasing energy expenditure.
     For this reason, venturi scrubber applications generally include a
variable throat system (enabling control of pressure drop) to allow a
                                                             37
constant efficiency to be maintained at varying boiler loads.    Pressure
drops across venturi scrubbers generally range from 1.5 to 30 kPa (5 to
100 in w.g.) depending upon the application and the desired removal
                                     4-25

-------
                       COLLECTION EFFICIENCY VS. PARTICLE SIZE
                                                                                   99.99
X
o
LLJ
5
>
m
z
UJ
LL
U.
01

z
O


o
01
O
U
                                                                                   90.00
                                                                                   80.00
                                                                                10.0

                             PARTICLE DIAMETER IN MICRONS


                                                                                39
    Figure 4.1-11.   Venturi scrubber  comparative  fractional efficiency curves.
                                        4-26

-------
efficiency.   The emissions data for venturi  scrubbers presented later in
this chapter show that pressure drops range  from about 2.4 to 5.0 kPa (8 to
20 in w.g.)  for applications on coal-fired industrial boilers.   In general,
gas velocities through the venturi  throat range from 61  to 183  m/s (200 to
                                                                        3
600 ft/s) while liquid-to-gas ratios (L/6) vary from 1.0 to 2.0 liters/m
(8 to 15 gal/1000 ft ).    At gas-side pressure drops of less than 5.0 kPa
(20 in w.g.), good initial liquid distribution is important to  achieving
                                         38
high particulate collection efficiencies.
     The collection efficiency of a venturi  scrubber decreases  when the size
                                                           41
of the particles to be collected is in the submicron range.    Thus it is
important to take the size distribution of the particles to be  removed into
account when designing the scrubber.  Figure 4.1-12 demonstrates the
relationship between aerodynamic cut diameter and the pressure  drop, with
the liquid to gas ratio as a parameter.  The aerodynamic cut diameter is the
particle size.that is collected with an efficiency of 50 percent by a device
such as an Anderson, Pilot, or Brinks impactor.  Figure 4.1-11  also
illustrates  the relationship between wet scrubber performance and particu-
late size distribution at constant gas-side  pressure drop.
     In a plate tower, an effective way of increasing particulate removal is
to increase the velocity of gas through the  plates (trays).  Adding trays
does not necessarily improve particulate removal, but increasing the
pressure drop across a single tray does.
     The transient, nonsteady state periods  of industrial  boiler operation
are critical in terms of the control system's performance.  Variations in
temperature, airflow, and particulate loadings which affect system
performance  are typical of the varying load  conditions often encountered
                        40
with industrial boilers.    However, with a  system designed for maximum load
and particulate loading, outlet emissions during low or transient load
conditions will be less than the design emission rate.
4.1.4  Multitube Cyclones
     The collection mechanism, status of development, applicability to
industrial boilers, and factors which affect the performance of multitube
cyclones are discussed in this section.
                                   4-27

-------
                                            PRESSURE DROP, inches w.c.
ro
Oc
         5.0
        a.

       oc
       UJ
       I-
       LU


       < 1.0
       O

       IO.B
O
O
cc
UJ
          0.1

           0.25
                                 3.94
                                                                      39.4
                                     MM
              I     I    I   I  I  I ill
I    J  J   IJJJ 11
                                  2.5
                    25
                                               PRESSURE DROP, kPa
                                                                                            I-H
I     J   J J  I  I 11
                    250
                          Figure 4.1-12.   Aerodynamic cut diameter versus gas pressure drop with
                                         liquid-to-gas  ratio (L/G)  as  a parameter.41

-------
     4.1.4.1   Process Description
     4.1.4.1.1   System.   Cyclones are mechanical  collectors which remove
participates  from a gas  stream by an inertial  impaction mechanism.  At the
entrance of the cyclone  a spin is imparted to  the particle-laden gas.  This
spin creates  a centrifugal  force which causes  the particulate matter to move
away from the axis of rotation and towards the walls of the cyclone.
Particles which contact  the walls of the cyclone tube are directed to a dust
collection hopper where  they are deposited.
     In a typical single cyclone the gas enters tangentially to initiate the
spinning motion.  In a multitube cyclone the gas approaches the entrance
axially and has the spin imparted by a stationary "spin" vane that is in its
path.  This allows the use  of many small higher efficiency cyclone tubes,
with a common inlet and  outlet, in parallel to the gas flow stream.
Figure 4.1-13 illustrates the configuration of the individual tube and an
assembly of such tubes in a multitube cyclone.
     One variation of the multitube cyclone is two similar mechanical
collectors placed in series.  This system is often referred to as a dual or
double mechanical collector.  The collection efficiency of the dual
mechanical collector is  theoretically improved over that of a single
mechanical collector.
     4.1.4.1.2  Development status.  Fly ash collection by multitube
cyclones is a well established technology.  It has been used for many years
to limit particulate emissions from coal-fired industrial and utility
boilers and to reduce erosion of downstream ductwork and equipment.
Multitube cyclones were  the most common type of mechanical  collector used
for fly ash control before more stringent emission regulations were enacted.
However, in many cases they now function as precleaning devices to reduce
grain loading to the primary collection device.
     4.1.4.1.3  Applicability to industrial boilers.  Because of their
modular configuration, multitube cyclones are applicable to all sizes of
coal- and oil-fired industrial boilers.  There are several  operational
factors associated with  industrial boilers that affect mechanical collector
                                     4-29

-------
 GAS
OUTLET
                          PARTICLE
                          DISCHARGE
                                                    GAS
                                                   INLET
                                                           ^-DUST-LADEN
                                                                       GAS
        Figure 4.1-13.  Schematic of a multiple cyclone and detail
                        of an individual tube.
                                    4-30

-------
performance and limit applicability as the sole PM control  device.   These
and other factors are discussed in the next section.
     Application of the dual  mechanical  collectors is currently limited to a
few coal-fired boilers operating under relatively steady steam demand
conditions.
     4.1.4.2  Factors Affecting Performance.   The most important design
factors affecting performance for a cyclone are the inlet gas velocity, the
diameter of the tubes, the number and angle of axial  vanes, the construction
materials, and the system pressure drop.   Most multitube cyclones are
axial-gas entry units designed for gas velocities of  25.4 to 35.6 m/sec
(5,000 to 7,000 ft/min) in the entry vane region.  Such high velocities
require the use of hard alloy materials  for the vanes (gray or white iron or
                                           45
chromehard steel) to minimize vane erosion.    Figure 4.1-14 is a
theoretical curve that presents the variation of the  collection efficiency
resulting from the variation  of the inlet gas velocity.
     The performance of any mechanical collection system is significantly
affected by the particle size distribution of the particulate matter to be
collected.  Figure 4.1-15 shows that the collection efficiency of a cyclone
increases as the percentage of larger particles increases.   Particle collec-
tion efficiency for most cyclonic devices varies inversely with the diameter
of the collecting tube.  A reduction in  tube diameter increases the radial
force acting upon the particles so that  their transit to the wall region and
                             45
their removal  is accelerated.    Figure  4.1-15 illustrates comparative
collection efficiencies for two axial-entry cyclones  with diameters of 15.2
and 30.5 cm (6 and 12 inches), respectively, as a function of the percent of
                 4ft
dust under 10  ym.
     Operational procedures related to the boiler/control device system that
hamper mechanical collector performance  include transient operations such as
startup, shutdown, or emergency upsets and load variation.     In addition,
air in leakage, cyclone corrosion, particle reentrainment, tube plugging,
pressure drop  and the degree of flyash reinjection will affect mechanical
collector outlet emissions.     Large load swings significantly affect
removal efficiency.  Changes  in the sulfur content can alter the acid
                                     4-31

-------
  100
  10
           0.5          1.0           1.5          2.0
                VELOCITY RELATIVE TO DESIGN CONDITION
2.5
Figure 4.1-14.   Variation  of  a single cyclone collection efficiency  with
               gas  velocity.46
                                4-32

-------
            100
          > 90
          O


          8-
          u.
          u.
          u 80
          Z
            75
            70
            05
          U
          Ul
          s
                PRESSURE DROP

                  I    i    i
•3 in. WATER GAUGE

 I    I    I    I	
                  10  20  30  40  50  60  70


                    pcrnm OF DUST UNDER 10 /4n
                80
Figure 4.1-15.   Typical overall  collection efficiency of
                axial-entry  cyclones.48
                            4-33

-------
concentration in the flue gas which can result in corrosion of the cyclone.
At constant load and inlet particle size distribution, outlet emissions will
be proportional to inlet mass loading.  Therefore, a large increase in fly
ash loading (which could result from variations in load, coal ash content,
soot blowing or fly ash reinjection) will increase emissions.
     Proper mechanical collector maintenance is essential in sustaining the
desired removal efficiency.  To avoid efficiency losses due to corrosion of
the cyclone from acid condensation or particle abrasion, the cyclone should
be constructed of materials that will withstand the highest expected loading
of potentially corrosive flue gas components.  Primary considerations to be
                                                         44
used in evaluating the construction materials needed are:
       •  Gas temperature
       •  Abrasiveness of the dust particles
       •  Corrosiveness of the gas stream
If the gas stream is corrosive or the dust particles are abrasive it may be
necessary to use a stainless steel alloy instead of carbon steel in the
construction of the cyclone.
     It is important to accurately monitor the pressure drop across the
cyclone so that any plugging can be detected.  In addition, the interior
should be inspected on a regular basis for corrosion damage, plugged tubes,
or defective gaskets.  Another area of maintenance that is critical to
efficient mechanical collector performance is the discovery and remedy of
air leakage into the collector.  Leakage can occur at the hopper access
door, hopper discharge valve, hopper casing, or the lower tube sheet.  Air
leakage into a collector hopper can result in reentrainment of collected
particles, thus reducing collector performance.
     One of the most detailed sources of information on mechanical collector
performance is a study conducted jointly by the American Boiler
Manufacturer's Association (ABMA), the Department of Energy  (DOE), and EPA.
Several stoker-fired boilers equipped with mechanical collectors were tested
in this study and particulate emissions tests were conducted at both the
boiler and the mechanical collector outlets.  Based on a review of these
                                     4-34

-------
data,  the following  conclusions  can  be  made  about  the  effect of boiler
                                                        47
operating parameters on  mechanical collector performance:

       •  Figure  4.1-16  shows  that,  for 3  similar  coals, mechanical
          collector  efficiency remained relatively constant with changes in
          boiler  load above  about 60 percent.  However, there was
          significant drop in  collector efficiency at  loads of approximately
          50 percent and less.

       •  There was  considerable scatter in  the test data for some units as
          a  result of variable process  conditions  and  fuel types.  However,
          the results showed that particulate matter emissions from both the
          boiler  and mechanical  collector  (in terms of lb/10  Btu) tended to
          increase as the boiler load increased.   This trend can be seen in
          Figures 4.1-17 and 4.1-18  where  boiler and mechanical collector
                                                                   231
          outlet  emissions are plotted  as  a  function of boiler load.
          Although these figures illustrate  emissions  from a single boiler,
          they are representative of the overall trends from the data set.
          Figures  4.1-18  also  illustrates that controlled emissions from
          this  boiler  remained fairly steady, but showed a trend of
          increased  emissions  at  boiler loads greater than 50 percent.  This
          trend was  also  seen  for other boilers.  The sharp increase in
          emissions  at very  low loads was attributed to the reduced
          mechanical collector efficiency at the unusually low firing rate
          obtained at  this one site.

          In  general,  no  significant correlations were observed between
          mechanical collector performance  and overfire air levels, or
          excess air levels.
                                    4-35

-------
LU

O
I—
o
       100--
         90"
         80-
         70--
         60"
         50"
                      A
         10"
                        20
                           Legend

       ©  Coal  A* - 7.7%  Ash,13320 Btu/lb

       B  Coal  B* - 8.1%  Ash,12869 Btu/lb

       A  Coal  C* - 7.3%  Ash,12832 Btu/lb
  40           60

BOILER  LOAD(%)
80
100
           Figure  4.1-16.  Mechanical  collector  efficiency versus  boiler load
                            (spreader stoker boilers).47
       Note: Data shown Is from two different boiler collector systems.
           Coal A was fired In one, while coals B and C were fired in
           the other .
                                           4-36

-------
                     Boiler Type:  Spreader Stoker  75,000 Ib/hr steam
                     Coal  Properties:
      10.0-
                                     O- 8.04% Ash, 12869 Btu/lb
                                     +- 4.42% Ash, 13860 Btu/lb
                                     A- 7.32% Ash, 12832 Btu/lb
10
 o
 c
 o
 •I—
 I/)
 10
 •^

 LU
(O


u
•^
•p


Q.

•o

-------
      1.0 -
                         Boiler Type:   Spreader Stoker  75,000 Ib/hr steam
                         Coal Properties:
                       O- 8.042 Ash, 12869 Btu/lb
                       -r- 4.42S Ash, 13860 Btu/lb
                       A- 7.322 Ash, 12832 Btu/lb
~    0.8 _
J-:

•S-
     0.6 -
     0.4 -
     0.2 -
                  '9-
V " " <•/''
4 / s
&*&£'' ,„ >
            •'/,
                               •'•»?,
                  fjrss* f .    j-v -'i-y ,j>-r x
                  ••   ?*&*%£,•'  * 7 s*  s j-
                   r  ' f ^•SJ}&      s  •%• ^-^
                  /^-   //&%&   '/ * /' fSf&

                  -?A% wAft« -X
                                                                     •^,1
                    —r
                     20
         SHADED AREA EMPHASIZES THE DATA TREND
                                 	1	
—T
 40
~nr
 60
                                                           80
                                                    100
                                      Boiler Load
                                (2 of rated capacity)
                   Figure  4.1-18.
                  Controlled particulate  emissions  vs.
                  boiler  load."1
                                           4-38

-------
       •  The data did show that mechanical  collector collection efficiency
          was lower when there were relatively high percentages of small
          particles (less than 10 microns in diameter) at the inlet to the
          collector.  However, no correlations were observed between boiler
          load, excess 02, or overfire air levels and the resulting particle
          size distribution.

4.1.5  Side Stream Separator
     The collection mechanism, status of development, applicability to
industrial boilers, and factors which affect the performance of side stream
separators are discussed in this section.
     4.1.5.1  Process Description
     4.1.5.1.1  System.  The side stream separator system consists of a
single multitube cyclone and a small  pulse-jet baghouse as shown in
Figure 4.1-19.  The boiler exhaust stream is ducted to the cyclone where a
portion (approximately 15 to 40 percent) of the gas is drawn from the
cyclone at the bottom of the tubes just above the ash hopper and ducted to a
fabric filter.  The gas flow from the cyclone and baghouse are then
recombined and exhausted to the stack.  The gas stream from the cyclone to
the baghouse is believed to have a higher concentration of small particles
relative to the total gas stream thereby removing the particles of the size
that are generally not collected efficiently by the mechanical collector
 ,    50
alone.
     As individual units, the cyclone and fabric filter operate as described
in Sections 4.1.2 and 4.1.4.  Together, the fabric filter adds additional
overall removal of particulate to the cyclone thereby improving overall
removal efficiency.  The side stream separator design is based on the use of
a single- or multi-compartment pulse-jet fabric filter.
     4.1.5.1.2  Development status.  The side stream separator system is a
recent development in particulate removal from industrial boilers.  Its
application is currently limited to retrofitting existing mechanical
collectors on spreader stokers firing a limited range of coal types.  As a
result, the existing data base for side stream separators is limited.  In
                                   4-39

-------
                                          STACK
                                          EXHAUST
                                          STREAM
     MECHANICAL
     DUST  COLLECTOR
BOILER
EXHAUST
                      SIDE
                      STREAM
                      FLOW
                                             SIDE STREAM
                                             BAGFILTER
                  LARGE
                  PARTICULATE
FINE
PARTICULATE
                  Figure 4.1-19. Side stream separator.
                                           50
                            4-40

-------
addition,  because the existing installations  are relatively new,  the long
term performance of side stream separators  cannot be documented.   There  are
currently  2 to.3 vendors and an independent consultant offering  similar
devices as retrofits to upgrade existing mechanical  collector performance.
However, vendors do not offer performance guarantees on this device.
     4.1.5.1.3  Applicability to industrial  boilers.  The side stream
separator  is applicable to coal-fired stoker boilers,  but has not been
applied to pulverized coal units.   Application  of this technology is
currently  limited to spreader stokers firing  low or  medium ash coals
(<10 percent).  Application of this device  to other  stoker types  and  coal
types has  not occurred to date.
     4.1.5.2  Factors Affecting Performance.  Most of  the factors that
affect performance of mechanical collectors and fabric filters (previously
discussed  in Sections 4.1.4 and 4.1.2, respectively) also affect  the
performance of the side stream separator.  The  performance of the mechanical
collector  is affected by the diameter of the tubes,  the number and angle of
entry vanes, construction materials, and pressure drop.  Fabric  filter
performance is affected by air-to-cloth ratio,  filter  fabric, cleaning
mechanism, baghouse temperature and fuel properties.
     The performance of mechanical  collectors is also  affected by the
proportion of small fly ash particles (less than 10  microns in diameter) at
the inlet  to the collector.  However, this  factor should have less impact on
side stream separators since the fabric filter  used  with the mechanical
collector  is relatively efficient with respect  to fine particles.
     As discussed in Section 4.1.4, mechanical  collector efficiency drops
off rapidly at low boiler loads.  This factor will result in decreased  side
stream separator efficiency at low loads, unless uncontrolled emissions  at
                                                                     49
low loads  are reduced enough to compensate  for  the reduced efficiency.
Currently, side stream separators are equipped  with  constant flow rate  fans.
Therefore, as boiler load decreases a higher percentage of the total flow is
routed to  the side stream baghouse.  This affect may act to compensate  for
reduced mechanical collector efficiency at  low  loads.   However,  present data
are insufficient to adequately assess the performance of sidestream
separators at lower loads.
                                    4-41

-------
4.1.6  Emission Data
     Available emission data for ESPs, fabric filters, wet scrubbers,
multitube cyclones, and side stream separators are presented in this
section.  The sites from which this test data were gathered are referred to
here as Plant A, B, C, D, etc.  A description of each test site including
the particulate control equipment tested, the complete test data, and any
unusual emission testing or control device operational factors that impact
the validity of the test results are presented in Appendix C.  Appendix C
also contains the  references for complete test reports on each site.
     The data base gathered during the course of this study was reviewed
extensively to assure  that each test met two important criteria.  First, all
of the PM emission tests were reviewed to determine if the test methods
complied fully with EPA Method 5 specifications.  Those data found to have
been collected with emission test methods not meeting Method 5 specifica-
tions have not been included in this chapter or Appendix C.
     Secondly, a thorough analysis of the remaining valid test data was
conducted to assure that no unusual boiler or control device operating
conditions affected the test results.  The data was also evaluated in an
effort to characterize, as fully as possible, the important design and
operating parameters of each emission control system.  Emission test data
collected under nonrepresentative conditions or data collected from systems
where critical control device design and operating parameters could not be
documented are generally not presented in this section.  Exceptions to this
procedure were made in a few cases, however.  Test data that can be used to
demonstrate an important point about control system performance (for
example, the performance of mechanical collectors at low boiler load) were
included in this section.  Also, where no complete emission test data was
available, such as for ESPs on industrial oil-fired boilers, data from other
studies are presented  and used to characterize control system performance.
     Appendix C provides further information for the majority of the data
presented in this  section.  Appendix C also includes data that was collected
with valid test methods but was considered not to be representative of well
designed and operated  systems.  Specific documentation of why these tests
                                     4-42

-------
were not considered to be representative is specified in Appendix C.
Following such a procedure allows this section to focus primarily on
emission data that represents the PM control  levels achievable with well
designed, operated, and maintained systems.
     Method 5 tests are normally run at a sample box temperature of about
120°C (248°F).  Method 5 specifications state that tests can be run at
higher temperatures as specified in individual emission standards [Subpart D
for fossil  fuel  fired industrial boilers larger than 73.3 MW (250 x 10
Btu/hr) allows temperatures up to 160°C (320°F)].  In many cases, variations
in sample box temperature across this range have little or no effect on the
amount of PM matter emissions measured.  However, it appears that in
measuring particulate matter emissions from boilers firing high sulfur coal,
low sample box temperatures can lead to condensation of gaseous S03.  This
condensation may result in a difference in measured emissions, depending  on
coal sulfur content and sample box temperatures.  Some of the emission data
in this section were taken at higher sample box temperatures of [up to 177°C
(350°F)] in an attempt to prevent SO., interference.  Data collected at the
high sample box temperature is referred to as high temperature Method 5
data.  High temperature data is presented and discussed where available.
     This section concludes with a discussion of the available data on the
performance of post-combustion PM controls with respect to fine particulate
(Section 4.1.6.6) and data on visible emission (Section 4.1.6.7).
     4.1.6.1  ESP Emission Data.  This section presents data from emission
tests performed on oil- and coal-fired boilers equipped with ESPs.  The only
data available for ESPs on oil-fired units were gathered in a study of
utility boilers, but the technology is directly transferable to industrial
oil-fired boilers.
     Particulate emission data from coal-fired industrial boilers equipped
with ESPs are shown in Figure 4.1-20.  Method 5 data were collected for both
spreader stokers and pulverized coal-fired units.  Specific collection area
(SCA) is the most important control system design and operating parameter
for ESPs.  Figure 4.1-20 lists both the design and operating SCA, as well as
                                    4-43

-------
                                           O  Individual  Tests
                                           -H  Average of  Tests

30
(.070)"
f 3
VI h-
in ca
18 "2 20.
3 ^ (.047)
3
O "S
n>
10.
(.023)
Plant
,. Spreader Stokers
w/ Upstream MC
0
•• ^

O
Q
... Pulverized Coal -Fired ...„
Boilers w/ No Upstream MC
o
"o^
0
o ._
o
X O o
O ~tr
1 1 1 1 1 1 1 1 1 1 1 1 1
KKKPNbNbZZZZWWW
Boiler Number 9 7 8 26 27 25 29 RC BB PG
Design Capacity 125 75 100 200 300 300 430 430 430 430 380 300 300
(103 Ib steam/hr)
Operating Capacity 99- 103- 93- 87- 76 52- 95- 96- 99- 99- 72- 86 91
(% of Design) 102 106 98 89 59 98 97 100 100 85
SCA (ft2/103 acfm)
Design 128 132 152 349 344 344 90 96 98 98 300 369 325
Operating 128 128 160 397 542 634 90 96 98 98 348 348 364
Fuel Typec BBB BSBSBBBBB---
Fuel Sulfur .57 - *>l .73 .54 .63 ^1 •*,! *1 •».! -
(Wt %)
Fuel Ash 11.4 12.0 11.2 6.6 8.3 5.4 -x-12 -\-12 -^12 -\-12
(Wt %)
Fly Ash
Reinjection No No No Yes Yes Yes No No No No No No No
             Figure 4.1-20. Electrostatic precipitator emission data.'
aAll  tests ordered from left to right by increasing SCA
bAll  tests done on a hot side ESP
cB-Bituminous coal, SB-Sub Bituminous coal
                                         4-44

-------
coal  sulfur content, boiler load during the test,  and whether or not  there
was fly ash reinjection during the emission test.
     All but one of the tests were conducted on  boilers  with  the ESP  located
downstream of the air preheater (cold-side ESP).   At Plant  N, however,  the
ESP is located upstream of the air preheater (hot-side ESP).
     All of the emission tests shown in Figure 4.1-20 were  conducted  on
boilers firing low sulfur coals (1 percent sulfur  or less).   As  discussed
earlier in Section 4.1.1, a larger collection area is generally  required to
achieve a given particulate collection  efficiency  on low sulfur  coal  units
than on high sulfur coal units.  Thus,  the achievable emission control
levels shown in Figure 4.1-20 would be  achievable  on boilers  firing high
sulfur coal with SCAs equal to or less  than those  shown.
     Average emissions were less than 20 ng/J (0.047 Ib/million  Btu)  in each
of the six tests of spreader stoker boilers equipped with ESPs.   Each of the
spreader stokers tested had mechanical  collectors  operating upstream  of the
precipitator.  Operating specific collection areas of the cold side ESPs on
spreader stokers ranged from 419 to 1302 m2/(103m3/s) (128  to
397 ft2/103 acfm).  The hot side ESP at Plant N  operated with SCAs of 1774
to 2075 m2/(103m3/s) (542 to 634 ft2/103 acfm).
     Six of seven tests on pulverized coal-fired boilers equipped with  ESPs
averaged 23 ng/J (0.053 Ib/million Btu) or less.   A seventh test averaged
30 ng/J (0.070 Ib/million Btu).  Operating SCA's ranged  from  295 to
1199 m2/103m3/s) (90 to 364 ft2/103 acfm).  The  highest  average  emissions
were observed from the ESP with the lowest SCA:  Boiler 26 at  Plant Z  has an
SCA of only 295 m2/(103m3/s) (90 ft2/103 acfm).
     In summary, average emissions were 30 ng/J  (0.07 Ib/million Btu) or
less in all 13 tests.  These emission levels were  achieved  despite
relatively low SCAs in four of the tests and despite the fact that the
boilers tested were burning low sulfur  coal (1 percent or less).
     The available emission data for oil-fired boilers equipped  with  ESPs
are considerably less well characterized with respect to SCA  and boiler load
during the tests.  Table 4.1-2 presents the emissions data, boiler size,  and
fuel  characteristics for seven utility  boilers equipped  with  ESPs.
                                    4-45

-------
           TABLE 4.1-2.
SUMMARY OF PARTICIPATE EMISSION  TEST  DATA  FOR
ESPs  ON OIL-FIRED BOILERS48
Boiler Controlled
Capacity Participate Emis;
Company (MWe) ng/J (lb/10° B1
Polaroid Corp.
New Bedford
Boston Edison .
Mystic Station0



Boston Edison
Mystic Station



Control
;ions Efficiency
Fuel
tu) (%) % Sulfur
10 23.7 (0.055) 40 0.
10 30.1 (0.070) 51 0.
48 48.6 (0.113) 38 2.
48 64.5 (0.150) 57 2.
48 12.9 (0.033
48 66.2 (0.154
48 66.2 (0.154
71 2.
2.
2.
593 28.0 (0.065) 83T 2.
17.6 (0.041)6 -, 2.
595 43.9 (0.102
21.1 (0.049
589 30.1 (0.070
69' 2.
e -f 2.
78T 2
19.4 (O.'045je - 2.
Hartford Electric
Light Co.
Middletown Station
United Illuminating Co.
Bridgeport Harbor '
Consolidated Edison
Ravenswood
Astoria3
119 30.1 (0.070) - 1.
117 24.5 (0.057) - 1.
119 28.8 (0.067) - 1.
406 64.5 (0.150) - 1.
405 54.2 (0.126) - 1.
600 7.2 (0.017) 16 0.


320 3.5 (0.008) 51 0.
350 5.2 (0.012) 54 0.
355 5.2 (0.012) 40 0.
385 5.2 (0.012) 45 0.
7
7
4d
4.
45
3d
3
2
2
2
2
2
2
95d
8Fd
79d
80"*
77d
3

3
37
3
37
% Ash
_
-
_
-
-
-
-
-
-
.-
-
-
—
0.09
".07
0.07
0.08
0.09
0.02

_
_
_
~
Test Sponsor
Industry

Industry




EPA





Industry


Industry
Industry

Industry



 ESP originally designed for coal.
 ESP originally designed for coal,  later modified for oil.
CSCA =  375 ft2/103 acfm (design).
 Oil additives used to prevent boiler fouling and corrosion.
eBased  on EPA Method 5 high temperature method (320°F).
 Efficiency calculation based on low  temperature  Method 5 inlet  and outlet data

-------
     Most of the test data presented were performed by industry, however, as
noted in the table one series of tests were performed by EPA for the purpose
of this study.   Most of the precipitators were designed to collect coal  fly
ash.   Generally, the collection efficiency of the ESP is lower when it is
used  to control  fly ash from oil  combustion than when it is used to collect
coal  fly ash.   The lower collection efficiency is due primarily to
differences in  particle resistivity, size distribution, and surface
properties between oil and coal fly ash.   Thus, larger ESP collection areas
may be required to achieve a given level  of control  when the boiler is
switched from coal to oil.
     The Method 5 data in Table 4.1-2 shows controlled emissions ranging
from  3.5 to 66.2 ng/J (0.008 to 0.154 lb/106 Btu).   The high temperature
Method 5 data collected at Boston Edison  indicates  that ESPs can achieve
emission levels of below 22 ng/J (0.05 lb/10 Btu).   The boiler was firing a
high  sulfur oil  (about 2.3 percent) and the ESP was originally designed to
collect fly ash generated from oil combustion.  Average emissions for the
three EPA test runs at Boston Edison was  20 ng/J (.045 lb/106Btu).
     4.1.6.2  Fabric Filter (FF) Emission Data.  Data presented in this
section are for coal-fired boilers equipped with fabric filters.  No data
were  available for FF applications to oil-fired boilers.
     Figure 4.1-21 shows emission test data for both spreader stoker and
pulverized coal-fired boilers.  Included  in Figure 4.1-21 are boiler load,
design and operating air-to-cloth (A/C) ratios, and percent ash in the fuel.
All tests were conducted on reverse-air cleaned fabric filters.
     Average controlled emissions were less than 15 ng/J (0.035 lb/10  Btu)
in all four tests on spreader stoker boilers equipped with fabric filters.
Air-to-cloth ratios of the fabric filters tested ranged from 2.3 to 3.5
ft/min.  Two tests on pulverized coal-fired boilers showed controlled PM
emissions of less than 16 ng/J (0.037 lb/10  Btu) with operating A/C ratios
of 1.5 and 2.2 ft/min.
     4.1.6.3  Wet Scrubber Emission Data.  Particulate emission test data
for spreader stoker and pulverized coal-fired industrial boilers equipped
                                    4-47

-------
  EPA  Sponsored  Test
r * i i\du.j i»i jr I *=.J W
- Average of Tests
20

w
C
*WJ h- —
in co
T—
c tp
tQ o
»— i

-------
with wet scrubbers are presented in Figure 4.1-22.  All of the data are for
wet scrubbers designed for combined SCL and PM removal.
     As discussed in Section 4.1.3, venturi and tray scrubbers are the most
suitable type of scrubbers for PM removal; test data were available for both
types of scrubbers.   Higher gas-side pressure drop across the scrubber
generally results in lower controlled emissions.
     Emission test data for wet scrubbers applied to pulverized coal-fired
industrial  boilers are shown in the right sections of Figure 4.1-22.
Average controlled emissions in all 4 tests were less than 35 ng/J
(0.081 lb/106 Btu).
     Emission test data for wet scrubbers applied to coal-fired spreader
stokers are shown in the left three sections of Figure 4.1-22.  High
temperature Method 5 data are available for venturi  scrubbers and tray-type
scrubbers.   Average high temperature controlled emissions from the tray-type
scrubber tested (Plant AAA) is below 35 ng/J (0.08 lb/106 Btu) while  average
high temperature controlled emissions from the venturi scrubber (Plant LL)
range from 45 ng/J (0.10 lb/106 Btu) to 38 ng/J (0.08 lb/106 Btu) for two
similar boiler scrubber combinations.  Both plants have upstream mechanical
collectors  as particulate precleaners prior to final particulate removal in
the S02 scrubbers.
     Average controlled emissions from two entrainment type scrubbers
(Plant 0) were 82 and 104 ng/J (0.191 and 0.241 lb/106 Btu), at operating
scrubber pressure drops of 3 kPa (12 in. w.g.).
     Limited test data is also available for wet scrubber systems applied to
utility coal-fired boilers.  This data is presented in Table 4.1-3.  The
scrubber systems tested are all designed for SOp as well as PM removal.
     4.1.6.4  Mechanical Collector Emission Data.  Particulate emission data
for coal-fired boilers equipped with single and dual mechanical collectors
are presented in this section.
     Figure 4.1-23 shows Method 5 data for single and dual mechanical
collectors  installed on five spreader stokers, while Figure 4.1-24 shows
data for single mechanical collectors installed on three mass fed stokers.
Dual mechanical data were available only for one spreader stoker under
                                    4-49

-------
f Method S - Low
Temperature
A Method 5 - High EPft Test
Temperature
Q Industry Test
"T- Average of Tests




120
(.279)


100
(.233)

i~
« 3 80
« « (.186)
«i ^.
»• ^
It 60
r ? (.wo
M
(k
40
(.093

20
(.047

Plant
Boiler No.
^ „—..._, ,._„-- ^
•




9 J L




•
_ _
o
.6 6 |

o
in r

,,
h 7
0
nui 1 1 *en tun
Scrubbers







O
9 T
o 4-
•¥- o
T


T i


i
A

1 1 1 1 1 1 1 1 1 1 1 1
QQ QQ 00 L MM LL LL MM LL LL LL MM SS AAft A
4 5 5 3 2 20 20 3 19 19 19 1 4 3 1





>^ D*i1u*t*49A<4 ^nm^ R/i4 1 AVC
f— *. — * 	 A
•>














i


i
i

ciiwi a IIMIICI|I.
Scrubbers


Q



Q

o v

™ ^

o







U......4
Sieve Tray
Tower
Combination
Scrubbers















fQ
TT"


• CIIVUI 1
Spray Tower
Combination
Scrubbers
















O
JL

r i i i i
HA 00 00 00 TT TT
132
Design Capacity
(10e lb steam/hr) - - - 100 	 100 60 60 100 100 100 80 - - - -
(10° Btu/hr) 202 202 202 295 236 236 295 236 236 236
137 137 100 100
Operating Capacity 80- 80- 95 93- 75 85- 85- 75 100 73 73 86- 79- 81 92 92 80- 88 100 100 100 100
(» of Design) 100 95 97 91 91 92 83
Emission Controls1* 55527777777333 3
Scrubber Pressure
Drop
100
344 6611


(In H-0 guage)
Design 22 22 22 10 17 17 17 17 17 17 17 13 13 13 13 13 12 12 - - -. -,
Operating 8 8 8 10 16 17.3 17.3 17.5 18.1 19.3 19.3 7.5 7.5 7.5 7.5 7.5 12 12 12/4SC 17/4SC 3° 9°
Design L/6 11.4 11.4 11.4 10 10 10 10 10 10 10 20 20
(gpn/103 acf«i)
20 20 4 17
Fuel Sulfur 2.4- 2.4- 2.4- 0.8 2.4- 2.54 2.54 2.4- 2.4- 2.6 2.6 2.3 2.4 2.14 1.33 1.33 2.33 2.33 3.5 3.5 3.9 3.9
(wt I) 3.4 3.4 3.4 3.4 3.4 3.4

Fuel Ash (wt S)
10   10
Fly Ash ReInjectIon Yes   Yes
10   7.2   10  10.4 10.4   10   10  11.4  11.4  9.4  8.0   5   4.4
No   No  Yes   Yes  Yes  Yes  Yes  Yes   Yes   No   No  No   No
4.4  10.5   9.9 12.3  12.3 14.7 14.7
 No   No   No   No   No  No  No
                    Figure 4.1-22.    Emission  data  for  wet scrubbers.'
                                                      4-50

-------
Notes for Figure 4.1-22.

aVenturi tests ordered by increasing operating pressure drop.

All  other tests ordered by decreasing percent ash in fuel.


 PM and SO  control  devices

     1.   Ventur/spray tower
     2.   95 percent efficient mechanical  collector, FMC venturi dual
          alkali scrubber.
     3.   Mechanical collector, multi-venturi flex tray dual  alkali
          scrubber.
     4.   Mechanical collector, Zurn entrainment type scrubber.
     5.   80 percent efficient mechanical  collector, venturi  dual  alkali
          scrubber.
     6.   Venturi/seive tray scrubber.
     7.   Mechanical collector, venturi  dual  alkali scrubber with cyclonic
          separators.


cVenturi Ap/ sieve tray Ap.


 Ap for venturi only.
                                    4-51

-------
                TABLE 4.1-3.  DATA ON PARTICULATE  SCRUBBING, SYSTEMS INSTALLED ON COAL-FIRED UTILITY  STEAM

                              GENERATORS 48
en
ro
Bo Her Type*
and Stzeb
(X Sulfur)
125HW-PC
(P.5)
125MH-PC
(0.5)
125MW-PC
(0.5)
350MW-PC
(0.7)
116MU-PC
(0.5)
380HH-PC
(2.1)
150MH-PC
(2.1)
Partlculate
Control
Before Scrubbing
Mechanical
Mechanical
Mechanical
None
Mechanical
Mechanical
and ESP
Mechanical
Separate Particulate
Control by Scrubbing
None
None
None
None
Flooded Disc Scrubber
None
None
SO? or SO; and Syslea Pressure Drop
Partlculate Control Kilopascals
Dy Scrubbing (Inches II^O)
Venturi Scrubber
Carbonate
Venturi Scrubber
Carbonate
Venturi Scrubber
Carbonate
Venturi Scrubber
Lime
Packed Absorber
Lime
Venturi
Venturi
3.8 (15)
3.8 (15)
3.8 (15)
4.5 (18)
3.0 (12)
2.5 (10)
2.5 (10)
Partlculate Emissions
at the Scrubber Outlet
ng/J (lb/106 BTU)
20-24 (0.046-0.055)
22 (0.051)
14-18 (0.033-0.041)
8-10 (0.019-0.023)
16 (0.037-0.038)
19-21 (0.045-0.048)
9-30 (0.022-0.070)
Number
of Test
Tests Method
2 EPA 5
2 EPA 5
3 EPA 5
3 EPA 5
2 EPA ^5
2 EPA 5
10 EPA 5
        DMH • HH Output

-------
 O  Industry Test

 •f  Average of Tests
Single Mechanlcal/Relnjectlon In Use
Single Mechanical/  Dual Mechanical
No Re1nject1on     Collector/No

600
(1.39)
500
(1.16)
VT
| -? 400 _
a « (0.93)
a S
to •—
3 ^ 300_
5 ^ (0.70)
L. c
10
a.
200 _
(0.47)

(0.23P
Plant



8 
-------
                                O   Industry Test

                                4-   Average of Tests
400
(0.93)
300
£ ~ (0.70)
O 3
I/) CO
IS)
•f- 10
E 0
UJ t-H
S 5 200 _
* ~ (0.47)
1 ?>
i * (_
^* &
rtJ
100 _
(0.23)
Plant


8
O ?
1
*
1 1 1
R HH R
Boiler No. -
Boiler Typeb VG CG VG
Design Capacity 90 70 90
(10J Ib steam/hr)
Operating Capacity 45 48- 61-
(% of Design) 50 69
Fuel Sulfur (Wt %) 2.23 1.82 2.26
Fuel Ash (Wt %) 8.2 9.0 8.1

r
i
i

©
i i i
R HH H
_
VG CG U
90 70 35
79- 73- 77-
89 103 90
1.89 1.65 0.57
7.2 7.0 8.1
 Figure 4.1-24.  Mechanical  collector emission data for mass
                   fed stokers without fly ash  reinjection.
 All tests ordered from left to right by increasing operating capacity.
JVG-vibrating  grate stoker, CG-chain grate stoker, U-underfeed.
                                4-54

-------
relatively steady steam demand conditions.   Although dual  mechanical
collector data is limited, the available data falls within the range  of
performance for single stage mechanical  collectors.
     Figures 4.1-23 and 4.1-24 suggests  that mechanical  collector perfor-
mance is extremely variable from plant to plant.   This may be a result of
different boiler loads, cyclone tube diameters,  coal types and operation and
maintenance procedures.  Because of this variability, an estimation of
mechanical collector performance from this  data  is difficult.  Figure 4.1-23
shows that average outlet emissions from mechanical collectors applied to
spreader stokers cover a wide range from 50 ng/J (0.12 lb/10  Btu)  to
617 ng/J (1.43 lb/106 Btu).
     Average controlled emissions in 10  of  12 tests on spreader stokers
equipped with single mechanical collectors  were  340 ng/J (0.79 lb/10   Btu)
or less.  The highest average emissions  of  these 10 tests  occurred  at low
boiler load (16 to 17 percent).  Two other  tests on a spreader stoker
equipped with single mechanical collector averaged more than 430 ng/J
(1.0 lb/106 Btu).
     Particulate emission data for single mechanical collectors applied to
chaingrate, vibrating grate and underfeed stokers is presented in
Figure 4.1-24.  Average emissions from the  vibrating grate stoker range from
180 to 230 ng/J (0.42 to 0.53 lb/106 Btu) while  the limited underfeed stoker
data averages 31 ng/J (0.07 lb/10  Btu). Average particulate emissions from
the chaingrate stoker tested range from  65  to 79 ng/J (0.15 to
0.18 lb/106 Btu).
     Only one test was available for mechanical  collectors on oil-fired
boilers.  Two Method 5 tests were performed at Plant ZZ on a 55,000 Ib
steam/hr boiler firing No. 2 oil.  The boiler was operated at 67 percent of
capacity during the tests.  The results  show an  average emission of 9 ng/J
(0.02 lb/106 Btu).  Fuel oil sulfur content in this test was less than
1 percent, and ash content of the oil was reported as "nil".
     4.1.6.5  Side Stream Separator Emission Data.  Particulate emissions
from spreader stoker boilers equipped with  side  stream separators are
presented in this section and Figure 4.1-25.  No EPA tests were performed,
                                     4-55

-------
O Industry Test
4- Average of Tests
80 _
(0.186)

in
1 § 60—
2 « (0.140)
0) — .
3 ^ 40—
5 g (0.093)
a
Q.
20—
(0.047)

O

9 §
"5" @ Q
a f o ® f
y" V "®
^ O 0

1 III
Plant ODD CCC 6G6 EEE FFF EEE BBB
Boiler Number - 1333
Design Capacity 45 70 60 40 100 55 52
(10J Ib steam/hr)
Operating Capacity 68b 71- 74- 84- 85- 99- 97-
(% of Design) 80 80 93 97 105 108
Average Opacity (%) Oc - - 6.9 0 Oc 6C
Fuel Sulfur 0.82 0.80 0.94 1.79 1.67 2.09 0.80
(Wt X)
Fuel Ash 9.7 10.1 4.3 9.0 6.1 8.8 7.8
(Wt X)
% of Flow to 16b 31 30 37 15 15 17
Baghouse
   Figure 4.1-25.   Side stream separator emission data.'
^11  tests ordered from left to right by increasing operating capacity
•'Data presented are averages for all  tests

                                 4-56

-------
but industry has provided test results from seven stoker boilers using
retrofitted side stream separators.   The side stream separator has not been
applied to pulverized coal-fired boilers or mass fed stokers.
     The results show that under relatively steady state conditions,  average
emissions from newly installed and adjusted collectors were less than
72 ng/J (0.17 lb/10 Btu)  at all  seven locations.  Average emissions during
the tests ranged from 52  ng/J (0.12  lb/106Btu)  to 72 ng/J (0.17 lb/106Btu).
All emissions tests were  performed using Method 5.  The boilers tested
operated under relatively steady state conditions and at boiler loads at or
above 68 percent.  No data was collected for low load or variable load
operations.  Percent ash  in the fuel  varied from site to site  and ranged
from 4.3 to 10.1 percent.  The percent of the total  flow sent  to the
baghouse also varied from site to site and ranged from 15 to 37 percent.  It
should be noted that extensive adjustment of the existing mechanical
collectors was required to achieve the emission levels shown in
Figure 4.1-25.53
     4.1.6.6  Fine Particulate Collection Efficiency.   In addition to the
overall collection ability of post-combustion PM control  devices, another
important factor characterizing their performance is the ability to collect
fine and inhalable particulate matter.  In general,  inhalable  particulate
matter is defined as that particulate matter having  an aerodynamic diameter
of 15 microns or less, while fine particulate matter is defined as that
having of aerodynamic diameter of 2.5 microns or less.  (These definitions
are used for discussion purposes only.)
     This section presents the available data for the fine particulate
control capability of ESPs, fabric filters, wet scrubbers and  dual
mechanical collectors.  The limited  data available for single  mechanical
collectors were obtained  with two different particle size methods and the
results were generally inconsistent.   Therefore, data for single mechanical
collectors are not presented here; the ability of mechanical collectors to
capture fine particulate  was discussed qualitatively in Section 4.1.4.  No
data on the fine particulate collection efficiency of side stream separators
(SSS) were available.  However, particle size testing on this  device  has
                                    4-57

-------
indicated that the slipstream to the fabric filter has a high percentage of
particles less than 10 microns, and other data shows fabric filters to have
high collection efficiencies on small particles.    Thus, the SSS should
have higher fine particulate collection efficiencies than a mechanical
collector.
     The data presented in this section indicate that high efficiency ESPs
and fabric filters show the greatest degree of control of fine particulate.
Venturi scrubbers offer limited control of fine particulate from spreader
stokers, but are fairly effective in controlling fine particulate from PC's.
     Table 4.1-4 shows the available data for ESPs, fabric filters and wet
scrubbers.  Also shown in Table 4.1-4 is the boiler type, fuel, and
available control device design or operating parameters.  Data were
available for industrial spreader stoker and residual oil-fired boilers and
for utility pulverized coal-fired units.
     Two ESPs, both operating on low sulfur pulverized coal-fired utility
boilers, showed 97.4 and 98.5 percent removal of fine particulate, respec-
                                                    2  3
tively.  [Specific collection areas were 79 and 96 m /m /s, respectively].
Three tests were performed on a utility size spreader stoker equipped with a
reverse air fabric filter system.  Fine particulate removal efficiency was
99.8 percent and above for all three tests.  These data include the effect
on outlet emissions from the reverse air/mechanical shaker bag cleaning
system.
     Fine particulate collection efficiencies of venturi scrubbers on
utility pulverized coal-fired boilers ranged from 51.8 to 91.8 percent; two
of the three units were operated at 2.3 to 2.5 kPa (9 and 10 in. H20)
pressure drop.  Data from one spreader stoker equipped with a venturi
scrubber, operating at 3 kPa (12 in. H20) pressure drop, showed 35.3 percent
efficiency on fine particulate.  Venturi scrubbers on two boilers firing
residual oil showed fine particulate collection efficiencies of 44.7.and
49.5 percent.
     An EPA-tested dual mechanical collector, operating on a high sulfur
coal-fired spreader stoker, collected virtually no fine particulate,
(<2.5 ricrons) but did collect 23.9 percent of the inhalable particulate
                                    4-58

-------
                         TABLE  4.1-4.
FINE PARTICULATE CONTROL  EFFICIENCY FOR VARIOUS
          PM CONTROL  DEVICES*
 i
en
10
Control
Device
ESP
Fabric Filter
Fabric
Fabric
n
Filter
Filter

it
Vcnturi
Venturi
Venturi
Venturi
Venturi
Venturi
Scrubber
Scrubber
Scrubber
Scrubber
Scrubber
Scrubber
Boiler Typec
Utility - PC
Utility - PC
Utility - PC
Utility - SP
n
i
Utility PC
Utility PC
Utility PC
Spreader Stoker
Industrial
Industrial
Fractional Collection
Efficiency %
Fuer of Particles <2.5
LSC
LSW
ANTH
LSC
H
It
LSE
Coald
Coald
HSE
Residual 011
Residual Oil
97.4
99.4
99.9
99.9
99.9
99.8
91.8
79.3
51.8
35.3
49.5
44.7
Control Device Parameters
SCA = 79 m2/m3/s
( ft2/103 acfm)
Teflon coated glass fabric; reverse-air cleaned
Si 1i cone coated
1
1

AP = kPa (9
AP = kPa (10
AP = kPa (12


glass; A/C =
; A/C -
: A/C =
—
in. H20)
in. H20)
In. H?0)
~
—
1.9:1
2.5:1
2.8:1






            aSource: Sedman, Charles B.  Memo and attachments to Industrial Boiler files.
             Performance of Emission Control Systems on Fine Partlculates.  April 21, 1981.
             (Reference 231).
            ''Fine particulate defined as that particulate matter with an aerodynamic diameter of
             2.5 microns or less.

            CPC = pulverized coal-fired.
             LSC = low sulfur coal; LSW  = low sulfur western coal.
             ANTH = anthracite; LSE = low sulfur eastern coal
             HSE - high sulfur eastern coal.
             SP = spreader stoker

             Coal sulfur not specified.

-------
matter (<15 microns).  As discussed earlier, these tests may not be fully
representative of system performance due to air leakage, therefore, the
tests are not included in Table 4.1-4.
     4.1.6.7  Visible Emissions.  This section presents the available
visible emissions data for ESPs, fabric filters, wet scrubbers, mechanical
collectors and side  stream separators.  Table 4.1-5 lists data obtained
using continuous transmissometers.  Table 4.1-6 lists data obtained using
the EPA Method 9.  Tests in which soot blowing occurred are noted in
Table 4.1-6.
     The opacity of  flue gas exiting the stack of industrial boilers
utilizing fabric filters, ESPs, and wet scrubbers for particulate emission
control was less than 10 percent for all data presented.  Opacity data from
spreader stokers equipped with  side stream separators showed opacities
ranging from 0 to 10 percent.   The opacity of the stack gas from industrial
boilers utilizing mechanical collectors for particulate emission control
ranged from 5 to 35  percent depending to a large degree on the PM emission
level.  The lower opacities were observed from a small underfeed stoker;
underfeed stokers generally have much lower uncontrolled emission rates than
spreader stokers.
     Opacity evaluations in Table 4.1-6 indicate that, when soot blowing is
continuous or scheduled on a frequent and regular basis, soot blowing has
little effect on opacity.  Additional data indicate that, during soot
                                                            55
blowing, opacity was not increased more than 0 to 4 percent.
4.2  POST-COMBUSTION TECHNIQUES FOR S02 CONTROL
     Post-combustion techniques for controlling S02 emissions from
industrial boilers are discussed in this section.  The flue gas desulfuri-
zation (FGD) processes discussed are:
       •  Sodium Scrubbing
       •  Dual Alkali
       •  Lime and Limestone (with and without adipic acid addition)
       •  Dry Scrubbing
Each of these FGD systems is currently being used commercially to remove S0«
from industrial boiler flue gases with the exception of adipic acid enhanced
                                     4-CO

-------
TABLE 4.1-5.  OPACITY TRANSMISSOMETER DATA

Parti cul ate
Boiler Load Mass Loading
Type of Boiler 10J lb/hra Control Equipment ng/J lb/10D Btu
Pulverized Coal
(Plant KK)







Spreader Stoker
(Plant UU)







Spreader Stoker
(Plant VV)










Spreader Stoker
(Plant EE #2)

Spreader Stoker
(Plant EE #4)

168
166
164
215
173
189
167
185
170
94
96
95
94
94
88
95
93
95
70
70
72
71
56
61
60
70
69
49
52
16
50
49
49
77
78
78
Fabric Filter 12.8
8.4
7.8
7.8
6.4
4.3
2.5
3.2
3.2
Mechanical Collector 670
610
600
570
540
500
450
450
420
Mechanical Collector 400
360
360
350
300
260
250
240
220
220
180
160
Mechanical Collector 3.9
and Fabric Filter 6.5
•8.6
Mechanical Collector 3.0
and Fabric Filter 4.3
5.6
0.030
0.020
0.018
0.018
0.015
0.010
0.006
0.007
0.008
1.55
1.42
1.40
1.34
1.26
1.16
1.05
1.05
0.99
0.931
0.839
0.842
0.827
0.690
0.596
0.577
0.553
0.516
0.513
0.426
0.380
0.009
0.015
0.020
0.007
0.010
0.013
Opacity
Percent
0
0
0
0
0
0
0
0
0
35
35
25
30
25
25
25
25
25
10
10
10
10
10
12
11
10
10
10
10
11
<10
<10
<10
<10
<10
<10
                   4-61

-------
                        TABLE 4.1-5.   (CONTINUED)

Boiler Loa
Type of Boiler 1CT Ib/hr
Spreader Stoker
(Plant EE #5)
Vibrating Grate
Stoker (Plant R)












Spreader Stoker
(Plant BBB)






Spreader Stoker
(Plant EEE)
Boiler #1
Spreader Stoker
(Plant EEE)
Boiler #3





145
144
78
78
55
77
58
80
57
79
71
78
59
57
59
58
55
53
50
56
55
54
51
55
37
34
36
40
41
42
42
40
40
41
40
Parti cul ate
d Mass Loading
Control Equipment ng/J lb/10D Btu
Mechanical Collector 7.7
and Fabric Filter 16
Mechanical Collector 320
290
260
250
240
210
210
180
180
170
160
150
140
140
Sidestream- Separator 75
74
74
72
72
71
66
65
Sidestream Separator 53
52
50
Sidestream Separator 71
66
64
62
61
59
54
53
0.018
0.038
0.754
0.667
0.595
0.574
0.557
0.490
0.488
0.424
0.421
0.393
0.372
0.354
0.328
0.319
0.175
0.173
0.171
0.167
0.166
0.164
0.154
0.151
0.123
0.120
0.117
0.165
0.153
0.149
0.144
0.143
0.136
0.126
0.123
Opacity
Percent
<10
<10
35
19
11
23
30
29
12
19
19
32
12
12
12
12
6
6
6
6
6
6
6
6
10
5
5
0
0
0
0 .
0
0
0
0
Steam output from boiler.
                                  4-62

-------
TABLE 4.1-6.  OPACITY-ERA REFERENCE METHOD 9

Boiler Loaj
Type of Boiler 10"3 lb/hrc
Pulverized Coal
(Plant C)

Spreader Stoker
(Plant JJ)
(Pulse Jet Cleaning
Mode)
Spreader Stoker
(Plant JJ)
(Reverse Air
Cleaning Mode)
Spreader Stoker
(Plant J2)


Pulverized Coal
(Plant II)


Residual Oil Fired
(Plant HHH)

Spreader Stoker
(Plant K-Boiler #9)

Underfeed Stoker
(Plant H)

Spreader Stoker
(Plant XX)


250
250
250
80



75



45



52



3744
3789
3735
124
126
124
31
27
28
75
75
75
60
Particulate
i Mass Loading
Control Equipment ng/J lb/10 Btu
Fabric Filter 18
15
14
Fabric Filter 6



Fabric Filter 5
4
4

Fabric Filter 9
9
10
23
Scrubber 67
47
28
21
ESP 44
30
28
ESP 5.6
5.2
4.3
Mechanical Collector 30
30
26
Mechanical Collector 220
170
210
110
0.043
0.034
0.032
0.013



0.011
0.010
0.009

0.020
0.021
0.023
0.054
0.157
0.109
0.066
0.048
0.102
0.070
0.065
0.013
0.012
0.010
0.09
0.07
0.06
0.506
0.392
0.494
0.253
Opacity.
Percent
2.5C
2.5
2.5
0



<1
0
0

or

-------
          TABLE 4.1-6.  OPACITY-ERA REFERENCE METHOD 9  (CONTINUED)
Type of Boiler
Boiler Load
 10J lb/hra
                     Particulate
                     Mass Loading     Opacity.
Control  Equipment   ng/J lb/10  Btu   Percent
Spreader Stoker
 (Plant FFF)
     90
Sidestream Separator   70  0.156
aSteam output from boiler.
 Average of six-minute readings.
clncluded a soot blow cycle.
 Soot blown continuously.
<1
Spreader Stoker
(Plant ODD)


31
31
31
31
Sidestream Separator 56
55
50
45
0.130
0.128
0.116
0.104
0
0
0
0
                                    4-64

-------
FGD.   Each system relies on either a calcium- or sodium-based sorbent to
react with S02 to form sulfite and sulfate salts, thereby removing S(L from
the flue gas stream.
     The following sections present a description of each system, and a
brief evaluation of their development status, applicability, and design and
operating characteristics.   Continuous monitoring test data for each system
is presented in Section 4.2.5.
4.2.1  Sodium Scrubbing
     Sodium scrubbing processes are capable of achieving high SCL removal
efficiencies over a wide range of inlet SCL concentrations.  However, these
processes consume a premium chemical (NaOH or Na^CO,,) and produce an aqueous
waste for disposal which contains sodium sulfite and sulfate salts.
     4.2.1.1  Process Description
     4.2.1.1.1  System.  Sodium scrubbing processes currently being  used in
industrial boiler FGD applications employ a wet scrubbing solution of sodium
hydroxide (NaOH) or sodium carbonate (Na2C03) to absorb SCL from the flue
gas.   The operation of the scrubber is characterized by a low liquid-to-gas
ratio [1.3 to 3.4«,/m  (10 to 25 gal/1000 ft )] and a sodium alkali sorbent
which has a high reactivity relative to lime or limestone sorbents.
Further, the scrubbing liquid is a solution rather than a slurry because of
the high solubility of sodium salts.  The SO/> absorption reactions which
                               56
take place in the scrubber are:

          2 NaOH + S02   +   Na2S03 + H20                    (4.2.1-1)

          Na2C03 + S02   +   Na2S03 + C02                    (4.2.1-2)

          Na2S03 + S02 + H20   -»  2NaHS03                    (4.2.1-3)

Simultaneously some sodium sulfite reacts with absorbed oxygen from the flue
gas to produce sodium sulfate:
                                     4-65

-------
          Na2S03 + 1/202   ->-   Na2S04                        (4.2.1-4)

The scrubber effluent, therefore, consists of a mixture of sodium salts.
     Solids storage and handling equipment are auxiliaries associated with
sodium scrubbing systems.  Sodium reagent handling requirements include dry
storage, usually in silos.  A conveyor system is generally used to transport
the reactant from the silo to a mixing tank, where the sodium alkali is
dissolved to produce the scrubbing solution.  The solution from the mix tank
is pumped to a larger hold tank where it combines with the scrubber
effluent.  The majority of the hold tank liquor is recycled to the scrubber
with a slip stream going to waste treatment and disposal.  A simplified
process flow diagram is presented in Figure 4.2-1.
     4.2.1.1.2  Development status.  Sodium scrubbing systems are
commercialized technology; operating systems are in use on industrial
boilers ranging in size from 10 to 125 MW (35 to 430 x 10  Btu/hr) thermal
input.  Table 4.2-1 presents a summary of operating sodium scrubbing systems
applied to U. S. industrial boilers.  Currently 102 sodium FGD systems are
in operation on domestic industrial boilers, and 23 are in the planning or
construction stage.
     4.2.1.1.3  Applicability to industrial boilers.  Sodium scrubbing,
because it is simple both chemically and mechanically, can be applied to
boilers of varying size and type.  The process has been applied to oil-
fired boilers as well as stoker and pulverized coal-fired boilers.
     Future applications of sodium scrubbing systems may be limited by the
need to dispose of the sodium sulfite/sulfate waste liquor.  As shown in
Table 4.2-1 the majority of sodium scrubbing systems in use today are
located in the California oil fields where the wastes are disposed of in
evaporation ponds or by deep-well injection.  Systems in use at industrial
plant locations either reuse the waste liquor in various plant processes or
dispose of it in city sewers or by mixing it with fly ash and either ponding
or landfill ing the waste liquor/fly ash mix.  If wastes from future sodium
scrubbing systems cannot be disposed of by treating them in existing waste
water or ash disposal facilities, or by use as a plant process make-up
                                     4-66

-------
                        ABSORBER
           FLUE GAS
FRESH SORBENT
                MIX
                TANK
   WATER

                                   HOLD
                                   TANK
                                                                            STACK
                                                         MAKEUP WATER
                                                                     WASTE TO
                                                                    TREATMENT
     Figure 4.2-1.   Simplified flow  diagram of a sodium scrubbing system.

-------
TABLE 4.2-1.   PERFORMANCE DATA  FOR OPERATING  SODIUM  SCRUBBING  SYSTEMS.5?
No. of
l,f utl Start-up FCD
Installation/location
Alytska Pipeline
Valdez. Alaska
American Thread
Martin, NC
belridge Oil
McKlttrlck, CA
Canton Textiles
Canton, CA
Chevron
Eakersfleld, CA
FMC
Creen River, WY
General Motors
Dayton, OH
General Motors
Pontiac. Ml
General Motors
St. Louis, MO
General Motors
Tonawanda, NY
Georgia Pacific
Orosett, AK
Getty Oil
Sakeisfield. CA
Great Southern
Cedar Springs, CA
ITT Rayonler
Fernandina. FL
Kerr-McCee
Trona. CA
Mead Paperboard
Stevenson, AL
Mobil Oil
San Ardo, CA
Nekoosa Papers
Ashdowe, AK
Northern Ohio Sugar
Freenont, OH
St. Regis Paper
Cantonment, FL
Texaco
San Ardo. CA
Texasgulf
Granger. VY
(1) C-coal
O-oll
B-bark
PC'petroleum coke
Sorbent
KaOH

Caustic waste

*NaOH

Caustic waste

Hi. CO,

Ka,CO,

KaOH

NaOH

KaOH

NaOH

Caustic waste

NaiCO]

Caustic waste

Cau*Cle viaate

NajCOj

KajCO,

Na, CO, /NaOH

Caustic wast*

NaOH

KaOH

KaOH

Ka,CO,





Type"*
0

C

0

c

0

c

c

c

c

c

B.C.O

0

B.C.O

B.O

0

0

0

C

C

B.O

0

C





IS Date Units
<0.1 6/77

1-1.5 1973

1.1 6/78

0.8 6/74

1.1 7/78

1 5/76

0.7-2.0 9/74

O.B 4/76

3.2 1972

1.2 6/75

1.5-2 7/75

1.1 6/77-12/78

1-2 1975

2-2.5 1975

0.5-5 6/78

1.5-3 1975

2-2.5 1974

1-1.5 2/76

1 10/75

<1 1973

1.7 11/73

0.7 9/76





1

2

2

1

3

2

2

2

2

4

1

6

2

2

2

1

28

2
_
2

1

32

2





so<»
Inlet (ppm)
150

500

500

500

700

800

1.43*/104BTU

-

2000

1*/10* BTU

500

600

1000

1200

-

1500

HOC

600

-

-

1000

860





Percent
Removal
96

70

90

70

90

95

86

—

90

90

80

90-96

85-90

80-85

98

95

90

90

-

80-90

73

90





(2)
Uaste Disposal
oxidation/dilution

pond

waste >rater treatment

pond/waste trealnent

pond/waste treatment

pond

clarify/adjust pH/
to strwer
combine with ash/
landfill
oxidise/neutralize/
discharge
coablne with ash/
landfill
to city sewers

pond

ash pond

to paper process

pond

to paper process

pond

watte treatment

pond

clarification/
aeration
pond/we 11s/ so f lening
and retuse
pond





(2) SO, Inlet (ppm) and percent SO, removal are as reported to PEDCo
bv2the FGD system operator. Values reported nay represent anything
-' • - - -i — «--»i.. j-.~— -I-.H n,,*iK»r< tn rnnHnuous
 frotr. single point wet cneroicaiiy oei-crnimcu ,,v,,^,f  „ .	
 monitoring results and may or may not be obtained by approved
 EPA nvethoOs.
                                      4-68

-------
stream, costs associated with achieving a zero discharge waste will  more
                                           58
than likely limit the system's application.     Treatment and disposal  of
sodium scrubbing system wastes is further discussed in Chapter 7.
     4.2.1.1.4  Avai1abi1ity/reliabi1ity.  The three indices used  in the
EPA Industrial Boiler FGD Survey to reflect this  aspect of system  perform-
ance are availability, operability, and reliability.  These indices  are
defined as follows:
     Availability -  Hours the FGD system was available (whether
                    operated or not)  divided by the hours in the
                    period,  expressed  as a percentage.
     Operability  -  Hours the FGD system was operated divided by
                    boiler operating  hours in the period, expressed  as a
                    percentage.
     Reliability  -  Hours the FGD system was operated divided by  the
                    hours the FGD system was called upon to operate,
                    expressed as a percentage.
Overall reliability  of sodium scrubbing systems applied to industrial
boilers has generally been quite high.   Data reported in the EPA  Industrial
Boiler FGD Survey indicate that of the 22 industrial boiler installations
which have operating sodium scrubbing  systems, 15 reported quantitative
reliability or operability indices that ranged from 89 to 100 percent with
an average of 97.8 percent.   Of the 15 responses, 9 reported a 100 percent
reliability/operability and all  but two reported  reliabilities of greater
                59
than 95 percent.
     Of the seven installations that did not report quantitative  reliability
indices, two reported that the FGD system had no  problems, two reported
erosion/corrosion problems,  one had down-time due to reconstruction, one had
mechanical problems  with pump packings, and one system had no reported
comments.
     4.2.1.2  Factors Affecting Performance.  For a given set of  boiler
operating conditions, the S02 removal  performance of a sodium scrubber
depends on two main  factors:  the amount of scrubbing liquid circulated
                                    4-69

-------
through the scrubber (this is termed the liquid-to-gas ratio or L/G)  and the
sorbent feed rate.  Although design L/G ratios are dependent on the type of
gas-liquid contactor used by the process vendor, sodium scrubbing systems
have relatively low L/G ratios due to the high reactivity of the sodium
                                                                           o
alkali.  Sodium scrubbing L/G's are generally in the range of 1.3 to 3.4t/m
(10 to 25 gal/1000 ft3) whereas typical L/G's for lime and limestone
scrubbers are in  the range of 5 to 158,/m3 (35 to 100 gal/1000 ft3).61
     The amount of fresh sorbent added to the system should be sufficient to
replace the spent sorbent discharged with the process waste-water stream.
If insufficient sorbent is added, the SO^ removal performance of the
scrubber will decrease.  If more than the required amount of sorbent is
added, its concentration will build-up in the system and may eventually
result in chemical scale.  In addition, adding too much fresh sorbent will
increase process  operating costs.  A pH controller is used to monitor the
sorbent feed rate.  A  pH measurement below a specified set point will result
in an  increase in the  sorbent rate whereas a high pH measurement will
decrease the sorbent feed rate.
4.2.2  Double Alkali
     The double or dual alkali process uses a clear sodium alkali solution
for S02 removal and produces a calcium sulfite and sulfate sludge for
disposal.  Although double alkali processes produce a throwaway byproduct, a
regeneration step is employed to regenerate the active alkali for S0£
sorption.
     4.2.2.1  Process  Description
     4.2.2.1.1  System.  The double alkali processes developed in the U.S.
use lime as the calcium alkali, but other processes developed in Japan and
                                                          fi?
still  in the development stage in the U. S. use limestone.    A simplified
flow diagram of a typical double alkali system is given in Figure 4.2-2.
The process can be divided into three principal areas:  absorption,
regeneration, and solids separation.  The principal chemical reactions for a
sodium/lime double alkali system are illustrated by the following
          62
equations:
                                     4-70

-------
                   SCRUBBED GAS
                                                      GAS TO STACK
         SCRUBBER
FLUE GAS
                                                                        WASTE
                                                                       CALCIUM
                                                                        SALTS
 Figure  4.2-2.   Simplified flow diagram for a sodium/lime double-alkali  process.63

-------
     Absorption

     2 NaOH + S02   •*•   Na2$03 + H20                             (4.2.2-1)

     Na2C03 + S02   -*   Na2S03 + C02                             (4.2.2-2)

     Na2S03 + 1/2 02   +   Na2S04                                (4.2.2-3)

     Regeneration

     Ca(OH)2 + 2NaHS03   ->•   Na2$03 + CaS03 1/2 H20 + 3/2 H20    (4.2.2-4)

     Ca(OH)2 + Na2S03 + 1/2 H20  •>   2NaOH + CaS03 1/2 H20(s)    (4.2.2-5)

     Ca(OH)2 + Na2S04 + 2H20  •»•   2NaOH + CaS04 2H20(s)          (4.2.2-6)

     In the scrubber, S02 is removed from the flue gas by reaction with NaOH
and Na2C03, according to Equations 4.2.2-1 and 4.2.2-2.  Because oxygen is
present in the flue gas, oxidation also occurs in the system, according to
Equation 4.2.2-3.  Most of the scrubber effluent is recycled back to the
scrubber, but a slipstream is withdrawn and reacted with slaked lime in the
regeneration reactor according to reactions 4.2.2-4, 4.2.2-5, and 4.2.2-6.
The presence of sulfate in the system is undesirable in that it converts
active sodium to an inactive form, thus lowering S02 removal or increasing
sodium consumption for a fixed SCL removal.
     The regeneration reactor effluent, which contains calcium sulfite and
sulfate is sent to a thickener where the solids are concentrated.  The
thickener overflow is returned to the system, and the underflow containing
the calcium solids is further concentrated in a vacuum filter (or other
device) to about 50 percent or greater solids content.  The solids are
washed, to reduce the soluble sodium salts in the adherent liquor prior to
                                                         64
disposal, and the wash water is returned to the scrubber.
                                    4-72

-------
     4.2.2.1.2  Development status.   Several  process vendors currently offer
double alkali  systems commercially in the United States.   Double alkali
systems are currently operating or planned for use at ten  industrial  boiler
sites, with the smallest application treating 230 Nm /min  (8100 scfm)  and
                            o                           gc
the largest treating 8640 Mm /min  (305,000 scfm) of gas.     Table 4.2-2
presents a summary of double alkali  scrubbing systems applied to U.  S.
industrial boilers.
     4.2.2.1.3  Applicability to industrial  boilers.  A potential limitation
of this technology,  although not as  severe as with the once through  sodium
systems, is the need to dispose of the solid  waste byproduct.  The waste
consists of calcium sulfite and sulfate salts and generally contains from 30
to 50 weight percent water.  Because of the high concentration of soluble
species in the scrubbing solution, the wastes will also contain soluble
salts (such as Na^SO^, Na^SO,, and NaCl) as well as the relatively insoluble
calcium salts.  However, the soluble salts content of the  waste can  be
reduced to less than 1 weight percent when the waste is washed to recover
the sodium.
     4.2.2.1.4  Reliabi1ity/operabi1ity.  Since there are  few double alkali
systems with long-term operating histories in the U. S.,  it is difficult to
assess the overall reliability of this technology.  A limited amount of  data
has, however,  been reported in the EPA Industrial Boiler FGD Survey  for
seven different industrial  boiler sites, and that data indicates that
reported double alkali system reliability averages slightly higher than
90 percent.67
     4.2.2.2  Factors Affecting Performance.   Fuel characteristics such  as
the sulfur and chlorine content can  have major impacts on  the design and
operation of a double alkali system.  Major operating variables include  the
L/G ratio and  alkali addition rate.
     Combustion of low sulfur coal results in a higher ratio of oxygen to
sulfur dioxide in the flue gas than  does combustion of high sulfur coal.
The high relative oxygen content promotes the oxidation of a high percentage
of sodium sulfite to sodium sulfate.  Since sodium sulfate does not react
with hydrated  lime in the presence of sodium sulfite, some active sodium is
                                    4-73

-------
    TABLE 4.2-2.  SUMMARY OF OPERATING AND PLANNED  INDUSTRIAL BOILER  DOUBLE  ALKALI SYSTEMS65
Installation/Location
ARCO Polymers
Monaca, PA
Caterpillar Tractor Co.
East Peonia, ILL
Caterpillar Tractor Co.
Juliet, ILL
Caterpillar Tractor Co.
Mapleton, ILL
Caterpillar Tractor Co.
Morton, ILL
Caterpillar Tractor Co.
Mossville, ILL
Firestone Tire and
Rubber
Pottstown, NY
General Motors, Corp.
Parma, OH
Grissom Air Force Base
Runlffiy* Mill IN
uuiiKtri niii) in
Santa Fe Energy Corp.
Bakersfield, CA
Vendor or
Developer

FMC

FMC

ZURH

FMC

ZURII

ZURH


FMC

G.M.
Neptune/
Airpol


FMC
Size
(SCFM)

305,000

210,000

67 ,000

236,000

38,000

•140,000


8070

128,400

"v> nnn
O£ , UUU

70,000
No. of
FC.I) Units

3

4

2

5

2

4


1

1
•
i
i

1
T.yi»

C

C

C

C

C

C


C

C




0
Fuel
e %S 1

3

3.2

3.2

3.2

3.2

3.2


2.5-3.0

2.5

i n i 
-------
lost in the regeneration step.   This loss has the same effect as  reducing
the sodium alkali  feedrate.   Oxidation can be minimized in  low sulfur fuel
(<1 percent S)  applications  by  using a dilute absorbing solution  (active
sodium concentration less than  0.15 Molar).   At the  resulting low sulfite
concentrations, the sulfate  will  react with  calcium  to regenerate the
scrubbing liquor.   For higher sulfur applications,  (>1  percent S) oxidation
can be minimized by using a  concentrated absorbing  solution (active  sodium
concentration greater than 0.15 Molar) and sulfate can be copreci pita ted
                     CO
with calcium sulfite.
     Chlorides  absorbed from the flue gas are difficult to  remove and can
cause problems  if they build up in  the system.   The  only mechanism for
chlorides to leave the system is in the liquor contained with the solid
waste.  However, chlorides are  recovered and recycled  to the absorber when
the waste is washed to recover  sodium.  In addition  to decreasing the
concentration of active alkali  in the absorber, high levels of chlorides can
result in stress corrosion.   A  solution proposed by  one vendor is to use a
prescrubber to  remove chlorides before the double alkali system.     The  use
of a prescrubber with a separate liquor loop, however, could cause water
balance problems in the system.  Since all the evaporation  loss would occur
in the prescrubber, the only water  loss from the double alkali system would
be the water occluded with the  solid waste.   This small water loss would not
allow enough water addition  for the normal cake washing (more than one
displacement wash), demister washing, pump seals, and  lime  slaking.
     The effects of variable L/G, pH, and pressure drop on  double alkali
process operation are shown  in  Figures 4.2-3 and 4.2-4 respectively.
Figure 4.2-3 illustrates the increase in SO, removal performance  due to
                                                                          o
increased L/G.   Typical double  alkali L/G's  range from about 1.3  to
(10 to 25 gal/1000 ft3).   The effects of pH are shown in Figure 4.2-4.   The
operating pH of the system can be adjusted by changing the sorbent feed rate
and/or adjusting the pH of the regenerated liquor.   In general, as shown by
Figure 4.2-4, S02 removals decrease rapidly below pH 6.   High pH levels
(pH 9 or above) will result in calcium carbonate formation which can result
                                    4-75

-------
               100
                95
                90
              Q
              UJ
                85
              O
              CO
                80
                75
                 f
                                        2/ACTUAL m3
                                  234
                                              FUEL = COAL
                                              pH = 5.8-7.1
                                              SINGLE STAGE ABSORBER
                            10
                        20        30
                       L/G-Gal/IOOOacf
40
Figure 4.2-3.
S0? removal versus L/G ratio for the  Envirotech/Gadsby Pilot Plant
with a single stage polysphere absorber.71
                                         4-76

-------
Figure 4.2-4.
                 100
                  90
                Q
                HI

                080
                5
                LU
                DC
                 «
                o
                CO
                  70
                  60
                                                  o o
                             = 2.52/m3
                            = 4inH2O
                         TWO-STAGE ABSORBER
                                       1
             3456
               SCRUBBER EFFLUENT pH
                                                           8
SCL removal  versus  scrubber effluent pH for the Envirotech/Gadsby
Pilot Plant  with a  two-stage absorber. 72
                                       4-77

-------
in scale formation.  Consequently, the operating pH of double alkali  systems
                                     CO
is generally in a range of pH 6 to 8.
4.2.3  Lime and Limestone
     The lime and limestone FGD processes use a slurry of calcium oxide or
calcium carbonate to absorb SCL in a wet scrubber.  A calcium
sulfite/sulfate sludge is produced for disposal.
     4.2.3.1  Process Description
     4.2.3.1.1  System.  The absorption of SCL from flue gases by a lime or
limestone  slurry involves both gas-liquid, and liquid-solid mass transfer.
The  chemistry is complex, involving many side reactions.  The overall
reactions  are those of SCL with lime (CaO) or limestone (CaCOg) to form
calcium sulfite  (CaS03 1/2 H,,0) with some oxidation of the sulfite to form
calcium sulfate  (CaSO^ 2H20).  These reactions can be represented as
follows:
      Lime

      S02 +  CaO  +  1/2  H20   ->  CaS03  1/2  H20                       (4.2.3-1)

      S02 +  1/2  02 + CaO + 2H20   -»•  CaS04  2H20                    (4.2.3-2)

      Limestone

      S02 +  CaC03  + 1/2 H20  •*  CaS03  1/2  H20 +  C02               (4.2.3-3)

      S02 +  1/2  02 + CaC03 + 2 H20  +   CaS04 2H20  +  C02           (4.2.3-4)

 The  calcium sulfite and sulfate crystals  precipitate  in  a  reaction vessel or
 hold tank which is designed to  provide adequate residence  time for solids
 precipitation as  well as  for dissolution  of the alkaline additive.  The hold
 tank effluent is  recycled to the scrubber to absorb additional S02.  A slip
 stream from the hold tank is sent to  a solid-liquid separator to remove the
 precipitated solids from  the system.   The waste solids,  which may vary from
                                    4-78

-------
35-70 weight percent solids,  are generally disposed of by ponding or
landfill.   A simplified  flow  diagram is  presented in Figure  4.2-5.
     Auxiliary equipment associated with this  process includes  a  reagent
preparation system.   Reagent  preparation may consist of limestone grinding
and/or lime production.   However, for most industrial  boilers,  due to  their
small size, preground lime  and  limestone may be  purchased and the feed
preparation system will  then  consist of  storage  silos and either  lime
slaking or limestone slurrying  equipment.
     Addition of adipic  acid  to the FGD  slurry can  enhance SC^  removal  and
improve the reliability  and economics of lime  and limestone  FGD systems.
Adipic acid addition provides a buffering action which limits the drop in  pH
that normally occurs at  the gas/liquid interface during SCL  absorption.
This stabilized pH results  in an increased mass  transfer rate of  SCL into
the liquid phase.  In addition, the capacity of  the scrubbing liquor
available  for reaction with S09 is increased by  the formation of  calcium
                    74
adipate in solution.    Adipic  acid addition also increases  lime  or
limestone  utilization.  As  a  result, limestone grinding requirements and
solid waste generation are  somewhat lower than those for a conventional
                     75
limestone  FGD system.
     4.2.3.1.2  Development status.   Both lime and  limestone FGD  technology
is demonstrated and  commercially available.  Lime FGD technology  was first
used to control  S09  emissions on commercial  boiler  pilot plants in England
                  76
about 40 years ago.    As shown by Table 4.2-3,  there are currently two
operating  systems  on industrial boilers  in the U.  S.; one lime  system
                3
treating 2380 Nm /min (84,000 scfm)  of gas,  and  one limestone system
                3                           78
treating 1560 Mm /min (55,000 sfcm)  of gas.
     In addition to  industrial  boiler use, some  34,000 MW of coal-fired
electrical generating capacity  in the United States has been committed to
lime or limestone  scrubbing.  This figure includes  28 facilities  in
operation, 35 under  construction, and another  16 in the planning  stages
(i.e., contract awarded, letter of intent signed, or requesting/evaluating
bids).76
                                   4-79

-------
cs
o
                                               S02ABSORBER
                                     FLUE GAS
                                                               TO STACK
                                                             MAKE-UP WATER
             LIME	

             OR

   	J*
                            LIME
                           SLAKER
            LIME
           STONE
CRUSHING
  AND
GRINDING
SLURRY
                                ^e
EFFLUENT HOLD TANK
                                                                          SECOND STAGE
                                                                          SOLID-LIQUID
                                                                           SEPARATOR
                                                                              OR
                                                                         SETTLING POND
                                                                                                      SOLID-LIQUID
                                                                                                       SEPARATOR
                                                                                                1
                                                                                            SOLID WASTE
               Figure 4.2-5.  Process flow diagram for a  typical  lime or limestone  wet  scrubbing system. 73

-------
                            TABLE 4.2-3.   SUMMARY  OF  OPERATING LIME AND LIMESTONE SYSTEMS
                                    FOR U.S.  INDUSTRIAL  BOILERS AS OF MARCH 1978 77
I
CO

M~.., ~~ c4-~ rUel
Process
Lime
Lime and
Limestone
Vendor
Koch Engineering
Research
Cottrell-Bahco
new \ji -J i i.c
Company/Location retrofit scfm Type
Armco Steel R 84,000 Coal
Middle town, OH
Rickenbacker Air R 55,000 Coal
Force Base
Columbus, OH
Sulfur (%)
0.8
3.6
        190 percent design S02 removal  with  lime,  lower with limestone.

-------
     Emission test results from an EPA test facility at the Shawnee Power
Station in Paducah, Kentucky have demonstrated an average S0? removal  of
97 percent for an industrial boiler-size, adipic acid enhanced, venturi/FGD
system.  A 30 day test at the Rickenbacher AFB in Columbus, Ohio
demonstrated an average S02 removal efficiency of 94 percent for an adipic
acid enhanced limestone FGD system.  This test is discussed in more detail
in Section 4.2.5 and complete test data are presented in Appendix C.  A
demonstration of this technology on a full scale utility boiler is currently
underway at Springfield City Utilities' Southwest Power Plant.
     4.2.3.1.3  Applicability to industrial boilers.  Both lime and
limestone processes are applicable to industrial boilers.  The processes use
readily available moderate priced sorbents.  As with the double alkali
process, a potential limitation of the lime and limestone processes is the
requirement for disposal of the waste sludge byproduct.  While the problem
associated with the presence of highly soluble salts in the waste is much
less severe than for the double alkali or once through sodium processes, the
increased land requirements associated with scrubber -sludge disposal could
be limiting for some applications and must be evaluated on a site specific
basis.
     The presence of adipic acid on the EPA's hazardous materials list
should not exclude its use as an FGD additive.  Bioassay tests run on sludge
samples from the Shawnee facility show no significant difference in toxicity
between adipic acid enhanced system sludge and sludge samples from systems
without adipic acid.  Additional studies on leachate toxicity have indicated
that sludge generated from systems using adipic acid show toxicity to be
                       79
well within EPA limits.
     4.2.3.1.4  Rel1abi1ity/operabi1ity.  Reliability of lime and limestone
FGD systems for industrial boiler applications is difficult to assess since
there are only two installed systems and only one of those, the Bahco system
located at Rickenbacker Air Force Base (RAFB), has been operational over a
long period of time.  Scrubber performance at the RAFB facility has
generally been quite good except for the early stages of operation in which
several startup problems resulted in significant amounts of downtime.   From
                                   4-82

-------
November 1976 through December 1978,  the  RAFB system demonstrated that an
industrial  boiler FGD system can  operate  with high  reliability as it
operated 95 percent or more of the time during that period except for the
months of January,  February and March 1978.   During those  three months,
system downtime was caused by a severe blizzard which resulted in the
                           80
freeze-up of several  lines.    This problem  can be  mitigated  or avoided by
insulating exposed  lines and by keeping the  slurry  circulating through the
lines whenever possible during periods of downtime  in severely cold  weather.
     4.2.3.2  Factors Affecting Performance.   The removal  of  S02 from
industrial  boiler flue gas in a lime or limestone FGD system  involves a
gas-liquid-solid mass transfer process and thus is  more complex than the
once through sodium or double alkali  FGD  systems which involve only  gas-
liquid mass transfer in the scrubbing step.   As a rule, a  large portion of
the alkalinity required for SO^ removal in lime and limestone systems is
derived from solids dissolution in the scrubber. Since solid-liquid
reactions tend to be significantly slower than do liquid-liquid reactions,
it is advantageous  to minimize the amount of solids dissolution required by
maximizing the amount of liquid phase alkalinity in the scrubber feed
liquor.  For this reason systems  which operate with high magnesium and
sodium concentrations but low chloride levels exhibit higher  S09 removals
                                                   81
than systems which  are lower in soluble alkalinity.
     Gas maldistribution can be a major problem in  lime and limestone FGD
systems, particularly in large units. Unlike once  through sodium and double
alkali systems, lime and limestone FGD systems normally utilize "open"
contactors such as  spray chambers.  While this practice helps to minimize
potential scaling and plugging problems often associated with lime and
limestone systems,  it is susceptible to gas  distribution problems.  Portions
of the scrubber can become liquid phase alkalinity- limited due to gas
maldistribution even though the total alkalinity entering  the scrubber is
sufficient for good S02 removal.   Scrubber design should therefore
incorporate straightening vanes and/or open  packing to promote good  gas
             op
distribution.
                                    4-83

-------
     Several design and operating variables should be considered in the
design of a lime or limestone F6D process.  The effects of the following
major variables on SC^ absorption efficiency and/or overall process
operations are briefly discussed:
          L/G Ratio - Higher SCL removal efficiencies are achieved at higher
L/G ratios up to the point where flooding and poor gas distribution
occurs.82  Typical L/G's range from 5-15 i/m3 (35-100 gal/1000 ft3).
          Slurry pH - Higher S0? removal efficiencies are achieved with
higher pH levels.  Since scaling can occur at high pH's (pH greater than 9)
typical control points for a lime system are in the pH 8-9 range.  Because
limestone systems are buffered, they typically operate in the pH 5-6
      83
range.
          Effects of Soluble Species - The concentration of dissolved ions
in the scrubbing slurry directly affects the liquid phase alkalinity and
hence the system's ability to remove sulfur species from flue gas.  For a
given set of operating conditions, high concentrations of Na+ and Mg++ will
improve the S09 removal efficiency and high concentrations of Cl will reduce
n.84
          Ash Removal - Although fly ash can be removed simultaneously with
SO^, the trend has been to remove it upstream for the following reasons:  to
decrease erosion in the scrubber and associated equipment such as pumps,
piping, nozzles, and fans; to provide dry fly ash for sludge fixation; and
to avoid particulate emission excursions during periods of scrubber
inoperation.
          Oxidation - Forced oxidation systems increase the amount of
calcium sulfate (gypsum) in the waste which is produced by sparging air into
the system.  A high sulfate sludge is more easily dewatered and has better
structural  properties than does the more difficult to handle thixiotropic
calcium sulfite sludge.86  Application of forced oxidation to FGD systems
using adipic acid additive may result in degradation of the adipic acid in
the slurry.
                                     4-G4

-------
4.2.4  Dry Scrubbing
     Dry scrubbing processes that appear to be applicable to industrial
boilers include spray drying of a lime or sodium sorbent, and firing of a
pelletized or pulverized coal and limestone mixture.  Each of these
processes produce a dry waste product for disposal.  The use of the
coal/limestone fuel mixture is discussed in Section 4.6.
     4.2.4.1   Process Description
     4.2.4.1.1  System.  In a spray drying process, flue gas is contacted
with a solution or slurry of alkaline material in a vessel of relatively
                                      87
long residence time (5 to 10 seconds).    Generally the particulate matter
(fly ash) has not been removed prior to entering the absorber, and the spray
drying process acts as a combined particulate/SCL removal system.   The flue
gas SOp reacts with the alkali solution or slurry to form liquid phase salts
which are dried to about one percent free moisture by the heat in  the flue
gas.  These solids, along with fly ash are entrained in the flue gas and
carried out of the dryer to a particulate collection device such as an ESP
or baghouse.   Systems using a baghouse for particulate removal report
additional S02 sorption occurring in the baghouse.  A generalized diagram
for a typical spray drying process is shown in Figure 4.2-6.
     Reaction between the alkaline material and flue gas S(L proceeds both
during and following the drying process.  The mechanisms of the SCL removal
reactions are not well-understood.  It has not been determined whether S02
removal occurs predominantly in the liquid phase, by absorption into the
finely atomized droplets being dried, or by reaction between gas phase SCL
and the slightly moist spray-dried solids.  The overall chemical reactions
                                 89
for this process are shown below.

