United States Office of Air Quality EPA-450 3-83-01 Oa
Environmental Protection Planning and Standards April 1983
Agency Research Triangle Park NC 27711
__ ___ — —
v>EPA Inorganic Arsenic Draft
Emissions from EIS
Low-Arsenic
Primary
Copper Smelters -
Background
Information for
Proposed Standards
-------
EPA-450/3-83-010a
Inorganic Arsenic Emissions from
Low-Arsenic Primary Copper Smelters -
Background Information
for Proposed Standards
Emission Standards and Engineering Division
U S ENVIRONMENTAL PROTECTION AGENCY
Office of Air, Noise, and Radiation
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
April 1983
-------
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
™nstitLi;e endorsement °r recommendatlon f°r use. Copies of this report are available through the Library Serv.ces
urnce (MD-35), U S Environmental Protection Agency, Research Triangle Park, North Carolina 27711; or, for a fee from
the National Technical Information Services, 5285 Port Royal Road, Springfield, Virginia 22161
Publication No. EPA-450/3-83-010a
-------
ENVIRONMENTAL PROTECTION AGENCY
Background Information and Draft
Environmental Impact Statement
F H M „ for1Pr/Imary Copper Smelters Processing
Feed Materials Containing Less Than 0.7 Percent Arsenic
As^~)
fl /// / Prepared by:
"Txr^MAA-1
x— ... .-armer~
Director, Emission Standards and Engineering Division
U. S. Environmental Protection Agency
Research Triangle Park, NC 27711
1.
2.
3. The comment period for review of this document is 60 days
period"6 ^ be contacted regarding the date of the comment
4. For additional information contact:
Mr. Gene W. Smith
Standards Development Branch (MD-13)
U. S. Environmental Protection Agency
Research Triangle Park, NC 27711
telephone: (919) 541-5624.
5. Copies of this document may be obtained from:
U. S. EPA Library (MD-35)
Research Triangle Park, NC 27711
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
-------
TABLE OF CONTENTS
Section page
1.0 SUMMARY 1_1
1.1 Statutory Authority 1-1
1.2 Regulatory Alternatives 1-1
1.3 Environmental Impacts 1-2
1.4 Economic Impacts 1-3
2.0 THE PRIMARY COPPER INDUSTRY 2-1
2.1 General 2-1
2.1.1 Raw Materials 2-1
2.1.2 Process Description 2-5
2.2 Arsenic Behavior and Distribution in Copper Smelters. . . 2-19
2.2.1 Arsenic Behavior During Roasting 2-19
2.2.2 Arsenic Behavior in Smelting Furnaces 2-22
2.2.3 Arsenic Behavior During Converting 2-27
2.2.4 Arsenic Balance 2-30
2.3 Arsenic Distribution and Emissions at Domestic Low-
Arsenic Throughput Copper Smelters 2-33
2.3.1 Process Arsenic Emissions 2-34
2.3.2 Fugitive Arsenic Emissions 2-36
2.4 References 2-60
3.0 CONTROL TECHNOLOGY 3_1
3.1 Alternative Control Techniques 3-1
3.1.1 Process Emission Controls 3-1
3.1.2 Fugitive Emission Sources and Controls 3-17
3.2 Performance Capabilities of Alternative
Control Techniques for Arsenic and Total
Particulate Emissions 3.49
3.2.1 Process Control Systems 3-49
3.2.2 Fugitive Control Systems Evaluation 3-64
3.2.3 Conclusions 3_8i
3.3 References 3.35
-------
TABLE OF CONTENTS
(continued)
Section
Page
4.0 MODEL PLANTS, REGULATORY BASELINE, AND REGULATORY
ALTERNATIVES 4.4
4.1 Regulatory Considerations 4-1
4.1.1 Clean Air Act 4.3
4.1.2 Arsenic Regulation by the Occupational
Safety and Health Administration 4-7
4.1.3 Clean Water Act 4_g
4.1.4 Resource Conservation and Recovery Act
(RCRA) 4.9
4.2 Baseline and Regulatory Alternatives 4-10
4.2.1 Definition of Baseline 4-10
4.2.2 Description of the Regulatory Alternatives .... 4-10
4.3 Baseline Configuration, Baseline Arsenic Emissions,
and Regulatory Alternatives for Model Plants 4-12
4.3.1 Baseline Arsenic Emissions 4-12
4.3.2 Model Plant Baseline Configurations and
Regulatory Alternatives 4-16
4.4 References 4.39
5.0 ENVIRONMENTAL IMPACTS 5_1
5.1 Introduction 5_1
5.2 Air Pollution Impacts of Regulatory Alternatives 5-1
5.2.1 Baseline Emissions 5-1
5.2.2 Arsenic Emission Reductions Under the Regulatory
Alternatives 5_2
5.3 Energy Impacts of the Regulatory Alternatives 5-5
5.4 Solid Waste Impacts of the Regulatory Alternatives. ... 5-6
5.5 Water Pollution Impacts of the Regulatory Alternatives. . 5-8
5.6 References 5-10
6.0 COSTS 6_!
-------
TABLE OF CONTENTS
(continued)
Section
Page
6.1 Baseline Controls .... c ,
D-j
6.1.1 Baseline Costs . . c „
b-4
6.2 Process Controls. ... c ,
b-4
6.2.1 Process Control Costs 5,4
6.3 Fugitive Controls 6_1Q
6.3.1 Converter Controls 6_10
6.3.2 Matte and Slag Tapping Controls 6_18
6.4 Costs of Regulatory Alternatives 6-23
6.5 Cost Effectiveness c 00
b-
-------
TABLE OF CONTENTS
(concluded)
Section Page
APPENDIX C - SUMMARY OF TEST DATA C-l
C.I ASARCO-Tacoma C-2
C.2 ASARCO-E1 Paso C-6
C.3 Anaconda C-10
C.4 Phelps Dodge-Ajo C-12
C.5 Phelps Dodge-Hidalgo C-15
C.6 Phelps Dodge-Douglas C-17
C.7 Kennecott-Magna, Utah C-18
C.8 Kennecott-Hayden C-20
C.9 Tamano Smelter (Hibi Kyodo Smelting Co.,) Japan C-25
C.10 Test Data (Tables) C-26
C.ll References C-109
APPENDIX D - TEST METHODS D-l
APPENDIX E - QUANTITATIVE EXPRESSIONS OF PUBLIC CANCER RISKS FROM
EMISSIONS OF INORGANIC ARSENIC FROM LOW-ARSENIC
PRIMARY COPPER SMELTERS E-l
APPENDIX F - ARSENIC DISTRIBUTION AT U.S. COPPER SMELTERS F-l
vm
-------
LIST OF TABLES
Page
Assessment of Environmental And Economic Impacts for Each
Regulatory Alternative Considered 1-4
2-1 Domestic Primary Copper Smelters 2-2
2-2 Major Copper-bearing Minerals 2-3
2-3 Arsenic Input in the Feed to Domestic Copper Smelters . . . 2-4
2-4 Arsenic Elimination in Roasters 2-21
2-5 Elimination of Arsenic in Reverberatory Furnaces 2-24
2-6 Elimination of Arsenic in Electric Furnaces 2-26
2-7 Elimination of Arsenic in Flash Furnaces 2-28
2-8 Elimination of Arsenic During Converting 2-29
2-9 Arsenic Balance Weight Percent as Reporting in Smelter
Products 2-31
2-10 Measured Arsenic Collection Efficiencies
of Control Devices 2-35
2-11 Summary of Process Arsenic Emission Estimates in Absence
of Control for Low-Arsenic Throughput Primary Copper
Smelters 2-37
2-12 Potential Sources of Fugitive Arsenic Emissions 2-38
2-13 Fugitive Arsenic Emissions During Calcine Transfer
From Multihearth Roasters 2-41
2-14 Matte Tapping Emissions from Copper Smelters 2-47
2-15 Slag Tapping Fugitive Arsenic Emissions from
ASARCO-Tacoma 2-50
2-16 Reverberatory Furnace Slag Analysis for Arsenic
Content at ASARCO-Tacoma 2-56
2-17 Summary of Potential Fugitive Arsenic Emission Estimates in
Absence of Control for Low-Arsenic Throughput Primary Copper
Smelters 2-58
3-1 Summary of AS.Og Vapor Pressure Data and Corresponding
Arsenic Concentration at Various Temperatures 3-2
3-2 Arsenic Data for Hot ESP 3-6
IX
-------
LIST OF TABLES
(continued)
Table
3-3 Estimated Approximate Maximum Impurity Limits For
Metallurgical Offgases Used to Manufacture Sulfuric
3-4
3-5
3-6
3-7
3-8
3-9
3-10
3-11
3-12
3-13
3-14
3-15
3-16
3-17
3-18
3-19
Acid
Summary of Design Data for the ASARCO-Tacoma Converter
Secondary Hooding/Air Curtain System
Arsenic Performance Data for the Roaster Baghouse at
ASARCO-Tacoma
Arsenic Performance Data for the Arsenic Plant Baghouse
at ASARCO-Tacoma
Arsenic Performance Data for Spray Chamber/Baghouse at
the Anaconda-Anaconda Smelter
Particulate Performance Data for Spray Chamber/Baghouse at
the Anaconda-Anaconda Smelter
Arsenic Emissions at Outlet of Reverberatory Furnace
Electrostatic Precipitator at ASARCO-Tacoma
Arsenic Performance Data for Spray Chamber/Electrostatic
Precipitator at ASARCO-E1 Paso
Particulate Performance Data for the Spray Chamber/
Electrostatic Precipitator at ASARCO-E1 Paso
Particulate Performance Data for the Spray Chamber/
Electrostatic Precipitator Outlet at ASARCO-E1 Paso ....
Arsenic Performance Data for Venturi Scrubber
At Kennecott-Hayden
Arsenic Performance Data for Double-Contact
Acid Plant at ASARCO-E1 Paso
Arsenic Performance Data for Single-Contact
Acid Plant at Phelps Dodge-Ajo
Summary of Visible Emission Observation Data for Capture
Systems on Fugitive Emission Sources at ASARCO-Tacoma . . .
Air Curtain Capture Efficiencies at ASARCO-Tacoma
Using Gas Tracer Method - January 14, 1983
Air Curtain Capture Efficiencies at ASARCO-Tacoma
Using Gas Tracer Method - January 17-19, 1983
Air Curtain Capture Efficiencies at ASARCO-Tacoma for
Special Gas Tracer Injection Points - January 18-20, 1983 .
3-13
3-42
3-50
3-52
3-53
3-53
3-55
3-57
3-58
3-59
3-61
3-63
3-64
3-66
3-70
3-71
3-72
-------
LIST OF TABLES
(continued)
Table paqe
3-20 Visible Emissions Observation Data for Converter
Secondary Hood System During Matte Charging
At the Tamano Smelter 3.75
3-21 Visible Emissions Observation Data for Blister
Discharge at the Tamano Smelter 3-78
3-22 Arsenic Data for Converter Building Baghouse at
ASARCO-E1 Paso 3_80
3-23 Particulate Data for Converter Building Baghouse
at ASARCO-E1 Paso 3-80
4-1 State Implementation Plans (SIP's) for Sulfur Dioxide
Affecting Copper Smelters and Compliance Status 4-4
4-2 State Implementation Plans (SIP's) for Total Suspended
Particulates Affecting Copper Smelters and
Compliance Status 4_6
4-3 State Implementation Plans (SIP's) for Lead Affecting
Copper Smelters and Compliance Status 4-8
4-4 Summary of Baseline Process Arsenic Emission
Estimates for Low-Arsenic Throughput Primary
Copper Smelters 4_13
4-5 Summary of Baseline Fugitive Arsenic Emission
Estimates for Low-Arsenic Throughput Primary
Copper Smelters 4_14
5-1 Arsenic Emissions from Low-Arsenic Throughput
Copper Smelters by Emission Source and
Regulatory Alternative 5.3
5-2 Nationwide Annual Arsenic Emissions by Regulatory
Alternative and Emission Reductions
From Baseline 5_4
5-3 Nationwide Annual Energy Requirements by
Regulatory Alternative 5_7
5-4 Nationwide Annual Solid Wastes Generated by
Regulatory Alternative 5_g
6-1 Regulatory Alternative Control Requirements 6-2
6-2 Equipment Considered in Baseline Cost Analysis 6-5
-------
LIST OF TABLES
(continued)
Table paqe
-^
6-3 Estimated Capital and Annualized Costs of Baseline
Controls for Primary Copper Smelters 6-6
6-4 Design Parameters for Add-on Process Particulate Matter
Control for Converters at Kennecott-McGill 6-8
6-5 Estimated Annualized Cost of Add-on Process
Particulate Matter Control System for Converters
at Kennecott-McGill 6-11
6-6 Design Parameters for Air Curtain Secondary Hood
Capture System for Primary Copper Smelters 6-13
6-7 Estimated Capital Costs of Air Curtain Secondary Hood
Capture Systems for Primary Copper Smelters 6-15
6-8 Estimated Capital Costs of Air Curtain Secondary Hoods
and Fabric Filters for Primary Copper Smelters 6-17
6-9 Estimated Annualized Costs of Air Curtain Secondary
Hoods and Fabric Filters for Primary Copper Smelters . . . 6-19
6-10 Estimated Capital Costs of Add-on Fugitive Emission
Capture and Collection Systems for Matte and Slag Tapping
Operations 6-21
6-11 Estimated Annualized Costs of Add-on Fugitive Emission
Capture and Collection Systems for Matte and Slag Tapping
Operations 6-22
6-12 Summary of Incremental Costs of Regulatory Alternatives
Over Baseline for Control of Arsenic Emissions for
Low-Arsenic Throughput Primary Copper Smelters 6-24
b-13 Emission Reduction and Cost Effectiveness Impacts
for Low-Arsenic Throughput Primary Copper Smelters .... 6-26
7-1 Smelter Ownership, Production and Source Material
Arrangements 7_3
7-2 United States and World Comparative Trends in Copper
Production: 1963-1981 7-7
7-3 U.S. Copper Consumption 7-9
7-4 U.S. Copper Demand By Market End Uses 7-10
7-5 Average Annual Copper Prices 7-14
7-6 Increase In Cost of Producing Copper Due to Arsenic
Controls for Low-Arsenic Primary Copper Smelters 7-25
xn
-------
LIST OF TABLES
(continued)
Page
Maximum Percent Price Increase for Arsenic Controls
for Low-Arsenic Primary Copper Smetlers 7-26
7-8 Business Segment Return on Sales for Copper Companies . . . 7-30
7-9 Macimum Percent Profit Decrease for Arsenic
Controls for Low-Arsenic Primary Copper Smelters 7-31
7-10 Review of Smelters 7-33
7-11 Capital Costs of Arsenic Controls for Primary
Copper Smelters 7-36
7-12 Number of Employees at Companies that Own Primary
Copper Smelters 7-39
C-l Summary of Emission Tests C-27
C-2 Index to Arsenic and Particulate Test Data Tables by Process
Facility and Sample Type C-29
C-3 Summary of Arsenic Test Data -- Roaster Baghouse Inlet,
ASARCO-Tacoma Smelter C-31
C-4 Summary of Arsenic Test Data — Roaster Baghouse Outlet,
ASARCO-Tacoma Smelter C-32
C-5 Summary of Arsenic Test Data — Arsenic Kitchen Baghouse
Inlet, ASARCO-Tacoma Smelter C-33
C-6 Summary of Arsenic Test Data -- Metallic Arsenic Baghouse
Inlet, ASARCO-Tacoma Smelter C-34
C-7 Summary of Arsenic Test Data -- Arsenic Baghouse Outlet
(Metallic and Kitchen), ASARCO-Tacoma Smelter C-35
C-8 Summary of Arsenic Test Data — Reverb ESP Outlet,
ASARCO-Tacoma Smelter C-36
C-9 Summary of Arsenic Test Data — Calcine Discharge,
ASARCO-Tacoma Smelter C-37
C-10 Summary of Arsenic Test Data — Matte Tapping,
ASARCO-Tacoma Smelter C-38
C-ll Summary of Arsenic Test Data ~ Slag Tapping,
ASARCO-Tacoma Smelter C-39
C-12 Summary of Arsenic Test Data — Converter Slag Return,
ASARCO-Tacoma Smelter C-40
C-13 Summary of Arsenic Test Data — R & R ESP Inlet (Roaster),
ASARCO-E1 Paso Smelter C-41
XI 11
-------
LIST OF TABLES
(continued)
Table
C-14
C-15
C-16
C-17
C-18
C-19
C-20
C-21
C-22
C-23
C-24
C-25
C-26
C-27
C-28
C-29
C-30
C-31
_F
Summary of Arsenic Test Data — R & R ESP Inlet (Reverb-
North), ASARCO-E1 Paso Smelter
Summary of Arsenic Test Data — R & R ESP Inlet (Reverb-
South), ASARCO-E1 Paso Smelter
Summary of Arsenic Test Data — R & R ESP Inlet (Total),
ASARCO-E1 Paso Smelter
Summary of Arsenic Test Data — R & R ESP Outlet,
ASARCO-E1 Paso Smelter
Summary of Particulate Test Data ~ R & R ESP Inlet
(Roaster), ASARCO-E1 Paso Smelter
Summary of Particulate Test Data -- R & R ESP Inlet
(Reverb-North), ASARCO-E1 Paso Smelter
Summary of Particulate Test Data -- R & R ESP Inlet
(Reverb-South), ASARCO-E1 Paso Smelter
Summary of Particulate Test Data — R & R ESP Inlet (Total),
ASARCO-E1 Paso Smelter
Summary of Particulate Test Data — R & R ESP Outlet,
ASARCO-E1 Paso Smelter
Summary of Arsenic Test Data -- DC Acid Plant Inlet,
ASARCO-E1 Paso Smelter
Summary of Arsenic Test Data -- DC Acid Plant Outlet,
ASARCO-E1 Paso Smelter
Summary of Arsenic Test Data — Converter Building Baghouse
Inlet, ASARCO-E1 Paso Smelter
Summary of Arsenic Test Data — Converter Building Baghouse
Outlet, ASARCO-E1 Paso Smelter
Summary of Particulate Test Data — Converter Building
Baghouse Inlet, ASARCO-E1 Paso Smelter
Summary of Particulate Test Data -- Converter Building
Baghouse Outlet, ASARCO-E1 Paso Smelter
Summary of Particulate Test Data ~ Roaster/Reverberatory ESP
Outlet, ASARCO-E1 Paso Smelter
Summary of Particulate Test Data -- Calcine Discharge Duct,
ASARCO-E1 Paso Smelter
Summary of Particulate Test Data ~ Calcine Discharge Duct,
'age
C-42
C-43
C-44
C-45
C-46
C-47
C-48
C-49
C-50
C-51
C-52
C-53
C-54
C-55
C-56
C-57
C-58
ASARCO-E1 Paso Smelter C-59
xiv
-------
LIST OF TABLES
(continued)
Table
C-32
C-33
C-34
C-35
C-36
C-37
C-38
C-39
C-40
C-41
C-42
C-43
C-44
C-45
C-46
C-47
C-48
C-49
Summary of Arsenic Test Data — Matte Tapping Duct,
ASARCO-E1 Paso Smelter
Summary of Parti oil ate Test Data — Matte Tapping Duct,
ASARCO-E1 Paso Smelter
Summary of Arsenic Test Data -- Spray Chamber/Baghouse
Inlet-West, Anaconda-Anaconda Smelter
Summary of Arsenic Test Data — Spray Chamber/Baghouse
Inlet-East, Anaconda-Anaconda Smelter
Summary of Arsenic Test Data — Spray Chamber/Baghouse
Inlet (Total), Anaconda-Anaconda Smelter
Summary of Arsenic Test Data -- Spray Chamber/Baghouse
Outlet, Anaconda-Anaconda Smelter
Summary of Particulate Test Data — Spray Chamber/Baghouse
Inlet-West, Anaconda-Anaconda Smelter
Summary of Particulate Test Data -- Spray Chamber/Baghouse
Inlet-East, Anaconda-Anaconda Smelter
Summary of Particulate Test Data -- Spray Chamber/Baghouse
Inlet (Total), Anaconda-Anaconda Smelter
Summary of Particulate Test Data -- Spray Chamber/Baghouse
Outlet, Anaconda-Anaconda Smelter
Summary of Arsenic Test Data -- Reverberatory ESP Inlet,
Phelps Dodge-Ajo Smelter
Summary of Arsenic Test Data -- Reverberatory ESP Outlet,
Phelps Dodge-Ajo Smelter
Summary of Arsenic Test Data -- Converter ESP Inlet No. 1,
Phelps Dodge-Ajo Smelter
Summary of Arsenic Test Data -- Converter ESP Inlet No. 2,
Phelps Dodge-Ajo Smelter
Summary of Arsenic Test Data — Converter ESP Outlet (Acid
Plant Inlet), Phelps Dodge-Ajo Smelter
Summary of Arsenic Test Data ~ Acid Plant Outlet,
Phelps Dodge-Ajo Smelter
Summary of Arsenic Test Data -- Matte Tapping Hood Outlet,
Phelps Dodge-Ajo Smelter
Summary of Particulate Test Data -- Matte Tapping Outlet,
Phelps Dodge-Ajo Smelter
Page
C-60
C-61
C-62
C-63
C-64
C-65
C-66
C-67
C-68
C-69
C-70
C-71
C-72
C-73
C-74
C-75
C-76
C-77
XV
-------
LIST OF TABLES
(continued)
Page
Summary of Arsenic Test Data — Converter Secondary Hood
Outlet, Phelps Dodge-Ajo Smelter C-78
C-51 Summary of Particulate Test Data -- Converter Secondary Hood
Outlet, Phelps Dodge-Ajo Smelter C-79
C-52 Summary of Arsenic Test Data -- Converter Secondary Hood
Outlet, Phelps Dodge-Hidalgo Smelter c-80
C-53 Summary of Arsenic Test Data — Calcine/Roaster Fugitives
Baghouse Inlet, Phelps Dodge-Douglas Smelter C-81
C-54 Summary of Arsenic Test Data — Calcine/Roaster Fugitives
Baghouse Outlet, Phelps Dodge-Douglas Smelter C-82
C-55 Summary of Particulate Test Data — Calcine/Roaster Fugitives
Baghouse Inlet, Phelps Dodge-Douglas Smelter C-83
C-56 Summary of Particulate Test Data — Calcine/Roaster Fugitives
Baghouse Outlet, Phelps Dodge-Douglas Smelter C-84
C-57 Summary of Arsenic Test Data — Concentrate Dryer Scrubber
Outlet, Kennecott-Magna Smelter c-85
C-58 Summary of Arsenic Test Data -- Acid Plant Inlet
Kennecott-Magna Smelter c_86
C-59 Summary of Arsenic Test Data — Matte Tapping Duct,
Kennecott-Magna Smelter c_87
C-60 Summary of Arsenic Test Data — Slag Tapping Duct,
Kennecott-Magna Smelter Q_88
C-61 Summary of Arsenic Test Data — Converter Fugitives (Full
Cycle), Kennecott-Magna Smelter C-89
C-62 Summary of Arsenic Test Data — Rollout Converter Fugitives
Kennecott-Magna Smelter I C-90
C-63 Summary of Arsenic Test Data — Venturi Scrubber Inlet
Kennecott-Hayden Smelter c_91
C-64 Summary of Arsenic Test Data — Venturi Scrubber Outlet
Kennecott-Hayden Smelter c_92
C-65 Summary of Arsenic Test Data — Acid Plant Outlet,
Kennecott-Hayden Smelter c_93
C-66 Visible Emissions Observation Data, EPA Method 22—Roaster
Calcine Discharge Into Larry Cars, ASARCO-Tacoma C-94
C-67 Visible Emissions Observation Data, EPA Method 22—Matte
Tap Port and Matte Launder, ASARCO-Tacoma c-95
xvn
-------
LIST OF TABLES
(concluded)
Table Paqe
C-68 Visible Emissions Observation Data, EPA Method 22~Matte
Discharge Into Ladle, ASARCO-Tacoma C-96
C-69 Visible Emissions Observation Data, EPA Method 22--Slag
Tap Port and Slag Launder, ASARCO-Tacoma C-97
C-70 Visible Emissions Observation Data, EPA Method 9—Slag
Tap and Slag Launder, ASARCO-Tacoma C-98
C-71 Visible Emissions Observation Data, EPA Method 22--Slag
Tapping at Slag Discharge into Pots, ASARCO-Tacona C-99
C-72 Visible Emissions Observation Data, EPA Method 9~Slag
Tapping at Slag Discharge into Pots, ASARCO-Tacoma C-100
C-73 Visible Emissions Observation Data, EPA Method 22—Converter
Slag Return to Reverberatory Furnace, ASARCO-Tacoma C-101
C-74 Visible Emissions Observation Data, EPA Method 9--Converter
Slag Return to Reverberatory Furnace, ASARCO-Tacoma C-102
C-75 Visible Emissions Observation Data, EPA Method 9—Blister
Discharge From Converter at the Tamano Smelter in Japan. . . C-103
C-76 Summary of Average Observed Opacities for Blister Discharge
At the Tamano Smelter in Japan C-104
C-77 Summary of EPA Method 9 Visible Emissions Data--Individual
and Total Matte Charges to Converter Observed at the Tamano
Smelter in Japan C-105
C-78 Summary of Visible Emissions Observation Data—Copper Blow
At the Tamano Smelter in Japan C-106
C-79 Summary of Visible Emissions Observation Data—Slag Blow
At the Tamano Smelter in Japan C-107
C-80 Summary of Visible Emissions Observation Data—Converter
Slag Discharge At the Tamano Smelter in Japan C-108
E-l Identification of Low-Arsenic Primary Copper Smelters . . . E-15
E-2 Input Data to Exposure Model Low-Arsenic Primary Copper
Smelters E-16
E-3 Total Exposure and Number of People Exposed E-17
E-4 Public Exposure for Low-Arsenic Copper Smelters as
Produced by the Human Exposure Model E-18
E-5 Maximum Lifetime Risk and Cancer Incidence for Low-Arsenic
Primary Copper Smelters E-23
xvn
-------
LIST OF FIGURES
Pacje
2-1 Primary Copper Smelter ................... 2-6
2-2 Primary Copper Smelting Process .............. 2-7
2-3 Calcine Roaster ................... 2-g
2-4 Reverberatory Smelting Furnace ............... 2-12
2-5 Copper Converter ...................... 2-16
2-6 Fugitive Emission Sources at Primary Copper Smelters. . . . 2-39
3-1 Arsenic Trioxide Vapor Pressure and Saturated
Vapor Concentration with Temperature ............ 3.3
3-2 Contact Sulfuric Acid Plant ................ 3_15
3-3 Types of Exhaust Hoods ................... 3_2Q
3-4 Uses of Air Curtains .................... 3_2i
3-5 Spring-Loaded Car Top and Ventilation Hood, ASARCO-
Hayden ........................... 3_24
3-6 Matte Tapping Fugitive Control System (Plan View),
ASARCO-Tacoma ..................... 3_25
3-7 Matte Tapping and Ladle Hoods ............... 3_27
3-8 Launder Cover ..................... 3_2g
3-9 Slag Tapping Fugitive Control System (Plan View),
ASARCO-Tacoma ............ ..... \ ..... 3_30
3-10 Typical Converter Fixed Secondary Hood ........... 3.33
3-11 Conceptual Design for Converter Mechanical Secondary
Hood System ........................ 3_34
3-12 Converter Air Curtain Control System ............ 3.37
3-13 Converter Air Curtain Secondary Hood, Onahama and
Naoshima Smelters ..................... 3_38
3-14 Air Curtain System at the Tamano Smelter .......... 3.40
3-15 Controlled Airflow from a Heated Source .......... 3.44
3-16 Uncontrolled Airflow from a Heated Source ......... 3.44
3-17 Anode Furnace Movable Hood ................. 3.47
3-18 SF6 Tracer Injection Locations ............... 3_68
3-19 Tracer Injection Test Ports ................ 3.59
3-20 Control Device Arsenic Collection Efficiencies ....... 3-82
xvi
-------
LIST OF FIGURES
(continued)
Page
4-1 ASARCO-E1 Paso Smelter Baseline Configuration ....... 4-17
4-2 ASARCO-Hayden Smelter Baseline Configuration ........ 4-19
4-3 Tennessee Chemical Company Smelter Baseline Configuration. . 4-20
4-4 Inspiration-Miami Smelter Baseline Configuration ...... 4-22
4-5 Kennecott-Garfield Smelter Baseline Configuration ...... 4-23
4-6 Kennecott-Hayden Smelter Baseline Configuration ....... 4-25
4-7 Kennecott-Hurley Smelter Baseline Configuration ....... 4-27
4-8 Kennecott-McGill Smelter Baseline Configuration ....... 4-28
4-9 Magma-San Manuel Smelter Baseline Configuration ....... 4-30
4-10 Phelps Dodge-Ajo Smelter Baseline Configuration ....... 4-31
4-11 Phelps Dodge-Douglas Smelter Baseline Configuration ..... 4-33
4-12 Phelps Dodge-Hidalgo Smelter Baseline Configuration ..... 4-34
4-13 Phelps Dodge-Morenci Srnelter Baseline Configuration ..... 4-36
4-14 Copper Range-White Pine Smelter Baseline Configuration . . . 4-38
F-l(a) Arsenic Distribution at ASARCO-E1 Paso Smelter ....... F-3
F-l(b) Overall Arsenic Material Balance At ASARCO-E1 Paso
Smelter ........................... p_4
F-2(a) Arsenic Distribution at ASARCO-Hayden Smelter ........ F-5
F-2(b) Overall Arsenic Material Balance at ASARCO-Hayden Smelter. . F-6
F-3(a) Arsenic Distribution at TN Chemical Co.-Copperhill
Smelter ........................... p_8
F-3(b) Overall Arsenic Material Balance at TN Chemical Co.-
Copperhill Smelter ..................... p_g
F-4(a) Arsenic Distribution at Inspiration-Miami Smelter ...... F-10
F-4(b) Overall Arsenic Material Balance At Inspiration-Miami
Smelter .......................... p_-^
F-5(a) Arsenic Distribution at Kennecott-Garfield Smelter ..... F-12
F-5(b) Overall Arsenic Material Balance at Kennecott-Garfield
Smelter ..................... F-13
F-6(a) Arsenic Distribution at Kennecott-Hayden Smelter ...... F-15
F-6(b) Overall Arsenic Material Balance at Kennecott-Hayden
Smelter ..................... ^ p_16
F-7(a) Arsenic Distribution at Kennecott-Hurley Smelter ...... F-17
xix
-------
LIST OF FIGURES
(concluded)
page
F-7(b) Overall Arsenic Material Balance at Kennecott-Hurley
Smelter p_18
F-8(a) Arsenic Distribution at Kennecott-McGill Smelter F-19
F-8(b) Overall Arsenic Material Balance at Kennecott-McGill
Smelter P_2Q
F-9(a) Arsenic Distribution at Magma Copper Company-San
Manuel Smelter P_22
F-9(b) Overall Arsenic Material Balance at Magma Copper Company-
San Manuel Smelter p_23
F-10(a) Arsenic Distribution at Phelps Dodge-Ajo Smelter F-24
F-10(b) Overall Arsenic Material Balance at Phelps Dodge-Ajo
Smelter P_25
F-ll(a) Arsenic Distribution At Phelps Dodge-Douglas Smelter. . . . F-27
F-ll(b) Overall Arsenic Material Balance at Phelps Dodge-Douglas
Smelter p_28
F-12(a) Arsenic Distribution at Phelps Dodge-Hidalgo Smelter. . . . F-29
F-12(b) Overall Arsenic Material Balance at Phelps
Dodge-Hidalgo Smelter p_30
F-13(a) Arsenic Distribution at Phelps Dodge-Morenci Smelter. . . . F-31
F-13(b) Overall Arsenic Material Balance at Phelps Dodge-
Morenci Smelter . p_32
F-14(a) Arsenic Distribution at Copper Range Company Smelter. . . . F-34
F-14(b) Overall Arsenic Material Balance at Copper Range Company
Smelter P_35
xx
-------
1.0 SUMMARY
1.1 STATUTORY AUTHORITY
National emission standards for hazardous air pollutants are
established in accordance with Section 112(b)(l)(B) of the Clean
Air Act (U.S.C. 7412), as amended. Emission standards under Section
112 apply to new and existing sources of a substance that has been
listed as a hazardous pollutant. This study examines inorganic arsenic
emissions from primary copper smelters which process feed material with
an annual average inorganic arsenic content of less than 0.7 percent by
weight. This category of primary copper smelters is defined as "low-
arsenic throughput smelters." There are currently 14 primary copper
smelters in this category, and none of them is expected to increase
the annual average inorganic arsenic content of its feed materials
to or above 0.7 percent. The single existing high-arsenic throughput
smelter, owned and operated by ASARCO, Incorporated, and located in
Tacoma, Washington, is not expected to decrease the annual average
inorganic arsenic content of its feed materials to below 0.7 percent
and no new smelters are projected to be built during the next 5 years.
