EPA-450/3-79-006
Control Techniques
for Carbon Monoxide Emissions
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
June 1979
-------
This report has been reviewed by the Emission Standards and Engineering Division of the Office of
Air Quality Planning and Standards, EPA, and approved for publication. Mention of trade names or
commercial products is not intended to constitute endorsement or recommendation for use. Copies of
this report are available through the Library Services Office (MD-35), U.S. Environmental Protection
Agency, Research Triangle Park, N.C. 27711, or from the National Technical Information Services,
5285 Port Royal Road, Springfield, Virginia 22161.
Publication No. EPA-450/3-79-006
11
-------
TABLE OF CONTENTS
PAGE
1. Introduction and Summary
2. Characterization of Carbon Monoxide Emissions 2'1
2-1
2.1 Formation of Carbon Monoxide
2-3
2.2 Sources of Carbon Monoxide Emissions
2-5
2.3 Carbon Monoxide Emission Estimates and
Emission Factors
2.4 Carbon Monoxide Emission Trends and Projections 2-5
2-20
2.5 Sampling and Analytical Methods
. 4. i 3-1
3. Mobile Source Control
3.1 Background - Engine Design Variables 3-4
3.2 Description of Light Duty Vehicle, Light Duty 3-10
Truck, and Heavy Duty Truck Industry
3.3 Description of the Aircraft Industry 3-16
3.4 Vehicle CO Emission Standards 3"20
3-22
3.5 In-Use Experience
3-29
3.6 CO Emission Factors
3.6.1 The Effect of Cold Weather on CO Emissions 3-42
3.7 Carbon Monoxide Control for New Mobile Sources 3-43
3.7.1 Types of CO Controls for New Mobile Sources 3-45
3.7.2 Carbon Monoxide Emission Reduction Benefits 3-48
3.7.3 Costs for New Mobile Source Controls 3-48
3.7.4 Energy Requirements for New Mobile Source 3-51
;' Control s
-------
TABLE OF CONTENTS
PAGE
3.7.5 High Altitude Control for New Mobile 3-54
Sources
3.7.6 Environmental Impact of New Model Source 3-55
Controls
3.8 Carbon Monoxide Controls Applied to Vehicles After 3-56
Sale and Other Measures Available to States and/or
Local Governments
3.8.1 Inspection/Maintenance Control Techniques 3-57
3.8.1.1 Types of I/M Control Strategy 3-57
Approaches
3.8.1.2 Costs for I/M Programs 3-58
3.8.1.2 Benefits of I/M Programs 3-61
3.8.1.4 Energy Requirements for I/M Program 3-66
3.8.2 Transportation Control Programs 3-66
3.8.2.1 Transportation Control Strategy 3-69
Approaches
3.8.2.2 Emission Reduction Benefits of Trans- 3-74
portation Control Programs
3.8.2.3 Costs of Transportation Control 3-76
Programs
3.8.2.4 Energy Requirements of Transportation 3-77
Control Programs
-------
TABLE OF CONTENTS
PAGE
3.8.2.5 Environmental Impact of Transporta- 3-77
tion Control Programs
3 77
3.9 Special Bibliography for Chapter 3
3-79
3.9.1 Types of Control Techniques
3-79
3.9.2 Emission Reduction Benefits
3-80
3.9.3 Costs
3-80
3.9.4 Energy Requirements
3-80
3.9.5 Environmental Impacts
4. Stationary Internal Combustion Source Control 4-1
4-1
4.1 Process Description
- n • 4-1
4.1.1 Engine Design
4-3
4.1.2 Engine Applications
4-5
4.2 Emission Sources
4-7
4.2.1 Gas Turbine Engines
4-9
4.2.2 Spark Ignition Engines
4.2.3 Compression Ignition Engines
4.3 Emission Factors and Nationwide CO Emissions 4-15
4-15
4.3.1 Gas Turbine Engines
4.3.2 Reciprocating Internal Combustion Engines 4-17
, . 4-17
4.4 Control Techniques
4.4.1 Oxidation of CO in the Exhaust Gas 4-20
4.4.2 Design Changes and Operating Adjustments 4-22
m
-------
TABLE OF CONTENTS
PAGE
4.5 Economic, Environmental, and Energy Impacts of 4-23
Control Techniques
5. Stationary External Combustion Source Control 5-1
5.1 Utility and Large Industrial Boilers 5-1
5.1.1 Process Description 5-1
5.1.2 Process Emission Sources and Factors 5-3
5.1.3 Control Techniques 5-4
5.1.3.1 Automatic Excess Air Rate Control 5-7
5.1.3.2 Proper Firing Rate 5-7
5.1.3.3 Burner Maintenance 5-7
5.1.3.4 Reduced Fuel Consumption 5-7
5.1.4 Cost of Controls 5-8
5.1.5 Impact of Controls 5-8
5.1.5.1 Emission Reduction 5-8
5.1.5.2 Environment 5-9
5.1.5.3 Energy Requirements 5-11
5.2 Industrial Boilers 5-11
5.2.1 Process Description 5-11
5.2.2 Process Emission Sources and Factors 5-14
5.2.3 Control Techniques 5-14
5.2.4 Cost of Controls 5-14
5.2.5 Impact of Controls 5-16
5.2.5.1 Emission Reduction 5-16
-------
TABLE OF CONTENTS
PAGE
5-16
5.2.5.2 Environment
5-17
5.2.5.3 Energy Requirements
5 3 Residential, Commercial, and Institutional Heaters 5-17
5-17
5.3.1 Process Descriptions
5-17
5.3.1.1 Residential Heating
5-18
5.3.1.2 Commercial and Institutional Heating
5-18
5.3.2 Process Emission Sources and Factors
5-23
5.3.3 Control Techniques
5-24
5.3.3.1 Effect of Maintenance
5-25
5.3.3.2 Fuel Substitution
5-26
5.3.4 Cost of Controls
5-27
5.3.5 Impacts of Controls
5-27
5.3.5.1 Emissions Reduction
5-27
5.3.5.2 Environment
5-28
5.3.5.3 Energy Requirements
5-29
5.4 Solid Waste Incinerators
5-29
5.4.1 Municipal Incinerators
5-30
5.4.1.1 Process Description
5.4.1.2 Process Emission Sources and Factors 5-31
5-32
5.4.1.3 Control Techniques
5-33
5.4.1.4 Cost of Controls
5-33
5.4.1.5 Impact of Controls
-------
TABLE OF CONTENTS
PAGE
5.4.2 Commercial/Industrial Incinerators 5-34
5.4.2.1 Process Descriptions 5-35
5.4.2.2 Process Emission Sources and Factors 5-36
5.4.2.3 Control Techniques 5-37
5.4.2.4 Cost of Controls 5-38
5.4.2.5 Impact of Controls 5-38
6. Industrial Process Source Control Systems 6-1
6.1 Incinerators 6-1
6.1.1 Equipment and Design Parameters for Thermal
Incinerators 6-2
6.1.2 Equipment and Design Parameters for Catalytic 6-9
Incineration
6.1.3 Incinerator Control Efficiency 6-15
6.1.4 Applicability 6-16
6.1.5 Energy Requirements 6-16
6.1.6 Environmental Impact 6-25
6.1.7 Costs (Mid-1978 Dollars) 6-26
6.2 Carbon Monoxide Boilers 6-33
6.2.1 Equipment and Design Parameters for Carbon 6-33
Monoxide Boilers
6.2.2 Control Efficiency 6-34
6.2.3 Applicability 6-34
-------
TABLE OF CONTENTS
PAGE
fi OC
6.2.4 Energy Requirements
r OC
6.2.5 Environmental Impact
6.2.6 Costs (Mid-1978 Dollars) 6-36
6-36
6.3 Flares an* Plume Burners
7. Industrial Process Source Control
7.1 Carbon Black Industry
7-2
7.1.1 Process Description
7.1.2 Process Emission Sources and Factors 7~3
7.1.3 Control Techniques
7.1.3.1 CO Boilers 7"
7 8
7.1.3.2 Flares
7.1.3.3 Thermal Incinerator 7"8
7.1.3.4 Pellet Dryers
7.1.3.5 Catalytic Incinerator
7-10
7.1.3.4 Pellet Dryers 7"9
7-9
7-9
7.1.4 Cost of Controls
7.1.5 Impact of Controls
7.1.5.1 Emissions Reduction
7.1.5.2 Environment
7.1.5.3 Energy Requirements 7-11
7.2 Charcoal Industry
7-13
7.2.1 Process Description
7.2.1.1 Batch Process 7"13
-------
TABLE OF CONTENTS
PAGE
7.2.1.2 Continuous Process 7-15
7.2.2 Process Emission Sources and Factors 7-19
7.2.3 Control Techniques 7-20
7.2.3.1 Control of Batch Processes 7-20
7.2.3.2 Control of Continuous Processes 7-22
7.2.4 Cost of Controls 7-22
7.2.5 Impact of Controls 7-24
7.2.5.1 Emission Reductions 7-24
7.2.5.2 Environment 7-24
7.2.5.3 Energy Requirements 7-24
7.3 Organic Chemical Industry 7-25
7.3.1 Acrylonitrile 7-26
7.3.1.1 Process Description 7-26
7.3.1.2 Process Emission Sources and Factors 7-28
7.3.1.3 Control Techniques 7-29
7.3.1.4 Cost of Controls 7-31
7.3.1.5 Impact of Controls 7-33
7.3.2 Formaldehyde 7-35
7.3.2.1 Process Description 7-35
7.3.2.2 Process Emission Sources and Factors 7-37
7.3.2.3 Control Techniques 7-40
7.3.2.4 Cost of Controls 7-42
-------
TABLE OF CONTENTS
PAGE
7.3.2.5 Impact of Controls 7"42
7.3.3 Maleic Anhydride 7~44
7.3.3.1 Process Description 7~46
7.3.3.2 Process Emission Sources and Factors 7-48
7.3.3.3 Control Techniques 7~49
7.3.3.4 Cost of Controls 7"49
7.3.3.5 Impact of Controls 7-50
7.3.4 Phthalic Anhydride 7"51
7.3.4.1 Process Description 7~51
7.3.4.2 Process Emission Sources and Factors 7-55
7.3.4.3 Control Techniques 7"55
7.3.4.4 Cost of Controls 7~58
7.3.4.5 Impact of Controls 7~58
7.4 Iron and Steel 7"60
7.4.1 Basic Oxygen Furnace 7~61
7.4.1.1 Process Description 7~62
7.4.1.2 Process Emission Sources and Factors 7-62
7.4.1.3 Control Techniques 7-63
7.4.1.4 Cost of Controls 7~63
7.4.1.5 Impact of Controls 7~66
7.4.2 Blast Furnace 7"66
7.4.2.1 Process Description 7"67
-------
TABLE OF CONTENTS
PAGE
7.4.2.2 Process Emission Sources and Factors 7-68
7.4.2.3 Control Techniques 7-68
7.4.2.4 Cost of Controls 7.70
7.4.2.5 Impact of Controls 7.70
7.4.3 Submerged Electric Arc Furnace 7-70
7.4.3.1 Process Description 7.71
7.4.3.2 Process Emission Sources and Factors 7-71
7.4.3.3 Control Techniques 7-72
7.4.3.4 Cost of Controls 7-80
7.4.3.5 Impact of Controls 7-30
7.4.4 Direct Electric Arc Furnace 7-8!
7.4.4.1 Process Description 7-31
7.4.4.2 Process Emission Sources and Factors 7-82
7.4.4.3 Control Techniques 7-32
7.4.4.4 Cost of Controls 7.34
7.4.4.5 Impact of Controls 7.34
7.4.5 Gray Iron Cupola 7.35
7.4.5.1 Process Description 7.35
7.4.5.2 Process Emission Sources and Factors 7-86
7.4.5.3 Control Techniques 7-36
7.4.5.4 Cost of Controls 7.37
7.4.5.5 Impact of Controls 7.33
-------
TABLE OF CONTENTS
PAGE
7-89
7.4.6 Sintering Furnace
7 ftQ
7.4.6.1 Process Description '"^
7.4.6.2 Process Emission Sources and Factors 7-89
7.4.6.3 Control Techniques 7'90
7.4.6.4 Cost of Controls 7"91
7.4.6.5 Impact of Controls 7'91
7-93
7.5 Petroleum Refining
7-QA
7.5.1 Catalytic Cracking
7.5.1.1 Process Description and Emissions 7-94
7.5.1.2 Control Techniques 7-103
7.5.1.3 Cost of Controls 7"104
7.5.1.4 Impact of Controls 7"106
7.5.2 Fluid Coking 7"108
7.5.2.1 Process Description and Emissions 7-108
7.5.2.2 Control Techniques 7""m
7-11?
7.5.2.3 Cost of Controls ' "*
7 119
7.5.2.4 Impact of Controls '~n£-
7 11 ^
7.5.3 Sulfur Plants ' MJ
7.5.3.1 Process Description and Emissions 7-113
7.5.3.2 Control Techniques 7'120
7.5.3.3 Cost of Controls 7"121
7.5.3.4 Impact of Controls 7-121
XI
-------
TABLE OF CONTENTS
PAGE
7.6 Primary Aluminum Industry 7-123
7.6.1 Process Description 7-123
7.6.1.1 Prebake Anode Cell 7-126
7.6.1.2 Soderberg Cells 7-127
7.6.2 Emission Sources and Factors 7-130
7.6.2.1 Potline Emissions 7-130
7.6.2.2 Anode Bake Furnaces 7-134
7.6.2.3 Miscellaneous Sources 7-135
7.6.3 Control Techniques 7-135
7.6.3.1 Thermal Incinerators 7-136
7.6.3.2 Potline Off Gas Recycle 7-136
7.6.4 Cost of Controls 7-137
7.6.5 Impact of Controls 7-137
7.6.5.1 Emissions Reduction 7-137
7.6.5.2 Environment 7-137
7.6.5.3 Energy Requirements 7-138
7.7 Pulp and Paper Industry 7-138
7.7.1 Process Description and Emission Factors 7-138
7.7.1.1 Process Description - Kraft Pulping 7-138
7.7.1.2 Emissions 7-143
7.7.2 Control Techniques 7-145
xn
-------
TABLE OF CONTENTS
PAGE
7-146
7 7.3 Cost of Controls
7-147
7.7.4 Impact of Controls
n A 4.- 7~147
7.7.4.1 Emissions Reduction
7-147
7.7.4.2 Energy Requirements
7-147
7.7.4.3 Environment
xm
-------
LIST OF TABLES
PAGE
2-1 Dissociation of C02 to CO 2-2
2-2 Bond Energies of Some Simple Chemical Substances 2-3
2-3 Summary of 1977 Nationwide Carbon Monoxide Emissions 2-6
From All Sources - 106 Metric Tons Per Year
2-4 Summary of 1977 Nationwide Carbon Monoxide Emissions 2-7
From Transportation Sources - 103 Metric Tons Per Year
2-5 Summary of 1977 Nationwide Carbon Monoxide Emissions 2-8
From Vehicles - 103 Metric Tons Per Year
2-6 Summary of 1977 Nationwide Carbon Monoxide Emissions 2-9
From Combustion Sources - 103 Metric Tons Per Year
2-7 Summary of 1977 Nationwide Carbon Monoxide Emissions 2-11
From Industrial Sources - 103 Metric Tons Per Year
2-8 Summary of 1977 Nationwide Carbon Monoxide Emissions 2-13
From Solid Waste Disposal and Wildfires - 103 Metric
Tons Per Year
2-9 EPA Uncontrolled Carbon Monoxide Emission Factors for 2-14
Selected Stationary Sources
2-10 Carbon Monoxide Emission Trends, 1970-1977 2-19
3-1 Light Duty Vehicle, Light Duty Truck and Heavy Duty 3-12
Vehicle Factory Sales From U.S. Plants
3-2 New Vehicle Registrations 3_14
3-3 Motor Vehicles in Use by Age as of July 1, 1977 3-15
xiv
-------
LIST OF TABLES
PAGE
O TO
3-4 Summary of Aircraft Engine Regulations
3-5 Commercial Aircraft Source CO Emissions as a Percentage 3-19
of Total Air Quality Control Region Emissions
3-6 Federal Vehicle Exhaust Emission Standards for CO 3-23
3-7 California Vehicle Exhaust Emission Standards for 3-24
Light Duty Vehicles
3-8 Federal Vehicle Exhaust Emission Standards for CO: 3-25
Heavy Duty Gasoline and Diesel Vehicles
3-9 California Vehicle Exhaust Emission Standard for CO: 3-25
Heavy-Duty Gasoline and Diesel Vehicles
3-10 U.S. Vehicle Exhaust Emission Standards for Motor- 3-26
cycles - 50 States
3-11 Comparison of Exhaust Emission Levels Between the 49- 3-26
State, Low-Altitude Vehicles in the Restorative Main-
tenance and Emission Factors Programs
3-12 Effect of Engine Component Operation on Emissions 3-30
3-13 Exhaust Emission Rates for All Areas Except 3-32
California and High-Altitude - Light Duty Vehicles
3-14 Idle Emission Rates for All Areas Except California 3-32
and High-Altitude - Light Duty Vehicles
3-15 Exhaust Emission Rates for All Areas Except 3-33
California and High-Altitude - Light Duty Trucks
xv
-------
LIST OF TABLES
PAGE
3-16 Idle Emission Rates for All Areas Except California 3-34
and High-Altitude - Light Duty Trucks
3-17 Exhaust Emission Rates for All Areas Except California 3-35
and High-Altitude - Heavy Duty Gasoline Fueled Vehicles
3-18 Idle Emission Rates for All Areas Except California 3-35
and High-Altitude - Heavy Duty Gasoline Fueled Vehicles
3-19 Exhaust Emission Rates for All Areas Except California 3-36
and High-Altitude - Heavy Duty Diesel Fueled Vehicles
3-20 Idle Emission Rates for All Areas Except California 3-36
and High-Altitude - Heavy Duty Diesel Fueled Vehicles
3-21 Exhaust Emission Rates for All Areas Except California 3-37
and High-Altitude - Motorcycles
3-22 Idle Emission Rates for All Areas Except California 3-37
and High-Altitude - Motorcycles
3-23 Carbon Monoxide Control Techniques for New Mobile Sources 3-46
3-24 Light Duty Vehicle Emission Control Component Retail Cost 3-49
3-25 CO Control Costs for Different Federal Levels of Control 3-52
for New Gasoline Fueled Power Plants
3-26 Federal Regulations for Light-Duty Vehicle Fuel Economy 3-53
3-27 Characteristics of Idle Mode and Loaded Mode Testing 3-59
3-28 Distribution of the Types of Repairs Required for 3-60
Vehicles Failing Inspection
xvi
-------
LIST OF TABLES
PAGE
3-65
3-29 CO FTP Emission Levels and Emission Reductions in 1987
Due to I/M Program Implemented in 1982
3.30 Illustrative Transportation-Related Air Pollution Problems 3-70
3-31 Summary of Estimated Impacts for the Localized Prototype 3-72
Scenarios
3-32 Sugary of Estimated Impacts for the Regional Prototype 3-73
Scenarios
4_! Applications of Stationary Reciprocating 1C Engines 4-6
and Energy Production by Fuel Use Category
4.2 CO Emissions From Spark Ignition Reciprocating Engines 4-11
at Rated Load
4.3 CO Emissions From Compression Ignition Reciprocating 4-13
Engines at Rated Conditions
4.4 Composite CO Emission Factors for the 1971 Population 4-15
of Electric Utility Gas Turbines
4.5 Composite CO Emission Factors for the 1971 Population 4-16
of Electric Utility Gas Turbines on a Fuel Basis
4.6 CO Emission Factors for Reciprocating Internal Combustion 4-18
Engines
4.7 Estimated 1975 Nationwide CO Emissions from Installed 4-19
Reciprocating 1C Engines
4-8 Percent of Total 1975 Nationwide Emissions of NOX, CO, 4-24
and Hydrocarbons for Stationary Internal Combustion
Engines
xvii
-------
LIST OF TABLES
PAGE
5-1 CO Emission Factors for Utility and Large 5-5
Industrial Boilers
5-2 Summary of 1977 Nationwide Carbon Monoxide Emissions 5-6
From Utility and Large Industrial Boilers
5-3 Representative Effects of N0x Controls on CO 5-10
Emissions From Utility Boilers
5-4 Carbon Monoxide Emission Factors for Industrial 5-15
Boilers With Capacities of 3-30 MW
5-5 Carbon Monoxide Emission Factors for Residential, 5-19
Commercial, and Institutional Heating
5-6 Estimated 1977 Nationwide Carbon Monoxide Emission 5-20
From Residential and Commercial/Institutional Heaters
5-7 Estimated Annual Average Composition of Municipal 5-30
Refuse
5-8 Carbon Monoxide Emission Factors for Selected Commercial/ 5-36
Industrial Incinerators
5-9 Estimated 1977 Carbon Monoxide Emissions From 5-37
Commercial/Industrial Incinerators
6-1 Sample Annualized Cost Calculations for Thermal 6-31
Incineration (Mid-1978 Dollars)
6-2 Annualized Cost Bases 6-32
7-1 Typical Vent Gas Composition for Carbon Black 7-5
Furnace Oil Process
xvm
-------
LIST OF TABLES
PAGE
7-2 Characteristics of Off-Gases From a Herreshoff 7-23
Furnace Charcoal Plant
7-3 Mass Emission Estimates for Carbon Monoxide From 7-26
Four Organic Chemical Processes, 1977
7-4 Composition of Main Process Vent Gas From Acrylonitrile 7-32
Production via the Sohio Process
7-5 Energy Requirements for CO Emission Controls in 7-34
Acrylonitrile Production
7-6 Absorber Vent Gas Composition in the Mixed Oxide 7-39
Catalyst Process for Formaldehyde
7-7 Absorber Vent Gas Composition in the Silver Catalyst 7-40
Process for Formaldehyde
7-8 Model Plant Data for Formaldehyde Production With 7-43
the Silver Catalyst and Mixed Oxide Catalyst Processes
of Formaldehyde Production
7-9 Energy Requirements for CO Emission Controls in 7-45
Formaldehyde Production
7-10 Product Recovery Condenser Vent Gas Composition 7-48
in Maleic Anhydride Production
7-11 Typical Main Process Vent Gas Composition From 7-57
0-Xylene Based Phthalic Anhydride Production
7-12 Typical Main Process Vent Gas Composition From 7-57
Naphthalene-Based Phthalic Anhydride Production
xix
-------
LIST OF TABLES
PAGE
7-13 Energy Requirements for CO Emission Controls in 7-60
Phthalic Anhydride Production
7-14 Mass Emission Estimates for Carbon Monoxide From 7-61
the Iron and Steel Industry, 1977
7-15
Thermal Incineration Costs in Sinter Plants, 1978 7-92
7-16 Domestic Catalytic Cracking Capacity, 1978 7-94
7-17 EPA Estimated 1977 Uncontrolled CO Emissions From 7-98
U.S. Catalytic Cracking Units
7-18 Typical Operating Conditions for Fluid Catalytic 7-101
Cracking
7-19 Emission Rates From FCC Unit Regenerators, Before 7-104
and After CO Boiler
7-20 Current Domestic Fluid Catalytic Cracker Regeneration 7-105
Techniques (August 1978)
7-21 Effect of Controls on CO Emissions From FCC 7-106
Regenerators
7-22 Typical Fluid Coker Operating Conditions 7-110
7-23 CO Controls on Domestic Fluid Coking Units 7-111
7-24 Typical Claus Tail Gas Compositions 7_H6
7-25 Representative Tail Gas Compositions for the Beavon 7-119
Sulfur Removal Process
xx
-------
LIST OF TABLES
PAGE
7 1 ^1
7-26 Carbon Monoxide Emissions From Primary Aluminum
Production
. . 7-132
7-27 Average Potline Emissions
. . 7-134
7-28 Anode Furnace CO Emissions
7-29 Reported Compositions of Exhaust Gases From Two Gas- 7-145
Fired Lime Sludge Kilns
xxi
-------
LIST OF FIGURES
PAGE
2-1 Stationary Manmade Sources of Carbon Monoxide 2-4
2-2 Grab Sampling System for CO Collection 2-21
2-3 EPA Integrated Sampling Train for Carbon Monoxide 2-22
2-4 Nondispersive Infrared Gas Analyzer 2-23
3-1 Relationships of CO, N0x, and HC Emission Concentrations 3-2
and Air-Fuel Ratio
3-2 Estimates of Aircraft Source CO Emissions at Major 3-21
National Airports
3-3 Pass/Fail Outcomes of the Initial Test on 1975 and 3-28
1976 Vehicles
3-4 Past and Projected CO Emissions From Motor Vehicles 3-40
3-5 Past and Projected U.S. Vehicle Travel 3-41
3-6 Relative CO Violations Vs. Mean Temperature 3-44
3-7 Estimated Impacts for Nine Regional Scenarios in a 3-78
Large Urban Area: Regional Highway Fuel Consumption
4-1 Specific Emissions of CO as a Function of Load for Gas 4-8
Turbine-Powered Generators, Composite of Several Makes
and Models
4-2 Effect of Air/Fuel Ratio on Emissions From a Gasoline 4-10
Engine
4-3 Diesel Engine Part-Load Carbon Monoxide Emissions 4-14
5-1 Watertube Boiler 5-12
xxn
-------
LIST OF FIGURES
PAGE
5-2 Firetube Boiler 5~13
5-3 General Trend of Smoke, Gaseous Emissions, and Efficiency 5-22
Versus the Percent Excess Air for Oil-Fired Residential
Heaters
6-1 Steps Required for Successful Incineration of Dilute Fumes 6-6
6-2 Coupled Effects of Temperature and Time on Rate of 6-7
Pollutant Oxidation
6-3 Schematic Diagram of Catalytic Afterburner Using Torch- 6-10
Type Preheat Burner with Flow of Preheated Waste Stream
Through Fan to Promote Mixing
6-4 Volume of Catalyst/Volumetric Flow Rate of Waste Stream 6-12
6-5 Effect of Temperature on Catalytic CO Conversion 6-13
6-6 Effect of Exchanger Recovery Factor and Waste Gas 6-19
Temperature on Inlet Temperature to Thermal Incinerator
6-7 Thermal Incinerator Energy Requirements With No Heat 6-20
Recovery Oxygen From Outside Air
6-8 Thermal Incinerator Energy Requirements With No Heat 6-21
Recovery Oxygen From Waste Gas
6-9 Effect of Exchanger Recovery Factor and Waste Gas 6-22
Temperature on Inlet Temperature To Catalytic Incinerator
6-10 Catalytic Incinerator Energy Requirements With No Heat 6-23
Recovery Oxygen From Waste Gas
xxm
-------
LIST OF FIGURES
PAGE
6-11 Catalytic Incinerator Energy Requirements With No Heat 6-24
Recovery Oxygen From Outside Air
6-12 Installed Capital Costs for Thermal Incinerators 6-27
6-13 Installed Capital Costs for Catalytic Incinerators 6-28
6-14 Annualized Costs for Thermal Incinerators 6-29
6-19 Annualized Costs for Catalytic Incinerators 6-30
6-16 Installed Equipment Cost for Carbon Monoxide Boilers 6-37
6-17 Annual Costs for Carbon Monoxide Boilers 6-38
7-1 Flow Diagram for a Furnace Type Carbon Black Plant 7-4
7-2 Typical Missouri-Type Charcoal Kiln With Multiple 7-14
Exhaust Stacks
7-3 Typical Beehive Kiln 7-16
7-4 Exterior View of a Herreshoff Multiple Hearth Furnace 7-17
7-5 Cross Sectional View of a Herreshoff Multiple Hearth 7-18
Furnace, With Plume Burning
7-6 Flow Diagram for Acrylonitrile Production Via the Sohio 7-27
Process
7-7 Schematic Diagram for a Combination By-Product Incinerator/ 7-30
Absorber Vent Gas Thermal Oxidizer System
7-8 Flow Diagram for Silver Catalyst Process for Formaldehyde 7-36
Production
xxiv
-------
LIST OF FIGURES
PAGE
7-9 Flow Diagram for Mixed Oxide Catalyst Process for 7-38
Formaldehyde Production
7-10 Flow Diagram for Production of Maleic Anhydride From 7-47
Benzene
7-11 Flow Diagram for 0-Xylene Based Phthalic Anhydride 7-54
Process
7-12 Basic Oxygen Furnace With Open Hood and Gas-Cleaning 7-64
and Storage System
7-13 Basic Oxygen Furnace with Closed Hood and Gas-Cleaning 7-65
System
7-14 Schematic Diagram of a Blast Furnace 7-69
7-15 Open Furnace Controlled by a Venturi Scrubber 7-74
7-16 Semi-Enclosed Furnace Controlled by a Venturi Scrubber 7-75
7-17 Sealed Furnace Controlled by Venturi Scrubber 7-76
7-18 Roof Hood 7-78
7-19 Side Draft and Direct Evacuation Hoods 7-79
7-20 Direct Shell Evacuation (DSE) System Open Roof 7-83
7-21 Fluid Catalytic Cracking Unit 7-96
7-22 Thermoflor Moving-Bed Catalytic Cracker 7-97
7-23 Fluid Coking Process 7-109
7-24 Claus Sulfur Plants 7-115
7-25 Major Processing Phases in Primary Aluminum Reduction 7-124
xxv
-------
LIST OF FIGURES
PAGE
7-26 Prebake Reduction Cell, Schematic Arrangement 7-128
7-27 HSS Soderberg Cell, Schematic Arrangement 7-129
7-28 Flow Diagram for Primary and Secondary Emission 7-133
Control Systems
7-29 Kraft Pulping and Recovery Process 7-140
xxvn
-------
1. INTRODUCTION AND SUMMARY
The document "Control Techniques for Carbon Monoxide Emissions from
Stationary Sources (AP-65)," was published by the U.S. Environmental Protec-
tion Agency in March 1970. It was one of a series of documents which sum-
marized control technique information for criteria air pollutants. Section
108(b) of the Clean Air Act, as amended, 42 USC paragraph 7401 et. seq.,
instructs the Administrator to issue information on air pollution control
techniques simultaneously with the issuance of new or revised air quality
criteria, as follows:
"... the Administrator shall, after consulation with appropriate
advisory committees and federal departments and agencies, issue
to the States and appropriate air pollution control agencies
information on air pollution control techniques, which informa-
tion shall include data relating to the cost of installation
and operation, energy requirements, emission reduction benefits,
and environmental impact of the emission control technology.
Such information shall include such data as are available on
available technology and alternative methods of prevention and
control of air pollution. Such information shall also include
data on alternative fuels, processes, and operating methods
which result in elimination or significant reduction of emissions."
This control techniques document was written in compliance with Section 108(c),
which requires the Administrator to review, and where appropriate, modify
and reissue any air quality criteria or information on control techniques.
Thus, this document represents an updated and expanded version of AP-65.
1-1
-------
It includes information on significant stationary sources of carbon monoxide
emissions as well as available methods for control of carbon monoxide emis-
sions from mobile sources.
This document characterizes carbon monoxide emission sources and controls
for use by states in revising State Implementation Plans (SIP's). It is
intended for use by state and local air pollution control engineers to pro-
vide basic available information on carbon monoxide emissions from mobile
sources, stationary combustion sources, and industrial process sources.
Both demonstrated and feasible control strategies are presented for each
source. Information is also provided on emission reduction benefits, energy
requirements of controls, and annualized and operating costs of controls.
Chapter 2 of this document presents background information on carbon
monoxide formation and lists significant sources of CO. Recent estimates
of carbon monoxide emissions and current emission factors are listed. This
chapter also includes a brief discussion of sampling and analytical methods
for carbon monoxide.
Chapter 3 summarizes current measures to control carbon monoxide emissions
from mobile sources. Information is included which will assist states in devel-
oping transportation measures for CO control through State Implementation Plans
Chapter 4 (internal combustion) and Chapter 5 (external combustion) dis-
cuss methods for control of carbon monoxide emissions from combustion sources.
Chapter 6 describes the techniques employed to control carbon monoxide
emissions from industrial sources and gives information on the energy require-
ments, costs and environmental aspects of these techniques.
Chapter 7 describes the techniques used for control of specific indus-
trial sources and gives information on energy, cost, and environmental aspects.
1-2
-------
The control methods described in this document represent information
from many technical fields. The proper choice of a method of controlling
carbon monoxide emissions from a specific source depends on several factors
other than source characteristics. No attempt has been made in this docu-
ment to review all the possible combinations of control techniques that may
be used.
1-3
-------
-------
2. CHARACTERIZATION OF CARBON MONOXIDE EMISSIONS
Most of the material presented in this section provides background infor-
mation on manmade carbon monoxide sources and emissions. Information on
natural sources of carbon monoxide is given in a companion document "Air
Quality Criteria for Carbon Monoxide," revised 1979. Also included is infor-
mation on the formation of carbon monoxide and a description of standard
methods for analysis of source and ambient CO concentrations.
2.1 FORMATION OF CARBON MONOXIDE
Carbon monoxide is formed as an intermediate product of reactions between
carbonaceous fuels and oxygen.1 When less than the theoretical amount of
oxygen required for complete combustion is supplied, CO is a final product of
the reaction. Under these conditions, CO concentrations may exceed 50,000 ppm,
Formation of the oxides of carbon is a simple process only when pure
carbon and pure oxygen are involved. The burning of carbonaceous fuels, in
general, is a very complicated process involving formation of CO before C02
is formed.1 If the temperature of combustion is high enough, dissociation of
the C02 begins:
C0 * CO + 0
2-1
-------
Table 2-1 shows the percentage dissociation of C02 to CO as a function
of temperature.
TABLE 2-1
DISSOCIATION OF C02 TO CO
TEMPERATURE
727°C
1127°C (2,060°F)
1527°C (2,780°F)
1627°C (2,960°F)
1727°C
PERCENTAGE DISSOCIATION
2 x 10
1.5x10
5.5 x 10
1 .0
5.0
-5
-2
-1
Source: Reference 2
Actually, CO is a very stable substance at high temperature, as indi-
cated by Table 2-1. In order for a chemical reaction to take place, chemical
bonds must be broken and formed. Bond energies are a measure of the diffi-
culty in breaking a chemical bond. Table 2-2 indicates a higher bond energy
for CO than for acetylene, which is notorious for its stability at electric
arc temperatures; CO is indeed known to be stable at very high temperature.
Conversely, propane is easily cracked or decomposed at moderate temperatures,
and the bond energy is seen to be low. The bond energy for C02 is moderately
low, and experience shows that it is not difficult to remove an atom of oxy-
gen from C02 by dissociation to form CO. For these reasons then, a second
mechanism of CO formation is high-temperature dissociation of C02, or
2-2
-------
hindering of the combination of CO and oxygen by virtue of temperature. Thus,
raising the temperature increases the concentration of CO in the thermodynamic
sense.
TABLE 2-2
BOND ENERGIES OF SOME SIMPLE CHEMICAL SUBSTANCES
SUBSTANCE BOND BOND ENERGY, Kcal/mol
n t~fi ~l
Carbon monoxide C-0
Carbon dioxide 0 — C-0 128
Propane C3H7-H 9
Acetylene HC == CH
Source: Reference 3
The reaction rates increase with temperature. Increase of oxygen con-
centration tends to decrease the CO concentration by affording a greater
chance for collision of CO and oxygen molecules (actually, hydroxyl radicals)
to form C02-1
2.2 SOURCES OF CARBON MONOXIDE EMISSIONS
There are numerous manmade sources of carbon monoxide emissions. These
sources can be categorized into two broad groups-mobile and stationary.
Figure 2-1 shows a breakdown of the stationary sources of carbon monoxide
which are investigated in this report. Chapter 3 discusses in detail the
sources of CO within the mobile category. Chapters 4, 5, and 7 investigate
the sources within the stationary source category. These sources were
chosen based upon their contribution to the total yearly tonnage of carbon
monoxide emissions in the U.S.
2-3
-------
External combustion
Solid waste incineration-
Reciprocating internal
combustion engines
Carbon black industry
Charcoal industry
Chemical industry
Iron and steel industry
Petroleum refining
Primary aluminum industry
Pulp and paper industry-
Utility & large industrial boilers
Industrial boilers
Residential & commercial heaters
Municipal incinerators
Industrial/commercial incinerators
Acrylonitrile
Formaldehyde
Maleic anhydride
Phthalic anhydride
Basic oxygen furnaces
Blast furnaces
Electric arc furnaces
Grey iron cupolas
Sintering plants
Catalytic cracking
Fluid coking
Sulfur plants
Sulfur recovery furnaces
Lime kilns
FIGURE 2-1. STATIONARY MANMADE SOURCES OF CARBON MONOXIDE
2-4
-------
2.3 CARBON MONOXIDE EMISSION ESTIMATES AND EMISSION FACTORS
EPA estimates of 1977 nationwide CO emissions are given in Tables 2-3
through 2-8. As these tables indicate, a wide variety of transportation,
combustion, industrial, and solid waste disposal sources contribute to the
total mass emissions of carbon monoxide. Table 2-3 shows that about 83 per-
cent of all nationwide CO emissions are from transportation sources. As
shown in Table 2-4, about 90 percent of the CO emissions from transportation
sources are from motor vehicles. CO emissions from gasoline powered passen-
ger cars comprise about 63 percent of the CO emissions from motor vehicles
(Table 2-5). CO emissions from combustion, industrial, and solid waste dis-
posal categories are on the order of hundreds of thousands of metric tons as
compared with millions of metric tons from motor vehicles.
Table 2-9 summarizes EPA uncontrolled carbon monoxide emission factors
for various stationary sources. Chapter 3 includes information on emission
factors for mobile sources.
2.4 CARBON MONOXIDE EMISSION TRENDS AND PROJECTIONS
Nationwide carbon monoxide emission estimates have been made by the
EPA's Office of Air Quality Planning and Standards for the years 1970
through 1977.4 These estimates are presented in Table 2-10.
Although the categories are not as detailed as the ones in Tables 2-3
through 2-8, they are sufficient to show the year-to-year trends in CO emis-
sions for the recent past. These estimates were made by EPA from internally
consistent sets of data based on current emissions factors.4
As Table 2-10 indicates, changes in annual CO emission estimates are
slight for the period 1970 through 1977. Emission estimates for highway
2-5
-------
TABLE 2-3
SUMMARY OF 1977 NATIONWIDE CARBON MONOXIDE EMISSIONS^
FROM ALL SOURCES - TO6 METRIC TONS PER YEAR
(106 tons/yr)
SOURCE CO EMISSIONS
Transportation 85.7 ( 9^4.5)
Combustion 1.2 ( 1.3)
Industrial 8.3 ( 9.2)
Sol id Waste Disposal
and WiIdfires 7.5 ( 8.3)
Total Emissions 102.7 (113-3)
Source: Reference k
Does not include carbon monoxide from natural sources
2-6
-------
TABLE 2-4
SUMMARY OF 1977 NATIONWIDE CARBON MONOXIDE EMISSIONS FROM
TRANSPORTATION SOURCES - 103 METRIC TONS PER YEAR
(103 tons/yr)
SOURCE
Motor Vehicles^
Ai rcraft
Commercial
General Aviation
Mi 1i tary
Rai1 roads
Diesel and Distillate
Residual Oil
Coal
Vessels
Residual Oil
Diesel Oil
Gasoline
Coal
Off-Highway Use Farm
Tractors
Gasoline
Diesel
Other Farm Equipment
Gasoline
Diesel
Construction
Gasoline
Diesel
Snowmobiles
Smal1 Uti1ity Engines
Heavy Duty Engines
Gasoline
Diesel
Motorcycles
Total Mobile Source Emissions
*see Table 2-5 for breakdown of
Source: Reference 4
CO EMISSIONS
151-3 ( 166.8)
261.8 ( 288.6)
238.5 ( 262.9)
259-5 ( 286.0)
0.7 ( 0.8)
10.2 ( 11.2)
0.7 ( 0.8)
29.8 ( 32.9)
(1613-2)
( 5.3)
2179.2 (2404.2)
111.8 ( 123.2)
232.4 ( 256.2)
5.6 ( 6.2)
734.5 ( 809.6)
223.1 ( 245.9)
109.0 ( 120.2)
1262.8 (1392.0)
1125.9 (1241.1)
51.3 ( 56.6)
87.5 ( 96.5)
emi ssions.
2-7
77170.6 (85066.0)
651.6 ( 718.3)
270.4 ( 298.0)
1498.8 ( 1652.2)
2291.0 ( 2527.4)
238.0 ( 262.4)
957.6 ( 1055.5)
109-0 ( -120.2)
1262.8 ( 1392.0)
1177.2 ( 1297.7)
87.5 ( 96.5)
85714.5 (94486.2)
-------
TABLE 2-5
SUMMARY OF 1977 NATIONWIDE CARBON MONOXIDE
EMISSIONS FROM VEHICLES - 103 METRIC TONS PER YEAR
(103 tons/yr)
SOURCE TYPE
Gasoline
Passenger Cars
Light Duty Trucks
Heavy Duty Trucks
Motorcycles
Total Gasoline
CO EMISSIONS
Urban
Rural
Total
38,315 (42,235) 10,147 (11,185) 48,462 (53,420)
8,726 ( 9,619) 2,231 ( 2,459) 10,957 (12,078)
9,937 (10,954) 5,973 ( 6,584) 15,910 (17,538)
476 ( 525) 163 ( 180) 640 ( 705)
57,455 (63,333) 18,514 (20,408) 75,969 (83,741)
Heavy Duty Trucks -
Diesel
Total From Motor
Vehicles
494 ( 545) 708 ( 780) 1,202 ( 1,325)
57,949 (63,878) 19,221 (21,188) 77,171 (85,066)
Source: Reference 4
2-8
-------
TABLE 2-6
SUMMARY OF 1977 NATIONWIDE CARBON MONOXIDE EMISSIONS FROM
COMBUSTION SOURCES - 103 METRIC TONS PER YEAR
(103 tons/yr)
SOURCE
Anthracite Coal
Electric Uti1ities
Industrial
Commercial-Insti tutional
Resident ial
Bituminous and Lignite Coal
Electric Uti1ities
Industrial
Commercial - Insti tutional
Res ident ial
Residual Oil
Electric Uti1ities
Industrial
Commercial - Insti tutional
Resident ial
Disti1 late Oi1
Electric Uti1ities
Industrial
Commercial - Insti tutional
Resident ial
Natural Gas
Electric Uti1i ties
Gas Pipelines and Plants
Industrial
Commercial - Institutional
Resident ial
Kerosene
I ndustri al
Heat i ng
Liquid Propane Gas
Industrial
Domest i c-Commercial
CO EMISSIONS
0.6 ( 0.7)
0.5 ( 0.5)
0.1 ( 0.1)
77.6 ( 85.5)
212.8 (234.6)
26.8 ( 29.5)
5.0 ( 5.5)
73.5 ( 81.0)
50.7 ( 55.9)
24.0 ( 26.5)
20.7 ( 22.8)
0(0)
5.6 ( 6.2)
9.5 ( 10.5)
17.3 ( 19-1)
38.8 ( 42.8)
22
395
52
( 25.0)
(436.0)
4 ( 57.8)
24.2
46.1
( 26.7)
( 50.8)
1.5 (
4.5 (
3.1 (
6.7 (
1.6)
5.0)
3.4)
7-4)
78.7 ( 86.8)
318.1 (350.6)
95.4 (105.2)
71.3 ( 78.6)
541.0 (596.3)
6.0 ( 6.6)
9.8 ( 10.8)
2-9
-------
TABLE 2-6 (Continued)
SOURCE
CO EMISSIONS
Wood
Process Gas
Bagasse
Total
Source: Reference 4
41.7 ( 46.0)
3.3 ( 3.6)
8.8 ( 9.7)
1174.1 (1294.2)
2-10
-------
TABLE 2-7
SUMMARY OF 1977 NATIONWIDE CARBON MONOXIDE EMISSIONS FROM
INDUSTRIAL SOURCES - 103 METRIC TONS PER YEAR
(103 tons/yr)
SOURCE CO EMISSIONS
Iron Foundries 1020.8 (1125.2)
Petroleum Refineries 2425.6 (2673-8)
FCC Units 2384.7 (2628.7)
TCC Units 40.4 ( 44.5)
Fluid Coking 0.5 ( 0.6)
Process Heaters 24.5 ( 27.0)
Oil-Fired 5.1 ( 5-6)
Gas-Fired 19-4 ( 21.4)
Asphalt Roofing 11 .9 ( 13-1)
Carbon Black 2184.2 (2407-7)
Gas 442.3 ( 487.6)
Oil 1741.9 (1920.1)
Channel 0 ( 0 )
Thermal 0(0)
Steelmaking 929-7 (1024.8)
Sintering 624.7 ( 688.6)
BOF 99-2 ( 109-4)
Electric Arc 205-7 ( 226.8)
Coke Production 43-7 ( 48.2)
Beehive 0.4 ( 0.4)
Byproduct 43.4 ( 47-8)
Kraft Pulp and Paper 1105-7 (1218.8)
Charcoal 97-3 ( 107-2)
Petrochemicals 481.1 (530.3)
Acetic Acid 8.2 ( 9-0)
Acrylonitrile 130.4 ( 143-7)
Cyclohexanol 39-0 ( 43-0)
Source: Reference 4
2-11
-------
TABLE 2-7 (Continued)
SOURCE
Dimethylterephthalate
Ethylene Dichloride
Formaldehyde
Maleic Anhydride
Phthalic Anhydride
Total Industrial Emissions
CO EMISSIONS
55.7 ( 61.4)
14.2 ( 15.7)
64.9 ( 71.5)
117.8 ( 129.9)
50.9 ( 56.1)
8324.4 (9176.1)
Source: Reference 4
2-12
-------
TABLE 2-8
SUMMARY OF 1977 NATIONWIDE CARBON MONOXIDE
EMISSIONS FROM SOLID WASTE DISPOSAL AND WILDFIRES
103 METRIC TONS PER YEAR (103 tons/yr)
SOURCE CO EMISSIONS
Municipal Incinerators '55.6 ( 171-5)
Conical Incinerators 530.7 ( 585.0)
Other Incinerators &55.0 ( 722.0)
Open Burning 1291.6(1*23.7)
Prescribed Forest Burning 1016.0 (1120.0)
Prescribed Agricultural Burning ^99.0 ( 550.0)
Forest Wildfires 3255-3 (3588.M
Structure Wildfires 135-5 ( 1*9.*)
Total Emissions 7538.7 (8310.0)
Source: Reference
2-13
-------
TABLE 2-9
EPA UNCONTROLLED CARBON MONOXIDE EMISSION FACTORS
FOR SELECTED STATIONARY SOURCES
SOURCE/FUEL TYPE
Boilers, Heaters, and Incineration
Bituminous Coal
Large Boilers [>29MW (>100xl06 Btu/hr)]
Intermediate Boilers
[3-29MW (10-100xl06 Btu/hr)]
Small Combustion Units
[<3MW (<10xl06 Btu/hr)]
Hand-Fired Units
Fi replaces
Lign ite
Pulverized Coal and Cyclone Units
Stoker Units
Anthracite Coal
Pulverized Coal
Traveling Grate Stokers
Hand-Fired Units
Fi replaces
Fuel Oil
Residual-Fired Large Boilers
[>73MW (>250xl06 Btu/hr)]
Residuel-Fired Small and
Intermediate Boilers
[0.15-73MW (0.5-250xl06 Btu/hr)]
Distillate-Fired Small and
Intermediate Boilers
[0.15-73MW (0.5-250xl06 Btu/hr)]
Domestic Units [<0.15MW (<0.5xl06 Btu/hr)]
Orchard Heaters
EMISSION FACTOR
0.5 kg/metric ton (l Ib/ton)
1 kg/metric ton (2 Ib/ton)
5 kg/metric ton (10 Ib/ton)
45 kg/metric ton (90 Ib/ton)
45 kg/metric ton (90 Ib/ton)
0.5 kg/metric ton (1 Ib/ton)
1 kg/metric ton (2 Ib/ton)
0.5 kg/metric ton (1 Ib/ton)
0.5 kg/metric ton (1 Ib/ton)
45 kg/metric ton (90 Ib/ton)
45 kg/metric ton (90 Ib/ton)
0.63 kg/103 1iter (5 lb/103 gal)
0.63 kg/103 liter (5 lb/103 gal)
0.63 kg/103 liter (5 lb/103 gal)
0.63 kg/103 liter (5 lb/103 gal)
2.8 kg/heater/hr (6.2 1b/heater/hr)
Source: Reference 5
2-14
-------
TABLE 2-9 (Continued)
SOURCE/FUEL TYPE
EMISSION FACTOR
Boilers, Heaters, and Incineration (Cont'd)
Natural Gas
Power Plant
Industrial Boilers
Domestic and Commercial Units
Liquid Propane Gas
Industrial Units
Domestic and Commercial Units
Liquid Butane Gas
Industrial Units
Domestic and Commercial Units
Wood
Wood and Bark Boilers
Wood Stoves
Fi replaces
Conical Incinerators
Solid Waste Incineration
Municipal Refuse
Multiple Chamber
Con Ical
Industrial/Commercial
Multiple Chamber
Single Chamber
Flue-Fed Single Chamber
Open Burning
General
Agricultural Waste
272 kg/106m3 (17 lb/106 ft3)
272 kg/106m3 (17 lb/106 ft3)
320 kg/106m3 (20 lb/106 ft3)
0.18 kg/103 liter (1.5 lb/103 gal)
0.23 kg/103 liter (1.9 lb/103 gal)
0.19 kg/103 liter (1.6 lb/103 gal)
0.24 kg/103 liter (2 lb/103 gal)
1-30 kg/metric ton (2-60 Ib/ton)
130 kg/metric ton (260 Ib/ton)
60 kg/metric ton (120 Ib/ton)
65 kg/metric ton (130 Ib/ton)
17.5 kg/metric ton charged
(35 Ib/ton)
65 kg/metric ton (130 Ib/ton)
5 kg/metric ton charged (10 Ib/ton)
10 kg/metric ton charged
(20 Ib/ton)
10 kg/metric ton charged
(20 Ib/ton)
42 kg/metric ton (85 Ib/ton)
16-154 kg/metric ton
(33-309 Ib/ton)
Source: Reference 5
2-15
-------
TABLE 2-9 (Continued)
SOURCE/FUEL TYPE
Boilers, Heaters, and Incineration (Cont'd)
Automobile Body Burning
Sewage Sludge Incineration
Reciprocating Internal Combustion Engines
Gasoline Engines
Small, 2-Stroke, General Utility
Small, 4-Stroke, General Utility
Farm Equipment (wheeled tractor)
Farm Equipment (non-tractor)
Heavy-Duty Construction Equipment
Industrial Engines
Diesel Engines
Farm Equipment (wheeled tractor)
Farm Equipment (non-tractor)
Heavy-Duty Construction Equipment
Industrial Engines
Natural Gas-Fueled
Heavy Duty Engines
Electric Utility Turbines
Gas-Fueled
Oil-Fueled
Industrial Process Sources
Asphalt Roofing Manufacturing
Asphalt Blowing
EMISSION FACTOR
1.1 kg/automobile (2.5 1b/
automobile)
Negligi ble
652 g/kWhr (486 g/hp-hr)
374 g/kWhr (279 g/hp-hr)
192 g/kWhr (143 g/hp-hr)
292 g/kWhr (218 g/hp-hr)
190-271 g/kWhr (142-202 g/hp-hr)
267 g/kWhr (199 g/hp-hr)
4.48 g/kWhr (3.34 g/hp-hr)
5.47 g/kWhr (4.08 g/hp-hr)
2.41-5.90 g/kWhr (1.80-4.40
g/hp-hr)
4.06 g/kWhr (3.03 g/hp-hr)
7020 kg/106m3 (430 lb/106 Ft3)
1842 kg/106m3 (115 lb/106 ft3)
1.85 kg/103 liter (15.4 lb/103 gal)
0.14 kg/metric ton asphalt
(0.27 Ib/ton)
Source: Reference 5
2-16
-------
TABLE 2-9 (Continued)
SOURCE/FUEL TYPE
EMISSION FACTOR
Industrial Process Sources (Cont'd)
Felt Saturation
Carbon Black Manufacturing, Furnace
Process
Charcoal Manufacturing
Chemical Industry
Ad i p i c Ac i d
Ammon i a
Lime Manufacturing
Phthalic Anhydride
0-Xylene Process
Naphthalene Process
Glass Manufacturing (melting furnace)
Iron and Steel Production
Basic Oxygen Furnace
Blast Furnace
Electric Arc Furnace
Cupola Furnace
Sinter Production
1.A5 kg/metric ton felt
(2.9 Ib/ton)
300 kg/metric ton product
(2600 Ib/ton)
60 kg/metric ton product
(320 Ib/ton)
58 kg/metric ton product
(115 Ib/ton)
100 kg/metric ton product
(200 Ib/ton)
1 kg/metric ton product
(2 Ib/ton)
151 kg/metric ton product
(301 Ib/ton)
50 kg/metric ton product
(100 Ib/ton)
0-0.2 kg/metric ton product
(0-0.5 Ib/ton)
70 kg/metric ton steel
(140 Ib/ton)
875 kg/metric ton pig iron
(1750 Ib/ton)
9 kg/metric ton steel
(18 Ib/ton)
72.5 kg/metric ton charge
(1^5 Ib/ton)
22 kg/metric ton product
(kk Ib/ton)
Source: Reference 5
2-17
-------
TABLE 2-9 (Continued)
SOURCE/FUEL TYPE
EMISSION FACTOR
Industrial Process Sources (Cont'd)
f-atroleum Refining
Fluid Catalytic Cracking Units
Moving Bed Catalytic Cracking Units
Pulp and Paper Manufacturing
Kraft Pulping
Recovery Furnace
Lime KiIns
39.2 kg/m3 feed (13,700 lb/103 bbl)
10.8 kg/m3 feed (3,800 lb/103 bbl)
1-30 kg/metric ton dried pulp
(2-60 Ib/ton)
5 kg/metric ton dried pulp
(10 Ib/ton)
Source: Reference 5
2-18
-------
PA PA \o
PA— —
en —
— CM LA
O ' — — '
CM PA v£>
— do
— CM CN
— ooorgrsi
— o o r-
en
i
o
en
— LACO
-3- O O O CT>
— OO OOOCNCNCN —
— — r-»
do c^
LA CM PA LA PA vO LAOO-JLA— OOO CO P-~ — —
_«r _f r-» — O O O CT\ PA CM CM — -3-3 PAO OO
J'PAvC LACMr^.CMPAOva-3- LAr^ •— — LA
— O O OOOfMfMCM — P^-3- PAO OO PA
I
CM
00
<
co
CO
PA PA O — O O OCOPMtMfM— JTLA JTO OO J"
PACMsD LAO(T>— — CT\CMCM PALA ef\ — PSI
d d en — do ooocMrMCMOvo^o -a-— oo rg
— TO ^ d.
*J .- O — 1/1
O-i C — 0) TD
— Q ^ — C
X 2 TO UJ —
2-19
-------
vehicles have increased about 6.4 percent from 1970 to 1977. Emission esti-
mates for other source categories have remained relatively stable.
Projections
Future nationwide CO emissions from stationary sources will depend in
large measure upon future Federal, State, and local air regulatory action.
Since this document and concurrent air quality criteria documents will pro-
vide an important basis for determining regulatory action, it is not possible
to make meaningful predictions of future nationwide CO emissions from station-
ary sources. The Clean Air Act, as amended in 1977, specifies a course of
action for future control of CO emissions from mobile sources. The effect
of the Act on future emissions from mobile sources is discussed in Chapter 3.
2.5 SAMPLING AND ANALYTICAL METHODS
Detailed information is available in the open literature on sampling
and analysis of carbon monoxide emissions. The following is a brief review
of the subject. Three categories of carbon monoxide monitoring are addressed;
(1) stationary source emissions, (2) mobile source emissions, and (3) ambient
air.
Two general methods of sample collection may be used for these monitor-
ing categories, grab (instantaneous) sampling and integrated (continuous)
sampling. The choice of collection method must be coordinated with the
analytical method which will be used to determine carbon monoxide content.
Grab samples for carbon monoxide measurement are typically taken using an
apparatus such as that shown in Figure 2-2. A certain quantity of gas is
pumped into a sample bag over a short time interval. This represents an
2-20
-------
instantaneous sample of the gas. A more representative grab sample may be
obtained by taking several such samples over several intervals and combining
them. An integrated or continuous sample can be taken using a sampling train
similar to the one shown in Figure 2-3. This is the EPA sampling train which
incorporates a gas conditioning section to remove moisture and carbon dioxide.6
This conditioning minimizes interferences with the analytical method. The
EPA analytical method for carbon monoxide is the non-dispersive infrared
(NDIR) analysis method.7
ROTAMETER-i
i
VALVE PI
AIR COOLED CONDENSER
PROBE
GLASS WOOL
RIGID
CONTAINER
Figure 2-2. Grab Sampling System for CO Collection
Analytical methods currently used to determine the quantity of carbon
monoxide present in a sample of gas are either instrumental methods or wet
chemical methods. Instrumental methods include non-dispersive infrared
(NDIR) and gas chromatographic analysis.
NDIR analysis, the EPA reference method, has been used to obtain most
of the available air quality data for carbon monoxide.7 This instrument
2-21
-------
relies on the principle of selective absorption of infrared radiation by car-
bon monoxide. Infrared radiation is passed through two parallel cells, a
sample cell containing the gas to be analyzed, and a reference cell. The
net radiation passing through the two cells is then passed into carbon monox-
ide detectors. Carbon monoxide present in the sample cell absorbs some of
the infrared radiation, reducing the amount of radiation reaching the detector
cell. The detector cell senses the difference in temperature and pressure
between the sample detector cell and the reference detector cell and produces
a signal corresponding to the concentration of carbon monoxide in the sample
gas. This system is shown in Figure 2-4. The NDIR instruments have a typical
minimum sensitivity of 20 ppm for carbon monoxide.
ICE BATH
ZERO
GAS
GAS
Figure 2-3. EPA Integrated Sampling Train for Carbon Monoxide
2-22
-------
LIGHT
CHOPPER
x
3
ir 'i
3
-»
MMi
MWM
•*
SAMPLE SAMPLE
Jo SAMPLE CELL fl p DETECTOR
->
-»
LIGHT
SOURCES
— ,, -» -+ -^
REFERENCE CELL
*
*
h-
-«- DIAPHRAGM
Figure 2-4. Nondispersive Infrared Gas Analyzer
Gas chromatographic analysis offers greater sensitivity than the NDIR
method, with measurement capabilities down to 0.05 ppm.7 This method in-
volves separation of carbon monoxide from methane using a molecular sieve.
The carbon monoxide is then quantitatively converted to methane (typically
using hydrogen gas over a nickel catalyst), and analyzed using a flame
ionization detector (FID). Other advantages over the NDIR method are that
the response to carbon monoxide is linear over the entire concentration
range and the method is specific to carbon monoxide.
Wet chemical analytical methods for carbon monoxide analysis depend
upon one of three classes of reactions: (1) reduction of a metal, (2) catalytic
oxidation to carbon dioxide, or (3) complexation.8 The reduction method is
attractive because of its simplicity although it is limited by the low
solubility of carbon monoxide in aqueous solutions and the long reaction
time required for completion of the reduction reaction. These factors limit
2-23
-------
the use of this analytical method to grab samples. This method is used as
the basis for carbon monoxide detector tubes. These tubes typically contain
silica gel impregnated with reagents which undergo a chemical change upon
reaction with carbon monoxide.
