United States Office of Air Quality EPA-450/2-78-007a
Environmental Protection Planning and Standards July 1978
Agency Research Triangle Park NC 27711
_
x>EPA Electric Utility
Steam Generating
Units
Background
Information for
Proposed S02
Emission Standards
-------
Background Information and
Draft Environmental Impact Statement
for Proposed Sulfur Dioxide Emission Standards for
Electric Utility Steam Generating Units
Type of Action: Administrative
Prepared by:
-f
Don R. Goodwin (Date)
Director, Emission Standards and Engineering Division
Environmental Protection Agency
Research Triangle Park, North Carolina 27711
Approved by:
(Date)
Director, Office of Air Quality Planning and Standards
Environmental Protection Agency
Research Triangle Park, North Carolina 27711
Draft Statement Submitted to EPA's - .^ ... , ,
Office of Federal Activities for Review on (Date)
This document may be reviewed at:
Central Docket Section
Room 2903B, Waterside Mall
401 M Street
Washington, D. C. 20460
Additional copies may be obtained at:
U. S. Environmental Protection Agency Library (MD-35)
Research Triangle Park, North Carolina 27711
National Technical Information Service
5285 Port Royal Road
Springfield, Virginia 22161
-------
EPA-450/2-78-007a
Electric Utility Steam Generating Units
Background Information for Proposed SC>2
Emission Standards
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
July 1978
-------
This report is issued by the Environmental Protection Agency to report
technical data of interest to a limited number of readers. Copies are avail-
able - in limited quantities - from the Library Services Office (MD-35) ,
U.S. Environmental Protection Agency, Research Triangle Park, North
Carolina 27711; or, fur a fee, from the National Technical Information
Service, 5285 Port Royal Road, Springfield, Virginia 22161.
Publication No. EPA-450/2-78-007a
-------
TABLE OF CONTENTS
Page
LIST OF ILLUSTRATIONS viii
LIST OF TABLES xii
1.0 INTRODUCTION 1-1
1.1 Background 1-2
1.1.1 Present Standard 1-3
1.1.2 Revisions of the Standard 1-5
1.2 Statutory Authority 1-7
2.0 DESCRIPTION OF AND RATIONALE FOR THE PROPOSED ACTION 2-1
3.0 LEGAL ALTERNATIVES 3-1
3.1 No Action 3-1
3.2 Delayed Action 3-2
3.3 Nature of Standard and Stringency of Controls 3-2
3.4 Control Practices 3-2
4.0 ALTERNATIVE CONTROL TECHNOLOGIES 4-1
4.1 Burning Low-Sulfur Coal 4-1
4.1.1 Availability of Acceptable Coal 4-2
4.2 Fuel Treatment Processes 4-7
4.2.1 Physical Coal Cleaning 4-8
4.2.2 Chemical Coal Cleaning 4-17
4.2.3 Solvent Refined Coal Process 4-31
4.2.4 Summary and Conclusions 4-40
4.3 Fluidized Bed Combustion 4-42
4.3.1 Overview 4-42
4.3.2 FBC System 4-47
4.3.3 Status of FBC 4-57
4.3.4 FBC Vendors 4-58
4.3.5 Summary 4-60
iii
-------
TABLE OF CONTENTS
(Continued)
4.4 Flue Gas Desulfurization 4-60
4.4.1 Overview of Flue Gas Desulfurization Processes 4-60
4.4.2 Sulfur Dioxide Removal 4-64
4.4.3 FGD Process 4-76
4.4.4 FGD Wastes 4-134
4.4.5 Status of Flue Gas Desulfurization Technology 4-137
4.4.6 Vendor Capabilities 4-140
4.4.7 Availability 4-148
5.0 DESCRIPTION OF THE EXISTING ENVIRONMENT 5-1
5.1 The Electric Power Industry in the United States 5-1
5.1.1 Generating Capacity 5-2
5.1.2 Production of Electrical Energy 5-3
5.1.3 Supply of Coal 5-4
5.1.4 Origin and Destination of Coal for New Units 5-6
5.1.5 Long-Range Projections 5-13
5.2 Coal Resources of the United States 5-16
5.2.1 Geographical Distribution of Coal Deposits 5-18
5.2.2 Sulfur Content of U.S. Coals 5-21
5.2.3 Coal-Producing Regions 5-23
5.3 Air Quality 5-23
5.3.1 Ambient S02 Concentrations 5-23
5.3.2 S02 Emissions 5-28
5.3.3 Air Quality Modeling Results 5-30
5.4 Present Water Environment 5-34
5.4.1 Water Quality 5-35
5.4.2 Water Quality 5-48
5.5 Land Use 5-52
5.5.1 Land Used for the Physical Plant 5-53
5.5.2 Land Used for Solid Waste Disposal 5-54
5.5.3 Current Land Requirements 5-55
5.5.4 Projected Land Requirements 5-56
IV
-------
TABLE OF CONTENTS
(Continued)
Page
5.6 Energy Consumption Associated with Control Measures 5-60
5.6.1 Flue Gas Desulfurization 5-60
5.6.2 Other Control Options 5-64
5.6.3 Energy Penalty Projections (1987-1997) 5-67
6.0 ASSESSMENT OF ENVIRONMENTAL IMPACTS 6-1
6.1 Impacts on Coal Resources and Transportation 6-2
6.2 Air Quality 6-4
6.2.1 SO^ Emissions 6-4
6.2.2 Ambient SO Concentrations 6-11
6.3 Water 6-13
6.3.1 Water Quantity 6-13
6.3.2 Water Quality 6-17
6.4 Land Use 6-17
6.4.1 Land Use for the Physical Plant 6-17
6.4.2 Land Used for Solid Waste Disposal 6-18
6.5 Ecology 6-21
6.5.1 Ecology at the Physical Plant 6-21
6.5.2 Ecology at the Disposal Site 6-22
6.6 Energy Penalities Associated with Alternate Strategies 6-23
6.7 Noise 6-27
6.8 Secondary Impacts 6-28
7.0 ECONOMIC IMPACT ANALYSIS 7-1
7.1 Industry Profile 7-1
7.1.1 General Industry Background 7-1
7.1.2 Predominance of Coal in Electric Power
Generation 7-6
v
-------
TABLE OF CONTENTS
(Continued)
Page
7.2 Cost Analysis of Alternative Emission Control Systems 7-10
7.2.1 New Facilities 7-10
7.2.2 Basis of Cost Analysis 7-10
7.2.3 Estimated Control Costs 7-18
7.3 Other Cost Considerations 7-22
7.3.1 Additional Capital and Operating Costs 7-22
7.3.2 Energy Penalty Costs Associated with SO
Control 7-24
7.4 Economic Impact of Alternative Control Systems 7-24
7.4.1 Increased Costs to Utility Industry 7-24
7.4.2 Financial Impact on Utility Industry 7-36
7.4.3 Effects on Price of Electricity to Consumer 7-40
7.4.4 Secondary Economic Impacts 7-50
7.5 Cost-Effectiveness of Revised NSPS 7-83
7.5.1 Costs of SO Reduction on a Ton-Per-Year
Basis 7-83
7.5.2 Limitations of Cost-Effectiveness 7-86
APPENDIX A: SO REMOVAL MECHANISMS AND EFFICIENCY A-l
APPENDIX B: ENERGY REQUIREMENTS B-l
APPENDIX C: REHEAT OF SCRUBBED FLUE GASES C-l
APPENDIX D: FGD SYSTEM PERFORMANCE D-l
APPENDIX E: PLANNED AND OPERATING FGD SYSTEMS E-l
APPENDIX F: CONSTRUCTION SCHEDULE F-l
APPENDIX G: ASSUMED PARAMETERS IN FGD COSTS G-l
APPENDIX H: MEASURES TO IMPROVE FLUE GAS DESULFURIZATION
AVAILABILITY AND OPERATING PROBLEMS AND
SOLUTIONS H-l
VI
-------
TABLE OF CONTENTS
(Continued)
Page
APPENDIX I: EFFECT OF COAL PROPERTIES ON FGD SYSTEMS 1-1
APPENDIX J: FORECASTS OF FUTURE ELECTRIC UTILITY INDUSTRY
STRUCTURE J-l
REFERENCES K-l
vii
-------
LIST OF ILLUSTRATIONS
Figure Number
4-1 Meyers/TRW Process Flow Diagram
4-2 Battelle Process Flow Diagram
4-3 Hazen Process Flow Diagram
4-4 KVB Process Flow Diagram
4-5 LOL Process Flow Diagram
4-6 BOM/ERDA Process Flow Diagram
4-7 Solvent Refined Coal Process
4-8 Percent of United States Bituminous Coal
Cleanable to a Given Sulfur Content for
Various Cleaning Scenarios 4-41
4-9 Control of Atmospheric Pollution by FBC 4-45
4-10 Generic Process Flow Sheet for Category I
Atmospheric FBC of Coal 4-48
4-11 Effect of Ca/S Mole Ratio on Sulfur
Retention 4-51
4-12 Generic Process Flow Sheet for Category II
Pressurized, Combined Cycle FBC of Coal 4-53
4-13 Generic Process Flow Sheet for Category III -
Pressurized, Combined Cycle FBC of Coal
(Adiabatic Combustor) 4-54
4-14 Comparison of S0£ Removal Results -
Dolomite Sorbent 4-56
4-15 Flue Gas Desulfurization Processes Tested
on Coal-Fired Boilers 4-61
4-16 Schematic of Three-Bed TCA 4-69
4-17 Schematic of Venturi Scrubber and Spray
Tower 4-72
4-18 Station Electrical Loss As a Function L/G
Ratio and Nozzle Pressure 4-74
4-19 Station Electrical Loss As a Function of
Draft Requirements 4-75
4-20 Typical Process Flow Diagram for Lime/-
Limestone Scrubbing 4-77
4-21 Scrubber System Operability - Green River
No. 1, 2 and 3 4-85
4-22 Effect of Circulating Liquor Flow Rate on
S02 Removal at Constant Gas Flow 212
(450,000 SCFM) Mohave Plant 4-89
4-23 La Cygne Availability History 4-94
4-24 Availability History Sherburne No. 1 and
No. 2 4-96
4-25 Simplified Process Diagram for Sodium
Carbonate Scrubbing System 4-98
Vlll
-------
LIST OF ILLUSTRATIONS
(Continued)
Figure Number
4-26 Simplified Process Diagram for Double
Alkali System 4-103
4-27 Simplified Diagram for Magnesium Oxide
Recovery System 4-114
4-28 Simplified Process Diagram for Wellman
Lord Recovery System 4-123
4-29 Inlet and Outlet SO- Concentrations During
Run No. 1 4-127
4-30 Inlet and Outlet S02 Concentrations During
Run No. 2 4-128
4-31 Inlet and Outlet S02 Concentrations During
Run No. 3 4-129
4-32 Flow Sheet - Two-Stage Dry Scrubber/S02
Absorber 4-131
4-33 Average Availability for Selected FGD Systems 4-150
5-1 Flow of Coal to New Generating Units from
the Western Regions of the Northern
Great Plains (in 1000 Tons) 5-11
5-2 Flow of Coal to New Generating Units from
the Appalachian Region, from U.S. Bureau
of Mines District 15, and from ttie
Mountain Region (in 1000 Tons) 5-12
5-3 Air Quality Control Regions; Status of
Compliance with Ambient Air Quality
Standards for Sulfur Dioxide 5-27
5-4 System //1-Once-Through Water Management 5-36
5-5 System //2-Partial Recirculatory Water
Management 5-37
5-6 System #3-Recirculatory Water Management 5-38
5-7 System #4-Zero Discharge Water Management 5-39
6-1 National Power-Plant S02 Emissions under
Alternative Control Scenarios, High
Growth 6-7
6-2 National Power-Plant S02 Emissions
Under Alternative Control Scenarios,
Moderate Growth 6-8
6-3 Primary Ambient Air Quality Standards 6-12
IX
-------
LIST OF ILLUSTRATIONS
(Continued)
Figure Number Page
7-1 Percentage Growth Rate Over Previous Year
Reported by Major U.S. Utility Systems 7-2
7-2 Annual Peak Reserves as a Percentage of
Total Availability Capacity 7-3
7-3 Annual Capacity Factor for the Major U.S.
Electric Utilities 7-5
7-4 Aggregate Annual Number of Installed Coal-
Fired Units Over 25 MWe on a 5-Year Running
Average 7-8
7-5 Average Size of Newly Installed Coal-Fired
Units on a 5-Year Running Average 7-9
7-6 Actual and Projected Petroleum Demand and
Domestic Production (1950-1985) 7-77
A-l Effect: of Ca/S Mole Ratio on Sulfur Retention A-3
A-2 Comparison of Performance of Greer and
Germany Valley Limestones A-4
A-3 Comparison of SO Removal Results - Limestone
Sorbent A-6
A-4 Sulfur Retention as a Function of Superficial
Gas Velocity A-7
A-5 Effect of Gas Residence Time on Ca/S Ratio
Required to Meet Present EPA SO Emission
Standard A-10
A-6 Comparison of Dolomite No. 1337 and Limestone
No. 1359 as SO Sorbents on a Mass Feed
Rate Basis A-14
B-l Energy Requirements for S02 Control - 520 ng/J
at 500 MW Plant B-5
B-2 Energy Penalties for S02 and Particulate
Control - 90% S02 Removal Control Level,
500 MW Plant B-7
B-3 Energy Requirements for S0? and Particulate
Control - 220 ng/J Control Level, 500 MW
Plant B-8
B-4 Energy Penalties for S0? Control - Summary
of Effects of SO Control Level B-10
-------
LIST OF ILLUSTRATIONS
(Concluded)
Figure^ Number
D-l
D-2
D-3
D-4
D-5
D-6
D-7
D-8
D-9
D-10
F-l
F-2
J-l
Effect of Inlet SC>2' Concentration on S02
Removal Efficiency for Fixed Design and
Operating Conditions D-3
Effect of Liquid-to-Gas Ratio on S02
Removal Efficiency with Low Sulfur Coal
at the Mohave Power Station D-5
Effect of Liquid-to-Gas Ratio on S02
Removal Efficiency - TCA with Limestone D-6
Effect of Gas Velocity on S02 Removal
Efficiency - TCA with Limestone D-7
Effect of Scrubber Inlet pH on S02 Removal
Efficiency - TCA with Limestone Dr8
Effect of Bed Height on S02 Removal
Efficiency - TCA with Limestone D-10
Effect of Liquid-to-Gas Ratio on S02
Removal Efficiency - TCA with Limestone
and Magnesium D-l2
Effect of Scrubber Inlet pH on S02 Removal
Efficiency - TCA with Limestone and
Magnesium D-13
Effect of Magnesium on S02 Removal
Efficiency - TCA (No Spheres) with
Limestone D-14
S02 Absorption Efficiency for Two
Scrubbers in Series D-15
Construction Schedule for a Typical
(500 MW) Power Plant F-2
Construction Schedule for a Typical Power
Plant Equipped with FGD System F-3
Projections of Electric Utility Coal
Consumption J-27
XI
-------
LIST OF TABLES
Arable Number Page
4-1 Distribution of Coal Samples by Region
and State 4-5
4-2 Estimates of Recoverable Reserves of Raw
Coal Characterized by Emission Rate of
Sulfur Dioxide from Uncontrolled Combustion 4-6
4-3 Physical Coal Cleaning Process Environmental
Problems 4-11
4-4 U.S. Recoverable Reserves to Meet the NSPS,
Raw and Prepared Coal to Meet the 1985
Annual Demand from New and Existing Electric
Utilities (Standard - Ib S02/1Q6 Btu) 4-13
4-5 Typical Analyses of Coals Used 4-36
4-6 Typical Operating Conditions and Results 4-37
4-7 Typical Analyses of Solvent Refined Coal 4-38
4-8 Major Coal Cleaning Process Considerations 4-43
4-9 Selected List of Operational Fluidized
Bed Combustors 4-59
4-10 Lime/Limestone Process Evaluation 4-80
4-11 Lime Based FGD Systems in the United States 4-81
4-12 Power Plant and FGD System Design Data 4-83
4-13 Green River Power Station Operational Data
FGD Unit 4-84
4-14 Power Plant and FGD System Design Data 4-86
4-15 Power Plant and FGD System Design Data 4-88
4-16 Operational Data - Mohave Horizontal FGD
Unit 4-90
4-17 Major Domestic FGD Installations - Limestone
Slurry 4-92
4-18 Power Plant and FGD System Design/Operating
Data, La Cygne No. 1 4-93
4-19 Sherburne County Generating Plant - Unit 1 -
Performance Data 4-97
4-20 Sodium Carbonate Scrubbing Evaluation 4-100
4-21 Double Alkali Process Evaluation 4-104
4-22 CEA/ADL Double Alkali Prototype Scrubber
Performance History: Operation and
Viability Parameters 4-111
4-23 Magnesium Oxide Scrubbing 4-116
4-24 Operability of MgO System at Mystic No. 6 4-118
4-25 SOX Emissions Test Results for MgO FGD
System - Dickerson 4-120
4-26 Operability Data for Dickerson No. 3 4-120
4-27 Sodium Sulfite Scrubbing Evaluation 4-12%
XII
-------
LIST OF TABLES
(Continued)
Table Number
4-28 Spray Dryer/Fabric Filter Process Evaluation
4-29 Volume of Scrubber Wastes Produced by
Typical Nonregenerable Systems over a
30-Year Period 4-135
4-30 Breakdown of FGD Units 4-139
4-31 FGD Applications to Coal-Fired Boilers 4-139
4-32 Approximate Process Distribution of Planned
FGD Systems on New Coal-Fired Utility
Boilers 4-140
4-33 Projected Utilization of Flue Gas
Desulfurization on New Coal-Fired Units 4-141
4-34 Comparison of Supply Versus Demand for
FGD Systems on New Coal-Fired Utility
Boilers under Present NSPS 4-143
4-35 Comparison of Supply Versus Demand for
FGD Systems on Coal-Fired Utility
Boilers Under More Stringent NSPS 4-143
4-36 Time Required for FGD System Design,
Installation, and Startup 4-144
4-37 Lead Time and Delay Frequency of Various
Items in the Design and Installation
of an FGD System 4-145
4-38 Guarantees Offered by Manufacturers for
SO- Removal 4-147
4-39 FGD System Performance DataAverage Values 4-151
5-1 Incremental Coal Demand in 1980 and 1985
Attributed to New Units Scheduled for
Operation Between 1978 and 1985 5-7
5-2 Projected Movement of Coal for New Units
Scheduled to Become Operational Between
1976 and 1985 5-9
5-3 Predicted National Coal Production 5-14
5-4 Regional Coal Production in High Growth
Scenario 5-16
5-5 Reserve Base and Recoverable Reserves of
Coal in the United States 5-20
5-6 Reserve Base of Coal in the United States,
by State and Sulfur Content 5-24
5-7 Coal Producing Regions of the United States 5-25
5-8 Regional S02 Emissions 5-29
5-9 Air Quality Impact - yg/m (% Federal
Ambient Air Quality Standard - S02) 5-32
Kill
-------
LIST OF TABLES
(Continued)
Table Number Page
5-10 Sluice Water Requirement 5-41
5-11 Base Case: Model Power Plant Water
Consumption 5-43
5-12 Base Case: FGD System Make-Up Water
Requirement 5-44
5-13 Model Plant System Water Requirements 5-46
5-14 Water Consumed by FGD Systems 5-47
5-15 Range of Concentration of Constituents in
Scrubber Liquors Studied 5-51
5-16 Projected Land Requirements for Coal-Fired
Electric Generating Plants 5-57
5-17 Range of Concentrations of Chemical
Constituents in FGD Sludges from Lime,
Limestone, and Double-Alkali Systems 5-59
5-18 Energy Penalties for Model S02 Control
Systems (520 ng S02/J) 5-63
5-19 Analysis Assumptions for the Energy Penalty
Associated with Coal Cleaning and Western
Coal Transportation 5-65
5-20 Energy Associated with 520 ng/J Standard
(10i8 Joules) 5-66
6-1 Impacts on Regional Production 'of Coal 6-5
6-2 Regional and National Power-Plant S02
Emissions Assuming High Growth 6-9
6-3 Regional and National Power-Plant S02
Emissions Assuming Moderate Growth 6-10
6-4 Model Plant System Water Requirements 6-14
6-5 Water Consumed by FGD Systems 6-16
6-6 Projected Land Requirements for Coal-Fired 6-
Electric Generating Plants - 90 Percent
Scrubbing 6-19
6-7 Projected Land Requirements for Coal-Fired
Electric Generating Plants - 220 ng/J
(0.5 lb S02/106 Btu) 6-20
6-8 Energy Penalties for Model S02 Control
Systems 6-24
6-9 Energy Consumed by FGD Systems in 1995
(KP Megajoules) 6-26
6-10 Energy Consumed in Transporting Coal to
Electric Generating Plants (109 Megajoules) 6-27
xiv
-------
LIST OF TABLES
(Continued)
Table Number
7-1 Orders for Coal-Fired Boilers 7-4
7-2 Projected Electric Capacity Mix 7-7
7-3 Costs of SO Control Alternatives for
Level of 1.2 lbs/106 Btu 7-15
7-4 Costs of SO Control Alternatives for 90%
SO Removal 7-16
7-5 Costs of SO Control Alternatives for Level
of 220 ng/J (0.5 lb/106 Btu) 7-17
7-6 Incremental Costs of Removing 90% S02
(Compared to Costs of Meeting 1.2 lb/10
Btu) 7-19
7-7 Environmental Capital Costs for Representa-
tive New Plant (1975 dollars) 7-22
7-8 Capital Expenditures 1975-1985 by Type of
Pollution Control Equipment 7-23
7-9 O&M Expenses for the Industry 7-23
7-10 Energy Penalty in Mills/KWH for Selected
Control Processes, Scenarios and Plant
Sizes 7-25
7-11 Alternative Teknekron NSPS Scenarios 7-26
7-12 Nationwide Costs of Generating Electricity
under Alternative NSPS 1986-1995 -
Moderate Rate of Power Growth 7-30
7-13 Nationwide Costs of Generating Electricity
under Selected Alternative of NSPS 1986-
1995 - High Rate of Power Growth 7-31
7-14 Pollution Control Costs by Region for
Alternative NSPS Revisions, 1986-
1995 (Moderate Growth of Power) 7-34
7-15 Pollution Control Costs by Region for
Alternative NSPS Revisions, 1986-
1995 (High Growth of Power) 7-35
7-16 Capital Investment 7-37
7-17 Long-Term External Financing 7-39
7-18 Return on Equity (Return on Common Stock) 7-41
7-19 Interest Coverage 7-42
7-20 Quality of Earnings 7-43
7-21 Regional Price Impacts on the Electric
Utility Industry of Alternative NSPS
Revisions, 1995 7-44
xv
-------
LIST OF TABLES
(Continued)
Table Number
7-22 Per Capita Cost of Alternative NSPS
ReA/isions for Investor-Owned Utilities 7-46
7-23 Per Capita Cost of Alternative NSPS for
Investor-Owned Utilities 7-47
7-24 1990 Coal Production under Alternative New
Source Performance Standards (Electricity
Growth Rate of 5.8 Percent Per Year Until
1985 and 5.5 Percent Thereafter) 7-51
7-25 1990 Coal Distribution under the Current
New Source Performance Standard of 1.2
Ibs. of SO (High Electricity Growth Rate) 7-55
7-26 1990 Coal Distribution under an Alternative
New Source Performance Standard of 90
Percent Removal of SO. (High Electricity
Growth Rate) 7-56
7-27 1990 Ton-Miles of Coal Shipments under the
Current New Source Performance Standard
of 1.2 Ibs. of SO (High Electricity
Growth Rate) 7-58
7-28 1990 Ton-Miles of Coal Shipments under an
Alternative New Source Performance Standard
of 90 Percent Removal of SO (High
Electricity Growth Rate) 7-58
7-29 Coal Industry Employment (in Thousands of
Employees) Electricity Growth Rate of 5.8
Percent Per Year Until 1985 and 5.5 Percent
Thereafter 7-60
7-30 Estimated Regional Income Differential in
1990 Resulting from 90 Percent SO
Reduction 7-62
7-31 Required Capacity of Electric Power with
FGD Systems 7-66
7-32 Additional Income Resulting from Increased
Employment in Power Plant Construction
with FGD Systems Moderate Growth Rate of
Power 7-68
7-33 Additional Income Resulting from Increased
Employment in Power Plant Construction with
FGD Systems High Growth Rate of Power and
90% S02 Reduction 7-69
xvi
-------
LIST OF TABLES
(Continued)
Table Number
7-34 Industry Capacity to Meet FGD Requirements 7-70
7-35 Distribution of FGD Industry Employment Per
GW Required for Installation of Additional
FGD Equipment 7-71
7-36 Utility Oil and Gas Consumption 7-73
7-37 1990 Generation Capacity (GW) 7-75
7-38 The Share of Imports in U.S. Domestic
Petroleum Demand 7-76
7-39 Imports-Petroleum Products and Natural Gas 7-80
7-40 Estimated Increased Import Costs 7-81
7-41 U.S. Balance of Foreign Trade (1969-1973) 7-82
7-42 FGD Cost Effectiveness on a Unit Basis
Full Scrubbing 500 MW Plant 7-85
A-l Estimated Ca/S Mole Ratio to Achieve
Varying Sulfur Retention Levels A-2
A-2 Sorbent Requirement for AFBC to Meet EPA
S0? Emission Standards Based on Pilot
Plant Data A-8
A-3 Results of Runs at Turndown Conditions A-13
A-4 Sorbent Requirements for PFBC to Meet EPA
SO,., Emission Standards Based on Pilot
Plant Data A-15
B-l Process Design Bases for FGD Processes B-2
B-2 Design Assumptions for Physical Coal Cleaning
Facility B-3
B-3 Energy Requirements for the Processing
Operations in FGD Systems B-4
B-4 Total SO and Particulate Energy Penalty
Associated with Different Methods of
Controlling SO Emissions - 500 MW Plant,
3.5% Sulfur Coal B-9
D-l Plants Reporting 90 Percent or Greater SO
Removal D-28
xvi i
-------
Table Number
LIST OF TABLES
(Concluded)
G-l Flue Gas Desulfurization Units G-2
G-2 Analyses of Coals Used as the Cost
Estimating Basis G-4
G-3 Design Parameters for the FGD Systems G-5
J-l Key Scenario Elements Held Constant
Throughout the Analysis J-3
J-2 National Electricity Demand Growth Rates J-4
J-3 Alternative NSPS Scenarios J-6
J-4 Definition of Geographic Regions J-9
J-5 Utility Generating Capacity as of
December 31, 1975 J-10
J-6 Scaled Energy Demand Growth Rates, by Region J-ll
J-7 Projected Capacity Mix for Selected
Scenarios J-14
J-8 Projected Capacity Mix, by Region, for the
Baseline Scenario with Moderate Growth J-16
J-9 Projected Coal-Fired Capacity by Regulatory
Category J-20
J-10 Projected Coal Capacity Using Flue Gas
Desulfurization J-21
J-ll Regional Breakdown of Installed FGD
Capacity in 1995, Scenario HI.2(90)0.03 J-24
xviii
-------
1.0 INTRODUCTION
The present New Source Performance Standard (NSPS) limiting the
emissions of sulfur dioxide (802) from coal-fired electric utility
steam generators has been in effect since December 1971. Under this
standard, S02 may not exceed 520 nanograms per Joule (ng/J) heat
input (1.2 Ib/million Btu).
The U.S. Environmental Protection Agency (EPA) proposes to re-
vise the existing standard in view of technological advances in
controlling S02 emissions from coal-fired steam generators and the
requirements contained in the Clean Air Act Amendments of August 7,
1977. The revised standard is described in the preamble to the
regulation published in the Federal Register.
Environmental impacts associated with the proposed standard
for S02 emissions from coal-fired electric utility steam gene-
rators are described in this document. Revisions to the new source
performance standards for the emission of particulate matter and
oxides of nitrogen (NOX) from coal-fired electric utility steam
generators are also being proposed. These actions and their impacts
are the subjects of 3 separate volumes and their supplements.
Several alternatives, ranging from the retention of the present
standard to proposing a very stringent standard, have been consider-
ed by the EPA. The analysis includes several levels of fractional
reduction of S02- In addition, a control level of 220 ng/J (0.5
Ib 502/10^ Btu) has been considered. Discussion of these alter-
natives and their attendant impacts serves to put into perspective
1-1
-------
the beneficial and adverse effects expected to result from the
promulgation of the revised standard. Information to support the
Agency's analysis is derived from in-depth studies sponsored by the
Agency to examine the implications associated with each alternative
standard, including the proposed standard. This document summarizes
the results of those studies up to 15 February 1978 and subsequent
minor corrections. These studies are referenced throughout this
document. Developments which occurred between 15 February 1978 and
the date of proposal are discussed in the supplement.
1.1 Background
In accordance with the provisions of the Clean Air Act,* the
Administrator of the U.S. Environmental Protection Agency (referred
to as the Administrator in this document) is authorized to promul-
gate standards of performance for new stationary sources of air pol-
lutants. Under provisions of the Clean Air Act Amendments of 1970
the Administrator was to establish standards of performance that re-
flect "the degree of emission limitation achievable through the ap-
plication of the best system of emissions reduction which (taking
into account the cost of achieving such reduction) the Administrator
determines has been adequately demonstrated (42 U.S.C. 1857C-6)."
*The Clean Air Act (42 U.S.C. 1857 et seq.) includes the Clean Air
Act of 1963 (P.L. 88-206) and amendments made by the Motor Vehicle
Air Pollution Control ActP.L. 89-272 (20 October 1965), the Clean
Air Act Amendments of 1966P.L. 89-675 (15 October 1966), the Air
Quality Act of 1967P.L. 90-148 (21 November 1967), the Clean Air
Amendments of 1971P.L. 92-157 (18 November 1971) the Energy
Supply and Environmental Coordination Act of 1974P.L. 93-319 (24
June 1974, and the Clean Air Act Amendments of 1977P.L. 95-95 (7
August 1977).
1-2
-------
Under the Clean Air Amendments of 1977, the definition has expanded
the scope of a standard of performance and, with respect to any air
pollutant emitted from a category of fossil fueled stationary
sources to which the standard applies, and stipulate that the stan-
dard "reflect the degree of emission limitation and the percent-
age reduction achievable through application of the best techno-
logical system of continuous emission reduction which (taking into
consideration the cost of achieving such emission reduction, any
nonair quality health and environmental impact and energy require-
ments) the Administrator determines has been adequately demon-
strated (Section Ill(a)(1)(C))." In establishing a regulatory frac-
tional reduction in emissions resulting from the combustion of
fuels, the Administrator may credit "any cleaning of the fuel or re-
duction in the pollution characteristics of the fuel after extrac-
tion and prior to combustion."
1.1.1 Present Standard
On August 17, 1971 (36 FR 15704), the Administrator proposed
regulations establishing standards of performance for airborne
emissions from new stationary sources in several source categories.
Among these were standards for emissions of particulate matter,
sulfur dioxide (SC^), and nitrogen oxides (NOX) from fossil
fuel-fired steam generators. Interested parties were afforded an
opportunity to participate in the rule-making by submitting comments
1-3
-------
and private sectors. Following a review of the proposed regulations
and consideration of the comments, the regulations were revised and
promulgated on December 23, 1971 (36 FR 24876, 40 CFR 60).
The standards of performance established by the regulations
were based on field testing conducted by the Agency and its contrac-
tors and on data derived from various other sources, including the
available technical literature. In comments to the proposed regu-
lations, many questions were raised as to costs and demonstrated
capability of control systems to meet the standards. These comments
were given due consideration, and the Administrator judged that the
standards corresponded to levels of performance that could be met
with demonstrated control systems at reasonable costs.
The regulation pertaining to the emission of SC>2 required
that emissions not exceed 520 ng/J heat input (1.2 lb/ million Btu)
derived from solid fossil fuel (40 CFR 60.43(a)(2)).* In accor-
dance with definitions contained in the Act or established by
regulation, the standard applied to steam generating units of capa-
city greater than 73 MW heat input (250 million Btu/hour), the con-
struction or modification of which commenced after August 17, 1971.
The S02 standard required the application of control systems
on plants burning high sulfur coal; and allowed the use of coal of
*Periods of excess emissions are defined as any two consecutive
hourly periods during which average emissions of sulfur dioxide
exceed 1.2 pounds per million Btu of heat input (40 CFR 60.45(g)
(2)).
1-4
-------
sufficiently low sulfur content and high calorific value whose emis-
sions of sulfur dioxide remained below the regulatory standard.
1.1.2 Revisions of the Standard
Provisions of the Clean Air Act in effect before enactment of
the amendments of August 7, 1977 require the Administrator to review
and if necessary revise established standards of performance for new
sources as new knowledge and technology became available. On August
6, 1976, the Agency was petitioned by the Oljato and Red Mesa Chap-
ters of the Navajo Tribe and the Sierra Club to revise the standard
of performance for the emission of SC>2 from power plants and to
require a reduction in emissions of 90 percent from uncontrolled
levels. The petition included detailed information to support the
claim that a revision of the standard was necessary in view of re-
cent advances in control technology. The Agency agreed to inves-
tigate the matter thoroughly and gave notice of its intent to review
the new source performance standard for S02 (42 FR 5121; Appendix
B*). Interested persons were invited to participate in the Agency's
efforts by submitting written data, opinions or arguments.
On January 27, 1977, the Agency gave notice of a public
hearing to be held on May 25 and 26 (42 FR 18884; Appendix B). The
public hearing was held at the General Services Administration
Auditorium, in Washington, D.C. A panel was formed consisting of
*Appendix A is reserved for public comments on this environmental
statement.
1-5
-------
representatives of the Oljato and Red Mesa Chapters of the Navajo
Tribe, Sierra Club, the Illinois Pollution Control Board, the
Utilities Air Regulatory Group, and of the U.S. Environmental Prot-
ection Agency. The panel discussion was made part of the public
hearing. Oral presentations were made at the meeting by re-
presentatives of Federal, state, and local governmental agencies,
industry and commerce, citizens and industrial groups, and
individuals in the private sector. The record includes written
statements submitted prior to June 6, 1977, by all interested
parties.
Pursuant to the amendments of August 7, 1977 to the Clean Air
Act, the Administrator is now required to review and, if appro-
priate revise established new source performance standards at least
every 4 years (Section Ill(b)(1)(B)). Revised standards reflecting
a fractional reduction in emissions resulting from the combustion of
fuels are to be promulgated within 1 year of the enactment of the
amendments (i.e. by August 7, 1978).
Proposed revisions to the present standard for SC>2 emissions
from electric utility steam generators have been discussed at two
meetings of the National Air Pollution Control Techniques Advisory
Committee (NAPCTAC) held in Alexandria, Virginia, on September 28
and on December 14 and 15, 1977. Public comments in both oral and
written form are included in the records of these meetings.
1-6
-------
1.2 Statutory Authority
Authority to promulgate and revise standards of performance for
new sources is derived from Section III of the Clean Air Act. The
Administrator is directed to establish standards relating to the
emission of air pollutants from new stationary sources and is ac-
corded discretionary power to:
1. Identify categories of stationary sources that cause or
contribute significantly to air pollution, where it may
reasonably be anticipated that public health or welfare
would be endangered.
2. Distinguish among classes, types and sizes within cate-
gories of new sources for the purpose of establishing
standards.
3. With respect to any air pollutant emitted from a cate-
gory of fossil fuel fired stationary sources to which the
standard applies, to establish a standard of performance
that reflects the degree of emission limitation and the
percentage reduction achievable through application of the
best technological system of continuous emission reduction
which (taking into consideration the cost of achieving such
emission reduction, any nonair quality health and environ-
mental impact and energy requirements) the Administrator
determines has been adequately demonstrated.
The term "stationary source" encompasses all buildings,
structures, facilities or sources that emit or may emit any air
pollutant. A source is considered new if its construction or mod-
ification commences after the publication of proposed regulations.
Modifications subjecting an existing source (any stationary source
other than a new source) to regulatory standards are considered to
be physical changes in the stationary source or changes in the meth-
od of operation, provided that such changes lead to an increase in
the amount of any air pollutant not previously emitted.
1-7
-------
2.0 DESCRIPTION OF AND RATIONALE FOR THE PROPOSED ACTION
The material intended to be presented in this section, appears in
the preamble to the regulation published in the Federal Register.
2-1
-------
3.0 LEGAL ALTERNATIVES
Under the present provisions of the Clean Air Act incorporating
the amendments of August 7, 1977, the Administrator is directed to
review and, if appropriate, revise established standards governing
the release of airborne pollutants from new stationary sources (PL
95-95, Section III(b)(1)(B)). With respect to fossil fuel-fired
stationary sources to which a revised standard may apply, the
standard must:
1. Reflect a fractional reduction of emissions relative to the
corresponding emissions from the combustion of untreated
fuel (PL 95-95, Section Ill(a)(1)(A)(ii))
2. Reflect the application of the best technological system
of emission reduction (PL 95-95, Section Ill(a)(1)(C))
3. Take effect 1 year after the enactment of the amendments
of August 7, 1977 (PL 95-95, Section Ill(b)(1)(B)(6))
3.1 No Action
Under the Clean Air Act Amendments of 1977, no action is required
by the Agency in revising the new source performance standard for the
emission of sulfur dioxide from coal-fired electric utility steam
generators if the existing standard reflects the application of the
best technological system of emission reduction.
On the basis of the information and the analysis set forth in
this statement, the Administrator has determined that demonstrated
technology is available to limit the emissions of sulfur dioxide from
coal-fired electric utility steam generators to levels lower than
might be allowable under the present standard. Accordingly, the
3-1
-------
Administrator has the responsibility for revising the existing
standard and establishing one that reflects the application of the
best technological system of emission control.
3.2 Delayed Action
The Administrator has no discretionary power to delay the estab-
lishment of a revised standard.
3.3 Nature of Standard and Stringency of Controls
The Administrator has no discretionary power to establish a
revised standard other than one expressed in terms of a fractional
reduction in emission levels. A proposed fractional reduction of 85
percent, with added provisions for maximum allowed emissions and
maximum required levels of control, has been selected based on:
the availability of demonstrated technology to achieve this reduc-
tion, the properties of U.S. coal reserves, and the environmental
considerations documented in this report.
3.4 Control Practices
The proposed fractional reduction in SC>2 emissions can be
achieved by treatment of the flue gases from the combustion of coal
or by a combination of flue gas treatment and coal treatment to
reduce sulfur content prior to combustion. Technologies under
development, are the conversion of coal to clean fuels and the
combustion of coal in a fluidized bed containing sorbent material.
The proposed standard allows credit to be taken for any and all
3-2
-------
techniques applied to reduce the emission of sulfur dioxide and,
therefore, conforms with the provision of the Clean Air Act stipula-
ting that such credit be given (PL 95-95, Section Ill(a)(1)(O). No
alternative form of standard would comply simultaneously with this
and the other stipulations of the Act discussed above.
3-3
-------
4.0 ALTERNATIVE CONTROL TECHNOLOGIES
Sulfur is a natural constituent of practically all coal.
During coal combustion, most of its sulfur content is converted to
gaseous sulfur compounds. In the absence of emission control devices,
these compounds* and other products of combustion are released.
Methods of limiting the emission of sulfur compounds fall into four
broad categories:
Burning low-sulfur coal.
Cleaning coal before combustion to remove part of its sulfur
content.
Retaining sulfur during or immediately following combustion
in sorbent material mixed with the fuel coal.
Processing of flue gases (flue gas desulfurization). In
principle, combinations of two or more of these techniques
could be applied to achieve a given degree of sulfur reten-
tion.
4.1 Burning Low-Sulfur Coal
Burning of coal leads to the volatilization of its sulfur
content and, in the absence of control equipment, the release of
practically all of the sulfur from the combustion system. A small
fraction of the sulfur liberated from the coal, generally 5 percent
of the amount present, is retained within the ash or other deposits
in the system. The remainder is released predominantly as sulfur
dioxide, with a small amount of sulfur trioxide.
*As a general rule, 95 percent by weight of the sulfur present
in bituminous and subbituminous coal is released as gaseous sulfur
compounds from a utility boiler without sulfur control devices
(U.S. Environmental Protection Agency, 1974). Sulfur is released
predominantly as sulfur dioxide. A very small portion of the
emissions may consist of sulfur trioxide.
4-1
-------
Clearly, the quantity of sulfur dioxide released from a steam
generator burning coal varies with the sulfur content. The feed
rate of coal in a given system is governed by the capacity of the
system, its thermal efficiency, and the heat content or calorific
value of the coal. These parameters effectively determine the rate
at which sulfur dioxide is emitted from a given generator. More
specifically, the emission rate of sulfur dioxide can be expressed in
terms of the mass of sulfur dioxide released per unit of heat input
to the system or as the number of pounds of sulfur dioxide released
per million Btu of heat input. Both the sulfur content and the
heating value of the fuel, therefore, influence the normal release
rate of sulfur dioxide. These two properties determine whether the
burning of a particular coal meets regulatory limitations.
4.1.1 Availability of Acceptable Coals
Surveys of the U.S. coal reserve base show low sulfur coal
is present predominantly in the western states. However, most
western coals are of a lower rank than eastern coals and, when
allowance is made for differences in heating value, the estimated low
sulfur fraction of the reserve base in the west drops from 84 percent
to a maximum of 80 percent (U.S. Department of the Interior, Bureau
of Mines, 1976). With respect to rank, 22 percent of the coal with
a sulfur content lower than 1 percent is of high rank (anthracite
and bituminous) and 78 percent is of low rank (subbituminous and
lignite).
4-2
-------
States with the largest quantities of low sulfur coal are
Alaska, Montana, and West Virginia. Montana has an estimated reserve
base of 102 billion tons, or 51 percent of the total reserves; West
Virginia has 7 percent; and Alaska has 6 percent. Virtually all of
the Montana and Alaskan coals are of low rank; whereas all of the
West Virginia coals are high rank bituminous coals. Of the high rank
low sulfur coals 82 percent of the reserve base is amenable to
underground mining, while only 58 percent of the reserves of low rank
low sulfur coals could be recovered.
Certain low sulfur coals with low contents of mineral matter
(generally no greater than 8 percent) are suitable for coking and
subsequent use in metallurgical production and other industrial
processes (U.S. Department of the Interior, Bureau of Mines, 1976).
It is estimated that 90 percent of all coking coals are in the
Appalachian coal region. West Virginia has the largest quantities of
premium quality coking coal. Of the 14.1 billion ton reserve base of
low sulfur bituminous coal in West Virginia, 10.5 billion tons are
considered to be of premium grade suitable for use in coke production
(U.S. Department of the Interior, Bureau of Mines, 1975, 1976).
Other states with substantial deposits of coking coal are eastern
Kentucky and Pennsylvania.
Little information is available on the sulfur content and
heating value of coals distributed throughout the U.S. reserve base.
A sampling program to estimate the potential reduction in sulfur
4-3
-------
dioxide emissions that could be realized by washing U.S. coals
yielded data characterizing coals on the basis of emission rates of
sulfur dioxide per unit of heat input (U.S. Environmental Protection
Agency, 1976). The program analyzed 455 coal samples collected from
surface and deep mines currently producing coal primarily for use by
electric utilities. Samples included in the survey are drawn from mines
producing in aggregate more than 70 percent of the current total annual
consumption of the utility industry. The distribution of samples by
region and state is shown in Table 4-1. The number of samples varies
greatly among regions and states. Since the intent was to characterize
steam coal currently being produced, information for the eastern regions
is more precise than that for reserves in the western midwest and
western regions.
The results of the survey are combined with data on the recover-
able reserves of coal in the United States to yield the estimates
shown in Table 4-2. As indicated in the table, approximately 110
billion tons of coal or 42 percent of all recoverable reserves
could be burned without pretreatment or controls and without exceed-
ing the present regulatory standard of performance of 1.2 pounds of
sulfur dioxide per million Btu of heat input. Approximately 70
percent of the recoverable reserves in the western region meet this
criterion, 12 percent in the eastern region and 5 percent or less in
the eastern and western midwest regions. Reducing the standard to
4-4
-------
TABLE 4-1
DISTRIBUTION OF COAL SAMPLES BY REGION AND STATE
REGION
Eastern
Eastern Midwest
Western Midwest
Western
STATE
Alabama
Eastern Kentucky
Maryland
Ohio
Pennsylvania
Tennessee
West Virginia
Total
Illinois
Indiana
Western Kentucky
Total
Arkansas
Iowa
Kansas
Missouri
Oklahoma
Texas
Alaska
Arizona
Colorado
Montana
New Mexico
North Dakota
South Dakota
Utah
Washington
Wyoming
Total
Total
Total, All Regions
NUMBER OF SAMPLES
10
7
34
58
103
8
44
272
40
20
35
95
3
17
8
9
77
0
44
0
6
11
5
9
1
0
8
0
4
44
455
4-5
-------
CM
1
Sj"
W
tJ
5
H
>-i
PQ
O
W
N
CTERI
CHARA
j
<
0
o
3
P?
fa
o
RESERVES
w
hJ
3
W
^
CO
O
w
Pi
fa
c
CO
w
2
1 ]
H
CO
w
§
M
H
CO
£3
1
O
p
r-3
_3
0
Pi
H
23
Q
O
55
3
s
O
Pi
fa
DIOXIDE
pi
3
fa
H-]
O
CO
fa
o
w
^H
O
rH
CN
O
CO
PQ
r-3
sl-
O
p
H
PQ
vc
O
iH
CN
o
CO
pq
>J
^o
0
3
H
pq
^o
0
iH
CN
O
CO
pq
_]
oo
o
D
c-i
PQ
vO
0
rH
CN
O
to
PQ
,-J
CN
.
rH
CO CO
53 55
O O
M H
J
,-J fa
M O
a
TT f"
fa PS
O H
CO
B^ fc
Pi
CO CO
S 'Z
0 0
H H
_1
>J fa
g°
W
fa oi
o w
CO
6-S W
pi
CO CO
23 23
O O
M H
>J
1 rv ,
1 I 1^4
H O
S
W
% OF
RESERV
CO CO
S3 23
0 0
IH H
_J
»J fa
|_J / )
SB
W
>
fa Pi
o w
CO
r>S W
oi
CO
z
W O
s3 * H
PQ rn
00
m
CN
CN
o
o
o
r~
O
.H
m
0
m
VD
o
o
o
oo
o\
0
o
o
0
iH
o
i
O
o
o
oo
un
o
0
o
iH
m
o
0
o
m
tH
o
o
o
o
sr
rH
o
0
o
CN
vO
CN
M
0)
4-1
tn
d
tn
0)
c
S-i
0)
4-1
cn
ca
w
en
a;
13
s
S-J
01
en
OJ
C
M
0)
4-t
en
CO
H
O
H
C3
E-i
c?.
3
4-1
H
4-1
CD
C
II
o
0)
s
4-1
CO
M
O
i-i
M-l
OJ
4-1
Q,
0)
o
!-(
Q
to
4-6
-------
0.8 pounds would reduce the fraction of recoverable reserves that
could meet regulatory requirements to 2 percent or less in the
eastern, eastern midwest, and western midwest regions and 41 percent
in the western region. Nationwide, 58 billion tons or 22 percent of
recoverable reserves of coal would have the requisite properties with
respect to sulfur content and heating value. Further reductions of
the standard would eliminate practically all coals except those of
western origin. At a standard of 0.6 pounds, 16 percent of coals in
the western region, or 22 billion tons of reserves, would be accept-
able. At 0.4 pounds, 2 percent of western reserves, or 2.8 billion
tons, could meet regulatory requirements.
4.2 Fuel Treatment Processes
The effectiveness of fuel treatment processes as a sulfur
dioxide emission control technology depends on the cleanability of
the coal (the amount of sulfur that can be removed). Sulfur is
present in the coal as either inorganic sulfur, such as pyrite, or
organic sulfur. The principal categories of fuel treatment are
physical and chemical coal cleaning and the application of the
solvent refined coal (SRC) process.
The physical coal cleaning methods are the conventional tech-
nologies that are commonly applied; however, sulfur removal is
limited. These technologies are typically employed to remove car-
bonaceous shale, gypsum, Kaolin, calcite, and pyrite to produce
better quality fuel rather than strictly for sulfur removal.
4-7
-------
The chemical coal cleaning processes being developed remove
more of the pyritic sulfur and, in many cases, some of the organic
sulfur. The chemical methods or processes that are not limited in
terms of sulfur removal are currently in the developmental stages and
have produced little or no operating data.
The SRC process under development involves dissolving pulverized
raw coal in a coal-derived solvent. In the process, mineral matter
(ash) and pyrite and organic sulfur are removed, producing a liquid
which when cooled to ambient temperature becomes a solid material
that can be burned in modified pulverized-coal boiler.
Extensive studies show that fuel treatment processes result
in reduced sulfur content for many medium to high sulfur coals. The
reduction and total amount of coal that can be cleaned depends on
many variables which are discussed in the following sections.
4.2.1 Physical Coal Cleaning
Physical cleaning can be defined generally as the separation of
waste or unwanted "refuse" material from coal by techniques based on
the differences in the physical properties of coal and refuse. The
most common physical property used in coal cleaning is density.
Specific gravity ranges are generally:
Coal (1.2 to 1.8)
Carbonaceous shale ( 4)
Pyrite (5)
Gypsum, kaolin, calcite (2.3 to 2.9)
Density separation is done using hydraulic jigs, laundering tables,
cyclones, dense medium vessels, or air classifiers. In such equipment,
4-8
-------
ground coal is suspended in a fluid, the refuse material falls to
the bottom of the separating unit, and the cleaned coal floats or
moves to the top of the unit for removal. Froth flotation, a related
technique, also uses the surface properties of coal particles to
enhance separation. Physical cleaning removes mineral sulfur such
as pyrite, which has a high density, but not organic sulfur, which
is an integral part of the coal. The amount of mineral sulfur
removed depends on the crystal size; the smaller the crystals, the
smaller run of the mine (ROM) coal must be crushed to achieve effec-
tive separation. Large amounts of coal will be lost with the refuse
if the particle sizes of the mineral sulfur and pulverized coal are
not matched well and if a large fraction of the mineral sulfur is to
be removed. As the coal is pulverized to smaller and smaller par-
ticles, costs of pulverization rise quickly. These costs vary widely
depending on the type of coal.
The Btu recovery rate of the cleaning process is usually based
on the input heating value. The heating value of the coal lost
in the refuse is counted as an energy loss. Physical cleaning
generally has a Btu recovery of 80 to 95 percent of the ROM coal,
with the largest losses associated with coal lost with the refuse and
with the coal required to operate the thermal drier. One can expect
physical cleaning to remove 35 to 70 percent of the mineral sulfur
(inorganic) in ROM coals, depending on the amount of size reduction
and the other physical characteristics of the coal. However, while
4-9
-------
while physical coal cleaning removes much of the mineral sulfur, the
organic sulfur (not removed by physical cleaning) can make up 30 to
70 percent of the sulfur in a particular coal.
Several other techniques can be used in physical cleaning, such
as magnetic separation of iron pyrite (FeS^), oil agglomeration, and
electrophoretic and electrostatic separation. Either for economic or
processing reasons, they have not been developed sufficiently to war-
rant a detailed discussion.
Physical coal cleaning reduces sulfur and ash content. While
this is done primarily to improve fuel quality and consistency the
result enhances the environmental acceptablity of burning the cleaned
coal. However, physical cleaning has its own set of environmental
problems. The refuse is usually gob piled. These piles can be
a source of pollution similar to acid mine drainage and may require
a collection and lime treatment system for the drainage. Gob piles
can also be sources of fugitive dust. Table 4-3 gives generalized
potential environmental problems associated with the various process
technologies.
Physical coal cleaning reduces the sulfur and ash contents and,
hence, reduces costs of transportation and particulate removal and ash
handling at the power plant. In addition, SO and particulate emis-
sions are reduced. Process reliability is a minor problem, since
most of the systems have been used in the mining industry for years.
However, for certain coals, the reduction of sulfur content is limited
4-10
-------
W
KJ
PQ
<
H
O
Pi
PM
W
s
S3
O
W
CO
w
w
L3
O
ftf
(X
w
J
O
rJ
*
P-,
Q
O
33
H
W
S
II
O
03
H
^
O
O
^
w
rJ
PQ
o
D^
PM
^
o
o
rJ
O
a
pa
u
w
H
CD
i l
CJ
^
o
^_l
OJ
4J
CO
£5
T)
0)
CO
0
O
rJ
0)
4-1
CO
£&
CO
co
(LI
CJ
O
l-i
(X
-0
0)
4-1
CO
c
H
E
CO
4-J
C
o
O
-i
^
l-i
-o
c
3
CO
i 1
\
60
i-l
>~)
Q)
, 1
o
cfl
4-1
*
01
c
0
t 1
CJ
^
a
0)
CO
C
0)
^
0)
4-1
Cfl
^
cn
co
0)
U
O
l-i
a.
T3
0)
4-1
cfl
C
H
E
td
-U
C
O
, 1
0)
CO
CO
0)
p>
E
3
H
-o
01
E
0)
co
C
0)
Q
0)
i l
O
£*"}
o
r-l
0)
4-1
CO
U
T3
01
CO
0
f^.j
O
cn
CO
O
r-l
CO
r-l
-a
0)
E
o>
C
o
I 1
u
£**
o
-O
c
cfl
01 r-l
CD O
3 4J
O cfl
XI 4-1
H
60 CL
cfl -H
XI O
0)
i-i a
o
4-1 O
O -H
01 4-1
i 1 Cfl
i 1 4-1
O CO
0 O
^1
0) 4-1 ^
C O p..
O O! 03
I 1 rH W
CJ Ol ^^^
0
4J
CO
p
"O
r-H
cfl
O
CJ
O)
£>
l-l
4-1
H
60
3
ta
v^
0)
H
M-l
i-l
CD
CD
Cfl
i 1
CJ
r-l
l-l
1 1
w
4-1
H
3
o
M
r-l
O
r-l
01
4-J
Cfl
d
01
CO
o
0
H
a)
4-1
cfl
^
co
CO
Q)
CJ
O
r-l
P-
T-J
0)
4-1
CO
c
r-l
e
CO
4-1
O
CJ
*,_4
4-1
Ol
O
XI
a.
o
4-J
CJ
a>
r 1
W
4-J
H
3
CJ
r-l
l-l
O
l-l
0)
4-1
cfl
^
13
0)
CO
O
U
^
0)
4J
CO
^
cn
CO
a>
CJ
O
l-i
cx
T3
01
4-1
CO
C
H
E
CO
4-1
d
o
o
o
4-1
O)
C
60
CO
s
4-J
r-l
3
U
r-l
r-l
CJ
r-l
H
O
T3
C
cfl
j_,
0)
4-1
cfl
T3
-------
because the organic sulfur cannot be removed. As a result, not
all coals can be cleaned to meet the new source performance standard
(NSPS), and as it is lowered, the applicability of physical coal
cleaning diminishes. Table 4~4 shows the effect of changes in
the standard on the cleanable coal reserve. These figures are also
optimistic, since it is assumed that the utilities have access to 100
percent of the coal (typically a portion of the more desirable
coals, i.e., metallurgical coals, goes to industry at premium prices)
and that production can be shifted to produce this coal. However,
physical coal cleaning might be used with another control option
such as flue gas desulfurization (FGD). If physical coal cleaning, a
relatively low-cost process, is used to reduce the sulfur content to
near the NSPS compliance level, then FGD, a relatively high-cost
process, can be used to treat a portion of the flue gas stream to
achieve NSPS compliance. The size of the FGD unit and, hence, its
cost would be reduced,. A further benefit would accrue from using FGD
on only a portion of the flue gas if the recombined treated and
untreated flue gas streams retain sufficient buoyancy so that reheat-
ing is not required to achieve plume rise.
The economics of combined Physical Coal Cleaning (PCC) and FGD
have been analyzed by Hoffman-Munter Corporation in a study for the
Bureau of Mines, and by PEDCo Environmental Incorporated in a study
for the U.S. Environmental Protection Agency. These studies show that
a lower cost can be expected by using the combined technologies if the
4-12
-------
CO
W
M
H
M
,-4
h- 1
H
** l"~*
CO
P-l CJ
CO rH
Z PC1
W H
W PS O
re H w
EH («J x x
H W »
H W H
W W O PQ
w si z
5S II ^o
OHO
O H CO rH
H M -~~
C_> Q
W ftS Z PQ
iJ [d Q < ,-4
M CO W
< W M S 1
H pti <2 W
PU Z Q
W W ptj
iJ Dd g <3
PQ PJ O Q
5 0 CO
O Z ^
O 13 o^ oo
1~4 CO i-H
»\ % M
m o oo
i ( f i
CN >
M rH
H rH
CO tfl
M 3
X C
W C
CO to
H C
M O
Z H
iJ O
< CO
t-I W
P_l Qj
r= w
co co ^o cj> in
W -d- r~ o
W Oi rH
m w
)J PQ
M w
o
fn U
o w
CO
as g
tfl
0) -i-l
rH M
O
0 0
C a)
cd S
e
a; a;
T3 i~~*
rH
m D
OO 4J
t i ni
CQ
C
O
T3 Q)
0) O
CO M
cti 3
PQ O
* CO
4-13
-------
sulfur content of the cleaned coal is near compliance levels. As
the difference between sulfur content and compliance level increases,
FGD must be used on a greater percentage of the flue gas. At some
point, a crossover occurs where it is no longer economical to use PCC
in conjunction with FGD. This is because scrubbers have a size/cost
exponential relationship of 0.8. As the scrubber size becomes
larger, economics of scale make it more attractive to totally utilize
scrubbers as the cleaning system and dispense with the coal cleaning.
In addition, as more of the flue gas is scrubbed, reheating becomes
necessary and the benefit of the combined technologies also di-
minishes. It is difficult to generalize, because the necessary
analysis of costs and effectiveness must be performed specifically
for each power plant, but if more than 50 percent of the flue gas
must be scrubbed to achieve compliance, it is likely that combined
PCC and FGD will not be the lowest cost option. In conclusion, the
more stringent the NSPS, the less useful physical coal cleaning
becomes as an alternative control option.
Battelle Memorial Institute (1977) has studied the costs of
physically cleaning easily cleaned northern Appalachian coals, present-
ing cost data for coals cleaned at a top size of 0.95 cm (3/8 inch), and
high yield factors (a range of 85 to 95 percent of input product yield,
weight basis). Other Appalachian coals generally have a 60 to 70
percent weight yield (Battelle Memorial Institute, 1977), and the
associated costs would be higher on a cleaned-coal basis. Capital
4-14
-------
investment for a physical cleaning plant larger than 454 kkg (500
tons) per hour capacity at the mine mouth (a lower practical economic
limit) can cost between $9,920 and $49,600 per kkg ($9,000 and
$45,000 per short ton) per hour capacity (Battelle Memorial Insti-
tute, 1977). The higher value, $49,600, includes rail spurs, and
coal handling equipment normally associated with mine facilities
costs. The mean cost range is $16,500 to $19,800 per kkg ($15,000 to
$18,000 per short ton) per hour capacity. These mean costs are
incremental to mine facility costs, e.g., rail spurs, conveyors.
Assuming the following:
(1) 15-year capital write-off
(2) 13 productive hours per day, 260 days per year operation
(3) interest rate of 10 percent
(4) 90 percent product yield
(5) $19,800 per ton per hour of capacity capital cost
one can expect a capital charge of $.845 per kkg ($.767 per ton)
of ROM coal processed for a 454 kkg (500 short tons) per hour plant,
and an operating and maintenance cost of $.76 to $.94 per kkg ($.65
to $.80 per short ton) of ROM coal processed, depending on the site
and coal specifics of the cleaning plant. The operating and mainten-
ance cost includes an allowance for disposal costs of the refuse.
Because of the loss of rejects material in cleaning, and because the
heating value is an important factor in selling the cleaned coal,
costs are usually reported in dollars per million Btu. If the coal
4-15
-------
is assumed to go from a ROM heating value of 25.58 MJ per kg (11,000
Btu per pound) to a product heating value of 27.91 MJ per kg (12,000
Btu per pound), with a ROM coal price of $19.80 per kkg ($18 per
ton), and a 90 percent weight yield, the cost of cleaning would be
calculated as follows:
$19.80 per kkg
Raw Coal Cost = (25.58 MJ/kg) (1000 kg/kkg) = $'774/GJ
Cleaned Coal Cost = $19.80 ROM coal cost
.84 capital charge
.94 O&M cost
$21.58 per kkg ROM coal
$21.58
.9 yield
= $23.98/kkg cleaned coal
$23.98/kkg
_
(27.91 MJ/kg) (1000 kg/kkg)
Cleaning cost = $.859 - $.774 = $.085/GJ or $.09/10 6Btu
$ 19 57
'-r- u = $21.74/ton cleaned coal
.9 yield
$21.74/ton (10 ) >, nn,,,nf> -. , -.
= $.906/10 Btu cleaned coal
(12,000 Btu/lb) (1000 Ib/ton)
Cleaning cost = $.906 - $.818 = $.087/10 Btu
4-16
-------
This cost is for a plant using hydraulic jigs, washing tables,
cyclones, froth flotation units, filters, screens, and mechanical
and thermal driers. Using the cleaned coal as a basis, the cleaning
cost is then $2.37 per kkg (2.09 per ton). If the cleaning yield is
assumed to be a more typical 60 weight percent, the ROM heating
value of the coal is 18.61 MJ per kg (8,000 Btu per pound), and a ROM
coal price of $11 per kkg ($10 per ton) is used, the capital charges
and operating and maintenance costs used above then give a cleaned
coal processing cost of $4.80 per product kkg ($4.27 per product
short ton), and $.172 per GJ ($0.178 per million Btu's). These
figures do not include any profit for the operation.
4.2.2 Chemical Coal Cleaning
Chemical coal cleaning has an advantage over physical methods
in that it has the potential for removal of nearly all of the inor-
ganic sulfur, and some of the organic sulfur as well. However, most
of the systems are in the development or pilot stages and have not
yet been totally demonstrated. As a result, no acceptable reliability
data are available. Further, some of the processes have environmental
problems which are difficult to resolve.
There are currently about 25 chemical cleaning processes under
active development and many more in conceptual stages. There is
economic information available for eight of these processes.
4.2.2.1 Meyers/TRW Process. The Meyers process is the most
highly developed of the chemical cleaning processes. This process
4-17
-------
leaches - 149 )j.m (-100 mesh) coal containing iron pyrite (FeS2)
with ferric sulfate Fe2(SO^)3, converting the pyrite to sulfuric
acid, ferrous sulfate, and elemental sulfur. The process operates
2
at moderate temperatures and pressure, 70°C to 120°C and 100 Kn/m
to 550 Kn/m (15 to 80 psia). Leaching times are 5 to 10 hours.
The process has no proven organic sulfur removal. Elemental sulfur
produced is solvent extracted or vaporized and recovered by conden-
sation. Figure 4-1 indicates the layout involved in the Meyers/TRW
process .
Dow Chemical has performed an extensive design and economics
study of this process for a 420 kkg (380 short tons) per hour plant.
Their total capital cost for this design was $145 million (mid-1975
dollars) plus or minus about 20 percent. This includes limited
physical cleaning facilities for removal of rock aggregate and shale.
Dow feels that based on this design and 95 percent removal, of pyritic
sulfur, a cleaning cost of $11 to $15.50 per kkg ($10 to $14 per ton)
of cleaned coal would be appropriate currently. Bechtel Corporation
has studied the economics of a 300 kkg (330 short ton) per hour plant
suggesting a total capital cost of $131 million and a cleaning cost
of $.78 per GJ ($0.82 per million Btu's), or $20.90 per cleaned kkg
($19 per cleaned short ton). The costs of both companies contain
no profit margins, and Dow's cost is based on cleaning a Pennsylvania
Lower Kittanning coal. Bechtel "s design is based on using a Pitts-
burgh A bituminous coal. Dow indicates that, based on their design,
4-18
-------
W
O
M
Pn
-------
the process can achieve a 90 percent Btu recovery, while Bechtel
indicates 98 percent Btu recovery.
The Meyers process is one of the more troublesome chemical
cleaning processes from an environmental standpoint. It uses organic
solvents in contact with process wastes to extract the elemental
sulfur. A portion of the solvent is left in the cleaned coal.
The waste products of the processferrous sulfate, sulfuric acid,
and physical cleaning refusehave to be disposed of properly with
pH adjustment. This refuse is obviously much more acidic than
just physical cleaning refuse alone. Internally, the process must
use a closed water circuit with solvent recovery to avoid further
effluent problems. The Meyers process probably could be commercial
in 5 to 6 years. An 8 ton per day pilot plant is currently being
built, which should provide scale-up information.
4.2.2.2 Battelle Hydrothermal. The Battelle process (Battelle
Memorial Institute, 1977) leaches - 149 m + 74 m (-100 + 200 mesh)
coal with sodium and calicum hydroxide solutions at elevated tempera-
tures and pressures, 98°C to 170°C (200°F to 340°F) and 1.55 to
17.25 MN/m2 (225 to 2500 psia). The process removes up to 99
percent of the mineral sulfur and has demonstrated 24 to 72 percent
organic sulfur removal, depending on the specific coal processed.
Btu recovery ranges from 75 to 90 percent, depending on process
operation. Figure 4-2 indicates the process layout and unit opera-
tions. The capital cost of the process suffers due to the elevated
4-20
-------
CLEAN
" -
OA
NER
0 i:
U
_j
<
O
O
_j
ELEMENT
CO
.j
c
»
L
1
*
> u
tt °
Q
_J
<
O
U
*
1
>
j
CC
LU
H
"^ *""
"^""""N ^
^y
/\"^ CO
2
£ /f
i^
2LI >
^^§
m ^> "
^ W LU
J fN*
in c
t
w
Ztf /
A
1
I CO
^ '
LU G_D " '
= r
PO
O
o
1 .
if _ra
|'\- ,
l^/
^
LU
C.
I CMC-
1
O
H =
+ -^ LU "* "
silf
0 9 m
<
CJ
g
o
*i
LU
U
w
o
O
CO
w
u
o
&,
p-l
w
w
H
i r-^aTVTr-^ iT-gi-'iFJ"-T'^TTjr'^ V
V?
3?t4
L__
4-21
-------
temperatures and pressures used in the system, and the need for
leachant regeneration equipment to close the process water loop,
preventing the loss of leachant.
Battelle currently feels that an operating cost of $19.80 to
$27.50 per kkg ($18 to $25 per short ton) of cleaned coal or about
$.95 per GJ ($1.00 per million Btu) is a good estimate (Battelle
Memorial Institute, 1977) based on the regeneration of leachant,
0.25 hour leaching time, and processing a lower Kittanning coal from
2.4 to 0.9 percent sulfur. Under these conditions, a capital cost of
$134 to $145 million has been estimated for a 360 kkg (400 short
tons) per hour plant, the cost depending on the coal to leachant
ratio (2 to 1, or 3 to 1). No profit margin is included in these
figures.
With leachant regeneration, internal process water loops are
closed, so that the only water effluent is in the wet coal. Hydrogen
sulfide (l^S) is produced in the process, and protection against H~S
leakage would be necessary both from a processing and a safety view-
point. The process is known to leach out many heavy metals in coal.
Any effluents containing high concentrations of these metals may
require special disposal. The Battelle hydrothermal process could be
commercialized in 4 to 6 years.
4.2.2.3 Hazen Process. The Hazen process, shown in Figure 4-3,
is a totally dry process. The process reacts iron pyrite with
gaseous iron pentacarbonyl:
4-22
-------
*
S. x"
n) [c
a:
"N 9 0
0 m5
«*X K 2 ^
L-T g<
O
O
,-1
CO
w
u
o
N
4-23
-------
3 FeS2 + Fe(CO)5 -2 Fe2S3 + 5 CO
The Fe2S3 is much more magnetically susceptible, enabling it to be
magnetically separated from the coal. Thus, this process can remove
only mineral sulfur, and it requires very fine grinding of the coal
to liberate the pyrite particles. This factor may restrict appli-
cation of the Hazen process. The process is simpler than others,
and uses fewer unit operations and process steps at mild temperatures
and pressures. The process does have severe process monitoring
requirements due to the use of highly toxic iron pentacarbonyl.
Results reported to date have been limited to coal ground to
1.19 mm (14 mesh) because there are no magnetic separators available
to handle dry, fine-pulverized materials. Thus development of the
process will be hindered.
Bechtel has estimated costs of a 300 kkg (330 short ton) per
hour plant for a Pittsburgh bituminous coal at a capital cost of
$48 million, and operating and maintenance costs of about $15.40 per
kkg ($14 per short ton) cleaned. They indicate a cleaning cost of
$.57 per GJ ($.60 per million Btu) , with a Btu recovery of 76 percent.
Few aspects of Bechtel1s design are specified; one is the Fe(CO)r
cost. Hazen estimates its cost at $.10 per pound with a consumption
of 32 pounds per ton of coal (whether ROM or cleaned is not specified).
Private vendor prices for Fe(CO)r run as high as $3.30 per kkg ($1.50
per pound). This higher price changes the cleaning costs dramatically.
4-24
-------
Along with monitoring FetCO);. levels in the plant area, the
disposal of the refuse will be of environmental concern. Problems
will be similar to those of refuse from physical cleaning, except
that Hazen refuse will create severe dusting problems because of its
small particle size.
Hazen is considering a 0.9 kkg (1 ton) per day plant, so commer-
cialization might be in 6 to 8 years.
KVB Process. This process shown in Figure 4-4, oxidizes sulfur
components of dry pulverized -1.19 |am + 595 |j.in (-14 + 28 mesh) coal
with NC>2 followed by caustic leaching to solubilize and remove the
sulfur compounds formed in the oxidation step. The soluble sulfur
compounds are mixed with lime to regenerate caustic and precipitate
gypsum (CaSO/), and iron oxides, which would be landfilled. The
advantages of the KVB process are its claim to removal with oxidation;
87 percent with additional caustic leaching, the simplicity and low
costs of dry oxidation; and the moderate temperatures, pressures, and
vessel residence times. A problem in the system is the uptake of
nitrogen by the coal.
Bechtel has developed cost information on the KVB process, based
on the KVB patent and limited nonproprietary information (no litera-
ture is available and little bench-scale work has been done). Bechtel
indicates a capital cost of $68 million for a 300 kkg (330 short tons)
per hour plant with an operating and maintenance cost of $25 per kkg
of cleaned coal ($23 per cleaned short ton). They indicate a cost of
' 4-25
-------
w
«
8
g
CO
00
w
CJ
O
4-26
-------
$.93 per GJ ($.98 per million Btu) , for a Pittsburgh bituminous coal,
with 90 percent Btu recovery.
Environmentally the KVB process poses one major problem; it is an
NO producer. No information is available on expected effluent levels
X
of NO . The other waste product is gypsum, for which established
disposal technologies are available.
Ledgemont Oxygen Leaching (LOL). The LOL process (see Figure 4-5)
is based on the following reaction:
FeS2 + H20 + 3.502 FeS04 + H2S04
High temperatures and pressures must be used to speed the reaction
rate for a commercially viable process. Strong oxidizing conditions
in the reactor cause some coal loss and volatization in the reactor.
This results in loss of heating value. The process has no significant
organic sulfur removal capability. Sulfur is removed from the system
by mixing the reaction products with lime, producing gypsum and iron
oxides which would be landfilled. Kennecott Copper Company claims
95 percent pyritic sulfur removal in the LOL process, with 93 percent
Btu recovery.
Dynatech and Bechtel have studied the economics of the LOL
process. Dynatech1s study gives an operating cost of $7.60 per
kkg ($6.90 per short ton) cleaned, but no capital costs. Bechtel1s
study shows a capital cost of $155 million for a 300 kkg (330 short
tons) per hour plant with an operating cost of $20.90 per kkg ($19
per cleaned short ton) or $.77 per GJ (.81 per million Btu). Dynatech
4-27
-------
i
-------
does not indicate what coals were used as a design base, or what type
of preparation facilities were included in the cost case. Bechtel
indicates a Pittsburgh A bituminous coal pulverized to 80 percent
minus 74 |j.m (200 mesh).
Bureau of Mines/ERDA. This process (see Figure 4-6) uses wet
oxidation, employing air instead of oxygen as used by LOL. The
process operates at higher temperatures and pressures than LOL,
generating iron sulfates and sulfuric acid. Because of the extreme
operating conditions, both pyritic and organic sulfur removal are
claimed, and the process can be expected to show coal loss similar
to the LOL process. Lime is used to convert iron sulfates to iron
oxides and gypsum.
Bechtel has studied the economics of this process using a
Pittsburgh bituminous coal. With pulverization facilities, grinding
to 80 percent minus 74 fj.m (200 mesh), Bechtel estimates a capital
cost of $130 million and an operating and maintenance cost of $20.90
per kkg of cleaned coal ($19 per cleaned short ton), or $.80 per GJ
($.84 per million Btu) with a Btu recovery of 94 percent. These
costs are for a 300 kkg (330 short tons) per hour plant.
Environmentally, the process will be very similar to the LOL
process. The process is under bench-scale development, so commer-
cialization would be in about 6 to 9 years.
Dynatech Process. This process uses microbial action at 38°C
(100°F) and 1 atmosphere pressure. There is little information but
4-29
-------
Z M
3
O
i-3
a
'8
ca
s
o
CO
4-30
-------
Dynatech does indicate using minus 74 fim (200 mesh) washed coal;
complete pyritic and some organic (amount unknown) removal; and
gypsum, sulfuric acid, and elemental sulfur products. Dynatech has
released limited cost data for a 300 kkg (330 ton) per hour plant
with coal preparation facilities, indicating a cost of $4.15 per kkg
($4.05 per ton) of cleaned coal. Other details are not available.
General Electric Process. GE is developing a process that
radiates coal with microwaves, gasifying the sulfur. Information is
limited, but GE claims 52 percent reduction in pyritic and organic
sulfur, and the possibility of reducing sulfur in most coals to
0.7 percent. Products of the process are H2S; COS, S02; H20, C02
and traces of CH/ , C2H,-, and H~. GE's preliminary cost data for
a 440 kkg (400 ton) plant claims a cost of $7.30 per kkg of cleaned
coal ($6.60 per cleaned ton).
4.2.3 Solvent Refined Coal Process
Development of the SRC process originated in Germany prior
to World War II, and was based upon the research of two German
scientists, Pott and Brocke, who patented the basic process in 1932.
Further development work was conducted in the U.S. from 1962 to 1965
by the Spencer Chemical Company, sponsored by the Office of Coal
Research (OCR), Department of the Interior. The Spencer Chemical
Company was subsequently acquired by Gulf Oil Corporation and devel-
opment activity was continued by the Pittsburg and Midway Coal Mining
Company, another Gulf subsidiary. A 50-ton per day SRC pilot plant
4-31
-------
is currently in operation at Ft. Lewis, Washington, under the
sponsorship of the U.S. Department of Energy; and under the sponsor-
ship of the electric utility industry, a 6-ton per day pilot plant is
in operation on the site of an Alabama Power Company steam plant near
Wilsonville, Alabama.
The SRC process involves dissolving of pulverized raw coal in
a coal-derived solvent in a hydrogen atmosphere at elevated tempera-
ture and pressure, as shown in Figure 4-7. In the process, pulver-
ized feed coal is first slurried with two or three parts of a solvent
fraction that is generated internally in the process. This recycled
solvent fraction has a boiling range of about 177°C (350°F) to Hydro-
gen is then added to the slurry of coal and solvent, and the mixture
is preheated and transferred to a single-stage reactor or dissolver.
In the reactor, the temperature Ls raised to 427°C (800°F) to 468°C
(875°F)pressure is elevated to approximately 1700 psig. Under these
conditions of temperature and pressure, approximately 93 percent of
the carbonaceous material in the coal is dissolved during a residence
time of approximately 30 minutes in the reactor. Approximately 60
percent of the organic: sulfur in the coal is converted to hydrogen
sulfide in the reactor, with hydrogen consumption of about 2 to 3
percent of the weight of the coal processed. The effluent from the
reactor is then passed to a high pressure separator where the liquid
and gas phases are separated. The liquid from the separator is next
subjected to a mineral separator step where undissolved solids are
4-32
-------
Pd
ftf
D
O
w
o
o
M
P-i
i-J
<
O
O
a
M
w
o
CO
4-33
-------
removed by filtration. The filtrate, from which the solids have
been removed, is then flashed in a vacuum distillation column.
Process solvent is recovered from the column and recycled to slurry
the coal feed. The bottoms from the vacuum column form the solvent
refined coal product which solidifies when cooled to about 177°C
(350°F). The solid filter residue from the mineral separation step
contains a substantial quantity of wash solvent as an absorbed
liquid. This solvent is removed by passage of the residue through a
rotary dryer where the solvent is extracted in vapor form, condensed
to a liquid, and recycled to the process. The solid residue from the
rotary dryer contains most of the mineral matter and some undissolved
carbon. When the process is operated on a commercial scale, it is
expected that this residue will be fed to a gasifier to produce
hydrogen required in the process.
Evaluation of the SRC process at the Ft. Lewis, Washington
facility has focused primarily on process-variable tests to generate
information needed to establish optimal coal-processing conditions.
Tests at this facility have utilized a Kentucky coal from Pittsburg
and Midway's Colonial Mine (a mixture obtained from the No. 9 and No.
14 seams). SRC material produced at the facility has been stockpiled
for use in burn tests at a 22-MW boiler at the Mitchell Plant Station
of the Georgia Power Company.
Five major U.S. coals have been tested at the Wilsonville,
Alabama, facility. The analyses of the coals tested is shown in
4-34
-------
Table 4-5. The operating conditions used in the tests and the
results produced (in terms of yield and sulfur content) are shown in
Table 4-6. The typical analyses of the material produced by the SRC
process with these five coals ^re shown in Table 4-7.
Extensive pulverization and combustion tests have been conducted
on the product from the SRC process to provide information on utili-
zation as a fuel in electric utility systems. In general, the
material can be easily pulverized with minor modification required to
the ball- and race or bowl-mill machinery commonly used in pulverized-
fuel boilers. However, the pulverized SRC material has the unusual
property of tending to agglomerate. Accordingly, it is necessary to
redesign most burner systems (normally with water jacketing) to keep
the temperature of the pulverized SRC at about 66°C (150°F) despite
boiler windbox temperatures of 260°C (500°F) to 316°C (600°F).
Additional burner changes involve the use of a venturi to control
fuel flow and stabilize the burner flame (as opposed to conventional
nozzle impellers which would be easily fouled by the SRC material).
In terms of burning characteristics, the SRC material ignites
like an oil, but requires a longer burnout time (similar to an
anthracite). In some tests, an improved boiler heat transfer rate
has been observed when SRC material is utilized over that attained
when coal is fired. (This is possibly attributable to the fact that
SRC combustion does not produce a slag layer on furnace surfaces
because of its low ash content.)
4-35
-------
O
-a-
w
o
w
r )
r-J
CO
i-j
<
a
f*
o
cn
w
KH
] 1
S
<3
, i
6
i i
P^J
H
CU
O
t-
£J
0
oo
E
(/>
cn i
C "O
r- C
E f0
0 r-
>> O
r
O
Cl /-V
Nt-J
f «
O
r
{"^
r- _C
CT) CT>
t- s-
i- 3
in
4-> 4->
to 4->
CD !-
3 0.
r
0
c to
(
r- 0
>-H 2^
^j.
r-H
o
J*1 C
O
3 CT>
4->
C
CD O
*^ 2C
CU
+-> E
CD
to t/>
^_
£
CD
r
r
CD
CO
CU
4.
CD
4->
C
0
CD
O)
T3
£
CD
>
0
_i
CM
O
z.
t_
it)
to
cn
C
C
i.
