EPA-450/3-81-005b
Control Techniques
for Participate Emissions
from Stationary Sources —
Volume 2
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
September 1982
For sale by the Superintendent ol Documents, U.S. Government Printing Office, Washington, D.C. 20402
-------
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 for sale by the Superintendent of Documents,
U.S. Government Printing Office, Washington, D.C. 20402, and the National
Technical Information Services, 5285 Port Royal Road, Springfield,
Virginia 22161.
it
-------
SUMMARY TABLE-OF CONTENTS
: .. VOLUME i
Section Page
1. INTRODUCTION 1-1
2. BACKGROUND 2-1
2.1 Trends and Projections in Particulate Emissions 2-1
2.2 Sources of Suspended Particulate Matter 2-3
2.3 Various Approaches to Limiting Particulate Emissions 2-11
3. ALTERNATIVE PARTICULATE CONTROL APPROACHES 3-1
3.1 Energy Source and Fuel ^Selection 3-1
3.2 Process Optimization ; 3-3
3.3 Exhaust Gas Cleaning ;, 3-5
4. PARTICULATE CONTROL SYSTEMS 4-1
4.1 Selection and Application of Particulate Control
Systems 4.1-1
4.2 Mechanical Collectors 4.2-1
4.3 Electrostatic Precipitators 4.3-1
4.4 Fabric Filter 4.4-1
4.5 Wet Scrubbers . 4.5-1
4.6 Incinerators 4.6-1
5. FUGITIVE EMISSION CONTROL 5-1
5.1 Sources of Fugitive Emissions 5-1
5.2 Control of Industrial Process Fugitive Emissions 5-2
5.3 Control of Fugitive Dust 5-16
6. ENERGY AND ENVIRONMENTAL CONSIDERATIONS 6-1
6.1 Energy Requirements 6-1
6.2 Secondary Pollutant Generation 6-14
6.3 Liquid Waste Management 6-15
6.4 Solid Waste Management; 6-21
6.5 Noise Management 6-25
6.6 Radiation Control 6-25
7. COSTS OF PARTICULATE CONTROL EQUIPMENT AND FUGITIVE
EMISSION CONTROL TECHNIQUES ; 7-1
7.1 Particulate Control Equipment Cost Analysis 7-2
7.2 Methodology for Analyzing Cost of Particulate Control
Systems 7-5
7.3 Cost Curves for Various. Particulate Control Systems 7-10
7.4 Cost of Fugitive Emission Control 7-28
-------
SUMMARY TABLE OF CONTENTS (Continued)
Section Page
8. EMERGING TECHNOLOGIES .8-1
8.1 Advanced Scrubbing Techniques 8-1
8.2 Advanced Electrostatic Precipitation Techniques 8-15
8.3 Advanced Filtration Techniques 8-28
8.4 High Gradient Magnetic Separation 8-34
8.5 Agglomeration Techniques 8-38
IV
-------
TABLE OF CONTENTS '......:.. /
VOLUME 2
Section . : : Page
LIST OF FIGURES . • : vi
LIST OF TABLES xii
CONVERSION FACTORS xxii
, TABLE OF SYMBOLS ; xxiii
SUMMARY TABLE OF CONTENTS - VOLUME 1 xxvi
9. SOURCES OF PARTICULATE EMISSIONS AND CONTROL
TECHNIQUES 9.0-1
9.1 Stationary Source Selection... 9.1-1
9.2 Stationary Combustion Sources 9.2-1
9.2.1 Pulverized Coal-Fired Boilers....... 9.2-1
9.2.2 Stoker Fed Coal-Fired Boilers 9.2-12
9.2.3 Coal-F-ired Cyclone Furnaces ....' '. 9.2-19
9.2.4 Nonfossil Fuel-Fired Boilers ; 9.2-21
9.2.5 Oil-Fired Utility Boilers 9.2-45
9.2.6 Oil-Fired Industrial/Commercial Boilers.... 9.2-49
9.3 Refuse Incinerators 9.3-1
9.3.1 Municipal Incinerators 9.3-1
9.3.2 Industrial and Commercial Incinerators 9.3-11
9.3.3 Sludge Incinerators 9.3-24
9.4 Open Burning... „ 9.4-1
9.4.1 Agricultural Burning.. .* 9.4-1
9.4.2 Prescribed Burning.. 9.4-5
9.5 Chemical Process Industry 9.5-1
9.5.1 Charcoal Plants.... 9.5-1
9.5.2 Carbon Black (Furnace Process) 9.5-3
9.5.3 Detergent Manufacturing Plants 9.5-6
9.5.4 Explosives Industry 9.5-8
9.5.5 Thermal Process Phosphoric Acid
Manufacturing.. 9.5-9
9»5.6 Sulfuric Acid.; 9.5-12
9.5.7 Phthalic Anhydride 9.5-18
9.5.8 Hydrogen Fluoride (Hydrofluoric Acid)...... 9.5-21
9.5.9 Boron Compounds 9.5-23
9.5.10 Pesticide Manufacturing... 9.5-27
9.5.11 Sodium Carbonate (Natural Process)......... 9.5-27
9.5.12 Potash 9.5-33
-------
TABLE OF CONTENTS.(Continued)
Section - . Page
9.6 Food and Agricultural Industry 9.6-1
9.6.1 Ammonium Nitrate Fertilizer. 9.6-1
9.6.2 Ammonium Sulfate Fertilizer... 9.6-7
9.6".3 Urea Fertilizer 9.6-12
9.6.4 Diammonium Phosphate Fertilizer 9.6-15
9.6.5 Grain Handling and Storage 9.6-16
9.6.6 Grain Processing 9.6-20
9.6.7 Alfalfa Dehydrating 9.6-25
9.6.8 Cotton Ginning 9.6-27
9.6.9 Starch Manufacture 9.6-32
9.6.10 Vegetable Oil 9.6-34
9.7 Mineral Products 9.7-1
9.7.1 Nonmetallie Mineral Processing 9.7-1
9.7.2 Metallic Minerals Mining and Processing.... 9.7-25
9.7.3 Brick and Related Clay Products 9.7-42
9.7.4 Beneficated Clay Products Manufacturing.... 9.7-47
9.7.5 Gypsum 9.7-51
9.7.6 Lime 9.7-62
9.7.7 Cement-Manufacturing. 9.7-71
9.7.8 Concrete Batching..! 9.7-83
9.7.9 Asphalt Concrete Plants 9.7-86
9.7.10 Asphalt Roofing Plants.. 9.7-96
9.7.11 Glass Manufacturing. 9.7-105
9.7.12 Fiberglass Manufacturing 9.7-128
9.7.13 Mineral Wool 9.7-133
9.8 Metallurgical Industry 9.8-1
9.8.1 Iron and Steel Plants 9.8-2
9.8.1.1 Coke Ovens 9.8-2
9.8.1.2 Sintering 9.8-27
9.8.1.3 Blast Furnace 9.8-46
9.8.1.4 Open Hearth Furnace 9.8-50
9.8.1.5 Basic Oxygen Furnace 9.8-56
8.8.1.6 Electric Arc Furnace 9.8-62
9.8.1.7 Rolling, Shaping, and Finishing... 9.8-67
9.8.1.8 Scarfing 9.8-69
9.8.2 Ferroalloy Production 9.8-79
9.8.3 Gray Iron Foundries 9.8-92
9.8.3.1 Cupola 9.8-92
9.8.3.2 Electric Furnaces 9.8-94
9.8.4 Steel Foundries 9.8-98
9.8.5 Primary Aluminum 9.8-103
9.8.6 Primary Copper Smelters. 9.8-119
9.8.7 Primary Lead Smelting 9.8-138
9.8.8 Primary Zinc Smelting 9.8-150
9.8.9 Secondary Aluminum Operations..... 9.8-163
9.8.10 Secondary Copper Smelting and Alloying 9.8-171
VI
-------
TABLE OF CONTENTS (Continued)
Section . Page
9.8.11 Secondary Lead Smelting.................... 9.8-183
9.8.12 Secondary Zinc Processing 9.8-192
9.9 Petroleum Industry 9.9-1
9.9.1 Catalyst Regeneration Processes............. 9.9-1
9.9.2 Coking Processes, Emissions, and Control
Techniques....... ... 9.9-7
9.9.3 Air Blowing Operations, Emissions and
Control Techniques 9.9-9
9.9.4 Sludge Incineration, Emissions, and
Control Technique..... 9.9-9
9.9.5 Process Heater, Boiler, and Flare •-."
Emissions and Control, Techniques 9.9-10
9.10 Forest Products Industry 9.10-1
9.10.1 Kraft Pulping.... 9.10-1
9.10.2 Sulfite Pulping 9.10-16
9.10.3 Plywood Manufacture „ 9.10-21
9.11 Lead-Acid Battery Manufacturing 9.11-1
9.11.1 Storage Battery Manufacturing Processes
and Particulate Emissions ..... 9.11-1
9.11.2 Control Techniques 9.11-6
9.11.3 Secondary Environmental Impacts 9.11-12
9.12 Fugitive Dust Sources. 9.12-1
9.12.1 Agricultural Sources... 9.12-1
9.12.2 Transportation Sources 9.12-6
9.12.3 Aggregate Storage Piles and Waste
Disposal Heaps 9.12-18
9.12.4 Construction... 9.12-25
GLOSSARY. : G.l-1
vii
-------
LIST OF FIGURES -
F-igure Page
9.2.1 Pulverized Coal-Fired Boiler with Burner? Mounted
in Front Wai 1 i...'...... 9.2.2
9.2-2 Flyash Generation from Coal-Fired Boilers. 9.2-11
9.2-3 Traveling Grate Spreader-Stoker with Front Ash Discharge.. 9.2-13
9.2-4 Single-Retort, Horizontal-Feed, Side Ash Discharge
Underfeed Stoker , 9.2-14
9.2-5 Chain Grate Overfeed Stoker 9.2-15
9.2-6 Cyclone Furnace Side View 9.2-20
9.2-7 Size Distribution of MSW Flyash 9.2-25
9.2-8 Size Distribution of MSW Flyash from a Controlled Air
Boiler 9.2-27
9.2-9 Flyash Size Distribution for a Typical RDF/Coal Cofired
Boiler 9.2-30
9.2-10 Size Distribution of Wood Flyash 9.2-32
9.2-11 Dutch Oven Furnace and Boiler 9.2-33
9.2-12 Small Spreader-Stoker Furnace 9.2-34
9.2-13 Typical Relationships Between Cyclone Collection Effi-
ciency and Particle Diameter for Large and Small Tubes
(Same Inlet Velocity) 9.2-38
9.2-14 Variations of Single Cyclone Collection Efficiency with
Gas Flow Velocity 9.2-39
9.2-15 Electrical Resistivity of Flyash from Three Different
U.S. Municipal Incinerators 9.2-42
9.2-16 Schematic of Integral Cyclone Gravel Bed Filter 9.2-43
9.2-17 Oil-Fired Cast Iron Boiler.. 9.2-50
9.4-1 Annual Forest Fire Particulate Production by Region and
Season „ 9.4-6
9.5-1 Charcoal Briquetting Flow Diagram 9.5-2
vm
-------
LIST OF FIGURES (continued)
Figure , ; Page .
9.5-2 Carbon Black Process......... 9.5-5
9.5-3 Detergent Spray-Drying with Tower Equipped for Either
Cocurrent or Countercurrent Operation - Dampers in
'.. Countercurrent Mode of Operation.... ... 9.5-7
9.5-4 TNT Production by Batch Process 9.5-10
9.5-5 Flow Diagram for Typical Thermal-Process Phosphoric
Acid PI ant. 9.5-11
9.5-6 Contact-Process Sulfuric Acid Plant Burning Elemental
Sulfur ' 9.5-16
9.5-7 Badger-Sherwin-Williams Process for Manufacture of
Phthalic Anhydride from Naphthalene 9.5-19
9.5-8 BASF Process for Manufacture of Phthalic Anhydride from
0-xyl ene ... 9.5-20
9.5-9 Schematic Flow Diagram for the Manufacture of Hydrogen
Fl uoride ;. 9.5-22
9.5-10 Procedures for Recovering Borax--Trona Processes 9.5-24
9.5-11 Boric Acid from Borax by Acidulation. 9.5-25
9.5-12 Direct Carbonation Process... 9.5-35
9.5-13 Sesquicarbonate Process ; 9.5-36
9.5-14 Monohydrate Process 9.5-37
9.5-15 Simplified Flow Sheet for Production of Muriate of
Potash from Sylvite Ore 9.5-42
9.5-16 Simplified Flow Diagram for Production of Langbeinite,
(K-Mg-S) I 9.5-43
9.5-17 Simplified Flow Diagram for Production of Sulfate of
Potash 9.5-44
9.6-1 Flow Diagram of Ammonium Nitrate Production 9.6-3
9.6-2 Flow Diagram for Ammonium Sulfate Processes 9.6-8
9.6-3 Uncontrolled Ammonium Sulfate Dryer Emissions-Particle
Size Distribution 9.6-10
-------
LIST OF FIGURES (continued)
Figure Page
9.6-4 Diammonium Phosphate Production . 9.6-17
9.6-5 Generalized Flow Diagram for Alfalfa Dehydration Plant.... 9.6-26
9.6-6 Typical Ginning Operation 9.6-29
9.6-7 Corn Wet Milling Flow Diagram 9.6-35
9.6-8 Schematic of a 1,000 Ton Per Day Soybean Processing
PI ant 9.6-43
9.7-1 Flowsheet of a Typical Crushing Plant 9.7-5
9.7-2 General Schematic for Nonmetallic Minerals Processing 9.7-6
9.7-3 Summary of Visible Emission Measurements from Best
Controlled Fugitive Primary Crushing Source (Portable-
Facility T) by Means of Wet Supression (According to
EPA Method 9) 9.7-13
9.7-4 Summary of Visible Emission Measurements from Best
Controlled Fugitive Secondary Crushing Source (Portable-
Facility R) by Means of Wet Supression (According to
EPA Method 9) 9.7-14
9.7-5 Summary of Visible Emission Measurements from Best
Controlled Fugitive Secondary Crusher (Small,
Stationary-Facility S) by Means of Wet Supression
(According to EPA Method 9) 9.7-15
9.7-6 Summary of Visible Emission Measurements from Best
Controlled Fugitive Primary Crushing Source (Stationary-
Facility S) by Means of Wet Supression (According to
EPA Method 9) ' 9.7-16
9.7-7 Summary of Visible Emission Measurements from Best
Controlled Fugitive Secondary Crushing Source (Large,
Secondary-Facility S) by Means of Wet Supression
(According to EPA Method 9) 9.7-17
9.7-8 Particulate Emissions from Nonmetallic Minerals
Processing Operations 9.7-21
9.7-9 Scope of Mining Activities 9.7-32
9.7-10 General Schematic for Initial Steps in Ore Processing 9.7-34
9.7-11 Beneficiation Processes 9.7-35
-------
LIST OF FIGURES (continued)
Figure
9.7-12
9.7-13
9.7-14
9.7-15
9.7-16
9.7-17
9.7-18
9.7-19
9.7-20
9.7-21
9.7-22
9.7-23
9.7-24
9.7-25
9.7-26
9.7-27
9.7-28
9.7-29
9.8.1-1
'
Common Brick Manufacturing Using the Continuous Stiff-
Particle Size Distribution for Emissions from a Brick
Kiln and Dryer
Flow Diagram, Kiln-Fired Refactories Manufacturing
Generalized Lime Manufacturing Plant
Basic Flow Diagram of Portland Cement Manufacturing
Process Flow Diagram for Concrete Batching Showing
Potential Industrial Process Fugitive Participate
Material Flow for Representative 159-Mg/hr Asphalt
Concrete Plant ;
Particle Size Distribution in Uncontrolled Saturator
Schematic Diagram of Vertical Asphalt Blowing Still
Schematic Diagram of Asphalt Roofing Line
Typical Flow Diagram for the Manufacture of Soda-Lime
Glass
Regenerative Side Port Glass-Melting Furnace
Log-Probability Distribution of Particle Sizes Present
Typical Flow Diagram of Textile-Type Glass Fiber
Product ion Proces s
Typical Flow Diagram of Wool-Type Glass Fiber Production
Flow Diagram of Mineral Wool Process ing.
A Composite Flow Diagram for a Steel Plant
Page
9.7-43
9.7-4:6
9.7-49
9.7-50
9.7-53
9.7-63
9.7-73
9.7-85
9.7-89
9.7-97
9.7-99
9.7-101
9.7-110
9.7-111
9.7-116
9.7-131
9.7-131
9.7-135
.' 9.8-3
XI
-------
LIST OF FIGURES (continued)
*•
Figure Page
9.8.1-2 Flow Sheet Showing the Major Steps Involved in the
Carbonization of Coal Using the By-Product Process
and the Subsequent Recovery of Coal Chemicals from
the Gases Generated at the Ovens.. 9.8-7
9.8.1-3 Schematic Diagram of a Coke Battery 9.8-8
9.8.1-4 Schematic Diagram of Three Common Flue Designs 9.8-10
9.8.1-5 A Typical Aspiration System Operating in the Return
Bend of the Gas-Collection Duct... 9.8-15
9.8.1-6 Scrubber Efficiency Vs. Coal Feedrate Particulate Data... 9.8-29
9.8.1-7 Particle Sizing from Inlets of Windbox Secondary Control
Devices 9.8-32
9.8.1-8 Particle Sizing from Windbox Secondary Control Outlets
of Sampled Facilities 9.8-36
9.8.1-9 Particle Sizing from Baghouse Outlet Controlling Sinter
Processing Emissions at a Sintering Plant 9.8-45
9.8.1-10 Typical Blast Furnace 9.8-47
9.8.1-11 The Relationship Between Clean-Gas Dust Loading and
Pressure Drop Across Venturi Scrubber 9.8-51
9.8.1-12 Cross-Section of a Basic Open Hearth Furnace 9.8-53
9.8.1-13 Relationship of Electrostatic Precipitator Collecting
Surface to Collection Efficiency for Open Hearth
Emi ssi ons 9.8-57
9.8.1-14 Relationship Between Clean-Gas Dust Loading and Pressure
Drop for a Wet Scrubber on an Open Hearth Furnace 9.8-58
9.8.1-15 Basic Oxygen Furnace 9.8-60
9.8.1-16 Direct-Arc Electric Furnace 9.8-64
9.8.1-17 Ventilation Systems for Electric Arc Furnaces 9.8-68
9.8.2-1 Submerged Arc Furnace for Ferroalloy Production... 9.8-80
9.8.2-2 Open Furnace Controlled by a Venturi Scrubber..... 9.8-81
9.8.2-3 Sealed Furnace Controlled by Venturi Scrubber 9.8-82
xii
-------
LIST OF FIGURES (continued)
Figure Page
i
9.8.3-1 Conventional Lined Cupola. 9.8-93
9.8.5-1 Schematic Diagram of Primary Aluminum Production
Process 1 9.8-104
9.8.5-2 Aluminum Reduction Cell ^ 9.8-105
9.8.5-3 Average Composite Particle Size Distribution by
Weight for Reduction Cell Facility Roof Ventilator
Em ss ions 9.8-111
9.8.5-4 Schematic Layout of Alcoa 398 Process Reactor............ 9.8-115
9.8.6-1 Typical Primary Copper Smelter Flowsheet 9.8-121
9.8.6-2 Reverberatory Flue System Particulate Size Distribution.. 9.8-126
9.8.6-3 Average Cumulative Inlet and Outlet Mass Loading vs.
Particle Size, Copper Reverberatory Furnace .;...... 9.8-127
9.8.6-4 Converter Flue System Particulate Size Distribution 9.8-128
9.8.7-1 Primary Lead Smelting Process 9.8-139
9.8.7-2 Lead Blast Furnace : 9.8-141
9.8.8-1 Primary Pyrometallurgical Zinc Smelting Process.......... 9.8-155
9.8.8-2 Primary, Electrolytic Zinc Smelting Process 9.8-156
9.8.10-1 Raw Material and Product Flow Diagram for the
Secondary Copper Industry. 9.8-173
9.8.10-2 Air Pollution Control System in the Brass and Bronze
Industry 9.8-176
9.8.11-1 Secondary Lead Smelter Process » 9.8-184
9.8.11-2 Control Schemes for Secondary, Lead (a) Blast and
(b) Reverberatory Furnaces.... 9.8-187
9.8.11-3 Controlled Lead Pot and Ventilation System, Baghouse..... 9.8-189
9.8.12-1 Diagram Showing One Bank of a Belgium Retort Furnace..... 9.8-195
9.8.12-2 Diagram of a Distillation-Type Retort Furnace 9.8-196
9.8.12-3 Diagram of a Muffle Furnace and Condenser 9.8-197
xm
-------
LIST OF FIGURES (continued)
Figure Page
9.9-1 Plan of a Complex Refinery..., 9.9-2
9.9-2 Fluid Catalytic Cracking Process 9.9-5
9.9-3 Flow Diagram of Modern Fluid Coking unit 9.9-8
9.10-1 Typical Kraft Sulfate Pulping and Recovery Process.. 9.10-2
9.10-2 Indirect-Contact Recovery Furnace System 9.10-4
9.10-3 Particulate Concentrations in Control Systems Exhaust
from Kraft Recovery Furnaces 9.10-13
9.10-4 Particulate Emissions in Control System Exhaust from,
Smelt Dissolving Tanks used in the Kraft Pulping
Industry 9.10-14
9.10-5 Particulate Concentrations in Control System Exhaust
from Lime Kilns Used in the Kraft Pulping Industry 9.10-15
9.10-6 Simplified Process Flow Diagram of Magnesium-Base
Sulfate Pulp Process Employing Chemical and Heat
Recovery 9.10-19
9.10-7 Plywood Plant Process Diagram 9.10-23
9.11-1 Process Flow Diagram Showing Uncontrolled Lead
Emission Factors for Lead-Acid Battery Manufacture 9.11-3
9.12-1 Diagram of Street Surface/Atmospheric Exchange of
Particulate Matter 9.12-10
xiv
-------
LIST OF TABLES
Table Page
9.2-1 Selected Stack Parameters for Pulverized Coal-Fired
Utility and Industrial Boilers 9.2-3
9.2-2 Analyses of Flyash (FA) and Bottom Ash (BA) from Utility
Plants 9.2.5
i , • .
9.2-3 Design Parameters for Particulate Control Equipment .
Applied to Pulverized Coal-Fired Boilers 9.2-7
9.2-4 Size Specific Emissions from Pulverized Coal-Fired
Boi 1 ers...."... 9.2-9
9.2-5 Stack Parameters for Typical Stoker Coal-Fired Boilers.... 9.2-17
9.2-6 Ultimate Analysis of a Typical General Solid Waste 9.2-22
9.2-7 Uncontrolled Emissions from a Typical Large 525 GJ/Hr
(500 X 106 Btu/Hr) Heat Input1 MSW-Fired Boiler... 9.2-24
9.2-8 Uncontrolled Particulate Emissions from MSW-Fired
Boi ler 9.2-26
,9.2-9 Typical Characteristics of Refuse-Derived Fuels 9.2-28
9.2-10 Ultimate Analysis of Moisturer-Free Wood-Waste Fuel.. 9.2-31
9.2-11 Average Stack Parameters for Typical Wood/Bark-Fired
Boi 1 ers 9.2-35
9.2-12 Key Considerations of Particulate Emission Control
Devices for Wood-Fired Boilers.;... , 9.2-37
9.2-13 Efficiency of Granular Bed Filters in Series with
Cyclones on Hogged Fuel Boilers 9.2-44
9.2-14 Selected Stack Parameters of Oil-Fired Utility
and Industrial Boi 1 ers : 9.2-47
9.2-15 Selected Stack Parameters for Typical Oil-Fired
Industrial/Commercial Boilers. 9.2-51
9.3-1 Summary of Particulate Emission Test Data from
incinerators . 9.3-2
9.3-2 Municipal Incinerator Stack Parameter Data 9.3-5
9.3-3 Typical Electrostatic Precipitator Design Parameters for
Inci nerator Appl i cat ions. 9.3-8
xv
-------
LIST OF TABLES (Continued)
Table
9.3-4
9.3-5
9.3-6
9.3-7
9.3-8
9.3-9
9.4-1
9.5-1
9.5-2
9.5-3
9.5-4
9.5-5
9.5-6
9.5-7
9.5-8
9.5-9
9.5-10
9.5-11
9.5-12
Operating and Design Parameters for Fabric Filter
Industrial and Commercial Incinerator Stack Parameter
Data
Summary of Emission Test Data from Incinerators Burning
Summary of Uncontrolled Emission Test Data from Electric
Governmental, Commercial, and Industrial Sludge
Summary of 1973 State and Nationwide Agricultural Open
Typical Thermal Process Phosphoric Acid Stack Parameters..
Emission and Operating Data for Thermal Process
Particle Size Distributions in Selected Sulfuric Acid
PI ant Absorber Ef f 1 uents
Typical Applications Air Pollution Control Equipment......
Size Specific Emissions from a Borax' Fusing Furnace.......
U.S. Production of Synthetic Organic Pesticides, by
Usage Category, in 1974
Summary of Principal Air Emissions
Summary of Air Emission Control for Five Major Pesticides.
Sodi urn Carbonate Natural Processes
Typical Particulate Emission Sources and Controls for a
Emission Sources and Control Equipment for the
Page
9.3-10
9.3-12
9.3-19
9.3-20
9.3-21
9.3-26
9.4-2
9.5-13
9.5-14
9.5-17
9.5-26
9.5-28
9.5-29
9.5-30
9.5-31
9.5-32
9.5-34
9.5-38
9.5-39
XVI
-------
LIST OF TABLES (Continued)
Table '. Paqe
9.5-13
9.5-14
9.5-15
9.5-16
9.6-1
9.6-2
9.6-3
9.6-4
9.6-5
9.6-6
9.6-7
9.6-8
9.6-9
9.6-10
9.7-1
9.7-2
9.7-3
9.7-4
9.7-5
9.7-6
Emission Sources and Control Equipment for the Direct
Carbonati on Process
U.S. Potash Producers
Particle Size Distribution During Muriate Production
Size Specific Emissions fronv Potash and Salt Dryers
Summary of Neutralization Emission Data
Emission Test Results for Controlled Dryers in Ammonium
Sulfate Manufacturing Plants
Di ammonium Phosphate Production Stack Parameter Data
Grain Handling Operations Emission Factors
Calculated Energy Requirements to Operate Alternate
Grain Elevator Control Systems
Particulate Control Devices Applicable to Cotton Ginning
Operati ons
Cotton Ginning Stack Parameter Data
Emission Factors for Starch Manufacturing
Starch Manufacturing Stack Parameter Data
Nonmetallic Minerals Production Statistics
1977 Nationwide Particulate Emissions from the
Nonmetallic Minerals Mining and Processing Industry.......
Nonmetallic Mineral Unit Operations and Possible Sources
Particulate Emission Sources'for the Extraction and
Processing of Nonmetallic Minerals.....
Summary of Visible Emission Measurements from Fugitive
Noncrushing Sources Controlled by Wet Suppression
(According to EPA Method 22).
Air-to-Cloth Ratios for Fabric Filters Used for Exhaust
Emi ssion Control
9.5-40
9.5-41
9.5-45
9.5-47
9.6-4
9.6-11
9.6-18
9.6-21
9.6-22
9.6-23
9.6-31
9.6-33
9.6-36
9.6-37
9.7-2
9.7-3
9.7-7
9.7-9
9.7-12
9.7-19
xvn
-------
LIST OF TABLES (Continued)
Table
9.7-7
9.7-8
9.7-9
9.7-10
9.7-11
9.7-12
9.7-13
9.7-14
9.7-15
9.7-16
9.7-17
9.7-18
9.7-19
9.7-20
9.7-21
9.7-22
9.7-23
9.7-24
9.7-25
Baghouse Units Tested by EPA
Summary of Visible Emission Measurements from Fugitive
Sources at Non-Metallic Mineral Plants
Energy Requirements for Model Nonmetallic Minerals
Plants Having Crushing and Grinding Operations
Energy Requirements for Model Nonmetallic Minerals Plants
Having Crushing Operations Only...
Types of Emissions Generated by Metal Ore Processing
Methods of Particulate Emission Control Used at Metallic
Mineral Processing Plants...,
Metallic Minerals Processing Industry Characteristics
Particulate Emission Sources and Control Options for the
Processing of Metallic Minerals
Particulate Concentrations and Particle Size Data for
Metallic Minerals Processes
Uncontrolled Particulate Emissions from a Brick Kiln
Dryer
Ore Drying Emissions Reported in a Permit Application.....
Dry Mining Emissions
Composition of Particulate Matter from Natural Gas Fired
Lime Kilns
Lime Production Emission Size Distributions
Control Technologies Applicable to Emission Sources in
Rotary Lime Kiln Emissions
Lime Hydrator Emissions
Page
9.7-20
9.7-23
9.7-26
9.7-27
9.7-28
9.7-29
9.7-31
9.7-39
9.7-41
9.7-45
9.7-55
9.7-58
9.7-59
9.7-61
9.7-66
9.7-66
9.7-67
9.7-68
9.7-70
xvm
-------
LIST OF TABLES (Continued)
Table ; Page
9.7-26 Energy Consumption and Solid Waste Production Factors for
Control Devices Applied to Rotary Lime Kilns 9.7-72
9.7-27 Size Specific Emissions from A Wet Cement Kiln 9.7-75
9.7-28 Sources of Air Emissions in Cement Manufacturing Plants... 9.7-76
9.7-29 Size Distribution of Dust Emitted from Kiln Operations
Without Controls ; 9.7-77
9.7-30 Distribution of Kiln Dust Collection Systems in Wet and
Dry Process Cement Plants 9.7-79
9.7-31 Analysis of Portland Cement Plant Compliance Test Data 9.7-80
9.7-32 Advantages/Disadvantages of Control Devices for Various
Cement Manufacturing Operations 9.7-82
9.7-33 Emission Source Control's : 9.7-87
9.7-34 Asphalt Concrete Plant Exhaust Gas and Stack
Characteristics for the Rotary Dryer 9.7-90
9.7-35 Size Specific Emissions from A Rotary Dryer Asphalt
. Batchi ng PI ant : 9.7-91
9.7-36 Primary and Secondary Control Devices Used in the Asphalt
Hot Mix Industry 9.7-93
9.7-37 Hot Mix Asphalt Concrete Plants 9.7-94
9.7-38 Control Equipment Used on Asphalt Saturators 9.7-102
9.7-39 EPA Test Data at Asphalt Roofing Plants 9.7-103
9.7-40 Summary of Fugitive Emissions Data from Capture Systems.... 9.7-106
9.7-41 1976 Production Rates and Values of Shipments.. 9.7-109
9.7-42 Emissions from Uncontrolled Glass Melting Furnaces for
Each Industry Category... - 9.7-114
9.7-43 Standards of Performance for Gas-Fired Glass Melting
Furnaces ,. 9.7-117
9.7-44 All Electric Glass Melting Furnace Particulate Emissions
Tests 9.7-119
xix
-------
LIST OF TABLES (Continuted)
Table Page
9.7-45 Particulate Emissions Test Results for Glass Melting
Furnaces Equipped with Fabric Filters 9.7-121
9.7-46 Particulate Emission Test Results for Glass Melting
Furnaces Equipped with Venturi Scrubbers 9.7-123
9.7-47 Particulate Emission Test Results for Glass Melting
Furnaces Equipped with Electrostatic Precipitators 9.7-125
9.7-48 Representative Particulate Emissions from Glass Melting
Furnaces 9.7-127
9.7-49 Glass Type, Composition, Properties and Usages. 9.7-129
9.7-50 Mineral Wool Industry Emission Sources and Selected
Control s 9.7-137
9.8.1-1 Coke and Coal Chemicals Produced by U.S. Coke Plants in
1978 , 9.8-4
9.8.1-2 Stack Parameters for Byproduct Metallurgical Coke
Manufacturing Emission Sources.. 9.8-12
9.8.1-3 Size Specific Emissions from Coke Pushing 9.8-13
9.8.1-4 Comparison of Electrostatic Precipitator Installations.... 9.8-19
9.8.1-5 Performance Data for Electrostatic Precipitator Control
of Coke Oven Battery Stack Particulate Emissions...: 9.8-20
9.8.1-6 Summary of Test Results on Fabric Filters at Kaiser
Steel 9.8-22
9.8.1-7 Data on Uncontrolled Particulate Emissions from Pushing... 9.8-25
9.8.1-8 Particulate Emissions and Size Distribution from
Coke Pushing 9.8-28
9.8.1-9 Stack Parameters for Sintering Operations 9.8-31
9.8.1-10 Average Dry ESP Design Performance Parameters for
Collecting Sinter Machine Windbox Exhaust Emissions 9.8-34
9.8.1-11 Baghouse Operating Parameters for Collecting Sinter
Machine Windbox Exhaust Emissions 9.8-40
9.8.1-12 Average Scrubber Design Performance Parameters for
Collecting Sinter Machine Windbox Exhaust Emissions 9.8-42
xx
-------
LIST OF TABLES (Continued)
Table ; Page
9.8.1-13 Stack Parameters for Blast Furnace Iron Production
Operat i ons.. 9.8-48
9.8.1-14 Size Analysis of Flue Dust U.S. Blast Furnaces 9.8-49
9.8.1-15 Chemical Composition by Dry,;Blast-Furnace Flue Dust 9.8-49
9.8.1-16 Stack Parameters for Open Hearth Steel Production
Furnaces 9.8-54
9.8.1-17 Size Specific Emissions from,Open Hearth Steel
Production 9.8-55
9.8.1-18 Stack Parameters for Basic Oxygen Steel Production
Furnaces 9.8-61
9.8.1-19 Stack Parameters for Electric Arc Steel Production
Furnaces 9.8-65
9.8.1-20 Typical Chemical Analysis of;Electric Arc Furnace Fume.... 9.8-66
9.8.1-21 Stack Parameters for Steel Production Finishing
Operati ons. 9.8-70
9.8.1-22 Stack Parameters for Scarfing..............:. 9.8-71
9.8.2-1 Stack Parameters for Ferroalloy Production 9.8-83
9.8.2-2 Emission Factors for Ferroalloy Production in Electric
Smelting Furnaces 9.8-85
9.8.2-3 Particle Size Distribution of Particulates from
Ferroalloy. i 9.8-86
9.8.2-4 Standards of Performance for New Sources Ferroalloy
Production Facilities i... 9.8-87
9.8.2-5 Performance of Particulate Controls on Ferroalloy
Furnaces. 9.8-90
9.8.3-1 Size Specific Emissions from a Gray Iron Cupola 9.8-95
9.8.4-1 Particle-Size Distribution for Particulate Emissions
from Three Electric Arc Furnace Installations Melting
Gray Iron i 9.8-99
9.8.5-1 Results of EPA Source Tests for Particulates from Primary
Aluminum Reduction Cells and Anode Bake Plant...... 9.8-107
-------
LIST OF TABLES (Continued)
Table Page
9.8.5-2 Representative Particle Size Distributions of
Uncontrolled Emissions from Prebaked and Horizontal
Stud Soderberg Cells 9.8-108
9.8.5-3 Atmospheric Pollutants from Secondary Sources in
Al umi num PI ants 9.8-109
9.8.5-4 Industrial Process Fugitive Emission Sources and
Contami nants 9.8-110
9.8.5-5 Aluminum Ore Electroreduction Stack Parameter Data 9.8-113
9.8.5-6. Air Pollution Controls for Primary Aluminum Reduction
Potline Emissions 9.8-114
9.9.5-7 Summary of In-Plant Energy Consumption per Ton of
Aluminum for Pollution Control by Plant Type
(kil owatt-hours/ton) 9.8-117
9.8.6-1 Emission Factors for Primary Copper Smelters...... 9.8-125
9.8.6-2 Reverberatory Furnace ESP Performance Data 9.8-131
9.8.6-3 Performance Data for Spray Chamber/ESP 9.8-134
9.8.6-4 Performance Data for Spray Chamber/Baghouse 9.8-135
9.8.7-1 Emission Factors for Primary Lead Smelting Processes
Without Controls 9.8-142
9.8.7-2 Sintering Machine Flue Dust Size Distribution.. 9.8-143
9.8.7-3 Concentrations of Lead, Cadmium, and Zinc in Fugitive
Particulate Emissions of Various Primary Lead Smelting
Operations' Emissions Sources 9.8-145
9.8.7-4 Primary Lead Smelters—Sintering Machine Baghouse Test
Data 9.8-146
9.8.7-5 Lead Smelters--Blast Furnace Baghouse Test Data..... 9.8-147
9.8.8-1 Domestic Slab Zinc Capacity 1968-1972 9.8-151
9.8.8-2 Domestic Slab Zinc Consumption 9.8-152
9.8.8-3 Supply of Slab Zinc by Source 9.8-153
9.8.8-4 Emission Factors for Primary Zinc Smelting Without
Control s 9.8-157
xxii
-------
LIST OF TABLES (Continued)
Table Page
9.8.8-5 Typical Zinc Roasting Operations. 9.8-158
9.8.8-6 Components of the Oust from Multiple-Hearth, Suspension,
and Fluidi zed-Bed Roasters..... 9.8-159
9.8.9-1 Primary Air Emission Sources and Pollutants for
Secondary Aluminum Manufacturing Operations 9.8-164
9.8.9-2 Scrubber Collection Efficiency for Emissions from
Aluminum Chlorine Fluxing... 9.8-168
9.8.10-1 Composition of Particulate Matter Collected from a
Secondary Brass and Bronze Smelter.. 9.8-174
9.8.10-2 Summary of Air Pollution Control Equipment in Use in the
Secondary Copper Industry.., 9,8-177
9.8.10-3 Secondary Copper Industry Stack Parameter Data 9.8-178
9.8.10-4 Operating Specifications and Uncontrolled and Controlled
Emission Rates for the Typical Secondary Copper Smelting
and Refining Plant 9.8-179
9.8.10-5 Summary of Available Emission Data by Plant,and
Emission Source 9.8-181
9.8.11-1 Stack Parameters for Emission Sources Within the Lead
Smelting Industry ;....' 9.8-190
9.8.12-1 Stack Parameters for Emission Sources Within the
Secondary Zinc Processing Industry 9.8-198
9.8.12-2 Zinc Sweating Furnace Control Information 9.8-199
9.9-1 Petroleum Industry Emission^Sources and Controls. 9.9-3
9.9-2 Petroleum Industry Stack Parameters.. 9.9-4
9.10-1 Particulate Property Values
-------
LIST OF TABLES (Continued)
Table
9.10-5
9.10-6
9.10-7
9.10-8
9.11-1
9.11-2
9.11-3
9.11-4
9.12-1
9.12-2
9.12-3
9.12-4
9.12-5
9.12-6
9.12-7
9.12-8
9.12-9
9.12-10
Estimated Impact of Kraft Pulp Mill Parti cul ate Matter
Energy Impacts of Control Techniques for Kraft Pulp Mills
Stack Parameters for a Typical Sulfite Pulp Mill
1975 Nationwide Emissions of Lead from the Manufacture
of Lead-Acid Storage Batteries
Particulate Control Devices Used in the Lead-Acid
Battery Manufacturing Industry
Selected Control Alternatives for Lead-Acid Battery
Manufacturing Industry.
Summary of Alternative Control Systems Costs and Control
Particle Size Distribution of the Dust Emissions from
Agricultural Ti 1 1 i ng
Fugitive Dust Control Methods for Agricultural Sources....
Cost and Control Efficiencies for Fugitive Dust Control
Techniques for Open Fields.
Particle Size Distribution of the Dust Emissions from
Fugitive Dust Control Methods for Transportation Sources..
Initial Cost and Maintenance Cost of Alternative Road
Surfaces Applied by Maricopa County Highway Department....
Effectiveness and Cost of Alternative Measures to
Control Dust Emissions from Unpaved Parking Lots or Truck
Stops
Particle Sizes of Fugitive Dust Emissions from Aggregate
Particle Size Distribution for Dust Emissions from
Fugitive Dust Control Methods for Stockpiles and Waste
Page
9.10-11
9.10-17
9.10-20
9.10-22
9.11-2
9-11-7
9.11-9
9.11-10
9.12-3
9.12-4
9.12-7
9.12-11
9.12-13
9.12-15
9.12-16
9.12-19
9.12-21
9.12-22
XXIV
-------
LIST OF TABLES (.Continued)
Table :
9.12-11 Possible Control Technology Applications for Open Storage
Piles. 9.12-24
9.12-12 Effectiveness and Cost of Control Measures for Emissions
from Tailings Piles......... 9.12-26
9.12-13 Fugitive Dust Control Methods for Construction Sources.... 9.12-28
9.12-14 Cost of Alternative Dust Control Measures for
Construction Emissions in Phoenix Area 9.12-30
XXV
-------
CONVERSION FACTORS
1 meter (m) = 3.281 ft
1 meter (m) = 3.937 x 101 in.
1 meter2 (m2) = 1.076 x 101 ft2
1 meter3 (m3) = 1.308 yd3
1 meter3 (m3) = 3.532 x 101 ft2
1 meter/second (m/s) = 196.8 ft/min
1 meter/second (m/s) = 3.281 ft/s
1 meter3/second (m3/s) = 2.119 x 103 ft3/min
1 meter3/second (m3/s) = 1.585 x 105 gal (U.S. 1iquid)/min
1 meter3/second (m3/s) = 2.282 x 107 gal (U.S. 1iquid)/day
1 kilogram (kg) = 2.205 Ib
1 kilogram (kg) = 1.102 x 10~3 short tons (2000 Ib)
1 kilogram/meter3 (kg/m3) = 1.284 x 10"2 lb/ft3
1 kilogram/meter3 (kg/m3) = 8.98 x 101 grains/ft3
1 joule (J) = 9.479 x 10~4 Btu (mean)
1 joule (J) = 2.778 x 10"7 kWh
1 watt (W) = 1.340 x 10~3 hp
1 pascal (Pa) = 1.45 x 10~4 lbf/in.2 (psi)
1 pascal (Pa) = 4.019 x 10~3 in. H20
1 pascal second (Pa-s) = 0.672 Ib/ft2-s
1 kilopascal (kPa) = 0.4019 in. H20
XXVI
-------
TABLE OF SYMBOLS
Symbol Definition
a cross-sectional area
A collection plate area;of an electrostatic precipitator
A wetted surface area
C Cunningham correction factor
c dust loading
d diameter
d . drop diameter
df fiber diameter
d aerodynamic particle diameter
,D gas diffusivity
i
D particle diffusivity
E charging field strength
C*
E precipitation field strength
g acceleration of gravity
h height
Hd liquid hold-up
K inertia parameter i
K. inertia parameter at throat velocity
pi
K2 resistance coefficient of dust cake
1 bed depth
n number of plates or stages
xxvi i
-------
Symbol Definition
NRE Reynolds number
p total pressure
p static pressure
p corona power
P. penetration
q particle charge
Q, liquid throughput
Q gas throughput
r . drop radius
r. radius of hole
T absolute temperature
t time
tn residence time
v gas velocity
v pickup velocity
a solid fraction in fiber bed
A static pressure change
H efficiency
u gas viscosity
u. liquid viscosity
p gas density
p. liquid density
p particle density „
pn bulk resistivity
xxvi
-------
Symbol Definition
Pj In-situ resistivity
e porosity
w migration velocity
xxix
-------
GLOSSARY
ABSORPTION.1 Transfer of molecules from the bulk of the gas to a liquid
surface followed by diffusion to the bulk of the liquid.
ADIABATIC SATURATION.1 A process by means of which an air or gas stream is
saturated with water vapor without adding or subtracting heat from the
system.
AERODYNAMIC DIAMETER. The diameter of a unit density sphere having the same
aerodynamic properties as an actual particle.
AEROSOL. A dispersion of solid or liquid particles of microscopic size.
AGGLOMERATION. The combination of smaller particles due to collisions.
AIR, DRY. Air containing no water vapor.
AIR-TO-CLOTH RATIO (A/C). The volumetric rate or capacity of a fabric
filter; the volume of air (gas) cubic meter per minute, per square
meter of filter medium (fabric).
ATOMIZATION. The reduction of liquid to a fine spray.
BACK CORONA. Localized electrical breakdown of a dust layer, producing
positive ions, which degrade or neutralize the intended charging
process.
BAROMETRIC SEAL.1 A column of liquid used to hydraulically seal a scrubber,
or any component thereof, from atmosphere or other part of the system.
BLAST GATE.2 A sliding plate installed in a supply or exhaust duct at right
angles to the duct for the purpose of regulating air flow.
BLINDING (BLINDED).2 The loading, or accumulation, of filter cake to the
point where capacity rate is diminished.
BURNER.1 A device for the introduction of fuel and air into a furnace at
the desired velocities, turbulence, and concentration to establish and
maintain proper ignition and combustion of the fuel.
CASCADE IMPACTOR. A particle-sizing device in which progressively increas-
ing inertia! forces are used to separate progressively smaller particle
sizes.
xxx
-------
CHEVRON MIST ELIMINATOR.1 Series of diagonal baffles installed in a gas
stream, designed to separate fine droplets of liquid from the gas by
means of inertial impaction on the surfaces.of the baffles.
COCURRENT. Flow of scrubbing liquid in the same direction as the gas
stream.
COLLECTION EFFICIENCY.1 The ratio of the weight of pollutant collected to
the total weight of pollutant entering the collector.
CONDENSATION.1 The physical process of converting a substance from the
gaseous phase to the liquid or solid phase via the removal of heat, the
application of pressure, or both.
CONTACT CHARGING. Charging of particles by contacting them and then re-
leasing them from a charged surface.
CORONA CURRENT. Measure of the current flow from the transformer to its
electrical section in an electrostatic precipitator (ESP).
COUNT.2 The number of warp yarns (ends) and filling yarns (picks) per inch.
Also called thread count,
CROSSFLOW. Flow of scrubbing liquid normal to the gas stream.
CROWFOOT SATIN.2 A 3/1 broken twill arranged 2 threads right, then 2
• threads left. Also called 4 shaft satin, or broken crow weave.
CUNNINGHAM FACTOR. A correction factor to account for slippage of fine
particles moving through a discontinuous gaseous medium.
CURRENT DENSITY. Corona current level per unit area of collection surface
of an electrostatic precipitator (current per plate).
CYCLONE. A device in which the velocity of an inlet gas stream is trans-
formed into a confined vortex from which inertial forces tend to drive
particles to the wall. , •
DAMPER.2 An adjustable plate installed in a duct to regulate gas flow.
DEHUMIDIFY.1 Reduction of water vapor content of a gas stream.
DEMISTER. A mechanical device used to remove entrained water droplets from
a scrubbed gas stream.
DENIER.2 The number, in grams, of a quantity of yarn, measuring 9000 meters
in length. Example: A 200-denier yarn -measuring 9000 meters weighs
200 grams. A 200/80-yarn indicates a 200-denier yarn composed of 80
filaments. Usually used to describe continuous multifilament yarns of
silk, rayon, Orion, Dacron, Dynel, Nylon, and similar materials.
xxxi
-------
DENSITY.2 The ratio of the mass of a specimen of a substance to the volume
of the specimen. The mass of a unit volume of a substance.
DIELECTRIC STRENGTH. The maximum potential gradient that may exist in a
material without the occurrence of electrical breakdown.
DIFFUSION (AEROSOL). Random motion of particles caused by repeated colli-
sions of gas molecules.
DIFFUSION (LIQUID).1 The spontaneous intermingling of miscible fluids
placed in mutual contact, and accomplished without the aid of mechani-
cal mixing.
DIFFUSION CHARGING. Process of transferring electrical charge to particles
by random movement of electrons and ions; the effective charging mech-
anism for submicrometer-sized aerosols.
DIFFUSIOPHORESIS. Force acting on a particle, effecting movement due to a
vapor condensation gradient, resultant of differences in molecular
impacts on opposite sides of a particle.
DIMENSIONAL STABILITY.2 Capability of fabric to retain finished length and
width, under stress, in hot or moist atmosphere.
DRAFT.1 A gas flow resulting from the pressure difference between the
incinerator, or any component part, and the atmosphere, which moves the
products of combustion from the incinerator to the atmosphere. (1)
Natural draft: the negative pressure created by the difference in
density between the hot flue gases and the atmosphere. (2) Induced
draft: the negative pressure created by the vacuum action of a fan or
blower between the incinerator and the stack. (3) Forced draft: the
positive pressure created by the fan or blower, which supplies the
primary or secondary air.
DRAG FORCE. Resistance of a viscous medium due to relative motion of a
fluid and object.
DUST.2 Solid particles less than 100 micrometers created by the attrition
of larger particles.
DUST LOADING.2 The weight of solid particulate suspended in an airstream
(gas), usually expressed in terms of grains per cubic foot, grams per
cubic meter, or pounds per thousand pounds of gas.
DUST PERMEABILITY.2 The mass of dust (grains) per square foot of media
divided by the resistance (pressure drop) in inches water gauge per
unit of filtering velocity, feet per minute. Not to be compared with
cloth permeability.
ELECTROSTATIC FIELD. The position-dependent electrostatic force per unit
charge, made up of two components—one related to applied voltage and
electrode geometry, the other related to space change due to the pres-
ence of electrons, ions, and charged particles.
xxxi i
-------
ENTRAINMENT SEPARATOR (DEMISTER). That part of a gas scrubber designed to
remove entrained liquid droplets from a gas stream by centrifugal
action, by impingement on internal surfaces of the scrubber, or by a
bed of packing, mesh, or baffles at or near the scrubber gas outlet.
ENTRY LOSS.2 Loss in total pressure caused by air (gas) flowing into a duct
or hood. j
EXCESS AIR.1 Afr supplied for combustion in excess of that theoretically
required^, for complete combustion; usually expressed as percentage of
theoretical air (130% excess air).
FABRIC.2 A planar structure produced by interlacing yarns, fibers, or
filaments. (1) Knitted fabrics produced by interlooping strands of
yarns, etc. (2) Woven fabrics:are produced by interlacing strands at
more or less right angles. (3) Bonded fabrics or a web of fibers held
together with a cementing medium which does not form a continuous sheet
of adhesive material. (4) Felted fabrics or structures built up by the
interlacing action of the fibers themselves without spinning, weaving,
or knitting. :
FEEDSTOCK.1 Starting material used in a process. Can be raw material or an
intermediate product that will undergo additional processing.
FIELD CHARGING. Process of transferring electrical charge to particles
induced by high electric field* strengths in the interelectrode space;
the effective charging mechanism for particles greater than 1 micro-
meter. • . - - :
FIELD STRENGTH. A force field created by a large potential difference
between surfaces of different polarity; measured by the potential
difference divided by distance between surfaces.
FILAMENT.2 A continuous fiber.
FILL.2 Crosswise threads woven by loom. .
FILL COUNT.2 Number of fill threads per inch of cloth.
FILTER MEDIUM. The substrate support for the filter cake; the fabric on
which the filter cake is built.;
FILTER VELOCITY. The velocity at which the air (gas) passes through the
filter medium, or the velocity of approach to the medium. The filter
capacity rate. \
FLY ASH. Finely divided particles;0f ash entrained in flue gases arising
from the combustion of fuel. The particles of ash may contain unburned
, fuel and minerals. ;
xxxi i i
-------
FUGITIVE DUSTS. A type of participate emission made airborne .by forces of
wind, man's activity, or both—such as construction sites, tilled land,
or windstorms.3
FUGITIVE EMISSIONS. Particles generated by industrial or other activities
which escape to the atmosphere not through primary exhaust systems but
through openings such as windows, vents and doors, ill-fitting oven
doors, or poorly-maintained equipment.3
FUGITIVE EMISSIONS. Industrial process such as emissions from a point,
area, or line source other than a stack, flue, or control system.
Emissions escape to the atmosphere from a defined industrial process
flow stream because of leakage, materials charging/ handling, inade-
quate operational control, lack of reasonably available control tech-
nology, transfer, or storage.
FUME.1 Fine solid particles predominately smaller than 1 micrometer in
diameter suspended in a gas. Usually formed from high-temperature
volatilization of metals or by chemical reaction.
GALVANIC SERIES.1 A list of metals arranged according to their relative
tendencies to corrode. When dissimilar metals are joined together in
an electrolytic solution, the one closest to the "active" end of the
galvanic series corrodes preferentially to the one closest to the
"passive" end.
GRAVITY, SPECIFIC.2 The ratio of the mass of a unit volume of a substance
to the mass of the same, volume of a standard substance at a standard
temperature. Water is usually the standard substance. For gases,- dry
air at the same temperature and pressure as the gas is often the stan-
dard substance.
GRID.1 A stationary support or retainer for a bed of packing in a packed
bed scrubber.
HEADER.1 A pipe used to supply and distribute liquid to downstream outlets.
HUMIDITY, ABSOLUTE.2 The weight of water vapor carried by a unit weight of
dry air or gas.
HUMIDITY, RELATIVE.2 The ratio of the absolute humidity in a gas to the
absolute humidity of a saturated gas at the same temperature.
HYDROPHILIC MATERIAL. Particulate matter that adsorbs moisture.
INERTIA. Momentum; tendency to remain in a fixed direction, proportional to
mass and velocity.
INTERCEPTION. A type of aerosol collection related to impaction, in which
an aerosol impacts the side of an obstacle because of reduced mobility
across streamlines.
xxxiv
-------
INTERELECTRODE SPACE. The space between the discharge electrode and the
collection plate; the active particle-charging region in an electro-
static precipitator. '
INTERSTICES.2 The openings between the interlacings of the warp and filling
yarns; the voids.
ION, GASEOUS. A gas molecule that loses or gains one or more electrons.
IONIZATION. Generation of free electrons that become attached to gas-mole-
cules, forming ions.
ISOKINETIC SAMPLING. Matching the gas velocity at the sampling probe en-
trance to the gas velocity of the localized gas stream to collect a
representative particle size distribution.
LIQUOR.1 A solution of dissolved ^substance in a liquid (as opposed to a
slurry, in which the materials are insoluble).
LOG-NORMAL DISTRIBUTION. A series of points that can be defined by a geo-
metric mean value and a geometric standard deviation.
MEAN FREE PATH. The average distance between successive collisions of gas
molecules; related to molecular size and number per unit volume.
MIGRATION VELOCITY. The average drift velocity of charged particles normal
to the direction of gas movement; also known as precipitation rate
parameter, a measure of- the efficiency of collected particles to the
volume of gas treated and the area of the collection plate.
MOBILITY. A measure of response per unit force; the ease of motion relative
to the magnitude of the force-inducing motion.
MONOFILAMENT.2 A continuous fiber; of sufficient size to serve as yarn in
normal textile operations.
MULLEN BURST. The pressure necessary to rupture a secured fabric specimen.
MULTIFILAMENT (MULTIFIL).2 A yarn bundle composed of a number of filaments.
NAPPING PROCESS.2 A process to raise fiber Of filament ends (for better
coverage and more surface area), accomplished by passing the cloth over
a large revolving cage or drum of small power-driven rolls covered with
card clothing (similar to a wire brush).
NEEDLED FELT.2 A felt made by the placement of loose fiber in a systematic
alignment, with barbed needles moving up and down, pushing and pulling
the fibers to form an interlocking of adjacent fibers.
NONWOVEN FELT. A felt made by needling, matting of fibers, or compression
with a bonding agent for permanency.
xxxv
-------
OPACITY. Measure of the fraction of light attenuated by suspended particu-
late.
PARTICLE. Small discrete mass of solid or liquid matter.
PARTICLE SIZE. An expression for the size of liquid or solid particle.
PARTICIPATE MATTER. As related to control technology, any material except
uncombined water that exists as a solid or liquid in the atmosphere or
in a gas stream as measured by a standard (reference) method at speci-
fied conditions. The standard method of measurement and the specified
conditions should be implied in or included with the particulate matter
definition.
PARTICULATE MATTER, ARTIFACT. Particulate matter formed by one or more
chemical reactions within the sampling train.
PENETRATION. Fraction of suspended particulate that passes through a col-
lection device.
PERMEABILITY, FABRIC. The capability of air (gas) to pass through a fabric.
Measured on Frazier porosity meter'or Gurley permeometer. Not to be
confused with dust permeability.
PENTHOUSE (ESP).1 Weatherproof gas-tight enclosure over the electrostatic
precipitator that contains the high-voltage insulators.
pH.1 A measure of acidity-alkalinity of a solution; determined by calculat-
ing the negative logarithm of the hydrogen ion concentration.
PLAIN WEAVE.2 Each warp yarn passing alternately over each filling yarn.
The simplest weave, 1/1 construction; also called taffeta weave.
PLATE AREA. The effective area of both sides of the collecting surfaces in
an electrostatic precipitator.
POLYDISPERSITY. A particle size distribution consisting of different size
particles.
PRESSURE, STATIC. The pressure exerted in all directions by a fluid; mea-
sured in a direction normal to the direction of flow.
PRESSURE, TOTAL. The algebraic sum of the velocity pressure and the static
pressure.
PRESSURE, VELOCITY. The kinetic pressure in the direction of gas flow.
PRIMARY PARTICULATE MATTER. Particulate matter emitted directly into the
air from identifiable sources.
PRIMARY STANDARD. The national primary ambient air quality standard which
defines levels of air quality that are necessary to protect public
health.
xxxvi
-------
PRIME COAT (PRIMER).1 A first coat of paint applied to inhibit corrosion or
to improve adherence of the next coat.
QUENCH.1 Cooling of hot gases by rapid evaporation of water.
RAPPER (ESP). Device for imparting acceleration of the collecting surface
to dislodge the deposited particles.
RAPPER INSULATOR. A device that electrically isolates a rapper from the
high-voltage system of an electrostatic precip-itator, yet permits the
transmission of mechanical forces.
REFRACTORY.1 Ceramic material used for the lining of vessels, ducts, and
pipe for protection from heat, abrasion, or corrosion; also used for
insulation.
RESISTIVITY. The impedance offered to charge transfer across a dust layer;
defined by the ratio of electric field intensity to the current per
unit area passing through the dust layer.
REYNOLDS NUMBER, FLUID. A dimensionless quantity in fluids to describe the
ratio of inertial to viscous forces.
REYNOLDS NUMBER, PARTICLE. A dimerisionless quantity in aerosol science to
describe the ratio of inertial to viscous forces relative to the parti-
cle.
SATEEN.2 Cotton cloth made with a 4/1 satin weave, either as warp sateen or
filling sateen.
SATURATED GAS.1 A mixture of gas and vapor to which no additional vapor can
be added, at specified conditions. Partial pressure of vapor is equal
to vapor pressure of the liquid at the gas-vapor mixture temperature.
SATIN WEAVE.2 A fabric usually characterized by smoothness and luster.
Generally made warp face with a great many more ends than picks. The
surface consists almost entirely of warp (or filling) floats in con-
struction 4/1 to 7/1. The intersection points do not fall in regular
lines, but are shifted regularly or irregularly.
SECONDARY PARTICULATE MATTER. Particulate matter formed in the atmosphere
by physical and/or chemical gas-to-aerosol conversion mechanisms.
SECONDARY POLLUTANT. A pollutant not emitted into the air from a pollution
source, but formed in .the air |from the reactions of primary pollutants
(often photochemically).
SEEPAGE. The migration of particles through a freshly cleaned fabric.
SIZE DISTRIBUTION. Distribution of particles of different sizes within a
matrix of aerosols; numbers of particles of specified sizes or size,
ranges, usually in micrometers.:
xxxvi i
-------
SLURRY.1 A mixture of liquid and finely divided insoluble solid materials.
SMOKE. Small gasborne particles resulting from i incomplete combustion;
particles consist predominantly of carbon and other combustible mate-
rial; present in sufficient quantity to be observable independently of
other solids.
SNEAKAGE. Portion of a gas stream that bypasses the intended collection
area in an electrpstatic precipitator.
SOOT. An agglomeration of carbon particles impregnated with "tar," formed
in the incomplete combustion of carbonaceous material.
SPECIFIC GRAVITY.1 The ratio between the density of a substance at a given
temperature and the density of water at 4°C.
SPRAY NOZZLE.1 A device used for the controlled introduction of scrubbing
liquid at predetermined rates, distribution patterns, pressures, and
droplet sizes.
SPUN FABRIC.2 Fabric woven from staple (spun) fiber; same as staple.
»
STAPLE FIBER.2 Manmade fibers cut to specific length (lh in., 2 ft, 2% in.,
etc.); natural fibers of a length characteristic of fiber, animal
fibers being the longest.
STOKES NUMBER. Descriptive of the particle collection potential of a speci-
fic system; the ratio of particle-stopping distance to the distance a
, particle must travel to be captured.
STREAMLINE. The visualized path of a fluid in motion.
SUSPENDED PARTICULATE MATTER. Particulate matter in the ambient atmosphere,
as determined by a specific reference method; material generally refer-
red to as total suspended particulate (TSP); consists of particles
within the size range of 100 to 0.1 micrometer in diameter.
TENSILE STRENGTH.2 The capability of yarn or fabric to resist breaking by
direct tension. Ultimate breaking strength.
TEMPERATURE, ABSOLUTE.2 Temperature expressed in degrees above' absolute
zero.
TERMINAL SETTLING VELOCITY. The steady-state speed of a falling particle
after the equilibration of gravitation, drag, and buoyant forces has
occurred.
TRANSFORMER-RECTIFIER SETS. Electrical device used in electrostatic precip-
itators to rectify a.c. to d.c. and to transform low voltage to high
voltage.
xxxvi i i
-------
THREAD COUNT. The number of ends and picks per inch of a woven cloth.
TURBULENT FLOW. A type of flow in which the fluid passes in a nearly ran-
dom, fluctuating motion. :
TWILL WEAVE.2 Warp yarns floating.over or under at least two consecutive
picks from lower to upper right, with the point of intersection moving
one, yarn, either outward and upward or downward on succeeding picks,
causing diagonal lines in the cloth.
VAPOR. The gaseous form of substances that are normally in the solid or
liquid state and whose states can be changed either by increasing the
pressure or by decreasing the temperature.
WARP.2 Lengthwise threads in loom or cloth.
WARP COUNT.2 Number of warp threads per inch of width.
WET/DRY LINE.1 The interface of hot, dry particulate-laden gas and cooling
or scrubbing liquid, at which an accumulation of solids can occur.
WOVEN FELT.2 Predominantly a woven' woolen fabric heavily fulled or shrunk,
with the weave completely hidden by the entanglement of the woolen
fibers.
GLOSSARY
REFERENCES
1. Industrial Gas Cleaning Institute. Wet Scrubber Terminology. Publica-
tion WS-1, July 1975. .
2. Industrial Gas Cleaning Institute. Fundamentals of Fabric Collectors
and Glossary of Terms. Publication F-2, August 1972.
i
3. PEDCo Environmental, Inc. Technical Guidance for Control of Industrial
Process Fugitive Particulate Emission. EPA-450/3-77-010, March 1977.
xxxix
-------
Page intentionally Blank
-------
SECTION 9
SOURCES OF PARTICULATE EMISSIONS AND CONTROL TECHNIQUES
Section 9, which is Volume 2, describes specific stationary sources and
applicable control technologies. To the degree available, information is
provided regarding: '
o Industry, process, or source description
o Emission characteristics and applicable control technology
o New Source Performance Standards (NSPS) promulgated under Section
111 of the Clean Air Act
o Secondary environmental impacts
Costs of particulate control systems, their energy requirements, and
methods of handling and disposing of liquid and solid wastes are not pre-
sented in detail in Section 9. These subjects are discussed in detail in
Sections 6 and 7 of Volume 1 and, more specifically, in a number of the
references cited for the industry being described.
Criteria for selection of source categories are presented in Section
9.1. Detailed descriptions are presented in Sections 9.2 through 9.12.
9.0-1
-------
Page Intentionally
-------
9.1 STATIONARY SOURCE SELECTION
Approximately eighty (80) industrial and/or process-specific sources of
particulate matter have been selected for discussion. The selection was
based on one of the following guidelines:
o The average plant/source emitted more than 91 megagrams
(100 tons) of particulate matter per year.
o The EPA has promulgated or is planning to promulgate a New Source
Performance Standard (NSPS) for the source.*
The selected sources are grouped! into eleven categories, namely:
o Stationary combustion sources o Metallurgical industry
o Refuse incineration
o Open burning
o Chemical process industry
o Food and agriculture processes
o Mineral products industry
o Petroleum industry
o Forest products industry
o Lead-acid battery manufacturing
o Fugitive dust sources ,
9.1-1
-------
REFERENCES FOR SECTION 9.1
Monarch, M. R., et al. Priorities for New Source Performance Standards
Under the Clean Air Act Amendments of 1977. Argonne National Laboratory.
Publication No. EPA-450/3-78-019. April 1978.
9.1-2
-------
9.2 STATIONARY COMBUSTION SOURCES
This section reviews particulate! control technologies for stationary
combustion sources which are categorized by primary fuel (coal, refuse,
oil), firing mechanism, and application. Six categories of stationary com-
bustion sources are reviewed. These Include: pulverized coal-fired boilers,
stoker fed coal-fired boilers, coal-fired cyclone furnaces, nonfossil-fuel-
fired boilers, oil-fired utility boilers, oil-fired industrial/commercial
boilers.
9.2.1 Pulverized Coal-Fired Boilers >
Pulverized coal-fired boilers are used to produce steam for electric
power generation and for large industrial operations. As of late 1977, the
total capacity of fossil fuel-fired electric utility steam generators was
380 gigawatts.l During 1977, coal provided about 62 percent of the electric
power derived from fossil fuel-fired steam generators.1 Individual units
range from 15 to 1300 megawatts of power output. In 1978, particulate emis-
sions from electric utility coal-firejd steam generators were about 2,400
gigagrams.2 Nearly all new coal-fired electric utility steam generators
are pulverized coal-fired units.3. Figure 9.2-1 shows the components of a
typical pulverized coal-fired boiler.4
9.2.1.1 Source Description. In: general, particulate matter is produced
along with thermal energy and other combustion products in the boiler combus-
tion chamber. The walls of the combustion chamber are lined with water-filled
tubes where thermal energy is absorbed and steam is generated. Combustion
products flow from the combustion chamber to superheat and reheat sections
where further thermal energy is transferred to the steam. Upon leaving
these sections, the combustion products typically flow to an economizer and
then an air preheater before passing through pollution control devices, an
induced draft fan, and out the stack., Table 9.2-1 gives selected stack para-
meters for utility pulverized coal-fired boilers.5,6
9.2.1.2 Emission Characteristics and Applicable Control Technologies.
Coal combustion produces solid wastes* These materials consist of inorganic
mineral constituents in the charged fuel as well as any unburned organic mat-
ter. The inorganic material can be present in the fuel in concentrations
of from 3 to 30 percent. During combustion, the solid inorganic material
(ash) is divided into two fractions: bottom ash, collected from the bottom
9.2-1
-------
Figure 9.2-1 Pulverized coal-fired boiler with burners mounted in front
wall. t
9.2-2
-------
Table 9.2-1. SELECTED STACK PARAMETERS FOR PULVERIZED COAL-FIRED
UTILITY AND INDUSTRIAL BOILERS5.6
Stack Stack • Stack Flue gas volumetric
Boiler Size, height, diameter, temperature, flowrate,
type MW m m K AnrVmin
Utility 300 175 5.82 ' 400 32,000
Industrial 70 55 2.8 470 9,800
9.2-3
-------
of the boiler in the ash hopper; and flyash, collected in the mechanical
collectors, electrostatic precipitators, or high efficiency participate
matter control devices. Organic and inorganic gas phase pollutants are also
contained in the combustion gas. Gas phase organics include low molecular
weight unburned hydrocarbons. Inorganic constituents in the gas phase in-
clude nitrogen oxides, hydrochloric acid, and sulfur dioxide. Vaporized
metals such as mercury can also be present. Discussion of the control of
these pollutants is outside the scope of this document. References 7 and 8
provide additional information on the products of coal combustion.
The distribution of ash between the bottom ash and flyash fractions is
a function of the type of boiler, the coal type, and the boiler configura-
tion. A comparison of the typical chemical composition of each fraction is
presented in Table 9.2-2. Pulverized coal-fired boilers produce 60 to 85
percent flyash and 15-40 percent bottom ash. This range is dependent upon
the ash fusion temperature and whether the boiler is a wet or dry type. Wet
bottom boilers are designed to process more slag (bottom ash before cooling)
and therefore generate less flyash.9
Flyash is primarily made up of particles ranging in size from 0.5 to
100 micrometers. Approximately 20 percent by volume of the flyash consists
of light particles called cenospheres. Cenospheres are hollow silicate
glass spheres filled with nitrogen and carbon dioxide gas. They range in
size from 20 to 200 micrometers and float when sluiced to a settling pond.
The major chemical constituents of flyash and cenospheres are silicon,
aluminum, iron, and calcium. These elements make up approximately 95 to 99
percent of the total weight of the ash.
Of the control technologies described in Section 4 of Volume 1, the most
commonly used high efficiency device on pulverized coal-fired utility boilers
is the electrostatic precipitator. Fabric filters or scrubbers are also
used. Mechanical collectors, such as settling chambers and cyclones, have
only a limited effect on fine particulates; they will not reduce particu-
late emissions sufficiently to meet state and local regulations.5 However,
mechanical collectors are sometimes installed upstream of high efficiency
control devices in order to reduce the burden on subsequent equipment.
The selection of a particulate control technology for a particular ap-
plication is influenced by the physical and chemical constituents of the coal
9.2-4
-------
Table 9.2-2. ANALYSES OF FLYASH (FA) AND BOTTOM ASH (BA) FROM UTILITY PLANTS^
ro
en
Compound
or
element
Si02, %
A1203,%
CaO, i
S03,%
MgO
NaoO
K20
P205,%
T]02,%
As, ppm
Be, ppm
"CA; ppra
Cr, ppm
Cu, ppm
Hg, ppm
Mn, ppm
Ni, ppm
Pb, ppm
• Se, ppm
V, ppm
Zn, ppm
B
Co, ppm
F, ppm
Plant
FA
59.0
27.0
3.8
3.8
0.4
0.95
1.88
0.9
0.13
0.43
12.0
4.3
0. 5
20.0
54.0
0.07
267.0
10.0
70.0
6.9
90.0
63.0
266.0
7.0
140.0
1
BA
58.0
25.0
4.0
4.3
0.3
0.88
1.77
0.8
0.06
0.62
1.0
3.0
0.5"
15.0
37.0
0.01
366.0
10.0
27.0
0.2
70.0
24.0
143.0
7.0
50.0
Plant
FA
57.0
20.0
5.8
5.7
0.8
1.15
1.61
1.1
0.04
1.17
8.0
7.0
0.5
50.0
128.0
0.01
150.0
50.0
30.0
7.9
150.0
50.0
200.0
20.0
100.0
2
BA
59.0
18.5
9.0
4.8
0.3
0.92
1.01
1.0.
0.05
0.67
1.0
7.0
0.5
30.0
48.0
0.01
700.0
22.0
30.0
0.7
85.0
30.0
125.0
12.0
50.0
Plant
FA
43.0
21.0
5.6
17.0
1.7
2.23
1.44
0.4
0.70
1.17
15.0
. 3.0 ..
0.5
150.0
69.0
0.03
150.0
70.0
30.0
18.0
150.0
71.0
300.0
15.0
610.0
3
BA
50.0
17.0
5.5
13.0
0.5
1.61
0.64
0.5
0.30
0.50
3.0
2.0
0.5
70.0
33.0
0.01
150.0
15.0
20.0
1.0
70.0
27.0
70.0
7.0
100.0
Plant
FA
54.0
28.0
3.4
3.7
0.4
1.29
0.38
1.5
1.00
0.83
6.0
. .7.0
1.0
30.0
75.0
0.08
100.0
20.0
70.0
12.0
100.0
103.0
700.0
15.0
250.0
4
BA
59.0
24.0
3.3
3.5
0.1
1.17
0.43
1.5
0.75
0.50
2.0
. 5.0 .
1.0
30.0
40.0
0.01
100.0
10.0
30.0
1.0
70.0
45.0
300.0
7.0
85.0
Plant
FA
NR
NR
20.4
3.2
NR
NR
NR
NR
NR
NR
8.4
8.0 ..
6.44
206.0
68.0
20.0
249.0
134.0
32.0
26.5
341.0
352.0
NR 0
6.0
624.0
5
BA
NR
NR
30.4
4.9
0.4
NR
NR
NR
NR
NR
5.8
7.3
1.08
124.0
48.0
0.51
229.0
62.0
8.1
5.6
353.0
150.0
NR 0
3.6
10.6
Plant
FA
42.0
17.0
17.3
3.5
NR
1.76
1.36
2.4
NR
1.00
110.0
NR - -
8.0
300.0
140.0
0.05
298.0
207.0
80.0
25.0
440.0
740.0
NR
39.0
NR
6
BA
49.0
19.0
16.0
6.4
NR
2.06
0.67
1.9
NR
0.68
18.0
NR -
1.1
152.0
20.0
0.028
295.0
85.0
6.2
0.08
260.0
100.0
NR
20.8
NR
-------
to be burned, governing regulations, and disposition of the collected material.
The present energy policy of the United States (i.e., increased coal consump-
tion), coupled with the required control of sulfur oxides in the flue gas, has
resulted in the increased use of low sulfur coal. Combustion of low sulfur
coal usually leads to the formation of an ash highly resistant to collection
in conventional electrostatic precipitators located after the air preheater
(see Volume 1, Section 4.3). Existing electrostatic precipitators often
operate at significantly reduced efficiencies when users switch from high
sulfur to low sulfur coal.-'-0
To overcome this problem, the user has had to take one or more of the
following steps: (1) install a larger precipitator, (2) install the precipi-
tator upstream of the air preheater (because of reduced resistivity at gas
temperatures at this point), (3) condition the ash, and (4) install different
control hardware. The size of the precipitator must be increased to compen-
sate for the decreased migration velocity. Ash conditioning reduces resis-
tivity through the addition of chemicals such as sulfur trioxide, sulfuric
acid, or ammonia to the gas stream; these chemicals are adsorbed on the
surface of the ash, thereby increasing its conductivity.
Finally, the use of other control devices may be an alternative to pre-
cipitator collection. Scrubbers have been used to control particulate matter,
although the typical use of wet scrubbers is in the control of sulfur dioxide.
The greatest drawbacks to scrubbers are the large energy requirement necessary
to attain the efficiency levels required (see Volume 1, Section 4.4). Fabric
filters also have recently experienced increased use for particulate emissions
control; they are unaffected by flyash resistivity and can meet present stan-
dards for performance.
Regardless of the control device selected, the effectiveness of the
applicable control technology is dependent on the equipment's design. Table
9.2-3 gives characteristic operating information for high efficiency electro-
static precipitators, fabric filters and scrubbers.5*11.12,13,14 Tne effi-
ciency information given in Table 9.2-3 applies to total particulate emissions
control.
Table 9.2-4 presents controlled and uncontrolled emission data from four
coal-fired boilers.15 These limited data show that the efficiency of the
control devices used on these sources generally decreases with decreasing
9.2-6
-------
Table 9.2-3. DESIGN PARAMETERS FOR PARTICIPATE CONTROL EQUIPMENT APPLIED
TO PULVERIZED COAL-FIRED BOILERS5'11'12'13'14
Control Design
technique parameter
Electrostatic SCA (high sulfur coal,
precipitators cold side application)
SCA (low sulfur coal ,
cold side application)
SCA (low sulfur coal,
hot side application)
Collection area per
rapper
Plate spacing j
Plate height
Pressure drop
Migration velocity
Rapping method
Efficiency
Fabric filters Filtering velocity |
(air-to-cloth ratio)
Bag fabric
Cleaning method
Typical or
range of values
75 to 100 m2 per
Am-Vmin
115 to 130 m2 per
Am-Vmi n
60 to 70 m2 per
Am^/mi n
135 to 230 m2
0.23 to 0.35 m
7.3 to 13.7 m
Less than 0.25 kPa
15 to 60 cm/sec
Pneumatic, electro-
magnetic, or
mechanical
Greater than 99 percent
0.6 to 0.9 m/min
Glass fabrics
Mechanical shaking,
reverse air flow
Reference
5
5
5
—
11
11
5,12
__
11
•
5
5
13
9.2-7
-------
Table 9.2-3 Concluded. DESIGN PARAMETERS FOR PARTICIPATE CONTROL
EQUIPMENT APPLIED TO PULVERIZED COAL-
FIRED BOILERS
Control
technique
Fabric filters
(continued)
Wet scrubber
(venturi)
Wet scrubber
(moving bed)
Wet scrubber
(performed
spray)
Design
parametar
Pressure drop
Bag diameter
Bag length
Efficiency
Liquid-to-gas
ratio
Pressure drop
Efficiency
Liquid-to-gas
ratio
Pressure drop
Efficiency
Li quid- to- gas
ratio
Pressure drop
Efficiency
Typical or
range of values
Less than 1.5 kPa
0.2 to 0.3 m
6.1 to 12.2 m •
Greater than 99 percent
1.3 to 2.0 L/Am3
4.0 to 7.C kPa
95 to 99 percent
7.0 L/Am3
2.5 to 4.0 kPa
95 to 99 percent
1.3 to 6.7 L/Am3
1.0 to 1.5 kPa
95 to 99 percent
Reference
5
13
13
--
14
5, 14
_-
14
6
--
1
6
--
9.2-8
-------
Table 9.2-4 SIZE SPECIFIC EMISSIONS FROM PULVERIZED COAL-FIRED BOILERS15
ro
i
ID )
Control
device
operating
parameter Mass concentration, liig/DNCM (mass percent less than stated size)
Total 15.3 m 12.9 m 10.1 m 7.28 m 5m 2.5 m 1.01 m
Scrubber Uncontrolled 3925 993(25.3) 848(21.6) 675(17.2) 506(12.9) 372(94.8) 201(5.12) 57.5(1.47)
2.5 kPa Controlled 105 101(95.7) 100(95.4) 100(94.9) 99.7(94.2) 98.7(93.3) 96(90.8) 87.4(82.6)
pressure drop3
Efficiency 97.0 89.8 88.1 85.2 80.3 73.5 52.2 NA
Scrubber Uncontrolled 4360 1330(30.5) 1180(27.1) 1070(24.5) 966(22.1) 836(19.1) 352(8.07) 87.5(2.0)
4.2 kPa Controlled 31.5 29.9(94.9) 29.3(92.9) 29.1(92.4) 29.1(92.4) 29.1(92.3) 29.0(92.0) 25.5(80.9)
pressure dropb
Efficiency 99.3 97.7 97.5 97.3 96.9 96.5 91.8 70.9
ESP
60 m2/
M3/min
specific
collection
areac
Uncontrolled
Controlled
Efficiency
2933
54.9
98.1
1100(37.5)
49.2(89.6)
95.6
1040(35
45.3(82.
95.6
.5) 990(33
4) 36.4(66
96.3
.6)
.2)
918(31.2)
31.9(58.1)
96.5
826.3(28.0)
28.1(51.2)
96.6
519(17.
21.5(39.
95.8
7)
2)
400(13.6)
11.8(21.4)
97.1
Fabric filter Uncontrolled 3164 1060(33.5) 883(27.8) 591(18.7) 358(11.3) 262(8.27) 145(4.6) 85.4(2.7)
0.85 Controlled 8.01 6.42(80.1) 4.77(59.6)4.19(52.4)3.43(42.8)2.56(32.0)1.6(20.4)0.983(12.0)
M3/M2 min
air to cloth Efficiency 99.7 99.4 99.4 99.3 99.0 99.0 98.8 98.8
ratiod
3The Fine Particulate Emissions Information System (FPEIS) number is 51.
bThe FPEIS number is 130.
CThe FPEIS number is 122.
dThe FPEIS number 1s 35.
-------
particle size. More extensive data on participate emissions and emission
control are given in References 5 and 16.
New Source Performance Standards (NSPS) have been promulgated by EPA for
fossil fuel-fired steam generators with heat inputs greater than 73 megawatts
(250 million Btu/h). Standards promulgated in 1971 limit particulate emissions
to 43 nanograms/joule (0.10 Ib/million Btu) heat input. The NSPS was revised
in 1979 as it applies to steam used to generate electric power. For these
units, particulate emissions are limited to 13 ng/j (0.03 Ib/million Btu)
heat input.
The NSPS also established limits for emissions of S02 and NOX from the
same boilers. These restrictions have the effect of preventing the formation
of secondary particulates in the atmosphere.
9.2.1.3 Secondary Environmental Impacts. Implementation of particulate
emissions control has secondary environmental impacts on land and water quali-
ty. The amount of ash to be disposed of is a direct function of: the concen-
tration of the ash in the fuel; the quantity of fuel consumed; the distribu-
tion of ash between the bottom ash and flyash fractions; and the effectiveness
of control. Figure 9.2-2 shows flyash generation as a function of ash content
and control system efficiency.1? The primary environmental impact from dis-
posal of ash is the land requirement associated with disposal. The prevalent
method for disposal of flyash collected by dry collectors is wet-sluicing it
from ESPs or fabric filters to on-site ponds. The water requirements for wet-
sluicing can range from 5,000 to 165,000 L/Mg of ash. Scrubber sludge can
also be disposed of in on-site ponds, or it may be combined with collected
flyash and used as a landfill.18
Ash ponds may be lined or unlined. The usual lining material is clay;
synthetic plastic or rubber liners are also commercially available. Poten-
tial leaching of trace elements in the ash to ground water is a major con-
cern, and proper lining is essential to prevent this from happening.
Ash utilization is becoming more prevalent. In 1974, 8.4 percent of
the flyash, 20.3 percent of the bottom ash, and 50.0 percent of the boiler
slag generated in the United States was used primarily in: Portland cement
and asphalt concrete mixtures, road surfacing, and other miscellaneous
products.18
9.2-10
-------
400
300
s-
(O
CD
2 200
CfL
LU
CD
s
100
Dry Solids Basis
10% Bottoms
10,000 Btu/lb
10 Tons Coal/MW-day
Collection Efficiency of
Electrostatic Precipitator
5 10 15 20
ASH CONTENT, percent by weight
25
Figure 9.2-2 Flyash generation from coal-fired boilers.17
-------
Sludge disposal from wet scrubbers usually involves transporting the
slurry to settling ponds, then recycling the separated scrubber solution.
Again, proper lining of settling ponds should reduce leaching of trace ele-
ments into the ground or ground water. Operation of these scrubber systems,
as well as flyash sluicing systems, in total recycle (closed loop) will help
reduce contamination of streams or ground water.
9.2.2 Stoker Fed Coal-Fired Boilers
Stoker boilers account for nearly all coal-fired boilers used by indus-
try in the range of 2.9 to 73 MW heat input.19 Small pulverized coal-fired
boilers represent the balance. Stokers are classified as follows by the
methods used to introduce fuel into the furnace:
o Spreader
o Underfeed
o Overfeed
9.2.2.1 Source Description. Spreader stokers are generally larger than
all other stoker types. In spreader stokers, coal is injected midway into
the furnace above a burning fuel bed. Spreader stokers can be distinguished
by the type of grate design: stationary and dumping grate, traveling grate,
or vibrating grate. Stationary and dumping grates are used mostly in small
and medium-sized boilers. The traveling grate, illustrated in Figure 9.2-3,
is generally used in large spreader stokers. The use of spreader stokers as
industrial equipment to burn coal has increased constantly in recent years,
and this use should continue to rise in future years as oil and gas supplies
dwindle.19'20
Underfeed stokers introduce 'coal through a retort beneath the burning
fuel bed. These stokers can be classified as either: single-retort horizon-
tal-feed, or multiple-retort gravity-feed. Figure 9.2-4 is a schematic of a
single-^retort underfeed stoker. The coal is moved from the trough (retort)
toward the rear of the boiler, then upward to spread over the air-admitting
tuyeres. Spent fuel is forced to the side dumping grates.
Overfeed stokers feed the coal from above onto a continuously moving
bed. The coal enters the furnace on the grate and continues to burn as it
moves along the fuel bed, becoming progressively thinner until only ashes
remain. There are two basic kinds of overfeed stokers: chain or travel-
ing grate (Figure 9.2-5), and water-cooled vibrating grate.
9.2-12
-------
- ro
i
FEEDER
COAL
HOPPER
Figure 9.2-3 Traveling grate spreader-stoker with front ash discharge.^
-------
ro
i
FORCED-
DRAFT
FAN
USHER I PUSHER
BLOCO^ROD
Figure 9.2-4
Single-retort, horizontal-feed, side ash discharge
underfeed stoker.^
-------
s-
OJ
-*:
o
+J
t/1
S-
OJ
o
OJ
+•>
(O
Ol
(O
^:
o
LO
CM
• '
cr>
9.2-15
-------
With each type of stoker, combustion products pass through or around
tubes which transfer thermal energy to the boiler water or steam. Use of
superheat sections, reheat sections, economizers, or air preheaters varies
with the size and application of the boiler. Thus, combustion product con-
ditions vary both prior to a particulate emissions control device and at the
stack exit. Table 9.2-5 presents typical stack parameters for stoker coal-
fired boilers.
9.2.2.2 Emission Characteristics and Applicable Control Technologies.
Particulate emissions from stoker fed coal-fired boilers may consist of un-
burned carbon, condensable tars, and flyash. Generally, coal-fired stokers
produce less particulate matter than pulverized coal-fired units. Typically,
about 65 percent of the total ash from such boilers is emitted as flyash with
the balance discharged as bottom ash.21
There are four major equipment types available for controlling particu-
late emissions from stoker fed coal-fired boilers: electrostatic precipita-
tors, mechanical collectors, fabric filters, and wet scrubbers. As in the
case of the pulverized coal boilers, combinations of the above may be used,
usually with a cyclonic collector upstream of one of the higher efficiency
devices.
Selection of the appropriate particulate emissions control technique is
more difficult than for the larger pulverized coal-fired boilers. In indus-
tries where stoker boilers are used, operating loads and requirements are
variable. In addition, industry is less likely to purchase coal on long-term
contracts. Thus, the type of coal being burned and the load on the boilers
may vary drastically.
Fabric filters and electrostatic precipitators are used for high effi-
ciency particulate control. In the past, when greater particulate emissions
were allowed, lower efficiency mechanical collectors such as cyclones were
used extensively. Often it is practical to leave such collectors in place
and add a high efficiency fabric filter or an electrostatic precipitator
downstream. Scrubbers are not widely used for energy and cost reasons.22,23
Effectiveness of the four types of particulate control techniques is dis-
cussed in the following paragraphs. More extensive data are given in
Reference 24.
9.2-16
-------
Table 9.2-5. STACK PARAMETERS FOR TYPICAL STOKER COAL-FIRED BOILERS^
Boiler
type
Spreader
Overfeed
Underfeed
Size,
MW
>29.3
2.9-29.3
<2.9
>29.3
2.9-29.3
<2.9
2.9-29.3
<2.9
Stack
height,
m
45
36
21
68
42
35
41
21
Stack
: diameter,
\ m
2.5
1.9
0.9
2.9
2.2
, 1.7
! 2.2
1.0
Stack
temperature,
K
475
515
415
490
510
470
510
455
Flue gas
volumetric
flowrate,
Am-Vmin
2575
1065
405
4010
1550
1560
1295
290
9.2-17
-------
Mechanical Collectors—Multitube cyclones, which represented the most
common type of inertial collector used for flyash collection before stricter
emission regulations were enacted, are relatively inefficient, especially
for particles of less than 10 micrometers.25 Efficiencies over 85 percent re-
moval by weight are uncommon. Efficiencies of 65 to 70 percent and pressure
drops of 1 kPa are typical.
Fabric Filtration—Fabric filtration systems are gaining in favor be-
cause they provide high efficiencies at moderate pressure drops.10 Filter
bags usually range from 14 to 18 cm in diameter and 3.5 to 6.0 m in length.
Pressure drops of 0.8 kPa to 1.5 kPa and air-to-cloth ratios in the range of
3.5 to 1 and 5 to 1 are typical. Efficiencies of fabric filtration systems
are generally around 99.8 percent removal by weight. Fiberglass is the most
widely used bag material, and reverse air or pulse jet cleaning is usually
used.13
Electrostatic Precipitators--ESP modules can be furnished in sizes
down to 140 Am3/min; high efficiencies (99.8 to 99.9 percent by weight)
are obtainable. Improperly designed or operated rapper systems can lead
to significant mass emissions through reentrainment of coarse particles
during the plate rapping (see Volume 1, Section 4.3.3).26
Plate spacings range from 20 to 30 cm, with plate heights between 7.6
and 13.7 m; rappers are generally mechanical, although they may be pneumatic
or electrical. Specific collection areas vary, depending upon the removal
efficiency desired, but they are roughly 100 m^ per Am^/s.H
Wet Scrubbers—These devices are not widely used to control flyash
emissions from stokers. Three types are applicable, namely, venturi, high-
pressure spray, and moving bed absorbers. Generally, particulate matter
removal efficiencies can exceed 95 percent by weight with a pressure drop
of 2.5 to 5 kPa and liquid-to-gas ratios around 2 L/Am3.10>14
9.2.2.3 Secondary Environmental Impacts. Secondary environmental im-
pacts of particulate matter collection from stoker boilers are comparable to
impacts from pulverized coal boilers (see Section 9.2.1.3). Water pollution
and landfill problems at flyash disposal sites are the primary concerns. If
the flyash is transported by a hopper sluicing system, a settling pond or
other solid/liquid separation device is used. Pond liners and controlled
procedures for discharging, evaporating, or recycling liquids are used to
9.2-18
-------
lessen water pollution impacts. Dry flyash handling and disposal require
similar precautions since water runoff can cause leaching of metals into the
i .
water table. Aside from outright disposal, flyash is also used in road
embankments and in concrete mixes.27 :
9.2.3 Coal-Fired Cyclone Furnaces
Coal-fired cyclone furnaces are not widely used. Only 3.3 percent of
all utility and large industrial boilers are cyclone units, and very few
have been installed in recent years.28 Their future use appears limited
to firing certain kinds of lignite.29, One reason for their declining use
is the intensity of combustion and the resultant high NOX content in the
gas stream.
9.2.3.1 Source Description. The cyclone furnace is a water-cooled,
horizontal cylinder into which partially dried crushed coal of approximately
0.63 cm diameter is fired in a tangential or vortex pattern into the firebox
(Figure 9.2-6).30 Bottom ash is continuously removed through a slag
tap in the furnace floor. .
9.2.3.2 Emission Characteristics and Applicable Control Technologies.
Particulate emissions from cyclone furnaces are less than one-eighth of that
from pulverized coal-fired units.31 Eighty to 85 percent of the ash is col-
lected as slag. The balance of the ash is discharged as flyash. Approxi-
mately 85 to 90 percent of the flyash produced is less than 10 micrometers
in diameter.^»30
A cyclone-fired utility boiler of average size produces 21,850 Am-Vmin
of flue gas; it has a stack height of 114 m, a stack diameter of 4.4 m, and a
stack gas temperature of 430 K.14 Although no specific effectiveness data
were identified, electrostatic precipitators are generally used to control
cyclone furnace particulate emissions'.2^
9.2.3.3 Secondary Environmental Impacts. Flyash collected from cyclone
boilers is difficult to dispose of due to its small particle size. Unsuited
for landfill, the flyash is generally reinjected into the cyclone furnace and
converted into the more easily disposed slag.30 The slag is usually disposed
of by landfilling.
9.2-19
-------
PRIMARY
AIR
Figure 9.2-6 Cyclone furnace side view.30
9.2-20
-------
9.2.4 NonfossJl Fuel-Fired Boilers
Nonfossil fuel-fired boilers burn municipal solid waste (MSW), wood,
wood waste and bark (hogged fuel), and refuse-derived fuel (RDF). MSW
consists of refuse and garbage from residences, commercial establishments,
and industries. Boilers firing MSW are found in district and municipal
heating plants and privately owned'waste-to-energy facilities. RDF is
municipal solid waste that has been processed to remove noncombustibles
such as glass and metals. RDF can :be burned alone but is often burned as
a supplement to coal. In such cases, 10 to 50 percent of boiler heat input
(Btu/hour) may come from RDF. Boilers currently burning-RDF are operated by
utilities or municipalities; however, industrial firms are studying the
feasibility of burning RDF with coal in their boilers.
Wood, wood waste and bark are used to fire boilers in the forest pro-
ducts industry, furniture industry,1 and paper and .allied products industry.
The forest products industry (e.g.,; lumber mills) burns log waste and limbs
after they are broken into chips in a hammer mill (hogged). The paper indus-
try burns bark that is removed from! logs prior to chipping the wood for pulp-
ing. Bark is often cofired with coal or oil in boilers to make process
steam. Sawdust, sander dust, and wood scraps are burned in boilers in the
furniture industry.
Wood is beginning to be harvested and chipped or pelletized for use as
a fuel in any boiler.32,33
9.2.4.1 Source Description
9.2.4.1.1 Municipal solid waste-fired boilers. Large MSW-fired boilers
have been operated in Europe since the late 1940s. Five or six are current-
ly operating in the USA, and several more are planned for completion before
1985. These MSW-fired boilers are typically waterwall furnaces with overfeed
stokers and traveling or vibrating grates for ash removal. The MSW is typi-
cally burned without being classified; only the largest noncombustibles (e.g.,
refrigerators) are removed. Following combustion, ferrous metals are removed
from the ash, and the residue is landfilled.
A typical large MSW furnace burns 38,200 kg/hr (85,000 Ib/hr) of solid
waste.34,35,36 This waste is received on site with a moisture content
(wet basis) of 35 percent. Ultimate analysis (Table 9.2-6) of the waste
shows a sulfur content of 0.06 percent and a nitrogen content of 0.6 percent
9.2-21
-------
Table 9.2-6. ULTIMATE ANALYSIS OF A TYPICAL GENERAL SOLID WASTE37
Moisture 35.00%
Carbon 20.00%
Oxygen (02) 18.00%
Hydrogen 2.50%
Nitrogen 0.60%
Sulfur 0.06%
Noncombustibles 23.84%
100.00%
9.2-22
-------
by weight.37 The heating value of this waste is typically 9287 kJ/kg (4000
Btu/lb) of waste as received.38 Ash content of the MSW is high (24 percent)
because of the large fraction of noncombustibles in the waste.
Uncontrolled emissions from a typical large MSW boiler are presented
in Table 9.2-7, A representative size distribution of uncontrolled PM emis-
sions is shown in Figure 9.2-7.39 .
Combustion of MSW in small shop-assembled (modular) boilers was intro-
duced in the late 1960s.40 These units are constructed in a factory on a
package basis and are typically hopper- or ram-fed. To provide easy expan-
sion of burning capacity for small towns and industries, the modular
boiler system is designed to allow installation of additional units in
modules two to eight as refuse generation increases.41
A typical small modular boiler consists of an incinerator with primary
and secondary combustion chambers. Units of this type are commonly referred
to as "controlled-air" or "starved-air" boilers because these terms denote
the control and regulation of the air flowing into the two combustion cham-
bers. 40 Uncontrolled emissions based; on actual boiler emissions measurements
are shown in Table 9.2-8. Controlled-air boilers commonly burn refuse at
816°C (1500°F) in the primary chamber and at 1038°C (1900°F) in the second-
ary chamber.42 A representative size distribution curve for uncontrolled
PM (flyash) from a controlled air boiler is shown in Figure 9.2-8.4°
9.2.4.1.2 Refuse-derived fuel-fired boilers. RDF has a relatively uni-
form quality. The major RDF fuel typies are as follows:
o Fluff from a wet pulping process.
o Fluff from dry processing—size reduction, air classification.
o Screened fluff from dry processing—size reduction,
air classification, and screening.
o d-RDF (densified RDF)—pelletjization of fluff or screened fluff.
o Powdered RDF—proprietary commercial process, fuel characterized
as a fine dustlike material.44
The combustion properties of RDF; depend upon the degree of processing for
the materials described above; some properties of interest are given in Table
9.2-9.45 Choice of the most appropriate type of RDF depends on the type of
boiler used. For example, powdered RDF would be burned in a boiler designed
for suspension firing. Briquetted RDF could be burned in a stoker fed boiler.
9.2-23
-------
Table 9.2-7. UNCONTROLLED EMISSIONS FROM A TYPICAL LARGE 525 GJ/hr
(500 x 106 Btu/hr) HEAT INPUT MSW-FIRED BOILER36
Emission
PM
NOX
so2
co2
Mass,
kg/hr (Ib/hr)
576
340
57
41,500
(1270)
(750)
(125)
(91,500)
Concentration,3 Heat input,
g/Nm3 (gr/scf) ng/J (lb/106 Btu)
2.36
1.39
0.23
170.15
(1.03)
(0.61)
(0.10)
(74.30)
1092.91
645.41
107.57
78,740.82
(2.54)
(1.50)
(0.25)
(183.00)
Concentration values adjusted to 12 percent C02.
9.2-24
-------
10(39.0)
o
c
LO
I
o
CO
S-
-------
Table 9.2-8 UNCONTROLLED PARTICIPATE EMISSIONS FROM MSW-FIRED BOILERS
Plant
U38
£ AA43
CT)
BE*"
.
Boiler type
Traveling grate
stoker
Traveling grate
stoker
Cont rolled-air
combustion
Design
heat input,
GJ/hr
(106 Btu/hr)
127 (121)
1210 (1149)
11 (10)
Emissions
On the basis of On the basis of
concentration at 12% C02, heat input,
Fuel Type g/Nm3 (gr/scf)
MSW PM 3.34 (1.46)
MSW PM 3.37 (1.47)
MSW PM 0.18 (0.08)
ng/J (lb/106)
1879 (4.37)
4300 (10.00)
85 (0.20)
66 Controlled-air 19(18,1) MSW PM 0.35(0.16) 142(0.33)
-------
99
90
50
O
c;
LU
D_
10
4
2
1.0
0.1
(0.4)
l.Oi
(3.9)
10
(39)
100
(390)
PARTICLE SIZEi, micrometers (xlQ-5 inches)
Figure 9.2-S Size distribution of MSW flyash from a controlled air laoiler.
9.2-27
-------
Table 9.2-9 TYPICAL CHARACTERISTICS OF REFUSE-DERIVED FUELS'^
<£>
PO
PO
CO
As-received
refuse-
derived fuel3
Fluff-wp
Fluff-dp
Screened fluff
d-ROF pellets
(from fluff)
d-RDF briquettes
(from fluff)
Powdered RDF
Density,
kg/m3 (Ifa/ft3)
224
80-144
16-80
380-720
750-900
480
(14)b
(5-9)
(1-5)
(24-45)
(47-56)b
(30)
Heating value,
kJ/kg (Btu/lb)
8,140 (3500)
12,090 (5200)
16,750 (7200)
12,090 (5200)
12,090 (5200)
18,011 (7750)
H20
f,
50
25
16
15
14
2
Ash
20
19
10
18
18
10
Approximate particle size
2.5 cm (1.0 in)
2.5-5.7 cm (1.0-2,5 in)
2.5-5.2 cm (1.0-2.0 in)
1.3 cm dia. x 2.5 cm long
(0.5 in dia. x 1.0 in long)
3.0 x 3.0 x 7.5 cm
(1.25 x 1.25 x 3.0 in)
0.15 mm (0.006 in)
awp = wet processed; dp = dry processed; d-RDF = densified refuse-derived fuel.
^Estimated. '
-------
Figure 9.2-9 shows the size distribution of uncontrolled PM (flyash) from a .
boiler burning 50 percent coal and 50 percent RDF (heat input).^6
9.2.4.1.3 Wood, wood waste, and bark-fired boilers. Wood and bark
fuels usually have approximately 50 percent moisture, a heating value of
10,227 KJ/kg (4400 Btu/lb), and an ultimate analysis as shown in Table
9.2-10.^7 The low available energy input of the wood results from the high
moisture content of the fuel. Heat energy is required to evaporate the
water contained in the wood before! combustion can take place. Once evapo-
rated, the water vapor is discharged up the stack, resulting in a larger
exhaust mass than would be expected from the combined air input rates.
A typical size distribution curve for uncontrolled wood fly ash
emissions is shown in Figure 9.2-10.48 Such emissions have a primary resis-
tivity of 1.7 x 105 Qcm (6.7 x lO^Qin) at 204°C (400°F).49
Wood and bark have been burned in dutch ovens, spreader stokers,
suspension-firing boilers and fluidized bed combustors.
9.2.4.1.4 Dutch ovens. Figure 9.2-11 is a cross-section of a dutch
oven, which is primarily a large rectangular box lined on the sides and top
with fire-brick.50 The fuel rests'on a grate through which underfire air
is fed. Overfire air is introduced around the sides of the fuel pile. By
!
design, combustion in a dutch ovenior primary furnace is incomplete. Com-
bustion products pass between the bridge wall and the drop-nose arch into a
secondary furnace chamber, where combustion is completed before gases enter
the heat exchange section. Construction of new dutch ovens was phased out
in the 1950s because of their high construction costs, low efficiency, and
inability to follow load load swings.51
9.2.4.1.5 Spreader stoker. Since the mid 1940s, nearly all of the
wood-fired boilers constructed in the United States have been spreader
stokers. The fuel is burned in the base of the water wall boiler unit
rather than in a refractory chamber (Figure 9.2-12). Spreader stokers are
preferred because of their ease of ^operation and relatively high thermal
efficiency: 65-80 percent of the energy available in the fuel. Table 9.2-11
summarizes average stack parameters for typical wood/bark-fired boilers.
9.2.4.1.6 Direct firing applications. Over the past five years in
the United States, nonfossil fuel-fired boilers have been installed in
9.2-29
-------
100 (390)
in
O)
u
IT)
o
X
in
O)
+>
cu
o
o
o
LO
o
LU
M
*—<
CO
o;
-------
Table 9.2-10. ULTIMATE ANALYSIS'OF MOISTURE-FREE WOOD-WASTE FUEL^7
Material ; Percentage
Hydrogen •• : 5.80
Carbon " . , . 52.20
Sulfur 0.05
Oxygen , 40.20
Nitrogen ] 0.05
Ash 1.70
Total 100.00
9.2-31
-------
1000(3900.0)
to
o>
LO
i
o
I/J
100(390.0)
£ 10(390.0)
o
o
1(3.9)
0.1(0.4)
Wood fly ash
without
reinjection
Wood fly ash
with
reinjection
0.01
10 40 90
SMALLER THAN, %
99.9
Figure 9.2-10 Size distribution of wood flyash.1+&
9.2-32
-------
FUEL IN
TO CINDER
COLLECTORS,
AIR HEATER,
a STACK
AUX FUEL
BURNER
UF USED)
UNDERFtRE
' AIR IN
Figure 9.2-11. Dutch oven furnace and boiler.50
9.2-33
-------
•STEAM OUT
FUEL CHUTE
STACK
STEAM AIR MULTIPLE
DRUM HEATER CYCLONE
COLLECTOR
OVERFIRE
AIR
SPREADER
GRATES ASH PIT /MUD
DRUM
ID FAN
figure 9.2-12. Small spreader-stoker furnace.50
9.2-34
-------
Table 9.2-11. AVERAGE STACK PARAMETERS FOR TYPICAL WOOD/BARK-FIRED
BOILERS6 ..
Stack Stack Stack
Fuel height, diameter, temperature, Flue gas volumetric
fired Size m mi K flowrate,
Bark All 42 2.4 470 940
Wood/bark All 30 1.6 ; 495 1590
Wood All 24 1.1 445 615
9J2-35
-------
which the hot gases from burning bark and wood are used directly for heat.
Applications involving direct firing of wood and bark include veneer dryers,
drying kilns for lumber, and dryers for wood and bark particles. (Section
9.10 discusses the forest products industry).
9.2.4.2 Applicable Control Technologies. Particulate emissions from
nonfossil fuel-fired boilers have been controlled with mechanical collectors
(cyclones and multiple cyclones), water scrubbers, electrostatic precipita-
tors (ESPs), gravel bed filters (GBF), and fabric filters (baghouses). Table
9.2-12 summarizes key considerations in choosing particulate emission control
devices for wood-fired boilers. The systems operating in series provide the
best performance but their costs are roughly double that of single collector
systems.
9.2.4.2.1 Mechanical collectors. Mechanical collectors have been used
extensively for particulate control on wood- and bark-fired boilers and MSW-
fired waterwal1 incinerators. They are often used as primary collectors
ahead of other devices which collect fine particulates more efficiently. The
typical relationship between particle size and cyclone collection efficiency
is illustrated in Figure 9.2-13.52 The collection efficiency of particles
greater than 5 microns in diameter is good for multiple cyclones. Below this
approximate diameter, the collection efficiency decreases rapidly.
Cyclones are designed to operate within a set range of inlet gas veloci-
ties. If the inlet-velocity decreases below the intended range, the centrif-
ugal force on the particles will not be great enough to separate them from
the gas stream and the collection efficiency of the cyclone will decrease.
This change in collection efficiency is illustrated in Figure 9.2-14 for a
given cyclone and dust.53 The velocity of the boiler exhaust gas can vary
with changes in boiler load (firing rate) and excess combustion air.
9.2.4.2.2 Hater scrubbers. Scrubbers have been used primarily on wood-
and bark-fired boilers. They have been avoided on MSW-fired boilers because
of odor and corrosion problems that result from wetting the ash from MSW.
9.2.4.2.3 Fabric filters. Baghouses have been used on two MSW incin-
erators and seven wood-fired boilers. The greatest detriment to applying
baghouses to nonfossil-fueled boilers has been the potential for fires due
to the high carbon content of the fly ash. A collection efficiency of 99.8
percent was achieved for one of the above-mentioned units operating under
9.2-36
-------
Table 9.2-12. KEY CONSIDERATIONS OF PARTICIPATE EMISSION CONTROL DEVICES FOR WOOD-FIRED BOILERS50
ro
oo
Cost,
$ per
Collector type Ani3/min
Single cyclone
Multiple cyclone
Granular bed filter _
Baghouse
Multiple cyclone
plus ESP
Multiple cyclone
plus dry scrubber
Multiple cyclone
plus baghouse
19
57
57
77
153
115
134
Power
required,
W per
An)3/min
17
25
35
42
37
62
67
Expected
performance
Pressure drop,
kPa
0.25 to 0.50
0.38 to 0.75
1.25
0.75
0.50
1.75
1.25
Temp.
limit,
k
819
810
810
530
810
810
530
Effic.,
%
80
90
.95.
99
99.5
99
99.5
Outlet
loading,
g/Nm3
0.9
0.5
.0...2
0.001
0.02
0.11
0.002
Disposal of collected
particulate
Dry: landfill or
charcoal
Dry: landfill or
charcoal
Dry: .landfill or- _. .
charcoal
Dry: landfill or
charcoal
Dry: landfill or
charcoal
Dry: landfill or
charcoal
Dry: landfill or
charcoal
a!978 dollars.
-------
CL)
O
CL)
O.
O
i— I
U_
u_
LU
O
I— I
O
O
O
100
85
60
40
20
SINGLE
LARGE
CYCLONE
MULTIPLE
SMALL
CYCLONES
I
20 40
PARTICLE SIZE, micrometers
Figure 9.2-13
Typical relationships between cyclone collection efficiency
and particle diameter for large and small tubes (same inlet
velocity).52
9.2-38
-------
o.
>-
100
90
80
70
60
50
40
30
20
10
I
0.5 1.0 1.5 2.0
VELOCITY RELATIVE TO DESIGN CONDITION
2.5
Figure 9.2-14 Variations of single cyclone collection efficiency with
gas flow velocity.53
9.i2-39
-------
the following conditions:5^
Gas flow: 85 n?/s at 260°F
A/C ratio: 0.010 m/s
Fabric: Coated fiberglass
Average pressure drop: 0.50 to 0.75 kPa
Bags cleaned by mechanical shaking with reverse air. ...
Fabric filter applications to cofired or MSW-fired boilers seem particu-
larly appropriate, since the solid waste input varies considerably. Fabric
filter systems are much less sensitive to variations in fuel quality than ESPs
and can perform efficiently with poorer grades of fuel.
9.2.4.2.4 Electrostatic precipitators. ESPs have been used on MSW- and
RDF-fired boilers. The resistivity of RDF fly ash is very high, ranging up
to 2 x 1012 ncm--well above the 2 x 1010 flcm upper limit of the range of
optimum resistivities. In addition, RDF combustion requires greater excess
air than does coal combustion, thus resulting in larger exhaust-gas volumes.
Consequently, the collection efficiency of an ESP with RDF fly ash is lower
than that of an ESP with coal fly ash. The most promising methods for im-
proving the collection efficiency on RDF are pretreating the exhaust gas to
reduce particle resistivity and designing the ESP to accept the larger ex-
haust volumes.55
Recently, ESPs have been installed on boilers that are cofired with
wood or bark and coal and on boilers fired with wood and bark only. Flyash
from wood-fired boilers has a resistivity that can be less than the 10? ftcm
lower limit for optimum particle charging and collection in an ESP. Also,
wood flyash loses its charge rapidly because it is very carbonaceous and,
therefore, is reentrained into the gas stream readily. An ESP designed for
a wood-fired boiler would need a larger specific collection area than an ESP
for a similar coal-fired boiler in order to overcome these design problems.
Typical operating parameters are as follows:5^
Gas flow: 194 m3/s
Plate spacing: 22 cm
SCA: 27 m2 per m3/s
Plate area: 5,180 m2
Migration velocity: 11 to 15 cm/s.
9.2-40
-------
Figure 9.2-15 presents resistivity data for participates from three U.S.
municipal incinerators.56 The wide variation in resistivities is due to dif-
ferences in the constituents of the waste—both noncombustible matter that is
emitted directly as particulates and substances which, when emitted as gase-
ous compounds, alter the surface conduction properties of the particles by
adsorption onto them. |
9.2.4.2.5 Gravel bed filters. jGBFs are used on about ten wood-fired
boilers. The system consists of a cylindrical vessel containing two concen-
tric, louvered tubes. The annular space between the tubes is filled with
gravel media. Particles in a dirty gas stream are removed by impaction on
the gravel as the gas passes throughithe filter. The particulate-laden fil-
tering media is continuously removedjfor cleaning from the bottom of the
filter. The clean gravel is then returned to the top of the filter and re-
cycled. In some applications, a low-energy cyclone is incorporated as an
integral part of the outside wall of;the filter in order to remove very large
particles; this technique is particularly useful in the case of wood-fired
boiler applications where coarse, carbonaceous particulate can be collected
for reinjection into the boiler. A schematic of a gravel bed filter with an
integral cyclone is presented in Figure 9.2-16.57
Variations in fuel properties or boiler load do not affect gravel bed
filter performance. The simple design, the self-cleaning moving bed, and the
inertness of the filter make it insensitive to fluctuations in temperature,
gas flow rate, and chemical composition of the particulates. Changes in gas
loading can be compensated for by regulating the recirculation rate of the
filtering media. Corresponding changes in fan power will also be required.58
Efficiencies and design parameters for a typical GBF operating in series with
a mechanical collector on a hogged fuel boiler are shown in Table 9.2-13.
9.2.4.3 Secondary Environmental! Impacts. All of the control technolo-
gies outlined above will result in the need to dispose of many tons of partic-
ulate matter per year. A comparison ;of the average properties of the flyash
from coal and MSW showed the major difference to be increased concentration
of the following trace elements from MSW: antimony, arsenic, barium, cadmium,
chromium, copper, lead, mercury, zinc, bromine, and chlorine. The changes
in the major components of the flyash are not so large that the disposal of
flyash collected from the burning of MSW would pose any more of a problem
than the disposal of the flyash collected from the burning of coal. (See
i
9*2-41
-------
10
13
10
12
j=
o
co
UJ
o
LU
Jl
JO
10
8
250
(-10)
I
350
(170)
450
(350)
550
(530)
TEMPERATURE, k (°F)
I
OPTIMUM
COLLECTION
RANGE
650
(710)
A -
,B -
C — TYPICAL URBAN DOMESTIC/COMMERICAL RESIDUE
- SUBURBAN LOCATION, HIGH GRASS/TREE CLIPPINGS
CONTENT
- TYPICAL URBAN DOMESTIC/COMMERICAL RESIDUE, BUT
WITH HIGH PAPER CONTENT
Figure 9.2-15 Electrical resistivity of flyash from three different U.S.
municipal incinerators.56
9.2-42
-------
MEDIA
CLEANUP UNIT
Figure 9.2-16 Schematic of integral cyclone gravel bed filter.57
9.2K43
-------
Table 9.2-13. EFFICIENCY OF GRANULAR RED FILTERS IN SERIES WITH CYCLONES ON HOGGED FUEL BOILERS50
to
IN3
£
0.
o.