The oxidation methods of carbon monoxide analysis rely upon catalytic
oxidation to carbon dioxide.8 Two methods of analysis can be used, one
which determines the quantity of carbon dioxide produced; the other deter-
mines the quantity of species reduced by reaction with carbon monoxide.
Each of these methods requires certain species to be removed to minimize
interferences. The major advantage of the oxidation method is that a con-
tinuous integrated sample is used, insuring a more representative sample
than a grab sample. The drawbacks include the necessity of a complex sampl-
ing train to condition the sample gas or to remove possible interferences.
Complexation methods for certain carbon monoxide analysis rely on two
techniques, volumetric analysis, through absorption, or blood methods rely-
ing on the carbon monoxide—hemoglobin complex.8 The absorption method most
widely used is the Orsat analysis, which gives gas concentrations on a dry
basis. Orsat analysis relies upon an apparatus which exposes a known
quantity of gas to reagents which absorb oxygen, carbon dioxide, and carbon
monoxide. The volumetric change resulting from the absorption of these
species is read on a scale typically graduated in 0.2 percent increments
which can be read with reasonable accuracy to 0.1 percent.9'10 All of the
above methods lack the sensitivity necessary to measure low levels of carbon
monoxide, i.e., below 100 ppm. They are also limited to use on grab samples.
2-24
-------
The type of sampling technique and analytical method used for carbon
monoxide determination depends upon the category of monitoring being per-
formed, i.e., stationary sources, mobile source, or ambient, and the reason
for monitoring, i.e., compliance, background level determination, etc.
For stationary source monitoring, the NDIR method utilizing an inte-
grated sampling train is the EPA reference method.7 Gas chromatographic
analysis may also be performed on an integrated sample with better sensi-
tivity than NDIR. Grab samples may also be taken and analyzed by reduction
or complexation wet chemical methods, or either instrumental method.
For mobile source monitoring, i.e., exhaust gases from vehicles, NDIR
methods are primarily used for carbon monoxide analysis.7 The samples may
be either grab samples or integrated samples. Gas chromatography may also
be used with either sampling technique.
For ambient monitoring, the use of gas chromatographic analysis offers
greater sensitivity which is important if low levels of carbon monoxide are
expected.7
2-25
-------
REFERENCES FOR CHAPTER 2
1. Fristrom, R.M. The Mechanism of Combustion in Flames. Chem. Eng.
News 41(41):150-160. October 14, 1963.
2. Control Techniques for Carbon Monoxide Emissions from Stationary
Sources. Pub. No. AP-65, U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina, March 1970.
3. Handbook of Chemistry and Physics, 47th ed. Cleveland, Ohio. The
Chemical Rubber Co., 1966. 1856 p.
4. National Air Quality Monitoring and Emission Trends Report, 1977,
EPA-450/2-78-052, and supporting background information. U.S.
Environmental Protection Agency, Research Triangle Park, North
Carolina, December 1978.
5. Compilation of Air Pollutant Emission Factors, 2nd edition with
supplements. AP-42. U.S. Environmental Protection Agency, Research
Triangle Park, North Carolina. 1972 through 1977.
6. U.S. Environmental Protection Agency. Determination of Carbon Monox-
ide emissions from stationary sources. Fed. Reg. 38(111):15112, 1973.
7. Stern, Arthur C., ed. Air Pollution, Vol. 3, Measuring, Monitoring,
and Surveillance of Air Pollution, 3rd edition. Academic, New York,
1976.
8. Driscoll, John N. Flue Gas Monitoring Techniques. Manual Determination
of Gaseous Pollutants. Ann Arbor Science, Ann Arbor, Michigan. 1974.
2-26
-------
9. Perry, Robert H. et al., eds. Chemical Engineer's Handbook, 4th ed.
McGraw-Hill, 1969, pp. 9-12.
10. Curtin Mathison Scientific, Inc. Handbook of Scientific Instruments
and Laboratory Supplies. Catalogue #122-267. 1975. p. 469.
2-27
-------
-------
3. MOBILE SOURCE CONTROL
Estimates of 1977 nationwide emissions from mobile sources are given
in Tables 2-3, 2-4, and 2-5. As tnese tables snow, about 83 percent of all
nationwide emissions are from transportation sources. About 90 percent of
the CO emissions from transportation sources are from motor vehicles. Table
2-10 shows that CO emissions from transportation sources have increased
from 80.5xl06 metric tons (88.6x106 tons) in 1970 to 85.7xl06 metric tons
(94.3xl06 tons) in 1977.!
The relationship between CO emissions and air/fuel ratio is shown in
Figure 3-1. A simplified description of CO and C02 production during the
combustion process is shown in the following steps:
2C + 02 + 2CO
2CO + 02 -> 2C02
The first reaction proceeds at a much greater rate than the second.
Hydrogen in the fuel is oxidized to H20 quite easily, provided sufficient
oxygen and heat is available locally for combustion. Hydrocarbons (HC)
present in the fuel are also typically oxidized faster to CO than to C0?.
Poor distribution and mixing of fuel and air (which is more likely when
3-1
-------
I THEORETICAL AIR-FUEL RATIO
MAXIMUM CHEMICAL
EFFICIENCY
FIGURE 3-1. RELATIONSHIPS OF CO, NOX, AND HC EMISSION
CONCENTRATIONS AND AIR-FUEL RATIO
3-2
-------
fuel droplets rather than fuel vapors are present) can result in incomplete
combustion, and also produce CO that is emitted in the exhaust gas.
Either a chemically correct air/fuel mixture (stoichiometric) or an ex-
cess of air provides CO control. When the air/fuel ratio is richer than
chemically correct, substantial amounts of CO appear in the exhaust. When the
mixture is chemically correct or leaner than stoichiometric, CO concentration
usually does not drop to zero. This happens because of a combination of cycle-
to-cycle and/or cylinder-to-cylinder air/fuel charge maldistribution and slow
CO reaction kinetics. Fuel injection, better carburetion or better overall fuel
distribution are approaches to low CO emission from the engine. When a
hydrocarbon fuel is burned with an amount of air containing enough oxygen to
oxidize it completely, the following basic chemical reaction is assumed to
occur:2
1.00 CHK85 + 1.46 02 + 5.50 N2 ->
0.925 H20 + 1.00 C02 + 5.50 NZ
This chemical reaction assumes: 1) hydrocarbon fuels are accurately rep-
resented and contain an average of 1.85 hydrogen atoms for each carbon atom;
2) the volume ratio of nitrogen to oxygen in the air is 3.76:1; 3) the fuel
is burned completely to water and carbon dioxide; and 4) nitrogen is inert
and does not react with any other substances in the combustion chamber.
Assumptions 1 and 2 are quite true in practice, but the formation of HC, CO
and NOx in the combustion process indicates that assumptions 3 and 4 are not
wholly correct.
3-3
-------
Although the overall mixture is stoichiometric, local conditions at any
particular point in the combustion chamber may be far fcom;stOijchiometric.
An air-rich mixture (high air/fuel ratio) would provide excess air to par-
tially offset the increased CO emissions that result from poor distribution
and vaporization. A relatively large excess of air is normally found in
stratified charge engines, diesel engines, gas turbine engines, and some turbo-
charged gasoline engines. This accounts for the relatively low CO emissions
which can be found from these types of powerplants. Another factor that may
contribute to increased emissions is flame quenching at the relatively
cool combustion chamber boundaries. Quenching can occur ewen if the fuel is
perfectly vaporized and distributed throughout the combustion chamber. Gross
malfu i of the ignition or fuel induction systems can increase CO and
HC emissions from spark-ignition engines. A misfire, for example, allows
an entire air/fuel charge to be emitted into the exhaust system. A sticking
automatic choke system, or a restriction in the air intake system can also
have an adverse effect on the air/fuel ratio, generally increasing both CO
and HC emissions.
3.1 BACKGROUND - ENGINE DESIGN VARIABLES
It is often impossible to isolate the effect of any single design variable
or operating parameter on engine emissions. Some of these factors for spark-
ignition engines are included as follows:3
1) air-fuel ratio
2) load or power level
3) speed
3-4
-------
4) spark/injection timing
5) exhaust backpressure
6) valve overlap
7) intake manifold pressure
8) combustion chamber deposit build-up
9) surface temperature
10) surface to volume ratio
11) combustion chamber design
12) stroke to bore ratio
13) displacement per cylinder
14) compression ratio
In the following discussion of these design variables, HC and CO are
treated together because, once formed, both can be influenced by chemical
oxidation in either the cylinder or exhaust system if excess oxygen is
present. The importance of a lean air/fuel ratio for CO reduction is obvious,
and the gain in emission reduction by operating vehicles lean suggests the
importance of minimum carburetor/fuel injection tolerances and good manu-
facturing control to achieve uniform fuel distribution. Significant after-
reaction can occur in the exhaust system with lean overall air/fuel ratios
or with air injection when the average exhaust temperature exceeds 650°C (1200°F)
but after-reaction might not continue to lower emissions as the mixture be-
comes even leaner because the exhaust temperature can become too low to
achieve a significant reaction rate.
3-5
-------
At a fixed air/fuel ratio there is no effect of power output on carbon
monoxide emission concentration. However, the mass emission of CO will in-
crease directly with increasing power output and air consumption. Therefore,
a smaller, lighter vehicle may have the advantage of lower CO mass emission
due to its reduced power demand to drive the cycle, all other things being
equal. However, all other things are generally not equal, especially when
the standards are based on mass.
Increased exhaust port turbulence at higher engine speeds promotes ex-
haust system oxidation reactions through better mixing. This promotes after-
oxidation of the quenched layer and one would expect mass emissions of HC to
decrease with an increase in speed; however, the decrease will be less than
expected because of the increased flow volume required to overcome higher
engine friction at higher speeds. Speed has no effect, however, on CO con-
centration because oxidation of CO in the exhaust is kinetically limited
rather than mixing limited at the normal exhaust temperatures.
Spark retard has little effect on CO concentration except at very re-
tarded timing where the lack of time to complete CO oxidation leads to in-
creased CO emissions. In actual operation increased throttle is required
to maintain constant power levels and thus the mass of CO emitted from the
cylinder tends to increase. The increase is off-set to some extent by the
higher exhaust temperatures which result in some CO clean-up in the exhaust
system.
3-6
-------
Increasing backpressure increases the amount of residual exhaust gas
left in the cylinder at the end of the exhaust cycle. If this increase in
residual does not increase the percentage of dilution of the fresh charge
to a level where the combustion is adversely affected, the HC and probably
the CO emissions will be lowered. The reduction arises from leaving the tail
end of the exhaust in the cylinder and subsequently oxidizing it in the next
cycle. At idle, dilution is already high and combustion is usually marginal
so the engine cannot tolerate much more exhaust dilution.
Increasing valve overlap has a similar effect to increasing the back-
pressure. The charge is further diluted with residual gases. Deterioration
in combustion can result with lean mixtures as the residual is greater with
increased valve overlap. If the mixture ratio must be enriched to provide
stable idle and off-idle performance, then CO emissions will be increased.
There is no effect on carbon monoxide concentration at a constant mixture
ratio, but any increase in throttle opening to overcome the increased charge
dilution will increase the CO as well as the HC emissions.
Intake manifold pressure is essentially an indicator of engine power.
Since carburetor and distributor settings are variable in the vehicle, there
is a change in emission concentration as the throttle is varied at constant
speed. In the intermediate power range, at constant speed, minimum HC and
CO from the engine are achievable for lean air-fuel calibrations. At wide
open throttle, the power valve is normally actuated and the mixture is en-
riched. The resulting enrichment forces an increase in HC and CO emissions,
3-7
-------
but the increase is limited somewhat by exhaust cleanup arising from increased
exhaust temperatures. At light loads and low manifold pressure, increased
emissions result from increased wall quenching accompanying the rich mixtures
as well as incomplete flame propagation. Dash-pots or other throttle cracking
devices are often used to limit intake manifold vacuum during deceleration.
Another approach is to include a fuel shut-off device, commonly used with
fuel injection systems, to minimize emissions during the deceleration mode.
Combustion chamber deposit buildup acts to increase the surface area of
the combustion chamber because of the resulting irregular, porous surface
deposits. Deposits also act as a sponge to trap raw fuel which remains
unburned and thus adds to the exhaust HC. In addition, exhaust deposits tend
to increase compression ratio which also increases HC emission. There is a
negligible effect of deposit build-up in the combustion chamber on CO emission.
Surface temperature changes the thickness of the combustion chamber
quench layer and the degree of after-reaction. Increased surface temperature
decreases HC emissions by increasing fuel evaporation and distribution, com-
bustion chamber temperature, exhaust system temperature, and therefore, ex-
haust gas reaction. However, an increase in surface temperature by engine
modification is expected to have an adverse effect on engine octane require-
ment, volumetric efficiency and lubrication. Hydrocarbon emissions arise
primarily from quenching at the combustion wall surface.
The ratio of surface area to volume in the combustion chamber is useful
for interpreting the effects of many design and operating variables on HC
emission concentration. CO emission concentration, however, is not necessarily
affected by surface-to-volume ratio changes.
3-8
-------
The combustion chamber design is an important consideration to reduce
the surface area for a given clearance volume. For example, a 10 centimeter (4 inch)
bore engine maintaining a fixed clearance volume, can have surface-to-volume
ratios of 8.0:1 for the pot-in-piston design, 7.2:1 for the pancake design,
6.6:1 for the hemisphere in head design, and 6.4:1 for the double-hemisphere
design. The stroke-to-bore ratio is another design factor used to minimize
the surface-to-volume ratio by increasing the stroke-to-bore ratio. Unfor-
tunately, this modification is opposed to modern engine design practice which
favors short strokes for lower friction and lower engine silhouette.
Larger displacement per cylinder suggests the possibility that for the
same displacement, engine emissions can be reduced by decreasing the number
of cylinders but increasing the displacement per cylinder. On the other hand,
for a given number of cylinders, increasing engine displacement can reduce the
surface-to-volume ratio, but mass emissions might increase because of in-
creased engine friction and intake charge volume.
A large reduction in compression ratio can also significantly reduce the
surface-to-volume ratio. This increases the clearance volume with little in-
crease in surface area. However, reducing the compression ratio results in
lower thermal efficiency and reduced engine power. Some of the major causes
of high CO emission are the direct result of improper maintenance for any
specific engine design combination which results in maladjusted carburetors,
3-9
-------
air/fuel mixture imbalances and general malfunction of emission control de-
vices.
3.2 DESCRIPTION OF LIGHT DUTY VEHICLE, LIGHT DUTY TRUCK, AND HEAVY
DUTY TRUCK INDUSTRY
A light duty vehicle (LDV) is currently defined as a passenger car or
passenger car derivative capable of seating 12 passengers or less.4
A light duty truck (LOT) is any motor vehicle rated at 3856 kg (8,500 Ib.)
gross vehicle weight rating (GVWR) or less and under 2720 kg (6,000 Ib.)
vehicle curb weight, has a basic vehicle frontal area of 4.27 m2 (46 ft2)
or less, and which is: a) designed primarily for purposes of transportation
of property or is a derivative of such a vehicle, or b) designed primarily for
transportation of persons having a capacity of more than 12 persons, or c)
available with special features enabling off-street or off-highway operation
and use.4
A heavy duty vehicle (HDV) is defined as any motor vehicle that has a
vehicle curb weight of more than 2720 kg (6000 Ib.) or that is rated at more
than 3856 kg (8500 Ib.) GVWR, or that has a basic vehicle frontal area in excess
of 4.27 m2 (46 ft*).4
U.S. manufacture of light duty vehicles is almost entirely done by the
five major motor vehicle manufacturers: General Motors Corp., Ford Motor
Company, Chrysler Corp., Volkswagen, and American Motors Corp. In 1977
factory sales of passenger cars exceeded 10.4 million of which 9.2 million
were of domestic origin.5 The major foreign importers were Toyota, Nissan,
Volkswagen, Honda and Fiat.
3-10
-------
The manufacture of light duty trucks sold in the U.S. is primarily ac-
complished by the major domestic passenger car producers. General Motors
Corporation (Chevrolet and GMC divisions), Ford Motor Company and Chrysler
Corporation (Dodge Truck division) all have separate truck divisions which
produce light duty as well as heavy duty trucks. American Motors Corporation
operates the Jeep division which manufactures light duty trucks.
The other major domestic manufacturer of LDT's is the International
Harvester Corporation (IHC). International does not produce light duty pas-
senger vehicles but does produce a line of light and heavy duty trucks.
Some LDT's sold in the U.S. are imported. The majority of U.S. imports
of trucks come from the Canadian plants operated by U.S. domestic producers.
Some imports, primarily light pick-up trucks, under 1814 kg (4,000 Ib.)
GVWR, come from Japanese producers. The major importers are Nissan
(Datsun), Toyota, Isuzu, and Toyo Kogyo.
Table 3-1 shows unit factory sales for light duty vehicles, light duty
trucks, and heavy duty vehicles from U.S. plants. Most data available on
light duty trucks are presented in two categories, based on GVWR. There is
a 0-2722 kg (0-6,000 Ib.) and a 2722-4536 kg (6,001-10,000 Ib.) category.
Since the new definition of light duty trucks includes only trucks up to
3856 kg (8,500 Ib.) GVWR, some adjustment to the 2722-4536 kg (6,001-10,000
Ib.) category was necessary for this analysis. The 1975 industry production
data available to EPA indicate that only five percent of all trucks with
3-11
-------
Q
>-
Q <
Z _J
< 0.
— o:
>- o
h- a:
OQ
<;
•> cc.
j O
o o
— <
x: LU
O ^
U J3
- O)—
W "O ^O
— 4) 1
_J U O "
3 —
o \o
LA O
>-OO O
> • un
(D
-------
GVWR's less than 4536 kg (10,000 Ib.) have GVWR's of more than 3856 kg (8,500
lb.). This five percent figure is used in Table 3-1 to adjust production
data to fit the LOT definition.
Heavy Duty Vehicles only represent on the order of 5 to 6 percent of the
total annual U.S. motor vehicle factory sales, but 70-75% of these vehicles
are powered by gasoline engines, most of which are derivatives of their LDV
engine counterparts. The majority of these gasoline powered trucks are used
in GVWR classes less than 14969 kg (33,000 lb.) GVWR and the majority of
trucks rated greater than 14969 Kg (33,000 Ib.) GVWR are powered by diesel
engines. The total population of motor vehicles in these categories is
presented in Table 3-2.
Table 3-3 presents data on the number of passenger cars and trucks in
use by age. This information, when compared to past carbon monoxide standards,
can give an indication of the number of vehicles in the United States subject
to a given standard. This is important since the air quality goal of a
control program based on exhaust emission standards will not be achieved
until most vehicles are equipped with controls that can meet the standards.
The data from Table 3-3 indicates that there are approximately 23% of the
passenger cars in-use which are uncontrolled. Approximately 42% of the trucks
in-use are uncontrolled.
3-13
-------
TABLE 3-2
NEW VEHICLE REGISTRATIONS
Source
LDV
LOT AND HDV
Total
Source
LDV
LDT and HDV
Total
Excludes the State of Oklahoma
Source: Reference 5
New Vehicle Registrations
1976 1977
9,751,485
3,058,009
12,809,494
10,751,924
3,465,193
14,217,117
Total Vehicle Registrations
1976 1977
10,351,327
27,719,597
38,070,924
114,113,000
29,230,000
143,343,000
3-14
-------
TABLE 3-3
MOTOR VEHICLES IN USE BY AGE
AS OF JULY 1, 1977
Age in Years
Under 1
1 - 2
2 - 3
3 - 4
4 - 5
5 - 6
6 - 7
7 - 8
8 - 9
9-10
10 - 11
11 - 12
12 - 13
13 - 14
14 - 15
15 - 16
16 and older
Passenger
7,
9,
7,
9,
10,
9,
7,
7,
6,
5,
4,
3,
3,
1,
1,
2,
Cars (1000's)
177
557
477
59^
854
563
866
449
963
859
416
887
023
969
315
818
093
Trucks (1000's)
2,177
2,746
2,109
2,689
2,752
2,291
1,639
1,573
1,645
1,267
1,129
1,096
922
736
566
442
2,422
Source: Reference 5
3-15
-------
3.3 DESCRIPTION OF THE AIRCRAFT INDUSTRY
EPA has established the following classes of aircraft and corresponding
power-plant classes to which different sets of standards would apply as
determined by the technical, economic, and safety constraints which are rel-
evant to each class:
Class
PI
P2
Tl
T2
T3J4
T5
APU
Type
Piston Engines
(excluding radials)
Turboprop engines
Small turbojet/fan
engines
Large turbojet/fan
engines intended
for subsonic flight
Special classes
applying to specific
engines for the purpose
of instituting early
smoke standards
Large turbojet/fan
engines intended for
supersonic flight
Gas turbine auxiliary
power units
Aircraft Application
Light general aviation
Medium to heavy general
aviation; some commercial
air transport
General aviation jet
aircraft; some commercial
air transport
Commercial subsonic
transport
Commercial subsonic
transports
Supersonic transport
Many turbojet/turboprop
The emissions levels permitted by the standards are described by an EPA
parameter (EPAP) which is defined in the aircraft regulations. Briefly, it
is a measure of the total emission of a particular pollutant produced by
3-16
-------
an engine over a typical landing/takeoff (LTO) cycle normalized with respect
to the total power output of the engine over that cycle. As such, larger
engines performing greater useful work are permitted proportionally larger
amounts of total emissions over smaller engines.
The standards, promulgated in July 1973 for all classes but T5 and in
July 1976 for that class, are summarized in Table 3-4.6
In addition, there has been proposed (FR Vol. 38, N. 136, July 17, 1973,
p. 19050) a regulation which, if promulgated, would require all (including
those already in service as of January 1, 1979) large i.e., > 129 kilonewtons
(29,000 Ibs.) thrust in-use engines of the T2 class to comply with the T2
class standards of 1979 for HC, CO, NOX, and smoke. As this would effective-
ly require a retrofit program for the older engines (pre-1979), the com-
pliance date was proposed to be January 1, 1983, thus allowing four years
for that retrofit to be accomplished.
On a nationwide basis, however, all aircraft are estimated to contribute
only 0.63percent of the total CO as shown in Table 2-4. This includes
commercial transport, military and general aviation. General aviation
includes a wide variety of aircraft which are used for business, training,
and pleasure flying. Commercial transport aircraft source CO is shown as a
percentage of the total impact for different Air Quality Control Regions in
Table 3-5. With the relatively small percentage of the total CO inventory
attributable to aircraft sources, it is not meaningful or perhaps even
possible within the accuracy of any existing air quality computer model to
discuss the impact of aircraft source CO emissions from a nationwide
3-17
-------
TABLE 3-4
SUMMARY OF AIRCRAFT ENGINE REGULATIONS
Newly Manufactured Engines
Class
Tl
T2
T3
T4
T5h
P2 h
APU°
HC
45.3
22.7
22.7
22.7
10.5
3.0
0.2
(1.6)
(0.8)
(0.8)
(0.8)
(3.9)
(4.9)
(0.4)
266
122
122
122
853
16.
3.0
(9.4)
(4.3)
(4.3)
(4.3)
(30.1)
(26.8)
( 5.0)
105
85
85
85
255
7
1
The standards for advanced engines are:
Newly Certified Engines
Class
T2
T5
H£
11.3 (0.4)
28.3 (1.0)
EPAP
CO
85
221
(3.0)
(7.8)
85
142
NO,
NO,
Compliance Date
(3.7)
(3-0)
(3.0)
(3.0)
(9.0)
(12.9)
(3.0)
January 1
January 1
January 1
January 1
January 1
January 1
January 1
1979
1979
1979
1979
1979
1979
1979
Compliance Date
(3.0) January 1, 1981
(5.0) January 1, 1984
aMicrograms of pollutant per Newton thrust seconds (pounds of pollutant per
1000 pounds thrust hours) over the LTO except as noted.
Grams of pollutant per kilowatt hour (pounds of pollutant per 1000 horsepower
hours) over the LTO cycle.
Source: Reference 6
3-18
-------
TABLE 3-5
COMMERCIAL AIRCRAFT SOURCE CO EMISSIONS AS A
PERCENTAGE OF TOTAL AIR QUALITY CONTROL REGION EMISSIONS
Percentage of AQCR Emissions
AQCR Attributable to Commercial Aircraft
£2.
Los Angeles 0.22
San Francisco 0.37
NY-NJ-Conn. 0.32
Chicago 0.19
St. Louis 0.3**
Cincinnati 0.14
Baltimore 0.32
Boston 0.35
Houston 0.32
S.E. Wisconsin 0.19
Washington, D.C. 0.46
Atlanta 1.08
Source: Reference 7
3-19
-------
standpoint.8 EPA has monitored the progress of aircraft technology since
1973 and has reviewed the impact of various types of aircraft on ambient air
quality. As a result, it is currently being proposed that aircraft emission
standards for commercial turbine engines be relaxed and implementation be
delayed considering the status of control technology and lead time considera-
tions.7
CO violations attributable to aircraft are occurring, however, at some
airport terminal boarding gate areas and at the end of the runways. These
situations will require CO control technique strategies to preclude such
point-source violations. A point-source violation is defined by EPA to be
one with emissions of any pollutant greater than 100 tons/year.7 By this
definition, aircraft operating at the major commerical airports must be
considered as a major source of CO as shown in Figure 3-2. Yet, as shown in
Table 3-5, commercial aircraft contribute only 1.1 percent or less of the
total CO in any particular air quality control region. The general conclu-
sion is that the aircraft source for CO is significant even though the over-
all percentage contribution may be small. New data and models are currently
being gathered and evaluated to determine if this conclusion is valid. For
information on those special cases where aircraft source CO control measures
are required, the reader is referred to Reference 6.
3.4 VEHICLE CO EMISSION STANDARDS
Motor vehicle emission standards on passenger cars and light duty trucks
have been enforced in California since 1966 and the remaining states since
1968.16 CO standards for medium- and heavy-duty trucks were implemented in
3-20
-------
18,100
(20,000)
13,600
(15,000)
9,100
(10,000)
4,500
(5,000)'
Carbon Monoxide
_l
Ref. 4
. 13
Carbon Monoxide
Ref. 10
Ref. 11
1970
1975
18,100
(20,000)
13,600
(15,000)
en cu 9,100
0^(10,000)
4,500
(5,000)
1980
'^s.Ref.
1970
1975
1980
18,100
(20,000)
13,600
(15,000)-
9,100 _
(10,000)
4,500 -
(5,000)
Comparison of Estimates of Aircraft
Emissions at O'Hare. 1970-1980
Monoxide
Comparison of Estimates of Aircraft
Emissions at J.F.K., 1970-1980
3,600
(4,000)
2,700
(3,000)'
•H o 1,800
iib (2,000)'
900 _
(1,000)
Carbon Monoxide
I
Ref. 10
Ref. 12
1970
1975
1980
1970
1975
i960
Comparison of Estimates of Aircraft
Emissions at L.A. International,
1970-1980
Comparison of Estimates of Aircraft
Emissions at Washington National,
1970-1980
Source: Reference 7
FIGURE 3-2. ESTIMATES OF AIRCRAFT SOURCE CO EMISSIONS AT MAJOR NATIONAL
AIRPORTS
3-21
-------
California in 1969 and for the 49 states in 1970. Table 3-6 summarizes the
standards for CO exhaust emissions from non-California light-duty vehicles
and light-duty trucks. Table 3-7 summarizes the California standards for
light-duty vehicles, light-duty trucks, and medium-duty trucks. Other CO
exhaust emission standards are presented in Tables 3-8, 3-9 and 3-10. These
tables apply to non-California heavy-duty vehicles (HDV), California HDV's,
and motorcylces, respectively. For detailed descriptions of testing proce-
dures and methodologies, refer to the Special Bibliography at the end of
this chapter.
3.5 IN-USE EXPERIENCE
Results from a 1973 surveillance program17, the In-Use Compliance Pro-
gram, indicated that seven classes of 1973 and 1974 model year vehicles were
significantly exceeding the emission standards in use. Subsequently, it was
found from the analysis of the Fiscal Year 1974 (FY74) Emission Factor Pro-
gram (EFP) that 63 percent of the 1975 model year (MY) vehicles that were
tested failed to meet the standards for one or more pollutants.18 Of 587
1975 MY vehicles tested, 52 percent failed because of high CO levels only or
in combination with other pollutants. The FY75 EFP results for 1976 MY
vehicles were not statistically different from the 1975 MY vehicles tested
in the 1974 EFP in terms of mean HC and CO emissions,19 Another study, called
the Restorative Maintenance Project was initiated to better evaluate why such
a large percentage of vehicles had excessive emissions and to determine if
normal emissions could be restored.20
3-22
-------
TABLE 3-6
Federal Vehicle Exhaust Emission Standards for CO
Model Year
Light Duty Vehicles
Pre-1968
1968 - 1969
1970 - 1971
1972°
1973 - 1974^
1975 - 1976°
1977,- 1979
1980d
1981 and later
Light Duty Trucks
Less than 2720 kg (6000 Ib) GVWR
Pre-1975 ,
1975 - 1978°
1979,- 1982°
1983
2720-3856 kg (6001-8500 Ib) GVWR
pre-1979
1979,- 1982d
1983d
CO Standard
No standard
a2.3% by volume for 820-1639 cc
displacement (50-100 CIDJ
a2.0% by volume for 1640-2294 cc
displacement (100-140 CID)
a!.5% by volume for >2294 cc
displacement (>140 CID)
14.3 g/km
24.2
9.3
9-3
4.3
2.1
g/km
g/km
g/km
g/km
g/km
(23 g/mi)
(39 g/mi)
(39 g/mi)
(15 g/mi)
(15 g/mi)
(7.0 g/mi)
(3.4 g/mi)e
Same standard as automobiles
12.4 g/km (20 g/mi)
11.2 g/km (18 g/mi)
to be determined
Same standard as heavy duty
gasoline vehicles (see Table 3~8)
11-2 g/km (10 g/mi)
to be determined
aEmission standard varied with vehicle's volumetric displacement using
7-mode driving cycle test
by-mode Test Procedure
cCVS-72 Test Procedure
dCVS-75 Test Procedure
eA waiver of the 2.19 g/km (3-4 g/mi) CO standard is possible for 1981 and
1982 at a level not to exceed 4.35 g/km (7 g/mi).
Source: Title 40 CFR
3-23
-------
TABLE 3-7
MODEL YEAR CO STANDARD
Automobi les
1966 - 1967 1.5% by volume
1968 - 1969 Same as U.S. standard
1970 - 1971 14.3 g/kmc (23 g/mi)
1972 14.3 g/kmc or 24.2 g/kmd (23 g/mi or 39 g/mi)
1973 - 1974 2k. 2 g/kmd (39 g/mi)d
1975 - 1980 5-6 g/kme (9-0 g/mi)e
1981 b 4.3 g/kmc or 2.1 g/kme (7-0 g/mi^or
3.4 g/mi )
1982 and later 4.3 g/kme (7.0 g/mi)e
Light Duty Trucks
Less than 1814 kg (4000 1b) GVWR and 1815-2722 kg (4001-6000 lb) GVWR
Pre-1975 Same as automobiles
1975 12.4 g/kme (20 g/mi)
1976 - 1978 10.6 g/kme (17 g/mi)
1979 and later 5-6 g/kme (9-0 g/mi)
Medium Duty Trucks
2723 - 3856 kg (6001 - 8500 lb) GVWR
1969 - 1977 Same as Heavy Duty Standards
1978 - 1980 10.6 g/kme (17 g/mi)e
1981 and later 5-6 g/kme (9-0 g/mi)e
aStandard applies to passenger cars and light duty trucks through 1974.
After 1975, standards apply only to passenger cars.
b4.3 g/km (7-0 g/mi) CO and 0.43 g/km (0.7 g/mi) NOX or 2.1 g/km (3.4 g/mi)
CO and 0.62 g/km (1.0 g/mi) NOX [0.93 g/km (1.5 g/mi) NOX optional with
161,000 km (100,000 mile) durability]
7-mode test procedure
dCVS-72
eCVS-75
Source: Title 13> California Administrative Code
3-24
-------
TABLE 3-8
FEDERAL VEHICLE EXHAUST EMISSION STANDARDS FOR CO:
HEAVY DUTY GASOLINE AND DIESEL VEHICLES
Model Year CO Standard
Pre-19708 No standard
1970 - 1973a 1.5% by volume ,
1974 - 1978 53-6 g/kw hr (40 g/BHP-hr)^
1979 - 1982 33.5 g/kw hr (25 g/BHP-hr)
1983 and later New standard and test procedure
being developed
Gasoline Only
Brake horsepower-hour
Source: Title 40 CFR
TABLE 3-9
CALIFORNIA VEHICLE EXHAUST EMISSION STANDARD FOR CO:
HEAVY-DUTY GASOLINE AMD DIESEL VEHICLES
Model Year CO Standard
1969 - 1971a 1.5% by volume
1972 1.0% by volume
1973 - 1974 53.6 g/kw-hr (40 g/BHP-hr)
1975 - 1976 40.2 g/kw hr (30 g/BHP-hr)
1977 and later 33.5 g/kw hr (25 g/BHP-hr)
From 1969 - 1972, standards apply to gasoline-powered vehicles only.
After 1973, standards apply both to gasoline- and diesel-powered vehicles
Source: Title 13, California Administrative Code
3-25
-------
TABLE 3-10
U.S. VEHICLE EXHAUST EMISSION STANDARDS
FOR MOTORCYCLES - 50 STATES
Model Year
Pre-1978
1978-- 1979
1980 and later
CO Standard
No standard
17 g/km (27.4 g/mi)
12 g/km (19.3 g/mi)
Source: Title 40 CFR
TABLE 3-11
COMPARISON OF EXHAUST EMISSION LEVELS BETWEEN THE 49-STATE
LOW-ALTITUDE VEHICLES IN THE RESTORATIVE MAINTENANCE
AND EMISSION FACTORS PROGRAMS
Model Year Program
1975/1976
1976
1975
RM
EF
EF
1975/1976 Federal
Standards
N
300
515(
587
_
Average
Mi leage
12,900 km
(8,000 mi)
18,500 km
11 ,500 mi)
14,200 km
(8,800 mi)
_
HC
g/km
fa/mi)
0.81
(1.3)
0.81
(1.3)
0.81
(1.3)
0.93
(1.5)
CO
g/km
(q/mi)
12.6
(20.3)
11.4
(18.3)
14.2
(22.9)
9.3
(15)
NO*
g/km
(d/mi )
1.74
(2.8)
1.62
(2.6)
1 .49
(2.4)
1.93
(3-D
% Meeting
Standards
42
45
37
Source: Reference 20
3-26
-------
A summary of exhaust emission results from the initial test on the
300 vehicles of the 1975 and 1976 model years in Chicago, Detroit, and
Washington, D.C. is shown in Table 3-11.20 These values are compared to the
performance of 1975 and 1976 models tested in the Emission Factor programs
as well as to 1975/1976 Federal Standards.
Table 3-11 indicates that this sample of Restorative Maintenance vehi-
cles is similar to the Emission Factors fleet in terms of the initial test
with regard to emission levels and pass/fail performance. Although the
average levels of HC and N0v are below the standards, the scatter of the
X
individual data points combined with an average value of CO which was above
the standard, allowed only 42 percent of the total fleet to meet the standards
(Figure 3-3.) The inspection which followed the initial test sequence re-
vealed that 74 percent of the 1975 and 1976 models which failed to meet the
standards had some form of malperformance in their emission control systems.
Although few actual defects were discovered, many maladjustments and disable-
ments were found. The primary area of malperformance was in the Carburetor/
Choke/Exhaust Heat Control Valve System with a 66 percent failure rate over
the entire sample. Limiter caps were missing or broken on 45 percent
of the 300 vehicles; idle speed was maladjusted (more than +_ 100 rpm from
specification) on 25 percent and the choke adjustment was out of production
tolerances on 10 percent of the vehicles tested. The ignition system was
the second largest area for malperformance with a 27 percent overall rate.
Most of this was basic ignition timing maladjustment at 19 percent. The
remaining major area was the exhaust gas recirculation (EGR) system.
3-27
-------
Fail All 6%
C0/N0x 4%
HC 0.5%
Source: Reference 20
HC/NOx 0.5%
FIGURE 3-3. PASS/FAIL OUTCOMES OF THE INITIAL TEST
ON 1975 AND 1976 VEHICLES
3-28
-------
Fifteen percent of the vehicles were found to have malperformance in this
area. In testing 1977 models, fewer vehicles (58 percent) were found to have
malperforming systems although the pattern discovered on the older vehicles
was still present.
In general, the effect of engine component operation on CO and HC emis-
sions is shown in Table 3-12. The effect on HC emissions is included here
since CO formation is an intermediate product of combustion of hydrocarbon
fuels.
3.6 CO EMISSION FACTORS
EPA has administered programs to determine how well vehicles perform in
actual use by administering a series of exhaust emission surveillance pro-
grams. Test fleets of consumer-owned vehicles within various major cities
are selected by model year, make, engine size, transmission and carburetor/
fuel injection system in such proportion as to be representative of both the
normal production of each model year and the contribution of that model year
to total vehicle miles traveled. In the case of heavy duty vehicles, fuel
type and gross vehicle weight are key items in the stratification scheme.
The data collected in these programs are analyzed to provide an estimate of
mean emissions with accumulation of age, percentage of vehicles complying
with standards, and to assess the effect on emissions of vehicle parameters
(engine displacement, vehicle weight, etc.).
These surveillance data, along with prototype vehicle test data, assem-
bly line test data, and technical judgement form the basis for the existing
and projected mobile source emission factors presented here.22 For localized
3-29
-------
TABLE 3-12
EFFECT OF ENGINE COMPONENT OPERATION ON EMISSIONS
CHANGE IN EMISSIONS
COMPONENT
Decreased air-fuel ratio
Decreased engine idle speed
Restricted PCV valve
Restricted air fi1ter
Choke malfunctions
Carburetor malfunctions
Ignition system malfunctions
Advanced spark timing
Stuck heat riser valve
Exhaust valve leak
Intake manifold leaks
Emission control device
malfunct ion
Catalytic converter malfunction
Carbon Monoxide
Increase
Increase
I ncrease
Increase
Increase
Large Increase
NSC
NSC
I ncrease
NSC
Increase
Increase
Large Increase
Hydrocarbon
Increase
Increase
Increase
Increase
I ncrease
Increase
Large Increase
I ncrease
NSC
Increase
I ncrease
Increase
Large Increase
NSC = No Significant Change
Source: Reference 21
3-30
-------
pollutants such as CO, the ability of the test procedure to predict changes
in emissions depends on the similarity of the localized driving pattern and
associated operating conditions to those in the test procedure. The EPA,
therefore, has developed a series of correction factors to expand upon the
LDV and HDV test procedures and to predict emissions from a large number of
user-specific scenarios. These are contained in Reference 22. Data required
to develop these correction factors have been generated using statistical
studies with consumer-owned vehicles.
The base CO exhaust and idle emission factors for LDV's, LDT's, MDT's,
and HDV's and motorcycles are shown in Tables 3-13 through 3-22 and represent
the mean emission factors for July of any particular calendar year. The
emissions testing for light-duty vehicles, light-duty trucks and medium-duty
trucks is performed according to the 1975 Federal Test Procedure (FTP) as
stipulated in the Federal Register (Vol. 137, No. 211, November 15, 1972).
Light-duty trucks in the range of 0-2720 kilograms (0-6000 Ibs.) Gross Vehicle
Weight Rated (GVWR) and 2721-3856 kilograms (6001-8500 Ibs.) GVWR are also
tested according to the 1975 FTP. However, until the 1979 model year (MY),
the trucks in the 2721-3856 kilogram (6001-8500 Ibs.) GVWR range were certi-
fied under the less stringent Heavy-Duty Truck procedures.
EPA test programs for determining in-use heavy-duty vehicle (HDV) emis-
sion factors use both the heavy-duty FTP, which is a steady state engine
dynamometer procedure, and an actual urban road test, referred to as the San
Antonio Road Route (SARR). The SARR is a 11.65-kilometer (7.24-mile) test
course and includes arterial and local/collector highway segments. The
3-31
-------
TABLE 3-13
EXHAUST EMISSION RATES FOR ALL AREAS.
EXCEPT CALIFORNIA AND HIGH-ALTITUDE'
Light Duty Vehicles
Pollutant
CO
CO
CO
CO
CO
Model Year
Pre-1968
1968-1974
1975-1979
1980
1981 +
A q/km (q/mi)
New Vehicle
Emission Rate
42.44 (68.30)
19.35 (31-1*0
11.56 (18.60)
1.86 ( 3-00)
0.87 ( 1.40)
B g/km (g/mi)
Deterioration Rate
Per 16,100 km (10,000 mi)
1.90 (3.06)
3.82 (6.15)
1.74 (2.80)
1.43 (2.30)
1.24 (2.00)
The Exhaust Emission Factor is calculated from the linear equation C = A + BY,
where C is the exhaust emission factor for a vehicle with cumulative mileage
M, A and B are the factors listed in the above table, and Y = M/l6,100 (M/10,000)
Source: Reference 22
TABLE 3-14
IDLE EMISSION RATES FOR ALL AREAS
EXCEPT CALIFORNIA AND HIGH-ALTITUDE'
Light Duty Vehicles
Pollutant
CO
CO
CO
CO
CO
Model Year
Pre-1968
1968-1974
1975-1979
1980
1981 +
A g/km (g/mi
New Vehicle
Emission Rate
10.20 (16.42)
7.91 (12.73)
3-37 ( 5.43)
0.55 ( 0.88)
0.25 ( 0.41)
B g/km (g/mi)
Deterioration Rate
Per 16,100 km (10,000 mi)
1.58 (2.55)
1.81 (2.92)
0.52 (0.83)
0.42 (0.67)
0.37 (0.59)
The Idle Emission Factor is calculated from the linear equation C = A + BY,
where C is the idle emission factor for a vehicle with cumulative mileage
M, A and B are the factors listed in the above table, and Y = M/l6,100 (M/10,000)
Source: Reference 22
3-32
-------
TABLE 3-15
EXHAUST EMISSION RATES FOR ALL AREAS
EXCEPT CALIFORNIA AND HIGH-ALTITUDE8
Light Duty Trucks: Both Weight Categories
Pollutant
CO
CO
CO
CO
CO
CO
Model Year
Pre-1968
1968-1969
1970-1974
1975-1978
1979-1982
1983+
A g/km (g/mi)
New Vehicle
Emission Rate
43.73 (70.38)
26.15 (42.08)
19.56 (31.^8)
14.57 (23.44)
9.01 (14.50)
2.40 ( 3.87)
B g/km (g/mi)
Deterioration Rate
Per 16,100 km (10,000 mi)
1.90 (3.06)
3.38 (5.44)
3.82 (6.15)
3.54 (5.70)
3.32 (5.34)
1.24 (2.00)
CO
CO
CO
CO
CO
Light Duty Trucks 0-2720 kg (0-6000 Ib) GVWR
Pre-1968
1968-1974
1975-1978
1979-1982
1983+
42.44 (68.30)
19.35 (31.14)
10.00 (16.10)
9.01 (14.50)
2.40 ( 3.87)
1.90 (3.06)
3.82 (6.15)
(5.34)
(5.34)
(2.00)
3.32
3-32
.24
CO
CO
CO
CO
Light Duty Trucks 2721-3856 kg (6001-8500 Ib) GVWR
Pre-1970
1970-1978
1979-1982
1983+
48.90 (78.70)
20.13 (32.40)
9.01 (14.50)
2.40 ( 3.87)
1.90 (3.06)
3.82 (6.15)
3.32 (5.34)
1.24 (2.00)
The Exhaust Emission Factor is calculated from the linear equation C = A + BY,
where C is the exhaust emission factor for a vehicle with cumulative mileage
M, A and B are the factors listed in the above table, and Y = M/16,100 (M/10,000)
Source: Reference 22
3-33
-------
TABLE 3-16
IDLE EMISSION RATES FOR ALL AREAS
EXCEPT CALIFORNIA AND HIGH-ALTITUDE3
Light Duty Trucks, Both Weight Categories
Pollutant
CO
CO
CO
CO
CO
CO
Model Year
Pre-1968
1968-1969
1970-1974
1975-1978
1979-1982
1983+
A g/km (g/mi)
New Vehicle
Emission Rate
10.30 (16.58)
8.56 (13.77)
8.90 (14.32)
5.90 ( 3.k3)
1.13 ( 1.82)
0.30 ( 0.49)
B g/km (g/mi)
Deterioration Rate
Per 16,100 km (10,000 mi)
58
76
81
(2.55)
(2.83)
(2.92)
1.35 (2.17)
0.97 (1-56)
0.16 (0.25)
CO
CO
CO
CO
CO
Light Duty Trucks 0-2720 kg (Q-6000 Ib) GVWR
Pre-1968
1968-1974
1975-1978
1979-1982
1983+
10.20 (16.42)
7-91 (12.73)
1.26 ( 2.02)
1.13 ( 1-82)
0.30 ( 0.49)
1.58 (2.55)
1.81 (2.92)
(1.56)
(1.56)
(0.25)
0.97
0.97
0.16
CO
CO
CO
CO
Light Duty Trucks 2721-3856 kg (6001-8500 Ib) GVWR
Pre-1970
1970-1978
1979-1982
1983+
10.71 (17-24)
11.57 (18.62)
1 .13 ( 1.82)
0.30 ( 0.49)
1.58 (2.55)
1.81 (2.92)
0.97 (1.56)
0.16 (0.25)
The Idle Emission Factor is calculated from the linear equation C = A + BY,
where C is the idle emission factor for a vehicle with cumulative mileage
M, A and B are the factors listed in the above table, and Y = M/l6,100 (M/10,000)
Source: Reference 22
3-34
-------
Pollutant
CO
CO
CO
CO
CO
TABLE 3-17
EXHAUST EMISSION RATES FOR ALL AREAS
EXCEPT CALIFORNIA AND HIGH-ALTITUDE3
Heavy Duty Gasoline Fueled Vehicles
Model Year
Pre-1970
1970-1973
1974-1978
1979-1982
1983+
A g/km (g/mi)
New Vehicle
Emission Rate
B g/km (g/mi)
Deterioration Rate
Per 16,100 km (10,000 mi)
169.6
132.2
136.0
119.2
(272.9)
(212.7)
(218.8)
(191.9)
9.56 ( 15.38)
1.90 ( 3.06)
3.82 ( 6.15)
3.82 ( 6.15)
3.82 ( 6.15)
6.55 (10.54)
The Exhaust Emission Factor is calculated from the linear equation C = A + BY,
where C is the exhaust emission factor for a vehicle with cumulative mileage
M, A and B are the factors listed in the above table, and Y = M/l6,100 (M/10,000)
Source: Reference 22
Pollutant
CO
CO
CO
CO
CO
TABLE 3-18
IDLE EMISSION RATES FOR ALL AREAS
EXCEPT CALIFORNIA AND HIGH-ALTITUDE;
Heavy Duty Gasoline Fueled Vehicles
Model Year
Pre-1970
1970-1973
1974-1978
1979-1982
1983+
A g/km (g/mi)
New Vehicle
Emission Rate
15.30 (24.63)
9.76 (15.70)
13.62 (21.92)
11.95 (19.23)
0.96 ( 1.54)
B g/km (g/mi)
Deterioration Rate
Per 16,100 km (10,000 mi)
1.58 (2.55)
1.81 (2.92)
(2.92)
(2.92)
(5.00)
1.81
1.81
3.11
The Idle Emission Factor is calculated from the linear equation C = A + BY,
where C is the idle emission factor for a vehicle with cumulative mileage
M, A and B are the factors listed in the above table, and Y = M/16,100 (M/10,000)
Source: Reference 22
3-35
-------
Pollutant
CO
CO
CO
CO
TABLE 3-19
EXHAUST EMISSION RATES FOR ALL AREAS.
EXCEPT CALIFORNIA AND HIGH-ALTITUDE'
Heavy Duty Diesel Fueled Vehicles
Model Year
Pre-197**
1974-1978
1979-1982
1983+
A g/km (g/mi)
New Vehicle
Emission Rate
21.81 (35.10)
16.78 (27.00)
16.78 (27.00)
16.78 (27.00)
B g/km (g/mi)
Deterioration Rate
Per 16,100 km (10,000 mi)
The Exhaust Emission Factor is calculated from the linear equation C = A + BY,
where C is the exhaust emission factor for a vehicle with cumulative mileage
M, A and B are the factors listed in the above table, and Y = M/l6,100 (M/10,000)
Source: Reference 22
TABLE 3-20
IDLE EMISSION RATES FOR ALL AREAS
EXCEPT CALIFORNIA AND HIGH-ALTITUDE'
Heavy Duty Diesel Fueled Vehicles
Tollutant
CO
CO
CO
CO
Model Year
Pre-197**
1974-1978
1979-1982
1983+
A g/km (g/mi)
New Vehicle
Emission Rate
0.82 (1.32)
0.41 (0.66)
0.41 (0.66)
0.41 (0.66)
B g/km (g/mi)
Deterioration Rate
Per 16,100 km (10,000 mi)
The Idle Emission Factor is calculated from the linear equation C = A + BY,
where C is the idle emission factor for a vehicle with cumulative mileage
M, A and B are the factors listed in the above table, and Y = M/l6,100 (M/10,000)
Source: Reference 22
3-36
-------
Pollutant
CO
CO
CO
CO
TABLE 3-21
EXHAUST EMISSION RATES FOR ALL AREAS.
EXCEPT CALIFORNIA AND HIGH-ALTITUDE'
Motorcycles
Model Year
Pre-1978
1978-1979
1980-1982
1983+
A g/km (g/mi)
New Vehicle
Emission Rate
21.38 (34.40)
12.60 (20.27)
9-23 (14.86)
1.68 ( 2.71)
B g/km (g/mi)
Deterioration Rate
Per 16,100 km (10,000 mi)
0.96 (1.54)
2.49 (4.00)
2.49 (4.00)
1.24 (2.00)
The Exhaust Emission Factor is calculated from the linear equation C = A + BY,
where C is the exhaust emission factor for a vehicle with cumulative mileage
M, A and B are the factors listed in the above table, and Y = M/l6,000 (M/10,000)
Source: Reference 22
TABLE 3-22
IDLE EMISSION RATES FOR ALL AREAS
EXCEPT CALIFORNIA AND HIGH-ALTITUDE'
Motorcycles
Pollutant
CO
CO
CO
CO
Model Year
Pre-1978
1978-1979
1980-1982
1983+
A g/km (g/mi)
New Vehicle
Emission Rate
5.14 (8.27)
3.03 (4.87)
2.22 (3.57)
0.40 (0.65)
B g/km (g/mi)
Deterioration Rate
Per 16,100 km (10,000 mi)
0.80 (1.28)
0.70 (1.12)
0.70 (1.12)
0.30 (0.48)
The Idle Emission Factor is calculated from the linear equation C = A + BY,
where C is the idle emission factor for a vehicle with cumulative mileage
M, A and B are the factors listed in the above table, and Y = M/l6,100 (M/10,000)
Source: Reference 22
3-37
-------
average speed is around 32 km/hr (20 mi/hr) with about 20 percent of the time
spent at idle. Since emissions from the steady state dynamometer tests are
generally not easy to convert to on-the-road emissions, regression equations
were developed so that on-the-road emissions (SARR) could be predicted. It
is not known, however, whether the SARR accurately represents the average
HDV driving patterns. Preliminary analysis of Los Angeles urban truck
operation data indicates an average speed of around 42 km/hr (26 mi/hr),
10 km/hr (6 mi/hr) higher than the SARR average speed. However, the road
route does have similar characteristics to the representative light duty
driving schedule with respect to average road speed and percent time at idle.
Since traffic is likely to be the major constraint within urban environment,
it is not surprising that truck and car schedules would be similar, but the
SARR (and the current LDV FTP) makes no attempt to account for the time that
trucks spend idling as a result of deliveries, special operations (buses,
garbage trucks, etc.), auxiliary power equipment, etc.
Since operational data have not been completely analyzed and trucks have
not been fully tested on transient cycles developed from the operational data,
the projected emission factors for heavy-duty vehicles shown in Tables 3-17
to 3-20 are based on the SARR driving schedule. The data for the HDV tables
were assembled from emission factors contracts involving the testing of 35
gasoline and ten diesel in-use heavy duty trucks by chassis dynamometer
versions of the FTP as well as over the SARR, and a sensitivity study of
18 gasoline and 12 diesel in-use heavy-duty trucks.
Motorcycles have become more popular and their numbers have been increas-
ing in recent years. The majority of motorcycles are powered by either 2-
3-38
-------
stroke cycle or 4-stroke cycle air-cooled engines. Currently, the nationwide
population of motorcycles is approximately 49 percent 2-stroke and 51 percent
4-stroke.22 Emission rates given in Tables 3-21 and 3-22 are composites of
six different categories of motorcycles (small, medium, and large for 2- and
4-stroke cycle). Composite exhaust emission factors are calculated according
to the 1975 FTP as stipulated in the Federal Register (Vol. 40, No. 205,
October 22, 1975).
These mean composite exhaust emission rates for the different vehicles
reflect the national average mileage accumulation rates of greater than
16,100 kilometers (10,000 miles) per year for newer vehicles and decreasing
mileage accumulation as vehicles age. An additional series of correction
factors to predict specific scenarios to reflect such variables as temperature,
average speed, air-conditioning, vehicle loading, trailer towing, inspection/
maintenance credits, etc. are covered in greater detail in Reference 22,
entitled Mobile Source Emission Factors For Low-altitude Areas Only, EPA
400/9-78-006, March 1978.
Carbon monoxide emission factors for mobile sources provide useful infor-
mation for projecting the CO impact on ambient air quality from mobile sources.
The results of one such study are shown in Figure 3-4. The curve is based
upon current and proposed CO standards as well as upon a control program for
new vehicles. It is not known whether the effects of vehicle and control
equipment degradation were considered in this study. Figure 3-5 presents the
results of another study which projected the total number of vehicle kilo-
meters (miles) traveled through 1990. In the development of these curves,
3-39
-------
o
o
Q >>
to
i- -O
-------
\
o
.CT>
cn
cn o
•— o
oo o
r\ T\
CO CM
o
00
cr>
-
<* o
i— O
«=3- LO
f\ rv
CvJ i—
O O
i— O
vo o
LO o
o o
co i-n
o
CM
CO
u
c
O)
O)
M-
O)
Qi
CO
U
S-
3
o
00
LU
00
O
UJ
o
o
C£
D_
O
00
<
Q_
Lf)
I
CO
UJ
SUOLLU8
(S9L.LW)
3-41
-------
it was assumed that the average passenger car is driven 15,100 kilometers
(9400 miles) per year. Although this curve was generated in 1963, its pre-
diction of 1976 passenger car vehicle kilometers (miles) traveled deviates
from the actual number by only 3 percent.
3.6.1 The Effect of Cold Weather on CO Emissions
The Federal Test Procedure (FTP) employed by the EPA to determine com-
pliance with specific model year emission standards specified that vehicle
temperature be stabilized in a temperature environment of 20 - 25.6°C (68 -
78°F) prior to the test. While the starting-up and running of these vehicles
for the first part of the test cycle constitutes a "cold start" with respect
to engine coolant temperature, the cold start typically experienced under
ambient temperature and considered a "cold start" by most people is not, in
fact, simulated under the FTP conditions.
EPA has studied the effects of colder ambient temperatures on CO emis-
sions. Quantitative information is included in References 22 through 26 listed
at the end of this chapter. The emissions of CO are shown to increase signi-
ficantly under non-FTP, low ambient temperatures. For example, in one study
where 84 vehicles were selected for Tow temperature tests, 87 percent pro-
duced more CO in the low temperature FTP than in the norraa-1 FTP.?6 The first
group of 14 vehicles tested at temperatures from -8.9 to -3.9°C (-5QC average)
[16°F to 25°F (23°F average)] showed an 82 percent increase in CO; the second
group of 26 vehicles tested at temperatures from -3.3 to 1.7°C (0°C average)
[26°F to 35°F (32°F average)] showed a 74 percent increase in CO; and the
third group of 13 vehicles tested at temperatures from 7.8 to 12.8°C (10°C
3-42
-------
average) [46°F to 55°F (50°F average)] showed a 21 percent increase from the
normal 23.9°C (75°F) average FTP.2^
Some vehicles of course are more sensitive than others. For example,
a 1976 model year vehicle from one study produced 1.74 g/km (2.89 g/mi) when
tested according to standard FTP conditions [approximately 25°C (77°F) cold
soak], but when tested under non-FTP cold soak conditions of -12.2 to -3.9°C
(10°F to 25°F) produced CO emissions of 11.29 q/km (18 q/mi).24
It is not surprising then, that the National Ambient Air Quality Standard
(NAAQS) for CO is violated during cold weather conditions. Figure 3-6 shows
the relative CO violations versus mean temperature. EPA is working to refine
the results shown in Figure 3-6 and is also considering whether control of CO at
temperature conditions other than those represented by the current FTP is
warranted.
In addition to temperature, the type of driving cycle also affects CO
emissions. EPA is studying this effect as well, but currently some of the
driving cycle effects are less quantified than are the temperature effects.
What is known is that if vehicles are operated in higher engine speed/load
modes that are not well represented on the EPA tests, the emissions of CO and
other pollutants can be higher than would be indicated from the Federal Test
Procedure (FTP) results. Work is underway to quantify the magnitude of these
effects on CO and other pollutants.
3-7 CARBON MONOXIDE CONTROL FOR NEW MOBILE SOURCES
The control of CO emissions from new mobile sources provides an impor-
tant and effective approach to improving air quality with respect to CO.
3-43
-------
00
00
o
00
m
CM
CO
o
co en
i i
o m
co oo
CM oo
i I
-3- m
CM r-
co -3-
CM r^
I I
CTi IT)
CXL
«=C
cm
LU
Q.
21
O
O
•o
Q)
O
to
-o
O
CD
Q.
i-
O)
-Q
OO
I I
ro m
O
O
LU CO
cm •>•
*
oo
i i
CM m
CO
I I
I CM
o
o
CO
-a
o
•r~
s-
O)
Q.
s-
O)
-Q
E
ZJ
c:
O
*
o
m
m
co
o
CO
m
CM
o
CM
(%) SNOI1V10IA 3AIlV13d
3-44
-------
Control of new mobile sources has received significant developmental efforts
in recent years. The driving force for this has been the implementation and
enforcement of increasingly stringent CO exhaust emission standards. The
objective of this section is to identify and to provide information on con-
trol techniques applicable to the reduction of CO from new mobile sources.
The information on the controls is general in nature. More detailed dis-
cussions may be found in the references listed in the Special Bibliography
at the end of this chapter.
3.7.1 Types of CO Controls for New Mobile Sources
The literature reports that there are basically four alternative approaches
for controlling carbon monoxide emissions from new mobile sources. The first
and currently one of the more effective methods is treatment of the engine
exhaust gases for the removal of the CO. The second method is to reduce the
formation of CO in the vehicle engine by improving fuel/air mixture distri-
bution and control. The third is to replace the conventional premixed charge
spark-ignition gasoline-fueled engines with alternative types of engines which
produce less CO. The fourth method is the use of alternative fuels, such as
liquid petroleum gas (LPG), liquid natural gas (LNG), hydrogen, etc. Table
3-23 contains a list of specific controls under each of these general methods
and summarizes the status of development of each. The following sections
briefly discuss the emission reduction benefits, costs, energy requirements,
and environmental impacts associated with the application of these controls.