For this reason, only the 14 existing low-arsenic throughput primary
copper smelters are analyzed in this document with respect to the
environmental, energy, and economic impacts of regulating the low-arsenic
throughput smelter category. The existing high-arsenic throughput
copper smelter is analyzed in the document, "Inorganic Arsenic Emissions
from High-Arsenic Primary Copper Smelters - Background Information for
Proposed Standards" (EPA-450/3-83-009a).
1.2 REGULATORY ALTERNATIVES
Review of the technical support data led to the development of
five regulatory alternatives. Alternative I would require no additional
regulatory action. This alternative would rely on existing regulations
and existing controls to limit emissions of inorganic arsenic.
Alternative II would require the control of process arsenic emissions,
This alternative is based on the use of flue gas cooling followed by a
particulate control device to collect process arsenic emissions.
1-1
-------
Sources of process arsenic emissions include the roasters, smelting
furnaces, and copper converters.
Alternative III would require the capture of fugitive arsenic
emissions from converter operations and the collection of these fugitive
emissions in a particulate control device. Fugitive emissions are
those that escape capture and control through the primary control
equipment. This alternative is based on the use of an air curtain
secondary hood consisting of a fixed enclosure and an air curtain
system followed by a particulate control device (baghouse or equivalent
technology).
Alternative IV would require the capture of fugitive emissions
from furnace slag tapping operations and the collection of inorganic
arsenic emissions from furnace slag tapping and matte tapping operations.
This alternative is based on the use of localized hoods to capture the
slag tapping fugitive emissions and a particulate control device (bag-
house or equivalent technology) to collect the inorganic arsenic emissions
from both furnace slag tapping and matte tapping.
Alternative V would require the elimination of all arsenic emissions
at copper smelters. To accomplish this alternative the smelters would
be forced to process ores which were virtually free of arsenic content.
1.3 ENVIRONMENTAL IMPACTS
Under Alternative I there would be no additional inorganic arsenic
reduction from the baseline because there is no additional regulatory
action associated with Alternative I. Under Alternative II, inorganic
arsenic emissions from process sources would be reduced by 163 megagrams
per year (Mg/yr). This represents a 30 percent reduction in process
inorganic arsenic emissions and a 22 percent reduction in overall
inorganic arsenic emissions from the low-arsenic copper smelter category.
Energy requirements at the one smelter affected by Alternative II would
increase by 4.9 x 10® kWh. This represents a 33 percent increase in
the energy requirements at the one affected smelter. Solid waste increases
at the affected smelter would amount to 16,300 Mg/yr. This represents
a 0.5 percent increase in solid waste which must be handled by the smelter.
Under Alternative III, fugitive arsenic emissions from converter
operations would be reduced by 118 Mg/yr. This represents a reduction
of 60 percent in fugitive inorganic arsenic emissions and a reduction
1-2
-------
of 16 percent in overall inorganic arsenic from the source category.
Energy requirements for this alternative would increase by 1.8 x 10^ kWh.
This represents a 0.35 percent increase in energy requirements for the
source category. Solid wastes generated would increase by 11,800 Mg/yr,
which represents a 0.4 percent increase for the source category.
Under Alternative IV, fugitive inorganic arsenic emissions from
matte and slag tapping operations would be reduced by 11 Mg/yr. This
represents a reduction of 5 percent in fugitive arsenic emissions and
a reduction of 1.5 percent in overall inorganic arsenic emissions from
the source category. Energy requirements for the source category would
increase by 9 x 10^ kWh, which represents a 0.02 percent increase in
energy requirements. Increases in solid wastes generated would amount to
1,100 Mg/yr, or an increase of 0.03 percent for the source category.
The air quality, energy, and solid waste impacts, as well as all
environmental and economic impacts, are summarized in Table 1-1.
The control systems for the regulatory alternatives are dry
systems; consequently, no incremental increase in water discharges
is anticipated. If scrubbers are used, increases in wastewater
discharges would result and the scrubber discharge would be treated
within existing water pollution control systems. Therefore, even if
scrubbers are used, no adverse water pollution impact is anticipated.
The regulatory alternatives would result in negligible impacts
on noise, space, and availability of resources.
1.4 ECONOMIC IMPACTS
The total capital cost for the process controls specified under
Alternative II is $9.8 million. The annualized cost for the controls
at the Kennecott-McGill smelter would be $4.1 million. The total
annualized cost for this alternative, coupled with the annual emission
reduction expected under Alternative II, would yield an annualized
cost-effectiveness of $25,200/Mg of inorganic arsenic reduced. The
total capital cost for the fugitive emission controls for the converter
operations specified under Alternative III is $109.9 million. The
annualized cost of these converter controls would be $29.2 million.
The total annualized costs associated with the converter controls,
1-3
-------
Table 1-1. ASSESSMENT OF ENVIRONMENTAL AND ECONOMIC IMPACTS FOR EACH
REGULATORY ALTERNATIVE CONSIDERED
Regulatory
Alternative
I
II
III
IV
V
Air
Impact
0
+2**
+2**
+2**
+4**
Water
Impact
0
-1*
-1*
-1*
+4**
Solid Waste
Impact
-1*
-1*
-1*
-3**
Energy
Impact
0
_!**
_!**
_!**
+4**
Noise
Impact
0
-1*
-1*
-1*
+1**
Economic
Impact
0
.4**
-3**
-3**
_4***
Key: + Beneficial Impact
- Adverse Impact
0 No impact
1 Negligible Impact
2 Smal1 Impact
3 Moderate Impact
4 Large Impact
* Short-Term Impact
** Long-Term Impact
*** Irreversible Impact
-------
coupled with the emission reduction expected under Alternative III,
would yield an annualized cost-effectiveness of $247,000/Mg of inorganic
arsenic reduced.
The total capital cost for the fugitive emission controls
for the matte and slag tapping operations specified under Alternative IV
is $14.8 million. The annualized cost associated with these fugitive
emission controls would be $4.3 million. The total annualized cost of
these matte and slag tapping fugitive emission controls, coupled with
the emission reduction expected under Alternative IV, would yield an
annualized cost-effectiveness of $389,000/Mg of inorganic arsenic reduced.
In 1982, copper producers experienced one of the worst years in
recent history. Such a situation cannot be used as the foundation to
examine the long-term economic impact of the potential arsenic NESHAP.
Therefore, the economic analysis is based on a more normal condition for
the industry. The principal economic impacts analyzed are: the ability
of the smelters to increase copper prices in response to an increase in
costs due to the arsenic standard; and, the impact on profits if part
or all of the costs cannot be passed on in the form of price increases.
If each smelter attempts to maintain its normal profit margin and
pass control costs forward in the form of a price increase, the price
increases would range from 0.1 percent to 15.2 percent. However, competi-
tion will prevent the existence of such a broad variation. If control
costs are absorbed and profit margins reduced, again a broad range
exists. The profit reductions would range from 0.4 percent to 151.9
percent. At a 100 percent capacity utilization rate and a price of
187
-------
Alternative V, which would eliminate arsenic emissions from the
low-arsenic throughput smelter category would result in closing of all
the smelters in the source category. Although this results in the
greatest emission reduction, the economic impacts and hardships associated
with this alternative are severe.
1-6
-------
2.0 THE PRIMARY COPPER INDUSTRY
2.1 GENERAL
Currently, there are 15 primary copper smelters operating or
temporarily closed in the United States. Of these, seven are located
in Arizona, two in New Mexico, and one each in Nevada, Texas, Utah,
Tennessee, Michigan, and Washington. The concentration of copper
smelters in the Southwest is due mainly to the local availability of
copper-bearing ores. For the 15 copper smelters, smelting capacity
totals approximately 1.72 million Mg (1.9 million tons) of smelter
product (99 percent "blister" copper) per year. Table 2-1 lists
these 15 smelters, their locations, and estimated capacities.
2
Primary copper production in 1982 was 975,437 Mg (1,075,400 tons).
This represented approximately a 60 percent utilization of domestic
primary copper smelting capacity. Smelter capacity in the United
States appears to be relatively steady, with minimal prospects for
substantial additions before 1985. The Bureau of Mines forecasts the
total demand for copper in the year 2000 to be between 3.5 and 6.0 teragrams
(Tg) (3.9 x 10 and 6.6 x 10 tons). This forecast represents an
annual growth of 3.6 percent. However, recycling of scrap and the
expansion of hydrometallurgical facilities are expected to accommodate
3
a substantial portion of this increased demand.
2.1.1 Raw Materials
Copper ores are generally classified as sulfide, oxide, or native
depending on the predominant copper-bearing minerals they contain.
Although copper occurs in at least 160 minerals, only a few of these
have any commercial importance. Table 2-2 lists the composition of
the more important sulfide and oxide minerals from which copper is
extracted. Of the primary sulfide minerals, chalcopyrite is the most
abundant, followed by bornite and chalcocite. Oxide minerals are
produced by the oxidation of primary sulfide minerals under certain
2-1
-------
Table 2-1. DOMESTIC PRIMARY COPPER SMELTERS
Company
Location
Annual capacity9
Megagrams(Tons;
ASARCO, Incorporated
Tennessee Chemical Company
Inspiration Consolidated
Copper Company
Kennecott Copper Corporation
Magma Copper Company
Phelps Dodge Corporation
Copper Range Company
TOTAL
El Paso, Texas
Hayden, Arizona
Tacoma, Washington
Copperhill, Tennessee
Miami, Arizona
Garfield, Utah
Hayden, Arizona
Hurley, New Mexico
McGill, Nevada
San Manuel, Arizona
Ajo, Arizona
Douglas, Arizona
Hidalgo, New Mexico
Morenci, Arizona
White Pine, Michigan
91,000
182,000
91,000
13,600
136,000
254,000
71,000
73,000
45,000
(100,000)
(200,000)
(100,000)
(15,000)
(150,000)
(280,000)
(78,000)
(80,000)
(50,000)
181,000 (200,000)
64,000
115,000
163,000
191,000
(70,000)
(127,000
(179,000)
(210,000)
52,000 (57,000)-
1,722,600 (1,896,000)
Production of "blister" copper (99 percent Cu)
2-2
-------
Table 2-2. MAJOR COPPER-BEARING MINERALS
Type
Sulfide
Oxide
Mineral
Chalcopyrite
Bornite
Chalcocite
Covellite
Malachite
Azurite
Chrysocolla
Cuprite
Formula
CuFeS2
Cu^FeS-
Cu2S
CuS
CuC03'Cu(OH)2
2CuCO,'Cu(OH)9
O C-
CuSi03'2H20
Cu20
climatic conditions. When present, these are usually found in the
upper portions of copper ore deposits. Native copper consists of
almost pure metallic copper. Although found in small amounts in many
copper ore deposits, it is essentially unique to the upper peninsula
of Michigan.
Copper ores consist of one or more of these copper-bearing minerals
disseminated within relatively large quantities of siliceous and other
earthy matter. In addition, they also contain varying amounts of
other metals including sulfides and oxides of iron, arsenic, antimony,
lead, zinc, etc. In the United States, low grade sulfide ores account
4
for 85 to 95 percent of the total primary production. The average
tenor of these ores (copper content) is less than 1 percent. Virtually
all copper ores processed are beneficiated at the mine. Sulfide ores
are crushed, finely ground, and concentrated by froth flotation.
Oxide ores are leached with acid, and the dissolved copper is re-
covered by chemical precipitation on scrap iron. Typically, copper
ore concentrates contain about 15 to 30 percent copper.
In addition to copper, metals such as arsenic are also concentrated
in the concentration process. The arsenic content of copper concentrates
processed at U.S. copper smelters is highly variable, ranging from a
few parts per million to several percent. The impurities, mainly
arsenic and others such as antimony, bismuth, lead, and zinc exert a
strong influence on the selection of smelting technology. Table 2-3
2-3
-------
Table 2-3. ARSENIC INPUT IN THE FEED TO
DOMESTIC COPPER SMELTERS
Plant
ASARCO-Tacoma
ASARCO-Hayden
ASARCO-E1 Paso
Kennecott-Garfield
Kennecott-McGill
Phelps Dodge-Ajo
Inspiration-Miami
Phelps Dodge-Hidalgo
Phelps Dodge-Douglas
Kennecott-Hayden
Phelps Dodge-Morenci
Magma-San Manuel
Tennessee Chemical Company -
Copperhill
Kennecott-Hurley
Copper Range - White Pine
Arsenic
Percent
4.0
0.6
0.5
0.14
0.4
0.3
0.033
0.018
0.03
0.015
0.006
0.006
0.0004
0.0005
0.008
content of
kg/hr
991
170
142d
118e
81f
47
189
14
11
8.0
4.5h
2.0
1.3
1.0
0.71"
feeda'b
(Ib/hr)
(2,185)
(375)
(314)
(261)
(179.3)
(103)
(41.1)
(30.6)
(24)
(17.7)
(9.99)
(4.39)
(2.9)
(2.14)
(1.53)
The feed is a mixture of concentrates, precipitates, lead smelter
by-products, and smelter reverts.
Does not include recycled flue dusts and other intermediates.
50 kg/hr (111 Ib/hr) of this amount is fed directly to the arsenic plant,
51 kg/hr (112 Ib/hr) of this amount is fed directly to the converters.
/>
3.5 kg/hr (7.8 Ib/hr) of this amount is fed directly to the converters.
6.5 kg/hr (14.2 Ib/hr) of this amount is fed directly to the converters.
90.3 kg/hr (0.6 Ib/hr) of this amount is fed directly to the converters.
0.2 kg/hr (0.35 Ib/hr) of this amount is fed directly to the converters.
0.1 kg/hr (0.22 Ib/hr) of this amount is fed directly to the converters.
2-4
-------
presents the amount of arsenic input to the domestic primary copper
smelters based on information received from the smelters in early
iqoo °»'>o,y,1U,11,1^ nfitn^n
' Tne ASARCO-Tacoma smelter, which introduces
more arsenic in the feed material to the smelter than the combined
total from all of the low-arsenic throughput copper smelters, is
analyzed in a separate background information document (EPA-450/3-83-009a).
2.1.2 Process Description
The pyrometallurgical process used for the extraction of copper
from sulfide ore concentrates is based on iron's strong affinity for
oxygen as compared to copper's weak affinity for oxygen. The purpose
of smelting is to separate the copper from the iron, sulfur, and
gangue materials. Conventional practice includes three operations:
1. Roasting (optional) to remove a portion of the concentrate
sulfur content.
2. Smelting of roasted calcines or unroasted ore concentrates
and fluxes in a furnace to form slag and copper-bearing matte.
3. Converting (oxidizing) of the matte in a converter to form
blister copper (about 99 percent pure copper).
Figure 2-1 presents a pictorial representation of the copper
smelting process. Figure 2-2 illustrates the three basic operations
employed as well as materials entering or leaving each operation.
Briefly, the smelting of copper concentrates and precipitates is
accomplished by melting the charge and suitable fluxes in a smelting
furnace. Part or all of the concentrates may receive a partial roast
to eliminate some of the sulfur and impurities such as arsenic. In
the smelting furnace, the lighter impurities combine and float to the
top as slag to be skimmed off and discarded, while the copper, iron,
most of the sulfur, and any contained precious metals form a product
known as matte which collects in and is drawn off from the lower part
of the furnace. The molten matte is transferred to a converter where
air blown through the matte burns off the sulfur, oxidizes the iron
for removal in a slag, and yields a 99 percent blister copper product.
Typically, the blister copper is further refined in an anode furnace
prior to the casting of copper anodes for electrolytic refining. A
more detailed discussion of roasting, smelting, and converting operations
is presented in succeeding paragraphs.
2-5
-------
EMISSION CONTROL EQUIPMENT
CONVERTER
WASTE HEAT
BOILERS
REVERBERATORY
FURNACE
Figure 2-1. Primary Copper Smelter
-------
HTEHDIS THE SYSTEM
LEIYING THE SYSTEM
Raw canontratts
Fuel
lir
ROASTER
Flui and
fettl ing material
Fuel
Air
SMELTING FURNACE
Siliceous f!ux
Mi see Ilaneous
material high in copper
Air
CONVERTER
Gases, roiatiIt oxldei.
and dust to dust recovery
and stack
Gases and dust
to Haste heat boilers,
dust recovery, and stack
Slag to dump
Gases to stack
Jlisler copper
to refinery
Figure 2-2. Primary Copper Smelting Process
2-7
-------
2.1.2.1 Roasting. In roasting of copper sulfide ore concentrates,
concentrates are heated to a high temperature (but below the melting
point of the constituents) in an oxidizing atmosphere to eliminate a
portion of the sulfur contained as sulfur dioxide (S02); to remove
volatile impurities such as arsenic, antimony, and bismuth; and to
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. Representative reactions include the following:
COUI COp
FeS2 —
S
4FeS H
+ 02-
^ 702 -
^ VyUo
— ^ FeS
— *• so2
— »- 2Fe
O 1 £-1 CO
+ S
2°3 + 4SO;
Currently, 7 of the 15 existing primary copper smelters roast
concentrates prior to smelting. Two types of roasters are used:
multiple-hearth roasters and fluid-bed roasters (refer to Figure 2-3).
With both types, the roasting process is generally autogenous. The
roaster operating temperature is typically about 650°C (1,200°F).
In multi-hearth type roasters, the hearths are constructed of
refractory brick with a slight arch. The external portion of the
furnace is a brick-lined, steel shell with hinged doors and inspection
plates at each level. The moist concentrate enters the roaster through
an annular opening to the top-most or dryer hearth. Rabble arms,
attached to the hollow central shaft, rotate as the shaft turns and
plow through the charge to continuously expose fresh surfaces to the
oxidizing air. The rabble blades are set at an angle and, in addition
to stirring the material, move it alternately from the center of the
hearth to the periphery where it falls to the next lower hearth.
Finished calcine is discharged through holes on the circumference of
the bottom hearth.
The air required for roasting is admitted through the central
shaft and, by means of valves, the air supply to each hearth may be
2-8
-------
OFF
GAS
FEED
4
DRYING
HOT AIR
TO EXHAUST
RABBLE
ARM
RABBLE
BLADE
CALCINE
SLURRY
FEED
TUYERE
HEADS
AIR
OFF-GAS
fl
MULTI-HEARTH ROASTER
FLUID-BED ROASTER
Figure 2-3. Calcine Roaster
-------
regulated. Roaster gases are drawn off through gas outlets located
just below the dryer hearth. The discharge from the dryer hearth to
the top roasting hearth prevents the escape of gas from the interior
of the roaster.
Fluid-bed roasters are cylindrical, refractory-lined vessels
equipped with diffusion plates in the bottom containing tuyeres or
bubble caps through which air is blown from the bottom. Finely ground
material (60 percent minus 200 mesh) is introduced either as a slurry
through a feed pipe or relatively dry (6 to 12 percent moisture)
through a screw conveyor. The feed is continuously delivered into the
combustion chamber. Roasting occurs as the sulfide particles fall
through the oxidizing air.
Combustion air from the windbox passes through distribution
plates at the bottom of the chamber into the combustion zone. Because
of the large surface area of the finely ground material exposed to the
air stream, the residence time in the oxidizing atmosphere is short.
The reaction is self-sustaining. Oil, gas, or pulverized coal burners
are required only to preheat the roaster to combustion temperature.
Roaster gases are drawn off through flues at the top of the
chamber and immediately pass to cyclone collectors, followed by cooling
and final dust collection. As much as 85 percent of the feed is
carried with the gas stream and, hence, cyclones are an integral part
of the roasting operation. Screw conveyors collect this finished
calcine as well as that from the bottom of the combustion chamber for
discharge into hoppers.
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
offgases. Sulfur dioxide concentrations in the roaster offgases are
considerably higher for fluid-bed roasters than for multi-hearth
roasters due to the lower total air volume. Average stack gas S0?
concentration is about 12 percent with a maximum of 18 percent for
fluid-bed roasters and only 3 to 6 percent for multi-hearth roasters.14
2.1.2.2 Smelting. Smelting is the pyronetallurgical process in
which solid material is melted and subjected to certain chemical
2-10
-------
changes. During copper smelting, hot calcines from the roaster or
raw, unroasted concentrates are melted in a smelting furnace with
siliceous or limestone flux. Converter slag, collected dust, oxide
ores, and any other material rich in copper may be added to the furnace
charge. Copper and iron which are present in the charge combine with
sulfur to form a stable cuprous sulfide. Excess sulfur unites with
iron to form a stable ferrous sulfide, FeS. The combination of the
two sulfides, known as matte, collects in the lower area of the furnace
and is removed. Such mattes may contain from 15 to 50 percent copper,
with a 40 to 45 percent copper content being most common. Mattes may
also contain impurities such as sulfur, antimony, arsenic, iron, and
precious metals.
The remainder of the molten mass containing most of the other
impurities is known as slag. Slag is of lower specific gravity,
floats on top of the matte, and is drawn off and discarded. Slags in
copper smelting are ideally represented by the composition 2FeO'SiO?,
but contain alumina from the various charge materials and calcium
oxide which is added for fluidity. Since slags are discarded, the
copper contained in the slag is a major source of copper loss in pyro-
metallurgical practice. Copper concentration in the slag increases
with increasing matte grade. This behavior limits the matte grades
normally obtained in conventional practice to below 50 percent copper.
Currently, conventional reverberatory furnaces are used at 11 of
the 15 existing primary copper smelters. Two smelters employ electric
furnaces, while one smelter employs the Outokumpu flash furnace and
another a Noranda continuous smelter. Two smelters will be modifying
one or more of their reverberatory furnaces to convert them to oxygen-
sprinkle smelting, while two more are planning to retire their reverberatory
furnaces within the next few years and install Inco flash smelting
furnaces. In a reverberatory furnace (Figure 2-4), fossil fuels such
as oil or natural gas are burned above the copper concentrates being
smelted. The furnace is a long, rectangular structure, generally
about 11 m (36 ft) in width and 40 m (131 ft) in length, with an
arched roof and burners at one end. Flames from the burners may
extend half the length of the furnace. Temperatures at the firing end
of the furnace exceed 1,500°C (2,730°F). Part of the heat in the
2-11
-------
I
ro
CALCINE
V
FUEL
FETTUNG DRAG
CONVEYOR
OFPGAS
AIR AND
OXYGEN
CONVERTER
SLAG
BURNERS
MATTE
SLAG
SLAG
FETTLING PIPES
MATTE
Figure 2-4. Reverberatory Smelting Furnace
-------
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. The roofs of the older
reverberatory furnaces are sprung-arch silica roofs, while almost all
newer furnaces have suspended roofs of basic refractory.