15
15
Media
size,
cm
6 x 0.3
6 x 0.3
x 20
x 20
Media
gas
velocity,
cm/sec
64.5
86.4
76.2
63.5
Media
pressure
drop,
kPa
1.5
2.3
3.0
2.4
Cyclone
pressure
drop,
kPa
0.3
0,5
0.4
0.3
Total Loading,
g/Nm3 at 12% C02
Cyclone in
2.768
1,486
2.542
4.719
Media in
0.875
0.609
0.800
0.618
Media out
0.075
0.080
0.070
0.026
Collection efficiency
Cyclone
68.4
59
68.5
86.9
Media
91.4
86.9
91.3
95.7
Total
97.3
94.6
97.3
99.4
-------
Section 9.2.1.3 for the secondary environmental impact of control techniques.
for pulverized coal firing.) The changes in trace element concentration
might result in leaching problems if the collected flyash is placed in a
landfill, but no assessment of the relative impacts has been made.49
Potential water impacts arise when sluicing water is used to carry the
collected flyash to settling ponds. l In comparing coal-only to coal-MSW
firing data for the ash pond water, bnly total dissolved solids increased
with the burning of MSW. Analysis for trace constitutents showed little
change for coal-MSW sluice water compared to coal-only water.59
9.2.5 Oil-Fired Utility Boilers ;
During 1977 about 21 percent of: total fossil-fuel electric power was
generated with oil-fired steam generators.1 Residual fuel oil is the
predominant fuel used in oil-fired utility steam generators.1 Some dis-
tillate oil is blended with the residual oil to meet specifications.
Distillate oil also is used for startup of coal-fired steam generators.!
Nationwide emissions from oil-fired electric utility boilers in 1978 were
estimated at 140 gigagrams.2 !
9.2.5.1 Source Description. All oil-fired utility boilers are qf
watertube design,.i.e. water flows through the heat transfer tubes. The
typical furnace is of waterwall design, backed by refactory or insulation.
The tubes, which form the furnace waterwall, are an integral part of this
boiler. Oil-fired utility boilers are similar to pulverized coal units,
except for burners and equipment to handle fly ash and bottom ash. Since
the inorganic ash content is much lower with oil than coal, soot blowing
needs are less severe and flyash collection and disposal systems handle
considerably lower volumes. ;
The primary firing methods for oil-fired utility boilers are as
follows:
o Single-wall firing, in which; a bank of burners mounted on a plane
wall ejects fuel in one direction only.
o Opposed firing, in which two. banks of burners are directed toward
one another.
o Tangential firing, in which the burners are located at the
corners of a square and eject oil in such a direction as to give
a rotational motion to the combustible mixture.
9.2-45
-------
Common burner designs employ steam, air, or mechanical atomization. Air
atomization apparently leads to more complete combustion than steam, and
steam provides more complete combustion than mechanical atomization.60
9.2.5.2 Emission Characteristics and Applicable Control Technologies.
Particulate emissions from utility boilers are generally a function of boiler
size. Measurements indicate that the average particulate emission levels
from a sample of uncontrolled boilers decreased from 36.6 ng/J to 19.4 ng/J
as the boiler size went from 1 to 500 MW. Emissions from the newer, larger
boilers are lower than those from older units because of improved combustion
controls. Selected stack parameters for oil-fired utility boilers are pre-
sented in Table 9.2-14.
On a mass basis, particulate emissions from an uncontrolled residual
oil-fired boiler are on the same order of magnitude as those from a highly
controlled coal-fired boiler. Stack tests indicate that between 85 and 90
weight percent of the particles liberated by uncontrolled residual oil com-
bustion are less than 1 micrometer in diameter; usually less than 10 percent
of the particles liberated by coal combustion are smaller than 1 micro-
meter. 21 »61
Finally, the size and mass of particles emitted varies with boiler
maintenance. If a boiler is well-maintained and properly adjusted, there
will be little carbonaceous matter released; emissions will consist almost
entirely of inorganic ash particles, a substantial portion of which will be
less than 1 micrometer in diameter. However, it is common to find poorly
maintained oil-fired boilers whose emissions are dominated by larger par-
ticles.
Four types of particulate matter collectors are available for control-
ling emissions: electrostatic precipitators, mechanical collectors, fabric
filters, and wet scrubbers. Of these, only the ESP and the multiple cyclone
(one type of mechanical collector) are used to any degree at oil-fired facili-
ties.^ Fabric filters and wet scrubbers are seldom used.
9.2.5.2.1 Electrostatic precipitators. ESPs are the most common col-
lectors now in use on oil-fired boilers. Normally, they are designed to
remove about 90 percent of particulates by weight.6^,64,65
Precipitators that were originally designed to collect coal flyash
experience a marked drop in efficiency when the boiler is fired with oil.62.66
9.2-46
-------
Table 9.2-14. SELECTED STACK PARAMETERS OF OIL-FIRED UTILITY
AND INDUSTRIAL BOILER$25
Stack Stack Stack Flue gas volumetric
Fired Size, height, diameter, temperature, flowrate,
fuel MW m m K AnvVmin
Residual
Greater
than
29.3
48
2.9
470
6070
9.2-47
-------
Changes in mass loading particle resistivity, size distribution, and surface
properties are the primary reasons for this reduction. Units which were
installed on coal-fired boilers require modification if the boiler is con-
verted to oil.
More extensive data on electrostatic precipitators applied to oil-fired
steam generators are given in Reference 67.
9.2.5.2.2 Mechanical collectors. Typical efficiencies of mechanical
collectors are only 75 to 85 percent for particles 10 micrometers in diameter
or larger.64 Their efficiency is much lower for submicrometer particles gen-
erated by oil furnaces. Mechanical collectors are ineffective on oil burning
units even if all burners are properly maintained and adjusted. Small parti-
cles account for more than one-half of participate emissions by weight from
well maintained sources.63,68,69,70 if ac-j(j smut (which is composed mostly
of large particles) is a major pollution problem, mechanical collectors are
sometimes used. With older units, mechanical collectors are sometimes em-
ployed, primarily to collect large particulates released during soot blowing.62
9.2.5.2.3 Fabric filters. Only one full-scale fabric filter system
is known to have been installed on an oil-burning utility boiler.62 Many
problems were encountered when the baghouse was in service, the most promi-
nent being the deterioration of the bags by the acidic oil and ash, and
plugging of the bags due to the hygroscopic nature of the oil ash.64 Explo-
sion hazards are also possible.
9.2.5.2.4 Wet scrubbers. In the U.S., flue gas desulfurizer (FGD)
wet scrubbers are used by utility operators mainly for sulfur oxide removal
at coal-fired boilers. In Japan, FGD units have been used to control $03
and particulates from oil-fired boilers,.
Some wet scrubbers under development have shown removal efficiencies
of greater than 99 percent by weight during pilot and prototype programs.70
Wet scrubbers are capable of providing highly efficient removal of particulate
matter from flue gases, but there are the following potential problems:62,70
o High draft loss, up to 15 kPa for higher efficiencies resulting
in higher power requirements.
o Moist and cooled exit gas, potentially leading to corrosion of
downstream equipment, visible vapor plumes and reduced buoyancy
at the stack.
9.2-48
-------
o Generation of scrubber slurries (the bulk of which, however, are
sulfur wastes, not ash from fuel oil).
Therefore, even though scrubbers canjbe used efficiently to remove particu-
late emissions from oil-fired boilers, they are not widely employed.
9.2.5.3 Secondary Environmental Impacts. The amount of flyash col-
lected from oil-fired utility boilers will be less than that for coal-fired
utility boilers. Section 9.2.1.3 discusses in greater detail the secondary
environmental impacts of the control techniques for coal-fired boilers.
9.2.6 Oil-Fired Industrial/Commercial Boilers
Emissions from oil-fired industrial/commercial boilers vary over a wide
range, depending on the size of the boiler and the type of fuel burned.^
Except for combustion modifications and proper maintenance, controls for
commercial sized oil-fired boilers are seldom used.
9.2.6.1 Source Description. Commercial boilers may be categorized as
cast iron, firetube, or watertube types; industrial boilers are exclusively
firetube or watertube types. Cast iron sectional units are designed to
supply low pressure steam or hot water and are used primarily for hydronic
heating.62 These boilers consist of an assembly of cast iron sections.
Generally, cool water enters at the bottom of the sections, and hot water or
steam exits at the top. Figure 9.2-17 illustrates an oil-fired cast iron
boiler. In a firetube boiler, the products of combustion are directed
through tubes which are submerged in water. Firetube boilers are used when
steam demand is relatively small. i
9.2.6.2 Emission Characteristics and Applicable Control Technologies.
Emissions vary with the size of the b'oiler and the oil atomization tech-
nique. As with any oil-fired boiler, particulate emissions are hygro-
scopic and relatively light (one-quarter the weight of pulverized coal
ash); therefore, they tend to plug particulate emissions control devices
and pack if stored in hoppers. Table 9.2-15 presents selected stack para-
meters for oil-fired industrial/commercial boilers.
An important parameter in controlling particulate emissions is the amount
of swirling motion impacted to the gases in the combustion region.71.72,73
There is an optimum value of swirl that yields low particulate emissions.
Other aerodynamic factors influencing particulate emission include recircu-
lation in the gas flow and relative air/fuel velocity.
9.2-49
-------
Figure 9.2-17 Oil-fired cast iron boiler. (Courtesy of the Utica Radiator
Company.)
9.2-50
-------
Table 9.2-15. SELECTED STACK PARAMETERS FOR OIL-FIRED INDUSTRIAL/COMMERCIAL
BOILERS^
Flue gas
Stack Stack Stack volumetric
Fuel Size, height, diameter, temperature, flowrate,
fired MW m m K Am^/min
Residual 2.9-29.3 27 1.5 500 730
oil less than 21 0.9 470 380
2.9
Distillate 2.9-29.3 20 1.2 490 705
oil less than 13 0.7 475 175
2.9
9.2-51
-------
Although dust collectors can be used to control commercial units, they
are not used generally. One study has shown that dust collectors can be
effective in reducing emissions from some commercial boilers, but the
investigation did not report the condition of the burners, or the attention
paid by the operator to the control device.74
Additives containing certain alkaline-earth and transition metals in
residual oils reportedly are effective in reducing carbon particulate by
as much as 90 percent.^5 Similar reduction can be achieved with good
burner design and adjustment, but for certain commercial and industrial
boilers operating in the field that cannot receive the maintanence service
needed to reduce emissions, additives may be an effective particulate
control technology.
Particulate emission control technologies for industrial oil-fired
boilers are virtually identical to those for oil-fired utility boilers.
These technologies include electrostatic precipitators, fabric filters,
scrubbers, and mechanical collectors. The efficiencies of these collectors
would be roughly the same for industrial and utility boilers.
9.2.6.3 Secondary Environmental Impacts. The secondary environmental
impacts of particulate emissions control technology applied to oil-fired
industrial/commercial boilers are very similar to those of oil-fired utility
boilers. Section 9.2.5.3 contains further discussion of these impacts.
Environmental health considerations make the use of fuel additives in
commercial and industrial boilers questionable because the additives create
more potentially harmful new emissions.
9.2-52
-------
REFERENCES:FOR SECTION 9.2
1. Steam Electric Plant Factors 1978. National Coal Association.
Washington, DC. 1978. !
I
2. U.S. Environmental Protection Agency, Office of Air Quality Planning and
Standards, National Air Data Branch. Research Triangle Park, NC. 1980.
i
3. Stationary Watertube Boiler Sales, 1976. American Boiler Manufacturers
Association. Arlington, VA. 1977.
4. Steam - Its Generation and Use, 38th Edition. Babcock and Wilcox Co.
New York, NY. 1975. i
5. Background Information for Proposed Particulate Matter Emission Stan-
dards, NSPS. U.S. Environmental Protection Agency. Publication No.
EPA-450/2-78-006a. July 1978. |
6. National Emission Data System (Source Classification Codes). U.S. Envi-
ronmental Protection Agency. Research Triangle Park, NC.
7. Niessen, W. R. Combustion and Incineration Processes - Applications in
Environmental Engineering. Marcel Dekker, Inc. New York, NY. 1978.
8. Review and Assessment of the Existing Data Base Regarding Flue Gas
Cleaning Wastes. Electric Power Research Institute. Palo Alto, CA.
EPRI FP-671, Project 786-2, Volumes 1 and 2. January 1979.
9. Coltharp, W. M., et al. Reviewiand Assessment of the Existing Data Base
Regarding Flue Gas Cleaning Waste. EPRI FP-671, Volume I. Radian Cor-
poration. Austin, TX. January ^1979.
i
10. Reigel, S. A., and R. P. Bundy. i Why the Swing to Baghouses. Power.
January 1977. >
11. The ESP Manual, Volumes 1 and 2. The Mcllvanie Co. Northbrook, IL.
12. Particulate and Sulfur Dioxide Emission Control Costs for Large, Coal-
Fired Boilers. PEDCo Environmental, Inc. U.S. Environmental Protection
Agency. Publication No. EPA-450/3-78-007. February 1978.
13. The Fabric Filter Manual. The Mcllvanie Co. Northbrook, IL.
14. The Scrubber Manual. The McHvanie Co. Northbrook, IL.
15. Fine Particulate Emissions Information System: Annual Report (1978).
U.S. Environmental Protection Agency. Publication No. EPA-600/7-79-126.
May 1979. ;
16. Guidelines for Determining Best Available Retrofit Technology for Coal-
Fired Power Plants and Other Existing Stationary Facilities. National
Technical Information Source. Springfield, VA.
9:.2-53
-------
17. Hart, F. C.} and B. T. Delaney. The Impact of RCRA (PL 94-58) on
Utility Solid Wastes. Fred C. Hart Associates, Inc. New York, NY.
Publication No. 78-779. August 1978.
18. Dvorak, A. J., et al. Impacts of Coal-Fired Power Plants on Fish,
Wildlife, and Their Habitats. U.S. Department of the Interior.
Publication No. FWS/OBS-78/29. March 1978.
19. Devitt, T., et al. The Population and Characteristics of Industrial/
Cominerical Boilers. PEDCo Environmental, Inc. May 1979.
20. Lim, K. J., et al. Environmental Assessment of Industrial Boiler Com-
bustion Modification NOX Controls. Acurex Corporation. Mt. View, CA.
Publication No. TR-79-10/EE. July 1979.
21. State of the Art: Oil-Fired Precipitator Applications. Research
Cottrell, Inc., Air Pollution Control Division. Bound Brook, NJ.
22. Roeck, D. R., and R. Dennis. Technology Assessment Report: Assessment
of Particulate Collection Technology for an Industrial Boiler New Source
Performance Standard. 6CA Corporation, Technology Division. Bedford,
MA. Draft Final Report. June 1979. p. 51.
23. Gard, Inc. Capital and Operating Costs of Selected Air Pollution Control
Systems. U.S. Environmental Protection Agency. Publication No.
EPA-450/3-76-014. May 1976.
24. Draft of Fossil Fuel Fired Industrial Boilers - Background Information
for Proposed Standards, Chapters 3 and 5, and Appendix C. U.S. Environ-
mental Protection Agency, Emission Standards and Engineering Division.
Research Triangle Park, NC. July 1980.
25. Horzella, T. I. Selecting, Installing, and Maintaining Cyclone
Dust Collectors. Chemical Engineering Journal. January 1978.
26. Electric Power Research Institute test data reported in the "Precip
Newsletter." The Mcllvanie Co. Northbrook, IL. Number 21.
October 20, 1977.
27. Surprenant, N. F., et al. Preliminary Emissions Assessment of Conven-
tional Stationary Combustion Systems, Volume II, Final Report. U.S.
Environmental Protection Agency. Publication No. EPA-600/2-76-046b.
March 1976.
28. Lim, K. J., et al. Environmental Assessment of Utility Boiler Combus-
tion Modification NOX Controls. Acurex Corporation. Mt. View, CA.
Publication No. TR-78-105. April 1978.
29. Standards Support and Environmental Impact Statement, Volume I: Proposed
Standards of Performance for Lignite-Fired Steam Generators. U.S.
Environmental Protection Agency. Publication No. EPA-450/2-75-030a.
October 1976.
9.2-54
-------
30. Ctvrtnicek, T. E., and S. J. Rusek. Applicability of NOX Combustion
Modifications to Cyclone Boilers -(Furnaces). U.S. Environmental Pro-
tection Agency. Publication No. EPA-600/7-77-006. January 1977.
31. Compilation of Air Pollution Factprs, Second Edition. U.S. Environ-
mental Protection Agency. Publication No. "EPA-AP-42. August 1977.
32. Feasibility Study for a National Wood Energy Data Base, Final Report.
Ultrasystems Incorporated. McLeain, VA. (U.S. Department of Energy,
Contract No. EI-78-2-01-1951.) September 1978. p. 5.
33. Wood Waste for Fuel. Engineering1 Times. National Society of Profes-
sional Engineers. Washington, DC;. September 1979.
I
34. Resource Recovery Activities. In:: NCRR Bulletin, National Center
for Resource Recovery, Incorporated. Washington, DC. March 1979.
I
35. Plehn, S. W. RCRA Leads Attack on Solid Waste Problems. Consulting
Engineer. March 1979. p. 86.
36. Memo from Blackwell, C. D., Acurex Corporation, to Acurex Nonfossil-
Fueled Boiler file. October 31, f|979. Combustion calculations for
general solid waste boilers. 1
37. DeMarco, J., D. J. Keller, J. Lechman, and J. L. Newton, Municipal-
Scale Incinerator Design and Operation. U.S. Department of Health,
Education, and Welfare. Washington, DC. Publication No. 2012.
1969. p. 6. :
38. Bozeka, C. G. Nashville Incinerator Performance Tests. Babcock and
Wilcox Company. North Canton, OH;. 1976. p. 221-227.
39. Reference 38 above, p. 227. ;
40. Frounfelker, R. A Technical, Environmental, and Economic Evaluation
of Small Modular Incinerator Systems with Heat Recovery. Systems
Technology Corporation (SYSTECH).: Xenia, OH. (U.S. Environmental
Protection Agency, Contract No. 68-01-3889.) 1979. p. 13.
41. Evaluation of Small Modular Incinerators in Municipal Plants. Ross
Hofman, Associates. Coral Gables!, FL. (U.S. Environmental Protection
Agency, Contract No. 68-01-3171).. 1976. p. 7.
42. Compliance Test for Particulate Emissions, Garbage Incinerator System
with Energy Recovery. Hayes, Seay, Mattern, and Mattern. Roanoke, VA.
(City of Salem, VA.) March 1979. p. 13.
43. Hall, E.H., et al. Refuse Combustion in Fossil Fuel Steam Generators.
Battell-Columbus Laboratories. Columbus, OH. (U.S. Environmental
Protection Agency, Contract No. 68-02-0611.) September 1974.
44. Niemeyer, W., and T. C. Derbidge.; Investigation of Waste-Fired Indus-
trial Steam Generators. Acurex Corporation. Mt. View, CA. (U.S.
Environmental Protection Agency, Contract No. 68-02-26711.) September
1978. pp. 2-16, 2-17, 2-18. '
9.2-55
-------
45. Reference 44, pp. 2-18, 2-21.
46. Hall, J. L., et al. Evaluation of the Ames Solid Waste Recovery System:
Refuse Processing Plant and Stoker-Fired Steam Generators. U.S.
Environmental Protection Agency. Cincinnati, OH. EPA Grant No.
R803903-01-0. February 1978. p. 211.
47. Hall, E. H., et al. Comparison of Fossil and Wood Fuels. Battelle-
Columbus Laboratories. Columbus, OH. U.S. Environmental Protection
Agency, Contract No. 68-02-1323. Publication No. EPA-600/2-76-056.
March 1976. p. 39.
48. Reference 47, p. 63.
49. The Electrostatic Precipitator Manual, Volume III. The Mcllvanie
Company. Northbrook, IL. October 1977. p. 92.
50. Boubel, R. W. Control of Particulate Emissions from Wood-Fired Boilers.
U.S. Environmental Protection Agency. Publication No. EPA-340/1-77-026.
1977.
51. Junge, D. C. Boilers Fired with Wood and Bark Residues. Plywood Re-
search Foundation. Tacoma, WA. Research Bulletin 17. November 1975.
p. 2.
52. Boubel, R. W. Control of Particulate Emissions from Wood-Fired Boilers.
PEDCo Environmental, Inc. EPA Publication No. EPA-340/1-77-026, NTIS
PB-278 483. 1977. p. 6-6.
53. Stern, A. Air Pollution, 3rd Edition, Volume IV. Engineering Control
of Air Pollution. Academic Press. New York, NY. p. 117.
54. Devitt, T. W., et al. Air Pollution Emissions and Control Technology
for Waste-As-Fuel Processes. PEDCo Environmental, Inc. February 1979.
55. Reference 54, p. 92.
56. Walker, A., and F. Schmitz. Characteristics of Furnace Emissions from
Large, Mechanically Stoked Municipal Incinerators. In: Proceedings of
the 1966 National Incinerator Conference. New York, NY. May 1966.
57. Reese, R., and D. Marsh. An Update of Dry Scrubber Utilization and
On-Line Test Results. In: Proceedings of the 1978 TAPPI Environmental
Conference. Washington, DC. April 1978. p. 47.
58. Reference 22, p. 51.
59. Gorman, P., et al. St. Louis Demonstration Project Final Report: Power
Plant Equipment, Facilities, and Environmental Evaluations. U.S.
Environmental Protection Agency. Publication No. EPA-600/2-77-155b.
December 1977.
60. McGarry, F. J., and C. J. Gregory. A Comparison of the Size Distri-
bution of Particles Emitted from Air, Mechanical, and Steam Atomized
Oil Fired Burners. Journal of the Air Pollution Control Association.
22:8. August 1972.
9.2-56
-------
61. Vandegrift, A. E., et al. Partibulate Pollutant Systems Study.
Midwest Research Institute. Publication No. APTD 0743.
62. Offen, G. R., et al. Control of,Participate Matter from Oil Burners
and Boilers. U.S. Environmental;Protection Agency. Publication No.
EPA-450/2-76-005. April 1976. j
63. Hartshorn, W. T. Electrostatic Dust Collection from Oil-Fired
Boilers. TAPPI. 5j5:6. June 1973.
64. Bagwell, F. H., and R. G. Velte., New Developments in Dust Collecting
Equipment for Electric Utilities. Journal of the Air Pollution
Control Association. ^:12. December 1971.
65. Pinheiro, G. Precipitators for Oil-Fired Boilers. Power Engineering.
_75:4. April 1971. ;
i
I
66. Sahagian, J., R. Dennis, and N. Surprenant. Particulate Emission
Control Systems for Oil-Fired Boilers. U.S. Environmental Protection
Agency. Publication No. EPA-450/3-74-063.
i
67. Study of Electrostatic Precipitators Installed on Oil-Fired Boilers.
Electric Power Research Institute. Palo Alto, CA. EPRI FP-792,
Project 413-1, Task 1.4, Volume 2, Final Report. June 1978.
68. Higgins, R. S., et al. Experience in the Use of Brown Coals from
the Latrobe Valley, Australia. (Presented at Technology of Use of
Lignite Proceedings: Bureau of Mines. University of North Dakota,
Symposium. Grand Forks, ND. May 9-10, 1973.) IC-8650.
69. Fernandes, J. H., et al. Boiler Emissions and Their Control.
(Presented at Conference on Air Pollution Control. Mexico City.
April 28, 1976.)
70. Burdock, J. L. Centrifugal Collectors Control Particulate Emissions
from Oil-Fired Boilers. TAPPI. :&5:6. June 1973.
71. Safford, D. E. Burner Changes vs. Fuel Changes. Hydrocarbon
Processing. £9_:2. February 197Q.
72. Schreter, R. E., et al. Industrial Burners - Today and Tomorrow.
Mechanical Engineering. June 1970.
73. Drake, P. F., and E. H. Hubbard. ; Combustion System Aerodynamics and
Their Effects on the Burning of Heavy Fuel Oil. Journal of the
Institute of Fuel. _39:302. March 1966.
74. Robison, E. B. Applications of Dust Collectors to Residual Oil-Fired
Boilers in Maryland. Maryland Bureau of Air Quality Control.
Publication No. BAQC-TM-74-15. December 1974.
75. Giammar, R. D., et al. The Effects of Additives in Reducing Particu-
late Emissions from Residual Oil 'Combustion. (Presented at Symposium
on Stationary Source Combustion. U.S. Environmental Protection Agency.
Atlanta, GA. September 24-26, 19,76.)
i
i
9.!2-57
-------
Page Intentionally Blank
-------
9.3 REFUSE INCINERATORS
Refuse incinerators are furnaces;that burn wastes. In addition, they
do not recover usable heat. This failure to recover heat in the form of
steam and/or hot water generation distinguishes incinerators from nonfossil-
fueled boilers (NFFB) and other heat recovery equipment using heat transfer
principles to recover heat from combustion gases. Hot water generators
and process heaters are examples of non-NFFB heat recovery equipment that
can use wastes as fuel. i
Refuse incinerators discussed in-this section are categorized as
municipal incinerators, industrial and commercial incinerators, and sludge
incinerators. Table 9.3-1 presents uncontrolled and controlled emissions
data and emission factors for all types of refuse incinerators.!^ i^-\s
table is referenced throughout Section 9.3.
9.3.1 Municipal Incinerators ;
Municipal incinerators may be classified as large or small units.
Large municipal incinerators (processing greater than or equal to 45
megagrams of refuse per day) are declining in use for the following reasons:
o The increased cost of construction or upgrading existing
incinerators to meet pollution regulations makes other waste
disposal options more attracti;ve.
o Resource recovery systems and
-------
Table 9.3-1. SUMMARY OF PARTICIPATE EMISSION TEST DATA FROM INCINERATORS*
IO
IN3
Source No. of sources
tested
Municipal
Incinerators
Pathological
Incinerator
Multiple-hearth
sewage sludge
Fluidlzed-bed
burning
sewage sludge
Other Incinerator
designs burning
sewage or
industrial sludge
Trash
Incinerators
Cotton gin
waste Incinerator
Liquid waste
Paint racks
Film
Hood scrap
Paper drums'5
13
37
5
4
3
64
1
5
1
1
2
1
1
Uncontrolled emissions, Emission factor ,2 Controlled emissions,
kg/Mg feed [mean] kg/Mg kg/Mg feed [mean]
0.2 - 19.5
C5.8]
0.52 - 16.3
[3.11]
no data
no data
1.3 - 7.3C
0.06 - 4.05
[0.67]
4.38 -'9.84
[6.72]
4.15 - 7.63
[5.8]
0.10 g/m3
3.1
0.06 g/m3
0.41 g/m3
7.5 - 15.0 0.37 - 0.91
[0.48]
4.0 0.05 - 17.1
[2.98]
50.0 0.14 - 2.0 kg/Mg dry
sludge [0.84]
50.0 0.35 - 2.8
[1.3]
50.0 0.09 - 3.3
[0.86]
3.5 - 7.5 0.35 - 1.0
[0.62]
3.5 - 7.5 9.25 - 12.7
[11.5]
3.5 - 7.5 0.0 - 9.72
[2.91]
3.5 - 7.5
3.5 - 7.5
3.5 - 7.5 .
0.07 g/m3d
3.5 - 7.5
aWet spray collector, a low efficiency collection device.
''Paper drums burned with smokeless powder and activated charcoal.
cMean not available.
-------
The reasons for the popularity of small municipal incinerators are low •
cost, relative simplicity, and relatively easy pollution control requirements.
Most small municipal incinerators are located in the South and Northeast.4
Total particulate emissions fronj all municipal incinerators have been
estimated at 34 Gg for 1977.5 This value may include emissions from boilers
burning mixed municipal refuse. ;
9.3.1.1 Source Description. Se;veral furnace designs can be used for
municipal incinerators. These furnaqe designs utilize grates and combustion
chambers in a variety of configurations. While no design can be considered
best, certain configurations are more suitable for combusting wastes with
specific characteristics. j -
Furnace varieties commonly used |for combusting municipal wastes include:
vertical circular, multicell rectangular, rectangular, and rotary kiln. These
furnace types are normally designed for heat release rates of approximately
680 MJ per hour per cubic meter of furnace volume. Heat release rates can
vary from 470 to 940 MJ per hour per ;cubic meter. Details on these types of
municipal incinerator furnaces are giiven below.
9.3.1.1.1 Vertical circular furnaces. The vertical circular furnace is
usually a refractory lined, circular 'chamber. Solid waste is charged through
a gate or lid in the upper part (usually the ceiling) and drops onto a rotat-
ing, central cone grate which is surrounded by a stationary circular grate.
The primary combustion air is underfire forced-air which also serves to cool
the grates. Rabble arms extend from the cone grate and, as the cone rotates
slowly, agitate the solid waste bed. Residue is displaced to the sides
where it is discharged, manually or mechanically, through a dumping grate
on the periphery of the stationary circular grate. Stoking doors provide
for manual agitation and assistance in residue dumping, if necessary. Over-
fire air is usually introduced into the upper portion of the circular chamber.
9.3.1.1.2 Rectangular furnaces.; The rectangular furnace is the most
common modern furnace for municipal wiaste disposal. Several, grate systems
are adaptable to the rectangular furnace. Two or more grates are usually
arranged in tiers to agitate the solid waste as it drops from one level to
the next. Each furnace has only one charging chute. Secondary combustion
of gases often occurs in the rear of the furnace behind a curtain wall. This
9.3-3
-------
wall also radiates heat toward the charging grate (the first grate upon which
the waste drops) to augment drying, promote ignition, and increase combustion
gas velocity and the level of turbulence.
The multicell rectangular furnace may be water-cooled or refractory
lined; it is also known as a mutual assistance furnace. Two or more furnace
cells, each with rectangular grates, are arranged side-by-side. These cells
are ordinarily connected to a common secondary combustion chamber and residue
disposal hopper. Solid waste is charged through openings in the top of each
cell.
9.3.1.1.3 Rotary kiln furnaces. The rotary kiln furnace is a slowly
revolving inclined kiln into which the waste is fed after it has been dried
and partially burned in a rectangular chamber. Tumbling action in the kiln
exposes unburned material to additional burning. Upon exiting the kiln, com-
bustion gases and entrained particulate matter are then burned in a mixing
chamber. Ash falls from the end of the kiln into a quench trough and is
mechanically removed.
9.3.1.2 Emission Characteristics and Applicable Control Technology.
The quantity, size distribution, and composition of particulate emissions
from municipal incinerators varies widely depending on furnace design, type
of refuse fired, method of feeding, operating procedures, and completeness
of combustion. The important properties of particulate matter, from the
standpoint of control equipment design, are mass loading, particle size
distribution, specific gravity, electrical resistivity, and chemical com-
position. These properties are detailed for a number of installations in
References 6 and 7. Average stack parameter data for single and multiple
chamber municipal incinerators are presented in Table 9.3-2.8
Optimum control of incinerator particulate emissions begins with proper
furnace design and careful operation. A proper design includes: (1) a fur-
nace/grate system appropriate to the waste, (2) an adequate combustion gas
retention time and velocity in the secondary combustion chamber, (3) a suit-
able underfire and overfire air system, and (4) appropriate construction
materials to maintain temperatures with minimal auxiliary fuel.9 Operational
procedures used to minimize particulate emissions include: (1) using furnace
loading rates appropriate to the waste type, (2) maintaining proper tempera-
ture, (3) supplying combustion air in sufficient volume and velocity, and
9.3-4
-------
Table 9.3-2. MUNICIPAL INCINERATOR STACK PARAMETER DATA3
Emission sources
Number of sources
Average stack height, m
Average stack diameter, m
Average temperature, °C
Average gas flowrate, Am3/s
Average annual quantity burned,
Gg/incinerator
Single chamber
1
| 277
12.8
i
, 0.731
520.6
! 7.91
5.58
Multiple chamber
209
33.2
1.95
312.8
28.1
28.8
Caution: The total number of municipal incinerators in this table
(486 units) varies considerably from the total number of
large and small municipal incinerators (approximately 140
units) identified in References 3 and 4, respectively.
The difference may be that the NEDS data (Reference 8)
used for this table is older and does not reflect the
decline in the municipal,incinerator population over the
past few years, and units capable of some form of boiler-
type heat recovery were not included from References 3
and 4 as municipal incinerators.
Therefore, the above data should be used only as a general
guide for whatever purpose. Specific data, if available,
should be used for modeling considerations.
9.3-5
-------
(4) establishing the optimum underfire/overfire air ratio. These operational
procedures are quantified in Reference 10.
Although optimizing the combustion process through proper furnace design
and careful operation yields some reduction of particulate emissions, add-on
control devices are required for municipal incinerators to meet emission stan-
dards. Wet scrubbers and electrostatic precipitators have found wide use for
this purpose. Fabric filters are still primarily experimental. Settling
chambers, flooded or spray-wetted baffles, mechanical collectors, and after-
burners have been used as final particulate control devices. However, be-
cause they have relatively low efficiency on small particles, these devices
are now generally used only as gas precleaners. Exceptions are small munic-
ipal incinerators, where afterburners were the only control technology em-
ployed as of August 1979.4
Wet scrubbers applied to municipal incinerators have the following char-
acteristics:
o Submerged entry of gases
o Spray-wetted-wall cyclones
o Venturi or orifice scrubbers
The first two configurations are moderate-to-low pressure drop scrubbers
(generally less than 2.5 kPa) capable of only moderate collection efficiency.
To meet current standards,9 scrubbers with pressure drops exceeding 3.7 kPa
and collection efficiencies above 95 percent are required.^ The most success-
ful, high efficiency scrubber design for municipal incinerator applications has
been the venturi scrubber; however, flooded disk scrubbers can be used for the
same applications. Pressure drops normally range from 3.7 to 12 kPa, and
liquid use varies between 0.7 and 2 liters per actual cubic meter of exhaust
gas.
aAn example is the current New Source Performance Standard of 0.18 grams per
dry standard cubic meter at 12 percent carbon dioxide (0.08 grains per
dry standard cubic foot at 12 percent carbon dioxide) for large municipal
incinerators. For typical solid wastes burned in the United States, this
limit equates to 0.77 kg/Mg (1.5 Ib/ton).
9.3-6
-------
In addition to high energy requirements, the main disadvantages of high
efficiency scrubbers, such as the verituri and flooded disk, are serious cor-
rosion/erosion problems and the need Ifor a liquid waste treatment system.
Corrosion problems can be solved by using corrosion resistant materials and
pH control. Liquid waste treatment requirements can be reduced by recircu-
lation. However, solids-monitoring instrumentation, recycle tanks, and addi-
tional pumps are required.
9.3.1.2.1 Electrostatic precipi'tators. Electrostatic precipitators
(ESPs) installed on municipal incinerators are usually the single-stage,
duct type with horizontal gas flow. Normally, insulation is required on the
shell to minimize corrosion due to acidic liquid condensation- Discharge
electrodes are either of the weighted wire or supported frame variety. Col-
lection electrodes are parallel stud plates spaced 20 to 30 cm apart. Rappers
are the impact type—either mechanical, pneumatic, or electromagnetic. Water
sprays may also be installed under the ESP's roof to rinse the collection
electrodes. i
Waste properties and the temperajture of the gases entering the ESP are
of utmost importance. These parameters affect particle resistivity and,
hence, influence collection efficiency. Since moisture lowers the resis-
tivity of the particles, it increases! ESP efficiency. Moisture is often
added to exhaust gases to promote thijs effect while at the same time lowering
gas temperature. For typical installations, an optimum temperature range
for precipitators is 200 to 315°C.H ;
Other factors that may affect the particulate matter collection effi-
ciency of an ESP are gas velocity and the condition of the unit. Excessive
velocities may re-entrain flyash and increase particulate emissions. Non-
uniform flow may occur within the unit at lower flowrates, thereby creating
increased particulate emissions due tb channeling. Maintenance is extremely
important to sustain optimum collection efficiency in such complex devices.
Table 9.3-3 presents typical ESP design parameters for incinerator
applications.12
Corrosion of electrostatic precipitators installed on municipal inciner-
ators is a serious problem which must receive adequate consideration during
design. Incinerator flue gas contains significant amounts of hydrochloric
9.3-7
-------
Table 9.3-3. TYPICAL ELECTROSTATIC PRECIPITATOR DESIGN PARAMETERS
FOR MUNICIPAL INCINERATOR APPLICATIONS 12
Plate spacing, cm 20-30
Velocity through precipitator, m/s 0.9-1.8
Vertical height of plates, m 3.6-10
Horizontal length of plates9 0.5-1.5 x height
Applied voltage, volts 30,000-80,000
Gas temperature, °C 177-343
Gas residence time in precipitator, sec 3-6
Draft loss, kPa 0.03 to 0.2
Fields (electrical stages) in direction
of gas flow 1-4
Total power for precipitator, KW 7-35 KW
per m3/min (0.2-1 KW/1000 ACFM)
Collection area, m2 per 1000 m3/min 400-1000
Efficiency, % ' 93-99
Gas flow per precipitator, m3/min 850-8500
Migration velocity,*5 cm/s 6-12
aAspect ratio s (total horizontal length [depth] of collection plates)
(height of collection plates) = 0.5-1.5.
bMean average effective migration velocity of a particle toward the
collection electrode. Sometimes called precipitation rate or drift
velocity.
9.3-8
-------
acid and lesser amounts of S02; the S02 reacts with water to form sulfuric
acid. These compounds, coupled with the high moisture content of the flue
gas (5 to 20 percent), are responsible for creating the corrosive environment.
There are three ways to minimize corrosion: (1) precipitator gas temperature
is maintained above the acid dewpoint '(generally above 1750C) by using ade-
quate insulation, sealing off atmospheric air infiltration, and using auxil-
iary heaters; (2) precipitator shutdowns, in which cooling can occur, are
minimized and the acid laden gas is purged immediately after shutdown; and
(3) corrosion resistant materials are used.13
9.3.1.2.2 Fabric filters. The fabric filter, or baghouse,.when proper-
ly designed and operated, is considered the most efficient particulate emis-
sion control device. The limited number of fabric filters installed on
municipal incinerators all use si 1 iconic-treated fiberglass bags to with-
stand high temperatures; bags last fro'm 1 to 3 years. Control efficiencies
have been measured from 98 to greater than 99 percent. Pressure drops are
between 0.75 and 1.75 kPa; air-to-cloth ratios (filter velocities) range
between 0.6 and 1.2 meters per minute.14
Fabric filters are very sensitive to operating temperatures. The maxi-
mum satisfactory range for gas temperature is 120 to 29QOC; the range can be
much narrower for certain bag fabrics and exhaust gas compositions. Tempera-
tures in excess of 260°C will: damage1 the silicone coating, accelerate bag
deterioration, distort the metal frame|, and potentially cause fires. Below
150°C, condensation and caking can lead to blinding of the filter and bag
failure. The unit should also be sealed against infiltration of cool, am-
bient air.
All free moisture (entrained droplets or moist flyash) should be removed
from the gas stream to prevent blinding of the filter. Since incinerator
i
flue gases are usually cooled to an acceptable baghouse inlet temperature by
water spray, adequate residence time must be provided following gas cooling
to allow for complete evaporation of the water before the gases reach the
baghouse.
Pertinent design and operating data for a successful commercial baghouse
incinerator installation are presented1 in Table 9.3-4.15
9.3^-9
-------
Table 9.3-4. OPERATING AND DESIGN PARAMETERS FOR FABRIC FILTER
BAGHOUSE ON MUNICIPAL INCINERATOR15
Unit capacity, Mg/day N/A
Air flow, m3/min 5090
Air temperature, °C 260
Fabric Glass fiber
Air/cloth ratio, m^/min per m^ 0.61
Bag size ;
diameter, m 0.14
length, m 4.27
Number of bags, approximately 4350
Method of cleaning Reverse air
Design pressure drop, kPa 0.5 to 0.75
9.3-10
-------
9.3.2 Industrial and Commercial Incinerators
The various types of industrial1 and commercial incinerators are listed
and quantified in Table 9.3-5.16 Each type of incinerator is distributed
fairly evenly throughout the United States, with the exception of apartment
incinerators and conical (teepee) incinerators. The bulk of apartment in-
cinerators are concentrated in EPA Rjegion II, particularly in New York City,
while teepee incinerators are primarily located in the Southern and Western
U.S. Although presented in Table 9.3-5, incinerators used in apartments,
stores, schools, and woodworking processes (teepee incinerators) are not
specifically addressed in this sectibn. They are declining in use and are
projected to be essentially phased out by 1988.16
Particulate emissions from industrial and commercial incinerators are
! *
quite variable and depend principally on the type of waste processed. Total
nationwide particulate emissions from industrial and commercial incinerators
have been estimated at 158 Mg for 1977.15
9.3.2.1 Source Description and!Emissions. The following paragraphs
describe the types of incinerators generally considered as industrial and
commercial incinerators and their emissions.
9.3.2.1.1 Single chamber incinerators. Single chamber incinerators are
usually furnace boxes consisting of a combustion chamber and an ash pit. A
grate separates the chamber from theipit. Waste is fed through a charging
door which is located above the grate. An auxiliary fuel burner is normally
located below the grate. An ash cleanout door provides access to the ash pit.
Openings are incorporated in the charging door for overfire air and in the
cleanout door for underfire air. Units are usually natural draft, although
mechanical-draft excess-air units do exist. Natural-draft units are normally
controlled by a barometric damper.
The advantages of the single chamber incinerator are simplicity and low
cost. The disadvantages, due principally to design, include inadequate mixing
i
of combustion gases, short retention !time, extreme temperature fluctuations,
and entrainment of particulate matter in the flue gas. Incomplete combustion,
due to various combinations of the above,disadvantages results in excessive
emissions of particulate and gaseous hydrocarbons. The emissions often create
9.3-11
-------
Table 9.3-5. INDUSTRIAL AND COMMERCIAL INCINERATOR TYPES16
CO
!-»
ro
Incinerator types
Commercial and
institutional
o Hospital, nursing
home
o Store, school, etc.
Apartment
Pathological
Industrial
Conical (teepee)
Number of
operating
incinerators
in 1978
3,498
8,161
6,388
1,990
4,136
551
Current
average size
of incinerator,
Mg/y
135
135
140
79
450
7,864
Projected
number of
operating
incinerators
in 1983
3,291
3,084
4,114
2,260
4,098
231
Projected
average size
of incinerator
in 1983,
Mg/y
163
163
140
79
495
7,864
-------
dark smoke, foul odors, and carbon monoxide. Particle size distributions
for single chamber incinerators vary considerably according to the design
and condition of the unit and the type of waste charged.
9.3.2.1.2. Multiple chamber incinerators. Multiple chamber incinera-
tors are built with two or more interconnected refractory-lined chambers
designed to provide optimum combustion of the charged solid waste. A rep-
resentative multiple chamber incinerator has a primary or ignition chamber
which is divided into a combustion settion and an ashpit by a grate on which
the solid waste is oxidized. An auxiliary fuel burner is normally located
beneath the grate. The primary chamber has both overfire and underfire air
controls. ; . ,
A mixing or downpass chamber folilows the primary chamber. The mixing
and primary chambers are separated by a bridge wall which has an opening,
the flame port, at its top. This port allows combustion gas to flow from
the primary chamber into the mixing chamber. Secondary air ports in the
mixing chamber supply air for completion of combustion. The secondary
I
burner is located in this chamber. j
The last chamber is the secondary combustion chamber. It is separated
from the mixing chamber by a curtain wall. This wall has an opening at the
bottom which allows mixing chamber combustion gases to enter the secondary
combustion chamber for final combustion. Some particulate removal occurs in
this last chamber due to wall impingement and simple settling.
Multiple chamber incinerators are usually natural draft. However, me-
chanical-draft excess-air and starved-air units also exist. Multiple chamber
incinerators can be field-erected or of the modular variety depending on the
size desired. Units can also be of the in-line or retort configurations.
Each configuration offers certain advantages which are fully described in the
literature. Particle size distributions for multiple chamber incinerators de-
pend on the unit's design and condition, and the type of waste charged.
9.3.2.1.3 Control 1ed-air incinerators. Controlled-air incinerators
consist of two distinct refractory lined chambers. A reducing atmosphere is
maintained in the primary chamber, and an oxidizing atmosphere is maintained
9.3-13
-------
in the secondary chamber. Starved-air incinerators could be considered a
subset of controlled-air incinerators because no overfire air is supplied in
the primary chamber.
Waste is charged into the primary chamber where, under a reducing at-
mosphere, only a portion of the waste is oxidized. Usually, it is the fixed
carbon (the char) which burns and releases heat. The heat pyrolyzes the
volatile portion of the waste, releasing a dense combustible smoke which
then passes into the secondary chamber.
Therefore, the primary chamber can have the following four zones:
o Ash bed
o Char bed
o Pyrolysis
o Overfire
These zones are present for all waste materials regardless of chemical com-
position, physical state, and water and ash content. However, the size of
the zones may vary depending on the waste. The ash bed at the bottom of
the chamber is the inert region where the inorganic, incombustible part of
the waste collects. The char bed is where the char portion of the waste is
oxidized. The pyrolysis zone contains waste in various stages of gasifica-
tion. The overfire zone is the region through which the gases pass to the
secondary chamber.
Upon entering the secondary chamber the gases are mixed with additional
air; the mixture is then ignited, and the smoke is oxidized. Auxiliary fuel
is a requirement only if the mixture does not support combustion. Temper-
atures are maintained between 760 and 1370°C to ensure complete combustion,
minimize nitrogen oxide production, and protect equipment. Proper air control
by controlled air dampers and temperature control by auxiliary fuel control
valves ensure that the desired secondary chamber temperature is maintained.
Controlled-air incinerators were developed because single and multiple
chamber incinerators could not satisfy federal, state, or local particulate
emission requirements without add-on pollution control equipment such as
scrubbers. As with single and multiple chamber incinerators, particulate
matter distribution varies according to unit design condition and the type
of waste charged.
9.3-14
-------
The next few pages describe special purpose industrial and commercial
incinerators that are used to a significantly lesser degree than those
incinerators previously discussed. '
9.3.2.1.4 Rotary kiln incinerator. The rotary kiln incinerator is a
horizontal, refractory-lined slightly inclined hollow cylinder that slowly
rotates axially. Waste fed at the higher end traverses the length of the
kiln due to the revolving motion andithe incline. Liquid and gaseous wastes
are charged through auxiliary burners at the kiln's lower end. The waste is
incinerated by this countercurrent charging, and the noncombustibles drop
out at the lower end. The rotating action continually exposes new surfaces
for oxidation and removes ash. Often the discharge end is hooded to allow
combustion gases to pass to a secondary combustion chamber.
All combustion requirements (retention time, air supply, temperature,
and turbulence) can be controlled. Waste retention time is controlled by
varying the kiln's rotational speed..• Gas stream retention time is mandated
by variable air supply controls. Narrow temperature ranges can be maintain-
ed by varying the waste feed rate and the auxiliary burner controls. Com-
bustion gas is mixed by the turbulence caused by the kiln's rotating action
as that gas stream passes from the kiln to the secondary combustion chamber.
A mixing chamber is often located directly between the kiln and the second-
ary chamber to ensure complete combustion.
The kiln's inside diameter must be sufficiently large to keep the veloc-
ity of the gas stream low enough to prevent entrainment of particulate matter.
i '
Rotary kiln incinerators are well suited for all combustible waste including
hazardous organic wastes and sewage sludges.
9.3.2.1.5 Rotary hearth incinerators. The rotary hearth incinerator
is a circular, refactory-lined combustion chamber with a rotating hearth.
Solid waste is charged on the outer perimeter of the slowly rotating hearth.
The added material forces the partly ;burned waste to move in a special path
toward an ash pit in the center of the hearth. Waste charge retention time
is controlled by the rotational speed of the hearth. At a point in the
rotation, the noncombustible ash is scraped off the hearth into the ash
discharge pit by a plow mechanism. Liquid and semi-liquid waste are usually
9.3-15
-------
Auxiliary burners supplement combustion chamber heating needs. Hearth
rotation permits continuous waste feeding and maintains a uniform bed on the
hearth. An optimal constant temperature can be maintained by adjusting the
feed and rotational rates.
Unit design normally provides a cyclonic path for the combustion gases
to pass through the combustion chamber to the outlet. Complete mixing of
the combustion gases is accomplished by this cyclonic action.
9.3.2.1.6 Multiple hearth incinerators. The multiple hearth incinerator
is very similar to the rotary hearth except that it usually consists of 4 to
13 hearths in a vertical stack. Waste is fed to the outer perimeter of the
top hearth and is slowly raked by rabble arms toward the hearth's center
where it falls onto a second hearth. The waste is then raked to the outer
perimeter of the second hearth and falls to a third hearth and so on. Final-
ly, the noncombustible ash falls out of the incinerator's bottom into an ash
removal system.
Three phases of incineration occur at different levels. Evaporation of
moisture and oxidation of volatiles occur on the upper hearths. The devola-
tilized waste is then burned on the middle hearths. Finally, the bottom
hearths cool the ash prior to discharge. Rabble arm teeth continually agi-
tate the waste to expose new surfaces to evaporate moisture and oxidize vol-
atiles. Adequate combustion gas mixing is ensured by the stirring of the
rabble arms and by the directional changes as the gas travels around the
hearths which serve as baffles.
Multiple hearth incinerators are well suited for most organic wastes.
Solids are usually fed on the top hearth, semi-liquids through side ports,
and liquids and gases through auxiliary burners.
9.3.2.1.7 Liquid injection incinerators. Liquid injection incinerators
are vertical or horizontal refractory-lined combustion chambers in which
atomized liquid wastes are burned. Liquid injection is limited to pumpable
liquids and slurries.
Combustion is enhanced by atomizing the liquid to droplets of less than
40 micrometers in diameter. It may be necessary to heat or mix viscous liq-
uids or slurries in order to pump them through atomizing nozzles.
9.3-16
-------
A forced draft to the combustion chamber mixes the combustion gases and
creates turbulence. Frequently, this result is also produced by firing the
atomized waste tangentially to create a cyclonic effect in the chamber.
The required combustion oxygen is supplied by the compressed air used for
waste atomization. Heated, pressurized air can also be injected near the dis-
i • •
charge of the combustion chamber to create an afterburner effect. Additional
\
required heat is provided by auxiliary burners.
Usually, particulate emissions from liquid injection incinerators are of
little consequence since there is little inorganic material in the waste that
would create such emissions.
9.3.2.1.8 Fluidized bed incinerators. The fluidized bed incinerator is
a refractory-lined, hollow metal cylinder with a grid in its lower section
which supports a bed of inert, granufar particles such as sand. Blower-driven
air enters the bottom of the unit and agitates and expands (fluidizes) the
bed, causing it to behave like a boiling liquid. Wastes are then fed pneuma-
tically, mechanically, or by gravity iinto the bed where rapid, relatively
uniform mixing occurs. ,
During combustion, heat is transferred from the bed media to the injected
i
waste materials. Bed temperatures are typically between 760°C and 870°C.
Because fluidized-bed incinerators have up to three times the heat capacity
of conventional incinerators in the s'ame temperature ranges, they are often
i
self-sustaining and need no auxiliary; fuel after start-up. Waste combustion
heat is transferred back to the bed. Solid materials stay in the bed until
they have become light enough to be carried off by the flue gas as particu-
lates. Auxiliary burners heat the bed to the required temperature during
start-up and then supply heat when necessary.
Fluidizing air flow must be properly balanced to fluidize the bed and
provide the required oxygen for combustion. Too much air would blow bed
media and incomplete combustion products out of the bed with the flue gas.
Too much air also depletes the stored!heat energy of the bed.
Advantages of the fluidized bed incinerator are that it: (1) has a
minimum of mechanical components; (2);is relatively simple to operate; and
(3) will incinerate all organic wastes, particularly liquids. It is also
very attractive for intermittent operations since the bed serves as a large
heat reservoir, minimizing the amountiof fuel necessary for start-up follow-
ing shutdown. ;
9.3-17
-------
Stack parameter data for single, multiple, and controlled-air chamber
industrial and commercial incinerators are presented in Table 9.3-6.
9.3.2.2 Emission Control Techniques. Particulate emissions are the
only air pollutants emitted from incinerators for which control devices are
generally considered appropriate. The nature and quantity of particulate
emissions from industrial and commercial incinerators vary depending on:
(1) the waste type (e.g., solids or liquids), quantities, and character-
istics; and (2) the design and operation of the incinerator. Tables 9.3-1
and 9.3-7 provide ranges of uncontrolled and controlled emissions from a
number of industrial incinerators. Table 9.3-8 provides ranges of uncon-
trolled emissions from electric motor incinerators.
The three mechanisms principally responsible for these particulate emis-
sions are:
o Combustion gas entrainment of particles from the burning waste.
o Incomplete combustion, above the burning waste, of the carbon formed by
the cracking of volatiles produced during pyrolysis.
o Condensation of inorganic salts or oxides as the flue gas cools.
Commonly used particulate emission control devices for incinerators include
settling chambers, afterburners, mechanical separators (e.g., cyclones),
scrubbers (e.g., wetted baffles, spray chambers, venturi scrubbers), and, to
a lesser degree, electrostatic precipitators and fabric filters.1
Collection or removal of small diameter particles (less than 5.0 micro- .
meters) is difficult and requires sophisticated and efficient pollution control
equipment. For that reason, several of the above devices by themselves would
be unable to satisfy State regulations on allowable particulate emissions.
Fundamental understanding of particle dynamics and the physical concepts of
the various types of devices is necessary to evaluate these devices for col-
lection or removal of particulate emissions from specific incinerators. De-
vice selection, once it has been determined that applicable regulations can
be satisfied, is a compromise of: (1) particulate matter collection effi-
ciency, (2) annual operating cost, and (3) initial capital investment.
Pollution control devices and their applicability to reducing incinerator
particulate emissions are briefly discussed below.
9.3-18
-------
Table 9.3-6. INDUSTRIAL AND COMMERCIAL INCINERATOR STACK PARAMETER DATA8
CO
i—«
ID
Emission sources
Number of sources
Average stack Height, m
Average stack diameter, m
Average temperature, C
Average gas flowste, Affif'/s"
Average annual quantity burned,
Multiple
chamber
Industrial Commercial
470
13.1
0.79
474.0
- — - 5.39
1.53
1,280
12.8
0.61
450.0
2.21
1.75
Single
Industrial
447
Z2.3
1.2
471.0
27.8
5.42
chamber
Commercial
541
16.1
0,67
345.0
12.2
0.237
Control led-air chamber
Industrial
136
19.2
1.1
519.0
";.: 8.26
2.23
Commercial
44
10.4
0.64
588.0
1.61
0.707
Gg/Incinerator
-------
Table 9.3-7. SUMMARY OF EMISSION TEST DATA FROM INCINERATORS BURNING WIRE INSULATION,
SI UNITS (ENGLISH UNITS)!
Pollutant No. of Uncontrolled emissions, Emission factor2 Controlled emissions,
sources kg/Mg (Ib/ton) kg/Mg (lb/ton) kg/Mg (Ib/ton) kg/Hg (Ib/ton) kg/Mg (lb/ton)
tested feed combustible feed combustible
[mean] [mean]
Wire
insulation 11 7.7 - 12.4 36.1 - 122.5 3.5 - 7.5 0.63 - 7.45 12.9 - 28.5
[10.5] [58.8] [3.43] [21.7]
GO
rj - {15.5-24.9} (72.2-245.0) (7.0-15.0) (1.26-14.9) (25.8-57.1)
0 [21.0] [117.6] [6.86] [43.4]
-------
Table 9.3-8.
SUMMARY OF UNCONTROLLED
MOTOR INCINERATORS, SI
EMISSION TEST DATA FROM ELECTRIC
UNITS (ENGLISH UNITS)1
Pollutant
No of
sources
tested
Emissions,3
kg/Mg (Ib/ton)
[mean]
Emission factor,
kg/Mg (Ib/ton)
Participates
11
1.5 - 5.22
[2.62]
(3.0 - 10.44)
[5.24]
3.5 - 7.5
(7.0 - 15.0)
aBased on combustible feed.
9.3-21
-------
9.3.2.2.1 Settling chambers. Settling chambers are the simplest, least
expensive and least effective participate emission control devices for incin-
erators. The principle employed is to provide enough volume in the chamber
to decrease the combustion gas velocity to 3 meters per second or less. This
velocity reduction allows particles greater than 100 micrometers to settle
out by gravity. Baffles are used in some settling chambers to promote in-
ertial settling, while other chambers wet down the particulate material which
has settled to prevent reentrainment in the gas stream. Depending on the de-
sign, operation, and maintenance, such chambers can achieve, at most, 30 to 35
percent by weight particulate matter removal. This settling chamber method of
particulate emission control is no longer used by itself, but is used in con-
junction with other more efficient particulate emission control devices.
9.3.2.2.2 Afterburners. An afterburner can be either an add-on particu-
late emission control device or an integral component included in the original
incinerator design. The latter is often referred to as the secondary burner
in the secondary combustion chamber. The purpose of the afterburner is to
oxidize organic particulate emissions, gases, and odors. There are two types
of afterburners: direct flame and catalytic. Afterburners have been used
extensively on incinerators and provide excellent organic particulate emis-
sion control. For inorganic particulate matter removal, however, other
control devices must be used.
9.3.2.2.3 Mechanical separators. Mechanical separators include the
most commonly used particulate emission control device, the cyclone separator.
In this device, centrifugal force is used to separate particles from the ex-
haust gas. Large diameter cyclones have low collection efficiencies (down
to 30 percent), especially for particles of less than 30 micrometers. They
are, however, low in capital cost and operate at a low pressure drop of 0.25
to 0.75 kPa. Multiple-tube cyclones can achieve efficiencies of over 90 per-
cent on particles greater than 10 micrometers. Pressure drops of between 0.75
and 1.25 kPa are common with multiclones. Multiclones also have problems with
plugging and erosion. All cyclones, regardless of tube diameter, must contend
with acid corrosion and gas leakage.
9»3.2.2.4 Wet scrubbers. Wet scrubbers commonly used for particulate
matter removal in incinerator flue gases include the following types:
9.3-22
-------
o Wetted baffles
o Spray chambers
o Venturi scrubbers
In each type, participate matter collection is a combination of inertia!
interception, impingement, diffusion, thermal gradients, and electrostatic
attraction. Collection is also affected by particle wetting properties,
moisture condensation, and drop evaporation. Interception and impingement
i
are the two most important mechanisms. r- ;
Wetted Baffles—Wetted baffles are the simplest kind of wet scrubber and
consist of metal or brick screens or plates wetted by flushing sprays or over-
flow weirs. Although wetted baffle? are often installed in the effluent gas
duct, they can also be housed in a separate chamber. Water usage varies be-
tween 1.7 and 6.9 liters per minutejper megagram of waste burned per day. A
pressure drop between 0.075 and 0.15 kPa is common. Efficiencies are quite
low since only particles greater than 50 micrometers are removed.16
Spray Chambers--A spray chamber is a round or rectangular chamber into
which nozzles spray water. Three different water spray configurations are
possible: cocurrent, countercurrent, and crosscurrent. Water usage varies
between 670 and 2,000 liters per actual cubic meter of fl.ue gas per minute.
Pressure drops are normally between 1.25 and 1.75 kPa. Efficiencies of 90
percent or greater are possible, witjh efficiencies declining rapidly for
particles of 5 micrometers or less.
Venturi Scrubbers—Venturi scrubbers promote particulate matter removal
by impacting a high velocity gas stream with a finely atomized water spray.
A number of physical phenomena are involved during this impaction. High
pressure drops of greater than 3.7 kPa make this an energy expensive particu-
late emission control option. However, efficiencies of up to 99 percent or
more are possible for particles as small as 2 micrometers.
9.3.2.2.5 Electrostatic precipitators. Particulate matter collection
in an electrostatic precipitator (ESP) is based on particle charging and
collection on oppositely charged elebtrodes. The pressure drop through a
typical ESP is less than 0.125 kPa. Energy requirements are between 7,000
9,3-23
-------
and 14,000 J/sec/m^ of exhaust gas per minute. In the United States and
Europe, efficiencies- of between 90 and 99.6 percent have been achieved on
incinerator exhaust gases.
9.3.2.2.6 Fabric filters. Fabric filters collect particulates by im-
pinging them on a filter surface. Present fabric filter systems cannot
continually withstand gas temperatures above 260°C; however, metal, carbon,
and ceramic fibers offer promise for operation at higher temperatures.
Also, fabric filters are fire prone if the combustible content of the col-
lected material is sufficient to support combustion.
Typical pressure drops through fabric filters range from 1 to 1.5 kPa.
Filtering velocity (air-to-cloth ratio) ranges from 3 to 6.1 m^/min/m^ of
cloth for units that are cleaned by reverse jet and 0.46 to 0.9 m-Vmin/m^
of cloth for units cleaned by shaking J6
Minimal research, outside of the scope of the previously discussed par-
ticulate emission control devices, is being conducted in new particulate
emission control technologies. Some technology is being borrowed from other
emission control areas that use combustion to recover heat (e.g., nonfossil
fuel-fired boilers); examples of these other areas include dry scrubbers and
gravel-bed filters. However, this emission control technology is in the
experimental stage and is poorly documented.
9.3.3 Sludge Incinerators
Sludge incinerators have been developed because solid waste and liquid
waste incinerators cannot adequately handle sludges or slurries. Solid waste
incinerators can only destroy very small proportions of sludges because of the
sludges' low heat content and high water content. Liquid waste incinerators
cannot handle sludges because of solids handling problems. The incinerator
design options available are limited; these options are discussed below.
The four broad classes of sludges processed in sludge incinerators are
as follows:
o Flocculent sludges from primary sedimentation of effluents.
o Biological sludges from secondary sedimentation of biological
treatment processes.
o Chemical sludges from neutralization and precipitation processes.
o Oil and hydrocarbon sludges from the mineral and petrochemical
industries.
9.3-24
-------
Most of the ash produced during sludge incineration is discharged with the
stack gas.^ ;
9.3.3.1 Source Description and Emissions. Multiple hearth and fluidized
bed incinerators are the most common incinerators used for sludge incineration.
The multiple hearth incinerator is used to dispose of most forms of com-
bustible wastes. It can incinerate combustible sludges, tars, granulated
solids, liquids and gases; and it is especially well suited to the disposal
of spent biological treatment facility sludges. For this reason, a disposal
facility, especially one which contains biological treatment capability, could
contain a multiple hearth unit. About 120 such multiple hearth units have
been incorporated as part of biological treatment disposal facilities.. The
units are designed with diameters varying from 1.83 to 6.70 meters; they are
capable of handling from 5 to 1,130 Mg of waste per 24 hours, with the number
of hearths usually ranging between 4 and 12.
The fluidized bed incinerator disposes of combustible solid, liquid, and
gaseous wastes. For this reason, fluidized bed incinerators are especially
suitable for the disposal of sludges.- Standard combustion units rely on
heat transfer from hot gases that contain only 600 J/m3. However, the ex-
panded bed of the fluidized bed incinerator has 60,000 J/m^, and therefore,
more heat to drive off the liquid phase.
Other less frequently used incineration or thermal techniques include
flash drying with incineration, wet air oxidation, atomized suspension, and
cyclonic and infrared incineration. ;
Stack parameters for sludge incinerators are presented in Table 9.3-9.
9.3,3.2 Emission Control Techniques. New Source Performance Standards
(NSPS) regulating discharges from municipal sludge incinerators have been
promulgated by the Environmental Protection Agency (EPA). These standards
limit discharge of particulate matter; to 0.65 grams per kilogram of dry
sludge input from both new and modified sewage sludge incinerators. Parti-
culate matter collection efficiencies of 98.5 to 99.5 percent are required
to meet the NSPS.18 Particulate emissions from older, non-modified incinera-
tors not subject to NSPS are regulated by State and local agencies through
incinerator codes or process weight restrictions. Table 9.3-1 provides
ranges of uncontrolled and controlled emissions from sewage sludge
incinerators. i
91.3-25
-------
Table 9.3-9. GOVERNMENTAL, COMMERCIAL, AND INDUSTRIAL SLUDGE
INCINERATOR STACK PARAMETER DATA8
Emission sources
Number of sources
Average stack height, m
Average stack diameter, m
Average temperature, QC
Average gas flowrate, AnrVs
Average annual quantity burned,
Gg/Incinerator
Governmental
75
23.8
1.5
146.0
15.9
13.1
Commerci al
18
28.4
2.2
419.0
31.0
283.0
Industrial
69
12.5
0.67
685.0
6.58
2.52
9.3-26
-------
The wet scrubber is the most common participate emission control device
used in sludge incineration. Venturi; scrubbers and impingement scrubbers
used in conjunction with an oxygen meter that automatically regulates fuel
burning rate are capable of meeting the NSPS.19 Electrostatic precipitators
could also provide more than adequate'control and are doing so in Japan,
where their performance has formed thfe basis for setting emission stan-
dards. 16 -.'!'.
9.3-27
-------
REFERENCES FOR SECTION 9.3
1. Source Category Survey: Industrial Incinerators. U.S. Environmental
Protection Agency. Publication No. EPA 450/3-80-013. May 1980.
2. Compilation of Air Pollution Factors, Second Edition. U.S. Environmental
Protection Agency. Publication No. EPA-AP-42, August 1977.
3. Personal Communication from Van Noordwyk, H., Acurex Corporation, Mt.
View, CA, to Alvarez, R. J., Hofstra University, Hempstead, NY.
October 1979.
4. Blackwell, C. D. Survey of Small Municipal Incinerators. Acurex
Corporation. Research Triangle Park, NC. August 1979.
5. OAQPS Data File of Nationwide Emissions. U.S. Environmental Protection
Agency, Office of Air Quality Planning and Standards. Research Triangle
Park, NC. February 1979.
6. Niessen, W. R. et al. Systems Study of Air Pollution From Municipal
Incineration, Volume I. Arthur D. Little, Incorporated. Cambridge, MA.
U.S. Department of Health, Education, and Welfare, National Air Pollution
Control Administration. Contract. No. CPA-22-69-23. March 1970.
7. Walker, A. B., and F. W. Schmitz. Characteristics of Furnace Emissions
from Large, Mechanically-Stoked Municipal Incinerators. (Presented at
the 1966 National Incinerator Conference. American Society of
Mechanical Engineers. New York, NY. May 1-4, 1966.)
8. Atmospheric Modeling Data from National Emission Data System (NEDS).
U.S. Environmental Protection Agency, Office of Air Quality Planning
and Standards. Research Triangle Park, NC. May 1979.
9. DeMarco, J., et al. Municipal-Scale Incinerator Design and Operation.
U.S. Department of Health, Education, and Welfare, Bureau of Solid
Waste Management, Public Health Service. Publication No. 2012. 1969.
10. Field Operations and Enforcement Manual for Air Pollution Control,
Volume II. U.S. Environmental Protection Agency, Office of Air Programs.
Publication No. APTD-1101. August 1972.
11. Municipal Incinerator Enforcement Manual. U.S. Environmental Protection
Agency, Office of General Enforcement. Washington, DC. Publication
No. EPA 340/1-76-013. January 1977.
12. Weinstein, N. J. and R. F. Toro. Thermal Processing of Municipal Solid
Waste for Resource and Energy Recovery. Ann Arbor Science Publishers
Inc. Ann Arbor, MI. 1976.
13. Franconeri, P. Electrostatic Precipitator, Corrosion on Incinerator
Applications. (Presented at the 68th Annual Meeting of the Air Pollu-
tion Control Association. June 1975.)
9,3-28
-------
14. Inspection Manual for Enforcement of New Source Performance Standards,
Municipal Incinerators. U.S. Environmental Protection Agency, Office of
General Enforcement. Washington] DC. Publication No. EPA 340/1-75-003.
February 1975. I
15. Bergmann, L. New Fabrics and Their Potential Application. Journal of
Air Pollution Control Association. 24:12, 1187-1192. December 1974.
16. Mclnnes, R. G., et al. Screening Study to Determine the Need for
Standards of Performance for Industrial and Commercial Incinerators.
GCA Corporation. Bedford, MA. January 1979.
17. The Mcllvaine Scrubber Manual, rjlcllvaine Company. Northbrook, IL.
1974. ;
18. Helfand, R. M. A Review of Standards of Performance for New Stationary
Sources - Sewage Sludge Incinerators. The MITRE Corporation. McLean,
VA. U.S. Environmental Protection Agency. Contract No. 68-02-2526.
April 1979. ' j
19. Background Information for New Source Performance Standards, Volume III.
U.S. Environmental Protection Agency. Publication No. 450/2-74-003.
APTD-1352C. February 1974. i
9J3-29
-------
Page Intentionally Blank
-------
9.4 OPEN BURNING ;
In the U.S., solid waste treatment by open burning includes: (1) agri-
cultural crop residue burning, (2) prescribed burning of naturally occurring
forest materials and forest slash, and (3) municipal solid waste burning.
The burning of the first two categories of waste contributes about 503 Gg
of particulate matter to the atmosphejre annually, while emissions from munic-
ipal solid waste burning in open dumps is much less significant. 1>2 Although
emissions from municipal solid waste .burning can be a significant factor in
localized air quality problems, the quantity of emissions are minimal on
a national basis when compared to tho'se from agricultural and prescribed
burning. In addition, municipal solijd waste disposal via open burning is be-
ing conducted less each year. For these reasons, only emissions and control
technologies characteristic of agricultural and prescribed burning are dis-
cussed here.
9.4.1 Agricultural Burning
Agricultural open burning is usecl for the disposal of residues from field
crops, row crops, and fruit trees. These residues include: rice straw and
stubble, barley straw and stubble, wheat residues, orchard prunings and nat-
ural attrition losses, grass straw and stubble, potato and peanut vines, to-
bacco stalks, soybean residues, hay residues, sugarcane leaves, and farmland
grass and weeds. Several important advantages are realized from agricultural
open burning:
o Low cost residue removal and disposal
o Preparation of farmlands for cultivation
o Clearing of vines or leaves to facilitate harvesting
o Disease control !
o Direct weed control by incineration of weed seeds and plants
o Indirect weed control by providing clean soil surface for
soil-active herbicides ;
o Destruction of certain mites, iinsects, and rodents
It has been estimated that 13.2 Tg of vegetation were burned in the U.S.
in 1973, resulting in the emission of [US'Gg of particulate matter.^ A sum-
mary of the quantities of agricultural; residues burned and particu-late emis-
sions produced in each state is presented in Table 9.4-1. Most burning occurs
9.4-1
-------
Table 9.4-1.
SUMMARY OF 1973 STATE AND NATIONWIDE AGRICULTURAL
OPEN BURNING DATA-"*
State
AT abama
Arizona
Arkansas
California
Col orado
Del aware
Florida
Georgia
Hawaii
Idaho
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Mexico
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
South Carolina
South Dakota
Tennessee
Virginia
Washington
Wisconsin
Wyoming
National Total
Area burned,
107 m2/yr
36
4.5
26
309
32
0.1
109
39.5
44.7
4.8
243
13
142
15
0.6
0.6
21
61.2
138
40.5
34
67.6
0.8
0.5
138
97
32
36
107
15
8
59
14
9.5
57
32
7.3
2^350
Amount burned,
Gg/yr
161
20
115
2,075
143
2
1,716
883
1,175
22
544
58
1,904
33
7
3
93
274
617
181
152
303
5
2
619
438
142
161
479
69
38
265
61
43
256
145
34
13,242
Parti cul ate
emissions,
Mg/yr
1,400
170
980
26,000
1,200
17
15,000
7,500
10,000
190
4,600
490
16,000
280
60
26
790
2,300
5,200
1,500
1,300
2,600
43
17
5,300
3,700
1,200
1,400
4,100
590
320
2,300
520
370
2,200
1,200
290
113,000
9.4-2
-------
in the Western and Southern States, and Hawaii. The quantity of agricultural
wastes disposed of by open burning has been declining over the past 10 years,
primarily due to: (1) increasing awareness of its contribution to air pollu-
tion; (2) the development of waste utilization; (3) chemical weed, pest, and
disease controls; and (4) modern tillage and fertilization practices. In
some cases, however, no economical alternatives to open burning exist at this
time. ; . • -
i
Particulate emissions from burning are formed as a consequence of poor
mixing between the fuel and air and the quenching of the combustion gases by
the surrounding cool air. Particles consist of carbon, flyash, and condensed
organic materials emitted during the pyrolysis stage of the burning process.
The majority of these particles are in the submicrometer size range and thus
can have adverse health and environmental effects.^ Emission factors for
agricultural open burning are presented in References 3, 5, and 6.
9.4.1.1 Methods of Burning. The major burning method utilized in agri-
cultural residue disposal is burning by headfire or backfire. , A.headfire
progresses in the same direction as the wi.nd and the fire, front is typically
fast-moving. Headfire burning can be jused under varying fuel loading condi-
tions. With backfire burning, the fire front progresses in the opposite
direction from the wind and is typicalily slow-moving. Consequently, backfire
burning provides a greater residence tjime, thereby allowing for more thorough
combustion. Backfire burning produces1 significantly less particulate emis-
i ' ' •- ' ' '
sions than headfire burning; however, backfire burning cannot be used when
the fuel loading is low because the fi're front cannot be sustained.
9.4.1.2 Control Techniques. .Control techniques can be used either to
reduce the quantity of particulate matjter emitted or minimize the adverse
impact of open burning. These techniques include proper fire and fuel manage-
ment, appropriate burning operations under optimum meteorological conditions,
and alternative residue disposal procedures.
9.4.1.2.1 Fire and fuel management. Backfire burning results in fewer
particulate emissions than headfire burning. For the burning of rice residues,
this reduction has been shown to be approximately 50 percent. However, back-
fire burning is more costly to implement than headfire burning.
The reduction of particulate emissions is also dependent on combustion
t - - '
residence time, air temperature, humidity, wind speed, fuel loading, and fuel
9.4-3
-------
residual moisture content. Of these variables, fuel residual moisture content
has the greatest influence on particulate emissions. Residue with a 10 per-
cent moisture content has been shown to emit approximately one-third as much
particulate matter upon burning as residue with a 25 percent moisture con-
tent.? The residue moisture content is strongly related to solar radiation,
relative humidity, air temperature, wind speed, and residue loading. In order
to minimize moisture content, the day and time of burning should be chosen
carefully. Residue drying time can be reduced by spreading the residues
evenly over the field, which also improves disease control.
9.4.1.2.2 Meteorological considerations. Although not a particulate
reduction technique, meteorological dispersion can be used to reduce the
effect of pollutants generated from agricultural open burning. Burning
should be permitted only when meteorological conditions are conducive to
good dispersion. Maximum dispersion occurs when the inversion base, maximum
mixing height, wind velocity, and wind direction are at specified levels.
The values of these various parameters are site-specific and depend on the
season, time of day, and topography of the area.
9.4.1.2.3 Alternatives to agricultural burning. Alternatives to agri-
cultural open burning fall into the general categories of: mechanical or
chemical residue and crop treatment, and controlled incineration.
The most common method of mechanical treatment involves incorporation of
the waste material directly into the soil. This is accomplished by clipping
or shredding the waste material for subsequent cultivation. Several factors
must be considered before implementing mechanical soil incorporation techni-
ques in areas where residues are now burned. These factors include whether
or not there is: (1) a poor soil conditions situation for cultivation where
the water table is high or where residue occurs late in the year when tempera-
tures are low and biodegradation is slowed, (2) the occurrence of bulky waste
material not capable of incorporation, such as large prunings, (3) a require-
ment for rapid removal of the residue to permit planting of the following
crop, and (4) possible soil nitrogen depletion due to residue decomposition.
Another method of mechanical residue treatment is the physical removal of
the residue to an offsite location. This method is generally expensive
unless the residue can be sold at a price sufficient to cover the cost of
collection and transport.
9.4-4
-------
Chemical methods may be used to control weeds, disease, and pests in
the event that open field burning is npt used. The subsequent effect of the
i
addition of fungicides, herbicides, anjd pesticides to the air and water must
be evaluated. i
Another alternative method of agricultural open burning is the con-
trolled incineration of agricultural residue with a mobile field sanitizer.
Development of such a device was started at Oregon State University in 1970.
Particulate emissions from this incinerator were reduced by 80 to 90 percent
. •- ' i • • •
compared to emissions from open burning. However, the reliability of the
sanitizer and its effectiveness in the control of weeds, pests, and disease,
and in the thermal stimulation of soil! for various crop species is still not
fully known. !
9.4.2 Prescribed Burning |
Prescribed burning is conducted in the United States to: (1) reduce the
hazard of wildfires posed by excessive! fuel accumulations, (2) aid in silvi-
cultural activities, and (3) improve grazing forage and wildlife habitat.
Prescribed burning is the controlled burning of naturally occuring forest
materials and forest slash under prescribed conditions to keep fire intensity
and rate of spread within prescribed limits. This resource management practice
is carried out in the forested regionsiof the Southeast, Pacific Northwest,
and Rocky Mountain States. Particulate emissions from prescribed burning
and wildfires (fires set by natural or|unplanned human activities) are
quantified by region in Figure 9.4-1.8' National estimates of the annual
production of particulate emissions from wildfires are 32 Tg; emissions
from prescribed fires are 390 Gg.2 i
9.4.2.1 Methods of Burning. Prescribed burning techniques may be di-
vided into broadcast burning, pile burning, and understory burning. Broad-
cast burning is an "in place" method of slash disposal and brushland con-
version (burning off the brush and later planting grass or trees). The size
of the area burned, ignition devices, and burning pattern used depend on the
i ' • '
specific environmental conditions and treatment objectives for each site.
The basic ignition patterns utilized to initiate broadcast burning are strip,
ring, center, and area. i
Strip ignition is used to initiate headfires, backfires, strip headfires,
and flank fires. Headfires and backfires are described in Section 9.4.1.
9.4^5
-------
REGIONS AND REGIONAL TOTAL (Gg)c
500
400
§"
21
O
£300
o
"~*
o
o
a.
LU
fe
= 200
CJ
t—
<:
a.
TOO
0
SOUTH
WF -- 960
PF — 202
-
-
_
_
-
/
/
\
'<
'/
'/
/
/
/
/
/i
i
'i
j
/
_
mAU
m
n
/
/
//
X
X
J
'/
X
X
1 2 1 3 1
PACIFIC
WF
PF
1
— 526
— 90
\
2
j-
X
X
',
'
'/
/
/
',
/
'
1
/
/
\
3
^
T\
ROCKY MTN.
WF — 763
PF -- 95
7i ra
i
/
/
t
f>
f
'.
1
/
X
1
/
1
tj
/
/
2
f
X
f
/
/
X
/
/
/
y
/
X
/
J.
,1
NORTH CENT.
WF --
PF --
iiiJ
1
X
X
/
175
0
2
.3
I
3
4
EAST
WF
PF
1
_
/
- 119
- 1.4
H
|3|4.
LEGEND -
SEASONS:
1 -- JAN TO MAR
2 -- APR TO JUNE
3 -- JULY TO SEP
4 — OCT TO DEC
FIRE TYPE:
V/A WILDFIRE (WF)
•PRESCRIBED
FIRE (PF)
PARTICULATE EMISSIONS FROM WILD AND PRESCRIBED FIRES IN ALASKA ARE
587 AND 0 Gg, RESPECTIVELY.
Figure 9,4-1 Annual forest fire particulate production by region & season.8
-------
Strip headfires are parallel headfiresjacross the intended burn area from.
the downwind side toward the windward side. Flank fires are set in strips
parallel to the wind and allowed to spread at an angle to the wind. Ring
ignition is accomplished by firing the(perimeter of the intended broadcast
burn area and allowing the fire to burn towards the center. This type of
ignition is generally used in moderately flat terrain and accomplished with
light fuels. Center ignition is accomplished by igniting the center of the
burn area and allowing the fire to burrt towards the perimeter. Area ignition
is accomplished by checkerboard firing|or spot ignition of the burn block
. i
(area to be burned). i
i
Pile burning is utilized when insufficient fuel is available to support
a broadcast burn or during wet or snowy periods. Unmerchantable material.is
moved by hand or mechanical means (yarded) to partially cut, thinned or clear-
cut areas and concentrated into piles or windrows. PUM (piling unmerchantable
material) is accomplished by hand or tractor and is limited to slopes of 30
to 35 percent where soils are not adversely affected by tractor compaction.
!
YUM (yarding unmerchantable material) is accomplished by cable logging tech-
i
niques on slopes inaccessible to tractprs. Slash material is pulled to the
j
log landing or road, concentrated in pi!les, and later burned.
Understory burning can be used to jreduce undesirable light fuel loads
without damaging desirable residual vegetation. Strip ignition techniques
are commonly used, but flame heights arje kept to 1 to 2 meters so that
scorching of the tree crowns does not q'ccur.
9.4.2.2 Control Techniques. The majority of control techniques
applicable to minimizing particulate emissions from agricultural open burning
are also applicable to forest slash burning. These techniques consist of
alternative burning methods, smoke management programs, and alternative
residue treatment techniques that do not require field burning of forest
fuels. < . .
Alternative burning techniques involve extended burn periods, optimal
burning procedures, and new burning technology. Present broadcast burning
activities are concentrated in the Fall! when fuel conditions are optimal for
ignition and the risk of spot fires in [the surrounding forest is minimal.
Extending the burn period throughout the year provides more flexibility for
optimal smoke dispersion conditions and| reduces emission concentrations
9.4^7
-------
expected during any one period. Implementation of optimal burning procedures
can minimize the potential impact of forestry burning on air quality. These
procedures optimize fuel arrangement and fire ignition for rapid, complete
combustion. Fuel preparation by PUM or YUM techniques prior to burning can
be used to remove larger fuel components which produce intense heat and a
prolonged residual fire. Fuels which are not piled must be sufficiently
concentrated to burn efficiently. Rapid ignition and mop-up techniques can
also reduce particulate emissions. Complete fire mop-up activities started
immediately after the flaming front of the burn has subsided will reduce
residual smoldering.
Smoke management programs are designed to minimize the impact of emis-
sions from forest slash burning on populated areas. These programs restrict
burning activities to periods when specified meteorological conditions occur.
These meteorological parameters, in conjunction with local topographical
characteristics, are critical to the effective dispersion of pollutants.
9.4.2.2.1 Alternatives to prescribed burning. Alternatives to pre-
scribed burning include: (1) on-site and off-site incineration, (2) the
use of mechanical or chemical slash treatments, (3) slash utilization, and
(4) improved logging techniques.
On-site incineration by an air curtain combustor has been used in the
disposal of forest slash.9 This device is specifically designed for wood
waste combustion with little or no emissions if used properly. Slash is
loaded into a long pit and ignited. Complete combustion is promoted by a
blower system which directs a curtain of air diagonally downward across the
burning slash. This configuration supplies sufficient combustion air and
allows for the secondary combustion of gaseous and particulate emissions.
Combustion temperatures are typically 750°C to 1500°C. This burning technique
is not widely utilized at present because of extremely high operating costs.
Offsite incineration is also available to dispose of logging
residues. This technique produces minimal emissions compared to open burning,
but it may not be cost-effective because of high mechanical handling and trans-
portation costs.
Mechanical techniques for treating slash are technically feasible and
versatile. These techniques do not eliminate slash material; instead, they
change the size and shape of the slash components to satisfy silvicultural
9 ,,4-8
-------
and environmental considerations. Me'chanical methods include mastication,
chipping, piling, scarification, and 'burying. Mastication is used to reduce
I ;
materials less than 6 inches in diameter to a mat of wood chips and chunks
by crushing or shredding. Chipping Involves the transformation of both
large and small materials into chips.; Smaller slash residues are chipped by
mobile chippers, while larger materials require PUM or YUM support operations
I
in conjunction with large timber processer-type chippers. Piling of forest
slash is used to break up the continuity of slash concentration to reduce
fire hazard. Ground scarification techniques expose mineral soil for regen-
eration planting and break up the continuity of slash fuels to reduce fire
hazards. Burying slash materials in !on-site pits is feasible for smaller
components. Necessary tractor support limits this technique to use in rela-
tively flat terrain. Burying slash mjay have short- and long-range environ-
mental effects that would reduce its ^feasibility.
Chemical herbicides can be used for temporary control of undesirable
vegetation that would otherwise have to be mechanically treated or burned.
Broad spectrum applications.are used in brushland conversion for preparation
of seedling sites. However, chemical herbicides provide only temporary vege-
tation control, usually less than 2 or 3 years, and they do not reduce the
slash concentration. In addition, the effects of broad applications of her-
bicides on air'and water must be considered in relation to the environmental
impact of burning the forest slash. ;
Increased slash utilization can also reduce the need for additional
treatment by mechanical, chemical, oriopen burning techniques. Forest resi-
dues may be used as a raw material for the production of wood pulp and
timber products (particle board, flakeboard, chemicals) or as a source of
supplemental fuel. All of these uses* however, require transportation- of
the slash to commercial areas, which may be very costly.
Present logging techniques generate considerable forest residues, which
can be substantially reduced by implementing alternative logging systems.
Logging techniques which reduce unusable forest slash include the following:
o Directional felling |
o Multistage logging '
i
o Minimum bucking ,
o Whole tree yarding !
o Optimal material handling techniques
9J4-9
-------
Directional felling helps minimize logging slash by reducing log break-
age and thus increasing use potential. Multistage logging operations can be
used to recover low-grade material that would normally remain as slash. This
technique, which is used preceding or following normal logging operations,
involves the use of light material handling systems designed to recover
small diameter material. Minimizing the bucking of logs into uniform length
classes will also optimize the utility of low-grade materials. Shattered
log ends and extraneous log lengths that are bucked prior to yarding are not
easily handled by standard yarding machines and, therefore, they remain
on-site as slash. Whole tree yarding may be used to eliminate the need for
any bucking. Optimal material handling techniques improve opportunities
for slash utilization. New systems are being developed which allow yarding
preprocessing and transporting of slash materials.
9.4-10
-------
REFERENCES FOR SECTION 9.4
I
j "
1. Geomet, Inc. Impact of Forestry Burning Upon Air Quality, A State-of-
the-Knowledge Characterization in'Washington and Oregon, Final Report.
U.S. Environmental Protection Agency, Region X. Seattle, WA. Publi-
cation No. EPA 910/9-78-052. October 1978.
2. Fox, D. G., J. M. Pierovich, E..wJ Ross, and D. V. Sandberg. Effects
of Fire on the Air Resource. USDA Forest Service. Publication No.
FS-8520. April 1978. '
i
3. Monsanto Research Corp. Source Assessment of Agricultural Open Burning,
State-of-the-Art. U.S. Environmental Protection Agency, Industrial
Environmental Research Laboratory.! Research Triangle Park, NC.
Publication No. EPA 600/2-77-107A.; July 1977.
4. Carroll, J. J., E. F. Darby, J. R.1 Goss, G. E. Miller, and J. F.
Thompson. Minimizing Visible Emissions from Agricultural Burning.
California Air Environment. '4_:1, jp 4-6. 1973.
i •
5. McQueary, M. L., and L. C. Wayne. ' Calculation of Emission Factors for
Agricultural Burning Activities. Pacific Environmental Services,
Inc. Santa Monica, CA. November 1975.
i '
6. Darley, E. F., S. L. German. Air Pollutant Emissions from Burning
Sugar Cane and Pineapple Residues Ifrom Hawaii. Statewide Air Pollution
Research Center. Riverside, CA. July 1975.
1
7. Goss, J. R., and G. E. Miller. Stpdy of Abatement Methods and Meteoro-
logical Conditions for Optimum Dispersion of Particulates from Field
Burning of Rice Straw — Spring Opfen Field Burning Trials. NTIS
Publication No. PB 235 796. June 1973.
i - •
8. Johnson, R. W., C. K. McMahon, andjD. T. Ward. An Update on Particulate
Emissions from Forest Fires. (Presented at 69th Annual Meeting of the
Air Pollution Control Association., Portland, OR. June 1976.) p. 14
9. Geyer, 0. W., and E. A. Rudulph. Minimizing Air Pollution from Open
Burning with an Air Curtain Destructor. (Presented at 63rd Annual
Meeting of Air Pollution Control Association. St. Louis, MO.
June 1970.)
9.4-11
-------
Intentionally Blank
-------
9.5 CHEMICAL PROCESS INDUSTRY i
The chemical process industry includes a wide variety of processes and
products. The processes that have been selected for discussion here are
those involving typical particulate Emission rates exceeding 45 to 90 Mg per
year; these processes include production of charcoal, carbon black, deter-
gents, explosives, sulfuric acid, and chemicals.1
9.5.1 Charcoal Plants i.
Charcoal is produced by the destructive distillation or pyrolysis of wood
or wood waste products in a continuous or batch process, after which the char-
coal can be processed into briquets.?>3 There are about 150 charcoal produc-
ing plants in the United States; over 50 percent of the plants are located in
Missouri, and the rest are scattered throughout the country. In 1975, about
55 percent of the 590 Gg of charcoal!produced was made by the continuous pro-
cess and 45 percent was made by the batch process. The total nationwide
particulate emissions, consisting of'wood chips, char, soot, tar, oils, and
pyroacids, were estimated in 1977 atj79 Gg.3,4
9.5.1.1 Process Description. In the batch process, wood is first placed
in a kiln, which takes one to two days for 165 m3 of wood. Next, a fire is
started and the wood is pyrolyzed for five to eight days. Air flow is then
stopped, and the kiln is sealed and allowed to cool for 10 to 14 days. A
typical yield is 25 percent of the original charge.3 The major emission
points are the off-gas stacks from the kiln. Possible minor emissions result
from raw material and product handling.
A typical continuous process consists of a multiple hearth furnace in
which input material, such as hogged wood, is fed into the top as air is
fed into the bottom. The typical unit produces 2.5 Mg/h with a yield of
25 percent of the original charge. Emissions range from 28 to 406 kg/Mg.3
The major emission points are the off-gases from the furnace.
Figure 9.5-1 shows a charcoal briquetting operation in which the
charcoal is crushed, mixed with a binder, and briquetted. Solids handling
is the major source of particulate emissions.3
9.5.1.2 Emission Characteristics and Applicable Control Technologies.
Because of the cyclic nature of the batch process, controls are difficult to
implement. As indicated in Reference! 4, about one third of the batch type
kilns are controlled by oil- or gas-f]ired afterburner systems. Although
9.5-1
-------
LUMP
CHARCOAL
STORAGE
GROUND
CHARCOAL
STORAGE
CHARCOAL
PROPORTIONING'
FEEDER -H
CONVEYOR
— STARCH
_STORAGE AND,
PROPORTIONING
FEEDER
COOLING ELEVATOR
STORAGE
J
Figure 9.5-1 Charcoal briquetting flow diagram.3
9.5-2
-------
some scrubbers are used to control pafticulates from batch type kilns,
scrubber applications are unusual. The temperature and combustible charac-
teristics of off-gases from continuous processes are sufficient to support
i ,-..". :
combustion. Consequently, the mass o,f combustible particulates from con-
tinuous type processes is reduced whejn the off-gases are oxidized upon
mixing with air. The exact particulate emission reduction effected by after-
burners installed on batch type kilnsi or by the flaring of off-gases from
continuous type processes has not beejn quantified. Because of the high
moisture content of the gas emitted djuring the early stages of drying the
wood charge, supplementary fuel is required to sustain the afterburner effec-
tiveness. As the wood charge becomesj sufficiently dry, the water content of
the off-gas decreases and the supplementary fuel is no longer needed. The
afterburner, as used in Wisconsin and' Minnesota facilities, can reduce emis-
sions 80 to 90 percent. Although these controls are designed primarily for
CO and HC control, particulate matter is also reduced through combustion.3
In 1977, the total annualized cost fo'r the control device was $4 to $43/Mg
of charcoal, which is 7 to 71 percent of the selling price of charcoalJ
Emissions from continuous processes are easier to'control than those
from batch processes, since the flow rates and compositions are nearly
constant. Again, afterburners can be; used, but an additional fan may be
required because of the added pressure drop. Sometimes the gases from the
afterburners are scrubbed for added control. A reduction of 95 percent for
! . • • _ b '
total emissions is reported for this technique, but the reduction of par-
ticulate matter was not singled out.3' The gases emitted from the hearth,
with an average heat content of 29 GJ^/IO^ kg of charcoal produced, have
been used in steam generators and to fire wood and briquet dryers.
Centrifugal collection devices (jcyclones) with collection efficiencies
of 65 percent or fabric filters with Collection efficiencies of 99 percent
have been suggested for control of emissions from briquetting processes.
9.5.2 Carbon Black (Furnace Process)!
Carbon black is currently produced by two main processes: (1) the oil
furnace process, used for over 90 percent of carbon black production; and
(2) the thermal process. Plants rang,e in size from 23 to 174 Gg/year and
9J.5-3
-------
typically operate at 80 percent of capacity; total U.S. capacity is 1.9 Tg/
year. There are 19 plants in Texas and Louisiana, and other plants in
California, Ohio, West Virginia, Kansas, Oklahoma, and Arkansas.5
9.5.2.1 Process Description. Figure 9.5-2 shows a simplified flow
diagram for the furnace process.6 Carbon black is produced by burning a
mixture of hydrocarbon and/or heavy aromatic oil with a limited amount of
air. After the gases are cooled by a water quench, the carbon black is
filtered out for further processing. The emissions sources are the main
process vent, dryer exhaust, pneumatic conveyor exhaust, bagging operations,
and fugitive emissions (Figure 9.5-2).
9.5.2.2 Emission Characteristics And Applicable Control Technologies.
National emissions data published in 1977 for carbon black production list
particulate emissions at 7.5 Mg/year.7 Particulate size distribution data
were not given. However, the particulate type emitted is the product carbon
black; the mean particulate diameter for the product varies from 0.01 to
0.4 micrometers.8
Almost all plants now use a bag filter in the product recovery system to
achieve up to 99.95 percent recovery of carbon black, resulting in emissions
of 0.07 to 0.30 g/Nm3 (1 to less than 5 g/kg product). The bags are usually
constructed of a graphite-silicone, film-coated fiberglass material, which
must be kept below 230°C; the bags have an average life of 12 to 18 months
and are usually cleaned by reversing the gas flow.^
The gases from the main process vent may be fed from the baghouse to an
incinerator and/or a CO boiler to remove CO. This procedure also removes
additional particulate matter and at one plant results in average emissions
of particulate matter of 0.04 g/Nm3.9
The dryer exhaust may also be equipped with a baghouse but does not have
the economic incentive of product recovery.6 Fiberglass bags are normally
used because the operating temperature generally exceeds 200°C. One manufac-
turer uses a water scrubbing system to control dryer emissions.6 The scrub-
ber reduces the particulate emissions to 0.7 g/kg product. The scrubber is
a combined venturi and tangential entry vertical cylinder scrubber. Some
of the scrubber slurry is recycled to the scrubber, and the rest is used
as quenching water for the reactor off-gases. The scrubber uses about
300 L water/1000 Nm3 gas and has a pressure drop of 3.7 kPa.
9.5-4
-------
vo
in
QUENCH ^p-1
L STOMtt Q^CH WATER WATER " j
___~ X— 1 > .. . _
PREHEATER J BLOWER • OIL 1— J
(OPTIONAL) S PREHEATER QUEHC
n M W (OPTIONAL) TOME!
[--] — >T /i^x > L_> -^ *<
"~^™J 1 SfcACT.QR ~ ~
A fttn
FUEL PREHEATER
CYCLONE r^l< PNEUMATIC CONVEYOR
AND/OR rH DRI£R
f^°??f^D.V7 PURSE GAS
_ BAG FILTER ...Nr. -T[ — ^ - - —
<;URGE I I.
TANK J [BAG FILTER! nDTP» TMnrnFrr
^Y EXHA^?l,iiv>i VEN^GAS
- U?T MIXER >4. T. ,
(PELLEHZER 1. L — -!---_ J
CONVFYnR _A_ r
* — _JT
CS: I ATS
H
?
.^
>^
\
.
>.
__
4AIN
'ROCESS
ffKT
A
1
Q MICRO-PULVERIZER
' - ••• fiTD
s.
., ^f , ,,. , " ^ '-.....
STORAGE STORAGE ,
\ / \ /"
\.,^S \— /
w f
i, 1 ^ 1
i-.spsa.j p==r?n
FUEL GAS
BUCKET ELEVATOR BAG PACKAGING HOPPER CAR
*If bag filter is employed, filtered gas is vented to atmosphere.
Figure 9.5-2 Carbon black process.6
-------
The pneumatic conveyor exhaust may also be equipped with a baghouse.
However, because of the lower operating temperatures (essentially ambient),
wool, cotton, or or!on bags can be used.5 The filtered air is recycled to
the compressor, resulting in a closed-loop air system.5 Cyclones may be
used in place of a fabric filter depending on particulate loading and size
distribution.
Other particulate sources are bagging operations and fugitive emissions
(from torn bags and leaks). Emissions from these sources can be controlled
with improved operating methods, preventive maintenance, and a vacuum clean-
ing system. The vacuum cleaning system, used to clean up spills, can also
be used to vacuum-package the bags. A cyclone and filter can be used in the
exhaust of the vacuumed system.
9.5.3 Detergent Manufacturing Plants
The main source of particulate emissions from a detergent plant is the
spray drying operation in which the detergent slurry is converted to a pow-
der. Some particulate matter is also emitted in the process used to make
the slurry and in the post-dryer operation where the powder is cooled,
blended, and packagedJ°
9.5.3.1 Source Description and Applicable Control Technologies. Figure
9.5-3 shows a flow diagram for a spray-dryer. The detergent surfactant is
produced by sulfonation of a detergent alkylate, from which there are pos-
sible acid mist (S02, S03, H2S04) emissions.10
Methods suggested for controlling emissions from the spray-dryer
include a wet scrubber and demister, two wet scrubbers in series, a cyclone
and wet scrubber, a cyclone and scrubber-ESP unit, or a cyclone and fabric
filter.10,11,12,13 in all cases the collected slurries and powders from the
primary collectors may be recycled back to the process. Typical particle
loadings from the spray dryer are around 0.2 g/Nm3, with the temperature of
the stream around 100°C.
Predicted collection efficiencies for a wet scrubber used alone are
98.0 percent at a pressure drop of 2.5 kPa across the scrubber and
99.8 percent at a pressure drop of 10..0 kPa. The scrubber operates with a
recirculation slurry of 40 to 45 wt percent solids, with the slurry in the
recycle tank being maintained at a temperature of approximately 40°cJ1
9 ,,5-6
-------
SYNTHETIC DETERGENT PRODUCT MANUFACTURING EQUIPMENT 1
' ' - -' .; 1
uJif 2?
E"S
Ul
|g
h-
1- Z ,-.
5£ '
o o >- =
o: t a: 0
.UJ U. CC. »
o
ox^ S
z
111 5
O —IUJ t-
TO FAN AND
AIR POLLUTION
CONTROL,
ID
cn
L i
ALTERNATE PROCESS
\ A !!
\ / \ !t:
CRUTCHER
I
SURGE TANK
HIGH • PRESSURE I
PUMP
-SPRAYS
CYCLONE OR
GRAVITY SEPARATOR
FINES AND
OVERSIZE
,TO RECYCLE
TEMPERING
AIR
FURNACE
-ADDITIVE EQUIPMENT
TO
STORAGE
AND
PACKAGING
N
STARTUP
BY PASS STACK
Figure 9.5-3
Detergent spray-drying with tower equipped for either
cocurrent or countercurrent opeV-ation - dampers in
countercurrentmode of operation.10
-------
A second scrubber or demister may be used to capture emissions from the first
scrubber. The percent solids in this scrubber is considerably less than in
the first scrubber.
For a dryer exhaust cleaning system consisting of a cyclone followed
by a fabric filter, predicted operating conditions and equipment type are
available. The fabric filter is the pulse jet type with a maximum air-to-
cloth ratio of 6 to 1 and a pressure drop of 1.5 kPa. The cyclone is
designed to have a collection efficiency of 85.2 percent and the fabric
filter 95.9 percent for a total collection efficiency of 99.4 percent.11
A cyclone followed by a combined scrubber-ESP device is used by one
large detergent manufacturer. Emission test data indicated a cyclone
control efficiency of 98.2% and a scrubber-ESP control efficiency of 53.7%
for a total collection efficiency of 99.2%.14
In some plants, the dried detergent beads are transferred from the
spray dryer to the dry product storage via a pneumatic conveying system.
The pneumatic air also acts to cool the detergent beads. Most systems use
a cyclone to separate the detergent beads from the pneumatic air. A
typical particulate loading downstream of the cyclone is listed as
0.085 g/Nm3 with the gas at 40°C. If further treatment of the exhaust
gas is desired, a fabric filter may be used.11 This pulse jet fabric
filter is specified at a maximum air-to-cloth ratio of 6 to 1 and a pres-
sure drop of 1.5 kPa; collection efficiency is predicted to be 99.8 percent.
The collected fines can be returned to the product storage tank. This
control technique has little environmental impact on water and solid waste
because the collected particulate matter can be recycled back to the process.
9.5.4 Explosives Industry
There are two main types of explosives: (1) detonating (high explo-
sives) and (2) deflagrating (low explosives). High explosives explode with
great violence. Low explosives do not explode but burn rapidly. Examples
of these two explosives are as follows:2
1. Detonating:
a. Primary or initiating (detonators): lead azide, mercury
fulminate, lead styphnate (lead trinitroresorcinate),
diazonitrophenol, nitromannite.
9.5-8
-------
b. Secondary: TNT-AN (trinitrotoluene-ammonium nitrate),
TetylP (2,4,6-trinitrop'henyl-methylnitramine), PETN
(pentaerythritol tetranitrate), RDX (sym-trimethylene
trinitramine), TNT (trinitrotoluene), ammonium picrate,
picric acid, DNT (dinitrotoluene), and EDNA
(ethylenedinitramine). ; ,
2. Deflagrating: Smokeless powder, black powder, NG (nitroglycerin),
DNT, nitrocotton, ammonium 'nitrate fuel oil blasting compounds.
. There is little available liter'ature that identifies particulate matter
emissions during the manufacture of explosives, except for the production of
TNT. Some of the processes emit suljfuric acid mists which can be controlled
by the methods discussed under sulfuric acid (Section 9.5-6). Most of the
processes involve some type of nitration, and sulfuric acid is usually used
as part of the process and/or to help recover nitric acid, HNO^.^ Ammonium
nitrate production, which can be a source of particulate emissions, is dis-
cussed in Section 9,6.
TNT Production—Figure 9.5-4 shows a flow diagram for the production
of TNT by the batch process.16 TNT is produced from toluene and nitric
acid. In addition to the batch process, TNT can be produced by a continuous
process.I? A waste stream called reel water is generated during production
via either the batch or continuous process. The waste is disposed of by
incineration, which results in emissions of particulate matter. Wet scrub-
bers have been used to reduce these emissions J5 Reference 18 discusses
red water incineration but not particulate control.
9.5.5 Thermal Process Phosphoric Acid Manufacturing
Essentially all the phosphoric acid (^04) produced in the U.S. is
manufactured by either the wet process or the thermal process. Wet process
i ' •
acid is manufactured by treating phosphate rock with sulfuric acid; the
technique is used primarily in fertilizer manufacturing. Thermal process
phosphoric acid is produced from elemental phosphorus; the acid is used
where a high-quality product is required such as for food-grade acid and
industrial phosphates. i
9.5.5.1 Process Description. In the thermal process, phosphoric acid
is manufactured using elemental phosphorus, air, and water. Figure 9.5-5
presents a flow diagram for the thermal process and subsequent acid purifi-
cation.^ | .
9.5-9
-------
or
o
ELECTROSTATIC
PRECIPITATOR
SULFURICACID
CONCENTRATOR
60V.HN03 OLEUM
L,
TRI-ACID
BI-HOUSE
FUEL
VENT
BI-OIL
FUMES , FUMES
OXIDATION
CHAMBER
1
r
BUBBLE CAP
TOWEH
'
VENT
FURNACE
RED
HOT GAS
WASH
HOUSE
WATER
N»2C03
SOLUTION PLANT
t
PURIFIED TNT | OLE"M EXHAUST GAS
EVAPORATORS
1
WASTE LIQUOR
ROTARY
KILNS
EXHAUST
FUEL
Mi2S04
Figure 9.5-4 ThIT production by batch process.16
-------
Stack
Effluent
(Air + H3P04 Mist)
Acid Purification
Stack Effluent
(Air
Blower
Hydrogen Sulfide,
Sod i urn Hydros ul fi de,
or Sodium Sulfide
Phosphorus
Combustion
Chamber
Product
JAcid to
IStorage
Burning and Hydration
acid to storage
ACID PURIFICATION
(Used in the Manufacture of Acid
for food and special uses)
Figure 9.5-5 Flow diagram for typical thermal-process phosphoric acid
plant.19
-------
In the combustion chamber, phosphorus combines with oxygen to form
phosphorus pentoxide by the following reaction:
P4 + 502 —*- P40io
The phosphorus pentoxide passes to the hydrator where it reacts with
water and/or an aqueous solution of weak phosphoric acid. The hydration
reaction is as follows:
P40io + 6H20 —> 4H3P04
The product acid is drained from the bottom of the hydrator and pumped to
storage.
The 1978 production of thermal process phosphoric acid was about
627,000 Mg as 100% P205.20 Average plant production is about 27,250 Mg/yr.20
Typical thermal process plant parameters are shown in Table 9.5-1.
9.5.5.2 Emission Characteristics and Applicable Control Technologies.
The primary pollutant emitted from the thermal process is particulate matter
in the form of phosphoric acid mist. Particle size of the acid mist is
reported to range from 0.4 to 2.6 micrometers with a mass median diameter of
1.6 micrometers.19 Several control devices can be applied to agglomerate
the mist particles and capture the liquid particulates. These devices in-
clude packed towers, electrostatic precipitators, scrubbers, fiber mist
eliminators, and wire mesh contactors. Operating parameters and performance
of selected control devices installed on thermal process phosphoric acid
plants are presented in Table 9.5-2.19
9.5.6 Sulfuric Acid
About 150 sulfuric acid plants operate in the U.S. with a current
annual production capacity of 42 Tg.7 Of the 41 Tg manufactured in 1978,
99.9 percent was produced by the contact, process.21 Sixty-eight percent of
this sulfuric acid was produced from elemental sulfur; 4.5 percent from
iron pyrites; 9 percent from smelter tail-gas; and 18.5 percent from hydro-
gen sulfide, spent alkylation acid, and acid sludge from refineries.22
9.5-12
-------
TABLE 9.5-1. TYPICAL THERMAL PROCESS PHOSPHORIC ACID STACK PARAMETERS20
Stack height j 23 m (75 ft)
Stack diameter j 1.2 m (4 ft)
Stack gas temperature ; 60°C (140°F)
I
Stack flow rate 1
range: 900 to 4100 m3/Mg product
(35,000 to 160,000 scf/ton product)
1.6 to 14.3 m3/s !
(3400 to 30,200 scfn))
average: 7.1 m3/s (15,000 acfm)
915-13
-------
Table 9.5-2. EMISSION AND OPERATING DATA FOR THERMAL PROCESS PHOSPHORIC ACID PLANTS*9
Packed
Control device tower
Phosphorus feed 2600
rate - Kg/hr
Gas flow rate 3.27
m3/sec
Stack gas temp 79
= °C
<£> Collector pressure 0.37
^ drop kPa
i
£ Superficial gas
velocity in 0.61
collector m/sec
Acid mist to 23
collector g/Nm3
Acid mist from 0.16
collector g/Nm3
Plume opacity medium
Collector 99.4
efficiency %
Venturi
scrubber
2721
3.27
61
12.9
61.33
1.8b
211.6
0.24
light
99.95
2404
3.11
94
8.2
100a
22.1
0.53
100%
97.5
Glass fiber
mist eliminator
1829 5035
2.9 9.17
82 83
f
5.5 5.3
3.96 0.12
8.9 54.5
0.32 0.009
medium 0%
96 . 99.9
W1 re-mesh
contactor
3629
5.78
78
9.7
8.23
94.3
0,02
100%
99.9
a Venturi.
" Separator.
-------
9.5.6.1 Source Description. Figure 9.5-6 is a flow sheet of a process
for producing sulfuric acid from burnjng elemental sulfur.23 The elemental
sulfur is burned to S02, the S02 is catalytically oxidized to $63, and then
the $03 is absorbed in 98 percent H2S04 to produce sulfuric acid. Water
is mixed with the strong acid to maintain the desired acid concentration.
Some plants also produce oleum, which'is a solution of 863 and sulfuric acid.
9.5.6.2 Emission Characteristics And Applicable Control Technologies.
i •
The uncontrolled acid mist emissions from the absorption tower for typical
plants are 70 to 700 mg/Nm3 or 0.2 to; 2 g/kg acid from a plant producing
no oleum and 175 to 1750 mg/Nm3 or 0.5 to 5 g/kg acid from a plant producing
oleum.23 Emission factors given in Reference 24 for five different types of
i
sulfur-bearing process feeds are within these ranges.
Table 9.5-3 shows a particle size distribution for uncontrolled acid
mist emissions. It can be seen that oleum production results in mists
with smaller particle sizes, and that stronger oleums emit smaller particle
sizes than weaker oleums. . . • j
Standards of performance have been set for S02 and acid mist from
new and modified control process sulfuric acid and oleum facilities that
burn elemental sulfur, alkylation acid, hydrogen sulfide, organic sulfides,
or acid sludge. j
The standard does not apply to acid plants used as S02 control systems,
to chamber process plants, to acid concentrators, or to oleum storage and
transfer facilities. ;
Standards of performance for acid mist limit the discharge into the
atmosphere from any affected facility of any gases which:
(1) Contain acid mist, expressed as H2S04, in excess of 0.075 kg
per metric ton of acid produced (0.15 Ib per ton), the production
being expressed as 100 percent H2S04.
(2) Exhibit 10 percent opacity pr greater.
The U.S. EPA guidelines for States for the control of sulfuric acid
mist emissions from existing sulfuric acid plants limit these emissions to
0.25g acid mist per kg (0.50 Ib/ton) of 100 percent H2S04 produced.23
The best available control technique for sulfuric acid mist is the
fiber mist eliminator. Three types of fiber mist eliminators are currently
in use: (1) vertical tubes, (2) vertical panels, and (3) horizontal dual
9i5-15
-------
HEAT 1 EXCH. .TO
98% ACID PRIMARY HEAT CONVERTER ECONOMIZER SECONDARY 98% ACID
ABSORBER EXCHANGER ABSORBER
Figure 9.5-6 Contact-process sulfuric acid plant burning elemental sulfur.23
9.5-16
-------
Table 9.5-3. PARTICLE SIZE DISTRIBUTIONS IN SELECTED SULFURIC ACID
PLANT ABSORBER EFFLUENTS.23
Cumulative weight percent smaller than
I stated size
Particle diameter,
micrometers
Acid production
only \
20% oleum
production
32% oleum
production
0.2
0.4
0.6
0.8
1.0
1.5
2.0
1
7
12
21
40
0.4
2.0
4.8
8.0
11.6
48.0
84.5
3.6
16.0
30.0
42.0
53.0
86.5
97.0
,5-17
-------
pads. Vertical tube eliminators require superficial gas velocities of 0.1
to 0.2m/s, operate at pressure drops of 1.3 to 3.7 kPa, and have a large
turn-down ratio. Collection efficiencies of 99.3 percent are reported for
these collectors. For vertical tube eliminators installed on single absorp-
tion plants, emissions as low as 7.1 mg/Nm3 have been reported.23 Vertical
panels are designed for superficial gas velocities of 2.0 to 2.5 m/s and
operate at pressure drops of 2.0 kPa. A collection efficiency of 90 percent
is reported for vertical panels. For vertical panels installed on single
absorption plants, emissions as low as 19 mg/Nm^ have been reported. Hori-
zontal dual pads have a superficial gas velocity of about 160 m/min and pres-
sure drops of 2.2 kPa. Collection efficiencies of 90 percent are reported
for horizontal dual pads. Emissions as low as 7.1 mg/Nm^ have been reported
for the use of these pads on a single absorption plant.23
In all cases, since the collected acid mist can be recycled to the
absorption column, there are no additional environmental effects.
9.5.7 Phthalic Anhydride
Phthalic anhydride is manufactured by the catalytic vapor phase oxida-
tion of o-xylene or naphthalene at 10 plants in the continental U.S. and one
plant (type not identified) in Puerto Rico. Seven plants in the United States
use o-xylene and three use naphthalene. Plants located in New Jersey, Penn-
sylvania, Illinois, California, Texas, and Louisiana have a total capacity of
500 Gr/yr. The total emissions of particulate matter from o-xylene processes
are 909 Mg; from naphthalene, 244 Mg per year. Included in particulate matter
emissions are organic vapors which, upon cooling, may form particulate matter
downstream from the pi ant.25
9.5.7.1 Source Description. Flow sheets for the manufacture of phtha-
lic anhydride from o-xylene or naphthalene are shown in Figures 9.5-7 and
9.5-8. The o-xylene process uses a fixed bed reactor while the naphthalene
process uses a fluid bed. The major particulate emissions points for both
processes are the main process incinerator and the secondary incinerator.
Reference 25 lists average controlled emission factors for particulate matter
as 1.3 g/kg product. Flaker and bagging operations give uncontrolled emis-
sions of 0.20 g/kg product which are usually controlled to 0.002 g/kg product.
Reference 25 does not list any particulate size distribution.
9.5.7.2 Application Control Technologies. Ore control technique for the
9.5-18
-------
ATMOSPHERIC EMISSIONS
PROCESS ©
AIR JT
J^ ST£AM"* (r\
© *F WATER-*- G ®
i ®
-. -Crufc ^r"^
. * ?