The Special Bibliography at the end of this chapter lists sources containing
more detailed information on new mobile source controls.
3-45
-------
TABLE 3-23
CARBON MONOXIDE CONTROL TECHNIQUES
FOR NEW MOBILE SOURCES
Type of Control
Fuel/Air Mixture
Improved fuel metering
Cold-Start Control Approaches
Quick chokes, exhaust heated
intake charge, improved cold
start vaporization/distribu-
tion, start catalysts, etc.
Air Injection
Improved EGR*
Electronic control for spark
timing, EGR, cold enrichment,
idle speed, etc.
Exhaust Gas Treatment
3-way catalyst
Oxidation catalyst
3-way plus oxidation catalyst
Thermal reactors
Status of Development
Extensive efforts currently underway
by virtually every auto manufacturer;
for example, bypass feedback carburetion
and feedback fuel injection
A key part of a system to control CO
since much of the CO is emitted during
the first few minutes of vehicle opera-
tion after startup. Active development
work by all manufacturers.
Has been in use for several years.
Ford, GM and Chrysler are all developing
electronic EGR systems. Not primarily a
CO control technique, but this can be
used to improve CO performance.
Systems are currently in use on some
vehicles and will be used nearly across
the board by 1983.
Currently available and receiving con-
siderable development work.
Currently available and receiving con-
siderable development work.
Currently available and receiving con-
siderable development work.
Currently used in some exhaust control
systems.
3-46
-------
TABLE 3-23 (Cont'd)
Type of Control
Al ternat ive Enq ines
Stratified charge
"Fast Burn" (May "Fireball",
MCA-JET, NAPS-Z)
Diesel
Gas turbine
Steam engine
Electric
Alternate Fuels
Status of Development
One variation is currently available
through Honda and other types are cur-
rently receiving extensive development
work (Ford PROCO and and Texaco TCCS).
Several manufacturers are considering
"fast burn" concepts.
Numerous models available.
Currently undergoing extensive develop-
ment by several major manufacturers.
Has been tested by several investigators.
Currently available via special produc-
t ion.
Liquified gaseous and gaseous fuels are
considered to have practical problems
like storage and availability. Some
advanced research is ongoing, however, on
hydrogen generators. The use of ethanol/
gasoline blends ("gasohol") is currently
receiving widespread attention.
"Exhaust gas recirculati
ion
Source: References 16 and 27
3-47
-------
3.7.2 Carbon Monoxide Emission Reduction Benefits
The CO emission reduction potential for those controls listed in
Table 3-23 are variable. The literature, however, does not quantify the
CO emission reductions for individual control elements. The effectiveness of
a vehicle's control system depends upon numerous factors including specific
engine design characteristics and the target emission standard. The reader
is referred to the Special Bibliography for reports containing more detailed
discussions on emission reduction benefits of the various CO controls for new
vehicles.
3.7.3 Costs for New Mobile Source Controls
Costs are available for many of the LDV emission control subsystems and
components listed in Table 3-23. These are shown in Table 3-24. The actual
costs associated with CO control for a given motor vehicle, however, depend
upon such factors as the particular design characteristics of the vehicle and
its engine, the actual control technology used, the type and rate of produc-
tion of the components, and the target emission standard. With the vari-
ability from one engine to the next, as well as the available CO control
options, more definitive costing is beyond the scope of this chapter. The
cost attributable to CO control alone is difficult to determine. There are
several reasons for this. First, emission control systems are typically de-
signed to meet emission standards that include HC, CO, and NO requirements.
Therefore, the system is designed to provide acceptable control of all three
pollutants. Second, some components and subsystems control more than just
one pollutant. For example, an oxidation catalyst can control both CO and
3-48
-------
TABLE 3-24
LIGHT DUTY VEHICLE EMISSION CONTROL
COMPONENT RETAIL COST
Component/Subsystem Consumer Cost3
Feedback Controlled Carburetor 49~75
Electronic Fuel Injection System 95~550
Mechanical Fuel Injection System ^70
Electronic Ignition System 22-30
Air Injection System 45-120
Aspirator System 3.23
Closed Loop Control System 133-172
Electronic Control Unit (ECU) 32-84
Oxygen Sensor 16-35
Throttle Position Sensor 1_y
Coolant Temperature Sensor 2-5
Crankshaft Position Sensor yb
MAP/BAP Sensor0 15
Inlet Air Temperature Sensor 5
Wiring Harness for Electronic Controls 19-21
Oxidation Catalyst 58-140
3-way Catalyst 113-200
Heat Shield for 3~way Catalyst 8
Deceleration System 21
Idle/Deceleration System 4
8 Dollar basis (e.g. 1978 dollars) was generally not specified.
Includes requisite engine modifications.
Manifold Absolute Pressure/Barometric Atmospheric Pressure Sensor
Source:Reference 27
3-49
-------
HC emissions, and the cost for just CO control could range from the entire
cost of the catalyst (all the cost apportioned to CO control) and zero (all
the cost apportioned to HC control). Extending the relative apportioning
ranges to other components will yield a large overall range of costs for con-
trol of any given pollutant, including CO. An alternative procedure could be
to take the entire cost of the emission control system and apportion it
equally to all the pollutants. For a system which is designed to control
three pollutants (HC, CO, N0x) the entire system cost would be divided by
three. It is realized that the major advantage of this approach is simplicity,
Fourth, the components and subsystems used on vehicles, in addition to con-
trolling pollutants, may also be used for other purposes; for example, drive-
ability and/or performance and/or fuel economy improvements. An example of
this is fuel injection, which in addition to providing emission control bene-
fits may be able to provide driveability/performance/fuel economy benefits.
Unfortunately, as is the case with emission control components and subsystems,
there is no universally accepted way to apportion these costs. An example of
the issues involved in a cost analysis for a given pollutant can be found in
the Rulemaking Docket for EPA's revision of the oxidant (ozone) standard.
During the rulemaking on this standard mobile source costs to control oxidants
was an issue. In a memorandum from EPA's Office of Mobile Source Air Pol-
lution Control Program to EPA's Office of Air Quality Planning and Standards
dated 19 December 1978, the apportioning of the cost for all mobile sources
to oxidant control is treated. This document can be found in the Rulemaking
Docket on the revised oxidant standard as Docket Number OAQPS 78-8 and it is
3-50
-------
included as Reference 28 in the list of references for Chapter 3. Using
the same method, cost apportionment for mobile source CO controls on a per
vehicle or per engine basis are shown in Table 3-25, for gasoline fueled
power plants. Cost estimates for new arid in-use gas turbine aircraft for
point source episodes are more complex and the reader is referred to Ref-
erence 40 for appropriate cost information. Other references to cost infor-
mation are included as References 27 through 39 at the end of this chapter.
3.7.4 Energy Requirements for New Mobile Source Controls
The energy requirements for new mobile source controls are measured as
either an increase or decrease in vehicle fuel economy. The impact on fuel
economy due to control of one or more than one exhaust pollutant is a func-
tion of the level of control, the technology used, the lead time, the
emphasis given to fuel economy by the designers, etc. etc. Therefore,
apportioning the changes in fuel economy (either positive or negative) to
control of a pollutant or pollutants is difficult. For example the average
new-car fleet fuel economy for model year 1974 was about 5.95 kilometers
per litre (14 MPG) and the CO emission standard (1975 FTP basis) was
about 14.3 g/km (23 g/mi) CO. In model year 1975 the average new-car fuel
economy was over 6.38 kilometers per litre (15 MPG) and the CO emission stan-
dard was 9.32 g/km (15 g/mi) CO. Considering only the CO difference and
the fuel economy difference might lead to the conclusion that tighter CO
control results in fuel economy improvements. However, because of the
other factors noted above, it would not be appropriate to take credit for
fuel economy improvements due to emission control.
3-51
-------
Another factor which must be taken into account in evaluating fuel
economy/emissions interactions is that fuel economy is not a free variable.
Fuel economy is now regulated under the Energy Policy and Conser-
vation Act, and car and light truck manufacturers have to meet fleet fuel
economy standards that were in effect for model years 1978 and 1979 and will
become increasingly stringent for model years 1980 through 1985. Table
3-26 summarizes these regulations. It appears now that if the appropriate
technical approaches are used, both the fuel economy standards and the
emission standards can be met, thereby making the positive or negative
impacts of emission control on fuel economy a moot question.
TABLE 3-25
CO CONTROL COSTS FOR DIFFERENT FEDERAL LEVELS
OF CONTROL FOR NEW GASOLINE FUELED POWER PLANTS
LDc ! LDI r , Progressive A Cost Increase Over Uncontrolled Engines
Federal Standard $ (1978 basTil ~~
2k.2 g/km (39 g/mi) f3
9-32 g/km (15 g/mi)a +55.32
4.35 g/km (7 g/mi) +72.32
2.11 g/km (3.4 g/mi) +88.66
HDV
Federal Standard
1.5% by volume I 8
53.6 g/kwhr (40 g/bhp-hr) ' +5.82
33.5 g/kwhr (25 g/bhp-hr) +3-99
Motorcycles
17 g/km (27.4 g/mi) 15 25
12 g/km (19-3 g/mi) +4.5
3 LOT only
Source: References 20 and 28
3-52
-------
TABLE 3-26
FEDERAL REGULATIONS FOR LIGHT-DUTY VEHICLE FUEL ECONOMY
Mode1 Year Minimum Fuel Economy
Kilometers/Litre (Miles/Gallon)
Combined Urban and Highway Cycle
J978 7.59 (18.0)
979 8.01 (19.0)
8.43 (20.0)
9.27 (22.0)
983 n.O (26.0)
n.*» (27.0)
11.6 (27.5)
Source: Energy Policy and Conservation Act.
It should be pointed out that CO control and its effects on fuel
economy has been a less controversial subject than either HC or NO con-
X
trol. This is because many of the techniques used to control CO from the
engine tend to be directionally the same as those that improve fuel econ-
omy. For example, the CO control approaches to reduce cold start emissions
are directionally positive for fuel economy since when the engine is run-
ning rich and producing a large quantity of CO on cold start, it is also
running richer than may be considered desirable from the fuel economy
standpoint. However, as discussed above, this approach and others, such
3-53
-------
as operation with high charge dilution, are not specifically credited with
fuel economy benefits. Several manufacturers have demonstrated the capa-
bility to improve fuel economy while achieving very low exhaust emission
levels.27 The benefits of electronic emission control systems have not
been adequately quantified yet, but their adaptation may also be utilized
to support combustion with highly dilute mixtures and lean air/fuel ratios
which complement effective CO control. The Special Bibliography contains
references which provide more detailed information on energy requirements
for control alternatives.
3.7.5 High Altitude Control for New Mobile Sources
For 1979 and 1980 only eleven manufacturers have reported that they
will offer high altitude compensation systems and several have stated that
these will be offered only as options on a limited number of engine/vehicle
combinations at extra costs.27 Consequently, there is a great potential
that many of the new models will be sold with low altitude calibrations
during model years 1979 and 1980. The Clean Air Act (CAA) amendments pro-
vide that EPA may promulgate proportional reduction standards for high alti-
tude during the 1981 to 1983 model years. EPA anticipates proposing light
duty vehicle high altitude proportional reduction standards for these model
years of about 0.30 g/km (0.48 g/mi) HC, 3.1 g/km (5.0 g/mi) CO and 0.62 g/km
(1.09 g/mi) N0x. A great deal of work will be necessary in the near future
for many of the manufacturers to develop appropriate control technology
for control of CO at high altitude.
3-54
-------
3.7.6 Environmental Impact of New Model Source Controls
Three of the new mobile source CO control alternatives have potentially
adverse secondary emissions. These are the oxidation catalyst, the three-
way catalyst, and the diesel engine. Oxidation catalysts can oxidize a por-
tion of the sulfur dioxide in the exhaust to sulfuric acid. Although the
quantity of sulfuric acid formed is relatively small, it may be possible to
have high localized levels of sulfuric acid along heavily traveled roads.
Extensive work sponsored by the EPA and major auto manufacturers has been
done to examine this problem. More definitive actions await health effects
data from EPA's Office of Research and Development, which has been studying
the problem for several years. Unfortunately, definitive answers have not
been generated. In addition, work is continuing by EPA and the auto manu-
facturers to evaluate other unregulated emissions from catalyst equipped
vehicles. The Special Bibliography contains sources which present the
results of much of this work.
Three-way catalysts can produce reduced species if operated in a rich
air/fuel mode. Reduced species such as HCN have been studied by EPA and
no specific action is contemplated at this point in time. Ammonia (NH3)
emissions have also been studied. If a system containing a 3-way catalyst
operates too lean, the environmental concerns are similar to those of the
oxidation catalyst, discussed above.
Diesel engines are a source of airborne particulates. Diesel particu-
lates are currently being investigated by EPA. Tests conducted to date
show that diesel engines discharge many times the amount of particulates
3-55
-------
generated by comparably sized gasoline engines. The Special Bibliography
contains sources which discuss this problem in more detail.
3.8 CARBON MONOXIDE- CONTROLS APPLIED TO VEHICLES AFTER SALE AND OTHER
MEASURES AVAILABLE TO STATES AND/OR LOCAL GOVERNMENTS
Inspection and Maintenance (I/M) Programs and Other Transportation
Control Measures are two examples of approaches to apply controls to
vehicles after their initial sale. I/M Programs are discussed separately
from other Transportation Control Programs for two reasons: (1) I/M pro-
grams are treated in a general manner in the Clean Air Act compared to
other Transportation Control, and (2) The office within EPA that is respon-
sible for I/M Programs is different from the office that is responsible for
other Transportation Control Measures.
Section 172(b)(ll) of the Clean Air Act gives three requirements for
a state to meet, if the state wishes to obtain a delay (from 1982 to 1987)
in meeting the National Ambient Air Quality Standards. One of these is a
requirement that the state establish a specific schedule for implementation
of an I/M Program.
I/M Programs, therefore, have been given special consideration by
Congress. Since carbon monoxide is primarily a mobile source pollutant,
I/M Programs can be considered an important control technique, and EPA is
committed to assist states in the design, development, implementation, and
evaluation of I/M Programs.
The discussion of I/M Programs in this Chapter provides a general over-
view of the subject of I/M. Since each I/M Program will be to some extent
3-56
-------
unique, specific details of all possible I/M Programs cannot be included.
However EPA will provide technical assistance to states in their efforts
to implement I/M Programs, to ensure that the most effective benefits are
obtained, and that the programs are tailored for any specific local situa-
tions that may exist.
Assistance in the I/M area can be obtained from:
Director, Emission Control Technology Division
Attention: I/M Staff
U.S. Environmental Protection Agency
Motor Vehicle Emission Laboratory
2565 Plymouth Road
Ann Arbor, Michigan 48105
3-8-1 Inspection/Maintenance Control Techniques
This section focuses on inspection/maintenance (I/M) techniques and
provides information on the emission reduction approaches, costs, benefits,
energy requirements, and environmental impacts.
3.8.1.1 Types of I/M Control Strategy Approaches
There are five recognized inspection alternatives for an inspection/
maintenance program.21 They are:
1) idle mode test conducted at state inspection stations,
2) idle mode test conducted at inspection stations operated by a
contractor to the state,
3) idle mode test conducted at privately owned service stations
and garages,
4) loaded mode test conducted at state inspection stations, and
3-57
-------
5) loaded mode test conducted at inspection stations operated by
a contractor to the state.
Table 3-27 summarizes the characteristics of idle and loaded mode test-
ing procedures.21 EPA and private research organizations have found idle
mode testing to be virtually as effective as the loaded mode test in identi-
fying gross HC and CO emitters, and thus a viable inspection technique.
The maintenance phase of an I/M program involves the repair of those
vehicles which were identified during inspection as high emitters. The
average quantity of repair work required on those vehicles failing inspec-
tion depends on the emission standards and the level of preventive mainte-
nance provided by vehicle owners. Information compiled by existing I/M
programs indicates the major causes of high carbon monoxide exhaust emis-
sion are:
1) carburetor out of adjustment,
2) air/fuel mixture imbalances, and
3) malfunction or disablement of emission control devices.
Table 3-28 contains information reported by the Portland, Oregon I/M
program on the types of maintenance required for vehicles failing inspec-
tion. Reference 21 contains more detailed information regarding maintenance
and its role in a successful I/M program.
3.8.1.2 Costs for I/M Programs
There are two kinds of costs for an I/M program:
1) the initial investment and operating costs for the inspection
facilities, and
3-58
-------
d>
c
D
Loaded Mode:
Transient Test
Engine operated
0,
1
L.
C
TO
-Q
L.
ZJ
•a
d>
TO
— a)
E 73
in o
i
O
75
0)
-a
o
a
0
4-1
0)
4-1
O
d>
a.
X
LU
Q_
h-
U_
-C
4-1
•~
est correlation
TJ
C
TO
4J
L.
d)
C
X)
TO
TO
1
TO
C
-o
power absorption
meter required
4-1
O
<4-
14—
TJ
d>
O
U
cn
c
>
o
d)
cn
TO
• ~ L.
>• d)
c u
c
TO 4-1
O in
O
•M E
in
d) c
O
u_
Computer needed 1
jx
TO TO
C —
TO 4-J
TO C
TO
"O -C
CO
c XT
1
c
" H
d)
-0 d>
0 4-J
3! TO
4-1
TJ CO
0)
TJ >J
TO TJ|
0 TO
4-J 1
CO
in
1 C
3 O
— 4-J
in —
i_ c
<1J O
-a o
c
u
4-J
TO
C
O
4-J
TO
1—
rt)
cr JZ
cc o
-o
4-1
TO
U
,n
^
Q.
T3
0)
-Q
4-J
0
c
c
TO
O
^ 1
in
0)
h-
in
(!)
4-J
|
• «.
0
TO
14-
L.
TO
Q.
0)
L- C
-C *c
o —
— TO
-C L.
in
4-1
c
d)
4-J
in
• — i
Carburetor a
4-J
in
d>
4-1
cn
c
T3
d)
-a
TO
d>
-Q
C
TO
O
d>
cr
c
d)
d)
0
in
14-
O
in
in
O
c
cn
TO
0
)
c
TO
in
4-J
c in
d) c
E 0
4-J —
in 4-J
rj o
— i C
"O Z)
TO <4-
TO TO
E E
1_
d)
JZ
4-J
*v.
^*
-Q
-a
d>
4->
TO
jj
Q.
C
TO
O
ist systems
• \u
d)
4-J
TO
>
Q.
O
a_
d)
E
4-J
4-1
in
d)
4-J
TO
E
c £
— a)
E E
Q.
in —
d> D
L- cr
— d)
cr-o
d) C
a: ro
o
u
O
c
TO
O
in
c
O
4-1
U
c
14-
TO
21
in -Q
c d)
O 4-J
— O
4-1 d)
— 4-J
"O d)
c -o
o
O d)
-Q
-a
"o o
TO C
O
TO
J- E
d)
_^_J ^..J
C C
ZJ TO
(U
cc
LA
O
CO
3-59
-------
2) the repair costs incurred for those vehicles which do not meet
the emission standards.
The costs of inspection facilities vary significantly according to the
sophistication of the program and the type of safety program existing in
the area. These costs are borne by the state or, if a contractor approach
is selected, by the private firm. The operating costs and repayment of the
initial investment would be covered by revenues derived from a fee charged
the owner when the vehicle is inspected. Experience has shown that most
inspections cost between $4 and $10, with the higher figure including both
emissions and safety inspection.21
TABLE 3-28
DISTRIBUTION OF THE TYPES OF REPAIRS
REQUIRED FOR VEHICLES FAILING INSPECTION
Repair Needed Percent Undergoing Repair
Carburetor adjustment 78
Tune-up ]k
Engine overhaul 1
Valves 1
Other 6
TOTAL 100
Source: Reference 21
3-60
-------
In addition to the inspection fee, those individuals whose vehicles
do not meet the emission standards will incur repair costs. The average
cost of repair has been reported for several existing I/M programs. In
New Jersey, the average cost of repairs has been $32.40; for Arizona,
$23.40; and for Oregon, $16.00.21 The actual number of vehicles requiring
maintenance as well as the cost is determined by the stringency of the emis-
sion standards established by the state.
3.8.1.3 Benefits of I/M Program
In order to obtain full benefits from an I/M program certain minimal
requirements must be met:
1) all vehicles for which emission reductions are claimed must receive
regular, periodic inspections
2) to ensure that failed vehicles receive the maintenance necessary
to achieve compliance with the inspection standards, they should be
required to pass a retest following maintenance
3) quality control measures, such as routine maintenance, calibra-
tion and inspection of all I/M equipment, and routine auditing of inspec-
tion results, must be followed to ensure the reliability of the inspection
system and accuracy of the equipment.
Beyond the minimum requirements, various other facets of an I/M program
can influence the emissions reductions to be achieved.
Type of Inspection
While currently available data indicate no overall difference in the
CO or HC emission reductions obtained through the use of loaded or idle
3-61
-------
mode testing, loaded mode testing is considered to be a better indicator
of the actual emissions of the vehicle in-use and it provides better
diagnostic information.
Inspection
Various engine component and emission control devices can deteriorate
or be disabled and have no noticeable effect on the way a car drives or
on its fuel consumption. The performance of periodic inspection provides
a suitable deterrent to either maladjustment or disablement because of the
threat of not meeting the required standards.
Mechanics Training
The air quality benefit from an I/M program is dependent, in part, on
the ability of the service industry to properly perform the repair work
necessary to lower emissions. Some savings in repair costs may also result
from the proper training since the mechanics would be more familiar with
the problems and the best solutions for them.
Vehicle Exemptions
The total emission reductions that result from an I/M program are
directly dependent on the number and types of vehicles inspected and the
requirement that maintenance be performed. In some cases, it may be
desirable to exempt vehicles that include different control technology
(diesels, Stratified charge, LPG/LNG, etc.). In some cases, it may also
be desirable to exempt vehicles when the estimated repair cost is a major
percentage of the vehicle value.
3-62
-------
Frequency of Testing
Most existing I/M programs require annual inspection. This frequency
is justified on the basis that it minimizes costs and maximizes public
acceptance while maintaining a reasonably high level of emission reduction.
When annual inspection is required for vehicle registration it helps enforce-
ment of an I/M program. A semi-annual program would involve substantially
higher program costs arising from the need for a greater number of inspect-
tion lanes, as compared to an annual inspection program. A biennial program,
while certainly providing some emission benefits, will lose some of the
effectiveness of an annual program because cars may be allowed to deteriorate
to a higher level.
Emission Standards
Most importantly, the I/M emission standards, or "cut points," deter-
mine the overall emission reduction potential of the program. The cut
point is the level of emissions which distinguishes between those vehicles
requiring emissions-related maintenance and those that do not. The cut
points that are selected define a "stringency factor" which is a measure
of the rigor of the program based on the estimated fraction of the vehicle
population whose emissions would exceed cut points for carbon monoxide in
the absence of an I/M program.
There are two basic concerns that constrain the selection of I/M emis-
sion standards to determine the emission reduction potential. While I/M
standards or "cut points" should be set to achieve a desired emission
reduction, the cut point should be limited to a level that will be accept-
able to both the general public and the repair industry. As experienced
3-63
-------
by other programs, negative public sentiments may result if an excessive
volume of vehicles do not comply with I/M standards at first inspection.
Further difficulties will arise if the total of the noncomplyinq vehicles
exceed the available capacity of the repair industry. The necessary
vehicle maintenance will be compromised under these conditions. Cut points
must be set at a level where potential emission reduction benefits are
maximized while impacts to the public are minimized. As stated above, emis-
sion reductions achieved with any particular I/M program are a result of a
combination of the emission reductions obtained through the optimal selec-
tion of various options. Table 3-29 lists credits for CO in percent emis-
sion reductions that can be achieved in 1987 through an inspection/mainte-
nance program which was implemented in 1982. The "basic" reductions (i.e.,
those that are achieved through an annual inspection of light-duty vehicles)
are broken down by Technology I and Technology II vehicles and by Technology
III and Technology IV vehicles.
Technology I vehicles include those light-duty vehicles subject to pre-
1975 federal emission standards; Technology II vehicles are subject to 1975
and later model year federal exhaust emission standards; Technology III and
Technology IV vehicles are subject to 1980 and 1981 federal exhaust emission
standards, respectively. A review of these data indicates that a 20 percent
stringency factor I/M program implemented on all light-duty vehicles (LDVs)
would achieve the policy required 25 percent reduction in CO for LDVs, and
that larger emission reductions are possible with mechanic's training. (The
reader is referred to the proposed revision of Appendix N of Reference 41
3-64
-------
en
CM
OO
LU
_J
CQ
t_
f^
tn
— Q
< -J
03
CO CM
Z OO
o en
K- "~~ >
o z —
Q to
LU Q
C£ LU —
h- —
Z Z — m
O LU >
<" LU cnfj
co — i o
— D_ , —
21 ^ Q
LU — C
Q Z u
Z < jj
< a: |_
CD
CO O
_i a:
LU CL j
uj 2: .i
— ' "^ 4
z~ - i
oo — -j
' — c
CO t*5 a
CO LU
— ID —
Z a in
LU >- > ^~
r>» CDQ —
Q- OO O — 1 E
I- en -- — 0 E
H- vs
O
>
O
c
A - cnoo o
vo en i — , -3" -3-
LA -3- -3- -3- -3-
— • — •— , — _
— o en o LA
r*v CA i— o en
en en en cnoo
VO vo vo
r^ CM LA r**. oo
— CM CM CM CM
— vo en r^. o
o o CM CM r^»
r^ \o LA -3- -3-
r**^ oo o r^ oo
LA en LA co —
o en en oo en
^_
o o o o o
— CM OO -3" LA
O CD
— c
c •—
03 C
r- _t- .—
•M O OJ
•— : .i
L- Q. r^x D t
c co CM' "E .-
O — — -3 f
~^ c
"~ C u
I/I 1— (/)
w 1 — w a
— LU — ,
Ei i — *•
— 1 L- -C L
LU — O 4J -
CO 4-1 ?
o o c is
' 2Z OJ O
f^ "+- (L
00 E T3 2.
cn oj c '- CD
L. (/) 5
=> 0 0)
• a:
i
) -o
> c
OJ
1^
0
o
1
CO
i
cn
Q
O
-3"
1
<
CL
LU
^
CO
cn
4_»
_
CD
3
"^
3-65
-------
for a more detailed discussion.) The final revised Appendix N should be
consulted when it appears as a final rule in the Federal Register.
Warranty Provisions
The Emission Control System Performance Warranty contained in Section
207(b) of the Clean Air Act provides warranty coverage to motorists in areas
having an I/M program. The Emission Performance Warranty, upon promulgation
of a regulation by EPA, will require the automobile manufacturer to bear the
cost of repair of any properly maintained and operated vehicle which fails
an EPA established emissions test within 24 months or 38,600 kilometers
(24,000 miles), whichever occurs first, of the original sale to the ulti-
mate purchaser. After this period, the warranty applies only to catalytic
converters, thermal reactors or other components installed on or in a
vehicle for the sole or primary purpose of reducing vehicle emissions.
These warranty provisions are thus an additional benefit to individuals
residing in areas with an I/M program.
3.8.1.4 Energy Requirements for I/M Program
A slight energy benefit is likely to result from the application of
an I/M program rather than an energy penalty, particularly if mechanics
have been trained in emission oriented maintenance. Fuel savings can
result on those vehicles that are in need of repair or in a state of
maladjustment. The extent of such benefits have recently been quantified
by EPA.42
3.8.2 Transportation Control Programs
In addition to I/M programs, there are several other Transportation
Control Measures that could possibly be used by state and/or local
3-66
-------
authorities to control motor vehicle-related carbon monoxide emissions.
A list of some of these measures can be found in Section 108(f) of
the Clean Air Act:
1) programs to control vapor emissions from fuel transfer and
storage operations and operations using solvents;
2) programs for improved public transit;
3) programs to establish exclusive bus and carpool lanes and areawide
carpool programs;
4) programs to limit portions of road surfaces or certain sections
of the metropolitan areas to the use of common carriers, both as to time
and place;
5) programs for long-range transit improvements involving new trans-
portation policies and transportation facilities or major changes in exis-
ting facilities;
6) programs to control on-street parking;
7) programs to construct new parking facilities and operate existing
parking facilities for the purpose of park and ride lots and fringe parking;
8) programs to limit portions of road surfaces or certain sections of
the metropolitan area to the use of nonmotorized vehicles or pedestrian us,,
both as to time and places;
9) provisions for employer participation in programs to encourage
carpool ing, vanpooling, mass transit, bicycling, and walking;
10} programs for secure bicycle storage facilities and other facili-
ties, including bicycle lanes, for the convenience and protection of
bicyclist, in both public and private areas;
3-67
-------
11) programs of staggered hours of work;
12) programs to institute road user charges, tolls, or differential
rates to discourage single occupancy automobile trips;
13) programs to control extended idling of vehicles;
14) programs to reduce emissions by improvements in traffic flow;
15) programs for the conversion of fleet vehicles to cleaner engines
or fuels, or to otherwise control fleet vehicle operations;
16) programs for retrofit of emission devices or controls on vehicles
and engines, other than light-duty vehicles, not subject to regulations
under section 202 of Title II of this Act; and
17) programs to reduce motor vehicle emissions which are caused by
extreme cold start conditions.
EPA is in the process of preparing reports, in conjunction with the
U.S. Department of Transportation, that cover each of these areas. At the
time of the preparation of this document, only one has been completed:
report EPA 400/2-78-002a, Air Quality Impacts of Transit Improvements, Pref-
erential Lane, and Carpool/Vanpool Programs.
Questions about the status of other reports on the above-listed
subjects, and requests for information and assistance in this general sub-
ject area can be directed to the EPA office listed below:
Director
Office of Transportation and Land Use Planning (AN-445)
U.S. Environmental Protection Agency
401 M. St., S.W.
Washington, D.C. 20460
3-68
-------
3-8.2.1 Transportation Control Strategy Approaches
Transportation-related air quality problems can be either localized or
regional. Localized problems generally result in CO concentrations exceeding
either the one-hour, or more likely, the eight-hour CO National Ambient Air
Quality Standard. Localized violations of the standards are usually asso-
ciated with high traffic volumes and congested traffic conditions frequently
found in densely populated urban areas. Regional transportation-related air
quality problems are typically a result of vehicle and stationary source
hydrocarbon and nitrogen oxide emissions reacting in the atmosphere to pro-
duce oxidant pollutants. Transportation-related air pollution problems of
localized and regional types are illustrated in Table 3-30.
The distinction between the pollutants CO and oxidant is important.
Transportation control programs designed for localized problems are dif-
ferent than those for regional air quality problems. For example, a trans-
portation systems management (TSM) program to implement a reserved lane for
carpools and buses on a particular freeway may reduce CO emissions in the
vicinity of the freeway, but is unlikely to have a noticeable impact on
regional oxidant emissions. Similarly, a regional car pool program may
contribute to a reduction in hydrocarbon and nitrogen oxide emissions, but
generally may have less impact on localized CO concentrations.
Four transportation control programs have been identified as having
the greatest potential for controlling localized violations of the CO
standards in a cost-effective manner.^ These programs were identified
through a comprehensive review of both operational and proposed transporta-
tion control programs. They are:
3-69
-------
0
1
00
UJ
CO
1—
CO
UJ
CO
0
a:
0.
z
o
1—
0
Q_
a:
Q
UJ
h-
<^
_J
UJ
cd
z
O
h-
1—
O
Q_
CO
•y
<£
fY*
|—
UJ
>
<
1-
co
^3
_J
_J
__
E
i-
03 cn
L. C
4-1
•a "3
a) jo
U l-
0) 4-<
— c
(U O
co o
— 03
U-
JZ M- D.
cn 03 O
._ 1_ 4-J
X h- CO
• •
4-J
o
03
a.
E
03
r- 0)
03 1-
0 <
Q.
h-
C.
o
03
C TJ
O <
4-1 in
0 c
(U O
in —
V- 4-J
<1) 03
4J O
c O
• •
oo
L.
>^
4-J
— T3
. — L.
03 03
D TD
0* C
03
U 4-J
— CO
•— Nfc
cn
c
—
T3
•>
«
cn
•
(U
^-^
in
2
0
F—
u_
T3
C
03
in
to Freeway
Arterial s
"O
i_
(U
L.
03
^~
3
0
•—
f~
0)
>
jr
cn
•—
X
•
03
(1)
C
03
JO
03
L.
0)
o
•
m
0)
E
ZI
— in
o -o
> 03 ^^
c 03 in
O L- ~0 C
4-J (1) O
-o c »- —
(U d) 3 4-J
in o in 03
03 C 03 O
CO O 0) O
-— o z: _j
4-J 4-J OO
Q)
E
~^^
cn
^— ;
s~~^
»- o 2; »-
Zf 0 0- 3
O 0 0- O
x " x
o cr\
OO r— *— ' —
0)
T)
•—
4-J
C
03
4-1
_
0
a.
X
o
c
o
c
o
JO
^_
03
O
-a
M- E
o a)
r^.
0) JO
Q. 0
\— O-
Q)
N
•—
. —
03
O
O
0)
E
^v.
cn
y
0 0_
O Q-
0
O ro
Mr s-x
L.
Z!
O
X
•~"
>—
03
0
•—
E
_c
O
O
4-1
o
JZ
o_
•—
03
C
o
•—
cn
0)
ct:
1_
(1)
4-J s-**
a) 2:
E Q.
\ o_
cn
71 CO
0
0 •
vD O
•~
^J
c
03
-o
X
0
OO
-3-
(!)
U
C
O
i—
-------
1) freeway priority treatment for high occupancy vehicles;
2) arterial priority treatment for high occupancy vehicles; and
3) areawide carpool and vanpool programs
4) transit service improvement programs.
In order to quantitatively assess the air quality and related impacts
of interest, 20 prototype scenarios were analyzed.43 These prototype sce-
narios were designed to provide representative findings on the range of
travel, air quality/emission, fuel consumption, cost and economic impacts of
TSM programs which appear to have potential for localized or regional air
quality improvement. These scenarios are presented in Tables 3-31 and 3-32
respectively for localized and regional prototypes. The strategies
considered have the potential for achieving improvements in regional air-
quality -- especially when considerations of strategies which include strong
incentives and nonincentives (e.g., auto restricted zones, limited idle/engine
off, pricing, etc.) not within the scope of this report are included in the
total transportation plan. The strategies which appear to have the greatest
potential for achieving improvements in localized CO air quality in a cost
effective manner include:1*3
1) with-flow freeway lanes reserved for buses and carpools;
2) contraflow bus lanes on freeways;
3) metered freeway access ramps with bus by-pass lanes;
4) contraflow bus lanes on major one-way arterial pairs;
5) provision of high level express bus service with reduced fares,
operating in mixed traffic on major arterials or freeways;
3-71
-------
CO
O
,**,
0
o
c o
in X
O i/i
U L-
It
in —
g —
lu O
17>
O VC
a. cr>
—
C 4>
4_)
It 1-
1- 4>
41 O-
0 ""'
a>
c
_ i'L
It — C
Q. 4) 4) 0
It C — —
u o^ a *j
a:
CO
CO
LU
CO
<£
LU
O
CO
LU
a.
*^
\—
2
o
QX
a.
o
1 i 1
LJU
M
t—t
_J
O
O
LU
a:
h—
a:
o
CO
o
^jl^
a.
I-H
O
LU
•
r—
^p-»
in
41 C
o 0
4> in
u in
O «) —
o >*• E
J4j i"
< „.
^ m •—
It E U
41 ^ It
a. cnu.
• ^ -o
. — *j
< u
in 41
c c u-
0 .2 <
u it E
Impa
Concent r
ptor Fro
5
cc
L.
c
a. O
— (ft
It 1-
O 4)
Q. I/I
K"O
sJ
3
51
4)
3
It
4)
in
it
CD
T-,
J§
Typical
Dispersi
-
el
1\
o|
1
it
X
It
CO
J<
V
°" 0 1
• TS 3
. T "5
0 *
0 —
u O
o o !c
It X 4)
I *
u
41 Ol
U C
I- It
41 £.
It
3 3
ui X* "3
it >
CO
2
oo
O
>-
CO
g I
41 .-
Q.U-
X It
0) 1-
4^
"S X
£
O>
C
It
in
in
it
a
X
_o
in
3
JD
T3 in
C w
It U
It
It
^V
I-
E it
o» o
a: w_
L. 4--
U is>
4-1 QJ .
C 0
t..
in
t. 41
• it
41 —
C
it in
— in
it
— a
2 '
0 X
a -Q
in in
ii
T3 T)
Oi c in
> It *J
1- U
01 -
E it
l"
it in
L. 4)
.s
01 —
C
it in
— in
it
— a
5 '
o x
a -o
•^
in in
3 3
-a J3
•D "D
4> C in
> It 4-1
u U
41 - It
m 01 a.
41 -D
•C It
4) "O *J
C U
- It It
2 -I
It 01 —
— o
.- o
i/l > —
3 V- JD
JD 4) It
4)
•O in —
c 4^ a
— 3 >» —
«- XI ** ^1
01 it
S I
I O
— it
S r
3 l. D. (t
u — e
It — T>
JO. C
O 11 3 0)
£ 0 2 .0 2
It It *J fl)
u in i- 3 _
« 4) O O .-
3-72
-------
co
o
UJ
O_
O
OL
< o a:
l- 2 <
2 — LU
LU >- >-
C£ LU LU
<_> Q- Q-
20^-
O» LU LU —
< 2 _l
0 O 0,
O 2 CO CO CO
< -J t- — LU O O
t3 U. O — Zj Ij <
z 2 r u. r °
O 2 O ^- u.
<_) — o
•!: ^
I •— LA —
O O"1
— CM
OO CM
PA JT
CM UJ
f^\ Z
1 h-
o:
uj o
oo
**
< ae o o
£ K 0, — LA
!S > < ° *~
DC
£1 E.
UJ LU LU <
23 oe o K LA
< LU K X
X 0, > _
co o
<
PROTOTYPE SCENARIO
BRIEF TITLE*
Carpool/Vanpool Program, Medium
Size City; Favorable Impacts
PA
CO
o
LA
1
**
LA
—
Carpool/Vanpool Program, Large
City; Favorable Impacts
PA
d
00
d
i
**
LA
CM
d
z
Reserved Bus/Pool Lanes, Ramp
Metering, and Bus By-Pass Lanes <
All Appropriate Freeway; Modest
Impacts
LA
CM
LA
1
u
01 -D
L. 1)
.cenarios except #11 are for a "la
a t ! on ) .
lated at 75°F assuming uninterrupt
— a 'Z
3-73
-------
6) provision of high level express bus service (possibly with reduced
fares), combined with a reserved lane for buses and carpools on the appro-
priate freeway facility; and
7) provision of high level express bus service (possibly with reduced
fares), combined with a reserved median lane for buses and bus preemption of
traffic signals on an appropriate arterial.
3.8.2.2 Emission Reduction Benefits of Transportation Control Programs
The freeway-based localized prototype scenarios (Scenarios 1-8, Table
3-30) are likely to achieve reductions on overall peak hour corridor traffic
volumes ranging between 1.5 percent and 7 percent. The arterial scenarios
analyzed (Scenarios 9 and 10) can also promote 4 to 15 percent reductions in
peak hour vehicular volumes. As is true for the freeway scenarios, the
attainment of such reductions is highly dependent upon the specific setting
in which such strategies may be implemented. However, the percentage reduc-
tions in vehicular volumes for arterials are based on smaller base volumes
and are not fully comparable to the corridor volumes in the freeway scenarios.
Generally the relative reductions in peak hour CO concentrations (under
typical, good dispersion conditions) shown in Table 3-31 are several percen-
tage points higher than the corresponding reductions in peak hour corridor
vehicle volumes but are generally several percentage points lower than the
corresponding reductions in peak direction freeway vehicle volumes. In Scenarios
6 and 7, CO concentrations are estimated to increase relative to the base condi-
tions. The increase in CO concentrations in several contraflow reserved freeway
lane scenarious reflect the travel and meteorological conditions assumed in those
scenarios. The results do not indicate that contraflow lanes, per se, have
3-74
-------
undersirable air quality effects, but rather illustrate the importance of
carefully analyzing the potential air quality effects of implementing a
contraflow lane on freeways carrying heavy traffic volumes in the "off-peak"
direction.
Scenarios 13 through 17 (Table 3-32) which involve the implementation
of reserved lanes on multiple radial freeways or arterials in a region,
generally resulted in total regional and work trip vehicle miles traveled
(VMT) reductions of less than 0.5 percent and 1.5 percent, respectively.
The small reductions in VMT are in large part related to the limited size of
the peak period radially-oriented central business district (CBD) travel
market in most large urban areas. For example, home to work trips and VMT
comprise approximately 20 percent and 30 percent of total weekday regional
person trips and VMT, respectively. Travel survey data suggest that only
15 percent of home to work trips are oriented to the CBD of urban areas
exceeding 1 million population. However, those urban areas with especially
large percentages of CBD-oriented travel could experience higher reductions
in VMT than those estimated in this study.
Despite their limitations in reducing regional air pollution emissions,
the freeway reserved lane strategies show considerable potential for reducing
peak period travel congestion along radial travel corridors when applied
under appropriate travel conditions. These strategies can contribute to
reductions in CO concentrations along heavily traveled freeways and can also
contribute to reductions of vehicular travel with CBD's.
3-75
-------
3.8.2.3 Costs of Transportation Control Programs
Table 3-31 presents the estimated capital and annual operating costs for
the localized scenarios. They represent order of magnitude estimates based on
costs published in the 1iterature.43
The largest individual cost item for all of the scenarios is for improve-
ments to express bus service. Generally, the geographic coverage and the fre-
quency of express bus service were assumed to increase significantly in order
to complement the reserved high occupancy vehicles (HOV) lanes and attract
large numbers of auto travelers. The annual cost of bus service shown in
Table 3-31 represents the incremental cost of providing bus service above that
assumed in the base case (i.e., "before" case).
The costs of implementing ramp metering and park-and-ride facilities
are also significant. With regard to the cost of park-and-ride lots, two
conditions are assumed. If use can be made of existing parking facilities
at shopping centers or other locations, the capital cost of such facilities
would be negligible. However, such arrangements may not be feasible in many
locations, so the full capital cost of constructing the park-and-ride facili-
ties is also presented. For both of these conditions, the cost of operating
and maintaining the park-and-ride lots is assumed to be a public cost.
Based on analyses of express bus operations in Minneapolis and Seattle,
annual operating revenues may only offset approximately 50 percent to 66 per-
cent of the annual operation and maintenance costs of express bus service
shown in Table 3-32. Consequently, sizeable annual operating subsidies may
be required to operate express bus services such as those assumed in the
3-76
-------
localized scenarios. If fare reductions are implemented, the subsidy require-
ments are likely to be even more significant. The economic impacts of the
regional scenarios are likely to be small. More details on the economic
impacts and the nature and magnitude of the impacts are contained in Reference
43.
3.8.2.4 Energy Requirements of Transportation Control Programs
Transportation control programs by their very nature promote lower fuel
consumption for the areas where they are implemented. Actual quantification
of this decrease is not available for the localized prototype scenarios shown
in Table 3-31. Estimated impacts for nine of the regional scenarios in a
large urban area are shown in Figure 3-7 with the most significant gains being
accomplished with carpool/vanpool program variations (7.2 to 14.2 million gal-
lons per weekday saved in highway fuel consumption).
3.8.2.5 Environmental Impact of Transportation Control Programs
The only potential adverse environmental impact associated with imple-
mentation of the scenarios listed in Tables 3-30 and 3-31 would be increased
particulate emissions and odor problems associated with the use of Diesel -
powered vehicles, i.e., buses. Diesel engine discharge much larger quantities
of particulates than gasoline engines. Odor is another problem resulting from
diesel engines. See Reference 22 for a more detailed discussion of diesel
engine emissions.
3.9 Special Bibliography for Chapter 3
The objective of this bibliography is to furnish more detailed
and basic information on each of the topics covered in this chapter. The
/
reference numbers refer to the references for Chapter 3.
3-77
-------
2.0%-
(-14.2 10«
GALLONS)
^ESTIMATED ABSOLUTE REGIONAL CHANGE IN ANNUAL HIGHWAY FUEL CONSUMPTION FOR
PROTOTYPE URBAN REGION OF APPROXIMATELY 2,500,000. 3,000,000 SMSA POLLU-
TION AND A BASE ANNUAL HIGHWAY FUEL CONSUMPTION OF 4955 MILLION LITRES
(1.309 MILLION GALLONS) FULL 365 DAYS, INCLUDING WEEKENDS AND HOLIDAYS)
Source: Reference 43
FIGURE 3-7 ESTIMATED IMPACTS FOR NINE REGIONAL SCENARIOS IN A LARGE
URBAN AREA: REGIONAL HIGHWAY FUEL CONSUMPTION
3-78
-------
3.9.1 TYPES OF CONTROL TECHNIQUES
3.9.1.1 New Mobile Source Controls
References 3, 16, and 27
3.9.1.2 In-Use Mobile Source Controls
References 2, 3, 20, 21, 27, and 43
3.9.1.3 Inspection/Maintenance Programs
Reference 21
3.9.1.4 Transportation Control Programs
Reference 43
3.9.2 EMISSION REDUCTION BENEFITS
3.9.2.1 New Mobile Source Controls
References 3, 16, and 27
3.9.2.2 In-Use Mobile Source Controls
References 2, 3, 20, 21, 22, 27, 41, 42, and 43
3.9.2.3 Inspection/Maintenance Programs
Reference 21
3.9.2.4 Transportation Control Programs
Reference 43
3-79
-------
3.9.3 COSTS
3.9.3.1 New Mobile Source Controls
References 18 through 29
3.9.3.2 In-Use Mobile Source Controls
References 21, 27, and 43
3.9.3.3 Inspection/Maintenance Programs
Reference 21
3.9.3.4 Transportation Control Programs
Reference 43
3.9.4 ENERGY REQUIREMENTS
3.9.4.1 New Mobile Source Controls
Reference 27
3.9.4.2 In-Use Mobile Source Controls
References 21 and 43
3.9.4.3 Inspection/Maintenance Programs
Reference 21
3.9.4.4 Transportation Control Programs
Reference 43
3.9.5 ENVIRONMENTAL IMPACTS
3-80
-------
3.9.5.1 New Mobile Source Controls
References 17, 18, 19, 21, 27, and 43
3.9.5.2 In-Use Mobile Source Controls
References 17, 18, 19, 21, 27, and 43
3.9.5.3 Inspection/Maintenance Programs
Reference 21
3.9.5.4 Transportation Control Programs
Reference 43
3-81
-------
REFERENCES FOR CHAPTER 3
1. National Air Quality and Emission Trends Report, 1976. EPA 450/1-7-002.
U.S. Environmental Protection Agency, Research Triangle Park, North
Carolina, December 1977.
2. Control Techniques for CO, N0x and HC Emissions from Mobile Sources.
Publication No. AP 66, U.S. Environmental Protection Agency, Research
Triangle Park, North Carolina, March 1970.
3. D.J. Patterson, N.A. Henein, Emissions From Combustion Engines and
Their Control. U.S. Environmental Protection Agency, Ann Arbor, Michigan,
1974
4. Title 40 Code of Federal Regulations, Protection of Environment,
July 1, 1977.
5. Motor Vehicle Manufacturers Association, Motor Vehicle Facts and
Figures, Detroit, Michigan, 1978.
6. Aircraft Technology Assessment Status of the Gas Turbine Program,
U.S. Environmental Protection Agency, Ann Arbor, Michigan, December 1976.
7. Review of Past Studies Addressing the Potential Impact of CO, HC, and
NOX Emissions from Commercial Aircraft on Air Quality, Technical Sup-
port Report for Regulatory Action, U.S. Environmental Protection Agency,
Ann Arbor, Michigan, March 1978.
8. An Assessment of the Potential Air Quality Impact of General Aviation
Aircraft Emissions, U.S. Environmental Protection Agency, Research
Triangle Park, North Carolina, June 1977.
3-82
-------
9. The Potential Impact of Aircraft Emissions Upon Air Quality, U.S.
Environmental Protection Agency, Ann Arbor, Michigan, December 1971.
10. Aircraft Emissions, Impact on Air Quality and Feasibility of Control,
U.S. Environmental Protection Agency, Ann Arbor, Michigan, 1972.
11. D.M. Rote, et. al., Argonne National Laboratory, Energy and Environ-
mental Systems Division, Airport Vicinity Air Pollution Study, Report
No. FAA-RD-73-113, December 1973.
12. I.T. Wang, et. al., Argonne National Laboratory, Energy and Environ-
mental Systems Division, Airport Vicinity Air Pollution Study -
Model Application and Validation and Air Quality Impact Analysis at
Washington National Airport, July 1974.
13. Technical Support Report - Aircraft Emissions at Selected Airports
1972 - 1975, Report No. AC 77-01, U.S. Environmental Protection Agency,
Ann Arbor, Michigan, January 1977.
14. Study of Jet Aircraft Emissions and Air Quality in the Vicinity of the
Los Angeles International Airport, Air Pollution Control District,
County of Los Angeles, Contract CPA 22-69-137, April 1971.
15. Howard M. Segal, Boeing Company, Pacific Northwest International
Section - Air Pollution Control Association, Paper No. 73-AP-48,
November 30, 1973.
16. Stern, Arthur C., ed., Air Pollution, Vol. 5, Air Quality Management,
3rd edition, New York, Academic, 1977.
17. Automobile Exhaust Emission Surveillance Analysis of the FY'73 Program,
EPA 460/3-75-007, U.S. Environmental Protection Agency, Ann Arbor,
Michigan, July 1975.
3-83
-------
18. Automobile Exhaust Emission Surveillance Analysis of the FY'74 Program.
EPA 460/3-76-019, U.S. Environmental Protection Agency, Ann Arbor,
Michigan, September 1976.
19. Automobile Exhaust Emission Surveillance Analysis of the FY'75 Program,
EPA 460/3-77-022, U.S. Environmental Protection Agency, Ann Arbor,
Michigan, December 1977.
20. John T. White, III, An Evaluation of Restorative Maintenance of Exhaust
Emissions From In-Use Automobiles, SAE Technical Paper #780082, pre-
sented at the SAE Congress and Exhibition, February 27 - March 3, 1978.
21. Information Documents On Automobile Emissions Inspection and Maintenance
Programs, Final Report, EPA 400/2-78-001, U.S. Environmental Protection
Agency, Ann Arbor, Michigan, February 1978.
22. Mobile Source Emission Factors (For Low-Altitude Areas Only), Final
Report. EPA 400/9-78-006, U.S. Environmental Protection Agency,
Washington, D.C., March 1978.
23. Ambient Temperature and Vehicle Emissions, EPA 460/3-74-028, U.S.
Environmental Protection Agency, Ann Arbor, Michigan, October 1974.
24. CO Hot Spot Preliminary Investigation, TAEB77-13, U.S. Environmental
Protection Agency, Ann Arbor, Michigan, December 1977.
25. Emissions Under Non-FTP Temperature and Speed Conditions, U.S. Environ-
mental Protection Agency, Ann Arbor, Michigan, July 1978.
26. Effects of Low Ambient Temperature on the Exhaust Emissions and Fuel
Economy of 84 Automobiles in Chicago, U.S. Environmental Protection
Agancy, Ann Arbor, Michigan, October 1978.
3-84
-------
27. Automobile Emission Control - The Developmental Status, Trends, and
Outlook as of January 1978. U.S. Environmental Protection Agency,
Ann Arbor, Michigan.
28. Memorandum from Environmental Protection Agency, Office of Mobile
Source Air Pollution Control Program to the Office of Air Quality
Planning and Standards, Regulatory Analysis Review Group (RARG) Review
of Proposed Revision to the National Ambient Air Quality Standard for
Oxidants.
29. Analysis of Technical Issues Relating to: California's Request for
Waiver of Federal Preexemption with Respect to Exhaust Emission Stan-
dards and Test Procedures for 1981 with Subsequent Model Years Light-
Duty Vehicles, Environmental Protection Agency, Ann Arbor, Michigan,
March 1973.
30. Cost Estimations for Emission Control Related Components/Systems and
Cost Methodology Description, EPA 460/3-78-002, U.S. Environmental
Protection Agency, Ann Arbor, Michigan, March 1978.
31. Manufacturability and Costs of Proposed Low-Emissions Automotive
Engine Systems, Consultant Report to the: Committee on Motor Vehicle
Emissions, Commission on Sociotechnical Systems, National Research
Council, September 1974.
32. Revised Evaporative Emission Regulations for the 1978 Model Year,
Environmental and Economic Impact Statement, U.S. Environmental Protec-
tion Agency, Ann Arbor, Michigan.
33. Revised Evaporative Emission Regulations for 1981 and Later Model
Year Gasoline-Fueled Light-Duty Vehicles and Trucks, Environmental
3-85
-------
and Economic Impact Statement, U.S. Environmental Protection Agency,
Ann Arbor, Michigan, August 1978.
34. Draft Environmental and Economic Impact Statement for 1981-1983 High
Altitude Emission Standards, U.S. Environmental Protection Agency,
Ann Arbor, Michigan, September 1978.
35. Draft Report to Congress in Response to Section 206(f)(2) of the
Clean Air Act as Amended in August, 1977, U.S. Environmental Protec-
tion Agency, Ann Arbor, Michigan, October 1978.
36. Revised Gaseous Emission Regulations for 1983 and Later Model Year
Heavy-Duty Engines, Draft Environmental and Economic Impact Statement,
U.S. Environmental Protection Agency, Ann Arbor, Michigan, October 1978.
37. Draft Environmental Impact Statement for Gasoline-Fueled, Heavy-Duty
Vehicles - Notice of Proposed Rulemaking, U.S. Environmental Protection
Agency, Ann Arbor, Michigan, October 1978.
38. Exhaust and Crankcase Regulations for the 1978 and Later Model Year
Motorcycles, Environmental and Economic Impact Statement, U.S. Environ-
mental Protection Agency, Ann Arbor, Michigan, December 1976.
39. Cost-Effectiveness Analysis of the Proposed Revisions in the Exhaust
Emission Standards for New and In-Use Gas Turbine Aircraft Engines
Based on EPA's Independent Estimates, Technical Support Report for
Regulatory Action, U.S. Environmental Protection Agency, Ann Arbor,
Michigan, December 1976.
40. Cost-Effectiveness Analysis of the Proposed Revisions in the Exhaust
Emission Standards for New and In-Use Gas Turbine Aircraft Engines
3-86
-------
Based on EPA's Independent Estimates, U.S. Environmental Protection
Agency, Ann Arbor, Michigan, February 1978.
41. Appendix N - Emission Reduction Achievable Through Inspection and Main-
tenance of Light-Duty Vehicles, Motorcycles, and Light and Heavy-Duty
Trucks. Proposed Rule. Federal Register, 24(84): 22177-22183, Monday,
May 2, 1977.
42. Effects of Inspection and Maintenance Programs on Fuel Economy, U.S.
Environmental Protection Agency, Ann Arbor, Michigan, March 1979.
43. Air Quality Impacts of Transit Improvements, Preferential Lane, and Carpool/
Vanpool Programs, Final Report, U.S. Environmental Protection Agency,
Office of Transportation and Land Use Policy, in Cooperation with U.S.
Department of Transportation, EPA 400/2-78-002a, March 1978.
3-87
-------
-------
4. STATIONARY INTERNAL COMBUSTION SOURCE CONTROL
4.1 PROCESS DESCRIPTION
4.1.1 Engine Design
One of the oldest forms of combustion engines is the gas turbine which
pre-dates, by far, the reciprocating piston engine. The main components of
the gas turbine consist of a compressor, a turbine, and a combustion chamber.
In operation, air is drawn into the compressor, compressed, and then passed,
in part, through the combustion chamber. The high temperature gases leaving
the combustion chamber mix with the main body of air flowing around the
combustor. This hot gas, with greatly increased volume, is led to a nozzle
ring where the pressure is decreased and the velocity is increased. The high
velocity gas is directed against the turbine wheel and the kinetic energy of
the gas is utilized in turning the drive shaft, which also drives the com-
pressor.1 The gas turbine can be operated at much higher speeds than other
engines because of the absence of reciprocating parts. This continuous flow
system, as contrasted to the intermittent flow of the piston engine, produces
a high specific power output from a small machine. The sizes of gas turbines
can range from about 150 to 60,000 kilowatts (200 to 80,000 horsepower) all
operating at high speeds.
4-1
-------
Reciprocating (piston) engines produce power by combustion of a fuel/
air mixture confined in a small space between the head of a piston and the
surrounding cylinder. Expansion of the high pressure combustion gases
pushes the piston producing a linear force which is converted to rotary
torque by a crank shaft. Fuel/air mixtures are ignited in reciprocating
engines by either compression ignition (CI) or by spark ignition (SI). Com-
pression ignition engines usually burn diesel fuel or dual fuel (diesel fuel
plus natural gas). Ignition occurs spontaneously when the fuel is injected
into the cylinder containing compression-heated air or an air/gas mixture.
Spark ignition engines usually burn gasoline, liquid petroleum gas (LPG),
or natural gas, and combustion is initiated by the spark of an electrical
discharge in the combustion chamber. Reciprocating engines are character-
ized by their: (1) cylinder arrangement and number of cylinders, (2) dis-
placement, (3) method of ignition, (4) fuel type, (5) number of piston
strokes per power cycle, (6) compression ratio, (7) rated speed and output
(8) method of cooling, (9) method of aspiration, and (10) fuel metering
method.
Air can be introduced either by natural aspiration or under pressure.
In natural aspiration, air is forced into the cylinder by the vacuum created
by the moving piston. The pressurized method of air introduction is called
supercharging or turbocharging. In the type of supercharging called turbo-
charging, an exhaust gas-driven turbine powers a compressor which boosts the
pressure of the inlet charge. This allows more fuel to be processed through
the engine in a given amount of time, and since the combustion is usually not
impaired, more power results. Since air temperature increases with an increase
4-2
-------
in pressure, the air charge is often cooled to offset charge density losses
from heating during compression and/or to prevent premature autoignition
(called intercooling). Although the Roots-type blowers, typically used on
2-stroke cycle blower scavenged engines, supply air at higher pressure than
atmospheric, the main reason for their use is for exhaust gas scavenging.
Higher cylinder inlet charge densities, therefore, can be obtained with
other types of supercharging such as turbocharging or turbocharging in
series with Roots-type blowers.
Spark-ignition engines are usually of the open chamber design although
some spark-ignition engines may be of the divided chamber or pre-combustion
chamber type (e.g., the Honda CVCC). Carburetion or port injection are
typically used in spark ignition engines although direct fuel injection may
also be used (e.g., the Ford PROCO and Texaco TCCP stratified charge combus-
tion systems). For compression-ignition engines, direct fuel injection is
commonly used with open-chamber engines and indirect fuel injection (injec-
tion into the secondary chamber) is commonly used with divided chamber engines,
Examples of divided chamber engines are the pre-chamber, swirl-chamber and
energy cell or La Nova chamber engines.