3
CO
r-
C
O
t
o
o
CD
C
r
z:
,
ro
O
t i
X
03
£^
QJ
-t-> r
C ro
0 0
s: o
c
0
4_>
rO
-o
r- O
O
01 i
C ro
0 0
c_> c_>
C
0
f
4_>
ft3
T3
i- O
I C_>
o
1/1 1
C ro
0 0
C_J C_)
>,
r- O
:E c_>
ca cn
C"
OV--
J- C
jQ £
in
!-» i
4^ (T3
r- O
Q. 0
^>l
C
ro
Q.
e
o
w
s
«\
. ^
l/l
tn
ro
^
i_
"^3
-
in
ra
C
c£
Ci>
4-J
ro
^E
X
o
S-
0.
to r^
T-H 0
* CO
in ro
*^" ri
o co
«d- in
*xt" "^
CO «3"
cn r~-
cn i i
ro in
p*^ ro
CM Ol
CO O
ro in
ro «a-
0 0
CO ro
ro in
OJ
^j
4-> C
ro O
>£! r~^
i.
CU ro
r O
«r-
*J T3
rtl CD
r- X
O -r-
^« LL.
r-^ ro
r~^ to
in c\j
> ( CM
o to
r-H
CO 0
CO ID
CO tO
O O
CO CO
o m
i i
ro I-H
cn CM
CO «*
a>
^_
^3
4->
to
^1 !-
1/1 O
< 2:
^^
0
ro
,_"
i i
J£J
CNJ
CM
i 4
to
IT)
CO
ro
i i
CM
O
to
CM
i r
CD
to
O
CO
I-H
^5
-Q 4-*
S
3 -
CQ r-
in
CO i
3 ro
C
ro «i
CO
CT> -!->
C ro
e
ro +->
CD ^~
m *
cn cn
^"*- *-^t*
to
o cn
co ro
^j- in
to
rH CM
r^ I-H
in in
r^
^- to
cn O
o in
(^
^j1 r^-
cn CM
CM in
r^
C
CD
C CJ>
0 0
.a t-
?*- *T3
ro >>
C_> Z
to
0
< 1
O
CM
,_<
CM
ro
1 1
in
CM
,_<
=*
i i
< i
C
CD
cn
O
l_
4-^
r«
z
< i CM r~^ in
O to r-~ to
O O r~- r^
i i
r^- in in «^-
O CO I I
O «* O *d~
T-H 1 t
r-i ro co ro
i < UD CM CO
O CM CO tO
cn ro o ro
O CM CO tO
O ro CD co
r-H
ro P1^ ro CM
i o cn in
O co co co
4J
^
to
O
co u.
c
r- S_ C i-
V. 3 CD 3
O v cn 4-
i ^- -C >, r
^: 3 to x 3
o t/i
*
A
_p^
t/)
*i
U U C
^_> fQ C"
r- if- ro C
i r cn o
>> 3 S_ fc.
Q.
-------
00
01 i
c -o
-C
£ re
O r
>i O
CO
w
CO
2
O
CO CO QJ
S- t- O)
r- 3 TD
> -0 -r-
l/> 1-
+-> -M QJ
>
O
O un o
O c\j o
LO o
r-v O LO
CO CD LO
r-. co
CO
CM
CM
I OO
QJ T-
0.
w
Q
2
O
O
o
2
ii
H
3
w
DH
O
t-
C O
c re r
QJ o
M CT> O
O <^ O
O CM O
to o o
co o co
r^ CO
CM
CTl
CVJ
VD
00
re
o
o
<+-
o
>>x:
C«t-
J- o
(/>
* to
U. E QJ
re oj c
4-> OJ -r-
oo oo z:
o
o
a) o: -r-
o u.
OJ TO C
,
LJL. ft^
)^a
' T^l
i-f
tt rt)
ti_ O E
00 3
. _. V^> f/«
^^ CT^- VI
Q. CU * O
S*. d t_J r"~"
*^ J
QJ 4-J
i- re oo
~> < J- I
-j >-- ^ '
to dJ i
00 Q. 3
QJ E oo
i- QJ s^ <:
c T3 z:
0 >>
0 f
re
o
CJ
u_
g
<4-
O
V5
Ol
t3
QJ
f
>-
r- O
re o u.
O LOO
o oo o
o
O co 3: co
OJ -O O 1
c - > o
r- 3 QJ IT)
M- CT-Q CO
QJ !
a: _i en o>
c: c
4-> U -I -r-
C T- 1
QJ C !-!-
>
-------
CJ
M
Pu
H
en i cC
c -v
r- C CU
E fO
O r
>, O CU
3 «: CQ
cvj
o
O
en
I CO O IT)
in «o- t <
co
VO l-v I fI f(
-i O O
ii o i
o o i
o o o
m
co
co
r- O)
O i-
C VO CU
i O O
co
o
o
co
CM r^ O in
1^ CM t1
i I O CT> I CO i I
. . .| ..
vo r i O O
co
O O r^
o o o
O
o
ro
r-
l
o
o
g
a
w
H
a
w
o
CO
cn CD O)
j_ s- 01
1-3-0
> JD -r-
Irt S-
4-> 4-> CD
in 4J >
O) T- O
3D 1
O
O
01
i oo
co o
co
. in i i l- o O co
I CM S-
r O 3
-" Z: CO
in
.
oo
CO
o
o
' O m
O)
co « i
co
co
O O CO
co vo co o
o o vo co
co
o o o
O 'I-
4-> C O
C
CU >- i-
V5
O 4_>
3
3
M to
oa <-
J_ 10
cu >.
-4-> O)
4_> C 3 (T3
fO O i C
*y o n3 e^
i- S>
cu 13 a:
i O CD -fcJ
r- c: ID
+j -a .- £
f^3 CU -i-5 »
, x -c: re -t->
o > in ai i
5» ij_ < 3: rD
&£
4^
S
**
to
E
L.
0
c c: cu u.
cu cu c:
C CD CD-I- S_ C L.
O O O i- 3 CD 3
J3 S_ S_ O <*- CD 14-
^_ "O -^ ^~ ^~ f* >> ^-*
ns >,-t- J= 3 01 X 3
v_)Z2:c_)oo> 3 i. O>
Q- 00 O El
4-38
-------
The normal nitrogen content of the SRC material can range
from 1.5 to 1.9 percent compared with 1.0 to 1.2 percent in the
natural coal. This is due to the fact that the SRC process does not
remove any of this constituent from the coal. However, more thorough
combustion testing is required, particularly at higher more repre-
sentative furnace temperatures, before it can be determined if this
characteristic of the process would present any special problem with
respect to NO emissions.
Although specific current costs estimates have not been pub-
lished for the SRC process, several general statements have been
offered by the industry based on estimates made by Southern Company
Services for two new power plants (one with raw coal firing and
scrubbing, and the other fired with SRC material).
The industry has stated that if one takes into account the
difference in expected "forced outage rates" in the two options,
achievement of the same electrical system reliability would require a
greater reserve capacity for the scrubber option. Industrial repre-
sentatives have further stated that the energy used in connection
with the scrubber operation will require greater generating capacity
for the scrubber option to achieve the same net electrical output.
Further, the capital requirements per unit of total output are said
to be less for the SRC option because of claimed inherent savings
associated with the quality and heating value of the fuel (i.e., less
4-39
-------
fuel storage area is required, the boiler can be smaller, the pul-
verizers can be smaller, the ash storage area can be significantly
reduced if not eliminated, etc. , and no provisions are required for
scrubber wastes). The industry has stated that when these factors
are taken into account, the capital requirements for an SRC-fired
generating plant will be only about 60 to 70 percent of those for a
coal-fired plant equipped with scrubbers. However, much of this
capital difference will be required by others for investment in the
SRC refinery itself, although this will reduce the huge capital
burden of the utility industry. SRC firing is also claimed to bring
about reductions in operating cost since cost savings are claimed
because there is no need to provide for scrubber reactants, for
personnel and energy to operate and maintain the scrubber system, and
for disposal of scrubber wastes.
4.2.4 Summary and Conclusions
While physical and/or chemical coal cleaning has some potential
to extend the range of usable coals under the current NSPS, the impact
is much less when a new, more stringent NSPS is considered. Figure
4-8 graphically projects the reduction in sulfur content based on
four cleaning scenarios and plots them against the percent of mines
sampled with coal sulfur content less than or equal to industrial
percent sulfur. An alternative might be the use of coal cleaning
with another SO,, control system such as FGD. This is especially the
case with physical coal cleaning since it is a low-cost, well estab-
lished technology that provides a consistent product in terms of
4-40
-------
00
w
Pi
O
o o
01 'ivr1' i vo !:vi!.i. '/rn
4-41
o
H CO
O
W M
W CO
O O
i-J M
< a
o <;
o w
CO U
O CO
M O
s S
H <
i-i p>
PO
W O
u u
c
H CJ
H- 1
53 Pi
M W
O >
ort M
w o
03
jj
Q
0)
d
H
4-1
o
CO
^)
C
o
OJ
(C
PQ
-------
moisture, ash and caloric value and requires little or no R&D.
The chemical processes, however, are still in the development stage.
The projected costs and sulfur removal for a number of coal cleaning
processes are summarized in Table 4-8.
4.3 Fluidized Bed Combustion
The removal of sulfur compounds can be achieved during the
combustion process by application of fluidized bed combustion (FBC).
This process has potential advantages over conventional coal-burning
processes:
The ability to burn a large variety of fuels
The capability of directly controlling the emission of sulfur
oxides (SOX), as a step in the combustion process
The potential for simultaneous reduction in emission of
nitrogen oxides (NO )
X
Small system size per given capacity
High thermal efficiency (heat transfer rate)
Projected lower capital and operating costs.
4.3.1 Overview
Fluidized bed combustion involves the burning of a fuel (coal)
in a bed of inert ash and/or an active sorbent for the control of
sulfur oxide emissions. The bed is fluidized (i.e., the solid
particles of coal, ash, and sorbent are held in suspension) by air
injected at controlled rates through an air distributor plate that
supports the bed material.
4-42
-------
>-l
Pi
w
K^
c5
CJ
W &-S
Pi
3
Ij
PQ
O
M
H ^
C3
0 Pi
rS ^^
PL)
pi
»i
P3 CJ
PH M
rJ H
PD M 6-8
CO Pi
>-l
Pw
CO 3
Z 4-1
O M
M
HvO
VA-*
M CO
CO W
S CO CJ
O HO
cj co Pi c
O PH O
CO CJ 4-1
OO CO --.
1 W
-* u
O
w Pi
rJ PM
PQ >-4
< O
W CJ
rJ
CJ
i-4
<
O
U M
CO 3
pi MO
O CO ,fl
r-J -
cn pa
Cfl r<
rH 0) 0) <1 £
cd >~, rH Q O
CJ 01 rH pi 0)
H S 0) C Pd 4-J
CO -~^. 4J CU ~-^ cfl
>^ & 4-1 NPQ^g C
42Pi cfl cfl>OO >,W
pMEHMptiWrJPQOO
cd
4-J
CO
T3
rl
01
42
4-1
o
T3
fi
CO
CO
4-)
cn
o
o
>4-l
o
cn
H
cn
Cfl
4"^
(-1
O
UH
4-1
X
OI
4-J
OI
0)
CO
cd
r~l
cd
o
CJ
-a
OJ
C
rd
0)
rH
O
O)
r<
Cfl
CO
4-)
cn
0
CJ
rH
O
4-J
CO
H
CO
CO
43
r>
CO
4-'
CO
o
CJ
0)
o
(2
CO
c
0)
4-1
c
H
cd
0
T3
a
cd
*-.fl
cxu
C
H
4-1
cd
M
a)
CL,
o
CO
3
rH
CX
OI
6D
M
cd
43
O
rH
cd
4-J
H
PX
cd
o
CO
0)
T3
3
rH
CJ
C
M
J2
4-43
-------
The two major categories of fluidized bed combustion are atmos-
pheric and pressurized fluidized bed combustion (AFBC and PFBC). In
both types of FBC, the reactive bed materials generally employed and
most thoroughly studied are calcined dolomite (MgO CaO) or calcined
limestone (CaO). These calcium compounds are calcined under heat,
yielding metallic oxides that react the sulfur oxides from the coal
combustion to form sulfates and, hence, the sulfur oxides are removed
from the flue gas.
The amount of SC>2 removed by a CaO bed is not limited by ther-
mo-dynamic equilibrium. The partial pressure of S02 in equilibrium
with CaO, CaS04, and 02 is 1.25 x 10~7 atm at 902°C (1656°F),
or about 0.125 ppm, This very low value is not achieved in practice
because equilibrium is not reached. Relatively high S02 removal
has been obtained in experimental units by using greater than
stoichiometric quantities of CaO. EPA sponsored research has de-
monstrated that the FBC can achieve more than 90% sulfur removal (see
Figure 4-9). Furthermore, developmental PFBC units have burned mod-
erately high-sulfur coals, and achieved 0.3 Ibs S0x/million Btu in
the flue gas stream. For high sulfur removal rates (90" or greater)
Ca/S ratios of 2 or greater are required, which leads to large
quantities of spent limestone. In a once-through system design
this material must be disposed of in an environmentally acceptable
manner. Research on methods of regenerating the spent bed material
is in progress. The excess CaO required for efficient S02 re-
moval complicates the regeneration processes by increasing the
4-44
-------
- Sulfur-retention capabilities of
additives compared, molar basis
lOOi
001 OMiTE, I650T
(I INf DHArfN Ff
-------
quantity of material to be processed. For these reasons considerable
research is being devoted to keeping the Ca/S ratio requirement at a
minimum. A detailed discussion of factors and data relating to S02
removal in a FBC is presented in Appendix A.
The solid wastes generated by the FBC process consist of ash
and spent limestone or dolomite particles. This granular matter
is withdrawn from the bed or removed as finer particulate matter from
the effluent gases by the final dust collection. The sorbent require-
ments and disposal problems can be reduced by a regeneration process.
When a regeneration process is utilized, sulfated sorbent is withdrawn
from the bed and regenerated to produce an SO,, or H«S rich stream
which is subsequently fed to a conventional sulfur recovery operation
producing sulfur or sulfuric acid. The regenerated sorbent is then
returned to the bed.
These FBC wastes are of two types: ash and sulfur containing or
desulfurization waste. This is the case with conventional combustion
and FGD or coal cleaning as well as FBC. The desulfurization wastes
from AFBC and PFBC generate a dry material consisting of calcium sul-
fate; calcium carbonate; calcium oxide; and, in some cases, magnesium
oxide.* Several other minor constituents are also present. While
the wastes from FBC are of a magnitude similar to FGD, because of the
differences in physical characteristics and the higher calcium
sulfite content of the FGD wastes, the disposal of FBC wastes is
*Percentage depends on amounts of limestone and/or dolomite.
4-46
-------
expected to be of a somewhat lesser magnitude. Because the waste is
dry, reclamation of disposal land should be possible with minimal
preparation. The FBC wastes are also expected to have lower potential
for trace element toxic effects; however, a further investigation in
the area of trace elements and leachate effects will be required to
verify these expectations. Additional research in waste utilization
and sorbent regeneration will also aid the development of the FBC.
4.3.2 FBC Systems
Fluidized bed combustion systems may be categorized according
to the presence or absence of heat transfer surfaces in the bed
(or the excess air levels) and the operating bed pressure. The
three basic systems are: combustion at atmospheric pressure (tubes
in the bed and excess air levels of 15 to 25 percent), pressurized
bed (tubes in the bed and excess air levels of 15 to 25 percent) and
adiabatic pressurized combustion (no tubes, heat exchangers, hence no
Q change, in the bed and excess air levels of 300 percent). The low
excess air pressurized system can be further subdivided into those
systems using air and those using water as the working fluid in the
tubes in the bed. The system descriptions and flow diagrams presented
here are subdivided on the basis of the combustion pressure alone.
The descriptions are for a generic AFBC and PFBC.
4.3.2.1 Generic Atmospheric FBC of Coal. A schematic flow
diagram for a generic atmospheric FBC unit is presented in Figure
4-10. Atmospheric fluidized-bed combustion occurs in the temperature
4-47
-------
f
I
I
B
B
B
B
B
E
H « c&
< O.
CU O
O
tH
h*
s
U.1
i'.
U4
O
M
U O
Ul >-l
«s
as o ^J
W H 3 U
CO C
a- = c; a.
r o f-^ . -K p^ ^
ox < ':
H u r>
O
W I!
3 a
O 0
0
u ^
to)
erf
a
M
ttS
t/3
i
H
ta
w
CO
oo
CO
w
CJ
o
06
u
w
2
W
O
4-48
-------
range of 788°C to 843°C (1450°F to 1550°F) with excess air values of
15 to 25 percent, at normal atmospheric pressure. Steam produced in
the bundles and/or water walls located within the fluidized region is
converted to electrical energy in a conventional steam turbine cycle.
Tube bundles immersed within the bed are expected to result in an
increased heat transfer over conventional boilers due to the constant
agitation of the bed material and gases in control with the steam
tubes. Typical fluidization velocities are 6 to 8 ft/sec.
Most of the ash present in the coal feed is normally elutriated
from the bed and must be removed (along with attrited limestone and
other particulates) before release of flue gas to the atmosphere.
Since this ash may be high in carbon content, its direct disposal
would result in a lowered combustion efficiency. To remedy this
problem, the atmospheric unit can employ a carbon burnup cell (CBC),
a separate high-temperature, high-excess-air bed to which the collec-
ted ash is fed and combusted. Products from the CBC would then
undergo an additional particulate removal operation prior to flue
gas release to the atmosphere. Alternatively, reinjection of collected
ash back into the combustor may be adopted to improve combustion
efficiency. It should be noted that the "particulate removal operation"
blocks indicated in the generic flow sheets are general and may
include combinations of cyclones, filters, baghouses, precipitators,
and other devices. Normally a cyclone or series of cyclones is
employed initially to remove coarse particles from the flue gas,
while final cleanup is accomplished by other methods.
4-49
-------
The generic units are intended to be representative of antici-
pated commercialized systems and, as such, it is projected that there
will be on-site fuel preparation, sorbent preparation, and sulfur
recovery operations.
The fuel preparation operations involve drying, size reduction
and size classification. These operations are similar to those found
in conventional systems. The sorbent preparation operations consist
of size reduction and classification of raw material and should pose
no new problems. Similarly, the steam turbine cycle and water treat-
ment operations are similar to those of conventional steam turbine
systems.
Available data from pilot and bench scale studies indicate
that generally 90 percent sulfur retention can be obtained with a
Ca/S ratio of 3 to 4 (Figure 4-11). While the data from small scale
units are promising, these experimental units are orders of magnitude
smaller than proposed commercial units. Extrapolation of pilot and
bench-scale studies may not be reliable for estimating SO,, emissions
for larger demonstration and commercial scale units because scale-up
of gas-solids interactions is especially difficult. The DOE and
EPA development programs will be collecting these data as larger
pilot plants and demonstration plants come on stream. In the event
that S02 control is found to be poor, it can be improved by deeper
beds or lower superficial velocity.
4-50
-------
o
»-
UJ
a:
ce
ID
u.
_J
ILLINOIS
ILLINOIS
ILLINOIS
ILLINOIS
ILLINOIS
TEMP. ADDITIVE
°F NO.
1359
1359
1359
1359
1350
1GC-U 137?
AVERAGE PARTICLE SIZE RANGE FOR ADDITIVE' 490-630 >K
GAS VELOCITY IN CO'nBUSTOR: 2.6 TO 2.8 It sec
3 4
Ca/S MOLE RATIO
FIGURE 4-11
EFFECT OF Ca/S MOLE RATIO ON SULFUR RETENTION
4-51
-------
The additional limestone requirement to meet reduced S0? levels
would aggravate the sorbent supply and disposal problems and impact
FBC process costs in an undetermined way. Moreover, since much of
the limestone is elutriated from the bed, the load on the particulate
handling systems could be expected to increase substantially.
Attempts to improve limestone utilization by crushing and reinjection
could further complicate the particulate collection problem. Process
information needed to gauge these effects is currently unavailable.
4.3.2.2 Generic Pressurized, Combined Cycle FBC of Coal. A
schematic flow sheet for a generic, pressurized, combined-cycle
unit is presented in Figure 4-12. As stated earlier, pressurized
units are further subdivided in terms of their heat transfer surfaces,
for instance, the adiabatic PFB (Figure 4~13) generates steam fron
a waste heat boiler after the gas turbine in contrast to Figure 4-12
which utilizes in-bed heat transfer surfaces.
Combustion for the pressurized bed again occurs in a fluidized
bed of fuel, ash and sorbent, with excess air ranges similar to those
found in the atmospheric boiler and at temperatures approximately 93°C
(200°F) higher. Pressure within the combustor, however, is maintained
at a design value of 4 to 10 atmospheres, resulting in a dramatic
reduction in combustor size requirements and, thus, combustor cost.
The elevated pressure of the PFBC allows the use of deeper beds,
resulting in greater combustion efficiency (99 percent as compared
with 98.5 for AFBC).
4-52
-------
-------
O
H
to
PQ
ro
r-H
I
d-
W
&
8
H
Pn
u
I
CJ
Pn
o
u
pa
PH
w
ij
u
><
CJ
p
w
s
u
SI
to
g
!
o
H
to
O
ffj
W
!E5
W
O
4-54
-------
In contrast, the combined cycle adiabatic combustion utilizes
about 300 percent excess air at 816°C to 982°C (1500°F to 1800°F) and
four atmospheric pressure. The system will require a somewhat larger
combustion than the other pressurized system.
The primary power for the adiabatic combustion (about 80 percent)
will be from the gas turbine and the remainder from the waste heat
steam turbine generator. The pressurized (nonadiabatic) combined
cycle system would develop the primary generating power from the
steam cycle and secondary power from the gas turbine.
The fuel and sorbent preparation, the steam turbine cycle and
water treatment for the pressurized system are similar to correspond-
ing atmospheric operations. The particle control systems for the
combined cycles must maintain levels sufficient to meet gas turbine
requirements, which are generally more stringent than requirements
for the atmospheric particle control system, which must meet current
control standards.
In pressurized fluidized bed combustion, small pilot plant
data have shown that sulfur reductions of 90 percent may be obtained
on high sulfur Eastern coals with the use of dolomite sorbent (Figure
4-14). The activity of limestone is less than that of dolomite
under pressurized conditions and it is completely inactive at low
temperature turndown conditions. Unless some way of increasing its
activity can be found, limestone will not be used in a PFB combustor
to achieve high level sulfur reduction because of excessive stone
4-5'
-------
100
o
1 2
Ca/S (MOLE/MOLE)
FIGURE 4-14
COMPARISON OF S02 REMOVAL RESULTS - DOLOMITE SORBENT
4-56
-------
requirements. Under present standards, limestone may still be used
for sulfur reductions up to 70 percent (~3 percent sulfur coal) since
the higher calcium content of limestone compensates for its lower
activity. Above this level, the higher activity of dolomite is
needed to achieve the necessary sulfur reduction.
Although for some AFBC designs the full scale system may consist
of many small modules close in size to those units tested, the PFBC
pilot plants are orders of magnitude smaller than proposed commercial
units. Efforts in PFBC have concentrated on the use of high sulfur
coals, and little if any information is available regarding sulfur
removal efficiencies with low sulfur coals «1 percent S). A 90
percent reduction may be more difficult for low sulfur coals since
the rate at which calcium reacts under fluidized bed combustion
conditions is first order in S0« concentration.
4.3.3 Status of FBC
The atmospheric process is inherently simpler and, thus, more
fully developed than the pressurized process. Widespread indus-
trial use of the AFBC is projected for the mid-1980s whereas the
PFBC is projected to reach commercialization about 1985. EPA spon-
sored research indicates, however, that the pressurized process has
many advantages over the atmospheric process in terms of reduced
number of coal feed points, better pollution control, and higher
energy conversion efficiency. Emissions of carbon monoxide, hydro-
carbons, SO and NO are inherently less from PFBC than from an
X X
4-57
-------
AFBC. Cost estimates show that for utility-industry applications
an AFBC and PFBC boiler plant may provide savings on both capital
and annualized costs over the costs of a conventional pulverized
coal-fired plant with a stack-gas cleanup system. However, this
potential will be affected by the results of the current research
program as well as any NSPS revision.
While there are no commercial utility FBC plants operating in
the U.S., there are several test facilities operating, under construc-
tion, or planned. Table 4-9 lists the number of operating units,
some of which were used to develop the data discussed here.
Though no actual reliability data are available based on the
simplicity of the FBC concept, good reliability is expected. Major
problem areas are expected to be:
Hot gas clean-up
Materials of construction
Current lack of design data for scale-up.
Various development programs are addressing the first two areas,
and it is anticipated that information from the 30-MWe AFBC unit
Rivesville, West Virginia pilot plant and other systems will resolve
the third.
4.3.4 FBC Vendors
Though not yet commercially proven, at least two manufacturers'
are currently marketing industrialized FBC steam generating systems.
While they are making some operating performance guarantees, the
4-58
-------
TABLE 4-9
SELECTED LIST OF OPERATIONAL
FLUIDIZED BED COMBUSTORS
Unit Name and Location Size/Capacity
Scale
Type
Pope, Evans, and Robins 9 sq. ft.
DOE, Alexandria, Va. 0.5 MWe
PDU
AFBC
Pope, Evans and Robins 4 sq. ft.
DOE, Rivesville, W. Va. 3.0 MWe
Pilot
Scale
AFBC
Babcock and Wilcox
Renfrew, Scottland
6.1 MWe
PDU
AFBC
Argonne National
Laboratory, Argonne, 111.
6" diameter
Bench
PFBC
Exxon
EPA, Linden, N.J.
12.5" diameter
0.63 MWe
Miniplant PFBC
British Coal
Utilization Research
Association, England
6sq. ft.
0.5 MWe
Bench
PFBC
Process Development Unit
Atmospheric Fluidized Bed Combustion
Pressurized Fluidized Bed Combustion
4-59
-------
status of environmental performance guarantees is still somewhat
undetermined. A greater number of manufacturers are offering FBC
incinerators with waste heat boilers attached. No commercial utility
size systems are expected until about 1985.
4.3.5 Summary
Theoretically, FBC can be employed as a control technique to
meet reduced emission standards. Whether or not this can be achieved
practically and economically depends on the course of the research
and development activities.
4.4 Flue Gas Desulfurization
Sulfur compounds can be removed from fossil fuel combustion
gases through the application of a large number of specific chemical
processes. Many have been evaluated as potential means of treating
the flue gases from coal-fired boilers. Some processes have been
rejected as impractical, unreliable, or uneconomical, while others
are in various phases of development or have been applied com-
mercially.
4.4.1 Overview of Flue Gas Desulfurization Processes
The principal types of flue gas desulfurization processes
that have been tested or are currently being tested can be classified
in accordance with the scheme illustrated in Figure 4-15. A major
distinction is made between throwaway processes, in which all waste
streams are discarded, and regenerable processes in which the sorbent
is regenerated and recycled. In certain regenerable processes,
4-60
-------
FGD
PROCESSES
DEMr/:,£.T
.-"LL SCALE
PHOTOTYPE
PILOT PLANT
Source: Bechcei
p THROlAWAf
4
UTED ON-
35 'IW
10 - 35 MW
1 - 10 MW
1977.
p
WET
SEMI DRY
(SPRAY DRIER)
WET L.
SEMI DR\
(SPRAY DRIER)
CLEAR
LIQUOR
r
SLURRY
CLEAR
LIQUOR
L
L
r-
-
I.1KL
Lixi.sro'..
\L'^\LP.C
L L^ \Sfl
LMl.STU-.
BOILER
i>,jrcTio;,
SODILM
CARBONATE
N'OMRECtNER^BLE
DOUBLE
ALKALI
LIME CKLuklDC
DILUTE
ACID
SODIUM
CAR30-CATE
NONREGENEKABLE
NAHCOLITE
LIHESTO'.E
BOILER
INJECTION
MAGNESIUM
OXIDE
SODIUM
SULFITE
AMMONIA
CITRATE
PHOSPHATE
POTASSIUM
THIOSULFATE
SODIUM
C,^BOSATE
-JECL'.'ERABLE
CARBON
ADSORPTION
COPPER
OXIDE
CATALYTIC
OXIDATION
MOLTEN
CARBONATE
PILOT PLANT/COMMERCIALLY AVAILABLE
FULL SCALE/COMMERCHLLY AVAILABLE
PROTOTYPE 'COMMERCIALLY AV \ILAaL£
PILOT PLANT COMMERCIALLY AVAILABLE
PILOT PLATT/COMMCRCI iLLY AVAIL,\BLE
FULL SCALE/UOTETERMINED
PILOT PLANT/ABANDONED
FIGURE 4-15
FLUE GAS DESULFURIZATION PROCESSES TESTFD ON COAL-FIRED BOILERS
-------
elemental sulfur or sulfur compounds can be recovered from waste
streams as marketable products.
Flue gases can be treated through wet, semidry or dry desulfuri-
zation processes of both throwaway and recovery types. Wet processes
are subdivided further into the categories of slurry processes
and clear liquor processes. In slurry processes a suspension of
active sorbent is used to contact or scrub the gas stream. Examples
of slurry processes are lime and limestone scrubbing and the regen-
erable magnesium oxide (MAGOX) process.
Wet scrubbing processes can remove both fly ash and sulfur
dioxide simultaneously from a gas stream. In practice, however,
there may be good reasons for collecting fly ash separately, general-
ly by means of electrostatic precipitators or fabric filter (bag-
house). Possible interference with the process reactions is avoided
by removing the fly dsh upstream of the desulfurizing unit, and
erosion of the desulfurization process equipment is reduced. The
volume of sludge is also minimized when the fly ash is removed prior
to the desulfurization process. In addition, contamination of the
reagents and by-products is prevented.
Potential problems with scaling, plugging, and erosion can be mini-
mized in clear liquor scrubbing processes. Clear liquor desulfurization
processes involve various scrubbing solutions (Figure 4-15), including
sodium carbonate (used in the throwaway, single alkali process), sodium
sulfite (double alkali process and Wellman-Lord process), and ammonia.
4-62
-------
All wet processes cause a considerable cooling of the treated
flue gas and an increase in its moisture content. Reheat of the gas
prior to discharge may be desirable in certain applications to avoid
condensation and corrosion in ducts, fans and stacks downstream of
the scrubber and to restore the buoyancy of the flue gas entering the
stack. Avoidance of a visible stack plume (due to condensation) may
be an added incentive to reheat the gas. These drawbacks are largely
circumvented in semidry and dry processes. Further, the disposal of
solid wastes generated in semidry and dry throwaway processes may be
easier than the disposal of sludges and liquid wastes.
In the semidry process, the flue gas is contacted by small
quantities of spray containing sodium carbonate. The spent sorbent
may be either discarded or regenerated. Dry, throwaway processes
that have been tested on coal-fired boilers entail the use of nah-
colite (a mineral containing natural sodium bicarbonate) or limestone
(consisting mainly of calcium carbonate) injected into the boiler.
The two dry regenerable processes that have been demonstrated on a
pilot plant or prototype scale are the carbon absorption and copper
oxide processes. In the latter, flue gases are treated at tempera-
tures above 371°C (700°F), a feature that tends to enhance the
attractiveness of the system in new rather than retrofit applica-
tions. Nahcolite can be injected at temperatures above or below
149°C (300°F), although better sulfur removal efficiency is achieved
4-63
-------
at higher temperatures. Flue gases are desulfurized by the carbon
adsorption process at temperatures that may be lower than 149°C
(300°F), making the process amenable to retrofit installations
downstream of air heaters in power plants.
Though there is research and development work being done on wet,
semidry, and dry processes, the most successful accomplishments to
date have been with the wet processes in both the throwaway and
regenerable categories. As a result the six most highly developed
and applied systems are all wet systems:
Throwaway Regenerable
Lime scrubbing Magnesium Oxide
Limestone scrubbing Sodium Sulfite
Sodium Carbonate
Dual Alkali
These processes are expected to make up the largest portion of
new scrubber installations in the near future. Therefore, the
major portion of this text will address these wet systems.
4.4.2 Sulfur Dioxide Removal
In practice, the effectiveness of a particular wet scrubbing system
for removing sulfur dioxide from flue gases is dependent on a number of
design features and operating parameters. Principal among these are:
1. Liquid-to-Gas Ratio. Removal of sulfur dioxide from flue gases
is generally improved by a high ratio of liquid flow to gas flow since,
with a large circulating volume of scrubbing medium, the concentration
of dissolved sulfur dioxide is kept low and a large gas-liquid interface
area can be provided. An increasing penalty, in terms of the power
4-64
-------
expended in circulating the scrubbing medium, is incurred as the liquid-
to-gas ratio is increased. Liquid-to-gas ratios, expressed as gallons
of liquid pumped through the scrubber per thousand cubic feet of gas
flow, are typically of the order of 20 in venturi scrubbers, 50 in
turbulent contact absorbers, and 80 in spray towers (Bechtel, 1977).
2. Gas Velocity. Flue gas desulfurization is generally improved
if the velocity of the flue gas is kept low and the residence time of
the gas in the scrubber is lengthened. For a given throughput of gas,
a low velocity can be attained at the expense of larger scrubbing
equipment. However, the disadvantage of a high gas velocity can be
offset in part, by a high gas-liquid interface area. An upper
limit to the gas velocity may then be set in consideration of the
quantity of mist that is entrained with the gas. Gas velocities
are typically of the order of 5 to 30 feet per second in tower
scrubbers and in excess of 100 feet per second in venturi scrubbers
(Bechtel, 1977). The gas residence time in the venturi scrubbers may
be as low as a few hundredths of a second.
3. Gas Turbulence. A high degree of turbulence in the gas stream
is advantageous in promoting the rapid diffusion of sulfur dioxide from
the bulk flue gas to the gas-liquid interface.
4. Slurry Holdup. A relatively long holdup residence time of the
scrubbing medium in the scrubber is desirable to allow a more complete
transfer of sulfur dioxide from the flue gas. In slurry scrubbing with
lime or limestone, the holdup time is of particular importance since
4-65
-------
the uptake of sulfur dioxide is controlled by the slow dissolution
of suspended alkaline material. Residence times of slurries in
towers with packing may range up to 5 seconds or more (Bechtel,
1977). At the other extreme, the residence time of slurries in
venturi scrubbers may be of the order of a few hundredths of a second
per pass, a feature that constrains such systems to somewhat lower
removal efficiencies. However, it can be increased by multiple staged
scrubbers.
5. Scrubber Internals. The function of the internals or
packing in tower-type scrubbers is to provide a high gas-to-liquid
interface area. Several types of internals are in common use,
including staged moving spheres on grids, closely spaced rods or
grills and perforated trays. Fixed packings are generally not used
in slurry service because of the tendency to plug or scale. The
absorbent is often sprayed into the towers through several stages of
spray headers.
6. Flow Configuration. Scrubbers can be designed to operate so
that the gas and scrubbing medium flow is countercurrent, crosscurrent,
or concurrent. Countercurrent and crosscurrent flows have an inherent
advantage in that the cleanest gas comes in contact with the freshest
absorbent, but comparable performance can be attained in concurrent
flow with multiple stages of spray.
In addition to the parameters that govern the desulfurization
performance of a particular device, other characteristics of the
4-66
-------
system are important in determining the operating costs associated
with the system and its ability to function under the variations in
load during normal operation. One major factor that determines the
energy costs associated with a scrubber system is the pressure drop
in the gas across the scrubber and the system. Additional factors
include thermal energy for the reheat system and the pumping power
required for the circulation of the scrubbing medium. Energy require-
ments for the FGD systems and the impact on steam/electric generation
are discussed in Appendix B.
Varying loads on a boiler are accompanied by variations in the
flow rate of the flue gases, and scrubbers are required to maintain
satisfactory levels of performance under partial load conditions. Some
scrubbers are designed to be "turned down" to 50 percent of design load,
while others are constructed of individual sections that can be isolated
and closed off. A venturi can have a variable throat area to accommo-
date turndown. In big installations with more than one scrubber, in-
dividual modules can be taken out of service as the load is reduced.
4.4.2.1 Scrubbing Equipment. Scrubbing equipment that is in
general use with wet desulfurization processes falls generally into
the broad categories of packed towers, spray towers, tray columns,
and venturi scrubbers.
A packed scrubber is a device consisting of a tower filled with
one of many available packing materials between pairs of support and
restraining grids. Conventional packing materials such as raschig
4-67
-------
rings, berl saddles, intalox saddles, and ball rings are of little
use in slurry scrubbing because of their propensity for plugging.
With clear liquor scrubbing, high absorption efficiencies can be
achieved in conventional packed towers operating with liquid-to-gas
ratios of 15 to 30 gallons per thousand cubic feet (Bechtel, 1977).
The turbulent contact absorber (TCA) is a countercurrent multi-
stage scrubber consisting of retaining for grids that both support
and restrain mobile packing spheres. Good gas-liquid contact and
scale removal result from the turbulent movement of the spheres. Two
to four stages may be used and high sulfur dioxide removal can be
attained in slurry scrubbing at liquid-to-gas ratios of 40 to 60
gallons per ACFM of gas flow (Bechtel, 1977). The pressure drop per
stage is approximately 2 to 2.5 inches of water (Bechtel, 1977). A
schematic of the turbulent contact absorber is shown in Figure 4-16.
In a marble bed absorber, a fixed bed of glass spheres (marbles),
typically 4 inches in diameter, is kept in slight vibrating motion
and creates a turbulent layer of liquid and gas above the spheres.
Pressure drops across marble bed scrubbers are generally between 4
and 6 inches of water and the operating liquid-to-gas ratio is in
the range of 25 to 30 gallons per thousand cubic feet (Bechtel,
1977).
A spray tower is a countercurrent type of scrubber in which
the absorbent is sprayed through several headers and nozzles. Spray
towers can be designed to operate, with a relatively low pressure
4-68
-------
MIST ELIMINATOR
WASH WATER
Retaining Bar-grids
GAS IN
TCA SCRUBBER
GAS OUT
Mist / \
itor
Jt
-------
drop, but liquid-to-gas ratios of the order of 80 gallons per thousand
cubic feet are needed to attain a high removal of sulfur dioxide from
the gas streams.
Tray columns are designed to provide contact between gas and
liquid in a series of trays or plates. At each tray, the gas is
dispersed through a layer of liquid and the number of trays (stages)
required in a particular application depends on the ease with which
the mass transfer operation can be effected and the overall degree
of S0« removal required. The liquid residence time in the column
is long, and a high degree of sulfur dioxide removal can be achieved
with a relatively low pressure drop (Bechtel, 1977). A liquid-to-gas
ratio of 40 gallons per thousand cubic feet is typical in tray
columns (Bechtel, 1977). Scaling presents a potential problem in
this type of device and undersprays at the trays are required to wash
off soft scale.
Several other types of scrubber towers or columns have been
applied to flue gas desulfurization, including the cross-flow absorber
and screen, or grid, tower. Cross-flow absorbers are installed
horizontally and have been tested both with packing and sprays.
Pressure drops across the absorbers are low, but high liquid-to-gas
ratios are needed to effect a high degree of sulfur dioxide removal
(Bechtel, 1977). Screen scrubbers consist of stacks of five to
ten screens, typically with 7/8 inch openings. Pressure drops
in screen scrubbers are low and liquid-to-gas ratios generally exceed
50 gallons per thousand cubic feet (Bechtel, 1977).
4-70
-------
Venturi scrubbers are being used for the removal of both particu-
late matter and sulfur dioxide from flue gas streams. These devices
generate a high degree of liquid-gas mixing and contact but have
the disadvantages of a relatively short contact time and a high
pressure drop. In fly ash removal applications, a liquid-to-gas
ratio of 10 to 30 gallons per thousand cubic feet produces a reduction
in particulate loading from typical values encountered in coal-fired
boilers down to 0.02 grains per standard cubic feet. The associated
drop in pressure would be 10 to 15 inches of water (Bechtel, 1977).
Because of the short contact time between liquid and gas, removal
of sulfur dioxide with a single stage of venturi scrubbing is limited.
Using lime or limestone slurry as the scrubbing medium, approxi-
mately 40 to 50 percent of the sulfur dioxide present in a flue
gas stream can be removed by a single-stage venturi scrubber (Bechtel,
1977). To attain higher removal efficiencies, two or more stages
of venturi scrubbers would be required. Alternatively, magnesium
oxide can be added to the slurry to improve its desulfurization
properties or an after absorber can be added downstream of the
scrubber. This latter arrangement is illustrated schematically in
Figure 4-17, which shows a venturi scrubber combined with a spray
tower. In this instance, the venturi scrubber has an adjustable
throat area, a feature that permits operation of the device over a
wide range of flow conditions.
4-71
-------
YENTURI SCRUBBER AND SPRAY TOWER
GAS OUT
Chevron Mist
Eliminator
SPRAY TOWER
INLET SLURRY
Adjustable Plug
VENTURI SCRUBBER
MIST ELIMINATOR
WASH WATER
MIST ELIMINATOR
WASH LIQUOR
^
APPROX. SCALE
EFFLUENT SLURRY
Source: Bechtel, 1977.
FIGURE 4-17
SCHEMATIC OF VENTURI SCRUBBER AND SPRAY TOWER
4-72
-------
4.4.2.2 Energy Requirements. The major energy requirements of
the wet flue gas desulfurization processes are those associated with
the circulation of the liquid absorbing medium and the compensation
for the (gas) pressure drop across the device. In addition, a
further expenditure of energy is incurred if the flue gas is reheated
to reduce corrosion in the stack, to restore plume buoyancy, or to
prevent the formation of a visible plume at the stack exit. Appendix
C briefly describes reheat systems and their operations.
Typical relationships between the energy lost in circulating
the scrubbing medium and liquid-to-gas ratio are illustrated in
Figure 4-18. Energy losses are expressed in terms of the fraction
of a power plant's gross electrical output that is expended in
operation of pumps, fans, small auxiliaries, additional process
equipment and thermal energy for exhaust gas reheat. These losses
increase both with increasing liquid-to-gas ratio and with the pres-
sure at which the scrubbing absorbing medium is injected into the
scrubber. During conditions representing the operation of venturi
scrubbers, pumping losses amount to approximately 0.25 percent of
gross station capability. Greater pumping losses are incurred with
turbulent contact absorbers (0.75 percent) and spray towers (1 percent),
Fans are used to compensate for the drop in pressure across a
scrubbing device and to maintain an appropriate flow of flue gases
through the combustion system. The energy utilization in fans
increases with increasing pressure drop, as indicated in Figure
4-19. Spray towers typically give rise to the smallest pressure drops
4-73
-------
u
O
u.
O
LL.
O
O
2
I 1-
20
I ' I ' I ' I ' I
40 60 80 100 120
l/G RATIO,GAL/1000 ACFM (INLET)
140
160
Source: Bechtel, 1977.
FIGURE 4-18
STATION ELECTRICAL LOSS AS A FUNCTION
L/G RATIO AND NOZZLE PRESSURE
4-74
-------
O
oc
O
U-
O
z
<,
!0
20 30
PRESSURE DROP,INCHES OF WATER
50
Source: Bechtel, 1977
FIGURE 4-19
STATION ELECTRICAL LOSS AS A FUNCTION
OF DRAFT REQUIREMENTS
4-75
-------
and their operation involves a loss of approximately 1 percent of
gross station capability. Fan losses associated with the operation
of turbulent contact absorbers amount to 1.5 percent of gross station
capability, and those associated with venturi scrubbers followed by
spray towers amount to 2.5 percent of gross station capability. The
aggregate of pumping and fan losses then ranges roughly from 2
percent of gross station capability with spray tower equipment to 4
percent with venturi and spray tower combinations.
4.4.3 FGD Processes
The process descriptions in this section are limited to those
that are commercially successful.
4.4.3.1 Lime/Limestone Processes. Lime and limestone scrubbing
operations are similar; both employ a slurry containing suspended
alkali as the SC^ absorbent medium. The basic difference is that
the lime process utilizes a lime-slaking process to produce a
lime (Ca(OtL)) slurry for scrubbing the flue gas, and the limestone
process utilizes finely ground limestone in its slurry. A typical
flow diagram for the lime/limestone processes is presented in
Figure 4-20.
The flue gas from the boiler enters the scrubber where it
is scrubbed by the respective slurry. The scrubbed gas then passes
through the mist eliminator which removes any entrained mists or
particulate matter in the gas. The cleaned flue gas is reheated to
increase buoyancy and prevent condensation of the moisture in the gas.
4-76
-------
o
CN
I
w
Pi
S3
o
si
O -J
0)
U
o
a)
o
O
CO
o
Pi M
o PP
<; PQ
M £D
Q Pi
U
o
i-l W
PM iz;
o,
to E-f
W W
W
i< M
U i-l
M
PH Pi
>-4 O
4-77
-------
The absorbed 862 reacts with dissolved alkali and results in a
calcium sulfite and calcium sulfate precipitate that make up the
major portion of the waste products. The slurry from the scrubbers
drains to holding tanks where the reaction reaches or approaches
completion. Makeup lime or limestone is added to the slurry in the
tank and reusable slurry is recycled to the scrubbers. This recycled
slurry contains from 5 to 15 percent suspended solids including fresh
alkali, reacted waste products, and fly ash. To remove the solid
waste products, a portion of the recycled material is withdrawn to
the solids separation equipment. A clarifier (thickener or settling
tank) separates the suspended solids and clear liquid. The liquid is
then returned to the scrubber loop. The waste product stream for the
clarifier containing 20 to 40 percent solids is directed through a
filter to further reduce volume and increase the solids concentration
to greater than 60 percent. These wastes are then discharged to a
disposal pond area. The clarifier and/or filter are optional;
however, elimination of these steps results in a greater volume of
sludge sent to the disposal area. In this case, once settling
occurs, the clear liquid may be removed from the top of the ponding
area and returned to the system.
Makeup water is added to the slurry to replace evaporated water
and water entrained in the waste stream. The water is added as mist
eliminator wash water through pump seals and the lime slaker (for the
lime process only).
4-78
-------
A summary evaluation of the. lime and limestone process is
presented in Table 4-10. These systems are the most popular because
of their proven operation and lower cost. A discussion of factors
affecting removal efficiency appear? in Appendix D. The appendix
afso includes detailed discussions of the mist eliminator and chemi-
cal scaling problems and techniques employed to reduce or eliminate
the problems.
The lime and limestone systems make up the largest portion
(greater than 90 percent) of operating scrubbers (PEDCo, 1977).
While early lime or limestone scrubbers had some operational prob-
lems, the operability and reliability of the newer systems has been
very good.
In 1972 the Phillips Power Station of Duquesne Light Company
began operation as the first major domestic lime FGD system. Since
that time, 11 other major stations have installed lime scrubbers
(Table 4-11). Each of the systems that has been tested is operating at
or above the S0~ efficiency required to meet S0« emission legislation.
Three lime units, Green River, Bruce Mansfield, and Mohave,
are briefly discussed here since they are examples of the avail-
ability and/or SO^ removal efficiencies of lime systems for both high
and low sulfur coal applications. The Mohave unit, is a demonstration
unit and is not representative of availability data. It is discussed
because of its S0« removal efficiency.
4-79
-------
TABLE 4-10
LIME/LIMESTONE PROCESS EVALUATION
PARAMETERS
COMMENTS
Process complexity, operability,
and reliability
Process performance
Wastes/product
Development status
Advantages/disadvantages
Relative simplicity and lower
costs make systems currently
most popular. Some problems
with scaling, plugging, erosion
and corrosion.
Capability of greater than 90
percent SOo removal has been
shown.
Systems produce large quantities
of waste sludge.
Commercially available for full
scale commercial operation on
coal-fired boilers.
Process can tolerate fly ash;
requires high liquid to gas
ratios and/or substantial gas
pressure drop.
-------
CO
w
H
CO
Q
W
H
I
^
^1
COO
C7> G C
H 01 41
kl -H
01 O
a C*H
O VM
CO u-l
01
kl
3
rH VM
fl -H
O 3
U a
»
JJ
kl 1
O u-i
O
> ki
41 JJ
z oi
14
09
3
JJ
10
JJ
a
3
1 0)
JJ JJ
kl «J
a -o
jj
CO
*
V -~
N 2
H Z
CO
>,
c
(0
a
6
o
u
c
0
4
JJ
10
jj
CO
01
a
o
vo
r~
at
r-l
iH
rW
kl
a
<
in
m
CO
O
U
kl
§1
o
a.
a
H
C
10
>
~H
>*
01
c
c
01
a.
a
01
H
VW
O
C
HI
Z
41
O
3
ki
m
m
00
^
i
in
m
Z
10
o
-H
JJ
10
^J
01
a
0
vo
r~
CM
JJ
n
3
IT
3
<
CO
r~
rH
O
H
rJ
JJ
U
41
rH
U
JJ
01
10
u
0)
rH
rH
rJ
>
n
r4
3
0
J
C
05
4)
C
10
u
0
to
CO
o
Z
10
c
o
*4
JJ
SJ
l>
W
01
a
o
in
i
en
rH
ki
0)
A
41
U
0)
a
0
vo
m
O
CJ
ki
01
J
o
a.
a
c
BJ
jj
c
o
a
H
kl
JJ
01
rH
0
U
0
en
en
«r
I
in
»'
Z
10
c
o
H
JJ
fl
4)
Q.
O
r
r-
a\
rH
>1
kl
10
3
kl
ja
£1
&L,
o
0
V
o
--H
.e
o
c
IH
01
JJ
3
O
CO O
0)
01
c
o
CJ
in
CO
CO
(V
1
rH
K
10
0
*-!
iJ
10
kd
01
Cu
o
in
r~
-H
|J
0)
c
10
01
3
cr
3
Q
0
§
kl
rH
U
<
z
r-
o
(X
>l
JJ
rH
H C
0 S
10 O
*w 13
4J
^ 13
0 A
di 01
l-
to
r-
ot
rH
>,
kl
O
3
ki
j3
01
Cx
o
to
rH
01
O
-H
>
kl
41
CO
u
-H
rH
£>
3
O.
10
c
o
N
-H
kl
<
01
kl
01
C
kl
o
o
kl
3
o
h.
o
CO
«»
1
CO
n
e
fM
(3
C
O
-H
JJ
fl
V4
41
a
o
in
r-
a\
rH
kl
1
01
JJ
a
O1
CO
^
to
0)
01
T4
JJ
-H
rH
-H
JJ
3
>.
Jtf
O
3
jj
C
01
X
kl
01
>
H
cc
c
01
41
ki
U
<
z'
CO
O
tx
><
JJ
rH
H C
0 S
10 0
'w *C
JJ
4^ 3
n s:
4) 01
H
m
r»
Ov
rH
kl
1
01
>
O
z
o
to
rH
C
O
CO
-H
o
b]
10
iH
C
kl
o
VM
«H
rH
JI
01
IQ
C
01
rH
rH
-H
>
a
H
3
Q
J
K
9
>i
W
o
fl)
a.
m
CD
CO
r*
I
rH
K
<8
C
0
H
JJ
10
Ut
01
a
o
(N
r»
cn
rH
rH
H
kl
a
<
o
rH
*r
.
O
U
jj
Si
a*
«4
J
41
C
ID
01
3
tr
3
O
a
a
-H
rH
rH
-4
J3
a.
o
m
to
r>
at
^H
10
o
H
JJ
1
k4
4)
a
o
vo
r-
»t
rH
Si
o
kl
«J
o
(N
ftu
S
CO
a
LI
o
it
o
10
i
41
J4
u
H
K
<
Z*
en
,
JJ
rH
*!
U
10
01
41
fH
§«
«
41
§
a
.c
CO
O
U
Q
W
P-i
CU
CJ
S-l
3
O
CO
4-81
-------
The lime slurry unit at the Green River Station of Kentucky
Utilities is attached to boilers 1, 2, and 3 which supply steam for
two-turbines with a total capacity of 64 MW. These generating units
are peaking units and normally operate 5 days per week, with one or
more of the boilers at reduced capacity. The boilers use high sulfur
coal with about 25,100 KJ/Kg (10,800 Btu/lb) heat content (Table 4-12).
Commercial operation of the system started in the fall of 1975.
After shakedown tests and discovery and correction of minor problems,
the closed loop full capacity operations began in March 1976. To date
system performance has been good. Mechanical reliabilty has been
excellent as shown by the average system availability, operability
reliability, and utilization data in Table 4-13. The operability
is plotted through May 1977 in Figure 4-21. The SCU removal effi-
ciency has been about 90 percent, well above the 80 percent design
value (PEDCo, 1977).
The Bruce Mansfield facility is the largest scrubber in the
world and is attached to the 835-MW Bruce Mansfield No. 1 steam
generator. The system burns 4.7 percent sulfur coal and must
maintain 260 ng/J (0.6 Ib SO /10 Btu) or an equivalent greater than
90 percent SC>2 control. Table 4-14 shows design related information.
This facility reported 100 percent operability (hours FGD
operated/per hours boiler operated) during the first months after
startup from May to December 1976. During the exceptionally cold
4-32
-------
TABLE 4-12
POWER PLANT AND FGD SYSTEM DESIGN DATA
Green River - Kentucky Utilities
Boiler data
Generating capacity, MW
Year placed in service
Boiler manufacturer
64
1949
Babcock & Wilcox
Coal data
FGD system
data
Heat value
Ash content
Sulfur content
S02 removal efficiency
Particulate removal
efficiency
Startup date
Flue gas rate
Flue gas temperature
Stack height
FGD vendor
25,100 kJ/kg
(10,797 Btu/lb.)
13 to 14 percent
3.8 percent
80 percent design
99.7 percent
9/75
170 m Is
(360,000 acfm)
149°C (300°F)
24 m (78 ft.)
American Air Filter
Source: PEDCo, 1977.
4-83
-------
CO
rH
I
w
rJ
pq
H
o
Pn
O
I
O
M
H
23
O
in
pi
u
^
o
pi
w
ft
w
w
pi
VD
r--
O-v
rH
1
10
N
< -H C
Q -H -H
4J 4J
3
X
1 4J
Ll -H
\ <-<> a H
Q O XI
1
o x
C -P
U Q -1
Q tn H
D XI
Li 10
1 >,
rH 4J
< -H -H
m > -H
10 XI
10
01 73
01 Ll 01
U 3 rH 10
0 H Ll
X 0 0.'
xi a
0
Q
U
Cv. TJ
E 01
Li -U 10
3 tn rl
0 X 01
x 01 a
0
Q
O
U.
a
en u c
C Ll rH 0
3 rH a
O 10 3
x u
Q 01
U rH
U. E XI
CO 01 10
LI tn -H
3 X 10
O 01 >
X 10
TI
01 o
Ll -H
3 C l-l
< O ~H 0)
x a
Month
vO
00
fN
rH
rH
o
*
CJv
i
in
i-H
o
o
VC
0
o
in
o
o
s
f^
January
ri (jv
m in
IN »
IN TT
T CO
IN in
IN T
en o
S S:
OD n
m in
IA
LTl OO
r- m
-H CD
rN rn
CO ^
ty^ oo
c^ o
r^ o
in in
m IN
CTi O
VD r^
0 0
o o
W O
^r
-------
100
90
80
70
S60
£50
Di
UJ
"-40
30
20
10
0
SHUT DOWN FOR
STACK REPAIR
JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC JAN FEB MAR APR MAY
1976 1977
MONTHS
Source: PEDCo, 1977.
FIGURE 4-21
SCRUBBER SYSTEM OPERABILITY - GREEN RIVER 110. 1, 2 AND 3
4-85
-------
TABLE 4-14
POWER PLANT AND FGD SYSTEM DESIGN DATA
Bruce Mansfield No. 1 - Pennsylvania Power Co.
Boiler data
Coal data
FGD system
Generating capacity
Year placed in service
Boiler manufacturer
Heat value
Ash content
Sulfur content
S02 removal efficiency
Particulate removal
efficiency
Start-up date
Flue gas rate
Flue gas temperature
Stack height
FGD vendor
839 MW
1976
Foster-Wheeler Corp.
27,700 kJ/kg
(11,900 Btu/lb.)
12.5 percent
4.5 to 5.0 percent
92 percent
99.8 percent
4/76
1580 m3/s
(3,350,000 acfm)
196°C (385°F)
290 m (950 ft.)
Chemico
Source: PEDCo, 1977.
4-86
-------
winter months of January and February 1977, the boiler lost 11 and 24
percent of generation capability as a result of FGD problems. In
March the system was taken out of service for a 10-week turbine over-
haul. During this time, repairs also began on the chimney flue for
A, B, and C modules. The polyester flakeglass lining had failed
and was being replaced. The other chimney flue for the D, E, and F
modules also needs repair. Roughly 1 year will be required to
complete the work and the boiler is being held to half load for that
time. Only three FGD modules are required for operation at half load
while the repairs are in progress. The D, E, and F modules continued
performing and had very good operability during the repair period.
The Mohave generating station is a demonstration facility
which has been used for extensive scrubber configuration tests with a
scrubber module capacity of approximately 170 MW. The unit employs
low sulfur coal (about 0.4 percent). Table 4-15 gives design para-
meters for the unit. Sulfur dioxide removal efficiency as excellent
for all types of absorbers tested. Although the S0« inlet concentra-
tion was 200 ppm, all of the absorbers were capable of removing 95
percent of the inlet S02. Outlet S02 loadings ranged from 1 to 10 ppm.
The S02 removal efficiency was strongly dependent on L/G for all
three modules shown in Figure 4-22. Note that the L/G shown for the
horizontal module represents the ratio in each stage. Table 4-16
shows the performance history of the Mohave system. Since this was a
test facility, several design changes were made during the period and
4-87
-------
TABLE 4-15
POWER PLANT AND FGD SYSTEM DESIGN DATA
Mohave Test Plant - Southern California Edison
Boiler data
Coal Data
FGD system
data
Generating capacity
Year placed in service
Boiler manufacturer
Heat value
Ash content
Sulfur content
SC>2 removal efficiency
Particulate removal
efficiency
Startup date
Flue gas rate
Flue gas temperature
Stack height
FGD vendor
790 MW
1971
Combustion Engineering
26,800 kJ/kg
(11,500 Btu/lb.)
10 percent
0.4 percent
95 percent
93 percent
11/73
212 m3/s
(450,000 scfm)
149°C (300°F)
152 m (500 ft.)
Southern California
Edison
Stearns Roger
Source: PEDCo, 1977.
4-88
-------
TOO
CIRCULATING LIQUOR FLOW RATE PER STAGE
(1/S)
0.5 0.75 1.0 1.25 1.5
1 .7!>
99
98
97
96
CM
O
IS)
fe 95
LU
94
93
92
91
HORIZONTAL
4 STAGES
LIME
90
5 10 15 20 25
CIRCULATING LIQUOR FLOW RATL PER STAGE
(1000 GPM)
FIGURE 4-22
EFFECT OF CIRCULATING LIQUOR FLOW RATE ON SO REMOVAL
3
AT CONSTANT GAS FLOW 212 M /S (450,000 SCFM) MOHAVE PLANT.
30
4-89
-------
o
fl'
I
w
H
iI
2
c-0
^ i
Pi
O
H
O
S
H
Q
|
O
M
H
p-l
o
Oi
«
W
fn
O
H
I
ON
<
1-1
ro
4J
I 4-)
(0 -iH
l-J rH
(1) -H
Q.J2
O (0
H -H
H i-l
QJ -r4
-r-l
< -Q
ft)
Q) M
^H 3
H 0
O X
I
CD tT>
^ C
Q) -H
Q.4J
O
tn
3 T3
O 0)
Q fO
U U
b
I
rH Q)
-H rH
cd xi
> CO
to
M
3
O
K
T!
O
C -H
H M
Q)
0.
Q
td
V)
C
O
0)
O
d
O
4-90
-------
this work contributed to the unavailability. As a result, oper-
ability data for the Mohave unit are not characteristic of a lime
system. It should be emphasized that the problems with the horizontal
mist eliminator gas flow distribution were solved and are not a
problem to future modules of this type.
A list of domestic limestone scrubbing units appears in Table
4-17. Design specifications for the units usually call for 60 to
80 percent SO* removal efficiency, depending on local regulations,
and all the units that have undergone performance testing have met or
exceeded design specifications. Three units that demonstrate the
excellent availability that can be achieved by these systems are La
Cygne No. 1 and Sherburne No. 1 and No. 2.
The Kansas City Power and Light La Cygne Power Station Unit No.
1 steam generator is a 820-MW (net) system and has one of the earli-
est limestone scrubber systems installed in the U.S. (1973). The
sulfur content of the coal ranges from 5 to 6 percent. System design
data are shown in Table 4-18.
The La Cygne FGD was plagued with numerous startup problems,
many of which were not due to FGD operation. However, despite the
problems at startup, the availability of the system improved steadily.
This system is now one of the most reliable large domestic utility
FGD systems. Figure 4-23 summarizes availability data. As shown,
availability for 1976 averaged 91 percent; the first half of 1977
averaged about 93 percent. This system was designed for 76 percent
4-91
-------
as
K
3
O
H
00
nJ
I
s
u
frl
W
i
a
PS
o
-H
*8 ^i
^ o
C 0 C
0»E 0>
* 01 -H
kl U
01 -H
Q INU-i
O **J
U 0>
O)
KJ kl
r-l JJ 3
« C «w
O 0> r-l
U 0 3
kl IS
C.
M 1
*s
01 JJ
Z 01
kl
a
4J JJ
kl «3
t3 13
w
01
H £
«T-{ ^,
w>-
SN
.w e
c a
C E
X O
C U
-Su
0 *
too.
kl
I
u
PM 0
H «
0 XI
kl 10
V
> 01
ox
jj in
in r-
e
00 *
in
fM
en
<-< «>
i i
T ft
»
0 tM
-*
£ £
jj
V
a
*
n TO
f^ ^*
x \
O 00
"-<
in o
r-l O
iH "V
O JJ
U f.
*iM O^
> ~<
w >j
V <
W H o
B O
H iH «-< .
o in o <*^ o
l O Cf\ iH
eo 1-1 v r- r~
4J «
X O
o> o
H
J kl
JJ JJ t)
kl »< -H ~* O.
fl) «J «J
5 . o M a
HO » m « 001 0)
(i Z -H JJ
u tj jj o s
o>i oo> oo> CI-H z JJ
Z Z * Z * Ji -< CO
-H O O 0 -H B
au o> o. 0)0, j jj c c
C U U D C kl
tK« co co e sot
>i«J 0) fl «0 --< -O £
(JO) kl B UB JJ « kl JJ
c * e > c kix 01 ki
00 no 00 0v j:o
i^K 4K i4K S H WZ
0
in
00
o
01
f-
X
en
o
to
vo
t
B
kl
o
a,
(N CO
0)
JJ
o a
z -u
t<
55
3 Oi
J3 J=
ki JJ
01 kl
.c o
w z
o
00
in
fT)
£
r»
^
«v
o
e
(M
a
0)
,4
r*4
^
>,
*H 4-*
^
* ^^
o
Z -3
jj e>
0) ^1
* e"
£ C
JJ -4
S '*
o a
W CO
o
GO
^
f*>
JJ
H
O
kl
JJ
01
K
r-
\
in
o
in
in
>,
jj
*H
kl
0
JJ
00 <
. N.
o o>
Z «H
J«tf *5
01 >
0)
ki S)
^s
n a
* 01
O C
o c
H 0)
S H
M
00
O
«r
jj
-H
0
kl
JJ
0>
K
W
^
<>4
f
\e>
ff4
0
i-t n
H
"D
O M
£
jS
«H JJ
jj rH
f; flj
3 0)
0 *
^i
r-l S
-< O
S U
o
r»
o
i-t
(U
f
\
f^
o
CD
f^
0)
o
>
4)
CO
U
3
a.
0
jj
«N -H
iH
O
O ki
z«
0 C.
C 3
H 0
* w
r-
r-
o\
H
H
>
Q.
O
^J
«O
iH
t4
m
JQ
E
0)
a
-------
TABLE 4-18
POWER PLANT AND FGD SYSTEM DESIGN/
OPERATING DATA, LA CYGNE NO. 1
Maximum generating capacity
Boiler manufacturer
Year placed in service
Maximum coal consumption
Maximum heat input
Unit heat rate
Stack height above grade
Flue gas ratemaximum
Flue gas temperature
Particulate
Removal efficiency (actual)
so2
Removal efficiency (actual)
No. of FGD modules
Process vendor
820 MW (net)
B & W
1973
366 metric ton/hr.
(404 ton/hr.)
8,105 106 kJ/hr.
(7,676 106 Btu/hr.)
9880 kJ/kWh
(9,360 Btu/kWh)
213 m (700 ft.)
1,297 m3/s
(2,760,000 acfm)
141°C (285°F)
97 to 99 percent
70 to 83 percent
B & W
Source: PEDCo, 1977.
4-93
-------
CO
r^.
cr»
cr>
PC
o
o
o
CTi
O
CO
% '
CN
I
E-H
II
HJ
r-
cr>
ur>
w
o
>
CJ
<
4-94
-------
SO- removal efficiency. Actual SO efficiency has been 80.18 percent
with the seven modules operating on 720 MW. Under maximum load, the
removal efficiency averaged 76.2 percent. Efficiencies under both
conditions should improve now that eight modules are operating.
The Northern States Power Co. Sherburne County Station No.
1 and No. 2 units both have 700 MW net capacity and burn 0.8 percent
sulfur coal. Availability for Unit No. 1 averaged 85 percent for the
four months of operation after startup. For the past 12 months,
availability has been in excess of 90 percent. Unit No. 2 has
shown even better startup performance, with operabilities averaging
about 95 percent for the first 4 months. These data are shown
in Figure 4-24. Table 4-19 shows pertinent operating data for the
first 8 months of operation of No. 1 unit. The S0« removal efficien-
cy was 50 to 55 percent, which was sufficient to meet local regula-
tions .
Based on the operating experience of lime and limestone systems,
(see Appendix D) there seems to be sufficient evidence to show that
these systems can operate at 90 percent SCu removal or greater
and that they can operate reliably (90 percent operability) with
proper design and maintenance.
4.4.3.2 Sodium Carbonate Scrubbing. The sodium carbonate
process accomplishes the removal of S0~ from the stack gas by employ-
ing a clear water solution of sodium carbonate. As can be seen in
Figure 4-25, a solution of soda ash (Na^CCO reacts with SCL to form
4-95
-------
.-I CO
o o
z: z:
-4
-3
>-
Di
01
s:
CO
UJ
o
o
o
H-
Q-
CD 2
>-
_J
13
O
>-
-------
TABLE 4-19
SHERBURNE COUNTY GENERATING PLANT
Unit 1 - Performance Data
I. Unit Data (based on May 1 through December 31, 1976 data)
Electrical output
Overall capacity factor
On-line duration
3,068,130 total MW-hr
60 percent
5,176 hr
II. Scrubber System Data (Averages)
Particulate concentration:
Inlet
Inlet
Outlet
Outlet
Removal efficiency
Sulfur dioxide concentration:
Inlet
Inlet
Outlet
Outlet
Removal efficiency
4.6 to 9.2 g/dry m
(2 to 4 gr/dscf)
1.7 to 3.4 g/kJ
(4 to 8 Ib/MM Btu)
0.080 to 0.10 g/dry m3
(0.035 to 0.044 gr/dscf)
0.032 to 0.036 g/kJ
(0.075 to 0.085 Ib/MM Btu)
98 to 99 percent
400 to 800 ppm
0.730 to 0.859 g/kJ
(1.7 to 2.0 Ib/MM Btu)
200 to 400 ppm
0.370 to 0.41 g/kJ
(0.85 to 0.95 Ib/MM Btu)
50 to 55 percent
Source: PEDCo, 1977.
4-97
-------
MAKE-UP WATER
FLUE GAS
Na2C03
1
SODA
LIQUOR
STORAGE
TO CHIMNEY
BLEED
ABSORBER
WASTE
LIQUOR
SURGE
TO
SEALED
DISPOSAL
POND
Source: Bechtel, November 1977.
FIGURE 4-25
SIMPLIFIED PROCESS DIAGRAM FOR
SODIUM CARBONATE SCRUBBING SYSTEM
4-98
-------
sodium sulfite/bisulfite. Both the reactant and the reaction
products are highly soluble in water. If fly ash is removed prior to
S0~ removal, the absorber can be either a packed tower or a tray
tower, which has very high efficiency with low pressure drop. If fly
ash is not removed, or is only partially removed, a venturi scrubber
may be used for both particulate and S0« removals with somewhat lower
absorption efficiency and higher presure drop. In either type of
absorber, a recirculating liquid stream as well as fresh soda makeup
is required to effect better gas-liquid contacting and better
S02 removal. Since the sodium alkali is very reactive with SC^,
the required L/G ratios are generally low (in the 10 to 25 gal/mcf
range). The system responds rapidly to changes in SO* loadings, and
the soda feed rate can be controlled by pH signal from the absorber
effluent, the pH changing with S0« loadings.
Table 4-20 is a summary of the sodium carbonate system, an
important process consideration is the purging of the spent alkaline
solution (sodium sulfite/bisulfite) in order to maintain the chemical
balance. The purge rate can be controlled by the liquid density.
This purge stream, usually slightly acidic, is neutralized with more
soda alkali before disposal. Process water makeup is required to
compensate for the water evaporated in the flue gas and lost in the
purge stream.
In some arid areas, the spent alkali purge stream may be
discharged to a sealed evaporation pond for drying. Alternative
disposal methods include fixation of the scrubber effluents and
4-99
-------
TABLE 4-20
SODIUM CARBONATE SCRUBBING EVALUATION
PARAMETERS
COMMENTS
Process complexity, operability,
and reliability
Process performance
Waste/product
Development status
Advantages/disadvantages
Extremely simple, easy to
operate, high degree of
reliability
S02 removal capability high
(better than 90 percent)
Disposal of spent alkali
solution requires extensive
evaluation
Full-scale operation on coal-
fired boilers
No scrubber scaling, low L/G
ratio, minimal corrosion and
erosion, can tolerate fly ash
in the system; limited appli-
cability because of expensive
alkali
4-100
-------
the recovery of sodium sulfate (salt cake) for sale. If permitted by
local regulations, ponding may be the most economical method.
The system consumes a premium chemical, either caustic soda
or soda ash; therefore, its application is limited to small indus-
trial boilers or utility boilers located near an inexpensive source
of the alkali. System advantages are simplicity, very high SC>2
removal efficiency, low capital cost, and good system operability and
reliability. Disposal of the spent alkali solution requires careful
consideration.
A prototype unit, serving two industrial coal-fired boilers
(equivalent to 25 MW) at the General Motors assembly plant in St.
Louis, has been in operation since 1972. The system availability has
been greater than 90 percent. Three 125 MW units at Nevada Power
Company's Reid Gardner Station are operating with low sulfur coal.
System availability has ranged from 70 to 99.4 percent (Bechtel,
1977) typically about 90 percent.
4.4.3.3 Double Alkali (Soda-Lime) Scrubbing. This FGD process
is a combination of sodium carbonate and lime or limestone processes.
The double alkali system is similar to the lime or limestone system
in that lime or limestone is used and a calcium sulfite/sulfate and
fly ash wet solid product is the result. However, a number of inter-
mediate steps are added. The SO^ is absorbed by a clear sodium
sulfite solution to produce soluble sodium bisulfate which is later
reacted with lime or limestone to produce the system waste product in
4-101
-------
the form of an insoluble calcium salt and to regenerate the sodium
sulfate. Hence, absorption and waste product functions are separated.
The results of the two-stage system is that scrubbing can be accom-
plished by a solution instead of a slurry. This will increase
reliability by reducing scale and plugging within the scrubber. The
system also increases both the utilization of the sorbent and SO
removal efficiency. The waste is essentially the same as the lime/
limestone sludge. Figure 4-26 shows a typical process flow chart
for a double alkali system.
The double alkali system is designed to combine desirable
properties of other FGD processes. This FGD process has the
high SOj absorption and nonscaling characteristics of the clear
liquid sodium carbonate process and avoids disposal problems of sodium
salt waste.
A summary of the double alkali system is presented in Table
4-21. One of the main problems associated with the double alkali
system is the regeneration of sodium sulfate (^280^). Sodium sulfate
does not react well with hydrated lime in the presence of sodium
sulfite (Na2S03).
The sodium sulfate is formed by oxidation of sodium sulfite in
the absorber. Regeneration of Na?SO, can be improved in two ways:
Minimize sodium sulfate formation by minimizing oxidation
through the use of a concentrated absorbing solution.
Employ a dilute absorbing solution to reduce the amount of
sodium sulfite, and increase oxidation since in the absence
of the sulfite sodium sulfate will react with lime to
precipitate calcium sulfate.
4-102
-------
I!
CN
I
s
pi
p
o
i-i
cu
,0
I
>
o
a
0)
4-J
x:
o
0)
u
S-i
o
CO
g
4-103
-------
TABLE 4-21
DOUBLE ALKALI PROCESS EVALUATION
PARAMETERS
COMMENTS
Process complexity, operability,
and reliability
Process performance
Waste/product
Development status
Advantages/disadvantages
Process is somewhat complex
and requires more components
than the lime/limestone pro-
cesses; operability and re-
liability are expected to be
greater than lime or limestone
FGD processes.
SO
2 removal capability greater
than 90 percent.
Disposal of calcium sludge;
additional problem of sodium
sulfate purge is required.
Full-scale operation on oil-
fired boilers in Japan; proto-
type operation coal-fired in
the U.S.; full-scale units
under construction in the U.S.
Minimizes scrubber scaling;
low liquid to gas ratio; no
product to market; requires
two or three separate solids
handling systems.
4-104
-------
Most operating experience to date has been based on 700 MW
from about 12 installations operating on oil or coal-fired industrial
boilers in the U.S. and Japan, three utility oil-fired industrial
boilers in Japan totaling 1,050 MW and a 20-MW prototype coal-fired
utility boiler developed for testing by Gulf Power Company in Florida.
There are currently three double alkali processes scheduled for
operation in 1979 on coal-fired utility boilers.
Several successful bench-scale, pilot plant and prototype
double alkali FGD systems have been tested on boiler flue gas appli-
cations in the U.S. The success of these programs has resulted in
commitments by three separate electric utility companies to install
full-scale double alkali FGD systems on coal-fired boilers. As yet
no full-scale system is operating on a utility boiler in the U.S;
but several systems are working on coal-fired industrial boilers, and
one pilot plant system and one prototype have been tested on utility
units.
Applications on industrial boilers are as follows:
Company: General Motors, Inc.
Plant: Chevrolet Parma
Location: Parma, Ohio
Stream treated: Off-gas from coal-fired boilers
System size: 124 m3/s (262,000 acfm) (32 MW)
S02 inlet: 800 to 1300 ppm (1.5 to 3.0 percent S coal)
Startup date: March 1974
4-105
-------
Company: Caterpillar Tractor Co.
Plant: Joliet Plant
. Location: Joliet, Illinois
Stream treated: Off-gas from coal-fired boilers
System size: 48.8 m3/s (103,500 acfm) (18 MW)
SC>2 inlet: 2300 ppm (4 percent sulfur coal)
Startup date: September 1974
Company: Firestone Tire and Rubber Co.
Plant: Pottstown Plant
Location: Pottstown, Pennsylvania
Stream treated: Off-gas from a oil-fired boiler
System size: 6.6 m3/s (14,000 acfm)
S02 inlet: 1,000 ppm
Startup date: January 1975
Company: Caterpillar Tractor Co.
Plant: Mossville Plant
Location: Mossville, Illinois
Stream treated: Off-gas from 4 coal-fired boilers
System size: 113 m3/s (240,000 acfm) (57 MW)
Fuel properties: Coal, 3.2 percent sulfur average
Startup date: October 1975
One prototype and one pilot plant double alkali system have
operated on utility coal-fired boilers:
Utility: Utah Power and Light Co.
Unit: Gadsby Station, Unit No. 3
Location: Gadsby, Utah
Unit size: 1.2 m3/s (2500 acfm) ( 0.6 MW)
Fuel properties: Coal, 0.4 percent sulfur average
Startup date: 1971
Note: Terminated 1973
Utility: Gulf Power Co.
Unit: Scholz, Unit No. 1
Location: Chattahoochee, Florida
Unit size: 35 m3/s (75,000 acfm) (20 MW)
Fuel properties: Coal, 3 to 5 percent sulfur
Startup date: February, 1975
Note: Terminated July 1976
4-106
-------
As a result of the success of pilot and prototype systems, three
full-scale double alkali systems are scheduled for operation soon on
new coal-fired utility boilers.
The GM Parma system has performed well with regard to S0~
removal. Results of a 1-week test in 1974 indicate SC>2 removal
efficiencies in the 94- to 99-percent range, with relatively low
inlet SO,-, levels (600 to 1200 ppm) and high excess air rates.
A test was conducted by A. D. Little, Inc., and General Motors (GM)
from August 19, 1974 to May 14, 1976. It consisted of three 1-month
intensive test periods and 18 months of lower-level tests. Removal
of SO reflects the variations in operating modes employed by GM
during the period, but removal efficiencies were at 90 percent
for the viable operating modes. Operation during April and May
1976 was excellent and A. D. Little, Inc., recommended continued
operation in the mode used during this period.
The operability (hours the FGD system was operated per boiler
operating hours in a period expressed as a percentage) of the Parma
system for the 1-year period from May 1976 through April 1977 was
about 70 percent. The system's best period of operation was May
through August 1976, when operability averaged 94 percent. The GM
Parma plant has several unique characteristics that affect operability.
Each boiler is equipped with its own separate scrubbing module with
no provision for crossflow between modules. The plant is not needed
during the summer months, because operations are shut down during
4-107
-------
automobile model changes and because there is no need for heating.
United Automobile Workers personnel operate the scrubber plant, which
precludes operation of the scrubbers when they are not on the site.
The GM plant is a developmental system, and as such is subject to
modifications. Many of the low operability periods were due to
mechanical outages or outages for modifications to accommodate
and test new modes of operation. Several different operating modes
have been investigated, and significant improvements have been
obtained in both process and mechanical performance. Although it has
yet to be proved over an extended test period, it is believed that in
the latest operational mode the system is capable of long-term
reliability.
The Joliet system has achieved excellent SO-} removal efficiencies
of between 85 and 95 percent under various operating conditions.
Sulfur dioxide inlet concentrations are high, about 2300 ppm. The
system was designed to attain an emission level of 860 ng/J (1.9
Ib S0?/106 Btu) (75 percent S0? removal), but has consistently
performed much better than designed.
The operability of the FGD system has been improving steadily.
Process availability for the period October 24, 1975, through June
1976 has been 100 percent. Most problems at the Joliet plant are
mechanical; the majority are solved while still on-stream or during
scheduled shutdowns. As a consequence, there have been few forced
outages.
4-JOS
-------
The Firestone-Pottstown system has exhibited excellent SO
removal efficiencies of 90 percent on high-sulfur oil, but no data
are available for its performance on coal. It has also achieved a
very high availability: 99 percent for the first 12 months of opera-
tion. Most downtime periods were due to mechanical component
failure or to maintenance, and not to unwanted chemical changes
or side reactions. No scaling problems have been experienced.
The Gadsby scrubbing system has performed well with respect to
SO,, removal. Various modes of operation were tested using two types
of absorbers. With the polysphere absorber, S02 removals of 90 per-
cent were achieved, giving outlet concentrations of 15 to 40 ppm S02»
With the venturi absorber, efficiencies ranged from 80 to 85 percent
S02 removal.
With the exception of the first 3-month operating period,
during which some gypsum scaling problems were encountered, dilute
mode operations were conducted for almost 2 years without any major
problems. No operating problems causing shutdown were experienced
between October 1972 and August 1974. For convenience, the system
was shut down on weekends, but no drainage of solution or cleaning of
equipment took place during these shutdowns.
The Gulf-Scholz prototype system started up February 3, 1975,
and operated continuously through July 18, 1975, when it was shut
down for repairs and modifications. The second period of operation
was from September 16, 1975, through January 2, 1976.
4-109
-------
The system exhibited excellent SOj removal capabilities, of 90
percent and greater. Using the combined venturi/absorber configura-
tion at a venturi liquor pH above 5.2, outlet SO,, concentrations
below 50 ppm were achieved, which corresponds to greater than 95-
percent removal. Raising the pH of the venturi liquor above 6.0
resulted in SOj removal efficiencies greater than 98 percent.