          S02 + Na2 C03  -»•  Na2S03 + C02                         (4.2.4-1)

                        or

          S02 + CaO + 1/2 H20  -»  CaS03 1/2 H20                  (4.2.4-2)

-------
oc
cr>
                                                                                        Clean Gas to

                                                                                        Atmosphere
                                    Hot or Warm Gas Bypass
lHot IWarm
*v ' * '
	 J Flue Gasr / ^
Rnllnr — — _ ^1
. fc uoiier Dr^r^riT/ 1
—^ Preheater J
1
Air
Combustion Air

Sp
So


^4
]
Spr
Dry
V
ent
ids
^
>
ay
er
/^
Clean Gas. / \
O*l \
Stack
Fan
1
I r
T ^
Flue Gas'' Baghouse
& Solids or ESP
X )
4
^
r
A

Partial Recycle of Solids i r
(Lime Reagent) I
Sorbent Product Solids &
Slurry Fly Ash Disposal
Tank
1
                                                    Sorberrt Storage
                   Figure 4.2-6.  Typical  spray dryer/particiilate collection process flow diagram.

-------
In addition to these primary reactions,  sulfate salts will  be produced by
the following reactions:

          Na2S03 + 1/2 02   +  Na2S04                             (4.2.4-3)

          S03 + Na2C03 -»•   Na2S04 + C02                           (4.2.4-4)
                       or

          S02 + CaO + 1/2  02 + 2H20  +  CaS04 2H20               (4.2.4-5)

     Liquid to gas (L/G)  ratios for spray drying are typically 0.03 to
0.04fc/m3 (0.2 to 0.3 gal/1,000ft3).  This low liquid rate is not sufficient
to saturate the gas.  Gas  exit temperatures are typically in the 65-93°C
(150 to 200°F) range which provides a safe margin against water
condensation.
     4.2.4.1.2  Development status.  Spray drying technology for removing
S02 from boiler flue gas  has b.een limited to pilot-scale testing of
industrial  boiler sized systems [280 to  560 m3/min (10,000 to 20,000 acfm)]
at several  utility locations burning low sulfur western coals.  This
technology is being commercially offered by several  vendors, and five spray
drying FGD systems have been sold for industrial boiler applications.  These
systems are being applied  to boilers burning coals with a fairly wide range
of sulfur contents (0.6 to 3.5 percent S).  Table 4.2-4 summarizes the
commercial  spray drying systems sold for application to industrial boilers.
In addition eleven full-scale utility systems have been sold.  The utility
systems are being applied  to low sulfur  (less than 2 percent) coal-fired
units and S02 removal guarantees from the vendors are as high as 90 percent.
However, it still remains  to be shown whether spray dryer systems will be
able to achieve high SO^  removal efficiencies when applied to full scale
industrial  boiler installations firing a range of coal types.
     4.2.4.1.3  Applicability to industrial boilers.  Spray drying
technology is an applicable S02 control  method for all industrial boilers
firing low to medium sulfur fuels (less  than 3 percent sulfur).  However,
                                    4-67

-------
               TABLE 4.2-4.  SUMMARY OF INDUSTRIAL BOILER SPRAY DRYING SYSTEMS 91
                                                                  Fuel             S02 Removal
     Company                                      Size      	       Guarantee
     Location            Vendor      Sorbent  (Ib steam/hr)  Type  % Sulfur           (%)


Strathmore Paper Co.  Mikropol        Lime        85,000     Coal  2 to 2.5    75% on 3% S coal
Woronoco, MA
(operating)

Celanese              Wheelabrator-   Lime       110,000     Coal   1  to 2     85$ on 2% S coal
Cumberland, MD        Frye/
(operating)           Rockwell Int.

University of         Carborundum     Lime     2 units @     Coal   0.6 to            70%
  Minnesota           Environmental            120,000 acfm         0.7
Minneapolis, MN       Systems, Inc.            each

Department of Energy  Niro Atomizer,  Lime       170,000     Coal    3.5              80%   6
Argonne, IL           Inc./Joy-                                              (1.21bS02/10 Dtu)
                      Western
                      Precipitation
                      Division

Container Corp.       Ecolaire, Inc.  Lime       170,000     Coal     1               NA
Pittsburgh, PA


NA = Not available.
aVendor design guarantees under specific operating conditions.

-------
the technical  and economic viability of this process is not clear for
applications requiring high S(L removals for coals containing greater than
three percent sulfur.
     The potential  for condensation in downstream particulate collection
equipment, especially  during system upsets, is also a concern.   Condensation
problems may be avoided by bypassing the fabric filter during system upsets
and by maintaining spray dryer outlet temperatures at an adequate margin
above the adiabatic saturation point.  The effects of condensation on
downstream equipment and system performance using varying quality coals are
questions that will be resolved only after additional operating experience
is obtained in either  utility or industrial boiler applications.
     4.2.4.1.4  Reliabi1ity/operabi1ity.  Since dry scrubbing is  a
relatively recent innovation in industrial boiler FGD, no data is available
on the long-term commercial reliability or operability of these systems.
However, since they are less complex mechanically and no more complex
chemically than wet calcium or sodium-based scrubbing systems, they should
ultimately prove to be at least as reliable and operable.
     4.2.4.2  Factors  Affecting Performance.  The performance of a spray
dryer FGD system depends on several factors, the two most important being
the L/G and the stoichiometric ratio of sorbent to StL.  Unlike a wet
scrubbing system the amount of water that can be added (the L/G)  is set by
heat balance considerations for a given inlet flue gas temperature and
                                                                      3
approach to saturation.  Typical L/G ratios range from 0.03 to 0.04j,/m
                       o
(0.2 to 0.3 gal/1000 ft ).  The sorbent stoichiometry is varied by raising
or lowering the concentration of a solution or slurry containing this set
amount of water.  As sorbent stoichiometry is increased to raise the level
                                                           87
of S0£ removal, there  are two potentially limiting factors:
        • Sorbent utilization may decrease, raising sorbent and disposal
          costs per unit of S0« removed.
        • An upper limit on the solubility of the sorbent in the solution,
          or on the weight percent of sorbent solids in a slurry may be
          reached.
                                    4-89

-------
     Methods of circumventing these limitations include recycling sorbent,
either from solids dropped out in the spray dryer or from the particulate
                 92
collection device   and operating the spray dryer at a lower outlet
                                                         OQ
temperature; that is, at a closer approach to saturation.
     Based upon pilot unit test results, high SCL removals (up to
90 percent) can be achieved for low-sulfur coal applications, using either
lime or sodium-based sorbents.  Stoichiometric ratios of 2.3 to 3.0 were
required for lime operations whereas Stoichiometric ratios of only 1.0 to
1.2 were required to achieve the same S02 removal for sodium operations.  It
has also been reported that 90 percent S0« removal may be achieved with a
Stoichiometric lime requirement of 1.3 to 1.7 by recycling some of the
                  93
unreacted sorbent.    A sodium-based system should be able to achieve higher
S02 removals than lime based systems on high sulfur coals due to the greater
reactivity of sodium hydroxide or sodium carbonate compared to lime.
     Spray dryer design can also be affected by the choice of the particu-
late collection device.  Bag collectors may have an advantage over ESPs in
that unreacted alkalinity in the collected waste on the bag surface can
react with the remaining SO, in the flue gas.  Some process developers have
                                                                94
reported S02 removal on bag surfaces on the order of 10 percent.    A
disadvantage of using a bag collector is that since the fabric is somewhat
sensitive to wetting, a safe margin above the saturation temperature (on the
order of 20 to 35°F) must be maintained for bag protection.  Some vendors
claim that ane ESP is less sensitive to condensation and hence can be
operated closer to saturation (less than a 25°F approach) with associated
increase in spray dryer performance.  However, they feel that S09 removal
                                                                  95
within the collector is not likely to be as high as in a baghouse.
4.2.5  Emission Reduction Data
     This section presents continuous S02 emission monitoring data for five
wet FGD systems and a lime spray drying system.  Emission data for the wet
F6D systems are representative of the S02 removal capability of well
designed, operated and maintained industrial boiler FGD systems.  All
sampling and analyses were conducted in accordance with the procedures
specified in 40 CFR 60 Appendix A.
                                     4-90

-------
     As with the particulate matter emission data,  tests not considered to
be representative of well  operated FGD systems are  not presented in this
chapter, but are included  in Appendix C along with  documentation of the
reasons why they were not  considered to be representative.   Three such tests
of wet FGD systems are discussed in Appendix C.
     4.2.5.1  Emission Reduction Data for Wet FGD Systems.   This section
presents the results of five continuous S0« emission monitoring tests  of
industrial boiler wet FGD  systems.   All of the tests were conducted by EPA.
Data were collected for two dilute double alkali  systems, one sodium
throwaway system, a lime system, and a limestone  system with adipic acid
addition.  Table 4.2-5 summarizes the five test programs and daily average
results are shown in Figures 4.2-7 to 4.2-11.  Hourly results and detailed
descriptions of tests procedures can be found in  the references cited  in the
Appendix C discussions of  each of the test sites.  Figures 4.2-7 to 4.2-11
show the 24-hour average $02 removal, boiler load,  and scrubbing slurry pH.
Only days with 18 hours or more of test data are  presented; missing days
(days where 18 hours of data were not obtained are  indicated by a break in
data shown in Figures 4.2-7 to 4.2-11.
     Table 4.2-5 shows that each system averaged  more than 90 percent  S0«
removal over the test period.  In addition, average outlet S0« concentra-
tions for each test period were 192 ng/J (0.45 lb/10  Btu) or less.
     Thirty days of continuous emissions data were  gathered at the sodium
throwaway scrubbing system at Location I.  Figure 4.2-7 shows consistent
high S02 removal, averaging 96.2 percent for the  test period.  Table 4.2-5
shows that daily average inlet S02 concentrations ranged from 1961.to
2480 ng/J (4.6 to 5.6 lb/106Btu).  The scrubbing  solution pH was
consistently maintained at about pH 8.  As discussed in Section 4.2.1.1,
proper pH control is important maintaining the sorbent feed rate required
for the desired S02 removal.
     Figures 4.2-8 and 4.2-9 show daily average results for two similar
double alkali systems at Location III.  The two systems averaged 91.6  and
92.2 percent S02 removal over the respective 17-  and 24-day test periods.
Daily average inlet S02 concentrations ranged between 1235 and 2000 ng/J
                                    4-91

-------
                                          TABLE 4.2-5.   SUMMARY  OF CONTINUOUS S02 EMISSION  DATA
                                                           AT FIVE  INDUSTRIAL  BOILER WET  FGD SYSTEMS
4i
 I
<£)
Location
I
Ill/No. 1
Ill/No. 3
IV
IV
System No. of
Type Days of Data
Sodium Throwaway
Double Alkali
Double Alkali
L1me
Limestone with
Adlplc Add Addition
30
17
24
29
30

Inlet S02 (ng/J)c
Range Average
. 1961-2480 2348
1235-2000 1646
1180-2285 1606
1927-2432 2250
1333-2765 2125
24-hr Average Results
' Outlet S02 (ng/J)c % S02 Removal
Range Average Range Average Comments
54-267 87 88-98 96 Tray & quench liquid scrubber;
coal sulfur = 3.6J
81-213 138 88-95 92 Two Tray scrubber; Design
pH o 5.5 to 7,5; Design
L/6 • 2.7 i/m;
37-446 128 74-97 92 Same design as Location HI/11.
94-294 192 88-96 91 Two "Inverted venturl" stages;
Coal Sulfur - 3.5X.
56-262 122 90-97 94 Coal sulfur 2.2 to 3.52;
Adlplc Acid concentrations of
1770 to 3000 ppm.
               aHore complete descriptions, data testings, and references for test reports can be found 1n Appendix C.
                Only days with 18-hrs or more of test data are reported.
               C01v1de by 430 to convert to lb/106 Btu.
                Arithmetic mean of 24-hr averages for test period.

-------
   lOOr
S   90
 CM

O
00
    80
                                                Average  S02 Removal = 96.2%
    70
                           10
15
20
25
30
    90




    80




    70
•o

-------
  lOOr
•5  90
o>
   80
                          Average S02 Removal = 91.6%
                          10
   15
20
25
30
    90



    80



    70
•o


2  60

s_
0)

£  50
o
CO


**  40



    30
                          10
   15
20
25
30
                          10
   15

Test  Days
20
25
30
     Figure 4.2-8.   Daily average  S0« removal, boiler  load, and.
                     slurry pH for  the dual alkali scrubbing
                     process at Boiler No.  1, Location  III.
                                     4-94

-------
100
 90
 80
                                         Average S02 Removal = 92.2%
                    10
                                15
                  20
25
30
   80


   70
•o

-------
  TOO
>

o
o
   90
                                               Average S09 Removal = 91.5?
10
15
                     20
                                                     25
                                                                    30
   90-


   80-


   70-
-o
>
7 •
6-
c

'
10         15         20
        Test Days
                                                     25
                                                                    30
    Figure 4.2-10.
                         Daily average S0_ removal, boiler  load, and
                         slurry pH for line slurry scrubbing  process
                         at Location  IV.
                                   4-%

-------
  o
  01



  ^
       100r
        90'
                                           Average S0_ Renewal - 94.3?
                                                    20
                         25
                      30
      3000
^ §.  250C
£~


22
a. s-
••- 3
•o •—
«Ci«
       7.0




       6.G




       5.( '



       4.0'




       3.0
                              10
   15

Test Days
20
25
30
         Figure 4.2-11.   Daily average S02 removal,  boiler load,  adipic

                          acid concentration, and  slurry pH for linestone

                          system at  Location IV.
                                    4-97

-------
(2.9 and 4.7 lb/106 Btu) at Boiler No. 1 and between 1180 and 2285 ng/J
(2.8 and 5.3 lb/106 Btu) at Boiler No. 3.  The scrubbing slurry pH for both
systems was maintained close to pH 6 during the test periods.  The desired
operating pH of most double alkali systems is pH 6 to 8 (Section 4.2.2).
The design pH for the systems at Location III is pH 5.5 to 7.5 and the
design L/G ratio is 2.7ii/m3 (20 gal/103 ft3).
     The lowest S02 removals observed at Location III, Boiler No. 3 (Test
days 9 and 10 in Figure 4.2-9) were during FGD system start-up after the
scrubber had been taken off-line due to low boiler load requirements at  the
plant.
     Figure 4.2-10 shows the daily average results of tests of a lime
scrubbing system at Location IV.  Average S02 removal for the period was
91.5 percent and daily average inlet S09 concentrations ranged between 1927
                                c      f-
and 2432 ng/J (4.5 and 5.7 lb/10  Btu).  The lowest S02 removals were
observed during the last few days of testing when the scrubbing slurry pH
dropped below pH 6.  As discussed in Section 4.2.3, typical control points
for lime systems are more often in the pH 8 to 9 range.  Figure 4.2-10 shows
generally higher S0? removals for the periods during which slurry pH was
maintained near pH 8.
     Figure 4.2-11 presents the results of 30-days of testing at Location IV
during which limestone reagent was used (instead of lime) and adipic acid
was added to the scrubbing solution.  These data show an average S02 removal
of 94.3 percent for the test period.  High S0« removals were obtained over a
wide range of boiler loads.  Adipic acid concentrations in the slurry ranged
from 1770 to 3000 ppm and slurry pH was maintained near pH 5.  Inlet S02
concentrations ranged from 1333 to 2765 ng/J (3.1 to 6.4 lb/106 Btu).
     The data in Figure 4.2-11 indicate that adipic acid addition contri-
butes to high S02 removals and, with proper pH and adipic acid addition
control, low variability in system performance.  Previous testing of the  FGD
system at Location IV with limestone slurry had shown S02 removals between
50 and 70 percent.  It should be noted that adipic acid addition may not
have been solely responsible for the improved S02 removal efficiency since
                                    :-98.

-------
the limestone only tests appeared to have been conducted at conditions
outside the design range of the system (See Appendix C).
     4.2.5.2  Emission Reduction Data for Lime Spray Drying System.
Figure 4.2.12 illustrates the daily average results for S02 emission
monitoring of the lime spray drying system at Location VI.   Removal
efficiencies ranged from 46 to 80 percent.   Inlet S02 removal  efficiency
averaged 68.4 percent over the test period.   S(L  concentrations averaged
1492 ng/J (3.5 lb/106Btu) and ranged from 1118 to 1905 ng/J (2.6 to
4.4 lb/106 Btu).   Outlet concentrations  had a range of 339  to  702 ng/J (0.8
to 1.6 lb/106Btu) while averaging 460 ng/J (1.1 lb/106Btu).  Figure  4.2-12
shows S09 removal efficiencies averaging 75 percent on the  days when average
        f-                                         c
daily S02 concentrations were 1720 ng/J  (4.0 lb/10  Btu) or greater.  The
somewhat variable performance of the spray dryer can be attributed in part
to various system upsets that occurred throughout the testing  period.  These
upsets include slurry pump problems, spray dryer plugging and  boiler load
fluctuations.  Over the last six days of the testing program,  a period in
which no upsets occurred, the average daily S09 removal remained near
           232
80 percent.
     The average sulfur content of the coal  fired during the test was near
2 percent, which is the coal sulfur content the system was  designed  for.  No
data were available for spray drying systems applied to high sulfur  coal-
fired boilers.
4.3  COMBUSTION MODIFICATION TECHNIQUES  FOR NITROGEN OXIDE  (NOV) CONTROL
                                                              /\
     NO  emissions from industrial boilers are generally classified  as one
       A
of two types:
       •  thermal NO  (formed by the reaction of atmospheric nitrogen and
                    A
          oxygen in the combustion zone)
       •  fuel NO  (formed by the reaction of fuel nitrogen and oxygen in
                 A
          the combustion zone).
N0x includes both NO and N02.  The latter species is typically about
5 percent of the total NO  emissions, although some data indicates that the
                         A
N02 fraction may be somewhat lower for coal- and oil-fired units than for
gas-fired units.97
                                    4-99

-------
O
CU
 CM

O
oo
     80
      70
 60
      50
      40
                             10
                                    15
25
TO
 O>
 c
 CM

O
oo
 01
200C




1800




160C




140C




120C



100C




 80(T
                              10         15          20


                                     Test Days
                                                          25
                                                                            5.0
             4.0   5-

                  in
                  o
                 r\j



             3-0
                                                                                 o
                                                                                  en

                                                                            2.0   5
                                                                            1.0
           30
                 Figure 4.2-12.   Daily average S02 removal, inlet S02  for

                                  lime spray system at  Location VI.
                                   4-100

-------
     The formation of thermal  NO  increases with increases in excess air,
                                A
flame temperature, and residence time in the high temperature zone of the
boiler.     Fuel NO  formation  occurs  at a much lower flame temperature than
                  A
those required to form thermal  NO  and thus emissions do not generally vary
                                 A
with flame temperature.
     The rate of formation of  both thermal  and fuel  NO  is dominated by
                                                      A
combustion conditions and thus is amenable to suppression through modifica-
tion of  the combustion process.   The  following combustion modifications have
been investigated as NO  control measures for industrial boilers:
                       A
      •  Low excess air (LEA)
      •  Staged combustion (SC)
      •  Flue gas recirculation (FGR)
      •  Low NO  burners (LNB)
                A
      •  No combustion  air preheat or reduced air preheat (RAP)
      •  Ammonia injection
     The mechanism by which each of these techniques reduces NO  formation
                                                               A
and/or emissions, the applicability of the technique to new industrial
boilers, the design or operating factors which influence the NO  reduction
                                                               A
performance of the control technique  on an industrial boiler, and any impact
these controls may have  on the design and operation of the boilers is
discussed in Subsections 4.3.1  through 4.3.6.  Data of the type reported for
FGD systems is not available for NO  combustion modification reliability/
                                   A
operability.  However, a number of qualitative factors or concerns which may
impact the operability of these techniques are discussed in the following
sections.  Both short- and long-term  performance data are available for LEA,
SC, FGR, and RAP applied to various industrial boiler types and fuels.
These data are presented and discussed in Subsection 4.3.7.  No performance
data for ammonia injection were available for operating commercial boilers.
4.3.1 Low Excess Air
     Burner and boiler manufacturers  usually recommend the lowest excess air
level consistent with safe operation  and prevention of smoke for a given
burner/boiler/fuel combination.   However, industrial boilers normally
operate  with excess air levels above  those recommended by the burner or
                                    4-101

-------
boiler manufacturers.  For example, a gas/oil burner designed to operate
                                                                        no
with 10 to 15 percent excess air often operates with 20 percent or more.
This additional combustion air provides a safety margin designed primarily
to prevent smoke emissions during sudden load surges.  It also allows for
minimal operator supervision and simple combustion air control equipment.
This additional excess air, however, provides extra oxygen to the flame
zone, and results in increased NO  formation.  Excess air levels higher than
                                 A
the manufacturer's specification also reduce the thermal efficiency of the
                                                                         QQ
boiler by increasing the volume of heated gas released to the atmosphere.
     In low excess air (LEA) operation the primary combustion air flow is
reduced; with less combustion air, both thermal and fuel NO  formation are
        100                                                *
reduced.     In general, the further the excess air is reduced on a given
boiler, the lower the NO  emissions.
                        /\
     4.3.1.1  Process Description
     4.3.1.1.1  System.  LEA operation is achieved by reducing combustion
air flow to the windbox serving conventional burners.  Air flow control to
the windbox of gas- and oil-fired firetube and packaged watertube boilers is
accomplished by closing the inlet vanes of constant speed forced draft fans
or by closing the vanes at the windbox inlet (if these are provided), or
both.  Figure 4.3-1 illustrates the location of the fan and windbox inlet
vanes on two typical arrangements of single-burner packaged boilers.  In
larger gas- and oil-fired industrial boilers equipped with variable speed
induced and forced draft fans, the speed of both fans is controlled to vary
                                                                 97
the airflow while maintaining the design pressure in the furnace.    Since
pulverized-coal and large gas- and oil-fired boilers are generally
multiburner units, combustion air control requires a compartmented windbox,
and the desired excess air level is obtained by altering the speed of the
fans.  For stoker coal-fired boilers, LEA operation can be achieved by
reducing the undergrate air flow.  This is accomplished by adjusting the air
vanes and the speed of the fans.
     4.3.1.1.2  Development status.  Low excess air controls are currently
being applied to many types of boilers to improve thermal efficiency and
reduce fuel costs.
                                    4-102

-------
                                         FAN
         MOTOR
                           CONTROL
                           LINKAGE

                              BURNER
                             COMBUSTION
                           AIR ADJUSTMENT
                              ASSEMBLY
                                                              FAN
                                                             INLET
                                                             VALVE
  (A) CONTROLLING AIR FLOW TO THE
     BURNER WITH FORCED DRAFT FAN
     INLET VANES
                                                            WINDBOX
                                                           INLET VANES
                                 BURNER
                   FAN/MOTOR
                   ASSEMBLY
                                                  COMBUSTION
                                                AIR ADJUSTMENT
                                                   ASSEMBLY
 .(B) .CONTROLLING AIR FLOW TO THE
     BURNER WITH WINDBOX INLET VANES
Figure 4.3-1.
Schematics of two  single burner units  for packaged
boilers  showing location and  control of combustfbn
airflow  vanes.
                                  4-103

-------
     Recently, manufacturers have been marketing oil- and gas-fired burners
designed specifically for LEA firing.  These are termed LEA burners
(distinct from the low NO  burners discussed in subsection 4.3.4).   These
                         A
burners can safely support complete combustion at oxygen levels lower than
                              98
those of conventional burners.    These burners are already being installed
on new industrial boilers, primarily to improve boiler thermal  efficiency  in
                               98
light of escalating fuel costs.
     4.3.1.1.3  Applicability to industrial boilers.  Low excess air
techniques are applicable to all industrial boiler types and fuels.  In all
cases, an oxygen trim system is recommended to ensure safe, efficient,
continuous operation of the boiler with no smoke.     Commercially available
oxygen trim systems permit automatic LEA operation throughout the boiler's
load range.  Excess air in the boiler is measured by the excess oxygen
concentration in the flue gas.  A discussion of the relationship between
excess air and excess CL is presented in Reference 102 (p.6-9).  The oxygen
trim system, which consists of in-stack 02 and CO monitors that control
airflow to the windbox, is currently being used in the field.  Proper
maintenance of these monitors is important to maintaining good combustion.
     For stokers clinker formation is a potential concern during LEA
operation.  If the undergrate airflow is maintained sufficiently high
                                      103
clinkers should not form on the grate.     The use of an oxygen trim systems
will help ensure reliable continuous operation at the proper 02 level.
     4.3.1.2  Factors Affecting Performance.  For a given set of boiler
operating conditions the NO  reduction performance of LEA depends directly
                           X
on the excess combustion air setting -- the larger the reduction in excess
air, the greater the decrease in NO  emissions.  For coal the reduction of
                                   A
one percentage point in 02 concentration, from a typical normal operating
level of between 5 and 10 percent 02> represents about a 10 percent
reduction in excess combustion air requirements.  For oil and gas the one
percent drop in 0? concentration (from a normal operating level baseline of
between 4 and 8 percent 0?) represents a 5 percent reduction in excess air
             101
requirements.
                                   4-104

-------
     Long-term emission tests conducted by EPA demonstrate the NO  reduction
                                                                 A
performance of LEA.   The continuous monitoring data for residual  oil  and
spreader stoker boilers, discussed in Section 4.3.7, show that the partial
correlations between NO  emissions and excess oxygen (that is, the correla-
                       A
tions between emissions and excess oxygen with boiler load variations
factored out) are statistically significant and positive.
     The effectiveness of LEA in reducing NO  emissions from industrial
                                            7\
boilers varies with  fuel type and boiler design.   Operation at LEA levels is
                                                               104
generally more effective in reducing the thermal  NO  component.      Since a
                                                   A
large part of the total NO  emissions from coal-  and residual  oil-fired
                          A
boilers are due to fuel NO  formation and since LEA operation  primarily
                          A
impacts thermal NO , there is a limit to the degree of total NO  reduction
                  X                                            A
that can be achieved through LEA on coal- and residual oil-fired boilers.
This point is discussed further in Section 4.3.7-
     Although it is  desirable to reduce combustion air as much as possible
for NO  control purposes and to reduce fuel costs, the excess  air level  must
      A
be maintained above minimum levels to avoid incomplete combustion and
corresponding higher emissions of CO, HC, and smoke.  The minimum excess air
levels achievable without measurable increases in CO, HC, and  smoke
emissions vary according to the fuel burned, the  firing mechanism, and the
boiler operating load.  Table 4.3-1 presents the  excess oxygen levels, based
on numerous field tests of existing industrial boilers, that are considered
indicative of minimum levels for safe operation at firing rates above 80
                          102
percent of rated capacity.

                 TABLE 4.3-1.  SAFE OPERATING LEVELS FOR LEA
                           Minimum excess Op           Typical normal
 Fuel/firing type               (percent)             excess Op (percent)
 Natural gas                    0.5-3                      4-8
 Oil                              2-4                      4-8
 Coal/Pulverized                  3-6                      5-9
 Coal/Stoker                      4-8                      6-11
                                    4-105

-------
Also listed for comparison, are typical normal operating excess 02 levels
for industrial boilers.  The actual minimum excess 02 concentration marking
the onset of incomplete combustion varies with boiler load, and generally
                                 97
increases as the load is reduced.    Fuels with low carbon/hydrogen ratio,
such as natural gas, can achieve lower excess air levels than heavy oil  and
                                                            105
coal before increases in CO and carbon soot formation occur.
     Variations in ambient conditions, such as temperature, pressure and
moisture, that alter the density of the combustion air affect burner excess
air and excess 02 at a given setting.  The forced draft fans introduce a
constant volume of air to the furnace, but the mass flow changes according
to its density.  In the absence of compensating controls, temperature
variations of 10K (20°F) would change the 02  concentration by about one
percentage point for gas-, oil- or pulverized coal-fired boilers, which
normally operate at about 4 percent excess 0« (20 percent excess air).  The
same temperature variation would also change  the Op concentration by about
one percentage point in a stoker boiler which normally operates with about
8 percent excess 02 (60 percent excess air).  These changes in excess 0«
concentration could result in changes in NO   emissions on the order of about
                                           A
5 percent for all fuel/boiler types.  However, new industrial boilers
equipped with flue gas monitors and oxygen trim systems automatically adjust
combustion air flow to offset these ambient variations and maintain a
                                        98
constant excess 02 level in the firebox.
4.3.2  Staged Combustion
     A second combustion modification technique applicable to industrial
boilers is staged combustion (SC).  This technique is often used in
combination with LEA firing and involves diverting a fraction of the
combustion air from the burner(s) and injecting it into the furnace beyond
the burner.  Depending on the amount of combustion air that is diverted, the
burners can be made to operate near or below  stoichiometric conditions.  (At
stoichiometric conditions, 100 percent of the air theoretically needed for
complete combustion is injected through the burner windbox.)
                                     4-106

-------
     4.3.2.1   Process  Description
     4.3.2.1.1   System.   Like  LEA,  SC  reduces  oxygen  availability  and  flame
temperatures  in  the  primary combustion zone,  resulting  in  lower  thermal  and
fuel  NO  formation.  The  additional  staged  air permits  the combustion
       A
process to go to completion, oxidizing any  unburned fuel and  CO  formed in
the air-deficient combustion zone.   The ports  used to inject  the staged  air
downstream from  the  primary combustion zone are normally referred  to as
overfire air  (OFA) ports, sidefire  air (SFA)  ports, or  simply NO  ports.
                                                                A
Depending on  the boiler design,  either OFA  or  SFA ports can be used to
                  no gg
inject staged air.  '     Figure  4.3-2 illustrates schematically the
application of OFA for large units  and SFA  for packaged units.   In SC  for
multiburner units the  OFA ports  are located above the top  burner level.
Unlike OFA, SFA  is injected from the sides, the top,  or the bottom of  the
furnace.  In  other respects the  two techniques are the  same.
     Although staged combustion  is  mainly applicable  to boilers  with burners
(pulverized coal-fired, oil-fired,  and gas-fired boilers), staged  combustion
also occurs in stoker  boilers.   Stokers generally achieve  some degree  of
staged combustion by their inherent design.  Fuel is  burned relatively
slowly on a grate as compared  to the rapid  suspension burning which occurs
at burners.  Staged  combustion is also encouraged through  the use  of OFA
ports which are  used on most stokers to reduce smoking. OFA  tends to  reduce
undergrate air creating a locally oxygen deficient zone at the fuel bed.
Further staging  of combustion  air with larger  fractions of air introduced at
the OFA ports has been attempted (see  Section  4.3.7.1).
     For oil- and gas-fired boilers, a common  method  of achieving  staged
combustion is to take  one or more burners out  of service  (BOOS).  Burners no
longer firing fuel can then be used as OFA  or  SFA ports.   However, many oil-
and gas-fired boilers  use only a single burner, making  BOOS impossible for
these units.   For single  burner units, staged  combustion must use  separate
OFA ports. Many single burner oil- and gas-fired boilers  include  provisions
for OFA or SFA air ports  allowing staged combustion  controls  to  be used.
     4.3.2.1.2  Development status. The development  status of SC  for
various fuels and equipment types is summarized in Table  4.3-2.   SC  has been
                                    4-107

-------
  OVERFIRE
  AIR PORTS
 BURNERS
WINDBOX
WINDBOX
              SIDEFIRE AIR PORTS
                                       BURNER	:n*r-^"\
                                                  OOOOQOOOOOOQO QOQCOOQOQ]
                  (A) OFA PORTS
                  (B)SFA PORTS
   Figure 4.3-2.  Schematic of industrial watertube boilers equipped with (A) OFA
                ports and (B) SFA ports.

-------
     TABLE 4.3-2.  DEVELOPMENT STATUS OF STAGED COMBUSTION  FOR  APPLICATION TO  INDUSTRIAL BOILERS

Boiler Type

Packaged
erected

COAL
and field
stokers


Field erected
pulverized


OIL
Field erected
watertube
AND NATURAL GAS
Packaged
watertube

Packaged
firetube
Status         Available but not      Available and     Available  and     Available    Not  available
               implemented            implemented       implemented           and       R&D  status
                                                                          implemented


aMeans that the control technique is commercially offered,  but  is  not presently being  implemented
 for emission control.

-------
demonstrated and used commercially on large field-erected pulverized coal-,
oil-, and gas-fired industrial boilers.  For example, new coal-fired units
with heat input capacity greater than 73.3 MW (250 x 106 Btu/hr) thermal
input sold since 1971 are equipped with OFA injection ports to meet
40 CFR 60 Subpart D New Source Performance Standards for NO .   In California
                                                           A
OFA is used routinely on large residual oil- and gas-fired utility boilers,
generally larger than 73.3 MW (250 x 106 Btu/hr) thermal input, to meet
State and local NO  regulations for existing units.     Since large field-
                  A
erected industrial boilers are very similar in design and operation to
utility units, they exhibit similar NO  emission levels and are amenable to
                            107
the same control techniques.     OFA ports are a common design feature of
industrial stoker-fed boilers, primarily to complete combustion and control
smoke.  Staged combustion has also been being used on new and existing oil-
                                                            108
and gas-fired steam generators in the California oil-fields.
     4.3.2.1.3  Applicability to industrial boilers.  SC is applicable for
all fuel types but not to all existing boiler types.  Increased combustion
staging in existing stokers has been attempted in recent field tests to
lower NO  emissions.  The technique involved a reduction in undergrate air
        A
flow.  Although NO  emissions were reduced in most cases, consistent NO
                  X                                                    X
reductions were not demonstrated for all stokers with increased combustion
staging.  In general, OFA ports on existing stokers are neither specifically
                                        109
designed nor positioned for NO  control.
                              A
     Automatic controls which maintain prescribed airflows to the OFA ports
and individual burners to allow more precise operation are commercially
available.  For example, automatic control systems have been installed
recently on two 190 MW (650 x 10  Btu/hr) thermal input pulverized
coal-fired industrial boilers.
     Potential impacts of OFA for PC boilers include increased smoke and
particulate matter emissions.  Increased furnace slagging and corrosion can
also occur when severe SC conditions are implemented.  These impacts can be
avoided by proper maintenance of excess air levels and proper distribution
of air between burners and OFA ports.  The combustion air metering system
requires a flue gas monitoring system which includes, as a minimum,
                                     4-110

-------
continous 02 and opacity monitors.   A compartmented windbox is  also required
to assure equal  distribution of windbox air to each burner.   These  control
features are commercially available and are already being  implemented  in  the
r- 1A 110
field.
     With stoker-coal-fired boilers, undergrate air flow needs  to be
maintained high  enough to prevent clinker formation, or the bed needs  to  be
poked periodically to  break up any forming clinkers.  For  oil-  and  gas-fired
units, potential problems of smoke and combustible  emissions can be avoided
by operating the unit  with an oxygen trim system, and maintaining a minimum
                                                                       102
of 3 percent excess oxygen for oil-firing and 2 percent for gas-firing.
     4.3.2.2  Factors  Affecting Performance.   The success  of the staged
combustion technique depends primarily on the location of  the secondary air
injection ports  and the careful control of the airflow between  the  OFA or
SFA ports and the windbox.  Utility boiler experience with staged combustion
has shown that ports located too close to the convective section may cause
high steam temperatures, incomplete combustion, or  both.  Conversely,  OFA or
SFA ports located too  close to the burners (or fuel bed in the  case of
stokers) may decrease  the NO  reduction performance.    Manufacturers of
                            A
large industrial boilers have relied on utility boiler experience to locate
OFA or SFA ports to lower NO  formation without affecting  steam temperatures
                                 no
or causing incomplete  combustion.
     The partitioning  of combustion air between the OFA or SFA  ports and  the
windbox, together with the overall  excess oxygen level, determine the  burner
stoichiometry (or undergrate air in the case of stokers).   With combustion
of coal or heavy fuel  oil, operational and safety problems caused by
slagging and corrosion can be avoided by maintaining burner air feed rates
slightly above stoichiometric conditions (e.g., 5 percent  excess air at the
         112
burners).
     Commercially available airflow controls can be used to maintain the
required staged  air injection and windbox combustion air flowrates
                                 112
throughout the boiler load range.     With distillate oil  and gas,  burner
stoichiometries  as low as 90 percent (i.e., combustion air 10 percent  below
that required for complete combustion) are often possible, but  careful
                                   4-111

-------
control of the operating parameters is required to avoid losses in boiler
           112
efficiency.
4.3.3  Flue Gas Recirculation
     A third technique for NO  control by combustion modification is flue
                             A
gas recirculation (FGR).  This technique involves extracting a portion of
the flue gas and returning it to the furnace through the burner windbox.
     4.3.3.1  Process Description
     4.3.3.1.1  System.  Figure 4.3-3 shows schematics of FGR installations
on both a  large and a small packaged industrial boiler.  The systems consist
primarily  of an FGR fan assembly, an apportioning and mixing systeir, and
associated ducting connecting the stack (or flue gas duct) to the windbox.
The forced draft fan has to be larger when FGR is used than without recircu-
lation to  overcome the increase in pressure drop caused by the recirculation
of flue gas through the burners.  The recirculated flue gas absorbs some  of
the heat released during combustion.  This lowers the bulk furnace gas
temperature, resulting in a reduction of thermal NO  formation.  Further-
                                                   A
more,  the  addition of flue gas reduces the oxygen concentration in the
combustion air.  The effect is to reduce NO  formation by decreasing the
                                            114
oxygen available to react with the nitrogen.
     4.3.3.1.2  Development status.  FGR is commercially available and
applicable to all gas- and distillate oil-fired industrial boiler types.
For example, in 1978, FGR was installed on two new 15 MW (50 x 106 Btu/hr)
thermal input packaged watertube gas/oil-fired boilers which are now in
operation  in Southern California.
     FGR is not, however, as effective for residual oil- and coal-fired
boilers.115  When these fuels are burned, as much as 40 to 60 percent of the
total  NO   emissions may be attributed to formation of NO  from fuel-bound
        x                                               *
nitrogen.  Limited test data have shown that recirculation rates of up to
15 percent decreased NOV emissions by 17 percent when firing high nitrogen
                       A
fuels  whereas a similar recirculation rate decreased NO  by as much as
                                                       A
50 percent for gas and distillate oil-fired boilers with no air
   .    . 107,111,116
preheat.    '
                                    4-112

-------
BURNERS  ::      ::
                                     AIR
                  APPORTIONING
                    DAMPERS
                     FORCED DRAFT FAN


FLUE GAS RECIRCULATING FAN
             (A) LARGE INDUSTRIAL BOILER APPLICATIONS
                 STACK
                                                 FGR DUCTS
                  FRONT VIEW
               (B) SMALL PACKAGED BOILER APPLICATIONS
 Figure 4.3-3.  Schematics  for FGR  systems  for industrial  boilers.113
                                      4-113

-------
     4.3.3.1.3  Applicability to industrial boilers.  Implementation of FGR
for NO  control requires extra fan capacity and ducting.  Fans are reported
      A
to erode rapidly at the high operating temperatures encountered which may
increase safety hazards and operating problems.
     By designing the burner and windbox to account for the increased gas
flow, and by maintaining maximum FGR rates at a safe 20 to 25 percent flame
stability can be maintained.  Some burner designs are capable of with-
standing slightly higher FGR rates without incurring flame instabili-
ties.   '     Flame sensors should be located and their sensitivity adjusted
to detect the onset of combustion instability.
     4.3.3.2  Factors Affecting Performance.  The recirculation rate is the
only FGR operating parameter that can be varied to control NO  reductions,
                                                             A
and, as shown in Figure 4.3-4, NO  emissions decrease as the recirculation
                                 A
rate is increased.  It is important to note that these curves indicate
percentage reductions in emissions rather than absolute reductions (ng/J or
lb/10  Btu).  Thus the absolute emission reduction may actually be higher
for the residual oil-fired boiler compared to the natural gas-fired boiler
due to different uncontrolled emissions.  The potential for flame
instability at high FGR rates generally limits recirculation to 25 to
30 percent.106'111
4.3.4  Low NO  Burners
             A
     4.3.4.1  Process Description.  New burner designs are being developed
for industrial boilers which alter the mixing of air, fuel, and combustion
products within the burner flame zone to reduce N0v formation.  Lower NO
                                                  X                     A
emissions are obtained by peak flame temperature reduction, staging, and
local combustion product recirculation.  For example, commercially available
LNB's for coal-fired utility boilers use delayed fuel/air mixing and low
turbulent flames to produce a staging effect.  The oil-fired LNB's that are
currently being developed may use a combination of cooling and staging.
Flame surface area is increased for greater heat dissipation.  Local gas
recirculation is promoted to cause rapid quenching of the flame and cool the
                                   4-114

-------
    100
    80
   1
   0)
   c

   "55

   8

   £60
   o
   +••

   0)
   S
   0>
   a
  a
  01
  cc
    20
                                     NATURAL GAS D


                                    DISTILLATE OIL A


             50% NATURAL GAS AND 50% RESIDUAL OIL 0

                                     RESIDUAL OIL O
                              a
         a
                  10           20          30

                 FLUE GAS RECIRCULATION, percent
                                       40
Figure  4.3-4
FGR test  results on a 5.1 MW (17.5 x 106  Btu/hr)

packaged  watertube boiler. 106   (NO air preheat)
                           4-115

-------
combustion process, and controlled air/fuel mixing is used to provide
staging.118
     Low NO  burners have been classified by a variety of schemes.   For
           A
purposes of this evaluation, a low NO  burner is considered to be any  burner
                                     A
that internally achieves either of the following NO  reduction techniques:
                                                   A
       •  Staged combustion (distributed mixing, multiple stage combustion,
          or off stoichiometric combustion)
       •  Self-recirculation
Burners designed to operate under LEA conditions but which do not incor-
porate either of the above techniques are not considered low NO  burners.
                                                               A
     In general, commercial demonstration experience with low NO  burners
                                                                A
applied to industrial boilers is very limited.  Field testing of small
boilers (3 MW or 10 x 10  Btu/hr) firing gas and oil has occurred    and low
NO  burners (self-recirculation type) are currently in use at three 15 MW
  x     /-                 ?^3
(50 x 10  Btu/hr) boilers.     However, several vendors are offering low NO
                                                                          A
burners for certain applications, and widespread commercial utilization of
this technology could occur with the next few years.
     The major factor inhibiting widespread application of low NO  burners
                                                                 X
to industrial boilers appears to be burner size.  Use of low NO  burners on
                                                               A
process heaters, where burner size is generally smaller than 5.9 MW
(20 x 10  Btu/hr), is widespread.  Industrial boilers, on the other hand,
may use burners as large as 73 MW (250 x 106 Btu/hr).     There is no
technical constraint to using multiple small low NO  burners on large
                                                   A
industrial boilers, however, the cost of the multiple burner boiler may be
higher than an equivalent single burner boiler.  This cost difference  is
partly due to the current trend in which single burner boilers can be  shop-
erected rather than field-erected.
     4.3.4.2  Factors Affecting Performance.  Preliminary results suggest
that NOV emissions increase from gas/oil-fired LNB as oil temperature  is
       T?l
raised.     No final assessment has been made of this effect nor of attempts
to resolve it.  Since some LNB designs under development may lead to
                                   4-116

-------
extended flame zones,  it may be necessary to use an enlarged firebox to
avoid flame impingement on the back wall  of the boiler.   However,  the need
                                                  112 iiQ
for an enlarged firebox is not clear at this time.    '
4.3.5  No Combustion Air Preheat or Reduced Air Preheat
     4.3.5.1   Process  Description.   Using combustion air at ambient tempera-
tures instead of preheating it results  in a lower peak  temperature in the
primary combustion zone.   This in turn  lowers thermal  NO  production.  Most
                                                        A
industrial  watertube boilers with design  heat input capacities greater than
15 MW (50 x 10  Btu/hr) recover some flue gas heat  in combustion air
preheaters  or feedwater economizers to  maximize thermal  efficiency.      The
installation  of an economizer instead of  a combustion air preheater on new
boilers will  result in lowered peak temperatures while  still  allowing for
                                 123
effective flue gas heat recovery.    Lowering peak temperatures  is
primarily effective for reducing thermal  NO , but has little effect on fuel
                                           A
NO .   Hence,  the technique of no combustion air preheat  will  result in
  A
higher percent reductions for low nitrogen fuels — distillate oil and
            12?
natural gas.
     4.3.5.2   Factors  Affecting Performance.  The only factor affecting the
NO  reductions achievable by reducing combustion air preheat is the degree
  A
of air preheat reduction.  Limited testing on distillate oil- and  natural
gas-fired boilers has  indicated that the  reduction  of combustion  air preheat
is effective  in reducing NO  emissions  over a wide  range of combustion air
temperatures  — 300 to BOOK (80° to 440°F).106  RAP is  not as effective on
residual oil- and coal-fired industrial boilers due to  the fuel nitrogen
contribution  to the total NO  emissions and the ineffectiveness of RAP in
                            A
reducing fuel NO  emissions.  The effectiveness of reducing NO  emissions by
                A                                             A
using an economizer instead of an air preheater can be seen in the
difference in NO  emissions between boilers with and without air preheat.
                A
(See Sections 4.3.7.3  and 4.3.7.4).
4.3.6  Ammonia Injection
     4.3.6.1   Process  Description.  Ammonia (NH^) injection involves the
noncatalytic decomposition of NO  in the flue gas to nitrogen and water
                                A
using ammonia as the reducing agent.  This technique is often referred to as
                                    4-117

-------
selective noncatalytic reduction or thermal DeNO .   At a molar ratio of
                                                /\
1.5 moles NhL per mole NO   over 40 percent of the NO can be reduced if the
            0            X
reaction is designed to take place at a location in the boiler where the
temperature ranges from 1200 to 1260 K (approximately 1700° to 1800°F)
Outside the range of 1175 to 1350 K (approximately 1650° to 2000°F)  less
than 10 percent of the NO  in the flue gas can be reduced to nitrogen and
                           124
water by ammonia injection.     Since ammonia must be injected into  the
section of the boiler that is within the narrow optimal temperature  window,
some curtailment of load following capability may result.  Investigations
with multiple NhL injection ports are under way to seek a resolution of this
problem.  H9 injection with the NH~ can also be used to increase the
                                                      125
temperature range over which the process is effective.
     Sulfur-containing fuels present another potential problem.  The
formation of ammonium sulfate or ammonium bisulfate can cause plugging of an
air preheater or corrosion of boiler parts.  Increased frequency of water
                                   124
washing will minimize this problem.     To insure that ammonia emissions to
the atmosphere are minimized, ammonia sensors and feedback control systems
for the injectors may be required.
     Ammonia injection is applicable to all industrial boiler types and
fuels where there is access in the proper temperature range.  Although this
technique is commercially offered, it is not currently applied to any
                                     124
domestic operating industrial boiler.     Ammonia injection has been
installed on three gas- and oil-fired boilers ranging in size from about 16
to 79 MW (55 to 270 x 106 Btu/hr) thermal input in Japan.126  In the U.S.
this technique has been investigated only on pilot-scale facilities, except
for one commercial installation on a crude oil-fired thermal enhanced oil
                         127
recovery steam generator.     This installation is not currently operating
because of problems experienced with the steam generator.  Ammonia injection
is scheduled for application on large residual oil-fired utility boilers in
Southern California by 1982.
     4.3.6.2  Factors Affecting Performance.  The required reaction tempera-
tures for noncatalytic decomposition of NO with ammonia are found in
different areas of the boiler depending on its design and operating load.
                                    4-118

-------
For example,  at full  load these temperatures occur in the convective section
of both packaged and  field-erected watertube boilers.  Changing boiler load,
however, causes a shift in the temperature profile through the boiler,
reducing NO  removal  to below 30 percent.   For small  firetube boilers,
           A
optimal ammonia injection temperatures  occur directly in the firebox.   In
this area of  the boiler, cross-sectional  flue gas  temperatures are often not
uniform, causing significant degradation  of the NO  reduction performance to
                 124
below 10 percent.
     For new  units, multiple ammonia  injection grids  can be strategically
designed and  located  to compensate for  temperature gradients and shifts in
temperature profiles  with changing loads.   This technique, however, has not
                      124
yet been demonstrated.
     Other factors affecting performance  include NhU  injection rate and
residence time at optimal temperature.  The optimal  NHL/NO  molar ratio has
                                                      O   A
been established to be  approximately  1.5,  with no  additional NO reduction
gained by increasing  the ratio to 2.0.  Maximization  of the residence  time
at optimal temperature  can be achieved  by proper location of the multiple
injection grids.  A cross-sectional temperature profile will be required for
                                               124
each boiler design to identify these  locations.
4.3.7  NO.. Emission Reduction Data
         A
     This section presents available  NO  emission  data for combustion
                                       A
modifications in coal-, oil-, and natural  gas-fired industrial boilers.  The
data are presented by fuel type, boiler type, and  combustion modification
technique used.  All  data were collected  using EPA approved methods as
specified in  40 CFR 60, Appendix A.
     Each subsection  contains both continuous monitoring data and "short-
term" data.  The continuous monitoring  data were obtained on specific  units
during test periods ranging from 17 days  to 24 months.  The short-term data,
however, were collected during 30-minute  to 2-hour test periods at a large
number of industrial  boiler sites. Considerable variation is evident  in the
short-term data due to  variations in  fuel  nitrogen contents, boiler heat
release rates, burner designs, and combustion air  temperatures between
boilers.  Thus, the short-term data is  used primarily to illustrate trends,
                                    4-119

-------
whereas the continuous monitoring is more representative of the NO  emission
                                                                  A
levels that can be achieved for a specific boiler/fuel combination over a
range of operating conditions.
     Appendix C contains a more detailed listing of short-term data used to
construct plots presented in this section.  All the short-term data were
taken from Reference 128.  Appendix C also contains hourly and daily
emission data and more information on each of the continuous monitoring
tests.
     4.3.7.1  Coal-Fired Boilers.  NO  combustion modification data for
              ^—^^——__^_^_^_^_^^__     ^
coal-fired boilers are presented according to boiler type (i.e., pulverized,
spreader stoker, and other stokers) and combustion modification technique.
The combustion modification techniques for which coal-fired boiler data are
available are LEA and staged combustion.
     NO  emissions of two pulverized coal-fired boilers at Location I (1976
       A
start-up) firing low-sulfur coal were monitored continuously for 24 one-
month periods.  At this installation, two 88 MW (300 x 106 Btu/hr) boilers
share a common stack and NO  monitor.  The control techniques used at this
                           A
installation are excess oxygen control (LEA) and staged combustion (SC)
using manually adjustable overfire air compartments.  The first six months,
which are representative of the test period, are shown in Figures 4.3-5
through 4.3-10.  During the entire test period, individual 24-hour averages
ranged from 108 to 344 ng/J (0.25 to 0.8 lb/106 Btu); however, all but one
of the monthly averages were at or slightly below 258 ng/J (0.6 lb/10  Btu).
During the test period, one boiler had an average load of 71 percent of
capacity with daily loads ranging from 33 to 94 percent of capacity.  The
second boiler, discounting one extremely low load day, averaged 57 percent
of capacity, with daily fluctuations from 26 to 94 percent of capacity.  The
vendor NO  emission guarantee for these boilers is 301 ng/J (0.7 lb/10  Btu)
         A
heat input when burning coal.  A typical coal analysis indicated a nitrogen
content of about 1.6 percent and a heat content of (14,000 Btu/lb) on a dry
basis.
     In a 1977 study, the presence of oxygen in the coal fuel was
hypothesized as a contributer to increased NO  emissions in tangentially
                                             A
                                    4-120

-------
"-»
>»»

c
     350
     300
     250
     200
     150
     100
      50
                        10       15       20

                                TEST DAYS
                             25     • 30
      Figure 4.3-5
Continuous monitoring data for LEA/OFA

combustion modification on a pulverized

coal-fired boiler (month #1).
                                   4-121

-------
o»
   350
   300
   250
   200
   150
    100
    50
                       10       15       20

                             TEST DAYS
                              25
30
    Figure  4.3-6.
Continuous monitoring data for LEA/OFA
combustion modification on a pulverized
coal-fired boiler (month #2).

-------
•"3
*v.
e
    350
    300
    250
    200
    150
    100
     50
                        10
             15

          TEST DAYS
20
25
30
   Figure 4.3-7-
Continuous monitoring data for LEA/OFA
combustion modification on a pulverized
coal-fired boiler (month #3).
                               4-123

-------
Ol
    350 rl
    300
    250
    200
    150
    100
     50
                        10       15       20

                              TEST DAYS
                              25
30
   Figure 4.3-8.
Contlnous monitoring data for LEA/OFA
combustion modification on a pulverized
coal-fired boiler (month
                                  4-124

-------
cn
   350
    300
    250
    200
    150
    100
     50
                       10
    15

TEST DAYS
20
25
    Figure 4.3-9.  Continuous  monitoring data for  LEA/OFA
                  combustion  modification on a pulverized
                  coal-fired  boiler (month #5).
                             4-125

-------
       -3



        C
        X
       o
              350
               300
               250
               200
               150
               100
                50
                                    10       15


                                        TEST DAYS
                                      20
25
Figure 4.3-10.
Continuous monitoring  data  for LEA/OFA combustion modification
on a pulverized coal-fired  boiler (month #6).
                                       4-126

-------
                              129
fired pulverized coal  boilers.      Figure 4.3-11  presents  the results  of
this study.   This figure predicts  the fuel  NO  fraction  of NO  emissions as
                                             A                A
a function of coal  nitrogen  content and coal  oxygen/coal nitrogen  ratio.
This study indicates  that western  sub-bituminous  coal  may  actually result in
slightly higher NO emissions  despite the lower fuel  nitrogen content  due to
                  A
the higher coal  oxygen/coal  nitrogen ratio.   However,  data presented on  coal
properties indicates  that coals  with high coal  oxygen/coal nitrogen ratios
tend to have lower coal  nitrogen contents.  Thus,  the  two  influences tend to
balance one  another resulting  in reasonably similar fuel NO  emissions for a
                                                           A
variety of coal  types.
     As stated earlier,  staged combustion is  effective in  reducing fuel  NO
                                                                          A
emissions since it reduces the available oxygen in the flame  zone.  The four
major pulverized coal-fired  boiler manufacturers  (C-E, Babcock and Wilcox,
Foster Wheeler, and Riley Stoker)  now produce new boilers  equipped with
overfire air provisions  (staged  combustion) that  are guaranteed to emit  NO
equal to or less than  the 1971 NSPS of 301  ng/J (0.7 Ib  NO/106 Btu).130
                                                          X
Spreader Stoker Boilers
     Continuous NO emission monitoring was conducted  by EPA  on two spreader
                  A
stoker boilers equipped  with low excess air controls.  These  data, including
daily average NO  emissions, percent 0« in  the  flue gas, and  boiler load are
                A                     £
shown in Figures 4.3-12  and  4.3-13.
     The boiler at Location  II (Figure 4.3-12)  is a spreader  stoker with a
rated steam capacity  of  45,400 kg/hr (100,000 Ib/hr) which fires a high
sulfur coal.  Coal  analyses  showed an average  nitrogen  content of
1.3 percent nitrogen  and a heating value of 27,940 kJ/kg (12000 Btu/lb).
Daily average NOX emissions  ranged between  154  and 189 ng/J (0.36  to
0.44 lb/106  Btu), averaging  170  ng/J (0.40  lb/106 Btu) for the 30-day  test
period.  During the first 20 test  days, Figure  4.3-12  shows NO  emissions
                                                              A
decreasing as excess  02  is decreased at relatively constant boiler load.
Test days 27 through  30  show that  NO  emissions did not  increase signifi-
                                    A
cantly during lower load operation despite  increases in  excess air levels.
     Other results of EPA testing  at Location II  showed  that LEA operation
resulted in  a 24 percent decrease  in particulate  emissions measured at the
                                    4-127

-------
 I
I—»
r>o
Co
                   0.61-
                                                                       30  20   15
                                                                                      COAL OXYGEN
                                                                                      COAL NITROGEN
                                                                                            10
                                                                                NORMAL OPERATION
                                                                                20% EXCESS AIR
                                          0.6    0.8     1.0     1.2     1.4
                                          COAL NITROGEN CONTENT, LBS N/106 BTU
1.6
1.8
2.0
                       Figure 4.3-11   Fuel NO  formation  as a function of  the  coal  oxygen to
                                        nitrogen ratio and  the coal  nitrogen content.

-------
T


C
    190

    182


    174

    166


    158

    150
 CM
O
    8.0

    7.6

    7.2


    6.8

    6.4

    6.0
O

O
_l


Ul
     95

     85


     75

     65

     55
      45 I-
                                                          V
                   \
                              10
                                         15

                                      TEST DAYS
20
                                                               25
         Figure 4.3-12.  Continuous monitoring data  for  LEA combustion
                         modification on a spreader  stoker coal-fired
                         boiler at Location II.
30
                                         4-129

-------
Ol
c
235



225


215


205


195


185
 CM
o
9.5r-


9.0


8.5


8.0


7.5


7.0
                 I
                                            ;
O
§
o
CQ
      90
      80
 70


 60


 50

                              10

                           TEST
                                    15
  Figure 4.3-13. Continuous monitoring  data  (8-hour average) for
                 LEA combustion modification  on  a  spreader stoker
                 coal-fired boiler at Location  III.
                                  4-130

-------
outlet of the mechanical collector relative to normal operation.   Other
observations during LEA operation relative to normal operation included:  no
effects on plume opacity (constant at 10 percent), no discernable effect  on
polycyclic organic matter emissions (ROMs), and little effect on  boiler
efficiency or carbon monoxide emissions.  (See Appendix C for reference).
     At Location III (Figure 4.3-13), LEA conditions were maintained only
while the test contractor was onsite, i.e., eight hours per day,  five days
per week.  Thus, the data plotted in Figure 4.3-14 represent averages of  the
eight hour period when LEA conditions were maintained.  NO  emissions during
                                           fi              ^
LEA operation averaged 208 ng/J (0.48 lb/10  Btu) ranging from 190 to
231 ng/J (0.44 to 0.54 lb/106 Btu).  During the test period, hourly boiler
loads ranged from 37 to 76 percent of capacity, with an average load of
60 percent.  Other effects of LEA operation included a decrease of
23 percent (614 ng/J down to 474 ng/J) in particulate emissions,  a reduction
in percent opacity from 35 to 25 percent, an increase of approximately
0.5 percent in boiler efficiency, and no observable effect on polycyclic
organic matter or carbon monoxide emissions.  (See Appendix C for
reference).  For these tests the spreader stoker boiler rated at
72,600 kg/hr (160,000 Ib/hr) steam output was firing a coal of about
0.8 percent nitrogen with a heat content of 14790 kJ/kg (8500 Btu/lb).
     Figure 4.3-14 shows short-term NO  emissions data from several
                                      A
different spreader stoker boilers as a function of excess oxygen.  These
data were collected on four spreader stoker boilers ranging in size from
22,500-91,000 kg/hr (50,000-200,000 Ib/hr) of steam capacity at loads
between 35 and 100 percent.  The data in Figure 4.3-14 clearly show that  NO
                                                                           y\
emissions tend to decrease as excess oxygen is reduced.  Scatter in the test
data may be attributed to the fact that fuel characteristics and  boiler heat
release rate varied between the boilers tested.  However, all the data taken
at excess 09 levels of 7 percent or less (50 percent excess air)  fall below
           t-        C
256 ng/J (0.60 lb/10  Btu).  Seven percent excess 02 is typically within  the
range of safe excess 0^ levels for industrial stoker-fired boilers, as
discussed in Section 4.3.1.
                                   4-131

-------
      800
      700
      600
 e\j
O
(£,
o.
CL
      500
      400
      300
.  v
                                    .     •
                                                                          1.0
                                                                          0.9
                                                                          0.8
                                                                          0.7 -
                                                                             CO
                                                                          0.6
                                                                         0.5
                                                    0.4
             4    56    7   8    9   10   11    12   13   14   15   16
                                  Excess 02 %
 *Assunring 0.1 lb/10D Btu = 75  ppra NO  ? 32 0? dry.
                                   ^       «
           Figure  4.3-14.  NOX emissions vs. excess 02 - Short-term

                            data for coal-fired  spreader stokers.
                            (unstaged combustion)
                                   4-132

-------
     Figure 4.3-15 shows short-term data for a 56,750 kg/hr steam
(125,000 Ib/hr steam) spreader stoker operated under staged combustion
conditions.  Data taken under normal operating (unstaged) conditions are
also shown for comparison.  Boiler loads during these tests varied from
40 to 70 percent.  This figure shows that staged combustion has little or no
effect on NO  emissions.  By virtue of their inherent firing technique,
            ^
spreader stokers appear to achieve some degree of staged combustion without
the use of additional staging air.  Volatile matter is driven off the fuel
bed as the coal is fed onto the grate creating a fuel rich combustion zone
at the grate with lower combustion intensity and relatively slow burning.
     Figure 4.3-16 shows short-term data from two mass-fed stoker boilers.
One unit of 27,000 kg/hr (60,000 Ib/hr) steam capacity fired coal  with a
relatively low nitrogen content of 0.9 percent; NO  emissions from this unit
                                                  A
were generally lower than those from the second unit of 97,000 kg/hr
(215,000 Ib/hr) steam capacity which fired coal with a nitrogen content of
about 1.4 percent.  Both sets of data show uncontrolled NO  emissions from
                                                          ^
mass-fed stokers to be lower than those from spreader stokers - less than
215 ng/J (0.5 lb/10  Btu) under all conditions.  No reduction in emissions
was noted during staged combustion tests.  Loads during these tests ranged
from 25 to 100 percent of rated capacity.
     4.3.7.2  Residual Oil-Fired Boilers.  Figure 4.3-17 shows the results
of continuous NO  emission monitoring tests conducted at a 35,900 kg
                A
steam/hr (79,000 Ib steam/hr) residual oil-fired boiler at Location IV.
Tests were conducted using low excess air (LEA) and LEA in combination with
staged combustion.  Staged combustion conditions were simulated by removing
one of three burners from service.  Controlled NO  emissions averaged
112 ng/J (0.26 lb/106 Btu) for the 29-day test period.  Emissions during
16 days of LEA testing averaged 123 ng/J (0.29 lb/10  Btu).  Average
emissions during the remaining 13 days, when staged combustion was used,
were 98 ng/J (0.23 lb/10  Btu).  These data show that staged combustion in
combination with LEA achieves greater reduction in NO  emissions than LEA
                                                     ^
alone.  Other effects noted during LEA/staged combustion operation (relative
to unstaged combustion) included an increase in particulate matter emissions
                                    4-133

-------
                     O  unstaged combustion
                     ^  staged combustion
DUU
>> 500
-a
CM
o
8 400
 %
10
                                                          _  0.7
                                                          —  0.6
                                                          —  0.5
                                                          —  0.4
                                                          —  0.3
11
                                                                     o
                                                                      a*
                                                                     oo
Figure 4.3-15.  Short-term  emission  data  for staged
                combustion  in  a  spreader  stoker boiler.
                           4-134

-------
bUU

400

>>
^ 300
CM
O
fc*
CO
E
a 200
X
o

100

• Fuel nitrogen =
0.9%
_ o Fuel nitrogen =
1.4%
o
o
- oo —
O 0
o o
o
o o*
o —
o
° o
o •
v. •;•• •
..\
—
1 I 1 1 1 1 1 1 1 1 1 1 1 1

0.6
0.5
2=
O
X
_,
0.4 ^
o
c
oo
c+
0.3
0.2

0.1
1 23 45 6 78 9 10 11 12 13 14 15
                           Excess 0_, %
Figure 4.3-16>  Short-term emission data for two mass fed stokers.
                        (unstaged and stage combustion)
                               4-135

-------
     150

     135

     120
ox   105
      90

      75
                  • - LEA only
                  A- LEA/SCA
 00
o
11

10

 9

 8

 7

 6
               • - LEA only
               A- LEA/SCA
O
03
95-

85-

75-

65-

55- '

45-
  I
                   • - LEA only
                   A- LEA/SCA
                             10
                                 15
                              TEST DAYS
20
25
   Figure 4.3-17.
              Continuous monitoring data for LEA/SCA
              combustion modification on a residual
              oil-fired boiler at Location IV.
                            4-136

-------
(29 ng/J to 43 ng/J), a slight decrease in POM emissions,  no  effect  on
visible emissions or carbon monoxide emissions, and little change  in boiler
efficiency.
     The nitrogen content of the fuel  fired during the test was
0.26 percent.  Although this nitrogen  content is typical  of many residual
oils currently used as industrial  boiler fuels, combustion of higher
nitrogen oils (up to 0.8 to 1  percent  nitrogen) may become more common  in
the future in certain areas of the country.  Uncontrolled  nitrogen oxides
emissions tend to be higher when higher nitrogen fuels are fired due to the
increased levels of fuel NO  evolved.   No continuous monitoring data were
                           A
available for industrial units burning high nitrogen oil.   However,
short-term data were available for boilers firing residual and distillate
oils with varying nitrogen contents.
     These data were obtained  on several  different boilers at varying levels
of excess air, combustion air  temperatures, furnace heat  release rates  and
boiler loads.  Boiler capacities ranged from 7,900 to 90,000  kg/hr
(17,500 to 200,000 Ib/hr) of steam and loads varied from  18 to 100 percent.
Using standard statistical techniques, a regression was developed to relate
NO  emissions to the nitrogen  content  of the fuel at a given  level of excess
                               104
air and combustion air preheat.
     Excess air (excess oxygen), combustion air temperature (degree  of  air
preheat), and nitrogen content of the  fuel were found to  be the major
variables influencing NO  emissions from residual oil-fired boilers. The
                        A
correlations (1) between NO  and boiler load and (2) between  NO  and furnace
                                                                       104
heat release rate were found not to be significant on the  units tested.
The lack of a correlation between boiler load and NO  is  expected.   As
                                                    A
boiler load is decreased, heat release rate and thus NO  formation tends to
                                                       A
decrease.  However, the excess air rate to the boiler must be increased as
load drops, and the increased  0? leads to more NO  formation. Therefore,
                               £•                 A
the increase in excess air at  lower boiler load offsets the benefits of
reduced combustion intensity,  and there is little or no net change in NO
                                                                       A
emissions with boiler load.
                                    4-137

-------
     Figure 4.3-18 shows NO  emissions as a function of fuel  nitrogen
                           A
content for 3 percent excess 02 (low excess air) and no combustion air
preheat.  This figure was constructed using the following regression:

          E = 24.2 T°'34 A0'24 + 1055 N1'06                      (4.3-1)
where
     E = NOX emissions  (ppm at 3 percent Op, dry),
     T = Combustion air temperature (°R),
     A = excess oxygen  (mole fraction 09 in flue gas), and
                                     6
     N = fuel nitrogen  content (lb/10  Btu)
This regression was developed from 208 short-term data points (see
Appendix C - Tables C.4-17, C.4-18, and C.4-20).  These data were obtained
from several boilers both with and without air preheat.  The nitrogen
content of the oil fired in the boilers tested ranged from near zero to
about 0.8 weight  percent.
     Equation 4.3-1 shows that NO  emissions from residual oil units can be
                                 A
reduced by any (or a combination) of three methods:
     (1) reducing or eliminating combustion air preheat,
     (2) reducing the excess air (LEA operation), or
     (3) burning  oil with a low nitrogen content.
Figure 4.3-18 represents the NO  emissions expected at a LEA level of
                               A
3 percent Oo  (15  percent excess air) and no combustion air preheat.  As the
figure shows, NO  emissions are still a strong function of the fuel nitrogen
                A
level.  This results from the fact that neither LEA operation nor lower
combustion air temperatures are effective  in reducing fuel NO  formation.
                                                             /\
Other than burning lower nitrogen content  oil, the most effective technique
for reducing fuel NO  emissions is staged  combustion (see Section 4.3.2).
                    A
     Short-term data were available for 6  residual oil-fired boilers using
staged combustion (See  Appendix C).  These units burned oils with nitrogen
contents of from  0.14 to 0.49 weight percent.  These data were normalized to
3 percent excess  oxygen and no air preheat and a factor was developed to
relate the reduction in NO  emissions with staged combustion relative to
unstaged, LEA operation.     The results are shown in Figure 4.3-19.  The
                                     4-138

-------
              *To convert lb/10  Btu to
               ng/J, multiply by 430.
                           predicted NO  emissions
                                            Basis:  3% excess 0 •,

                                                    No air preheat
                                                       \	1
     0.1   0.2   0.3   0.4   0.5   0.6   0.7    0.8   0.9   1.0

                         Fuel nitrogen (%)
Figure 4.3-18.  Predicted NO  emissions from residual oil units as a
                function of fuel nitrogen content - LEA controls.
                           4-139

-------
    1.0 -T
                     LEA controls only (3% excess  (L)
--  LEA/staged combustion controls
                                   95% confidence level
                        LEA/SC predicted NO  emissions
                                           A
                                                                  Basis: 3% 0,
                                                                         No air preheat
                                     95% confidence level  LEA/SC
              0.1   0.2   0.3   0.4   0.5

                                   Fuel  nitrogen  (%)
                          0.6   0.7   0.8
 {-- H
0.9   1.0
          Figure 4.3-19.   Predicted NO  emissions  from residual oil-fired
                          boilers  vs.  fuel  nitrogen  -  LEA and  staged
                          combustion controls.
*To convert to ng/J,
 multiply by 430.
                                   4-140

-------
factor describing the effectiveness of staged combustion was found to be a
function of fuel nitrogen content.  As the nitrogen content of the fuel (and
thus potential fuel NO  emissions) increases, the NO  reduction achieved by
                      A                             A
staged combustion increases.
     Short-term emission data was also available for a small boiler
[7,900 kg/hr (17,500 Ib/hr) steam)] firing residual oil with a 0.14 percent
nitrogen content and using flue gas recirculation (FGR).  Loads ranged from
80 to 85 percent.  Sixteen tests with FGR showed average emissions of
78 ng/J (0.18 lb/10  Btu).  Emissions under low excess air operating
conditions (3 percent excess 02, no FGR) were 95 ng/J (0.22 lb/106 Btu).
These data indicate that FGR is somewhat effective in reducing NO  emissions
                                                                 A
below the levels achieved with LEA (Reference 128, pp. 315- 367).
     4.3.7.3  Distillate Oil-Fired Boilers.  No continuous NO  emission
              ^^^^^^^^^^^^^^^^™^^^~"-•     ^^^^™   - --  ^^^^^^^™~'                   ^
monitoring data were available for distillate oil-fired boilers.  However,
short-term data, under both unstaged and staged combustion conditions were
available.  The unstaged data are included with the short-term residual oil
data presented in the previous section and used to develop Figure 4.3-18.
These data are also shown in Figure 4.3-20.
     Fuel NO  formation in distillate oil-fired boilers is minimal because
            A
most distillates have very low nitrogen contents.  The most important effect
on NO  emissions from distillate units, is combustion air temperature.  The
     A
data in Figure 4.3-20 are separated into those tests conducted on boilers
equipped with combustion air preheat and those conducted on boilers with
little or no air preheat.  Nearly all units with combustion air preheat had
higher NO  emissions than boilers without air preheat.  Figure 4.3-20 shows
         A
that on the average, NO  emissions from boilers with air preheat were about
                       A
twice those from boilers without air preheat at a given excess 02 level.
Figure 4.3-20 shows NO  emissions from boilers without air preheat to be
less than 86 ng/J (0.2 lb/106 Btu).
     Figure 4.3-21 shows short-term data from a distillate oil-fired boiler
(without air preheat) under staged combustion conditions.  The one available
emission test under unstaged operating conditions is shown for comparison.
                                     4-141

-------
400,.
300
en


=   200
TOO
           O Preheated combustion air
           • No air preheat      *
                                  O
                                 oo


                               o *o
                •    t
                                                              0.5
                                                              0.4
                                                              0.3
                                                                   o
                                                                    ert
                                                                   03
                                                                   rt-
                                                              0.2
                                                              0.1
        1    2    3456    7    89     10    11   12

                            Excess 02, %

     Figure 4.3-20.   Short-term emission data for distillate
                     oil-fired boilers.