4.1.2 Engine Applications
Stationary gas turbine and reciprocating internal combustion engines
are widely used by the oil and gas industry for production and pipeline
applications, in electric power generation, and in industrial and
agricultural applications. Gas turbine engines are more commonly used in
electric utility power plants and as a standby source of electric power
generation and in pipeline transport systems.
4-3
-------
The applications of spark ignition engines depend on engine size (horse-
power) and fuel type. Small gasoline engines in the range of 1 to 8 kw (1 to
10 hp) are used for domestic, agricultural, and commercial power tools and
equipment (power saws, lawn mowers, and portable compressors, pumps, and
electric generators). Medium-size gasoline engines in the range of 40 to
150 kw (50 to 200 hp) are found in commercial and construction site compres-
sors, pumps, blowers, lift trucks, and electric power generator units. Medium-
large spark-ignition engines in the range of 150 to 750 kw (200 to 1000 hp)
are usually fueled by natural gas. Most are of the naturally-aspirated
type. They are used for heavy-duty, medium-speed applications such as gas
compressors or standby power generators. Large spark-ignition engines of
750 kw and up (1000 hp and up) are always operated on gaseous fuels and
are both 4- and 2-stroke cycle, low-speed (300 to 400 rpm) engines. They are
used for compressor drives, gas recompression (in transmission lines), gas
plant compressors, refinery process compressors, water pumping, sewage
pumping, and electric power generator drives for continuous operation. The
total number of gasoline and natural gas-fueled spark ignition engines in use
is much larger than the number of diesel and dual fuel (compression ignition)
engines.2
Diesel engines are widely used in electric power generation, oil and gas
production and transport, and in operation of small electric power and pump-
ing stations. Electric utilities employ diesel engines as prime movers of
continuous and peaking-power generators and in standby power installations.
The transmission line and process compressors used in the petroleum industry
are usually powered by diesel engines. They are frequently used to drive oil
4-4
-------
and gas well drilling and pumping equipment, water pumps, and electric
generators. Municipalities and commercial firms use diesel engines to supply
part of their electric power needs and to power total energy systems and
water and sewage pumping units.
Large low-speed diesel engines above 750 kw (1000 hpj are designed for
continuous operation. Medium, 75 to 750 kw (100 to 1000 hp), and small,
below 75 kw (100 hp), stationary diesel engines are usually derivatives of
engines developed for motor vehicle use.2 They are used mairrly for general
industrial and agricultural applications.
Table 4-1 summarizes the applications of stationary reciprocating engines
by fuel category. It shows the average rated power of engines in each fuel use
category and gives the estimated energy production in kwhr/yr and shows
that natural gas-fueled engines account for 70 percent, diesel and dual-
fuel engines account for 20 percent, and gasoline engines produce 10 percent,
of total reciprocating 1C engine stationary energy production. The energy
production estimates in Table 4-1 are based on average power, load factors,
operating hours (duty cycles), and engine population data for engines in each
•3
category.
4.2 EMISSION SOURCES
CO is emitted in internal combustion engine exhaust due to incomplete
combustion. CO formed in the combustion process is converted to C0? by
combustion with oxygen at temperatures above 625°K (1160°F,. But conversion
of CO to C02 is inhibited if there is insufficient oxygen present during or
after combustion (fuel-rich combustion zones), or if the combustion products
cool to temperatures below 625°K (1160°F) before CO oxidation is complete.
4-5
-------
CJ
O
z
CO
LU
z
— 05
C3 >-
Z OC
LU O
O \O O LA O
O oo \D r*- i""*
CO O — -3- —
CSI CM \O —
O LU
C3 O
to
-3- O
O -J
I I I lV ILJ
_J Q_ ZD
CD — U.
^ O
I— LU >-
al OQ
5- -Z.
— O
oo LA — mo
LA LA \A CO
i~-\ — — J- OO
o o o o o
O LA O LA LA
O «"^ LA P^ r^-
o o o o o
O sO O VC vO
LA CM CA LA LA
tf\ ff\ O O •—
O LA O O
en LA
— en
— j-
§ ^
CO
Z
o
<
o
Q_
Q_
— 3 —
3 U II
O *J J
— Q. 41 —
re 3 in in
._ in 3 L.
i- O
in *-> w. 41 in
w i/i 41 c in
41 3 *-> — 4>
in -Q re L. L.
41 41 3 — c
._ . 41 41
c c •"-< *J re
— o LU
L. *J in —
— C 3
3 Q) O CD Oi
L. .— re — •— o
41 -o
C C
41 re
o>
•D
c — in
4) •—'41
— C 3 in
— re o "O L.
u 4) re 41 m
— c L. i c
4-6
-------
CO emission rates from gas turbines are extremely low above 50% of
rated power. CO emission rates from reciprocating internal combustion
engines are quite variable. The rates depend on both engine design and how
the engine is operated. Important design factors include the number of
strokes per power cycle, combustion chamber design, the methods of air
charging (aspirated, turbocharged, blower-scavenged), and the method of fuel
charging (direct and indirect injection and carburetion). Significant
operating variables include fuel type, ignition, air/fuel ratio, engine
speed and load, and maintenance practices.
The following sections discuss the effects of engine design and operating
variables on CO emissions from gas turbines, spark ignition, and compression
ignition engines. Mass emission rates are given for specific engine designs,
sizes, and fuels at rated and reduced load and speed. Then average emission
factors are presented, and these are used to estimate total nationwide emis-
sions of CO from stationary engines.
4.2.1 Gas Turbine Engines
CO emissions from gas turbine engines, used in electric utility service,
expressed in terms of energy, are shown to be very low when the gas turbine
is operated under load, as shown in Figure 4-1. It has been postulated that
the average load factor for gas turbine engines during operation is about
86.8 percent based on 1196 hours of operation per year, or about 4.8 hours
per operating day. It is further assumed that time spent at off-design
conditions includes 15 percent at zero load, and 2 percent each at 25 percent,
50 percent and 75 percent load. Then the percentages of operating time at
4-7
-------
CO
C
O
•H
W
w
•H
I
O
u
Source:
50 75
% $ RATED LOAD
Reference 4
100
125
FIGURE 4-1.
SPECIFIC EMISSIONS OF CO AS A FUNCTION OF LOAD
FOR GAS TURBINE-POWERED GENERATORS, COMPOSITE
OF SEVERAL MAKES AND MODELS
4-8
-------
rated load (100 percent) and peak load (assumed to be 125 percent of rated
load) can be calculated to produce an 86.6 percent load factor. These
percentages turn out to be 19 percent at peak and 60 percent at rated load.
CO emission factors developed for electric utility gas turbines are presented
in Section 4.3.
4.2.2 Spark Ignition Engines
Spark ignition engines burn gasoline or natural gas, and CO emissions
from gasoline engines are an order of magnitude higher than those from gas
engines. The air/fuel (A/F) ratio of the combustible mixture is the most
important variable. Figure 4-2 shows the effect of the air/fuel ratio on
N0x, HC, and CO emissions from gasoline engines.3 It shows that when the
air/fuel ratio is adjusted to produce low engine out CO emissions, the NO
A
emissions produced by the engine can range from relatively high to relatively
low values.
Since gaseous fuels typically allow stable combustion at leaner air/fuel
ratios, the CO emissions from gaseous fueled spark ignition engines are
considerably lower than they are from gasoline-fueled spark ignition engines.
Table 4-2 summarizes data on CO emissions from heavy duty, 4-stroke,
naturally aspirated gasoline engines and medium and large gas engines of
different designs.2 It shows the effects of engine design, fuel type, and
air/fuel ratio on emission rates at rated loads. Emissions are given for
continuous duty (steady state) operating conditions and as composite (modal)
values. The composite values are the result of standard test cycles at
specified load/speed modes of operation.
4-9
-------
CO
I-J
w
S3
O
h-l
CO
CO
W
i-J
W
10
12
14 16 18
AIR-TO-FUEL RATIO
20
22
Source: Reference 3
FIGURE 4-2. EFFECT OF AIR/FUEL RATIO ON EMISSIONS FROM
A GASOLINE ENGINE
4-10
-------
Q
<
O
<
o:
<
CO
CM
I
O
o:
a.
O
I-
z
CJ
a:
O_
CO
O
a:
u_
oo
•z.
O
I/)
CO
I
QJ
O
•T-S
Jt C
o ~
*•• >.
o —
*-• >»
v -o . en
ra .c
I- O
3 0
•u -O
1C 3
Z t-
II II
4-11
-------
While detailed emission data for smaller gasoline engines are not
included in Table 4-2, average emission factors are presented later in this
section.
4.2.3 Compression Ignition Engines
CO is formed by the same mechanisms in compression ignition engines as
in spark ignition engines, but in compression ignition engines fuel is
injected independently of air so fuel/air mixtures are more heterogenous.
Fuel distribution can be controlled by injector design, and thus wall
quenching effects can be minimized. Compression ignition engines are usually
unthrottled and are designed to operate fuel lean (high excess air) so CO
emissions are relatively low.
CO emissions from compression ignition engines are more clearly
dependent on engine design and variations in emission rates are quite large.
The lowest CO emissions are produced by large, low speed engines, and smaller
engines usually have higher emission rates. Divided chamber turbocharged
diesel engines produce the lowest emissions. Table 4-3 summarizes data on
CO emissions from compression ignition diesel engines of different designs
and sizes at rated conditions.4 Average emissions vary from 0.3-14.6 g/kwhr
(0.2 to 10.9 g/hphr) depending on engine design.2
Emissions from compression ignition engines are dependent on engine load
and speed. Figure 4-3 shows normalized data variations in CO emissions against
engine load at rated speed.2 The data are expressed in terms of CO/CO as a
percent of rated power (mass emissions at reduced power output divided by
4-12
-------
CO
<
CD
z
o
o
O
LU
CO
Z Z
o o
Z Q
CD Z
— O
O
Z
O Q
— UJ
CO h-
co <
LU o:
a:
a. h-
2: <
o
O CO
o —
oc CD
CO
z
o
CO
CO
— m
O O
r^ so
— -T
O
CJ
a o
a> —
en i 4-1
< 1-
Z 0)
J3
- E
0 U
4-> C
in 0)
I Q.
-3- O
(_> u
1— 0)
.a
<0
-T T!
.a
TJ
j:
u
(U
O D
i- -a
4J • —
Ifl >
1 —
-3- T3
to
CQ
O
i/i
CNI
4-13
-------
o
o
o
M
H
55
O
M
CO
CO
M
§
CO
CO
CJ
M
Fn
M
CJ
W
PM
CO
O
O
3.5
3.0 '
2.5
2.0 -
1.5 -
1.0 -
0.5 H
O Four-stroke, naturally aspirated, open chamber
QFour-stroke, turbocharged, open chamber
A Four-stroke, naturally aspirated, divided chamber
O Four-stroke, turbocharged, divided chamber
QTwo-stroke, blower scavenged
BTwo-stroke, turbocharged
0 30
I
40
50
I
60
i
70
80
90
100
PERCENTAGE OF RATED POWER
SOURCE: Reference 2
FIGURE 4-3. DIESEL ENGINE PART-LOAD CARBON MONOXIDE EMISSIONS
4-14
-------
emissions at full load) for six engine designs. In general, CO emissions
decrease as load is reduced, but they tend to increase as the load is reduced
to less than about 60 percent of rated power. When engine speed is reduced
as well as load, CO emission rates can be reduced by as much as 50 percent.3
4.3 EMISSION FACTORS AND NATIONWIDE CO EMISSIONS
4.3.1 Gas Turbine Engines
Emission factors developed for electric utility gas turbines are pre-
sented in such a form as to yield mass emissions in pounds of mass per unit
time.4 CO emission factors are assumed to be uniform for the different types
of turbines because of the limited amount of information that is available.
Factors for CO are found in Table 4-4.
TABLE 4-4
COMPOSITE CO EMISSION FACTORS FOR THE 1971
POPULATION OF ELECTRIC UTILITY GAS TURBINES ON A FUEL BASIS
GRAMS PER CUBIC METRE GRAMS PER LITRE
(lb/106 ft3) gas (lb/103 gal) oil
Composite 1.84 (115) 1.85 (15.4)
E.F.
Source: Reference 4
4-15
-------
Other useful emission factors for electric utility gas turbines are shown
in Table 4-5. These factors can be used to estimate nationwide CO emissions
by multiplying the composite emission factor and the total rated capacity (MW)
of all U.S. gas turbines and assuming both gas and oil-fueled turbines operate
75 percent of the time. On a national basis, electric utility turbine sources
account for less than 1/2 of 1% of the CO contribution from all sources.
Although CO emissions from electric utility turbines are not a large part of
the national or even regional impact, this source of CO can be a major source
in urban or heavily populated areas and therefore may require CO control
measures.
TABLE k-5
COMPOSITE CO EMISSION FACTORS FOR THE 1971
POPULATION OF ELECTRIC UTILITY GAS TURBINES
Electrical
Output
% Rated
Power
0*
25
50
75
100
125
CO
Kg/Hr per
Emi ss
ions
MW Rated Capacity
(ib/hr per
3.
1.
0.
0.
0.
0.
9
5
k
k
5
5
MW
(8
(3
(0
(0
(1
(1
Rated Capacity)
.6)
.2)
.8)
.9)
.0)
.0)
Wei
Wei ght ing
Factor
0
0
0
0
0
0
.15
.02
.02
.02
.60
.19
(Ib/hr
0.
0.
0.
0.
0.
ghted
Kg/Hr
per
59
03
01
01
27
09
MW
(1
(0
(0
(0
(0
(0
CO Emissions
per MW
Rated
.29)
.06)
.02)
.02)
.60)
.19)
Capaci ty )
.00
1.00 (2.18)
Composite E.F.
Source: Reference
"Spinning reserve
4-16
-------
4.3.2 Reciprocating Internal Combustion Engines
CO emissions from reciprocating engines can vary from less than one to
hundreds of g/kwhr depending on engine design, operating conditions, and
fuel. Engine population data are available by fuel and rated power, but
not by engine design. There are also wide variations in CO emission rates
among engines in the same fuel-size categories. All of these factors make
it very difficult to define accurate emission factors for reciprocating
internal combustion engines. Table 4-6 summarizes "brake-specific" fac-
tors.2»3»5 The emission factors are based on engine application, fuel, and
rated power. Annual emissions can be calculated from the product of the
emission factor, the number of hours per year of operation, the rated power,
and the load factor (output produced divided by output available).
Selected emission factors combined with the data in Table 4-1 were
used to estimate nationwide annual CO emissions from stationary reciprocat-
ing internal combustion engines. Table 4-7 summarizes the estimate and
shows that reciprocating internal combustion engine emissions are 3.6 mil-
lion metric tons/year (4 million tons/year).
4.4 CONTROL TECHNIQUES
CO control techniques for stationary internal combustion engines are
in the developmental stages. There are few techniques currently in routine
use to control CO emissions. CO control technology has been developed for
mobile applications in response to California and Federal limits on vehicular
emissions. The techniques are now being considered for stationary engines.
Differences in duty cycles, engine size and weight, and fuels for stationary
engines mean that testing is required to demonstrate how the techniques can
4-17
-------
Z
o
CO
ID
CD
^.
O
O
0) *J
3 O
TO
> -
C
0) O
ro ij
0) —
cn a
en c
< a
CD
Z
O
O
a:
-J a.
CQ —
LU
a:
o
u.
CO
OL
O
h-
o
u_
Z
o
co
CO
o
o
) — — — .- — ._
TO tj I
o TO «-
•— L. «-
—' W O
TO (D 3 (0 u 3
i- ai cr k. u i.
^ X 0> •*-• If 4-J
3 ^- 3 -^
4) &>
4-18
-------
TABLE 4-7
ESTIMATED 1975 NATIONWIDE CO EMISSIONS FROM INSTALLED
RECIPROCATING 1C ENGINES
RANGE OF RATED POWER
FUEL
Diesel
Subtotal
Natural Gas
Subtotal
Dual Fuel
Subtotal
Gasoline
Subtotal
TOTAL
Percent of all sources
(kW)
15-75
76-370
>370
<370
>370
All
>75
(hp)
20-100
01-500
>500
<500
>500
Al
5-99
>100
EMISSIONS
O3 Metric tons/yr
(103 tons/yr)
30.7 (33.8)
47.6 (52.5)
16.9 (18.6)
95.2 (104.9)
113.0 (125.0)
242.0 (267.0
355.0 (392.0)
21.1 (23.3)
21.1 (23.3)
1940.0 (2140.0)
856.0 (994.0)
328.0 (362.0)
3124.0 (3446.0)
3595.3 (3966.2)
3.4
Source: Reference 3
4-19
-------
be successfully transferred. Stationary engines do not have the space and
weight limitations of mobile engines. Since they usually operate at steady-
state conditions, it is easier to optimize some operating and design param-
eters for emission control.
Selection of control techniques for internal combustion engines is a
very complex problem. Engine size and design, fuel, and duty cycle as well
as the desired level of reduction must be considered. The effects of con-
trol methods on fuel consumption, engine maintenance requirements, durability
of engine components, performance, and emissions of other pollutants such as
NOX, hydrocarbons, and fine particulates are also important.
In general, there are four ways to reduce CO emissions from stationary
reciprocating internal combustion engines: exhaust gas treatment to oxidize
CO to C02; adjustments to the fuel/air mixture controls; replacement of the
engine with alternative engines; and use of alternative fuels. This general
discussion of control methods complements the more detailed presentation in
Chapter 3.
4.4.1 Oxidation of CO in the Exhaust Gas
Exhaust manifold air injection, thermal reactors, and catalytic converters
all control CO emissions by oxidizing CO in the exhaust to C02. The gas
temperature, oxygen concentration, catalyst parameters and CO concentration
are the important operating variables. Secondary air injection and temperature
control are often required. Two kinds of thermal reactors have been developed
for automotive (gasoline spark ignition) engines: the Rich Thermal Reactor (RTR)
for fuel rich air/fuel ratios and the Lean Thermal Reactor (LTR) for lean ratios.
The thermal reactor is a container which, by its size and configuration,
4-20
-------
increases the residence time and turbulence of exhaust gases, thereby provid-
ing a chamber for the high-temperature oxidation reaction. High temperatures
are maintained by the exothermic oxidation of CO and HC in the exhaust gas?
The rich thermal reactor operates at temperature from 870 to 1040°C (1600 to
1900°F) and is designed for fuel rich operation. At rich air/fuel ratios of
11-12 to 1, N0x emissions are reduced to less than 6 g/kwhr (4.5 g/hphr),
but fuel consumption penalties are incurred, Secondary air injection is
normally injected into the thermal reactor for complete oxidation, and con-
struction materials such as Inconel 601 are needed for the inner core, baffles
and port liners. Temperature control devices are required to protect the
reactor construction materials against overtemperature.
The lean thermal reactor operates at higher air/fuel ratios (17-19 to 1)
and lower operating temperatures, 760-870°C (1400 to 1600°F), than the rich
thermal reactor. Secondary air-injection is not usually required and con-
struction materials have less severe durability requirements than do the
materials for rich thermal reactors. Oxidation catalysts and 3-way catalysts
are being used extensively in the control of CO from automotive engines.
This CO control strategy can be equally effective in the control of CO from
stationary engine sources. Recent literature describes a patented platinum
catalyst on a ceramic honey comb support that has withstood 50,000 hours of
stationary engine testing. The catalytic converter has also been used for
small Diesel, LP gas, and gasoline engines in sizes up to 13.1 litres (800
cubic inches) displacement and is applicable to 2- and 4-cycle naturally
aspirated or turbocharged engines. Applications include Diesel powered
mining and tunneling equipment, locomotives, loaders, forklift trucks
operated in enclosed spaces, and electric generators located near
4-21
-------
air-conditioning intakes. For oxidation catalysts to be an effective means
of controlling CO and HC emissions, the engine must be properly tuned and
unleaded fuel must be used. Also, the control system should ideally be
adjusted to preclude the formation of sulfate emissions which can be formed
in the catalyst due to excess oxynen in the exhaust gases and sulfur content
of the fuel. Alternatively, sulfur can be removed from the fuel. In the
case of 3-way catalysts, rich mixtures are conducive to the formation of
HCN and ammonia.
Air injection into the exhaust manifold can reduce CO emissions by a
factor of 55 percent from baseline emissions on some engines with modifica-
tions to the air/fuel ratio, compression ratio, and spark ignition timing
schedule.2
4.4.2 Design Changes and Operating Adjustments
The air/fuel ratio is the operating variable that determines CO emis-
sions, and it has a significant effect on N0x emissions. Operation at
air/fuel ratios that produce low CO emissions can produce high or low NO
emissions depending on the exact value of the air/fuel ratio used. Since NO
emissions from stationary reciprocating internal combustion engines are con-
sidered more of a problem than CO emissions, design and operating changes
are expected to be made in these sources primarily for NO control. Care must
X
be taken to ensure that the entire emission control system provides adequate
control of all emissions that need to-be controlled. This sometimes leads to
more sophisticated systems. Derating, turbocharging, and improved fuel
injection nozzles can be used to control CO emissions from compression
ignition engines. The addition of a turbocharger is normally used to
increase specific power output but it also can increase the air/fuel ratio
4-22
X
-------
in the power modes of operation. This usually improves specific fuel consump-
tion but also causes an increase in NO emissions. Retarded injection timing
X
(diesel) and/or intercooling the boosted inlet air charge can be used to
offset the N0v penalty. Improved diesel fuel injectors (e.g., low sac
X
nozzles) can be used to reduce CO and HC emissions but, again, NOV emissions
X
may increase due to more efficient and higher temperature combustion.
Measures that could be used to increase the air/fuel ratio for gasoline
spark ignition engines include charge homogenation and air/fuel stratifica-
tion. Both approaches are under consideration and may provide some potential
for lowered CO emissions. In general, adjustments to increase the air/fuel
ratio for gasoline engines will require design changes to insure a uniform
air/fuel mixture in each cylinder and to achieve stable engine operation.
4.5 ECONOMIC, ENVIRONMENTAL, AND ENERGY IMPACTS OF CONTROL TECHNIQUES
The only existing regulations for CO emissions from internal combustion
sources are the California and Federal standards for automotive engines (see
Chapter 3). As a result of these standards, most CO control technology has
been developed for automotive engines. Suggested standards of performance
for new stationary engines do not now require CO control so there is little
incentive for developing stationary engine CO controls.6 Catalytic (oxida-
tion) converters are currently marketed for small engines, mostly on wheeled
equipment used in enclosed spaces. This is the only example of a CO control
technique currently available for application to a "stationary" engine. Some
testing has been done, however, to determine the applicability of automotive
engine controls to stationary engines.2
4-23
-------
Since CO control methods for stationary engine sources are still in
the developing stages, there is no quantitative information on cost CO reduc-
tion efficiencies for controls applied to classes of stationary engines, or
environmental and energy impact. The status of development for different
control techniques and qualitative information on environmental and energy
impacts for new mobile sources are summarized in Chapter 3. Many of these
CO control techniques are also applicable to the stationary source powerplants.
Internal combustion engines also produce significant emissions of
nitrogen oxides, hydrocarbons, odorous organic compounds, and fine particu-
lates (smoke). Table 4-8 shows that internal combustion engines contribute
quite significantly to the nationwide emissions of N0x, CO, and hydrocarbons.3
The N0x and reactive hydrocarbon emissions due to the application of CO con-
trol techniques are important because these pollutants participate in oxidant-
forming reactions. In developing standards of performance for new internal
combustion engines, more emphasis has been placed on controlling NO emis-
sions than CO or hydrocarbons. This is relevant because control techniques
usually influence the emissions of NO and hydrocarbons as well as CO.
TABLE 4-8
PERCENT OF TOTAL 1975 NATIONWIDE EMISSIONS OF NO , CO,
AND HYDROCARBONS FOR STATIONARY INTERNAL COMBUSTION ENGINES
NOX CO TOTAL HYDROCARBONS
Percent of all sources 8.4 3.4 3.8
Source: Reference 3
4-24
-------
There are some changes in engine design or operating variables that
result in lower CO emissions, but in some cases those reductions are achieved
at conditions which produce increased NO emissions. Since controlling NO
X X
emissions from internal combustion engines has a higher priority than con-
trolling CO emissions, there are probably few situations in which CO emis-
sions would be controlled at the expense of increasing NO emissions.
X
Consequently, aftertreatment devices such as catalytic systems would appear
to be one of the control approaches that would be considered if both NO
X
and CO are to be controlled to the lowest levels.
4-25
-------
REFERENCES FOR CHAPTER 4
1. Obert, Edward F., Internal Combustion Engines, Third Edition,
December, 1968.
2. Assessment of the Applicability of Automotive Emission Control
Technology to Stationary Engines, EPA-650/2-74-051, U.S. Environmental
Protection Agency, Washington, D.C., 1974.
3. Standard Support Document and Environmental Impact Statement - Stationary
Reciprocating Internal Combustion Engines. Aerotherm Project 7152.
Contract 68-02-1318, U.S. Environmental Protection Agency, Research
Triangle Park, North Carolina, 1976.
4. Exhaust Emissions From Uncontrolled Vehicles and Related Equipment Using
Internal Combustion Engines, Part 6 Gas Turbine Electric Utility Power
Plants, AR 940, U.S. Environmental Protection Agency, Ann Arbor,
Michigan, February, 1974.
5. Compilation of Air Pollutant Emission Factors Including Supplements
Through No. 8, U.S. Environmental Protection Agency, Research Triangle
Park, North Carolina, February, 1977.
6. Engelhard Industries, PPC Purifier to Control Engines' Exhaust Fumes,
EM-A-168 Rev. 4/78; PTX Purifier to Control Diesel Engine Exhaust Fumes,
EM-A-131 Rev. 4/78; PTX Diesel Catalytic Purifiers Cut Harmful Air
Pollution, EM 11459.
4-26
-------
5. STATIONARY EXTERNAL COMBUSTION SOURCE CONTROL
This chapter describes carbon monoxide emissions and controls from sig-
nificant stationary combustion sources. Combustion sources discussed include
utility and industrial boilers, residential and commercial heaters, and
solid waste incinerators. Process descriptions are given in enough detail
to indicate where emissions are produced, and emission quantities are esti-
mated for each source. Currently applied control technology and feasible
control methods are discussed, as are control efficiencies, energy require-
ments, costs, and environmental impacts.
5.1 UTILITY AND LARGE INDUSTRIAL BOILERS
This category includes the majority of the utility and industrial elec-
tric power generating boilers. The thermal input of boilers in this category
ranges from 30 MW (100 x 106 Btu/hr) up to 3500 MW (120 x 108 Btu/hr).
5.1.1 Process Description
Utility and large industrial boilers may be fueled with coal, oil, or
gas. The principle distinction between these boilers is the type of fuel
fired and the firing mode, although such factors as furnace volume, operating
pressure, and configuration of internal heat transfer surface differ as well.
Firing mode includes the type of firing equipment, the fuel handling equip-
ment, and the placement of the burners on the furnace walls. The major types
of firing modes are:
5-1
-------
1. single- or opposed-wall fired,
2. tangentially fired,
3. turbo fired,
4. cyclone fired, and
5. stoker fired.
Each of the major firing modes except stoker fired can be used in boil-
ers burning gas, oil, or pulverized coal. However, the cyclone mode is
usually designed to fire coal as the principal fuel.
In single- or opposed-wall fired furnaces, the burners are mounted
horizontally on the walls of the combustion chamber. These units can burn
gas, oil, pulverized coal, or a combination of these fuels. Opposed-wall
firing is used in larger units and capacities generally exceed 1200 MW
(4 x 109 Btu/hr) heat input.1 Turbo fired units are similar to horizontally
opposed units except that the burners are set at an angle in the vertical
plane. The intermixing of the opposing streams produce highly turbulent
conditions and virtually complete combustion takes place below the furnace
throat.2
Tangential fired units have a furnace characterized by a square cross-
sectional shape with burners mounted in two or more corners. The burners
are fired tangentially to a small, imaginary circle in the center of the
furnace so that the flames exhibit a rotating or spinning motion.1*2
In cyclone fired units, fuel and air are introduced circumferentially
into a water-cooled cylindrical combustion chamber. Cyclone burners were
originally designed to burn crushed, low ash-fusion temperature coals. How-
ever, because of difficulties in obtaining suitable low sulfur coals and the
inability of this design to adapt to low NOX operation, cyclone furnaces are
5-2
-------
no longer being constructed. Many existing cyclone units have been convert.
to burn fuels other than coal.2*3
Vertical fired furnaces were developed for pulverized coal burning
prior to the advent of water-walled combustion chambers. These units pro-
vide long residence times and burn low volatile content coals such as anthr
cite. Vertical fired units are no longer sold and relatively few of these
units are found in the field.3
Stoker-fired boilers are designed to burn solid fuels in a bed. This
bed is either a stationary grate through which ash falls, or a moving grate
which dumps the ash into a hopper. The two most common types of stoker r' -
signs are underfeed (single and multiple retort) and overfeed (spreader)
stokers. In the underfeed designs both fuel and air move in the same rela-
tive direction. Rams force the new fuel into the furnace from beneath the
fuel bed as ash is pushed aside and collected.1'3 Spreader stokers are over-
feed designs which distribute the fuel by projecting it evenly over the fuel
bed. A portion of the coal burns in suspension. The upper limit of spreader
stoker size is about 180 MW (600 x 106 Btu/hr) heat input.1* All larger si/^d
units are pulverized coal or cyclone designs. Either pulverized coal or
spreader stoker-type units are used in the size range of 30-180 MW (100-600
x 106 Btu/hr) heat input depending upon local economics and customer pref-
erence.5
5.1.2 Process Emission Sources and Factors
The formation of carbon monoxide in boilers and subsequent emission in
the flue gas results primarily from the partial oxidation of the fuel. In
some cases, however, high temperature dissociation may contribute to the emis-
sions of Cu in boiler flue gas, particularly if the unit is being operated
5-3
-------
above design load. Improperly operated stoker boilers may also emit ex-
cessive amounts of CO by the reduction of COz. This occurs when the fuel
bed is allowed to build too deep, creating a reduction zone where the CO is
formed.
Estimates of 1977 fuel consumption were obtained from an inventory of
combustion-related emissions from stationary sources, published by the EPA.1
These estimates were based in part on data contained in the National Emis-
sions Data Service file and agree with data obtained from other references.3
CO emissions were estimated from the fuel data using AP-42 emission factors
as listed in Table 5-1.6 EPA estimates of 1977 CO emissions for utility
boilers and large industrial boilers are given in Table 5-2. The total es-
timated emissions from both utility and large industrial boilers are 405,600
metric tons/yr (447,200 tons/yr).7 These sources contribute slightly more
than 2 percent of the total estimated CO Emissions from all stationary
sources.
5.1.3 Control Techniques
Control strategies for reducing CO emissions from utility and large in-
dustrial boilers can be divided into two groups:
1) Control strategies which reduce CO concentrations in boiler
flue gas, and
2) Control strategies which reduce CO emissions by decreasing
boiler fuel consumption through increased unit efficiency.
It should be noted, however, that CO emissions from well-operated units are
usually quite low (less than 50 ppm) so that implementation of further con-
trols in many cases offers very little potential for further reduction.8
The following is a summary of the various control techniques.
5-4
-------
CO
LU
_J
O
CD
i
—
J-
CO
~~^
Q
Z
LU
C3
o:
«^
1
Q
•z.
<
x»
1—
— •
_J
—
1—
ID
o:
o
LO-
CO
Cd
O
h—
O
v^
O
o
-Q
f— .
r—
N— '
C
O
4-1
o
._
u.
4_)
CL)
E
"v^
0
O
cn
LA
•
o
TO 4->
— CL
L— d)
4-1 O
in X
U <\)
~o
c —
._ r—
TO
(U ^~-
cn
TO TO
— o
U
TJ
C 1
TO -—«
I/) (/)
>, L. l_
4-J 0) 0)
• — c— _i£
— — o
— O -M
4-1 JD m
^
vO
cr\
, — ^
c
O
x^
0
o
-Q
i—
CM
N — '
C
O
4-J
O
._
i_
4-1
~ U.
4-J 0)
• — r—
F— .—
— O
4-J _Q
^
l^v. v£>
CM CO
^ —
TO TO
cn cn
ro ro
0 0
x^ x^
0 0
o o
jQ -Q
F— r—
Lf\ LA
"— ' *~-s
in in
^_
en cn
.*: ^:
ro co
vO vO
o o
i
in
i_
(1)
>— i—
•— .—
O O
-Q
1
t/1 TO
1_ •—
CL) 1-
• — 4_)
— i/l
O D
-D "D
C
>~ •—
4-J
• —
O
\
o
o
-Q
p^.
,_
%_^
on
E
•Z.
10
O
\
cn
•^
CM
CM
TO
— TJ
I- C
4-J TO
l/l
3 l/)
TJ TO
c cn
._
^_
CD TO
cn L.
TO -M
— TO
C l/>
T3 TO
c i cn
TO
in i/>
>» J- in
4-J 0) (L)
— r— 0
r— .- O
— o ^~
•M -Q QL
ID
vO
CL)
O
C
s_
M-
d>
, .
CL)
0
L-
^
0
CO
5-5
-------
TABLE 5-2. SUMMARY OF 1977 NATIONWIDE CARBON MONOXIDE
EMISSIONS FROM UTILITY AND LARGE INDUSTRIAL BOILERS
CO Emissions
Fuel Type
103 metric tons/yr
103 tons/yr
Anthracite Coal
Electri c Uti1i ties
Industrial Boilers
Bituminous Coal and Lignite
Electric Utili ties
Industrial Boilers
Residual Oil
Electric Uti1i ties
Industrial Boilers
Disti1 late Oi1
Electric Uti1i ties
Industrial Boilers
Natural Gas
Electric Uti M ties
Industrial Boilers
TOTAL
0.6
0.5
212.8
26.8
50.7
2k.Q
5.6
9.5
22.7
52.4
405.6
0.7
0.5
234.6
29.5
55.9
26.5
6.2
10.5
25.0
37.8
447.2
Source: Reference 7
5-6
-------
5.1.3.1 Automatic Excess Air Rate Control
In normal boiler operation, it is often necessary to operate at excess
air rates somewhat higher than what is necessary for complete combustion.
This is to provide a "cushion" against minor variations in process conditions
such as fuel heating value, steam pressure, ambient temperatures, etc. With-
out such a cushion, fluctuations in the air/fuel ratio can result in periodic
smoke and/or high CO emissions.8 By employing automatic excess air control,
the boiler can be operated at low excess air rates, resulting in less fuel
consumption and reduced NOX emissions, while still assuring that CO emissions
are held to a minimum.
5.1.3.2 Proper Firing Rate
Components of the combustion system should be chosen to handle any
future increases in load requirements. Firing in excess of design capacity
can result in premature cooling of combustion gases by decreasing the resi-
dence time of these gases within the combustion zone. A similar quenching
effect is observed if the flames are allowed to impinge on any relatively
cold surfaces within the combustion chamber. Cooling of the combustion gases
by these mechanisms can result in increased emissions of smoke and CO.
5.1.3.3 Burner Maintenance
Damaged or clogged burners can result in high CO emissions by disturbing
proper air/fuel distribution. Both proper installation and maintenance of
burners and other combustion equipment is required for clean and efficient
operation and minimum CO emissions.
5.1.3.4 Reduced Fuel Consumption
Devices for improving the thermal efficiency of a boiler system, such as
added insulation, low excess air burners, air preheaters, soot blowers, and
5-7
-------
load management techniques, can be implemented to reduce CO emissions. A
decrease in fuel consumption will usually result in a proportional decrease
in CO emissions.
5.1.4 Cost of Controls
Many of the CO control techniques mentioned above involve operations or
maintenance-related functions, such that capital cost requirements are low
or negligible. In many cases increased maintenance costs due to CO control
efforts are offset by fuel savings through more efficient operation.
Sophisticated combustion control systems, such as the automated excess
air control mentioned above, can be quite expensive to implement. Costs
vary substantially depending on the complexity of the system. However, a
control system which controls excess air rates at a minimum will result in
overall fuel savings, which can help offset high first costs.
5.1.5 Impact of Controls
5.1.5.1 Emission Reduction
Total CO emissions from utility and large industrial boilers are esti-
mated at 405,600 metric tons/yr (447,200 tons/yr).7 The potential for sig-
nificant reduction of these emissions by the applications of additional CO
control techniques is not large. Factors which contribute to this are:
1. CO emissions from most utility and large industrial boilers are
quite low (generally lower than 50 ppm in the flue gas).8
2. Oil and coal-fired units will usually emit smoke or soot
when the amount of excess air is decreased. Conditions
which result in smoke formation are avoided, resulting in
corresponding low CO levels.3
5-8
-------
3. Several of the common NOX control techniques result in
increased CO emissions. In general, a NO control method
/\
is applied until flue gas CO levels reach 200 ppm. Further
application is then curtailed.3 Table 5-3 illustrates the
change in CO emissions which results from application of
NOX control measures to several boilers.
4. CO emissions from coal-fired units are usually higher than
those from oil or gas-fired units.6 Many utilities are
converting their oil and gas units to coal, reflecting
anticipated shortages of these fuels. Hence, CO emissions
can be expected to increase accordingly.
5.1.5.2 Environment
Reducing CO emissions from combustion sources usually involves tech-
niques which improve combustion. Examples of such techniques include
checking oil burners for proper fuel atomization or improved control over
excess air levels. These same techniques are also useful in reducing the
level of combustible particulates.9
Sulfur dioxide emissions are not directly affected by CO control
techniques as most all of the sulfur in the fuel exits with the flue gas.
There is some evidence, however, which suggests that lowering excess air
levels (by using a better combustion control system, for example) can result
in reduced sulfate emissions.3 Total sulfur emissions, though, can be de-
creased proportionately by any efficiency improving technique which results
in lower fuel consumption rates.
5-9
-------
TABLE 5-3. REPRESENTATIVE EFFECTS OF NOX CONTROLS ON
CO EMISSIONS FROM UTILITY BOILERS
CO Emissions (ppm at 3% 02)
NOV Control
Low Excess Ai r
Staged Combustion
Flue Gas Recirculation
Fuel
Natural Gas
Oil
Coal
Natural Gas
Oil
Coal
Natural Gas
Oil
Basel ine
14
86
12
p
0
14
19
85
15
19
42
20
2k
27
27
14
86
12
14
19
85
15
28
2k
27
17
31
175
21
With NOX Control
68
7k
61
}k
k2
53
20
19
93
60
283
81
225
16
67
13
}k
21
85
21
37
20
26
40
45
65
9
Source: Reference 3
5-10
-------
5.1.5.3 Energy Requirements
Generally, approaches to CO control involve maximizing fuel efficiency.
Consequently, implementation of most CO control measures results in a net
fuel savings.
5.2 INDUSTRIAL BOILERS
The industrial boilers discussed in this section differ from the utility
and large industrial boilers described .in Section 5.1, in that the thermal
input of these boilers is smaller [3-30 MM (10-100 x 10s Btu/hr)], the de-
signs are less complicated, and the fuels consumed are more varied. In
general, operation of industrial boilers is less controlled than that of
utility boilers.
5.2.1 Process Description
Industrial boilers with 3-30 MW (10 x 106 - 100 x 106 Btu/hr) capacities
are either field-erected or package units.3 Usually, field-erected units
have larger capacities and are similar in design to the boilers described in
Section 5.1.
Packaged boilers (shipped complete with fuel-burning equipment) are
mainly watertube or firetube designs, although other types such as cast
iron or shell designs are occasionally used in applications where low pres-
sure steam is all that is needed. In watertube boilers, hot gas passes over
water- or steam-filled tubes which line the combustion chamber walls (Fig-
ure 5-1). In firetube boilers, hot gas flows directly through tubes which
are submerged in water (Figure 5-2).
Most packaged boilers with capacities greater than 8.8 MW (30 x 106 Btu/
hr) are watertube boilers.3 Upper pressure limits on firetube boilers range
5-11
-------
FIGURE 5-1. WATERTUBE BOILER
5-12
-------
LU
O
CD
CO
CM
to
LiJ
o:
5-13
-------
from 1.1 - 1.8 megapascals (150-250 psig).5 Small watertube boilers have
been built for operation at up to 4.2 megapascals (600 psig).5
Packaged boilers of both types are primarily single-burner fired, using
either natural gas or fuel oil. About 15 percent of packaged boilers were
reportedly stoker-fired in 1975.3 Boiler firing modes are discussed in
greater detail in Section 5.1.1.
5.2.2 Process Emission Sources and Factors
The factors that contribute to carbon monoxide production in utility
boilers (Section 5.1.2) also contribute to CO formation in industrial boilers
Although industrial boilers have less sophisticated combustion monitoring
systems than larger utility boilers, carbon monoxide emissions may be slight-
ly less because industrial boilers are generally fired with greater amounts
of excess air.3
Reported carbon monoxide emission factors for industrial boilers are
given in Table 5-4.6 The total 1977 carbon monoxide emissions from both
large and small industrial boilers were estimated at 117,800 metric tons
(129,900 tons).7 Emissions from industrial boilers contributed approxi-
mately 0.8 percent of the carbon monoxide emitted from stationary sources
in 1977.7
5.2.3 Control Techniques
Methods of controlling carbon monoxide emissions from industrial
boilers are similar to those discussed for utility boilers in Section 5.1.3.
5.2.4 Cost of Controls
The carbon monoxide control techniques applicable to industrial boilers
are based primarily on maintenance and operational procedures. Capital
5-14
-------
CO
LU
_J
O
CO
_l
J
(U
c
D
C
O
4-J
O
L.
4-1
cu
E
•v^
cn
•—
C
o
4-J
\
-Q
T3
0)
C
-Q
C
O
4-J
O
l_
4-1
CU
\
cn
"7
0
c
o
4-1
\
-Q
-a
c
XI
c
o
4-J
U
L.
4-J
cu
\
cn
LA
o
03
ro
0
^s
x»
LA
CU
C
XI
CO
s_
0)
4-1
^_
0
\
cn
£
o
03
cn
ro
O
x»
LA
0)
C
XI
CO
1_
E
ID
O
cn
CM
CM
03
cn
ro
O
Vs^
XI
LA
^Z.
T3
C
1
XI
(L)
4-J
ro
O
cn
oo
o
03
cn
ro
O
^"
XI
^
C3-
-a
1/1
0)
L.
03
5>
CO
0)
s_
03
•^=>
DQ
<
0) 03
a. o
>- o
4-1
CO
(U O
=3 C
'E
4-J
CO
03
O
U
Cl)
4-J
o
03
x:
4-J
c
<
03
T3
CO
V
—
O
, —
-------
costs for CO control in these units are therefore negligible. Maintenance
costs can possibly be recovered by the fuel savings resulting from more
efficient boiler operation. Control costs are discussed in more detail in
Section 5.1.4.
5.2.5 Impact of Controls
5.2.5.1 Emission Reduction
Because the potential for carbon monoxide emissions reduction from in-
dustrial boilers is small, it is doubtful that the estimated 117,800 metric
tons (129,900 tons) of CO emitted per year can be substantially reduced.
Contributing factors to this situation are similar to those discussed in
Section 5.1.5.
1. Carbon monoxide in the flue gas signifies decreased fuel
combustion efficiency. Therefore, most industrial boilers
are operated to keep CO emissions at a minimum.
2. Smoke emissions resulting from low excess air firing occur
before significant CO emissions are produced. Operating
with too low excess air can therefore be easily diagnosed
and corrected before CO emissions become excessive.
5.2.5.2 Environment
Environmental effects of carbon monoxide emissions reduction from in-
dustrial boilers are similar to the effects described for utility boilers in
Section 5.1.5. However, specific data regarding the trade-offs between NOX
and CO controls were not available for industrial boilers.
5-16
-------
5.2.5.3 Energy Requirements
Excessive carbon monoxide emissions are an indication of inefficient
boiler operation and therefore poor fuel usage. The application of CO con-
trols, which generally improve boiler fuel combustion, will result in in-
creased unit efficiency. Specific data to estimate energy savings were
unavailable.
5.3 RESIDENTIAL, COMMERCIAL, AND INSTITUTIONAL HEATERS
Small-scale combustion units consume a considerable amount of the
total fuel burned in the United States. These combustion sources include
forced air, hot water, and steam space heating systems as well as hot water
heaters. The majority of these sources are fired with gas and oil although
some coal burning equipment, primarily the coal stoker, is still in use.
Total CO emissions from these sources have been estimated at 314,500
metric tons per year (346,700 tons/yr)? Residential fuel burning accounts
for about 79 percent of this total while the remainder is composed of emis-
sions from the commercial/institutional sector.3' 6' 10
5.3.1 Process Descriptions
5.3.1.1 Residential heating
There were an estimated 60 million fuel burning residential heating
plants in operation in 1974.* These units consumed an estimated 8.2 x 1018
Joules/yr (7.8 x 1015 Btu/yr) of fuel.3' 10
The firing capacity of these units is quite low with maximum firing
rates seldom exceeding 117 kilowatts (400,000 Btu/hr).3 The most common
fuels used for residential heating include natural gas and distillate fuel
oil which account for roughly 69 and 28 percent, respectively, of the total
5-17
-------
fuel consumption for this category. The use of coal in residential heating
units has been declining since 1945 due to the availability of cleaner,
more readily utilized fuels.11 As a result, coal accounts for less than 3
percent of the total amount of fuel consumed in these units.3 Small amounts
of other fuels including LPG and wood are also used.
5.3.1.2 Commercial and Institutional Heating
Commercial and institutional systems are used for space heating and hot
water generation. The equipment consists mainly of oil and gas fired warm
air furnaces and firetube boilers.3 The firing capacities of these units
range from 88 kilowatts (300,000 Btu/hr) to 3 megawatts (1 x 107 But/hr).
The total amount of fuel used for commercial and institutional space
heating in 1974 has been estimated to be 4.9 x 1018 Joules/yr (4.6 x 1015
But/yr).3'10 Fuels burned in commercial and institutional heaters include
residual and distillate fuel oil, natural gas, and occasionally coal. Re-
sidual fuel oil use is generally limited to larger units.
5.3.2 Process Emission Sources and Factors
Carbon monoxide emission factors for small combustion sources are listed
in Table 5-5.5 EPA emission estimates for residential, commercial, and insti-
tutional heaters are shown in Table 5-6.
Carbon monoxide is formed as an intermediate product of reactions be-
tween carbonaceous fuels and oxygen. If the conditions necessary for com-
plete combustion are not provided, CO will be included in the combustion
products.12 In general, the conditions required for complete combustion
are:
5-18
-------
TABLE 5-5. CARBON MONOXIDE EMISSION FACTORS FOR RESIDENTIAL,
COMMERCIAL, AND INSTITUTIONAL HEATING
Fue
Emission Factor
Bituminous coal
Stokers
Hand fi red
Anthracite coal
Stokers
Hand fi red
Fuel Oil
Natural gas
LPG
Butane
Propane
Wood
5 kg/metric ton
5 kg/metric ton
0.5 kg/metric ton
45 kg/metric ton
0.63 kg/103 liters
320 kg/106 Nm3
(10 Ib/ton)
(90 Ib/ton)
( 1 Ib/ton)
(90 Ib/ton)
( 5 lb/103 gal)
(20 lb/106 scf)
0.24 kg/103 liters ( 2 lb/103 gal)
0.23 kg/103 liters (1.9 lb/103 gal)
60-130 kg/metric ton (120-260 Ib/ton)
Source: Reference 6
5-19
-------
TABLE 5-6. ESTIMATED 1977 NATIONWIDE CARBON MONOXIDE EMISSION
FROM RESIDENTIAL AND COMMERCIAL/INSTITUTIONAL HEATERS
CO Emissions
Fuel/Heater Type
Anthracite Coal
Res idential
Commercial/Institutional
Bituminous Coal and Lignite
Res i dential
Commercial/Institutional
Residual Oil
Res idential
Commercial/Institutional
Disti1 late Oi1
Res idential
Commercial/Institutional
Natural Gas
Res idential
Commercial/Insti tutional
Keros ine
Res idential
Liquid Propane Gas
Residential/Commercial
TOTAL
metric tons/yr
77.6
0.1
73.5
5.0
0
20.7
38.8
17.3
46.1
2k. 2
6.7
314.5
tons/yr
85.5
0.1
81.0
5.5
0
22.8
42.8
19.1
50.8
26.7
5.0
7.4
346.7
Source: Reference 7
5-20
-------
1. High combustion temperatures,
2. Proper excess air levels for the fuel being fired,
3. Rapid mixing of the fuel and the combustion air, and
4. Sufficient residence time of the combustion gases within
the combustion chamber.
CO emissions are sensitive to the amount of combustion air supplied
to the burner. Figure 5-3 shows the general trend of CO, smoke, NO , and
X
fuel efficiency as a function of the excess air level for a typical oil
burning unit. As excess air is increased from theoretical, emissions of
smoke, CO, and unburned hydrocarbons pass through a minimum while fuel
efficiency and N0x emissions pass through a maximum.3 As indicated in the
diagram, proper excess air levels can result in high fuel efficiency and
low CO and smoke emissions. At excess air levels below this point, CO
and smoke emissions increase because the concentration of oxygen at the
flame is too low to permit complete combustion during the residence time
provided. Too much excess air results in increasing CO and hydrocarbon
emissions because the additional combustion air cools the flame to tempera-
tures below that required for complete combustion. Improperly adjusted
excess air levels are one of the major causes of CO and smoke emissions from
small combustion sources.13s1'"1 5
Before any fuel can be burned, it must be mixed with combustion air.
This is accomplished in oil burning units by atomization of the fuel. Fuel
is delivered under pressure to the burner nozzle where it is atomized into
fine droplets. In larger units, steam or air may be used to aid in fuel
atomization. The combustion air is introduced through swirl vanes located
5-21
-------
(0.002)
1.50
^ (0.0015)
0)
0»
^ 1.00
01 (0.001)
Optimum setting for
minimum emissions and
maximum efficiency
0.50
(0.0005)
0.00
Percent Excess Aira
Values vary for various fossil fuels and combustion unit
characteristics
Source: Reference 3
FIGURE 5-3. GENERAL TREND OF SMOKE, GASEOUS EMISSIONS, AND EFFICIENCY
VERSUS THE PERCENT EXCESS AIR FOR OIL-FIRED RESIDENTIAL
HEATERS
5-22
-------
in the burner throat. The swirl vanes promote rapid mixing between the air
and the atomized fuel. Uneven fuel/air distribution can lead to high CO
emissions. This most often occurs because of improper fuel pressure or a
worn, damaged, or clogged burner nozzle.
High CO emissions may be encountered when burning coal if the coal is
not evenly distributed on the grate. Since coal is a solid, it is more
difficult to obtain good fuel/air mixing. Hence, the excess air levels re-
quired for coal burning are higher than those used for either oil or
natural gas.11
Unlike utility and industrial boilers, many residential and commercial
heaters are fired in cycles and CO emissions during burner startup and shut-
down can be very high. This is because air continues to flow through the
combustion chamber due to natural draft during the burner off period. At
burner startup, the cold combustion chamber walls cool the combustion gases
before complete combustion can occur.16 Besides cooling the combustion
chamber, the heat carried away by the air contributes to a decrease in ovar-
all fuel efficiency.17
A source of post burn emissions for oil fired equipment is fuel leakage
from the nozzle.14 The nozzle absorbs heat from the hot refractory causing
increased CO emissions.16 In coal fired stokers, the coal bed continues
to smolder during the off cycle. Since only a limited amount of air (that
supplied by natural draft) is present, high CO and/or smoke emissions
usually result.ll
5.3.3 Control Techniques
The following paragraphs discuss the principles used in reducing CO
emissions from residential, commercial, and institutional heaters. It is
5-23
-------
recommended that the measures discussed be implemented by qualified service
personnel who are specially trained and who are experienced with the com-
bustion system. Sources for verifying the expertise of service personnel
are 1) the vendor of the combustion system, 2) building safety regulatory
agencies, and 3) local fuel vendors.
The most practical technique for reducing CO emissions from residential,
commercial, and institutional heaters is proper unit maintenance. Several
studies have shown that old, worn out, poorly constructed, or maladjusted
burners are responsible for unnecessarily high levels of air pollutant
emissions.
Other methods of reducing CO emissions are:
1. Reduce unit fuel consumption by improving steady state
and cyclic efficiency,
2. Prevent the cooling off of the combustion chamber in
between heating cycles by dampers,
3. Equip new heaters with combustion modification designs
such as flame retention burners and flue gas recirculation,
and
4. Fuel substitution.
5.3.3.1 Effect of Maintenance
Guidelines for proper maintenance and tuning of residential and commer-
cial heating units are available from many sources including government
agencies, equipment manufacturers, and various trade groups.13'14'15 In
summary, these guidelines recommend the following maintenance procedures
for oil and gas fired residential and commercial heaters:
5-24
-------
1. Clean burner and heat transfer surfaces
2. Clean fuel delivery system
3. Set excess air.
In addition to minimizing CO emissions, a burner tune-up such as de-
scribed above can improve fuel efficiency. An annual tune-up is recommended
by burner manufacturers to maintain good operation.16
Improvements in the heating system fuel efficiency can result in lower
total emissions of all pollutants as less fuel is consumed to supply a given
heating load. A variety of techniques is available which can result in
modest improvement in efficiency. Some of these techniques are listed below:
1. Flame retention burners
2. Added insulation
3. Flue gas recirculation
4. Reduced firing rates.
Reduced firing rates have the added benefit of reducing spike (or sharply
increased) CO emissions. At reduced firing rates, cycle fired equipment
tends to run a greater percentage of the time, thus reducing off-cycle heat
losses and reducing the number of cold start-ups. Since the quenching effect
of combustion gases touching cold areas in the combustion area upon start-up
is a major contributor to spike emissions, any decrease in off-cycle heat
losses will have the tendency of reducing these emissions.
5.3.3.2 Fuel Substitution
As indicated in Table 5-5, CO emissions from small coal-fired units
are significantly higher than CO emissions from oil or gas units. Therefore,
substitution of oil- or gas-fired equipment for small coal-fired equipment
5-25
-------
could result in substantial reductions in total CO emissions from that equip-
ment.
Although modern coal burning units are designed to reduce routine main-
tenance and achieve efficiencies approaching that of oil-fired equipment, CO
and smoke emissions are still quite high, particularly during the units off-
cycle. X1
5.3.4 Cost of Controls
The most effective technique for reducing CO emissions is proper mainte-
nance of the heating unit.16 A general tune-up of an oil-fired residential
furnace including nozzle cleaning or replacement, changing the filters, and
adjustment of the proper excess air level costs in the range of $60 to $80
(1978 dollars).18 The cost for tuning a gas-fired furnace is somewhat less;
no information was available on maintenance costs for coal-fired heaters.
Unit efficiency generally increases with tuning and savings in fuel costs can
often offset the tuning cost. In addition, increased unit life and trouble
free operation act as incentives to keep these units properly tuned. Burner
sales and service organizations recommend that these units be tuned once per
year, preferably at the start of the heating season. 13>16
New burners may be required in units for which normal maintenance
procedures fail to reduce emissions or improve efficiency. New flame
retention burners can be installed in these units for around 250 to 300
dollars in 1978.18 Since these burners can operate at lower excess air
levels than conventional high pressure burners, the resulting improvement
in efficiency can result in substantial fuel savings.18
5-26
-------
5.3.5 Impact of Controls
5.3.5.1 Emissions Reduction
Several studies have shown that old, worn out, or damaged burners are
responsible for unnecessarily high CO emission levels. In the residential
heating sector, the number of units which would require replacement due to
low efficiency, high smoke or CO emissions, or other poor performance char-
acteristics, has been estimated to be in the range of 9 to 30 percent.16'19
Tuning or replacing the burners in commercial and institutional heating
units can also reduce CO emissions. The actual reduction in emissions
resulting from these measures was not determined. However, their effect is
probably less significant than for residential heaters because CO emissions
from commercial heating units are typically lower than those from residential
heaters due to more frequent maintenance and more efficient design.
As mentioned previously, a number of techniques are available which can
provide modest increases in fuel efficiency. The application of these tech-
niques can result in substantial fuel savings while simultaneously reducing
total CO emissions.
Even though coal accounts for less than 3 percent of the total amount
of fuel burned in small combustion sources, CO emissions from coal burning
equipment represent over 70 percent of the total estimated TO emissions.
Hence, a reduction in the use of coal could provide a significant reduction
in total CO emissions from these sources.
5.3.5.2 Environment
The application of controls for CO emissions from small combustion
sources will have both positive and negative impacts with respect to other
5-27
-------
pollutant discharges. Many of the control techniques discussed above result
in improvements in the combustion characteristics of the system. As a re-
sult, these same techniques often provide a reduction in the emission rates
of other combustibles such as smoke and unburned hydrocarbons.
Sulfur dioxide emissions are not directly affected by CO control tech-
niques as most all of the sulfur in the fuel exits with the flue gas. Total
sulfur emissions, however, can be reduced by any technique which results
in improvements in fuel efficiency.
Increased NO emissions may result from the application of CO controls.
X
Those techniques which produce an increase in combustion intensity generally
result in higher flame temperatures with increased NO production.3
X
A considerable amount of effort has been directed toward developing
techniques which reduce NO emissions from combustion sources. In general,
X
these techniques depend on reducing the maximum flame temperature, limiting
the availability of oxygen at the flame, or a combination of these factors.
Unfortunately, these techniques may result in increased CO emissions.3
5.3.5.3 Energy Requirements
The energy impacts of applying CO control techniques to small combustion
systems occur primarily through effects on fuel efficiency rather than the
energy requirements of the control method itself. The most promising CO
control techniques (i.e., tuning, replacement of poor units, firing rate
reductions, and flame retention burners) can all result in improved effi-
ciency and reduced fuel consumption. These improvements in efficiency
result from decreased losses of combustibles such as smokes and CO and a
decrease in both on- and off-cycle stack heat losses.
5-28
-------
5.4 SOLID WASTE INCINERATORS
Incinerators, combustion systems that burn waste materials, are used
to reduce the weight, volume, and volatile contents of refuse. Because re-
fuse characteristics vary widely, methods of incineration must be adjusted
to fit specific types of waste material. In general, refuse differs from
fossil fuels in that refuse grate-loading rates are much lower, and ex-
cess air rates are higher.
Carbon monoxide is a significant pollutant from most incineration pro-
cesses. The greatest CO emissions are produced by municipal, industrial,
and commercial incinerators.20 Although emission rates from residential
incinerators are high, total carbon monoxide emissions are low because of
the low volume of waste burned in residential units.6
The following sections give process/design descriptions for different
types of municipal, industrial, and commercial solid waste incinerators.
Process emission sources and factors are included, as are discussion of
control techniques, control costs, and the impact of controls on carbon
monoxide emission reduction, the environment, and energy requirements.
5.4.1 Municipal Incinerators
Municipal incinerators are designed to dispose of combustible wastes
from residential, commercial, and industrial sources which do not maintain
their own waste disposal facilities. (Heavy industrial, agricultural, and
oversize bulky wastes are not usually treated in municipal incinerators.)
Municipal incinerator capacities range from 45 to 900 metric tons/day (50-
1000 tons/day).21 The estimated average composition of municipal incinera-
tor feed is shown in Table 5-7.