H H &
t
LH - CP-
_- © L
*iri &r '
^ ^PLEKSH.^1
**" L— ^WATFR
STEAM \ /
PROCESS rff>£ T
jb- &^^_CJ-J
•Y1-. ® '
— • OIL
to, y 4
' VACUUM TO ®
SCRUBBER OR li
INCINERATOR,!-1— v
Oil
-H « i I
OIL^; _,-ii
— -*M N
1 _ .„ , ^ .. ,_,.— w.S-'.J^i trUllftBrB UTMT Mt
i T
i f~^\^
1
J II ^-»- FUGITIVE EMISSIONS
T) ' — H * (3) ®
^*- -*= FUEL •- CATALYST STORAGE HOPPER VENTS
IT ® ' '
,. .. *K ^ CRUDE PHTHAIIC ANHYDRIDE
I — "~ STORAGE TANK VENT GAS
•
VACUUM TO
SCRUBBER OR
INCINERATOR
®jj® .
i fc/D(V.
*^y^
^ 1
3^~O)
I.YPf- @ .
© ^ REFINED PHTHALIC ANHYDRIDE
i STORAGE TANK VENT GAS
PHTHALIC ANHYDRIDE
® P r*" TRANSPORT LOADING
' *• I-ACILIIY
' 1
'" *• TO TANK CARS
' /^ RESIDUE ^
-------
cn
ro
o
'
r^-
A
STEAM
^LJLJ ^
B
|_J STEAM
-V
F
!1
®jj®
^ G .*•
H
fe":jTDM
^WATFJ)
G
H '
<
i
*.( i
i
•
T£u
^
i
JL
K
' »
i
®r
!
®
•4-WATER
®.
lOCESsC&X.
AIR ^y~
PROCESS
(I)
OIL
TO SCRUBBER K -*-]
TO SECONDARY 5
SCRUBBER -*-H,
OR INCINERATOR -"-v
@| VA
VACUUW
@ WATER
0
J
BASF PROCESS FOR THE MANUFACTURE OF
PHTHAUC ANHYDRIDE FROM 0-XYLENE
A. O-XYIENE STORAGE
B. SO 2 STORAGE
C. 0-XYUNE PREHEATER
D COMPRESSOR
E. AIR PREHEATER
F. REACTOR IFIXEO BED)
L. INCINERATOR
M. HEATER
N. CRUDE PRETREATMENT TANK
0. PRECOOLER
P. STRIPPER
Q. EVAPORATOR
G. MOLTEN-SALT HEAT EXCHANGER R. EVAPORATOR
H. WASTE HEAT BOILER 5. RECTIFIER
1. SVVITC" CONDENSERS T. PHTHAUC ANHYDRIDE STORAGE
J. CRUDE PRODUCT STORAGE U. FLAKER
K. SCRUBBER V. BAGGING MACHINE
-------
main process vent is a scrubber plus:a thermal incinerator. In this con-
! •
trol method, the off-gases pass throijgh a wet scrubber with the scrubber
solution being fed to an incinerator.' This technique is used primarily for
hydrocarbon emissions which could condense as the gas cools, causing particu-
late emissions downstream. Some particulate matter is emitted from the
scrubber vent and the incinerator due to impurities in the make-up water.
i
Incinerators can also be used on the ^off-gas stream. The advantage of using
a wet scrubber for the off-gas is that the size of the incinerator required
is reduced. However, CO emissions, w|hich are emitted from the scrubber over-
head, are not controlled.26 Heat recbvery systems can be added to the incin-
erator to save energy costs. '
I - ' -,-..•-- - . --;-..
Based on actual measurements, emissions from a 59 Gg/year plant equipped
- i ' '•
with a wet scrubber and scrubber solution incinerator for control of hydro-
carbon emissions are expected to be as follows:
i
o From the scrubber vent, 0.076 kg/min of particulate matter
from the make-up water impurities and 0.30 kg/min of con-
densed hydrocarbons. The gas flow rate from the scrubber
overhead is 3460 Nm3/min at 38PC.
o From the incinerator stack, 0.05 kg/min of particulate
matter from the make-up water inpurities and 0.06 kg/min
of condensed hydrocarbons. Thfe gas flow rate from the
incinerator is 275 Nm^/min at 927°C.26
!
For controlling emissions from the product storage vents, a condenser can
be placed on the vent to collect the phthalic anhydride and return it to the
system. , Again, this is a hydrocarbon(control technique that could prevent
particulate emissions from forming downstream.26
9.5.8 Hydrogen Fluoride (Hydrofluoric Acid)
At present, there are 11 hydrofluoric acid plants in the U.S.—three in
Louisiana, three in Texas, and one each in California, New Jersey, Ohio, West
Virginia, and Kentucky. In 1977, these plants produced 240 Gg of acid.27
9.5.8.1 Source Description. A flow diagram for the manufacture of
hydrofluoric acid is shown in Figure 9.5-9. Hydrofluoric acid is produced
by reacting fluorspar with sulfuric acid.27 The sources of particulate emis-
sions are spar (fluorspar) drying and ;the fugitive emissions from handling
and storing the spar. There are also |fluor.ide emissions from the process
9J5-21
-------
cn
ro
ro
HF. S02( SIFV C02
8
ti
8
O
u
S.
p-T
Oleum jr |
1
Makeup
Sulfuric acid 1
_««_.«»« — b-
•*
§
M
X
S
u?
tf> <
V*
%
O
t*
X
u."
Kiln
Crude HF
Calcium sulfate
H2S04.25% HF
Intermediate storage
SiF4. S02. C02, HF
H2S04
99.98% HF
i. SOj
1 '
K
•§
a
&
r\
X
t-t
\
K
a
?
K
O
I
c
04
-*- Vent
• Water
30-35% H2SiF6
Figure 9.5-9 Schematic flow diagram for the manufacture of hydrogen
floride.27
-------
and possible combustion related emissjions from the kiln.
9.5.8.2 Applicable Control Techhologies. Most plants use a cyclone or
baghouse to control particulate emissions from the dryer.27 The baghouse,
usually cleaned by pulsed air, gives 99 percent collection efficiency. Based
on plant measurements, particulate emissions are 1.8 to 2.4 kg/h for gas flow
rates of 140 to 370 m3/min. i
For emissions which occur during storage, the recommended control is
a silo with a baghouse on the vent.27i At one plant, the particulate emission
rate from the silo baghouse was 9 kg/h for a gas flow rate of 45 m3/min. For
both sources, dryer and silo, the collected dust can probably be recycled.
9.5.9 Boron Compounds ;
Borax (^26407.10^0) is produced from mines located near Boron, Cali-
fornia, and from the brines of Searles Lake, also in California.28 The borax
could be used as is, converted to boric acid, or used as a source of boron
for other compounds. On a 6203 basis; 650 Gg of boron compounds were pro-
duced in 1975.29 j
9.5.9.1 Source Description. In producing borax from mining operations,
several solutions and thickening steps are involved.30 If anhydrous borax
is needed, the borax is fed to a bora>j; fusing furnace. Figure 9.5-10 shows
a flow diagram for producing borax from lake brines.2 Figure 9.5-11 shows
a flow diagram for producing boric acid from borax and sulfuric acid.28
Boric acid and boric oxide (an anhydrous form of boric acid) are used to
produce boron carbide, metal borides,iboron alloys, and other boron com-
pounds.30 !
!
The various processing steps that emit particulate matter and the equip-
ment used to control emissions are summarized in Table 9.5-4. The particu-
late size varies from 1 to 45 micrometers.
9.5.9.2 Emission Characteristics anci Applicable Control Technologies.
References 31 and 32 describe scrubber systems for controlling emissions
from borax fusing furnaces. As the borax dries, the water vapor pressure
inside the crystal ruptures the crystal, producing small particles. Refer-
ences 29 and 32 describe an impact type scrubber and a venturi scrubber that
have been used, respectively, to control emissions from borax fusing furnaces.
An impact type scrubber which cari be added to an existing plant has been
developed for particulate control. In the first section of the scrubber, the
9.:5-23
-------
Row loke brine-fr borox mother liquors
I
Wormed by condensing vapors in vacuum crystallizers
I
Evaporated in triple-effect evaporators
(IMa2C03-2IMo2S04)-**NaCl+Li2NaP04 1 KCH-Na2B407 both in hot solution
1
Halite: No Cl, coarse crystals
Burkeite (Na2COj-2NOjS04) and Li2r
fine crystals
Separated by countercurrent washing
washed an
Overflow filtered and washed with la
» Brine
Burkeite, dissolved in HjO, cooled,
and LijNoP04 froth floated
Ja P04:
NoCl,
toy '
Burkcilf (to start 4)
Cooled to 30*C and filtered
\ *-Some NaCl
Cooled to 5*C and filtered
\ >• Brine
No2C03-IOHjO
Recrystallized hot
No2C03- H20
Calcined
1
Sodo Ash 58% Na20
1
NajSCvlOHjO
1
NoCl odded to lov
transition to 17"C
to Na2S04
Filler
*-NaCl mo
1 liquor
Na2S04 refined
Refined sent cake
Dried
1
Mother liquor. Quick vacuum
cooling to 38*C
i
KCl centnfuged,
dried and shippe
i
cooled to 24'C, seeded,
and crystallized
Crude borax
Recrystallized
1
Refined borox
Li2NaPO«(20%Li02)
Dried
1
Acidified with
cone, HjS04
Li2S04
Treated with
No2C03 solution
1
LijCOj, centnfuged,
dried and shipped
Figure 9.5-10 Procedures for recovering borax -"Trona Processes.2
9.5-24
-------
Water
Sulfuric or end
acid liquors
Granular
borax
I
or
solution
Acidulator
Crystallizer
Filter
Sodium
sulfate
solution
U-
1
Drypr UMH
i mil i '
Water— ^
Dissolver
-Boric acid
(technical)
BoriCj add End liquor
(USP)
Figure 9.5-11 Borac acijd from borax by acidulation.28
9;5-25
-------
Table 9.5-4. TYPICAL APPLICATIONS AIR POLLUTION CONTROL EQUIPMENT30
Procedure
Control equipment
Borax refining
Anhydrous borax
Borax calcining and melting
Boric acid drying
Borax drying
Borax dehydration
Borax packing
Dehydrated borax packing
Dehydrated borax grinding
Boric oxide production
Boron trichloride production
Boron tribromide production
Cyclone, scrubber, and baghouse
Cyclone, scrubber, and baghouse
Cyclone, baghouse, and
electrostatic precipitator
electrostatic precipitator
Cyclone and scrubber
Baghouse
Cyclone and electrostatic
precipitator
Cyclone and baghouse
Baghouse
Baghouse ;
Scrubber and baghouse
Scrubber
Scrubber
9.5-26
-------
gas is quenched with water, helping to form large particles. Next, the gases
go through a low velocity impingement section where the larger particles are
. ! '
removed. Then they enter a high velocity impingement section and leave
through a mist eliminator. A 97 percent collection efficiency can be achieved
at a pressure drop of about 5.5 kPa.! This control effectiveness results in
an emission level of 0.2 g/Nm3 from the furnace.31
A venturi scrubber has been developed with a 97.5 percent collection effi-
ciency. For a gas flow rate of 20 m^/s at 80°C, 33 L/s of scrubbing liquor is
used and the pressure drop is 10.8 kPa. This control method reduces emissions
~ i
to 0.02 to 0.04 g/Nm-3 with a mass median aerodynamic particle diameter of 0.2
to 1.1 micrometer. i
I i ,.•_ _ -
Table 9.5-5 represents combined emission tests of a borax-fusing fur-
nace controlled with a venturi scrubber. The design pressure drop of the
scrubber was 11 kPa at 32 NM-Vsec. The furnace was operating between 5.6 and
8.4 Tons/hr, which is 50 to 75% of capacity. These computer modeled data are
from the Fine Particulate Emissions Information System (FPEIS) repository.
9.5.10 Pest. 1 cide Manufacturirig ! •
In 1974, 643 Gg of synthetic organic pesticides were produced. Inorganic
pesticides account for an additional 130 Gg. Producers are scattered through-
out the U.S. Formulator plants, wherje the active pesticides are mixed with
inert materials for use by the consumer, are located in almost every State.
Tables 9.5-6 and 9.5-7 identify some jof the pesticides produced and their
uses.33,34,35 j
i .•
9.5.10.1 Source Description. The emissions from the various processes
are not well quantified. Table 9.5-8, lists some emissions for the production
of certain pesticides.34 No data are1 given for formulator plants, but since
a large amount of pesticides are used as dusts, there may be emissions from
these plants also. i
i
Table 9.5-9 lists some control techniques that have been applied to
particulate and gas phase pollutants.34 As can be seen, baghouses, water
scrubbers, venturi scrubbers, mist eliiminators, and packed columns have been
used with a claimed reduction in particulate emissions greater than 90 per-
cent. Actual design data are not available.
9.5.11 Sodium Carbonate (Natural Process)
i
Sodium carbonate, or soda ash, is produced from natural deposits that
9.5-27
-------
Table 9.5-5. SIZE SPECIFIC EHISSIOHS FROM A BORAX FUSING FURNACE
Mass concentration, mg/DNCM
(mass percent less than stated size)
Total
15.3
12.9 pm 10.1 urn
7.28/tm
5 jim
2.
1.01
Uncontrolled 784 600(76.5) 598(76.3) 596(76.1) 593(75.7) 585(74.5) 531(67.7) 248(31.6)
Controlled 24.3 21.5(88.3) 21.1(86.8) 20.6(84.7) 204(82.7) 19.8(81,2) 19(78.3) 17.2(70.7)
Efficiency 96.8 - 96.4 96.5 96.5 96.6 96.6 96.4 93.1
en
ro
CO
-------
Table 9.5-6. U.S. PRODUCTION OF SYNTHETIC ORGANIC PESTICIDES,
BY (USAGE) CATEGORY, IN 197434
Pesticides usage categories
1974 production,9
10 Mg
Fungicides I
Pentachlorophenol and sodium salts;
Naphtenic acid, copper salt !
Other cyclic fungicides !
Dithiocarbamic acid salts j
Other acyclic fungicides '..\
j
Total fungicides ;
Herbicides and plant hormones ;
- . • j
Maleic hydrazide !
2,4-D acid,!3 dimethylamine salt j
Other cyclic compounds '
All acyclic compounds i
i
Total herbicides and plant hormones
Insecticides, rodenticides, soil conditioners
and fumigants ;
Aldrin-toxaphene group j
Methyl parathion ; |
Other cyclic organophosphorus insecticides
Methoxychlor ;
Other cyclic insecticides and rodenticides
Methyl bromide |
Acyclic organophosphorus insecticides
Chloropicrin i
Other acyclic insecticides, rodenticides, soil
conditioners, and fumigants
Total
Total synthetic organic pesticide
production, 1974
23.8
0.9
31.8
16.1
1.3
T378
2.6
6.6
212.0
52.8
274.0
64.3
23.3
25.6
1.5
72.8
13.8
35.7
2.2
55.8
29479"
642.7
aData may not add to totals due to
^2,4-Dichlorophenoxyacetic acid.
independent rounding.
9.5-29
-------
Table 9.5-7. PRINCIPAL INORGANIC PESTICIDE FORMULATIONS^
Insecticides
Herbicides
Fungicides
Calcium arsenate
Calcium cyanide
Lead arsenate
Sodium cyanide
Sodium fluoride
Ammonium sulfamate
Arsenic acid
Borates
Magnesium chlorate
Potassium chlorate
Sodium arsenite
Sodium chlorate
Cadmium chloride
Copper carbonate
Copper chloride
Copper oxide
Copper oxychloride sulfate
Copper sulfate
Mercuric chloride
Mercurous chloride
Sodium polysulfide
Sulfur
Zinc oxide
Zinc suTfate
9.5-30
-------
Table 9.5-8. SUMMARY O'F PRINCIPAL AIR EMISSIONS25
Pesticide manufactured
Type of pollutant
Methyl parathion
MSMA
Trifluralin
Pentachlorophenol
Captan
DDT
Toxaphene
Sulfur dioxide (gas)
Arsenic trioxide
(particulate)
Nitrate (particulate)
Sulfate (particulate)
Chloride (particulate)
Sulfur dioxide (gas)
Sulfur trioxide (gas)
Hydrogen fluoride (gas)
Hydrogen chloride (vapor)
Nitrogen oxide (gas)
Pentachlorophenol
(particulate)
Sodium pentachlorophenol
(particulate)
Phenol (vapor)
Captan (particulate)
DDT (particulate)
Hydrogen chloride (vapor)
9.5-31
-------
Table 9.5-9. SUMMARY OF AIR EMISSION CONTROL FOR FIVE MAJOR PESTICIDES3^
Pesticide
Control device
Emissions controlled
Reported
efficiency,
to
i
CO
INS
Methyl parathiona
ToxapheneC
Trifluralinf
PentachlorophenolS
Incinerator
Water scrubber
BrinkR mist eliminator
Baghouse
Baghouse
Water scrubber
Acidifier vent scrubbers
1- and 2-stage venturi scrubber
and Tri-mer wet scrubber
Packed and venturi scrubber
Baghouse
Hydrogen sulfide, sulfur, mercaptan
Phosphorus pentoxide, hydrogen chloride
Phosphorus pentoxide (for visibility)
Toxaphene
Arsenic trioxide
Arsenic trioxide
Chlorine, phenol, acids
Pentachlorophenol
—b
95
99.9
90
90 to 100
95 to 99
information reported by Monsanto.
^Blanks indicate data not available.
Information reported by Hercules.
^Information reported by Diamond Shamrock.
Information reported by Ansul.
fInformation reported by Eli Lilly.
SInformation reported by Reichhold.
-------
are located in California and Wyomijng and via the synthetic Solvay process.
• !
Synthetic production has declined sharply since the mid-1960s and only one
synthetic plant is currently (August 1979) in operation.36 Table 9.5-10
shows the location and size of plants that produce sodium carbonate by the
8 - '•
natural process. !
9.5.11.1 Source Description, i Figures 9.5-12, 9.5-13, and 9.5-14 are
I
flowsheets for the three processes used to produce sodium carbonate from
brine or ore.3? The sources of particle emission include ore handling
equipment, calciners and dryers, purification steps, and product handling.
i
In the calciners and dryers, small particles are entrained in the hot gases
I
traveling past the ore or product, iIn the purification steps, particles are
entrained in rising vapors through the'crude soda ore. There are also emis-
sions from trucks traveling on dusty roads. Calcining and drying are the
largest sources of particulate emission. Tables 9.5-11, 9.5-12, and 9.5-13
list emission sources and control techniques for the three processes used.36
9.5.11.2 Applicable Control Technologies. Cyclones, wet scrubbers,
precipitators, and baghouses can all be used for particulate control. The
I
choice of one device over another would be site dependent. To control emis-
sions from storage piles and conveyer belts, hoods and baghouses must be
constructed (see Volume 1, Section 5). When using baghouses, care must be
taken to prevent blinding because of the hydroscopic nature of soda ash.3^
9.5.12 Potash i
The potash industry produces three products: (1) muriate of potash
(KC1), (2) langbeinite (K2S04 • 2MgSJ04), and (3) sulfate of potash (K2S04).
Plants where these compounds are produced are described in Table 9.5-14.
9.5.12.1 Source Description. JFigures 9.5-15, 9.5-16, and 9.5-17 des-
cribe the various processes used to produce potash.3^ Processing steps that
emit particulate matter include crushing, screening, conveying, drying, com-
pacting, evaporating, and product st'pring and loading. These steps are com-
mon to all the processes. Table 9.5|-15 lists particle sizes for emission
from the screening, drying, and compacting steps of muriate production.
' 9.5.12.2 Emission Characteristics and Applicable Control Technologies.
Dryers are the major sources of partjiculate emissions in the potash indus-
tries. Control equipment for dryers,includes dry cyclones, wet scrubbers,
and cyclones with baghouses. Emission levels of 3 to 5 g/Nm3 can be achieved
9.5-33
-------
Table 9.5-10. SODIUM CARBONATE NATURAL PROCESSES36
to
Capacity
Owner
Kerr-McGee
Kerr-McGee
Allied Chem.
FMC Corp.
Stauffer Chem.
Texasgulf, Inc.
Plant name
Trona
West End
Trona
Westvaco
Big Island
Location
Trona, CA
San Bernardino, CA
Green River, WY
Green River, WY
Green River, WY
Granger, WY
106 metric tons
per year
1.2
0.14
2.0 .
1.14
1.14
1.5
0.9
106 short tons
per year
1.3
0.15
2.2
1.25
1.25
1.65
1,0
Process
Direct carbonation
Direct carbonation
Monohydrate
Monohydrate
Sesqui carbonate
Monohydrate
Monohydrate
-------
o
>— I -z.
1—0
•< ,_,
Q£ (—
LlJ S
Q£ CQ
n cv
>=C
LlJ O
KM Q
a: -z.
CO <=C
<
^ >-
O C£.
2: CD LU
rs oc >
i— i ==C O
O <-J <-J
O i— i UJ
oo CQ o:
' O
1 — 1
^
00 O
.<•-,
U-
=a: K-I
Q QC
o rs
cs
•z.
i— i C3
>- ^
Qi ^i
Q
<_j «a:
g^
O Q
C£ ^
Q- «a:
•BRINE
! i
PRECftRBONATION
! 1
PRIM:. & SECON.
CARBONATION
CRYSTALLIZATION
; -1
FILTERING
r
| T
CALCINING-DRYING
. _. ^ j, _
i ^1
BLEACHING
i
. 1
RECRYSJALLIZATION
: i
WASHING
> V
CENTRIFUGATION
• 1
DRYING
1
SHIPPING
i
Figure 9.5-12 Direct carbonation process.37
9.5-35
-------
CD
2-
0»-i
%£
CD^
CE Z
t— i
52
ac o
1
PURIFICATION
1
1
1
1
cs
h- Z
o »-«
Bel
S5
Q- 3=
MINE
i
r
ORE
STOICKPILE
1
SCREENING &
CRUSHING
~)
r
DISSOLUTION
)
f
CLARIFYING AND
/OR THICKENING
, ' \
t
FILTRATION
>
r
CRYSTALLIZATION
i
r
CENTRIFUGATION
>
r
CALCINING
• )
r
COOLING
^
f
SHIPPING
•,/
Figure 9.5-13 S6squicarbonate proccess.*7
9.5-36
-------
CD
i O
cz.
>- ^
Qi I—I
Q _l
c_3 =a;
ra a:
a
cj a
c£ 2:
a.
-------
Table 9.5-11. TYPICAL PARTICULATE EMISSION SOURCES AND CONTROLS FOR A
MONOHYDRATE PROCESS PLANT36
Average outlet
grain loading,
Process equipment Control g/Nm^ (dry basis) kg/Mg
Coal-fired calciner Cyclone-ESP 0.06 0.11
Dissolver Venturi scrubber 0.09
Rotary steam tube dryer Venturi scrubber 0.09 0.45
-------
Table 9.5-12.
EMISSION SOURCES AND CONTROL EQUIPMENT FOR THE
SESQUICARBONATE ;PROCESS36
Process equipment
Controjl equipment
Outlet
grain loading,
g/Nm3 (dry basis) kg/Mg
Screening and crushing
Rotary steam tube calciner
Gas-fired calciner
Fluid bed steam tube calcine
Product storage
Product shipping
Bag collector
Wet scrubber
Cyclone-ESP
Wet scrubber
Bag collector
0.05
0.06
0.06
0.06
0.05
0.05
0.11
0.11
0.11
9.5-39
-------
TABLE 9.5-13. EMISSION SOURCES AND CONTROL EQUIPMENT FOR THE DIRECT
CARBONATION PROCESS35
Process equipment
Outlet grain loading.
Control equipment g/Nm3 (dry basis) kg/Mg
Gas-oil-fired dryer-calciner
Gas-oil-fired dryer
Carbonation bleacher
Product storage
Product shipping
Cyclone and wet
scrubber
Wet scrubber
Cyclone-venturi
scrubber
Bag collector
Bag collector
0.06
0.09
0.02
0.05
0.05
0.11
0.45
0.03
9.5-40
-------
Table 9.5-14. U.S. POTASH PRODUCERS38
10
-yr
i
Capacity
Company
New Mexico
AMAX Chemical Corporation
Duval Corporation
International Minerals
and Chemical Corp.
Kerr-McGee Chemical Corp.
Mississippi Chemical Corp
National Potash Co.
Potash Co. of America
Utah
Great Salt Lake Minerals
and Chemical Corp.
Kaiser Alum, ft Chemical
Corp.
Texas Gulf, Inc.
California
Kerr-McGee Chemical
Corp.
Location
Carlsbad
Carlsbad
Carlsbad
Bobbs
Carlsbad
Carlsbad
Carlsbad
Ogden
Wendover
Moab
Searles
Lake
Parent company Products
American Metal Climax, Inc. KC1
Pennzoil Co., Inc. KC1,K?S04,
2MgS04,K2S04
KC1,K2S04,
2KgS04,K2S04
KC1
KC1
Freeport Minerals Co. KC1
Ideal Basic Industries, Inc KCL
Gulf Resources and K2S04
Chemicals Corp.
Kaiser Industries Corp* KC1
KC1
KCL,
K2S04
Tg K20/year
410
" 225a
320
320
165
210
480
no
55
165
180
103 short tons,
KgO/year
450
250a
350
350
180
230
530
120
60
180
200
aK2SO, production was terminated in late 1977, KC1 production to be terminated in 1978
K2S04MgS04 product to be expanded to 100 Tg per year (110,000 short tons per year).
-------
Sylvite Ore-
Sat'd.
Irine Sol'n.
Flotation
Agents
(Tails)
Screens
V
Coarse
Product
+
Granular
Product
*
Standard
4 t
Compactors
Granular
Product
Figure 9.5-15 Simplified flow sheet for production of muriate of potash
from sylvite ore.39 (Estimated emission points - specific
points not identified in ref.39.)
9.5-42
-------
Langbeinlte Ore *j , Crush
-14 Mesh
90% NaCl
Brine
Screen
+14 Mesh
Leach
40%
Water
Debrine
NaCl Brine
Leach
\ Debrine j
K-Mg-S
K-Mg-S
I Dry
Screen
Granular Standard
Product Product
Figure 9..5-16 Simplified flow'diagram for production of langbeinite,
(K-Mg-SJ.?-8 . „ (Estimated" emission pqints, -specific
points ..not identified in ref. 3.8..)
9.5-43
-------
Langbeinite
Fines
KC1 Fines
Crush
-200 Mesh
Slurry
Water
KC1
Brine
REACT
Dewater
Solution
Evaporate
Solids
Centrifuge
KC1 Crvstals
MgCl2
Solution
Screen
Granular Standard
Product Product
Figure 9.5-17 Simplified flow diagram for production of sulfate of potash.38
(Estimated emission points - specific points not identified
in ref.38.)
9.15-44
-------
Table 9.5-15.
PARTICLE SIZE DISTRIBUTION DURING MURIATE
PRODUCTION38 j
Emission source
Particle size,
micrometers
Percent particles
coarser than
(by weight)
Screening section
20.0
10.0
5.0
2.0
1.6
2
9
26
60
83
Dryer
20.0
10.0
5.0
2.0
1.0
7 to 2
23 to 7
50 to 15
83 to 35
95 to 54
Compactor
20.0,
10.0
5.0
2.0
1.0
7 to 1
20 to 5.5
42 to 12
73 to 41
89 to 64
9.5-45
-------
using dry cyclones, which are the most commonly used control technique. ,Wet
scrubbers can reduce the emission level to 0.4 g/Nm3 with a pressure drop
of 2 to 5 kPa; however, there is a potential water disposal problem, espe-
cially if the plant is under zero discharge constraints. In some cases the
water can be recycled to the crystallizer.39
Table 9.5-16 is a summary, by particle size, of data from an emissions
test of a potash dryer and a salt dryer. The control device for the potash
dryer was a multivane scrubber operating at 0.78 kPa pressure differential.
The control device for the salt dryer was a wetted fiber scrubber.
Since KC1 is quite hydroscopic and could cause caking, baghouses on
dryers are in limited use. They have been used, however, on a sulfate of
potash dryer because sulfate is much less hydroscopic than chloride.39
Using steam-tube dryers instead of direct-fired dryers could reduce the
total gas flow and particulate emissions.39
The flue gas from submerged combustion evaporators contains 50 to 60
percent moisture and high loadings of particulate matter which make this
stream difficult to clean. Also, in one test, it was found that 85 percent
of the particles are less than one micron in diameter. At present, no con-
trols are being used in.the U.S.39
Wet scrubbers with pressure drops of 1.5 to 2.5 kPa, venturi scrubbers,
or tray towers are used to control emissions from compactors. When baghouses
are used, caking can become a problem. Removal efficiencies with these con-
trols can be greater than 99 percent. A dry cyclone with a control effi-
ciency of 90 to 98 percent is another possible control device for compactors.
These control efficiencies are based on actual measurements but Reference 39
does not list the particulate loading.39
For bagging and loading operations, a cyclone with control efficiencies
of 90 to 97 percent can be used. Some type of shed or hood arrangement with
a fan is" required. A wet scrubber could be used for higher control effi-
ciencies. 39 (See Volume 1, Section 5 for more details.)
9.5-46
-------
Table 9.5-16. SIZE SPECIFIC EMISSIONS FROM POTASH AND SALT DRYERS12
Ul
I
Mass concentration, mg/DMCM
(mass percent less than stated size)
Total
15.3 Aim
12.9 /Am
10.1 fjum
7.28/>tm
5 fj^rn
2.5 ju,m
1.01 jura
Potash Uncontrolled 802 499(62.2) 493(61.4) 466(58.0) 412(51.3) 298(37.2) 153(19.1) 75.1(9.4)
dryer3
Controlled 233 201(86.3) 198(84.9) 190(81.7) 175(74.9) 155(66.5) 107(46.0) 48.8(20.9)
Efficiency
70.9 59.7
59.8
59.2
57.5
48.0
30.1
34.7
3.0 1.93(64.3) 1.90(63.2) 1.81(60.1) 1.60(53.4) 1.29(43.0) 0.64(21.2) 0.16(5.5)
0.24 0.22(90.2) 0.22(90.1) 0.22(90.0) 0.22(89.6) 0.22(89.0) 0.20(82.2)0.15(60.6)
Efficiency 92 88.6 88.4 87.8 86.3 82.9 68.8 6.3
Salt Uncontrolled
dryerb
Controlled
aFPEIS test series number is 49.
&FPEIS test series number is 52.
-------
REFERENCES FOR SECTION 9.5
1. OAQPS Data File of Nationwide Emissions. U.S. Environmental
Protection Agency, Office of Air Quality Planning and Standards,
Research Triangle Park, NC. February 1979.
2. Shreve, R. N. Chemical Process Industries, 3rd edition. McGraw-
Hill Book Company. New York, NY. 1967
3. Moscowitz, C. M. Source Assessment: Charcoal Manufacturing - State
of the Art. U.S. Environmental Protection Agency. Publication
No. EPA-600/2-78-042. December 1978.
4. Hulman, P. B. Screening Study on Feasibility of Standards of
Performance for Wood Charcoal Manufacturing. Radian Corporation.
Austin, TX. Publication No. 78-200-187-32-06. August 1978.
5. Source Assessment - Carbon Black Manufacture. U.S. Environmental
Protection Agency. Publication No. EPA-600/2-77-107k. October 1977.
6. Schwartz, W. A., et al. Engineering and Cost Study of Air Pollution
Control for the Petrochemical Industry - Volume I: Carbon Black
Manufacture by the Furnace Process. U.S. Environmental Protection
Agency. Publication No. EPA-450/3-73-006a. June 1974.
7. Drabkin, M., and K. J. Brooks. Review of Standards of Performance for
New Stationary Sorces - Sulfuric Acid Plants. U.S. Environmental
Protection Agency. Publication No. EPA-450/3-79-003. January 1979.
8. Gerstle, R. W., et al. Industrial Process Profiles for Environmental
Use, Chapter 4, Carbon Black Industry. U.S. Environmental Protection
Agency. Publication No. EPA-600/2-77-023d. February 1977.
9. Draft SSEIS Report on Carbon Black. U.S. Environmental Protection
Agency, Emissions Standards and Engineering Division. Research
Triangle Park, NC. April 1976.
10. Danielson, J. A. Air Pollution Engineering Manual. U.S.
Environmental Protection Agency. Publication No. AP-40. May 1973.
11. Sittig, M. S. Particulates and Fine Dust Removal. Noyes Data
Corporation. Park Ridge, NJ. 1977.
12. Source Category Survey: Detergent Industry. U.S. Environmental
Protection Agency. Publication No. EPA-450/3-80-030. June 1980.
13. Ibid.
14. Trip report on Procter and Gamble detergent plant. U.S. Environmental
Protection Agency, Office of Air Quality Planning and Standards,
Standards Documentation Section. Research Triangle Park, NC.
File AID 8-1-24. April 1980.
9.5-48
-------
15. Hudak, C. E., and T. B. Parsons. Industrial Process Profiles for
Environmental Use, Chapter 12, The Explosives Industry. U.S.
Environmental Protection Agency. < Publication No. EPA-600/2-77-023c. .
February 1977. j
16. Compilation of Air Pollution Emission Factors, 2nd edition. U.S.
Environmental Protection Agency. ; Publication No. AP-42. April 1973.
17. Nelson, T. P., and R. E. Pyle. Screening Study to Determine the Need
for New Source Performance Standards in the Explosives Manufacturing
Industry. Radian Corporation. Austin, TX. July 1976.
I
18. Happel, J., and M. A. Hantow. Feasibility Study of a Process to Treat
Flue Gas from Red Water Incinerators. Pictinny Arsenal. Contract
No. DAAA-2I-74-C-0537. February 1976.
19. Atmospheric Emissions from Thermal-Process Phosphoric Acid Manufacture.
U.S. Public Health Service. National Air Pollution Control Adminis-
tration. Publication No. AP-48. j October 1968.
20. Source Category Survey: Thermal Process Phosphoric Acid Manufacturing
Industry. U.S. Environmental Protection Agency. Publication No.
EPA-450-3-80-018. May 1980. j
I . •
21. Bury, J. I., et al. Potential Abatement, Production and Marketing
of Byproduct Sulfuric Acid In The1 U.S. U.S. Environmental Protection
Agency. Publication No. EPA-600/7-78-070. April 1978. pp 17-18.
22. Gerstle, R. W., et al. Industrial Process Profiles for Environmental
Use, Chapter 23: Sulfur, Sulfur Oxides, and Sulfuric Acid. U.S.
Environmental Protection Agency. ; Publication No. EPA-600/2-77-023w.
February 1977. i
23. Final Guideline Document: Contro;! of Sulfuric Acid Mist Emissions
from Existing Sulfuric Acid Production Units. U.S. Environmental
Protection Agency. Publication No. EPA-450/2-77-019. September 1977.
24. Compilation of Air Pollutant Emission Factors. U.S. Environmental
Protection Agency, Office of Air Quality Planning and Standards.
Research Triangle Park, NC. Publication No. AP-42, Supplement A.
July 1979. |
25. Serth, R. W., and T. W. Hughes. Source Assessment: Phthalic Anhydride
(Air Emissions). U.S. Environmental Protection Agency. Publication No.
EPA-600/2-76-032d. December 19761.'
26. Schwartz, W. A., et al. Engineerjing and Cost Study of Air Pollution
Control for the Petrochemical Industry - Volume 7: Phthalic Anhydride
Manufacture from Ortho-Xylene. U.S. Environmental Protection Agency.
Publication No. EPA-450/3-73-006g; July 1975.
27. Boscak, V. Screening Study on Feasibility of Standards of Performance
for Hydrofluoric Acid Manufacture. U.S. Environmental Protection
Agency. Publication No. EPA-450/3-78-109. October 1978.
9.5-49
-------
28. Lowenheim, F. A., and M. K. Moran. Faith, Keyes, and Clark's
Industrial Chemicals, 4th edition. Wiley-Interscience Publication.
New York, NY. 1975. U.S. Environmental Protection Agency. Publi-
cation No. EPA-450/3-78-109.
29. Muehlberg, P.E., et al. Industrial Process Profiles for Environ-
mental Use, Chapter 15: Brine and Evaporite Chemicals Industry. U.S.
Environmental Protection Agency. Publication No. EPA-600/2-77-023o.
February 1977.
30. Davis, W. E. National Inventory of Sources and Emissions: Boron-1969.
U.S. Environmental Protection Agency. Publication No. APTD-1159.
June 1972.
31. Lemon, E. D. Wet Scrubbing Experience with Fine Borax Dust. Journal
of the Air Pollution Control Association. Pittsburg, PA. 27:(1080),
1977.
32. Calvert, S. American Air Filter Kinpactor 10 X 56 Venturi: Scrubber
Evaluation. U.S. Environmental Protection Agency. Publication No.
EPA-600/2-77-209b. November 1977.
33. Kelso, 6. L., et al. Development of Information on Pesticides
Manufacturing for Source Assessment. U.S. Environmental Protection
Agency. Publication No. EPA-600/2-78-100. May 1978.
34. Archer, S..R. Source Assessment: Pesticide Manufacturing Air
Emissions - Overview and Prioritization. U.S. Environmental Protection
Agency. Publication No. EPA-600/2-78-004d. March 1978.
35. Patterson, J. W. State-of-the Art for the Inorganic Chemicals
Industry: Inorganic Pesticides. U.S. Environmental Protection Agency.
Publication No. EPA-600/2-74-009a. March 1975.
36. Blythe, 6. M. Screening Study to Determine Need for Standards of
Performance for the Sodium Carbonate Industry. Radian Corporation.
Austin, TX. Publication No. DCN 78-200-187-34-08. October 13, 1978.
37. Sodium Carbonate Industry - Background Information for Proposed
Standards. U.S. Environmental Protection Agency. Publication No.
EPA-450/3-80-029a. August 1980.
38. Blythe, 6. M., and K. N. Trade. Final Report - Screening Study to
Determine Need for Standards of Performance for the Potash Industry.
Radian Corporation. Austin, TX. Publication No. DCN 78-200-187-35-08.
October 13, 1978.
39. Wilson, D. B. An Engineering Analysis of Particulate Recovery from
Selected Potash Refining Operations. New Mexico State University,
Department of Chemical Engineering. Las Cruces, NM. November 1977.
9.5-50
-------
9.6 FOOD AND AGRICULTURAL INDUSTRY i
The production of agricultural fertilizers and the processing of food
and feed from agricultural crops are activities which, for purposes of this
discussion, are categorized as the Food and Agricultural Industry. Food and
feed products undergo a number of processing steps, such as refinement, pre-
servation, and product improvement, as well as storage, transfer, packaging,
and shipping before being used by the 'consumer. Particulate emissions from
these processes usually consist of agricultural waste materials, grain dust,
seeds, hulls, and dirt. Production processes, emissions characteristics,
and particulate matter abatement technologies are presented for those fer-
tilizer, food, and feed industries that have the potential to contribute
significant quantities of particulate Emissions to the atmosphere. These
industries include ammonium nitrate, ammonium sulfate, urea, and diammonium
phosphate fertilizer production, as well as grain handling, grain processing,
alfalfa dehydration, cotton ginning, and starch and vegetable oil manufac-
turing. !
9.6.1 Ammonium Nitrate Fertilizer j
i
Ammonium nitrate, an important so'urce of nitrogen in fertilizers, is
also used for a variety of nonagricultiural purposes, including the production
of nitrous oxide and as a component iri certain explosives. In the United
States there are 63 ammonium nitrate plants, located primarily in the Central
and Southeastern States. In 1979, theise plants produced 7.0 Tg of ammonium
nitrate.1 Pollutants emitted during ammonium nitrate production include
ammonia and nitric acid mist and solid ammonium nitrate. The solid particles
are typically either "greater than! millimeter or less than 3 micrometers
in size.2 Ammonium nitrate is water soluble; therefore, particulate emis-
sions can become a water pollution problem on rainy days.
9.6.1.1 Process Description. Ammonium nitrate is formed from the
reaction of ammonia with nitric acid in a vessel called a neutralizes
This ammonium nitrate solution is then1concentrated to a 95+ percent solu-
tion (ammonium nitrate in water) in an!evaporator to form a melt. The am-
monium nitrate thus formed is either marketed as a solution or solidified
from the melt by graining, granulation, or prilling (pelletizing). Among
these processes, prilling is the most commonly used and the largest source
of particulate emissions, principally because all granulators in use are
9.6-1
-------
controlled to reduce process losses. Prills are formed when the hot melt is
sprayed through orifices at the top of the prilling tower and falls counter
current to a flow of cooling air. Solid ammonium nitrate may be cooled,
dried, and coated (to improve shelf life) before being bagged and stored.
Main emission points in the solution part of this process are the neu-
tral izer and the atmospheric vents on the evaporator which emit ammonium
nitrate particles. Fugitive particulate emissions are produced during
solid product screening, coating, bagging, and bulk loading. A flow dia-
gram indicating major emissions points is given in Figure 9.6-1.
9.6.1.2 Emissions and Controls. In 1977, nationwide particulate emis-
sions from the manufacture of ammonium nitrate amounted to 7.4 Gg.3 Process
steps responsible for these emissions are as follows: (1) neutralization;
(2) evaporation and concentration; (3) solids formation; (4) product finish-
ing; and (5) product screening, coating, bagging and bulk loading.4
9.6.1.2.1 Neutralization. The neutralization reaction is exothermic
and produces steam that may contain particulates, ammonia, and/or nitric
acid. During normal plant operations, emissions can be eliminated by total
condensation. However, even with total condensation it is necessary to vent
steam during startup, shutdown, or upset conditions. Uncontrolled particu-
late emissions for conventional neutralizes usually fall in the range of
0.25 to 3.7 kg/Mg of 100% ammonium nitrate.2>4,5 Average emission rates
after control, as shown in Table 9.6-1, range from 0.14 to 0.5kg/Mg.
Mississippi Chemical Corporation neutralizer emissions typically range from
0.07 to 0.5 kg/Mg.2,4
9.6.1.2.2 Evaporator/concentrator. Approximately 75 percent of the
industry utilizes film-type evaporators to concentrate the ammonium nitrate
to the levels required for subsequent prilling or granulation. Ammonium
nitrate particulate emissions in the vapor streams off the evaporators gener-
ally fall in the range of 0.3 to 1.0 kg/Mg of ammonium nitrate.4*5 About 70
percent of particulate emissions from the neutralizer, evaporator/concentrator,
and prill tower are less than 3 m in size.2 Emissions from the evaporator/
concentrator are commonly controlled by a scrubber. In a few plants, the
neutralizer and the evaporator/concentrator are ducted to a mist eliminator.
9.6.1.2.3 Solid formation. Approximately 60 percent of the ammonium
nitrate produced in the United States is sold as a solid product. Eighty
9.6-2
-------
to
en
i
co
PARTICULATES,
AMMONIA,
NITRIC ACID
NHi
HN03
f
NEUTRALIZER
1
PARTICUUTES .
EVAPORATOR/
CONCENTRATOR
r
PARTICLE
FORMATION
PARTICULATES PARTICULATES
i
COATING
MATERIAL
MARKET
MARKET
MARKET
Figure 9.6-1 Flow diagram of ammonium nitrate production.1
-------
Table 9.6-1. SUMMARY OF NEUTRALIZATION EMISSION DATA
Particulate emissions
4,5
Emission control method g/m3 kg/Mga of product
Mist eliminator 0.50 0.42
Partial condensation 1.23 0.50
Mississippi Chemical
Corporation neutralizer 0.36 0.17
Venturi scrubber 2.40 0.14
aGrams per wet standard cubic meter.
9.6-4
-------
percent of the solid ammonium nitrate is produced by prilling and 10 percent
by granulation; the remaining 10 percent is manufactured by other methods.
Prill Towers—The prilling process, which involves prills falling through
a counter-current airstream, is highly conducive to particulate entrainment.
The particulates in the submicrometen range, referred to as "fumes," result
from the evaporation and subsequent condensation and solidification of the
material being prilled; they are particularly difficult to control. The
amount of fumes produced is temperature dependent; to minimize fume forma-
tion, melt temperature should be kept! as low as possible and melt composi-
tion carefully controlled. \ '
l
Approximately 70 percent of the low density ammonium nitrate prill
towers operate without emissions control equipment; 14 percent use a col-
lection cone in combination with a mist eliminator; 7 percent use wet
scrubbers; and the remainder use meshj pads or similar-'devices. Most
high density prill towers are controlled, with about 60% utilizing a mist
eliminator/collection cone system, and 18% using wet scrubbers. Tests for
uncontrolled ammonium nitrate prill towers indicate an emission rate of
0.81 to 2.7 kg/Mg for high density prills and 0.21 to 0.69 kg/Mg for low
density prills.1 The combined cone/mi'st eliminator collection system
appears to reduce emissions to below 20 percent opacity.
Industry-supplied test data for six different high density prill
towers with cone/mist eliminators indicate a controlled emission rate range
of 0.03 to 0.51 kg/Mg.1 (In this case, the controlled emission rate refers
to the sum of the emissions from the mist eliminators and the by-pass stream
from the prill tower. Uncontrolled particulate emissions from three dif-
ferent granulators range from 138 to 152 kg/Mg.1) Test data for two low
density prill towers with cone/mist eliminators indicate a controlled emis-
sion rate of 0.06 and 0.24 kg/Mg.1 i
The performance of wet scrubbers n's especially sensitive to the partic-
ulate size distribution, which, for a 'typical prill tower, is as follows:
o 30 percent by weight: > 3 //,m
o 20 percent by weight: 1 to 3 jum
o 35 percent by weight: 0.5 to. :1 ^m
o 15 percent by weight: < 0.5//,m
9.-6-5
-------
The large fraction below 1.0 /tm creates a difficult control problem and a
high opacity even at low concentrations. A low-energy scrubber reduces
emissions to 0.625 kg/Mg at a concentration of 0.098 g/m3.
Granulatoi—Because of the limited use of drum granulators in the am-
monium nitrate industry, less data are available regarding emissions from
this source. Tests on two plants indicate an emission range of 0.22 to
0.61 kg/Mg of ammonium nitrate produced for a granulator equipped with a
scrubber.1 Due to the relatively high uncontrolled emission rates from
granulators, a scrubber is considered to be an integral part of the process.
Product Finishing—Drying removes water from solids that have been
formed with a high moisture content melt. Typically, this step is performed
by two rotary drum dryers in series. Cooling is usually accomplished in
another rotary drum, although fluidized bed coolers are beginning to gain
some acceptance. It is common practice to use a coating and/or an additive
to enhance prill shelf life and to suppress dust emissions from solid ammo-
nium nitrate particles. Additives include magnesium oxide, calcium oxide,
and magnesium nitrate.
Uncontrolled emissions from dryers and coolers are reported to range
from 1 to 38 kg/Mg of product, but are readily reduced by scrubbers.1
Emissions from, coating operations are fugitive emissions. Based on the
estimate of 10 percent loss of coating material during coating operations,
there is an emission of 3 kg/Mg of ammonium nitrate (for a coating level of
3 percent). Most of the material actually settles to the floor and only a
small percentage escapes to the atmosphere.
Predryers, dryers, and coolers are usually very similar, except that
warm air is used for predryers/dryers and cold air is used for coolers.
Wet scrubbers are practically the only type of equipment used to control
emissions from these sources. Emission rates for low energy scrubbers
controlling predryers, dryers, and coolers range from 0.02 to 0.65 g/kg of
product.1»4
Product size is controlled primarily by screening. Oversize and under-
size material is removed from the product solids and recycled. Only a small
quantity of particulates escape as fugitive emissions from the building in
9.6-6
-------
which screening is performed. i
One source of emissions of airborn fines is the series of transfer
points (often by conveyor belt) in the process. The severity of this
emission source will depend on the characteristics of the material.
Approximately 90 percent of all qtmmonium nitrate solids are handled
in bulk. Because of the small quantities of fines present in the ammonium
nitrate solid, particulate entrainment is low in these operations. One
estimate is that less than 0.1 kg of particulates per megagram of ammonium
nitrate product is entrained in bulk handling.4
9.6.1.3 Environmental Impact. The extensive use of wet scrubbers to
reduce particulate emissions in the ammonium nitrate industry may create
wastewater problems. The wastewater can be treated with lime; however, such
treatment results in a solid waste disposal problem.2,6 Caustic scrubbers
produce sodium nitrate, which also creates disposal problems.?
9.6.2 Ammonium Sulfate Fertilizer !
Ammonium sulfate, although used primarily as a fertilizer, is also used
in water treatment, Pharmaceuticals, fermentation, food processing, fire-
proofing, and tanning. It can be manufactured from the neutralization reac-
tion between sulfuric acid and ammonia, similar to the way in which ammonium
nitrate is produced. However, most ammonium sulfate is obtained as a by-
product from other processes. Total production of ammonium sulfate in 1977
was 2.2 Tg, 51 percent of which was manufactured as a by-product of capro-
lactam production, 21 percent by the synthetic process, and 20 percent as a
by-product of coke manufacture.8 The(balance of ammonium sulfate (AS) is
produced as by-products from miscellaneous production processes.
9.6.2.1 Process Description. A;generalized flow diagram for the three
major AS manufacturing processes is shown in Figure 9.6-2. Although the
processes differ in the manner of AS crystal production, the subsequent
operations of crystal dewatering and drying followed by screening are quite
i
similar for these processes. j
9.6.2.2 Emissions and Controls.' In 1977, nationwide particulate emis-
sions from the manufacture of ammonium sulfate amounted to 17.2 Gg.3 The
primary process emission point is the(ammonium sulfate dryer.9 Only rotary
dryers are reported to be used in the |synthetic manufacturing plants while
9J6-7
-------
Caprolaetam By Product: Process
10
CO
Cryotallircr
•m*
Vacuum
System
Steam Cond.
To ATH
To ATH
TT
Steam Cond.
Roots
Blower
As Product(s)
J ATH
(Alt.)
S team
— J l— *
Vacuum
Pilfer
Dryer
Enclosed
Storage
1
Cond ,
Notes:
Heat N— Air
Steam—' l-*> Cond.
AS Product
1. Dryer tnay be rotary or Eluldlzed bed type
2. Coke oven plants may integrate cetvtrlfuging and drying or centrifuge only
3. Coke oven plant product not screened
Figure 9.6-2 Flow diagram fcr ammonium sulfate processes,9
-------
caprolactam by-product plants utilize fluidized bed or rotary dryers.9 In
i
coke oven gas by-product plants, dewatering drying is accomplished by using
rotary vacuum filters, combination centrifuge-dryers, or by centrifuges
followed by a rotary dryer.9 i
Industrial process fugitive emissions from materials handling or product
screening were not found to be major emission sources in a recent study and,
for that reason, control technology for these sources is not discussed
herein.9 ;
Particle size distribution test: data of uncontrolled particulate emis-
sions for several dryers are presented in Figure 9.6-3. The particle size
distribution for the one fluidized bed dryer which was sampled indicates a
larger percentage (greater than 99,9-f- percent) of particles greater than 1
micrometer in diameter than the rotary dryer test results, which ranged from
93 to 99.9 percent.9 |
i
The uncontrolled emission rate for one fluidized bed dryer was measured
i
at 111 kg/Mg, compared to an averagejof 23 kg/Mg (ranging from 0.4 to 77
kg/Mg) for 3 rotary dryers. This finding may be a consideration in control-
ling emissions since fluidized bed dryers would generally appear to require
more efficient control in order to comply with applicable air pollution
regulations.9 ;
The AS salt exhibits a moderately high solubility in water and, conse-
quently, wet scrubbing has frequently been used to control dryer emissions.9
Organic impurities may contaminate the AS: caprolactam at caprolactam by-
product plants and tars at coke oven'by-product plants account for these
impurities. Wet scrubbing has an advantage compared to baghouse control at
a caprolactam plant because test data have indicated up to 85 percent collec-
tion of caprolactam vapor emissions from the AS dryer.9
Venturi, centrifugal, spray-type, and packed tower designs of scrubbers
have been used in the industry. The:only fabric filter baghouse in use is
located at a synthetic AS plant. Colce oven facilities employ cyclones,
scrubbers, or no emission control systems.
Five particulate control systems, have been tested in a study of the AS
industry. The results of these tests are summarized in Table 9.6-2. The
baghouse and wet scrubbers with pressure drops in the 2.5 to 3.2 kPa range
9.6-9
-------
VD
r
CD
O
0.1
0.01 .05.1.2 .512 5 10 20 30 40 50 60 70 80 90 95 98 99 99.1 99.9 99.99
CUMULATIVE PERCENTAGE, % weight xdiameter
Figure 9.6-3 Uncontrolled ammonium sulfate dryer emissions
- particle size'distribution.9
-------
Table 9.6-2. EMISSION TEST RESULTS FOR COHTROU.EP DRIERS IN AMMONIUM SULFATE
MANUFACTURING PLANTS.9
Production
rate,
Plant Mg/h
A 13.9
B N/A
C 15.2
D 8.45
E 8.9
Dryer
type
Rotary
Fluldlzed
bed
Rotary
Rotary
Fluidlzed
bed
Stack
gas
Control temperature,
technology °C
Baghousec 46
Venturl, 85
scrubber
Centrifugal 83
scrubber
Venturl 45
scrubber
<# -
Cyclones and 46
centrifugal
scrubber
Particulate
concentrations ,a
(g/Nm3)
Inlet
N/A
39.0
8.87
98.29
N/A
Outlet
0.0490
0.0430
0.206
0.190
0.0577
Controlled
emission
factor,
kg/Mg
0.009
0.14
0.8
0.16
0.14
Average Control
collection equipment
L/G ratio,
efficiency, pressure drop, liters, oer
% kPa
98. 7d N/Ab
99.9 2.5
98.3 1.5
(design)
99.9 3.2
(design)
3.3
(design)
nrVmin
—
3.21
0.67
(design)
3.61
(design)
0.27
(design)
aEach result Is an average of three emission tests.
''N/A — Data were not collected or are not available.
cTwo sets of tests (three tests/set) were collected. The first set is not reported due to discovery of
punctures resulting in abnormally high particulate mass loadings.
^Efficiency calculated from average Inlet concentration from first set of test results.
-------
were highly efficient, removing 98.7 percent of the participate matter for
the baghouse and 99.9 percent for the two scrubbers. For these three collec-
tors plus Plant E where inlet concentration was not measured, the controlled
emission factor was less than 0.15 kg/Mg; Plant D was only slightly higher,
0.16 kg/Mg.a
9.6.3 Urea Fertilizer
Urea is used as a nitrogen fertilizer in tropical and temperate
climates, as a protein supplement in animal feeds, and in plastics manu-
facturing. The 43 urea plants in the United States are located next to
ammonia plants, which supply both the ammonia and carbon dioxide feed
streams necessary in urea manufacturing. Urea production in 1979 was 7.16
Tg.10 More than 99 percent of the particulate matter emitted during urea
manufacture consists of urea dust. The other particulates are coating
materials and ammonia mist. Particle sizes are either fine (diameters
ranging from 2 to 200 micrometers) or coarse (1 to 2 millimeter
diameters).11
9.6.3.1 Process Description.4 Urea is produced by reacting ammonia
(NHs) and carbon dioxide (C02) to form ammonium carbamate (NH4C02NH2). The
carbamate is then dehydrated to yield urea. The final product is distri-
buted either as a 70 to 75 percent urea solution or as a solid.
9.6.3.2 Emissions and Controls. The overall urea manufacturing process
can be broken down into the following steps:
1. Solution formation
2. Solution concentration
3. Solids formation
4. Solids cooling
5. Coating and/or additives
6. Screening
7. Bagging, storage, bulk shipping
aDuring the testing of plant D, the crystallizer was operating during a
fines cycle. The inlet loading to the scrubber was relatively high and
particle sizes were fairly small.
9.6-12
-------
9.6.3.2.1 Solution formation a'nd concentration. There are three
methods for producing urea: (1) once-through processes, (2) partial recycle
processes, and (3) total recycle processes. The most important of these
three classes is the total recycle process.
Process vents from the solution formation process are often scrubbed to
recover ammonia and other chemicals.! There are no data available regarding
the control or even the occurrence of particulate emissions from the solu-
tion formation process. ,
Plants using the total recycle process typically integrate the solution
formation and concentration steps to| conserve energy and to minimize emis-
sions. Particulate emissions from solution formation and concentration are
generally small but depend on the specific type of process. Uncontrolled
particulate emissions from three different plants range from 0.002 to
0.005 kg/Mg.l2 However, test data from another plant indicates an uncon-
trolled emission factor of 8.55 kg/Mg.10
There are two methods of concentrating urea solution prior to solid
formation: crystallization and evaporation. The method chosen depends on
the acceptable level of biuret [(NH2GO)2NH], an impurity formed by a side
reaction. To obtain technical grade.urea (<0.4 percent biuret), the more
expensive alternative, crystallation^ is necessary.
Typically, the crystallizers and the evaporators are operated under a
steam ejector vacuum. The airswept atmospheric evaporator is the only
type of evaporator that is a significant source of particulates. About 40
percent of the urea evaporators are controlled by condensation, 10 percent
by wet scrubbing, and 5 percent by demisting. The remaining 35 percent are
currently operating without control equipment.
Some evaporators are controlled 'by v«nturi scrubbers. Available data
show a particulate emissions rate, after such control, of 0.24 g/kg. The
exhaust from an evaporator controlled by a wet-scrubber may be recycled in
some instances, thus eliminating both' particulate and ammonia emissions.
9.6.3.2.2 Solids formation. There are essentially three methods of
producing solid urea: prill towers, !drum granulators, and pan granulators.
About 18 drum granulators are in operation in the United States. Drum
granulation accounts for roughly 50 percent of the solid urea produced.
9.6-13
-------
The remaining 50 percent is produced mainly by prilling, with only a small
percentage produced by pan granulation. Pan granulation is a recent develop-
ment and its overall cost is somewhat less than that of drum granulation.
Prilling of urea has emissions analogous to the prilling of ammonium
nitrate; the fumes are particularly difficult to eliminate. Uncontrolled
emissions are typically 1.0 to 3.0 kg/Mg of urea product.2*12
Approximately 45 to 50 percent of urea plants use wet scrubbers for
particulate emission control on prill towers. The other plants modify
production rates to meet State regulations. Many facilities can meet
existing mass emission rate regulations but have difficulty with opacity
standards because of the relatively large fraction of fine particles.
In the granulation technique, particles are built up to granules by
accretion. The granules produced are larger, with greater abrasion resis-
tance and two to three times the crushing strength of standard prills.
These properties result in less crushing, dust formation, and caking upon
handling.
Cooling air passing through the drum granulator entrains 15 to 20
percent of the product, but this airstream is smaller (approximately one-
third the airflow used in prill towers) and easier to treat than corre-
sponding prill tower airflows. With scrubbers, most of the process drum
granulator emissions are relatively low, generally in the range of 0.05 to
0.50 kg/Mg of product.4'12
The pan granulator consists of a tilted rotating circular pan. Feed
material desposited at the top falls through a fine spray of liquid urea,
and the larger granules thus formed spill over the lower edge of the pan
onto a conveyor belt. TVA has developed a low temperature (100 to 107°C)
process; Norsk-Kydro uses a high temperature (113 to 121°C) process. Cooling
is typically accomplished in a rotary drum cooler.
Emissions from pan granulators in the urea industry are expected to be
very low compared to those from drum granulators, based on a comparison with
similar units in the ammonium nitrate industry. Uncontrolled emissions from
one pan granulator in the ammonium nitrate industry were measured to be
1.36 kg/Mg compared to 138-152 kg/Mg for three drum granulators.1 Controlled
emissions can be much lower than those from drum granulators because of the
lower airflows.
9.6-14
-------
9.6.3.2.3 Product finishing. Urea product finishing covers cooling,
screening, incorporation of additives,; and coating. As the cooling drum
rotates, particulates are entrained. IParticulate emissions from this unit
are generally lower than those from a drum granulator. Uncontrolled
emissions from one rotary drum coolerlwere measured to be 3.90 kg/Mg.12
Wet scrubbers are commonly used to control cooler emissions. Industry test
measurements on four low to medium energy scrubbers controlling coolers
indicate an emission range of 0.01 tojO.l kg/Mg.12 The principal means of
product size control is screening. Oversize and undersize material is re-
moved from the product-size solids and recycled. Emissions from screening
are difficult to assess but are believed to be low.12
The primary purpose of coatings and additives is to reduce caking and
dust formation. The most common additives are formaldehyde and phosphate-
based compounds. In most plants, material commonly is transported by con-
veyors from one process step to another. Urea shipment is either by bag or
bulk. Data provided by industry indicate the uncontrolled emissions from
bagging operations are about 0.1 kg/Mg.12 The trend has been toward bulk
handling. Solution bulk shipment of urea is in tank cars.
It is somewhat difficult to predict or determine emissions from urea
product handling activities, since most are fugitive sources. For bagging
operations, a "worst case" value of 0.15 kg particulates/Mg of urea handled
has been estimated on the basis of the! fraction of fines. 13
9.6.3.3 Secondary Environmental 'Impacts. Wet scrubbing produces waste-
water that needs treatment. The treatment includes sending the effluent
stream to a hydrolyzer where the urea ;is decomposed.6
9.6.4 Pi ammonium Phosphate Fertilizer
Particulate emissions from the manufacture of diammonium phosphate
fertilizer include ammonia mist and solid diammonium phosphate; such emis-
sions amounted to 2.5 Gg in 1977.3 Both of these typically contain fluo-
rides, which are released from the phosphate rock and dilute phosphoric acid
scrubbing solution used to recover ammonia.14
9.6.4.1 Process Description. Dijammonium phosphate is formed by react-
ing ammonia gas with phosphate rock to/form a slurry. This slurry is further
9.6-15
-------
reacted in the granulator with ammonia and recycled diammonium phosphate to
form solid diammonium phosphate. The product is dried, cooled, and screened
prior to storage. Dilute phosphoric acid scrubbing is used to recover
ammonia from the reactor and granulator exhaust gases.
The granulator, dryer, and screens are the main emission sources of
solid particulates. Ammonia mist is emitted by the scrubbers on the
reactor.^ A flow diagram of diammonium phosphate production is given in
Figure 9.6-4-15 ancj stack parameter data are listed in Table 9.6-3.13
9.6.4.2 Emission Control Techniques. Ammonia emissions are very effec-
tively controlled by scrubbers using a dilute (30 to 40 percent) phosphoric
acid scrubbing solution. These scrubbers are either venturi or two-stage
cyclonic scrubbers. Venturi scrubbers are more commonly used because of
their higher efficiency in small particle removal.15 Most plants already
recover ammonia from the effluent gas streams because of the high cost of
ammonia. Therefore, ammonia emissions tend to be quite low. Emissions from
all sources at the plant can be treated by the phosphoric acid scrubbers,
but are usually passed through cyclones for product recovery first. Par-
ticulate efficiency data for these controls are not readily available.
9.6.5 Grain Handling and Storage
Grain handling operations include the shipping, receiving, weighing,
transfer and conveying, cleaning, drying, and storage of grains. These
operations are carried out at country and terminal elevators, as well as at
grain milling plants. The main pollutant emitted during grain handling
operations is grain dust; such emissions amounted to 670 Gg in 1977.3
Other emissions include sand, dirt, and trash. Although particulate, sizes
vary widely, 99.5 percent of the suspended particulate matter is less than 2
micrometers in diameter, and 50 percent is less than 0.03 micrometers.^6
The characteristics of the particles vary with the type of grain handled.
Corn handling produces "bees wings" emissions—large, low density particles
which are easily airborne. Dirt particulate matter, collected while
harvesting short-stemmed soybean plants, is emitted during soybean handling.
Country elevators are usually located within 15 to 30 km of the grain
fields and operate only during the harvesting season (June to November).16
9.6-16
-------
PHOSPHORIC ACID
I
^J
EMISSIONS
PHOSPHORIC
RECYCLE
If
AMMONIA
n
PREREACTOR
(REACTOR
GRANULATOR
| VENTURI
I SCRUBBERS
•-'1
I
_/*
I
PRODUCT TO
STORAGE
Figure 9.6-4 Diammonium phosphate production.is
-------
Table 9.6-3. DIAMMONIUM PHOSPHATE PRODUCTION STACK PARAMETER DATA13
Average Average Average
Number stack stack Average gas
Emission of height, diameter, temperature, flowrate,
sources sources mm °C
Dryer and cooler 59 26 1.3 64 20
Ammoniator and 71 26 1.1 53 14
granulator
9.6-18
-------
Approximately 85 percent of the grain, harvested is sent to country elevators;
the remaining portion is shipped to terminal elevators or grain mills. Termi-
nal elevators are located throughout the country, frequently in metropolitan
i
areas. These elevators operate year-round and have storage capacities aver-
aging 1.4 x 1Q5 cubic meters of grain1.!6
9.6.5.1 Process Description. Grain is unloaded from trucks, trains,
or barges into large receiving hoppers and is then conveyed to storage silos.
To prevent rotting during storage, some grains are dried in gas-fired rack
i
or column dryers. Corn and soybeans typically need to be dried. Before
being transported to the next purchaser, grain may be screened to remove
sticks, stones and other large trash.; It is then weighed and loaded into
trucks, trains, barges or ships. I
9.6.5.2 Emission Characteristics And Applicable Control Technology.
Industrial process fugitive particulate emissions are produced during all
phases of grain handling. The most common control strategy is to enclose
and hood the processing equipment or area with ventilation to cyclones and
filters (see Volume 1, Section 5). Cyclones used in this application operate
with pressure drops of 0.75 to 1.2 kPa and have control efficiencies of 85
to 95 percent.16 Fabric filters are more frequently used at terminal ele-
vators, which are usually located in metropolitan areas where control regula-
tions are more stringent. These filters typically use felted synthetic
fabric and operate at filtering velocities of between 3 and 4.6 m/min.15
i •
Unloading operations produce large quantities of airborne grain dust,
especially if power shovels or front end loaders are used to assist in the
• - i
process. Emissions can be reduced by|unloading the grain slowly, reducing
the free-fall distance, and enclosing;the unloading area with ventilation to
either cyclones or fabric filters.16 :
Industrial process fugitive emissions from conveyer belts occur
i
primarily at transfer points, when the grain is transferred from one belt to
the next. Placing a hood over the transfer point and venting to cyclones or
fabric filters reduce emissions from this source. The capture speed of
hoods over conveyer transfer points should be approximately 100 times the
speed of the conveyer belt to overcome the laminar layer of air conveying
particulates away from the hood.16 The same technique can be used to
9.6-19
-------
reduce emissions from screw conveyers, which tend to produce more emissions
than belt conveyers.
Emissions from both rack and column dryers can be controlled by vacuum
cleaned screens built around the dryer. Column dryers, in which the grain
flows down between two perforated metal sheets, emit less particulate matter
than rack dryers. Rack dryers have baffles or racks around which the grain
and hot air must flow, creating a cascading motion of the grain which results
in increased particulate emissions. The emissions from an uncontrolled
column dryer are approximately equal to the emissions from a rack dryer con-
trolled with a vacuum cleaned 50-mesh screen.
Typical controlled and uncontrolled emission factors for grain handling
operations are shown in Table 9.6-4, while average emission source character-
istics are summarized in Table 9.6-5.
9.6.5.3 Secondary Environmental Impacts. In general, all acceptable
control alternatives involve the use of dry type particulate matter collec-
tion devices; therefore, no liquid wastes are generated. With regard to
solid waste impacts, it is estimated that currently 68 percent of the particu-
late matter collected by emission control devices at elevators is returned
to the grain, 30 percent is sold for use in feed manufacturing, and 2 percent
is disposed of as solid waste. The additional particulate matter collected
by more efficient control devices can be sold as feed or landfilled.^
Energy requirements for systems to control particulate emissions are
proportional to the volume of air that must be moved, the pressure drop of
the systems, and the amount of time each system operates. Table 9.6-6 pre-
sents an estimate of the energy required to operate model control systems
installed on elevators of various typical sizes.
9.6.6 Grain Processing
Grain processing involves the operations of cleaning, drying, and
grinding. The milled grains, used mainly for flour production, include
wheat, dry corn, durum, rye, and oats. Information on wet corn or soybean
milling is in Sections 9.6.9 and 9.6.10. Since dry corn and wheat milling
are the largest segments of the grain milling industry, emissions from the
milling of these two grains will be stressed in this discussion.
9.6.6.1 Process Description. Grains may be contaminated with such
impurities as sticks, stones, metal, and dirt, which are removed by a variety
9.6-20
-------
Table 9.6-4. GRAIN HANDLING OPERATIONS EMISSION FACTORS
(COMPILED;FROM REFERENCE 16)
Process
Grain unloading
o Truck
o Railcar
o Barge
Screens
Cleaning
operations
Handling
operations
Dryers
o Rack
o Column
Grain loading
o Truck
o Railcar
o Barge
Emission factors,
(kg/Mg of grain)
)
Uncontrolled j Controlled
;
0.3 . 0.03
j 0.003
0.65 ' 0.05
0.0001
0.85 ! 0.1
; 0.0003
1.6 0.15
0.0015
3.0 0.3
0.007
3.0 0.05.
0.0001
2.0 0.015
j -0.025
0.13 '- 0.009
0.025
i
0.14 j 0.015
' 0.0005
0.14 • 0.015
0.0005
0.6 ' 0.03
0.0005
Control method
Cyclone
Fabric filter
Cyclone
Fabric filter
Cyclone
Fabric filter
Cyclone
Fabric filter
Cyclone
Fabric filter
Cyclone
Fabric filter
Vacuum cleaned
50-mesh screen
Vacuum cleaned
100-mesh screen
58-mesh screen
Use 12.7— mm diameter
perforations in
column sheeting
Cyclone
Fabric filter
Cyclone
Fabric filter
Cyclone
Fabric filter
9.6-21
-------
Table 9.6-5. FEED AND GRAIN ELEVATORS STACK PARAMETER DATA13
Emission
sources
Terminal
elevator
Shipping and
receiving
Transfer and
conveying
Screening and
cleaning
Drying
Country
elevator
Shipping and
receiving
Transfer and
conveying
Screening and
cleaning
Drying
Number
of
sources
959
840
328
339
5113
2666
1154
2430
Average
stack
height,
(m)
17
29
20
20
16
22
15
14
Average
stack
diameter,
(ra)
0.77
0.65
0.71
1.4
2.6
2.2
3.2
1.6
Average
temperature,
(°C)
23
29
21
52
22
22
22
30
Average gas
flowrate,
(Am3/s)
8.1
5.6
5.9
16
65
62
44
98
9.6-22
-------
Table 9.6-6. CALCULATED ENERGY REQUIREMENTS TO OPERATE ALTERNATE
GRAIN ELEVATOR CONTROL SYSTEMS*6
Facility
Energy required for
control system3,
kWh/yri
Percent increase in energy
required due to control system,
County elevator
(1 million bu/yr)
High throughput terminal
(3.5 million bu/yr)
Inland terminal
(15 million bu/yr)
Part terminal
(40 million bu/yr)
Process storage
System 1
32,000
112,000
619,200
716,400
840,000
System ;2
i
40,000
140,000
634,000
.763,0!00
1,050,000
System 3
42,000
147,000
636,500
766,000
1,052,000
System 1
12.5
12.5
22.5
18.2
21.5
System 2 -
15.6
15.6
23
19.4
26.9
System 3
16.4
16.4
23.1
19.5
27
aSystem 1 is designed to meet typical State standards and consists of:
o High-efficiency cyclones on all affectjad facilities (excluding dryers), except railcar
unloading at port terminals, barge and, ship loading at inland terminals, and barge and
ship unloading where fabric filter controls are required.
o No screens (filters) on column dryers and 20 to 30 mesh screens on rack dryers.
System 2 is designed to meet NSPS and consists of:
o Fabric filter control on all affected facilities excluding dryers.
o No screens (filters) on column dryers and 50 or finer mesh vacuum-cleaned screens on
rack dryers. j
o Three-sided shed on truck unloading and truck loading.
o Shed with two open ends for boxcar and|hopper car loading.
o Totally enclosed shed for railcar unloading.
o Totally enclosed leg for barge and ship unloading.
System 3 represents the best control technology possible not considering costs. System 3 is
identical to System 2 except for the following items:
o 100 mesh vacuum-cleaned screens (filters) on column and rack dryers.
o Totally enclosed sheds on truck unloading, truck loading, boxcar loading and hopper car
loading operations. '•
9.6-23
-------
of cleaning devices. Vibrating screens remove the larger trash while smaller
items are removed with magnetic separators, scourers, and disk separators.
Corn is frequently washed in a washing-destoning device and then dewatered.
Dust, chaff, dirt, and seeds may all be emitted during the grain cleaning
operation.
Drying removes excess moisture to prevent rotting and conditions the
grain for grinding. Particulate emission rates from grain driers are de-
pendent on several factors, including the type of dryer used and the grain
being dried.17 Oats are frequently dried in either a rotary steam tube
dryer or in 3 to 3.7 m diameter pan dryers which are stacked on top of each
other in a steam jacketed vessel. These dryers are generally not significant
sources of particulate emissions. Other grains are dried in column or rack
dryers. Column dryers, using recirculated air, tend to emit less particulate
matter than rack dryers.1? Corn dryers emit primarily large flaked particles
known as "bee's wings."
After drying, breakrolls and hammermills are used to remove the endo-
sperm from the bran and germ and then grind it into flour. The endosperm is
separated from the rest of the grain by sifters. In this process, the break-
rolls emit bran and grain dust while flour is emitted from other milling
equipment.
9.6.6.2 Emission Control Techniques. In 1977, particulate emissions
from the grain processing and milling industry amounted to 7.8 Gg.3 The
most serious particulate emission problems are created by the grain cleaning
equipment and the dryers.18 Grinding and milling operations also emit
particulate matter but to a lesser degree. Controls used in the drying and
cleaning processes are similar to those used at grain elevators during grain
handling operations. Venting the emissions to fabric filters provides the
best controT for cleaning-house and milling emissions; however, cyclones are
used more frequently. Milling emissions are also controlled by fabric fil-
ters and cyclones. Dryer emissions can be reduced by using self-cleaning
screens or by redesigning the dryer with small perforations in the dryer
walls rather than with louvered openings for air flow.15,17,18 (Refer to
Section 9.6.5 for more details on these controls.) Fabric filters may be
used to control dryer emissions that do not contain excessive moisture;
however, this is not often the case.
9.6-24
-------
9.6.6.3 Secondary Environmental Impacts. Devices used to control
emissions produced in grain milling have minimal secondary environmental
effects, except that additional poweif is required to operate them. The
solid waste collected by control equipment can be sold as animal feed or
used for landfill.
9.6.7 Alfalfa Dehydrating ;
Dehydrated alfalfa, used as animal feed, is produced at over 30 plants
in the United States.17 These plants are located primarily in Nebraska
and the northern plains States. Alfalfa dust, emitted in the drying and
grinding operations, is the major particulate pollutant. Dust emissions
range in size from 0.2 to 20 micrometers in diameter.19
9.6.7.1 Process Description. Alfalfa is fed to a rotary dryer, and
i
then pneumatically conveyed to a hammermill where it is ground into a fine
meal. This meal is bagged, stored, or pressed into pellets. Product re-
covery cyclones collect entrained alfalfa dust from the dryer and hammermill
(grinder) exhaust air streams. An alfalfa plant process flow diagram is
given in Figure 9.6-5. ;
Fine alfalfa dust is emitted by the primary cyclone, which follows the
dryer in the process. Other emission;sources include the grinder and bagger
cyclones, the pellet cooler, and the handling operations.
9.6.7.2 Emission Controls Techniques. More than 75 percent of the
particulate emissions from alfalfa processing are emitted from the dryer.18
The quantity and size distribution of,dryer emissions are highly dependent
on the quality of the alfalfa hay (i.e., the protein content, age, insect
damage, and moisture content).!9 Emissions from the dryer and from all
other sources are significantly increased when the alfalfa is overdried, a
common occurrence during periods of high productivity. The first step in
controlling dryer emissions is to maintain all equipment properly. Harvest-
ing equipment should have sharp blades and the dryer should be monitored to
ensure that the flame does not impinge on the green chops and that the chops
are not overdried. In addition, cyclones should be designed and maintained
to achieve the highest collection efficiency. Another way to reduce dryer
emissions is to recycle a portion of the primary cyclone exhaust back to the
dryer furnace, a step which also decreases the net amount of gases to be
treated if additional control is necessary. Finally, a medium energy wet
9.6-25
-------
Note:
Secondary cyclone collectors
may not be used and in some
cases the cyclone effluents
may be ducted back to the
primary cyclone.
CD
ro
Fresh Cut
Alfalfa (Green Chops)
from the field
\
Figure 9.6-5 Generalized flow diagram for alfalfa dehydration plant.17
-------
scrubber operating at a pressure drop of 1 to 1.25 kPa can be used to reduce
dust emissions from the dryer by at least 50 percent.20 Smoke emissions
from the dryer, typically between 0.2 ;to 0.5 micrometers in diameter, are
not effectively controlled by a medium energy scrubber.20 Several scrubber
!
designs have been tested at alfalfa plants; and Reference 20 describes these
systems in detail. ;
Hammermill cyclone emissions can.be controlled by using either a wet
scrubber or baghouse. If a scrubber is used, emissions from both the hammer-
mill and the dryer cyclones can be ducted to a common scrubber. With bag-
houses, a "burn out" loop of approximately 18 meters between the cyclone and
the baghouse should be used to reduce fire potential.I9 Emissions can also
be controlled by recycling a portion of the exhaust back to the hammermill.
Good engineering design and maintenance should be sufficient in control-
ling pellet cooler emissions. Industrial process fugitive emissions from
handling operations may be controlled :by enclosure and ventilation of the
area to a baghouse or scrubber. ;
Another technique to decrease the production of particles while increas-
ing their collectibility is liquid fat injection. Injecting 2 or 4 percent
fat prior to the hammermill has been shown to reduce fine particle formation
in addition to reducing total particulate emissions from the cyclones by
96.4 or 99.4 percent, respectively.^9 . The cyclones used to control emis-
sions from the grinders operate at pressure drops between 0.25 to 0.5 kPa.
Liquid fat injection requires additional equipment to store heat and inject
the fat.21
9.6.7.3 Secondary Environmental Impacts. The scrubbers used in treat-
ing dryer exhausts can produce approximately 3800 liters/min of sludge water,
creating a formidable disposal problem.21 Settling and evaporation ponds
can be used to concentrate the sludge.; The surface water can then be drained
off and used in field irrigation, while the built up sludge layer can be used
as fertilizer. Alternatively, the scr,ubber effluent can be clarified with a
chemical treatment of lime and synthetiic polymers, sodium hydroxide, or soda
ash.21
9.6.8 Cotton Ginning
Cotton gins separate cotton fiber
There were 2771 gins in the United States in 1976, located in the South and
9.6-27
(lint) from cottonseed and trash.
-------
Southwest.22 Gins located in Texas, California, and Missisippi handled
over half of the 2.3 Tg of cotton ginned in 1976.22 Although the ginning
season lasts from August through February, the average cotton gin operates
10 hr/day, 6 days/week for 10 weeks in the late autumn. The industry trend
is to extend this operating period by using fewer, larger gins which operate
continually.22
Emissions from cotton ginning consist of dust, fine-leaf trash, lint,
cotton dust, and other trash. The amount and nature of trash to be found in
the cotton depends heavily on the harvesting technique. Spindle picker
machines remove the cotton from the burrs with a rotating spindle, picking
up a minimum of dirt and leaves and leaving unopened bolls (pods) behind.
Stripper machines strip the cotton off the plants along with the burrs,
unopened bolls, leaves, stems, and soil. In arid areas of the Southwest,
where plants are smaller, shorter-stemmed, and have a lower yield, stripper
harvesting is used to minimize harvesting expenses. Most harvesting (65
percent) is accomplished with picker harvesters.22
Cotton dust may contain trace amounts of pesticides, such as DDT or
toxaphene, desiccants, or defoliants. Defoliants are sprayed on cotton
plants to reduce the amount of green leaves and stems-on the cotton stalk to
picker-harvested. Sodium chlorate and tributylphosphorotrithioates (DEF and
Folex) were the predominant (greater than 90 percent) defoliants used in
1971. A desiccant, generally arsenic acid or paraquat, is sprayed on cotton
fields to be stripper-harvested before a frost. This spray reduces the
moisture content of the stems and leaves. Uncontrolled emissions from this
industry were 15.6 Gg in 1977.3
9.6.8.1 Process Decription. Cotton ginning involves five basic steps:
(1) unloading, (2) seed cotton drying and cleaning, (3) overflow storage and
distribution, (4) lint cotton cleaning and handling, and (5) baling. Also
involved is the separation of lint from seed in gin stands. The cotton is
usually conveyed pneumatically from one step to the next. A simplified
diagram of ginning operations, showing emission sources, is presented in
Figure 9.6-6.
Each of these steps produces particulate emissions. Fine leaf trash and
dust are typically emitted from the unloading fan, inclined cleaner, trash
9.6-28
-------
EMISSIONS
EMISSIONS
missions
EMISSIONS
EMISSIONS
EMISSIONS
EMISSIONS
EMISSIONS
EMISSIONS
Figure 9.6-6 Typical ginning operation.22
3.6-29
-------
fan, and overflow system. Lint fly and cotton dust are emitted from the gin.
stands, lint cleaner condensers, battery condensers, and mote fan (a mote is
an immature seed with short immature fibers attached).
9.6.8.2 Emission Control Techniques. The partic.ulate emissions, pro-
duced during each phase of cotton ginning, are commonly reduced by cyclones
and filters. Wet scrubbers and, in some instances, process modifications
have also been used. The trash collected in these devices can be incinerated
or plowed back into the fields. Table 9.6-7 lists controls used for cotton
gin emission sources.23,24
Cyclones are the most frequently used particulate control device and
are used to reduce larger particulate emissions from the unloading fan,
inclined cleaners, extractor feeders, and gin stands. Even where other
controls are used, exhaust streams are typically first passed through a
cyclone. More than 90 percent of the cyclones used are small in diameter
(less than 0.96 m), high in efficiency, and capable of collecting more than
99 percent of particles which are larger than 125 micrometers.22 A typical
0.86 m diameter cyclone operates at a back pressure of between 1.0 kPa to
1.2 kPa.
Filters are used less frequently than cyclones. Filters are used pri-
marily for controlling lint cleaner and battery condenser emissions, but
recent studies have shown that they can be quite effective in reducing emis-
sions from all discharge points when such emissions are ducted to a common
filter. These filters consist of a fixed or rotating screen mounted in an
enclosed housing. The mesh can be cleaned by either a wiping arm or a line
of vacuum nozzles which continually remove trash. The air stream discharged
from the vacuum is conveyed to high efficiency cyclones. Collection effi-
ciencies for in-line filters are typically about 99 percent for lint fly and
80 percent overall on stripper-harvested cotton.22 The vacuum filter used
to control emission from all discharge points has a reported efficiency of
95 percent.25
Scrubbers and skimmer and spray columns are used less frequently in
controlling cotton gin emissions. As of January 1978, less than five cotton
gins used wet scrubbers.22 The reported collection efficiency of wet scrub-
bers for lint cleaner emissions ranges from 74.1 percent to 96.2 percent.25
Higher collection efficiencies can be achieved with skimmer and spray columns
which have reported efficiencies of 98.3 to 99.8 percent.
9.6-30
-------
Table 9.6-7. PARTICULATE CONTROL DEVICES APPLICABLE TO COTTON
GINNING OPERATIONS' (ADAPTED FROM REFERENCES 23,24)
Emission source
Control devices
Efficiency,
Unloading separator
Inclined cleaner
Extractor feeder and
gin stands
Lint cleaner
Battery condenser
Process heater
Cyclone, process modification 99
Cyclone, wet scrubber 72 to 96
Cyclone 99
Wet scrubber, in-line filter 72 to 96
Wet scrubber, in-line filter 72 to 96
[
(No controls used at present)
9.6-31
-------
Ginning emissions can be reduced by process modifications which involve
the proper adjustment of harvest equipment so that less trash is entrained
in the cotton during harvesting. Stack parameter data for typical emission
sources are presented in Table 9.6-8.
9.6.8.3 Secondary Environmental Impacts. Solid waste collected by the
use of filters and cyclones can be incinerated or used as landfill. Incin-
eration, however, generates additional air pollutants such as particulate
matter, nitrogen oxides, and sulfur oxides. Wastewater from wet scrubbing
devices needs to be treated to avoid the odor produced by excessive bacte-
rial growth.
9.6.9 Starch Manufacture
The starch industry includes primarily corn wet milling, potato starch
milling, and wheat starch milling processes. The corn wet milling process,
the largest of the three operations, produces corn starch, speciality
starches, corn syrup, high fructose corn syrup, animal feed by-products,
and corn oil. Wheat milling processes yield animal feed, starch, flour,
and wheat gluten for making breads. The potato starch industry manufactures
both potato starch and animal feed. Starch products, although primarily
used in the food industry, are also used in paper, textiles, absorbents,
insecticides, plastics and in rubber.
There are 24 corn wet milling plants in the United StatesJ? These
plants produced approximately 9.8 x 10^ megagrams of corn products in
1977.17 The particulate emissions produced during starch production are
primarily grain dust, sand, dirt, and fine starch particulate matter.
9.6.9.1 Process Description. Starch is separated from corn in a
process consisting of three basic steps: (1) corn steeping, (2) milling,
and (3) dewatering. Steeping the corn in hot water conditions the grain for
further processing, softens the kernel, and helps to break down the protein
(gluten) which contains the starch particles. After the water is evaporated,
the corn is ground, washed, screened, and centrifuged to separate the starch
from the fibrous corn material. The spent corn is then dried and sold as
animal feed (gluten feed and meal). The starch slurry from the centrifuges
is dewatered in vacuum filters or basket centrifuges and then dried. Between
40 and 70 percent of this starch is used in corn syrup and sugar production;
9.6-32
-------
Table 9.6-8. COTTON .GINNING STACK PARAMETER DATA13
Average Average Average
Number stack stack Average gas
Emission of height, diameter, temperature, flowrate,
sources sources m m °C Am3/s
Unloading fan 870
Cleaner 837
Stick and burr 847
machinery
5.8 0.83
6.5 0.89
5.8 0.71
21
22
21
8.1
8.9
6.0
9.6-33
-------
the rest is bagged for shipment.18 Figure 9.6-7 shows a flow diagram of
corn wet milling processes.26
9.6.9.2 Emissions and Controls. Particulate emissions produced in
the corn receiving and handling operations are controlled by the strategy
described for grain handling operations in Section 9.6.5. See Table 9.6-9
for uncontrolled and controlled emissions.27 Emissions produced from
gluten feed, gluten meal, germ, and starch dryers consist of both solid
particles and possibly aerosols formed from condensible gases. Product
recovery cyclones, typically used to collect valuable particulate matter
from the dryer exhaust streams, are actually the source of emissions. High
energy cyclones operating with efficiencies ranging from 95 percent to 98
percent are usually used to reduce dryer emissions.18 Emissions from
starch and animal feed dryers can be reduced further by using a wet scrubber
'after the cyclone. Scrubbers used in this application achieve efficiencies
of 98 to 99 percent.18 Baghouses may also be used for starch dryers if
adequate precautions are taken to prevent explosions. Other particulate
emission sources—the feed pellet mills, feed house, and dry starch
grinders—can be controlled with fabric filters (99+ percent efficent) J8
Stack parameter data for. typical emission sources are given in Table 9.6-10.
9.6.10 Vegetable Oil
Vegetable oil is extracted from soybeans, cottonseeds, peanuts, and
corn. Soybean oil constitutes more than 80 percent of the vegetable oil
market; therefore, Section 9.6.10 is limited to this segment of the vege-
table oil industry. Processing techniques and emission control techniques
for the other types of vegetable oils are similar.
There are approximately 100 soybean processing plants in the United
States, 88 of which crush soybeans nearly 100 percent of their processing
time. Eighty-three of the 100 plants are full-time solvent extraction
operations and are owned and operated by approximately 30 companies.
Processing plants are located in 21 States. Iowa and Illinois are the
two largest soybean processing States with 14 and 13 processing plants,
respectively. Geographically, processing plants are found in the Midwest,
South, and Middle Atlantic regions of the United States.
Domestic soybean production increased from 15 million metric tons in
1960 to 41.4 million metric tons in 1975 and up to 50.1 million metric tons
9.6-34
-------
Steep Water Evaporation
I
Steepwater Concentrate
Degerminator Mills
I
Hydroclone Separators
Germ
Dryer —»J[)ust and Vapor|
Starch, Gluten, Feed
Grinding Mills Oil Extraction
Washing Screens
I
r i
— —— — Feed Gluten
••••iMH^MMM^^
JDust and Vapors}-**- Dryers Dryers- >.|0ust and Vapors|
Centrifugal Separators Genn Heal Corn Oil
Slurry Starch
• To Refining
Acid or
Aci d/Enzymes
Gluten Feed Gluten Meal
| D'jst|<— Dryer
Dextrose
Acid Roasting.
Dextrins
Acid
Amines
Oxidation
Cooking
Modified Starch
IDustM Dryer
Dry Starch
Evaporation
Coirmon
Starch
Clarifier
Enzymes->•
Corn Syrup
Fructose
Ton-exchange*^
Resins
55S Fructose
Ion-exchange^
Resins
Storage and I Load in I Storage and Storage and I Load in
Emissions"^ Bagging Bag9ing->-| Emissions
90S Fructose
Clarifier
I
HFCS
Figure 9.6-7. Corn welt milling flow diagram.
9.6-35
-------
Table 9.6-9. EMISSION FACTORS FOR STARCH MANUFACTURING27
EMISSION FACTOR RATING: D
Particulates,
Type of operation kg/MT Ib/ton
Uncontrolled
Controlled3
4 8
0.01 0.02
aBased on centrifugal gas scrubber.
9.6-36
-------
Table 9.6-10. STARCH MANUFACTURING STACK PARAMETER DATA13
Dryer stack parameters Average value
Number of sources ' 109a
Stack height, m 16
Stack diameter, m ; 0.61
Stack temperature, °C 43
Gas flowrate, Am3/s 6.12
aRepresents number of sources surveyed.
9.6-37
-------
in 1978/79. Soybean production is expected to expand to 58 million metric
tons by 1985. U.S. soybean oil production is projected to be slightly less
than 5.6" million metric tons by 1985.
With the soybean processing industry currently operating at an 80 to 85
percent capacity, a large portion of the projected growth in crushing will
likely take place in existing soybean mills by their expanding to 90 to 95
percent operating capacity. The remainder of the growth will be in new soy-
bean plants, and an increase of about one new plant each year is projected.
9.6.10.1 Process Description. Processing techniques used in manufac-
turing vegetable oils from soybeans by the solvent extraction method include:
cleaning, drying and storage processes which are described in Section 9.6.5;
cracking and dehulling; conditioning and flaking; solvent extraction and
desolventizing-toasting; meal drying and cooling; and meal grinding, screen-
ing and loadout.
Soybeans are usually cracked into six or eight parts by a cracking mill
before dehulling. Cracking mills consist of two pairs of rolls approximately
30 cm (11.8 in) in diameter and 130 cm (51.18 in) long. Rolls this size are
capable of cracking 449.6 metric tons per day of soybeans. Proper cracking
aids the dehulling, extraction, and flaking processes, permitting the release
of more oil.28 A particulate recovery device such as a baghouse is usually
adapted to the cracker.
The purpose of dehulling and removing the hulls from the solvent ex-
traction process stream is to prevent hulls from absorbing and retaining
soybean oil through the remainder of the process. The dehulling process
consists of air suction and screening to separate hull from meat by density.
(The suction air is passed through a particulate collector.) Once separated,
the hulls may be ground and used for feed. The hulls are screened and passed
over a gravity table cyclically to screen out larger particles. The gravity
table typically has a particulate collection device such as a baghouse
adapted to it. The hulls are heated, ground, and then ground again with
meal derived from extraction.29 Additional particulate collection devices
may be added to the two systems as needed.
Conditioning is a process in which the dehulled soybeans are raised to
a temperature of approximately 72°C (160°F), and the moisture content is
9.6-38
-------
adjusted to 10 to 11 percent in a horizontal, steam-jacketed tube.29 By
optimizing temperature and moisture content, the soybeans obtain a plas-
ticity that permits good flaking.
Flaking mills compress the soybeans into flakes with thickness of from
0.020 cm (0.008 in) to 0.030 cm (0.012 in). Flaking causes rupturing of oil
cells and increases porosity, thus decreasing extraction time and desolventi-
zation time. The large surface area and ruptured cells increase the chances
of oxidation and hydrolysis. Thus, as flake thickness increases, the rate
of extraction decreases significantly.
Upon completion of flaking, the flakes should have a 10 percent mois-
ture content and a temperature of approximately 60°C (140°F). The condi-
tioner and flaker usually have their own particulate control system, e.g.,
cyclones.
The soybean flakes are conveyed to a continuous solvent extraction
system. Solvent, most commonly solvent grade hexane, flows countercurrent
to soybean flakes, extracting soybean oil. The solvent-oil mixture, mis-
cell a, is transferred to other stages of processing where oil is recovered
from the miscella and eventually refined.
The solvent-wet flakes are transported to a desolventizer-toaster. In
this device, solvent is steam-evaporated out of the flakes for recovery, and
the flakes are .toasted. :
Following desolventization, the flakes proceed to a dryer with natural
or forced draft that typically operates at a temperature of 110°C (230°F).
The dryer decreases meal moisture from 20 percent to less than 12 percent.
A common dryer type is the rotary steam tube dryer. This dryer type is
favored for large capacity processing; it requires less floor space and is
more economical than dryers such as the vertical stack dryer. A particulate
control device such as a cyclone may be an optional addition.
The dried meal exits the dryer at a temperature of 110°C (230°F) and
is transported to a cooler that reduces its temperature to less than 40°C
(104°F). A pneumatic cooling system, : a rotary cooler, and the French louver
meal cooler are three coolers commonly used. A cyclone is usually added to
the cooler.
The meal proceeds through a cyclic system of grinding and screening to
assure proper particle size for meal flakes. The ground meal is then prepared
9.6-39
-------
for shipment (which may include bagging) and conveyed to a loadout area.
Meal may be transported by truck, rail, or ship. Cyclones or baghouses are
usually applied in these areas to recover dust which could escape during the
processes.
Particulate emission point sources in soybean processing plants may be
either controlled or uncontrolled. Some process equipment at plants vent
particulate emissions directly to the atmosphere without any form of con-
trol; but, more often, process equipment emissions are vented through low
to medium energy cyclones, which are used primarily for product recovery.
For example, a medium energy cyclone may be added to a meal cooler because
of the economic incentive to recover effluent product. These low-to-medium
energy cyclones are not the best choice for emission control and would
result in an incidental, perhaps marginal gain in particulate emissions
control.
Fugitive particulate emissions—which are those particulate emissions
that are not collected by either controlled or uncontrolled venting—are
also found in the soybean and soybean meal preparation areas of processing
plants. Sources of fugitive emissions include maintenance openings in the
solids handling areas and leaks in the solids transport lines.
Fugitive particulates from soybean and hull storage, handling, and
grinding processes often have particulate matter larger in size than particu-
late matter emitted from cyclones and baghouses. These emissions can be
decreased by eliminating window and roof openings and by ventilating the
work-area air through baghouses and cyclones. However, no data are available
on actual quantities of fugitive dust emissions in the soybean processing
industry.
9.6.10.2 Emission Control Techniques. The current practice in soybean
processing plants is to rely almost exclusively on cyclones and fabric filters
to control particulate emissions. (Screens are often incorporated into the
bean drying apparatus to control emissions at that point.) However, wet
scrubbers have been used in the past and may continue to be used at some
processing plants in the future. Soybean processing plants feature enclosed
conveyors and extensive hooding with pneumatic conveying systems leading to
dust collectors. The performance of these systems is partially determined
by maintenance and operational practices, and there appears to be consider-
able variance in the effectiveness of systems among different plants.
9.6-40
-------
Although visible emissions are rare, there is accumulated particulate matter-
around some plants, suggesting either that emissions do occur on occasion or
that particulate matter accumulates during repairs and is not subsequently
cleaned up. Particulate matter may also accumulate because source control
measures such as covers, enclosures, and regulated air flows are circum-
vented by improper use or maintenance.. By contrast, other plants appear to
have very little accumulation of dust.
The bulk loadout of meal is a particulate source that appears difficult
to control, and the success of the control apparatus is largely dependent on
operational practices. A typical truck loadout, for example, would be located
in an enclosed building with large pick-up ducts distributed overhead, espe-
cially in the vicinity of the meal loadout duct. The loadout duct itself
may be fitted with a plastic- or canvas-type hood that lowers with the duct
'to the loadout point; the hood would have dust collection ducts. The col-
lecting ducts would be routed to a fabric filter. The successful operation
of this control system would depend upon operational practice, e.g., whether
or not the doors were closed and the canvas hood lowered. Similar opera-
tional considerations apply to railcar and barge loadout points. Illustra-
tive sketches of typical systems can be found in Reference 30. In each
case, the control apparatus is a fabric' filter.
Beginning -with the receipt of cleaned and dryed whole soybeans from
storage, particulate sources within a soybean processing plant include crack-
ing and dehulling, hull grinding, soybean conditioning, flaking, meal drying,
meal cooling, and bulk loading. (Other potential sources that could be
listed separately are cleaning or scalping prior to cracking and meal grind-
ing.) Dust from these sources is collected in a variety of combinations and
routed to control devices. A 1972 study of particulate emissions from soy-
bean processing plants30 estimated that!, in a typical plant, about 75 percent
of the particulate sources were controlled with cyclones and 25 percent with
fabric filters. Further, this study estimated the adequacy of those controls
at 6 on a scale' in which 0 represents complete inadequacy and 10 represents
complete adequacy. Recent plant visits.suggest that the use of fabric fil-
ters has increased significantly in recent years with an attendant increase
in adequacy.
9.6-41
-------
Figure 9.6-8 shows a schematic diagram of a 1,000 ton per day soybean
plant that highlights the particulate issue. It includes typical flows
through what is asserted to be a "best" control system.30 With the excep-
tion of the wet scrubber on the meal dryer effluent, it is very similar to
systems observed in plants recently visited.
Particulate emissions from soybean production are primarily hull and
seed dust. The size range and distribution are not known. A range of 2 to
50 micrometers in diameter has been estimated for hull and seed dust emis-
sions of other oil seeds.31
Cyclones are used extensively both for product recovery and for control-
ling emissions in soybean processing. Cyclones are classified as either high
efficiency or high throughput (less efficient) units. The less efficient
units are less expensive and are used primarily for product recovery. Less
efficient units should be followed by high efficiency cyclones or fabric
filters for adequate particulate emissions control. A typical efficiency
for collecting grain dust is 95 percent for the high throughput units. Modi-
fied high efficiency units (e.g., the Aerodyne Type S) has an estimatd col-
lection efficiency for grain dust of 99 percent.30
Fabric filters are currently used to control several soybean processing
particulate emission sources. The only sources where they are not used are
those in which the effluent has a high moisture content—i.e., the bean con-
ditioner, the flaker, and the dryer. Blinding of the fabric occurs when the
collected dust is moist, thus reducing the flow of air through the unit.
Also, the cleaning mechanisms (shaker, reverse air, pulse jet) generally will
not remove wetted particulates from the fabric surface.
Cotton sateen, wool felt, and dacron felt have all been used as filter
fabrics for grain industry emission sources. However, dacron felt is the
fabric now recommended by essentially all of the filter manufacturers for
these sources. Fabric weights of 16 to 22 ounces per square yard are
typical.30 .
The air-to-cloth ratio, sometimes referred to as the filter rate, is a
key design parameter for fabric filters. Typical values for soybean plants
are about 3 meters per minute.
Bag life is on the order of 18 to 36 months and varies with the type of
cleaning cycle. A complete cleaning of the bags (i.e., dry cleaning) is
9.6-42
-------
Plant Capacity - 1,000 tons/day
CTl
I
Bean Influent
from storage
•A- High Energy Cyclone
FF - Fabric Filter
WS - Het Scrubber
Miscella
Figure 9.6-8. Schematic of a 1,000 ton per day soybean processing plant.
30
-------
sometimes required to restore original bag efficiency and operating charac-
teristics. Complete cleaning of the bags would occur about once or twice a
year in a plant with a good preventive maintenance program..
Fabric filters, when properly designed and operated, can operate rela-
tively trouble free with efficiencies in excess of 99.9 percent and with no
visible emissions.
Although wet scrubbers are applicable to soybean processing sources,
they offer limited use and therefore are rarely used. The limited use of wet
scrubbers is due to the following reasons:
o The material collected by a wet scrubber is rarely suited for
reuse in the process.
o Use of a wet scrubber requires treatment of the scrubbing liquor
effluent to prevent water pollution and sanitation problems.
o Particulates are the only emissions of concern from most sources.
If a level of control higher than that attainable by a cyclone is
required, a fabric filter can usually be used.
For these reasons new installations are using cyclones and fabric filters
rather than wet scrubbers, and some processing plants are converting from
scrubbers to cyclones and fabric filters.
9.6-44
-------
REFERENCES FOR SECTION 9.6
1. Memo from Noble, E., to files, U.S. Environmental Protection Agency,
Office of Air Quality Planning and Standards. November 14, 1980.
Draft chapters 3 and 4, prepared for a background information document
on the ammonium nitrate industry.
2. Search, W. J., and R. B. Reznik. Source Assessment: Ammonium Nitrate
Production. U.S. Environmental Protection Agency. Publication No.
EPA-600/2-77-107. September 1977.
3. OAQPS Data File of Nationwide Emissions. U.S. Environmental Protection
Agency, Office of Air Quality Planning and Standards. February 1979.
4. Hineman, P. S., and P. Spaun. An Evaluation of Control Needs For The
Nitrogen Fertilizer Industry. U;S. Environmental Protection Agency.
Publication No. EPA-600/2-79-186. August 1979. EPA-600/2-77-107.
September 1977. -
5. Ammonium Nitrate Emission Test Report. C. F. Industries. Harrison,
TN. U.S. Environmental Protection Agency. EMB Report 79-NHF-10.
November 1979. . ,
6. The Fertilizer Institute Environmental Symposium. January 13-16, 1976.
7. Environmental Considerations of Selected Energy Conserving Manufac-
turing Process Options, Volume XV: Fertilizer Industry Report. U.S.
Environmental Protection Agency. Publication No. EPA-600/7-76-034.
December 1976.
8. Chemical Economics Handbook, Stanford Research Institute. Menlo Park,
CA. .December 1976.
9. Ammonium Sulfate Manufacturing Background Information Document for
Proposed Emission Standards. U.S. Environmental Protection
Agency, Office of Air Quality Planning and Standards. Research
Triangle Park, NC. Publication No. EPA-450/3-79-034.
December 1979.
10. Chemical Marketing Reporter. Schnell Publishing Co. New York, NY.
October 13, 1980.
11. Search, W. J., and R. B. Reznik. Source Assessment Urea Manufacture.
U.S. Environmental Protection Agency. Publication. No.
EPA-600/2-77-107L. November 1977.
12. Memo from Noble, E., to files, U.S. Environmental Protection Agency,
Office of Air Quality Planning and Standards. Draft chapters 3 and 4
prepared for a background information document on the urea industry.
13. Atmospheric Modeling Data from National Emission Data System (NEDS).
U.S. Environmental Protection Agency, Office of Air Quality Planning
and Standards. May 1979.
9v6-45
-------
14. Parker, A. Industrial Air Pollution Handbook. McGraw Hill.
15. Background Information for Standards of Performance: Phosphate
Fertilizer Industry, Volume 1: Proposed Standards. U.S. Environmental
Protection Agency. Publication Mo. EPA-450/2-74-019a. October 1974.
16. Standards Support and Environmental Impact Statement, Volume 1:
Proposed Standards of Performance for Grain Elevator Industry.
U.S. Environmental Protection Agency.
17. Shannon, L. J., et al. Emissions Control in the Grain and Feed
Industry: Volume 1—Engineering and Cost Study. U.S. Environmental
Protection Agency. Publication Mo. EPA-450/3-73-003a.
18. Schrag, M. P., et al. Source Test Evaluation for Feed and Grain
Industry. U.S. Environmental Protection Agency. Publication Mo.
EPA-450/3-76-043.
19. Annis, J. C., et al. Evaluation of the Liquid Fat Injection Method
of•Controlling Alfalfa Dehydrator Effluent. Journal of the Air
Pollution Control Association. January 1970.
t
20. Smith, K. D. Particulate Emissions from Alfalfa Dehydrating Plants—
Control Costs and Effectiveness. U.S. Environmental Protection Agency.
Publication No. EPA-650/2-74-007. January 1974.
21. First, M. W., et al. ControT of Odors and Aerosols from Spent Grain
Dryers. Journal of the Air Pollution Control Association. 24:7,
July 1974.
22. Rawlings, G. D., and R. B. Reznik. Source Assessment: Cotton Gins. U.S.
Environmental Protection Agency. Publication No. EPA-600/2-78-004a.
January 1978.
23. Feairheller, H. R., and D. L. Harris. Particulate Emission Measure-
ments from Cotton Gins—Plant Tested: J. G. Boswell Company El Rico
#9. Corcoran, CA. Monsanto'Research Corporation. Report No. 72-MM-19.
Submitted November 1974. Undated.
24. Rawlings, G. D., and H.B.M. Cooper. Air Pollution Control at Cotton-
seed Oil Mills. (Presented at Specialty Conference on Control
Technology for Agricultural Air Pollutants. Air Pollution Control
Association. March 1974.)
25. Roddy, W. J. Controlling Cotton Gin Emissions. Journal of the Air
Pollution Control Association. June 1978.
26. Draft Report, Source Category Survey for the Starch Manufacturing
Industry. Energy and Environmental Analysis, Inc. Durham, NC.
U.S. Environmental Protection Agency. Contract Mo. 68-02-3061.
April 1980.
27. Storch, H. L. Product Losses Cut With A Centrifugal Gas Scrubber.
Chem. Eng. Progr. 62: 51-54. April 1966.
9.6-46
-------
28. Control of Volatile Organic Emissions from Manufacture of Vegetable
Oils. U.S. Environmental Protection Agency Guidelines Series. U.S.
Environmental Protection Agency, Office of Air Quality Planning and
Standards. Research Triangle Park, NC. Publication No.
EPA-450/2-78-035. June 1978. p. 2-2.
29. Becker, K. W. Processing of Oilseeds to Meal and Protein Flakes.
Journal of the American Oil Chemists' Society. 48:300.
June 1971.
30. Shannon, L. J., et al. Emission Control in the Grain and Feed Industry.
Volume 1. Engineering and Cost Study. U.S. Environmental Protection
Agency. Research Triangle Park, NC. Publication No.EPA-450/3-73-003a.
December 1973.
31. Feairheller, W. R., and D. L. Harris. Particulate Emission Measurements
from Cotton Gins—Plants Tested; U.S. Environmental Protection Agency.
Research Triangle Park, NC. EPA Project Report No. 72-MM-19.
9.6-47
-------
Page Intentionally Blank
-------
9.7 MINERAL PRODUCTS ;
The conversion of naturally occurring minerals into salable products
involves both physical and chemical, processes. These processes, if inade-
quately controlled, can result in the generation and emission of signifi-
cant quantities of particulate matter. Several unit processes (e.g., mining,
crushing, screening, drying, materials handling, and mixing) are common to
many mineral production procedures. Mineral ores are usually extracted and
processed to some degree (beneficated) before being shipped to another loca-
tion where further processing occurs or a salable product is manufactured.
The mining and benefication of both nonmetallic and metallic minerals and
the corresponding emission sources and applicable control technologies are
discussed in Sections 9.7.1 and 9.7.2, Processes and applicable control
technologies for certain mineral products are discussed in Sections 9.7.3
through 9.7.13; the products discussed are those which are subjected to
additional production processes which have a potential for generating and
emitting particulate matter.
9.7.1 Nonmetallic Mineral Processing
Crushed stone, sand and gravel, and coal are by far the largest segments
of the nonmetallic mineral and processing industry, accounting for an annual
U.S. production rate of over 2000 teragrams. Other minerals produced in
large quantities are clay, phosphate, Islag, rock salt, gypsum, sodium com-
pounds, potash, pumice, borate, and barite. Production rates and other
pertinent data are listed in Table 9.7-1.1j2,3
Nonmetallic mineral processing involves extracting from the ground,
loading, unloading and dumping, conveying, crushing, screening, milling, and
classifying. Some mineral processing also includes washing, drying, cal-
cining, or flotation operations. The operations performed depend on the rock
type and the desired product. Essentially all mining and mineral processing
operations are potential sources of particulate emissions. The 1977 particu-
late emissions from the mining and processing of a number of nonmetallic
minerals are given in Table 9.7-2.4 The reported quantities include emis-
sions from numerous operations ranging from the mining of the raw materials
to the manufacture of a salable product from the listed commodity.
9.7-1
-------
Table 9.7-1. HONtCTALLIC MINERALS PRODUCTION STATISTICS^2.3
I
ro
Mineral
Reported 1975
production level,
Major producing States
in order of production
Number of
active operations
Estimated 1985
Annual production
growth rate, level,
% Gg
Crushed and
broken stone
Sand and gravel
Coal
Clay
Phosphate
Slag
Rock salt
Gypsum
Sodium compounds3
Potashb
Pumice
Asphalt and
related bitumens
Talc
Borate
Barite
Fluorspar
Feldspar
Pyritesc
Diatomite
Per lite
Vermiculite
Mica
Kyanite
819
716
588
44
44
27
14
8
4
2
3
1
1
,020
,100
,000
,490
,300
,120
,928
,845
,530
,270
,530
90
875
,060
,165
125
610
367
520
640
233
122
85d
PA,
AK,
KY,
IN,
GA,
FL,
UT,
PA,
TX,
CA,
CA,
NM
OR,
UT
VT,
CA
NV,
IL
NC
TN
CA,
NM
MT,
NC,
VA,
IL,
CA,
WV,
WY
TX,
NC,
MT
OH
NY,
MI,
TX
CA,
TX,
MO
KS,
SC
NM
GA
TX, FL, OH
MI, IL, TX, OH
PA, IL, OH, VA,
OH, NC
TN, ID, WY,
LA
IA, TX
AZ, NM
CA
NV
4800 (quarries)
5500 (plants)
5247 (mines)
120
36
21
69 (mines)
1
11
235
1
52
6
31
15
15
3
16
13
2
17
3
4
1
.0
.0
NA
3
4
2
2
2
2
3
3
2
4
5
2
3
4
.5
.0
.0
.0
.0
.5
.0
.5
.tie
.0
.0
.2
.0
.0
1,212
790
773
62
33
16
10
5
3
4
1
1
1
,620
,925
,000
,750
NA
,060
,510
,780
,800
,050
,980
110
,300
,080
,450
170
900
Neg.
5
4
4
4
6
.5
.0
.0
.0
.0
890
950
445
180
155
aNatural soda ash.
''Potassium equivalent.
C8ased on sulfur production by assuming a 40 percent content.
Estimate for 1974.
eAssumed.
NA—not available.
-------
Table 9.7-2. 1977 NATIONWIDE PARTICIPATE EMISSIONS FROM THE NONMETALLIC
MINERALS MINING AND PROCESSING INDUSTRY4
Mineral Emissions,
Gg
Sand and gravel 41
Stone and rock 1372
crushing
Phosphate rock
Drying 18
Grinding 24
Material handling 8
Clays 114
Lime 115
Perlite 2
Coal cleaning
Pneumatic dry 1
Thermal dry 12
Gypsum 70
9.7-3
-------
9.7.1.1 Process Description. Nonmetallic mineral mining includes
surface mining, dredging, and underground mining. Surface mining involves
removing overburden, recovering the desired mineral, transporting the mineral
to the processing operation where the ores or minerals may be beneficated,
and reclaiming the mined area. Deposits recovered by this method may be in
any rock type and are usually less than 150 meters deep. Some minerals
require blasting while others can be removed by bulldozer. Dredging is a
type of placer mining (mining of glacial or alluvial deposits containing
minerals) that involves the underwater extraction of minerals from placer
deposits. Underground mining involves the extraction and removal of a
mineral from a natural deposit beneath the earth's surface.
The nonmetallic minerals are normally delivered to the processing plant
by truck and dumped into a hoppered feeder (usually a vibrating grizzly type)
or onto screens, as illustrated in Figure 9.7-1. The screens are used to
separate materials that do not require further crushing from those materials
that require processing. Crushing and screening operations are conducted in
stages until the materials are of the proper size. At this point, some
mineral end products of the desired grade are conveyed directly to finished
product bins, or stockpiled in open areas by conveyors or trucks.
Most nonmetallic minerals require additional processing, depending on
the rock type and product requirements. Some minerals, especially certain
lightweight aggregates, are washed, dried, and sintered or treated prior to
primary crushing. Others are dried following secondary crushing or milling.