The operability of the Scholz plant for the period February
1975 through June 1976 is presented in Table 4-22. The Scholz
plant was designed to demonstrate the viability of the double alkali
process technology for application on utility coal-fired boilers.
As such, this prototype plant had less spare equipment than would
be normal in full-scale applications. The operability of the system
has been steadily improved; during the last 4 months of operation,
it was 94 percent.
Although no full-scale double alkali FGD systems are in opera-
tion on coal-fired utility boilers, it is possible to predict oper-
ability of the systems, based on experience with smaller units on
coal-fired industrial boilers described here. The operability of
double alkali FGD systems on coal-fired utility boilers and prototype
utility installations has been improved steadily; it is now 90
percent and above. Most operability problems were due to design-
related equipment shortcomings in these prototype installations. It
should be pointed out that most installations did not have spare
equipment; however, this would be included in full-scale utility
systems.
4-110
-------
TABLE 4-22
CEA/ADL DOUBLE ALKALI PROTOTYPE SCRUBBER
PERFORMANCE HISTORY: OPERATION AND VIABILITY PARAMETERS
Period
Feb.
Mar.
Apr.
May
Jun.
Jul.
Aug.
Sep.
Oct.
Nov.
Dec.
Jan.
Feb.
Mar.
Apr.
May
Jun.
75
75
75
75
75
75
75
75
75
75
75
76
76
76
76
76
76
Total
period ,
hr
672
744
720
744
720
744
744
720
744
720
744
744
696
744
720
744
720
Boiler
operation,
hr
459
507
604
598
720
683
744
577
559
620
732
0
0
480
642
735
656
FGD
operation,
hr
454
485
336
375
720
221
0
254
559
560
732
0
0
445
616
651
641
FGDa
operability,
percent
98
95
55
63
100
32
0
44
100
90
100
0
0
92
95
88
97
.9
.7
.6
.2
.0
.4
.0
.0
.3
.0
.7
.9
.6
.7
FGDb
utilization
percent
67
65
45
50
100
29
0
35
75
77
98
0
0
59
85
87
89
.6
.2
.2
.4
.0
.7
.3
.1
.8
.4
.8
.6
.5
.0
FGD operability: The number of hours the FGD system was in
operation divided by the number of hours the boiler was in
operation, expressed as a percentage.
FGD utilization: The number of hours the FGD system was in
operation divided by the number of hours in the period, expressed
as a percentage.
Source: PEDCo, 1977d.
4-111
-------
Equipment-related problems were solved at each of the double
alkali installations and the experience gained will benefit later
installations. The vendors of double alkali systems have developed
confidence in their reliability: as evidenced by guarantees of
90-percent availability for the first year of operation and 100
percent for the life of the plant (based on a boiler operating rate
of 70 percent for some of the new, full-scale utility applications).
The systems are all guaranteed to achieve 85 to 95 percent S0~
removal efficiency on high-sulfur coal applications. No new low-
sulfur coal applications are planned, but similar guarantees would
be expected for such systems.
Corrosion, erosion, and scaling problems have not been impor-
tant factors at double alkali FGD installations. Full-scale versions
of these systems are not expected to experience these problems either.
The double alkali system has demonstrated the ability to perform
well under fluctuating S0~ inlet concentrations. At the Scholz plant,
the design inlet SC>2 concentration was 1,800 ppm. At inlet concentra-
tions varying from 800 to 1,700 ppm, removal efficiencies were above 90
percent.
4.4.3.4 Magnesium Oxide Scrubbing. The magnesium oxide scrub-
bing process is a wet slurry scrubbing regenerable process that differs
from the lime process in two basic ways: the magnesium sulfite pro-
duced when S02 is absorbed from the flue gas can be calcined to recover
the S02 in concentrated form, and during calcining the magnesium oxide
4-112
-------
(MgO) is regenerated and can be recycled. The results are the process
produces a marketable product with no sludge to dispose of; and secondly,
there is only a small chemical makeup required. The magnesium oxide
process has not had the scaling problems of the lime/limestone pro-
cesses because magnesium sulfite and sulfate are orders of magnitude
more soluble at normal scrubber temperatures than their calcium coun-
terparts .
Figure 4-27 is a typical flow chart of the magnesium oxide
process. The magnesium oxide scrubbing requires particulate matter
removal prior to the flue gas entering the absorber. The flue gas
is next scrubbed with a slurry of 10 percent solids by weight. The
MgO reacts with the SC^ to form magnesium sulfite hydrates and as
portions of the sulfite are oxidized to the sulfate form.
The hydrated magnesium sulfite/sulfate (MgSO,,/MgSO, ) crystals
are withdrawn from the scrubbing cycle in a side stream containing
about 10 percent solids. This portion of the slurry goes to a
dewatering system (centrifuge) where the separate liquid is
recycled to the absorber and the solids in the form of a wet coke
are transferred to a dryer. The dried crystals are reacted with
a reducing agent (caibon) and calcined in a reducing atmosphere
at about 1500°F to produce concentrated S0,j and MgO. The reducing
agent is required to reduce the sulfate. The S02 rich stream can
then be used (after dust removal) for production of a sulfur or
sulfuric acid by-product. The MgO is recycled to the scrubber
system.
4-113
-------
, * CHIMNEY
SCRUBBER
RECOVERED SULFUR
FLY ASH-FREE
FLUE GAS
SULFUR OR ACID PLANT
Source: Bechtel, 1977.
FIGURE 4-27
SIMPLIFIED DIAGRAM FOR MAGNESIUM
OXIDE RECOVERY SYSTEM
4-114
-------
A summary of the magnesium oxide system is presented in Table
4-23. The major problems encountered with the magnesium oxide
system have been: (1) the formation of trihydrated sulfite crystals
in the scrubber as opposed to the hexahydrate crystals that are
easier to handle, (2) erosion of pumps valves and piping, (3) dust
from the dryer, and (4) excessive wear on internal parts of the
centrifuge. However, proper material choice and good operation and
maintenance practices will minimize some of these problems.
An added advantage for the magnesium oxide process is the
potential for a central regeneration facility to serve several FGD
systems. This is possible because the dry magnesium sulfite is
stable and easy to transport.
Three full size units have been operated in the United States:
Utility name: Boston Edison
Unit name: Mystic Station, No. 6
Unit location: Everett, Massachusetts
Unit size: 155 MW
Fuel properties: No. 6 Fuel Oil, 2.5 percent sulfur
Startup date: April J972
Demonstration terminated: June 1974
Utility name: Potomac Electric and Power
Unit name: Dickerson No. 3
Unit location: Dickerson, Maryland
Unit size: 95 MW
Fuel properties: Coal, 2.0 percent sulfur
Startup date: September 1973
Demonstration terminated: August 1975
Utility name: Philadelphia Electric
Unit name: Eddystone No. 1A
Unit location: Eddystone, Pennsylvania
Unit size: 120 MW
Fuel properties: Coal, 2.5 percent sulfur
Startup date: September, 1975
Suspended operation in January 1976 pending relocation of
calciner after acid plant shutdown. Restarted in June 1977.
4-115
-------
TABLE 4-23
MAGNESIUM OXIDE SCRUBBING
PARAMETER
COMMENTS
Process complexity, operability,
and reliability
Process performance
Wastes/product
Advantages/disadvantages
A complicated chemical opera-
tion unfamilar to the utility
industry, however the regen-
eration system can be separated
from the scrubber at the power
pi ant.
SOo removal capability greater
than 90 percent.
Produces sulfuric acid or sul-
fur for sale; sulfur poses
minimal disposal problems.
No scaling in system; oxidation
can be tolerated; must operate
acid plant; marketing of acid
may be problem; fly ash must
be kept out of regeneration
system; losses and deactivation
of MgO may occur by repeated
regeneration.
4-116
-------
Since the Mystic Station Unit No. 6 employed oil as its fuel,
no particle collection was required prior to the scrubber. The
scrubber did an excellent job of S0~ removal. Test data indicate
an average removal efficiency of over 91.6 percent. This was
o
achieved at gas flow rates ranging from 12,036 m /min (425,000 acfm),
Q
which was the design value, to 18,640 m /min (658,000 acfm), more
than 54 percent in excess of the design value. Outlet S02 con-
centrations averaged 82.7 kg/J (0.192 lb/106 Btu).
The operability of the No. 6 Unit at Mystic for the entire test
run is presented in Table 4-24. Operability is defined as hours
of FGD operation divided by hours of boiler operation in a given
period, expressed as a percentage. The unit worked best during its
last 4 months, when operability was about 80 percent. It would
have been approximately 85 percent but for a 2-week outage of the
off-site sulfuric acid plant. The FGD system had to shut down since
MgO could not be regenerated. From April 12 to May 10, 1974, the
system achieved 100 percent operability.
The operation showed that magnesium oxide scrubbing for sulfur
oxide removal is technically feasible. The scrubber performed at or
above the design efficiency (90 percent) at gas flow rates 50 percent
over design. Slurry solids separation was achieved and magnesium
sulfate concentration controlled. No plugging and scaling occurred.
The boiler at the Dickerson Station Unit No. 3 burned 2 percent
sulfur coal. Flue gas from the boiler normally passed through an
-------
CM
I
vC
O
u
M
w
IS
H
CO
O
K.
H
ii
nJ
ca
f«
o
CO
JJ
S
o
Cj
jj
f^
H
IB
kl
01
0
c
ki O
0) -H
JO JJ
jg « ki
3 ki JZ
kl 0)
u a
cn o
c
o
kl -H
01 4J ~
PH 10 kl
H kl £
g oi
a a
0
£,
J=
| 1
C
£
IM
*M
H
o
iH
10
u
H
C
10
"o
01
vu
O
01
0
3
IS
o
0>
>^
pH
JJ
c
0)
JJ
JJ
H
E
(U
fjj
c
-H
8
JJ
IB
ki
a
o
01
p-i
3
D
O
E
01
c.
^
fH
IN
r-
j^
a
i
ti
a
U
01
ki
o
B
pH
10
JJ
CO
kl
U
01
Jj
H
VM
fH
3
CO
E
-H
CO
01
c
Cn
A
C
01
jj
VM
0
5
-H
JJ
Ifl
kl
01
a-
tn
«
*-*
4J
. (
3
U
jj
E
CJ
pH
8
M
a
ki
O
I?
E
IB
"0
OJ
C
CJ
CO
?
J,,
X.
I
kl
kl
ooeu
».
cc
^,
IB
o
in
r-
»
c
o
-H
JJ
IB
ki
OJ
a
o
CO
3
o
3
c
«-<
JJ
c
o
o
IM
o
o
o
r-l
kl
01
a
jj
10
OJ
C"
C
0
pH
01
t-
00
vo
*
<-J
O
C
IB
01
C
3
r"3
Cn
C
H
kl
3
D
pH
VO
fH
in
ff^
in
f^
in
m
r"
fH
S
>*
JJ
rH
-H
n
ki
0)
U
O
E
Oi
JJ
CO
£
f
fH
D
IB
.C
IH
0
pH
a
3
c
c
10
01
JJ
kl
o
VM
C
0
o
OJ
IS
3
IH
0)
P-I
-H
O
J3
01
EH
. C
3
O
O
JJ
3
J=
CO
kl
01
pH
fH
0
03
rn
^
en
3
*
IS
o>
£
0
01
0
3
10
U
JS
kl
10
01
>f
01
5
VM
o
kl
OJ
Jj
kl
10
3
cr
CO
IS
pH
0)
jj
Cn
C
-H
kl
3
TJ
D
01
co
10
01
u
OJ
o
00
fi
n
{N
f^
n
vo
!">
^
a
4>
M
tl
en
3
IM
-H
M
JJ
C
01
o
D
c
IS
n
i*
a
c
o
fH
JJ
a
fH
3
(J
H
U
01
kl
kl
0
3
cr
H
**
01
JZ
JJ
c
H
c
0
H
co
o
kl
o
u
V.
c
o
fH
CO
o
kl
OJ
o vo
VO IN
r- IN
r- vo
PI fH
r- cn
IN CN
VO ^D
m r-,
r~ r~
*J >
o o
0 X
o
c
a
^
Jy
IB
C
IB
»~3
C
-H
CO
§
D
JJ
3
CO
JJ
c
01
3
cr
01
ki
VM
a
OJ
a
3
10
U
CO
E
0>
fH
kl
a
5
10
pH
0)
H ;>,
i It
kl 10
HI 3
pH kl
T-I n
o oj
C b,
n co in
fH IN (N
VO IN OD
oo in m
fH fH
co in fH
in in »r
vo in m
m f f
r* r~ r~
CJ C XI
tl N tl
Q i-j U,
.
£
JJ
C
0)
JJ
en
c
H
kl
3
O
T3
ki
ki
3
O
U
O
CO
o
0
kl
0)
a
c
0
H
JJ
IB
ki
01
a
0
in
3
O
3
C
H
JJ
C
O
o
k
^
IS
a
i
r»
0
t
r~ 1-1
00 CO
in (>
M "»
CD m
O CO
*r m
V f
r- r~
U »H
« a
0
S
0)
3
O
a
«a
JS
c
0
E
0)
JJ
M
O
VM
01
3
iH
IB
X
0)
a
c
H
JJ
.H
fH
-H
j3
10
ki
01
a
o
c
01
o
kl
01
a
i
in
01
j=
^
in
o
CO
CN
CO
CD
*r
«c
J^
£
.
41
^
3
-H
C
*M
!
CO
CO
8
o
JJ
c
10
jj
ki
OJ
IS
kl
Jj
c
IS
fH
a
a
-H
u
IB
01
£
JJ
VM
O
01
Cn
IB
3
O
v
01
2
1
a
o
01
^
Cn
C
-H
kl
3
o
o
01
IH
01
JJ
C
3
0
O
c
01
tH
IB
JJ
CO
U
0)
-H
VM
pH
3
CO
01
JJ
IS
kl
D
1
X
OJ
£
VM
O
a
10
OJ
CO
c
0>
JJ
10
IH
O
-H
IH
JJ
VM
0
C
o
-H
JJ
IS
I
o
D
tl
D
3
U
e*
U
^
JJ
«e
0)
o>
H
01
IS
Cn
V
CJ
IS
JJ
CO
VM
o
y
O
IB
fc
PH
01
IH
D
0)
JJ
C
(D
E
fH
O
kl
a
n
3
O
C
o
fH
n
O
IH
CO
o3
pH
j-i
O
a
OJ
a
IS
a
JH
0)
3
o
pH
_J
o
IS
JJ
CO
01
J=
Jj
c
o
JJ
IB
m
c
01
o
o
u
jj
c
01
a
-H
3
rv
&
^
IB
H
O
c
Q
kl
0
1
*o
c
fc
01
3
VM
-H
kl
JJ
C
01
o
T3
C
10
en
c
H
a
H
a
^
CO
a
a
.
n
01
ki
3
a
4-118
-------
ESP for fly ash removal before entering a two-stage venturi scrubber/
absorber. The MgO system was tested both with and without the ESP,
which can be bypassed. Dickerson Unit No. 3 achieved an average S0~
removal efficiency of 88.9 percent during performance testing (Table
4-25). If the one low result were omitted, the average would be 90.4
percent.
The operability of this MgO FGD system, defined as hours of
operation divided by boiler operating hours, is shown in Table 4-26.
The unit was a prototype trial installation, built to obtain operating
data and not necessarily for long-term operability. As such, equip-
ment and materials were used that would not have been used in a
long-term installation on a new plant. This operation fulfilled its
purpose by showing areas where improvement was needed. The opera-
bility data shows the downtime caused by mechanical and material
failures.
4-119
-------
TABLE 4-25
SO,, EMISSIONS TEST RESULTS FOR
X
MgO FGD SYSTEM - DICKERSON3
Test
Series
5A
5B
6
7
8
cm (in.)
38.4
16.8
13.0
37.3
13.2
H 0
2
(15.1)
(6.6)
(5.1)
(14.7)
(5.2)
so2,
Inlet
779
1373
800
1418
1419
ppm
Outlet
78
157
137
88
156
S02 Removal ,
percent
90
88.7
82.9
93.9
89.0
so3,
Inlet
34 . 6
47.5
2.9
1.8
ppm
Outlet
3.56
3.31
0.64
0.41
Test results abstracted from York Research Corporation, Final Report,
Y-8513, January 31, 1975.
Second Stage (absorber) pressure drop.
Source: PEDCo, 1977d.
TABLE 4-26
OPERABILITY DATA FOR DICKERSON NO. 3
TIME PERIOD
OPERABILITY, PERCENT
September 13, 1973 - January 14, 1974
December 9, 1973 - January 14, 1974
April 15, 1974 - May 1, 1974
August 1, 1974 - August 31, 1974
November 1, 1974 - November 30, 1974
November 15, 1974 - November 30, 1974
December 1, 1974 - December 31, 1974
Test Runs Completed August, 1975
27.4
58.3
40.5
43.5
44.6
67.9
57.9
Source: PEDCo., 1977d.
4-120
-------
The Eddystone facility scheduled for startup in 1975 had to
be temporarily shut down in January of 1976 because Olin Chemical
closed the acid plant serving the MgO calciner. The regeneration
equipment was relocated to Essex Chemical, Newark, New Jersey.
Although the Eddystone FGD system operated only a short time, it
removed more than 90 percent of the SO,, when both trays of venturi
rods in the SOo absorber were used and when the L/G was 6689 litres
3
per am /min (50 gal/1000 acfm). Eddystone must meet an S0~ emission
standard of 260 ng/J (0.6 Ib S02/106 Btu).
The operability data available are limited to the startup
down phase. Over the period October 2, 1975, through December
31, 1975, operability of the SO scrubber was only 33 percent.
During the recent startup the main problems were with ancillary
equipment; the major equipment, scrubbers and absorber, etc., worked
properly.
The two Chemico MgO FGD systems (oil-fired at Mystic; coal-
fired at Dickerson) were prototype units, built to demonstrate
the potentials of the process and to determine the major areas for
improvement. The venturi was adopted for use in these prototypes.
The units were built on a low budget and include little redundancy,
to enable a high percentage of on-stream time.
4.4.3.5 Sodium Sulfite. The sodium sulfite (Wellman-Lord)
process utilizes a clear liquid of sodium alkali with thermal regen-
eration of the sorbent and generation of sulfuric acid or elemental
4-121
-------
sulfur as a by-product. Figure 4-28 is a flow chart of the sodium
sulfite regenerable FGD process.
The flue gas entering the absorber is scrubbed with a sodium
sulfite liquid. The sodium sulfite reacts with the S0_ in the flue
gas to yield sodium bisulfite. The cleaned flue gas is reheated and
directed to the stack.
The sodium bisulfite is decomposed to sodium sulfite (solid) and
SO (gas) in a forced-circulation evaporator-crystallizer. The sodium
sulfite crystals are separated in a clarifier and redissolved in
water prior to recycling to the absorber. Sodium sulfate formed by
oxidation of sodium sulfite in the system cannot be decomposed and
must be purged from the system. The concentrated SO stream (90
percent SO , 10 percent HO) is directed to a sulfuric acid plant or
sulfur plant.
Table 4-27 is a summary of the sodium sulfite process,. The
removal of particulate matter from the flue gas prior to the sodium
sulfite absorber is a necessity. The liquid may have to be filtered
along with good particulate removal in order to assure reliable
operation. The oxidation of sodium sulfite to sodium sulfate is also
a problem. Five to 10 percent of the incoming sulfur is lost as
soluble sodium sulfate in the purge stream along with expensive
reactant. The lost reactant must be made up with soda ash or caustic
soda and the purge stream must be disposed.
4-122
-------
00
CM
I
st
W
O
M
a
a)
eq
ai
o
j-t
3
O
CO
4-123
-------
TABLE 4-27
SODIUM SULFITE SCRUBBING EVALUATION
PARAMETER
COMMENTS
Process complexity, operability,
and reliability
Process performance
Wastes/products
Development status
Advantages/disadvantages
This process is the most success-
ful regenerable process for oil-
fired boilers, availability has
been excellent; the system is
being demonstrated on coal.
S0« removal capability greater
than 90 percent.
System produces sulfuric acid
on sulfur for sales plus a small
amount of liquid waste.
The system is full demonstration
oil-fired units; full scale demon-
strations on coal-fired boilers
are in progress.
No scaling in systems; low liquid/
gas ratio in absorber; fly ash
must be kept out of system; corro-
sive process environment requires
expensive materials of construc-
tion; high steam consumption;
requires soda make up.
4-124
-------
The sodium sulfite (Wellman-Lord) process has been in operation
on two oil-fired 35-MW industrial boilers in Japan since August
1971. The availability has been close to 100 percent. Two larger
oil-fired systems, a 220-MW utility boiler and a 125-MW equivalent
industrial boiler started up in Japan in 1973, have both been
operating successfully. The first full-scale coal-fired EPA sup-
ported demonstration of the process is operational at Northern
Indiana Public Service Company 115-MW Mitchell Station. Two other
units totaling 715 MW are under construction in New Mexico.
Seven Wellman-Lord systems are currently in operation in the
United States. Six units are installed on sulfuric acid or Glaus
sulfur recovery units. The gas flow rates on these are small, 51,000
to 133,000 nm/hr (30,000 to 78,000 scfm), in comparison with a new
500-MW coal-fired boiler with a gas flow rate of 1,700,000 nm/hr
(1,000,000 scfm). The inlet SO concentrations on the seven small
units range from 2,700 to 10,000 ppm, which is about one to five
times the SO concentration expected with a 3.5-percent sulfur
coal-fired boiler. The smaller units, however, operate on fairly
clean, dry streams with low oxygen concentrations in comparison with
boiler flue gases.
The SO removal efficiency of the six units is typically 90
percent or greater; removal efficiencies in excess of 97 percent are
reported. The collected SO is either recycled to the sulfuric acid
or Glaus sulfur unit. Little operational data are available; it is
4-125
-------
reported, however, that the six units have absorber on-stream times
of greater than 97 percent.
The No. 11 unit at the D.H. Mitchell Generating Station of
Northern Indiana Public Service Company (NIPSCO) is currently the
only utility operation in the U.S. and the only coal-fired applica-
tion.
Initial startup of the NIPSCO Wellman-Lord absorber was on
July 19, 1976. An extended shutdown period began on November 28, 1976,
when high-pressure steam supply failures from the boiler and from
emergency backup systems to the FGD plant resulted in freeze damage
to the FGD plant. The shutdown lasted until early January 1977.
During the period from July through November 1976, the Unit 11 boiler
operated for 121 full days and 10 partial days, while the SO removal
system of the FGD plant operated for 71 full days and 23 partial days
and was down for 38 days. The steam supply failures previously men-
tioned were responsible for 28 of the 38 days that the system was down.
During the three sustained operating periods, the absorber demon-
strated the capability of greater SO removal than specified in the
performance criteria. The efficiencies are shown in Figures 4-29,
4-30, and 4-31.
The inlet flue gas contained as high as 2,800 ppm SO ; the
normal inlet SO concentration, however, ranged from 2,100 to 2,300
ppm. Flue gas volume exceeded that expected. Outlet SO concen-
trations normally ranged from 170 to 190 ppm, demonstrating the
capability of SO removal in excess of 90 percent.
4-126
-------
oo
Z3
LU O
LU LU
CO CO
LU If)
O
-=c
QL
Of.
rs
o
O£.
Iff
co
00
>
C\J
o
o
o
o o
o
CM
O
LU
"
D:
O
o
CM
O
LU
~T~
1
w
H
O
f:
f
eC
O
O.
-------
LO
CM
CD
00
LO
>>
fO
o
o;
o
i
r
o
ro
I
W
O
CO
2;
o
ii
H
H
5Z
W
U
H Pi
i-l O
H 53
£3 M
O Pi
5=i
23
-------
Q
Q Q UJ
O UJ LU
ti OO U_
LiJ
Q.
oo
oo
O
03
T3
Q;
ID
Q
__ O et
i^ P O
CJ3 1
z. o uj
>i oo :n "
o: ( <: uj
^» >- uj _j
o I zr o; i-i
CD
oo .__
UJ et 00
o a:
CTUJ >i
^> a: CQ
o o o
u o o:
uj n. Q.
CQ
re LU >->-<; z:
I c£ ca oo o: ii
P-t
cu
ro
I
w
O
H
^
a
o
oo
CNI
O 25
Pi
H
W O
iJ S
H M
O CD
Q
H
W
Aiucid
4-129
-------
There is only operating data for one large U.S. utility Wellman-
Lord system, but 17 systems are operating in Japan. As of mid-1976
all reported greater than 90 percent (some more than 98 percent)
SO^ removal efficiencies. The on-stream time for the absorption
area has been about 98 percent. None of the systems are on coal-
fired boilers.
4.4.3.6 Spray Dryer/Fabric Filter. This semidry process removes
SC>2 in two stages utilizing an alkaline slurry in a spray dryer in
conjunction with a fabric filter to collect the reacted alkali.
Spent sorbent may be disposed of (lime-limestone) or regenerated
(soda ash, etc.) by chemical reduction to produce by-product ele-
mental sulfur.
Figure 4-32 shows a typical flow sheet for the two-stage process.
Flue gas enters the dryer and flows downward through a finely atom-
ized spray of scrubbing solution containing an alkaline surry, there-
by removing a large fraction of the S02« The flue gas then leaves
the spray dryer by particulate loaders and enters the second stage
fabric filter where additional SC>2 is removed by reaction with
unused sorbent; the fabric filter also serves to remove particulate
matter. A summary evaluation of this process is given in Table 4-28.
The SC>2 removal efficiency varies with the sorbent used. Tests
have shown that efficiencies are highest for sodium sulfite which, at
stoichiometric ratios of 1.0 and 1.5, removes over 90 percent of the
S02 (>80 percent in the spray dryer and another 10 percent; in the
fabric filter). Dry injection of sodium bicarbonate or nahcolite
4-130
-------
,
h-
CL cr
U. C/D C/5 Q
w
o
O
CO
w
pq
u
CO
O
H
W
4-131
-------
TABLE 4-28
SPRAY DRYER/FABRIC FILTER PROCESS EVALUATION
PARAMETER
COMMENTS
Process complexity, operability and
reliability
Process performance
Wastes/products
Development status
Advantages/disadvantages
Combines two technologies spray drying
and fabric filtration which have had
wide commercial application in other
industries.
S02 removal capability greater than 90
percent.
Produces a dry powder mixture of sodium
or calcium sulfite and sulfate, unreacted
absorbent and flyash; potential regeneration
product is sulfur.
Demonstration unit scheduled to be oper-
ational on 100 MW coal fired power plant
by late 1979. (Eighteen are now under order.)
Minimal process development required
both technologies commercially available:
wet scrubber not required; low liquid/gas
ratio; expensive absorbent (if soda ash is
used, not if limestone is used) and fabric
filters; lower S02 removal efficiencies.
4-132
-------
, a one-stage fabric filter process gave 89 percent SC>2
removal using a stoichiometric ratio of 1.5. Lowest efficiency was
obtained with lime (CaO), which showed only 75 percent removal (50
percent in the spray dryer and 25 percent in the fabric filter) at a
stoichiometric ratio of 1.2; this is due to the comparatively low
solubility and reactivity of CaO. Overall efficiencies for the three
sorbents under identical test conditions (1.0 stoichiometric ratio;
1240 ppm 802) were: sodium carbonate, 92 percent; sodium bicar-
bonate, 74 percent; and lime, 71 percent.
If sorbent regeneration is used, the spent sorbent is reduced
with a carbonaceous reducing agent (petroleum coke or coal) at
1800°F. This regenerates the sorbent and produces hydrogen sulfide
which is then converted to elemental sulfur in a Glaus plant.
A program is now underway to design, build, test and operate a
demonstration 100-MW spray dryer using an electrostatic precipitator
rather than a fabric filter for particulate removal. The system will
use a sodium carbonate absorbing solution and will regenerate the
spent sorbent with a solid carbon reducing agent to produce hydrogen
sulfide that will be converted to by-product sulfur in a Glaus plant.
Test operations on this coaJ-fired unit are expected to begin in late
1979. The S02 removal efficiency is expected to exceed 90 percent,
using the electrostatic precipitator for particulate control. It is
4-133
-------
estimated that fabric filtration could be used at no additional
costand could improve both the 862 and particulate removal effi-
ciencies.
4.4.4 FGD Wastes
A more stringent NSPS would result in either or both an increase
in the amount of FGD wastes and/or FGD by-product. The waste pro-
ducers from lime/limestone and double alkali systems are virtually
identical (mixtures of FGD sludge and fly ash). Wastes from regen-
erable systems are primarily fly ash and purged liquid effluents.
The volume of waste from a typical nonregenerable scrubber system
over a 30-year period is shown in Table 4-29 for 1,000-, 500-, and
25-MW plants burning various coals under the alternative standards.
The wastes are in general proportional to size.
The chemical and physical characteristics of sludge vary based
upon a number of factors including coal, sorbent, scrubber, scrub-
ber operating parameters, and ash collection. The primary properties
of the sludge affected by these factors are:
Constituents Concentration
pH Total dissolved solids
Leaching characteristics Water retention
Bulk density Compressive strength
Permeability Viscosity
Compaction Porosity.
Because of high concentrations of salts and total dissolved
solids, the presence of trace elements, and in some cases extreme
4-134
-------
CN
I
W
PQ
H
01
0)
fa
I
0!
< O
O <
II ^^
PM
Ix cd
H O
O
> i-l
W
u ai
o ^
PH C
CO
CO
w <
H
CO rt
O
W CO
3 H
rt co
u >-<
CO CO
fa W
w
a
w
o
w
rt
rt
O
U
c
s-i
0)
0}
*
r^
"V
rt
" O
U
c
^
03
rt-
H
CD
ID
rt
C
Jj
_j
11
*^
to
p-
CO'
Z
^ s
CO ^
^ ^^2
rt O
in ^
u
o u
a
fvj ,_(
O £
^ c
e- P
o 3
tf
M ^
o a
CO ^
5 "o
rt
i-,
CP
CX
O
ID
ercent averag
a.
o
in
rt
CO CO
U-| C3O
f) O
J2 J2
H -H
CO CO
iH r-l
« rt
O 0
u o
0 O
U-l M-l
50 percent
60 percent
m
a
\o
i i
£
o
o
o
CO
00
o
"rt
o
u
C
u
CO
IV
rt
^
14
3
a
o
c
o
fS]
co"
IT)
ro
"rt
o
u
C
!-,
V
tn
/
to
c
,-,
3
(^
U
CO
^
Js
SO2 removal
^o
0
oo
u
f*
s
c
ro
O
CO
f-H
SO
0*
T)
85% SC>2 removal by scrubbing
w
c
*w
r*
"co
rt
^
rt
O
u
^
rt
sulfur remov
fe°
o
QC
C
_o
3
^H
u
CO
>*
J3
SO2 removal
fe^
o
%-<
f\)
0
CO
ro
o
CD
C
,2
3
U
CO
SO2 remova
m
r-
i
ON
ted
H
4-1
CO
-------
values of pH and COD, care must be taken in the choice of disposal
methods for untreated sludges. Secondly, because of its highly
water-retentive property, the material requires special handling,
conditioning or chemical treatment in its disposal to make the
disposal site reclaimable. Regardless of the type of handling or
treatment in disposal, consideration must be given to seepage to
groundwater, runoff to streams, intrusion into irrigation systems,
direct impact on vegetation, and impact on ocean life if disposed of
at sea.
Various forms of disposal are available and a selection depends
not only on cost but also on the following factors which are
generally site-specific: characteristics of the waste, climate,
geology, topography, hydrology, and disposal site availability and
proximity. Possible types of disposal are: ponding on Indigenous
clay soil; ponding with a flexible liner or a liner of impervious
soil; ponding with underdrainage; surface or deep mine disposal;
ocean disposal; and chemical treatment with landfilling. There are
specific cases where some of these methods are applicable,
environmentally and structurally. Although the chemical treatment
approach is universally applicable, it is not necessarily the best
choice in all cases if a ponding or mine disposal approach is
environmentally acceptable and less expensive. All disposal methods
require monitoring and site management throughout the active life of
4-136
-------
the site, and special provisions such as covering the site with soil
and the growth of vegetation to either prevent rewetting the material
or to prevent runoff problems, as applicable.
Three major products can be produced from flue gas scrubbing:
gypsum from nonregenerable systems; and sulfur and sulfuric acid
which are direct by-products of regenerable systems. Economically,
gypsum is not directly competitive; however, in consideration of
credits for disposal under certain conditions, it can be shown to be
a cost-effective commercial item. Sulfuric acid would have to
compete in an industry that is currently capable of producing 30
percent over demand. Attempts are being made to develop other pro-
ducts from sulfur sludge such as fertilizer and building materials
(Aerospace Corporation, 1977).
4.4.5 Status of Flue Gas Desulfurization Technology
Several aspects of flue gas desulfurization technology have been
the subject of continuous investigations. Rapid progress has been
made in expanding the data base needed to support the design of
commercial devices to treat flue gases from coal-fired boilers.
Through research, development and demonstration, information has been
acquired on all aspects of flue gas desulfurization, including the
basic chemistry of the various processes, the design and performance
of equipment, and the selection of appropriate construction
materials.
4-137
-------
Table 4-30 summarizes the number and capacity of FGD systems on
utility boilers in the U.S. as of August 1977. Of these systems, 29
were operational (8,914 MW), 28 were under construction (11,810), and
68 systems were planned (32,628 MW). This table omits 16 instal-
lations (8,592 MW) whose operators are considering FGD as well as
other control systems (e.g., low sulfur coal). Some 12 to 15 boilers
(6,000 MW) that are definitely planning to use FGD systems are
excluded, because the information is not ready for public release.
Not shown in the table are 16 systems (1,488 MW) that have been shut
down for various reasons. Several of these were demonstration sys-
tems; others were based on first-generation technology.
Flue gas desulfurization systems have been applied to new boilers
and as retrofits in existing installations. Appendix E specifies the
types of FGD systems associated with coal fired boilers in the U.S.
A summary list of these applications (by system type) is provided in
Table 4-31. Boilers equipped with flue gas desulfurization devices
vary in capacity from relatively small units of 100 megawatts or less
to large new units in the 800- to 850-MW class. The sulfur content
of the coal and lignite that is burned or scheduled to be burned in
these units ranges from 1 percent or less to approximately 5 percent.
As shown in Table 4-31, lime and limestone slurry scrubbing are
currently the predominant processes in flue gas desulfurization
applied to coal-fired utility boilers in the U.S.
4-138
-------
TABLE 4-30
BREAKDOWN OF FGD UNITS
Status of Units, August 1977
Number
of Units
Capacity,
Megawatts
Operational
Under construction
Planned:
Contract awarded
Letter of intent
Request ing/evaluating bids
Considering only FGD systems
TOTAL
29
28
23
5
5
35
125
8,914
11,810
11,880
1,892
2,825
16,031
53,352
Source: PEDCo, 1977e.
TABLE 4-31
FGD APPLICATIONS TO COAL-FIRED BOILERS
Process
Number
of Units
Capacity,
Megawatts
Limestone scrubbing
Lime scrubbing
Lime/fly ash scrubbing
Lime/1imestone scrubbing
Double alkali scrubbing
Sodium carbonate scrubbing
Magnesium oxide scrubbing
Wellman Lord/Allied Chemical
Wellman Lord
Aqueous carbonate scrubbing
Process not selected
Regenerative, not selected
Throwaway, not selected
41
24
8
2
3
5
4
3
1
1
28
3
2
125
18,003
9,910
4,047
20
1,102
1,009
846
830
180
100
14,675
1,650
980
53,352
Source: PEDCo, 1977
4-139
-------
4.4.6 Vendor Capabilities
4.4.6.1 FGD System Supply and Demand. Rased on the new coal-
fired boilers now planned for construction and a projected growth
rate of 5.56 percent per year, approximately 510,000 MW of coal-fired
boiler capability will be built between 1978 and the year 2000. The
ability of FGD manufacturers to meet this demand was evaluated in
terms of the present and proposed alternative NSPS of 0.5 Ib SC>2/
10^ Btu and 90 percent control. For the alternative NSPS, it was
assumed that a new standard would result in the use of FGD's on all
new power plants, and that the distribution of types of FGD processes
would be the same as the distribution of systems already planned and
shown on Table 4~32. Table 4-33 shows the demand for FGD based on
the projected new coal-fired units.
TABLE 4-32
APPROXIMATE PROCESS DISTRIBUTION OF PLANNED
FGD SYSTEMS ON NEW COAL-FIRED UTILITY BOILERS
Percent Application
FGD Process to New Units
Nonregenerable
Lime scrubbing 25
Lime/alkaline fly ash scrubbing 13
Limestone scrubbing 52
Double alkali 3
Sodium carbonate 2
Regenerable
Sodium solution 3
Magnesium oxide 2
Source:PEDCo, 1977c.
4-140
-------
CO
I
pa
z
o
t 1
H
<
IS!
od
CD
fa
1 J
CD
CO
W CO
Q H
H~l
CO Z
-l Z
nJ 0
H
3
Q
W
H
O
w
> >
o
Pi
PH
l-W
O
C
0
, ^-1
4-1
CO
N
r-l
t 1
r-l
4-1
CD
T3
01
4-1
O
CD
o
i_i
3-i
i i
CO
4-J
o
H
T3
a)
4-1
0
a)
r
o
^_i
PM
vQ
O
. i
CNI
O
CO
_a
i i
LA
o
CO
t-l
01
-a
C
Q
O
[T,
Q
O
fa
<4-l
O
C
Q
H
4-1
CO
N
H
i 1
H
4-1
CD
T3
0)
4-1
O
O)
1)
o
t~
PH
, l
CO
4-1
O
H
i i.
0
(-1
4-1
q
o
o
6-S
O
ON
)-,
0
3
CQ
CO
*\
CO
PH
CO
Z
4-1
0
0)
cn
o>
M
PH
\-t
0)
13
C
^-i
i '
n
CD
C
o
r-l
4-1
r-l
-a
-o
-.
4-1
H
O
CB
a
CO
O
M
CO
0)
>-!
d
cn
C
O
H
4-1
CO
3
60
Q)
04
CO
4-1
4-1
CO
S
CO
60
OJ
2
cn
4-1
4J
cfl
s
CO
60
O)
2
ooooooooooooooooooooooo
LTiOoOOOOOoOOOOOOOOOOOOOO
ONir~.Lnf")cNc-OLni
OOOOOOOOOOOOOOOOOOOOOOO
unOoOOOOOoOQOOOOOOOOOOOO
ON H I mc"NCNfN.m~3-LrN.vor--ooaNOi t
ON ON ON ON ON
ON ON ON ON ON ON ON ON ON ON ON ON ON ON ON ON O
Q
O
fa
U-l
o
C
o
H
4J
cfl
N
>r_j
i I
H
3
4-J
fl
0)
u
t-l
0)
(X
o
vD
T~l
U
-j
0)
4-1
M-l
CO
cn
01
)-i
3
60
i-l
fa
CO
cn
V-i
0)
1 1
r-l
O
42
T3
0)
(-1
iH
4--I
1
CO
O
o
s
0)
C
C!
O
Q
0
pc_i
4-1
O
q
o
4-1
CO
N
llj
H
4-1
3
4-1
q
01
o
1-1
01
ex
o
o
C
o
T3
QJ
cn
CO
«
-D
CJ
r~~
r^-
ON
1
%
O
o
Q
W
OH
0)
0
1-1
D
O
CO
4-141
-------
Responses received from 13 out of 18 FGD systems manufacturers
surveyed indicated they believe they will be capable of supplying the
design personnel and equipment for the projected demand for FGD
systems. The capability of the manufacturers to meet demand is
flexible and increases in proportion to demand. Tables 4-34 and 4-35
give the comparison of supply versus demand for FGD systems under the
current NSPS and two more stringent alternative levels220 ng/J
(0.5 lb S02/106 Btu) or 90 percent S02 reductions.
Ample limestone and to a lesser extent lime supplies exist in
this country to supply all FGD systems. Shortages in specialized
construction personnel are a possibility; however, the added
personnel needs for new FGD systems are a very small portion of the
total labor requirement. By about 1990, shortages in large scrubber
modules and fans are also predicted by several of the suppliers
depending on the sizes required at that time (PEDCo, 1977).
4.4.6.2 Construction Time and Guarantees. Table 4-36 gives the
design, construction and startup time requirements for an FGD process
based on the estimates of 13 FGD process manufacturers. Table 4-37
lists the items from the manufacturer's experience that can delay in-
stallations as a result of long lead times. A typical construction
schedule for a 500 MW unit with and without FGD is shown in Appendix
F. The schedules exclude the preliminary study which is estimated at
18 to 24 months.
4-142
-------
TABLE 4-34
COMPARISON OF SUPPLY VERSUS DEMAND
FOR FGD SYSTEMS ON NEW COAL-FIRED UTILITY BOILERS
UNDER PRESENT NSPS
Time
period
1978-1982
1983-1987
1988-1992
TOTAL
FGD
Vendor Capability
with Present Staff,
Megawatts
205,710
212,885
218,540
637,135
Projected Demand,
Megawatts
48,200
39,200
65,400
152,800
Differential
Capacity,
Megawatts
+ 157,510
+ 173,685
+ 153,140
484,335
Source: PEDCo, 1977c.
TABLE 4-35
COMPARISON OF SUPPLY VERSUS DEMAND
FOR FGD SYSTEMS ON COAL-FIRED UTILITY BOILERS
UNDER MORE STRINGENT NSPS
Time
Period
1978-1982
1983-1987
1988-1992
TOTAL
FGD Manufacturers'
Capability
with Present Staff,
Megawatts
371,500
421,890
434,990
1,228,380
Projected
Demand ,
Megawatts
75,550
65,400
109,000
249,950
Differential
Capacity ,
Megawatts
+295,950
+356,490
+325,990
978,430
Source: PEDCo, 1977c.
4-143
-------
cfl
,-1
t-M
3
H
Pi
<3
H
CO
Q
2j
z
O
M
H
Q
3
4-1
(-1
CO
4-)
CO
1-1
0
4-1
T3
cu
i-i
r-(
3
CT
CU
OS
01
a
S
H
co
CU X!
00 -t-1
d d
cfl O
(S -2.
cu co
oo -d
CO 4-1
1-1 d
01 o
> S
.5
d
o
1-1 -r-l
0 4J
4-1 Cfl
i 1
"O i 1
CU CO
1-1 4-1
H 03
3 d
cr M
CU
DJ T3
d
cu to
I d
H 00
rH
CO
CU
Q
CO
cu X!
00 W
d d
<5 o
oi S
cu
oo co
cfl X!
1-1 W
0) d
> 3
<< s
03
4-J
4J
* Cfl
cu S
N Cfl
H 00
CO CU
^
CO 03
XJ XI
4J 4-1
d d
o o
B B
vO vO
0 0
4-1 4->
03 03
XJ X!
4J 4J
d d
o o
s e
m m
o o
00 CO
i 1 CM
vO v£i
co co
1 1
0 OO
CN 3-
CN
CN
0)
Ol t
3 QJ
4-1 i 1
XI
Ol cfl
X! 4-1
4-J CX
CU
03 O
Cfl CJ
CO
T3
0) C
d co
t(_J
4-1 4-1
0) CO
T3
cu
CO 4-1
r-l CO
M
CX CU
3 CX
4J O
M
CO C
4J CO
CO O
XI
o
r-»
r-^
ON
i i
^
O
C_3
Q
W
PM
.
CU
ourc
CO
4-144
-------
W
PQ
H
fn
o
>1
u
z
w
O"
w
QQ
ta
<
w
Q
Q
^
«^
W
a
^
Q
^
W
1-1
a
w
H
CO
Q
O
pc-i
£2
^
pr_)
O
25
0
M
H
<
J
\3
^
H
CO
z
t 1
Q
Z
-<2
S3
CJ5
M
CO
W
°
uj
t~r]
£1
Z
M
CO
2
W
H
M
CO
£3
O
HH
OS
o
60
d
H
r 1
cx
0)
^
CO
J^J
0>
4-1
O
CO
4-1
3
d
cd
s
U-l
o
g_j
01
Pi
g
3
^
>,
o
d
Ol
3
cr
01
4-1
>-,
r5
0)
Q
S
01
4J
r-l
f-;
4J
CO
a,
i i
CO
O
r-l
4J
H
j^j
U
^
0)
E
H
4_)
T3
CO
0)
1 1
Ol
oo
cO
j^
01
<
o
13
Ol
60
2
0)
>;
. ^
O
CO
0)
£*-*
43
4-1
d
o
e
E
Ol
4-1
r 1
l/"°^ CO >^J" -^" CNJ CO CO -^|~ "~H
^cj" ^O vO -^" vO un CO *^i" CT^
' t -H OCNCNCN-^CNJO
CN ' i \£)roOr^oo^^O
oo O^ -«d" f^- O co CN -vj- r^-
i i
vO O O\0 cd
^ d r-l \
SO O O) i i d
d *H O) 0) O
r-l 4-1 4-1 CD 'I
.J cd en to 4-1
dl cj Ol cO
0) -H r-| > 4-1
d d >-i co d
60 -i-* 43 t-l t-t Ol
H ^D cO 3 dl E
CO d 4-1 4-* O 3 CO 00
o) a) 0,0 coi-iMc!
O 4-J 33COCX4JO-H
T3 d Vj)-idgc04JDH
co a) 01 4-icjco3do-H
to , t E co co PLI p-i h-i a n i
01 -H a,
O CO -r-l
!-i o) tr
p| f^| fTl
d
o
H
en
d
01
CX
H
3
ST
W
4-1
0) 0)
01 o
M 4-1
O 01
O _S
& r-^
d
CO 60
cO
&
CO
»-l
O)
o
CO
4-1
3
d
cd
U-l
o
CO
Ol
to
d
o
cx
tn
OJ
&,
cd
CO
r-l
O)
^
4-t
o
CO
4-1
3
d
CO
U-l
o
en
0)
en
d
O
OH
co
0)
rt
XI
r-~
C31
Q
w
O
co
4-145
-------
As noted in Appendix F an FGD system can be constructed with
minimal impact on the overall construction schedule of the power
plant. The increase in construction time is 6 months for a 3-year
schedule. However, an increase in construction personnel can elim-
inate this increase in construction time. While the elapsed time
needed to construct a plant is a function of manhours, the actual
number of men that can be used during any one stage of erection is
1 imited.
The survey of manufacturers also indicated that generally, they
are willing to guarantee 90 percent SC>2 removal and some are pre-
pared to guarantee better than 90 percent SC>2 removal, on a case-
by-case basis. Table 4-38 summarizes the terms of guarantees.
Seven out of 12 FGD manufacturers surveyed will guarantee
performance (availability of 90 percent) of the FGD systems they
market. Vendors of the double alkali system have enough confidence
in their systems to offer a 90 percent availability for the first
year of operation and 100 percent for the life of the plant (based on
a boiler operating rate of 70 percent for some of the new full-scale
utility applications) and these systems are for 85 to 95 percent
S02 removal of high sulfur coal (PEDCo, 1977).
All responding manufacturers would guarantee the costs of the
FGD system:
Four manufacturers would guarantee cost subject to an
escalation clause
4-146
-------
CO
ro
I
o
o
4J
c
ra
2
CT>
r-l
ic
S
O
E
VtJ
0s
V,
U-J
C
.-[
C>
>
J
J
H
UP
o
c^
A
dP
0
Oi
o
V
a
IS
c
19
Q.
0
U
C
O
V
01
c
<0
kl
w
3
C*
in
ki
0) «
U-4 -H
**-i m
O O
O
*> ti
IT,
CP r.
C (J
r* 1
M 1*1
rH .0
-*H (
3 a
n it)
M U
.
C
01
^
r<
tr
u
E
3
E
C
H
z
4$
CM
O
W
jj
flj
,»)
c
-r4
C
0
a
V
o
10
a
H
dp
O
o>
A
c
B 0) JJ X
W X 0 0)
0 4J 10 JS
«s a 4J
u c
rt 4)
r-l J= 01
rH *P 4J 03
ON 01
0> C 0
c "< o
* o k.
4J a
* o
in c. O
er> 3 A f.
4J
a> »o *c *
ft) CJ .t CJ C «»
*j CJ ^ 4) O O
C *J W jj Q, c\
*5 C ^3 C 3 *
14 (0 O*3
4 .-' | V4 £T> C.
3 O ^fl C GJ
IT 3 ,_Q -3 -H -*-1
"° i c c
l-j > !2.> Si «
0 <0 *fl 0> 3
S E <->= Q tr»
,
c
^
i-H
tJ^
1^1
( t
rH
C
g
k<
o
c
a
-r4
o:
01
4J
C
10
^i
n
CP
B
j;
£-t
U Q U b,
#
4J
B
a
0)
£
4J
C
-H
^
al
O
a
3
0
01
flj
4J
C
10
ll
fl
3
CT
OJ
>
10
=
(M
o
10
w o
kl 4J
01
C 0
> 0
in
c
oi n
> -H
O^ C
o
B -H
H 4J
<0
01 ki
01 4J
4J C
C 01
<3 0
kl C
0 O E
3 ci a
e* a
B 01 O
-C C 0
u
^) *
kl
* N
? *H
H
C-H
01 4-J
> 3
en
0)
0! -H
H
rH
0) W
01 O
4J 0
C
3 kl
ki 3
3 <"
3 «H
&* 3
01
01 |
J= 0
01
o>
4J
c
t,
10
3
CK
>,
r-l
1
O «rt
C CO
0 0
tH 4J
9
00
Jt CO
s
n
-r4
^0 Jj
O -1 B) -H
O\ <0 01 X
O 01 01
C O rH
« c
JS kl 01 -H
4J 3 01
IM 4J C
kl r^ C 0
01 3 10 -H
4J B kl 4J
4J 1 10 10
JS"& p-i
H C
01 O Cl
>" ki C C
1M O O
O -0 Cl
1 rH
0 rt 3 rM
4J O O O
1 3 W
o
Cl ki 4J E
kl 0' 3 Q.
u x .a a
a 4J
OI -H » O
^4 01 "O LTl £
a oi n
c c c 01
01 01 ki Q U
ki JS 3 £ 4J
< i J3 4J o]
c
ft)
> kl
T) 3
CT1 IM
rH
>i 3
r-l 01
C rH
0) 10 *>
> 3V
T« CD
CJi 30
01 n
oi -H n
C 01 JS
} 01 4J
kl 4J -H
10 C 3
3 0
C
01
J
«e
in
Ch
01
01
4J -O
C 01
10 C
kl kl
10 3
3 J3
tTl
0)
rS -rl
01
^ *~*
4J 10
O
0) O
01
in ki
O 3
Cl U-J
rH
>* 3
C 01
10 1
E JS
cr
C -H
M £
C
01 ki
> 3
CT-H
3
X 01
rH 1
r-l .C
10 D^
E -H
ki JZ
o
K ki
O
V)
rt 1
S
4) O -D
01 rH 01
4J C
C ki ki
fl 01 3
u .c .Q
e 41
3 -H ea
C/1 01 *r4
» C rH
*! 01 1C
.C .C O
f- .« 0
fc£
w
n
|Q
U
1
X
1
01
a
u
o
c
o
in
en
0
a
3
01
01
4J
C
fl
kl
0
3
pi 05
.rl
X 01
m a
z a
.
c
01
^
H
ty
c
01
4J
C
e
^t
Q
3
C*
E
3
E
H
C
32
j
en
k.
0)
3
Cl
C
U-l
c
a
E
B
4J
01
m
Cl
a
01
K
fO J
o
rH
QJ
£
4J
H
^
>
0
4J
C
kl
H
H
rH
01
D
01
W
U
0>
a
c
>*
c
>3
a
E
O
U
Q
O
H
w
CJ
r-l
D
O
4-147
-------
One manufacturer would negotiate the terms of the guarantee
None of the other manufacturers specified provisions.
Eight of the 12 manufacturers responding stated they would offer
operation and maintenance service. They further indicated that they
would issue a guarantee but did not specify the provisions.
4.4.7 Availability
System operating performance is discussed separately for each
FGD system description in Appendix D. This section describes oper-
ating availability of FGD systems in general.
Four parameters are commonly used in reporting FGD system
operating data:
1. Availability = hours the FGD system was available
for operation
hours in the period
2. Reliability = hours the FGD system was operated
hours the system was required to
operate
3. Operability = hours the FGD system was operated
hours the boiler was operated
4. Utilization = hours the FGD system was operated
hours in the period
To develop availability data that would be pertinent to the
utility industry, the following criteria were used:
1. The FGD system must treat flue gas from a utility generating
station greater than 50 MW.
2. The system has been operating for 1 or more years
3. The system is not a demonstration or test unit
4-148
-------
Unfortunately, availability data for plants fitting these criteria
were only available for seven systems, all of which are lime or
limestone systems. Performance data for these systems are presented
in Table 4-39.
The average modular availabilities showed a wide range among
systems. Five systems did show availabilities greater than 70
percent and three were greater than 88 percent. Four of the units in
the table had utilization factors of approximately 50 percent or
greater indicating the load in the FGD units has been large enough to
quantify operating history.
Figure 4-33 gives more comprehensive availability data on four
of the seven systems. While individual modular availability varied
over a wide range, the annual average availability for all the FGD
system was generally well over 80 percent. Furthermore, for 1977,
three of the four systems operated at better than 90 percent avail-
ability.
FGD system availability is dependent on both system design and
the manner in which the system is operated. There is a trend in
overall system availability, as a function of the year the system was
started up. Continuing improvement in availability is evident as the
newer, improved units come on line. Although some recent installa-
tions have not shown particularly high availability, a statistically
significant correlation does exist between startup date and average
availability. In addition, only one of these newer stations had a
4-149
-------
100
90-
80-
»*
co
<
70-
60-
50-
40-
30
-
-
-
MM
P
LD
cn
cr>
1
r-
r-^
cn
LaCYGNE
(
5.
4%
S)
r-.
en
ur>
CT>
k
>£>
o:
0
u.
u.
>
CM
l/l
Ul
_J
Q
O
n
MODULES
1
OL
0
U-
Ul
^
CT^
-
r-^
CTi
-
-
«
CHOLLA SHERBURNE BRUCE
(0.6% S) (0
.8%
S) MANSFIELD
( 3.3% S MAX)
ANNUAL AVERAGE
AVAILABILITY FOR
ENTIRE FGD SYSTEM
RANGE IN INDIVIDUAL
MODULE AVAILABILITY
Source: PEDCo, 1977d.
FIGURE 4-33
AVERAGE AVAILABILITY FOR SELECTED
FGD SYSTEMS
4-150
-------
cri
CO
w
p
rJ
w
o
<^
1
1
^^
O^ r i
M 5
1 Q
*^r
W
W c_)
rH- 5S
W -, FU
3 4-1
-a -H
O rH
g -H
0) cfl 60
60 rH d
cfl -H -H
rJ Cfl 4-10)
Oi > cfl 4-J
£> CN O
tt *3~ %5 *£> r-~-
vO r-^
r^ r^ i^
CJ "^^ '"-^^ f"^
O co ~-~
Crf r-H Pi 1 CO
1 LO 1
co r^ VD
!S r^ & -^ r-.
^ CN ~-~-
CO rH LO
QJ
^O v^ LO "O O>
*
CO O ^H C\l O
co ^o o*> r^- c>
LO in r^.
r~^« r^- r** r^ r~*-
r**- **^~, ^^-^ "^--. r^.
-\ O LO rn -^^
CO rH | | CO
1 | CO UP, |
^" co r~~* r~^ vO
r^ r ' -. '*"--. r^
--^ ~\ CN CN ~--
rH CO rH rH LO
CO
co co r^ LO ^D
r^> r^ "^-^ r . r^
~~~~ o ~-~. -~~-
CN r-^ rH c\ cn
, I
I QJ
d & "z -A d
QJ ex ed s-i
60 -H cfl 3
>^ rH rH d J3
O rH rH QJ (-1
H O QJ O)
Cfl ,d rd rJ ,d
rJ PH U O CO
00
*«£)
r^
vD
r^*
CN
rH
1
**o
r^
^^
LO
LM
CX)
.
r^
[-^
f-^
f-^
-- _
00
1
"^O
r-^,
--^
LO
vD
p^.
^^
<3"
0
LO
1
LO
^J.
LO
CN
00
1 1
.
o
z
T-J
rH
QJ
H
14-1
cn
d
cfl
g
QJ
CJ
3
r-l
pp
60
pi
H
C
H
rH
r^
o
cfl
4-J
cn
o
4-J
cn
Vj
H
cfl
CX
0)
r-J
4^ 0
a 4J
cfl
4-1 01
cn 3
T3
QJ
d r--
H r^
rH CT\
Q) rH
^ ^
O rH
4-1 3
I]
1 ^ I
r^- J2
rH !-l
CO
rH g
H
^ s
CX O
4H
Cfl -H rH
r^ cfl
r^ QJ ^H
cr. to
rH Cfl O
1 1 1 1
41 * '
3
d o d
o o
H rd -H
4-J . 4-1 4J
cO 0) C O
Li rH 0 3
o Td Td ^3 B TJ
CX r^- a) cfl QJ
i-t r~- 4-1 a CN r-l
O CT» r-1 "H
CJ rH O rH CO Cn
CX CX QJ QJ
0 - o) cx -a id
co O rJ cfl 3 3
H CJ rH rH
TJ P 4-J U CJ O
CO K! O O d d
Pi r-< ^ p^; h t hH
Cfl ,-Q CJ rrj (U L|_|
4-151
-------
spare module available for use in the event of malfunction of the
operating modules. A spare scrubbing module has a significant effect
on overall system availability, since it can replace a module that
may be shut down for any reason.
It should be emphasized that the major portion of the perform-
ance data is based on lime and limestone systems, and many second
generation FGD systems as well as newer lime/1imestone systems, are
expected to have greater availability. This concept is reenforced by
an increase on the part of manufacturers to guarantee performance
(both availability and removal efficiency) and some commitment for
sparing of scrubber modules. In any case, based on current operating
data, it should be possible for FGD systems to achieve near 90 per-
cent or better availability irrespective of fuel sulfur content.
Various measures have been and can be used to improve on main-
taining high FGD availability. Such items as separate maintenance
crews for the FGD, a dedication to operation, frequent cleaning and
inspection, and use of a successfully demonstrated design concept can
have a significant effect on FGD performance. Measures to improve
FGD availability can be categorized in terms of three areas: mainte-
nance methods, operating techniques and design concepts.
4-152
-------
5.0 DESCRIPTION OF THE EXISTING ENVIRONMENT
This section describes the industries that would be affected by
the proposed revision to the NSPS, and the air, water, land, and
energy conditions which relate to the NSPS prior to the revision.
5.1 The Electric Power Industry in the United States
The electric power industry is made up of many utility systems
that vary greatly in size, type of ownership and functions. The
industry is made up of investor-owned companies, non-Federal public
agencies, Federal agencies and cooperatives. The investor-owned or
private sector is by far the largest, accounting for 77 percent of
the 1.9 billion kWh of electrical energy generated in 1974 (U.S.
Federal Power Commission, 1976). Nearly all of the approximately
200 major investor-owned utilities operate integrated generation,
transmission and distribution systems.
Five Federal agencies* market electrical power generated at
facilities owned and operated by the Federal Government. The
Tennessee Valley Authority is the largest Federal electric system as
well as being the largest electric system in the nation. Electrical
energy marketed by the Department of the Interior is generated
hydroelectrically and, together with the energy generated by the
Tennessee Valley Authority, represents approximately 12 percent
*The Tennessee Valley Authority, Bonneville Power Administration,
Southwestern Power Administration, Southeastern Power Administration
and the Bureau of Reclamation. The Secretary of the Interior is the
marketing agent for power produced by all Federal power projects
except that produced by the Tennessee Valley Authority.
5-1
-------
of the total energy generated in the U.S. Public non-Federal electric
systems, including the systems of towns and cities, a few counties,
special utility districts and various state authorities, generate 9
percent of the total production. Electric cooperative systems supply
power in many of the rural areas of the country and account for 2
percent of the total production of electric energy.
5.1.1 Generating Capacity
The total electrical generating capacity presently installed in
the U.S. is 545,364 MW (U.S. Federal Energy Administration, 1977).
Coal-fired steam turbine generators make up 38 percent of the total
capacity; oil-fired units 25 percent; gas-fired units 14 percent;
hydroelectric units 12 percent; nuclear units 9 percent and other
units, including units of unrecorded types, 3 percent.
Assuming no revision of the present NSPS, long range forecasts of
future additions to generating capacity have been developed as part of
a study (Teknekron, 1978a) sponsored by the U.S. Environmental Protec-
tion Agency. According to these forecasts, the distribution of capacity
additions by energy source, or the mix of additions, will be sensitive
to the rate of growth in installed capacity that will be required to
meet increasing demand for electrical energy. If a moderate rate of
growth is experienced between 1977 and 1995, nuclear units will
constitute 59 percent of total additions and coal-fired units 36
percent. In a high growth scenario the corresponding figures are 45
percent for nuclear units and 47 percent for coal units. Installed
5-2
-------
capacity in the U.S. by 1995 would reach 1,081.5 GW (35 percent
nuclear, 37 percent coal) with moderate growth and 1,311.7 GW (31
percent nuclear, 44 percent coal) with high growth. These scenarios
underlie the analysis of impacts presented in later sections of this
report. A description of the scenarios and derivation of the above
forecasts is given in Appendix J.
5.1.2 Production of Electrical Energy
The total net production of electrical energy in the U.S.
reached 2,036.5 billion kWh in 1976 (U.S. Federal Power Commission,
1977). Approximately 46.3 percent of the energy was derived from
coal, 12.7 percent from oil, 14.5 percent from gas, 9.4 percent from
nuclear fuel and 0.2 percent from other sources. During this year,
the electric utility industry consumed more than 448.1 million tons*
of coal (U.S. Federal Power Commission, 1977).
Current projections (U.S. Federal Power Commission, 1977b;
Edison Electric Institute, 1977) indicate that coal will continue to
supply 47 to 48 percent of the primary energy from 1977 to 1986. On
the basis of data pertaining to 1975, the U.S. Federal Power Commission
(1977b) estimates that 111,600 MW of new coal-fired generating
capacity will be brought on line by the utility industry between 1976
and 1985 (U.S. Federal Power Commission, 1977). Electrical energy
derived from coal would amount to 1,227 billion kWh in 1980 and 1,690
billion kWh in 1985. Later projections based on actual data for 1976
*0ne ton = 0.9842 metric tons.
5-3
-------
(Edison Electric Institute, 1977) are in substantial agreement with
the forecasts of the Federal Power Commission. For example, the
amount of electrical energy that will be derived from coal, according
to these later projections, is predicted to be 1,230 billion kWh in
1980 and 1,595 billion kWh in 1985 (Edison Electric Institute, 1977).
5.1.3 Supply of Coal
Short range forecasts based on the present NSPS indicate that
consumption of coal by the electric utility industry will increase
from the level of 406 million tons* in 1975 and reach 570 million
tons in 1980 and 770 million tons in 1985 (U.S. Federal Power Com-
mission, 1977). Historical patterns of coal demand and supply are
expected to change over the decade. Departures from these patterns
have occurred for several reasons, among which is the promulgation
in 1971 of the existing NSPS SO emissions for coal-fired steam
generators (U.S. Federal Power Commission, 1977). The present stan-
dard has accelerated the procurement of low-sulfur coal to fuel the
generating units required to comply with the standards. Included in
this category is the major portion of coal-fired additions scheduled
to become operational on or before 1985, although some of the later
units could become subject to the revised standard. Other factors
expected to increase U.S. reliance on coal over the next several
years include the anticipated substitution of coal for oil under
the provisions of the Energy Supply and Environmental Coordination
*0ne ton = 0.9842 metric tons.
5-4
-------
Act of 1974 and the phase out of natural gas in areas where it is now
a primary source of energy (U.S. Federal Power Commission, 1977).
The increased demand for low-sulfur coal to supply new coal-fired
units is expected to affect primarily the development of coal resources
in western states. It is anticipated that more than 55 percent of
the demand created by new coal-fired units coming on line between
1976 and 1985 will be supplied by coal originating in the western
states of Arizona, Colorado, Montana, New Mexico, North Dakota, South
Dakota, Utah and Wyoming (U.S. Federal Power Commission, 1977). As a
consequence, projections indicate coal from these states will supply
approximately 35 percent of the total coal needs of the electric
utility industry, increasing from a level of 16.9 percent in 1975. A
high rate of growth in the production of steam coal is expected to
prevail in the Western Midwest Region (Bureau of Mines Districts 12,
14, and 15). Approximately 15 percent of the demand created by new
coal-fired units is expected to be met from resources in this region,
raising its share of the total production of steam coal from 3.9
percent in 1975 to 10 percent in 1985. The supply of coal from other
coal-producing regions is expected to increase over the same years,
although the relative share of the overall production of steam coal
originating in each of these regions is expected to decline (U.S.
Federal Power Commission, 1977).
Requirements for coal to fuel new units scheduled for service
between 1976 and 1985 are expected to vary markedly throughout the
5-5
-------
country. Details of the projected demand attributable to new units
in the various regions of the country are given in Table 5-1. As the
data indicate, the largest incremental demands will be in the West
South Central and West North Central states. The rate of growth is
expected to remain high throughout the decade in the West South
Central states, whereas a moderation in the rate of growth is expected
to prevail in the West North Central states in the earlier part of
the period. Rapid growth in demand between 1980 and 1985 is expected
in the South Atlantic and East South Central states.
5.1.4 Origin and Destination of Coal for New Units
The Appalachian Region has been the traditional major source of
coal for the nation's electric utility industry. For reasons cited
previously, new supply patterns are forming in the case of coal for
generating units scheduled to become operational between 1976 and
1985. A relatively small fraction, 14.5 percent, of the quantity of
coal needed to fuel these units is projected to be of Appalachian
origin (U.S. Federal Power Commission, 1977). Based on the existing
NSPS, the region would remain as the largest source of steam coal in
1985, but its share of the total supply would decrease from 44.8
percent in 1975 to an estimated 35 percent in 1985. The entire
incremental production of coal from Appalachia would be consumed by
new units in the East.
New units in the West will depend almost entirely on bituminous
and subbituminous coal and lignite produced in states west of the
5-6
-------
TABLE 5-1
INCREMENTAL COAL DEMAND IN 1980 AND 1985 ATTRIBUTED TO NEW UNITS
SCHEDULED FOR OPERATION BETWEEN 1978 AND 1985
REGIONS/STATES
New England
(Connecticut, Maine, Massachusetts,
New Hampshire, Rhode Island, Vermont)
Mid Atlantic
(New York, Pennsylvania,
New Jersey)
East North Central
(Illinois, Indiana, Michigan, Ohio,
Wisconsin)
West North Central
(Iowa, Kansas, Minnesota, Missouri,
Nebraska, North Dakota, South Dakota)
South Atlantic
(Delaware, Florida, Georgia, Maryland,
North Carolina, South Carolina, West
Virginia, District of Columbia,
Virginia)
East South Central
(Alabama, Kentucky, Mississippi,
Tennessee)
West South Central
(Arkansas, Louisiana, Oklahoma,
Texas)
Mountain
(Arizona, Colorado, Idaho, Montana,
Nevada, Utah, Wyoming)
Pacific
(Oregon, California, Washington)
TOTAL
INCREMENTAL COAL DEMAND
IN THOUSANDS OF TONS
1980
1985
10,300
24,214
40,025
6,711
9,076
57,113
26,085
400
173,924
18,100
47,004
67,330
23,792
23,249
124,207
52,889
1,200
357,771
Source: U.S. Federal Power Commission, 1977.
5-7
-------
Mississippi River if the NSPS for coal-fired utility boilers remains
at 520 ng/J (1.2 Ib S02/106 Btu). These units represent the largest
incremental demand for coal for two reasons: the large fraction (58.8
percent) of new coal-fired capacity scheduled to be installed in the
western states, and the lower average heat content of western coal.
An estimated 68.7 percent of the total incremental production of coal
will be required in these states with the remainder (31.3 percent)
crossing the Mississippi River to eastern destinations.
Details of the projected origin and destinations of coal to
supply new units in 1980 and 1985 are given in Table 5-2 and graphical
displays of coal movements in these years are shown in Figure 5-1 and
5-2. The importance of the western regions of the Northern Great
Plains is underscored by the information presented in Figure 5-1. As
indicated, most of the incremental coal production in the region will
be delivered to adjacent areas of the country, particularly to the
West North Central and the West South Central regions. This region
will supply 141 million tons of coal annually, or 39.4 percent of the
overall incremental demand by 1985, with relatively small quantities
of coal reaching as far as the South Atlantic Region (U.S. Federal
Power Commission, 1977).
Figure 5-2 shows that the incremental production of coal in all
of the other coal-producing regions will be used primarily to satisfy
the demand generated by new units within the region (U.S. Federal
Power Commission, 1977). Coal from the four states in the Mountain
-------
TABLE 5-2
PROJECTED MOVEMENT OF COAL FOR NEW UNITS
SCHEDULED TO BECOME OPERATIONAL BETWEEN 1976 AND 1985
REGION STATE
OF OF
ORIGIN DESTINATION
East Alabama
Florida
Delaware
Georgia
Kentucky
Ohio
Michigan
Massachusetts
North Carolina
Pennsylvania
South Carolina
Eastern Midwest Florida
Georgia
Illinois
Indiana
Iowa
Kentucky
( Michigan
Missouri
Oklahoma
Western Midwest3 Iowa
Missouri
Oklahoma
Texas
West Arizona
Arkansas
Colorado
Georgia
Idaho
Indiana
Iowa
Kentucky
Louisiana
Michigan
DEMAND IN
1980
1,980
0
800
317
1,800
4,250
1,650
506
1,410
10,300
463
23,476
987
1,234
2,485
6,640
20
3,010
0
4,800
4,046
23,222
20
350
0
21,419
21,789
5,885
7,260
5,415
0
0
3,700
5,081
6,093
2,500
1,099
1,000 TONS
1985
6,398
651
800
6,517
6,275
10,000
2,230
1,272
4,464
12,200
1,407
52,214
987
1,234
6,490
11,700
270
7,524
420
4,500
3,001
36,358
20
1,250
900
51,845
53,655
5,885
12,960
5,905
1,000
1,600
3,700
7,731
11,690
6,000
5,499
5-9
-------
TABLE 5-2 (Concluded)
REGION STATE
OF OF
ORIGIN DESTINATION
West Minnesota
(Continued) Massachusetts
Montana
Nebraska
Nevada
North Dakota
New Mexico
Oklahoma
Oregon
Texas
Utah
Wisconsin
Wyoming
Unknown Arizona
Colorado
Florida
Illinois
Missouri
New Mexico
New York
Ohio
West Virginia
TOTAL U.S.
DEMAND
1980
5,500
1,780
2,000
4,411
365
8,700
3,220
6,000
400
14,888
2,400
3,730
4,600
91,077
2,200
0
0
660
0
0
0
0
1,500
4,360
173,924
IN 1,000 TONS
1985
9,400
1,780
7,000
6,770
8,746
18,100
5,450
7,303
1,200
38,849
5,238
6,470
7,100
197,684
2,200
1,730
2,000
345
1,500
1,035
5,900
150
3,000
17,860
357,771
aNo shipments anticipated from Arkansas, Iowa, or the Oklahoma
counties of Haskell, LeFlore and Sequoyah.
Source: U.S. Federal Power Commission, 1977.
5-10
-------
5-11
-------
5-12
-------
RegionArizona, Colorado, New Mexico and Utahwill supply 10.8
percent of the coai requirements for new units by 1985. Most of
this coal will remain within the area, where many of the new units
will be located close to producing mines. The demand tor coal from
the western Midwest (Bureau of Mines Distiict 15, including Kansas,
Missouri, Texas and part of Oklahoma) is projected to increase
rapidly. Coal from this region will supply 15 percent of coal demand
for all new units by 1985; and 96 percent of the incremental produc-
tion in this area will be consumed in the state of Texas. As men-
tioned previously, 14.5 percent of incremental demand will be met by
Appalachian coal, with deliveries remaining within the eastern
region.
5.1.5 Long-Range Projections
Long-range projections based on the existing standard indicate
that growth in national coal production beyond 1985 will depend
strongly on the demand for coal generated by the electric utility
sector (ICF, Inc., 1978). In a scenario of moderate growth in demand
for electrical energy (5.8 percent annually through 1985 and 3.4
percent beyond 1985), consumption of coal by utilities is forecast
to reach 1 billion tons per year by 1995. In a high growth scenario
(5.8 percent annually through 1985 and 5.5 percent beyond 1985) the
corresponding projection would be 1.4 billion tons. Pertinent details
of predicted national coal production are shown in Table 5-3.
5-13
-------
TABLE 5-3
PREDICTED NATIONAL COAL PRODUCTION
National Coal Production
(Millions of Tons)
out: licit j.u
Moderate Growth
Nonutility coal
Utility coal
Total production
High Growth
Nonutility coal
Utility coal
Total production
1975
223
424
647
223
424
647
1985
416
802
1218
416
802
1218
1990
611
973
1584
611
1157
1768
1995
763
1000
1763
763
1438
2201
Source: ICF, Inc., 1978.
Regional production forecasts indicate that the greater part of
incremental production through 1995 will be in the West, especially in
the Northern Great Plains. The substantial growth in the West is attri-
buted to the: (1) vast resources of coal that can be produced at low
prices, (2) availability of coal that can be burned without FGD while
meeting existing standards, (3) projected growth in consumption within
the western region, and (4) shipment of coal across the Mississippi
River.
5-14
-------
Western coal consumed in the east is forecast to be primarily
subbituminous low-sulfur coal from Montana and Wyoming and, to a
lesser degree, bituminous coal from Colorado. Predictions show
that these coals will be consumed mostly by the utility sector in
the states east of the Mississippi River but west of the Appalachian
Mountains. The increase in shipments is attributed to new power
plants that would be required to meet the existing standard and older
plants switching to low sulfur western coals in response to state
implementation plans. It is predicted that Appalachian production
will remain fairly level because resources of low-sulfur coal in the
region, particularly in Central Appalachia, are limited and expen-
sive to exploit; and low sulfur coals produced in this region are
generally used for metallurgical purposes. The growth in production
of Northern Appalachian high-sulfur coal is expected to be modest
because of the limited increase in demand for this coal generated by
growth in the utility sector and the large increase in nuclear
capacity in the geographic markets of Northern Appalachia coal.
Growth in coal production in the Midwest is predicted because coal
from this area, although high in sulfur content, can be mined and
transported to many large markets at competitive prices. Details of
projected regional coal production in a high growth scenario are
presented in Table 5-4.
5-15
-------
TABLE 5-4
REGIONAL COAL PRODUCTION IN HIGH GROWTH SCENARIO
Region
Northern Appalachia
Central and Southern Appalachia
Midwest and Central West
Northern Great Plains
Rest of West
National Total
Western Coal Consumed in East
Source: ICF, Inc., 1978.
5.2 Coal Resources of the United
Regional Coal Production
(Millions of Tons)
1975
179
218
151
55
44
647
21
States
1985
172
236
243
424
143
1218
206
1990
205
237
298
810
218
1768
455
1995
223
241
331
1160
247
2201
601
The vast resources of coal in the U.S. have been identified by
the U.S. Geological Survey. Deposits at depths of less than 3,000
feet contain about 1,700 billion tons (U.S. Department of the Interior,
Bureau of Mines, 1977). Based on geological knowledge and theory it
is thought that an additional coal resource of even larger size
exists. Estimates of the quantities of coal in relatively thick beds
and formations that are amenable to mining by conventional surface
and underground methods have been developed by the U.S. Department of
5-16
-------
the Interior, Bureau of Mines. From information available on January
1, 1976, the Bureau estimates that the demonstrated reserve base* of
coal in the U.S. is 438 billion short tons.
The reserve base refers to coal that is technically and economi-
cally minable under present conditions. It is not a fixed quantity,
but one that increases with discovery and additional development,
decreases with mining, and changes as the criteria underlying the
estimate of its extent are modified.** No consideration is given
to factors affecting the marketability of specific coals in computing
the reserve base. Nevertheless, the criteria applicable to bed
thickness and depth generally correspond to the characteristics of
deposits presently being mined.
*The reserve base is composed of all coals in deposits that meet
certain criteria related to the thickness and depths of the deposits.
Criteria applicable to thickness are: 28 inches or more for
bituminous coal and anthracite, 60 inches or more for subbituminous
coal and lignite. Deposits of all ranks at depths greater than
3,000 feet from the surface are excluded from the reserve base, and
only the lignite beds that can be mined by surface methods are
included. These beds occur generally at depths no greater than 200
feet. Certain coal beds that do not meet the depth and thickness
criteria are included in the reserve base because they are presently
being mined or could be mined commercially at this time. The term
"demonstrated" denotes both measured and indicated categories as
defined by the Geological Survey and the Bureau of Mines. Coal
deposits are included in these categories where, on the basis of
geological projections and engineering evaluation, there is a high
degree of certainty regarding their existence (U.S. Department of
the Interior, Bureau of Mines, 1977).
**The current estimate of 438 billion tons of coal as the reserve base
represents an increase of 1.5 billion tons over the estimate derived
previously from information available on January 1, 1974 (U.S. Depart-
ment of the Interior, Bureau of Mines, 1975).
5-17
-------
5.2.1 Geographical Distribution of Coal Deposits
The reserve base of 438 billion tons of coal is distributed
widely throughout the U.S. with 46 percent of the base found in
states east of the Mississippi River, and 54 percent in western
states and Alaska (U.S. Department of the Interior, Bureau of Mines,
1977). Quantities of coal of different rank as well as quantities
amenable to production by underground and surface mining methods
differ markedly in the various coal rich areas of the country. An
evaluation of coal deposits by rank shows that 52 percent of the
total reserve base is composed of bituminous coal, 38 percent sub-
bituminous coal, 8 percent lignite, and 2 percent anthracite.
Approximately 85 percent of the bituminous coal and virtually all of
the anthracite are found east of the Mississippi River. Most of the
subbituminous coal and lignite is found in the West.
One-third of the reserve base (141 billion tons) is in beds so
close to the surface or in beds so thick that underground mining is
impractical (U.S. Department of the Interior, Bureau of Mines, 1977).
Of this quantity, nearly three-quarters is in states west of the
Mississippi River. Over one half of the coal that can be mined by
underground methods is in states east of the Mississippi River.
States with the largest coal reserves, ranked in order, are Montana,
Illinois, Wyoming, West Virginia and Pennsylvania. These five states
contain 71 percent of the nation's available coal. Deposits in
Montana and Wyoming, represent 40 percent of the total reserve base
5-18
-------
and consist principally of lower rank subbituminous coals. Coals of
the three eastern states are all of bituminous or higher rank.
Pertinent details of the reserve base of coals in the U.S. are given
in Table 5-5.
The fraction of coal that can be recovered from the reserve
base is the "reserve" or the "recoverable reserve." Recoverability
varies within a range of 40 to 90 percent of Lhe reserve depending on
the characteristics of the coal bed, the mining method, and restraints
on mining a deposit imposed by natural and man-made features and
restrictions. Mining experience in the U.S. indicates that at least
one half of the U.S. reserve base of coal may be recovered. Estimates
of recoverable reserves have been derived by applying a recoverability
factor of 50 percent to underground deposits and 85 percent to
surface deposits.* The recoverable reserves of coal in the U.S.
amount to 270 billion tons; 150 billion tons in the western states
and 120 billion tons in the eastern states. Details of the reserve
base and recoverable reserves by state are given in Table 5-5.