5-29
-------
TABLE 5-7. ESTIMATED ANNUAL AVERAGE COMPOSITION
OF MUNICIPAL REFUSE
Component Mean Weight Percent
Glass 9-9
Metal 10-2
Paper 51.6
Plastics I-1*
Leather and rubber 1-9
Textiles 2-7
Wood 3.0
Food wastes 19-3
100.0
Source: Reference 21
5.4.1.1 Process Description
Municipal waste is usually transported to the incinerator via truck.
After being weighed, the waste is dumped into storage bins or charging hop-
pers. At times the waste is shredded prior to incineration. Refuse is
either batch-fed or continuous-fed into the furnace. Process combustion
control is improved when continuous firing is employed.
A variety of furnace types are currently used in U.S. municipal incin-
erators. Nearly all municipal incinerators are multiple-chambered. Most
municipal batch-fed incinerators consist of vertical cylindrical or rec-
tangular chambers, into which refuse is charged at regular intervals. The
charging doors in vertical batch incinerators are located directly above
the grates; in rectangular batch furnaces the doors are in the rear of the
roof, and refuse travels from rear to front as it burns.
Underfire air is forced up through the incinerator grates, while over-
fire air is introduced through furnace wall ports in the primary combustion
5-30
-------
chamber. The amount of overfire air must be controlled to maintain combus-
tion temperatures of about 980°C-1090°C (1800°F-2000°F) to avoid quenching.22
Flue gases pass from the primary combustion chamber to the secondary chamber,
where oxidation is completed. Gases from the secondary combustion chamber
usually flow to a particulate emission control system.
In continuous-fed incinerators, refuse moves from the charging hopper
down the feed chute into the primary combustion chamber. Fresh refuse en-
tering the primary chamber is ignited by the burning waste and hot combus-
tion gases. Continuous-fed incinerators are similar to batch-fed incinera-
tors with the exception of their charging mechanism. In both types of in-
cinerators, furnace temperatures range from 650°C to 870°C (1200°F-1600°F),22
Flue gases usually remain in the secondary chamber at 870°C (1600°F) for
approximately two seconds.22 Flue gases are cooled by one or a combination
of three methods: (1) direct injection and vaporization of water; (2) with
a heat exchanger (waterwall or convection boiler, air-cooled refractory, or
air preheater); (3) direct dilution and mixing with cool atmospheric air.
Flue gases exit the stack at temperatures of 315°C-370°C (650°F-700°F),22
5.4.1.2 Process Emission Sources and Factors
Carbon monoxide is emitted from municipal incinerator stacks. Emissions
of CO result from improper incinerator design or operating conditions, in-
sufficient secondary combustion chamber temperatures, and disruptions in
burning conditions (e.g., during start-up and shutdown, or after charging
in a batch-fed incinerator).6'23 No carbon monoxide emission control de-
vices are currently applied to municipal incinerators. Uncontrolled emis-
sions of carbon monoxide from mutliple chamber municipal incinerators have
5-31
-------
been estimated at 17.5 kilograms per metric ton of refuse charged (35 Ib/ton).
Emissions vary with refuse composition and furnace operating conditions.
EPA estimates of total CO emissions from multiple chamber municipal in-
cinerators were 155,600 metric tons (171,500 tons) in 1977.7 Another source
gave a much higher estimate of 265,000 metric tons (292,000 tons), calculated
from published emission factors and the amount of solid waste processed.21*
5.4.1.3 Control Techniques
The CO content of incinerator flue gas is reduced through control of the
combustion process. Incinerator furnace design and operation must be care-
fully controlled so that exhaust gas residence time, furnace temperature,
and turbulence are sufficient to achieve complete combustion of CO in the
exhaust gas.2*5 Although afterburners would reduce CO emissions, this type
of system is not applied to municipal incinerators. The incinerator furnace
should be designed so that exhaust gas residence times in the secondary com-
bustion chamber are sufficient to achieve oxidation of carbon monoxide. If
the incinerator is not operated at a high enough temperature [760°C (1400°F)],
25
increased CO emissions will result.
Sufficient combustion air is necessary to achieve optimum incineration
conditions. The underfire air system should provide at least 150 percent of
stoichiometric air requirements and the overfire air jets should be able to
supply approximately 100 percent of stoichiometric air requirements.25'26
Jets must be positioned so that full penetration of the furnace gases and
uniform mixing are achieved. It has been reported that sidewall jets are
more effective than roof jets in promoting maximum mixing.25 Thorough mix-
5-32
-------
ing ensures that sufficient oxygen for complete combustion is available in
all parts of the furnace. Cold gases from the burnout zone of the furnace
must be mixed with hot gases from the burn zone to prevent gas stratifica-
tion and quenching. Controlled underfire air, forced up through the furnace
grate, produces turbulence in the burning refuse bed and thus ensures a more
uniform ignition of the waste.2-1
Continuous-fed incinerators are more easily operated within design
parameters than batch-fed incinerators because the characteristics of refuse
reaching the furnace are more uniform. If too much fresh charge is loaded
into a batch-fed incinerator, the gases from the burning refuse already in
the furnace may be quenched, thus producing high levels of carbon monoxide.
Excessive charge may also increase the rate of burning exceeding air supply
capabilities. When this occurs, carbon monoxide emissions increase because
residence time in the secondary combustion chamber is insufficient and be-
cause there is not enough air for combustion of CO in the exhaust gas.
5.4.1.4 Cost of Controls
No additional equipment, labor, or fuel is used to control carbon
monoxide emissions from municipal incinerators. Therefore, no capital or
operating costs are incurred.
5.4.1.5 Impact of Controls
Emission Reduction—Carbon monoxide emissions from municipal incinera-
tors are minimal if the incinerators are operated according to design speci-
fications. Although CO emissions would be reduced by more careful control
of combustion conditions, it is not known how much CO emissions can be re-
duced by improving operating practices. It is estimated that afterburners
5-33
-------
would reduce CO emissions by as much as 90 percent for cases where combus-
tion temperatures would otherwise be less than 760 C (1400 F).
Environment--0peration of municipal incinerators so that carbon monox-
ide emissions are controlled would not affect the emission rate of nitrogen
oxides (NOX) from the incinerator. Because incinerators operate at rela-
tively low temperatures, most of the nitrogen oxides are formed by direct
conversion of chemically-bound nitrogen in the refuse rather than by the
high temperature reaction of nitrogen in the combustion air. In general,
good operating practice should result in lower emissions of particulates
and hydrocarbons as well as carbon monoxide.
Energy Requirements—Municipal refuse has a similar heating value to
that of peat or lignite.27 The heat content of refuse has been estimated to
range from 9.2-10.4 megajoules/kilogram refuse (3,935-4,450 Btu/pound).
No supplemental fuel is necessary to maintain refuse combustion. The carbon
monoxide content of the furnace exhaust gas varies with refuse content and
furnace operating conditions; no exhaust gas heat contents were reported.
No estimates are available for the energy requirements of afterburner sys-
tems used on municipal incinerators.
5.4.2 Commercial/Industrial Incinerators
Many commercial and industrial operations (e.g., grocery stores, apart-
ment complexes, textile and woodworking industries) use small incinerators
to burn refuse. Most of the units are batch-fed, and many are of single
chamber design.21 The following paragraphs describe several of the more
widely used furnaces.
5-34
-------
For purposes of this discussion, waste gas streams are divided into
two groups-those which require supplemental fuel for incineration and
those which do not. Incineration of those streams which can support combus-
tion and therefore do not require supplemental fuel is straightforward. It
can be treated as a fuel quality stream and burned in a normal waste gas
burner. The resulting temperature, greater than 1200°C (2200°F), is suf-
ficient to completely oxidize any CO.1 In some cases it may be possible to
use the waste gas as fuel in a boiler or process heater and thereby recover
its heating value.
The incineration of a waste gas which cannot support combustion and so
requires supplemental fuel needs careful design of the incinerator equip-
ment to ensure good CO removal. Temperature, residence time, and the degree
of mixing all directly influence the performance of the afterburner. Figure
6-1 diagrams the sequence of steps required for successful incineration of
dilute waste gases.
Temperature and residence time requirements for dilute waste gas incin-
eration are discussed together since they are interchangeable to some degree.
A higher operating temperature allows use of a shorter residence time com-
bustion chamber and longer residence times allow lower temperatures. This
flexibility is limited due to the strong temperature dependence of oxida-
tion rates. Figure 6-2 shows the general effects of temperature and resi-
dence time on oxidation rates in a flow-through reactor.1 These curves do
not represent carbon monoxide specifically, but instead give an indication
of how combustible pollutants respond to these operating variables. After-
burner experience shows that temperatures of 76G-790°C (1400-1450°F) are
6-5
-------
4J
C
C 0)
CT) 3
CU rH
H M-l
O 4-<
W
4-1
0
i-i
Retentio
^
WD
•H
ffi
4-1
CO
cn
0)
1
h
^
o
4-1
CU
«-t
Temperatl
cu
S
•H
H
4-J
r-<
Suf ficiei
i
o c
•H
oo
UJ
u_
UJ
h-
ID
_l
i— i
O
u_
o
•z.
o
4-1
C -H
CU >-i
60 fi T3 O
>•> O CU
O 4J CU 0)
cn cu a
CQ CU
Ml
cn >>,
c
o
•H
rH 4-J
cu cn
PH
6
o
o
ex B *-i
CU O -H
CO rH X
rH 0) 3 O
o cu -d o
4-1 3 -H P4 B^S
FT^ U^ ^.Q
CU 4J (U rH
3 O O P
fe 4-1 v-^ P4 V
O)
o
E
O)
t-
cu
^-
cu
cu
o
O
oo
CU
§
rH
ex
co
oo
UJ
o
o
ZD
00
o
Q
UJ
cr
OO
Q_
OO
cu i cu
T3 ' cn
•H 3
CO I
4_) 4-1
3 I-H
O 1^
CU
4-J
D
rH
•H
Q
6-6
-------
The combustion chamber and heat recovery equipment are the major pieces
of equipment for an incineration system. Auxiliary equipment includes
blowers, ducts, supporting structure, and de-entrainment devices. Blowers
are needed if the waste gas is at insufficient pressure to move it through
the ductwork and the combustion chamber. The blower may be either forced
draft or induced draft. Each type of fan has advantages and disadvantages
depending on the specific application.
The design and layout of the ductwork depends primarily upon the source
of the waste gas and the location of the incinerator. Careful attention
should be paid to its design for safety and economic reasons. Long duct
runs can cost more than the afterburner itself. Condensation of combustible
material can occur even in insulated ducts, causing a fire hazard. For those
applications where the waste gas is at a concentration above 25 percent of
the lower explosive limit (LEL) of the gas, but below the upper explosive
limit (UEL), provision must be made to prevent flashback through the ductwork
to the process source.1 This is done by providing high velocity sections
where the waste gas velocity is higher than the flame propagation velocity.1
Another preventive measure is to dilute the waste gas with air to below
25 percent of the LEL. If concentrations are above the UEL, the waste gas
may be ducted without the need for air dilution.1 It is essential that
air be excluded at all points between the waste gas source and the incinera-
tor to prevent an explosive mixture from forming.1
The supporting structure for the afterburner represents an important
piece of auxiliary equipment insofar as installation is concerned. If the
system is mounted on a concrete pad on the ground, its weight will have
6-3
-------
little influence on installation. However, long duct runs can be avoided if
the incinerator can be located close to the waste gas source. This arrange-
ment results in a safer and less expensive system. Roof mounting is therefore
frequently done since besides avoiding long duct runs it also saves space
within the building and eliminates the need for a tall stack on the incin-
erator. The primary disadvantage of this location is that for roofs not
strong enough to take the additional load, a special (and expensive) support-
ing structure will be required; or if this cost is prohibitive, lightweight
afterburner designs or ground level installation will be needed.1
In some applications, the CO-containing waste gas may contain liquid
or solid particulate matter which may significantly affect operation of an
afterburner. Provisions for removal of this must be made in equipment
design and selection to ensure proper operation of the incinerator. There
are a large number of different types of equipment for removing particles
and mists including fabric filters, electrostatic precipitators, cyclones,
demisters, etc. Depending upon the nature of the solid or liquid, suitable
devices can be installed upstream of the afterburner.
Operating principles - Good removal of the carbon monoxide in a waste
gas simply requires contacting the gas with sufficient oxygen at high
enough temperature for the CO to be oxidized in the time available. There-
fore, the three principles of good combustion-time, temperature, and
turbulence-hold true for waste gas incineration as well. The difficulty
does not come in recognizing their value, but in actually putting them into
practice. The following discussion presents information on the conditions
necessary for proper operation of a thermal incinerator to control carbon
monoxide.
6-4
-------
6. INDUSTRIAL PROCESS SOURCE CONTROL SYSTEMS
This section examines those control systems which are used to control
carbon monoxide emissions from industrial process sources. The specific
controls examined include:
1. incinerators (thermal and catalytic)
2. flares and plume burners, and
3. carbon monoxide boilers.
A technical and economic assessment is presented for each of the controls
listed above. The technical assessment includes discussions on equipment and
operating principles, control efficiencies, and feasible areas of application.
The economic assessment includes both capital and annualized cost curves for
representative systems.
6.1. INCINERATORS
Incineration is the most applicable and efficient control technology for
reducing carbon monoxide emissions from most industrial process sources.
There are two basic designs currently used in the pollution control field for
incinerators (or afterburners)-thermal and catalytic. Both have advantages
in certain applications and both have been used extensively to destroy com-
bustible pollutants in waste gas streams by oxidation to C02 and water. The
main use of afterburners in the past has been for odor, hydrocarbon, and
6-1
-------
smoke control. There are some applications, however, in carbon monoxide
control. The remainder of this section examines the application of inciner-
ators specifically for the control of carbon monoxide emissions.
6.1.1 Equipment and Design Parameters for Thermal Incinerators
Equipment—Carbon monoxide emissions are controlled in thermal incinera-
tors by heating in the presence of oxygen the CO-containing waste gas to a
temperature sufficient to allow complete oxidation in the residence time
available. The incinerator itself is a steel shell, refractory-lined
combustion chamber. A burner is located at one end through which the waste
gas is introduced into the chamber along with supplemental fuel, should it be
needed. Alternatively, the fuel may be burned with air and the hot combus-
tion gases mixed with the waste gas just after the burner. This arrangement
is usually used when the waste gas does not contain enough oxygen to oxidize
all the fuel, carbon monoxide, and other combustible pollutants present in
the waste gas.
As fuel costs have risen in recent years, the incentive for recovering
available heat in the incinerator flue gas has become strong. This has led
to the application of numerous heat recovery techniques. Recovery methods
include heat exchange between hot flue gas and incoming cool waste gas,
recycling a portion of the hot flue gas back to the process to supply heat
directly, and using the heat to generate steam for other processing or heat-
ing loads in the plant. Fuel savings from employing any of these alternatives
can usually pay for the cost of the heat recovery equipment.1
6-2
-------
23. Jahnke, J.A. et al. A Research Study of Gaseous Emissions from a
Municipal Incinerator. J. APCA 27(8): 747, 1977.
24. Achinger, William C. and Richard L. Baker. Environmental Assessment
of Municipal-scale Incinerators. Open File Report SW-111. U.S.
Environmental Protection Agency, Research Triangle Park, North
Carolina, 1973.
25. Incinerator Overfire Mixing Demonstration, Final Report. EPA 600/2-
75-016, PB 245 015. U.S. Environmental Protection Agency, Research
Triangle Park, North Carolina, August 1975.
26. Mohn, C. Michael, Richard H. Stephens, and Thomas J. Lamb. Applica-
tion of Incinerator Jets to Municipal Incinerators. Paper No. 73-225,
presented at the 66th Annual APCA Meeting, Chicago, Illinois.
June 1973.
27. Chansky, Steven H. et al. Systems Study of Air Pollution from Munici-
pal Incineration, Vol. 1. Arthur D. Little, Inc., Cambridge,
Massachusetts. March 1970.
28. Air Pollution Engineering Manual, 2nd ed. AP-40, PB 225 132. U.S.
Environmental Protection Agency, Research Triangle Park, North
Carolina. 1973.
29. Cross, Frank L., Jr. Controlled Air Incinerators. Pollution Eng.
1973 (December), 30.
30. Rolke, R.W. et al. Afterburners Systems Study. EPA-R2-72-062,
PB 212 560, U.S. Environmental Protection Agency, Research Triangle
Park, North Carolina, 1972.
5-43
-------
-------
5.4.2.1 Process Descriptions
Flue-fed incinerators are single-chamber, rectangular furnaces in
which the stack also serves as a charging chute for refuse. Refuse is dried
by gas burners located below the grates. Refuse is ignited through a charg-
ing door above the grates, and ash is removed through a cleanout door at the
bottom of the furnace. Overfire and underfire air jets are usually installed
in both doors.
Conical incinerators are used by some lumber, wood product, and textile
industries to burn wood or fiber waste. Combustion control is difficult in
this type of incinerator because the addition of combustion air is not con-
trolled. A typical conical burner consists of a cone-shaped sheet metal
shell with a mesh screen on top. Refuse is charged through a door near the
top of the burner and falls to a fuel pile where it is ignited. Air is
supplied through small tangential inlets near the base of the burner.
Silo incinerators are vertical steel cylinders which are sometimes lined
with refractory brick. They are charged and fired similar to conical burners.
but operate at higher temperatures because of the refractory-lined chamber.
Both single- and multiple-chamber units are in current use. Combustion air
is supplied through louvers located at the base of the incinerator.
Temperatures in the combustion chambers of the above incinerators will
vary with the amount of combustion air, charging method, and type of refuse
burned. In general, temperatures range from 540°C-980°C (1000°F-1800°F) .28
In the single chamber incinerators described above, turbulence and gas resi-
dence time are difficult to control and vary widely.
The controlled ("starved") air incinerator is a relatively recent de-
velopment. This type of unit is always two-chambered. The concentration
5-35
-------
of carbon monoxide-rich exhaust gas produced in the burner's primary chamber
is reduced when additional air is added in the incinerator's secondary cham-
ber. Controlled air incinerators may be batch- or continuous-fed, and typi-
28
cally operate at temperatures of 1090°C-1200°C (2000°F-2200°F). ° Secondary
chamber residence time is longer than in conventional incinerators (1.25-1.60
seconds).29 An efficient starved air incinerator is equipped with a primary
burner to initiate incineration and with a secondary burner to oxidize the
combustibles in the off-gases when temperatures are less than 870°C (1600°F).
5.4.2.2 Process Emission Sources and Factors
Uncontrolled emission factors for various types of commercial/industrial
incinerators are given in Table 5-8.
TABLE 5-8. CARBON MONOXIDE EMISSION FACTORS FOR SELECTED
COMMERCIAL/INDUSTRIAL INCINERATORS
Emission rates (uncontrolled)
Incinerator type Kilograms/metric ton Pounds/ton
Industri al/commercia1
Multiple chamber 5 10
Single chamber 10 20
Flue-fed single chamber 10 20
Controlled air negligible negligible
Source: References 6, 29
Emissions estimated for 1977 are shown in Table 5-9. As the table indi-
cates, conical incinerators produced almost 50 percent of the carbon monoxide
emitted from industrial and commercial incinerators.
5-36
-------
TABLE 5-9. ESTIMATED 1977 CARBON MONOXIDE EMISSIONS
FROM COMMERCIAL/INDUSTRIAL INCINERATORS
Incinerator type
Conical , al1 fuels
Other, all fuels
TOTAL
Total Mas Emissions
metric tons
530,700
655,000
,185,700
tons
585,000
722,000
1,307,000
Source: Reference 7
Control techniques (e.g., afterburners and draft controls) are appli-
cable to flue-fed incinerators and other types of single- and multiple-
chamber incinerators.
5.4.2.3 Control Techniques
The more simple design characteristics of most commercial/industrial
incinerators make carbon monoxide control through good operating practices
difficult. In single chamber incinerators, exhaust gas mixing and residence
times are insufficient to achieve complete combustion of CO in the exhaust
gas. Conical and silo burners have virtually no means of combustion air
control, so temperatures and burn rates will vary.
Direct flame afterburners are reportedly applicable to flue-fed incin-
erators and other types of commercial/industrial incinerators,6'21 This
type of afterburner typically operates at temperatures of 650-980°C (1200-
1800°F), with residence times ranging from 0.3-0.6 seconds.21 Control
5-37
-------
efficiencies of 90 percent CO removal can reportedly be achieved if an after-
burner is operated at temperatures of at least 760°C (1400°F).28 Catalytic
afterburners are not feasible because exhaust gas from burning refuse con-
tains substances which foul the catalyst.
Installation of controlled air incinerators as replacements for less
sophisticated units would result in substantial carbon monoxide emission
reductions. These units can be used to combust a variety of wastes and are
designed for capacities of 180-1360 kilograms/hr (400-3000 pounds/hr),29
Emissions of CO from controlled air incinerators have been reported as
negligible.29
5.4.2.4 Cost of Controls
Chapter 6 contains a detailed presentation of the capital and annua-
lized costs for thermal incinerators. To accurately determine the costs for
applying this control to refuse incinerators, flow rates and composition of
the flue gas are needed. Due to the variations in operation of existing
refuse incinerators, flow rates and composition of the flue gases from the
units will change significantly not only from one unit to the next but also
from time to time for a given incinerator. Data characterizing compositions,
flow rates, and their variations were not available. Without this informa-
tion accurate costs cannot be determined for thermal incineration of the
flue gas from this source.
5.4.2.5 Impact of Controls
Emissions Reductions — If afterburners were applied to existing commer-
cial/industrial incinerators, or if existing units were replaced by effi-
cient controlled air incinerators, carbon monoxide emissions from these
5-38
-------
sources would be substantially reduced. Based on 1977 emissions data, an
emissions reduction of 1,207,000 metric tons (1,331,000 tons) could be
achieved assuming these controls had removal efficiencies of 90 percent.
Environment—The use of afterburners will increase the amount of nitro-
gen oxides (N0x) emissions from commercial and industrial incinerators. Un-
less afterburner operating temperatures exceed 980°C (1800°F), however, NOX
emissions will remain relatively small (20-30 ppm).30 Sulfur oxides emis-
sions may increase if fuel oil rather than natural gas is used as supplemen-
tary afterburner fuel. The use of better-designed incinerators, such as
controlled air incinerators, as well as afterburners, should reduce emissions
of combustible particulates and hydrocarbons in addition to carbon monox-
ide.29
Energy Requirements—Supplementary fuel will be required to maintain
combustion in afterburners applied to incinerator stacks. The amount of fuel
will depend on the type of refuse burned and the operation of the incinera-
tor. Typical afterburner fuel requirements are described in Chapter 6.
If controlled air incinerators are installed, the small quantities of com-
bustion air required results in reduced amounts of fuel necessary to fire
the incinerator itself.29
5-39
-------
REFERENCES FOR CHAPTER 5
1. Inventory of Combustion-related Emissions from Stationary Sources,
2nd Update. EPA 600/7-78-100. U.S. Environmental Protection Agency
Research Triangle Park, North Carolina, June 1978.
2. Shields, Carl D. Boilers: Types, Characteristics, and Functions.
McGraw-Hill, New York. 1961.
3. Control Techniques for Nitrogen Oxide Emissions from Stationary
Sources. Final Report, 2nd edition. EPA 450/1/78-001, U.S. Environ-
mental Protection Agency, Research Triangle Park, North Carolina,
January 1978.
4. Putnam, A.A., E.L. Kropp, and R.E. Barrett. Evaluation of National
Boiler Inventory, Final Report. EPA Contract No. 68-02-1223, Task 31.
Battelle Columbus Lab., Columbus, Ohio. October 1975.
5. Stationary Watertube Boiler Sales Data, American Boiler Manufacturers
Association, Arlington, Virginia, Updated.
6. U.S. Environmental Protection Agency. Compilation of Air Pollution
Emission Factors. Second Edition with supplements. AP-42. U.S.
Environmental Protection Agency, Research Triangle Park, North Carolina,
1972-1977.
7. National Air Quality Monitoring and Emission Trends Report, 1977,
EPA-450/2-78-052, and supporting background information. U.S.
Environmental Protection Agency, Research Triangle Park, North
Carolina, December 1978.
5-40
-------
8. Stationary Source Combustion Symposium, Vol. 3, Field Testing and
Surveys, Proceedings. EPA 600/2-76-152c, PB 257 146. U.S. Environ-
mental Protection Agency, Research Triangle Park, North Csrolina,
June 1976.
9. Proceedings of the Second Stationary Source Combustion Symposium,
4 vols. EPA 600/7-77-073a-d. U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina, July 1977.
10. Source Assessment: Overview Matrix for National Criteria Pollutant
Emissions. EPA 600/2-77-107c. U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina, July 1977.
11. Emissions From Residential and Small Commercial Stoker-Coal-Fired
Boilers Under Smokeless Operation, Final Report. EPA 600/7-76-029,
PB 263 891. U.S. Environmental Protection Agency, Research Triangle
Park, North Carolina, October 1976.
12. Control Techniques for Carbon Monoxide Emissions from Stationary
Sources. Pub. No. AP-65, U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina, March 1970.
13. Himmel, Robert L., Douglas W. DeWerth and David W. Locklin. Guidelines
for Adjustment of Residential Gas Heating Equipment for Low Emissions
and Good Efficiency. Paper No. 78-49-4. Presented at the 71st Annual
APCA Meeting, Houston, Texas. June 1978.
14. Guidelines for Residential Oil-burner Adjustments. EPA 600/2-75-069a,
PB 248-292. U.S. Environmental Protection Agency, Research Triangle
Park, North Carolina, October 1975.
5-41
-------
15. Guidelines for Burner Adjustments of Commercial Oil-fired Boilers.
EPA 600/2-76-088. U.S. Environmental Protection Agency, Research
Triangle Park, North Carolina, March 1976.
16. Study of Air Pollutant Emissions from Residential Heating Systems,
Final Report. EPA 650/2-74-003, PB 229 667. U.S. Environmental
Protection Agency, Research Triangle Park, North Carolina. January 1974.
17. Residential Oil Furnace System Optimization—Phase I, Final Report.
EPA 600/2-76-038, U.S. Environmental Protection Agency, Research
Triangle Park, North Carolina, February 1976.
18. Katzman, L. and D. Weitzman. A Study to Evaluate the Effect of Reduc-
ing Firing Rates on Residential Oil Burner Installations. Abcor Inc.,
Walden Research Division, Wilmington, Massachusetts. Undated.
19. Field Investigation of Emissions from Combustion Equipment for Space
Heating, Final Report. EPA-R2-73-084A. PB 223 148. U.S. Environ-
mental Protection Agency, Research Triangle Park, North Carolina, 1973.
20. National Emissions Data Systems (NEDS) by Source Classification Code.
National Air Data Branch, U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina. Feburary 24, 1978.
21. Field Surveillance and Enforcement Guide: Combustion and Incineration
Sources, Final Report. APTD-1149, PB 226 324. U.S. Environmental
Protection Agency, Research Triangle Park, North Carolina, June 1973.
22. Chansky, Steven H, et al. Systems Study of Air Pollution from Municipal
Incineration, Vol. 2. Arthur D. Little, Inc., Cambridge, Massachusetts.
March 1970.
5-42
-------
4. antimony
5. mercury
6. lead
7. zinc
8. tin
9. sulfur
10. halogens.
All except sulfur and halogens form alloys with the metal catalyst and
therefore permanently deactivate the catalyst.1 However, sulfur and halo-
gens, in most cases, combine in a reversible chemical reaction with the
metal. Catalyst activity is usually restored when the sulfur or halogen-
containing species is removed from the waste stream.1
6.1.3 Incinerator Control Efficiency
The control efficiency of carbon monoxide in dilute quantities in a
waste stream by thermal incineration depends primarily upon three factors:
residence time, temperature, and degree of mixing. Proper design Df an
incinerator taking these three factors into consideration can result in a
thermal incinerator capable of consistent removal of CO at efficiencies
exceeding 90 percent. Higher efficiencies (greater than 95 percent) can
be designed for at the expense of higher capital and operating costs to
achieve longer residence times and higher operating temperatures.
Control of carbon monoxide by catalytic incineration depends primarily
upon the operating temperature and bed volume. Properly designed and
operated, a catalytic incineration system can consistently achieve CO
6-15
-------
removal efficiencies of greater than 90 percent. Higher efficiencies
(greater than 95 percent) will require greater capital outlays, mainly for
increasing catalyst bed volume. Compensation for deactivation of the
catalyst will have to be included in the initial design and during opera-
tion to ensure good CO removal over a period of time.
6.1.4 Applicability
Thermal incinerators are applicable to virtually all sources of carbon
monoxide containing waste gases which are below the lower explosive (combus-
tion limit. (As mentioned earlier, gases which can support combustion
would not be disposed of in an incinerator, but rather flared through a
waste gas burner or sent to a boiler or process furnace for heat recovery.)
Catalytic incinerators would be limited somewhat in their application to
dilute waste gases. This is due to the presence of catalyst poisons in
some gases.
6.1.5 Energy Requirements
In general, the energy requirements for thermal or catalytic incinera-
tors depend upon the following factors:
1. concentration of CO and other combustibles in the waste gas,
2. waste gas temperature,
3. oxygen content of waste gas,
4. incinerator operating temperature, and
5. amount of heat recovery employed.
The concentration of carbon monoxide and other combustibles in the
waste gas can have a significant effect upon energy requirements for thermal
6-16
-------
or catalytic incinerators. The heat released upon oxidation of CO at a
concentration of 25 percent of the lower explosive limit in a waste gas is
sufficient to raise the temperature of a normal cubic meter (0.3 scf) of that
gas by 340°C (650°F).
The temperature of the waste gas also affects the amount of energy
required for its incineration. Most if not all of the supplemental fuel con-
sumed for thermal and catalytic incinerators is used to raise the temperature
of the waste gas up to the design operating temperature of the unit.
If the oxygen content of the waste gas is sufficient (16 percent or
greater) to oxidize the supplemental fuel and combustibles in the waste gas,
significant energy savings will result. This is because the use of outside
air for the oxygen will require fuel to be consumed to heat the air up to
the operating temperature of the incinerator.
As mentioned, the heat required to raise the waste gas (and air if
needed) to the operating temperature of the incinerator is the primary
energy requirement for incineration. Therefore the incinerator operating
temperature as well as the waste gas temperature affect the amount of
supplemental fuel needed.
Heat recovery techniques can lower the amount of supplemental fuel
required for incineration significantly. The simplest and probably most
common form of heat recovery employed in incinerators is the use of the
hot flue gas from the incinerator to heat up the incoming waste gas. This
is referred to as primary heat exchange and a simple diagram of an incinera-
tion system utilizing this technique is shown in Figure 6-1. Another heat
recovery technique, commonly referred to as secondary heat recovery,
6-17
-------
utilizes the remaining heat in the incinerator flue gas after primary heat
recovery. Application of this technique is limited to plants which have
a use for additional heat. Secondary recovery involves further heat
exchange with a process stream or use of the hot gases for drying.
To accurately estimate the energy requirements for thermal and catalytic
incinerators, each of the above factors must be considered. Due to the
potential for wide variation in each, reporting a single energy requirement
or set of requirements would not provide an accurate representation. Plots
are presented which should yield reasonable estimates of the energy require-
ments for incineration. Figures 6-6, 6-7, and 6-8 can be used to determine
the energy requirements for a wide variety of thermal incinerator applica-
tions. Figures 6-9, 6-10, and 6-11 can be used similarly for catalytic
incineration. These plots were taken from the Shell Afterburner Systems
Study and modified to reflect conditions representative of incinerators
designed to control waste gases containing carbon monoxide.1
Within the graphs, provisions are made to account for factors affect-
ing the energy requirements for thermal and catalytic incinerators. The
operating temperatures of both types of incinerators are fixed and all cal-
culations are based on these temperatures. For thermal incinerators, the
temperature chosen was 870°C (1600°F) and for catalytic, 480°C (900°F).
These temperatures should be sufficient to oxidize not only all CO in a
waste gas but virtually all organics as well.
To calculate the energy requirements for a particular application, it
is first necessary to assume a heat exchanger recovery factor. Typical
6-18
-------
TB,°F
1000-
900-
800-
700-
600-
500-
400-
300
Exchange Recovery Factor
90°C(200°F)
40°C(100'F)
TC»870'C(1600'F)
20
100
I 1 1
40 60 80
Exchanger Recovery Factor
Percent Reduction of Supplemental Fuel Requirements with No Heat Recovery
Source: Reference 1
FIGURE 6-6. EFFECT OF EXCHANGER RECOVERY FACTOR AND WASTE GAS TEMPERATURE
ON INLET TEMPERATURE TO THERMAL INCINERATOR
6-19
-------
60
(2020)
50
(1680)
40
(1345)
« 30
i (ioio)
•»-J
o
20
(670)
10
(335)
Operating Temperature: 870'C(1600'F)
Combustion Oxygen from Outside Air
—T~
200
400
600
800
1000
Incinerator Inlet Temperature.T
Source: Reference 1
FIGURE 6-7 THERMAL INCINERATOR ENERGY REQUIREMENTS WITH NO HEAT
RECOVERY OXYGEN FROM OUTSIDE AIR
6-20
-------
60
(2020
50
(1680)
40
(1345)
30
(1010)
* 20.
(670)
10
(335)
Operating Temperature: 870°C(1600'F)
Combustion Oxygen from Waste Gas
2°0 400 600 800
Incinerator Inlet Temperature, °F
Source: Reference 1
1000
FIGURE 6-8. THERMAL INCINERATOR ENERGY REQUIREMENTS WITH NO HEAT
RECOVERY OXYGEN FROM WASTE GAS
6-21
-------
700
600
500-
400-
300-
200-
100-
Exchange Recovery Factor
40-C(100'F)
0
"T"
20
T
40
"T"
60
T~
80
100
C 480°C(900"F)
Exchanger Recovery Factor
Percent Reduction of Supplemental Fuel Requirements with No Heat Recovery
Source: Reference 1
FIGURE 6-9. EFFECT OF EXCHANGER RECOVERY FACTOR AND WASTE GAS
TEMPERATURE ON INLET TEMPERATURE TO CATALYTIC
INCINERATOR
6-22
-------
40
(1345)
Operating Temperature: 480 C(900*F)
Combustion Oxygen from Waste Gas
30
(1010)
200
Incinerator Inlet Temperature,aF
Source: Reference 1
FIGURE 6-10.
WITH
6-23
-------
40
(1345)'
CQ
~ 30
(1010)
2. 20 .
M (670)
-------
\
o
o
00
o;
o
o
o
t— I
O)
O
O)
S_
O)
-------
Deactivation may occur due to several mechanisms. Thermal aging is
probably the most common. It involves micro-structure changes in the
active metal or the porous support and loss of active metal by erosion,
attrition, and vaporization. Proper operating temperatures can slow this
aging and allow satisfactory performance from a unit for three to five
years.1 However, thermal aging may be accelerated by increasing bed temp-
erature. Upper limits of 590°C (1100°F) for alumina-based catalysts and
810°C (1500°F) for all-metal catalysts are recommended by manufacturers
for maximum bed life.1 To keep bed temperatures below these levels, it is
generally recommended that catalytic incinerators be limited to waste gases
with combustible concentrations below 25 percent of the lower explosive
limit.2
A second mechanism for deactivation is the buildup of coatings on the
surface of the catalyst. These are commonly condensed (and polymerized or
partially charred) organic material and/or layers of inorganic particulates.
They deactivate the catalyst by inhibiting contact between the gas phase and
the catalyst surface. Unlike thermal aging, which is irreversible, periodic
cleaning is usually effective in restoring up to 90 percent of the initial
catalyst activity where surface coating is the deactivstion mechanism.1
The final mechanism for deactivation is poisoning by specific contami-
nants in the waste stream. These contaminants either combine chemically
with the active metal or form alloys with it.1 These poisons include:
1. phosphorus
2. bismuth
3. arsenic
6-14
-------
per unit flow rate of waste gas.1 Temperature is important because of its
influence on the effective rate constant for the oxidation of carbon monox-
ide. The catalyst bed volume is important in that it determines the operat-
ing capability of the system and the overall CO removal efficiency. Figure
6-4 shows the relative effect of catalyst bed volumes on pollutant conver-
sion.1 It shows that about twice the volume of catalyst is required for
90 percent conversion as for 66 percent conversion. And twice again is
required to go from 90 percent to 99 percent. This figure is not based on
carbon monoxide specifically, but the general relationship should be repre-
sentative of that expected for carbon monoxide. Figure 6-5 shows the effect
of catalyst bed temperature on the conversion efficiency of carbon monoxide.1
Control efficiencies greater than 90 percent can be achieved at temperatures
above 430°C (800°F).1
Besides temperature and bed volume another factor affecting the CO
oxidation performance of a catalytic incinerator is the deactivation of the
catalyst with age and exposure. This must be compensated for in the initial
design and also during subsequent operation of the system.1 This compensa-
tion may include:
1. initial overdesign in catalyst bed volume,
2. raising preheat temperatures as catalyst activity decreases,
3. cleaning the catalyst during periodic shutdowns,
4. replacement of the catalyst, or
5. treating the waste gas for removal of potential poisons prior
to feeding into the incinerator.
6-11
-------
100
60
o
2
40
0.5
1.0
1.5
2.0
2.5
Source: Reference 1
FIGURE 6-4. VOLUME OF CATALYST/VOLUMETRIC FLOW RATE OF WASTE STREAM*
*Does not apply quantitatively to carbon monoxide.
6-12
-------
6J'2 Equipment and Design Parameters for Catalytic Incineration
" The basic equipment used for a catalytic incineration
system is shown in Figure 6-3. This consists of a combustion/mixing
chamber upstream of the catalyst bed. A preheat burner is usually located
in this chamber to bring the temperature of the waste stream up to
required oxidation temperature. The chamber is also designed to achieve
a uniformly distributed mixture of the combustion gases from the preheat
burner and the waste gas. The catalyst bed is located at the end of the
chamber. It usually consists of a metal mesh-mat, ceramic honeycomb, or
other ceramic matrix structure with a surface deposit or coating of finely
divided particles of platinum or other platinum family metals. The metal
acts as the catalyst while the matrix structure serves to support the
catalyst. The support is designed for high surface area for relatively
small bed volumes to maximize the number of active sites where the catalyzed
oxidation reaction can take place. Relatively small catalyst bed volumes,
0.014 - 0.057 m3 (0.5 - 2.0 ft3), are required per 27 Nm3/min (1000 scfm)
of waste gas.1 This small volume and the low density of the catalyst bed
contribute to relatively small sizes and light weights for catalytic versus
thermal units. Heat recovery from the flue gas out of the catalyst bed may
be included in the overall system design. It will be similar to that for
a thermal unit; however, because of the lower operating temperatures and
supplemental fuel requirements, less energy can be recovered.1"
Design Paramete_rs_ - Catalytic incineration of carbon monoxide depends
primarily upon two factors, operating temperature and catalyst bed volume
6-9
-------
CLEAN, HOT GASES
TO HEAT RECOVERY
OR STACK
CATALYST
ELEMENTS
WASTE GAS
PREHEATER BURNER
FIGURE 6-3. SCHEMATIC DIAGRAM OF CATALYTIC AFTERBURNER USING TORCH-TYPE
PREHEAT BURNER WITH FLOW OF PREHEATED WASTE STREAM THROUGH
FAN TO PROMOTE MIXING
6-10
-------
*
•ZL
O
X
O
O
Q_
OJ
O
E
cu
S-
CU
M-
CU
o
oo
0.
CD
oo
I—
LU
o
LU
Q-
ZD
O
CNJ
I
CC
ID
CD
CU
-a
X
o
o
O
i-
(J
O
4->
>>
CU
o.
Gi-
ro
CO
CU
O
o
6-7
-------
required with an actual residence time at this temperature of 0.2-0.4
seconds after mixing of the waste gas and the hot combustion gases.1 These
conditions should result in nearly complete oxidation of CO.1 However, due
to difficulties in achieving complete mixing of the gases in the combustion
chamber temperatures of 870-980°C (1600-1800°F) and residence times of 0.5
seconds are often designed for in actual applications to ensure good CO
removal.l
As just noted, incinerators with designs which achieve good mixing are
needed not only to ensure adequate CO removal but also to allow operation of
the system as close to the ideal (and least expensive) conditions of 760°C
(1400°F) and 0.2-0.4 second residence times. Operation at this lower tempera^
ture and time requires the fuel to be burned as rapidly as possible and the
hot gases to be thoroughly mixed with the waste gas.
Thorough mixing can be achieved by using either distributed burners or
discrete burners with internal baffles. Distributed burners are placed
directly in the waste gas stream and divide the flame into many individual
jets surrounded by waste gas. This subdivision greatly enhances the mixing
of the waste and hot combustion gases. Distributed burners have certain
limitations which make them unavailable for some applications. They are sub-
ject to fouling, have somewhat limited turndown, can burn only gaseous sup-
plemental fuels, and are difficult to use when outside air is used to supply
oxygen for combustion.1 Where distributed burners are not feasible, dis-
crete burners are employed. Because of their design (one burner versus
several for the distributed design) mixing is more difficult to achieve.
Internal baffles and/or longer residence times are needed for sufficient
mixing of the gases in the combustion chamber.1
6-8
-------
heat exchanger recovery factors for primary heat exchange are 35 to 45 per-
cent. Higher recovery factors (up to 85 percent) are possible with second-
ary heat exchange if potential exists for utilizing this technique at a
given site. Knowing the waste gas temperature (TA) the temperature into
the incinerator (TB) may be determined from Figure 6-6 for thermal incinera-
tion or Figure 6-9 for catalytic incineration. Then, entering the appropriate
graph, Figure 6-7 or 6-8 for thermal, or Figure 6-10 or 6-11 for catalytic
at temperature TB, and knowing the heat content of the waste gas, the sup-
plemental fuel requirement can be read. Allowing for heat losses from the
incinerator may add approximately 5 percent to the fuel requirement shown
on the graph.l
6.1.6 Environmental Impact
Incineration of waste gases can increase emissions of S09 and NO
Z X*
The primary source of the S02 is the sulfur contained in the supplemental
fuel used in the incinerator. Depending upon specific conditions and sul-
fur content of the fuel and waste gas, S02 emissions may vary from neglig-
ble to over 50 ppm. This is not considered significant, however.
The N0x emissions result from the oxidation of any nitrogen compounds
in the waste gas as well as to a limited extent the reaction between atmos-
pheric nitrogen and oxygen. However, due to design and operation dif-
ference, incinerators (particularly catalytic ones) have relatively low
N0x emissions. Reported levels of N0x in the flue gas from thermal after-
burners fired with gas at temperatures up to 980°C (1800°F) are 40-50 ppm
and for catalytic afterburners, 15 ppm.1,? N0x emissions from oil-fueled
6-25
-------
thermal afterburners fired at the same temperatures were reported to be
from two to three times higher.1
Incineration of waste gases containing halogen compounds can result in
the formation of corresponding acids, e.g. chlorine will form hydrochloric
acid. Provisions must be made to remove this from the incinerator flue gas.
Usually this is done by wet scrubbing.1
6.1.7 Costs (Mid-1978 Dollars)
The capital and annualized costs for thermal and catalytic incinerators
are presented in this section. Capital costs for incinerators depend pri-
marily upon the flow rate of the waste gas being incinerated, but also are
affected by the presence of corrosive compounds in the waste gas which neces-
sitate expensive construction materials. Capital costs will vary to a lesser
extent depending upon whether the unit is a package or custom design. Addi-
tional capital expenditures will also be incurred if the system is designed
for secondary heat recovery. Figures 6-12 and 6-13 present installed capital
cost estimates for thermal and catalytic incineration systems.3 These in-
clude costs for the basic equipment as well as all auxiliary equipment such
as ducts, blowers, instrumentation, demister, piping, etc., and installation
charges.3
Annualized costs are presented for thermal and catalytic incinerators
in Figures 6-14 and 6-15. These costs include operating and maintenance
costs as well as capital-related charged. Table 6-1 shows an example calcu-
lation for determining the annualized costs for a thermal incinerator. Basis
for the calculation is given in the table. The installed capital cost for
the unit was taken from Figure 6-12. Bases for the annualized costs are
given in Table 6-2.1>3»u
6-26
-------
3
o
00
a:
o
uu
2 §
o o n
O U U
l/> QIC
«/l «
.
S £ «
§ s s;
Q. 3
« O
O •/>
LU
00
oo
o
o
00
O
o
CM
0001$
o;
cu
6-27
-------
-r
§
00
00
C£
O
o <*»
2 c
o i-«
CM
O^
o
a:
o
00
o
Q_
O
Q
00
CO
UJ
a:
e?
0001$ *
-icne.i3Uj.3Ui
6-28
-------
o
o
oo
o
00
o
«—t
o"
00
a:
o
s i
S n
o 4?
oo £.
s I
o o
LU -o
oo oo
h- r^
oo cr>
o •—
O •!-
vo
LU
o:
CD
CM rH
0001$
6-29
-------
2 c
S |
CO
o" °
S O
•^r o
3
oo
Qi
O
I—
<
CtL
t— CO
cr>
o i—
C_) I
-o
O T-
LO
\
CD
0001$ '
6-30
-------
TABLE 6-1
SAMPLE ANNUAL I ZED COST CALCULATIONS FOR THERMAL
INCINERATION (Mid-1978 Dollars)
Design Bases:
Incinerator
Operating temperature
Exchanger recovery factor
Operating time
Waste Gas
Flow rate
Heating value
Temperature
870°C (1600°F)
0.35
4,000 hrs/yr
20 NmVsec (40,000 scfm)
555 kilojoules/Nm3
(15 Btu/scf)
150°C (300°F)
COMPONENT
Operating and Maintenance
Fuel
Electricity
Labor
Maintenance
Administrative overhead
Fixed Costs
Capital recovery
Taxes, insurance, etc.
Annualized Cost
COST
$ 68,000
Neg
Neg
6,200
3,400
50,500
12,400
$140,000
6-31
-------
TABLE 6-2
ANNUAL I ZED COST BASES
OPERATING AND MAINTENANCE COSTS
FIXED COSTS
Fuel
Electricity
$2.^0/gigajoule
(2-50/MM Btu)
$0.03/kWh
Operating Labor
Direct $10/man-hour
Supervision 15% of direct
Capital Recovery
(10 yr life,
interest)
Taxes, Insurance,
etc.
16.28% of
installed cost
k% of installed
cost
Maintenance
Overhead
Plant
Payrol1
2% of installed cost
50% of labor and
maintenance
20% of labor
Source: References 1,3, and
6-32
-------
6.2 CARBON MONOXIDE BOILERS
The control of carbon monoxide emissions by oxidation in the furnace of
a boiler represents an effective and in some cases economical control tech-
nique. This method is generally applied only when the CO-containing waste
gas possesses a relatively high heating value. The following sections pre-
sent information on the equipment and design parameters, CO control effi-
ciency, applicability, energy requirements, environmental impact, and
economics of CO boilers.
6'2'1 Equipnient and Design Parameters for Carbon Monoxide
A CO boiler is essentially a typical gas-fired steam generating boiler.
A few modifications are necessary, however, due to the potential for large
variations in the concentrations of combustibles and oxygen in the CO-con-
taining waste gas. Provisions must be made so that the amount of excess
oxygen leaving the unit can be determined directly. = This may be done inter-
mittently by an Orsat or continuously by an oxygen recorder.
It is also necessary to provide for independent operation of the CO
boiler so that its operation will not interfere with that of the process or
unit which produces the CO. Water-seal tanks are installed to act as shut-
off valves. They permit the CO gases to be sent to the boiler or be passed
directly to the stack if the boiler is down.s
Supplemental fuel is required to ensure stable operation of the boiler
as well as to provide high enough temperatures in the firebox to assure com-
plete burning of the combustibles in the CO-gas stream. The following design
criteria have been established for proper operation of CO boilers:'
6-33
-------
1. supplementary firing should be capable of raising the temperature
of the CO-gas stream to over 790°C (1450°F), which is the minimum tempera-
ture needed for CO ignition.
2. the furnace temperature should be about 980°C (1800°F) for stable
operation.
3. at least two percent excess oxygen in the flue gas should be
supplied.
Sizes of the CO boilers may vary from those producing less than 23,000
kg/hr (50,000 Ib/hr) of steam to those producing greater than 230,000 kg/hr
(500,000 Ib/hr).5 The smaller units will typically be standard pre-engineered
boilers; the larger ones will be fully field-erected customized units.
6.2.2 Control Efficiency
The carbon monoxide emissions from a properly operated CO boiler should
be below 200 ppm in the flue gas. Numerous applications of CO boilers in
the refining industry have consistently achieved this level.5 Since the
concentrations of the carbon monoxide in the gases to the CO boiler are in
the range of 5 to 10 volume percent, control efficiencies of greater than
99 percent are achievable by this method.
6.2.3 Applicability
The application of CO boilers to controlling carbon monoxide emissions
from industrial sources is limited. These limitations are due to the follow-
ing reasons:
1. the fuel value of the waste gas should be sufficient so that
large quantities of supplemental fuel are not required. A plant or process
6-34
-------
will be limited in the amount of steam it can use. Fuel consumption in
excess of this for the purpose of incinerating low-heat waste gases is
expensive. Incinerators will be able to provide adequate control of these
gases at substantially lower costs.
2. the waste gas should be free of species that will foul, attack, or
deposit upon boiler internals. Sodium salts, unsaturated aromatics, potas-
sium, vanadium, halogenated compounds, and phosphorous all can result in
expensive construction materials, high maintenance, and formation of plumes.
3. the waste gas source must be able to operate independent of the
CO boiler.
However, there are several industrial processes which have had CO
boilers applied to controlling their waste gases. These include petroleum
refining fluid catalytic cracker regenerators, fluid cokers, and carbon
black plants. These applications are discussed in Chapter 7.
6-2-4 Energy Requirements
Control of CO emissions by CO boilers will result in an energy savings
or credit rather than a penalty. The magnitude of the credit will depend
directly upon the temperature and combustibles content of the waste gas
Assuming a boiler efficiency of approximately 75 percent, then 75 percent
of the heat content of the waste gas can be recovered in the steam produced.
6-2-5 Environmental Impact
The operation of a CO boiler will result in about the same environmental
lmpacts as a regular boiler. Increased S02 emissions will originate from
the sulfur contained in the supplemental fuel and increased NOX emissions
6-35
-------
will result from any nitrogen compounds in the waste gas as well as thermal
fixation of nitrogen contained in the combustion air to the boiler.
6.2.6 Costs (Mid-1978 Dollars)
The installed equipment and annualized costs for carbon monoxide boilers
are presented in Figures 6-16 and 6-17, respectively. The installed equip-
ment costs are based on information provided by a manufacturer of CO boil-
ers.5 The cost curves reflect data for three separate types of units. For
steam flows up to 12.6 kg/sec (100,000 Ib/hr), the unit would be a standard
pre_engjneered boiler with combustor. For a steam flow range of 12.6 to
27.7 kg/sec (100,000 to 220,000 Ib/hr), the unit would be a customized pre-
engineered boiler. Above this capacity, the unit would be a fully field
erected customized boiler.
The annualized costs for CO boilers were developed according to EPA
factors as shown in Table 6-2.1>3»l+'5 According to Figure 3-19, annualized
costs decrease (i.e., a net savings is realized) as the steam flow rate
increases. This savings results from the steam credit figured into the
costs. Although the graph does not show it, at very low steam rates (cor-
responding to relatively small CO boilers) the annual costs are expected to
be positive.
6.3 FLARES AND PLUME BURNERS
Flares and plume burners are devices which thermally incinerate waste
gases, in this case carbon monoxide, with no recovery of heat. The primary
distinction between a flare and a plume burner is the amount of supplemental
fuel necessary to maintain combustion. A flare requires some degree of sup-
plemental fuel for continued operation, while a plume burner is completely
6-36
-------
6.0
300 400
(103lb/hr)
Steam Flow
Source: Reference 5
FIGURE 6-16. INSTALLED EQUIPMENT COST FOR CARBON MONOXIDE BOILERS
(mid-1978 dollars)
6-37
-------
o
(1.0)
(2.0) -
(3.0)
(4.0)
« (5.0) .
(6.0) .
(7.0)
13
25
38 51
kg/sec
63
76
100 200 300 400
(10 Ib/hr)
Steam Flow
Source: Reference 1, 3, 4, 5,
500
600
FIGURE 6-17. ANNUAL COSTS FOR CARBON MONOXIDE BOILERS
(mid-1978 dollars)
6-38
-------
self-supporting. In the past flares and plume burners have been most com-
monly used as safety devices to incinerate waste gases from petroleum refin-
ing and petrochemical manufacturing operations. More recently other industries,
such as carbon black manufacturing, have also been using flares and plume
burners for disposing of waste gases.
The effectiveness of flares or plume burners for reduction of CO emis-
sions is uncertain because there is no data on emission control.6 This is
because the combustion gases are discharged into the atmosphere making it
difficult to sample the unconfined gases.6
6-39
-------
REFERENCES FOR CHAPTER 6
1. Rolke, R.W., et al. Afterburner Systems Study. EPA-R2-72-062,
PB 212560, U.S. Environmental Protection Agency, Research Triangle
Park, North Carolina, 1972.
2. Control Techniques for Volatile Organic Emissions from Stationary
Sources, Final Report. EPA-450/2-78-022, U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina, May 1978.
3. Capital and Operating Costs of Selected Air Pollution Control Systems.
EPA 450/3-76-014, U.S. Environmental Protection Agency, Research
Triangle Park, North Carolina, May 1976.
4. Control Techniques for Lead Air Emissions. EPA 450/2-77-012, U.S.
Environmental Protection Agency, Research Triangle Park, North Carolina,
December 1977.
5. Babcock & Wilcox. Steam: Its Generation and Use, 38th edition. New
York, 1972.
6. Flare Systems Study, PB-251-664, U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina, March 1976.
6-40
-------
7. INDUSTRIAL PROCESS SOURCE CONTROL
Carbon monoxide emissions and controls for industrial process sources
are discussed in this chapter. The industrial sources chosen for character-
ization include carbon black production, charcoal manufacture, the organic
chemical industry, the iron and steel industry, petroleum refining, primary
aluminum smelting, and the pulp and paper industry. Process descriptions
are given in enough detail to'indicate where emissions are produced, and
emission quantities are estimated for each source. Currently applied con-
trol technology and feasible control methods are discussed, as are control
efficiencies, energy requirements, costs, and environmental impact.
7.1 CARBON BLACK INDUSTRY
Carbon black is produced by the partial oxidation of hydrocarbons in a
limited supply of air. The primary use of carbon black is in the production
of rubber where it acts as a reinforcing agent. Currently about 95 percent
of all carbon black produced in the U.S. is used by the rubber industry.1
It is also used as a colorant for printing ink, paint, paper, and plastics.1
The most recent estimates available indicate that in 1977 about 2.2
million metric tons (2.4 million tons) of carbon monoxide were emitted from
carbon black production in the U.S.2 The following sections include a brief
7-1
-------
process description of carbon black production and an assessment of carbon
monoxide control technology for the carbon black industry.
7.1.1 Process Description
There are three basic processes used in the United States for the pro-
duction of carbon black. They are: the furnace process, the channel pro-
cess, and the thermal process. Production from the furnace process
accounts for about 90 percent of the total tonnage of carbon black pro-
duced.3 Almost 10 percent is produced from the thermal process and less
than 0.1 percent from the channel process.3
Thermal process plants use a relatively clean feedstock and can recycle
almost all of the off-gas to reactors to recover the energy in the gas.
Because recycle is a part of the thermal process, carbon monoxide emissions
from this process are insignificant.4'5
In 1974, only one plant producing carbon black via the channel process
was still in operation, and it was subject to a court order requiring
gradual closure by 1979.3 Because this process has been almost totally
phased out, carbon monoxide emissions from it are not discussed in this
document.
In the furnace process, mixed feeds of a light hydrocarbon gas and a
heavy oil are used in most plants. The best oil to use for the production
of modern high structure carbon blacks is highly aromatic, low in sulfur,
contains high molecular weight resins and asphaltenes, and is substantially
free of suspended ash and water.1 The mixed feed is preheated and injected
with a limited supply of combustion air into the reactor or furnace.
7-2
-------
Internal reactor temperatures vary from 1300-1700°C (2400-3100*F),
depending on the grade of carbon black being produced.6
The flue gases and entrained carbon from the reactor are cooled to 540°C
(1000°F) by heat exchange with the furnace feed and sent to a water quench
tower.6 The carbon black laden gas stream is then sent to a fabric filter
unit for product recovery. The gaseous effluent contains approximately 50
percent water vapor and 35 percent nitrogen. The remaining 15 percent is
made up of CO, C02, and H2, with small amounts of methane and acetylene.6
The recovered carbon black is sent to a small collecting cyclone and
is then fed to a micropulverizer to break up any hard agglomerates present.
The pulverized carbon black is sent to a finishing area where final process-
ing yields a pelletized or bead product. Figure 7-1 is a simplified flow
diagram for the furnace type carbon black process.
7.1.2 Process Emission Sources and Factors
In the furnace process, the gas stream containing the carbon black also
contains significant quantities of carbon monoxide. After the carbon black
has been removed, this stream is usually discharged to the atmosphere
through a vent stack. This vent is the source of the carbon monoxide emis-
sions from the furnace process. Table 7-1 contains a representative vent
gas composition.1 Actual vent gas composition can vary considerably from
the average figures shown, depending primarily upon the grade of carbon
black being produced. CO emissions tend to be higher for small-particle
carbon black production.1
The uncontrolled carbon monoxide emission factor for the furnace pro-
cess as reported by EPA is 1300 kilograms/metric ton of black produced
7-3
-------
-
CXL
01
O
-------
TABLE 7-1
TYPICAL VENT GAS COMPOSITION FOR CARBON
BLACK FURNACE OIL PROCESS
RANGE IN COMPOSITION TYPICAL COMPOSITION
COMPONENT MOLE % MOLE %
Hydrogen 5.5 - 15 5.7
Carbon Dioxide 3-6.5 2.5
Carbon Monoxide 6-14 5.5
Hydrogen Sulfide 0.01 - 0.2 0.1
Methane 0.2 - 0.7 0.2
Acetylene 0.1 - 1.0 0.2
Nitrogen & Argon 65 - 80 35-5
Oxygen 0-4.9 0.3
Nitrogen Oxides (N02) 15 - 200 ppm^ 44 ppm
Water (b) 49.0
High values represent values from two plants. Most producers
believe actual value is toward low end of range shown.
Dry basis. Stream typically contains 42-50 mole % water.
Source: Reference 1
7-5
-------
(2600 lb/ton).5 With a CO boiler or thermal incinerator for a control
device, the reported emission factor is 5 kg/metric ton (10 Ib/ton). These
represent efficiencies of over 99 percent for the boiler or incinerator.
There are no significant CO emissions from the thermal process.4'5
7.1.3 Control Techniques
In 1976, 29 carbon black plants were operating in the U.S. Of that
number, three were equipped with CO boilers, two had thermal incinerators,
and two were equipped with flares.7 At that time, three additional plants
were installing CO boilers.7 Besides these control techniques, catalytic
incineration represents a feasible but undemonstrated control alternative.
The remainder of this section discusses the application of these control
techniques to carbon black plants.
7.1.3.1 CO Boilers
The CO boiler is one method of controlling combustible gaseous emis-
sions. However, the addition of a CO boiler to any existing carbon black
plant would be costly. This is because most plants have electric-powered
motors to drive their equipment and thus cannot use the generated steam
without a large expenditure for turbine drivers.