                        (unstaged combustion)
                           4-142

-------
              O  Normal  excess air

              •  Staged  combustion
    300
 CM
O
CO
a.
Q.
    200
    100
                                                         0.3
                                                         0.2
                                                         D.I
                                                                o*
4     5

Excess
                                               8
10
         Figure 4.3-21.   Short-term data for distillate
                         oil-fired boiler without air preheat.
                         (staged combustion)
                                4-143

-------
This data base is considered too limited to draw any conclusion concerning
the effectiveness of staged combustion on distillate oil-fired boilers.
     4.3.7.4  Natural Gas-fired Boilers.  Continuous NO  emission data were
                                                       A
available for a small firetube natural gas-fired boiler at Location V with  a
steam capacity of 3130 kg/hr (6960 Ib/hr).  As shown in Figure 4.3-22, NOX
emissions ranged from 27 to 33 ng/J (0.06 to 0.08 lb/106 Btu)  over the
21-day test period even though the excess 02 levels were never less than
5.5 percent.
     These data demonstrate the relatively low NO  emission rate from small
                                                 A
natural gas units without air preheat.  However, NO  emissions from larger
                                                   A
natural gas units, both with and without combustion air preheat, are
generally greater than 43 ng/J (0.1 lb/10  Btu), even at low excess air
levels.  Figure 4.3-23 presents short-term NO  emission data collected on
                                             A
several natural gas-fired industrial boilers under unstaged combustion
conditions.
     The use of combustion air preheat on natural gas-fired boilers can  have
a significant impact on both uncontrolled NO  emissions and the NO  levels
                                            A                     A
achievable at low excess air levels.  Although there is considerable scatter
in the short-term data, (due to variations in heat release rate, combustion
air temperature and burner design), the lowest NO  emissions are generally
                                                 A
observed at excess 0^ levels of less than 2.5 percent (approximately
11 percent excess air).  [As discussed in Section 4.3.1, the recommended
minimum safe excess 02 levels for gas-fired units range from 0.5 to
3 percent].
     Figure 4.3-23 indicates that elimination of combustion air preheat
leads to lower NOV emissions as evidenced by the difference in NO  emissions
                 X                                               A
between units with and without air preheat.  At excess Q^ levels of less
than 3 percent, NO  emissions from units without air preheat are about
86 ng/J (0.2 lb/10  Btu) or less.  In contrast, emissions from units with
air preheat range from 86 to 151 ng/J (0.2 to 0.35 lb/10  Btu) at excess 02
levels of less than 3 percent.
     Figure 4.3-24 shows the available short-term emission data for staged
combustion in natural gas units.  Emissions under normal operating
                                   4-144

-------
     40 r
     301
 o>
     ZD
     10
                             10
15
20
25
30
 
-------
     500,
     400
     300
o
44
CO
     200
     100
                    O Preheated combustion air
                    • No air preheat
        ocP o
         00
         o o
       o  o
°o   o  o0o«
                    On
                                                        o°o
                                           _J	I    J
                                                             I	I  	J
                                                                          0.7
                                                                           0.6
                                                                           0.5    3
                                                                           0.4
                                                                           0.3
                                                                           0.2
                                                                          0..1
            1    2    345    6    7   8    9    10   11  12   13   14   15
                                 Excess 02,  %
     Figure 4.3-23.  Short-term emission data for natural  gas fired boilers.
                      (unstaged combustion)
                                    4-146

-------
              O normal operating conditions
              • staged combustion
HUU
300
£
•o
JM 200
0
%«
n

-------
conditions are shown for comparison.  These data were obtained from five
boilers, four of which had air preheaters.  Staged combustion is somewhat
effective in reducing NO  emissions although considerable scatter is
                                                       fi
evident.  Controlled emissions were 129 ng/J (0.3 lb/10  Btu) or less in
each test conducted under staged combustion conditions.
     The only emission data for flue gas recirculation was obtained on a
small firetube unit [8170 kg steam/hr (18,000 Ib steam/hr)].   Short-term
data showed that FGR reduced NO  emissions from 28 ng/J (0.07 lb/106 Btu)
                                                                6
at normal operating conditions to an average 13 ng/J (0.03 lb/10  Btu).
(Reference 128, pp. 315-367).  This data is considered too limited to draw
any conclusions concerning the performance of FGR on natural  gas-fired
boilers.
4.4  POST COMBUSTION TECHNIQUES FOR N0v CONTROL
                                      A
     Post combustion flue gas treatment (FGT) techniques for control of NO
                                                                          X
emissions use either a gas phase reaction or liquid absorption to treat the
flue gas.  In most cases the gas phase reaction is between NO  and NFL in
                                                             A       w
the presence of a solid phase catalyst.  The catalyst is contained within a
fixed or moving bed reactor.  The NO  is converted to N9 which exits with
                                    A                  C-
the flue gas.  Systems using a liquid absorption technique contact flue gas
with the absorbent in conventional scrubbers.  The absorbed NO  either
                                                              A
remains in the scrubbing liquor and is treated in the liquid phase or reacts
with a reductant to form N2 which is liberated and is discharged with the
flue gas.131
     The NO  FGT systems discussed in subsections 4.4.1 through 4.4.3
           X
include systems designed for NO  removal only  (NO -only) as well as
                               A                  A
processes designed for simultaneous removal of S09 and NO  (NO /SO ).  The
                                                 c.       XXX
NO -only processes described are fixed bed, moving bed and parallel flow
  A
selective catalytic reduction; the NO /SO  processes described are wet
                                     X   X
scrubbing, and electron beam irradiation.  The mechanism by which each of
these techniques reduces flue gas NO  concentration, the applicability of
                                    X
the technique to new industrial boilers, the design or operating factors
which influence the NO  reduction performance of the control technique on ar
                      X
industrial boiler, and any impact these controls may have on the design and
                                     4-146

-------
operation of the boilers are briefly discussed.  No performance data taken
by approved EPA methods was available for any of the techniques discussed.
4.4.1  Selective Catalytic Reduction
     Selective catalytic reduction (SCR) is a technique involving removal  of
the flue gas NO  by reacting NO  with NHL in a catalytic reactor to form
               A               A        O
elemental nitrogen.  With the exception of the use of a catalyst it is
similar to the ammonia injection NO  control technique discussed in
                                   A
Section 4.3.
     4.4.1.1  Process Description
     4.4.1.1.1  System.  A generalized SCR process flow diagram is shown in
Figure 4.4-1.  In this process, ammonia, taken from a liquid storage tank
and vaporized, is injected at molar ratios of 0.7-1.2 moles NFL per mole of
NO   and mixed with the flue gas prior to the catalytic reactor.  The flue
  /\
gas passes through the catalyst bed where NO  is reduced to N9.  Typically,
                                            A                L.
a 1.0 mole ratio of NH~ to NO yields a 90 percent reduction in NO
                      *5                                          X
emissions.  The flue gas exits the reactor and is sent to the air preheater
and, if necessary, further treatment equipment for removal  of particulates
and S02.  Flue gas must enter the reactor at 350-400°C (662 - 752°F) since
it is in this temperature range that catalysts show the optimum combination
of activity and selectivity.  The catalysts used in most SCR processes are
oxides of non-noble metals which have shown the best combination of high
reactivity and resistance to S02 and SO., poisoning.
     The type of fuel burned in an industrial boiler plays  an important role
in the selection of the catalyst bed configuration.  The following
discussion presents three common bed configurations and the appropriate
application of these bed types to coal-, oil-, and gas-fired boilers.
     Moving Bed Reactor.  Moving bed systems for selective  catalytic reduc-
tion of NOX are applicable only to flue gas streams containing particulate
concentrations less than .998 g/dNm  (.437 grain/SCF).  Particulate concen-
trations for all coals are higher, on the order of 0.998-4.99 g/dnm
(.437-2.18 grain/SCF).  In moving bed reactors the catalyst circulates
through the reactor and is screened to remove particulates.  A second
possibility would be the use of a moving bed design which would permit the
                                    4-149

-------
-Pi
I
I—»
en
o
                                 NH
                                        REACTOR
                       BOILER
J  350-

  400°C
                         j
                                     200-250°C
                                                         AIR
                                                                                        STACK
                                Figure 4.4-1.  Typical industrial boiler SCR system.132

-------
periodic removal of catalyst for cleaning.  However, it is not expected that
moving bed systems will be used for coal-fired industrial boiler
             133
applications.
     Parallel Flow Reactor.  The distinguishing aspect of this process is
that the catalyst is produced in a variety of shapes including honeycomb,
pipe, or plate configurations.  The catalyst shapes allow particulate-laden
flue gas to pass through the reactor with no inertial  impaction of the
particles while the NO  is transported to the catalyst surfaces by basic
                      A
diffusion.  These catalysts can handle the particulate levels of boilers
firing all fuels.
     The reactors used are similar to standard fixed packed bed units
discussed below.  The catalyst is usually prepared in  small modules and
manually stacked within the reactor.  The specific arrangement will depend
                                              134
on the particular process under consideration.
     Fixed Packed Bed Reactor.  Fixed packed bed systems for selective
catalytic reduction of NO  are applicable only to flue gas streams
                                                        3
containing particulate emissions of less than .021 g/dNm  (0.009 grain/SCF).
Particulate emissions for all coals are higher than this level.  For this
reason, fixed packed bed SCR systems are not considered applicable to
                                      135
coal-fired boilers by process vendors.
     Although most SCR processes are NO -only, one parallel flow reactor
                                       X
arrangement using a copper based catalyst is capable of simultaneous NO /SO
                                                                       A   A
emission reduction.  In this case, the copper-based catalyst functions as an
SO,, adsorbent as well as a NO  reduction catalyst.  A  generalized flow
  £•                          A
diagram of this type process is shown in Figure 4.4-2.  In the reactor, S02
reacts with CuO and oxygen to form CuSO,.  Copper sulfate then promotes the
reduction of NO with NHg.  Several reactors are operated in "swing"
operation, that is, when one reactor is saturated it is taken offline  for
regeneration and a freshly regenerated reactor is brought on line.
     4.4.1.1.2  Development status.  SCR is not considered to be a
commercially demonstrated control technology for coal-fired sources in the
United States.  SCR processes have been used commercially in Japan on  gas-,
distillate oil-, and residual oil-fired industrial boilers and SCR processes
                                    4-151

-------
en
ro
                                                  PARTICIPATE BEMOVAL


                                                     AND STACK
                                                                             OFF
                                                                             GAS
                                                                                           PRODUCT

                                                                                      (S, S02 (i), OR H2S04)
                          Figure 4.4-2.   Flow  diagram  of a simultaneous  NOX/SOX SCR system.
                                                                                                   136

-------
on coal-fired utility boilers are currently under construction in that
country.  Ongoing studies in the United States are investigating NO -only
                                                                   A
and NO /SO  SCR performance with coal  combustion in pilot-scale facilities.
      X   A
There is no full-scale U.S. or Japanese SCR installation with documented
performance in accordance with EPA test methods, although removals in  excess
of 90 percent have been reported for Japanese gas- and oil-fired boiler SCR
applications.  The earliest U.S. commercial demonstrations are planned for
                                                                        135
1982 on two low-sulfur oil-fired utility boilers in Southern California.
SCR technology is also expected to be applied to steam generators involved
in thermally-enhanced oil recovery projects in California.  EPA is
sponsoring two pilot scale evaluations of the technology on coal-fired
utility boilers.  The Electric Power Research Institute (EPRI) is also
sponsoring a coal-fired utility boiler SCR pilot plant.
     4.4.1.1.3  Applicability.  SCR is applicable to all industrial  boiler
types.  Japanese SCR technology relies primarily on two catalyst reactor
designs: fixed bed and parallel flow.   The fixed bed reactor which consists
of tightly packed catalytic granules,  can become plugged if there is high
grain loadings in the flue gas.  Therefore, the Japanese developers
                                        138
recommend it only for gas-fired boilers.     In the parallel flow reactor,
the catalyst arrangements resist plugging and blinding of catalytic  surfaces
due to dust in the flue gas and are recommended when burning residual  oil
      n.1J"
       137
         134
and coal.     All  three U.S.  pilot-scale studies use the parallel  flow
design.
     The flue gas flow rate from the boiler and the design NO  control  level
                                                             A
determine the catalyst volume necessary.  Increases in either will  increase
the required reactor size.  The uncontrolled NO  concentration is  primarily
                                               A
a function of fuel type used to fire the boiler.  Higher NO  concentrations
                                                           A
require larger NH- storage and vaporization equipment; reactor size is  not
significantly affected by NO  concentration for a constant control  level.
                            A
Boiler load can affect several parameters including flue gas temperature,
flow rate and NO  concentration.  Temperature control  equipment may be
                X
necessary to accommodate large boiler load variations.  Where such
                                    4-153

-------
variations are expected, some equipment overdesign may be warranted to
                                1 "3Q
insure a constant control level.
     The impacts of parallel flow and fixed packed bed SCR systems on  boiler
operation and maintenance should be minor.  The primary impact is on the  air
preheater since dry residual NHL will react with SCL as flue gas temperature
decreases to form ammonium bisulfate and ammonium sulfate.  Bisulfate  is
formed by a one-to-one reaction between NHL, SCL, and hLO in the flue  gas:
               NH3(g) + S03(g) + H20(g)   -   NH4HS04 (s)

Bisulfate is corrosive when it condenses on unprotected surfaces; therefore,
use of corrosion-resistant material is warranted where bisulfate deposits
are probable.  A minimum NhL injection ratio is also recommended for low NFU
emissions and bisulfate formation.  Heat exchanger temperatures must be kept
above bisulfate formation and acid condensation points and should be
equipped with a cleaning apparatus to remove any deposits of these
        .  140
compounds.
     There are some potential adverse environmental impacts associated with
the use of SCR, including gaseous NhU emissions, disposal of ammonium
bisulfate or ammonium sulfate and disposal of spent catalyst.
     4.4.1.2  Factors Affecting Performance.  An important design variable
with respect to SCR performance is the space velocity which is expressed as
the volume of catalyst required to treat one volume per hour of flue gas.
Space velocity requirements vary with catalyst formulation, catalyst shape,
and control level.  Typical values of space velocity for various catalyst
shapes are shown in Table 4.4-1.  Also shown are other catalyst design
variables such as catalyst dimensions, gas velocities, bed depth and
pressure drop.  Ranges of values are used since specific values are
different for each catalyst.  The values shown are for a design NO  removal
                                               141
of 90 percent and an NH-,/NO  mole ratio of 1:1.
                       O   A
     Both NH-/NO  ratio and space velocity will change with removal level.
            0   A
The NH-/NO  mole ratio will range from 0.7-1.2 for control levels of 70 to
      *3   A
                                    4-154

-------
 TABLE 4.4-1.   CATALYST DESIGN VARIABLES FOR VARIOUS  CATALYST SHAPES        141

                (Basis:  90% NO   removal  at NH-/NO   ratio  of 1:1, 350-400°C)
                               X               *3   X


Catalyst size (mm)
Thickness
Opening
Gas velocity (m/sec)a
Bed depth (m)
SV (1,000 hr~])b
Pressure drop (mmHgO)
Honeycomb
(metallic)

1
4-8
2-6
1-2
5-8
40-80
Honeycomb ,
tube (ceramic)

2.3-5
6-20
5-10
1.5-5
4-8
40-160
Parallel Plate
(Ceramic)

8-10
8-14
5-10
4-6
1.5-3
80-160
Metallic

1
5-10
4-8
2-5
2-4
60-120
 Velocity at 350-400°C in open column (superficial velocity).
L             «J                         «J

 Gas volume (Mm /hr)/catalyst bed volume  (m ).
                                   ,4-155

-------
           142
90 percent.     The operating temperature range for most of these processes
is about 300-500°C (527-932°F), though more efficient NO  removal usually
                                                        A
occurs in the higher portion of this range.  To maintain the reactor
temperature at desirable operating levels during periods of reduced boiler
load, most process vendors recommend bypassing a part of the flue gas around
the economizer.  In some pilot plant and larger operations, auxiliary
heaters have been used to maintain reactor temperatures during turndown.
     4.4.1.3  Emissions Data.  While there are a number of commercial SCR
systems presently treating oil-fired flue gas in Japan, the data on these
units are limited mostly to a single reported removal level.  These tests
give only point values of removal and not a set of continuous data.  In
addition, the test method and boiler operating conditions are not given.
     The results from several catalyst life tests conducted in Japan
demonstrate NO  removal efficiencies of 70 to 90 percent or greater for low
              A           t                   .
sulfur, high sulfur, and heavy oil-fired utility boilers.  NO tNH., ratio  for
these tests was equal to 1.  Other operating parameters were not specified.
4.4.2  Wet Scrubbing
     Wet FGT processes are, in most cases, designed to take advantage of
technology already available from previously developed FGD systems.  Most
wet FGT processes were originally designed as simultaneous NO /SO  systems.
                                                             A   /\
Unfortunately, NO, which represents the majority of industrial boiler NO
                                                                        A
emissions has an extremely low solubility in aqueous solutions.   N02, which
is the lesser component of industrial boiler NO  emissions, is much more
                                               X
soluble than NO although the solubility of N02 is poor relative to S02-
Therefore, the major task associated with any wet NO  removal process is  the
                                                    A
absorption of the NO  by the scrubbing solution where it can be concentrated
                    A
and converted into other nitrogen compounds.
     There are two common methods of removing NO  from flue gas, direct
                                                A
absorption of NO  in the absorbing solution and gas-phase oxidation to
                X
convert the relatively insoluble NO to N02 followed by absorption.  Wet NOX
removal processes are generally classified as absorption or oxidation
processes, depending on whether or not the flue gas is treated with a
gas-phase oxidant before absorption.  Additionally, each of these
                                    4-156

-------
classifications is divided based on the fate of the NO  after it has been
                                                      A
absorbed by the scrubbing solution.  Processes which reduce the absorbed NO
                                                                           A
either partially or completely to molecular nitrogen or complex
nitrogen-sulfur compounds are classified as reduction processes.  Processes
which do not reduce the absorbed NO  are absorption processes.   Absorption-
                                   /\
oxidation processes involve a absorption of NO and liquid phase oxidation to
nitrates which must be removed by wastewater treating techniques.   Thus, the
wet NO  processes can be categorized into one of the following  groups:
      A
oxidation-absorption-reduction, oxidation-absorption, absorption-reduction,
or absorption-oxidation.  A very simplified flow diagram for a  wet NO /SO
                                                                     A    A
process is shown on Figure 4.4-3.  The various processes developed in each
of these categories are described in the technology assessment  report on NO
                                                                           A
flue gas treatment listed in Table 4-1.
     Development to date for all of these processes has not proceeded beyond
the pilot plant stage.  Numerous processes have been piloted on coal  and
oil-fired utility boilers but to date prototype plans have all  been
abandoned in favor of SCR development.  Major problems encountered include
high energy penalties, difficult water treatment problems, and  high sorbent
replacement rates.
     At least two vendors are currently offering absorption-oxidation
systems for oil-fired steam generator applications for thermally enhanced
oil recovery operations.  No commercial applications exist however and  no
pilot or prototype data have been published.
4.4.3  Electron Beam Irradiation
     This dry process utilizes an electron beam to bombard the  flue gas,
removing NO  and S09 in the process.  A block flow diagram for  the process
           A       £
is shown in Figure 4.4-4.
     Flue gas downstream of the air preheater is passed through a "cold
side" ESP to remove particulates.  After a small amount of ammonia is added,
the gas enters a reactor, with a residence time of 1-20 seconds, at 373 K
(100°C) where it is bombarded with an electron beam at the rate of
105-106 rad/sec.
                                    4-157

-------
                                                                                        FUEL
                                                                        60°C
                                                                                                     *- TO STACK
t-«
CJi
CO
                                                             ABSORBER
                       FLUE GAS
                      FROM BOILER
                                                   150°C
                                                 O3 or CIO2
                                                                                       MaOH or
                                                                                       Na2CO3
                                                                                               Na2SO4 to
                                                                                              BY-PRODUCT
                                                                                              TREATMENT
                            Figure  4.4-3.   Generalized  flow diagram for wet  N0/S0« processes. 144
                                                                                   A   £

-------
                             NH3
                    FLY ASH
                                       ELECTRON BEAM
                                       ACCELERATOR
                                         REACTOR
                                                    OFF-GAS
 SOLID
RESIDUE
                                                              BY-PRODUCT
                                                              TREATMENT
                                                             DISPOSABLE OR
                                                           SALABLE BY-PRODUCT
Figure 4.4-4.   Process flow  diagram for  the Ebara-JAERI  electron  beam process. 145
                                          4-159

-------
     The key subsystem of this process is the electron beam accelerator.
Control of this unit's power supply is based upon inlet composition, flow
rate, and temperature of the flue gas.  (The penetration of the gas stream
by the beam requires a unique discharge pattern and other special  design
considerations.)  A powder containing both ammonium nitrate and sulfate is
generated by an unknown reaction mechanism.  The gas is then passed through
a second ESP to remove the solid by-product.  The by- product treatment
system is still being developed.  Various methods being investigated include
thermal decomposition in the presence of an inert gas, steam roasting with
CaO, or steam roasting with H90.  The byproduct may eventually be useful  as
             150
a fertilizer.
     The Ebara Manufacturing Company in conjunction with Japan Atomic Energy
                                                 o
Research Institute (JAERI) has operated a 1000 Nm /hr pilot plant treating
flue gas from an oil-fired boiler.  In 1976, Ebara began operating a 3000
Nm /hr pilot plant on the off-gas from an iron ore sintering furnace at
Nippon Steel.  This process is licensed in the U.S. by Avco- Everett
Research Laboratory.  The U.S. Department of Energy (DOE) is funding
development of an electron beam process offered by Research- Cottrell.
Pilot unit tests with flue gas are scheduled, however, the details of the
program are not yet available.
4.5  PRE-COMBUSTION TECHNIQUES FOR PM, NO. AND S07 CONTROL
                                         A        £
     Pre-combustion techniques considered for reducing PM, NO , and S09
                                                             /\        t.
emissions from industrial boilers include the use of naturally occurring
clean fuels, physically or chemically-cleaned fuels, and synthetic
(coal-derived liquid or gaseous) fuels.  A technique for reducing
particulate emissions from oil-fired industrial boilers, involving use of an
oil/water emulsion, is also considered as a pre-combustion emission control
technique.
     Naturally-occurring clean fuels discussed in this section are raw low
sulfur coal and raw low sulfur oil which are low enough in sulfur content to
meet S02 emission limits with no additional controls.  The fuel  cleaning
processes discussed in this section are physical coal cleaning (PCC) and
hydrodesulfurization (HDS) of oil.  These processes are primarily designed
                                    4-1CO

-------
to control SCL emissions by reducing the sulfur content of the fuel.
However, they may also aid in the control  of particulate emissions by
simultaneously reducing the ash content of the fuel.   Oil cleaning may
result in reduced NO  emissions due to reduction of fuel nitrogen content by
                    A
hydrotreating.
     The synthetic fuels discussed are low-Btu gas (LBG) and solvent  refined
coal (SRC).  LBG is derived from the gasification of  coal and may be  burned
in a gas-fired boiler as an alternative to a coal-fired boiler with
conventional emission controls.  SRC is either a solid or a liquid boiler
fuel derived from noncatalytic coal  liquefaction processes which produce a
fuel substantially reduced in ash and sulfur and potentially low in
fuel-bound nitrogen.  These fuels may replace coal and residual  oil use in
some industrial boilers.
     The water/oil emulsion technique involves preparing an oil  fuel  with a
sufficient amount of water to increase the fuel atomization.  Unburned
carbon particulate emissions are reduced as a result  of the improved
combustion conditions which result.
     The applicability of each of these pre-combustion PM, S09 and NO
                                                             £-       A
emission control techniques to industrial  boilers, the design or operating
factors which influence their pollutant reduction performance, and the
mechanism by which they reduce emissions is discussed in the following
subsections.  No performance data are presented for any of the
pre-combustion emission control techniques because their performance  is
either obvious (in the case of naturally-occurring clean fuels)  or has yet
to be proven (e.g., synfuels or oil/water emulsions).
4.5.1  Naturally-Occurring Clean Fuels
     The naturally occurring clean fuels of interest  are low sulfur coal and
low sulfur fuel oil.  Low sulfur coal is defined as run-of-mine (ROM) coal
which can comply with a given emission standard.  Where no emission standard
has been delineated, coals with sulfur contents of less than 1 percent by
                                 146
weight are considered low sulfur.
     The sulfur content of United States coals is quite variable.  While
46 percent of the U.S. total reserve base can be identified as low sulfur
                                   4-161

-------
coal because its sulfur content is less than 1  percent,  21  percent  ranges
between 1  percent and 3 percent in sulfur, and  an additional  21  percent
contains more than 3 percent sulfur.  The sulfur content of 12  percent of
the coal reserve base is unknown, largely because many coal  beds have not
been mined.
     Nearly 85 percent of the reserve base of less than  1  percent sulfur
coal is located in states west of the Mississippi River.  The bulk  of the
western coals are, however, of a lower rank than the eastern coals.  On a
heat content basis, it is estimated that at least 20 percent of the nation's
reserve of low sulfur coal is in the east.
     Low sulfur western coals can be burned in  underfeed and traveling grate
stokers as long as they are designed with sufficient control of undergrate
air to handle any caking that may occur.  Caking causes  an uneven ash layer
to form on the grate which reduces combustion efficiency unless undergrate
air can be distributed properly.  It has been reported that current designs
of some spreader stokers cannot handle caking coals because they lack the
                                               147
ability to control undergrate air distribution.     Since design changes to
incorporate the necessary air distribution system have not been
demonstrated, the use of those low sulfur coals which cake or have  a low ash
fusion temperature is not applicable to these stokers.  Other low sulfur
coals such as eastern bituminous, which do not  cake or have a low ash fusion
temperature, can be burned in underfeed and traveling grate stokers.  The
demonstrated reserve base of low sulfur eastern bituminous coal  as  of
                                                         148
January 1, 1974, was greater than 24 billion metric tons.
     Some spreader stokers of current design also cannot handle coals with
ash fusion temperatures below 1477 K (2200°F),  which are typical for many
low sulfur western coals (e.g., the Wyoming subbituminous, Utah bituminous
                 .149
and the lignites.)1
     Pulverized coal boilers can be designed for almost any type of coal.
The initial choice of coal will determine the type of pulverizer used, the
tube spacing in the boiler and superheater (low ash fusion temperature coals
require greater spacing), and the type of materials used in the furnace
  n
wall.
                                    4-162

-------
     In 1976 domestic refinery capacity for producing fuel  oil  from low
sulfur crude was 231,000 m3/day (1,452,000 bbl/day),  with the difference
made up by imports.  In contrast to low sulfur coal,  low sulfur fuel  oil
derived from naturally-occurring low sulfur crude is  readily applicable to
                                                             151
all boiler types and sizes that burn a similar grade  of fuel.
     There are no factors affecting the applicability of naturally-
occurring low sulfur coal or oil to reduce S02 emissions, except the  actual
sulfur content of the fuel.  However, the higher resistivity of the fly ash
from the combustion of low sulfur coal will  affect the design of an ESP
relative to that for medium to high sulfur coal.  The effect of resistivity
on ESP performance is discussed in Subsection 4.1.1.
4.5.2  Physical  Coal Cleaning
     Physical coal cleaning is the generic name for all  processes which
remove inorganic impurities from coal, without altering the chemical  nature
of the coal.  Basically, a coal cleaning plant is a continuum of
                                                 152
technologies rather than one distinct technology.     Each  coal  cleaning
plant is a uniquely-tailored combination of different unit  operations
determined by the specific coal characteristics and by the  commercially
dictated processing objectives.
     Overall process design philosophy in coal cleaning plants  is to  use
step-wise separations and beneficiations, with a goal  of eventually treating
small, precise fractions of the feed with the more sophisticated and
specific unit operations.  In this way, the least costly technologies are
applied to large throughputs and the more costly to much smaller through-
puts.  A characteristic of this design philosophy is  that multiple product
streams evolve,  each with its own set of size and purity properties.   In
conventional cleaning plants the separate product streams are blended prior
to shipment, to produce a composite coal meeting the  consumer's specifica-
tions.  Within the context of supplying industrial boilers  with small
quantities of relatively low-sulfur fuel, opportunities exist for premium
low-sulfur coals to be segregated from the final blending operation and
                               153
targeted for specialty markets.
                                    4-163

-------
     4.5.2.1  Process Description
     4.5.2.1.1  System.  In a modern PCC plant coal  is typically subjected
to: size reduction and screening, separation of coal from its impurities,
and dewatering and drying.  Commercial PCC methods are currently limited to
separation of the impurities based on differences in the specific gravity  of
coal constituents (gravity separation) and on the differences in surface
properties of the coal and its mineral matter (froth flotation).154  A
generalized physical coal cleaning schematic is shown in Figure 4.5-1.
     Five general levels of coal cleaning are used to categorize the degree
of treatment to which a coal has been subjected.  These levels are:
          Level 1 — Crushing and sizing
          Level 2 -- Coarse size coal beneficiation
          Level 3 -- Coarse and medium size coal beneficiation
          Level 4 — Coarse, medium, and fine size coal beneficiation
          Level 5 -- "Deep cleaning" coal beneficiation
     Level  1 processes are generally used to size raw coal to user specifi-
cations, and to remove overburden.  No washing is done and the entire
process is dry.
     Levels 2 and 3, in addition to crushing and screening raw coal also
perform a minimum of cleaning.  Level 2 provides for removal of only coarse
pyritic sulfur.  Level 3 is basically an extension of Level 2 in that both
the coarse and medium size fractions obtained from screening are washed
                                                        155
whereas in Level 2 only the coarse fractions are washed.
     Level 4 systems provide high efficiency cleaning of both coarse and
medium coal fractions with lower efficiency cleaning of the fines.  The
primary difference between Level 4 and the lower cleaning levels is the use
of heavy media processes for cleaning specific size fractions above 28 mesh.
For particles smaller than 28 mesh, cleaning by froth flotation is most
commonly used.  Level 4 systems accomplish free pyrite rejection and
improvement of heat content.
     Level 5 coal preparation systems are unique in that two products are
produced, a high quality, low sulfur, low ash coal called "deep cleaned"
coal and a middlings product with higher sulfur and ash content.  Level 5
                                     4-164

-------
tn
                       LOWEST
                      CLEANING

                    RAW COAL
                        I
                     CRUSHING
                     SCREENING
           PRODUCT
REFUSE
                       INTERMEDIATE
                         CLEANING

                         RAW COAL
                                                          CRUSHING
                                                         SCREENING
                                               PRODUCT
                                                                       SINK ,   SINK
 "DEEP"
CLEANING

RAW COAL
                                                                                              CRUSHING
                                                                                             SCREENING
                                                                                               -K"
                                                                                     GRAVITY
                                                                                    SEPARATION
                                                                                                           +J&"
                                                              FLOAT FLOAT
                                                                            GRAVITY
                                                                          SEPARATION
                                                                                     SINK
                                                                     REFUSE I   FINE
                                                                              REFUSE
                                                              PRODUCT
                         COARSE
                         REFUSE
                            Figure 4.5-1.  Physical coal  cleaning  unit operations employed to  achieve
                                            various levels of cleaning.

-------
provides the most advanced state-of-the-art in physical coal cleaning with
large reductions in pyrite and ash content and improvement of heat content
at high yields.  In addition, this system is flexible relative to the types
of coal that can be processed.  Variations in raw coal and product
specifications can be handled by varying the heavy medium densities and
careful control of coal sizes treated in various circuits.
     Level 5 coal cleaning plants use the techniques and principles utilized
in the first four levels, but combine them in unique ways to maximize mass
and energy recovery.  Major operations involved are crushing, screening or
sizing, heavy media separation, secondary separation, dewatering and removal
of fines from process water.  The high efficiency of Level 5 is due to the
repeated use of these operations to produce the desired products.
     4.5.2.1.2  Development status.  There are currently over 460 physical
coal cleaning plants in the U.S.  In 1976 about 340 million tons of raw coal
was processed by these plants.  This represents 58 percent of the total 1976
U.S. coal production of 590 million tons.  The majority of these plants were
designed for ash removal rather than sulfur removal although many do take
out 20-30 percent of the sulfur in the raw coal.  The status of coal
cleaning plants operated in 1976 is summarized in Table 4.5-1.  Some plants
use only one major cleaning process, while the majority use a series of
cleaning processes.  The capacity of individual plants varies widely from
less than 200 metric tons per day to more than 25,000 metric tons per
 .   159
day.
     Levels 1 through 4 are currently in use in operating commercial plants
which produce steam coal.  There are examples of Level 5 systems at
metallurgical coal plants where both a low sulfur, low ash metallurgical
grade product and a middling  (higher sulfur and ash content) combustion
grade by-product are produced.  All unit operations proposed for a Level 5
plant are presently used in commercial plants.  However, the unit operations
have not yet been combined to form a commercial Level 5 plant for producing
steam coal.
     4.5.2.1.3  Applicability to industrial boilers.  Firing of physically
cleaned coal in industrial stoker-fired boilers is not expected to have a
                                    4-166

-------
         TABLE 4.5-1.   PHYSICAL COAL CLEANING  PLANTS  CATEGORIZED  BY STATES  FOR 1976.158
Estimated
Total
Coal Production
State 1000 tons
Alabama
Arkansas
Colorado
Illinois
Indiana
Kansas
Kentucky
Maryland
Missouri
New Mexico
Ohio
Oklahoma
Pennsylvania
(Anthracite)
Pennsylvania
(Bituminous)
Tennessee
Utah
Virginia
Washington
West Virginia
Wyoming
Total
21,425
670
8,160
59,251
24,922
568
146,900
2,792
5,035
9,242
44,582
2^770
5,090

81,950

9,295
6,600
36,500
3,700
110,000
23,595
603,055
Nunber
of
Coal-
, Cleaning
Plants
22
1
2
33
7
2
70
1
2
1
10
2
24

66

5
6
42
2
152
1
459
Total
Nurber of Daily
Plants Capacity
for Which of
Capacity Deporting
Data Plants,
Reported Tons
10
0
0
20
6
2
40
0
1
1
13
1
14

50

4
4
29
1
113
1
318
40,600
-
-
136,775
42,000
3,800
245,700
-
3,500
6,000
102,750
550
13,000

205,010

8,520
23,100
143,550
20,000
577,375
600
1,652,030
Es Li ma bed
Annual
Cnpaci ty
.of
Reporting
Plants, (a)
1000 tons
10,150
-
-
34,195
10,500
950
61,425
-
875
1,500
25,690
140
3,250

71,255

2,130
5,775
35,090
5,000
144,345
150
413,210
Nviiber of Plants Using Various
Cleaning Methods
Heavy
Media
0
1
2
17
2
-
43
-
-
1
6
1
21

30

1
2
26
1
104
-
266
Jigs
10
-
-
20
5
2
27
-
2
-
11
1
4

19

1
4
15
1
55
-
177
Flotation Air
Unl ta Tables
6
-
1
4
1
-
16
-
-
1
-
-
4

16

1
2
9
-
59
-
121
1
-
-
1
-
-
4
-
-
-
1
-
-

20

2
2
8
-
12
1
52
Washing
Tables
12
1
-
1
1
-
20
-
- •
-
2
-
3

15

-
-
15
-
55
-
125
t
Cyclones
6
1
-
8
3
-
24
1
-
1
5
1
2

19

I
2
11
-
59
-
144
(a)  The estimated annual-capacity values for the reporting plants were calculated  fran the claily-cnpacity values by assuming an
    average plant operation of 250 days per year (5 days per week for 50 weeks per year).

-------
significant effect on boiler maintenance requirements.   In industrial
pulverized coal-fired boilers, firing of physically cleaned coal  may reduce
boiler maintenance costs.
     Physical cleaning of coal should improve the overall  performance  of a
stoker-fired boiler provided the resultant coal size is acceptable for
stoker firing (1-1/2" x 1/4" with minimal fines).  Physical cleaning
partially removes pyrites, ash, and other impurities, thus reducing both SOp
and particulate emissions.  As compared to raw coal, physically cleaned coal
is easier to handle and feed, burns more uniformly with less chance for
                                              i fi?
clinkering, and reduces ash disposal problems.     As an example, both a raw
and the corresponding physically cleaned coal were fired in a steam plant
spreader stoker boiler.  When firing the raw coal, the boiler could operate
only at about one half capacity.  The high ash content of this coal resulted
in nonuniform combustion caused by feeding problems, excessive ash buildup
and clinker formation on the fuel bed.  In contrast, the physically cleaned
                                                             1 cp
coal was fired at full capacity with no operational problems.
     4.5.2.2  Factors Affecting Performance.  Sulfur reduction by physical
cleaning varies depending upon the distribution of sulfur forms in the coal.
There are three general forms of sulfur found in coal; organic, pyritic, and
sulfate sulfur.  Sulfate sulfur is present in the smallest amount
(0.1 percent by weight or less).  The sulfate sulfur is usually water
soluble, originating from in-situ pyrite oxidation, and can be removed by
washing the coal.  Mineral sulfur occurs in either of the two dimorphous
forms of iron disulfide (Fe$2) - pyrite or marcasite.  The two minerals have
the same chemical composition, but have different crystalline forms.
Pyritic sulfur occurs as individual particles (0.1 micron to 25 cm. in
diameter) distributed through the coal matrix.  Pyrite is a dense mineral
(4.5 g/cc) compared with bituminous coal (1.3 g/cc) and is quite water-
insoluble thus the best physical means of removal is by specific gravity
separation.  The organic sulfur is chemically bonded to the organic carbon
of the coal and cannot be removed unless the chemical bonds are broken.  The
amount of organic sulfur present defines the lowest limit to which a coal
can be cleaned with respect to sulfur removal by physical  methods.  Physical
                                   4-168

-------
cleaning typically can remove about 50 percent of the pyritic sulfur,
although the actual removal  depends on the washability of the coal,  the unit
                                                            I CO
processes employed and the density of the separating medium.
     A trade-off between product yield and purity exists for any one unit
operation of a physical coal  cleaning process.  Product yield is defined as
the ratio of the clean product heating value divided by the heating  value of
the raw coal and can vary from 0 to 1.  Product purity refers to the amount
of sulfur retained in the clean product - the lower the sulfur content, the
higher the purity.  One unit operation cannot achieve both performance goals
— either yield is maximized, or purity is maximized, or a compromise is
made between yield and purity.  This basic limitation on performance also
applies to an entire plant if that plant only produces one clean coal
product.  However, the designer of a multiproduct plant may achieve  both
performance goals.  As an example, one unit operation may be selected for
maximizing product purity although the quantity of this clean product is
relatively small.   In this case, a fine fraction (28 x 0 mesh) may be
produced with a pyritic sulfur content reduced by up to 90 percent,  but with
a yield of less than 50 percent.  If the rejected portions are washed again
at a relatively high specific gravity in another (sequential) unit
operation, a "middling" product with somewhat higher pyritic sulfur  content
may be recovered with an overall recovery (between the two products) of the
                                       164
majority of the original heating value.
     The inherent design advantages of a multi-product plant do have special
significance for industrial  boilers.  Since the coal quantities used by
industrial boilers are a small fraction of the total coal demand, it might
be quite attractive for a coal cleaning plant to produce a very clean
product for new industrial boilers and a middling product suitable either
for consumers subject to less stringent emission standards or for large
consumers (i.e., utilities)  with additional site-specific S02 controls.
4.5.3  Oil Cleaning
     Hydrotreating or hydrodesulfurization (HDS) processes are used  to
produce oil fuels substantially reduced in sulfur, nitrogen and ash  content.
They are chemical  processes, which involve contact of the oil with a
                                    4-169

-------
catalyst and hydrogen to convert much of the chemically-bonded sulfur and
nitrogen to gsseous hydrogen sulfide (HgS) and ammonia (NH-).   These gases
are separated from the fuel and then collected.
     4.5.3.1  Process Description
     4.5.3.1.1  System.  In a typical hydrotreating process, oil  to be
treated is filtered to remove suspended material.  The oil is then mixed
with hydrogen, heated to 340 to 450°C (650° to 850°F), and passed over one
or more catalytic reaction beds.  The most widely-used catalysts  are
composites made up of cobalt oxide, molybdenum oxide, and alumina, where
alumina is the support and the other agents are promoters.
     Numerous chemical reactions occur which lead to removal of most of the
sulfur as H9S.  Table 4.5-2 illustrates some of the types of compounds and
                   1 fi7
reactions involved.     In an HDS process, hydrogen also reacts with other
species besides sulfur compounds.  For example, nitrogen compounds break
down to liberate ammonia from the oil.  This is referred to as denitrogena-
tion or denitrification.  Nickel and vanadium in the oil, which are bound as
organo-metal compounds, are also liberated by reaction with hydrogen.  This
is generally referred to as demetallization.  Most of the liberated metals
deposit (as the sulfide) on the catalyst surface or in its pores  and slowly
deactivate the catalyst.  Other reactions which take place break  up large
complex molecules such as asphaltenes and lead to a reduction in  carbon
residue for the product oil.
     Many companies are engaged in developing and using catalytic hydro-
treating or hydrodesulfurization processes.  All are similar in basic
concept but vary in specifics such as the type of catalyst employed, the
process conditions, and the process  complexity.  Figure 4.5-2 represents a
simplified flow diagram of an HDS process currently being commercially
marketed.  Its basic elements are a  feed filter, a heater, a single- stage
catalytic reactor, a gas/liquid separator, a fractionating column, and a gas
treatment section.  This system is capable of producing fuel oil  of approxi-
mately 1 percent sulfur from a feedstock such as atmospheric residual oil
containing 2 percent sulfur.  To produce a lower sulfur content product,
additional catalytic reaction stages must be added.  A system with two
                                    4-170

-------
          Table 4.5-2.   CHEMISTRY  OF HYDRODESULFURIZATION REACTION'S  IN
                                       PETROLEUM CRUDE OIL 167
Name
Thiophenes
Structure
Typical reaction
Thiols (mercaptans)
Disulfides
Sulfides
ROI1
RC C R'
R S R'

R.
R_

-------
              HYDROGEN
                           r
                                                 RECYCLED HYDROGEN GAS
           REDUCED
           CRUDE
                    FILTER
ro
 A
                                                         COMPRESSOR
1
1
** —
J
1
^
n
u
r<
u
ft
^1 >




i
HEATER
                                                   | CATALYST
                                                     VESSEL
                                                          FRACTIONATING COLUMN

                                                                     OFF GAS
A
0
GAS
Xr~~^
h«77j7l GAS/LIQUID SEPARATOR



1 UNREACTED HYDROGEN PLUS H2S


1
1
1
1
1
1
LIQUID |
1
1
1
1
|
1
1
1

RECYCLE GAS
TREATMENT











1
	
	
	
T





^
NAPHTHA ^
GAS OIL ^
(OPTIONAL) f
\
FLOW SULFUR
FUEL /
\
>» Af*in f2AC (H Ql

|fc 1 lfI14T U P

                                                Figure  4.5-2.  Basic HDS process.
                                                                                    168

-------
catalytic reaction stages can produce a fuel  of approximately 0.3 percent
sulfur content from a 2 percent sulfur feedstock.   A more advanced process
using three catalytic reactors can produce fuel oils with sulfur contents as
low as 0.1 percent.
     4.5.3.1.2  Development status.   Over 30  hydrotreating processes are
actively in use, and more than 250 processes  have  been described in the
patent literature since 1970.     Many of these processes have been in
commercial existence for over 10 years.  The  particular process selected by
a refinery depends on the existing or planned refinery products.  In
existing facilities, a fuel desulfurization process is usually chosen to
minimize modification or retrofit and/or satisfy refinery product mix goals
and feedstock purchase expectations.   Hence,  the desulfurization process
selected depends on the required sulfur content of the product and the
feedstock properties.
     4.5.3.1.3  Applicability to industrial boilers.  Like low sulfur fuel
oil produced from naturally-occurring low sulfur crude, oil  that has been
treated by an HDS process is readily  applicable to all boiler types and
sizes that burn a similar grade of fuel.  Use of this cleaned oil  should not
adversely affect the operation of the boiler.  In  fact, boiler performance
may even be improved due to the potential for less corrosion and deposit
formation in the boiler due to the chemical composition changes in the oil
as a result of hydrotreating.
     4.5.3.2  Factors Affecting Performance.   The  composition of the feed-
stock to a hydrotreater strongly influences the amount of hydrogen and
catalyst consumption in the process.   Major feedstock variables are density
(expressed as °API), sulfur content,  and metals content.
     Hydrogen consumption has been correlated with sulfur reduction for a
variety of residual oil feeds.  Figure 4.5-3  illustrates these results on
feedstocks varying from 4 - 18° API gravity.   It can be seen that to obtain
90 percent reduction in sulfur for a  19° API  feedstock, about 0.1 Mm  of
hydrogen are consumed per liter of oil processed (650 scf/barrel); whereas,
a 4° API feed would require 0.2 Nm3/liter (1200 scf/barrel).173
                                    4-173

-------
   1,200,-
   1,100
   1,000
  CD
  CO
    900
  O
  c/3
    800
  t-
  o.
    7on
  O
  O

  £u 600
  O
  O
  CC
  Q

  I 500
    400
    300
    200
          12
          14
            16
       30
40
50       60       70       80

    SULFUR REDUCTION %
90
100
        NOTE: 1. REDUCE BY 9% FOR FIXED-BED PROCESSES.
              2. APPLY CORRECTION FOR HIGH-METALS FEEDS.


Figure 4.5-3.  Hydrogen consumption in desulfurization of residual oil.172
                                   4-1/4

-------
     As previously discussed, removal  of metals by hydrotreating results in
metals deposition on the catalyst surface or in the pores.   This leads to
deactivation of the catalyst, which can be overcome by a temperature or
pressure increase to maintain acceptable processing rates.   The increase in
required severity of process conditions leads to more hydrocracking with a
                                            174
subsequent increase in hydrogen consumption.     Further complication from
the metals content of the feed is a shortening of catalyst  life.  Even
though some deactivation can be tolerated, the resultant increase in
hydrogen consumption means the catalyst must be changed out more frequently.
     The effect of metals is shown in  Figure 4.5-4.  This figure shows that
for 90 percent sulfur removal from a 25 ppm metals content  feedstock, about
27 barrels of oil can be processed per pound of catalyst; to achieve the
same sulfur removal with a 100 ppm metals content feedstock, only 4.5
barrels can be procesed per pound of catalyst; a feedstock  containing
300 ppm metals requires almost 1 pound of catalyst per barrel.   Clearly,
high metal feedstocks are a problem to the refiner.  Therefore, many
refiners are using a separate stage of lower cost catalyst  material prior to
the special hydrodesulfurization catalysts.   These separate stages may be
packed with a material such as alumina or clay, which collects  the metals
and "guards" the subsequent high activity catalyst.  For this reason, some
refiners refer to this stage as a "guard reactor" or "guard vessel".
4.5.4  Low-Btu Gasification
     Converting coal into a "clean" low-Btu gas with subsequent combustion
in a boiler, reduces S09, NO   and particulate emissions (versus direct coal
                       £    X
combustion) by removing the pollutant's precursors.  With respect to
particulate emissions, the coal ash content is physically separated from the
gas when coal is gasified.  Any entrained ash or coal particles are
subsequently removed from the gas in hot cyclones and in the gas quenching
and cooling steps.  S02 emissions are reduced by removing sulfur species
such as H2S and COS from low-Btu gas prior to combustion.  Nitrogen oxide
emissions are reduced because low-Btu gas contains only small quantities of
nitrogen compounds (NH~ and cyanides)  which are oxidized to NOV (fuel N is
                      O                                       X
                                    4-175

-------
30
40
 50       60       70        80
PERCENTAGE OF SULFUR REMOVED
10U
Figure 4.5-4.   Effect of metals content on  catalyst consumption.*75
                            4-176

-------
low compared to coal).   Moreover,  low-Btu gas burns  with a low flame
temperature which helps reduce the formation of NO  by thermal  fixation.
                                                  /\
     4.5.4.1  Process Description
     4.5.4.1.1   System.  As shown  by Figure 4.5-5 a  low-Btu gasification
system consists of three basic process steps:  coal  pretreatment,  coal
gasification, and product gas purification.  Coal pretreatment is  necessary
to supply uniformly size coal to the gasifier.   In the coal gasification
step, pretreated coal is reacted with a steam/air mixture to produce a
low-Btu gas with a heating value of apprxoiamtely 5.6  MJ/m  (150 Btu/scf).
In the gas purification step, particulate matter, sulfur, and nitrogen
compounds may be moved from the product gas.  If not removed, the  sulfur  and
nitrogen compounds would be oxidized to S02 and NOX  in the boiler  and the
particulate matter would erode the burner.
     Close to 70 different low and medium Btu gasifier types have  been  used
commercially in the past or are currently under development.  Among the
important characteristics which distinguish one gasifier from another are:
       •  Bed type,
       •  Operating conditions,
       •  Gasification media,
       •  Coal  feeding technique,
       •  Ash removal process,
       •  Energy input, and
       •  Type of gas produced.
     To produce a clean fuel, the  critical  parts of  a  coal gasification
system are the gas purification and acid gas removal (A6R) operations.
Removal of coal dust, ash, and tar aerosols entrained  in the raw product  gas
leaving the gasifier is accomplished with cyclones,  ESP's, and water or oil
scrubbers.  In the gas quenching and cooling section,  tars and oils are
condensed and particulates and other impurities, such  as ammonia and
cyanides, are scrubbed from the raw product gas.
     Acid gases such as h^S, COS,  CS^, mercaptans, and S0« are also removed
with varying effectiveness from the raw product gas  in the gas quenching and
cooling section.  Either low sulfur coal or AGR systems must be used to
                                    4-177

-------
             -COALPRETREATMENT '
                                     • COAL GASIFICATION'
  COAL
              DRYING
           PARTIAL OXIDATION
              CRUSHING
               SIZING
             PULVERIZING
              STORAGE
             CONVEYING
                                                            • GAS PURIFICATION -
                                                                                  POLLUTION CONTROL OPERATIONS




LOW/MED-
Btu GAS
PARTICULATE
REMOVAL
QUENCHING/COOLING
ACID GAS REMOVAL





PARTICULATE CONTROL
SULFUR CONTROL
HYDROCARBON CONTROL
NITROGEN OXIDES CONTROL
OIL/WATER SEPARATION
SUSPENDED SOLIDS REMOVAL
DISSOLVED ORGANICS REMOVAL
DISSOLVED INORGANICS REMOVAL
. CHEMICAL FIXATION
SLUDGE REDUCTION
LANDFILL
                             AIR
Figure 4.5-5.  Low-Btu coal gasification system process  and pollution control  modulesJ78

-------
avoid excessive sulfur emissions.   Commercially available  AGR techniques
include physical  and chemical  solvent processes, direct conversion  and
                                                                   178
catalytic conversion processes,  and fixed-bed  adsorption processes.     The
specific process used depends  on the major acid gas  constituents.
     4.5.4.1.2  Development status.  LBG from  coal has  been  produced both  in
the United States and overseas for many years.   It is estimated  that at one
time there were some 11,000 coal  gasifiers in  use in the U.S.  As the
availability of natural  gas increased, the number of operating gasification
systems declined significantly.   At the present time there are only a few
                                                                    179
coal gasifiers operating in the  United States  on a commercial  basis,    and
all of these are used to fuel  process furnaces.   Some of these furnaces have
a heat transfer medium to transfer the energy  from the  combustion operation
to the process, and, hence, they are similar in design  to  boilers.   They
also operate at combustion temperatures typical  of industrial  boilers, which
indicates that low-Btu gas can be burned in industrial  boilers.  None of
these systems incorporate an AGR process for gas cleanup.  However,  AGR
processes are comrnercially available from a number of process  licensors and
vendors.
     4.5.4.1.3  Applicability  to industrial boilers. Low-Btu  gasification
systems are applicable to any  size industrial  boiler.   For an  8.8 MW
(30 x 10  Btu/hr) industrial boiler, one 3 m (10 ft) diameter Wellman-
Galusha gasifier is required.   For larger boilers, multiple  gasifiers would
be used (ten 3 m diameter gasifiers are required for a  boiler with  a thermal
input of 117.2 MW or 400 x 106 Btu/hr).  All  of the  low-Btu  gasification
systems examined in the individual technology  assessment report  for
synthetic fuels are sources of gaseous emissions, liquid discharges, and
solid wastes.  However, with suitable precautions there do not appear to  be
any uncontrollable adverse environmental impacts associated  with the
production and use of low-Btu  gas.
     The use of coal-derived low-Btu gas in new industrial gas-fired boilers
has several advantages over the use of coal in direct  coal-fired boilers.
First, a gas-fired boiler is a much simpler piece of equipment to  operate
than a coal-fired boiler.  There is no need for ash  handling equipment  and
                                    4-179

-------
the only fuel handling equipment required is piping.  However, ash handling
equipment will be required at the gasifier location.  Second, due to the
less complex nature of gas-fired boilers, maintenance requirements will  be
                                 •ion
less than for coal-fired boilers.
     On the other hand, the production and use of coal-derived gases at  an
industrial site can have adverse impacts.  The primary concerns are the
reliability/operability of the gasification system and how that affects  the
operability of the boiler.  In order to minimize adverse impacts, installa-
tion of spare capacity or sparing of key process units in the gasification
system may be required.  Another alternative would be to provide a backup
fuel source (such as distillate fuel oil) for the boiler.  The incorporation
of either of these options into a gasification/steam generation system
design must be done on a case by case basis, taking into consideration the
particular requirements of the system.  In addition, in the selection and
design of the boiler, consideration must be given to the different
combustion characteristics (e.g., heat release rate and flame temperatures)
                                       180
of coal-derived gas versus natural gas.
     4.5.4.2  Factors Affecting Performance.  The performance of LBG as  an
emission control technique for industrial boilers depends on the performance
of the gas purification system operation.  More specifically, it depends on
the performance of the acid gas removal unit in removing H£$ and organic
sulfur compounds (predominantly COS) from the product gas.  The demonstrated
acid gas removal processes are capable of removing over 90 percent of the
                                  181
sulfur species from the raw gases.
     S0? emissions can be predicted accurately from the producer gas
analysis, which is available from several gasifier/acid gas removal system
combinations.  An upper limit on particulate emissions can also be
predicted, based on the particulate content of the cleaned gas and
experience with gas-fired boilers.  Particulate emissions are estimated  to
approach those for natural gas-fired boilers.  NO  emission data for
                                                 A
specific coal and gasifier types which are necessary to accurately predict
NO  emission levels from industrial boilers firing low-Btu gas are not
  /\
available.
                                    4-180

-------
4.5.5  Solvent Refined Coal
     The Solvent Refined Coal  (SRC) process is a fuel  pretreatment process
designed to produce clean solid (SRC-I) and liquid (SRC-II)  fuels.  Both the
SRC-I and SRC-II processes use a noncatalytic hydrogenation  step in which
coal is partially dissolved in a hydrogen-rich solvent to produce a fuel
substantially reduced in sulfur and ash content compared to  the raw coal.
Fuel nitrogen content may also be reduced.
     4.5.5.1  Process Description
     4.5.5.1.1  System.  The two SRC processes are shown schematically in
Figure 4.5-6.  In the SRC-I  process, slurried coal is  liquified and the
product is separated from the unreacted residue by filtration.    Recycled
solvent for coal-slurry preparation is recovered from  the product mixture by
distillation.  The rest of the liquid product is solidified  to  produce a
boiler fuel.  In the SRC-II  process, more hydrogen (almost double the amount
required in the SRC-I process) is added to the coal  in the liquefaction
reactor.  The unreacted solids are separated from the  product by vacuum
distillation., and a fraction of the liquid product is  recycled  for slurry
preparation.  The product liquids may be hydroprocessed for  further
upgrading, depending on the product quality desired.
     4.5.5.1.2  Development status.  Systems to produce SRC  fuels are in
advanced stages of development and could be commercially available in the
late 1980's.  Both solid and liquid boiler fuels are currently  being
                                                        183
produced in DOE sponsored SRC-I and SRC-II  pilot plants.     Table 4.5-3
presents a comparison of the status of development of  the two SRC processes.
     4.5.5.1.3  Applicability to industrial boilers.   Preliminary results
from SRC-I handling and burning tests indicate that some industrial boiler
modifications may be required for the operation of the fuel  handling and
storage equipment, pulverizers, burners, and the combustion  process if SRC
               185
fuels are used.     In pulverized coal-fired boilers,  for example,
pulverizer temperatures must be lowered to prevent the SRC-I fuel from
melting during pulverization.   Pulverizer temperatures can be lowered by
reducing the amount of air that the pulverizer receives from the air
          •I QC
preheater.     In addition,  the solid fuel  produced by the SRC-I process is
                                    4-181

-------
SOLVENT REFINED COAL-I PROCESS
                                   RECYCLED GAS
                                    COMPRESSOR
                      COAL
                      COAL
                     SLURRY

DON

Y
ION
LURR
FEED
PUMP
\ 	
Y
REC
Rl
YCLEHjV_^
CHGAS
f


REACTOR
(DISSOLVED)

LET
AND
SYi
V

FIRED
PR EH EATER

SOLID
SEPA
(Fl


                          SOLVENT RECYCLE
          SOLVENT REFINED COAL-II PROCESS
VACUUM
L AT ION
8

HYDROGEN
PRODUCTION

MINERA
                                                                      STEAM
                                                                      OXYGEN
                                                    HYDROGEN
                                                              MATTER
                                                ATM
                                               DISTIL-
                                               LATION
                                                                        LIQUID
                                                                      ' PRODUCTS
Figure 4.5-6.   Flow diagram of the  SRC-I  and  SRC-II  liquefaction processes.
                                                                                         182
                                           4-182

-------
Table  4.5-3.    GENERAL COMPARISON AND RELATIVE  TECHNICAL
                  STATUS OF  THE  SRfe-I AND SRC-11
                  LIQUEFACTION PROCESSES 184
      Technical Status
                                   SRC-I
                                                    SRC-II
            Pilot unit
       Scale of Operations
    (metric tons/day,  coal)

    Size of Pilot Plant
    (metric tons/day,  coal)

    Pilot Plant began
    Operat ion

    Fuel Types
    Coal Feed


    General Comparisons

    Fuel Flexibility
    Reactor Operating Severity

    Process Scale-up Risk

    Number of New Components
    and Design of Commercial
    Equipment

    Reactor Complexity

    Fuel Utilization,
    Combustion

      Raw Product, Stability,
      Compatibility

      Combustion Experience
0.9
a) 45
b) 5.5

a) Late 1974
b) Mid-1976

Refined coal
(solid fuel)

Eastern
Western
Moderate
Moderate
Moderate
0.9


27.3


Mid-1977


Distillate oil
Eastern (high
pyrites only)
Developed to
produce
substitute
solid boiler
fuel only
Moderate
Moderate
High
Poor


High
High
High
Moderate
Poor
Limited
                              4-183

-------
not applicable to all industrial coal-fired stoker boilers.   SRC-I  solids
have a low melting point (approximately 615 K or 310°F) and  would melt on
                                                                        187
the grate of current fixed-bed stoker boilers before they are combusted.
     Fuels from the SRC-II process may be used to replace residual  fuel  oil
as an industrial boiler fuel with minor modificatons in the  combustion
        188
process.
     4.5.5.2  Factors Affecting Performance.  The primary operating
variables which could affect the conversion of sulfur and nitrogen in the
raw coal to hydrogen sulfide and ammonia for removal from the fuel  are:
       •  Characteristics of the raw coal,
       •  Process operating variables,
       •  Hydrogen consumption,
       •  Reactor space velocity, temperature, and pressure, and
       •  The degree of hydroprocessing of the raw product fuel.
     The higher the sulfur and nitrogen content of the coal  processed, the
greater the hydrogen consumption will be.  Nitrogen removal  is more
difficult than sulfur removal because the reaction between nitrogen and
hydrogen does not take place easily.  Nitrogen removal from the feed coal
ranges from approximately zero to 40 percent with about half of the nitrogen
                                             189
removed going into the production of ammonia.
     Reactor space velocity, temperature, and pressure all affect hydrogen
consumption.  An increase in reactor temperature and pressure results in
increased reaction rates with an increase in hydrogen consumption.   An
increase in residence time  (decrease in space velocity) also increases the
consumption of hydrogen.  Since hydrogen consumption is influenced by all
these variables, it is used as an indicator of the sulfur and nitrogen
      , 190
removal.
     Reactor temperature and residence time have greater effects on sulfur
                              190
removal than reactor pressure.     At low temperatures, the relationship
between sulfur removal and hydrogen consumption is approximately linear.
However, as the temperature is increased, more sulfur is removed but at a
lower rate.  Sulfur removal can also be increased by increasing reactor
pressure and residence time.
                                    4-184

-------
     An increase in reactor temperature also increases hydrogen consumption
to react with nitrogen; however, nitrogen removal at higher temperatures
                              190
does not significantly change.     It appears that the effect of reactor
pressure changes on hydrogen consumption for nitrogen removal is small, on
the order of 0.1 percent nitrogen content change for a change in pressure
from 8.86 to 10.49 MPa (1280 to 1520 psi).
     Raw coal-derived liquid fuels differ from petroleum-derived fuels in
that they are very aromatic and, as such, are hydrogen deficient.   Hydro-
processing can be used to increase the hydrogen content of coal-derived
fuels by catalytically reacting hydrogen with the fuel.  Hydrogenation also
further decreases the sulfur and nitrogen content of the coal-derived liquid
fuel produced.  The hydroprocessing variables that affect sulfur and
nitrogen removal from coal-derived liquids are catalyst type and hydrogen
consumption.
     There is limited storage and combustion data on coal liquefaction
products from either of the two SRC processes.   The Ft. Lewis,  Washington,
SRI-I pilot plant with a capacity of 45 metric tons per day, produced a 2725
metric ton sample of solid SRC-I fuel  for combustion testing at the 22 MW
Plant Mitchell power station of the Georgia Power Company.  The combustion
                                                   192
tests were performed in the second quarter of 1977.     Small-scale tests on
home heating units and industrial  boilers, and some limited laboratory tests
                                             1 no
have been performed with SRC-II liquid fuels.
4.5.6  Oil/Water Emulsions
     Oil/water emulsions can be fired in distillate and residual  oil-fired
boilers to enhance the atomization of the fuel  and obtain improved
combustion.  As a result of improved combustion, the firing of  an oil/water
emulsion in an industrial  boiler can result in decreased particulate
emissions, and, in some cases, decreased NO  emissions.
                                           A
     4.5.6.1  Principle of Operation.   The oil/water emulsion process is
based upon the firing of a stable emulsion in a conventional oil-fired
boiler.  Emulsion preparation equipment commercially available  uses
ultrasonics or mechanical  means to produce stable emulsions. Surfactants
are required to produce a  stable emulsion with distillate oil;  whereas,
                                   4-135

-------
residual oil, because it typically contains natural  surfactants,  will  form a
                                            1°3 194
stable emulsion without surfactant addition.   '
     When firing an oil/water emulsion, each fuel  droplet contains one or
more small droplets of water.  During combustion  the internal  water droplets
vaporize, causing mini-explosions of the fuel droplets, leading to a much
finer atomization and a very thorough mixing of air  and fuel.   This allows
complete combustion with much less excess air and results in a dramatic
reduction in soot production.  Use of less excess air means that  less heat
is carried out the stack by the exhaust gases and the reduction of soot
formation keeps the boiler heat transfer surfaces clean.  Thus, boiler
efficiency is improved.   '
     Improved combustion conditions result in less unburned carbon being
emitted from the boiler with a resulting decrease in particulate  emissions.
Some tests have, however, shown that particulate emissions from firing an
emulsion may have a smaller size distribution resulting in increased visible
                                               197
emissions although the mass emissions decrease.     In addition to
reductions in particulate emissions, use of oil/water emulsion technology
has been reported to lower NO  emissions for distillate oil firing due to
                             A
the lower excess air used and the reduced combustion temperatures which
        1 98
result.     However, no significant reduction was observed for residual oil
emulsions due to the high nitrogen content of the residual oil fuel
        1Q9
tested.Iyy
     4.5.6.1.2  Development status.  Oil/water emulsifiers have been
marketed commercially in the United States and Europe since the early
1970's.     Emulsifiers sold to date have been primarily used for the
purpose of increasing boiler efficiencies through improved combustion.
Environmental benefits have, apparently, not been a major factor in sales of
emulsification systems, and consequently, actual  performance data concerning
emission reductions achieved by this technology are not available.
     4.5.6.1.3  Applicability to industrial boilers.  Oil/water emulsion
systems are generally applicable to industrial boilers burning either
distillate oil or residual oil.  Applications to date have been retrofits
for the primary purpose of improving the .combustion efficiency of older
                                    4-186

-------
boilers.  Improved combustion conditions typical  of new boilers, with
improved burner designs and instrumentation, will result in essentially the
same benefits that emulsion firing has been shown to provide.   Consequently,
this technology will probably continue to have as its major application the
improvement of the performance of existing installations.
4.6  COAL/ALKALI COMBUSTION TECHNIQUES FOR S02 CONTROL
     Both combustion of coal/alkali fuel mixtures or coal  in a bed of
alkaline sorbent are being developed as alternatives to post-combustion S02
control.  Two of the most promising alternatives  are combustion of coal/
limestone fuel mixtures and combustion of coal in a fluidized bed of
limestone.  With these combustion techniques, fuel sulfur is converted to
S0« which reacts with calcium oxide and excess oxygen in the fuel bed
according to the following overall reaction.

               S02 (g) + CaO (s) + * 02 (g)   ->   CaS04 (s)

The CaO is produced by a rapid calcining of calcium carbonate (limestone) in
the fuel bed using the heat of combustion.  Most  of the calcium sulfate
formed stays in the fuel bed and is removed from  the system along with the
bottom ash.  Some CaSO. may become entrained in the flue gas and
subsequently be collected in a downstream particulate control  device.
     This section describes two methods that may  be used to burn coal/alkali
fuel mixtures for S0« control.  Fluidized bed combustion (FBC) is discussed
first followed by coal/limestone pellet (CLP) combustion.   It should be
noted that CLP technology is still in the developmental stage and any
information presented should be considered as preliminary and subject to
change.  FBC technology has been used on a limited basis with its use
expected to increase in the future.
     Development of another process involving combustion of coal/limestone
fuel mixtures is currently being funded by EPA, with future plans for a
joint EPA/DOE development program being considered.  In this process, a
pulverized mixture of coal and limestone is fired in a low-NO  burner to
                                                             X
reduce S09 and NO  emissions (relative to the combustion of coal in a
         £       A
                                    4-187

-------
conventional burner.)  However, development of this process had not
progressed beyond bench scale at the time of this report.
4.6.1  Fluidized Bed Combustion for S00 and NO  Control
                                      £       X
     Fluidized bed combustion (FBC) is a boiler design option which,  because
of the nature of its operation, results in lower SCL and NO  emissions.
                                                   £       A
Because of its S09 and NO  emission reduction potential, FBC is discussed
                 ^       /\
here as a pollution control technique rather than as a boiler type in
Chapter 3.  FBC technology offers a variety of advantages over conventional
boiler designs, including S0? emission reduction without use of FGD systems,
smaller more compact boilers, and flexibility in fuel use.
     Although both pressurized and atmospheric fluidized-bed designs  are
currently being developed, it appears that atmospheric fluidized bed  combus-
tion (AFBC) will dominate the FBC market for industrial boiler applications
in the near future.  Apparently, the additional complexity of the
pressurized designs  (and associated cost) is not offset by increased
performance in industrial boiler applications.  Pressurized designs may,
however, prove to be economical in utility, co-generation, and combined
                   ?m
cycle power plants.      In the following discussion, only AFBC designs will
be considered.
     4.6.1.1  Process Description
     4.6.1.1.1  System.  A simplified schematic diagram of an AFBC boiler is
presented in Figure  4.6-1.  The unit is comprised of a bed of sorbent (or
inert material) which is suspended or "fluidized" by a stream of air at
0.3  to 4.6 m/sec (1  to 15 ft/sec).
     Coal is injected into this bed and burned.  A  sorbent (usually lime-
stone or dolomite) is also injected to react with the S02 formed upon
combustion.  The gas velocity  is set so that the bed particles are suspended
and  move about in random motion.  Boiler tubes submerged  in the bed remove
heat at a high rate  to maintain bed temperatures in the range of 760° to
                         onq
870°C (1400° to 1600°F).
     Particulate matter  emitted from the boiler passes to a primary cyclone
where 80 to 90 percent of the  larger carbon containing particles are
removed.  This collected material can be recirculated back to the FBC unit,
                                     4-188

-------
CONVECTION
  SECTION
  HEAT
TRANSFER
 BAFFLE
  TUBES
   FUEL
   FEED
   SORBENT
    FEED
     AIR-
                                         PRIMARY
                                         CYCLONE
    FLUIDIZED BED
>~~~xxxxxx~
      PLENUM
                                  FREEBOARD
                                   1
                               CARBON
                               RECYCLE
                      HEAT TRANSFER
                       WATER TUBES
                                    SPENT STONE WITHDRAWAL

                                    AIR DISTRIBUTION
                                     SUPPORT GRID
   Figure 4.6-1.   Typical  industrial  FBC boiler.202
                         4-189

-------
fed to a carbon burnup cell (CBC) to maximize combustion efficiency, or
disposed of.  A carbon burnup cell is a separate FBC reactor which is
operated at higher temperatures [1090°C (2000°F)] than the main FBC to
                                   203
achieve maximum carbon utilization.
     In addition to the cyclones normally incorporated in an atmospheric FBC
design, additional particulate control equipment such as "hot-side" or
"cold-side" ESPs or fabric filters (see Section 4.1) may be used to further
reduce particulate emissions.
     FBC technology can reduce S02 emissions by up to 90 percent or more
depending upon the rate of sorbent addition to the bed and the FBC design
and operating conditions.  Nitrogen oxide (NO ) emissions from FBC are
                                             A
inherently lower than uncontrolled emissions from conventional combustion.
Combustion temperature is considerably lower in FBC (815° to 930°C [1500° to
1700°F]) than conventional combustion (1500°C [2700°F]).  The lower FBC
combustion temperature results in lower NO  emissions due to reduced
                                          A
fixation of atmospheric nitrogen.  Formation of NO  at the lower
                                                                204
temperatures is primarily due to the oxidation of fuel nitrogen.

                    2N (fuel) + 02   -»   2NO

The NO is formed rapidly as the coal burns and is thought to be reduced in
the presence of carbon monoxide and other products of incomplete combustion,
                                    204
by a reaction such as the following:

                    2CO + 2NO  *  2C02 + N2

     Some combustor design and operating conditions tend to increase NO
                                                                       A
emissions.  For example, increasing bed temperature, increasing excess air,
decreasing gas residence time, and possibly increasing fuel nitrogen content
can all contribute to increased NO  emissions.  However, the influence of
                                  A
these variables on NO  emissions has not been quantified or correlated; and
                     A
the mechanisms of NO  formation and decomposition in FBC are not well
                    A
                                   4-190

-------
understood.  Hence, it is not possible to design FBC's for low NO  emissions
                                           205
with the same reliability possible for S02.
     4.6.1.1.2  Developmental status.   Development of coal-fired FBC
industrial boilers is continuing on several  fronts.   Much of the work is
being conducted with funding and guidance from the U.S.  Department of Energy
(DOE) as part of the National Energy Research, Development,  and
Demonstration Program.  Recently the State of Ohio has supported FBC
development.  In addition, several vendors now offer commercial  FBC
industrial boilers independent of government funding.   Finally,  industrial
boiler FBC development is being supported through utility FBC development
work:  The Electric Power Research Institute (EPRI)  is actively involved in
this program.
     DOE lists four major demonstration FBC  boilers  currently operating or
                   ?0fi
under construction.     The installation at  Georgetown University has been
operating since mid-1979.  This unit is a two-bed design with a total
capacity of 44,840 kg/hr (100,000 Ib/hr) of  saturated steam.   Another major
demonstration project is the 22,420 kg/hr (50,000 Ib/hr) boiler at the Great
Lakes Training Center in Illinois.  This unit is currently scheduled for
start-up in early 1981.  The remaining DOE demonstration projects are part
of an investigation into the use of anthracite culm (mine tailings) in
industrial boilers.  A 8,970 kg/hr (20,000 Ib/hr) boiler is  under construc-
tion in Paxinos, Pennsylvania and will supply a paper reprocessing plant.
Finally, a larger 44,840 kg/hr (100,000 Ib/hr) boiler is planned to supply
the City of Wilkes-Barre, Pennsylvania with  steam for heating and air
conditioning.
     At least three manufacturers now offer FBC industrial  boilers on a
commercial basis.  Units as large as 50,000  Ib/hr are available as package
boiler units.  The largest of the industrial FBC boiler manufacturers
reports 14 sales of coal-fired industrial boilers, 10 of which will burn
pure coal with the remaining four burning mixtures of coal  and other
fueU.207
     Despite the availability of commercial  units, FBC is still  an emerging
technology.  Long term data on the performance of FBC units is lacking.
                                    4-191

-------
Future work is currently being directed toward the confirmation of long-term
S0x removal efficiency in large scale units.  Documentation of the influence
of gas phase residence time and sorbent particle size on SCL removal  are
other major areas of research.  Other investigations are required to  assess
limestone characteristics and availability as well as alternative
sorbents.
     Experimental work is continuing in an effort to gain a better under-
standing of NO  formation/reduction mechanisms in FBC, and of the
              A
correlation between emissions and the key FBC design/operating conditions
which can influence emissions.  The goal of these studies is to provide the
capability to better predict and control NO  emissions through adjustment of
                                           A
standard design/operating conditions.  Also, several investigators are
beginning to address combustion modifications, deliberately aimed at
reducing NO  emissions from FBC, such as staged combustion, flue gas
           A
recirculation, ammonia/urea injection, and stacked beds.  It is necessary to
define the effects of such combustion modification techniques, not only on
NO  emissions, but on other system parameters, such as combustion efficiency
  A
and materials corrosion and the potential increase of S09 particulate
   .   .    209
emissions.
     4.6.1.1.3  Applicability to industrial boilers.  Fluidized bed combus-
tion can be used in place of practically any type of industrial boiler
(stoker, pulverized coal, gas/oil).  FBCs can be used for saturated/
unsaturated steam, process heating (water, air, crude oil), and direct/
indirect heating applications.
     In the industrial boiler capacity size range of less than 73.3 MW
(250 x 106 Btu/hr) it is expected that most, if not all, FBC units will
operate at atmospheric pressure with a once-through sorbent processing
scheme.  Most industrial FBC boiler users probably will not have sufficient
need for onsite electric power generation to justify the additional capital
and operating costs and operational complexity associated with pressurized
FBC systems.  A similar argument of economics, operational complexity, and
technological demonstration holds for sorbent regeneration systems.  It is
expected that the typical industrial user will select a once-through  sorbent
                                    4-192

-------
operating scheme, due to its demonstrated-simplicity and lower costs, at
                                             210
least for first generation FBC installations.
     The concensus of opinion indicates that widespread application of
coal-fired FBC industrial boilers will  be limited to systems greater than 15
to 30 MW (50 to 100 x 10  Btu/hr) thermal input.  This is due primarily to
the high cost of related coal and ash handling equipment for smaller units.
However, there does not appear to be any lower capacity technical  limit to
                                211
coal firing with FBC technology.
     Fuel flexibility is an important advantage of FBC use in the  industrial
sector due to the incentive to burn industrial byproducts and low-grade,
high sulfur fuels not easily burned in  conventional  boilers.  FBC  boilers
have multifuel capability and can burn  all  ranges of coal, oil, and gas and
some industrial wastes.
     FBC industrial boilers produce higher amounts of solid waste, relative
to conventional cumbustion, since spent sorbent as well as ash must be
disposed of.  It is possible that waste disposal requirements for  FBC may
limit its use in areas with severe solid disposal limitations.  For most
installations, solid disposal will not  be a major factor influencing the use
of FBC boilers.
     4.6.1.2  Factors Affecting Performance.
     SOp Control.  Of the factors which affect S0« emission control, the
calcium to sulfur molar feed ratio (Ca/S) has the greatest impact.  As the
calcium content of the bed is increased, greater S0« removal is achieved.
Westinghouse Research and Development Center has developed a model which
projects sorbent requirements to attain certain levels of S02 removal
efficiency.  Figure 4.6-2 illustrates the rapid increase in sulfur retention
with increasing Ca/S based on the model.  For sorbents with a particle size
of approximately 500 ym, the relationship is nearly linear below about
75 percent S02 removal.  Above this level, sulfur retention approaches
100 percent asymptotically.  However, further data from larger systems and
for high levels of S09 removal are required to fully support the model
            213
projections.
                                   4-193

-------
 100
  90
  80
O
I-
o
  70
2 60
I 50
LU
DC
  40
O
DC
£30
  20
  10 —
   0
                             GREER | BUSSENl
WESTERN    $  /   /

         ./  •'   '
                                 •GROVE •'
                                 A
                                / MENLO
      LIMESTONE
        TYPES
              ..* OPERATING CONDITIONS

                PRESSURE = 101.3 KPc (Ictm)

                TEMPERATURE = 841°C (1546°C)
               AVERAGE SORBENT= 500
               PARTICLE SIZE

               BED DEPTH = 1.2m (4ft.)
               SUPERFICIAL  = 1.8 m/sec (6 ft/sec)
               GAS VELOCITY                   —

               PRIMARY SORBENT/FLY ASH RECYCLE
                           3       4
                       Ca/S MOLAR RATIO
Fiaure 4  6-2.  Projected desulfurization performance of FBC
        '      based upon a model  developed by Westinghouse.212
                           4-194

-------
     Sorbent particle size is also an important factor influencing SC^
reduction.  As the particle size of a given sorbent is decreased, the
calcium utilization is increased.  Thus, with the same Ca/S molar feed
ratio, the S02 reduction efficiency can be increased significantly by
decreasing the sorbent particle size.  The increased reactivity of smaller
                                                       214
sorbent particles is due to their greater surface area.
     A third major factor which affects the sulfur removal efficiency of the
system is the time the gas phase remains in the bed and is defined as the
ratio of the expanded bed height to the superficial gas velocity.
Figure 4.6-3 illustrates the calculated relationship between gas phase
residence time and Ca/S molar feed ratio required to achieve 90 percent SCL
removal, at various particle sizes for two types of limestone.   As gas
residence time is increased, the required calcium to sulfur molar feed ratio
decreases.  Figure 4.6-3 also indicates that there is a critical gas
residence time (0.6 to 0.7 sec) below which sulfur retention efficiency is
                 ?l fi
severely reduced.
     These three control factors are interrelated and can be varied to
obtain the optimum S0« removal efficiency.  A trade-off must be made among
                                              ?l fi
these factors in designing the optimum system.     There are, however,
factors other than these which affect emission control.  Overbed feeding is
technically simpler than underbed feeding, but solid and gas residence time
may be less than desirable.  SO,, released above the bed could be captured
with reduced efficiency and sorbent may be entrained in the flue gas before
                         217
it has a chance to react.
     The temperature within the bed may have a direct effect on the
efficiency of the reaction between sulfur dioxide and calcium oxide.
Several investigators have shown that optimum temperatures for calcium use
are between 760° and 870°C (1400° and 1600°F), depending upon the coal and
sorbent used.  Figure 4.6-4 shows pilot scale results comparing sulfur
retention versus temperature for two coals.  The lower temperature limit is
determined by the temperature at which calcination occurs; that is, CaC03
releases C02, forming CaO, the reactive form of the sorbent.  Below 760°C
(1400°F) calcination is not complete.  The lower sulfur retention observed
                                    4-195

-------
i
i—«
vo
                                                                         CARBON LIMESTONE
                                                                         PARTICLE SIZE
                                                                        GROVE LIMESTONE
                                                                        PARTICLE SIZE
                                                 0.8     1.0      1.2     1.4
                                               GAS RESIDENCE TIME, sec
                 Figure 4.6-3.  Ca/S molar feed required to maintain 90 percent sulfur removal  in  AFBC,
                               as projected by the  Westinghouse Model.215

-------
        90
        80
      S 70
      o
      0>
       : 60
        50

      LU
      DC
      tr 40
      ID
30



20
        10 —
         1300
                           TEMPERATURE, °C

                     760          8" 3          871
                         /
o/
 /
6
                     GAS VELOCITY, 3 ft/sec
                     EX ESS OXYGEN, 3%
                     LIMESTONE NO.  1359
                   O  ILLINOIS COAL, ca/S 2.5

                   •  PITTSBURGH COAL, Ca/S 4.0
             1400         1500        1600
                   TEMPERATURE, °F
                                                  927
                                   1700
Figure 4.6-4.  S02 reduction as a  function of bed temperature
                              4-197

-------
at temperatures greater than the optimum temperature may be caused by the
release of SO,, after capture due to local reducing conditions in the bed, or
by slight changes  in other variables.219
     NQx Control.  Design and operating factors which influence the
formation and control of NO  in AFBC boilers include:
                           A
       •  Temperature
       •  Excess air
       •  Gas residence time
       •  Fuel nitrogen
       •  Coal particle size
       •  Factors  affecting local reducing conditions
     The kinetics  and mechanisms of NO  reduction in AFBC boilers are not
                                      A
well understood.   Research to date indicates over 98 percent of NO
                                                                  A
emissions are NO and, furthermore, over 90 percent of the NO emitted is
derived from fuel  nitrogen.  Surprisingly, however, NO  emissions appear to
                                                      A
be relatively independent of fuel nitrogen content.  It is thought that NO
is formed near the base of the bed and is then reduced to elemental nitrogen
as the gases rise  through the bed.  Many of the factors above affect this
reduction.  In summary, AFBC boilers emit considerably less NO  than
                                                              A
conventional boilers because of the lower combustion temperatures.  However,
the further reduction of emissions by combustion modifications will probably
                                    220
need to await further investigation.
     Particulate Matter (PM) Control.  For the most part, factors affecting
the generation of  PM emissions and the performance of control devices are
similar to those affecting conventional boilers.  FBC boilers can use fabric
filters, ESP's and cyclones for PM control .  Cyclones are commonly used for
recycling elutriated bed material back to the boiler or to a separate carbon
burnup cell.
     4.6.1.3  Emission Test Data.  Nearly all the available emission data
for AFBC industrial boilers was obtained from tests run on small pilot plant
or demonstration projects.  Because these units are primarily research and
development facilities, this test data may not be characteristic of full-
scale industrial  size units.   In addition, the majority of the data has been
                                    4-198

-------
obtained using sampling and analytical  techniques other than EPA reference
methods.  Rigorous testing with established EPA reference test methods is
usually done to determine whether a boiler is complying with specific
emission standards.  So far, the need for this type of testing has been
limited.
     The situation, insofar as the availability of standard test data
obtained from full scale, continuously operating units, is changing rapidly.
Several large scale commercial units are scheduled for start up within the
next year and will likely require compliance tests.
     SOg Emission Data Summary.  Figure 4.6-5 is a summary of SCL data
obtained at eight AFBC test facilities under a wide variety of test
conditions.  The bounded area is an indication of the range of performance
expected from FBC systems at high gas-phase residence times and small
sorbent particle size.  Much of the experimental data falls within these
boundaries.  Deviations from the band are noted in the data from the B&W
3 ft x 3 ft unit and the PER-FMB unit.   If the units and test conditions are
considered closely these deviations from the band are expected.  The B&W 3
ft x 3 ft unit has a shallow bed which allows less than optimum sorbent/gas
contact.  Gas phase residence times are approximately one- third of the
recommended 0.67 sec.  The PER-FBM data were also obtained using low
                                                           222
gas-phase residence times in the range of 0.13 to 0.26 sec.
     A continuous emission monitoring program for S0? was conducted at the
Georgetown University 45,400 kg (100,000 Ib) steam/hr coal/limestone feed
fluidized-bed boiler.  Because this system was still in an extended
shakedown phase, several key operating conditions (e.g., level of excess
air, percent fly ash recycle) were not operating in the intended design
range.  On a daily average basis, desulfurization was greater than
75 percent on all 14 days of record, greater than 85 percent on 9 days, and
greater than 90 percent on 5 days.  Throughout the duration of the testing
program boiler load varied between 50 to 60 percent.  Coal feed properties
                                                               234
ranged from 10 to 15 percent ash and 1.2 to 2.5 percent sulfur.     The
complete S02 data set is presented in Appendix C.
                                   4-199

-------
   100
   90
   80
   70
2  60
UJ
er

-------
     NO  Emission Data Summary.  The composite diagram of NO  emission data
       X                                                    X
measured over the range of normal FBC operating conditions is shown in
Figure 4.6-6.  In the temperature range of interest (800° to 900°C), most of
the data points are below 215 ng/J (0.5 lb/10  Btu).  However, about
10 percent of the test results in the temperature range of interest show N0v
                                   6
emissions above 300 ng/J (0.7 lb/10  Btu).  All of these higher values are
                                                 224
from the Argonne 6 in. diameter bench-scale unit.
     These data are reported from experimentation with units where there was
generally no intentional variation of design of operating conditions to
reduce NO  emissions.  Increased gas residence times to enhance S09 control
                                                                     225
may contribute to additional reduction of NO  emissions even further.
                                            A
     A continuous emission monitoring program for NO  was conducted at the
                                                    /\
Georgetown University 100,000 Ib steam/hr coal/limestone feed fluidized-bed
boiler.  NOX emissions ranged from 441 ng/J (1.0 lb/106 Btu) to 218 ng/J
(0.5 lb/106 Btu) and averaged 281 ng/J (0.7 lb/106 Btu) for a 16 day period.
Boiler load ranged from 48 to 61 percent capacity and percent oxygen values
ranged from 8.8 to 12.3 percent on a dry basis.  Fuel nitrogen was
                                       235
consistent at about 1.5 weight percent.     The complete test results  are
presented in Appendix C.
     PM Emissions Data Summary.  Available data concerning emissions from
primary cyclones, which are considered part of the FBC boiler process,
indicate that emissions at the cyclone outlet are in the range of 215  to
2150 ng/J (0.5 to 5.0 lb/10  Btu) with a mass mean particle size of 5  to
      ooc
20 ym.     These emission characteristics are comparable to those for
conventional mass feed spreader stokers.  Thus, it is expected that
controlled emissions from FBC boilers would be generally equivalent to those
stokers equipped with the same controls (i.e., ESPs or fabric filters).
     Table 4.6-1 summarizes the results from particulate emission tests
conducted at the Georgetown University 45,400 kg (100,000 Ib) steam/hr
coal/limestone feed fluidized-bed boiler.  The FBC unit is equipped with a
mechanical collector and baghouse for particulate control.  The percent ash
in the fuel was high throughout the duration of the testing program and
ranged between 10.6 and 15.0 percent on an as received basis.  Average
                                    4-201

-------
        I WO
      300
                   I4TO
BED  TEMPERATURE,1

  1*50        1830
                                2010
                                                              21*0

-------
                                TABLE 4.6-1.  RESULTS OF PARTICULATE EMISSION TESTING AT

                                              THE GEORGETOWN FBC UNIT.234a
PO
o
CO

Run number
Boiler load (Ib/hr)
C02, %
02, %
Excess air, %
Avg. stack temp., °F
Stack gas volume, dscfm
Isokinetic ratio, %
Total particulate
ng/J
lb/106 Btu
Average ng/J
lb/106 Btu

P-l
53,600
8.0
11.5
117.9
337.2
18,890
96.74

47.04
0.1094


August 23
P-2
52,000
8.0
11.5
117.9
337.1
19,245
95.87

37.79
0.0879
36.5
0.0848

P-3
51,000
8.0
11.5
117.9
336.2
18,296
98.58

24.58
0.0572



P-4
54,000
8.9
10.1
89.5
347.8
17,607
98.56

32.31
0.0751


September 13
P-5
47,000
8.0
10.6
97.3
349.4
18,121
98.40

20.92
0.0487
24.3
0.0565

P-6
50,000
8.0
10.6
97.3
348.8
18,177
98.36

19.62
0.0456



-------
emissions ranged from 36.5 ng/J (0.0848 lb/106 Btu) to 24.3 ng/J
(0.0565 lb/10  Btu) for the August and September tests, respectively.
During the August set of runs there was noticeable puffing from the stack at
regular intervals, indicating leakage in a compartment of the baghouse.
After replacing several bags the extent of the puffing was reduced but not
completely eliminated.  An inspection of the baghouse prior to conducting
the September runs revealed that extensive blinding of the teflon bags had
occurred.  After repairs were made, overall baghouse performance improved as
evidenced by the results presented in Table 4.6-1.  It should be noted that
throughout the testing program mechanical collector plugging occurred which
                                                    poc
may have affected the inlet loading to the baghouse.
4.6.2  Coal/Limestone Pellets
     4.6.2.1  Process Description
     4.6.2.1.1  System.  Coal/limestone pellet (CLP) technology is an S0«
removal technique currently being studied by the EPA.  In this process,
coal/limestone pellets are fired as ordinary fuel  in stoker boilers.  The
S02 formed during combustion reacts with limestone present in the fuel
pellets to form calcium sulfite and calcium sulfate salts.  The majority of
calcium salts remain  in the ash bed and are discharged from the boiler along
with the bottom ash.  This system does produce an  increase in boiler
particulate emissions which may affect the design  of fly ash control
          227
equipment.
     There are several processes available for the manufacture of CLP.
These processes include the pellet mill process, briquette production
process, auger extrusion process, and disk production process.  In all cases
the pellets are composed of coal, limestone and a  cement or organic binder.
     Pellet production studies have been conducted with the goal of pro-
                                         228
ducing a CLP suitable for  industrial use.     Ideally, these pellets would
have the mechanical strength, durability and weatherability characteristics
comparable to those of raw coal.  Table 4.6-2 compares the physical
properties of coal and coal/limestone pellets (Ca:S ratio of 3.5) produced
by two different processes using different binders.  The mill production
method creates a pellet with physical properties that exceed raw coal in all
                                     4-204

-------
               TABLE  4.6-2.   COMPARISON OF PHYSICAL PROPERTIES  OF RAW COAL AND FUEL PELLET229
Pellet Formulation





PO
o
Ul
Production
Method
Raw coal
Raw coal
Raw coal
Raw roal
CPH lab mill
Banner
extrusion

Coal
Type
Illinois 16
E. Kentucky
Lignite
Rosebud
Illinois §6
Illinois 16

Llmentone Durability
Z Type Z Dlnder Index 
100 — ~ — 85 ± 2
100 — ~ — 85 1 2
100 — — — 77+4
100 — ~ ~ 84 ± 2
70 Piqua 30 2Z All bond 4 1Z 87
Poly co 2136
70 Plqua 30 1.5Z Allbond 200 4 94
1Z H-167 .01

Comp renal on
Strength, Weather
Ib Index (b)
74 i 12 89 t 1
83 t 22 94 ± 1
92 i 22 80 t 4
50+15 79 t 2
112 100
84 100

Post Weathering
Durablllty(l')
Index
75
83
34
20
85
62

Strength.
Ib
58
94
45
68
>112
60

(a)  Water a-.'tled as needed.




(b)  Percent survival - 100 - percent fines.

-------
respects while the extrusion process produces a pellet with properties
comparable to raw coal.
     4.6.2.1.2  Developmental status.  The use of CLP as an S02 control
technique for industrial boilers is still in the developmental stage.   It
must be shown to be economically viable, applicable to industrial  boilers
and effective at removing S02 before commercialization can begin.
     As part of an EPA-funded program to evaluate this technology, studies
aimed at resolving these questions are currently underway.  These studies
include:
       •  Battelle Columbus Labs:  assess the technologies SO,, removal
          capabilities for various boiler types, fuel types and quality.
          Development of a suitable pelletizing process.

       •  Versar, Inc.:  address the impacts of this technology on boiler
          design.  Evaluation of Battelle pelletizing process.

       •  Charles River Associates and Versar:  address the cost to produce
          the pellets, marketing aspects and ability of potential  pellet
          vendors to supply the projected demand.

     At this time final results from the studies mentioned above are not
available.  However, some preliminary information has been supplied in the
form of pellet production and emission data studies conducted by Battelle.
These preliminary data indicate that Battelle-Columbus Laboratories has
developed a pellet (Ca:S 3.5:1) suitable for industrial use with sulfur
retention capability of about 50 percent.  However, problems with the
pelletizing process have impeded continued development.
     4.6.2.1.3  Applicability to Industrial Boilers.  Coal/limestone pellet
S02 control technology is applicable to any type of coal-fired stoker
boiler.  Preliminary data suggest that bed temperature, steam production
rate, fly ash loading, and bottom ash loading will be affected to some
                                                    229
degree when firing coal/limestone in stoker boilers.
                                   4-206

-------
     Based on preliminary data from a "demonstration test" conducted at the
Battelle Lab steam plant, it was found that CO levels from pellet firing
were relatively high (usually greater than 100 ppm) compared to those from
the firing of conventional stoker coals.  These higher CO levels may be
related to the nature of the burn and/or to the fact that the overfire air
flow rate was decreased during the pellet firings.  Because of the
compactness of the pellet and the limited access of air into the pellet, the
capture process first involves the formation of calcium sulfide via

                    2 CaO + FeS2   -»•   2CaS + FeO + CO

which can account for the part of this increase in CO.