Sand and gravel, crushed and broken stone, and most lightweight aggregates
normally are not milled but are screened and shipped to the consumer after
secondary or tertiary crushing. Minerals such as talc or barite may require
air classification to obtain the required mesh size, and treatment by flota-
tion to obtain the necessary purity and color. Figure 9.7-2 shows a simpli-
fied diagram of the typical process steps required for nonmetallic mineral
processing. Table 9.7-3 lists the various unit process operations and emis-
sions sources for each industry.5
Most mining and mineral processing operations are potential sources of
particulate emissions. Two categories of emission sources have been defined:
process fugitive emission sources and fugitive dust sources. Process
9.7-4
-------
SURGE PILE
FINISHING
SCREENS
Figure 9.7-1 Flowshfeefc .of a typical crushing, pi ant.
9.7-5
-------
SECONDARY
CRUSHER
SIZE
CLASSIFIER
STOCKPILE
OR BIN
13
STOCKPILE
OR BIN
#4
Figure 9.7-2 General schematic for nonmetallic minerals processing.i
9.. 7-6
-------
Table 9.7-3. NONMETALLIC MINERAL UNIT OPERATIONS AND POSSIBLE SOURCES OF EMISSIONSM.S
Type of plant
Crushed 8 broken stone
Sand 8 gravel
Coal
Clay
Phosphate
Gypsum
Pumice
Feldspar
Boron
Talc
Barite
Diatomite
Lightweight aggregate
Rock salt
Fluorspar
Gilsonite
Sodium compounds
Mica
Kyanite
Crushers
X
X
X
X
X
X
X
X
X
.X
X
X
X
X
X
X
X
X
X
Screens
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Transfer
points
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Grinders
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Loading
operation
. X
X
X
x •
X
X
Bagging
operation
X
X
X
X
X
X
X
X
Dryers or
calciners
X
X
X
X
X
X
X
X
X
X
X
X
X
Drilling
operation
X
X
X
X
X
X
-------
fugitive emission sources include any source where emissions can be collected
and controlled. Examples include drilling, crushing, screening, grinding,
conveyor transfer points, and loading operations. Fugitive dust sources are
not amenable to control by conventional systems. Source operations include
blasting, hauling, and conveying; source sites include the mine, haul roads,
stockpiles, and the plant yard.
The common factors that affect emissions from all mineral processing
operations include the type (hardness) of ore processed, the moisture content
of the ore, the type of equipment employed, the amount of ore processed, and
various geographical and seasonal factors. In conjunction with these common
factors, operation-specific factors also affect emissions from each mineral
processing or mining operation. In drilling operations, these factors in-
clude the type of drill used, the diameter of the hole, and the penetration
rate. Emissions from blasting depend on the size of the shot, blasting
practices, mineral type, and meteorological conditions (especially wind).
During screening operations, the level of uncontrolled dust emissions depends
largely on the particle size of the material screened and the amount of
mechanically induced energy.^
9.7.1.2 Emission Control Techniques. The diverse particulate emission
sources in mineral mining and processing operations require the implementa-
tion of a variety of control methods and techniques. Dust suppression tech-
niques, designed to prevent particulate matter from becoming airborne, are
applicable to both process fugitive emissions and fugitive dust sources.
When process fugitive emissions can be contained and captured, collection
systems may be used. Emissions sources and applicable control options avail-
able in the nonmetallic mineral mining and processing industry are listed in
Table 9.7-4.
Methods used to reduce emissions include wet dust suppression, dry col-
lection and a combination of the two. Wet dust suppression systems control
dust by applying moisture in the form of water or water plus a wetting agent
sprayed at critical dust producing points in the process flow. This method
is designed to prevent particulate matter from becoming airborne. The method
is used extensively to control emissions from fugitive dust sources, but it
is also used for fugitive process emissions.
9.7-8
-------
Table 9.7-4. PARTICULATE EMISSION SOURCES FOR THE
EXTRACTION AND PROCESSING OF NONMETALLIC
MINERALS1
Operation or source3
Control options
Drilling
Blasting
Loading (at mine)
Hauling
Crushing
Screeni ng
Conveying (transfer points)
Stockpiling
Grinding
Storage bins
Conveying (other than
transfer points)
Windblown dust from
stockpiles
Windblown dust from roads
and plant yard
Loading (product into RR
cars, trucks, ships)
Bagging
Magnetic separation
Liquid injection (water or water
plus a wetting agent)
Capturing and venting emissions to a
control device
Adopt good blasting practices
Water wetting
Water wetting of haulage roads
Treatment of haulage roads with
surface agents
Soil stabilization
Paving
Traffic control
Wet dust suppression systems
Capturing and venting emissions to a
control device
Same as crushing
Same as crushing
Stone ladders
Stacker conveyors
Water sprays at conveyor discharge
Pugmill
Same as crushing
Capturing and venting to a control
device
Covering
Wet dust suppression
Water wetti ng
Surface active agents
Covering (i.e., silos, bins)
Windbreaks
Water wetting
Oiling
Surface active agents
Soil stabilization
Paving
Sweeping
Wetting
Capturing and venting to control device
Capturing and venting to control device
Capturing and Venting to control device
aDoes not include processes involving combustion.
9.7-9
-------
9.7.1.2.1 Wet dust suppression. The wet dust suppression method has
been used on a wide variety of stone including limestone, traprock, granite,
shale, dolomite, and sand and gravel. Water sprays are not practical in all
cases because moisture may interfere with further processing such as screen-
ing or grinding, where agglomeration cannot be tolerated. The capacity of
dryers used in some processing steps may limit the amount of water that can
be sprayed onto the raw materials.
A typical wet suppression system utilizes several points within the
process flow scheme when the suppressant material is applied. These points
may be at the primary crusher truck dump, at the discharge of the primary
crusher, at secondary and tertiary crushers, at feeders located under surge
or reclaim piles, at screens, at conveyor transfer points, and at conveyor
screen discharges to bins and storage piles. If properly conditioned at the
initial processing steps, continued application of the wetting agent can be
minimized. The wetted material should exhibit some carry-over dust control
effect that will last through a number of material handling stages.
In order to enhance the effectiveness of the wet suppression technique,
wetting agents are added to the water to lower its surface tension and con-
sequently improve its wetting efficiency. These chemical wetting agents
have the following advantages:
o Lower cost because of less frequent application.
o Better wetting of fine particles and longer retention of the
moisture film.
o Flexibility of application, since the agents can be applied
directly to the surface being controlled or worked into the
material being treated.
o Greater reduction of particulate emissions from fugitive sources
—up to 90 percent efficiency versus 50 percent efficiency with
untreated water.6
Due to the unconfined nature of emissions from facilities controlled by
the wet suppression technique, the quantitative.measurement of mass particu-
late emissions is not possible. Thus, no mass emission data are available
which permit a quantitative comparison of the control capabilities of wet
dust suppression versus dry collection techniques. Visible emission obser-
vations were conducted by EPA at six crushed stone, and sand and gravel
9.7-10
-------
facilities (which EPA designated F, P, Q, R, S, T); EPA used wet dust
suppression techniques to control particulate emissions generated at these
plant process facilities. Emissions generated by 13 crushers, 14 screens,
7 transfer points, 1 impact mill and 1 storage bin were visually measured
by EPA Methods 9 and 22. Facilities R and T are portable crushing facili-
ties. Facilities P, Q, R, and T process crushed limestone, while facility
F processes crushed traprock, and facility S produces crushed granite.
The results of the tests for non-crushing sources (e.g., screens, trans-
fer points, and storage bins) are summarized in Table 9.7-5. These results
indicate that visible emissions occur less than 10 percent of the time. The
results of the tests for crushing sources from the best controlled station-
ary (facility S) and portable (facility R) plants are summarized in Figures
9.7-3 to 9.7-7. The data are reported in 6-minute averaging of Method 9
data. For each testing set (approximately 1 hour), the results of the two
observers simultaneously measuring visible emissions are indicated by a
solid and a dashed line. Even though facility R is designated the best con-
trolled portable crushing plant, the secondary crusher exceeded 15 percent
opacity several times, according to one observer. This is attributed to the
fact that during the test, there was no spray bar located near the crusher
outlet. Had the spray bar for the crusher been relocated closer to the
crusher than its present position some 5 feet from the crusher, emissions
might have dropped below 15 percent opacity for all observer readings.
In general, the positioning and number of spray bars at all plants
except facilities F and S were judged to be inadequate. Even though all the
facilities tested were reasonably controlled, they might have achieved the
low level of emissions seen at facilities F and S with more careful placement
of the bars, and in some cases, additional spray bars.
9.7.1.2.2 Dry collection. Particulate emissions from process operations
(such as dust from crushing, screening equipment, and transfer points) may
be controlled by capturing and ducting the emissions to a collection device.
Collection systems consist of an exhaust system utilizing hoods to capture
emissions, fan and ducting to convey the emissions to a collection device,
and the collection device to remove the particulate emissions prior to ex-
hausting the air stream. Collection devices include fabric filters, cyclones,
and low-energy scrubbers.
9.7-11
-------
TABLE 9.7-5. SUMMARY OF VISIBLE EMISSION MEASUREMENTS FROM FUGITIVE NONCRUSHING SOURCES
CONTROLLED BY WET SUPPRESSION (ACCORDING TO EPA METHOD 22)1
Accumulated
observation
Rock type Date of Process time,
Plant processed test facility minutes
P Crushed limestone (S)b 10/02/79 Secondary screen
Transfer point
Q Crushed limestone (S) 10/10/79 Three process screens
R Crushed limestone (P)c 10/15/79 Three process screens
Two transfer points
S Crushed granite (S) 10/23/79 Two process screens
Two transfer points
T Crushed limestone (P) 10/29/79 Process screen
Transfer point
Storage bin
60
60
270
210
120
240
240
120
120
120
Accumulated
emission
time,
minutes9
0
< 1
2
11
1
10
< 1
0
3
0
Percent of
time with
visible
emissions
0
1
< 1
5
< 1
4
0
0
2
0
aData from observer with highest readings.
b(S) = Stationary plant.
C(P) = Portable plant.
-------
18
16
14
12
10
15 percent OPACITY
o
a
0.
O
10 percent OPACITY
6 -
4
OBSERVER I
OBSERVER
/•«*-OBSERVER 2
OBSERVER 2
SET I
SET 2
0 -
L 1
12 24 36 48 60
12 24 36 48 60 0
TIME., minutes
Figure 9.7-3. Summary of visible emission measurements from best controlled fugitive primary
crushing source (portable-facility T) by means of wet suppression (according
to EPA Method 9).1
-------
18
OBSERVER 2
•OBSERVER 2
15 percent OPACITY
c
V
u
8
O
OBSERVER I
10 percent OPACITY
OBSERVER I
SET I
SET 2
12 24 36 48 60
12 24 36 48 60
TIME, minutes
Figure 9.7-4 Summary of visible emission measurements from best controlled fugitive secondary
crushing source (portable-facility R) by means of wet supression (according
to EPA Method 9). l
-------
<£>
e
0
O
t
O
-------
CD
18
16
14
_ 12
c
o
u
I 10
§ 8
0.
O
6
4
2
15 ptrcent OPACITY
10 perc«nt OPACITY
/V^
OBSERVER 2
A
OBSERVER 1
SET i
j i
SET 2
12 24 36 48 60
12 24 36 43 60
TIME, minutes
Figure 9.7-61 Summary of visible emission measurements from best controlled fugitive primary
crushing source (stationary-facility S) by means of wet supression (according
to EPA Method 9).1
-------
!8r
16
15 percent OPACITY
14
*. 12
u
a 10
OBSERVER I
OBSERVER 2
10 percent OPACITY.
a.
o
OBSERVER
OBSERVER
SET I
SET 2
24 36 48 60
24 36 48 60
TIME, minutes
Figure 9.7-7 Summary of visible emission measurements from best controlled fugitive secondary
crushing source (large, secondary-facility S) by means of wet supression
(according to EPA Method 9).1
-------
To maximize collection system efficiency, exhaust systems must be prop-
erly designed and balanced. Process equipment should be enclosed as com-
pletely as possible, allowing access for operation and maintenance. Indraft
velocities should be maintained at a minimum of 61 meters per minute.^ Prop-
er hood design will minimize the effects of cross drafts and reduce power
consumption by minimizing exhaust volumes. Ducting must be designed for
adequate conveying velocities to prevent settling of dust particles. Based
on data for the crushed stone industry, conveying velocities recommended for
mineral particles range from 1100 to 1400 meters/min.7 For proper dust con-
trol from process sources, hoods should be installed at conveyor transfer
points, screens, crushers, grinders, and bagging operations. Hood configura-
tions and exhaust rates are discussed in Reference 1.
Fabric filters are generally the most effective dust collection devices
used in the mineral mining and processing industry. In most crushing plant
applications, mechanical shaker type collectors which require periodic shut-
down for cleaning (usually after 4 to 5 hours of operation) are used. Col-
lector bags are made of cotton sateen and operated at a filtering velocity
of 0.6 to 0.9 m/min. When it is impractical to turn off the collector for
cleaning, fabric filters with continuous cleaning are used. Fabric filters,
using wool or synthetic felts as filtering media, provide continuous cleaning
and may be operated at a filtering velocity of 1.8 to 3.0 m/min. Table 9.7-6
gives standard air-to-cloth ratios suggested for the collection of different
material dusts. Efficiencies for well maintained baghouses, regardless of
type (jet pulse or mechanical shaker), are greater than 99 percent, even on
submicrometer particle sizes.7
Particulate emission measurements were conducted by EPA on 16 baghouse
collectors used to control emissions generated at crushing, screening, and
conveying (transfer points) operations at five crushed stone installations,
one kaolin plant, one fuller's earth installation, and on one baghouse col-
lector used to control emissions generated at grinding, classifying, and
fine product loading operations at a feldspar installation. Table 9.7-7
briefly summarizes the process operations controlled by each baghouse tested,
along with specifications for each baghouse. The results of these measure-
ments are summarized in Figure 9.7-8.
9.7-18
-------
Table 9.7-6. AIR-TO-CLOTH RATIOS FOR FABRIC FILTERS USED FOR EXHAUST
EMISSION CONTROLl
filtering velocity,9
Industrial segment meters/min
Sand and gravel 2.1
Clay 1.8
Gypsum 1.8
Lightweight aggregate 2.3
(Perlite and
Vermiculite)
Pumice 1.4
Feldspar 1.2
Borate 1.5
Talc and soapstone 1.5
Barite 1.4
Diatomite 1.2
Rock salt 1.5
Fluorspar 1.5
Mica 1.8
Kyanite 1.8
Sodium compounds 1.4
Gilsonite N.R.b
Crushed and broken stone 7.0
aRatio is based on operating surface required to
obtain a particulate concentration of 0.05 g/nP in
the outlet stream from the filter. In all cases,
the filter is a pulse-jet type operating at
1.5 kPa w.c. differential pressure. The filtering
medium is felted polypropylene or polyester.
bNo recommendation for this segment.
Q 7_1 Q
-------
TABLE 9.7-7. BAGHOUSE UNITS TESTED BY EPA*
Rock type
Facility processed
Al
A2
A3
A4
Bl
B2
i° Cl
--j
8 C2
Dl
D2
El
E2
Ml
M2
61
LI
L2
Limestone
Limestone
Limestone
Limestone
Limestone
Limestone
Limestone
Limestone
Traprock
Traprock
Traprock
Traprock
Fuller's earth
Fuller's earth
Feldspar
Kaolin
Kaolin
Baghouse specifications
Type
Jet pulse
Jet pulse
Jet pulse
Jet pulse
Shaker
Shaker
Shaker
Shaker
Shaker
Shaker
Jet pulse
Jet Pulse
Reverse air
Reverse air
Reverse air
Unknown
Unknown
Filtering
ratio
5.3 to 1
7.0 to 1
7.0 to 1
5.2 to 1
3.1 to 1
2.1 to 1
2.3 to 1
2.0 to 1
2.8 to 1
2.8 to 1
5.2 to 1
7.5 to 1
6.0 to 1
5.2 to 1
3.0 to 1
Unknown
Unknown
Capacity
sens
12.5
7.5
1.1
5.0
2.7
8.6
3.5
3.1
15.0
' 12.3
7.0
10.0
0.9
1.6
1.9
6.6
3.3
scfm
,(26,472)
(15,811)
(2,346)
(10,532)
(5,784)
(18,197)
(7,473)
(6,543)
(31,863)
(25,960)
(14,748)
(21,122)
(1,800)
(3,300)
(3,960)
(14,040)
(6,960)
Process operations controlled
Primary impact crusher
Primary screen
Conveyor transfer point
Secondary cone crusher, screen
Primary impact crusher
Scalping screen, secondary cone crusher, two
finishing screens, hammer mill, five storage
bins, six conveyor transfer points
Primary jaw crusher, scalping screen, hammer mill
Two finishing screens, two conveyor transfer points
One scalping and two sizing screens, secondary
crusher, two tertiary cone crushers, several
conveyor transfer points
cone
Finishing screen, several conveyor transfer points
Two sizing screens, four tertiary cone crushers
several conveyor transfer points
Five finisheing screens, eight storage bins
Raymond mill system
Fluid energy system
9
Pebble mill, bucket elevator, two conveyor transfer
points, fine product loading
Raymond impact mill
Roller mill
-------
0.02
+,
o
£
o
f~l
3
O
•2 0.015
(0
•o
c:
(O
1/1
s-
•o
i-
Q.
£0.01
ra
s-
O)
-
t2
o
1— <
GO
GO
1 — 1
J , 1
w 0.005
_1
1— 1
•a:
O-
0
_
_
-
&
i
j_
r* ~
1
-1
KEY
eft AVERAGE
V
• EPA TEST METHOD
0 OTHER TEST METHOD
ft
-
) — )
^
«
1 ' Q
ii J-L Qr>
(^ II , ,
V ^ i^
^ '
1 1 T 1 1 1 1
•
|D ••
II
1
1
1
I
1
y
« ft i
C 1 F
n If |
1 1 i * i
.ft. ft i
fC-Lj H
ii 'n* |
1 1 1 1 LJ
1 1 i i nj
W II
ll
• « w f!
'
Mi i
O€ hU<
U|
1 1
1
1
ft !• Pi
1 .
i gto w ' L
1
i fi w M
i ^ [1
' fi Hfj1
M-M 99 |
II hnH ' 1
U W oo
r 1 1 1 I I 1 1 1 I
On /i c
.046
0.034 ^
i)
<^~>
0)
E
O
2
3
O
T-,
s-
(O
•o
0.023 S
t/)
>>
"O
s-
Ol
n.
(/>
(O
s-
0.011 °*
0
Facility AT A2 A3 A4 Bl B2 B3 Cl C2 01.02 El E2 Gl Li L2 Ml. N2
Rock Type LLLLLLLLLTTTTFKKFEFE
Figure 9.7-8
Parti cul ate emissions from non-metal He minerals
processing operations.1.
9.7-21
-------
Observations for visible emissions were also made at hoods and enclo-
sures to record the presence and opacity of emissions escaping capture.
The results of these measurements are summarized in Table 9.7-8. In most
instances, essentially no visible emission were observed at adequately
hooded or enclosed process facilities.
The other collection devices used in the industry, cyclones and low-
energy scrubbers, have efficiencies of 95 to 99 percent for coarse particles
(greater than 40 micrometers). These efficiencies, however, drop to less
than 85 percent for particles under 20 micrometers in size.1
Wet dust suppression and dry collection techniques are often used in
combination to control particulate emissions. A typical control method
combines wet dust suppression at primary crushers, screens, transfer points,
and crusher inlets, with dry collection at the discharge of secondary and
tertiary crushers, where new dry surfaces and fine particles are formed. A
specific combined method controls fugitive dust from abandoned tailing dumps
by using combinations of water, chemical stabilizers, and vegetative cover.
Efficiency is reported at 90 to 95 percent.6 For drilling operations, water
injection or aspiration connected to-a control device such as a cyclone or
fabric filter, preceded by a settling chamber, is common.8
For crushing, screening, and transfer operations, controls include the
use of wetting.agents and fine water sprays at critical transfer and unload-
ing points.
9.7.1.3 Secondary Environmental Impact. The utilization of dry collec-
tion techniques (particulate capture combined with a dry emission control
device) for control generates no water effluent discharge. In cases where
wet dust suppression techniques could be used, the water is absorbed by the
material processed. Thus, wet suppression systems do not result in a water
discharge.^
The impact of solid waste upon the environment depends on the type of
control used. With the use of fabric filters, approximately 1.4 Mg of solid
waste is collected per 250 Mg of mineral processed. 'Often this material can
be recycled back into the process, sold, or used for a variety of purposes.1
If no market exists for the collected wastes, they^are often discharged to
the tailings pond or disposed of in the mine or an.isolated location in the
9.7-22
-------
TABLE 9.7-8. SUMMARY OF VISIBLE EMISSION MEASUREMENTS FROM FUGITIVE SOURCES AT NON-METALLIC MINERAL PLANTS!
ro
oo
Accumulated
Date of observation
Plant Rock type processed test Process facility time,
minutes
A Crushed limestone 7/9/75 Baghouse discharge to conveyor
Primary impact crusher discharge
Conveyor transfer point
B Crushed limestone 7/1/75 Scalping screen
Surge bin
Secondary cone crusher No. 1
Secondary cone crusher No. 2
Secondary cone crusher No. 3
Hammer mill
3-deck finishing screen (L)
3-deck finishing screen (R)
6/30/75 Two 3-deck finishing screens
0 Crushed stone 7/8/75 No. 1 tertiary gyrasphere
cone crusher
No. ?. tertiary gyrasphere
cone crusher
Secondary standard cone crusher
Scalping screen
Secondary (2-deck) sizing screen
Secondary (3-deck) sizing screen
F Traprock 8/26/76 Two tertiary crushers
Four processing screens
Conveyor transfer points
240
240
166
287
287
231
231
231
287
107
107
120
170
170
170
210
210
210
65
180
179
Accumulated
emission
time,
minutes
0
4
3
45
3
23
0
0
0
4
0
86
0
0
0
0
0
0
0
0
/
0
Percent of time
with visible
emissions
0
1
2
15
1
10
0
0
0
4
0
72
0
0
0
0
0
0
0
0
0
(continued)
-------
TABLE 9.7-8. (concluded)
vo
ro
Accumulated
Date of observation
Plant Rock type processed test Process facility time,
minutes
6 Feldspar 9/27/76 Conveyor transfer point No. 1
Conveyor transfer point No. 2
Primary crusher
Secondary crusher
Conveyor transfer point No. 4
Ball mill (feed end)
Ball mill (discharge end)
Indoor transfer point No. 1
Indoor transfer point No. 2
Indoor bucket elevator
Truck loading
Rail car loading
80
87
60
60
84
60
60
60
60
60
.13
32
Accumulated
emission
time,
minutes
0
0
1
0
0
0
0
0
0
0
0
5
Percent of time
with visible
emissions
0
0
2
0
0
0
0
0
0
0
0
15
H Gypsum
I Mica
J Talc
10/27/76 Hammer mill
9/30/76 Bagging operation
10/21/76
Vertical mill
Primary crusher
Secondary crusher
Bagger
Pebble mill
298
60
90
90
150
150
90
0
20
4
13
6
0
22
3
9
7
Kaolin
12/7/78 Rail car loading
Test 1
Test 2
Test 3
144
99
154
17
2
9
12
2
6
-------
quarry. The amount of solid waste produced from a typical crushing plant
(544 Mg/h) is about 28 Mg over an 8-hour period. This waste represents less
than 0.7 percent of the plant throughput and does not create a significant
solid waste disposal problem.1
The implementation of emission control measures in the nonmetallic
mineral mining and processing industry results in an increase in energy
usage. The energy requirements both with and without air pollution controls
for six typical plant sizes are shown in Tables 9.7-9 and 9.7-10. (See
Reference 1 for additional details and assumptions.)
9.7.2 Metallic Minerals Mining and Processing
The metallic mineral mining and processing industry involves the mining
and processing of ores containing metals such as iron, aluminum, copper,
lead, zinc, silver and gold. This industry is highly diversified, employing
a variety of extraction, benefication, and emission control techniques. The
principal extraction methods include underground mining, surface mining, and
dredging. Major emission sources at the mine site include haul roads and
the drilling and blasting operations. Benefication processes involve the
crushing, classification, concentration, and drying of the metallic ores to
enhance their usefulness. Air pollution emissions in the mineral mining
industry consist primarily of particulate matter emitted from the above
processes.
Emissions may be characterized as process, process fugitive, or fugitive
dust. (See Table 9.7-11 for the type of emissions generated by ore proces-
sing operations.) Process emissions are point source emissions generated by
a process operation; they are normally captured and ducted to the atmosphere
as part of the process. Process fugitive emissions are also point source
emissions generated by a process operation; however, they are normally not
captured and ducted to the atmosphere as part of the process. Fugitive dust
emissions are non-process emissions usually generated by wind action or the
passing of vehicles; they are emitted from non-point sources. Control tech-
niques for these emissions are shown in Table 9.7-12.
Factors that affect emissions from most mineral operations include the
hardness of the ore processed, the type of equipment and operating practices
9.7-25
-------
Table 9.7-9. ENERGY REQUIREMENTS FOR MODEL NONMETALLIC MINERALS PLANTS
HAVING CRUSHING AND GRINDING OPERATIONSl
(kilowatts)
Plant size, Fabric filter Percent energy increase,
Mg/hr Uncontrolled controlled %
9 187 216 16.0
23 410 445 8.4
68 895 953 6.5
136 1567 1649 5.2
272 3134 3289 4.9
544 6342 6626 4.5
9.7-26
-------
Table 9.7-10. ENERGY REQUIREMENTS FOR MODEL WONMETALL1C MINERALS PLANTS
HAVING CRUSHING OPERATIONS ONLY1
(kilowatts)
Plant size, Fabric filter Percent energy increase,
Mg/hr Uncontrolled controlled %
9 48 69 43.8
23 150 174 16.0
68 300 345 15.0
136 375 435 16.0
272 750 856 14.4
544 1570 1761 12.4
9.7-27
-------
Table 9.7-11. TYPES OF EMISSIONS GENERATED BY METAL
ORE PROCESSING OPERATIONS
Process operation Emission type
Drying
Pelletizing Process
Calcining
Crushing
Screening
Conveying (transfer points) Process fugitive
Storage in bins
Dry grinding
Product loading
Wind action on the following:
Haul roads
Stockpiles Fugitive dust
Wastepiles
Tailings pond beach and dam areas
9.7-28
-------
Emission
category
Table 9.7-12. METHODS OF PARTICIPATE EMISSION CONTROL USED
AT METALLIC MINERAL PROCESSING PLANTS ,
Soiirce
Method of control
Dust collection
Dry Wet
cyclone scrubber Baghouse ESPa
Water
Water wetting Soil
Enclosure wetting plus stabilizer paving
i surfactant
ro
Crushers
Screens
Conveyor belt transfer
Process points
fugitive Storage bins
Dry grinders
Product packaging
Product loading
Process
Dryers
Calciners
Pelletizers
Stockpiles
Fugitive Wastepiles
dust Haul roads
Tailings pond beach
and dam areas
X
X
X
X
X
X
X
X X
X
X X
X
X
X
X
X
X
X
X
X
X
x.
X
aElectrostatic Precipitator.
-------
employed, the moisture content of the ore, the amount of ore processed, and
a variety of geographical and seasonal factors. Dust suppression techniques
for preventing particulate matter from becoming airborne are used to control
both fugitive dust and process fugitive point sources. Collection systems
are used to control point sources where particulate emissions can be con-
tained, captured, and removed from exhaust streams.
Table 9.7-13 presents industry characteristics for each of the metals
considered. Ore production in a given year depends primarily on current
market value for the particular metal. Production at a specific mine, how-
ever, depends on the cost/benefit ratio which depends on such factors as
location and extent of the ore body (i.e., surface or underground), metal-
lurgical complexities of extraction process, distance to the market, and
costs of environmental compliance. Copper ore is by far the largest quantity
produced; iron ore and titanium/zirconium sandtype are second and third,
respectively. These ores alone represent 85 percent of the total.10 ,
9.7.2.1 Process Description. Figure 9.7-9 displays the three phases
of metallic mineral mining and processing. Phase II, the extraction process,
and Phase III, ore processing or benefication, are described below. Particu-
late emissions from these processes are also cited.
9.7.2.1.1 Extraction processes. Drilling, blasting, ore loading, and
ore transporting are common to all mining procedures (i.e., underground and
surface mining, dredging). Drilling consists of boring blast holes into the
mineral deposits. The holes are subsequently charged with explosives and
detonated. Blasting is used to displace minerals from their deposits and to
fragment them into sizes that require a minimum of secondary breakage and
allow ease of handling and loading. The excavation and loading of broken
minerals is normally performed by shovels and front end loaders.
At underground mines, crude ore is transported to the surface in buckets,
trucks, conveyers, or cars called "skips." Ore is transported to the skip by
conveyors or hauling vehicles. At many surface mines, large haulage vehicles
with a capacity of 18 to 68 megagrams are used to transport minerals from
the mine to the primary crusher. 10,11
In general, a surface mine operation includes removing the overburden
material that covers the deposit, removing the mineral being recovered,
9.7-30
-------
TABLE 9.7-13. METALLIC MINERALS PROCESSING INDUSTRY CHARACTERISTICS^,
Metal
Aluminum
Antimony
Beryl 1 i urn
Copper
Gold
Iron
Lead/Zinc
Molybdenum
Nickel
Silver
Titanium/Zirconium
Tungsten
Uranium
Vanadium
Zinc
Major
AL,
ID,
UT
AZ,
AZ,
MI,
CO,
CO,
OR
CO,
FL,
CA,
CO,
AK,
ID,
processing
States
AR, TX, LA
MT
MT, NM, UT
NV, SD, UT
MN
ID, MO UT
NM
ID
NJ
NV
NM, WY
CO, ID
MO, NJ
Number of
active
operations
9
2
1
34
17
31
35
3
1
9
5
4
30
3
35
Production,
MT
2,0139
1,0969
__b
1,3649
34.29
56.7a
5373
55.53
13,0163
1,228
~b
2.733
8,344C
4.723
4083
aMine production 1977.
bConfidential.
GMine production 1976.
9.7-31
-------
(PHASE I !
BEFORE MINING I
PHASE 11
MINING
! PHASE III 1
| AFTER MINING . |
10
CO
FINDING
PROVING
PLANNING
OPENING AND
DEVELOPING
DRILLING
SAMPLING
SHAFTING AND/OR TUNNELING
EXTRACTION
SURFACE
UNDERGROUND
DREDGING
BREAKING
LOADING
TRANSPORTING
UNLOADING
ORE TO
PROCESSING
PROCESSING
TO FURTHER PROCESSING
OR CONSUMER PRODUCTS
SIZE REDUCTION
SCREENING
CLASSIFYING
DESIGN AND ENGINEERING
SHAFT SINKING AND TUNNELING
CLEARING AND GRUBBING
STRIPPING
UNDERGROUND AND SURFACE
CONSTRUCTION
DEWATERING
THERMAL DRYING
LEACHING
Figure 9.7-9 Scope of mining activities.2
-------
transporting the mineral to the benefication processing operations and
reclaiming the disturbed area. Dredging is a type of placer mining (a
subcategory of surface mining.) that involves the underwater extraction of
minerals from placer deposits (alluvial or glacial deposits of sand and
gravel containing particles of valuable minerals). The deposits dredged are
usually low-grade and lie near the surface.10,11- The major particulate pol-
lutant in these operations is fugitive dust emissions. Drilling, blasting,
loading, hauling, dumping, storage piles, waste piles, overburden removal,
wind erosion of unprotected surfaces, and land reclamation activities all
contribute to fugitive dust and process fugitive emissions.
Particulate emissions from drilling operations are caused primarily by
air flushing the bottom of the hole to remove cuttings and dust. Factors
affecting emissions from blasting include the size of the shot, blasting
procedures, rock type, and meteorological conditions. Considerable fugitive
dust emissions may result from loading, hauling, and dumping operations.
The most significant factor affecting emissions during loading is the wetness
of the ore. Factors affecting emissions from hauling operations are type of
road surface, wetness of the surface, volume and speed of vehicle traffic,
and vehicle characteristics. Storage piles are another major source of
fugitive dust. The emissions from waste and tailing piles are similar in
mineral identity to those from primary storage piles, but because the par-
ticles are finer, they travel further.
9.7.2.1.2 Metallic mineral preparation processes (beneficiation). A
mineral deposit commonly contains many distinct mineral phases which are
closely interlocked to form the ore. Thus, mineral processing begins by
freeing the desired mineral from the undesired gangue by pulverizing and
concentrating it. A schematic diagram of mineral processing is shown in
Figure 9.7-10. Figure 9.7-11 displays a process flow diagram for the
beneficiation process.
An open or closed circuit may be used in ore treatment. Open circuits
use screens to divert finer ores to subsequent stages rather than passing
all of the ore through one crusher. This practice lessens the mechanical
burden on a piece of equipment, improves operating efficiency, and lowers
maintenance requirements. Closed circuit design, which is still used in
9.7-33
-------
oo
Figure 9.7-10 General schematic for initial steps in ore processing.
12
-------
I
CO
en
SIZE
REDUCTION
SCREENING
CLASSIFICATION
CONCENTRATION
PRITORV CRUSHING DRY SCREENING WCHANICAI EROTH FLOTATION
SECONDARY CRUSHING UET SCREENING HYDROCLONES GRAVITY CONCENTPATION
DRY GRINDING MAGNITIC SEPARATION
HET GRINDING FIlnRCISTATIC SIPARATION
EXTRACTIVE HETAl.HIRGY
AGGIOHIRATION
1 (ACHING
ION EXCHANGE
DEWATERING
THE DUAL DRYING
SCREENS ROTARY DRYING
CENTRirur.rs FLASH DRYING
ClASSIEILRS CONIINI10IIS TRAY DRYING
SLnlHtNIAIION TLUlDI/rD BED DRYING
riLIRATION
ElOfUILATION
ff. ORE
CONCENTRATION
Figure 9.7-11 Benefielation processes.2
-------
many plants, requires that every piece of ore be passed through each unit
of equipment in the process.
Metallic mineral processing typically consists of several stages of
crushing, screening, conveying, separating, bagging and bulk loading, and
waste disposal. These processes are described below along with their
emission characteristics.
Crushing—Crushing is the operation of reducing the crude ore to the
fineness necessary for mechanical separation or metallurgical treatment.
Primary crushing reduces the ore to a maximum size of 10 to 15 cm; it is
often accomplished at the base of the ore shaft in underground mines.
Secondary crushing is usually a surface operation to reduce the ore further
to about 2.5 cm.^ In some ore dressing sequences, tertiary crushing may
be included, although it is common practice to slurry the ore after the
secondary stage and introduce it to a wet grinding stage. There are a
number of mechanisms employed in the size reduction of mineral ores, and
these mechanisms are exploited to various degrees depending on the crush-
ing equipment used. Crushing equipment types include jaw, gyratory, impact,
and roll crushers.
Particulate emissions are unavoidable in crushing operations; therefore,
deliberate control methods are usually employed. The level of particulate
emissions before control depends on the moisture content of the incoming
ore, the type of crusher used, the type of ore processed, and the ore size
range at the crusher exit. Particle size depends on the crushing mechanism
employed and the amount of energy imparted to the fines. Impact-type
crushers produce more fines and impart more energy to the particles than do
compression crushers. These crushers consequently generate larger quanti-
ties of particulate emissions per unit weight of ore processed than any
other type of crusher.12 Rotating crushers produce an initial draft which
further increases particulate emissions through dust entrainment.
Screening—Screening is used only for relatively coarse material at the
primary and secondary size reduction stages because the rate of screening
becomes considerably slower at fine mesh sizes. Industrial dry screening is
rarely carried out below 20 mesh. Screening surfaces may be constructed of
metal bars, perforated or slotted plates, or woven wire cloth. The capacity
9.7-36
-------
of screens is dependent on the open area of the screening surface and the
physical characteristics of the feed. Screening equipment commonly used in
the metallic minerals industry includes grizzlies, shaking screens, vibrating
screens, and revolving screens.
Screens generate dust emissions from agitation of the dry material.. The
amount of dust emitted depends on the size of the particles being agitated,
the moisture content of the ore, and the agitation frequency and amplitude.
Conveying—The movement of ore between the various processing units
is accomplished by feeders (apron, belt, reciprocating belt, vibrating, and
Wobbler(TM))} belt conveyors, elevators (bucket and scraper), screw
conveyors, and pneumatic conveyors (pressure and vacuum).
Ore conveying systems can be a source of process fugitive emissions if
controls are not installed. Particulate emissions can be generated at any
point where ore is transferred from one system to another by free fall, such
as at transfer points between belt conveyors or from a tripper to storage.
Conveying ore over long distances in the open air can be a source of emis-
sions due to wind action. The extent of particulate emissions depends
largely on the fineness of the ore being transported, moisture content, belt
speed, and free fall distance.
Separating and Classifying—Mechanical air separators of the centrif-
ugal type find.wide acceptance in the industry for classification of dry
materials in a relatively fine state of subdivision. In commercial practice,
the separators usually follow vibrating screens extending from about 40 to
60 mesh and smaller. Forced air is used to classify fine particles. The
air is either vented or recirculated; if vented, this air is a source of
particulate emissions.
Bagging and Bulk Loading—In the metallic minerals industry, a valve
type paper bag is sometimes used for shipping fine materials. Dust is
emitted during the final stages of filling when dust-laden air is forced out
of the bag. Fine product materials that are not bagged for shipment are
either bulk-loaded in tank trucks or enclosed railroad cars or shipped. The
usual method of loading is by gravity feeding through plastic or fabric
sleeves. Bulk loading of fine material is a source of particulate emissions
because dust-laden air is forced out of the truck or railroad car during the
loading operation.
9.7-37
-------
Waste Disposal—The metallic minerals mining and processing industry
generates two types of waste material: the waste rock and overburden which
are produced at the mine to expose the ore, and the tailings which are the
waste from the ore concentration process. Pebble and boulder-sized waste
material is usually piled, whereas fines are slurried and discharged to a
pond. Both types of wastes can be a source of fugitive dust if left exposed
to the action of wind, rain, or evaporation.
9.7.2.2 Emission Control Techniques. There are two broad categories
for dealing with particulate emissions in the metallic mineral mining indus-
try: suppression and collection. Dust suppression is the control of dust _,
emissions by preventing the particles from becoming airborne. This technique
is used primarily for fugitive dust emissions control and is accomplished by
wetting systems or by installing some form of coverage. Dust collection
usually involves exhausting the dust generating source(s) to a device that
removes dust from the airstream passing through it. Table 9.7-14 lists
several particulate emission sources common to the metallic mineral industry
with the applicable control options for these sources.
Wet suppression systems are usually employed at the truck dump or crush-
ing pit, and at the inlet and exit of the primary crusher. There are several
wetting agents available to the industry. Their chemical compositions vary
but they are usually composed of a hydrophobic group (long-chain hydrocarbon)
and a hydrophillic group (usually a su'lfate, sulfonate, hydroxide, or ethy-
lene oxide) which, when added to water, reduce its surface tension and allow
deeper penetration. One wet suppression system manufacturer claims 90 per-
cent emission reduction from primary crushers to stockpile and reclaim.13
It should be noted that the common industry practice is to use a wetting
system as a supplement to dry collection systems; however, some plants use
dry collection systems only.12
Dry collection consists of enclosing a particular source of emission
with a hood. Dust-laden air emissions are transported to a central collec-
tion area or to several control devices. Hoods are generally classified
into three broad groups: enclosures, receiving hoods, and exterior hoods.^
Enclosures are commonly used at conveyor transfer points, crushers, and
screens; they are usually the only type of hooding used in the industry.
9.7-38
-------
Table 9.7-14. PARTICIPATE EMISSION SOURCES AND CONTROL OPTIONS FOR THE
PROCESSING OF METALLIC MINERALS^
ID
co
vo
Method of control
Wet
suppression
Source system
Crushing X
Dry grinding X
Screening X
Conveying X
Conveyor transfer X
points
Drying concentrate
Concentrate loading
and transfer
Stockpiling
Storage bins X
Haul roads and
plant yard
fugitive dust
Wind erosion of X1^
stockpiles
Wind erosion of tailing
dams and beach areas
of tailing ponds
Dust
collection
system3
X
X
X
X
X
X
X
X
Coverage Dust
of Oiling, agglomerates, Vegetative
equipment paving Windbreaks wetting cover Other
X
X
x
X Xl>
X
X
X X Xc
X
X X
X X
X X X ' X'e
aA system designed to capture and vent emissions to a control device.
bSpaced capture points exhausted to control device.
cStacker or radial stacker conveyors.
^Effective use of system prior to stockpiling.
eSoil stabilization program.
-------
Receiving hoods are used for high velocity dust collection; exterior hoods
are rarely used in the mining industry. For detailed hooding exhaust rates
and design considerations, see Section 5 of Volume 1.
Once the dust has been captured, it must be removed from the airstream
by a collection device before final discharge to the atmosphere. These
collection devices include wet scrubbers, dry cyclones, mechanical collec-
tors, fabric filters, and electrostatic precipitators. Table 9.7-15 is a
summary of emission tests of several of these collection devices.15,16,17,18,19
Factors which affect the type of dust control device selected include parti-
cle and carrier gas characteristics, process parameters such as participate
concentration and flow rate, and operational characteristics such as pressure,
temperature, moisture, corrosion service requirements and other
physical limitations.
Wet scrubbers are the most common dust collectors in the metallic min-
eral processing industry. The chief scrubber designs used are wet cyclone,
mechanical-centrifugal with water spray, and venturi. Wet cyclones have been
operated using 0.5 to 0.8 liters of water per cubic meter of gas and have
achieved collection efficiencies of 98 percent.2° Venturi scrubbers have
achieved efficiencies averaging 99.9 percent with throat velocities of 4570
to 6090 meters per minute and water consumption of 0.4 liters per cubic meter
of gas. These, high efficiencies are also evidenced by the low particle size
ranges collected (less than 1 micrometer).21
Fabric filters are used for particulate emission control from point
sources in the metallic mineral industry because of their high collection
efficiencies and competitive annual costs. Electrostatic precipitators are
rarely used in the metallic mineral processing industry. Their use is in
iron ore pelletizing and for such aluminum industry applications as kiln
drying and calcination operations involving refined bauxite and alumina.
Efficiencies of 70 to 98 percent have been reported.20
Dry inertia! separators are also commonly used in alumina plants to
recover alumina product and emissions from kilns just prior to the passage
of the airstreams into the electrostatic precipitators. The range of col-
lection efficiency is from over 90 percent for greater than 30 micrometer
sized particles to 60 percent for the 12 to 15 micrometer size range; effi-
ciency is lower for smaller particles.22
9.7-40
-------
Table 9.7-15. PARTICULATE CONCENTRATIONS AND PARTICLE SIZE DATA
FOR METALLIC MINERALS PROCESSES*5.^,17,18,19
I
-p.
Unit
Primary
crusher
Secondary
crusher
Screening
Plant
A
B
C
D
C
E
B
Screening and D
milling
Ore storage
Product
handling
truck
loadout
Conveyor
transfer
B
E
C
C
C
Control
device
Baghouse
Scrubber
Scrubber
Scrubber
Scrubber
Baghouse
Scrubber
Scrubbers
Scrubber
Scrubber
Baghouse
Scrubber
Operating
parameter
0.88 m/mirt
2 kPa
1.5 kPa
1.5 kPa
1.5 kPa
2.77 m/min
2 kPa
2 kPa
1.5 kPa
1.5 kPa
1.5 kPa
Inlet
concentration,
g/DSCM (gr/DSCF)
3.07 (1.34)
Unknown
unknown
0.302 (0.132)
2.75 (1.2)
0.137 (0.06)
0.009 (0.004)
0.073 (0.032)
0.169 (0.074)
1.60 (0.70)
Size less than
stated diameter
16.655 (4.5 Aim)
unknown
unknown
20%<5 /urn
5%<5 Aim
78%<5 A
-------
9.7.3 Brick and Related Clay Products
The manufacture of brick and related clay products such as clay pipe,
tile, pottery, and some types of refractory brick involves the mining,
crushing, drying, dry grinding, and concentrating of the raw materials, and
the blending, forming, cutting or shaping, drying or curing and firing of
the final product. Since clay suitable for commercial use is found in all
50 States, this industry is fragmented with mining and manufacturing opera-
tions in most of the States. The structural clay products industry alone
was composed of 850 firms in 1967; and in 1974, over 500 manufacturing
plants were identified as producers of common brick and face brick.23,24
Particulate matter is the primary emission in the manufacture of brick
and related clay products. The main source of dust is the materials
handling procedure, which includes drying, grinding, screening, and storing
the raw material. In addition, combustion products are emitted from the
fuel consumed in the curing, drying, and firing portion of the process.
Particulate emissions from the manufacture of brick and related structural
clay products amounted to 515 Gg in 1977.4
9.7.3.1 Process Description. Structural clay products are manufac-
tured from a variety of common clay or shale raw materials. The primary
types are kaolinite and montmorillonite; they contain varying amounts of
impurities, depending upon the location of the clay supply. After the clay
is mined, it is crushed to remove stones and stirred before it passes onto
screens in preparation for processing into brick or structural clay products
(see Figure 9.7-12).25 Particulate emissions arising from mining, crushing,
and screening operations and the applicable control techniques are discussed
in Section 9.7.1.
Crushed and processed clay is manufactured into brick and tile by
forming the clay by extrusion or in molds, drying, and then firing the ware
in a kiln. The blended clay raw materials are mixed with water and formed in
an extruder or in molds to the desired shape. The formed product is then
dried. Common brick can be air dried, but more frequently the brick and tile
products are dried in a batch or tunnel dryer. Waste heat from the cooling
section of the firing kiln is usually used for drying, but makeup heat can
be provided by fuel combustion. If the dry press formation method has been
9.7-42
-------
CLAY
FEED
STORAGE
HOPPER
I
DUST TO ATMOSPHERE
OR CONTROL DEVICE
DUST TO ATMOSPHERE
OR CONTROL DEVICE
DUST TO
.ATMOSPHERE
OR CONTROL
DEVICE
SCREENS
•
RETURN TO
SUPPLY
CRUSHER
STORAGE
HOPPER
DUST TO
ATMOSPHERE
OR CONTROL
DEVICE
CRUSHER
SCREENS
n
<• o
DUST TO
ATMOSPHERE
OR CONTROL
DEVICE
RECYCLE
MIX
STORAGE
HOPPER
WATER
1-D
EXTRUDER CUTTER
TO ATMOSPHERE OR
POLLUTION CONTROL
DEVICE
FUEL
BRICKS
TO
STORAGE
COOLING
SFrTTON
T
FIRING
HEAT-ING
SFrTTON
1
COMBUSTION
AIR
TUNNEL
KILN
Figure 9.7-12 Common brick manufacturing' using the continuous
stiff-mud process.25
9.7-43
-------
used, the formed product is fired without first being dried.
Brick and tile products are then fired in either a continuous tunnel
kiln (usually) or a periodic kiln. In the tunnel kiln, cars are loaded
with the formed pieces and pass continuously through the kiln in a counter-
flow to the air and combustion products. In the first chamber, the pieces
contact the spent combustion air and are heated while the air is cooled
prior to being exhausted from the kiln. The combustion of the fuel and more
intense heating or firing of the pieces occurs in the second chamber where
they are fused to final hardness. The fired pieces are cooled in the third
chamber by preheating the incoming combustion air. The total time required
for the pieces to pass through the kiln varies from about 40 hours to 270
hours. Temperatures in the hottest zone are in the 800 to 1100°C range.
Approximately 4.3 to 5.7 Am^/min of exhaust gas is generated for each mega-
gram of daily kiln production. Flue gas temperatures are in the 150 to 315°C
range. Kiln production is typically in the 45 to 225 Mg per day range.24
Uncontrolled particulate emissions from a brick kiln and dryer are
summarized in Table 9.7-16.26 /\ particle size analysis for the kiln emis-
sions is summarized in Figure 9.7-13.26
9.7.3.2 Particulate Control Techniques. Raw material crushing,
grinding, drying, calcining, blending, conveying, and stockpiling in the
processing of clay products all cause particulate emissions. For applicable
control techniques, see Section 9.7.2 and Section 5 of Volume 1.
Particulate emissions are also produced by the combustion of oil or
solid fuels in tunnel kilns at brickmaking plants. In addition, other emis-
sions such as sulfur oxides, fluorides, and hydrocarbons may be generated
from building brick or refractory tunnel kilns. The amount and character-
istics of these emissions are dependent on the chemical composition of the
clay. Sometimes, other materials are added to the clay to improve the
quality of the brick or tile; for example, sawdust, coal, ashes, and ground-
fired bricks are added.
The type of control equipment used depends to a large extent on the type
of fuel being burned. Coal, sawdust, natural gas, and oil are used in the
building brick industry. Only natural gas and oil are presently used at
refractory plants. Emissions from gas-fired kilns are about 5% of those
9.7-44
-------
TABLE 9.7-16. UNCONTROLLED PARTICIPATE EMISSIONS FROM A BRICK KILN AND DRYER26
Particulate emission rates
Facility
Ib/ton brick kg/MT brick gr/dscf mg/dscm
Comments
Firing kiln
Dryer
Firing kiln
Dryer
1.61
0.0671
2.31
0.0592
0.805
0.0336
1.16
0.0296
0.0636 145 Approximately
75% coal-fired
0.0012 2.81 and 25% gas-
fired; 4.3% ash
in coal.
0.0893 204 Approximately
75% coal-fired
0.0011 2.52 and 25% gas-
fired; 6.9% ash
in coal
9.7-45
-------
CD
+J
CD
O
O
13
e;
<:
O.
LU
10.0
8.0
6.0
4.0
3.0
2.tt
1.0
0.8
0.6
0.4
0.3
0.2
0.1
I 1 I
©-Uncontrolled emissions
from a firing kiln burn-
ing 75% coal w/4.3% ash
content and 25% gas.
ED-Uncontrolled emissions
from a firing kiln burn-
ing 75% coal w/6.9% ash
content and 25% gas.
i i i
j I
I i
0.01 0.1
12 5 10 20 304050607080 90 9"5 9899
CUMULATIVE PERCENTAGE LESS THAN INDICATED 'DIAMETER, by weight
Figure 9.7-13
Particle size distribution for emissions from a brick
kiln and dryer.26
9.7-46
-------
from coal-fired kilns.24 At building brick plants that have minimal gaseous •
pollutants, fabric filters may be used to control particulate emissions from
coal-fired tunnel kilns.2? However, these devices may be subject to bag
blinding if the temperature is not maintained above the dew point. Gas-fired
tunnel kilns emit such low levels of particulate emissions that the majority
are uncontrolled. Wet scrubbing may be used at those sites that have usually
high levels of hydrocarbons, sulfur, and flouride in the clay. At one re-
fractory plant, a wet ionic scrubber is used to control gaseous fluorides,
chlorides, and ammonium bisulfuric acid formed from the reaction of ammonia
and sulfur oxides.28 LOW energy scrubbers also have low efficiency for the
removal of volatized organic matter due to its small particle size.
In addition to "add on" control devices, particulate emissions from
coal-fired brick and tile kilns are greatly reduced with the use of oil or
gas--to 5 percent of the coal level.24
9.7.4 Beneficated Clay Products Manufacturing
The production of products from beneficated clays utilizes processes
similar to those applicable to the manufacturing of structural clay products
(see Section 9.7.3). These processes involve the grinding, screening, cal-
cining, and blending of the raw clay materials and the forming, drying or
curing, and firing of the ware. While structural clay products are generally
manufactured from common crushed clay, beneficated clays are used to manu-
facture numerous products including ceramics and refractories, adhesives,
drilling mud, filter material, foundry sand, cosmetics and Pharmaceuticals,
paper coating and fillers, pesticides and gypsum products.23
Particulate emissions occur during the handling of raw materials,
grinding, screening, drying and firing operations. In 1977, these emissions
amounted to 114 Gg.4
9.7.4.1 Process Description. The production of beneficated clay
products involves the conditioning of the basic ores by several methods.
These include the separation and concentration of the minerals by screening,
floating, wet and dry grinding and blending of the desired ore varieties.
The multiplicity of beneficated clay products results in a number of highly
variable process flowsheets. A description of the processes applicable to
the manufacture of each product is beyond the scope of this discussion.
9.7-47
-------
However, since many beneficated clays are used for ceramic and refractory
products, their production procedures are described briefly.
The basic raw materials in ceramic clay manufacture are kaolinite and
montmorillonite clays. These clays are refined by separation and bleaching
and then blended, kiln dried, formed into the final product, and kiln fired.
The benefication and bleaching processes for ceramic clay are illustrated in
Figure 9.7-14.29
Refractories are those materials that are used to withstand the thermal,
chemical, and physical effects that occur in furnaces. Refractories are
sold in the form of firebrick, silica brick, magnesite brick, chromite brick,
magnesite-chromite brick, zircom'a, and others. The usual operations in manu-
facturing kiln-fired refractories include grinding and screening, calcining,
mixing, pressing or molding and repressing, drying, and burning. Depending
upon the desired product, raw materials may be calcined or dried prior to
mixing and blending. Figure 9.7-15 illustrates an overall flowsheet for a
typical plant producing a kiln-fired refractory. The decision to calcine or
dry the raw material depends upon tts end use. The type of clay or refrac-
tory brick and its ultimate density are among the factors that influence the
decision.
Fused-cast refractories are manufactured by: carefully blending such
components as alumina, zircom'a, silica, chrome, and magnesia; melting the
mixture in an electric-arc furnace at temperatures of 1760 to 2480°C;
pouring it into molds; and slowly cooling it to the solid state. Fused
refractories are less porous and more dense than kiln-fired refractories.
Castables also refer to unformed dry mixes which are later mixed with
water and applied in position, after which they solidify.
9.7.4.2 Emission Control Techniques. Particulate matter from the
processing and manufacturing of beneficated clay products is emitted from
the raw materials handling procedures, grinding, drying, calcining and firing
operations. Factors affecting these emissions include: the type and quan-
tity of clay processed, the type of grinding (wet or dry), the temperature
within the dryers, gas velocities and flow direction within the kilns, type
of dryer and firing-kiln used, final clay moisture content, and impurities
within the raw material ores.
9.7-48
-------
RAW
MATERIAL
THICKENER
EENEF1CIATION
TROMMEL
SCREEN
WASTE
ER
1
FIL-
TER
1 JLLC.I-
i»ps
TIVE
ETTLING
V
y DI r^uuuui
WASTE
PANOx F
DRYER — ^CLAy
*
WASTE
BLEACHING
P,NOx, F,
CLAY
BAGPI KG
STORAGE
SHIPMENT
P, NO , F
REACTED GAS*
* REACTION GAS MAY INCLUDE
CHLORINE AND CARBON TETROCHLORJDE
OR OTHER BLEACHING AGENTS.
Figure 9.7-14 Ceramic clay manufacturing processes.29
9.7-49
-------
CALCINER
QUARRY
BLASTING,MINING
STOCKPILE
PRIMARY
CRUSHING
RAW CLAY
MIXING
SECONDARY
GRINDING OR
MILLING
BLENDING
(USUALLY WET)
BURNING
DRYER
PRODUCT
Figure 9.7-15. Flow diagram, kiln-fired refractories manufacturing.29
9.7-50
-------
Participate emissions from the manufacture of castable refractories are
created by the drying, crushing, handling and blending phases of this process,
as well as by the actual melting process and the molding phase. These emis-
sions are affected by the amount of material handling and pretreatment re-
quired before melting and by the components in the melt. Generally, increas-
ing concentrations of silicon will increase particulate emissions. Emissions
from the electric arc furnace are condensed fumes consisting of very fine
particles, the majority of which are 2 micrometers or smaller.8
Common control techniques for the beneficated clay manufacturing pro-
cesses include mechanical collectors, wet scrubbers, electrostatic precipi-
tators, and fabric filters. Cyclones for the coarser material, followed by
wet scrubbers, fabric filters, or electrostatic precipitators for dry dust,
are the most effective control systems. A variety of control devices may be
used to reduce both particulate and gaseous emissions. Almost any type of
particulate control system will reduce emissions from the materials handling
process. However, good design and hooding are required to capture the emis-
sions (see Section 5 of Volume 1 for additional details). Emissions from
the electric arc furnace used in castable refractory production are largely
condensed fumes and consist of very fine particles. Fabric filters may be
used to control these emissions. Multicyclones,- baghouses, and electrostatic
precipitators have been used on rotary and vertical kilns in kiln-fired
refractory plants.
9.7.5 Gypsum
Gypsum, or hydrated calcium sulfate, is mined in open pits and under-
ground mines and then processed into the following product-categories (each
category's approximate percent of the total produced is given):
o Crushed or pulverized gypsum (cement retarder, agricultural
gypsum)—27 percent.
o Pulverized calcined gypsum (various types of wall plasters and
specialty plasters)--6 percent.
o A variety of prefabricated gypsum-core board products (wallboard,
rocklath, sheathing, and wallboard)—67 percent.
In 1978, the industry included aproximately 100 facilities which either
mined and mechanically processed crude gypsum rock, or calcined and produced
9.7-51
-------
prefabricated products. Thirty-six of these facilities were integrated min-
ing plus fabricating installations and 41 were mines. Twenty-nine companies
sell sized, ground, or crushed gypsum rock. The industry produced about
19 Tg of gypsum products from 12 Tg of domestic gypsum plus 7 Tg imported
gypsum.30 A typical calcining plant produces 20.4 Mg per hour.31
Emissions of particulate gypsum or calcium sulfate are generated by
grinding equipment, calciners, and dryers; and in 1977, these emissions were
estimated to amount to 70 Gg.4
9.7.5.1 Process and Emission Control Techniques. All gypsum processing
operations are similar, except for differences in product mix. Depending on
the final product, the sequence of processing steps is as follows:
o Mining
o Crushing/grinding9
o Ore drying
o Calcining (either in kettles or other types of calciners)
o Blending
o Fabricating
o Packaging
The mining and processing of a number of nonmetallic minerals, including
gypsum, are described in Section 9.7.1. Figure 9.7-16 illustrates the pro-
cess flow sheet for various gypsum products, as well as the particulate
emission sources.
Particulate control technologies applicable to process and fugitive
emissions from the mining and crushing/grinding operations (if used) are
similar to those for nonmetallic minerals. Generally, cyclones, bag fil-
ters, and electrostatic precipitators, each individually or in combination,
if properly designed and operated, are capable of meeting the most restric-
tive emission limitations.32
9.7.5.1.1 Ore drying. Crushing is followed by ore drying; emissions
from ore drying consist of particulates from both drying and fuel combustion,
and gases from fuel combustion. Particulates are the most significant
aGrinding usually follows ore drying when drying is employed.
9.7-52
-------
<£>
I
cn
HEAT
CRUSHING
GRINDING
2
Figure 9.7-16 Gypsum products process flow diagram.
-------
emissions from ore dryers. Such particulates are generated mainly by the
entrainment of ore fines in the drying gas stream; however, some particu-
lates are emitted from fuel combustion. Emissions from the entrainment of
ore fines depend mainly on the gas flow through the dryer. Thus, indirect-
heated dryers that limit the air flow to the amount needed to carry away
moisture will have less entrainment than direct-heated dryers.
Dryers used in the gypsum industry are normally direct-fired, with the
combustion gases directly contacting the gypsum ore. Indirect heating with
combustion gases would still result in the same emissions to the atmosphere.
Uncontrolled particulate emissions are estimated from the reported
controlled emissions and from the reported particulate control efficien-
cies. Table 9.7-17 gives uncontrolled and controlled emissions for an ore
dryer, as reported in a permit application.33
From data reported in the U.S. EPA's National Emissions Data System
(NEDS) the uncontrolled particulate emission factors for ore dryers range
from 1.3 g/kg dried (3 Ib/ton dried) to 95 g/kg dried (190 Ib/ton dried).
The average uncontrolled emission factor, based on a production-weighted
average of available data, is about 25 g/kg dried (49 Ib/ton dried). This
average value is comparable to the average value of 20 g/kg dried (40 Ib/ton
dried) reported in the Compilation of Air Pollutant Emission Factors (U.S.
EPA Publication No. AP-42).
The only emission control systems used on gypsum dryers are particulate
control devices. Sulfur oxide and nitrogen oxide controls are not needed
when clean-burning fuels are used. Emission controls used in current facili
ties include cyclones, electrostatic precipitators, fabric filters, and
wet scrubbers. The reported controlled particulate emission factors range
from 0.011 g/kg dried (0.021 Ib/ton dried) to 4.0 g/kg (8.0 Ib/ton dried),
with a production-weighted average of 0.25 g/kg (0.50 Ib/ton). Test data
for emissions controlled by fabric filters show emission factors as low as
0.012 g/kg dried (0.023 Ib/ton dried) and grain loadings as low as 0.016
(0.0071 gr/acf). These lower controlled levels demonstrate control effi-
ciencies exceeding 99.9%.
9.7.5.1.2 Calcining. Emissions from calcining operations consist of
particulates from both fuel combustion and calcining of the gypsum ore as
well as of gases from fuel combustion.
9.7-54
-------
Table 9.7-17. ORE DRYING EMISSIONS REPORTED
IN A PERMIT APPLICATION9'33
Controlled participate emissions'5 Mg/yr (ton/yr) 11 (12)
.g/kg (Ib/ton) 0.06 (0.12)
g/m3 (gr/acf) 0.05 (0.02)
Uncontrolled participate emissions Mg/yr (ton/yr) 18,000 (20,000)
g/kg (Ib/ton) 95 (190)
g/m3 (gr/acf) 92 (40)
aDryer fires No. 6 fuel oil with 1.5% sulfur; participate emissions
estimates include particulates from drying and from fuel combustion.
bFabric filter control; air flow = 14,000 acfm (6.6 m3/s).
9.7-55
-------
Emissions from the entrainment of raw and calcined gypsum depend on the
gas flow through the calciner and on the agitation within the calciner.
Indirect-heated calciners that limit the air flow to the amount needed to
carry away moisture could have less entrainment than direct-fired calciners
but data supporting this comparison are unavailable.
Both direct- and indirect-heated calciners are used in the gypsum indus-
try. In direct-heated calciners, as in flash calciners, hot combustion
gases directly contact the gypsum. Only clean fuels can be burned to avoid
contaminating the calcined product. Indirect-heated calciners such as
kettles predominate in the gypsum industry. In these calciners, heat for
calcining is provided by hot combustion gases flowing outside the kettle
shell and in flues through the kettle. The combustion gases are normally
vented separately from the kettle off-gas.
Estimates of controlled and uncontrolled emissions from gypsum calciners
are available from NEDS data. Uncontrolled particulate emissions are esti-
mated from the reported controlled emissions and from the reported particu-
late control efficiencies.
Uncontrolled particulate emission factors for calciners range from
7.5 g/kg calcined (15 Ib/ton calcined ) to 65 g/kg calcined (130 Ib/ton
calcined). The average uncontrolled emission, based on a production-weighted
average of available data, is about 28 g/kg calcined (56 Ib/ton calcined).
This average compares to the average value of 45 g/kg calcined (90 Ib/ton
calcined) reported in the Compilation of Air Pollution Emission Factors.
Estimated uncontrolled emissions from typical calciner facilities range from
360 Mg/yr (400 tons/yr) to 5.7 Gg/yr (6300 tons/yr).
The only emission control systems used on gypsum calciners are particu-
late control devices. Particulate emission controls used in current facili-
ties include cyclones, electrostatic precipitators, fabric filters, and wet
scrubbers. Combustion gases from the indirect-heated calciners circumvent
the control device and are combined with the cleaned off-gas downstream of
the emission control device. The reported controlled particulate emission
factors range from 0.003 g/kg calcined (0.006 Ib/ton calcined) to 3.7 g/kg
calcined (7.3 Ib/ton calcined), with a production-weighted average of
1.2 g/kg calcined (2.4 Ib/ton calcined). The relatively high average
9.7-56
-------
controlled emission factor results from the large emissions from several -
uncontrolled calciners. Tests data for emissions controlled by fabric
filters show emission factors as low as 0.003 g/kg (0.006 Ib/ton calcined)
and grain loadings as low as 0.0092 g/m3 (0.0040 gr/acf). These controlled
levels demonstrate control efficiencies exceeding 99.9%.
9.7.5.1.3 Dry mixing. Emissions from dry mixers occur during feeding
and blending of gypsum and various additives. These emissions consist of
gypsum and additive dusts. Continuous dry mixing in screw conveyors is used
for gypsum board manufacture, and dry mixing in batch equipment is used for
plaster manufacture.
Available data on the uncontrolled and controlled particulate emis-
sions from dry mixing, presumably batch dry mixing, are summarized in
Table 9.7-18.34*35'36 These data include data from the NEDS (Plant LL),
a permit application (Plant Q), and a source test report (Plant AA). Data
from NEDS include estimated uncontrolled emissions based on the reported
controlled emissions and the reported particulate control efficiency. The
permit application contains a gypsum producer's estimates of the uncon-
trolled and controlled emissions.
The only two estimates of the uncontrolled particulate emission fac-
tors for dry mixing are 10 g/kg mixed (19 Ib/ton mixed) and 33 g/kg mixed
(65 Ib/ton mixed). The average uncontrolled emission factor based on a
production-weighted average of available data is 10 g/kg mixed (20 Ib/ton
mixed). Estimated uncontrolled emissions from two facilities are 2.9 Gg
(3200 tons/yr) and 91 Gg (100 tons/yr).
The reported particulate emission factors for two dry mixers controlled
by fabric filters are 0.011 g/kg mixed (0.022 Ib/ton mixed) and 0.65 g/kg
mixed (1.3 Ib/ton mixed), with a production-weighted average of 0.017 g/kg
mixed (0.033 Ib/ton mixed), Test data for emissions controlled by fabric
filters show an emission factor of 0.026 g/kg mixed (0.052 Ib/ton mixed)
and a grain loading of 0.015 g/Nm3 (0.0065 gr/scf).
9.7.5.1.4 Scoring and chamfering. Emissions from scoring and chamfer-
ing consist of paper fibers. Available data on the emission of fibers from
scoring and chamfering are shown in Table 9.7-19.33>35»37 These data are
derived from the NEDS (Plant N) and two permit applications (Plants KK
9.7-57
-------
Table 9.7-18. DRY MIXING EMISSIONS34,35,36
10
*
--J
tn
00
Gas flow rate,
Plant ID
Plant Q
Plant LL
Plant AAC
Operation Control device m3/s (acfm)
Dry mixing FFb
Blending FF
Blending/ FF
handling
2.8
1.6d
2.0d
1.7d
1.8d
6000
3374
4200
3608
3727
Controlled participate emissions,
Mg/yr (ton/yr)a g/kg
3.4
1.8
0.6
0.8
1.0
0.8
3.7
2
0.6
0.9
1.1
0.9
0.011
0.65
0.018
0.028
0.033
0.026
(Ib/ton) g/m3
0.022
1.3
0.035
0.055
0.065
0.052
0.039
0.011s
0.014s
0.019s
0.015s
(gr/acf)
0.017
0.0049
0.0061
0.0084
0.0065
aAssumes 8322 hours/year.
bpF = fabric filter, Cyc = cyclone.
CEPA-5 used.
dNm3/s (scfm).
(gr/scf).
-------
Table 9,7-19. SCORING AND CHAMFERING EMISSIONS DATA33'35'37
Plant ID
Plant KK
Plant N
Plant 00
Control device
Gas flow rate
acfm
m3/s
2700
1-3
Controlled particulate emissions
ton/yr
Mg/yr
1b/tonb
g/kgb
gr/acf
0.058b
0.053b
0.00054
0.00027
0.0006
0.0013
Cyca
5298
2.5
60
55
0.37
0.85
Cyc/FFa
3000
1 -4
0.42b
0.38b
0.0029
0,0015
0.0039
0.0089
Uncontrolled particulate emissions
ton/yr
Mg/yr
lb/tonc
g/kgc
gr/acf
g/m3
Control efficiency
58t>
53b
0.54
0.27
0.60
1.4
99.9
300
270
2
5
80.0
57b
51b
, 0.40
0.20
0.53
1.2
99.3
aFF = fabric filter, Cyc = cyclone.
bAssumes 8322 hrs/yr.
cExpressed as units participate per unit wallboard formed.
9.7-59 '
-------
and 00). Data from the NEDS include estimated uncontrolled emissions based
on the reported controlled emissions and reported control efficiency.
The only two estimates of the uncontrolled particulate emission factors
for scoring and chamfering are 0.27 g/kg wall board formed (0.54 Ib/ton) and
0.20 g/kg (0.40 Ib/ton), with a production-weighted average of 0.23 g/kg
(0.46 Ib/ton). Estimated uncontrolled emissions from three facilities range
from 51 Mg (56 ton) per year to 270 Mg (300 ton) per year.
Emissions from scoring and chamfering are usually controlled by cyclones
and/or fabric filters. The reported controlled emission factors for two
facilities controlled by fabric filters are 0.00027 g/kg (0.0054 Ib/ton) and
0.0015 g/kg (0.0029 Ib/ton), with control efficiencies of 99.9 and 99.3
percent.
9.7.5.1.5 Board end sawing. Emissions from board end sawing consist of
gypsum and fibers generated when dried board or block is sawed to give smooth,
straight ends.
Data from the NEDS and two permit applications include estimated uncon-
trolled emissions based on the reported controlled emissions and the reported
particulate control efficiency. The permit applications contain gypsum pro-
ducer estimates of the uncontrolled and controlled emissions. Table 9.7-20
summarizes source test data from one board end-sawing operation controlled
by fabric filters.^
The uncontrolled particulate emission factors range from 0.5 g/kg wall-
board processed (1 Ib/ton) to 50 g/kg wall board processed (100 Ib/ton), with
a production-weighted average of 23 g/kg (47 Ib/ton). Estimated uncontrolled
emissions from typical facilities range from 36 Mg (40 tons) per year to
15 Gg (17,000 tons) per year.
Emission controls applied to board end-sawing operations include
cyclones and fabric filters. The reported controlled particulate emission
factors range from 0.011 g/kg wallboard processed (0.021 Ib/ton) to 0.7 g/kg
wallboard processed (1.4 Ib/ton), with a production-weighted average of
0.35 g/kg (0.70 Ib/ton). None of these emission factors has been verified
by source testing. Test data are available only to estimate the grain
loading in exhausts from fabric filter-controlled operations. As shown in
Table 9.7-20, one series of tests showed controlled grain loadings of about
9.7-60
-------
Table 9.7-20. END SAWING EMISSIONS MEASURED IN SOURCE TESTINQa>38
ID
Gas
rn^/s
0.90
0.93
0.90
0.91
flow rate
acfm
1912
1977
1908
1932
Gas
°C
32
32
33
. 32
temperature
°F
90
90
91
90
g/s
0.0058
0.0019
0.0066
0.0048
Participate
Ib/h
0.046
0.015
0.052
0.038 -
emission
g/m3
0.0064
0.0021
0.0074
0.0053
gr/acf
0.0028
0.0009
0.0032
0.0023
aEPA-5 testing of fabric filters.
-------
0.0053 g/m3 (0.0023 gr/acf). A single test at another board end-sawing
operation showed a grain loading of 0.011 g/m3 (0.0048 gr/scf).39 These
levels probably correspond to control efficiencies exceeding 99.5%.
9.7.6 Lime
Lime is the high-temperature product of the calcination of limestone.
Chief products of the lime industry are quicklime, slaked or hydrated lime,
dolomite, and hydrated dolomite. In 1974, total lime production by 172
plants was 19.6 Tg. Of this total, 17.3 Tg were quicklime (calcium oxide)
and 2.3 Tg were hydrated lime (calcium hydroxide).4^
Industrial processes emit limestone and lime dust, flyash from fuel
combustion, and soot and tars resulting from the incomplete combustion of
fuels. Total particulate emissions from lime production in the U.S. in
1977 amounted to 155 Gg.4
9.7.6.1 Process Description. The basic processes in the production of
lime are: (1) quarrying the limestone raw material, (2) preparing the lime-
stone for kilns by crushing and sizing, (3) calcining the limestone, and (4)
processing the quicklime further by hydration.41 Figure 9.7-17 shows a gen-
eralized lime manufacturing process and the potential particulate emission
points.
Lime is made from limestone which is subjected to temperatures of about
1100°C (2000°F). to break it down chemically, producing quicklime and releas-
ing C02- Calcining at this temperature produces a soft, porous, highly
reactive lime.
In the United States, calcination is carried out in a variety of kilns,
including the long rotary kiln, the short rotary kiln with external stone
preheater, the vertical or shaft kiln, the rotary hearth or Calcimatic
kiln, and the fluidized bed kiln. Each type has its own advantages, but the
U.S. lime industry apparently favors rotary kilns; almost 90 percent of
U.S. lime production is processed in rotary kiln systems.^ Virtually all
kilns built in 1974-1975 were rotary kilns, and this trend is expected to
continue in the future.43 One factor that makes the rotary kiln attractive
for processing lime in the future is that it is the only kiln that can
presently use coal and still maintain product quality.44
The operation of long and short rotary kilns is basically identical.
These kilns are slowly rotating, slightly inclined cylindrical furnaces
9.7-62
-------
—
1
£
CONTROL _^
. FUEL-*-
£
CONTROL
DEVICE ^^
r~
uvnPATFn i
LIME
^-'-nJV MILL/AIR
SEPARATOR
STORAGE/
SHIPMENT
LIMESTONE
QUARRY/MINE
t
PRIMARY
CRUSHER
1
SECONDARY
CRUSHER
i '
SCREENS AND
CLASSIFIERS
1
STONE
PREHEATER
(LIMESTONE
KILN
LIME
,< r
PRODUCT
COOLER
(LIME
WATER/DUST SLURRY
•\ |- - rt^
\s*' "•"^
^^
^^-
^—
^^K_
^
KILN
EXHAUST
J
^-«.
. STORAGE/
' SHIPMENT
WATER SPRAY/ \r^ a.
WETSCRUBBER
^ ,__ _. 5TONE
..r POTENTIAL
"* ^^^/v EMITTING POINTS
^1 AIR/EXHAUST
Figure 9.7-17 Generalized lime manufacturing plant.lf°
9.7-63
-------
made of heavy steel plate lined with refractory brick. They are fired by
one or combinations of three available fuels: natural gas, pulverized coal,
or oil. The largest kiln now in operation in the U.S. has a production
capacity of almost 1000 tons of lime per day. Kilns vary in size, ranging
from about 2 to almost 5 meters in diameter and from 18 to 137 meters in
length.
Limestone is fed into the elevated end of the kiln and is discharged as
quicklime at the lower end into the cooling system. No more than 10 percent
of the kiln is filled with limestone or lime as it moves slowly through the
long cylindrical furnace in a gentle tumbling motion. Usually, cooling air
is induced into the discharge end of the product cooler and into the kiln as
secondary combustion air. The combustion gases flow countercurrent to the
flow of the stone toward the charging end, where they are used to preheat
the kiln feed. In the long rotary kiln, the exhaust gas temperature ranges
between 593 and 760°C. In the external limestone preheater of the short
rotary kiln, the exhaust gas temperatures range between 926 and 1148°C.
Although most lime produced is sold as lime, a small amount (10 percent
in 1974} is converted into slaked lime or hydrated lime.
In most hydration plants, water is added to the lime in a pug mill pre-
mixer where there is thorough blending of the lime and water. The lime-water
mix then goes to the agitated hydrator where most of the chemical reaction
takes place. The reaction is exothermic and the heat of reaction converts
part of the water in the mix to steam. A fan maintains a slight negative
pressure in the hydrator, and the steam is discharged to the atmosphere
along with any air that enters the hydrator through the charging port.
Hydrator emissions are normally controlled by the use of either water sprays
in the hydrator stack OF by wet scrubbers. The resulting slurry or milk of
lime is usually returned to the premixer as part of the slaking water. Vir-
tually all hydrators have this equipment integrally installed.45
9.7.6.2 Emission Characteristics and Applicable Control Technologies.
Nearly every lime production process emits dust. Fugitive and process emis-
sions of particulate limestone are generated by mining, handling, crushing,
and screening operations. Applicable control techniques for these operations
are discussed in Section 9.7.1 and Section 5 of Volume 1. Lime dust is
9.7-64
-------
emitted in the hot kiln gases and from the hydrator. Flyash, soot, and tars
may be generated by fuel combustion used as a heat source for calcining in
the kiln. Fugitive emission sources of lime dust include screening and
pulverizing operations, storage silo vents, packaging and loading equipment,
materials handling, transfer and conveying operations, and plant roads.
Chemical analysis of solid particulate matter emitted from the stacks
of natural gas-fired rotary kilns is listed in Table 9.7-21. These parti-
cles average 5 to 6 micrometers in size.41 The size distributions of emis-
sions from three other lime manufacturing emission sources are given in
Table 9.7-22. A summary of emissions sources and control methods is given
in Table 9.7-23.
Rotary kiln emissions can be controlled by fabric filters, wet scrub-
bers, ESPs or gravel bed filters. Table 9.7-24 is a summary of emission
tests from rotary lime kilns. Baghouses or high-energy scrubbers are most
frequently used for controlling dust emissions from rotary kilns. Bag-
houses generally offer the highest particulate matter collection efficiency
of any gas treatment method. Glass fiber bags with graphite and silicone
finishes are commonly used. Some form of gas cooling is required ahead of
fabric filters since conventional fabrics cannot withstand temperatures in
excess of approximately 300°C. Kiln exhaust temperatures generally exceed
538°C. Cooling is achieved by: (1) evaporative water sprays, (2) indirect
radiation convection heat exchange, (3) ambient air dilution, (4) external
stone preheating, or (5) a combination of the aforementioned devices. Pres-
sure drops of 1.25 kPa and a filtering velocity of 0.67 m/min are typical.41
The most common high pressure drop scrubber used for controlling emissions
from rotary lime kilns is the venturi scrubber. For high removal efficiency,
a pressure drop of 5.5 kPa is required.41
Precipitators for lime kiln applications are of the dry, horizontal flow
construction. Since they are constructed of carbon steel, the kiln gas must
be cooled by methods,similar to those described for the baghouse. Precipi-
tators generally can achieve 90 to 99 percent particulate removal efficien-
cies. Higher efficiencies can be attained by increasing the precipitator
sUe.41
Gravel bed filters have only recently been applied to rotary kilns in
the U.S. They clean exhaust gases in three steps. First, the gas enters
9.7-65
-------
Table 9.7-21. COMPOSITION OF PARTICULATE MATTER FROM NATURAL GAS FIRED
LIME KILNS40
Emission component
Acid insoluble
Heavy metal oxides (R20s)
CaCQs
CaO
MgO
CaS04
Ca(OH)2
Chemical
High-calcium
lime,
wt. %
0.66
0.97
23.06
66.32
1.40
1.22
6.37
analysis
Dolomitic
, 1 i me ,
wt. %
0.45
0.35
64.30
7.23
28.20
0.27
"•"•""••
Table 9.7-22. LIME PRODUCTION EMISSION SIZE DISTRIBUTIONS6
Production operation
emission source
Hammer mill (crusher)
Screening
Bagging house
Particle size,
micrometers
3
5
10
20
40
3
5
10
20
40
5
10
20
40
Weight %,
less than stated size
30
47
60
74
86
46
72
85
95.8
98.8
71
87.3
96
98.8
9.7-66
-------
Table 9.7-23. CONTROL TECHNOLOGIES APPLICABLE TO EMISSION SOURCES IN THE
LIME INDUSTRY40.41
Emission Source
Kiln
Hydrator
0
0
0
0
0
0
0
0
Control
Fabric filter
ESP
Venturi scrubber
Gravel bed filter
Cyclone
Impingment scrubber .
Wet cyclone
Wet scrubber
Quicklime screening
Quicklime or hydrated
lime pulverizing with
leaks from mill and
feed discharge exhaust
systems
Grinding mills
Lime product
silo vents
Packaging
quicklime
and hydrate lime
Materials handling,
conveying, transfer
Truck; rail, ship,
or barge loading of
quicklime or hydrated
lime
Plant roads
— Central wet fan scrubber
— Water sprays in the hydrator
stack
— Direct spray scrubbers and
condensers
o Baghouse
o Wet suppression (might impair
product quality)
o Enclose screens, and vent exhaust
to fabric filter
o Hoods or covers, and vent to
fabric filter
o Better control of operating
parameters and procedures
o Improved maintenance
o Hood and exhaust to fabric filter
o Fabric filter
o Fabric "sock"
o Enclosure; partial or complete
o Choked feedsystem
o Fixed hoods or covers exhausted to
control device
o Movable hoods with flexible ducts
to control device
o Filling spout with outer
concentric aspiration duct to
fabric filter
o Fabric filter (with capture system)
o Fabric sock (with capture system)
o Refer to Section 5 of Volume 1
o Enclosure; partial or complete
o Choked feed systems
o Fixed hoods and covers
o Movable hoods with flexible ducts
o Filling spout with outer concentric
aspiration duct to fabric filter
o Refer to Section 5 of Volume 1
9.7-67
-------
Table 9.7-24 ROTARY LIME KILN EMISSIONS41
Plant
B C D (1) D (2)
Control Baghouse ESP ESP ESP
equipment
AP, kPa 1.09 0.5 0.07 0.7
Air Cloth m3/m2sec 0.5 — — «
ratio
Specific m^/tnin/m3 — 1.06 1.45 1.45
collection
area
g/m3 0.03 0.008 0.03 0.03
g/dry standard ra3 0.05 0.016 0.08 0.07
kg/hr 6.0 2.5 • 3.9 3.4
kg/Mg of feed 0.111 0.068 0.133 0.141
E F
Baghouse Scrubber
0.6 3.7
1.0
--
0.009 0.04
0.013 0.06
0.7 10.1.
0.041 0.216
9.7-68
-------
the filter and its velocity is decreased, which results in the dropout of
large particles. The medium-sized particles are then removed by cyclonic
separation. Finally, the smallest; particles are removed by agglomeration as
they pass through a filter medium of crushed stone. The clean gas is vented
to the atmosphere. Pressure drops on the order of 2.49 kPa are typical for
this system.4!
Lime hydrators are usually controlled by simple scrubbers, most com-
monly the wet fan type with centrifugal separation. If a fabric filter is to
be used, the nearly saturated gas stream must be superheated to avoid conden-
sation. Table 9.7-25 summarizes emission tests of lime hydrators.
Fugitive emissions are controlled by enclosure of the source, or hoods
or covers with exhaust ducts connecting to a control device, usually a fabric
filter. Wet suppression has limited application because of potential impair-
ment to product quality. Emissions during packaging and loading of lime can
be minimized by using a choked-feed system or a gravity-feed fill spout
mechanism with outer concentric aspiration ducts to fabric filters.
Emissions from new rotary lime kilns are limited to no more than
0.15 kg/Mg of limestone feed. Emissions from new lime hydrators are limited
to no more than 0.075 kg/Mg of lime feed.
9.7.6.3 Secondary Environmental Impact. The reduction of particulate
emissions from lime production processes could result in the following secon-
dary environmental impacts:
o Increased solid waste disposal requirements that result from the
collection of particulate matter.
o Increased waste water treatment requirements if wet scrubbers are
used.
o Increased electrical energy consumption.
Dust collected by electrostatic precipitators, scrubbers, fabric fil-
ters, and gravel~bed filters represents a solids disposal problem which must
be resolved. Some producers recycle the collected dust to the lime kiln
while others use the lime dust as a raw material in cement kilns. Other
uses for the collected dust are as an agricultural soil conditioner, a
fertilizer additive, a metallurgical production aid, and a neutralizing
chemical.4! Wet sludges from scrubbers are dredged from the treatment
9.7-69
-------
Table 9.7-25. LIME HYDRATOR EMISSIONS41
Plant H-A H-B
Lime feed rate Mg/hr 12.7 12.7
Water feed rate I/mm 106 170
Hydrated lime
production Mg/hr 15.4-16.3 15.4
Particulate emissions
g/dry standard m3 0.065 0.055
g/m3 0.029 0.025
kg/hr 0.53 0.43
kg/Mg feed 0.042 0.034
9.7-70
-------
plant settling ponds and disposed of in landfills or in mined out quarries.
The wastewater impact of emission control alternatives for lime production
plants is minimal or nonexistent if available technologies are fully em-
ployed (see Table 9.7-23). Normally, lime plants that use water scrubbers
operate closed water systems with total recycle; therefore, zero waste
water effluent is achieved.41
The energy impact of various control devices is discussed in Section 6
of Volume 1. Energy consumption and solid waste generation factors are
presented in Table 9.7-26 for control options applicable to the rotary kiln.
9.7.7 Cement Manufacturing
Cement manufacturing consists of producing complex calciurn-si!icate-
aluminate-ferrite materials which, when mixed with water, form a binding
material for aggregates (crushed stone, gravel, and sand) in "concrete."
Products include a variety of portland cements, masonry cement, and calcium
aluminate cement. The portland cements are dominant in this industry and
account for approximately 95 percent of the total volume produced. Masonry
cement and calcium aluminate cement account for the remaining 5 percent.46
In 1978, 164 cement plants produced 72 Tg of portland cement, and minor
quantities of masonry and calcium aluminate cement.47
The size of portland cement plants, as gauged by production capacity in
1979, ranged from 50 Gg per year to 2.16 Tg per year.47 Mean plant pro-
duction capacity was 480 Gg per year.
Particulate matter is the primary emission in the manufacture of port-
land cement; it amounted to 208 Gg in 1977 on a nationwide basis.4
9.7.7.1 Process Description. Portland cements of several types are
manufactured using two processes known as the "dry process" and the "wet
process" (see Figure 9.7-18). As of 1973, 36 percent of the plants used the
dry process and 62 percent used the wet process, with the remaining 2 percent
using both.48 Recent trends in the industry have been toward increased
use of the dry process because it is less energy intensive.
The major steps in the portland cement manufacturing process include:
o Grinding and blending
o Kiln operations
o Finish grinding and packaging
9.7-71
-------
Table 9.7-26. ENERGY CONSUMPTION AND SOLID WASTE PRODUCTION FACTORS
FOR CONTROL DEVICES APPLIED TO ROTARY LIME KILNS23
Energy consumption, Solids disposal,
Control device kilowatt hours per kilograms per
Mg of lime Mg of lime
Baghouse 5.34 170
High efficiency precipitator 7.03 170
Medium efficiency precipitator 4.87 170
5.5 kPa water scrubber 33.5 211
3.7 kPa water scrubber 23.8 211
2.2 kPa water scrubber 14.2 211
Gravel bed filter 8.9 170
9.7-72
-------
QUARRYING
RAW
MATERIALS
PRIMARY AND
SECONDARY
CRUSHING
RAW
MATERIALS
STORAGE
DRY PROCESS
RAW
MATERIAL
PROPORTIONED
to
*
-•J
CO
SLURRY MIXING
AND
BLENDING
•
STORAGE
STORAGE |—
SHIPMENT
Figure 9.7-18 Basic flow diagram of portland cement manufacturing
process-31
-------
Cement is made from a combination of calcareous materials (limestone or
other calcium carbonate substances), argillaceous materials (clay or similar
substances), siliceous materials (sand), and ferrous materials (iron ore).
In the dry process, these materials are dried, ground, and mixed in powder
form; in the wet process, water is added, and the raw materials are mixed
and ground to form a slurry. The dry or wet materials are charged into the
upper end of a rotary kiln to be calcined or burned, by heating to approxi-
mately 1500°C to form a material called "clinker." After the clinker is
discharged from the kiln, it is cooled, mixed with approximately 5 percent
gypsum, ground to a fine (size distribution unknown) powder, and packaged
for shipment.
Masonry cement is made by mixing crushed limestone and gypsum with
clinker and grinding to a fine powder. Calcium aluminate cement is made by
fusing a mixture of limestone and bauxite in a kiln and then grinding the
kiln product.
9.7.7.2 Emission Characteristics and Applicable Control Techniques.
Table 9.7-27 is a computer combined and modeled, size specific, emission
data set from the Fine Particulate Emission Information System (FPEIS).49
Original emission data were from a source test of a gas-fired wet cement
kiln operating at 97 to 110% of full capacity. The control device was a wet,
hot side, low voltage electrostatic precipitator. Although other data are
available within the FPEIS, the data used were determined to be most suitable
for this publication.
Sources of dust at cement plants include quarrying and crushing, raw
material preparations, grinding and blending (dry process only), clinker
production, finishing grinding, and packaging. These sources are listed in
Table 9.7-28. Control techniques applicable to process and fugitive emis-
sions from the quarrying and crushing operations, and the preparation of the
raw materials are discussed in detail in Section 9.7.1 and Section 5 of
Volume 1.
Exhaust gases from the clinker production kiln contain substantial quan-
tities of particulate matter and are generally the largest source of air
emissions in the plant. The size distribution of uncontrolled emissions is
detailed in Table 9.7-29.50,51
9.7-74
-------
Table 9.7-27. SIZE SPECIFIC EMISSIONS FROM A WET CEMENT KILN
^J
en
Uncontrolled
Controlled
Efficiency,
percent
Mass
Total 15.3 fim
2.02 E7a 1.76 E4
(80.9)
67.4 68(100)
99.6 99.6
concentration, mg/DNCM (mass percent
12.9/im
1.65 E4
(76.1)
67.7(100)
99.6
10.1 /tm
1,48 E4
(68.3)
67.1(99.5)
99.5
7.28Mm
1.25 E4
(57.5)
66.7(98.9)
99.5
less than stated size)
5 (j.m
1.0 E4
(45.8)
67.7(98.9)
99.3
2.5 fj.m
6.0 E3
(27.8)
63.1(93.7)
98.9
l.Ol^m
1.12 E3
(5.18)
14.1(20.9)
98.7
aE7 means 10?, etc.
-------
Table 9.7-28. SOURCES OF AIR EMISSIONS IN CEMENT
MANUFACTURING PLANTS46
1. Preparation of raw materials
2. Crushing operations
3. Preparation of raw materials
4. Kiln operation
5. Clinker cooling
6. Finish grinding
a.
b.
c.
d.
a.
b.
c.
d.
e.
a.
b.
c.
a.
b.
a.
b.
a.
b.
c.
d.
e.
Drilling
Blasting
Loading broken rock
Transporting or conveying to
cement plants
Unloading rock from quarry
Crushing rock
Screening rock
Conveying to and from storage
Storage
Drying operations
Conveying and feeding to grinding
circuit
Grinding of raw materials and
conveying of ground material
(dry process)
Feeding raw material to kiln(s)
— dry process
Gases exhausted from kiln(s)
Excess air exhausted from clinker
cooler(s)
Conveying clinker from cooler(s)
to finish-grinding mill(s)
Conveying clinker from storage to
finish-grinding mill(s)
Finish grinding of clinker,
gypsum, and additives
Air classification of finished
product and conveying to storage
Storage
Bulk loading operations
9.7-76
-------
Table 9.7-29. SIZE DISTRIBUTION OF DUST EMITTED FROM
KILN OPERATIONS WITHOUT CONTROLS50,51
Kiln dust finer than corresponding
Particle size particle size,
micrometers %
60 93
50 90
40 84
30 ' . 74
20 58
10 38
5 23
1 3
9.7-77
-------
The cement industry uses mechanical collectors, electrostatic precipi-
tators, gravel bed and fabric filter collectors, or combinations thereof,
depending upon the operation and exhaust gas temperatures. Although high-
energy wet collectors (venturi scrubbers) are used in several plants, they
are not generally used in the portland cement industry.52
The distribution of kiln dust collection equipment in 101 cement plants
surveyed in 1975 is shown in Table 9.7-30. The effectiveness of the listed
control devices is dependent on the characteristics of the gas stream and
the particulate matter—specifically the size of the particles, the moisture
content of the gas, the resistivity of the dust, and the concentration and
composition of the dust. Mechanical collectors are not effective on subnri-
crometer particles and, therefore, are used only as a precleaner to a fabric
filter or ESP. The dust collected by these precleaners is recycled to the
kiln when its chemical composition does not alter that of the final product.
Table 9.7-31 contains a summary of compliance test data for various processes
and control technologies.53
Nhen baghouses are used to control dry process kilns, gas temperatures
are of primary concern. Kiln exhaust gases must be cooled to at least 315°C
before entering fabric filters. Glass and Nomex(R) fabrics, which withstand
290°C and 230°C, respectively, are the most commonly used materials for bags.
Higher temperatures accelerate the aging of bag fabrics.52
When either electrostatic precipitators or fabric filters are used on
wet-process kilns, extensive thermal insulation must be provided to prevent
condensation of water vapor within the device. Although some precipitators
are specified to withstand a maximum temperature of 370°C, the usual
operating range is 150 to 260°C. Wet-process kiln gases exhibit the proper
moisture and temperature characteristics for effective electrostatic
precipitation.54 Water conditioning improves particle resistivity in dry
kilns, and reduces the temperature. Several preheater installations
utilize the kiln exhaust gases to dry the raw material. This increases the
moisture content of the gas and reduces its temperature. Fabric filters
applied to kilns are designed with filtering velocities of:
0.37 to 0.46 m/min reverse air cleaning
2.1 to 3.0 m/min for pulse jet cleaning
0.61 to 0.91 m/min for mechanical shaker cleaning
9.7-78
-------
Table 9.7-30. DISTRIBUTION OF KILN DUST COLLECTION SYSTEMS IN WET AND
DRY PROCESS CEMENT PLANTS3!
Type of process
and
number of plants
Kiln-dust collection system
Single dust collector
Cyclones
Precipitators
Baghouses
Wet scrubbers
Settling chamber
Combinations of dust collectors
Wet
2
31
3
1
1
' Dry
2
3
3
0
0
Precipitators and wet scrubbers
Cyclones and wet scrubbers
Cyclones and precipitators
Cyclones and baghouses
Cyclones, baghouses, and precipitators
Baghouses and precipitators
Baghouses and wet scrubbers
1
1
14
4
2 .
1
0
0
0
12
16
2
1
1
9.7-79
-------
Table 9.7-31. ANALYSIS OE PORTLAND CEMENT PLANT COMPLIANCE TEST DATA53
Particulate
emissions,
kg/Mg feed
Number
of
Variable tests
Dry process
ESP
Baghouse
Wet process
ESP
Baghouse
All processes
ESP
Baghouse
All controls
Dry process
Wet process
All kiln data
Clinker cooler
Baghouse
Gravel bed
Wet scrubber
All controls
8
10
6
4
14
14
18
10
28
16
3
1
20
Minimum
0.021
0.013
0.020
0.049
0.020
0.013
0.013
0.020
0.013
0.005
0.023
0.022
0.005
Maximum
0.125
0.123
0.142
0.132
0.142
0.132
0.125
0.142
0.142
o.oeia
0.045
0.022
0.061
Mean,
kg/Mg
feed
0.061
0.070
0.084
0.091
0.070
0.076
0.066
0.087
0.073
0.022
0.034
-
0.024
Statistic
Standard
error
0.012
0.013
0.021
0.021
0.011
0.011
0.009
0.014
0.008
0.004
0.006
-
0.003
Standard
deviation NSPS
0.033 0.15
0.041
0.051
0.043
0.041
0.041
0.037
0.046
0.040
0.016 0.05
0.011
-
0.015
aA single compliance test shows 0.061 kg/Mg. The EPA states that the
source is in compliance.
9.7-80
-------
Electrostatic precipitators are designed for a drift velocity of 0.061 to
0.091 m/s.55
The clinker cooler is the second largest air pollution source in cement
plants. Because of the relatively small particle size of this dust, fabric
filters, electrostatic precipitators, and granular bed filters are used.56
Granular bed filters are popular because of their cost, relatively low main-
tenance requirements, and ability to withstand higher temperatures than
conventional fabric filter collectors and electrostatic precipitators.57
There are numerous other particulate emission sources within a cement
manufacturing facility, although they are less significant than the kiln or
clinker cooler. Baghouse collectors appear_to be most frequently used to
control emissions from these various sources. Advantages and disadvantages
of the use of various control devices on these sources as well as on the
kiln and clinker coolers are given in Table 9.7-32.
For dust control from emission points other than the kiln, most tech-
niques involve the capture of dust by drawing ambient air through a hood or
enclosure. To assure capture of the particles, an air intake velocity of
1.0 to 1.25 m/s is required. The dust-laden air is ducted to dust collec-
tors. To prevent dust from falling out within the capture and transport
system, an air velocity of above 18 m/s, and preferably about 20 to 23 m/s,
should be maintained.46
Fugitive emissions from materials handling, storage, and loading and
unloading operations can be reduced by using a variety of controls. These
include enclosures and hoods ducted to dust collectors, sprinkling systems
for dust suppression (using water, foam, or chemicals), improved handling
techniques, better housekeeping, and combinations of these and other con-
trols. Plant roads can be paved, watered or oiled, treated with chemicals,
or swept regularly to minimize dust reentrainment. For additional details
on fugitive emission control techniques, see Section 5 of Volume 1.
9.7.7.3 Secondary Environmental Impacts. Disposal of collected dust
creates a major solid waste problem. Dust with a high alkali content cannot
be recycled into the cement manufacturing process and therefore must be
disposed of. Proper disposal methods are needed to avoid potential ground
water contamination by the leaching of alkaline salts. The collected kiln
9.7-81
-------
TABLE 9.7-32. ADVANTAGES/DISADVANTAGES OF CONTROL DEVICES FOR VARIOUS CEMENT
MANUFACTURING OPERATION$52
i
co
ro
Operation
Quarrying
Crushing and
grinding
Raw material
storage
Integral
preheater
and kiln
Mechanical collectors
Not applicable
Not applicable
Not applicable
Integral part of preheater
countercurrent gas and
material flow; high energy
controls necessary to
meet opacity requirements
Fabric filters
Not applicable
Very good
Very good
Very good; must contend
with temperature
reduction, dewpoint
Electrostatic precipltators
Not applicable
Not economically feasible; low
flow volumes
Not economically feasible
Very good, if gas stream is
properly conditioned
Kiln
Clinker
cooler
Finish
grinding
Finished
material
storage
Packaging and
shipping
Used as precleaners for
high energy devices
Used as precleaners for
high energy devices
Cannot meet opacity
requirements
Cannot meet opacity
requirements
Cannot meet opacity
requirements
Very good; must contend
with temperature
reduction, dewpoint
Very good; must contend
with abrasive particu-
late gravel-bed filters
recently introduced
Very good
Very good
Very good
Very good; must contend with
dewpoint particle resistivity,
and explosion potential problems
Precipitator design must contend
with combination of clinker dust
and moisture, possibly coating
ESP interior with cement
Very good on large mills
Not economically feasible; low
air flow volumes
Not economically feasible; low
air flow volumes
-------
dust may be used as a substitute for agricultural limestone, fertilizer, or
mineral filler. Alternatively, the dust is often hauled to landfill sites or
abandoned quarries or storage piles. The piles of dust should be protected
from erosion by using covers or enclosures, or spraying with water to form a
surface crust.
9.7.8 Concrete Batching
Concrete batching involves the proportioning of sand, gravel, and cement
by means of weigh hoppers and then conveyed into a mixing receiver such as a
transit mix truck. The required amount of water is also discharged into the
receiver along with the dry materials. In some cases, the concrete is pre-
pared for on-site building construction work or for the manufacture of con-
crete products such as pipes and prefabricated construction materials.
Particulate emissions consist primarily of cement dust, but some sand
and aggregate gravel dust emissions do,occur during batching operations.
There is also a potential for dust emissions during the unloading and con-
veying of concrete and aggregates at these plants and during the loading of
dry-batched concrete mix. Another source of dust emissions is the traffic
of heavy equipment over unpaved or dusty surfaces in and around the concrete
batching plant. In 1977, nationwide.particulate emissions amounted to 15 Gg.4
9.7.8.1 Process Description and Emission Control Techniques. Concrete
batching plants store, convey, and measure the materials to make concrete;
they then blend'them and transfer the mixture to trucks for shipment. There
are three types of plants—wet batching, dry batching, and central mix
plants. They all receive, store, transfer, and blend the solid raw materials
in similar ways, but they.add water to the mix at different points in the,
process. The average plant will produce 59 Gg of concrete per year.31
Crushed and sized raw materials are delivered to the plant by truck or
rail, and then transferred to elevated storage silos and bins. Cement is
usually transferred pneumatically or, less frequently, by bucket elevator,
while sand and aggregate are conveyed by belt conveyor or bucket elevator.
The raw materials are weighed by a weigh hopper to proportion the proper
amounts for mixing. In the wet batching plant, the sand, aggregate, and
cement are mixed with water, and then poured into transit mix trucks which
mix the concrete on the way to the site where it will be used. In dry batch
9.7-83
-------
plants, the dry mixture of aggregate and cement is transferred in flat bed
trucks to paving machines at the site, where water is then mixed in. Most
plants that do dry batching also do wet batching. A central mix plant makes
wet concrete in a central mixer, and transfers it by open bed dump trucks to
the job site. A process flow diagram showing emission points is presented
in Figure 9.7-19.
Particulate matter is emitted in significant quantities from the receiv-
ing and conveying of cement, sand, and aggregates and from the load-out of
concrete.32 Fugitive dust can be emitted from the following points:
o Sand and aggregate storage piles on the ground
— loading onto the pile
— vehicular traffic
— loading out from the pile
— wind erosion of the pile
o Transfer of sand and aggregate to elevated bins
o Cement unloading to elevated storage bins
o Weigh hopper loading of cement, sand, and aggregate
o Mixer loading of cement, sand, and aggregate (at a central mix
plant)
o Loading of transit mix trucks (at a wet-batching plant)
o Loading of batch truck (at a dry-batching plant)
o Plant roads
In a wet batching plant, almost all the dust generated is cement dust,
since most of the sand and aggregate is damp. However, dry light aggregates
create considerable dust when handled. Typically, between 10 and 20 percent
by weight of the dust is less than 5 micrometers in size, depending on the
grade of the cement.58 fhe dust that dry batching plants generate has
similar characteristics.
The extent of emissions and control methods vary with the type of plant
and the characteristics of the aggregate used. Dust can be controlled by
preventive procedures and operating changes, or by capture methods using
hoods and covers, and then venting to dust removal equipment.
The amount of dust emitted during the transfer of sand and aggregate
depends on their moisture content. Wet suppression with water sprays
9.7-84
-------
SAND AND
AGGREGATE
STORAGE
to
00
LEGEND:
ELEVATOR
E
ELEVATED
STORAGE
BINS
SAND
AGGRE
GATE
/ WEIGH
\ HOPPERS
WATER (WET-BATCH)
.V
TRANSIT
MIXER
(WET BATCH)
FLAT-BED TRUCK (DRY BATCH)
MIXER
H
ELEVATED
CEMENT
SILO
±
cc
o
et
>
LU
I
s-
1
•«
i
•x
PNEUMATIC
(RAILCAR)
TRUCK
(CENTRAL-
MIX PLANTS)
EMISSION POINT
PROCESS" FLOW
9
DUMP TRUCK
(CENTRAL MIX)
Figure 9.7-19 Process flow diagram for concrete batching showing
potential industrial process fugitive particulate
emission points.7
-------
temporarily prevents dusting. Conveyors and transfer points used during the
loading of the mixer can be enclosed to prevent dust loss. Using a pneumatic
transfer system, rather than bucket conveyors, eliminates emissions during
the transfer of cement to storage silos. Fabric filters or fabric "socks"
are used to control dust discharged through cement silo vents. The filter
on these vents is made of cotton sateen with a filtering velocity of approxi-
mately 0.91 meters/minute. This same filtering rate can be used in fabric
filters applied to the weigh hopper and dry batching or central mix plant
mixers.59
Fugitive emissions can be controlled by using movable or fixed hoods
over the discharge end of the mixer or over the receiving hopper of a truck
or weigh hopper. The collected dust can then be vented to a fabric filter
or scrubber. See Section 5 of Volume 1 for additional details regarding
fugitive dust control techniques. A summary of the controls applied to the
emissions sources within a concrete batching plant is given in Table 9.7-33.
9.7.9 Asphalt Concrete Plants
Asphalt and crushed stone (aggregate) are combined at asphalt hot-mix
plants to form asphaltic concrete, which is used in the paving of highways,
streets, and parking surfaces. As of April 1977, there were 4539 asphalt
concrete plants distributed throughout the United States; approximately
60 percent of these plants were stationary and the remaining portion were
portable plants.60 National production is approximately 270 Tg per year,
with a continued increase in use projected for the future. Types of plants
include batch-mix (90.8 percent), continuous mix (6.6 percent), and dryer-
drum mix (2.6 percent).2 Typical plants produce between 45 and 320 Mg/hr
of hot mix, and operate 1200 hr/yr.58'60
Condensed hydrocarbons (asphalt fumes) and mineral aggregate dust
are both emitted during asphalt concrete batching. Nationwide emissions
in 1977 have been estimated at 136 Mg/year.4 The size of the mineral
particles ranges from submicrometer to approximately 100 micrometers,
the precise range depending on the type of aggregate being used. Crushed
limestone, sandstone, ore tailings, basalt, granite, sands, and gravels
can all be used. In most cases, the emitted particles are less than
74 micrometers, of which 10 to 50 percent are less than 5 micrometers.
9.7-86
-------
Table 9.7-33. EMISSION SOURCE CONTROLS58.59
Emission source
Control
1. Sand and aggregate storage
2. Transfer of sand and aggregate
to elevated bins
3,
4.
Cement transfer to elevated
storage silos, and silo
vents
Weigh hopper loading of
cement, sand, and aggregate
5.
6.
'Mixer loading of cement, sand,
and aggregate (central mix
plant)
Loading of transit mix
truck (wet batching)
7. loading of flatbed truck
(dry batching)
8. Plant roads
(Refer to Chapter 5 of Volume 1 for discus-
sion of controls for storage piles.)
o Wet suppression with water of the feed,
transfer, and discharge points
o Enclose partially or completely the
conveyor system
o Exhaust dust laden air from transfer
points to fabric filter
o Maintain conveyor equipment to
prevent leaks
o Enclose bucket elevators
o Maintain conveyor equipment
o Use pneumatic transfer
o Fabric filters on cement silo vents
o Fabric sock on cement silo vents
Fixed hoods, curtains, partitions, or
covers (canvas shroud)
Fabric filter
Scrubber
Vent the displaced air to the storage
bins and silo, or vent it to a
central collecting system (filter)
Movable hoods with flexible ducts
Fabric filter
Scrubber
o Enclose the rear of the mixer with a
shroud
o Fixed hoods, curtains, partitions or
covers
0 Movable hoods with flexible ducts which
enclose the receiving hopper
o Fabric filter
o Scrubber
Same types of controls as wet batch loading
(Refer to Chapter 5 of Volume 1 for discus-
sion of controls for plant roads.)
9.7-87
-------
9.7.9.1 Process and Emission Control Techniques. Asphalt hot-mix pro-
duction consists of mixing a combination of aggregates with liquid asphalt.
The asphalt plant is used to heat, mix, and combine the aggregate and asphalt
in measured quantities to produce the required paving mix. Figure 9.7-20
gives a detailed process and materials flow diagram for a representative
159-Mg/hr asphalt concrete plant.
Aggregate of appropriate mix is fed into a rotary dryer (Stream 1) at a
controlled rate. The aggregate, generally composed of locally available
material, will contain both coarse-sized crushed rock and fines. The rotary
dryer is an inclined rotating cylinder (usually employing oil or gas as fuel)
into which the aggregate is fed at the raised end and discharged at the
lower end.
As the aggregate leaves the dryer, it drops into a bucket elevator and
is transferred to a set of vibrating screens where it is classified by size.
The classified hot materials then enter the mixing operation. After all the
material is weighed, the sized aggregates are dropped into a mixer and dry
mixed for about 30 seconds. The asphalt, which is solid at ambient tempera-
tures, is pumped from heated storage tanks, weighed, and then injected into
the mixer. The hot mixed batch is then dropped into a truck and hauled to
the job site.
The rotary' dryer is the principal process point source of particulate
emissions in a hot-mix asphalt plant (Stream 2). The vibrating screens,
bins, weigh hopper, and mixer are also emission sources that need to be con-
trolled. These areas are normally enclosed and the dust emitted is carried
by ventline to the control system (Stream 7A or 7B). Typical exhaust gas
and stack characteristics are listed in Table 9.7-34.61 Fugitive dust emis-
sions arise from fine aggregate stockpiles, loading operations, cold storage
bins, the cold aggregate conveyor, and truck traffic. These emissions are
caused by natural elements, poor housekeeping, exposed aggregate stockpiles
and storage bins, and uncontrolled traffic conditions. For a discussion of
fugitive particulate control methodologies, see Section 5 of Volume 1.
Table 9.7-35 lists uncontrolled size specific emissions from a rotary
dryer at an asphalt batching plant operating at 136 to 159 Mg/hr (150
to 175 Tons/hr). This computer modeled data from the FPEIS was derived
from source test results of one plant.62
9.7-88
-------
ID
*
••vl
I
c»
PRIMARY COL-
LECTOR (CYCLONE)
(96%
EFFICIENCY)
DISCARD
143 kg/H
CYCLE * 141'3 k9/H
H
t
S~\ ^>-
1.43 kc
SECONDARY COLLEC-
TOR (FABRIC FILTER]
(99% EFFICIENCY)
SECONDARY
TOR (VENTL
SCRUBBER)
EFFICIENCY
COLLEC-
IRI
(98%
)
y>
/H
2.<
kg/
i j
)
'H*
140.0 kg/H
WATER AND MUD
DISPOSAL AS VIA
SETTLING POND
Figure 9.7-20 Material flow for representative 159-Mg/hr asphalt concrete
plant.57
-------
Table 9.7-34. ASPHALT CONCRETE PLANT EXHAUST GAS AND STACK
CHARACTERISTICS FOR THE ROTARY DRYER61
Stack parameters
Number of sources 2825
Average stack height, m 10.2
Average stack diameter, m 1.5
Average stack gas temperature, °C 93
Average gas flowrate, A m3/s 16.7
9.7-90
-------
Table 9.7-35. SIZE SPECIFIC EMISSIONS FROM A ROTARY DRYER
ASPHALT BATCHING PLANT62
Mass concentration, mg/DNCM (mass percent less than stated size)
Total 15.3/xm 12.9/xm 10.1/*m 7.28 ^m 5 fm 2.5 pm 1.01/im
Uncontrolled 55500 6220(11.2) 4560(8.2) 3390(6.1) 2170(3.9) 1300(2.4) 401(0.7) 112(0.2)
•vl
I
U)
-------
A single control system is typically used to control emissions from
the rotary dryer, vibrating screens, storage bins, weigh hopper, and mixer.
This system usually consists of a product recovery collector (primary) fol-
lowed by a secondary collector. An individual survey of the asphalt hot-mix
industry indicates that the primary collectors preferred are usually dry
collectors. Table 9.7-36 lists the types of primary and secondary collectors
and the extent of industry usage.6^ Table 9.7-37 is a summary of tests show-
ing the capabilities of these collectors.64
The primary collectors usually employed to control emissions from the
rotary dryer and sometimes from the mixing operations, have relatively low
efficiencies for particles less than 20 to 30 micrometers. Therefore, a
primary collection alone will not suffice to meet current New Source
Performance Standards (NSPS), a but may be used to facilitate the recycling
of larger particles and to improve the performance of the secondary control
system.60 A secondary control device of the type specified in Figure 9.7-20
is usually necessary for meeting current NSPS.
Wet scrubbers used in the asphalt concrete industry consist of:
o Gravity spray towers
o Cyclonic scrubbers
o Centrifugal fan wet scrubbers
o Venturi scrubbers
o Orifice scrubbers
As with most wet scrubbers, those with low pressure drops also are charac-
terized by low efficiencies. Higher efficiencies can be achieved by in-
creasing the power input and to some extent the liquid to gas ratio.6^
Fabric filters are considered by many to be the most effective parti-
culate control device for this application. The fabric filters commonly
used are of the pulse-jet type, utilizing compressed air to sharply reverse
the air flow for bag cleaning. These fabric filters use typical filtering
velocities of about 2 M^/min of air per square meter of cloth.
aThe NSPS requires emissions of no greater than 0.04 grains/dscf, and
opacity at the stack of no greater than 20%.
9.7-92
-------
Table 9.7-36. PRIMARY AND SECONDARY CONTROL DEVICES USED IN
THE ASPHALT HOT MIX INDUSTRY^
Type of control equipment Percent of industry
Primary collectors
Settling or expansion chambers 4
Centrifugal dry collectors 58
Multicyclones 35
Other 3
Secondary collectors
Gravity spray tower .8
Cyclone scrubber 24
Venturi scrubber 16
Orifice scrubber 8
Baghouse 40
Other 3
9.7-93
-------
Table 9.7-37.' HOT MIX ASPHALT CONCRETE PLANTS64
Plant
Design
capacity
Production
rate
Fuel
Paniculate
emissions
Probe t fil-
ter catch
g/DHCH
(gr/dscf)
g/ACH
(gr/acf)
Kg/hr
(Ib/hr)
Total
catch
g/ONCH
(gr/dscf)
g/ACH
(gr/acf)
Kg/hr
(Ib/hr)
Control
device
"Tests used to
"Inspection of
CFabrlc filter.