Approximately 50 to 60 percent of the steam generated by off-gas com-
bustion in a new carbon black plant can be used in the process to drive
steam turbines and to supply steam for other uses.7 Consequently, CO
boilers are not used to generate steam beyond this level, unless other out-
lets for the steam are available.
To ensure complete combustion of the CO in the vent gas, boilers are
normally designed for combustion zone operating temperatures of 870-980°C
7-6
-------
(1600-1800°F). 1 If all the energy in the vent gas is not needed for steam
production, a CO control system can be installed. In typical systems,
part of the vent gas is used as CO boiler fuel and the remaining portion
is sent to a thermal incinerator or flare. The excess gas may also be used
as fuel for drying carbon pellets.
Testing at two carbon black plants found the carbon monoxide emissions
from CO boilers ranged from 0.001 to 0.005 kg/metric ton of carbon black
(0.002-0.010 lb/ton).7
Additional problems associated with the application of CO boilers to
carbon black plants include:1
a) The vent gas is at low pressure, and has a high water vapor content.
b) The gas stream is corrosive.
c) Up to 35 percent of the total heating value of the gas burned
in the boiler must be added as supplemental fuel in order to achieve complete
combustion.
d) Flameouts causing safety problems are possible due to flame control
difficulty.
e) A dependable steam supply may require a spare boiler.
f) Frequently, the type of carbon black produced is changed. This
requires the complete plant system to be purged. During this time total
supplemental fuel firing of the boiler is necessary because the vent gas
has no heating value. Bringing the boiler back on line when the new type
of carbon black is first being produced is difficult.
7-7
-------
7.1.3.2 Flares
As of 1976 two domestic carbon black manufacturers reported the use
of vent gas flares.7 According to the operator of one plant, the minimum
self-supporting heating value is about 1.87 megajoules/Nm3 (50 Btu/scf).1
The typical heating value for carbon black vent gas has been reported to
be only 1.49 megajoules/Nm3 (40 Btu/scf).1
With respect to their application to carbon black plants, flares have
the following limitations:1
a) The burner could be extinguished due to relatively small changes
in the vent gas composition if supplemental fuel and adequate instrumenta-
tion are not provided.
b) The CO control effectiveness of a flare cannot be measured accurately
because it is necessary to sample and measure gas flow after the gases
leave the stack outlet and mix with ambient air. (See Section 6.3.)
7.1.3.3 Thermal Incinerator
A thermal incinerator which utilizes heat recovery by preheating the
air and vent gas in a heat exchanger with the products of combustion will
not require supplemental fuel.1 This is true for virtually all carbon black
plants.
To achieve adequate oxidation of the carbon monoxide, the combustion
zone temperature should be between 870-980°C (1600-1800°F).1 These
temperatures should result in CO removal efficiencies of over 95 percent.
In 1976, two U.S. plants were known to use thermal incineration for
control of the vent gas emissions.7 At one plant, more excess air is used
7-8
-------
than is normal (7 vs. 4 mole % 02) and there is no air preheat. For these
reasons, supplemental fuel is required for this unit.
7.1.3.4 Pellet Dryers
Vent gas may also be used as a fuel for drying carbon pellets. However,
supplemental fuel would be necessary to maintain combustion. Carbon monoxide
concentrations of less than 10 ppm have been measured from the exhaust of a
pellet dryer using vent gas as a fuel.7
7.1.3.5 Catalytic Incinerator
As of 1976 no carbon black plants were using catalytic incineration to
burn their process vent gases.7 However, it has been reported that one
attempt was abandoned some years ago because of catalyst poisoning.1 If a
catalyst is used that is not poisoned by sulfur, and adequate control instru-
mentation is employed to prevent high bed temperatures, it should be possible
to use a catalytic incinerator.1
A 490°C (900°F) inlet temperature to the catalytic bed should be suf-
ficient to oxidize almost all carbon monoxide in the vent gas.1 Maximum
temperature within the bed should be limited to 650°C (1200°F) in order to
prevent damage to the catalyst and a resulting loss in catalyst activity.
As the catalyst ages, though, its combustion efficiency will gradually
decrease due to a loss in activity. At the time when excessive pollutant
concentrations begin to be discharged from the incinerator, the catalyst
bed must be replaced.
7.1.4 Cost of Controls
Chapter 6 contains a more detailed presentation of capital and annualized
costs for the carbon monoxide control techniques described above. Both types
7-9
-------
of costs are presented graphically in terms of dollars per normal cubic
meter per second ($/scfm) with several curves per graph showing the effect
of the heating value of the gas being oxidized on the annualized costs.
Therefore, given a representative flow rate and heating value for the vent
gas from a carbon black plant, various control costs can be estimated.
As mentioned earlier, the heating value of the vent gas is typically
1.49 megajoules per normal cubic meter (40 Btu/scf). A representative vent
gas flow rate for a 41,000 metric tons/yr (45,000 tons/yr) carbon black
plant is about 27 Nm3/sec (57,400 scfm).1 This corresponds to a vent gas
flow rate of approximately 20.9 x 103 Nm3 per metric ton (0.67 x 106 scf/
ton) of carbon black produced. l
7.1.5 Impact of Controls
The following presents information on the impact of applying the con-
trol techniques discussed earlier to the vent gas stream from carbon black
production. Potential reductions in carbon monoxide emissions, environ-
mental impact, and energy requirements for each of the controls are
addressed.
7.1.5.1 Emissions Reduction
The main process vent is the primary source of carbon monoxide emissions
from carbon black plants. As of 1976, 25 percent of the plants (7 plants)
employed controls for this source.7 Assuming retrofit of the remainder of
the plants with control devices such as a CO boiler or incinerator with a
CO control efficiency of 99 percent, annual carbon monoxide emissions could
be reduced from this industry by about 2.18 x 106 metric tons (2.38 x 106
tons).
7-10
-------
7.1.5.2 Environment
The application of controls for the CO emissions from carbon black
plants will have both positive and negative impacts with respect to other
pollutant discharges. The positive impact will include the oxidation of
the combustible components other than CO in the vent gas (hydrogen, H2S,
methane, acetylene and most of the particulate carbon black which pene-
trated the fabric filters).6
The negative impact will include increased emissions of NO from all
the control techniques described, conversion of some S02 in the gas to S03
in the catalytic incinerators if noble metals are used, and increased S02
emissions if oil is used as a supplemental fuel in the oxidation systems.
Increased NOX emissions will depend on the operating temperature of
the oxidation system being used. Reported increase in the NO levels in the
X
vent gas after being oxidized in a thermal incinerator is about 4.8 grams
of N0x per Nm3 of vent gas (3 lb/10,000 scf).1 For catalytic incinera-
tion it is about 1.6 grams per Nm3 (1 lb/10,000 scf).1 Because CO
boilers perform the function of providing plant energy as well as pollution
reduction, no incremental emissions are attributed to CO boilers. Without
the CO boiler, energy would have to be generated elsewhere and purchased
by the plant. This would result in roughly equivalent quantities of combus-
tion-related pollutant emissions.
7.1.5.3 Energy Requirements
The energy requirements associated with the application of CO controls
to carbon black plants will vary significantly from plant to plant due to
the variations in vent gas composition and heating value. Assuming a typical
7-11
-------
vent gas heating value of 1.49 megajoules/Nm3 (40 Btu/scf), energy require-
ments for the various controls can be calculated.
Thermal and catalytic incinerators with heat recovery designs will
require no supplemental fuel. A CO boiler fueled with vent gas with a heat-
ing value of 1.49 megajoules per Mm3 (40 Btu/scf) will require approximately
7.9 megajoules of supplemental fuel per kilogram of carbon black (3,400 Btu/
Ib) produced at the plant.1 If the vent gas is sent to a flare stack for
oxidation of the CO, approximately 42.8 megajoules of supplemental fuel per
kilogram of carbon black (18,400 Btu/lb) will be required.1 These calcula-
tions are based on an average vent gas flow of 19 Nm3/kg of carbon black
(300 scf/lb).1
7.2 CHARCOAL INDUSTRY
Charcoal is manufactured by the pyrolysis (carbonization or destructive
distillation) of carbon-containing materials. Raw materials can be almost
any carbon-containing material but are principally medium to dense hardwoods
such as beech, birch, hard maple, hickory, and oaks. Wood charcoal is used
primarily as a recreational cooking fuel.
The most recent national emission estimates indicate that in 1977 about
97,300 metric tons (107,200 tons) of carbon monoxide were emitted from char-
f\
coal manufacturing/ Calculations based on these numbers and the uncontrolled
carbon monoxide emission factors for charcoal manufacturing indicate that
more than seventy percent of U.S. charcoal plant production has no carbon
monoxide emission controls. The following sections include a brief process
description, identification of charcoal plant carbon monoxide emission
7-12
-------
sources, and an assessment of carbon monoxide control technology for the
charcoal industry.
7.2.1 Process Description
Two basic processes exist in the charcoal manufacturing industry:
batch kilns and continuous multiple-hearth furnaces. Of the total yearly
production of charcoal in 1975, approximately 55 percent was produced by
the continuous process and 45 percent by the batch process.8 Because the
two differ significantly, two process descriptions are given.
7.2.1.1 Batch Process
The present day batch process incorporates two types of charcoal kilns.
The most widely used is the Missouri type shown in Figure 7-2.
The Missouri type kiln is usually constructed of concrete, typically
processing 45 to 50 cords of wood per cycle. A cycle includes loading the
kiln, carbonizing the wood, allowing the charcoal to cool, and unloading the
kiln. Time requirements for each component of the cycle differ greatly from
plant to plant; however, the overall time period involved in a normal cycle
is about 6 to 25 days.9
Once started, maintaining proper conditions in the kiln is the primary
requirement for satisfactory carbonization. Sufficient heat must be gene-
rated to first dry the wood and then to maintain temperatures necessary for
efficient carbonization. Combustion of a part of the wood volatiles gene-
rates the heat to sustain the carbonization process. By varying the size
of the air port openings providing air for the combustion of these wood
volatiles, control of the kiln temperature is achieved. Kiln temperatures
7-13
-------
00
o
I—
oo
I—
oo
o
CJ
CJ
I
I — i
o;
^>
o
oo
oo
DL.
>-
C\J
I
o:
7-14
-------
of from about 840-950°C (1540-1740°F) are required for the production
of good quality charcoal.
The second type of batch kiln used presently is the beehive kiln which
is shown in Figure 7-3. This kiln is usually constructed of concrete and
consists of a cylindrical wall with a dome-shaped ceiling. The kiln struc-
ture includes ground-level air and mid-level exhaust ports located around
the periphery of the wall, a steel door in the side of the wall for loading
and unloading, and an opening in the dome-shaped ceiling for loading and
firing. Beehive kilns typically process 50 to 90 cords of wood per cycle.
The time period involved in a normal cycle is about 10 to 20 days.
7.2.1.2 Continuous Process
The Herreshoff multiple hearth furnace is the predominant continuous
charcoal process in use today. This process is gaining a larger share of
the total charcoal production yearly.9
The Herreshoff multiple hearth furnace consists of several hearths or
burning chambers stacked one on top of the other as shown in Figures 7-4
and 7-5. The hearths are contained in a cylindrical, steel, refractory-
lined shell, and are divided by refractory decks which function as the floor
of one hearth and the roof of the hearth below. Passing up through the
center of the furnace is a shaft to which two or four rabble arms per hearth
are attached. As the shaft turns (usually 1 to 2 rpm), the hogged (chipped)
material resting on the hearth floors is continually agitated, exposing
fresh material to the hot gases being evolved. Another function of the rabble
arms is to move material through the furnace. On alternate hearths the teeth
7-15
-------
EXHAUST PORTS
AIR PORTS
FIGURE 7-3. TYPICAL BEEHIVE KILN
7-16
-------
FIGURE 7-4. EXTERIOR VIEW OF A HERRESHOFF MULTIPLE HEARTH FURNACE
7-17
-------
EXHAUST GASES
GAS COMBUSTION
RABBLE ARMS
FURNACE SHELL
AND HEARTHS
PRODUCT CHARCOAL
COOLING AND COMBUSTION AIR
FEED
MATERIAL
FIGURE 7-5. CROSS SECTIONAL VIEW OF A HERRESHOFF MULTIPLE HEARTH
FURNACE, WITH PLUME BURNING
7-18
-------
are canted to spiral the material from the shaft toward the outside wall
of the furnace or from the outside wall toward the center shaft. Around
the center shaft is an annular space through which material drops on alter-
nate hearths, while on the remaining hearths material drops through holes
in the outer periphery of the hearth floor. In this way, material fed at
the top of the furnace moves alternately across the hearths at increasing
temperatures until it discharges from the floor of the bottom hearth.
Furnace temperatures range between 450°C and 650°C (840°F and 1200°F).
All off-gases exit from above the top hearth. These gases are either
flared directly to the atmosphere through stacks located on top of the fur-
nace as shown in Figure 7-5, or they may be further processed to use the
available heat for predrying the incoming feed material, drying briquettes
produced at an adjacent briquetting plant, or for producing steam in an
adjacent waste heat steam boiler.
Multiple hearth furnaces require a large and steady source of raw
materials. This limits their use to areas where many small or a few large
sawmills and other wood waste producers are located. This criteria also
eliminates the chance of replacing all batch-type processes with multiple
hearth furnaces since most batch-type plants as well as their raw material
sources are located in isolated areas.
7.2.2 Process Emission Sources and Factors
Large amounts of carbon monoxide are formed by the partial oxidation
mechanisms within both batch kilns and continuous operating furnaces. An
emission factor of 160 kilograms of CO per metric ton charcoal (320 Ib/ton)
7-19
-------
has been reported for carbon monoxide emissions from charcoal production.5
No distinction is made between carbon monoxide emission rates from batch
and continuous processes. However, it is reasonable to expect that due to
the higher operating temperatures in the continuous process, there will be
lower carbon monoxide emissions per unit weight of charcoal produced.
7.2.3 Control Techniques
Conditions for CO control on batch processes are different than for
continuous processes. Gas or oil-fired thermal incinerators are the only
methods employed for control of emissions from batch process plants. Most
batch process plant emissions are uncontrolled. CO emissions from continuous
process plants can be controlled with thermal incinerators. At times flares
are used. All continuous process plants employ either one or a combination
of these two methods. Applications of controls to batch and continuous
processes are discussed separately below.
7.2.3.1 Control of Batch Processes
Control of emissions from batch charcoal kilns is difficult due to
the cyclic nature of the process and, as a result, the cyclic nature of the
emissions. During the carbonization cycle, both the emission composition
and discharge rate vary. Typically, emission rates peak early in the cycle
at a flow rate over 40 percent greater than the flow rate near the end of
the cycle.8 Variations in the type of feed material, the moisture content
of the feed material, and the operating practice also influence emission
composition and rate.
7-20
-------
A direct fired thermal incinerator is the only method used to control
emissions from batch kilns. Afterburner temperatures of over 750°C (1400°F)
with a residence time of 0.2 to 0.4 seconds are required to achieve efficient
oxidation of carbon monoxide.10
Existing control systems have been designed primarily to reduce visible
emissions (particulates and hydrocarbons) instead of CO. In a typical opera-
tion, each incinerator, directly fired with natural gas or oil, services two
or more kilns. A temperature of about 650°C (1200°F) is maintained in the
incinerator during the kiln burn by automatic controls which cycle the fuel
fed to the afterburner on and off. The afterburner is then shut down as
soon as the kiln burn is complete.9 To provide at least 90 percent effi-
cient CO control, these systems would have to be modified to operate at a
temperature of about 980°C (1800°F).9 This modification would increase the
requirement for supplementary fuel and perhaps require incinerator redesign.
Problems associated with the application of incinerator systems to
batch kilns include the following:
a) The design and operation of batch kilns must be modified to
accommodate the application of incinerators. The multiple exhaust pipes
or ports (as shown in Figures 7-2 and 7-3) must be converted to one large
exhaust manifold. Because of this requirement, applications to beehive
kilns would be costly.
b) Plants with kilns mounted far apart or on unlevel land must
install long lengths of costly ductwork to connect the kilns to the
incinerators.
7-21
-------
c) The control systems consume large quantities of supplementary fuel
during periods of the kiln burn.
7.2.3.2 Control of Continuous Processes
Herreshoff furnaces generate an off-gas with a relatively constant
composition and flow rate. As a result, control of emissions is easier
with the Herreshoff furnaces than with Missouri kilns.
The furnace off-gas can be burned in refractory-lined stacks on top
of the furnace by admitting combustion air through adjustable doors in the
base of the stack, as shown in Figure 7-5.8>n With dry feed the heating
value of the off-gas is sufficient to maintain temperatures ranging from
750 to 850°C (1400-1550°F).12 The most efficiently controlled plants are
equipped with wood dryers for removing free moisture prior to the Herreshoff
furnace. Table 7-2 shows off-gas characteristics for a plant equipped with
a wood dryer. The off-gases from the Herreshoff furnace are used for both
wood and briquette drying. The remainder of the off-gas is discharged from
the furnace stack. The accuracy of the CO emission data shown are poor
because Orsat CO analyses are inaccurate at low CO concentrations. If a
continuous type plant is not equipped with a wood dryer, it would be neces-
sary to apply an afterburner to achieve the outlet temperatures shown in
Table 7-2.
7.2.4 Cost of Controls
The control technique identified for batch and continuous processing
charcoal kilns was thermal incineration. Chapter 6 contains a detailed
presentation of capital and annualized costs for this control technique.
7-22
-------
TABLE 7-2
CHARACTERISTICS OF OFF-GASES FROM A
HERRESHOFF FURNACE CHARCOAL PLANT
Capacity: 1.9 Kilograms of Dry Wood per
Second (7.5 tons/hr)a
CO-PPM
Stack gas volume
Actual cubic meters
per second
(ACFM)
Stack gas velocity
Actual meters per
second
(AFPM)
Stack diameter
Meters
(Inches)
FURNACE
STACK
3000-5000
6.0-9.9
750-850
(1400-1550)
WOOD
DRYER
STACK
9800
16.2
65
(150)
BRIQUETTING
MACHINE
STACK
o
18-19.7
72
(162)
49.6-73.1 11.0-12.0
(105,000-155,000) (23,200-25,400)
3.28
(129)
0.86C
(34)
Feed free moisture-50 percent by weight
Orsat analysis
'Dimensions of one of two stacks
Source: Reference 12
10.6
(22,500)
5.9-8.7
(1160-1700)
18.7-20.5
(3700-4000)
13.8
(2700)
1.17x0.66°
(46 x 26)
7-23
-------
To accurately determine the costs of applying these controls to batch
type charcoal plants requires data on the off-gas flow rates and composi-
tion. No information was found from which flow rates of batch type charcoal
kiln off-gases could be calculated.
7.2.5 Impact of Controls
The following discusses potential reductions in carbon monoxide emis-
sions from the carbon monoxide control techniques identified in Section
7.2.3, as well as the environmental impacts and energy requirements of
these controls.
7.2.5.1 Emission Reductions
The current level of control of charcoal kiln off-gases is unknown.
Consequently, even an approximation of the potential reduction of CO
emissions from the charcoal industry cannot be made.
7.2.5.2 Environment
The application of controls on charcoal plants for CO will result in
the oxidation and control of virtually all hydrocarbons in the gas as well
as most of the combustible particulates.
The operation of these controls, though, will result in an increase
in NO emissions. However, this increase is not expected to be substantial
if the flame temperatures are kept below 980°C (1800°F).
7.2.5.3 Energy Requirements
The application of CO controls to batch kilns will require fuel. Sup-
plementary fuel requirements vary depending on the moisture content of the
raw material used, the type of fuel used in the afterburner, climatic factors,
7-24
-------
and operating methods. Fuel oil consumption has been reported to average
about 3.3 megajoules/kilogram of char (2.8 x 106 Btu/ton of char) during
the summer and about 6.6 megajoules/kilogram of char (5.7 x 106 Btu/ton
of char) during the winter.9
The off-gases from continuous charcoal kilns are of a high enough heat
content so that no supplementary fuel is required for their oxidation. The
recoverable heat content of the gas is about 29 megajoules per kilogram of
charcoal produced (25 million Btu/ton). "'I* This heat can be used to pre-
dry raw material fed to the carbonizer or for briquette-drying.
7.3 ORGANIC CHEMICAL INDUSTRY
Substantial amounts of carbon monoxide are emitted from organic chemical
processes, which partially oxidize hydrocarbons derived primarily from petro-
leum, coal, and natural gas into organic intermediates and products. The
processes producing the largest amounts of CO are acrylonitrile, formal-
dehyde, maleic anhydride, and phthalic anhydride production. CO is also
produced from incineration of unmarketable by-products.
Mass carbon monoxide emissions from these processes are shown in Table
7-3. Smaller amounts of carbon monoxide are emitted from many other organic
chemical processes which are not discussed in this report.
Carbon monoxide emissions from the four organic chemical processes
discussed in this section comprised about 76 percent of 1977 CO emissions
from the U.S. petrochemical industry, 4.4 percent of the CO emitted from
U.S. industrial processes, and 2.1 percent of the total amount emitted in
the U.S. from stationary sources.2
7-25
-------
TABLE 7-3
MASS EMISSION ESTIMATES FOR CARBON MONOXIDE FROM FOUR ORGANIC
CHEMICAL PROCESSES, 1977
CARBON MONOXIDE EMISSIONS
SOURCE
Acrylonitrile production
Formaldehyde production
Maleic anhydride production
Phthalic anhydride production
TOTAL
Metric tons
130,400
64,900
117,800
50,900
364,000
Tons
143,700
71,500
129,900
56,100
401 ,200
Source: Reference 2
7.3.1 Acrylonltrile
Acrylonitrile is an important feedstock in the production of synthetic
fibers and in the treatment of natural fibers to improve their properties.
Acrylonitrile is also used extensively in the production of low cost, multi-
purpose plastics, barrier resins, and nitrile rubber.
1977 EPA estimates indicate that 130,400 metric tons (143,700 tons)
of carbon monoxide were emitted in the United States.2 The extent and type
of emission control varies widely within the industry.
7.3.1.1 Process Description
Acrylonitrile is produced in the U.S. by the Sohio fluid bed catalytic
process. Figure 7-6 is a simplified flow sheet of the process. Air,
ammonia, and propylene are fed to a reactor at 140-310 kilopascals (5-30
psig) and 420-530°C (780-980°F) to form acrylonitrile. The chemical reaction
is shown in the equation below.
2 CH2 = CH - CH3 + 2 NH3 + 302-> 2 CH2 = CH - CN + 6 H20
7-26
-------
Product
Aurvl on i. f- T"i
0)
rH ,
•H
4J
•H
o
rH
£J
CU
TJ
—
CO
ca
i
o c
OJ
c >
•H
O
CJ
4
1
1
I
!
%
k
(
X
& ^
i
r
rH
CU 4-1
a -H
3 C
^ , -..,-,
J 4J '
CU
|
j
^ / ?
w I
f
1
1
1
1
>k
A
.
i
S ' cu
„
^^
^ ^
( —
£
M
•H!
M
1>
H
J
H '
1-1
1
A
_i
O
B
,
A
a
n) d
H r— I
c >•
I §
rH
O
U
4J
a
TJ
O
Pi
P-I
co
'SH
C O
W o
•H
M
4J
•H
C
O
4-1
•H O
C O
o
-------
No recycle is required, as the reaction is virtually complete. Reaction
products are recovered in a water absorber-stripper system. Acrylonitrile
is then separated from by-products in a series of distillations. The first
fractionation of crude acrylonitrile usually removes HCN as an overhead
stream. The acrylonitrile is then purified to 99+ percent in further dis-
tillation steps. The wet acetonitrile by-product is subjected to extrac-
tive distillation using water as the extractive solvent.
By-product streams may be processed to recover high purity HCN and
acetonitrile for sale. The by-product streams which are not sold are
incinerated. Currently, two acrylonitrile producers market acetonitrile.15
All of the producers market HCN. Fifty percent of the HCN is sold, and the
remaining 50 percent is incinerated or disposed of in deep wells.16
There have recently been two ammoxidation catalysts in use: Catalyst
21 and Catalyst 41. Although the yields are about the same for the two
catalyst systems, Catalyst 41 provides for better utilization of ammonia
and requires less oxygen. All U.S. acrylonitrile producers have switched
to Catalyst 41.16
7.3.1.2 Process Emission Sources and Factors
The major source of CO emissions within acrylonitrile plants is the
main process vent, which vents from the absorber. Currently, three acryloni
trile plants out of six in the U.S. apply CO emission control technology to
emissions from their main process vents.7 Absorber vent gas composition is
affected by catalyst type, reactor operating conditions, absorber overhead
temperature, reactor feed rates, and feed material composition. Catalyst
7-28
-------
type can especially influence CO emission rate. Prior to 1973, uncontrolled
CO emissions from processes using Catalyst 21, a uranium-based catalyst,
were estimated at 0.178 kg/kg (0.178 Ib/lb) acrylonitrile. 'V7 When manu-
facturers switched to Catalyst 41, a bismuth phosphomolybdate catalyst,
emission factors were reduced to 0.079 kg/kg (0.079 Ib/lb) acrylonitrile.17
Controlled emission factors were derived using reported control device
efficiencies and uncontrolled emission factors.^,17 Both thermal and cata-
lytic incinerators used as CO control devices have reported CO removal
efficiencies of greater than 95 percent.is.i? when this factor was applied
to uncontrolled emission rates, controlled CO emissions from the main pro-
cess vent were estimated to be less than 0.004 kg/kg (0.004 Ib/lb) acryloni-
trile when Catalyst 41 was used.
7-3.1.3 Control Techniques
Three U.S. acrylonitrile plants currently control CO emissions from
their main process vents.7 All use combustion devices (i.e., a catalytic
incinerator or a thermal incinerator) to reduce emissions.7 These two types
of demonstrated controls are discussed in the following paragraphs.
Thermal Incinerators - A schematic diagram of one of the three thermal
incinerators currently used in U.S. acrylonitrile plants is shown in Figure
7-7.17 This device is used for combustion of by-product acetonitrile and
hydrogen cyanide as well as main process vent gas. The incinerator operates
at 870°C (1600°F) and reportedly achieves >95 percent combustion of CO in
the vent gas. Natural gas is used as a supplemental fuel because of the
relatively low heating value of the vent gas (0.75-1.49 megajoules/m3
[20-40 Btu/ft3]).17
7-29
-------
HYDROGEN CYANIDE
ACETONITRILE
COMBUSTION AIR
ABSORBER VENT GAS
NATURAL GAS
870°C
FIGURE 7-7 SCHEMATIC DIAGRAM FOR A COMBINATION BY-PRODUCT INCINERATOR/
ABSORBER VENT GAS THERMAL OXIDIZER SYSTEM
7-30
-------
A similar thermal incinerator could be used to control CO emissions
from the main process vent only. Typical main process vent gas composi-
tion is shown in Table 7-4. Operating temperatures range up to 980°C
(1800°F); more complete combustion can be achieved with higher temperatures,
but N0x emissions increase rapidly at temperatures above 980°C (1800°F).
Catalytic Incinerators — One U.S. acrylonitrile producer uses a
catalytic incinerator to oxidize off-gas from its main process vent.17
Operating parameters for this device have not been reported, but typical
catalytic incinerators operate at temperatures ranging from 480-650°C
(900-1200°F).16 The effectiveness of this catalytic incinerator for
reducing CO is not reported. The effectiveness of the unit for reducing
hydrocarbon emissions is reported to be 42.5 percent.7
Because of their lower operating temperatures, catalytic incinerators
use less supplemental fuel and tend to emit lower levels of NO than thermal
X
incinerators. Their principal drawbacks are the moderate length of catalyst
life, the tendency toward catalyst poisoning by off-gas components, and
their increased operating and maintenance costs.
7.3.1.4 Cost of Controls
A detailed presentation of annualized costs for the above-mentioned
carbon monoxide controls is given in Chapter 6. The following describes
how this information can be applied to estimate the costs for controlling
carbon monoxide emissions from acrylonitrile production. This can best be
accomplished by taking a model plant and describing those parameters which
will determine the annualized costs for controlling its CO emissions. These
7-31
-------
TABLE 1-k
COMPOSITION OF MAIN PROCESS VENT GAS FROM
ACRYLONITRILE PRODUCTION VIA THE SOHIO PROCESS
COMPONENT VOLUME PERCENT
Carbon dioxide 2.6
Carbon monoxide 1.5
Propylene 0.3
Propane 0.5
Hydrogen cyanide <0.1
Acryloni trile <0.1
Acetoni trile 0.1
Nitrogen 80.9
Oxygen 0.8
Water 13.3
Nitrogen oxides <0.1
Source: Reference 17
7-32
-------
parameters are vent gas flow per unit weight of acrylonitrile produced and
the energy content of the vent gas.
The model is based on a new plant producing 9.1 x 10^ metric tons (2.0
x 108 Ib/yr) of acrylonitrile. Representative flow rates for the process
vent gas from a plant this size have been reported to be 21 Nm3/sec (45,000
scfm).16 The energy content of this gas has been reported to be within a
range of from 0.75-1.49 megajoules/m3 (20-40 Btu/ft3) with 0.89 megajoules/
m3 (24 Btu/ft3) reported to be the most representative number.17 Using
this information and the information in Chapter 6, annualized costs can be
estimated for the various applicable control techniques for an individual
plant of a given size or for the entire industry.
7.3.1.5 Impact of Controls
Emissions Reduction -- The main process vent is the primary source of
carbon monoxide emissions from acrylonitrile plants. Currently, emissions
from three plants, or about 47 percent of the U.S. acrylonitrile capacity,
are reportedly controlled.16 It was calculated that application of inciner-
ators or the other feasible CO controls could result in a reduction of
annual carbon monoxide emissions of about 62,000 metric tons (68,600 tons),
if 90 percent removal efficiency were achieved.
Environment -- When incineration is used as a means of CO emission con-
trol , the amount of N0x in the incinerator flue gas increases. In general,
higher incinerator temperatures result in higher NO emissions. No data
X
were available regarding N0v formation in catalytic incinerators. NO
X X
emissions from this device should be lower than for thermal incinerators
because of the lower operating temperature.
7-33
-------
At the present time, natural gas is generally used as supplemental
incinerator fuel. If future shortages of natural gas require the use of
fuel oil as supplemental incinerator fuel, an increase in sulfur oxides
(SO) emissions would result. The magnitude of the SO emissions would
X X
depend on the sulfur level in the fuel oil and the total quantity of oil
consumed.
Energy Requirements -- Both the demonstrated and undemonstrated tech-
niques for CO emission control require the use of supplemental fuel. Energy
content of the main process vent gas ranges from 0.75-1.49 megajoules/m3
(20-40 Btu/ft3).16 The amount of supplemental fuel needed will vary with
vent gas energy content and with the type of control device used.
Table 7-5 lists the amount of energy needed for thermal incinerators
and waste heat boilers when used to control CO emissions from the main
process vent. The calculations were based on a process vent gas energy
content of 0.89 megajoules/m3 (24 Btu/ft3), from a 9.07 x 104 metric ton/yr
(2.0 x 108 Ib/yr) acrylonitrile plant.17
TABLE 7-5
ENERGY REQUIREMENTS FOR CO EMISSION CONTROLS
IN ACRYLONITRILE PRODUCTION
/ v ENERGY REQUIRED/
CONTROL DEVICE ENERGY REQUIRED^ UNIT PRODUCT
Thermal incinerator without 5.87 megajoules/sec 1.9 megajoule/kg
heat recovery or waste heat (20 x 106 Btu/hr) (800 Btu/lb)
boiler
Data from Reference 17, based on 8000 operating hours per year.
7-34
-------
No information regarding catalytic incinerator energy requirements
was available. However, this device uses less supplemental fuel than the
other devices discussed above because its operating temperatures are sub-
stantially lower.
7.3.2 Formaldehyde
Formaldehyde is manufactured by two processes. One employs a silver
catalyst and the other a mixed metal oxide catalyst. Approximately 23 per-
cent of U.S. formaldehyde capacity is based on the mixed oxide process and
77 percent is based on the silver catalyst process.*M9 Both processes
are described below.20
7.3.2.1 Process Description
The overall reaction for making formaldehyde from methanol with a silver
catalyst is shown in the following chemical equation:
2 CH3OH + % 02 -> 2 CH20 + H2 + H20
Figure 7-8 is a simplified flow diagram of the silver catalyst process.
The feedstocks are prepared before they are introduced into the
reactors. Air is washed with caustic to remove C02 and sulfur compounds
and heated to about 80°C (180°F). The treated air and vaporized methanol
are combined and sent to a battery of catalytic reactors. Some plants use
a feed vs. effluent heat exchanger as the next step. Otherwise, effluent
gases containing the formaldehyde go directly to the primary absorber for
product recovery. The sorbent is an aqueous solution of formaldehyde and
methanol, part of which is recycled back to the absorber. The other portion
7-35
-------
Q
O
O.
UJ
Q
o:
o
GO
GO
UJ
O
o
C£
Q_
GO
>-
o
C\J
O)
o
c.
-------
goes to an intermediate storage facility. Noncondensibles and uncondensed
vapors are sent to a secondary absorber for further product recovery. Dis-
tilled water is used as a sorbent. The resulting solution of formaldehyde
and methanol is used as makeup for the primary absorber. Noncondensibles
and associated vapors (methanol, formaldehyde, methyl formate, CO) from the
secondary absorber are vented overhead. The methanol and formaldehyde solu-
tion from the primary absorber is fractionated to yield 99 percent methanol
and a 37 percent (weight) solution of formaldehyde containing less than 1
percent methanol. The formaldehyde product may undergo additional treatment
to remove formic acid and to prevent polymerization during storage.
The reaction for making formaldehyde from methanol using the mixed
metal oxide catalyst is shown in the following chemical equation.
CH3OH + % 02 + CH20 + H20
Methanol is mixed with air and recycled vent gas and heated to 105-
177°C (220-350°F). The reaction takes place in the presence of a mixed
oxide catalyst at temperatures between 343°C and 472°C (650°F and 880°F).
The heat of reaction is removed by circulating coolant. A heat exchanger
cools the effluent gases to 105°C (220°F) before they are quenched in the
absorber. Water is used as a sorbent to form a 37-53 weight percent formal-
dehyde solution. Part of the noncondensibles are vented from the top of
the absorber, and the remaining portion is recycled. Figure 7-9 is a
simplified flow-sheet of the mixed oxide catalyst process.
7.3.2.2 Process Emission Sources and Factors
The main source of carbon monoxide emissions from both silver catalyst-
and mixed oxide catalyst-based plants is the process absorber vent. In the
7-37
-------
LU
Q
>-
X
UJ
Q
DC
o
Q.
o
Q
O
CC
LLJ
H
O
I-
CO
a:
o
LJJ
>
UJ
_J
o
o
UJ
DC
DC
UJ
CD
DC
O
CO
CD
y*
O
o
UJ
DC
00
O)
o
-
LU
Q
i—i
X
o
Q
LU
X
o:
o
Q
Z2
O
< i
x
H-
LU
i
r^
LU
oz
ZD
CD
7-38
-------
mixed oxide-based process, absorber vent gas stream composition is dependent
primarily on gas recycle ratio. Other factors influencing absorber vent gas
composition in this process are strengths of formaldehyde produced, catalyst
formulation, catalyst age, and absorber operating temperature. The carbon
momoxide emissions from the mixed oxide catalyst-process have been estimated
at 0.16 kg/kg (0.16 Ib/lb) 37 percent formaldehyde.18 Table 7-6 presents a
representative composition for the vent gas from the mixed oxide catalyst-
process.18
The composition of the absorber vent gas stream from the silver catalyst
process varies with catalyst age and activity. Uncontrolled emissions of
carbon monoxide from this process have been estimated at 0.018 kg/kg (0.018
Ib/lb) of 37 percent formaldehyde solution.19 Controlled emissions from
this process were calculated to be 10 percent of uncontrolled emissions:
0.002 kg/kg (0.002 Ib/lb) of 37 percent formaldehyde solution. Table 7-7
presents a representative composition for the vent gas from the silver
catalyst process.
TABLE 7-6
ABSORBER VENT GAS COMPOSITION IN THE MIXED OXIDE
CATALYST PROCESS FOR FORMALDEHYDE
COMPONENT VOLUME PERCENT
Formaldehyde 0.1
Methanol 0.1
Dimethyl Ether 0.1
Oxygen 7.7
Nitrogen 86.4
Carbon Dioxide 0.1
Carbon Monoxide 1 .1
Water k.k
Source: Reference 1
7-39
-------
TABLE 7-7
ABSORBER VENT GAS COMPOSITION IN THE SILVER
CATALYST PROCESS FOR FORMALDEHYDE
COMPONENT VOLUME PERCENT
Formaldehyde 0.1
Methanol 0.3
Hydrogen 17.9
Carbon Dioxide 3.7
Carbon Monoxide 0.7
Oxygen 0.3
Nitrogen 7^.2
Water 2.8
Source: Reference 19
EPA data indicate that carbon monoxide emissions from both formaldehyde
processes were 64,900 metric tons (71,500 tons) in 1977.2 Process-specific
emissions data were not available.
7.3.2.3 Control Techniques
The majority of U.S. formaldehyde manufacturers do not currently con-
trol emissions of carbon monoxide from their process absorber vents. When
surveyed in 1975, none of the producers using the mixed oxide process
reportedly controlled CO emissions.19 Four out of 35 plants using the silver
catalyst process reportedly controlled CO emissions: two incinerated the
waste gas without heat recovery and two used the waste gas as supplemental
boiler fuel.19 The following paragraphs describe both demonstrated and
undemonstrated techniques for CO emission control.
7-40
-------
Thermal Incinerator -- Although no performance data have been reported
for thermal incinerators used on absorber vent gas streams, silver catalyst-
based producers using this device have estimated carbon monoxide removal
efficiency to be greater than 95 percent.19 The thermal incinerators in
use have operating parameters similar to those described in Chapter 6 but
are specially designed to sustain combustion using gas with a heat content
of as low as 2.24 megajoules/m3 (60 Btu/ft3).19 Incinerator design details
were considered proprietary. No thermal incinerator has been demonstrated
in a mixed oxide plant, but the technique is also a feasible control method
for this process.
The problems associated with applying thermal incineration to absorber
vent gas streams are similar to those described in previous discussions of
thermal incinerators. In addition, the relatively high hydrogen content of
the gas in a silver oxide-based plant may pose some unique hazards.
Boiler Firebox -- Two plants producing formaldehyde via the silver
catalyst process reportedly use absorber vent waste gas as supplemental
boiler fuel.19 Performance data from these plants were proprietary, but
combustion of carbon monoxide should be essentially complete. A reduc-
tion in CO emissions of more than 95 percent should be achieved.19
It is not economically attractive to use vent gas from mixed oxide
processes as supplemental boiler fuel because its energy content is very
low [0.19 megajoules/m3 (5 Btu/ft3)].18
Catalytic Incinerator -- Catalytic incineration may be a feasible car-
bon monoxide control technique in formaldehyde manufacturing. Since no
plants currently employ this technique, it is not known whether catalyst
7-41
-------
poisons are present in the vent gas. Estimated CO emissions reductions
from a catalytic incinerator are comparable to those achieved by thermal
incinerators. A more detailed description of catalytic incinerators is
found in Chapter 6.
7.3.2.4 Cost of Controls
Annualized cost information for the above mentioned carbon monoxide
controls is presented in detail in Chapter 6. Control costs for the
formaldehyde industry may be determined as described in Section 7.3.1.4.
Model plant capacities, representative absorber vent gas flow rates,
and average vent gas energy contents for the formaldehyde industry are
shown in Table 7-8. The average flow rate for the absorber vent gas is
650 Mm3 per metric ton of formaldehyde product (21,000 scf/ton) for the
silver catalyst process and 1020 Nm3 per metric ton (33,000 scf/ton) for
the mixed catalyst process.18,^ Using this information and the information
and graphs in Chapter 6, annualized costs for the control techniques dis-
cussed in Section 7.3.2.3 can be estimated for an individual plant or for
the formaldehyde industry as a whole.
7.3.2.5 Impact of Controls
Emissions Reduction - As of 1975, carbon monoxide emission control
systems were operative in only four formaldehyde plants.^ Combined produc-
tion from these plants, all of which use the silver catalyst process, repre-
sented 15 percent of total annual silver and mixed-catalyst based formal-
dehyde production.18,^,20 Therefore, approximately 85 percent of the
industry is uncontrolled with respect to carbon monoxide. The application
of any of the control systems identified earlier for formaldehyde plants
7-42
-------
TABLE 7-8
MODEL PLANT DATA FOR FORMALDEHYDE PRODUCTION WITH
THE SILVER CATALYST AND MIXED OXIDE CATALYST PROCESSES OF
FORMALDEHYDE PRODUCTION3
SILVER CATALYST
PROCESS
MIXED OXIDE CATALYST
PROCESS
Model Plant Capacity
4.54 x 10^ metric
tons/yr
(50,000 tons/yr)
k.5k x 10^ metric
tons/yr
(50,000 tons/yr)
Representative Flow Rate
Absorber Vent Gas
1.02 NmVsec
(2,170 scfm)
1.60 NmVsec
(3,390 scfm)
Energy Content of Gas
2.2^ megajoules/m3
(60 Btu/ft3)
0.19 megajoules/m3
(5 Btu/ft3)
'Data from References 18 and 19
7-43
-------
on the uncontrolled production capacity could potentially reduce annual
emissions of carbon monoxide by approximately 49,700 metric tons (54,700
tons), assuming an overall control efficiency of 90 percent. (See Section
7.3.2.2 for basis of estimate.)
Environment -- The environmental impact of the devices used to control
carbon monoxide emissions from formaldehyde plants would be similar to that
described in Section 7.3.1.5.
Energy Requirements — Energy requirements of carbon monoxide control
devices will vary with the type of device and the manufacturing process
used. The low energy content of absorber vent gas from the mixed oxide
catalyst process requires the use of substantial amounts of supplemental
fuel for all feasible control devices. However, the energy content of
vent gas from the silver catalyst process is high enough that specifically
designed self-sustaining incineration devices may be used.
The amount of supplemental fuel needed for the control devices dis-
cussed in Section 7.3.2.3 is shown in Table 7-9. Calculations for the mixed
oxide catalyst process were based on a plant producing 4.54 x 10^ metric
tons/yr (50,000 tons/yr) of a 37 percent formaldehyde solution, with a vent
gas energy content of 0.19 megajoules/m3 (5 Btu/ft3).18 Data for the silver
catalyst process were calculated for a plant producing 4.54 x 104 metric
tons/yr (50,000 tons/yr) with a vent gas energy content of 2.24 megajoules/m3
(60 Btu/ft3).19
7.3.3 Maleic Anhydride
Maleic anhydride is a white crystalline solid whose major use is in
the formulation of polyester resins.21 It is also an intermediate in the
production of fumaric acid, agricultural pesticides, and alkyd resins.21
7-44
-------
O
O
CO O
CO —
O O
(T> O CC
I Q-
UJ
o uj
U. O
—i >-
CQ CO X
< I- LU
I— Z Q
LU —I
— O
LU Z
0
LU
CC.
— •
ZD
0>
UJ
oc
>-
o
OL
LU
Z
LU
CJ
0
0
cc
Q_
U_
O
u-
"T"
C3
—
LU
^
cn
"s^
in
0)
^-B
3 x— x
O -Q
.— ^ ^—
ro ^*»
cn 3
<0 4->
E oo
cn o
-3* -^"
• \O
ro
O
OJ
QZ.
—
3D
CD*
LU
CC
>^
O
a:
LU
z
LU
o
V
in
*"•«>, **~ ^
in i—
D -C
i— ^v.
3 3
O 4-J
.—,00
ro
CTXD
0) O
E —
LA X
(V^
• CO
CM '—'
(U
-o
00
CO
LU
O
0
C£.
o_
•—
X
o
-a
a)
x
^
0)
o
i- O
o
LU
a
_i
o
a:
z
0
(_)
4-1 U.
ro
J_ 4-1
<1> ro
c (U
— JC
0
C 4-J
— 3
O
_ (—
ro 4-1
E
L— 32
(U
JZ
h-
en
in
a)
r—
3
0
•™"^x^*^*
ro JQ
O^ '"^
o ^^^
E 3
cn oo
CM
O fA
• i—*
O ^-"
O X— x
—
CO
cn
•Sx,
•—
ro ^x
cn 3
E OQ
ro O
cn o
• -a-
o "—•
o
a)
in
XN^ s~-+>
in L.
0) JZ
r— "**^
3 3
O -M
— ,00
ro
cnio
«
O L-
i-
Q. O
0
O (1)
-3" i-
4-» ro
.— D
IS JZ
cn
*s^
in
0)
r**.
D x-^
O ^Q
.— ^ ^-.
H3 ^^^
en 3
<1> -M
E OQ
r^\ O
cn o
• -3"
O •*-"
0
a)
U1
•x^ ^— >.
(/> L.
. CM
CO r—
o .
• 0
O ^- '
4-J
in
ro
4-J
ro
i_
(U
>
—
CO
in
en JQ
-------
7.3.3.1 Process Description20
Maleic anhydride is produced by the catalytic oxidation of benzene.
The reaction is shown in the following chemical equation.
H
\
C — C
P + 2 K20 + 2 C
I 02
H
Processing variations exist within the industry; however, the following pro-
cess sequence is typical.22 A mixture of benzene and air is introduced into
a reactor containing vanadium pentoxide and molybdenum catalyst. Tempera-
ture control is achieved through circulating heat transfer fluid or molten
salt. The reactor effluent is cooled before it passes through a partial
condenser and separator. The overhead material is passed through an absorber
for recovery of the anhydride as maleic acid. Maleic acid is generally
dehydrated by azeotropic distillation with xylene. Some producers use thermal
dehydration. The resulting anhydride is combined with maleic anhydride from
the condenser. Purification is accomplished by vacuum distillation. The
solid product is tableted or flaked before packaging or storage. The product
may also be shipped in bulk liquid form. Figure 7-10 is a simplified flow
sheet of the maleic anhydride process.
There are alternative processes using butane and butene feed. They
are used by at least two U.S. producers and are used in several other
countries. With the exception of raw material storage and some reactor
modifications, the debased system is about the same as the benzene-based
7-46
-------
o
UJ
LiJ
00
CXL
O
O
C\J
CD
(J
E
O)
s-
OJ
M-
QJ
QJ
O
S-
o
o
H-l
I—
o
Q
O
D;
o
CD
i—i
Q
:s
o
o:
7-47
-------
process. Small amounts of maleic anhydride are commercially produced as a
by-product of phthalic anhydride production.23
7.3.3.2 Process Emission Sources and Factors
The only source from which carbon monoxide emissions have been reported
in maleic anhydride production is the product recovery condenser vent
gas.22»2tf The waste stream comes from the product recovery scrubber which
is used to recover maleic acid from separator exit gas.
Uncontrolled emission estimates from the recovery condenser vent range
from 0.44 to 0.87 kg CO/kg maleic anhydride (0.44-0.87 lb/lb.).22 Controlled
emission factors were calculated assuming 90 percent control efficiency:
0.087 kg/kg (0.087 lb/lb) maleic anhydride. Table 7-10 shows a representa-
tive composition of the vent gases containing carbon monoxide. Total mass
carbon monoxide emissions from U.S. maleic anhydride production were esti-
mated at 117,800 metric tons (129,900 tons) in 1977.2
TABLE 7-10
PRODUCT RECOVERY CONDENSER VENT GAS COMPOSITION
IN MALEIC ANHYDRIDE PRODUCTION
COMPONENT VOLUME PERCENT
Oxygen ^.5
Nitrogen 81.9
Carbon dioxide 2.4
Carbon monoxide 2.0
Benzene 0.1
Source: Reference 2k
7-48
-------
7.3.3.3 Control Techniques
Control devices for carbon monoxide emissions are reportedly employed
in three U.S. maleic anhydride plants.24 The following paragraphs describe
demonstrated CO control techniques as well as feasible, but undemonstrated
CO control techniques.
Thermal Incinerators -- One U.S. maleic anhydride plant uses a thermal
incinerator with 30 percent heat recovery to burn waste gas from the product
recovery condenser vent.22 The incinerator operates at 760°C (1400°F) and
reportedly removes 95 percent of the carbon monoxide in the vent gas.24
Approximately 25.4 megajoules/sec (86.7 x 106 Btu/hr) of supplementary fuel
are required to maintain combustion in this device.22
Waste Heat Boiler -- At one U.S. maleic anhydride plant the vent gas
is used as the primary air supply for a waste heat boiler.22 Carbon monoxide
removal efficiency for this device is reportedly greater than 95 percent.
Catalytic Incinerator -- A catalytic incinerator similar in design and
operating parameters to the one described in Section 7.3.1.3 is used by one
U.S. maleic anhydride producer to control emissions.22 This device reportedly
removes 80 to 85 percent of the CO present in the product recovery condenser
vent stream. The problems and advantages of catalytic incinerators are dis-
cussed in Section 7.3.1.3.
7.3.3.4 Cost of Controls
Annualized cost information for the carbon monoxide control devices
described above is presented in detail in Chapter 6. Control costs for
the maleic anhydride industry may be determined as described in Section
7.3.1.4.
7-49
-------
The reported flow rate for the condenser vent gas for a plant producing
23,900 metric tons per year (26,300 tons/yr) of maleic anhydride is 18 Nm3/
sec (38,000 scfm).22 This amounts to a flow rate of approximately 21,700
Nm3/metric ton (693,500 scf/ton) of product. A vent gas energy content of
0.56 megajoules/m3 (15 Btu/scf) was estimated from reported material balance
data.23 Using this information and the information and graphs in Chapter 6,
annualized costs for the control techniques discussed in Section 7.3.3.3 can
be estimated for an individual plant or for the maleic anhydride industry as
a whole.
7.3.3.5 Impact of Controls
Emissions Reduction -- As of 1977, only three U.S. maleic anhydride
plants used carbon monoxide emission control systems.22 Combined produc-
tion from these plants represented 32 percent of the total annual produc-
tion of maleic anhydride. Therefore, approximately 68 percent of the
industry is uncontrolled with respect to carbon monoxide. Assuming appli-
cation of demonstrated control technology with 90 percent CO removal effi-
ciency, annual emissions could potentially be reduced by 72,100 metric tons
(79,500 tons).
Environment -- The environmental impacts of carbon monoxide controls
used in maleic anhydride plants are similar to those discussed in Section
7.3.1.5.
Energy Requirements — The low energy content of the product recovery
condenser vent gas from maleic anhydride plants requires the use of supple-
mental fuel in carbon monoxide emission control devices. The amount of
energy required depends primarily on the type of device used.
7-50
-------
Complete data regarding energy requirements of controls were not
reported. However, data for the plant using a thermal incinerator with 30
percent heat recovery indicated that 25.4 megajoules/sec (86.7 x 106 Btu/hr)
were necessary to maintain combustion temperatures at near-optimum levels.24
This equivalent to 4.4 megajoules/kg (1900 Btu/lb) maleic anhydride.
No information was available on the energy requirements for waste heat
boilers or catalytic incinerators used as control devices in maleic anhy-
dride production. It is likely that energy requirements for waste heat
boilers would be somewhat higher than those for thermal incinerators with
heat recovery. Catalytic incinerators, however, should require substantially
less supplemental energy because of their lower operating temperatures. How-
ever, if the plant can use the steam, a waste heat boiler is more energy effi-
cient than a thermal or catalytic incineration system with heat recovery.
7.3.4 Phthalic Anhydride
Phthalic anhydride is produced by the vapor-phase oxidation of o-xylene
or naphthalene. Approximately 67 percent of domestic-produced phthalic
anhydride is produced from o-xylene; 33 percent is produced from naphtha-
lene.25 Since the o-xylene process is more economical (i.e., this process
uses a cheaper raw material and yields slightly more product on a weight
basis), future phthalic anhydride plants will probably be designed to use
o-xylene as a feedstock.26
7.3.4.1 Process Description
There are basically two processes used for phthalic anhydride produc-
tion in the United States. Processes using naphthalene as a feedstock use
7-51
-------
fluidized bed reactors; whereas, o-xylene-based plants use tubular fixed
bed reactors. Except for the reactors and catalyst handling and recovery
facilities used, the two processes are similar.
The following reaction describes the conversion of o-xylene to phthalic
anhydride.
catalyst I 0 I Y) + 3 H 0
^^ V ^K.<-> *^
o-xylene Oxygen phthalic water
anhydride
Naphthalene is converted to phthalic anhydride via the following reaction
+ 4% 02 catalyst
2 CO-
naphthalene oxygen phthalic carbon water
anhydride dioxide
In both processes, a vanadium oxide catalyst is used. Small amounts of
phthalic anhydride produced are oxidized to maleic anhydride, C02, and water
7-52
-------
Figure 7-11 is a flow diagram for an o-xylene based phthalic anhydride
process. In both the o-xylene and naphthalene-based processes, filtered
air is compressed to a range of 170 to 200 kilopascals (10-14 psig) and pre-
heated.26 Liquid o-xylene is mixed with reaction air and vaporized before
it enters the fixed tubular bed reactors; whereas, liquid naphthalene is
injected directly into a fluidized bed reactor and vaporized. Reactors in
both processes operate at 340-385°C (650-725°F). A small amount of sulfur
dioxide (S02) is added to the reactor feed to maintain catalyst activity.
Reactor effluent is used to generate low pressure steam in a waste
heat boiler and then flows through a series of condensers (a parallel series
of tubular condensers which are alternately heated and cooled). Crude
phthalic anhydride is condensed as solid crystals on the condenser tube
fins. It is then melted, removed from the condenser tubes, and sent to pre-
treatment. In this step, phthalic acid is dehydrated to the anhydride form,
and impurities (water, maleic anhydride, and benzoic acid) are partially
evaporated. The pretreated liquid stream is then sent to a vacuum distilla-
tion section where pure (99.8 percent) phthalic anhydride is obtained as a
distillate. The pure product may be stored in a molten state or solidified
to flakes and bagged for shipment.
All future phthalic anhydride industry growth is expected to be based
on o-xylene feed. In 1977, only three of the ten phthalic anhydride plants
in the U.S. were naphthalene-based.25 Projected production capacity from
naphthalene-based plants is expected to remain the same through 1985.26
7-53
-------
z
o
z
o
H
O
<
cc
u.
oo
00
UJ
o
o
o:
Z
UJ
2
UJ
c
H
UJ
c
0.
CO
z oc
o uj
t °>
Z
o
o
o eo
O
< Ul
O C
O
CNJ
-------
7.3.4.2 Process Emission Sources and Factors
The major source of carbon monoxide emissions from phthalic anhydride
plants is the main process vent which comes from the switch condensers.
Only 50 percent of U.S. plants control CO emissions from this point. Three
plants reportedly use thermal incinerators and two others use a combination
thermal incinerator/waste heat boiler to control emissions.27
Uncontrolled carbon monoxide emissions from the main process vent have
been estimated at 150 kilograms of CO per metric ton of phthalic anhydride
(300 Ib/ton) in o-xylene-based plants, and 50 kilograms of CO per metric
ton of phthalic anhydride (100 Ib/ton) in naphthalene-based plants.5 Inciner-
ation reportedly controls CO emissions to 0.125 g/kg (0.25 Ib/ton) phthalic
anhydride in o-xylene-based plants, and 0.05 g/kg (0.10 Ib/ton) phthalic
anhydride in plants where naphthalene is used as a feedstock.25
Recent EPA data indicated that carbon monoxide emissions from U.S.
phthalic anhydride production were 50,900 metric tons (56,100 tons) in
1977.2 Process-specific emissions data were not available.
7.3.4.3 Control Techniques
As was previously mentioned, only 50 percent of U.S. phthalic anhydride
plants employ carbon monoxide control devices on their main process vent
streams. The following paragraphs describe demonstrated CO control techni-
ques as well as undemonstrated control techniques, for both o-xylene- and
naphthalene-based processes.
Thermal Incinerator -- Three U.S. phthalic anhydride manufacturers use
thermal incinerators to control carbon monoxide emissions from their main
7-55
-------
process vents. One plant is naphthalene-based and two are o-xylene based.
Operating at 649°C (1200°F), the incinerator in one o-xylene-based plant
reportedly removes greater than 90 percent of the CO in the switch condenser
of gas.25 Removal efficiencies of only 80-85 percent have been reported
for a similar incinerator used in the naphthalene-based plant.26'27
Thermal incinerators may operate at higher temperatures than the ones
currently in use (760-860°C [1400-1580°F]). Under these conditions, CO con-
trol efficiency could increase to 95 percent.25
Tables 7-11 and 7-12 list typical main process vent compositions for
o-xylene and naphthalene-based plants, respectively. Because of the low
energy content of the main process vent gas (0.075-0.112 megajoules/m3
[2-3 Btu/ft3]), supplemental fuel is needed to achieve complete combustion
in a thermal incinerator.25 Fuel requirements can be reduced if vent gas
is preheated before being incinerated by heat exchange with the incinerator
flue gas. However, preheating increases the danger of explosion if slugs
of condensed phthalic anhydride are present in the vent gas.25»26>27
Thermal Incinerator/Waste Heat Boiler -- A thermal incinerator with a
waste heat boiler is used to control carbon monoxide emissions from the main
process vent in two U.S. phthalic anhydride plants.27 This control technique
reportedly achieves greater than 99 percent reduction in CO emissions.27
The vent gas is not preheated prior to incineration, thereby avoiding the
danger of explosion. Using an incinerator plus a waste heat boiler as a
control technique requires more supplementary fuel than using an incinerator
alone; however, as Table 7-13 shows, energy is recovered in the process
stream produced.
7-56
-------
TABLE 7-11
TYPICAL MAIN PROCESS VENT GAS COMPOSITION FROM
0-XYLENE BASED PHTHALIC ANHYDRIDE PRODUCTION
COMPONENT VOLUME PERCENT
Sulfur dioxide <0.1
Carbon monoxide 0.6
Carbon d ioxi de 1.3
Nitrogen 76.9
Oxygen 15-7
Phthalic anhydride <0.1
Maleic anhydride <0.1
Benzoi c acid <0.1
Water .4
Source: Reference 27
TABLE 7-12
TYPICAL MAIN PROCESS VENT GAS COMPOSITION FROM NAPHTHALENE'
BASED PHTHALIC ANHYDRIDE PRODUCTION
COMPONENT VOLUME PERCENT
Phthalic anhydride <0.1
Maleic anhydride <0.1
Naphthoquinone <0.1
Oxygen 12.2
Ni trogen 78.1
Carbon d ioxide 5.1
Carbon monoxide 0.4
Water 4.
Source: Reference 27
7-57
-------
Catalytic Incinerator -- Catalytic incinerators have reportedly been
used to control carbon monoxide emissions from other chemical processes.
Emission reductions of 99 percent have been reported. It is not known
whether any components which could poison the catalyst are present in the
vent gases from phthalic anhydride plants. Since catalytic incinerators
operate at lower temperatures than thermal incinerators (410-525°C [800-
1000°F]), supplemental fuel requirements are somewhat less than require-
ments for an incinerator.
7.3.4.4 Cost of Controls
Annualized cost information for the above-mentioned carbon monoxide
controls is presented in detail in Chapter 6. Control costs for the
phthalic anhydride industry may be determined as described in Section
7.3.1.4.