     Particulate emissions at the outlet of the mechanical collector during
the "demonstration test" from the firing of the 3.5:1 Battelle pellet were
258 ng/J (0.6 lb/10  Btu).  The smoke opacity was only 20 percent which
would appear low for a particulate loading of 258 ng/J (0.6 lb/10  Btu) if
the fly ash were from coal firing alone.  Fly ash from pellet firing is
about 50 percent more dense and considerably more coarse than from firing
           228
coal alone.     For equivalent mass loadings, optical density varies
inversely with particle size and density.  Thus, the apparent discrepancy
between smoke opacity and particulate loading is explained partially by the
laws of optics.  A change in particle size distribution over that of raw
coal firing could affect the design of fly ash collection equipment.
However, additional data are needed (particle size distributions) in order
to quantify this effect.
     Boiler thermal efficiency may be affected by the addition of limestone
to the boiler feedstock.  Limestone present in the bed will absorb thermal
energy that would normally be used to produce steam.  Numerical estimates
for the potential efficiency reduction are not currently available.  In
addition the calcination of limestone is an endothenffic reaction which will
further reduce the thermal efficiency.
                                    4-207

-------
     For an existing coal-fired boiler using CLP technology,  the steam
production rate is expected to be reduced by 20 percent of the rated
         229
capacity.     This is due primarily to a decrease in the heating value
(Btu/lb fuel) of the coal/limestone mix relative to that of raw coal.   In
order to produce the energy equivalent of 100 kg (220 Ib) of coal,  163 kg
                                                       229
(359 Ib) of coal/limestone mix would have to be burned.     Therefore, it  is
unlikely that a boiler could achieve its rated steam production capacity
unless modifications to the fuel feeder mechanism prove successful  in
providing increased mass feed rates to the boiler.
     Flue gas volumetric flow rate is expected to increase by 8 percent over
                        229
that of raw coal firing.     This increase can be attributed to the CQ^
produced when calcining of calcium carbonate (limestone) occurs by  the
reaction:

                    CaC03' -»  CaO (sorbert) + C02

This increased flow is expected to affect the design and cost of new boilers
and could affect the performance of existing boilers and controls using CLP.
Wider tube spacing will be required on new boilers in order to maintain the
standard design velocity past the tubes.  In existing boilers using
coal/limestone pellets this increased velocity could affect the heat
transfer rate while simultaneously increasing corrosion of the tubes.
Increased I.D. fan horsepower will also be required in order to accommodate
the higher flue gas volume for both new and existing installations.
     Total ash loading on the boiler and controls can be expected to
increase with CLP use.  Initial estimates show that 3 to 4 times as much  ash
                                              228
may be generated over that of raw coal firing.     The impact resulting from
this could take the form of increased bottom ash capacity and/or an
increased number of bottom ash removal cycles, both which would affect
boiler capital and/or maintenance costs.
     The factors mentioned in this section could affect the applicability of
this technology to industrial boilers with respect to boiler design,
operation, maintenance and cost.
                                    4-208

-------
     4.6.2.2  Factors Affecting Performance.  An important factor affecting
the performance of coal/limestone pellets as an SC^ control technique is the
calcium/sulfur ratio.  A preliminary test using pellets with a 7:1  ratio
demonstrated between 70-80 percent sulfur removal.   Subsequent tests using
pellets with a 3.5:1 ratio show removal  efficiencies of from 45-67  percent.
The calcium/sulfur ratio also affects the physical  properties of the fuel
(strength, durability and weatherability) and thus, the handling
characteristics of the pellets.  When conducting tests on the 7:1 ratio
pellets, it was found that they lacked strength and broke under the stress
from the fuel feed system.  Although broken pellets may not affect  S02
removal, particulate emissions may be increased.
     A second factor affecting S02 removal  efficiency appears to be the
combustion or bed temperature.  In preliminary tests on a small spreader
stoker using pellet ratios of 7:1 and 3.5:1, removal efficiencies from
52-72 percent were reported (see emissions  data section).  Thermochemistry
suggests that sulfur capture is reduced  at  higher bed temperatures.
Additionally, an analytical model developed to serve as an interpretive tool
suggests that maximum sulfur capture occurs for a minimum pellet surface
                  228
area/volume ratio.     Thus, it is not surprising that the pillow-shaped
briquet with a relatively high surface area/volume  ratio has the lowest
sulfur capture of any of the production  techniques.
     4.6.2.3  Emissions Data.  A series  of  preliminary emissions tests have
been conducted using coal/limestone pellets developed by Battelle-  Columbus
Labs.  Initial model spreader tests used both the 3.5:1 and 7:1 Battelle
pellets as the feedstock to a 20 brake horse power model spreader stoker
boiler.  Subsequent demonstration and checkout tests were conducted on an
11,340 kg/hr (25,000 Ib steam/hr) spreader  stoker at Battelle Laboratories
using the 3.5:1 Battelle pellet only.  In all of the tests a high sulfur
coal was used (3-4 percent sulfur) to produce the pellets.  Because these
units are primarily research and development facilities, these test data may
not be characteristic of full-scale commercial units.
     Results from the model spreader stoker tests are summarized in
Table 4.6-3.  A preliminary test, using a raw coal  feedstock was run to
                                    4-209

-------
TABLE 4.6-3.  MODEL SPREADER STOKER EXPERIMENTS228

Fuel
Illinois No. 6
Cement-bound pellets
_^ Cement-bound pellets
I
£2 Methylcellulose-bound pellets
O
Ditto
Ditto
Ditto
Ditto
Ca/S
0
7
3-1/2

3-1/2

Ditto
Ditto
Ditto
Ditto
Production Technique
—
Pellet mill (cylinders)
Ditto

Ditto

Ditto
Briquettes
Auger extrusion (cylinders)
Disc (spheres)
Hens u red
S02.<«>
ppm
3700
1040
1220

1480

1260
1780
1370
Pellets did
Predicted
S02.
ppm
3700
3700
3700

3700

3700
3700
3700
not have adequate
Sulfur
Retention,
percent
0
72
67

60

67
52
63
strength

-------
TABLE  4.6-4.   EMISSION  DATA SUMMARY FOR  FUEL PELLET DEMONSTRATION
                                                                         228

Smoke CO at Fuel N Fuel S
Load, Oj. CO?. CO. NO, SO?. Opnclty, 3ZOj, NO at 3Z 0?, ppm Converted, SO? at 3Z 0?. ppm Emitted
ppli Z Z ppm ppm ppm Z ppm Computed Menmircd Z Computed Mensured X
20,000 8.4 10.5 300 310 1600 20 420 2250 440 20 4100 2250 55
i
rartlciilatcfl,
Ih/H Rtu
0.6
                      TABLE 4.6-5.   SULFUR  BALANCE
      Computed Fuel S In,
          lb/106  Btu
Emitted aa S02
  lb/106 Btu
Sulfur Retained In
 Bed Aali .as S02.
   lb/106 Btu
                                       4.1
                        3.3

-------
document uncontrolled emissions.  Sulfur retention values for the  model
spreader tests were calculated using this as the baseline emission level.
The 7:1 Battelle pellet achieved 72 percent sulfur retention while the  3.5:1
pellets demonstrated retentions of from 52-67 percent.   Variations in sulfur
capture for the 3.5:1 pellet were attributed to variations in bed  tempera-
tures, with sulfur capture tending to be reduced at higher bed temperatures.
Further testing of the 7:1 Battelle pellet was not conducted due to its  lack
of physical strength.
     A demonstration test was conducted using twenty tons of CLP with a  Ca:S
ratio of 3.5.  The pellets were fired in an 11,340 kg/hr (25,000 Ib/hr)
steam spreader-stoker boiler at the Battelle steamplant.  Two types of
pellets were used, a lower density (0.9 to 1.2 g/cc) pellet produced by
Banner Industries using auger extrusion and a higher density pellet
(1.4 g/cc) produced by Alley-Cassetty Coal Company using a pellet  mill.
Both types of the pellets were fired under a variety of boiler conditions.
During the demonstration tests, the pellet feed rate was maintained at
approximately 1.36 Mg/hr (1.5 tons/hr) at a boiler load of 80 percent.
Tables 4.6-4 and 4.6-5 summarize the results of this test.  As indicated in
Table 4.6-4, the sulfur capture was 45 percent during the demonstration
test.  The greater sulfur retention of the earlier model spreader  tests  was
attributed to the lower bed temperatures which were seldom higher  than
1260°C (2300°F).  The bed temperatures during the demonstration tests were
seldom less than 1371°C (2500°F) and as high as 1455°C (2650°F).
Additionally with a pulsating ash discharge stoker, the fuel bed is
violently disturbed.  Ash can therefore be recirculated back into  the hot
zone.  Thus, if sulfur is retained in the ash at a lower bed temperature, it
may be released when the ash is exposed to a higher temperature.
     The Battelle steam plant boiler facility uses a mechanical collector to
control particulate matter.  Depending on the ash and sulfur content of the
coal, earlier experiments had shown that particulate loadings varied between
86 and 258 ng/J (0.2 and 0.6 lb/106 Btu).  Generally, for low S, low ash
coals, particulate loadings were less than 129 ng/J (0.3 lb/10  Btu).   The
                                   4-212

-------
particulate loading during the  demonstration  test was  258  ng/J
(0.6 lb/106 Btu).
                                   4-213

-------
4.7  REFERENCES

1.   Roeck, D.R. and R. Dennis (GCA Corporation).  Technology Assessment
     Report for Industrial Boiler Applications:  Participate Collection.
     (Prepared for U.S. Environmental Protection Agency.)  Research  Triangle
     Park, N.C.  Publication No. EPA-600/7-79-178h.  December 1979.   p.  24.

2.   Reference 1, p. 26.

3.   Offen, G.R., et al.  (Acurex Corporation).  Control  of Particulate
     Matter From Oil Burners and Boilers.  (Prepared for U.S. Environmental
     Research Triangle, Park, N.C.  Publication No. EPA-450/3-76-005.
     April 1976.  p. 4-40.

4.   Reference 1, p. 27.

5..  White, H.J.  Electrostatic Precipitation of Fly Ash.  Journal  of Air
     Pollution Control  Association.  Vol. 27.  March 1977.  p. 20.

6.   Reference 1, p. 32.

7.   Reference 1. p. 29.

8.   Reference 1, p. 31.

9.   Reference 1, p. 34.

10.  Southern Research  Institute.  Study of Electrostatic Precipitators
     Installed on Oil-Fired Boilers.  (Prepared for Electric Power Research
     Institute.)  Palo  Alto, California.  EPRI Report FP-792.  p.  3-35.

11.  Reference 1, pp. 44,45.

12.  Reference 10, p. 3-21.

13.  Meterology Research, Inc.  Determination of the Fractional Efficiency,
     Opacity Characteristics, Engineering, and Economics of a Fabric Filter
     Operating on a Utility Boiler.  (Prepared for Electric Power Research
     Institute.)  Palo  Alto, California.  EPRI Report FP-297.  p.  3-6.

14.  Oglesby, S., et al.  A Manual of Electrostatic Precipitator Technology,
     Part I — Fundamentals.  Southern Research Institute.  Birmingham,
     Alabama.  NTIS PB  196 380.  August 1970.  p. 369.

15.  The Mcllvaine Company.  The Mcllvaine Electrostatic Precipitator
     Manual, Volume I.  Northbrook, Illinois, The Mcllvaine Company, 1976.
     Chapter II, p. 40.2.
                                    4-214

-------
16.  Telecon.   Offen,  G.,  Acurex Corporation,  with  L.  Coe,  Joy  Western
     Precipitation.   October 23 and November 13,  1979.
17.  Reference 14, pp. 152-158.
18.  Reference 14, p.  166.
19.  Reference 15, p.  40.4.
20.  Reference 15, p.  40.7.
21.  Reference 15, p.  40.9.
22.  Reference 1.  p.  49.
23.  Reference 1,  p.  48.
24.  Reference 1,  p.  51.
25.  Reference 1,  pp.  55-57.
26.  Letter from Orem, S.R., Industrial  Gas  Cleaning  Institute, to
     Mobley, J.C., EPA:PTB.   August 14,  1979.   Comments  on  particulate
     control report.
27.  Reference 1,  p.  64.
28.  Trip report.   Piccot, S.C., Radian  Corporation,  to  file.   July 7, 1981.
     Report of visit to DuPont fabric filter.
29.  Reference 1,  p.  58.
30.  Reference 13, p.  3-4.
31.  The Mcllvaine Company.   The Mcllvaine Fabric Filter Manual, Volume  I.
     Northbrook,  Illinois, The Mcllvaine Company, 1979.   Chapter III, p.
     119.0.
32.  Reference 31, p.  40.9.
33.  Reference 13, p.  xiv.
34.  Reference 1,  p.  65.
35.  Theodore, L.  and A.J. Buonicore.  Industrial Air Pollution Control
     Equipment for Particulates.  New Haven, Connecticut, Chemical  Rubber
     Company Press.   1976.  p. 203.
                                   4-215

-------
36.  Calvert  S., et al.   (A.P.T., Inc.).  Wet Scrubber System Study:
     Volume 1, Scrubber Handbook.  (Prepared tor U.S.  Environmental
     Protection Agency.)   Research Triangle Park, North Carolina.
     Publication No. EPA-R2-72-118a.   NTIS PB 213 016.   July 1972.   p.  3-8.

37.  Reference 1, p. 68-69.

38.  The Mcllvaine Company.  The Mcllvaine Scrubber Manual.   Volume  I.
     Northbrook, Illinois, The Mcllvaine Company, 1974.  Chapter III,
     p. 40.0.

39.  York Research Corporation.  Emission Test Program:  Sodium Carbonate
     Manufacturing Plant Conducted at FMC Corporation.   Draft Report.
     (Prepared for U.S. Environmental Protection Agency.)  Research  Triangle
     Park, N.C.  EPA Contract No. 68-02-2819.  August 1, 1979.

40.  Reference 1, p. 70.

41.  Reference 1, p. 74.

42.  Perry, R.H. and C.H. Chilton.  Chemical Engineers' Handbook,  Fifth
     Edition.  New York, McGraw-Hill  Book Company, 1973.  p. 18-83.

43.  Reference 1, p. 83.

44.  Reference 1, p. 86.

45.  Reference 1, p. 85.

46.  Stern, A.C.  Air Pollution:  Volume IV, Engineering Control of  Air
     Pollution, Third Edition.  New York, Academic Press, 1977.  p.  117.

47.  Memo from Kelly, M.E., Radian Corporation, to Industrial Boiler File.
     Review of ABMA/DOE/EPA study on particulate emissions from stoker-
     fired boilers and mechanical collectors.  April 1, 1981.

48.  Reference 1, p. 88.

49.  Meeting Notes.  Jennings, M.S., Radian Corporation.  Meeting  between
     representatives of U.S. EPA, Radian, and General  Motors Corporation to
     discuss the side stream separator.  Durham, North  Carolina.
     February 4, 1981.46.

50.  General Motors Corporation.  Report on the Side Stream Separator.
     (Prepared for U.S. Environmental Protection Agency and Ohio EPA.)  Re:
     Innovative Technology Order 45 Federal Register 10344 (February 15,
     1980).  General Motors Corporation.  Detroit, Michigan.  Undated.
                                    4-216

-------
51.  Meeting Notes.  Radian Corporation.  Meeting between representatives of
     Radian and EPA to discuss EPA's Office of Research and Development
     (ORD) position on Industrial Boiler NSPS.  Durham, North Carolina.
     November 10, 1980.

52.  U.S. Environmental Protection Agency.  Electric Utility Steam
     Generating Units.  Background Information for Proposed Particulate
     Matter Emission Standards.  Research Triangle Park, North Carolina.
     EPA 450/2-78-006a.  Emissions Standards and Engineering Division.
     July 1978.  p. 4-43.

53.  Memo from Sedman, C.B., EPA:ISB, to Durkee, K.R., EPArlSB.   Meeting
     with General Motors to discuss side stream separators.  December 22,
     1981.

54.  Memo from Sedman, C.B., EPArlSB, to Industrial  Boiler Files.   Perfor-
     mance of Emission Control Systems on Fine Particulates.  April  21,
     1981.

55.  Letter and Attachments from Chamberlin, B.A., Tennessee Eastman Company
     to Jennings, M., Radian Corporation.  November 23, 1981.  4p.   Informa-
     tion about soot blowing practices.

56.  Dickerman, J.C. and K.L.  Johnson (Radian Corporation).  Technology
     Assessment Report for Industrial Boiler Applications:  Flue Gas
     Desulfurization.  (Prepared for U.S. Environmental Protection Agency.)
     Research Triangle Park, North Carolina.  Publication
     No. EPA-600/7-79-178i.  November 1979.  p. 2-147.

57.  Reference 56, p. 2-153.

58.  Reference 56, p. 2-154.

59.  Reference 56, p. 2-159.

60.  Reference 56, pp. 2-159 and 2-160.

61.  Leivo, C.C. (Bechtel Corporation).  Flue Gas Desulfurization  Systems:
     Design and Operating Considerations.  Final Report, Volume I.
     Publication No. EPA 600/7-78-030a.  March 1978.  p. 3-19.

62.  Reference 56, p. 2-80.

63.  Reference 56, p. 2-81.

64.  Reference 56, p. 2-82.

65.  Reference 56, p. 2-84.
                                    4-217

-------
66.  Reference 56, p. 2-88.
67.  Tuttle, J., et al.  (PedCo Environmental, Inc.).   EPA Industrial  Boiler
     FGD Survey:  First Quarter 1979.  (Prepared for U.S. Environmental
     Protection Agency.)  Research Triangle Park, North Carolina.
     Publication No. EPA-600/7-79-067b.  April  1979.   p.  31-122.
68.  Reference 56, pp. 2-88 through 2-90.
69.  Reference 56, p. 2-90.
70.  Reference 56, pp. 2-90 through 2-91.
71.  Reference 56, p. 2-98.
72.  Reference 56, p. 2-97.
73.  Reference 56, p. 2-9.
74.  Ayer, F.A. (Research Triangle Institute).   Proceedings:   Symposium  on
     Flue Gas Desulfurization — Las Vegas, Nevada, March 1979;  Volume 1.
     (Prepared for U.S. Environmental Protection Agency.)  Research Triangle
     Park, North Carolina.  Publication No. EPA-600/7-79-167a.   July 1979.
     p. 342.
75.  Mobley, D.J., EPA/IERL.  Memo to Larry Jones, EPA/OAQPS.  Memo on the
     status of adipic acid enhanced FGD.  February 12, 1981.
76.  Reference 56, p. 2-15.
77.  Reference 56, p. 2-26.
78.  Reference 56, p. 2-16.
79.  Reference 75, p. 3.
80.  Reference 56, p. 2-72.
81.  Reference 56, p. 2-33.
82.  Reference 56, p. 2-34.
83.  Reference 56, pp. 2-34 through 2-43.
84.  Reference 56, p. 2-52.
85.  Reference 56, p. 2-55.
86.  Reference 56, p. 2-31.
                                   4-218

-------
87.  Blythe, G.M., et al.  (Radian Corporation).  Survey of Dry S0? Control
     Systems.  (Prepared for U.S. Environmental Protection Agency.)
     Research Triangle Park, North Carolina.  Publication
     No. EPA-600/7-80-030.  February 1980.  p. 9.

88.  Reference 87, p. 11.

89.  Reference 56, p. 2-162.

90.  Reference 56, p. 2-163.

91.  Kelly, M.E.  and S.A.  Shareef (Radian Corporation).  Third Survey of Dry
     S02 Control  Systems.   (Prepared for U.S.  Environmental  Protection
     Agency.)  Research Triangle Park, North Carolina.   Draft Report.  EPA
     Contract 68-02-3171.   February 18, 1981.   p. 15.

92.  Reference 87, p. 10.

93.  Reference 56, p. 2-166.

94.  Reference 87, pp. 10-12.

95.  Reference 87, pp. 9-11.

96.  Martin, G.B. and J.S. Bowen.  Nitrogen Oxides Control.   In:   Energy/
     Environment IV; Proceedings of the Fourth National Conference on the
     Interagency Energy/Environment R&D Program.   Jacoff, F.S. (ed.).
     Washington,  D.C.  U.S.  Environmental Protection Agency.   Publication
     No. EPA-600/9-79-40.   October 1979.  pp.  69-100.

97.  Cato, G.A.,  et al.  (KVB, Inc.).  Field Testing:   Application of
     Combustion Modification to Control Pollutant Emissions  from Industrial
     Boilers — Phase I.   (Prepared for U.S. Environmental Protection
     Agency.)  Research Triangle Park, North Carolina.   Publication No.
     EPA 650/2-74-078a.  October 1974.  p. 73.

98.  Schweiger, B.  Industrial Boilers — What's  Happening Today.   Power
     Magazine. J21_(2).1-S.24.  February 1977.

99.  Mason, H.B., et al.  (Acurex Corporation).  Preliminary Environmental
     Assessment of Combustion Modification Techniques:   Volume II, Technical
     Results.  (Prepared for U.S. Environmental Protection Agency.)
     Research Triangle Park, North Carolina.  Publication
     No. EPA-600/7-77-119b.   October 1977.  pp. 4-22 through 4-27.
                                    4-219

-------
100.  Lim, K.J., et al.  (Acurex Corporation).  Technology Assessment Report
     for Industrial Boiler Applications:  NO  Combustion Modifications.
     (Prepared for U.S. Environmental Protection Agency.)  Research Triangle
     Park, North Carolina.  Publication No. EPA-600/7-79-178f.
     December 1979.  p. 2-16.
101.  Lim, K.J., et al.  (Acurex Corporation).  Environmental  Assessment of
     Industrial Boiler Combustion Modification NO  Controls.  (Prepared for
     U.S. Environmental Protection Agency .)  Resiarch Triangle Park,  North
     Carolina.  Draft Report TR-79-10/EE, Publication No. EPA Contract
     68-02-2160.

102.  McElroy, M.W. and D.E. Shore (KVB, Inc.).  Guidelines for Industrial
     Boiler Performance Improvement.  (Prepared for U.S. Environmental
     Protection Agency.)  Research Triangle Park, North Carolina.
     Publication No. EPA-600/8-77-003a.  January 1977.  pp.  41-44.

103.  Maloney, K.L., et al.  Low Sulfur Western Coal Use in Existing Small
     and Intermediate Size Boilers.  (Prepared for U.S. Environmental
     Protection Agencv.)  Research Triangle Park, North Carolina.
     Publication No. EPA-600/7-78-153a.  July 1978.  p. 26.

104.  Burklin, C.E. and W.D. Kwapil.  (Radian Corporation).  Correlations for
     NO  Emissions from Oil-Fired Industrial Boilers.  (Prepared for
     U.S. Environmental Protection Agency.)  Research Triangle Park, N.C.
     EPA Contract  No. 68002-3058.  March 1982.

105.  Reference 100, p. 4-22.

106.  Cato, 6.A., et al. (KVB, Inc.).  Field Testing:  Application  of
     Combustion Modifications to Control Pollutant Emissions from Industrial
     Boilers, Phase II.   (Prepared for U.S. Environmental Protection
     Agency.)  Research Triangle Park, North Carolina.  Publication
     No. EPA-600/2-76-086a.  April 1976.

107.  Carr, R.C. (KVB,  Inc.).  Effectiveness of Gas Recirculation and Staged
     Combustion in Reducing NO  on a 560-MW Coal-Fired Boiler.  (Prepared
     for Electric  Power Research Institute.)  Palo Alto.  Publication
     No. PB-260-282.  September 1976.

108.  Memo from Berry, R., Radian Corporation, to Industrial  Boiler Files.
     August 31, 1981.  The use of staged combustion and low NO  burners on
     gas- and oil-fired industrial boilers and the use of ESP's on oil-fired
     industrial boilers.

109.  Reference 99, p. 2-30.

110.  Harstine, G.E. and D.C. Williams.  Industrial Pulverized Coal-Fired
     Boiler Designed to Meet Today's Challenges.  (Presented at the American
     Power Conference.  Chicago.  April 24-26, 1978.)  p. 5.
                                   4-220

-------
111.  Carter, W.A., et al.  (KVB,  Inc.).   Emission Reduction  on  Two  Industrial
     Boielrs with Major Combustion Modifications.   (Prepared for
     U.S.  Environmental Protection Agency.)   Research  Triangle Park,  North
     Carolina.  Publication No.  EPA-600/7-78-099a.   June 1978.   p.  94.

112.  Lim,  K.J., et al. (Acurex Corporation.)   Environmental Assessment  of
     Utility Boiler Combustion Modification  NO  Controls.   (Prepared  for
     U.S.  Environmental Protection Agency.)   Research  Triangle Park,  North
     Carolina.  EPA Contract No.  68-02-2160.   April  1978.   p.  5-102.

113.  Reference 99, pp. 2-22, 2-46.

114.  Frey, D.J. and M.S.  McCartney.   The Formation,  Measurement and Control
     of Nitrogen Oxides in Existing Fossil-Fuel  Fired  Steam Generators.
     (Presented at the 28th TAPPI Engineering Conference.   Boston.
     October 1, 1973.)  p. 182.

115.  Telecon.  Milligan,  R., Acurex Corporation, with  Weeks, E.  and
     B. Morton, E. Keller Company.  July 26,  1978.   Information about a
     steam boiler with F6R and special  burner.

116.  Cichanowicz, J.E., et al.   Pollutant Control Techniques for Package
     Boilers:  Phase I Hardware  Modifications and Alternate Fuels.
     (Prepared for U.S. Environmental  Protection Agency.)   Research Triangle
     Park, North Carolina.  EPA Contract No.  68-02-1498.  November  1976.
     pp. 70-81.

117.  Memo from Sedman, C., EPA:ISB, to Industrial Boiler Files.  Emission
     control capabilities of mechanical collectors.  March  2,  1982.

118.  Ando, J., et al. (PedCo Environmental,  Inc.).   NO  Abatement for
     Stationary Sources in Japan.  (Prepared  for U.S.  Environmental
     Protection Agency.)   Research Triangle  Park, North  Carolina.
     Publication No. EPA-600/7-77-103b.  September  1977.  pp.  8, 15.

119.  Telecon.  Lim, K.J.,  Acurex Corporation, with  Boughton, M., TRW, Inc.
     May 21, 1979.  Information  about TRW low NO burner.
                                                J\

120.  Memo  from Burklin, C. and M. Jennings,  Radian  Corporation, to
     Industrial Boiler Team.  Status of low-NO  burners  for process heaters
     and boilers.  March  16, 1982.            x

121.  Koppang, R.R.  A Status Report on the Commercialization and Recent
     Development History of the TRW Low NO  Burner.  January 1976.  p.  13.
                                          A
122.  Reference 99, p. 2-47.

123.  Palazzolo, Michael.   Air Preheat vs. Economizers.  Technical Note.
     Radian Corporation.   Durham, North Carolina.   September 2, 1981.
                                   4-221

-------
124. Castaldini, C., et al.   (Acurex Corporation.)  Technical  Assessment of
     Thermal DeNO  Process.   (Prepared for U.S. Environmental  Protection
     Agency.)  Research Triangle Park, North Carolina.  Publication
     No. EPA-600/7-79-117.  May 1979.  pp. 2-18, 3-8 to 3-12.

125. Maxwell, J.D., et al.  (Tennessee Valley Authority.)  Preliminary
     Economic Analysis of NO  Flue Gas Treatment Processes.  (Prepared for
     U.S. Environmental Protection Agency.)  Research Triangle Park,  North
     Carolina.  Publication  No. EPA-600/7-80-021.

126. Exxon Research and Engineering Company.  Exxon Thermal DeNO  Process.
     New Jersey, Exxon Technology, April 1978.  pp. 1-2.

127. Reference 126, p. 1.

128. Hunter, S.C. and H.J. Buening.  (KVB Engineering, Inc.)  Field Testing:
     Application of Combustion Modifications to Control Pollutant Emissions
     from Industrial Boilers - Phases I and II (Data Supplement).  (Prepared
     for U.S. Environmental Protection Agency.)  Research Triangle
     Park, N.C.  Publication No. EPA-600/2-77-122.  June 1977.  643 p.

129. Habelt, W.W.  The Influence of the Coal Oxygen to Coal Nitrogen Ratio
     on NO   Formation.  (Presented at the 70th Annual AIChE Meeting.)
     Mew YoVk.  November  13-17, 1977.

130. Environmental Protection Agency.  Electric Utility Steam Generating
     Units:  Background Information for Proposed NO  Emission Standards.
     (Prepared for U.S. Environmental Protection Agency).  Research Triangle
     Park, N.C.  Publication No. EPA-450/2-78-005a.  July 1978.  p. 6-1.

131. Jones,  G.D. and K.L. Johnson.  (Radian Corporation.)  Technology
     Assessment Report for Industrial Boiler Applications:  NO  Flue Gas
     Treatment.  (Prepared for U.S. Environmental Protection Agency.)
     Research Triangle Park, North Carolina.  Publication
     No. EPA-600/7-79-178g.  p. 2-56.

132. Ando, J.  NO  Abatement for Stationary Sources in Japan.  (Prepared for
     U.S. Environmental Protection Agency.)  Research Triangle Park, North
     Carolina.  Publication No. EPA-600/7-79-205.  August 1979.  p. 239.

133. Reference 131, p. 2-81.

134. Reference 131, p. 2-7.

135. Faucett, N.L., et al.  (Tennessee Valley Authority.)  Technical
     Assessment of NO  Removal Processes for Utility Application.  (Prepared
     for U.S. Environmental Protection Agency.)  Research Triangle Park,
     North Carolina.  Publication No. EPA-600/7-77-127.  November 1977.
     p. 240.
                                    4-222

-------
136.  Reference 131,  p.  2-25.
137.  Rule 475.1.   South Coast Air Quality Management District El  Monte,
     California.   January 22, 1979.   p.  38.
138.  Reference 131,  p.  2-6.
139.  Reference 131,  p.  2-73.
140.  Reference 131,  p.  6-2.
141.  Reference 131,  p.  2-74.
142.  Reference 131,  p.  2-79.
143.  Reference 125,  p.  29.
144.  Corbett, W.E.,  et  al.  (Radian  Corporation.)   Assessment of  the  Need
     for NO  Flue Gas Treatment Technology.   (Prepared for U.S.
     Environmental Protection Agency.)   Research Triangle  Park, North
     Carolina.  Publication  No. EPA/7-79-178c.   December 1979.  p. 43.
145.  Reference 131,  p.  2-135.
146.  Buroff, J.,  et  al.  (Versar, Inc.)   Technology Assessment for
     Industrial  Boiler  Applications:  Coal  Cleaning and Low Sulfur Coal.
     (Prepared for the  U.S.  Environmental  Protection Agency.) Research
     Triangle Park,  North Carolina.   Publication No. EPA-600/7-79-178c.
     December 1979.   p. 43.
147.  Reference 103,  p.  26.
148.  Reference 143,  p.  44.
149.  Reference 103,  p.  8.
150.  Reference 146,  p.  95.
151.  Comley, E.A., et al.  (Catalytic,  Inc.)  Technology Assessment Report
     for Industrial  Boiler Applications:  Oil Cleaning. (Prepared for
     Environmental Protection Agency.)   Research Triangle  Park, North
     Carolina.  Publication  No. EPA-600/7-79-178b.   July 1980.  p. 61-63.
152.  Reference 146,  p.  53.
153.  Reference 146,  p.  54.
154.  Reference 146,  p.  120.
                                    4-223

-------
155.  Reference 146,  p.  124.

156.  Reference 146,  p.  127.

157.  Reference 146,  p.  130.

158.  Reference 146,  p.  135.

159.  Reference 146,  p.  134.

160.  Reference 146,  p.  145.

161.  Reference 146,  p.  185.

162.  Reference 146,  p.  184.

163.  Reference 146,  p.  56.

164.  Reference 146,  p.  142.

165.  Reference 146,  p.  143.

166.  Reference 151,  p.  41.

167.  Reference 151,  p.  42.

168.  Reference 151,  p.  44.

169.  Reference 151,  p.  43.

170.  Ranney, M.W.  Desulfurization of Petroleum.  Noyes Data Corporation.
     Park Ridge, New Jersey.   1975.   pp.  3,  31.

171.  Reference 151,  p.  64.

172.  Reference 151,  p.  49.

173.  Reference 151,  p.  47.

174.  Reference 151,  p.  51.

175.  Reference 151,  p.  52.

176.  Reference 151,  p.  53.
                                   4-224

-------
177. Thomas, W.C.  (Radian Corporation.)  Technology Assessment Report for
     Industrial Boiler Applications:   Synthetic Fuels.   (Prepared for
     U.S. Environmental Protection Agency.)  Research Triangle Park,  North
     Carolina.  Publication No. EPA-600/7-79-178d.   November 1979.   pp.  1-12
     and 1-13.

178. Reference 177, p. 2-6.

179. Spaite, P.W., and G.C. Page.   Low- and Medium-Btu  Gasification  Systems:
     Technology Overview.   (Prepared  for U.S. Environmental  Protection
     Agency.)  Research Triangle Park, North Carolina.   Publication
     No. EPA-600/7-78-061.  March 1978.  p. 3.

180. Reference 177, p. 2-35.

181. Reference 177, p. 2-34.

182. Reference 177, p. 2-48.

183. Reference 177, p. 2-42.

184. Reference 177, p. 2-49.

185. Callen, R.B., et al.   Upgrading  Coal  Liquids to Gas Turbine  Fuels:
     1. Analytical Characterization of Coal Liquids.  Industrial  and
     Engineering Chemistry Product Research and Development.   15(4):222-233.
     December 1976.

186. Schrieber, R.J., et al.  (Acurex Corporation.)   Boiler  Modification
     Cost Survey for Sulfur Oxides Control by Fuel  Substitution.   (Prepared
     for U.S. Environmental Protection Agency.)  Research Triangle Park,
     North Carolina.  Publication No. EPA-650/2-74-123.   November 1974.
     p. 6-8.

187. Reference 177, p. 2-50.

188. Reference 177, p. 2-51.

189. Reference 177, p. 2-57.

190. Reference 177, p. 2-59.

191. Reference 177, p. 2-62.

192. Reference 177, p. 2-69.

193. Hall, R.E.  The Effect of Water/Residual Oil Emulsions  on Air Pollutant
     Emissions and Efficiency of Commercial Boilers.  (Presented  at  the ASME
     Annual Meeting.  Houston.  November 30 - December  5, 1975.)   p.  2.
                                   4-225

-------
194.  Hall, R.E.   The Effect of Water/Distillate Oil  Emulsions  on  Pollutants
     and Efficiency of Residential  and Commercial  Heating Systems.
     (Presented at the 68th Annual  Air Pollution Control  Association
     Meeting.   Boston.  June 15-20, 1975.)   pp. 4-5.

195.  Barrett,  R.E., et al.  (Battene Memorial  Institute.)   Summary Report
     on Residual Fuel Oil-Water Emulsions.   (Prepared for the  National Air
     Pollution Control Administration.)  Cincinnati,  Ohio.   Publication No.
     PB-189-076.  January 12, 1970.

196.  Dooher, J.P.  Investigations of Combustion of Ultrasonically Generated
     Water/Oil Emulsions.  (Presented at the 1975 Annual  NASA  Facilities
     Conference.  Pasadena.  October 21-24, 1975.)  p.3.

197.  Reference 193, p. 6.

198.  Reference 194, p. 4.
                  /
199.  Reference 193, p. 4.

200.  lammartino, N.R.  Can Water Help Fuel  Burn?  Chemical  Engineering.
     81_(25):84.  November 11, 1974.

201.  Young, C.W., et al.  (GCA Corporation.)  Technology Assessment Report
     for  Industrial Boiler Applications: Fluidized-bed combustion.
     (Prepared for U.S. Environmental Protection Agency.)  Research Triangle
     Park, N.C.  Publication No. EPA-600/7-79-178e.   November  1979.   p. 125.

202.  Reference 201, p. 32.

203.  Reference 201, p. 33.

204.  Reference 201, p. 35.

205.  Reference 201, p. 36.

206.  Letter and attachments from Byam, J.W., U.S. Department of Energy, to
     Jennings, M., Radian Corporation.  December 11, 1980.   11 p.
     Information about atmospheric fluidized bed demonstrations.

207.  Letter and attachment from Wallish, J.W., Johnston Boiler company, to
     Jennings, M., Radian Corporation.  December 11, 1980.   11 p.
     Information about atmospheric fluidized bed demonstrations.

208.  Reference 201, p. 59.

209.  Reference 201, p. 61.

210.  Reference 201, p. 34.
                                    4-226

-------
211. Reference 201, pp.  52-53.
212. Reference 201, p. 67-
213. Reference 201, p. 66.
214. Reference 201, p. 68.
215. Reference 201, p. 73.
216. Reference 201, p. 71.
217. Reference 201, p. 44.
218. Reference 201, p. 81.
219. Reference 201, p. 80.
220. Reference 201, pp.  88-92.
221. Reference 201, pg.  100-101.
222. Reference 201, p. 77.
223. Reference 201, p. 105.
224. Reference 201, p. 174.
225. Reference 201, p. 93.
226. Reference 201, pg.  110-124.
227. Dickerman, J.C. and K.L.  Johnson, Radian Corporation.   Technology
     Assessment Report for Industrial  Boiler Applications:   Flue Gas
     Desulfurization.  (Prepared for U.S.  Environmental  Protection  Agency.)
     Research Triangle Park, N.C.  Publication No.  EPA-600/7-79-178i.
     November 1979.  pp. 2-161.
228. Giammar, R.D., et al.  Limestone/Coal Fuel  Pellet - A Viable Method to
     Control S0« Emissions from Industrial Boilers.   (Presented at  the 73rd
     Annual Meeting of the Air Pollution Control  Association.   Montreal.
     June 22-27, 1980.)
229. Telecon.  Piccot, S., Radian Corporation, with  Davis, B., Versar
     Incorporated.   December 5, 1980.   Conversation  about factors effecting
     boiler operations.
230. Reference 15,  Chapter 3,  p. 2.07.
                                   4-227

-------
231. American Boiler Manufacturers Association.  Emission and Efficiency
     Performances of Industrial Coal Stoker Fired Boilers.  (Prepared for
     U.S. Department of Energy.)  Washington, D.C.  DOE Report
     No. DOE/ET/10386-TI (Vol. I),  pp. 64,65.

232. Kezerle, J.A. and S.W. Mulligan, TRW, Inc.  Performance Evaluation of
     an  Industrial Spray Dryer for S02 Control.  (Prepared for U.S.
     Environmental Protection Agency.7  Research Triangle Park, North
     Carolina.  Publication No. EPA-600/7-81-143.  August 1981.  pp.  4-4 -
     4-7.

233. South Coast Air Quality Management District.  Public Meeting to
     Consider a Suggested Control Measure for the Control of Emissions of
     Oxides of Nitrogen from Boilers and Process Heaters in Refineries.
     Report No. SS-81-016.  October 1981.  p. 38.

234. Young, C., et al.  Continuous Emission Monitoring at the Georgetown
     University Fluidized-Bed Boiler.  (Prepared for U.S. Environmental
     Protection Agency.)  Research Triangle Park, North Carolina.  Contract
     No. 68-02-2693.  Technical Directive No. 5.  September 1981.  pp. 46,
     47.

235. Reference 234, p. 9.

236. Reference 234, pp. 30, 46, 47.
                                    4-228

-------
                    5.0  MODIFICATION AND RECONSTRUCTION

     Standards of performance are applicable to facilities whose construc-
tion, modification, or reconstruction commenced after proposal  of the
standards.  Such facilities are termed "affected facilities."  Standards of
performance are not applicable to "existing facilities" which are facilities
whose construction, modification, or reconstruction commenced on or before
proposal of the standards.  However, an existing facility may become an
affected facility and therefore subject to standards, if the facility under-
goes modification or reconstruction.
     Modification and reconstruction are defined under 40 CFR 60.14 and
60.15, respectively.  The definition of "commenced" appears in 40 CFR 60.2.
Modification and reconstruction provisions are summarized in Section 5.1 of
this chapter.  Section 5.2 discusses the applicability of the provisions to
fossil fuel-fired industrial boilers.
5.1  SUMMARY OF MODIFICATION AND RECONSTRUCTION PROVISIONS
5.1.1  Modification
     With certain exceptions, any physical or operational change to an
existing facility that would result in an increase in the emission rate to
the atmosphere of any pollutant to which a standard of performance applies
would be considered a modification within the meaning of Section 111 of the
Clean Air Act.  The key to a modification determination is whether total
emissions to the atmosphere (expressed in kg/hr) from the facility as a
whole would increase as a result of the change.  For example, if the
affected facility is defined as a group of pieces of equipment, then the
aggregate emissions from all the equipment must increase before the facility
will be considered modified.
                                   5-1

-------
     Exceptions which allow certain changes to an existing facility without
it becoming an affected facility, irrespective of an increase in emissions,
are listed below:
     1.  Routine maintenance, repair, and replacement.
     2.  An increase in production rate without a capital  expenditure
         (as defined in 40 CFR 60.2).
     3.  An increase in the hours of operation.
     4.  Use of an alternate fuel or raw material if, prior to the
         standard, the existing facility was designed to accommodate
         that alternate fuel or raw material.
     5.  The addition or use of any system or device whose primary
         function is the reduction of air pollution, except when an
         emission control system is removed or is replaced by a
         system determined by EPA to be less environmentally beneficial.
     6.  Relocation or change in ownership of the existing facility.
     Once an existing facility is determined to be modified, all of the
emission sources of that facility are subject to the standards of perfor-
mance for the pollutant whose emission rate increased and not just the
emission source which displayed the increase in emissions.  However, a
modification to one existing facility at a plant will not cause other
existing facilities at the same plant to become subject to standards.
     An owner or operator of an existing facility who is planning a physical
or operational change that may increase the emission rate of a pollutant  to
which a standard applies, shall notify the Administrator 60 days prior to
the change, as specified in 40 CFR 60.7(a)(4).
5.1.2  Reconstruction
     An existing facility may also become subject to new source performance
standards if it is determined to be "reconstructed."  As defined in
40 CFR 60.15, a reconstruction is the replacement of the components of an
existing facility to the extent that (1) the fixed capital cost of the new
components exceeds 50 percent of the fixed capital cost of a comparable new
facility and (2) it is technically and economically feasible for the
facility to meet the applicable standards.  Because EPA considers
                                   5-2

-------
reconstructed facilities to constitute new construction rather than
modification, reconstruction determinations are made irrespective of changes
in emission rate.  Determinations are made on a case-by-case basis.  If the
facility is determined to be reconstructed, it must comply with all of the
provisions of the standards of performance applicable to that facility.
     If an owner or operator of an existing facility is planning to replace
components and the fixed capital cost of the new components exceeds
50 percent of the fixed capital cost of a comparable new facility, the owner
or operator shall notify the Administrator 60 days before the construction
of the replacements commences.
5.2  APPLICABILITY OF MODIFICATION AND RECONSTRUCTION PROVISIONS TO
     FOSSIL FUEL-FIRED INDUSTRIAL BOILERS
5.2.1  Modification
     Actions which may increase emissions and therefore may be considered
modifications include changes in the type of fuel  fired and changes in the
boiler components.  These changes are discussed below.
     5.2.1.1  Fuel Switching.  The combustion of an alternate fuel will not
be deemed a modification so long as an existing boiler was designed to
accommodate the alternate fuel as discussed in 40 CFR 60.14(e)(4).  Any
other switch in fuel which increases the emissions of a regulated pollutant
may constitute a modification, with the exception of fuel switches described
in Section lll(a)(8) of the Clean Air Act and those specifically excluded by
the standard.
     5.2.1.2  Physical and Operational Changes.  Physical changes could be
made to many components of a fossil fuel-fired industrial boiler.  This
section summarizes some of the changes which may result in emissions
increases.
     Combustion Air System.  The air flow in a boiler's draft system can be
increased by changing fans and air nozzles in order to correct combustion
problems and to reduce tubing corrosion.  This change could result in
greater excess air and higher air velocities which in turn could increase
particulate matter and NO  emissions.  Other changes in air flow include
                                     5-3

-------
altering the ratio of air added over (overfire air) and under (underfire
cir) the grates.  Increasing the velocity of underfire air may also result
ir increased particulate matter carryover.
     Flue Gas Handling System.  Alterations can be made in the flue gas
handling system by adding an economizer and/or air preheater, or by
replacing the primary fan.  The addition of an economizer would not increase
the emission rate of any pollutant and thus would not be termed a modifica-
tion.  The addition of an air preheater, however, could increase furnace
temperatures and NO  formation.  The likelihood of an owner or operator
                   x                1
installing an air preheater is high.
     Fly Ash Reinjection.  A system to reinject fly ash or unburned carbon
particulate matter from stoker-fired boilers can be added to improve the
overall fuel combustion efficiency and reduce overall operating costs.  Fly
ash reinjection increases the boiler particulate loading and therefore may
increase emissions.  Rapidly rising fuel costs tend to make this alternative
more attractive and may cause some existing facilities to either add injec-
tion systems or increase  injection rates in the future.
5.2.2  Reconstruction
     In a reconstruction determination, when components are replaced as part
of a maintenance program the capital expenditures for each component are
first adjusted by the annual asset guideline repair allowance percentage
(Internal Revenue Service Publication 534) as specified in 40 CFR 60.2.
Replacement of single boiler components would not likely require sufficient
capital to subject an existing facility to the reconstruction provisions
but, replacement of groups of components (e.g., retubing and rebricking) may
result in sufficient expenditures to subject the facility to these
provisions.
     It does not appear likely that existing boilers that undergo normal
repair and maintenance practices will become affected facilities by virtue
of the reconstruction provisions.  The National Board Inspection Code
                                       2
defines repairs as the following items:
                                    5-4

-------
     •  Replacement of sections of boilers tube, provided the
        remaining part of the tube is not less than 75 percent
        of its original thickness.
     •  Seal welding of tubes.
     •  Building-up of certain corroded surfaces.
     •  Repairs of cracked ligaments of drums or headers within
        certain definite limits.
The types of maintenance that will usually require substantial  amounts of
time are boiler cleaning and repair or replacement of various parts.
Primary maintenance  areas for solid fuel-fired boilers are the fuel  feed
system and the fuel firing mechanism.
5.2.3  Summary
     Modification determinations depend upon a physical or operational
change that results in an increased emission rate.  Reconstruction determi-
nations, made by the Administrator on a case-by-case basis, depend on the
level of capital expenditures and on the technological and economic
feasibility of meeting the standard.
     It appears that the reconstruction provisions could cause  some existing
boilers to be reclassified as affected facilities.  In addition, there are
boiler modifications that could result in an existing boiler becoming
classified as an affected facility subject to new source performance
standards.  Likely examples are additions of a fly ash reinjection system or
an air preheater or some types of fuel switching.
                                     5-5

-------
5.3  REFERENCES

1.   Marx, W.B., President, Council of Industrial Boiler Owners, personal
     correspondence with Larry D. Broz, Acurex Corporation.
     February 16, 1980.

2.   Bornstein, M. et^ ^1_.  Impact of Modification/Reconstruction of Steam
     Generators on S09 Emissions.  GCA Corporation.  Bedford, Massachusetts,
     EPA-450/3-77-048:  December 1977.  pp. 12-14.
                                    5-6

-------
                6.0  MODEL BOILERS AND CONTROL ALTERNATIVES

     The impacts of various control  alternatives applied to fossil
fuel-fired industrial  boilers are assessed through an analysis  of "model
boilers".  The model  boiler evaluation provides a boiler-specific analysis
of the environmental,  energy, and cost impacts resulting from the applica-
tion of different emission control techniques to various types  and sizes  of
industrial boilers.  This chapter defines the model  boilers.  Chapters  7  and
8 provide the environmental, energy, and cost impact analyses for the model
boilers.
     Figure 6-1 presents a simplified illustration of the three-step
approach used in developing model boilers.  The first step, discussed in
Section 6.1, is to select a group of "standard boilers".   These standard
boilers are boilers without emission controls that represent the population
of new industrial boilers expected to be built.  The second step, discussed
in Section 6.2, is to select appropriate control alternatives for each
standard boiler.  The alternatives are combinations  of emission levels
and/or emission reduction requirements for NOX, S02> and PM.  Each level
and/or reduction requirement is based on the performance of control methods
as presented in Chapter 4.  The last step, discussed in Section 6.3,  is to
combine the standard boilers with emission control methods to form model
boilers.  Each model  boiler represents a standard boiler controlled to  the
emission levels and/or reduction requirements specified in a control
alternative.  The environmental, energy, and cost impacts associated  with
each model boiler provide an estimate of the impacts of applying a specific
control method to a specific boiler type.  In general, the model boilers  are
selected to cover a range of boiler sizes, fossil fuel types, and control
methods.
     This model boiler selection process results in the generation of
61 model boilers representing the application of various NO , SO^, and  PM
                                    6-1

-------
                         Boiler
                         Capacity
                      (Thermal  Input)
Fuel Type
                                                            Emission  Control  Levels
                                                           Based on Performance Data
                                                                (N0x,  S02,  PM)
                                Boiler  Type
                                    i
                 Select Standard  Boilers
                     Representative
                         of  New
                         Sources
                                                                Combine  Emissions
                                                                 Limits  to  Form
                                                              Control Alternatives
cr>
i
ro
                                Standard
                                 Boilers
                             Specify Control Methods
                             Applicable to Standard
                             Boilers Which Meet
                             Emission Levels in
                             Control Alternatives
                                                          1
   Control
Alternatives
                                                     Model  Boilers
                                   Figure 6-1.   Logic  leading  to  selection  of model  boilers,

-------
control technologies presented in Chapter 4.  In addition to these model
boilers, five "emerging technology" model boilers are defined in Appendix E.
These model boilers are used to estimate the environmental,  energy, and cost
impacts of various technologies that are still  in the development stage,  but
have the potential to become commercially viable control  technologies.   The
emerging control technologies are discussed in Chapter 4  and the model
boiler impacts are presented in Appendix E.
6.1  SELECTION OF STANDARD BOILERS
     Standard boilers are selected to represent the new industrial  boiler
population.  Factors considered in their selection include boiler fuels,
firing methods, heat transfer configurations, and boiler  distribution  by
capacity.  A summary of the standard boilers selected for evaluation is
presented in Table 6-1.
6.1.1  Capacities and Fuel Type
     The boiler capacities and fuels reflected in the standard boilers
represent current and future designs based on the industrial  boiler popula-
tion data presented in Chapter 3.  The principal industrial  boiler fossil
fuels are coal, residual  oil, distillate oil, and natural gas.  Standard
boilers are selected to represent each of the basic fuel  types.   Since  coal
properties such as sulfur and ash content can vary considerably, separate
standard boilers are selected for both low sulfur coal  (LSC)  and high  sulfur
coal (HSC) applications.   Representative boiler design capacities within
each fuel type are then selected to cover the range of expected capacities
for the new industrial boiler population.
     Many new industrial  boilers are expected to fire mixtures of fossil  and
nonfossil fuels.  Mixed-fuel boilers are evaluated in a companion study for
which a separate Background Information Document (BID)  has been prepared.
     As discussed in Chapter 3, capacities of industrial  boilers range  from
less than 0.4 MW (1.5 x 106 Btu/hr) to greater than 146.5 MW
(500 x 10  Btu/hr) thermal input.  The majority of boilers at the lower end
of the capacity range are used for space heating whereas  the boilers at the
                                   6-3

-------
TABLE 6-1.   STANDARD BOILERS SELECTED FOR EVALUATION

Boiler
Code
N6-30
NG-150
DIS-30
DIS-150
RES-30
RES-150
RES-400
HSC-30
HSC-75
HSC-150
HSC-400
LSC-30
LSC-75
LSC-150
LSC-400
Fuel
Natural Gas
Distillate Oil
Residual Oil
High-Sulfur Coal
Low-Sulfur Coal
Heat
MW,
8
44
8
44
8
44
117
8
22
44
117
8
22
44
117
Input Thermal
(106Btu/hr)
.8
.0
.8
.0
.8
.0
.2
.8
.0
.0
.2
.8
.0
.0
.2
(30)
(150)
(30)
(150)
(30)
(150)
(400)
(30)
(75)
(150)
(400)
(30)
(75)
(150)
(400)
Boiler Configuration
Package, Firetube
Package, Watertube
Package, Firetube
Package, Watertube
Package, Firetube
Package, Watertube
Field-Erected, Watertube








Package, Watertube, Underfeed Stoker
Field-Erected, Watertube, Chaingrate Stoker
Field-Erected, Watertube, Spreader Stoker
Field-Erected, Watertube, Pulverized Feed
Package, Watertube, Underfeed Stoker
Field-Erected, Watertube, Chaingrate Stoker
Field-Erected, Watertube, Spreader Stoker
Field-Erected, Watertube, Pulverized Feed

-------
upper end of the capacity range are generally used for process  steam and,  in
some cases, electricity generation.
     The industrial  boiler population capacity range is segmented  into  four
size categories, with appropriate standard boilers chosen  to  represent  each
capacity interval, as noted in Table 6-2.   These four categories span a
range of capacities from 8.8 to 117.2 MW (30 to 400 x 106  Btu/hr).   Two
capacities are selected to represent the range of natural  gas-fired  and
distillate oil-fired boilers while three are selected for  residual oil-fired
boilers.  The natural gas- and distillate oil-fired units  are represented  in
the small sizes by a 8.8 MW (30 x 106 Btu/hr) boiler and in the large
capacities by a 44 MW (150 x 106 Btu/hr) boiler.  The residual  oil-fired
boiler population tends to be larger and is represented by a 8.8 MW
(30 x 106 Btu/hr), a 44 MW (150 x 106 Btu/hr), and a 117.2 MW
(400 x 10  Btu/hr) boiler.  All the natural gas- and oil-fired  boilers  are
package units with the exception of the largest residual oil-fired standard
boiler which is a field-erected unit.  The construction of large oil-fired
boilers is expected to be quite limited due to the high cost of oil  in
comparison to coal.   However, a large residual oil-fired standard boiler is
included to represent possible cases in which combinations of low  residual
oil and/or high coal prices make oil-firing economical.
     In contrast to oil- and natural gas-fired boilers, coal-fired units
vary greatly in firing methods and emission characteristics across their
capacity range.  As a result, coal-fired boilers have been selected  for
evaluation as standard boilers at all capacity intervals with thermal input
capacities of 8.8 MW (30 x 106 Btu/hr), 22 MW (75 x 106 Btu/hr), 44  MW
(150 x 106 Btu/hr), and 117.2 MW (400 x 106 Btu/hr).
6.1.2  Standard Boiler Configurations
     In addition to fuel type and capacity, industrial  boilers  also  vary
according to heat transfer configuration.   The three basic heat transfer
configurations presented and discussed in Chapter 3 are cast  iron, firetube,
and watertube.  No cast iron boilers have been selected for evaluation  due
primarily to their very small size and corresponding low emissions.   Cast
iron boilers are typically found in the small capacity sizes  (less than
                                   6-5

-------
                              TABLE 6-2.   REPRESENTATIVE STANDARD  BOILER  CAPACITIES

Capacity range
(thermal Input)
7.3-14.7 MWK
(25-50 x 10° Btu/hr)
14.7-29.3 MW,
(50-100 x 10b Btu/hr)
29.3-73.3 MW ,
(100-250 x 10° Btu/hr)
>73.3 MW ,
(>250 x 10b Btu/hr)
Coal (HSC & LSC)
8.8 MW ,
(30 x 10° Btu/hr)
22 MW ,
(75 x 10° Btu/hr)
44 MW fi
(150 x 10° Btu/hr)
117.2 MW K
(400 x 10° Btu/hr)
Residual
oil
8.8 MW ,
(30 x 10° Btu/hr)

44 MW ,
(150 x 10° Btu/hr)
117.2 MW ,
(400 x 10° Btu/hr)
Distillate
oil
8.8 MW ,.
(30 x 10° Btu/hr)

44 MW ,-
(150 x 10° Btu/hr)

Natural
gas
8.8 MW ,
(30 x 10°


Btu/hr)

44 MW ,
(150 x 10° Btu/hr)


CTI
CT>

-------
0.1 MW or 0.4 x 10  Btu/hr) and are primarily fired with natural  gas and/or
distillate fuel oil.  Firetube boilers are generally larger than  cast iron
boilers and tend to fire more coal, however, most units larger than 8.8 MW
(30 x 10  Btu/hr) are watertube boilers.   Therefore, all the standard
boilers are of the watertube type except for the smallest 8.8 MW
(30 x 10  Btu/hr) gas- and oil-fired model boilers.
     The firing mechanisms or burners for natural gas- and oil-fired boilers
are essentially the same across their capacity ranges.  (Exceptions have
been noted in Chapter 3).  As a result, no differentiation in the firing
methods has been made for the fuel  oil- and natural gas-fired standard
boilers.  Coal-fired boilers, however, may be equipped with one of several
different firing mechanisms or methods.  The uncontrolled emissions, costs,
and energy requirements of these boilers  are influenced by these
differences.
     Underfeed stokers typically occupy the lower end of the capacity range,
and pulverized coal the upper end,  with other stoker types occupying the
intermediate range between underfeed and  pulverized coal-fired units.
Exceptions do occur, with some boiler types appearing across the  capacity
range.  Because more than 60 percent of the total number of coal-fired
industrial boilers in the 2.9 to 14.7 MW (10 to 50 x 106 Btu/hr)  thermal
input capacity range are underfeed  stokers, an 8.8 MW (30 x 10 Btu/hr)
underfeed stoker has been selected  as representative of that range.  The
chaingrate stoker and spreader stoker are the common firing mechanisms for
medium-sized industrial boilers, with the chaingrate stoker being the more
common firing method for boilers represented by the 22 MW (75 x 10  Btu/hr)
unit.  More than 60 percent of the  stoker-fired boilers in the 29.3 to
73.3 MW (100 to 250 x 10  Btu/hr) thermal input capacity range are spreader
stokers; thus a spreader stoker firing mechanism has been selected for the
44 MW (150 x 10  Btu/hr) capacity coal boiler.  Pulverized coal-fired units
account for only 15 percent of the  coal-fired boilers in the thermal input
                                                               re
                                                               2
capacity range from 29.3 to 73.3 MW (100-250 x 106 Btu/hr)  however,  the
percentage increases toward the upper end of the capacity range.'
Pulverized coal-fired boilers comprise 58 percent of the coal-fired boilers
                                     6-7

-------
in the 73.3 to 147 MW (250 to 500 x 106 Btu/hr) thermal input capacity
range.  Since the lower end of the 29.3 to 73.3 MW (100 to 250 x 106 Btu/hr)
range is represented by the spreader stoker, pulverized coal  firing has been
selected for the 117.2 MW (400 x 106 Btu/hr) capacity coal boilers.
     Seventy-five percent of all boilers in the 29.3 to 73.3  MW (100 to
        C                                                              n
250 x 10  Btu/hr) thermal input capacity range are field-erected units.
The percentage is even higher for coal-fired units.  As a result, coal-fired
boilers selected as standard boilers are all field-erected units, with the
exception of the 8.8 MW (30 x 10  Btu/hr) underfeed stoker unit which is a
package boiler.
6.1.3  Standard Boiler Specifications
     The specifications for the standard boilers are used in  the "model
boiler" environmental, energy, and cost analyses.  The primary specifi-
cations relevant to these analyses are:
       •  Fuel type and quality
       •  Steam capacity and load factor
       •  Flue gas characteristics
Each of these factors are discussed in the following sections.  The
specifications for all of the standard boilers are presented  in Tables 6-3
through 6-7.  Additional specifications required for cost analysis,
including control device specifications, are presented in Chapter 8.
     6.1.3.1  Fuels.  The fuel specifications have been chosen to represent
currently available alternatives for industrial boiler fuels  and are
presented in Table 6-8.  The fuel characteristics, including  heating value
and chemical analysis, are used to determine the combustion-related charac-
teristics of the standard boilers.  Natural gas, distillate oil, and
residual oil are each represented by one type of fuel.  The fuel charac-
                                                                          4
teristics presented for these fuels are based on data for "average" fuels.
The values selected for distillate oil represent No. 2 fuel oil and have
been selected from average values.  One exception is the value for sulfur
content which is chosen from the upper part of the range for  distillate oil.
The analysis for the residual oil has been selected from the  range of values
given for No. 6 fuel oil; again, all values are taken from the middle of the
                                    6-8

-------
                      TABLE  6-3.   SPECIFICATIONS FOR NATURAL GAS-FIRED STANDARD  BOILERS
                                                   (NG-30, NG-150)
at
10

Thermal input, MM (106 Btu/hr)
Fuel rate, m3/hr (ft3/hr)
Analysis
% sulfur
% ash
Heating value, kJ/kg (Btu/lb)
Excess air, %
Flue gas flow rate, m /s (acfm)
Flue gas temperature, K (°F)
Load factor, %
Flue gas constituents,0 kg/hr (Ib/hr)
Fly ash
NO*
COX
HC as CH4
Thermal output, MW (106 Btu/hr)
Steam
Losses
Efficiency (%)
Steam quality
Pressure, kPa (psi)
Temperature, K (°F)
Steam production kg/hr (Ib/hr)
8.8 (30)
850 (30,000)
Trace
Trace
50,707 (21,800)
15
5.28 (11,200)
450 (350)
45
Trace
Trace .
1.63 (3.6)°
0.26 (0.52)
0.04 (0.09)
7.04 (24.0)
1.76 ( 6.0)
80.0
1170 (170)
464 (375)
10,580 (23,300)
44.0 (150)
4250 (150,000)
Trace
Trace
50,707 (21,800)
15
26.44 (56,000)
450 (350)
55
Trace
Trace .
16.34 (36.00)°
1.15 ( 2.56)
0.20 ( 0.46)
38.28 (130.5)
5.72 ( 19.5)
87.0
5170 (750)
672 (750)
47,800 (105,300)
             aLosses include flue gas sensible heat, flue gas water vapor latent heat,  and boiler radiative and
              convective losses.
              Assuming a saturated condensate return at 10 psig.

             cUncontrolled emissions.
              Uncontrolled NOx emissions on  a ng/J (lb/10 Btu) basis are higher for NG-150 boiler due to use of
              air preheater on larger unit.

-------
                     TABLE 6-4.  SPECIFICATIONS FOR  DISTILLATE  OIL-FIRED STANDARD BOILERS
                                                   (DIS-30, DIS-150)
 I
I—*
o

Thermal input, MW (106 Btu/hr)
Fuel rate, m/hr (gal/hr)
Analysis
% sulfur
% ash
Heating value, kJ/kg (Btu/lb)
Excess air, %
Flue gas flow rate, m/s (acfm)
Flue gas temperature, K (°F)
Load factor, %
Flue gas constituents,0 kg/hr (Ib/hr)
Fly ash
so
NO*
COX
HC as CH4
Thermal output, MW (106 Btu/hr)
Steam
Losses3
Efficiency (%)
Steam quality
Pressure, kPa (psi)
Temperature, K ( °F)
Steam production, kg/hr (Ib/hr)
8.8 (30)
0.818 (216)
0.5
Trace
45,346 (19,500)
15
5.38 (11,400)
450 (350)
45
Trace
6.75 (15.3) .
1.63 (3.60)d
0.49 (1.08)
0.10 (0.22)
7.04 (24.0)
1.76 (6.0 )
80.0
1170 (170)
464 (375)
10,580 (23,300)
44.0 (150)
4.09 (1080.0)
0.5
Trace
45,346 (19,500)
15
26.9 (57,000)
450 (350)
55
Trace
34.71 (76.50).
16.34 (36.00)°
2.44 ( 5.40)
0.49 ( 1.08)
38.28 (130.5)
5.72 ( 19.5)
87.0
5170 (750)
672 (750)
47,814 (105,300)
           Losses include flue gas sensible heat, flue gas water vapor latent heat, and boiler radiative
           and convective losses.
           Assuming a saturated condensate return at 10 psig.
          Uncontrolled emissions.
          Uncontrolled NOx emissions on a ng/J  (lb/10 Btu) basis are higher for DIS-150 boiler due to
           use of air preheater on larger unit.

-------
                   TABLE  6-5.  SPECIFICATIONS  FOR RESIDUAL OIL-FIRED  STANDARD  BOILERS

                                          (RES-30,  RES-150,  RES-400)
cr»
i
Thermal Input,
MW (106 Btu/hr) 8.8 (30)
Fuel rate, m3/hr (gal/hr) 0.76 (200)
Analysis
% sulfur
% ash
Heating value
Excess air, %
3.0
0.1
, kJ/kg (Btu/lb) 43,043 (18,500)
Flue gas flow rate, m3/s (acfm) 5.17 (10,950)
Flue gas temperature, K (°F) 478 (400)
Load factor, %
55
Flue gas constituents,0 kg/hr (Ib/hr)
Fly ash 3.13 ( 6.90
SO 43.77 (96.30
NOX 5.44 (12.0
COX 0.45 ( 1.0
HC as CH4 0.09 ( 0.20
Thermal output, MW (106 Btu/hr)
Steam 7.48 (25.5 )
Losses* 1.32 ( 4.50)
Efficiency (%)
Steam quality
Pressure, kPa
Temperature ,
85.0
(psi) 1030 (150)
K(°F) 458 (365)
44.0 (150) 117.2 (400)
3.79 (1001) 10.11 (2670)
3.0 3.0
0.1 0.1
43,043 (18,500) 43,043 (18,500)
15 15
25.80 (54,740) 68.89 (145,960)
478 (400) 478 (400)
55 55
15.65 34.5) 41.73 ( 91.92)
218.47 481.5) 582.59 (1283.24)
27.22 60.0) 72.59 (159.89)
2.27 5.0) 6.05 ( 13.33)
0.45 1.0) 1.20 ( 2.64)
37.40 (127.5) 99.73 (340.38)
6.60 22.5) 17.60 ( 60.07)
85.0 85.0
5170 (750) 5170 (750)
672 (750) 672 (750)
Steam production, kg/hr (lb/hr)b 11,368 (25,044) 48,815 (107,760) 130,440 (287,360)
         aLosses include flue gas sensible heat,  flue gas water vapor

          and convective losses.

         Assuming a saturated condensate return  at 10 psig.

         Uncontrolled emissions.
latent heat, and boiler radiative

-------
                          TABLE 6-6.   SPECIFICATIONS FOR  HIGH-SULFUR COAL-FIRED STANDARD  BOILERS
                                                  (HSC-30,  HSC-75,  HSC-150,  HSC-400)
CTl

I—»
ro
Thermal input, MW (106 Btu/hr)
Fuel rate, kg/s (ton/hr)
8.8 (30)
0.32 (1.27)
22.0 (75)
0.80 (3.18)
44.0 (150)
1.60 (6.36)
117.2 (400)
4.27 (16.95)
Analysis
  % sulfur
  % ash
  Heating value, kJ/kg (Btu/lb)

Excess  air, %

Flue gas flow rate, m /s (acfm)

Flue gas temperature,  K (°F)

Load factor, %
  3.5
 10.6
 27,477  (11,800)

.50

 5.76  (12,224)

 478   (400)

  60
 3.5
10.6
27,447  (11,800)

50

14.42  (30,560)

478    (400)

 60
 3.5
10.6
27,447 (11,800)

50

28.85  (61,120)

478    (400)

 60
 3.5
10.6
27,447  (11,800)

30

66.84  (141,600)

478     (400)

 60
                  Flue gas constituents,   kg/hr (Ib/hr)
Fly ash
SO,
NO;
COX
HC as CH4
Output, MW (106 Btu/hr)
Steam
Losses
Efficiency (%)
Steam quality
Pressure, kPa (psi)
Temperature, K (°F)
Steam production, kg/hr
30.22 ( 66.60)
77.59- (171.00)
4.75 ( 10.50)
1.15 ( 2.54)
0.58 ( 1.28)

6.86 ( 23.4)
1.94 ( 6.6)
78.0

1030 (150)
458 (365)
(lb/hr)b 10,315 (22,723)
75.54 (166.50)
193.96 (427.50)
11.23 ( 24.75)
2.88 ( 6.36)
1.44 ( 3.18)

17.57 ( 59.9)
4.93 ( 15.1)
79.9

1030 (150)
458 (365)
26,440 (58,247)
396.10 (873.00)
387.93 (855.00)
42.88 ( 94.50)
5.76 ( 12.72)
2.88 ( 6.36)

35.58 (121.3)
8.42 ( 28.7)
80.9

3100 (450)
589 (600)
48,502 (106,850)
1304.3 (2876.00)
1034.0 (2280.00)
137.9 ( 304.00)
7.7 ( 16.95)
2.3 ( 5.09)

97.39 ( 332.4)
19.81 ( 67.6)
83.1

5170 (750)
672 (750)
127,010 (280,000)
                  aLosses  include flue gas sensible heat,  flue gas water vapor latent heat,  and boiler radiative
                   and convective losses.
                   Assuming a saturated condensate return  at  10 psig.
                  Uncontrolled emissions.

-------
                        TABLE 6-7.   SPECIFICATIONS  FOR LOW-SULFUR COAL-FIRED  STANDARD  BOILERS
                                             (LSC-30,  LSC-75,  LSC-150,  LSC-400)
CT.
I
I—»
CO

Thermal input, MW (106 Btu/hr)
Fuel rate, m /hr (gal/hr)
Analysis
% sulfur
% ash
Heating value, kJ/kg (Btu/lb)
Excess air, %
o
Flue gas flow rate, m /s (acfm)
Flue gas temperature, K (°F)
Load factor, %
8.8 (30)
0.39 (1.56)

0.6
5.4
22,330 (9,600)
50
5.92 (12,535)
450 (350)
60
22.0 (75)
0.99 (3.91)

0.6
5.4
22,330 (9,600)
50
14.79 (31,339)
450 (350)
60
44.0 (150)
1.98 (7.82)

0.6
5.4
22,330 (9,600)
50
29.58 (62,677)
450 (350)
60
117.2 (400)
5.25 (20.83)

0.6
5.4
22,330 (9,600)
30
68.71 (145,572)
450 (350)
60
Flue gas constituents,0 kg/hr (Ib/hr)
Fly ash
S02
N0*
COX
HC as CH4
Output, MW (106 Btu/hr)
Steam
Losses
Efficiency (%)
Steam quality
Pressure, kPa (psi)
Temperature, K (°F)
Steam production, kg/hr (Ib/hr)
19.08 (42.00)
16.13 (35.60)
4.75 (10.50)
1.41 ( 3.12)
0.71 ( 1.56)

6.89 (23.5)
1.91 ( 6.5)
78.3

1030 (150)
458 (365)
10,331 (22,760)
47.64 (105.00)
40.49 ( 89.25)
11.23 ( 24.75)
3.54 ( 7.82)
1.77 ( 3.91)

17.72 ( 60.4)
4.28 ( 14.6)
80.5

1030 (150)
458 (365)
26,672 (58,760)
248.36 (547.50)
80.99 (178.50)
42.88 ( 94.50)
7.08 ( 15.62)
3.54 ( 7.81)

35.85 (122.2)
8.15 ( 27.8)
81.5

3100 (450)
589 (600)
49,044 (108,044)
816.3 (1800.00)
215.9 ( 476.00)
137.9 ( 304.00)
9.4 ( 20.83)
2.8 ( 6.24)

97.89 (334.1)
19.31 ( 65.9)
83.5

5170 (750)
672 (750)
126,740 (279,200)
          aLosses  include flue gas sensible heat, flue gas water vapor latent heat, and boiler radiative
           and convective losses.
           Assuming a saturated condensate return at 10 psig.

          Uncontrolled emissions.

-------
           TABLE 6-8.   ULTIMATE ANALYSES  OF FUELS SELECTED  FOR THE  MODEL  BOILER  ANALYSIS'

Composition, % by weight3


Fuel Moisture
Natural Gas
Distillate Oil
Residual Oil
Eastern high-sulfur
high-ash coal
0.02
0.05
0.08
8.79
Western low-sulfur
low-ash coal 20.80

Carbon
69.26
87.17
86.62
64.80
57.60

Hydrogen
22.67
12.28
10.20
4.43
3.20

Nitrogen
8.05b
Trace
0.3C
1.30
1.20

Oxygen
Trace
Trace
Trace
6.56
11.20

Sulfur
Trace
0.5
3.00
3.54
0.60

Ash
0
Trace
0.10
10.58
5.40
Heating value9
kj/kg (Btu/lb)
50,707 (21,800)
45,346 (19,500)
43,043 (18,500)
27,447 (11,800)
22,330 ( 9,600)
 All analyses  are based on engineering judgements by PEDCo about information provided by Babcock & Wilcox,
 reference 5.

 Unbound nitrogen, not fuel N2 that can be  converted to "Fuel" NOX emissions.  See Chaper 3.

°Fuel  nitrogen contents of residual oils can vary widely and can have  a  significant effect on NO
 emissions (See Chapter 3).  0.3% has  been  chosen for the model  boiler analysis, but a separate analysis
 of the effect of fuel N2 on achievable NO  emissions is discussed in  Chapter 4.

-------
ranges except the sulfur value, which is from the upper part of the  range.
The sulfur contents are taken from the upper end of the range in order to
provide a worst case analysis and because boiler operators would select
lower cost, higher sulfur fuels in the absence of constraints limiting SCL
emissions.
     Two types of coal  are used to represent the range of coals available in
the United States.  These two coals bound the primary factors which  affect
emission characteristics and control  device performance;  sulfur content,  ash
content, and heating value.  The coal chemical analysis data in Table  6-8
are based on the following fuels:
       •  Eastern high-sulfur, high-ash, bituminous coal  (HSC)
       •  Western low-sulfur, low-ash, subbituminous coal  (LSC)
Although there are several other types of coals suitable  for industrial
boiler use, the two fuels selected for this analysis adequately bound  the
range of impacts.  Practical limits on the number of model  boiler cases that
could be analyzed is also a factor in the selection of HSC and LSC as  the
coal fuels for the model boiler analysis.
     6.1.3.2  Steam Capacities and Load Factors.  The capacities of  the
standard boilers selected in Section  6.1.1 are based on the maximum  heat
input to the boiler.  The heat input  determines the fuel  firing rate using
the heating value of the fuel.  Capacities of industrial  boilers, however,
are often based on steam output.  To  quantify the steam output, the  boiler
thermal efficiency and the steam quality are specified.  The boiler  thermal
efficiency, which is a measure of the boiler heat input transferred  to the
steam cycle, is a function of the fuel properties, firing method, flue gas
characteristics, and boiler heat losses.  The quality of  the steam produced
is specified in terms of temperature  and pressure.  The steam quality  varies
with the intended steam use.  The steam temperatures and  pressures specified
for the standard boilers are those commonly found in industrial applications
for the selected capacities.
     The capacities of the standard boilers represent maximum firing rates.
Boilers, however, seldom operate at maximum capacity year-round.  To analyze
impacts on an annual operating basis, an appropriate estimate of actual
                                   6-15

-------
boiler usage must be provided.  The load factor (or capacity utilization
factor) is the actual annual fuel consumption as a percentage of the
potential annual fuel consumption at maximum firing rate.   Lower load
factors generally represent "non-process" boilers or boilers used mainly for
seasonal space heating and cooling, while the higher factors represent
process boilers whose output is tied directly to plant production.  For each
standard boiler, representative average values from this range are selected
based on previous boiler studies and on data for typical load factors.
Load factors selected for the standard boilers are presented in Tables 6-3
through 6-7 for each boiler type.
     6.1.3.3  Flue Gas Characteristics.  Temperature, composition, and
volumetric flow rate are the main flue gas characteristics upon which the
design of emission control technologies are based.  These characteristics
are affected primarily by fuel composition and boiler excess air.  Fuel
analyses were presented earlier in Table 6-8.  Table 6-9 presents ranges
(percent by weight) of excess air common to different boiler types assuming
no NO  control by reduced excess air.  A representative excess air value for
     A
each standard boiler is specified for each boiler type.
     The pollutant concentrations in the flue gas exiting the boiler are
calculated based on the excess air rate, the chemical composition of the
fuel, the fuel firing rate, and the emission factors developed in Chapter 3.
Tables 6-3 through 6-7 report emission rates on kg/hr (Ib/hr) basis for each
standard boiler.  In Chapter 7, emission factors for the uncontrolled
standard boilers, and for the model boilers controlled to various emission
limits are presented on a ng/J (lb/10  Btu) basis.
     The flue gas flowrates and NO  emission rates are based on the excess
                                  A
air conditions specified in Table 6-9.  The uncontrolled NOX emission rates
(on a ng/J or lb/106 Btu basis) for the NG-30 and DIS-30 boilers are much
lower compared to the larger boilers since the firetube units used in this
size range do not use air preheaters.  Use of air preheaters tends to
increase flame temperatures and NOV emission rates as detailed in Chapter 4.
                                  A
                                     6-16

-------
                        TABLE 6-9.   TYPICAL EXCESS AIR REQUIREMENTS FOR INDUSTRIAL BOILERS
I
^J


Fuel
Coal (pulverized)
Coal (stoker)


Fuel oil
Natural gas

Type of burners
Partially water-cooled
for dry ash removal
Spreader stoker
Chaingrate and
traveling-grate stokers
Underfeed stoker
Multifuel and flat-flame
Multifuel
Typical Range
for
Excess air,
(% by weight)
15-40
30-60
15-50
20-50
10-20
7-15
Specified
excess air,
for standard
boiler
(% by weight
30
50
50
50
15
15

-------
6.2  SELECTION OF CONTROL ALTERNATIVES
     Control alternatives are defined as sets of emission limits and/or
percent reduction requirements for NO , S09, and PM applied to standard
                                     A    u.
(uncontrolled) boilers.  The limits and/or reduction requirements are based
on the performance of various emission control methods as presented in
Chapter 4.  The emission levels selected for the control  alternatives
include a baseline level and several levels involving increasingly stringent
emission reductions from this baseline level.  By comparing the impact at
the baseline level to the impact at other emission levels, it is possible  to
estimate the incremental impacts associated with application of a given
control system to a given boiler type.
6.2.1  Baseline Alternative
     The baseline alternative represents the highest level of emissions
expected under the current mix of existing regulations (SIPs and Subpart D).
The control method selected to meet the baseline alternative generally
represents  the least degree of control applicable to a particular pollutant
and standard boiler.  In most cases, the baseline control method also
represents  the least expensive control method which might be used.
     Table  6-10 details the emission levels selected for the baseline
control alternative (other control alternatives are discussed in the next
section).   The following discussion reviews how the baseline levels were
selected.
     For boilers larger than or equal to 73 MW (250 x 106 Btu/hr) in
capacity, the existing NSPS defines the highest levels of NOX, S02, and PM
expected.   Table 3-19 presented earlier, outlines the emission limits
specified in the existing NSPS (Subpart D) for large fossil fuel-fired
boilers.  Since all new boilers larger than 73 MW (250 x 106 Btu/hr) must
comply with these standards, the emission levels in the existing NSPS have
been adopted for the large boiler baseline alternative.
     For boilers smaller than 73 MW (250 x 106 Btu/hr), the selection of a
baseline alternative is complicated by a variation in SIPs among states.
This variation can be seen in Chapter 3 where SIP regulations are presented.
                                   6-18

-------
  TABLE  6-10.   CONTROL  ALTERNATIVES  SELECTED  FOR  EVALUATION
                          Emission Levels, ng/J (lb/10 Btu)
Alternatives
Coal-fired, *73
NO/
S02a*b PM
.3 MW (250 x 106 Btu/hr)
Baseline 151-271 (0.33-0.
Alt. 1 215 (0.5
Alt. 2 215 (0.5
Alt. 3 215 (0.5
Alt. 4 215 (0.5
Alt. 5 215 (0.5
Coal-fired, ^73. 3 MW (250 x 10
Baseline
Alt. 1
Alt. 2
Alt. 3
Residual -fired,
Baseline
Alt. 1
Alt. 2
Alt. 3
Alt. 4
Residual-fired,
Baseline
Alt. 1
Alt. 2
301 (0.7)
258 (0.6)
258 (0.6)
258 (0.6)
<73.3 MW (250
172 (0.40)
129 (0.3)
129 (0.3)
129 (0.3)
129 (0.3)
£73.3 MW (250
129 (0.3)
129 0.3)
129 (0.3)
63) 1076 (2.5) 172-258 (0.40-0.60)
860 (2.0) 86 (0.2 )
860 (2.0) 22 (0.05)
50% Removal 43 (0.1 )
90% Removal 43 (0.1 )
90% Removal 22 (0.05)
6 Btu/hr)
516 (1.2) 43 (0.1 )
50% Removal 43 (0.1 )
90% Removal 43 (0.1 )
90% Removal 22 (0.05)
x 106 Btu/hr)
688 (1.6) 99 (0.23)
344 (0.8) 43 0.1
129 (0.3) 43 0.1
90% Removal 43 0.1
90% Removal 22 (0.05)
x 106 Btu/hr)
344 (0.8) 43 (0.1
90% Removal 43 (0.1
90% Removal 22 (0.05
 Distillate oil-fired

 Baseline       52-104 (0.12-Q.24)C
 Alt.  1         43-86  (0.1-0.2)C

 Natural gas-fired

 Baseline       52-104 (0.12-Q.24)C
 Alt.  1         43-86  (0.1-0.2)C
219 (0.51)
129 (0.3)
0.43(0.001)
0.43(0.001)
  6 (0.015)
  6 (0.015)
4.3 (0.01)
4.3 (o.Ol)
 NO  and S02 emission  limits and percent removal requirements shown  are
 on a long-term average basis and do not represent requirements for  a
bshorter Interval  such as 24-hrs.
 Percent removal  limits indicate percent S02 emissions are reduced from
 uncontrolled.                                            	    _
£
 NO  emission levels depend on use of air preheat.  Larger boilers
 usi air preheat and consequently have higher NO  emissions.
 Lower level is for 8.8 MW and 22 MW boilers; higher level
 is for 44 MW boiler.
                                    6-19'

-------
     Under the average SIP, S02 emissions are limited to 688 ng/J
(1.6 lb/106 Btu) for oil-fired boilers and 1075 ng/J (2.5 lb/106  Btu)  for
coal-fired boilers.  These levels are selected to represent the baseline
alternative for SC^ for coal and residual oil-fired boilers smaller than
73 MW.  Distillate oil- and natural gas-fired boilers have uncontrolled
emission rates below these levels.  Therefore, the uncontrolled emission
level is selected to represent the baseline alternative for these  boilers.
     NO  emissions from new industrial boilers smaller than 73 MW  are
       A
generally not subject to emission limits under current SIPs.  Therefore,  the
baseline alternative selected for NO  for these boilers is the uncontrolled
                                    A
case for coal-, oil-, and gas-fired model boilers.
     SIP particulate matter emission levels for coal-fired boilers are
generally more variable than the NO  and SO,, emission levels.  Variations in
                                   /\       C-
emission levels between states, boiler types, and boiler sizes make an
average difficult to apply to this set of model boilers.  However, in  many
states, the particulate matter emission limits for coal-fired boilers  are
met with the application of single mechanical collectors.  Therefore,  the
single mechanical collector is selected to represent the control  method
applied under the baseline alternative.  The emission level for mechanical
collectors presented in Table 6-10 are based on the performance data
presented and discussed in Chapter 4.
     Particulate emissions from new oil- and gas-fired boilers are not
subject to emission limitations under current SIPs.  Therefore, the baseline
alternative selected is the uncontrolled case.  The emission levels
presented in Table 6-10 are based on the uncontrolled emission rates for
oil- and gas-fired boilers.
     Because the emissions at baseline vary with respect to boiler size  and
fuel type, the 15 standard boilers are grouped into six classes of boilers
using these parameters.  These six classes are:
     (1)  < 73.3 MW (250 x 106 Btu/hr) coal-fired,
     (2)  > 73.3 MW (250 x 106 Btu/hr) coal-fired,
          < 73.3 MW (250 x 106 Btu/hr) residual oil-fired,
                                     6-20

-------
     (4)  > 73.3 MW (250 x 106 Btu/hr) residual  oil-fired,
     (5)  distillate oil-fired, and
     (6)  natural gas-fired.
Separate control alternatives are developed for each class.
6.2.2  Other Alternatives
     For each of the six classes of standard boilers, up to  five  alterna-
tives involving further emission reductions beyond baseline  are specified.
These alternatives are shown  in Table 6-10.  The six categories of  boilers
in Table 6-10 correspond to the six classes of model boilers defined  above.
In general, the number of alternatives is reduced for the larger  boilers
since the available degree of control beyond baseline is reduced.
Similarly, the number of alternatives for the distillate and natural
gas-fired boilers are limited due to low uncontrolled emissions.
     The emission levels and  alternatives presented in Table 6-10 do  not
represent every possible type and form of an emission standard.   There  is a
practical limit to the number of cases which can be examined on an
individual boiler basis.  The selected levels were chosen to permit environ-
mental, energy, and cost impacts to be evaluated over a range of  control
alternatives and associated emission reductions.
6.3  SUMMARY OF CONTROL SYSTEMS AND MODEL BOILERS
     A controlled standard boiler is termed a model boiler and is used  to
evaluate environmental, energy, and cost impacts of NO   S09, and PM
                                                      A   C.
control.  Results of these evaluations are presented in Chapters  7  and  8.
     Control methods selected to meet each emission limit or percent
reduction requirement are based on the performance of the control method  as
presented in Chapter 4.  In many cases, more than one emission control
method or combination of control methods can achieve a specified  control
level.  As general guidelines, model boiler control methods  were  selected
based on the technology's ability to meet a specified emission limit, the
development status and commercial availability.    In the case of  sidestream
separator controls, the available emission data is very limited.  The
emission level assumed for purposes of this analysis may not be indicative
of long term emission levels  on individual boilers.
                                    6-21

-------
     Control methods selected for model boiler evaluations include single
mechanical collectors, sidestream separators, wet scrubbers (also used for
S02 removal), electrostatic precipitators, HDS cleaned oils, and fabric
filters for particulate control; low excess air, staged combustion air, and
reduced air preheat for N0v control; and HDS cleaned oils, low sulfur coal,
                          A
double alkali scrubbing, dry scrubbing, and sodium throwaway scrubbing for
SCL control.  Since the cost and environmental impacts of the double alkali
      •                                                                    8
and lime/limestone SCL control  systems have been shown to be very similar,
either of the processes could be used to evaluate impacts for wet SO,,
controls producing a sludge.  The dual alkali process was selected to
represent wet sludge-producing  processes over the lime/limestone process
since there are more industrial boiler dual alkali systems than lime/lime-
stone systems.
     The control systems selected to achieve the emission levels in each
control alternative are shown in Tables 6-11 and 6-12 for coal-fired and
oil/gas-fired boilers respectively.  Abbreviations used in these tables are
defined in Table 6-13.  The emission limits in each regulatory alternative
are repeated from Table 6-10.   For  each standard boiler/control system
combination, a model boiler is  defined.  A total of 61 model boilers are
defined in this manner.
      In subsequent chapters, model  boilers are often referred to by a code
consisting of abbreviations for fuel type, boiler size, and control system.
As an example, the following code,
                    HSC -  150 - SCA, FGD, ESP
refers to a high-sulfur coal-fired, 44 MW (150 x 10  Btu/hr) model boiler
with  staged combustion air, double  alkali flue gas desulfurization
scrubbing, and electrostatic precipitator controls.  Similar codes are used
for all model boilers.
                                     6-22

-------
                                          TABLE 6-11.   COAL-FIRED MODEL BOILERS
 I
ro
CA>
Emission Levels or Removal Requirements
ng/J (lb/106 Btu)
Standard. Control
Boilerf Alternative NO SO,,
A £


HSC-30
HSC-75
HSC-150
B 151-271(0.
1 215 0.
2 215 0.
3 215 0.
4 215 0.
\ 5 215(0.
B 151-271(0.


LSC-30
LSC-30
LSC-150
33-0.63)° 1076 (2.5)
5) 860 (2.0)
5) 860 (2.0)
5) 50% Removal
5) 90% Removal
5) 90% Removal
33-0.63)c 1076 (2.5)
1 215 (0.5) 860 (2.0)
2 215 (0.5) 860 (2.0)
3 215 (0.5) 50% Removal
4 215 (0.5) 90% Removal
5 215 (0.5) 90% Removal
B 301 (0
HSC-400
1 258 (0
2 258 (0
3 258 (0
.7 516 (1.2)
.6 50% Removal
.6 90% Removal
.6 90% Removal
B 301 (0.7) 516 (1.2)
LSC-400

1 258 (0.6) 50% Removal
2 258 (0.6) 90% Removal
3 258 (0.6) 90% Removal
PM
172-258 (0.40-0.60)c
86 (0.2)9
22 (0.05)
43 (0.1)
43 (0.1)
22 (0.05)
172-258 (0.40-0.60)c
86 (0.2)9
22 (0.05)
43 (0.1)
43 (0.1)
22 (0,05)
43 (0.1)
43 (0.1)
43 (0.1)
22 (0.05)
43 (0.1)
43 (0.1)
43 (0.1)
22 (0.05)
Control
N0x
Uncb
SCAb
SCAb
a h
SCAK
SCAb
Unch
SCAb
SCA?
SCAb
SCAb
SCAb
LEA
a
SCA
SCA
LEA
SCA
SCA
SCA
System6
so2
cc
cc
cc
a
FGD
FGD
Unc
Unc
Unc
DS
FGO
FGD
FGDd
a
FGD
FGD
Unc
DS
FGD
FGD

PM
SH a
SSS9
ESP
a
FGD/PM
ESP
SM
SSS9
FF
DS/PM
FGD/PM
FF
FGD/PM
a
FGD/PM
ESP
FF
DS/PM
FGD/PM
FF
           50% S02 removal alternative not applicable for HSC  standard boilers since this removal would not meet  baseline
           emission  level.   Therefore, no model boiler is analyzed  for this alternative.