*A study of 200 underground mines shows an average recoverability of
57.0 + 1.7 percent (Lowrie, 1968). With respect to the coal reserve
base, average recovery by underground mining methods is expected
to be about 50 percent owing primarily to coal left unmined to
support the surface (U.S. Department of the Interior, Bureau of
Mines, 1975). Recovery of coal by strip mining depends primarily on
the ratio of the thickness of the overburden to that of the coal
bed. Local topography is another factor that affects the recover-
ability of coal. Recovery, depending on the type of mining (contour
stripping or area stripping) is expected to range between 80 and 90
percent (U.S. Department of the Interior, Bureau of Mines, 1975).
5-19
-------
CO
<^
H
C/>
P
M
^
W
PC
H
Z
t I
rJ
pi
w
W cn
! 1 [V|
pq fin
W
cn
W
Pi
cn
W
H
H
cn
§
W
H
CO
W
CO
[>1
E-i
c/n
^_,
w
CO
w
3
W P-
i-3 C
^rQ r.
-i, <
> ^ h
O t-
0 w -
g S t
CO
f, 5
w g
^
M 0
^
w ^
P5 f >
w
C/} T
W Jj
a
cd
H
H
K P
(~1_C ^
i-Q'
"^ r ^
W^* f
c£
O .-« h
0 w >
pj £~3 }-
S I*
Tf\
** ^
W H
« 0
W r/,
&1
w S
w 2
W .q
Pi r 1
r~1
a
cd
W
H
^
f~^
Cn
^
D
0 cn
3 g
H 2
J H
J
H
hJ
U
inrH
r^jHo-t^-o-J-cTi-d-r^ooc^
^D ~,>^ra > cu ni t-i
C8 -H CO ^ A! C rH CO -H -H
cd 0 cd 3 3 < 1 CO CU *H
43-H-H4J4J >,O C C 6C4J
Cd rH T3 Cd (S ^-) «rl C P* J-l CO
rH rH (S 0) 0) Cd ,£< QJ CU *H CU
rS
ooooooooooo
ooomoomoooovo
COCNCMincOOOCOOCNvDC^
C^ ON rH CO 00 CO 00
r-^
CNinCN^CNCNOC^. O in o o> t i ^ o> ^ i i
rH CO CO CM CN O> O m in ' 1 ^O
f\ r.r. r. «.*».«>
VO vDCN inC- ^Q Q^ Q*\ T\ rH IrH
in CN rH rH O CN rH
i i r^
cd
4J
o o
cj .y
CO O -H -H Cd cd
cd cd T3 &-t cd x P 6
cdcwcd C03CQ) o
r^J OC^-I CdOCd^^ (~]
cnNcdocdcncn4-i 4-icd
Cd 'H ,y i 1 |5 C3 CO PJ ^ M i t
i 1 J-iL-iOOcd*HOCUO ^
<< i i r
CN CO v.
*-d~ i 1 c
M
c
x.
0
1 1 r
VI
cd
4-1
o
cd
r£ CO
4-1 Cd
3 X
O CU
cn H C
O
o
0
0
CM
ro
i-^
co
00
3
o
rH
vD
CO
rH
-i oo r-~ oo
o in co co
»s »- »\
D t i m co
m m
CM
^ in r^ ^
D s s
-a cu
c o
« cd cd
cn ^ M
c! cn 3
o £ cn
4-1 O
4-1 MH
C 0
r-- O (S
r-. -HO 4J
0^ rH -H C
rH t-H rH CU
H I-H a
E -H r4
o) sr p,
fi . ^
H sr m
^ ^ rH 00
CO
IH O >-' CO
O 43 3
Cfl Cd rH
,3 T) C! ft
cd M -H
CU rH 01
s-i o cn
3 ^ M cd
M CO Cd rQ
C 0
O 01
cn 4J 42 >
P CM CU
O O cn
rt 53 CU
H rH )-l
O rH >
H -H s-^ 13
M e cu c
01 CO 3
4-J C7> Cd O
Ir-i O 60
CU r4
CU CU £> CU
i-C CO M 'O
4J cd a) C
43 CO 3
O CU M CO UH
t> C! O
4-1 r4 M3 O
0 CU 0) 4-1 4-1
ni rn r-* f*
tU UJ p C
6
W 3 cn O 3 l-i
U rH -H W) ft CU
PH a 3 cu B to
P X O rJ O CU
O W i-4 O CJ M
-------
5.2.2 Sulfur Content of U.S. Coals
Practically all coals contain sulfur found in one or more of
three basic forms: organic combinations as part of the coal substance,
inorganic pyrites or marcasite compounds, and sulfates (Lowry, 1963).
Coals in the U.S. vary markedly in both the fractional content by
weight of sulfur and the chemical form in which the sulfur is present.
The pyritic sulfur and total sulfur content of coals is highest in
bituminous coals of Pennsylvania age in the Appalachian and Interior
coal basins (Averitt, 1973). Subbituminous coals and lignite of the
Rocky Mountain and Northern Great Plains regions are characterized by
a relatively low content of pyritic and total sulfur. The fraction
of sulfur present in coal as sulfates, mainly sulfates of calcium and
iron, rarely constitutes more than a few hundredths percent of the
weight of coal (Lowry, 1963).
Coal with a sulfur content of 1 percent or less by weight is
generally referred to as low-sulfur coal. Medium-sulfur refers to
a sulfur content in the range of 1.1 to 3.0 percent, and high-sulfur
refers to a sulfur content in excess of 3.0 percent (U.S. Department
of the Interior, Bureau of Mines, 1975). On the basis of these
definitions, 46 percent of the total reserve base of the United
States is identified as low-sulfur coal, 21 percent as medium-sulfur
coal, and an additional 21 percent as high-sulfur coal. The sulfur
content of 12 percent of the reserve base is undetermined.
5-21
-------
The major portion, or 84 percent, of the reserve base of low-
sulfur coal is found in states west of the Mississippi River. (U.S.
Department of the Interior, Bureau of Mines, 1975). Roughly 40
percent of all low-sulfur coal is amenable to surface mining, with
the bulk of surface deposits again being in the West. In the East,
only 16 percent of the reserve base of low-sulfur coal is recoverable
by surface mining techniques. With respect to rank, 22 percent of
the reserve base of low-sulfur coal is composed of bituminous coals
and anthracite, and 78 percent is composed of subbituminous coals and
lignite.
States with the largest quantities of low-sulfur coal are
Alaska, Montana and West Virginia (U.S. Department of the Interior,
Bureau of Mines, 1975). Montana with a reserve base of 102 billion
tons of low-sulfur coal contains 51 percent of the base, West Virginia
7 percent, and Alaska 6 percent. Virtually all of the Montana and
Alaska coals are of low rank, whereas all of the West Virginia
reserve base is composed of high rank bituminous coal. An estimated
29 percent of the coal reserve base consists of coals with a sulfur
content of 0.7 percent or less; and 17 percent of the base consists
of coals with a sulfur content of 0.5 percent or less. The major
portion of these coals is located in western states, principally
Montana and Alaska. In the East, the reserve base of coal with a
sulfur content less than 0.7 percent is estimated to be 8 billion
tons, representing 6 percent of the national reserve base of such
5-22
-------
coal. Pertinent details of the characteristic sulfur content of U.S.
coals are given in Table 5-6. The quantities shown in this table
correspond to estimates of the reserve base derived from information
available on January 1, 1974. These may differ somewhat from the
later estimates of the reserve base given previously in Table 5-5.
5.2.3 Coal-Producing Regions
In various analyses presented throughout this statement, states
with substantial reserves of coal are grouped into four regionsthe
Eastern, Eastern Midwest, Western Midwest, and Western. Individual
states are included in a particular region on the basis of geographical
location and pertinent characteristics of the coal reserve base. The
state of Kentucky is divided into Kentucky-east and Kentucky-west;
and these segments are included in the Eastern and Eastern Midwest
regions, respectively. The states making up each of the four regions
are listed in Table 5-7.
5.3 Air Quality
5.3.1 Ambient SC>2 Concentrations
A comparison of 1975 ambient S02 monitoring data with data
for recent years shows that SC>2 concentrations in urban areas have
decreased by an average of 30 percent since 1970. Sulfur dioxide
levels improved rapidly in the 1970-1973 period and then leveled off
as many areas came into compliance with the National Ambient Air
Quality Standards (NAAQS) for S02 (primary: 80 p.g/m^annual
arithmetic mean; 365 (j.g/m^maximum 24 hour concentrations not to
5-23
-------
TABLE 5-6
RESERVE BASE OF COAL IN THE UNITED STATES, BY STATE AND SULFUR CONTENT
STATE*
WESTERN STATES
Alaska
Arizona
Arkansas
Colorado
Iowa
Kansas
Missouri
Montana
New Mexico
North Dakota
Oklahoma
South Dakota
Texas
Utah
Washington
Wyoming
TOTAL
EASTERN STATES
Alabama
Illinois
Indiana
Kentucky-east
Kentucky-west
Maryland
Ohio
Pennsylvania
Tennessee
Virginia
West Virginia
TOTAL
UNITED STATES
RESERVE BASE IN MILLIONS OF TONS
BY SULFUR CONTENT IN PERCENT
<1.0
11,458.4
173.3
81.2
7,475.5
1.5
0
0
101,646.6
3,575.3
5,389.0
275.0
103.1
659.8
1,968.5
603.5
33,912.3
167,305.0
624.7
1,095.1
548.8
6,558.4
0.2
135.1
134.4
7,318.3
204.8
2,140.1
14,092.1
32,852.0
200,181.1
1.1-3.0
184.2
176.7
463.1
786.2
226.7
309.2
182.0
4,115.0
793.4
10,325.4
326.6
287.9
1,884.6
1,546.7
1,265.5
14,657.4
37,540.6
1,099.9
7,341.4
3,305.8
3,321.8
564.4
690.5
6,440.9
16,913.6
533.2
1,163.5
14,006.2
55,381.2
92,997.6
<3.0
0
0
46.3
47.3
2,105.9
695.6
5,226.0
502.6
0.9
268.7
241.4
35.9
284.1
49.4
39.0
1,701.1
11,244.2
16.4
42,968.9
5,262.4
299.5
9,243.9
187.4
]2,634.3
3,799.6
156.6
14.1
6,823.3
81,406.4
92,671.1
UNKNOWN
0
0
74.3
6,547.3
549.2
383.2
4,080.5
2,166.7
27.5
15.0
450.5
1.0
444.0
478.3
45.1
3,060.3
18,322.9
1,239.4
14,256.2
1,504.1
2,729.3
2,815.9
34.6
1,872.0
2,954.2
88.0
330.0
4,652.5
32,476.2
50,837.7
TOTAL**
11,645.4
350.0
665.7
14,916.5
2,884.9
1,388.1
9,487.3
108,396.2
4,394.8
16,003.0
1,294.2
428.0
3,271.9
4,042.5
1,954.0
53,336.1
234,458.6
2,981.8
65,664.8
10,622.6
12,916.7
12,623.9
1,048.2
21,077.2
31,000.6
986.7
3,649.9
39,589.8
202,162.2
436,725.5
SOURCE: U.S. Department of the Interior, Bureau of Mines, 1975.
*Excludes Georgia, Idaho, Louisiana, North Carolina and Oregon
**Totals may not correspond exactly to the sum of entries because of
rounding errors.
5-24
-------
TABLE 5-7
COAL PRODUCING REGIONS OF THE UNITED STATES
OTHER
REGION DESIGNATION
Eastern Northern Appalachia
Southern Appalachia
and Alabama
Eastern Interior Basin
Midwest
Western Bureau of Mines
Midwest Districts 12, 14,
and 15
Western Northern Great Plains
The Rockies and
The Pacific
United States
COAL RESERVE
BASE
STATES MILLIONS OF
ENCOMPASSED TONS
Alabama
Kentucky-east
Maryland
Ohio
Pennsylvania
Tennessee
Virginia
West Virginia
TOTAL
Illinois
Indiana
Kentucky-west
TOTAL
Arkansas
Iowa
Kansas
Missouri
Oklahoma
Texas
TOTAL
Alaska
Arizona
Colorado
Montana
New Mexico
North Dakota
South Dakota
Utah
Washington
Wyoming
TOTAL
2,000
8,300
570
12,000
16,000
600
2,500
21,000
62,970
39,000
6,000
7,600
52,600
250
1,300
850
3,800
960
2,700
9,860
3,300
280
9,500
78,000
3,200
8,600
360
3,400
960
36,000
143,600
270,000
5-25
-------
be exceeded more than once per year; secondary 1300 fag/mmaximum
3-hour concentration not to be exceeded more than once per year). The
available data indicate that SC>2 levels were relatively stable for
the nation as a whole during 1975. Trends in S02 appear to have
leveled off or in some cases increased slightly, apparently because
of the failure or inability to use clean fuels in some areas of the
country or the installation of more advanced control measures in an
effort to meet air quality standards (U.S. Environmental Protection
Agency, 1976).
The status of compliance with ambient air quality standards
for 862 is shown in Figure 5-3. This map should not be used to
determine nonattainment areas because it has been created to reflect
the highest measured ambient concentrations in the 3-year period
(1974-1976). More current information may show that the counties are
or are not violating standards. This figure was prepared using data
that were available from the U.S. Environmental Protection Agency's
National Aerometric Data Bank (NADB) in September 1977. These data
were supplemented with updated information from state reports provided
by the EPA Regional offices.
The second maximum 24-hour average measured in the country in
the period 1974-1976 was used as the summary statistic. This average
relates to the short-term 24-hour average standard of 365 (j.g/m , which
is not to be exceeded more than once per year. This was used instead
of the annual mean, which could be compared with the SC>2 annual mean
5-26
-------
w
Q
M
X
o
O
u
CO
H
OT
5-27
-------
primary NAAQS, because many SC>2 monitors did not collect sufficient
data to meet the NADB validity criteria for calculating an annual
mean. The criteria require that at least 75 percent of the total
possible data be available to calculate an annual mean. Further, the
24-hour aveage NAAQS is more likely to be violated than the annual
standard.
Data are available for most of the nation except some areas of
the West and Northwest. An examination of the sulfur dioxide map
indicates that most areas are not showing violations of the short-
term NAAQS. Of the 834 counties with S02 data, 60 had second
maximum 24-hour averages violating the 24-hour primary standard.
Areas not in compliance are generally in industrial areas of the
midwest and the western part of the country where the principal
sources are smelting operations.
5.3.2 SO? Emissions
In 1976, the SC>2 emissions from electric utilities were esti-
mated to be 13.6 million metric tons, which represented about 64 per-
cent of the estimated national total of S02 emissions (Teknekron,
1978; EPA 1976).
An Integrated Technology Assessment computerized model (Tek-
nekron, 1977) was used to project 862 emissions from all operating
utility steam generators. This model calculated the total national
and regional S(>2 emissions from these sources and the results are
presented in Table 5-8 and Figure 5-4.
5-20
-------
TABLE 5-8
REGIONAL S02 EMISSIONS
(Million Metric Tons Per Year)
Region
1985
1990
1995
2000
a. Moderate-Growth Rate
NE
MA
SA
ENC
ESC
WNC
WSC
NM
SM
PA
Nation
0.25
1.70
3.62
3.96
2.44
1.61
0.95
0.09
0.23
0.31
15.2
0.31
1.66
3.53
3.79
2.38
1.58
1.52
0.13
0.30
0.31
15.5
0.25
1.62
3.73
3.69
2.28
1.70
1.73
0.18
0.28
0.33
15.8
0.26
1.69
3.80
3.61
2.00
1.83
1.91
0.24
0.28
0.29
15.9
b. High-Growth Rate
NE
MA
SA
ENC
ESC
WNC
WSC
NM
SM
PA
Nation
0.25
1.67
4.04
4.^9
2.43
1.56
0.94
0.07
0.23
0.32
15.9
0.26
1.62
4.19
4.75
2.28
1.63
1.84
0.14
0.31
0.43
17.5
0.32
1.72
4.67
5.45
2.26
1.93
2.71
0.28
0.38
0.60
20.3
0.44
1.87
5.18
6.14
2.44
2.29
3.49
0.42
0.56
0.90
23.8
Source: Teknekron, 1978.
5-29
-------
These results lead to the following conclusions about total
national emissions from electric power generation assuming no re-
vision to the NSPS.
National emissions of SC>2 from electric power plants
will increase from 1976 partial compliance levels
(about 13.6 million metric tons) at about 2 percent
per year until 1985, if electricity demand grows at
5.8 percent per year until 1985.
Under moderate demand growth and the present emission
standards national SC>2 emissions will increase at
approximately 0.4 percent per year from 1985 to 2000.
With moderate demand between 1985 and the year 2000
eemissions are projected to decline slightly or
remain relatively constant in all regions of the
country except the South Atlantic (SA) and western
midwest and mountain states (WNC, WSC, MM, SM).
Under high demand growth after 1985 (5.5 percent per
year in total demand and roughly 6 percent per year
in coal-fired generation), national S(>2 emissions
under current standards will increase at approxi-
mately 2.5 percent per year.
With a high growth between 1985 and the year 2000,
SC>2 emissions are projected to increase in every
region of the country except the east south cen-
tral (ESC) region. The western midwest and moun-
tain states again would show the largest percent:
increase in SC>2 emissions, but large increases
in the magnitude of emissions would occur in the
eastern north central (ENC) states as well.
5.3.3 Air Quality Modeling Results
The relationship between emissions and resultant air quality is
complex and dependent upon many conditions such as emission height,
5-30
-------
temperature and velocity, wind speed and direction, and terrain
topography. To relate emissions to predicted air quality levels in
a geographic area, various mathematical computer models have been
developed. These models are useful in indicating air quality concen-
trations at points where no actual measurements are available or for
predicting air quality impacts based on future emissions. However, a
process of calibration of the model through comparisons with real
data at selected geographic locations is desirable to reflect actual
conditions at a location. When actual data are not available, as in
predicting the effects of a new source, the results may be less
reliable. In addition, factors such as atmospheric stability, changing
terrain topography (e.g., water to land or flat to mountainous)
and variations in wind direction and speed over time result in
uncertainty, especially over longer averaging periods. Because of
these uncertainties, the results obtained from these models must be
used cautiously. They do illustrate trends and the relative air
quality impact from steam generator emissions in a localized area and
should be treated as such.
A series of dispersion analyses to estimate the effects of
typical power plants meeting the current emissions standard on
resulting air quality is presented in Table 5-9 (EPA, 1977a; 1977b).
Hypothetical plants of three sizes were considered: 25 MW, 500 MW,,
and 1000 MW. The entries in the table are estimates of the concen-
trations at ground level for the highest 3-hour average, the highest
5-31
-------
4-1
cfl
01
x:
s~^
0*
cn
O
Q
EH
cn
h- 1
rJ
t>
M
EH
sa
Oi p3
1 g
pq p2
< W
H Q
M
fa
bo
i
1
U
M
CX
C
r
CO
P
£
cc
EH
C
cc
tr
P*
4-1
CO
CO CC
-H £3
rH 0
(0 Xi
O rt
O
O
n
4-f
M-
rQ
c
c,
0
1
01 0)
rH rH
rH rH
H -i-t
CO XI
C 03
CC C
w
cn
e
01
x:
4-
CC
C
n
r-
4-
cn
0)
bC
cO
0
JI
CC
o
r
O 0)
Z Pi
rH ^ ^^x-, ,7^^
O^-'OO O^^rH O-^-^m
-sf O rH O O OO .
rHtOiH OI in OI OJrH rH-sfOJ COincO
x*x x~-
tO
01
XJ
0)
pi
XN OI x--. CO X-N x^
x^x-vOJ rH-^sOJ rHOCO
^ 'oi O^-'1J"> O^-^oi
r-rHrH .HCSrH OJCOOl
JJ
cO
01
o ot
!s P-
^ X-S O'X-NX- O"X-NX-N
CO'-^CS OJOrH COinrH
O^^ X-N n~. oJiHOi ^asto
in* 'O ov-'r-- O^^C
rH -31 OJ rH ^O O
rMOioi co-^-rH mr^oi
4-1
CO
O)
01
x-^ rH X^ O x-v x"v
x~vx-,OJ Olx^H OJOOJ
^^x -s^ vo o v ' rH c ~*-s ' **-x OO N^ " ' rH ^-^
*»- ^-x tn O N " ^O O * * CN
-^(HN-^ v^cN^
OS-'O o^xr~- o ' rH
vo o * 01 in oo i i
rH ^J OJ CO 1-O O4 -J- CO CO
4-)
£
01
Pi
OQx-soi rHojoi oimco
V^CO^' N-X r_) ^x v^rHX.x
Ov-'- O^-'rH
rHO* ojco- r*>m.
iHCOrH OJ-J-rH cOinOJ
^ V4 g M
OJ
3OtO S3OtO 2OCQ
OXI3 OX!3 OCX!"
x:ic oxiic OXSIP:
i<-C o i vt S o i ^
> 01
to a
£
J-i OJ
3 EH
O
1 0
co x:
bo
4J -H
to OJ
QJ EC
XI
^ bO ^
O -H U
fO E cfl
4_J s_x U
cn cn
00 OJ OJ
i 1 \ID vD
co
OJ 4J
rH CO
bOO
fi
H rH x-v
W CO S
CJ ^
TJ -H 4J
CO O bO
rH -H
rH C 0 01
OJ -H rW ffi
TJ cd o
0 )-i 0) ^
S M 4-J O
at 01 to
w cn
^
rH
in in u~i <3
F^- r*-. r~* PH
rH rH td
to
H 4-J
U PH rH rH
g
to ^^
a QJ
O N
H -H
4-1 CO
S 4J
d C
to td
CO rH
^
CM
LJ-I O O Ol
oi o a a
mo vj
rH 3
o
cn
5-32
-------
24-hour average, and the annual average concentration for different
plant sizes, emission concentrations, and geographic areas. These
concentrations are incremental and would be in addition to the
ambient concentration of the location (e.g., that concentration
present if the plant were not there).
With reference to Table 5-9, several aspects of the results
bear mentioning. First, it is apparent that the air quality impact
is not linearly proportional to plant size. This phenomenon reflects
the greater effective stack height (taller stack and greater plume
rise) of the larger plants. Another significant feature of the
results is the lower impact for the reheated exhaust streams. This
reflects the greater plume rise of the hotter gas. Finally, it is
evident that the estimated impacts vary considerably from one mete-
orological data base to another. This last result may be peculiar to
the year 1964, which was used, Thus, several different years' data
would have to be analyzed before geographical differences could be
definitely established.
The highest 3-hour average concentration indicated in Table 5-9
is 560 (J-g/m^ for the Cleveland-Buffalo meteorological data, a 1000-MW
plant, and no reheat case. This is 43 percent of the allowable 3-hour
Federal secondary standard. The 24-hour worst case result is for the
Dallas-Oklahoma City meterology, a 1000-MW plant and no reheat. The
89 )JLg/m^ concentration is 24 percent of the Federal 24-hour primary.
Finally, for the annual primary Federal air quality standard, the
5-33
-------
3
highest value in the Table is 3.6 to 3.9 (jig/m , or 5 percent for
the Dallas-Oklahoma City 25-MW and 1000-MW cases. Even with the
inaccuracies that are inherent in the results, the table indicates
that a power plant is more likely to cause a violation of the 3-hour
SO standard than either of the longer term standards. The specific
likelihood and number of violations would be a function of the pre-
existing ambient air quality at the time and place that these worst case
meteorological conditions occur. Should multiple units be employed,
consideration of the distances between stacks and their relative loca-
tions with respect to worst case wind patterns becomes important.
The additional impact upon ambient air quality from multiple
stack emission sources is difficult to quantify since the results
would be very sensitive to site-specific factors such as the relative
location of the stacks to wind patterns, the relative stack heights,
and the effective plume heights. If the stacks were in a line with
the wind and all other factors were equal, a substantial additive
effect would occur with the magnitude of the additional concentration
dependent on meterological conditions and distance between the
sources. A much smaller impact would occur if the same wind were
perpendicular to the same line of stacks.
5.4 Present Water Environment
Essentially four kinds of water management systems are available
to the power industry. These water management systems differ in the
relative use of once-through cooling and recirculating water cooling
s y s t ems.
'5-34
-------
System #1. All water used in the power plant is managed in a
once-through system (Figure 5-5).
System #2. Recirculating cooling water at 2.5 cycles
of concentration with once-through ash handling
system and once-through general services water is
used (Figure 5-6).
System #3. Recirculating cooling water at 5.0 cycles of concen-
tration, 50 percent recirculating of ash handling
water and the circulating of general service water
blowdown to ash handling and the recirculating of
general service water blowdown to ash handling
system is used (Figure 5-7).
System #4. All water of the power plant is recirculated
(Figure 5-8).
5.4.1 Water Quality
5.4.1.1 Unit Power Plant Water Requirements. In coal-fired
steam/electric power plants, the heat of combustion produces steam
to power turbine generators. The steam is subsequently condensed and
returned to the boiler for further service. Approximately 45 percent
of a fossil-fuel fired generating station's energy is removed and
ultimately discharged to the environment by the condenser cooling
system. To calculate the total cooling water requirement, a power
plant efficiency of 37 percent was used. For a 500-MWe power plant,
610 MW (35 MM Btu/min) heat removal capacity is required. If a 10°C
(20°F) rise in cooling water temperature is assumed in the condenser,
3
a circulating flow of 13 m /s (210,000 gpm) is required.
In once-through cooling systems, the makeup water requirement is
3
equal to the circulating rate, i.e., 13 m /s (210,000 gpm).
5-35
-------
WATER SOURCE
RETURN TO
WATER SOURCE
RETURN TO
WATER SOURCE
Source: Radian, 1977c,
Figure: 5-4 System #l-Once-Through Water
' Management.
5-36
-------
WATER
SOURCE
SURGE POND
GENERAL
SERVICE
WATER
J_
WATER
TREATMENT FOR
BOILER MAKE-UP
DRIFT
COOLJNG
TOWER
EVAPORATION
ASH POND
RETURN TO
WATER SOURCE
Source: Radian, 1977r.
Figure: 5-5 System //2-Partial Recirculatory Water
Management.
5-37
-------
WATER
SOURCE
SURGE
POND
T
GENERAL
SERVICE
WATER
WATER
TREATMENT FOR
BOILER MAKE-UP
RETURN TO
WATER SOURCE
4
DRIFT
EVAPORATION
ASH POND
BOTTOM
ASH
SLUICE
FLY
ASH
SLUICE
'
RETURN TO
WATER SOURCE
Source: Radian, 1977c.
Figure 5-6 System #3~Recirculatory Water Management,
5-38
-------
DRIFT
VAPCRAT1ON
WATHR
SOURCE"
SURGE
PONO
GENERAL
SERVICE
WATER
T
WATER
TREATMENT:
FOR 8OIL£?(
MAKE-UP I
ASH
PONO
BOTTOM
ASH
SLUICE
FLY
ASH
SLUICE
1
i
'
Source: Radian, 1977c.
Figure 5-7 System /M-Zoro Discharge Water
Management.
5-39
-------
The ash handling system of a coal-fired power plant uses water to
make ash slurry consisting of 5 weight percent fly ash and 1 weight
percent of bottom ash. Sluice water requirements for a number of
different coals are found in Table 5-10. The water of the ash handling
system can be managed several different ways:
1. Once-through ash sluicing water system where water makeup
requirements are equal to water sluicing requirements.
2. The use of blowdown from the cooling tower in a once-through
ash sluicing water system. Sometimes a high ash coal is
burned which requires more water for ash sluicing than is
available from the cooling tower blowdown. This additional
water requirement is filled with raw water. Sometimes less
water is required for ash sluicing than is provided by
the blowdown of the cooling tower; excess blowdown water
is ponded before it is discharged.
3. Cooling tower blowdown and general service water blowdown
are the sources of water for ash sluicing. Fifty percent
of the sluicing water is recycled. As with the previous
ash handling water management system, certain coals (high
ash coals) will require the addition of raw water for makeup.
4. Cooling tower blowdown is the source of water for ash sluic-
ing makeup and the cooling tower operated at 13.5 cycles of
concentration with the sluicing system being in a recycling
mode. No raw water is required as sluicing water makeup
when coal ash is low. Excess cooling tower blowdown is
available for general service water.
The generjil services water system includes water used for
condenser and boiler cleaning, water conditioning, boiler fireside
and air preheater washing, auxiliary cooling system and general
power plant water use. According to the date of a water recycle/reuse
study (Noblett, 1976; Christman, 1977; Gathman, 1976), a reasonable
estimate of the use of general service water is 95 ficm-Vs (1.5 gpm)
5-40
-------
1
ry^
))
o o-
1 ft
LO
Pi
W W
£ ^
^ ts
r i
SLUICE
cement
H
rr
01
CU
4-1
cd
Sluice
rd
4-1
0
H
43
CO
6
o
4J
pq
43
CO
CX
M
CO
CO
e
e
a
60
CO
CO
(3
5.
M
CO
m
e
cd
o
o
o o o o o o
o in o o o m
rH O SO OO O O
CO CM CO m CM CM
LO
o co co rC pin rC i-C ,-C r^
rH CO CO CO CO CO CO
M-) cd cd cd cd cd cd
CO B^S B^S B^S B-S B-S B-?
Cfl SO sD CM CM CM CM
rH rH rH rH
O CO CO CO CO CO CO
M-l -H -H -H -H -H -H
43 cd cd cd cd cd cd
CO 42 42 42 42 42 42
cd
4J
LW 42 43 42 42 42 42
O rH rH rH rH rH rH
4J 333333
0 4J 4J 4J 4-1 4-> 4J
OJ pq pq m pq pq pq
0
}-i O O O 0 0 O
01 O O O O O O
& o o o o o o
LO OO rH CM CM CM CM
^ co co w w co cn
CO CO CO CO CO CO
cu cd cd cd c$ CTJ pj
0 42 42 42 42 42 43
S ^ S ' ' ^ s^^ \^s s,^ /
CO
Ol C/J CO CO LO CO CO
3
rH43 B-SB-i6-SB-S6^B~S
cdco ooooinoioo
^^ ^D co r^*. co r^-s.
aj 6
O4-1 rHCMCO~d"LOsD
cd o
43 rH rH rH rH rH rH
01 cd cd ccj cd cd cd
43CO OOOOOO
H'H CJCJCJCJOU
o
rH
*1
rH
TJ
^
Source
5-41
-------
per megawatt. Thus, for a 500-MW power plant, 0.048 m3/s (750 gpm)
of water would be used for general service water. It has been
determined from a study of Georgia Power Company's Bowen Plant (Noblett,
1976) that 75 percent of the general service water could be used to
meet ash sluicing or cooling tower makeup water requirements. This
amounts to 0.035 m-Vs (560 gpm). The general service water system
can have three types of water management systems:
1. Service water system operates as a once-through system.
2. The service water is recirculated to ash sluicing.
3. The service water is recirculated in the cooling tower.
Since the boiler water has to be blown down periodically to
reduce the concentration of impurities, makeup water is required.
For a drum type steam boiler, the blowdown rate is 0.1 percent of the
steam generating rate. For a 500-MW plant operating at 37 percent
efficiency, the makeup water requirement is 50 cm^/s (9 gpm).
The total amount of makeup water required for the four alterna-
tive water management systems of a model power plant is shown in
Table 5-11.
5.4.1.2 Water Requirements for FGD Systems. The makeup water
required for five alternative FGD systems for the base case 500-MW
power plant is shown in Table 5-12. Water comsumptions for the
manageable FGD system (lime, limestone and double alkali.) is quite
similar, 0.035 to 0.038 m3/s (560 to 613 gpm). The regenerable FGD
5-42
-------
z
o
H
6=
1
Z
o
c:
1
g
E
H fe
1 j 3
LO Rj
ft^ C£
c= 3
< O
H C«
H;
A
g
!*3
CO
<£
O
u
c/:
cc
J- C
O Q
re §
3 £
^ 3
Cfl CT
& CJ
V Ctf
a.
re cu
0 TO
H E
V- CH
T 1 0<
O TO
pa S
o:
;j
^j
>
^
^'
j
rt £
DC CJ}
C £*i
C
o
0
c
g I-1
0) U
4J a
c£ *^
cn o
to
0
a
6 l-i
0) QJ
w E
>, 3
CO Z
E
c.
C/j
m
£
'g
at
cr.
E
'5
C£
tf
" .
c")
E
p:
6C
%-
U"
r-T~"
E
^_^
£
a
bO
-^,
E
g
O
lA
rsi
x^^
m
^^
m
o
0
=>
0
^
0
r^
C
o
§
ii£)
C*^
r^
CN
C
s-*.
C^
O
o
o
CN
X2
X
M
3
0
^j
1
!1J
c
0
^^
c
CO
00
vD
O
a^
c
o
o
c
r^
r
0
O
O
c
r-^
v£
O
o
^^
c
e
>_^
o
u
o
(C
3
u
TO
.,
W
C
D
U-J
T 1
1C
6~-S
..
M'
ro
VW
0
?N
C
Q;
H
o
-H
U-i
U-i
0,
c
TO
OC
C
*
J-j
TO
O
C-
o
4_i
C
TO
cx
3
0
C-
|S
1
c
0
TO
cn
-ri
QJ
cn
TO
U
to
TO
OJ
f-<
TO
01
C
x:
OJ
o
c
o
0.
TO
cn
E
0)
cn
cn
V-i
. QJ
^~* 4-1
.£> TO
i-H ^
P 4-1
4J C
CC TO
o 'a
o »-<
^ 0)
fsi 3
rH O
*~*s p-
T i
^3
i-
a,
TO
3
05
a
c,
>
OJ
tf-
tv
u
Qi
C
cu
x^
^
o
^
x:
V
OJ
u
c
c
TO
oc
c
-o
c
x:
"K
cc
x:
6C
a
n
j=
4-J
!
O
C
o
,
c
H
«
J^
c
c
o
l_J
o
tt
CJ
^-1
u
u
CN
TO
to
C
H
r 1
O
O
O
>>
o
4J
re
3
CJ
H
CJ
0)
J 1
j_.
4-'
a,
u
-H
>
^
a
i-t
TO
C
a-
c
01
u
CJ
OJ
^
T!
C
*v
rH
II
1
TO
x;
xt
cn
TO
X
C
to
3
CJ
H
U
^
O
c
c
*rl
nj
-U
C
OJ
o
c
o
M-J
o
cn
i i
o
u'
o
TO
00
C
i 1
O
o
u
>*
o
TO
U
-H
U
Di
3
^
O
TO
i I
ID
^ U
OJ -r-t
4-1 CJ
cn 01
!>i }^
cr
OJD l-j
C TO
H
5 1
C Oi
re 4->
xc w
x: cn
CO
T3 Vi
OJ
GJ 4J
x: TO
O 4J
4-> C
TO
C -H
3 a.
0
*o ^
S 01
° 1
H
^
(U
5-43
-------
£
W
M
cy
^
«
H
S
~*
£
E
to
c/:
Q
O
[H
U^
CO
to
C
H
J^
3
CJ
CO
QJ
3:
H
i
CC
T
"^
CJ
,£)
Q
X
H
K
0)
C
M
cc
Ll
c
1
c
rH
^
J_)
QJ
Q.
C
C
K
QJ
£
T-1
~
Q)
^
QJ
E
i-J
QJ
C
O
t£
OJ
£
H
e
H
c
0
H
ij
E
0
X
_C
<
SI
H
.0
3
y
w
QJ
u
H
W
00
c
.c
3
y
to
00
c
H
-^
3
y
V3
T!
Li
O
iJ
cO
3
4j
C
re
0)
c
01
01
4J
C
CO
0)
c
01
a
cu
as
4J
C
CJ
E
0)
L.
3
C"
a>
E
or
c/;
^^
,-^
G.
00
03
ro
E
e
c.
oc
c-T"
E
c
a
Of
u:
"i
/-*,
c-
cc
-^.
E
D*
60
VI
-^
£
o >^v be
X-N x-v /-N x-v m x-v
CO r^ m rH 7 rH
1
r~ 0 0 m c 0 0
C C C C C O O
000 O O O O
U ^-. 6C
^~v /* *~ N ^v in /^
O ' ^c m ^ '
- ' s ^^
1
O <^ rn oo C CM CM
rn C C r-i c C C
C C C O C O C
O C C C C C O
T)
^Q QJ 00
C m m N \D ' CN
m c O COsrmcoc:
C O C OOOCOOC
C O C CCCCCOC
QJ
Tj ^ iw OC'
C ' i i
i
O u^ m m m
ro C m O rH
CO COO
c c o o c
-o
CJ
E 4J )u
we -H
CD QJ 3
no i- E crm
C C 4J QJ CQJC10 LiLiLiC (3C
-HWX1S-C&'1)3QJ OLiO
ci-cc6o o -H
M
r^!
y:
e-i
u-,
C
rl
^r
"?
r1
cj;
E: w
u c
J^ -H
w
u-! "Tc
o y
J C
C cC
QJ
u * -
H x-v *H CT
"*-, ,C Cb >,
a) -^- c C
C -u ~
cc cc L- rr o
J-i C ^ k '-
o "a *9 "
&C * QJ C
C fN CT, L . U,
H r ' 3 C. "C O
cc K tr o ^
GJ j«; w -c cc
&, "~*- c u; ~ c O
O ^ O E T- 1-
^ -H Q) U C- L,
4J iJ iJ ft C
C cc co ie cr i*-
TO CM i 1 >, 0 -r-
CL U_ O L, L, O
Ll CC rH ij-i U-, f-|
QJ QJ y to * !"* i>
O rH JJ rH Cfi C/: L,
Q- CO C CC Q.
3 iH QJ T- -H e.
S 00 & *^ *~
C /^ j_i j_j r"
O -H rH -J C C
O AJ GJ C CJ CJ tT
u^ CO T3 T3 E E d.
GJ C COO
CO pC G G r-« J-" O
L- 01 Li 4J 3 3
O 6C O CT CT CC
"4-t CO U- 1 QJ O 01 C
LJ j: L s-
tfl QJ -O _I _C
iH > QJ Li Li .C
Q) 3 C 4J 4-1 3
tfi C tc cc r: u
CC CC LC CJ ^ S tJ
u cc dc en
ai 4-j tn u cd ct c
rO > > O O rt
^j QJ crj >j 4-1 L
-H 3 tr
01 nj i i QJ o OJ c-
j: o co j= .c ^ o
H W > £H H H «
5-44
-------
systems use more water, 0.042 m3/s (709 gpm) for the magnesia slurry
system and 0.058 m^/s (923 gpm) for the Wellman-Lord system.
The makeup water requirements for various model plants with FGD
systems which meet the present standard of performance are presented
in Table 5-13. The power plant makeup water requirements are for
System #3 which includes recirculating cooling water at 5.0 cycles of
concentration, 50 percent recirculation of ash handling water and
reuse of general service water in ash handling. The total makeup
water requirement is directly proportionaly to plant size but varies
little with the type of FGD system.
5.4.1.3 National Power Plant Water Requirements. Presently,
there are 29 operational FGD systems, which have a total generating
capacity of 8914 MW. Only two plants with a total capacity of 235
MW, have regenerable FGD systems. Using the data in Table 5-13, the
approximate consumption of water by the nonregenerable FGD systems is
o
estimated to be 0.600 m /S (9570 gpm). The approximate water require-
ments for the magnesia slurry FGD system with a 120-MW generating
capacity is estimated to be 0.0105 m /S (168 gpm), and for the 115 MW
of generating capacity which has a Wellman-Lord FGD system the water
3
requirement is estimated to be about 0.0143 m /S (172.5 gpm).
The utility simulation model was used to project the quantity
of water that would be consumed by FGD systems to meet the existing
NSPS in future years. Table 5-14 shows projections by region for
5-45
-------
EH
"Z.
t:
EH
s
Q
g
CO
o
H
rH
5
o
01
AJ
CD
'ower
CJ
JJ
m
>,
to
rH
>
c
CO
M
OJ
3
O
f"
c
8!
4J
C
a
rH
Makeup Wate
OJ
CO
3
CL.
3
OJ
^
CO
CO
0.
3
0)
ro
apacity Control Coal"
Requirement
4-1
lequiremet
lequiremen
01
a.
e
01
jj
w
>
cya
01
to
0
a.
00
00
e
a.
00
"*^
--~
~s
E
C
O
CN
00
o
o
c
o
o
o
o
o
CN
h-
o
f~>
01
H
J
O
o
o
rH
C
o
CN
00
c
o
c
o
c
o
o
CN
i
0
01
c
o
4J
CO
01
o
0
c
CN
o
o
00
00
o
o
o
o
c
c
c
o
CN
r-
c
Ul
o
rJ
t
5
E
rH
iH
o
o
o
r-i
ocoooooooocoooo
°. ° ^ *. °°. °°. ^, 1 m. -* ^ ^ C 3 C
CNCNOOOOCNrHrHrH
CNCNCNCNCr^mfNOOOOOOCOOOcNrgcN
ooooooccoooccoc
OOCCCOOCOOOOT-^r^oO
oio-.inoovcrvr^
-------
TABLE 5-14
WATER CONSUMED BY FGD SYSTEMS
(Thousand acre-feet per year)'
REGION
1990
1995
2000
a. Moderate-Growth Rate (520 ng/J-1.2 Ib S09/10 Btu)
NE
MA
SA
ENC
ESC
me
wsc
NM
SM
PA
Nation
3.23
21.5
13.6
12.4
5.73
negligible
11.4
5.07
20.0
negligible
92.9
3.84
23.1
11.3
12.7
5.59
negligible
11.4
5.51
18.9
negligible
92.3
4.64
25.6
10.5
12.9
4.67
negligible
11.5
5.36
17.1
negligible
92.3
b. High-Growth Rate (520 ng/J-1.2 Ib S02/10 Btu)
NE
MA
SA
ENC
ESC
WNC
WSC
NM
SM
PA
Nation
3.77
9.28
14.6
13.8
0.918
negligible
11.3
5.39
19.9
negligible
79.0
6.52
6.59
11.5
29.2
0.282
negligible
11.5
5.44
18.3
negligible
89.3
11.9
4.11
12.4
48.5
0.0856
negligible
11.5
5.50
16.8
negligible
111.
Metric units are not used because of the convention
of using acre-feet for water supply data; 1 acre-foot=
1,234 cubic meters.
Source: Teknekron, 1978.
5-4y
-------
both the moderate- and high-growth scenarios. The assumptions
used in making these projections and descriptions of the regions
appear in Appendix J.
There are only two areas in the U.S. where the water consumed
by FGD systems is negligible, the West North Central and Pacific
regions. The impact that these water demands would have upon the
water supplies in the respective regions, cannot be determined from
the above data, but critical areas in terms of the limited supply of
water would include areas in California to the north, the southern
mountain regions, and southern and central inland parts of the
state.
5.4.2 Water Quality
A power plant has a number of different water systems producing
wastewater streams that vary in quality according to the type of
water management system. The water quality of these wastewater
streams reflects the water use within the system. A once-through
cooling water system produces an effluent that is similar in water
quality to the influent; the effluent may contain only small amounts
of corrosive products, corrosion inhibitors and biocides. In a
recirculatory cooling water system, the effluent water quality is not
the same as that of the influent. Dissolved solids build up to 1500
to 10,000 ng/1 (1500 to 10,000 ppm). Soluble gases and particulates
enter the water from the air. Corrosion and scale inhibitors (chro-
mate, zinc, phosphate, silicates, certain proprietary organics for
5-48
-------
corrosion inhibitors, inorganic polyphosphates, chelating agents,
polyelectrolyte antiprecipitants, and organic/polymer dispersants for
scale-inhibitors) may also be present. The quality of bottom ash
sluice water effluent is similar to its influent water, but fly ash
sluice water effluent differs in water quality from its influent
water as its turbidity is greatly increased and as it contains salts
of sodium, potassium, calcium, and magnesium dissolved from the fly
ash. The sedimentation clarifier underflow water from the water
conditioning system has high concentrations of suspended solids and
traces of flocculent and coagulant (alum, aluminate, copper or ferric
chloride). Filtration backwash water in the water conditioning
system has high concentrations of suspended solids. The lime/lime-
soda softening clarifier underflow of the water conditioning system
has a hardness of about 50 mg/1 (CaCOo), a pH of 1C and traces of
flocculants and coagulants (alum, aluminate, copper or ferric chlor-
ide). The ion exchange regeneration waste stream of the water condi-
tioning system has high concentrations of suspended solids in the
backwash, and the spent regenerate has an extreme pH and high concen-
trations of eluted ions. The evaporation blowdown of the water
conditioning system contains the same concentrated impurities that
are found in the feedwater with a total dissolved solids concentra-
tion of 1000 to 2000 mg/1 and a pH of 9 to 11. The boiler blowdown
from the stream generation system has high concentrations of dissolved
solids, traces of corrosion products, and chemicals used for scale
5-49
-------
control such as inorganic phosphates, EDTA or NTA; the pH is between
8.0 and 9.5. Equipment cleaning and washing waste streams of the
general service water may have high suspended and dissolved solids,
an extreme pH, high BOD and/or COD, and detergents. Coal pile runoff
of the general service water may have high suspended and dissolved
solid concentrations and a pH of 2 to 3.
With the addition of flue gas desulfurization (FGD) systems
to a power plant, more water is used. The water quality of effluent
streams from these FGD systems differs markedly. Properly operated
lime/limestone wet scrubbers would not have a wastewater stream.
However, when the system has to be purged, the content of the purge
liquid would be equivalent to the scrubbing liquor. An example of
the chemicals found in the scrubbing liquor of lime/limestone FGD
systems are found in Table 5-15.
The prescrubbing system blowdown in a Wellman-Lord sulfide
scrubber may contain 10,000 to 20,000 mg/1 chloride ions, suspended
solid concentrations of 5 percent, and trace amounts of the fly ash
chemicals and scrubbing liquor. The condenser cooling water system
blowdown of the Wellman-Lord system would have a water quality
similar to that of the power plants cooling water blowdown. Pre-
scrubber blowdown of the magnesia slurry system may have a chloride
content of 10,000 to 20,000 mg/1, 5 percent suspended solids, and
trace amounts of fly ash and scrubbing liquor. An intermittent purge
of the magnesia slurry system may contain MgSCs, MgSO/ and trace
5-50
-------
TABLE 5-15
RANGE OF CONCENTRATION OF CONSTITUENTS
IN SCRUBBER LIQUORS STUDIED
Range of Constituent Concentrations
at Potential Discharge Points
(me/I)
Constituents
Minimum
Maximum
Aluminum
Antimony
Arsenic
Beryllium
Boron
Cadmium
Calcium
Chromium (total)
Cobalt
Copper
Iron
Lead
Magnesium
Manganese
Mercury
Molybdenium
Nickel
Potassium
Selenium
Silicon
Silver
Sodium
Tin
Vanadium
Zinc
Carbonate
Chloride
Fluoride
Sulfite
Sulfate
Phosphate
Nitrogen (total)
Chemical oxygen demand
Total dissolved solids
Total alkalinity (as CaCO )
Conductance mho /cm
Turbidity, Jackson Units
pH
0.03
0.09
<0.004
'0.002
8.0
0.004
520.
0.01
0.10
40.002
0.02
0.01
3.0
0.09
0.000 A
0.91
0.05
5.9
40.001
0.2
0.005
14.0
3.1
<0.001
0.01
<-l.Q
420.
0.07
0.8
720. 10
0.03
iO.OOl
60.
3200. 15
41.
0.003
il.
3.04
0.3
2.3
0.3
0.14
46.
0.11
3000.
0.5
0.7
0.2
8.1
0.4
2750.
2.5
0.07
6.3
1.5
32.
2.2
3.3
0.6
2400.
3.5
0.67
0.35
ao.
4800.
10.
3500.
,000.
0.41
0.002
390-
,000.
150.
0.015
0.0.
10.7
Samples obtained from: EPA/TVA, Shawnee, Steam Plant - venturi and
spray tower; EPA/TVA Shawnee Steam Plant - turbulent contact absorber;
Arizona Public Service Cholla Station - flooded disk scrubber and
absorption tower; and Duquesne Light Phillips Station - single - and
dual-stage venturi.
Includes all soluble species.
Source: Radian, 1977c.
5-51
-------
impurities. Prescrubber blowdown of the double alkali wet scrubbing
system may have a chloride concentration of 10,000 to 20,000 mg/1, 5
percent suspended solids, and trace amounts of fly ash and scrubbing
liquor. A purge stream from the double alkali wet scrubbing system
would have high sodium sulfide, sodium sulfate and non-sulfur calcium
salts. Solid-waste water of the double alkali wet scrubbing system
would be similar to the solid-waste water of lime/limestone systems
but not identical.
In the regenerable FGD systems, the enriched SO product stream
can be used to make elemental sulfur or sulfuric acid. Elemental
sulfur plants using the allied process would have no wastewater
stream. Product acid cooling water system blowdown in sulfuric acid
production would have the same water quality as the cooling water
system blowdown of the power plant.
The effluents from all of the above systems can be treated and
the purified water made available for recycling. Leaching of chemi-
cals from solid wastes of lime/limestone and double alkali scrubbing
systems can be controlled to some degree by chemical fixing and dump
management procedures.
5.5 Land Use
Coal-fired steam generating plants require land for facilities
used in power generation and for the disposal of solid wastes. The
amount of land utilized is variable and depends upon the generating
capacity of the plant, type of cooling system, sulfur and ash content
5-52
-------
of the coal, and the types of flue gas control systems employed.
Plants using cooling ponds as the primary cooling system require more
acreage than those using cooling towers; while power plants with
cooling towers utilize more acreage than once-through cooling
systems. The higher the concentration of ash and sulfur in coal, the
more acreage required for solid waste disposal. In general, the
larger the generating capacity of the coal-fired plant, the more
acreage required for the site.
5.5.1 Land Used for the Physical Plant
A coal-fired electric generating plant with an SC^ emission
control system is composed of an administrative and boiler facility,
turbine facilities an FGD system, an ash sluicing pond, a coal pile,
a solid waste disposal area, a switchyard, a terminal line, a primary
cooling facility, acreage for parking storage and other miscellaneous
purposes. Excluding land which might be used for a cooling pond or
cooling tower and a FGD disposal area, a typical 500-MW coal-fired
unit with an FGD system would usually require less than 100 acres of
land. The acreage occupied by mechanical draft circular cooling
towers for the 500-MW plant would be approximately 1.4 acres (Shafer,
Troxell and Howe, Co, 1977). Natural draft cooling towers would
require 3.6 acres; a cooling lake would have a surface area of about
1000 acres. The acreage utilized in transmission lines is highly
variable, but makes up a substantial part of the total acreage used
for power generation.
5-53
-------
5.5.2 Land Used for Solid Waste Disposal
All coal-fired electric generating plants need land for the
disposal of solid wastes. Generating plants which have nonregener-
able FGD systems (lime, limestone, and double alkali) produce a solid
waste containing fly ash, unreacted lime or limestone, calcium
sulfite and calcium sulfate and many chemicals introduced from fly
ash. Regenerable FGD systems (Wellman-Lord/Allied and magnesia
slurry) produce a little waste of purged solids which usually results
from operator error. With the present S02 emission limitation of 520
ng/J (1.2 S02/10 Btu), the amount of solid wasteproduced by a 500-MW
plant burning coal containing 3.5 percent sulfur and 14 percent ash
is 2.334 x 10 tons/year when a limestone FGD system is utilized
(Aerospace, 1977). In this case, 80 percent of the S02 in the flue
gas is removed, and 302 acres of land with a solid waste depth of 30
feet are required for disposal of the solid waste for a 30-year
period. However, 50 percent (dry weight) of this solid waste is
composed of fly ash and should not be considered a product of coal-
desulfurization. If 40 percent of the sulfur were removed prior to
combustion by physical coal cleaning (PCC), the dry coal wash tailings
would be 0.409 x 10 tons/year. Ash at the plant would be reduced to
0.7 x 10 tons/year and removal of the remaining sulfur by limestone
scrubbing could produce 0.5 x 10 tons/year of dry scrubber wastes.
Disposal of these solid wastes would require 182 acres of land at a
waste depth of 30 feet for a 30-year period. Disposal of the combined
5-54
-------
PCC and limestone FGD wastes requires approximately 36 percent less
acreage than does the disposal of FGD limestone wastes alone. Solid
wastes produced by double alkali FGD systems would require approxi-
mately the same acreage as that required for the lime/limestone FGD
systems. Regenerable FGD systems would have minimal solid wastes.
5.5.3 Current Land Requirements
Presently there are 120 coal-fired electric generating plants
(50,243 MW) which either have an operational FGD system or are in
some stage of planning, construction, or operation. Only 29 of these
plants (8914 MW) have operational FGD systems. The total amount of
coal-fired electric generating capacity in the U.S. is 206,258 MW
(Federal Energy Administration, 1977). Thus 4 percent of the total
coal-fired electric generating capacity of the U.S. at this time has
an operational FGD system. If 302 acres of land are required for
solid waste disposal at a 500-MW plant burning 3.5 percent sulfur
coal with an ash content of 14 percent and a limestone FGD system,
the computed land requirements for the solid waste presently being
produced in the operational FGD systems would be 4,350 acres. If 192
acres are required for solid waste disposal by a system combining PCC
and limestone FGD, the computed acreage required would be 2,275. A
substantial portion of this area would be required for ash disposal
even if the plants did not have FGD systems.
5-55
-------
5.5.A Projected Land Requirements
With the present SC>2 emission limitation, the projected annual
installed capacity of coal-fired generation will increase from 15,303
to 33,000 MW during the period 1978 to 1998 (see Table 5-16). This
represents a 222 percent increase in annual installed capacity of
coal-fired generation during this period. Based on this projec-
tion, the amount of dry solid waste generated in 1998 was calculated
to be about 156 million tons. A computerized simulation model using
more elaborate factors for projecting growth (including siting
criteria) predicted that about 162 million tons of dry sulid waste
would be generated in the year 2000 if the NSPS remained at 520
ng/J (1.2 Ib S02/106 Btu) (Teknekron, 1978). The acreage estimated
to be required for disposal of the solid wastes as shown in Table
5-16 would increase from 7,324 acres in 1978 to 202,390 acres in 1998
if only low sulfur coal or limestone FGD systems were used to meet
the existing NSPS. The projected amount of acreage required for the
disposal of solid waste from coal-fired generating facilities from
1978 to 1998 is substantial and may increase the total land use by
coal-fired electric power plants by 300 percent. As of February
1978, 92 percent of the generating capacity having operational FGD
systems has limestone/lime scrubbers, and 80 percent of all scrubbers
that should be operational by 1986 will be of the lime/limestone
variety. Assuming that this trend continues, most of the FGD systems
installed from 1978 to 1998 will be the solid waste generating
variety (limestone, lime, and double alkali).
5-56
-------
i-H
1
W
EH
W
H
nJ
PM
O
!3
M
H
3
W
13
w
o
u
M
ctf
u
w
tJ
W
Q
W
prf
M
fe
3
O
a
eJ
0
fn
CO
H
S3
W
S
PH
M
£>
o-
s
,_J
Q
W
H
U
W
!->
O
Pd
PM
^
~
£
E
2
1
i
o
o
LLj
0
ce
<
<
z
< -s
o ^ c-
< a £
t_U vO
t 0
1^1
SI
o
< c
^ LU Lj"
|- c
a i
o
ct >
I^H:
3* < O &
^ tt < g
ex
UJ
~ ."? Ic e c" t: "J5 " $ £ S ": u-" f ^ - S ~ § S ?
^^^.,a-^^^0,, ^ou-so-r-C^r-O-
^ "" "^ ^^ -^^fj2S,-^S^wS^SoS
5£SS = iSSSasSi«8g«SBS
iS5^S?*?5?8aS?«SSSSsi
, ^^-tvjts.t^.-s.r^r^mr^-w-c-T
5KSS£Si=SSCSt£5S2S22S
RrsjgjgS-^^jp-eoo^.c^p^g^og
. " . ^ . °. *" ^~
C>'X'r^r-<^O^-^S^rsj-^r-\^-^'or^Sr^r^Co
=--~r^~*°°°S°°°°°°»oSS
§5o§2ISS§§§§§§§l§§§§§
. , tvj ^. , , ^,-rsj-N.^^r^cvjf^t^r^
Sg|ggSS2|g|gfggg|||g|
[^
^
1-1
1
0)
o
cd
a.
o
4!
W
1
5-57
-------
Along with the actual environmental impact of taking lands out
of some other kind of production and using it for solid waste dis-
posal, there are some potential environmental impacts associated with
placing these solid wastes on the land. These potential environmen-
tal impacts would be mainly due to the chemical components of the
coal and absorbent and would be associated with nonregenerable FGD
systems (limestone, lime and double alkali) more than they would be
with regenerable systems (Wellman-Lord/Allied and MgO).
The amount of solid waste generated by the regenerable systems
is negligible. Limestone, lime and double alkali FGD systems produce
a solid waste which is mostly composed of unreacted lime/limestone,
calcium sulfite, calcium sulfate and chemicals from fly ash. In
addition to these chemicals, the double alkali process has sodium
carbonate in its waste. The relative amounts of these chemicals
depends upon the control system, its design and operating variables
and the type of coal burned. Ranges of chemical constituents found
in FGD sludges from lime, limestone, and double alkali systems are
shown in Table 5-17.
The chemical characteristics of ash depend largely on the geo-
logic factors related to the coal used. The major constituents of
ash are silicon, aluminum, iron and calcium; minor constituents are
magnesium, titanium, sodium, potassium, sulfur and phosphorus. There
can also be trace concentrations in the ash of elements such as arse-
nic, barium, beryllium, lead, mercury, cadmium and zinc. Studies
5-58
-------
TABLE 5-17
RANGE OF CONCENTRATIONS OF CHEMICAL
CONSTITUENTS IN FGD SLUDGES FROM LIME,
LIMESTONE, AND DOUBLE-ALKALI SYSTEMS
Scrubber
Constituent
Sludge Concentration Range
Solid, (mg/kga)
Minimum
Maximum
Aluminum
Arsenic
Beryllium
Cadmium
Calcium
Chromium
Copper
Lead
Magnesium
Mercury
Potassium
Selenium
Sodium
Zinc
Chloride
Fluoride
Sulfate
Sulfite
Chemical oxygen demand
Total dissolved solids
0.6
0.05
0.08
105,000
10
8
0.23
0.001
2
45
35,000
1600
52
6
4
268,000
250
76
21
17
48,000
430
9,000
473,000
302,000
Solids analyses were conducted on six samples from six power plants
burning eastern or western coal and using lime, limestone, or double-
alkali scrubbing processes.
Source: Aerospace, 1977.
5-59
-------
on the chemical composition of eastern and western coal reveal that
eastern coal tends to have higher concentrations of arsenic, cadmium,
mercury, and zinc than does western coal (Aerospace, 1977). It is
expected that these differences in the coal will be reflected in the
chemical makeup of the solid wastes resulting from the burning of
these two kinds of coal.
5.6 Energy Consumption Associated with Control Measures
The energy requirements of different SOo emission control
strategies are dependent upon the method of control, level of
control, and coal composition (Radian, 1977b).
5.6.1 Flue Gas Desulfurization
In FGD systems there are six basic energy consuming processes.
Raw material handling and feed preparation is that part of
the process that involves receiving, storing, and preparing
makeup reagents for the FGD system. Areas of energy utiliza-
tion include the powering of conveyors, grinders, mixers, and
pumps associated with the above operations.
Particulate/chloride removal is a necessary step in regener-
able systems to avoid the buildup of corrosive materials in
the scrubbing liquors. The principal energy penalty for
preremoval of particulates and chlorides involves the flue
gas pressure drop and equipment power associated with venturi
scrubbers and, in some cases, electrostatic precipitators or
baghouses required to remove both substances.
S02 scrubbing involves the actual removal of S0~ from the
flue gas. Various techniques are available with the princi-
pal energy penalties involving the operation of pumps,
agitators, etc., to operate the absorber materials and
pressure drops in the flue gas passing through the equipment.
5-60
-------
Reheat of the flue gas may be necessary to raise the gas
temperature to that value that will prevent the formation of
sulfuric acid misst and to provide a sufficient plume buoy-
ancy. Several techniques such as steam injection, combustion
of auxiliary fuels (oil or gas), and direct injection of hot
combustion gases could be used to raise the flue gas tempera-
ture. All of these involve energy penalties that are very
dependent on the required reheat temperature.
Operation of fans, either induced or forced draft, is used
to maintain gas flow through the scrubbing systems. The
energy penalty to operate those fans is proportional to the
exit gas velocity and temperature (which in turn is dependent
on the FGD system design).
Disposal/recovery of sulfur is dependent on the process used.
Nonregenerable processes generate a sludge that can be
disposed of at on-site holding ponds or, following sludge
stabilization, at an off-site disposal facility. For the
on-site disposal, energy is required to pump water and
material to and from the settling pond as well as to operate
agitators in the feed tanks. Off-site disposal would include
the energy cost of transportation and power to operate
holding tank agitators. Regenerable processes require energy
to operate the sulfur recovery processes. In the Wellman-
Lord/Allied process the energy requirements during the
evaportation step are critical; while in the MgO process,
oil-fired dryers and calciners are used to decompose MgSOo
into MgO and the dilute S0« stream.
In general, the various FGD control alternatives have the following
characteristics:
Regenerable systems have a higher overall "energy penalty"
than nonregenerable systems. This is due to the sulfur
recovery stage of the process.
Regenerable system energy costs are very sensitive to
the sulfur content of the coal. Again these costs are most
sensitive to the sulfur recovery stage of the process.
Nonregenerable system energy costs are relatively independent
of the sulfur content of the coal and the SO,, removal level.
The energy required to reheat the flue gases, operate the
draft fans, and to scrub to remove particulate matter and
5-61
-------
chlorides, is independent of the coal sulfur content: and SC>2
removal required. This energy comprises 65 to 90 percent of
the total energy penalty.
Table 5-18 summarizes the calculated energy penalties, as a
function of plant size, coal sulfur content, and the 520 rig SC^/J
removal level. The values in the table are normalized to indicate
the energy penalty per kilowatt generated. Numbers in parentheses
are the penalties as a percentage of the power plant net heat rates
(total plant heat rate divided by the generating capacity). The
following points emerge from Table 5-18.
There is relative independence of the energy penalty percent-
age with respect to power plant capacity. Regardless of the
control type/coal sulfur content/process combination the
energy penalty percentage on a kilowatt-generated basis is
nearly constant across all the various power plant sizes.
The smallest energy penalty is imposed by the use. of non-
regenerable FGD processes.
The limestone process requires from 10 to 35 percent more
energy than the other two nonregenerable process and indi-
cates a slightly greater sensitivity to sulfur content and
SOo control level. Doubling the coal sulfur content: from
3.5 percent to 7.0 percent would cause a 10 percent increase
in the energy penalty of a lime FGD system and a 30 percent
increase in the energy penalty of a limestone-FGD process.
Of the two regenerable processes, the Wellman-Lorcl/Allied
process requires over twice the energy as the MgO process
when 3.5 percent sulfur coal is burned.
The 7 percent sulfur coal is an extreme worst case example.
Most coal currently used by United States utilities is in
the 2 to 3 percent sulfur range with a maximum of 5 percent.
5-62
-------
r-j
CM
o
CO
60
C
O
C\l
v~x
CO
0
H
CO
CO
<&
H
53
O
u
00
T-t CN
1 0
LT| W
6J i-4
rJ W
PQ Q
<; o
EH S
f^
0
CO
w
M
H
<£
> O
4-1 O
, j
CO O
CJ 4J
CD
PM 4-1
OP- i
t-t
S-l -H
4-1
C >-,
O 60
C_) S-i
01
O 01
CO
o
$3
g
e
o
o
rH
s
0
m
S
0
rH
g
In
CN
0)
£3
rj*
H
£
r^
O
i 1 OO O^ OO CO CT*
rH i 1 CM
-a-cMoi^o cor^~ -cMC"*ooo mcM r^-in
rH rH rH CO rH rH
0)
6
H
r 3
n Q
o o
[*T j [~T j
-\ ^,
0) CU 60 00
fi C rH rH a S
CO CO CO -H 0) -H
4-J 4J O O C C C
cncn Q cjocoocd
ujcuaiciJ o a)4-JD
B66BQ Q f^ CCrHWrH
H -H -H -H CJ O M M U OJ CJ
t-3 1-4 rJ r- 1 fe [-W<1 0)0) S
1 1 1 1 ~^ 4J 4-J rH -H rH
QQPO ^ -^^ --^
^H "r i <^ < v^-i ^-^ v^ ^r* *^ *^r* ^T* Se^t
^-^4 r^H (^-4 <--! rt-' le^ <^ 1 r^-< ^-t r^< 1^-4
CTi O^ O^ 0> O^ O^ O> 0s* LO CN CN
r^f^.r^r-.r^ r-r-- o^oc^ a>
CNCNCNCNCN CNCN CNCNCN CN
a>
^3 S'S-iMMSH M>-4 C J-iMS-i S-i
CO rPPPP 0) pp O ppp 3
,_( n | l^-( n ] n ( ij | , 1 ^^ 14-^ %|_f n j i ) ) n ) 14-}
ID rHrHrHrHrH ,£> rHrH 4-J rHlHrH rH
C 33PPP cfl pp & ppp 3
Q) COCOCOCOCO >-l COCO O COCOCO CO
00 0)
QJ B°-5 ^? &^° &~^2 &^5 3 B^ o^ V-t &~S &^x &^? &^
i-i u"i <_i u"i o LTI QJ m m QJ -cf yj CN CN
O cor~-oor-~co QJ coco 4-i OOcN CM
53 Pi 0
QJ
4-1
Cfl
J_j
4-1
CO
QJ
4-J
QJ
CO CO CO CO
4-1 1 1 1 1
C g 3= S IS
PJ >~> "~3 tti "-3
)-i O O O O
£g i^. (^, v£) in QJ
O CM CM CM CM fi
CM "H
I 1 1 1 &
\ t t 1 P5
QJ ."2 P* 12 JI5 X
i 1 S § S § QJ
H m o o o
O CM O O O 4-1
s-i rH m o o
4-1 rH 53
c
0 1
G W
;=> 53
5-63
-------
5.6.2 Other Control Options
The use of low sulfur western coal and physical coal cleaning
in conjunction with nonregenerable FGD are two other options for
which the energy penalties associated with the western coal option
are its generally lower heat content per kilogram of coal and the
energy required to transport the coal to locations in the midwest and
east. (The energy required to develop the transportation network
such as trackbed improvements, roads, hopper cars, coal trucks, and
handling facilities is not included in the analyses.) With physical
coal cleaning, the energy penalty involves the energy required to
clean the coal, the loss in some original heat content, and the
probable need for some FGD system. The FGD system would require less
energy than that required for raw coal due to the lower particulate
removal required and the lower sulfur content of the cleaned coal.
The coal cleaning and coal transportation energy penalties
were calculated using the assumptions indicated in Table 5~19. Using
these assumptions a summary of energy penalties for these options was
generated as a function of plant size, sulfur content, and SC)2 con-
trol level.
A comparison of the results shown in Table 5-20 show that both
coal cleaning/FGD and the use of western coal have higher energy pen-
alties than any of the nonregenerable FGD alternatives regardless of
sulfur content. Physically cleaning coal in conjunction with a non-
regenerable FGD system is shown to require about three times the en-
ergy demand when only the FGD process is used. The coal cleaning/FGD
5-64
-------
TABLE 5-19
ANALYSIS ASSUMPTIONS FOR THE ENERGY PENALTY ASSOCIATED
WITH COAL CLEANING AND WESTERN COAL TRANSPORTATION
Assumptions for Physically Cleaned Coals
Physically cleaned Physically cleaned
3.5% sulfur coal 7.0% sulfur coal
Moisture-Ash
Ash, wt. %
H20, wt. %
Sulfur, wt. %
Heating Value
Free (MAP) Coal, wt. %
, MJ/kg
87.1
6.6
6.3
2.2
29.2
87.1
6.6
6.3
4.4
29.2
(1) 278 Kg/S (500 ton/hour) plant size
(2) 50% ash removal
(3) 40% sulfur removal
(4) 95% energy recovery efficiency
(5) 50% of product coal is thermally dried
(6) Heat for thermal dryer is supplied by combusting product
coal - 434 KJ/Kg (230 Btu/lb)
(7) Electric power required for a 278 Kg/S (500 ton/hour) plant is 2980 KW
Assumptions for Western Coal Transportation
(1) Distance is 2100 Km (1300 miles) - approximate distance between Four Corners
area of New Mexico or eastern Montana and central Ohio.
(2) Unit train capacity is 91000 Kg (100 tons)/car - 100 cars.
(3) Five locomotives: .00021 M3/S (200 gals/hr)/locomotive - full power
.000029 M /S (28 gals/hr)/locomotive - reduced power
(4) Eight hours loading/unloading - reduced power
(5) One hour per large city or federal inspection - reduced power
(6) One large city every 180Km (110 miles); one inspection every 800 Km (500 miles)
(7) Loaded speed: 48 Km/hr (30 mph); empty return speed: 96Km/hr (60 mph)
(8) One percent loss due to coal dust blow-off
(9) Heat content deisel fuel: 38 GJ/m3 (138000 Btu/gallon)
SOURCE: Radian, 1977 b.
5-65
-------
w
&
o
oo
rH
O
I
-20
STANDARD
in >->
W M
hJ C
<3 O
H CM
m
H
H
^
Q
T7I
[H
^t
| |
O
o
CO
H H
co O O
3 s
H 0 v_x
/_N
H
CD
PH
O
W <
W O
H
P
PH v~ '
H*
£>-* PH
O 13
O Pi M
2 Pd *
M ;3 P t-J
H W O .' *S_X
I 1 *sO C^ <]"
co co co CT> O
rH rH rH CM
O
:H SCENAR3
S
O
S3
O
M
33
X * X-N X"N X*X
rH CM CM CM
CM CM 0) CM
S * V_X X«x s '
m
CO rH
c <~> >
H a
4-1 6C O
r-- m in oo
rH CM CO
-------
option is relatively insensitive to doubling of the coal sulfur con-
tent. Raising the western coal sulfur content from 0.4 to 0.6 per-
cent would lower the energy penalty by about 15 percent due to the
higher energy content of the coal.
It should be noted that the sensitivity to the western coal en-
ergy penalty to transportation is such that halving the assumed dis-
tance (e.g., 650 miles) for transportation reduces the energy penalty
by about one-half, or doubling the assumed distance (e.g., 2600
miles) causes an approximate doubling of the penalty.
5.6.3 Energy Penalty Projections (1987 and 1997)
While there are many variables and uncertainties associated with
the mix of coals, the mix of SC>2 control methods, and future gener-
ating capacity growth, it is possible to obtain estimates of the
relative impacts of using a specific coal type and control techology.
The fact that the energy penalties of the various control techniques
are relatively insensitive to the plant size when considered on a per
kilowatt basis allows the analysis to be conducted as independent of
the mix of various plant sizes that will be built in the future.
An analysis of the total installed generating capacity predicted
for the years 1985, 1990, 1995, and 2000, has been performed (Tek-
nekron, 1978). The analysis included consideration of regional coal
availability and coal sulfur content differences, competing generat-
ing technologies, and costs. Based upon this study, projections of
the generating capacity that would be controlled by FGD and overall
5-67
-------
transportation energy that would be required using the current
standard NSPS are shown in Table 5-20.
Both moderate and high growth scenarios were used to project
coal-fired electrical energy requirements. Of this energy require-
ment, only a certain portion would be controlled by FGD and is
labelled in Table 5-20 as "Generating Input Energy With FGD." The
FGD energy calculation assumed that nonregenerable processes using
lime or limestone would predominate with regenerable processes ac-
counting for 5 percent of the FGD control. The coal transportation
energy calculation was based on an assumed "supply node" in a coal-
supply area and centrally located "consumer nodes" in each consuming
state.
The results indicate that the energy that would be expended
for FGD under the current standard would amount to between 0.6 and
0.9 percent of total electrical energy generation with the values
decreasing with time. The difference in the FGD controlled energy
input between the moderate growth and high growth scenarios in 1985
is due to differences in the allocation of low sulfur coal to pre-
viously operating units and those subject to the NSPS.
5-68
-------
6.0 ASSESSMENT OF ENVIRONMENTAL IMPACTS
The environmental impacts that may result from a revision to
the NSPS for S02 emissions from coal-fired utility boilers vary
with the level of the revised standard. Basically, the environmental
impacts resulting from a revised standard are expected to be the same
as those from the present standard, only their magnitude would
change.
The following sections discuss the potential environmental
impacts that may result from three alternative levels of the stan-
dard: 90 percent reduction of potential 862 emissions, 80 per-
cent reduction of potential S02 emissions, and an emission limit of
220 ng/J (0.5 Ib S02/106 Btu). The computerized utility simula-
tion model employed to quantify environmental impacts resulting from
the existing standard of 520 ng/J (1.2 Ib S02/106 Btu) discussed
in Chapter 5 was also used to quantify impacts of the alternative
standards addressed in this chapter. The same forecasts for the
electric utility industry were used (see Appendix J). The model pre-
dicts the environmental impacts that would result if alternative re-
vised levels of the standard were met. The existing standard must be
met if it is not revised; therefore, the impact of revising the
standard is the difference between meeting the present NSPS and the
revised NSPS. Consequently, in the following sections many refer-
ences are made to the discussion of the existing environment pre-
sented in the previous chapter.
6-1
-------
6.1 Impacts on Coal Resources and Transportation
Promulgation of the proposed standard is expected to give rise
to changes in established patterns of coal resource development and
coal movement throughout the country. While the present standard has
been in effect, utilities have exercised the option of procuring low
sulfur coal for the fuel steam generators that are subject to the
standard. Economic and other considerations have tended to promote
the movement of substantial quantities of these low sulfur coals from
the western states to the East (Section 5).
A utility's decisions concerning the (1) selection of fuel for
future power plants, (2) siting of these plants, and (3) logistics of
fuel supply, are based on a set of complex technical, economic, regu-
latory, environmental and other issues. With regard to coal as a
fuel, the costs of transportation from mine to power plant are impor-
tant considerations influencing the expansion plans of many utility
companies. These costs (discussed further in Chapter 7) increase
with increasing distance over which coal is transported and with de-
creasing heat content of the coal. Notwithstanding the financial
penalties, there is an increased reliance on low sulfur western coal
with a relatively low heat content. Projections indicate that coal
from the major producing western states will supply 35 percent of the
total needs of the electric utility industry in 1985. Roughly one-
third (31.3 percent) of the incremental production of coal in the
West will cross the Mississippi River to the East (Section 5).
6-2
-------
The revised standard is not expected to affect substantially the
projected flow of coal to new generating units scheduled to be in
operation on or before the early 1980s. In the ensuing years, how-
ever, an influence on regional patterns of incremental coal produc-
tion and distribution is anticipated. The revised standard would
preclude burning of most coals without flue gas desulfurization or
without a combination of flue gas desulfurization and coal clean-
ing. This will tend to eliminate or diminish the economic and other
incentives associated with the low sulfur option. As a consequence,
both production and movement of low sulfur coals beyond 1985 would be
affected.
These general conclusions are supported by a detailed study (ICF
Inc., 1978) of the impacts of the proposed revised standard on coal
markets and utility expansion plans. The study shows that such im-
pacts would be sensitive to the forecast rate of growth in demand for
electrical power beyond 1985, but are insensitive to whether the re-
vised standard is set at a level of 90 percent reduction, 80 percent
reduction, or a limit of 220 ng/J (0.5 Ib S02/106 Btu). The
greatest effects of the revised standard would be experienced if a
high rate of growth in consumption of electrical energy is sustained
beyond 1985. Comparisons are made in the study between a reference
case based on a 5.8 percent annual rate of growth between 1975 and
1985 and a 3.4 percent growth thereafter; and a second case based on
the same growth of 5.8 percent between 1975 and 1985 and a growth of
5.5 percent thereafter.
6-3
-------
In the high growth scenario, national coal production under each
of the three alternative revised standards is forecast to be less
than the projected production under the current standard by 30 to 50
million tons per year in 1990, representing a reduction of about 2.5
percent (ICF, Inc., 1978). Some of this reduction results from a
higher national average heat content of coal as production is shifted
from western coals of lower heat content to eastern coals of higher
heat content. Furthermore, more utility oil and gas is expected to
be consumed as higher costs of coal-fired power plants make it eco-
nomically competitive to increase the use of oil and gas in existing
oil and gas steam plants.
In 1990 western low sulfur coal will not be competitive with
locally available medium and high sulfur coals in new eastern and
midwestern power plants required to install scrubbers to comply with
a revised standard. Therefore, the amount of western coal shipped to
and consumed in the East is predicted to be lower by 150 million tons
in 1990, representing a decrease of 2.5 percent. The forecast on re-
gional production, which indicates a shift from western coals to
eastern and midwestern coals, is shown by the predictions in Table
6-1.
6.2 Air Quality
6.2.1 SC>2 Emissions
A computerized simulation model has been used to project SC>2
emissions in order to examine the effectiveness of various alterna-
tive standards (Teknekron, 1978). Several emission standard
6-4
-------
TABLE 6-1
IMPACTS ON REGIONAL PRODUCTION OF COAL
Regional Production
Existing
Region
Northern Appalachia
Central and Southern
Appalachia
Midwest and Central
Northern Great Plains
Rest of West
Total
Western Coal to East
1975
179
218
151
55
44
647
21
1985
172
236
243
424
143
1218
206
of Coal,
Millions of Tons
Standard Revised Standard"*
1990
305
237
298
810
218
1768
455
1995
223
241
331
1160
247
2201
601
1990
258
218
364
650
220
1711
298
Q
90 percent reduction of potential emissions.
Source: ICF, Inc., 1978.
6-5
-------
alternatives were considered: 90 percent removal of potential SOo
emissions, 80 percent removal of potential SC>2 emissions, and an
emission limitation of 220 ng/J (0.5 Ib SC^/IO^ Btu). To account
for uncertainty in future energy projections a moderate electrical
demand growth reflecting future conservation measures and a high
demand growth were assumed. The resulting estimated national SC>2
emissions for the alternative standards and the two growth rates are
shown in Figures 6-1 and 6-2 and quantified for the years 1985, 1990,
1995, and 2000 in Tables 6-2 and 6-3.
Salient features of these data can be summarized as follows:
The maximum reduction in national S02 emissions from the
level projected with the current NSPS is projected to be 35
percent, obtained in the year 2000 under high growth
conditions with a 90 percent removal revised standard.
However, the maximum impact will occur after the year 2000
when more of the older plants will be closing down.
Relaxing the SC>2 removal requirement from 90 to 80 percent
reduces the maximum projected reduction to 21 percent.
Assuming a moderate electricity demand growth rate after
1985, the maximum projected reduction in S02 emissions in
2000 at the national level is 20 percent.
More stringent new source standards have a more substantial
impact at the regional level: emission of S02 in the
Mountain and West Central states will be reduced by 49 and 39
percent, respectively, by 1990 assuming the 90 percent re-
moval requirement.
Given the coal sulfur levels used in this analysis, the
amount of S02 emitted under the 80 percent removal standard
and the 215 ng/J (0.5 lb/106 Btu) standard are nearly the
same in most regions. Nationally, emissions differ by a max-
imum of 4 percent. In the Mountain states, where relatively
low sulfur coals are used, the 80 percent removal requirement
further reduces emissions by about. 30 percent in 2000.