A flow rate for the process vent gas from a model phthalic anhydride
plant producing 5.9 x 101* metric tons/yr (1.3 x 108 Ib/yr) is 56 Nm3/sec
(119,000 scfm).26 An energy content of 0.112 megajoules/m3 (3 Btu/ft3) has
been reported for vent gas containing carbon monoxide.26 Using this infor-
mation and the information and graphs in Chapter 6, annualized costs for
the control techniques discussed in Section 7.3.4.3 can be estimated for an
individual plant or for the phthalic anhydride industry as a whole.
7.3.4.5 Impact of Controls
Emissions Reduction -- As of 1977, carbon monoxide emission control sys-
tems were operating in five phthalic anhydride plants in the United States.26
Four plants used o-xylene as a feedstock and one was naphthalene-based.26
7-58
-------
Combined production from these plants represented 46 percent of the total
annual production of phthalic anhydride. Therefore, approximately 54 per-
cent of the industry is uncontrolled with respect to carbon monoxide. One
source has estimated that if controls with at least 90 percent efficiency
were applied industry-wide, carbon monoxide emissions could be reduced to
less than 1,000 metric tons/yr (1,100 tons/yr).25
Environment -- Incinerators operating at the upper limits of their
temperature range will produce more N0x emissions than those operating at
lower temperatures. It has been reported that emissions of NO will increase
by approximately 15 percent when operating temperatures increase from 760-
860°C (1400-1580°F).10 N0x emissions from catalytic incinerators should be
negligible, since operating temperatures for this type of incinerator are
considerably less than those of thermal incinerators.
If it becomes necessary to use fuel oil rather than natural gas as
supplementary incinerator fuel, sulfur oxides (S0x) emissions may increase.
The amount of S0x emitted will depend on the sulfur content of the fuel oil
and the quantity of oil consumed.
Energy Requirements -- As was previously discussed, the low energy
content of the main process vent gas from phthalic anhydride plants necessi-
tates the use of supplementary fuel in the operation of any of the carbon
monoxide emission control devices. The amount of energy required will depend
primarily on the type of control device used.
The amount of supplemental fuel needed for several of the control
devices is shown in Table 7-13. The calculations were based on a plant
producing 5.9 x 10^ metric tons/yr (1.30 x 108 Ib/yr ) phthalic anhydride
7-59
-------
with a process vent gas energy content of 0.112 megajoules/m3 (3 Btu/ft3).26
Supplemental fuel data for catalytic incinerators were not available; how-
ever, because of their lower operating temperatures, substantially less
energy would be required.
TABLE 7-13
ENERGY REQUIREMENTS FOR CO EMISSION CONTROLS
IN PHTHALIC ANHYDRIDE PRODUCTION
ENERGY REQUIRED PER kg(lb)
CONTROL DEVICE ENERGY REQUIRED8 OF PHTHALIC ANHYDRIDE
Thermal incinerator 19-5 megajoules/sec 9-5 megajoules/kg
without heat recovery (66 x 106 Btu/hr) (4.1 x 103 Btu/lb)
Thermal incinerator 55-3 megajoules/sec 27.0 megajoules/kg
+ waste heat boiler (189 x 1O6 Btu/hr) (11.6 x 103 Btu/lb)
a
Data from Reference 25, based on 8000 operating hours per year.
Steam production 12.2 kilograms per second (97,000 Ib/hr) at 3.2 megapescals
(450 PSIG) and 400°C (750°F)
7.4 IRON AND STEEL
Four methods used in making steel or smelting ferrous ore contribute
heavily to the amount of carbon monoxide emitted from industrial processes.
These four methods include steelmaking with basic oxygen furnaces (BOF's),
ferroalloy and steel production using submerged arc and electric arc fur-
naces, respectively, ore dust agglomeration using sintering furnaces, and
gray iron production from cupolas.
Table 7-14 lists mass carbon monoxide emissions from the processes
described above. Estimated carbon monoxide emissions from these sources
totaled 1.95 x 106 metric tons (2.15 x 106 tons) in 1977.2 These emissions
7-60
-------
comprised about 23 percent of the CO emitted from industrial processes and
about 11 percent of the total amount emitted from stationary sources.2 The
two major emitters of CO in the iron and steel industry are gray iron cupolas
and sintering furnaces. Emissions from these sources represented about 84
percent of the total CO emissions from the iron and steel industry in 1977.2
TABLE 7-14
MASS EMISSION ESTIMATES FOR CARBON MONOXIDE FROM THE
IRON AND STEEL INDUSTRY, 1977
CARBON MONOXIDE EMISSIONS
SOURCE Metric Tons Tons
Sinter plants 624,700 688,600
Basic oxygen furnaces 99,200 109,400
Electric arc furnaces 205,700 226,800
Gray iron cupolas 1,020,800 1,125,200
TOTAI- 1,950,400 2,150,000
Source: Reference 2
The following sections discuss the processes, emission factors, control
techniques, and impact of cost of controls for each of the methods listed
above. A discussion of blast furnace CO emissions, which are almost com-
pletely controlled, is also included.
7.4.1 Basic Oxygen Furnace
The basic oxygen furnace (BOF) process, also known as the Linz-Denowitz
(L-D) process, is used to produce a major portion of steel in the U.S. The
furnace is a pear-shaped, refractory-lined vessel, open at the top for
7-61
-------
charging while vertical and for pouring while tilted. This process is being
increasingly used because of its high production rates, simplicity, and
efficient operation.
7.4.1.1 Process Description
The feed metal used in the BOF process is typically 70 percent molten
blast furnace iron and 30 percent scrap.28 The furnace is also charged with
fluxes, such as burnt lime, limestone, burnt dolomite, and fluorspar.
Oxygen is blown into the charge through a water-cooled lance under pressure
ranging from 1.1 to 1.3 megapascals (140 - 180 psi).28 The process converts
the hot metal into steel by oxidation of carbon, phosphorus, silicon, sulfur,
and other impurities in the iron. This reaction occurs at approximately
2000°C (3600°F) and atmospheric pressure.28 The steel is tapped into a ladle
where desired alloying materials may be added. The molten steel is usually
poured into ingot molds. The slag is tapped into slag pots and sent to the
slag dump yard.
7.4.1.2 Process Emission Sources and Factors
Large amounts of carbon monoxide are generated by the oxidation reactions
occurring in the BOF process. The exhaust gas at the surface of the molten
liquid has a carbon monoxide content ranging from 87 to 95 percent.29
Exhaust gas flow rates range from 570 to 940 Nm3/sec (1.2 x 106 -1.99 x 106
ft3/min).28 Typical exit temperatures range from 1600°C - 1900°C (2900-
3500°F).28
Uncontrolled carbon monoxide emissions from the BOF process are
estimated to be 70 kg/metric ton (140 Ib/ton) steel produced.5 When control
7-62
-------
methods are applied, emissions are reduced to less than 1.5 kg carbon mon-
oxide per metric ton steel (3 lb/ton).5 Total mass CO emissions from U.S.
basic oxygen furnaces were estimated to be 99,200 metric tons (109,400 tons)
in 1977.2
7.4.1.3 Control Techniques
Most basic oxygen furnaces in the United States control carbon monoxide
emissions by burning the waste gases with excess air in an open hood (Figure
7-12). A few U.S. facilities inhibit combustion with a retractable closed
hood and flare the off-gas. Some foreign facilities collect it as fuel
after cleaning (Figure 7-13).
In an open hood system, space is provided between the furnace and the
hood to admit air for the combustion of carbon monoxide. Closed hood sys-
tems use retractable skirts or other methods to limit the quantity of air
entering the hood. Hoods are water cooled, using either hot or cold water
or steam. When either type of hood is used, reductions in carbon monoxide
emissions exceed 98 percent.5 During charging and pouring, the furnace
and hood are disengaged. However, most of the CO is emitted during blowing.
7.4.1.4 Cost of Controls
The hooding design affects the cost of the total system. Open hoods
draw in air on a relatively uncontrolled basis, thus increasing the capital
and operating costs of the particulate collection equipment.30
Closed hood systems are more difficult to fabricate and maintain. In
addition, provision must be made for gas accumulation or flaring. However,
particulate collection costs less for closed hood systems. Economics of the
7-63
-------
UJ
h-
oo
oo
LU
CD
O
OO
Q
I
00
CD
Q
Z
«C
o
o
Q_
O
x
o
oo
<
CO
OsJ
I
o;
7-64
-------
co
>-
CO
CD
O
I
CO
CD
Q
Z
<
O
O
O
UJ
OO
O
_J
O
UJ
CD
X
O
O
I—I
CO
CQ
CO
r—
I
CD
7-65
-------
entire process and particulate emission regulations will determine the more
appropriate method.
7.4.1.5 Impact of Controls
Emissions Reduction -- Open hood combustion reduces CO in the furnace
exhaust gas to less than 1.5 kg/metric ton (3 lb/ton).2 If closed hood sys-
tems are used, 98 percent of the carbon monoxide produced can be recovered
and used as waste heat boiler fuel.30
Environment -- The industry-wide acceptance of burning waste gas in
steel production via the BOF process has significantly reduced CO emissions.
Nitrogen oxide (N0x) emissions during combustion of the waste gas under the
open hood are about 180 to 500 micrograms of NO per metric ton (0.36 to
1.0 pound per ton) of steel produced.30 There would probably be lower NO
emissions from closed hood collection since no incineration occurs. However,
there would be N0x emissions from flaring or from burning the gases in a
boiler. Particulate emissions are also greater with open hoods than with
closed hoods.
Energy Requirements -- The energy content of BOF exhaust gas is high
enough so that no supplemental fuel is necessary to maintain combustion in
an open hood or flare. When the carbon monoxide is burned, about 470 kilo-
joules/kilogram (400,000 Btu/ton) are produced.30 If closed hoods are used
and the exhaust gas is cleaned and recovered, it may be used to produce
steam for other process units.
7.4.2 Blast Furnace
Blast furnaces are vertical, refractory-lined shaft furnaces up to
36.6 meters (120 feet) tall and 8.5 meters (28 feet) in diamter. They
7-66
-------
reduce iron ore to molten pig iron, most of which goes directly to steel
furnaces.
7.4.2.1 Process Description
Blast furnaces are so called because air preheated from 760°C to 1150°C
(1400°F to 2100°F) is blown into the furnace near the bottom to burn the
coke.28 Iron ore, sinter, iron or steel scrap, coke, and flux (limestone)
are charged into the top of the furnace.28 In 1973, an average of 1.5
metric tons (1.7 tons) of charge was consumed per ton of pig iron produced.28
Blast furnaces operate at pressure ranging from 170 to 580 kilopascals (10
to 70 psi).28 When temperatures inside the furnace exceed 1450°C (2640°F),
the combustion product, C02» reacts as follows to produce carbon monoxide:
C + C02 -> 2CO
The carbon monoxide is necessary to reduce the iron oxides present in the
ores to elemental iron. As the metals descend, they are heated by the
reducing gases.
As the elemental iron moves toward the furnace fusion zone, it becomes
molten and collects in the hearth (See Figure 7-14). The limestone flux
reacts with impurities in the ore and coke and forms a molten layer of slag
on the pool of iron. Periodically, the molten iron and slag are tapped from
the blast furnace. The molten pig metal typically contains 4.1 percent
carbon, 0.9 percent silicon, 0.026 percent sulfur, 0.30 percent phosphorus
and 0.35 percent manganese.31
7-67
-------
7.4.2.2 Process Emission Sources and Factors
Exhaust gases leave the blast furnace at temperatures of 180°C to 280°C
(350 F to 540°F). The gas flow rate increases linearly with the coke feed
rate. One source estimated that 2.2 to 3.5 kilograms of exhaust gas are
generated per kilogram of pig iron produced (2.2 to 3.5 lb/lb).30 As much
as 30 percent of the exhaust gas volume may be carbon monoxide.32 Uncon-
trolled carbon monoxide emissions in furnace exhaust gas average 875 kilo-
grams of CO per metric ton of pig iron (1750 pounds per ton).5 However,
relatively little carbon monoxide is vented to the atmosphere, since 99.9
percent of the CO generated is normally collected, cleaned, and used as
process fuel.5
Occasionally, conditions within the furnace such as "slips" (sudden
movements of the charge into the furnace) generate high pressures which
open the furnace's pressure relief valves. Uncontrolled amounts of carbon
monoxide escape through the relief valves and the furnace charging enclosure
during "slips". No emissions estimates for CO have been reported for "slip"
conditions.
7.4.2.3 Control Techniques
The technique for controlling carbon monoxide emissions from blast
furnaces is part of the system used to control particulate emissions. A
typical system is shown in Figure 7-14. Initially, exhaust gas passes
through a settling chamber or a dry cyclone, where about 60 percent of the
dust is removed. Next, the gas undergoes a one- or two-stage cleaning
operation, in which the remaining particles are removed by a wet scrubber
7-68
-------
XHOMdiXDBHO X3IU8
«=c
2:
or
CO
CO
-------
or electrostatic precipitator. The cleaned gas is then ready to be used
as low energy process fuel.
7.4.2.4 Cost of Controls
There are no additional costs for controlling carbon monoxide emissions
from blast furnaces. The exhaust gas is used as a fuel.
7.4.2.5 Impact of Controls
Emissions Reductions -- As was previously discussed, carbon monoxide
emission control from blast furnaces is relatively complete throughout the
industry. Any remaining CO emissions result from escaping gas during high
pressure "slips." Improved charging techniques and operating practices
which closely adhere to furnace design specifications have significantly
reduced the number of "slips."
Environment — Since blast furnace gas is used as a fuel, nearly all
the carbon monoxide produced is oxidized to C02 before it reaches the
atmosphere.32
Energy Requirements -- The energy content of the blast furnace exhaust
gas is approximately 3.73 megajoules/m3 (100 Btu/ft3).32 It is therefore
economical to use the exhaust gas for process fuel. About 30 percent of the
cleaned gas is typically used to fire the stoves in which blast furnace air
is preheated.32 The remaining gas is used as fuel for other in-plant
purposes.32
7.4.3 Submerged Electric Arc Furnace
Submerged arc furnaces are used in the production of ferroalloys. The
basic raw materials used are metallic ores, limestone, and a reducing agent
(coke or low-volatile coal).3t* The exact composition of the charge depends
7-70
-------
on the product desired. Iron, silicon, manganese, chromium, calcium, and
zirconium are some of the metals which may be alloyed or reacted in the
furnace.
7.4.3.1 Process Description
Submerged arc furnaces of the same general design are used throughout
the ferroalloy industry. The cylindrical steel furnace shell has a flat
bottom and is supported on an open foundation that permits air cooling and
heat dissipation. The furnace shell's interior walls are lined with
refractory brick. One or more tapholes for removing slag and metal exist
at hearth level.3k
Graphite electrodes in electric submerged arc furnaces extend three to
five feet into the charge. The coke in the charge reacts with the metal
oxides and reduces the ores to base metal. Maximum furnace temperature
is 1570°C (2860°F). Most furnaces operate at atmospheric pressure.35
Continuous power is supplied to the furnace electrodes, whose depth
is varied during the process to maintain a uniform electrical load through-
out the charge. Oxidation begins to occur when molten metal begins to
form and continues until the entire charge is in solution. At the end of
the process, the electrodes are raised, and the molten product is tapped
into ladles and further treated, as desired. Slag removal may occur prior
to or during tapping, or at the end of the tap.
7.4.3.2 Process Emission Sources and Factors
The composition of exhaust gas from submerged arc furnaces varies with
hooding practices, slagging practices, process stage, and whether or not
7-71
-------
oxygen lancing is used. Major constituents of the exhaust gas include carbon
monoxide, carbon dioxide, oxygen and nitrogen. Fluorides and other vaporized
metallic compounds may also be present, depending on the type of ferroalloy
being produced.35
Emission points in electric submerged-arc furnaces include the electrode
ports in the furnace roof, the tapping spout, the slagging door, and the open
furnace top during charging. Uncontrolled carbon monoxide emissions from
direct electric arc furnaces have been estimated at 9 kilograms CO per metric
ton of ferroalloy produced (18 lb/ton).5 Exhaust gas from a number of facil-
ities tested contained between 60 and 95 percent CO.36 Carbon monoxide con-
centrations of 80 to 90 percent are common during short periods of each
cycle.35 Typical gas volumes range from 50 to 190 normal cubic meters per
second (100,000-400,000 scf/min).35
Recent EPA emissions estimates indicate that 205,700 metric tons
(226,800 tons) of carbon monoxide were produced from both direct- and
submerged-arc furnaces.2 No process-specific data were available.
7.4.3.3 Control Techniques
A number of techniques exist for controlling carbon monoxide emissions
from electric submerged-arc furnaces. The following paragraphs describe
these techniques.
Carbon monoxide reduction in electric arc furnaces is achieved by
inducing air into the exhaust hood. In a few cases the gases are collected,
then burned. There are three hood configurations for submerged arc fur-
naces: the open, the semi-enclosed, and the sealed furnace. The type of
hooding system used has an important effect on CO emission reduction. A
7-72
-------
few installations also supplement their hoods by shrouding or enclosing the
furnace area to capture the pollutants emitted during charging and tapping
operations.34 The CO emissions which escape the hoods are emitted in the
exhaust from the furnace building.
The open electric submerged arc furnace configuration (Figure 7-15)
employs a water-cooled canopy hood 2 to 2.7 meters (6 to 8 feet) above the
furnace rim. Air surrounding the furnace burns the CO as it combines with
the hot gases under the hood, diluting them by as much as 50 to I.34
In the semi-enclosed electric submerged arc furnace (Figure 7-16) emis-
sions are drawn from beneath a water-cooled cover that completely seals the
furnace except for annular spaces around the three electrodes through which
the raw materials are charged. Because very little air enters the semi-
enclosed furnace, gases from the furnace are concentrated in carbon monoxide
and can be used as fuel or flared after cleaning.35
Emissions leaking through the charging holes around the electrodes can
be minimized by maintaining a negative pressure within the furnace. This
involves using a fan to draw gases into the dust-cleaning device. The in-
duced air also oxidizes some of the carbon monoxide, reducing its fuel
value and raising the gas exit temperature.35
Another way of reducing emissions from sealed furnaces (Figure 7-17)
is by packing seals around the electrodes and charging chutes. In this
case, the fuel value of the exhaust gas is preserved because a slight posi-
tive operating pressure is maintained, preventing leakage of air into the
furnace. Gases withdrawn from sealed furnaces may be as little as 2 to 5
7-73
-------
O
CO
QJ
O
c
O)
S-
O)
OJ
u
S-
13
O
00
CO
CO
ID
C£
GO
t—i
CXL
>-
CQ
O
CXL
O
O
UJ
a.
o
un
CD
7-74
-------
7-75
-------
cc:
<:
o
<
to
<£>
CO
O)
O
c
o
CO
CO
CQ
UJ
Q
•=c
LU
CO
r^
i
CD
I—H
LL.
7-76
-------
percent of the volume handled in open furnaces.35 Gases from sealed fur-
naces are flared or are used for fuel.
Production of silicon metal or alloys containing over 75 percent silicon
are limited to open furnaces with canopy hoods because the techniques which
need to be used to prevent crusting and bridging of the charge and "blows"
(jets of extremely hot gas) cannot be employed with semi-enclosed and sealed
furnaces.35
Some of the specific types of hood systems used are as follows: The
roof hood or the "plenum roof" (Figure 7-18) covers the furnace roof with
openings for the electrodes and overhangs above the charge door and tapping
spout.35 The direct shell evacuation or "fourth hole" system (Figure 7-19)
(so-called due to the three electrode holes already in the furnace roof)
ducts the exhaust gases from beneath the furnace roof. A gap in the duct
elbow aspirates air to burn the waste gases. This system is totally in-
effective when the roof ring is swung aside for charging and during tapping.
The advantages are similar to those for sealed ferroalloy furnaces (i.e.,
less exhaust gas).35
The side draft hood (Figure 7-19) mounted on the furnace roof draws a
high velocity indraft of 31 to 190 Nm3/metric ton (1000 to 6000 ft3/ton) to
capture emissions around the electrodes. No extra air is needed to burn
the escaping carbon monoxide as the hood only partially surrounds the elec-
trodes, hence the name side-draft. However, carbon monoxide destruction
may not be as complete as achieved in the direct shell evacuation system
during meltdown since the side-draft hood draws in a large amount of cool
air, possibly lowering the temperature of the exhaust draft below the
ignition point.35
7-77
-------
^x
FIGURE 7-18. ROOF HOOD
7-78
-------
on
SIDE DRAFT
-o-
v
y
DIRECT EVACUATION-FOURTH HOLE
FIGURE 7-19. SIDE DRAFT AND DIRECT EVACUATION HOODS
7-79
-------
7.4.3.4 Cost of Controls
The problem of cost development for carbon monoxide emission control
is similar to that described in Sections 7.4.1.4 and 7.4.2.4. Since the
ventilation and transport systems used in CO control are also part of the
particulate control system, it is difficult to separate costs for CO con-
trol alone.
7.4.3.5 Impact of Controls
Emissions Reduction -- When applied, the emission control techniques
described in Section 7.4.3.3 are effective methods of reducing carbon monox-
ide emissions from electric submerged arc furnaces. If controls were employed
on all submerged arc furnaces, emissions reduction of more than 90 percent
should be achieved.35
Environment -- No nitrogen oxides (NO ) are formed during the carbon
reduction of oxidic ores.35 Any N0x formed as a result of carbon monoxide
emission control would be due to fixation of atmospheric nitrogen. If
closed systems are used and the CO-rich exhaust gas is recovered and used
as process fuel, N0x emissions should not be any greater than if natural
gas were used as fuel.
Energy Requirements -- If submerged-arc exhaust gas is burned in an
open hood, no supplementary fuel (other than air) is necessary to maintain
combustion. Neither is supplementary fuel needed if the gas is flared.
Because the exhaust gas is 60-90 percent carbon monoxide, it can be
used as a process fuel.35 The energy content of the gas was calculated to
be approximately 10 megajoules/m3 (270 Btu/ft3).
7-80
-------
7.4.4 Direct Electric Arc Furnace
Direct electric arc furnaces are used in the production of high-alloy
steels (e.g., carbon and stainless steels) and considerable amounts of mild
steel.28 Steel production in direct arc furnaces has steadily increased,
due to the increased availability of steel scrap.37
7.4.4.1 Process Description
Typical electric arc furnaces range in diameter from about 1 meter (3
feet) up to 4 meters (12 feet) with holding capacities of 230 kilograms (500
pounds) to 23 metric tons (25 tons), and production rates from 115 kilograms
(250 pounds) to 10.9 metric tons (12 tons.) per hour.38 Modern furnaces up
to 5.2 meters (17 feet) in diameter may hold 59 metric tons (65 tons) and
have production rates of over 18 metric tons (20 tons) per hour.38
Electric arc furnaces are basically refractory-lined crucibles with a
steel shell. In almost all applications, the furnace roof can be swung
aside for top charging. The roof is also refractory-lined, with ports allow-
ing the insertion of three graphite electrodes into the furnace just above
the surface of the charged metals. Maximum furnace temperature is 1570°C
(2860°F). Most furnaces operate at atmospheric pressure.35
The charge for iron or steelmaking usually consists of steel scrap,
cast iron scrap, pig iron, alloying elements, and flux. Preheating the
steel scrap is not a common practice when direct electric arc functions are
used. Addition of oxygen (oxygen lancing) during the melting process
reduces energy consumption and increases production rates.
The oxidation process in direct arc furnaces is similar to that described
for submerged arc furnaces (Section 7.4.3.1). Similar tapping and slagging
procedures are employed in both processes.
7-81
-------
7.4.4.2 Process Emission Sources and Factors
Carbon monoxide is generated by reaction of the carbon electrodes or
carbon in the steel scrap with blown oxygen or iron oxides. Major exhaust
gas components include oxygen, nitrogen, carbon dioxide, carbon monoxide,
and gaseous fluoride.28 Exhaust gas composition is influenced by the stage
of the heating process: typically, the CO content rises sharply at the
beginning of the melt and again during oxygen lancing. The exhaust gas
leaves the furnace at temperatures of 650°C to 980°C (1200°F to 1800°F).39
Data describing carbon monoxide emissions from direct arc furnaces are
limited. However, testing at one source indicated that carbon monoxide emis-
sions may be as high as 3 kilograms per metric ton (6 Ib/ton) of steel pro-
duced.40 Recent EPA estimates of carbon monoxide emissions from both sub-
merged- and direct-electric arc furnaces were 205,700 metric tons (226,800
tons) in 1977.2 No process-specific emissions data were available.
7.4.4.3 Control Techniques
The only known technique for controlling carbon monoxide emissions from
direct-arc furnaces is the direct shell evacuation system.37 This system,
shown in Figure 7-20, withdraws all potential emissions from the furnace
before they escape and mix with the ventilation air. The furnace roof is
constructed so that it can be elevated and rotated aside during top charging
and tapping and slagging. During furnace operation, the direct shell evacua-
tion system maintains a negative pressure within the furnace. As a result,
air is drawn into the furnace around the electrodes and through a small gap
in the roof. It then flows through the exhaust duct, where it not only cools
7-82
-------
BUILDING
MONITOR
EMISSIONS^
F?i
FURNACE
O
DSE
n
CLEAN AIR
EXHAUST GAS
FABRIC FILTER
Source: Reference 37
FIGURE 7-20. DIRECT SHELL EVACUATION (DSE) SYSTEM OPEN ROOF
7-83
-------
the exhaust gas but also promotes combustion of large amounts of carbon monox-
ide present in the gas.37 On small steel furnaces direct evacuation is not
always a viable option because of (1) lack of space for fourth hole in the fur-
nace roof and (2) pressure fluctuations in furnace which are too rapid for
automatic control of dampers in the exhaust duct.41
One source has estimated that direct shell evacuation systems achieve
about 85 percent carbon monoxide emission reduction.37 However, these sys-
tems cannot be used in producing some types of alloy steels. During the
production of some alloys, a second "reducing" slagging takes place. Air
will oxidize these slags and prevent their removal.37
An additional problem with direct shell evacuation systems is their
inability to function during top charging, tapping, and slagging. When the
roof is rotated during these times, much of the carbon monoxide in the
exhaust gas is not oxidized and rises directly through the roof of the shop.37
7.4.4.4 Cost of Controls
As discussed in previous sections, it is difficult to separate costs
of carbon monoxide emission controls from costs of particulate control sys-
tems. In almost all cases, the same ventilation and transport systems will
be used for both pollutants.
7.4.4.5 Impact of Controls
Emissions Reduction -- Direct shell evacuation systems have been found
to achieve up to 85 percent reductions in carbon monoxide emissions from
direct arc furnaces. When applied, these systems should substantially
reduce total mass emissions of carbon monoxide from this source.
7-84
-------
Environment -- Industry data indicate that nitrogen oxide (NO ) emis-
sions from direct arc furnaces are less than 0.05 kilograms per metric ton
(0.1 Ib/ton) of steel produced.37 Thus, almost all N0x formed during com-
bustion of carbon monoxide in a direct shell evacuation system would result
from the fixation of atmospheric nitrogen. N0x emissions should not increase
to significant levels as a result of carbon monoxide emissions reduction.
Energy Requirements ~ If exhaust gas from direct arc furnaces is burned
in a direct shell evacuation system, no supplementary fuel (other than air)
will be necessary to maintain combustion. Although the exhaust gas may
contain up to 20 percent CO during parts of the furnace cycle, average con-
centrations are too low for the exhaust gas to be used as process fuel.37
7.4.5 Gray Iron Cupola
7.4.5.1 Process Description
Cupolas, the most common furnaces for making iron castings and ingots,
may be water cooled or refractory lined. Air blown through a bed of coke
near the bottom of the cylindrical furnace rises through alternating charges
of pig iron and scrap, limestone flux and coke. Descending charges are pre-
heated by rising gases which may vary between 260° and 1200°C (500° and
2200°F), depending on the blast air rate, the preheat temperature, the
charge door induced draft rate and the cycle of operation.^2 Temperatures
of the cupola exhaust gases drop with the addition of each charge and are
cooled considerably from cold outside air induced through the charge door.
Molten iron and slag are tapped below the ports which introduce the blast
air into the furnace. Furnaces which preheat the combustion air are called
hot-blast cupolas. The air may be heated from an external source or with
7-85
-------
an off-gas heat recovery system. One system supplies the heat by recuperat-
ing the heat of the flue gases after combusting the CO.42
Cupolas range in size from 70 centimeters to 395 centimeters (27 inches
to 155 inches) in diameter producing one ton per hour in the smallest jobbing
foundries to more than 90 metric tons per hour (100 ton/hr) in captive foun-
dries. k3 Blast air is usually supplied at the rate of 935 cubic meters per
metric ton (30,000 cubic feet per ton) of melt capacity.42
7.4.5.2 Process Emission Sources and Factors
Exhaust gases from the cupola furnace are a significant source of CO
emissions. Recent EPA estimates indicate that 1,020,800 metric tons
(1,125,200 tons) of carbon monoxide were produced in cupola furnaces in
1977.2 Average carbon monoxide emissions have been estimated to be 72.5
kg/metric ton of metal charged (145 lb/ton).5 Actual carbon monoxide emis-
sions may vary with the quality of charge material, the volume and rate of
combustion air, and the melting zone temperature.38
7.4.5.3 Control Techniques
Afterburners are applied to cupola furnaces to reduce CO emissions.38
Besides reducing carbon monoxide emissions to 4 or 5 kilograms per metric
ton (8 to 10 pounds per ton) of iron melted, afterburners also reduce the
hazard of explosion and consume oil vapors and coke breeze, minimizing
damage and maintenance on particulate collection devices.38
The afterburner chamber is located in the top part of the cupola stick
above the charge door. For best gas-flame contact without quenching, the
off-gas, multiple burners are installed just below the charge door. Induced
7-86
-------
drafts from the charge door are essential to insuring sufficient mixing and
providing ample combustion air. To avoid stratification of the gas stream,
the exhaust from the large cupolas requires a mixing aid, such as an in-
verted cone in the afterburner chamber, with burners angled to encourage
swirling.
Recent laboratory research indicates that the carbon monoxide content
of the flue gas may be reduced to one percent or less without an afterburner,
achieving control efficiencies greater than 90 percent.1*" The study suggests
injecting the proper amount of air at a point in the furnace below the charge
door where temperatures are at least 700°C (1300°F). More details may be
obtained from Reference 44.
7-4-5.4 Cost of Controls (Corrected to 1978)
Reported installed costs for afterburners were $12,000 to $20,000,«
depending mostly on the size of the cupola. Fuel for the natural gas after-
burners makes up the major part of the annual operating expense which is
estimated at $12,000 for an average sized foundry and may exceed $75,000
for the cupolas melting 45 metric tons (50 tons) per hour or more, assuming
32 kilojoules required per second per metric ton of metal melted (100,000
Btu/hr/ton), a 6,000 hour per year operation and natural gas purchased at
$2.50 per gigajoule (10* Btu). Afterburners installed on water-cooled
cupolas require more heat than refractory-lined cupolas since they maintain
a hotter contact area. Charging height also affects control costs. Increas-
ing charging height reduces off-gas temperatures. Consequently, larger
afterburner systems are required which use more fuel.1*3
7-87
-------
Charge door enclosures can decrease afterburner heat loads by reducing
the amount of cold air mixed into the stack gases. The true value of an
enclosure, however, depends on its reliability. A poorly constructed enclo-
sure may interfere with the charging mechanism, demanding constant repairs
and costly delays. Installing charge door enclosures may produce other
undesirable side effects. Reducing the amount of induced air may affect
afterburner efficiency by restricting combustion air under the necessary
volume and by inhibiting stack gas mixing. Increasing charging height in-
creases afterburner fuel costs because more fuel must be used to compensate
for the lower gas temperatures.43
7.4.5.5 Impact of Controls
Emissions -- Only four percent of the facilities operating in 1975
reportedly controlled carbon monoxide emissions.3U As mentioned previously,
use of an afterburner could reduce carbon monoxide emissions from 72.5 kilo-
grams per ton (145 Ib/ton) of metal charged to 4 to 5 kilograms per metric
ton (8-10 Ib/ton). This reduction would result in estimated national emis-
sions of 66,000 metric tons/yr (72,700 tons/yr), based on recent emissions
data.2
Environment -- The application of afterburners would reduce emissions
of hydrocarbons by combusting them along with the CO. The need for supple-
mental fuel introduces the possibility of S02 emissions from the fuel source.
As with any combustion device used as a control device, N0x emissions will be
increased.
Energy Requirements -- A range of burner duties was reoorted at 1.5
megajoules per second per cupola (5,200,000 Btu/hr/cupola) to 4.7 megajoules
7-88
-------
per second per cupola (16,000,000 Btu/hr/cupola).^ Typical cupola produc-
tion data may be obtained from Reference 44. The same source indicates
that fuel requirements are quite varied and suggests that in some cases the
CO combustion might be self-sustaining.^ Increasing charging height in-
creases energy requirements since more fuel must be used to compensate for
the lower temperature of the off-gases."3
7.4.6 Sintering Furnace
Sinter plants prepare small particles of iron ore and recycled flue
dust for blast furnace smelting by agglomerating them into larger particles
(sinter) suitable for blast furnace use. In 1976, over 40 sinter plants
were operating in the U.S., with a total production capacity of over 54 x
106 metric tons (60 x 106 tons)."
7-4.6.1 Process Description
The sintering process converts fine ore concentrates, coke fines, lime-
stone fines, blast furnace flue dust, and miscellaneous fines into an agglom-
erated product that is large enough and strong enough to be charged to a
blast furnace. The mixture is placed on a travelling grate. Combustion air
is added and the mixture is ignited. Temperatures of 1300-1500°C (2400-
2700°F) are maintained as the mixture burns and forms a fused mass. The sin-
ter product is then cooled, crushed and screened for use in the blast furnace.
7'4'6-2 Process Emission Sources and Factors
The major source of carbon monoxide emissions from sintering furnaces
is incomplete combustion of coke fines. The CO exhausts through the wind-
box, a compartment under the sinter bed which provides uniform distribution
7-89
-------
of combustion air as it passes through the sinter bed. Exhaust gas may
leave the windbox at rates of 120 to 250 Nm3/sec (250 x 103 to 530 x 103
scfm).45 Gas temperature is typically 200°C (400°F) or less.45 Uncontrolled
emissions from this point have been estimated by one source to be 22 kilo-
grams CO per metric ton of sinter product (44 lb/ton).5 Another more recent
source gave a higher estimate of 26 kg/metric ton (52 lb/ton).45
No techniques were reportedly used by U.S. sinter producers to control
carbon monoxide emissions. Only one state regulates carbon monoxide emis-
sions, and none of the affected sintering plants have properly complied with
its control regulations.46
A recent EPA estimate of carbon monoxide emissions from sinter produc-
tion indicated that total mass emissions were 624,700 metric tons (688,600
tons) in 1977.2 The amount of carbon monoxide actually emitted from each
plant depends on the coke content of the sinter charge, processing size, and
the completeness of combustion.45
7.4.6.3 Control Techniques
Little data were reported regarding carbon monoxide controls used in
sinter production. As was previously mentioned, no controls are reportedly
currently in use in the U.S. The only applicable control devices appear to
be afterburners or thermal incinerators, although these would be costly.
Catalytic incinerators are not feasible because trace amounts of phosphorus
in the exhaust gas would foul the catalyst.45
Carbon monoxide concentrations in the windbox exhaust gas can be
reduced by 90 percent if an incinerator (afterburner) combustion chamber
7-90
-------
temperature of 680-800°C (1250-1500°F) is maintained" The energy content of
the exhaust-gas is too low to maintain combustion at these temperatures,
so supplementary fuel is required.
A new development in sinter plant design may benefit the operation of
afterburners for CO control. This method produces a 65 to 75 percent !ower
exhaust-draft than conventional sintering processes." Waste gases also
leave the process around 340°C (650°F), which, combined with the lower blow
rate, could reduce the incineration energy load." Further details are
given in Reference 47.
7.4.6.4 Cost of Controls
Control costs for new plants have been taken from estimates for a ther-
mal incinerator with and without heat recuperation installed after gas clean-
ing equipment." Table 7-15 gives costs corrected to 1978 dollars. The
annualized capital cost is small compared to the annual operating cost,
largely because of the quantity of natural gas necessary to heat the enor-
mous exhaust gas flow. Control costs for existing facilities do not differ
markedly from those given in the table for new plants. As the table indi-
cates, the cost of afterburner use is high, even with heat recovery.
7.4.6.5 Impact of Controls
Mssions^eduction - If control systems with 90 percent carbon monox-
ide removal efficiency were applied industry-wide to the sintering industry,
annual emissions could be reduced by 562,000 metric tons (620,000 tons).
This reduction would result in total annual CO emissions of 62,500 metric
tons (68,900 tons) based on 1977 emissions data.
7-91
-------
oo
oo
Q_
OO
1^. OO
UJ OO
_1 O
CQ O
O
01
S-
CU
-p
c
•r~
CO
M-
h- 0
OO
O C '
CO O
-p
ea; o •
| r- A
0 i- -
\— -P
CU
E
•faO-
^— «. X— x
"c" ^ ^ c c c
3 2 3 2 S B
O r^" ^- «vf V£> LO
^- ^- LO CO CO LO
c: vo <>o ^o ^ *d~ *d~ ^ ^
Q ' ^—^ ^
S gSS S5S %
- ^ r-: r^ ^. ^. in fc
Q.
1
O
-p
E
-zr
0
h-'c*
5 °
M .,-
1— ^
5 "^
^ -kf*
>
_J
^t 1
Z3 OO
Z CZ5
z. o
M-
co
00
1^
OO
03
to r-» r» i— LO r-- "O
VD LO CO LO i— LO I
000 r- r- 0 £-
Q-
1
o
-p
0
•r—
•P
CU
CD C
z o
1— 1 «r-
\— r—
iii «L
UJ 3E.
D-
**»**
_J
< 1—
ZD OO
•sz. o
z o
^c
E
o
CU
CO
CO
E
a> cr> r- LO r- LO z
u_
r-^ W CXJ ,J CD r^ LO x-O
00 r- CM r- 0 ^0
0 0 ««*•
Cv
(V
CU
-p t-
O3 O
J 1 —
•) CO
" T3
0 C CO
> 0 O S-
0 0 03
1_ O l£) CU CU
u:
2 r-
^-^ o
>
03
T3
•^^
&.
OJ
4-
c
•r-
(/
1— 4-
o
i
LO LO
i ^> *'s_"* ,^ JL* ^^ ^^ 14— • • CU CO
^ 1 t ^ t 5 o^^i;
^ ^"^, \J v-/ ^** ^*^ ti^ ^ ^ «« i > . . LI
^clo"^ ctoco cos--+
^^*> W- *" w* T™ _. ... *— _i_\ /•
« s § § ^ s •§ § Iii
,1 1 1 i i I 1 § i If si
^ LO « ' • OLO " « (/),,CJ
i^jjr— r>.i— cu,_r^.n- CO-PT--I
•oloC.C^C' j-^-s^^. a3CUOO<
. O cULOcocTi •>—
- | co vo" ^ co « £ _• c
5 5 ^ g.
= ^ 5 o
J T™' * T —
J O3 CU -r- LO
a s- E •— •*
>. CU -r-
U O-+-> S- CU
3- E. cu a
= cu ai c: c
1) -P O S- CU
-> E Z3 S-
•P CU J3 CU
-> CU "O t- M-
3J i — T- CU CU
— -p CO -P OtL
C =5 CU M-
— i Q rv cj"
CU
. • . • O
M CO ^± LO i-
O
OO
7-92
-------
Invironment - The use of incineration devices for carbon monoxide emis-
sion control would increase emissions of nitrogen oxides (NO ) from sinter-
ing furnaces. This increase would be due to the large amount's of natural gas
necessary to maintain combustion of the exhaust gas.
If it becomes necessary to use fuel oil rather than natural gas as sup-
plementary incinerator fuel, sulfur oxides (SO,) emissions would increase
The amount of S0x emitted would depend on the sulfur content of the fuel oil
and the quantity of oil consumed.
iCSraOe^uJrements - An incinerator, operating at 90 percent efficiency
in a typical 900 metric ton/day (1,000 ton/day) sinter plant, would require
31.3 megajoules/sec (1.1 x 108 Btu/hr).« Assuming an exhaust gas flow rate
of 41 Nm3/sec (88,000 scfm), the energy required per normal cubic meter
would be 0.8 megajoules (21 Btu/scf). These amounts would be reduced if the
exhaust gas was preheated.
7.5 PETROLEUM REFINING
Petroleum refining is the process of converting crude oil into salable
products. Currently there are over 240 refineries in the United States pro-
cessing over 2.2 million cubic meters (14 Dillon barrels) of crude oil per
day.^ Refineries are located 1n 3g stfltes wuh t^ maJQ^ ^ ^^
capacity found near the coasts." Refinery sizes vary considerably from a
processing rate of 500 mVday (3,000 bbls/day) to more than 64,000 mVday
(400,000 bbls/day)>8
There are several significant sources of carbon monoxide from petroleum
refining. These are catalytic cracker regenerators, fluid coking, and sulfur
7-93
-------
plants. The following sections provide a brief process description of these
sources and also an assessment of carbon monoxide control technology for
the petroleum refining industry.
7.5.1 Catalytic Cracking
7.5.1.1 Process Description and Emissions
Catalysts are utilized by the refining industry in the operations of
cracking, reforming, hydrotreating, isomerization, hydrocracking, alkyla-
tion, and polymerization. Of these, cracking catalysts are the only types
which require regeneration frequently enough to produce significant amounts
of CO.49
Several types of catalytic cracking units are presently in operation;
fluid catalytic cracking (FCC) units and moving bed designs such as Thermofor
(TCC) and Houdriflow (HCC) cracking units. Table 7-16 gives a breakdown of
catalytic cracking capacity in the United States as of January 1978.
TABLE 7-16
DOMESTIC CATALYTIC CRACKING CAPACITY, 1978
UNIT
TYPE
FCC
TCC
HCC
FRESH
mVstream day
742
37
8
,
,
»
700
200
190
FEED
(bbl
(4
/stream day)
,670
(233
(51
,000)
,800)
,500)
% OF TOTAL
FEED CAPACITY
94
4
1
.2
•7
.0
# OF UNITS
IN OPERATION
123
17
3
Source: Reference 48
7-94
-------
Figure 7-21 shows a diagram of a typical FCC unit.50 Hot regenerated
catalyst, mixed with hydrocarbon feed, is transported into the reactor. The
reactor contains a bed of powdered catalyst which is kept in a fluidized
state by the flow of vaporized feed material and steam. Cracking of the
feed, which occurs in both the riser leading to the reactor and in the
fluidized bed, causes a deposit of coke to form on the catalyst particles.
A continuous stream of spent catalyst is withdrawn from the reactor. The
catalyst is steam stripped to remove hydrocarbons and is conveyed to the
regenerator by airflow. The hydrocarbon vapor from the reactor is fraction-
ated into a variety of products including light hydrocarbons, cracked gaso-
line, and fuel oil while a portion of the fractionator bottoms is recycled
to the reactor.50
Additional air is injected into the regenerator to burn off the coke
deposit and the regenerated catalyst is continuously returned to the
reactor. Heat added to the catalyst during coke burn-off furnishes much of
the required heat for the cracking reaction.51
Thermofor and Houdriflow catalytic cracking units utilize beaded or
pelleted catalysts. Regenerated catalyst and vaporized feed enter the top
of the reactor chamber and travel concurrently downward through the vessel.
The catalyst is purged with steam at the base of the reactor and travels by
gravity into the regenerator chamber. Combustion air is admitted at a
controlled rate to burn off coke deposits. From the bottom of the regene-
rator, the catalyst is conveyed by airlift to a surge hopper above the
reactor. A diagram of a typical Thermoflor catalytic cracking unit is
given in Figure 7-22.
7-95
-------
V) r—
IO -r-
u o
o o
o
OJ
O
c
O)
t.
O)
M-
O)
o
to
o
o
C\J
I
UJ
o:
£ o>
7-96
-------
o
o
o
oo
Q
QQ
I
o
—I
u.
o
UJ
00
OJ
I
CD
I—H
U_
7-97
-------
Carbon monoxide is formed when the coke deposits are burned off the
cracking catalyst during regeneration. EPA emission factors for conventional
uncontrolled fluid catalytic cracking units and moving bed units are 39.2
and 10.8 Kg CO/m3 of fresh feed (13,700 and 3,800 pounds CO per 1,000 bar-
rels of fresh feed), respectively.5 The exact amount of CO produced depends
on the method of regeneration employed by the refiner. EPA estimates of
total CO emissions from fluid and moving bed catalytic cracking operations
are given in Table 7-17.
TABLE 7-17
EPA ESTIMATED 1977 UNCONTROLLED CO EMISSIONS FROM
U.S. CATALYTIC CRACKING UNITS
UNIT TYPE
Fluid Catalytic
Cracker
Thermoflor Cata-
lytic Cracker
TOTAL U.S. CAPACITY
mVstream day
(bbl/stream day)
EMISSION FACTOR
Kg CO/m3 feed
(Ib CO/1000 bbl)
CATALYTIC CRACKING
CO EMISSIONS
metric tons/year
(tons/year)
742,700
(4,671 ,000)
37,170
(233,800)
39.2
(13,700)
10.8
(3,800)
2,385,000
(2,629,000)
40,400
(44,500)
2,425,100
(2,673,200)
Source: References 2, 5
With the advent of new catalysts, major design and operational changes
have been incorporated in fluid catalytic cracking unit operation. By con-
trast, no major changes in moving bed type units have been observed and
these units are being phased out.48 Several of the operational changes in
7-98
-------
fluid catalytic cracking units that directly affect CO emissions are dis-
cussed below.
Conventional Fluid Catalytic Cracker Operation -- Coke is removed from
cracking catalysts to restore their activity. Spent catalyst contains
roughly 6 percent coke while coke levels on the regenerated catalyst vary
from 0.2-0.3 percent.54 The amount of air supplied to the regenerator is
insufficient for complete combustion which results in flue gas CO concen-
trations of 5-10 percent.54 The oxygen level in the flue gas is low enough
so that combustion does not continue in the regenerator "dilute phase" where
no catalyst heat sink is available. Combustion in the dilute phase, called
afterburning, can result in damage to the catalyst, the cyclones, and other
regenerator equipment due to high temperatures. To avoid equipment damage,
the regenerator is operated below 620°C (1150°F).53,54,55
High Temperature Regeneration — Zeolite catalysts first appeared on
the market in the mid-1960's. The major features of these catalysts are
summarized below:54*56
1) naphthenes and paraffins are cracked rapidly with excellent
selectivity,
2) aromatic nuclei crack slowly with poor selectivity,
3) high hydrogen transfer rates are observed,
4) the rate of cracking is relatively unaffected by boiling range, and
5) catalyst activity is adversely affected by coke deposits which
limit zeolite availability.
The use of zeolite catalysts has accelerated the trend to more fully
regenerate these coke sensitive catalysts as even slight improvements in
7-99
-------
regeneration can provide substantial yield benefits. Very low carbon on
regenerated catalyst (CRC) levels have been achieved using a technique
called high temperature regeneration (HTR). The key to this process is
complete conversion of coke to C02 within the regenerator. This situation
is quite different from that of conventional regeneration where conversion
of CO to C02 is minimized.55 High temperature regeneration can be utilized
in new units, or applied as a retrofit to existing units. The major features
of high temperature regeneration are:
1) Extremely low levels of coke on the regenerated catalyst are pos-
sible. Typical values are 0.05-0.1 percent coke. Amoco Oil Company
reported regenerated catalyst levels of 0.01 percent with their UltraCat
regeneration technique.50'55
2) CO emission levels of 500 ppm in the regenerator flue gas can be
obtained. This level is sufficiently low to meet federal New Source Per-
formance Standards and most state emissions regulations.55,57,53,59
3) Complete regeneration increases catalyst activity which means
a lower catalyst-to-oil ratio is possible. Thus, unit capacity can be
increased if bottlenecks are removed from the rest of the process50,5^,55,56,5?
4) Temperatures in the regenerator vary from 540-730°C (1000-1350°F).
This is 40-65°C (100-150°F) higher than conventional regeneration. Since
CO afterburn is possible, flue gas temperatures in the dilute-phase can be
several hundred degrees higher than the dense-bed temperature.51'55'57'58
5) The extremely active catalyst produced from HTR is most effectively
used in a short contact time riser cracking reactor. The advantage of
riser cracking over bed cracking lies in avoiding secondary reactions such
as the recracking of gasoline.54*57
7-100
-------
6) Recovery of thermal energy in waste heat boilers.
7) increased catalyst selectivity and the use of riser cracking can
result in a 20 to 30 percent reduction in the amount of coke produced.
Therefore, the increase in combustion air required to completely burn CO
and coke can be offset in some cases by lower coke production such that
overall combustion air usage can remain essentially constant.50
The operating conditions for conventional fluid catalytic crackers and
units using high temperature regeneration are compared in Table 7-18.
TABLE 7-18
TYPICAL OPERATING CONDITIONS FOR FLUID CATALYTIC CRACKING
Reactor Temperature, °C (°F) ^70 - 550 (885 - 1025)
Regenerator Temperature, °C (°F)
Conventional Regeneration 5/40 - 590 (1000 - 1100)
High Temperature Regeneration 590 - 730 (1]00 - 1350)
Coke Content of Spent Catalyst, Wt %
Conventional Regeneration 6
High Temperature Regeneration 5
Coke Content on Regenerated Catalyst, Wt %
Conventional Regeneration 0.2 - 0.3
High Temperature Regeneration 0.01 - 0.1
Source: References 50, 57
Existing fluid catalytic cracking units may be revamped to incorporate
high temperature regeneration. The required changes to convert to high
temperature regeneration depend on the design of the unit and the desired
coke content on the regenerated catalyst. To withstand higher regenerator
temperatures, steel components within the regenerator may require replace-
ment by components made with more heat resistant materials such as chromium-
7-101
-------
nickel alloy stainless steel. Other modifications may include an improved
combustion air distribution system or in the installation of a riser crack-
ing reactor. In general, switching to high temperature regeneration in-
creases the capacity of the process and some modifications in downstream
equipment may be required to remove bottlenecks.58'59
Combustion Promotion Catalysts -- The most recent development in fluid
catalytic cracking technology is the use of "promotion" catalysts to com-
pletely convert CO to C02.59 The first type to become available (1975) was
a fluid catalytic cracker catalyst modified with a small concentration of
noble metal promoting agent.59 In 1977, a number of manufacturers began
producing a solid promoter. This powder is mixed with make-up catalyst,
roughly 0.5-5 kg/metric ton (1-10 Ib/ton) of fresh catalyst. Liquid pro-
moters, injected directly into the regenerator, are also available.59
The advantage of using combustion promoters is that CO is converted to
C02 within the dense-phase of the regenerator. This avoids the problem of
CO afterburn in the regenerator dilute phase, a common problem in units
using high temperature regeneration. Thus, in units where temperature limi-
tations prohibit the use of high temperature regeneration, CO emissions
below 500 ppm can be obtained using combustion promoting catalysts.55
Essentially complete conversion of CO can be obtained with bed temperatures
of 620-650°C (1150-1200°F).55>57 However, regeneration of the catalyst is
not quite as effective at the lower temperature and selectivity of the
catalyst is slightly poorer in that more coke is produced.57 The thermal
energy from the regenerator is usually recovered through steam production.
7-102
-------
7.5.1.2 Control Techniques
There are a variety of ways to control CO emissions in the flue gas of
catalyst regenerators. The most widely used method is burning the flue gas
in a carbon monoxide waste-heat boiler. In addition to reducing CO emis-
sions, valuable thermal energy is recovered from the flue gas. The CO
boiler produces steam from sensible heat in the flue gas as well as from hea
produced by CO combustion. Carbon monoxide emissions from a properly
operated CO boiler can be very low. In one study in which five CO boilers
were sampled, CO levels in the flue gas of 0, 0, 5, 10 and 25 ppm were
obtained.57 Typical fluid catalytic cracker regenerator flue gas composi-
tions before and after incineration in a CO boiler are listed in Table 7-19.
Thermoflor and Houdriflow catalytic cracking unit regenerators produce
significantly less flue gas than fluid catalytic cracking unit regenerators
and may not justify the installation of a CO boiler. Flue gas from these
sources can be incinerated in a process heater.57
Another method of limiting CO emissions described earlier is high
temperature regeneration. High temperature regeneration can be used with
conventional catalysts or with combustion promotion techniques. CO emission
levels of less than 500 ppm have been reported for fluid catalytic cracking
units using this type of regeneration.57*58
Exxon Corporation has reported using a medium temperature regeneration
technique in their fluid catalytic cracking units to obtain the benefits
from improved catalyst selectivity without requiring the replacement of the
regenerator internals. However, lower CO content in the flue gas means addi-
tional auxiliary fuel consumption in their CO boilers.57'59
7-103
-------
TABLE 7-19
EMISSION RATES FROM FCC UNIT REGENERATORS, BEFORE AND AFTER CO BOILER
EMISSIONS
S02, ppm
N0x (as N02), ppm
CO, % Vol.
C02, % Vol.
H20, % Vol.
N2, % Vol.
Hydrocarbons, ppm
Ammonia, ppm
Aldehydes, ppm
Cyanides, ppm
Particulates, grains/scf
g/m3
Temperature, C
°F
BEFORE CO BOILER
140 - 3300
8 - 394
7.2 - 12.0
10.5 - 11.3
13-9 - 26.3
78.5 - 80.3
98 - 1213
0 - 675
3 - 130
0.19 - 0.94
0.08 - 1.39
0.18 - 3.18
538 - 645
1000 - 1200
AFTER CO BOILER'
Up to 2700
Up to 500
0-14 ppm
11 .2 - 14.0
13-4 - 23.9
82.0 - 84.2
0.017 - 1.03
0.039 - 2.36
250 - 440
485 - 820
Emissions after the CO boiler will be affected by the operating conditions
and the type of auxiliary fuel.
Source: Reference 57
Industry acceptance of high temperature regeneration and combustion
promotion techniques has been very good given the short length of time that
these methods have been available.55 The types of regeneration methods cur-
rently employed by U.S. refiners are listed in Table 7-20.
7.5.1.3 Cost of Controls
Although the cost of new CO boilers is quite high, the associated fuel
savings can make this an attractive investment, particularly if additional
steam generating capacity can be utilized. A typical CO boiler, operating
with a conventional fluid catalytic cracking unit, can recover approximately
7-104
-------
400 megajoules/m3 (60,000 Btu/bbl) of fresh fluid catalytic cracker feed.20
Information on investment and operating costs for CO boilers is given in
Section 6.2. In all but small refineries, the cost of CO boilers can be
recovered in a few years.
The cost of converting a conventional fluid catalytic cracking unit to
high temperature regeneration or promotion catalysts can vary over a wide
range depending on the original design of the unit and the degree of regene-
ration desired. Insufficient information was available, however, to ade-
quately develop capital and annualized costs for these control methods.
TABLE 7-20
CURRENT DOMESTIC FLUID CATALYTIC CRACKER REGENERATION
TECHNIQUES (August 1978)
REGENERATION TECHNIQUE
Conventional regeneration
High temperature
regenerat ion
Combustion promoting
catalysts
Combustion promotion,
separate from catalyst
% OF ALL FLUID
CATALYTIC CRACKING
UNITS THAT USE
THIS TECHNIQUE
53
26
10
REMARKS
Most units have CO boilers
May be used in conjunction
wi th a CO boiler
May be used in conjunction
wi th a CO boiler
May be used in conjunction
wi th a CO boiler
Source: Reference 60
7-105
-------
7.5.1.4 Impact of Controls
Emissions Impact -- CO emission levels from fluid catalytic cracker
regenerators are summarized in Table 7-21.
TABLE 7-21
EFFECT OF CONTROLS ON CO EMISSIONS FROM FCC REGENERATORS
CONTROL TECHNIQUE
Conventional regeneration,
(uncontrolled)
Conventional regeneration,
(CO boiler)
High temperature regeneration
or combustion promotion
TYPICAL EMISSIONS LEVEL
5-10% CO in regenerator flue gas;
AP-42 emission factor 39.2 kg/m3
(13,700 Ib CO/1000 bbl) feed
<50 ppm in CO boiler flue gas
200-2000 ppm CO in regenerator flue
gas; <500 ppm CO can usually be
obtained
Source: Reference 57
CO emissions from a properly operated CO boiler are nearly zero. This
represents a control efficiency of greater than 99.5 percent.20 The emis-
sions from units utilizing high temperature regeneration or combustion pro-
motion catalysts are roughly one percent of those from conventional units of
the same feed capacity.20 Assuming that roughly 50 percent of all FCC units
use high temperature regeneration or combustion promotion and that all the
remaining catalytic cracking units were controlled by CO boilers or other
forms of CO incineration, annual CO emissions from this source could be
reduced to an estimated 47,800 metric tons per year (52,500 tons/yr).
Environmental Impact -- Hydrocarbon emissions are reduced by the appli-
cation of CO control techniques. Hydrocarbon levels of less than .10 ppm
7-106
-------
have been reported in the flue gas of high temperature regenerators as well
as from CO boilers.57 The combustion reactions seem to be rate-limited by
the combustion of carbon monoxide, not the combustion of hydrocarbons.57
Temperatures within the CO boiler are above 980°C (1800°F) in order to
.promote complete combustion of carbon monoxide.57 This is considerably
hotter than the maximum temperatures observed in high temperature regenera-
tion. Hence, N0x emissions could be higher for fluid catalytic cracking
units that utilize CO boilers due to production of thermal NO . Also, nitro-
X
gen compounds present in the auxiliary fuel supply can also contribute to
N0x emissions. Typical N0x emission levels from CO boilers are 100-300 ppm.
N0x emissions from high temperature regeneration units are somewhat lower.57
The amount of sulfur oxides emitted from catalytic cracking units depends
on the sulfur content of the feed and the amount of coke burned. Adding a
CO boiler to an existing unit will result in increased SO production due to
X
sulfur compounds in the auxiliary fuel. A unit utilizing high temperature
regeneration produces a more selective catalyst which can reduce coke yield.
In addition to reducing total S0x emissions, lower coke yield can result in
reduced particulate emissions as well.57
Energy Impact — The flue gas from uncontrolled fluid catalytic crack-
ing units contains from 5-10 percent CO which represents a substantial energy
loss if released to the atmosphere.20 This energy is recovered in a CO
boiler by producing steam. Often, the entire cost of a CO boiler can be
justified on the basis of steam production alone.
Energy recovery from high temperature regeneration is about the same
as for CO boilers, estimated at 400 megajoules/m3 (60,000 Btu/bbl) of fresh
7-107
-------
feed.20 This energy manifests itself in the increased yield of valuable
liquid products and increased waste heat boiler steam production resulting
from higher flue gas temperatures.20
7.5.2 Fluid Coking
7.5.2.1 Process Description and Emissions
Coking processes convert residual oils, tars and resins into lighter,
more valuable liquid products and coke. Two processes, delayed and fluid
coking, account for most of the domestic petroleum coke production. However,
only fluid coking results in a discharge of carbon monoxide.k8 There are
only five fluid cokers currently in operation in the U.S.48
Fluid coking is a continuous thermal cracking process that involves the
injection of feed into a fluidized bed of hot coke particles. The hot oil
is cracked and carbon is deposited in thin layers on the coke particles.