Over the years, two types of reverberatory furnaces have evolved,
each with its own specific charging methods. The older type is the
deep bath reverberatory furnace which contains a large quantity of
molten slag and matte at all times. In modern, deep bath reverberatory
furnaces, the molten material is held in a refractory crucible with
cooling water jackets along the sides, which greatly diminishes the
danger of a breakout of the liquid material. In deep bath smelting,
several methods exist for charging. Wet concentrates can be charged
using slinger belts (high-speed conveyors) that spread the concentrates
on the surface of the molten bath. Dry concentrates or calcines from
the roaster can be charged through the roof or via a Wagstaff gun (an
inclined tube). Roof charging (side charging) is rarely practiced in
conjunction with deep bath smelting because of dusting problems with
fine dry calcine and explosion potential with green charge. Wagstaff
guns minimize these problems and are commonly used.
Side or roof charging is usually used with green charge. With
the charge dropped through a series of feed holes in the arch near the
walls, a buildup of material forming banks results. The banks slowly
melt and serve as protection to the side walls, eliminating the requirement
for special cooling.
Combustion gases 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, with averages varying from 0.5 to
2.0 percent.15 These lean S02 mixtures, unlike offgases from fluid-bed
roasters, converters, and other types of smelting furnaces, cannot be
economically utilized as feed for sulfuric acid plants.
Electric smelting furnaces provide the heat necessary for smelting
copper ore concentrates by allowing carbon electrodes to come into
contact with the molten bath within the furnace. The electrodes dip
into the slag layer of the bath, forming an electrical circuit. When
2-13
-------
an electric current is passed through this circuit, the slag resists
its passage, generating heat and producing smelting temperatures.
Charge concentrates and fluxing materials are fed through the roof,
and a layer of unsmelted charge covers the molten bath. Heat is
transferred from the hot slag to the charge floating on its surface,
and as the copper concentrates and fluxes are smelted, they settle
into the bath forming slag and matte. The chemical and physical
changes occurring in the molten bath are similar to those occurring in
the molten bath of a reverberatory furnace.
In flash smelting, copper sulfide ore concentrates are smelted by
burning a portion of the iron and sulfur contained in the concentrates
while they are suspended in an oxidizing environment. As such, the
process is quite similar to the combustion of pulverized coal. The
concentrates and fluxes are injected with preheated air, oxygen-enriched
air, or even pure oxygen, into a furnace of special design. Then,
smelting temperatures are attained as a result of the heat released by
the rapid, flash combustion of iron and sulfur. Flash smelting technology
has been developed by two companies: International Nickel Company
(INCO) in Canada and Outokumpu Oy in Finland. The major difference
between the two technologies is in the design of the smelting furnace
and the oxidizing environment within the furnace. The INCO furnace
uses pure oxygen, while the Outokumpu furnace employs preheated air or
oxygen-enriched air as the oxidizing medium.
Two smelters are currently planning to convert one or more of
their reverberatory furnaces to oxygen-sprinkle smelting. In essence,
this conversion will allow the existing reverberatory furnaces to
behave like flash furnaces. Specially designed burners positioned on
the furnace roof are used to introduce and disperse a mixture of
primarily dried concentrates and oxygen. The heat required for smelting
is generated from the flash combustion of the sulfur in the mixture.
The result is an increase in furnace efficiency and the production of
a strong S02 gas stream capable of being treated in a sulfuric acid
plant.
2.1.2.3 Converting. Matte produced in the reverberatory furnace
is transferred in ladles to the converters using overhead cranes.
Fourteen of the 15 smelters use converters of the cylindrical Pierce-Smith
2-14
-------
type, the most common size being 4 by 9 m (13 by 30 ft). Figure 2-5
is a sketch of a Fierce-Smith copper converter. An alternative to the
Pierce-Smith converter is the newer Hoboken or "siphon" converter.
The Hoboken converter, currently used by one of the domestic smelters,
is essentially the same as the Pierce-Smith converter except that it
is fitted with a side flue located at one end of the converter and
shaped as an inverted U. This flue arrangement permits siphoning of
the converter gases from the interior of the converter, using variable-speed
fans and dampers, directly to the offgas collection system. By maintaining
a slightly negative pressure at the converter mouth, it is possible to
minimize or eliminate emissions. Problems in improper draft at the
converter mouth have been reported by Inspiration Consolidated Copper
Company, the only domestic user of Hoboken converters. One of the
five Hoboken converters operated by Inspiration has been modified to
eliminate the siphon area. The modification has resulted in improved
performance, and Inspiration intends to modify its four remaining
converters in similar fashion.16
In both Pierce-Smith and Hoboken converters, air is blown from
the side through a series of openings called tuyeres. During the
initial blowing period (the slag blow), FeS in the matte is preferentially
oxidized to FeO and Fe304> and sulfur is removed with the offgases as
S02. Flux is added to the converter to combine with iron oxide and
forms a fluid iron silicate slag. When all the iron is oxidized, the
slag is skimmed and poured off from the furnace at various times
during slag formation, leaving behind "white metal" or molten Cu S.
During this stage, fresh matte is charged into the converter, and
the slag blowing continues 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 cast or subjected to additional fire
refining prior to casting. Converter blowing rates can vary between
340 and 855 normal m3/min (12,000 to 30,000 scfm) of air.
In general practice, the matte is added to the converter in two
to six steps, each step followed by oxidation of much of the FeS from
the charge. The resulting slag is poured from the converter after
2-15
-------
OFF-GAS
ro
i
TUYERE
PIPES
SILICEOUS
FLUX
PNEUMATIC
PUNCHERS
Figure 2-5. Copper Converter
-------
each oxidation step, and a new matte addition is made. In this way,
the amount of copper (as matte) in the converter gradually increases
until there is a sufficient amount for a final copper-making "blow."
At this point, the FeS in the matte is blown down to about 1 percent,
a final slag is removed, and the resulting white metal (impure Cu?S)
is oxidized to blister copper. The converting process is terminated
when copper oxide begins to appear with the liquid copper. The offgas
flow rate leaving the primary head of converters typically ranges from
850 to 1,260 Nm3/min (30,000 to 45,000 scfm). The average S02 concentration
in these gases is normally in the range of 4 to 5 percent during the
slag blow and 7 to 8 percent during the copper blow. Values of the
overall average S02 concentration in the offgases (after gas cleaning)
from existing domestic converting operations fall in the range of
1.6 to 6.5 percent on a dry basis. The industry-wide average value,
as weighted by the respective plant flow rates, is 4.6 percent.17
2.1.2.4 Refining. Virtually all copper produced by matte smelting
is subsequently electrorefined. For this reason, the final liquid
copper product of the smelter must be suitable for the casting of
strong, thin anodes. In a majority of cases, anode copper is fire
refined directly from molten blister copper.
Fire refining is performed in rotary-type refining furnaces
resembling Pierce-Smith converters or in small hearth furnaces. The
rotary-type predominates when molten blister copper is treated directly,
while hearth furnaces are used when melting of solid charges is practiced.
The temperature of operation is about 1,130 to 1,150°C, which provides
sufficient superheat for the subsequent casting of anodes. There is
very little heat produced by refining reactions, and some combustion
of fuel is necessary to maintain the temperature in the furnace.
Dimensions of the rotary-type refining furnaces vary; however, a
4 by 9 m (13 by 30 ft) furnace may be regarded as typical. Gas flow
rates are generally low so as to accurately control the metal composition.
Gas pressures at the tuyeres are near 250 to 600 kPa (2.5 to 6 atm).
Refining a 250-ton charge of copper requires 3 to 5 hours: 1/2 to
1 hour for the oxidation step, and the remainder for the deoxidation
("poling") step.
2-17
-------
A typical sequence of events in rotary anode furnace refining is:
1. Molten blister copper is added to the anode furnace as it
becomes available from the converters until about 150 to 300 Mg (165 to
331 tons) have been accumulated;
2. The accumulated charge is then oxidized by blowing air through
the tuyeres until the sulfur content is lowered to 0.001 to 0.003 percent S,
at which time a small ingot sample of copper shows a slight contraction
or small hole; and at which time the oxygen level in the copper is
about 0.6 percent;
3. The oxygen is then removed from the copper by blowing natural
gas, reformed natural gas, or propane through the tuyeres. The oxygen
level in the anode copper is 0.005 to 0.2 percent after this operation,
which gives a "flat set" to the anodes when they are cast. The correct
"end point" is determined by casting a sample of the copper (which
should set to a flat surface) or by continuously analyzing for oxygen
with an oxygen probe;
4. Finally, the liquid metal may be covered with low sulfur coke
to prevent reoxidation of the copper.
Several older modifications of this process are still being used
in which the air is introduced via steel lances, and in which the
oxygen-removal step is performed by lowering large logs or poles of
green wood into the copper (thereby providing the necessary hydrocarbons).
Both of these steps are clumsy, and are being discontinued.
Similarly, the use of the hearth type of anode furnace has by and
large been discontinued except where scrap (including anode scrap)
and/or blister copper are melted. A typical anode hearth furnace
resembles a small reverberatory furnace [width 5 m (16 ft), length
15 n (49 ft), height 3 m (10 ft), inside dimensions] capable of holding
3UO Mg (331 tons) of copper. In hearth furnaces, the air and hydrocarbons
are introduced into the copper by submerging steel lances into the
molten bath. Wood poles are often used for the oxygen removal step.
Since this study analyzes 14 model plants representing the 14 existing
low-arsenic throughput copper smelters, plant-by-plant descriptions of
operating and emission control parameters are presented in Section 4.3.2
of this document.
2-18
-------
2.2 ARSENIC BEHAVIOR AND DISTRIBUTION IN COPPER SMELTERS
Arsenic elimination during copper smelting occurs primarily by
volatilization and slagging. The bulk of the arsenic is volatilized
in most pyrometallurgical processes and removed with the offgases. In
the presence of oxygen, elemental arsenic or arsenic sulfides oxidize
to arsenic trioxide which is extremely volatile. However, in an
oxidizing atmosphere, the arsenic trioxide (As,,03) may oxidize to the
higher oxide (As^OJ which is less volatile and forms stable nonvolatile
(— J
arsenates with other metallic oxides.
The elimination of arsenic by slagging is dependent upon the
1 O
concentration or partial pressure of the arsenic trioxide. If the
partial pressure of the oxide is sufficient to allow it to penetrate
the slag layer and thus escape, volatilization will occur. However,
if the partial pressure is not large enough to allow passage through
the slag layer, the arsenic will be mechanically bound in the slag and
thus be eliminated on slag disposal. The magnitude of the partial
pressure is dependent upon the amount of arsenic present in the process
streams.
2.2.1 Arsenic Behavior During Roasting. Arsenic elimination during
roasting is indirectly dependent upon the grade of matte required
during smelting in the furnace. This requirement limits sulfur elimination
during the roasting operation to establish the matte controlling
sulfur-to-copper ratio for the calcine feed to the smelting furnace.
Arsenic elimination during the roasting process is by volatilization
and, therefore, is dependent upon the amount or degree of roasting.
As previously described, the roasting operation is primarily one
of oxidation of solid material by means of oxidizing gases. The
sulfides present in the copper ore that are oxidized include arsenic
sulfides such as FeSAs2S3, As2$3, and As2$5. The As2$3 boils at 707°C
(1,305°F) and As2S5 sublimes at 500°C (934°F) with decomposition.
Arsenic trioxide sublimes at 457°C (855°F) and melts at 310°C (590°F).
Thus, if the solids are heated to above 700°C (1,290°F), the volati-
lization of arsenic would seem to be assured, but such a high temperature
is not practical. At the beginning of roasting, when there is a large
portion of easily fusible sulfides present, high temperature melts the
particles which effectively reduces the surface-to-volume ratio available
2-19
-------
for sulfur elimination. However, if a low temperature is maintained
throughout the entire roasting period, sulfates of metal would form,
since the heat of formation of sulfates is higher than that of the
corresponding oxides.19 This would decrease the elimination of sulfur.
To minimize the formation of sulfates, it is desirable to maintain a
higher temperature (up to 554°C or 1,000°F) at the end of the roast
than at the beginning. The ideal temperature condition would therefore
be low initially and increase gradually during roasting.
Conditions for the removal of arsenic differ from those of sulfur
elimination. Maximum removal of arsenic occurs at a low temperature
(500° to 600°C or 930 to 1,110°F) under reducing conditions (when
sulfide concentrates are roasted). A higher temperature oxidizing
roast causes the formation of arsenic pentoxide, which can react with
calcium or iron to form stable and nonvolatile arsenates which remain
in the calcine. However, an oxidizing roast is required for the
sulfur removal. It is therefore desirable to alternate oxidation and
reduction several times to maximize the volatilization of arsenic.
Arsenic elimination in the multi-hearth roaster tends to be
greater than in a fluidized-bed unit for any given degree of sulfur
elimination. In a multi-hearth roaster, it is possible to vary oxidizing
and reducing conditions, temperature, and gas composition on each
hearth. The residence times of concentrate particles in a multi-hearth
roaster generally range from 1 to 2 hours. The fluidized-bed roaster
can provide only one set of conditions at a time, either oxidizing or
reducing. The temperature remains constant throughout the bed, and
there are a minimum of hot spots or uncontrolled variations in condi-
tions. The residence time of concentrate particles in the fluidized-bed
roaster is much smaller compared to a multi-hearth roaster.
Table 2-4 lists data from various sources on the amount of arsenic
volatilized on a 100 percent input basis in multi-hearth and fluidized-bed
roasters. The elimination of arsenic in the roasting of copper ores
and concentrates depends to a considerable extent upon the type of
roasting unit used. The third data point indicates that in multi-hearth
roasters, as much as 70 to 90 percent arsenic can be driven off when a
moderate air consumption rate is used. When concentrates are roasted
2-20
-------
Table 2-4. ARSENIC ELIMINATION IN ROASTERS
1.
2.
3.
4.
5.
6.
7.
8.
M A"
Type of Roaster
Multi-hearth roaster9
Multi-hearth roaster
Multi-hearth roaster0
Multi-hearth roaster
Fluid-bed roasterd
Fluid-bed roaster6
Fluid-bed roaster0
Fluid-bed roaster
Percent
Arsenic
in Feed
3.8
.22
N.A.
0.01 - 0.15
.20
0.99
N.A.
0.02
Percent of Arsenic in
Feed Volatilized
25
27
70 - 90
5-30
4-10
60 - 65
15 - 50
15
Information received on the ASARCO-Tacoma smelter from Mr. K.W.
Nelson, ASARCO, Incorporated, March 17, 1976.
Information on ASARCO-E1 Paso obtained from EPA testing (refer to
Appendix C).
Rozlovskii, A.A., "Behavior of Arsenic in the Production of Nonferrous
Metals," Tsvetenye Metally/Nonferrous Metals, NO. UDC 669.778.
Stankovic, D., "Air Pollution caused by Copper Metallurgy Assemblies in
Bor, ' Institute for Copper, Bor, Project No. 02-513-1, U.S. EPA.
p
Information on Anaconda obtained from Mr. Richard Sloane, Director of
Technology, Anaconda Company, October 11, 1978.
Information on Kennecott-Hayden received from Mr. 1.6. Pickering, Vice
President, Environmental Affairs, Kennecott Copper Corporation, May 9, 1978.
2-21
-------
in fluidized-bed roasters (seventh data point) with a great excess of
air, as little as 15 to 50 percent of the arsenic is eliminated. The
fourth and fifth data points also indicate that arsenic elimination in
multi-hearth roasters (5 to 30 percent) is greater than in fluidized-bed
roasters (4 to 10 percent). However, the amount of arsenic volatilized
in dependent upon the amount of roasting done or allowable, since
ASARCO-Tacoma using a "high" arsenic feed eliminates only 25 percent
of the arsenic in the feed, although more arsenic elimination is
possible. This, as ASARCO has pointed out, is due to the production
of low grade matte which is essential for the removal of impurities
20
during the converting operation. A relatively low grade matte leads
to higher converter temperatures which are essential for the vaporization
of impurities in the converter. Also, low grade matte tends to increase
the blowing time or "sweep" in the converter which facilitates impurity
elimination. Conversely, in the fluidized-bed roaster at the formerly
operated Anaconda smelter (sixth data point in Table 2-4), between 60
and 65 percent of input arsenic was volatilized. This was due to the
longer roast performed at Anaconda which also eliminated a large
amount of sulfur in the feed. Consequently, a high grade matte (50 percent)
is produced in the electric furnace.
2.2.2 Arsenic Behavior in Smelting Furnaces
Arsenic elimination in smelting furnaces is by volatilization and
slagging. Arsenic behavior in the smelting furnace is controlled by
the fact that smelting occurs in the presence of copper. As has long
21 22
been known, and confirmed by recent thermodynamic data, ' arsenic
is more stable in copper than in Cu-S, with a decrease in its thermo-
dynamic activity and, hence, volatility. As a result, in the smelting
furnace the volatility of arsenic will change with the grade of matte
being produced.
Since volatilization plays a major role in the elimination of As
during the production of matte in smelting furnaces, any factors which
enhance volatilization will further improve its elimination. These
are:
• Mineralization: some As compounds have higher vapor pressures
than others and will therefore volatilize to a greater
degree.
2-22
-------
• Green charge or calcine charge: a calcine charge will have
the more-easily volatilized As compounds removed in roasting;
thus, for a given original mineralization, a green-charged
furnace should volatilize a greater proportion of the As.
• Smelting temperature: the higher the temperature, the
greater the degree of volatilization.
• Exposure of slag surface free to charge material cover: the
greater the exposure, the greater the volatilization of As
from the slag.
The amount of arsenic slagged or volatilized differs from one
furnace to another. The behavior of arsenic in the different types of
furnaces is described in the following sections.
2'2'2-1 Reverberatorv Furnarps. The arsenic input into the
reverberatory furnace is from the arsenic in the new charge, slag
returned from the converters, and recycled dusts and reverts. This
arsenic is eliminated by either volatilization or slagging. The
amount of arsenic slagged or volatilized differs from one smelter to
another, as indicated in Table 2-5.
Data presented in Table 2-5 have been organized in two groups
Group A represents those furnaces which have a concentration of arsenic
of greater than 0.2 percent in the feed. Group B represents smelters
having a smelter feed with less than 0.2 percent arsenic content. All
percentages presented in Table 2-5 are on a 100 percent input basis in
the furnace.
Examining the data in Group A from ASARCO-Tacoma,23 ASARCO-El
Paso, Phelps Dodge-Ajo,25 and Anaconda26 (these data are for the old
Anaconda smelting configuration having a green reverberatory furnace)
the amount of arsenic volatilized and leaving the gas phase ranges
from 55 to 75 percent. However, all of these smelters have greater
than 0.2 percent arsenic in the feed. The ASARCO-Tacoma and ASARCO-El
Paso smelters feed roasted calcine into their reverberatory furnaces
but still have high volatilization. This is because only 25 percent
of the arsenic is volatilized in the roaster at the ASARCO smelters
The charge to the reverberatory furnace is therefore still high in
arsenic content.
2-23
-------
Table 2-5. ELIMINATION OF ARSENIC IN REVERBERATORY FURNACES
(100 percent input basis)
Type of Feed
Calcine? Group A
Calcine
Green^
Green
Green
Calcinef
Calcine^ Group B
Calcine^
r n
Green •
Calcine
GreenJ
Percent
Arsenic
in Feed
3.8
0.22
0.3
0.81
0-2
0-2
0.02
0.01-0.2
0.13
N.A.
0.0035
Percent Arsenic
Volatilization
69
62
55
73
70-90
15-70
37
5-30
29
12
Eliminated by
Slagging
13
20
25
10
5-15
26-60
16
35-55
26
54
51
N.A. - Not Available
NOTE: The percent arsenic content in calcine feed to the smelting
furnace is not known. The concentrations listed above are for
the feed into the roaster and do not represent arsenic concentration
in the calcine.
alnformation received on ASARCO-Tacoma from Mr. K.W. Nelson of ASARCO,
Incorporated, March 17, 1976.
bInformation obtained from testing at ASARCO-E1 Paso (refer to
Appendix C).
cSchwitzgebel, K., et al., "Trace Element Study at a Primary Copper
Smelter," Prepublication copy, EPA Contract Number 68-01-4136,
January 1978.
d"The Proposed Arsenic Standard: Feasibility and Estimated Costs of
Compliance for Three U.S. Copper Smelters," prepared by D.B. Associates
for OSHA Contract B-9-F-5-1663, March 25, 1976.
eRozlovskii, A.A., "Behavior of Arsenic in the Production of Nonferrous
Metals," Tsvetnye Metally/Nonferrous Metals No. UDC 669.778.
Information on Kennecott-Hayden received from Mr. 1.6. Pickering,
Vice President, Environmental Affairs, Kennecott Copper Corporation,
May 9, 1978.
9Stankovic, D., "Air Pollution Caused by Copper Metallurgy Assemblies
in Bor," Institute for Copper, Bor, Project No. 02-513-1, U.S. EPA.
hlnformation received from Mr. Hank Hansen of Kennecott Copper Corporation
on the old Kennecott-Utah smelter.
fuddle, R.W., "The Physical Chemistry of Copper Smelting," Institute
of Mining and Metallurgy, London, 1953.
JTurnbull, D.L., "Converter Practice at Mufulira," Seventh Commonwealth
Mining and Metallurgical Congress, 1961.
2-24
-------
The data obtained in Group B from Kennecott-Hayden,27 Bor-
Yugoslavia, Kennecott-Utah29 (for the old smelter configuration
utilizing the reverberatory furnace), and the infonnation obtained by
Ruddle indicate that the amount of arsenic volatilized and leaving
in the gas phase ranges from 5 to 27 percent. Also, the total elimination
in the furnace ranges from 40 to 60 percent of the input arsenic
This is reaffirmed by the Mufulira data.31 However, the aforementioned
smelters have less than 0.2 percent arsenic in the feed.
This leads to two basic observations for the reverberatory furnace:
1. With greater than 0.2 percent arsenic in the feed, 55 to
75 percent of the arsenic generally leaves in the gas phase and 10 to
25 percent is slagged out.
2. With less than 0.2 percent arsenic in the feed, 5 to 37 percent
of the arsenic generally leaves in the gas phase and 16 to 55 percent
is slagged out.
The aforementioned observations are consistent with those reported
in the U.S.S.R. (Group A) where the amount of arsenic volatilized
was lower for roasted feed (15 to 70 percent volatilized) since it has
low-arsenic content, as compared to unroasted feed (70 to 90 percent
volatilized) which has higher arsenic content. The amount of arsenic
eliminated by slagging ranged from 25 to 60 percent for roasted feed
and 5 to 15 percent for unroasted feed.
2-2.2.2 Electric Furnaces. Table 2-6 summarizes the available
information on arsenic elimination in electric furnaces. Paulson et
al., used calcine from the ASARCO-Tacoma smelter and ran tests on a
laboratory electric furnace. They found the behavior of arsenic to be
similar to that in the reverberatory furnace. However, some questions
remain regarding the manner in which these smelting tests were performed
In actual practice, copper smelting electric furnaces are operated
with a cold top - that is, a layer of solid charge is maintained over
the entire surface of the melt at all times. Paulson et al., carried
out their tests with the molten slag surface uncovered by the charge
material. In the latter case, the slag is hotter, and all the arsenic
vaporized from it will leave with the flue gases rather than be recondensed
when passing through the relatively cold charge layer.
2-25
-------
Table 2-6. ELIMINATION OF ARSENIC IN ELECTRIC FURNACES
(100 percent input basis)
Type of
Furnace
Electric3
Electric
Electric0
Type of
Feed
Calcine
Calcine
Calcine
Percent
Arsenic
in Feed
3.8
0.99
1.8
Percent Arsenic
Volatilization
38.3-68.2
8
24
Eliminated by
Slagging
1.5-11.2
78
50.5
NOTE- The percentage of arsenic content in calcine feed to the smelting
furnace is not known. The concentrations listed above are for
the feed into the roaster and do not represent arsenic concentration
in the calcine.
aPaulson, D.L., et al., "Smelting of Arseniferous Copper Concentrate
in an Electric Arc Furnace," United States Bureau of Nines Report of
Investigation 8144, 1976.
blnformation received on the Anaconda smelter from Mr. Richard Sloan,
Director of Technology, Anaconda Company, October 11, 1978.
c"Economic Impact of New Source Performance Standards on the Primary
Copper Industry: An Assessment," Final Report to U.S. Environmental
Protection Agency, EPA Contract No. 68-02-1349, October 1974.
2-26
-------
2.2.2.3 Flash Furnaces - The Noranda reactor flash smelting
furnace has been introduced by Kennecott at their Utah smelter. In
this process, copper concentrates and flux are fed into a horizontal
cylindrical reactor where smelting and converting occurs under the
dynamic action of oxygen-enriched air introduced through tuyeres below
the bath. Heat for smelting is provided by exothermic converting
reactions supplemented by burners. As indicated by Mackey et al.,34
the elimination of As in the Noranda process is more favorable at'
lower matte grades. The amount of arsenic volatilized is much greater
under Noranda matte-making conditions compared to arsenic volatilized
under regular copper-production conditions. Also, the amount of
arsenic reporting in the matte is much lower.
Table 2-7 shows that the amount of arsenic volatilized from four
different flash furnaces is near 76 to 85 percent, and the amount
slagged ranges from 7 to 17 percent.
2.2.3, Arsenic Behavior During Converting
Arsenic distribution in a copper converter between gases and slag
varies widely, as shown in Table 2-8. When a high grade matte is
converted, metallic copper appears early in the cycle and acts as a
collector for arsenic because arsenic is more stable and less volatile
in metallic copper than in copper sulfide. Depending upon the matte
grade, as much as 70 percent of the arsenic may report in the blister
copper, and only 30 percent is eliminated by volatilization or slagging.35
When a low grade matte is converted, the formation of metallic copper
is delayed, and a much larger proportion of the arsenic is eliminated.
Up to 92 percent » of the arsenic will be volatilized in such
cases. In normal practice, it is neither that low nor that high;
approximately 70 percent of the arsenic is volatilized in the converter,
and nearly 16 percent reports in the slag.
In some domestic copper smelters such as ASARCO-E1 Paso, it is
common to recycle low grade secondary materials and copper precipitates.
If materials to be smelted have high impurity content (particularly
lead speiss or other recycled materials), a low grade matte is necessary
to produce sufficient heat for elimination of these materials. Decreasing
matte grade allows time for the volatilization of impurities since the
slagging cycle increases with lower matte grade. Also, the longer
2-27
-------
Table 2-7. ELIMINATION OF ARSENIC IN FLASH FURNACES
(100 percent input basis)
Type of
Furnace
Noranda
Reactor
"KCS"b
Process
Noranda
Reactor
Otokumpu
Type of
Feed
Green
Green
Green
Green
Percent
Arsenic
in Feed
0.06
N.A.
0.14
0.17
Percent Arsenic
Volatilization
85
79
82
76
Eliminated by
Slagging
7
11
13
17
N.A. - Not Available
ariackey, P.O., et al., "Minor Elements in the Noranda Process," Paper
presented at the 104th Annual AIME Meeting, New York, February 16-20,
1975.
bGeorge, D.B., et al., "Minor Element Behavior in Copper Smelting and
Converting," Joint AIHE-MHIJ Conference, 1976.