6-6
-------
O
H
M
4-1
(1)
2
o
rl
30
25
S 20
0)
CO
G
O
H
15 -
10 H
5 -
Current Standard
90% Standard
1975
i960
1985
i
I co
i
1995
2000
Year
FIGURE 6-1
NATIONAL POWER-PLANT S02 EMISSIONS UNDER
ALTERNATIVE CONTROL SCENARIOS, HIGH GROWTH
6-7
-------
M
g
H
c
o
H
r-1
30-s
25-
% 20-
-------
TABLE 6-2
REGIONAL AND NATIONAL POWER-PLANT S02 EMISSIONS
ASSUMING HIGH GROWTH
(Million metric tons per year)
Region
NE
MA
SA
ENC
ESC
WNC
WSC
NM
SM
PA
National
NE
MA
SA
ENC
ESC
WNC
WSC
NM
SM
PA
National
NE
MA
SA
ENC
ESC
WNC
WSC
NM
SM
PA
National
1985
0.25
1.67
4.04
4.29
2.43
1.56
0.94
0.07
0.23
0.32
15.9
0.24
0.00
3.75
3.66
2.40
1.30
0.75
0.06
0.20
0.26
14.3
0.23
1.65
3.70
3.60
2.39
1.30
0.73
0.06
0.19
0.25
14.1
1990
Current Standard
0.26
1.62
4.19
4.75
2.28
1.63
1.84
0.14
0.31
0.43
17.5
80% Standard
0.24
1.67
3.96
4.11
2.30
1.34
0.98
0.09
0.20
0.26
15.2
90% Standard
0.20
1.54
3.62
3.91
2.27
1.31
0.82
0.07
0.17
0.24
14.2
1995
0.32
1.72
4.67
5.45
. 2.26
1.93
2.71
0.28
0.38
0.60
20.3
0.29
1.97
4.04
4.78
2.25
1.65
1.22
0.12
0.21
0.26
16.8
0.22
1.59
3.46
4.29
2.37
1.59
0.90
0.08
0.15
0.22
14.9
2000
0.44
1.87
5.18
6.14
2.44
2.29
3.49
0.42
0.56
0.90
23.8
0.38
2.16
4.09
5.56
2.51
1.91
1.44
0.15
0.24
0.32
18.8
0.24
1.60
3.27
4.67
2.47
1.75
0.98
0.09
0.15
0.24
15.5
Source: Teknekron
6-9
-------
TABLE 6-3
Region
NE
MA
SA
ENC
ESC
WNC
WSC
NM
SM
PA
National
NE
MA
SA
ENC
ESC
WNC
WSC
NM
SM
PA
National
NE
MA
SA
ENC
ESC
WNC
WSC
NM
SM
PA
National
REGIONAL AND NATIONAL POWER-PLANT S02
ASSUMING MODERATE GROWTH
(Million metric tons per year)
1976 1985 1990
Current Standard
0.23 0.25 0.31
2.06 1.70 1.66
3.09 3.62 3.53
3.50 3.96 3.79
2.64 2.44 2.38
1.32 1.61 1.58
0.11 0.95 1.52
0.12 0.09 0.13
0.34 0.23 0.30
0.20 0.31 0.31
13.6 15.2 15.5
80% Standard
0.24 0.29
1.68 1.59
3.93 3.82
4.36 3.96
2.44 2.39
1.43 1.37
0.78 1.02
0.07 0.07
0.20 0.20
0.27 0.25
15.4 15.0
90% Standard
0.23 0.25
1.65 1.49
3.88 3.60
4.30 3.90
2.44 2.35
1.43 1.35
0.74 0.84
0.06 0.05
0.19 0.17
0.26 0.22
15.2 14.3
EMISSIONS
1995
0.25
1.62
3.73
3.69
2.28
1.70
1.73
0.18
0.28
0.33
15.8
0.22
1.52
3.82
3.64
2.28
1.53
1.12
0.09
0.20
0.21
14.3
0.18
1.36
3. ,49
3. .58
2. .25
1..49
0.79
0.06
0..16
0..19
13.6
2000
0.26
1.69
3.80
3.61
2.00
2.83
1.91
0.24
0.28
0.29
15.9
0.22
1.53
3.68
3.22
2.05
1.76
1.14
0.11
0.18
0.16
14.1
0.18
1.29
3.24
3.17
2.00
1.69
0.76
0.07
0.13
0.13
12.7
Source: Teknekron
6-10
-------
6.2.2 Ambient SC>2 Concentrations
Atmospheric dispersion modeling for several broad geographic
meteorological conditions and power plant size combinations was
performed to estimate ambient air quality. For current standard of
520 ng/J (1.2 Ib SC>2/106 Btu) the predicted ambient air quality
near a power plant was discussed in Section 5.3.
Since SC>2 has a relatively low reaction rate and the dis-
persion pattern due to emissions is independent of the SC>2 emission
rate, a change in the standard would cause a directly proportional
change in the ambient air quality concentration if current state
parameters are not changed (i.e. reheat). Therefore, the effect of
changing the 520 ng/J (1.2 lb/106 Btu) emission standard to 220
ng/J (0.5 lb/10" Btu) all other factors remaining the same would be
to reduce all values (see Table 5-8) by 58 percent. Since SC>2 emis-
sions corresponding to a 90 percent reduction standard are lower than
emissions for 220 ng/J (0.5 lb/106 Btu) standard, ambient S02
concentrations would be lower for a 90 percent standard.
Figure 6-3 illustrates the potential reduction in local ambient
air quality with a 85 percent SC>2 emission reduction standard for a
single 1,000-MWe plant and for a grouping of three 100-MWe boilers
with three stacks. The effectiveness of reheat as a means of in-
creasing plume height, thereby causing greater dispersion and re-
duced ground concentrations, is evident in Figure 6-3. Reductions
6-11
-------
r~*
a-
o
3
§
E-
'-! Q
5 §
^. .-i
"" f
£ c
E-
12
PL
pc
(X
p.
,ui/Bry'|\|oii\/biN33l\l03 2OS
6-12
-------
of 20 to 25 percent in local concentration would occur with reheated
stack exhausts compared with no FGD systems without reheat. Given a
pristine background concentration, the 90 percent SC>2 emission re-
duction standard would result in an approximate halving of the
24-hour and annual local ambient air quality concentration for either
case (one or three utility boilers). In areas where SC>2 background
concentrations are not negligible, Figure 6-2 can be interpreted to
indicate that the probability of a 24-hour primary standard viola-
tions for a given background concentration is greatly reduced under
the 90 percent standard. The probability of annual standard viola-
tions is relatively insensitive to the revision of the standard.
6.3 Water
6.3.1 Water Quantity
The various types of FGD systems differ in the amount of water
each consumes in order to scrub an equivalent stream of flue gas.
Assuming 90 percent removal of sulfur dioxide the water required for
five types of FGD systems (lime, limestone, Wellman-Lord, magnesia
slurry, and double alkali) has been calculated for a number of dif-
ferent size generating plants, Table 6-4. Comparing water con-
sumption of FGD systems for 520 ng/J (1.2 lb/106 Btu) S02 emis-
sion standard (see Table 5-12) and for the 90 percent SC>2 removal
shows that the water consumption for the same size FGD system is ap-
proximately the same for these two SC>2 emission limitations. With
6-13
-------
Fd
O
H
rH
O
r-J
4J
C
o
CJ
>
o
co
CO
^*t
6
QJ
4-J
Cfl
CO
6
4J
CO
rH
OJ
*"O
o
s
K*>
60
OJ
4J
cd
si-i
4_)
CO
4-J
CO
rH
^-!
CU
^
0
PM
O
CO
J_j
QJ
J5
0
P-i
>->
4J
S
ft
3
QJ
vj
Cfl
g
a)
tf
CX
;3
OJ
r^i
cfl
S
n i
u/
!3t
ft
3
Q.t
r^J
Cfl
^
^
3
4-4
rH
£j
CO
^v°
rH
O
4-1
C
o
o
4J
PJ
CO
rH
P-i
1)
C
OJ
J_4
H
3
CT
0)
Pi
4-J
QJ
£i
0)
^_i
H
3
cr
QJ
Pi
u 1
QJ
e
CD
H
3
cr
QJ
Pi
rH
CO
£3
0
B
0)
Pi
K*^
60
OJ
4-1
cd
4-1
CO
4-1
H
O
CO
ft
CO
CJ
CO
cfl
CJ
13
P.,
60
CO
Y-J
S
ft
OJJ
CO
o"
e
1i
60
w
irT
e
S
^
0 0
0 0
o o
CO ro
ctd
CN CN
00 00
0 0
0 0
0 0
r 1 i i
iH i I
V-^ ^-S
0 0
0 O
0 0
CD S
0 O
0 O
CN CN
rH rH
^O VD
1^- h-
0 0
O O
QJ
rj
O
4-J
03
QJ QJ
e e
H -H
rJ J
O O
o o
0 0
rH rH
rH CN
-3" co r^* ?H rH CN CN
OO r~- co cO <} ~^ CO CO O O O O O O O O
oooooooooooooooo
oooooooooooooooo
oooooooooooooooo
OOOOOOOOOOOO-HrHrHrH
OOrHrHrHrHrHCNCNCNCNCNC-OCOrOCO
rH rH "» ~ ' V^ - s_ ' *~s - - ^-^ '
^sO ^O ^O VO *sO C~J C3 CD CD
^o *^D oo co oo co oo r^- r^» r~^ r^^ r~- cxj CN CNJ CN]
T^r^rommcnmooooooooo
oooooooooooocr;ooo
oooooooooooooooo
O*\ O> O^i 0s! O*^ O^ O^ CTi O"i O^ CTi O> O*i O^ O*1 O^
CdCO
QJrH COEQJrH COEQJrH WEQJ
d rQ QJ QJ rH Cd i"1 QJ QJ i 1 CIl o CLJ QJ t fl C
Cfl O -H -H QJ Cd O -H -H QJ Cfl O -H -H a) Cfl
SO.j^&SQr-l.JSgPHJ.JtSg
O O O CO O O O O O O O O LO to LO LO
OOOOOOOOOOOOCNCNCNCNi
O O L^ IJ") lO LO LO T i rH i i < 1 i !
iH rH
-3~ u~i »sO r~*^ co crs o i i CN ro -
O
CN
PI
5^"
^O
(M
. -.
S^t
CO
cd
&^2
CN
r-H
3
rH
r?
4-J
O
o
o
CJ
rH
\_x
M
^
^
CO
c-J
. ^
C/3
6"^
in
co
rd
'2J
C!
H
cd
4J
C
O
u
r)
QJ
E
^
rH
CO
O
0
4-1
o
QJ
ft
^>
4J
QJ
H
.
CJ
{^
rH
pT
cd
H
T3
ClJ
pi
QJ
U
3
O
CO
6-14
-------
90 percent S02 removal the water consumption increases proportion-
ally with the size of the plant and represents approximately ten pre-
cent of the plant make-up water requirement.
A 500-MW power plant that burns 0.8 percent sulfur coal having a
heating value of 19 MJ/kg (8,000 Btu/lb) and that uses a limestone
scrubber to meet the 220 ng/J emission limitation would consume 0.032
nrVs (500 gm of water). Whereas it is possible to meet the present
standard without using an FGD systems by burning low sulfur western
coal, compliance with the 90 percent S02 removal and the 220 ng/J
(0.5 lb 802/10^ Btu) standard will require the use of
scrubbers. Thus, there would be consumption of water by FGD systems
to meet revised SC>2 emission standards at some plants (mostly those
in the West) where none was required with the existing S02 emission
standard.
Projections of the consumption of water by FGD systems on a re-
gional basis were made for a 90 percent removal of sulfur from flue
gas for both high and moderate growth (Table 6-5). A comparison of
regional water consumption of FGD systems for the present SC>2 emis-
sion limitations (see Table 5-14) and for the 90 percent removal
standard shows that substantially more water would be required in
most regions for increased removal of sulfur from flue gas with
either high or moderate growth. Also, the amount of water required
nationally and for most regions would increase with time to the year
2000.
6-15
-------
TABLE 6-5
WATER CONSUMED BY FGD SYSTEMS
o
(Thousand acre-feet per year)
REGION 1990 1995 2000
a. Moderate-Growth Rate (90 Percent S00 Removal)
NE
MA
SA
ENC
ESC
WNC
WSC
NM
SM
PA
Nation
3.83
20.5
34.0
16.3
4.49
5.78
58.4
11.2
33.0
7.82
195.
4.44
26.9
52.4
15.6
5.56
10.7
89.6
15.9
30.1
9.79
261.
5.37
36.2
65.2
13.6
6.09
19.1
108.
19.5
31.2
9.49
314.
b. High-Growth Rate (90 Percent S0? Removal)
NE
MA
SA
ENC
ESC
WNC
WSC
NM
SM
PA
Nation
4.40
32.2
51.8
30.4
3.17
10.9
75.1
16.1
30.6
18.0
273.
7.62
60 . 8
83.6
66.6
10.7
28.4
141.
25.8
37.6
29.7
492.
15.0
83.2
116.
109.
24.5
53.0
209 .
34.3
46.6
58.3
751.
cl
The metric units are not used because of the
established convention of using acre-feet for
water supply data; 1 acre-foot=l,234 cubic
meters.
Source: Teknekron, 1978.
6-16
-------
6.3.2 Water Quality
FGD systems need not have effluent discharges that would impact
existing water existing water quality. However, if any water were
discharged into contiguous streams, it would have to meet water qual-
ity standards. Such discharges may occur when the system has to be
purged. These purges are necessitated by process problems of (1)
water imbalances, (2) changes in operation from design conditions,
(3) catastrophic blowdown of the system to prevent scaling, and (4)
operator errors of various kinds. The composition of the purge
liquid is as follows: calcium sulfite, calcium sulfate, sodium
chloride and trace elements in lime and limestone FGD purge liquid;
sodium sulfite, sodium sulfate and sodium chloride in the double
alkali purge liquid; and silica, ferric oxide, aluminum chloride and
sulfate ions, calcium oxide, calcium sulfate and calcium chloride,
magnesium sulfite, magnesium sulfate, and trace elements of the mag~
nesia slurry FGD purge liquid. The purge liquids can be treated and
the water can be returned to the system.
6.4 Land Use
6.4.1 Land Use for the Physical Plant
The land upon which the scrubber and the storage facility for
the absorbent are constructed would necessarily be close to the
generating plant buildings. Construction of a power plant would usu-
ally cause adjacent lands to be devoid of most of the vegetation and
animals typical of the region. Therefore, construction of FGD on
6-17
-------
land near the generating plant buildings would not result in any ad-
ditional ecological impact. It is likely that this land would not
have any other commercial or agricultural value while the generating
plant is operating. Presumably, the entrire plant site, including
that used for flue gas desulfurization, would be available for com-
mercial use or left for invasion by indigenous flora and fauna when
the plant is decommissioned.
6.4.2 Land Used for Solid Waste Disposal
The land used to dispose of solid wastes generated by limestone,
lime and double alkali fuel gas desulfurization systems would be lost
to other commercial or agricultural purposes as long as the disposal
site is operated. For a 500-MW unit coal-fired plant burning eastern
and western coal, projected use of disposal site land for limestone
scrubber waste for all of the installed generating capacity (for the
years 1978 through 1998) would be slightly greater if the 90 percent
emission limitation were adopted than if the present standard of 520
ng/J (1.2 Ib S02/106 Btu) remained in effect (see Tables 5-16 and
6-6). The amount of land predicted to be required if the standard
were revised to 220 ng/J (0.5 Ib S02/106 Btu) for the years 1978
through 1998 is shown in Table 6-7. The apparent paradox that the
220 ng/J limit would result not only in lower emissions but also in
less total waste and lower land requirements than either of the other
two S02 emission limitations is the result of the assumption that
coal washing of eastern coal is used to remove 40 percent of its
6-18
-------
o
H
Q CO
I
W
H
O W
O O
pi W
O P-i
fn
O
I
to
-) M
O (A
& H
PH O
W
ta
ct
_ M
UJ ', ^ CZ
*^ u-\ ur- i_^ sc <;>
l/> O LT\ u~\ "3:
u^ \c C^ f^- C^ '' OO f ^o ^' CT^ r^i ^c i' (~-J i^ "^ i^"- "^ ^T ro
CO O QO OD O
oooooogciogo
f^ oo c?* O i '
(D
U
cd
ex
tn
o
OJ
O
^1
3
O
to
6-19
-------
K
II
a
Pi
H
cj
w
W O
i-l
P ^~
W CN
PS O
M CO
i eo
O
o
i/l O
EH ti
&
u o
w CN
H I
B-co
23 PH
J O
H
O
W
o
I <
a. o
o ~~
3S
OCr^O T"^ CO O * fM
r^iOo--o.c^O.'^o^csj
-*
Sr fsj i
^^
^ K.
< o
c--6oc; -
O^
TT LTi 3*
OJ
o
to
o
0)
o
6-20
-------
sulfur (Aerospsace, 1977). The coupling of coal washing with a non-
regenerable scrubber system reduces the amount of land required for
disposal of solid wastes at the power plant since some portion of the
solid waste is disposed of at the coal washing site. The land re-
quirement attributed only to desulfurization wastes for any of the
alternative standards discussed is substantially less than that shown
in the tables5 since the data presented for total solid wastes in-
cludes the contributions of fly ash and solid wastes produced by the
particulate removal systems.
6.5 Ecology
6.5.1 Ecology at the Physical Plant
An ecosystem that exists on the land prior to construction and
operation of a power plant would be destroyed at the time of con-
struction with some favored species of plants eventually replacing
indigenous varieties. Continued minor encroachment of wild varieties
of plants and animals would occur but could be controlled. As a re-
sult, the ecology at the site of an existing power plant would be
simpler both qualitatively and quantitatively than the one it re-
placed. The use of ornamental species of plants for aesthetic pur-
poses and erosion control is a well developed art applied to dif-
ferent portions of land used in power generation. However, any at-
tempt at growing plants on ash and flue gas solid wastes must be
characterized as experimental. There are some apparent successes in
the use of flue gas solid wastes as fertilizers, such as its use with
the growing of rye grass by TVA.
6-21
-------
6.5.2 Ecology at the Disposal Site
The exact nature of the ecological, commercial, or agricultural
losses resulting from solid waste disposal would be site specific,
but virtually all flora and fauna would be removed from this land.
Whether or not these losses can be confined to just the acreage used
for a disposal dump is another site specific problem that depends
upon many factors in the ambient environment, how well the disposal
dump is managed, and whether ecologically sensitive areas such as an-
imal migration paths or nesting grounds exist. It is clear that many
of the negative ecological effects and problems with regional de-
velopment resulting from the solid waste disposal site could be
greatly minimized if proper consideration is given to ecology and re-
gional planning in a site selection study and if a good disposal site
management plan is adopted. Similar basic problems can be expected
to arise if the disposal site is in the ocean or some large body of
water; however, solid waste disposal sites in large bodies of water
appear to be less ecologically promising because of the diffi-
culty in stabilizing the solid wastes in a fluid environment. If the
solid waste disposal site is located in a rock quarry or mine, there
would be an opportunity for improving a highly stressed environmental
condition. This approach to the disposal of solid waste could
potentially restore some of the ecology of the area and reestablish
the original contour of the land. Also, more valuable land, which
6-22
-------
has been less disturbed ecologically, would not have to be used for a
solid waste disposal site. A good dump management program as well as
proper consideration of the existing environment and its surrounding
ecology is still required if the environmental impact is to be
minimized.
6.6 Energy Penalties Associated with Alternate Strategies
An analysis similar to that described in Section 5.6 was per-
formed to examine the manner in which various alternative levels of
control could affect the energy penalty associated with various con-
trol techiques and coal mixes.
Table 6-8 summarizes the calculated energy penalties as a func-
tion of plant size, coal sulfur content, and S02 removal level
(current standard versus 90 percent removal). The values in the
table are normalized to indicate the energy penalty per kilowatt
generated. The penalties as a percentage of the power plant net heat
rates (total plant heat rate divided by the generating capacity) are
shown in parenthesis.
Three points are of particular interest. First, 90 percent
removal results in about a 10 percent higher energy use with respect
to both nonregenerable and regenerable processes for a given coal
sulfur content. Second, doubling the sulfur content of the coal from
3.5 percent to 7.0 percent causes about a 60 percent increase in the
energy penalty using the MgO process and a doubling of the energy
penalty using the Well-Lord/ Allied process. (The 7.0 percent
6-23
-------
TABLE 6-8
GY PENALTIES FOR MODEL S02 CONTROL SYSTEMS
w
w
^
a
rH
cx
I rH
2 0
"--. r-i
enalty, J
to uncont
4-1
0 &
4J -H
C
O >,
O 00
CN > O>> O>>
:/NQO c/DOO WOO
e e Be e e
OOCUQJ &OQJCU OOOJS
CN CN CNJ
in&-s&-2 me^B^ infeNge^?
o o o o .00
O ON ON O ON ON O ON ON
60O060 OOOOOO OOOOOO
'-I 1-1 1-1 '->'->>-! >-l >~1 '-I
sss ass sss
CNCNCN1 CNCNCM CNJCNCM
1-l^lV-l Mrlrl ^V^^H
333 333 333
HrHiH iHrHrH rHMi-H
333 333 333
woouN. c/]Wcyi C/NUN.C/]
in in o in in o in in o
-H rH CN
ooo ooo
rH rH CN CN O~) \D
ro rH in r-. co co
LO vD O*i rH P") lO
rH rH CN
OOO OOO
st v£) LO rH LO CO
rH rH CN m n~) ^O
rH rH CN
OOO OOO
LO \D ^D CN £> O
LO O"i vO OO CN rH
LO lO CTv rH CO v£>
rH rH CN
OOO OOO
LO ^O ^O CN ^D rH
rH rH CN 0-) r-
Q O Q
OOO
O O Q pn PL, PH
OOO
W PL, tn <\ > O > >
en o o en o o
S S Eg
00 QJ CU OC CD CO
CN CN
O O O O
00 00 ClO 00 OO 00
!SS III
CM CN CN CM CN CN
333 333
rH rH rH rH rH rH
333 333
en en en en en en
LO LO O LO LO O
WWW
00 CO CN]
cr\ o~> o
rH
OOO
WWW
13 JS £5
0 O
/MJ Coal cleaning/FGD-Lime NI
/MJ Coal cleaning/FGD-Lime 270 (
/MJ Coal cleaning/FGD-Lime 280 (.
CN t-N (^1
OOO
en en en
00 t»o bi.
CN CN CN
LO CN CN
ooo
00 00 00
o> CTN cy\
CN CN CN
M rJ r-<
333
l4_| M_| LH
iH rH rH
333
tn en en
B^ t>e &-s
CM CM ^J-
o o
OO O
(U QJ OJ
e e e
co nl
01 0)
CD CO CT3
OOO
OOO
3
333
CN I-O
^D iO
33
I I I I
3 3 3- 3
A! J^ .li ^i
-> 1-3 >n ^J
O O O O
CN ^H 0
6-24
-------
sulfur coal is an extreme worst case example. Most coal would be in
the 2 to 3 percent sulfur range with a maximum of 5 percent.) This
result is due to the sulfur recovery step associated with these
processes. Finally, as indicated in Section 5.6, the various alter-
natives are relatively independent of plant size when normalized to a
per kilowatt-second (KW-S) basis.
Using the methodology described in Section 5.6.3 projections of
the energy penalty associated with the 80 and 90 percent SC>2 re-
duction standards were calculated (Teknekron 1978). Table 6-9 gives
the results of the projections for the baseline (current standard)
case and the two alternative standards in the year 1995. The highest
energy impact would occur with the 90 percent removal standard. Com-
pared with the current standard, three times as much energy would be
expended for FGD with the moderate energy growth scenario and five
times as much with the high energy growth rate if the 90 percent
standard were in force.
A second energy impact, which could be significant if a revised
standard resulted in large shifts in coal supplies, is the change in
the energy consumed in transporting the coal. A shift away from the
use of western coal in the Midwest in favor of more local supplies,
for example, would be expected to reduce coal transport energy. Re-
sults of this impact for the 90 percent control scenarios are shown
in Table 6-10.
6-25
-------
TABLE 6-9
ENERGY CONSUMED BY FGD SYSTEMS IN 1995
(10 Megajoules)
Energy Fraction of Energy
Growth Rate Consumption for FGD for Generation (Percent)
Moderate
Current standard 187
80 percent removal 534
90 percent removal 588
High
Current Standard 214
80 percent removal 548
90 percent removal 1150
0.95
2.7
3.0
0.71
1.8
3.8
Source: Teknekron, 1978.
TABLE 6-10
ENERGY CONSUMED IN TRANSPORTING COAL
TO ELECTRIC GENERATING PLANTS
Megajoules)
Year
1976
1990
1995
2000
Moderate
Current
Standard
100
250
300
350
Growth
90 Percent
Removal
100
230
250
260
High
Current
Standard
100
370
560
780
Growth
90 Percent
Removal
100
320
440
580
Source: Teknekron, 1978.
6-26
-------
These results show a significant reduction in fuel consumed with
imposition of the more stringent controls, due primarily to a shift-
ing of demand away from western coals delivered to states bordering
east of the Mississippi River. Note in particular that the energy
savings in 1995, 50 x 109 MJ (4.7 x 1013 Btu) for the moderate
growth case and 120 x 109 MJ (11.0 x 1013 Btu) for the high
growth case, serve to offset about 10 percent of the direct FGD en-
ergy requirements projected for that year (588 x 109 MJ and 1150 x
109 MJ for moderate and high growth, respectively).
6.7 Noise
Noise produced by an operating power plant is substantial. An
increase in noise at power plants is expected as a result of the
operating FGD system, the transporting of the solid waste generated
by the FGD to the disposal site, and the operation of the disposal
dump. The amount of noise produced by either the power plant or the
FGD system would depend upon the type of system, sizes and levels of
operation. The extent to which the sound would be perceived by an
individual would depend upon many site specific environmental
factors, including wind speed, wind direction and topography. The
noise produced by an operating FGD system would not significantly
contribute to the degradation of the environment of the plant, since
noise abatement techniques for power plants are available.
Depending upon the site and the means of transport, noise asso-
ciated with transporting solid waste to the disposal site may also
6-27
-------
prove to be a nuisance. Transporting of solid wastes by truck, train
or conveyor belts to the disposal site could result in these con-
veyances passing through small population centers. Thus, people
could be affected by the increased noise levels. Also, ecologically
sensitive areas such as nesting grounds for birds or animal migration
routes could be disturbed by the noise and the barrier the trans-
portation system would present. Transporting solid wastes by pipe-
line would alleviate the noise problems; however, if the pipeline is
not buried, it would still be a potential barrier to animal migra-
tion. Noise produced by the operation of the disposal site has
potential environmental consequences similar to those produced in the
transportation of waste to the disposal site.
6.8 Secondary Impacts
A revision to the NSPS for coal-fired utility boilers is expected
to result in secondary impacts affecting the industries, and the pop-
ulations supported by these industries, which supply and transport,
coal, metal, hardware, and chemicals for FGD systems. Changes in de-
mand for eastern and western coals would impact on employment in the
coal mining and transportation industries. A revision of the
standard would result in increased demand for FGD systems, thereby
effecting those industries which provide such systems. Impacts on
all thee industries could result in impacts on employment and in
other secondary impacts including shifting needs for housing,
schools, and medical facilities. Section 7 presents a discussion of
6-28
-------
some anticipated secondary economic impacts including the effect on
the price of electricity to the consumer.
6-29
-------
7.0 ECONOMIC IMPACT ANALYSIS
7.1 Industry Profile
7.1.1 General Industry Background
The electric utility industry is by far the predominant user of
large coal-fired boilers. Since 1960 over 88 percent of all coal-
fired units installed have been for the purpose of generating power
for electric utilities. These 404 units have provided some 159,000
MW of capacity, or slightly over 98 percent of the rated megawatts
installed.
Significant market perturbations have recently characterized
the electric utility industry (Figure 7-1). Following a relatively
stable growth rate of 5 to 9 percent per year for both energy and
peak-load demand in the 1960s, peak-load demand rose sharply in 1972.
The marked drop in demand that occurred in 1973 was ascribed to a
worldwide recession and the Arab oil embargo. There was virtually
no growth in energy demand in 1974. In 1975 and 1976, however,
demand began to rise again but at a rate lower than during the
1961-1972 period.
Plant construction cycles in this industry represent long lead
times and immediate responses to changes in demand for power are not
possible. Utilities must attempt to predict future needs in order to
have time to raise the needed capital and build the plants to meet
requirements for electric power. Market perturbations can make it
difficult to provide capacity commensurate with future demands. On
7-1
-------
(S
o
§
o
12
11
10
9
8
7
6
5
4
3
2
1
0
-1
December peak load
Annual energy requirements
61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77
YEAR
NOTE: Represented by the sum of the individual peak loads of the
Regional Council's report to the FPC.
Source: U. S, Federal Power Commission, 1974.
FIGURE 7-1
PERCENTAGE GROWTH RATE OVER PREVIOUS YEAR REPORTED
BY flAJOR U.S. UTILITY SYSTEMS
7-2
-------
H
M
O
<
%
o
H
O
H
W
O
-------
the one hand, power needs which turn out to be less than those pro-
jected when plant construction is undertaken can result in a tempo-
rary excess of reserve capacity. On the other hand, demands which
are significantly greater than those planned for earlier cannot be
quickly met because of the time needed to raise capital and to build
additional facilities. After 1968 reserves in the power industry
increased steadily for several years, culminating in 1974 in a high
of more than 50 percent excess capacity over winter peak demand
(Figure 7-2). Coincidentally, there was a marked drop in the indus-
try's collective system capacity factors (Figure 7-3). Consequently,
utilities found themselves with over-committed capital programs at
the same time that lower plant utilization was generating less
revenue.
Orders for coal-fired boilers decreased significantly in 1975
and 1976 (Table 7-1). Demand rose in 1977, and the market should
continue strong with increased annual energy requirements accompanied
by pressures on utilities to convert to coal from oil and gas.
TABLE 7-1
ORDERS FOR COAL-FIRED BOILERS
Year
1974
1975
1976
1977a
Projected
b . .
Boilers
Ordered
69
22
13
20
-i /.
Megawatt;
Capacity
32,964
10,774
6,312
13,000b
-------
c
QJ
-------
7.1.2 Predominance of Coal in Electric Power Generation
Historically, coal has provided a greater amount of rated mega-
wattage than any other electric power source. As Table 7-2 shows, in
1976 coal provided 189 GW or about 39 percent of the total rated
capacity. During the period 1977-95 coal and nuclear fuel are expec-
ted to share the preponderant role in providing additions to the U.S.
electric utility system. Two scenarios are projected in Table 7-2.
The moderate growth scenario, in which the 1976 generating capacity
is slightly more than"doubled by 1995, projects coal as providing for
some 28 percent of the growth, ranking second to nuclear power. In
the high growth scenario, coal would provide about 20 percent of the
additions (Teknekron, Inc., 1978).
The growth of electric power from coal is shown from a different
perspective in Figure 7-4, which presents the total number of coal-
fired units (over 25 MWe) installed or projected between 1960 and
1978. As can be seen, cancellations and delays resulted in a steep
drop in 1973; a slow recovery is apparent in the trend since that
date.
The average size of coal-fired units installed (and projected)
between 1963 and 1978 appears in Figure 7-5. Steady growth in size
during the 1960s and early 1970s led to near stabilization at slight-
ly over 500 MW by 1975. The new generation of boilers expected for
the 1980s will average over 500 MW and be somewhat larger than those
installed 10 years ago.
7-6
-------
CM
w
S'
X
M
S
^"*
£-* '""N
M £S
U O
^
PL,
<) co
CJ W
M
CJ H
M M
(V, r 3
H M
CJ pq
W }
0
Pi
PU
X
in
o\
ON
rH
0
H
CO
^
H
0
H
Pd
o
PS
o
M
«
PS
o
pL^
o
PL)
H
S
o
Q
s~^
LTi
o\
0^
TH
r
p---.
r-x
o\
<~~t
**~s
CO
£5
0
M
lr~*
> i
|
PS
^
Q
PH
O
P3
O
M
PS
£C
FH
£g
o
o
w
£~*
3
w
o
o
2
£**
E~*
t (
CJ
*^ ^"*"»
**o
f^-
rH
W
P.;
^H
H
^~*
H
H
CJ
<3j
PM
*^
CJ
^- in cj>
r^ CTS m oo
00 OO CM O ^O O rH
en in i I r-H 00 rH
f
O CM CTi
oo en in « >
r^ o CM oo >£> o rH
m st- 1-1 co oo
r >
en in rH ^o
r~ o
* CT*. VO i 1
en en en rH -£>
f ^ \
en OO OO rH VD
co sr ....
en o r-^ \c oo oo rH
en CM en rH en
*
*^ o^ f^* "^ CM **o **3~ en **o
*......
^C^iHOOtHLOr-JOOO
vj~ oo oo ^ vo ^"i
i i
0)
rH
a
^i
CJ rH
ed
M d) Qj M
co d cs
OcSrHW0T3MO,G
3O-HcflO^S)Q)4J
^CJOOCJffiHOO
OO
o
en
rH
CM
00
o
a)
g
3
CO
CD
Cfl
0
4-1
rH *rl
o 6
00 -H
T-H
rH
0
$-1
4->
0
O
CJ
CD
^
4-1
O
r£-»
in 4-1
r^ *H
m ?
^>
rH
4J
(~;
60
H
rH
CO
CO
01
0 -H
>-)
r~- r3
00 >
^j-
in
CTi
rH
(-4
O
MH
X
H
e
JS ^
O 4J
^-^ -H
CJ
hJ CO
oo
cr»
rH
^
QJ
4-1
I4H
C8
Q>
CS
H
rH
fi
O
M
4-1
rl
C
^3
rH
Cfl
O
0
a
o
rH
ed
>
O
e
a)
P^
fsj
O
g^o
o
CT.
0)
S
3
en
CO
cd
a)
QJ
&
CD
>
o
B
OJ
S
M
fS
H
tH
CU
rl
^,
pq
.
oo
r^
c^
rH
*
O
CS
H
r\
C
O
M
j^
0)
C
01
H
QJ
U
M
3
O
CO
7-7
-------
g
d
H
Q
W
PS
H
Fn
fa
o
W
SQ
W
53
o
w
OS
o
NOTES:
34
32
28
26
24
22
20
18
16
14
12
10
8
6
ACTUAL
PROJECTED
J I
64 65 66 67 68 69 70 71 72 73 74 75 76 77 78
INITIAL OPERATION YEAR
(1) Includes units under construction but not yet in commercial operation.
(2) In this and subsequent figures, 5-year running averages are used to
smooth out variations caused by relatively small annual sample sizes.
Source: Foster Wheeler Corp., 1976.
FIGURE 7-4
AGGREGATE ANNUAL NUMBER OF INSTALLED COAL-FIRED UNITS OVER
25 MWe ON A 5-YEAR RUNNING AVERAGE
7-1
-------
tfl
SO
QJ
e
Q
os
M
I
o
3
w
w
Csl
to
Ed
O
750
700
650
600
550
500
450
400
350
300
250
200
150
100
50
ACTUAL
PROJECTED
i ii
63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78
INITIAL OPERATION YEAR
NOTE: Includes units under construction but not yet in commercial
application.
Source: Kidder Peabody Co., 1977.
FIGURE 7-5
AVERAGE SIZE OF NEWLY INSTALLED COAL-FIRED UNITS ON A
5-YEAR RUNNING AVERAGE
7-9
-------
7.2 Cost Analysis of Alternative Emission Control Systems
7.2.1 New Facilities
At issue are NSPS, which by definition apply to S02 emissions
from new, modified and reconstructed coal, oil and gas fired steam
generators firing more than 73 MW heat input. Facilities established
or those that begin construction prior to imposition of the standard
are not subject to NSPS. Hence, discussion of control systems and
the cost analysis will be limited to new electric generating plants
and those modified as defined in the proposed standard.
7.2.2 Basis of Cost Analysis
7.2.2.1 Key Variables. For each SC>2 system discussed in
Chapter 4, cost variation is governed principally by the following
variables:
Size of the generating plant
Coal used (i.e., sulfur content and precombustion costs from
source)
Averaging time over which the plant must meet S02 limitations
Level of control maintained.
In particular, the following control levels were analyzed as
scenarios:
Reduction to 90 percent of what the uncontrolled emissions
would be
Similar reduction to 80 percent
Maximum emission of 220 ng/J of heat input (0.5 lb/106 Btu)
of S02.
7-10
-------
These were compared with the base case: cost of providing the
control level specified by current standard of 520 ng/J heat input
(1.2 Ib S02 per 106 Btu heat input).
The effects of these variables and of particular ranges of
values chosen for them are discussed in the analysis reported in
Section 7.2.3. Cost is also influenced by the method chosen for
disposal of sludge from the scrubber. This consideration is also
discussed in Section 7.2.3.
Another consideration is the redundancy required in FGD equip-
ment. FGD systems installed must be reliable. Because system avail-
ability is increased through provision of a spare module, backup
equipment is included as a necessary item of the cost estimate. The
study incorporated one spare module of the FGD system which included
absorbers, pumps, tanks and associated equipment, for units larger
than 25 MWe.
7.2.2.2 Modeling Methodology. The method of determining costs
was to define typical plants and to model the expenses of operation
when values were applied to the variables listed above. This analy-
sis was carried out by PEDCo Environmental, Inc., under contract to
the U.S. Environmental Protection Agency. Flue gas desulfurization
systems, physical coal cleaning, and use of lower sulfur coal were
considered singly and as combined techiques. Cost differentials
Other control alternatives were analyzed in connection with
particulate controls.
7-11
-------
of boilers designed for western subbituminous coals versus eastern
coals and transportation costs for coal were taken into account: in
the analysis (PEDCo, 1977a; 1977b).
A typical new coal-fired plant was defined as a basis for
.%
calculations, assuming a midwestern location. PEDCo then calculated
costs for each control alternative, using computer programs developed
for the purpose. Mid-1976 costs were used as a basis with an annual
escalation rate of 7.5 percent through project completion. The cost
estimates obtained were expressed in August 1980 dollars. Five plant
capacities were selected for cost modeling: 25, 100, 200, 500 and
1,000 MWe. Other parameters and assumptions of the model plants are
detailed in the PEDCo reports (PEDCo, 1977a; 1977b).
Both eastern and western coals and lignite were among, the re-
presentative range of fuels considered in the cost analyses. Sulfur
content of the coals ranged from a low of 0.4 for western lignite to
about 7 percent for eastern bituminous. This high-sulfur coal was
used for analysis of a boundary condition although 7 percent sulfur
coal is rarely if ever burned for power generation in the U.S.
Results for coals with specific sulfur content are presented in
Tables 7-3 and 7-4 discussed in Section 7.2.3. Additional details
may be found in the PEDCo reports (1977a, 1977b).
East North Central Region (PEDCo, 1977a, Table 4-1, Page 4-8).
Different types of coal were (as noted below) considered in calcu-
lating costs.
7-12
-------
The costs of SC>2 control by an FGD system are affected by the
averaging time over which the emission regulation must be met. As
the averaging period decreases, the FGD systems must be designed to
cope with a higher average sulfur content of coal. This feature
results from the variability of sulfur content in the coal; over a
short period of time there is a greater likelihood of high average
sulfur content for that period than over a longer period during which
variability in the coal used tends to level out. Results within a
specific averaging period are affected also by variations in system
load and by features of the pollution control system itself,
particularly its efficiency, flexibility and reliability. For
determining most FGD system costs, an averaging time of 3 hours was
assumed. To assess the impacts of different averaging times in the
cost of FGD controls, a lime FGD system was modeled using four
different averaging periods, as well as a variety of plant sizes and
coal types. It was found that as the averaging time lengthens, costs
decrease irore strikingly from the increased variability of sulfur
reflected in the lesser amounts used during a short averaging time.
Capital cost for a 25 MWe plant was estimated to decrease by about
3.7 percent from the figures shown in Table 7-4 (from about $290 per
kW to about $279 per kW) with coal having a nominal content of 3.5
per-cent sulfur if the averaging time were lengthened from 3 hours to
a year. In contrast, the decrease reflected between these two
averaging times when the same coal is used in a 1000 MW plant was
7-13
-------
estimated to be 2.8 percent (representing the difference between
costs of about $116 and $112.66 per kW). Also, the cost penalty for
reduced averaging times is decreased as the sulfur content of the
coal decreases. For 0.8 percent sulfur coal cost decreases between
a 3-hour and a one-year averaging period are 2.8 percent for a 25 MW
plant and less than one percent for a plant of 100 MW capacity,.
7.2.2.3 Components of Cost. The study considered capital costs,
both direct and indirect, and annual operating and maintenance (O&M)
costs. Direct costs represent purchase of equipment items and the
costs of labor and material necessary to install the facilities and
connect the systems (e.g., site development, sludge disposal, piping
and electrical work). Indirect costs include engineering costs,
freight, interest, taxes, spare parts, land required for sludge
disposal, and other necessary outlays that cannot be charged against
any particular equipment items. It should be noted that replacement
capacity required as a result of the energy penalty associated with
SO control is taken into account. The total direct and indirect
2
costs are expressed as capital investment required in dollars per
kilowatt ($/kW) of rated capacity. Capital costs are also translated
into annual fixed charges classified under the components of depreci-
ation, taxes, insurance and capital charges, i.e., annual interest.
The fixed charges can then be expressed as mills per kilowatt hours
(mills/kWh) of electrical energy provided as annual output.
7-14
-------
o- vc f-i o s:
i (SJ -H i
7-15
-------
< OiT,
i CO vC
O f^- O O1* vT
-
-3- -H CN siJ CT'
IA CO CO
O r- in
E * -if
E STS!
H 3
fc. ~^
g""
^
1
o!
i i r-- vD
O Is- CO
£> \D O
Q *_l r^j
O (N m
££3
Si rn r5
* «* n0>r^j5Ln OIA«T
Sg ggSSS S2S
^or*"> 'ninLAoOLn c^f*^
S3 KSSSS CSS
OO-i mOvDtNCl CTsrHCO
_ cNjfM.H,-,^ _^ CC CO
r*- m ccm-HCOf^ &
-------
. 4
5 I
S L* "3 '
5 i ^* » i
S j
-------
Annual O&M costs of an SO control system represent the costs of
utilities required, including fuel; operating labor; maintenance and
repairs; and energy penalty costs (Section 7.3.2) resulting from the
fact that the control equipment uses energy and results in a require-
ment to generate additional power output. Also included are overhead
expenses for safety, employee benefits, engineering and legal services
that cannot be attributed to any specific part of the control process.
7.2.3 Estimated Control Costs
Estimates of the costs of the various control alternatives to
utility electric generating plants are summarized in Tables 7-3, 7-4,
and 7-5. These tables show the capital costs in terms of dollars per
kilowatt, which are then amortized as fixed costs expressed in mills
per kilowatt-hour at approximately 4 percent. Fixed costs are added
to the O&M to get the total costs in mills per kilowatt-hours.
Table 7-3 provides costs for meeting current SO control standards of
520 ng/J (1.2 lb/10 Btu) as a base case, which may be compared with
the costs of providing 90 percent reduction (Table 7-4) and meeting
an absolute level of 220 ng/J (0.5 lb/10 Btu) (Table 7-5). It is
seen from the tables that economies of scale result in considerably
lower (but non-linearly decreasing) costs for larger plant sizes.
The incremental costs of providing 90 percent SO reduction over those
of the baseline case are listed for lime and limestone in Table 7-6.
*
Costs for 80 percent SO reduction were not supplied by PEDCo.
7-18
-------
3
4J
PQ
O
O
vO
w
rJ
ca
O
Ed
s
fa
o
CO
H
CO
o
o
H
Q
Ed
Pi
S
O
U
CM
O
cn
O
o
Ed
Ci
fa
O
CO
o
a
H
Z
a
s
ft!
z
Ed
Ed
z
o
o
4=
3
cd -^
4J en
Ol
r~l
E-i r-l
H
e
a U "§
fa X ^
w fa "s
o
H
|*|
-1 0 "@
cd
a*>^
^~«
o
3
O rH
E * rH
iH
e
0 JS
o» 3
«v< O
(^ T tl
H -*-
3* e
fa
Ed "3
S ^3 ^i
rJ 0 ~6
1 1
rH J4
cd
3
r3 >.
CO U
5\ O
* « 2,
cu cd co
a O 4J
I>i 4-i
H v-< re
CU 3
1 rH C3
C\3 *rH tMj
3 0 i>
rH L£) lH LO T 1
oo oo
rH iH rH iH O
ej\ O CJN CO o^
O O r*^ LO ^
iH rH O O O
cn rH tH ^O iH
vo *o LO r^ -*
CM tH rH tH iH
O O O O O
LO -* cn cn CM
O 0 0 0 O
O O O O O
LO LO cn -* CM
rH rH tH rHrH
O O O OO
CM rH LO ^O LO
rH rH O O C
cn c*i i*^ r*^ v^-
CM rH iH CM iH
o o o o o
o o o o o
o o o o o
-
oo LO
,f^
c i
CO
^o
VJ
CTJ
"^
Ci
cd
CO
4J
C
OJ
r-l
Xj
3
CJ
4J
CU
CD
£
O
4J
a
o
fa
o
c
CU
r4
rH
3
eT
cu
M
0
4H
-a
cu
S
3
CO
Cfl
cd
C
0
H
4J
cd
3
4-J
H
cn
cu
C
H
rH
CU
en
cd
.fl
cu
.C
-
cu
o
^
rH
C3
JJ
C
0»
e
0)
|u
CJ
G
H
,
CU
r4
cd
CM
O cd
co r*^
r^-
4-1 CJ>
C rH
0)
CJ
i-l *
cu
a o
cj
0 Q
(^ Ed
a.
Ofl
2
H
> CU
O 0
S i-
CU 3
!- O
00
7-L9
-------
Combined precombustion cleaning of coal with use of FGD was also
considered and is discussed later.
A striking, although not unexpected feature of the PEDCo analy-
sis, is that the incremental costs vary inversely with the sulfur
content of the coal. That is, it is much less expensive incremental-
ly to go from a 520 ng/J (1.2 lb/10^ Btu) standard to a 90 percent
reduction of SC>2 from high sulfur coal (e.g., 7.0 percent) than
* from coal containing smaller amounts of sulfur (e.g., 3.5 percent).
These results are summarized in Table 7-6 for the lime and limestone
FGD alternatives.
Here it is seen that the incremental capital costs (over the
baseline situation of maintaining current NSPS) of providing 90
percent 862 reduction for eastern coal containing 3.5 percent sul-
fur range from $8.30 to $21.63 per kilowatt of capacity for using
lime FGD or on the order of 1/10 the total costs (as given in Table
7-4) for this scenario. On a mill per kilowatt-hour basis, the
incremental costs of this method add roughly 1/10 of a cent to the
cost of each kilowatt-hour of output. For high sulfur (7 percent)
eastern coal the incremental costs are almost negligibleless than
$l/kW capitalization and a fraction of a mill per kWh in the extreme
s ituation.
fFor low-sulfur western coal (assumed in the study to require no
controls in order to meet current standards) the total costs of re-
moving 90 percent of 862 would represent incremental expenses that
7-20
-------
add from about 1 to 2 cents to the cost per kilowatt-hour (depending
on generating plant size). The projected total costs of meeting
the 220 ng/J (0.5 lb/10 Btu) SO limit (as shown in Table 7-5) for
western coal represent incremental control costs over those required
by current standards. It should be noted that for low-sulfur western
coal it is less costly to meet the 220 ng/J (0.5 lb/10 Btu) level
than to reduce the SO emissions by 90 percent because the 220 ng/J
level requires less than 90 percent reduction of S09.
It is also noteworthy that the combination of precombustion
cleaning of coal to remove pyritic sulfur followed by FGD was not
found to be cost-effective when other alternatives are applicable.
The study indicates that the combined methodology would be appropri-
ate only with coals so high in sulfur that FGD alone would not meet
NSPS. This situation is represented for 7 percent sulfur-content
eastern coal (an extreme case) in relation to the 220 ng/J (0.5
lb/10 Btu) standard in Table 7-5.
Costs of SO control are also influenced by the method used
to dispose of scrubber sludge. Ponding and landfilling are the two
basic methods. The costs developed by PEDCo are based on an assump-
tion of sludge disposal in an on-site pond lined with clay. The
sludge would be stabilized by addition of fly ash and lime. This
method is calculated to add 1.15 mills/kWh to the cost of electri-
city generation. Alternatives such as the use of synthetic lining,
chemical fixation and pumping are estimated to add additional costs
7-21
-------
from 0.15 mills/kWh for proprietary fixation to more than 3 mllls/kWh
for trucking over extended distances.
7.3 Other Cost Considerations
7.3.1 Additional Capital and Operating Costs
In addition to control of SO emissions, a number of environ-
mental regulations apply to coal-fired generating plants. This
section considers the principal capital and operating costs associ-
ated with these plants. The analysis is based on a model 600-MWe
coal-fired unit, corresponding to the size chosen for the represen-
tative plants analyzed in determining costs of SO control (see
Section 7.4.1). Estimated capital and operating costs for the major
items of control considered for such a plant are shown in Table 7-7.
TABLE 7-7
ENVIRONMENTAL CAPITAL COSTS FOR REPRESENTATIVE NEW PLANT3
(1975 dollars)
Control Device
Unit
Cost/kW
Total Industry
Cost, millions
of dollars
Annual O&M Costs,
mills/kWh
Chemical Effluent
Treatment
Mechanical Cooling
1.52
0.9
a
Based on a new 600-MW coal-fired unit.
Source: Temple, 1976.
7-22
0.3
Tower
Entrainment Screens
Total
5.77
4.08
11.37
3.46
2.45
6.81
0.2
0.1
0.6
-------
TABLE 7-8
CAPITAL EXPENDITURES 1975-1985 BY TYPE OF
POLLUTION CONTROL EQUIPMENT
(Excluding equipment built for reasons other
than compliance with Federal regulations)
Water Regulations
Capital Expenditures
billions of 1975 dollars
Chemical Treatment
Cooling Towers
Entrainment Screens and
Cooling Towers
Total Costs
1.2
2.6
0.6
4.4
Source: Temple, 1976.
TABLE 7-9
O&M EXPENSES FOR THE INDUSTRY
Impacts
1975-1985
billions of 1975 dollars
Water Regulations
Air Regulations
Total Costs
6.1
19.0
25.1
Source: Temple, 1976.
7-23
-------
Table 7-8 shows the total capitalization requirements attribut-
able to control costs of water regulations in the electric utility
industry for the period 1975-85. The cumulative operating and main-
tenance expenses associated with this equipment between 1975 and 1985
total $6.1 billion as shown in Table 7-9.
7.3.2 Energy Penalty Costs Associated with SO Control
Table 7-10 presents the energy penalty costs in mills per
kilowatt-hour resulting from the imposition of selected control
systems for particular plant sizes and coal types under specified
scenarios. It should be noted that these costs have been included in
the cost tables of Section 7.2 as part of the total costs expressed
in mills per kilowatt-hour. Therefore, they are not additive to
other costs shown but are presented here for consideration as part of
the control costs.
7.4 Economic Impact of Alternative Control Systems
7.4.1 Increased Costs to Utility Industry
7.4.1.1 Method of Calculation. The economic and financial
impacts of the NSPS on the electric utility industry reported here
are based on results developed by Teknekron, Inc. (1978). Impacts
were calculated for the nation as a whole and regionally for indi-
vidual measures under specific scenarios that differ as to assumed
growth rates for production of electric energy and imposed control
standards.
7-24
-------
to
<
a
z
U
o
g
1
o
H
c
J
"Q
c
CO
cn
en
0)
]_,
eu
Q
Cx.
o
o
1
c
CO
£
rH
0)
H
H
flj
rH
(U
rH
,0
0
p
OJ
H
O
3
iH
01
OJ
C
it
rt)
1£
0)
g
jj
fn
OJ
g
3
o
s
H
*-5
CN o
O 'H
cn ij
u
O T)
0, U
CN
O 3
LO -U
ca
PQ
o
£N rH
-!
C
CN O
0 -H
cn u
u
5-^ 3
0 -0
c^i dJ
CN
O 3
p3
ca
^j \O
o
CN rH
rH
C
CN O
0 -H
CO JJ
o
o -3
<7\ i)
Q-
^
CN
O 3
cfl
O
^
CN
^ 3
03
rJ^D
O
.
O
c
CN O
CO u
o 3
, ^
Ofi
-^
° 5
33 sq
j^^
h 2
C3 -J
* 1 '3 1
rHOOcTi O mCNCN O
rH-Hi-HO * i-Hr-t'-H rH
tO LH CN O
i-H O O O
P*» CO I/1! ^ lO ^O ^ CN ^>
ONCOCOCX5 CO rHOO CO
OOOO O IrHrH O
00 O r^ U"!
CO cO r^ r-^
- (t ct) ctj CQ
OOOO
sO t/^ CN O^ CN ij\ ^O CN CN
r-tOOCTi O flCNCN O
^-HrHr-HO «-H i-H rH « 1 rH
in i/*i CN O
O O^ C^ C3>
r-1 O O O
r~^ 3O u^
O\ CO CO
O O O
OCQCOCQ CO CO r-JOO CO
O'OOO O rHr-^r-i O
CO O r-. ul CO r-^ r-
_ "°_ "^ ^ t ^ ^
o" o o d CN d o
r (
tnooo o o moo o
CNOOO O O CNOO O
"^ ft
1 """'
I j_,
3
DX3 -,
0
LO O tj^ CO CO xj1
A If *
~~ ~~ ^
! .0' C C ~ T L. il .tj
i ^' -^ ^ ^j ^ ~
o o
Z CJ
-------
TABLE 7-11
ALTERNATIVE TEKNEKRON NSPS SCENARIOS
Teknekron Scenario Label
Revised NSPS Maximum Emission Level in
lb/106 Btu (% Removal)
Moderate Growth Scenarios :
Ml.2(0)0.1 (Baseline)
S02 -1.2 (0)
NO
x
0.7
Particulates - 0.1
Ml.2(90)0.1
S02 -1.2 (90)
NO ==0.6
x
Particulates = 0.1
Ml.2(90)0.03
Same as Ml.2(90)0.1 but with
Particulates = 0.03
Ml.2(80)0.03
S02 =1.2 (80)
NO
x
0.6
Particulates = 0.03
MO.5(0)0.03
so2
NO
x
Particulates
0.5 (0)
0.6
0.03
High Growth Scenarios :
HI.2(0)0.1 (Baseline)
S02 - 1.2 (0)
NO
0.7
HI.2(90)0.1
Particulates =0.1
S02 - 1.2 (90)
NO
0.6
Particulates = 0.1
HI.2 (90)0.03
Same as HI.2(90)0.1 but with
Particulates =0.03
HI.2(80)0.03
S02 =1.2 (80)
NO =0.6
x
Particulates = 0.03
5.8% per year to 1985; 3.4% thereafter.
35.8% per year to 1985; 5.5% thereafter.
Source: Teknekron, Inc., 1978.
7-26
-------
The present standard of a maximum emission rate of 520 ng/J
(1.2 Ib SO./10 Btu) heat input is used as a baseline case in order
to assess the effects of the following assumed standards:
o A reduction of 90 percent of 862 emissions over what the
uncontrolled level would be with a ceiling of 520 ng/J
heat input
o A corresponding reduction of 80 percent with a ceiling of
520 ng/J heat input
o A maximum of 220 ng/J (0.5 Ib SO /10 Btu) regardless of
the degree of reduction.
Scenarios for which calculations were performed by Teknekron
*
are listed in Table 7-11.
Regional results are allocated among 10 regions, nine of which
correspond to those used by the U.S. Bureau of the Census. However,
the Mountain Region is divided into northern and southern portions
which are separately allocated. A listing of the states in each
region is provided in Table J-4 of Appendix J.
In performing the calculations, Teknekron used a computerized
utility simulation model developed for EPA's Integrated Technology
Assessment program (Teknekron, Inc., 1977a). The financial module of
To provide a more consistent basis for comparison with the base-
line situation, results in this report generally reflect for the 90
percent SO^ reduction scenario the Teknekron calculations assuming
0.1 Ib of particulates/10° Btu. Although this particulate level was
not calculated by Teknekron for other scenarios, differences between
the 0.1 and 0.03 Ib level for particulates in the 90 percent 862
reduction scenario were so slight as to indicate that no significant
discrepancies were introduced by using the 0.03 Ib particulates/10
Btu level elsewhere.
7-27
-------
the model is an accounting structure that treats financial transac-
tions (in this case, those associated with construction and operation
of electric generating plants) in a prespecified way. It simulates
the treatment of these transactions by regulatory and tax authorities
and readjusts revenue requirements for the coming year. Pollution
control cost data were adjusted to be consistent with those developed
by PEDCo, as discussed in the preceding section.
The computer simulations were based on meeting the requirements
for power as projected by each scenario (Table 7-2). For existing
generating units and new units that have already been scheduled, the
simulations used data provided by the utility companies concerned
which included size and authorized or projected site locations. For
the years after 1985, for which data on plant schedules are largely
nonexistent, the simulation determined the need for plants in par-
ticular locations to meet power growth requirements. In simulations
involving the post-1985 period, an average coal-fired unit of 600-MWe
rated capacity was assumed for new units and site locations were
selected, with counties containing Class I or nonattainment areas
excluded.
Included in the output from the Teknekron simulation runs were:
Component costs of supplying electricity and the effects
that the various proposed NSPS will have on capitalization
and costs for pollution control, for fuel, and for other
operation and maintenance. (Section 7.4.1.2.)
Effects that alternative revisions of the NSPS can be
expected to have on key parameters indicating the financial
status of the electric utility industry. These results are
discussed in Section 7.4.2.
7-28
-------
National and regional effects in terms of relative increases
in the price of electricity (Section 7.4.3).
7.4.1.2 Effect on Costs of Generating Electricity. The nation-
wide costs to the utility industry of generating electricity for the
baseline situation (assuming that the current maximum of 520 ng/J
of SO is continued) and for the alternative revisions considered
for NSPS are shown in Tables 7-12 and 7-13. These tables show for
each scenario the total costs (on a national basis) that the industry
must meet in the period of 1980-1995. Also shown is the incremental
cost of each alternative in a monetary unit and as a percentage
increase. Incremental monetary cost is determined as the difference
between costs under the assumed NSPS and costs in the baseline situa-
tion. The percentage increase is determined as the ratio of this
difference to the baseline cost. Further information includes the
allocation of costs to pollution control, fuel and other expendi-
tures. Table 7-12 gives the results for moderate power growth and
Table 7-13, a high growth rate.
Costs to the industry of alternative NSPS are measured in part
by the proportion of yearly expenditures that must be allocated to
pollution control as compared with other major outlays such as fuel
and O&M. The distribution of costs are shown nationwide in Tables
7-12 and 7-13. Both investment requirements and operating costs
(i.e., operation and maintenance of pollution control equipment) at-
tributable to pollution control are compared with other expenditures.
7-29
-------
><
H
M
M in
c^ ON
is*
rJ 1 O
W PCj
vO O
O oo
g o> erf
H rt 3
rH 2 PM &<
1 W «
C£^ r_1 fij
1-J > U
CB ru M H
< O H -i
H -i > OJ
u o to
0 ffl
>-> CO
cO JJ
rH tfl
Q
^J
c; a)
i u5
0) QJ rH
H > CJ
o o 10
C ffl
H PS
.
rH O
0 O
Q
^
CO
CO
rH
o
o
H
«
tj
0
rH
rH
LP m
O L-l
CM CM
O *O
CM CN
\C \&
, r
rH O
f*- m
rH O
rH rH
m o>
c> n
^ CN
.-H 0
O
d
Cl 4J
> to
01 o
Q- U
rH rH
CS fO
4J 4J
o o
H H
j
ON C o rn m in
1 rH |
cr> < so CN CN CM
m rH en (H
m CM m o CN <£>
O f^" O f^ *J "-fl"
1 rH 1
m vo ON o mo
. *
\D-3- vO CN CM CM
CO »H fH rH
CO -f VT 0 m M3
ON vD r* co in m
C*") rH C^1 rH
G ^ 0
JJ O U -W
C *H C C
O JJ tO O f-f
C> OJ C L> W
r-( QJ JJ 4J 4J
C OJjj CC K c -H
O CUC OQJ (DQJ >w
^H O*H -HE XE O
3 V-t E ,3 CO OCO CM
H rH CU rH OJ QJ
0) rH XT3 rH> rH> 4J
3 O JJC OC rHC QJ
fan pi OtO ChrH
rt
Q
cx
CO
en
to
H
J=
O
QJ
X
3
cc
o
(J
QJ
C
H
rH
O
n
ca
o
jj
c
E
QJ
M
e
H
14H
O
O
H
4J
CO
c;
o
CO
O".
rt
CJ
c
IK
c
c
QJ
C
QJ
f-
41
Q)
t-
3
O
7-30
-------
I
l-v
w
03
U I
rH SO
Ctf CO
f-l CJ>
U rH
W S
J en p-i
W PJ 3
en c
a a, A
z o
M PL,
H o a:
o
W H
ej < fe
o w w
H H
O Q EC
O W O
H M
W O 35
Q W
I-H _3
3 Ed
2 e/5
rH pi
EH W
Q
CU
PS
CN
0
00
o
co
'4H
rH
CO
>
o
a
CU
CN
o
b
ON
cu
d
iH
rH
0)
CO
CO
«
M
> /-N
O CU CO
d IH
4J iH CO
d rH rH
CD CJ rH
S CO O
CU CO TJ
M « v-'
O
d
M
Mp£
CO U
rH CO
rH 0
0 CJ
Q
CU
O cu en
d )H
4J lH CO
0) CU rH
S CO 0
cu ca T3
W pO ^"^
CJ
d
&HfJ3
CO 4J
1-1 cn
rH O
O U
Q
CO
CO
rH
rH
O
Q
C!
o
CO
u
o
rH
*^
CU
m co
.
rH CN
-a- to
0)
rH O
r~- O
co m
CO CM
*. A
rH rH
0)
rH rH
01
rH rH
CO CO
4-1 4J
O O
H H
r~~ rH 00 r-l CO CO
*
CN rH O O CN r- CN rH
.
m rH ^o t** r~- co
co in ox * u-i 0)0)
H rH O 'H -rl C O S^ S
4-1O CO 4JrH )-l 4JiJ
3 >H IH S 3 PU CM
rH r 1 4-1 0) rH rH 0)
0) r-HC -S"C rHO 4J rH>
3 GO 4JC OU CU rHC
^ Q-.CJ OC3 CU4H 22 ^CrH
3
u
ca
o
rH
CN
O
C/5
i 1
CN
rH
00
c
o
CN
in
o
co
a
d
CO
ij
CO
4-1
d
cu
to
0)
)H
a.
U-l
0
d
0
H
4-1
CO
3
d
rl
4-1
d
o
CJ
3 .
Ul
}_l
CO
rH
rH
o
a
m
r^
CT>
rH
14H
O
X
d
0
rH
i 1
H
S3
J2
rH
O
rl
4-1
d
o
CJ
d
0
H
4J
3
rH
rH
0
Cu
to
d
rl
T3
3
i 1
O
^
Cd
O
cu
3
rH
CO
>
cu
d
rl
rH
CU
CO
CO
O
4J
4J
cu
^
01
u
CJ
d
H
MH
O
O
H
^j
CO
T)
totals because miscellaneous costs were not included
o
4J
e
3
CO
4J
O
d
o
CO
S
CU
4J
iH
a
01
rH
iH
CO
4J
0)
01 l
3
4J
M
vO
O
rH
£1
rH
rH
0
M-l
0
rH
CU
OJ
rH
01
4J
CO
rH
3
CJ
H
4J
M
CO
D.
CO
00
d
'e
3
TO
CO
4-1
CO
r^»
ON
rH
t.
.
O
d
I-H
fs
d
Q
^
11
d
0)
P-nj
.,
^i
CJ
v^
a
o
co
7-31
-------
It should be stressed that allocation of costs to pollution con-
trol provides a somewhat different representation of the financial
impacts of NSPS from that given by an incremental increase in total
costs. Differences in the kinds and amounts of fuel and in O&M
requirements occasioned by NSPS are not attributed directly to the
cost of the revised control standards; instead, these differences
appear as part of the fuel and O&M costs.
It can be seen from Table 7-12 that there is little variation
in the effects of the alternatives. Under moderate growth all are
projected to increase total costs by about $26 billion, a percentage
increase of 2.52 to 2.61. As would be expected, a major component
of the increase is for pollution control. Under the assumption of a
high growth rate of power, total costs increase by about 4.5 percent.
Tables 7-14 and 7-15 expand the operational pollution control
&
costs to regional figures. Baseline (i.e., assuming maintenance of
present standards) costs are compared to those under alternative NSPS
revisions and incremental expenditures of the revisions are shown.
For the moderate growth of power, alternative revisions considered are
90 percent SO^ reduction, 80 percent SO reduction, and a ceiling of
220 ng/J (0.5 Ib S02/10 Btu). Only the first two of these scenarios
have cost estimates shown for the high growth assumption. Here marked
disparities appear among the various regions in the scenarios. The
Regional costs to consumers in the price of electricity expressed as
mills/kWh are discussed in subsection 7.4.3.
7-32
-------
burden (on a percentage increase basis) will fall heavily on the
utilities in the West South Central and Northern Mountain regions
and, secondarily, on those in the South Atlantic and Pacific regions
and (under high growth) also in the Mid-Atlantic and North Central
regions. In both the moderate and the high growth scenarios, maximum
costs occur under the assumption of 90 percent renewal of S0_. In
several regions, costs for the ceiling of 220 ng/J (0.5 Ib SO /106 Btu)
(moderate growth assumption) are estimated to equal those of 90 percent
SO reduction.
Under moderate growth, costs will increase in the West South
Central region by as much as 80.5 percent, up to $3.3 billion; and
under high growth as much as 84.8 percent, up to $3.8 billion. In
the Northern Mountain region under high growth maximum incremental
pollution control costs would be $0.4 billion, an increase of 133.33
percent. Under high growth, the Pacific region could face cost
increases up to 58.3 percent ($1.4 billion) and under moderate growth
increases of more than 26 percent ($0.5 billion). South Atlantic
region increases run up to 38.8 percent ($2.1 billion) under moderate
growth and up to 38.8 percent ($3.1 billion) under high growth.
East North Central increases under high growth could amount to $2.6
billion, an increase of 25.2 percent. Under high growth, the Mid-
Atlantic region is projected to increase as much as 69.8 percent
($3 billion). By contrast, New England and the East South Central
regions will be relatively unaffected (increases from virtually
7-33
-------
r-J
m
cd
Pi in
O ON
fa ON
O ^"O
M OO
O ON
td H
>H CO
O
CO H
H CO
CO M
8£
Pi
hj
O co
jyj p..|
H co
IS
O H
H S3
O
PH
W
H
^
CU
o
P-I
14-1
4-1
O
o
cu
4-1
cd
rl
cu
-O
o
3
4J
M
O
rH
^^
CM
0
CO
rH
.
O
c
0
H
1 t
O
3
CU
Pi
CM
O
CO
b
oo
G
o
H
1 i
CJ
3
T3
ru
CM
O
CO
b
Oi
cd
CU
G
H
rH
cu
CO
cd
U Q)
4-> CO
G cd
cu o)
CJ r-4
rJ O
O
00
CM
O
ON
I 1
f^
CM
Os
OO
f~-
.
,
CO
rH
ON
0
r-
o
^f
CJ
H
tlant
*3j
CO
in
CO
00
00
m
ro
CO
oo
in
.
CO
oo
00
m
oo
CO
in
rH
cd
r-J
4J
C
0)
u
£5
.
w
m
CM
rH
,3-
in
CM
rH
-*
O
m
CM
^j-
o
J3
OJ
^
OJ
CO
^c
CM
-*
Oi
ro
41
^O
OJ
-------
in
T 1
1
I--
w
, *]
CQ
<^
H
O
pL)
O
M
O
w
Pi
>-<
«
CO
H
CO
O
O
t -1
O
PH
H
X
a
CJ
IS
o
M
m
01
01
rH
1
**o
oo
01
rH
CO
0
M
CO
M
>
td
CO
PH
£^J
w
>
H
H
H <
iJ Id
O H
a)
0
PH
MH
O
43
1
O
43
00
H
33
rH
>
O
e
cu
CN
0
CO
oo
rH
cfl
O
e
CU
tf
CM
O
01
rd
cu
H
rH
01
CO
cfl
pq
0)
4J CD
C CO
cu cu
Ol,
M
s-i a
PH M
m
01
I
00
01
rH
CU
4-1 CO
f3 cfl
01 01
O 5-i
00
oo
a
si-
00
m
00
00
OO
CJ
f->
vo
in
CJ
0
O
<^
O
rH
01
r-H
fl
0
H
4J
cfl
8
cfl
1
1
rH
r-l
rH
O
CN
rH
rH
rH
00
0
*d
d
cfl
rH
60
[3
H
S
S
13
in
o
vD
01
vD
00
o\
vO
00
r^
00
si-
00
CN
CJ
H
^j
C
cO
rH
4J
<£
1
XI
rH
g
in
CN
CO
VD
O
rH
00
oo
OO
rH
rH
H
0
OO
0
si-
O
H
4J
cfl
rH
| 1
"^
.
CO
sr
rH
CN
m
CN
rH
CN
in
CN
01
CN
H
00
O
. 1
r""i
O^
sr
rH
CO
5-1
4-1
Pi
CU
CJ
£2
Id
00
m
o
CJ
*
C/D
Id
0
CN
00
00
00
0
CN
00
00
oo
m
CN
CN
rH
rH
Cfl
5-1
4J
C
U
&
^
r-.
rH
1 .
01
|^
OO
st-
oo
m
00
vO
sf
Oi
rH
rH
CO
S-I
4-1
ti
cu
U
»
CO
«
3
^
^«o
vD
in
o
00
00
00
00
r-H
r^
o
00
o
rH
d
c
H
cd
4J
C
3
)5d
*
oo
^j-
T-l
rH
00
in
00
! *
CN
00
1^-
CN
in
rH
C
H
cfl
4-1
C
3
O
CO
0
o
m
^
oo
oo
oo
m
oo
00
sl-
CN
OO
O
a
H
MH
H
O
CO
PM
r SC>2 , NOX, TSP and related pollution abatement equipmen
tandard-520ng/J (1.2 Ib S02/106 Btu).
o
MH
CO
0)
CO
(3
cu
Cu
X
0)
0)
-d
3
rH
O
C
H
CO
4_J
Cfl
0
CJ
a)
CJ
rjj
Cfl
C
a)
4J
rjj
r-l
Cfl
e
n-j
c
cfl
c*
o
-H
4-J
c<3
5-i
cu
fx
o
CO
4-1
n
cu
CO
CU
y_l
ex
MH
0
e
o
H
4-1
CO
3
C
H
4-1
n
o
u
.
Cfl
5-1
cfl
rH
rH
0
d
in
r^
01
rH
MH
O
CO
rjj
O
H
rH
rH
H
M
3 43 a T
ine scenario.
rH
01
to
cfl
43
6
O
l_i
14H
CU
&0
cfl
O
4-1
£3
01
U
^_j
cu
4-1
C
cu
CJ
5-i
cu
a
rH
MH
O
O
rH
rH
C
cfl
4J
CO
CO
CU
CO
01
rH
A
O
c
H
*
d
0
^
CU
y
CU
H
CU
0
£_(
3
0
CO
3 0)
7-35
-------
nothing up to about 13 percent and $0.5 billion). Under moderate
growth, the East Central regions will receive among the lightest
impacts (percentage increases from between 2.5 and 5.0 percent and
dollar increases not exceeding $0.3 billion).
7.4.2 Financial Impact on Utility Industry
The impacts of the alternative control levels under NSPS on the
electric utility industry are measured by several key parameters
indicative of the industry's financial status. These parameters were
calculated by the Teknekron model. Table 7-16 shows the capital
investment requirements for the Electric Utility Industry (EUl) on a
nationwide basis under assumptions of both moderate and high growth.
The total capital investment required for the baseline scenarios
(maintenance of present standards) is given in billions of 1975
dollars. The required increments are shown for pollution control and
other expenses under the selected NSPS revision. As shown, the over-
all percentage of changes attributable to revised NSPS is smallabout
3 percent for moderate growth and slightly over 5 percent for high
growth. However, the increase in capital investment for pollution
control is more than 43 percent (about $14.6 billion) for moderate
growth and nearly double the percentage to 82.3 ($34.4 billion addi-
tional) for the high growth scenario.
Incremental impacts of alternative NSPS revisions on total U.S.
capital investment, Gross National Product (GNP) and new debt and
equity issues do not appear significant. Even under the high growth
7-36
-------
VD
r-H
I
ta
H
W
S
E *
CO
W
^
H
^
H
H
p -
§
icipa
d
3
S
T3
d
cd
T3
C
H
s_ ^
E^
M
CO
3
T3
d
M
>,
4-1
rl
H
4-1
^
O
H
4J
CJ
0)
in
ON
VO
i^
ON
1 1
X
cn
d
o
cn
H
>
cu
Pi
cn
PH
CO
£^j
01
rl
4-1
CO
e
01
M
1 1
Vi
01
d
J^3
/*,
CO
J_J
cd
rH
rH
O
T3
in
r>^
ON
rH
MH
O
CO
d
0
H
rH
rH
rl
^
W
CO
i PH d
i-H 4-1 CO CO O
H d o) o d a -H
cfl cu oo 4-) ^ cn
M O C 0) rl
o) t-i cd 01 4J > >
> 0) -C 3 rH -H 0)
O P i O T3 ^ 4-1 Pi
CO
1 PH
1 1 4-J | CO
d -H co o> M is d
r-H CU o. CU 3 0) O
rHScO>T34-IOl-H
CO 01 CJ C rH > CO
>J )-l M 4-1
K* d CO rH CU O Cd CU
C 5 hH 4-1 Cd B 4-J Cj p^l
CO "
1 PH d
CO CO O
0) 0 d 5S -rl
oo 4J M cn
d cu cu -H
Cd 0) 4-1 > >
fl 3 rH -H 01
U T3 <3j 4J c2 6*S
id cu i
d -H 3 CO
M P! 13 d CO
4-> O SH PH d
rH C -r-i rH 0) CO O
CO 0) 4-1 O 4J 53 -rl
4-1 6 3 S-l rH W
H 4-1 rH 4J <3 0) vH
ft cn rH d > >
Cd OJ O O Q -r( 0)
CJ > p., C_> 4-1 4J K
o d
4-1 O
CO
3 -rl -H
T3 4J >
cd cu
QJ d pi
00 i-J
d 0) OT
cfl 4-) PH
& rH CO
O B -H o rH cn
H 4-1 13 -H rH >
cdd)rH3>HO-HQJ
0>OrH4J4->4JP*!
d
01 O
d 'H 1 d -H rH
H CO 4J !-) 4-1 O
rH 4J Cn 3 V4
01 VH 0) 4J rH 4-1
CO ft K" d i 1 d
cd cO d CU O O
PQ U H 6 PH CJ
fl
0) 1 O
d rH 1 X OO vH rH
H CO 4-1 Cl> d 4-1 O
rH 4-1 W -H 3 Vl
0) -H 0) 4-1 TJ rH 4-1
10 ft > d 3 i-H d
cd CO d 11 t-H o O
pq U M g 0 PL, U
r~\
4J
> QJ
O 4-1
V4 cd
i
i
i
I
i
I
in
co
CO
ON
rH
CO
m
cd
0) 0)
4-1 d
CO -rl
JH rH
CU 0)
""O CO
o cd
S W
iH
,
CO
^j-
r--
i-j
'^O
cO
_-^
CO
CO
r-1
m
,
o
OO
CM
1
1
d
o
CN»H
O 4J
w o
3
S-S TJ
O CU
00 Pi
CM
CO
»^J-
OO
rH
^J-
CO
^__
1^
^f
M
r*>.
.
0
O"\
v
CO
1
1
d
o
O 4-1
C/l CJ
3
§>,£
1
1
1
1
1
1
Cj\
(
rH
*
OO
vf
CM
CO
(D
d
H
rH
d cu
60 cn
-H cd
ft
d
o
H
4-1
cd
3
d
H
4-1
d
o
u
OO
o
c
rH
C
O
CD
H
0)
O
V)
3
O
C/3
7-37
-------
scenarios, the ratio of Incremental Capital Investment clue to NSPS
to GNP amounted to 1 percent by the year 1990. These findings
were based on the TRFNDLONG 0877 forecast of GNP provided by Data
Resources, Inc.* (Teknekron, Inc., 1978).
Table 7-17 shows the expected long-term external financing
by the industry in providing for the capitalization requirements.
Relative financial impacts are shown nationally and for each of the
10 regions under the scenarios in terms of three selected parameters
characterizing the FUI in Tables 7-18, 7-19 and 7-20. The following
are indicative of the industry's financial health and the degree of
difficulty likely to be encountered in financing the required growth
in electric power under NSPS.
Peturn on common (FOG) - equity returns or what stockholders
may expect in return for their investment in the EUI.
Interest coverage (1C) - ratios that tend to vary directly
with company earnings. If these ratios are low, they
adversely affect bond ratings, interest rates and prices of
stock shares, creating difficulty for the EUI in raising
the capital needed.
Famines quality (FO) - a measure of the extent to which
earnings represent cash and indicative of the extent to
which the FUI can expect to attract funds through investment
in stock.
The measures shown in these tables indicate that for the most
part the impacts in the FITI will be small. Whereas the estimated
mean value of ROC (Table 7-18) is for all alternative NSPS depressed
The year 1990 is the latest provided by the Data Resources Inc,
forecast.
7-38
-------
o
w
w
H
W
H
I
O
M
C
CJ O
-Q LO
ra !-
3 QJ
r~i ry
r
S- GO
-t-> a_
4-> GO
C <-
ro -(->
.C ra
O C
s_
4-> CJ
f~ ^J
CJ r
U ->
rd t
rj^ i
^- ^x
CJ
-t-> O
X 4-" 1.0
i 1 r-*
CJ O
S r- -r-
s- _a oo
QJ ro -r
-M + ' ~>
1 3 CT) CU
c c: >
O) -r- -r-
CJ C ro
1- 0 C
o c s_
E - CJ
i . U_ 4->
cn
c:
1
(^J
c~
rc
U-
,_
ra
^~
t.
aj
X
UJ
CJ
c
1
,
C)
to
ra
ca
1
-o
01
i- O
a> o
CJ GO
s_
Q-
C v^
o o
E 0
O GO
C )
E
S-
CJ -t-J
1 -Q
1 QJ
cno
c
o
I
a
QJ
S ^
S- 0
QJ O
t4 4_3
CJ GO
Q-
c ^
o o
£ O
^- 4_>
O GO
o
r^
i-
CJ -t-J
1 o
i cy
cr>Q
c:
0
1
QJ
^_ ^^
t- CJ
OJ GO
a.
J_ _^
o o
^ o
- -f *
O GO
C i
E
I
CJ -t->
r -Q
1 QJ
CJiQ
c~
o
1
n)
Moderat)
Growth
co
.
0
co
CD
r^.
CO
CM
.
LO
LT>
CM
crj
QJ
Basel in
;;-!
CO
,-J
c^^
CTi
1
to
^9
CO
1
0
0
to
<^-
.
CO
c
CM O
80% SO
Reducti
0 '-
LO
,_'
?, n
o
i\
u ^
LO
r
C\J
r
r
to
co
CO
1
1
1
1
£
CM O
90% SO
Reducti
CO
.
CO
r_
CO
co
in
i
f-^v
.
CO
to
CO
^
4-J ra
S 01
O £
o In
. 0)
.C CO
60 rrj
ffi ^
£ '
o
CO
0 r'
^o
*
03
0 2
r~~
CM
CO
CM
**o
CO
1 '
f^.
.
f^.
c
90% S02
Reductio
3
4-1
O
to
TJ
i-l
o
M
C
C
O
C
QJ
cn
Cu
C
o
H
4-1
tr)
G
H
C
O
u
G
c
c
G
Tekne
0>
O
(-<
3
O
to
7-39
-------
from that projected for the baseline situation, under moderate growth
assumptions virtually all means fall within the range of values
predicted under maintenance of the current NSPS. The only exception
is the West South Central region where the mean falls slightly below
the lower boundary of the baseline range. In the high growth situa-
tion, the effects are more pronounced. The national mean of 10.8
under 90 percent SO reduction is 0.2 below the lower level of the
projected range. This drop reflects the same condition projected for
three of the 10 regions, and the fact that in the populous Mid and
South Atlantic regions the projected mean for the 90 percent scenario
is near the baseline lower boundary.