           SCA required on 44 MW (150 x 10  Btu/hr)  size only; smaller boilers meet NOx  level without control.

         cBaseline emissions depend on boiler size  and type (see  text and Chapter 7).

         d78.9% S0? removal efficiency required at  baseline.

         Abbreviations defined in Table 6-13.  Unc (uncontrolled)  indicates no control system is required to  meet  emission
           levels.

           Alternatives shown define model boilers for each standard boiler.  For example, six model boilers are  defined
           for HSC-30, six are defined for HSC-75, etc.
         9SSS emission level  based on limited  emission  data (see Chapter 4).

-------
                                           TABLE 6-12.   OIL-  AND  GAS-FIRED MODEL BOILERS
ro


Emission
Standard Control
Boi1erc Alternative N0x

RES-30
RES- 150





DIS-30
DIS-150
N6-30
NG-150
B 172 (0.4)
1 129 (0.3)
2 129 (0.3)
3 129 (0.3)
4 129 (0.3)
(B 129 (0.3)
1 129 (0.3)
< 2 129 (0.3)
(B 52 (0.12)
| 1 43 (0.1)
\ B 104 (0.24)
[ 1 86 (0.20)
(B 52 (0.12)
f 1 43 (0.1)
fB 104 (0.24)
(1 86 (0.20)
Levels or Removal Requirements
ng/J (lb/106Btu)
so2
688 (1.6)
344 (0.8)
129 (0.3)
90% Removal
90% Removal
344 (0.8)
90% Removal
90% Removal
219 (0.51)
129 (0.3)
219 (0.51)
129 (0.3)
0.43 (0.001)
0.43 (0.001)
0.43 (0.001)
0.43 (0.001)
PM
99 (0.23)
43 (0.1)
43 (0.1)
43 (0.1)
22 (0.05)
43 (0.1)
43 (0.1)
22 (0.05)
6 (0.015)
6 (0.015)
6 (0.015)
6 (0.015)
4.3 (0.01)
4.3 (0.01)
4.3 (0.01)
4.3 (0.01)
Control System
N0x
Unc
LEA
LEA
LEA
LEA
LEA
LEA
LEA
Unc
LEA
Unc
LEA/RAP
Unc
LEA
Unc
LEA/RAP
so2
HDS (1.6)
HDS (0.8)
HDSJ0.3)
FGDa
FGDa
F6Dd
FGD
FGD
Unc
HDS (0.3)
Unc
HDS (0.3)
Unc
Unc
Unc
Unc
PM
Unc
HDS/PM
HDS/PM
FGD/PM
ESP
FGD/PM
FGD/PM
ESP
Unc
Unc
Unc
Unc
Unc
Unc
Unc
Unc
             aDouble alkali scrubbingfi(FGD)  used on 44 MW (150 x 10   Btu/hr), sodium throwaway (FGD/Na)
              used on 8.8 MW (30 x 10° Btu/hr).

              Abbreviations defined in Table 6-13.  Unc (uncontrolled)  indicates no control  system  is required
              to meet emission levels.

             Alternatives shown define model boilers for each standard boiler.  For example,  five  model
              boilers are defined for RES-30, five for RES-150, etc.
              75% removal efficiency required at baseline.

-------
              TABLE 6-13.  ABBREVIATIONS FOR CONTROL SYSTEMS
NO  Control Systems
  /\

     SCA - Staged combustion air (overfire air) used in combination
             with LEA
     LEA - Low excess air
     RAP - Reduced air preheat

S0« Control Systems
     CC
     F6D

     FGD/Na
     DS
Compliance coal
Double alkali  scrubbing flue gas desulfurization (90% removal
  unless noted)
Sodium throwaway flue gas desulfurization (90% removal)
Dry scrubbing (50% removal)
     HDS(x) - Hydrodesulfurized oil (x percent sulfur)

PM Control Systems

     SM    - Single mechanical collector (multitube cyclone)
     SSS   - Sidestream separator
     ESP   - Electrostatic precipitator
     FF    - Fabric filter
     FGD/PM- Particulate removal via  FGD scrubber
     DS/PM - Particulate removal via  DS fabric  filter
     HDS/PM- Particulate removal via  low ash  HDS  cleaned  oil
 Compliance coal is defined as a coal with a sulfur content allowing an
 emission limit to be met without control.  The sulfur content is less
 than HSC but greater than LSC (actual sulfur content depends on emission
 limit, see Chapter 8.)
                                    6-25

-------
6.4  REFERENCES

1.   U.S. Environmental Protection Agency.  Background Information Document
     for Nonfossil Fuel Fired Boilers.  Research Triangle Park, N.C.
     Publication No. EPA-450/3-82-007.  March, 1982.

2.   Devitt, T.  (PEDCo Environmental, Inc.)  Population and Characteristics
     of  Industrial/Commercial Boilers in the U.S.  (Prepared for
     U.S. Environmental Protection Agency.)  Research Triangle Park,  N.C.
     Publication No. EPA-600/7-79-178a.  August 1979.  p. 17.

3.   Reference 2, p. 91.

4.   Useful Tables for Engineers  and Steam Users, Thirteenth Edition.  New
     York, Babcock and Wilcox,  1978.  p. 39.

5.   Steam, Its Generation  and  Use, 38th Edition.  New York, Babcock  and
     Wilcox, 1975.   pp.   5-1  to 5-22.

6.   Reference 2, pp. 33-37,  p. 110.

7.   Reference 2, pp. 93-102.

8.   Dickerman, J.C. and  J.L. Johnson.   (Radian Corporation).  Technology
     Assessment Report for  Industrial Boiler Applications:   Flue Gas
     Desulfurization.  (Prepared  for U.S. Environmental  Protection Agency.)
     Research Triangle Park,  N.C.   Publication No.  EPA-600/7-79-178i.
     November 1979.  p. 1-3.
                                     6-26

-------
                   7.0  ENVIRONMENTAL AND ENERGY IMPACTS

     An analysis of the environmental and energy impacts that result from
applying various emission control  technologies to individual  fossil
fuel-fired industrial  boilers is presented in this chapter.   Environmental
and energy impacts of the emerging control technologies are  presented in
Appendix E.  National  and regional environmental and energy  impacts
resulting from application of various control technologies to the projected
new industrial boiler population were analyzed using the Industrial  Fuel
Choice Analysis Model  (IFCAM) and the results are presented  in Chapter 10.
     The environmental and energy impact analysis is based on an evaluation
of the model boilers presented in Chapter 6.   The objective  of this  chapter
is to determine the incremental increase or decrease in air  pollution, water
pollution, solid waste generation, and energy usage for various alternative
control levels compared to the baseline level. The baseline  control  level
corresponds to no change in existing regulations and represents the  controls
required under the current mix of industrial  boiler SIP and  NSPS regulations
(40 CFR Subpart D).
     Tables 7-1 and 7-2 specify the NO , S0?, and PM emission factors for
                                      rt    Cm
the model boilers using no controls, baseline controls and various control
alternatives outlined in Chapter 6.  Emission factors for the control
alternatives are developed in Chapter 6.  The technologies used to meet
these control alternatives are identified in  Chapter 6 (Tables 6-11  and
6-12) and described in Chapter 4.
7.1  AIR POLLUTION IMPACTS
     This section presents an analysis of the air pollution  impacts
associated with each model boiler.  The air pollution impact analysis is
divided into two main subsections as outlined below:
                                    7-1

-------
                   TABLE 7-1.   SUMMARY OF  EMISSION FACTORS  FOR COAL-FIRED  MODEL BOILERS
IX)

Standard
Boiler

HSC-30


HSC-75


HSC-150


HSC-400


LSC-30


LSC-75


LSC-150


LSC-400

Capacity
(106 Btu/hr)

30


75


150


400


30


75


150


400

Emission
Specie
NO
SO*,
PM2
NO
SO*
PM2
N0v
SO*,
PM2
N0y
so*,
PM2
NO
SO*,
PM2
NO
so*,
PM2
NO
SO*,
PM2
NO
SO*
PM2
Emission Factors
Uncontrolled
0.35
5.70
2.22
0.33
5.70
2.22
0.63
5.70
5.82
0.76
5.70
7.19
0.35
1.19
1.40
0.33
1.19
1.40
0.63
1.19
3.65
0.76
1.19
4.50
Baseline
0.35
2.50
0.40
0.33
2.50
0.40
0.63
2.50
0.60
0.70
1.20
0.10
0.35
1.19
0.40
0.33
1.19
0.40
0.63
1.19
0.60
0.70
1.19
0.10
Alt lb
0.35
2.00
0.20
0.33
2.00
0.20
0.50
2.00
0.20
NA
NA
NA
0.35
1.19
0.20
0.33
1.19
0.20
0.50
1.19
0.20
0.60
0.60
0.10
(lb/10
Alt 2
0.35
2.00
0.05
0.33
2.00
0.05
0.50
2.00
0.05
0.60
0.57
0.10
0.35
1.19
0.05
0.33
1.19
0.05
0.50
1.19
0.05
0.60
0.12
0.10
6 Btu)a
Alt 3b
NA
NA
NA
NA
NA
NA
NA
NA
NA
0.60
0.57
0.05
0.35
0.60
0.10
0.33
0.60
0.10
0.50
0.60
0.10
0.60
0.12
0.05

Alt 4
0.35
0.57
0.10
0.33
0.57
0.10
0.50
0.57
0.10
NA
NA
NA
0.35
0.12
0.10
0.33
0.12
0.10
0.50
0.12
0.10
NA
NA
NA

Alt 5
0.35
0.57
0.05
0.33
0.57
0.05
0.50
0.57
0.05
NA
NA
NA
0.35
0.12
0.05
0.33
0.12
0.05
0.50
0.12
0.05
NA
NA
NA
        *To convert to ng/J, multiply by 430.
        50 percent S02 removal alternative not evaluated for HSC-f1red model boilers.

-------
                    TABLE 7-2.   SUMMARY  OF  EMISSION  FACTORS  FOR OIL- AND GAS-FIRED MODEL BOILERS
i
CO

Standard
Boiler

RES-30


RES-150


RES- 400


DIS-30


DIS-150


NG-30


NG-150

Capacity Emission
(106 Btu/hr) Specie

30


150


400


30


150


30


150

Wv
SO?
PM^
mv
so;
PM2
W
SO?
PM2
Wv
so?
PM2
m
so?
PM2
NO
SO?
PM^
NOU
SO?
PM2

Uncontrolled
0.40
3.21
0.23
0.40
3.21
0.23
0.40
3.21
0.23
0.12
0.51
0.02
0.24
0.51
0.02
0.12
trace
0.01
0.24
trace
0.01
Emission
Basel ine
0.40
1.60
0.23
0.40
1.60
0.23
0.30
0.80
0.10
0.12
0.51
0.02
0.24
0.51
0.02
0.12
trace
0.01
0.24
trace
0.01
Factors
Alt 1
0.30
0.80
0.10
0.30
0.80
0.10
0.30
0.32
0.10
0.10
0.30
0.02
0.20
0.30
0.02
0.10
trace
0.01
0.20
trace
0.01
(lb/10
Alt 2
0.30
0.30
0.10
0.30
0.30
0.10
0.30
0.32
0.05
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
6 Btu)a
Alt 3
0.30
0.32
0.10
0.30
0.32
0.10
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA

Alt 4
0.30
0.32
0.05
0.30
0.32
0.05
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
             To convert to ng/J multiply by 430.

-------
     Primary Air Impacts (Section 7.1.1)
       •  model boiler annual emissions and emission reductions
       •  model boiler dispersion analysis.
     Secondary Air Impacts (Section 7.1.2)
       •  annual emissions from facilities supplying power to operate
          pollution control devices.
     The primary air impacts subsection presents the annual emissions for
each control alternative and discusses the impacts of increasingly stringent
control alternatives.  Both source emissions and ambient air impacts are
discussed.  The secondary air impacts subsection evaluates emissions that
result from facilities supplying electrical power to operate the pollution
control devices as they are applied to individual boilers.
7.1.1  Primary  Impacts
     7.1.1.1  Model Boiler Emissions and Emission Reductions
     Numerical  emission factors for each model boiler control alternative
are developed in Chapter 6 and presented in Tables 7-1 and 7-2.
Uncontrolled model boiler emission factors are also included.  Based on
these emission  factors, annual emissions of SCL, PM and NO  are tabulated in
                                              £           A
Tables 7-3 and  7-4 for all coal-, oil-, and gas-fired model boilers.
Calculation of  annual emissions is based on the percent of the boiler
capacity used on an annual basis.  These capacity utilization or load
factors are as  follows:

          Boiler Capacity and Fuel                          Load Factor
      8.8 MW (  30 x 10? Btu) Natural Gas & Distillate Oil      0.45
       44 MW (150 x 10° Btu) Natural Gas & Distillate Oil      0.55
      All residual-fired boilers                               0.55
      All coal-fired boilers                                   0.60
     From the annual emissions presented in Tables 7-3 and 7-4, the annual
emission reductions achieved under each control alternative are calculated.
These emission  reductions are quantified in two ways including:  incremental
annual emission reductions achieved over the baseline alternative and
incremental percent reductions achieved over the baseline alternative.
                                    7-4

-------
                                  TABLE 7-3.   COAL-FIRED MODEL  BOILER ANNUAL  EMISSIONS
cn

Standard
Boiler

HSC-30


HSC-75


HSC-150


HSC-400


LSC-30


LSC-75


LSC-150


LSC-400

Capacity Emission
(106 Btu/hr) Specie
N0tf
30 SOj
PM2
N0v
75 SO*
PTT
NO
150 SO*,
PM2
nov
400 SO*,
PM2
NO
30 so;
PfT
NO
75 so;
PM2
NO
150 so;
PM2
NO
400 SO*,
PM2
Annual Emissions (Tons/yr)a
Uncontrolled
28
449
175
65
1123
438
248
2247
2294
799
5992
7558
28
94
110
65
235
276
248
469
1439
799
1251
4730
Basel 1 ne
28
197
32
65
493
79
248
986
237
736
1261
105
28
94
32
65
235
79
248
469
237
736
1251
105
Alt lb
28
158
16
65-
394
39
197
788
79
NA
NA
NA
28
94
16
65
235
39
197
469
79
631
631
105
Alt 2
28
158
4
65
394
10
197
788
20
631
599
105
28
94
4
65
235
10
197
469
20
631
126
105
Alt 3b
NA
NA
NA
NA
NA
NA
NA
NA
NA
631
599
53
28
47
8
65
118
20
197
237
39
631
126
53
Alt 4
28
45
8
65
112
20
197
225
39
NA
NA
NA
28
9
8
65
24
20
197
47
39
NA
NA
NA
Alt 5
28
45
4
65
112
10
197
225
20
NA
NA
NA
28
9
4
65
24
10
197
47
20
NA
NA
NA
                To convert to Mg/yr multiply by 0.908.
               '50 percent S02 removal  alternative  not calculated for HSC-f1red model boilers.

-------
                            TABLE 7-4.  OIL- AND GAS-FIRED MODEL BOILER ANNUAL EMISSIONS
i
en

Standard
Boiler

RES-30


RES-150


RES-400


DIS-30


D IS- 150


N6-30


NG-150

Capacity Emission
(106 Btu/hr) Specie
NO
30 SO*
PM2
NO
150 S0$
PM2
NO
400 SO*,
PM2
NO
30 SO*
PM2
N0v
150 so;
PM2
NO
30 SO?
PM2
NOV
150 SO?
PM2
Annual Emissions (Tons/yr)
Uncontrolled
29
232
17
145
1160
83
385
3093
222
7
30
1
87
184
5
7
trace
1
87
trace
4
Basel ine
29
116
17
145
578
83
289
771
96
7
30
1
87
184
5
7
trace
1
87
trace
4
Alt 1
22
58
7
108
289
36
289
308
96
6
18
1
72
108
5
6
trace
1
72
trace
4
Alt 2
22
22
7
108
108
36
289
308
48
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Alt 3
22
23
7
108
116
36
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Alt 4
22
23
4
108
116
18
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
               To  convert to Mg/yr multiply by 0.908.

-------
Emission reductions over baseline can be interpreted as the amount of the
baseline emissions captured by applying a more stringent control
alternative.
     Annual emission and percent reductions over baseline are presented in
Tables 7-5 and 7-7 for coal-fired model boilers, and Tables 7-6 and 7-8 for
oil- and gas-fired model boilers.  The following discussion focuses on these
reductions and identifies trends across the range of impacts presented.
SOo Emission Reductions
     Prior to a discussion of the trends shown, two general points are noted
concerning application of the control alternatives.  First, the baseline
alternative for LSC-fired model boilers does not require application of
controls.  Uncontrolled SC^ emissions from LSC-fired model  boilers are below
the levels required under the mix of existing regulations.   Second, for
HSC-fired model boilers, the alternatives requiring 50 percent SCL removal
are not applied because the baseline alternatives for HSC-fired boilers are
more stringent.
     Several trends are evident in Tables 7-5 and 7-6 regarding SCL emission
reductions from coal-fired model boilers.  For model boilers smaller than
73 MW, HSC-fired boilers have the highest actual emission reductions.  This
occurs because the SCL emissions available for capture at baseline are about
two times greater for HSC-fired boilers smaller than 73 MW (see Table 7-3).
For Alternatives 3 and 4 where 90 percent SO^ removal by flue gas
desulfurization is applied, the emission reduction over baseline for the
44 MW (150 x 106 Btu/hr) HSC-fired model boiler is 691 Mg/yr (761 tons/yr).
The emission reduction for the 44 MW LSC-fired model boiler is 383 Mg/yr
(422 tons/yr).  Under Alternatives 3 and 4 percent reduction over baseline
values are 77 and 90 percent for the HSC- and LSC-fired model boilers,
respectively.
     For coal-fired model boilers larger than 73 MW (250 x 10  Btu/hr) this
trend is reversed as a result of the relatively stringent baseline
alternative applied to HSC-fired model boilers.  Because 80 percent of the
uncontrolled emissions from HSC-fired model boilers are captured at
baseline, the alternatives requiring 90 percent removal result in a
                                     7-7

-------
              TABLE  7-5.   COAL-FIRED MODEL BOILER ANNUAL
                          EMISSION REDUCTIONS OVER BASELINE

Emission Reductions (Tons/yr)a
Standard
Boiler

HSC-30


HSC-75


HSC-150


HSC-400


LSC-30


LSC-75


LSC-150


LSC-400

Capacity
(106 Btu/hr)

30


75


150


400


30


75


150


400

Emission
Specie
NO
so;;
PM
NO
so;
PM2
NO
so«
PM^
NO
SO,
PM2
NO
sol
PM2
NO
so«
PM^
NOV
so;
PM2
NOV
so;
PM2
Alt lb
0
39
16
0
99
39
51
197
158
NA
NA
NA
0
0
16
0
0
39
51
0
158
105
620
0
Alt 2
0
39
28
0
99
69
51
197
217
105
662
0
0
0
28
0
0
69
51
0
217
105
1125
0
Alt 3b
NA
NA
NA
NA
NA
NA
NA
NA
NA
105
662
53
0
47
24
0
116
59
51
233
197
105
1125
53
Alt 4
0
152
24
0
380
59
51
761
197
NA
NA
NA
0
84
24
0
211
59
51
422
197
NA
NA
NA
Alt 5
0
152
28
0
380
69
51
761
217
NA
NA
NA
0
84
28
0
211
69
51
422
217
NA
NA
NA
aTo convert to Mg/yr multiply by 0.908.
b
 50 percent S02 removal  alternative not calculated  for HSC-fired
 model  boilers.
                                   7-8

-------
     TABLE 7-6.   OIL-  AND GAS-FIRED MODEL  BOILER ANNUAL
                 EMISSION REDUCTIONS  OVER  BASELINE

Emission Reductions (Tons/yr)a
Standard Capacity Emission
Boiler (106 Btu/hr) Specie

RES-30 30


RES-150 150


RES-400 400


DIS-30 30


DIS-150 150


NG-30 30


NG-150 150

NO
SO*
PMZ
NOV
SO*
PM
NO
SO*
PMZ
NO
SO*,
PM2
NO
SO*,
PM
NOV
SO*,
PIT
NO
soj
PM2
Alt 1
7
58
9
36
289
47
0
463
0
1
12
0
14
76
0
1
0
0
14
0
0
Alt 2
7
94
9
36
470
47
0
463
48
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Alt 3
7
93
9
36
463
47
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Alt 4
7
93
13
36
463
65
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
aTo convert to Mg/yr multiply  by 0.908.
                                  7-9

-------
            TABLE 7-7.   COAL-FIRED MODEL BOILER PERCENTAGE
                        EMISSION REDUCTIONS OVER BASELINE

Standard Capacity Emission
Boiler (106 Btu/hr) Specie

HSC-30 30


HSC-75 75


HSC-150 150


HSC-400 400


LSC-30 30


LSC-75 75


LSC-150 150


LSC-400 400

NO
SO*
PM
NO
SO*
PM
NO
SO*
PM
NO
so;
PM2
NOV
SO*
PM
NO
so*,
PMZ
NO
so*2
PM2
NO
SO*,
PM2
Reductions (Percent)
Alt la
0
20
50
0
20
50
21
20
67
NA
NA
NA
0
0
50
0
0
50
21
0
67
14
50
0
Alt 2
0
20
88
0
20
88
21
20
92
14
53
0
0
0
88
0
0
88
21
0
92
14
90
0
Alt 3a
NA
NA
NA
NA
NA
NA
NA
NA
NA
14
53
50
0
50
75
0
50
75
21
50
83
14
90
50
Alt 4
0
77
75
0
77
75
21
77
83
NA
NA
NA
0
90
75
0
90
75
21
90
83
NA
NA
NA
Altx5
0
77
88
0
77
88
21
77
92
NA
NA
NA
0
90
88
0
90
88
21
90
92
NA
NA
NA
a
 50 percent S02 removal alternative not evaluated for HSC-fired model
 boilers.
                                   7-10

-------
TABLE 7-8.  OIL- AND GAS-FIRED MODEL BOILER PERCENTAGE
            EMISSION REDUCTIONS OVER BASELINE

Reductions (Percent)
Standard Capacity Emission
Boiler (106 Btu/hr) Specie

RES-30 30


RES-150 150


RES-400 400


DIS-30 30


DIS-150 150


NG-30 30


NG-150 150

NO
SOo
PM2
NOV
S0£
PM
NOV
so;
PM2
NO
SO*
PM
NO
so*.
PM
NO
SO*,
PM2
NO
SO?
PM2
Alt 1
25
50
57
25
50
57
0
60
0
17
41
0
17
41
0
17
0
0
17
0
0
Alt 2
25
81
57
25
81
57
0
60
50
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Alt 3
25
80
57
25
80
57
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Alt 4
25
80
78
25
80
78
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
                          7-11

-------
relatively small incremental reduction over baseline in comparison to
LSC-fired model boiler incremental reductions.
     As a result of the low uncontrolled SO,, emissions from natural  gas- and
distillate oil-fired model boilers, only one control alternative is
evaluated which applied to distillate oil-fired model  boilers.   Alternative
1, where hydrodesulfurized fuel oil is applied, results in an 11 to  69 Mg/yr
(12 to 76 tons/yr) emission reduction over baseline across the  boiler size
range.
     Residual oil-fired boilers have higher uncontrolled S0? emissions due
to the relatively higher sulfur content of residual fuel oil.  From
Table 7-4 it can be determined that baseline controls reduce SCL emissions
by 50 percent for boilers smaller than 73 MW (250 x 106 Btu/hr), and by
75" percent for boilers larger than 73 MW (250 x 106 Btu/hr).  As a result of
this difference, the emission reductions over baseline for the  most
stringent control alternative (Alternative 2) are relatively high for
boilers smaller than 73 MW.  For Alternative 2, an emission reduction over
baseline of 420 Mg/yr  (463 tons/yr) is shown for the 117 MW (400 x 106
Btu/hr) model boiler while a 420 Mg/yr (463 tons/yr) reduction  is also shown
for the 44 MW model boiler.  This represents percent reductions over
baseline of 60 and 80  percent for the large and small model boilers,
respectively.  As with the small coal-fired boilers, this is a  direct result
of the high percentage of emissions available for capture at baseline for
boilers smaller than 73 MW.  These boilers are not subject to the more
stringent NSPS for industrial boilers.
PM Emissions Reductions
     As with SOp emissions, the natural gas- and distillate oil-fired
boilers have low uncontrolled emissions and are not impacted by the control
alternatives.  There are no baseline controls required for residual
oil-fired boilers smaller than 73 MW.  Boilers larger than 73 MW are subject
to the existing industrial boiler NSPS which requires emission  reductions
from the uncontrolled  level.  The most stringent control alternatives
examined for residual oil-fired boilers result in incremental emission
reductions of from 50 to 78 percent over baseline.  As is the case with S02
                                     7-12

-------
emission reductions, the highest emission reductions occur for boilers not
subject to the existing NSPS for industrial boilers (smaller than 73 MW)
250 x 106 Btu/hr)).
     Baseline control alternatives for coal-fired model  boilers vary as a
function of boiler type and size range.  As a result, percent reductions
over baseline are variable as well, with the highest percent reductions
occurring for model boilers not subject to the existing  NSPS for industrial
boilers (boilers smaller than 73 MW).  Existing baseline controls for
coal-fired boilers reduce uncontrolled PM emissions by 50 to 99 percent,
with the highest percent removals occurring where electrostatic precipita-
tors or fabric filters are applied (boilers larger than  73 MW).  The most
stringent control alternatives evaluated result in emission reductions over
baseline of from 25 to 197 Mg/yr (28 to 217 tons/yr) for model  boilers
smaller than 73 MW (Alternative 5) and 48 Mg/yr (53 tons/yr) for model
boilers larger than 73 MW (Alternative 3).
NO  Emission Reductions
  A"""
     Unlike the S02 and PM impact analyses where several  control  alterna-
tives and control  methods were evaluated, the NO  impact analysis examines
                                                A
only a limited number of emission levels achievable through combustion
modifications.
     No NO  controls are required at the baseline control level  for model
                                    6
boilers smaller than 73 MW (250 x 10  Btu/hr) whereas boilers larger than
73 MW are subject  to the existing NSPS for industrial boilers.   In addition,
N0x controls are not applied to underfeed and chaingate stokers  since the
uncontrolled NO  emission rate for these boilers is less than the emission
               A
rate specified for any of the control alternatives examined.
     In general, combustion modification reduces NO  emissions  from small
                                                   A
(less than 73 MW)  coal-fired model boilers by 14 to 21 percent using low
excess air (LEA).   For the 117 MW (400 x 106 Btu/hr) pulverized  coal-fired
model boilers, Table 7-7 shows a 14 percent reduction over baseline using
staged combustion  air (SCA).  This represents an overall emission reduction
of 21 percent over the uncontrolled emission rate.  Emission reductions from
the oil- and gas-fired model boilers investigated range from 17  to
                                     7-13

-------
25 percent over baseline by using low excess air (LEA) and low excess
air/reduced air preheat (LEA/RAP).  A 25 percent reduction from the
uncontrolled level occurs for the 117 MW (400 x 10  Btu/hr) residual
oil-fired boiler.
     7.1.1.2  Model Boiler Dispersion Analysis.  In order to assess the
ambient air impacts of the model boiler air emissions, dispersion analyses
were performed by the Source Receptor Analysis Division of the EPA.  The
ambient levels of NO , S09 and PM, resulting from the model boilers were
                    A    C~
estimated, using the single source CRSTR model.  When the inputs (emission
rates, meteorological data) are accurate, the computed concentrations have
been found to have an uncertainty factor of about two.  Based on this
factor, the actual ambient concentrations could be greater by a maximum of
two or less by a minimum of i, than the concentrations calculated from the
CRSTR model.
     As a basis for the dispersion analysis, it was assumed that:
       •  The pollutants modeled displayed the dispersion behavior of a
          non-reactive gas.
       •  Sources were located on flat or gently rolling terrain in urban
          locations.
       •  Prevailing meteorological conditions were unfavorable to the
          dispersion of effluents.
       •  All model boiler stacks were modeled as continuous point sources
          of emissions.
       •  Receptors were located at plant grade (same elevation as the base
          of the stack).
        • All emissions were emitted from one stack.
          1964 meteorological data for St. Louis was used.
     The dispersion modeling results are presented in Table 7-9, and were
based on the emission rates presented in Tables 7-1 and 7-2.  However,
                                    7-14

-------
TABLE 7-9.  MODEL BOILER DISPERSION MODELING RESULTS
Maximum Downwind Ambient A1r Concentration at Averaging Period ug/m (10" gr/dscf)
NO,0 SO " Paniculate"
Control
Model Boiler Alternative Annual Mean Annual Mean Max.- 24 Hr. Annual Mean Max.- 24 Hr.
NG-30 UNC, UNC, UNC B mrna]
LEA, UNC, UNC 1 mrna1
DIS-30 UNC, UNC, UNC B nrnal
LEA, HDS, UNC 1 mrna'
RES-30 UNC, HDS, UNC B 1.820*:
LEA, HDS, HDS 1 1.380;
LEA, HDS, HDS 2 .1350*;
LEA, FGO(90)-Na, FGD 3 mrna!
LEA, FGD(90)-Na, ESP 4 mrna
mrna] mrna]
mrna ) mrna
mrna) nrna.
mrna ) mrna
.7859
.5959
.0583
mrna
mrna
HSC-30 UNC, CC, SM B mrna] ( mrna
LEA, CC, SSS 1 mrna.
LEA, CC, ESP 2 mrnai
LEA, FGD(90 , FGD 4 .2855^
LEA, FGD(90 , ESP 5 .2855°
LSC-30 UNC, UNC, SM B .2140?
LEA, UNC, SSS 1 .2140!
LEA, UNC, FF 2 .2140=
LEA, DS(50), DS 3 mrna}
LEA, FGD(90 , FGD 4 .2855;
LEA, FGDJ90), FF 5 .2855T
HSC-75 UNC, CC. SM B mrna]
LEA, CC, SSS 1 mrnaj
LEA, CC, ESP 2 mrna'
LEA, FGD(90), FGD 4 .5379*
LEA, FGD(90), ESP 5 .5379
LSC-75 UNC, UHC, SM B .3300;!
LEA, UNC, SSS 1 .3300»
LEA, UNC, FF 2 .3300?
mrna
mrna) nrna]
mrna) mrna
mrna) mrna]
mrna) mrna
7.330*: (3.165
3. 680; (1.589
1.380*; (.5959
77.90?.
38.93,
14.60*;
mrna, mrna) mrna'.
mrna ( mrna) mrna
mrna]
mrna.
mrna) mrna,,
.1233 .4670?
.1233) .4670°
.0924
.0924
.0924
mrna
.1233
.1233
mrna
mrna
mrna
mrna
mrna
mrna
.2017
.2017
mrna.
mrna.
mrna'
5.820,
5.820C
mrna) mrna]
mrna ) mrna
mrna) mrna.
mrna) mrna
33.64
16.81
6.304
1.040?
.4500,
.4500*;
mrna) mrna.
mrna
mrna
mrna
mrna
2.514
2.513
mrna
mrna-|
mrna.
mrnaj
.0807*!
.0404"
mrna) mrna]
mrna) mrna
mrna) mrna-.
mrna ) mrna
.4491
.1943
.1943
mrna
mrna
mma
mrna
mrna
0349
.0174
11.1<£
4.767,
4. 767?
mrna.
mrna
mrna
mrna
mrna
mrna
4.793
2.058
2.058
mrna
mrna
mrna-j ( mma
mrna, ( mrna
mrna,
1.007,
.5033C
.7210?, (.3113) 8.790*: (3.796) .2720*: (.1174) 3.310*:
.7210?,
.7210?
mrnal
.09701
.0970T
mrna]
mrnaJ
mrna'
.2323 .9480"
.2323) .9480e
.1425
.1425
.1425
LEA, DS(50) DS 3 mrna' { mrna
LEA, FGD(90 , FGD 4 .5379*
LEA, FGD(90 , FF 5 .5379e
.2323
.2323
NG-150 UNC, UNC, UNC B .3795? (.1638
LEA/RAP, UNC, UNC 1 mrna' ( mrna
DIS-150 UNC, UNC, UNC B .3790?
LEA/RAP, HDS, UNC 1 mrna'
RES- 150 UNC, HDS, UNC B .SOOO9
LEA, HDS, HDS 1 .4490*
LEA, HDS, HDS 2 .4490*
LEA. FGD(90), FGD 3 .8610=
.1636
mrna
.2591
.1939
.1938
.3719
1.2009,
1.200*
1.200?
mrna'
.2000?,
.2000e
.0015?
mrna
.8060?
mrna
2.400;>
1.1859
.44959
.9184!
.3113
.3113
mrna
.0419
.0419
8. 790,
8. 790*;
mrna'
1.210^
1.210C
mrna) mrna.
mrna
mrna
.4093
.4093
.5182
.5182
.5182
rnrna
.0864
.0864
mrna.
mrna
3.796
3.796
mrna
.5225
.5225
mrna
mrna
mrna
12.90^ (5.570
.1209^
.0302?
mma*
.0807,
.0403T
.0522
.0130
mrna
.0348
.0174
mrna. ( mrna
mrna.
mrna'
mrna
mrna
.1630*: .0704
12.90C (5.570) .0815C .0352
14.60*;
14.60u
14. 60^
6.304
6.304
6.304
mrna mrna
.45119
.20059
.0502?
mrna..
2.720, (1.174) .1630^
2.720C (1.174) .081 5e
.0006) .0194J
nrna) mrna
1.471;
.3678*;
____ 1
mrna
1.007;
.5033C
i
mrna.
mrna.
mrna
2.210C
mrna
.4350
.2173













1.429)
.6352)
.1588)
mrna)
.4348)
.2173)
mrna)
mrna)
mrna)
.9543)
1.105C (.4771)
.1949) 5.490£
.0866) 2.440P
.0217) .6100,
mrna
.0704
.0352
mrna
2.210^
1.105C
2.372
1.054
.2635
mrna


.9543)
.4771)
.0083) .10539, (.0454) .188o] (.0811)
mrna) mrna mrna) mrna ( mrna)
.3480) 9.890J (4.271) .0230?
mrna) mrna ( mrna) mrna
1.036) 29.00]
.5115) 14.93
.1940 5.60;
.3964) 12.91^
12.87
6.44
2.42
5.57
.34509,
.15009
.1500'
.2813*;
.0099) .2820]
mrna ) mrna
.1489
.0648
.0648
.1214
4.290]
1.860!
1.860'
3.93°
.1218)
mrna)
1.852
.8031
.8031
1.69



LEA, FGD 90 , ESP 4 ' .86106 .3719) .9184' .3964) 12.91" 5.57) .1407- .0607) 1.97' .8503)

-------
                                              TABLE 7-9.   (Continued)
•J e
Maximum Downwind Ambient Air Concentration at Averaging Period »ig/m (10" gr/dscf)
NO,0
Pnntrnl *-
SO,"
2
Model Boiler Alternative Annual Mean Annual Mean Max.- 24
HSC-150 UNC, CC, SM B mrna] mrna) mrna] m
LEA, CC, SSS 1 mrna'
LEA, CC, ESP 2 mrna1
mrna mrna. m
mrna mrna m
LEA, FGD(90), FGD 4 1.325® (.5721) .510® .6
LEA, FGD(90), ESP 5 1.3256 (.5721) .510e .6
LSC-150 UNC, UNC, SM B .7940-j
LEA, UNC, SS 1 .6302J
LEA, UNC, FF 2 .630K
LEA, DS(50), DS 3 mrna1
LEA, FGD(90), FGD 4 1.3076
LEA, FGD(90), FF 5 1.3076
RES-400 LEA, FGD(75), FGD B mrna]
LEA, HDS, HDS 1 mrna]
.3428) .500J .6
.2721) .500} .6
.2721) .50W (.6
mrna) mrna ( m
.5644) .3130^ (.1
rna) mrna, r
rna) mrnaj r
rna) mrna r
520) 21.60^ 9
520) 21.60C 9
477 18.70] (8
477 18.701 (8
477 18.70] (8.
rna mrna ( n
352) 4.490C (1
.5644) .3130° (.1352) 4.490C (1
Particulate"1



Hr. Annual Mean Max.- 24 Hr.
nrna mrna] f mrna) mrna.
nrna mrnaj I mrna) mrna^
nrna mrna ( mrna) mrna
327) .2577^ .1113) 3.700^
327) .1288e .0556) 1.850C
075) .7613J (.3287) 9.544^
075) .2538^ .1096 3.18l]
075) .0634^ .0274 .7953]
nrna) rnnra mrna mrna
939) .2550® (.1101 3.667*:
939) .1275e (.0550 1.833C
mrna) mrna, ( mrna) mrna, ( mrna) mrna, mrna) mrna.
mrna) mrna, i mrna) mrna, ( mrna) mrna, mrna) mrna.
LEA, FGD(90), ESP 2 mrna ( mrna) mrna1 ( mrna) mrna1 ( mrna) mrna1 mrna) mrna1 '
HSC-400 LEA, FGD(90)a,F60 B 2.1203
LEA/OFA, FGD(90), FGD 2 2.7209
LEA/OFA, FGD(90), ESP 3 2.7209
LSC-400 LEA, UNC, FF B 1. 140*1
LEA/OFA, DS(50), DS 1 mrna1
LEA/OFA, FGD(90), FGD 2 2.6709
LEA/OFA, FGD(90), FF 3 2.6709
.9154) 3.6403 (1.
1.174 2.6009 (1.
1.174) 2.6009 (1.
.4923) 1.940|[ f.8
mrna) mrna ( m
1.153) .53309 (.2
1.153) .53309 (.2
572) 51. OO] (2!
123) 38.201, (1(
123) 38.201 (1<
377 24.10? (1(
rna mrna ( r
301 7.820^ 3
301 7.820C 3
?.02) .3040^ .1313 4.260^ I
5.49) .45009 .1943 6.600^ i
5.49) .22509 .0972 3.300C i
| mrna)
mrna)
mrna)
1.598)
. 7988)
4.121)
1.374)
.3434)
mrna)
1.584)
.7918)
!mrna)
mrna)
mrna)
1.839)
2.850)
1.425)
).41) .1630![ (.0704) 2.030? (.8766
nrna) mrna ( mrna) mrna ( mrna
377) .44009 (.1900) 6.467^ (2.792
377) .22009 (.0950) 3.233C (1.396
'Partial scrubbing. ""National Primary Ambient Air Quality Standards for particulate matter:
0.3 Downwind distance from stack (km) Annual Mean 75 w/m3 (32.38 x 10 gr/dscf)
C0>5 „ Maximum - 24 Hr 260 v/m* (112.26 x 10'6 gr/dscf)
d, ft , "National Primary Ambient Air Quality Standards for SO,:
1 • U
e, ., „ Annual
Artihmetic Mean 80 u/m3 (34.54 x 10"6 gr/dscf)
f Maximum - 24 Hr 365 u/mj (157.60 x 10"° gr/dscf)
J'° " °National Primary Ambient Air Quality Standard for NO,:
"l fi n
h''° Annual
V7
10.8
^2.0 "
k, n »
Arithmetic Mean 100 v/m3 (43.18 x 10'6 gr/dscf)












1
 Modeling results not available.

-------
several  of the emission control  alternatives presented in Tables  7-1  and 7-2
are not represented in the dispersion analysis because of their subsequent
addition to the study after these results were generated.
     Table 7-9 presents the maximum downwind ambient air concentrations of
NO , SCL and PM over the same averaging times that are used to define the
Primary National  Ambient Air Quality Standards (NAAQS).   Downwind distances
from the stack to the receptor are indicated with footnotes.   A footnote to
Table 7-9 presents the Primary NAAQS for NO , S09 and PM. Comparison  of
                                           A    £
these values to the modeling results reported in Table 7-9 shows  that the
model boilers, located in a pristine atmosphere, comply with  the  NAAQS, as
described in the 1971 Federal Register.  For the alternative  control  levels
investigated the dispersion analysis shows that ground level  concentrations
of NO , S0?, and PM, range from 0.1 to 20 percent of the concentrations
     A    £-
specified by the ambient air quality standards.
     The dispersion analysis also shows the ambient air impact in going from
the baseline control level to a  more stringent control level.  As  an example,
for the 44 MW (150 x 10  Btu/hr) model boiler burning residual oil, the
24-hour average ground level concentration of S09 is reduced  by 81 percent
                                                               fi
in going from the baseline control level of 688 ng/J (1.6 lb/10  Btu) to a
more stringent emission control  level of 129 ng/J (0.3 lb/10   Btu).  In
addition, for the same model boiler, ambient PM concentrations are reduced
by 57 percent as the result of a 57 percent decrease in the PM emission
level and ambient NO  impacts are reduced by 25 percent corresponding to a
                    A
25 percent decrease in the NOV emission level.
                             A
     Results from the dispersion analysis indicate that where flue gas
reheat is not applied, the use of wet FGD scrubbers to control S02 emissions
can have an impact on the ground level concentrations of SO,,, PM, and NO .
                                                           ^             A
Wet FGD scrubbers cause a cooling of the flue gas which results in reduced
plume buoyancy.  When plume buoyancy is reduced, dispersion of the
pollutants in the upper atmosphere is inhibited and ground level  concen-
tration is increased.
     To illustrate the effect of reduced plume buoyancy on ground level
concentrations, the ground level concentrations of NO , S09, and PM are
                                                     /\    £
                                    7-17

-------
compared for the RES-150 model  boilers  using HDS and FGD to control S02
emissions (Alternatives 2 and  3,  respectively).  NO  and PM emission rates
                                                   A
are the same for both alternatives while  the S02 emission rates vary by only
6 percent (0.32 lb/106 Btu  for Alternative  3, 0.30 lb/106 Btu for
Alternative 2).  Table 7-9  shows  that when  the wet FGD system is used,
ground level N0x, S02, and  PM  concentrations are about 2 times greater than
the concentrations associated  with using  HDS oil.  In addition, the modeling
results show that the higher ground  level concentrations associated with FGD
applications occur at receptors which are located close to the stack when
compared with  the receptor  distances where  HDS oil is used.  It should be
noted that the use of FGD does not increase the ground level concentrations
of S02 or PM over those values that  are estimated to represent the current
model boiler ambient air  impacts  (impacts at baseline).  However, NO
                                                                    A
concentrations do tend  to  increase by 40  to 50 percent over baseline with
the application of LEA  and  SCA controls.
7.1.2  Secondary Air  Impacts.
     Secondary air emissions result  from power boilers supplying electricity
to the industrial boiler  control  devices.  The power required to operate
pollution control equipment will  ultimately result in greater emissions at
the electric power generation  facility.
     Tables 7-10 and 7-11  present the estimated incremental NO , S09, and PM
                                                              A    £
emissions from a coal-fired electric power  generation facility supplying
power to operate model  boiler  pollution controls.  Natural gas-fired model
boilers are not  included  since virtually no electrical energy is required
for pollution  control.   Incremental  N0x,  SO,,, and PM emissions at the power
generating facility were  calculated  using the control device power require-
ment and assuming that  the  power  boilers  comply with the NSPS for utility
boilers.
     Tables 7-10 and  7-11  show that  the emissions caused by auxiliary power
generation are very  small when compared to  the emission reductions from the
model boilers  that were presented in Section 7.1.1.  For example, a 117 MW
 (400 x 10  Btu/hr) model  boiler burning high sulfur coal with S02 and PM
emissions controlled  to the most  stringent  Alternative 3 level would
                                      7-18

-------
                    TABLE  7-10.   SECONDARY  AIR  POLLUTION IMPACTS  FOR  COAL-FIRED MODEL BOILERS

Power Boiler Emissions [Mg/yr (tons/yr)]
Boiler Specie Baseline Alt. lc Alt.
HSC-30 NO .12
SO? .17
PV\d .01
HSC-75 NO .30 (
SO? .45 (
PMZ .02 (
HSC-150 NO .61
SO? .91 1.
PHZ .04 (
HSC-400 NO 2.73 3.
SO? 4.10 4.
WF .16 .
LSC-30 NO .12 (.
sol -17 •
PMZ .01 ( .
LSC-75 NO .30 .
SO? .45 .
PMZ .02 .
LSC-150 NO .61 .
SO? .91 1.
PMZ .04 ( .
LSC-400 NO 2.12 (2.
SO? 3.19 (3.
PMZ .13 (.
13) .15 (.17) -04 (
.19) .23 (.25) .05
01) .01 (.01) 0
33) .42 (.46) -12 (
50} .63 (.69) -17
02) .03 (.03) -01
67 .80 (.88) -26 (
0 1.20 (1.32) -39
04) .05 (.05) -02 (
2 Alt. 3 Alt. 4 Alt. 5
.04) .19 ( .2
.06) a .28 ( .3
0 ) .01 ( .0
1) .23 ( .25)
1 .35 ( .38)
1 .02 ( .02)
.13) .49 ( .54) .16 ( .67)
.19) a .74 ( .81) .91 (1.0 )
.01) .03 ( .03) .04 ( .04)
.29) .87 ( .9
.43) a 1.31 (1.4
.02) .05 ( .C
01 5.08 (5.60) 5.77 (6.35)
51 a 7.62 (8.39) 8.64 9.52) b
18 .31 ( .34) .35 .38)
13) .15 (.17) -19 (
19 .23 (.25) -28 i
01 .01 (.01) -01 I
33 .38 (.42) -45 i
50 .57 (.63) -68 i
02) .03 (.03) -03 (
67) .76 .84) -91
0 ) 1.14 (1.25) 1-36
04) .05 (.05) -05
6) 1.14 (1.25
4) 1.71 (1.88
6) .07 ( .08)
b
.21) .23 ( .25) .19 ( .21) .38 ( .42)
.31) .35 ( .38) .28 ( .31) .57 ( .63)
.01) .02 ( .02) .01 ( .01) .03 ( .03)
.50) .68 ( .75) .45 ( .50) .91 (1.0
.75) 1.03 (1.13) .68 ( .75) 1.36 (1.5
.03) .05 ( .05) .03 { .03) .05 ( .06
1.0 ) 1.03 1.13) .80 ( .f
1.5) 1.53 1.69) 1.20 (1.:
.06) .06 .07) .05 ( .(
34) 5.04 5.55) 4.81 (5.30) 6.94 ( 7.64)
51) 7.56 8.33) 7.33 (7.95) 10.41 (11.46) b
14) .30 .33) .30 ( .32) .42 ( .46)
J8 .80 ( .88)
12 1.20 (1.32)
)5 .05 ( .05)
b
 50% scrubbing not applied since this will not meet the baseline emission limit.
faThere are no alternatives 4 and 5 for the 117 MW  (400 > 106 Btu/hr) boilers.

-------
                                   TABLE 7-11.
SECONDARY AIR POLLUTION IMPACTS FOR

GAS-  AND OIL-FIRED MODEL BOILERS
i
ro
o
Model Emission
Boiler Specie
RES-30


RES- 150


RES-400


DIS-30


D IS- 150


N0x
so2
PM
N0x
so2
PM
N0x
so2
PM
N0x
S02
PM
N0x
so2
PM
Power Boiler Emissions [Mg/yr (tons/yr)]
Baseline
.17
.25
.01
.88
.32
.05
1.60
2.40
.10
0
0
0
0
0
0
( .19)
( -28)
( -01)
( .97)
( -35)
( -06)
(1.76)
(2.65)
( .11)
(0)
(0)
(0)
(0)
(0)
(0)
Alt. 1
.38
.56
.02
1.92
2.88
.12
5.68
8.51
.34
.08
.13
.01
.22
.33
.01
( .42)
( .62)
( .02)
( 2.11)
( 3.17)
( .13)
( 6.26)
( 9.38)
( 0.37)
( .09)
( .14)
( .01)
( -24)
( .37)
( .01)
Alt. 2
.50 { .55)
.75 ( .83)
.03 ( .03)
,2.46 (2.71)
3.70 (4.07)
.15 ( .17)
2.18 (2.40)
3.27 (3.60)
.13 (0.14)






Alt. 3 Alt. 4
.46 ( .46)
.69 ( .69) DNAa
.03 ( .03)
.72 ( .67) 1.03 (1.13)
1.08 (1.00) 1.53 (1.69)
.04 ( .04) .06 ( .07)









             Data not available (See Energy Impacts Table 7-15).

-------
indirectly result in the following incremental  emissions  from the  power
boiler:
               N0x   -    5.47  Mg/yr  (6.02  tons/yr)
               S02   -    8.20  Mg/yr  (9.03  tons/yr)
               PM    -    0.3 Mg/yr  (0.33 tons/yr)
These power boiler emissions would be offset  by the  following emission
reductions from the 117 MW (400  x  10  Btu/hr)  coal-fired  model  boiler.
               NOX   -   152.7  Mg/yr  (168.2 tons/yr)
               S02   -    4898  Mg/yr  (5394  tons/yr)
               PM    -    6814  Mg/yr  (7504  tons/yr)
     A similar relationship between power boiler emissions  and model  boiler
emission reductions is  evident for all  other  model boilers.
7.2  LIQUID WASTE IMPACTS
     Water pollution impacts or  the need for  additional water treatment can
result from controlling industrial  boiler air emissions if  the control
technologies used to achieve the various control  levels produce aqueous
discharge streams.  Control  of NO   by combustion modification, as  discussed
                                A
in Chapter 4, does not  result  in aqueous discharges.   Likewise, control of
PM or S02 emissions by  use of  hydrodesulfurized fuel  oils or low sulfur coal
does not result in any  waste water streams.   Consequently these technologies
are not considered further.
     Dry particulate controls  (ESP, FF, MC) also do  not result in  water
discharges, but incremental  water  pollution impacts  from  PM controls  can
result if the collected particulate material  is sluiced to  disposal  ponds.
However, the sluiced ash stream  from a PM control  device  can be treated in
                                                                           p
existing facilities along with the boiler ash stream, and the water  reused.
     The control  of S02 by FGD can result in  liquid  waste discharges  while
dry scrubbing processes are designed not to generate liquid wastes.   Once-
through sodium scrubbing systems (FGD/Na) result in  direct  liquid  discharges
of sodium sulfite/sulfate salts.  Dual-alkali,  lime  and limestone  FGD
systems can be designed on a closed-loop basis  so that the  only water losses
during normal operation occur  with the sludge going  to landfill.  Purging of
                                    7-21

-------
either of these systems due to water imbalances or other operating upsets,
system blowdown to prevent scaling, or operator error will result in
discharge of an. aqueous waste stream which can be contained and treated.
However, during normal operation, there should be no water pollution impact
from lime, limestone and dual-alkali (FGD) systems designed on a closed-
                           o
loop, zero discharge basis.
     Since the sodium throwaway  (once-through) system is the primary system
resulting in liquid discharges,  the remainder of this section focuses on  the
water pollution impacts of the sodium throwaway FGD system.  Potential  water
pollution impacts were assessed  by considering the following:
     • effluent quantity and characteristics,
     • effluent treatment and disposal, and
     • applicable regulations.

7.2.1  Effluent Quantities and Characteristics
     Aqueous emissions from a sodium throwaway FGD process contain about
5 percent (by weight) dissolved  solids.   The composition of the dissolved
solids will vary depending upon  the extent of sulfite oxidation; with the
major compounds consisting of sodium sulfate (Na^O^), sodium sulfite
(Na2S03), and sodium carbonate (Na^CO.,).  In addition, chlorides and trace
elements absorbed from the flue  gas will be present in smaller amounts.
     The only model boiler using a sodium throwaway FGD control (FGD/Na)  is
the  RES-30 model boiler.  This model boiler is typical of oil-field steam
generators where the majority of the sodium throwaway systems are used.
FGD/Na systems were not considered for other model boilers because potential
regulations may limit the discharge of FGD/Na effluents into municipal  water
treatment facilities.  The estimated liquid waste impacts resulting from
applying Alternatives 3 and 4 (90 percent S02 control) to the RES-30 model
boiler are effluent discharge rates of 25.3 liters/min (6.7 gpm).  On an
annual basis the discharge rate  is 8.0 x 10  ji/yr (1.88 x 10  gal/yr).
     Table 7-12 shows effluent discharge rates for HSC boilers using FGD/Na
systems to reduce S02 emissions  by 90 percent.  As can be seen, for a fixed
                                     7-22

-------
  TABLE 7-12.  WATER POLLUTION IMPACTS FOR THE SODIUM THROWAWAY SYSTEM

Model Boiler
Heat Input
MW
8.8
22
44
117
(106 Btu/hr)
(30)
(75)
(150)
(400)
Fuel Type3
HSC
HSC
HSC
Pulverized HSC
Effluent Discharge Rate
Ji/Min
51.0
130.8
262.8
702.0
(gpm)
13.4
32.6
69.4
185.0
Average Dissolved  Solid  Compositions     Na2S03  -  77 percent
                                        Na2S04  -   9 percent
                                        Na9CO,  -  14 percent
                                          {.   O
aHSC is 3.5% sulfur
 Based on 90% removal  of S02
                                 7-23

-------
control level the effluent discharge rate increases directly with the boiler
heat input capacity.  Therefore, with a 100 percent increase in boiler heat
input capacity, (22 MW to 44 MW) the discharge rate could also be expected
to increase by approximately 100 percent.
7.2.2  Effluent Treatment and Disposal
     The dissolved solids content and pH-imbalances are the two main areas
of concern for which treatment may be required for wastes from a sodium
scrubbing system.  Discharge to an evaporation pond or to an existing
centralized wastewater treatment facility is commonly practiced.  Of the
102 sodium scrubbings systems in use today, about 80 use evaporation ponds
(over 30 of these in conjunction with well injection), and 10 use
centralized water treatment for disposal of FGD wastes.
     If the scrubber effluents are being discharged directly to a receiving
stream, the water quality standards applicable to that stream will govern
the degree of treatment required.  Also, if the scrubber effluents are
discharged to a publicly owned treatment works (POTW), then the pretreatment
requirements contained in the guidelines for that POTW will determine the
degree of treatment necessary.  Treatment methods available to reduce total
dissolved solids include: ion exchange, electrodialysis, reverse osmosis,
and distillation.   Neutralization of the wastewater may be necessary to
achieve proper pH.  The treatment method employed at a centralized treatment
facility will depend upon the characteristics of the industry's process
waste streams with which the scrubber effluent is being combined.
     Some industries (e.g., textile and paper mills) can use process waste
streams containing sodium as a feed to the scrubber.  The aqueous stream
from the FGD system is then recombined with the industrial process waste
streams and discharged to an on-site centralized waste treatment facility.
The treatment processes in such a centralized treatment facility vary with
the specific industry.  Typically, the treatment is designed to remove the
dissolved and suspended solids and attain a neutral pH.
     The adverse impacts of discharging aqueous scrubber wastes to the
environment are potential degradation of the water quality (both surface and
ground) of the receiving stream and the subsequent impact on users of that
                                    7-24

-------
water.  Improper treatment or disposal  practices can allow aqueous wastes
with high total  dissolved solids to be  introduced into streams and aquifers
that may serve as sources of water for  other users.
7.2.3  Applicable Regulations
     The applicable regulations relative to liquid waste discharge will  be
governed by the disposal technique being used.   Discharges to a surface
facility (a receiving stream, centralized industrial wastewater treatment
facility or publicly owned treatment works) will have to satisfy the
requirements of the Clean Water Act. Disposal  by deep-well  injection must
satisfy requirements of the Safe Drinking Water Act and the Underground
                          o
Injection Control Program.
     If the scrubber effluent is discharged to  an on-site centralized
industrial wastewater treatment facility and is treated along with other
process waste streams, specific effluent standards applicable to the
industry with which the boiler facility is associated must be satisfied.
The scrubber effluent will be considered a contributing source to the
central treatment facility and will be  listed as such on the National
Pollutant Discharge Elimination System  (NPDES)  permit for that treatment
facility.
     When the scrubber effluent is discharged directly to a waterway, the
effluent must be treated to meet specific instream water quality standards
at the location of the discharge.  If the scrubber discharge is directed to
a POTW, pretreatment guidelines must be met, so that these effluents do  not
interfere with the operation and performance of the POTW.
     When effluents are being disposed  of by well injection, steps must  be
taken to assure that contamination of any drinking water sources does not
occur.  The Underground Injection Control Program proposed regulations
specify the procedures to be followed to protect any sources of drinking
                                                                o
water and specify how those sources of  water will be identified.
7.3  SOLID WASTE DISPOSAL IMPACTS
     Industrial  boiler air pollution control techniques produce two main
types of solid wastes: fly ash collected by the PM control devices, and
                                    7-25

-------
waste solids (both sludge and dry scrubbing products) from the control  of
SQ^ emissions.  No incremental solid waste results from NO  emission control
by combustion modification.  In this section, the impacts of the incremental
solid wastes produced from PM and SCL controls are discussed by considering
the following:
        • solid waste quantities and characteristics,
        • waste treatment and disposal, and
        • applicable regulations.

7.3.1  Solid Waste Quantities and Characteristics
     The primary constituents of coal fly ash are silicon, aluminum, iron,
and calcium, with lesser quantities of magnesium, titanium, sodium,
potassium, sulfur, and phosphorus.  In addition, fly ash contains trace
concentrations of from 20 to 50 elements (depending on the specific coal),
including lead, arsenic, and cadmium, and radionuclides of several
         g
elements.
     Dual alkali scrubber sludges are composed primarily of calcium sulfite/
sulfate solids.  Also present are dissolved sodium salts and trace elements
(e.g., lead, arsenic and cadmium), which may contaminate the groundwaters
and surface waters due to runoff and leaching from sludge disposal sites
(see Section /.2.3).  The,chemical composition and concentration of FGD
sludge varies with the different coal types used in industrial boilers.
When a particulate collection device is not used upstream of the FGD system
and the FGD system is being used to control both SO,, and PM emissions,  the
trace element concentrations in the scrubber sludge are increased due to the
addition of fly ash to the sludge.
     The dry solid waste produced from dry scrubbing FGD processes consists
primarily of calcium or sodium salts, depending upon the type of alkali used
as the S02 sorbent.  Significant quantities of fly ash will also be present,
because the PM collection device is located downstream of the spray dryer
and removes fly ash along with the spray dried solids.
                                     7-26

-------
     Tables 7-13 and 7-14 show the quantities of solid wastes for each of
the model boilers equipped with controls that result in a solid waste.
Waste production rates are graphically illustrated in Figure 7-1  for
coal-fired model boilers.  For the ESP, FF, and mechanical  collector control
techniques, the solid waste quantities presented are the quantities of fly
ash collected on a dry basis.   For SCL control, the quantities of sludge
presented are for a FGD system with a sludge quality of 60  percent solids.
The solid wastes shown for the dry scrubbing (DS) control consist of fly
ash, sulfate/sulfite salts, and unreacted sorbent collected by the fabric
filter system downstream of the dry SCL scrubbing system.  Sludge quantities
presented for the combined S02/PM systems are based on a sludge concen-
tration of 60 percent solids and include fly ash collected  in the scrubber.
     Table 7-13 and Figure 7-1 can be used to show the relative increase in
solid waste resulting from increasingly stringent control alternatives.  For
example, the HSC-30 model boiler exhibits more than a 10 fold increase in
going from baseline to control Alternative 5, while the LSC-30 model  boiler
shows about a 5 fold increase.  In both cases, this solid waste increase can
be attributed primarily to the FGD system applied at the more stringent
control Alternatives 4 and 5.   In general, increases such as the ones demon-
strated by this example can be expected for boilers where no baseline FGD
systems are required.  This includes all  LSC and HSC boilers smaller than 73
MW (250 x 106 Btu/hr), and RES boilers smaller than 73 MW.   In addition,
Table 7-13 illustrates that where FGD is used to collect fly ash as well as
S02» (i.e., Alternative 4 for coal-fired boilers, Alternative 3 for
oil-fired boilers) overall solid waste loading will increase over systems
that collect fly ash by dry collection methods (i.e., Alternative 5 for
coal-fired boilers, Alternative 4 for oil-fired boilers).  This is a direct
result of the water associated with the fly ash collected in a wet FGD
system.  For the combined S02/PM FGD systems, 80 percent of the total fly
ash is removed with an upstream mechanical collector, while the remaining
fly ash is collected in the FGD system to meet the PM emission limit of 43
ng/J (0.1 lb/106 Btu).  The fly ash collected by the FGD unit results in a
solid waste that is 60 percent solids (40 percent water).  This additional
                                     7-27

-------
                                     TABLE 7-13.   SOLID  WASTE IMPACTS  FROM  COAL-FIRED MODEL  BOILERS
--4
 I
ro
CO
Model
Boiler
HSC-30


HCS-75


HSC-150


HSC-400


LSC-30


LSC-75


LSC-150

LSC-400


Type of
Solid
Waste
Fly Ash
Sludge
Total
Fly Ash
Sludge
Total
Fly Ash
Sludge
Total
Fly Ash
Sludge
Total
Fly Ash
Sludge
Total
Fly Ash
Sludge
Total
Fly Ash
Sludge
Total
Fly Ash
Sludge
Total
Amount of Solid Haste,
Baseline
no

130
326 |

326
143)

143)
359)

359)
1867 (2058)

1867
6767
12560
19327
71

71
179

179

(2058)
7452)
13832)
21284)
( 78)

( 78)
197)

197)
1090 (1202)
1090 (1202)
4200 (4625)


4200 (4625)
AH.
145

145
362

362
2012

2012

f

86

86
215

215
1235
1235
4876c>e

4876C>e
1
( 160)

( 160)
( 399)

( 399)
(2216)

(2216)



( 95)

( 95)
( 237)

( 237)
(1360)
(1360)
(5370)

(5370)
Alt.
155

155
388

388
2065

2065
541 4K
22420°
27834
97

97
242

242
1289
1289
3360u
10606°
13966
2
( 171)

( 171)
( 427)

( 427)
(2274)

(2274)
(5961)
(24686)
(30648)
( 107)

( 107)
( 267)

( 267)
(1420)
(1420)
(3700)
(11678)
(15377)
Hg/yr ( ton/yr )a
AH. 3 Alt. 4

122^ (
Alt.
134) 155
f 1539" (1695) 1489
1661 (1829) 1644

303!1 (
334) 388
f 3907" (4302) 3780
4210 (4636) 4168



6815
20164
26979
161C

161C
640C

640C
1873C
1873C
4248
9206
13454
1638^ (1
f 8182° (9
9820 (1
(7505)
(22207) g
(29712)
fd( 177) 74b
, 323b
>d( 177) 397
>d( 704) m*
. 799b
804) 2065
5
( 171)
(1640)
(1811)
( 427)
[4163)
(4590)
(2274)
009) 7500 (8260)
0815) 9565 (10534)

g

81) 97
356) 291
437) 288
204) 242
880 721
•d( 704) 985 (1085) 963
fd(2063) 10171; (1
x 2028° (2
•"(2063) 3044 (3
(4678)
(10139) g
(14817)



[ 107)
321)
317)
267)
794)
1061)
120 1289 (1420)
233 1605 (1768)
352 2894 (3187)

g




                aFly Ash  -  Mg/yr (ton/yr), dry basis; Sludge - Hg/yr  (ton/yr)  060 percent  solids.
                 Scrubber also removes  fly ash.   This fly ash is included with the sludge
                 960 percent solids.
                cTotal fly ash and alkali salts.
                 Sodium sorbent.
                eL1me sorbent.
                 50 percent  removal alternative  not applicable for HSC  boilers since this removal
                  would not meet baseline emission limit.
                9No alternative 4 and 5 for 117  HW boilers.
                 SOX of the  total fly ash is collected dry by a mechanical collector upstream of the scrubber.

-------
                              TABLE 7-14.  SOLID WASTE IMPACTS FROM OIL-FIRED MODEL BOILERS
i
ro
10


Model
Boiler

RES- 30


RES-150


RES-400


Type of
Solid
Waste

Fly Ash
SI udge
Total
Fly Ash
Sludge
Total
Fly Ash
Sludge
Total
Amount of Solid


Baseline Alt. 1
.A
c c


c c


91b ( 100>b 91b < 10°)d
8440° (9296)° 10120° (11147)
8531 (9397) 10211 (11247)
Waste,


Alt







157.6
10082
10240
Mg/yr (tons/yr)a


. 2 Alt. 3

c c


c
3821 ( 4208)
3821 ( 4208)
(173.6)
(11104)
(11278)



Al

11.8

11.8
59.1
3756
3815






t. 4

(13.0)

(13.0)
(65.1)
(4137)
(4202)



      aFly Ash - Mg/yr (ton/yr); Sludge - Mg/yr (ton/yr) @60% solids.

      DFly Ash included @60% solids.

      cNo  solid wastes generated.

       80% of total  fly ash is removed in an upstream mechanical  collector.

-------
                        30000  -
                        25000  -
                        20000  -
u>
o
>-

O»




O

+J
o
3

o
J-
o.

01
4->
in
                O
                10
                         15000  -
                         10000  *
                          5000
                                                                                             HSC Model  Boilers

                                                                                               (90X FGD,ESP)a
                                                                                                                  LSC Model Boilers

                                                                                                                  (90% FGD,  FF)a
                                                                           HSC Model Boilers (CC, ESP)U

                                                                           LSC Model Boilers (UNC, FF)C
                                                 22
                                                (75)
                                                 r
                                                 44
                                               (150)
117.2
(400)
                                                              Boiler Size MW (10b BTU/Hr)
                                             Figure  7-1.   Solid Waste Production (Fly Ash and  Sludge)
                a)   90% S02 removal via Double Alkali  FGD
                    S02 control via compliance coal, Participate control  via  Electrostatic Precipitator.
                    Uncontrolled S02, Particulate control via Fabric Filter

-------
water Increases the solid waste loading over that resulting from systems
that accumulate a dry solid waste.
     Figure 7-1 shows that where no FGD systems are required (PM control by
SSS/DM, FF, ESP), percent increases over baseline of from 10 to 40 percent
are typical for coal-fired boilers.  The parabolic shape of the curves shown
in Figure 7-1 can be attributed to the nonlinear increase with respect to
boiler size in collectable particulate emissions from coal-fired model
boilers.  This can be seen in Table 7-1 where the uncontrolled emission
rates from spreader stokers and pulverized coal units are greater than those
from underfeed and chaingrate units, on a lb/10  Btu basis.  It should be
noted that these curves are simplified somewhat and as such, should be used
to illustrate solid waste loading trends only.   Any one curve is actually a
set of discontinuous straight lines, with the slopes of these lines
increasing with boiler size.  The discontinuity occurs at the boiler sizes
that correspond to a change in boiler type (i.e., chaingrate to spreader
stoker) and subsequent change in uncontrolled particulate emissions.
7.3.2  Waste Treatment and Disposal
     Ponding and landfill ing are currently the  primary methods used for
disposing of collected fly ash.  An alternative to landfill ing is the
commercial utilization of fly ash in road embankments, concrete mixture, and
sludge stabilization.    Current Federal, State and local regulations would
govern the disposal practices at the landfills.
     Solid wastes from spray dryers (dry scrubbing) may be handled in the
same manner as fly ash.  Off-site landfill ing has been selected as the
disposal method for the first two dry scrubbing systems installed on indus-
trial boilers.11
     The main sludge disposal options for wet FGD systems include ponding
and landfill ing.  Ponding is the simpler of the two methods, but is poten-
tially more harmful to the environment than landfill ing.  Ponding involves
slurrying the sludge to a pond, allowing it to settle and pumping the
supernatant liquor either to a treatment process or back to the facility for
reuse.  Because there is always a hydraulic head on the waste in the bottom
of the pond, the potential for leachates reaching ground-water sources
                                     7-31

-------
beneath the pond is greater than for a landfill.  Use of the pond area may
be limited after disposal ceases, mainly because of the poor load bearing
                                                                   I O
capabilities of the sludge compared to the original soil structure.
     Landfill disposal of FGD wastes in a specially prepared site requires
some processing of the wet scrubber sludge (either stabilization or
fixation) to obtain a soil-like material that may be loaded, transported and
placed as fill.  Stabilization refers to the addition of fly ash or other
similar material to the  sludge to produce only physical changes without any
chemical reactons.  Fixation is a type of a stabilization which involves the
addition of reagents  (such as lime) to cause chemical reactions with the
       13
sludge.    The objective of these treatment methods is to increase the load
bearing capacity of the  raw sludge and to decrease the permeability and
correspondingly the mass transport rate of contaminants leaching out of the
sludge.14
     Proper design of both ponds and landfills is required to assure minimum
environmental impact  of  the solid waste disposal.  Contaminants that are
contained in ponds and landfills or accidentally spilled on the surface can
enter ground-water systems by leakage or leaching.  As the term implies,
leakage refers to migration of fluids that are deposited on the surface to
the subsurface.  Leakage is of more concern for ponds and spills than
landfills.  Leaching, on the other hand, denotes the introduction of water
(usually infiltrating precipitation) into the waste after it has been
landfilled, so that contaminants are dissolved and elutriated or leached out
of the solid material.
     Transport of trace elements and other potential pollutants from the
disposal site via leaching or run off is determined by many factors,
including: (1) the chemical form and concentration of the potential
pollutant in the waste,  (2) the permeability, sorption capacity, and
porosity of the substrate, (3) soil and leachate pH, (4) the permeability
and porosity of the waste, (5) the proximity of the disposal site to the
ground-water table and/or surface water, (6) the presence or absence of clay
or plastic liners or  other methods of enclosing the wastes in materials of
low permeability, and (7) climatic factors such as precipitation,
                                     7-32

-------
                                   15
temperature, and relative humidity.     However, if a landfill  site is
properly designed and operated, these leaching and runoff problems can be
averted and the landfill area eventually reused either for recreational or
building use purposes.
7.3.3  Waste Disposal Regulations
     At the present time, the regulations governing solid waste disposal  are
not fully defined.  EPA recently (May 2, 1980) issued Phase I  final  RCRA
regulations covering the framework for management of solid wastes.    In
addition, Congress is currently considering legislation that would exempt
certain "special wastes" (as defined in the proposed regulations)  from the
possibility of being classified as hazardous until  more data are gathered
about their characteristics (2 to 3  years).
     The Phase I RCRA regulations exempt fly ash, bottom ash,  slag,  and air
pollutant emission control  sludge produced in the combustion of fossil fuels
from consideration as hazardous wastes.  This exemption applies to
industrial  boiler FGD sludges.
     Since  the wastes are currently  exempt from hazardous waste regulations,
they may be considered non-hazardous.   Non-hazardous waste disposal  manage-
ment and techniques will be governed by Section 4004 of RCRA.   This  section
requires states to implement disposal  programs that will protect the
environment (especially ground water)  from contamination.  EPA has also
published Landfill Disposal of Solid Waste, Proposed Guidelines that wi 11
act as a guide to the states as to what their disposal  management  programs
               18
should contain.
     Disposal  of non-hazardous wastes will  require at a minimum that a clay
liner be used at the disposal site,  that daily cover be applied, that access
to the site be controlled,  that ground-water quality at the site boundary be
monitored,  and that a final impermeable cover be placed and revegetation
      18
occur.    These activities  are required, primarily, to protect ground water
in the disposal area.
                                     7-33

-------
7.4  ENERGY IMPACT OF CONTROL TECHNOLOGIES
     All control systems used for PM, and S02 emission control will require
electrical energy.  The major portion of the electrical energy is needed to
operate the fans installed to overcome  the pressure drop across control
systems.  Lesser amounts of electrical  energy are needed for motors that
operate the pumps in wet scrubbing  systems and bag cleaning mechanisms in
fabric filters.  For ESP's, energy  is also required to create the corona
discharge and to run auxiliary equipment such as collection plate
        19 20
rappers.  '    Use of HDS cleaned oils  results in energy penalties in the
form of power requirements at the HDS facility.  These energy requirements
include demands for electricity, fuel and steam at the hydrogen plant, oil
heaters and miscellaneous processes.  Use of low excess air (LEA) for NO
                                                                        A
emission control results in improved boiler efficiency, and therefore an
overall net gain in energy for the  industrial boiler.  However, use of other
NO  combustion modifications techniques [flue gas recirculation (FGR), and
  x                                                         21
staged combustion air (SCA)] may result in energy penalties.    The energy
requirements for combined S02/PM systems include energy for operating the
wet scrubber along with the energy  associated with slurry pumping and sludge
handling.  Combined SO^/PM systems  use  venturi scrubber configurations with
an estimated pressure drop of 5 kPa  (20 inches of water).
     Table 7-15 shows the annual energy demand of the control devices
associated with each model boiler.  The steam and electrical demands are
expressed in thermal megawatts and  10   Btu/hr of net heat input to the model
boiler.  Control device energy demands  were derived from information
supplied in the Individual Technology Assessment Reports for each control
method.  A 33 percent heat to electrical energy conversion efficiency was
assumed.
     Hydrodesulfurization of fuel oil is shown to be the most energy inten-
sive control technology considered.  For all residual oil-fired boilers,
2.4 percent of the boiler heat input is required to achieve 50 percent
sulfur reduction while 5.8 percent  is required for 75 percent reduction and
8.0 percent is required to achieve  90 percent reduction.  In comparison, for
                                      7-34

-------
                                 TABLE  7-15.   MODEL  BOILER ENERGY REQUIREMENTS
CO
01
Model Boiler3
HSC-30




HSC-75




HSC-1BO




HSC-400


LSC-30




LSC-75




LSC-150





LSC-400



- Unc,
Unc,
Unc,
Unc,
Unc,
- Unc,
Unc,
Unc,
Unc,
Unc,
- SCA,
SCA,
SCA,
SCA,
SCA,
- LEA,
SCA,
SCA,
- Unc,
Unc,
Unc,
Unc,
Unc,
Unc,
- Unc,
Unc,
Unc,
Unc,
Unc,
Unc,
- Unc,
SCA,
SCA,
SCA,
SCA,
SCA,
- LEA,
SCA,
SCA,
SCA,
CC, SM
CC, SSS
CC, ESP
FGD (90)
FGD (90)
CC, SM
CC, SSS
CC, ESP
FGD (90)
FGD (90]
CC, SM
CC, SSS
CC, ESP
FGD (90]
FGD (90]
FGD (90)
FGD (90)
FGD (90)
Unc, SM
Unc, SSS
Unc, FF
DS (50),
FGD (90)
FGD (90)
Unc, SM
Unc, SSS
Unc, FF
DS (50),
FGD (90)
FGD (90)
Unc, SM
Unc, SSS
Unc, FF
DS (50),
FGD (90)
FGD (90)
Unc, FF
DS (50),
FGD (90)
FGD (90)


i
, FGD1
, ESP


4
, FGD1
, ESP


4
, FGD1
, ESP
, FGD]
, FGD1
, ESP

4
DS1 .
, FGD1
, FF

J
DS1 ,
, FGD1
, FF


4
DS1 .
, FGD1
, FF
i
DS1 .
, FGD1
, FF
Control
Alternative
Basel ine
1
2
4
5
Baseline
1
2
4
5
Baseline
1
2
4
5
Baseline
2 .73
3 .73
Baseline
1
2
3
4
5
Baseline
1
2
3
4
5
Baseline
1
2
3
4
5
Baseline
1 .73
2 .73
3 .73
Energy Demand MWt (106
NOX










0
0
0
0
0
09
(2.48)
(2.48)











0
0
0
0
0
0"
(2.48)
2.48)
(2.48)
so2



.11
.05



.27
.13



.53 1.
.23
.54 1.
1.26 4.
.61 2.


.15 (
.11 .
.05 { .


.32 (1.
.26
.12


h
,60h 2.
.48 1.
.21

1.35 (4.
1.21 4.
.54 1.

.03
.04
.005
37
17 .009
.08
.11
.03
92
44 .03
.16
.21
.07
80)
78) .07
84 .11
28
07 .14
.03
.04
.05
51
37
17 .05
.08
.10
.12
09)
88
41) .12
.16
.20
h '24
04)"
63
71) .24
.56
59)
11
84 .56
Btu/hr) h
PM
( .10)
.14)
(.015J
c
(.027)
( .27)
( .37)
( .10)
c
( .10)
( -54)
( -71)
( -24)
c
( -24)
( .33)
c
( .42)
( .10)
( .14)
( -17)
c
c
( .17)
( .27)
( .34)
( .41)
c
c
( -41)
( -54)
( .68)
( -82)
c
c
( .82)
(1.90)
c
c
(1.90)
Percent of
N0x S02



1.25
.57



1.23
.59



1.21
.52
.46
.30 1.08
.30 .52


1.70
1.25
.57


1.45
1.18
.55



1.36
1.10
.48

.62 1.15
.62 1.03
.62 .46
Boiler Heat
PM
.34
.45
.06
c
.10
.36
.50
.14
c
.14
.36
.48
.16
c
.16
.09
c
.19
.34
.45
.57
c
c
.57
.36
.45
.55
c
c
.55
.36
.45
.55
c
c
.55
.48
c
c
.48
Input
Total
.34
45
.06
1.25
.67
.36
.50
.14
1.23
.73
.36
.48
.16
1.21
.68
.55
1.38
.94
.34
.45

1.70
1.25
1.14
.36
.45
.55
1.45 .,
1.18
1.10
.36
.45
.55
1.36
1.10
1.03
.48
1.77
1.41
1.56

-------
                                                              TABLE  7-15.    (Continued)
 I
CO
en

Energy Demand
Model Boiler Alternative NO SO.
RES-30 -




RES-150 -




RES-400 -

DIS-36 -

DIS-150 -
Unc
LEA
LEA
LEA
LEA
Unc
LEA
LEA
LEA
LEA
LEA
LEA
LEA
Unc
LEA
Unc
1
»
»
»
*
»
*
»
>
9
t
>
1
t
»
*
HDS,
HDS,
HDS,
FGD
FGD


HDS,
HDS
HDS
FGD
FGD
,
,


FGD5
FGD
FGD
Unc
HDS
Unc
LEA/RAP,
NG-30 -

NG-150 -
Unc
LEA
Unc
»
»
»
Unc
Unc
Unc

]
,
,

t
,
,
LEA/RAP,
Unc Baseline -21^'t(
HDS 1 O9 .5iyt(l
HDS . 2 O9 .70d'f(2
(90)-MA, FGD1 3 O9 .08 (
(90J-MA, ESP 4 O9 .03 (
Unc Baseline 1.06;. ',(3
HDS 1 O9 2.55V I 8
HDS . 2 O9 3.52d>t 1
(90), FGD1 3 C9 .55
(90), ESP 4 O9 .16
, FGD1 . Baseline O9 .36
(90), FGD1 1 O9 .99
(90), ESP 2 O9 .38
Unc Baseline .
Unc 1 O9 ,10d (
Unc Baseline ,
HDS, Unc 1 O9 ,47T (1
Unc Baseline
Unc 1 O9
Unc Baseline O9
Unc, Unc 1
MHt (106 Btu/hr) Percent of Boiler Heat
PH N0x S02 PM
.74)d'£ 2.4
.79)7* c 5.8 c
.45)d>t c 8.0 c
.27 c .88 c
.10) DNAe .37 DNAe
.71)d'£ 2.4
.93)°'% c 5.8 c
2.32)a>T 8.0 c
1.85) c .88 c
.54) .11 ( .37) .36 .25
1.22 c .31 c
3.33 c p .85 c
1.28 DNAe .32 DNAe

.34)d 1.14

.59)f 1.14




Input
Total
2
4
5.8
8.0
.88

2.
5.
8.
.
.
•


1.

1.





4
8
0
88
61
31
85

0
14
0
14
0

0

                Nomenclature definitions can be found in Section 6.3
                Steam and electricity  demands are expressed as net heat input to  the model boiler.
               clncluded in SO,  demand or percent.
               j              £.
                Includes heater  and hydrogen plant demands.
               eData not available.
               f3.4% S feedstock for residual and 0.5S feedstock for distillate.
               9When LEA control  is applied to uncontrolled boilers, a small  increase in fuel efficiency is realized.
                The purpose of this table is to present energy penalties associated with various control techniques,
                and therefore,  the  energy penalty shown for LEA is zero.
                where LEA/RAP are used  together the effects are balanced.
               ^Sodium alkali.
                Venturi  type  FGD used for combination S02/PM control.
RAP tends  to reduce  efficiency.  Therefore,
               j
                75% SO. removal.

-------
a wet FGD system to achieve 90 percent control,  about 0.5 percent of the
boiler heat input is required.
     Combined S02/PM control  techniques are the  second most energy intensive
systems considered.   Across the coal-fired boiler size range,  energy demands
range from 1.03 to 1.25 percent of the boiler heat input for combined
systems using.venturi-type double alkali  scrubbers (Alternative 4),  and 1.15
to 1.70 percent for boilers using dry scrubbing  systems (Alternative 3).   In
general, the energy requirements for combined S02/PM systems  exceed the
requirements for model  boilers using separate SO? and PM controls (e.g., FGD
for S02, FF for PM).  This increase in energy demand is a result of the 5
kPa (20 inches of water) gas-side pressure drop  assumed for combined
systems, compared to an overall 3 kPa (12 inches of water)  pressure drop
assumed for the FGD tray type scrubbers with fabric filter particulate
control.
     Table 7-15 shows that electrostatic precipitators (ESP) require less
energy to maintain the  corona, overcome system pressure drop and operate
plate rappers, than fabric filter systems require for fan and  bag cleaning
operations. For example, over the model  boiler size range,  the energy
required to operate the ESP at control  Alternative 5 [22 ng/J  (0.05 lb/10
Btu)] ranges from 0.10  to 0.19 percent of the boiler heat input for all HSC
boilers, while 0.48 to  0.57 percent is required  to operate the fabric filter
systems on all LSC boilers.  At the less stringent Alternative 2 [43 ng/J
(0.1  lb/10  Btu)], the  energy required to operate ESP's on HSC boilers
ranges from 0.06 to 0.16 percent of the boiler heat input, while the energy
demand from fabric filter systems on LSC boilers ranges from 0.55 to
0.57  percent.
     In conclusion, application of the control alternatives to the model
boilers may require less than a total  of 2 percent of the boiler heat input
to achieve the respective control levels.  The exceptions to this are the
control alternatives where HDS is applied.  Energy requirements for HDS
range from 2.4 to 8.0 percent of the boiler heat input.
                                     7-37

-------
7.5  OTHER IMPACTS
     An increase in noise at the industrial boiler site is expected as  a
result of the operation of the various control techniques.  For FGD's,  the
higher level of noise would result from fans, pumps, and agitators.  For
ESPs, the higher noise levels are due to the fans, pumps, compressors,
electrode rappers, etc.  For FF's, the bag cleaning mechanisms result in
increased noise levels.  Noise-abatement techniques, such as design changes
(redesign of cams, gears, and housings, or provision for vibration
absorption), use of absorbing materials placed on walls to absorb sound
after it has been generated, and sound barriers or silencers for fans should
mitigate the increased noise levels effectively.
7.6  OTHER ENVIRONMENTAL CONCERNS
7.6.1  Long-Term Gains/Losses
     Increased emission control of the air pollutants resulting from the
operation of industrial boilers would result in reduced air emissions and
increased energy, water (if sodium scrubbing systems are used), and solid-
waste impacts.  The solid-waste and water impacts are mitigated by other  EPA
regulatory programs.  The long-term gains achieved result from reducing PM,
S09, NO  , trace metals, radionuclides, inhalable particulates, and POM
  L.    X
emissions to the ambient air.  Another important long-term benefit will be
the application of control technology which makes possible the use of coal
in an environmentally acceptable manner.
7.6.2  Environmental Impact of Delayed Standard
     As analyzed in Section 7.1, there are significant air quality benefits
achieved by emission reductions at the alternative control levels compared
to baseline emissions.  Large quantities of pollutants are reduced and a
significant incremental ambient air quality benefit is achieved.  Therefore,
the impact of a delayed standard would be negative to the extent that the
incremental benefit discussed in Section 7.1 would not be achieved as long
as the standard was delayed.
                                     7-38

-------
7.7  REFERENCES

1.   Environmental  Protection  Agency.   New Stationary  Sources  Performance
     Standards; Electric Utility Steam Generating  Units.   Federal  Register.
     June 11, 1979.

2.   Roeck, D.R., and R. Dennis.  Technology  Assessment  Report for Industrial
     Boiler Applications:  Particulate  Control.   GCA/Technology Division.
     Bedford, Massachusetts.   June 1979.   p.  239.

3.   Dickerman, J.C.  and K.L.  Johnson. Technology Assessment  Report  for
     Industrial Boiler Applications:  Flue  Gas Desulfurization.
     EPA-600/7-79-178i.   Radian Corporation.  Durham,  North  Carolina.
     November 1979.   pp. 6-14, 6-27.

4.   Reference 3, p.  6-15.

5.   Reference 3, p.  2-153.

6.   Nemerow, Nelson  L.  Industrial  Water Pollution;  Origins, Characteristics
     and Treatment.   Reading,  Massachusetts,  Addison-Wesley  1971.
     pp. 134-141.

7.   Survey of the Application of Flue Gas Desulfurization Technology in the
     Industrial Sector.   Energy and Environmental  Analysis,  Inc. Arlington,
     Va.  NTIS PB - 270-548.   December 1976.  p. 26.

8.   Environmental  Protection  Agency.   Water  Programs; State Underground
     Injection Control Programs; Minimum Requirements  and  Grant Regulations.
     Federal Register.  44(78): 23738-23766.  April  20,  1979.