V V
120 120
99.7 lllj
oil gas
0.153 0.048
(0.0067) (0.021)
0.0101
(0.0044)
0.43 1.27
(0.94) (2.8)
0.0497
(0.0217)
0.0323
(0.0141)
1.37
(3.02)
FFC FF
Ba
300
212.3
oil
0.019
(0.0031)
0.010
(0.0044)
0.66
(1.46)
0.132
(0.0575)
0.0684
0.0299
4.57
(10.07)
FF
C»
200
125.5
oil
0.039
(0.017)
0.032
(0.014)
2.44
(5.37)
0.041
(0.018)
0.037
(0.016)
2.58
(5.68)
Venturl
(5 kPa)
0'
240
222.5
gas
0.041
(0.0178)
0.024
(0.0104)
1.65
(3.64)
0.206
(0.0899)
0.120
(0.0524)
8.37
(18.45)
FF
E"
180
—
gas
0.043
(0.0189)
0.0247
(0.0108)
1.59
(3.51)
0.0617
(0.027)
0.0345
(0.0151)
2.31
(5.10)
FF "
H,'
240
180
oil
0.072
(0.0315)
0.0588
(0.0257)
3.33
(7.35)
0.163
(0.0713)
0.133
(0.0582)
7.66
(16.88)
Venturl
(4.6 kPa)
H2a I" J»
120
200
oil gas
0.025 0.057 0.023
(0.011) (0.025) (0.01012)
—
2.93 1.07
(6.45) (2.362)
0.050 0.059
(0.022) (0.026)
—
2.29 3.06
(5.05) (6.75)
Venturl Venturl FF
(4.9 kPa) (6.5 kPa)
. Ka Lb Mb
..
120 200 153
oil
0.247 0.016 0.098
(0.108) (0.0072) (0.043)
-
8.35 0.794 3.05
(18.4) (1.75) (6.735)
_
-
—
FF , Fr FF
support a new source performance standard.
dust collectors showed evidence of poor maintenance or operation.
-------
Fabrics currently used within the asphalt hot mix industry include
glass yarns treated with lubricants such as silicone to prevent fibers from
breaking due to self-abrasion during flexing (good for continuous operation
up to 260°C), polyesters (maximum temperature of 1320C), Nomex(R)-type
nylon (good for temperatures up to 234°C), and glass/Nomex(R) web on a
Nomex(R) scrim. Exhaust gases from the dryer contain large quantities
of water vapor within the 121° to 177°C temperature range. As a conse-
quence, the fabric filter unit should be insulated so that the exhaust gas
remains above the dew point.
Several process modification techniques can also be used to reduce emis-
sions from the aggregate dryer. Since particle entrainment in the exhaust
air stream is due in part to the air stream velocity, reducing this velocity
decreases uncontrolled particulate emissions. Another method that is becom-
ing increasingly popular is dryer-drum mixing. In this technique, atomized
asphalt is sprayed onto the undried, cold aggregate. The mixture is then
dried and mixed in a rotary dryer. Emissions:are reduced since the asphalt
coats the aggregate particles, preventing them from becoming airborne.
Fabric filters used to control emissions from this process must be designed
to withstand high temperatures to avoid asphalt clogging.
9.7.9.2 Secondary Environmental Impacts. In addition to the reduction
of particulate emissions due to the application of control equipment, there
are the following principal environmental effects:
o Solid waste disposal
o Water pollution concerns if wet scrubbers are used
o Increased electrical energy consumption
Approximately 50 percent of the solids collected by dry collection de-
vices can be recycled directly to the asphalt concrete mixing process; dispo-
sal must be made of the rest.60 All of the particulate matter collected by
wet scrubbers is in such a form that disposal must be made of it. The most
common disposal method for wet scrubber effluent is to divert the effluent
to settling ponds.
9.7-95
-------
9.7.10 Asphalt Roofing Plants
The asphalt roofing industry manufactures asphalt-saturated felt rolls,
shingles, rolls with surface mineral granules, and smooth rolls that may
contain a small amount of surface mineral dust or mica. While most of these
products are used in the construction of roofs, a relatively small number
are used in walls and in other building applications. In 1977, 8.6 Tg of
asphalt and tar roofing products were produced. In 1978, there were 118
asphalt roofing plants scattered throughout the United States.66
Asphalt shingles are prepared by: (1) impregnating roofing felt with an
asphalt saturant; (2) coating both sides of this product with a harder,
rougher asphalt; (3) embedding mineral granules into one side of the sur-
face; and (4) cutting the surface into strips or shingles. Regardless of
the weight of the asphalt saturant, the product is approximately 40 percent
percent dry felt and 60 percent saturant when it leaves the saturator.67
Particulate emissions from the asphalt roofing industry emanate from
two primary sources: aspahalt blowing stills and saturators. Blowing stills
are used to dehydrogenate the liquid asphalt. The progressive loss of hydro-
gen results in polymerization of the asphalt to a desired consistency.
Saturators consist of dip tanks and sprayers where the saturant is applied
to the felt.
Particulate emissions from the asphalt roofing industry in 1977 were
estimated to be 21 Gg.4 Figure 9.7-21 shows the size distribution of par-
ticulate matter from the uncontrolled saturator exhaust. Minimal data are
available to characterize emissions from the blowing stills; however, such
emissions emanate at higher temperatures (94° to 153°C) and at higher mass
loadings than saturator emissions. Opinions have also been expressed that
the asphalt emissions from air blowing are of a more tar-like nature than
those from saturator operations.^
9.7.10.1 Process Description. Asphalt flux is the "bottom" material
from the petroleum refining process. It can consist of the residues from a
single crude or from a blend of many crudes. A number of products are pro-
duced for the asphalt roofing industry; the principal products, however, are
the "saturant" asphalt and "coating" asphalt used in the production of
asphalt roofing and siding. The main difference in these two asphalts is
9.7-96
-------
10.0
9.0
8.0
7.0
6.0
5.0
4.0
3.0
o>
o
00
UJ
2.0
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
I I I
I
J I
I I t
I
I
I
5 10 20 30 40 50 60 70 80 90
PERCENT BY WEIGHT SMALLER THAN INDICATED SIZE
98 99
Figure 9.7-21 Particle size distribution in uncontrolled saturator
exhaust.6^
9.7-97
-------
their softening point. Saturants usually have a softening point between
40° and'74°C (104° and 165°F), while coating asphalts soften at about
110°C (230°F).
Asphalt is blown with air in an asphalt blowing still (Figure 9.7-22),
which is a tank fitted near its base with a sparger (air lines in a spider
arrangement). Air blown through the sparger rises through the asphalt,
participating i,n an exothermic oxidation reaction. Oxidizing the asphalt
has the effect of raising its softening temperature, reducing penetration,
and modifying other characteristics. Sometimes catalysts are added to assist
in this transformation. The time required for airblowing asphalt depends on
a number of factors, including the characteristics of the asphalt flux, the
characteristics desired for the finished product, the reaction temperature,
the type of still used, the air injection rate, and the efficiency with
which the air entering the still is dispersed throughout the asphalt. Blow-
ing times may vary in duration from 30 minutes to 12 hours.
Asphalt blowing is a highly temperature-dependent process because the
rate of oxidation increases rapidly with increases in temperature. Asphalt
is preheated to a range of 204° to 243°C (400° to 470*F) before blowing is
initiated to assure that the oxidation process will start at an acceptable
rate. Asphalt temperature is normally kept at about 260°C (500°F) during
blowing by spraying water onto the surface of the asphalt; however, external
cooling may also be used to remove the heat of reaction. The allowable upper
limit of the reaction temperature is dictated by safety considerations, with
the maximum temperature of the asphalt usually kept at least 28°C (50°F)
below the flash point of the asphalt being blown. Vertical stills, because
of their greater head (asphalt height), require less air flow for the same
amount of asphalt-air contact. Both vertical and horizontal stills are
still in use, but where new design is involved, a vertical type is preferred
by the industry because of the increased asphalt-air contact and consequent
reduction in blowing times. Asphalt losses from vertical stills are also
reported to be less than those from horizontal stills. All recent blowing
still installations have been of the vertical type. Asphalt blowing can be
either a batch process or a continuous operation. All stills at roofing
plants are believed to use the batch process, as do most of the asphalt
processing plants, but the ratio among refineries is unknown.
9.7-98
-------
to Atmosphere
CD
vo
to
fD
10
•v4
ro
ro
CO
O
0>
3
fu
rt
_j.
O
Q.
n>
o
-h
n>
-s
o
fa
•a
3-
EU
rt
Cr
O
_u.
3
(Q
(/I
Air
Blower
Charge Asphalt
from Heated
Storage
232 °C
50-150Nm3/min
45 Nrt\3/m',n
700 g/cm2
s~
•<-« *-i •» MI .. r-l
«*~—~-+*\
|
J
—
r^^
L-.
Control
Device
Fan
Sparger
232-270°C
Recovered
Asphalt
H2O
(Steam)
40,000 Liter
Capacity
Blowing Time
2-4hrs.
Blown Asphalt
to Storage
Pump
-------
The emissions from the blowing still are primarily organic particulates
with a fairly high concentration of gaseous hydrocarbons (6,000 to 7,000 ppm)
and polycyclic organic matter [112,308 g/Nm3 (0.00007 lb/ft3)]. The blowing
still has the highest total emissions of any of the emissions sources in an
asphalt roofing plant.
Figure 9.7-23 illustrates a roofing line. Roofing lines contain satura-
tor dip tanks or sprays, or both, where the saturant is applied to felt. The
saturated felt then passes through a drying-in section (wet looper), followed
by a coating operation where coating asphalt is applied to both sides of the
saturated felt. Although mechanical problems or breakage of the felt may
cause intermittent shutdowns of the roofing line, it is essentially a con-
tinuous operation.
These saturating, drying-in, and coating operations may be partially or
totally enclosed, with air and fumes being exhausted from the enclosure to a
control device (or directly to the atmosphere). The volume of gas exhausted
is on the order of 560 Nm3/min and consists primarily of air, water vapor,
asphalt liquid droplets (fumes), and gaseous hydrocarbons.
Several parameters that affect the emission rate from the saturation
process include in-draft air and hooding arrangements, characteristics of
the asphalt and felt, variations in the spraying/dipping process, line
speeds, and temperatures.
9.7.10.2 Control Techniques. Within the asphalt roofing industry,
approximately 40 percent of the plants do their own asphalt blowing; the
remainder buy their asphalt from refineries or asphalt processing plants.68
All such plants which have blowing stills and were contacted by EPA were
found to use some form of emissions control. Fumes are either ducted to a
direct-fired process heater or to an afterburner.68
Major control systems presently used in the asphalt roofing industry
include afterburners (or fume incinerators), high velocity air filters
(HVAFs), electrostatic precipitators, and wet scrubbers. A breakdown of
control device usage from a survey of 76 asphalt roofing plants is summa-
rized in Table 9.7-38. Results of particulate emission test data from
plants tested by EPA are summarized in Table 9.7-39.
9.7-100
-------
I
CD
Return
Asphalt
Storage
Saturant-Asphalt
Dry
Felt
Dry
Looper
OJ
Coating
Asphalt
i Sand
MixerV /
\f 280-560Nm3/min
D O DOQ
ft
Drying-ln
Section T°P
Coating
Continue to:
Top Surfacing
Backcoating
Cooling
Drying
Cutting & Packing
Saturating Tank
Figure 9.7-23 Schematic diagram of asphalt roofing line.65
to Atmosphere
Fan
-------
Table 9.7-38. CONTROL EQUIPMENT USED ON ASPHALT SATURATORS^S
Number of saturators of Total Control devices
28 37 Afterburners
18 24 High velocity air filters
10 13 Electrostatic precipitators
9 12 Wet scrubbers
11 14 Uncontrolled
76 100
9.7-102
-------
Table 9.7-39. EPA TEST DATA AT ASPHALT ROOFING PLANTS66
.— Emission source (2)
^ control device
measurement parameter
1
I-1 Parttculate
CO g/Hm3
g/n.3
kg/h
kg/Kg shingle
kg/«g felt
Control efficiency <
Volume flow rates
Nm3/sec
mVsec
Control device temp "C
Shingle production rate
Felt usage rate
Inlet
0.1494
0.1300
6.7585
0.2380
2.0270
12.33
14.14
Hg/hr
Hg/hr
Plant A ' Plant C
Saturator ft coater
storage tanks
Saturator 8 coater high velocity
electrostatic preclpitator air filter
Outlet 1 Outlet 2 Total outlet Inlet Outlet
0.0117
0.0101
0.2585
0.0190
0.1580
92.2
6.12
7.14
0.0089
0.0076
0.1814
0.0130
0.1110
94.5
5.72
6.64
52
27.85
3.27
0.0103
0.0089
0.4399
0.0160
0.1350
93.35
11.84
13.78
0.9565 0.0160
0.8146 0.0137
29.94 0.50
1.5700 0.0270
—
98.3
8.71 9.29
10.21 10.66
43
19.05
Plant 0
Saturator
high velocity
air filter
Inlet Outlet Inlet 1 Outlet 1
0.1442 0.0297
0.4119 0.0549
6.93 1.53 0.3318 0.0275
0.1600 0.0350 5.4880 1.2250
1.2500 0.2800
—
77.9
77.6
13.27 13.89 3.72 6.38
4.66 12.49
69 538
43.27 37.0
5.53 4.3
Plant 8
Saturator ft coater
afterburner
Inlet 2 Outlet 2 Total In
0.6018a 0.0229 0.5080
0.4554 0.0114 0.3936
7.0760 0.4990 12.565
0.3270
2.7700
92.9
3.19 6.13 6.91
4.25 13.88 8.92
649
Total out Storage ta
0.0389 0.0776
0.0183
1.724 1.00
0.0450
0.3800
86.3
12.52 0.38
26.37 0.43
__
-
-
Plant E
Blowing stills .
afterburner
Saturant asphalt Coating asphalt
nks Inlet
27.87
10.47
80.01
3.30
0.9
2,31
Outlet
0.364
0.185
5.58
0.230
93.40
4.21
8.15
816
Inlet
33.41
13.52
98.61
12.21
0.91
2.28
Outlet
0.210
0.117
3.27
0.405
96.70
4.29
8.10
816
-------
Low-voltage electrostatic precipitators have been installed on satura-
tors, wet loopers, and coaters. Particulate collection efficiencies of
approximately 93 percent have been demonstrated.^ Maintenance costs are
generally high on ESPs in this application since the exhaust air stream
contains cohesive tar-like particles. Additionally, water sprays used in
the ducts to lower the temperature form oil emulsions that are difficult to
disperse; however, detergents have been used with some success. Low-voltage
ESPs have received renewed interest by the industry as a result of fuel
shortages, afterburner costs, and the water pollution aspects of wet
scrubbers.
Low-energy scrubbers have demonstrated collection efficiencies in the
70 percent range when applied to saturators. This is not sufficient to
eliminate plume opacity and odors. High energy venturi scrubbers have
generally been avoided since operating costs are excessive and water clean-
up systems are required.
High velocity air filters (HVAFs) have been used for particulate control
on saturators, wet loopers, coaters, coater mixers, and storage tanks. Par-
ticulate collection efficiencies up to 98 percent have been demonstrated,
with a pressure drop of 7 kPa.^6 HVAFs are capable of up to 99 percent re-
moval efficiency if preceded by a precooling section of water sprays. This
may require liquid effluent clean-up. HVAFs consist of a fiber pad in roll
form which is advanced to expose a clean segment to the exhaust gases when
needed. Because of the resulting oil content of these rolls, disposal may
pose a solid waste problem.
The most popular control scheme currently for particulate matter con-
trol on blowing stills, saturators, wet loopers, and coaters is the use of
afterburners (fume incinerators). Additionally, afterburners have been used
to control emissions from storage tanks. Particulate control efficiencies
of over 93% have been demonstrated using combustion temperatures of 816°C.
Increased control efficiencies can be achieved by increasing combustion
temperature, residence time, and turbulence.66
Afterburners are installed both with and without heat recovery, accord-
ing to the needs of the individual plant. In general, heat recovery is more
economical for new facilities. However, it is not always possible to achieve
9.7-104
-------
maximum recovery since the afterburner exhaust often contains more heat than
the process requires. Maximum heat recovery is realized only when the roof-
ing manufacturing facility also produces the paper (dry felt), since the
felt-drying process requires copious quantities of heat in the drying drums.
The 95° to 760°C afterburner exhausts can be used for generating the steam
for this process or can be diluted with ambient air and blown directly over
the drying drum. 'Data have shown maximum heat utilization from this process
in the range of 40 to 50 percent.67 Recovered heat can also be used for
preheating the fumes entering the afterburners, preheating the asphalt flux
for blowing, and preheating asphalt in the saturators and/or in the blowing
operation.
Afterburners now in use combust either natural gas or No. 2 fuel oil.
The availability and cost of fuel is one disadvantage of this system. After-
burners are less expensive to install than other particulate collectors.
Neither HVAF nor ESP units have been used for the control of airblowing
emissions. It is recognized that the higher particulate mass loadings in
the air-blowing exhaust will increase the filter mat usage rate in a HVAF
and will seriously compound the buildup and fouling problems in an ESP.69
The handling of the mineral stabilizer and the sand, talc, and mica
parting agents causes emissions of dust particles during receiving opera-
tions, transfer, and application. These emissions are well controlled with
fabric filters. The use of a fabric filter with pneumatic materials receiv-
ing and handling systems is usually an integral part of the plant process.
Particulate control efficiencies in excess of 99 percent are common for
fabric filter devices applied to these emissions.67
Fugitive emissions from saturators, wet loopers, coaters, and transfer
points are usually captured with enclosures. Table 9.7-40 summarizes fugitive
emissions test data from four plants.
9.7.11 Glass Manufacturing
9.7.11.1 Types of Glass. The glass manufacturing industry is classi-
fied by industry definitions in the Standard Industrial Classification (SIC)
system. Under this system of classification, an industry is generally
9.7-105
-------
Table- 9,7-40. SWSWRY OF FUGITIVE EMISSIONS DATA FROM CAPTURE SYSTEMS66
-vj
I
O
CD
Plant
A
D
01
C-2
D-l
0-2
Number of
Source observers
Asphalt
saturator
Het looper
and coater
Asphalt
saturator,
wet looper
and coater
Asphalt
saturator
Het looper
and coater
HVAF,
1n1*t duct
Asphalt
saturator
Het looper
and coater
Asphalt
unloading
tanker
transfer to
2
2
2
2
2
2
2
2
2
Total
observation
tlrae (hours)
12
12
12
6
6
12
10.8
1.0
6
•storage tanks
Emission
location
Saturator
and wet
looper
Coater
Top 1 ft
of door
openings
Saturator
Coater
HVAF/duct
Interface
Coater end
of hood
Top of hood
(2 holes)
Pipe
connection
Hatch
cover
Emissions
Type and
color
None
White
fumes
Gray
fumes
White
puffs
Hhlte
fumes
White
puffs
Gray
fumes
Gray
fumes
None
-Gray
fumes
Frequency
N/A
Constant
Intermittent0
Intermittent
Variable
but constant
Intermittent
Almost
constant
Constant
N/A
Intermittent*
Opacity
range
percent3
0
0
10
5
0-22
0-22
0-2
0-2.3
0.2-5.8
1.5-6.3
0
0-146
0-15
0-10
20
0
o-iof
Vlslblen
0
0
100
100
34
31
7
38
100
100
0
20
98
93
100
0
10
5X
Opacity
0
ft
100
100
17
5
0
0
1&
20
0
t>
S2
39
100
0
10
101 .
Opacity!*
0
0
100
0
5
3
0
0
0
a
«
0
2r
2
ie&
0
10
Cements
Enclosure houses saturator and wet
looper. Wet looper door open
during tests. Coater hooded. Data
recorded only during normal
operation.
Enclosure houses saturator, wet
looper, and coater. Coater door
open during tests. Data recording
halted only when line stopped.
Enclosure houses saturator and wet
looper, Coater hooded. Testing
Interrupted when, more than one
door in enclosure was open.
Ducting not tightly sealed
at HVAF 1nTet.
Saturator, wet tooper, and coater
are hooded {mxnclosure).
Data records* -only during
nomiaT operation
Tanker solid-coupled to piping
through' a flexible tiose. Tanker
hatch cover normally open 1-2 Inches
for -venting;
JRange of opacities after data collated Into 6-mfnute averages.
"Percent of total observation time opacities were visible, were 5 percent opacity or higher, or were 10 percent or higher.
cEmissfons were visible only when doors were open. Unlike other tests, these were-not halted when doors were open to correct
process problems unless Hne was shut down.
•'Emissions were visible only when hatch cover was open more than 1-2 In.
epuffs lasted less than 15 seconds. Maximum opacity was 10 percent.
fMaximum opacity (30 percent) was experienced when hatch was opened fully for 2 mln.
-------
defined as a group of establishments producing a single product or a more or
less closely related group of products. Accordingly, for the glass industry
there are four SIC codes:
SIC 3211—Flat glass
SIC 3221--Container glass
SIC 3229--Pressed and blown glass, not elsewhere classified
(N.E.C.)
SIC 3296--Wool fiberglass
The flat glass industry produces window glass, sheet glass, plate glass,
and laminated glass. The glass container industry produces bottles, jars,
and other glass packaging items. Wool fiberglass manufacturers make insula-
tion and glass wool.
The pressed and blown glass segment of the glass industry makes a vari-
ety of products from a very large number of formulations. Among the large
number of products made are art and table glassware; oven glassware; chemical
glassware; glass blanks for electric light bulbs and TV tubes; insulators;
construction glass; colored signal glass; pressed lenses for lighting, bea-
cons, and lanterns; and tubing. Textile glass fibers are also made.
The major raw materials used in all types of glass manufacturing are
glass, sand, soda ash, limestone, and cullet. (Cullet consists of recycled,
crushed glass.) The raw materials melted to make glass can be categorized
as formers, fluxes, and stabilizers. Formers account for the random three-
dimensional atomic structure characteristic of glass. Fluxes are added to
lower the melting points and the working temperatures which must be main-
tained in the furnace. Stabilizers improve the chemical durability of the
glass product by lowering the coefficient of expansion and preventing glass
crystallization. Borates increase the thermal durability of the glass pro-
duct by lowering the coefficient of expansion; lead increases the refractive
index and density; aluminum increases glass strength; feldspar, reportedly,
lowers the mixture melting point and prevents devitrification; sodium accel-
erates the melting process; and arsenic compounds aid in fining (removing
bubbles from the melt). In addition to these compounds, trace amounts of
various metal oxides are added to the batch to change the color of the glass
9.7-107
-------
by either imparting a color or neutralizing the tints caused by batch
contaminants.
9.7.11.2 Industry Statistics. Glass manufacturing facilities are
located throughout the United States and are usually situated in areas that
ensure the availability of raw materials. These plants are found in 34
States, with almost three-quarters of them in the following 10 States:
California, Illinois, Indiana, New Jersey, New York, Ohio, Oklahoma, Penn-
sylvania, Texas, and West Virginia. In early 1978, there were 129 primary
glass-producing companies which together operated 338 individual plants.
There were 32 flat glass plants; 117 container glass plants; 165 pressed and
blown, N.E.C., plants, including 13 textile fiberglass plants; and 24 wool
fiberglass plants.70,71,72,73,74
Recent production rates and dollar values of shipments for each segment
of the industry are summarized in Table 9.7-41.75576>77»78,79,80,81 A sig-
nificant result of these statistics is that over 90 percent of the total
glass produced in 1976 is soda-lime glass, assuming 77 percent of the
pressed and blown glass produced is soda-lime glass, as it was in 1973.82
9.7.11.3 Process Description. Glass is manufactured by a high tempera-
ture conversion of raw materials into a homogeneous melt capable of fabrica-
tion into useful articles. This process can be broken down into three sub-
processes: raw materials handling and mixing, melting, forming and finishing.
Figure 9.7-24 gives a typical flow diagram for the manufacture of soda-lime
glass;83 however, the flow diagram has general application to other commer-
cial glass formulations.
The raw materials are mixed according to the desired glass recipe. For
instance, borax will be added to make a low-expansion borosilicate glass if
oven glassware is to be molded.
Pot type melting furnaces are used if only a few tons of a specialty
glass are to be produced; the continuous tank type furnace is used for larger
production quantities. By far the larger amount of glass is melted in
furnaces, and only these furnaces are considered here in connection with
particulate control. The glass melting process is the largest source of
fine particulates in a glass manufacturing plant.
Continuous tank furnaces have a holding capacity of up to 1400 tons and
a daily output of as much as 300 tons. Figure 9.7-25 is a cut-away view of
9.7-108
-------
Table 9.7-41. 1976 PRODUCTION RATES AND VALUES OF SHIPMENTS75"81
Segment
Production rate
in 1976
Dollar value of
shipments in 1976,
in millions of dollars
Flat glass
Container glass
Pressed and blown
(N.E.C.)
Wool fiberglass
2.56 Tg (2.91 MM Tons)75
11.8 Tg (13.0 MM Tons)77
1.73 Tg (1.95 MM Tons)75
0.896 Tg (0.986 MM Tons)80
64576
3,25178
1,59879
81781
9.7-109
-------
OTHER ADDITIVES
FOR K?0, HgO,
ZnO, BaO, PbO,
ETC. AND THOSE
FOR LINING,
OXIDIZING,
COLORING, AND
DECOLORING
RAW MATERIAL
BATCHING AND
HANDLING
SIDE-PORT
CONTINUOUS TANK,
LOOKING DOWN
THROUGH TOP
SUBMERGED
THROAT IN
BRIDGEWALL
MELTING
AND
FINING
TEMPERATURE = 1500-2000 F
DEPENDING ON ARTICLE
AND PROCESS
FORMING HOT, VISCOUS GLASS
SHAPED BY PRESSING,
BLOWING, DRAWING
OR ROLLING
60-90 MINUTES IN
CONTINUOUS BELT
TUNNEL LEHR: HOT ZONE
900°F
INSPECTION AND
PRODUCT TESTING
FORMING
FINISHING
PACKING, WAREHOUSING,
AND SHIPPING
Figure 9.7-24 Typical flow diagram for the manufacture of soda-lime
glass.
83
9.7-110
-------
GLASS SURFACE IN REFINER
REFINER SIDE WALL
GLASS SURFACE MELTER SIDE WALL THROAT
•MELTER BOTTOM
IN MELTER
NATURAL DRAFT STACK
BACK WALL
COMBUSTION AIR
BLOWER
RIDER ARCHES
MOVEABLE REFRACTORY BAFFLE
Figure 9.7-25 Regenerative side port glass-melting furnace.
9.7-111
-------
a typical side-port, regenerative container glass furnace.84 it is also
representative of some specialty glass furnaces. The "batch" feed is intro-
duced at the left, or "melting end," which is maintained at as high a tem-
perature as the production requires. The glass is heated by direct radia-
tion from the flames and the refractory melter crown. As the mass fuses, it
passes into the "fining zone" and finally through the submerged throat of
the bridgewall into the "working zone."
The combustion gases, on leaving the melting zone, retain, a considerable
amount of heat. This is reclaimed in a regenerator or brick checker chamber.
When the firing cycle is reversed, combustion air is preheated by being
passed through the brickwork. Preheating saves fuel but increases the flame
temperature.
A variation on side-port firing is the end-port fired furnace, in which
a single burner replaces multiple burners at one end of the tank. The com-
bustion gases follow a U-shaped path and enter a checker chamber for heat
reclamation. Reversal of the cycle is then similar to that in si deport
firing. Numerous variations on construction and firing include direct firing,
and regenerative firing.
Coal is not used in glass melting. Since molten glass is conductive,
electric heating is used as a booster to supplement fuel firing whenever
technically and economically practical. Gas and, to a lesser extent, fuel
oil are the preferred fuels.
9.7.11.4 Emission Characteristics. The evolution of particulates and
other pollutants from a glass melting furnace depends, among other things,
on combustion gas volume and melt temperature. Much of the fine particulate
is condensed sodium sulfate formed by vaporization of compounds of sulfur
and compounds of sodium.85 Testing shows that over 75 percent of the par-
ticulate catch has a size less than one micrometer.86
The chemical composition of the particulate emitted from the manufac-
turing of glass depends on the raw materials processed through the furnace.
The particulate emitted in borosilicate glass manufacture consists of boric
acid and alkali borates. In the production of opal glass, 8203, NaF, and
Na2SiF6 appear in the particulate catch. For lead glass production irr a
natural gas-fired furnace, the chemical composition of the particulate is
lead oxide and lead sulfate.87
9.7-112
-------
Other operating parameters affect the levels of pollutants emitted from
the glass furnace, such as: the amount of cullet in the raw batch, the use
of electric boosting, the surface area of the molten glass bed, the produc-
tion (or pull) rate of glass exiting the furnace, and the type of fuel being
burned.
The surface area of molten glass exposed to combustion gases has been
shown to affect particulate emissions. With all other parameters constant,
a larger exposed area generates more particulate than a smaller area.88
For a furnace producing a single type of glass, increasing the pull
rate requires more energy, which, if supplied by the combustion of fossil
fuels, causes an increase in furnace temperature with a concomitant increase
in emissions. This dependence of emission rates on furnace throughput is
incorporated within the compliance regulations of several States. These
compliance regulations indicate that particulate emissions per kilogram of
glass produced decrease as production rate increases. In the limiting case
of zero pull rate, data show that particulates are still emitted from the
molten glass bed.8^ For this case, the emission levels at zero pull rate
were roughly 20 percent of those at the normal pull rate with both measure-
ments being taken at the same temperature.
Gaseous and particulate emissions from uncontrolled glass melting fur-
naces are depicted in Table 9.7-42 for each industry category. Values of
gaseous emissions are taken from source assessment documents.90,91,92,93,94
Particulate emissions are based on the results of: emission tests performed
for the EPA, on the results of emission tests provided by the glass industry
in response to questionnaires, and on the emissions reported in source assess-
ments of the screening of study documents.
The largest mass emissions from glass melting furnaces are nitrogen
oxides. Changing from natural gas firing to fuel oil will increase sulfur
oxide emissions in proportion to the sulfur content of the fuel oil. There
is roughly a 10 percent increase in particulate emissions from fuel oil
fired furnaces compared with natural gas firing.95,96 Particle size distri-
bution was not dependent on type of fuel fired.
The total amount of particulate matter emitted by the entire nationwide
glass manufacturing industry was 23.4 Gg in 1977.97 A significant fraction
9.7-113
-------
Table 9.7-42. EMISSIONS FROM UNCONTROLLED GLASS MELTING FURNACES
FOR EACH INDUSTRY CATEGORY90»91.92,93,94
Industry Particulate, Fluorides,
category g/kg (Ib/ton) g/kg (Ib/ton)
Flat glass 1.5 (2.0)
Container 1.25 (1.5)
glass
Pressed and 1.25 (2.5)
blown:
soda-lime
Pressed and 5.0 (10.0) 10 (20)
blown:
other than
soda-lime *
Wool fiber- 5.0 (10.0) 0.06 (0.12)
glass
9.7-114
-------
is in the respirable range. The median particle diameter from glass melting
and processing operations is generally submicrometer, with 79 to 83 percent
of the particles less than 3.0 micrometers in size.98 The size distribution
of particulate matter emitted is shown for two specific glasses in Figure
9.7-26. These values change with the chemical composition of the glass
(flint glass versus amber glass).
9.7.11.5 Applicable Control Techniques.
9.7.11.5.1 Current regulations. Table 9.7-43 contains Federal stan-
dards of performance, as they apply to particulates from gas-fired glass
melting furnaces for each of the glass manufacturing categories. These
standards are based on analyses of costs and other factors that show the
ability of each category of glass manufacturing furnace to achieve such a
level of control through the use of systems of continuous emission reduction.
An increment 30-percent greater than the promulgated emisssion limits
for natural gas-fired furnaces is allowed for fuel-oil fired glass melting
furnaces, and a proportionate increment is allowed for glass melting fur-
naces simultaneously firing natural gas and fuel oil. Both allowances apply
to glass furnaces melting other than flat glass. The flat glass standard is
based solely on emission tests conducted on a liquid-fired furnace while the
other standards are based on emission tests conducted on both liquid and
gas-fired furnaces.
State particulate regulations for the glass industry are largely based
on process weight.9^ California limits are those for allowable particulate
emissions in the South Coast Air Quality Management District. New Jersey
allows for an increase in the use of cullet and has a special concentration
limit for lead glass.
9.7.11.5.2 Control of particulates from glass melting. Examination of
the three major operations of glass manufacturing—namely, raw material
handling, glass melting, and forming and finishing—shows that essentially
100 percent of the oxides of nitrogen, 98 percent of the particulates, and
essentially all of the oxides of sulfur are generated in the melting of
glass.99 Because emissions are centered in the glass melting operations,
the emission control techniques described in this section deal with the,
reduction of airborne emissions in the furnace exhaust.
-------
i,
O)
•p
OJ
I
o
Q
LU
0.6
0.5
0.4
0.3
0.2
0.1
0.05
0.04
0.03
0.02
FLINT GLASS
AMBER GLASS
1
J__J \ I L
1
J L
i__J L
0,05 0.2 12 5 10 20 30 50 70 80 .. 90 95 98 99
PERCENT BY NUMBER EQUAL TO OR LESS THAN INDICATED SIZE
99.8
Figure 9.7-26 Log-probability distribution of particle sizes present
in glass furnace effluent.69
-------
Table 9.7-43. STANDARDS OF PERFORMANCE FOR
GAS-FIRED GLASS MELTING FURNACES
Standard
Glass category (g of particulate/kg of glass produced)
Container glass 0.1
Pressed and blown glass:
Borosilicate 0.5
Soda-lime or lead 0.1
Other than borosilicate,
soda-lime, or lead 0.25
Wool fiberglass 0.25
Flat glass 0.225
9.7-117
-------
Process Modifications—Process modifications employed in the manufac-
turing of glass in order to lower emissions include reducing the amounts of
materials in the feed that vaporize at furnace temperatures, increasing the
fraction of recycled glass in the furnace feed, installing sensing and con-
trolling equipment on the furnace, modifying the burner design and firing
pattern, and utilizing electric boosting.
Because emission tests are not available to document the lowering of
particulate emissions by using process modifications, the evidence substan-
tiating the efficacy of these methods is not quantitative as is that for
the other control techniques discussed below. Nevertheless, these control
methods and the approach to particulate emission control warrant considera-
tion, whenever it is feasible to employ them.
All Electric Melting—In contrast to conventional fuel-fired furnaces,
the surface of the melter in a cold top electric furnace is maintained at
ambient temperature, and fresh raw batch materials are fed continuously over
the entire surface. As molten glass is withdrawn from the melter, raw batch
drops in the melter, gradually heating and finally reacting in the liquid
phase. This processing minimizes losses from vaporization. The gases dis-
charged through the batch crust consist of carbon dioxide and water vapor.
Since there is no combustion taking place, fuel-derived pollutants are elimi-
nated. The only air emissions are from the decomposition of carbonates,
sulfates, and 'nitrates, with the majority of the exhausts being C02- Finer
control of the glass melting process has meant lower emissions, since elec-
tric melters retain more borates, phosphates, and fluorides than fossil
fuel-burning furnaces.100 In addition, there is no solid disposal problem
as with fabric filters or with electrostatic precipitators, and no water
disposal problem as with scrubber systems.
Actual emission test results are presented in Table 9.7-44. All tests
were performed by EPA Method 5 except the soda-lime melts. These results
demonstrate that particulate emission levels equivalent to or less than
approximately 0.1 g/kg (0.2 Ib/ton) can be maintained in the production of
soda-lime and borosilicate glasses.
9.7-118
-------
Table 9.7-44. ALL ELECTRIC GLASS MELTING FURNACE PARTICULATE
EMISSIONS TESTS"
Emission
test Glass
reference industry Mass emissions
number category Glass type g/kg (Ib/ton)
92 Wool Soda-lime
fiberglass borosilicate 0.05 (0.10)
92 Wool Soda-lime
fiberglass borosilicate 0.07 (0.14)
92 Wool Soda-lime
fiberglass borosilicate 0.09 (0.18)
93 Glass
container Soda-lime 0.12 (0.24)
9.7-119
-------
Fabric Filters—Several glass manufacturing facilities utilize fabric
filter systems to collect particulates in the glass melting furnace exhaust.
In these systems, the furnace exhaust is first cooled and then passed through
/
a fabric filter which retains particulate and allows the gases to vent to
the atmosphere. The physical characteristics of the filtering fabrics and
the agglomerating tendency of submicrometer particles have made the fabric
filter systems viable control techniques for the collection of glass melting
furnace particulates.
Fabric filter systems are claimed to have the advantages of high collec-
tion efficiency (99 percent),101 low pressure drop across the system, and
low energy requirements.102 Collection efficiencies are not affected by the
electrical resistivity of the particles. In addition, bag life is about two
years depending on the bag construction material.103 There are certain dis-
advantages to the application of fabric filters to glass melting furnace
gases; for example, the temperature of gases entering the fabric filter must
be below a maximum value to inhibit attack on the filtering media as well as
above a minimum value to prevent condensation of sulfur trioxides and too
high a moisture content of the gases can form an irremovable plug within a
filter bag.
Table 9.7-45 lists emission test results for glass melting furnaces
using baghouse-controls. EPA Method 5 was used except in the first test
listed, which used the Los Angeles particulate sampling method.
Particulate emissions for the tests listed in Table 9.7-45 range from
0.12 g/kg (0.24 Ib/ton) to 0.55 g/kg (1.1 Ib/ton). The high collection
efficiency claimed for fabric filters is substantiated in the soda-lead
borosilicate glass test. The particulate collection efficiency of the fabric
filter treating the soda-lime furnace exhaust may be lower than the effi-
ciency which is technically feasible because particulate collection was never
maximized in this system. The unit was designed only to meet local opacity
regulations. Since the unit met the regulations after startup, no improve-
ment of particulate collection was attempted.
In conclusion, fabric filters have demonstrated reductions of particu-
late emissions to levels equivalent to less than 0.2 g/kg (0.4 Ib/ton) for
glass formulations in two glass industry categories--"wool fiberglass" and
9.7-120
-------
Table 9.7-45. PARTICULATE EMISSION TEST RESULTS FOR GLASS MELTING FURNACES
EQUIPPED WITH FABRIC FILTERS^
vo
IN3
Glass
industry
category
Pressed and blown:
soda-1 ime
Pressed and blown:
other than soda-lime
Wool fiberglass
Glass
type
Soda- lime
Soda-lead
borosilicate
Borosilicate
Air/
cloth
ratio
0.65 : 1
0.6 : 1
0.85 : 1
Parti cul ate
removal
efficiency,
percent
72
94.8
Fabric filter outlet
parti cul ate emissions,
g/kg (Ib/ton)
0.12
0.17
0.20
(0.24)
(0.34)
(0.40)
Wool fiberglass
Soda-lime
borosilicate
0.5 : 1
0.55
(0.10)
Wool fiberglass
Soda-lime
borosilicate
0.26
(0.52)
-------
"pressed and blown, other than soda-lime." Additionally, based on the assess-
ment of test 36, appropriately sized and optimized fabric filter systems can
be expectd to reduce participate emissions from soda-lime melting furnaces
to levels of 0.1 g/kg (0.2 lb/ton).99
Venturi Scurbber Systems—Although scrubber systems have been built to
control particulate emissions in the glass industry, only a few devices are
in use presently to control container glass emissions. The most common sys-
tem in operation is the venturi scrubber. In a venturi scrubber, particle-
laden gases are accelerated by a high power I.D. fan through a restriction
in the ducting where water is injected into the gas stream. The velocity of
the gas stream provides the dual function of atomizing the scrubbing fluid
while at the same time providing a differential velocity between particles
and the resulting liquid droplets. Since the particulates are mostly water
soluble, the scrubber provides a means of removing these emissions. Addi-
tionally, some gases are absorbed as condensables. The pressure drop to
obtain high velocities in the throat of a scrubber is directly proportional
to the gas velocity squared and the liquid-to-gas ratio; therefore, high
velocities are possible only at substantial pressure drops which result in
high fan energy expenditures. Typical pressure drops are approximately
7.5 kPa (30 inches of water).104
Table 9.7-46 lists emission test results for furnaces using venturi
scrubbing systems. The EPA Method 5 test was used except for test 43, which
used the Los Angeles particulate test method." Pressure drops were 30 to
34-in. water, and sulfur dioxide removal varied from 76-90 percent. Tests
43 and 45 are from the same furnace. These tests demonstrate that venturi
scrubbers can lower the particulate emissions from uncontrolled container
glass melting furnaces to a level equivalent to or less than 0.20 g/kg
(0.4 lb/ton).
Electrostatic Precipitators—As of 1979, electrostatic precipitators
were installed on more than 19 glass furnace exhaust systems throughout the
country, making ESPs the most popular control technique."
The fundamental steps of electrostatic precipitation are particle char-
ging, collection, and removal and disposal of the collected material. Par-
ticulate charging is accomplished by generating charge carriers which are
9.7-122
-------
Table 9.7-46. PARTICULATE EMISSION TEST RESULTS FOR GLASS MELTING
FURNACES EQUIPPED WITH VENTURI-SCRUBBERS99
Emission
test
reference
number
98a
99
100 a
101
Glass
industry
category
Container
Container
Container
Container
P articulate .
removal
Glass efficiency,
type percent
Soda-lime 82.5
Soda-lime
Soda-lime 79.6
Soda- lime
Venturi-scrubber outlet
Mass emissions,
g/kg (Ib/ton)
0.37
0.12
0.14
0.20
(0.74)
(0.24)
(0.28)
(0.40)
aOil fired.
9.7-123
-------
driven to the participates by an electric field. Collection occurs as the
charged participates migrate and adhere to electrodes. Applying a mechan-
ical force to the collection electrodes dislodges the collected material,.
which then falls into hoppers. Effective transfer of dust to the hopper
depends on the formation of chunks or agglomerations of dust, which fall
with a minimum of reentrainment.
Electrostatic precipitators can be designed and guaranteed to collect
99 percent of the particulate in the glass melting furnace exhaust.105
Table 9.7-47 lists emission test results for electrostatic precipitator-
controlled glass melting furnace exhaust. In some plant configurations, one
or more electrostatic precipitators collect particulates from several fur-
naces. In these cases, the table entries list the total pull rates from all
furnaces whose exhausts are controlled during testing and the sum of the
particulate emissions of all electrostatic precipitators in the plant.
EPA Method 5 testing was used for all tests with the following excep-
tions: the Los Angeles test method was used for tests 47 and 48; tests 62
through 64 were by plants using unspecified test methods. Test 53 was on a
furnace firing number 5 fuel oil.
The particulate emissions for soda-lime formulations in the "container"
glass category and for the lead, fluoride-opal, and potash-soda-lead formu-
lations in the."pressed and blown, other than soda-lime" category (see Table
9.7-47) range from 0.3 g/kg to 0.27 g/kg (0.6 Ib/ton to 0.54 Ib/ton). For
borosilicate glass formulations manufactured in the "pressed and blown,
other than soda-lime" category and in the "wool fiberglass" category, the
particulate emission test results range from 0.09 g/kg to 0.57 g/kg (0.17 to
1.14 Ib/ton). Two factors that could explain the higher values for borosili-
cate emissions despite the larger special collection area are: (1) the higher
electrical resistivity of borosilicate dusts, and (2) the tendency for the
collected dusts to bridge in the precipitator.106 Since the resistivity
of the lead dusts is nearly equal to the resistivity of borosilicate dusts
and since the lead particulate is collectable, the second factor may control
the collection of borosilicate glass melting furnace emissions.
In conclusion, electrostatic precipitators have demonstrated particu-
late emission control levels of 0.06 g/kg (0.12 Ib/ton) for soda-lime,
9.7-124
-------
Table 9.7-47. PARTICIPATE EMISSION TEST RESULTS FOR GLASS MELTING
FURNACES EQUIPPED WITH ELECTROSTATIC PRECIPITATORS99
Emission
test
reference Glass Industry
number category
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
Container
Container
Container
Pressed and blown:
other than soda-lime
Pressed and blown:
other than soda-lime
Pressed and blown:
other than soda-lime
Pressed and blown:
other than soda-lime
Pressed and blown:
other than soda-lime
Pressed and blown:
. other than soda-lime
Pressed and blown:
other than soda-lime
Pressed and blown:
other than soda-lime
-' Pressed and blown:
other than soda-lime
Pressed and blown;
other than soda-Hme
Pressed and blown:
other than soda-Hme
Pressed and blown:
other than soda-Hme
Wool fiberglass
Wool fiberglass
Wool fiberglass
Glass type
Soda- 11 me
Soda-lime
Soda-Hme
BoroslHcate
BoroslHcate
BoroslHcate
BoroslHcate
Fluoride/opal
Lead
Lead
Lead
Lead
Lead
Lead
Potash-soda-
lead
BoroslHcate
BoroslHcate
BoroslHcate
Percent
of
design
Specific SCFH Partlculate
collection area during removal
™2/fta3/s
a
138
237
225
138
290
179
379
233
337
183
195
237
220
222
216
test efficiency,
(Ft2/SCFM) X
a 91
(0.65) 83
(1.12) 116
(1.06) 100
(0.65) 89
(1.37) 43
(0.85')
(1.19) 64
(1.09) 75 93
(1.59) 117
(0.56) 91
(0.92) 80
97
(1.12) 122
(1.04)
(1.05) :
(1.02)
Preclpltator outlet
mass emissions.
g/kg
0.06
0.07
0.06
0.45
0.57
0.48
0.48
0.17
0.06
0.08
0.08
0.07
0.18
0.27
0.03
0.36
0.09
0.09
(Ib/ton)
(0.12)
(0.14)
* (0.12)
(1.14)
(0.96)
(0.96)
(0.34)
(0.12)
(0.16)
(0.16)
(0.14)
(0.36)
(0.54)
(0.06)
(0.72)
(0.19)
(0.17)
aCla1med proprietary.
9.7-125
-------
lead, and potash-soda-lead glass formulations, and control levels of about
0.2 g/kg (0.4 Ib/ton) for borosilicate glass formulatitms.
Summary of Particulate Control Techniques—Table 9.7-48 presents the
levels of particulate emissions from the control systems discussed in this
section for each industrial glass category except flat glass manufacturing.
The emission levels listed in Table 9.7-48 represent particulate control
technically achievable as substantiated by test reports, and, therefore,
these levels reflect the lowest values from the previous tables.
All-electric melting of glass has been shown to greatly reduce the par-
ticulate emissions from glass melting furnaces without the addition of add-on
control equipment. This technique is not applicable to the entire glass
industry because, at present, only formulations of appropriate resistivity
and furnaces of relatively moderate production rates can utilize all-electric
melting.
Fabric filters have been installed on existing furnaces classified in
both the "pressed and blown" and "wool fiberglass" categories.
Venturi scrubbers have been installed on existing container glass fur-
naces. Scrubbers have not been used to control borosilicate emissions
because the chemicals discharged in the liquid effluent present more of a
disposal problem than those from soda-lime glasses.107
Electrostatic precipitators have been installed widely in the glass
manufacturing industry. Significant amounts of emission tests substantiate
the values listed in Table 9.7-48.
Switching fuels from natural gas to fuel oil adds particulate formed in
combustion to the particulate formed in producing glass. The add-on control
devices discussed above would be expected to be equally efficient in con-
trolling particulate emissions with either fuel. As demonstrated in Tables
9.7-43 and 9.7-44, venturi scrubbers and electrostatic precipitators have
previously been used on oil-fired glass melting furnaces.
Although, as of June 1978, no add-on control system continuously con-
trols particulate emissions from a flat glass manufacturing furnace, there
is no technical evidence to preclude their use. The flat glass furnaces
produce more soda-lime glass than container furnaces, but the physical and
.9.7-126
-------
Table 9.7-48. REPRESENTATIVE PARTICIPATE EMISSIONS FROM GLASS MELTING FURNACES^
vo
ro
Glass industry
category
Container
Pressed and blown:
soda-lime
Pressed and blown:
other than soda-lime
Pressed and blown:
other than soda-lime
Pressed and blown:
other than soda-lime
Wool fiberglass
All-electric
Glass type Melting
g/kg (Ib/ton)
Soda-lime 0.12 (0.24)
Soda-lime
Lead
Fluoride/
opal
Borosilicate
Borosilicate 0.07 (0.14)
Fabric filter
g/kg (Ib/ton)
0.12 (0.24)
0.17 (0.34)
0.25 (0.50)
Electrostatic
Venturi scrubber precipitator
g/kg (Ib/ton) g/kg
0.20 (0.40) 0.06
0.08
0.17
0.50
0.10
(Ib/ton)
(0.12)
(0.16)
(0.34)
(1.0)
(0.20)
-------
chemical nature of the'resulting particulates is identical. Because of the
greater glass production in flat glass furnaces and concomitant larger
exhaust volume than in container glass furnaces, an electrostatic precipi-
tator would probably best control the particulate emissions. One flat
glass manufacturer is presently installing an electrostatic precipitator.
9.7.12 Fiberglass Manufacturing
Fiberglass is manufactured by melting various raw materials to form
molten glass, drawing the molten glass into fibers, and coating these fibers
with various organic materials. There are many different glass compositions
used to make commercial fiberglass. These compositions are selected to
produce finished products with material properties that are suitable for
given applications. Types of glass, along with their composition, proper-
ties, and uses, are shown in Table 9.7-49.108
Textile fibers and wool fibers are the two basic types of fiberglass
products; they are made by different forming processes. Textile fibers,
often referred to as yarns, are formed in continuous fibers on spools. Wool
fibers are collected as mats and formed in various lengths.
In 1972, there were a total of 33 fiberglass manufacturing plants in
the U.S., operated by 10 manufacturers. These plants are located primarily
in the north-central, mid-Atlantic, and southeast regions of the U.S., with
three plants in each of Texas and California.
In 1970, 210 Mg of textile fibers and 550 Mg of wool fibers were pro-
duced in the United States. The annual growth rate during the 10 years
before 1970 was 12.2 percent for textile fibers and 4.2 percent for wool
fibers.109
The major emission from the fiberglass manufacturing process is particu-
late matter from the glass-melting furnace, the forming line, the curing
oven, and the product cooling line.HO For textile fiber manufacturing, the
melting furnace produces 89 percent of total uncontrolled particulate emis-
sions from this process (an average of 85 percent for regenerative furnaces
and 91 percent for recuperative furnaces). The major source of emissions
from wool/fiber production is the forming process, which accounts for 66
percent of the total emissions; glass melting accounts for 29 percent of
9.7-128
-------
TABLE 9.7-49. GLASS TYPE, COMPOSITION, PROPERTIES AND USAGES108
Glass
name
E-Glass
T-Glass
C-Glass
SF-Glass
S-Glass
Glass
type
Low alkali,
Lime-alumina
borosilicate
Soda-lime glass
Soda-lime
borosilicate
o 59.5% SiOg
o 14.5% NaaO
o 8% Ti02
o 7% 8203
o 11% other
o 65% Si02
o 25% A1203
o 19% MgO
Properties
o Good durability
o Moderate cost
o Low cost
o Thermal and
acoustic
properties
o Resistive to
aci ds
o Excellent
weathering
properties
o High strength
o High Young's
modulus
Uses
o Textiles
o Electical
insulation
o Plastic
rei nf orce-
ment
o Mats
o Coarse fiber
mats for
air filters
o Thermal
insulation
o Acoustic
insulation
o Mats for
storage
battery
retainers
o Acid chemical
filters
o Filter cloths
o Anode bags
p Low density
thermal and
acoustic
insulators
o Paper
additives
o High efficiency
all glass
filter papers
o Aerospace
applications
9.7-129
-------
the total. Electric induction furnaces can be used in both textile and wool
production; such furnaces reduce particulate emissions to negligible
quantities.110
9.7.12.1 Process Description. Fiberglass production plants can be
divided into five process stages, with each stage having different emission
characteristics. These stages are outlined in Figures 9.7-27 and 9.7-28 for
textile and wool fibers, respectively. The general process consists of:
o Batch mixing and conveying
o Glass melting
o Forming operations
o Curing
o Cooling and fabrication
Batch mixing and conveying systems are usually commercial equipment of
standard design. This equipment is customarily contained in an enclosed
structure called the "batch plant," which is separate from the melting
furnace.
The average particle size of the raw materials used In the production
of glass fiber is about 300 micrometers. A small percentage of these
materials are less than 50 micrometers in size and can cause emission prob-
lems during conveying, mixing, and storage operations.108
The glass-melting reaction takes place in a large rectangular gas- or
oil-fired reverberatory furnace. These melting furnaces are equipped with
either regenerative or recuperative heat-recovery systems. Electric induc-
tion has also been used for wool fiber production because it is competitive
with regenerative furnaces in areas with low electric power costs.
Particulate emissions from melting furnaces result from the complete physi-
cal and chemical reactions inside the furnace. Factors affecting particle
size and emission rates include the furnace design (fuel oil, natural gas,
or electric), raw material size and composition, and type and volume of the
furnace heat-recovery system.110 The generation of particulate matter is
also affected by the bubbling of carbon dioxide which propels particles
from the melting batch.& Reverberatory furnaces generally produce more
particulate emissions than melting furnaces; emissions from electric induc-
tion furnaces are negligible.10^
9.7-130
-------
RAW MATERIALS
—
BINDER
ADDITION
-
*
DRYING OR
CURING
RAW MATERIAL
STORAGE
FORMING BY
DRAWING,
STEAM JETS,
OR AIR JETS
COLLECT AN
OR
CUT AND FAB
BATCHING
—»•
GLASS MELTING
AND
REFINING
(FURNACE)
*
MARBLE
REMELT
FURNACE
-
MARBLE
FORMING
D WIND
*ICATE
PRODUCTS:
CONTINUOUS TEXTILES,
STAPLE TEXTILES,
MAT PRODUCTS, ETC.
Figure 9.7-27
Typical flow diagram of textile-type, glass fiber
production process.40
RAW MATERIAL
STORAGE
BATCHING
COMPRESSION
(OPTIONAL DEPENDING
UPON PRODUCT)
*
CURING
(OPTIONAL DEPENDING
UPON PRODUCT)
COOL
ADDITION OF
BINDERS, LUBRICANTS
AND/OR ADHESIVES
PACK OR
FABRICATE
FORMING BY AIR
BLOWING, STEAM
BLOWING, AND
CENTRIFUGE
PRODUCTS: LOOSE WOOL
INSULATION, BONDED
WOOL INSULATION, WALL
AND CEILING PANELS,
INSULATION BOARD, ETC.
Figure 9.7-28 Typical flow diagram of wool-type glass fiber production
process.^ •
9.7-131
-------
Forming operations differ for textile and wool fiber production. Wool
fiber mats are produced in large volumes by feeding molten glass directly
from a glas.s tank through platinum bushings, attenuating the glass streams
into fibers, collecting the fibers as mats, and spraying these mats with
lubricants and binders. Flame blowing is used to produce fibers less than
250 micrometers in diameter.
For textile fibers, molten glass from the forehearth passes through
platinum bushings and is attenuated by an air blower. Staple fibers are
collected in a revolving perforated drum and wound as a "sliver" onto a
spool rotating at higher speeds than the collecting drum. For very fine
fibers (less than 6.5 micrometers in diameter), glass makers form the
molt en1-refined products into marbles, which are inspected for quality and
then remelted in an electric furnace.
Emissions from the forming line are caused by evaporation and subse-
quent condensation of the resin binder that is sprayed on the hot glass
fibers as they emerge from the forming nozzles. This fine condensed par-
ticulate material which escapes from the operation is largely submicrometer
in size. The particulate emissions from the forming line are primarily
affected by the composition and quantity of the binder and by the spraying
techniques used to coat the fibers; a fine spray and volatile binders in-
crease emissions. ^^
Curing and cooling operations also differ for wool and textile fiber
production. Wool fiber blankets enter a curing oven where the thermo-
setting binder is cured at temperatures ranging from 200°C to 260°C. Cer-
tain products must be restricted to an established thickness and density,
and this is accomplished in the curing oven by pressure rolls and plates.
For textile fibers, tubes from the winding operation are sent to either a
curing oven where the thermosetting binder is cured, or to a yarn condi-
tioning room where the binder sets from 1 to 16 hours under conditions of
high humidity. The curing vaporizes additional amounts of resin binder,
which condenses upon cooling and causes a visible emission. The amount of
emissions depends largely on oven temperature and binder composition.
After curing and cooling, wool fiberglass is slit and cut to specified
dimensions or rolled into packages. Textile fiber slivers are twisted into
9.7-132
-------
coarse or fine yarns and then chopped and packaged for shipment. Participate
emissions from this process are negligible.
9.7.12.2 Control Techniques* Participate control devices for batch
mixing and conveying of raw materials and for glass melting in the fiberglass
manufacturing industry are essentially the same as those used in the glass
manufacturing industry. The reduction of particulate emissions during these
two production stages can be accomplished by several methods, including control
and treatment of raw material, efficient combustion of fuel, proper design of
glass furnaces, and installation of control equipment. Control techniques
and options for these two particulate emission sources are outlined in
Section 9.7.11.
The control of the fine condensed fumes from the forming lines can be
accomplished by process modifications or control devices. Process modifica-
tions include a change in resin composition, a change in the method of
application, and a reduction of forming zone temperatures.
Many exhausts from forming operations produce submicrometer organic,
sticky particles that present serious problems with fabric filtration. High
energy scrubbers may require secondary water treatment systems. Electrostatic
precipitators experience fouling from the sticky particulate matter. Incin-
eration becomes costly for large volume gas flows.
The gaseous and particulate emissions from the curing process can be
controlled with gas-fired afterburners.
9.7.13 Mineral Wool
Mineral wool, a fibrous material made from natural rock and metallurgi-
cal slag, is used for thermal and acoustical insulation and as a gas filtra-
tion medium. A typical mineral wool plant produces 2 Mg/hr or 17.5 Gg/yr of
product.m The wool manufacturing process has three primary sources of
particulate emissions: the cupola furnace stack, the blow chamber, and the
curing ovens. Some particulate matter is also emitted during the cooling
stage; however, these emissions are not well characterized. Particulates
emitted during mineral wool production include mineral wool particles, fumes,
oil vapors, and binding agents.
9.7.13.1 Process Description. Mineral wool is usually made in a cupola
furnace that is charged with slag from steel blast furnaces and basalt, lime-
stone, or silica rock. Slag from copper or lead furnaces may also be used.
9.7-133
-------
The mixture of rock and slag is heated to a molten state (1650°C) using coke
as fuel. A mechanical spinner is used in most current operations.1Q8 The
molten minerals are fiberized on a spinning rotor using a high velocity
stream of air to assist in fiber attenuation. An oil or binding agent is
applied to the fiber before it is collected on a wire mesh conveyor in an
area known as the blow chamber. Phenolic resins may be applied to some wool
products as binding agents. These binding agents are cured in ovens after
application, and the mineral wool is then cooled prior to shipment or storage
(see Figure 9.7-29).
9.7.13.2 Particulate Control Techniques. Emissions from the cupola or
furnace stack consist of submicrometer sized fume particles. These emissions
can be controlled with dacron or orlon fabric filters having a maximum fil-
tering velocity of 0.8 m/min.6 Most of these particles can be inhaled—about
97 percent (by number) are less than 15 micrometers in diameter and 92 per-
cent are of submicrometer size.59 Glass fiber fabrics are not recommended
because of the fluorides present in the cupola exhaust gases.
Fumes, oil vapors, binding agent, and wool fibers are emitted from the
blow chamber. When the blow chamber temperature is maintained below about
80°C, wool fiber emissions are the major component of any emissions. At
these lower temperatures, oily mists are not formed. The most common con-
trol devices on blow chambers are low energy scrubbers, primarily baffled
spray chambers.m One reference reports the use of a wet scrubber/ESP
combination on blow chamber exhausts.32 Wire mesh lint cages are sometimes
used to collect larger wool fibers. Fabric filters are not advisable
because the binding agents may clog the filters.
Blowing operations may generate large amounts of fibrous particles or
"flywool." At the point of fiber formation, an oil is applied to suppress
dust and cause the fibers to agglomerate. The contact of oil with hot sur-
faces during spinning may generate decomposition products and/or aerosols of
oil.
Emissions from curing ovens, usually binding agent and oil particles,
may create an opacity problem. The particles are not emitted in large
quantities and since they are combustible, the most common form of control
is direct flame incineration.m The use of electrostatic precipitators and
catalytic incinerators has also been reported.
9.7-134
-------
•-4
I
GO
cn
PACKING
AND
STORAGE
Figure 9.7-29 Flow diagram of mineral wool processing.32 NOTE: This is
a batt line; for loose wool products, the oven and batt
machine are eliminated.
-------
Asphalt fumes generated during binder application can become a problem
if the temperature of this operation exceeds approximately 200°C. Careful
control of the asphalt temperature is usually sufficient; however, the use
of a two-stage electrostatic precipitator has been reported to reduce these
emissions. A precleaning device must precede the ESP to prevent plate
fouling.59
Table 9.7-50 summarizes the emission sources and controls used in the
mineral wool industry. It should be noted that the controlled emission
data shown in the table are results from a limited number of emission tests
on different plants using undefined test methods. Therefore, the data may
not reflect the relative performance of the various control devices.
Modeling parameters for a typical mineral wool cupola are listed
Stack height: 15.25 m
Stack diameter: 0.91 m
Stack gas temperature: 150°C
Stack flow rate: 4.55 m^s
Modeling parameters for other mineral wool processes are not available.
9.7-136
-------
Table 9.7-50. MINERAL WOOL INDUSTRY EMISSION SCOURGES AND SELECTED CONTROLS111
GO
--J
Source
Cupola
Blow
chamber
Curing
oven
AP-42
Uncontrolled
Type of particulate
particulate emission factor,
Combustion products 11
and mineral fines
Wool fiber, 8.5
oil & binding agent
aerosols
Oil and 2
binding agent
aerosols
Control Controlled
kg/Mg device mg/Nm^
Cyclone
Wet scrubber
Failure fitter
Wet scrubber
Wet scrubber and
ESP
Direct-flame
afterburner
Catalytic
330
451
46.7
49.4
25.2
73.3
163
emissions,
kg/Mg
1.15
1.1
0.21
0.75
0.42
0.71
0.95
-------
REFERENCES FOR SECTION 9.7
1. Draft - Non-metallic Mineral Processing Plants, Background Information
for Proposed Emission Standards. U.S. Environmental Protection Agency,
Office of Air Quality Planning and Standards. Research Triangle Park,
NC. March 1980.
2. Jutze, G., et al. Draft - Evaluation of Fugitive Dust Emissions from
Mining. U.S. Environmental Protection Agency. EPA Contract No.
68-02-1321. 1979.
3. Augenstein, D. M. Air Pollution Control Techniques for Phosphate
Rock Processing Industry. U.S. Environmental Protection Agency.
Publication No. EPA-450/3-78-030. June 1978.
4. OAQPS Data File of Nationwide Emissions. U.S. Environmental
Protection Agency, Office of Air Quality Planning and Standards.
Research Triangle Park, NC. February 1979.
5. Rusek, S. J., et al. Source Assessment: Open Mining of Coal - State
of the Art. U.S. Environmental Protection Agency. Publication No.
EPA-450/3-78-030. June 1978.
6. Assessment of Environmental Impact of the Mineral Mining Industry.
U.S. Environmental Protection Agency. EPA Contract No. 68-03-2479.
July 1979.
7. Hankin, M. Is Dust the Stone Industry's Next Major Problem? Rock
Products. Chicago, IL. April 1967.
8. Sittig, M- Particulates and Fine Dust Removal. Noyes Data Corporation.
Park Ridge, NJ. 1977.
9. Draft Development Document for Effluent Limitations Guidelines and
Standards of Performance - Mineral Mining and Processing Industry -
Volume I. U.S. Environmental Protection Agency. EPA Contract No.
68-01-2633. January 1975.
10. Metallic Minerals Processing Plants: New Source Performance Standards,
Chapter 3. U.S. Environmental Protection Agency. May 18, 1979.
11. Mineral Commodity Summaries. U.S. Bureau of Mines. Washington, DC.
1980
12. Draft - Screening Study for New Source Performance Standards for Metal-
lic Processing Plants. U.S. Environmental Protection Agency. EPA
Contract No. 68-02-2607. October 1978.
13. Costel, D. M. Environmental Regulation of the Metal Industry. Journal
of Metals. Warrendale, PA. ^0_:1. January 1978.
14. Duncan, L. J. Analysis of Final State Implementation Plans - Rules
and Regulations. Mitre Corporation. McLean, VA. EPA Contract No.
68-02-0248. July 1972.
9.7-138
-------
15. Emission Test Report. Homestake Mining Company. Lead, SD. U.S.
Environmental Protection Agency. Research Triangle Park, NC. EMB
Report Number 80-MET-7. May 1980.
16. Emission Test Report. Anaconda Copper Company. Butte, MT. U.S.
Environmental Protection Agency. Research Triangle Park, NC. EMB
Report Number 79-MET-3. January 1980
17. Emission Test Report. Exxon Mineral Company. Converse County, WY.
U.S. Environmental Protection Agency. Research Triangle Park, NC.
EMB Report Number 79-MET-l. " December 1977.
18. Emission Test Report. Cyprus Bagdad Copper Mine. Bagdad, AZ. U.S.
Environmental Protection Agency. Research Triangle Park, NC. EMB
Report Number 79-MET-4. January 1980.
19. Emission Test Report. Climax Molybdenum Company. SiIverthorne, CO.
U.S. Environmental Protection Agency. Research Triangle Park, NC.
EMB Report Number 79-MET-2. December 1979.
20. Environmental Assessment of Primary Nonferrous Metals Industry Except
Copper, Lead and Zinc, and Related By-products Metals. Battelle
Columbus Laboratory. Columbus, OH. EPA Contract No. P68-02-1323.
February 1977.
21. MacDonald, B. I. Outlook for Copper Industry - Not Bright. Chemical
Engineering Progress. New York, NY. .74-:9. September 1978.
22. Kosakowski, M. W. Development Document for Interim Final Effluent
Limitations Guidelines and Standards of Performance - Mineral Mining
and Processing Industry. U.S. Environmental Protection Agency.
Publication No. EPA 440/l-76-059a. June 1976.
23. Reding, J. T. Industrial Process Profiles for Environment Use, Chapter
19: The Clay Industry. U.S. Environmental Protection Agency. Publica-
tion No. EPA-600/2-77-023s. February 1977.
24. A Screening Study to Develop Background Information to Determine the
Significance of Brick and Tile Manufacturing. Research Triangle
Institute. Research Triangle Park, NC. EPA Contract No. 68-02-0607.
December 1972.
25. Hardison, L. C., and C. A. Greathouse. Air Pollution Control Technology
and Costs in Nine Selected Areas. U.S. Environmental Protection Agency.
Publication No. APTD-1555. September 1972.
26. Emission Test Report. Building Brick and Structural Clay Industry - Lee
Brick and Tile Co. Stanford, NC. U.S. Environmental Protection Agency.
Research Triangle Park, NC. EMB Report Number 80-BRK-l. April 1980.
27. Trip Report on General Shale Products Corp. - Brick Plants. U.S. Environ-
mental Protection Agency, Office of Air Quality Planning and Standards,
Standards Development Section. File AIP 8-1-14-2. July 19, 1979.
9.7-139
-------
28. Ceilcote Ionizing Wet Scrubber Evaluation. U.S. Environmental
Protection Agency. Publication No. EPA-600/7-79-246. November 1979.
29. Vandegrift, A. E., et al. Particulate Pollutant Systems Study, Volume
III: Handbook of Emission Properties. Midwest Research Institute.
May 1971.
30. Muehlberg, P. E., et al. Industrial Process Profiles for Environmental
Use, Chapter 17: The Gypsum and Wallboard Industry. U.S. Environmental
Protection Agency, Industrial Environmental Research Laboratory.
Research Triangle Park, NC. Publication No. EPA-600/2-77-023q.
February 1977.
31. Hopper, T. G., and W. A. Marrone. Impact of New Source Performance
Standards on 1985 National Emissions from Stationary Sources. U.S.
Environmental Protection Agency. Publication No. EPA-450/3-76-017.
April 1977.
32. Formica, P. N. Controlled and Uncontrolled Emission Rates and Appli-
cable Limitations for Eighty Processes. U.S. Environmental Protection
Agency. Publication No. EPA-450/3-77-016. September 1976.
33. Gold Bond Building Products, Division of National Gypsum Co. Charlotte,
NC. Application for Permit to North Carolina Division of Environmental
Management, Air Quality Section. February 22, 1978
34. United States Gypsum Company. Chicago, IL. Construction Permit Appli-
cation to Texas Air Control Board. June 19, 1979.
35. National Emissions Data System. U.S. Environmental Protection Agency.
36. Nevada Testing Laboratories, Ltd. Las Vegas, NV. Report of Determina-
tions: Stack Emission Investigation Tests No. 3, No. 4, and No. 5, Blue
Diamond Plant. NTF Lab No. E-2. September 13, 1973.
37. Flintkote Company. Stamford, CT. Permit Application for Construction at
the Sweetwater Texas plant. Submitted to the Texas Air Control Board.
November 5, 1979.
38. New Jersey Department of Environmental Protection. Report of Emission
Tests. National Gypsum Co. Burlington, NJ. September 6, 1973.
39. Isokinetic Sampling: Empire Gypsum Mill. United States Gypsum Company.
Chicago, IL. Report No. 23074, File No. 1-0015. January 28, 1976.
40. Doumas, A. C., et al. Industrial Process Profiles for Environmental
Use, Chapter 18: The Lime Industry. U.S. Environmental Protection
Agency. Publication No. EPA-600/2-77-023r. February 1977.
41. Standards Support and Environmental Impact Statement, Volume I: Proposed
Standards of Performance for Lime Manufacturing Plants. U.S. Environ-
mental Protection Agency. Publication No. EPA-450/2-77-007a. April
1977.
9.7-140
-------
42. Schwartzkopf, F. Lime Burning Technology - A Manual for Lime Plant
Operators. Kennedy Van Saun. Danville, PA. 1974.
43. Bunyard, F. L. EPA Trip Report of visit with F. Schwartzkopf of Kennedy
Van Saun. U.S. Environmental Protection Agency. Research Triangle Park,
NC. September 29, 1975.
44. Schwartzkopf, F. A Comparison of Modern Lime Calcining Systems. Rock
Products. Chicago, IL. July 1970.
45. Minnick, L. J. Control of Particulate Emissions from Lime Plants - A
Survey. Air Pollution Control Association Journal. Pittsburgh, PA.
April 1971.
46. Reding, J. T., et al. Industrial Process Profiles for Environmental
Use, Chapter 21: The Cement Industry. U.S. Environmental Protection
Agency. Publication No. EPA 600/2-77-023u. February 1977.
47. Mineral Commodity Summaries - 1979. U.S. Bureau of Mines. Washington,
DC. 1979.
48. Gagan, E. W. Air Pollution Emissions and Control Technology. Cement
Industry Report. Environmental Protection Service, Department of the
Environment. Ontario, Canada. Publication No. 3-AP-74-3. 1974.
49. Fine Particle Emissions Information System: Annual Report (1978). U.S.
Environmental Protection Agency. Publication No. EPA-600/7-79-126.
May 1979.
50. Kreichelt, T. E., D. A. Kemnitz, and S. T. Cuffe. Atmospheric Emis-
sions from the Manufacture of Portland Cement. U.S. Public Health
Service. Washington, DC. Publication No. 999-AP-17. 1967.
51. Davis, T. A. and D. B. Hooks. Disposal and Utilization of Waste Kiln
Dust from Cement Industry. U.S. Environmental Protection Agency.
Publication No. EPA-670/2-75-043. 1975.
52. Kulujian, N. J. Inspection Manual for Enforcement of NSPS, Portland
Cement Plant. U.S. Environmental Protection Agency. Publication No.
EPA 340/1-75-001. September 1975.
53. Barrett, K. W. A Review of Standards of Performance for New Stationary
Sources, Portland Cement Industry. U.S. Environmental Protection
Agency. Research Triangle Park, NC. Publication No. EPA 450/3-79-012.
March 1979.
54. Manufacturing Process Options, Volume X: Cement Industry Report. U.S.
Environmental Protection Agency. EPA 600/7-76-034J. December 1976.
55. Capital and Operating Costs of Selected Air Pollution Control Systems.
U.S. Environmental Protection Agency. Publication No. EPA 450/3-76-014.
May 1976.
9.7-141
-------
56. Multimedia Assessment and Environmental Research Needs of the Cement
Industry. A. T. Kearney, Inc. U.S. Environmental Protection Agency,
Industrial Environmental Research Laboratory. Cincinnati, OH. EPA
Contract 68-03-2586. January 1979.
57. Ford, R. Practicable Methods of Controlling Particulate Emissions in
the Cement Industry. Proceedings of the Air Pollution Control Associa-
tion Symposiums, Quebec Section. Air Polution Control Association.
Pittsburgh, PA. 1975.
58. Technical Guidance for Control of Industrial Process Fugitive Particulate
Emissions. U.S. Environmental Protection Agency, Office of Air Quality
Planning and Standards. Research Triangle Park, NC. Publication No.
EPA 450/3-77-010. March 1977.
59. Air Pollution Engineering Manual, Second Edition. U.S. Environmental
Protection Agency. Publication No. AP 40. May 1973.
60. Brooks, K. J., et al. A Review of Standards of Performance for New
Stationary Sources - Asphalt Concrete Plants. Mitre Corporation.
McLean, VA. Publication No. MTR 7826. January 1979.
61. Atmospheric Modeling Data from National Emission Data Systems (NEDS).
U.S. Environmental Protection Agency, Office of Air Quality Planning
and Standards. Research Triangle Park, NC. May 1979.
62. Fine Particulate Emissions Information System: Annual Report (1978).
U.S. Environmental Protection Agency. Publication No. EPA-600/7-79-126.
May 1979.
63. Khan, Z. S., and T. W. Hughes. Source Assessment: Asphalt Hot Mix.
U.S. Environmental Protection Agency. Publication No. EPA-600/2-77-107n.
December 1977.
64. Background Information for Proposed New Source Performance Standards:
Asphalt Concrete Plants, Volume 2. U.S. Environmental Protection Agency.
Publication No. APTD 1532b. June 1973.
65. The Operation of Exhaust Systems in the Hot Mix Plant - Efficiency and
Emission Control. National Asphalt Pavement Association. Riverdale, MD.
Information Series 52. 1975.
66. Background Information for Proposed Standards, Asphalt Roofing Manu-
facturing Industry. U.S. Environmental Protection Agency. Publication
No. EPA-450/3-80-021a. June 1980.
67. Laster, L. L. Atmospheric Emissions from the Asphalt Industry. U.S.
Environmental Protection Agency. Publication No. EPA-650/2-73-046.
December 1973.
68. Gorman, P. G. Control Technology for Asphalt Roofing Industry. U.S.
Environmental Protection Agency. Publication No. EPA-600/2-76-120.
April 1976.
9.7-142
-------
69. Spinosa, E. D., and R. A. Newton. Chemical Analysis of Particle Size
Fractions of the Particle Catch from Glass Melting Furnaces. U.S.
Environmental Protection Agency, Industrial and Environmental Research
Laboratory. Cincinnati, OH. EPA Contract No. 68-03-2552. March
1979.
70. Schorr, J. R., D. T. Hooie, M. C. Brockway, P. R. Sticksel, and D. E.
Niesz. Source Assessment: Pressed and Blown Glass Manufacturing
Plants. U.S. Environmental Protection Agency, Industrial and Environ-
mental Research Laboratory. Research Triangle Park, NC. Contract No.
68-02-1323. NTIS 600/2-77-005, Task 37. January 1977. Appendix A.
71. Reznik, R. B. Source Assessment: Flat Glass Manufacturing Plants.
U.S. Environmental Protection Agency, Industrial and Environmental
Research Laboratory. Research Triangle Park, NC. Contract No.
68-02-1874. NTIS 600/2-76-032b. March 1976. Appendix B.
72. Schorr, J. R., D. T. Hooie, P. R. Sticksel, and C. Brockway. Source
Assessment: Glass Container Manufacturing Plants. U.S. Environmental
Protection Agency, Industrial and Environmental Research Laboratory.
Research Triangle Park, NC. Contract No. 68-02-1323. NTIS 600/2-76-269,
Task 37. October 1976.
73. The Glass Industry Directory Issue 1976-1977. The Glass Industry. New
York, NY. £7(10). 1976-1977.
74. Final Report of Screening Study to Determine Need for Standards of Per-
formance for New Sources in the Fiberglass Manufacturing Industry. U.S.
Environmental Protection Agency. Industrial Studies Branch. Research
Triangle Park, NC. Contract No. 68-02-1332, Task 23. December 1976.
Tables 4 and 5.
75. U.S. Department of Energy. Voluntary Industrial Energy Conservation,
Progress Report No. 5. July 1977. Page 81.
76. U.S. Bureau of Census. Current Industrial Reports, Flat Glass Fourth
Quarter 1976. MQ-32A(76)-4. February 1977.
77. Glass Packaging Institute 1977 Annual Report. Glass Packaging
Institute. Washington, DC.
78. U.S. Bureau of Census. Current Industrial Reports, Glass Containers
Summary for 1976. M 32G(76)-13. May 1977.
79. U.S. Bureau of Census. Current Industrial Reports, Consumer, Scienti-
fic, Technical, and Industrial Glassware. MA-32E(76)-1. 1976.
80. Reference 73, Table 1.
81. U.S. Bureau of Census. Current Industrial Reports, Fibrous Glass
1976. MA-32J(76)-1. June 1977.
82. Reference 69, page 4.
9.7-143
-------
83. Hutchings, J. R., and R. V. Harrington. Glass. In: Encyclopedia of
Chemical Technology, Second Edition. Kirk and Othmer. John Wiley and
Sons. New York, NY. 1966. Page 549.
84. Danielson, J. A. (ed.). Air Pollution Engineering Manual. National
Center for Air Pollution Control. Cincinnati, OH. Public Health
Service Publication No. 999-AP-40. 1967.
85. Santy, M. J. Particulate Control Through Process Modification Chemical
System Screening. Prepared for Glass Container Manufacturers Institute.
New York, NY. December 1971. Phase II, page 2.
86. Stockham, J. D. The Composition of Glass Furnace Emissions. Journal
of the Air Pollution Control Association. Z^:713-175. November 1971.
87. Reference 69, page 45.
88. Reference 84, page 730.
89. Ryder, R. J., and J. J. McMackin. Some Factors Affecting Stack
Emissions From a Glass Container Furnace. rhe Glass Industry. 50:
307-311. June 1969.
90. Reference 69, Table 7.
91. Reference 70, Table 8.
92. Reference 71, Table 8.
93. Reference 73, Table 9.
94. Reference 73, Table 11.
95. O'Sullivan, W., Krauss, C., and Londres, E. Fuel Oil Particulate
Emissions From Direct Fired Combustion Sources. New Jersey Bureau
of Air Pollution Control. Trenton, NJ. January 26, 1976.
96. Preliminary Data from Emission Tests on Uncontrolled Glass Furnaces.
U.S. Environmental Protection Agency, Industrial and Environmental
Research Laboratory. Cincinnati, OH. April 1978.
97. OAQPS Data File of Nationwide Emissions. U.S. Environmental Protection
Agency, Office of Air Quality Planning and Standards. Research
Triangle Park, NC. February 1979.
98. Spinosa, E. D. and R. A. Newton. Chemical Analysis of Particle Size
Fractions of the Particle Catch from Glass Melting Furnaces. U.S.
Environmental Protection Agency, Industrial and Environmental Research
Laboratory. Cincinnati, OH. EPA Contract No. 68-02-2552. March 1979.
99. Draft Background Information Document. Proposed Standards of Perfor-
mance: Glass Manufacturing Plants. U.S. Environmental Protection
Agency, Office of Air Quality Planning and Standards. Research
Triangle Park, NC. June 1979.
9.7-144
-------
100. Losel, R. E. Practical Data for Electric Melting. The Glass Industry.
February 1976. Page 26.
101. Schorr, J. R., et al. Source Assessment: Pressed and Blown Glass
Manufacturing. Plants. EPA Contract 68-03-1326. NTIS 60/2-77-005.
Task 37, Page 103.
102. Teller, A. J. Control of Furnace Emissions. The Glass Industry. New
York, NY. February 1976. Page 22.
103. Corning Glass Response to 114 Questionnaire by EPA. Corning, NY.
October 12, 1977.
104. Glass Container Corporation Response to 114 Questionnaire by EPA.
Fullerton, CA. October 14, 1977.
105. Reference 71, page 79.
106. Custer, W. Electrostatic Cleaning of Emissions from Lead, Borosili-
cate, and Soda-Lime Furnaces. In: Collected Papers of the 35th Annual
Conference on Glass Problems. University of Ohio. Columbus, OH. 1974.
107. Communication at meeting of Owens/Corning Fiberglass representatives
and Dave Powell of PES. Pacific Environmental Services. Santa Monica,
CA. September 29, 1977.
108. Liu, C. T. Screening Study for Background Information and Significant
Emissions from Fiberglass Manufacturing. Vulcan-Cincinnati, Inc.
Cincinnati, OH. December 1972.
109. Environmental Considerations of Selected Energy Conserving Manufacturing
Process Options, Volume XI: Glass Industry Report. U.S. Environmental
Protection Agency. Publication No. EPA-600/7-76-034. December 1976.
110. Compilation of Air Pollution Factors, Second Edition. U.S. Environ-
mental Protection Agency. Publication No. AP-42. April 1973.
111. Source Category Survey: Mineral Wool Manufacturing Industry. U.S.
Environmental Protection Agency. Publication No. EPA-450/3-80-016.
March 1980.
9.7-145
-------
Page Intentionally Blank
-------
9.8 METALLURGICAL INDUSTRY
Fugitive and process particulate emissions from the following segments
of the metallurgical industry are discussed in this section:
9.8.1 Iron and steel
9.8.2 Ferroalloy production
9.8.3 Gray iron foundries
9.8.4 Steel foundries
9.8.5 Primary aluminum
9.8.6 Primary copper smelter
9.8.7 Primary lead smelter
9.8.8 Primary zinc smelter
9.8.9 Secondary aluminum operations
9.8.10 Secondary copper smelting and alloying
9.8.11 Secondary lead smelting
9.8.12 Secondary zinc processing
The metallurgical industry is a major source of particulate emissions.
Furnace operations in the industry are the primary source of process par-
ticulate emissions. In most cases, the furnaces employed in the metal produc-
tion operations discharge high temperature exhaust effluents which must be
cooled and often further conditioned before ducting to a control device.
The control device must usually be capable of high efficiency collection of
submicrometer particles, especially in processes where oxygen lancing is
used. Iron and steel plant furnaces, such as basic oxygen furnaces, emit as
much as 80 percent of particulates below 1 micrometer in diameter.
Fugitive emissions from metallurgical processes are also a significant
source of particulate emissions. In general, fugitive emissions occur during
loading and unloading of the raw materials on the plant premises, transfer
and conveying, storage, charge to the furnace, travel on unpaved plant haul
roads, and truck traffic to and from the processes. Section 5 of Volume 1
discusses the control of industrial process fugitive emissions common to
many industrial operations (e.g., storage and plant haul roadways). A dis-
cussion is presented in each subsection below of the control of fugitive
emissions sources that are more specific to the industry segment being
evaluated.
9.8-1
-------
Particle size distribution data are included in each subsection. In
many cases, when available, chemical characteristics of the emissions are
also provided for different processes as well as secondary environmental
impact information.
9.8.1 Iron and Steel Plants
This section discusses sources of particulate emissions and their
control techniques. The processes include the following:
9.8.1.1 Coke ovens
9.8.1.2 Sintering
9.8.1.3 Blast furnace
9.8.1.4 Open hearth furnace
9.8.1.5 Basic oxygen furnace
9.8.1.6 Electric arc furnace
9.8.1.7 Rolling, shaping, and finishing
9.8.1.8 Scarfing
Figure 9.8.1-1 shows a flow diagram for the above processes in a
typical integrated iron and steel pi ant.1
9.8.1.1 Coke Ovens. Coke is a carbon material produced by the de-
structive distillation of coal. Coke is manufactured by heating coal in
the absence of oxygen to drive off the volatile compounds. It is used in
iron and steel foundries and, more often, in integrated steel mill blast
furnaces. Coke plants are either an integral part of steel mills or are
located in the vicinity of iron and steel manufacturing facilities. About
99 percent of the United States coke production is made in "by-product" (or
slot-oven) coke ovens, with the remaining coke being made in beehive ovens
from which by-products are not recovered.2 Sixty-five plants with 13,324
ovens grouped in 213 batteries produce coke. These plants tend to be con-
centrated north of the Ohio River and east of the Mississippi River.
Coke oven batteries range in size from relatively small batteries to
newer, larger batteries. Table 9.8.1-1 lists typical products from a by-
product coking plant. About 40 percent of the coke Oven gas is returned to
the oven and used as process fuel to heat the oven. Because of the rela-
tively high heating value of this gas, it is an important source of fuel gas
for other operations in an integrated iron and steel mill.
9.8-2
-------
.FLUE DUST FROM,
BLAST FURNACE
10
bo
i
co
BALLING DRUM
\-GRATE FEEDER
ROLLED TOl
SLABS. BILLETS
BY-PRODUCT
RECOVERY
STEEL FURNACE
TO SINTER
PLANT
QUENCHING
TOWER
COKING
FURTHER
PROCESSING
Figure 9.8.1-1 A composite flow diagram for a steel plant.1
-------
Table 9.8.1-1. COKE AND COAL CHEMICALS PRODUCED BY U.S. COKE
PLANTS IN 19782
Yield
Product Per megagram coal charged
Coke 0.6878 megagram
Breeze 0.0531 megagram,
Crude tar 34.75 liters
Crude light oil 10.22 liters
Ammonia 6.77 kilograms
(sulphate equivalent)
Coke oven gas 331 cubic meters3
Estimated.
9.8-4
-------
Approximately 82 Tg of coal is consumed per year in the production of
coke. The compound annual growth rate for this industry is estimated to be
one percent.3 Nationwide participate emissions in 1977 were estimated
to be 108 Gg per year from the by-product coking process and 54 Gg per year
from the beehive process.4
9.8.1.1.1 Process description and emissions. The by-product plant coke
production process can be divided into the following three subprocesses:
o Coal preparation
o Coke production (thermal distillation)
o By-product production
Coal preparation involves the blending and pulverization of coal.
Blending is done in order to develop the desired coke properties, to obtain
the optimum types and quantities of by-products. The coal is pulverized in
order to produce the desired physical properties in the coke. The coal is
then either charged into the ovens by the use of a larry car which fills the
ovens through charge holes in the roofs of the ovens, or the coal is heated
and dried before charging. Traditionally, coke ovens have been charged with
"wet" coal, which contained 6 to 11 percent water by weight. In the dry-coal
charging process, moisture is driven off in the drying process, and the hot
coal is then charged into the coke oven by means of a pipeline, a mechanical
conveyor, or a hot larry car. Dry coal charging has only recently been
applied commercially. However, of the nine coke oven batteries recently
built, five of these employ dry coal charging.5
The coal is typically heated 15 to 18 hours to produce blast-furnace
coke and 25 to 30 hours for foundry coke. The heating occurs in long narrow
ovens in the absence of oxygen until a temperature of about 900 to 1100°C
is reached.6 The ovens have typical dimensions of 3.6 to 6.7 meters in
height, 12.2 to 16.8 meters in length, and 0.36 to 0.51 meters in width.
The ovens share common walls and are heated by a regenerative combustion
system located underneath.2 ,
The volatile components in the coal are driven off, leaving the high
carbon content, non-volatile coke in the oven. At the end~of the heating
cycle, the doors are opened on both sides of the coke oven, and the
incandescent coke is pushed from the oven into a special hopper rail car
9.8-5
-------
(called a quench or hot coke car) by a mechanical ram. The hot coke in the
quench car is then moved to a "quench tower" where the hot coke is cooled by
the addition of large quantities of water. The coke is then discharged from
the quench car, allowed to drain and cool, and then crushed and screened.
The coke can then be used as a fuel and a reducing agent for the production
of iron. Any small size coke (breeze) produced as a result of the crushing
and screening is usually used in other steel-plant processes or sold.
The volatile components or gases driven off by the coke oven heating
process are transferred from the oven by exhaust offtake pipes (called stand-
pipes or ascension pipes) into the collector main and directed to the by-
product plant, where the gases are cooled to 80 to 100°C by spraying with
water. About 20 to 35 percent of the initial coal charge is volatilized.
Tars and other viscous materials are condensed in the water, and separated
from the water or flushing liquor in a decanter. Ammonia can be recovered
either as an aqueous solution or as a salt (e.g., ammonium sulfate). The
remaining gas is rich in hydrogen and methane and has a heating value of
about 20.5 MJ/Nm3. This gas is then combusted in the oven underfiring
combustion system and used to heat the ovens; the gas is also used in other
steel production processes.
Figure 9.8.1-2 presents a flow sheet showing the major steps in the
coking of wet coal using the by-product process, including the by-product
recovery steps. Figure 9.8.1-3 presents a schematic of a coke battery, with
emission points shown. The primary sources of dust and smoke in the
coal by-product coking process described are as follows:
o Unloading, handling, and stockpiling coal
o Handling, crushing, screening, and blending coal
o Drying and heating of coal (dry coal charged batteries only)
o Charging of coal into the coke ovens
o Smoke leakage from charging lids and standpipes
o Underfiring of coke ovens (battery stacks)
o Smoke leakage around coke oven doors
o Pushing of coke from the ovens
o Quenching of hot coke
o Handling, crushing, and screening of coke
9.8-6
-------
Coal Haiit I
00
-vj
«.». Cor
Dumper
1 . Co
Coo
Slorogt*
X
J r I
— •>> '' _
Mi. ing ~*
*
7
Cool t.idac
XZJ
Fluihing Liquor
Foe Further
frocvtling
or For Ftwl
f luiKing
Liquor
Oecartler
Tar
Cole
Coke
Outlier
Metallurgical Coke
Coke Screening!
' Row
* 1 iGai
U li
J
I
I Gat For Underlying Coke Ovent
AciH
I
NH3
T
Ligt.1 Oil
Scrvt.ber
1
- »
Debe
Sul'ide
Scrubber
iioliied Wot
-1*
Oil
kntoliiod Woth Oil
Got
Holder
Gat 14
Sleel Flaot
f«9fn Stripper
Figure 9.8.1-2 Flow sheet showing the major steps involved
in the carbonization of coal using the by-product
process and the subsequent recovery of coal,
chemicals from the gases generated at the
ovens.2
-------
IO
ba
CO
Quenching
Emissions Coa| Rorfs
Waste Gas
Stack
Pusher Car
Coke Guide
Figure 9.8.1-3 Schematic diagram of a coke battery.7
-------
The processes of unloading, handling, stockpiling, crushing, screening
and blending of coal and the handling, crushing, and screening of coke are
all potential sources of fugitive emissions of coal and coke dust particles
to the atmosphere. Control techniques applicable to these emissions are
discussed in Section 5 of Volume 1.
The charging of coal into a hot incandescent coke oven results in the
formation of a large volume of steam, gases, and smoke, which is forced from
the oven by the resulting increase in pressure. The steam and the gases
result from the evaporation of any moisture present and from other volatile
components. The smoke results from particles and fumes entrained with the
gases. Emissions will occur at any oven opening to the atmosphere, such as
out an open charge port. Particles emitted during the charging cycle have
been identified as coal coke balls, pyrolitic carbon, high-temperature coke,
char, coal, mineral matter, and flyash.
After charging, the charging lids and standpipes must be sealed. Im-
properly fitted lids, cracks, and broken seals in the standpipes will result
in particulate emissions. These emissions are expected to have properties
similar to those resulting from charging, without the larger particles of
coal and coke.2 ~
In the by-product plant, the various tars, light oils, phenol and
ammonia are removed from the raw coke oven gas, and the remaining combusti-
ble gas is then used to heat the ovens. In some plants, combustible by-
product gas from blast furnaces is used to heat coke ovens, and the coke
oven gas of higher heating value is used in other plant processes.? Almost
all coke batteries are modified using natural drafts provided by a tall
stack. Figure 9.8.1-4 presents a schematic of several types of oven heating
flue designs. The particulate matter emitted by coke oven modified combus-
tion stacks is very small in size. Data from impactor sampling showed that
40 to 95 percent of the particulate matter is smaller than 1 micrometer in
diameter with an average of 90 percent smaller than 3 micrometers.3 These
emissions are primarily the result of cracks in oven walls which cause dust
and coke pyrolysis products to leak into the exhaust flues.
The end of each coke oven has refractory-lined doors that are opened
during the coke discharge or pushing operation and closed again during the
9.8-9
-------
Charging Hole*
Dividing Wall
"Of •"<>> I • —Coking
LFuel Gas Inlets Chamber
EARLY KOPPERS OVEN
Charging Ho lei
Corrbuttion Occurs
In Vertical Fluej'
(Not Shown}
•Crossover
Flue.
J
H
s
H
j
X.
1
•^1
^
t
V
*»*,
J
}
~~^
L
^
r1^
Heated Ai
from Reg
1 % t
SFm
eneraton \ T*
r^
^
Fuel GOJ Inlet-
H^
oking
Chamber
KOPPERS-BECKER OVEN
'Charging Holes
Combustion Occurs
in Vertical Flues
(Not Shown)
Fuel Gas Inlets
Coking Chamber
Combustion
Occurs Hera
Figure 9.8.1-4 Schematic diagram of three common flue designs.
9.8-10
-------
charging and coking operations. If these doors are not sealed properly,
participate emissions will leak from any gaps between the door and the oven
doorjam. The pushing of the hot incandescent coke from the ovens tnto the
charge car may result in a significant quantity of particulate emissions
(mostly coke dust). Emissions are particularly severe if the coal has not
been adequately coked and the remaining volatile products burn as the charge
is exposed to oxygen during the pushing operation.8 The particulate matter
from inadequately coked coal will contain coal dust and coke dust mixed with
condensed tars.6
The rail car containing the hot coke charge is moved to a quench tower,
where it is cooled by the application of water to the hot coke. This re-
sults in the formation of a steam plume which tends to mask the particulate
in the plume and makes particulate sampling difficult. The limited work
that has been performed to size quench tower particulate emissions shows
that the majority of particles from a controlled quench tower (i.e., a tower
with a baffle system) are less than 10 micrometers in diameter; particulate
emissions are greater than 10 micrometers in. diameter for an uncontrolled
quench tower.9
The stack parameter data for emission sources in metallurgical coking
operations are summarized in Table 9.8.1-2.10 Table 9.8.1-3 gives an indi-
cation of the size distribution of emissions from coke pushing during the
maximum emission periods. At this time, these are the only data available
in the EPA Fine Particulate Emissions Information System (FPEIS) data base.
9.8.1.1.2 Particulate control techniques.
Charging—The control of charging emissions can be accomplished by the
following three methods:
o Stage charging
o Sequential charging
o Scrubber control systems
Scrubber control systems which represented initial efforts to curtail
charging emissions utilize a scrubber mounted on the larry car to control
emissions captured by hooding devices at the charging ports. EPA concluded
that such systems are costly, complex, and susceptible to severe maintenance
problems, and that they have the additional problem of scrubber water
disposal.2 Few if any such systems are still in operation.
9.8-11
-------
Table 9.8.1-2. STACK PARAMETERS FOR BYPRODUCT METALLURGICAL COKE
MANUFACTURING EMISSION SOURCESl°
Facility or
operation
Oven charging
Oven pushing
T3 Quenching
ro
Unloading
Underfiring
Coal crushing/handling
Nationwide
number of
facilities
or
operations
92
65
84
7
60
9
Average
emission
source,
m
62
67
34
38
70
42
Average
emission
source
diameter,
m
2.8
2.9
5.3
6.1
3.0
1.0
Average
temp.,
oc
291
380
107
64
228
21
Average
flow,
AITH/S
40.8
45.6
131.0
81.8
44.3
7.53
Average operating rate,
Gg of load charged
per facility,
per year
435
383
511
345
387
609
-------
Table 9.8.1-3. SIZE SPECIFIC EMISSIONS FROM COKE PUSHINGa
Mass concentration,
mg/DNCM(mass percent less stated size)
Total 15.3/tm 12.9//m 10.1/-im 7.28/urn 5/*m 2.5/nm 1.01/wn
Uncontrolled 283 274(97.0) 273(96.6) 271(95.8) 266(94.0) 257(90.9) 211(74.6) 176(62.2)
aCoke oven pushing during 2-6 minute maximum emission period. No control device.
.
00
1—"
CO
-------
Stage charging is a procedure for charging coal into coke ovens in such
a manner that the passageway to the offtake gas collection system remains
open. The gases and particulate emissions generated in the oven are drawn
into the collector main by steam aspiration (or aspiration using a liquor)
at the standpipe and exhausted by the regular gas handling equipment into
the by-product recovery plant. This technology is identified as the best
control technology for wet-coal charging.2 Sequential charging differs
from stage charging only in that the oven is charged more rapidly than in
stage charging, thus producing gases and particulate matter are generated
at a faster rate.
Providing either an adequate degree of aspiration-or a slightly nega-
tive oven pressure is the key to a successful stage charging operation.
Aspiration efficiency is very sensitive to flow and pressure maintained at
the aspirator nozzle. Also, for the system to work properly, the door and
charge openings must be minimized, and the standpipe leading to the aspirator
must be kept clean. Figure 9.8.1-5 shows a sketch of a typical oven aspira-
tion system.
Batteries which have ovens, fitted with gas offtake standpipes on only
one side must be modified for stage charging to provide adequate aspiration of
emissions into the collector main. This can be achieved in two ways: (1) by
retrofitting the entire battery so that every oven has another offtake standpipe
at the other side, connected to a second collector main installed along the
length of the battery; or (2) by retrofitting or replacing the charge car so
that a jumper pipe is installed which permits emissions generated during
charging to be drafted into an adjacent oven and thus be directed, into the
single collector main through the standpipe of the adjacent oven.
Topside Leaks—The best control technology for the control of emissions
from charging lids and standpipes (topside leaks) essentially involves proper
maintenance and replacement of oven closure devices, sealing (luting) of
lids and standpipes, inspection, and resealing. The following list details
these considerations:^
o Replacement of warped lids.
o Cleaning carbon deposits or other obstructions from the mating
surfaces of lids or their seats.
9.8-14
-------
PIPINfi FOR STEAM AND/OR
WEAK LIQUOR
-ASCENSION PIPE
DOOR
OVEN
Figure 9.8.1-5 A typical aspiration system operating in the
return bend of the gas-collection duct.2
9.8-15
-------
o Patching or replacing cracked standpipes.
o Sealing lids after a charge or whenever necessary with a slurry
mixture of clay, coal, and other materials (commonly called lute).
o Sealing cracks at the base of a standpipe with a slurry mixture.
Battery Stacks—Control techniques for battery stacks are broadly grouped
into two categories: (a) operating and maintenance practices, and (b) add-on
control devices.11
Operating and maintenance practices encompass inspection, diagnosis, and
adjustment or maintenance, on some scheduled or routine basis. These procedures
vary from plant to plant; most plants utilize some of the practices to some
degree. Also, the needs of one plant may differ from another, depending on
battery age and condition and other factors. These procedures are aimed at
preventing emissions by maintaining proper firing conditions and by minimizing
leaks between the oven space and the flue system.
Other means of control ing stack emissions involve the use of add-on
control devices. Even with the use of such devices, however, some of the
operating and maintenance techniques discussed would probably still need to
be employed to maintain battery emissions at levels that would not overload
the control device and maintain coke quality.
Only two types of add-on control devices are presently used on coke oven
battery stacks: dry electrostatic precipitators and fabric filters. Other
add-on devices that have been pilot tested on battery stacks include: a wet
electrostatic precipitator, a charged droplet scrubber (pilot and full scale),
and an afterburner.
While operating and maintenance practices are used at almost all plants,
the diligence or frequency of their use varies, and some practices are unique
to one or a few plants. The following paragraphs describe the most common
operating and maintenance practices.
Probably the most common operating practice, for minimizing battery
stack emissions, is adjustment of the fuel/air mixture. Such adjustments may
be made by changes in the stack draft damper or changes in the "fingerbars"
that limit the air drawn into the flues of each oven. These adjustments may
be made on the basis of visual observations of the stack, or on the basis of
the continuous oxygen monitor that some plants use. Sometimes the excess air
9.8-16
-------
level is increased in an attempt to reduce stack emissions, but this may be
only a temporary measure, until the cause of the problem can be diagnosed.
Another operating practice that affects stack emissions is "decarboniza-
tion." This is a common practice, intended to remove buildup of carbon at
the top of the ovens; it is important for the control of emissions during oven
charging. Some carbon buildup, however, helps minimize wall and roof leaks.
Therefore, the decarbonization may be decreased at times to increase the car-
bon buildup and thus help seal leaks.
A common, routine maintenance practice is the cleaning of fuel nozzles
that flue cap inspections indicate may be partially plugged. The frequency
with which this is done depends, of course, on the inspection frequency.
Nozzle problems may also lead to excess fuel in the flues, especially in
gun-flue batteries.
Probably the most common coke oven maintenance practice for maintaining
coke quality and minimizing wall leakage is end-flue patching. End-flue
repairs usually involve a scheduled maintenance program in which end-flue
cracks are patched routinely using a hand-held slurry patching gun. Patching
is usually done routinely, and whenever wall inspections indicate additional
patching is needed. Thus, the frequency of patching varies from plant to
plant. For example, ovens at U.S. Steel-Fairfield are patched at least once
every 3 months, while those at Kaiser Steel are patched every 1 to 1-1/2
months. At CF&I, the coke side is patched twice; the pusher side, once every
5 weeks. In addition, at CF&I, each oven is inspected once every 3 days and
additional patching done if necessary.
Some of the above common operating and maintenance practices are also
used in conjunction with other special maintenance techniques, such as silica
dusting and mobile gunning. Silica dusting is a supplementary maintenance
technique used for sealing small cracks"in oven walls by blowing finely ground
silica powder into a sealed oven. The silica dust fills the cracks in all
wall areas of the oven. Mobile gunning is an alternate patching technique
used in conjunction with inspection, diagnosis, and other common maintenance
practices. Mobile gunning supplements hand-held spray patching of wall
cracks. Cracks in the interior portion of oven walls and roof not reachable
9.8-17
-------
by hand-held patching can be patched by mobile gunning. In recent years,
mobile gunning has been placed in service at a number of plants.
Diligent and rigorous use of most of these common practices would help
minimize stack emissions. The most diligent use of such practices, defined
as "systematic operation and maintenance," involves more frequent end-flue
patching plus additional manpower for other inspections, diagnoses, and
maintenance.
Battery stack emission test data were available for only one battery that
used systematic operation and maintenance—a 31-oven, gun-flue battery rebuilt
in 1960. Battery stack tests (three runs each) were conducted in June 197812
and August 1979.13 Average particulate concentrations by EPA Method 5 were
0.055 g/NnP and 0.089 g/NnP, respectively. Mass emissions of the 1978 test
were 1.67 kg/hr or 81 g/Mg of coal based on a coking time of 25.5 hours. Mass
emissions of the 1979 test were 3.15 kg/hr or 109 g/Mg of coal based on an
18-hour coking time. These tests showed the highest 6-minute average opacity
to be only 11 percent.
Dry electrostatic precipitators (ESPs), one of two types of control de-
vices used on coke oven battery stacks, are basically similar to fly ash
precipitators used in coal-fired power plants. Full-scale dry ESPs have been
installed on five coke oven battery stacks in the United States. These batteries
are listed in'Table 9.8.1-4 along with some of the operating and design para-
meters for each installation. The parameters reflect the fact that control
of battery stack effluents by dry ESPs requires a relatively high specific
collection area, multiple stages, and low gas velocity, in order to offset:
(a) low inlet dust concentration, (b) small particle size, and (c) low resis-
tivity. The paticulate emission test results presented in Table 9.8.1-5 of
full-scale battery stack ESPs shows a wide variation in collection effi-
ciencies. 14,15,16 The negative efficiency for the National Steel battery may
be partly due to an error in measurement of gas flow rate, which averaged 21
percent higher at the ESP outlet than at the inlet.
Particulate removal by fabric filters is usually highly efficient, even
for fine particles; therefore, this add-on device has been used in many ap-
plications. In addition, fabric filters have the advantages of dry dust
recovery and minimal equipment corrosion problems. There are, however, some
9.8-18
-------
TABLE 9.8.1-4 COMPARISON OF ELECTROSTATIC PRECIPITATOR INSTALLATIONS
Operating and
design
parameters
BATTERY NUMBER
Number of ovens
Age of battery
Year installed
IO
^ Gas flow rate, anvVmin
\o Inlet loading, g/Nm3
Outlet loading, g/Nnr3
Efficiency, % • •
Gas velocity, m/sec
SCA, m2/l,000 m3 min"1
Number of stages
Lone Star
Steel
A & B
78
35 yr
1973
3,767
0.135
' 0.087
35
1.3
1,122
3
Armco Houston Works
1
47
36 yr
1976
1,869
0.0183
0.014a
25
0.6
1,909
3
2
15
25 yr
1976
425
0.1693
0.0233
86
0.7
1,909
3
National Steel
Granite City,
USS Clairton Illinois
21
87
Rebuilt in 1972
March 1979
4,616 (design)
0.061 (tested 1975)
Not yet tested
90 (vendor guarantee)
0.7
1,912
Unknown
C
61
18 yr
March
2,605
0.648
0.487
Neg.b
0.9
1979
1,955-2,936
4
aTests were conducted when battery was underfired with natural gas,
showed negative efficiency.
-------
TABLE 9.8.1-5. PERFORMANCE DATA FOR ELECTROSTATIC PRECIPITATOR CONTROL OF COKE
OVEN BATTERY STACK PARTICIPATE EMISSIONSa.b.c
10
•
00
I
re
o
Particulate "
loading (q/Nm3)a
Test
Lone Star Steel I4
E. B. Germany Works
Batteries A and B^
(February 1973)
Test No. 1
Test No. 2
Test No. 3
Armco Steel 15
Houston Works
Battery 1
(November 1976)
Test No. 1
Test No. 2
Test No. 3
Armco Steel 15
Houston Works
Battery 2
(November 1976)
Test No. 1
Test No. 2
Test No. 3
National Steel16
Granite City
Battery C
(July 1979)
Test No. 1
Test No. 2
Test No. 3
Inlet
0.1156
0.1050
0.1849
0.0211
0.0178
0.0156
0.1721
0.0746
0.2590
0.923
0.371
--C
Outlet
0.0995
0.0714
0.0883
0.0211
0.0110
0.0087
0.0291
0.0126
0.0254
0.847
0.375
0.240
Particulate emis-
sions (kq/hr)a
Inlet
7.39
6.67
11.48
1.38
1.18
1.03
2.42
1.07
3.97
63.8
25.1
— c
Outlet
6.44
4.49
5.76
1.35
0.78
0.56
0.48
0.22
0.42
70.4
25.1
19.1
Collection
efficiency
(weight %)
12.9
32.7
49.8
2.2
33.9
45.6
80.2
79.4
89.4
Neg.
Neg.
--
Outlet particulate
emissions
g/wg
143
99
128
35.9
20.7
14.9
41.9
19.2
36.7
2,505
1,160
g/hr/Hg
6.30
3.68
4.72
1.71
0.99
0.71
1.90
0.87
1.67
129
58
35
aprobe, cyclone, and filter catch (EPA Method 5).
bCombined stack.
clnvalid sample.
-------
disadvantages. Since filter units are usually rather large, installation
space may be a problem. Fabric filters also involve much higher pressure
drops: on the order of 1.0 to 2.3 kPa, compared to 0.5 kPa or less for dry
ESPs. Also, when handling combustible dust such as that emitted from a
battery stack, there is a potential fire hazard. For example, at the Kaiser
Steel Company's Fontana plant, some fire damage occurred in part of the com-
partments on one of the fabric filters. Kaiser believes the fire was caused
by misoperation of the battery, which increased the oxygen level in the flue
gas. Kaiser has taken precautions to protect the filters from such
occurrences.
Four fabric filter units have been placed in service at Kaiser Steel-
Fontana, and the company plans to install fabric filters on all seven bat-
teries at that plant, the only plant that has installed fabric filters for
control of battery stack emission.
Operating parameters in Kaiser's graphite-silicone treated, glass-fiber,
reverse air-cleaned filters are: a net air-to-cloth ratio of between 0.5 and
0.6 M-Vmin per M^, pressure drops between 1.5 and 2.2 kPa and temperature
between 177 and 243°C.
Particulate emission test results of these fabric filter controlled
batteries are shown in Table 9.8.1-6.17,18,19,20,21 The compliance tests on
battery B show 'a higher average emissions due partly to a lower coal-feed
rate and capacity during the tests because 12 ovens were out of service* The
higher emissions reported at this battery during the EPA test may have been
due to poor cleaning of the bags as evidenced by the small change in pressure
drop of less than 0.3 kPa after completion of each cleaning cycle. Opacity
data on all filters showed a maximum 6-minute average opacity of 5 percent.
Door Leaks—The oven doors on both ends of each oven can account for a
large percentage of visible coke oven emissions.22 The volatilization of
the coal results in a positive oven pressure that will push emissions through
improperly sealed doors and door jambs into the atmosphere. One study con-
cluded that door jamb warpage was the most serious problem.23 This study
recommended that door jambs should be as close to square or rectangular as
possible, with a low moment of inertia and not excessively stiff, so that
internal stresses will not cause warpage of the jamb when heated. It was
9.8-21
-------
TABLE 9.8.1-6. SUMMARY OF TEST RESULTS ON FABRIC FILTERS AT KAISER STEELl7»18,19,20,21
IQ
•
00
ro
[S3
EPA test
Test period
Battery Capacity (Hg)
Particulate
concentration,
g/dNm3
Average
Flue gas flow,
dNm-Vmin
Average
Average flue gas
flow per unit of
capacity (Nm^/hr
per Mg)
Particulate emissions
(kg/hr)
Average
Particle emission rate,
g/Mg of coal
Average
Particulate emission rate,
g/Mg of coal
Average
Filter total
pressure drop
(cm 1)20)
Average
BI:
Inlet
Sept.
0.2481
0.1626
0.1233
7)71780
1,000
999
1.023
ITUOT
14.89
9.75
7.57
T077T
J
Outlet
79
420
0.0625
0.0788
0.0323
0.0579
1,033
1,043
1.123
1,033
148
3.87
4.94
1.98
TTSO
157
80
200
146
9.21
11.76
4.71
~8756
21.6
19.6
21.1
2078"
Outlet
July 79
420
0.0215
0.0272
0.0244
978
1,123
T755T
150
1.27
1.83
iriir
79.9
115.1
97.5
3.02
4.36
JTeV
16.4
17.7
T77T
Compliance tests3
Outlet
Mar. 79
572
0.0135
0.0128
0.0108
070l24~
1,044
1,052
1,056
1,051
110
0.844
0.807
0.676
0.776
27.4
27
22.7
2577
1.48
1.41
1.18
1.36
14.4
14.4
15.8
T379
D20
Outlet
Nov. 78
572
0.0208
0.0229
0.0176
oToloT
994
982
1.031
T.SOT
105
1.24
1.35
1.09
T72T
37.2
40.5
35.6
37.8
2.17
2.36
1.91
ITIF
NA
NA
NA
£21
Outlet
Feb. 79
572
0.0025
0.0014
0.0030
0.0023
1,439
1,389
1,314
T738T
145
0.22
0.12
0.23
07T9"
7.2
3.9
7.1
0.22
0.12
0.23
TJ73T
21.3
20.7
19.7
2076"
batteries were fired with COG during the tests, except for battery E which was fired with BFG.
-------
determined that gray cast iron is not a satisfactory jamb material; ferritic
ductile iron was recommended as a suitable material. The study concluded
that upgraded metal seals have the best potential for meeting all of the
emission control and retrofit criteria. A spring type seal, with no point
loading, made from material that can withstand temperatures as high as 427°C
is best for the construction of effective seals. CF&I Steel Corporation re-
placed 316 stainless steel door sealing edges with an edge fabricated from
"NiCuti" material which proved superior for resisting damage and for ease of
repair.24 A buildup of carbon between the door and oven can interfere with
door seals, and a systematic program to remove this material is recommended.
Data gathered by the U.S. EPA at coke oven batteries with aggressive operation
and maintenance programs to control door emissions indicate that emissions
from as few as 12 percent of the doors is achievable on a continuous basis.2
Short batteries with superlative control have achieved emissions from an aver-
age of as few as 4 percent of the doors, whereas tall batteries have achieved
emissions from an average of as few as 7 percent of the doors.2.