The bed is kept fluidized by the injection of steam. The coke particles
travel from the reaction to a burning chamber where approximately 25 percent
of the coke is burned to provide process heat. The heated coke particles
(600-650°C, 1110-1200°F) circulate back to the reactor section. Since more
coke is produced in the reactor than is burned, a coke product stream is
withdrawn. The coke produced in fluid coking is unsuitable for most indus-
trial uses. Consequently, most of this coke is sold as fuel or is used in
the refinery to produce steam.52'61 A diagram of a fluid coking unit is
given in Figure 7-23.
Typical operating conditions for fluid coking are presented in Table
7-22.
7-108
-------
CM
LO
O)
O
c
0)
S-
cu
M-
OJ
O)
u
S-.
^
O
00
00
00
LU
C_)
O
o:
Q.
CD
O
CJ)
CO
C\J
I
7-109
-------
TABLE 7-22
TYPICAL FLUID COKER OPERATING CONDITIONS
REACTOR BURNER
Temperature, °C
°F
Pressure, kilopascals
psig
Bed Velocity, m/sec
ft/sec
Bed Depth, m
510
950
170
10
.30 - .91
1 - 3
9.1 - 15
30 - 50
620
1150
180
11
.61 - .91
2 - 3
3.0 - k.6
10 - 15
Source: Reference 62
Carbon monoxide is formed in the burner section where coke is burned
in limited air. It is estimated that CO emissions average 86 kg/m3 (30
pounds per barrel) of fresh feed.49 The energy content of the flue gas can
be recovered by burning in a CO boiler, or, if the CO content is high enough,
the flue gas could be used to fire a process heater. All five domestic
fluid cokers presently in operation utilize one or the other of these
methods.
The most recent advancement in coking processes is Flexi coking, developed
by Exxon Research and Engineering.20 Flexicoking integrates conventional
fluid coking with coke gasification. The gaseous products are referred to
as coke gas. The coke gas contains considerable quantities of carbon monox-
ide, hydrogen, carbon dioxide and water vapor and it may be substituted for
refinery fuel gas or natural gas to fire process heaters or boilers. No
commercial Flexicokers have yet been installed in the United States.20
7-110
-------
7.5.2.2 Control Techniques
Control techniques for CO emissions from fluid cokers consist of burn-
ing CO in either a CO boiler or a process heater. As summarized in Table
7-23, all domestic fluid coking capacity is controlled by one or the other
of these methods.
TABLE 7-23
CO CONTROLS ON DOMESTIC FLUID COKING UNITS
REFINERY
LOCATION
Exxon
Ben ic ia, Ca.
Bill ings, Mont.
Tosco
Avon, Ca.
Bakersfield, Ca
Getty
Delaware City,
Del .
bbl/calendar day
Source: Reference
REFINERY CRUDE
CAPACITY,
mVstream day
(bbl/stream day)
16,200
(102,000)
7,300
(46,000)
20,000
(126,000)
6,360
(40,000)
23,850
(150,000)
FLUID COKING
CAPACITY,
m3/stream day
(bbl/stream day)
3,910
(24,600)
1,190
(7,500)
6,680
(42,000)
1 ,110
(7,000)
7,000
(44,000)
CO CONTROL METHOD
Flue gas used to fi re
crude pipest ills
CO boiler
CO boiler
CO boiler
CO boiler
7-111
-------
7-5.2.3 Cost of Controls
Chapter 6 contains a detailed presentation of capital and annualized
costs for CO boilers. These costs are presented graphically in terms of
dollars per normal cubic meter per second ($/scfm) with several curves per
graph showing the effect of the heating value of the gas on annualized costs
Therefore, given a representative flow rate and heating value of the burner
off-gas, control costs for CO boilers can be estimated.
The flow rate and heating value of the off-gas was estimated, based
on the following assumptions:
1) coker feed density - 1.0 g/cm3 (360 lb/bbl)62
2) coke production - 28 wt % of fresh feed62
3) coke burnoff rate - 25% of total coke production62
4) CO production rates - 85 kg/m3 (30 lb/bbl) of fresh feed50
Using these values, the off-gas flow rate is estimated at 534 cubic
meters of gas per cubic meter of fresh coker feed (3000 scf/bbl feed). The
heating value of the gas is 1.61 megajoules per normal cubic meter (43 Btu/
scf).
7-5.2.4 Impact of Controls
Emissions — At the present time, CO emissions from all five domestic
fluid coking units are controlled, either by CO boilers or by incineration
in a process heater.61
Environmental Impact -- The application of CO boilers or other methods
of gas incineration to control CO emissions will have both positive and
negative impacts with respect to other pollutant discharges. The positive
impact includes the combustion of some of the particulates which escape
7-112
-------
from the burner section cycle. The negative impacts include increased
levels of S02 and NO .
^ X
Increased S02 emissions will occur if supplemental fuel is required.
Most of the sulfur in this fuel will be discharged as S02.
Temperatures within the CO boilers are above 980°C (1800°F) in order
to promote complete combustion of carbon monoxide. At this temperature, NO
can be formed from elemental nitrogen and oxygen which are present during
the combustion process. In addition, nitrogen compounds in the burner off-
gas or the supplemental fuel can also form N0x. Typical N0x emission levels
from CO boilers are 100-300 ppm.57
Energy Impact - The burner off-gas from fluid coking units contains
substantial quantities of CO which would represent a considerable energy
loss if released to the atmosphere. This energy is recovered in a CO
boiler by producing steam. Often the entire cost of a CO boiler can be
justified on the basis of steam production alone.
7.5.3 Sulfur Plants
7.5.3.1 Process Description and Emissions
Claus sulfur plants — Many refineries utilize a Claus sulfur plant to
recover elemental sulfur from H2S laden gas streams produced within the
refinery. The first step in the process is the oxidation of part of the
H2S stream to S02
-------
As indicated in Figure 7-24, several flow schemes are available for
the Claus process.6t+ In the most common type, the "once through" design,
the H2S feed stream is burned in a limited amount of air to convert one-
third of the H2S to S02. The Claus reaction is initiated in the combustion
step and continues in the catalytic reactors. After each step, sulfur is
condensed and is removed as a liquid.
In the bypass or split-flow designs, only one-third of the feed stream
is burned. This stream is burned more completely and most of the H2S is
converted to S02. No sulfur is formed in the combustion step using this
flow scheme. The hot gas from the furnace is cooled and combined with the
bypass stream which then enters the reactor section. The split flow scheme
is useful when the H2S content of the feed is below 50 percent.63 Addi-
tional fuel is necessary to support stable combustion at lower H2S concen-
trations and the split flow design reduces fuel consumption by reducing
the amount of inert gas which must be heated. Most refinery sulfur plant
feed streams contain H2S concentrations greater than 50 percent and the once-
through design is the most prevalent.63*65
Carbon monoxide is formed in the combustion furnace from small amounts
of hydrocarbon and carbon dioxide present in the feed stream. Since only
partial combustion of the H2S is desired, not enough oxygen is supplied to
convert all the CO formed to C02. CO produced in the combustion process
proceeds through the reactor-condenser section and ends up in the tail gas.
The composition of the tail gas from a typical Claus unit is given in
Table 7-24.
7-114
-------
Waste Heat
Boiler
Catalytic
Reactors
Condensers
Tan
STRAIGHT THROUGH CLAUS PROCESS
Sulphur Pit
Liquid Sulphur
Waste Heat
Boiler
Catalytic
Reactors
SPLIT FLOW CLAUS PROCESS
fan~)
Sulphur Pit
Liquid Sulphur
Source: Reference 64
FIGURE 7-24. CLAUS SULFUR PLANTS
7-115
-------
TABLE 7-24
TYPICAL CLAUS TAIL GAS COMPOSITIONS'
COMPONENT
SOUR GAS FEED
VOLUME %
CLAUS TAIL GAS
VOLUME %
THERMALLY INCINERATED
TAIL GAS
VOLUME %
H2S
S02
Sg vapor
SB aerosol
COS
CS2
CO
C02
02
N2
H2
H20
H.C.
Temperature, C
°F
Pressure
Ki lopascal s
Psig
Total Gas Volume
89.9
0.0
0.0
0.0
0.0
0.0
0.0
4.6
0.0
0.0
0.0
5.5
0.0
100.0
40
104
150
6.6
--
0.85
0.42
0.10 as Sj
0.30 as $!
0.05
0.05
0.22
2.37
0.00
61 .04
1 .60
33.00
0.00
100.00
140
284
110
1.5
3.0 x feed
gas volume
0.001
0.89
0.00
0.00
0.02
0.01
0.10
1.45
7.39
71.07
0.50
18.57
0.00
100.00
400
752
100
0
5.8 x feed
gas volume
Two catalytic reactors - overall efficiency of 94%
Gas volumes compared at standard conditions
Source: Reference 66
7-116
-------
The tail gas still contains substantial quantities of H2S which can
pose a serious health hazard. Consequently, most refiners incinerate the
tail gas before discharge to the atmosphere.63 Incineration converts all
sulfur values to S02 and simultaneously converts CO to C02.
Tail Gas Cleaning -- Claus plant sulfur removal efficiency depends on
many factors including the concentration of H2S in the feed, the number of
reactor stages, and the level of impurities such as C02, water vapor, and
hydrocarbons in the feed. Claus plant efficiency can range from 90 to 97
percent; however, increasingly strict state and Federal emission regulations
can require up to 99.9 percent sulfur removal.67 To achieve this efficiency
tail gas cleaning is required.
Many different processes have been developed which can reduce the sulfur
level in the tail gas. Several of these use incinerated tail gas as feed.
Incineration converts sulfur species such as elemental sulfur, H2S, COS, and
CS2 into S02, which is removed in the tail gas cleaning unit. The Wellman
S02 recovery process, Shell's flue gas desulfurization process, and the SNPA
Wet Contact Aid process are of this type.67
Carbonyl sulfide and carbon disulfide are produced from side reactions
occurring in the thermal reactor section of the Claus plant.64 Even with
improved Claus unit catalysts, these contaminants are present in the tail
gas and account for a sizable portion of the total sulfur loss. As an
alternative to incineration, followed by the so-called "oxidation-scrubbing"
systems, several tail gas cleaning processes have been designed which reduce
all sulfur compounds to H2S. Examples of this type of process are the
7-117
-------
Beavon, SCOT, and Clean Air processes.67 The reaction takes place at 260-
320°C (500-610°F) over a cobalt/molybdenum catalyst with H2, H20, and CO
as reducing agents. Carbonyl sulfide and carbon disulfide are removed by
hydrolysis;
COS + CS2 + 3 H20 £ 3 H2S + 2 C02
while S02 is hydrogenated:
S02 + 2 H2 J S + 2 H20
S + H2 $ H2S
The same catalyst is effective for hydrolysis of carbon monoxide via
the water-gas shift reaction:
CO + H20 J H2 + C02
The hydrogen produced here, together with that initially present in tail
gas, is usually sufficient to convert all sulfur species to H2S.68 If not,
additional hydrogen can be supplied from other units or from fuel rich com-
bustion of natural gas ahead of the hydrogenation reactor. The H2S is then
removed using conventional H2S removal techniques. For example, the Beavon
process consists of a catalytic hydrogenation reactor followed by a Stretford
H2S removal system.
Carbon monoxide emission levels can be reduced using these "reduction-
scrubbing" processes. Actual sampling data was limited; however, the devel-
oper of the Beavon process reported that tail gas CO levels of a few hundred
ppm were typical.69
Tail gas compositions for a representative Beavon unit are given in
Table 7-25.
7-118
-------
TABLE 7-25
REPRESENTATIVE TAIL GAS COMPOSITIONS FOR THE
BEAVON SULFUR REMOVAL PROCESS
CLAUS TAIL GAS BEAVON PROCESS
COMPQNENT 'NPUT, VOL % TAIL GAS, VOL %
H2 2.5 Varies
CO i.o o.2
C°2 10.0 14.0
N2 56.2 80.8
H20 26.0 5.0
5 -07 o.O
H2S 2.0 o.O
S°2 1.0 o.O
COS °-3 Less than 250 ppm
CS2 0.3 0.0
Source: Reference 69
7-119
-------
Two additional tail gas cleanup methods, the IFF and the Sulfreen pro-
cesses, are continuations of the Claus reaction. Carbonyl sulfide and car-
bon disulfide are not removed by these processes and the tail gas usually
requires incineration.67
7.5.3.2 Control Techniques
Carbon monoxide emissions from refinery sulfur plants can be reduced
by incinerating the tail gas. The incinerator is a refractory lined vessel
with one or more burners. Temperatures in excess of 650°C (1200 f) with
residence times of 0.5-0.6 seconds were recommended by several manufacturers
to assure complete conversion of H2S to S02.68 An auxiliary fuel supply
such as natural gas or fuel oil provides the heat necessary for incineration
as the heating value of the tail gas, estimated from the data in Table 7-24,
is only 0.37-0.75 MJ/m3 (10-20 Btu/scf).66 Excess air levels of 20 - 30
percent are used and the flue gas is vented through a tall stack to disperse
S02.
The recommended temperature and residence time given above is effective
for conversion of H2S to S02. However, higher temperatures, in the range of
870-980°C (1600-1800°F), are required to oxidize CO to C02. Therefore,
additional auxiliary fuel may be necessary to provide a sufficiently high
temperature for complete CO oxidation.
7-120
-------
The primary motivation for installing an incinerator is to remove H2S,
not carbon monoxide. Although other methods of gas incineration such as
flares or existing process heater could reduce CO emissions, these methods
are not recommended for H2S disposal because of inadequate gas residence
time, insufficient stack height, or safety considerations.
Some tail gas treating processes have the capacity to reduce CO levels
in the tail gas (see Table 7-25). These "reduction-scrubbing" systems
utilize CO in the tail gas as a reactant in the catalytic reduction of all
sulfur species to H2S.
7.5.3.3 Cost of Controls
A detailed presentation of annualized costs for waste gas incinerators
is given in Chapter 6. Capital costs are based on the volume of gas that
requires incineration. An estimate of the tail gas volume, calculated from
the data in Table 7-24, is 2.5 cubic meters per kilogram of sulfur recovered
(40 scf/lb sulfur). The heating value of this gas, also estimated from the
data in Table 7-24 is 0.37-0.75 megajoules/m3 (10-20 Btu/ft3). Using this
information and the information in Chapter 6, annualized costs can be
estimated for Claus plant tail gas incinerators.
7-5.3.4 Impact of Controls
Emissions- " Uncontrolled CO emissions from refinery sulfur plants have
been estimated at 28800 metric tons per year (31700 tons/yr).™ Based on a
total refinery sulfur plant capacity of 8500 metric tons per day (9300 tons/
day),™ and a tail gas production estimate (calculated from data in Table
7-24) of 2.5 cubic meters per kilogram of sulfur (40 scf/lb) the average
7-121
-------
level of CO in the tail gas was estimated at 0.3 volume percent. Although
only a limited amount of actual sampling data were located, typical CO
levels from incinerated tail gas averaged approximately 0.1 volume percent.
Assuming this level of CO in the incinerated tail gas with the incinerator
fired at 25 percent excess air, controlled CO emissions from all refinery
sulfur plants would be 11000 metric tons per year (12100 tons/year). This
represents a reduction in total CO emissions of 62 percent. Further reduc-
tion in total CO emissions could be obtained by operating the incinerators
at higher temperatures, although the benefits obtained would have to be
balanced against higher fuel consumption and the possibility of increased
N0x emissions.
Environmental Impact -- The primary effect of Claus tail gas incinera-
tion is to convert all sulfur species to S02 before discharge to the
atmosphere. Although actual sulfur emissions are not reduced, S02 is the
least toxic of the sulfur compounds produced.
As is the case with all combustion operations, additional pollutants
may be generated. Sulfur in the auxiliary fuel will oxidize to S02, adding
to total sulfur emissions while nitrogen in the fuel, the tail gas, and
the combustion air may be converted to N0x. N0x emission levels of 40 - 50
ppm have been reported from non-catalytic hydrocarbon vapor incinerators.20
Claus incinerators are operated at higher temperatures, however, and N0x
emissions may be slightly higher.
Energy Requirements -- Auxiliary fuel must be used when incinerating
Claus-unit tail gas. Part of the cost of this fuel can be offset by
recovering heat from the incinerator flue gas. This heat may be utilized
7-122
-------
to preheat the incinerator feed or to generate steam. Heat recovery from
the incinerator flue gas offers a way to reduce incinerator energy require-
ments at the expense of increased equipment costs. However, care must be
taken in the design and operation of incinerators utilizing heat recovery
to avoid corrosion problems which would occur at temperatures below the
dew point of the flue gas.
7.6 PRIMARY ALUMINUM INDUSTRY
Aluminum is produced from alumina (A1203) which is contained in its
hydrated form in bauxite ore. Alumina, after it has been separated from
the ore, is reduced electrolytically to form aluminum metal.
Significant emissions of CO to the atmosphere result from the reduction
process. Estimates for 1976 were 220,000 metric tons CO emitted per year
(242,000 tons/yr).7i No control methods expressly designed for CO control
are currently in use.
This section contains a discussion of electrolytic reduction plant
operation; CO emission sources; control methods for those sources; and cost,
environmental impact, and energy requirements for possible control methods.
7.6.1 Process Description
The production of aluminum metal from alumina by electrolytic reduction
is shown diagramatically in Figure 7-25. Alumina is decomposed in reduction
plants by a continuous current flowing through an electrolytic cell which
contains alumina dissolved in molten cryolite (Na3AlF6). The aluminum metal
is deposited at the cathode, while oxygen passes to the carbon anode. The
reaction between carbon and oxygen at the anode is one major source of CO
emissions in the aluminum industry.
7-123
-------
Petroleum Coke
Pitch
r
I
T
1
Calcined cocnercial
Alumina
Electrical
Supply
(Direct current)
I
Air
Anthracite Pitch
Aluminum (pig,
billet, ingot,
rod)
Dry-Process Any
Solids Wet-Scrubber
I returned to liquor to
cell treatment
Spent Potliners (to cryolite recovery
or disposal)
FIGURE 7-25. MAJOR PROCESSING PHASES IN PRIMARY ALUMINUM REDUCTION
7-124
-------
A large number of reduction cells are usually linked together electri-
cally in parallel to form a potline, the basic production unit of the
reduction plant. Potline configuration, cell types, and cell dimensions
vary according to the design and capacity of the individual aluminum reduc-
tion plants. A typical late design potline may consist of 180 cells. Such
a potline operating at 83,000 kW would produce approximately 125 megagrams
(275,000 pounds) of aluminum per day.72
The reduction cell, or pot, is a strongly reinforced steel box, lined
with heat insulation and either prebaked carbon blocks or a rammed monolithic
carbon liner inside the insulation. The carbon liner forms the cathode of
the electrolytic cell and provides high electrical conductivity and good cor-
rosion resistance to the highly reactive molten electrolyte. The carbon
lining contains steel electric current collector bars that extend through
the sides of the steel shell. The collector bars are connected to a ring
collector bus which is connected to the main bus. The main bus is usually
made of aluminum bars and serves as the electrical connection for a line of
cells connected in parallel.
The anode, also made of carbon, is suspended over the steel pot shell
and is immersed in the molten electrolyte. It is connected to the main
bus system through flexible conductors.
Reduction cells are of two basic types, the prebake cell using multiple
prebaked carbon anodes, and the Soderberg cell using one large self-baking
anode.
7-125
-------
7.6.1.1 Prebake Anode Cell
Modern prebake cells use a number of anodes suspended in the electro-
lyte. The anodes are press-formed or vibration molded from a carbon paste
mixture of coke and pitch. They are then baked in anode bake furnaces,
sometimes termed "ring furnaces." The off gases from the anode bake fur-
naces are one source of CO emissions in the prebaked anode plants.72
The anode bake furnaces are sunken pits with surrounding brick flues.
Anodes are packed into the pits with a blanket of coke or anthracite filling
the space between the anode blocks and the pit walls. A blanket of calcined
petroleum coke fills the top of each pit, 25 to 30 cm (10-12 in) above the
top layer of anodes.
The pits are heated with natural gas or oil fired manifolded burners
for a period of about 40 hours. The flue system of the furnace is arranged
so that hot gas from the pits being fired is drawn through the next section
of pits to gradually preheat the next batch of anodes. The anodes are fired
to approximately 1200°C (2200°F), and the cycle of placing green anodes,
preheating, firing, cooling, and removal is approximately 28 days. The
baked anodes are stripped from the furnace pits by means of an overhead
crane which may also be used for loading and removing the coke pit packing.
The ring-type furnace flues are under draft. Most volatile hydrocarbon
materials released from the anodes during the baking cycle are drawn into the
flue with the combustion gases. These hydrocarbons burn and are a source of
CO along with any CO formed as a result of incomplete fuel combustion.72
Flue gases may be passed through scrubbers and perhaps electrostatic precipi-
tators before exhausting to a stack. Typically, there are no special con-
trols for CO removal.
7-126
-------
After baking in a ring furnace, the baked anode blocks are moved to a
rodding plant where steel stub electrodes are bonded into preformed holes
in the blocks. Completed anode assemblies are delivered to the potlines,
ready for the replacement of consumed anodes. Figure 7-26 shows a sectional
view of a typical prebake reduction cell with a hood for cell effluent col-
lection. The newer design prebake cells use up to twenty-six anode assemblies
per cell.
The sacrificial carbon anodes are replaced periodically by new anode
assemblies, the total operating time being dependent on the size of the
anode blocks and the amperage of the potline.
7.6.1.2 Soderberg Cells
There are two types of Soderberg cells, each having a single large car-
bon anode, but differing in the method of anode bus connection to the anode
mass. They are termed the horizontal spike suspension (HSS) Soderberg and
the vertical spike suspension (VSS) Soderberg. The HSS Soderberg cells are
being completely modified at all three operating plants. No information is
available on the new process at this time. In both, the anode material is
a paste of carbon and pitch which is fed periodically into the open top of
a rectangular steel compartment and baked by the heat of the cell to a solid
coherent mass as it moves down the casing. This casing is mounted on the
steel superstructure of the cell and is raised or lowered by means of powered
jacks. Paste is added to the upper section to replenish the anode as it is
consumed. Figure 7-27 shows a schematic design of the HSS Soderberg cell
design.
7-127
-------
ALUMINA (ORE) BIN
ANODE ROD
CRUST BREAKER
RISER BUS TO
NEXT CELL
ANODE BUS
rr^
s ^
CARBON
ANODE
yfr-.v
ALUMINA INSULATION
STEEL CRADLE
SIDE HOOD FOR
VENT CONTROL
ALUMINA
CRUST
CRYOLITE BATH
CATHODE
RING BUS
STEEL CATHODE
COLLECTOR BAR
FIGURE 7-26. PREBAKE REDUCTION CELL, SCHEMATIC ARRANGEMENT
7-128
-------
PASTE COMPARTMENT
COVER
ALUMINA
HOPPERS
REMOVABLE
CHANNELS
ALUMINA
CRUST
«AJUL*AAAAAAAAAAAAAAAAJL*AAAAAJUU^^
STEEL SHELL
INSULATION
PASTE
COMPARTMENT
CASING
POT ENCLOSURE
DOOR
ANODE STUDS
GAS & FUME
EVOLVING
CATHODE
COLLECTOR BAR
MOLTEN ALUMINUM
FIGURE 7-27. HSS SODERBERG CELL, SCHEMATIC ARRANGEMENT
7-129
-------
In both types of Soderberg cells, CO, C02, and hydrocarbons are released
as the pitch binder of the paste mixture bakes.72 These products are a com-
ponent of the Soderberg cell effluents and are essentially absent from those
of the prebaked cells. Their tarry nature requires modification of the con-
trol treatment techniques applied to the effluents, as it interfers with
pollutant removal devices. With HSS Soderberg cells hydrocarbons and CO are
collected at the cell in a hood and exit in the primary off-gases.
7.6.2 Emission Sources and Factors
The three significant sources of emissions of CO in the primary alumi-
num industry as pointed out in the preceding section are:
1) the reaction of oxygen with carbon anodes during both types of
cell operation,
2) baking of the pitch binder in Soderberg cell operation, and
3) baking of the anodes for the prebaked anode cells.
Emissions from the first two sources are found at the potlines; anode baking
emissions occur in the baking furnace flue gases. In addition, there are
miscellaneous sources of smaller amounts of CO emissions within aluminum
plants (see Section 7.6.2.3). Limited data concerning emissions of CO from
the primary aluminum industry are summarized in Table 7-26.
CO emissions from potlines, from anode baking furnaces, and from mis-
cellaneous sources are quantified in the following discussion.
7.6.2.1 Potline Emissions
Little CO emission data are available for potline operations. Table
7-27 presents data on CO emissions collected by EPA while measuring fluoride
7-130
-------
emissions from several potline operations.72 There are two emissions points
from potlines: primary and secondary as shown in Figure 7-28. The reported
primary CO emission rates for prebake cell potlines range from 250 to 960
kilograms CO per metric ton of aluminum produced (500 to 1900 Ibs CO/ton Al).72
No CO was detected in the primary outlet for either the VSS or HSS Soderberg
cell plants.72 The validity of these data is questionable.
Two types of secondary emissions v/ere reported. CO emissions for one
VSS plant were reported to be 340 kilograms CO per metric ton of aluminum
produced (680 Ibs CO/ton Al).72 No CO was detected in the secondary outlet
of two other plants (two measurements at one VSS plant, one measurement at
an HSS plant).72 The validity of these data is questionable.
TABLE 7-26
CARBON MONOXIDE EMISSIONS FROM PRIMARY ALUMINUM PRODUCTION
CO EMISSIONS3
Metric 1
PLANT TYPE
Prebake anode
Horizontal stud
Soderberg
Vertical stud Soderberg
Anode bake furnace
Other
TOTAL
Metric Tons CO
IL
117,000
57,900
28,400
12,500
3,600
219,^00
Tons CO
yr_
128,700
63,700
31 ,200
13,800
4,000
241 ,400
Based on 1973 production of 4,117,300 metric tons (4,529,000 tons)
of aluminum.
Source: Reference 72
7-131
-------
fO
oo
CM
r»-
CO
c —
o o
LJ >
o
Q-
CD
•<
cc:
CO CO CO CO
to oo oo oo
O^ CM CM
O O O SO
o o o o
PA PA
o o
O O O
000
O O O O O
o o o o o
OO CM — O —
-------
00
00
>-
oo
O
O
OO
oo
o:
o
o
o
LJ
00
o
z:
«=c
>-
DC:
<
D_
O
CD
00
CM
ce:
7-133
-------
Two measurements were reported for prebake plants which had no controls
on the roof monitor emissions.72 In one case, no CO was detected in the
roof monitor emissions.72 For the other plant, CO emissions were reported
to be 6,300 kilograms CO per metric ton of aluminum produced (12,600 Ibs CO/
ton Al).72 The validity of these data is also questionable.
The foregoing data make it obvious that more study is needed to charac-
terize primary and secondary CO emissions from both prebaked anode and Soder-
berg cell potlines.
7.6.2.2 Anode Bake Furnaces
CO emissions data for anode bake furnaces are also scarce. Table 7-28
presents data collected by EPA at one anode plant.72 The CO emission factor
for this plant ranged from 150 to 180 kilograms CO per metric ton (300 to 400
Ib/ton) of anode produced. The average emission factor was 160 kilograms CO
per metric ton (320 Ib/ton) of anode produced.
TABLE 7-28
ANODE FURNACE CO EMISSIONS
TEST NUMBER
PARAMETER 111 AVERAGE
Anode production: kg/s 2.30 2.30 2.30 2 30
(Ib/hr) 18,200 18,200 18,200 18,200
Gas flow (dry): Nm3/s 18.3 21.1 17.5 |8 6
(scfm) 40,000 45,000 37,000 40,000
CO concentration ].6
(Volume %, dry)
CO emission factor: kgCO/metric 160 180 150 160
ton anode
(Ib CO/ton anode)320 360 300 320
Source: Reference 72
7-134
-------
7.6.2.3 Miscellaneous Sources
Most aluminum reduction plants have a casthouse on-site. The casthouse
usually has several reverberatory furnaces which are used for holding and
fluxing the molten aluminum prior to casting. These furnaces are oil- or
gas-fired and do emit small quantities of CO in the off-gases. All off-
gases from the casthouse are vented uncontrolled to the atmosphere.
Prebake plants all have a rodding room associated with the carbon plant.
In the rodding room, the copper rods which conduct electricity to the anode
are fastened to the carbon anode with cast iron. The cast iron melting
furnaces are small CO sources.
The only CO emission data from these sources were found in the 1973
National Emissions Data System (NEDS) file.71 The total CO emissions
reported were 3600 metric tons CO per year (4000 tons/yr). This translates
to an emission factor of 0.88 kilograms CO per metric ton of aluminum (1.76
Ib/ton Al) based on the 1973 U.S. primary aluminum production of 4,117,300
metric tons (4,529,000 tons) per year.73 These emissions are small compared
to those for anode bake furnaces and potlines.
7.6.3 Control Techniques
The primary aluminum industry does not presently use any techniques
designed specifically for CO control. Should CO control become necessary,
two control alternatives might be considered for primary emissions from the
potlines. The first is thermal incineration of the CO present in the pri-
mary emissions in an afterburner. The second would be, in the case of pre-
bake plants, recycle of the primary emissions to the anode bake furnace
combustion air fan.
7-135
-------
Low CO concentrations and huge gas volumes would make thermal or cata-
lytic incineration of secondary CO emissions from potlines very costly.
Catalytic conversion of either the primary or secondary CO emissions
might be impractical because of catalyst sensitivity to the particulate
and gaseous fluorides present in the gas streams.
7.6.3.1 Thermal Incinerators
A thermal incinerator as described in Chapter 6 could possibly be used
to combust CO present in the primary potline emissions. The incinerator
would treat the gases after they exit either the wet or dry particulate
removal devices used at most aluminum reduction plants. Supplemental fuel
would be required to incinerate the primary potline emissions because of the
low heating value of the gas [38 to 76 kilojoules/m3 (1 to 2 Btu/ft3)].72
An incinerator operating temperature between 870°C and 980°C (1600 to
1800°F) would be required to achieve adequate CO combustion efficiencies.
Higher temperatures would result in more complete CO combustion but NO
X
formation increases rapidly at temperatures above 980°C (1800°F). More
study is needed to predict the effectiveness of thermal incineration for
reducing low concentration CO emissions.
7.6.3.2 Potline Off Gas Recycle
At prebake anode reduction plants, it might be possible to duct the
primary potline off-gases to the suction of the anode bake furnace combus-
tion air fan. The duct length and fan size required would vary considerably
from plant to plant. No supplemental fuel would be required other than the
fuel currently used in the anode bake furnaces. Trace quantities of fluo-
rides present in the gas stream pose a potential fan corrosion problem
7-136
-------
which should be investigated if this control option is considered.72 More
study is needed to predict the effectiveness and cost of this technique.
7.6.4 Cost of Controls
As mentioned earlier, the primary aluminum industry does not presently
use any CO control technology. As a result, there are no cost data for
either thermal incinerator or potline gas recycle installations at aluminum
reduction plants. The size, layout, age, gas flow, and pot type all vary
considerably between plants. Cost estimates would be very site specific.
Furthermore, because of the sparse data on CO emission rates, it is not
possible to calculate representative costs for CO control at this time.
7.6.5 Impact of Controls
7.6.5.1 Emissions Reduction
There are not enough data to estimate the potential effectiveness of
thermal incineration for reducing CO emissions from primary aluminum plants.
7.6.5.2 Environment
If incineration were used to control CO emissions, N0x emissions in the
incinerator flue gas would increase. N0x emissions increase as a function
of temperature. Both the burner flame temperature and the average incinera-
tor operating temperature affect the quantity of N0x generated. Average
incinerator temperatures of 980°C (1800°F) can be expected to cause signifi-
cant quantities of N0x to form.
At the present time, natural gas is generally used as supplemental
incinerator fuel. If future shortages of natural gas necessitate the use
7-137
-------
of fuel oil as supplemental fuel, an increase in sulfur oxide emissions (SOX)
can be expected. The magnitude of the S0x emissions would depend on the
sulfur content of the fuel and the total amount of fuel consumed.
7.6.5.3 Energy Requirements
Because the potline off-gases have such a low heating value, only 38
to 76 kj/m3 (l to 2 Btu/ft3), most of the heat for thermal incineration
would have to be supplied by supplemental fuel.72 Approximately 4 Mm3
natural gas/Nm3 off-gas (4 scf/scf) would be required to incinerate potline
off-gases.72 This represents between 46.3 and 220.1 megajoules/metric ton
Al (40 to 190 x 103 Btu/ton Al), based on data from Reference 72.
7.7 PULP AND PAPER INDUSTRY
Although the pulp and paper industry is comprised of three distinct
segments (pulp, primary paper and paperboard, and converted paper and paper-
board products), the only segment which has the potential for contributing
significant CO emissions to the atmosphere is the pulping segment. Further-
more, of the commercially used pulping processes, only one, the kraft pro-
cess, is significant with respect to CO emissions. CO emissions from the
kraft process were estimated by EPA at 1,105,700 metric tons/yr (1,218,700
tons/yr) in 1977.2
7.7.1 Process Description and Emission Factors
7-7.1.1 Process Description - Kraft Pulping
In the kraft or sulfate pulping process, cellulose fibers (i.e., pulp)
are separated from the binding material called lignin. This is accomplished
by chemical digestion at elevated temperature and pressure in a "white
7-138
-------
liquor" solution of sodium sulfide and sodium hydroxide. Then the pulp is
separated by filtration, and the spent liquor, now referred to as "black
liquor," is sent to a chemical recovery system along with pulp wash water.
It is this recovery system which is the source of CO emissions of interest
in this industry.
A simplified flow diagram of the kraft process is presented in Figure
7-29. The entire process may be considered in eight parts:
1. Digester
2. Brown stock washer system
3. Multiple-effect evaporation
4. Recovery furnace system
5. Smelt dissolving tank
6. Lime kiln
7. Black liquor oxidation system
8. Condensate stripping system
Digestion -- Digestion of the wood chips is carried out in batch, con-
tinuous or, in a few cases, rotary digesters. While usage of continuous
units is increasing, most pulping at this time is still carried out in
batch digesters. The wood chips are cooked with white liquor at about
170-175°C (340-350°F) and 0.8-1 megapascal (100-135 psig) for two to five
hours. Gases formed during digestion are periodically vented to maintain
proper process pressure. In batch processes, when the cooling is complete,
the pressure is reduced to 0.7 megapascal (80 psig) and the contents are
discharged to an atmospheric blow tank where the pulp is drained. The steam
7-139
-------
T
Vent Gases
Wood —;
White Liquor —*
(NaOH + Na,,S)
•Condensate
DIGESTER
SYSTEM
•Pulp
4-
Exhaust Gas
RECOVERY
FURNACE
SYSTEM
BROWN STOCK
(PULP)
WASHERS
PulD
Water
Heavy
.Black ^
Liquor
-------
and other gases released here are sent to a heat accumulator recovery unit.
This blow of the digestor does not pertain to continuous digesters.
Brown Stock Washer System -- Chunks of undigested wood are removed,
and the remaining pulp is washed countercurrently in several stages. Vacuum
filters are used to dewater the washed pulp.
Multiple-Effect Evaporators -- The brown stock wash water and spent
liquor are combined to form a weak black liquor. This stream is concentrated
from 12-18 percent solids to 40-55 percent solids in a series of five or six
evaporation units. Further concentration steps may be taken to increase the
solids content to 63 percent, which is the level needed for combustion in
the recovery furnace.
Recovery Furnace System -- The concentrated black liquor from the
evaporative system is then burned in the recovery furnace. Combustion in
this manner allows for recovery of sodium and sulfur, production of steam,
and disposal of unwanted dissolved wood components of the liquor. The fur-
nace can theoretically be divided into three zones: drying, reducing and
oxidizing. The black liquor is sprayed into the drying zone where evapora-
tion takes place. The spray nozzles are located on one furnace wall and
oscillate automatically so that the sheet spray covers the other walls.
The frequency and extent of oscillation may be adjusted to optimize the
operation and to minimize emission of objectionable gases. Emphasis is
placed on minimizing reduced sulfur species, but CO emissions are also
affected.
The solids fall to the hearth, forming the char bed where combustion
begins. In the lower furnace (reduction) zone inorganic sulfate and other
7-141
-------
sodium compounds are reduced. These compounds, mainly sodium sulfide and
sodium carbonate, settle out in a smelt on the furnace grate. Organic sul-
fur compounds are oxidized in the upper, or oxidizing zone. Combustion air
is supplied by a forced-draft system through lower (primary) air ports in
the reduction zone and through secondary and tertiary ports in the upper
zone.
There are two types of furnaces in use. The majority in use at this
time employ a direct contact evaporator to provide an evaporation step
necessary for concentrating the 55 percent solids black liquor to 63-65
percent solids prior to combustion. In this type of furnace, black liquor
is contacted directly with furnace exhaust gases. The other type is a non-
contact, direct-fired, "low odor" or indirect-contact system.
Smelt Dissolving Tank — This is a large tank located below the
recovery furnace. Molten smelt discharged from the furnace floor is dis-
solved in water, forming "green liquor" in the stirred tank. A steam or
liquid shatterjet system is used to break up the smelt stream before it
contacts the water.
L1me Kl'1" — This unit is a source of CO as well as particulate emis-
sions. The kiln is a part of the closed-loop system that converts green
liquor to white liquor. In the kiln calcination of the lime mud (calcium
carbonate which precipitates in the causticizer) is carried out to produce
calcium oxide for recausticizing the green liquor discharged from the smelt
dissolving tank. Large rotary kilns with capacities of 36-360 megagrams
(40-400 tons) of quicklime per day are typically used, although there are
a few fluidized bed calciners also in use.74 The lime sludge typically
7-142
-------
enters as a slurry containing 55-60 percent solids. The quicklime produced
is then sent to a slaker to form a calcium hydroxide solution for the
causticizing reaction.
Black Liquor Oxidation System -- The purpose of black liquor oxidation
is to raise the oxidation state of sodium sulfide in either weak or strong
black liquor, thereby decreasing reduced sulfur species emissions from the
direct contact evaporator. Air, or in a few cases, oxygen is used to
oxidize the sodium sulfide to sodium thiosulfate or a more oxidized form.
The process can be carried out in sparging reactors, packed towers and bubble
tray columns in single or multiple stages.
Condensate Stripping System -- Condensation of off-gases from the
digester and multiple-effect evaporators results in dissolution of some
total reduced sulfur gases in the condensate. To avoid odor problems,
these compounds are stripped either by air or steam before the condensates
are discharged to the pond.
7.7.1.2 Emissions
In 1977, CO emissions from the kraft process were estimated at 1,105,700
metric tons (1,218,700 tons).2 The major reported source of CO in this pro-
cess is the sulfite recovery system. The conventional recovery system con-
sists of a furnace and a direct-contact evaporator. Newer systems have, in
some cases, a modified furnace and an indirect-contact evaporator. In the
furnace, reduction of sulfate to sulfide takes place, with accompanying
formation of reduced gaseous sulfur species and carbon monoxide. Air is
admitted above this reduction zone to oxidize these combustible gases. If
7-143
-------
the furnace is operated within design capacity, CO emissions are very low.
If furnaces are operated above their design capacity, there is an insufficient
supply of air for complete combustion of the furnace gases, causing increased
emissions of CO. Emissions of CO in the recovery furnace flue gas can vary
from negligible under proper operation to nearly two volume percent with an
inadequate air supply.1*9 EPA emission factors range from 1-30 kg/metric ton
(2-60 Ib/ton), the higher number characterizing CO emissions from an over-
loaded furnace.5 CO emissions measured by EPA from two recovery furnaces
were about 1.3 kg/metric ton pulp (2.5 lb/ton).7t|
The quantity of carbon monoxide emitted from lime kilns deoends upon
the following factors:
1) kiln operating temperature,
2) amount of excess air, and
3) type of fuel used.
Table 7-29 presents reported compositions of exhaust gases from two
rotary kilns of comparable throughput but operating at different tempera-
tures and excess air levels. The type of fuel used in the kiln also
affects the amount of CO emitted. When coal or coke are used, carbon monox-
ide concentrations in the exhaust gases may range up to one volume percent.
For kilns using natural gas or fuel oil, CO concentrations are much less
and may be negligible if the excess air and kiln operating temperature are
high.
The reported CO emission factor from lime kilns is 5 kg/metric ton
of air-dried unbleached pulp (10 Ibs/ton).5
7-144
-------
TABLE 7-29
REPORTED COMPOSITIONS OF EXHAUST GASES FROM TWO GAS-FIRED LIME
SLUDGE KILNS
VOLUME %
COMPONENT
H20
C02
CO
02
N2
KILN Aa
37.1
10.4
0.0
3.2
^9.3
100.0
KILN Bb
30.0
15.3
0.5
0.2
5^.0
100.0
Kiln operated at very high excess air and exhaust temperature of
210°C
Kiln operated at less than 5% excess air and exhaust temperature
of 175°C (350°F)
Source: Reference
7.7.2 Control Techniques
Currently there are not measures applied for CO control in the pulping
industry. However, since the primary sources of CO emissions are recovery
furnaces operating above design limits, the best control would simply be
proper operation of these furnaces. As mentioned earlier, furnaces operat-
ing within design limits emit little or no carbon monoxide. Alternately,
operation of furnaces above design capacity with low CO emissions may be
possible with some modifications of furnace operation. Adjusting primary
and secondary air rates to the furnace may provide the required amount of
oxygen to oxidize the CO before it escapes. However, the effectiveness of
this control method may be limited due to the decreased residence times of
the gases in the furnace. CO concentrations in the off-gas will almost
certainly depend upon this parameter. No data were available on this effect
7-145
-------
With increasingly strict regulation of total reduced sulfur emissions
from the pulping industry, many plants are converting their recovery systems.
These modifications usually include replacement of the furnace itself by one
of more efficient design and/or conversion to an indirect-contact evaporator.
These sulfur control measures should reduce total CO emissions.
The energy content of the exhaust gases from the recovery furnace is
very low, less than 37 kilojoules/Nm3 (1 Btu/scf).1*2 For this reason, incin-
eration of such a large volume, low energy content gas would be costly.
Lime kiln emissions of carbon monoxide can be most effectively con-
trolled by operating the kiln at sufficient temperatures and excess air
levels to eliminate the CO in the exhaust gases. However, the effectiveness
of this technique on the CO from kilns fired with coke or coal is unknown.
Based on the data reported in Table 7-29, high excess air levels and temp-
eratures can reduce CO emissions from gas-fired kilns substantially (over
99 percent) compared to kilns operating with low excess air levels and lower
temperatures.
7.7.3 Cost of Controls
Estimates for the costs of the controls outlined above are not available,
Proper operation of these recovery furnaces operating above design limits
should result in no additional costs. It may be argued, though, that this
is in effect a derating of the furnace. Increasing primary and secondary
air rates to the furnaces operating above design limits may require the
addition of another blower to the furnace air supply system.
Costs for increasing excess air levels and temperatures in lime kilns
will consist primarily of the cost for additional fuel to raise operating
7-146
-------
temperatures. Also, additional air supply capacity will need to be added to
increase the excess air level in the lime kilns.
7.7.4 Impact of Controls
7.7.4.1 Emissions Reduction
Assuming that proper operation of recovery furnaces will result in one
kg CO/metric ton pulp (2.0 lb/ton)5 and applying this factor to total produc-
tion (29 teragrams [32,000,000 tons] in 1974)74 results in a total annual
emission reduction of 29 gigagrams (32,000 tons CO) per year.
7.7.4.2 Energy Requirements
The additional fuel required to raise lime kiln operating temperatures
will be the only significant energy requirement of the controls identified.
No data were available to estimate this requirement.
7.7.4.3 Environment
No adverse environmental impacts are anticipated from modification of
operating procedures for either recovery furnaces or lime kilns.
7-147
-------
REFERENCES FOR CHAPTER 7
1. Schwartz, W.A. et al. Engineering and Cost Study of Air Pollution
Control for the Petrochemical Industry, Vol. I: Carbon Black Manu-
facture by the Furnace Process, Final Report. EPA-450/3-73-006-a,
U.S. Environmental Protection Agency, Research Triangle Park, North
Carolina, June 1974.
2. National Air Quality Monitoring and Emission Trends 1977, EPA-450/2-
78-052, U.S. Environmental Protection Agency, Research Triangle Park,
North Carolina, December 1978.
3. Chemical Marketing Reporter, 205(10). March 11, 1974.
4. Industrial Process Profiles for Environmental Use, EPA-600/2-77-23d,
U.S. Environmental Protection Agency, Research Triangle Park, North
Carolina, February 1977.
5. Compilation of Air Pollution Emission Factors, AP-42, U.S. Environ-
mental Protection Agency, Research Triangle Park, North Carolina,
May 1978.
6. Carbon Black (Oil Furnace Black). Hydrocarbon Processing, November
1977, p. 138.
7. U.S. Environmental Protection Agency, Emission Standards and Engineer-
ing Division, Research Triangle Park, North Carolina. Unpublished
Data. April 1976.
8. Moscowitz, L.M. Source Assessment: Charcoal Manufacturing State-of-
the-Art. Monsanto Research Corporation. Prepared for the Office of
Research and Development, U.S. Environmental Protection Agency,
December 1977. 77pp.
7-148
-------
9. Hulman, P.S., R.D. Delleney, S.M. Killingsworth. Screening Study on
Feasibility of Standards of Performance for Wood Charcoal Manufacturing
Radian Corporation. Prepared for the U.S. Environmental Protection
Agency. Research Triangle Park, North Carolina. August 1978. 47 pp.
10. Afterburner Systems Study, EPA-R2-72-062, U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina, August 1972.
11. Kroneberger, G.F. Applications Involving Multiple Hearth Processing
and Agglomeration of Briquetting. Envirotech Systems, Inc. (Presented
at the 12th Biennial Conference of the International Briquetting Asso-
ciation. Vancouver, British Columbia, Canada. August 1971) pp 31-38.
12. Particulate Emission Rate and Size Study on the ACC Outlet Stack, the
Aerodyne Cyclones, and the Briquette Dryer Vent. Harman Engineering,
Auburn, Alabama, June 28, 1977.
13. Gallagher, P.P. Utilization of Off Gases from Herreshoff-Furnace
Charcoal Production. Nichols Engineering and Research Corporation.
(Presented at the llth Biennial Conference of the International
Briquetting Association. Sun Valley, Idaho, August 27-29, 1969).
pp 27-29.
14. Rienks, J. Charcoal Burner and Waste Heat Systems. Olson Lawyer
Lumber, Inc. (Presented at the Forest Products Research Society
Energy Workshop. Denver, Colorado. September 3-5,1976) pp 104-106.
15. Stanford Research Institute, Intl. 1978 Directory of Chemical Pro-
ducers, U.S.A. Menlo Park, California, 1973. p 418, 427.
16. Source Assessment: Acrylonitrile Manufacture (Air Emissions),
EPA-600/2-77-107J, U.S. Environmental Protection Agency, Research
Triangle Park, North Carolina, February 1977.
7-149
-------
17. Engineering and Cost Study of Air Pollution Control for the Petro-
chemical Industry, Volume 2: Acrylonitrile Manufacture, Final Report,
EPA-450/3-73-0065, U.S. Environmental Protection Agency, Research
Triangle Park, North Carolina, February 1975.
18. Engineering and Cost Study of Air Pollution Control for the Petro-
chemical Industry, Volume 5, Formaldehyde Manufacture with the Mixed
Oxide Catalyst Process, Final Report, EPA-450/3-73-006e, U.S. Environ-
mental Protection Agency, Research Triangle Park, North Carolina,
March 1975.
19. Engineering and Cost Study of Air Pollution Control of the Petro-
chemical Industry, Volume 4, Formaldehyde Manufacture with the Silver
Catalyst Process, Final Report, EPA-450/3-73-0061, U.S. Environmental
Protection Agency, Research Triangle Park, North Carolina, March 1975.
20. Control Techniques for Volatile Organic Emissions from Stationary
Sources, EPA-450/2-78-022, U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina, May 1978.
21. Hawley, Gessner, G., rev. Condensed Chemical Dictionary, 8th ed. Van
Nostrand-Reinhold, New York, 1971. p. 538.
22. Survey Reports on Atmospheric Emissions from the Petrochemical Industry,
Final Report, Volume 3, EPA-450/3-73-005 a-d, U.S. Environmental Pro-
tection Agency, Research Triangle Park, North Carolina, March 1974.
23. Maleic Anhydride, BASF AG. Hydrocarbon Proc. 56 (11): 179. 1977.
24. Source Assessment: Maleic Anhydride Manufacture. Monsanto Research
Corp., Dayton, Ohio. Prepared for EPA, Contract No. 68-02-1874.
April 1978.
7-150
-------
25. Source Assessment: Phthalic Anhydride (Air Emissions). EPA-600/2-76-
032d, U.S. Environmental Protection Agency, Research Triangle Park,
North Carolina, December 1976.
26. Engineering and Cost Study of Air Pollution Control for the Petro-
chemical Industry, Volume 7: Phthalic Anhydride Manufacture from
Ortho Xylene, Final Report. EPA-450/3-73-006-g, U.S. Environmental
Protection Agency, Research Triangle Park, North Carolina, July 1975.
27. Phthalic Anhydride Plant Air Pollution Control, EPA-600/2-77-188,
U.S. Environmental Protection Agency, Research Triangle Park, North
Carolina, September 1977.
28. Industrial Process Profiles for Environmental Use, Chapter 24, Iron
and Steel Industry. EPA-600/2-77-023X, U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina, February 1977.
29. Wheeler, O.H. Fume Control in L-D Plants. J. APCA 18 (2): 98, 1968.
30. Control Techniques for Nitrogen Oxides Emissions from Stationary
Sources, 2nd Edition, EPA-450/1-78-001, U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina, January 1978.
31. The Pneumatic Steelmaking Process. In: The Making, Shaping and Treat-
ing of Steel, Ninth Edition, McGannon, H.E. (ed.). U.S. Steel Company,
Pittsburgh, Pennsylvania, 1971. pp. 473-479.
32. A Systems Analysis Study of the Integrated Iron and Steel Industry,
PB 184 577, U.S. Environmental Protection Agency, Research Triangle
Park, North Carolina, May 1969.
33. American Iron and Steel Institute. The Making of Steel. Washington, DC
7-151
-------
34. Cerepaka, Bruce. An Air Pollution Compliance Analysis Report on Nine
Industries, Executive Report, 9 Vols. Radian Project No. 200-C45-16.
Austin, Texas, September 1975.
35. Background Information for Standards of Performance: Electric Submerged
Arc Furnaces for Production of Ferroalloys, Vol. 1. EPA-450/2-74-018a,
U.S. Environmental Protection Agency, Research Triangle Park, North
Carolina, October 1974.
36. Background Information for Standards of Performance: Electric Sub-
merged Arc Furnances for the Production of Ferroalloys. Volume 2,
EPA-450/2-74-018b. U.S. Environmental Protection Agency, Research
Triangle Park, North Carolina, October 1974.
37. Background Information for Standards of Performance: Electric Arc
Furnaces in the Steel Industry. Volume 1, EPA-450/2-74-017a. U.S.
Environmental Protection Agency, Research Triangle Park, North
Carolina, October 1974.
38. Kearney, A.T. & Co., Inc. Air Pollution Aspects of the Iron Foundry
Industry. APTD 0806, PB 204-712, U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina, 1971.
39. Brough, J.R. and W.A. Carter. Air Pollution Control of an Electric
Furnace Steel Making Shop. J. APCA 167-171, March 1972.
40. Roy F. Weston, Incorporated. Source Testing Report for Plant A.
EPA Contract No. 68-02-0240. January 1973.
41. Air Pollutant Control Techniques for Electric Arc Furnances in the
Iron and Steel Foundry Industry, EPA-450/2-78-024, U.S. Environmental
Protection Agency, Research Triangle Park, North Carolina, June 1978.
7-152
-------
42. Air Pollution Control Technology and Costs in Seven Selected Areas,
EPA-450/3-73-010, PB 231 757, U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina, December 1973.
43. Letter W.B. Huelsen, American Foundry Men's Society to D.R. Goodwin,
U.S. Environmental Protection Agency, Research Triangle Park, North
Carolina, January 1979.
44. Draper, Allen. (Professor, Penn State University). Private communica-
tion with R. L. Cook, Radian Corporation, July 31, 1978.
45. An Investigation of the Best Systems of Emission Reduction for Sinter
Plants in the Iron and Steel Industry, Draft Standards Support and
Environmental Impact Statement, U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina, April 1976.
46. Chaterbury, Rama. (Permit Division, Illinois Environmental Protection
Agency). Private Communication with R. L. Cook, Radian Corporation,
July 11, 1978.
47. Ban, Thomas E. Ore-Sintering Process Reduces Air Pollutants. Chem.
Eng. 1978 (June 19), 81.
48. Cantrell, A. Annual Refining Survey. Oil and Gas Journal. March 20,
1978. pp. 108-146.
49. Control Techniques for Carbon Monoxide Emissions from Stationary
Sources. AP-65, U.S. Environmental Protection Agency, Research
Triangle Park, North Carolina, March 1970.
50. Stover, R.D. Control of Carbon Monoxide Emissions from FCC Units
by UltraCat Regeneration. In: Ind. Proc. Des. Poll. Control, Proc.
AIChE Workshop 6:80-85, 1975.
7-153
-------
51. Gary, James H. and Glenn E. Handwerk. Petroleum Refining, Technology
and Economics. Marcel Dekker, New York, 1975.
52. Knight, W. N. N., and M. L. Peniston-Bird. Modern Petroleum Tech-
nology. Applied Science Publishers, Ltd., Essex, England, 1975.
293 pp.
53. Hemler, C. L. Developments in Fluid Catalytic Cracking. Nafta
(Zagreb) 27(4):207-216, 1976.
54. Murphy, J. R., and M. Soudek. Modern FCC Units Incorporate Many
Design Advances. Oil and Gas Journal 1977 (January 17): 70-76.
55. Rheaume, L. et al. New FCC Catalysts Cut Energy and Increase
Activity. Oil Gas J. 74(20):103-107, 110, 1976.
56. Murphy, J. R. Catalyst and Design Spur FCC Revival. Oil Gas J. 1970
(Nov. 23): 72.
57. Arthur D. Little, Inc. Screening Study to Determine Need for NSPS for
New FCC Regenerators. Draft Final Report. EPA 68-02-1332, Task 22.
Cambridge, Massachusetts, June 1976.
58. Bruch, H. R. Refiners Focus on New FCC Technology. NPRA Question
and Answer Session. Oil and Gas J. 1976 (March 15): 87-93.
59. Davis, John C. FCC Units Get Crack Catalysts. Chem. Eng. 84(12);
77-79, 1977.
60. Grace, W. R., and Co. Davison Chemical Co. Private Communication
with J. B. Jarvis, Radian Corporation, September 1978.
61. Foulkes, P. B. and M. D. Harper. Prospects for Coking May Brighten.
Oil and Gas J. 1978 (March 20): 85-93.
7-154
-------
62. Busch, R.G. Fluid Coking: Seasoned Process Takes on New Jobs. Oil
and Gas J. 1970 (April 6): 102-111.
63. Bailleul, M. R., P. Berthier, and G. Guyot. Pollution and Sulfur
Plants. Tech. Paper No. 49. In: Proc. Annual Conv. Natural Gas
Processors Association, pp. 89-92.
64. Beers, W. D. Characterization of Claus Plant Emissions. Final Report
EPA-R2-73-188, PB 220 376/8, EPA Contract No. 68-02-0242. Processes
Research Inc., Industrial Planning and Research, Cincinnati, Ohio.
April 1973.
65. Institute of Gas Technology, Illinois Institute of Technology. LNG:
A Sulfur-Free Fuel for Power Generation. Final Report. Contract No.
PH 22-68-58. Chicago, Illinois, 1969.
66. Groenendaal, W., and H. C. A. Van Meurs. Shell Launches its Claus
Off-Gas Desulfurization Process. Petrol. Petrochem. Int. 12(9):
54-58, September 1972.
67. Barry, Charles B. Reduce Claus Sulfur Emission. Hydrocarbon Proc.
51(4):102-106, 1972.
68. Hyne, J. B. Methods for Desulfurization of Effluent Gas Streams.
Oil Gas J. 1972 (August 28), 64.
69. Beavon, David K. Add-On Process Slashes Claus Tailgas Pollution.
Chem. Eng. 78(28):71-73, 1971.
70. Source Assessment: Overview Matrix for National Criteria Pollutant
Emissions. EPA 600/2-77-107c. U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina, July 1977.
7-155
-------
71. NEDS Data File, National Air Data Branch, U.S. Environmental Protection
Agency, Research Triangle Park North Carolina, March 1979.
72. Primary Aluminum Draft Guidelines for Control of Fluoride Emissions
from Existing Primary Aluminum Plants, EPA-450/2-78-049a, U.S. Environ-
mental Protection Agency, Research Triangle Park, North Carolina,
February 1979.
73. U.S. Bureau of Mines. Minerals Yearbook 1973. Vol. I, Metals, Min-
erals, and Fuels. Government Printing Office, Washington, D.C., 1975.
74. Standards Support and Environmental Impact Statement, Volume 1:
Proposed Standards of Performance for Kraft Pulp Mills. EPA-450/2-
76-014a, U.S. Environmental Protection Agency, Research Triangle Park,
North Carolina.
7-156
-------
1. REPORT NO.
EPA-450/3-79-006
TECHNICAL REPORT DATA
(ncasc read Instructions on the reverse before competing]
2.
4. TITLE ANDSUBTITLE
Control Techniques for Carbon Monoxide Emissions
3. RECIPIENT'S ACCESSION NO.
5. REPORT DATE
June 1979
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS ~~~
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
10. PROGRAM ELEMENT NO.
12-SPONSORING AGENCY NAME AND ADDRESS
DAA for Air Quality Planning and Standards
Office of Air, Noise and Radiation
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
15. SUPPLEMENTARY NOTES ' " ~~
This document is issued per the
Clean Air Ammendments of 1977.
Fl6. ABSTRACT
11. CONTRACT/GRANT NO.
68-03-2608 Task No. 43
Radian Corporation
13 TYPE OF REPORT AND PERIOD COVERED
Final
14 SPONSORING AGENCY CODE
EPA/200/04
requirements of Section 108 of the
Contro Techniques for Carbon Monoxide Emissions presents recent developments of
control techniques which have become available since preparation of the first
r on °! Control Techniques for Carbon Monoxide, Emissions from Stationary
Sources (AP-65) and those sections of the first edition of Contro'
for Carbon Monoxide Nitrogen Oxide, and Hydrocarbon Emissions from Mobil
bource:
•66) that pertain to Carbon Monoxide. This edition presents
——_ - , -_... _„ „„. ~v., . .v,i ivssx i ui.. IIMCJ cuiuiun u rebcii tb
available data on control techniques including description, effectiveness
costs, and energy and environmental aspects. '
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
Carbon Monoxide
Control Techniques
Internal Combustion-Mobile
Internal Combustion-Stationary
Combustion
Costs
Industrial Sources
18. DISTRIBUTION STATEMENT
Unlimited
EPA Form 2220-1 (Rev. 4-77) PREV.OUS ED.T.ON is OBSOLETE
-------
------- |