Information on Kennecott-Garf i eld received from Mr. I.G. Pickering,
Vice President, Environmental Affairs, Kennecott Copper Corporation,
May 9, 1978.
dKoh, Shiroh, "Extractive Metallurgy of Kuroko," Paper presented at
the joint meeting of the MMIJ-AIME, Tokyo, May 24-27, 1972.
2-28
-------
Table 2-8. ELIMINATION OF ARSENIC DURING CONVERTING
(100 percent input basis)
Type of Matte
Producing Furnace
Reverberatory?
Reverberatory
Reverberatory *j
Reverberatory
Reverberatory?
Reverberatory
Outokumpu Flash9
KCS Smelting
Converter
Shaft Furnace1
.A. - Not Available
Percent
Cu in
Matte
N.A.
40
40
35-55
N.A.
35-55
45
70
N.A.
Percent Arsenic
Volatilization
73.0
92.0
92.0
76.0
50-85
55.0
43 0
73 0
70.2
Eliminated by;
Slagging
11.0
n
u
c n
o. u
19.0
10-35
23.0
90 n
<-.y . u
29.5
Ruddle R.W. "The Physical Chemistry of Copper Smelting," Institute
of Mining and Metallurgy, London, 1953. institute
bljS™at!°n received on ASARCO-Tacoma from Mr. K.W. Nelson,
ASARCO, Incorporated, March 17, 1976.
CInformation obtained from tests performed at ASARCO-E1 Paso (refer
to Appendix C).
"i t "J1 r/0"ut1on Caused by C°PP^ Metallurgy Assemblies
, Institute for Copper, Bor, Project No. 02-513-1, U.S. EPA.
eRozlovskii, A. A., "Behavior of Arsenic in the Production of
Nonferrous Metals, "Tsvetnye Metal ly/Nonferrous Metals, Number UDC
Dt K«'.et a1" "Trace Ele^nt Study at a Primary Copper
Smelter, Prepublication copy, EPA Contract Number 68-01-4136, January 1978.
Koh, Shiroh, "Extractive Metallurgy of Kuroko," Paper presented at
the joint meeting of the MMIO-AIME, Tokyo, May 24-27, "eritea at
i-7 / L. •
George, D.B., et al . , "Minor Element Behavior in Copper Smelting
and Converting," Joing AIHE-MMIJ Conference, 1976.
'information received from Technika, Sofia, Bulgaria, September 24, 1978.
2-29
-------
converting time allows a greater duration of gas sweep across the bath
for impurity removal. Hence, the concentration of matte grade governs
the recycle of intermediates in the converter.
2.2.4 Arsenic Balance
There is a wide variation in the distribution of arsenic among
the various products of a particular process from one low-arsenic
throughput smelter to another. This is due not only to the difference
in feed composition but also to factors such as temperature, blowing
rates, gas composition, and the analysis and relative quantities of
products. The data presented in Tables 2-5, 2-6, 2-7, and 2-8 are
summarized in Table 2-9 on a 100 percent input basis through the
smelting circuit. Table 2-9 shows that the wide variation in arsenic
elimination exists not only for different smelter configurations but
also for smelters having similar circuits. Detailed causes of these
variations are complex, and relatively little information can be found
in the literature. There are, however, relationships between process
parameters that affect arsenic and other minor element distributions
that can be observed from the data in Table 2-9.
The distribution of arsenic within the various smelting processes
is specifically controlled by the characteristics of the arsenic and
its environment. There appear to be two fairly distinct input levels
of arsenic which result in different distributions. Refering to
Table 2-9, it can be seen that for smelting processes with levels of
input arsenic greater than or equal to 0.2 or 0.3 percent, the resulting
arsenic distribution appears to be directly influenced by temperature
and the presence of copper. The copper tends to fix the arsenic. The
partial pressure of the arsenic trioxide is high enough to allow
volatilization of the arsenic through the slag layer in the smelting
furnace. This results in a lower proportion of the input arsenic
leaving with the slag. However, when the level of input arsenic is
less than 0.2 percent, the data in Table 2-9 indicate that a higher
proportion of the input arsenic is transferred to the slag. The
hypothesis is that in the case of lower arsenic input, the arsenic is
mechanically bound or locked into the feed, and therefore, not readily
released. This prevents it from being volatilized and tends to allow
it to enter the slag.
2-30
-------
I
co
Table 2-9. ARSENIC BALANCE -
WEIGHT PERCENT AS REPORTING IN SMELTER PRODUCTS (AMOUNT INPUT = 100)
PROCESSES
Reverberatory - Converters3
Reverberatory - Converters'3
Reverberatory - Converters0
Reverberatory - Converters'1
Reverberatory - Converters6
Multi-hearth - Reverb - Converters f
Multi-hearth - Reverb - Converters9
Multi-hearth - Reverb - Converters'1
Fluidized-Bed - Reverb - Converters'1
Fluidized-Bed - Reverb - Converters'
Fluidized-Bed - Electric - Converters1^
Multi-hearth - Electric
Multi-hearth - Electric - Converters'
Noranda Reactor - Converters'"
Noranda Reactor - Converters"
Fierce-Smith Reactor - Converters0
(KCS process)
Outokumpu Furnace - Converters'3
Input in
smel ter
feed
Wt. % As
in
charge
0.3
0.81
0.13
0.0035
N.A.
3.8
0.22
0.01-0.15
0.2
0.02
0.99
3.8
1 8
0.06
0.14
N.A.
0.17
Roasting
% As
volatilized
N.R.
N.R.
N.R.
N.R.
25.0
27.0
5-30
4-10
15
63.3
25.0
81
N.R.
N.R.
N.R.
N.R.
Smelting
% As
slagged
25.0
9.7
26
54.2
10.0
15.0
30-45
35-40
13
28.9
1.13-8.4
7.0
13.0
11.0
17.0
% As
volatilized
55.0
73.3
29
51
11.8
52.0
45.0
5-15
25-30
31
2.7
28.73-61.15
.0
85.0
82.0
79.0
76.0
% Cu
in matte
35-45
40.0
40.0
N.A.
N.A.
40 0
40.0
35-55
39-43
40
50
40
36
70.0
70
70.0
45.0
Converting
% As
% As
slagged volatilized
4.5
3.4
9
3.75
o
0.6
5-10
0-10
12.1
1 4
Experimental
11.0
11.3
AQ
24.8
1 7 n
11.5
20-40
15-20
28.7
^ fi
study done
.
Blister
% As
in copper
4.5
2.3
5.45
.u
0.9
1.5-3
1-2
0.2
. 1
on a pilot plant
1 0
2
o
o
2.0
1 9
A
O c
7 7
3.0
. C
.5
. /
2.0
N.A. - Not Available
N.R. - No Roasting
*The input into the furnace is a calcine feed.
-------
Footnotes for Table 2-9.
Schwitzgebel, K., et al., "Trace Element Study at a Primary Copper
Smelter," Prepublication copy, EPA Contract No. 68-01-4136, January
1978.
Burton, D.J., "The Proposed Arsenic Standard: Feasibility and Estimated
Costs of Compliance for Three U.S. Copper Smelters," prepared for
OSHA, Washington, D.C., Contract B-9-F-5-1663, March 25, 1975.
Conversation with Mr. Hank Hansen, Environmental Engineer, Kennecott
Copper Corporation, Salt Lake City, Utah, September 26, 1978.
Herneryd, 0., et al., "Copper Smelting in Boliden's Ronnskar Works
Described." Journal of Metals, March 1954.
eRuddle, R.W., "The Physical Chemistry of Copper-Smelting," Institute
of Mining and Metallurgy, Lond, 1953.
Correspondence from Mr. K.W. Nelson, ASARCO, Incorporated to
Mr. J. Padgett, Environmental Protection Agency, March 17, 1976.
^Harris, D.L., "Particulate and Arsenic Emission Measurements From a
Copper Smelter," Volume I (text), Monsanto Research Corporation, EMB
Project Report No. 77-CUS-6, EPA Contract No. 68-02-1404, Task No. 36,
1977.
Stankovic, D., "Air Pollution Caused by Copper Metallurgy Assemblies
in Bor," Institute for Copper, Bor, Project No. 02-513-1, U.S. EPA.
Data for Kennecott-Hayden received in the correspondence from
Mr. I.G. Pickering, Vice President Environmental Affairs, Kennecott
Copper Corporation, to Mr. D.R. Goodwin, Director, ESED, U.S. EPA,
May 9, 1978.
Information received on the Anaconda smelter fron fir. Richard Sloan,
Director of Technology, Anaconda Company, October 11, 1978.
L.
Paulson, D.L. et al., "Smelting of Arseniferous Copper Concentration
in an Electric-Arc Furnace," United States Bureau of Mines Report of
Investigations 8144, 1976.
"Economic Impact of New Source Performance Standards on the Primary
Copper Industry: An Assessment," Final Report to U.S. Environmental
Protection Agency, EPA Contract No. 68-02-1349, October 1974.
"Vlackey, P.J., McKerrow, G.C., Terassoff, P., "Minor Elements in the
Noranda Process," Paper presented at the 104th Annual AIME meeting,
New York, February 16th - 20th, 1975.
nData for Kennecott-Hayden received in the correspondence from
Mr. I.G. Pickering, Vice President Environmental Affairs, Kennecott
Copper Corporation, to Mr. D.R. Goodwin, Director, ESED, U.S. EPA,
May 9, 1978.
°George, D.B., Donaldson, J.W., Johnson, R.E., "Minor Element Behavior
in Copper Smelting and Converting," 1976 Joint AIME-MMIJ Conference.
pKoh, S., "Extractive Metallurgy of Kuroko," Joint Meeting MMIJ-AIME,
Tokyo, May 1972.
2-32
-------
While it has been theorized by some that multi-hearth roasters are
required to eliminate arsenic, this is not necessarily confirmed by
the available information. In general, with the combined processes of
the roaster and reverberatory furnaces, it appears that in the case of
the multi-hearth roaster, a larger amount of arsenic is volatilized
than in the reverberatory furnace. In the case of the fluidized-bed
reverberatory furnace combination, less arsenic is volatilized in the
fluidized-bed than in the reverberatory furnace. However, if the sum
of the roaster and furnace arsenic volatilization is considered, then
the total seems to be the same with either type of roaster (refer to
eighth and ninth data points in Table 2-9).
Arsenic distributions at domestic copper smelters were determined
by using (1) arsenic mass balances supplied by the smelters, (2) the
previously cited arsenic distribution information for foreign and
domestic copper smelters available in the literature, and (3) control
equipment performance data obtained by EPA testing at various U.S.
copper smelters. The estimated arsenic distributions for each domestic
copper smelter are presented in Appendix F. Complete recycle of the
collected flue dusts from the existing control devices at each smelter
was assumed in these balances. The arsenic removal efficiency of
existing control devices was taken into account in determining the
arsenic material balances for 10 of the 14 low-arsenic throughput
smelters. For the remaining four smelters, Phelps Dodge-Ajo, -Morenci,
ASARCO-Hayden, and Kennecott-Hurley, material balances were generated
for the smelter configurations and control equipment arrays projected
after completion of modernization programs. The modernization programs
for these smelters are discussed in Section 4.2.1.
Arsenic emissions from primary copper smelters can be categorized
as process and fugitive emissions. Process emission sources include
primary offgas emissions from roasting, smelting, and converting
operations. Fugitive emissions include those emissions which escape
from process flow streams due to leakage through process exhaust
systems, material transfer systems, or ineffective capture at the
source of generation.
2-33
-------
2.3.1 Process Arsenic Emissions
Information indicating the distribution of arsenic in the smelter
circuit was received from most domestic copper smelters. Based on
this information, arsenic mass balances for each smelter were performed.
Typically, the information obtained consisted of process material
balances and estimates or measurements of the arsenic content in these
materials. However, the information received was incomplete for a
number of copper smelters. In these cases, assumptions were made
regarding the behavior of arsenic based on the information presented
in Section 2.2 on the behavior of arsenic at copper smelters.
For those smelters for which information on arsenic distribution
was not obtained or the information provided was incomplete, an arsenic
distribution for the smelter was developed based on the type of smelting
configuration at that particular smelter, the level of arsenic input
to that smelter, and the availability of arsenic distribution information
from smelters having similar configurations. For example, lack of
complete arsenic distribution data for the projected configuration of
ASARCO-Hayden after installation of an INCO-flash furnace necessitated
use of an arsenic balance provided by Kennecott for the projected
arsenic distribution at the Hurley smelter after modification to INCO
smelting technology.
The control equipment efficiencies used in the arsenic mass
balances were based on tests performed at several existing units.
Tests performed on baghouses and electrostatic precipitators have
indicated that the collection efficiency of the control equipment for
arsenic is dependent upon the inlet temperature of the offgases and the
arsenic concentration of the gas stream, and that maximum collection
of arsenic trioxide occurs at reduced temperatures. As a result, the
collection efficiency can vary from 30 to 99.0 percent. The results
obtained are summarized in Table 2-10. Where company-supplied data
were not available, an arsenic removal efficiency of 99 percent was
ascribed to acid plants, due to the gas precleaning and conditioning
required for effective acid plant operation. Detailed test data are
available in Appendix C, and a discussion on the performance of these
control devices is presented in Section 3.2.1.
2-34
-------
Table 2-10. MEASURED ARSENIC COLLECTION EFFICIENCIES
OF CONTROL DEVICES9
Sine! ter
Phelps Dodge-Ajo
ASARCO-E1 Paso
Anaconda
Kennecott-Hayden
ASARCO-Tacoma
Detailed test data
Device
Reverb ESP
Roaster & Reverb -
Spray Chamber/ESP
Spray Chamber/Baghouse
Venturi Scrubber
Roaster Baghouse
available in Appendix C.
Operating
Temperature
(°C)
Inlet
327
212
263
329
91
Outlet
313
104
101
39
86
Collection
Efficiency
Percent
27.8
96.2
98.9
98.3
99.7
2-35
-------
The arsenic distributions for all domestic copper smelters are
given in Appendix F. These distributions are based on a full recycle
of flue dusts and converter slag (except at Kennecott-Hayden where
converter slag is not recycled). In most cases, the final arsenic
distributions have been obtained after performing iterative recycles
until steady-state conditions are reached. Steady-state conditions
are indicated by stabilizing the arsenic content of the blister copper
and the converter slag.
The estimated potential process arsenic emissions in the absence
of control for each of the domestic low-arsenic throughput copper
smelters are summarized in Table 2-11.
2.3.2 Fugitive Arsenic Emissions
Fugitive emissions may be characterized as emissions which escape
directly from the process area to the atmosphere rather than through a
flue or exhaust system. They result from leakage in and around process
equipment and from material handling and transfer operations. These
emissions may be considered as low level emissions as compared to
process emissions, since they usually leave the smelter at or near
ground level, whereas process emissions are discharged through a tall
stack.
Listed in Table 2-12 and shown in Figure 2-6 are potential sources
of fugitive arsenic emissions. These emissions depend upon the particular
types of equipment and operating practices employed by the copper
smelter.
2.3.2.1 Roaster.
2.3.2.1.1 Charging. Fugitive emissions during charging of
multi-hearth roasters seldom occur because of the water content (8 to
10 percent) in the feed, and the "choke feed" mechanism used on the
charging hoppers. Fugitive emissions during charging are seldom
emitted from a properly designed and operated fluidized-bed roaster
because of the enclosed feed and discharge system.
2.3.2.1.2 Leakage. Fugitive arsenic emissions from multi-hearth
roasters may be emitted from leaks that can occur at the doors located
at each one of the hearth levels, from holes in the actual shell of
the roaster, or from leaks around the shaft that holds the rabble
arms. Under normal operating conditions, these emissions are minimized
2-36
-------
ARMrnciTnn °F PROCESS ARSENIC EMISSION ESTIMATES IN
ABSENCE OF CONTROL FOR LOW-ARSENIC THROUGHPUT PRIMARY COPPER SMELTERS
Smel ter
ASARCO-E1 Paso
ASARCO-Hayden
Tennessee Chemical Co. -
Copperhill
Inspiration-
Miami
Kennecott-Garfield
Kennecott-Hayden
Kennecott-Hurley
Kennecott-McGill
Magma- San Manuel
Phelps Oodge-Ajo
Phelps Dodge-
Douglas
Phelps Dodge-
Hidalgo
Phelps Dodge-
Morenci
Copper Range -
White Pine
Sources: MHR - Multi hearth Roaster
REV - Reverberatory Furnace
CONV - Converters
FF - Flash Furnace
Emission
Source
MHR
REV
CONV
FF
CONV
FBR
EF
CONV
EF
CONV
NOR
COW
FBR
REV
CONV
FF
CONV
REV
CONV
REV
CONV
OXREV
CONV
MHR
REV
CONV
FF
CONV
OXREV
CONV
REV
CONV
FBR
EF
NOR
OXREV
Exit Gas
Arsenic Content
kg/hr
86.6
54.9
76.2
157.0
45.0
0 ?
u • c.
0.4
0.5
4.5
1.5
115.0
6.0
1.9
5.3
5.0
1.3
0.4
18.6
35.6
1.8
0.5
45.3
2.3
2.2
0.9
3.2
13.7
0.9
4.3
5.3
0.4
0.2
- Fluid Bed Roaster
- Electric Furnace
- Noranda Reactor
- Oxygen-Sprinkle
Reverberatory Furnace
2-37
-------
Table 2-12. POTENTIrtL SOURCES OF FUGITIVE
ARSENIC EMISSIONS
Roaster
Charging
Leakage
Hot calcine discharge and transfer
Smelting Furnace
Charging
Leakage
Matte tapping
Slag tapping
Converter slag return
Converters
Charging (matte, reverts, flux, lead smelter by-products, cold
dope)
Blowing (primary hood leaks)
Skimming
Holding
Pouring of slag and blister
Converter leaks
Anode Furnace
Charging
Blowing
Holding
Pouring
Miscellaneous
Dust handling and transfer
Ladles (matte and slag)
Slag dumping
2-38
-------
ro
i
UD
<•(. II 5IURAGC
UNLOADING I
COMIC HANOIING
ROASTEH
/ CHARGING
"OAJffH LEAKAGE
ANODE
FURNACE
CHAM! Of
10 ANODE FURNACE
» SLAG
» HANOI INS
LIMESTONE
UNLOADING l—k'
&
COffEU
» CASriNG
10 vnnm
REVERBERATORY FURNACE
Figure 2-6. Fugitive Emission Sources at Primary Copper Smelters
-------
by operating the multi-hearth r asters under a slight negative pressure
required to supply induced air for oxidation.
A fluidized-bed roaster is essentially a vertical cylinder of
steel plate lined with insulation and fire bricks. This type of
roaster is fed by and discharges into closed systems. Therefore,
leakage from fluidized-bed roasters is negligible.
2.3.2.1.3 Hot calcine discharge and transfer. Fugitive emissions
may be generated during discharging and transfer of hot calcine from
roaster to smelting furnaces. Smelters with multi-hearth roasters
usually use larry cars (small rail cars) to transport calcines to the
furnace. When the material is dropped from the calcine hopper located
under the roaster into the covered car through a feed opening, large
quantities of dust are generated as a result of material movement and
pressure changes within the car. Some fugitive emissions can also
occur during the transportation of the roaster calcines to the smelting
furnace. In the case where larry cars are used, their feed opening is
usually covered to minimize this effect.
Hot calcine discharge from fluidized-bed roasters primarily
occurs by entrainment of the calcine in the gas going from the top of
the roasters into a series of cyclones. The material drops into the
cyclone hoppers. Generally, about 80 to 85 percent of the calcine
passes through the top of the roasters. The "underflow" or remaining
calcine flows through an opening near the base of the roaster. The
roaster and cyclones are operated under positive pressure and, therefore,
must be airtight and free of leaks. Consequently, fugitive emissions
are seldom emitted from a properly operated fluid-bed roaster during
calcine transfer operations since it is essentially a closed system.
42 43
Tests were performed by EPA at ASARCO-Tacoma, ASARCO-E1 Paso,
and Phelps Dodge-Douglas44 to obtain fugitive arsenic emission estimates
from the calcine transfer systems used at these smelters. The results
of these tests are presented in Table 2-13.
The calcine transfer systems in the aforementioned smelters
differ in design. At the ASARCO-Tacoma smelter, an enclosed spring-sealed,
apron-type unloading system is used. The transfer of calcine in this
system is completely enclosed, and a minimum of fugitives are emitted.
The calcine transfer systems used at the ASARCO-E1 Paso and Phelps
2-40
-------
fSD
I
Table 2-13. FUGITIVE ARSENIC EMISSIONS DURING CALCINE TRANSFER FROM MULTIHEARTH ROASTERS
Smelter
ASARCO-Tacoma
ASARCO-E1 Paso
Phelps Dodge-
Douglas
Sample
Run
1
2
3
Avg.
1
2
3
Avg.
1
2
3
Avg.
Arsenic
in Calcine
(Ib/hr) kg/hr
719
719
1,073
130
130
130
10.5
1.0
2.3
326
326
987
59
59
59
4.8
0.5
1.0
English
ppm
69.7
109.3
153.6
110.9
3.7
1.03
2.5
2.41
0.0
0.0
0.0
0.0
Arsenic
Units
Ib/hr
1.0
1.7
2.7
1.8
0.33
0.09
0.22
0.22
0.007
0.026
0.064
0.032
Emissions
Metric
Mg/m3
217.3
340.9
479.0
345.7
11.53
3.23
7.81
7.52
0.059
0.22
0.642
0.307
Units
kg/hr
0.46
Q.lf
1.24
0.82
0.15
0.04
0.10
0.10
0.003
0.011
0.029
0.014
Comments
Holding aprons on hopper exit and
larry car. Probable 100 percent
collection efficiency. Intermit-
tent testing on one set or roasters.
Probable 50 percent collection
efficiency. Testing was done on a
continuous basis.
Probable overall collection
efficiency 70 percent. Testing on a
continuous basis. Calcine transfer
8-hour shifts.
-------
Dodge-Douglas smelters are, howe^r, not as effective. A description
of these systems is provided in Section 3.2.2.2.
Fugitive arsenic emission estimates were developed based on the
measurements presented in Table 2-13. EPA testing of the calcine
transfer system at ASARCO-Tacoma was performed only when the system
was in operation. At the ASARCO-E1 Paso and Phelps Dodge-Douglas
smelters, the systems were tested on a continuous time basis. To
compare test results on a similar basis, the ASARCO-Tacoma test results
were adjusted from an intermittent to a continuous basis. This was
performed by averaging the measured arsenic emission rate obtained
from intermittent testing over the total time period of the testing.
Also, since the test results at ASARCO-Tacoma represented only half
the roasters typically operated at any one time, adjustments were also
made to account for the other roasters, in operation, but not tested.
For example, sample run number 1 at ASARCO-Tacoma indicated an
arsenic emission rate of 0.46 kg/hr (1.0 Ib/hr). Test data45 indicated
an actual testing time of 15 minutes over an elapsed time period of
1.76 hours. The total arsenic emitted during the 15 minutes was
therefore 0.12 kg, which was equivalent to an average emission rate of
0.07 kg/hr (over a time period of 1.76 hours), assuming that the
second row of roasters emitted the same amount of fugitive arsenic
emissions as the first set of roasters over this same time period.
The total uncontrolled fugitive arsenic emission rate during calcine
transfer was 0.14 kg/hr.
Corrections were also made to the ASARCO-E1 Paso and Phelps
Dodge-Douglas smelters' test results to account for the collection
efficiencies of the calcine transfer systems in use at these smelters.
The results thus obtained indicated the average uncontrolled fugitive
arsenic emission rates for the ASARCO-Tacoma, ASARCO-E1 Paso, and
Phelps Dodge-Douglas smelters to be 0.16 kg/hr, 0.10 kg/hr, and
0.014 kg/hr, respectively.
For the purpose of developing an emission factor for fugitive
arsenic emissions occurring during calcine transfer from multi-hearth
roasters, the fugitive arsenic emissions from the three smelters were
compared to the quantity of arsenic reporting in the calcine during
the test period. The ASARCO-Tacoma, ASARCO-E1 Paso, and Phelps Dodge-Douglas
2-42
-------
test results indicated that an enrage of 0.05, 0.33, and 2.6 percent
respectively, of the arsenic in the calcine reports in the fugitives
A correction factor for the probable collection efficiencies during
calcine transfer at the three smelters is included in the aforementioned
results. However, there is some question regarding the accuracy of
the arsenic emission estimates made at the Phelps Dodge-Douglas and
ASARCO-E1 Paso smelters. Calcine samples obtained at the Phelps
Dodge-Douglas smelter indicated small amounts of arsenic present (less
than 0.015 percent). Accurate readings for small arsenic contents are
difficult to obtain. Transfer of calcine from roaster to larry cars
at the ASARCO-E1 Paso smelter takes place in a long, rectangular shed.
The shed is open at one end and has an exhaust duct at the other end
The configuration of the system is such that not only are fugitive
emissions captured by the exhaust system during calcine transfer, but
also reentrained fugitive dust generated by the movement of the larry
cars and by wind action.
In view of the aforementioned, the ASARCO-Tacoma smelter results
were used only to compute the calcine transfer fugitive arsenic emissions
factor. The ASARCO-Tacoma test results indicated that an average of
0.005 kg of arsenic reports in the fugitive emissions per kg of arsenic
contained in the calcine, or 0.05 percent. This factor was used to
estimate calcine transfer of fugitive arsenic emissions from multi-hearth
roasters at all domestic copper smelters. The amount of arsenic
reporting in the calcine for these estimates was obtained from the
process arsenic mass balances provided in Appendix F.
2-3-2-2 Smelting Furnace. As noted previously, three basic
types of smelting furnaces are used. These include reverberatory
type, electric, and flash furnaces. A discussion on the fugitive
emission sources associated with the operation of these furnaces
follows.
2.3.2.2.1 Charging. Reverberatory furnaces can be charged by
any of the following methods: (1) by dropping the calcine through the
arch of the furnace, (2) by using retractable or fixed Wagstaff guns
located at the furnace side walls, (3) by charging through ports in
the side walls which extend some distance toward the middle of the
furnace, and (4) through green charge slingers that cover the entire
bath in the smelting zone.
2-43
-------
In general, when a green o>" calcine charge is dropped into the
smelting furnace, a rapid increase in gas volume occurs because of
metallurgical reactions occurring within the furnace. This can cause
a temporary excess pressure in the furnace depending upon the response
of the pressure control system. If the pressure exceeds atmospheric
level, fugitive emissions can be generated through all the furnace
openings. With side-charged reverberatory furnaces, there exists the
possibility of a portion of the charge bank flowing, caving in, or
sloughing into the molten bath and creating a rapid reaction between
the charge and the bath. In such instances, generation of gas can
occur so rapidly that che furnace arch can be seriously damaged by the
excessive pressure in the furnace.
The seal between the retractable Wagstaff gun and the furnace
wall, and the furnace opening required for slinger feed type charging,
are additional openings through which fugitive emissions can be emitted
when these types of charging techniques are used.
Electric furnaces are charged continuously either from the roof
or from the side. The charge almost always consists of dried concentrates
or calcine since moisture tends to cause steam explosions. Oxygen
requirements for electric furnaces are much smaller compared to rever-
beratory furnaces because no fuel is being directly burned. There is
Ucs likelihood of openings being present to allow excess air as is
the case with the reverberatory furnace, thereby minimizing fugitive
emissions as a result of pressure surges. When emissions do occur,
they generally appear at the seals of the electrodes.