In regard to interest coverage (Table 7-19), all projected means
fall within the baseline ranges. It is noteworthy that under all
scenarios as well as the baseline situations, the average ratio for
Mew England is projected to fall below 2.0 for at least 1 year of the
period 198^-95. Ouality of earnings (Table 7-20) are shown as little
affected by revised NSPS. It may, therefore, be considered that the
effects of revised NSPS on the industry under moderate growth would
be small and local. Under high growth, however, return would be
depressed nationally.
7.4.3 Effects on Price of Electricity to Consumer
A significant measure of the impacts of NSPS on the average
American is the change that can be expected in the price of electric
power to the wholesale and retail consumer. This effect has been
7-40
-------
"i
K
fl
CO
td
^ o
O oo m
£-« Ci> Cn
DO c:: i
2 TO 00
O Cn
£ C * H
^ °
U ro ic
01 o
2 "H *H
co O J-> i-J
-H -H TO
1 2 r-i C
r- oi -H o
ID J-J CJ
W H ;=> V.
J td
OS ci 0 C
E-" S-t -H
>H 4J 4-1
H U «
1 Q) c
O* W 0)
W 4J
2 OJ <
0 C
3 i-l
2 O cu
(V^ | --£]
;=> )- c
H OS
H -u
PS cn
0)
C
t 1
o
fa
A-
_
re
-
Z^
> "
i ,J
! -^
*r~
'
,
£
> ->
O"! Z
; CJ
i .1
* ^T~
i
'P
i ^
CO C
* o
1
-*J
"~. CJ
LiJ C__>
1
1
ir.
,
i
01
LU
^~
o
1
4->
ro
01
u j:
td AJ
QJ O
*U J-i
S °
'
) CNJ
LT1
1
^-. , . ,
"" ^ ^ L-~J
OO i i
1
LT> LJ"j
CC
c-< i O
CM 1
, " r '
C r-
&-; i co
un CM .
LD
V< ) ^C
0 I .
ex,1 ro
O r
r- CC
[
>-i I O
CTi C^i
CM ^ ^-
r r
S^ 1 CO
i l
CD 1
i cn
^ ^ ^ *
ro ex; i
i
cn i
ro r CM
' '
l
61 l "5-
«3- OD
. ro
CM i P
IU
C 01
ro c
£ 3.
- uo i
cn rv.
i
1 CO
CO CC
CM ' l
i r
1 CO
'
CD ^
CC
1
l ro
IO CM
CM1 CM i
1
Cn O .
ro . CM
r l
t
1 UD
cn ro .
CM
QJ
C Ul
z. a
C
o
T-l
4-1
6-5 CM U
O O 3
O CO TD
a
, i
CM C^
; ro :M r
1
1 '
, in CM
1 . .ro
1 ^ IU '"
i
i
i '
1 «5T
1 Ol '
i «-r CM ,
i
l
i 0
i
| LO r- O
i 01 cn
1
i 1
I CM
' ' ^ «-r
i ro CM i
1
1
j 1 LT)
CM 1
i CM CM
O f
i ^- a-.
i
' ^ j_co
| r l r
l r
1
1
1 ; ^
O r
1 2 o *"
{
t
1 ! ,-
1
I
1
, "~- CM
' UO r- CM
1
1
, 1
i cn
i LfJ O
. -CM
1 ' '
1
1
' OJ
^ 1 OJ ra
| S CC
^J I
0 ,
J2 a)
to ' c
H , 3
i «
1
cn cr
, ^Z '
i O
ro i
i ro
^z ^ ^~
i O^
LD 1
. cr o
i
i r^^
i
i ro
co ir>
. ro
i r--
cn co
i «s-
CC CO
O Or
'
1
r- cc O
cn co
l O
«-r Ln
. ro
r- O r
i
1 CM
IT) O
. CM
LT> J CM
1 Ol
CO f*") *
O O r
r '
QJ
c cn
CD ro
2: cx
C
0
H
6-5 cvl CJ '
003
CT> W -O '
CU
05
CC
cn
»i (
CJ
" C
,M
C
O
^
-------
C3
<
a:
W
J I
03 00
< LU
rn
ca
ra
c
o
cu en
C£ I
LO
rO CO
en
l/l l/l
O O
+-> s-
-r- rC
CL>
U
CJ C
r S_
T3 r
O <
2 i~
O QJ
i -a
1/1
CD
o
'£
r
U
ra
Q_
.
c:
oo 2:
.
c
s: 2:
r
2
4->
OO C
CD
3 0
_
rg
4-J
P" (77
eu
3 0
f
ro
-i->
OO C
cu
1 1 1 <^_)
r
i.
4->
7^ C"
0)
Lu CJ
*J
ej;
1
g
00
,__'
1
S
CD
z.
o
T3
-z.
CU
u
rfl j-j
J-i ^
11 o
T3 1-1
0 C
o
^ ,-j.
o ^r co
1 LO
. CO
OO CO
0
en
CNl CO LO
LO
. CO
CO CO
o
) 1
^-«-
CO r-~ O
r-. co
. CO
CM CM
o
4-J
«^- £\j ij£
«3- CO .
. CO
CO CO
o
^
CM CO O
co «=r
CO
CM CM
o
CM
O en r-»
LO CO
. CO
CO CO
CO
CM O SI-
CM *
O
o
^ r«- "3".
. CO
ro CM
0)
^ CD
CD ro
>' Cd
4)
C
T-l
r~j
4)
a
o
CO
oo LO co
. CO
CO CO
o
en
CO CM LO
CO CM
. CO
CO CO
o
CO
in en m
CO *~j"
LO
*3- CO
O
LO
CM co r-
l-O CO
CM
CM CM
O
+J
CO
CM UD CTl
U3 CO
. CO
CO CO
0
[ i
^
CO CO O
r^ LO
. CO
CM CM
o
UD
CO >-O LO
«3- CM
CO
CO CO
o
+J CM
r 00 O
co «=r
. CO
CM CM
O
O
r--. «^j- to
a- co .
CO
CO' CO
CO
CM O «-
LO +-> .
CM *
O
4-J
T CM CO
. CO
CO CM
CD
C en
CD ro
2: o:
c
o
u
B^? CN! O
O O 3
oo en T3
D
o
co
CO LO CO
. CO
CO CO
0
4-1 LO
rn i en
CO CM
. CO
CO CO
o
CM
r~-
CM LO LO
«* CM
CO CO
o
4~>
CM -K
O
<3- CM CO
i r-.
. CO
CO CM
CD
C CD
rO CT
CD ro
2: o;
c
o
H
1 j
5^ CM O
C O 3
en C/3 T3
0)
0
4_)
O
r-^ CQ in
00 r
. CO
CO CO
o
CO
3- en co
. en
. CO
CO CM
o
co
i CM i
CO CO
o
co
LO r l£3
LO "5J- '
CM
CM CM
0
co
o co ^r
CO i
. CO
CO CO
o
r*^
10 O en
U3 LO
CM
CM CM
O
o
r-. vo «3-
CM i
. CO
CO CO
o
CO
r- LO oo
VO ^J
CM
CM CM
O
CO
T O LO
CM r
CO
CO CO
CO
co o "=r
CM -M
CO *
o
CM
i en CM
O LO
. co
CO CM
CJ
c: en
rO CT
O ro
-G "^ Q"
4_l
^
O 41
^ c
CJ '"
1 1
ao en
H rd
=
0
i 1 rH OX
ro CN .
. CO
m co
O
CM
o en co
, co
. CO
CO OJ
o
co
CO U3 CM
'^O ^" <
ro co
O
r~.
cn i m
«5J- CO
. CM
CM CM
O
-U
co
LO O «3-
CM t~-
. CO
ro co
O
US
vo UD en
'^O ^3*
CM
CM CM
O
O
cn o co
i /
. CO
.-o co
0
+ CM
CC CM CO |
'411 VJ
. . CM
CM CM
O
CM
. CO
ro co
CO
CNJ o cr
CM -^
ro *
O
r- co ^
cn LO .
CO
(XI CM
CD
w^ O"i
ra C
CD ro
S ££
C
O
H
4-1
5-5 CM O
C C 3
O> CO T3
CU
ro
CL>
OJ
C
o
1/1
re
OJ
ro
C
O
CM
O
o
*->
ro
S-
O
l/l
CD
o
£
CD
~i_)
LTi
03
O
C
O
s.
C
o:
-------
o
i
r
W
C
o
OJ I
o: Ln
cc
ro C1
O -
l/l O
O T-
P- $-
r- C
< CJ
r- (_)
0
>
"- CJ
i- S_
ro 4J
UJ O
CJ
C
S-
Qj
"O <
QJ
C t-
s o
co ~o
i c
^- ^
o
+->
CO
QJ
.X
*
*
, )
zl
p
-*->
uO C
GJ
:s o
,
-J "*~^
«^- C
).
'i
. 4->
CO C
o
ro
£_
(->
CJ
LJ O
t '
*->
<
1
to
^3
i
CD
(
^
C
o
4-J
ro
^y
0)
-U 0=
Mod era
Growt
o
^
r^- oo
CC CM
CM
c
CM'
--\ cn ^r
,-~. CM
ro
O
s--- CM oo
CM i
CM
0
t->
OO CO
LO
0
*
^5 CO' 00
t CM
CO
o
co
s-a CO Ln
VJD OO
"^
o
-1
^~
B< Ln =3-
o
-e< CTi Ln
OO OO
"^
o
cxj
t-= o «a-
CM oo
oo
I ^
s-s Ln
to o Ln
O
0^ ^3"
co Ln ^3"
ro oo
QJ
C CD
ro C
QJ ro
'2L C£.
Baseline
o
s-"- Ln oo
CO CM
CM
O
OO
fr"^ O ^3-
to oo
oo
O
o
>-* «s- oo
OO i
CM
o
r
S ^ ^
UO
o
t-
t-a co oo
O CM
oo
O
,
B--S CM tO
O ^J"
Ln
0
co oo
en
o
CO
c-5 O L.O
^r "sd"
^
O
Ln
c"-5 CTl OO
. CM
00
O
Ln
s^ cjn Ln
CO OO
*
o
^ ^
cr-S tO «3"
CT^ OO
OO
QJ
c cn
ro c
QJ ro
SL &L
80%
S02
Reduction
o
^A CC OO
CO Csj
1 ^
~c
OO
l^. CM
00
ro
CM
O
O
en
o
^
K-: to oo
O CM
oo
0
4_J
CN)
c-s f to
O *3"
en
o
4~>
a-? to *
CO 00
00
o
4J
CTl
&5 O Ln
Ln ^3~
^
o
CO
6>s cn oo
i CM
oo
o
&*S CNJ
I
^
^-^ 1
*4 i
^f <^r
I OJ
C CT)
I ro C
QJ ro
High Grow
Baseline
r--.
ro '
L.O OO
OO OO
in
uo
LO cn
^ 00
O
cn
6-^ 1
OO OO
00 CM
CM
CM CO
tO <3-
00
^ t
r~*, 1**^
^ 00
CM
r
r^ r
o>
S-S 1
en oo
«=r -53-
to
c^^ 1
*3-
cn r*^
oo oo
to
to
b oo
LT>
5-S 1
«3- «3"
OJ
C cn
re c
QJ ro
2: o:
90%
S02
Reduction
00
rj
C
C
O
0)
C
a)
H
0)
U
M
O
CO
7-43
-------
cs
1
w
rJ
03
^
prf
H
C/l
uO
ON
H r-H
o co
H 52
erf o
H M
O 00
W M
rJ >
fjj fv;
erf
w
w in
£H pLi
C/l
W
H H
O H
& ^
S erf
M pd
Cd i -1
O
-------
calculated by Teknekron. Table 7-21 shows for the nation as a whole
and for the 10 regions individually the additional mills per
kilowatt-hour which will be paid for electricity in 1995 under the
selected scenarios for both high and medium growth rates. The table
also shows the percentage increase that each increment represents
over the baseline case (no revision in current S02 control stand-
ard,). Tables 7-22 and 7-23 show (for moderate and high growth,
respectively) the average additional cost per capita nationally and
regionally of the selected alternative revisions of NSPS (based on
results for investor-owned utilities).
As can be seen from the tables, most price increments represent
small increases on a percentage basis, particularly under the assump-
tion of moderate growth. The price increases projected for the West
South Central region (Table 7-21) stand out as relatively very large,
more than twice the percentage increase for any other region. Dif-
ferences among the various scenarios for the moderate growth (shown
on a per capita basis in Table 7-22) are slight in most cases. They
amount per capita to about $1 per month maximum on a typical utility
fill in the West South Central Region. In the mountain regions, with
plentiful supplies of low sulfur coal that can be burned with minimal
controls to meet a standard of 220 ng/J (0.5 Ib S02/106 Btu), the
revision based on such a level without regard to reduction would save
some money on electric power compared with 80 and 90 percent reduc-
tion scenarios.
7-45
-------
CO
w
H
H
H
j )
H
£3
Q
W
O
1
Pi
O
H
CO
[V] s-^
> CO
*f£t r^i
H m H
ft) pi o> W
rJ r-l EH
PQ CO &
O CM
**O CM vD
^O VJ"
2
G
H
G
O
UJ -r-t
T) 4-,'
cd i )
G a
cd c
4-1 &! '
CO CO
>i !-l
4-1 * O cd
C rH
0) T3 rH
w cu o
r-l -H
a, > m oci
*r~( r*^* f^*"
M J *"£j &\ O"*
O rH r-l
CO
C 0) C '"
O 3 -H "
H 5 q
4_» a] T3 c
Cd > CU r-l
3 CU W
G r-l CO ">
H CU C
4-1 rH >-l O
G cd p^ M
o 4-J R j
1 0) rH 2
o <: c
cd -K + V.
7-46
-------
CO
CM
I
W
r4
cd
CO
w
M
rJ
H
H
O
W
o
I
o
H
CO
w
o
fe
CO
PH
CO
E5
W
H
CO
o
U
m
o
rH
o
D
PH
H
£2
O
O
o
w
C
o
O 4J
CO O
3
6-S t3
O 0)
0", *
Q>
H cd
rH W
0) 4-1
M W
cd O
ca u
G
O
H
W)
0)
PH
w
cd
r-l
rH
O
13
^~^
03
cd
rH
rH
o
13
^"^
in CM rH
rH CO CM
["">. *^Q \£)
m CM c^
OO co o
^O 'O vO
o
T) -rl
G 4-i
ctf G
rH Cd
M rH
G G 4J
o w
CN
CS|
f^
rH
G
H
cd
4-1
§
O
S
CO
CO
00
l>^
r^
VO
O
rl
<4-t
rl
CJ
Cd
w
PH
m
O'\
t
r*n
G
H
C
O
H
4-1
cS
rH
Oi
O
CX
,0
tJ
iH
13
to
0)
3
G
(U
(U
j-i
rH
cd
U
0
4-1
CO
cd
13
0)
G
H
IH
0)
n
cd
CM
O
CO
^
CM
rH
^ x
>-3
t>0
G
o
CM
in
I
CO
13
J-|
cfl
13
G
cd
4-)
W
4J
G
(1)
}-l
a
MH
O
G
O
H
4J
cd
3
G
H
4-1
G
O
U
.0
,
OO
r-
CA
rH
O
G
M
G
0
M
*v4
cy
M
Q)
H
CJ
O
3
0
CO
7-47
-------
Percentage per capita price increases are higher for the high
growth everywhere (Table 7-23). The percentage differences in the
South Atlantic and the Pacific regions are less than 2 percent in
all scenarios.
It should be noted that the per capita figures in Tables 7-22
and 7-23 do not distinguish classes of users (e.g., commercial
versus residential) or individual establishments served. The dollar
values merely represent total cost (nationwide or within each region)
divided by the appropriate population estimate for 1995.,
The effects of increasing the average price of electricity to
the consumer by the amounts projected will be relatively small in
terms of the cost of living. A measure of the effect o £ price of
electricity is provided by the weighting attached to this item of
expenditure in the Consumer Price Index (CPI) computed periodically
by the U.S. Department of Labor, Bureau of Labor Statistics.
The relative importance of a component of the Consumer Price
Index is its expenditure or value weight expressed as a per-
centage of all items. At: the time of their introduction,
after a major weight revision, the value weights for groups
of commodities represent average annual expenditures of urban
wage-earner and clerical worker consumers and relative impor-
tances indicate how these consumers actually allocate their
expenditures to the various groups (U.S. Department of Labor,
1976).
As of December 1976 (the latest date for which CPT information
was available at the time of writing) a weight of 1.367 was assigned
to the price of electricity in the CPI on a nationwide basis. This
figure reflects the fact that on the average national expenditures
7-48
-------
for electricity represented 1.367 percent of all consumer expendi-
tures. The CPI is computed by multiplying for each item included the
price change (essentially as a percentage basis) of that item by the
percentage weight, then summing all of the resultant products and
dividing by the base data aggregate. Comparison of the CPI thus
obtained for a given month with that computed for a previous time
period enables an estimate to be made of increase or decrease in this
specific measure of the average cost of living. Each increase of 100
percent in the nrice of electricity would raise the CPI by 1.367 per-
centage points. An increase of 2 percent in the price of electricity
would be reflected as an increase of about 0.027 percentage points in
the CPI.
The relative importance of the cost of electricity to the
average consumer may be indicated by comparing the weight of 1.367
with weights of 23.667 for food, 9.194 for apparel and upkeep, 3.206
for gasoline in private transportation and 19.013 for health and
recreation (among selected categories of items) (U.S. Department of
Labor, 1976).
The CPI is also computed for regions of the country, such as
individual Standard Metropolitan Statistical Areas (SMSA). Results
for selected SMSAs indicate that price increase in electricity would
have somewhat greater impact (percentage-wise) on some localities
such as SMSA in Ohio and in the areas of Pittsburgh, Buffalo, Atlanta
and Detroit than on the national average. Conversely, in Boston,
7-49
-------
Seattle, Chicago and Baltimore the percentage impact would be rela-
tively less.
It may also be noted that the cost of electricity (along with
that of other items reflecting energy sources) has risen recently.
The weight of 1.367 assigned in December 1976 represents an increase
of 1.46 percent over the weight of 1.347 assigned in December 1975.
In view of the factors that tend to increase energy costs at a rate
exceeding price rises in other commodities, it may be conjectured
that the weight assigned to electricity in future years will be still
higher. However, there is no valid basis for estimating the weight
in any future year. Fven conjecturing that the relative escpenditures
for cost of electricity increased annually by 2 percent (the percentage
increase noted between 1975 and 1976) would leave the weight assigned
for electricity in the CPT at less than 2 by 1995.
7«4.4 Secondary Economic Impacts
7.4.4.1 Economic Impacts on the Coal Industry. Secondary
effects of the revised NSPS on the coal industry will be chiefly
a slight reduction (over the baseline situation) in the total amount
of coal to be produced and a shift in the pattern of regional produc-
*
tion. These effects are shown quantitatively in Table 7-24 for the
year ]990.
In all situations the total amount of coal produced in the future
will be much greater than current production.
7-50
-------
TABLE 7-24
1990 COAL PRODUCTION UNDER ALTERNATIVE
NEW SOURCE PERFORMANCE STANDARDS
(ELECTRICITY GROWTH RATE OF 5.8 PERCENT PER YEAR
UNTIL 1985 AND 5.5 PERCENT THEREAFTER)
(in 106 tons)
Region *
1.2 Ib S02/10
Btu
INCREMENT IN TOTAL
f £ t
90% Removal ! 10 Tons 1 Percent
Northern Appalachia
Low Sulfur
Medium Sulfur
High Sulfur
Total
Central and Southern Appalachia
Low Sulfur
Medium Sulfur
High Sulfur
Total
Midx^est and Central West
Low Sulfur
Medium Sulfur
High Sulfur
Total
Northern Great Plains
Low Sulfur
Medium Sulfur
High Sulfur
Total
Rest of West
Low Sulfur
Medium Sulfur
High Sulfur
Total
National
Low Sulfur
Medium Sulfur
High Sulfur
Total
28.0
111.2
65.5
204.8
198.4
35.0
3.1
236.5
2.6
97.9
197.5
298.0
547.3
262.0
0.3
809.6
94.4
124.5
-
219.0
870.6
630.8
266.5
1,767.9
20.2
128.0
109.5
257.6
176.4
32.3
3.1
211.8
1.6
99.7
271.1
372.4
321.1
329.9
0.3
651.4
79.0
138.6
-
217.7
598.3
728.6
384.0
1,710.8
)
t
+52.8
+25.8
1
-51.7 -21.9
+74.4
-158.2
-1.3
-57.1
+25.0
-19.5
-0.6
-3.2
* These regions differ from those discussed under costs and prices of electricity in
the preceding subsections. The Mountain Region, as used for census purposes, combines
the Northern and Southern Mountain Regions used in Teknekron. Low sulfur includes coal
that has less than 0.6 Ib sulfur per million Btu or is of metallurgical quality.
Low sulfur coal can meet the current NSPS of 1.2 Ib SO without scrubbers. Medium
sulfur coal has between 0.61 and 1.67 Ib of sulfur per million Btu or roughly 0.7
to 2 percent sulfur coal. High sulfur coal has above 1.67 Ib sulfur or is roughly
more than 2 percent sulfur.
Source: ICF, Inc., 1978a.
7-51
-------
More stringent control of SO with increased use of FGD systems
extracts a penalty in the effective energy output derived from the
coal and, hence, more fuel is required to achieve a given level of
power production. Surprisingly, for 1990, the 90 percent reduction
of SO * is predicted to lead to a decrease in the total amount of
coal required, from 1,767.9 million tons under current standards to
1,710.8 million in 1990 under the proposed revision. This reduction
in tonnage results from two causes. First, imposition of FGD systems
under the 90 percent reduction scenario would make it economical to
burn more eastern coal that has a higher average btu content than
most western coal, consumption of which is accordingly reduced.
Average btu content of all coals consumed for power generation,
therefore, would increase, requiring a slightly smaller total amount
of coal. Secondly, more utility oil and gas is expected to be con-
sumed as higher costs of coal-fired power plants make it economically
competitive to increase the use of oil and gas in existing oil and
gas steam plants (ICF, Inc., 1978).
*
Projections under the scenarios of 80 percent S02 reduction and
a maximum emission rate of 0.5 lb S0~/10° Btu showed coal produc-
tion in 1990 differing by less than 1 percent from estimates for
90 percent SO* reduction in any of the regions listed in Table
7-24 (ICF, Inc., 1978).
7-52
-------
Table 7-24 gives coal production in millions of tons by sulfur
** _ . . ***
content within each region. Noteworthy changes under 90 percent
SO control are production increases in high sulfur coal in Northern
Appalachia, the Midwest and Central West. The requirement for
scrubbers in all new plants would enable higher sulfur coal to be
more economically competitive with the transport and use of low
sulfur coal. Medium sulfur coal production would also increase,
whereas mining of low sulfur coal would decline everywhere. The
effects on overall coal production are far more significant region-
ally than nationally (where the total decrease is only 3.2 percent).
As most western coal is low sulfur, the decline there would be over
15 percent or about 160 million tons in 1990. Also, the higher
sulfur coals of the Midwest, Central West and Northern Appalachia
would be produced in greater quantity; whereas a drop particularly in
low sulfur coal production in Central and Southern Appalachia would
result in 1990 in an overall decrease of nearly 22 percent (over 51
million tons).
Low sulfur includes coal that has less than 0.6 Ib sulfur per
million btu or is of metallurgical quality. Low sulfur coal can
meet the current NSPS of 520 ng/J (1.2 Ib S02) without scrubbers.
Medium sulfur coal has between 0.61 and 1.67 Ib of sulfur per
million Ptu or roughly 0.7 to 2 percent sulfur coal. High sulfur
coal has above 1.67 Ib sulfur or is roughly more than 2 percent
sulfur. (IGF, Tnc.,1978a).
***
These regions differ from those discussed under costs and prices
of electricity in the preceding sections. The Mountain Region,
as used for census purposes, combines the Northern and Southern
Mountain Regions used in Teknekron calculations.
7-53
-------
7.4.4.2 Economic Impacts on Coal Transportation. The proposed
revisions of NSPS will considerably alter the pattern of coal trans-
portation. In particular, the amount of western coal shipped to
the fast is projected to be about 455 million tons in 1990 under
continuation of the present 520 ng/J (1.2 Ib SO /10 Btu) and
slightly under 300 million tons under the requirement for 90 percent
*
SO removal. There will also be a net decrease in the 90 percent
reduction scenario of 226 billion in total ton-miles of coal ship-
ments (i.e., tonnage moved between any two points multiplied by the
distance as compared with continuation of present standards). The
principal cause of this relative decrease in shipping will be the
reduced shipment of western coal to the East. As a result, the
average shipment will be 687 miles under 90 percent SO removal,
as contrasted with 791 miles if present standards are continued
(TTF, Inc., ]978a).
These results are shown quantitatively in Tables 7-25 through
7-?8. Table 7-25 gives coal distribution from each of the five
jf${
production areas or coal supply regions to the nine census regions
for consumption as projected for 1990 under the baseline situation
*
Variations in coal shipment from West to Fast as projected for
scenarios of 80 percent SO- reduction and of a maximum emission
level of 220 ng/J (0.5 Ib S02 Btu) differ from those shown in Table
7-28 for the 90 percent reduction scenario by less than 1 percent in
1990.
**
These regions differ from those discussed under costs and prices
of electricity in the preceding sections. The Mountain Region,
as used for census purposes, combines the Northern and Southern
fountain Regions used in Teknekron calculations.
7-54
-------
W
*-£
K
25 n:
§ B
g§ §
U < U
Q
W 2 >H
E-t H H
CO U
OS H
H w K; ^~.
Q U EH en
s z u C
D g w o
^ J P
2; K u
in o O ^>
CN M u, x o
1 H K 15 H
r~ D W M
,, PQ f^ a c
W M - -H
sS a; w
« H U 04
f"1 M D in
a o
in fn
J O
< S
O W
u 2 in
CQ
O >J
CTv Ol
, j *
rH
0
in
C
o
H
CP
(1)
P""1
rH
a
a
P
in
rH
rd
0
u
i (
fd
4J
O
^
O
P
W
0)
C
S-l
CD P
P 0)
O O
23
rH
4J
o
H
rfj
C
nj rH
rd
4-1 1H
in 4-1
Q) C
3 OJ
o o
H
T!
c
rd C
T-t 0)
fd ^
4--) P
C O
a) co
u
£
M
0)
r<~;
4->
^
O
2
T
s
.3
4J
cd
P
in
Q)
*^
P
01
0)
in
c
H
rd
a!
in
rd
W
P
01
Ci)
2
rd
r-i
U
rH
rd
a
a
«:
rd
H
U
rd
, )
to
%
U)
C
O
H
tT1
CD
«
rj7>
r^
H
g
in
£
O
u
^r ro r- (» co
CM C^ rH m CO
rH CO «^ rH
oj CM r~ m r^
o CM r- ^ o
^0 0 CO O
rH rH rH
i 1 O"N
1 1 1
i i tec; P i-i P
CP o P in in
0) -H 0 rd rd
2 S cn W W
m
VD
rH
rH
CO
£
*3"
o
m
rH
ro
rH
5
CO
IV,
o
r-
ir
r"*»
v£>
CM
^>
m
co
CS)
CO
"3*
0
CM
P
01
rd
W
rH
rd
0
H
in ^ i-
(n rH ^
r-
to r~-
rH
tO O
O ID
r^ oj
in O
rH
ro <3*
T-H m
O rH
(M OJ
CSJ \.Q
en o
VO rH
ro co
^0 °°
rH ^
CO M
r~-
*^" O
0 r-
CO CTi
CM
CN ^>
rH VD
no
CM
ro
1 "^
O
CM
P
01
01
£5
rH
rd #
4J rH
O «J
EH C!
O
4-1
ccj
!a
pi
o
rH
4-1
C«
S_i
iw
i
cd
H
(U
TJ
\X)
:"~*
r-i
*-
O
H
*
U
M
*
QJ
CJ
iM
3
O
Crt
7-55
-------
CM
I
i
^§3
§53 §
= 15 -
g£~£
W 3 tJ
M 5 <
a w> s*
j
« « CD
w d w
3 y 3
3 CJ
H *J -H
u 3 fl
e o &
O lA p
1
r a t
en
t
r
1/1 In
>N
s
in o( ^*
f-i iH
? 7
I ^
1 -3
00
r^
CTv
7-56
-------
(continuation of current standards). Table 7-26 shows the same
interarea distribution as projected for 1990 under the assumption
of 90 percent SO removal, together with the increase or decrease
in coal shipped from each supply region to each consuming region as
contrasted with the baseline situation. Ton-mile shipments between
Fast and West are contrasted under the two scenarios in Tables 7-27
and 7-28.
As seen in Table 7-27, the West is projected in the baseline
situation to ship coal it has produced a total of nearly 700 billion
ton-miles in the East and over &00 billion ton-miles to points for
consumption in the West. Conversely, the East is estimated (under
continuation of present NSPS) to ship its coal product for use in
the West, a total of only 11 billion ton-miles; whereas coal both
produced and burned in generating electric power in the East will
move a total of 266 billion ton-miles. Coal consumed in the East will
(wherever it is produced) represent a movement of nearly 1 trillion
ton-miles (as shown in the last column of Table 7-27). The changes
reflected under the 90 percent SO reduction as shown in Table 7-28
are almost entirelv due to the decreased shipment of western coal.
Coal burned in producing electric power in the East will represent a
drop in shipment of over a quarter of a trillion ton-miles. Changes
from the baseline scenario in the shipment of coal for generation of
electric power in the West will be negligible. Total movement of
7-57
-------
TABLE 7-27
1990 TON-MILES OF COAL SHIPMENTS UNDER THE CURRENT
NEW SOURCE PERFORMANCE STANDARD OF
1.2 LBS. OF S02 (HIGH ELECTRICITY GROWTH RATE)
(in 10y ton-miles)
Producing Regions
Consuming Regions
East
West
East
266
11
West
699
420
National
965
431
National
277 1,119 1,396
Average ton moves 791 miles.
Source: ICF, 1978a.
TABLE 7-28
1990 TON-MILES OF COAL SHIPMENTS UNDER AN ALTERNATIVE
NEW SOURCE PERFORMANCE STANDARD OF
90 PERCENT REMOVAL OF S02
(HIGH ELECTRICITY GROWTH RATE)
(in 10^ ton-miles)
Producing Regions
Consuming Regions
East
West
East
300
12
West
443
415
National
743
427
National
312 858 1,170
Average ton moves 791 miles.
Source: ICF, 1978a.
7-58
-------
coal in ton-miles to the West from anywhere in the nation is projected
to drop by less than 1 percent.
7.4.4.3 Effect on Coal Mining Employment
Total production of coal for use in generation of electricity
will of course increase substantially by 1990, but the increase under
90 percent SO reduction will be slightly less than that projected
under maintenance of current NSPS. Thus, the effect is a net decrease
under the revised NSPS when compared to the baseline. However, the
requirement for miners is projected to increase in all situations.
Under 90 percent reduction a rise in employment (over the baseline
situation) would result because much more eastern coal would be mined
and less western coal. The mines of the west can be operated at a
higher rate of production per miner. Table 7-29 contrasts 1990 pro-
jected employment requirements for the five coal production regions
under the baseline situation (no change from current standards of 520
ng/j) and the proposed NSPS for 90 percent removal.
Major employment increases in Northern Appalachia the Midwest,
and the Central West would not be fully offset by decreases in Central
and Southern Appalachia and in the Great Plains. The rest of the West
is relatively unaffected. The net increase projected to occur over
actual employment figures for 1975 is also shown in Table 7-29. As
can be seen, the work force needed to produce coal in Central and
Southern Appalachia for generating electric power is expected to
remain nearly stable (about 1 percent increase in the 90 percent
7-59
-------
TABLE 7-29
COAL INDUSTRY EMPLOYMENT* (IN THOUSANDS OF EMPLOYEES)
ELECTRICITY GROWTH RATE OF 5.8 PERCENT PER YEAR UNTIL 1985 AND 5..5 PERCENT THEREAFTER
1990
Net Changes
Baseline
Region
Northern Appalachia
Central and
Southern Appalachia
Midwest and Central
West
Northern Great
Plains
Rest of West
National
1975 Actual
59.1
91.6
29.7
2.8
6.7
1.2 Ibs. SO + 90% Removal
189.9
67.3
104.0
64.9
26.4
23.1
285.7
88.9
93.2
82.5
20.3
22.7
Increment Percent
307.6
+21.6
-10.8
+17.6
-6.1
-0.4
+21.9
29.8
1.6
52.8
17.5
16.0
117.7
Miners and mine supervisors, exclusive of employees at mining company headquarters and of
workers in coal preparation plants.
+ - Baseline
Source: ICF, Inc., 1978a.
7-60
-------
reduction scenario over 1975); whereas huge increases are projected
for all other production regions. The net gain in employment will
greatly exceed the 1975 work force in all regions except Appalachia.
Employment is projected to increase in the Great Plains by a factor
of 8.42 in the baseline situation and of 6.25 in the 90 percent
reduction scenario.
The further impacts generated by this overall increase in em-
ployment and by the changes in distribution of the labor force are
difficult to predict and certainly cannot be accurately quantified.
In classical economics, jobs are not created but filled at the
expense of jobs elsewhere. But many examples can be cited to
indicate such a theory represents a drastic oversimplification, and
that its resulting assumptions, particularly about the mobility of
the labor force, may be at variance with prevailing conditions.
However, it is true that coal mining is a specialized occupation
requiring skills that may not be possessed by those unemployed.
Some indication of the immediate follow-on effects of this shift
in employment may be seen from the changes in the total payrolls to
coal miners as given in Table 7-30. Coal mine payrolls will be
greater in the Northeast and Midwest by the amounts shown than
they would otherwise have been, whereas coal mine payrolls will be
reduced in the West and in Central and Southern Appalachia. These
potential losses may or may not be compensated by other sources of
income. Changes in the induced effects resulting from employment
7-61
-------
o
m
I
H H
W W
PS u
w a:
in 3
O O
c*
la z
u « H
Z ta O
i-i a
o R
iJ z; w
«c M as
Z H
O iJ CM
l-i 3 C
U u to
H U
« of
a o
W en
H cr.
M IS
H M
LC
2
O
M
a
w
pi
JJ
CO
QJ
3
c
J-l
CO
OJ
K:
to
c:
H
cfl
D-,
4-1
CO
0)
i-l
c
4J
CO
OJ
-a a
d
CO rH
cfi
13 l-j
H JJ
E d
a>
o
w
H
JZ
o
13 CO
C rH
CO Cfl
a.
rH CL,
Cfl H
l-l CO
o a.
E (X
*H
E
CO
l_i
O
JJ
o
CO
u.
cn
i- C
CO 0) O
rH 3 -H
rH rH rH
O CO rH
Q > -H
x-^
W
J-l
o
JJ
u
cc
pu
CO
rJ d
CO QJ O
H 3 -H
rH rH rH
O Cfl rH
a > *H
£
^-x
w
n
o
JJ
y
£
CO
i-i d
CO QJ O
rH 3 -H
rH rH iH
O Cfl rH
Q > lH
E
\_^
CO
t-l
o
JJ
y
CO
U.
CO
i- d
CO 0) O
H 3 -H
rH rH M
O CO rH
D > -H
E
w
i-i
O
JJ
y
n)
ESH
CM
CM
o co LD
1 CC 1 CO
rH
O rH
O 1 1 1
--> L? rH rH 13
O rH PL. C
rH r 1 O T-1 rH
Cu Cfl W JJ
E 3 ^ rH t3
w d to 3 c
C P-t £ co
0)
CO 01 O Ki
rH W 3 -H i-l
(C 0> rH M T3
O > CO QJ d
O < > Cd M
rH CM en
rH
1
(
1
r-~
0
r-.
CM
+
1
v£>
O
O
^5
rH
+
d
o
*o
0)
CO
cfl
CQ
QJ
3
rH
CO
>
I-i
CO
rH
rH
o
o
y
cd JJ
JJ W
o o
H rJ
\D
o
ON
cn
rH
1
in
r*-
O
rH
T3
O
H
VJ
QJ
P.
M
CO
OJ
{*
I
LO
rH
QJ
^
JJ
P^
O
14-4
00
O
o
v^
J-J
cfl
i-J
cfl
4J
cO
O
4J
s
4J
E rH CM
H
4J C C
co E E
o 3 3
rH rH
QJ O O
p U O
CO
oj en
CO
OJ OJ 01
GC d C
rfl -H -H
IS rH rH
X f<
CO CM rH LTi
1 i i s
rH 3 3 'H
O rH r-i rH
tJ O O
0 U +
l/~l
r-. rH
O>
rH
< 0
u
4J
d
OJ c/;
E
CL
O X>
rH
1)
>
0) r-v
C ^O
OA
U rH
H
E
0 JSl
d o
o o
U CO
H
ca
U-l JJ
O cfl
Q
CO
d >-
O JJ
-H d
4-> 3
o o
(y u
O -0
M C
Pi TO
>i
J2 JJ
O -H
k U
cfl
QJ
W CO
QJ 3
Pi LO
d
u ,
W k( U-i
CO O
T3 a
d c 3
ct) ^J Cfl
r-3 QJ
CO >-J
CO * 3
o; nj cfl
c c
H CO *
CO W QJ
3 c a
PC O M
S OJ
*O E
CO O
3 rH O
Cfl 3
QJ C WH
rJ CO O
3 OS
P3 -H J-J
E C
» QJ
CO E
d 4J
CO CO M
4-» d Cfl
d to a .
O -u OJ U
E C Q
o
u-i £ K Q
O 01
C J-J -
^ -H ffl C
JJ JJ O
-H >, c^: JJ
CO JJ b£l
tJ -H T3 C
QJ > 0> -H
> -H JJ .C
H AJ -H LO
CO C CO
^ ^
en
w
o
Pi
P
ac
d
*H
j_j
c
H
}_i
a,
j-j
^ d
r--
M O
QJ CJ)
iH
rH
ex cn
-H
4J
rH S
3
0
i xC
cfl r^
iH a-
j_- t-i
jj
w jj
3 CO
"O 3
C bC
-H 3
<
U-j
o
CO
jj be
3 C
0 -H
JJ C
d rJ
H CD
VJ U
CL
o
VJ C
QJ trj
JJ
3 Jj
tc
OJ
o E
u >^
0
rH
a
co E
^ W
CO
r*l *
H K
tfl U
C "H
< J-J
K
U -H
-H 4-J
E tfl
O JJ
C W
o
0 l-(
u o
J2
AW TO
0 rJ
3 tw
Cfl O
OJ
LJ 3
3 CC
aq a-
p
01 C£
y
u
-------
shifts appear in Table 7-30. Induced effects result from the fact
that money provided to workers (and similarly for other production
costs) creates a demand for additional goods and services. For
example, housing must be provided; retail sales increase in the area
when coal mines expand; new stores, restaurants, and other businesses
move in; and service activities produce induced income.
To the extent that a revised NSPS of 90 percent SO removal
would increase mining employment in regions with traditionally high
production of coal and decrease it in the West, the socioeconomic
consequences may be less than would result by maintaining the current
standards. In 1975, mining employment in the West represented only
about 5 percent of the national total. This small fraction of the
total western labor force is projected to increase more than five-
fold by 1990 under the baseline scenarios. Much of the western coal
lies in localities where very little mining has occurred in the
recent past and is not part of the way of life. Communities there
lack infrastructures for dealing with large influxes of miners and
with the secondary growth associated that can create boom towns in
farming and ranching communities.
Mine employment in the West is projected to grow in any event
by over 30,000 workers in 1990. Therefore, it is far from clear
what differences in sociologic conditions would occur if this labor
force does not increase by a further 6500 miners. It can be noted,
however, that such regions find their problems of adjusting to
7-63
-------
industrial growth increasingly complicated as the size of the outside
work force grows. The rate of financial growth in the West may (as
implied by Table 7-30) be somewhat slowed under 90 percent SO
reduction; however, the resulting socioeconomic problems are likely
to be diminished. Eastern coal regions would of course face increased
socioeconomic problems as the labor force grows more under the 90
percent reduction scenario than in the baseline situation. However,
the increase, even though sizable, will represent a much smaller
percent of the 1975 base. The eastern regions can also be expected
to have an advantage in dealing with the growth because coal mining
in these locations has long been a way of life. However, many other
economic and sociologic variables will strongly influence the extent
of impacts experienced.
7.4.4.4 Plant Construction. Increased installation of FGD
svstems under the revised NSPS will have a significant economic
impact through the direct and indirect effects resulting from the
additional employment.
The average numbers of man-years of construction forces required
to build power plants under existing and proposed revisions to NSPS
have been calculated by PEDCo (1977c). The work force required varies
somewhat with the size of the plant; economies of scale are possible
so that as the rated capacity of the plant increases, the ratio of
man-years to MWe decreases slightly. For the 600 MWe-plant (the size
modeled in the Teknekron simulations discussed in Section 7.4), the
7-64
-------
work force requirement is slightly over 1.1 man-year per MWe. Specif-
ically, the estimates (expressed in gigawatts) are as follows (PEDCo,
1977).
Scenario* Man-Years per GW
Current NSPS (Baseline) 1150
90 Percent SO Reduction 1170
220 ng/J (0.5 Ib S02/1Q6 Btu) 1186
Using these figures, the incremental man-years required under
90 percent SO reduction and under a ceiling of 220 ng/J (0.5 Ib SO
/10 Btu) (compared with baseline) were calculated together with the
resulting additional payrolls and secondary income. The gigawatt
capacities projected as requiring FGD at 5-year intervals under
baseline and under the alternative scenarios were taken from Tek-
nekron calculations as shown in Table 7-31.
Results of calculating additional payroll and secondary income
under revised NSPS are shown in Table 7-32 for the moderate growth
rate and in Table 7-33 for the high growth rate. It was assumed
that the 1980 capacity with FGD was the same in all scenarios, as
capacity affected by revised NSPS would probably not be operational
until after that date. Results for the moderate growth scenarios are
straightforward. These figures show the additional income generated
under the 220 ng/J (0.5 Ib SO /10 Btu) scenario to be on the order
*
Calculations for 80 percent reduction were not reported,
7-65
-------
TABLE 7-31
REQUIRED CAPACITY OF ELECTRIC POWER WITH FGD SYSTEMS
GW CAPACITY WITH FGD~
Scenario 1985 1990 1995
Moderate-Growth Baseline
Maintenance of Present NSPS 52.5 61.3 67.1
90% S02 Reduction 74.2 145 207
Increment Over Baseline 21.7 83.7 139.9
0.5 Ib S02/106 Btu 71.3 132 188
Increment Over Baseline 18.8 70.7 120.9
High-Growth Baseline
Maintenance of Present NSPS 45.6 59.4 76.5
90% Reduction 65.0 216 403
Increment Over Baseline 19.4 156.6 326.5
Source: Teknekron, Inc., 1978.
7-66
-------
of 1 3/4 times as great as that under 90 percent reduction. Total
additional income generated (over baseline) will exceed $100 million
during the period 1990-1995.
The high growth situation introduces a slight complication as
may be seen in Table 7-33. A study of firms providing FGD systems
showed that at present industry would be capable of installing all
of the capacity required under any of the scenarios except that of
high growth with 90 percent SO reduction. For this situation sub-
stantial staff increases would be required. It was, therefore,
assumed that employment in the FGD industry itself (not included in
the PEDCo study of construction forces required for power plants)
would be essentially the same for all scenarios under moderate growth,
Significant additional employment was assumed for the scenario of
high growth with 90 percent SO reduction, according to Tables 7-34
and 7-35. Table 7-34 indicates significant employment increases
that would be necessary within the industry to meet FGD requirements.
These amount to slightly over 12 man-years per additional MWe capa-
city, derived from estimates obtained in an industry survey (IGC,
Inc., 1977). The distribution of additional manpower by type of
employee is shown in Table 7-35. These basic figures have been
incorporated into the employment estimates for the 90 percent reduc-
tion scenario under high growth in Table 7-33. Average annual pay
for craftsmen is assumed to be the same as that for the construction
worker. Medium income for the other employee types was calculated in
7-67
-------
I
H
jS « pa
o & w
j S 3
fl4 f- O
p o
Ed o
cn u w
w fe <
!Z t-l X
tH S E-
sss
PS rH CS
tfc H
O 3 H
25 OS
H M-l CO
rH O fH
rH rH
^H O
Z (2
>T C"
CM co m ^
o o cr.
m CM rH CM
co m
r-- o
-H 00
CO
o
4-1
CJ
CO
CO CJ
C CO
O l-l
i-H 14-1 CO
rH 0 rH
rH rH
£ &
en - *
u-, ,_, rH CM
CO rH
o "n
m ON
cn oo
CO
o
4J
U
[*
CO CJ
c cn
0 >*
1-1 14J CO
rH O rH
rH «H
£ &
^ 0> rH
. c^ r-*
rH O ^
in CS rH CN
^ 3
0^1 cn
r-H m
0)
0
o
A
fa
O CT- rH
>* r»»
CM CM in .
\D rH rH CM
to r-t m
C CO
O (-<
H M-i Cd
rH O rH
rH rH
iH O
s a
rH -J-
-y CM
O CO
rH CM
cn| m
o
o
ca
co a
c cn
O PJ
H *4H CIS
rH O rH
i s
-j- cn
CO OO rH
CM * r-
oo m cn
rH J rH CM
^O rH
co >a-
cr> CM
^ %
CO
o
U
o
c2
cn
^ ^r ^ ?
rH cn cn -
CM -s.fr rH CM
1
C 04
c a o rH
o & r-ta
H CO O W
o « e X 1-1
h 3 a co co ,-1
cu M cn ex a.
> 4-1 >. 3 -H 11
O CO CO O rH 4J 6
C CO CL -C f ^ CO rH O
3 O v 4j to c 3 'j
O CJ fflll-^UOSC
rHCUrH ^ CO 1* rH 4-) rH
COCCd t l-l ITJ rH 1-» CO >,
B *H C c« G) 0) O ^3 'H ^4
H OJ -H G S < <4JTJ
4J CO 4J iJ CO C
H CO -H O 3 O
T3 PQ -3 Cn 1 1 1 -o O
"O T3 C <1J
32 < M C/J
1 in
rH
*3"
rH
m
rH
0
00
0>
m
m
CM
CM
rH
£>
0
CM
m
\o
in
CO
cn
r-.
CM
CM
CM
l-l
QJ
0
£
e
01 0
>-l £1
CO
CO T3
j= e
4J CO
a h
.H O
CO
c -a
O 01
H 4J
4_l CQ
1-4 >-l
o cu
a c
CO CU
CO
a
3 01
cu g
n cj
CO i-l
O rH W
CJ CO 14
fe C CO
O r-|
e w o
V4 TJ
IH *o in
3 co r-~
cr o\
U CU -H
4J O
JJ CO
cj C
a c
3 0) O
O OS O
i i t
crj j2 o
lerce
§
o
^J
14-)
O
C
E
u
CO
a.
cu
a
c/3
iH
01
g
i-i
01
g
8
0
B
S
CO
a
cu
a
3
01
CJ
0
cy}
7-68
-------
cn
cn
i
r-.
W
j
M
H
W &J
^ ^
b o
2^ r^ i
M Q
E- W
2 H
W <
S 05
O K
rJ H
0-, 3
S O
W Pi 2
O O
Q M
W 33 H
cn o O
< M 3
M re o
(-^ Ci3
u co pa
2 S
M Ld CM
EH O
S Crt CO
C >H
M CO B-S
fo C
d ON
cj cj
2 PM O
t? ^
c/: 3
W
(X 2
O
W VH
S H
q o
M |-i
CO
< o
2 C
m
ON
ON
rH
1
o
ON
ON
rH
1
i
O
rH
i
m
CO
ON
rH
,
m
co
ON
rH
1
O
CO
ON
< 1
CO
C m co
C r-v j_i
T-( ON CO
i i rH rH
rH rH
H OH O
SOP
X
^4
0
4-1
CJ
CO
P*u
CJN
ON
NO
rH
CO
C m cn
O r**« t*t
r» O Cfl
r-i rH rH
rH rH
rH 14-1 Q
IS 0 0
cn
^
o
4-J
CJ
to
fc
CN
f-
cn
rH
CO
C m cn
O f-~ M
H ON CO
rH t-t i-H
rH rH
H UH O
IS 0 Q
cnl
M
o
4J
u
CO
fa
sr
ON
rH
0)
CO O
01 )-i
c o
H fo
rH
CU C
03 O
CO -H
pa 4J
a
S-i 3
CU M
Q cn
3 0
O U
CO (0
c e
c o
H -H
4-1 iJ
H -i-l
"u -a
""^1 ~O
< <
ON.
cn
m
sr
sr
CN
CO
00
cn
r^-
rH
r*»
NO
si-
cn
»
o\
r 1
O
en
NO
cn
ON
*
m
rH
CM
m
cn
m
CO
ON
r^
»
cn
rH
U-H
14-1
CO
cn
ON
rH
CO
NO
0
4-*O
CO
rH
rH >,
CJ C >-i
M M 4-1
an-years
2£
1
ay per man-yea
dditional pay-
onal FGD Indus
C-J S
OJ 1
C C rH
M CO rH
H 60
co M
CU M >,
ngineers and D
Man-years
Median pay pe
Additional pa
raftsmen
w i i i u
\o
m
cn
PJ
sr
cn cn
cn *
rH rH
m
(Hi
p-
ON
ON
rH
rH m
\O rH
O
Ut
CC
0>
1
C rH
e "o
Man-years
Median pay pe
Additional pa
dministrative
i ; i <
CM
NO
CN
CM
CO
rH
ON
rH rH
ON
^f
^
CO
f
CM CM
in rH
a
%4
CO
cu
1
C rH
CO rH
E O
S-l ^
Man-years
Median pay pe
Additional pa
i i i
NO rH
cn CM
ON oo
^5 rH
r-i cn
cn
rH
^
.
CM
sr CM
CM NO
CN ON
CO P^
rH
cn
rH
r«».
.
CM
CN ON
in rH
cn in
m sr
rH
cn
rH
.
CN
rH
i 1
oc
Additional Pay
rial Multiplie
ary Income
4-J T3
rH « C
W 3 O
O C 3
r-i W C/3
r^
«n
^,
,
NO
CM
NO
CO
rH
vC
rH
z i *
CO T-l ^
CU
Q! _^
rH ,C °
CO 4U ^
'H pH
C/^ M J>i /^
0-t 4J ^Q
C/3 CO
23 CO
TTJ CU
4-1 C 03 L~
c -H co £;
CU CU
M CU i-i '""'
MX: o
34J C
U -H *
***\
4H O M-j "
O M-l ' '
^ CO
CU r-l 4-< i_T
fj I. f« f1
C CJ g
CO CU 4J ^
CM c M
CU tH CO
4-i T5 O
e -H
H TD "^H rf
CO CU -H H^
CN 1 !-) Cu -H
s § B ?:
rH
CO
ON
rH
SJ
rH
Ctl
CO
ca
INCREMENT OVER
C 03 O ^
= 4-1 C i
T3 CO .C g
CU ^ \ _i f-
M 4-J -H ,9
T-i 3 "
3 CU
cr > -a
cu o cu °
1- -C -H
CC rH ^
CO -0 £ c
-C C 3S
4J to cn £
CM cu «
> ^ 5
r- C C J
to to
C "tv CJ *
c cu ^
rH 4 4J ^
4- CO CO "^
-rH M if1
O CU 4-J
^C C "
CO CU >N ?S
cn -H 5
e cu cj ^
cu E cfl
4-1 C Cu
« u -co ?*.
i>, C x a r
03 -i- l-i ^
w -a £
C r- CU S
C^ CO t i 4_j ^-^ jr
ft. c c co sr R
c '^ 4 * cn ^^
cc « tn i
C 4- ui r^
H -H 1^ CU °
M -C O1 4=
-------
TABLE 7-34
INDUSTRY CAPACITY TO MEET FGD REQUIREMENTS
(High Growth Rate, 90 Percent S02 Reduction)
CUMULATIVE CAPACITY IN GW PROVIDED WITH FGD
YEAR
1982
1985
1987
1990
1992
1995
FGD INDUSTRY WITH3
PRESENT STAFF
51.4
87d
110.85
148d
173.15
210e
PROJECTED REQUIREMENTS15
(c)
64.4
123d
216
(c)
403
SHORTAGE AT
END OF 5-YEAR
PERIOD
-
(c)
-
68
-
183
aTeknekron, 1978.
"Industrial Gas Cleaning Institute, Inc., 1977.
cNot given by source.
dlnterpolated.
eExprapolated.
7-70
-------
the same way from income data in the Statistical Abstract of the
United States.
7.4.4.5 Increased Consumption of Oil and Gas. A revised NSPS
is projected to increase the costs of burning coal for generation of
electricity. Consequently, oil and gas will be economically more
attractive as a fossil fuel than they would otherwise be, i.e., under
maintenance of the current standards of 520 ng/J (1.2 Ib SCV/10 Btu).
TABLE 7-35
DISTRIBUTION OF FGD INDUSTRY EMPLOYMENT
PEP GW REQUIRED FOR INSTALLATION OF ADDITIONAL FGD EQUIPMENT
EMPLOYEE
Designers
Engineers
Craftsmen
Administrative
TOTAL
MAN-YEARS
PER GW
1.0
1.28
8.98
0.77
12.03
Source: IGC, Inc., 1977.
This situation could lead to increased imports of petroleum products
(as much as the equivalent of about 550,000 barrels of oil per day
*
by 1995) as discussed in the following paragraphs. Such an outcome
This quantity barrels of oil per day represents approximately 0.7
percent of projected U.S. energy demand in 1995 and about 1.7 per-
cent of estimated domestic fossil energy production for the same
year. (The Nation's Energy Future, Report to the President by the
Pay Committee, 1 December 1973).
7-71
-------
is predicted in one analysis of the economic effects of revised NSPS
(ICF, Inc., 1978b). Increased petroleum imports to meet the need
would adversely affect U.S. balance of payments in foreign trade and
would exacerbate competition for increasingly scarce derivatives of
petroleum, especially if a sudden shortage should develop, such as
that resulting from the Arab oil embargo of 1973.
Scrubber requirements result in higher capital costs for burning
coal.
The effect of increasing the costs of burning coal is
to increase the minimum capacity factor at which it is
economic (in terms of minimizing generation costs) to build
and operate a new coal plant... The higher capital costs
(as a result of the scrubber requirement) would have to be
allocated over more kilowatt hours to maintain the same
total generation costs (i.e., capital plus operating plus
fuel costs), where the breakeven total generation costs
between coal and oil would be set by the cost of burning oil
in existing steam plants and new turbines (which does not
change as a result of the alternative NSPS). This results in
a) less coal capacity constructed, b) this coal capacity being
operated at higher capacity factors, c) existing oil and gas
steam capacity being utilized at higher capacity factors (and
hence burning more oil), and d) more turbines being built to
satisfy daily and seasonal peak load requirements that would
have been provided by existing oil and gas steam plants. (ICF,
Inc., January 6, 1978).
Estimates of the amount of oil and gas consumed at 5-year inter-
vals under the alternative NSPS for both moderate and high growth
rates of energy appear in Table 7-36. These estimates are based on
scenarios in which power generation grows at the rates and with the
distribution shown in Table 7-37. Each alternative NSPS can be seen
to result in decreased generation of electric power from coal and
7-72
-------
TABLE 7-36
UTILITY OIL AND GAS CONSUMPTION
1015 Btu (Quads)3
Year
1975
1985
1990
1995
1.2
6
8
5
5
Ib
.5
.2
.8
.2
Reference
90%
_
8.6
6.4
6.1
Case
80%
_
8.7
6.2
5.9
1° Reference
0.5
8
6
5
Ib
.7
.4
.9
1.2
6
8
6
7
Ib
.5
.2
.4
.2
90%
_
8.6
7.1
8.3
Case IIC
80% 0
_
8.1
7.1
8.2
.5 Ib
8,7
7.1
8.3
aOne quad is equivalent to approximately 5 x 10^ bbl/day of oil.
^Assumes growth rate of 5.8 percent per year 1975-85 and 3.4 per-
cent thereafter.
cAssumes growth rate of 5.8 percent per year 1975-85 and 5.5 per-
cent thereafter.
Source: IGF, Inc., 1978b.
7-73
-------
increased generation from methods using oil and/or gas (ICF, Inc.,
1978b).
The results calculated in Table 7-36 reflect scenarios in
which new oil-fired plants such as those using a combined cycle are
prohibited for nonpeaking purposes, except in Southern California
where they are assumed permitted for environmental reasons. The use
of combined-cycle operation would have resulted in even more oil
consumption. It should be noted that a 1976 study using the PIES
model of the Federal Energy Administration predicted that the use of
combined cycle operation under standards requiring 90 percent SO
reduction would increase oil consumption by as much as 1 million
barrels per day (ICF, Inc., 1978b). It should also be noted that the
projections in Table 7-37 differ from those shown in Table 7-2. As
discussed below, the projections in Table 7-37 are based on an
assumption that the generation of power from nuclear sources will
remain constant under all scenarios.
As seen from Table 7-36, the alternative NSPS are projected as
increasing oil and gas consumption for power generation by 0.2 quads
(about 10 barrels per day of oil) in 1985. Thereafter, the increase
varies with the scenario. In 1990 the range is projected as the
equivalent of 200,000 barrels per day for the 80 percent reduction
scenario under an assumption of moderate growth to 350,000 barrels per
day with a high rate of energy growth. In 1995 the additional con-
sumption ranges from about 350,000 barrels per day (for 80 percent
7-74
-------
co
I
o
&
o
M
1
w
o
O>
ON
H
vO
O
rH
CM P^ rH O CM 00 rH
CO vO OO OO CM VO in
r-» r~~ J r^- ON O i 1 vo ON
o vo m co OM cxi in
co r--. ON , 03
U VJ
03 JS
Q) 0) 4J
SCO
e -H -H
cd FQ jD *^
a & t-i &
i-i 03 4-1 O 3 CO"
cO co CO c_) EH
CD O O
rH rH -^ J-4
O CO rH TJ
3 O -H >,
g; O o tri
P
H
pq
vO
0
rH
ON vo CM f^ in in ON ^3~ CM
rH O CO vo<(- COin>3-vO
oo
CO cO
J-i !-(
r« ,13
4J 4J
IS 3
O O
V-i S-i
M) 00
CO 03
(IJ 0)
S B
3 3
cn en
cn w
o
C
M
01
o
J-l
3
O
CO
7-75
-------
TABLE 7-38
THE SHARE OF IMPORTS IN U.S. DOMESTIC PETROLEUM DEMAND
(Thousands of Barrels)
YEAR
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975b
1976b
U.S.
ANNUAL
TOTAL
159,389
188,144
235,559
310,261
308,194
348,507
377,499
383,955
455,564
525,591
574,589
620,589
649,583
664,111
699,666
759,793
774,713
826,736
900,772
939,162
925,991
1,039,369
1,155,551
1,248,062
1,432,880
1,735,314
2,283,493
2,230,947
2,210,335
2,669,929
IMPORTS
DAILY
AVERAGE
437
514
645
850
844
952
1,034
1,052
1,248
1,436
1,574
1,700
1,780
1,815
1,917
2,082
2,123
2,259
2,468
2,573
2,537
2,840
3,166
3,419
3,926
4,741
6,256
6,112
6,056
7,295
IMPORTS
AS A %
OF
DEMAND
8.0%
8.9
11.1
13.0
11.9
13.1
13.6
13.5
14.7
16.3
17.8
18.6
18.7
18.5
19.2
20.0
19.8
20.5
21.4
21.3
20.2
21.2
22.4
23.3
25.8
29.0
36.1
36.7
37.2
41.8
Revised
Preliminary
Source: U.S. Bureau of Mines, 1976.
7-76
-------
ADDITIONAL IMPORTS
UNDER $7/BBL OIL
PRICE ASSUMPTION
PORTION OF DEMAND
SATISFIED OUT OF
DOMESTIC SOURCES
ADDITIONAL IMPORTS
i UNDER S7/BBL OIL
PRICE ASSUMPTION
1950
1955
1960
1965 1970
YEARS
1975
1980
DEMAND
AT $11/BBL
DOMESTIC SUPPLY AT
$11 BBL
DOMESTIC SUPPLY AT
$7 BBL
1985
Sources: U.S. Bureau of Mines Petroleum Statement, Annual and December.
American Petroleum Institute, Basic Petroleum Data Book.
Washington, D.C. October 1976.
Federal Energy Administration, Project Independence Blueprint,
"Project Independence" (Title of Volume), November 1974, pages
23 and 48.
U.S. Energy Research and Development Administration, Final
Environmental Impact Statement, Alternative Fuels Demonstration,
Vol. I, 1977.
FIGURE 7-6
ACTUAL AND PROJECTED PETROLEUM DEMAND
AND DOMESTIC PRODUCTION
(1950-1985)
-------
reduction under moderate growth) to some 550,000 barrels per day (for
high growth with 90 percent reduction or with a ceiling of 220 ng/J
0.5 Ib ?0?/106 Btu.
The above results do not take into account the user taxes and
rebates to stimulate use of coal and nuclear generation instead of
oil and gas as proposed recently in the President's National Energy
Plan (U.S. Congress, 1977). An assessment of the effects of this
program indicates that under continuation of the present standards
the proposed program of taxes and rebates could save an additional
£50,000 barrels per day compared with the quantity used under 90
percent SO reduction (IGF, Inc., 1978).
Tt is likely that any sizable increase in requirements for oil
and gas would be met through foreign imports. The U.S. shortfall in
domestic production required to meet demand (Table 7-38) has increased
from about 8 percent in 19A7 to over 41 percent in 1976. The need
for imported petroleum is projected to grow even more in the period
to 1995 (Figure 7-6). Of course it is possible that by 1995 synthetic
fuels from coal may be available for generation of electricity. But
it should be noted that the projections of power generation are based
on an assumed price of petroleum with which synthetic fuels are un-
likelv to be economically competitive. Introduction of synthetic
fuels as a source of electric power generation would change the
estimates of power generation from coal as well as oil and gas.
7-78
-------
Actual and projected costs of imported petroleum products
(expressed in dollars per barrel of oil equivalent) are shown in
Table 7-39. Total expenditures for the projected increase in
imported petroleum products for key years of the 5-year intervals
under each scenario appear in Table 7-40.
If not obtained through input of additional foreign petroleum
products, the increased gas and oil consumed would be in competition
with other uses such as #2 heating oil for residences and commercial
establishments. Again the assumption of synthetic fuels from coal
could change the picture.
It should be noted that the ICF results were calculated under
scenarios in which the contribution of nuclear sources to total
power generation was held constant. The projections by ICF (1978) in
the 90 percent reduction scenario of 31 percent power generation from
oil and gas and 17 percent from nuclear are very nearly a reversal of
the percentages of 20 and 35 percent, respectively, calculated by
Teknekron (1978) under the assumption that nuclear contribution was
not held constant. But it is not necessarily a valid inference from
this fact that increased use of nuclear sources would obviate the
need for additional power generation using oil and gas. "Most
studies show nuclear trading off with coal in baseload and coal
trading off with oil in intermediate load, but not nuclear trading
off with oil" (ICF, Inc., 1978b) . The assumptions used in the two
7-79
-------
TABLE 7-39
IMPORTS - PETROLEUM PRODUCTS AND NATURAL GAS
NET DEFICIT
VALUE BALANCE
YEAR ($ BILLIONS) ($ BILLIONS)
1965
1967
1969
1970
1971
1972
1973
1974
1975 (1st qtr.)
2.2
2.5
2.8
3.0
3.6
4.7
7.9
2.4
6.4
- 1.7
- 1.6
- 2.2
- 2.3
- 2.8
- 3.9
- 7.1
-23.4
- 6.2
Source: U.S. Department of Commerce, 1974.
7-80
-------
3
u:
B:
z
c" S " c
cr E- L" o r"; o
Z fri 5 C
t- c c;
_^ CC C=3
< *b
CO DC H
C£
_£ > clb
*~ ^ ~^J ^
< S S """
O W CC 0}
-o> CQ » v^ f*i cv;
'l^s
tJ <
wfw £
P2 ' £ ^
o
> 0*
c
0 1
o c
u'', O f~i O*-
c
J- CNJ O O>
p" i/^ r^-
o
coo
Cvj QQ vC CO
*v cc tfi cn
C/J J 01 «1
CO . O Q
m 3 ID u O o
CO CT rH C^
rH M CU QJ i 1
rH rJ « W
H (6 CO fO
O CO 01 Q>
rH ^ CJ U
CC « CC C C
C C r-t r-t
J-J JJ d CC
*0 "C C C
TD -O C C
sC «M -C
Is-- fSI CC CC
^-i CC f*>
c
-U> V-
r- CM cc cc
e " w ^
t->
^. K o S
C m r-
co so co
v£5 O t-T\ CO
O *~"
r-i
o o o
in * *-
rH 1^- rH
rH
T3
ID rH
QJ )-i
l-< ct
CO CQ
CC
\-t fii
OJ C.
p-
H
« £
U CO CO
L-
^ co w cn
CO OJ 4-J AJ
T3 >-» tC W
CC *~^- O O
3 to u u m
O* rH O
W OJ O rH
rH l-i IE CO
H Cfl CO CO
O W OJ QJ
rH r-t 0 O
CO cc p C
c o
4-J *J CQ CO
*H -H 2 3
T; T3 c C
c c c
o oc ^ oc
r-l rs| tT1
r- * *
-tTr V>
C C C
rH C C C
OJ V>
r* CNJ cc co
rH 00 ^
O
rH
ft r^ fH
rH rH f^-
o -
rH rH
000
CM
vD O
rH
r*< 0)
0) W
CC fC,
PC
i- a
a» ex
a
1 rH
^>
U co to
^. CC Ul tA
Cfl QJ 4-i JJ
-O >- Ift W
'w QJ 13 tJ
rH tw 05 C/)
-H CO CS CO
O CQ OJ OJ
rH rH U U
TO ftj C ^C
5 -2
*J 4J CQ ^
H M 2 3
*O tJ C C
< < < <
i_l
a)
u
CC
CJ
u
a-
j:
4H
4J
S
CJ
P
(U
a
>
lu
OJ
ex
J_)
c
QJ
U
t-i
OJ
a
CO
in
0
QJ
CC
u
4-J
O
en
3
cn
05
<;
LJ
a>
4-t
cc
0)
V4
QJ
4-i
4J
§
u
14
QJ
a
in
in
"C
c
1C
in
oc
ey\
I
in
r^
o
^
TC
OJ
>,
>^
o
a.
u
c
0)
u
Ul
OJ
c.
00
m
o
d> Eft CO
re QJ CD
l-i U I
^ -H
x re o
O IM uj
lJ O O
C£
co cn
en c C
0) O O
6 -H TH
W 1 ~4
W -H v-l
<5 ^ £
V-
JZ.
4J
^
M
oc
^
a
f.
CO
^H
£
14-4
O
C/}
C
o
T<
JJ
Ct
S
CO
CO
re
i-i
QJ
O
C
3
CO
1U
?>
^
4-1
CD
C
^
0)
rH
til
O
3
u
cu
01
J=
^
o
CO
4U
CO
01
to
u
J_l
c
o
t-^
^1
00
tH
"
o
c
M
b
rH
u
3
o
to
7-81
-------
TABLE 7-41
U.S. BALANCE OF FOREIGN TRADE
(1969-1973)
YEAR
1960
1961
1962
1963
1964
1965
1966
1968
1969
1970
1971
1972
1973
VALUE OF TOTAL
EXPORTS
($ BILLIONS)3
20.58
21.00
21.70
23.35
26.50
27.48
30.32
34.63
38.01
43.22
44.13
49.78
71.31
VALUE OF TOTAL
IMPORTS
($ BILLIONS)
15.02
14.71
16.38
17.14
18.68
21.37
25.54
26.81
36.04
39.95
45.56
55.56
69.12
VALUE OF EXPORTS
MINUS VALUE OF
IMPORTS
($ BILLIONS)
5.56
6.29
5.32
6.21
7.82
6.11
4.78
7.82
1.97
3.27
-1.43
-5.78
2.19
Excluding Department of Defense shipments.
Source: U.S. Department of Commerce, 1976.
7-82
-------
calculations by ICF, Inc. and Teknekron were different, and resolu-
tion of divergent results cannot be achieved by so simple a means
as ascribing them to the single factor of constant versus variable
contributions from nuclear sources.
Taking all of these uncertainties into account it should be
stressed that the extent of increased consumption of oil and gas and
additional imports of petroleum reflects more contingencies than do
any other potential impacts discussed in this chapter.
7.5 Cost Effectiveness of Revised NSPS
Cost-effectiveness, or cost-benefit analysis, represents an ex-
tremelv useful methodology for assessing the relative advantages and
disadvantages of a set of alternatives, such as revisions in NSPS.
By making explicit the prices associated with the anticipated bene-
fits, cost-effectiveness aids the decision maker(s) in focusing on
critical issues. More generally, cost-effectiveness seeks to quan-
tify the effects of each alternative and present them on a basis for
comparing gains of each with the costs.
7.5.1 Costs of SO,, Reduction on a Ton-Per-Year Basis
An important partial measure of the cost-effectiveness of alter-
native NSPS revisions is provided by considering the price calculated
for each scenario at which a unit amount (such as one ton per year)
of SO is removed. That is, typically two alternatives are estimated
to prevent emission of different amounts of SO at costs to the elec-
tric utility industry (and hence ultimately to the consumer) which
7-83
-------
when calculated on a common basis represent different monetary
values. It is then instructive to examine the marginal costs of the
more expensive alternative: i.e., whether the cost of removing each
ton per year (TPY) which would otherwise be emitted is higher or
lower than the cost of each TPY removed under the cheaper option.
A specific measure of these marginal costs has been provided for
two coal types and for selected scenarios as shown in Table 7-41.
All calculations reflect a single power plant of 500 MWe generating
capacity. Thus the specific question addressed is "What are the mar-
ginal costs per TYP of decreased S02 emissions under different NSPS
for each 500-MWe plant erected?" Inferences as to cost-effectiveness
on this basis should be readily extensible to an individual plant of
any size; although the cost of each TPY can be expected to vary
according to the size of the plant, it may be reasonably expected
that a difference in plant capacity will not affect which of two NSPS
alternatives is more cost effective.
Estimates in Table 7-42 readily show that for a coal with 3.5
percent sulfur content the TPY cost (on an annualized basis) for 90
percent removal of SC>2 is achieved at a lower cost for each TPY
than for the baseline situation (i.e., maintenance of the present
level of 1.2 Ib S02/106 Btu). What is even more striking, is
that the additional 8,300 TPY captured under the 90 percent scenario
are achieved at a cost only 75 percent of that required at which each
TPY is removed in the baseline situation. On this basis, the extra
7-^84
-------
CO
M
CO
H
M
4-1 ^>
CO t-l
o cd
O 0)
^-1
T3 ***>
a) 0
N O
H -rl
r-l r-l
Cd rH
3 -H
Pi
<|
-a
j-i
3 **"N
4J ^
D. f^
CT) H
U v^
CM
O
CO
cu
a
H
H
«
O
O
o m o
o m m oo co m
O 00 1 <3~ -J" O
"A * CO H rH H
oo a\ r-4 oo oo o
f- O CO CSI
^*N ^**\ /~S
iH rH r-l
cfl cfl cd
r* K^ ^
O O O
e e B
S 53 ft
CO P^ fyj p^
x^
CM 6-5 6-S B^S
. 0 00
i i c?> oo o>
^ ' V^ ' N^^ S^/
O O O O O O
0 O O O O O
^O O^ cO
-------
cost of removing an additional 8,300 TPY through a 90 percent reduc-
tion level may appear to represent a bargain.
Similar results are indicated when the same kind of calculations
are applied to 0.8 percent coal under the two scenarios of 90 and 80
percent SC>2 removal. Again it is seen that the greater quantity of
SC>2 captured from a 500-MWe power plant under the 90 percent reduc-
tion standard is achieved at a lower cost in TPY. Also, the cost of
preventing emission of an additional 1,900 TPY represents a marginal
cost of $1050 per ton - about 70 percent of the cost at which each
ton is captured under the 80 percent removal scenario.
This particular measure adds a highly important dimension to the
economic impacts of revised NSPS for SC^. It is noteworthy because
it is based on two representative coal types and a plant-size which
may be fairly typical for the future. However, neither it nor any of
the other aspects discussed in this section, whether singly or in
combination, can provide a fully comprehensive basis for a final de-
cision as is discussed below.
7.5.2 Limitations of Cost-Effectiveness
There are inherent limitations in cost effectiveness; and final
decisions among important alternatives (such as revised NSPS) must be
supplemented by the expertise of the decisionmakers. Some of the
limitations reflect the present state-of-the-art and will likely be
remedied by further scientific investigation. Others are due to
constraints on the resources of time and human effort that can be
7-86
-------
devoted to the required data gathering and interpretation. Other
areas lie outside the range of objective analysis and reflect dif-
ferences in interests within the population affected and variations
in values.
Cost-effectiveness seeks to quantify the effects of each alter-
native as a basis for comparing gains of each with the costs. When
the objectives of a set of alternatives are limited and can be pre-
cisely defined, and when few indirect or secondary impacts are ex-
pected, the problem is simplified. Many variables that govern the
extent of both gains and losses can be successfully measured and
numerical values assigned, as is shown in this report. Costs of com-
ponents of the operations involved can be quantified, as can the
prices of end-products, the physical effects such as tons of pol-
lutants released into the atmosphere, and the resources required.
Many secondary impacts, however, cannot be successfully measured be-
cause there is no adequate data base to support the development of
reliable measures. It is not possible within the scope of the pres-
ent study to measure, for example, impacts from variations in the
amount of oil and gas imported or ramifications of changes in the
quantities of scrubbers produced. There is no way to assign a real-
istic measure to the possible increases in agricultural production
that may result from "clean air", or to cost savings to property
owners from pollutant reduction or to human health improvements. It
is not possible to assess the potential effects on migration of
7-87
-------
workers and their families resulting from changes in anticipated
coal production under the revised NSPS.
A major difficulty in the use of cost-effectiveness is the ab-
sence of a common base for costs and gains. To the extent that ad-
vantages and disadvantages can be reduced to the same unit of mea-
surement, such as dollar-value, trade-offs among alternatives are
simplified. It becomes possible to determine the net. gain or loss of
each and to consider the marginal costs and benefits.
A final limitation on cost-effectiveness is due to the uncer-
tainty of projections into the future. Best estimates are subject to
error from unforeseen events or conditions. By making explicit the
assumptions about the future on which the analysis is based, the
cost-effectiveness methodology serves the decisionmaker(s) who can
then allow for possible future changes.
7-88
-------
APPENDIX A
S02 REMOVAL MECHANISMS AND EFFICIENCY
Material in this appendix was extracted from "Technical Document
on Applying Tentative NSPS for Coal Fired Steam Generators to Fluid-
ized Bed Combustion" a draft prepared for the U.S. DOE, November 1977,
SO,., Removal Overview
When coal and calcined limestone or dolomite are burned in a
fluidized bed, the sulfur released from the coal reacts with the
limestone or dolomite. In the presence of excess air, the sulfur in
the coal is oxidized to SO- and the reaction with the sorbent pro-
\
duces calcium sulfate. To replace the used sorbent (limestone or
dolomite) new limestone or dolomite is usually added into the bed
where in-situ calcination takes place by the following reactions to
produce lime (CaO) or calcined dolomite CaO-MgO:
CaC03 - -CaO + C02
Ca C03 MgC03 - -CaO MgO + 2C02
The reaction of the lime with S02 appears to be a fairly simple
reaction
CaO + S02 + 1/2 02
lime + sulfur dioxide + excess air - ^Calcium Sulfate.
However, it has been shown that the reaction in the bed with coal
combustion is very complex. Many factors affect the retention of
sulfur by the lime,
A-l
-------
The most important factors include: Ca/S ratio, type of sorbent,
bed temperature, bed pressure, and residence time. The discussion on
sulfur removal is based on data from bench scale and small pilot
plant operations, and it is difficult to estimate the effect scale-up
to a commercial size unit would have on these data. The results are
further limited since most of the data were gathered on only two
sorbents and there is little information on low sulfur coal.
Atmospheric Fluidized Bed Combustors
The sorbent ratio (moles Ca/rooles S) required to achieve a
90 percent reduction of S02 emission from small pilot scale
AFB combustors is shown in Figure A-l, ranges from 3 to 4. This
range of Ca/S values has been found to be independent of the sulfur
content for higher sulfur coals, but is dependent on the specific
sorbent employed as shown in Figure A-2. Table A-l shows the Ca/S
mole ratio required for a range of sulfur retention levels for two
specific coals.
TABLE A-l
ESTIMATED Ca/S MOLE RATIO TO ACHIEVE VARYING SULFUR
RETENTION LEVELS
Sulfur Retention Level
92
90
85
80
75
70
Sewickly Coal
(2.4% Sulfur)
Ca/S Ratio
3.21
3.03
2.85
2.67
2.49
Pittsburgh Coal
(2.7% Sulfur)
Ca/S Ratio
3.75
3.53
3.83
3.18
2.99
2.78
A-2
-------
100
90
80
70
60
50
(.0
40
30
10
I
O EXXOM BATCH UNIT
A NCB
B ARGONNE.N.L.
D EXXON MIN1PLANT
600-800 UPa
7 50-9 20 °C
2 3
Ca/S (MOLE/MOLE)
FIGUKE A-l
EFFECT OF Ca/S MOLE RATIO ON SULFUR RETENTION
A-3
-------
53
O
CJ
o
w
CN
o
w
Germany Valley
Limestone
0
U
for
RANGES
FOR DATA
Datu Uncorrected
Di ffer ence s in
Operating Variables
OF VARIABLES
POINTS SHOWN
Gas Superficial Velocity
Gas Residence Tiae
Static Bed Depth
Coal Feed Hate
0 2 in Flue Gases
Bed Temperature
12. 5 -I1'. 6 ft/sec
.20 2 -.266 sec
15-5-22.8 inches
630-720 Ib/hr
3.0-3.6?
1500-1530°F
1.0
2.0
3.0
4.0
5.0
Ca/S
FIGURE A-2
COMPARISON OF PERFORMANCE OF GREER AND GERMANY VALLEY LIMESTONES
A-4
-------
Figure A-3 indicates the variation in sulfur retention with
change in bed temperature. The drop in sulfur retention at the lower
temperatures is related to the failure of the sorbent to calcine. At
higher temperatures, the decomposition reaction of CaSO/ is favored,
particularly if a minimum level of oxygen is not maintained (e.g.,
in carbon burnup cells, higher operating temperatures tend to expel
S0~ from the spent sorbent mixture if a minimum of 3.5 percent oxygen
content is not maintained). As shown in Figure A-3 the sulfur
retention has been observed to reach a maximum at about 1450°-1550°F.
It should be noted that no satisfactory thermodynamic or kinetic
explanation exists, at this time, for the occurrence of this maximum.
The effect of superficial gas velocity (gas residence time) on
sulfur retention levels is shown in Figure A-4. The vertical scale
indicates a reduction in the sulfur retention as the superficial
velocity is increased, that is to say, as the residence time of the
gases in the bed is reduced. In general, the sulfur retention
increases as the bed depth is increased. This is expected since both
the solids and the gas residence time are increased. There is, how-
ever, a practical limit on bed depth since the bed pressure drop
increases directly as the bed depth increases, and the final bed depth
may be dictated by the maximum allowable pressure drop for available
fans.