9.   Torrey, S. (ed.).  Coal Ash Utilization; Fly  Ash, Bottom  Ash  and Slag.
     Park Ridge, New  Jersey.   Noyes Data Corporation,  1978.  p.  5.

10.  S0« Emissions from Coal-Fired Steam-Electric  Generators:  Solid Waste
     Impact.  The Aerospace  Corporation.   Los Angeles, California.
     EPA-600/7-78-044a.   March 1978.   p. 43.

11.  Reference 6, p.  6-24.

12.  FGD Sludge Disposal Manual, FP-977 Research Project 786-1, Electric
     Power Research  Institute, Palo Alto,  California,  January  1979.  pp. 5-1
     to 5-5.

13.  Reference 12, pp. 12-1.

14.  Reference 12, pp. 5-5  to  5-7.

15.  Reference 12, pp. 7-19  to 7-29.
                                  7-39

-------
16.   Reference 12, pp. 8-128 to 8-133.

17.   Personal  Communication, Alan Corson, Hazardous Waste Management
     Division, Office of Solid Waste, U. S. Environmental Protection Agency,
     May 1980.

18.   Environmental Protection Agency.  Landfill Disposal  of Solid  Waste;
     Proposed Guidelines.  Federal Register. 44 (59): 18138-18148.
     March 26, 1979.

19.   Reference 3, pp. 5-3 - 5-10.

20.   Reference 2, pp. 203-207.

21.   Lim, K.J., et al.  Technology Assessment Report for Industrial  Boiler
     Applications: NO  Combustion Modification.  Acurex Corporation.
     Mountain View,  California.  EPA-600/7-79-178f.  December 1979.  pp.  5-2
     and 5-73.
                                      7-40

-------
                              8.0  COSTS

     This chapter presents an analysis of the cost impacts  of various
control alternatives applied to industrial  boilers.   The costs associated
with uncontrolled boilers and emission control  systems  are  evaluated for the
model boilers described in Chapter 6.   The emphasis  is  to quantify the
individual boiler cost impacts associated with  control  of NOX> S02> and PM
to various emission levels.   In addition to this  cost  impact analysis, an
analysis of the economic impacts of various emission control  levels on
boiler users, boiler manufacturers, and emission control  system vendors is
presented in Chapter 9.  A further analysis of  regional  and national cost
impacts using the Industrial Fuel Choice Analysis  Model  (IFCAM) is presented
in a separate report.    The IFCAM analysis accounts  for regional  variations
in projected fossil fuel prices, impacts of national energy policy, and the
impacts of local, State, and Federal  air quality regulations to generate
aggregate economic impacts on the industrial boiler  population.  Chapter 10
describes the IFCAM methodology and summarizes  the major results.
     The cost analysis presented in this chapter provides an individual
boiler analysis of the cost impacts of various  control  alternatives.  Total
capital and annualized costs are presented individually, and relative  to the
uncontrolled boiler case and the baseline case.  The uncontrolled case is
defined as a new boiler without any emission controls while the baseline
case is defined as a new boiler with controls designed  to meet the highest
level of emissions expected under the mix of existing regulations (see
Chapter 6).
     The following sections present the methodology  used to develop and
analyze the cost impacts for various control technologies.   Results are
presented as a function of boiler type and size, fuel type, and control
alternative.  Cost impacts for emerging technologies are presented in
Appendix E.  Costs for emerging technologies are based  on limited cost data
                                    8-1

-------
since these technologies have not generally been commercially applied to
full-scale industrial boilers.  As such, the costs presented for emerging
technologies in Appendix E should not be considered as accurate as the costs
presented in this chapter.
     All costs are reported in June 1978 dollars.  The costs presented in
this chapter do not include costs of emission testing, compliance,
monitoring, and reporting that may be incurred under the control alterna-
                                                       p
tives.  These costs are addressed in a separate report.
8.1  COSTING APPROACH
     The cost impacts of control alternatives, including the baseline
alternative, were assessed using the concept of model boilers.   As discussed
in Chapter b, model boilers are combinations of standard boilers and partic-
ulate matter (PM), nitrogen oxides (NO ), and sulfur dioxide (S0~) control
                                      A                         C-
methods designed to meet specified emission levels.  Cost impacts for each
model boiler were calculated  in terms of:

        •  capital costs of boilers and control systems
          (capital investment required),
        •  annual!zed costs of boilers and control systems
          (annual operation and maintenance costs plus
          capital related charges),
        •  incremental capital and annual costs for boilers
          and control systems over the uncontrolled and
          baseline alternatives.
     The model boilers selected in Chapter 6 for analysis of the cost,
energy, and environmental impacts of control technologies are shown in
Tables  8.1-1 and 8.1-2 (emerging technologies are addressed in  Appendix E).
Table 8.1-3 defines the abbreviations used to denote the various control
systems.  Boilers represented by alternatives 1 thru 5 were selected to
allow evaluation of the impacts of NOX, S02, and PM controls across a range
of boiler types and sizes, fuel types, and emission control methods.
     Cost impacts were analyzed for boilers firing natural gas  (N6),
distillate fuel oil (DIS), residual fuel oil (RES), low sulfur  coal (LSC)
and high sulfur coal  (HSC).  The range of costs developed for boilers firing
                                     8-2

-------
                                TABLE  8.1-1.   COAL-FIRED MODEL BOILERS
CO
CO

Emission Levels or Removal Requirements
ng/J (lb/10°Btu)
Standard Control
Boiler Alternative
B

HSC-30
HSC-75
HSC-150

1
2
3
4
5
B


LSC-30
LSC-30
LSC-150
1
2
3
4
5
B
HSC-400

1
2
3
B
LSC-400

1
2
3
NO •
X
151-271(0.33-0.63)°
215(0.5)
215(0.5)
215(0.5)
215(0.5)
215(0.5)
151-271(0.33-0.63)°
215 (0.5)
215 (0.5)
215 (0.5)
215 (0.5)
215 (0.5)
301 (0.7)
258 (0.6)
258 (0.6)
258 (0.6)
301 (0.7)
258 (0.6)
258 (0.6)
258 (0.6)
so2
1076 (2.5)
860 (2.0)
860 (2.0)
50% Removal
90% Removal
90% Removal
1076 (2.5)
860 (2.0)
860 (2.0)
50% Removal
90% Removal
90% Removal
516 (1.2)
50% Removal
90% Removal
90% Removal
516 (1.2)
50% Removal
90% Removal
90% Removal
PM
172-258 (0.40-0.60)C
86 (0.2)
22 (0.05)
43 (0.1)
43 (0.1)
22 (0.05)
172-258 (0.40-0.60)C
86 (0.2)
22 (0.05)
43 (0.1)
43 (0.1)
22 (0.05)
43 (0.1)
43 (0.1)
43 (0.1)
22 (0.05)
43 (0.1)
43 (0.1)
43 (0.1)
22 (0.05)
Control Methods6
N0x
Unc.
SCAb
SCAb
3 K
scAb
SCAb
Unc.
SCA.
SCAb
SCAb
SCAb
SCAb
LEA
a
SCA
SCA
LEA
SCA
SCA
SCA
so2
cc
cc
cc
a
FG3
FGD
Unc
Unc
Unc
OS
FGD
FGD
FGDd
a
FGD
FGD
Unc
DS
FGD
FGD
PM
SM
sss
ESP
a
FGD/PM
ESP
SM
SSS
FF
DS/PM
FGD/PM
FF
FGD/PM
a
FGD/PM
ESP
FF
DS/PM
FGD/PM
FF
         a50% SO- removal  alternative not applicable for HSC standard  boilers  since this removal would not meet baseline
          emission level.   Therefore, no model boiler is analyzed for this  alternative.
          SCA required  on  44  MW  (150 x 10  Btu/hr) size only; smaller  boilers  meet NOx  level without control.
         cBaseline emissions  depend on boiler size and type (see Chapter 6 and Chapter 7).
         d78.9% S0? removal efficiency required at baseline.
         eAbbreviations defined  in Table 8.V3.  Unc (uncontrolled) indicates no  control  system  is required to meet emission
          levels.
          Alternatives  shown  define model boilers for each standard boiler.   For example,  six model boilers are defined
          for HSC-30,  six  are defined for HSC-75, etc.

-------
                                 TABLE 8.1-2.   OIL- AND  GAS-FIRED  MODEL BOILERS
00
 I









Emission Levels or Removal Requirements
ng/J (lb/10°Btu)
Standard Control
Boiler Alternative NO

RES-30
RES-150


\
RES-400 j


"
DIS-150
NG-30
NG-150
B 172
1 129
2 129
3 129
4 129
(B 129
1 129
'2 129
B 52
1 43
IB 104
(1 86
IB 52
(1 43
IB 104
(1 86
(0.
(o.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
4)
3)
3)
3)
3)
3)
3)
3)
12)
1)
24)
20)
(0.12)
(0.1)
(0.
(0.
24)
20)
so2
688
344
129
(1.6)
(0.8)
(0.3)
90% Removal
90% Removal
344
(0.8)
90% Removal
90% Removal
219
129
219
129
0.43
0.43
0.43
0.43
(0.51)
(0.3)
(0.51)
(0.3)
(0.001)
(0.001)
(0.001)
(0.001)
PM
99
43
43
43
22
43
43
22
6
6
6
6
4.3
4.3
4.3
4.3
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(o.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
23)
1)
1)
1)
05)
1)
1)
05)
015)
015)
015)
015)
01)
01)
01)
01)



Control Methods5
NO
x
Unc
LEA
LEA
LEA
LEA
LEA
LEA
LEA
Unc
LEA
Unc
LEA/RAP
Unc
LEA
Unc
LEA/ RAP
so2
HDS (1 .6)
HDS (0.8)
HDS (0.3)
FGD3
FGD3
FGDd
FGD
FGD
Unc
HDS (0.3)
Unc
HDS (0.3)
Unc
Unc
Unc
Unc
PM
Unc
HDS/PM
HDS/PM
FGD/PM
ESP
FGD/PM
FGD/PM
ESP
Unc
Unc
Unc
Unc
Unc
Unc
Unc
Unc
^n/\i .kin •% 1 L -» 1 •» f^M.tUk.-tMA / Cf* r\\ ..«•«. .4 ~. _ A A LJII 1 i r r\ . . ir\0 t\ *- n _\ 	 _i • 	 j_i 	 _ _ / r-/*n »• x
         Double alkali scrubbing,(FGD)  used on 44 MW (150 x 10  Btu/hr), sodium throwaway (FGD/Na)
         used on 8.8 MW (30 x 10  Btu/hr).

         Abbreviations defined in  Table 8.1-3.  Unc (uncontrolled)  indicates no control system is  required
         to meet emission levels.

         Alternatives shown define model  boilers for each standard boiler.  For example, five model
         boilers are defined for RES-30,  five for RES^T50", etc.

         75X removal efficiency required  at baseline.

-------
            TABLE 8.1-3.   ABBREVIATIONS FOR CONTROL METHODS
NO  Control  Methods
     SCA - Staged combustion air (overfire air)  used in combination with LEA
     LEA - Low excess air
     RAP - Reduced air preheat

S0? Control  Methods
     CC     - Compliance coal
     F6D    - Double alkali  scrubbing  flue gas desulfurization  (90% removal
                unless noted)
     FGD/Na - Sodium throwaway flue gas  desulfurization (90% removal)
     DS     - Dry scrubbing  (50% removal)  using  lime spray  drying
     HDS(x)  - Hydrodesulfurized oil (x percent sulfur)

PM Control Methods
     SM     - Single mechanical collector  (multitube cyclone)
     SSS    - Sidestream separator
     ESP    - Electrostatic precipitator
     FF     - Fabric filter
     FGD/PM - Particulate removal via FGD scrubber
     DS/PM  - Particulate removal via DS fabric filter
     HDS/PM - Particulate removal via low ash HDS cleaned  oil
 Compliance coal  is defined as a coal  with  a  sulfur content allowing an SO,,
 emission limit to be met without control.  The actual  sulfur content
 depends on the emission limit (see Table 8.1-9).
                                   8-5

-------
these fuels are expected to illustrate the range of control costs that
boiler operators could experience.
8.1.1  Cost Bases
     Capital investment and annual operating and maintenance (O&M) costs
were calculated for each model boiler and its associated control  method(s).
In general, these cost calculations were carried out using a variety of
"cost algorithms" developed for boiler and control method cost estimation.
Each algorithm represents a particular boiler or control method cost
component as a algebraic function of key system specifications.  A separate
report documents all model boiler costing algorithms and other background
     2
data.   Table 8.1-4 summarizes the various sources of information used to
develop costs.
     The boiler specifications presented in Chapter 6 (Tables 6-3 thru 6-7)
provide the specifications required to cost boiler systems.  In addition, a
number of control device specifications are required in order to cost
control devices.  Table 8.1-5  lists the general specifications for the
control devices evaluated.  These specifications are typical for industrial
boiler control devices currently  in use.
8.1.2  Capital Costs
     Table 8.1-6 shows the bases  and methodology for developing capital cost
estimates for uncontrolled boilers and control methods.  Specific equipment
lists and assumptions regarding the capital cost bases for individual types
of boilers and control systems are detailed in the appropriate Individual
Technology Assessment Reports  (ITARs) and the references listed in
Table 8.1-4.  Boiler capital costs are for new individual boilers of the
specified design capacity.  Savings associated with multiple boiler units or
adding a boiler to an existing powerhouse are not considered.  Retrofit
control costs are addressed separately in Section 8.2.3.
8.1.3  Operating and Maintenance  Costs
     Table 8.1-7 lists the components of boiler and control method operating
and maintenance (O&M) costs.   Specific assumptions with regard to O&M cost
bases for individual types of  boilers and control methods are detailed in
the appropriate ITARs.
                                   8-6

-------
                                 TABLE 8.1-4.  SUMMARY OF SOURCES OF COSTING INFORMATION
00

Costed Item
All boilers and control systems
Uncontrolled boilers
FGD systems
FGD systems
FGD systems
SM, SSS's
ESP, FF's
NO controls
/\
HDS oils
HDS oils
Compliance coals
Developing
Type of Information Organization Date(s)
General summary of cost
development and results
Algebraic cost algorithms
Algebraic cost algorithms
Technology Assessment Report
giving individual system costs
Revisions to cost algorithms
developed by Acurex
Technical memo developing
cost algorithms from vendor
quotes
Algebraic cost algorithms
Technology Assessment Report
giving individual system costs
Technology Assessment Report
on oil cleaning estimating
costs of HDS oil
Issue paper on oil cleaning
Technical memo estimating
Radian
PEDCo
Acurex
Radian
Radian
Radian
PEDCo
Acurex
Catalytic
Radian
Radian
4/82
1/80,
6/79
12/79
11/79
2/80,
6/80
1/81
4/80
12/79
1979
11/80
2/80
Reference
2
3,4
5
6
7,8
9
10
11
15
16
17
                                                costs of coal with reduced
                                                sulfur content

-------
                         TABLE 8.1-5.   EMISSION CONTROL  SYSTEM  GENERAL  DESIGN  SPECIFICATIONS1
           Control  Device
          Item
               Specification
           Single
           Mechanical  Collectors
Material of construction

Pressure drop
Carbon steel

1.0 kPa (4 in. O gauge)
CO
I
00
           Side Stream Separators
Material of construction

Pressure drop
Amount of gas flow treated
in fabric filter
Fabric filter

Bag life
Mechanical  collector and fabric filter:
carbon steel
1.5 kPa (6 in.  H20 gauge)
20%

Multi-compartment pulse-jet with Teflon
coated glass felt bags
2 years
           Electrostatic
           Precipitators (ESP)
Material of construction

Specific collection areas0
(plate area per gas volume
 for 21.5 ng/J
(0.05 lb/10°Btu) control
 levels
                                    Pressure drop
                                    Power demand
Carbon steel  (insulated)

Underfeed and chaingrate  Stokers:
     33.2 mVirf/s (169 ft /Kracfm)
Spreader Stekeps:        9   o
     46.4 rrr/m /s (236 ftr/KTacfm)
Pulverized Coal:          9   ~
     50.0 nr/nr/s (254 fir /Kracfm)
Oil -Fired: „  7          9-1
     78.7 nT/nr/s (400 fr/10 acfm)
0.25 kPa (1 in. H
32 W/m2 (3 w/ft2)
                                                  gauge)

-------
                                               TABLE 8.1-5.   (CONTINUED)
00
I
VO
           Control  Device
          Item
               Specification
           Fabric  Filter
Material of construction
Cleaning method
Air to'cloth ratio
Bag material
Bag life     .
Pressure drop
Carbon steel (insulated)
Reverse-air (multi-compartment)
1 cm/s (2 ft/min)
Teflon-coated fiberglass
2 years
1.5 kPa (6 in.  HO gauge)
           Double Alkali  FGD
           (S02 removal only)
Scrubber type
Pressure drop
Scrubber sludge
Sludge disposal
Tray tower

1.5 kPa (6 in. H20 gauge)

60% solids

Trucked to off-site landfill
           Sodium Throwaway  FGD
           (either S0« removal
            only or combined SO,
            & PM removal)      '
Material of construction

Scrubber type

Pressure drop
Waste water treatment
316 stainless steel

Variable throat venturi

2.0 kPa (8 in. HgO)

Treated in existed facility
           Double Alkali FGD
           (S0« and PM removal)
Material of construction

Scrubber type
System design
                                    Pressure drop
                                    (over SM and scrubber)

                                    Sludge disposal
316 stainless steel

Variable throat venturi

Includes 80% efficient single mechanical
collector upstream of scrubber

5.0 kPa (20 in. H20 gauge)


Dry particulate collected in single mechanical
combined with 60% solids scrubber sludge and
trucked to off-site landfill

-------
                                    TABLE 8.1-5.  (CONTINUED)
Control Device                     Item                               Specification

Dry Scrubbing            Materials of construction     Carbon steel spray dryer and fabric filter
(spray drying, S0?                                     (insulated)
 and PM removal)
                         Reagent                       Lime; no solids recycle

                         Fabric filter                 Reverse-air (same design as previous
                                                         fabric filter

                         Pressure drop                 1.5 kPa (6 in. H20 gauge)
                         Solids disposal               Trucked to off-site landfill


 For more detail on system design and operating parameters see Individual Technology Assessment
 Reports (ITARS).
K
 All pressure drops refer to gas side pressure drop across entire control system.
 Values shown are for sulfur content of 3.5% in coal feed to boiler.   Boilers firing coals with
 lower sulfur content have somewhat higher SCA values.

-------
                TABLE 8.1-6.   CAPITAL  COST COMPONENTS3
(1)  Direct Costs
                    Equipment
                    Installation

                    Total  Direct Costs

(2)  Indirect Costs12
          Engineering (10% of direct costs  for boiler,  NO ,  and  PM
                      controls;  For boilers with  heat inputs ^58.6 MW,
                      FGD  system engineering costs  are  taken as  10% of
                      direct costs  for an  FGD system  at 90%  removal  on
                      58.6 MW unit.   For 117.2 MW boilers, engineering
                      costs for  SOp controls are  10%  of direct costs)

          Construction and Field Expenses     (10% of  direct  costs)
          Construction Fees                  (10% of  direct  costs)
          Start Up Costs                      ( 2% of  direct  costs)
          Performance Costs                  ($2000 for NO   systems, 1%  of
                                              direct  costs for boilers,
                                              FGD systems, and PM systems)

                      Total  Indirect Costs
                  1 p
(3)  Contingencies   = 20% of (Total  indirect + Total Direct Costs)

(4)  Total  Turnkey Cost =  Total  Indirect Cost + Total Direct Cost +
                          Contingencies

(5)  Working Capital12 = 25% of  Total  Direct Operating  Costs (See Table  8.1-7)

(6)  Total  Capital  Cost =  Total  Turnkey + Working Capital


 Boiler and each control system  costed separately;  factors apply to cost
 of boiler or control  system considered;  i.e., the  engineering cost
 for the PM control  system is 10% of the direct cost  of the  PM control
 system.
                                   8-11

-------
        TABLE 8.1-7  OPERATING AND MAINTENANCE COST COMPONENTS
(1)  Direct Operating Costs3
          Direct Labor
          Supervision
          Maintenance Labor, Replacement Parts and Supplies
          Electricity
          Water
          Steam
          Waste Disposal
            Solids (Fly ash and bottom ash)
            Sludge
            Liquid
          Chemicals	

          Total Non-Fuel 0 & M
          Fuel	

          Total Direct Operating Costs

(3)  Indirect Operating Costs (Overhead)b'c

          Payroll (30% Direct Labor)
          Plant   (26% of Direct labor + Supervision + Maintenance costs)

(3)        Total Annual Operating and Maintenance Costs =

                   Total Direct + Total Indirect Costs


aFor HDS and CC the total direct operating cost (DOC) is taken as:
     nnr   /fuel usex ,8760 hr% /boiler load*  /incremental costv
     UUL " (  hour  ' {   yr  ' l  factor   '  v    of fuel     '
 where incremental cost of fuel is the cost of CC or HDS fuel minus the
 cost of the fuel used in the uncontrolled boiler.

bBoilers and each control systems are costed separately; factors apply
 to boiler or control system being considered, (i.e., payroll overhead
 for F6D system is 30% direct labor requirement of F6D system).

°Factors recommended in Reference 12, p. 117.
                                   8-12

-------
     In addition to their dependence on boiler size,  fuel  type,  and the flow
rate and composition of the flue gas to be treated, the O&M costs for the
boiler and control methods are a function of capacity utilization (load
factor), utility unit costs (steam, electricity,  water), and unit costs for
raw materials, waste disposal, and labor.  Table  8.1-8 lists the values
selected for these parameters in this analysis.   Fuel  costs are  a major
component of boiler operating costs.  The prices  used for boiler fuels are
presented in Table 8.1-9.  As the table indicates,  1990 fuel prices (in 1978
dollars) are used in the cost analysis to account for the  expected escala-
tion of fuel prices above the general inflation  rate  during the  period in
which a regulation would be effective.  The costs of  uncleaned oils, natural
gas, HSC, and LSC are based on fuel prices developed  for the IFCAM
      13 14
model.  '    The prices for the HDS cleaned oils  are  based on oil cleaning
                          15
costs developed in an ITAR   and updated by Radian  to 1990 fuel  prices in a
subsequent memo.    The prices for the intermediate sulfur coals (compliance
coals) use estimates of the cost premium associated with obtaining coals
with lower sulfur content compared to high sulfur coal.
     Combustion modification techniques used to control  NO  emissions can
                                                          A
affect the magnitude of the fuel cost component.  Operation with low excess
air (LEA) to control NO  tends to increase boiler thermal  efficiency,
                       /v
resulting in fuel savings and a reduction in the  fuel  cost component.  The
use of staged combustion air (SCA) to control  NO  emissions (low excess air
                                               ^
in combination with overfire air) can result in  increased  fuel use, and
therefore increased fuel costs.  Reduced air preheat  (RAP) control
techniques may reduce boiler efficiency, increasing fuel use. The
incremental increase or decrease in fuel costs associated  with combustion
modifications is reported as an operating cost for NO  control in subsequent
                                                     A
sections.
     Likewise, use of more expensive intermediate sulfur coals (compliance
coals) or HDS cleaned oils for S02 and/or PM control  results in  increased
fuel costs compared to the use of HSC or uncleaned  oils.  These  incremental
fuel costs are reported as an operating cost for  S02  and/or PM control in
subsequent sections.
                                    8-13

-------
   TABLE 8.1-8.   LOAD FACTORS AND UTILITY AND UNIT OPERATING COSTS1
(1)  Load Factors (Capacity Utilization)12
  Boiler Capacity and Fuel
8.8 MW ( 30 x 106. Btu) Natural Gas & Distillate Oil
 44 MW (150 x 10° Btu) Natural Gas & Distillate Oil
All residual-fired boilers
All coal-fired boilers
                                                            Load  Factor

                                                               0.45
                                                               0.55
                                                               0.55
                                                               0.60
(2) Utility Costs
                 12
          Electricity
          Water
          Steam
                              $0.0258/kwh       -
                              $0.04/m3 (SO.U/IQ-3 gal)
                              $3.01/GJ ($3.5/103 Ib)
(3) Raw Material, Labor and Waste Disposal Costs

     Waste Disposal3'6
                                                12
          Solids (Ash)
          Sludge
          Liquid

     Chemicals

          Na2C03
          Lime
          Limestone
     Labor
          Direct labor
          Supervision labor
          Maintenance labor
                               $0.0166/kg  ($15/ton))     ,.,,«.,   .,.„
                               $0.0166/kg  ($15/ton Mocked  to landfill

                               $0.47/m3  ($1.79/103 gal)
                               $0.099/kg      ($90/ton)
                               $0.039/kg      ($35/ton)
                               $0.00883/kg    ($8/ton)
                               $12.02/man-hour
                               $15.63/man-hour
                               $14.63/man-hour
 'June 1978 dollars
                                   8-14

-------
         TABLE 8.1-9.  FUEL PRICES (June 1978
                                                      Price3
             Fuel                                 $/GJ ($/106 Btu)

     Natural Gas                                  $5.12 ($4.85)
     Distillate Oil (0.5% S)                      $6.39 ($6.06)
     Distillate Oil (0.3% S W/HDS)                $6.83 ($6.47)
     Residual Oil (3.0% S)                        $5.12 ($4.85)
     Residual Oil (1.6% S W/HDS)                  $5.58 ($5.29)
     Residual Oil (0.8% S W/HDS)                  $5.87 ($5.56)
     Residual Oil (0.3% S W/HDS)                  $6.15 ($5.83)
     Low Sulfur Coal (0.6% S)                     $2.54 ($2.41)
     High Sulfur Coal  (3.5% S)                    $1.91 ($1.81)
     Intermediate Sulfur Coal (1.6% S)b           $2.35 ($2.23)
     Intermediate Sulfur Coal (1.2% S)c           $2.43 ($2.30)

aPrice is projected 1990 price in 1978 $ and includes transportation
 costs to Midwest boiler location.
bUsed to achieve 1076  ng/J (2.5 lb/106 Btu) emission limit.
GUsed to achieve 860 ng/J (2.0 lb/106 Btu) emission limit.
                                 8-15

-------
8.1.4  AnnualIzed Costs
     The total annualized costs ($ per year) for uncontrolled boilers,  each
control system, and the total annualized cost of boilers and controls are
calculated for each model boiler based on the boiler capacity utilization.
As depicted in Table 8.1-10, the total annualized cost is the sum of the
annual O&M costs and annualized capital charges.
     The capital recovery factors used in this study are based on equipment
lives specified in the ITARs and an interest rate of 10 percent.  The
10 percent interest rate should not be considered as the actual cost of
borrowing capital since this analysis is not intended as an economic
feasibility study.  Rather, 10 percent was selected as a typical nominal
rate of return on investment to provide a basis for calculation of capital
recovery charges.  The capital recovery factors used for the boiler and
control equipment investments are presented in Table 8.1-10.
8.2  ANALYSIS OF COST IMPACTS
     This subsection presents the results of the model boiler cost impact
analysis.  This analysis focuses on the incremental cost impacts in going
from the baseline control alternative to more stringent alternatives.
Capital costs, annualized costs, and cost effectiveness of emission control
methods are discussed in the following subsections.  In addition, a brief
discussion of cost impacts for retrofit installations is also included since
certain modified or reconstructed boilers may become subject to Federal
regulations as discussed in Chapter 5.
8.2.1  Capital Costs
     The capital costs for each model boiler are presented in Tables 8.2-1
through 8.2-3.  Individual costs for each uncontrolled boiler, each control
system, and the total model boiler costs are  given.  Also included is the
"normalized" total capital cost calculated by dividing the total cost of the
model boiler by the boiler capacity.  Normalized capital costs provide a
measure of the capital investment required per unit of installed boiler
capacity.
                                     8-16

-------
               TABLE 8.1-10.   ANNUALIZED COST COMPONENTS
1.  Total  Annual ized Cost = Annual  Operating Costs + Capital  Charges
2.  Capital  Charges
         =  Capital recovery + interest on working capital  +
              miscellaneous (G&A,  taxes and insurance)
3.  Calculation of Capital -Related  Cost Components
   A.  Capital  Recovery = Capital  Recovery Factor (CRF) x Total  Turnkey Cost
                Number of years of
       System       Useful  Life - ,       CRF '          * *" 1  " 10ii
Boiler
PM
FGD
N0x
30
20
15
30
                                                  0.1061
                                                  0.1175
                                                  0.1315
                                                  0.1061
   B.   Interest on Working Capital  = 10% of working capital
   C.   G&A,  taxes and insurance = 4% of total  turnkey cost
                                   8-17

-------
                                            TABLE  8.2-1.
CAPITAL  COSTS  OF  HSC MODEL BOILERS'
(JUNE  1978$)
00
I—»
00
Capital Costs ($1000)
Control
Alternative
B
1
2
3
4
5
B
1
2
3
4
5
B
1
2
3
4
5
B
1
2
3
Model
Boiler
HSC-30-Unc, CC, SH
HSC-30-Unc, CC, SSS
HSC-30-Unc, CC, ESP
a
HSC-30-FGD, FGO/PM
HSC-30-FGD, ESP
HSC-75-Unc, CC, SH
HSC-75-Unc, CC, SSS
HSC-75-Unc, CC, ESP
a
HSC-75-Unc, FGO, FGD/PM
HSC-75-Unc, FGD, ESP
HSC-150-Unc, CC, SM
HSC-150-SCA, CC, SSS
HSC-150-SCA, CC, ESP
a
HSC-150-SCA, FGD, FGD/PH
HSC-150-SCA, FGO, ESP
HSC-400-LEA, FGDb, FGD/PM
a
HSC-400-SCA, FGD, FGD/PM
HSC-400-SCA, FGD, ESP
Uncontrolled
Boiler
1922
1922
1922

1922
1922
3533
3533
3533

3533
3533
8015
8015
8015

8015
8015
19059
19059
19059
N0x
Control
0
0
0

0
0
0
0
0

0
0
0
22.0
22.0

22.0
22.0
44.0
87.0
87.0
so2
Control
16.6
19.3
19.3

901
841
41.4
48.3
48.3

1355
1223
82.7
96.6
96.6

1842
1657
2756
2816
2576
PM
Control
62.6
111
362

w/SO,
289 i
125
229
693

w/SO
546 2
217
400
1475

w/SO.
1369 t
w/S02
w/SO_
1848 i
Total
2002
2053
2304

2903
3052
3700
3810
4275

4888
5303
8314
8533
9608

9878
11063
21859
21962
23569
Normalized0
Total
66.7
68.4
76.8

96.8
101.7
49.3
50.8
57.0

65.2
70.7
55.4
56.9
64.1

65.9
73.8
54.6
54.9
58.9
Percent Increase 1n Costs
Over Uncontrolled
Boiler
4.2
6.8
19.9

51.0
58.8
4.7
7.8
21.0

38.4
50.1
3.7
6.5
19.9

23.2
38.0
14.7
15.2
23.7
Over Baseline
Controlled Boiler
0
2.5
15.1

45.0
52.4
0
3.0
15.5

32.1
43.3
0
2.6
15.6

18.8
33.1
0
0.5
7.8
        Alternative requiring  50% SO- removal not applicable  to HSC boilers since resulting emissions would not meet baseline requirements.
         Baseline requires 78.91 SO- removal; other alternatives require 90% removal.
        cNormal1zed total Is capital cost divided by boiler capacity ($1000/106Btu/hr).

-------
                                      TABLE  8.2-2.
CAPITAL  COSTS  OF LSC MODEL BOILERS'
(JUNE 1978$)
03
I
I—•
UD
Capital Costs ($1000)
Control Model Uncontrolled
Alternative Boiler Boiler
B
1
2
3
4
5
B
1
2
3
4
5
B
1
2
3
4
5
B
1
2
3
LSC-30-Unc,
LSC-30-Unc,
LSC-30-Unc,
LSC-30-Unc,
LSC-30-Unc,
LSC-30-Unc,
LSC-75-Unc,
LSC-75-Unc,
LSC-75-Unc,
LSC-75-Unc,
LSC-75-Unc,
LSC-75-Unc,
LSC-150-Unc,
LSC-150-SCA,
LSC-150-SCA,
LSC-150-SCA,
LSC-150-SCA,
LSC-150-SCA,
LSC-400-LEA,
LSC-400-SCA,
LSC-400-SCA,
LSC-400-SCA,
Unc, SM
Unc, SSS
Unc, FF
OS, DS/PH
F6D, FGD/PM
FGD, FF
Unc, SM
Unc, SSS
Unc, FF
OS, DS/PM
FGD, FGD/PM
FGD, FF
Unc, SM
Unc, SSS
Unc, FF
OS, DS/PM
FGD, FGD/PM
FGD, FF
Unc, FF
OS, DS/PM
FGD. FGD/PM
FGD, FF
2326 '
2326
2326
2326
2326
2326
4274
4274
4274
4274
4274
4274
8690
8690
8690
8690
8690
8690
19924
19924
19924
19924
NO
Control
0
0
0
0
0
0
0
0
0
0
0
0
0
22.0
22.0
22.0
22.0
22.0
44.0
87.0
87.0
87.0
Control
0
0
0
653
768
604
0
0
0
1108
1067
077
0
0
0
1748
1488
1191
0
3341
2285
1850
PM
Control
63.6
113
269
H/SO-
H/SO;
269 i
127
233
667
H/SO.
u/so;
667 i
218
405
1142
W/SO.
w/so;
1142 i
2147
U/SO-
H/SO'
2147^
Total
2390
2440
2595
2979
3094
3199
4401
4507
4941
5382
5340
5817
8908
9118
9854
10461
10200
11045
22116
23353
22182
24009
Percent Increase in Costs
Normalized3 Over Uncontrolled
Total Boiler
79.7
81.3
86.5
99.3
103.1
106.6
58.7
60.1
65.9
71.8
71.2
77.6
59.4
60.8
65.7
69.7
68.0
73.6
55.3
58.4
55.5
60.0
2
4
11
28
33
37
3
5
15
25
'4
36
2
4
13
20
17
27
11
17
11
20
.8
.9
.6
.1
.0
.5
.0
.5
.6
.9
.9
.1
.5
.9
.4
.4
.4
.1
.0
.2
.3
.5
Over Baseline
Controlled Boiler
0
2.1
8.6
24.6
29.5
33.8
0
2.4
12.3
22.3
21.3
32.2
0
2.4
10.6
17.4
14.5
24.0
0
5.6
0.3
8.6
        'Normalized total is total capital cost divided by boiler capacity ($1000/10  Btu/hr).

-------
                           TABLE 8.2-3.
CAPITAL COSTS OF OIL- AND GAS-FIRED MODEL BOILERS'
                  (JUNE 1978 $)
Control
Alternative
B
1
2
3
4
B
1
2
3
4
B
1
» 2
^3
3 B
1
B
1
B
1
B
1
Model
Boiler
RES-30-Unc, HDS(1.6), Unc
RES-30-LEA, HDS(0.8), HDS/PM
RES-30-LEA, HDS(0.3), HDS/PM
RES-30-LEA, FGD/Na, FGD/PM
RES-30-LEA, FGD/Na, ESP
RES-150-Unc, HDS 1.6 , Unc
RES-150-LEA, HDS 0.8 , HDS/PM
RES-150-LEA, HDS 0.3 , HDS/PM
RES-150-LEA, FGD, FGD/PM
RES-150-LEA, FGD, ESP
RES-400-LEA, FGDb, FGD/PM
RES-400-LEA, FGD, FGD/PM
RES-400-LEA, FGD, ESP
DIS-30-Unc, Unc, Unc
DIS-30-LEA, HDS(0.3), Unc
DIS-150-Unc, Unc, Unc
DIS-150-LEA/RAP, HDS(0.3), Unc
NG-30-Unc, Unc, Unc
NG-30-LEA, Unc, Unc
NG-150-Unc, Unc, Unc
NG-150-LEA/RAP, Unc, Unc

Uncontrolled
Boiler
707
707
707
707
707
2735
2735
2735
2735
2735
14039
14039
14039
871
871
2927
2927
835
835
2709
2709
Capi
NO
Control
0
12.0
12.0
12.0
12.0
0
17.0
17.0
17.0
17.0
27.4
27.4
27.4
0
14.0
0
19.7
0
14.0
0
20.9
tal Costs
so2
Control
15.9
25.6
35.4
391
391
79.5
128
193
1475
1338
2246
2313
2125
0
12.1
0
74.1
0
0
0
0
($1000)
PM
Control
0
W/SO?
W/SO,
W/SO,
401 i
0
W/SOp
W/SO,
w/so;
1057 *
W/SO,
W/SO,
1692 *
0
0
0
0
0
0
0
0

Total
723
745
754
1110
1511
2815
2864
2912
4210
5130
16285
16352
17856
871
897
2927
3021
835
849
2709
2730

Normalized
Total
24.1
24.8
25.1
37.0
50.4
18.8
19.2
19.5
28.2
34.3
40.8
40.9
44.7
29.0
29.9
19.5
20.1
27.8
28.3
18.1
18.2
Percent
Over Uncontrol
Boiler
2.3
5.4
6.6
57.0
113.7
2.9
5.3
7.1
54.6
88.2
16.2
16.7
27.4
0
3.0
0
3.2
0
1.7
0
0.8
Increase in Costs
led Over Basel ine
Controlled Boiler
0
3.0
4.3
53.5
109.0
0
2.3
4.0
50.2
82.8
0
0.4
9.6
0
3.0
0
3.2
0
1.7
0
0.8
'Normalized total is total  capital cost divided by boiler capacity ($1000/106 Btu/hr).
 75% SO- removal efficiency required at baseline control alternative.

-------
     The last two columns in Table 8.2-1 thru 8.2-3 present percent
increases in the total  model boiler capital  cost.   These percent increases
are calculated with respect to:
       •  Uncontrolled boiler capital  cost,  and
       •  Baseline controlled boiler capital  cost  (cost of boiler
          and control  method required under  the baseline control
          alternative).
For those cases where  a control  device is used, the percent increase in
costs over the uncontrolled case provides a  measure of the additional
capital required to construct the control device(s).   The percent increase
over the baseline case provides  a measure of the additional capital required
to meet emission limits more stringent than  the emission limits based on
existing regulations (baseline).  In the discussions  of capital costs, the
major emphasis will be the comparison of percent increases in capital  cost
over baseline for various control alternatives.
     It should be noted that non-capital intensive control methods, such as
compliance coal and HDS cleaned  oils, do require working capital in order to
purchase more expensive fuels.   Working capital costs are reported as
capital costs in Tables 8.2-1 through 8.2-3.   Also, many model  boilers use
control methods which  simultaneously control  both  S02 and PM emissions.  In
Tables 8.2-1 thru 8.2-3, the total control system  cost is reported as an SCL
control cost with an appropriate note in the PM control  cost column.
     8.2.1.1  Small Coal-Fired Model Boilers.  This subsection discusses the
capital costs of controls for coal-fired boilers with thermal input
capacities of 73 MW (250 x 106 Btu/hr) and less.  The total normalized
capital costs for each HSC and LSC model boiler in Tables 8.2-1 and 8.2-2
are graphically represented in  Figures 8.2-1 and 8.2-2.   The small insert
table in each figure provides a  key to the S02 and PM control technologies
used by each model boiler in achieving the emission limits in the control
alternatives.  NO  controls are  not included in the insert tables.  In
                 A
general, NO  capital costs are  very small and do not  have a significant
           A
impact on overall capital costs.  Capital costs for NO  control by
                                                      X
                                    8-21

-------
100

90


80

70

_^
t! i! 60
o z
•4-1
 1 CC.SSS
'Z 2 CC.ESP FGD,FGD/PM
c 3 - FGD.ESP
« 4 FGD.FGD/PM
" 5 FGD.ESP

c
1
\ g
Unc


































5




2

x 1

\ B
XtJnc



























^ 4
•^ 2


^x 1

x Unc












-










j
.. 2
x B

Unc








      HSC-30
      HSC-75
HSC-150
HSC-400
                         Model Boiler
Figure 8.2-1
Capital costs  of control alternatives applied  to
HSC-fired model  boilers.
                             8-22

-------
no



100


90


80


70



^^
•/I tl 60
O •«•
tj^
Its s.
•»-> c
33 50
CD
•^
ono
N O
55
|§ 40
ss^
30
20
10
0
Boiler Size
30,75,150 400


M


.


—







_


~



-
-
f.
m












—— «"^—


















B Unc.SM Unc.FF
4 § 1 Unc.SSS OS.DS/PM
"Z 2 Unc.FF FGD.FGD/PM
3 S3 DS.DS/PM FGD.FF
fe 4 FGD.FGD/PM
^ 5 FGD-.FF
<

2
^ 1

^ B
^ unc



























t^mn^m^mtm^m












5
^ 3
"x, 4

2

_ 1

^ B
\ Unc






























5
^ 3
4
^ 2
1 ,
^ ' 3
^r
\ Unc











•••^•iH^










<. 1
^ 2
^ B
Unc









LSC-30
                    LSC-75
LSC-150
LSC-400
Figure 8.2-2.
                     Model  Boiler

           Capital costs of  control  alternatives applied to
           LSC-fired model boilers.
                      8-23

-------
combustion modification result in less than a 0.3 percent increase over
baseline costs for all coal-fired model boilers.
     Single mechanical collectors and sidestream separators are the least
expensive PM control methods for coal-fired boilers smaller than 73 MW
(250 x 10  Btu/hr).  Compared to the baseline alternative (which uses a
single mechanical collector), capital cost increases of 3 percent and less
are incurred in using the sidestream separator  (Alternative 1) control. A
relatively larger capital cost increase is incurred when an ESP or fabric
filter is required as is the case with Alternative 2.  For HSC model boilers
smaller than 73 MW, which require an ESP to meet the emission limits in
Alternative 2, cost increases over baseline are approximately lb to 16
percent.  Fabric filters used on LSC model boilers under Alternative 2
result in increases of 8 to 12 percent over the size range of 8.8 MW
(30 x iU6 Btu/hr) to 44 MW (150 x iU6 Btu/hr).
     An apparent diseconomy of scale for the HSC-150 model boiler ESP can be
noted when compared to the ESP's used on the smaller boilers.  The
relatively high cost for the larger ESP is explained by two factors:
(1) the uncontrolled particulate matter emission rate of the larger spreader
stoker is considerably higher than the smaller  underfeed and chaingrate
stokers at 2500 ng/J (5.82 lb/106 Btu) vs. 955  ng/J  (2.22 lb/106 Btu), and
(2) ESH costs are very sensitive to collection  efficiency requirements.
Since the ESP for the spreader stoker requires  a larger collection area
relative to the flue gas flow, the end result is a more expensive ESP on a
normalized cost basis.  Unlike ESP's, fabric filters are not sensitive to
collection efficiency and thus do not exhibit this anomaly.
      ihe largest capital costs are associated with the control alternatives
which require F6D systems.  Capital costs for SO., control via compliance
coal on HSC model boilers are relatively small  since working capital for the
higher priced fuel is the only capital cost.  For the HSC model boilers the
increases in capital costs over baseline associated with the use of
compliance coal are less than one percent of the model boiler cost.  Capital
costs jump sharply under Alternatives 4 and 5 which  require 90 percent S02
                                     8-24

-------
removal  with double alkali  FGD.   Alternative 4 requires a  combined S02  and
PM scrubbing system in which PM  is removed in the double alkali  scrubber.
Capital  costs for this system vary from a 19 percent to 45 percent increase
over baseline, with the largest  percent increase associated with the 8.8 MW
(30 x 10  Btu/hr) boiler.   Alternative 5 requires an ESP for PM  removal  in
addition to the FGD system  required for 50^ removal.  The  separate SO,,  and
PM systems required under Alternative 5 are the most expensive alternatives
evaluated.  Cost increases  range from 33 percent to 52 percent over baseline
for the boiler sizes examined, with the percent increasing as boiler size
decreases.
     The costs of FGD systems applied to boilers firing LSC are  similar to
costs for boilers firing HSC. The capital  costs of the dry scrubbing system
used in Alternative 3 are similar to the costs of the combined double alkali
SOp and PM control systems  despite the higher SO^ removal  of the double
alkali system (90 percent vs. 50 percent).   In fact, the dry scrubbing
system is more expensive than the double alkali system on  all  but the
smallest LSC model boiler.   This result is attributable to the capital  cost
of the fabric filter which  is included with the dry scrubbing system.  The
fabric filter collects waste solids generated in the spray dryer and fly
ash.  As with the HSC model  boilers, the maximum capital cost impacts are
incurred with use of separate S02 and PM control systems under
Alternative 5.  These capital cost increases over baseline range from
21 percent to 32 percent depending on boiler size with the percentage
increasing with decreasing  boiler size.
     All the FGD systems show strong economies of scale.   The combined  S0?
and PM control system used  on the HSC-150 model boiler costs only 35 percent
more than the same system applied to the HSC-75 model boiler, yet treats
over twice the gas flow and removes over five times as much PM.   The
economies of scale of FGD systems result in a narrowing of the cost
difference between Alternatives  2 and 4 as boiler size increases.  For the
small HSC-30 unit, use of a combined S02 and PM scrubber (Alternative 4)
results in a capital cost increase of 26 percent compared  to use of
compliance coal and an ESP  (Alternative 2).  At the HSC-150 size, however,
                                   8-25

-------
the same comparison reveals only a three percent increase in costs.   At the
larger size, most of the FGD cost is recouped by elimination of the
relatively expensive ESP since the combined system relies on the FGD
scrubber for PM removal.
     8.2.1.2  Large Coal-Fired Model Boilers.  Boilers larger than 73 MW
(250 x 10  Btu/hr) are subject to the existing NSPS, resulting in more
stringent baseline control requirements for the HSC-400 and LSC-400  model
boilers compared to the smaller coal-fired boilers.  Thus, the alternatives
examined for the large coal-fired boilers show much smaller percent
increases in cost over baseline since the baseline alternatives already
require considerable costs for emission control.
     For the HSC-400 model boiler, the baseline S00 emission limit of
                   c                              <-
516 ng/J (1.2 lb/10  Btu) requires an S0? removal efficiency of 79 percent.
Increasing this removal efficiency to 90 percent, as required under  Alterna-
tive 2, results in a 0.5 percent increase in capital costs.  Both the
baseline and Alternative 2 model boilers use combined S02 and PM scrubbing
with double alkali systems.  The most stringent alternative, Alternative 3,
requires a separate PM control system (ESP) in addition to a FGD system.
Alternative 3 results in a 7.8 percent increase in capital costs over
baseline for the HSC-400 boiler.
     For the LSC-400 model boiler, an additional alternative requiring
50 percent S02 removal via dry scrubbing was examined.  As was the case for
the LSC-150 model boiler, capital costs for dry scrubbing at this removal
are greater than the cost of a double alkali combined S02 and PM control
system at 90 percent S02 removal.  The most expensive system is a separate
S0? and PM control scheme with double alkali scrubbing for S02 and a fabric
filter for PM.  This system results in an 8.6 percent increase in costs over
baseline.  The higher percent increases over baseline for LSC compared to
HSC for equivalent emission reduction requirements are due to the different
systems used at baseline:  HSC boilers require FGD systems at the baseline
level; LSC boilers do not.  Thus, the alternatives beyond baseline for LSC
boilers require the addition of an FGD system resulting in greater capital
cost increases over baseline.
                                    8-26

-------
     8.2.1.3  Oil- and Gas-Fired Model  Boilers.   The capital  costs for
residual  oil-fired model  boilers, presented earlier in Table  8.2-3, are
graphically illustrated in Figure 8.2-3.   NO  controls are not specified in
                                            A
the insert table.  As is  the case for coal-fired boilers, capital  costs for
NO  control on residual-fired boilers are generally small.  In all cases,
  A
NO  capital costs are less than 1.5 percent of the total  model boiler cost.
  /\
     The sharp increase in capital  costs  associated with  the  use of FGD
systems is also evident for residual  oil-fired boilers.   For  boilers smaller
than 73 MW (250 x 10  Btu/hr) increases in capital costs  over baseline for
model boilers firing HDS  oils for S02 control  (Alternatives 1 and 2) range
from 3 percent to 4 percent for the range of boiler sizes examined.  This
cost increase is primarily due to increased working capital  costs associated
with the purchase of HDS  cleaned oils.  Application of FGD systems to the
same size residual oil-fired boilers  results in  substantially greater
capital cost increases relative to  capital costs of HDS oil use.
     Combined S02 and PM  systems (Alternative 3) result in capital cost
increases of 50 to 54 percent for boilers smaller than 73 MW.  The RES-30
model boiler uses a sodium throwaway  FGD  system  while the RES-150 uses a
double alkali system.  The separate S0? and PM control systems using FGD
scrubbers and ESP's are the most capital  intensive control systems
evaluated.  Increases in  capital cost over baseline are 109 percent and
83 percent for the RES-30 and RES-150 model boilers, respectively.
     For the large residual  oil-fired model boiler (RES-400)  a combined S02
and PM scrubbing system is required to meet the  baseline  S00  emission limit
                      6
of 344 ng/J (0.8 lb/10  Btu).  To meet this limit, an S02 removal  efficiency
of 75 percent is required.  Increasing the removal efficiency to 90 percent,
as required under Alternative 1, results  in a 0.4 percent increase in
capital costs.  Under Alternative 2,  an ESP is required for PM control in
addition to the FGD system.   Alternative  2 results in a capital cost
increase over baseline of 9.6 percent.
     An apparent diseconomy of scale  can  be noted when comparing the
normalized capital costs  of the uncontrolled RES-150 and  RES-400 model
boilers.   This result is  due to the different types of boilers used.  The
                                    8-27

-------
                                                    Boiler Size
                                               30.150
50
45
40
35
S£ 30
t_) ^.
4->
r- 3
 +j CO
CQ
•a
OJU3
N O
r— ^
>a o
e o
o° 20
•z. *«»•
15
10
5
0
-
-
-
-
-
-











> B HDS(1.6),Unc
'•£ 1 HDS(0.8),HDS/PM
. = 2 HDS(0.3),HDS/PM
4 « 3 FGD.FGD/PM
5 4 FGD,ESP
3
xf
X Unc













4
3
^i
\ Unc











2
/I
^B
Unc




RES-30 RES-150 RES-400
Model Boiler
                                                 400

                                              FGD(75),FGD/PM
                                              FGD.FGD/PM
                                              FGD.ESP
Figure 8.2-3.
Capital  costs of control  alternatives  applied
to residual  oil-fired model  boilers.   (June 1978 $)
                              8-28

-------
smaller boiler is a package unit while the larger boiler is  a field-erected
unit.  Since field erected units are more expensive than shop erected units,
the net result is a relatively expensive RES-400 boiler.
     For the distillate oil-fired model  boilers, cost increases  of 3.0 to
3.2 percent are incurred in going from baseline to Alternative 1.   These
capital costs are for the LEA and/or RAP systems used for NO  control, and
                                                            A
incremental working capital for purchase of HDS cleaned oils.  For natural
gas-fired boilers, only NO  controls are used in Alternative 1.  The cost
                          A
increases over baseline of the LEA and/or RAP systems used are 1.7 percent
and 0.8 percent for the NG-30 and NG-150 model  boilers, respectively.
8.2.2  Annualized Costs
     Model boiler annualized costs are presented in Tables 8.2-4 through
8.2-6.  The total model boiler annual  costs have been normalized by dividing
by the total annual heat input to the boiler.   The resulting numbers provide
a measure of the total  cost of firing a  unit of fuel  for a given model
boiler.  Percent increases in annualized costs  for each model boiler over
the uncontrolled case and the baseline case are also provided.   In the
discussions of annualized costs, the major emphasis will be  the  comparison
of percent increases in capital  cost over baseline for various control
alternatives.
     As indicated earlier in Table 8.1-10, annualized costs  are  the sum of
capital charges, operating costs, and maintenance costs.  Included in the
boiler operation costs  are fuel  costs.  The annualized costs provide a
measure of the total annual cost to build, operate, and maintain a boiler
and control system.
     Annualized costs of the boiler and  emission controls are a  function  of
boiler capacity factor.  All costs are presented on the basis of the
capacity factors specified in Table 8.1-8.  An  analysis of control costs  for
                                                                      p
boilers with lower capacity factors is contained in a separate report.
     8.2.2.1  Small Coal-Fired Model  Boilers.   This subsection discusses  the
annualized costs for coal-fired boilers  with thermal  input capacities of
73 MW (250 x 10  Btu/hr) and less.  The  total  normalized annual  costs for
each HSC and LSC model  boiler are graphically represented in Figures 8.2-4
                                    8-29

-------
                                         TABLE  8.2-4.
ANNUALIZED  COSTS OF  HSC MODEL  BOILERS'
(JUNE  1978$)
00
CO
o
Annual Ized Costs ($1000/yr)c
Control
Alternative
B
1
2
3
4
5
B
1
2
3
4
5
B
1
2
3
4
5
B
1
2
3
Model
Boiler
HSC-30-Unc,
HSC-30-Unc,
HSC-30-Unc,
a
HSC-30-FGD,
HSC-30-FGD,
HSC-75-Unc,
HSC-75-Unc,
HSC-75-Unc,
a
HSC-75-Unc,
HSC-75-Unc,
HSC-150-Unc
HSC-150-SCA
HSC-150-SCA
a
HSC-150-SCA
HSC-150-SCA
HSC-400-LEA
a
HSC-400-SCA
HSC-400-SCA

CC, SM
CC, SSS
CC, ESP

FGD/PH
ESP
CC, SM
CC, SSS
CC, ESP

FGD, FGD/PH
FGD, ESP
, CC, SM
, CC, SSS
, CC, ESP

, FGD, FGD/PM
, FGD, ESP
, FGDb, FGD/PM
, FGD, FGD/PM
, FGD, ESP
Uncontrolled
Boiler
1027
1027
1027

1027
1027
2076
2076
2076

2076
2076
3575
3575
3575

3575
3575
8817
8817
8817
N0x
Control
0
0
0

0
0
0
0
0

0
0
0
5.4
5.4

5.4
5.4
-8.4
53.9
53.9
so2
Control
67.9
79.2
79.2

458
419
169
197
197

616
572
339
396
396

873
778
1461
1540
1328
PM
Control
14.6
30.5
92.4

w/SO,
80. 0*
31.1
60.0
158

w/SO?
133 Z
77.8
126
331

w/SO?
311 2
w/S02
w/SO,
524 i
Total
1110
1137
1199

1485
1527
2276
2334
2432

2692
2781
3992
4103
4307

4453
4669
10270
10411
10722
Normalized
Total
7.04
7.21
7.60

9.42
9.68
5.77
5.92
6.17

6.83
7.05
5.06
5.20
5.46

5.65
5.92
4.88
4.95
5.10
Percent Increase In Costs
Over uncontrolled
Boiler
8.1
10
16

44
48
.7
.7

.6
.7
9.6
12.4
17.1


29.7
34.0
11.
14.
20.

24.
30.
16.
18.
21.
7
8
5

6
6
5
1
6
Over Baseline
Controlled Boiler
0
2.4
8.0

33.8
37.6
0
2.5
6.9

18.3
22.2
0
2.8
7.9

11.5
17.0
0
1.4
4.4
        Alternative requiring 502 SO- removal not applicable to HSC boilers since resulting emissions  would not meet baseline requirements.
        Baseline requires 78.9% S02 removal; other alternatives require 90% removal.
       cNegat1ve numbers Indicate net savings.
        Normalized total annual cost Is total annual cost divided by total annual heat Input to boiler ($/10 Btu).

-------
                                       TABLE  8.2-5.   ANNUALIZED COSTS  OF LSC MODEL BOILERS'

                                                        (JUNE  1978$)
CD
I
co
Annual Ized Costs ($1000/yr)a
Control
Alternative
B
1
2
3
4
5
B
1
2
3
4
5
B
1
2
3
4
5
B
1
2
3
Model
Boiler
LSC-30-Unc, Uhc, SH
LSC-30-Unc, Uhc, SSS
LSC-30-Uhc, the, FF
LSC-30-Unc, OS, DS/PH
LSC-30-Uhc, FGD, FGD/PH
LSC-30-Unc, FGD, FF
LSC-75-Unc, Uhc, SH
LSC-75-Unc, Unc, SSS
LSC-75-Unc, Unc, FF
LSC-75-Unc, OS, DS/PM
LSC-75-Unc, FGO, FGD/PH
LSC-75-Unc, FGD, FF
LSC-150-Unc, Unc, SH
LSC-150-SCA, Unc, SSS
LSC-150-SCA, Unc, FF
LSC-150-SCA, DS, DS/PH
LSC-150-SCA, FGD, FGD/PH
LSC-150-SCA, FGD, FF
LSC-400-LEA, Unc, FF
LSC-400-SCA, DS, DS/PH
LSC-400-SCA, FGD, FGD/PH
LSC-400-SCA, FGD, FF
Uncontrolled
Boiler
, 1199
• 1199
1199
1199
1199
1199
2448
2448
2448
2448
2448
2448
4135
4135
4135
4135
4135
4135
10180
10180
10180
10180
N0x
Control
0
0
0
0
0
0
0
0
0
0
0
0
0
5.4
5.4
5.4
5.4
5.4
-14.8
59.0
59.0
59.0
so2
Control
0
0
0
349
376
333
0
0
0
455
470
415
0
0
0
621
622
517
0
1050
964
756
m
Control
13.9
30.0
83.4
w/SO?
w/so;
83.4^
29.1
58.6
169
H/SO-
w/SO;
169 t
65.7
115
295
w/SO?
w/SO;
295 i
607
w/SO,
w/SO|
607 i
Total
1213
1229
1282
1548
1575
1616
2477
2507
2617
2903
2918
3032
4200
4255
4435
4760
4762
4952
10771
11289
11203
11602
Normalized6
Total
7.69
7.79
8.13
9.82
9.99
10.25
6.28
6.36
6.64
7.36
7.40
7.69
5.33
5.40
5.63
6.04
6.04
6.28
5.12
5.37
5.33
5.52
Percent Increase In Costs
Over Uncontrolled
Boiler
1.2
2.5
6.9
29.1
31.4
34.8
1.2
2.4
6.9
18.6
19.2
23.9
1.6
2.9
7.3
15.1
15.2
19.8
5.8
10.9
10.0
14.0
Over Baseline
Controlled Boiler
0
1.3
5.7
27.6
29.8
33.2
0
1.2
5.7
17.2
17.8
22.4
0
1.3
5.6
13.3
13.4
17.9
0
4.8
4.0
7.7
         aNegat1ve numbers Indicate net savings.

          Normalized total annual cost Is total annual cost divided by total annual heat Input to boiler ($/10 Btu).

-------
                           TABLE 8.2-6.
oo
 i
CO
ro
ANNUALIZED COSTS  OF OIL-  AND GAS-FIRED  MODEL BOILERS'

                        (JUNE  1978 $)
Annual 1 zed Costs
Control
Alternative
B
1
2
3
4
B
1
2
3
4
B
1
2
B
1
B
1
B
1
B
1
Model Uncontrolled
Boiler Boiler
RES-30-Unc, HDS(1.6), Unc
RES-30-LEA, HDS(0.8), HDS/PM
RES-30-LEA, HDS(0.3), HDS/PH
RES-30-LEA, FGD/Na, FGD/PM
RES-30-LEA, FGD/Na, ESP
RES-150-Unc, HDS(1.6), Unc
RES-150-LEA, HDSfO.8), HDS/PM
RES-150-LEA, HDS(0.3), HDS/PM
RES-150-LEA, FGD, FGD/PM
RES-150-LEA, FGD, ESP
RES-400-LEA, FGDd, FGD/PM
RES-400-LEA, FGD, FGD/PM
RES-400-LEA, FGD, ESP
DIS-30-Unc, Unc, Unc
DIS-30-LEA, HDS(0.3), Unc
DIS-150-Unc, Unc, Unc
DIS-150-LEA/RAP, HDS(0.3), Unc
NG-30-Unc, Unc, Unc
NG-30-LEA, Unc, Unc
NG-150-Unc, Unc, Unc
NG-150-LEA/RAP, Unc, Unc
1070
1070
1070
1070
1070
4368
4368
4368
4368
4368
12472
12472
12472
1117
1117
5260
5260
970
970
4364
4364
N0x
Control
0
-5.7
-5.7
-5.7
-5.7
0
-39.0
-39.0
-39.0
-39.0
-108
-108
-108
0
-0.5
0
71.3
0
-0.7
0
75.9
soz
Control
65.2
105
142
322
322
326
526
773
639
598
961
1018
952
0
49.7
0
304
0
0
0
0
($1000/yr)
PM
Control
0
w/SO_
w/SO*
w/SO<
96. 12
0
w/SO
w/so;
w/SO*
225 i
w/SO?
w/SO£
385 i
0
0
0
0
0
0
0
0
a
Total
1136
1170
1210
1386
1482
4694
4855
5055
4968
5153
13325
13382
13700
1117
1166
5260
5635
970
969
4364
4440
Percent Increase in Costs
Normalized0
Total
7.86
8.09
8.37
9.59
10.25
6.50
6.72
6.99
6.87
7.13
6.91
6.94
7.11
9.44
9.86
7.28
7.80
8.20
8.19
6.04
6.14
Over Uncontrolled
Boiler
6.2
9.3
13.1
29.5
38.5
7.5
11.1
15.7
13.7
18.0
6.8
7.3
9.8
0
4.4
0
7.1
0
negligible
0
1.7
Over Baseline
Controlled Boiler
0
3.0
6.5
22.0
30.5
0
3.4
7.7
5.8
9.8
0
0.4
2.8
0
4.4
0
7.1
0
negligible
0
1.7
        'Negative numbers Indicate net savings.

         Negative numbers Indicate alternative 1s less costly than baseline alternative  (see text).

        formalized total annual cost Is total annual cost divided by total annual heat  input to boiler ($/10 Rtu).

         75% efficiency required at baseline.

-------
                                         Boiler Size
                                   30.75.150
                                      400
10


9
8

7


tn ,
••-> 6
VI
O -P
3
r— Q.
3 I—"
C
< -5 5
CO
•o
IM O
•r" r™
to-bty
f: 4
0
3
2
1
0
B CC.SM FGD(79),FGD/PM


v





^






_




-
-
-





























, §! 1 CC.SSS
5 U 2 CC,ESP FGD.FGD/PM
4 S3 - FGD.ESP
% 4 FGD.FGD/PM
^ 5 FGD.ESP

2
C.
Unc





































5

X2
^ 1 q


n

Unc

























.
^^
2
^ — k
xX 1 •?

\R
D
Unc
















x^
x' 2
X B

Unc





HSC-30 HSC-75 HSC-150 HSC-400
Model Boiler
Figure 8.2-4.
Annualized costs of control  alternatives applied
to HSC-fired model boilers.  (June 1978$)

          8-33

-------
Boiler Size

10




9


8



7

5 6
o • — •
O -M
3
.— Q.
C
c 3 5
00
•o
OIVO
N O
ia «/>
I" ^
3
2
1
0






-


-



••

w

-


•.
-
-
-



























5 30,75,150 400
A
, ... B Unc.SM Unc.FF
J > 1 Unc.SSS DS.DS/PM
'•C 2 Unc.FF FGD.FGD/PM
= 3 DS.OS/PM FGD,FF
55 4 FGD.FGD/PM
5 5 F6D-.FF
2
X
/ ]

\" B
\ Unc













^^i^K^^^^^Bi^













_ 4
X 3

^

\ Unc




















4,3
2
/
NB
\ Unc








_









^Unc







  tSC-30
LSC-75
LSC-150
LSC-400
                       Model Boiler
Figure 8.2-5.   Annualized costs of control  alternatives
                applied to LSC-fired model  boilers.  (June 1978$)
                           0-34

-------
and 8.2-5.  The small  insert tables are repeated from the capital  cost
tables and provide a key to the technologies used to meet the emission
limits specified in each control  alternative.
     Most of the trends present in the capital  cost data are carried over in
the annualized costs.   In general, the percent  increases in annualized cost
over baseline are reduced compared to capital  cost increases over baseline.
This is primarily due to the effect of including relatively high boiler fuel
costs in the total annualized cost.  Including  fuel costs increases the
total cost of the uncontrolled boiler; thus, reducing the emission  control
costs as a percentage of the baseline or uncontrolled boiler cost.
     For the HSC model boilers, the sidestream  separator control system show
a cost advantage compared to ESP systems.  Increases in annualized costs over
baseline attributable to PM control for Alternative 1 (which requires a
sidestream separator)  are less than 1.5 percent.  Under Alternative 2 (which
requires an ESP) PM control costs are increased by 6.0 percent to
7.5 percent. As was mentioned in the discussion of capital  costs, the
normalized annual cost of ESP control  is relatively high at the 44  MW
(150 x 10  Btu/hr) boiler size due to the  higher uncontrolled emissions of
the spreader stoker compared to the smaller boilers.
     Percentage cost increases over baseline resulting from PM controls
applied to LSC boilers are slightly less than  are the increases for the same
controls applied to HSC boilers.   The primary  reason is the higher  annual
costs for uncontrolled LSC boilers compared to  uncontrolled HSC boilers.
Alternative 1 annualized cost increases are less than 1.5 percent over
baseline, while Alternative 2 cost increases are less than  5.7 percent over
baseline for all boiler sizes.  The normalized  cost of fabric filters does
not exhibit the cost anomaly at the 44 MW  (150  x 10  Btu/hr) boiler size as
was pointed out in the previous discussion of  ESP costs.  Uncontrolled
emissions rates have only a slight effect  on fabric filter costs.
Therefore, the difference in uncontrolled  emissions rates between boiler
types has little effect on fabric filter costs.
                                     8-35

-------
     The annualized costs of SCL control account for a higher percentage of
total model boiler costs than was the case for capital costs.  This  results
primarily from the higher annualized costs associated with utilization  of
compliance coal or F6D systems.  For the HSC model boilers, the costs of
obtaining coals which will meet the 860 ng/J (2.0 lb/106 Btu) emission  limit
specified in Alternatives 1 and 2 is significant, and accounts for 6.6  to
11.1 percent of the uncontrolled boiler annualized costs.  However,  these
costs are still less than FGD costs for boilers with thermal input
capacities of 44 MW (150 x 10  Btu/hr) or less.  Combined S02 and PM control
systems (Alternative 4) applied to HSC model boilers result in increases in
annualized cost over baseline of 12 to 34 percent, with the percentage
increasing as the boiler size decreases from 44 to 8.8 MW (150 to 30 x  10
Btu/hr).  The most expensive control systems for HSC model boilers are  the
separate S02 and particulate matter systems (Alternative 5) which result in
annualized cost increases over baseline of from 17 to 38 percent.  A
comparison of the combined vs. the separate SO,, and PM control systems
applied to HSC model boilers indicates that the separate systems cost 3 to
b percent more than the combined systems on an annualized basis.
     Costs of S02 control for LSC model boilers show the same trends as the
costs of control for HSC model boilers.  However, the percent increases over
baseline are somewhat  lower for the LSC-fired units.  This is to be expected
since much of the S0? control is inherent in the firing of LSC and the
increased cost of this fuel is included in the uncontrolled boiler cost.
For a given boiler size, SO,, control level and type of control system,  the
total cost of producing steam ($/unit mass of steam) is approximately the
same for the LSC and HSC units.  Dry scrubbing at 50 percent S02 removal
(Alternative 3) and double alkali scrubbing at 90 percent S02 removal
(Alternative 4) have virtually identical costs.  The only minor exception is
a slight (less than two percent) annualized cost advantage for dry scrubbing
at the smallest boiler size.  The combined S02 and PM systems show cost
increases over baseline of 13 to 30 percent for the LSC boilers.  The
combination of an FGD and a fabric filter results in increases of 18 to
33 percent.  (In both cases the percent increases as .the boiler size
                                     8-36

-------
decreases from 44 to 8.8 MW.)  The FGD and fabric filter systems cost 2 to
4 percent more than the combined S02 and PM systems (FGD or DS)  on an
annualized basis.
     8.2.2.2  Large Coal-Fired Model Boilers.   Because of more stringent
baseline control  requirements, and the economies of scale for large FGD
units, annualized cost increases over baseline for large (greater than 73 MW
(2bO x 10  Btu/hr)) coal-fired boilers are generally less than for the
smaller model  boilers.  For the HSC-400 boiler, increasing SC^ removal
efficiency from the baseline level of 79 percent to 90 percent required
under Alternative 2 results in a 1.4 percent increase in annualized costs.
Alternative 3  requires separate double alkali  FGD and ESP control systems
and results in a 4.4 percent increase over baseline for the 117  MW
(400 x 106 Btu/hr) unit.
     For the LSC model boilers, three alternatives beyond baseline were
evaluated.  Alternative 1 requires dry scrubbing with 50 percent $62 removal
while Alternative 2 requires a double alkali combined S02 and PM control
system with 90 percent S02 removal.   Consistent with annualized  costs of
controls for smaller boilers, Alternatives 1 and 2 show approximately equal
annualized costs, with increases over baseline of four to five percent for
the 117 MW (400 x 10  Btu/hr) boiler.  Alternative 3 requires a  separate S02
and PM control system (FGD plus a fabric filter) and results in  the greatest
cost impact.  The annualized cost increase over baseline for this
alternative is 8 percent for the 117 MW boiler.
     8.2.2.3.   Oil- and Gas-Fired Model  Boilers.  The annualized costs for
residual oil-fired model boilers presented earlier in Table 8.2-b are
graphically illustrated in Mgure 8.2-6.  NO  controls are not specified in
                                            A
the insert table.  NO  control costs using combustion modification are
                     A
generally less than one percent of the total model boiler annualized cost
for all  boiler sizes.
     At the smallest boiler size (8.8 MW or 30 x 106 Btu/hr) application of
FGD results in a sharp increase in annualized costs over baseline.  At this
size, cost increases over baseline for Alternatives 1 and 2, which require
HDS cleaned oil, are 3 and 6.5 percent, respectively.  Alternative 3, which
                                     8-37

-------
                                                        Boiler Size
      10 -
       9 _
       8 -
 nj Q.
 3 C
 C I— I
 C
-o ca
 01
 rsiiO
•r- O
       6 .
       5 -
       4 -
       3 .
       2 -
       1  _
-
-

—


"



-











4
3
1
B
Unc






<
•r
+
fl
C
I
a
+.
f—
<









£ 8
3 1
3 *>
- Z
-> 3
; «



^^
,. . 2

^1
^B
llnr

JU» 1
HDS(
HDS(
HDS(
FGD,
FGD,









JU
1.6), Unc
3.8),HDS/P
.3) ,HDS/P
FGD/PM
JSP



/?
CR
^^ B
Unc


                                                     400

                                                 FGD(75),F6D/PM
                                                 FGD,FGD/PM

                                                 FGD,ESP
                 RES-30
               RES-150



             Model  Boiler
RES-400
Figure  8.2-6
Annualized costs of control alternatives  applied
to residual oil-fired model boilers.  (June  1978 $)
                                  8-38

-------
requires a sodium throwaway FGD scrubber,  results  in  an annualized cost
increase over baseline of 22 percent.   However, at the 44 MW
(150 x 106 Btu/hr) size, the economies of  scale of FGD systems  tend to make
the costs of HDS and FGD comparable.   The  net result  is that annualized
costs under Alternative 3 (combined SO,, and PM FGD system) are  slightly less
than those for Alternative 2 (HDS cleaned  oil). The  most expensive emission
control  system is an I-GD system (90 percent SO,, removal)  and an ESP for PM
control.  This system, required under  Alternative  4,  results in a 10 percent
increasi
boiler.
increase  in  annualized  cost  over baseline for the 44 MW  (150 x  10  Btu/hr)
     At the 117 MW (400 x 106 Btu/hr)  boiler size the baseline alternative
requires 75 percent S02 removal  to meet a  344 ng/J (0.8 lb/10  Btu)  S02
emission limit.  A slight (0.4 percent) annualized cost increase is  incurred
at a 90 percent S0« removal  level  as  required under Alternative 1.   Double
alkali scrubbing with PM removal  is used at both  the baseline alternative
and Alternative 1.  To meet  the  PM emission limit of 11 ng/J
(O.Ub lb/10  Btu) required under Alternative 2, an tSP is  required  in
addition to the double alkali FGD system for S02  control.   This control
scheme results in a 2.8 percent  increase in annualized costs  over baseline.
     The annualized cost data for the  distillate  oil-fired boilers  shown in
Table 8.2-6, indicates cost  increases  over baseline ot 4.4 percent  and
7.1 percent for the DIS-30 and DIS-150 model  boilers, respectively.   These
cost increases are for the LEA and LEA/RAP systems for NO   control  and use
                                                        A
of HDb cleaned oils for S02  control.   The  higher  percent increase is
associated with the larger boiler. This is due to the HDS costs which are a
progressively larger percent of total  boiler costs as the  boiler size (i.e.
fuel consumption) increases.
     Costs of NOX control for natural  gas-fired boilers are small.   At the
8.8 MW (30 x 106 Btu/hr) size, LEA control results in a negligible
annualized cost impact.  At  the 44 MW (150 x 106  Btu/hr) size, the  LEA/RAP
control required under Alternative 1  results in a 1.7 percent increase in
annualized cost over baseline.  No S02 or  PM controls are  used on the
natural gas-fired model boilers.
                                     8-39

-------
8.2.3  Retrofit Cost Impacts
     Under the provisions of 40 CFR 60.14 and 60.15, an "existing facility"
may become subject to standards of performance if deemed modified or recon-
structed.  In such situations control devices would have to be installed  for
compliance with new source performance standards.
     Due to special considerations, the cost for installing a control  system
in an existing boiler facility is generally greater than the cost of
installing the control system on a new facility.  Since retrofit costs are
highly site-specific, they are difficult to estimate.  Examples of these
site-specific factors are availability of space and the potential need for
additional ducting.
     Configuration of equipment in the plant governs the location of the
control system.  For instance, if the boiler stack is on the roof of the
boiler house, the control system may have to be placed at ground level,
requiring long ducting runs from the ground level to the stack.  If the
available space at the plant is inadequate to accommodate the control
equipment, it may be necessary to install the equipment on the roof of an
adjacent building, thus requiring the addition of structural steel support.
It has been estimated that roof top installation can double the structural
costs for installation of the control system.  Foundations and structural
support costs typically amount to 2-3 percent of the control system capital
costs.
     Other capital cost components that may increase because of space
restrictions and plant configurations are contractor and engineering fees
                                                             18
(typically 15-25 percent of the control system capital cost),   construction
and labor expenses, and interest charges during construction (because of
longer construction periods).
8.3  OTHER COST CONSIDERATIONS
     This section addresses additional cost considerations that may be
incurred by boiler operators and/or regulatory agencies that have not been
addressed in Section 8.2.  Additional cost impacts are likely in two areas:

       •  Liquid and solid waste disposal, and
                                    8-40

-------
       •  Impact of compliance and reporting requirements.
     The major liquid and solid waste streams from an  uncontrolled boiler
are:  water softening sludge,  condensate blowdown, bottom ash  disposal,  and
coal pile runoff.  Bottom ash  collection, handling, and disposal  costs  have
been incorporated into the uncontrolled boiler cost estimates.   Bottom  ash
disposal costs were estimated  based on a non-hazardous waste classification
under RCRA regulations.   If industrial  boiler wastes are classified as
hazardous in the future,  then  the disposal  costs  and overall boiler control
costs (for coal-fired boilers) could increase significantly.
     Costs for treating the other three waste streams  were  not  quantita-
tively evaluated in this  study.   The costs  associated  with  the  disposal
problems are highly site-specific and are influenced by the following:

       •  Water softening sludge - raw water quality,  steam quality,
          water makeup rate.
       •  Condensate blowdown  -  effluent discharge quality  requirements,
          raw water quality, condensate blowdown  quantity.
       •  Coal  pile runoff - coal  quality,  meterological  conditions,
          effluent discharge quality requirements.
However, these costs would be  associated with the boiler itself and would
not affect the analysis  of incremental  cost impacts of air  pollution
controls.
     Impacts of compliance and reporting have been addressed in separate
        2
studies.
                                   8-41

-------
8.4 References

1.   Energy and Environmental Analysis, Inc.  "Impact Analysis of Selected
     Control Levels for New  Industrial Boilers.  (Energy, Environmental  and
     Cost Impacts)"  Prepared for U.S. Environmental Protection Agency,
     Research Triangle Park, NC.  Arlington, Virginia.  May 30, 1980.

2.   Kelly, M.E. (Radian Corporation).  Model Boiler Cost Analysis (Draft).
     Prepared for U.S. Environmental Protection Agency, Research Triangle
     Park, NC.  EPA Contract No. 68-02-3074.  February 17, 1981.  57 p.

3.   PEDCo Environmental,  Inc.  "Cost Equations for Industrial Boilers".
     Final report.  Prepared for U.S. Environmental Protection Agency,
     Research Triangle Park, NC.  EPA Contract No. 68-02-3074.
     January 1980.

4.   PEDCo Environmental.  "Capital and Operating Costs for Industrial
     Boilers".  Final report.  EPA 450/5-80-007.  Cincinnati, Ohio.
     June 1979.

5.   Gardner, R., R. Chang,  and L. Broz.  "Cost, Energy and Environmental
     Algorithms for NO , S02, and PM Controls for Industrial Boilers.   Final
     Report.  Prepared for U.S. Environmental Protection Agency, Research
     Triangle Park, NC.  EPA Contract No. 68-03 2567.  Acurex Corporation.
     Morrisville, NC.  December 1979.

6.   Dickerman, J.C. and K.L. Johnson.  Technology Assessment Report for
     Industrial Boiler Applications:  Flue Gas Desulfurization.  Final
     Report.  EPA 600/7-79-78i.  Radian Corporation.  Austin, Texas.
     November 1979.

7.   Kelly, M.E.  "Cost Calculations for FGD Costs Applied to Industrial
     Boilers".  Memo to Industrial Boiler File.  Radian Corporation.
     Durham, NC.  February 1980.

8.   Dickerman, J.C.  "Revisions to FGD-double alkali costs".  Memo to
     Industrial Boiler File.  Radian Corporation.  Durham, NC.  June 1980.

9.   Tighe, S.C. and M.S.  Jennings (Radian Corporation).  Mechanical
     Collectors for Particulate Control Stoker Coal-Fired Boilers.  Prepared
     for U.S. Environmental  Protection Agency, Research Triangle Park, NC.
     EPA Contract No. 68-02-3074.  January 13, 1981.  18 p.

10.  PEDCo Environmental,  Cost Algorithms for Particulate Control.  Prepared
     for U.S. Environmental Protection Agency, Research Triangle Park, NC.
     EPA Contract No. 68-02-3074.  Cincinnati, Ohio.  April 1980.
                                    8-42

-------
11.  Lim, K.J., _et aj_.  Technology Assessment Report for Industrial Boiler
     Applications:  NO  Combustion Modification.  Acurex Corporation.
     Mountain View, California.   EPA 600-7-78-178f.   December 1979.

12.  Devitt, T., P. Spaite, and  L. Gibbs.   "Population and Characteristics
     of Industrial/Commercial  Boilers in the U.S."  EPA-600/7-79-789a.
     PedCo Environmental.   Cincinnati, Ohio.  August 1979.

13.  Energy and Environmental  Analysis, Inc.  "Industrial  Fuel  Choice
     Analysis Model:   Primary  Model  Documentation".   Prepared for
     U.S. Environmental  Protection Agency,  Research  Triangle Park, NC.
     EPA Contract No.  68-02-3330.   June 1980.

14.  Energy and Environmental  Analysis, Inc.  "Impact Analysis  of Alterna-
     tive New Source  Performance Standards  II:   Energy,  Environmental, and
     Cost Impacts".  Prepared  for U.S. Environmental  Protection Agency,
     Research Triangle Park, NC.  EPA Contract No. 68-02-3330.
     December 19, 1980.

15.  Catalytic, Inc.   Technology Assessment Report for Industrial Boilers:
     Oil Cleaning.  EPA-600/7-79-178b.  Charlotte, NC.   1979.

16.  Menzies, W.R. (Radian Corporation).   Issue Paper No.  5:   S02 Standard
     for Oil.  Prepared  for U.S. Environmental  Protection  Agency,  Research
     Triangle Park, NC.   EPA Contract No. 68-02-3074.   November 3, 1980.
     34 p.

17.  Jennings, M.S.  "Adjustments  to Fuel Prices for Industrial  Boilers".
     Memo to Industrial  Boiler File.   Radian Corporation.   Durham, NC.
     February 1980.

18.  Neverill, R.B.  "Capital  and  Operating Costs of Selected Air Pollution
     Control  Systems".   CARD,  Inc.  Miles,  Illinois.   EPA-450/5-80-002.
     December 1978.  p.  3-3.
                                   8-43

-------
                            9.0  ECONOMIC IMPACT
     This chapter presents  economic  impacts  for a  range of alternative
regulatory options (ARO's)  analyzed  in  the development of the proposed
standards for new industrial  boilers.   The ARO's reflect an upper and lower
bound on the stringency of  the emission standards  which have been studies by
EPA.  The actual  proposed standards  resemble most  closely the emission
regulations analyzed under  ARO I  (the  least  stringent  ARO), which is
outlined below in Section'9.2.1 along with the  base  case and ARO V (worst
case) regulations.
     The discussion of impacts for industrial users  of steam includes the
impacts for both  ARO I and  V.   The results for  ARO I are essentially  the
same as the potential  impacts  for the proposed  standards.  The most
stringent regulation (ARO V)  illustrates a worst case  scenario to determine
whether potentially severe  economic  impacts  could  occur to either the
industrial users  of steam or  the producers of industrial steam-generating
and pollution control  equipment.
     The impact analysis examines cost-related  impacts and capital
availability issues.  Cost-related impacts include impacts on product price,
changes in the competitive  position  of  an industry (firm), and closure.
                                    9-1

-------
Capital availability issues refer to the ability of a firm to obtain capital
to finance the costs of control required by alternative regulatory options.
     Because the number of industries that could be affected by the proposed
standard is large, a two-fold approach was used to assess the level and
nature of the economic impact without undertaking a detailed analysis of
every industry.  The first component is performed on the general  industry
level (i.e., food or steel industries) for major steam-using industries.
Eight industry groups, which account for approximately 70 percent of total
industrial steam consumption and, therefore, which will bear most of the
cost burden of alternative regulatory options, are examined.
     The second component of the user impact analysis focuses on  the
economic impact on selected four-digit Standard Industrial Classification
(SIC) industries.  This focus is necessary since the major steam user
analysis utilizes industry averages to assess economic impact.  Because each
two-digit SIC industry grouping is composed of many four-digit SIC
industries, the industry average may not capture the impact of regulatory
options on specific four-digit SIC industries.  In addition, four-digit SIC
industries that are not part of the eight industry groups analyzed under
major steam users may be affected severely.  To remedy the situation, the
economic impact on selected four-digit SIC industries is examined.  The
industries chosen for this component of the analysis were selected by a
screening process designed to identify the four-digit SIC industries most
likely to experience adverse economic impacts.  By evaluating the economic
impact on industry groups most likely to be affected adversely, the impact
on other industry groups can be inferred to be less severe.
                                      9-2

-------
     Chapter organization -- This chapter is presented in two parts.   The
first section (9.1) presents profiles of the industries that will  be covered
in the economic impact analysis.   The second section (9.2) presents the
economic impact analysis of users.
9.1  INDUSTRY ECONOMIC PROFILES
9.1.1  Major Steam Users
     9.1.1.1  Introduction.   The  major steam users consist of the  following
eight industry groups:
  •  Food
  •  Textiles
  •  Paper
  •  Chemicals
  •  Petroleum refining
  •  Stone, clay, and glass
  •  Steel
  •  Aluminum.
These industries are examined because together they account for approxi-
mately 70 percent of total  industrial steam consumption and, therefore,
will bear most of the cost burden of an alternative regulation.   Except for
steel and aluminum, the industries examined are identified by a two-digit
SIC code.
     These eight industries generally use approximately 50 percent or more
of their energy consumption to generate steam and/or have steam costs that
comprise a major percentage of production costs.   Table 9-1 shows  1976
total fossil fuel consumption (excluding raw material and feedstock uses)
and the percent of total consumption accounted for by boilers in each of
the major steam user groups.  Approximately 48 percent of all industrial
non-feedstock fossil fuel consumption in 1976 was in boilers.  The paper,
food, and textiles industries consumed significantly more of their fossil
fuel in boilers than in other uses; the paper industry consumed approximately
87 percent of its fossil fuel in  boilers, the food industry 83 percent,
while the textile industry used 80 percent.  The chemicals and aluminum
industries also were well above the average for industrial boiler fossil
fuel consumption.
                                 9-3

-------
     One indication of the percentage of production costs that accounts for
steam generation is the ratio of steam consumption to dollar value of
product.  This ratio is computed by finding the quotient of annual steam
consumption and the value of shipments.  Ratios for 1976 values are listed
in Table 9-1.  The average steam consumption per dollar value of product
for industry is 0.0035 GJ (0.0033 MMBtu).  Therefore, the average cost of
steam per dollar of product for each major steam user depends on the average
cost of its steam (i.e., $/GJ or $/MMBtu).  The paper and chemical indus-
tries show the greatest steam consumption per dollar of product at 0.0215
GJ (0.0204 MMBtu) and 0.0160 GJ (0.0152 MMBtu), respectively.
     9.1.1.2  Economic Profile.  The major steam users represent a large
segment of the industrial sector.  This subsection shows the relationship
of the major steam users to aggregate  industry on three measures:  value
added in manufacturing, number of establishments, and total employment.
     Table 9-2 shows the value added by manufacture in each industry for
1976.  Value added is the dollar amount by which an industry increases the
sum of  its material inputs to produce  a finished product.  The largest
value added within the major steam users occurs in the food and chemical
industries.
     Only 22 percent of the total number of industrial establishments
within the United States are accounted for by the major steam users.  Of
this 22 percent, over three-quarters are food, chemical, and stone, clay,
and glass industry establishments.
     Major steam users employ about 30 percent of all manufacturing workers.
The food industry employs the largest  number of employees of the major
steam users.
     9.1.1.3  Projected Growth.  For this analysis, a macroeconomic fore-
casting model developed by Data Resources, Inc. (DRI) provides projections
of industrial economic activity in 1985, 1990, and 1995.  The measure of
industrial activity is the value added by manufacture statistic.  The value
added statistic is the difference between the value of shipments and the
total cost of materials.
     Table 9-3 presents projected value added statistics for the major
steam user groups.  The chemical industry has the highest projected growth.
                                  9-4

-------
                   TABLE 9-1.   FOSSIL  FUEL  CONSUMPTION CHARACTERISTICS OF THE MAJOR STEAM USERS

Total fossil .
fuel consumption '
SIC Code
20
22
26
28
u29
en
32
3312,
3315-17
3334,
3353-55


Industry
Food
Textiles
Paper
Chemicals
Petroleum
refineries
Stone, clay
and glass
Steel
Aluminum
Otherd
Total
1012
797.8
224.9
1,191.9
2,537.2
1,395.3
1,128.8
967.4
275.4
2,476.1
10,994.7
kJ (1012 Btu)
(756.3)
(213.2)
(1,129.9)
(2,405.3)
(1,322.7)
(1,070.1)
(917.1)
(261.1)
(2,347.3)
(10,423.0)
Industrial boiler3
fossil fuel consumption
as a percent of total
fossil fuel consumption
83
81
87
66
26
4
30
64
32
48
Steam consumption pera'c
dollar product
106 kJ per
$ value of
shipment
0.0037
0.0050
0.0215
0.0160
0.0042
0.0015
0.0063
0.0155
0.0008
0.0035
(106 Btu per
$ value of
shipment)
(0.0035)
(0.0047)
(0.0204)
(0.0152)
(0.0042)
(0.0014)
(0.0060)
(0.0147)
(0.0008)
(0.0033)
aEnergy and Environmental Analysis, Inc.  The Industrial Sector Energy Consumption Data Base (ECDB) for 1975 and
 1976.  December 15, 1980.
 Excludes raw materials, feedstock, and use in vehicles.  Includes coal, distillate and residual fuel oil, and
 natural gas.
CU.S. Department of Commerce, Bureau of the Census.  Annual Survey of Manufactures.  1976.
 Includes agriculture, mining, miscellaneous manufacturing, and construction industries.