Feed to flash furnaces is usually from a concentrate dryer where
the concentrate-to-flux ratio is established by computer, based on the
chemical composition of the materials and the desired matte grade.
Dried concentrates from the dryer are discharged to the feed receiver
by a pneumatic air lift. Air injected into the receiver lifts the
dried furnace feed to an air expansion vessel. A drag conveyor from
the expansion vessel conveys the feed to the storage bin. The feed
from the storage bin to the flash furnace reaction shaft is conveyed
by variable speed drag conveyors. A carefully controlled mixture of
dry concentrate and pretreated air is then injected to the furnace
through concentrate burners from the roof of the reaction shaft. The
2-44
-------
charging systems for flash furnaces are gas-tight, and so fugitive
emissions are seldom emitted.
Depending upon the method used for charging, and the type of
smelting furnace, fugitive emissions can be generated. However, by
using add-on control systems which are activated at the time the
charge is dropped, effective reduction of potential fugitive emissions
can be obtained. Since most smelters are currently using such control
systems, it is reasonable to assume that fugitive arsenic emissions
during charging of smelting furnaces are negligible.
2.3.2.2.2 Leakage. The structure of most reverberatory furnaces
has to allow for thermal expansion. This leaves many leakage points
in the reverberatory furnace outer shell. Fugitive emissions from
these leakage points are emitted when the pressure in the reverberatory
furnace exceeds atmospheric pressure. This, as has been pointed out
earlier, can occur during charging and converter slag return.
Proper maintenance is a major factor in minimizing leaks. Hand
or spray sealing of all cracks between refractories, if practiced on a
routine basis, considerably reduces fugitive emissions due to leakage
in a reverberatory furnace.
Usually, reverberatory furnaces are operated under very slight
negative pressure at all domestic copper smelters. This is performed
not only to prevent fugitive emissions from the multitude of leakage
points around the furnace but to avoid overheating of the furnace by
providing the cooling effect of air infiltration. Pressure controls
are used to maintain a slight negative pressure and also to ensure
that pressure surges are kept under control.
In electric furnaces, there is much less of a requirement for
providing oxygen to the furnace as compared to the fuel-fired rever-
beratory furnace. Therefore, these furnaces are usually better sealed.
Thus, if a positive pressure is momentarily generated, fugitive emissions
due to leakage are considerably less.
The construction of flash furnaces is such that they are virtually
gas-tight. This prevents leakage of emissions and cooling by air
infiltration.
It is reasonable to assume that fugitive arsenic emissions due to
leakage from smelting furnaces are negligible provided they are operated
2-45
-------
and maintained properly. Based or this observation, it was concluded
that normally no fugitive arsenic emissions are discharged due to
leakage.
2.3.2.2.3 Matte tapping. Matte tapping is a principle fugitive
emission source at the smelting furnace. Smelting furnaces have from
one to three matte tap holes with associated launders on each side.
Normally, only one tap is used at a time. The launder directs the
flowing matte to a point where it can be collected in a large ladle.
Fugitive emissions are observable from the point at which the matte
leaves the furnace to the location where it enters the ladle. Typically,
a matte tapping operation takes 5 to 15 minutes.
EPA performed testing at the ASARCO-Tacoma, ASARCO-E1 Paso,
and Phelps Dodge-Ajo48 smelters to determine captured fugitive arsenic
emissions during matte tapping. The results are presented in Table 2-14.
Fugitive arsenic emission estimates for matte tapping were developed
based on the results of the arsenic emission measurements presented in
Table 2-14. It was assumed that complete capture of matte tapping
fugitive emissions was obtained by the exhaust hood systems. EPA
testing of the matte tapping operations at ASARCO-Tacoma was only
performed when actual matte tapping was being conducted. However, the
testing at the ASARCO-E1 Paso and Phelps Dodge-Douglas smelters was
performed on a continuous basis over a period of time. To compare the
test results on a similar basis, the ASARCO-Tacoma test results were
therefore adjusted from an intermittent to a continuous basis. This
was performed by averaging the emission rate of arsenic obtained from
testing during the time the matte tapping system was in operation over
a 24-hour time period. For example, sample run number 1 at ASARCO-Tacoma
(refer to Table 2-14) indicated an arsenic emission rate of 1.69 kg/hr
(3.73 Ib/hr). Process data49'50 indicated that 7 minutes are required
to fill a matte ladle, and 36 ladles were filled on the day sample run
number 1 was made. Therefore, actual time of matte tapping hood
system operation was 4.2 hours. Total fugitive arsenic matte tapping
emissions emitted were 7.1 kg/hr (15.6 Ib/hr). The average fugitive
arsenic matte tapping emission rate for sample run number 1 at ASARCO-Tacoma
was therefore 0.3 kg/hr (0.65 Ib/hr).
2-46
-------
Table 2-14.
MATTE TAPPING EMISSIONS FROM COPPER SMELTERS
Arsenic Emissions
Phelps Dodge-
Hood is only on one matte tap
hole. The 6 ft. launder was
uncovered.
Approximately 11 x 103 acfm
of additional flow was bypassed
52.42
243.05
236.18
177.22
The testing was done only during
the matte tapping operation.
N.A. - Not available
-------
At the ASARCO-E1 Paso smelt,r, approximately 11,000 acfm of flow
was being bypassed upstream of the test location. The ASARCO-E1 Paso
test results were therefore adjusted by the ratio of the total flow
rate (measured flow rate plus bypass flow rate) to the measured flow
rate. The results thus obtained, indicated the average fugitive
arsenic matte tapping emission rates from the ASARCO-Tacoma, ASARCO-E1
Paso, and Phelps Dodge-Ajo smelters to be 1 kg/hr (2.15 Ib/hr), 0.31 kg/hr
(0.67 Ib/hr), and 0.19 kg/hr (0.41 Ib/hr), respectively.
For the purpose of developing an emission factor for fugitive
arsenic matte tapping emissions, the fugitive arsenic emissions from
the three smelters were compared to the quantity of arsenic reporting
in the matte during the test period. The ASARCO-Tacoma and ASARCO-E1
Paso test results indicated Lhat an average of 1.9 and 1.5 percent of
the arsenic in the matte reports in the fugitives, respectively. The
Phelps Dodge-Ajo smelter process samples indicated extremely small
amounts of arsenic51 in the matte (refer to Table 2-14). In view of
the difficulty in obtaining accurate readings for such small arsenic
contents, the fugitive arsenic matte tapping test results for Phelps
Dodge-Ajo were not considered in the development of the emission
factor.
Based on the ASARCO-Tacoma and ASARCO-E1 Paso results, an emission
factor of 1.5 percent of the arsenic in the matte was assumed to
report in the matte tapping fugitive emissions. This emission factor
was then used to'estimate fugitive arsenic matte tapping emissions at
all the domestic copper smelters regardless of the kind of smelting
furnaces in use. The amount of arsenic in the matte at these smelters
was obtained from the process arsenic mass balances in Appendix F.
2.3.2.2.4 Slag tapping. Slag tapping is a principle fugitive
emission source at the smelting furnace. Slag tap ports and slag
launders have been observed to emit less fugitive emissions than those
emitted during matte tapping operations. However, fugitive emissions
are observable from the point of the slag leaving the furnace to the
location where it enters the ladle. Typically, a slag tapping operation
takes 10 to 20 minutes. 52
EPA performed testing at the ASARCO-Tacoma smelter to estimate
fugitive arsenic emissions during slag tapping and to evaluate the
2-48
-------
capture system in use at the sm^er. At the ASARCO-Tacoma smelter
both the slag tap hole and the slag pot are provided with exhaust
hoods, and the slag launder is covered. The captured slag tapping
emissions are ducted to the brick flue. The emissions sampling was
conducted in the duct leading to the brick flue. Results of this
testing are provided in Table 2-15.
Testing at the ASARCO-Tacoma smelter was only performed when slag
was being tapped from the reverberatory furnace. Three sample runs
were made, and during these runs the slag was also analyzed for arsenic
content. The actual test time for sample run number 1 was 60 minutes
On the day sample run number 1 was made, 120 pots of slag were dumped
Approximately 3 minutes were required to fill a slag pot.53 The
actual on-time for the slag tapping exhaust hood system was therefore
6 hours. At an emission rate of 0.25 kg/hr (0.56 Ib/hr) the arsenic
emitted over a 14 hour period was 1.5 kg (3.3 lb/) at an average rate
of 0.06 kg/hr (0.14 Ib/hr). Similar calculations for sample runs 2
and 3 were performed, and an average arsenic emission rate of
0.12 kg/hr (0.25 Ib/hr) was obtained for the three sample runs.
For the purpose of developing an emission factor for fugitive
arsenic slag tapping emissions, the arsenic reporting in the fugitives
was compared to the amount of arsenic in the furnace slag. It was
assumed that the arsenic in slag tapping fugitives-to-arsenic in slag
ratio remained constant regardless of the type of smelting furnace
Also, fugitive arsenic emissions discharged during slag tapping at the
ASARCO-Tacoma smelter were all captured by the exhaust hood system.
At the ASARCO-Tacoma smelter this comparison indicated that an
average of about 0.1 percent of the arsenic in the slag (samples taken
during test period) reported in the fugitive emissions. Based on
these results, a factor of 0.1 percent arsenic in the slag was assumed
to report in the slag tapping emissions at all smelters. The amount
of arsenic in the slag at each smelter was obtained for the process
mass balances discussed in Appendix F.
2'3'2-2'5 Converter slag return. Converter slag is returned to
the reverberatory and electric furnaces through the converter slag
return launder. This is either a simple channel with an opening in
the furnace wall or a mechanically operated chute.
2-49
-------
Table 2-15. SLAG TAPPING FUGITIVE ARSENIC EMISSIONS
FROM ASARCO-TACOMA
Sample
Run
1
2
3
Avg.
Arsenic in
Slag
Ib/hr kg/hr
208 94
275 125
290 132
Measured Emi
Concentration
g/m3
7.94
23.04
18.16
16.38
ssions
Mass Rate
Ib/hr
0.56
1.43
1.17
1.05
Comments
Testing was only
done when the slag
was being tapped.
2-50
-------
The number of times convert slag is returned to the furnace
depends upon the number of converters and the operating level of the
smelter. Each time the slag is returned, pressure fluctuations occur
in the furnace. These fluctuations are due to the agitation in the
bath and the rapid chemical reactions with the slag constituents.
This tends to generate fugitive emissions through the relatively large
converter slag opening.
To develop an emission factor for fugitive arsenic emissions
discharged during converter slag return to the smelting furnace, tests
were performed by EPA at the ASARCO-Tacoma smelter.54 The emissions
during converter slag return were sampled during 1 to 3 minute intervals
when slag returned to the reverberatory furnace from the converters.
Since this procedure occurs only for a short duration, testing was
performed for 3 days to obtain an adequate sample for analysis.
The test results over a sample period of 23 minutes (18 ladles of
converter slag returns) indicated 0.13 kg/hr (0.3 Ib/hr) of arsenic
was emitted. The total actual elapsed time during this test period
was 5 hours. Therefore, the average fugitive arsenic emission rate
during converter slag return was 0.01 kg/hr (0.02 Ib/hr). Analysis of
the converter slag indicated that during the 3 days in which the tests
were performed, an average of 25 kg/hr (56 Ib/hr) of arsenic was being
returned from the converters to the furnace with the slag. Comparison
of the arsenic in the fugitives to the arsenic in the converter slag
indicated 0.0004 kg of arsenic reports in the fugitives per 1 kg of
arsenic returned with the slag.
Based on these results, it was concluded that the amount of
arsenic emitted (0.01 kg/hr) during converter slag return was negligible
It should be noted that the arsenic input at the ASARCO-Tacoma smelter
is h1gher than the combined arsenic input at all the other smelters.
2.3.2.3 Converter.
2.3.2.3.1 Charging. During converter charging, the converter is
tilted by a drive mechanism until the mouth of the converter is approximately
45 degrees from the vertical. Fugitive emissions during this operation
result when matte or other materials such as reverts flux are poured
from a ladle into a converter. The gate on the primary hood is retracted
2-51
-------
to its highest position. An ovc.nead crane lifts the matte ladle
above the mouth of the converter and pours the matte into the converter
by tilting the ladle. During this operation, visible emissions are
heavy, but of short duration. When charging is completed, blowing air
is turned on while the tuyeres are above the liquid level of the
molten bath. The converter is rotated until its mouth is in its
topmost position, contained within the primary hood. The gate is then
lowered and the blowing cycle commences.
2.3.2.3.2 Blowing. Most domestic smelters have attempted to
provide relatively close fitting primary hoods on converter openings.
These hoods are used to contain and capture offgases generated during
blowing operations. However, these hoods at best do not completely
seal the opening since sufficient space has to be provided for easy
movement of the converter shell. This is to allow for converter shell
irregularities and the molten copper buildup on the shell due to
splashing during the blowing operation. Some fugitive emissions are
discharged through these openings, especially when pressure surges
occur during converter blowing, and rapid adjustments are not made by
the duct damper system.
2.3.2.3.3 Skimming. During skimming operations, the mouth of
the converter is rotated to a position between 65 to 85 degrees from
the vertical, depending upon the bath level. The blowing air remains
on as the converter is rolled out, until the tuyeres are above the
surface of the bath. This action results in significant quantities of
fugitive emissions, but is necessitated by equipment requirements.
Slag is skimmed from the converter mouth into a slag ladle. Fugitive
emissions are visible during this operation. The primary hood gate
may not be fully retracted during skimming; however, the hood is
isolated by dampers from the main duct system to keep dilution air
from mixing with the high S02 offgases from blowing converters. At
the completion of skimming, the converter mouth is again rotated to
its vertical position within the hood. The gate is fully extended,
and the blowing cycle resumes.
2.3.2.3.4 Pouring. As the blister copper is poured, the converter
is slowly rotated downward until its mouth reaches a position approximately
90 to 125 degrees from the vertical, depending upon the size of the
2-52
-------
pour and the buildup within th- converter. The hood gate may be
partially extended during this operation. Fugitive emissions during
this operation result as the blister copper is poured from the converter
into the ladle. After the blister pour, the converter mouth is rotated
upward to a position approximately 45 degrees from the vertical to
await a new matte charge and the start of a new cycle.
2.3.2.3.5 Holding. There are times during normal smelting
operations when material, either slag or blister copper, cannot be
immediately transferred from the converters to the ladles. This may
be due to such conditions as unavailability of the crane, refining
furnace operations not allowing additional feed material, or others.
It then becomes necessary for the converter to be placed in a holding
mode. In this mode the converter is rotated until the mouth of the
converter is 30 to 45 degrees from the vertical to keep the tuyeres
out of the bath. This results in fugitive emissions from the molten
material in the converter being discharged into the converter building.
2'3'2-3-6 Converter leaks. Since the ends of most Pierce-Smith
converters are joined by bolts and springs, they occasionally leak at
the end joint. When this leakage is located below the molten material
surface, it is usually repaired rapidly to prevent major erosion.
However, in those cases where it is located above the charge surface,
H may not be repaired. Thus, fugitive emissions may occur at this '
point.
EPA performed testing at the ASARCO-E1 Paso smelter55 to develop
a fugitive arsenic emission factor for fugitive emissions occurring
during converting. The results indicated an emission factor of 150 g
of arsenic per 1 kg of arsenic contained in the converter primary
offgas stream. Thus the emission factor of 15 percent of the arsenic
contained in the converter primary offgas, based on test data from
ASARCO-E1 Paso, was used for estimating converter fugitive emissions
at the remaining low-arsenic throughput copper smelters.
2.3.2.4 Anode Furnace. Refining of blister copper to anode
copper is performed in rotary-type refining furnaces which are similar
to Pierce-Smith converters. Unlike converters, however, these furnaces
have no control whatsoever. Offgases generated during charging,
2-53
-------
blowing, skimming, holding, and pouring are discharged directly to the
building atmosphere from the fuinace mouth. It should be noted,
however, that emissions generated in the anode or refining furnaces
are considerably less significant both in terms of volume and arsenic
content than from converters. This is because the quantity of impurities
removed consists of only about 1 percent of the molten material processed.
Tests by ASARCO-Tacoma were performed separately during the
various phases of the anode furnace operations. The results indicated
an anode furnace fugitive emission factor of 110 g of arsenic per 1 kg
of arsenic contained in the blister copper. Thus an emission factor
of 11 percent of the arsenic contained in the blister copper was
obtained, based on test data from ASARCO-Tacoma. This emission factor
was then used to estimate fugitive arsenic emissions from anode furnaces
at the other domestic smelters.
2.3.2.5 Miscellaneous Fugitive Emission Sources.
2.3.2.5.1 Dust handling and transfer. Dust handling and transfer
can generate fugitive emissions if carelessly performed. However,
most smelters take reasonable precautions to minimize fugitive emissions
from dust handling and transfer. Dust transfer from control devices
onto conveyors is typically performed by mechanically timed and activated
rotary valves or twin gravity self-closing gates. These conveyors are
generally covered by housings and discharge into storage bins from
which dust may be withdrawn as desired. Dust transfer from storage
bins is usually through dust-tight connections to surface transportation
units such as tank trucks and dumpsters. Cleaning and unloading of
dust from flues and settling chambers is performed by conveyors which
feed into hoppers provided at spaced intervals underneath the flues
and settling chambers. Both screw and drag-type conveyors are used.
These flue dusts are usually treated in a pugmill or pelletizing disc
where moisture is added. The wet dust is then transferred to a bedding
area, blended with other feed constituents, and recycled. Dust from
waste heat boilers and crossover flues is usually removed by hand
methods, water-bomb lances, slugger guns, and other methods. This
dust is handled and transported by surface vehicles to the smelter
flux crushing system. A more detailed description of dust transfer,
handling, and conveying is given in Section 3.1.2.5.
2-54
-------
In the transfer, handling, .nd conveying of dust from control
device storage hoppers, smelter flues, and dust chambers, it was
assumed that all transfer systems such as screw or drag type conveyors
are dust-tight. The uncontrolled fugitive arsenic emissions from
these systems were assumed to not exceed 0.1 percent of the arsenic in
the dust collected at arsenic drop-out points such as control device
storage hoppers, smelter flues, dust chambers, and waste heat boilers.
This emission factor was based on particulate (dust) emission rates
found in grain transfer, asphalt batching, and other industries involved
in transfer, handling, and conveying operations. In these industries,
0.1 percent of the dust transferred, handled, or conveyed is estimated
to appear as fugitives.57
2.3.2.5.2 Ladles. Normal process fluctuations may require that
ladles containing matte, converter slag, or blister copper be temporarily
set aside until needed. Because these ladles contain molten material,
some emissions can be observed due to fuming. These, however, are
short-lived. The exposed surface of the material cools rapidly,
forming a solidified layer or skull which greatly limits fugitive
emissions. It was therefore assumed that no fugitive arsenic emissions
are discharged from ladles once they are filled.
2-3-2.5.3 Slag dumping. Smelting furnace slag is disposed of by
water granulation or by transport in the molten state for dumping a
short distance from the smelter. Slag dumping is the more widely used
method. The slag is transported to the dump site by train or slag
hauler. The slag train is usually comprised of a number of slag pots
or ladles on flat cars. Solidification at the surface of the slag in
the pots is fairly rapid. Fugitive emissions during transporation to
the dumping site are therefore limited. However, during dumping of
slag at the dumping site, substantial fugitive emissions, although
short in duration (less than 1 minute), can be observed.
Testing was performed at the ASARCO-Tacoma smelter by EPA to
determine the magnitude of fugitive arsenic emissions during slag
dumping. Reverberatory slag from slag pots was analyzed for arsenic
content on exit from the reverberatory furnace and after the slag was
deposited on the dump site. Process sample analysis was performed on
four separate days. The test results obtained are listed in Table 2-16.
2-55
-------
Table 2-16. REVERBERATORY FURNACE SLAG
ANALYSIS FOR ARSENIC CONTENT AT ASARCO-TACOMA
Percent arsenic in slag at
exit from furnace slag Percent arsenic in
Sample Run launder slag at dump site
1 0.33 0-40
2 0.38 0.52
3 0.46 0.44
4 0.49 0.29
Avg. 0.42 0.41
2-56
-------
Test results indicated that m the first two of the four samples
the arsenic content of the reverberatory furnace slag increased in
concentration at the dump site. In the third sample there was a
slight decrease in concentration at the dump site. The fourth sample
indicated a larger decrease in concentration at the dump site. However,
comparison of the averaged results indicated approximately the same
concentration of arsenic in the slag at the exit from the furnace and
at the dump site. This indicates that fugitive arsenic emissions
during slag dumping are negligible.
Table 2-17 presents a summary of the potential fugitive arsenic
emission estimates in the absence of control for the 14 low-arsenic
throughput primary copper smelters.
2-57
-------
TABLE 2-17. SUMMARY OF POTENTIAL FUGITIVE ARSENIC EMISSION ESTIMATES
IN ABSENCE OF CONTROL FOR LOW-ARSENIC THROUGHPUT PRIMARY COPPER SMELTERS
Smel tec
ASARCO-E1 Paso
ASARCQ-Hayden
Tennessee Chemical
Co. - Copper-hill
Inspirafion-
Mi;ni
Kennecott-Garf ield
Kennecott-Hayden
Kennecott-Hurley
Emission
Source
CT
ST
MT
CONV
AF
FDH
ST
MT
CONV
AF
FDH
ST
MT
CONV
FDH
ST
MT
CONV
AF
FDH
ST
MT
CONV
AF
FDH
ST
MT
CONV
AF
FDH
ST
MT
CONV
AF
FDH
Arsenic Content
Present in
Calcine
Furnace Slag
Matte
Process Offgas
Anode Furnace
Transferred Dust
Furnace Slag
Matte
Process Offgas
Anode Furnace
Transferred dust
Furnace Slag
Matte
Process Offgas
Transferred Dust
Furnace Slag
Matte
Process Offgas
Anode Furnace
Transferred Dust
Furnace Slag
Matte
Process Offgas
Anode Furnace
Transferred Dust
Furnace Slag
Matte
Process Offgas
Anode Furnace
Transferred Dust
Furnace Slag
Matte
Process Offgas
Anode Furnace
Transferred Dust
Amount, kg/nr
146.5
54.0
48.1
76.2
12.1
260.0
99.0
97.0
45.0
20.0
194.0
0.4
0.7
0.5
0.6
14.1
5.4
1.5
0.7
1.9
33.0
15.7
6.0
13.1
189.5
2.2
7 9
5.0
0.09
4.1
0.8
A Q
0.4
0.16
1.6
Emission
Factor
(percent)
o'.i
1.5
15.0
11.0
0.1
0.1
1.5
15.0
11.0
0.1
0.1
1.5
15.0
0.1
0.1
1.5
15.0
11.0
0.1
0.1
1.5
15.0
11.0
0.1
0.1
1 5
15.0
11.0
0.1
0.1
1 5
15.0
11.0
0.1
Emission
Rate, Absence
of Control
kg/hr
u.o /
0.05
07
11.4
1.3
0.26
0.1
1.5
.8
2.2
.2
0.0004
0.01
0.08
0.0006
0.01
0.08
0.22
0.08
0.002
0.03
0.2
1 C
00
0.002
0.11
0.01
0.004
0.001
0.014
0.054
0.02
0.002
2-58
-------
Smelter
i^ennecott-McGil
Magma-San Manuel
Phelps Dodge-Ajo
Phelps Oodge-
Oouglas
Phelps Dodge-
Hi Idal go
•Phelps Dodge-
Mo renci
Copper Range -
White Pine
Emission
Source
MT
CONV
AF
FDH
ST
MT
CONV
AF
FDH
ST
MT
CONV
AF
FDH
CT
ST
MT
CONV
AF
FDH
ST
MT
CONV
AF
FDH
ST
MT
CONV
AF
FDH
ST
MT
CONV
AF
FDH
Arsenic Content
Present in
furnace Slag '
Matte
Process Offgas
Anode Furnace
Transferred Dust
Furnace Slag
Matte
Process Offgas
Anode Furnace
Transferred Dust
Furnace Slag
Matte
Process Offgas
Anode Furnace
Transferred Dust
Calcine
Furnace Slag
Matte
Process Offgas
Anode Furnace
Transferred Dust
Furnace Slag
Matte
Process Offgas
Anode Furnace
Transferred Dust
Furnace Slag
Matte
Process Offgas
Anode Furnace
Transferred Dust
Furnace Slag
Matte
Process Offgas
Anode Furnace
Transferred Dust
Amount, kg/hr
34.0
35.' 6
3.3
14.9
0.4
0 6
u • u
0.4
0.06
0.*7
10.1
4. 1
2.3
0.9
11.8
10.6
6.2
4 2
T • L.
3.2
0 ?
y • L.
1.9
3.0
1 "i
•L * J
0.9
0 fi
W »
-------
2.4 REFERENCES
1 Background Information Document for Review of New Source^Performance
Standards for Primary Copper Smelters (Draft). U.S. Environmental
Protection Agency. Research Triangle Park, North Carolina. EPA
Contract No. 68-02-3056. October 1982. p. 3-2.
2 Telecon. Whaley, 6., Pacific Environmental Services, with Butterman,
W., U.S. Bureau of Mines. Primary copper smelter production in
1982. March 2, 1983.
3. Reference 1, p. 3-3.
4 Field Surveillance and Enforcement Guide for Primary Metallurgical
Industries. U.S. Environmental Protection Agency. Research
Triangle Park, North Carolina. Publication No. EPA 450/3-73-002.
December 1973. p. 175.
5. Reference 4, p. 180.
6. Letter and attachments from J.W. Maksym, White Pine Copper Division,
Copper Range Company, to J.R. Farmer, U.S. Environmental Protection
Agency. March 17, 1983. Response to Section 114 Information
Request.
7 Letter and attachments from L.R. Judd, Phelps Dodge Corporation
to J.R. Farmer, U.S. Environmental Protection Agency. April /,
1983. Response to Section 114 Information Request.
8 Letter and attachment from R.A. Malone, Kennecott Minerals Company
' to J.R. Farmer, U.S. Environmental Protection Agency, March Ib,
1983. Response to Section 114 Information Request.
9 Letter and attachments from M.O. Varner, ASARCO, Inc. to J.R. Farmer,
U.S. Environmental Protection Agency, March 16, 1983. Response
to Section 114 Information Request.
10 Letter and attachment from J.W. George, Tennessee Chemical Company,
to J.R. Farmer, U.S. Environmental Protection Agency, April 6,
1983. Response to Section 114 Information Request.
11 Letter and attachment from J.H. Boyd, Magma Copper Company, to
" J R Farmer, U.S. Environmental Protection Agency, March 15,
1983. Response to Section 114 Information Request.
12. Letter and attachment from T.B. Larsen, Inspiration Consolidated
Copper Company, to J.R. Farmer, U.S. Environmental Protection
Agency. March 14, 1983. Response to Section 114 Information
Request.
13 Process Parameters for Primary Copper Smelters and Their Effects
on Arsenic Emissions (Draft). U.S. Environmental Protection
2-60
-------
n-rk, North
14. Reference 1, p. 3-11, 3-12.
15. Reference 1, p. 3-18.
16' [nsei?ationhC1ey' ^" Pac1fic Envi>°™ental Services, with
converters. March 4, 1983. a us o oboken
17. Reference 1, p. 3-36.
18' ??rP™t£;J't 51 aLJ!1!or Elements 1n the No™nda Process.