Pushing—The quantity of emissions from coke pushing is highly variable.
Particulate emissions include coke entrained iri the strong thermal current of
air that rises above the hot coke and the products of incomplete combustion
of the volatile residues in the coke. Coke that has not been completely car-
bonized (with a high percentage of residual volatile constituents) is called
"green" coke. All coke may have some residual volatile constituents, although
poor operation and maintenance can result in much larger quantities. As stated
previously, in Section 9.8.1.1.1, pushing emissions from "green" coke are
greater than from coke that has less volatile constituents remaining.
Three primary operating variables affect the greenness of the coke and,
hence, the quantity of emissions. The variables are: heating rate (or tempera-
ture of the oven), coking time (from when the coal is charged to when the coke
is pushed) and heat distribution within the oven. The first two are interrelated.
Within some range, the time required for the distillation of volatiles to near com-
pletion is inversely related to the fourth power of the inner wall temperature.2
The necessary time and heating rate are affected by the oven design and type
of coal charged. In general, these are operating parameters that can be adjusted
to achieve good coking. If not sufficient, coking time can be extended or the
heating rate of the battery can be raised to reduce emissions. There is, however,
9.8-23
-------
an upper temperature limit above which the refractory in the ovens may be
damaged.
The third variable, heat distribution in the oven, affects the greenness
of the coke as follows: if sufficient heat does not reach all portions of
the coal charged to the oven, there will be pockets or areas of green coke
when it is pushed, hence more emissions.
Causes of insufficient heat to portions of an oven include non-homogeneity
of the coal blend charged to the oven, its moisture content and bulk density.
Specifically, a pocket of coal with a high moisture content can cause the
temperature in that area to rise more slowly than in the rest of the oven.
As a result, when the remainder of the coal has completed coking and the
coke is pushed, that pocket of coke is still green.
A more significant cause of poor distribution is inadequate adjustment
or maintenance of the heating system. For example, a faulty gas nozzle or
a blocked passage in the system of heating flues can reduce the heat reaching
a section of an oven. Flues may break or become clogged with debris from
spelling or cracking of the refractory. Blocked flues are particularly a
problem near the ends of an oven where the periodic removal and replacement
of oven doors creates greater temperature changes which cause deterioration
of the oven refractory.
Table 9.8.1-7 presents available data on the quantity of uncontrolled
emissions from the pushing operation.25-38 j^g data are presented in three
categories: fine, coarse and fugitive. This breakdown corresponds to the
manner in which the data were collected. Since uncontrolled pushing emissions
are not confined to a stack (which presents extreme sampling problems),
measurements have been performed at plants where the emissions are captured
and exhausted through a duct. Some estimates have been made of the emissions
that escape capture; hence, the distinction on Table 9.8.1-7 between the
"fugitive" and other categories. A distinction between "fine" and "coarse"
is made when the emissions measured were captured by a shed. A shed segregates
the small and large particles by entraining only the small particles in the
exhaust gas stream and allowing the large particles to settle under the
shed.
Considerable variation in the data can be seen in Table 9.8.1-7. The
variation in the "fine" column probably reflects, among other factors, the
9.8-24
-------
TABLE 9.8.1-7. DATA ON UNCONTROLLED PARTICULAR EMISSIONS FROM PUSHINR25-38
CO
I
ro
en
Uncontrolled particulate emissions,
grams/kilogram of coke produced Numbe>
(pounds/ton of coke produced) a »b>c of
Test Reference Plant Test by Fine
la 25 L EPA 0.14 (0.28)
Contractor
Ib 25 L EPA 0.42 (0.83)
Contractor
2 26 L Plant 0.13 (0.25)
Contractor
3a 27 L Plant 0.22 (0.44)
0.26 (0.52)
0.43 (0.35)
3b 27 L Plant 0.34 (0.68)
4 28 L Plant
5 29 T EPA 0.39 (0.78)
Contractor
6 30,31 T Plant 0.24 (0.48)
7 30,31 T Plant " 0.40 (0.80)-
8 30,31 T Plant 0.31 (0.62)
9 30,31 T Plant
10 32 T Plant 0.38 (0.76)
11 33,34,35 P ' Plant 0.23 (0.46)
Contractor
12 36 K Plant 0.76
Contractor
13 37,38 V Plant 0.25
Multiply the values in this column by 0.7 to convert
Coarse Fugitive Total sampl<
0.8 (1.6) 0.03 (0.05) 1.0 (1.9) 3
1
5
3
, 7
1 5
1
30
0.7 (1.4)
0.32 (0.64) 0.10 (0.19) 0.8 (1.6) 3
2
7
2
0.34. (0.68)
23
2 (4) ' " . . 2
(1.5) 0.44 - 1.2 1.2 - 2.0 9
(0.88 - 2.4) (2.4 - 3.9)
(0.50) 3
to a denominator of weight of coal charged.
''For tests 1 through 11, the values in the "fine" column are for measurements in the exhaust duct
based on measurements of dustfall under the sheds.
size, velocity, etc., of unconfined plumes. For t-.ps
The entry in the "fugitive" column was estimated as
Those in the "fugitive" column are estimates
ts 12 an'1 n, the values in the "fine-coarse"
for the other tests.
emissions from oven door leaks durino the 1
Number
r of
pushes .
&s° sampled" Comments
36
5 The sample was collected when the
pushes were much greener than during
collection of the 3 samples above.
5
about 18 The average greeness of the pushes
about 42 sampled increased in the order the
about 30 values are listed.
about 180 Includes the 15 samples listed for
Test 3a
60
16-24
56-84
16-24
184-230
not known The plant was operating on extended
coking times.
168 The two values in the "fugitive"
column were obtained by two
different estimation techniques.
48
from a shed. Those in the "coarse" column are estimate;
based on rough measurements of the particulate concentr;
column are for measurements In the exhaust duct from a 1
to 3 minutes-per-push sampling time. Door emissions an
mixed with pushing emissions In the exhaust duct from the sheds tested. The door emissions should be less than 5 percent of the values shown (based on
measurements o* these emissions during tests 1 and 5). •
''The number of samples and pushes sampled refer only to the data in the "fine" column.
-------
greenness of the coke pushed. Due to this variation, the data are best
viewed in terms of ranges of the data.
The California Air Resources Board has identified three feasible
methods presently being used in this country for controlling coke pushing
emissions.22 These methods involve the use of a "one-spot (or enclosed)
quench car," the installation of a "full-length shed" over the coke exit
side of the oven, and "traveling hoods." The one-spot quench car provides
exhaust gas hooding during the pushing of the coke from the oven into the
car. The collected gases and particulate matter are drawn from the enclo-
sure into an exhaust gas cleaning car which employs a scrubber for particu-
late removal. The full-length shed collects and contains the emissions
generated from the pushing operation while the air is drawn from the shed
into a particulate control device. The traveling hood involves the use of a
mobile hood (attached to the quench car or coke guide) which is connected to
an overhead exhaust main that feeds the collected gases and particulate
matter into control equipment. The emissions generated from pushing green
coke are capable of overloading most control systems. Thus, it is critical
that the coal be properly coked to preclude the handling of green coke.
Captured particulate emissions and their size distribution from a coke
pushing operation controlled with a traveling hood system are shown in Table
9.8.1-8.35 The average hood capture efficiency was estimated to be 41 percent
based on measured dust concentrations above the hood. An engineer observer's
estimate of the hood capture efficiency for these pushes was 66 percent. The
coke being pushed was estimated to be moderately green by a panel of indepen-
dent, non-biased evaluators.
One source reports that it is feasible to use fabric filters to reduce
emissions captured with full-length shed and traveling hood control systems.6
The condensed tars that reach the filter material can cause blinding or
plugging of the bags unless the ratio of solid particulate to tar is 20 to 1
or greater. At these ratios the particulate matter will become sufficiently
mixed with the tar in the duct leading to the filter, thus preventing the tar
from coming in direct contact with the filter material. Where particulate to
tar ratios are less than 20 to 1 (which is the case with most domestic coking
9.8-26
-------
ID
*
CO
ro
Table 9.8.1-8. PARTICIPATE EMISSIONS AND SIZE DISTRIBUTION FROM COKE PUSHING35
Emission classification Mass emissions, kg/Mg (mass percent less than stated size)
Total
10.1 pm 7. 25 /tin 5 /xm 2.5 /zm 1 ftm
Fugitive 0.85a/0.315t> — — — — — |
Ducted (prior to control) 0.56
Ducted (after control dev1cec) 0.0036
0.084 (15). 0.0644 (11.5) 0.0504 (9) 0.028 (5) 0.011 (2)
0.0034 (87) 0.0032 (82) 0.003 (76) 0.0023 (60) 0.0014 (36)
aDetermined using hood capture efficiencies estimated from dust concentration above hood.
^Determined using hood capture efficiencies estimated by an engineer observer.
cControl device was a 15 kPa venturi scrubber.
-------
operations), the bag material must be protected by either a continuous injec-
tion or a batch precoat of solid material. Materials considered for such use
are coal, limestone, coke breeze, and flyash. Table 9.8.1-8 provides infor-
mation relating to these materials and their uses. Fabric filters employing
a precoating system have been in use at a coke plant in Japan since 1976,
with no significant operating problems reported.6
Quenching—The control of quench-tower particulate emissions is ac-
complished by the use of baffles. Baffles are used primarily as mist elimi-
nators and are designed to intercept particulate matter and water droplets
carried in the quench tower vapor updraft. It has been reported that partic-
ulate removal for baffles ranges from 50 to 95 percent, depending on the
baffle type used.9 It is also reported that the use of dirty quench water
can result in emissions 1.5 to 3 times greater than if clean water is used.
Using a mathematical model, one source developed a strategy for reducing
particulate matter emissions that employed the following techniques:-^
o Increase baffle angle
o Increase liquid particle growth by addition of cooling water
o Increase gas velocity
Dry Coal Charging—If the dry coal charging process is used, the gas
used to preheat the coal should be cleaned before exhausting it to the atmos-
phere. Low-energy scrubbers and dry electrostatic precipitators have been
used to control the preheater stack emissions.5 The results from only one
comprehensive preheater stack test are available; Figure 9.8.1-6 presents
scrubber collection efficiency as a function of coal feedrate and operation
temperature.
9.8.1.2 Sintering
9.8.1.2.1 Process description. Fine iron-ore particulate matter,
whether in natural or in concentrated ores, must be agglomerated to a size
and strength suitable for blast-furnace charging. Sintering is used to con-
vert iron ore fines and other iron-containing dusts into a product more
suitable for this purpose. The sintering process converts materials such as
fine ore concentrations, blast-furnace flue dust, mill scale, turnings, coke
fines, limestones fines, and miscellaneous fines into an agglomerated product
that is suitable for blast-furnace feed material. This is done by depositing
9.8-28
-------
96
95
94
93
« 92
91
90
'89.
88
87
86
85
84
83
82
81
80
o
o>
i.
a-s
z
LjJ
*—*
<_>
o
t—
O
a:
CO
CO
288 °C
82
95
COAL FEED RATE, Mg/h
109
Figure 9.8.1-6 Scrubber efficiency vs. coal feedrate parti-
culate data.5
9.8-29
-------
the mixture on a sinter machine traveling grate through which combustion air'
is drawn into a windbox. The mixture is ignited by natural gas or fuel oil
and burns to form a fused mass which is fed to a cooler, crushed, and then
screened in preparation for charging into a blast furnace.40 Average stack
emission parameters for sintering are presented in Table 9.8.1-9.41 Nation-
wide particulate emissions from sintering were estimated to be 95 Gg per year
in 1977.42
9.8.1.2.2 Emission characteristics and applicable control technology.
The sintering process can be a significant source of particulate emissions.
Uncontrolled particulate emissions from the sinter machine windbox are about
10 g/Kg of sinter produced.43 Particle size analysis conducted at the inlet
of a cyclone at one sinter plant indicated that about 13 percent of the
particulate was smaller than 10/xm. A summary of the particle size analysis
is shown in Figure 9.8.1-7.44,45,46 jne particle concentration for this un-
controlled emission stream was 2 g/Nm3 (1 gr/scf) or 3 g/kg (7 Ib/ton) strand
feed.46 Because the size of particulate from the sinter plant windbox is of
predominantly larger diameter, mechanical collectors are typically used for
product recovery and to reduce dust loads for the more efficient secondary
collectors. Secondary collectors which have been used to control sinter
plant emissions include wet and dry electrostatic precipitators, fabric
filters, scrubbers and gravel bed filters.
Dry Electrostatic Precipitators--The addition of dry ESPs downstream of
existing mechanical collectors reportedly increased overall dust removal ef-
ficiencies to over 97 percent.47 This is based on efficiencies reported to
the National Emission Data Systems by several sinter plants for control
systems using a mechanical collector as a primary control device and a dry
ESP as a secondary control device.
However, modern tendencies toward superfluxing have changed the ba-
sicity of the sinter which changes the electrical properties of the particu-
lates, causing them to have a higher resistivity. Basicity, or base-to-acid
ratio, is normally defined in the sintering industry as the ratio of the
"base" components in the feed (CaO and MgO) to the "acid" components in the
feed (SiOg and A^OS). The acid components are naturally present in iron
ores while most of the base components are added in the form of limestone
9.8-30
-------
00
co
Table 9.8.1-9. STACK PARAMETERS FOR SINTERING OPERATIONS4^
Nationwide number Average Average Average Average Average operating rate per
of facilities or stack ht., stack diam., temp., flow, facility or operation (Gg
Facility or operation operations m m °C (Am-Vmin) of sinter produced per yr)
Sinter machine 71 41 3 117 6900 790
operation ' „
-------
.10
9
8
7
6
5
4
cu
+•»
0)
o
o
to
UI
o
t—I
te
o.
o
1
0.9
0.8
0.7
0.6
S 0.5
<:
? 0.4
0.3
0.2
0.1
\ I I I I
I I I
] I
LT>
O O i— C\J LT>
O O O O, O i— CM
O
CM
o o o o o
CO «tf- LT> to 1^
o
oo
PERCENT BY HEIGHT SMALLER THAN INDICATED SIZE
Figure 9.8.1-7.
Particle sizing from inlets of windbox secondary control
devices.44 4S ^6
9.8-32
-------
and/or dolomite. In general, the efficiency of an ESP decreases as the
basicity of the sinter increases. It is difficult to predict the resistivity
of an individual sinter plant's particulate emissions. Some limited data
have been reported as follows:48
Parameter
Percent fluxing
material by weight
Resistivity of dust
225°F and 6% mois-
ture, ohm-cm
Approximate basicity
Sinter
Sample No. 1
-
9
1.5 x 109
1.0
Sinter
Sample No. 2
35
5 x 1013
3.5
Excessive sparking can occur when the dust reaches a resistivity of
about 2 x IQlO ohm-cm. When resistivity exceeds about 1QH ohm-cm, it
becomes very difficult to attain reasonable efficiencies.
Therefore, as a sinter plant moves toward making a superfluxed sinter
(basicity greater than 1.5), the resistivity of the windbox dust becomes
too high to maintain good ESP performance.
Resistivity is also affected by gas moisture content, various constit-
uents in the ores, and temperature. In general, however, the effects of
superfluxing far overshadow these other factors. The uncertainty regarding
this resistivity of the dust to be collected has resulted in the reluctance
of ESP manufacturers to state the efficiency of equipment proposed for sinter
plants. Changing from a normal sinter to a superfluxed sinter has reportedly
resulted in overall loss of ESP efficiencies in the range of 30 percent.
Average design performance parameters are shown in Table 9.8.1-10. The
9.8-33
-------
TABLE 9.8.1-10. AVERAGE DRY ESP DESIGN PERFORMANCE PARAMETERS FOR
COLLECTING SINTER MACHINE WINDBOX EXHAUST EMISSIONS49
Gas velocity
Inlet gas temperature
Electric field
Inlet dust loading
Precipitator power
1.3 m/s (4.3 fps)
118°C (245°F)
3.2 kV/cm (8.1 kV/in)
2.3 g/m3 (1.0 gr/acf)
28 watts/m3/m (71 watts/1000 cfm)
9.8-34
-------
size of the ESP is based on the collection efficiency desired, gas volume, '
and the precipitation rate parameter. The latter is dependent on the re-
sistivity of the dust.
EPA-contracted emissions tests have been conducted at one dry ESP con-
trolled sinter plant. This plant has four traveling-grate strands with a
total production capacity of 5440 Mg/day (6000 ton/day). Windbox exhausts
are controlled by multiclones at each strand followed by two ESPs. Each
ESP controls two strands. Tests were conducted on only one ESP. Outlet
concentrations for three samples taken by EPA Method 5 were 0.097, 0.046
and 0.064 g/Nm3 (0.042, 0.020 and 0.028 gr/scf), for a combined average of
0.069 g/Nm3 (0.030 gr/scf) or 0.1 g/kg (0.19 Ib/ton) strand feed.44 The
average total particulate load entering the precipitator was 0.069 g/Nm3
(0.30 gr/scf) or 0.83 g/kg (1.66 Ib/ton) strand feed. The efficiency of the
control device was calculated using the average of samples of the inlet and
outlet which were sampled concurrently. The average efficiency was 91.2
percent for the ESP at this plant.
Opacity readings of the windbox exhaust stack were taken during two of
the three particulate tests on the ESP outlet. The 6-minute averages ranged
from 0 to 20 percent.
The process was operating normally during all tests. Sinter basicity
during the tests ranged from 1.2 to 1.6. This is within the range normally
produced at the facility. The control system was also fully operational
during the tests.
Results of inlet particle size analysis after cyclone control for this
plant are summarized in Figure 9.8.1-7. The results of particle size analy-
sis after the ESP are summarized in Figure 9.8.1-8.44,45,50
Wet Electrostatic Precipitators—Wet electrostatic precipitators (ESPs)
have been demonstrated as capable of controlling windbox exhaust emissions.
A continuous water washing of the plates prevents an insulating layer of pol-
lutants from forming and nullifies the effect of the increased resistivity
of the particles caused by superfluxing the sinter.51 The gas stream is kept
saturated by continuous sprays and, as the particles and liquid droplets
collect on the plates, a film is formed which continuously drips off and
carries with it the collected dust, thereby preventing any buildup.52 The
9.8-35
-------
20...
PERCENT BY WEIGHT SMALLER THAN INDICATED SIZE
9.8.1-8. Particle sizing from windbox.secondary control outlets of sampled
facilities.44 45 5D~
9.8-36
-------
approximate water requirement for this system for present application is
0.67 - 1.2 L/m3 (5 to 9 gal/1000 scf).53 This requirement is comparable to
water usage for the high energy scrubbers applied to sinter plants.
The water sprays reduce the temperature of the inlet gases to 38 - 49°C
(100 - 120°F); therefore, the condensable hydrocarbon portion of the total
particulate emission is available for collection. Some hydrocarbons condense
into a fine mist and are collected on the plates. Because of their small
size and low dielectric constant, these hydrocarbon particles are not removed
as efficiently as dust particles with dielectric constants greater than 10.
A typical pressure drop across a wet ESP is approximately 0.13 kPa
(0.5 inches w.g.).51
As with other control devices, a wet ESP must be properly designed. A
wet ESP installed on a Canadian sinter plant was unsuccessful because the
inlet grain loadings were too high and the acidity of the recycled liquor
caused corrosion problems.53 The maximum inlet loading for proper operation
should be 0.57 g/Nm3 (0.25 gr/scf) which could be achieved by installing a
primary mechanical collector upstream.52 Also, the pH of the recycled water
should be monitored and a caustic solution injected to neutralize any
acidity.
Wet electrostatic precipitators have reported solid particulate removal
efficiencies in excess of 99.5 percent.51 One pilot test of a wet ESP in
series with a saturator (for cooling the gas stream) shows an overall ef-
ficiency of 95.7 percent for the removal of solid particulate.54
A full size module has been operated since July 1976 at one plant.55
The module handles 1370 ,Nm3/min (50,000 scfm) or about one-half the windbox
exhaust from a single sinter machine. After six months of continuous
operation, the company concluded that the technical feasibility of operating
the wet ESP had been demonstrated. Performance tests were conducted during
four different conditions of process operation. The baseline conditions
during 38 emission tests were full gas flow [1360 Nm3/min (49,600 dscfm)]
through the wet ESP and production of low basicity (0.85) sinter. Filterable
particulate emissions ranged from 0.007 to 0.05 g/Nm3 (0.003 to 0.22 gr/dscf)
with average emission of 0.023 g/Nm3 (0.010 gr/dscf). The sinter machine
produced 50,000 kg (55 tons) of sinter per hour with a total burden rate of
9.8-37
-------
94,000 kg (104 tons) per hour. Average filterable participate emissions were
0.045 g/kg (0.09 Ib/ton) of strand feed. The wet ESP was operated with a gas
flow rate of only 600 Nm^/min (22,000 dscfm) for three tests. Average fil-
terable particulate emissions dropped to only 0.009 g/Nm3 (0.004 gr/scf) or
0.035 g/kg (0.07 Ib/ton) of strand feed. A high basicity sinter (1.5) was
produced during ten tests with the wet ESP operating at 1200 Nm3/min (43,000
dscfm). Filterable particulate emissions ranged from 0.018 to 0.04 g/Nm3
(0.008 to 0.018 gr/dscf); the average was 0.03 g/Nm3 (0.014 gr/dscf). Sinter
production averaged 48,000 kg/hr (53 tons/hr) during these tests; therefore,
emissions averaged 0.06 g/kg (0.12 Ib/ton) of strand feed. The reason for
increased emissions with higher basicity sinter is not clear.
Baghouses—At least two sinter plants have successfully used baghouses
to control windbox emissions. The baghouses were selected rather than ESPs
because of the high resistivity of the sinter dust.
At other sintering facilities where baghouses have been considered as
a possible windbox control, there has been a reluctance to adopt fabric
filters. The reluctance is based on the concern that fabric filters would
experience abnormally high bag replacement rates and other problems due to
the condensation of moisture and condensible hydrocarbons on the bags. In
order to minimize bag condensation problems it is necessary to: (1) insulate
the baghouse and ductwork to limit the exhaust temperature drop to less than
10°C (50°F) under all ambient conditions, and (2) keep the oil content of the
exhaust gases low.56 This can be accomplished by careful selection of burden
materials, pretreatment of some materials or recirculation of the exhaust
gas.
Also, the oil content of the particulates captured must be kept low to
minimize heat buildup in the hoppers. Oil captured with the dust in the
baghouse hoppers can cause chemical reactions which create excessive temper-
atures.57 These high temperatures can cause partial sintering of the col-
lected materials and bridging can occur across the hopper outlet to prevent
the catch from being removed.57
Bag failures due to the abrasive nature of the dust or the possibility
of sparks igniting bags can be minimized by proper selection of bag material.
Adjacent bags can be damaged as the broken bags whip around. This problem
9.8-38
-------
can be minimized if damaged bags are repaired as quickly as possible or if
the baghouse is designed with greater distance between the bags.
The operating parameters for the two sinter plants at which baghouses
have been used successfully to control windbox emissions are shown in Table
9.8.1-11.58,59 Major considerations for baghouse design are the physical
properties and quantity of the particulates, and the temperature and volume
of the windbox exhaust.
Baghouses have the advantages of low energy requirements and low pres-
sure drop (i.e., 3.75 kPa (15 inches w.g.)]. Their efficiency is not affected
by the resistivity of the particulate, and no water pollution problems are
created.
A particulate removal efficiency of 99.9 percent was reported^ by one
plant for its baghouse which controls windbox emissions. The baghouse manu-
facturer guaranteed a minimum of 99 percent by weight particulate efficiency
during normal operation. Source tests conducted by the local county air
pollution control office showed the filterable particulate concentration to
be 23 mg/Nm3 (0.01 gr/scf). An EPA-contracted source test at this plant
using EPA Method 5 on the baghouse outlet shows an average concentration of
97 mg/Nm3 (0.042 gr/dscf) which is much higher than would be expected unless
some bags were leaking.^O Particle size results show over 20 percent of the
particulate was greater than 10 microns in diameter (see Figure 9.8.1-8).
However, this plant had low stack opacities during source testing. The 6-
minute opacity averages during the third test for particulates ranged from
0 to 1.8 percent. There were no opacity readings for the windbox stack
during the other two particulate tests because of overcast conditions.
This plant has two independent sinter strands with separate control
systems. Total plant production is normally 2720 Mg/day (3000 ton/day).
Only one sinter machine was tested. Windbox emissions are controlled by four
cyclones in parallel followed by a baghouse. The hood over the discharge
end of the strand and breaker extends back over the last windboxes so emis-
sions are drawn down into the windbox. Baghouse outlet concentrations were
69, 108, and 115 for an average of 97 mg/Nm3 (0.042 gr/dscf) or 160 nig/Kg
(0.32 Ib/ton) strand feed. Control efficiencies could not be determined
since samples were not taken of the inlet to the baghouse. The process, and
9.8-39
-------
TABLE 9,8.1-11. BAGHOUSE OPERATING PARAMETERS FOR COLLECTING
SINTER MACHINE WINDBOX EXHAUST EMISSIONS58.59
Parameter Plant A Plant B
Net air-to-cloth ratio 0.45 m/min (1.47 fpm) 0.41 m/min (1.36 fpm)
Gas volume 10194-11327 m3/mm (360.000-400,000 cfm) 12459 m3/min (440,000 cfm)
Moisture 6-10% by volume 6%
i° Particulate n.7-1.1 g/Nm3 (0.3-0.5 gr/scf)
00
_p* Gaseous sulfur 500 ppm
o
Pressure drop 3.25-3.5 kPa (13-14 inches w.g.)
-------
control equipment were operating normally during the test series. Sinter
basicities during the test week ranged from 1.9 to 2.1.
A baghouse was installed at another plant in September 1975 to control
windbox emissions.61 The single strand normally produces 3265 Mg/day (3600
tons/day). The windbox exhaust passes through multiclones and an ESP with
combined collection efficiencies of 85 percent prior to entering the bag-
house. Compliance tests run in February 1976 showed the average filterable
particulate concentration was 23 mg/Nm^ (0.01 gr/dscf). Mill scale is not
sintered at either of these plants and blast furnace flue dust is limited to
ten percent or less of the burden to avoid problems with condensable hydro-
carbons.
Scrubbers—Scrubbers have not been used frequently in the past due to
high energy requirements and problems associated with corrosion. Modern
sinter plants, however, are installing wet scrubbing systems for the windbox
exhausts, since the potential advantages of high collection efficiencies for
total particulates (solids and condensable hydrocarbons) appear to outweigh
operating problems.
Of all the devices used to control the windbox exhaust emissions, high
energy scrubbers require the greatest energy input. To effectively remove
total particulates, a pressure drop of 10 kPa (40 inches w.g.) or greater is
required.62 This necessitates the addition of more powerful fan motors
to overcome the added resistance of the system. Average scrubber design
performance parameters are shown in Table 9.8.1-12.63
The corrosive nature of the scrubber liquid and the high abrasiveness
of the particulate matter requires the use of special materials. Water
treatment with lime or soda ash is also necessary if the effluent is to be
recycled. Untreated scrubber liquor may typically contain up to 1600 ppm
chlorides and up to 520 ppm total sulfur and have a pH of 2.1 to 3.1.62,64
The low pH is caused mainly by the absorbed exhaust gases. Absorbed sulfur
dioxide, fluorides, chlorides, carbon dioxide, carbon monoxide, and even
oxides of nitrogen can react with water to form acids which result in the
low pH found in untreated scrubber effluent.
Because of the low pH, the effluent stream must be treated. In most
9.8-41
-------
TABLE 9.8.1-12. AVERAGE SCRUBBER DESIGN PERFORMANCE PARAMETERS FOR
COLLECTING SINTER MACHINE kflNDBOX EXHAUST EMISSIONS63
Liquid-to-gas ratio (L/G) 1.2 L/M3 (9 gal/1000 acf)
Pressure drop 11.25 kPa (45 inches water gauge)
Liquid recycle ratio 0.95
Throat velocity 137 m/sec (450 ft/sec)
9.8-42
-------
cases, water treatment facilities exist for treatment of other process ef-
fluents associated with steel-making processes. A typical recycle ratio for
a scrubber system installed on a sinter plant is 0.95 with scrubber bleed
ranging up to 1900 1pm (500 gpm).
Unlike dry ESPs, scrubbers are not affected by the resistivity of the
particulate collected. In fact, sinters with a higher basicity will tend,
to minimize corrosion problems. Re-entrainment of captured particulate in
the air is also not a problem since it is handled as a sludge. Scrubbers
cool the exhaust gases and condense hydrocarbons which would pass through
other types of dry collectors in vapor form.
Emission tests have been conducted at another plant that has one
traveling-grate sinter strand with a normal production of 2600 Mg/day (2880
ton/day). Windbox emissions are controlled by four cyclones in parallel
followed by a single venturi scrubber. The total particulate outlet data
for the first test was 76 mg/Nm3 (0.033 gr/dscf) or 80 mg/kg (0.16 Ib/ton)
strand feed. Subsequent to this test, the scrubber was modified to achieve
an increase in pressure drop across the venturi to approximately 13 kPa.
Total particulate concentrations for the next three tests were 71, 53 and 62
for a combined average of 62 mg/Nm3 (0.027 gr/dscf) or 90 mg/kg (0.18 Ib/ton)
strand feed.65 The inlet emission rate was 1.1 g/kg (2.2 Ib/ton) strand
feed.
Two newer sinter plants have been equipped with recirculating water
scrubbers for controlling windbox particulates.66»67 preliminary tests at
one of these plants showed filterable particulates averaged 62 mg/Nm3
(0.27 gr/dscf) or about 100 mg/kg (0.2 Ib/tori) strand feed. Filterable
particulates were 44 mg/Nm3 (0.019 gr/dscf) at the other plant and 50 mg/kg
(0.1 Ib/ton) of strand feed.
Typical scrubber guarantees are for an outlet dust loading of 46 mg/Nm3
(0.02 gr/scf) when based on the dry catch. One vendor reported that a scrubber
guarantee of 92 mg/Nm3 (0.04 gr/scf) could possibly be given when both solids
and condensable hydrocarbons are considered.68
Gravel Bed Filters—Gravel bed filters have been used to clean windbox
exhaust, also. The filtration media consists of several layers of sized grit.
9.8-43
-------
As the gas flows through the layers the impingement action causes the
particles to agglomerate and be caught on the grit and in the spaces in the
media. The filter is cleaned by b.ackflushing with clean air while raking the
media. The clean air is heated to avoid condensation of moisture which
could cause problems with the media. Precautions must also be taken to
avoid condensing hydrocarbons which could coat the media and plug the voids.
9.8.1.2.3 Sinter processing equipment control systems. Baghouse instal-
lations can capture over 99 percent of the particulate collected by the
equipment exhaust hoods. Baghouse vendors usually guarantee 99 percent
efficiency for their equipment under normal operating conditions and with
proper maintenance programs.
In developing emission standards for sinter plants, the U.S. EPA tested
a plant which controlled sinter processing operations. At this plant, one
baghouse controls emissions from the discharge hoods, breakers, and hot
screens for all four strands. The baghouse has two stacks both of which
were tested concurrently. The average grain loadings from the outlet of
this plant's two baghouse stacks were 0.046 g/Nm^ (0.02 gr/scf).^^ More
than 50 percent of the particulates were over 10 microns in diameter, while
less than four percent were under one micron (see Figure 9.8.1-9). These
results could .have been caused by leakage through damaged bags. The opacity
six-minute averages of visible emission readings never exceeded one percent,
however.
Baghouses have recently been installed at other plants to control
emissions from the discharge hoods, breakers, hot and cold screens, and con-
veyor transfer points. Average filterable particulate concentrations were
0.01 g/Nm3 (0.005 gr/dscf) at one plant69 and 0.02 g/Nm3 (0.009 gr/dscf)
at another pi ant.66
Visible emissions were observed from the discharge hoods, breakers, hot
and cold screens, conveyor transfer points and cooler at a number of
plants.70'71>72 Visible emissions averaged less than 10 percent for all
such sources observed at these plants.
Emissions averaging 10 percent opacity were observed from the point of
loading and discharge from the cooler at one plant, but there were r,^ visible
9.8-44
-------
00
10.0
9.0
8.0
7.0
6.0
5.0
4.0
3.0
Ol
E
§
M
Q_
CJ
o
o
a;
2.0
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
o
•
CO
I 1 I I I I I I I
i i i r
i i i
i i
tn
O i— CVJ
...
CD O CD
tn
•
CD
i— CSJ
o
o
CM
CD
CO
CD CD
^- in
CD
CO
CD
CT(
PERCENT BY WEIGHT SMALLER THAN INDICATED SIZE
Figure 9.8.1-9.
Particle sizing from baghouse outlet controlling sinter
processing emissions at a sintering plant.60
9.8-45
-------
emissions at a second plant. At a third plant, this point plus the first
quarter of the circular cooler were hooded and exhausted to mechanical col-
lectors. This hood could be seen while the discharge hood and hot screens
were observed, and no visible emissions were seen. The cooling air exiting
the top of the cooler was observed for three hours at the second plant and
no visible emissions were seen.
9.8.1.3 Blast Furnace. Pig iron or molten iron for steel-producing
furnaces is obtained by the reduction of iron ore, Fe203 (hematite) or
Fe3d4 (magnetite), to iron in the blast furnace. The iron ore is contained
in the iron ore pellets and in the sinter material. A diagram of a typical
blast furnace is shown in Figure 9.8.1-10.73
The blast furnace operations are continuous, and the furnace charge
consists of alternate layers of iron ore, coke, and limestone. The furnace
operates at about 1540°C.
The iron ore descends down the furnace and is reduced and melted by
the countercurrent flow of hot reducing gases created by the partial
combustion of coke. Hot metal from the furnace is tapped into torpedo cars
and weighed on the hot metal track scale. After the metal is transferred
to a charging ladle, a crane transports it to the steel-making vessel.
Molten slag is removed from the furnace through separate tapping holes
which are at a higher elevation than the molten-iron tap hole.74 Average
stack emission parameters for iron production blast furnaces are presented
in Table 9.8.1-13. Nationwide .particulate emissions from blast furnaces
were estimated to be 55 x 103 Mg per year in 1977.42 The particle size
distribution of the dust collected from a blast furnace is shown in Table
9.8.1-14. Table 9.8.1-15 presents chemical composition data for dry blast-
furnace flue dust.
Particulate emissions from blast furnaces tend to be minimized since a
high degree of particulate emission control is necessary to keep the stores
(heat exchangers) from plugging. About 83 kg of particulate matter per Mg
of product is emitted.43
Particulates of molten iron, from the blast furnace are also emitted
during each tap, and these emissions enter the atmosphere by passing through
9.8-46
-------
Figure 9.8.1-10. Typical blast furnace.77
9.8-47
-------
CO
CO
Table 9.8.1-13. STACK PARAMETERS FOR BLAST FURNACE IRON PRODUCTION OPERATIONS^
Nationwide number Average Average Average Average
of facilities or stack fit., stack diam., temp., flow
Facility or operation operations m m °C AmVmin
Average operating rate per
facility or operation (Gg
of sinter produced per yr)
Ore charged furnace
Agglomerate charged
furnace
70
68
57
61
3.3
2.5
353
285
3238
4650
610
571
-------
Table 9.8.1-14.
SIZE ANALYSIS OF FLUE DUST U.S. BLAST
FURNACESa,74
Size, microns
Range, %
683
589
414
295
208
147
104
74
2
3
7
10
10
10
7
5
.5
.9
.0
.7
.0
.2
.7
.3
to
to
to
to
to
to
to
to
20
10
11
12
15
16
12
8
.2
.6 '
.7
.4
.0
.8
.5
.8
aDust collected in participate
control devices.
Table 9;8.1-15.
CHEMICAL COMPOSITION BY DRY, BLAST-FURNACE
FLUE DUST74
Compound
Weight,
Iron
Ferrous oxide
Silicon dioxide
Aluminium oxide
Calcium oxide
Zinc oxide
Phosphorus
Sulfur
Manganese
Carbon
3.6.5 to 50.3
Not available
8.9 to 13.4
2.2 to 5.3
3.8 to 4.5
Not available
0.1 to 0.2
0.2 to 0.4
0.5 to 0.9
3.7 to 13.9
9.8-49
-------
the sides and roof of the cast house.
About 6 Mg of gases are evolved for every ton of iron produced from the
blast furnace. These flue gases contain dust concentrations of 27.5 g/m^
and leave the furnace at temperatures of 180 to 280°C.75 The flow rate of
the gases is a function of the coke feed rate. Total gas volume increases
linearly with an increase in the coke feed rate.75
Various types of control equipment are presently used to control par-
ticulate emissions from blast furnaces. Dry cyclones, wet scrubbers, and
electrostatic precipitators are common. Venturi scrubbers or electrostatic
precipitators are reported to clean blast furnace flue gas to particulate
concentrations of 0.23 grams per cubic meter.7^
The relationship between clean-gas dust loading and pressure drop
across a venturi scrubber is shown in Figure 9.8.1-11.73 As seen from the
figure, low output dust loadings can be achieved at higher pressure drops.
Venturi scrubbers of fixed- or variable-orifice types can be employed on
blast furnaces.
9.8.1.4 Open Hearth Furnace. In the open hearth furnace, steel is
produced from a mixture of scrap and hot metal in varying proportions. The
feed, consisting of limestone, light and heavy scrap, and (normally) molten
pig iron, is charged to the open hearth or reverberatory furnace and heated.
Cold pig iron is used when molten pig iron is not available. Impurities
such as carbon, manganese, silicon, sulfur, and phosphorous are reduced to
specified levels by oxidation. The refining operation produces a slag mate-
rial that forms a continuous liquid layer on the surface of the liquid metal
and combines with many of the iron impurities. The oxidized forms of chro-
mium, vanadium, aluminum, titanium,'tungsten, columbium, and zinc may be
present in the slag.40 Some carbon in the iron charge is converted to
carbon monoxide and boils off the molten bath. Calcination of limestone in
the bath produces carbon dioxide which also boils off the molten bath. In
order to shorten the heating time, oxygen may also be injected into the
bath.
When the molten steel bath is determined to be at the grade of steel
required, the molten contents of the furnace are tapped through the tap
9.8-50
-------
o
CM
Q.
O
UJ
o;
oo
UJ
100
10
0.001
\ \
I I I
J I
0.01
0.1
I I I
1.0
CLEAN GAS DUST LOADING, grains per standard cubic foot
Figure 9.8.1-11 The relationship between clean-gas dust loading
and pressure drop across venturi scrubber.73
9.8-51
-------
hole into the steel ladles. The alloying, recarbonizing and deoxidizing
materials are added to the molten metal before the slag starts flowing.
The molten metal is poured into ingot molds for the particular final product
(see Figure 9.8.1-12).
The furnace is a rectangular, shallow basin enclosed by refractory-
bricked walls and roof. Heat is supplied from burners at one end. The fuel
used includes coke oven or natural gas, oil, tar, or pitch. The flame trav-
els the length of the furnace above the charge which rests on the hearth.76
The open hearth furnace is regenerative, i.e., uses the heat removed
from the exhaust gases to preheat the combustion air in two regenerator or
checker work sections. Each section serves alternately as the exhaust
passage for the hot products of combustion and the inlet passage for
combustion air and fuel.
From the checker sections, the gases are normally directed to a waste
heat boiler, where the temperature is further lowered to an average of 260
to 315°C (500 to 600°F).77 Open hearth furnace capacities vary from as
little as 27 Mg to as much as 454 Mg per heat (i.e., batch of finished
steel).^O The median is between 91 and 181 Mg per heat. The time required
to produce a heat is commonly between 8 and 12 hours when normal amounts of
oxygen are used.78 Average stack emission parameters for open hearth fur-
naces are shown in Table 9.8.1-16. Table 9.8.1-17 provides an indication of
the size distribution of uncontrolled emissions from open hearth steel pro-
duction during oxygen lancing. Currently, this is the only data available
from the FPEIS data base that is considered suitable for publication. Other
data are available upon request. Nationwide particulate emissions from open
hearth furnaces were estimated to be 33 x 103 Mg per year in 1977.42
9.8.1.4.1 Emissions and controls. Emissions from open hearth furnaces
depend upon the furnace operation and type of charge in each furnace and may
vary during the cycle and from cycle to cycle. Oxygen lancing, used to
decarbonize the metal, generates considerably more particulate matter.
Usually up to 90 percent of the particulates from open hearth furnaces are
iron oxides, predominantly Fe203.40
Particle size distribution varies considerably with the heat. For a
furnace that is not oxygen-lanced, about 50 percent of the particulates
9.8-52
-------
Furnace Hearth
Figure 9.8.1-12. Cross-section of a basic open hearth furnace.73
9.8-53
-------
Table 9.8.1-16. STACK PARAMETERS FOR OPEN HEARTH STEEL PRODUCTION FURNACES^!
00
1
en
Nationwide number
of facilities or
Facility or operation operations
Oxygen lance 83
No oxygen lance 71
Average
stack ht.,
m
49
45
Average
stack diam.,
m
2.9
2.2
Average
temp. ,
°C
277
411
Average
flow,
Amvrain
9775
2941
Average operating rate per
facility or operation (Gg
of sinter produced per yr)
254
78
-------
Table 9.8.1-17. SIZE SPECIFIC EMISSIONS FROM OPEN HEARTH STEEL PRODUCTIONS
Mass concentration,
mg/DNCM (mass percent less than stated size)
Total 15.3 yum 12.9 ftm 10.1/x.m 7.28/xm 5/xm 2.5/u,m 1.01/u,m
°° Uncontrolled 8580 3580/41.8 3440/40.1 3350/39 3300/38.5 3270/38.2 3270/38.2 3270/38.2
tn
tn
aDuring oxlance, control device type not available.
-------
are less than 5 micrometers. The particle size distribution for an oxygen-
lanced furnace is reported as follows:77
Percent by weight Size (micrometer)
69 percent 10
45 percent 5
20 percent 2
The small size of the particulate matter emitted from open hearth fur-
naces necessitates the use of high-efficiency collection equipment such as
venturi scrubbers and electrostatic precipitators. Baghouses have also
been installed for particulate emissions control, but they require that the
gases be precooled.
A major problem with the efficiency of electrostatic precipitators on
the open hearth furnaces is the open hearth process itself. The problem
stems from the variation in the properties of emissions from the open hearth
furnace during operation. During a period of heat, the moisture content of
the bases may drop below the normal value of 18 to 20 percent, with a
resultant increase in resistivity and drop in precipitator efficiency.77
The relationship between the collection efficiency and size of a precipi-
tator is shown in Figure 9.8.1-13. The figure shows that removing the dust
from 150 m^/sec (315,000 cfm) of open hearth waste gas required 5,400 m2
(58,300 ft2) of collecting surface for an efficiency of 95 percent. An
Increase in the collection to 99.3 percent can be obtained by increasing
the collecting surface area to 9,000 m2 (96,500 ft2).
Wet scrubbers have been used in some cases, when the shop either had
no waste heat boilers or when the existing boilers could not lower gas
temperatures enough to warrant installation of electrostatic precipitators.
The relationship between the outlet grain loading and pressure drop is
illustrated in Figure 9.8.1-14.77
9.8.1.5 Basic Oxygen Furnace (EOF). The Basic Oxygen Process (BOP),
which uses the BOF, is being employed increasingly to produce steel because
of its high production rates, simplicity, and efficiency of operation. The
9.8-56
-------
CJ
I—I
LI-
LI-
too
90
80
70
60
50
40
30
20
10
0.5
COLLECTING SURFACE, ft'
Note: ft2x 0.093 = M*
166
Figure 9.8.1-13
Relationship of electrostatic precipitator
collecting surface to collection efficiency
for open hearth emissions (8920 Am3/min).77
9.8-57
-------
cCl
Oi
in
Q
to
en
0.10
Q08
Q06
004
002
a.
I—
=>
o
QOI
0008
Ore and lime boil
and working period
Charging, melt down__//
and hot metal
2628303334363840
PRESSURE DROP, In H0
Note: grains/SCF x 2.29 = g/SM , and inches x 2.54 = cm
Figure 9.8.1-14 Relationship between clean-gas dust loading and pressure
drop for a wet scrubber on an open hearth furnace (oxygen-
lancing used during the refining period).77
9.8-58
-------
BOF is a pear-shaped structure with a refractory lining (see Figure 9.8.1-15).
A water-cooled lance is used to supply pure oxygen (typical ratio of 62 m3/Mg~
steel produced) to a mixture of steel scrap, hot metal, and flux materials.40
The oxygen reacts exothermically with the carbon in the metal. No other fuel
is used, and this reaction provides enough heat so that steel scrap can be
melted to form about 30 percent of the charge. Furnace capacities range from
68 to 295 Mg per heat, and the time required per cycle is very short. Typi-
cal 136 Mg BOF operations take the following times:75
. •" Charge scrap 1 minute
Charge hot metal 2 minutes
Oxygen blow 20 minutes
Chemical tests 5 minutes
Tapping time 5 minutes
Total time 33 minutes
Molten iron from the blast furnace is brought to the basic oxygen fur-
nace shop in railroad submarine ladle cars, and steel scrap is brought in
by both rail and truck.
The BOF has displaced the open hearth as the major steel production
process, but it is much less flexible because only a limited amount of scrap
(25 to 30 percent) can be used in the charge.
Average stack emission parameters for BOFs are presented in Table
9.8.1-18. Nationwide particulate emissions from BOFs were estimated to
be 18 x 103 Mg per year in 1977.42
9.8.1.5.1 Emissions and controls. Particulate emissions from BOFs range
from 4.6 g/m3 to 22.9 g/m3.43 j^g effluent gas may have a temperature of
290° to 1660°C (560° to 3000°F) depending on utilization of a waste heat
boiler. About 85 to 95 percent of these particulates are less than 1 micro-
meter in size.^O Transfer of hot metal to the BOF creates additional par-
ticulate emission problems. During the transfer, known as reladling, sig-
nificant amounts of kish (precipitated graphite flakes) and iron oxide
particulates are generated. Fugitive emissions occur primarily during
charging, tapping, and slagging. During charging, the emissions occur when
the primary hood is removed. Emissions from slagging and metal tapping are
a result of violent mixing of the molten metal. Roof ventilation and canopy
ventilation collection systems can control such emissions.
9.8-59
-------
TYPICAL BASIC OXYGEN FCE
(BOFJ
Figure 9.8.1-15 Basic oxygen furnace.73
9.8-60
-------
Table 9.8.1-18. STACK PARAMETERS FOR BASIC OXYGEN STEEL PRODUCTION FURNACES**
Nationwide number Average Average Average Average Average operating rate per
of facilities or stack ht., stack diam., temp., flow, facility or Operation (Gg
Facility or operation operations m m °C Am3/min of sinter produced per yr)
10 Basic oxygen furnace 57 49 3.4 148 10,561 1.189
JJQ operation
i • . ;
' ' '
-------
The particle size distribution of the emissions from a BOF are reported
as follows:
Percent by weight Size micrometer
33 149
54 74
84 10
97 5
Venturi scrubbers and electrostatic .precipitators are commonly used to
control particulate emissions from BOFs. The emissions are usually routed
through either an open or a closed hood. The closed hood is designed to
minimize infiltration air and, thus, reduce the exhaust gas volume as well
as reclaim carbon monoxide. However, the high concentration of combustible
carbon monoxide often limits the choice of cleaning equipment to a high-
energy venturi scrubber because of safety considerations related to possi-
ble carbon monoxide caused explosions.79
Variability in gas flows, and the moisture content and temperature of
the entering gas are the design considerations used in selecting an electro-
static precipitator. Electrostatic precipitators are selected over venturi
or wet scrubbers when there is a water-treatment problem in a plant. Prob-
lems can include inadequate water-treatment facilities or lack of sufficient
water. The problems relating to improper performance of wet scrubbers can
be attributed to a lack of material to withstand the abrasive and corrosive
nature of the dust-laden water, or a misapplication of construction mate-
rials.73
The particulate emissions from hot metal reladling are a mixture of
kish and submicrometer, iron oxide fumes. Cyclones and baghouses are the
only control equipment installations applied on a large scale.78 More
extensive data on control of fugitive particulates from hot metal reladling,
skimming and furnace operations are given in Reference 80.
9.8.1.6 Electric Arc Furnace. Steel scrap (and sometimes blast-furnace
produced iron) and flux materials are charged into the electric arc furnace
(EAF). The EAF is a cylindrical vessel with a refractory lining and three,
large carbon electrodes protruding through the roof. The heat required for
a metallurgical reaction is generated by the arc from three electrodes.
9.8-62
-------
Currents ranging from 10,000 to 20,000 amps are used.40 MOSt modern arc
furnaces for steelmaking are top-charged with molten blast furnace metal
(when available), light and heavy scrap, alloying materials, and fluxes.
The electrodes are moved out of the way while the furnace is being charged.
The roof is returned to close the furnace, and electrodes are lowered to
about 25 centimeters above the charge.40 AS current is applied through the
electrodes, the charge is melted (see Figure 9.8.1-16). During this period,
phosphorus, silicon, manganese, carbon, and other materials are oxidized and
generally combine with the slag that is carefully controlled throughout the
operation. Oxygen lancing is often used to increase production rates.
At the end of the process, the electrodes are raised, and the steel is
tapped from the furnace. In tapping, depending on whether the furnace is
fixed or tilting, the tap hole is either opened or tilted so that the steel
is tapped from the furnace into a ladle. Average stack emission parameters
for EAFs are presented in Table 9.8.1-19. Nationwide particulate emissions
from EAFs were estimated to be 20 x 10^ Mg per year in 1977.42
9.8.1.6.1 Emissions and controls. Particulate emissions from electric
furnaces consist primarily of oxides of iron, manganese, aluminum, calcium,
magnesium, and silicon. The typical composition of these emissions is pre-
sented in Table 9.8.1-20. Most emissions occur during the early "melting"
operation, although significant quantities are also emitted during charging,
tapping, and oxygen-blowing operations. Particulate emissions from electric
arc furnaces vary from cycle to cycle and are affected by contaminants in the
scrap steel, the amount of oxygen blown, and the amount of electrical power
used in the furnace. The size distribution of the particulate emanating
from electric arc furnace operation is as follows:40
Percent by weight Particle size (micrometer)
10 0 to 2
17 2 to 4
34 4 to 8
9 8 to 12
15 12 to 44
15 less than 44
9.8-63
-------
CARBON ELECTRODES
PORT FOR THIRD ELECTRODE
TAPPING SPOUT
SLAG SPOUT
Figure 9.8.1-16 Direct-arc electric furnace.73
9.8-64
-------
Table 9.8.1-19. STACK PARAMETERS FOR ELECTRIC ARC STEEL PRODUCTION FURNACES^l
IO
CO
Facility or operation
With oxygen lance
No oxygen lance
Nationwide number
of facilities or
operations
103
47
Average
stack fit.,
m
18.9
22.6
Average
stack diam.,
m
2.4
3.1
Average
temp.,
°C
73
122
Average
flow,
AnF/min
3723
3954
Average operating rate per
facility or operation (Gg
of sinter produced per yr)
173
89
O1
-------
Table 9.8.1-20. TYPICAL CHEMICAL ANALYSIS OF
ELECTRIC ARC FURNACE
Compound
Total iron
Loss of ignition
CaO
Si02
Total carbon
MgO
A1203
Cr203
MnO
V205
Total S
Na20
Ti02
F
K20
BaO
Pb
Weight,
%
32.2
10.1
18.6
10.8
8.3
7.1
5.7
1.3
1.2
1
0.60
0.59
0.50
0.49
0.37
0.20
0.15
9.8-66
-------
The particle-laden gases generated by electric arc furnaces can be
collected by any of several types of collection systems that direct the
emissions to particle-removal devices such as baghouses, scrubbers, etc.
Figure 9.8.1-17 shows schematics of four types of collection or ventila-
tion systems.81,82 jhe canopy hood system is not as efficient as the other
three systems shown in Figure 9.8.1-17. However, unlike the three other
systems, it can collect emissions during furnace charging and tapping.
For this reason, the canopy hood system is appropriate for use with one of
the other three systems. It is also possible to have a collection system
consisting of collection ducts in the roof of the electric arc shop build-
ing which collects air from the entire building. Such a system also col-
lects large volumes of air.
Fabric filters or baghouses are the most commonly used devices for
cleaning electric furnace gases, although venturi scrubbers and electrostatic
precipitators have been installed at some plants.40 The use of baghouses
on electric arc furnaces necessitates precooling of gases to protect the
bags. Water spray towers, radiant coolers, and dilution with ambient air
are commonly used to reduce the inlet temperature of the gases to the bag-
house. More extensive data on control of particulates from electric arc
furnaces are given in Reference 83.
The fugitive emissions resulting from charging operations are difficult
to control because the collection or ventilation system must be removed to
charge the furnace. Fugitive emissions can also be significant during
slag removal and metal tapping as a result of violent mixing of the molten
material. Canopy hoods and roof ventilation control systems are effective
controls. Particulate collected in the emission control systems is sometimes
recycled to the sintering operation. (Much of it is not recyclable because
of zinc contamination.) Water pollution can be a problem when a wet scrubber
is used.
9.8.1.7 Rolling, Shaping, and Finishing. Steel ingots are heated in
a soaking pit furnace to prepare them for hot working (rolling). In the
furnace, steel is heated until it is plastic enough for rolling to the
desired shape. After the ingots are rolled into billets, blooms, or slabs,
they are cooled and inspected. Surface defects are removed by grinding,
chipping, peeling, or scarfing (see Section 9.8.1.8 for discussion of
9.8-67
-------
Fourth hole
Side draft
Combination hood
Canopy hood
Figure 9.8.1-17 Ventilation systems for electric arc furnaces.81 82
9.8-68
-------
scarfing). Reheating furnaces are used for raising the temperature of the
slabs, blooms, or billets for rolling into sheets, coils, or other shapes.74
Acid treatment (called pickling) is used to clean the oxidized surface, hot-
rolled steel in preparation for cold rolling.
The finishing operation consists primarily of treating the semifinished
steel products by hot or cold working and processing to the final form. Some of
the sheet and tin finishing operations are tempering, tin plating, galvanizing,
chrome plating, coating, polishing, and continuous annealing. Average stack
emission parameters for finishing operations are presented in Table 9.8.1-21.
9.8.1.7.1 Emissions and controls. Particulate emissions from rolling
and shaping operations are considered negligible, unless the pits are fired
with fuel other than gas. Grinding and chipping operations generate particu-
lates which are confined to the buildings.
9.8.1.8 Scarfing. Slabs of steel, blooms, and billets are processed
to remove defects that could be detrimental to the finished product. This
can be done by chipping, grinding, or scarfing, Of these processes, scarfing
can produce the largest amount of emissions. Before the scarfing operation
takes place, the slag, bloom, or billet is heated to the rolling temperature.
The process consists essentially of burning the surface of the steel with a
jet of oxygen in combination with a fuel gas, such .as acetylene or natural
gas. The purpose of the fuel gas is to ensure that the steel is heated to a
sufficiently high surface temperature (about 870°C) to bring about rapid
oxidation and to localize melting of a thin layer of metal.74 Approximately
3 mm (1/8 inch) of metal is removed from all four sides of the red-hot bil-
lets, blooms, or slags as they travel through the machine in a manner similar
to the path through rolling mills.40 Average stack emission parameters
for scarfing are presented in Table 9.8.1-22. Nationwide particulate emis-
sions from scarfing were estimated to be 6 x 106 Mg per year in 197.7.42
9.8.1.8.1 Emissions and controls. Little information is available on
the pollutants generated by the scarfing process. Emissions of particulates
range from 0.9 to 1.8 g/m3.74
Scrubbers, cyclones, baghouses, and wet and dry electrostatic precipi-
tators are, used to control emissions from scarfing installations in the U.S.
Fugitive emissions from scarfing occur from leaks in the machine, the
scarfer's control equipment, and from open (outdoor) hand scarfing.84
9.8-69
-------
Table 9.8.1-21. STACK PARAMETERS FOR STEEL PRODUCTION FINISHING OPERATIONS*!
Facility or operation
Pickling
Soaking pits
Grinding, etc.
Nationwide number Average
of facilities or stack ht.,
operations m
13
123
12
15.2
36.9
14.9
Average
stack diam.
m
1.6
1.7
1.0
Average
, temp. ,
°C
41
384
120
Average Average operating rate per
flow, facility or operation (Gg
AmS/min of sinter produced per yr)
2762
4147
18158
143
122
68
-------
Table 9.8.1-22. STACK PARAMETERS FOR SCARFING45
Facility or operation
Nationwide number
of facilities or
operations
Average Average Average Average
stack ht., stack diam., temp., flow,
m m °C Am-Vmin
Average operating rate per
facility or operation (Gg
of sinter produced per year)
Scarfing operation
26
27
1.8
58
3850
1205
CO
I
-------
REFERENCES FOR SECTION 9.8.1
1. Bohn, R., T. Cosano, and C. Cowherd. Fugitive Emissions From Inte-
grated Iron and Steel Plants. U.S. Environmental Protection Agency,
Office of Research and Development. Washington, DC. Publication
No. EPA-600/2-78-050. March 1978.
2. Draft of Standards Support and Environmental Impact Statement. Volume
I: Proposed National Emission Standards By-Product Coke Oven Wet-Coal
Charging and Topside Leaks. U.S. Environmental Protection Agency.
Research Triangle Park, NC. June 1978.
3. Study of Coke Oven Battery Stack Emission Control Technology. Final
Report. Volume II: Control Methods. Midwest Research Institute.
Kansas City, MO. March 1979.
4. OAQPS Data File of Nationwide Emissions. U.S. Environmental Protection
Agency, Office of Air Quality Planning and Standards. Research Triangle
Park, NC. February 1979.
5. Kemner, D., D. Loudin, J. Smith, and G. Saunders. Control of Emissions
from Dry Coal Charging at Coke Oven Batteries. PEDCo Environmental,
Inc. Cincinnati, OH. October 1978.
6. Bratina, J. E. Fabric Filter Applications on Coke Oven Pushing Opera-
tions. Journal of the Air Pollution Control Association. Pittsburgh,
PA. 29:9, September 1979.
7. Study of Coke Oven Battery Stack Emission Control Technology. Final
Report. Volume I: Collection and Analyses of Existing Emission Data.
Midwest Research Institute. Kansas City, MO. March 1979.
8. McClelland, R. 0. Coke Oven Smokeless Pushing System Design Manual.
U.S. Environmental Protection Agency. Publication No. EPA-650/2-74-076.
September 1974.
9. Engineering Analysis of Emission Controls for Wet Quench Towers. Final
Report. Midwest Research Institute. Kansas City, MO. January 1979.
10. Atmospheric Modeling Data from National Emission Data System (NEDS).
U.S. Environmental Protection Agency, Office of Air Quality Planning
and Standards. Research Triangle Park, NC. May 1979.
11. Preliminary Draft, Coke Oven Battery Stacks, Background Information for
Proposed Standards. U.S. Environmental Protection Agency. Research
Triangle Park, NC. May 1980.
12. Test Report. CF&I, Battery D. York Research Corp. Denver, CO. CF&I
Steel Corp. July 24, 1978.
13. Test Report. CF&I, Battery D. Clayton Environmental Consultants, Inc.
Southfield, MI. For U.S. Environmental Protection Agency. Based on
testing done in August 1979.
9.8-72
-------
14. Source Emission Survey for Lone Star Plant - Coke Oven Electrostatic
Precipitator. Ecology Audits, Inc. Dallas, TX. For Lone Star Steel.
February 1973.
15. Sampling and Analysis of Particulate Emissions from Inlet and Outlet of
Coke Plant, East and West Precipitators. Armco, Inc. Houston, TX.
November 1976.
16. Test Report. National Steel-Granite City, Battery C. Clayton Environ-
mental Consultants, Inc. Southfield, MI. For U.S. Environmental
Protection Agency. Based on testing done in July 1979.
17. Test Report. Kaiser Steel-Fontana Battery B. TRW, Inc. Redondo Beach,
CA. For U.S. Environmental Protection Agency. Based on testing done
in September 1979.
18. Compliance Test Report. Kaiser Steel, Battery B. From James Good,
Kaiser Steel, to U.S. Environmental Protection Agency, Region IX,
Enforcement Division. San Francisco, CA. August 8, 1979.
19. Compliance Test Report. Kaiser Steel, Battery C. From James Good,
Kaiser Steel, to U.S. Environmental Protection Agency, Region IX,
Enforcement Division. San Francisco, CA. April 2, 1979.
20. Compliance Test Report. Kaiser Steel, Battery D. From James Good,
Kaiser Steel, to U.S. Environmental Protection Agency, Region IX,
Enforcement Divison. San Francisco, CA. December 21, 1978.
21. Compliance Test Report. Kaiser Steel, Battery E. For U.S. Environ-
mental Protection Agency, Region IX, Enforcement Division, San
Francisco, CA. February 1979.
22. Coke Oven Emissions, Miscellaneous Emissions, and Their Control at
Kaiser Steel Corporation's Fontana Steel Making Facility. State of
California Air Resources Board. Sacramento, CA. November 1976.
23. Phelps, R. G. AISI-EPA-Battelle Coke Oven Door Sealing Program. Jour-
nal of the Air Pollution Control Association. Pittsburgh, PA. 29:9.
September 1979.
24. Oliver, J. F., and J. T. Lane. Control of Visible Emissions at CF&I's
Coke Plant. Pueblo, Colorado. Journal of the Air Pollution Control
Association. Pittsburgh, PA. _29_:9> September 1979.
25. Draft report, Study of Coke-Side Coke Oven Emissions. Clayton Environ-
mental Consultants, Inc. For the U.S. Environmental Protection Agency.
Contract No. 68-02-1408, Task No. 14, Volume 1. January 16, 1976.
26. Gas Cleaning Tests on Coke Pushing Emissions Captured in the Coke-Side
Shed at Plant L. Appendix III to a letter from Edward Roe, Plant L, to
Donald Goodwin, U.S. Environmental Protection Agency. Research Triangle
Park, NC. April 14, 1975.
27. Letter from Edward Roe, Plant L, to Don Goodwin, U.S. Environmental Pro-
tection Agency. Research Triangle Park, NC. April 14, 1975.
9.8-73
-------
28. Letter from David Anderson, Corporate Office for Plant T to Don Goodwin,
U.S. Environmental Protection Agency. Research Triangle Park, NC.
April 18, 1975.
29. Draft report, Source Testing of a Stationary Coke Side Enclosure. Clay-
ton Environmental Consultants, Inc. For the U.S. Environmental Protec-
tion Agency, Contract No. 68-02-1408, Task No. 10, Volume 1. November 5,
1975.
30. Letter from David Anderson, Corporate Office for Plant T to Andrew
Trenholm, U.S. Environmental Protection Agency. Research Triangle
Park, NC. February 9, 1976.
31. Symons, C. Plant T memorandum titled, "Gas Cleaning Requirements for
Coke-Pushing Emissions," to Andrew Trenholm, U.S. Environmental Protec-
tion Agency. Research Triangle Park, NC. January 17, 1975.
32. "Report of Emissions Tests at Plant P Coke Oven Battery C Coke Side
Shed," by Clean Air Engineering, Inc. Palatine, IL. April 30, 1975.
33. Trenholm, A., U. S. Environmental Protection Agency. Research Triangle
Park, NC, memorandum titled, "Meeting with the American Iron and Steel
Institute." May 10, 1976.
34. Letter from John Brough, Plant P, to Andrew Trenholm, U.S. Environmental
Protection Agency. Research Triangle Park, NC. November 13, 1975.
35. Draft report, Emission Testing and Evaluation of Koppers Smokeless Coke
Pushing System. Clayton Environmental Consultants, Inc. For the U.S.
Environmental Protection Agency. Contract No. 68-02-0630. December
15, 1975.
36. Letter from William Smith, Plant V, to Andrew Trenholm, U.S. Environmen-
tal Protection Agency. Research Triangle Park, NC. December 18, 1975.
37. Letter from William Smith, Plant V, to Andrew Trenholm, U.S. Environmen-
tal Protection Agency. Research Triangle Park, NC. April 14, 1976.
38. "Coke Oven Emissions - Letter Report on IITRI Project No. C6333, Task 1,"
memorandum from IIT Research Institute to Kirk Foster, U.S. Environ-
mental Protection Agency. Research Triangle Park, NC. November 5,
1975.
39. Ertel, G. L. Quench Tower Particulate Emissions. Journal of the Air
Pollution Control Association. Pittsburgh, PA. j?9:9, September 1979.
40. Operation and Maintenance of Particulate Control Devices on Selected
Steel and Ferroalloy Processes. U.S. Environmental Protection Agency.
Research Triangle Park, NC. Publication No. EPA-600/2-76-037. March
1978.
41. Op. cit. Reference 10.
42. OAQPS Data File of Nationwide Emissions. U.S. Environmental Protec-
tion Agency, Office of Air Quality Planning and Standards. Research
Triangle Park, NC. 1977.
9.8-74
-------
43. Compilation of Emission Factors. U.S. Environmental Protection Agency.
Publication No. AP-42. August 1977.
44. Test Report of Sinter Plant Emissions at Bethlehem Steel Corporation,
Bethlehem Plant, Bethlehem, PA. York Research Corporation, Stamford,
CT. Report No. 75-SIN-l. December 1975.
45. Particle Size Analysis, Granite City Steel, Granite City, Illinois.
York Research Corporation. Stamford, CT. For the U.S. Environmental
Protection Agency, Emission Measurement Branch. Task No. 19, Contract
No. 68-02-1401. August 1975.
46. Emission Testing Report. Colorado Fuel & Iron. Pueblo, CO. U. S. En-
vironmental Protection Agency. EMB Report No. 75-SIN-5. February 1976.
47. National Emission Data System, Point Source Listing - Sintering. U.S.
Environmental Protection Agency file. Research Triangle Park, NC.
June 27, 1974.
48. Oglesby, S., and G. B. Nichols. A Manual of Electrostatic Precipitator
Technology, Part II - Application Areas. Prepared by Southern Research
Institute for the National Air Pollutin Control Administration.
Cincinnati, OH. Contract No. CPA 22-69-73. August 25, 1970. p. 445.
49. Reference 48, p. 499.
50. Particle Size Analysis, Kaiser Steel, Fontana, CA. York Research Cor-
poration. Stamford, CT. For the U.S. Environmental Protection Agency,
Emission Measurement Branch. Task No. 22, Contract No. 68-02-1401,
Project No. 75-SIN-3. August 1975.
51. Bakke, E. The Application of Wet Electrostatic Precipitators for Con-
trol of Fine Particulate Matter. Mikro Pul Division, United States
Filter Corporation. (Presented at the Symposium on Control of Fine
Particulate Emissions from Industrial Sources. San Francisco, CA.
January 15-18, 1974.)
52. Personal communication, memo to file on phone conversation between
Jack Roehr, Western Precipitator, and Fred Hall, PEDCo, Cincinnati, OH.
- January 29, 1975.
53. Personal communication, memo to file on phone conversation between
William Greg, Mikro Pul Divison, United States Filter Corporation, and
Donald J. Henz, PEDCo, Cincinnati, OH. January 29, 1975.
54. Precipitator Efficiency Tests - Mikropul Pilot Wet Precipitator Sinter
Plant Main Exhaust Jones & Laughlin Steel Corporation, Aliquippa, PA.
February 4-6, 1974. WFI Sciences Company. Pittsburgh, PA. March 7,
1974.
55. Mazer, M. R., S. T. Herman, S. A. Jaasund. Wet Electrostatic Precipita-
tor Demonstration for Sinter Plant Emissions Control. Bethlehem Steel
Corporation. Bethlehem, PA. December 28, 1976.
9.8-75
-------
56. Nowak, T. T. Sinter Plant Baghouse. Kaiser Steel. (AIME, 31st Iron
Making Conference Proceedings, Chicago, April 10-12, 1972.) p. 14.
57. Rounds, G. L. and G. Geminder. Problems of Recycling Waste Oxides
through the Sinter Strand (Baghouse Experience). Kaiser Steel Corpora-
tion. (Presented at the Minerals and the Environment Symposium, London,
England, June 1974). p. 23.
58. Reference 56, p. 4.
59. Woodard, K. R. "Trip Report to Inland Steel Company Sinter Plant, East
Chicago, Indiana", memorandum to J. V. Crowder, Emission Standards and
Engineering Division, U.S. Environmental Protection Agency. Research
Triangle Park, NC. August 19, 1976.
60. Emissions Source Test From a Baghouse Serving an Iron and Steel Sinter-
ing Plant at Kaiser Steel Corporation, Fontana, CA. Pacific Environmental
Services, Inc. Santa Monica, CA. For the U.S. Environmental Protection
Agency. Task Order No. 6, Contract 68-02-1405, PES Contract 098, Project
No. 75-SIN-3. November 1975.
61. Report of Emissions Tests for Inland Steel Company at the Sinter Plant
Main Stack, Baghouse, East Chicago, IN. Clean Air Engineering, Inc.
Palatine, IL. P.O. No. HX-95861. February 1976.
62. Ball, D. F., et. al. Environmental Control in Iron Ore Sintering.
(Presented at the Minerals and the Environment Symposium, London,
England, June 1974). p. 23.
63. Anonymous. Steel Mill Sinter Plant Control Device Permit Applictions.
64. Harris, E. R. and F. R. Beiser. Cleaning Sinter Plant Gas with Venturi
Scrubber. Journal of the Air Pollution Control Association 24^:10. October
1974.
65. Air Pollution Emission Test, Sinter Plant, Granite City Steel Division,
National Steel Corporation, Granite City, IL. Clayton Environmental
Consultants, Inc. Southfield, MI. U.S. Environmental Protection
Agency Contract No. 68-02-1408. EMB Report No. 75-SIN-4.
66. Sintering Plant Environmental Control Sintering Machine Windbox Gas
Cleaning System, Burns Harbor Plant, IN. System Description, Bethel em
Steel Corporation. Arthur G. McKee & Company. Cleveland, OH. Reference
C4110. Bethlehem Reference No. 327-37000. May 1974.
67* Application for Permit to Construct Gas Cleaning or Emission Control
Equipment at Sparrows Point Plant, Bethlehem Steel Corporation.
Maryland State Department of Health and Mental Hygiene, Environmental
Health Administration, Bureau of Air Quality Control. Baltimore, MD.
March 2, 1973.
68. Personal communication, memo to file on phone conversation between
Richard Schwartz and Fred Hall, PEDCo. Cincinnati, OH. January 31,
1975.
9.8-76
-------
69. Report of Emissions Tests for Inland Steel Company at Plant 2, The No. 3
Sinter Plant, East Chicago, IL. Clean Air Engineering, Inc. Palatine,
IL. P.O. No. KX-86730. July 1975.
70. Woodard, K. R. Visible Emission Observations at CF&I Steel Corporation
on January 20, 1976. A trip report to Mr. J. C. Berry, Emission Stan-
dards and Engineering Division, U.S. Environmental Protection Agency.
Research Triangle Park, NC. January 30, 1976.
71. Woodard, K. R. "Trip Report to Bethlehem Steel, Sparrows Point Sinter
Plant", memorandum to J. U. Crowder, Emission Standards and Engineering
Division, U.S. Environmental Protection Agency. Research Triangle Park,
NC. August 19, 1976.
72. Woodard, K. R. "Trip Report to Bethelehem Steel, Burns Harbor Sinter
Plant", memorandum to J. U. Crowder, Emission Standards and Engineering
Division, U.S. Environmental Protection Agency. Research Triangle Park,
NC. August 18, 1976.
73. Varga, J., Jr., and H. W. Lownie. A System Analysis Study of the Inte-
grated Iron and Steel Industry. Battelle Memorial Institute. Columbus,
OH. May 1969.
74. Industrial Process Profiles for Environmental Use. Chapter 24: The
Iron and Steel Industry. U.S. Environmental Protection Agency. Publi-
cation No. EPA-600/2-77-023x. February 1977.
75. Labee, C. J. Steel Making at Weirton. Iron and Steel Engineer.
Pittsburgh, PA. 46_(10):W1. October 1969.
76. The Making, Shaping and Treating of Steel,'Ninth Edition. U.S. Steel
Corporation. Pittsburgh, PA. December 1970.
77. Vandegrift, A. E., et al. Particulate Pollutant System Study, Volume
III. Midwest Research Institute. Kansas City, MO. U.S. Environmental
Protection Agency Contract No. 22-69-104. May 1971.
78. Background Information for Establishment of National Standards of Per-
formance of New Sources — Iron and Steel Industry, Draft Report.
Environmental Engineering, Inc. March 1971.
79. Background Information for Proposed New Source Performance Standards:
Iron' and Steel Plants, Volume 1. U.S. Environmental Protection Agency.
Publication No. APTD 1352A. June 1973.
80. Preliminary Draft, Revised Standards for Basic Oxygen Process Furnaces -
Background Information for Proposed Standards. U.S. Environmental
Protection Agency, Office of Air Quality Planning and Standards.
Research Triangle Park, NC. November 1980.
81. Air Pollutant Control Techniques for Electric Arc Furnaces in the Iron
and Steel Foundry Industry. U.S. Environmental Protection Agency.
Publication No. EPA-450/2-78-024. June 1978.
9.8-77
-------
82. Inspection Manual for the Enforcement of New Source Performance Stan-
dards: Steel Producing Electric Arc Furnaces. U.S. Environmental
Protection Agency. Publication No. EPA-340/1-77-007. May 1977.
83. Background Information for Standards of Performance: Electric Arc Fur-
naces in the Steel Industry. Volume 1: Proposed Standards. U.S.
Environmental Protection Agency. Research Triangle Park, NC.
EPA-450/2-74-017a. October 1974.
84. Bonn, R., et al. Fugitive Emissions from Integrated Iron and Steel
Plants. Midwest Research Institute. Kansas City, MO. U.S. Environ-
mental Protection Agency. Publication No. EPA-600/2-78-050. March
1978.
9.8-78
-------
9.8.2 Ferroalloy Production
A ferroalloy is an alloy of iron and one or more other elements such as
silicon and manganese. Ferroalloys are used for deoxidizing molten steels
and for making alloy steels. They are produced primarily by using electric
submerged arc furnaces and, to a limited extent, by metallothermic processes.
More than 75 percent of all ferroalloys produced (e.g., manganese, silicon,
and chromium ferroalloys) are produced in electric smelting furnaces (Figure
9.8.2-1).
Ferrosilicon is the major ferroalloy produced. It can be manufactured
in either a blast furnace or an electric arc furnace. The submerged arc is
more efficient in the reduction of oxides and is thus used instead of the
blast furnace.
Ferroalloys are made by reduction of suitable oxides in the, electric
arc furnace. For example, in making ferrochromium the charge may consist of
chrome ore, limestone, quartz (silica), coal and wood chips, along with
scrap iron. In this case, the silica and lime form a slag. For ferrosilicon,
the charge would consist mainly of iron scrap, silica, and coke. In every
case, carbon monoxide is formed copiously and escapes through the pores and
channels in the charge. The escaping gas carries large quantities of parti-
culates, which are essentially submicrometer in size (50 percent less than
0.1 micrometer) and present a notoriously difficult control problem.
Both open and sealed submerged electric arc furnaces are used. Figures
9.8.2-2 and 9.8.2-3 present schematics of such furnaces. Average stack emis-
sion parameters for ferroalloy production are presented in Table 9.8.2-1.1
The open furnaces are more prevalent; they have large gas flows, and the
furnace gas is diluted to a level that makes the exit gas temperature low
enough for fabric filtration.
Nationwide particulate emissions from ferroalloy production in 1977
have been estimated for ferroalloy material handling and for production
of the various alloys as follows:?
Ferroalloy material handling 18 x 103 Mg per year
Fe-Si alloy production 13 x 103 Mg per year
Si-Mn alloy production 2 x 103 Mg per year
9.8-79
-------
REFRACTORY
LINING
SHELL
CRUCIBLE
TAP HOLE
Figure 9.8.2-1 Submerged arc furnace for ferroalloy production.
9.8-80
-------
ELECTRODES
EXTENDING
/ I f:f
ID
00
I
CD
TAP
HOLE
I
I
-CD
THROUGH
HOOD
J I II I I I I
HOOD
.••'DUST;'
'
'
MIX FEED
CHUTE
(TYPICAL)
>FINDUCED AIR
1 1 II T 1 1
I T I I 1^ I I I I II
VENTURI
WATER
TO STACK
Figure 9.8.2-2 Open furnace controlled by a venturi scrubber.
-------
MIX FEED
(TYPICAL)
V\
ELECTRODES
PC)
ELECTRODE
SEAL
TAP
HOLE
WATER
FLARE
STACK
Figure 9.8.2-3 Sealed furnace controlled by venturi scrubber.
9.8-82
-------
Table 9.8.2-1. STACK PARAMETERS FOR FERROALLOY PRODUCTION!
00
I
CO
GO
Nationwide number Average
of facilities or stack ht.,
Facility or operation operations m
Open furnace:
50 percent iron silicon
75 percent iron silicon
90 percent iron silicon
Silicon metal
Silicomanganese
Screening
Ore dryer
Low carbon Cr reactor
Semi-sealed furnace:
Ferromanganese
Oeneral
26
15
2
13
13
7
5
1
15
31
24.4
31.7
15.2
89.0
23.2
12.2
12.5
33.8
24.7
108
Average
stack diam.
m
2.0
5.4
1.7
5.6
3.9
0.9
7.9
8.0
2.4
8.7
Average
temp.,
°r,
124
146
107
170
91
28
93
64
69
190
Average Average operating rate per
flow, facility or operation, Gg
AnvVmin of sinter produced per year
2243
4979
5806
3945
1501
574
2315
1642
1845
109
26
41
7
10
25
65
43
21
25 .
154
-------
Fe-Mn alloy production (electric furnace) 2 x 103 Mg per year
Fe-Mn alloy production (blast furnace) 1 x 103 Mg per year
Si metal alloy production 7 x 103 Mg per year
Other ferroalloys production 12 x 103 Mg per year
9.8.2.1 Emissions and Controls. Table 9.8.2-2 presents emission fac-
tors for ferroalloy production by the submerged arc electric furnace.3 As
shown in the table, uncontrolled furnaces generate large amounts of highly
concentrated particulate pollution.
Table 9.8.2-3 presents particle size distributions for the particulates
emitted from various ferroalloy manufacturing furnaces.4,5 The data of
Reference 4 are supported by a significant number of size measurements.
Table 9.8.2-4 shows the existing New Source Performance Standards for
the ferroalloy industry. The May 20, 1976, revision consisted of only five
minor corrections and wording adjustments. The main points of Table 9.8.2-4
are: (1) no gases shall escape the capture system at the tap and be visible
for more than 40 percent of the tapping period; (2) when a blowing tap
occurs, visible emissions are not limited; and (3) the standards require
that carbon monoxide emissions of 20 volume percent or more be flared or
otherwise burned.
Dust concentrations in the untreated furnace gas from covered and open
furnaces differ considerably. Use of an open submerged arc furnace requires
treatment of large volumes of hot gas—up to 300 m3/sec (500,000 acfm), at
temperatures up to 650°C (1200°F). These large volumes of gas and fumes
formed from the smelting process consist of carbon monoxide and evaporated
metallic oxides which rise through the charge bed to the surface of the
charge. At the surface, the gas is burned with oxygen from the air, and
very small'particles of oxide fumes are farmed.
Volumes of gases are considerably smaller from a covered electric fur-
nace. In a covered furnace, unburned carbon monoxide gas is collected under
the roof of the cover and withdrawn from the furnace without combustion.
This results in a gas volume which may be as little as 3 percent of that
from an. open furnace.6
The average size of fume particles from open furnaces is reported to
be less than 0.5 micrometers.6 Except for the larger dust particles of feed
9.8-84
-------
Table 9.8.2-2. EMISSION FACTORS FOR FERROALLOY PRODUCTION
IN ELECTRIC SMELTING FURNACESa»3
Type of furnace and product Participates
Ib/ton kg/MT
Open Furnace
50% FeSi 200 100
75% FeSi 315 157.5
90% FeSi 565 282.5
Silicon metal 625 312.5
Silicomanganese 195 97.5
Semi-covered furnace
Ferromanganese 45 22.5
aEtnission factors expressed as units per unit weight of specified product
produced.
9.8-85
-------
Table 9.8.2-3. PARTICLE SIZE DISTRIBUTION OF PARTICULATES FROM FERROALLOY OPERATIONS^
CO
I
00
CT1
Source operation
Fraction within particle size
7-3 Jim
Electric furnace
Ferrosilicon alloys
Ferromanganese alloys
Ferrochromium alloys
Miscellaneous ferroalloys
0
0
0
0
.012
.096
.140
.095
3-1 urn
0.120
0.475
0.290
0.295
1-0.5 urn
0
0
0
0
.180
.285
.185
.230
0.5-0.1 /Jt,m
0
0
0
0
.520 .
.1343
.240
.315
range
0.1-0.05 jum
0
0
0
0
.110
.0007
.025
.026
0.05-0.01 fjim
0
0
0
0
.050
.010
.009
-------
Table 9.8.2-4. STANDARDS OF PERFORMANCE FOR NEW SOURCES
FERROALLOY PRODUCTION FACILITIES
Source category
Affected facility
Pollutant
Emission level
Monitoring
requirement
00
I
00
Ferroalloy production
facilities
Proposed
10/21/74 (39 FR 37469)a
Promulgated
5/4/76 (41 FR 18497)
Revised
5/20/76 (41 FR 20659).
Electric submerged
arc furnaces
Particulate 0.99 Ib/Mw-hr (0.45
kg/Mw-hr)("high silicon
alloys") 0.51 Ib/Mw-hr
(0.23 kg/Mw-hr)(chrome
and manganese alloys)
No visible emissions may
escape furnace capture
system
No visible emissions may
escape tapping system
for >40% of each
tapping period
No requirement
Flowrate
monitoring
in hood
Flowrate
monitoring
in hood
Dust handling
equipment
Opacity
CO
Opacity
15%
20% volume basis
10%
Continuous
No requirement
No requirement
aVolume 39 of Federal Register^ page 37469,
-------
mix carried from the furnace, the particle fume size is generally below
2 micrometers.
The fugitive emissions occur from raw material handling and
preparation and are estimated to be less than 0.25 percent of the material
processed.6
Several methods are used to control emissions from electric submerged
arc furnaces. Emissions from open industrial furnaces are controlled by
wet scrubbers, fabric filters, and electrostatic precipitators. Emissions
from covered furnaces, however, are controlled almost entirely by wet
scrubbers (with only one known use of fabric filters)?, primarily because
of high gas temperature and safety hazard of handling carbon monoxide.
High-energy wet scrubbers on open furnaces producing silicomanganese,
high-carbon (HC) ferrochromium, and ferrochrome-silicon have been demon-
strated to achieve 96 to 99 percent efficiency with a pressure drop of
about 14 kPa (150 cm water gauge).6 However, large volumes of gas from
open furnaces and a high-pressure drop across the scrubber require high
energy. In some cases, the power requirements may be as high as 10 percent
of the power supplied to a furnace.
High energy venturi scrubber systems used on covered furnaces
producing ferromanganese, silicomanganese, and 50 percent ferrosilicon
were found to remove up to 99.9 percent of the particulates from the
collected reaction gases.6
Baghouses on open ferroalloy furnaces producing silicomanganese,
ferrochrome-silicon, and silicon metal have shown collection efficiencies
of about 99 percent.6 The ratio of air-to-cloth is about 1.5:1 to 2:1,
thus requiring substantial cloth area relative to the volume of gas to be
cleaned. Exhaust gas temperatures in excess of 200°C (500°F) require
preceding of the gas before cleaning.
Unequal baglife on treating silica fumes necessitates frequent bag
replacement; the bags last only from 18 months to 2 years. Dust collected
in the dry state is usually wetted before disposal.
Electrostatic precipitators have also been installed on some large
open furnaces producing ferrochromium and ferrochrome-silicon. The
overall collection efficiency (including tapping emissions) on the HC
ferrochrome furnace was more than 98 percent.6 Preconditioning the gases
9.8-88
-------
before they enter the precipitator increases precipitator efficiency.
Table 9.8.2-5 gives some of the performance data from actual source
tests of various control devices.8 EPA Method 5 was used; a high pressure
drop was often required for the venturi scrubbers at the listed performances,
9.8-89
-------
Table 9.8.2-5. PERFORMANCE OF PARTICIPATE CONTROLS OH FERROALLOY FURNACES*.**
Plant
A2
B
K
LI
«3 M
00
i
<£> M
o n
0
Q
R
P
Type of
furnace
Sealed
Seal ed
Sealed
Open
Open
Open
Open
Open
Sealed
Semi-
Power, I'M
Product Design
FeMn 30
SiHn 27
50* FeSi 42
SiHn
Si 17
SiMn
FeSi 75% 36
FeCrSi
FeCrSi 26
50% FeSi
Operating
27
23
31
7.5
7.2
22
20
18.6
46.5
enclosed
S
T
Sealed
Open
HCb FeCr
HC FeCr 40
18
33
AP, Gas
Control device Inches HgO T, °F
2 parallel
3-stage
Venturis
Venturi 50
2 series venturi -> ,
mist eliminators
-> wet FSP
2 parallel 57
Venturis
3 parallel 200
haghouses
Aeronetics
scrubber
Baghouse 379
Raghouse 172
2 Venturis ->
wet cyclone
2 parallel 75
Venturis
Baghouse -> 145
flare
ESP
Particulatc
emissions
Ib/MWhr
0.024
0.009
0.0026
0.36
0.61
1.5
1.02
0.44
0.0093
0.078
0.036
0.65
g/OSCF
0.016
0.010
0.0018
0.009
0.0020
0.078
0.017
0.0027
0.0107
0.058
0.032
0.016
Remarks
water-
activated
venturi
AP venturi 1 = 12"
AP venturi 2 = 24"
0.61 Ib/MW hr 0
47" H20 and
1.60 0 37"
AP Venturis,
8" and 80"
Gas condition-
ing, partial
collection of
tap fumes
3EPA Method 5 testing.
bHigh carbon = HC.
-------
REFERENCES FOR SECTION 9.8.2
1. National Emission Data System, Data File, SCC Codes. U.S. Environ-
mental Protection Agency. Research Triangle Park, NC.
2. OAQPS Data File of Nationwide Emissions. U.S. Environmental Protection
Agency, Office of Air Quality Planning and Standards. Research Triangle
Park, NC. 1977.
3. Compilation of Emission Factors. U.S. Environmental Protection Agency.
Publication No. AP-42. August 1977.
4. Particulate Pollutant System Study, Volume II: Fine Particle Emissions.
NAPCA Contract No. CPA-22-69-104. Midwest Research Institute. Kansas
City, MO. pp 231-335. August 1971 (updated 1973).
5. Weast, T. E., et al. Fine Particulate Emission Inventory and Control
Survey. U.S. Environmental Protection Agency. Research Triangle Park,
NC. Publication No. EPA-450/3-74-040. January 1974.
6. Engineering and Cost Study of the Ferroalloy Industry. U.S. Environmental
Protection Agency. Publication No. EPA-450/2-75-008. May 1, 1974.
7. Background Information for Standards of Performance: Electric Submerged
Arc Furnaces Producing Ferroalloys. Volume 3: Supplemental Information.
U.S. Environmental Protection Agency. Publication No. EPA-450/2-74-018c.
April 1976.
8. Background Information for Standards of Performance: Electric Submerged
Arc Furnaces Producing Ferroalloys. Volume 2: Test Data Summary. U.S.
Environmental Protection Agency. Publication No. EPA-450/2-74-018b.
1974.
9.8-91
-------
9.8.3 Gray Iron Foundries
Gray iron foundries produce iron castings. The production process
essentially involves the furnace melting of scrap iron to produce an appro-
priate molten iron alloy which is then poured into sand molds to produce
castings. Gray iron foundries depend primarily upon the cupola furnace for
the economical production of large quantities of gray iron. Electric fur-
naces and small reverbatory furnaces are also used in some foundries.
Nationwide particulate emissions in 1977 from gray iron foundry furnaces and
nonmelting operations were estimated to be 87 x 1C)3 Mg per year.l
9.8.3.1 Cupola
9.8.3.1.1 Description. The cupola is the oldest and still the most
universally used furnace for the production of gray iron. If not adequately
controlled, a cupola becomes a large contributor to air contaminants.
The cupola is a firebrick-lined, vertical, cylindrical steel shell,
approximately 68 cm to 270 cm (27 inches to 108 inches) in diameter, sup-
ported on structural steel legs.2 Air is supplied through a windbox and
tuyeres by either positive displacement blowers, centrifugal blowers, or
fan type blowers. Figure 9.8.3-1 presents a schematic of a cupola.3 As in
blast furnaces, air is the largest single raw material of the heat; the air's
weight depends mainly on the size of the furnace.
Cupola preparation for melting consists of securing the bottom drop
door, placing a layer of sand over the door to prevent heat damage, closing
the tap and slag holes, and charging coke for the bed. The bed is ignited
and allowed to burn through; then the charges of coke, flux limestone or
soda ash, and iron are placed in alternate layers up to the charge door.
The blast is turned on, and melting begins. Charging continues until the
desired tonnage of iron has been melted, after which the air is shut off,
and the furnace bottom is dropped to allow the remaining charges to fall on
the foundry floor. This material is recharged during the next operating
cycle.3
9.8.3.1.2 Emissions and controls. Gray iron cupolas emit dust, fumes,
smoke, gases up to 16 percent carbon monoxide, and oil vapors. Oil mist and
smoke in cupola exhaust gases are due largely to oil and grease contamination
of the scrap metal charges to the furnace. Factors affecting particulate
9.8-92
-------
Skip-ham roil
(lof2)
Stick lining—•'
Coil iron lining
Chorgmg floor-^".
Cnorging,
Wins boi
for iron
((lag holt n ISC'
eppetue)
Sand
C*fit>tnli«nol eupolo
Figure 9.8.3-1 Conventional lined cupola.3
9.8-93
-------
emissions from gray iron cupola foundries include the following:^
o Furnace design
o Charging practice
o Quantity and quality of the charge
o Quantity of coke used
o Melting zone temperature
o Volume and rate of combustion air
o Use of techniques such as oxygen enrichment and fuel injection
Table 9.8.3-1 provides data from the Fine Particulate Emissions Informa-
tion System (FPEIS) that are the most suitable for this publication.5 These
data represent an emission test of a gray iron cupola controlled with a ven-
turi scrubber having a pressure drop of from 10 to 23 kPa (40 to 94 inches wg),
The control techniques employed to reduce particulate emissions from cupolas
vary widely. The most commonly used were wet caps, which can be placed
directly on top of cupola stacks and therefore require no gas-conducting
pipes or induced draft blowers. However, due to their low collection effi-
ciencies, the particulates are not adequately controlled to meet existing
or new regulations.
Cyclonic dust collectors remove 70 to 80 percent of the particulate
matter from the gas stream, depending on the particle size range.4
Wet scrubbers or high energy venturi scrubbers are capable of removing
95 percent of the particulate emissions from cupolas. Variable throat ven-
turi scrubbers are especially useful for cupola operations because their
pressure drop can be adjusted to achieve a desired efficiency.
Fabric filters on cupolas necessitate cooling of the gas prior to
entrance to the baghouse. Radiant or water spray cooling or both are
employed for this purpose. The fabric filter installations achieve par-
ticulate removal efficiencies as high as 99 percent.4
9.8.3.2 Electric Furnaces. Electric arc furnaces are used in the
melting of ferrous scrap to produce molten gray iron or steel. An electric
furnace consists of a refractory-lined, cup-shaped steel shell with a
refractory-lined roof through which three graphite electrodes are inserted.
Scrap is charged to the furnace and melted; alloying elements and fluxes are
added as needed.
9.8-94
-------
Table 9.8.3-1. SIZE SPECIFIC EMISSIONS FROM A GRAY IRON CUPOLAS
!£>
•
00
1
ID
CJ1
Control Control device
Process device operating
and sections parameter
fuel type (in series) • Total
Cupola Prespray Ap Uncontrolled 2000
55 section 10-23 kPa
Venturi Controlled 39.57
rod
Collector Efficiency 98
and cooling
section
Mass concentration, mg/DNCM
(mass percent less than stated size)
15.3 urn
1780/87.9
35.2/88.7
98
12.9/im
1780/87.9
34.8/87.7
98
10.1 fiH\
1770/87.8
34.2/86.2
98
7 .28 ju.ni
1770/87.7
33.2/83.7
98.12
5 urn
1765/87.35
32.1/80.06
98.18
2.b /tm
1730/85.4
29.2/73.6
98.31
1 .01 fim
1360/67.3
23.6/59.4
98.26
-------
The high temperatures of the arcs attained while melting the scrap
produce dense fumes consisting of iron and other metal oxides plus organic
particulates from oil and other contaminants in the scrap.
Electric induction furnaces are being used increasingly by the foundry
industry for melting scrap iron. An electric inductive furnace consists of
a drum-shaped vessel that converts electrical energy into heat by setting
up a magnetic field when the primary coil of the transformer is energized.
Alternating current is passed through a primary coil with a solid iron
core. The molten iron is contained within a loop that surrounds the pri-
mary coil and acts as a secondary coil. The current flowing through the
primary coil induces a current in the loop, and the electrical resistance
of the molten metal creates the heat for melting. The heated metal circu-
lates to the main furnace chamber and is replaced by cooler metal.
Induction furnace shops feature a preheater in which all water and most
of the oil are removed from the scrap to preclude the possibility of an ex-
plosion hazard in the furnace. Emissions generated during preheating are
controlled. Additional emissions from induction furnaces are generated
during melting, backcharging, and tapping. Recently installed induction
furnaces also control melting emissions. Backcharging and tapping emis-
sions have not yet been successfully controlled.
9.8.3.2.1 Emissions and controls. Particulate matter, fumes, smoke,
and metal oxides (formed when a vaporized metal contacts air) are emitted
by electric furnaces. Emissions are captured using either roof or side
draft hoods and, in the case of steel furnaces, by direct furnace evacuation.
Baghouses are used to reduce emissions from electric arc furnaces at an
efficiency of 95 to 99 percent.6 Source tests by EPA at two foundries with
fabric filter controlled electric arc furnaces included inlet and outlet par-
ticulate measurements.6 Emissions at the new foundry showed a reduction from
715.4 mg/DNCM (3.42 kg/Mg charge) to 8.92 mg/DNCM (0.047 kg/Mg charge). Emis-
sions at the other foundry with retrofitted control equipment showed a reduction
from 886.3 mg/DNCM (6.27 kg/Mg charged) to 10.6 mg/DNCM (0.054 kg/Mg charged).
Recently installed induction furnaces are equipped with particulate
collection devices of the fabric filter types.
9.8-96
-------
REFERENCES FOR SECTION 9.8.3
1. OAQPS Data File of Nationwide Emissions. U.S. Environmental Protection
Agency, Office of Air Quality Planning and Standards. Research Triangle
Park, NC. 1977.
2. Field Operations and Enforcement Manual for Air Pollution Control.
Volume II: Inspection Procedures for Specific Industries. Pacific Envi-
ronmental Services, Santa Monica, CA. U.S. Environmental Protection
Agency Contract No. CPA-70-122. August 1972.
3. Systems Analysis of Emissions and Emissions Control in the Iron Foundry
Industry, Volume I. A. T. Kearney, Inc. Chicago, IL. U.S. Environmen-
tal Protection Agency Contract No. EPA-22-69-106. February 1971.
4. Controlled and Uncontrolled Emission Rates and Applicable Limitations
for Eighty Processes. U.S. Environmental Protection Agency. Publication
No. EPA-450/3-77-016. September 1976.
5. Calvert, S., et al. FPEIS Test Series 55. U.S Environmental Protection
Agency. Research Triangle Park, NC. Publication No. EPA-600/2-76-282.
December 1976.
6. Air Pollutant Control Techniques for Electric Arc Furnaces in the Iron
and Steel Foundry Industry. U.S. Environmental Protection Agency.
Publication No. EPA-450/2-78-024. June 1978.
9.8-97
-------
9.8.4 Steel Foundries
9.8.4.1 Process. The charge material in steel foundries is ferrous
scrap. Steel foundries produce castings as a finished product by melting
the scrap and pouring the hot metal into molds. Steel castings are made
for manufacturers of heavy industrial machinery such as bulldozer frames,
locomotive wheels, and farm machinery. Direct electric arc, induction, and
open hearth are common furnaces used in steel foundries. The melting opera-
tion involves furnace charging, melting, tapping the furnace into a ladle,
and pouring the steel into molds.
9.8.4.2 Emissions and Controls. While emissions from steel foundries
are similar to those from gray iron foundries, there are some differences.
The oxygen lance, when used, temporarily produces a large gas volume accom-
panied by increased particulate emissions and substantially large amounts of
carbon monoxide. The particle size distribution of the flue gas from three
gray iron electric arc furnaces is given in Table 9.8.4-1.1 The dust from
steel-producing furnaces .falls into the same range, with 80-85% of the par-
ticles below 5 micrometers in diameter.
Emissions are captured using roof hoods, side draft hoods, or direct
furnace evacuation for electric arc furnaces. Direct evacuation is accom-
plished under slightly negative pressure in the furnace. It is the most
effective method for collecting melting emissions and also results in the
lowest gas volumes. Unlike roof or side draft hoods, direct evacuation
requires greater cooling of the exhaust gases before they enter the gas
cleaning device. Cooling is usually accomplished by introducing dilution
air, although atomizing water spray chambers, radiant-convection coolers,
and air- or water-cooled ductwork may also be used. Lower exhaust gas vol-
ume reduces the cost of the particulate collection equipment.
Fabric filters are the most frequently employed device for reducing
particulate emissions from electric arc furnaces in steel foundries. Fabric
filters also collect dust in a dry form which facilitates disposal. The
typical filter media used for bag construction includes woven or felted
glass, Dacron or Orion, and other synthetic fibers.1
Particulate emissions from electric arc furnace steel foundries
are predominantly less than 5 micrometers in diameter, and fabric filters
9.8-98
-------
Table 9.8.4-1. PARTICLE-SIZE DISTRIBUTION FOR PARTICULATE EMISSIONS
FROM THREE ELECTRIC ARC FURNACE INSTALLATIONS MELTING
GRAY IRON*
Cumulative percent by weight for
Particle size, indicated particle diameter
micrometers Foundry AFoundry BFoundry C
Less than 1 58 18
Less than 2 15 54 61
Less than 5 28 80 84
Less than 10 41 89 91
Less than 15 55 93 96
Less than 20 68 96 96
Less than 50 98 99 99
9.8-99,
-------
are generally recognized as the most efficient collectors of these particu-
lates.2 Participate removal efficiency for properly designed and operated
fabric filters is 99.5 percent.1 Emission tests of a 33.2 Mg capacity fur-
nace obtained by EPA that used EPA Method 5 to measure particulates showed
emission concentrations of 5.73 mg/DSCM for an 80-minute period during back
charging, 2.8 mg/DSCM for an 80-minute period during the middle of the heat
and 8.71 mg/DSCM for a 30-minute period after back charging but prior to
oxygen lancing.2 The average outlet loading was 5.74 mg/DNCM. EPA obtained
EPA Method 5 test data from another foundry that showed emissions reductions
from 686 mg/DNCM to 6.63 mg/DNCM.2 This last test was for 4 hours and encom-
passed two consecutive heats.
Wet scrubbers, although applicable to electric arc furnace foundries,
have been installed at very few locations. Wet scrubbers are generally not
as efficient as baghouses for collecting the fine particles generated by
electric arc furnaces. A commercially available high energy scrubber, the
Steam-Hydro scrubber, achieves high removal efficiency of submicrometer
particles by means of steam injected at supersonic velocity in a mixing
section.2
Maintenance and capital costs are less for wet scrubbers than for bag-
houses. However, disposal of scrubber wastewater requires water treatment
and/or water recirculation systems at the plant. Efficiencies in'the range
of 98 to 99 percent are possible for particulate removal.
A wet scrubber on an electric arc furnace would produce about 0.21
m^/Mg (50 gal/ton) of wastewater if operated prudently. Available data
indicate that this wastewater would have a pH of approximately 8.0 and a
total organic carbon content of less than 25 mg/1.
Treatment to remove suspended solids would be required before the
wastewater could be discharged to a receiving stream. Water treatment
technology for removal of suspended solids is well established. "Best
practicable control technology currently available," as defined by EPA,
will reduce total suspended solids to 50 mg/1, a 99.8 percent removal.3
The use of the above treatment would require disposal of approximately
6.9 kg of solids per Mg of iron (13.8 Ib/ton) and 8.0 kg of solids per Mg
of steel (15.9 Ib/ton). These solids would be in the form of high-solids
sludge containing up to 60 percent solids.
. 9.8-100
-------
9.8.4.3 Solid Waste Handling and Disposal. Dust collected from found-
ries is usually placed in landfills. Economic recycling of this iron-bearing
dust has not been demonstrated. The most effective method for handling dust
is a pelletizing operation which practically eliminates re-entrainment pro-
blems. Dust from the baghouses can be emptied into sealed bags or containers
which would also serve to contain dusts at the disposal site. Another option
is to produce a slurry by injecting water into the dust handling system.
Many foundries will probably continue to handle loose dust, usually trans-
porting it by truck to the disposal site. Open-bodied trucks should have a
cover placed over the load, and vehicle speed should be limited to avoid
losses during transport.
9.8.4.4 Landfill Disposal. Care in disposal of electric arc furnace
dust is necessary because relatively high levels of trace elements, including
the toxic metals lead, cadmium, and arsenic, are often present as metal
oxides, which are insoluble in pure water but are slightly soluble in acidic
solutions. Landfill site design must preclude horizontal or vertical migra-
tion of these metals to surface or groundwaters. The Safe Drinking Water Act
of 1974 provides for protection of potential drinking water supplies and sets
limits on the concentration of certain toxic metals. Where geo-hydrological
conditions do not provide reasonable protection against leaching of these
elements, devices such as impervious liners are needed.
Where wet scrubbers are used, scrubber wastewater should be contained
in a settling pond and recirculated. Protection of groundwaters and surface
waters is essential, and the landfill disposal requirements for scrubber
sludge are the same as those for dust.
9.8-101
-------
3.
REFERENCES FOR SECTION 9.8.4
Varga, J. Jr., and H. W. Lownie. A System Analysis Study of the Inte-
grated Iron and Steel Industry. Battelle Memorial Institute, Columbus,
OH. May 1969.
Air Pollutant Control Techniques for Electric Arc Furnaces in the Iron
and Steel Foundry Industry. U.S. Environmental Protection Agency.
Publication No. EPA-450/2-78-024. June 1978.
Development Document for Effluent Limitations Guidelines and New Source
Performance Standards for the Steel Making Segment of the Iron and Steel
Manufacturing Point Source Category. U.S. Environmental Protection
Agency. EPA Report No. EPA-440/l-74-024a. June 1974. pp. 352-357.
9.8-102
-------
9.8.5 Primary Aluminum
Primary aluminum production involves the recovery of metallic
aluminum from bauxite ore. Bauxite ore is processed into an alumininum-
containing material called alumina. This processing generally occurs
in plants located in coastal areas, and it involves a calcining oper-
ation which can produce significant amounts of particulate emissions.
Processed alumina is shipped to reduction plants where it is electro-
lytically reduced to aluminum metal. Primary aluminum production plants
are located in the Pacific Northwest, the Gulf Coast region, and the Ohio
River Valley. These plants are located in areas with low electric power
costs since electric power requirements for this industry are high.
Between 1967 and 1972, domestic primary aluminum production capacity
increased from 3,200 to 4,340 Gg. However, during the same period, domes-
tic plant utilization decreased from 98 to 86 percent. Nevertheless,
estimates indicate that between 1972 and 1984, the production of aluminum
may double.1
Particulate emissions from aluminum material handling and electric
reduction were estimated to be 38,500 Mg in 1977.2
9.8.5.1 Process Description and Emissions. Raw bauxite ore contains
aluminum compounds plus a variety of other compounds including Si02,
Fe203, and Ti02- Bauxite is processed to produce an aluminum-containing
material called alumina (^203), which is shipped to primary aluminum-
production facilities for processing into aluminum metal by electrolytic
reduction. A schematic of the production process at a primary aluminum
production facility is shown in Figure 9.8.5-1.3 All domestic primary alumi-
num production is performed by electrolytic reduction of the alumina to
aluminum and oxygen; the reaction takes place in a molten salt bath of cryo-
lite (a double fluoride salt of ^AlFs). Figure 9.8.5-2 is a schematic
of an aluminum electrolytic reduction eel 1.4
The reduction process is carried out in a cell or pot made of carbon-
lined steel. A carbon block is extended into the pot, which contains the
mixture of alumina dissolved in the molten cryolite electrolyte. The pot
and the carbon block are electrically connected to serve as a cathode and
an anode, respectively. The cell is heated to between 950 and 1000°C by
9.8-103
-------
DILUTION ,
WATER I
I RED MUD I
(IMPURITIES) I
00
ALUMINA PRODUCTION
FROM BAUXITE USUALLY
OCCURS AT SITES WHICH
ARE REMOVED FROM
ALUMINUM REDUCTION
PLANTS
ANODE PASTE
Figure 9.8.5-1
Schematic diagram of primary aluminum production
process.3
-------
o
Ml
o
en
DIRECT CURRENT
ANODE
ALUMINUM
MOLTEN
ALUMINA
"PAD
CATHODE
\S.
MOLTEN CRYOLITE
/BATH
Figure 9.8.5-2 Alurtiinum reduction
-------
the heat generated from the electrical resistance between the electrodes.
Aluminum ions migrate to the cathode, where they are reduced to aluminum.
The aluminum settles to the bottom of the pot and is removed periodically.
The oxygen from the alumina compound migrates to the carbon anode, where
it reacts with the carbon to form CO or C02« The carbon block anode is
thus consumed and must be replaced periodically.
The two basic types of reduction cells, prebake (PB) and Soderberg,
differ in the process by which the carbon anode is prepared. The PB cell
anodes are replaced about every 10 to 20 days. The anodes are prepared by
mixing crushed carbon material with pitch, forming this mix into the anode
shape and slowly baking the anode to form a solid block. The Soderberg
cell anode is formed by periodically adding a carbon and pitch mixture to
the top of the electrode. Heat from the process then drives off the lower
boiling organics, and the new anode material is baked to the old anode.
Thus, the anode is continually replenished. The Soderberg cell can be
further classified into two cell types: the horizontal stud Soderberg
(HSS) and the vertical stud Soderberg (VSS).
The anode bake ovens and aluminum reduction pots are the major sources
of particulate emissions. Particulate emission test results from a number
of primary aluminum production facilities are given in Table 9.8.5-1.5
Reduction cell emissions contain particles which are composed of alumina,
carbon, cryolite, aluminum fluoride, calcium fluoride, chiolite, and ferric
oxide.6 Particle size distributions of PB and HSS cell emissions are
tabulated in Table 9.8.5-2.7 These emissions, as well as those from refining
and casting furnaces, range in size down to submtcrometer levels.6 -The
composition of particulate emissions from a variety of secondary emission
sources in primary aluminum plants is shown in Table 9.8.5-3.
Dust constituents from four sources of industrial process fugitive
emissions are shown in Table 9.8.5-4.8 About 10 to 25 percent by weight of
the generated particulate matter is composed of fluoride compounds.8
Average composite particle size distribution data for fugitive emissions
from a reduction cell facility are presented in Figure 9.8.5-3. Fugitive
emissions from the anode baking and metal refining operations are in the
submicroineter size range.8
9.8-106
-------
Table 9.8.5-1. RESULTS OF EPA SOURCE TESTS FOR PARTICULATES FROM PRIMARY ALUMINUM REDUCTION CELLS
AND ANODE BAKE PLANT, mg/DNCM (kg/Mg OF ALUMINUM OR ANODE PRODUCED)5
00
I
I—'
o
-•J
Facility
Potline
Anode bake
Cell type
Control3
Primary
Secondary
Participates
Front half
Primary inlet
Secondary inlet
Primary outlet
Secondary outlet
Total
Primary inlet
Secondary inlet
Primary outlet
Secondary outlet
A
VSS
BS-WESP
SS
1550 (32.9)
5.9 (10.9)
1.5 ( 0.03)
0.9 ( 1.64)
1553 (33.8)
7.2 (13.3)
1.5 0.05)
1.6 ( 2.9)
B BI
PB PB
FBDS FBDS
None None
352 (55) 316.2 (49)
40.8 ( 7.9) 3.5 { 0.615)
320.5 (49.7)
5.1 (0.9)
D
PB
FBDS
None
325.1 (26.3)
4.8 ( 0.38)
446.6 (36.1)
13.9 ( 1.11)
E
HSS
ST-WESP
None
163.1 (33.3)
7.3 ( 1.48)
200.2 (40.9)
14.6 ( 2.98)
F
VSS
ST-WESP
None
5972 (14.2)
5.3 ( 0.13)
750.5 (18.0)
27.7 { 0.62)
G
ESP
39.35 (0.39)
80.3 (0.78)
aControl: BS - "Bubbler" scrubber.
WESP - Wet electrostatic precipitator.
FBDS - Fluidized bed dry scrubber.
ST - Spray tower.
SS - Spray screen.
-------
Table 9.8.5-2. REPRESENTATIVE PARTICLE SIZE DISTRIBUTIONS.
OF UNCONTROLLED EMISSIONS FROM PREBAKED
AND HORIZONTAL STUD SODERBERG CELLS7'
Particles within size range,
wt %
Par-ticle
size range,
micrometers
Less than 1
1 to 5
5 to 10
10 to 20
20 to 44
Greater than 44
Prebaked
35
25
8
5
5
22
Horizontal stud
Soderberg
44
26
8
6
4
12
9.8-108
-------
Table 9.8.5-3. ATMOSPHERIC POLLUTANTS FROM SECONDARY SOURCES
IN ALUMINUM PLANTSa»6
Process and emissions source Participate emissions
Raw materials handling
Alumina unloading and transfer A1203
Cryolite unloading, transfer and grinding Na3AlF6, A1F3
Pot lining operation
Anthracite coal grinding and transfer Coal dust
Anode preparation (prebake plants only)
Coke unloading, transfer and 'grinding Coke dust
Pitch unloading and transfer Pitch dust
Anode baking furnaces Carbon dust
Cleaning of baked anodes Coke dust
Cleaning of copper rods and steel stubs Copper and iron dust
Electric arc cast iron furnace Iron oxide
Paste making (Soderberg plants only) Carbon dust
Aluminum refining
Ingot casting furnaces Aids, A1203
Excluding the pot line (reduction cells) and monitor (roof vent)
emissions.
9.8-109
-------
Table 9.8.5-4. INDUSTRIAL PROCESS FUGITIVE EMISSION SOURCES
AND CONTAMINANTS8
Emission sources Emissions3
Material handling A1203, A1F3
Anode baking Coke and pitch dust
Electrolytic
reduction cell A1203, A1F3, N32C03,
Soderberg anodes CaF2, NasAlsF^, carbon
dust, condensed HC, and tars
Refining A1C103, ^2°3> Cryolite
aAlumina (A1203), cryolite (Na3AlF6),
aluminum fluoride (A1F3), fluorspar (CaF2),
sodium carbonate (NaC03),
chiolite
9.8-110
-------
10.0
5.0
S-
OJ
•!->
OJ
O
O
S i.o
CJ
h—4
I—
Q;
CL.
0.5
0.2
J_
I
I
I
I
0.5 2 5 10 20 40 . 60 80 90 95
WEIGHT PERCENT LESS THAN STATED SIZE
Figure 9.8.5-3
Average composite particle size distribution by
weight for reduction cell facility roof ventilator
emissions.8
9.8-111
-------
Typical stack parameters for major emission sources in the primary
aluminum industry are summarized in Table 9.8.5-5.9 These data are
representative of nationwide operations and facilities and can be used
to estimate the impact that a specific source could have on ambient air
quality.
9.8.5.2 Particulate Control Techniques. Fluorides emitted from primary
aluminum-production operations are the pollutant of primary concern because
of well documented adverse environmental effects. Thus, the main control
emphasis is on fluorides, which are emitted in gaseous and particulate forms.
Source tests have shown that if fluoride emissions are well controlled, the
resulting control of particulate emissions is also good.4 Controlled emis-
sions of particulates determined by EPA source tests accomplished at facili-
ties that represent applications of best fluoride control technologies are
shown in Table 9.8.5-1.5
A variety of control devices have been used on many different kinds of
gases and particles emitted from the reduction cells. One or more types of
wet scrubbers (such as venturi, packed beds, and floating beds) have been
used to control gaseous and particulate emissions from PB, HSS, and VSS
cells, and from anode baking ovens. Particulate control methods such as wet
and dry electrostatic precipitators, multiple cyclones, and dry scrubbing
systems are being employed on anode baking ovens, and PB and VSS eel Is.5
Recently, however, the dry control systems have become the predominant
control system for all types of cells.