Sorbent Requirements to Meet a 90 Percent SO Reduction Requirement
and the Present NSPS Using an Atmospheric FBC System. Table A-2
A-5
-------
?,»
2-
O
H
UJ
LU
OS
K
U.
to
GAS VELOCITY. 3 fI/sec
EXCESS OXYGEN, 3%
LIMESTONE NO. 1359
O ILLINOIS COAL, Ca/S-2.5
PITTSBURGH COAL, Ca/S-4.0
T
1300
1400 1500
TEMPERATURE, °F
1600
'7QO
FIGURE A-3
COMPARISON OF S02 REMOVAL RESULTS - LIMESTONE SORBENT
A-6
-------
o
H
H
H
W
Pi
Pi
en
100
90
80
70
60
50
40
30
20
10
Ca/S -1
CRE DATA (1470 *F)
' " * "Q"*
AHL DATA (1 550 *F)
WELDECK COAL, BRITISH LIMESTONE i«0}!m}
ILLINOIS COAL, LIMESTONE NO. 1359 (>1000
A ILLINOIS COAL. POINT TAKEN FROM CURVE (FIGURE 2)
FOR LIMESTONES AND DOLOMITE (5630pm), Ca'S =4
2 4 6
SUPERFICIAL GAS VELOCITY, ft/sec
FIGURE A-4
SULFUR RETENTION AS A FUNCTION OF SUPERFICIAL GAS VELOCITY
A-7
-------
indicates the variation of limestone requirements (based on a Ca/S
performance curve) for Greer limestone for different sulfur contents
of the coal. The table compares the limestone quantities for 90
percent retention and for meeting the present 520 ng/J heat input
(1.2 Ib S02/10 Btu) standard for a fixed design atmospheric FBC
system.
TABLE A-2
SORBENT REQUIREMENT FOR AFBC TO MEET EPA S02 EMISSION
STANDARDS BASED ON PILOT PLANT DATA
Sulfur Content in Coal %
Limestone Required to Meet
90% S02 Reduction-lb/ton coal
Limestone Required to Meet
Present NSPS Ib/ton coal
Additional Limestone Required
for 90% Control-lb/ton coal
1
219
31
188
2
437
225
212
3
656
487
168
4
875
750
125
5
1093
1030
62
Assumptions:
(1) Coal HHV of 12,000 Btu/lb.
(2) Limestone is 100% CaC03.
Pressurized Fluidized Bed Combustors
Limestone and dolomite are inexpensive sorbents that are
widely abundant in eastern U.S. areas where much of the high sulfur
coal is found. It should be noted that the magnesium in dolomite
does not absorb S0~ under FBC conditions, and that to supply a given
weight of calcium requires a dolomite input 1.8 times that of
A-8
-------
limestone and disposal of a correspondingly larger amount of solid
waste. Differences in effectiveness of dolomite and limestone as
sorbents, especially in PFBC, can partially compensate for the weight
factor.
The activation of both limestone and dolomite for SCL absorption
depends on the creation of a porous structure with a large internal
surface area as a result of calcination reactions. Calcination of
CaCOn occurs readily at bed temperatures in AFBC, but is inhibited
at the higher pressures in PFBC, whereas calcination of MgCOo occurs
readily in either AFBC or PFBC. Because of the inhibition of lime-
stone calcination, dolomite is more effective than limestone at a
given value of Ca/S (moles of calcium fed in the sorbent to moles of
sulfur fed in the coal).
The effect of gas phase residence time, calculated as the
ratio of the expanded bed height to the superficial gas velocity, was
investigated at EXXON. The effect of residence time is more pro-
nounced at high S0~ retention levels and the magnitude is such that,
for 90 percent S0? retention, decreasing the residence time from 3 to
0.5 seconds would require doubling the Ca/S ratio from 1.5 to about
3. The dependence on gas residence time of the Ca/S ratio required
to meet the present EPA SO^ emission standards is shown in Figure A-5.
The effect is not large but evident. As for example, with a 2 percent
sulfur coal, an increase in gas residence time from 1 to 3 seconds
reduces the Ca/S requirement from 0.95 to 0.75.
A-9
-------
o
UJ
txJ
O
2:
tu
loJ
cc
CO
<
o
Q
I
CO
a
o
M
00
CO
M
a
w
r-J
O
CO
w
w
CO
w
oi
PH
H
W
W
S
Q
W
OS
O
M
En
o-
CO
«J
w
cj
O
M
CO
w
CO
-------
At temperatures between 1400° and 1750°F, very little effect of
temperature on sulfur retention for dolomite has been observed. SOj
retention levels measured at 1270° to 1400°F were slightly lower than
those measured at the higher temperature. However, with limestone a
marked effect of temperature was observed, with increasing tempera-
ture giving higher S0? retention levels. These S0~ retention levels
were lower than those observed using dolomite sorbent, and it is felt
that these effects were due to the inability of the limestone to
calcine completely under pressurized combustion conditions. At
higher temperature (1700° to 1740°F) limestone undergoes extensive
calcination, and, although the limestone is not as active as dolomite,
it is more active than at lower temperatures (1500° to 1650°F) where
the stone is largely in the carbonate form.
Although an FBC utility boiler would normally be expected to
operate in the temperature range of about 1550° to 1750°F, operation
at temperatures down to about 1300°F would be required to turn-down
the boiler output to match a decrease in the electrical power demand.
A series of runs was made by EXXON using dolomite and limestone
sorbent at temperatures near 1300°F to determine the behavior of the
FBC system at these lower temperatures. Some runs were also made at
temperatures as low as 1270°F to determine the lowest limit of
operability. The minimum temperature at which combustion was stable
was 1270°F. An attempt to decrease the temperature control in the
combustor became erratic and carbon monoxide emissions in the flue
gas increased sh'arply, denoting poor combustion.
A-ll
-------
Results of the tests are shown in Table A-3. A slight decrease
in S09 retention was seen using dolomite sorbent at low temperatures.
However, limestone was completely inactive and, therefore, cannot be
used in a pressurized FBC unit unless some means of increasing its
activity under low temperature "turndown" conditions can be found.
Precalcination of the limestone is one possible way to do this and
will be studied in the future.
Sorbent Feed Required to Meet the EPA SO Emission Standards
A plot of sulfur retention vs. the sorbent mass feed rate (Figure
A-6) shows that calcined Grove limestone (930°C/1700°F combustion
temperature) is far more effective than the partly calcined stone
(880°C/1600°F combustion temperature). It also shows that Pfizer
dolomite is more effective on a weight basis than calcined limestone
at sulfur retention levels greater than 70%. While far higher
utilizations can be achieved with dolomite than with limestone,
dolomite contains only 50 weight percent CaCO_, as compared with a
100 percent CaCOo content for limestone. Figure A-6 can be used to
estimate the sorbent feed rate required to meet the present EPA
standard for any sulfur coal by calculating the equivalent retention
level. Table A-4, which gives the sorbent feed requirements for
1 to 5 percent sulfur coals with 12,000 Btu/lb heating value, was
prepared in this manner. At a coal sulfur level of 2 percent, 20
percent less limestone is needed as compared with dolomite to meet the
present NSPS. At higher coal sulfur levels, it becomes increasingly
A-12
-------
00
1
^
w
PQ
H
to
O
H
M
Q
O
g
0
O
!S
OS
H
H
D-i
r
i !
vO
i-l O Q
OS P-i
OO
o
o
03 Q) -i-l
C ^ "
O 3 to
r-l 4J pe! C
4J CO O
r-t VJ T3 -H
*XJ QJ CD Wi
C CX 01 en
o 6 fc -H
cj oj E
H VJ W
cjO CO
C T3 i l en
!-( 0) O 4J co
4J PQ S C O
rt a>
V-t C/2 ,.Q Q)
a) oo ~~^ t-i 3
p. p> cfl O r-i
O < O CO fn
o
CO QO
o\ in o o c?
CN CN o r~-
C"> t-l r-<
»
O i 1 1
vO
Csl O O O CN
1 Csl iA
CT\ i I OO
m
CO csl O Csl O
v£> O O 00
r_f r ,( ^V
^f
vO
O vj O vO O
oo i o r~-
^-1 -< OO
oo
r^
co o^ in i~~^ ^*~^
CN Csl CN sf
CO i-l CO
CN
i i in m o CN
vO co r^^- *^
i i i i
B-S
f\
(3
O
H
4J ^
C 4 J
0) PP
4J ^
a> o
6 B f* ^
a. D- a »-i CN
a. 3 o
- y-i co
CN X " tH
o o o 3 ja
co a o co ij
^s
1
CO
CO
, l
O
a
01
4-J
i-(
£3
o
1 1
o
o
^1
01
bo
1-4
M-<
PM
II
Q
OH
A-13
-------
]00
80
60
2
w
ft,
40
20
0
DOLOMITE
Limestone - 930°C
(Calcined)
; Limestone - 880°C.
(Partly Calcined)
1 1
4 6
Kg FEED SORBENT
!
8
i
20
Kg COAL SULFUR
FIGURE A-6
COMPARISON OF DOLOMITE NO. 1337 AND LIMESTONE NO. 1359 AS SO,
SORBENTS ON A MASS FEED RATE BASIS ^
A-14
-------
TABLE A-4
SORBENT REQUIREMENTS FOR PFBC TO MEET EPA S02
EMISSION STANDARDS BASED ON PILOT PLANT DATA
Sulfur Content in Coal %
Dolomite Required to Meet
90% S02 Reduction (Ib/ton coal)
Dolomite Required to Meet
Present NSPS (Ib/ton coal)
Additional Sorbent Needed for
90% Reduction
1
170
23
147
2
340
200
140
3
520
400
120
4
640
580
60
5
820
800
20
*For coal with 12,000 Btu/ Ib HHV
more attractive. Thirty percent more limestone than dolomite is
required to meet the present NSPS with a 5 percent sulfur coal. With
this coal, the Ca/S molar feed ratios would be 3.2 for limestone and
1.3 for dolomite. It should be noted that 90 percent of the U.S.
coal reserves have a sulfur content under 5 percent. Also shown in
Table A-4 are estimates of sorbent feed rates required to meet 90
percent reduction levels based on the assumption that the Ca/S ratio
is independent of the coal sulfur content. At this level of sulfur
retention, dolomite is the most effective sorbent for all sulfur
levels.
A-15
-------
APPENDIX B
ENERGY REQUIREMENTS
The information summarized in this appendix shows how energy
requirements depend on SCL control method, level of control, and
coal sulfur content. The data show that energy requirements for
SQ~ control systems (expressed as energy required per unit of elec-
trical generating capacity) depend only slightly on plant size.
Design Assumptions
The energy requirements for operating flue gas desulfurization
systems were calculated based on the process designs summarized in
Table B-l. The design assumptions for coal cleaning processes are
shown in Table B-2. A unit train consisting of 100 coal cars and
five locomotives was the design basis for coal transportation. Fuel
consumption rates, transport distance, train speeds at full and
reduced power, and coal dust blow-off losses were specified.
Table B-3 shows calculated energy requirements for the six
processing operations in FGD systems. Particulate/chloride removal,
reheaters, and fans account for 65 to 90 percent of total energy
requirements for nonregenerable FGD processes. Sulfur recovery
operations account for the majority of energy requirements for
regenerable FGD processes.
Comparison of Energy Requirements by SO^ Control Methods
Figure B-l shows the energy required to meet the existing NSPS
of 520 ng/J (1.2 Ib S02/10 Btu) using different S02 control methods,
B-l
-------
O
Fn
w
I-J
CO
V T9
5 §
X
co
W
co
co
W
u
o
PS
P-i
Q
O
H
V
CO
W
P
CO
CO
w
o
o
S Si
S J
SJ5
S-S
B-2
-------
TABLE B-2
DESIGN ASSUMPTIONS FOR PHYSICAL COAL CLEANING FACILITY
40% sulfur removal
95% of energy recovery efficiency
The electric power requirements for a
278 kg/s (500 ton/hr) cleaning plant
are 2980 kW.
The heat duty of a thermal dryer is
534 kJ/kg (230 Btu/lb) of coal dried.
One half of the clean coal product (the coal
fines) is thermally dried.
Heat for the thermal dryers is supplied by
burning a portion of the clean coal
product.
50% of the ash content of the coal is removed.
The average heating value of the clean coal is
29.2 MJ/kg (12,500 Btu/lb).
B-3
-------
co
PQ
W
rJ
CO
H
P
O
§
M
H
W
PM
O
O
w
o
o
w
ffi
9-t
Pi
O
Fn
CO
H
2
W
ex
w
q
«i
w
Z
Fd
K
6 cn
CU 4-1
4-1 C
CO CU
r*> c
CO CU
rH 60 -H
CO }-l 3
4-1 CU CT
0 C CU
H W PS
^
r*^ rH
S-l CO
Vl CU CO
3 > 0
LM o Cu
rH O CO
3 01 -H
C/3 Crf P
CO
C
cfl
4-J
cn w
-~~. cu
t } rj
S Oj
Pi
c
o
H 60
4-1 C
Cfl -H
N rQ
H jn
M 3
3 CNi-J
MH O O
rH CO Cfi
3
cn
cu
P ~~~.
cu
CD 4-J
Cfl Cfl
O rH 0)
3 Td rH
CU CJ -rl CO
3 -H !-i >
rH 4-1 O O
f^ ^ rH E
tfl ^3 CU
s-i p., u ft
o
4H
rH ""O
cn 03 C C
4-1 -H Cfl O
Ci S-l -H
0) CU 60 4J
E 4-1 C Cfl
CU cfl -H S-i
S-I 2 rH CC
H Td Cu
3 £ C 01
CT cfl cfl S-i
CU PS 33 PJ
^
6C M
S-i 3
01 14-4
Cl rH
PJ 3 4-J
CO C=
01
rH 4-1
Cfl C
O O
U U
CD
U
' cu
o
o
S-I
(Xl
p
o
PH
CN co in
o>
O
o
co
CN
O
&*2
in
CO
Td
01
H
rH
CO
r-.
O
CT.
^j-
rH
CN
OO
rH
^3-
CN
O
co
co
O
6-5
m
.
CO
H
rH
Cfl
^
rH
<5
1
O)
rH
n
3
O
P
CU
4-J
C
H
Td
CU
Td
3
rH
CJ
a co
H rH
cfl
CD 4-1
rl O
4-1
rH
Cfl g
> 0)
O 4-1
6 CD
cu >-,
\-t cn
cu c
4-1 -H
cfl
rH CO
3 01
0 -H
H O
4-1 ti
(-1 tfl
cfl Cu
Cu CU
M
& CJ
4-1 CO
H vH
13 Td
Td cu
CU f,
4-1 4-J
Cfl
H )-*
0 O
O iJH
cn
CO 4-J
cfl C
3
Cu 0
O 0
i-l U
Td Cfl
oi cn
SJ 01
3 cn
cn cn
CO 01
01 o
}-l O
Cu !-i
Cu
cu
e P
o o
CJ Fn
M
CU 0)
> a &
O O 4-1
H
O 4J M
4-1 CO O
(-4 UH
to cu
C CX CD
cfl O 0)
>4H CJ
rH -iH
>> cfl >
,n > Ci
O 0)
T3 E co
cu cu
H c!
3 cu cfl
cr 4-1
cu to cn
SH rH 01
3 -H
>\ CJ 4J
60 -H vH
>-l 4J rH
Q) IM -H
C tfl 4-'
W Cu p
* K
lied process is used within the process
..
CO
C
O
H
4-1
Cu
S
3
cn
cn
<3$
s>~^
[>-,
4-1
H
O
rfl
p,
tfl
CJ
50
C
H
.U
cfl
0)
C
0)
M
4-J
0)
-_x
4-J
C
cfl
rH
Cu
i-l
CD
0
Cu
J2
^
O
0
m
rH
CO
0
E
CU
^i
c
O
CO
g\5
O
o*»
CO
1
Zz
A!
1-1
o
-<3"
*>O
CS]
II
CU
4-1
cfl
4-1
CO
OJ
Vf1
U
0)
n
4-1
C
tfl
rH
Cu
S-I
CU
[3
O
Cu
Td
a)
rH
rH
o
SSH
4-1
C!
o
o
s
rH
<
T3
S-i
o
fj
1
a
to
E
rH
rH
0)
rs
Td
C
CO
!>-,
}_i
^
3
rH
CO
cfl
rH
cn
0)
C
I3
i
rCI
Cfl
CO
cfl
S-i
3
MH
rH
3
Cfl
CO
CU
0
3
Td
o
^-4
Cu
cn
cn
cu
a
o
S-4
Cu
0)
H
rH
rH
<2
- .
Td
^_]
o
)_ 3
1
C
cfl
E
rH
rH
0)
£2
B-4
-------
LCMMO-
| OIL OK QAS
| COAL CLlANWa AM THAW OIMOACU IO*«M
STEAM AND CLECTRlCtTY
1 DIBSfU JUItL OIL
O zoo
c
H
z
O
I !
It:
1
i
g
ss
1
u
0
fc
HI
i
.t% 3ULFUR.
21 >M«I«
COAL
T OtSULFUR. O.B«SULrUfl. 0.8%SULFUK. 3 5% SULfUR.
27BMJ/4* 20«MJ'k« 2ft.ftMJIkt 1T»Mjr«t
COAL COAL COAL COAL
FIGURE B-l
ENERGY REQUIREMENTS FOR S02 CONTROL - 520 ng/J
AT 500 MW PLANT
B-5
-------
Total energy requirements for several coal compositions are
separated into the types of energy required, i.e., steam fuel oil,
natural gas, electricity, and coal losses. Nonregenerable FGD
processes impose the lowest energy requirements. The combination of
coal cleaning and nonregenerable FGD systems requires three times the
energy required by the FGD process alone. Transportation of low
sulfur western coal to the Midwest requires 25 to 100 percent more
energy than combusting a high sulfur-eastern coal and using a non-
regenerable FGD process. Figure B-2 shows the energy required
to meet a standard of 90% SC^ removal using different SO control
methods for several coal compositions. Figure B-3 shows energy
requirements for meeting a standard of 220 ng/J. The energy require-
ments for different control methods have the same relative variations
for the more stringent standards as those for meeting the existing
standard. Table B-4 summarizes the total SC>2 and particulate energy
requirements shown in Figures B-2 through B-4 for combusting a 3.5
percent sulfur coal.
Comparison of Energy Requirements for Alternative LeveIs of
the SOp NSPS. Figure B-4 shows the energy penalties associated
with two levels of S(>2 control: the existing standard, and the more
stringent standard calling for 90 percent SC^ removal. As shown in
Figure B-4, for most control methods the energy penalty for achieving
90 percent removal is about 10 percent higher than that required to
meet the existing NSPS.
B-6
-------
o
£ 40.^
1
O
ll
LfOCNO:
1 Oil. OH QAi
I
I AND
27.IMJ'>1 COAL
tT.l«U'«1 COAL
FIGURE B-2
0.8* SULFUH 0.«* tULFUM
24,»U4ffeg COAL afl.fttUUf COAL
ENERGY PENALTIES FOR S02 AND PARTICULATE CONTROL
90% S02 REMOVAL CONTROL LEVEL, 500 MW PLANT
B-7
-------
o
3
^^ COAL CLIANtNQ LO*9«
IHH 3TCAM AND IHCTRIClTI
0 0
2 2
So 3 o
g § 22
I 1 1 1
WM gm ** £
'i '*; " J
t 'f\ ;" S
?;' ^ J ^
|l ^
» « : 1
i
3
H.
z
i
:;
?'"
:.f
;'.<
1
§
i
o
o
u
o
u
i
kv;
*-:
vJ
t ^
'V
^
1
1
Ik
$
O
i
i
j
i
u
i
u
j
:/
-
;,
^'
^
i
f
?^
^
1
§
Hk
i
i
<
w
u
_i
o
u
1
*:'
:?
i;
1
1
''":
o a% SULFUR.
2*.0 Mjyh«
COAL
3.9% SULfUR
ZT.9 MJ/ht
COAL
r o» SULFUR.
zr.auj/kf
COAI.
FIGURE B-3
ENERGY REQUIREMENTS FOR S02 AND PARTICULATE
CONTROL - 220 ng/J CONTROL LEVEL 50Q MW
PLANT
BO
-O
-------
TABLE B-4
TOTAL S02 AND PARTICULATE ENERGY PENALTY ASSOCIATED
WITH DIFFERENT METHODS OF CONTROLLING S02 EMISSIONS - 500 MW
PLANT, 3.5% SULFUR COAL
S02 Control
Process
Energy Penalty (% Energy Input to
Equivalent Uncontrolled Power Plant)
0.5 Ib S02/
NSPS 90% MM Btu
Nonregenerable FGD
Limestone
Lime
Double-Alkali
3.4
3.0
3.0
3.8
3.4
3.0
NE
NE
NE
Regenerable FGD
Magnesia Slurry
Wellman-Lord/Allied
Coal Cleaning Plus
Lime/Limestone FGD
5.3
11.7
9.8
6.1
13.2
NE
NE
NE
9.8/10.2
NE = Not Examined
B-9
-------
CO
1
CU
C
2 CONTROL - SUMMARY
OF EFFECTS OF SO CONTROL LEVEL
B-10
-------
The results of the study also show that for combustion of low
sulfur western coal, 90 percent SO removal requires up to 10 percent
more energy than controlling emissions to 220 ng/J (0.5 Ib S0~/10
Btu) of heat input.
When flue gas reheat and particulate/chloride removal are
excluded from the limestone and lime systems, the energy requirement
of these systems is reduced by 50 to 60 percent. For the double-
alkali, magnesia slurry and Wellman-Lord/Allied processes, the
particulate/chloride removal operation is required to prevent buildup
of chlorides in the SO scrubbing liquor and to prevent contamination
of the chemical process by particulates. However, excluding flue gas
reheat requirements would reduce the energy required for operation of
the MgO system by 15 to 25 percent, the Wellman-Lord/Allied system by
about 10 percent, and the double-alkali system by about 50 percent.
B-ll
-------
APPENDIX C
REHEAT OF SCRUBBED FLUE GASES
A substantial amount of FGD energy is expended in the reheat
of flue gases. This appendix discusses the need for reheat systems,
and the problems and solutions associated with their usage. Mater-
ial in this appendix was extracted from "Flue Gas Desulfurization
Systems: Design and Operating Parameters, S0~ Removal Capabilities,
Coal Properties and Reheat" a draft prepared for the U.S. EPA, November,
1977.
Purpose and Need for Reheat
Flue gases are normally discharged to the stack at ~120° to
150°C (250° to 300°F). The temperature is selected to remain
above the dew point in order to reduce corrosion and permit carbon
steel to be used for fans, ducting, and stack lining.
When a wet scrubber is inserted between the air heater and stack
for SO,, removal, the flue gas exiting the scrubber is saturated with
water and cooled to the saturation temperature of about 50°C (125°F).
Discharge of the cool, wet gas to the stack can lead to:
Condensation of water vapor and sulfur oxides, resulting in
the acidic water corrosion of downstream ducts, fans, and
stack lining
Impaired plume rise and, hence, poorer dispersion of residual
pollutants for a given stack height
Deposition of scrubber residue on downstream fan blades,
resulting in imbalance
A visible plume as water vapor condenses
C-l
-------
Stack rain, or mist droplets, that can settle around the
power station.
To prevent corrosion of downstream components, the treated gas
may be reheated to a higher temperature before discharge.
Methods of Reheat
Flue gas can be reheated in many ways, and several approaches
have been developed. The basic differences in reheat methods are the
energy sources used and the methods of transferring that energy to the
flue gas. Reheat methods currently in use include:
Direct inline reheat - using steam or hot water heat exchangers
e Direct combustion reheat - using gas or oil in either inline
burners or external combustion chambers
Indirect hot air reheat - using steam to heat air which is
then mixed with the scrubbed gas
9 Bypass reheat - bypassing a portion of the untreated hot flue
gas to mix with the scrubbed gas.
In the U.S., the scrubbed gas is generally reheated by 15° to 40°C
(30° to 100°F). Except for bypass, the energy penalty for reheat may
range from 1 to 5 percent of the heat input to the boiler system.
Direct Inline Reheat. Inline steam reheat is the; most prevalent
method in the United States, although a few systems use hot water. An
inline reheater consists of a heat exchanger installed in the flue gas
duct, and is generally simple in design and installation. For a given
degree of reheat, the energy consumption is lower than for other types
of reheat, except bypass. The major problems encountered have been
plugging, corrosion, and vibration of the heat exchanger tube bundles.
C-2
-------
Plugging occurs from entrainment. Once the scrubber liquor
deposits on the reheater, the dissolved and suspended solids bake
onto it. This deposit continues to grow, blocking the gas flow,
increasing the pressure drop, and helping to induce corrosion from
the localized high temperature and concentrated salts. The effec-
tiveness of the heat transfer surface is progressively reduced as
the deposit builds up. To minimize problems, an efficient scrubber
mist eliminator is essential. Most direct inline reheaters incor-
porate a steam or air soot blower for maintaining clean reheat
surfaces. Corrosion and pitting have been attributed to periodic
acid conditions and to chlorides. Effective design of the mist
eliminator will minimize this. Operating experience suggests that
certain materials should not be used, such as carbon steel, 304SS,
316SS, and Corten. Structural design against vibration is essential.
Direct Combustion Reheat. The major advantage of this type
of reheat is operational reliability, especially if gas is used,
because there is no heat transfer surface on which fouling can
occur. The main drawback in the United States is the limited
availability and cost of the oil or gas required. Where oil is
used, problems have occurred in attempting to maintain the flame
within the main flue gas stream. This saves space, but typically
does not work well. An external combustion chamber of adequate size
is essential. Refractory failures have occurred in the combustion
chamber from flame impingement, attack from condensation during
downtime, vibration, and too rapid heating. Careful specification
C-3
-------
of refractory and subcomponents is essential and provision must be
made for the combustion chamber to be heated up slowly.
Indirect Hot Air Reheat. Indirect hot air reheaters have had
fewer problems than inline types. One advantage of this type of
reheat is that it provides desirable dilution for moisture content
and residual pollutant concentration in the stack gas. Disadvantages
are higher capital investment, higher steam consumption for the same
degree of reheat, and increased stack gas volume. The temperature
of the hot air before mixing is higher than 200°C (400°F) and can
destroy the usual coatings used to protect ducts and scrubber walls.
Hot air must be prevented from entering the system without cold gas
flow during startup and shutdown. For the same degree of reheat,
indirect hot air reheat has the highest energy requirement. However,
for the same amount of energy consumption, it may have equal or better
«
benefits than other types of reheat.
Bypass Reheat. Bypass reheat has the advantages of low capital
investment, negligible operating cost, and simple, reliable operation.
However, the maximum degree of reheat obtainable is limited by the
overall SO removal requirement versus the SO removal capability of
the FGD system. Separate fly ash removal is required to ensure that
the bypassed flue gas is sufficiently low in dust content to meet
particulate emission standards. Operating experience with bypass
reheat in the United States is limited to only one installation, but
additional FGD systems are under construction with provision for
bypass reheat.
C-4
-------
Alternatives to Reheat
A number of alternatives to reheating are available. Whether
or not they are feasible must be considered on a case by case basis.
To the extent that excessive ground concentration of residual pollu-
tants is a problem due to reduced plume buoyancy, one alternative
to reheating is to build a taller stack. Because there is no
energy penalty, the taller stack could be more economical than
reheating even though it involves a high capital cost. However, this
option is specifically limited by the Clean Air Act Amendments of
1977.
To limit corrosion, one may either select materials that are
inherently resistant to corrosion or use coatings to cover materials
subject to corrosion. If the purpose of reheat is to protect a down-
stream fan, an alternative is to place the fan upstream of the
scrubber, although this requires a precipitator to remove erosive
particulate matter. Wet, or washed, fans have also been used.
Using reheat to overcome the effect of liquid entrainment
from the scrubber is costly and not necessarily the best answer to
such problems. The use of more efficient mist eliminators would be
the preferred alternative. One form of liquid emission that would
not be affected by better mist elimination results from liquid
condensate forming inside the stack. Such an effect can be reduced
or eliminated by the combination of insulation on the stack so as to
reduce condensation likelihood, plus the use of a lower velocity
stack.
C-5
-------
There is very little, other than reheat, that is effective in
eliminating a visible vapor plume. However, plume appearance and,
in particular, the length of a visible plume are strong functions of
atmospheric conditions. When the gas leaves the stack, water vapor
condenses in the cooler atmosphere forming a visible plume. As the
plume disperses, condensed vapor evaporates at a rate depending on
the ambient humidity and temperature. With high external humidities
early in the day, the visible plume may travel long distances before
disappearing. In a desert environment at mid-day, on the other hand,
the visible plume vanishes rapidly. If the purpose of reheat is a
cosmetic one, one approach could be to use variable reheat. That
is, reheat could be limited to those periods of atmospheric conditions
during which a plume of objectionable length would otherwise occur.
Reheating should not be considered as a necessity, but as one
of a number of approaches for consideration in optimizing sulfur
dioxide absorption systems. Because of the high cost of reheat
installation and operation, some FGD system operators have selected
"no reheat," or "wet-stack," design.
C-6
-------
APPENDIX D
FGD SYSTEM PERFORMANCE
Flue gas desulfurization processes are categorized as regener-
able or nonregenerable depending on whether sulfur compounds are
separated from the absorbent as a by-product or disposed of as a
waste. Nonregenerable processes produce a sludge that requires
disposal in an environmentally sound manner. Regenerable processes
have additional steps to produce by-products such as liquid SC^,
sulfuric acid, or elemental sulfur. The nonregenerable group
includes lime and limestone, sodium carbonate and double alkali
scrubbing techniques. The regenerable systems currently in operation
are typified by the magnesium oxide and the Wellman-Lord systems.
The following sections briefly describe these processes, their
efficiency and reliability, and present information on their perfor-
mance at selected installations.
Lime and Limestone Scrubbing
Lime slurry scrubbing is a wet scrubbing process that uses a
lime slurry to react with S0~ in the flue gas. Lime is fed into the
system and combined with water to form a slurry, which is then contacted
with the flue gas to absorb SC^. Sulfur dioxide reacts with the
slurry to form calcium sulfite and sulfate, which are removed from
the system as sludge. The limestone slurry scrubbing process is
similar, although it uses limestone rather than lime as the reagent.
Facilities using lime and limestone systems have reported both long
D-l
-------
and short term SOo removal efficiencies in excess of 90 percent in
the United States. Both have successfully operated on high- and low-
sulfur coal-fired applications.
Many operating lime and limestone systems were designed for SCL
collection efficiencies of less than 90 percent, since this was all
that was required to meet an applicable regulation. Often an effi-
ciency in the range of 60 to 70 percent was sufficient, and such values
were used to establish system design.
Design of newer systems which are required to achieve high
efficiency must take into account a number of key design variables
including:
inlet 862 concentration
liquid to gas ratio
scrubber gas velocity
scrubber liquor inlet pH
type of absorber
Higher removal efficiencies can be more easily achieved at
lower SOo inlet concentrations because the amount of S02 that must be
absorbed per unit of scrubbing liquor to achieve a specified outlet
concentration is smaller. At low S02 concentrations the alkali in
the liquor can react with a greater percentage of the SO and affect
a greater removal efficiency under a given set of operating conditions.
Figure D-l shows removal efficiency vs SO,, inlet concentration for a
given set of conditions.
Higher efficiencies are realized at higher liquid to gas (L/G)
ratios for lime and limestone systems. For a given absorber, increased
D-2
-------
100
o
Z
UJ
o
IZ
LL
UJ
O
s
UJ
cc
CM
o
V)
K
u
o
cc
UJ
0.
O EPA PILOT TCA
SPHERE HEIGHT - 7 INCHES-BED, 3 BEDS
LIQUID TO GAS RATIO = 35 qal/Mcf
TCA GAS VELOCITY = 7 b ft/sec
<) TVA PILOT SPRAY TOWER
LIQUID TO GAS RATIO = 85 gal/Mcf
90
8b
0
80
70
LIMESTONE SCRUBBING
1.000 2.000 3,000
INLET SO2CONC., ppm
4,000
5,000
Source: Bechtel, 1977.
FIGURE D-l
EFFECT OF INLET S02 CONCENTRATION ON S02 REMOVAL
EFFICIENCY FOR FIXED DESIGN AND OPERATING CONDITIONS
Dr-3
-------
L/G ratios will yield higher efficiencies until flooding and poor gas
distribution occur. For new designs, absorbers which can accommodate
high L/G ratios can be selected and high efficiency maintained.
Higher liquid ratios also require larger pumps, pipes, and slurry
reaction tanks. Again, these can be designed into the system and
should cause no unusual operating problems. Figures D-2 and D-3 give
some typical data on L/G ratios vs collection efficiency.
The effects of changes in flue gas absorber velocity on S02
removal efficiency, when other variables are kept constant, vary with
the type of absorber. For a spray tower, the efficiency decreases at
a fixed L/G ratio. This effect is much less noticeable on packed
and turbulent contact type absorbers (TCA) (Figure D-4). For a new
plant, the scrubber would be designed for the required L/G when
considered along with other design parameters.
Increased efficiency is achieved at higher pH (Figure D-5) since
more alkali is available and higher dissolution rates are achieved.
Operation at very high pH, however, causes scaling problems. Mainten-
ance of the desired pH by careful measurement and close control of
reagent feed and mixing system will prevent the pH variations which
reduce efficiency (if too low) or cause scaling (if too high).
A large variety of absorber designs have been utilized to achieve
862 removal efficiencies as high as 99 percent. These include cross-
flow horizontal spray chambers (Weir), spray towers, packed-grid
towers, and turbulent contact (mobile bed) absorbers. The venturi
D-4
-------
99-
TCA4 STAGES
LIMESTONE
93
92-1
90-
iNl.n',s02- 200 PPM
GAS RAIL = 450,000 SCFM
20
40
60
LIQUID TO GAS RATIO, gal/Mcf
Source: Bechtel, 1977.
FIGURE D-2
EFFECT OF LIQUID-TO-GAS RATIO ON S02 REMOVAL EFFICIENCY
WITH LOWSULFUR COAL AT THE MOHAVE POWER STATION
D-5
-------
100
90 -
80 -
o
2:
UJ
u.
u.
u
o
ui
cc
CM
2
LU
o
K.
U,
a.
70 -
60
50 1
40 "
30
SCRUBBER INLET pH
© pH = 5.8 LONG-TERM TEST
O pH = 5.75.9 FACTORIAL TESTS
D pH = 5.4-5.6 FACTORIAL TESTS
A pH= 5.1-5.3 FACTORIAL TESTS
SCRUBBER GAS VELOCITY = 10.4 ft/s3C
TOTAL HEIGHT OF SPHERES = 15.0 in,
EFFECTIVE 1 IOUOR Mg*+ CONCENTRATION = 0 ppm
INLET SOo CONCENTRATION - 2/00-2,900 ppm
LIQUOR CJ~ CONCENTRATION = 3,000-7.000 ppm
20
30
40 50 60
LIQUID - TO - GAS RATIO, gal/Mcf
70
80
Source: Bechtel, 1977.
FIGURE D-3
EFFECT OF LIQUID-TO-GAS RATIO ON S02 REMOVAL
EFFICIENCY - TCA WITH LIMESTONE
D-6
-------
100 --
SLURRY FLOW RA1 E
@ 38 gal/min-ft2
O 38 gal/min-U 2
n 28 gal/min-ft"
^
A 19 gal/min-ft
LONG-TERM TESTS
FACTORIAL TESTS
FACTORIAL TESTS
FACTORIAL TESTS
<
>
o
LU
cc
CM
O
v>
>-
H
UJ
O
K
U
o.
90 -
I 80
I1!
u
u.
t'J
70 J-
60 -
a
A
FLOW RATE O = 38 gal/min-f
I i^J^J-»«»»^^^^
o
A 19 ga!,/mm-'t
-* ««iasMS»»*a«
O
O
50 -
40
TOTAL HEIGHT OF SPHERES = 15.0 in.
SCRUBBER INLET pH = 5.7-5.9
EFFECTIVE LIQUOR Mg"^ CONCErJTRATION - 0 porn
INLET SC2 CONCENTRATION - 2,000-3,000 ppm
LIQUOR C!~ CONCENTRATION * 2,000-6,000 pprn
8 9 10 11 12
SCRUBBER GAS VELOCITY, ft/sec
Source: Bechtel, 1977.
FIGURE D-4
EFFECT OF GAS VELOCITY ON S02 REMOVAL
EFFICIENCY- TCA WITH LIMESTONE
D-7
13
-------
100
o
z
ui
O
u.
UI
q
UJ
cc
co
H
M
O
CC
UJ
Q.
90 --
LIQUID-TO GASRATH-
FACrORIAL TES'IG
O CO gal/mcf
Q 45cj3i/tocf
A 30 gal/mcf
80
70
60
U
50
40 -
30
4.9
5.1
O
n a
a
SCRUB!,'.:\ GAS VELOCITY = 10.4 ft/sec
TOTAL HUGiiT OF SPHERES = 15.0 in.
EFFECTIVE LiaLIOil Mg++ CONCENTRATION = 0 ppm
INLET SO-7 COr'Cr.NTRATION = 2,300-2,700 pprn
LIQUOR CI" COiv'CENTRATION = 5,000-7,000 ppm
5.3 ' 5.5 5.7
SCRUBBER INLET pH
5.9
6.1
Source: Bechtel, 1977.
FIGURE D-5
EFFECT OF SCRUBBER INLET PH ON SO REMOVAL,
EFFICIENCY - TCA WITH LIMESTONE
D-8
-------
type has also been used, however, it is more useful as a particulate
removal scrubber an not as efficient for SO,, absorption ducts short
residence times (unless and additive such as MgO is used). The final
selection and design of an absorber are usually based on previous test
data and on the required liquid and gas flow rates. Spray towers
(either horizontal or vertical) offer a number of advantages including
simple internal design which decreases scaling potential, acceptance
of high liquid flows and decreased maintenance.
Variations in the alternative absorbers can also affect S0~
removal efficiency. For instance, Figure D-6 shows the effect of mobile
packing height on SC^ removal efficiency in a three-bed TCA. Increas-
ing bed heights by adding more spheres increases slurry holdup in the
scrubber. This has the combined beneficial effects of providing
additional liquor holdup time for alkali dissolution and a greater
gas-liquid contact area for improved efficiency. In the TCA, where
sphere retaining grids are about 4 feet apart, the static sphere
height per bed can be adjusted over a range of several inches. If
greater removal is desired, the number of beds can be increased.
Increasing the bed height, however, increases the gas pressure drop.
Other improvements in efficiency are achievable by increasing
the packing height in fixed packing towers and by adding more trays
in tray towers, although these types of scrubbers are not often used
in slurry service because of their susceptibility to plugging. These
improvements are made at the expense of increased gas pressure drop
across the scrubber and, therefore, increased fan power requirements.
D-9
-------
SLURRY FLOW RATE -
FACTORIAL TESTS
38 gal/min-ft2
100
40 4
-SS9
SCRUBBER GAS VELOCITY = 10.4 ft/sec
SCRUBBER INLET pH = 5.8
EFFECTIVE LIQUOR Mg++ CONCENTRATION = 0 ppm
INLET S02 CONCENTRATION = 2,300-2,700 ppm
LIQUOR Cl~ CONCENTRATION = 4,000-9,000 ppm
0 6 12 18 24
TOTAL HEIGHT OF SPHERES, inches
Source: Bechtel, 1977.
FIGURE D-6
EFFECT OF BED HEIGHT ON SO REMOVAL
EFFICIENCY - TCA WITH LIMESTONE
D-10
30
-------
The addition of relatively small amounts of magnesium compounds
(less than 1 percent by weight) to the scrubber liquor in the form
of magnesium oxide, magnesium sulfate, or dolomitic lime (in lime
systems) can greatly increase the S02 collection efficiency of the
system. Magnesium compounds are much more soluble, compared to
calcium, and can react rapidly in the liquid phase with SO^. Figures
D-7 through D-9 show the effect of magnesium, L/G and pH on SO^
removal.
As another means of achieving higher overall experiences, it
is possible to convert two scrubbers in series. The net overall
efficiency is substantially higher than that of either stage.
Figure D-10 shows the variations in two-stage absorption efficiency
as a function of individual scrubber efficiency. In this case the
first scrubber is a relatively inefficient (for lime or limestone
without additives) scrubber, such as a venturi. When the venturi and
spray tower have been operated together at the Shawnee Test Facility,
SC>2 removals greater than 90 percent have been realized during short-
term tests.
However, when two scrubbers are operated in series, the mechanical
complexity increases and additional energy penalties are incurred
for the greater pumping requirements and gas pressure drop. Japanese
practice with oil combustion and recent U.S. developments in coal
combustion indicate that the extra capital cost of two-stage scrubbing
may be offset somewhat by the use of forced oxidation techniques to
D-ll
-------
100 --
o
2
O
H
u.
Ill
O
in
o
IU
O
IX
HI
Q_
EFFECTIVE LlQUOR Mg"r CONCENTRATION
FACTORIAL TCSTS
O 7,000-10,000 ppm
D 3,500-5,500 ppm
A 0-500
90 -
GO -
60 -
A
50
40
SCRUDBER GAS VELOCITY = 10.4 h/:.ec
TOTAL HEIGHT OF SPHERES = 15.0 in.
SCRUBBER INLET pH = 5.4- 5 6
INLET SO-, CONCENT RATION = 2,200-2,800 ppm
LIQUOR cY~ CONCENTRATION = 6,000-16,000 ppm
20
30
40 50 60
LIQUID - TO - GAS RATIO, gal/Mcf
70
80
Source: Bechtel, 1977.
FIGURE D-7
EFFECT OF LIQUID-TO-GAS RATIO ON S02 REMOVAL
EFFICIENCY - TCA WITH LIMESTONE AND MAGNESIUM
D-12
-------
<
>
Q
90
20
TOO -~
1
SO -
50 4-
EFFECTIVE LIQUOR Mg^ CONCENTRATION
. FACTORIAL TESTS
O 7,000-10.000 parr,
a 3,500-5,500 ppm
A 0-600 ppm
5.0
o£
**
cC
.^.
^
Mfc^ *J
,o^o'
SCRUBBER GAS VELOCITY = 10.4 ft/sac
LIQUID - TO - GAS RATIO - 45 gal/Wcf
.TOTAL HEIGHT OF SPHERES = 15.0 in.
iNLET S02 CONCENTRATION = 2,300-2,700 ppm
LIQUOR Cl~ CONCENTRATION = 12,000-16,000 ppm
5.2
Source: Bechtel, 1977
5.4 5.6
SCRUBBER INLET pH
FIGURE D-8
5.8
6.0
EFFECT OF SCRUBBER INLET PH ON S0? REMOVAL
EFFICIENCY - TCA WITH LIMESTONE AND MAGNESIUM
D-13
-------
100
o
2
LU
O
O
2
LU
O
cc
UJ
O.
90 --
80 -
70 -
SO -
50 -
SCRUBBER INLET pH
FACTORIAL TESTS
O pH = 5.7-6.0
D pH=5.4-5.5
A pH =5.1-5.3
40 --
30
SCRUBBER GAS VELOCITY = 10.4 ft/sec
LIQUID TO - GAS RATIO = 45 gal/Mcf
TOTAL HEIGHT OF SPHERES = 0 in.
INLET S02 CONCENTRATION = 2,300-2,700 ppm
LIQUOR Cl~ CONCENTRATION = 5,000-14,000 pprr
1 j (
2,000 4,000 . 5,000 8,000 10,000
EFFECTIVE LIQUOR MAGNESIUM CONCENTRATION, ppm
Source: Bechtel, 1977.
FIGURE D-9
EFFECT OF MAGNESIUM ON S0? REMOVAL EFFICIENCY
TCA (NO SPHERES)"WITH LIMESTONE
D-14
-------
100
95
H
UJ
o
iZ
u_
(.'I
O
LU
£
CM
O
LU
O
£ 90
a.
85
30
40 50
PERCENT S02 REMOVAL FOR FIRST SCRUBBER
60
Source: Bechtel, 1977.
FIGURE D-10
S0? ABSORPTION EFFICIENCY FOR TWO SCRUBBERS IN SERIES
D-15
-------
bring down the high costs of sludge disposal and improve alkali
utilization. In lime and limestone scrubbing, the waste product is
normally a slurry of calcium sulfite, calcium sulfate (gypsum) and
fly ash (if removed by the scrubber). The calcium sulfite can be
oxidized to gypsum by air-slurry contact (forced oxidation). The
resultant product has improved properties including higher settling
rates, improved dewatering characteristics, and reduced total waste
volume. One successful approach is to oxidize the slurry in the
first of two scrubbing stages. The cost offset of two-stage forced
oxidation does not apply to coals of lower sulfur content where
oxidation and settled sludge density are normally high. The rela-
tionship between sulfur content, degree of forced oxidation, impounded
sludge behavior, and cost is not yet fully quantified.
Facilities at which high removal efficiencies have been obtained
are briefly described below:
(1) The Mohave Station of the Southern California Edison Company,
reported SC>2 removal efficiencies of 95 percent or more with
limestone, and of 98 percent with lime. The tests were con-
ducted intermittently over 1-year on low-sulfur coal. The
unit was a 170-MW equivalent, prototype scrubber.
(2) The packed module on the 115-MW Unit No. 1 at the Cholla
Station of Arizona Public Service shows 92-percent removal
of S(>2 using limestone slurry scrubbing. This is also a
low-sulfur-coal application (0.8%).
(3) Recent tests at the Paddy's Run Station of Louisville Gas
and Electric have shown SC^ removal efficiencies in excess
of 99 percent on 3-percent-sulfur coal. This extremely
high removal efficiency was due to the addition of magnesium
oxide to the lime slurry.
D-16
-------
(4) Several tests were conducted at the 10-MW TVA Shawnee Pilot
Plant, where SO removal efficiencies of 95 to 99 percent
were reported for lime-based systems, and of more than 90
percent for limestone systems. During one test run an
efficiency of 96 percent on a turbulent contact absorber
(TCA) unit, high-sulfur coal application, was achieved for
the limestone system.
A brief summary of three lime-based systems follows: 1) the
Green River facility of Kentucky Utilities, 2) the Bruce Mansfield
Station of Pennsylvania Power Company, and 3) the Mohave Station of
Southern California Edison. Two limestone slurry systems are also
discussed: l) the LaCygne Station of Kansas City Power and Light,
and 2) Sherburne No. 1 and 2 of Northern States Power Company.
Kentucky Utilities, Green River No. 1, 2, and 3
The FGD system is installed on three boilers which generate an
equivalent of 64 MW and burn coal with a sulfur content of 3.8
percent. This system is designed to remove 80 percent of the SO^ in
a turbulent contact scrubber and 99 percent of particulates. The
unit started up in September 1975, and commercial operation began in
the late fall of 1975. Before commercial service, the system went
through an extensive four-phase, prestartup evaluation.
Sulfur dioxide removal efficiency has been well above the design
value, averaging about 90 percent. After commercial startup, several
relatively minor problems were encountered and corrected. Closed-loop,
full-capacity operation began March 1976, with the initiation of a
6-month vendor qualification test. To date, performance of the
n-17
-------
system has been good; and mechanical reliability is excellent.
Average system operability has been above 90 percent since March
1976, with the exception of a period between February and April 1977,
when the unit was shut down for stack repair.
Pennsylvania Power Company, Bruce Mansfield No. 1
This two stage venturi FGD system is installed on Unit No. 1,
which is rated at 839 MW and burns coal with a sulfur content between
4.5 and 5.0 percent. The FGD system was designed for 92-percent S09
removal and 99.8-percent particulate removal. Unit No. 1 started up
in April 1976, and full commercial operation began in May 1976.
Availability was reportedly very high during the first 7 months
after startup; operating problems were solved without causing boiler
downtime. Since then, however, the unit has experienced serious
problems with the stack liner, and the load must be reduced by
approximately 50 percent for about a year for liner repairs.
Two performance tests were conducted in July 1977. The results
were 190 and 540 ng/J (0.44 and 1.26 Ibs SO /10 Btu) , representing
94-percent and 83-percent removal, respectively. The allowable
emission rate is 300 ng/J (0.6 Ibs S02/106 Btu). The variations in
emissions were apparently due to pH fluctuations which have since been
corrected.
Southern California Edison, Mohave Station
Participants in the Navaho/Mohave Power Project funded a full-
scale scrubber demonstration at the Mohave Generating Station. The
D-18
-------
170-MW demonstration facility was installed on a 790-MW boiler firing
coal with an average sulfur content of 0.4 percent. Two types of
scrubbers were installed for the demonstration tests: a horizontal
cross-flow scrubber and a vertical countercurrent unit. The vertical
module was operated both in a TCA and in a packed grid configuration.
Sulfur dioxide removal efficiency was excellent for all three absorbers.
Although the SO. inlet concentration was only 200 ppm, all three con-
figurations were capable of removing 95 percent of the inlet SO .
Calculated availability percentages for the horizontal and vertical
modules were 81.3 and 72.8 percent, respectively. Since this was a
test facility, several design changes that contributed to low avail-
ability were made during the period.
Kansas City Power and Light Company, LaCygne No. 1
The unit is rated at 820 MW and burns coal with sulfur content
ranging from 5 to 6 percent. The FGD system installed in 1972 con-
sists of eight identical scrubbing modules, each with a venturi
scrubber for particulate emission control and an absorber for SO,,
control. Particulate removal efficiency is from 97 to 99 percent.
The system was designed for 76-percent S0~ removal. Actual S0~
removal efficiency is 80 percent with seven modules operating on
729 MW. Under maximum load, the removal efficiency averaged 76.2
percent. Efficiencies under both conditions should improve now that
eight modules are operating.
I)-19
-------
The FGD installation was plagued with startup problems. However,
analysis reveals that nearly all of them were due to mechanical design
rather than to process chemistry limitations. The availability of
this system has improved steadily as solutions to the various problems
have been found. The system is now one of the most reliable FGD
systems on a large boiler in the United States. The availability for
1976 averaged 91 percent, and for the first half of 1977 averaged
about 93 percent.
Northern States Power Company, Sherburne Station No. 1 and No. 2
Each unit has a net generating capability of 700 MW and fires
a subbituminous western coal with a 28-percent moisture, 9-percent
ash, and 0.8-percent sulfur content. Each system has 12 scrubber
modules, 11 of which are required for full-load operation. Sulfur
dioxide removal is between 50 and 55 percent, which is sufficient
to meet local requirements and approximates the value for which the
system was designed. Availability for Unit No. 1, which started
up in March 1976, averaged 85 percent for the first 4 months of
operation. During the past 12 months, availability has been in
excess of 90 percent. Unit No. 2 started up in April 1977 and has
shown even better startup performance. Availabilities have averaged
about 95 percent for the first 4 months.
Wellman-Lord Process
The Wellman-Lord Process uses an aqueous sodium sulfite solution
to absorb S0« and form sodium bisulfite. The solution is regenerated
D-20
-------
and S0~ is released in an evaporator-crystallizer. The regenerated
sodium sulfite is dissolved for recycle in the absorber. The concen-
trated S07 stream is recovered as liquid S0~, sulfuric acid, or ele-
mental sulfur. Guidelines to obtain high efficiency for Wellman-Lord
Systems include:
- Installation of a prescrubber with a separate water recirculation
system for final particulate control and reduction of SO, and
chlorides.
o
- Use of a three to five tray absorber with an L/G of 1.0 to 1.3 1/m
(6 to 10 gal/1000 acf).
- A superficial gas velocity in the range of 2.7 to 3.1 m/sec (9 to
10 ft/sec).
- Maintenance of the required sodium sulfite scrubbing solution at
a pH of 6.0 at the absorber inlet.
- System make-up of fresh, 20 percent sodium carbonate solution
should be approximately 0.07 1/m (0.5 gallon/1000 acf) per tray.
- As 862 inlet concentration decreases, the number of trays
required to obtain high S02 removal should be increased.
Seven Wellman-Lord systems are operating in the United States.
Six units are installed on S02 or Glaus sulfur recovery plants. The
S0~ removal efficiency of these six is typically 90 percent or greater,
and removal efficiencies in excess of 97 percent have been reported.
On-stream time for the absorption area of these plants is more than
97 percent.
The No. 11 unit at the D. H. Mitchell Generating Station at
Northern Indiana Public Service Company (NIPSCO) is currently the
D--21
-------
only operational Wellman-Lord system on a utility boiler in the
United States. It is also the only coal-fired application in the
world. The process is designed to remove at least 90 percent of the
S0~ when firing coal containing up to 3.5 percent sulfur. The
supplier guarantees the mechanical soundness and product: quality of
the process, as well as water, electricity, and chemical consumption.
The initial startup of the NIPSCO unit began July 19, 1976, and an
extended shake-down period began November 28, 1976. During this per-
iod, the Unit 11 boiler operated for 121 full days and 10 partial
days, whereas the S0~ removal system operated for 71 full days and
23 partial days, and was down for 38 days. In the course of the three
sustained operating periods, the absorber demonstrated the capability
of greater SO^ removal than specified. A boiler-related mishap
occurred January 15, 1977, causing the unit to be shut down for
repairs until May 1977. The absorber resumed operation June 13, 1977.
Operation has been erratic since then, again primarily because of
boiler problems. Trials began on August 29, 1977 and were success-
fully completed on September 15, 1977. The equipment met the guar-
antee covering S0_ and particulate removal, chemical meikeup, and
utility usage.
Three Wellman-Lord systems are currently under construction. Two
of these will be on coal-fired boilers at the San Juan Station of the
Public Service Company of New Mexico. Each unit will be on a coal-
fired boiler with approximately a 350-MW rating. Both units are
D-22
-------
designed for 90-percent removal of SCL. The third unit is at ARCO/
Polymers in Monaca, Pennsylvania, where a single scrubber will receive
flue gases from three coal-fired boilers with a total equivalent
rating of 100 MW. The unit is designed for approximately 87.5-percent
SOo removal.
Magnesium Oxide Systems
This process uses a magnesium oxide slurry to react with SO,,.
The reaction product, magnesium sulfite, is dried and calcined to
regenerate magnesium oxide. Sulfur dioxide, liberated in the regen-
eration step, is recovered for conversion to sulfuric acid or for
reduction to elemental sulfur. Guidelines to achieve high efficiency
include:
- High efficiency particulate removal should precede the absorber^
- A prescrubber should be used to remove any remaining particu-
lates and most of the chlorides and SO-,.
- Venturi absorbers should be utilized typically operating at
a pressure drop of 25 cm (10 inches) of water or greater, or
Turbulent Contact Absorbers operating at approximately 20 cm
(8 inches) of water pressure drop, at an L/G of 5.3 to 6.6 1/m-^
(40 to 50 gal/100 acf).
- The absorber superficial gas velocity should not exceed approxi-
mately 3.0 m/sec (10 ft/sec) range.
- The slurry pH measured at the absorber discharge should be main-
tained in the 6.0 to 7.5 range.
Three full-scale units have been operated in the United States:
1) Mystic Station, Unit No. 6, of Boston Edison (oil-fired); 2)
Dickerson No. 3, of Potomac Electric and Power (coal-fired); and 3)
D-23
-------
Eddystone No. 1A, of Philadelphia Electric (both coal-fired). All
used fuel with 2 to 2.5-percent sulfur content. Sulfur dioxide
removal efficiencies at all three locations have been in excess of 90
percent. However, in general the three units experienced serious
problems which have limited operability to between 27 and 80 percent.
When reviewing these operability levels, however, several points
must be kept in mind:
Sulfur dioxide collection efficiencies were frequently over
90 percent during test periods.
Two units (Mystic and Dickerson.) were trial installations,
built to obtain operating data. As such, various construc-
tion materials were used that would not have been used in
a full-scale plant designed for long-term operation.
The sulfuric acid plant that was to receive SO^ from the
Eddystone MgO regeneration facility was shut down by its
owner and another had to be found.
The single regeneration facility at Rumford, Rhode Island,
could not process material from the Mystic and Dickerson
stations simultaneously because it was too small.
Many problems at the Eddystone installation are related to
particulate scrubbing and not to the SCU absorber section.
Many design and operating problems at these installations
were solved during these early programs and would not be
encountered in new designs.
In the past, the MgO systems installed by Chemico (Mystic and
Dickerson) and United Engineers (Eddystone) have not had overall
performance guarantees. Rather, the manufacturers of certain com-
ponents guaranteed them against manufacturing defects only. Now,
however, Chemico is willing to guarantee the entire MgO system
D-24
-------
mechanically, as well as specifying that the unit will meet applicable
SO,, emission regulations, including a 90-percent removal efficiency.
Double Alkali Flue Gas Desulfurization Systems
Double alkali scrubbing is an indirect lime/limestone process,
in which a soluble alkaline medium is used in the scrubbing vessel
to react with SO-- The scrubber effluent is then treated with lime
or limestone in a reactor outside the scrubber loop, where calcium
sulfites and sulfates are precipitated and the scrubbing liquor
regenerated and returned to the scrubber. This system greatly
reduces the problems of plugging and scaling. Various double alkali
process configurations are available and are described in the full
report. Guidelines to achieve high efficiency include:
- Utilization of a prescrubber with a separate water recirculat-
ing system for control of particulates and chlorides for high
chloride coal (>0,04 percent Cl by weight in the coal).
- Use of a two-stage tray or packed tower absorber with an L/G
in the 1.3 to 2.7 1/m3 (10 to 20 gal/1000 acf). Typically the
absorber pressure drop is 15 to 30 cm (6 to 12 inches) of
water.
- The absorber scrubbing liquor pH being recycled to the absorber
should be in the range 6.0 to 7.0 pH range.
- If lime regeneration is used, the reaction tank residence time
should be approximately 10 minutes.
- If limestone regeneration is used, the reactor tank residence
time should be approximately 30 minutes.
A number of successful bench-scale, pilot plant, and prototype
double alkali systems have been tested on both industrial and utility
boiler flue gas applications in the United States. The success of
D-25
-------
these programs has resulted in commitments by three separate utilities
to install full-scale, double alkali systems on coal-fired boilers.
As yet, however, no full-scale system is operating on utility boilers,
although several are in operation on coal-fired industrial boilers.
At the Cane Run No. 6 unit of Louisville Gas and Electric, a
277-MW coal-fired unit, the double alkali system is scheduled to start
up in February 1979. The unit is designed to have 200 ppm of SO or
less in the discharge from the scrubber, and 95-percent SO removal
2
when the sulfur content of the coal is 5 percent or greater. Coal
sulfur content is expected to be between 3.5 and 4 percent.
At the A. B. Brown No. 1 installation of Southern Indiana Gas
and Electric, the double alkali system will be applied to a 250-MW
boiler firing coal with an average sulfur content of 3.5 percent.
The unit is scheduled for startup in April 1979, and designed to
remove 85 percent of the SO,, when burning 4.5-percent sulfur coal,
the maximum sulfur content expected.
At the Newton No. 1 unit of Central Illinois Public Service, the
double alkali system will be installed on a 575-MW boiler firing coal
with an average sulfur content of 4 percent. The unit will start up
in November 1979. The design SO removal efficiency is 95 percent,
or less than 200 ppm in the exit gas.
Four double alkali systems have been installed on industrial
coal-fired boilers. These systems have operated with high removal
efficiency, ranging from 85 to 99 percent (mostly 90 to 95 percent).
D-26
-------
While some, have had mechanical problems, the systems have shown them-
selves reliable; generally operability has been over 90 percent. In
addition, two prototype double alkali systems were operated on utility
coal-fired boilers, one on low-sulfur coal and the other on high-
sulfur coal. Both had SC^ removal efficiencies above 90 percent, and
their success has resulted in the design of a full-scale system that
is expected to have high levels of operability and efficiency.
FGD System Efficiency Summary
Table D-l identifies facilities at which FGD systems have
removed 90 percent or more of S0«. In addition, many of the systems
listed in Appendix E are being designed for efficiencies of 90
percent or greater.
The major suppliers of systems are now offering S02 removal
guarantees. Levels of SOo removal which vendors will guarantee
exceed 90 percent, and in some cases 95 percent, often have a
lower limit on outlet SO* concentration (e.g., 50 ppm). For
lower sulfur coals, this lower limit, rather than efficiency, would
become the basis of the guarantee. Thus existing technology is
adequate for meeting a 90-percent S02 removal requirement.
D-27
-------
«
w
t-1
3
rH
CN IT)
O *> d*5
Ul O
0)
IH
rH
41
W 3
dp
C
H
(A
in
01
0
0
rH
0.
4)
VH
3
4->
ro
2
II
s
£
4J
C
3
c
0
rH
4-1
IT)
4J
cn
>,
c
13
cu
E
O
CJ
>^
4J
H
! --1
-H
4-"
I ^
tN
OS
O
rH
1
O
01
c
o
4,1
in
41
E
«*t
^
01
rH
ffj
U
in
rH
rH
3
&4
in
rH
rH
rH
O
2
IT)
rH
rH
O
JC
u
41
O
rH
^
IH
01
c/]
O
H
rH
£t
3
0,
IT)
C
O
N
H
kH
<
m
Os
t
0
41
E
rH
^
C
o
H
4-1
(3
4J
in
C
0
6
4)
Q
0
1X3
rH
m
0
Z
en
k.
41
C
kH
O
u
3
O
rV,
+
o
J
4'
rH
n)
0
in
i
r-l
rH
2
fc
m
r-
,-H
-T
0
z
c
3
pi
4)
C
U
o
rH
IH
4J
0
4)
rH
W
4
in
n)
o
4)
rH
r-H
>
Ul
rH
T
O
J
in
4
CM
OS
0
in
4
n
HI
e
H
^
c
0
H
4-1
<0
kH
4-1
Ul
C
O
e
01
o
in
U5
SO
O
2
C
3
fX
in
-
^,
a
a
rO
&,
o
OS
in
m
CN
.
m
a
VH
o
,-j
[
C
(TJ
c;
rH
41
^
C
o
.rH
4-1
tj
kH
U
cn
c
O
E
41
Q
m
rH
r-H
r-H
rH
O
2
rH
r-H
CD
X
U
rH
^
CJ
H
rH
13
3
a,
ITJ
c
13
H
a
M
*< dl
1) O
.C H
4J >
0 4)
2 '/)
00
OS
1
in
in
CN
X
O
1
CP
ITJ
E
C
o
H
4J
rfj
k^
4J
in
0
E
O
o
o
CN
*~*
r-H
0
01
c
o
4J
'fl
-a
D
CU
o
H
rH
4J
O
41
rH
W
10
H
r",
D,
rH
T)
rH
H
-C
a.
as
ON
1
in
ON
o
\
CO
.
o
41
C
O
AJ
Ul
41
g
H
rH
V
4)
rH
"-*
4)
a
4J
O
4J
O
kH
a.
o
rH
0
r".
O
2
CU
o
c
3
10
x:
co
4-1
H
IH
0
4-1
3
"3*
>^
d'
rH
r-H
rrj
^>
CU
01
n
U!
4)
C
c
4i
EH
O
as
m
CN
X
O
1
ty
rrt
*
C
0
H
4J
3
4J
cn
c
0
E
CU
D
o
l/l
rH
«>
0
2
O
H
4J
Ul
K
C
o
in
H
a
u
c
4>
i/l
Q
i-H
OS
1
0
n
o
4)
C
O
4J
in
E
H
"^
C
o
H
4-1
rt
kH
4J
W
C
o
E
a
ro
VO
rH
UD
O
2
kH
H
(3
r-H
U
4J
CO
£4
O
Ul
T3
Cli
4J
H
O
rH
4J
Q
Q
+
O
H
O
4->
O
E
rH
10
41
C
01
o
in
cn
o
m
H
rH
10
y
rH
10
1
0)
rH
§
O
Q
01
a
^J
o
4J
0
CU
0
CN
rH
o
2
r)
rH
O
U
CO
kH
41
8
a
!«-«
r-H
3
O
in
cn
0
ro
13
-a
o
_r;
U
41
a
.p
o
4J
O
&.
o
IN
CN
1
r-H
O
2
N
rH
O
O
co
H*
O
o
m
c
0
k-l
(3
CJ
4)
a,
^j
o
4J
O
kH
Cu
0
CN1
CN
O
2
N
r-H
O
£.
U
C/l
o
OS
0
IN
X
G
1
CP
(3
C
O
4J
D
4-1
U)
c
g
S
Q
in
Os
.T!
o
2
C
O
Ul
k^
4)
O
H
Q
rH
Ol
3
O
o.
U)
O
H
kH
4J
0
41
r-H
W
U
T3
f=
O
4J
o
a,
in
CTS
ID
O
4>
O
4-1
in
a)
rH
^
c
o
rH
4->
13
^J
Ul
C
O
4)
Q
O
r-
'~
rH
G
2
0)
X
o
o
cn
TD
W
ra
H
C
kH
O
H
rH
a
CJ
C
CU
4J
3
0
CO
m
OS
o
0)
H
J
c
o
H
4J
10
kH
4J
in
c
O
E
0)
D
o
(^
rH
r-4
O
2
4)
p.
13
x:
o
OS
OS
r-i
4)
e
H
1-1
4C
*
01
(0
o
in
i
rH
rH
3
fc,
_>
CN
Os
1
rH
O
IH
41
'6
13
C
01
cj
K
4)
CJ
IH
O
kH
f^
W
0
H-
O
ffs
CO
m
41
H
^
O
rH
O
Ul
1
rH
r-H
3
b.
rj.
'£)
r-i
1
r-H
O
2
IH
41
p»
rH
p£
C
4)
41
v~<
0
cn
41
-rH
4J
rH
rH
H
4J
^5
A:
u
3
4J
C
o;
X
41
in
(TJ
rH CO
fl T3
H >, CD
IH !H IH
4J 13 -,
Cfl 4J [t.
3 -r(
fl rH rH
C -H -H
rH E O
o
p
w
PL,
Q)
U
O
C'3
QJ
i-H
Cj
K
ct)
n
a)
G
H
G
O
OJ
CO
tn
H
CO
O
O
CO
CD
4-1
O
O-
01
w
H
O
D-28
-------
APPENDIX E
PLANNED AND OPERATING FGD SYSTEMS
This appendix gives a list of planned and operating FGD systems.
The data were obtained from the PEDCo FGD Status Reports.
E-l
-------
fri
X
ii
Q
PH
a.
a o j:
-j 3 -_.
o -n
i £
tt T3
u ^ U b
C U (80
"^ *w V £ "" "* w "« ~* * *
cere; r *- r c: 3
c oc^ooooowu
Q] K c a
: ci. ft. o o
» O C U ooooooo
O O
i O O C> O (
1 O O O O O O
O U O -5 O
3 ? :*
6C CC t* CO tO t>0
C C C C C
-2 ,n
-o x
~ ~ OC Ofi
M M tsl
<, < < :: 5,
£ 2;
< <
. . j j u - - o ut-:
COGCJ-W^J::
j; u, E, b a. ;~ f
^^t^^r^-t-oc.
ffiO»DOOjOQ
n. o. a. a.
,"*.s
C C^
-
XI J^ C C
-'CM 7- OOOOO D
r. y, ^ K '^ -* c - -
o^v-oo° -jrcooo
n j ^, .- o x. .: V- -*- -* -x ^ c oj y. ^ *. x
"oj^-'C;: ci-S-ja:vict:G.a.Q.ci.
,- i' L' ~ "" i .x ci L. C' u iJ -' (C d P u LI u
j a, f» z x .-= "- >~> -: ^ ^ u ^ u j ;: ;; i: d4 u 5: u
S£
^ ^
ex. r-
o o
Si
^ ^
«V -v
-o j
7r -L
c
r
<^
ji
0.
r
%
^
r5
u d. r
TOO
-. r, a,
-. y c t- o
v. L ii. f- : . ,._ D f" C, o /i LI " IA J-
a t.' * o ;', ;" t1, "M L "' "' !" ' ' 'J u * i IT o «
o , , i. u u u :' ^ ' [: : ,...:r ; > -;v:-
H ^ -' W *. *J * ,,,-.-. ^
i: i ^ io ,-u ^ m M; ' ~ ; :- ^ 1 : L ^ o u ^
-* 0 0 I t- 1 f, ' L L t!
> <1 U *j U C C tl t. CJ r_J C' O CJ C_J
, r e^:^^::;^Iiui;H
>! .. ,. , ; 1 i. u ,. ~ -'-"'" "J "' " ^' "' u
J .- . _ ... 0. ^ fc ^J 0 ^ 0 .« - m «j « » JJ w
; - - - -z ': £ :: ' '. < ^ ^ .i ,d: ?. s. ~
;,,,.'.:.'..'... u" * : .7 IJ , - .'i ,'' I1 ." J: J
>
.J
>-,
ej
n.
g
u
(«0
?
t
C)
,* ^
.
. ' n
f! ft
u u
"? ^ ^
" nU n6
-.tar.a
ncana
. X X
X K
C G
o. n.
U (J
t-i U
8 I
?3
n «
*J 4J
ill
E-2
-------
~ ftj i> ' - o5 '
_j _J .0 <' -J '
13 -V
J} ,n O
. « r. "i UUlZt'Jl-i*-"-1'1-'^-'
Ul -*J TD TJ '" M "1 <0 ^ TJ *" " l'
&. w
r > CL,
b
j CO CO CO O) 7\ .
* ^i ^j O O O O <
i O O ^) O O
V V W OJ
O i O C O
O O 0 C C C r.
mo i .o J3
U l-i J3 J3 ^
v>u oo sc oo oo cfl -C yj cc '-j J j f &)
C C C '-- C T: a r: C z '- V' c C
TJllj ^-(^J~-,-M--VVjI-..-.Jfl.--i.-l r- -i _< _-J _,
C*; Cfl 'J XI -O _> -O »J ,C I _: .'i J^ J-- -O _2 -O w'J _T J'l
^ OJ ^ -D .O -' -^ -1 ^ -^ -^ -^ -9i -^ -5 -"", ^ -C
w '/i o
- - o - r. i: s
< < Q ^ >" ?: r: " >, >> -sj N: ro yi C
*j nr, ^ >>
o o
K; r. :
c c
IJ U
c/; o/j on o~,
r r; r £ a,
c c: c r:
: I
X X > X >..',>,
e r c r r £ r
« « CJ » tr; TO -I
cv tx o- Ci, (i a. r.
E U C R i h h
O C o o c o c
o u o u u u a
X 4> L " «
*' > 5 < <
^ r*
woo
, ^ A. a.
0.' 'J u v- ?-, ^.
tL (x i r. " "
u « e ,1
»- Q. L , |
g e
E-3
-------
APPENDIX F
CONSTRUCTION SCHEDULE
This appendix presents two hypothetical power plant construction
schedules, one with FGD and one without FGD. While this comparison
shows 6 months' difference in a 3-year schedule (excluding preliminary
design about 18 to 24 months) the effect can be varied by increasing or
decreasing the construction force size.
F-l
-------
o
o
UT
«t
-D
'O
E
eC
^
U_
r~^
Q
z:
o
L/">
=r
^D
^)
s:
^O
S
«£
«:
LJ_
^_>
O
z.
_J » , CL. "
=£ h-
^-i «r c£
I o; UJ
= S o
i-. O 0- "
I
1
I
! i_L
is
i
!_
mi
^
i i r,"-j
Grading and roads
Buildings and utilities
Concrete plant
r-* OJ CO
Construction Plant
r_p_. I
i -
JL
ttl
*l~
sa-
ri
(.
s,
4-
t
c
r-
^
C
CO r
-o T
2 .<
cr> i
t
-o _i
ti ^
(3 C
D-. '
c :
ro <
QJ C
c3 c
C" u
fO
s~
QJ
C
O
c;
-a
c
fO
I/I
*O
^3
I
)
3
J
J
J
U
U
5
0
c
J
3
y
0
y
j
T
U3
Coal handling facility erection
U3
Cl
c
r
-o
c:
fT3
3i
ID
O
0
r^
£
C
4-
(.
C
s
c
c
£
-I
^
C
L
C
j;
i
c
^
c
r-
c
£
-i
Q
C
i
C
J
s
c
J
J
u
u
J^
3
3
D
U
0
3
D
J
s
J
r
3
3
n
3
5
D
E
D
u
I
i
I
l
I
' -HHM
rt^ r 2 ri
1 < I
8-1 1
t_ |_L
II
I i
LL i !_
f" i CM H
CJ ^ -|
1
I
2
c-
rH
VJ
s_
o
4J
fO
S- ^
GJ O-
£ §
en o.
O
n to
s_ -^
3 r- C
+-> 03 ^
(^ -*-1
T3 O
EC.
ro in i/l
.,_ tn S- cr
E 73 _^ OJ C
ra tn u 4-> -1-
C -C C ro f3 D-
4-> (/) «T3 -*-> (2J
OO «=C "-1- t/1 HI (X
co 01 CD f* c\j ro
Pouer House Mechanical
^
Vj
1
1
;
8
A
|
1
S
£
*cr
r4
Power house electrical connection
«=r
| Power House Electrical
~|
$
81
i i
Transformer and switchvard erection
LD
Transformer and Switchyard
uj
Service bay erection
tD
t i
>>
fO
CD
t.
OJ
CO
r~
Erection of condenser water supply
r--.
«t
Condenser Water
t
and discharge structure
CO
r 1
Erection of water treatment facility
CO
i i
Water Treatment
c
Miscellaneous wiring, grounding, and
Ol
i
[ Control Bui Id inq
1 ,
lighting
a.
cc
ui
I
a.
la.
a cc
S-
-------
QS
C3 "'
-5
-l -,
<(--!-
.0: z to
LU -a; >-
D 1 UI
~ O CL.
1 O
" °-
*" 4
r |
i JL
; i
1 ' I
ro
T"
i <>j
Grading and roads
Buildings and utilities
Concrete plant
t c\j ro
c
TO
c.
0
u
3
if!
C
3
+-LL" j
T^ .
sf
"[
p
\
1
i
I_ /J
T ^~
n
1
r>
^
J
3
J
j
>
;
j
5
5
3
i
3
3
j
i
j
3
)
i /2\
vy
Coal and limestone handling facility
erection
©
Coal and Limestone
Handling
(<*
C
C
4-
£
a
c
c
X
r
t-
£
c:
e:
u.
~c
c
c
E
^
^
5
t
3
T
3
J
5
3
1
i
3
U
> rt
. LI,
)(S
}
'
)
J
r=
c
t/
t
r
4-
^.f"
V^
A*"-
r6
=D
>
i/
c
E
C
t/
~c
'C
-U
l/
s-
c
4-
fC
0.
)(§
^
(^-^
'
L
L
rx
)©
I
i
i
!
(^
Power' house and i-GD electrical
©
c:
OJ
I/)
o
OJ
o
O-
5)
connection
1 Electrical
r
1
B
i
§
I
I
\
\
Vrv
Transformer and switchyard erection
©
Transformer a>id Switchyard
^£
C
O
4_>
CL
i
_o
0)
i-
OJ
s
5*1
3
o
S.
3\
treatment facility
r^
Miscellaneous wiring, grounding, and
©
| Control Building
lighting
0
UJ
o.
D.
8
UJ
OC
UJ
O
a.
< uj
C3 tt U.
E2£
o
UI
o
H
2
8
D
8
2
o
UI
o.
o
F-3
-------
APPENDIX G
ASSUMED PARAMETERS IN FGD COSTS
The following appendix gives the assumptions made by PEDCo in
developing the project FGD costs.
G-l
-------
CO
H
H
rH
I
O
fd
rJ
PQ
Nl
H
Pi
CO
CO
<3
O
3
CO
C
O
H
4-J
ft
s
3
CO
CO
cd
*rj
G
cd
CO
0
H
4-1
CO
H
n
0)
4-1
O
n
cd
u
CO
n
QJ
4-1
0)
B
cd
n
cd
ft
4-J
C
cd
rH
ft
H
0)
t)
0
S
/-* 4J
3 3
4-J M-l pL
m o G
-H
§ rH n
S cd 3 4J
^- o MH cd
^ CM CJ r-H QJ C
CO 0 3 X! 0
lH CO rH CO -H
oi cd 3 oo
rH ,d O &-S 4-1 0)
H rH -H l~^ PQ CSS
O Pu
,0 CM >-, MH X rH
4-1 O S Cd
Q) rH ~ M
H v ' Cd rH tO 4-1
60 cd £ C
G "~J G O O QJ
H o O -H U
CO 60 CO
^ C O rH CO JS
O O CO -H 4J
O CM Pu CJ B M CO
o m -H o) o 4J
o >, fx ia c
rH MH f> >-, CM CO
O 4-1 O 4-1 rH
T3 rH CO tO ft
C co cd cd co
Cd Pn > . W rH
CO O T3 CO rH
fe5 B C rQ 1 CO
O QJ CO rH
"~i ^ c! ri in o o
CM -H 4-J CM 3 -H M-l
> 1 4J 3 O MH O 4-1
o cj co ft co ,-H co in
O -H C 3 i-j o MD
c\l OJ X "H UH CO ^^ O *
rH 01 O SO rH O
" & 4J &« C
o cd oi cd fr-s in o 4J TJ
O EH ^ QJ (^} CM CO QJ
rH EH 41! cj*\ co CM 01 B
OI |S 3
> Ji 0) ,-s ^ ,-v, T3 CO
incucorH CM m -H co
CMg:^^ v_ v_S <;
CO
4J
c
0)
13 co B n
§ O Q) O
H n 4-1
* 4-j -H a
co co 3 cd
oj -H cr MH
H n QJ
4-1 QJ n >.
H CO 4-J 4-1
U 3 U rH Cd -H
Cd 4J Cd O 4-1 O
ft cd n n co cd
Cd 4-1 CO 4J C TJ ft
O CO J3 C O cd
0 O T-l n CJ
4-1 4J a 4J 0)
C C rH cd t-H
cd cd cd CM o -H
rH rH 0 O 00
P^ PJ CJ) CO hJ CQ
^n
r) >.
c
H ^
rd OJ
§M-(
H
p4 f I
Jj
O 3
rH g
MH ---
B
H MH
to o
cd cd
60
n
QJ OJ
3 4-1
rH Cd
pr | l_J
«\
H
QJ
4-1
CO .-fl
V-" §
4-1 -^
cd 3
0) 4->
PS M
«*
E>^
4-J
H
C-J k3
cd g
ft
cd
CJ
m m in m in
CO CO CO CO CO
o o in o o
o in r~- oo o
m co rH o o
*% f\ ft *% «1
CO CO CO CO CO
O O O 0 O
o o o o o
O in CM o r~-
O CT\ cy\ CTi OO
rH
m o o o o
CN O O O O
rH CM in O
«t
rH
QJ
£s *-W
0 -H
rH rH
M-l
60
co c3
cd -H
60 Pi
H
QJ cd
MH M
** ^
CO pi
o) cd
4-1
cd co
n QJ
4J
4-1 Cd
3JJ
ffl
*
CO
4J
f3
cd
rH
ft
rH
rH
cd
M
o
MH
pti
0
o
rH
CO
*X3
OJ
B
H
CO
CO
<^
CO
OJ
n
3
4-1
cd
M
Q)
ft
cD
4J
CO
cd
60
QJ
3
H
|~T !
QJ
n o cd
4J G >
C OJ
O 60 QJ
CJ 4J OJ
CO Pi
0) 01 -
co
CO
QJ
0
o
n
P*
C
0
H
4-1
cd
N
r-J
PI
M
3
MH
rH
3
CO
a)
Q
4-J
C
01
H
MH
M^
W
-a
0)
o
P!
cd
T3
n
o
MH
CO
QJ
cd
4-J
CO
w
4-J
CO
B
C
0
n
H
>
G
W
CO
h3
4-J
C
0)
B
ft
o
rH
0)
>
0)
o
G
cd
je
o
r-J
cd
QJ
CO
Q)
"
O
QJ
a
H
M-l
MH
°
n
o
4-1
CO
n
o
>>
j^
QJ
G
cd
rH
CJ
CJ
a
o
CJ
f^
,0
*s
<:
4J
^
cd
PH
T3
N«^
>
0
G
Q)
60
r~~
rH
>-,
cd
r^
v§
CD
**o
ex
PH
r
U rt QJ ».
OJ CO M -H
H S n o QJ
cd 4-1 QJ & u
4-1 co -o 4J n
' ^ C 3 3
co 3
-------
O
W
TJ
cu
c
(3
O
U
en
a
0
H
4J
p.