-------
                                  TABLE 9-2.   CHARACTERISTICS OF THE MAJOR STEAM USERS
1C

Value added by manufacture3
SIC Code
20
22
26
28
29
32
3312,
3315-17
334,
3353-55


Industry
Food
Textiles
Paper
Chemicals
Petroleum
refineries
Stone, clay
and glass
Steel
Aluminum
Other
TOTAL
106 $
52,760
14,495
20,604
51,408
13,169
16,773
16,984
3,371
321,907
511,471
Percent
of total
10
3
10
10
3
3
3
1
63
100
Number of
Number
24,113
6,580
5,891
11,032
1,982
15,713
1,110
304
243,908
310,633
establishments
Percent
of total
8
2
2
4
1
5
1
1
78
100
Total employment
103
1,535.8
875.9
614.9
850.9
144.4
598.9
523.8
26.3
12,511.0
17,681.9
Percent
of total
9
5
3
5
1
3
3
1
71
100
   a
    U.S.  Department of Commerce, Bureau of Census, 1976.

-------
            TABLE 9-3.  INDUSTRIAL PRODUCTION GROWTH RATE PROJECTIONS3

Value added (109 $
Industry
Food
Textiles
Paper
Chemicals
Petroleum refineries
Stone, clay & glass
Primary metals
1974
50.2
N.A.
21.9
50.9
8.0
16.6
43.1
1985
69.7
N.A.
25.3
76.1
8.8
22.7
36.5
1990
79.2
N.A.
29.5
94.8
8.9
26.9
42.0
1975)
1995
88.3
N.A.
33.3
112.1
9.1
28.3
44.8
Annual
1974-85
1.3
3.1C
1.3
3.7
0.9
2.8
(0.2)
growth rate (%)
1986-90
2.6
4.1C
3.2
4.5
0.4
3.5
2.8
1991-95
2.2
3.0C
2.5
3.4
0.5
1.0
1.3
Data Resources, Inc.  TRENDLONG.2005 Forecast, December 1980 (Department of Energy,
Energy Information Administration's Annual Report to Congress 1980, Medium Case).
Average compound annual increase.

Growth rate projections for other manufacturing.
                                     9-7

-------
9.1.2  Selected Industries
     This section contains profiles of the seven four-digit SIC industries
selected for analysis.  Due to the large number of four-digit SIC industries,
screening criteria were used to target those industries most likely to
experience cost-related impacts and/or capital availability constraints.
     Industries most  likely to experience cost-related impacts are those
with a high steam cost to production cost ratio.  A high ratio usually
stems from one of two factors:  1) the production process is steam-intensive
or 2) the firm or industry has cyclic steam requirements, resulting in a
low capacity utilization of the boiler equipment.  Low capacity utilization
causes the capital cost component of unit steam costs to rise, yielding
high annualized costs per unit of steam.  Therefore, capacity utilization
and percentage of steam costs to total product costs are used as selection
criteria.
     Capital availability constraints occur when the cost of acquiring
funds is so high that a firm considers a project to be uneconomic or finan-
cially unattractive.  Capital availability is most often a problem for
relatively small firms.  Although some large firms may have excessive debt
burdens, lack of access to organized capital markets is more often charac-
teristic of small firms.  Thus, size is used to identify firms with poten-
tial capital availability problems.
     The following seven four-digit SIC industries are profiled in this
section:
  •  Beet sugar refining
  •  Fruit and vegetable canning
  •  Rubber reclaiming
  •  Automobile manufacturing
  •  Petroleum refining.
  •  Iron and steel manufacturing
  •  Liquor distilling.
Each profile contains an industry description and a discussion of recent
production trends.
     9.1.2.1  Beet Sugar Refining Industry.  The U.S. beet sugar refining
industry (SIC 2063) is characterized by relatively few producers.  Twelve
companies operate 44  refineries generally located in the midwestern and
Pacific States.1
                                 9-8

-------
     Refining operations are highly seasonal,  usually commencing in mid-
September and ending by mid- to late March.   The length of the refining
season varies from 120 to 220 days, depending on beet crop conditions.
During the average refining season of 180 days,  plants operate 7 days a
week, 24 hours a day.   Annual capacity utilization of boilers at a plant is
typically in the range of 30 to 45 percent.
     In recent years,  the annual  total output of the beet sugar industry
has been decreasing.   The number of plants in the industry also has decreas-
ed;  firms have closed less profitable plants, run operations as a coopera-
tive (i.e., where the  firm is owned by the sugar beet farmers) or, as a
last resort, entirely  terminated operations.
     The major explanation for decreased production is the increasing level
of cane sugar imports, which in 1978 accounted for over 45 percent or 4.5
million megagrams (5 million tons) of total  U.S.  sugar demand.   In the same
year, domestic beet sugar production claimed only 32 percent or 3.2 million
                                       2
megagrams (3.5 million-tons) of demand.    In view of this foreign competi-
tion, American firms must keep their prices  aligned with world sugar prices.
Although substitutes for sugar exist in the  form of highly concentrated
fructose, the beet sugar industry maintains  that foreign imports pose the
greater threat to solvency.
     Between 1974 and  1978,  world and U.S. sugar prices fluctuated widely,
ranging from $0.18 to  $1.43 per kilogram ($0.08 and $0.65 per pound) of
refined sugar.   Record prices were set in 1974 because of the world sugar
beet crop shortage.  The price of sugar for 1979 production was $0.40 per
kilogram ($0.18 per pound).
     Production of beet sugar increased significantly in the post-sugar
shortage years.  Since 1976, however, total  production has decreased due to
plant shutdowns and decreased production per plant.  In addition, total
sales and sales per plant have decreased steadily during the 1975 to 1978
period.
     Due to highly volatile prices and declining sales, most domestic
producers are not considering expansion.  Instead, they are focusing on
plant (and boiler) maintenance and/or replacements as well as company
consolidations when economically practical.   Due to the industry's small
profit margin, only the larger, more profitable firms that benefit from
economies of scale would consider investing in a boiler replacement.

                                 9-9

-------
     9.1.2.2  Fruit and Vegetable Canning Industry.  The fruit and vegetable
canning industry (SIC 2033) is highly competitive; hundreds of firms operate
over 500 canneries which are located in the fruit and vegetable growing
areas of the United States and which specialize in canning local raw produce.
California has the highest concentration of canneries -- accounting for 45
percent of domestically produced canned goods -- and is followed closely by
States in the northeastern and midwestern regions.  Despite this production
                                                                   4
concentration, canneries exist in almost every State of the nation.
     Fruit and vegetable canneries have highly seasonal production patterns.
The peak of the canning season is reached at the end of the growing season
— in late summer and early fall.  Although the range of operations varies
with each cannery, the average plant cans fruits and vegetables approximate-
ly 120 days per year.  Some plants have secondary, non-seasonal product
lines produced during the "off" season that help decrease average fixed
production costs.
     In recent years, the number of firms in the fruit and vegetable can-
ning industry has declined as large corporations acquire smaller firms.  In
fact, the canning industry is characterized by large numbers of small
canneries competing with a growing number of large canneries.  The canning
industry is consolidating for several reasons.  First, diversification of a
company's product line eases fluctuations in sales and product price and,
hence, profit.  Second, product diversification, or canning produce with
varying processing seasons, tends to increase the capacity utilization of a
company's plants.
     The most intense product competition to the canning industry stems
from two domestic markets:  fresh fruits and vegetables and frozen fruits
and vegetables.  The perishable nature of fresh produce renders that market
a less direct threat to the canning industry than frozen fruits and vege-
tables.  Unlike canned goods, processing frozen items requires only minimal
cooking, resulting in more flavor and enhanced product quality.
     Fruit and vegetable canning is an unpredictable industry.  Variables
such as weather conditions and crop yields dictate a cannery's profitabil-
ity each year.  Unseasonable frost, insufficient rainfall, and poor crop
planning can restrict the supply of fresh produce available for canning.
These factors resulted in price fluctuations and production swings between
1974 and 1978.  Total production of canned fruits and vegetables decreased

                                 9-10

-------
from 581,383,000 cases in 1974 to 516,034,000 cases in 1976.   Production in
1977 increased, however,  to 548,728,000 cases.
     Similarly, total  sales and sales per plant in 1974 were  high,  totaling
$3.4 billion and $6.6  million, respectively.   These amounts decreased in
1975 and 1976 to a low of $2.98 billion total sales and $5.72 million sales
per plant.   By the end of 1977, total sales and sales  per plant began to
increase,  approaching  the 1974 level.
     9.1.2.3  Rubber Reclaiming Industry.   The  rubber  reclaiming industry
(SIC 3031)  consists of eight producers operating eight manufacturing estab-
lishments.    Although  the bulk of reclaiming occurs in Ohio,  manufacturing
plants are  located throughout the eastern States.   This geographic  distri-
bution is  explained partially by the  availability of energy supplies and
the relative proximity to the tire and automotive industries.
     Rubber reclaimers buy old rubber tires,  inner tubes,  and other scrap
rubber materials and recycle them into a reusable form of rubber,  notably
tires and  floor mats.   Approximately  60 percent of all  reclaimed rubber is
                           o
used in tire manufacturing.    For several  manufacturers,  unmolded reclaimed
rubber is  the sole output, while for  others,  especially the larger  integrat-
ed firms,  unmolded recycled rubber is not a primary product but part of an
internal subprocess producing various rubber goods.
     For the past decade, the rubber  reclaiming industry has  suffered from
volatile sales, production fall-offs, and plant reductions.   A series of
external factors account  for these conditions.   More stringent ceilings on
the amounts of reclaimed  rubber allowed in car  tires have been the  largest
obstacle to the industry's expansion.  With the increasing popularity of
radial tires, which contain a much smaller percent of  reclaimed rubber than
standard ply tires, reclaimed rubber  consumption has decreased.  Although
reclaimers  still are heavily involved in tire manufacturing (e.g.,  farm
equipment), their importance in the passenger tire market has lessened.
     American reclaimers  face little  product competition from foreign
producers,  because only a small amount of reclaimed rubber is imported from
overseas.   Instead, the major competitors are two higher quality substi-
tutes:  new, domestically produced synthetic rubber and imported natural
rubber.
     Industry production  fluctuated during the 1974-1978 period.  In 1974,
production  was relatively high at 143,330 megagrams (168,900  short tons).

                                 9-11

-------
The following year, industry production decreased by 49 percent to 78,200
megagrams (86,200 short tons).  Since 1975, total production climbed gra-
dually, reaching 119,200 megagrams (131,400 short tons) in 1978.   Produc-
tion per plant followed a similar trend.
     Prior to 1973, the price of reclaimed rubber fluctuated only slightly.
Prices remained around the 1967 base year price of $0.25 per kilogram
($0.11 per pound).  In 1973-1974, increasing input costs, such as increas-
ing fuel import prices and rising transportation and collection costs of
scrap rubber, began exerting upward pressure on industry price levels.  By
1978, the price of reclaimed rubber was $0.37 per kilogram ($0.17 per
pound).
     After an initial fall in 1975, total sales and sales per plant have
risen gradually.  Although production per plant in 1978 was significantly
less than that in 1974, nominal sales per plant in 1978 were higher ($5.56
million versus $5.34 million).  With production dropping, selling prices
fluctuating, and production costs mounting, the industry is not considering
expanding productive capacity, but instead is using available funds to
replace capital assets.
     9.1.2.4  Automobile Manufacturing  Industry.  The U.S. automobile
manufacturing industry (SICs 3711, 3713, and 3714) consists of four firms
engaged in manufacturing and assembling "American-made" vehicles.  These
vehicles are produced in many States (and sometimes in other countries),
with the majority of production occurring in Michigan, Missouri, Ohio,
California, New Jersey, and Wisconsin.
     In addition to producing automobiles, the four manufacturing firms
produce light trucks, commercial trucks, buses, and other motor vehicles.
The following description focuses on the domestic automobile segment of the
motor vehicle industry, which accounts  for approximately 75 percent of
industry production.
     The profitability of the automobile industry is affected by the econo-
mies of scale realized by each firm in  its manufacturing process.  Economies
of scale in the automotive manufacturing industry occur when a firm produces
enough automobiles to decrease the fixed cost per unit output and, conse-
quently, total costs per unit output.   The most efficient way for a firm to
capture economies of scale is to integrate vertically its production process.
The level of integration differs by firm, ranging from minimal levels to

                                 9-12

-------
near total  integration.   A firm with minimal  integration could produce one
or two of the components needed, such as the  engine or alternator,  as well
as assemble the automobile.   A totally integrated firm could produce all  of
the major automobile components as well  as  the inputs used in producing
these components,  e.g.,  steel, glass, plastics.   Such a firm would  tend to
be larger and more productive than a less integrated firm.
     The profitability of automobile manufacturers and new car dealers also
depends upon consumer demand for automobiles.   Manufacturers attempt to
stimulate demand for their products by producing more than one car  class.
Many assembly lines are  geared to producing many different models in dif-
ferent price ranges at the same assembly plant.   This allows the manufac-
turer to produce more low-priced cars or more high-priced cars,  depending
on consumer demand.
     Since 1975, total production and production per plant have  increased.
The average price of new cars for the four  domestic manufacturers rose from
$4,202 in 1974 to $6,249 in 1978, an annual compounded increase  in  average
sales price of 10.4 percent per year.
     Except for the industry leader, the market shares of all domestic
manufacturers declined between 1974 and 1978, due to the increased  market
share of the dominant domestic producer as  well  as the increased sales of
imported automobiles.   The share of total sales accounted for by imports
increased by about one-third between 1974 and 1977, from 13.8 percent to
18.3 percent.
     9.1.2.5  Petroleum  Refining Industry.  The petroleum refining  industry
(SIC 2911) consists of 153 companies operating 289 domestic refineries.
Of these firms, 19 control over 70 percent  of the U.S. refining capacity.
     Although refineries operate in 41 States, approximately 27 percent of
the crude distillation capacity is concentrated in Texas.   California and
Louisiana,  the second and third largest petroleum refining States,  respec-
tively, jointly account for another 27 percent of crude distillation capa-
city.
     While U.S. production of refined oil remained relatively constant at
3.0 to 3.1 billion barrels per year between 1974 and 1978, consumption has
increased.   Domestic consumption of crude oil for 1978 equaled 6.9 billion
barrels or more than twice the amount of crude oil refined domestically.14
Consequently, the United States depends on imported sources of refined oil.

                                 9-13

-------
     The U.S. petroleum refining industry suffers from an operating cost
disadvantage when compared to European oil companies.  U.S.  oil companies
face higher taxes and labor costs than their European counterparts.  This
partially explains why the United States imports such a large share of its
oil requirements in the form of refined products (as opposed to crude).
Consequently, almost 27 percent of world refinery capacity additions are
taking place in Europe, while the United States accounts for only 0.9
percent.
     Production per plant decreased slightly during the 1974-1978 period.
In 1974, average production per plant was 12.06 million barrels per year;
by 1978, this average fell to 10.85 million barrels.  One explanation for
this decrease is the addition of 30 new plants in the past 5 years without
a corresponding increase in total production.    It appears that new plants
are being built with a certain amount of planned excess capacity.  Future
petroleum refining growth is expected to be concentrated in utilizing
excess capacity.
     The price of and total revenue from refined petroleum products has
almost doubled from 1974 to 1978.  The rise in crude oil costs accounts for
most of the  recent increases in refined product prices and the correspond-
ing value of their sales.  Total sales in this period have increased from
$54.8 billion to $105.6 billion.  Sales per refinery have increased over 70
percent from $211 million to $365 million.
     Production capacity in the U.S. petroleum refining industry is pro-
jected to grow at a rate of 1.73 percent per annum.    Using this figure as
an estimate  of the level of expansion within the industry, it appears that
the petroleum refinery industry could invest in new boilers for both expan-
sion and replacement purposes.
     9.1.2.6  Iron and Steel Manufacturing Industry.  The iron and steel
industry (SIC 3312) consists of integrated establishments that produce
basic steel  shapes in the form of semi-finished products such as ingots,
billets, blooms, and slabs or finished products such as steel strips, bars,
shapes, heavy structurals, and rails.  Establishments primarily engaged in
producing finished products from purchased iron and steel (e.g., non-inte-
grated) are  considered separate industries and are classified under SIC
codes 3315, 3316, and 3317.18
                                 9-14

-------
     Typically, operations at integrated steel  works involve manufacturing
iron from raw materials, refining the iron into steel, casting and milling
the steel into semi-finished shapes, and either selling the shapes to
non-integrated finishing facilities or hot rolling into finished products
at the works.     Integrated steel works range in size from large plants
using several  steel-making processes and finishing mills to small plants
using a single process and selling a semi-finished product.
     The industry is dominated by a few major producers.   The seven largest
companies produce approximately 70 percent of all  domestic steel.  The
remaining 30 percent is produced by numerous smaller companies,  many of
which operate only one facility.   There are about  260 integrated iron and
                                         20
steel establishments, spanning 36 States,    most of which are located in
the middle Atlantic and northeastern central regions.   The States with the
highest concentrations are Pennsylvania, Ohio,  Indiana, Illinois, and
Michigan.  With the exception of Michigan, these States are all  major coal
producers.   Locating plants close to coal  regions  reduces the expense of
obtaining coal that is used extensively in steel making.
     Over the last decade, the steel industry has  suffered from recession-
ary trends rooted in the 1950's.   Spiral ing costs  and restrained prices
have reduced industry profits to low levels, leaving major steel producers
with little capital for maintenance or expansion.   As a result,  domestic
steel producers have been postponing large capital commitments,  closing or
selling unprofitable operations,  reducing production levels, and merging
with other companies.
     The steel industry attributes its profit deterioration to several
factors, the most important of which is the increasing amount of low-cost
foreign steel  on the domestic market.   Over the last 5 years, imports have
                                                               00
increased from 13 to 18 percent of total domestic steel demand.     In 1978,
almost 1 of every 5 megagrams of steel used in the United States was pro-
duced outside of the country.  Many steel  manufacturers claim foreign steel
is being dumped on the U.S. market and have responded by discounting their
steel prices.   Other factors cited by industry are costs to meet environ-
mental and safety standards, inflationary wage and energy costs, and govern-
ment restraint of steel prices, including direct price controls from 1971
        23
to 1974.    Low steel prices have caused the industry to absorb increased
costs rather than pass them on to the consumer.

                                 9-15

-------
     Steel production from 1974 to 1978 was volatile, while prices increased
steadily after controls expired in 1974.  Because steel inventories were in
short supply throughout the world in 1974, domestic steel production was
high.  Prices were low due to price controls, causing total sales to mask
this healthy production.  In 1975 and 1976, demand for steel was signifi-
cantly lower, causing production to decrease.  Then, in 1977 as costs
increased markedly, imports flooded the domestic market and the steel
industry experienced losses for the first time since the 1930's.  An upturn
in the industry's economic profile took place again in 1978 due to a higher
production level and rising prices.  In 1978, the steel industry operated
at 86.8 percent of capacity compared to the low levels (78.4 percent) in
1977.24
     Most of the capital  investments made in recent years have been piece-
meal expansions and replacements rather than large-scale capacity additions
or new plants.  Projects  have attempted to cut costs by improving produc-
tivity, or boosting yields through modernization.  New plants may be needed
to meet projected demand  for steel in the 1980's, but it is uncertain
whether steel manufacturers will have the discretionary income to make
costly investments.  It is likely that future investments will follow the
same route of piece-meal  expansions and replacements unless present market
conditions improve.
     9.1.2.7  Liquor Distilling Industry.  The liquor distilling industry
(SIC 2085) is made up of  those establishments that manufacture liquor by
distillation or rectification.  They produce cordial and alcoholic cocktails
by blending processes or  by mixing liquors and other ingredients.  All
liquors except brandy are included in this category.
     The  liquor distilling industry is comprised of approximately 50 firms
                              or OC
that operate 100 distilleries.   '    The greatest concentration of distil-
ling plants is located  in the east south central States, Federal Region 4.
While distilleries are  located in 25 States, Kentucky has 27 percent of the
total number of domestic  plants.  California follows second with 11 percent
of the total.27
     The  number of plants that these firms operate has decreased in recent
years.  In 1972, 121 distilleries were operating; by 1977, this sum had
                                                    28
fallen to 104, a loss of  17 distilleries in 5 years.    Apparently, no new
facilities have been constructed in recent years.  Several factors may

                                 9-16

-------
explain the decreasing number of operating distilleries:   some plants are
old, inefficient, and not equipped for the major production modifications
often necessary to satisfy the demands of a changing market; furthermore,
larger firms often find operating fewer plants more efficient.
     The liquor distilling industry has much intra-industry, as well  as
inter-industry, competition.   Intra-industry competition  is seen in growth
rate and market share statistics of industry segments.   During the period
1960-1978, growth rates within the industry segments varied.  The cocktails
and mixed drinks segment increased at an annual  rate of 15 percent during
the years studied; cordials and liquors grew 9 percent  annually; "white"
goods (e.g., vodka, gin) grew at a 7.5 percent rate; and  "brown" goods
                                                                    29
(e.g., whiskey, bourbon) increased moderately at 1 percent annually.
     The major inter-industry competition to the distilled spirits industry
arises from the beer and wine industries.   Beer and wine  consumption  has
grown at the expense, to some degree, of "hard"  liquors.   Intensive adver-
tising campaigns and brand proliferation has helped beer  consumption  grow.
Wine, especially white wine,  once just a dinner beverage, has become  a
cocktail beverage as well.
     With the exception of 1976, total industry output  grew consistently
each year during the 1974-1978 period.  In 1974, bottled  output was 1,339.6
million liters (353.9 million gallons); by 1978, industry output reached
1,483.4 million liters (391.9 million gallons),  a 9.7 percent increase in 5
years.  Output per plant increased steadily from 11.2 million liters  (2.9
million gallons) in 1974 to 14.3 million liters (3.8 million gallons) in
1978.  The weighted average producer price of distilled spirits in 1974, as
reported by the Bureau of Labor Statistics, was $1.31 per liter ($4.95 per
gallon).  By 1978, the price grew to $1.66 per liter ($6.27 per gallon), a
27 percent increase or 6.6 percent annually.  Industry  sales volume is
derived from total output and price per liter (gallon).   Due to price
increases, total sales have increased at a greater rate than total output.
During 1974 to 1978, total  sales grew from $1.75 billion to $2.46 billion,
an increase of 40 percent.   Similarly, sales per plant  grew from $14.60
million in 1974 to $23.63 million in 1978.
     The selling price of some distilled liquors normally reflects produc-
tion costs incurred several years prior to sale.  The time lag between
production and maturation of whiskey can range from 2 to 8 years, depending
upon the quality of the product desired.  Therefore, higher manufacturing
                                 9-17

-------
costs incurred in one year may be reflected in higher prices in subsequent
years.
9.2  ECONOMIC IMPACT ANALYSIS
9.2.1  Regulatory Options
     Three regulatory cases are  considered in the economic impact analysis:
the  results of two  regulatory options are compared against a third case
composed of current emission regulations.  Current emission regulations are
the  current New Source  Performance  Standards (NSPS) for industrial boilers
with heat input capacity greater than 73 MW (250 MMBtu/hr) and State
Implementation Plan (SIP) emission  regulations for all smaller boilers.
Large boilers also  are  subject to SIPs when these regulations are more
stringent than the  NSPS.  Current emission regulations, presented in
Table 9-4, hereafter are referred to as the base case.
     Regulatory Option  I, summarized in Table 9-5, requires coal-fired
boilers between 15  and  44 MW (50 and 150 MMBtu/hr) to meet a 43 ng/J
(0.1 Ib/MMBtu) emission limit for particulate matter  (PM) and a 258 ng/J
(0.6 Ib/MMBtu) emission limit for nitrogen oxides (NO ).  Because no sulfur
                                                      A
dioxide ($02) regulation is specified, the coal-fired boilers between 15 and
44 MW (50 and 150 MMBtu/hr) are  subject to S02 SIP's.  Coal-fired boilers
between 44 and 73 MW (150 and 250 MMBtu/hr) are subject to a 860 ng/J
(2.0 Ib/MMBtu) emission limit for S02 and a 43 ng/J (0.1 Ib/MMBtu) emission
limit for PM.  Stokers  in this size category are subject to a 258 ng/J
(0.6 Ib/MMBtu) emission limit for NO ; pulverized coal boilers are subject
                                    A
to a 301 ng/J (0.7  Ib/MMBtu) emission limit.  Coal-fired boilers with heat
input capacity greater  than 73 MW (250 MMBtu/hr) are  required to reduce S02
emissions by 90 percent and cannot  have an S0« emission rate that exceeds
430  ng/J (1.0 Ib/MMBtu).  Because no minimum percent  reduction is required,
S02  emissions need  be reduced only  to meet the floor which is 258 ng/J
(0.6 Ib/MMBtu).  The PM emission limit for the larger boilers is 22 ng/J
(0.5 Ib/MMBtu); the NOV emission limit is 301 ng/J (0.7 Ib/MMBtu).
                      x *
     Regulatory Option  V,  summarized in Table 9-6, requires that boilers
between 15 and 73 MW (50 and 250 MMBtu/hr) reduce S02 emissions by 90 per-
cent and that the S02 emission rate not exceed 860 ng/J (2.0 Ib/MMBtu).  If
90 percent reduction results in  an  emission rate below 258 ng/J
*Regulatory Options  I.  and  V  are  used  to  bound  the analysis, of economic impacts.

                                      9-18

-------
(0.6 Ib/MMBtu), emissions need to be reduced by less than 90 percent, though
a minimum of 50 percent reduction is required.   These boilers are subject to
a 43 ng/J (0.1 Ib/MMBtu) PM emission limit.   Stokers in this size category
are subject to a 258 ng/J (0.6 Ib/MMBtu)  emission  limit for NO ;  pulverized
                                                              A
coal boilers are subject to a 301 ng/J (0.7  Ib/MMBtu) emission limit.
Coal-fired boilers with heat input capacity  greater than 73 MW
(250 MMBtu/hr) are subject to an SO- regulation similar to the one covering
boilers between 15 and 73 MW (50 and 250  MMBtu/hr).   The only variation is
with respect to the maximum SO* emission  limit. For boilers between  15 and
73 MW (50 and 250 MMBtu/hr), the S02 emission rate cannot exceed  860  ng/J
(2.0 Ib/MMBtu); for boilers greater than  73  MW  (250 MMBtu/hr), the S02
emission rate cannot exceed 430 ng/J (1.0 Ib/MMBtu).   The PM emission limit
for the larger boilers is 22 ng/J (0.05 Ib/MMBtu);  the NO  emission limit is
301 ng/J (0.7 Ib/MMBtu).
     The emission regulations for natural  gas are  the same for Regulatory
Options I and V:  they consist of a NO emission limit of 86 ng/J (0.2
Ib/MMBtu) for boilers greater than 15 MW  (50 MMBtu/hr).   Distillate and
coal-fired boiler regulations in Tables 9-5  and 9-6.
9.2.2  Major Steam Users
     The economic analysis of the major steam users  focuses on cost-related
impacts.  Capital availability considerations are  best examined on a  firm
level and, therefore, are covered only in  the analysis of selected
industries in Section 9.2.3.
     The economic analysis of the major steam users  is designed to determine
the effect of the regulatory options upon  the major industrial  steam  users.
As in Section 9.1.1, the major steam users consist of the following eight
industry groups:  food; textiles; paper;  chemicals  petroleum refining;
stone, clay, and glass; steel, and aluminum.
     These industries account for approximately 70 percent of total
industrial steam consumption and, therefore, will  bear most of the cost
burden of a regulatory option.  A profile of the major steam users is
presented in Section 9.1.1.
                                    9-19

-------
                                     TABLE 9-4.  BASE CASE AIR EMISSION REGULATIONS
                                                         (ng/j)
                                                       [(Ib/MMBtu)]
VO
I
N>
O
Fuel type Coal
Boiler size, MW 15-44
(MMBTU/HR) (50-150)
S02
Pollutant PM
NOx
SI
SI
SI
44-73
(150-250)
P's
P's
>'s
273
(>250)
516
(1.2)
43
(0.1)
301
(0.7)
Residual oil
15-44
(50-150)
SI
SI
SI
44-73
(150-250)
P'S
"s
P's
>73
(2:250)
344
(0.8)
43
(0.1)
129
(0.3)
Distillate oil
15-44
(50-150)
—
—
—
44-73
(150-250)
—
—
—
>73
(>250)
344
(0.8)
43
(0.1)
129
(0.3)
Natural
gas
215
(>50)
—
—
86
(0.2)

-------
                                                   TABLE 9-5.   REGULATORY OPTION I

                                                                    (ng/J)

                                                                 [(Ib/MMBtu)]
Fuel type Coal
Boiler size, MW 15-44
(MMBTU/HR) (50-150)
S02
Pollutant PM
NOx
SIP's
43 (
258
44-73
(150-250)
860
(2.0)
3-D
^301(0.6
273
(>250)
90* reduction,
430 (1.0)
ceiling.
251(0.6)
door.
22
(0.05)
/o.?r
Residual oil
15-44
(50-150)
SIP's


44-73
(150-250)
344
43(0.1)
129(0.3)
273
(2250)
(0.8)


Distillate oil
15-44
(50-150)
SIP's


44-73
(150-250)
150 (
—
86 (0.2)
273
(>250)
).35)


Natural
gas
215
(>50)
—
—
86
(0.2)
V£>
 I
to
             258(0.6) for stokers, 301 (0.7)for pulverized coal boilers.

-------
                                                  TABLE 9-6.   REGULATORY OPTION V

                                                                   (ng/J)

                                                                [(Ib/MMBtu)]
NJ
ho
Fuel type Coal
Boiler size, MW 15-44
(MMBTU/HR) (50-150)
S02
Pollutant PM
NOx
44-73
(150-250)
8(0(2.0) and
90% reduction.
if emissions
<2M.O(OG)
SO* reduction.
43 (
258/<
U)
;oi* (0.6
273
(>250)
430(10) and
90% reduction:
If emissions
< 258 (0.6)
90% reduction
22
(0.05)
'0.7)*
Residual oil
15-44
(50-150)



44-73
(150-250)
344(0.8]
43(0.1)
129 (0.3)
273
(2250)



Distillate oil
15-44
(50-150)



44-73
(150-250)
L50(0.35
—
86 (0.2)
273
(>250)



Natural
gas
215
(>50)
—
—
86
(0.2)
            * 258(0.6)for stokers, 301 (0.7)for pulverized coal boilers.

-------
     9.2.2.1  Methodology of Economic Impact Analysis.   The analysis of
major steam users consists of four steps:
  •  Step One ~ Evaluate price impacts of alternative  regulatory options
     on a general industry level  assuming the costs of  the regulatory
     option .are passed completely to the consumer (i.e.,  full  cost pass-
     through).
  •  Step Two — If price impacts are significant,  evaluate the ability of
     the industry to pass through the additional  costs.
  •  Step Three — If industry is able to pass through  costs,  assess the
     macroeconomic impacts of the price increase.   If industry is unable to
     pass on the additional  costs,  assess the ability of  the industry to
     absorb the additional costs.
  •  Step Four ~ If cost impacts are significant and the industry is
     unable to absorb the costs,  further analysis is warranted for the
     impact on both the other industries (non-major steam users) and two-
     digit SIC industries.
     The first step of the analysis evaluates the price impacts of alterna-
tive regulatory options on major  steam users.   When price impacts are
determined to be significant, they are evaluated  in terms of the conditions
contained in Steps Two through Four-
     The effect of a regulatory option on product price is calculated by
finding the product of the change in the cost of  new steam, the share of
steam affected by the regulatory  option, and the  amount of steam consumed
per dollar of output (see Figure  9-1).   The cost  impacts  are stated in real
terms.   The only real cost increase is assumed to be due  to new boiler,
pollution control, and fuel  costs.   All other production  costs are held
constant in real terms.
     When regulatory options are  applied,  the first component of the product
price calculation (the change in  the cost of new  steam) is affected.  The
cost of new steam changes due to  an option's effect upon  annualized boiler
and pollution control capital costs, annualized non-fuel  operating and
maintenance (O&M) costs, and annualized fuel costs.  When this new steam
cost change is multiplied by the  ratio of annual  steam  consumed (per unit
of output) to annual dollar value of shipment (per unit output), a gross
change in product price is derived.  Because a certain  percentage of the
product is produced with steam generated from existing  boilers, the cost
estimate is reduced by the proportion of new boiler steam to total steam
used within each industry group,  which results in an average steam cost for

                                 9-23

-------
                     FIGURE 9-1.   CHANGE IN PRODUCT PRICE DUE  TO  REGULATORY OPTION
 Product
^Price
{(
GJ (MMBtu) Steam/
 Product Price
    Steam
/Affected By
 Regulatory
   Option
)}
((
k
Cost of
New Steam
)
Regulatory
Option
(
Cost of
New Steam
)
Current
Regulation
                                                                                                                   ]
                                                                                                                    >
                                                                                                                   J
VO
i
       •  GJ (MMBtu)/Steam Product Price

             Annual  GJ (MMBtu) Steam/Annual  Output (Total  Number  of  Units)
             Annual  $ Value of Shipments/Annual  Output (Total  Number of  Units)

       •  Steam Affected by Regulatory Option

             Steam Consumption Affected by Regulatory  Option
             Total Steam Consumption

       •  Cost of New Steam

             Total Annualized Cost of Steam
             GJ (MMBtu)

-------
the industry.  The average cost of steam, instead of the marginal cost of
the new steam, is used to reflect actual industrial cost accounting proce-
dures of spreading the new steam costs over the entire product line.
     The ratio of annual steam consumed (per unit output) to annual dollar
value of shipment (per unit output) is computed by finding the quotient of
annual steam consumption and the value of shipments.   The ratio of annual
steam consumed to annual dollar value of shipment by industry is assumed to
remain constant over time.   Ratios employed in this analysis are listed in
Table 9-7.
     Table 9-8 shows the amount of total steam potentially affected by a
regulatory option.  The amount of steam increases over time as new boilers
come on-line and, potentially, are subject to the control level.
     9.2.2.2  Economic Impacts.
     9.2.2.2.1  Steam cost impacts.   The cost of new steam, unlike the
other components of the change in product price equation, is dependent upon
the specific regulatory, option chosen.   The cost of new steam for 1985 and
1990 is projected by the Industrial  Fuel Choice Analysis Model (IFCAM), an
energy demand model developed by Energy and Environmental Analysis, Inc.,
IFCAM simulates fuel choice decisions for the industrial major steam users
under different regulatory options.

     The projected total annualized cost of new steam per GJ (MMBtu) is
presented in Tables 9-9  and 9-10 for Regulatory Options I and V, by indus-
try and year.  These steam costs are functions of the boiler size and
capacity utilization distribution of the individual industry, the region in
which the industry is located, and the number of new boilers subject to the
regulatory option.
     There is a strong relationship between the boiler size/capacity utili-
zation distribution of an industry and average steam cost per GJ (MMBtu)
for an industry.   Industries that operate predominantly small boilers of
low capacity utilizations,  such as the textiles and stone, clay, and glass
industries, exhibit the highest cost per GJ (MMBtu).   Lower steam costs are
found in industries, such as paper and chemicals, firing larger boilers
because they can capture economies of scale in steam production, allowing a
lower steam cost per GJ (MMBtu).
                                 9-25

-------
         TABLE  9-7.   1976 STEAM CONSUMPTION PER DOLLAR PRODUCT

Steam consumption per '
dollar product
SIC Code
20
22
26
28
29
32
3312,
3315-17
3334,
3353-55


Industry
Food
Textiles
Paper
Chemicals
Petroleum
refineries
Stone, clay,
and glass
Steel
Aluminum
Other
Total
106 kJ per
$ value of
shipment
0.0037
0.0050
0.0215
0.0160
0.0044
0.0015
0.0063
0.0155
0.0008
0.0035
106 Btu per
$ value of
shipment
0.0035
0.0047
0.0204
0.0152
0.0042
0.0014
0.0060
0.0147
0.0008
0.0033
 Energy and Environmental  Analysis, Inc.  The Industrial  Sector Energy
 Consumption Data Base (ECDB) for 1975 and 1976.   December 15, 1980.

3U.S.  Department of Commerce, Bureau of the Census.   Annual Survey of
 Manufactures.   1976.
                                9-26

-------
       TABLE  9-8.   PERCENT OF NEW STEAM IN TOTAL STEAM CONSUMPTION3
Industry
Food
Textiles
Paper
Chemicals
Petroleum refineries
Stone, clay, and glass
Steel
Aluminum
Other
Total
1985
17
5
8
19
23
ob
9
13
18
16
1990
40
21
26
40
36
ob
26
41
37
36
aEnergy and Environmental  Analysis,  Inc.   Industrial  Fuel  Choice Analysis
 Model.   May 1980.

 The small amount of new steam demand in  the stone,  clay,  and glass industry
 is projected to be met by boilers smaller than 15 MW (50  MMBtu/hr).
                                 9-27

-------
  TABLE 9-9.   NATIONWIDE AVERAGE ANNUALIZED COSTS FOR NEW INDUSTRIAL
                      BOILERS BY INDUSTRY:   1985a
                       [1978 $/GJ (1978 $/MMBtu)]


                                                  	Regulatory Option
Industry                           Base Caseb          Ic           Vd

Food                               5.61 (5.92)    5.74 (6.06)    5.77 (6.09)
Textiles                           5.77 (6.09)    6.01 (6.34)    6.14 (6.48)
Paper                              5.02 (5.30)    5.13 (5.41)    5.17 (5.45)
Chemicals                          4.92 (5.20)    5.06 (5.34)    5.07 (5.35)
Petroleum refining                 4.98 (5.26)    5.15 (5.43)    5.27 (5.56)
Stone, clay, and glass                  e              e              e
Steel                              4.99 (5.27)    5.16 (5.44)    5.39 (5.69)
Aluminum                           4.77 (5.04)    4.90 (5.17)    5.14 (5.42)
Other                              5.18 (5.47)    5.35 (5.64)    5.46 (5.76)
Total                              5.07 (5.35)    5.21 (5.50)    5.28 (5.57)

aIFCAM steam costs annualized over 15 years at a 10 percent discount rate,
 pre-tax.
 Base Case emission limits are in Table 9-4
cRegulatory Option I emission limits are in Table 9-5.
 Regulatory Option V emission limits are in Table 9~6.
eThe small amount of new steam demand in the stone, clay, and glass
 industry is projected to be met by boilers smaller than 15 MW (50 MMBtu/hr).
                               9-28

-------
   TABLE 9-10.   NATIONWIDE AVERAGE ANNUALIZED COSTS FOR NEW INDUSTRIAL
                     BOILERS BY INDUSTRY:   1990
                       [1978 $/GJ (1978 $/MMBtu)]

Regulatory Option
Industry
Food
Textiles
Paper
Chemicals
Petroleum refining
Stone, clay, and glass
Steel
Al umi num
Other
Total
Base Case
6.55(6.91)
6.54(6.90)
5.54(5.84)
5.38(5.68)
5.62(5.93)
e
5.70(6.01)
5.46(5.76)
5.73(6.05)
5.64(5.95)
I
6.68(7.05)
6.74(7.11)
5.67(5.98)
5.56(5.87)
5.79(6.11)
e
5.74(6.05)
5.57(5.88)
5.93(6.25)
5.80(6.12)
V
6.58(6.94)
7.00(7.38)
5.72(6.03)
5.71(6.02)
5.91(6.23)
e
6.08(6.41)
5.75(6.07)
6.09(6.43)
5.93(6.25)

 IFCAM steam costs annualized over 15 years at 10 percent discount rate,
 pre-tax.

 Base Case emission limits  are in Table 9-4.

 Regulatory Option I emission limits are  in Table 9-5.

 Regulatory Option V emission limits are  in Table 9-6.
g
 All new steam demand in the stone,  clay,  and glass industry is projected
 to be met by boilers smaller than 15 MW  (50 MMBtu/hr).
                             9-29

-------
     Rising fuel costs typically cause the total industry steam costs,
displayed in Tables g_g  and g_io , to increase over time for both regula-
tory options.
     The cost impact of the regulatory options on steam generation is more
severe in Option V than in Option I due to the mandatory scrubbing require-
ment in Option V for smaller boilers.
     9.2.2.2.2  Price impacts.  The change in product price from the base
case for the regulatory options for each industry assuming full cost pass-
through is listed in Table 9-11.  The change in product price is less than
one percent for each of the major steam users.  Option V generally results
in the greatest price impact for all years.  The percent change in product
price for all industries increases over time, as more new steam capacity
comes on-line and is subject to controls.
     The greatest change in product price is found in those industries
where steam is a large fraction of total product value, such as chemicals,
paper, and aluminum.
     The regulatory options examined do not affect product price signifi-
cantly.  As Table 9-11 shows,  the product price increase is less than 1
percent for all industries and regulatory options examined.  This result is
due primarily to the relatively small fraction of total product value
accounted for by steam.
     The major steam users in  aggregate will not experience a significant
impact from the regulatory options.  This does not mean, however, that a
component of the major steam user industry group will not be affected
adversely.  The focus of the next section of this chapter is on selected
industries both within the major steam users and from other manufacturing
groups to assess whether smaller industry groups may be affected.
9.2.3  Selected Industries
     The analysis of selected  four-digit SIC industries forms the second
part of the economic impact analysis.  The economic analysis of selected
industries focuses on cost impacts, capital availability, and profitability
indicators.
     The major steam users analysis in Section 9.2.2 considers industry
averages in assessing economic impact.  Since each two-digit SIC industry
grouping is composed of many four-digit industries, the industry average
may not capture the impact of  regulatory options on each four-digit SIC

                                 9-30

-------
industry.   Also, the industry average may not reflect the impact on some
four-digit SIC industries that are not considered in the major steam user
analysis but that may be affected severely.   Smaller four-digit SIC indus-
tries within a two-digit SIC industry may experience different pollution
control costs and may vary in terms of financial  indicators.
     The industries chosen for this analysis  were screened to identify
four-digit SIC industries most likely to experience adverse economic impacts.
These selected industries are:
       •  Beet sugar refining
       •  Fruit and vegetable canning
       •  Rubber reclaiming
       •  Automobile manufacturing
       •  Petroleum refining
       •  Iron and steel  manufacturing
       •  Liquor distilling.
By evaluating the economic impact on industry groups most likely to be
affected adversely, the impact on other industry  groups  can be inferred to
be less severe.
     The selected industry section (9.2.3) is organized  into a summary of
results (9.2.3.1) and description of methodology  (9.2.3.2), followed by
individual analyses of each of the seven selected industries (9.2.3.3 to
9.2.3.9).
     9.2.3.1  Summary of Economic Impacts on  Selected Industries.   The
economic impacts of the regulatory options on the seven  selected industries
are summarized in Table 9-12 for the base case,  Regulatory Option I, and
Regulatory Option V.   Regulatory Option V represents the most stringent
control level examined and, therefore, generates  the maximum economic
impacts. The change in product cost is 1 percent  or less for all industries
except beet sugar, which has a 3.9 percent increase under Regulatory Option
V.
     Return on assets does not vary significantly between the base case and
Regulatory Options I and V for the selected industries except for liquor
distilling and beet sugar refining.   Return on assets for liquor distilling
decreases from 1 percent in the base case to negative 0.5 percent under
Regulatory Option V.   Beet sugar changes from a positive 0.9 percent return
                                 9-31

-------
             TABLE 9-11.  CHANGE IN PRODUCT PRICE FROM THE BASE CASE
                                  (percent)

Industry
Food
Textiles
Paper
Chemicals
Petroleum refining
Stone, clay, & glass
Steel
Aluminum
Other
Total
1985
Regulatory
Ia
0.008
0.006
0.018
0.040
0.016
—
0.009
0.025
0.002
0.008

Option
Vb
0.010
0.009
0.024
0.043
0.029
--
0.023
0.073
0.004
0.013
1990
Regulatory
Ia
0.020
0.021
0.074
0.116
0.027
—
0.002
0.072
0.006
0.020

Option
Vb
0.004
0.027
0.101
0.207
0.045
--
0.062
0.187
0.011
0.036
 Regulatory Option  I  emission  limits are in Table 9.5.

'Regulatory Option  V  emission  limits are in Table 9-6.
                                  9-32

-------
     TABLE  9-12.   SUMMARY OF CHANGE IN PRODUCT COST AND RETURN ON ASSETS
                  FOR MODEL PLANTS IN SELECTED INDUSTRIES:   1990
                                (percent)
Increase in product
cost over base case
Regulatory Option

Beet sugar refining
Fruit & vegetable
canning
Rubber reclaiming
Automobile manufacturing
Petroleum refining
Iron & steel
manufacturing
Liquor distilling
Ia
0.40
0.05
0.10
0.00
0.08
0.01
0.24
vb
3.90
0.05
0.60
0.02
0.08
0.07
0.64
Return on assets
Base
Case0
0.86
2.32
4.09
8.10
5.94
3.37
1.26
Regulatory
Ia
0.56
2.32
3.58
8.09
5.92
3.36
0.68
Option
vb
(4.00)
2.32
1.02
8.04
5.92
3.32
0.50
Regulatory Option I emission limits are in Table 9-5.
Regulatory Option V emission limits are in Table 9-6.
Base Case emission limits are in Table 9-4.
                                9-33

-------
on assets under the base case to a negative 4.0 percent under Regulatory
Option V.
     The analysis of capital availability examines the ability of the model
firm to finance the new boiler investment.  The coverage ratios and debt/
equity ratios were calculated for the base case, Regulatory Option I, and
Regulatory Option V.  The  ratios did not vary significantly for any of the
seven selected industries.   If current financing schemes (i.e., split
between debt and equity financing) for each industry are assumed to continue,
the ratios are considered  to be above "acceptable" levels under Regulatory
Options I and V.  This indicates that the industries should be able to
absorb additional financing  of new boiler investments without undue weaken-
ing of the solvency position of the industries.
     9.2.3.2  Methodology.
     9.2.3.2.1  Cost and profitability impacts.  The following three steps
are used to estimate the cost impact of regulatory options on a selected
industry:
  •  Step One — Define a  model plant for the selected industry.
  •  Step Two — Evaluate  the cost impacts for the model plant, assuming
     full cost absorption.
  •  Step Three — Evaluate  the impacts on the profitability of the model
     plant.
Each step is described below.
     Model plant.  The selected industries analysis focuses on model plants
to measure the economic impact of regulatory options on each industry.  The
model plant represents a typical plant for the segment of each industry
that might be considering  a  boiler investment either as boiler expansion or
replacement.  A model plant  is used since it is difficult to obtain precise
details about the expansion  and replacement plans of actual firms.
     For this analysis, each plant within the firm is assumed to be identi-
cal in steam use and product output.  Each plant employs the same process,
produces equal amounts of  output, operates identically configured boilers,
and consumes equal amounts of steam.  The fuel type burned in the existing
boiler(s) of the model plant is determined by industry sources.  The fuel
type of the replacement or expansion boilers is based on industry trends
and projections from IFCAM based on the combustor's size, location, and
applicable energy and environmental regulations.

                                 9-34

-------
     The following production characteristics for the model plant are
 supplied:
  •  Plant Output/Year — average product output per year in those plants
     more likely to invest in new boilers.
  •  Price (Cost)AJnit of Output — the historic, average selling price per
     unit, in real 1978 dollars.
  •  Plant Sales/Year — plant output per year multiplied by price per unit
     of output.
  •  Plant Earnings/Year -- plant sales per year multiplied by a derived
     profit margin (percent return on sales).  This figure estimates the
     profitability of the model plant.
     Product cost calculation.   The effect of regulatory options on product
 cost is calculated by finding the product of the change in the cost of new
 steam, the share of steam affected by the new regulation, and the amount of
 steam consumed per dollar of output.   The cost impacts are stated in real
 terms.  The only real  cost increase is due to new boiler and fuel costs;
 all other real production costs are held constant.
     Profitability impacts.   The additional  costs due to a regulatory
 option will affect the profitability of an industry.   This impact will  be
 assessed by examining the following two financial indicators for the model
 plant:
  •  Net Profit After Taxes (Net Income).   Profit after all  costs and taxes
     have been deducted.
  •  Return on Assets.  Net income divided by total assets,  converted to a
     percent form.
 Both of these indicators are analyzed for the base case and for the impact
 cases (Regulatory Options I and V).   The change in indicators due to regula-
 tory options is  a measure of the ability of the model plant to absorb the
 additional  costs of a regulatory option.
     Net income  is calculated by subtracting expenses from total sales to
 derive gross profit and then taxes are subtracted from gross profit to
 equal net income.   Regulatory options could affect the amount of expenses,
which would alter net income.   Return on assets is derived by dividing net
 income by total  assets for the model  plant and converting to a percent
 form.  Alternative regulatory options could affect net income, resulting in
a change in return on assets.
                                 9-35

-------
     9.2.3.2.2  Capital availability.  Capital availability constraints may
result if regulatory options create a need for financing additional pollu-
tion control investments.  The following two steps are used to evaluate
whether capital availability will be a constraint on a selected industry:
  •  Step One — Define financial indicators for a model firm.
  •  Step Two — Evaluate the ability of a firm to finance pollution con-
     trol investments.
     The firm is the focus of the capital availability analysis because
decisions involving large capital expenditures are made at the corporate
level.  Depending  upon the state of corporate cash reserves and the rela-
tive costs  of various financing tools, a firm will choose a combination of
internal and external financing instruments to meet the additional invest-
ments required to  comply with regulatory options.
     The capital availability analysis focuses on the following two finan-
cial indicators, which measure each  industry's financing ability:
  •  Coverage Ratio — the number of times operating income (earnings
     before taxes  and interest expenses) covers fixed obligations  (annual
     interest on debt instruments and long-term leases).
  •  Debt/Equity Ratio — a measure of the relative proportions of two
     types  of external financing.

     These  two indicators are analyzed for both the base case and  the
regulatory  options.  The change in indicators due to regulatory options  is
analyzed to determine how difficult  it will be for the firm to meet finan-
cial requirements  for the pollution  control equipment investment.
     The cash flow coverage ratio is calculated by dividing operating
income by fixed obligations, both of which could change as a  result of
alternative regulatory options.  If  the coverage ratio remains above the
3.0  standard benchmark,  the cost of  capital can be assumed to be above
"acceptable" levels.  However, as the coverage ratio falls, the cost of
obtaining capital  will rise.
     The debt/equity ratio is calculated by dividing total debt by total
equity of the firm (book values).  The incremental debt incurred from
financing the pollution  control required by the regulatory options is added
to the base case debt; the incremental equity issued to finance the remainder
of the investment  is added to the base case equity.  A new debt/equity ratio
                                 9-36

-------
then is calculated and the change is analyzed to assess the effect of the
regulatory options on the firm's capital structure.
     To determine the coverage and debt/equity ratios under alternative
regulatory options, five financing strategies, which differ by the percent-
ages of the investment financed by debt versus equity, are considered.
(Note that for the changes in coverage ratios and debt/equity ratios, 100
percent external financing is assumed.)  The external financing scenarios
are:
  •  Zero percent new debt,  100 percent new equity
  •  25 percent new debt, 75 percent new equity
  •  50 percent new debt, 50 percent new equity
  •  75 percent new debt, 25 percent new equity
  •  100 percent new debt, zero percent new equity.
     The financial indicators.generated for this analysis were derived from
a variety of published sources.   Robert Morris Associates'  Annual  Statement
Studies was consulted for composite industry financial data.   More specific
corporate figures were collected from Moody*s Industrial  Manuals and Form
10-K's and Annual Reports on file at the Securities  and Exchange Commission.
     9.2.3.3  Beet Sugar Refining Industry.
     9.2.3.3.1  Model firm and plant description.  The major characteristics
of the model firm for the beet sugar refining industry are listed in Table
9-13.  The model firm is made up of four plants, which are located in the
north central United States  (Federal Region 8).   Each plant is identical in
its steam use and product output.
     Total annual firm production is 326,600 megagrams (360,000 tons) of
sugar, with annual sugar sales at $126 million,  assuming that sugar sells
for $38.60 per hundred kilograms ($17.50 per hundred pounds) and that none
of this sugar is added to existing inventories.   Annual profits are 1.74
percent of total sales or about $2.2 million.  Comparing these figures to
the 1978 U.S. sugar demand,  this firm satisfies slightly more than 3 per-
cent of total demand and constitutes about 10 percent of the beet sugar
market.
     The model plant boiler house consists of four fossil fuel-fired boilers
with a total heat input capacity of 132 MW (450 MMBtu/hr).  Table 9-13
describes the individual boilers.   The three new boilers are coal-fired
units replacing similarly sized oil-fired boilers.  Each new boiler has a
                                9-37

-------
             TABLE 9-13.  MODEL FIRM AND PLANT CONFIGURATION:
                           BEET SUGAR INDUSTRY
Model firm
       Financial data
       Average bond rating:        Baa
       Coverage ratio:             4.0
       Debt/equity ratio:          0.41
Model plant
       Production data
       Plant output/year:          81,650 megagrams (90,000 tons)
       Price/unit output:          $38.60/hundred kilogram wt.c
                                      ($17.50/hundred pound wt.)
       Plant sales/year:           $31.50 million
       Plant earnings/year:        $0.548 million '
       Boiler configuration
       Total firing rate:          132.1 MW  (450 MMBtu/hr)
       No. of boilers:             4
       Federal region:             8
       Characteristics  of  individual  boilers
                                                       Boiler
                                                                       4
       Capacity, MW                          40.1    40.1    40.1     11.7
                 (MMBtu/hr)                  (137)   (137)   (137)    (40)
       Fuel type                             coal    coal    coal   residual
                                                                    fuel oil
       Annual capacity  utilization,  percent   45      45      45       25
       Replacement, expansion  or
         existing                            	replacement	  existing
aBased upon 1978  values.
 Based upon the average production  of  the portion of the  industry most
 likely to invest in  a new  boiler.
Expressed in 1978 $.
dBased upon the 1977  return on  sales ratio of 1.74 percent.

                                 9-38

-------
heat input capacity of 40 MW (137 MMBtu/hr) and an annual capacity utiliza-
tion of 45 percent.  Approximately 95 percent of total steam generated for
the plant is from these new boilers.   These three new boilers also are used
for process heat and electricity generation.
     9.2.3.3.2  Financial analysis.   The financial analysis of the beet
sugar industry shown in Table 9-15 focuses on 1974 to 1977.   During this
period, the industry's average annual net profits were positive, ranging
from $4.2 million to $22 million.   The period under consideration does not
include 1978, since the negative profits realized in 1978 are considered
atypical for the industry.
     From 1974 to 1977, the industrial net profit margin ranged from 1.74
percent to 4.14 percent, averaging 3.4 percent.   Return on total assets
ranged from 2.27 to 11.71 percent.   The beet sugar companies realized
higher earnings during the post-sugar shortage years of 1975 and 1976.
     According to industry sources,  1978 was an unprofitable — and atypical
year for the beet sugar industry due to low sugar prices.   In that year,
the industrial "average" net earnings* were negative; consequently, the
industry's ratios (i.e., return on assets and net profit margin), all of
which are a function of the industry's net earnings, were negative.
     Capital availability is analyzed using debt/equity and coverage ratios.
The model firm has a debt/equity ratio ranging from 0.25 to 0.41 for 1974
to 1977 and a coverage ratio ranging from 24.6, in post-sugar shortage
1976, to 2.10 in 1978.
     9.2.3.3.3  Regulatory option results.   In the base case, all the
plant's boiler replacements are subject to SIP emission regulations.
Industry representatives expect that any new boilers of this size will fire
     25
coal.    IFCAM also projects that coal is the least-cost fuel type and that
a low sulfur western coal is the least-cost coal type.  A Venturi scrubber
will be installed to ensure that PM emissions do not exceed the level
allowed by local regulations.   In Regulatory Option I a Venturi scrubber is
also chosen, while in Regulatory Option V, FGD controls are required.
     Replacement boilers consist of three 40 MW (137 MMBtu/hr) units,
operating at 45 percent annual  capacity utilization.  Table 9.15 shows the
pre-tax 1990 boiler and pollution control costs for the regulatory options
^Defined as the average of those firms most likely to invest in a new
 boiler, i.e., the six largest firms in the industry.
                                 9-39

-------
                TABLE  9-14.   FINANCIAL ANALYSIS:   BEET SUGAR  INDUSTRY3

Financial
indicator
Capital expenditures
Total assets (106$)
Capital expenditures/
firm (106$)
Capital expenditures/
total assets (%)
Profitability
Net profit after
taxes (106$)
Return on assets (%)
Return on equity (%)
Return on sales (%)
Trends in dividends
($ per share)
Net earnings before
interest and taxes
(106$)
Capitalization
Interest on fixed
obligations (106$)
Coverage ratio
Rating on bonds
Long-term debt (106$)
Stockholders' equity
(106$)
Debt/capitalization (%)
Debt/equity ratio

1974
166.523
9.493
5.7
14.21
8.53
14.60
4.14
1.12
N.A.C
N.A.
N.A.
N.A.
36.37
89.27
28.95
0.4074

1975
188.404
13.411
7.1
22.07
11.71
16.70
4.00
2.00
N.A.
N.A.
N.A.
N.A.
29.99
101.94
22.73
0.2942
Year
1976
204.321
16.963
8.3
13.82
6.76
14.80
3.57
1.67
26.39
1.07
24.62
N.A.
28.32
115.13
19.74
0.2460

1977
185.006
10.720
5.8
4.20
2.27
4.70
1.74
0.94
17.09
3.68
4.65
Baa
34.38
89.23
27.81
0.3853

1978
186.971
8.153
4.4
(0.61)b
(0.32)
(0.70)
(0.25)
0.65
9.58
4.57
2.10
N.A.
35.74
87.27
29.05
0.4095
Average,
1974-1978
186.245
11.748
6.3
10.74
5.77
10.02
2.64
1.28
17.69
3.11
5.69
N.A.
32.96
96.57
25.45
0.3413
aAverage per firm estimates (Securities and Exchange Commission; EEA estimates).
 Nominal terms.
 Numbers in parentheses represent negative amounts.
CN.A.  = Not Available.
                                      9-40

-------
            TABLE 9-15.   1990 BOILER COSTS:   BEET SUGAR MODEL PLANT
Costs
 Base Case
                                                      Regulatory Option
Total  boiler and
  pollution control
  capital  costs, 1978 $
20,650,000
Annualized total
  boiler cost,  $/GJ ($/MMBtu)

   Capital


   O&M


   Fuel


   Total


Coal type, ng SOy/J
(Ib S02/MMBtu)   *

Control  Technology

   S00
   1.38
  (1.46)

   1.65
  (1.74)

   1.07
  (1.13)

   4.10
  (4.33)

 718.00
  (1.67)
   PM
Venturi
Scrubber
20,880,000
   1.40
  (1.48)

   1.72
  (1.81)

   1.07
  (1.13)

   4.19
  (4.42)

 718.00
  (1.67)
Venturi
Scrubber
23,475,750
   1.60
  (1.69)

   2.10
  (2.22)

   1.21
  (1.28)

   4.92
  (5.19)

 374.00
  (0.87)
                                      Double Alkali/
                                      Mechanical
                                      Collector
 Base case regulations are the applicable SIP's for all  pollutants.
 Regulatory Option I regulations for these boilers are SIP limits for
 S02; 43 ng/J (0.1 Ib/MMBtu)  for PM; and 258 ng/J (0.6 Ib/MMBtu) for NO .
 Regulatory Option V regulations for these boilers are 860 ng/J (2.0 Ib/MMBtu)
 and 90-50% reduction for S09; 43 ng/J (0.1 Ib/MMBtu) for PM; and 258 ng/J
 (0.6 Ib/MMBtu) for NO .     *
                                    9-41

-------
applicable to the beet sugar industry.  Boiler and pollution control capital
costs for Regulatory Option I amount to $20.9 million, compared to the base
case cost of $20.7 million.  Total annualized boiler cost for the base case
is $4.10 per GJ ($4.33 per MMBtu), compared to $4.19 per GJ ($4.42 per
MMBtu) for Regulatory Option I.  Total annualized cost of new steam for
Regulatory Option V is $4.92 per GJ ($5.19 per MMBtu), a 19 percent increase
over the base case.
     Using the total steam cost figures, the cost of steam per dollar of
output is calculated for the beet sugar industry.  Table 9-16 illustrates
the extent of product cost changes attributable to increased cost of new
steam.  Each kilogram (pound) of beet sugar produced requires approximately
19,500 kJ (8,378 Btu).  Assuming that the cost of beet sugar equals its
average selling price of $0.386 per kilogram ($0.175 per pound) and that
new steam accounts for 95 percent of total steam use, the cost of new steam
per dollar output ranges from $0.201 for Regulatory Option I to $0.2360 for
Regulatory Option V.  Resulting product cost changes range from 0.40 percent
in Regulatory Option I to 3.90 percent in Regulatory Option V.
     Table 9-17 presents the change in profitability levels as a result of
increased cost of new steam.  The analysis of the base case and the regula-
tory options assumes that sales are constant in real terms and that expenses
increase only as a function of the new boiler investment.  The incremental
expense is assumed to be absorbed by the firm and is not passed on to the
consumer.  The resultant net income figures for the regulatory options
range from $135,000 for Regulatory Option I to a net loss of $978,000 for
Regulatory Option V, compared to a net income in the base case of $208,000.
The adverse effect of the incremental costs associated with the new boiler
investment is due in part to the low profitability level (0.66 percent of
sales for the base case) existing in the industry.  Return on assets for
the regulatory options ranges from a positive 0.56 percent in Option I to a
negative 4.0 percent in Option V.  The negative return on assets suggests
that the industry would choose not to replace the boiler, a decision which
could result in a plant closure.  The plant closure possibility for the
beet sugar industry is the theoretical worst case that could develop.
However, from a practical standpoint, closure may not occur due to a number
of reasons.
                                 9-42

-------
     First, the model firm for the beet sugar industry consists of four
plants.   The other three plants are not considering replacement and may be
able to carry the loss burden resulting from the "impact" plant's replace-
ment decision.   Assuming a net income of $1.644 million for the three
plants,  a net loss of $978,000 from the impact plant would mean a decrease
in the previous amount to $666,000.   This is equivalent to a $222,000
profit per non-replacing plant, which translates to 0.7 percent return on
sales from a previous level  of 1.74 percent.
     Second, based on a strict assumption that each plant is an independent
cost center, and that other  profitable plants cannot subsidize this loss,
the conditions  theoretically would warrant plant closure.   However, viewing
the firm as a profit entity  composed of different profit-generating segments
means that losses in certain segments may not necessarily dictate closure.
Other factors must be considered,  including how long the loss situation
will be maintained, whether  the magnitude of the loss offsets the other
segments'  earnings, how large the  amount of sunk costs involved may be, and
whether opportunity costs of not being able to supply buyers due to closure
and/or loss of customers would affect the profitable segments'  earnings.
     Third, a net loss due to boiler replacement is attributable to in-
creased expenses, i.e., variable costs.   However, other significant fixed
costs cannot be discounted.   The cost savings that will be realized from
the plant shutdown may be more than offset by the capital  costs embedded in
the plant.  Plant closure may not  be viable due to the existence of these
fixed costs.  Closure may occur only when the loss due to replacement is
greater than the loss associated with the firm's inability to recover the
fixed costs already in the existing plant.
     Table 9-18 shows the effect of the regulatory options on coverage and
debt/equity ratios in the beet sugar industry.  Although there is little
variation in coverage ratios as a  function of regulatory options, there is
a significant decrease when  a higher debt level is assumed for the boiler
investment.  The biggest change occurs in Regulatory Option V, where cover-
age decreases from 4.0 in the zero percent debt level to 2.6 in the 100
percent debt level.
     Debt/equity ratios increase significantly as a function of financing
strategy.   In Regulatory Option V, debt/equity increases from 0.32 to 0.68,
                                9-43

-------
          TABLE 9-16.  CHANGE IN PRODUCT COST:   BEET SUGAR MODEL PLANT
                            Base Case
                                                       Regulatory Option
GJ steam.per kg (MMBtu/lb)   0.0195
  output0                   (0.0084)

Percent of new steam per       95
  kg (lb) product
 0.0195
(0.0084)

   95
 0.0195
(0.0084)

   95
Cost of new steam            4.10
  per GJ (MMBtu), 1978 $    (4.33)

Cost of new steam            0.076
  per kg (lb) output,       (0.034)
  1978 $

Average product cost         0.386
  per kg (Ib), 1978 $       (0.175)

Cost of new steam            0.197
  per $ output, 1978 $
 4.19
(4.42)

 0.0776
(0.0352)
 0.386
(0.175)

 0.201
 4.92
(5.19)

 0.0743
(0.0413)
 0.386
(0.175)

 0.2360
Percent increase (decrease)
  in steam cost per $ output

Percent increase (decrease)
  in product cost
 2.04
 0.40
19.80
 3.90
 Base case regulations are the applicable SIP's for all pollutants.

 Regulatory Option I regulations for these boilers are SIP limits for
 S09; 43 ng/J (0.1 Ib/MMBtu) for PM; and 258 ng/J (0.6 Ib/MMBtu) for NO .
c
 Regulatory Option V regulations for these boilers are 860 ng/J (2.0 Ib/MMBtu)
 and 90-50% reduction for S0?; 43 ng/J (0.1 Ib/MMBtu) for PM; and 258 ng/J
 (0.6 Ib/MMBtu) for NO .
H
 Estimated from industry contacts.

eBased on model plant configuration.

fSteam costs are 1990 pre-tax estimates.
                                    9-44

-------
      TABLE 9-17.  CHANGE IN PROFIT MARGIN DUE TO NEW BOILER INVESTMENT:
                            BEET SUGAR MODEL PLANT

Regulatory Option

Sales
Expenses
Gross profit
Taxes
Net income
Return on
assets, %
Base Case3
% of
106 $ sales
31.500 100.00
31.084 98.68
0.416 1.32
0.208 0.66
0.208 0.66
, 0.86
Ib
% of
106 $ sales
31.500 100.00
31.230 99.14
0.270 0.86
0.135 0.43
0.135 0.43
0.56
VC
% of
106 $ sales
31.500 100.00
32.478 103.10
(0.978) (3.10)
0.0 0.0
(0.978) (3.10)
(4.00)
aBase case regulations are the applicable SIP's for all  pollutants.

 Regulatory Option I regulations for these boilers are SIP limits for
 S02; 43 ng/J (0.1 Ib/MMBtu)  for PM; and 258 ng/J (0.6 Ib/MMBtu) for NO .

cRegulatory Option V regulations for these boilers are 860 ng/J (2.0 Ib/MMBtu)
 and 90-50% reduction for S09; 43 ng/J (0.1 Ib/MMBtu) for PM;  and 258 ng/J
 (0.6 Ib/MMBtu) for NO .     *
                                   9-45

-------
           TABLE 9-18.  CAPITAL AVAILABILITY:   BEET SUGAR MODEL FIRM
                                                        Regulatory Option
                                      3             h                  C
Financial indicator          Base Case             I                  V

Coverage ratio
  Percent financed
    by debt
         0                     4.00               4.00               4.00
        25                     3.60               3.60               3.55
        50                     3.27               3.26               3.19
        75                     2.99               2.98               2.89
       100                     2.76               2.75               2.64
Debt/equity ratio
  Percent financed
    by debt
0
25
50
75
100
0.34
0.40
0.47
0.55
0.65
0.33
0.40
0.47
0.56
0.65
0.32
0.40
0.48
0.57
0.68
aBase case regulations are the applicable SIP's for all pollutants.
 Regulatory Option I regulations for these boilers are SIP limits for
 SO,; 43 ng/J (0.1 Ib/MMBtu) for PM; and 258 ng/J (0.6 Ib/MMBtu) for NO .
r
 Regulatory Option V regulations for these boilers are 860 ng/J (2.0 Ib/MMBtu)
 and 90-50% reduction for S0?; 43 ng/J (0.1 Ib/MMBtu) for PM; and 258 ng/J
 (0.6 Ib/MMBtu) for NO
                      /x
                                    9-46

-------
compared to the base case increase of 0.34 to 0.65, for zero percent to 100
percent debt financing, respectively.
     With regard to financing capability,  the analysis of coverage ratios
indicates that new boiler investment can be funded with debt ranging from
50 to 75 percent of total investment cost without going beneath the 3.0
standard benchmark.   In terms of total  capitalization, debt/equity ratios
fall within the acceptable 1.0 benchmark.   Assuming a 50 percent debt
financing option (given that the 3.0 coverage benchmark is to be main-
tained), the debt ratio does not vary to a significant degree.   This indi-
cates a sufficient equity base to absorb additional financing of new boiler
investment without undue weakening of the  industry's financial  position.
     The profitability of the beet sugar industry in the base case is
determined using industry sales data for 1977, a year in which the industry
performed below average in terms of profitability.   The financial  impacts
thus represent a worst case analysis in terms of the industry's ability to
absorb the cost of replacing its steam plant.   During other years  in the
mid-1970's, the beet sugar industry averaged a 3 to 4 percent return on
sales, which is twice as high as depicted for the model firm.   Thus, the
typical firm actually may be able to make  the new boiler investment without
reducing its profit margin to near zero.
     In addition, faced with increasing steam costs due to rising energy
prices, the beet sugar firm likely would initiate energy conservation
measures to reduce the level of steam use  and overall production costs.
This would result in an improved profit margin for the plant and more
favorable conditions for investing in the  new boilers.
     The proposed standard is not expected to be the primary criterion in
the decision to install a new steam plant.   The firm will need to evaluate
the investment as being cost-effective with or without more stringent
emission regulations.  Given the low rate of return in these firms, the
major issue is whether any capital investment is justified, even with the
intent of reducing energy costs by installing a coal-fired boiler.  Under
ARO I (essentially the proposed standards), the industry still would save
energy costs by installing the coal-fired boiler, although savings would be
less than in the base case.
                                9-47

-------
     9.2.3.4  Fruit and Vegetable Canning Industry.
     9.2.3.4.1  Model firm and plant description.   Table 9-19 depicts the
firm and plant configuration of a typical fruit and vegetable canning
operation.  This analysis assumes that the model plant cans selected fruits
and vegetables in proportions similar to the national average.   This plant,
located in Federal Region 9, is one of six canning plants that the typical
firm operates.
     The typical firm produces 21 million cases of processed fruits and
vegetables each year and commands about 4 percent of the domestic canned
goods market.  The producer selling price per case is approximately $7.04
(1978 dollars), generating sales per firm of $147.8 million and earnings of
$5.6 million.
     The typical plant houses three fossil fuel-fired boilers that have a
combined heat input capacity of 69 MW (236 MMBtu/hr).  Each boiler operates
at 25 percent of annual heat input capacity.  The typical plant would
replace the two larger boilers, listed on Table 9-19, with two natural
gas-fired boilers of the same size.
     9.2.3.4.2  Financial analysis.  The financial indicators of the fruit
and vegetable canning industry are depicted in Table 9-20.   Historically,
the industry's profits earned have been low but stable.  Net profit margins
(i.e., return on sales) have remained at a steady 3.6 percent, even during
the 1974-1975 recession.  Net profits ranged from $46.2 million to $70.1
million, with a 5-year average of $58.0 million.  Return on assets averaged
8.3 percent over the 1977-1978 period.
     The total amount of long-term capitalization and each of its com-
ponents has increased over the years studied.  Between 1977 and 1978,
stockholders' equity increased by a greater percentage than did long-term
debt:  $465.0 million to $513.1 million (a 10 percent increase) compared
with $152.7 million to $160.8 million (a 5 percent increase).  The debt/
equity ratio averaged 0.32 for 1977 and 1978.  The coverage ratios of 10.99
for 1977 and 11.70 for 1978 indicate that, on average, operating income
(earnings before interest and taxes) is more than adequate to support debt
obligations.
     9.2.3.4.3  Regulatory option results.  The replacement boilers are
expected to burn natural gas because of local environmental regulations.
                                 9-48

-------
 Both of the regulatory options, but not the base case, have identical NOX
 regulations that require combustion modification.
      The difference in costs between the base case and regulatory options
 is due only to the cost of NO  combustion modification.  As Table 9-21
                              /\
 shows, the capital cost of the two new boilers varies from $2,335,000 for
 the base case to $2,428,000 for all others.  The before-tax annualized
 components of total new steam cost (capital, O&M, and fuel) also are pre-
 sented in the table.  The total annualized steam cost is $7.19 per GJ
' ($7.59 per MMBtu) in the base case and $7.36 per GJ ($7.77 per MMBtu) in
 the regulatory options.  The fuel cost of $5.12 per GJ ($5.40 per MMBtu) is
 a major component of the total cost in both options.
      Table 9-22 lists the variables needed to calculate the cost of new
 steam per dollar product.   Approximately 27,100 kJ (25,700 Btu) are needed
 to produce one case of canned goods.   An average weighted price of $7.04
 per case of output is assumed to be the 1978 producer price of canned
 fruits and vegetables.. The cost of new steam represents approximately 2.05
 percent of product cost in the base case and 2.10 percent in the regulatory
 cases.  These calculations are based on the assumption that new steam
 accounts for 74 percent of total steam demand and that costs are fully
 absorbed by the producer.
      Table 9-23 presents the change in profit margin due to a new boiler
 subject to a given regulatory option.   Assuming a 50 percent corporate
 income tax and constant sales, net income is $260,000 in the base case.  In
 Options I and V, net income declines $10,000 to $250,000.
      The coverage ratio and the debt/equity ratio for the fruit and vege-
 table canning industry are shown in Table 9-24.  These ratios do not vary
 between the base case and regulatory options due to the small capital cost
 difference between them.  Both ratios appear to be in a healthy range,
 suggesting that the firm will be able to finance a new boiler investment.
      The results of the analysis indicate that the regulatory options bring
 about little percentage change in product cost.  New steam cost per dollar
 output is approximately 2.10 percent for both options.  Profitability is
 affected only slightly by the environmental expenses of NO  combustion
                                                           ^
 modification required by the regulatory options.
                                  9-49

-------
            TABLE 9-19.  MODEL FIRM AND PLANT CONFIGURATION:
                 FRUIT AND VEGETABLE CANNING INDUSTRY
Model firm
      Financial dataa
        Average bond rating:       N.A.
        Coverage ratio:            11.7
        Debt/equity ratio:          0.31
Model plant
      Production data
       Plant output/year:          3.5 million cases
       Price/unit output:          $7.04 per casec>
       Plant sales/year:           $24.64 million01' e
       Plant earnings/year:        $0.94 million  '
      Boiler configuration
Total firing rate: 69.2 MW (236 MMBtu/hr)
No. of boilers: 3
Federal region: 9
Characteristics of individual boilers
Capacity, MW
(MMBtu/hr)
Fuel type
Annual capacity utilization, percent
Replacement, expansion or
existing


Boiler
1 2
25.5 25.5
(87) (87)
natural natural
gas gas
25 25
— replacement —



3
18.2
(62)
natural
gas
25
existing
a!978 values.
bN.A. = Not  available.
cPrice/case  of  output  is  a  weighted  average  (1974-1977) price of total
 canned fruits  and  vegetables  produced,  inflated to 1978 dollars.
Expressed in 1978  $.
ePlant sales/year is derived by  multiplying  plant output/year by
 price/case  of  output.
 Based upon  the 1978 return on sa^egsratio of 3.8 percent.

                                  9-50

-------
     TABLE  9-20.  FINANCIAL ANALYSIS:  FRUIT AND VEGETABLE CANNING INDUSTRY3
Financial
indicator
Capital Expenditures
Total assets (106 $)
Capital expenditures/
firm (106 $)
Capital expenditures/
total assets (%)
Profitability
Net profit after
taxes (106 $)
Return on assets (%)
Return on equity (%)
Return on sales (%)
Trends in dividends
($ per share)
Net earnings before
interest & taxes
(106 $)
Capitalization
Interest on fixed
obligations
(106 $)
Coverage ratio
Rating on bonds
Long-term debt (106 $)
Stockholders' equity
(106 $)
Debt/capi tal i zati on
Debt/equity ratio

1974

N.A.b
40.8
N.A.
46.2
N.A.
N.A.
3.4
1.09

N.A.

16.96
N.A.
N.A.
120.9
N.A.
N.A.
N.A.

1975

N.A.
58.4
N.A.
54.7
N.A.
N.A.
3.6
1.15

N.A.

19.23
N.A.
N.A.
139.4
N.A.
N.A.
N.A.
Year
1976

N.A.
51.4
N.A.
57.0
N.A.
N.A.
3.6
1.25

N.A.

18.45
N.A.
N.A.
141.1
N.A.
N.A.
N.A.

1977

749.6
46.6
6.2
61.6
8.2
13.2
3.6
1.34

190.2

17.31
10.99
N.A.
152.7
465.0
24.7
0.3284

1978

834.8
65.9
7.9
70.1
8.4
13.7
3.8
1.46

210.4

17.98
11.70
N.A.
160.8
513.1
23.9
0.3134
Average ,
1974-1978

792.2
52.6
7.1
57.95
8.3
13.5
3.6
1.26

200.3

17.99
11.35
N.A.
143.0
489.1
24.3
0.3205
Average per firm estimates (Securities and Exchange Commission; EEA estimates)
Nominal Terms.

N.A.  = Not available.


                                    9-51

-------
          TABLE 9-21.  1990 BOILER COSTS:   FRUIT AND VEGETABLE
                          CANNING MODEL PLANT

Costs
Regulatory Option
Base Case3 Ib Vc
Total boiler and
  pollution control
  capital costs, 1978$

Annualized total
  boiler costs, $/GJ
     ($/MMBtu)

  Capital
  O&M


  Fuel

  Total
2,334,760
   0.66
  (0.70)

   1.42
  (1.50)

   5.12
  (5.40)

   7.19
  (7.59)
2,427,760
   0.69
  (0.73)

   1.56
  (1.65)

   5.12
  (5.40)

   7.36
  (7.77)
2,427,760
   0.69
  (0.73)

   1.56
  (1.65)

   5.12
  (5.40)

   7.36
  (7.77)
Control Technology
     NO
                     Combustion
                     Modification
                  Combustion
                  Modification
 The base case contains no regulations for these boilers.

 Regulatory Option  I contains an 86 ng/J (0.2 Ib/MMBtu) N0x regulation
 for these boilers.

cRegulatory Option  I contains an 86 ng/J (0.2 Ib/MMBtu) NO  regulation
 for these boilers.                                       x
                                 9-52

-------
         TABLE 9-22.  CHANGE IN PRODUCT COST:  FRUIT AND VEGETABLE
                            CANNING MODEL PLANT


GJ (MMBtu) steam .
per case output
Percent of new steam
per case
Cost of new steam
per GJ (MMBtu), 1978 $
Cost of new steam
Base Case3
0.0271
(0.0257)
74
7.19
(7.59)
0.1443
Regulatory
Ib
0.0271
(0.0257)
74
7.36
(7.77)
0.1478
Option
VC
0.0271
(0.0257)
74
7.36
(7.77)
0.1478
  per case, 1978 $

Average product cost per
  case, 1978 $

Cost of new steam
  per $ output, 1978 $
7.04


0.0205
7.04


0.0210
7.04


0.0210
Percent increase (decrease)
  in steam cost per $ output

Percent increase (decrease)
  in product cost
                    2.44


                    0.05
                  2.44


                  0.05
 The base case contains no regulations for these boilers.

 Regulatory Option I contains an 86 ng/J (0.2 Ib/MMBtu) NO  regulation
 for these boilers.                                        x

 Regulatory Option I contains an 86 ng/J (0.2 Ib/MMBtu) NO  regulation
 for these boilers.                                        x

 Estimated from industry contacts.

eBased on model plant configuration.

 Steam costs are 1990 pre-tax estimates.
                                9-53

-------
        TABLE  9-23.   CHANGE IN PROFIT MARGIN DUE TO NEW BOILER INVESTMENT:
                    FRUIT AND VEGETABLE CANNING MODEL PLANT

Regulatory Option

Sales
Expenses
Gross profit
Taxes
Net income
Return on
assets, %
Base
106$
24.64
24.12
0.52
0.26
0.26

Case3
% of
sales
100.00
97.89
2.11
1.06
1.06
2.32
Ib
% of
106$ sales
24.64 100.00
24.13 97.93
0.51 2.07
0.25 1.01
0.26 1.06
2.32
vc
& of
106$ sales
24.64 100.00
24.13 97.93
0.51 2.07
0.25 1.01
0.26 1.06
2.32
aThe base case contains no regulations for these boilers.

bRegulatory Option I contains an 86 ng/J (0.2 Ib/MMBtu) N0x regulation
 for these boilers.

cRegulatory Option I contains an 86 ng/J (0.2 Ib/MMBtu) N0x regulation
 for these boilers.
                                9-54

-------
                  TABLE  9-24.  CAPITAL AVAILABILITY:
                FRUIT AND VEGETABLE CANNING MODEL FIRM

Regulatory Option
Financial indicator
Coverage ratio
Percent financed by debt
0
25
50
75
100
Debt/equity ratio
Percent financed by debt
0
25
50
75
100
Base Case3

11.70
11.66
11.62
11.58
11.55

0.31
0.31
0.32
0.32
0.32
Ib

11.70
11.66
11.62
11.58
11.55

0.31
0.31
0.32
0.32
0.32
VC

11.70
11.66
11.62
11.58
11.55

0.31
0.31
0.32
0.32
0.32
aThe base case contains no regulations for these boilers.