16-20? 1975!) 1Mth AnnUa1 AIME Meet1ng- New York- Feb^a^
of Copper. London. John
20. Correspondence from Mr. Jim Henderson, ASARCO, Incorporated, to
fir. U.R. Goodwin, Environmental Protection Agency, August 30,
21. Yazawa, A Thermodynamic Considerations of Copper Smeltina
Canadian Metallurgical Quarterly. .13(3):443. ^g^"16111"9'
22' DuHnn'rln a^ T1\Azakam^- Thermodynamics of Removing Impurities
8:257 ?Q6Q Smeltlng' Canadian Metallurgical Quarterly/
°ncence from K'W- Nelson' ASARCO, Incorporated, to J
, Environmental Protection Agency. March 17? 1976.
24. Harris, D.L. Air PolHuion Emission Test - Paniculate and
Arsenic Emission Measurements from a Copper Smeler Volume I
June 26-30 Ig3";? ReSe3rCh CorP°^tion- ™ Report No. 77-CUS-6
Contract No. 68-01-4136. January 1978.
27. Correspondence from "ckerlng, I.G., Vice President Environmental
cwn c ^ennecott. Copper Corporation, to D.R. Goodwin, Director
1 Pr°teCt10n A9en^' HW 9> »78r'nDaurftr0r>
2-61
-------
28. Stankovic, D. Air Pollute , Caused by Copper Metallurgy Assemblies
in Bor. Institute for Copper. Bor. Environmental Protection
Agency. No. 02-513-1.
29. Conversation with Mr. Hank Hansen, Environmental Engineer, Kennecott
Copper Corporation, Salt Lake City, September 26, 1978.
30. Ruddle, R.W. The Physical Chemistry of Copper-Smelting. Institute
of Mining and Metallurgy. London. 1953.
31. Turnbull, D.L., "Converter Practice at Mufulira," Seventh Commonwealth
Mining and Metallurgical Congress, 1961.
32. Rozlovskii, A.A., "Behavior of Arsenic in the Production of
Nonferrous Metals," Tsvetnye Metally/Nonferrous Metals, Number UDC
669.778.
33. Paulson, D.L., et al. Smelting of Arseniferous Copper Concentrate
in an Electric Arc Furnace. U.S. Bureau of Mines. Report of
Investigation No. 8144. 1976.
34. Reference 18.
35. Reference 18.
36. Reference 23.
37. Reference 24.
38. Schwitzgebel, K., et al. Trace Element Study at a Primary Copper
Smelter. Prepublication copy. U.S. Environmental Protection
Agency. Contract No. 68-01-4136. January 1978.
39 Harris, D.L. Air Pollution Emission Test - Particulate and
Arsenic Emission Measurements from a Copper Smelter. Volume I
(text). Monsanto Research Corporation. EMB Report No. 77-CUS-6.
June 20-30, 1977.
40. EPA testing at Anaconda (process emissions).
41. EPA testing at Kennecott-Hayden (process emissions).
42. TRW Environmental Engineering Division, Emission Testing of
ASARCO Copper Smelter, Tacoma, Washington. EMB Report No. 78-CUb-l
-------
45. Reference 42.
46. Reference 42.
47. Reference 43.
48. TRW Environmental Engineering Division. Emission Testing at
Pnelps Dodge Copper Smelter, Ajo Smelter. EMB Report No. 78-CUS-9
Februar
February 1979.
49.
1978
50.
Correspondence from Mr. K.W. Nelson, ASARCO, Incorporated, to
Goodwm, Environmental Protection Agency, September 14,
Correspondence from Mr. A.L. Labbe, ASARCO, Incorporated, regarding
process data for the ASARCO-Tacoma plant for the arsenic sampling
period September 12-24, 1978, to Mr. S.T. Cuffe, Environmental
Protection Agency. November 20, 1978.
51. Reference 48.
52. Reference 42.
53. Reference 50.
54. Reference 42.
55. Reference 43.
56. Godsey, E.S. Tacoma Plant Anode Furnace Emissions. ASARCO Salt
Lake City office. August 8, 1975.
C.L. Wilson. Metallurgy of Copper. London. John
58. Reference 42.
2-63
-------
3.0 CONTROL TECHNOLOGY
This chapter identifies alternative control techniques which can
be applied to process and fugitive emission sources at primary copper
smelters to control arsenic emissions. The amount of arsenic emission
reduction which can be achieved by the application of these control
techniques as well as major factors affecting their performance are
discussed.
3.1 ALTERNATIVE CONTROL TECHNIQUES
3.1.1 process Emission Control
3-1'1-1 General Considerations. As discussed in Section 2.0, much
of the arsenic entering the copper smelting process is volatilized and
eliminated as metallic oxide in the process offgas streams. This is a
result of the very high temperatures associated with the pyrometallurgical
processes used and the inherent volatility of arsenic and its prevalent
oxide, arsenic trioxide (As203). The extreme volatility of arsenic,
and especially arsenic trioxide, is the single most important factor
affecting the controllability of arsenic emissions from sources at
primary copper smelters.
Several studies have been made to determine the vapor pressure of
arsenic trioxide in air at various temperatures.1 Table 3-1 presents
vapor pressure data for arsenolite (As406), the more common form of
arsenic trioxide and the most abundant arsenic compound in smelter
offgases. ' Also presented are the arsenic concentrations at saturation
corresponding to the temperatures and vapor pressures listed. Figure 3-1
illustrates graphically the vapor pressure-temperature relationship
indicated by these data. This figure illustrates a significant logarithmic
increase in the vapor pressure of arsenic trioxide, and thus the
amount of arsenic which can exist in the vapor state, with temperature
3-1
-------
Table 3-1. SUMMARY OF As406 VAPOR PRESSURE DATA AND CORRESPONDING
ARSENIC CONCENTRATION AT VARIOUS TEMPERATURES
Temperature,
°C(°F)
457.2 (855.0)
412.2 (774.0)
332.5 (630.5)
299.2 (570.6)
259.7 (499.5)
212.5 (414.5)
200.0 (392.0)
175.0 (347.0)
150.0 (302.0)
125.0 (257.0)
110.0 (230.0)
100.0 (212.0)
90.0 (194.0)
Vapor pressure,
mm Hg
760
400
100
40
10
1.0
0.910
0.175
2.77 x 10~2
3.58 x 10"3
8.81 x 10"4
3.31 x 10"4
1.18 x 10"4
Arsenic concentration,
g/m3
5.0 x 103
2.81 x 103
7.94 x 102
3.36 x 102
90.20
9.90
9.25
1.88
0.315
0.043
0.011
0.0043
0.0016
3-2
-------
3.0-
I
w
z
o
S
O
Q_
0.01 L
100
1
,
5C
0.1
10
TEMPERATURE, *C
Figure 3-1. Arsenic Trioxide Vapor Pressure and Saturated
Vapor Concentration with Temperature
3-3
-------
Furthermore, the vapor pressure H;,ta indicate that arsenic trioxide
maintains an appreciable vapor pressure at relatively low temperatures.
For example, Table 3-1 shows that at 200°C (392°F), which is
typical of the outlet temperature of many of the control devices
currently used in the industry, 9.25 grams of arsenic trioxide per m
of gas (g/m3) (4.04 gr/ft3) could exist in the vapor form. Even at a
temperature as low as 125°C (257°F), a gas stream could contain 0.043 g/m
(0.019 gr/ft3) of arsenic trioxide as vapor. Any concentration higher
than 9.25 g/m3 in a gas stream at 200°C would lead to condensation of
some of the vapor, and the As405 would then exist in two phases: a
vapor or gaseous phase and a condensed or particulate phase. The
vapor would pass through a particulate control device without collection,
while the condensed material would be collected at about the same
efficiency as total particulate matter. The implication of these
facts with regard to the controllability of arsenic is important. The
temperature of the gas stream determines the amount of arsenic trioxide
which can exist as vapor and, consequently, the quantity of condensed
or particulate arsenic which can potentially be collected in a particulate
control device. A lower temperature gas stream is more likely to
contain more arsenic trioxide in a form which can be collected in such
a device.
An example will serve to illustrate these points. A 2,834 m /min
(100,000 acfm) gas stream at 150°C (302°F) with an arsenic trioxide
mass flow rate of 40 kg/hr would have an arsenic concentration of
0.235 g/m3 (0.103 gr/ft3). Table 3-1 shows that at 150°C, the saturation
concentration for As40g is 0.315 g/m3. Therefore, the stream is at a
subsaturation level and no condensation, and hence no particulate
collection, is predicted. If this stream were cooled to 125°C in a
spray chamber, the arsenic concentration at the new flow rate of
2,667 m3/min (94,100 acfm) would increase to 0.250 g/m (0.109 gr/ft ).
Since the arsenic saturation concentration shown in Table 3-1 for this
temperature is 0.043 g/m3, the amount of arsenic leaving the spray
chamber in condensed form would be 0.207 g/m3 (33.12 kg/hr), and the
remaining amount, 0.043 g/m3, would remain in vapor form. If this
stream was then sent to a baghouse collector with a control efficiency
for total particulate matter of 96 percent, a total of 0.96 x
3-4
-------
33.12 kg/hr = 31.80 kg/hr would * collected. The overall control
efficiency for arsenic at the lowered temperature would thus be
31.80/40 = 8U percent. From Table 3-1 it can be seen that as the
temperature is further lowered below 125°C, the saturation concentration
becomes very small. In this case, virtually all of the arsenic trioxide
present in the stream would condense into a collectible form and the
collection efficiency for arsenic would approach the particulate
efficiency of 96 percent.
Tests conducted by EPA across a hot electrostatic precipitator
(ESP) which controlled particulate emissions from a "green" charge
reverberatory smelting furnace further demonstrate this phenomenon.4
The electrostatic precipitator was operated at 315°C (600°F) or higher.
Measurements for particulate matter conducted using in-stack filters, '
at the operating temperature of the ESP, demonstrated an overall
collection efficiency for particulate matter of about 97 percent. In
contrast, arsenic measurements conducted using a modified EPA Method 5
sampling train operated at 121°C (250°F) indicated an average arsenic
collection efficiency for the ESP of less than 30 percent (see Table 3-2).
Calculations based on the arsenic concentration measured at the hot
ESP inlet and the saturation concentration data presented in Table 3-1
indicate that the application of a 97 percent effective ESP to the
same reverberatory furnace process gas stream which had been cooled to
110-C (230°F) could result in an overall arsenic collection efficiency
of 90 percent. These measurements and calculations suggest that,
while the subject ESP was reasonably effective in removing material
which existed as particulate matter at its operating temperature, any
material such as arsenic trioxide which would be predicted to exist in
the vapor state at the elevated temperature at which the ESP was
operated passed through the ESP with little removal. This demonstrates
the need to cool the gas stream to be treated to the extent practicable
to condense as much of the arsenic trioxide vapor as possible prior to
its entering a control device for collection.
As a result, the alternative control techniques for arsenic
process sources considered herein include precooling as an integral
part of the overall control system. These include the application of
baghouses, high voltage electrostatic precipitators, or high energy
3-5
-------
Table 3-2. ARSENIC DATA FOR HOT ESP
Sample
run
1
2
3
Avg.
Temp. ,
°C (°F)
.,.-. . - — • —
328 (622)
317 (602)
336 (639)
327 (621)
Inlet
Emissions
g/Nm3(gr/scf) kg/hr (Ib/hr)
__ — —
0.268 (0.117) 26.8 (59.1)
0.325 (0.142) 33.0 (72.7)
0.334 (0.146) 34.1 (75.2)
0.309 (0.135) 31.3 (69.0)
i einp. »
°C (°F)
313 (595)
321 (610)
304 (580)
313 (595)
Outlet
Emissions
g/Nm3 (gr/scf) kg/hr (Ib/hr)
0.211 (0.092) 24.3 (53.6)
0.181 (0.079) 20.2 (44.6)
0.199 (0.087) 23.2 (51.1)
0.197 (0.086) 22.6 (49.8)
— _ —
Efficiency,
percent
9.3
38.6
32.0
27.8
I
en
-------
venturi scrubbers, preceded by g .s stream cooling. In addition, these
include contact sulfuric acid plants, with their extensive gas precleaning
and conditioning systems, applied to process sources which generate
offgases containing high concentrations of S02 (greater than 3.5 percent).
Before proceeding with the discussions on control techniques, it
should be noted that the 30 percent arsenic collection efficiency
recorded for the hot ESP cannot be predicted using the vapor pressure
data for arsenic trioxide. In fact, given the operating temperature
of the ESP (315°C), no arsenic recovery would be predicted because the
arsenic concentration in the gas stream was several orders of magnitude
lower than that needed to achieve saturation. This suggests (1) that
other arsenic compounds less volatile than arsenic trioxide may be
present in the gas stream, and/or (2) that condensation, although the
principal mechanism in situations where more arsenic trioxide is
present than is needed to saturate the gas stream, is not the only
mechanism by which arsenic trioxide can be collected.
3.1.1.2 Gas Cooling. Methods commonly applied in the nonferrous
metals industries for cooling hot gas streams include radiative cooling,
evaporative cooling, and cooling by dilution with ambient air. The
dilution method consists of introducing sufficient quantities of
ambient air into the hot gas stream so that the resultant gas mixture
is at the desired temperature. The ambient air required may be introduced
by infiltration or forced draft. Evaporative cooling uses water to
cool hot gases in spray or quench chambers. Water is injected into
the hot gas stream where the heat contained in the gases vaporizes the
injected water, resulting in a temperature reduction. Radiative
cooling relies on heat loss due to natural convection and radiation to
effect cooling. These losses occur whenever a temperature gradient
exists between gases inside a duct and the surrounding air. Cooling
of gases by this method requires only that a sufficient heat transfer
area be available to obtain the desired cooling. All three methods
have advantages and disadvantages.
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
3-7
-------
to be treated, the amount of dil -non 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 draft fan and the control device required.
Due to the need for sufficient heat transfer area, radiative
cooling requires considerable space. Depending on the temperature of
the gas stream to be cooled, the heat transfer areas required could
exceed 1.6 m2/m3 (500 ft2/1000 ft3) of gas treated. Normally used in
the lead industry on zinc fuming operations, radiative cooling devices
consist of a series of 10 to 20 U-shaped tubes, each about 3 m (10 ft)
in diameter and 20 m (66 ft) in height. In addition, fan horsepower
requirements increase due to the increased resistance to gas flow
resulting from the added ductwork. The major drawback to radiative
cooling, however, is its limited flexibility for temperature control.
Cooling hot gases by evaporative cooling is relatively simple and
requires 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 entering electrostatic precipitators or baghouses for
particulate removal. Typically, the spray chambers have a cross-sectional
area of about 35 m2 (375 ft2) and are 30 to 60 m (100 to 200 ft) in
length. The large cross-sectional area results in a low flow velocity
ai,d a relatively long residence time. Water is introduced through a
series of sprays along the cross section of the chamber. Water require-
ments vary depending on the temperature of the stream to be cooled and
the desired end temperature.
The major difficulty in applying water spray chambers for cooling
smelter offgases is the potential for corrosion. The amount of cooling
achievable is limited by the dew point of the treated gas stream. As
the gases are cooled, water and sulfuric acid mist contained in the
gas stream may condense, creating a corrosion problem in the flue
system, control device, and stack. While the problem of corrosion
attack is potentially severe, it can be negated by the use of appropriate
construction materials. Acid resistant cement, stainless steel,
nickel, and. chromium alloys have been used in several cases. No
significant corrosion problems have been reported at primary copper
3-8
-------
and lead smelters where spray cb-,,oers are used for cooling prior to
electrostatic precipitators and baghouses used for particulate matter
control.
3*1'1-3 Ba3ho"ses (Fabric Filters). Baghouse particulate collectors
have historically achieved collection efficiencies in excess of 99 percent,
over a broad range of applications. Although extensively used in the
primary lead and zinc industries for the collection of particulate and
metallurgical fume, their application at primary copper smelters has
been limited. ASARCO-Tacoma is the only smelter which currently uses a
baghouse for the control of particulate matter contained in smelter
offgases.
In principle, particles contained in the treated gas stream are
initially captured and retained on the fibers of the fabric by a
number of mechanisms, including direct interception, inertial impaction,
diffusion, gravitational settling, and electrostatic attraction. Once
a mat or cake of dust is formed on the fabric, further collection is
achieved by simple sieving. Periodically, the fabric is cleaned by
mechanical or other means to allow for disposal of the collected
particulate and to maintain the pressure drop across the filter within
practical operating limits.
The filtration area required at a specified pressure drop is
dependent on the gas volume treated, particulate loading, permeability
of the fabric used, resistance properties of the particulate deposited,
and the cleaning mechanism used. The pressure drop commonly found in
baghouses applied in the nonferrous metals industries ranges from 0 5
to 2.0 kPa (2 to 8 inches H20).5 The filtering velocity, or air-to-cloth
ratio, generally ranges from 0.30 to 0.61 m3/min per m2 (1 to 2 acfm/ft2)
for conventional mechanical shaker cleaning type baghouses when applied
to metallurgical fume. Pulse jet cleaning type units generally operate
at higher air-to-cloth ratios, ranging from 1.8 to 3.0 m3/min per m2
(6 to 10 acfm/ft^).5
Although there is a variety of woven and felted fabrics available
woven Orion and Dacron bags should be suitable in smelter applications.'
Both exhibit good resistance to acid attack and may be operated at
temperatures ranging up to 135°C (275°F) with no significant deterioration 5
3-9
-------
Bag life, which varies considerably with operating conditions, should
be approximately 1 to 3 years. Closed pressure or closed suction
baghouse designs may be used in the applications considered. The
baghouse design selected depends on the acid dew point and the consequent
corrosion potential of the smelter offgases being treated.
3.1.1.4 Electrostatic Precipitators (ESP's). Single-stage
electrostatic precipitators are widely used in the primary copper
industry for the control of particulate emissions from smelting facilities.
Electrostatic precipitators use electrical forces for the removal of
suspended particulates in a gas stream. The process encompasses three
basic functions: the charging of particles, the collection of charged
particles, and the removal of the collected particles. Particles
suspended in the gas stream are charged while passing through a high
voltage, direct current corona established between a discharge electrode,
usually a small diameter wire which is maintained at high voltage, and
a grounded collecting surface (collecting electrode). As the particles
pass through the corona, they are bombarded by negative ions emanating
from the discharge electrode, and charged within a fraction of a second.
The charged particles, influenced by electric field forces, migrate
toward the grounded collecting surface where they are deposited and
held by electrical, mechanical, and molecular forces. Particulate
metter adhering to the collecting surface is periodically dislodged by
mechanical rappers or by flushing with water. The material is collected
in a hopper and periodically removed for disposal or recycle.
Two principal types of precipitator design are available, the
wire-and-plate and wire-in-tube types. In the more common plate-type
precipitators, the collecting surface consists of parallel vertical
plates spaced 15 to 30 cm apart with wire or rod discharge electrodes
suspended vertically between the plates. The plates are typically 4
to 12 m in height and 4 to 7 m in length.6 Plate-type ESP's are
generally applied directly to process sources for the control of dry
particulate matter at elevated temperatures, usually 200 to 340°C (400
to 650°F). The collecting surface in tube-type precipitators consists
of a cylinder with the discharge electrode centered along its longitudinal
axis. This type of ESP is used exclusively for acid mist elimination
3-10
-------
and fine particulate removal pn>, to acid manufacturing. With either
type, a complete preclpltator installation consists of several individual
subunits positioned both in series and in parallel to achieve the
desired collection efficiency.
There are several important factors to be considered in the
design and sizing of a preclpltator to obtain a desired efficiency.
These include the volume, temperature, and moisture content of the gas
stream to be treated, and the resistivity, size distribution, and
loading of the particulate to be collected. Of these, resistivity is
the most important. Resistivity (the reciprocal of conductivity) is
dependent on the chemical and physical properties of the particulate,
and the temperature of the stream. Because of the presence of metal
oxides, the resistivity of smelter dusts is relatively high at reduced
temperatures (90 to 200°C) in the absence of natural conditioning
agents such as moisture and S03. Too high a resistivity (greater than
10 ohm-cm) may result in excessive sparking, which can seriously
limit precipitator performance. Preconditioning with moisture and
sulfuric acid has been applied to decrease particle resistivity prior
to precipitation.
3-1-1-5 Venturi Scrubbers. In a venturi scrubber, flue gases
are passed through a venturi at high velocity, and the scrubbing
liquid is introduced at the venturi throat under low static pressure
where it is atomized and accelerated by the gas stream. Particles
suspended in the gas stream are removed by impaction with the atomized
liquid droplets. The wetted particles and entrained liquid are removed
by a cyclone or other type of entrapment separator. The collection
efficiency achieved, which can be in excess of 99 percent, is directly
related to the total amount of energy expended in forcing the gases
through the venturi and in atomizing and accelerating the scrubbing
liquid. The expended energy is reflected in the pressure drop across
the device, which may range from 2.5 kPa (10 in. H20) to over 20 kPa
(80 in. H20). Typical throat velocities for venturi units range from
75 to 100 m/s (15,000 to 20,000 fpm) and 1iquid-to-gas ratios range
from 0.4 to 2 liters/min per m3/min (3 to 15 gpm/103 acfm).8 It is
3-11
-------
important that the liquid-to-gas :utio be sufficiently high to guarantee
complete wetting across the venturi cross-section.
The application of venturi scrubbers at primary copper smelters
is limited to a tew smelters where they are used to augment gas stream
precleaning systems and to provide additional gas cooling prior to
acid manufacturing.
3.1.1.6 Sulfuric Acid Plants. As noted previously, smelter
offgases containing high concentrations of S02 (over 3.5 percent) are
generally treated in single- or double-contact sulfuric acid plants
for S0? removal. The presence of solid and gaseous contaminants, such
as acid mist, entrained dust, and metal fumes in the treated smelter
offgases can present serious difficulties in the operation of an acid
plant. The major difficultias caused by these contaminants include
the corrosion of heat exchanger tubes, plugging of catalytic beds,
deactivation of the catalyst, and contamination of the product acid.
As a result, rather extensive measures have to be taken to remove
contaminants to ensure that their concentrations are reduced to tolerable
levels prior to entering the acid plant.
Table 3-3 contains estimates of the maximum levels of impurities
that can be tolerated in smelter offgases used for sulfuric acid
manufacturing. The degree of catalyst deterioration experienced at
these various impurity levels can be tolerated by shutting down the
acid plant once per year to screen the catalyst and repair the equipment.
Table 3-3 also contains the estimated upper level of impurities that
can be removed by typical gas precleaning systems with prior removal of
coarse dust. Although complete removal of contaminants, such as
arsenic, from the offgases is not practical, over 99 percent removal
is considered achievable.
Both hot and cold gas cleaning devices are used. Generally, the
offgases are initially treated in a hot electrostatic precipitator
where the coarse particulate, which contains large amounts of metals,
is 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
3-12
-------
Table 3-3. ESTIMATED APPROXIMATE MAXIMUM IMPURITY LIMITS FOR
METALLURGICAL OFFGASES USEDJO MANUFACTURE
SULFURIC ACIDy
T . . Impurity limits at
c . . Impurity limits at inlet to gas puri-
Substance inlet to acid plant, fication system,
(mg/Um )3 (mg/Nm3)a'b
Chlorides, as Cl i<2 125c
Fluorides, as F o.25 25d
Arsenic, as As203 i.2e 200
Lead, as Pb Ie2 2QQ
Mercury, as Hg 0.25 2.5f
Selenium, as Se soe 100
Total solids 1>2 , Q00g
H2S04 mist, as 100% acid 50
Water 400 x 103
: dry offgas stream containing 7 percent sulfur dioxide.
typical gas purification system with prior coarse dust removal.
be reduced to 6 mg/Nm3 if stainless steel is used.
are Sp&^cI'rtJ" ^Ke'Uce"^^*^ SSTltt"
Can be objectionable in product acid.
5'°°° t0 10'°0° mg/Nm if weak acid settll'"9 tanks
3-13
-------
mist precipitators where acid nr.t, fine participates, 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.
The basic steps in the contact process for the manufacture of
sulfuric acid from sulfur dioxide-bearing gases are shown in Figure 3-2.
As noted, the offgases are cooled and cleaned to remove particulates
and volatile metals. Acid mist is removed in an electrostatic mist
precipitator, and the gases are dried with 93 percent sulfuric acid.
The dry gases then pass through a series of gas-to-gas heat exchangers
to heat the offgases to the optimum temperatures for the catalytic
conversion of sulfur dioxide (S02) to sulfur trioxide (S03). Single-contact
acid plants use three or four stages of catalytic converters, whereas
dual-contact plants use one, two, or three stages of catalyst before
the first absorption tower. Since the conversion of sulfur dioxide to
sulfur trioxide is exothermic, the converter outlet gases must be
cooled before passing through the absorption tower. These outlet
gases are passed countercurrent to the inlet gases in the heat exchangers
mentioned above. The sulfur trioxide is then absorbed by 98 percent
sulfuric acid in an absorption tower to yield the product. In a
single-contact acid plant, the remaining gases are then treated to
remove acid mist and spray, and then vented to the atmosphere.
In a dual-contact acid plant, the gases exhausted by the first
absorption tower are passed through a second series of heat exchangers
and catalytic converter stages to oxidize the sulfur dioxide remaining
in the gases. Normally, this step employs one or two stages of catalyst.
The gases then pass through a second absorption tower, where sulfur
trioxide is absorbed by sulfuric acid as in the first absorption
tower. The waste gases are then treated to remove acid mist and
spray, and vented to the atmosphere.
3.1.1.7 Factors Affecting Performance. The alternative arsenic
emission control techniques considered above share a common element.
They all consist of cooling the gas stream to condense arsenic (provided
a sufficient quantity of arsenic is present in the gas stream) and
collecting the resultant fume in a high efficiency collection device.
3-14
-------
GAS CLEANING
ACID PRODUCTION
S02 BEARING GAS
I
ELECTROSTATIC
PRECIPITATOR
OR BAGHOUSE
to
I
COOLING
AND
SCRUBBING
FACILITIES
WEAK ACID
AND
SOLIDS
DUST
ELECTRO
STATIC
MIST
PRECIP
ITATOR
TO ATMOSPHERE TO ATMOSPHERE
4
DRYING
TOWER
1
1
1
L_
L-*
HEAT
EX-
THANfiFfK
•*- -*•
•«- -*-
•« — >•
CONVERTER
FIRST
ABSORPTION
TOWER
SECOND
ABSORPTION
TOWER
SINGLE CONTACT
DOUBLE CONTACT
93% ACID
98% ACID
Figure 3-2. Contact Sulfuric Acid Plant
-------
Regardless of the control device o.iployed, the amount of achievable
arsenic emission reduction is dependent on four major factors. These
include the concentration of arsenic trioxide present in the gas
stream, the temperature to which the gas stream is cooled prior to
collection, the allocation of sufficient time between cooling and
collection to allow condensable arsenic compounds to condense, and the
overall collection efficiency of the control device.
As discussed in Section 3.1.1.1, because of the volatile nature
of arsenic (As203), the arsenic inlet concentration and the operating
temperature of the control device are critical factors in determining
its potential effectiveness in controlling the emission of arsenic.
The temperature of the gas stream to be treated determines the maximum
amount of arsenic present in the gas stream which can exist in the
vapor state and, consequently, the quantity of arsenic which can exist
in the solid state and be collected in a particulate control device.
As a result, the gas stream to be treated must be cooled to the extent
practicable and adequate time must be allowed to ensure that most of
the condensable fraction of the arsenic present is condensed prior to
entering the control device for collection.