Table 9.8.5-6 presents estimated particulate and fluoride removal
efficiencies for a number of control systems as applied to reduction cell
emissions. The dry scrubbing system outlined in this table involves the use
of fluidized alumina solids to absorb fluorides. Any entrained particles
are then collected in bag filters. A schematic of this system is shown in
Figure 9.8.5-4.!0
The minimum collection efficiency of a wet electrostatic precipitator
is 95 percent.H Based on particulate emission measurements and over 40
hours of visual emission observations by EPA, it was concluded that a dry
control system (such as alumina fluid bed) or a wet scrubber in series with
a wet electrostatic precipitator provides the best system of particulate
control.
9.8-112
-------
Table 9.8.5-5. ALUMINUM ORE ELECTROREDUCTION STACK PARAMETER DATA9
00
I
Average Average
Number stack stack
Emission of height, diameter,
sources sources m m
Prebake cells 50 36.6 4.57
Horizontal 83 21.3 5.36
stud Soderburg
Vertical 5 21.3 4.57
stud Soderburg
Materials 56 20.4 0.853
handling
Anode bake 20 40.5 1.77
furnace
Average
exhaust gas
temperature,
oc
58
54
39
79
' 112
Average Average
gas flowrate, operating rate,
Am3/s Gg
842 123
421 51.4
59.4 62.4
18.7 84.9
28.6 106
-------
Table 9.8.R-6. AIP POLLUTION CONTROLS FOR PRIMARY ALUMINUM REDUCFION
POTLINE EMISSIONS'1.6
00
I
Type of
cell
HSSh
Prebake
vssf
Est. removal
Existing
collectors Fluorides'1
Soderhern 80 to 90
spray scrubbers
Mul tic! ones 0
dry electrostatic
Spray scrubbers 8(1 to 90
Soderberg mult.i- 0
clones
Spray scrubbers 80 to 90
efficiencies, Est. removal
%
Particulate Latest
matter collectors
40 to 50 1. Floating bed
scrubber0
2. Venturi scrubber
followed by
wet electrostatic
precipitator
3, Dry alumina
scrubbers
-------
00
_3
01
CLEAN GAS OUT
i i i i
••MM
v-
MMMM
.UII
V
MM
ALU
•MH^M
MIN
ft
i "'«•
•Wi
/
r-
ALUMINA CONTAINING
FLOURIDE TO REDUCTION
CELLS
GAS DISTRIBUTION PLENUM
BAG FILTER
ALUMINA FEED
JUULT ,1
GAS AND DUST FROM
FAN EXHAUSTING
REDUCTION CELL HOODS
Figure 9.8.5-4 Schematic layout of Alooa 398 process reactor.10
-------
Industrial process fugitive emissions from materials handling can be
controlled by standard techniques such as water spray, use of enclosures,
and hooding/venting to particulate removal systems. The basic technique
for controlling fugitive emissions from anode baking, reduction cell, and
refining operations is to use adequate hooding/collection devices with
increased exhaust flow rates that capture and direct the emissions to
particulate control devices. In some cases, it may be more feasible to
contain the particulate emissions in the building and vent the building
exhaust through a control system.7 A study to control exhaust emissions
from a reduction cell building indicates that a wet scrubbing system can
reduce building exhaust particulate emissions from 164 kg/h to 88 kg/h.12
9.8.5.3 Secondary Environmental Impacts. Table 9.8.5-7 presents a
summary of energy consumption estimates for equipment designed to meet
various pollution control levels applicable to various aluminum production
unit processes. Two developments are currrently being evaluated which could
reduce the energy requirements of the electrical reduction process and also
the energy requirements for controlling air pollution from this operation.
These developments involve the use of titanium dibromide cathodes in the
present reduction cells and the use of a new electrical reduction process
called the Alcoa Chloride Electrolysis Process. The Alcoa process has been
described by Alcoa as offering the potential for producing aluminum with an
energy savings of about 10 percent while maintaining operating costs at
their present level. Both the Alcoa process and the use of titanium dibro-
mide cathodes are expected to reduce air pollution from the reduction cells
and from the anode production operations.
9.8-116
-------
Table 9.8.5-7,
SUMMARY OF IN-PLANT ENERGY CONSUMPTION PER TON OF
ALUMINUM FOR POLLUTION CONTROL BY PLANT TYPE
(kllowatt-hours/ton)a
Control
level
Federal new source performance
standards
Unit
process
1972:
base year
State
implementation
plans
New plants^
Existing plants
to achieve a
2 Ib. TF/TARC
CWPB
Potroom 199.6
Bake plant 8.4
Paste plant 7.8
Rod room 3.3
289.5
54.7
10.7
3.3
425.0
90.8
10.7
3.3
697.0
90.8
10.7
3.3
Totals
219.1
Totals
249.0
Totals
167.8
358.2
450.6
332.3
529 .8
801.8
SWPB
Potroom-
Bake plant
Paste plant
Rod room
211.6
16.3
7.8
3.3
419.8
16.8
10.7
3.3
530.0
18.4
10.7
3.3
562.4
vss
Potroom
Paste plant
160.0
7.8
321.6
10.7
2633.0
10.7
2654.7
HSS
Potroom 288.1
Paste plant 7.8
746.3
10.7
1091.0
10.7
Totals
295.9
757.0
1101.7
CODE: CWPB = Center-worked Prebake
VSS = Vertical Stud Soderberg
SWPB = Side-worked Prebake
HSS = Horizontal Stud Soderberg
aTo convert from kilowatt-hours/ton to kilowatt-hours/Mg, multiply by 1.1.
^Energy requirements for a new center-worked prebake plant to achieve
the New Source Performance Standard promulgated January 26, 1976.
cThis scenario assumes existing plants will achieve a 2-pound total
fluoride per ton of aluminum produced standard.
9.8-117
-------
REFERENCES FOR SECTION 9.8.5
1. Field Surveillance and Enforcement Guide for Primary Metallurgical
Industries. U.S. Environmental Protection Agency. EPA-450/3-73-002.
December 1973.
2. OAQPS Data File of Nationwide Emissions: U.S. Environmental Protection
Agency, Office of Air Quality Planning and Standards. Research
Triangle Park, NC. February 1979.
3. Compilation of Air Pollution Emission Factors, Third Edition. U.S.
Environmental Protection Agency, Office of Air Quality Planning and
Standards. Research Triangle Park, NC. August 1977.
4. Background Information for Standards of Performance: Primary Aluminum
Industry. Volume 1: Proposed Standards. U.S. Environmental Pro-
tection Agency. Publication No. EPA-450/2-74-020a. October 1974.
5. Air Pollution Control in the Primary Aluminum Industry, Volume II.
Singmaster & Breyer. New York, NY. EPA Publication No. 450/3-73-004B.
July 1973.
6. Vandegrift, A. E., et al. Particulate Pollutant System Study. Volume
III: Handbook of Emission Properties. Midwest Research Institute.
Kansas City, MO. Publication No. PB-203 522. May 1971.
7. Engineering and Cost Effectiveness Study of Fluoride Emissions Control,
Volume 1. TRW Systems and Research Corp. Reston, VA. January 1972.
8. Technical Guidance for Control of Industrial Process Fugitive Par-
ticulate Emissions. U.S. Environmental Protection Agency. Publi-
cation No. EPA-450/3-77-01D. March 1977.
9. Atmospheric Modeling Data from National Emission Data System (NEDS).
U.S. Environmental Protection Agency, Office of Air Quality Planning
and Standards. Research Triangle Park, NC. May 1979.
10. Parker, A. (ed.). Industrial Air Pollution Handbook. McGraw-Hill
Book Company, Ltd. London, England. 1978.
11. Proceedings: Particulate Collection Problem Using ESPs in the Metal-
lurgical Industry. Southern Research Institute. Birmingham, AL.
Publication No. PB-274 017. October 1977.
12. Particulate and Fluoride Emissions Control. Anaconda Aluminum Company.
Columbia Falls, MT. PEDCo-Environmental Specialists, Inc. Publication
No. PB-255 241. February 1974.
9.8-118
-------
9.8.6 Primary Copper Smelters
Pyrometallurgical smelting methods are utilized extensively in the
United States to produce copper from sulfide ores. These ores usually con-
tain less than 1 percent copper and therefore must be concentrated before
being transported to the smelter. Concentration to 15 to 35 percent cop-
per is accomplished by crushing, grinding, and flotation at the mine site.
Sulfur content of the concentrate is generally 25 to 35 percent. Most of
the remaining concentrate is iron (25 percent) and water (10 percent). Some
concentrates also contain significant quantities of arsenic, cadmium, lead,
boron, antimony, and other heavy metals.
Currently, there are 16 primary copper smelters operating in the United
States. Of these, seven are located in Arizona, two in New Mexico, and one
each in Nevada, Texas, Utah, Montana, Tennessee, Michigan, and Washington.
The concentration of copper smelters in the Southwest is due mainly to the
local availability of copper bearing ores. Domestic copper smelting capa-
city in 1976 totaled approximately 8.2 million Mg of charge, which is equi-
valent to about 1.85 million Mg of smelter product (99 percent "blister"
copper).! Projected production for 1985 and 2000 is forecast at 1.82 and
2.86 million Mg of copper, respectively, which reflects an annual growth
rate of 3.0 percent.
Particulate emissions from domestic primary copper smelting under
existing controls were estimated to be 14.7 Gg in 1977.2
9.8.6.1 Process Description and Emissions. The pyrometallurgical
process used for the extraction of copper from sulfide ore concentrates is
based on copper's strong affinity for sulfur and its weak affinity for oxygen
as compared to that of iron and other base metals in the ore. The purpose
of smelting is to separate the copper from the iron, sulfur, and gangue
materials. Conventional practice includes three operations:
o Roasting to remove a portion of the concentrate sulfur content.
o Smelting of the concentrate and fluxes in a furnace to form slag
and copper-bearing matte.
o Oxidizing of the matte in a converter to form blister copper
(about 99 percent pure copper).
9.8-119
-------
Typically, the blister copper product is further refined in a fire-
refining furnace prior to being cast into copper anodes and electrolytically
refined. Figure 9.8.6-1 illustrates the basic smelting operations employed
as well as the materials entering and leaving each operation.3 Currently,
half of the domestic smelters do not roast prior to smelting but instead
charge wet or "green" feed directly to the smelting furnace.
In the roasting of copper sulfide ore concentrates, the concentrates
are heated to a high temperature (but below the melting point of the con-
stituents) in an oxidizing atmosphere to: (1) eliminate a portion (20 to 50
percent) of the sulfur contained as S02; (2) remove volatile impurities
such as arsenic, antimony, and bismuth; and (3) preferentially convert a
portion of the iron sulfides present to iron oxides. The roasted concentrate
is called calcine. The degree of roast (i.e., the amount of sulfur and iron
oxidized in the roasting operation) is dependent on the desired quality of
the charge to the smelting furnace. Both multiple-hearth and fluidized-bed
roasters are used.
For either type of roaster, there are three major operating variables:
feed rate, combustion air flow rate, and temperature. A key difference
between the two is the S02 concentration in the roaster off-gases; the con-
centration is considerably higher for fluidized-bed roasters than for
multiple-hearth roasters due to the lower total air volume required. Average
stack gas S02 concentration is about 12 percent, with maximums near 18 per-
cent for fluidized-bed roasters and only 1.5 to 3 percent for existing
multiple-hearth roasters.
In the smelting operation, hot calcines from the roaster or raw, un-
roasted concentrates are melted in a smelting furnace with .limestone and
siliceous flux. The copper and iron which are present in the charge combine
with sulfur to form stable cuprous sulfide (Cu2S) and stable ferrous sulfide
(FeS). The combination of the two sulfides, known as matte, collects in
the lower area of the furnace and is removed periodically for further pro-
cessing. Such mattes may contain from 15 to 50 percent copper; 40 to 45
percent is the most common. The remainder of the molten mass, containing
most of the other impurities and known as slag, is of lower specific gravity,
and, therefore, floats on top of the matte from where it is drawn off and
discarded.
9.8-120
-------
MATERIALS ENTERING
THE SYSTEM
MATERIALS LEAVING
THE SYSTEM
RAW CONCENTRATES *»•
FIIFl . * ,1,1. , , ^
nir !-.'•• —
SILICEOUS FLUX ^
•MATERIAL HIGH
IN COPPER
ROASTER
(MULTIPLE-HEARTH
OR FLU1DIZED BED)
CALCINES
1
SMELTING FURNACE
(REVERBERATORY
OR ELECTRIC-ARC)
MATTE 1
I SLAG
CONVERTER
BLISTER
COPPER
t
FIRE-REFINING FURNACE
ANODE
COPPER
i
CASTING WHEEL
ANODES
1
ELECTROLYTIC REFINING
^ GASES DUST AND VOLATILE OXIDES
TO CONTROL EQUIPMENT AND STACK
BOILERS, CONTROL EQUIPMENT
AND STACK
»»-SLAG TO DUMP
AND STACK
*- bAoto lUoIALK
**oLAu 1 U LUIMVbnTcH
Figure 9.8.6-1 Typical primary copper smelter flowsheet.3
9.8-121
-------
Currently, conventional reverberatory furnaces are used at 11 of the 16
existing primary copper smelters. Three smelters employ electric furnaces;
one smelter employs the Outokumpu flash furnace; and another uses a Noranda
continuous smelter.
In reverberatory furnace operation, heat is supplied by the combustion
of oil, gas, or pulverized coal. The fuel is burned above the copper concen-
trates being smelted. Flames from the burners may extend half the length of
the furnace. Part of the heat in the combustion gas radiates directly to
the charge lying on the hearth below, while a substantial part radiates to
the furnace roofs and walls and is reflected down to the charge. Combustion
gases will contain from 15 to 45 percent of the sulfur in the original
charge, depending primarily upon whether or not the concentrate was roasted.
However, because of the high volume of combustion air, S02 concentrations
are low, varying from 0.5 to 1.5 percent. These lean SC>2 mixtures, unlike
off-gases from the roaters, converters, and other types of smelting furnaces,
are not economically utilized as feed for sulfuric acid plants.
Electric smelting furnaces provide the heat necessary for smelting
copper ore concentrates by placing carbon electrodes in contact with the
molten bath within the furnace. The electrodes dip into the slag layer of
the bath, forming an electrical circuit. When an electric current is passed
through this circuit, the slag resists its passage, generating heat and
producing smelting temperatures.
In flash smelting furnaces, roasting and smelting are combined in one
operation to produce a high-grade copper matte (about 70 percent copper).
The concentrates and fluxes are injected with preheated air, oxygen enriched
air, or even pure oxygen, into a furnace of special design, and smelting
temperatures are attained as a result of the heat released by the rapid,
flash combustion of iron and sulfur.
The final step in the production of blister copper is converting. Con-
verting is normally performed in a Fierce-Smith converter, which consists
of a cylindrical steel shell mounted on trunnions at either end and rotated
about its major axis. An opening in one side of the converter functions as
a mouth through which molten matte, siliceous flux, and scrap copper are
charged to the converter and gaseous products are vented. Air or oxygen-
enriched air is blown into the bath through a series of openings called
9.8-122
-------
tuyeres. During the initial blowing period (the slag blow), FeS in the
matte is preferentially oxidized to FeO and Fe^O^, and sulfur is removed
with the off-gases as SC^. Flux is added to the converter to combine with
iron oxide and form a fluid iron-silicate slag. When all the iron is oxi-
dized, the slag is skimmed from the furnace, leaving behind "white metal" or
molten Cu2S. Fresh matte is charged into the converter at this stage and
the slag blowing continued until a sufficient quantity of white metal has
accumulated. When this happens, the white metal is oxidized with air to
blister copper during the "copper blow." The blister copper is removed from
the converter and then cast or subjected to additional fire refining prior
to casting.
Cooling of the hot converter gases is necessary in order to prevent
thermal damage to dry-gas cleaning equipment. Normally, this is accomplished
by adding dilution air that can vary in volume from 1 to 4 times the converter
off-gas. With dilution air, S02 concentrations in the converter off-gases
can vary from 1 to 7 percent and typically average about 3.5 percent. With
close fitting hoods or with Hoboken converters, the off-gases average 5 to
10 percent S02- However, when dilution air is not used, cooling devices
such as waste heat boilers, air/gas heat exchangers, or water sprays are
necessary.
Blister copper usually contains from 98.5 to 99.5 percent pure copper.
Impurities may include gold, silver, antimony, arsenic, bismuth,, iron, lead,
nickel, selenium, sulfur, tellurium, and zinc. To purify the blister copper
.further, fire refining and electrolytic refining are used. In fire refining,
air is blown through the metal to oxidize remaining impurities; these are
removed as a slag, and the remaining metal bath is subjected to a reducing
atmosphere to reconvert cuprous oxide to copper. The fire-refined copper is
cast into anodes and further refined electrolytically.
Electrolytic refining involves separation of copper from impurities by
electrolysis in a solution containing copper sulfate and sulfuric acid.
Metallic impurities precipitate from the solution and form a sludge that is
removed and treated for recovery of precious metals. The copper produced is
99.95 to 99.97 percent pure.
Actual emissions from a particular smelter unit depend upon the configu-
ration of equipment in the smelting plant and the operating parameters
9.8-123
-------
employed. Table 9.8.6-1 summarizes the emission factors for uncontrolled
participate emissions from major process units used in various smelter con-
figurations.^ Other potential emission sources, which have not been quanti-
fied, include ore crushing and preparation, flux crushing, ore storage,
concentrate drying, slag dumping, fire refining, and copper casting. Sig-
nificant quantities of fugitive emissions are also generated during material
handling operations and charging and topping of furnaces.
As a general observation, particulate emissions from primary smelting
operations are predominantly metallic fumes in the submicrometer size range.
A variety of particulate contaminants are typically emitted during the roast-
ing process. They vary in composition depending on the particular ore being
roasted. Copper and iron oxides are the primary constituents but other
oxides such as those of arsenic, antimony, mercury, lead, cadmium, and zinc
may also be present with metallic sulfates and sulfuric acid.6 Combustion
products from fuel burning also contribute to the emissions from roasters.
The dust content of the waste gases is strongly influenced by the charac-
teristics of the copper concentrates as well as the volume of air aspirated
by the roasting furnaces.
Particulate emissions from reverberatory smelting furnaces contain
significant amounts of copper, zinc, and sulfur, along with trace elements
that correspond to the ore composition, including oxides of arsenic, anti-
mony, aluminum, silicon, and sulfur (sulfates)J The amount of dust gener-
ated depends on the fineness of the furnace charge, the amount of charge
agitation and the density of the charge material. Particle size distribu-
tion data from reverberatory smelting flue system exhaust are shown in
Figure 9.8.6-2.^ Another source reports particulate size characteristics
from a reverberatory furnace at the inlet and outlet of an electrostatic
precipitator control device. This information is presented in Figure
9.8.6-3.
Converter exhaust streams contain particulate matter composed of lead,
antimony, arsenic, bismuth, selenium, tellurium, zinc, cadmium, and thallium.
Particulate size distribution data from a converter exhaust flue are shown
in Figure 9.8.6-4. Particulate emissions from fire refining are minimal,
and atmospheric emissions from electrolytic refining are relatively minor.9
9.8-124
-------
Table 9.8.6-1. EMISSION FACTORS FOR PRIMARY COPPER SMELTERSa»M
Smelter
configuration
Unit
Control0
Particulatesd
Ib/ton kg/MT
Reverberatory furnace
followed by
converters
Multiple-hearth roaster
followed by reverber-
atory furnace and
converters
Fluidized-bed roaster
followed by reverber-
atory furnace and
converters
Fluidized-bed roaster
followed by electric
furnace and
converters
Reverb.
Converter
.Roaster
Roaster
and
reverb. d
Converter
Roaster
Reverb.
Converter
Roaster
None
ESP
None
ESP
ESP+SCAP
None
Baghouse
None
ESP
Spray
chamber +
ESP
None
ESP
ESP + SCAP
ESP + SCAP
None
Baghouse +
SCAP
ESP
ESP + SCAP
None
Baghouse +
SCAP
36
22
42
2.5
0.28
45
0.2
4.8
1.4
42
2.9
0.38
0.38
55
0.1
2.4
1.1
55
0.1
18
11
21
1.3
0,14
22.5
0.1
2.4
0.7
21
1.5
0.19
0.19
28
0.05
1.2
0.55
28
0.05
Total uncontrolled smelter '
None
135
66.5
a Emission factors are expressed as units per unit weight of
concentrated ore processed by the smelter. Approximately 4 unit
weights of concentrate are required to produce 1 unit weight of
copper metal.
" Other potential emission sources include (1) ore storage, crushing,
and handling, (2) flux crushing and handling, (3) concentrate
drying and handling, (4) slag dumping, (5) fire refining, and (6)
copper casting. Emission rates have not been quantified, however.
c ESP = electrostatic precipitator
SCAP = single-contact acid plant
DCAP = double-contact acid plant.
d Roaster and reverberatory furnace emissions are combined and
therefore a single set of emission factors is provided.
9.8-125
-------
in
-------
103
01
Q
et
O
oo
oo
TOO
I T I I I I I I I I I II I I I || I I M I I I 1
INLET
OUTLET
100
102
PARTICLE DIAMETER, micrometers
Figure 9.8.6-3
Average cumulative inlet and outlet mass loading
vs. particle size, copper reverberatory furnace.!
9.8-127
-------
o
i-
u
on
LU
fc
TO
5
4
3
2
0.5
0.4
0.3
0.2
0.1
PRIOR TO ELECTROSTATIC
PRECIPITATOR
FOLLOWING ELECTROSTATIC
PRECIPITATOR
SAMPLED AT CONVERTER MOUTH
i i i i i
10 30 50 70 90 99
PERCENT GREATER THAN STATED SIZE
Figure 9.8.6-4 Converter flue system particulate size distribution.8
9.8-128
-------
Fugitive participates emitted from primary copper smelting consist basi-
cally of metallic oxides and dust. Major sources of fugitive emissions are
ore concentrate unloading and handling, roaster calcine transfer operations,
furnace tapping operations, and converter charging and skimming operations.5
9.8.6.2 Control Techniques. Control devices for particulate emissions
from roasting, smelting, and converting operations include mechanical collec-
tors (cyclones and settling flues)3 hot and cold electrostatic precipitators,
fabric filters, and scrubbers. Electrostatic precipitators, usually preceded
by mechanical collectors and operated at elevated temperatures, are by far
the most commonly applied.
The control techniques actually applied vary, depending on smelter con-
figuration, process equipment mix, emission characteristics, and feasibility
for sulfur dioxide control. Off-gases from smelting equipment which produce
relatively high concentrations of S02 (>-4 percent), including fluidized-bed
roasters, non-reverberatory smelting furnaces, and converters, are generally
treated in single- or double-contact sulfuric acid plants for S02 removal.
The presence of solid and gaseous contaminants such as acid mist, entrained
dust, and metal fumes in these off-gases can present serious difficulties
in the operation of an acid plant. The major difficulties caused by these
contaminants include the corrosion of heat exchanger tubes, plugging of
catalytic beds, deactivation of the catalyst, and contamination of the pro-
duct acid. As a result, rather extensive measures are taken to remove these
contaminants to ensure that their concentrations are reduced to tolerable
levels prior to entering the acid plant for sulfuric acid production.
Both hot and cold gas cleaning devices are used. Generally, the off-
gases are initially treated in a hot electrostatic precipitator, where the
coarse particulates, which contain high metal values, are removed. The
gases exiting the precipitator are then scrubbed in one or more packed-bed
or impingement type scrubbers, where, in addition to undergoing further
particulate removal, the gases are humidified and cooled. The cooled,
gases then enter a series of electrostatic mist precipitators, where acid
mist, fine particulate, and volatile metals are removed prior to entering
the acid plant. If more elaborate cleaning is required, venturi type
scrubbers are used upstream of the cooling towers. Although complete removal
9.8-129
-------
of contaminants such as arsenic from the off-gases is not practical, 99+
percent removal is considered achievable.
The off-gas streams produced by existing multiple-hearth roasters and
reverberatory smelting furnaces are generally considered too lean in S02
concentration for the economic recovery of S02 using sulfuric acid plants.
However, the control of particulate matter emitted from these process
facilities is standard practice because of the value of the copper contained
in the dust and the presence in some instances of significant quantities of
potentially hazardous substances such as arsenic. Typically, reverberatory
furnace off-gases exit the furnace and pass through one or more waste heat
boilers, where the gas stream is cooled to about 370°C (700°F) and a signifi-
cant portion of the heat is recovered for pov/er generation. Generally, the
gases then pass through a large rectangular flue (balloon flue), where the
gas velocity is reduced sufficiently to allow the coarser particles (larger
than 40) to fall out prior to entering an electrostatic precipitator for
fine particulate removal. The electrostatic precipitators used are generally
operated at temperatures ranging from 200°C to 315°C (400°F to 600°F). De-
sign efficiencies normally range from 95 to 99 percent. Excluding the waste
heat recovery step, particulate emissions from multiple-hearth roasters are
treated similarly.
Table 9.8.6-2 presents performance data obtained using in-stack filter
measurements on a dry electrostatic precipitator used to control particu-
late emissions from a typical "green" charge reverberatory smelting fur-
nace. 1° The precipitator is designed to handle 4,250 actual m3/min (150,000
SCFM) at 315°C (600°F) and has a total collection area of 3,860 m2 (40,500
ft^). Simultaneous inlet and outlet measurements were conducted using
in-stack filters. As the data indicate, the overall collection efficiency
measured was about 96 percent. It should be noted, however, that while the
subject precipitator is reasonably effective in removing material that
exists as particulate at its operating temperature, any volatile materials,
such as metal oxides of arsenic or selenium that exist in the vapor state
at the precipitator operating temperature, would pass through the precip-
itator with little or no removal. Measurements conducted on the same pre-
cipitator specifically for arsenic (AS203) showed an arsenic collection
9.8-130
-------
Table 9.8.6-2. REVERBERATORY FURNACE ESP PERFORMANCE DATA*
Sample
run
1
2
3
4
5
Average
Mass concentration, g/dscm (gr/dscf)
Inlet ' Outlet
0.156 (0.358)
0.283 (0.647)
0.283 (0.647)
0.189 (0.433)
0.184 (0.422)
0.219 (0.501)
0.008
0.007
0.008
0.010
0.011
0.009
(0.018)
(0.017)
(0.019)
(0.022)
(0.025)
(0.020)
Efficiency
(*)
95.0
97.4
97.1
94.9
94.1
95.7
aBased on in-stack filter measurements.
9.8-131
-------
efficiency of less than 30 percent. As a result, participate controls ap-
plied at smelting facilities that process materials containing high levels
of volatile impurities should employ preceding as an integral part of the
overall control system. The gas stream to be treated should be cooled to a
sufficiently low temperature, and adequate time should be allowed to en-
sure that the bulk of the volatile constituents present in the gas stream
are condensed prior to entering the control device for collection.
Methods commonly applied in the nonferrous metal industry for cooling
hot gas streams include radiative cooling towers, water spray chambers, and
dilution with ambient air. All three methods have advantages and disadvan-
tages. Although cooling with dilution air is the simplest alternative, its
application for the gas volumes and temperatures under consideration may
not be economical. Depending on the temperature of the gas stream to be
treated, the amount of dilution air needed to effect cooling could result in
a two- to four-fold increase in the total gas volume to be treated, with a
corresponding increase in the size and cost of the control device applied
and overall fan requirements.
Radiative cooling towers require considerable space because of their
need for sufficient heat transfer area. In addition, horsepower require-
ments are high due to the increased resistance to gas flow resulting from
the added ductwork needed. The major drawback to radiative cooling, however,
is its limited flexibility for temperature control. Because the temperature
of the gas stream exiting the cooling tower is dependent only on the tempera-
ture and volume of the inlet gas stream and the ambient air temperature, any
variation in either of these will result in a lower or higher outlet tempera-
ture than intended.
Cooling hot gases by evaporative cooling is relatively simple and re-
quires little space. Water spray chambers are currently used at a number
of copper smelters for cooling process gases from a variety of sources prior
to the entry of the process gases into an electrostatic precipitator or bag-
house for particulate removal. Typically, the spray chambers used have a
cross-sectional area of about 35 m^ (375 ft^) and are 30 to 60 meters (100
to 200 feet) in length. The large cross-sectional area results in a low
flow velocity which allows for a longer residence time. Water is introduced
9.8-132
-------
through a series of sprays along the cross-section of the chamber. Water
requirements will vary depending on the temperature of the stream to be
cooled and the desired end temperature.
The major difficulty with applying water spray chambers for cooling
smelter off-gases is the potential for corrosion. The presence of moisture
and sulfur oxides in the smelter off-gases introduces a lower temperature
constraint under which further cooling cannot be tolerated without incur-
ring severe operating problems. Because $63 is hygroscopic, it will absorb
moisture at temperatures well above the moisture dew point and form sulfuric
acid mist which is highly corrosive. The temperature at which this acid
mist formation occurs (acid dew point) is highly variable, depending on the
$63 concentration present as well as other gas stream characteristics. Con-
tinued operation of a dry control device at or below the acid dew point
could result in a severe corrosion problem due to acid attack. Thus, it is
generally recommended that the operating temperature of a dry control device
be maintained 10°C to 25°C above the best estimate of the acid dew point.
Little data are available on the acid dew point of smelter off-gases. How-
ever, practical experience at two smelters indicates that an operating tem-
perature of 110°C (230°F) is certainly within tolerable limits.
Performance data obtained on a spray chamber/electrostatic precipitator
used to control the combined off-gases from four multiple-hearth roasters and
a reverberatory smelting furnace are presented in Table 9.8.6-3.H The com-
bined gas stream, which averages about 5100 Nm3/min (180,000 SCFM), enters
the spray chamber, where it is cooled from about 220°C (428°F) to less than
115°C (240°F) prior to entering the electrostatic precipitator for particu-
late removal. The precipitator consists of seven parallel chambers, each
containing four sections. Simultaneous inlet and outlet mass particulate
measurements were conducted using out-of-stack filters (EPA Method 5). As
the data indicate, a collection efficiency of greater than 96 percent was
achieved.
Similar performance data are presented in Table 9.8.6-4 for a spray
chamber/baghouse used to control particulate emissions from a fluidized-bed
roaster, electric smelting furnace, and several converters.12 The collection
system consists of a baghouse preceded by two parallel spray chambers which
9.8-133
-------
TABLE 9.8.6-3 PERFORMANCE DATA FOR SPRAY CHAMBER/ESP9'11
Sample
run
1
2
Average
(g/dscm)
4.53
5.65
5.10
Inlet
(kg/hr)
1019
1233
1126
Outl
(g/dscm)
0.112
0.085
0.098
et
(kg/hr)
40.7
33.6
37.1
Efficiency
(%)
96.0
97.3
96.7
aEPA Method 5 (front-half only),
9.8-134
-------
Table 9.8.6-4. PERFORMANCE DATA FOR SPRAY CHAMBER/BAGHOUSEa.l2
Sample
run
1
2
3
Average
Inlet
(g/dscm) (kg/hr)
14.74 4060
13.64 3743
14.05 3819
14.14 3868
Outlet
(g/dscm) (kg/hr)
0.050 14.6
0.037 10.0
0.053 14.6
0.046 13.1
Efficiency
(%)
99.6
99.7
99.6
99.7
aEPA Method 5 (front-half only).
9.8-135
-------
effectively cool the inlet gas stream from about 315°C (600°F) to less than
110°C (230°F) prior to entering the baghouse. The baghouse, which consists
of 18 compartments, each equipped with 240 Orion bags, has a net collection
area of 29,800 m2. It is designed to treat 5,664 Nm3/min (200,000 SCFM)
effectively at an air-to-cloth ratio of 0.38 m3/m2 (1.25 ft3/ft2). Bag
cleaning is effected by mechanical shakers. As shown, the collection effi-
ciency achieved was greater than 99.5 percent.
In both of the above cases, the smelter charge contained high levels of
volatile impurities, especially arsenic.
Fugitive emissions produced by the majority of smelter fugitive sources,
including ore concentrate handling, calcine transfer, and furnace tapping
(matte and slag), are readily controllable by hooding and enclosing the
fugitive emission points and exhausting the captured emissions to a control
device for collection. Fugitive emissions associated with converter opera-
tions are much more difficult to control. These emissions are substantial
and occur during charging, skimming, or pouring operations when the converter
mouth is rotated out from under the primary hood. Control techniques which
have been applied include the use of secondary mechanical hoods in various
design and converter building evacuation (general ventilation).
9.8-136
-------
REFERENCES FOR SECTION 9.8.6
1. Schroeder, H. J. Copper - 1977. U.S. Department of the Interior,.
Bureau of Mines. 1977. p.2.
2. OAQPS Data File of National Emissions. U.S. Environmental Protection
Agency, Office of Air Quality Planning and Standards. Research Triangle
Park, NC. February 1979.
3. Compilation of Air Pollutant Emission Factors - Supplement No. 8. U.S.
Environmental Protection Agency. Publication No. AP-42. May 1978.
p. 7.3-2.
4. Reference 3, p. 7.3-5.
5. Controlled and Uncontrolled Emission Rates and Applicable Limitations
for Eighty Processes. The Research Corporation of New England.
Wethersfield, CN. Publication No. PB-266 978. September 1976.
6. Hardison, L. C., and C. A. Greathouse. Air Pollution Control Technology.
and Costs in Nine Selected Areas. Industrial Gas Cleaning Institute,
Inc. Publication No. PB-22 746. September 1970.
7. Vandergrift, A. E., et al. Particulate Pollutant System Study, Volume
III: Handbook of Emission Properties. Midwest Research Institute.
Kansas City, MO. Publication No. PB-203 522. May 1971.
8. Proceedings: Particulate Collection Problems Using ESPs in the
Metallurgy Industry. U.S. Environmental Protection Agency. Publica-
tion No. EPA-600/2-77-208. July 1976.
9. Trace Pollutant Emissions from the Processing of Metallic Ores. PEDCo-
Environmental Specialists, Inc. Publication No. PB-238 655. October
1974.
10. Schwitzgebel, K., R. T. Coleman, R. V. Collins, R. M. Mann, and C. M.
Thompson. Trace Element Study at a Primary Copper Smelter, Prepublica-
tion copy. Radian Corportion. Austin, TX. EPA Contract No. 68-01-4136.
January 1978.
11. Emission Test Report - ASARCO, Incorporated. El Paso, Texas. Prepared
by Monsanto Research Corporation. Dayton, OH. EMB Report No. 77-CU-6.
June 1977.
12. Emission Test Report - Anaconda Copper Company. Anaconda, MT. Prepared
by Monsanto Research Corporation. Dayton, OH. EMB Report No. 77-CUS-5.
August 1977.
9.8-137
-------
9.8.7 Primary Lead Smelting
Primary lead smelting entails the recovery of metallic lead from lead
ore concentrates. Raw lead ore frequently contains large amounts of zinc as
well as a number of other metals including cadmium, antimony, copper, and
silver. Normally, the lead- and zinc-containing ores are separated and
concentrated for processing and recovery at lead and zinc smelters.
There are six primary lead smelters in the United States, and produc-
tion from these smelters was estimated to be 612 Gg in 1976.1 Considerable
amounts of lead are also imported into the United States.
Primary lead production in the United States involves three basic oper-
ations: (1) sintering of ore concentrate, (2) processing of the sintered
material in a blast furnace to produce lead bullion, and (3) refining of
lead bullion to produce specification lead and lead alloys that are used in
a variety of products. In 1971, a total of 1.3 Tg of lead was consumed; 47
percent of this lead was used in the manufacture of lead acid batteries.2
Lead consumption appears to be stable for the near term. Therefore, it
is not expected that any new smelters will be constructed in the immediate
future.2 One source estimates an anticipated annual growth of about 2
percent, while qualifying this with the observation that environmental and
economic considerations and changing use patterns could have a major impact
on demand rates.3
Particulate emissions from sintering machine, blast furnace, and rever-
beratory furnace operations were estimated at 3260 Mg for 1977.4
9.8.7.1 Process Description and Emissions. The lead contained in lead
ore is found primarily as lead sulfide. Lead compounds are separated from
zinc compounds which frequently are combined in natural ore. The lead com-
pounds are concentrated and sent to a smelting facility for processing into
metallic lead.
A simple schematic of the primary lead smelting process is shown in
Figure 9.8.7-1.5 The initial sintering process has two basic purposes:
(1) to produce a lead oxide material by driving off the sulfur in the form
of sulfur dioxide, and (2) to produce a material with the dense permeable
properties needed for processing in the blast furnace. The sintering reac-
tion is autogenous and creates temperatures of about 540°C. Up to
9.8-138
-------
CONCENTRATE
SINTERING
SINTERED MATERIAL
REDUCTION
LEAD BULLION
REFINING
LEAD
AND
LEAD ALLOYS
Figure 9.8.7-1 Primary lead smelting process.5
9.8-139
-------
20 percent of the sinter feed material (including ore concentrates, coke,
and fluxes) can be emitted as dust and fumes. Economics require that much
of this material be captured and recycled.6
The sintered material, containing lead oxide and some sulfur not removed
during the sintering process, is mixed with coke and other materials and
charged into a blast furnace in which the lead oxide is reduced to produce
metallic lead (lead bullion). A schematic of a lead blast furnace is shown
in Figure 9.8.7-2. Particulate material emitted from the blast furnace can
contain up to 65 percent lead as well as cadmium and arsenic. Therefore,
much of this material is captured and recycled back to the sintering
process.7
The lead bullion and slag material are discharged from the base of the
blast furnace. The slag material (which is less dense than the lead bullion)
is separated by gravity, collected continually from the furnace, and either
treated in the smelter or shipped to other facilities to recover the metal
content (such as zinc, copper, and antimony). The lead bullion is then
transferred to refining kettles. Slag that is high in zinc may be treated
in a fuming furnace to recover zinc oxide.
Drosses that contain impurities are formed in the refining kettles and
then selectively skimmed to remove impurities from the lead and to bring the
lead into the desired specification limits for antimony, tin, arsenic, and
other elements. Zinc may be removed by applying a vacuum to the kettle and
removing the zinc fume. The drosses are then treated in a reverberatory
furnace to recover the lead and other metals removed in the refining kettle
dressing operation.
Emission factors for uncontrolled emissions from primary lead smelting
processes are presented in Table 9.8.7-1.8 The particulate emissions from
the sintering machine consist primarily of lead and zinc compounds, with
traces of oxides of such elements as arsenic, cadmium, selenium, and tel-
lurium. The size distribution of flue dust from a sintering machine is
shown in Table 9.8.7-2. Particulate material emitted from the blast furnace
typically contains lead oxides, zinc oxide, cadmium oxide, quartz, limestone,
iron pyrites, iron-lime-silicate slag, arsenic compounds, and other compounds
containing metals associated with lead ores.7
The metallic fume emissions from primary lead production are generally
9.8-140
-------
CHARGE
BUCKET
GAS OFFTAKE
SLAG
FURNACE DIMENSIONS
BETWEEN TUYERES = 1 METER
LENGTH = 7 METERS
HEIGHT = 8 METERS
BLOWER
BUTTON MOLD
Figure 9.8.7-2 Lead blast furnace.
9.8-141
-------
TABLE 9.8.7-1. EMISSION FACTORS FOR PRIMARY LEAD SMELTING
PROCESSES WITHOUT CONTROLS3'8
Parti culates
Process
Sintering (updraft)
Blast furnace
Dross reverberatory furnace
Materials handling
kg/MT
106.5
180.5
10.0
2.5
Ib/ton
213.0
361.0
20.0
5.0
Sulfur
kg/MT
275.0
22.5
Neg
dioxide
Ib/ton
550.0
45.0
Neg
aEmission factors expressed as kg/MT (Ib/ton) of lead product.
9.8-142
-------
Table 9.8.7-2. SINTERING MACHINE FLUE DUST SIZE DISTRIBUTION5
Size, Percent by weight
micrometers
20 to 40 15 to 45
10 to 20 9 to 30
5 to 10 4 to 19
5 1 to 10
9.8-143
-------
submicrometer in size. Fumes from lead sintering can contain lead, antimony,
zinc, cadmium, germanium, selenium, tellurium, indium, thallium, chlorine,
fluorine, and arsenic. Fumes from blast furnace operations contain essen-
tially the same materials as that from sintering, with the following repre-
sentative concentrations of trace metals: 0.01 to 1 parts per million each
of cadmium, manganese, nickel, tin, and vanadium; 0.01 to 0.1 ppm of copper;
1.0 to 10 ppm of magnesium; and 0.1 to 100 ppm of lead.9
Fugitive emissions from lead smelting operations can be significant.
The greatest amounts result from the handling and recharging of recovered
sinter materials into the sintering machine. Other significant fugitive
emission sources are lead ore concentrate handling and transfer facili-
ties, zinc fuming furnace vents, and reverberatory and blast furnace
leakage and tapping operations. Particulate fugitive emissions from the
blast furnace consist primarily of lead oxides of which 92 percent are
less than 4 micrometers.^ Concentrations of lead, cadmium, and zinc in
fugitive particulate emissions from several primary lead smelting opera-
tions are shown in Table 9.8.7-3.
9.8.7.2 Control Techniques. Fabric filters and electrostatic precipi-
tators are the most commonly used particulate emissions control systems in
the primary lead smelting industry. Fabric filters are used for particulate
emissions control on all of the blast furnaces and five of the six sintering
machines found in domestic primary lead smelting plants. The sixth sintering
operation employs an electrostatic precipitator. In many cases, the blast
furnace and dross reverberatory furnace are served by the same fabric
filter.2,10
A fabric filter is normally chosen only when the $03 concentration in
the flue gas is low since high $03 concentrations can corrode fabric filter
structures and deteriorate the filter fabric.2
Large dust and some fume particulates from sintering machines are
removed from flue gas by gravitational settling in large flues or chambers
prior to flue gas entry into an ESP or fabric filter. Blast furnace par-
ticulate matter may be collected in fabric filter units with wool or fiber-
glass bags.^ The results of emission tests on fabric filters applied to pri-
mary lead sintering machines and blast furnaces are given in Tables 9.8.7-4
and 9.8.7-5.
9.8-144
-------
Table 9.8.7-3. CONCENTRATIONS OF LEAD, CADMIUM, AND ZINC IN
FUGITIVE PARTICULATE EMISSIONS OF VARIOUS
PRIMARY LEAD SMELTING OPERATIONS' EMISSIONS
SOURCES5
Percent by weight
Operations' emissions sources
Ore concentrate storage
Return sinter transfer
Sinter sizing and storage
Sinter product dump area
Sinter transfer to blast furnace
Bl ast furnace roof vents
Blast furnace upset
Lead refinery roof vents
Lead casting roof ducts
Zinc fuming furnace area
Lead
37
19
58
31
39
47
27
37
38
3
Cadmi urn
0.8
0.6
0.7
0.6
0.7
0.4
4.0
0.3
0.1
Zinc
8 ,
2
5
6
6
8
7
19
18
62
9.8-145
-------
Table 9.8.7-4. PRIMARY LP.AD SHELTERS—SINTERING MACHINE BAG11CIUSE TES1" DATA™
00
en
Parameter
Fabric filter
type
Fabric
Temperature,
°C
A/C ratio
AP, kPa
Inlet, g/Nm3
Outlet, g/Nm3
Cooling
S02, %
Bag life,
months
Dust removal
Test data
Shaker Shaker
Acrylic
93 102
20.1 20.1
0.84
4.6 3.2
0.002
Water
spray
5.5
: 36+
Screw Cellars
conv.
Shaker Shaker
Polyester
127 116
32.9 31.9
1.00
4.8
0.007
Water
spray
5.0
23+
Cellars Screw
conv.
Shaker
—
93
32.9
5.1
0.45
0.11
—
-r
0.1
30
Screw
conv.
Shaker Shaker
Wool
93 71
8.1 29.3
6.4 8.9
0.25
0.02
„
Infilt.
—
53
Cellars Cellars
Shaker
Acrylic
71
29.3
8,9
0.58
0.01
--
Mater
sprays
—
40
Cellars
Shaker
Acrylic
65
10.0
5.1
0.20
--
--
Infilt.,
water
sprays
—
56
Cellars
-------
Table 9,8.7-5. LEAD SMELTERS—BLAST FURNACE BAGHOUSE TEST
00
Parameter
Fabric filter
type
Fabric
Temperature,
°c
A/C ratio
AP, kPa
Inlet, g/Nm3
Outlet, g/Nm3
Cooling
SOg, %
Bag life,
months
Dust removal
Test data
Shaker
—
93
32.9
0.50
4.6
0.11
««.
0.1
30
Screw
conv.
Shaker Shaker
Wool
93 71
8.1 29.3
0.63 0.87
- -- . 2.5
0.02
Infilt.
.-
53
Cellars Cellars
Shaker
Aery 1 i c
71
29.3
0.87
5.9
0.01
Water
sprays
—
40
Cellars
Shaker
Acrylic
65
. 10.0
0.50
2.0
—
Infilt.,
water
sprays
--
56
Cellars
Shaker
Acryl i c
-- ,
23.8
0.87
2.8
0.022
"•"-
—
'
Screw
conv.
-------
Fugitive emissions derive from a variety of operations and processes;
consequently, their control requires measures applicable to each source.
Fugitive emissions from sintering operations and reverberatory furnaces can
be controlled through proper operating and maintenance procedures such as
not using excessive fuel and maintaining equipment. Ore handling and trans-
fer fugitive emissions can be controlled by using water sprays and confining
emissions by enclosure. Zinc fuming furnace and blast furnace fugitive
emissions can be effectively controlled via adequate hooding and ventilation
to a fabric filter.5
• New Source Performance Standards for primary lead smelting process
facilities (promulgated on January 15, 1976) limit particulate emissions
from blast furnaces, dross reverberatory furnaces, and sintering machine
discharge ends to 50 mg/dscm (0.022 gr/dscf). Sulfur dioxide emissions from
sintering machines are limited to 0.065 percent by volume (650 ppm).
9.8-148
-------
REFERENCES FOR SECTION 9.8.7
1. Background Information for New Source Performance Standards: Primary
Copper, Zinc, and Lead Smelters, Volume I: Proposed Standards. U.S.
Environmental Protection Agency. Publication No. EPA-450/l-74-002a.
1974.
2. Assessment of Fugitive Particulate Emission Factors for Industrial
Processes. U.S. Environmental Protection Agency. Publication No.
EPA-450/3-78-107. 1978. >
3. Atmospheric Modeling Data from Material Emission Data System (NEDS).
U.S. Environmental Protection Agency. May 1979.
4. Field Surveillance and Enforcement Guide for Primary Metallurgical
Industries. U.S. Environmental Protection Agency. Publication No.
EPA-450/3-73-002.
5. Technical Guidance for Control of Industrial Process Fugitive Particu-
late Emissions. U.S. Environmental Protection Agency. Publication No.
EPA-450/3-77-010. March 1977.
6. Weisburg, M. I. Field Operations and Enforcement Manual for Air Pol-
lution Control, Volume III: Inspection Procedures for Specific
Industries. Pacific Environmental Services, Inc. Elmhurst, IL.
Publication No. APTD 1102. August 1972.
7. Vandegrift, A. E., et al. Particulate Pollutant System Study, Volume
III: Handbook of Emission Properties. Midwest Research Institute.
Kansas City, MO. Publication No. PB-203-522. May 1971.
8. Compilation of Air Pollutant Emission Factors. U.S. Environmental
Protection Agency. Publication No, AP-42. August 1977.
9. Trace Pollutant Emissions from the Processing of Metallic Ores. PEDCo-
Environmental Specialists, Inc. Cincinnati, OH. Publication No.
PB-238-655. October 1974. ,
10. Caplan, K. J. Applications of Baghouses in Lead and Zinc Smelters.
In: Control of Particulate Emissions in the Primary Nonferrous Metals
Industries. U.S. Environmental Protection Agency. March 1979.
9.8-149
-------
9.8.8 Primary Zinc Smelting
Primary zinc smelting entails the recovery of zinc from zinc ore
concentrates and from zinc containing slag produced from lead blast
furnaces.
Primary zinc production in the United States uses either a pyrometal-
lurgical or an electrolytic extraction process. The pyrometallurgical
process involves three basic operations: roasting the zinc ore concen-
trates, sintering the roasted ore, and reducing the sinter material to
produce metallic zinc. The electrolytic process involves roasting the
zinc ore concentrates, followed by chemical leaching and electrolytic
extraction of metallic zinc.
Between 1968 and 1971, a number of domestic primary zinc producers
ceased operation, reducing the estimated domestic zinc production capacity
from 1210 Gg in 1968 to 695 Gg in 1972.1 Table 9.8.8-1 shows the industry
structure as it changed between 1968 and 1972.2>3 Domestic primary zinc
production in 1976 was estimated to be approximately 680 Gg.4 The domestic
consumption of slab zinc is given in Table 9.8.8-2. The supply sources of
slab zinc are shown in Table 9.8.8-3.
Even though increased importation of foreign slab zinc is expected,
it is also anticipated that at least one new domestic smelter will be
buflt.l One source estimates that the demand for zinc will increase at
an average of between 1 to 3 percent per annum.5
Particulate emissions from roasting, sintering, and retort furnace
operations were estimated at 1270 Mg for 1977.6
9.8.8.1 Process Description and Emissjons_. The zinc contained in
zinc ore is found primarily as zinc sulfide (ZnS), also called sphalerite.
The zinc-containing ore usually has impurities such as lead, cadmium, and
minor amounts of other trace elements. The zinc compounds are separated,
concentrated, and then sent to a smelting facility for processing into
zinc oxide or metallic zinc. Zinc oxide may also be recovered as a by-
product in the primary lead smelting process when slag from the lead blast
furnace is treated in a fuming furnace.
Zinc is recovered in domestic smelters by using either the pyrometal-
lurgical or the electrolytic process. Simple schematics of these two
9.8-150
-------
Table 9.8.8-1. DOMESTIC SLAB ZINC CAPACITY 1968-19722,3
Estimated capacity,
Gg
Company/smelter location 1968 1972
Asarco, Amarillo, Texas
Asarco, Corpus Christi, Texas
Blackwell Zinc, Blackwell, Oklahoma
National Zinc, Bartlesville, Oklahoma
New Jersey Zinc, Palmerton, Pennsylvania
St. Joe Minerals, Monaca, Pennsylvania
Bunker Hill, Kellogg, Idaho
American Zinc, Dumas, Texas
American Zinc, Sauget, Illinois
Eagle Picher, Henryetta, Oklahoma
New Jersey Zinc, Depue, Illinois
Mathiessen & Hegler, Meadowbrook, W. Virginia
Anaconda, Great Falls, Montana
Anaconda, Anaconda, Montana
50
98
80
57
107
204
99
53
76
50
63.5
41
147
82
,50
98
80
57
107
204
99
—
__
—
—
—
--
—
Totals 1207.5 695
9.8-151
-------
Table 9.8.8-2. DOMESTIC SLAB ZINC CONSUMPTION (Gg)2>4
Consumption 1969 1970 1971
Galvanizing 447 430 431
Biassay 162 116 136
Other alloys 522 421 468
Other 124 110 102
Totals 1255 1077 1137
9.8-152
-------
Table 9.8.8-3. SUPPLY OF SLAB ZINC BY SOURCE (Gg)1.3
Source
Domestic ores
Foreign concentrates
Scrap
Subtotal domestic production
Slab imports
Total slab zinc supplied for
domestic consumption, exports,
and stocks
1969
420
528
64
1012
298
1310
1970
405
430
70
905
236
1141
1971
370
340
70
780
294
1074
9.8-153
-------
processes are shown in Figures 9.8.8-1 and 9.8.8-2. Emission factors for
uncontrolled particulate emissions from primary zinc smelting operations
are presented in Table 9.8.8-4.7
Ore roasting is done in both production processes. The roasting is an
exothermic reaction during which sulfur is removed from the zinc ore in the
form of S02, and the zinc is oxidized to form a ZnO calcirve material. Pro-
duction and emission data for typical roasting operations and various roaster
types are found in Table 9.8.8-B.8 Typical components of flue dust from
roasters are shown in Table 9.8.8-6.9 Uncontrolled emissions are estimated
at 120 Ib/ton of concentrate processed.
For the pyrometallurgical process, after roasting, the calcine is pro-
cessed in a sintering machine. Sintering eliminates any remaining sulfur,
volatilizes lead and cadmium, and forms a dense permeable material suitable
for feeding to a reduction furnace. Uncontrolled emissions are estimated at
90 Ib/ton of concentrate processed. The composition of particulate material
emitted by the sintering process is reported to be 5 to 25 percent Zn, 30 to
50 percent Pb, 2 to 15 percent Cd, and 8 to 13 percent S; and the particulate
emissions are reported to be Tess than 10 micrometers in size.
For the pyrometallurgical reduction operations, three reduction furnace
systems can be used. These systems are the horizontal retort furnace, the
vertical retort furnace, and the electrothermic furnace. These reduction
systems can use scrap zinc material as feed; 30 percent of the recycled or
secondary zinc production in 1972 occurred at primary pyrometallurgical
smelters.4
The basic reactions that takes place in reduction furnaces are as
fol1ows:
ZnO + CO : ^ Zn (vapor) + C02
C02 + C •» 2 CO.
Carbon reduction material is supplied either as coke or coal, which is
mixed with the sintered ZnO and charged into the reduction furnace. The
flue gas from reduction furnaces contains particulate matter that ranges in
size from micrometer to submicrometer and normally consists of 50 to 70
9.8-154
-------
CONCENTRATE
ROASTING
Calcine (ZnO)
LEACHING
1
PURIFICATION
I
ELECTROLYSIS
I
CATHODE
STRIPPING
I
MELTING &
CASTING
SLAB ZINC
TO RESIDUE
TREATMENT
Figure 9.8.8-1 Primary pyrometallurgical zinc smelting process.1
9.8-15*
-------
OXIDE
FURNACE
ZINC OXIDE
CONCENTRATE
ROASTING
CALCINE (ZnO)
SINTERING
S02
SINTERED MATERIAL
REDUCTION
i
METALLIC
ZINC
Figure 9.8.8-2 Primary electrolytic zinc smelting process.1
9.8-156
-------
Table 9.8.8-4. EMISSION FACTORS FOR PRIMARY ZINC
SMELTING WITHOUT CONTROLS*.7
Participates
Type of operation Ib/tonkg/MT
Roasting (multiple-hearth)
Sintering
Horizontal retorts
Vertical retorts
Electrolytic process
120
90
8
TOO
3
60
45
4
50
1.5
^Approximately 2 unit weights of concentrated ore are
required to produce 1 unit weigirt" of zinc metal.
Emission factors expressed as units per unit weight
of concentrated ore processed., .
-------
Table 9.8.8-5. TYPICAL ZINC ROASTING OPERATIONS8
Operating
temperature,
Type of roaster °C
Multiple-hearth ... , 650 to 730
Multiple-hearth9 870 to 900
Roppb . 650
Fluidized-bedc 890
Fluidized-beda 900
Fluidized-bed (Lurgi) 930
Suspension 980
Fluid column 1000
Feed
capacity,
Mg/Day
45 to 110
230
36 to 45
130 to 200
220 to 320
220
110 to 320
200
Dust in
off gas,
% of Feed
5 to 15
. 5 to 15
5
70 to 80
75 to 85
50
50
17 to 18
aFirst stage is a partial roast in multiple-hearth; second stage
is a dry-feed dead roast in fluidized-bed.
^Partial roast.
cSlurry feed.
•9 .-8*158
-------
Table 9.8.8-6.
COMPONENTS OF THE DUST FROM
MULTIPLE-HEARTH, SUSPENSION,
AND FLUIDIZED-BED ROASTERS9
Component
Percent by weight
Zinc
Lead
Sulfur
Cadmium
Iron
Copper
Manganese
T1n;; ,
tlercury
54.02
1.4
7.0
0.41
7.0
0.4
0.21
,0.01
0.03
9,8-159
-------
percent zinc and up to 3 percent lead by weight. Uncontrolled emissions
from horizontal retorts are estimated to be about 8 Ib/ton of concentrate
processed, while emissions from vertical retorts are estimated to be 100
Ib/ton. Emissions from electrothermic reduction furnaces are negligible.
Zinc vapor may either be condensed to produce metallic zinc or oxidized
to produce ZnO product. Because ZnO product is collected in air pollution
control equipment, emissions from any well-operated ZnO production furnace
are relatively mi nor.10
The major sources of fugitive dust are the sintering operation, ore
handling and transfer, and casting of molten zinc. Casting produces nearly
34 percent of all fugitive emissions.4
9.8.8.2 Control Techniques. Fabric filters and electrostatic pre-
cipitators are the most common particulate control techniques used by the
zinc smelting industry. Both fabric filters and precipitators are used to
control particulate from domestic zinc sintering machines. A fabric filter
is normally chosen only when the $03 in the flue gas is low since high $03
concentrations can corrode fabric filter structures and deteriorate the
filter fabric.1 Use of wet scrubbers to control sinter plant particulate
emissions and use of cyclones and electrostatic precipitators in series for
roasting plant particulate emissions control has been reported.11 In addi-
tion, roaster gases are typically treated in a sulfuric acid plant for the
removal of sulfur dioxide.
Fugitive particulate emissions derive from a variety of operations and
processes; their control requires measures applicable to the source. Ore
handling and transfer fugitive emissions can be controlled by using water
sprays and confining emissions by enclosure. Sinter machine fugitive emis-
sions can be controlled by using water sprays, reducing the distance that
the sintered material falls when discharged from the sinter machine, enclos-
ing dusty areas, and collecting emissions by hooding and ducting to a fabric
filter. Emissions from zinc casting are also controlled by collecting the
emissions in hoods and ducting them to a baghouse.
Standards of performance for new and modified zinc smelting facilities,
promulgated on January 15, 1976, limit particulate emissions from zinc sinter-
ing machjnes to 50 mg/dscm (0.022gr/dscf). Emissions of sulfur dioxide
9.8-160
-------
contained in roaster gases, and from any sintering machine that eliminates
more than 10 percent of the sulfur initially contained in the zinc sulfide
concentrates processed, are limited to 0.065 percent by volume (650 parts
per million).
9.8-161
-------
REFERENCES FOR SECTION 9.8.8
1. Background Information for Standards of Performance, Primary Copper,
Zinc and Lead, Volume I: Proposed Standards. U.S. Environmental
Protection Agency. Publication No. EPA-450/2-74-002a. October 1974.
2. Minerals Yearbook. U.S. Bureau of Mines. Washington, DC. 1972.
3. Yearbook of the American Bureau of Metal Statistics for 1971. June
1972.
4. Assessment of Fugitive Particulate Emission Factors for Industrial
Processes. U.S. Environmental Protection Agency. Publication No.
EPA-450/3-78-107. September 1978.
5. Field Surveillance and Enforcement Guide for Primary Metallurgical
Industries. U.S. Environmental Protection Agency. Publication No.
EPA-450/3-73-002.
6. Atmospheric Modeling Data from National Emission Data System (NEDS).
U.S. Environmental Protection Agency. May 1979.
7. Compilation of Air Pollutant Emission Factors. U.S. Environmental
Protection Agency. Publication No. AP-42. August 1977.
8. Weisburd, M. I. Field Operations and Enforcement Manual for Air
Pollution Control, Volume III: Inspection Procedures for Specific
Industries. System Development Corporation. McLean, VA. Publication
No. APTD-1102. August 1972.
9. Technical Guidance for Control of Industrial Process Fugitive Particu-
late Emissions. U.S. Environmental Protection Agency. Publication
No. EPA-450/3-77-010. March 1977.
10. Trace Pollutant Emissions from the Processing of Metallic Ores. PEDCo-
Environmental Specialists, Inc. Cincinnati, OH. Publication No.
PB-238 655. October 1974.
11. Vandegrift, A. E. Particulate Pollutant System Study, Volume III:
Handbook of Emission Properties. Midwest Research Institute. Kansas
City, MO. Publication No. DPB-203 522, APTD 0745. May 1971.
9.8-162
-------
9.8.9 Secondary Aluminum Operations
Secondary aluminum operations entail refining scrap aluminum to produce
specification aluminum or aluminum alloy casting metal. These operations
involve scrap processing (e.g., sweating) and aluminum melting and refining.
Secondary aluminum production accounts for about 24 percent of the
total aluminum sales in the United States. There are about 86 secondary
aluminum production plants in the United States; about 40 companies account
for 95 percent of the country's secondary aluminum production. The major
product is casting material, which accounts for 90 percent of secondary
aluminum sales volume.1
Nationwide particulate emissions in 1977 from sweating and refining/
fluxing operations were estimated at 15,400 Mg/yr.2 in 1973, fugitive
particulate emissions were estimated at 1808 Mg.3
9.8.9.1 Process Description and Emissions. Aluminum scrap processing
involves the separation of aluminum from contaminants or attachments such as
iron, brass, and magnesium. Higher melting materials, such as iron and brass,
are typically separated from aluminum by charging these materials into a
sweating furnace where the aluminum melts and separates from the higher melt-
ing materials and then flows from the furnace into a mold. The higher melt-
ing materials and the hot aluminum dross formed in the furnace are removed
periodically. The hot aluminum dross contains suspended metallic aluminum,
which can be removed and recovered by mechanical separation processes. These
processes can result in emissions of fine dust. Scrap that is contaminated
with large amounts of paint, oil^ or grease is often dried in a rotary dryer
to remove the organic contamination before charging the scrap into a refining
furnace.
9.8.9.1.1 Emission sources. The secondary aluminum industry generates
air emissions primarily from its presmelting preparation and smelting opera-
tions. Presmelting preparation sources of air emissions include high iron
scrap sweating furnaces, borings and turnings scrap dryers, and dry residue
milling machines. Smelting operations that generate air emissions include
charging, fluxing, and demagging. In addition to these main sources, there
are a few miscellaneous sources of air emissions associated with the manufac-
turing process. Table 9,8.9-1 identifies the primary air emission sources
and pollutants.4
9.8-163
-------
Table 9.8.9-1 PRIMARY AIR EMISSION SOURCES AND POLLUTANTS
FOR SECONDARY ALUMINUM MANUFACTURING OPERATIONS4
Process operation
Air emission source
Pollutant(s)
Presmelting preparation High iron scrap
sweating furnace
Borings and turnings
scrap dryer
Smelting operations
Charging
Fluxing
Demagging
Dry residue milling
machines
Furnace forewell
Furnace
Furnace
Unburned hydrocarbons,
carbonaceous particu-
late, inorganic par-
ticulate, CO
Unburned hydrocarbons,
carbonaceous particu-
late, inorganic par-
ti cul ate, CO
Inorganic particulate
Unburned hydrocarbons,
carbonaceous particu-
late, inorganic par-
ticulate, CO
Particulate (such as
CaClg, NaCl)
Chlorine (Cl2), aluminum
chloride (Aids), hydro-
gen chloride (HC1),
magnesium chloride
(MgCl2), inorganic
particulate3
or
Hydrogen fluoride (HF),
aluminum fluoride
magnesium fluoride
(MgF2), inorganic
particulateb
aPollutants resulting from chlorine demagging.
^Pollutants resulting from fluoride demagging.
9.8-164
-------
Presmelting Sources—In theory, an aluminum sweating furnace can be
operated with minor emissions of air contaminants if clean, carefully hand-
picked metal free of organic material is processed. In practice, this
selective operation does not occur and high emissions periodically result
from the furnaces. Smoke is caused by the incomplete combustion of organic
constituents such as rubber, oil and grease, plastics, paint, cardboard,
and paper. Fumes are produced from the oxidation of zinc and magnesium
contaminants.
In order to avoid the production of black smoke during melting and re-
fining, many plants thermally treat their aluminum borings and turnings in a
dryer. These borings and turnings are contaminated with oils and other
organic compounds. The exhausts from these dryers can contain significant
amounts of fume composed of unburned hydrocarbons and other gaseous and par-
ti cul ate carbon compounds.
Residues used by the secondary aluminum industry are generally composed
of 10 to 30 percent aluminum, with occluded aluminum oxide fluxing salts
(mostly NaCl and KC1), dirt and various other chlorides, fluorides, and
oxides. In the dry milling process used to separate metals from nonmetals,
dust (inorganic particulate) is generated at the crusher, in the mill, at
the shaker screen and at points of transfer.
Smelting Sources—The major sources of air emissions from secondary
aluminum smelting operations occur during charging, fluxing, and demagging.
During charging, black carbonaceous smoke is emitted, resulting from incom-
plete combustion of paint, coatings and oil present in dirty scrap. The
smoke consists of oily carbonaceous particulate, submicron carbon and.
inorganic particulate, and hydrocarbons.
Cover fluxes and solvent fluxes are used to remove solid impurities
from the molten bath (such as dirt and various oxides) and also provide
alloying agents for the molten aluminum. Cover flux initiated emissions
include fumes composed of salts and oxides such as sodium chloride (NaCl),
calcium chloride (CaCl2). calcium fluoride (CaF2)> aluminum fluoride (AlFs),
aluminum oxide (A1203), magnesium oxide (MgO), etc. Solvent fluxes cause
the formation of aluminum chloride (Aids) vapor, which condenses to form
submicrometer-sized particulates.
9.8-165
-------
Demagging operations, either performed by the addition of aluminum
fluoride or gaseous chlorine to the molten bath, are a major source
of emissions. Aluminum fluoride addition results in the formation of a magne-
sium fluoride (MgF2) slag and is responsible for the generation of HF (due
to the hydrolysis of h^O and F~). Chlorine addition presents several prob-
lems: (1) free chlorine escape, (2)'aluminum chloride emissions, (3) reac-
tion of the hydroscopic AlCls with the moisture in the air to form HC1, and
(4) magnesium chloride (MgCl2) emissions from the fuming dross. The demagging
operation may be conducted in the charging zone, combustion zone, a third well
within the reverberatory furnace or a separate holding furnace.
Another process has been developed that produces lower emissions during
the fluxing and refining of aluminum alloys. This process utilizes a mixture
of an inert gas (nitrogen or argon) with 3 to 5% dichloro-difluro methane
(CC12F2) commonly known as Freon 12. The use of this mixture appears to
significantly reduce process, fugitive, and visible emissions.5
Miscellaneous Sources--In addition to the presmelting and smelting emis-
sion sources, there are a number of minor sources of air emissions throughout
the secondary aluminum manufacturing process. These minor sources are all
associated with the storing, sorting, loading, and transporting of scrap.
During these operations, fugitive dust is generated which results in some
particulate emissions. These emissions are minor since the majority of the
dust generated is quite large and falls to the ground immediately.
9.8.9.2 Control Techniques. There are three basic types of pollution
control systems used in the control of secondary aluminum operations. These
include wet systems, dry systems, and those systems involving process
adjustments.^
Wet Scrubber—The wet scrubber is very suitable for control of hygro-
scopic emissions from the aluminum smelter, particularly with the chlorination
demagging process. The spray quenching of the hot furnace gases creates steam
which reacts with the A1C13 gas to form hydrated aluminum oxide (A^Os) and HC1
Both of these are relatively easy to remove in an appropriately designed and
operated scrubber.
In order to obtain adequate collection efficiency, the use of high-
efficiency wet scrubbers, with a caustic solution as the scrubbing medium,
9.8-166
-------
has been found effective. Table 9.8.9-2 shows typical test data on the
collection efficiency for various wet scrubbers.? Scrubber collection effi-
ciency depends mainly upon scrubbing ratio [m^ of liquid/nv* to gas (gal/1000
cu ft)], velocity of gas in the scrubber, contact time and pressure drop. The
values given in Table 9.8.9-2 are typical efficiencies and do not reflect the
entire range of results.
Scrubber Plus Baghouse—To more efficiently control particulate emissions
from chlorinating aluminum, the wet scrubber may be followed by a baghouse or
electrostatic precipitator. At present, the trend in control equipment for
aluminum-fluxing emissions appears to be away from electrical precipitators and
toward the scrubber-baghouse combination. More precise controls need to be
developed for this technique to adapt the scrubber to the peak chlorine "burn-
off" cycle when the temperature of the aluminum furnace is raised to volatilize
all residual chlorine from the system.
Conventional Baghouse—Conventional baghouses (fabric filters) are used
extensively in the secondary aluminum industry for handling dust that emanates
from several of the operations. Hoods are placed over loading and unloading
areas, conveyors, and transfer points in order to capture the particulate
emissions. The dust-laden gas stream is then cleaned via a baghouse. These
systems generally operate in the 95 percent efficiency range (typically 96
percent capture efficiency and 99 percent control efficiency).
Conventional baghouses have not been effective, however, in the removal of
fumes from the demagging operations. Blinding occurs during the collection of
submicron particulates. These particles enter into the interstices of the
weave and create a barrier to gas flow. When blinding occurs, the pressure
drop rises rapidly and gas flow diminishes. Another problem is the presence
of sparks which cause bag perforations.
Coated Baghouse—A new approach for controlling pollution from secondary
aluminum smelters is the use of the coated baghouse. These baghouses have been
used in the primary industry for some time. The basic concept is to coat a
standard baghouse with a material that absorbs and neutralizes the acid gases
while simultaneously filtering out the fine particulates.
Afterburner—Afterburners are also extensively used throughout the
secondary aluminum industry in order to reduce hydrocarbon emissions. After-
burners are typically used for controlling emissions from the dryers and
9.8-167
-------
Table 9.8.9-2. SCRUBBER COLLECTION EFFICIENCY FOR EMISSIONS
FROM ALUMINUM CHLORINE FLUXING?
Scrubber collection efficiencies,3
percent
Contaminants
HC1
C12
Particulate
matter
Slot
Water
90 to 95
30 to 50
30 to 50
scrubber
10% caustic
solution
95 to 99
50 to 60
5U to 60
Packed-col
Water
95 to 98
75 to 85
70 to 80
umn scrubber
10% caustic
solution
99 to 100
90 to 95
80 to 90
aCollection efficiency depends mainly upon scrubbing ratio (liter/cu.
meter), velocity of gas in scrubber, contact time and, to a lesser
extent, on other aspects of the design. These values are typical
efficiencies obtained by actual tests but do not reflect the entire
range of results.
9.8-168
-------
furnaces. When afterburner temperatures are maintained at 1033 to 1144 K
(1400 to 1600°F) and exhaust gas residence times are 1/2 to 1 second,
essentially complete combustion will occur and there are virtually no hydro-
carbon emissions.
9.8-169
-------
REFERENCES FOR SECTION 9.8.9
1. Gerstein, S. M., and M. E. Franza. Control Technology for Secondary
Aluminum Smelters. (Presented at 68th Annual Meeting of the Air Pollu-
tion Control Association. Boston, MA. June 15-20, 1975.)
2. OAQPS Data File of Nationwide Emissions. U.S. Environmental Protection
Agency. Research Triangle Park, NC.
3. Assessment of Fugitive Particulate Emission Factors for Industrial
Processes. U.S. Environmental Protection Agency. Publication No. EPA-
450/3-78-107. September 1978.
4. Compilation of Air Pollutant Emission Factors. U.S. Environmental
Protection Agency. Publication No. AP-42. February 1972.
5. Control of Particulate Emissions in the Primary Nonferrous Metals
Industry - Symposium Proceedings. U.S. Environmental Protection Agency.
Publication No. EPA-600/2-79-211. December 1979.
6. Screening Study on Feasibility of Standards of Performance for Secondary
Aluminum Manufacturing. The Research Corporation of New England.
Wethersfield, CT. EPA Contract No. 68-02-2615. September 1978.
7. Vandergrift, A. E., et al. Particulate Pollutant System Study, Volume
III: Handbook of Emission Properties. Midwest Research Institute.
Kansas City, MO. Publication No. APTD-0745. May 1971.
9.8-170
-------
9.8.10 Secondary Copper Smelting and Alloying
Secondary copper smelting and alloying involves the processing of copper-
containing scrap metal and oxides to produce specification copper or copper
base alloys (e.g., brasses and bronzes). At least one furnace operation
must be used to produce usable secondary copper and copper alloys. Furnace
operations include sweating, burning/incineration, metal oxide reduction
(primarily in blast or cupola furnaces), melting, refining, and alloying of
copper and copper alloys.
The secondary copper smelting and alloying industry is concentrated in
the country's major population centers, primarily in the northeastern, mid-
western, and Pacific coast states. About 60 manufacturers produce approxi-
mately 272,000 Mg of refined brass and bronze ingot.1 Brass and bronze
ingot production experienced a decline after 1966, and excess capacity may
exist in the industry.2 •
Nationwide particulate emissions in 1977 were estimated to be 38 Gg.3
Fugitive emissions in 1976 were estimated at 766 Mg.4
9.8.10.1 Process Description. Of the five basic processes in this
industry, namely, scrap pretreatment, smelting, refining, alloying, and
pouring, only scrap pretreatment can be performed without the use of furnaces.
Certain types of scrap can be pretreated by hand sorting and/or mechanically/
magnetically removing iron contaminants. Other types of scrap may require
furnace treatment, such as the sweating of -lead-covered copper cable, the
incineration of plastic coated copper wire and cable, and the vaporization
of oil from chips and borings in heated kilns. Recently, industry has been
charging without pretreatment to take advantage of the Btu value of the
coatings, oil, etc., in reducing energy requirements.
Low grade copper scrap, including copper oxide, is usually smelted in a
blast furnace or cupola to reduce the oxide to a metallic copper-containing
material called black copper. This material may be further processed and
ultimately electrolytically- or fire-refined to produce a high purity metallic
copper. These copper refining processes are essentially the same as those
used in primary copper refining. If the blast furnace copper is to be used
in copper alloy production, further refining and alloying is performed in
other furnaces.
9.8-171
-------
To produce copper base alloys, pretreated scrap and/or blast furnace
copper is then melted in direct-fired reverberatory furnaces, electric induc-
tion furnaces, or indirect-fired crucible furnaces. The melted charge can
be refined (i.e., impurities removed or excess alloying elements removed) by
the use of fluxes and/or by the addition of oxygen to the molten bath to
oxidize and separate from the bath such materials as iron, manganese, silicon,
aluminum, and zinc. Alloying materials such as tin, zinc, and silicon may
then be added to bring the alloy within specification. The alloy is then
poured into ingots, or, in some cases, castings are made. Figure 9.8.10-1
shows the raw material and product flow for the secondary copper industry.
9.8.10.2 Emission Characteristics and Applicable Control Technologies.
Typical components of the particulate matter emitted from secondary copper
production operations include zinc and lead oxides, copper, and flyash;
depending on the metal alloy composition, silica, tin, cadmium, and copper
compounds may also be present. The metallic fumes are submicrometer in
size (0.03 to 0.5 micrometers) and agglomerate readily.5 Particulate matter
collected in a secondary brass and bronze smelter baghouse was analyzed and
the resulting chemical component breakdown is shown in Table 9.8.10-1.6
Fugitive particulate emissions result from all furnace operations. The
majority of emissions result from wire insulation burning and from reverbera-
tory furnace charging and tapping operations.4
Fabric filters are the principal devices used to control particulate
emissions from secondary brass and bronze furnace operations. The furnace
exhaust gases must be cooled sufficiently to prevent damage to bags. Venturi
scrubbers are also used, but to a limited extent, and electrostatic precipi-
tators are even less frequently used.
Scrap pretreatment operations such as kiln drying of oil/organic contami-
nated chips and borings and blast furnace operations can result in the emis-
sion of significant amounts of oily and sticky particulate matter. Oily and
sticky material should be incinerated, particularly if the furnace exhaust
is fed to a baghouse, because such material can blind the bag fabric.
Dacron is the most widely accepted fabric for bag construction and can
be used if gas temperatures entering the fabric filter are below 149°C.
Glass fabric bags can be used with gas temperatures as high as 260°C. A
9.8-172
-------
Depleted Slog
(Sell or landfill)
i
BLAST OR CUPOLA
MELTING FURNACE
Residual to
Low Grade. Scrap
INTERMEDIATE GRADE SCRAP
Total » 37 Classification,9.3.
Red Bras, Yellow Bra«i
Auto Rodiaton
Sludges to Free.
Met. Recov, Low
Grade Scrap, or
Sell
/No. 1 Copper Wire
I No. 1 Heavy Copper
(5)
(?)
No. 2 Copper
No. 2 Heavy Copper
Light Copper
SHIP
FIRE REFINED
COPPER INGOTS,
4 BILLETS
«• Residues to Low Grade Scrap
Figure 9.8.10-1
Raw material and product flow diagram for the secondary
copper industry.
9.8-173
-------
Table 9.8.10-1. COMPOSITION OF PARTICULATE MATTER COLLECTED FROM
A SECONDARY BRASS AND BRONZE SMELTER6
Particulate composition,
Component percent by weight
Zinc 45.0 to 77.0
Lead 1.0 to 12.0
Tin 0.3 to 2.0
Copper 0.05 to 1.0
Chlorine 0.5 to 1.5
Sulfur 0.1 to 0.7
9.8-174
-------
filtering velocity of 0.6 m/min is acceptable, and a pressure drop of 1 kPa
is common.1 An exhaust duct velocity of 760 to 950 m/min is desirable to
limit dust deposition in ducting.^ Baghouse particulate collection efficien-
cies of 93.7 and 96.2 percent are reported for crucible furnace operations,
and a collection efficiency of 96.0 percent is reported for an electric
furnace operation.5
Venturi scrubbers can have pressure drops that vary between 7.5 to 25 kPa.
A 15 kPa pressure drop corresponds to a throat velocity of about 61 m/s and
requires water at a rate of 0.4 L/s per cubic meter/min of gas.l Particulate
collection efficiencies for three scrubbers serving brass furnaces range from
53 to 63 percent.!
A schematic of a baghouse control system serving typical secondary brass
and bronze furnace operations is shown in Figure 9.8.10-2. Table 9.8.10-2
lists the types of emission control equipment being used by this industry.
The average stack parameter data for the secondary copper industry is pre-
sented in Table 9.8.10-3.7
Table 9.8.10-4 provides data from various furnaces at a typical secondary
copper smelting and refining plant.8 Table 9.8.10-5 provides a summary of
measured emission data from the various furnace process control systems of a
few plants in the United States.8
Fugitive emissions result primarily from furnace charging, metal tapping,
metal casting, and wire pretreatment incineration operations. Control tech-
niques include using properly designed collection hoods with adequate air
flow, maintaining molten copper alloys at the lowest possible temperature to
prevent the emission of low boiling constituents (e.g., zinc), and increasing
the exhaust rate of incineration control systems.6
9.8-175
-------
VQ
00
= EMISSION POINTS
REVERBERATORY
FURNAtt
- ELECTRIC
CRUCIBLE FURNACE
Figure 9.8.10-2 Air pollution control system in the brass and
bronze industry,*
-------
Table 9.8.10-2. SUMMARY OF AIR POLLUTION CONTROL EQUIPMENT IN USE IN
THE SECONDARY COPPER INDUSTRY1
Furnace
or facility
type
Reverberatory
Rotary
Electric
Crucible
Sweat
Cupola
Wire burning
Rotary dryer
Incinerator
Raw material
concentrator
Slag furnace
Number
18
12
16
. 9
5
2
16
' 4
1
4
1
1
1
1
1
1
2
1
Control equipment
Baghouse
No controls
Baghouse
No controls
Baghouse
Scrubber
No controls
No controls
Afterburner &
baghouse
Baghouse
No controls
Wet collector
Afterburner
chamber
No controls
Afterburner
Cycl one
No controls
Baghouse
9.8-177
-------
Table 9.8.10-3. SECONDARY COPPER INDUSTRY STACK PARAMETER DATA?
00
I
00
Facility or
operation
Blast furnace
Crucible furnace
Cupola furnace
Electric induction
furnace
Reverberatory
furnace
Nationwide no. of
facilities or
operations
9
97
26
90
45
Uncontrolled
emission factor,
kg of parti cul ate
per Mg of metal charged
18.0
12.0
73.0
2.0
70.0
Average
stack
height,
m
25
15
27
15
19
Average
stack
diameter,
m
1.2
1.1
0.98
1,1
0.98
Average
stack
temp.,
°C
42
131
146
127
403
Average
gas
flow,
-Am3/s
5.57
20.3
6.94
17.3
10.3
Average
operating rate
Gg of metal charged
to each
furnace per year
2.59
6.92
18.1
16.1
8.69
Rotary furnace
32
60.0
17
1.2 178 12.2
2.75
-------
TABLE 9.8.10-4. OPERATING SPECIFICATIONS AND UNCONTROLLED AND CONTROLLED EMISSION RATES
FOR THE TYPICAL SECONDARY COPPER SMELTING AND REFINING PLANTS
00
Parameters
Net Cu content
production,
Mg/yr (tons/yr)
Total process
charge rate,
Mg/hr (tons/yr)
Total hours of
operation
Gas effluent
rate,
ra3STP/min (SCFM)
Gas temperature,
°C (°F)
Stack height,
n (ft)
Fuel or charge
Furnace types
Cupola
22,680
(25,000)
8.2
(9.0)
8,400
1,698
(60,000)
93
(200)
23
(75)
coke
Rotary
converter
22,680
(25,000)
3.4
(3.7)
8,400
1,132
(40,000)
127
(260)
23
(75)
fuel oil
Reverberatory
anode furnace No. 1
22,680
(25,000)
4.5
(5.0)
4,200
682
(24,100)
78
(173)
17
(55)
Fuel oil
Natural gas
Coke
Reverberatory
anode furnace No. 2
22,680
(25,000)
4.5
(5.0)
4,200
682
(24,100)
78
(173)
17
(55)
Fuel oil
Natural gas
Coke
Reverberatory
fire refining furnace
9,070
(10,000)
1.1
(1.2)
8,400
566
(20,000)
78
(173)
17
(55)
Fuel oil
Natural oil
Coke
Shaft
furnace
54,430
(60,000)
6.5
(7.1)
8,400
730
(25,800)
247
(476)
17
(55)
Natural
gas
-------
TABLE 9.8.10-4. (concluded)
00
00
o
Parameters
Burning rate,
GJ/Hg
(106 Btu/ton)
of Cu content
Uncontrolled
participates
emission fac-
tor, kg/Hg
(Ib/ton) of
charge
Uncontrolled
particulates,
kg/hr (Ib/hr)
Controlled
particulates,
kg/hr (Ib/hr)
Sulfur
emissions,
kg/hr (Ib/hr)
Furnace types
Cupola
23.45
(20.16)
17
(33)
136
300)
4.1
(9.0)
3.6
(8.0)
Rotary
converter
3.99
(3.43)
22
(45)
76
(167)
2.3
(5.0)
23
(50)
Reverberatory
anode furnace No. 1
1.42 (1.22)
0.27 (0.23)
0.33 (0.28)
12
(24)
54
(119)
1.8
(4.0)
4.5
(10)
Reverberatory Reverberatory Shaft
anode furnace No. 2 fire refining furnace furnace
1.42 (1.22) 1.42
0.27 (0.23) 0.27
0.33 (0.28) 0.33
12 20
(24) (40)
54 22
(119) (48)
1.8 0.5
(4.0) (1.0)
4.5 1.8
(10) (4.0)
1.22) 2.27
0.23) (1.95)
0.28)
0.5
(1.0)
2.6
(5.7)
1.1
(2.4)
0
-------
TABLE 9.8.10-5.
SUMMARY OF AVAILABLE EMISSION DATA
BY PLANT AND EMISSION SOURCE8
Control
system
Baghouse
Baghouse
Baghouse
Baghouse
Baghouse
Baghouse
Baghouse
Baghouse
Settling chamber
Baghouse
None
High energy venturi
with mist eliminator
Baghouse
Baghouse
None
None
Baghouse
Baghouse
Baghouse
None
None
Type of
furnace
Cupola
Slag cleaning
Converter
Reverberatory
Reverberatory
Reverberatory
Reverberatory
Reverberatory
Shaft
Cupola
Holding
Kaldo
Kaldo
Reverberatory
Reverberatory
Billet
Shaft
Cupola
Converter
Reverberatory
Reverberatory
Reverberatory
Particulates,
kg/hr (Ib/hr)
16.3 (35.9)
4.9 (10.8)
2.8 (6.2)
0.6 (1.4)
4.5 (9.8)
2.9 (6.4)
0.3 (0.7)
1.1 (2.5)
1.5 (3.2)
4.6 (10.1)
0.3 (0.6) c
8.1 (17.8)
8.1 .(19.2)
1.7 (3.7)
4.3 (9.4)
4.8 (10.5)
1.2 (2.7)
4.1 (9.0)
2.3 (5.0)
4.5 (10.0)
15.1 (33.4)
30.2 (66.6)
9.8-181
-------
REFERENCES FOR SECTION 9.8.10
1. Secondary Brass and Bronze Ingot Production Plants. Inspection Manual
for Enforcement of New Source Performance Standards. U.S. Environmental
Protection Agency, Division of Stationary Source Enforcement. Washington,
DC. Publication No. EPA-340/1-77-003. January 1977.
2. Background Information for Proposed New Source Performance Standards.
U.S. Environmental Protection Agency, Office of Air Quality Planning and
Standards. Research Triangle Park, NC. Publication No. APTD-1352a.
June 1973.
3. OAQPS Data File of Nationwide Emissions. U.S. Environmental Protection
Agency, Office of Air Quality Planning and Standards. Research Triangle
Park, NC. June 1973.
4. Zoller, T., T. Bertke, and T. Janszen. Assessment of Fugitive Particulate
Emission Factors for Industrial Processes. U.S. Environmental Protection
Agency, Office of Air Quality Planning and Standards. Research Triangle
Park, NC. Publication No. EPA-450/3-78-.107. September 1978.
5. Vandegrift, A. E., et al. Particulate Pollutant System Study, Volume III:
Handbook of Emission Properties. Midwest Research Institute. Kansas City,
MO. Publication No. PB-203-522. May 1971.
6. PEDCo Environmental, Inc. Technical Guidance for Control of Industrial
Process Fugitive Particulate Emissions. U.S. Environmental Protection
Agency. Publication No. EPA-450/3-77-010. March 1977.
7. Atmospheric Modeling Data from National Emission Data System (NEDS). U.S.
Environmental Protection Agency, Office of Air Quality Planning and
Standards. Research Triangle Park, NC. May 1979.
8. Source Category Survey: Secondary Copper Smelting and Refining Industry.
U.S. Environmental Protection Agency, Office of Air Quality Planning
and Standards. Research Triangle Park, NC. Publication No. EPA-450/
3-80-011. May 1980.
9.8-182
-------
9.8.11 Secondary Lead Smelting
Secondary lead smelting involves the processing of scrap lead (primarily
from recycled car batteries) and scrap lead oxides to produce usable lead
and lead alloys that conform to specified composition requirements. In the
United States at the end of 1975, there were approximately 115 secondary
lead smelting plants operated by 85 firms. Two of these firms accounted for
over half of the total secondary lead output. The industry has experienced
an average annual growth rate of about 1.4 percent (Bureau of Mines, 1975).
During 1977, 543,000 to 544,000 Mg of secondary lead were recovered by pro-
cessing scrap lead materials.1 Secondary lead production is used primarily
in the manufacture of lead-acid storage batteries. Secondary lead is also
used in the manufacture of such products as solder and other lead alloys.
The particulate matter emitted from secondary lead processing operations
typically consists of PbO, SnO, ZnO, tar, flyash, coke dust, sulfates, and
sulfides. The metallic fume particulate emissions are submicrometer in size
with a mean diameter of about 0.3 micrometers for unagglomerated material.2
Nationwide particulate emissions from blast and reverberatory furnaces in
1977 were estimated by EPA to be 2180 Mg.3 Uncontrolled fugitive particulate
emissions were estimated in 1973 to be 4250 Mg.l
9.8.11.1 Process Description and Emissions. In the secondary smelting,
refining, and alloying of lead, the three types of furnaces most commonly
used are the blast or cupola, reverberatory, and pot. The grade of metal
produced (soft, semisoft, or hard) dictates the type of furnace to be used.
Scrap lead and lead-based metals are melted and the scrap oxides reduced to
their metallic state. The metal is refined (i.e., impurities removed) and
the lead or lead-based alloy is brought into specification by the addition
of appropriate alloying elements or the metal is oxidized to produce a
specified lead oxide material. The secondary lead smelting process is illus-
trated in Figure 9.8.11-1.4
Scrap lead and lead oxides used to produce "hard" lead are fed into a
blast furnace (or cupola), along with other charge materials including coke
and limestone. The charged coke is ignited and air is forced into the fur-
nace bottom and up through the charge material, which is continuously fed
into the furnace. The charges move down the,vertical- furnace as the lead is
9.8-183
-------
LEAD HOLDING,
MELTING
AND REFINING POTS
BLAST AIR
SLAG
BLAST FURNACE
CHARGE
TO BLAST FURNACE
CONTROL SYSTEM
.SLAG
TO VENTILATION
CONTROL SYSTEM
GAS OR
FUEL OIL
*!/
=v4 J
V^ ^
u
TO REVERBERATORY
FURNACE
CONTROL SYSTEM
LEAD
Figure 9.8.11-1 Secondary lead smelter process.14
9.8-184
-------
melted or reduced and the other charge materials are combusted, reacted, or
combined with the slag. The molten lead, as well as a slag material, are
tapped (removed) from the bottom of the furnace. Particulate emissions
consist of particles ranging in size from 1 to 100 micrometers.5 These
particles contain about 23 percent lead.6
Reverberatory furnaces, like blast-furnaces, are used for the production
of semi soft lead. The furnaces are charged with scrap lead and lead oxides,
along with other materials such as coke and limestone. Reverberatory fur-
naces can also be used to melt or sweat lead from lead-coated cable and other
lead-coated materials. Fossil fuel-fired burners are used to heat reverbera-
tory furnaces, and the furnaces are continuously charged. The lead is melted
or reduced, and the charge materials are combusted, reacted, or combined with
the slag. In the case of the sweating operation, the cable or other non-^lead
materials are removed and recovered. The semi soft lead is tapped from the
bottom of the furnace, and the slag covering is skimmed off the molten lead.
Particulates emitted from this operation are extremely fine (0.3 micrometers).
Pot furnaces are used for remelting, alloying, and refining and are
charged with scrap lead and/or lead tapped from blast or reverberatory fur-
naces. These furnaces are heated indirectly, usually by gas. In the refining
process, aluminum and other compounds are added to the metal to react with
impurities (e.g., antimony, arsenic, nickel, copper) which form complex com-
pounds that are skimmed from the metal surface. Since the pot furnace is
not direct-fired, and the charge is not subject to the same degree of turbu-
lence as in blast and reverberatory furnaces, the quantity of particulate
matter emitted is much less.
Lead oxide is produced from refined lead by the Barton.Process for use
in the manufacture of lead batteries, red lead, and paint. In this process,
air is drawn through agitated molten lead. The resulting lead-oxide contain-
ing air stream is exhausted through a baghouse in which the lead oxide is
captured. This process, by its nature, results in few emissions.2»7
Only limited data are available regarding quantitative estimates of
fugitive particulate emissions from secondary lead smelters. The limited
data do suggest that industrial process fugitive emissions are a factor in
contributing to elevated ambient particulate concentrations near secondary
lead smelters.6
9.8-185
-------
9.8.11.2 Particulate Control Techniques. Well controlled secondary
lead smelters reduce particulate emissions by the use of either fabric filters
or high-energy scrubbers, with the most widely used device being fabric fil-
ters. The control system selected is generally dependent upon the type of
furnace employed.
Blast furnace emissions contain metal oxides, bits of coke, and other
particulate matter present in the charge. Since these emissions may contain
oily and sticky particles, blast furnace fabric filters are usually preceded
by an afterburner to incinerate these materials. If allowed to enter the
filter units, oily and sticky particulate matter could blind the fabric. An
afterburner is not needed in the reverberatory furnace control system, since
the excess air and temperature are sufficient to incinerate these particles.
The gas temperature entering a fabric filter must not exceed about
149°C if Dacron bags are used or about 260°C if fiberglass bags are used.5
For blast furnace/afterburner and reverberatory furnace exhaust, it is usually
necessary to cool the gases by use of radiant cooling columns, evaporative
water coolers, or air dilution jets. A schematic of blast and reverberatory
furnace control systems is shown in Figure 9.8.11-2.
For baghouse control systems, a system pressure drop of up to 1 kPa is
common, and a gas volume to bag cloth area of 2 to 1 is the accepted ratio
to ensure efficient operation of the collection system. For venturi scrubber
control systems, the system pressure drop can vary between 7.5 and 25 kPa.
A 15 kPa pressure drop corresponds to a throat velocity of about 61 m/sec
and requires water at a rate of 0.39 liters/minute per m-Vmin of gas.5 Bag-
house particulate collection efficiencies of greater than 99 percent and 98
percent, for reverberatory and blast furnaces, respectively, have been
reported.2
Particulate emissions measured by EPA of three well controlled blast
furnaces averaged 5.9 mg/DNCM (0.09 kg/Mg lead produced), 18.1 mg/DNCM
(0.30 kg/Mg) and 32.7 mg/DNCM (0.57 kg/Mg) for facilities controlled with
afterburner/baghouse, afterburner/baghouse/venturi scrubber, and venturi
scrubber respectively.4 Baghouse controlled particulate emissions measured
by EPA of two well controlled reverberatory furnaces average 8.1 mg/DNCM
(0.11 kg/Mg lead produced) and 7.6 mg/DNCM (0.13 kg/Mg lead produced).4
9.8-186
-------
EMISSIONS
TO VENTILATION SYSTEM
JDDDG
BLASTFURNACE AFTERBURNER COOLING TOWERS f? BAGHOUSE|| FAN
COOLING BLEED AIR_y
DUST RECYCLED TO REVERBERATORS FURNACE
(a) Controlled lead blast furnace, afterburner and baghouse.
EMISSIONS
TO VENTILATION SYSTEM
REVERBERATORY FURNACE
DUST RECYCLE ,
(b) Controlled lead reverberatory furnace, baghouse.
Figure 9.8.11-2 Control sbhemes for secondary lead (a) blast and
(b) reverberatory furnaces.^
9.8-187
-------
Baghouses and scrubbers are also used to control pot furnaces (Figure
9.8.11-3). During melting and holding operations associated with pot fur-
naces, uncontrolled emissions are quite low because the vapor pressure of
lead is low at the melting temperature. During dross skimming and refining
operations, emissions are substantially increased, and adequate ventilation
must be provided to protect the health of the workers. The latter require-
ments govern the volume of exhaust gases.4
Emissions from blast and reverberatory furnaces are normally released
into the atmosphere through stacks with an average height of 90 to 100
meters; stacks from pot furnaces are shorter, averaging 45 meters. Stack
parameters from lead smelting emission sources are summarized in Table
9.8.11-1.8
Industrial process fugitive particulate emissions result from the re-
ceiving and handling of coke, limestone, and secondary lead materials, the
sweating of lead scrap, furnace charging and tapping operations, and lead
casting operations. Control of such emissions can be accomplished by apply-
ing a water spray during materials receiving and handling; improved hooding
and ducting, and better control of operating parameters and procedures for
charging, tapping, and sweating operations.9
9.8.11.3 Secondary Environmental Impacts. The predominant control
devices for the secondary lead industry are fabric filters, along with a
small number of high-energy scrubbers. Dust collected in baghouses can be
recycled directly back to the furnace. When wet scrubbers are used, settling
tanks and ponds are installed to precipitate the collected solids. The pre-
cipitate is removed, dried, and fed back to the furnace. Scrubbing water
will pick up sulfur dioxide from the gas stream, causing the water to become
acidic. Alkali can be added to the scrubber to control pH. Salts that pre-
cipitate with collected dust are also returned to the furnace and usually
become part of the slag.
9.8-188
-------
FROM FURNACES 1
\J
U\J
HOLDING, LEAD MELTING,
AND REFINING POTS
DUST RECYCLED TO REVERBERATORY FURNACE
t
FAN
Figure 9.8.11-3 Controlled lead pot and ventilation system, baghouse.1'
9.8-18*
-------
Table 9.8.11-1. STACK PARAMETERS FOR EMISSION SOURCES WITHIN THE
LEAD SMELTING INDUSTRY8
00
I
Facility or
operation
Pot furnace
Reverberatory
furnace
Blast (cupola)
furnace
Nationwide number
•of facilities or
operations
110
48
36
Average
stack ht.,
m
14
26
29
Average
stack dia.,
m
0.85
1.07
0.98
Average
temp.,
°C
133
116
81
Average
flow,
Am3/s
6.24
10.2
10.5
Average
operating rate,
Gg of metal charged
per furnace per year
9.03
35.8
15.4
Rotary
reverberatory
furnace
30
2.29
66
14.6
2.42
-------
REFERENCES FOR SECTION 9.8.11
1. Assessment of Fugitive Participate Emission Factors for Industrial
Processes. U.S. Environmental Protection Agency. . Publication No.
EPA-450/3-78-107. September 1978.
2. Vandegrift, A. E., et al. Particulate Pollutant System Study, Volume
III: Handbook of Emission Properties. U.S. Environmental Protection
Agency. Publication No. APTD-0745. May 1971.
3. OAQPS Data File of Nationwide Emissions. U.S.- Environmental Protection
Agency, Office of Air Quality Planning and Standards. Research Triangle
Park, NC. 1979.
4. Background Information for Proposed New Source Performance-Standards,
Volume 1: Main Text. U.S. Environmental Protection Agency. Publication
No. APTD-1352a. June 1973.
5. Inspection Manual For Enforcement of New Source Performance Standards
-- Secondary Lead Smelters. U.S. Environmental Protection Agency.
Publication No. EPA-340/1-77-001. January 1977.
6. A Method for Characterizations and Quantifications of Fugitive Lead
Emission from Secondary Lead Smelting, Ferroalloy Plants and Gray Iron
Foundaries (Revised). U.S. Environmental Protection Agency. Publication
No. EPA-450/3-78-003. August 1978.
7. Field Operations and Enforcement Manual For Air Pollution Control, Volume
III: Inspection Procedures for Specific Industries. U.S. Environmental
Protection Agency. Publication No. APTD-1102. August 1972.
8. Atmospheric Modeling Data from National Emission Data Systems (NEDS).
U.S. Environmental Protection Agency, Office of Air Quality Planning
and Standards. Research Triangle Park, NC. February 19,79.
9. Technical Guidance for Control of Industrial Process Fugitive Particulate
Emissions. U.S. Environmental Protection Agency. Publication No.
EPA-450/3-7-010. March 1977.
9.8-191
-------
9.8.12 Secondary Zinc Processing
Secondary zinc processing involves the use of one or more furnace re-
covery operations that process scrap zinc materials into usable zinc products.
Zinc is separated from scrap that contains lead, copper, aluminum, and iron
by carefully controlling the temperature in a sweating furnace and allowing
each metal to be removed at its melting temperature. Further refining of
the zinc can be performed in retort distilling or vaporization furnaces,
where the vaporized zinc is condensed to pure metallic form. Zinc galvaniz-
ing is sometimes included as a secondary zinc process, even though this opera-
tion involves zinc consumption rather than zinc recovery. For purposes of
this analysis, zinc galvanizing is not considered as a secondary zinc pro-
cessing operation.
Nationwide particulate emissions in 1977 from distillation and sweating
operations were estimated to be 1270 Mg.l A worst case estimate of secondary
zinc industry process fugitive emissions in 1973 was calculated to be as high
as 429 Mg.2
9.8.12.1 Process Description and Emissions. Secondary zinc processing
basically involves three operations: pretreatment (primarily sweating),
melting, and vaporization. Sweating is the scrap pretreatment method most
often used; it involves furnace operations that separate metallic zinc from
metals with higher melting points and other contaminants. Sweating can be
accomplished by using reverberatory, pot, rotary, muffle, or electric arc
furnaces. Higher melting materials are separated from the molten zinc,
while impurities such as plastic and oil are volatilized. The molten zinc
is then tapped (removed) from the furnace.
Sweating operations release air contaminants that consist mainly of
smoke and metal fumes. The smoke results from incomplete combustion of scrap
impurities such as oil and plastics, and the fumes consist primarily of vola-
tilized and condensed zinc compounds. If the furnace is operated properly at
low temperatures and the zinc does not reach a temperature much above 590°C,
heavy fuming will not result.3 size data from sweating operations is sparse
but it appears that the size distribution of the emitted particulate matter
is somewhat dependent upon the composition of the materials charged to the
furnace. The fumes generated during pot furnace sweating may have a geometric
9.8-192
-------
mean size of less than 2 micrometers.4 An analysis of a participate emissions
sample from the sweating of metallic scrap has shown 4 percent zinc chloride,
77 percent zinc oxide, 4 percent water, 4 percent other metal chlorides and ,
oxides, and 10 percent carbonaceous materials.5
The recovered molten zinc from sweating operations can then be fed
directly (or cast and then fed) to processing/melting furnaces or to a vapori-
zation furnace. The molten zinc may also be analyzed chemically, and appro-
priate alloying materials added to obtain a desired composition prior to its
being cast into salable ingots.
Skimmings scrap (residual scrap containing zinc oxide and zinc dross)
is not sweated, but is typically crushed and screened to separate metallic
constituents from pulverized nonmetallic materials. The metallics are then
used in the melting or vaporization process. Skimmings may also be physically
and chemically treated, resulting in the production of a Zn(OH)2 compound
which is further treated in a calcining furnace to produce zinc oxide. The
zinc oxide, in turn, is collected in polyester bag filters for further pro-
cessing. Collection of this material apparently poses no emissions problem.4-
In. the melting operation, zinc recovered from sweating and/or scrap
zinc is fed into a pot, reverberatory, or electric induction furnace and
melted. A fluxing compound is usually added to remove impurities. The
metal can then be fed directly to a vaporization type furnace and cast into
ingots; or the zinc can be chemically analyzed, brought into specification
by the addition of alloying materials, and then cast into salable ingots.
9.8.12.2 Emission Characteristics and Applicable Control Technologies.
Particulate emissions result from the melting operation when the charge con-
tains organic material. Emissions can also be caused by fluxing, but fluxes
are currently available whose use does not result in particulate emissions.
Metal temperatures above 593°C will result in excessive zinc vaporization
and the formation of particulate zinc oxide or zinc fumes. Normally, the
zinc can be treated and poured at temperatures below 593°C.3
Zinc is vaporized and condensed in either a furnace .(Belgian or
distillation-type) or muffle furnace condensation system. In this operation,
impure zinc metal scrap, molten or cast metal obtained from the sweating
and/or melting operation, or a zinc oxide preparation, is charged into the
9.8-193
-------
furnace. The charge is then heated indirectly (see retort and muffle furnace
schematics, Figures 9.8.12-1, 9.8.12-2, 9.8.12-3), resulting in the formation
of high purity vapors. The vapors are then condensed to form metallic zinc
or oxidized in the muffle and retort furnaces to form high purity zinc oxide.
The retort furnaces are operated on a batch basis while the muffle furnace
can be operated continuously for several days.
Particulate matter is emitted from the retort furnaces when residues are
removed and fresh batches are charged. These emissions consist mostly of
zinc oxide, with aluminum, copper, and other metals also present; the par-
ticles range in size from 0.05 to 1.0 micrometers.5 Zinc vapors, that leave
the muffle furnace condenser, form zinc oxide and are collected in a product
collection system. Average stack parameters for emission sources within the
secondary zinc processing industry are summarized in Table 9.8.12-1.6 Esti-
mates of uncontrolled emission factors for various secondary zinc smelting
sources are presented in Reference 7.
Fugitive emissions are produced in the skimming, crushing, and screening
operations, as well as in the various furnace charging and tapping operations.
In the zinc vaporization/condensation process, emissions can also result from
condenser upset conditions and from leaks between the vaporization furnace
and the condenser.
Sweat process emissions are alleviated by selection of processes that
appear optimum for the type of scrap being treated and by application of
established types of gas cleaning equipment. Afterburners are used to in-
cinerate combustible material emitted from low temperature sweating furnaces.
The incinerator exhaust, after being cooled, is then treated in a fabric fil-
ter. Design and operating characteristics of a fabric filter installed on a
sweating operation and test results are shown in Table 9.8.12-2.
Fabric filters are also used to control participate emissions from re-
verberatory and distillation retort furnaces. Experience in fabric filter
applications indicates polyester fabric is the most acceptable bag material
for distillation and sweating operations where the chloride concentration is
not excessive. A maximum zinc chloride content of between 2 and 5 percent
at an effluent temperature of 120°C (250°F) and a maximum air-to-cloth ratio
of 0.6 to 0.3 is recommended.4 By maintaining this temperature and a low
9.8-194
-------
FRONT WALL
OF FURNACE
10
bo
i
o
OT
GROUT JOINT
CONDENSED METAL
VAPORS
CERAMIC CONE
CONDENSER
POROUS LOOM
STUFFING
REAR WALL
OF FURNACE
CERAMIC RETORT
METALLIC OXIDE CHARGE
WITH REDUCING MATERIALS
FLAME FROM
COMBUSTIBLE
GASES
BURNER PORT
Figure 9.8,12-1 Diagram showing one bank of a Belgian retort furnace.3
-------
SPEISS HOLE
ID
00
i i i i i i i ' I '
' ' i i i i i i i
' i i i i i ' i i
Figure 9.8.12-2 Diagram of a distillation-type retort furnace;3
-------
STACK
CO
MOLTEN
METAL
TAP
HOLE
BURNER PORT
I.I.I
JL
I I
I I I I
I.I I I I I I I I. I . I
I I I I I I ' 1 I
MOLTEN METAL
METAL VAPORS __
MUFFLE —. /
•*_*.•-***• * •• **••/•.'/.*.*-*_•**...*'-.-•.*«-•.*• /-'•*-*.'/•*.'•'•'.'//*
1.1.1,1.1
11,1.1.1
'I ' '
i i r i i r~ i i i
iii i
JL_J_L_J_I
FLAME PORT
AIR IN
DUCT FOR
OXIDE
COLLECTION
RISER
CONDENSER
UNIT
MOLTEN
METAL
TAP HOLE
Figure 9.8.12-3 Diagram of a muffle furnace and condenser.3
-------
Table 9.8.12-1. STACK PARAMETERS FOR EMISSION SOURCES WITHIN THE
SECONDARY ZINC PROCESSING INDUSTRY6
10
00
I
Facility or operation
Retort furnace
Horizontal muffle. furnace
Pot furnace
Kettle sweat furnace
Calcining kiln
Reverberatory sweat furnace
Nationwide number
of facilities or
operations
9
9
20
12
6
14
Average
stack ht.,
m
12
5
8
16
49
8
Average
stack diam. ,
m
0.46
0.73
0.67
1.28
1.92
0.43
Average
temp.,
°C
128
1054
169
123
286
566
Average
flow (ACFH)
Am3/s
4.25
3.19
4.02
3.37
57,3
0.25
Average
operating rate
Mg of Zn metal
produced per year
2.07
3.62
8.08
2.34
27.9
4.28
-------
Table 9.8.12-2. ZINC SWEATING FURNACE CONTROL INFORMATION3
Furnace data
Type of furnace Reverberatory
Process weight, Mg/h 4.6
Material sweated Zinc castings
Baghouse data
Type of baghouse Sectioned tubular
Filter material Orion
Filter area, m2 482
Filter velocity, m/min 0.56
Precleaner None
Dust and fume data
Gas flowrate, Nm3/min
Baghouse inlet 715
Baghouse outlet 690
Average gas temperature, °C
Baghouse inlet 88
Baghouse outlet 78
Concentration, g/m3
Baghouse inlet 0.143
Baghouse outlet 0.0054
Dust and fume emission, kg/h
Baghouse inlet 6.1
Baghouse outlet 0.23
Control efficiency, % 96.3
9.8-199
-------
percentage of chlorides in the particulate matter being collected, blinding
of the bags and caking of the collected material due to deliquescence are
prevented. The bags are satisfactorily cleaned by means of a pneumatic air
shaker.4
Electrostatic precipitators (ESPs) have limited application for control-
ling emissions from sweating operations. The reason for this is that an ESP's
efficient use is limited to a certain range of particle compositions and
effluent temperatures. High zinc oxide concentrations and high temperatures
can reduce the electrical conductivity and cause inefficient operation while
high chloride concentrations cause excessive corrosion.4
A degree of particulate emissions reduction can be achieved by selecting
specific processing procedures and furnaces. For example, scrap containing
considerable organic material can be processed without flux in reverberatory
furnaces, and the resulting sweated crude zinc alloy then distilled to obtain
pure zinc. In this way, carbonaceous and chloride emissions are prevented
and noncarbonaceous emissions (free of chlorides) that occur can be collected
in conventional baghouses.4
Industrial process fugitive emissions can be controlled by improving
operating procedures and by improving particulate capture/control methods.
The use of properly designed hooding devices during the crushing, screening,
charging, alloying, and tapping operations with an exhaust system capture
velocity of about 0.5 to 1.0 meter/sec can effectively control fugitive
emissions.5 Improved maintenance and operating procedures can reduce fugi-
tive leaks from vaporization/condensation operations and from equipment up-
sets and breakdowns. Maintaining zinc temperatures below 590°C will essen-
tially eliminate fugitive zinc fume emissions during casting operations.5
9.8-200
-------
REFERENCES FOR SECTION .9.8.12
1. OAQPS Data File of Nationwide Emissions. U.S. Environmental Protection
Agency, Office of Air Quality Planning and Standards. Research Triangle
Park, NC. 1977.
2. Assessment of Fugitive Particulate Emission Factors for Industrial
Processes. U.S. .Environmental Protection Agency. Publication No. EPA-
450/3-78-107. September 1978.
3. Vandegrift, A. E., et al. Particulate Pollutant System Study, Volume III:
Handbook of Emission Properties. Midwest Research Institute. Kansas
City, MO. May 1971.
4. Secondary Zinc Industry Emission Control Problem Definition Study, Part I:
Technical Study. U.S. Environmental Protection Agency. Research Triangle
Park, NC. Publication No. APTD-0706. April 1971.
5. Technical Guidance for Control of Industrial Process Fugitive Particulate
Emissions. U.S. Environmental Protection Agency. Publication No. EPA-
450/3-77-010. March 1977.
6. Atmospheric Modeling Data from National Emission Data System (NEDS). U.S.
Environmental Protection Agency, Office of Air Quality Planning and Stan-
dards. May 1979.
7. Source Category Survey: Secondary Zinc Smelting and Refining Industry.
U.S. Environmental Protection Agency, Office of Air Quality Planning and
Standards. Research Triangle Park, NC. Publication No. EPA-450/3-8.0-012.
May 1980.
9.8-201
-------
Page Intentionally Blank
-------
9.9 PETROLEUM INDUSTRY
The petroleum industry extracts crude oil from production wells, refines
the oil into more than 300 products, and then markets the products. The par-
ticulate matter emitted by this industry is produced almost exclusively by
catalyst regeneration, coking operations, asphalt air blowing, and sludge
incineration at the refineries. Some particles are also produced by process .
heaters, boilers, and flares. Emissions from these sources include catalyst
fines, coke dust, sulfuric acid mist, asphalt aerosol, flyash, and ,soot.
Many of the emissions are less than 10 micrometers in diameterJ An esti-
mated 83 Gg of particulate matter was emitted by 266 refineries nationwide
in 1977.2,3 /\s Of March 1979, there were 289 refineries in operation.4 A
flow diagram of a complex refinery is given in Figure 9.9-1.5 Emissions
control arid stack parameter data for each of the particulate emission sources
are summarized in Tables 9.9-1 and 9.9-2.6
9.9.1 Catalyst Regeneration Processes
Catalysts are used in many refinery processes such as fluidized catalytic
cracking, reforming, hydrotreating, and alkylation. These catalysts need to
be regenerated to remove coke deposits. Most processes require regeneration
only periodically—anywhere.from once per week to once every other year. A
fluidized catalytic cracking (FCC) catalyst, however, needs to be regenerated
continuously. Particulate emissions from regenerators result from the mechani-
cal actions in the regenerator (catalyst attrition) and coke combustion.1
Emissions from batch regenerators are not usually controlled; particulate emis-
sions are not significant due to the infrequency of regeneration.^ Therefore,
this discussion will focus on controls for the continuous regenerators used
on FCC units. Figure 9.9-2 shows a diagram of the fluid catalytic cracking
process.