 Regulatory Option I contains an 86 ng/J (0.2 Ib/MMBtu)
 NO  regulation for these boilers.
c  x
 Regulatory Option I contains an 86 ng/J (0.2 Ib/MMBtu)
 NO  regulation for these boilers.
   /\
                                9-55

-------
     9.2.3.5  Rubber Reclaiming Industry.
     9.2.3.5.1  Model firm and plant description.  Table 9-25 presents the
model plant and firm for the rubber reclaiming industry.  Each plant within
the industry is assumed to be identical  in its production process; each
produces the same amount of output with  equal amounts of steam.  The typical
plant operates in the midwestern United  States and has a yearly output of
18,100 megagrams (20,000 tons).  The typical producer selling price is
$0.37 per kilogram ($0.17 per pound), yielding sales of $6.7 million.
Applying an industry-wide profit margin  of 5.2 percent of sales, the model
plant earns $350,000 in profit.
     The typical plant's boiler house contains three boilers that have a
combined firing capacity of 62 MW (211 MMBtu/hr).  The replacement boiler
is a coal-fired unit with a heat input capacity of 26 MW (87 MMBtu/hr).
All boilers are assumed to operate at 45 percent of annual rated capacity.
     9.2.3.5.2  Financial analysis.  The financial parameters for the
rubber reclaiming industry are listed in Table 9-26.  The financial para-
meters for the rubber reclaiming industry are derived from data from the
parts of the industry that are steam-intensive and whose reclaimed rubber
sales comprise a substantial portion of  total domestic corporate sales.  It
is assumed that the profit indicators for the entire corporation and the
rubber reclaiming plant are comparable for a given year.
     Although reclaimed rubber production has fluctuated from 1974 to 1978,
corporate-wide profit margins  (return on sales)  have remained around 5
percent.  Reclaimed rubber price increases have  helped offset decreased
sales, thus slightly increasing total sales revenue.  Return on total
assets has hovered around 8.8  percent for 1977 and 1978.  Net profits have
increased from $6.1 million per firm to  $10.2 million from 1975 to 1978.
     Uncertainty of product demand and resulting fluctuations  in plant
production have constrained new investment in production facilities.  For
this reason, investments in replacing capacity rather than extending plant
capacity typify the industry.  Although  capitalization data exist only for
1977 to 1978, these financial  indicators appear  to be stable.  Stockholders'
equity per firm ranged from $86 million  to $94 million.  Long-term debt was
between $23 million and $24 million for  the 2-year period.  The resulting

                                 9-56

-------
debt/equity ratio hovered around the 2-year average of 0.26.   This low
long-term debt proportion suggests that the industry has unused debt capa-
city.  Assuming that the firm will finance externally, the majority of the
funds may come from debt instruments.
     9.2.3.5.3  Regulatory option results.   The new boiler under both
regulatory options is projected to burn coal.   IFCAM projects that coal  is
the least-cost fuel largely because of the high relative prices of residual
oil and natural gas in this region.   In the base case, in order to meet
local PM emission regulations, and also under ARO I, the plant will install
an electrostatic precipatitor on the boiler.
     Table 9-27 presents the pre-tax 1990 boiler and pollution control
costs for the regulatory options for the typical boiler investment in the
rubber reclaiming industry.   Boiler and pollution control capital  costs
vary from $5,760,000 in the base case to $6,242,000 in Regulatory Option V.
Annual!zed, these capital costs vary from $1.81 per GJ ($1.91 per MMBtu)
for the base case to $2,03 per GJ ($2.14 per MMBtu) for Option V.   Regula-
tory Option I requires an annualized capital  cost of approximately $1.84
per GJ ($1.95 per MMBtu).
     Table 9-27 also shows annual O&M and fuel  costs which,  combined with
the capital costs, yield a total cost of new steam.   Annual  O&M costs vary
from $2.05 per GJ ($2.16 per MMBtu) for the base case to $2.87 per GJ
($3.03 per MMBtu) in Option V.  The annual  O&M cost for Option I is $2.11
per GJ ($2.33 per MMBtu).  Fuel costs are lowest in Option V because under
this option the lowest cost compliance strategy is to scrub a higher sulfur
coal.  The total cost of new steam is highest in Option V at $6.94 per GJ
($7.32 per MMBtu), and the lowest for the base case at $6.28 per GJ ($6.63
per MMBtu), with Option I at $6.39 per GJ ($6.74 per MMBtu).
     Using these total steam cost figures,  the cost of new steam per dollar
of output can be calculated for the rubber reclaiming industry.  Table 9-28
depicts the calculation and presents results for the regulatory options.
Each kilogram (pound) of reclaimed rubber produced requires approximately
10,500 kJ (4,513 Btu).  Assuming that the cost of reclaimed rubber equals
its real average selling price of $0.37 per kilogram ($0.17 per pound) and
that new steam accounts for one-third of total steam use, the cost of new
steam for the model plant ranges from 5.8 percent of the product cost in
                                9-57

-------
           TABLE 9-25.   MODEL FIRM AND PLANT CONFIGURATION:
                      RUBBER RECLAIMING INDUSTRY
Model firm
       Financial data8
       Average bond rating:
       Coverage ratio:
       Debt/equity ratio:
Model plant
       Production data
       Plant output/year:
       Price/unit output:
       Plant sales/year:
       Plant earnings/year:
N.A.
9.9
0.24
18,100 megagrams (20,000 tons)c
$0.37/kilogram ($0.17/pound)d
$6.7 million
$0.35 milliond>e
       Boiler configuration
       Total firing rate:
       No. of boilers:
       Federal region:
       Characteristics of  individual boilers
61.8 MW (211 MMBtu/hr)
3
5
                                                        Boiler
       Capacity, MW
               (MMBtu/hr)
       Fuel type
       Annual capacity utilization,
         percent
       Replacement, expansion or
         existing
             25.5    18.2     18.2
             (87)    (62)     (62)
             coal   oil/gas  oil/gas
              45      45       45

          replacement —existing	
 1978 values.
bN.A. = Not available.
cAverage of plant output/year for 1974 - 1978.
Expressed in 1978 $.
eBased upon the 1978 return on sales ratio of 5.2 percent.
                                 9-58

-------
         TABLE 9-26.  FINANCIAL ANALYSIS:  RUBBER RECLAIMING INDUSTRY*
	
Financial
indicator
Capital expenditures
Total assets (106$)
Capital expenditures/
firm (106$)
Capital expenditures/
total assets (%)
Profitability
Net profit after
taxes (106$)
Return on assets (%)
Return on equity (%)
Return on sales (%)
Trends in dividends
($ per share)
Net earnings before
interest and taxes
(106$)
Capitalization
Interest on fixed
obligations (106$)
Coverage ratio
Rating on bonds
Long-term debt (106$)
Stockholders' equity
(106$)
Debt/capitalization
Debt/equity ratio

1974

N.A.b
N.A.
N.A.
6.2
N.A.
N.A.
4.8
0.75
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.

1975

N.A.
N.A.
N.A.
6.1
N.A.
N.A.
4.9
0.79
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
Year
1976

N.A.
N.A.
N.A.
7.3
N.A.
N.A.
4.9
0.92
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.

1977

108.1
11.5
10.6
9.6
8.9
11.2
5.4
1.00
23.9
2.3
10.4
N.A.
23.6
85.5
21.6
0.2760

1978

116.6
7.5
6.4
10.2
8.7
10.9
5.2
1.14
26.7
2.7
9.9
N.A.
23.0
93.9
19.6
0.2444
Average
1974-1978

112.4
9.5
8.4
7.9
8.8
11.0
5.1
0.92
25.3
2.5
10.1
N.A.
23.3
89.7
20.6
0.2595
Average per firm estimates (Securities and Exchange Commission; EEA estimates).
Nominal terms.

N.A.  = Not available.
                                     9-59

-------
         TABLE 9-27.  1990 BOILER COSTS:  RUBBER RECLAIMING MODEL PLANT
   Costs
Base Case*
                                                       Regulatory Option
Total boiler and
  pollution control
  capital costs (1978 $)

Annualized total
  boiler costs, $/GJ
               ($/MMBtu)

  Capital
  O&M


  Fuel


  Total


Coal type, ng SO?/J
          (Ib SO^/MMBtu)

Control Technology

       SO,,
       PM
5,760,000
  1.81
 (1-91)

  2.05
 (2.16)

  2.43
 (2.56)

  6.28
 (6.63)

744.00
 (1.73)
Electrostatic
Precipitator
5,820,000
   1.84
  (1.95)

   2.11
  (2.23)

   2.43
  (2.56)

   6.39
  (6.74)

 744.00
  (1.73)
Electrostatic
Precipitator
6,241,610
    2.03
   (2.14)

    2.87
   (3.03)

    2.04
   (2.15)

    6.94
   (7.32)

  997.00
   (2.32)
                                       Double Alkali/
                                       Mechanical
                                       Collector
 Base case regulations are the applicable SIP's for all pollutants.

 Regulatory Option I regulations  for these boilers are SIP limits for
 SO,; 43 ng/J (0.1 Ib/MMBtu) for  PM; and 258 ng/J (0.6 Ib/MMBtu) for NO
r
 Regulatory Option V regulations  for these boilers are 860 ng/J (2.0 Ib/MMBtu)
 and 90-50% reduction for S0?; 43 ng/J  (0.1 Ib/MMBtu) for PM; and 258 ng/J
 (0.6 Ib/MMBtu) for NOV.
                                     9-60

-------
      TABLE 9-28.  CHANGE IN PRODUCT COST:  RUBBER RECLAIMING MODEL PLANT



GJ



(MMBtu) steam per kg
(lb) output

Base Case3
0.010
(0.005)


0.
(0.
Regulatory
Ib
010
005)
Option
VC
0.010
(0.005)
Percent of new steam per
   kg (lb)e
   33
   33
   33
Cost of new steam perf
   GJ (MMBtu), 1978 $T

Cost of new steam per
   kg (lb), 1978 $

Average product cost
   per kg (Ib), 1978 $

Cost of new steam per
   $ output, 1978 $
 6.28
(6.63)

 0.0218
(0.0099)

 0.37
(0.17)

 0.058
 6.39
(6.74)

 0.022
(0.010)

 0.37
(0.17)

 0.059
 6.94
(7.32)

 0.024
(0.011)

 0.37
(0.17)

 0.064
Percent increase (decrease)
   in steam cost per $ output

Percent increase (decrease)
   in product cost
                      1.72
                      0.10
                    10.34
                     0.60
 Base case regulations are the applicable SIP's for all pollutants.

 Regulatory Option I regulations for these boilers are SIP limits for
 S02; 43 ng/J (0.1 Ib/MMBtu) for PM; and 258 ng/J (0.6 Ib/MMBtu) for NO .

 Regulatory Option V regulations for these boilers are 860 ng/J (2.0 Ib/MMBtu)
 and 90-50% reduction for SO,; 43 ng/J (0.1 Ib/MMBtu) for PM; and 258 ng/J
 (0.6 Ib/MMBtu) for NO .     *
d                     x
 Estimated from industry contacts.
6
 Based on model plant configurations.

 Steam costs are 1990 pre-tax estimates.
                                    9-61

-------
the base case to 6.4 percent in Option V.  This calculation assumes that
this cost increase is absorbed fully by the producer.
     Table 9-29 presents the change in profit margin due to a new boiler
investment.  This table assumes that sales are constant in real terms and
that expenses rise only due to the new boiler investment.   Assuming a 50
percent corporate income tax, net income varies from $160,000 in the base
case to 40,000 in Option V.  Return on assets for the model plant varies
from 4.09 percent for the base case to 3.58 percent in Option I and 1.02
percent in Option V.
     Table 9-30 lists the rubber reclaiming industry's coverage ratio and
debt/equity ratio for five financing options for the regulatory options.
The coverage and debt/equity ratios do not vary significantly across regula-
tory options.  These ratios do vary, however, depending on financing stra-
tegy.  In Option V, for example, the coverage ratio decreases from 9.89 to
8.04 in the 100 percent debt financing option.  While this represents a 19
percent decrease, the average ratio is still above the 3.0 standard bench-
mark.  The debt/equity ratio for Option V varies from 0.23 to 0.31 for the
five financing options, which is well below the 1.0 threshold level.  The
low long-term debt proportion suggests that the industry may have unused
debt capacity.  Assuming that the firm will finance externally, the majority
of the external funds may come from debt instruments.
     The results of the analysis indicate that the regulatory options cause
percentage increases in product cost of 1 percent or less.  Profits are
positive in Option I and in Option V.  Return on assets is 3.58 percent in
Option I and 1.02 percent in Option V.  Capitalization data suggest that
the firm will be able to finance a new boiler investment under both regula-
tory options examined.
     9.2.3.6  Automobile Manufacturing Industry.
     9.2.3.6.1  Model firm and plant description.  The model firm and plant
configuration for the automobile manufacturing industry is presented in
Table 9-31.  The plant that operates in Federal Region 5 is assumed to be
part of a 26-plant firm.  Total firm production is 2,343,380 vehicles, with

                                 9-62

-------
annual car and light truck sales of $14.64 billion,  assuming an average
price (1978 dollars) of $6,249 per vehicle.   These production statistics do
not include foreign-made cars (such as the Dodge Colt or the Ford Fiesta)
normally part of United States automobile companies'  fleets.   Because these
cars are not produced in this country, their production costs would not be
affected by alternative regulatory options.   Net profit for the model firm
is assumed to be 4.28 percent on total sales of $626.7 million.
     The model plant boiler house consists of four coal-fired boilers with
a total heat input capacity of 102 MW (348 MMBtu/hr).   The boiler invest-
ment decision is to replace one of these units with  a similarly sized new
coal-fired boiler.
     9.2.3.6.2  Financial  analysis.   The financial  indicators of the auto-
mobile manufacturing industry in 1974 and 1975, as shown in Table 9-32,
reflect the apprehension of consumers to purchase new automobiles after the
oil embargo of 1973-1974.   Return on assets  was approximately 3 percent,
less than one-half of the  5-year average for 1974 to 1978.   Return on sales
reached a similar low of 2.09 percent compared to the 5-year average of
3.94 percent.  By 1976, however, the industry had recovered.   In fact,
return on assets and return on sales surpassed the industry's 5-year average
during the following 3 years.   Note that net profit  generally increased
between 1974 to 1978, from $316 million to $1.2 billion.
     Capital availability  does not seem to pose problems for the typical
automotive manufacturing plant.   Long-term debt has  remained relatively
constant over the past 5 years, usually between $850 million and $950
million.  Stockholders' equity, on the other hand, has increased from
approximately $5.5 billion to $7.6 billion.   This increase has caused the
debt/equity ratio to fall  from 0.16 to 0.11.  Because these ratios show a
low percentage of debt, future investments could be  funded largely from
debt, depending upon the interest rate and the industry's inclination
toward debt financing.
     The coverage ratio for the automobile manufacturing industry has been
rising over the past 5 years from 10.52 to 20.17.  The average 5-year
coverage ratio of 16.68 is considered to be sufficiently high that the
automobile manufacturing industry should not have difficulty obtaining  debt
financing.
                                9-63

-------
    TABLE 9-29.   CHANGE IN PROFIT MARGIN DUE TO NEW BOILER INVESTMENT:
                       RUBBER RECLAIMING MODEL PLANT

Regulatory Option

Sales
Expenses
Gross profit
Taxes
Net income
Return on
assets, %
Base
106$
6.70
6.39
0.31
0.15
0.16

Case3
% of
sales
100.00
95.37
4.63
2.39
2.39
4.09

106$
6.70
6.43
0.27
0.13
0.14

Ib
% of
sales
100.00
95.97
4.03
2.01
2.01
3.58
VC
% of
106$ sales
6.70 100.00
6.63 98.55
0.07 1.04
0.03 0.53
0.04 0.53
1.02
aBase case regulations are the applicable SIP's for all pollutants.

 Regulatory Option I regulations for these boilers are SIP limits for
 SO,; 43 ng/J (0.1 Ib/MMBtu) for PM; and 258 ng/J (0.6 Ib/MMBtu) for NO .
r
 Regulatory Option V regulations for these boilers are 860 ng/J (2.0 Ib/MMBtu)
 and 90-50% reduction for S09; 43 ng/J (0.1 Ib/MMBtu) for PM; and 258 ng/J
 (0.6 Ib/MMBtu) for NO .     *
                                 9-64

-------
    TABLE 9-30.   CAPITAL AVAILABILITY:   RUBBER RECLAIMING MODEL FIRM
                                                   Regulatory Option
Financial  Indicator         Base Case3          I                 Vc

Coverage ratio
  Percent financed
    by debt
       0                     9.89              9.89              9.89
      25                     9.40              9.37              9.34
      50                     8.93              8.93              8.87
      75                     8.53              8.50              8.42
     100                     8.14              8.14              8.04
Debt/equity ratio
Percent financed
by debt
0
25
50
75
100

0.23
0.25
0.27
0.29
0.31

0.23
0.25
0.27
0.29
0.31

0.23
0.25
0.27
0.29
0.31
 Base case regulations  are  the  applicable  SIP's  for all  pollutants.
 Regulatory Option I  regulations  for  these boilers  are  SIP  limits  for
 S02;  43 ng/J (0.1 Ib/MMBtu)  for  PM;  and  258 ng/J  (0.6  Ib/MMBtu)  for NO .
 Regulatory Option V regulations  for  these  boilers are  860 ng/J  (2.0 Ib/MMBtu)
 and 90-50% reduction for S09;  43 ng/J  (0.1 Ib/MMBtu) for PM;  and 258 ng/J
 (0.6  Ib/MMBtu)  for NO .    *
                                 9-65

-------
           TABLE 9-31.  MODEL FIRM AND PLANT CONFIGURATION:
                   AUTOMOBILE MANUFACTURING INDUSTRY
Model firm
       Financial dataa
       Average bond rating:
       Coverage ratio:
       Debt/equity ratio:
Model plant
       Production data
       Assembly plant output/year:
       Price/unit output:
       Assembly plant sales/year:
       Assembly plant earning/year:
       Boiler configuration
Aaa/B
20.17
 0.11
90,130 automobiles
$6,249.00/automobilec
$563.22 million0
$24.11 millionc'd
Total firing rate: 102.0 MW (348 MMBtu/hr)
No. of boilers: 4
Federal region: 5
Characteristics of individual boilers
Capacity, MW
(MMBtu/hr)
Fuel type
Annual capacity utilization,

le
25.5
(87)
coal
0
Boil
2
25.5
(87)
coal
25
er
3
25.5
(87)
coal
25

4
25.5
(87)
coal
25
         percent
       Replacement, expansion  or
         existing
         •existing-
replacement
a!978 values.
 Based upon average  industry estimates.
Expressed in 1978 $.
 Based upon the 1978 return on sales  ratio of 4.28 percent.
eBoiler number one is used as a standby boiler.
                                 9-66

-------
          TABLE 9-32.   FINANCIAL ANALYSIS:   AUTOMOBILE MANUFACTURING INDUSTRY*
Financial
indicator
Capital expenditures
Total assets (106$)
Capital expenditures/

1974

10,411
1,142

1975

10,741
916
Year
1976

12,069
957

1977

13,631
1,545

1978

15,169
1,955
Average
1974-1978

12,404.2
1,303
  firm (106$)

Capital  expenditures/
  total  assets (%)
10.97
8.53
7.93
11.33
12.89
10.50
Profitability
Net profit after
taxes (106$)
Return on assets (%)
Return on equity (%)
Return on sales (%)
Trends in dividends
($ per share)
Net earnings before
interest and taxes
(106$)
Capitalization
Interest on fixed
obligations (106$)
Coverage ratio
Rating on bonds
Long-term debt (106$)
Stockholders' equity
(106$)
Debt/capitalization (%)
Debt/equity ratio

316

3.03
5.79
2.09
1.89

1,452 1



138

10.52
N.A.b
853
5,460 5

13.52
0.1563

335 1

3.12
6.04
2.06
1.12

,730 3



183

9.45
N.A.
985
,556 6

15.06
0.1772

,042 1

8.63
16.93
4.89
2.02

,039 3



155

19.61
N.A.
904
,155 6

12.80
0.1468

,285 1

9.43
18.72
5.05
2.69

,557 3



149

23.87
N.A.
939
,867 7

12.02
0.1367

,232

8.12
16.14
4.28
2.59

,732 2



185

20.17
Aaa/B
850
,635 6,

10.01
0.1113

842

6.79
13.30
3.94
2.06

,702



162

16.68
	
906.2
334.6

12.51
0.1430
 Average per firm estimates (Securities and Exchange Commission; EEA estimates).
 Nominal terms.

5N.A. = Not Available.
                                        9-67

-------
     9.2.3.6.3  Regulatory option results.  In the base case, the plant's
boiler replacement must comply with SIP air emission limits.  Based on
historical industry trends, the fuel for the new boiler is projected to be
coal.  IFCAM projects coal as the least-cost fuel available given the
applicable SIPs.  To meet the SIP specifications in the base case, the
plant will install an electrostatic precipatator for PM control.  Option I
results in use of a compliance coal and a electrostatic precipatator.  Flue
gas desulfurization (FGD) and a single mechanical collector (MC) are used
for S02 and PM control, respectively, in Option V.
     Table 9-33 presents the pre-tax 1990 boiler and pollution control
costs for the regulatory options for the automobile industry.  Option V has
significantly higher costs due to the mandatory FGD requirement.  The
capital cost in Option V is $5.86 million as compared to $5.85 million in
the base case.  The before-tax annualized components of total new steam
cost also are presented in this table.  The capital and O&M components vary
significantly between Options I and V.  Annualized capital costs are $3.36
per GJ ($3.54 per MMBtu) and O&M costs are $3.64 per GJ ($3.84 per MMBtu)
in the base case.  In Option V, the annualized capital costs are $3.43 per
GJ ($3.62 per MMBtu) and O&M costs are $4.82 per GJ ($5.09 per MMBtu).
Fuel costs are lowest in the base case and Option I, since higher sulfur
coals arp used.
     Combining the above costs yields a total cost of new steam, which
ranges from $9.03 per GJ ($9.53 per MMBtu) in the base case to $10.68 per
GJ ($11.26 per MMBtu) in Option V.
     The calculation of the cost of new steam per dollar output is depicted
in Table 9-34.  The calculation of profits assumes sales to be constant in
real terms and expenses to rise only due to the new boiler investment.  The
additional costs are assumed to be fully absorbed by the firm.  Typically,
the industry consumes 1.735 GJ (1.645 MMBtu) for each new automobile pro-
duced.  Assuming that the real price of a new car is $6,249 and that one-
third of the steam consumed is new steam, the cost of new steam per car
represents considerably less than one-tenth of 1 percent in both of the
regulatory options.
     Table 9-35 presents the change in profit margin calculations for the
automotive industry.  Because the boiler investment is such a small fraction
of total  expenses, the net income changes due to a regulatory option when

                                 9-68

-------
compared to the base case are small.   Option V results in a less than 1
percent decline in net income compared to the base case.   Return on assets
falls only slightly from 8.10 percent in the base case to 8.04 percent in
Option V.
     The coverage and debt/equity ratios for the automobile manufacturing
industry are shown in Table 9-36.   The coverage ratio declines slightly
from 20.17 to 20.11 over the five financing options and shows little diffe-
rence between regulatory options.   The debt/equity ratio remains at around
0.11.  Neither of these ratios suggests problems in obtaining capital in
either of the regulatory options.   Since these rates show a low percentage
of debt, future investments could be  funded largely from debt, depending
upon the interest rate and the industry's inclination toward debt financing.
     The results of the analysis indicate that the regulatory options do
not significantly affect any of the above financial parameters.   The impact
on product cost is negligible due to  the low ratio of new steam cost to
total dollar output.   Net income as a percent of sales is approximately 4.0
percent in all cases, with returns on assets of 8.0 percent for both regula-
tory options examined.   Capital availability is not constrained by any of
these cases, suggesting that the firm will  be able to finance a new boiler
replacement.
     9.2.3.7  Petroleum Refining Industry.
     9.2.3.7.1  Model  firm and plant description.   Table 9-37 presents the
model firm and plant for the petroleum refining industry.   This plant,
operating in the Southwest (Federal  Region 6), is assumed to be part of a
seven-plant firm.   Total production  for the model  firm is 112.93 million
barrels of refined product per year.  Assuming a real price of $33.65 per
barrel and an annual net profit margin of 4.51 percent, the firm realizes
annual sales of $3.8 billion and annual profits of $171.4 million.   Compar-
ing these figures to 1978 U.S. refined product consumption, this firm
satisfies approximately 2 percent of total demand and accounts for about 4
percent of the domestically refined  petroleum products market.
                                 9-69

-------
     TABLE 9-33.   1990 BOILER COSTS:  AUTOMOBILE MANUFACTURING MODEL PLANT
    Costs
                             Base Case
                                                       Regulatory Option
Total boiler and
  pollution control
  capital costs, 1978 $

Annualized total
  boiler costs, $/GJ
               ($/MMBtu)

       Capital
       O&M
       Fuel
       Total
Coal type, ng 50,,/J
          (Ib SO^/

Control Technology
          (Ib SO/MMBtu)
                             5,850,000
  3.36
 (3.54)

  3.64
 (3.84)

  2.04
 (2.15)

  9.03
 (9.53)

997.00
 (2.32)
                    5,920,000
  3.39
 (3.58)

  3.75
 (3.96)

  2.04
 (2.15)

  9.18
 (9.69)

997.00
 (2.32)
                   5,863,460
  3.43
 (3.62)

  4.82
 (5.09)

  2.43
 (2.56)

 10.68
(11.26)

744.00
 (1.73)
                                                                  Double Alkali/
                                                                  Mechanical
                                                                  Collector
       PM
                             Electrostatic
                             Precipitator
                    Electrostatic
                    Precipitator
                Double Alkali/
                Mechanical
                Collector
aBase case regulations are the applicable SIP's for all pollutants.

 Regulatory Option I regulations for these boilers are SIP limits for
 SO,; 43 ng/J (0.1 Ib/MMBtu) for PM; and 258 ng/J (0.6 Ib/MMBtu) for NO .
r
 Regulatory Option V regulations for these boilers are 860 ng/J (2.0 Ib/MMBtu)
 and 90-50% reduction for S0?; 43 ng/J (0.1 Ib/MMBtu) for PM; and 258 ng/J
 (0.6 Ib/MMBtu) for NO
                                   9-70

-------
  TABLE 9-34.   CHANGE IN PRODUCT COST:   AUTOMOBILE MANUFACTURING MODEL PLANT


GJ (MMBtu) steam .
per unit output
Base Case3
1.74
(1.65)
Regulatory
Ib
1.74
(1.65)
Option
vc
1.74
(1.65)
Percent of new steam
  per unit6
33.3
33.3
33.3
Cost of new steamfper GJ       9.03
  (MMBtu), 1978 $T            (9.53)

Cost of new steam per          5.22
  unit output, 1978 $

Average product cost per       6,249.
  unit output, 1978 $

Cost of new steam per        0.00084
  $ output, 1978 $
                   9.18
                  (9.69)

                   5.30


                   6,249.


                  0.00085
                   10.68
                  (11.26)

                    6.17


                    6,249.


                   0.00099
Percent increase (decrease)
  in steam cost per $ output

Percent increase (decrease)
  in product cost
                    1.19
                   0.001
                     17.51
                     0.015
 Base case regulations are the applicable SIP's for all pollutants.

 Regulatory Option I regulations for these boilers are SIP limits for
 S02; 43 ng/J (0.1 Ib/MMBtu) for PM; and 258 ng/J (0.6 Ib/MMBtu) for NO .
•>  *•                                                                   .X
'Regulatory Option V regulations for these boilers are 860 ng/J (2.0 Ib/MMBtu)
 and 90-50% reduction for S07; 43 ng/J (0.1 Ib/MMBtu) for PM; and 258 ng/J
 (0.6 Ib/MMBtu) for NO .     *
1                     X
 Estimated from industry contacts.
>
'Based on model plant configurations.

 Steam costs are 1990 pre-tax estimates.
                                   9-71

-------
      TABLE 9-35.  CHANGE IN PROFIT MARGIN DUE TO NEW BOILER INVESTMENT:
                     AUTOMOBILE MANUFACTURING MODEL PLANT

Regulatory Option
Base Case3

Sales
Expenses
Gross profit
Taxes
Net income
Return on
assets, %
106 $
563.22
515.14
48.08
24.04
24.04
8.
% of
sales
100.00
91.48
8.52
4.26
4.26
10
Ib
106 $
563.22
515.17
48.05
24.02
24.03
8.

% of
sales
100.00
91.47
8.53
4.27
4.27
09
VC
106 $
563.22
515.47
47.75
23.87
23.88
8.

% of
sales
100.00
91.52
8.48
4.24
4.24
04
 Base case regulations are the applicable SIP's for all pollutants.
 Regulatory Option I regulations for these boilers are SIP limits for
 S09; 43 ng/J (0.1 Ib/MMBtu) for PM; and 258 ng/J (0.6 Ib/MMBtu) for NO .
c
 Regulatory Option V regulations for these boilers are 860 ng/J (2.0 Ib/MMBtu)
 and 90-50% reduction for S09; 43 ng/J (0.1 Ib/MMBtu) for PM; and 258 ng/J
 (0.6 Ib/MMBtu) for NO .     *
                                    9-72

-------
 TABLE 9-36.   CAPITAL AVAILABILITY:   AUTOMOBILE MANUFACTURING MODEL FIRM

Financial indicator Base Case
Requlatory
ib
Option
VC
Coverage ratio
  Percent financed
    by debt
       0
      25
      50
      75
     100
20.17
20.16
20.14
20.13
20.11
20.17
20.16
20.14
20.13
20.11
20.17
20.16
20.14
20.13
20.11
Debt/equity ratio
Percent financed
by debt
0
25
50
75
100

0.11
0.11
Q. 11
0.11
0.11

0.11
0.11
0.11
0.11
0.11

0.11
0.11
0.11
0.11
0.11
 Base case regulations  are  the  applicable  SIP's  for  all  pollutants.
Regulatory Option I  regulations  for  these boilers are  SIP  limits  for
 S02; 43 ng/J  (0.1 Ib/MMBtu)  for  PM;  and 258  ng/J (0.6  Ib/MMBtu) for NO .
^                                                                      /\
'Regulatory Option V  regulations  for  these boilers are  860  ng/J  (2.0 Ib/MMBtu)
 and 90-50% reduction for S09;  43 ng/J  (0.1 Ib/MMBtu) for PM;  and  258 ng/J
 (0.6 Ib/MMBtu)  for NO  .    *
                                9-73

-------
           TABLE 9-37.   MODEL FIRM AND PLANT CONFIGURATION:
                      PETROLEUM REFINING INDUSTRY
Model firm
       Financial dataa
       Average bond rating:
       Coverage ratio:
       Debt/equity ratio:
Model plant
       Production data
       Plant output/year:
       Price/unit output:
       Plant sales/year:
       Plant earnings/year:
Aaa/A
14.12
0.32
16,133,000 barrels
$33.65 per barre1b'c
$543.88 million0
$24.48 million
              c,d
       Boiler configuration
       Total firing rate:
       No. of boilers:
       Federal region:
       Characteristics of  individual  boilers
381.0 MW(1300 MMBtu/hr)
4
6
                                                         Boiler
       Capacity, MW
                (MMBtu/hr)
       Fuel type
       Annual capacity  utilization,
         percent
       Replacement,  expansion,  or
         exi sti ng
      95.2
     (325)
  refinery refinery  natural petroleum coke
     gas
      75
     gas
      75

•exi sting-
gas
 75
& residual oil
   75

 replacement
a!978 values.
 The price per unit output  is  based  upon  the average price for
 all refined products.
Expressed in 1978 $.
dBased upon the 1978 return on sales  ratio of 4.51 percent.
                                9-74

-------
     The model plant boiler house consists of four boilers, with a total
heat input capacity of 381 MW (1,300 MMBtu/hr).   Each boiler has a firing
capacity of 95 MW (325 MMBtu/hr) and is used at 75 percent of heat input
capacity.   Three of the boilers are existing units, firing refinery and
natural gas.   The fourth unit, firing a mixture of petroleum coke and
residual oil, will be replaced by a new coal-firing boiler in 1990.   Ap-
proximately 25 percent of boiler steam generation for this plant will be
provided by the new coal-firing boiler.
     9.2.3.7.2  Financial analysis.   The petroleum refining industry his-
torically has been able to recoup increased crude import costs and refinery
costs through higher product prices charged to retail establishments.
Since 1973, retail gasoline and fuel  oil prices  have kept pace with the
nominal cost increases in crude imports.  Federal price regulations create
ceiling prices at the pump but have interfered minimally with the trend of
higher retail fuel prices to maintain profit levels.   The demand for refi-
nery products has, in general, surpassed the supply capabilities of domestic
refineries, causing price levels to rise in response to a tight market.
     If the petroleum refining industry does face increased steam costs, it
is not likely to adjust its production process to reduce the level of steam
use in its production equipment.  Steam is an integral input to most of the
individual process elements in a refinery; thus, increased steam costs
cannot be mitigated by process changes.   Because steam is such an important
input, new or replacement boiler investments are not likely to be cancelled
due to an increase in steam costs.   Product demand is healthy and suffi-
ciently inelastic to cover these additional costs without having the refi-
nery experience decreased profits.
     Table 9-38 delineates the financial indicators of the petroleum re-
fining industry.  Because petroleum-derived products exhibited strong sales
during 1974-1978, the profitability indicators for the industry have been
high and stable.  Profits were extremely healthy, especially in 1974,
during the oil embargo.   Since 1975,  net profits ranged from $658 million
to $868 million per year and averaged $775 million per firm between 1974
and 1978.   Return on total assets was between 5.8 and 10.1 percent, with
the 5-year average at 6.93 percent.
     Capital  availability appears to be stable for the petroleum refining
industry.   Although the coverage ratio has fallen in recent years, as of

                                 9-75

-------
           TABLE 9-38.  FINANCIAL ANALYSIS:   PETROLEUM REFINING INDUSTRY*
Financial
indicator
                                              Year
  1974
1975
1976
1977
1978
 Average
 1974-1978
Capital expenditures
Total assets (106$)
Capital expenditures/
  firm (106$)
Capital expenditures/
  total assets (%)
Profitability
Net profit after
  taxes (106$)
Return on assets (%)
Return on equity (%)
Return on sales (%)
Trends in dividends
  ($ per share)
Net earnings before
  interest and taxes
  (106$)
Capitalization
Interest on fixed
obligations (106$)
Coverage ratio
Rating on bonds
Long term debt (106$)
Stockholders' equity
  (106$)
Debt/capi tal i zati on
Debt/equity ratio
8,434.20  9,321.50  10,606.00 13,266.60 14,320.10  11,189.68
1,153.10  1,302.00   1,371.10  1,271.30  1,400.20   1,299.54
   13.67
  855.0
 13.97
657.8
  12.90
   9.58
 733.80    762.50
   9.78
           867.70
  108.70    130.90

   30.53     20.04
   Aaa/A     N.A.
1,276.30  1,578.60
4,732.40  5,094.30
           156.70
           188.10    221.80
            16.38     15.46     14.12
            N.A.      N.A.      N.A.
         1,979.20  2,131.80  2,049.70
         5,552.00  5,970.00  6,449.00
   21.24
 23.66
  26.28
  26.31
  24.12
   11.61
            775.36
10.14
18.07
N.A.b
1.81
7.06
12.91
N.A.
1.88
6.92
13.22
N.A.
2.02
5.75
12.77
N.A.
2.22
6.06
13.45
4.51
2.38
6.93
13.95
N.A.
2.06
3,319.00  2,623.50   2,566.80  2,907.80  3,130.80   2,909.44
    0.2697    0.3099
             0.3565    0.3571    0.3178
  161.24

   18.04

1,803.12
5,559.54

   24.49

    0.3243
aAverage per firm estimates  (Securities and Exchange Commission; EEA estimates).
 Nominal terms.
bN.A. = Not available.
                                       9-76

-------
1978 it was still, at 14.12, extremely high.   This ratio is sufficiently
high to assume that the refinery will not have problems in obtaining ex-
ternal funds for a boiler investment in the base case.   Long-term debt has
nearly doubled since 1974, from $1.3 billion to over $2 billion.   This
increase is counteracted partially by an increase in stockholders'  equity
from $4.7 billion to $6.4 billion.   Consequently, the debt/equity ratio has
increased only from 0.27 to 0.32.
     9.2.3.7.3  Regulatory option results.   Historically,  a significant
share of the boiler fossil fuel demand in the petroleum refining  industry
has been met with the use of liquid, solid, and gaseous waste by-products
of refinery operations.   Because the focus of this analysis is on the
choice between coal, oil, and gas,  the fuel type for the new boiler is
limited to these fuels.   IFCAM projects that the new boiler under both
regulatory options will  burn coal.   The option cases require scrubbing and
PM control.
     Table 9-39 presents pre-tax 1990 boiler and pollution control  costs
under the regulatory options for the model  petroleum refining plant.   Costs
are equal for Options I  and V, as the cases have the same  compliance strate-
gies.  The base case capital cost is $18.3 million, while  the two options
have capital costs of $19.3 million.
     The annualized capital cost for the regulatory options is $1.00 per GJ
($1.05 per MMBtu), while the annualized base case capital  cost is $0.95 per
GJ ($1.00 per MMBtu).  O&M costs vary from $1.43 per GJ ($1.51 per MMBtu)
for the base case to approximately $1.72 per GJ ($1.81 per MMBtu) in Options
I and V.   Fuel costs range from $2.47 per GJ ($2.61 per MMBtu) in the base
case to $2.27 per GJ ($2.39 per MMBtu) for Options I and V.
     Combining the above components yields the total cost of new steam.
Total cost is $4.84 per GJ ($5.11 per MMBtu) in the base case and $4.98 per
GJ ($5.25 per MMBtu) in the two regulatory options.
     The cost of new steam per dollar of output is shown in Table 9-40.
Assuming that a barrel of refined output requires 196,440 kJ (186,200 Btu)
of steam and that the cost per barrel of oil  is $33.65, then the cost of
new steam represents 0.7 percent of product costs in the regulatory options.
This calculation assumes that new steam accounts for 24 percent of the
total steam requirements and that the firm fully absorbs this cost increase.
                                 9-77

-------
        TABLE 9-39.  1990 BOILER COSTS:  PETROLEUM REFINING MODEL PLANT
    Costs
 Base Case
    Regulatory Option
     Ib            Vc
Total boiler and
  pollution control
  capital costs, 1978 $

Annualized total
  boiler costs, $/GJ
               ($/MMBtu)

      Capital
      O&M
      Fuel
      Total
Coal type, ng SO«/J
          (Ib SO^/MMBtu)

Control Technology


      S02
      PM
18,335,450
    0.95
   (1.00)

    1.43
   (1.51)

    2.47
   (2.61)

    4.84
   (5.11)

  374.03
   (0.87)
19,335,990
    1.00
   (1.05)

    1.72
   (1.81)

    2.27
   (2.39)

    4.98
   (5.25)

 2575.23
   (5.99)
19,335,990
    1.00
   (1.05)

    1.72
   (1.81)

    2.27
   (2.39)

    4.98
   (5.25)

 2575.23
   (5.99)
               Double Alkali  Double Alkali

                   ESP            ESP
aBase case regulations for these boilers are 516 ng/J (1.2 Ib/MMBtu) for
 SO,,; 43 ng/J  (0.1 Ib/MMBtu  for PM; and 301 ng/J (0.7 Ib/MMBtu) for NO
   L.                                                                  ^

 Regulatory Option I regulations for these boilers are 430 ng/J (1.0 Ib/
 MMBtu) and 90-50% reduction for S0?; 22 ng/J  (0.05 Ib/MMBtu) for PM;
 and 301 ng/J  (0.7 Ib/MMBtu) for NOf.
                                   /\

GRegulatory Option V regulations for these boilers are 430 ng/J (1.0 Ib/
 MMBtu) and 90-50% reduction for SO,,; 22 ng/J  (0.05 Ib/MMBtu) for PM;
 and 301 ng/J  (0.7 Ib/MMBtu) for NO;.
                                  9-78

-------
       TABLE 9-40.  CHANGE IN PRODUCT COST:   PETROLEUM REFINING MODEL PLANT


GJ (MMBtu) steam.
per bbl output
Base Case3
0.19644
(0.18620)
Regulatory Option
Ib VC
0.19644 0.19644
(0.18620) (0.18620)
Percent new steam
  per bble
24
24
24
Cost of new steam       f    4.84
  per GJ (MMBtu), 1978 $T   (5.11)

Cost of new steam            0.2284
  per bbl,  1978 $

Average product cost        33.65
  per bbl,  1978 $

Cost of new steam            0.0068
  per $ output, 1978 $
                     4.97
                    (5.25)

                     0.2346


                    33.65


                     0.0070
                4.97
               (5.25)

                0.2346


               33.65


                0.0070
Percent increase (decrease)
  in steam cost per $ output

Percent increase (decrease)
  in product cost
                     2.94
                     0.0184
                2.94
                0.0184
 Base case regulations for these boilers are 516 ng/J (1.2 Ib/MMBtu) for
 S02; 43 ng/J (0.1 Ib/MMBtu for PM;  and 301 ng/J (0.7 Ib/MMBtu) for NO

 Regulatory Option I regulations for these boilers are 430 ng/J (1.0 Ib/
 MMBtu) and 90-50% reduction for S09;  22 ng/J (0.05 Ib/MMBtu) for PM-
 and 301 ng/J (0.7 Ib/MMBtu) for NO  .
                                  ^

 Regulatory Option V regulations for these boilers are 430 ng/J (1.0 Ib/
 MMBtu) and 90-50% reduction for S09;  22 ng/J (0.05 Ib/MMBtu) for PM:
 and 301 ng/J (0.7 Ib/MMBtu) for NO  .
                                  f\

 Estimated from industry contacts.

 Based on model plant configuration.

 Steam costs are 1990 pre-tax estimates.
                                 9-79

-------
     Table 9-41 presents the change in profit margin due to a new boiler
investment.  This calculation assumes that sales are constant in real terms
and that expenses rise only due to the new boiler investment.  Options I
and V reduce base case net income by less than one percentage point.  This
small percentage change in product cost is due to the small fraction that
new steam cost comprises of average product cost.
     The coverage and debt/equity ratios for the model petroleum refinery
for the regulatory options are presented in Table  9-42. As the table
illustrates, neither the regulatory option nor the financing strategy
affect these ratios significantly.  The coverage ratio decreases approxi-
mately from 14.12 to 14.00, or less than 1 percent.  The debt/equity ratio
remains around 0.32 under all financing and control levels.
     The results of the analysis indicate that neither of the regulatory
options result in significant cost impacts on the petroleum refining indus-
try.  New steam costs for the regulatory options comprise a relatively
small fraction of average product costs.  Profitability is affected slightly
by the incremental expenses due to new boiler investment.  Return on assets,
however, remains at 5.9 percent for both regulatory options.
     Capital availability appears to be stable for the petroleum refining
industry.  The coverage ratio is sufficiently high to assume that the
refinery will  not have problems obtaining external funds for a boiler
investment.
     9.2.3.8  Iron and Steel Manufacturing Industry.
     9.2.3.8.1  Model firm and plant description.  Table  9-43 depicts the
model firm and plant for the integrated iron and steel industry.  This
plant is assumed to be part of a five-plant firm, located in the midwestern
States.  Total production for the model firm is 8.2 million megagrams (9.0
million tons) of raw steel per year.  Assuming a real price of $384 per
megagram ($348 per ton) and an annual net profit margin of 2.9 percent, the
firm realizes annual sales of $3.1 billion and annual profits of $90.9
million.
     The model plant boiler house consists of four boilers with a total
heat input capacity of 216 MW (736 MMBtu/hr).   Three of the boilers have a

                                 9-80

-------
capacity of 40 MW (137 MMBtu/hr) and the fourth has a capacity of 95 MW
(325 MMBtu/hr).   All the boilers currently fire blast furnace gas and have
an annual capacity utilization of 55 percent.   The three 40 MW boilers will
be replaced by three similarly sized coal-fired boilers.   Approximately 55
percent of boiler steam generation for this plant will  be provided by the
new coal-fired boilers.
     9.2.3.8.2  Financial analysis.   The major financial  indicators of the
average firm in the iron and steel industry are shown in Table 9-44.   A
record level for profits was set in 1974, followed by declining profits in
1975 and 1976, and a net loss in 1977.   Profits increased in 1978,  but did
not approach the previous level  of 1974.
     Net profit declined from $259.6 million in 1974 to $143.2 million in
1976, followed by a net loss of $7.8 million in 1977.   However, 1978 wit-
nessed a return to pre-1977 profit levels,  with a net profit of $155.3
million.  Long-term debt has increased steadily from $666.8 million in 1975
to $1.03 billion in 1978.  Debt levels for 1977 to 1978 have remained at
slightly over one-third of total capitalization.
     9.2.3.8.3  Regulatory option results.   The three replacement boilers
for the iron and steel manufacturing industry are assumed to burn coal.   In
order to meet local emission regulations in the base case and to meet ARO
I, a fabric filter is installed for PM control.  A compliance coal  for S02
is chosen in the base case and Option I, while scrubbing is required in
Option V with the selection of a FGD/single mechanical  collector for S02
and PM control.   Table 9-45 shows pre-tax 1990 boiler and pollution control
costs for the regulatory options applicable to the iron and steel industry.
Boiler and pollution control capital costs range from $20.3 million in the
base case to $22.6 million in Option V.
     Annualized capital  cost in the base case is $1.11 per GJ ($1.17 per
MMBtu) compared to $1.27 per GJ ($1.34 per MMBtu) in Option V.  O&M costs
of $2.05 per GJ ($2.16 per MMBtu) are also higher in Option V, while fuel
costs are lower.  Fuel costs are lower since a high sulfur coal, requiring
scrubbing, is burned.  Total cost of producing new steam is highest in
Option V at $5.36 per GJ ($5.66 per MMBtu), compared to $5.10 per GJ ($5.38
per MMBtu) in the base case.
     As shown in Table 9-46, each megagram (ton) of iron and steel produced
requires approximately 1.705 GJ (1.465 MMBtu).  Assuming that the cost of

                                  9-81

-------
        TABLE 9-41.  CHANGE IN PROFIT MARGIN DUE TO NEW BOILER INVESTMENT:
                             PETROLEUM REFINING MODEL PLANT

Regulatory Option

Sales
Expenses
Gross profit
Taxes
Net income
Return on
assets, %
Base
106 $
542.88
494.93
47.95
23.97
23.98

Case3
% of
sales
100.00
91.12
8.83
4.42
4.42
5.94
Ib
% of
106 $ sales
542.88 100.00
495.02 91.55
47.86 8.82
23.93 4.41
23.93 4.41
5.92
vc
% of
106 $ sales
542.88 100.00
495.02 91.55
47.86 8.82
23.93 4.41
23.93 4.41
5.92
aBase case regulations for these boilers are 516 ng/J (1.2 Ib/MMBtu) for
 S02; 43 ng/J (0.1 Ib/MMBtu for PM; and 301 ng/J (0.7 Ib/MMBtu) for N0x>

 Regulatory Option I regulations for these boilers are 430 ng/J (1.0 Ib/
 MMBtu) and 90-50% reduction for SO,,; 22 ng/J (0.05 Ib/MMBtu) for PM;
 and 301 ng/J (0.7 Ib/MMBtu) for NCT.
                                   /\

GRegulatory Option V regulations for these boilers are 430 ng/J (1.0 Ib/
 MMBtu) and 90-50% reduction for S0?; 22 ng/J (0.05 Ib/MMBtu) for PM;
 and 301 ng/J (0.7 Ib/MMBtu) for NO^.
                                 9-82

-------
    TABLE  9-42.  CAPITAL AVAILABILITY:
                            MODEL FIRM
               PETROLEUM REFINING
Financial indicator
Base Case
                                          Regulatory Option
Coverage ratio
  Percent financed
    by debt
          0
         25
         50
         75
        100
  14.12
  14.09
  14.06
  14.03
  14.00
14.12
14.08
14.05
14.02
13.99
14.12
14.08
14.05
14.02
13.99
Debt/equity ratio
  Percent financed
    by debt
0
25
50
75
100
0.32
0.32
0.32
0.32
0.32
0.32
0.32
0.32
0.32
0.32
0.32
0.32
0.32
0.32
0.32
 Base case regulations for these boilers are 516 ng/J (1.2 lb/
 MMBtu) for S09i 43 ng/J (0.1 Ib/MMBtu for PM; and 301 ng/J
 (0.7 Ib/MMBtu) for NO .
)                     x
 Regulatory Option I regulations for these boilers are 430 ng/J
 (1.0 Ib/MMBtu) and 90-50% reduction for SO,,; 22 ng/J (0.05 lb/
 MMBtu) for PM; and 301 ng/J (0.7 lb/MMBtu)^for NO .
•>                                                 f\
'Regulatory Option V regulations for these boilers are 430 ng/J
 (1.0 Ib/MMBtu) and 90-50% reduction for SO,; 22 ng/J (0.05 lb/
 MMBtu) for PM; and 301 ng/J (0.7 lb/MMBtu)^for NO .
                                 9-83

-------
               TABLE  9-43.  MODEL FIRM AND PLANT CONFIGURATION:
                   IRON AND STEEL MANUFACTURING INDUSTRY
Model firm
     Financial data
     Average bond rating:
     Coverage ratio:
     Debt/equity ratio:
Model plant
     Production data
     Plant output/year:
     Price/unit output:
     Plant sales/year:
     Plant earnings/year:
N.A/
6.09
0.52
1,632,600 megagrams (1,800,000 tons)
$384 per megagram ($348 per ton)
$626.4 million0
$18.2 millionc'd
     Boiler configuration
     Total firing rate:
     No. of boilers
     Federal region:
     Characteristics of  individual boilers
215.6 MW (736 MMBtu/hr)
4
5
Capacity, MW
(MMBtu/hr)
Fuel type
Annual capacity utilization,

1
40.1
(137)
coal
55
Boil
2
40.1
(137)
coal
55
er
3
40.1
(137)
coal
55

4
95.2
(325)
blast
furnace
gas
55
       percent
     Replacement, expansion or
       existing
           —replacement	  existing
°1978 values.
bN.A. = Not available.
Expressed in 1978 $.
 Based on the 1978 return on  sales ratio of 2.9 percent.
                                9-84

-------
  TABLE  9-44.  FINANCIAL ANALYSIS:  IRON AND STEEL MANUFACTURING INDUSTRY3

Financial
indicator
Capital expenditures
Total assets (106 $)
Capital expenditures/

1974
N.A.b
234.9

1975
N.A.
373.2
Year
1976
N.A.
367.6

1977
N.A.
322.1

1978
4,370.1
275.4
Average
1974-1978
N.A.
314.7
  firm (106 $)
Capital expenditures/   N.A.     N.A.       N.A.      N.A.      N.A.        N.A.
  total assets (%)
Profitability
Net profit after taxes   259.6   166.1     143.2     (7.8)   155.3      143.3
  (106 $)
Return on assets (%)     N.A.     N.A.       N.A.      N.A.        3.55     N.A.
Return on equity (%)      15.2     9.2       7.5     (0.4)     7.9        7.9
Return on sales (%)    .    6.5     4.7       3.7     (0.2)     2.9        3.5
Trends in dividends        2.37    2.12      2.07     1.94     1.90       2.08
  ($ per share)
Net earnings before      626.3   393.4     360.8    173.8    489.7      408.4
  interest and taxes
  (106 $)
Capitalization
Interest on fixed         35.6    39.9      52.9     75.3     80.4       56.8
  obligations (106 $)
Coverage ratio            17.59    9.86      6.82     2.31     6.09       8.53
Rating on bonds          N.A.     N.A.       N.A.      N.A.      N.A.        N.A.
Long-term debt (106 $)  533.2     666.8     799.5    998.8  1,029.9      805.6
Stockholders' equity  1,703.9   1,801.6   1,903.9  1,847.0  1,967.7    1,844.8
  (106 $)
Debt/capitalization (%)  23.83    27.01     29.58    35.10    34.36      29.98
Debt/equity ratio         0.3129   0.3701     0.4199   0.5408   0.5234     0.4334
a
 Average per firm estimates (Securities and Exchange Commission;
 EEA estimates).   Nominal  terms.
 N.A.  = Not available.
                                 9-85

-------
 TABLE 9-45.   1990 BOILER COSTS:  IRON AND STEEL MANUFACTURING MODEL PLANT
   Costs
 Base Case
     Regulatory Option
     Ib            VC
Total boiler and
  pollution control
  capital costs, 1978 $

Annualized total
  boiler cost, $/GJ
              ($/MMBtu)

     Capital
     O&M


     Fuel


     Total


Coal type, ng S09/J
          (Ib SO^/MMBtu)

Control technology

     SO,,
20,260,000
20,490,000
22,610,790
    1.11
   (1.17)

    1.64
   (1.73)

    2.35
   (2.48)

    5.10
   (5.38)

  739.00
   (1.72)
    1.24
   (1.18)

    1.67
   (1.76)

    2.35
   (2.48)

    5.14
   (5.42)

  739.00
   (1.72)
    1.27
   (1.34)

    2.05
   (2.16)

    2.04
   (2.15)

    5.36
   (5.66)

  997.00
   (2.32)
                               Double Alkali/
                               Mechanical
                               Collector
     PM
 Fabric
 Filter
 Fabric
 Filter
 Base case  regulations  are  the  applicable SIP's  for all pollutants.

 Regulatory Option  I  regulations  for  these boilers are SIP  limits  for
 SO,,; 43 ng/J  (0.1  Ib/MMBtu)  for  PM;  and 258  ng/J (0.6 Ib/MMBtu) for NO  .
r
 Regulatory Option  V  regulations  for  these boilers are 860  ng/J (2.0 Ib/MMBtu)
 and 90-50% reduction for SO,,;  43 ng/J  (0.1 Ib/MMBtu) for PM; and  258 ng/J
 (0.6 Ib/MMBtu)  for N0x.
                                9-86

-------
          TABLE 9-46.
CHANGE IN PRODUCT COST:   IRON AND STEEL
 MANUFACTURING MODEL PLANT


GJ steam per .
Mg (MMBtu/ton) output0
Base Case9
1.705
(1.465)
Regulatory Option
Ib VC
1.705 1.705
(1.465) (1.465)
Percent of new steam
  per Mg (ton) output6
          55
55
55
Cost of new steam       f        5.10
  per GJ (MMBtu), 1978 $T       (5.38)

Cost of new steam                4.78
  per Mg (ton) output, 1978 $   (4.33)

Average product cost           384.00
  per Mg (ton), 1978 $        (348.00)

Cost of new steam                0.0124
  per $ output, 1978 $
                          5.14
                         (5.42)

                          4.82
                         (4.37)

                        384.00
                       (348.00)

                          0.0125
               5.36
              (5.66)

               5.03
              (4.56)

             384.00
            (348.00)

               0.0131
Percent increase (decrease)
  in steam cost per $ output

Percent increase (decrease)
  in product cost
                          0.81
                          0.01
               5.69
               0.07
 Base case regulations are the applicable SIP's for all  pollutants.
 Regulatory Option I regulations for these boilers are SIP limits for
 S02; 43 ng/J (0.1 Ib/MMBtu) for PM; and 258 ng/J (0.6 Ib/MMBtu) for NO .
cRegulatory Option V regulations for these boilers are 860 ng/J (2.0 Ib/
 MMBtu) and 90-50% reduction for SO •  43 ng/J (0.1 Ib/MMBtu) for PM; and
 258 ng/J (0.6 Ib/MMBtu) for NO .   *
H
 Estimated from industry contacts.

eBased on model plant configuration.

 Steam costs are 1990 pre-tax estimates.
                                 9-87

-------
iron and steel equals its real average selling price of $384 per megagram
($348 per ton) and that new steam accounts for 55 percent of total steam
use, cost of new steam per dollar output ranges from 0.0124 for the base
case to 0.0131 for Option V.  For Option V, product cost increases by
one-tenth of a percent.  The narrow variation and relatively low cost of
new steam per output for the regulatory options compared to average product
cost accounts for the small range in product cost changes.
     Table 9-47. indicates that there is no significant change in profit-
ability levels for the industry as a result of the changes in cost of new
steam.  The analysis for the base case and regulatory options assumes that
sales are held constant in real terms and that expenses increase only as a
result of new boiler investment.  This incremental expense is assumed to be
absorbed by the firm and is not passed on to the consumer.   Given this
assumption, net income is $17.24 million in Option I and $17.00 million in
Option V, compared to $17.28 million for the base case.  Return on assets
ranges from 3.36 percent in Option I to 3.32 percent in Option V.  Because
of the relatively large sales and expense base for the industry, the incre-
mental expense brought about by the increase in new steam cost of the
regulatory option does not significantly affect overall profitability.
     Table 9-48 shows that coverage and debt/equity ratios do not vary
significantly as a function of regulatory options.  However, there is a
slight change in coverage as a function of debt financing strategy.  The
coverage ratio decreases from approximately 6.09 with zero percent debt
financing to approximately 5.94 with 100 percent debt in both the base case
and Option I and to 5.92 in Option II.  The debt/equity ratios vary from
0.52 to 0.53 for the various financing strategies.
     The results of the analysis indicate that the regulatory options cause
low percentage increases in product cost.  New steam costs for the regula-
tory options comprise a relatively small portion of average product cost.
Profitability likewise is affected slightly by the incremental expenses due
to new boiler investment.  Although slight differences in net income exist
between the regulatory options and the base case, return on sales remains
at approximately 2.7 percent.
     With regard to financing capability, the analysis of coverage ratios
indicates that new boiler investment can be funded with up to 100 percent
                                 9-88

-------
debt.  The 3.0 coverage benchmark is always exceeded even when total debt
financing is assumed.   This firm's solvency position remains stable even
when total debt financing is undertaken.   Due to the industry's large
equity base, the debt ratios do not exhibit wide variances as a result of
the five financing options.
     9.2.3.9  Liquor Distilling Industry.
     9.2.3.9.1  Model  firm and plant description.   The model plant and
boiler configuration of the liquor distilling industry is shown in Table
9-49.  It is assumed that the typical firm operates three plants.   The
model plant is located in a southeastern  State and produces 17 million
liters (4.5 million gallons) of distilled liquor annually.
     The model plant operates two boilers, one rated at 25 MW (87  MMBtu/hr),
the other at 18 MW (62 MMBtu/hr), with a  total firing capacity of  44 MW
(149 MMBtu/hr) and 45 percent capacity utilization.   The model plant elects
to replace the two older natural gas/oil-fired boilers with identically
configured coal-fired boilers.
     9.2.3.9.2  Financial analysis.   The  domestic liquor distilling indus-
try appears to have performed moderately  well between 1976 and 1978, based
on the financial indicators shown in Table 9-50.   Although certain segments
of the industry have outperformed others, overall  profits have grown steadily.
     Between 1976 and 1978,  net profits before interest and taxes  almost
doubled over the same time period — from $10.83 million in 1976 to $18.36
million in 1978.  Relative profitability  indicators increased between 1976
and 1978 as well:  return on total assets grew from 2.1 percent to 3.9
percent, while return on sales improved from 2.4 percent to 4.3 percent.
     Between 1976 and 1978,  total assets  increased at an annual rate of
nearly 7 percent.  In 1976,  total assets  per firm averaged $136.87 million;
by the end of 1978, average assets reached $155.38 million.
     Long-term debt obligations fluctuated within the $21 to $27 million
range, comprising about one-fifth of total capitalization.  Coverage of
fixed obligations has continually improved from 1976 to 1978.  The debt/
equity ratio ranged from 0.24 to 0.31 during this period.
     9.2.3.9.3  Regulatory option results.  The model plant replacement
boilers for the liquor distilling industry are assumed to be coal-fired.
To meet SIP emission regulations in the base case, a single mechanical
collector is installed for PM control.  An electrostatic precipatator will

                                 9-89

-------
     TABLE 9-47.  CHANGE IN PROFIT MARGIN DUE TO NEW BOILER INVESTMENT:
                    IRON AND STEEL MANUFACTURING MODEL PLANT

Base Case3
% of
106$ sales
Regulatory Options
Ib Vc
% of % of
106$ sales 106$ sales
Sales                626.40   100.00    626.40   100.00    626.40   100.00

Expenses             591.84    94.48    591.92    94.50    592.40    94.57

Gross profit          34.56     5.52     34.48     5.50     34.00     5.43

Taxes                 17.28     2.76     17.24     2.75     17.00     2.71

Net income            17.28     2.76     17.24     2.75     17.00     2.71
Return on                   3.37               3.36               3.32
  assets, %

aBase case regulations are the applicable SIP's for all pollutants.

 Regulatory Option I regulations for these boilers are SIP limits for
 SO,; 43 ng/J (0.1 Ib/MMBtu) for PM; and 258 ng/J (0.6 Ib/MMBtu) for NO
c
 Regulatory Option V regulations for these boilers are 860 ng/J (2.0 Ib/
 MMBtu) and 90-50% reduction for SO,; 43 ng/J (0.1 Ib/MMBtu) for PM; and
 258 ng/J (0.6 Ib/MMBtu) for NO .  ^
                                 9-90

-------
           TABLE 9-48.  CAPITAL AVAILABILITY:   IRON AND STEEL
                        MANUFACTURING MODEL FIRM
                                                Regulatory Option
Financial indicator            Base Case3          I          Vc

Coverage ratio
  Percent financed
    by debt
          0                       6.09          6.09           6.09
         25                       6.05          6.05           6.05
         50                       6.02          6.01           6.01
         75                       5.98          5.98           5.96
        100                       5.94          5.94           5.92
Debt/equity ratio
Percent financed
by debt
0
25
50
75
100


0.52
0.52
0.53
0.53
0.53


0.52
0.52
0.53
0.53
0.53


0.52
0.52
0.53
0.53
0.53
 Base case regulations are the applicable SIP's for all pollutants.
 Regulatory Option I regulations for these boilers are SIP limits for
 S02; 43 ng/J (0.1 Ib/MMBtu) for PM; and 258 ng/J (0.6 Ib/MMBtu) for NO .
c
 Regulatory Option V regulations for these boilers are 860 ng/J (2.0 lb/
 MMBtu) and 90-50% reduction for S09; 43 ng/J (0.1 Ib/MMBtu) for PM; and
 258 ng/J (0.6 Ib/MMBtu) for NO .   *
                                9-91

-------
             TABLE 9-49.   MODEL FIRM AND PLANT CONFIGURATION:
                        LIQUOR DISTILLING INDUSTRY
Model firm
     Financial data3
     Average bond rating:
     Coverage ratio:
     Debt/equity ratio:
Model plant
     Production data
                         N.A.U
                         5.44
                         0.292
                              $28.22 million
Plant output/year:
Price/unit output:
Plant sales/year:
Plant earnings/year:
Boiler configuration
Total firing rate:
No. of boilers:
Federal region:
Characteristics of individual boilers
                              17.0 million liters (4.5 million gallons)
                              $1.66/liter ($6.27/gallon)c
                                            c
                                           c,d
                              $1.21 million
                              43.7 MW  (149 MMBtu/hr)
                              2
                              4
                                                            Boiler
     Capacity, MW
               (MMBtu/hr)
     Fuel type
     Annual  capacity  utilization,  percent
     Replacement,  expansion,
        or existing
1
25.5
(87)
coal
45
2
18.2
(62)
coal
45
                                                   -replacement-
d!978 values.
bN.A. =  Not Available.
Expressed in  1978  $.
 Based upon the  1978  return  on  sales  ratio of 4.3 percent.
                                 9-92

-------
         TABLE 9-50.   FINANCIAL ANALYSIS:   LIQUOR DISTILLING INDUSTRY*
Financial
indicator
Capital expenditures
Total assets (106 $)
Capital expenditures/firm
(106 $)a
Capital expenditures/
total assets (%)
Profitability
Net profit after taxes
(106 $)a
Return on total assets (%)
Return on equity (%)
Return on sales (%)
Trends in dividends
($ per share)
Net earnings before
interest and taxes
(106 $)
Capitalization
Interest on fixed
obligations (106 $)
Coverage ratio
Rating on bonds
Long-term debt (106 $)
Stockholders' equity (106 $)
Debt/capitalization (%)
Debt/equity ratio

1974

N.A.b
3.25
N.A.


N.A.
N.A.
N.A.
N.A.
N.A.

N.A.



N.A.

N.A.
N.A.
N.A.
N.A.
N.A.
N.A.

1975

N.A.
2.58
N.A.


N.A.
N.A.
N.A.
N.A.
N.A.

N.A.



N.A.

N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
Year
1976

136.87
2.32
1.70


2.8
2.1
3.3
2.4
N.A.

10.83



3.22

3.36
N.A.
26.42
85.29
0.236
0.310

1977

146.89
4.85
3.30


5.0
3.4
5.7
4.0
N.A.

14.35



3.06

4.69
N.A.
20.65
87.62
0.191
0.236

1978

155.38
N.A.
N.A.


6.1
3.9
6.7
4.3
N.A.

18.36



3.37

5.44
Ba
26.57
91.08
0.226
0.292
Average
1974-1978

146.4
3.25
2.50


4.6
3.1
5.2
3.6
N.A.

14.5



3.22

4.50
N.A.
24.55
88.0
0.218
0.279
a
  Average per firm estimates (Securities and Exchange Commission;
  EEA estimates).   Nominal terms.
  N.A.  = Not available.
                                    9-93

-------
be used in Option I.  An S02 compliance coal is chosen in the base case and
Option I, while scrubbing is required in Option V.
     Table 9-51 shows pre-tax 1990 boiler and pollution control  costs for
the regulatory option applicable to the liquor distilling industry.   Boiler
and pollution control capital costs for the typical boiler investment in
the industry range  from about $6.9 million in the base case to $9.0 million
in Option V.  Annualized capital cost in the base case is $1.26 per GJ
($1.33 per MMBtu) compared to $1.72 per GJ ($1.81 per MMBtu) in Option V.
     Replacements consist of two boilers, one rated at 25 MW (87 MMBtu/hr)
and the other at 18 MW (62 MMBtu/hr), each with a 45 percent annual  capa-
city utilization.   Annualized costs are weighted with the corresponding
boiler sizes to determine an average total cost of new steam.
     Total cost of  steam is highest in Option V at $6.85 per GJ ($7.23 per
MMBtu), compared to the base case  level of $5.38 per GJ ($5.68 per MMBtu).
The difference in total cost of new steam between these two levels,  however,
is primarily due to capital and O&M cost differentials.  Option V exhibits
an annualized cost  of $2.95 per GJ ($3.11 per MMBtu) compared to $1.95 per
GJ ($2.06 per MMBtu) in the base case.  Annualized fuel cost varies slightly
between $2.17 per GJ ($2.29 per MMBtu) in the base case and $2.19 per MMBtu
($2.31 per MMBtu) in Option V.
     On the basis of these total steam costs, the resultant cost of new
steam per dollar of output for the industry can be calculated, as shown in
Table 9-52.  The steam requirement per liter output is 0.00725 GJ (0.026
MMBtu per gallon).   Given an average cost of $1.66 per liter ($6.27 per
gallon) of output,  the increase in product cost for Option V represents a
0.64 percent increase over the base case level.  Option I exhibits an
increase in product cost of 0.24 percent.
     Table 9-53 illustrates the changes in profitability levels due to the
new boiler investment.  Sales are  assumed to be constant in both regulatory
options and expenses increase only as a result of the new boiler invest-
ment.  The incremental expense is  assumed to be absorbed by the firm and is
not passed on to the consumer.  After-tax returns on sales of 1.37 percent
in the base case declines to 0.74  percent in Option I, and becomes negative
0.46 percent in Option V.  Net income levels for the regulatory options
range from $210,000 in Option I to a net loss of $140,000 in Option V.
                                 9-94

-------
        TABLE 9-51.  1990 BOILER COSTS:  LIQUOR DISTILLING MODEL PLANT
     Costs
 Base Case
                                                   Regulatory Option
Total boiler and
  pollution control
  capital costs, 1978 $

Annualized total
  boiler cost, $/GJ
              ($/MMBtu)

     Capital
     O&M


     Fuel


     Total


Coal type, ng SO-/J
          (Ib SOg/

Control technology

     so2



     PM
  6,850,000
    1.26
   (1.33)

    1.95
   (2.06)

    2.17
   (2.29)

    5.38
   (5.68)

  743.00
   (1.73)
Single
Mechanical
Collector
8,500,000
  1.57
 (1.66)

  2.21
 (2.33)

  2.17
 (2.29)

  5.95
 (6.28)

743.00
 (1.73)
9,000,770
  1.72
 (1.81)

  2.95
 (3.11)

  2.19
 (2.31)

  6.85
 (7.23)

529.00
 (1-23)
                               Double Alkali/
                               Mechanical
                               Collector
Electrostatic
Precipitator
 Base case regulations are the applicable SIP's for all  pollutants.

 Regulatory Option I regulations for these boilers are SIP limits for
 S02; 43 ng/J (0.1 Ib/MMBtu) for PM; and 258 ng/J (0.6 Ib/MMBtu) for NO .

cRegulatory Option V regulations for these boilers are 860 ng/J (2.0 Ib/
 MMBtu) and 90-50% reduction for S00; 43 ng/J (0.1 Ib/MMBtu) for PM; and
 258 ng/J (0.6 Ib/MMBtu) for NO .   *
                               /\
                                9-95

-------
     TABLE 9-52. CHANGE IN PRODUCT COST:   LIQUOR DISTILLING MODEL PLANT
                              Base Case
                                                  Regulatory Option
GJ steam per liter      .         0.00725
  (MMBtu/gallon) output0        (0.02600)

Percent of new steam              100
  per liter (gallon) output6
 0.00725
(0.02600)

  100
 0.00725
(0.02600)

  100
Cost of new steam       f        5.38
  per GJ (MMBtu), 1978 $T       (5.68)

Cost of new steam per liter      0.0390
  (gallon) output, 1978 $       (0.1477)

Average product cost             1.66
  per liter (gallon), 1978 $    (6.27)

Cost of new steam                0.0236
  per $ output, 1978 $
 5.95
(6.28)

 0.0431
(0.1633)

 1.66
(6.27)

 0.0260
 6.85
(7.23)

 0.0497
(0.1880)

 1.66
(6.27)

 0.0300
Percent increase (decrease)
  in steam cost per $ output

Percent increase (decrease)
  in product cost
10.35
 0.244
27.12
 0.640
 Base case regulations are the applicable SIP's for all pollutants.

 Regulatory Option I regulations for these boilers are SIP limits for
 SO,; 43 ng/J (0.1 Ib/MMBtu) for PM; and 258 ng/J (0.6 Ib/MMBtu) for NO
c
 Regulatory Option V regulations for these boilers are 860 ng/J (2.0 Ib/MMBtu)
 and 90-50% reduction for S0?; 43 ng/J (0.1 Ib/MMBtu) for PM; and 258 ng/J
 (0.6 Ib/MMBtu) for NO      *
H
 Estimated from industry contracts.

eBased on model plant configuration.

 Steam costs are 1990 pre-tax estimates.
                                 9-96

-------
     TABLE 9-53.  CHANGE IN PROFIT MARGIN DUE TO NEW BOILER INVESTMENT:
                       LIQUOR DISTILLING MODEL PLANT

Regulatory Option

Sales
Expenses
Gross profit
Taxes
Net income
Return on
assets, %
Base
106 $
28.22
27.45
0.78
0.39
0.39
1.
Case8
% of
sales
100.00
97.26
2.74
1.37
1.37
26
I
106 $
28.22
27.80
0.42
0.21
0.21
0
b
% of
sales
100.00
98.52
1.48
0.74
0.74
.68
V
106 $
28.22
28.35
(0.14)
0.00
(0.14)
(0
c
% of
sales
100.00
100.46
(0.46)
0.00
(0.46)
.50)
aBase case regulations are the applicable SIP's for all pollutants.

 Regulatory Option I regulations for these boilers are SIP limits for
 S02; 43 ng/J (0.1 Ib/MMBtu) for PM; and 258 ng/J (0.6 Ib/MMBtu) for NO .

GRegulatory Option V regulations for these boilers are 860 ng/J (2.0 lb/
 MMBtu) and 90-50% reduction for S09; 43 ng/J (0.1 Ib/MMBtu) for PM; and
 258 ng/J (0.6 Ib/MMBtu) for NO .   *
                               xx
                                 9-97

-------
Return on assets is 0.68 percent in Option I and negative 0.50 in Option V
as compared to 1.26 percent in the base case.
     Table 9-54 presents comparative coverage and debt/equity ratios for
the regulatory options.  Change in coverage ratio as a function of debt
level assumed is greatest in Option V.  This option shows the coverage
ratio decreasing from 5.45 with zero percent new debt to 4.30 with 100
percent debt.  The base case coverage ratio decreases from 5.45 with zero
percent debt to only 4.52 with 100 percent debt.  However, coverage ratios
for all financing options used for each of the regulatory options still are
above the 3.0 coverage benchmark.
     The results of the analysis indicate that product cost is expected to
increase by slightly over one-half of a percent at most.  New steam costs
for the regulatory options comprise a relatively small portion of average
product cost.  Profitability shows a decline as a result of the regulatory
options when compared to the base case.  Return on assets, for example,
decreases from 1.26 percent in the base case to negative 0.50 percent in
Option V.
     With regard to financing capability, the analysis of coverage ratios
indicates that new boiler investment can be funded totally by debt while
still meeting the 3.0 coverage benchmark.  The industry maintains a rela-
tively stabilized solvency position even when 100 percent debt financing is
assumed, because of its moderate leverage position.
                                 9-98

-------
     TABLE 9-54.  CAPITAL AVAILABILITY:   LIQUOR DISTILLING MODEL FIRM
                                                   Regulatory Option
Financial indicator            Base Case3           I             Vc
Coverage ratio
  Percent financed
    by debt
          0                       5.45             5.45         5.45
         25                       5.19             5.13         5.10
         50                       4.95             4.83         4.81
         75                       4.73             4.59         4.53
        100                       4.52             4.35         4.30
Debt/equity ratio
  Percent financed
    by debt
0 0.27
25 0.29
50 0.32
75 0.34
100 0.37
0.27
0.30
0.32
0.35
0.39
0.27
0.29
0.33
0.36
0.39
 Base case regulations are the applicable SIP's for all  pollutants.
 Regulatory Option I regulations for these boilers are SIP limits for
 SO,; 43 ng/J (0.1 Ib/MMBtu) for PM; and 258 ng/J (0.6 Ib/MMBtu) for
 N0x-
r  A
 Regulatory Option V regulations for these boilers are 860 ng/J (2.0 lb/
 MMBtu) and 90-50% reduction for S09; 43 ng/J (0.1 Ib/MMBtu) for PM; and
 258 ng/J (0.6 Ib/MMBtu) for NO .   *
                               /\
                                 9-99

-------
9.3  REFERENCES

1.   U.S. Department of Commerce,  Bureau of the Census.  1977 Census of
     Manufactures.  Washington, D.C. ,  Subscriber Services, Bureau of the
     Census.  MC77-1-20F-3(P).  April  1979.

2.   Meeting.  Golding, Rex,  Energy  and Environmental  Analysis,  Inc. with
     Olsen, Van R., United  States  Beet Sugar Association.  July  1979.
     Characteristics of the domestic beet  sugar industry.

3.   U.S. Department of Labor,  Bureau of Labor Statistics.   Producer Price
     Indices.  Washington,  D.C.

4.   U.S. Department of Commerce,  Bureau of the Census.  1977 Census of
     Manufactures.  Washington, D.C.,  Subscriber Services, Bureau of the
     Census.   MC77-1-20C-2(P).  July 1979.

5.   The Almanac  of the Canning, Freezing  and Preserving Industries.
     Westminister, Maryland,  Edward  E. Judge & Sons,  Inc.    1978.   p. 490,
     511.

6.   The Almanac  of the Canning, Freezing  and Preserving Industries. West-
     minister, Maryland,  Edward E. Judge & Sons, Inc.   1978. p.  490, 511.

7.   U.S. Department of Commerce,  Bureau of the Census.  1977 Census of
     Manufactures.  Washington, D.C.,  Subscriber Services, Bureau of the
     Census.  MC77-1-30A(P).   March  1979.

8.   Telecon.  McGovern,  Joan,  Energy and  Environmental Analysis, Inc. with
     Scharf, Jerry, National  Association of Recycling Industries.   August
     13, 1979.  Characteristics of the reclaimed rubber industry.

9.   Telecon.  McGovern,  Joan,  Energy and  Environmental Analysis, Inc. with
     Spangeberg,  Bill, Rubber Manufacturers Association.   October 24, 1979.
     Reclaimed rubber  historic production  figures.

10.  U.S. Department of Labor.  Producer Price Indices.  Washington, D.C.,
     Bureau of Labor Statistics.

11.  Ward's Automotive Yearbook.   Ward's Communications.   Forty-first Edi-
     tion.  1978.

12.  Becker, Harold S.  Corporate  Strategies of Automotive Manufacturers.
     Volume II:   A Comprehensive Summary of Likely Actions by Domestic
     Companies Under Fuel-Economy  Regulations:  1978-1985.   Prepared for
     the U.S. Department  of Transportation, National  Highway Traffic Safety
     Administration by the  Futures Group,  Glastonbury, Connecticut.  Final
     Report.  November, 1978.

13.  Cantrell, Ailleen.   Annual Refining Survey.   The Oil  and Gas Journal.
     77(13):122-154.  March 26, 1979.
                                 9-100

-------
14.   McCoslin, John C.   Limited Supply to Reduce Demand Below 1978 Levels.
     Midyear Report, The Oil and Gas Journal.   77(31):152.   July 30, 1979.

15.   Refiners Adding Capacity in Global Building Surge.  The Oil and Gas
     Journal.  76(17):66.   April 24, 1978.

16.   The Oil and Gas Journal.  Various Annual  Refining Issues.

17.   Small Refiners Claim Majors Have Edge.   The Oil and Gas Journal.
     76(17):66.   April  24, 1978.

18.   Myers, J.G., et al.  The Conference Board.  Energy Consumption in
     Manufacturing:  A Report to the Energy Project of the Ford Foundation.
     Cambridge,  Massachusetts, Ballinger Publishing Co.  1974.

19.   Myers, J.G., et al.  The Conference Board.  Energy Consumption in
     Manufacturing:  A Report to the Energy Project of the Ford Foundation.
     Cambridge,  Massachusetts, Ballinger Publishing Co.  1974.

20.   U.S.  Department of Commerce, Bureau of the Census.  1977 Census of
     Manufactures, Preliminary Reports, Blast Furnaces and Steel Mills—SIC
     3312.  Washington, D.C., Subscriber Services, Bureau of the Census.
     MC77-1-33A-1(P).  September 1979.

21.   Energy and Environmental Analysis, Inc.  Coal - A Data Book.  Prepared
     for the President's Commission on Coal.  Arlington, Virginia.  1979.

22.   American Iron and Steel Institute.  Annual Statistical Report.  Washington,
     D.C.   1978.

23.   Wheeling-Pittsburgh Chief Discusses Steel's Profit Problems.  Iron and
     Steel Engineer.  55(12):83-86.  December, 1978.

24.   Wheeling-Pittsburgh Chief Discusses Steel's Profit Problems.  Iron and
     Steel Engineer.  55(12):83-86.  December, 1978.

25.   U.S.  Department of Commerce, Bureau of the Census.  1977 Census of
     Manufactures. Washington, D.C., Subscriber Services, Bureau of the
     Census.  MC 77-l-20H-4(P). August 1979.

26.   Standard & Poor's Directory of Corporations, Executives and Directors.
     1979 Edition.

27.   U.S.  Department of Commerce, Bureau of the Census.  1977 Census of
     Manufacturers.  Washington, D.C., Subscriber Services, Bureau of the
     Census.  MC 77-l-20H-4(P). August 1979.

28.   U.S.  Department of Commerce, Bureau of the Census.  1977 Census of
     Manufactures.  Washington, D.C., Subscriber Services, Bureau  of the
     Census.  MC 77-l-20H-4(P). August 1979.
                                 9-101

-------
29.  U.S. Department of Commerce,  Industry  and Trade Administration.  1980
     U.S. Industrial Outlook.  Washington,  D.C., Superintendent of  Documents,
     Government Printing Office.

30.  List of companies compiled  from:  Thomas Register of American  Manufac-
     turers.  New York, Thomas Publishing Company,  69th  edition 1979, and
     EEA survey of manufacturers.

31.  Stationary Watertube  Boiler Sales 1978.  American Boiler Manufacturers
     Association.  Arlington, Virginia.  1979.   p.  5.

32.  Arthur D. Little, Inc.  The Economic Effects of Environmental  Regula-
     tions on the Pollution  Control  Industry.   Prepared  for  Environmental
     Protection Agency.  Washington,  D.C.   Publication No. EPA/230/1-78-002.
     September 1978.  p.7.

33.  Arthur D. Little, Inc.  The Economic Effects of Environmental  Regula-
     tions on the Pollution  Control  Industry.   Prepared  for  Environmental
     Protection Agency.  Washington,  D.C.   Publication No. EPA/230/1-78-002.
     September 1978.  p.109.

34.  PEDCo Environmental,  Inc.   EPA  Industrial  Boiler FGD Survey:   First
     Quarter 1979.  Prepared for U.S.  Environmental Protection Agency.
     Washington, D.C.  April 1979.   227  p.

35.  U.S. Department of Commerce.  Bureau of the Census.  Selected  Indus-
     trial Air Pollution Equipment.   In:  Current Industrial Reports.   1973,
     1975, 1977.

36.  Acurex Corporation.   Technology Assessment Report for Industrial
     Boilers Application:  NO  Combustion Modifications.  Washington, D.C.,
     EPA, EPA-600/7-79-178f. December 1979.

37.  Telecon.  Chronowski, Robert, Cleaver  Brooks,  with  Vidas, E.H.,  EEA.
     February 22, 1980.

38.  Telecon.  Orvidas, Ed,  Vapor. Corp., with  Vidas, E.H.,  EEA.  June  6,
     1980.

39.  Letter from American  Boiler Manufacturers  Association to Mobley, J.D.,
     EPA.  July 24, 1979.  Comments  on NO   Combustion Modification  ITAR.
                                         J\

40.  Arthur D. Little, Inc.  The Economic Effects of Environmental  Regula-
     tions on the Pollution  Control  Industry.   Prepared  for  U.S.  Environ-
     mental Protection Agency.   Washington, D.C.  Publication No. EPA/230/
     1-78-002.  September  1978.   p.  109.

41.  Industrial Gas Cleaning Institute.  Assessment of Manufacturers'
     Capabilities to Meet  Requirements for  Control  of Emissions of  Sulfur
     Dioxide, Particulate  Matter,  and Nitrogen  Oxides from Industrial
     Boilers.  Prepared for  U.S.  Environmental  Protection Agency.   Research
     Triangle Park, N.C.   July 1979.   p. 85.
                                  9-102

-------
42.  Industrial Gas Cleaning Institute.  Assessment of Manufacturers'
     Capabilities to Meet Requirements for Control of Emissions of Sulfur
     Dioxide, Particulate Matter, and Nitrogen Oxides from Industrial
     Boilers.  Prepared for U.S. Environmental Protection Agency.   Research
     Triangle Park, North Carolina.   July 1979.   p. 27, 52, 67.

43.  PEDCo Environmental, Inc.   Effects of Alternative New Source Perfor-
     mance Standards on Flue Gas Desulfurization System Supply and Demand.
     Prepared for U.S.  Environmental Protection Agency:  Washington, D.C.
     Publication No.  EPA-600/7-78-003.  March 1978.   p.  4-4 to 4-11.

44.  Industrial Gas Cleaning Institute.  Assessment of Manufacturers'
     Capabilities to Meet Requirements for Control of Emissions of Sulfur
     Dioxide, Particulate Matter, and Nitrogen Oxides from Industrial
     Boilers.  Prepared for U.S. Environmental Protection Agency.   Research
     Triangle Park, North Carolina.   July 1979.   Figure 6-2, p. 104 (for
     FGD), Figure 6-4,  p. 106 (for FF and ESP).
                                 9-103

-------
                                    TECHNICAL REPORT DATA
                             (Please read Instructions on the reverse before completing)
  REPORT NO.

  EPA 450/3-82-006a,b
2.
                               3. RECIPIENT'S ACCESSION NO.
 . TITLE AND SUBTITLE
 Fossil  Fuel-Fired Industrial Boilers
 Background  Information
                               6. REPORT DATE
                                  March 1982
                               6. PERFORMING ORGANIZATION CODE
  AuTnORtS)
                                                              8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
 Office of Air Quality Planning and Standards
 U. S. Environmental  Protection Agency
 Research Triangle Park, North  Carolina   27711
                                                              10. PROGRAM ELEMENT NO.
                               11. CONTRACT/GRANT NO.
                                                                  68-02-3058
12. SPONSORING AGENCY NAME AND ADDRESS
 DAA  for Air Quality Planning and Standards
 Office  of Air, Noise, and Radiation
 U. S.  Environmental Protection Agency
 Research Triangle Park, North Carolina   27711
                               13. TYPE OF REPORT AND PERIOD COVERED
                               14. SPONSORING AGENCY CODE
                                  EPA/200/04
15. SUPPLEMENTARY NOTES
 16. ABSTRACT
              This document provides background  information for the fossil fuel-fired
         industrial  boiler source category.   Fossil  fuels considered include coal,
         oil  and natural gas.  Background  information  for industrial boilers includes
         a  survey of boiler types, sizes,  operating  characteristics, and existing
         State and Federal regulations.  Uncontrolled  emissions of particulate
         matter, sulfur dioxide and nitrogen  oxides  are quantified and factors
         affecting these emissions are discussed.  Control  technologies for
         particulate matter, sulfur dioxide and  nitrogen oxides are identified and
         discussed with respect to the technologies'  applicability to industrial
         boilers, developmental status,  and factors  affecting performance.  Emissions
         data for each technology are also presented.   Finally, environmental, energy
         and  cost impacts of applying these technologies to fossil fuel-fired
         industrial  boilers are presented  and discussed.  This information was
         developed in support of potential new source  performance standards for
         industrial  boilers.
17.
                                  KEY WORDS AND DOCUMENT ANALYSIS
                   DESCRIPTORS
                                                b.IDENTIFIERS/OPEN ENDED TERMS
                                             c.  COSATI Field/Group
  Air pollution
  Pollution control
  Standards of performance
  Fossil  fuel-fired industrial  boilers
                  Fossil  fuel-fired industri
                     boilers
                  Air  pollution  control
na. OISTRISUT.ON STATEMENT
  Release unlimited.   Available from EPA
  Library (MD-35), Research  Triangle Park,
  North Carolina  27711               	
                  19. SECURITY CLASS I This Report/
                    unclassified
21. NO. OF PAGES
   869
                  20. SECURITY CLASS (Thispagei
                    unclassified
                                              22. PRICE
 SPA Form 2220-1 (Rev. 4-77)    PPSVIOUS ec'TION is OBSOLETE

-------