£
P
en
en
cfl
T3
0
cfl
en
o
rl
4-1
en
rl
^
CU
4-1
CJ
CO
$-4
cO
r*
O
O
oo
4J
CO
3
&C
3
C5
en
^4
cu
.'^J
4J
en
0
a
C
CO
cfl
cfl
T)
O
C
O
44
-x^
O
o
o
in
4-1
rl
T)
CU
CJ
cu
l^,
cfl
CJ
4-1
rH
CO
cn
c
o
4J
^^
o
o
m
CM
CO-
44
H
T3
CU
!-i
a
*rj
rl
U
CO
CJ
H
v^
3
rH
3
CO
r!
c c 5
O O rH ^
44 4J J3 -^
"^^ -*-. fo en
O O -^. rH
O O 0 rH
O -rl
m in 0
oo CM in
rH rH rH in
co- -co- ^
^1
CU
^
o
u
cu
^J
rH
CO
44
H
&,
CO
4-i
c
eu
e
44
en
cu
K*
C
H
rH
CO
44
H
P.
cfl
CJ
rH
cO
44
O
44
MH
O
44
C
CU
CJ
<_|
CU
P.
4H
0
rH
cfl CU
CO rH
O 3
P. TJ
en O
H e
a
cu
cu V4
44 CO
H P-i
en co
i
C 0
O £S
CO
44
C
CU
B
cu
t VJ
Cd -rl
co 3
o cr
p. cu
en >J
rl
TlJ >^
CJ
0) ti
M cd g
T3 T) S
3 C
rH 3 m
CO T3 CM
CU
&
CU
rH
3
T3
o
e
CU
S-*
cfl
P.
en
rH
g
§
O
0
0
rH
r.
O
o
m
*
o
o
CM
*,
Q
O
rH
o
4-1
cu
4J
cd
rH
eO
O
en
0)
C
cfl
ej
3
en
3
O
rl
J-l
CO
0)
^:
4-1
MH
O
en
t-i
o
4J
3
J3
H
en
H
TJ
o
c
H
0
§
u
Q
s
G-3
-------
TABLE G-2
ANALYSES OF COALS USED AS THE COST ESTIMATING BASIS
Coal Type
Total
sulfur
Pyritic
sulfur
Ash
Heating
Value
Btu/lb
Eastern bituminous
Eastern bituminous
Western subbituminous
Western lignite
Anthracite
6.39
3.48
0.8
0.4
0.8
4.6
2.49
14
14
8
6
6
12,000
12,000
10,000
8,000
13,500
Source: PEDCo, 1977.
G-4
-------
I
o
w
a
c
o
4J
a
m
n
D
c
«
m
O
4J
Ifl
-.4
ract<
c
a
01
H 41
W *O
01 H
C X
en o
n)
*
c
S.S
T3 3
O rH
01 O
0)
4t -H
-t >H
_fl Hj
p .X
gs
01
c
o
+j
P)
1
-*-<
0)
c
,4
stems
\
o o
*r *-4
«tf no
CJ
CTi 00
>,
Hj IN V
M *
frn -H \ > n
« "fi tr « n *>
59 -H tr a -"TO
o w
10 OJ
oo oj «
-------
a)
C
H
4J
C
O
O
O
3
H 41
M -0
0> H
C X
fr o
it)
n c
c go
o 3 *
-H H 4J
4J "O 3
a o-i
e wo
3 m
n
n
TJ
c
c at -H
i-H *
« XI n)
U 3 M
^j r? ""*
ID
V4
(U
4-1
O
ki t
< " o o
U
H CD
~-f
iVi
ft) VU -V
\-t
t* <*i ^ rsj
,^.
o ^*
n co
^
«3 0 -V
W
S-4 f*l -V ~ -H
0) IQ n IM
a x> o o >.
P -H
«*w O 0 U
W< O *J O O
0) O r-t U
X) ^ *D -t ft) 0)
IM ft* -H \ > a
0 jQ P *-* x.
m g & « co 4J
XI 3 -HO1 «) »«-i
In ^-^
rj o
rH lft ^
. o
to
c-4 oo o r*<
in k. « E
no -*H
*** +J *P
0) U
v< u c *
3o O 4J
4J -H C
id - *J 4J at
U -u « C E
e*) "S t! Jl
it 3 ^ M p
*J o tr
*w J^ 0) *J
« 1-1 o c u u
10 0) 10 3
tT1 jQ 0) 4J Q. T)
ti 4) 36
01 O M -D tl U
3 « & ^-i ji a
r-i xt 41 o a ><
&. « o x x. m
G-6
-------
APPENDIX H
MEASURES TO IMPROVE FLUE GAS DESULFURIZATION
AVAILABILITY AND OPERATING PROBLEMS AND SOLUTIONS
Improvement Measures
Various measures have been or can be used to maintain high levels
of FGD availability. These measures which are discussed below can be
grouped into maintenance methods, operating techniques, and design
concepts.
Maintenance Methods
The maintenance methods applied by La Cygne and Sherburne County
have succcessfully maintained a high system availability. The impor-
tant factors in these maintenance programs are: (1) taking one or
more modules off-line each night for inspection and cleaning, (2) use
of a separate maintenance crew trained to work on the FGD system, and
(3) a general dedication to gaining a better understanding of the
system and how to maintain it better.
Operating Techniques
There are several operating techniques that have or can be
used to contribute to maintaining a high FGD system availability.
Over and underspray of mist eliminators (demisters) removes deposits
from the mist eliminators. Automatic pH and process control result
in more stable operation and tend to prevent major failures such as
massive scaling. Finally, a staff of operators and technicians to
work with the FGD system on a daily basis is very important.
H-l
-------
Design Concepts
Each of the FGD systems examined differs somewhat in design
concept. Some of the concepts that have been or potentially can be
successful in enhancing availability are: (1) particulate removal
before the FGD system with an electrostatic precipitator (ESP), (2)
dry flue gas booster fan between the ESP and scrubber rather than a
wet fan after the scrubber, (3) adequate redundancy of pumps, valves,
lime/limestone feed systems, packing gland water systems, etc., (4)
spray tower scrubber configuration, (5) adequate instrumentation for
pH, S02, additive use, etc. with automatic controls, (6) indirect
reheat of flue gas, and (7) adequate particle dropout area to reduce
solids carryover to the mist eliminators.
The areas of improvement discussed in this section represent a
composite of experience at several specific FGD units. Future FGD
units are expected to include many of these improvements.
Operating Problems and Solutions
There have been and still are problems associated with FGD
systems; however, many of these problems have been solved and the
methods of reducing the severity of the remaining items are much
better understood.
To date, the problems encountered with FGD systems and the
severity of these problems varied both with system type and within
units of the same system. The more common problems encountered
are listed below.
H-2
-------
t Formation of scale in the absorber and associated equipment
in lime and limestone systems leading to plugging and reduced
capacity.
Plugging of mist eliminators, lines, and some types of
absorbers.
9 Failure of ancillary equipment such as pumps, piping, pH
sensing equipment, reheaters, centrifuges, fans and duct
and stack linings.
Inadequate absorbent make-up preparation.*
4 Handling and disposal of sludge in nonregenerable systems.
Scaling and Plugging
In lime and limestone systems, scaling has been a particular
problem and has reduced operability. Both a soft sulfite scale and
a hard sulfate scale may form in the absorber, mist eliminator,
and ancillary tanks, pumps, and pipes. Specific process control
techniques which have produced significant improvements include:
t Use of Magnesium
Full-scale and test facilities in this country have effectively
reduced saturation and scaling by addition of magnesium to the cir-
culating slurry. The TVA Shawnee facility, the Phillips facility,
and the Paddy's Run facility demonstrated that the addition of
magnesium to the lime and limestone slurry eliminated scrubber scale
formation. The Bruce Mansfield and Conesville stations use lime
containing magnesium oxide to prevent scaling.
Operation at subsaturation levels for calcium sulfate and
sulfite
*Discussed in PEDCO Environmental report, Nov. 1977.
H-3
-------
By maintaining high liquid to gas (L/G) ratios, the proportion
of unreacted lime or limestone remains high relative to the absorbed
SO . There in thus less chance of creating a supersaturated solution
of sulfites or sulfates. The higher L/G ratio also improves overall
SO collection efficiencies. The actual L/G will vary with the type
3
of absorber, and values in excess of 10.8 1/m (80 gal/acf) have been
used in spray towers.
Increased reaction tank holding time will also decrease satura-
tion by allowing further reaction between the absorbed SO and the
lime or limestone slurry. Slurry residence time at the Green River
facility is greater than 20 minutes, and scale formation is not
a major problem.
pH Control
Work at the EPA-Shawnee test facility has shown that an important
parameter in controlling scale formation is solution pH. The measure-
ment of pH has also received considerable attention. More rugged and
dependable sensors are being used; they are located in the slurry
stream where they are subject to less breakage, are more accessible,
and where they yield data which is more reliable and responsive for
pH control.
Co-precipitation of sulfate
H-4
-------
Minimizing the oxygen content in the flue gas by reducing any
air in-leakage, favors co-precipitation of sulfate crystal. There-
fore, air exposure is reduced by covering open reaction tanks,
clarifiers, etc.
Plugging caused by deposition of solids on equipment surfaces has
sometimes restricted the passage of liquids or gas in FGD systems. It
is usually easily removed by flushing with water or steam. Plugging
in pipes can be prevented through designs which avoid low flow
velocities. Careful control of raw material particle size and
screening of the slurry also decrease plugging problems, especially
in spray nozzles, pipes and pumps. Since this problem is caused by
the deposition of solids from the recirculating slurry, reduction of
the overall amount of solids will reduce the plugging. The minimum
stoichiometry that will effect the required SO removal efficiency
should be used. This has been demonstrated at Shawnee and LaCygne.
Erosion and Corrosion
Many problems with ancillary equipment were due to corrosion
and erosion. Erosion in venturi prescrubbers has resulted from high
fly ash loadings. Likewise, prescrubbers remove the bulk of any
chlorides and sulfur trioxide in the gas stream; both of these
components are highly corrosive. Corrosion occurs more frequently in
areas after the absorber is subjected to wet saturated flue gas as
opposed to areas subject to alkaline slurry streams.
H-5
-------
There are so many factors involved in FGD operation which affect
corrosion rates, that generalizations regarding corrosion resistant
materials are difficult. A sufficient amount of data has been accumu-
lated, however, to provide general guidelines for the construction of
critical elements in FGD systems as summarized below:
(a) Some systems are incorporating such alloys as Hastelloy
C-276, Hastelloy G, Inconel 625, Incology 825, 317L stain-
less steel, 904L stainless steel and Jessop JS700 in
wet/day high temperature, high chloride environments, such
as in presaturators. The LaCygne Station has found that
these materials give excellent reheat service. The Bruce
Mansfield station has had good results with Hastelloy
wetted parts of the fan.
(b) Synthetic and natural rubber coatings predominate in recycle
tanks, pumps, and lines. These materials have been reported
to give superior erosion resistance once application problems
have been overcome. For instance rubber lined pumps have
been used successfully at the following facilities: Green
River, LaCygne, Bruce Mansfield, and Conesville.
(c) For liners in the absorbers, exhaust ducts and stacks, a
number of materials such as resins, ceramics, polyesters,
polyvinyls, polyurethanes, Carboline, and Guriite, have
been used with varying degrees of success. Although
successful applications have been reported, widespread
failures of the liners have been attributed to improper
application, instability of the materials at high tempera-
tures, inconvenience of repair, and cost-related factors.
These problems are especially evident on higher sulfur
coals. Extensive effort is continuing by FGD suppliers to
fully solve this problem.
Equipment Design
Approaches utilized to reduce problems with ancillary equipment
include:
Recirculating Pumps - Slurry recirculation pumps provide the
driving force for the liquid circuit in FGD systems. In their
design, special attention must be given to an accurate service
description (solution pH, specific gravity, solids content,
gas entrainment, flow rates, and head). A number of general
trends are evident and summarized below:
H-6
-------
(a) New systems must incorporate spare pumps. Spare capacity
from 50 percent (one spare for every two operational)
to 100 percent (one spare for every one operational) is
useful to avoid downtime. This type of spare equipment
is found at new large stations including Bruce Mansfield
and Conesville.
(b) Natural and synthetic molded rubber lining should be
specified for wetted parts in the pumps.
(c) Flush-water wash systems are needed to purge the pumps
of solids, which tend to settle out during periods of
inactivity.
Mist Elimination - Chevron and baffle-type mist eliminators
have been and are currently being used in virtually every FGD
system in the United States. The popularity of these collec-
tors is due primarily to design simplicity, high collection
efficiency (for moderate to large size drops), low pressure
drop, wide-open construction, and low cost. Within these two
preferred types of mist eliminators, a number of specific
design and construction innovations have been implemented:
(a) Chevron designs (continuous vane construction) are
predominant over baffle designs (discontinuous slat
construction).
(b) Fiberglass-reinforced plastic is now used at nearly all
facilities.
(c) The horizontal configuration (vertical gas flow) is also
used in almost all installations for cost reasons.
(d) Two-stage designs predominate over single-stage designs,
because they yielded higher elimination efficiencies.
(e) Operation at high alkali utilization.
(f) Bulk entrainment separators, perforated plates, impinge-
ment plates and other precollection devices are becoming
integral parts of mist elimination systems. These reduce
plugging and improve separation. The Conesville facility
employs this as well as LaCygne and Coal Creek.
(g) Mist eliminator wash systems that employ intermittent,
high-velocity sprays predominate over continuous wash
systems. These produce a hydraulic washing effect.
H-7
-------
Application of these approaches greatly diminishes mist elimina-
tor problems.
* Reheat - Virtually all the FGD systems coming on-line and
planned for future operation incorporate some type of stack
gas reheat system. These systems heat the flue gas to avoid
condensation with subsequent corrosion to downstream equip-
ment, ductwork, and stack and to suppress plume visibility
as well as enhance plume rise and pollutant dispersion. To
date, a number of "wet stack" FGD systems (no reheat) have
been installed and have encountered corrosion problems. The
trend in reheat systems is toward heating of ambient air and
mixing with the flue gas and mixing of hot untreated flue gas
with scrubbed gas. In-line reheat systems have been subject
to corrosion and solids deposition, the latter often occurring
because of inefficient upstream mist elimination. Application
of heated ambient air reheat systems essentially eliminates
reheater problems.
Fans - Fans installed immediately after an FGD system (wet
fans) have experienced corrosion, chloride attack, and solids
deposition problems. Deposition problems have caused fan
imbalance resulting in excessive bearing wear and damage to
the fan. Only two systems have this trouble: Phillips and
Bruce Mansfield. The problems associated with fans installed
upstream of the FGD system (dry fans) include operation at
higher temperatures (over 150°C) resulting in higher gas
velocities and abrasion by fly ash. Dry fan problems are
more easily solved, and the tendency is toward fans upstream
of the FGD system. Where necessary, however, the use of
various steel alloys have made wet fans a viable alternative.
H-8
-------
APPENDIX I
EFFECT OF COAL PROPERTIES ON FGD SYSTEMS
The major coal properties affecting FGD system design and opera-
tion are heating value and sulfur, ash, moisture, and chlorine content.
The effects include:
Heating value of coal. Affects flue gas flow rate - generally
higher for lower heating value coals which also contribute a
greater water vapor content to the flue gas
Moisture content. Affects the heating value and contributes
directly to the moisture content and volume of the flue gas
Sulfur content. The sulfur content together with the allowable
emission standards determines the required S02 removal effici-
ency, the FGD system complexity and cost, and also affects
sulfite oxidation
Ash content. May affect FGD system chemistry and increases
erosion. In some cases it may be desirable to remove fly ash
upstream from the FGD system
Chlorine content. May require high alloy metals or linings
for some process equipment and could affect process chemistry
or require prescrubbing.
The importance of these factors is described in this section.
Coal Heating Value and Moisture Content
Because a power plant using a low heating value coal must fire
at a higher burn rate to generate the same amount of power, such coals
produce a larger volume of flue gas and greater S02 emissions per unit
of generated power. The effect on the FGD system is twofold. First,
the flue gas handling equipment, including the scrubbers, must be of
a larger size to accommodate the greater gas volumes. Typically,
power plant flue gas volumes may range from 5,000 to more than 7,000
1-1
-------
m^/hr/MW (about 3,000 to 4,000 acfm/MW), depending on the coal compo-
sition, boiler heat rate, gas temperature, and power plant elevation
(or gas pressure). Secondly, the increased SO.-, emissions mean that on
a megawatt basis the FGD system must treat proportionally larger quan-
tities of S02» On a megawatt basis, therefore, the FGD system (as
well as the power plant) equipment capacity is greater and capital
and operating costs are higher for coal with lower heating value -
for a given coal sulfur content and SO^ removal efficiency.
A characteristic of lower heating value coals and coals of high
moisture content is a flue gas with a greater proportion of water
vapor. This leads to smaller amounts of water evaporation in the
scrubbers, which in turn affects the temperature to which the gas is
cooled. The overall effect is that S02 absorption takes place at
slightly elevated temperatures. This could affect absorption effici-
ency, depending on the chemistry of the particular process.
Sulfur Content of Coal
The sulfur content of the coal together with the allowable
emission standards determines the absolute removal rate of SC^ in
pounds per hour. For a given absorption efficiency the sulfur con-
tent of the coal directly affects the design of almost every piece
of equipment in the FGD system.
For example, a lime or limestone system designed for high as
opposed to low sulfur coal would have:
1-2
-------
Scrubbers with capacity for greater SC>2 removal
Higher L/Gs and therefore bigger pumps and piping and higher
pumping energy requirements
Bigger fans and greater energy requirements if the improved
scrubber design results in higher gas pressure drop
Larger sized alkali storage, preparation and feed equipment
Greater lime or limestone feed rates
Larger scrubber recirculation tanks to maintain residence
time for increased L/Gs and to provide additional time for
increased SGv absorption load
Greater capacity slurry solids separation equipment
Provision for disposing of the larger waste volumes
Increased power requirements for the larger equipment loads.
With proper design, operation and maintenance the FGD systems can
achieve good availability and removal efficiency for either high or
low sulfur coal. However, for higher sulfur coals the lime and lime-
stone FGD systems are more complex and have higher capital and
operating costs than a low sulfur application.
Other FGD processes are similarly affected by the sulfur content
of the coal. For systems using regenerable absorbents (double alkali,
magnesium oxide and Wellman Lord processes), the capacity of the regen-
eration section is directly proportional to the sulfur content. With
high sulfur coal the overall cost of these sections (capital and opera-
ting) represents a large portion of the total cost for the system.
With recovery processes there is relatively little waste but a large
by-product processing cost, although the greater amount of by-product
produced helps to offset these costs.
1-3
-------
The amount of sulfite oxidized to sulfate in scrubber solutions
is proportional to the relative amounts of oxygen and SO absorbed.
It also depends on the pH and temperature of the liquid as well as the
composition of particulate emissions which may contain iron or copper
that act as catalysts for the oxidation reaction. In general, however,
as the gas SO concentration becomes smaller, the fraction of sulfite
oxidation tends to increase. For this reason, when lime or limestone
scrubbing systems are used for low sulfur coal or for boilers opera-
ting on high excess air they may experience high sulfite oxidation
and produce waste solids that are mainly gypsum. This can be a desir-
able feature since gypsum solids are more easily dewatered due to
faster settling Tates and higher final settled densities. Conversely
the lower oxidation observed with high sulfur coals can lead to
solid wastes high in sulfite and difficult to dewater.
Ash Content of Coal
Most coals fired in U.S. utility boilers contain 5 to 30 percent
ash. After combustion, part of the ash falls to the bottom of the
furnace and the remainder is carried upward with the flue gas. The
fraction of the ash that is carried overhead is a function of the
boiler design and combustion parameters. With pulverized coal firing,
85 percent or more of the ash appears as fly ash; in cyclone boilers
about 20 to 30 percent of the ash goes overhead.
Fly ash can be removed upstream of the FGD system by a precipi-
tator, fabric filter, or prescrubber, or integrally within the FGD
1-4
-------
system itself. Not all FGD processes are suitable for combined
removal. Even when fly ash is removed upstream, residual ash becomes
entrained in the process liquor.
Fly ash is invariably abrasive; some is chemically inert, and
some is highly acidic due to SO adsorption. Fly ash can cause exces-
sive erosion, scaling and plugging of equipment. It contributes to the
waste volume of throwaway processes, the loss of absorbent for regen-
erative processes, and may contaminate the byproduct of recovery pro-
cesses. Certain coals from Wyoming, Montana, and North Dakota produce
alkaline fly ash with large amounts of reactive calcium, magnesium,
sodium, and potassium oxides. With combined removal of alkaline fly
ash and S0?, major reduction in the alkali makeup requirement can be
realized. The presence of calcium alkali in the ash can, however,
aggravate wet-dry interface problems by producing hard insoluble
deposits.
In general, only processes using nonregenerable absorbents (lime,
limestone, soda) can be used for the combined removal of fly ash and
SC^. The fly ash is then disposed of together with the spent absor-
bent. Venturi scrubbers are often used for this purpose at the
expense of increased pressure drop over other absorption systems.
However, the fly ash contributes to solids buildup at the wet-dry
interface and causes erosion of pipes, pumps, spray nozzles, and
scrubber internals.
1-5
-------
Chlorine Content of Coal
The small amounts of chlorine in coal are converted to gaseous
chloride in the boiler. The chloride is absorbed from the gas by
wet scrubbing processes. Its presence provides the potential for
chloride stress-corrosion, requiring in some places the use of high
alloy equipment wherever rubber or other protective coatings are not
applicable.
In wet scrubbing processes, dissolved chloride replaces active
calcium, magnesium or sodium alkalis by their chloride salts, which
are inactive in the absorption process. The alkali associated with
the chloride is then lost as dissolved solids in the water portion
of the waste sludge. From a cost standpoint this is particularly
objectionable for magnesium and sodium based processes (or magnesium
enhanced lime and limestone processes), because these alkalis are
relatively expensive. For such processes, prescrubbing may be used
to absorb chlorides from the flue gas upstream of the FGD system.
This minimizes both alkali loss and chloride stress-corrosion problems,
For lime and limestone processes, an equivalent amount of calcium is
used up by the chloride (the amount is small relative to that used
for SO absorption), but the calcium is relatively inexpensive.
1-6
-------
APPENDIX J
FORECASTS OF FUTURE ELECTRIC UTILITY
INDUSTRY STRUCTURE
This appendix describes the method utilized to forecast future
electric utility industry structure under two scenarios of possible
growth. Baseline forecasts are presented for moderate and high in-
dustry growth with no change in present new source performance
standards (NSPS). Additional projections are provided for growth
rates and for several possible revised NSPSs. The basic tool utilized
to provide these forecasts is the Utility Simulation Model, a large-
scale computerized model that simulates the response of the electric
utility industry to specified economic conditions, energy policies
and regulatory constraints. A detailed description of the model is
contained in Volumes I and II of "An Integrated Technology Assessment
of Electric Utility Energy Systems" by Teknekron, November 1977.
Material in this appendix was extracted from "Review of New Source
Standards for SO Emissions from Coal-Fired Boilers," a draft pre-
pared for the U.S. EPA by Teknekron, 1978.
Base Year Data
The results presented here were projected from a data base
containing a description of every electrical generating unit (nuclear,
oil and gas-fired, hydro, geotht-rmal and combustion turbine, as well
as coal-fired) operating as of 31 December 1975, plus announced plans
for new units through 1985. Beyond 1985, the model creates new
generating units and sites them, by county, as needed. In order to
J-l
-------
simulate the industry's response to a pollution control regulation,
including a particular NSPS SO emissions from coal-fired units, a
minimum set of scenario parameters must be specified.
Key among these are:
Future growth rate in peak and average power demand
"Future mix fractions," giving the breakdowns of new
generating capacity beyond 1985 by type of generation
Kinds of coal to be burned by each coal-fired unit,
including sulfur and ash content.
Other nonair-pollution related variables that must be specified
include an overall inflation rate, new plant construction costs, costs
of water pollution controls, assumptions about the rate of phase-
out of natural gas as a utility boiler fuel, the minimum generating
reserve margin to be maintained, the size and thermal efficiency
of future generating units, fuel prices and price trends, and the
order in which each utility system will "dispatch" the available
units to meet the projected demand.
Variables which relate directly to air pollution controls include
specification of the sulfur dioxide (SO ) particulate and nitrogen
oxide (NO ) emission limits that must be met by both old and new units;
X
the costs of the pollution-control devices used to insure compliance
(flue gas desulfurization systems, electrostatic precipitators and
fabric filters, modified boiler configurations for NO control); and
x
constraints on siting future units due to air quality considerations.
The basic results produced by the simulation are industry
composition and fuel consumption down to the county level which
J-2
-------
C/J
I"H
rf\
tjj
p^j
3
3
^N
*~7
5
W
JT-J
£_<
£_<
£3
o
33
S
3
o
<&
33
p-j
H
L <2
>-i H
cn
ta z
3 0
S CJ
*
td
«
-
T> to
-a c cu -a cu -a r
c
1/rt
1
to c
O
c
O cn
i- 4->
-
C as ^.
*^
CM
c -a i
o cu
f- -r- >>
4J r- J3
03 Q.
4-> Q.T3
C 03 CU
CU E
e co 3
cu cu to
i 4-> CO
Ct. «3 03
£.
« 3 CO
o o
cu t- c
4-1 4-) 03
n3 J- t
4-> 03 r
oo a. a.
. .
CO
c
o
4->
03
^
3
CT
cu
on
c
o
-r-
4J
3
r~
r»
o
D_
X-
r-
to 4J
i ->. CU Q.
i CTHJ CU
us c o o
O CO 3 -T-
4-> r U
Lf> t- 4-»
O 4J «5
CU X-
i «- «3 T3
i a> Q- cu
CX X- £
CX «3 S- 3
rtj O to
r- <4- to
i/> as r-
*l^
in OX
to -O
o CM z:
^- co o
cn x-
s_r cn o
o c «»-
l*- T3 ^~,
C CO ID
to (a . *»»
r>. c J- jo
en cu o r
r CU M-
> 3; r-v
4J - .
+-> QJ O O
C JC1 1
cu ca
s- cu ->
S_ CU3 ~^
3 -i- o cn
CJ r r C
a o
CU i
~a cn
3 CU
r 01
u
X r
cu o
S-
CU 4-> (O
X- C -
rtj O «/r
CJ 3
-O 4-1
CU >> 03
0 4-> +->
C -r- 10
3 i
o
C 3 C
c: o- cu
03 E
s- c
>><- -1-
3
t/T
(Si
aj
-o
s_
o
c
cu 4-> cu
E « 5-
10 t 4->
1/1 4-> CU
03 CO X-
cu
cu cu
s- <: x-
03 UJ iQ
U.
" CO
"*> C 4-»
w
^3"
CO
o>
X-
QJ
4->
«b
«3-
1
CO
co
co
x_
Q)
-a
o
-a
cu ~o
-M c:
S3 «5
o>
r r-»
3 r^
E cn
O i
0->>
-Q
to
J t "
O 1 O T- <* -r- QJ
f ' ''
O^i Q. C
05
+J r- 3 3
T
^3
J=
O
s_
Q_
o -a
o cu cu
ro to X_
n3 *-
CU -O t^-
c: i
3 CO CO
4->
*r^
3
.a
VI
T3 «ar «
«3 4->
CU C 4J
s- 4->
S- ii
o o
C Q- CO
O CT
>! OJ nJ
QJ i
JQ cncj
c
tl T- >,
4-> C S_
^- T- 0
C
3 4-> nj
c -o
o o c
CU (J «3
s- s:
r- >»
<»- 4-> £_
I C O
r- 3 »-r-
c c ^
(O
U
S-
»->
CO
CO
a:
cr
c
4->
«^
to
CU
Q-
co
n3
A
fM
03
O
CJ
O
4->
co
"3
cn
-O
c
03
fH.
'r
O
E
O
«*-
CO
c
o
r-
to
i_
OJ
>
c:
o
o
«
CO
CU
4->
«
t-
cu
CfT
a
c
13
CO
c:
o
CO
S-
cu
>
c
o
CJ
3 cn
<»- o
O 1 r
CU r-
to co «3 T3
«3 CJ1 3
X- i
4J r- O
CO ,<: cu
en x_
X- 03
ai-Q
c: -ir>
LU ir> r~-
oo cn
< en r
)T-
-a ^
CU T3
U_ T3 CU
CU -C vo
cu 4J to cr>
j:: u t- en
4-> CU r r
<4- -Q
>,H- 3 >>
J3 CU Q-^3
>K,
4J
r-
O
R3
CL
"3
O
cn
c
4->
03
CU
C
QJ
cn
o
cu
X-
sZ
I
CO
«3
Cn
X-
o
^
«^-
o
3
QJ
C
O
z
u
a.
u_
X-
QJ
a.
1/7
(O
CO
4->
C
cu
E
a>
X-
T
4->
cu
X-
T3
C
03
k
to
QJ
4->
03
X-
O)
-o
M
CO
cu
4->
»
J= U
4-J +J
to
a 3
C XJ
OS C
^
03 X-
U QJ
"E o
cu a.
jr
o o
J= X_
4-> 4-> -
i- U >>
3: cu r
r QJ
cu cu >
o -»
c E +J
n3 (O CJ
t- QJ CU
r 4-> CX
Q- CO CO
E cu
O O) S-
CJ -C
4->
r r
i X- CQ
3 o cn
a_ <*- t
^.
co
£;
Q
4_)
03
3
cn
cu
(V
C
Q
^J
3
^
o
j_
CU
4_l
03
OO
CO
c
O
*r-
4->
U
CU
*l~i
o
X-
CL.
* t
14
CO
QJ
il
CJ
to
3
ca
cu
s-
3
ca
CO
3
co
c
QJ
CJ
,^
c
o
4->
a3
3
a.
o
a_
ui
VI LfJ
cu r*
c 4-> cn
X- C r
cu cu
-o E c
1 C T
f-* T"
3: «5-o
-M X-
4-> aS
' «5 T3
r>. i c
r-- c 05
cn o 4->
r Z 00
<»- >>
O 4->
CO T~
CO CU t
4J X- «3
C U 3
QJ cn cu
U C "T-
CJ X- «3
(0 Z
QJ CU
.C >>
4-> i C
03 «3
j: c
4-> O
03 CO o3
-a cu «
i_ X- O
0 0 t-
o «a >
o
03 O ~O
o cu
c o -o
i- ~ X-
tr> o
-a o
cu cn co
c c: s-
n_ -a «3
QJ CU
-a QJ sz.
U 4J
QJ X T-
x_ cu 3;
03
CO CO
CO J
X- O. C
03 =>
r- 0
> < 03 O
co S. CU
1/1 O co
03 E O
r 5 J=
CJ E 4-J
>>TJ CU
I- C X-
O «3 as
4J
03 to co
T3 «3 03
C CU CO
OS X- t-
2: OS 03
03
CO
X-
03
X-
O)
4J
03
r
X_
O
p~-
r>-
cn
1
c
t
r^
03
0
O
O
4->
4->
x- to
CU O-
>tn
c z
o
u cu
-C
4-> 4-»
3
J2 4->
O
r-- c
t~-
cn -
i 4->
»»~
co E
X- ~r-
Oi
<4-
CU i
J3 T3
o
C 0
o
j- a.
4J i
03 tO
X-
co cu
O-i
O -Q
n3
C O
03'"-
CT>r
QJ Q.
J3 0.
«3
^=
O QJ
i- .C
J= 4->
3:
0
CO 4->
4->
T- 4J
C U
3 CU
<->
o jn
CO 3
X. to
<£ cu
i _a
CO
03 O
cr>4->
a -a
c cu
to S
3
i CO
T CO
O "3
JD
J-3
-------
consists of generating mix, fuel consumption, reserve margins and
capacity factors.
Specification of Scenarios
The set of key input variables, which must be specified before
a simulation can be run collectively, defines one scenario. These
variables can reflect both broad national policies or region-specific
assumptions such as the fraction of post-1985 baseload nuclear gene-
rating capacity in a particular state.
The quantity of data needed to fully define a scenario is too
extensive to be discussed here. Table J-l summarizes assumptions that
are common to all scenarios analyzed to date (Teknekron, Inc., 1978).
The key scenario elements that vary among the scenarios are
the assumed growth rate in demand for electricity and the revised
NSPS being analyzed, the latter being applied only to coal-fired units
of at least 25 MWe on-line in 1983 or later. Table J-2 summarizes
the two sets of demand growth scenarios considered.
TABLE J-2
NATIONAL ELECTRICITY DEMAND GROWTH RATES
(Percent per year)
1975-1985 1986-2000
Moderate growth
High growth
Peak
5.8
5.8
Average
5.8
5.8
Peak
3.4
5.5
Average
3.4
5.5
J-4
-------
The alternative NSPS's that were considered involve one change
in the particulate and NO standards, combined with three different
x
SO standards:
SO : Ninety percent post-combustion SO removal
with an upper limit ("cap") on emissions of
520 ng/J (1.2 lb/106 Btu)
Eighty percent post-combustion SO removal
with an upper limit on emissions of 520 ng/J
(1.2 lb/10 Btu)
No fixed percentage of SO removal with an
upper,limit on emissions of 220 ng/J (0.5
Ib/lCT Btu)
Particulates: A limit on emissions of 12.9 ng/J (0.03 lb/106
Btu)
NO : A limit on emissions of 260 ng/J (0.6 lb/
10 Btu).
The "moderate" growth cases are meant to reflect a successful
conservation effort as envisioned by the President's National Energy
Plan (U.S. Congress, 1977). (Assumptions about natural gas phase-outs
and oil and gas conversions to coal are also designed to reflect the
goals of the National Energy Plan.)
The nomenclature used to label results from the nine different
scenarios analyzed is: the letter "M" or "H" first indicates whether
the "moderate" or "high" growth assumption was used. This is followed
by three numbers which specify the SO emission "cap" (in lb/10 Btu).
Since the NO limit was set at 260 ng/J (0.6 lb/10 Btu) in all cases
x
but the baseline scenarios (no NSPS revisions), its value is not
indicated explicitly. The scenarios are summarized in Table J-3.
J-5
-------
TABLE J-3
ALTERNATIVE NSPS SCENARIOS
Scenario Label
Ml.2(0)0.1
(Baseline with moderate growth)
HI.2(0)0.1
(Baseline with high growth)
Ml.2(90)0.1
HI.2(90)0.1
Ml.2(90)0.03
HI.2(90)0.03
Ml.2(80)0.03
HI.2(80)0.03
MO.5(0)0.03
Revised NSPS in lb/10b BTU (% Removal)
S02 = 1.2 (0)
NO;
0.7
Participates = 0.1
Same as above
SOz = 1.2 (90)
NOX =0.6
Participates = 0.1
Same as above
Same as Ml.2(90)0.1 but with
participates = 0.03
Same as HI.2(90)0.1 but with
particulates = 0..03
SO? = 1.2 (80)
NOX = 0.6
Particulates = 0.03
Same as above
SOo = 0.5 (0)
NOX = 0.6
Particulates = 0.03
J-6
-------
Industry Projections
The following sections characterize the utility industry* in
the base year (1975), and then project that configuration into the
future. The characteristics are the capacity mix, i.e., aggregate
generating capacity broken down by type of generation (coal, oil,
nuclear, etc.); the distribution of generating units by regulatory
category (SIP, NSPS, or revised NSPS); and the amount of capacity
using FGD.
Because the Utility Simulation Model takes into account many
of the complex interactions that occur among utilities' pollution
control compliance strategies and their other planning and dispatching
decisions, projections of characteristics like capacity mix are not
made independent of the particular pollution control scenario con-
sidered. To illustrate, a decision to comply with an S02 emission
limit through use of FGD will result in a generating capability
reduction that must eventually be compensated for somewhere in the
system. If that utility system's reserve margin is ample, then the
lost capacity can be compensated for in the short term by running the
existing units at higher levels. If the reserve is already near
the safe minimum, however, the utility may be forced to plan for
increased capacity additions, either by building more combustion
turbines in the short term, or accelerating planned building schedules
*The investor-owned sector and the publicly owned sector (municipal
systems plus the Tennessee Valley Authority and other Federal pro-
jects) are treated together.
J-7
-------
in the long term. Regardless of the particular system, more capacity
will have to be added in the long run if a substantial number of units
are forced to use FGD because of a new air emissions regulation. Fuel
consumption also varies with emission control strategy. For
example, increased use of FGD in the Midwest and East will tend to
encourage the use of locally available medium and high sulfur coals
at the expense of more distant supplies of low sulfur western coal.
This in turn changes the average heating value of the fuel, resulting
in a change in the tonnage of coal burned by the industry.* It also
decreases energy consumption by rail transit systems since low sulfur
coal is not transported to mideast locations.
Capacity Mix in the Base Year
Table J-4 defines the geographical regions used in reporting
capacity mix and other industry characteristics. Table J-5 shows the
electrical generating capacity as of December 1975, included in the
data base. Two key scenario variables involved in projecting this
base-year capacity mix to any future year are the electrical demand
growth rates that apply to each region and the future fractions used
in adding new units once the files of announced units for a given
state have been exhausted. State-level growth rates are scaled from
the national average values shown in Table J-2 according to population
*The amount of coal that must be combusted to yield 1 kWh of electrical
energy is given by the unit's heat rate (a way of expressing thermal
conversion efficiency) divided by the coal's heating value. It takes
about 1 Ib of coal to produce one 1 kWh from a modern coal-fired
boiler. Variations in tonnage burned among the control scenarios
considered here were found to be insignificant (less than 1 percent
variation).
J-8
-------
TABLE J-4
DEFINITION OF GEOGRAPHIC REGIONS
New England (NE)
CT
RI
MA
NH
VT
ME
Mid-Atlantic (MA) S. Atlantic (SA) E.N. Central (ENC)
NY
PA
NJ
DE
MD/DC
VA
WV
NC
SC
GA
FA
WI
MI
IL
IN
OH
Central (ESC)
KY
IN
MS
AL
W.N. Central (WNC)
NO
SD
NB
IA
MO
MN
W.S. Central (WSC)
TX
OK
AR
LA
N. Mountain (NM)
ID
MT
WY
S. Mountain (SM)
NV
UT
CO
AZ
Nil
Pacific (PA)
WA
OR
CA
NOTE: The first seven and the last region are identical to Bureau of the
Census regions.
T_
J
-------
I
w
H
en
u
O
O
O
C
o
r- QJ
4-> C
5 "5
-0 S_
e 3
0 1
o
S-
>,
re
to
CJ3
r
o
o
(_ }
s_
to
QJ
r»
0
Z3
^^
03
C
0
f
a
QJ
a:
i VO O CO CO CO
i oo en r>. CM o
CM VD CO 00 «* CO
^- CM CO CO CM i
CD O O CM CM r
i «d" o «j- in o
i O CO "si" CD CO
r
LO r*- o i cn i
CM IO VO CO LO CO
CO «±
O O vo r^* en co
O O r r CM VO
ir> co co co co CM
i CM f-v *tf- r CD
t CM r
vo in co CM en CM
O CO VO CM t-v. CO
r <^" VO CM i~
o to tn co i ^t-
ID O CO CO CM CO
1
C_> C_3 CO
LU ^
CO VO
CM
i o r->. o «*
C 0 CM CO CO
ir> vo
CM CD "=± i r*^
co o ^- en r
. oo
vo cn co *^~ en
CM co co r en
r- CO
00 CO ^J-
it
O CD CD CM i
^j-
C
0
f
CJ +.1
oo 2: s:
a
Q.
1
QJ
E
C
QJ
"*"*
4-
O
^r^
VO
cn
crt
r*
R!
O
cu
I
10
01
4J
t
-r-
-Q
(O
Q.
(O
U -
to QJ
O QJ 3
*r~ t ^ [ *
1 * ^ ^_
c 3t to
i- C
t- -a > c
QJ C
"O
>r_
S- i- S-
O QJ 4->
t- E <->
E QJ
i =5 r
1 10 LU
CM
H- C
QJ O O
i to
_0 QJ -r
CO CT)-O
H- 10 LU
S-
QJ O) Q)
QJ > _C
oo o: -«->
(O jQ
w
J^
o
J_
OJ
c
OJ
01
.
QJ
to
QJ
-5
^* '
c
o
~4J>
to
.a
E
O
o
r
fO
c
01
-4_)
c
^B
+
OJ
O
O
"O
o
Ql
U
J-10
-------
growth. States whose growth is projected to be higher or lower than
the national level will have higher or lower demand growth rates,
respectively, with the scaling being done so as to maintain the orig-
inally specified national energy demand (or average power) growth.*
Average compound growth rates for the periods 1976-1985 and 1986-1995
derived by this process are given in Table J-6.
TABLE J-6
SCALED ENERGY DEMAND GROWTH RATES, BY REGION
(Average compound growth rate in percent per year)
Moderate Growth Scenarios
Region3
NE
MA
SA
ENC
ESC
WNC
WSC
NM
SM
PA
Nation
1976-1985
5.6
5.5
6.2
5.6
6.0
5.2
5.6
4.6
6.3
5.7
5.8
1986-1995
3.4
3.3
3.7
3.3
3.4
3.0
3.3
2.8
3.8
3.4
3.4
High Growth Scenarios
1976-1985
5.6
5.5
6.2
5.6
6.0
5.2
5.6
4.6
6.3
5.7
5.8
1986-1995
5.5
5.4
5.9
5.4
5.5
5.1
5.4
4.8
5.9
5.5
5.5
See Table J-4.
There is no tractable decision rule for predicting the proportions
of future units built beyond the base year planning horizon that will
be nuclear. Therefore, future-mix fractions are specified exogenously
to the model. These may be made to vary with the emission control
scenario, or held constant. The fractions used in deriving the
*A national peak growth rate has less physical meaning, since the time
of peak power demand varies widely across the country.
J-ll
-------
results presented may be found in "Effects of Alternative New Source
Performance Standards for Coal-Fired Electric Utility Boilers on
the Coal Markets and on Utility Capacity Expansion Plans" by ICF,
Inc., 1978. Coal assignments and coal-unit dispatch orders, by
regulatory category, were also taken from data contained in this
source. The approach used was to hold constant the amount (mega-
wattage) of nuclear capacity in 1995, i.e., independent of both the
post-1985 growth rate and the control scenario. The rationale used
to justify keeping the nuclear capacity the same under the moderate
and high growth cases is that a variety of regulatory and other
constraints are operating which would hinder the acceleration of
nuclear building schedules beyond those currently envisioned by 1990,
and that the amount of building assumed is already set at an opti-
mistic level. The reason for not attempting to quantify the shift to
nuclear units that might occur as a result of the imposition of more
stringent emission standards on coal units beyond 1990 is related:
this issue is too complex to model realistically, at least within the
context of the current study, because: (1) many of the important
determinants of a utility's decision whether to "go nuclear" are not
quantifiable (e.g., the expected licensing period); 2) those measures
that are quantifiable in principle, such as relative power generating
costs, are impossible to predict accurately in the 1990 time frame
due to great uncertainties in the cost data.
J-12
-------
Projections to 2000
Capacity mixes for the two baseline scenarios (Ml.2(0)0.1 and
HI.2(0)0.1) and the two 90-percent control scenarios (Ml.2(90)0.1 and
HI.2(90)0.1) are shown in Table J-7. Note that although the total
capacity in 1995 does not vary when the more stringent SO controls
are imposed, there is a slight increase in nuclear capacity with a
corresponding decrease in coal-fired capacity. Note again that this
shift is not due to conclusions about the relative economics of coal
vs. nuclear generation in the future. One factor that does operate
is that capacity penalties incurred when FGD systems are applied to
coal-fired units, are, over the long term, partially compensated for
by increased nuclear capacity. Note that the tabular values are net
"capability," i.e., generating capacity after reductions due to all
pollution control devices are taken into account; these may add to
10 percent of the "nameplate" capacity. The values for coal capacity
under the more stringent control scenarios would increase by roughly
5 percent in 1995. Since nuclear units have only water pollution
controls, the nameplate capacities of the nuclear units would increase
by a smaller fraction, and would be independent of the SO controls
ass ume d.
Two other aspects of the planning algorithms used by the model
to create new capacity, once the announcements data file for a given
state has been exhausted bear upon the amount of nuclear capacity
J-13
-------
r--
I
W
ffl
cn
O
M
w
o
co
O
W
H
U
W
hJ
W
CO
pi
o
X
E-'
M
O
o
W
H
U
W
Pi
P-I
ro
S
CD
re
CD-
re
o
QJ
, .
in
en
cn
i
.
*>.
cn
^3.
c
o
r
i
-o
"O
CI-
LD
cn
cn
CO
cn
r
23
s
-J
<:
re
-Q
re
oc
_c
**
1
o
cn
ro
ro
0
.--P
CD
CM
en
CM
cn
CO
un
,_"
CD
VD
ro
o
.
I"-
-
**
^
CO
CM
in
co
CM
ro
CO
CM
in
co
CM
1
O
o
ro
CM
ro
1
ro
CM
ro
i
ro
CM
ro
i
ro
CM
ro
i
in
CM
*~
in
CM
i
in
CM
r
»
in
CM
'
cn
in
*~
cn
in
cn
in
i
cn
in
1
CO
«=c
cn
e&
1
O
0
in
cn
0
cn
o
CO
vo
CD
co
VO
f-»
CD
i
VO
o
r~-
o
o
CO
o
o
*
CO
o
o
CO
o
o
CO
o
o
CO
o
0
CO
UJ
1
o
o
CO
0
0
cn
ro
CM
CO
CO
i^-
in
CD
CM
in
vo
CO
m
0
cn
CM
VD
CO
cn
p*^
co
vo
CO
r-
vo
co
<
VD
00
1
VD
CO
0
a:
Q
>
rc
o
ID
VO
o
cn
in
co
ro
CM
vo
ro
co
o
,
CD
r
0
CD
CO
f m
CO
r*^
co
oo*
VD
cn
co
o
cn
vo
CO
co
vo
UJ
T^
i i
CO
o:
^
l
o
VO
,
o
._!
o
VD
._"
o
VD
._!
o
cn
^1
O
cn
,_!
o
cn
-
O
cn
i
CD
r--.
,_!
o
._!
o
i
,_"
o
i
._!
_i
^~
cr*
UJ
rc
l
0
UJ
CJ3
l~^
t^J-
ro
cn
ro
ro
co
ro
ro
CO
in
ro
r--
00
CO
cn
ro
CO
r-^
ro
in
vo
ro
cn
ro
i
r^
CO
I
co
ro
1
co
ro
'
a:
<^
UJ
t
(__}
i
-Z
,_!
CD
co
ro
CD
00
in
i"-^
in
o
vo
co
o
ro
cn
o
ro
«
CM
oo
o
.
ro
CO
o
^
?3~
r-.
VD
«^j-
f^
in
<3-
"^
^
cj*
*"""
w_,l
ct
1
O
t
QJ
4->
re
i-
(O
CL
J_
Ol
O)
QJ
re
r
o
J-
(O
Q.
i.
O)
QJ
CM
CO
cn
4-^
10
O
a.
c
o
re
>
o
QJ
i-
CM
0
CO
4->
c
ai
u
i.
QJ
a.
* ^ CD
QJ cn
c
r S^
QJ *->
oo 2
ro o
-Q i-
cn
r~~
M CD
3 4-)
o ro
C71 QJ +J
T3-r-
QJ O E
4-^ E *i
ro
i.
QJ ro
-a o
o
E o
r
C
3
CM
oo
.__
M
O
CL
c:
C)
re
6
C"
a3
$..
CM
o
+j
c
cu
o
i.
OJ
o.
-CD
QJ cn
c
r JC
QJ 4->
X
ro o
JO i-
cn
-M .c: 4
0 -r- 1
S- .C -i
CTt- -* r
^1 CO
cn 'CD
*| *
-C O
o
O
o cn
CM CM CM CM
o
o '-
o
o cn
n
tQ
QJ
cn
QJ
M n
J3 CO
J-14
-------
added in the 1990s.* New baseload units are added in discrete sizes,
not in the exact amount of capacity needed to bring the reserve
margin up to the minimum.** Secondly, the specified future fractions
are used only in a probabilistic sense. For example, specifying that
70 percent of the post-1985 capacity built in a New England state
will be nuclear, is interpreted by the model as a 7 out of 10
probability that a new unit will be nuclear. As a result, the
exact amount of nuclear capacity installed by any year beyond 1985
cannot be "clamped" exogenously. This reflects planning uncertainties
in the real world, and complicates the process of isolating the im-
pacts of changing standards applied to fossil-fueled units. Aggregate
nuclear capacity can be adjusted by trial and error. This adjustment
was made for the HI.2(90)0.03 scenario, in which the initial model
runs produced higher nuclear values than shown in these results.
Table J-8 shows capacity mix by geographic region for the
baseline scenario with moderate growth (Ml.2(0)0.1). The last column
gives the capacity additions over the years 1985-1995, the period
over which new builds are determined primarily by the future-mix
fractions.
*Announced units are not necessarily put into operation on the date
the utility has projected: units are deferred if the specified
demand growth does not justify operation until a later date. It is
assumed, however, that construction schedules may not be shortened,
and combustion turbines are built in the short term if more announced
units are available at a later date.
**The sizes are: nuclear - 1,200 MW, coal - 600 MW, and oil - 500 MW.
J-15
-------
TABLE J-8
PROJECTED CAPACITY MIX, BY REGION, FOR THE BASELINE
SCENARIO UITTI MODERATE GROWTH
(Net generating capability, Gigawatts)
New England
COAL
OIL & GAS
COMB CYCLE
HYDRO
TURBINE
GEOTHERMAL
NUCLEAR
TOTAL
Mid Atlantic
COAL
OIL & GAS
COMB CtfCLE
HYDRO
TURBINE
GEOTHERMAL
NUCLEAR
TOTAL
South Atlantic
COAL
OIL & GAS
COMB CYCLE
HYDRO
TURBINE
GEOTHERMAL
NUCLEAR
TOTAL
1976
0.58
11.9
0.08
2.53
1.52
0.0
4.14
2.0.8
1976
20.3
24.5
0.13
6.69
10.6
0.0
7.38
69.6
1976
41.2
22.4
0.0
5.74
9.10
0.0
9.67
88.1
1985
1.34
13.2
0.31
2.53
2.28
0.0
9.87
29.5
1985
24.5
25.0
0.13
8.05
13.1
0.0
21.5
92.3
1985
53.8
24.1
1.40
9.72
15.1
0.0
27.5
131.6
1995
3.97
13.1
0.31
2.53
2.28
0.0
20.0
42.2
1995
38.2
24.1
0.13
8.05
13.1
0.0
51.6
135.2
1995
80.6
23.6
1.40
9.78
15.5
0.0
72.0
202.9
1985-1995
2.63
-0.10
0.0
0.0
0.0
0.0
10.13
12.70
1985-1995
13.7
-0.9
0.0
0.0
0.0
0.0
30.1
42.9
1985-1995
26.8
-0.5
0.0
0.06
0.4
0.0
44.5
71.3
J-16
-------
TABLE J-8 (Continued)
PROJECTED CAPACITY MIX, BY REGION, FOR THE BASELINE
SCENARIO WITH MODERATE GROWTH
(Net generating capability, Gigawatts)
East North Central
COAL
OIL & GAS
COMB CYCLE
HYDRO
TURBINE
GEOTHERMAL
NUCLEAR
TOTAL
East South Central
COAL
OIL & GAS
COMB CYCLE
HYDRO
TURBINE
GEOTHERMAL
NUCLEAR
TOTAL
West North Central
COAL
OIL & GAS
COMB CYCLE
HYDRO
.TWINE
GEOTHERMAL
NUCLEAR
TOTAL
1976
69.1
7.28
0.0
3.13
6.80
0.0
8.69
95.0
1976
30.5
4.32
0.0
5.98
2.69
0.0
2.30
45.8
1976
18.5
4.96
0.0
3.00
4.81
0.0
4.00
35.3
1985
89.5
8.62
0.0
3.17
11.7
0.0
28.8
141.8
1985
37.0
4.63
0.0
7.53
3.06
0.0
18.0
70.2
1985
31.7
2.77
0.09
4.00
7.88
0.0
6.28
52.7
1995
104.3
8.59
0.0
3.17
12.0
0.0
79.3
207.4
1995
44.6
4.33
0.0
7.53
3.13
0.0
39.0
98.6
1995
42.7
2.74
0.09
4.00
8.50
0.0
14.0
72.0
1985-1995
14.80
-0.03
0.0
0.0
0.30
0.0
50.5
65.6
1985-1995
7.60
-0.30
0.0
0.0
0.07
0.0
21.0
28.4
1985-1995
11.0
-0.03
0.0
0.0
0.62
0.0
7.72
19.3
J-17
-------
TABLE J-8 (Concluded)
PROJECTED CAPACITY MIX, BY REGION, FOR THE BASELINE
SCENARIO WITH MODERATE GROWTH
(Net generating capability, Gigawatts)
West South Central
COAL
OIL & GAS
COMB CYCLE
HYDRO
TURBINE
GEOTHERMAL
NUCLEAR
TOTAL
N. Mountain
COAL
OIL & GAS
COMB CYCLE
HYDRO
TURBINE
GEOTHERMAL
NUCLEAR
TOTAL
S Mountain
COAL
OIL & GAS
COMB CYCLE
HYDRO
TURBINE
GEOTHERMAL
NUCLEAR
TOTAL
Pacific
COAL
OIL & GAS
COMB CYCLE
HYDRO
TURBINE
GEOTHERMAL
NUCLEAR
TOTAL
1976
2.78
57.1
0.23
2.32
2.78
0.0
0.90
66.1
1976
1.27
0.07
0.0
3.15
0.08
0.0
0.0
4.57
1976
12.8
3.50
0.23
3.14
1.75
0.0
0.0
21.4
1976
1.37
21.4
0.57
31.7
1.77
0.32
3.39
60.5
1985
19.6
57.0
0.46
2.57
5.83
0.0
9.03
94.5
1985
3.03
0.07
0.0
3.89
0.60
0.0
0.0
7.59
1985
19.5
3.26
0.51
3.67
2.73
0.0
0.67
30.3
1985
4.51
20.7
5.08
41.0
6.56
1.72
16.2
95.8
1995
58.4
25.6
0.46
2.57
12.4
0.0
34.0
133.4
1995
5.27
0.07
0.0
3.93
0.61
0.0
0.0
9.88
1995
27.1
2.18
0.51
3.67
2.83
0.0
7.73
44.0
1995
12.3
20.5
5.08
42.6
7.66
1.93
47.0
137.0
1985-1995
38.8
-31.4
0.0
0.0
6.57
0.0
25.0
38.9
1985-1995
2.24
0.0
0.0
0.04
0.01
0.0
0.0
2.29
1985-1995
7.6
-1.08
0.0
0.0
0.10
0.0
7.06
13.7
1985-1995
7.8
-0.2
0.0
1.6
1.1
0.21
30.8
41.3
J-18
-------
The age distribution of coaL-fired units is particularly impor-
tant in this study since the NSPS revisions are applied only to those
units that come into operation in 1983 or later: Key dates are:*
Year On-Line Applicable Category of Emission Standards
1976 or earlier State Implementation Plan (SIP)
1977-1982 New Source Performance Standards
1983 or later Revised New Source Performance Standards
Table J-9 gives the age breakdown by category in 5-year intervals
from 1980 to 1995. In 1985, emission changes due to a NSPS revision
affect only 11.5 percent of net generating capacity. With the lower
growth rate, less than half of the coal-fired capacity would be
subject to the revised standard by 2000, the last year of the simu-
lation. With higher growth, the fraction in that year is 69 percent.
Table J-10 gives projections of flue gas desulfurization
(scrubber) capacities for all the SO control variants, assuming
the revised particulate limit of 12.9 ng/J (0.03 lb/10 Btu). The
numbers listed under "Capacity of FGD Systems" are measures of the
size of the scrubbers, not of the units being scrubbed. These two
capacities can differ, because the pollution control module allows
for partial scrubbing of the flue gas. More specifically, full
scrubbing is assumed only when the required SO removal equals or
exceeds 90 percent. Less than 90 percent removal is achieved by
*The Clean Air Act stipulates that the revised new source standard
will apply to those units whose construction commences after publi-
cation of the proposed revision. The definition of "commence
construction" is somewhat at issue, and construction periods vary:
a fixed year of implementation, 1978, is assumed.
J-19
-------
TABLE J-9
PROJECTED COAL-FIRED CAPACITY BY REGULATORY CATEGORY
(Net generating capability, Gigawatts)
Moderate Growth Scenarios
Year
1980
1985
1990
1995
2000
Year
1980
1985
1990
1995
2000
SIP
Units
206.8
(87.4%)
212.5
(74.7%)
212.1
(59.7%)
212.1
(50.8%)
212.1
(43.5%)
Units
206.9
(87.4%)
212.6
(74.7%)
212.3
(50.3%)
212.3
(35.3%)
212.3
(25.8%)
NSPS
Units
29.8
(12.6%)
39.3
(13.8%)
39.3
(11.1%)
39.3
(9.4%)
39.3
(8.1%)
High Growth
Units
29.8
(12.6%)
39.3
(13.8%)
39.3
(9.3%)
39.3
(6.5%)
39.3
(4.8%)
Revised
NSPS
Units
0.0
(0%)
32.7
(11.5%)
103.9
(29.2%)
166.0
(39.8%)
236.4
(48.5%)
Scenarios
Revised
NSPS
Units
0.0
(0%)
32.7
(11.5%)
170.8
(40.4%)
349.7
(58.2%)
570.1
(69.4%)
> . ,
Total
237
285
355
417
488
Total
237
285
422
601
822
J-2n
-------
TABLE J-10
PROJECTED COAL CAPACITY USING FLUE GAS DESULFURIZATION
(Net capability, Gigawatts)
Scenario
Ml. 2(0)0.1
Ml. 2(90)0. 03b
(Ml, 2(90)0,1)
HI. 2(0)0.1
Year
1985
1990
1995
2000
1985
1990
1995
2000
1985
1990
1995
2000
. Unit Category
SIP and NSPS
All
SIP and NSPS
All
SIP and NSPS
All
SIP and NSPS
All
SIP and NSPS
Revised NSPS
All
SIP and NSPS
Revised NSPS
All
SIP and NSPS
Revised NSPS
All
SIP and NSPS
Revised NSPS
All
SIP and NSPS
All
SIP and NSPS
All
SIP and NSPS
All
SIP and NSPS
All
Generating
Capacity
285
285
355
355
417
417
487
487
252
31.2
283
252
94.8
347
252
151
403
252
212
464
285
285
423
423
602
602
822
822
Capacity of
FGD Systems3
52.2
52.5
61.3
61.3
67.1
67.1
75.1
75.1
38.7
35.5
74.2
38.7
106
145
38.7
168
207
38.7
236
275
45.6
45.6
59.4
59.4
76.5
76.5
99.7
99.7
J-21
-------
TABLE j-rio (continued). PROJECTED COAL CAPACITY USIfiG
FLUE GAS DESULFURIZATION
(Net capability, Gigawatts)
Scenario
b
HI. 2(90)0. 03
(HI. 2(90)0.1)
ML 7(80)0. 03b
Year
1985
1990
1995
2000
1985
1990
1995
2000
Unit Category
SIP and NSPS
Revised NSPS
All
SIP and NSPS
Revised NSPS
All
SIP and NSPS
Revised NSPS
All
SIP and NSPS
Revised NSPS
All
SIP and NSPS
Revised NSPS
All
SIP and NSPS
Revised NSPS
All
SIP and NSPS
Revised NSPS
All
SIP and NSPS
Revised NSPS
All
Generating
Capacity
243
39.2
282
243
167
410
243
336
579
243
543
786
252
40.3
292
252
94.7
347
252
151
403
252
212
464
Capacity of
FGD Systems3
31.0
34.0
65.0
31.0
185.0
216.0
31.0
372.0
403.0
31.0
602.0
633.0
38.7
31.6
70.3
38.7
93.3
132
38.7
149
188
38.7
209
248
J-22
-------
TABLE J-10 (Concluded),
PROJECTED COAL CAPACITY USING
FLUE GAS DESULFUR1ZATION
(Net capability, Gigawatts)
Scenario Year Unit Category
HI. 2(80)0. 03b 1985 SIP and NSPS
Revised NSPS
All
1990 SIP and NSPS
Revised NSPS
All
1995 SIP and NSPS
Revised NSPS
All
2000 SIP and NSPS
Revised NSPS
All
MO. 5(0)0. 03 1985 SIP and NSPS
Revised NSPS
All
1990 SIP and NSPS
Revised NSPS
All
1995 SIP and NSPS
Revised NSPS
All
2000 SIP and NSPS
Revised NSPS
All
Generating
Capacity
243
40.5
282
243
167
410
243
336
579
243
543
786
252
40.5
292
252
95.9
348
252
152
404
252
212
464
Capacity of
FGD Systems
30.0
30.1
61.0
30.0
161.
191
30.0
321
351.
30.0
527.
557.
39.7
31.6
71.3
38.7
93.5
132.
38.7
149.
188.
38.7
208.
247.
See text.
Differences in these results for the two different particulate scenarios
are insignificant.
J-23
-------
scrubbing less than 100 percent of the gas at 90 percent removal
efficiency. The capacity of the FGD system for an individual
boiler is, therefore, the generating unit's nameplate capacity, times
the fraction of the gas scrubbed (a number between 0.3 and 1.0).*
This is a measure of the design size of the scrubber module. Finally,
note that the figures reported under "Generating Capacity" are the net
of all pollution control related capacity penalties: this explains
why the FGD capacities exceed the "Revised NSPS" net generating capa-
cities for units subject to the 90-percent removal requirement.
The _+20-percent variations in the FGD capacity numbers for SIP
and NSPS units in 1985 are due to differences among the scenarios in
TABLE J-ll
REGIONAL BREAKDOWN OP INSTALLED FGD CAPACITY
IN 1995, SCENARIO HI.2(90)0.03
(Gigawatts)
3
Region
NE
MA
SA
ENC
ESC
WNC
WSC
NM
SM
PA
Nation
Net Coal-Fired Capacity
6.75
59.2
109.0
158.0
54.4
50.6
87.5
13.9
32.4
25.2
597.0
FGD Capacity
6.52
42.5
75.5
81.5
23.2
31.5
84.1
13.9
24.5
20.1
403.0
aSee Table J-4.
*Coal sulfur values are adjusted downward to the compliance level
whenever less than 30 percent SO removal is required to comply
with the applicable limit.
J-24
-------
the sulfur levels of coal assigned to the units. The sulfur content
and region of origin of coals used in all the scenarios were derived
from the output of a coal supply model given in IGF, Inc. (1978c).
Table J-ll shows a regional breakdown of installed FGD capacity
in 1995 under the high growth scenario with the 90-percent removal
requirement (HI.2(90)0.03).
Salient features of these results are:
Given the coal assignments used in this analysis, and the
present SO emission limitations, installed FGD capacity
would amount to approximately 17 percent of net coal-fired
generating capacity by 1985, remaining at approximately
that level for the following decade.
Under the high growth scenario, the 80-percent standard
increases the installed FGD capacity from 59 GW to 191 GW
by 1990, and from 77 GW to 351 GW by 1995.
Increasing the removal requirement from 80 to 90 percent
increases the installed FGD capacity by 10 to 15 percent
by 1995, depending upon the post-1983 growth rate.
The amount of FGD installed in response to a revised NSPS
standard of 220 ng/J (0.5 lb/10 Btu) is nearly the same
as that projected under the 80 percent removal scenario.
The regions of the country with the highest installed
scrubber capacities by 1995, assuming a 90-percent removal
requirement, are West South Central (84 GW), East North
Central (82 GW), and South Atlantic (76 GW). These three
regions contain 60 percent of the total installed scrubbing
capacity in that year.
As indicated in the previous tables, projections of capacity
mix and scrubber usage are not sensitive to a revision of the current
new source performance standard for particulates. This is because
the coal assignments, capacity penalties, future-mix fractions and
J-25
-------
dispatch orders remain invariant. The costs of control do increase -
but not enough to affect these key determinants of industry behavior.
The small cost increase is due primarily to the assumption that units
burning low sulfur western coals would use fabric filters rather than
precipitators to comply with new source standards (IGF, 1978c).
Given the control costs used in this study, elimination of this
assumption might substantially increase the costs of meeting the
revised particulate standard, with some noticeable differences in the
industry projections.
Projections of utility coal consumption are shown in Figure J-l
for the high growth scenarios. (Variations in coal consumption due
to the different SO control assumptions are too small to be signi-
ficant.) The curve starts in 1976 at 404 million metric tons (445
tons), which is very close to the actual utility "burn" in that
year of 406 million metric tons (448 million tons) as reported by the
Federal Powir Commission (1977)." The curves illustrate rather
dramatically the substantial difference a few percent change makes in
the assumed demand growth rate. A 2-percent increase in the national
average compound growth rate after 1985 (from 3.4 to 5.5 percent)
results in almost double the coal consumed in the year 2000 from 910
metric tons to 1670 metric tons.
The simulation model accounts for differences between coal mined
and coal burned due to tonnage loss in coal preparation plants. (Changes
*The FPC reports both deliveries and consumption. These may differ
in any given year due to changes in stockpile levels.
J-26
-------
2000 1
1800 '
4>
U
O
U
1600
1400
1200
"HIGH GROWTH" SCENARIOS
g 1000
CL
800
600
400 -
"MODERATE GROWTH"
1975
1980
1985
Year
1990
1995
2000
FIGURE J-i
PROJECTIONS OF ELECTRIC UTILITY COAL CONSUMPTION
J-27
-------
in utility stockpiles arc not considered.) Given the sulfur levels
of the coals assigned and assumptions about the minimum sulfur levels
that are available in uncleaned coals, the model projects that
an additional 32 million metric tons of coal would have had to be
mined in 1976 to account for refuse from producing 114 million metric
tons (126 million tons) of clean coal, assuming dense media separa-
tion processes with 80 percent weight recovery. This compares with
the U.S. Bureau of Mines estimate of 79 million metric tons (87
million tons) of steam and metallurgical coals that were cleaned by
dense media separation processes in 1975 (U.S Department of the
Interior, Bureau of Mines, 1977b).
J-28
-------
REFERENCES
Abelson, H. and F. W. Lowenbach, 1977. Procedures Manual for Envi-
ronmental Assessment of Fluidized Bed Combustion Processes. MITRE
Corporation, McLean, Virginia.
Aerospace Corporation, 1977. Preliminary Draft-The Solid Waste
Impact of Controlling SC>2 Emission from Coal-Fired Steam Gen-
erators, Volume II. EPA-600/7-78-044b, March 1978 (Volume I:
Executive Summary EPA-600/7-78-044a, March 1978). Prepared for
the Office of Research and Development, U.S. Environmental Protection
Agency. Washington, D.C.
American Petroleum Institute. 1976. Basic Petroleum Data Book.
Washington, D.C.
Averitt, 1973. Coal. United States Mineral Resources. Geological
Survey Professional Paper 820. U.S. Department of the Interior,
Geological Survey. Washington, D.C.
Aynsley, E. and M. R. Jackson, 1971. Industrial Waste Studies: Steam
Generating Plants - Draft Final Report. Freeman Laboratories, Inc.
Rosemont, 111.
Babcock and Wilcox Co, 1972. Steam: Its Generation and Use. 38th
Edition.
Battelle Memorial Institute, 1977. Evaluation of Physical Coal
Cleaning as an S02 Emission Control Technique. EPA 600/7-78-034,
February 1978. Prepared for the U.S. Environmental Protection
Agency. Research Triangle Park, North Carolina.
Bechtel Corporation, 1977. Flue Gas Desulfurization Systems: Design
and Operating Parameters, S02 Removal Capabilities, Coal Properties
and Reheat, Volume II. EPA-600/7-78-030b, March 1978 (Volume I.
Executive Summary EPA-600/7-78-030a, March 1978). Prepared for
the Office of Research and Development, U.S. Environmental Protection
Agency. Research Triangle Park, North Carolina.
Blake, R. T. , 1970. Proper Feedwater Treatment Helps Minimize Pollu-
tion. Plant Eng. Vol. 34.
Christman, P. G., 1977. Water Recycle/Reuse Alternatives at the
(Pennsylvania Power and Light) Montour Station - Draft. Radian
Corporation. Austin, Texas.
K-l
-------
Donahue, J. M., 1970. Chemical Treatment. Ind. Water Eng. 7(5) :35.
Edison Electric Institute, 1977. Annual Electric Power Survey. EEI
Publication No. 77-29. New York, N. Y*
Foster Wheeler Corporation, 1976. Experience Sheets: Foster Wheeler
Corporation, Babcock and Wilcox Co., Riley Stoker Group, Combustion
Engineering, Inc. Prepared for the Office of Air Quality Planning
and Standards, U.S. Environmental Protection Agency. Research
Triangle Park, North Carolina.
Gilbert Associates and MITRE Corporation, 1977. Technical Document
on Applying Tentative New Source Performance Standards for Coal-Fired
Steam Generators to Fluidized Bed Combustors. Prepared for the
Division of Coal Conversion and Utilization Fossil Energy, U.S.
Department of Energy. Washington, D.C.
Guthman, W. and J. G. Nobleth, Jr., 1976. Water Recycle/Reuse Alter-
natives at the Public Service Station of Colorado Comanche Plant.
Radian Corporation. Austin, Texas.
IGF, Inc., 1978. Effects of Alternative New Source Performance
Standards for Coal-Fired Electric Utility Boilers on the Coal Markets
and on Utility Capacity Expansion Plans. Executive Summary.
Washington, D.C.
ICF, Inc., 1978a. Memorandum from Jerry Eyster to Tom Schrader.
ICF, Inc., 1978b. Memorandum from Hoff Stauffer to Tom Schrader.
Industrial Gas Cleaning Institute, Inc., 1977. Flue Gas Desulfuriza-
tion System Manufacturers Surveys. Stamford, Connecticut,
Kidder Peabody Company, 1977. A Status Report on Electric Utility
Generation Equipment, Fossil Boilers as of December 31, 1976.
Lowry, H. H., (Ed.), 1963. Chemistry of Coal Utilization Supplemen-
tary Volume. John Wiley and Sons, Inc. New York.
The Marley Company, 1969. Cooling Tower Fundamental and Application
Principles.
Marshall, W. L., 1972. Cooling Water Treatment in Power Plants. Ind.
Water Eng. 9(2):38.
Noblett, J. G., Jr., 1977. Water Recycle/Reuse Alternatives of the
Georgia Power Company Plant Bowen. Tech. Note 200-118-08, Radian
Project No. 200-118. Radian Corporation. Austin, Texas.
K-2
-------
PEDCo Environmental, Inc., 1977. Summary Report - Flue Gas Desul-
furization Systems - August - September 1977. Prepared for the U.S.
Environmental Protection Agency. Research Triangle Park, North
Carolina.
PEDCo Environmental, Inc., 1977. Particulate and Sulfur Dioxide
Emission Control Costs for Large Coal-Fired Boilers. EPA-45Q/3-
78-007. Prepared for the U.S. Environmental Protection Agency.
Cincinnati, Ohio.
PEDCo Environmental Inc., 1978. Effects of Alternative Sulfur
Dioxide New Source Performance Standards on FGD System Supply and
Demand. EPA-600/7-78-033. Prepared for the U.S. Environmental
Protection Agency. Research Triangle Park, North Carolina. March.
PEDCo Environmental5 Inc., 1978a. Flue Gas Desulfurization Systems
Capabilities for Coal-Fired Steam Generators, Volume II. EPA 600/
7-78-032b (Volume I. Executive Summary. EPA-600/7-78-032a).
Prepared for the Office of Air Quality Planning and Standards. U.S.
Environmental Protection Agency. Research Triangle Park, North
Carolina. March.
Radian Corporation, 1977a. The Effect of Flue Gas Desulfurization
Availability on Electric Utilities, Volume It. EPA-600/7-78-031b,
March 1978 (Volume I. Executive Summary. EPA-600/7-78~031b,
March 1978). Prepared for the U.S. Environmental Protection
Agency. Research Triangle Park, North Carolina.
Radian Corporation, 1977b. The Energy Requirements for Controlling
SO Emissions from Coal-Fired Steam Electric Generators. EPA-450/
3-77-05a, December 1977 (Executive Summary. EPA-450/3-77-050b).
Prepared for the U.S. Environmental Protection AGency, Research
Triangle Park, North Carolina.
Radian Corporation, 1977c. Water Pollution Impact of Controlling
Sulfur Dioxide Emissions from Coal-Fired Steam Electric Generators,
Volume II. EPA-600/7-78-045b, March 1978 (Volume I. Executive
Summary. EPA-600/7-78-045a, March 1978). Prepared for the U.S.
Environmental Protection Agency. Research Triangle Park, North
Carolina.
Rice, J. K. and S. D. Strauss, 1977. Water-Pollution Control in
Steam Plants. Power 120(4), Sl-20.
Shafer, Troxell and Howe Company (undated). Personal Communication
from B. Raymond to R. Douglas (MITRE Corp.).
Strauss, S. D., 1973. Water Treatment. Power 117(6). SI-S24.
K-3
-------
Teknekron, Inc., 1977. An Integrated Technology Assessment of Elec-
trical Utility Energy Systems, Volume II. Components of the Impact
Assessment Model. Berkeley, California.
Teknekron, Inc., 1977a. Preliminary Analysis of Alternative New
Source Performance Standards. Briefing Materials for U.S. Environ-
mental Protection Agency. Berkeley, California.
Teknekron, Inc., 1978. Review of New Source Standards for S02
Emissions from Coal-Fired Boilers. Vol. I: Emissions and Won Air-
Quality Impact Assessment. Berkeley, California.
Teknekron, Inc., 1978a. Draft - Review of New Source Standards for
S02 Emissions from Coal-Fired Boilers. Volume III: Economic/
Financial Assessment. Berkeley, California.
Temple, Barker and Sloane, Inc., 1976. The Economic Impact of EPA's
Air and Water Regulations on the Electric Utility Industry.
Temple, Barker and Sloane, Inc. 1978. Memorandum on NSPS Analysis
Results from Michael Allen, TBS, to Low Pugliaresi, EPA. April 18.
U.S. Congress, 1977. National Energy Act. U.S. House of Representa-
tives, 95th Congress, Document No. 95-138. Washington, B.C.
U.S. Department of Commerce, Bureau of the Census, 1967. City and
County Data Book, 1967. U.S. Government Printing Office, Washington,
D.C.
U.S. Department of Commerce, Bureau of the Census, 1974. Statistical
Abstract of the United States, 95th Edition. Washington, D.C.
U.S. Department of Commerce, Bureau of Economic Analysis, 1976.
Business Statistics. 19th Biennial Edition. U.S. Government
Printing Office. Washington D.C.
U.S. Department of the Interior, Bureau of Mines, 1975. Demonstrated
Coal Reserve Base of the United States by Sulfur Category, on January
1, 1974. Washington, D.C.
U. S. Department of the Interior, Bureau of Mines. Annual and
December, 1976. Petroleum Statement. Washington, D.C.
U.S. Department of the Interior, Bureau of Mines, 1977a. Demon-
strated Coal Reserve Base of the United States on January 1, 1976.
Washington, D.C.
K-4
-------
U. S. Department of Labor, Bureau of Labor Statistics, 1976.
Employ-ment and Earnings. U.S. Government Printing Office.
Washington, B.C.
U. S. Department of Labor, 1977. Relative Importance of Components
in the Consumer Price Index December 1976. Report 497. Bureau of
Labor Statistics. U.S. Government Printing Office. Washington, D.C.
U. S, Energy Research and Development Administration, 1977. Final
Environmental Impact Statement, Alternative Fuels Demonstration Pro-
gram. Vol. 1. Washington, D.C.
U.S. Environmental Protection Agency, 1974. Development Document for
Proposed Effluent Limitations Guidelines and New Source Performance
Standards for the Steam Electric Power Generating Point Source
Category. Office of Air and Water Programs. Effluent Guidelines
Division. Research Triangle Park, North Carolina.
U.S. Environmental Protection Agency, 1976. Health, Environmental
Effects and Control Technology of Energy Use. EPA-600/7-76-002.
Research Triangle Park, North Carolina.
U.S. Environmental Protection Agency, 1976a. National Air Quality
Emission Trends Report. EPA-450/1-76-001. Research Triangle Park,
North Carolina.
U.S. Environmental Protection Agency, 1976b. National Air Quality
and Emissions Trends Report, 1975. EPA-450/1-76-002. Research
Triangle Park, North Carolina.
U.S. Environmental Protection Agency, 1977a. Memorandum from Phillip
Youngblood to Stanley Cuffe. June 30, 1977. Research Triangle Park,
North Carolina.
U.S. Environmental Protection Agency, 1977b. Memorandum from Phillip
Youngblood to Stanley Cuffe. November 8, 1977. Research Triangle
Park, North Carolina.
U.S. Environmental Protection Agency, 1978. Physical Coal Cleaning
for Utility Boiler S02 Emisson Control. EPA-600/7-78-034.
Research Triangle Park, North Carolina.
U.S. Environmental Protection Agency. 1978a. First Interim Report:
Continuous Sulfur Dioxide Monitoring at Steam Generations. Emission
Measurement Branch, Emissions Standard and Engineering Division.
EMB-77-SPP 23A. Research Triangle Park, N.C.
K-5
-------
U.S. Environmental Protection Agency, 1978b. Emission Standard and
Engineering Division. Engineering notes furnished by private
communication. May 25.
U. S. Federal Energy Administration. November, 1974. Project
Independence. Washington, D.C.
U.S. Federal Energy Administration, 1977. Inventory of Power Plants
in the United States. FGA/G-77/202. Washington, D.C.
U.S. Federal Power Commission, 1971. The 1970 National Power Survey.
Washington, D.C.
U.S. Federal Power Commission, 1976. Statistics of Privately Owned
Electric Utilities in the United States, 1975. Washington, D.C.
U.S. Federal Power Commission, 1977. FPC News Release. No. 23017,
Washington, D.C.
U.S. Federal Power Commission, 1977a. Annual Summary of Cost and
Quality of Electric Utility Plant Fuels, 1976. Staff report by
Bureau of Power. Washington, D.C.
U.S. Federal Power Commission. 1977b. Status of Coal Supply Con-
tracts for New Electric Generating Units, 1976-1985. Washington,
D.C.
University of Montana, Bureau of Business and Economic Research,
1974. Projections of Economic Development Associated with Coal-
Related Activity in Montana. Missoula, Montana.
Watson, J.W., Jr. et al., 1977. Infrastructure and Highway Support
Needs of Regions Affected by Energy Activities. The MITRE Corpora-
tion, METREK Division. McLean, Va.
K-6
-------
4. H TLE AND SUBTITLE
Electric Utility Steam Generating Units Background
Information for Proposed S02 Emission Standards
TECHNICAL REPORT DATA
(Please read Instructions on the ceri-rse before
1. SEPCRT NO.
EPA 450/2-78-007a
3. RECIPIENT'S ACCESSION NO.
5. REPORT DATE
July. 1978
6. PERFORMING ORGANIZATION CODE
7 AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
MITRE Corporation
METREK Division
McLean, Virginia 22101
12. SPONSORING AGENCY NAME AND ADDRESS
U. S. Environmental Protection Agency
Office of Air Quality Planning and Standards (MD-13)
Research Triangle Park, North Carolina 27711
10 PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-02-2526 Task 7
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
15 SUPPLEMENTARY NOTES
16. ABSTRACT
The report discusses the legal alternatives to revising the standard of
performance for sulfur dioxide emissions from steam/electric generators with
heat inputs greater than 250 million BTU/hour. Alternative sulfur dioxide control
technologies are discussed. The environmental and economic impact of various
alternative sulfur dioxide standards are discussed, also.
The report contains 50 references to detailed technical reports discussing
all aspects of flue gas desulfurization at steam/electric generators.
17.
a.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COS AT l field/Group
18. DISTRIBUTION STATEMENT
Release Unlimited
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
497
20 SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form J220-1 (Rev. 4-77) PREVIOUS EDITION is OBSOL ETE
------- |