The catalyst used in fluidized catalytic cracking is usually an alumino-
silicate powder or molecular sieve. The powder particles are typically be-
tween 40 and 80 micrometers in diameter.1 Catalyst particles are emitted
in the regenerator exhaust, despite the catalyst recovery cyclones located
in the regenerator. More than 90 percent of the particles entrained in the
regenerator exhaust are less than 44 micrometers in diameter and approximately
79 percent are less than 10 micrometers in diameter.1»8 Some particles may
9.9-1
-------
_
o
„§
II
Propane
de-asphalter
Vacuum
1
Bitumen
plant
residue
Bitumen
Vacuum gas
oil
CO
ss
> J
$3
f*=
•gE
'Eg
u-£
Lubricating
oils
Waxes ^
Extracts B
Asphalt
Vacuum
residue
Figure 9.9-1 Plan of = a complex refinery.5
Atmospheric
residue
9.9-2
-------
Table 9.9-1. PETROLEUM INDUSTRY EMISSION SOURCES AND CONTROLS
Particulate Control
Emission source matter device
FCC unit catalyst Catalyst fines, ESP following
regenerator coke dust, ash multistage
cyclones
Venturi
scrubber
Gravel bed
filter
Multiple tube
swirl vane
separator
Fluid coking Coke dust, ESP
sulfuric acid
mist
Gravel bed
filter
Scrubber
Cyclone
Delayed coking Coke dust Water washing
Asphalt air blowing Tar, oil mist Incineration
Wet Scrubbers
Sludge incineration Ash Single plate
impingement
Fabric Filter
ESP
Control
efficiency
67-99.9 for
particle size
less than
40 micrometers
95
N/A
99.5
68-98
99.9
N/A
N/A
N/A
N/A
N/A
90-98 for
particles greater
than 1 m
Reference
1, 8, 9
1
9
9
8
8
8
8
10
11
12
12
9
9
9
9.9-3
-------
Table 9.9-2. PETROLEUM INDUSTRY STACK PARAMETERS6
10
Emission
sources
Process heater - oil
Process heater - gas
Process heater - oil
Process heater - gas
Flare
Fluidized catalytic
cracking
Fluid coking - general
Fluid coking -
Number
of
sources
214
710
721
2347
148
140
9
3
Average
stack
height,
m
10
10
34
27
30
50
70
0
Stack
diameter,
m
0.6
0.5
1.8
1.5
0.6
1.9
2.7
0
Average
temperature,
°C
193
183
324
341
571
370
188
77
Average
gas flowrate,
Am3/s
1
1
1227
832
363
4172
7886
0
transport
Asphalt oxidizer
56
16
1.2
286
262
-------
To atmosphere
Regenerator
Products to
fractionation
Steam to remove products
from catalyst
Air to burn
carbon off
catalyst
Hot feedstock
and steam
Figure 9.9-2 Fluid catalytic cracking process.5
9.9-5
-------
also be emitted when the catalyst is transferred from the regenerator to the
feed hopper. Particulate emissions from FCC units are on the order of 0.05
to 0.1 kg/Mg of catalyst recycled.9
9.9.1.1 Particulate Control Techniques. Several control options for
fluidized catalytic cracking regenerators exist but the most frequently used
controls are tertiary cyclones and electrostatic precipitators. In some
cases, however, wet scrubbers and granular bed filters can also be used.
The hot regenerator exhaust stream is usually passed through a heat recovery
device, both to recover heat and to reduce the gas temperature to the control
device inlet. Fluidized catalytic cracking regenerators are typically fol-
lowed by CO waste heat boilers. Cyclones applied to this source are only
about 85 percent efficient in removing the 10 micrometer diameter particles
that predominate in regenerator exhaust gas.9 Since cyclone collection effi-
ciency decreases exponentially with particle size, cyclones have low effi-
ciency for collecting particles less than 5 micrometers in diameter.8
Therefore, tertiary cyclones are usually supplemented with an electrostatic
precipitator.9
Electrostatic precipitators operating on fluidized catalytic cracking
catalyst regenerators have efficiencies that range from 67 to 99.8 percent
but generally average 99.5 percent removal of the particles that escape the
cyclones.1»8 The efficiency varies with particulate loading and dust resis-
tivity .8 The ESP may be located either ahead of or after the CO boilers.
When the ESP is installed between the regenerator and the CO boiler, a smaller
ESP can be used since the additional combustion gas from the boiler will not
contribute to the gas volume being treated. In addition, particle resistivity
tends to be lower, and collection therefore easier, than when the ESP is
located downstream of the CO boiler. This lower particle resistivity results
from the higher gas temperature, increased coke content of the particulate
matter, and presence of natural "conditioning agents" such as ammonia and
sulfur trioxide in the regenerator exhaust as compared to the CO boiler ex-
haust. However, a heat exchanger is necessary between the regenerator and
the ESP to lov/er the gas temperature to 260 to 315°C.1 Ammonia injection can
be used to increase particle resistivity when the ESP is located, downstream
of the CO boiler.1
9.9-6
-------
An alternative to the use of tertiary cyclones and ESPs is the Shell
process. This process uses a centrifugal separator of the multiple-tube swirl
vane type, and is followed by either a baghouse or wet scrubber. The separator
has been shown to operate with an efficiency of 99.5 percent for 10 micrometer
particles.9 Additionally, a turbo-expander may be operated in conjunction
with the separator to recover some power from the gas stream. Another control
option receiving recent application for reducing particulate emissions from
catalyst regenerators is the granular bed filter.9 Catalyst-laden gas is
passed through a chamber containing granular sand as the filter media. The
media is cleaned periodically by a reverse air pulse.
The catalyst solids collected in the particulate control equipment are
generally inert and suitable for disposal by landfill.9 Energy requirements
for operating an ESP range from 35 kVA to 140 kVA, depending on the size of
the control device. Generation of the electric field in the ESP requires
approximately 10 to 15 kW per 1000 m^; the forced-draft fans needed to
overcome pressure losses in the ESP use about 10 kW per 1000 m^, assuming a
pressure loss of 125 Pa.5
9.9.2 Coking Processes, Emissions, and Control Techniques
In coking operations, the residual oil bottoms from the vacuum distil-
lation tower are used to produce coke and additional feedstock for catalytic
cracking and hydrocracking. The two coking processes used for this purpose
are delayed coking and fluid coking.
In delayed coking, the older and more commonly used process, residual
oil is cracked as severely as possible in a single pass heater to form a
heavy liquid stream and light hydrocarbon gas. The liquid stream is placed
in a coking drum, and the gas is sent on for further processing. The coker
drums are decoked once every 24 hours by drilling a hole through the coke '.
and hydraulically removing the coke from the drum. In fluidized coking,
coke is built up on a fluidized pellet bed. Steam is injected into the
bottom of the coker to fluidize the bed, and when the coke particles are
large enough, they are removed. A flow schematic of the fluid coking process
is presented in Figure 9.9-3. While the delayed coker drums are being decoked,
particulate emissions in the form of coke dust are .produced. Sulfuric acid,
with droplet sizes ranging from 0.5 to 3 micrometers in diameter, is emitted
along with coke dust from the fluidized coker exhaust stream.8
9.9-7
-------
REACTOR PRODUCTS
TO FRACTIONATOR
REFLUX
SLURRY
RECYCLE
STOCK
BURNER
,- . QUENCH
' ELUTRIATOR
MATER
AIR
PRODUCT
COKE TO
STORAGE
ANGLE BENT-
Figure 9.9-3 Flow diagram of modern fluid coking unit.
9.9-8
-------
Product recovery cyclones collect coke particles from the fluidized
coker exhaust stream. These cyclones may be followed by a v/et scrubber,
ESP, or fabric filter. Emissions from delayed coking are primarily fugitive
in nature and can be minimized by washing down the coking equipment.H This
procedure, however, produces a wastewater stream that requires treatment.
Minimal data are available on coking emissions and appropriate controls.
9.9.3 Air Blowing Operations. Emissions and Control Techniques
Air is blown countercurrent to an asphalt stream in a packed tower to
harden the asphalt and increase its melting point. This process emits oil
and tar aerosol as well as malodorous gaseous pollutants. Since the hot gas
blowing is exothermic, a steam quench stream is used to control the tempera-
ture.12 Aerosol emissions are controlled either by incineration or wet
scrubbing.
Existing fireboxes are usually used to incinerate asphalt air-blowing
emissions. Such a system is capable of handling emissions from an air blow-
ing unit producing 16 m3 of asphalt per 12-hour day.12
Wet scrubbers employed in this application frequently use a sea water
scrubbing solution. The disadvantage of using scrubbers is that a high
liquid-to-gas ratio, on the order of 13.4 L/Nrn^, is needed for high effi-
ciency. In addition, the wastewater produced may overload the refinery
wastewater treatment system and, if the wastewater is exposed to the atmo-
sphere, vaporized hydrocarbons will be emitted.12
9.9.4 Sludge Incineration. Emissions, and Control Techniques
Oily sludges produced by refinery processes are frequently atomized and
incinerated in a multiple-hearth or a fluidized-bed incinerator. The com-
bustible fraction is burned, and the noncombustible fraction becomes entrained
in the exhaust as flyash. Uncontrolled emissions from fluidized bed incinera-
tors are almost 10 times greater than uncontrolled emissions from multiple-
hearth incinerators (18 g/Nm^ and 2 g/Nm^, respectively).9. Scrubbers, bag-
houses, and electrostatic precipitators are all effective in reducing incin-
erator emissions. (Section 9.3 presents a detailed discussion on
incinerators.)
Venturi or impingement scrubbers, usually followed by a cyclone, have a
combined scrubber/cyclone efficiency of 90 to 98 percent for particles
9.9-9
-------
greater than 1 micrometer in diameter.9 Fabric filters can achieve an
efficiency of approximately 99 percent.9 However, the capital costs for
filters are almost three times the cost for a scrubber, and maintenance
costs are slightly more expensive. Also, care must be taken in operating
fabric filters to ensure that ignited ash from the incinerator does not
reach the baghouse.
Refinery sludges may be disposed of via landfills or lagooning, or
conditioned for use as fertilizer rather than being incinerated. Some types
of sludge need chemical treatment to render them suitable for landfill. One
such chemical treatment involves a three-phase reaction which traps the
sludge in a chemical matrix, forming a pseudo-mineral material.9
9.9.5 Process Heater, Boiler, and Flare Emissions and Control Techniques
Particulate emissions from process heaters, boilers, and flares are
caused by incomplete combustion. Proper design and maintenance of these
devices to ensure complete combustion can reduce the emissions. Since the
concentration of particulate matter emitted from process heaters is low,
there is generally no incentive to use external control devices.7»9-,13
Process modifications and changes in the combustion parameters (fuel, resi-
dence time, excess air, burner design) are implemented to reduce emissions.
(For more details, see Section 9.2.).
9.9-10
-------
REFERENCES FOR SECTION 9.9
1. Hardison, L. C., and C. A. Greathouse. Air Pollution Control Technology
and Costs in Nine Selected Areas. U.S. Environmental Protection Agency.
Publication No. PB-222-746. September 1972.
2. OAQPS Data File of Nationwide Emissions. U.S. Environmental Protection
Agency, Office of Air Quality Planning and Standards. Research Triangle
Park, NC. February 1979.
3. Cantrell, A. Annual Refining Survey. The Oil and Gas Journal. Tulsa,
OK. March 28, 1977.
4. Cantrell, A. Annual Refining Survey. The Oil and Gas Journal. Tulsa,
OK. March 26, 1979.
5. Formica, P. N. Controlled and Uncontrolled Emission Rates and Applied
Limitations for Eighty Processes. U.S. Environmental Protection Agency.
Publication No. EPA-450/3-77-016. September 1976.
6. Atmospheric Modeling Data from National Emission Data System (NEDS).
U.S. Environmental Protection Agency, Office of Air Quality Planning
and Standards. Research Triangle Park, NC. May 1979.
7. Parker, A. Industrial Air Pollution Handbook. McGraw-Hill. New York,
NY. 1975.
8. Manual on Disposal of Refinery Wastes - Volume on Atmospheric Emissions.
American Petroleum Institute, Division of Refining. .Publication No.
931. Washington, DC. September 1965.
9. Cavanaugh, E. C., et al. Environmental Problem Definition for Petroleum
Refineries, Synthetic Natural Gas Plants, and Liquified Natural Gas
Plants. U.S. Environmental Protection Agency. Publication No. EPA-600/
2-75-068. November 1975.
10. Laster, L. L. Atmospheric Emissions from the Petroleum Refining Industry.
U.S. Environmental Protection Agency. Publication No. EPA-650/2-73-017.
August 1973.
11. Sims, A. V. Field Surveillance and Enforcement Guide for Petroleum
Refineries. U.S. Environmental Protection Agency. Publication No.
EPA-450/3-74-042. July 1974.
12. Cavanaugh, E. C. Control Techniques for Volatile Organic Emissions from
Stationary Sources. U.S. Environmental Protection Agency. Publication
No. EPA-450/2-78-022.
13. Dana, M. T., et al. Particulate and S02 Emissions from Process Heater
at Shell and Texaco Refineries. Anacortes, WA. Battelle, Pacific North-
west Laboratories. September 1975.
9.9-11
-------
Page Intentionally Blank
-------
9.10 FOREST PRODUCTS INDUSTRY
The production of pulp and building products from wood is categorized
as the forest products industry. This section presents particulate emissions
and control information for the kraft and sulfite pulping industries as well
as for emission sources which are found in the manufacturing processes for
plywood and veneers. Most forest products production facilities supply a
significant portion of their in-plant energy needs through the use of waste
wood-fired industrial boilers; consequently, these boilers are quite prevalent
throughout pulp mills and plywood plants. Emissions and applicable control
technologies for waste wood-fired boilers are characterized in Section 9.2.
9.10.1 Kraft Pulping
Currently there are about 120 kraft (sulfate) pulp mills located in
28 states throughout the United States. The areas of greatest density are
the Southeast, the Northwest, and the Northeast, in descending order.1
In 1978, the capacity of these mills in the United States was 34.7 Tg.^
The average kraft pulp mill produces 700 Mg/day of pulp.l Particulate
matter emitted from the manufacture of kraft pulp amounted to 222 Gg in
1977 on a nationwide basis.2
9.10.1.1 Process Description. The process for producing kraft pulp
from wood is shown in Figure 9.10-1.3 In the process, wood chips are cooked
(digested) at elevated temperature and pressure in "white liquor", a water
solution of sodium sulfide (Na2S) and sodium hydroxide (NaOH). The white
liquor chemically dissolves lignin (the material that bonds the cellulose
fibers together) from the wood.
When cooking is completed, the contents of the digester are released
into the blow tank. Here, the major portion of the spent cooking liquor,
which contains the dissolved lignin, is drained and the pulp enters the
initial stage of washing. .The pulp is then filtered from the spent cooking
liquor, washed, and, in some mills, bleached before being formed, pressed,
and dried into a marketable product.
It is economically attractive to recover both the inorganic cooking
chemicals and the heat content of the "black liquor" (a combination of spent
cooking liquor and dissolved lignin), which is separated from the cooked
9.10-1
-------
CHIPS
RELIEF
j^
CH3SH. CH3SCH3, H2S
NONCONDENSABLES
CONTAMINATED
-*• WATER
PULP 13% SOLIDS
SPENT AIR, CH3SCH3,-*-
AND CH3SSCH3
OXIDATION
TOWER
CH3SH. CH3SCH3, H2S
NONCONDENSABLES
t
H2S. CHsSH. CH3SCH3.
AND HIGHER COMPOUNDS
TURPENTINE
CONTAMINATED WATER
STEAM, CONTAMINATED WATER,
H2S, AND CHsSH
UN
R
1 •
m
<
3=>
-o
0
o
I— .
BLACK LIQUOR
50% SOLIDS
DIRECT CONTACT
EVAPORATOR
IBLACK
WHITE-
LIQUOR
t
I
f 1 CO
/ I 0
I 1 O
I 1 ro
S~\.
V2T-
FILTER
r
LIQUOR 70% SOLIDS
CaO Na2S°4
'WATER
RECOVERY
FURNACE
OXIDIZING
ZONE
REDUCTION
ZONE
NaOH
*
MJL1-UK
1 4
GREEN
LIQUOR
SMELT j
Na2$ + Na2CO
Figure 9.10-1 Typical kraft sulf^te pulping and recovery process,
-------
pulp. The balance of the process is designed to recover both cooking chemi-
cals and heat. Recovery is accomplished by concentrating the liquor in
multiple-effect evaporators to about 55 to 70 percent solids, and then feeding
the liquor to a recovery furnace, where combustion and chemical recovery takes
place. . .. -
Initial concentration of the weak black liquor, which contains roughly
15 percent.solids, occurs in the multiple-effect evaporator. Here process
steam is passed countercurrent to the liquor in a series of evaporator tubes
that increase the solids content to 40 to 55 percent. The next step involves
the recovery furnace system.
There are two principal types of recovery furnace systems in use in the
industry: the direct-contact evaporator system (Figure 9.10-1), and the newer
indirect-contact or "low-odor" system (Figure 9.10-2). About 75 percent of
the new furnaces that have been installed in the last 5 years are indirect-
contact systems.1 .Combustion of the wood lignin dissolved in the black liquor
provides heat for generating process steam and converting sodium sulfate
(Na£S04) to sodium sulfide (Na2S). To make up for chemicals lost in the
operating cycle, salt cake (sodium sulfate) is usually added to the concen-
trated black liquor before it is sprayed into the furnace. The inorganic
compounds in the black liquor fall to the bottom of the furnace during com-
bustion and are known as "smelt."
The smelt, consisting of sodium carbonate (^003) and sodium sulfide,
is dissolved in water to form green liquor, that is transferred to a caus-
ticizing tank, where quicklime (CaO) is added to convert the sodium carbonate
to sodium hydroxide. Formation of the sodium hydroxide completes the regen-
eration of white liquor, which is returned to the digester. A calcium
carbonate mud precipitates from the causticizing tank and is calcined in a
kiln to regenerate quicklime.
Many mills need more steam for process heating, product drying, and
generating electric power, than can be provided by the recovery furnace
alone. Thus, conventional industrial boilers that burn coal, oil, natural
gas and, in many cases, bark and wood waste are commonly employed. Process
descriptions and control techniques used to control particulate emissions
from these sources are discussed in Section 9.2.
9.10-3
-------
RECOVERY FURNACE
o
i
AIR
SMELT **~
COMBUSTION
GAS
MECHANICAL GAS
PRECLEANER
(OPTIONAL)
65% SOLIDS
r
PARTICULATE
CONTROL DEVICE
INDIRECT
CONTACT
EVAPORATOR
40-55% SOLIDS
EXHAUST GAS
BLACK LIQUOR
Figure 9.10-2 Indirect-contract recovery furnace system.1
-------
9.10.1.2 Emission Characteristics and Applicable Control Technologies.
Particulate emissions from the kraft process occur primarily from the recovery
furnace, lime kiln, and smelt dissolving tank. These emissions consist mainly
of sodium salts but include some calcium salts from the lime kiln. The dust
collected in the kraft industry, especially from indirect-contact recovery
systems, is more corrosive and sticky than that encountered in other indus-
tries. * This leads to some special problems with the particulate emission
control equipment. Recovery furnace exhaust gases and particulate emissions
have different characteristics depending on whether they are generated from
a direct-contact (conventional) or an indirect-contact (low-odor) system, as
shown in Table 9.10-1.4 The typical size distribution of particulate matter
produced by an indirect-contact recovery boiler is given in Table 9.10-2.5
For a direct contact recovery boiler, the particle mass mean diameter may
vary from 1 to 7 micrometers.5»6
The recovery furnace, lime kiln, and smelt dissolving tank are the pri-
mary sources of particulate emissions to which control devices are applied.
ESP and scrubber systems are employed on recovery boilers, with ESP systems
being used most frequently. Lime kilns generally utilize scrubber systems,
but occasionally ESP systems are used; demister pads and other low energy
scrubber systems are generally applied to control particulate emissions from
smelt dissolving tanks. Fabric filters are not used in kraft mills because
of the high moisture content of the exhaust gases and the fact that mechani-
cal collectors are not efficient enough by themselves, due to the size dis-
tribution of the particulate matter, to provide the degree of control
required.
9.10.1.2.1 Recovery boilers. Application and design of ESP systems for
recovery boiler emission control depends on whether the system is to be applied
to a recovery system that uses a direct-contact or an indirect-contact evapora-
tor. For a recovery system using a direct-contact evaporator, the evaporator
itself may serve to reduce the mass of particulate emissions by as much as 50
percent.1 This, along with the fact that the particles emitted by a low-odor
recovery boiler are generally smaller and less dense than particles emitted by
a direct-contact recovery boiler, means that ESP systems for indirect-contact
evaporators must be designed with more conservative sizing estimates. Table
9.10-3 lists the design parameters for a typical ESP applied to an indirect-
9.10-5
-------
Table 9.10-1. PARTICULATE PROPERTY VALUES FROM CONVENTIONAL
AND LOW-ODOR RECOVERY PROCESSES^
Conventional Low-odor
Temperature, K 410-435 445-505
Moisture content, percent 30 7-20
Particle size, micrometers 6-10 Less than 6
Density, kg/m^ 320-400 80-160
Tenacity Reasonable Difficult
Resistivity Low High
Sulfur content Low High
9.10-6
-------
Table 9.10-2. TYPICAL PARTICLE SIZE DISTRIBUTION FOR INDIRECT-
CONTACT RECOVERY BOILER5
Size interval, Percent by weight
micrometers in size range, %
Greater than 4.08 19.8
2.40-4.08 7.6
. 1.62-2.40 20.3
0.89-1.62 26.1
0.51-0.89 13.6
Less than 0.51 13.2
9.10-7
-------
Table 9.10-3.
TYPICAL INDIRECT-CONTACT KRAFT RECOVERY BOILER
ELECTROSTATIC PRECIPITATOR SYSTEM DESIGN PARAMETERS7
Compartments
No. of fields
Collection plate area
Residence time
Gas velocity
Power input
Electrode rappers
Collection plate rappers
Rake speed
Rake torque
Screw speed
Screw torque
Dust density
Inlet concentration
Dust volume
Dust compartment depth
Efficiency
2
6-8
1.2 - 1.5 m2/Am3 per min
Minimum 10 sees
Maximum 1.1 m/s
40 - 90 watts/1000 Am3/min
700 - 900 ma/1000 Am3/min
164 - 490 ma/1000 m electrode
550 - 1220 m/unit
140 - 230 m2/urrit
2-3 cm/s
Minimum 75 Nm/m2 (60 in-lbs/ft2)
20 - 40 rpm
Minimum 55 Nm/m (150 in-lbs/ft)
80 - 130 kg/m3
11-18 grams/Nm3
1.6 - 3.7 m3/Mg of pulp
1.8 - 3 m
99.5 - 99.8 percent removal by weight
9.10-8
-------
contact recovery boiler.? Further information on the design parameters for
ESP systems may be found in References 5, 8, 9 and 10,
Problems encountered in applying ESP systems to recovery boilers include
corrosion, plugging, and wire breakage.1 These problems are apparently due
to operating the equipment at conditions for which it was not designed (i.e.,
higher gas volumes, higher inlet loadings, or lower inlet temperatures). In
order to-prevent corrosion, the manufacturers install insulation or heated
shells to maintain the gas temperature above the gas dew point throughout
the precipitator.
Scrubbers applied to kraft recovery boilers are generally of the venturi
type, and are multiple stage .(i-6.,.venturi scrubbers connected in series).
These systems are not widely used, and removal efficiencies as high as 95
percent by weight are obtainable with pressure drops around 3 kPa.H
9.10.1.2.2 Lime kilns. Wet scrubbers are frequently applied to lime
kilns, with venturi and impingement designs being the most prevalent. Typical
operating characteristics of particulate liquid scrubbers on kraft lime kilns
are summarized in Table 9.10-4.12,13 Average collection efficiencies for ven-
turi and impingement scrubbers range from 92 to 95 percent removal by weight.1
9.10.1.2.3 Smelt dissolving tanks. Showered mist eliminators are used
almost exclusively on smelt dissolving tanks. Showered mist eliminators con-
sist of fine wire pads approximately 30-cm thick. Removal efficiencies are
roughly 70 to 80 percent by weight.1»12 Demister pads used in series with a
packed tower or scrubber attain efficiencies of 92 to 96 percent.
The effect upon ambient air quality in the vicinity of a kraft mill
utilizing the above particulate control technologies is discussed below.
Ambient concentrations of particulate matter resulting from the implementa-
tion of alternate levels of control are presented in Table 9.10-5. The
following assumptions were applied in making the calculations:
o There are no significant seasonal or hourly variations in
emission rates for these plants.
o The plants are located in flat or gently rolling terrain.
o The meteorological regime is unfavorable to the dispersion of
effluents. This assumption introduces an element of
conservatism into the analysis.
o An average plant size of 910 Mg/day was used.
9.10-9
-------
Table 9.10-4. OPERATING CHARACTERISTICS OF PARTICULATE LIQUID
SCRUBBERS ON KRAFT LIME KILNSl2,13
Scrubber type
Parameter Ventura Impingement
Liquid-to-gas ratio, 1/m3 1.73 - 3.21 0.54 - 2.0
Slurry solids, % by wt 10-30 1-2
Pressure drop, kPa 2.5 - 3.75 1.25 - 1.75
9.10-10
-------
Table 9.10-5. ESTIMATED IMPACT OF KRAFT PULP MILL PARTICIPATE
MATTER EMISSIONS UNDER NON-DOWNWASH ASSUMPTION
(910 Mg PER DAY KRAFT PULP MILL)!
Maximum
Control Averaging
alternative time
•[b 24
annual
2C 24
annual
3d 24
annual
combined
concentration
9.7
2.2
5.1
1.1
2.5
0.6
Contribution of
, each source,9 /xg/m3
RF SDT LK
Neg.
0.2
Neg.
0.1
Neg.
0.1
1.8
0.4
1.1
0.2
1.1
0.2
7.9
1.6
4.0
0.8
1.4
0.3
a RF = Recovery Furnace
SDT = Smelt Dissolving Tank
LK = Lime Kiln
Neg. = Negligible
bAlternative number 1 has the following control techniques:
o Recovery Furnace - 99.9 percent ESP for particulate control.
o Smelt Dissolving Tank - Demister.
o Lime Kiln - 3.75 kPa venturi scrubber.
o Brown Stock Washer Systems - Mo control.
o Black Liquor Oxidation System - No control.
o Condensate Stripper System - Incinceration.
cAlternative number 2 consists of the following control techniques:
o Recovery Furnace - 99.9 percent ESP plus process control; noncontact
evaporation.
o Smelt Dissolving Tank - Scrubber plus use of clean water (process
control).
o Lime Kiln - 7.5 kPa venturi scrubber with caustic addition to scrubber
water plus process controls.
^Alternative number 3 is identical to number 2 except that the venturi
scrubber used for control of particulate emissions from the lime kiln is
replaced with a high efficiency electrostatic precipitator.
9.10-11
-------
o There is no aerodynamic downwash (interference of the stack
plume by a natural or manmade obstacle resulting in increased
localized concentration of the pollutants being emitted).
9.10.1.3 Particulate Emissions Data.l Emission data presented here are
the results of tests conducted by EPA at 12 kraft pulp mills. These data
represent 19 particulate tests performed to justify new source performance
standards for kraft pulp mills. Eight emission tests for particulates were
performed on seven recovery furnaces and seven lime kilns; five smelt dis-
solving tanks were also tested. Opacity readings were taken during particulate
tests on four stacks at three recovery furnaces and on three smelt dissolving
tank stacks and also during two tests on one lime kiln. These opacity readings
and any correlations to emissions are available in Reference 1. Additional
data obtained from various kraft mills, state air pollution control agencies,
and other sources are also presented where pertinent.
9.10.1.3.1 Recovery furnace. Five recovery furnaces were tested by EPA.
Three of the furnaces had direct-contact evaporators; the other two furnaces
were indirect-contact (no direct-contact evaporator) type furnaces. The par-
ticulate emissions for the furnaces tested are shown in Figure 9.10-3. Data
obtained from the operators of mills with several of the furnaces tested by
EPA are also presented in Figure 9.10-3 for comparison purposes.
9.10.1.3.2 Smelt dissolving tanks. Four smelt dissolving tanks were
tested by EPA. The data from these tests are presented in Figure 9.10-4.
Monthly data obtained from a state agency on two of the smelt tanks are also
presented.
9.10.1.3.3 Lime kilns. Particulate data obtained on four lime kilns
tested by EPA are presented in Figure 9.10-5. Data obtained by the mills and
state agencies are also presented. The particulate emissions from each lime
kiln are controlled by a venturi scrubber.
Testing was performed on more than one type of fuel on several of the lime
kilns, since the results of the testing on lime kiln K indicated that the con-
trolled emissions depended on the type of fuel used. The difference in the
controlled particulate levels when using No. 6 oil and natural gas seems to be
the result of the added particulates produced by inefficient combustion of
No. 6 oil.
9.10-12
-------
0.26
0.24
0.22
0.20
5
SO. 18
*
g 0.16
D— *
| 0.14
1
1 0.12
UJ
h-1
_i n 10
__ W • | V
«— *
Sc 0.08
Q.
0.06-
0.04
0.02
0
KEY
_ . " •
EPA Other
P P Maximum Data Point
it n
• H-k -iW) Average
H n
W y Minimum Data Point
-
- •
-
. ' ' O
Ju1 i|
tJ-M
Proposed NSPS Level ' '
11"
U
P
J i
Id
P ft -
ii it
M ^ " L Q
o 1 1 ill -
J£ 5 Data p « ' ! !
** ^Points tft* TT
\. M ®
i i i i i ' , \? i i
FURNACE D
Furnace _,
Type DC
Control p
Equipment
0.110
0.099
0.088
0.077
0.066
Lu
0.055
0.044
0.033
0.022
0.011
J'l J'2 J"l J"2 Kl K2 LI L2
1C 1C 1C 1C 1C 1C DC DC
PP P PP P P P
IC-Indirect Contact
DC-Direct Contact
P-Electrostatic Precipitator
Figure 9.10-3 Particulate concentrations in control systems exhaust
from kraft recovery furnaces.1
9.10-13
-------
0.25
§ 0.20
D>
O)
z:
0
h^
CO
= 0.15
Ul
£
3
=3
o
>— 4
I—
°- o.io
0.05
0
Key
EPA Other
ft H Maximum Data Point
'« 1 1 H
h!-H »|-[l Average
, i ' ' j
W y Minimum Data Point f
f 1 P
1 > i
1 1 t
| 1 10
1 1
1
1 I '
Proposed NSPS Level 1 ( j i
T # !
' '
1 ' 1 1
! i1 '
Pi ' '
is1 ^ H U
8 !' '
P H '
M p u y>
I i 11 i i
1 • • . $
1 1 to
Ij °l
y * N
0;5
0.4
fX.
O
0.3 1
J3
0.2
0.1
0
Smelt Dissolving Tank D E Fl F2 Gl 62
Control Equipment C C PT PT PT PT
PT-Packed Tower
C-Cyclonic Scrubber
Figure 9.10-4 Particulate emissions in control system exhaust from
smelt dissolving tanks used in the kraft pulping industry.
9.10-14
-------
u.ou
0.55
0.50
0.35
§
a
W -
PARTI CULATE CONCENTRATION,
D 0 O C
•j ro ro c*
n o ui c
U . 1 J
0.10
0.05
0
Lime K11
Control
Fuel Typ
n «*
' ' EPA Other
j f\ p Maximum Data Point
frri H n -
ii it '
Id f-j-yi fj]i Average
II " "
v H b1 Minimum Data Point.
•>
^ \
Proposed NSPS Level
(oil-fired)
. P •
1 1
1 1
*
1 1
U
; 1 •
n Proposed MSPS
" Level FTTH
.. . ' ' o •
u I i H , • " ~ . . —
n H n (gas-fired)
1 ' • i . M
"^ a P
'! tf b1 M
w fi • • 4 w -
ii TJ 'Id
-I « 8 w
!: f-!!-
" H
y y -
1 1 I I 1 1 1 1 ' 1 I 1
n Kl K2 K3 LI L2 L3 Nl N2
Equipment VVVVVV VV
e Used 6 N N 2 N N 6 N
0.26
0.24
0.22
0.14
0.12
0.10
CT
0. 8
0.06
0.04
0.02
V-Venturi Scrubber
N-Natural Gas
-2-No. 2 fuel oil
6-No. 6 fuel oil
Figure 9.10-5 Particulate concentrations in control system exhaust from
lime kilns used in the kraft pulping industry.1
9.10-15
-------
9.10.1.4 Secondary Environmental Impacts. Utilization of participate
control equipment in the kraft pulping industry generally has minimal secon-
dary environmental impacts on water or solid wastes because the control device
by-products are usually recycled to the process. Slurries from wet bottom
elecrostatic precipitators on recovery furnaces and liquids from scrubbers
on smelt dissolving tanks are recycled. Scrubbing water and lime-mud wash-
water effluents from the lime kiln are normally recycled to the causticizing
system for chemical recovery.
Particulate matter collected as a dry mass results from the utilization
of dry-bottom electrostatic precipitators on recovery furnaces or lime kilns.
The dry particulate matter from the recovery furnace is primarily ^SCty,
which is reused by dissolving it in the black liquor and returning it to the
furnace for reduction to Na2S. The sodium salts, calcium carbonate, and
calcium oxide collected from the lime kiln exhaust gases are similarly re-
turned to the causticizing unit. A secondary impact concerning solid waste
may result when caustic solution is used in the lime kiln scrubber for simul-
taneous removal of particulate matter and reduced sulfur gases. If the mill
at which the control system is applied cannot accept the added sodium in the
form of caustic solution due to total mill chemical balance, some sodium
waste may have to be removed and disposed.
The energy requirements associated with the various control alternatives
are presented in Table 9.10-6.15 The total energy consumption values and
the values reported for scrubbers applied to recovery furnaces were obtained
from Reference 15. All the other values were obtained from Reference 1, and
were in general agreement with Reference 15. As can be seen from Table
9.10-6, scrubbers employed on recovery furnaces consume the most energy—
more than three times the amount consumed by ESPs on recovery furnaces.
9.10.2 Sulfite Pulping
There are approximately 26 sulfite pulp mills in the United States cen-
tered primarily in Washington, Oregon, Wisconsin, and Maine.i6 In 1978, the
capacity of these mills was 2.9 Tg on an annual basis, with a typical mill
size of 310 Mg/day.l6 Total emissions of particulate matter from these
mills amounted to 1.3 Gg in 1977 on a nationwide basis.2
9.10.2.1 Process Description and Emissions. The production of acid
sulfite pulp is similar to kraft pulping. The basic difference is that in
9.10-16
-------
Table 9.10-6. ENERGY IMPACTS OF CONTROL TECHNIQUES FOR KRAFT
PULP
Control device Incremental
required for Control device energy required
economic recovery required for to meet NSPS,D
Affected facility level a meeting NSPS kWh/Mg
Recovery furnace ESP, 90% efficiency ESP, 99.5% efficiency 8.3
Scrubber, 7.5 kPa 6
Smelt tank Demister Scrubber, 1 kPa 2.5
Lime kiln Scrubber Scrubber, 7.5 kPa 5.6
Total energy
consumption of
control device,0
kWh/Mg
9.9
35.6
3.3 - 4.4
7.7
aThe economic recovery level is the level of control that can be economically justified by
the value of the materials recovered.
blncremental energy required to meet NSPS as compared to that required for achieving the
economics recovery level.
cThe total energy consumption is the amount of energy consumed by control equipment.
-------
place of the sulfide-containing caustic solution used to dissolve the lignin
in the wood, a salt of sulfurous acid is employed. To buffer the solution,
a bisulfite of magnesium, ammonium, calcium, or sodium is used. A simplified
flow diagram of a magnesium-base process is shown in Figure 9.10-6.17 Calcium-
base systems are used only in older mills and are being replaced with new
magnesium or ammonium-base mills due to problems with disposal of spent liquor
and recovery systems. Only one sodium-base sulfite pulp mill exists today.16,18
Digestion is carried out under high pressure and elevated temperature in
the presence of a sulfurous acid-bisulfite cooking liquor. When cooking is
completed, the digester is either discharged at high pressure (blowing) into
a blow pit or its contents are pumped out at a lower pressure into a dump
tank (dumping). The spent sulfite solution (known as red liquor) is drained
and is either treated and disposed, incinerated, or sent to a plant for recov-
ery of heat and chemicals. The choice of whether or not chemical recovery is
desirable is dictated by the base employed in the cooking liquor. In calcium-
base systems, chemical recovery is not practical, due to the low solubility
of calcium salts and the formation of scale and calcium sulfate ash in the
recovery process; therefore, the spent liquor is either discarded or incin-
erated. In ammonium-base operations, heat and sulfur can be recovered from
the spent liquor through combustion and subsequent S02 absorption, but the
ammonium base is consumed in the process. In sodium or magnesium-base opera-
tions, heat, sulfur, and base recovery are feasible. If recovery is practiced,
the spent weak red liquor is concentrated to 55 to 60 percent solids and then
sprayed into a furnace and burned. This recovery furnace is the major source
of particulate emissions in the su-lfite pulp mill.
Table 9.10-7 presents emission factors for sulfite pulping.17 The type ,
of particulate matter emitted depends on the base of the cooking liquor.
Magnesium oxide is the major constituent of particulate matter emitted by
magnesium-base liquor combustion, while ammonium sulfate and ammonium sulfate
particles are emitted from ammonia-base recovery furnaces. Sodium carbonate
and sodium sulfate are the major components of the particulate matter emitted
by sodium-base liquor combustion. Limited information on particle size is
available, but one study reports the mass mean particle size for a sodium-base
sulfite mill to be 0.96 micrometers.I9
9.10-18
-------
RECOVERY FURNACE/
ABSORPTION STREAM
EXHAUST
STEAM FOR
PROCESS AND POWER
O
I
VO
Figure 9.10-6
Simplified process flow diagram of magnesium-base
sulfate pulp process employing chemical and heat
recovery.""
17
-------
Table 9.10-7. EMISSIONS FACTORS FOR SULFITE PULPINGa,17
Particulate emission factor'3
Source
Recovery systemd
Base
MgO
NH3
Na
Control lb/ADUTc
Multiclone and 2
venturi scrubbers
Ammonia absorption 0.7
and mist eliminator
Sodium carbonate 4
kg/ADUMT
1
0.35
2
scrubber
aAll emission factors represent long-term average emissions.
^Factors expressed in terms of Ib (kg) of pollutant per air dried
unbleached ton (MT) of pulp.
°ADUT = Air dried unbleached ton.
"The recovery system at most mills is a closed system that includes the
recovery furnace, direct contact evaporator, multiple-effect evaporator,
acid fortification tower, and S02 absorption scrubber. Generally, there
will only be one emission point for the entire recovery system. These
factors are long-term averages and include the high S02 emissions during
the periodic purging of the recovery system.
9.10-20
-------
Typical stack parameters for sulfite pulp mills are presented in Table
9.10-8.20 These parameters represent averages for 26 mills across the United
States.
9.10.2.2 Participate Control Techniques. Emission control equipment
applied to sulfite pulp mill recovery furnaces is generally dependent upon
the chemical base used in the-cooking liquor. Magnesium-base recovery furnaces
are frequently controlled with multicyclones and venturi scrubbers.16.18 often
multicyclones are followed in series by one or more venturi scrubbers, and
these systems operate to eliminate sulfur emissions in addition to particulate
matter emissions. In one plant, the multicyclone consists of 7640 tubes; each
tube is 7.6 cm in diameter. Efficiencies of these multicyclones range from
96 to 98 percent in removing magnesium oxide.16,21
Ammonia-base sulfite recovery furnace emissions are controlled by low
pressure drop tray scrubbers followed by glass-fiber packed filter units or
mist eliminators. The multiple tray scrubber is designed primarily for sul-
fur dioxide absorption; however, some particulate matter removal is also
achieved. A typical flue gas scrubber applied to an ammonium-base sulfite
recovery boiler operates with a pressure drop of 2.75 to 3.85 kPa, and removal
efficiencies range from 85 to 95 percent.22,23
A venturi scrubber followed by a cross-flow packed bed scrubber is in
use in the only sodium-base recovery system in the U.S. The system serves
several functions by reducing particulate and reduced sulfur gaseous emissions
as well as recovering heat from the recovery furnace exhaust gases. The ven-
turi scrubber operates at a pressure drop of 1.75 to 2.5 kPa; the total system
pressure drop varies between 2.5 to 3 kPa. Removal efficiencies range from
96.5 to 97 percent.8
9.10.3 Plywood Manufacture
In 1978, approximately 500 veneer and plywood mills produced 2000 million
m2 of 0.95 cm thick plywood. The majority of these plants are located in
Washington, Oregon, and North Carolina.24,25 Total emissions of particulate
matter from these plants amounted to 22.4 Gg on a nationwide basis in 1977.2
9.10.3.1 Process Description and Emissions. Figure 9.10-7 shows a
typical plywood plant flow diagram.26 During the manufacture of plywood, in-
coming logs are sawed to desired length, debacked, and then peeled into thin,
9.10-21
-------
Table 9.10-8. STACK PARAMETERS FOR A TYPICAL SULFITE PULP MILL.20
Stack height, m 51
Stack diameter, m 2.5
Flue gas temperature, K 385
Flue gas flow rate, Am^/min 2340
9.10-22
-------
PAIIIKUIAll MAI I IK
EMiSbHIN SUURU
PARTICULATt MATTER
LMISSION SOURCb
i-O
O
I
ro
THIS HASTE IS BURNED OK
CHIPPED FOR REUSE
Figure 9.10-7 Plywood plant: process diagram.26
-------
continuous veneers of uniform thickness (6 to 50 mm is common). These veneers
are then transported to special dryers, where they are subjected to high tem-
peratures until dried to a desired moisture content. After drying, the veneers
are sorted, patched, and assembled in layers, with some type of thermosetting
resin used as the adhesive. The veneer assembly is then transferred to a hot
press where the plywood product is formed. Finally, the plywood must be
trimmed and sanded before it is bundled and shipped as final product. The two
major sources of particulate emissions in the plywood mill are the veneer dry-
ing and sanding operations.
There are two major types of veneer dryers: longitudinal and jet. The
longitudinal, or conventional, design consists of a forced convection drying
system utilizing hot air that is in parallel to the flow of the veneer. The
veneer is moved through the zones of the dryer on a roller conveyor, and the
air is passed over and under the veneer at temperatures around 530 K. The
gas is recirculated, with only a fraction of the gas stream being exhausted
after each pass. The jet design dryer differs from longitudinal dryers in
that the heated air stream is delivered to the veneer at a right angle. The
air stream is, in essence, impinged on the veneer. The air circulation in a
jet zone is across the flow of veneer through the dryer; air is recirculated
in a similar manner and at roughly the same temperature as in longitudinal
dryers. As with the longitudinal design, the veneer is moved through the
dryer on a roller conveyor.
Veneer dryers may be either direct heated or steam heated. Direct heated
dryers may be gas-fired or waste wood-fired. In the gas-fired veneer dryer,
the burner utilized in most cases is a line type burner wi'th burner rails
spread uniformly across the duct. This design allows the air stream to be
heated evenly.27 A steam heated dryer has the same air circulation system
as a gas heated dryer but instead of having a gas burner in the upper ducts,
the steam dryer has banks of finned heating coils. Since not enough heat
can be imparted to the air stream through these coils, additional steam
coils are employed along the conveyor of the dryer.27
The major pollutants emitted from veneer dryers are organic compounds
consisting of two discernable fractions: (1) condensibles, consisting of
wood resins, resin acids, and wood sugars, which form a blue haze upon cooling
9.10-24
-------
in the atmosphere, and (2) volatiles, which are comprised of terpines and
unburned methane, the latter occurring when gas-fired dryers are employed.
In addition, negligible amounts of fine wood fibers are emitted during the
drying process.25 The condensible organics tend to be submicrometer in
size, while the wood fiber and other particulate matter is generally in the
1 to 10 micrometer size range.28,29
Sanding operations generate larger particles than veneer dryers, with a
mean particle size on a count basis of 22 micrometers and 99.8 percent by
weight of the particles in the 10 to 80 micrometer range.30 Pneumatic trans-
fer systems are used to capture the dust at the sanders and transport it to
storage. Roughly 50 m3/min of air is needed to transport 0.5 kg of sander
dust in a low pressure system.30
9.10.3.2 Particulate Control Techniques. The main control technologies
employed on veneer dryers are scrubber systems and incinerators. The scrubber
systems generally employ pressure drops between 2.5 and 10.0 kPa, and they
operate at 70 to 90 percent removal efficiency by weight.31,32,33 One new
scrubber system currently being used is the Becker Sand Filter. Its sand
bed is 2.5 m2, and it operates with a pressure drop of 6.3 to 10.0 kPa and a
liquid-to-gas ratio of roughly 9.2 L/m3 for an efficiency of 85 percent.32,33
Incinerator systems involve utilizing a waste wood-fired boiler to control
emissions from steam heated veneer dryers. The exhaust from the dryers is
passed into the incinerator to combust the organic compounds present.
Incinerators generally reduce the outlet opacity from 60 to 10 percent.31
In direct heated dryers, up to one-half of the dryer exhaust can be vented
to the waste wood-fired burner.
The most widely used method for control of sander dust emissions is a
bag filter. A primary cyclone followed by a bag filter results in the best
performance, with outlet loading well below 0.11 g/Nm3.30 Filtering velo-
cities should not exceed 3 m/min and reverse air bag cleaning is generally
employed in wood dust applications. The danger of fires and explosions is
the only major drawback with these units. Small diameter multicyclones have
recently been used to reduce outlet dust loading below 0.11 g/Nm3, and they
are not subject to the fire and explosion problems of fabric filters. The
pressure drops range between 1 and 2.5 kPa for these units.30
9.10-25
-------
REFERENCES FOR SECTION 9.10
1. Standards Support and Environmental Impact Statement. Volume 1:
Proposed Standards of Performance for Kraft Pulp Mills. U.S. Environ-
mental Protection Agency. Research Triangle Park, NC. Publication No.
EPA-450/2-76-014a. September 1976.
2. OAQPS Data File of Nationwide Emissions. U.S. Environmental Protection
Agency, Office of Air Quality Planning and Standards. Research Triangle
Park, NC. February 1979.
3. Compilation of Air Pollution Factors, Second Edition. U.S. Environmental
Protection Agency. Publication No. AP-42. April 1973.
4. Bump, R. L. Precipitator Design for Low-Odor Boilers Offers Special
Problems. Pulp and Paper. October 1976.
5. Paul, J. E. Application of Electrostatic Precipitators for the Control
of Fumes from Low-Odor Pulp Mill Recovery Furnaces. Journal of Air
Pollution Control Association. 25:2. February 1975.
6. Bosch, J. C., et al. Size Distribution of Aerosols From a Kraft Mill Recov-
ery Furnace. TAPPI. _54_:11. November 1971.
7. Caron, A. L. An Analysis of the Design Parameters and Inquiry into the
Performance of High.Efficiency Electrostatic Precipitators Installed on
Kraft Recovery Furnaces. NCASI Special Report. National Council of the
Paper Industry for Air and Stream Improvement, Inc. New York, NY.
8. Engelbrecht, H. L., and N. D. Graves. Electrostatic Precipitator Instal-
lation for a Lpw-Odor Recovery Boiler. (Presented at the 68th Annual
Meeting of the Air Pollution Control Association. June 1975.)
9. Szabo, M. F., and R. W. Gerstle. Operation and Maintenance of Particulate
Control Devices in Kraft Pulp Mill and Crushed Stone Industries. U.S.
Environmental Protection Agency. Research Triangle Park, NC. Publica-
tion No. EPA-600/2-78-210. October 1978.
10. Henning, K., W. Anderson, and J. Ryan. Improved Air Pollution Control
for a Kraft Recovery Boiler: Modified Recovery Boiler No. 3. U.S.
Environmental Protection Agency. Research Triangle Park, NC. Publica-
tion No. EPA-650/2-74-071a. August 1974.
11. The Electrostatic Precipitator Manual. The Mcllvaine Co. Northbrook, IL.
April 1977.
12. Sittig, M. Pulp and Paper Manufacturing, Energy Conservation and
Pollution Prevention. Noyes Data Corp. Park Ridge, NJ. 1977.
13. Environmental Pollution Control, Pulp and Paper Industry, Part I: Air.
U.S. Environmental Protection Agency, Technology Transfer. Cincinnati,
OH. Publication No. EPA-625/7-76-001. October 1976.
9.10-26
-------
14. Atmospheric Emissions from the Pulp and Paper Manufacturing Industries.
Cooperative NCASI-USEPA Study Project. U.S. Environmental Protection
Agency. Research Triangle Park, NC. Publication No. EPA-450/1-73-002.
September 1973.
15. Energy Requirements for Environmental Control in the Pulp and Paper
Industry. Roy F. Weston, Inc. U.S. Department of Commerce, Office of
Environmental Affairs. Washington, DC. March 1977.
16. Thompson, C. M., et al. Screening Study on Feasibility of Standards of
Performance for Two Wood Pulping Processes. U.S. Environmental Protection
Agency. Research Triangle Park, NC. Publication No. EPA-450/3-78-111.
November 1978.
17. Compilation of Air Pollution Factors, 3rd edition. U.S. Environmental
Protection Agency. Publication No. AP-42. August 1977.
18. Linero, A. Background Document: Acid Sulfite Pulping. U.S. Environ-
mental Protection Agency. Research Triangle Park, NC. Publication No.
EPA-450/3-77-005. January 1977.
19. Teller, A. J., et al. Emission Control for Sulphite Recovery Boilers.
Pulp and Paper. Canada. 78_:2. February 1977.
20. Atmospheric Modeling Data from National Emission Data System (NEDS).
U.S. Environmental Protection Agency, Office of Air Quality Planning
and Standards. May 1979.
21. Keef, R. C. Magnesium Bisulfite Startup. TAPPI. _54_:4. April 1971.
22. Benjamin, R. M, Air Pollution Control Technology, Wood Pulping Industry.
Unpublished Data. Environment Canada, Chemical Process Sources Division,
Abatement and Compliance Branch, Air Pollution Control Directorate.
Ottawa, Ontario, Canada.
23. Button, E. F., et al. A New Processing System for Sulfite Recovery
Boiler Flue Gas. (Presented at TAPPI Environmental Conference. Chicago,
IL. April 1977.)
24. American Plywood Association Management. Bulletin No. FA-200. American
Plywood Association. Tacoma, WA. April 10, 1979.
25. Development Document for Effluent Limitations Guidelines and New Source
Performance Standards for the Plywood, Hardboard, and Wood Preserving
Segment of the Timber Products Processing Point Source Category, U.S.
Environmental Protection Agency. Washington, DC. Publication No.
EPA-440/l-74-023a. April 1974.
26. Bodien, D. G. Plywood Plant Glue Waste Disposal: Progress Report.
U.S. Department of the Interior. Corvallis, OR. Publication No. PR-2.
February 1968.
27. Vranizan, 0. M. Veneer Dryers - Typical Construction, Operations and
Effluent Abatement Possibilities. (Presented at APCA 1972 Annual Meeting.
Eugene, OR. November 17, 1972.)
9.10-27
-------
28. Brown, D. J. Veneer Dryer Emissions—Problem Definition. Forest Products
Journal. 21:9. September 1971.
29. Investigation of Emissions from Plywood Veneer Dryers, Final Report.
Prepared for Plywood Research Foundation. Tacoma, WA. U.S. Environmental
Protection Agency. Durham, NC. Contract No. CPA-70-138. March 1971.
30. Tretter, V. J. Technology for the Control of Atmospheric and Waterborne
Emissions from Plywood and Lumber Manufacture. (Presented to 69th AICHE
Meeting. Chicago, IL. November 1976.)
31. Tretter, V. J. Plywood Veneer Dryer Emission Control Systems. (Presented
at the 69th APCA Meeting. Portland, OR. June 1976.)
32. Burkart, A. Veneer Drier Emissions Control Systems. Oregon Department
of Environmental Quality. Portland, OR. June 1975.
33. Mick, A. Current Particulate Emissions Control Technology for Particle-
board and Veneer Dryers. (Presented to Pacific Northwest International
Section of Air Pollution Control Association. November 1973.)
9.10-28
-------
9.11 LEAD-ACID BATTERY MANUFACTURING
Over 200 plants have been identified as primary producers of lead-acid
storage batteries. The largest concentration of these plants is in California,
Florida, Illinois, Pennsylvania, and Texas. The lead-acid battery industry is
the largest single user of lead in the nation, accounting for more than half
of the total lead consumption. In 1975, this total lead consumption amounted
to 1.19 Tg, of which 0.635 Tg were used in the manufacture of storage bat-
teries. The lead-acid battery represents the most frequently used storage
cell, and two major types are produced: starting-lighting-ignition (SLI)
batteries and industrial storage batteries for low voltage power systems.
SLI batteries account for more than 80 percent of the battery market, and
almost 80 percent of the SLI batteries are used in automobiles. In the past,
production plants for lead-acid batteries have been small and located close to
their markets, but the present trend is toward larger plants with increased
efficiency and capacity. The industry is expected to expand at an annual
rate of 3 to 5 percent over the next 5 to 10 years.1
Particulate emissions from the battery industry include lead, lead oxide,
and sulfuric acid mist. The fraction of particulate matter estimated to be
lead is 50 to 60 percent; however, published quantitative data to substan-
tiate this estimate are minimal.2»3 National lead emissions from the storage
battery industry were estimated to be 81.5 Mg in 1975, as detailed in Table
9.11-1.1 Because of the importance of the storage battery industry as an
emission source of high lead levels in the atmosphere in the vicinity of
battery plants, this discussion concentrates on lead emissions and applicable
control techniques. In general, emission control methods used to reduce
lead emissions will also reduce the emission of non-lead particulate matter.
9.11.1 Storage Battery Manufacturing Processes and Particulate Emissions
A lead-acid battery consists of lead electrodes, or plates, in an electro-
lytic solution of sulfuric acid and water. The plates are made of an inactive
lead grid, onto which a lead oxide sulfate paste is applied and bonded. The
lead grid provides both structural support and a conductive path for electric
current. A process diagram of battery manufacturing, showing emission points
and uncontrolled emission factors, is given in Figure 9.11-1. Processes and
their emissions are described below.
9.11-1
-------
Table 9.11-1. 1975 NATIONWIDE EMISSIONS OF LEAD FROM THE MANUFACTURE
OF LEAD-ACID STORAGE BATTERIESl
Process
Estimated
avg. control
Throughput,3 efficiency,*3
Gg Pb %
Uncontrolled Estimated
emission factor, actual
g/kg of lead emissions,
throughput Mg
Lead oxide
production
Grid casting
Lead
reclamation
Paste mixing
Three- process
operation
338
397
6.26
338
63'5
__c
50
80
90
90
0.01
0.07d
2.97
0.86d
0.56d
3.4
10.3
3.6
29.0
35.9
Total emissions for 1975
81.5
aBased on 1975 data from "Lead Industry Monthly Mineral Industry
Surveys," Bureau of Mines, May 1976.
^Based on information obtained from battery manufacturers and control
agencies.
cEmission factor is based on controlled emissions; fabric filters are
a part of the process.
"Based on test data in units of pounds of lead emissions per 1000
batteries and assumption of an average of 11.8 kg of lead per battery.
Half is assumed to be in the castings and half in the paste.
Estimated at 1 percent of total lead throughput.
9.11-2
-------
0.41b
(0.90)
t
(10.0)
6.6711
(14.7)
t
H2S04
DRY BATTERY LINE
LEAD
INGOTS *
1
t
1
i
r" a — ~"i
i i
i i
i , i
GRID
CASTING
FURNACE
GRID
CASTING
i
i
i
1
r ~~ - t
GRID
PASTING
—*• PLATE — • ELEMENT -
STACKING BURNING
t
1
1
I
•• ELEMENT
ASSEMBLY
— !
» FORMATION
f
ACID
__ ASSE^8LY INTO
BATTERY CAST
RINSING
DRYING
S04 Mist
»
ASSEMBLY IN
BATTERY CAS
TO
E
WA
PA
H
TE
FORMATION
|
L ACID .
ACID
REFILL
BOOST
CHARGE
aGRAMS (POUNDS) OF LEAD EMISSIONS Pl« KILOGRAMS (TON) OF LEAD CHARGED.
^KILOGRAMS (POUNDS) OF LL'AD EMISSIONS PER 1000 BATTERIES.
FRESH ACID WET BATTERY LINE
Figure 9.11-1
Process flow diagram showing uncontrolled lead
emission factors for lead-acid battery manufacture.1
-------
9.11.1.1 Grid Casting. Techniques for casting the grid vary with the
alloy used, type of mold, and mold preparation procedure. Generally, however,
one of two methods is used. The lead ingots can either be melted in several
melting pots which are attached directly to the casting machines, or the lead
can be melted in a central pot furnace from which it is transferred to the
casting machines. The grids are formed in the molds and then are ejected,
trimmed, and stacked.
Lead emissions from grid casting operations are generally minimal. Melt-
ing pot furnace emissions are typically vented to the atmosphere, but the
areas around the casting machines are usually unvented. Estimated uncon-
trolled emissions from grid casting in 1975 amounted to 0.07 g/kg of lead
processed. Approximately one quarter of the particulates emitted were less
than 15 micrometers in diameter.1
9.11.1.2 Paste Mixing. The paste mixing operation is a batch process
which uses a muller, Day, or dough-type mixer. Lead oxide is blended in the
mixer with a sulfuric acid solution and a trace amount of Dynel fiber to form
a stiff paste. Carbon black and approximately 1 percent by weight of expander
are added to paste batches intended for negative plates. The mixing cycle is
generally 15 to 60 minutes in duration.
Lead oxide particles are emitted by the paste mixers, and smaller'amounts
of the other paste constituents such as Dynel and carbon black are also emitted'.
These emissions are produced when the dry ingredients are charged to the mixer,
and are emitted during the first few minutes of the mix cycle. Uncontrolled
lead oxide emissions are estimated to be 0.86 g/kg of lead throughput.1 The
size range of participate matter emitted from paste mixing operations prior to
control has not been reported.
9.11.1.3 Three-Process Operation—Stacking/Burning/Assembly. After
curing, the plates are usually sent to the three-process operation, which
includes plate splitting and stacking, burning, and assembly of the elements
into the battery case. All three of these processes emit pollutants to a
common workroom area and, therefore, the process emissions are considered
cumulatively. After being split, plates are stacked in an alternating posi-
tive and negative block formation with insulating separators between the
plates. Separators are made of wood, treated paper, rubber, or plastic.
9.11-4
-------
In the burning operation, leads are welded to the tabs of each positive
and negative plate, fastening the element together. Alternatively, molten
lead can be poured around and between the plate tabs to form the connection.
The elements are completed by welding a positive and a negative terminal to
each element. The completed elements are then assembled into battery cases
either before or after formation.
The major pollutant emitted from these operations is lead from airborne
paste particles. Plate stacking and the burning process emit most of the
lead emissions from this operation; however, significant emissions are also ;
generated during the handling"of plates between processes. Stacks are
straightened by being struck against a grated surface, causing paste particles
to be released into the air. Work areas are usually vented to the atmosphere
to protect the workers' health. Emissions of lead from these processes are
estimated to be 0.56 g/kg of lead processed, the majority of which are in
the inhalable size range.
9.11.1.4 Formation. The formation process converts the inactive lead-
oxide-sulfate paste into an active electrode. Cathodes are formed by oxidiz-
ing the lead oxide in the paste to lead peroxide; anodes are formed by re-
ducing the paste to metallic lead. To accomplish this, inactive plates are
placed in a dilute (10 to 25 percent) sulfuric acid solution with the positive
plates connected to the positive pole of a direct current source, and the nega-
tive plates connected to the negative pole.
Two basic methods of formation are used: wet and dry. In the wet bat-
tery formation process, the elements are assembled into the battery case.
The plates are then formed within the case; this operation generally takes 1
to 4 days. After formation, the spent acid is replaced with fresh acid. Dur-
ing dry formation, battery plates are formed in large open vats of sulfuric
acid prior to battery assembly.
Sulfuric acid mist is emitted during the formation process. The hydrogen
generated by the oxidation-reduction process forms bubbles which carry sulfuric
acid with them as they break the surface of the acid solution, resulting in
an acid mist above the containers. No data are currently available on emis-
sions from the wet formation process. However, based on the slow rate of wet
formation, and the fact that there is usually a lid on the assembled battery,
emissions from wet formation are thought to be minimal.1
9.11-5
-------
Data from one dry formation process showed sulfuric acid emissions
towards the end of a 16-hour formation cycle to be 66 mg/m^. This formation
room produced 20,000 battery plates over a 16-hour period. Emissions are
highest near the end of the formation cycle, when hydrogen production is
greatest.1
9.11.1.5 Lead Oxide Production. The lead oxide used in battery paste
production is approximately 70 percent PbO and 30 percent metallic lead.
Generally, it is produced on-site only at plants of production capacity
greater than 500 batteries per day. Two processes, the ball mill process
and the Barton process, are used to produce lead oxide. In both processes,
a fine dust of lead oxide is formed and conveyed by a circulating air stream
to a hammer-mill for further grinding, and then to a baghouse for product
recovery. Lead oxide emissions in the ball mill exhaust stream are also
recovered in a fabric filter. Emissions from the ball mill process are
estimated to be 0.475 g/kg of lead input.1
9.11.1.6 Lead Reclamation. Clean lead scrap, primarily small parts or
defective grids and plates, can be remelted in a pot-type furnace and cast
into ingots. This is typically done sporadically, only when enough scrap is
available to charge the furnace. Small plants, with production capacity
less than 500 batteries per day, do not have lead reclamation facilities.
It is estimated, based on plant studies, that approximately 1 percent of all
lead entering the lead-acid battery industry nationwide is recycled through
lead reclamation. Uncontrolled lead emissions from this process are high,
estimated to be 298 g/kg of scrap input. Over half of the particles emitted
during lead reclamation are under 15 micrometers in diameter.1
9.11.2 Control Techniques
Particulate emissions from the lead-acid battery industry are currently
controlled by a variety of devices ranging in efficiency from 50 to 99.8 per-
cent. Baghouses, with efficiencies of 96 to 99.8 percent, comprise 60 percent
of the control devices currently in use in this industry. Other devices fre-
quently used are venturi scrubbers, packed-bed scrubbers, and impingement and
entrainment scrubbers. These devices have reported efficiencies ranging from
50 to 98 percent. Devices currently controlling emissions from each process
are briefly described in this section, and are summarized in Table 9.11-2.
9.11-6
-------
Table 9.11-2. PARTICIPATE CONTROL DEVICES USED IN THE LEAD-ACID
BATTERY MANUFACTURING INDUSTRY1
to
Process
Grid casting
Paste mixing
materials charging
Mixing
Three-process
operation
Lead-oxide
manufacturing
Lead reclamation
Formation
Generic
control device/
control technique
Impingement and
entrainment scrubber
Centrifugal and
impingement scrubber
Baghouse
Baghouse
Impingement and
entrainment scrubber
Cascade scrubber
Baghouse
Impingement and
entrainment scrubber
Baghouse
Centrifugal and
impingement scrubber
Impingement and
entrainment scrubber
Good housekeeping
Foam
Mist eliminator
Scrubber/mi st
eliminator
Specific type
of device
Type N roto-clone
Cascade scrubber
Pulse jet type
Shaker type
Type N roto-clone
-
Pulse jet type
Shaker type
Type N roto-clone
Pulse jet type
Shaker type
Cascade scrubber
Type H roto-clone
—
-
Packed tower
Heil fume washer
Overall pollutant
removal efficiency,
% .
90t (lead)
•••
99
98 (lead)
90 (lead)
86 (lead)
97 - 99.3 (lead)
98.6 (lead)
90 (lead) "
-'
98.3 (lead)
N/A
--
-
—
97.5
(sulfuric ncid)
Ai r- to-
cloth
ratio
• N/A
N/A
6:1
4:1 - 8:1
N/A
: N/A
6:1 - 7:1
3:1
N/A
2:1
4:1
N/A
N/A
—
—
—
-
Water-to-gas Pressure Make-up
ratio, drop, water regulrement,
l/m3 pa 1/m3
2.6 1245 - Less than 0.134
0.41
N/A
N/A 249 - 1494 N/A
2.6 1245 Less thian 0.134
500
N/A ' - N/A
N/A --. N/A
2.6 1245 Less than 0.134
N/A 249 - 498 N/A :
N/A 1494 N/A ;
0.53 - 0.7 498 - 747
0.4 - 0.7 2000
..
— — —
.
"— '.' ,
currently in use in this application.
-------
Emissions from several processes or process stages are frequently
vented to a common control device via a collection system of hoods and ducts.
Plant layout and the economics of product recovery are usually the factors
used to determine which processes will be ducted to a specific control device
to form a control system. Table 9.11-3 presents several alternative control
systems that have been proposed.1 These systems rely primarily on control
devices currently in use in the industry, with the exception of a heavier
reliance on baghouses. The use of fabric filters has been avoided in pro-
cesses where exhaust gases have a high moisture content or possible spark
hazards. However, with proper precautions, fabric filters can successfully
be applied to these processes, and operate with lower costs and energy re-
quirements than scrubbers. None of the emissions from lead-acid battery
manufacturing processes are treated by a series of control devices. In
instances where a cyclone or baghouse precedes another device, the first
device is designed for product recovery rather than emissions control. Con-
trol system costs and effectiveness for each of the alternative systems are
presented in Table 9.11-4 and described below.
9.11.2.1 Grid Casting. Emissions from grid casting operations are typi-
cally very low, and are often uncontrolled. Some plants have used impingement
and entrainment scrubbers, such as the Type N Roto-Clone, or cascade scrubbers
to control emissions from both the furnaces and the casting machines. Cascade,
or multiwash centrifugal, scrubbers used to control grid casting emissions can
treat up to 1415 m^/min (50,000 acfm) with water injection requirements as low
as 0.41 l/m3 (3 gal/1000 acf). Frequently, grid casting machines and furnaces
are vented along with other operations, such as small parts casting and lead
reclamation, to a single low-energy scrubber. Fabric filters are presently not
used for grid casting particulate control. Although fabric filters would pro-
vide a higher degree of control (99 percent at a filtering velocity of 1.8
m/min), their use may also require a slight change in casting operations.
9.11.2.2 Paste Mixer. Both baghouses and scrubbers can be used to con-
trol emissions from this source. Most plants use only a scrubber. However,
some plants use a baghouse to collect the dry particulate emissions from the
charging phase and a. scrubber to collect particulate matter emitted during
the wet mixing phase. Impingement entrainment scrubbers, such as the Type N
9.11-8
-------
Table 9.11-3.
SELECTED CONTROL ALTERNATIVES FOR LEAD-ACID
BATTERY MANUFACTURING INDUSTRYl
Control
alternative
Facilities3
Control systemb
MI
III
IV
V
VI
VII
VIII
A, B, F
C, E
G
D
B, C, E
F
A
G
D
C, E
A, B, F
G
D
A, B, C , .
E
F
G
D
A, B, C, F
E
G
D
A, B, C
E
G
A, B, C, E,
G
A, B, C
E
G
Fabric filter, 6/1 A/C
Fabric filter, 6/1 A/C
Mist eliminator
Fabric filter, 2/1 A/C ;
Fabric filter, 6/1 A/C
Impingement and entrainment scrubber
Impingement and entrainment scrubber
Mist eliminator
Fabric filter, 2/1 A/CC
Fabric filter, 6/1 A/C
Impingement and entrainment scrubber
Mist eliminator
Fabric filter, 2/1 A/C
Impingement and entrainment scrubber
Fabric filter 6/1 A/C
Impingement and entrainment scrubber
Mist eliminator
Fabric filter, 2/1
Impingement and entrainment scrubber
Fabric filter, 6/1 A/C
Mist eliminator
Fabric filter, 2/1
Fabric filter, 6/1 A/C
Fabric filter;, 6/1 A/C
Mist eliminator
Fabric filter, 6/1 A/C
Mist eliminator
Impingement and entrainment scrubber
Fabric filter, 6/1 A/C
Mist eliminator
facilities key: A - Grid casting furnace; B - Grid casting machines;
C - Paste mixer; D - Lead oxide manufacturing; E - Three-process operation
and assembly; F - Lead reclaim furnace; G - Formation.
&A11 facilities are vented to common control systems as shown. Air-to-
cloth ratios (A/C) are given for fabric filters.
cSmall (less than or equal to 500 bpd) plants are assumed to have no PbO
manufacturing facilities.
9.11-9
-------
TABLE 9.11-4. SUMMARY OF ALTERNATIVE CONTROL SYSTEMS COSTS AHO CONTROL
EFFECTIVENESS FOR LEAD-ACID BATTERY PLANTS!
Plant
Control capacity,
alternative3 bpd
I 500
2000
6500
II 500
2000
6500
III 500
2000
6500
IV 500
2000
6500
V 500
2000
6500
VI 100
250
VII 100
250
VIII 100
250
Lead emissions,
kg/day
Uncontrolled
6.2
25.0
81.6
6.2
25.0
81.6
6.2
25.0
81.6
6.2
25.0
81.6
6.2
25.0
81.6
1.22
3.04
1.22
3.04
1.22
3.04
Controlled
0.06
0.30
0.99
0.09
0.40
1.31
0.10
0.42
1.43
0.33
1.30
4.38
0.33
1.30
4.41
0.0122
0.0304
0.0122
0.0304
0.0615
0.154
Lead
removal,
%
99.0
98.8
98.8
98.6
98.4
98.4
98.4
98.3
98.3
94.8
94.8
94.6
94.8
94.8
94.6
99.0
99.0
99.0
99.0
94. P
94.9
New plant
control systems
cost, Sl.QOOb
Installed
170
366
838
196
383
859
165
355
808
155
309
680
114
274
637
101
120
136
161
106
125
Annual izedd
67.5
158
379
76.9
172
398
67.6
157
372
50.2
98.1
186
39.4
86.9
173
36.6
43.7
52.3
60.3
34.1
40.9
Existing plant
control systems
cost, $1,000
Installedc
205
447
1020
136 ,
468
1040
199
434
982
187
379
828
138
337
776
121
144
163
193
127
150
Annual izedd
76.7
193
428
87.2
208
449
76.5
192
420
58,5
131
228
45.8
118
122
41.8
49.6
59.2
68.2
39.6
47.1
3A description of each control alternative is presented in Table 9.11-3. There is no control alternative for acid mist; the
cost to control this pollutant was added to the cost of each lead control alternative.
bMid-1975 dollars.
cExcludes SIP compliance costs estimated at $35,000, $35,000, $91,000, $95,000, and $105,000 for plants with capacities of
100, 250, 500, 2000, and 6500 bpd, respectively.
^Excludes costs of controlling facilities that require controls to meet SIP regulations.
-------
Roto-Clone (used to control grid casting emissions) and cascade scrubbers,
are used to control paste mixer emissions.
Fabric filters used to control emissions from the charging phase of this
process usually employ bags made of Orion felt, Polyester, cotton sateen, or
wool. At present, fabric filters are not frequently applied to emissions
from the entire mixing process because of difficulties with blinding of the
bags by moist particulates emitted during the wet mixing phase. Emissions
from the slitting process, part of the three-process operation wherein the
double plates are split, can also be treated by the paste mixer baghouse.
Removal efficiencies from combined slitting and mixer emissions are on the
order of 98 percent.
9.11.2.3 Three-Process Operation. Emissions from stacking, burning,
and assembly operations are usually vented to a common duct prior to cleaning
by either a fabric filter or a scrubber. Effective hooding of the emission
sources is very important in reducing emissions from the three-process opera-
tions due to the fugitive nature of the emissions. In some instances, emis-
sions from the paste mixers are also ducted to the same control system..
Impingement scrubbers with the same pressure drop, makeup water, and
liquid-to-gas ratio requirements used to reduce grid casting emissions are
also used to control the three-process emissions. Collection efficiency is
approximately 90 percent. Shaker-type fabric filters are capable of.reducing
lead emissions from the three-process operation to less than 1.15 mg/nP, with
efficiencies ranging from 97 to 99.3 percent.
9.11.2.4 Lead-Oxide Production. Fabric filters are usually used to
collect the valuable fine lead oxide particles emitted during lead-oxide
production. The baghouses serve the dual purpose of reducing air pollution
and increasing product recovery. Wet scrubbers are not used to control emis-
sions from this process for economic reasons. Air-to-cloth ratios of from 2
to 1 to 4 to 1 are typical for baghouses in this application, thereby prevent-
ing "the collected material from blowing through from one side of the bag to
the other. Two baghouses are frequently installed in parallel to handle the
emissions. Test results from baghouses installed on ball-mill lead-oxide
production facilities indicate that emissions can be reduced to less than
1.1 mg/m3.1
9.11-11
-------
9.11.2.5 Lead Reclamation. Emissions from lead reclamation facilities
are usually controlled by either impingement and entrainment scrubbers, or
cascade scrubbers. Since the character of the exhaust gas stream is very
similar to the exhaust from grid casting operations, emissions from these
processes are often vented to a common scrubber.
9.11.2.6 Formation. Formation, as explained earlier, can take place
either in the battery case or in a large open vat. Emissions from formation
in the battery case are apparently very low, since most plants using this
process do not have ductwork to remove the emissions from the area. Emissions
from the open vat forming process are higher, and are typically controlled by
foam, mist eliminators, or scrubbers. Good housekeeping procedures are used
to control sulfuric acid mist emissions from both forming techniques.
When the plates are being formed in the battery case, sulfuric acid mist
emissions are minimized by forming the plates slowly and keeping the battery
tightly covered. Either the battery filler caps or a reusable battery cover,
to be replaced by the permanent battery top after formation is complete, are
used for this purpose. One manufacturer uses a patented battery filler cap
having a ceramic disc on the inside of the cap. The ceramic absorbs the
hydrogen being emitted during formation. Since the sulfuric acid is carried
into the atmosphere by the hydrogen bubbles, these caps virtually eliminate
acid mist emissions.
Foam covers for the formation vats have been found to reduce acid odors
around the tanks. However, emission measurements indicate that very little
reduction, if any, is achievable through application of this control technique.
Mist eliminators, in which emissions from the vats are ducted to a packed
tower which absorbs mist particles, are frequently used. The packing is
washed periodically, at least once a day and in some instances as often as
two or three times per shift.
Scrubbers are less frequently used to control, formation process emis-
sions. The scrubbers used are usually low energy scrubbers, such as the
Heil fume washer, which is a combined scrubber and mist eliminator. Scrubber/
mist eliminator devices have been found to reduce emissions ,of sulfuric acid
mist by 97.5 percent.1 '
9.11.3 Secondary Environmental Impacts
All control devices used at lead-acid battery plants operate by means
9.11-12
-------
of electrical energy, which is used primarily to drive fans installed to
overcome the pressure drop across the control devices. Electrical energy is
also used to drive scrubber pumps. Energy requirements for particulate pol-
lution control are estimated at 37, 119, and 318 MWh/year for 500, 2000, and
6500 bpd plants, respectively.1 These values.assume a pressure drop of 1245
Pa across all control devices but do not include ductwork pressure drop.
The generation of aqueous wastes by particulate control scrubbers repre-
sents a small fraction of the total wastewater flow and lead content of waste-
water produced by a lead-acid battery manufacturing facility. The incremental
pollutant loadings due to particulate pollution control systems have been
assessed in Reference 1. Incremental flows and lead loadings to be treated
by the wastewater system range from a negligible increase to an incremental
increase of several percent.
The impact of particulate control residues on solid waste production is
also slight in comparison to the total solid wastes generated. The largest
quantity of air pollution-generated solid waste is the sludge generated by
lime treatment of the blowdown from the formation facility control system.
For a 6500 bpd plant, the generation of this sludge represents a 0.15 percent
increase in the solid waste produced by the plant. The solid wastes generated
from the dry collection of lead air pollutants are sent to in-pi ant or outside
reclamation furnaces or smelters for lead recovery. Solid waste practices and
disposal methods for this industry are discussed in Reference 4.
9.11-13
-------
REFERENCES FOR SECTION 9.11
1. Lead-Acid Battery Manufacture - Background Information for Proposed Stan-
dards. Draft Environmental Impact Statement. U.S. Environmental Protection
Agency, Office of Air Quality Planning and Standards. Research Triangle
Park, NC. Publication No. EPA-450/3-79-028a. November 1979.
2. Compilation of Air Pollution Emission Factors. U.S. Environmental Protec-
tion Agency, Office of Air Quality Planning and Standards. Research
Triangle Park, NC. Publication No. AP-42. July 1979.
3. Personal Communication with Lee Beck, U.S. Environmental Protection
Agency, and A. Walter Wyss, Acurex Corporation, Mountain View, CA.
September 1979.
4. McCandless, L. C., et al. Assessment of Industrial Hazardous Waste
Practices - Storage and Primary Batteries Industries. U.S. Environmental
Protection Agency. Publication No. EPA-530/SW-102c. January 1975.
9.11-14
-------
9.12 FUGITIVE DUST SOURCES
Fugitive dust emissions are dusts which become airborne due to forces
of wind, man's activity, or both. Fugitive dust emissions can be categorized
as: (1) anthropogenic sources (those that result directly from and during
human activities) such as agricultural tilling, construction activities,
street dust, inactive tailing piles, and unpaved roads; and (2) wind erosion
sources such as agricultural fields, disturbed soil surfaces, and unpaved
roads. This section discusses fugitive dust sources, emissions, and control
techniques employed to prevent or reduce emissions from agricultural sources,
transportation sources, stockpiles and waste disposal heaps, and construction
sources. Specific control technologies common to a number of the aforemen- ;
tioned sources are discussed in detail in Section 5 of Volume 1.
9.12.1 Agricultural .Sources
It is estimated that agricultural sources such as tilling and wind
erosion of harvested cropland contribute particulate emissions on the order
of 24.1 Tg per year nationwide. As a point of comparison, total point source,
emissions in the U.S. are on the order of 18.1 Tg. per year.1 The dust emis-
sions in agriculture are primarily produced by wind erosion of open fields
and agricultural tilling.
Although it has been reported that wind erosion of non-irrigated open
fields accounts for 90 percent of all, the fugitive dust emissions from agri-
cultural sources, a study performed for Phoenix, Arizona, found that 75
percent of the dust emissions from agricultural sources are due to agricul-
tural tilling and 25 percent are caused by wind erosion.2»3 The latter
study included the irrigation effects on the soil. Periodic irrigation
during the growing season maintains soil moisture such that the soil remains
in an aggregated state. The effects of the irrigation are significant in
the off-growing season, when disconsolidation of the soil and exposure to
winds would reduce resistance to soil erosion.
9.12.1.1 Source Description and Emissions.
9.12.1.1.1 Open fields. The fugitive emissions produced by wind ero-
sion of agricultural soils depend on the soil type and moisture, wind velocity,
vegetation cover, and field and surface geometry. Although many equations
have been developed by researchers for estimating agricultural emissions,
9.12-1
-------
there seems to be a basic agreement that from 2.5 to 10 percent of all the
soil eroded due to wind erosion becomes airborne as suspendable participate
matter.1 The threshold wind velocity, a critical factor for wind erosion,
has been defined as "the wind speed at which the generation of dust begins
due to its aerodynamic forces being of sufficient magnitude to overcome
those forces holding individual particles in the soil."4 The wind forces
cause soil movement by three distinct mechanisms: surface creep, saltation
(jumping), and suspension.4 Minimal information is, however, available
relating these mechanisms and their individual contribution to the dust
generated due to wind erosion.
9.12.1.1.2 Agricultural tilling. Fugitive dust emissions that result
from agricultural tilling are estimated to be 3.0 Tg per year nationwide.1
During a tilling operation, dust particles from the loosening and pulveriza-
tion of the soil are injected into the atmosphere as the soil is dropped to
the surface. Dust emissions are greatest when the soil is dry and during
final seedbed preparation. The factors affecting agricultural tilling fugi-
tive dust emissions are silt content of the soil, implement speed, distribu-
tion of agricultural acreage, moisture,in the soil, and temporal distribution
of tilling activities. Although a variety of implements are employed, includ-
ing disk plows, moldboard plows, and listers, it has been found that the
emissions do not differ greatly from one implement to another.3
A typical particle size distribution of dust emissions from agricultural
tilling has been reported and is presented in Table 9.12-1. Although the
size distribution data were gathered through field measurements in Morton and
Wallace Counties in Kansas, particle sizes may vary according to local soil
characteristics.
9.12.1.2 Control Techniques. Control techniques employed for agricul-
tural sources generally prevent the emissions from becoming airborne rather
than capturing them. Continous cropping, crop residue-limited irrigation of
fallow fields, strip cropping, windbreaks, and chemical soil stabilizers are
the common control techniques used in preventing fugitive emissions from
open fields. Vehicle speed reduction and deflector attachments are the
techniques employed in reducing emissions from agricultural tilling. Table
9.12-2 summarizes these techniques and their relative effectiveness.5
9.12-2
-------
Table 9.12-1. PARTICLE SIZE DISTRIBUTION OF THE DUST EMISSIONS
FROM AGRICULTURAL TILLING3
Particle diameter,
Aim . Weight, %
Less than 2 35
2 to 30 . 45
Greater than 30 20
9.12-3
-------
Table 9.12-2. FUGITIVE DUST CONTROL METHODS FOR AGRICULTURAL SOURCES5
Source
Type of control
Relative
estimated
effectiveness9
Remarks or restrictions
Open fields
Wind breaks
Chemical
stabilizers
Crop plantings
Agricultural Wet suppression
tilling
VP
P
Vehicle speed
reduction and
deflector
attachments
Possible interactions with
plants. May be restrictive
due to cost and temperature.
May be restrictive due to
cost and lack of markets for
off-season crops.
Continual turnover leads to low
efficiency of control.
Additional problems fnclude
the possible short supply of
water and the inability of
cultivating equipment to
carry enough water.
Abbreviations
G = good.
used in this column are: VP = very poor, P = poor, F = fair,
9.12-4
-------
9.12.1.2.1 Open fields. Continuous cropping may be accomplished by
repeated plantings of a single specific crop or by a complex process of
rotating various crop types on a given field throughout the year. Continuous
cropping eliminates the barren period between crops when the soil is exposed
to wind erosion. This technique also increases a field's productivity. The
cost for continuous cropping depends on water availability, manpower and
equipment requirements, crop resource requirements, and crop market value.
Continuous cropping is most effective with crops that do not leave a protec-
tive stubble or residue, such as cotton, sugar beets, beans, or vegetables.
The key limiting factors to continuous cropping are the rainfall and regu-
lated water allocation which restricts the amount of water available for
continuous cropping.
Crop residue or stubble left standing after the crop has been harvested
can often protect a field from wind erosion. Crop residue also improves
soil structure by allowing water to soak into the soil more readily. The
degree of protection from the wind depends upon the quantity and type of
residue, and cropping practices used with the stubble mulching. No-tillage
farming is currently being used as an advanced farming method to prevent
soil erosion, increase cropland production, and reduce farming costs.3
Plowing in the spring instead of in the fall, and planting a new crop in old
stubble can reduce fugitive dust.
When a field is barren (after harvest, between crops, or after planting),
dust emissions can be reduced by irrigating at frequent intervals. Watering
the field forms a thin surface crust which protects the undisturbed soil for
some time after the surface has dried. However, the cost and the availability
of water for this procedure are important considerations.
Strip cropping and inter-row planting of grains protect erosion-
susceptible crops or fallow areas with erosion-resistant crops. For maximum
effectiveness, the strips or rows are planted as nearly perpendicular to the
prevailing wind direction as possible. This method, if well planned, does
not remove any land from cultivation, and may not require any change in
cropping practices. Strip cropping may be used most effectively during the
early months of crop development.
Windbreaks along the edges of cultivated fields can reduce surface wind
velocity and soil blowing. Various physical barriers and vegetation have
9.12-5
-------
been illustrated in a comprehensive U.S. Department of Agriculture publica-
tion, Windbreaks for Conservation.6
While a field is in the seedling stage or is barren, wind erosion can be
reduced considerably with chemical stabilization. Liquid petroleum resin-in-
water emulsion is the most effective, durable, and economical stabilizer.
Herbicides must be used with stabilizers since stabilizers provide surface
layer protection only, and normal weed removal practices would disturb the
protective layer.2 Wind-blown dust emissions from agricultural lands can
be reduced by about 90 percent if the surface layer is undisturbed.6 Table
9.12-3 presents control efficiencies and costs of controls in reducing fugi-
tive emissions.
9.12.1.2.2 Agricultural tilling. Fugitive dust emissions from tilling
operations can be controlled by using deflector attachments for farm implements
and by reducing the speed of the equipment in the fields. The degree of reduc-
tion in emissions by these sources has not been established. Another way to
control fugitive dust from tilling operations is to water the field before
plowing it. However, this procedure can make soil unworkable and can adversely
affect plowed soil.
No-tillage farming is currently being used as an advanced farming method
to prevent soil erosion, increase cropland production, and reduce farming
costs.^ Despite the economic benefits of no-tillage farming, there is sub-
stantial resistance by farmers to depart from accepted practice.
9.12.2 Transportation Sources
Unpaved roads, paved roads, unpaved airports, and transport of material
by truck or train largely make up fugitive dust emission sources that result
from transportation activities. These emissions are generated due to vehicu-
lar traffic and wind erosion.
9.12.2.1 Source Description and Emissions
9.12.2.1.1 Unpaved roads and airstrips. Fugitive dust emissions from
unpaved roads and airstrips are affected by surface texture of the road, road
material, surface moisture, and vehicle speed and type. Unpaved roads have
been identified as the single largest source of fugitive dust emissions,
contributing on the order of 295 Tg per year of particulate emissions.1
Fugitive emissions from unpaved roads can be estimated using emission
factors developed in EPA Publication AP-42, Compilation of Air Pollutant
9.12-6
-------
Table 9.12-3. COST AND CONTROL EFFICIENCIES FOR FUGITIVE DUST
CONTROL TECHNIQUES FOR OPEN FIELDS3
Control method
Control .
efficiency,
Unit cost,
$/unit
Continuous cropping
Crop residue
Limited irrigation
Stripcropping
Inter-row planting
Windbreaks
Spray-on chemical
stabilizer
25
10
20
27
15
6
40
Dependent on crop
No data
1.5 to 4/hectare per year
Dependent on crop
No data
No data
8 to 20/hectare
per application
9.12-7
-------
Emission Factors, if the silt content of the road surface (percentage by
weight of particles smaller than 75 micrometers in diameter), average vehicle
speed, and average daily traffic are known. Average vehicle speed and average
daily traffic are frequently difficult to determine since there is generally
little incentive to study traffic characteristics on roads carrying limited
traffic.
Fugitive dust emissions from unpaved roads generally exhibit a particle
size distribution consisting of 60 percent of the particles having a diameter
of less than 30 micrometers.3
Wind erosion and airplane landings and takeoffs cause fugitive dust
emissions on unpaved airstrips. Although these emissions are similar to the
unpaved roadways, annual landing/take-off (LTO) cycles are one technique
for estimating fugitive emissions from unpaved airstrips. The Federal Avia-
tion Administration estimates that approximately 400 to 500 small aircraft
operations per year occur at small airports, with a typical value being 500
operations, or 250 LTO cycles.7
Factors influencing the emissions from unpaved airstrips include the
following:
o Surface texture, measured as percent silt content
o Average LTO speed
o Surface soil moisture as measured by annual number of dry days
o Length of runway used for one complete LTO cycle
o Wind erosion
9.12.2.1.2 Off-road vehicles. Fugitive dust emissions generated due to
off-road vehicles occur mostly during weekends in less populated areas. These
dust emissions are affected by the silt content of the soil, vehicle speed,
and nuniber of wheels per vehicle. Minimal information is available, however,
regarding the quantity of fugitive dust generated by off-road vehicles nation-
wide. In one study, particle size distribution of the fugitive dust due to
off-road vehicles was estimated to be the same as that for unpaved roads.3
9.12.2.1.3 Unpaved parking lots and truck stops. The fugitive dust
emissions from unpaved parking lots and truck stops are influenced by the
number of vehicles using the lot/stop each day and the distance traveled by
each of the vehicles. The silt content of the soil, varying from 12 to 24
9.12-8
-------
percent depending on the surface, also affects the generation of dust emis-
sions. The particle sizes of the fugitive emissions from the unpaved parking
lots and truck stops are estimated to be similar to those from unpaved roads.3
9.12.2.1.4 Paved roads. It is estimated that fugitive dust emissions
from paved roads contribute 8 Tg per year nationwide.! In recent years, con-
siderable attention has been focused on fugitive dust emissions from urban
paved roads. Microscopic analysis of samples taken from urban ambient air
sampling stations, where total suspended particulate levels are higher than
expected, has identified dust emissions from paved streets as a major cause
of non-attainment of the primary national ambient air quality standard.8
Figure 9.12-1 shows the material balance of suspended particulate matter
near urban streets. Although primarily generated by vehicular traffic, the
dust emissions increase when the wind velocity exceeds the threshold value
of about 21 kilometers per hour.8 Some studies have also shown a direct
correlation between the traffic volume and dust deposition due to traffic.9
The major street surface contaminants are mineral-like matter similar to
common sand and silt. Typically, 78 percent of the material is located with-
in 15 cm of the curb and 88 percent within 30 cm of the curb.8 The sources
of fugitive dust on paved streets are as follows:
o Wind erosion of unpaved parking lots and other exposed areas
o Motor vehicle exhaust, lubricant leaks, and tire wear
o Truck spills
o Street repairs
o Winter sanding and salting
o Vegetation and litter
o Atmospheric dustfall
The silt content (particles smaller than 75 micrometers) is in the 5 to
15 percent range for surface dust from paved streets.8 The particles under
75 micrometers in diameter include a large percentage of the total heavy
metals and pesticides. An American Public Works Association study found that
approximately 5 Kg of the dust under 0.31 cm in size comes onto each 30 m
of curbless paved roads in Chicago each day; this amount is reduced by a
factor of four if curbs are added.8 Particle size distribution of the dust
emissions from paved roads is presented in Table 9.12-4.
9.12-9
-------
BACKGROUND PARTICIPATE MATTER
LOCAL
VEHICLES
GROUND-LEVEL
"SUSPENDED
PARTICULATES"
(h < 10m)
ACCUMULATED
STREET DEPOSITS
URBAN
SOURCES —
CONVENTIONAL
AND FUGITIVE
ENTRAPMENT
(BY WIND AND
VEHICLE MOTION)
VEHICULAR DEPOSITS
(CARRYOUT FROM UNPAVED AREAS,
TIRE WEAR, OIL, ETC.)
RUNOFF MECHANICAL REMOVAL
(SEWERS) (STREET CLEANERS)
h = height above
ground level.
Figure 9.12-1 Diagram of street surface/atmospheric exchange
particulate matter.8
of
9.12-10
-------
Table 9.12-4. PARTICLE SIZE DISTRIBUTION OF THE DUST EMISSIONS
FROM PAVED STREETS^
Particle size, Weight of the dust
emissions, %
greater than 30 10
less than 30 90
less than 5 50
9.12-11
-------
9.12.2.1.5 Transport of material by truck or train. Fugitive dust
emissions in the transport of material by truck or train occur due to loading,
unloading, and spillage. The control of dust emissions due to loading and
unloading is discussed in Section 5 of Volume 1. The dust emissions due to
spillage or leakage occur from uncovered trucks and railroad cars. Minimal
data are available on the quantity of emissions or particle size distribution
of the dust particles as a result of spillage.
The factors affecting fugitive dust emissions from transporting of
material, excluding loading and unloading operations, are vehicle speed, silt
content of the bulk material transported, moisture content of the material,
vehicle body configuration, and condition of the transport vehicle.
9.12.2.2 Control Techniques. Fugitive dust emissions from transporta-
tion sources can, in general, be controlled by either temporary methods such
as wetting the road surface or permanent methods such as paving the unpaved
roads. Table 9.12-5 presents different techniques employed for each type
of source and their relative effectiveness.
9.12.2.2.1 Unpaved roads and airstrips. Fugitive dust emissions from
unpaved roads can be controlled by paving, oiling, watering, or applying a
surface chemical treatment. Paving can be very costly if the roadway has a
low-traffic density and is fairly long. However, up to 85 percent control
efficiency can be achieved with paving.10 Maintenance costs could be sub-
stantially reduced since it would no longer be necessary to blade and regrade
the roadways. Even after paving the streets, reentrainment of particles, re-
quires implementation of control techniques described in subsection 9.12.2.2.4.
Oiling (using locally available petroleum by-products)-or surface treat-
ment (using stabilization chemicals on the roadbed) is less costly than
paving the roadway. Fugitive dust can be suppressed up to 50 percent by
using stabilization chemical's.2 In one study, chemical stabilizers were
sprayed on an unpaved road and mixed to a 7- to 5-cm depth. The stabilizers
proved most effective 5 months after application. Effectiveness in suppress-
ing fugitive dust was as high as 95 percent.6 Stabilizing the roadbed has
considerable potential as an interim control procedure, since the roadbed
can later be used as a base for paving.
Watering is a much less efficient method of controlling dust on unpaved
9.12-12
-------
Table 9.12-5. FUGITIVE DUST CONTROL METHODS FOR TRANSPORATION SOURCES^
CO
Source
Unpaved roads
and airstrips
Unpaved roads
Paved roads
Transport of
fines by
truck or train
Road shoulders
Type of control
Wet suppression
Stabilization
Paving
Speed reduction
Washing
Vacuuming
Wetting
Covering (tarps)
Enclosure
Stabilization
Vegetation
Relative
estimated
effectiveness9
VP
P
G
Variable
P
P
P
F
G
F
G
Remarks or restrictions
Temporary
Temporary
Costly
Costly, temporary
Costly, temporary
Temporary only
Problems occur during loading
and from leakage, also costly.
Abbreviations used in this column are: VP = very poor, P = poor, F = fair, G = good.
-------
roadways because of the high frequency of application required. Watering
can be effectively used for temporary control of emissions, or where the
watering equipment is already available and roads are confined to a single
site, such as construction access roads or mining haul roads.
Reducing vehicle speed on unpaved roads is another technique that can
be used for fugitive dust control. Particulate reductions of 62 percent can
be achieved by lowering the average speed from 56 km per hour to 32 km per
hour.^
The cost of controlling fugitive dust emissions from unpaved roads
varies from region to region. A study conducted by the City of Seattle
Engineering Department has shown that the most cost-effective method of dust
control on Seattle roadways is a chip seal when the average daily traffic is
over 100 vehiclesJl Other factors such as traffic density, the length of
the road, and the type of controls also affect cost. Table 9.12-6 shows
annual maintenance and initial cost per length of roadway in Maricopa County,
Arizona.3 Several studies have shown that chip-seal surfacing is more
cost-effective than other road surfacing dust control measures.
The control techniques employed in reducing fugitive dust emissions from
unpaved airstrips are similar to those employed in controlling dust from un-
paved roads, except for traffic control.
9.12.2.2.2 Off-road vehicles. Feasible control methods to reduce fugi-
tive emissions from off-road vehicles include chemical stabilization, vege-
tation, and physical covers. Chemical stabilization requires frequent appli-
cation to maintain the reduction in dust emissions. The effectiveness of
vegetation in reducing fugitive dust emissions depends on the density and
nature of the growth. In areas that will support heavy vegetation, the dust
emissions due to wind erosion may be eliminated. However, in areas less
hospitable to plant growth, such as the arid southwest, only native species
may be grown (sagebrush, Indian rice grass, sand dropseed). Physical covers
have limited application in reducing fugitive emissions from off-road vehicles
because of costs.
9.12.2.2.3 Unpaved parking lots and truck' stops. Fugitive dust emis-
sions can be controlled from unpaved parking lots by using control techniques
similar to those for unpaved roads, as described above. Table 9.12-7 presents
9.12-14
-------
Table 9.12-6. INITIAL COST AND MAINTENANCE COST OF ALTERNATIVE ROAD
SURFACES APPLIED BY MARICOPA COUNTY HIGHWAY DEPARTMENTS
Initial cost, Annual maintenance,
Road surface type $/km $/km
Gravel road 10,000 378
Oiled surface (low- 1,260 to 1,890 1,260 to 1,890
cost application)
Oiled surface (dust 3,340 3,340
control oil)
Chip seal coat 22,000 500
7.5-cm asphalt 34,650 to 63,000 100
9.12-15
-------
Table 9.12-7. EFFECTIVENESS AND COST OF ALTERNATIVE MEASURES TO CONTROL DUST EMISSIONS
FROM UNPAVED PARKING LOTS OR TRUCK STOPS3
ID
•
H-»
ro
Emission rate,3
Control kg/vehicle km
Gravel surface
Oil surface
(dust control oil)
Oil surface
(low cost application)
Chip seal coat
Asphalt
0.3
0.17
0.3
Ob
Ob
Efficiency Initial
of control, cost of control
% $/1000 m2
50
75C
50
100
100
1080
360
140 to 200
2400
3780 to 6880
Cost of
, maintenance,
$/1000 m2
40
360
140 to 200
55
13
aRefers to emission rate for normal passenger vehicle. For truck stops, multiply stated
emission rate by N/4, where N = number of wheels on trucks.
bThis emission rate does not include entrainment of dust loadings off the pavement.
cBased on field tests conducted by Arizona Department of Transportation.
-------
control techniques, efficiencies, and costs.
9.12.2.2.4 Paved roadways. The entrainment of the dust from paved road-
ways can be reduced by two methods: (1) controlling street dust origins, and
(2) street cleaning. As previously discussed, noncurbed roads generate more
fugitive dust emissions than curbed roads by a factor of four to one. There-
fore, providing roadway curbs is one possible control measure. Typical city
construction costs for street curbs are about $15 per meter.3 A further
reduction in fugitive emissions from wind erosion is possible by using soil
stabilizers in the soil adjacent to the curbs.
Street cleaning can also be used to control dust. In many cities, a
flusher attached to the conventional machine sweeper jets water onto the
streets, moving materials to the gutter. Broom sweeping is not effective on
small particles from reentrainment. Flushing, however, wets the street,
causing dust suppression until the surface is completely dry. In one study
performed in the New York - New Jersey area, it was found that the particulate
concentrations near the streets were consistently lower on days with flushing
or days after flushing than on non-cleaning days. The average reduction in
the particulate concentration was found to be 16.7 micrograms per m3 or 15.7
percent.9
Many cities have replaced their street cleaning equipment such as broom
sweepers with either vaccum sweepers or regenerative air units which are
similar to vacuum sweepers. In a study performed in Charlotte, North Carolina,
it was found that the air quality in the urban area generally improved after
the broom sweepers were replaced with regenerative air sweepers and the
streets were flushed following sweeping. The flushers, however, use 6,000
to 10,000 liters of water per kilometer of the street or up to 250,000 liters
per day. It is estimated that street flushing could consititute 1 to 2 per-
cent of a city's total water consumption.3
9.12.2.2.5 Transport of material by truck or train. Control techniques
for dust emissions occuring as a result of loading and unloading of the mate-
rials from truck or train are discussed in Section 5 of Volume 1. Dust emis-
sions due to leakage and spillage of the material from trucks can be controlled
using tarps as covers on the open trucks. Complete enclosure of the truck
body will allow total dust control, although costs may be prohibitive.
9.12-17
-------
9.12.3 Aggregate Storage Piles and Haste Disposal Heaps
Fugitive dust emissions from stockpiles and waste disposal heaps occur
due to wind erosion of the exposed surface of a stockpile or heap and the
turnover operations of the stockpile which fluctuate with daily or weekly
demands. Minimal information is available on the nationwide particulate
emission rate of fugitive dust from stockpiles and waste disposal heaps.
9.12.3.1 Source Description and Emissions
9.12.3.1.1 Aggregate storage piles. An inherent part of the operation
of many plants that utilize minerals in the aggregate form is the maintenance
of outdoor storage piles. Storage piles are usually left uncovered, primarily
because of the necessity for frequent transfer of material into or out of
storage. Dust emissions occur at several points in the storage cycle: dur-
ing loading of the material onto the pile, whenever the pile is acted on by
strong wind currents, during equipment and vehicle movement in the storage
area, and during loadout of material from the pile. These four major emission-
producing activities contribute to the fugitive dust emissions in the following
proportions:10
Percent of fugitive dust
Activity emissions from the storage pile
Loading onto piles 12
Equipment and vehicle movement
in the storage area 40
Wind erosion 33
Loadout from piles 15
Fugitive dust emission rates from the stockpile are dependent on: (1)
turnover rate for a pile, (2) operations for adding and removing material,
and (3) pile configuration. The dust emissions from the stockpile would be
chemically the same as the materials in the pile. Particle size distribution
of the uncontrolled fugitive emissions from the stockpiles has been found to
be somewhat independent of the material stored because only the smaller
particles (less than 100 micrometers in size) become airborne. Table 9.12-8
lists typical sizes.
9.12-18
-------
Table 9.12-8. PARTICLE SIZES OF FUGITIVE DUST EMISSIONS FROM
AGGREGATE STORAGE PILESl°
Size range, Percent by weight
tm of emissions
less than 3 30
3 to 30 23
greater than 30 47
9.12-19
-------
9.12.3.1.2 Waste disposal heaps. Mineral mining and benefication produce
wastes in the form of overburden and tailings. Coal mining and coal preparation
usually produce both fine and coarse waste materials. These materials consist
of low grade coal, flyash, carbonaceous and pyritic shale, slate, clay, and
sandstone. Metallic tailings such as copper tailing, uranium tailings, and
iron tailings often present varied and extreme problems in the application
of a control technique due to their variable pH and toxic properties.12
As in the case of open storage, emissions arise from dumping and wind
erosion across unprotected surfaces. Since the waste heaps are generally
not disturbed after dumping, there are no emissions from an activity compar-
able to loading out of a stockpile. However, there may be emissions from
transporting the waste material on-site or from a reclamation process such
as landfill covering associated with the waste disposal operation.
The particle size distribution of the fugitive emissions from tailing
piles is estimated to be similar to that from aggregate loadout operations;
the distribution is presented in Table 9.12-9.
9.12.3.2 Control Techniques. Fugitive dust emissions from stockpiles
and waste disposal heaps can be controlled by temporary methods such as
wetting or stabilization or by permanent methods such as enclosure for stock-
piles or vegetation for waste disposal heaps. Table 9.12-10 presents the
types of controls employed and their relative effectiveness.
9.12.3.2.1 Aggregate storage piles. Watering of the stockpiles and
surrounding areas is the most common technique, but its effects are quite
temporary and watering sometimes adversely affects the capability to handle
the material easily. A more effective, longer-duration method of dust control
is the addition of chemicals to the water sprayed onto the aggregate. Rather
than acting as chemical soil stabilizers to increase cohesion between particles,
most of these chemicals work as wetting agents to provide better wetting of
particles (smaller than 100 micrometers in size) and longer retention of the
moisture film. The system of application can be a continuous spray onto the
aggregate during processing or a water truck with hose and spray nozzle.
Enclosing the materials in storage is generally the most effective means
of reducing emissions from the stockpile because it allows the emissions
from the stockpile to be captured. However, storage bins or silos may be
9.12-20
-------
Table 9.12-9. PARTICLE SIZE DISTRIBUTION FOR DUST EMISSIONS FROM
TAILING PILES3
Particle size, Weight percent
m of dust emissions
less than 1 30
1 to 2 46
2 to 3 16
3 to 4 6
greater than 4 4
9.12-21
-------
Table 9.12-10. FUGITIVE DUST CONTROL METHODS FOR STOCKPILES AND
HASTE DISPOSAL HEAPS5
Source
Type of control
Relative
estimated
effectiveness3
Remarks or restrictions
••£>
ro
r>o
ro
Aggregate
storage pile
Waste disposal
heaps
Wetting
Stabilization P
Enclosure F to G
Wind screen VP
Separation of fines F
that are sent to
enclosed areas
Wetting P
Stabilization P to F
Vegetation F to G
Physical stabilization F to G
Continuous operations on
stockpiles preclude
effective control.
Same as wetting.
May not be practical for all
types of operations.
Extra cost.
Temporary only.
Efficiency depends on type
of material, stablizer,
etc. Temporary.
May be expensive due to
cost of pretreating
(fertilizing, etc.).
^Abbreviations used in this column are: VP = very poor, P = poor, F = fair, G = good.
-------
very expensive. One alternative to the enclosure of all the material is to
screen the material prior to storage, sending the oversize material to open
storage and the fines to silos. Windbreaks,or partial enclosure of storage
piles can reduce wind erosion losses; however, these techniques do not permit
capture of the remaining storage pile fugitive dust emissions.
Telescopic chutes, flexible chute extensions, and travelling booms are
used to minimize the free fall of material onto the pile, and thereby decrease
the resulting dust emissions. Dust emissions occuring as a result of loadout
activity can be reduced by reclaiming the material from the bottom of the
stockpile with a mechanical plow or hopper system. The use of telescoping
chutes and flexible chute extensions for piles with high material flow rates
may require closer attention because of the possiblity of jamming. Table
9.12-11 shows the control efficiencies for each type of control and the
stockpile activity.
Cost estimates for these control techniques vary widely. The capital
costs of enclosed storage vary from $107 to $255 per cubic meter of capacity.
The wetting agents and their application costs range from $0.01 to $0.05 per
Mg.10
9.12.3.2.2 Waste disposal heaps. Control methods for fugitive dust emis-
sions from mineral waste heaps include: (a) physical control, (b) chemical
binding, and (c) vegetation cover. The applicability and cost of these
controls vary, depending on the type of mineral waste and the region in
which it is located.
Physical stabilization methods can be used for controlling fugitive dust
from inactive waste heaps. It requires the covering of the exposed surface
with a material that prevents the wind from disturbing the surface particles.
Common physical stabilizer materials for'inactive waste heaps include rock
soil, crushed or granulated slag, bark, and wood chips. The control efficiency
of this technique depends on the type of material and type of stabilizer.
Physical stabilization of tailings with a cover rock or smelter slag
can provide complete control of wind-blown emissions. A mixture of soil and
rock available from adjacent lands is a more widely used cover material.
Soil cover is subject to wind erosion to a lesser degree than the tailings,
and permits a habitat for encroachment of local vegetation.3 The primary
9.12-23
-------
Table 9.12-11. POSSIBLE CONTROL TECHNOLOGY APPLICATIONS FOR OPEN STORAGE
PILES10
Emission points
Control Procedures
Efficiency,
Loading onto piles Enclosure
Movement of pile
Wind erosion
Chemical wetting
agents or foam
Adjustable chutes
Enclosure
Chemical wetting agents
Watering ,
Traveling booms to
distribute material
Enclosure
Wind screens
Chemical wetting agents
or foam
Screening of material prior
to storage, with fines sent
directly to processing or
a storage silo
70 to 99
80 to 90
75
95 to 99
90
50
No estimate
95 to 99
Very low
90
No estimate
9.12-24
-------
drawback to physical covers as wind controls is the high cost of application,
particularly when the cover materials are unavailable in the immediate area.
Chemical stabilizers are commercially available and have been employed
to create a crusted erosion-resistant layer on mineral waste heaps. Since
chemical layers create only a thin skin of protection, they offer only tem-
porary protection, and repeated applications are required periodically to
maintain the crust. Chemical stabilizers are typically used in combination
with vegetation to form long-term erosion-resistant surfaces over the tailing
piles. The chemicals promote the growth of vegetation and protect the seeds
during the germination period.
Vegetation can be effectively used to stabilize a variety of exposed
surfaces. In many cases, however, modifications must be made to the surface
or the surrounding terrain before effective stabilization can occur (e.g.,
fertilization, pH modification, and slope reduction). Vegetative stabiliza-
tion for the control of fugitive dust is restricted to inactive areas where
the vegetation will not be mechanically disturbed once it is started. These
sources can include coal refuse piles and mineral waste disposal heaps.
The difficulty encountered in the application of vegetation stabiliza-
tion of coal refuse piles occurs as a result of the acidic nature of the
wastes and from the slopes of the piles' sides. Thus, chemical or physical
treatment of the piles' components must be accomplished prior to effective
= »
vegetation stabilization. Chemical treatment usually involves the addition
of soil-neutralizing material such as agricultural limestone. Other materials
such as flyash, mined phosphate rock, or treated municipal sewage sludge have
also been used.12 Many species of plants have been used for the stablization
of waste heaps; e.g., grasses, legumes, trees, shrubs, and vines.
The control efficiency of vegetation stab!ilization varies considerably
with differences in the amount and type of cover established for waste heaps
or tailing piles. Estimated efficiencies in the range of 50 to 80 percent
have been reported.12 Table 9.12-12 summarizes the efficiencies and costs
of controls in reducing fugitive emissions from tailing piles.
9.12.4 Construction
Construction sources contribute particulate emissions, fugitive in na-
ture, on the order of 24.8 Tg per year nationwide according to one estimate.1
9.12-25
-------
Table 9.12-12. EFFECTIVENESS AND COST OF CONTROL MEASURES FOR EMISSIONS
FROM TAILINGS PILES3
Control measure
Percent of
emissions reduced
Cost of measure,
$/hectare
Rock or slag cover
Chemical stabilization
Vegetation
Chemical stabilization,
vegetation
100
80
25 to 100
85 to 100
140-180 (available local
385-415 (transported)
26-2603
40-180
40-60
iy)b
^Applications of chemical stabilizers are typically required on an annual
basis.
&Plus cost of stabilizing the borrow area. Too often the borrow area is
ignored and, subsequently, becomes more of a pollution source than the
mineral area.
9.12-26
-------
Fugitive dust emissions at construction sites are generated by such operations
as land clearing, blasting, ground excavation, cut and fill operations, and
construction. Demolition of the existing structures produces fugitive emis-
sions due to the free fall of the material demolished.
o
9.12.4.1 Source Description and Emissions
9.12.4.1.1 Excavating. Fugitive dust emissions in construction opera-
tions are mainly due to excavation, vehicle and equipment operation, and
wind erosion of the exposed earth surfaces. Although heavy construction is
usually of short duration, earth-moving activity is the major source of fugi-
tive emissions. The exposed earth is susceptible to wind erosion and to dust
emissions from infrequent traffic disturbance* The fugitive emissions from
excavating are affected by the amount of construction activity and weather
conditions.1
The dust generation from a mechanical contact process, such as excavating,
is generally insensitive to the ambient wind speed; however, wind speed does
determine the drift distance of large dust particles and, therefore, the
localized impact of the fugitive dust source. On the other hand, the genera-
tion of suspendable particles by wind erosion of exposed surface is very
sensitive to the wind speed.
9.12.4.1.2 Demolition. The fugitive dust emissions at demolition sites
result essentially from the same source as those found on construction sites.
These sources involve earth-moving activities and general disturbance of
soil. A significant portion of the dust associated with demolition activities
may also be generated by falling walls, and an additional significant emission
hazard would be the release of asbestos particles when demolition involves
friable asbestos materials. The potential asbestos emissions hazard has re-
sulted in the promulgation of demolition and renovation standards for insti-
tutional, industrial, and commercial buildings containing a specified amount
of friable asbestos material.
9.12.4.2 Control Techniques. Wetting and stabilizing are the common
control techniques employed in preventing arid/or reducing fugitive dust
emissions from excavating and demolition. Table 9.12-13 shows relevent
types of controls and their relative effectiveness.
9.12-27
-------
Table 9.12-13. FUGITIVE DUST CONTROL METHODS FOR CONSTRUCTION SOURCE$5
ro
CO
Source
Vehicle -travel
Demolition
Type of control
Relative
estimated
effectiveness3
Remarks or restrictions
Excavating
Heaping of
excavated
Wetting
Wetting
Stabilizing
VP to P
P
F to G
Continual working precludes
effective control.
Temporary only.
Stabilizing with a binder is
See Table 9.12-5
Unpaved roads
Wetting
P to F
an effective control method
that is applicable to short-
term heaping of excavated
material.
Demolition may cause high,
short-term exposure to
asbestos from building
materials.
Abbreviations used in this column are: VP = very poor, P = poor, F = fair, G =• good.
-------
9.12.4.2.1 Excavating. Wetting or watering of construction sites pro-
duces a wide variation in apparent control efficiencies. This variation is
partly due to the highly variable nature of the emission sources. However,
wetting the surfaces of unpaved access trails for construction vehicles and
trucks effectively controls dust emissions, provided the surface is kept wet.
A study of the effect of watering on construction sites indicates that exten-
sive watering of the soil may reduce emissions from existing construction
operations by 60 to 70 percent.3 The study suggested that wetting of access
roads twice a day with an application of 2 liters of water per square meter
will suppress dust emissions by 50 percent. Obviously, heavy wetting may
cause mud. Unless cleaned up at the site, the mud could be conveyed on to
adjacent streets where, after becoming dust again, it would be susceptible
to reentrainment by passing vehicles.
Wind erosion of the exposed earth is another major fugitive dust source
at a construction site. Soil stabilizers can be effective in reducing wind
erosion. Good construction management will only expose earth that is being
worked on, thereby reducing possible wind erosion.
Costs for controlling fugitive dust emissions vary from site to site,
depending on water availability, traffic, street-sweeping costs, chemical
soil stabilizer costs, and the degree of control. Table 9.12-14 shows con-
trol efficiencies and their costs in reducing fugitive emissions from
excavating activities in the Phoenix, Arizona, area.
9.12.4.2.2 Demolition. The control methods available for demolition
sources are essentially the same as those employed at construction sites, as
discussed above. Asbestos emissions are the major health concern. The pro-
mulgation of National Emissions Standards for Hazardous Air Pollutants requires
that asbestos material be first removed prior to wrecking activities by speci-
fied handling procedures, and that these materials be wetted prior to removal
and handling.3 The dust created by falling walls of brick, plaster, or
concrete may be mitigated by spraying the walls with water before teardown
and immediately after the fall. This control method is estimated to reduce
fugitive dust emissions from masonry demolition by 10 to 20 percent.3
9.12-29
-------
Page Intentionally Blank
------- |