Although cooling is basic to effective arsenic control, the
presence of moisture and sulfur oxides in the smelter offgases introduces
a lower temperature constraint below which further cooling cannot be
tolerated without incurring severe operating problems. Because S03 is
hygroscopic, it will absorb moisture at temperatures well above the
moisture dew point and form highly corrosive sulfuric acid mist. The
temperature at which acid mist formation occurs (acid dew point) is
highly variable, depending on the S03 concentration and other gas
stream characteristics. Continued 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 to
25°C (20 to 45°F) above the best estimate of the acid dew point.
Little data are available on the acid dew point of smelter offgases.
However, practical experience at several smelters, including the
three ASARCO smelters and the Anaconda smelter prior to its closing,
3-16
-------
indicates that operating temperatures in the range of 100°C (21?°F) to
HO°C(230°F) are within tolerable limits.
The quantity or concentration of arsenic in the gas stream is
very important in determining achievable arsenic emission reductions.
To achieve any arsenic trioxide emission reduction by condensation,
the quantity of arsenic trioxide in the gas stream must be sufficiently
high so that the resultant arsenic trioxide concentration at the
control device operating temperature exceeds the predicted saturation
concentration. If the arsenic trioxide concentration at the control
device operating temperature does not exceed the predicted saturation
concentration, little or no emission reduction is achievable. Conversely
if the arsenic trioxide concentration greatly exceeds the predicted
saturate concentration, arsenic emission reductions approaching the
overall performance capability of the control device for paniculate
matter can be achieved.
The effect of the overall collection efficiency of the control
device on achievable arsenic emission reduction is self-evident; the
higher the efficiency for total particulate matter, the higher the
efficiency for arsenic.
3<1'2 furtive Emission Sources and Controls
Fugitive emissions may be characterized as emissions which escape
erectly from the process area to the atmosphere rather than through a
flue or exhaust system. They result from leakage in and around process
equipment and from material handling and transfer operations.
Fugitive emissions from these sources are controlled by local
ventilation (i.e., use of localized hoods or enclosures) or general
ventilation techniques (i.e., building evacuation) to confine and
capture emissions. Once captured, the emissions may be vented directly
to a particulate control device or combined with process offgases
prior to collection in a control device. As with process sources
some consideration must be given to cooling prior to collection 'in
most instances, however, cooling occurs naturally as a result of air
dilution due to infiltration and, therefore, additional equipment is
not required.
Besides the use of add-on controls, fugitive emissions from some
sources may be minimized or eliminated by minor process changes and
3-17
-------
good operating and maintenance .ractices. A general discussion of
these and other control alternatives is presented in the following
subsections.
3.1.2.1. Local Ventilation. Local ventilation consists of using
localized hoods or enclosures to confine the fugitive emissions at the
source, and the use of induced air currents to entrain and capture the
fugitive emissions and divert them into an exhaust opening.
The design of a local exhaust hood involves specifying its shape
and dimensions, its position relative to the emission point, and its
rate of air exhaust. The rate of exhaust is dependent on the air
velocity required and on the size of the imaginary curving area of the
hood. Air contaminants originating within this area are drawn directly
into the exhaust opening. Methods for estimating the surface area of
an exhaust hood can be found in standard references dealing with
industrial ventilation. The capture velocity is the velocity of the
air at the hood face or entry plane necessary to overcome opposing air
currents and to capture the emission-laden air by causing it to flow
11 12
into the exhaust hood. '
Hoods can generally be classified into three broad groups: enclosures,
receiving hoods, and exterior hoods. Enclosures usually surround most
of the point of emission, though sometimes one side may be partially
oi- even completely open. Receiving hoods are those wherein the air
contaminants are injected into the hoods. For example, the hood for a
grinder is designed to be in the path of the high velocity dust particles.
Exterior hoods must capture air contaminants that are generated from a
point outside the hood itself, sometimes some distance away. A canopy
hood is a good example of an exterior hood.
Exterior hoods are the most commonly used hoods and are by far
the most difficult to design since they are the most sensitive to
external conditions. For example, a hood that works well in a still
atmosphere may be rendered completely ineffective by even a slight
draft through the area.
For exhaust hoods to be effective, sufficient ventilation must be
applied across the space between the source and the hood so that all
emissions are entrained. This involves overcoming the cross currents
of indoor air which could deflect the stream of fugitive emissions
3-18
-------
away from the hood. The ventilet.on rate for exhaust hoods applied to
hot sources must also take into account the thermal draft that results
from heat transfer from the source to the surrounding air. The exhaust
rate should be as uniform as possible over the entire plane of the
hood inlet. Various exhaust hood configurations are illustrated in
Figure 3-3.
So-called "air curtains" can be used as complementary capture
devices for local ventilation. Air curtains are basically air jets of
suitable geometric configuration with sufficient momentum to resist
the forces working against them and maintain their continuity across
the opening they protect. For ventilation purposes, they are used as
part of push-pull type systems, wherein the air jet or curtain is
blown across the emission zone, forcing the emissions into an exhaust
hood located opposite the air curtain slot. Typical examples of air
curtain applications are shown in Figure 3-4. As indicated in Figure 3-4(a),
hazardous fumes, e.g., vinyl chloride, can be contained by the use of
air curtains. Dust control by air curtains is shown in Figures 3-4(b)
and 3-4(c). A detailed discussion of air curtains is presented in
Section 3.1.2.7.2.
The following typical fugitive emission sources at copper smelters
can be controlled by local ventilation methods:
• Calcine transfer from roaster
• Matte tapping
• Slag tapping
• Converter
3>1'2'2 General Ventilation. This technique is required whenever
it is not possible or expedient to use local exhaust hoods because of
their handicapping operations, maintenance, or surveillance of the
process or equipment; or when the local exhaust does not significantly
reduce air requirements or has no worker exposure advantages.
General ventilation has historically taken the form of either
natural air changes due to wind and density differences, or mechanically
assisted air changes. Natural changes of air through a building in
the absence of mechanical ventilation occur by the action of either of
two forces, the play of wind through windows or other openings or the
3-19
-------
a)
c)
b)
\u\\\r
HOOD
FLOOR
SOURCE
FLOOR
e)
FAN (HOOD)
FLOOR OR
SOURCE BENCH
" A TOP »
\\ \
SOURCE
Figure 3-3. Types of Exhaust Hoods
12
NOTE- In all cases, the source of contamination is beyond
the boundaries of the hood structure; hence, control
action is effected by inducing velocities in the
adjacent space.
3-20
-------
AIR CURTAIN
(a)
AIR CURTAIN
MATERIAL FEED
(b)
DUST
COLLECTOR
^B-^C^^h^ -^ _ wt_ff«^»aa»
fT£B&l.r2_
CT
(c)
NOTE:
TO DUST
COLLECTOR
Figure 3-4. Uses of Air Curtains13'14
3-21
-------
buoyancy action resulting from a Difference in temperature between
outdoors and the inside of the building. Mechanical ventilation is
induced by motor-driven fans and is used when the contaminants cannot
be removed by natural ventilation. With mechanical ventilation the
contaminants can be forced out of the building by roof evacuation.
Mechanical ventilation as applied to copper smelters is discussed in
greater detail in Section 3.1.2.7.
Ventilation requirements for a building are generally defined in
terms of total building air changes per unit time. Although essential
in determining the ventilation requirements, the air change rate, also
referred to as the ventilation rate, is not the only factor which is
important. It is also important to consider the rate of generation of
emissions in the building.
The main factors on which successful control of general ventilation
depends are the layout and siting of the sources of heat, the configuration
of the building (number of spans, form and shape of the roof), and the
arrangement of ventilation openings in walls and roof bays. The most
satisfactory solutions are obtained when the architect and the engineer
collaborate and consider the problems of natural ventilation at the
design stage of the facility.
3.1.2.3 Collection Devices for Fugitive Sources. As discussed
in Section 3.1.1.1, the most important factor affecting the controllability
of arsenic emissions is the temperature of the uncontrolled offgas
stream. Fortunately, fugitive offgas streams are generally lower in
temperature than process streams and seldom have a temperature higher
than 93°C (200°F). (Refer to fugitive emission test results in Appendix C.)
Cooling of the fugitive offgases occurs as a result of air dilution
due to ambient mixing and/or infiltration. Also, cooling is automatically
achieved as a result of radiative cooling during the passage of the
fugitive offgases in the usually long ductwork leading to the control
device. Therefore, additional gas cooling prior to collection is not
required for fugitive emissions.
The captured fugitive emissions are usually exhausted into existing
process control systems. Only two domestic copper smelters use collection
devices exclusively for the control of fugitive emissions. ASARCO-E1 Paso
3-22
-------
uses a baghouse to control fugi^ve emissions from the converter
building evacuation system. Phelps Dodge-Douglas uses a baghouse to
collect fugitive emissions generated from the roaster calcine discharge
system. Both baghouses were tested by EPA to evaluate their performance.
Results of the testing at ASARCO-E1 Paso are presented in Section 3.3.2.3,
and in Tables C-25 through C-28 of Appendix C. Results of the Phelps
Dodge-Douglas testing are presented in Tables C-53 through C-56 of
Appendix C.
Although baghouses and electrostatic precipitators are currently
used by smelters to collect fugitive emissions, scrubbers could also
be used, but high operating costs and water handling problems make
their use less desirable.
3.1.2.4 Calcine Transfer from Roaster. The multiple-hearth and
fluidized-bed roaster are the two basic types of copper concentrate
roasters used. In the case of the multi-hearth, the calcine hopper
located at the bottom of the roaster, which drops the roaster calcine
into the transfer vehicle (larry car), is a source of fugitive emissions.
The use of close fitting hooding between the car and hopper outlet is
difficult because of the necessity of moving the car to and from the
roaster discharge hopper. Fluidized-bed roaster material transfer is
carried out primarily by air conveying and secondarily by material
dropout; transfer is usually well contained as long as the equipment
is maintained in good condition, resulting in negligible fugitive
emissions.
A schematic of the calcine transfer fugitive emission control
system at ASARCO-Hayden, which is essentially identical to the one at
ASARCO-Tacoma, is presented in Figure 3-5. A continuous flat apron
strip nearly 0.6 m (2 ft) wide is mounted directly below the row of
multi-hearth roasters at the hopper exits. Below each roaster, in the
apron, there are two ports connected to a 0.5 m (1.5 ft) wide duct. A
matching leaf spring-loaded flat apron is mounted on the larry car,
which is driven directly below the roaster and in line with the matching
holes on the apron connected to the roaster hopper. One hole is used
for transferring the calcine from the roaster to the larry car. The
other holes are connected to vent lines which go to vent hoods with
their own individual draft fans (a single fan is used at ASARCO-Tacoma).
3-23
-------
CO
I
ro
HOPPER
NOTE: CAR TOP AND HOOD - 18 GA. C.R.S.
FRONT ELEV.
SIDE ELEV.
Figure 3-5. Spring-Loaded Car Top and Ventilation Hood, ASARCO-Hayden
-------
Lach draft fan has a capacity of -nproximately 142 Nm3/min (5,000 scfm)
The captured fugitives are then combined with the roaster process
yases and treated in a baghouse.
In addition to the Ioca1 hooding and ventilation applied directly
at the calcine hopper discharge point, at ASARCO-Tacoma, the calcine
hopper area has been enclosed to for, a tunnel-like structure which is
ventilated. The ventilated enclosure, coupled with the local hooding
and ventilation applied at the actual calcine discharge point, is very
effective in capturing fugitive emissions during calcine transfer
operates. During visual observations, no fugitive emissions were
observed escaping from the tunnel-like enclosure.15
3-1.2.5 Matte Tapping. Reverberatory furnaces may have up to
four matte tap holes, two on each side of the furnace. Matte is
tapped from one hole at a time and conveyed through troughs or launders
into 4.9 to 9.2 m3 (175 to 324 cubic feet) ladles. Typically, fewer
than 30 taps per furnace are made each day, with each tapping operation
taking 5 to 10 minutes. In electric furnaces there are generally four
•natte tap ports. Normally, only one matte tap port is in use at a
t 1 m/-v
time.
Copper matte from the furnace-port travels through a launder
wh!ch dlrects the flowing matte to a point where it can be collected
m a large ladle. Emissions are observable from the point of the
matte leavlng the furnace to the point where it settles in the ladle
Matte tap ports and launders in most smelting furnaces have
hood.ng systems. Tap port exhaust hoods may be of any shape, as long
as they are designed to be close to and cover as much of the emission
area as possible. Launder hoods generally consist of covers mounted
on the launder in sections to allow manual removal for launder cleaning
Schemata of a matte tapping fugitive emission control system
used at the ASARCO-Tacoma smelter are presented in Figures 3-6 and
3-7. The matte tap hood has a 1.2 by 1.2 m (4 by 4 ft) square
cross-sect10n and is located less than 0.9 m (3 ft) above the tap
hole The ducts connecting the matte tap exhaust hoods to the 1 2 m
4 ft) dimeter main duct are approximately 0.6 m (2 ft) in diameter
During a tap, 283 Nm3/nnn (10,000 scfm) is exhausted at the tap hole'
3-25
-------
OJ
I
ro
TO BAGHOUSE
« 1T-0 DIA.-LADLE HOOD
(MOVABLE)
40'0 x 3/16 THICK
MATTE TAP
HOOD
50 0 x 3/16"
REVERB.
FURNACE
« 35'0
Figure 3-6. Matte Tapping Fugitive Control System (Plan View),
ASARCO-Tacoma
-------
CjO
I
CABLE TO WINCH
RETRACTABLE
HOOD
1/4
THICK
LAUNDER
I MATTE LADLE '
I I
\
3'-8"
MATTE TAP HOOD
Figure 3-7. Matte Tapping and Ladle Hoods
-------
hood and 850 Nm3/min (30,000 scf-;. is exhausted at the ladle. The
covered launder is 3 to 6 m (10 to 20 ft) long, 0.6 m (2 ft) wide, and
approximately 0.3 m (1 ft) deep. The launder covers (Figure 3-8)
r.imir in si/e from 0.(> to 1.1. m (?. to 1» ft) in U'n<)t,h ami h.we a
same i rcular cross-section.
A 3.4 m (11 ft) diameter retractable ladle hood is used to capture
eriissions generated at the ladle. The ladle hood is lowered over the
ladle prior to tapping and is raised after the tap is completed.
Fran testing and observations made at the ASARCO-Tacoma smelter,
the matte tapping fugitive emission capture system was observed to
achieve greater than 90 percent capture efficiency.
3.1.2.6 Slag Tapping. Generally, reverberatory furnaces have
just one slag tap hole. Typically, slag is tapped an average of
30 times per day for approximately 10 to 20 minutes per tap. In
electric furnaces there are usually two slag tap ports with one slag
tap port in use at a time. The duration of a slag tap is normally
10 minutes. The slag flows from the tap port, down an inclined launder,
and into one or more slag pots. The slag pots/ladles at various
smelters range in capacity from 2.8 to 17 m3 (100 to 600 cubic feet).
Fugitive emission capture techniques for slag tapping operations
are very similar to those used for matte tapping. Local exhaust hoods
arc used over slag tap port areas. Slag launders are either partially
or completely covered. Design volumes for exhaust hoods vary from one
smelter to another, ranging from 566 to 850 Nm3/>nin (20,000 to 30,000 scfm).
A schematic of the slag tapping fugitive emission control system
used at the ASARCO-Tacoma smelter is shown in Figure 3-9. Slag tap
hoods are pyramidical in shape, having a 1.2 m by 2.4 m (4 ft by 8 ft)
rectangular cross-section. They are less than 0.9 m (3 feet) above
the tap hole. A larger exhaust hood, 2.4 m by 4.3 m (8 ft by 14 ft),
is situated above the slag pot transfer point. Each launder is covered
with fixed hoods. During tapping, 142 Nm3/nin (5,000 scfm) is applied
at the tap hole and 566 Nm3/min (20,000 scfm) is applied at the slag
pot. Emissions along the launder run are vented to either of the
above hoods. As with matte tapping, a 90 percent capture efficiency
should be achievable.
3-28
-------
OJ
I
r-o
vo
5'-0
(SECTION)
Figure 3-8. Launder Cover
-------
OJ
I
oo
o
TO BAGHOUSE
I-
REVERB.
FURNACE
•35'-0-
7'-7
40V x 3/16"
65'-0
NOTE:
1. ALL DUCT
2. HOODS
3/16" THICK C.S
3/16" THICK C.S
Figure 3-9. Slag Tapping Fugitive Control System (Plan View),
ASARCO-Tacoma
-------
3'1'2'7 Converter Operatic,. Primary converter hoods capture
process emissions during converter blowing periods, except for some
emissions that escape due to primary hood leakage. However, during
converter charging, skimming, or pouring, the mouth of the converter
is no longer under the fixed primary hood, and extensive quantities of
fugitive emissions escape capture by the primary hood.
Another source of fugitive emissions during converting occurs
during the converter holding mode. This occurs under normal smelting
operations when material, either slag or blister, cannot be immediately
transferred from the converters to the ladles. During this period the
generated emissions are not evacuated by the primary hood.
There are currently three basic alternative control techniques
used to capture fugitive emissions during converting. They are:
(1) secondary mechanical hoods; (2) air curtain secondary hoods; and
(3) general ventilation/building evacuation.
3.1.2.7.1. Secondary mechanical hoods, in normal practice,
primary converter hood systems are used only when the converters are
in the blowing mode. At some smelters, secondary mechanical hoods are
being used to capture emissions generated from the converter mouth
during the other nodes, such as charging, skimming, holding, and
pouring The flow rates handled by these hoods range from 700 to
2,400 Mm /min (25,000 to 85,000 scfm).
There are three major types of secondary mechanical hoods:
1. Fixed type - Attached to the primary or uptake hood, and
currently used at Phelps Dodge-Ajo, Phelps Dodge-Hidalgo, Phelps
Dodge-Morenci, and Kennecott-Utah.
2. Swing-away type - This type is being used at the Saganoseki
smelter in Japan and consists of a swing-away type hood used as a
deflector and a retractable type secondary hood just above it.
3. Mechanical type - This is a combination hood system utilizing
a fixed hood, movable hood, gate hood, and swing-away hood. This
system is a conceptual design system for converter fugitive emissions
control.
Domestic copper smelters are using the fixed type of secondary
hooding to control fugitive emissions from converters. A typical
3-31
-------
converter secondary hood configu-jJon of the fixed type is shown in
Figure 3-10. These hoods are approximately 3 in (10 ft) long, 6.4 m
(21 ft) wide, 1.7 m (5.5 ft) high and are affixed to the upper front
side of the converter primary uptake hoods. Testing and observations
made at existing copper smelters utilizing fixed converter hoods
indicated that their effectiveness was only marginal.
The swing-away type converter hood, sometimes called a deflector
converter hood with a retractable secondary hood, is used at Japan's
Saganoseki smelter. The hood was observed to be very effective during
the blowing, slagging, and pouring of blister copper. During
charging of matte or rabbling, capture of the secondary emissions
depended solely on the retractable hood portion of the device.
The converter mechanical secondary hood system shown in Figure 3-11
19
is a conceptual design developed by an EPA consultant. This system
uses a fixed hood in combination with a movable hood, a gate hood, and
a swing hood. The elliptical fixed hood is made of steel and is
attached to the primary uptake hood of the converter. Its opening is
situated in a manner to avoid the hook of the overhead crane during
converter operations. The upper end of the fixed hood is attached to
the smoke plenum, which leads to dust bins on each side of the converter.
The elliptical movable hood (refer to Figure 3-11) is made of
stsel and fits over the fixed hood. It has its own track of movement.
In the retracted position, this hood does not extend further, horizontally,
than the fixed hood. In the extended position, it mates with the lip
of the fixed hood to provide continuity of ducting for secondary
emissions.
The elliptical gate hood fits under the fixed hood in the retracted
position. In the extended position, the gate hood continues ducting
of the secondary emissions to the movable and fixed hood.
The frustum-shaped swing hood can be rotated 180 degrees. The
width at the top is the same as that of the mouth of the gate hood.
This hood is made of steel with a castable refractory lining. Its
pillar-mounting is motorized. In the retracted position it is clear
of the aisle and slightly behind the converter.
Retrofitting the mechanical secondary hood system to an existing
smelter could pose a difficult problem. The operational safeguards
3-32
-------
TO SECONDARY 4
HOODING
MAIN DUCT
T.O. RAIL
SMOKE
HOOD
PLENUM
FIXED
SECONDARY
HOOD
Figure 3-10. Typical Converter Fixed Secondary
Hood
3-33
-------
SECONDARY HOOD DUCT
SMOKE PLENUM
SECONDARY HOOD DUCT
SECONDARY HOOD DUST BIN
DUST BIN
MAIN HOOD
SECONDARY
HOOD DUCT
MOVABLE HOOD
EOT RUNWAY
SECONDARY HOOD
DUST BIN
""MOTORIZED DRIVE
SWING HOOD DURING
TAPPING OR SLAGGING
POSITION
LADLE
Figure 3-11.
Conceptual Design for Converter Mechanical
Secondary Hood System
3-34
-------
that would be required to operat* it with a minimum number of breakdowns,
delays, and damage would require close supervision and extensive
effort. On the other hand, the maintenance and operation of the
relatively inefficient fixed converter secondary hood system currently
being used at domestic smelters and the swing-away converter secondary
hood system, as used in Japan, are quite simple.
The efficiency of fixed converter secondary hoods is generally
dependent upon the distance of the hood from the emission source and
on the capture and face velocity created by the fan at the mouth of
the fixed hood. Conversely, maintenance and durability are enhanced
as the hood is moved farther from the converter mouth. The capture
effectiveness of the fixed converter secondary hoods applied at the
Phelps Dodge-Ajo and Phelps Dodge-Hidalgo smelters was judged to be
low.
The swing-away type of secondary converter hood again causes
clearance problems with the crane hook and cables during collar pulling
or matte additions. Also, rugged drive mechanisms are needed for the
operation of the swing-away hood and added time is needed to complete
pouring and skimming operations due to the time involved in the retraction
of the swing-away hood to allow crane access to the ladle. However,
good fugitive emission control is obtained during pouring, blowing,'
and slagging. During matte addition or rabbling, the capture efficiency
is similar to that of a fixed hood.
Since the converter mechanical secondary hood system combines a
fixed and a swing-away hood in combination with a movable hood it
should be more effective than either a fixed hood or swing-away hood
used alone. The capture efficiency of the converter mechanical secondary
hood system during charging, pouring, skimming, and blowing is estimated
to be 40, 85, 70, and 85 percent, respectively.20 Assuming a typical
converter operation to consist of 80 percent blowing, 15 percent
charging, and 5 percent skimming and pouring, an overall average
capture efficiency for the mechanical secondary hood system is calculated
to be approximately 80 percent.
3'1'2-7'2 A1r curtain secondary hoods. Another method of controlling
fugitive emissions from copper smelter converting operations involves
3-35
-------
the use of an air curtain system along with a secondary hood system.
An air curtain secondary hood capture system has been installed on a
domestic primary copper smelter (see below), and such systems are
being used abroad14'21'22'23 and in other U.S. industries.
As discussed in Section 3.1.2.1, an air curtain is a suitably
shaped air jet with sufficient momentum to resist the forces of fugitive
gas streams working against it and to maintain its continuity across
the opening it protects. Figure 3-12 shows a schematic diagram of an
air curtain system controlling converter fugitive emissions. Consideration
in air curtain design must be given to secondary or entrained flows
which start forming as the air curtain jet stream leaves its slot or
nozzle. As the entrained flows become fully mixed with the air curtain
jet stream some distance from the nozzle, the hot or cold secondary
flows are carried from one side of the air curtain jet stream to the
other where they are ducted for suitable discharge. The greater the
entrained flow, the greater the energy loss of the jet. To minimize
energy loss, a relatively thick, slow moving jet stream with a large
air volume is required. A basic rule in the design of air curtains is
to use the thickest and lowest velocity air stream to be projected
across the shortest dimension of the opening. Air curtain design
methods are discussed in references 14, 25, and 26.
The air curtain system being used at the Onahama and Naoshima
primary copper smelters in Japan is shown in Figure 3-13. The capture/
shielding device includes two steel plate partitions, one on each side
of the converter. The air jet is blown from a slot at the top of one
of the plates across the opening to provide a sheet or curtain of air
that prevents fugitive emissions from escaping. The other plate is
equipped with an exhaust hood. The opening allows the crane cables to
move into position above the converter mouth.
A propeller fan is used to push the air through an elongated slot
on one side and a backward inclined fan provides suction on the opposite
side to pull in both the fugitive gases and the push air. Captured
gas passes through steel ductwork to a baghouse. The combined temperature
of the converter fugitive gases with 100 percent of the push air
3-36
-------
JET SIDE
AIR
CURTAIN
JET
<
••M^M
••«•*••
BAFFLE
WALL
FUG IT
X EMISS
'f '
\
V
EXHAUST SIDE
Alk CURTAIN
/
xx ^;
CONVERTER
(FUME SOURCE)
LADLE
BAFFLE
WALL
TO SUCTION FAN
Figure 3-12. Converter Air Curtain Control System
3-37
-------
OJ
I
OJ
CO
6.25
CONVERTER
Figure 3-13. Converter Air Curtain Secondary Hood, Onahama and
Naoshima Smelters
-------
entering the ductwork on the puP >ide of the air curtain is approximately
80°C (180°F), which makes gas cooling prior to the control device
unnecessary.
The inlet air forming the air curtain above the converters at the
Naoshima smelter has a flow rate of approximately 600 Nm3/min (21,000 scfm).
The exhaust hood on the opposite side pulls in approximately 1,000 Nm3/min
(35,000 scfm) of gas to the main system. The capacity of the total
pull system at this smelter is three times this value, or 3,000 Nm3/min
(105,000 scfm), to accommodate the operation of three hoods at a time.
According to Naoshima authorities, the collection efficiency of these
hoods for fugitive emissions is approximately 90 percent.27
The Tamano copper smelter in Japan uses a differently designed
air curtain system along witi, a fixed hood, which is essentially a
total enclosure, for controlling fugitive emissions from each of its
three converters (usually one converter is operated at a time.) A
sketch of the air curtain system installed at the Tamano smelter is
shown in Figure 3-14. The enclosure has two front doors and a
movable roof which is slightly inclined toward the front. The air
curtain duct (slot hood) is located at the top of the enclosure level
at a position to push air from one side of the converter to the other.
Ambient air is supplied by a ground fan rated at 1,2000 Nm3/min (41,000 scfm),
Two ventilation points (offtakes) are located at the inside wall of
the enclosure for capture of fugitive emissions; one is for dilute SO
gas and the other is for high concentration S02 gas. The inlet of the
ductwork for high SO,, gas is located near the converter mouth at a
level below the other ductwork for dilute S02 fugitive gas. For
convenience, the inlet of this ductwork is shown at the top of the
roof in Figure 3-14.
Subjective evaluation of the air curtain secondary hood by visible
observation at the Tamano smelter indicates the system to be at least
90 percent effective in controlling fugitive emissions.22
ASARCO's Tacoma facility recently installed a prototype air
curtain secondary hooding system to control fugitive emissions from
their No. 4 converter. In this system, walls are erected to enclose
the sides and the back of the area around the converter mouth, with a
3-39
-------
Roof Opening
for Fugitive Gas
Fugitive Gases
to Bag House
at toiler
>vabl« Boof
Air
Fugitive Gas"
to" _
Desulfurization
Plant
Off-gases To
Acid Plant
Hz Curtain
Converter Furnace
front Door
Figure 3-14. Air Curtain System at the Tamano Smelter
3-40
-------
portion of the enclosure back we1, formed by the primary hood. Openings
at the top and in the front of the enclosure allow for movement of the
ladle and the overhead crane cables and block. The edges of the walls
which contact the prinary hood and the converter vessel are sealed to
contain the emission plume.
When the converter is rolled out for charging or skimming, the
yate on the primary hood is moved up and away from the converter mouth
to provide clearance for the overhead crane and ladle. As noted
previously, significant amounts of dust-laden fumes (fugitive emissions)
escape the primary hood system during these modes of the converter
cycle. Heavy fugitive emissions are generated during the roll-out and
roll-in modes, when charging (including cold additions), slag skimming,
and blister copper pouring are taking place. The heaviest emissions
occur during the actual rolling out of the converter because the
injection of blowing air continues during roll-out until the molten
bath is below the tuyeres to prevent plugging by cooled material
During these periods of operation, the air curtain is utilized to
capture most of the emissions which would otherwise escape the primary
system.
Air volume control for the system is regulated automatically by
dampers in the air curtain jet, the exhaust duct, and the induced
draft fan. The dampers are manually set for a predetermined exhaust-side
flow and, when placed in the automatic control position, are activated
by movement of the primary hood and converter. When the primary hood
is lifted and the converter is rolled out, the system switches to a
high flow mode to control the heavy fugitive emissions generated
during roll-out activities. At the completion of the converter roll-out
operations, the converter is rolled in and the primary hood is lowered
over the converter mouth. At this point, the system switches to a
lower flow volume which is maintained during blowing and holding
periods.
The gases that are captured by the air curtain system are treated
m an ESP for particulate removal before being passed to the atmosphere
through the main stack.
Design data for the ASARCO system are summarized in Table 3-4.
3-41
-------
Table 3-4. SUMMARY OF DESIGN DATA FOR THE ASARCO-TACOHA CONVERTER
SECONDARY HOODING/AIR CURTAIN SYSTEM28
Mode Air Curtain Push Rate, Main Offtake Evacuation
of operation Actual m3/min Rate, Actual m3/min
Matte charging 510 (18,000 acfm) 2,322 (82,000 acfm)
Blowing -a L700 (60>000 acfffl)
Slag skimming 510 (18,000 acfm) 2,322 (82,000 acfm)
Holding 510 (18,000 acfm) 850 (30,000 acfm)
Worst conditions6 1,020 (36,000 acfm) 4,644 (164,000 acfm)
aAir curtain will not be used during the blowing mode.
bWorst conditions would consist of either (1) two converters being
charged simultaneously or (2) one converter being charged while
another was being skimmed.
3-42
-------
EPA conducted a program to evaluate the capture effectiveness of
the ASARCO-Tacoma air curtain secondary hood system in January 1983 2-
This program included arsenic and total particulate sampling, a gas
tracer study to determine the system's capture efficiency during
specific modes of converter operation, and observations of visible
emissions to assess the system's effectiveness. The results of this
program are presented in Section 3.3.2.2.
3A'2'7'3 Biding evacuation. As noted previously, ventilation
requirements for a building evacuation system are generally defined in
terms of air changes per unit time. The rate of air change method
estimates are based simply on room volume and do not consider the rate
of evolution of the contaminant, the number of heat sources, or the
natural draft due to building configuration. For example, a general
ventilation installation designed by the rate of air change method
can, under some conditions, actually cause the contaminant to be
spread throughout the building, thus increasing the volume of dilution
air required to maintain hygienic conditions. This occurs when the
distribution of the ventilation air supply is poorly controlled.
Uncontrolled air flowing into a building, due to negative pressure in
the building or because of poorly designed air supply distributors
may not only cause recirculation of the contaminant, but also upset
the local ventilation systems. It is, therefore, important that the
amount of air, the location of its entry into the building, and its
direction be controlled. For example, Figure 3-15 shows a controlled
air supply which results in a convective flow from a heat source (such
as a ladle of molten metal) rising to be exhausted through a roof
ventilator. Figure 3-16 shows an uncontrolled air supply which results
in a disrupted rising plume and recirculation of the contaminant
throughout the building.
Natural air changes take place when hot air from the ground level
heat sources rises due to its buoyancy. If, however, there exists a
point within the building where the temperature of the surrounding air
is equal to that of the rising column of hot air, buoyancy is lost
Therefore, natural air changes will take place only if the temperature
of the rising column of hot air is high enough to maintain the buoyancy
3-43
-------
Figure 3-15. Controlled Airflow from a Heated Source
30
Figure 3-16. Uncontrolled Airflow from a Heated Source
30
3-44
-------
of the column until it is dlscha-jed through the roof monitors. In
most hot metal workshops this is, however, not the case. Hot air
pools of some depth are formed under the building roofs. As a result
air entering the building will at times mix turbulently with pools of'
contaminated air and transport it downward to the occupied levels near
the floor.
The concept of controlled ventilation is being employed at the
ASARCO-E1 Paso smelter to capture emissions from the converter aisle
Several modifications (area isolation, vent location, air flow control,
etc.) were made in order to implement this control measure at the
facility. The present building evacuation rate is 16,800 Nm3/min
(bOO.OOO scfm). This corresponds to an air change rate of 18 changes
per hour. Although it is impossible to quantify, the building evacuation
system at El Paso, when properly operated and maintained, is believed
to be capable of achieving 95 percent capture. Particulate emissions
contained in the building ventilation gases are controlled by a baghouse
EPA has performed tests on this baghouse; the results are provided in
Section 3.3.2.3, and detailed test summaries are presented in Appendix C
Although the building evacuation system at ASARCO-E1 Paso should be
capable of achieving 95 percent capture, the capture effectiveness
actually being achieved is substantially less. Severe operating
problems with this system have been encountered, which have led to
unacceptable buildups of arsenic, lead, and heat in the building 31
Worker exposure to airborne arsenic and lead has continuously exceeded
the concentration limits set by the Occupational Safety and Health
Administration (OSHA) (10 tfg/m3 for arsenic, 50 ^g/m3 for lead) In
an attempt to alleviate the situation, the company has increased the
openings in the building and are presently operating roof ventilators
whlch discharge directly to the atmosphere. The roof ventilators were
originally installed for emergency use only in the event of a power
failure. The company is currently investigating the use of local
ventilation methods (including converter secondary hoods) for use at
its El Paso facility.
3.1.2.8 Anode Furnace. Anodes are fire refined directly from
molten blister copper at all domestic copper snelters. Fire refining
3-45
-------
is carried out in rotary-type reining furnaces resembling Pierce-Smith
converters or in small hearth furnaces. The rotary-type furnace is
used at 14 of the 15 existing smelters. ASARCO-Tacoma is the only
domestic smelter which employs hearth-type anode furnaces.
Fugitive arsenic emissions from anode furnaces contain little
arsenic since most of the arsenic has been eliminated in earlier
processing steps.
ASARCO-Tacoma is the only domestic copper smelter capturing
arsenic emissions from anode furnaces. Emissions from the hearth-type
anode furnaces at ASARCO-Tacoma are siphoned off and conveyed to an
ESP for collection.
A swing-up type hood is used for the control of fugitive emissions
from the rotary-type anode furnaces used in Japan. A schematic of
such a hood is presented in Figure 3-17. This hood has a flexible
duct approximately 0.7 m (2.25 ft) in diameter connected to an arc-shaped
hood. The hood is approximately 4 m (13 ft) wide and 3 m (10 ft)
long. The quantity of dilution air required to cool the gases is a
function of the position of the hood above the anode furnace mouth.
For an anode furnace temperature of 982°C (1,000°F), the quantity of
dilution air required to reduce the temperature to 121°C (250°F) will
be 198 Nm3/min (7,000 scfm). The collection system is designed for
42S Nm3/min (15,000 scfm) for one anode furnace under these conditions.
The emissions captured by the swing-up hood occur during the
oxidizing and reducing blows. The emissions collected during these
blows constitute the majority of the emissions generated. Some emissions
escape during the charging of blister from the ladle and during the
pouring of refined copper from the anode furnace, since the mouth of
the furnace during these operations is not under the hood system;
these emissions, however, are comparatively small.
3.1.2.9 Dust Transfer. Handling, and Conveying. Dust transfer,
handling, and conveying practices vary from smelter to smelter.
However, the dust transfer from control devices and smelter flues is
common practice at all smelters.
Dust collected by an electrostatic precipitator drops into a
storage hopper beneath the unit. A collecting conveyor, contained in
dust-tight housing, transfers this dust to a storage bin from which
3-46
-------
PNEUMATIC CYLINDER AUTOMATIC HOOD OPENER
AIR LINES
HOOD 1/4" C.S.
HOOD
FLEXIBLE DUCT 28"-0
Figure 3-17. Anode Furnace Movable Hood
3-47
-------
dust may be withdrawn as desired. These bins are usually equipped
with dust level indicators. Discharge from the storage hopper is
usually through a dust-tight connection to surface transportation
units. Normal practice is to distribute the dust on smelter feed beds
where it is sprayed with water and covered by damp concentrates or
precipitates and other furnace charge materials.
Dust collected by a baghouse drops into a holding hopper during
the shaking cycle and is removed through a double gravity gate or
rotary valve and into an enclosed collector conveyor (generally of the
screw type), which in turn discharges into a closed storage bin.
Depending on the nature of the collected dust and its metallurgical
values, it is either conveyed to the charge mixing area or is discarded.
In scrubbers, the dust is removed as a slurry, which is then
dewatered in a thickener or filter and/or passed to a settling pond.
Offgases from roasters, smelting furnaces, converters, and other
pyrometallurgical units are conveyed through air-tight flues and
chambers to their ultimate dispersal point. The coarser dusts tend to
settle in the flues and chambers. All flues and chambers are equipped
with hoppers spaced at intervals underneath. These hoppers provide
surge capacity and storage for the settled dust until it is withdrawn.
Previously, these hoppers were equipped with plain swing gates for
discharge onto the ground, into railroad cars, or to other surface
conveyance equipment. Naturally, there was some dust loss by this
method. Presently, these hoppers are equipped with discharge gates
which usually automatically feed into screw or drag type collector
conveyors for storage in a central receiving bin. The dust is then
conveyed to receiving points. Flue dust, however, is a coarser material
and may be dumped directly onto wet feed belts or conveyed to a desired
location by pneumatic conveyors. Waste heat boilers and crossover
flues present special problems because the dust from the smelting
furnace is at high temperature and builds up in fusions and large
accretions. These must be removed by manual means, water-bomb lances,
soot blowers, fluxing, sand blasting, slugger guns, or a combination
of methods. Due to the various sizes of materials dropping into the
boiler ash pits, this material is handled and transported by conventional
surface vehicle conveyance and is usually passed through a smelter
flux crushing system.
3-48
-------
control IT'0 ""^ ^ ^^ ^^ <*<»"»"** of alternative
control techniques for arsenic emissions and total particulate, EPA
conducted emission measurements and made visual observations at several
smelters including the ASARCO-Tacoma smelter. This section discusses
the results of these emission measurements and presents conclusions
regard, ng the performance capability of each of the control techniques
evaluated for process and fugitive emission sources.
3-2-l Process Control Systems
3.2.1.1^ Baahouses,. Tests were performed at the ASARCO-Tacoma
and Anaconda smelters to evaluate the performance of baghouses in
controlling arsenic emission,, Tests were also performed at the
naconda smelter to evaluate the performance of the baghouse in controlling
total particulate emissions. Two baghouses were sampled at ASARCO-Tacoma
and one at Anaconda.
3'2-1-1-1 fo3"ouses (ASARCO-Tacomal. Simultaneous inlet and
outlet arsenic emission measurements were performed by EPA across the
baghouse serving the multi-hearth roasters at ASARCO-Tacoma
at the in,et was performed in the duct carrying emissions
fro™ the
,ng process in four 10 cm (4 in.) ports on top of the
osera
-------
Table 3-5. ARSENIC PERFORMANCE DATA FOR THE
ROASTER BAGHOUSE AT ASARCO-TACOMA
Sample
run
1
2
3
Avg.
—
Ft
Arsenic measurements
Inlet
°C (°F)
94 (201)
85 (185)
95 (203)
91 (197)
mg/Nm
314.3
295.8
254.2
288.1
(gr/dscf)
(0.138)
(0.130)
(0.111)
(0.126)
kg/hr
92.7
88.1
79.5
86.8
(Ib/hr)
(204)
(194)
(175)
(191)
Outlet
°C(°F)
88 (191)
87 (189)
82 (180)
86 (187)
mg/Nm
0.6
0.7
1.5
0.9
(gr/dscf)
(0.0003)
(0.0003)
(0.0007)
(0.0004)
kg/hr
0.2
0.2
0.5
0.3
(Ib/hr)
(0.4)
(0.4)
(1.0)
(0.6)
Efficiency,
percent
99.8
99.8
99.4
99.7
i i- \
Concentration and mass rate data are based on measurements on the total catch (front and back half).
OJ
I
en
o
-------
include six Godfrey roasters, ar-nic kitchens, and a metallic arsenic
plant. The baghouse, which has been replaced with a new baghouse,
consisted of five compartments, each having 288 homopolymer acrylic
bags. The bags measured 320 cm (126 in.) in length and approximately
13 cm (5 in.) in diameter. The total filtering area was 1,860 m2
(20,000 ft ). It was designed to effectively treat 850 Nm3/min (30,000 scfm)
at an air-to-cloth ratio of 0.63 m3/min per m2 (2.06 cfm/ft2). Bag
cleaning was performed by mechanical shakers. The outlet sampling was
done approximately 150 m (500 ft) downstream of the baghouse. The
test results obtained are summarized in Table 3-6. 33
As indicated, the average arsenic inlet concentrations and corresponding
mass rate at the baghouse were 2,941 mg/Nm3 (1.28 gr/dscf) and 76.3 kg/hr
168 Ib/hr), respectively. The outlet arsenic loading was 60.6 mg/Nm3
(0.026 gr/dscf), and the mass rate was 3.3 kg/hr (7 Ib/hr). The
average arsenic removal efficiency for this unit as indicated by these
results was 95.7 percent.
chamber/baqhouse (Anar.nnriaV Inlet and
. n oue
measurements were made at the spray chamber/baghouse separately for
arsenic and total particulate emissions. The sampling locations
included the two inlet plenums to the spray chamber and the baghouse
exhaust duct. Three samples each for arsenic and particulate were
obtained at the inlet and outlet. The results of the arsenic emission
measurements and particulate emission measurements are summarized in
Tables 3-7 and 3-8, respectively.34
When the tests were conducted, the smelter had the following
process emission control configuration. Smelting facilities at the
Anaconda smelter consisted of a fluid-bed roaster, a single electric
furnace, and six converters. Except for a portion of the electric
furnace and converter offgases [about 2,070 Nm3/min (73,000 scfm)]
which were diverted to an acid plant for S02 removal, process gases
from all three major smelting operations were combined into a main
flue duct and transported to a baghouse system for particulate removal
prior to being discharged to the atmosphere through the main stack
3-51
-------
OJ
I
en
ro
Table 3-6. ARSENIC PERFORMANCE DATA FOR THE ARSENIC PLANT BAGHOUSE AT ASARCO-TACOMA
Sample
run
1
2b
3
Avg.
Arsenic Emissions
Tnlpf
_ — ' ~~
Arsenic kitchen
°C mg/Nm3 kg/hr
—
56 1,717 42.4
58 1,518 42.8
60 1,532 45.1
58 1,589 43.4
Metallic arsenic
°C mg/Nm3 kg/hr
117 1,931 42.6
112 1 0.04
97 2,125 55.9
109 1,352 32.9
Total
• — — — — —
mg/Nm3 kg/hr
^ _-- — , • — • — — ~
3,648 85.0
1,519 42.8
3,657 101.0
2,941 76.3
Outlet
°C mg/Nm3 kg/hr
— —
83 69.8 3.5
72 15.8 1.0
71 96.1 5.5
76 60.6 3.3
_ — _
Efficiency,
percent
..
95.9
97.7
94.5
95.7
'Concentration and mass rate data are uaseu Ui. „««„«.«.—-
Wing this sample run the metallic arsenic process may not have been operating.
-------
Table 3-7. ARSENIC PERFORMANCE DATA FOR SPRAY CHAMBER/BAGHOUSE
AT THE ANACONDA-ANACONDA SMELTER
San-
run
•— i. .,••.
pie —
°C mg/Nm3
1 274 1,071
2 269 895
3 244 687
Avg. 263 885
1 — — — • •
catch (front and back
kg/hr
276.7
236.8
186.1
232.3
uu u i e L
°C
99
102
101
101
mg/Nm
—
7.1
9.8
12.6
9.8
kg/hr
1.8
2.5
3.4
2.6
Efficiency,
percent
99.3
98.9
98.2
98.9
>Shr?f? data dre baSed °n measurements °n the total
Table 3-8. PARTICULATE PERFORMANCE DATA FOR SPRAY
AT THE ANACONDA-ANACONDA SMELTER
.—.._..., ...
Sample
run
1
2
3
Avg.
5
i"" - ,n -i — , __
CHAMBER/BAGHOUSE
Particulate Emissions
Inlet
°C g/Nm3
— ....
281 14.76
288 13.57
302 14.08
290 14.14
Concentration and mass rat
cyclone, and filter catch
kg/hr
- -„. ,
4,071
3,736
3,860
3,890
Outlet
°C
103
103
101
102
(front half}
g/Nm3
0.05
0.04
0.05
0.05
measuremer
kg/hr
14.6
10.0
14.7
13.1
Efficiency,
percent
99.6
99.7
99 6
99.7
its on the probe,
3-53
-------
The baghouse system consists of
-------
or pipe design consisting of 18 Actions with a total collection area
of about 6,619 m (71,250 ft2). Each section contains 84 pipes measuri,
30 cm (12 inches) in diameter and 4.6 m (15 feet) in length. The
second unit is of a plate-type design, consisting of seven parallel
chambers each with four fields in series and with a total collecti,
area of 7,710 r/ (82,992 ft2). The exiting gases, about 7,740 actual
m /mln (270,000 acfm) at 90°C (200°F), are discharged through a large
flue to the smelter main stack.
Since the configuration of the inlet duct was such that it was
not possible to sample, only outlet arsenic emission measurements were
made by EPA at the electrostatic precipitator. The outlet sampling
was performed in the duct 23 m (75 ft) downstream of the main stack.
Three sample runs were made. A summary of the results obtained at the
outlet is presented in Table 3-9.35
Table 3-9. ARSENIC EMISSIONS AT OUTLET OF REVERBERATORY
FURNACE ELECTROSTATIC PRECIPITATOR AT ASARCO-TACoS
•
Sample
run
-' "•• i • _ .—
1
2
3
Avg.
- " •• - i
""•""' ' -— - .
®T ( ®E\
*-> \ r j
• • -.
105 (220)
101 (214)
87 (188)
97 (207)
Arsenic Emissions9
mg/Nm3 (gr/dscf)
38.1 (0.016)
21.0 (0.009)
9.6 (0.004)
22.9 (0.010)
— __
kg/hr (Ib/hr)
28.7 (63.1)
11.7 (25.8)
7.1 (15.6)
17.3 (38.2)
based °n
As indicated, the average arsenic concentration and mass rate at
the outlet were 22.9 mg/Nm3 (0.01 gr/dscf) and 17.3 kg/hr (38.2 Ib/hr),
respectively.
3'2'1'2*2 sPray Chamber/Electrostatic precipitator (ASARCO-E1 Pa
Inlet and outlet arsenic emission measurements were made by EPA across
this spray chamber/ESP. Three runs were made at each of three inlet
3-55
-------
locations and at one outlet location. A summary of the test results is
*3 -j
presented in Table 3-10. Measurements were also made for total
particulate emissions on two separate occasions. Three inlet locations
and one outlet location were sampled. During the first test, two runs
were made at each location. During the later test, three runs were
made only at the outlet. Tables 3-11 and 3-12 present the test data. >
Smelting facilities at the ASARCO-E1 Paso smelter consist of four
multi-hearth roasters, a single reverberatory furnace, and three
converters. When the tests were conducted, the smelter had the following
process emission control configuration. Process gases from the rever-
beratory furnace passed through two parallel waste heat boilers where
they were cooled to a temperature of about 400°C (750°F). The existing
gas stream was then combined with the roaster offgases in a large
rectangular flue. The combined gas stream, which averaged about
5,100 Nm3/min (180,000 scfm), then entered a spray chamber where it
was cooled from about 220°C (428°F) to less than 115°C (240°F). After
cooling, the combined roaster and reverberatory gases entered an
electrostatic precipitator for particulate removal prior to being
discharged into the main stack.* The precipitator consists of seven
parallel chambers, each containing four sections, with a total electric
field volume of 516 m3 (18,228 ft3).
As indicated in Table 3-10, the average arsenic concentration
recorded at the inlet and outlet was 0.308 and 0.006 g/Nm (0.13 and
0.0026 gr/dscf), respectively. The average mass rates were 95.9 kg/hr
(211 Ib/hr) at the inlet and 2.1 kg/hr (4.6 Ib/hr) at the outlet. The
results indicate that the average arsenic removal efficiency of this
unit was 97.8 percent for the three runs conducted.
Table 3-11 presents the particulate matter test results obtained
during the first test series conducted on the spray chamber/ESP at
ASARCO-E1 Paso. The average particulate concentration and mass rate
recorded at the inlet were 5.11 g/Nm3 (2.23 gr/dscf) and 1,134 kg/hr
(2,495 Ib/hr), respectively. The average particulate concentration
and mass rate at the outlet were, respectively, 0.098 g/Nm3 (0.043 gr/dscf)
*In current practice, reverberatory furnace gases are treated separately
in one ESP while roaster gases are treated in another ESP.
3-56
-------
Table 3-10. ARSLNIC PERFORMANCE DATA FOR SPRAY CHAMBER/
ELECTROSTATIC PRECIPITATOR AT ASARCO-EL PASO
— — -
Sample
run
™
1
2
3
Avg.
Nm /min
-" ™ • - - — • —
5636
5983
3467
5028
Concentration and mass
back half).
In!
°C
194
191
252
212
rate
et
g/Nm3
— - — -"-• -
0.100
0.338
0.594
0.308
data are
Arseni
kg/hr
33.8
123.1
123.8
95.9
c emissions3
•"' •— ' -i- — — .
Nm /min
4054
4233
4490
4259
based on measurements
°C
102
104
105
104
Outlet
—
g/Nm3
0.005
0.009
0.004
0.006
on the total
-
kg/hr
1.5
3.4
1.5
2.1
Efficiency,
percent
95.6
97.2
98.8
97.8
catch (front and
en
—i
-------
I
CJ1
00
Table 3-11 PARTICIPATE PERFORMANCE DATA FOR THE SPRAY CHAMBER/
ELECTROSTATIC PRECIPITATOR AT ASARCO-EL PASO
Sample
run
1
2
Avg.
Particulate Emissions
Inlet
°C
226
231
229
Concentration
filter catch
Mm /min
3,741
3,631
3,686
and mass rate
(front half).
g/Nm3
— — ~
4.54
4.87
5.11
data
kg/hr
1,022
1,248
1,134
°C
105
105
105
Nm /min
4,346
4,700
4,523
are based on measurements on
Outlet
g/Nnr
0.111
0.085
0.098
the probe,
kg/hr
40.8
33.8
37.2
Efficiency,
percent
96.1
97.3
96.7
cyclone, and
-------
— —
Particulate Emissions'
Concentration and mass rate data are based on
rtlPa<;iiy-i-t-c- n^ •!•!._ i , *"'
- - _ v . vl . i v i i «, ^ VII I* I I ^
catch (front half).
3-59
-------
and 37.2 kg/hr (81.8 Ib/hr). The average particulate removal efficiency
measured for the two sample runs performed was 96.7 percent.
Table 3-12 presents particulate matter test results obtained at
the outlet of the spray chamber/ESP during the second test series.
The average outlet particulate concentration and mass rate values
recorded during these latter tests were 0.15 g/Nnf5 (0.066 gr/scf) and
50.9 kg/hr (112 Ib/hr), respectively; this is about one and one-half
times higher than the values recorded during the earlier tests. The
higher concentrations and mass rates recorded in the latter tests were
likely due to the higher smelter production rates during those tests.
3.2.1.3 Venturi Scrubbers (Kennecott-Ha.yden). Arsenic emission
measurements were conducted by EPA at the Kennecott-Hayden smelter to
evaluate the performance of the venturi scrubber used in series with a
spray chamber equipped with impingement plates to preclean and condition
fluid-bed roaster offgases prior to acid manufacturing. The roaster
process gases pass through a series of primary and secondary cyclones
where an estimated 95 percent of the entrained calcine is recovered
and the gas stream is cooled to about 315°C (600°F). The cyclone
exhaust, consisting of about 565 Nm3/min (20,000 scfm) with an estimated
dust loading of 57 g/Nm3 (25 gr/scf), then enters the venturi scrubber
where most of the particulate matter is collected. Weak acid scrubbing
liquor is injected into the venturi throat at a rate of about 1,457 liter/min
(385 ypm), resulting in a pressure drop of about 41 cm (16 in.) of
water across the throat. Gases exiting the scrubber then enter a
spray-type scrubber equipped with perforated plates, where they are
humidified and cooled to about 46°C (115°F) prior to being combined
with the converter process gases, and subsequently treated in a double-
contact acid plant. The pressure drop across both scrubbers is about
61 cm (24 in.) of water.
Both inlet and outlet arsenic emission samples were obtained.
The inlet sample was collected upstream of the venturi scrubber while
the outlet sample was collected downstream of the spray-type scrubber.
It was not possible to sample directly downstream of the venturi
scrubber because of the system configuration. Table 3-13 presents a
38
summary of the results.
3-60
-------
Table 3-13. ARSENIC PERFORMANCE DATA FOR VENTURI SCRUBBER
AT KENNECOTT-HAYDEN
— — — _•____
Sample
run
1
2
3
Avg.
a
Arsenic emissions3
•• — ___
°C
336
328
324
329
Inlet
mg/Nm
29.53
25.87
22.90
26.10
kg/hr
0.85
0.74
0.75
0.78
°C
46
44
28
39
Outlet
mg/Nm3
0.64
0.27
0.32
0.41
kg/hr
0.02
0.01
0.01
0.01
• — .
Efficiency,
percent
97.9
98.9
98.8
98.4
based on
As the results indicate, the arsenic inlet loading to the scrubber
was quite low, averaging 26.10 mg/Nm3 (0.0114 gr/dscf) and 0.78 kg/hr
(1.72 Ib/hr). The outlet concentration was very low, averaging 0.41 mg/Nm:
(0.0002 gr/dscf). The arsenic mass emission rate at the outlet averaged
0.01 kg/hr (0.023 Ib/hr). The average collection efficiency observed
was 98.4 percent. It should be noted that the three inlet runs exceeded
isokinetic tolerances (refer to Appendix C) . This was due primarily
to the large fluctuations in the moisture content of the gas stream
frum run to run. As a result, the actual inlet concentration was
probably somewhat higher than that measured. Consequently, the actual
arsenic collection efficiency of the system is probably slightly
higher than the 98.4 percent recorded.
3'2'1-4 Sulfuric Acid Plant... Tests were performed at the
Kennecott-Hayden, ASARCO-E1 Paso, and Phelps Dodge-Ajo smelters to
evaluate the performance of acid plants in controlling arsenic emissions.
Gas precleaning and conditioning of the smelter offgases used for
sulfuric acid manufacturing is absolutely necessary for effective acid
plant operation. Both hot and cold gas cleaning devices are used.
3'2'1-4'1 Double-contact acid plant (Kennecott-HayriPnK The
double-contact acid plant operated at this smelter treats a combination
of flU1d-bed roaster and converter process gases. Acid production Ts
typically about 935 Mg/day (850 tpd) of 93.5 percent sulfuric acid.
3-61
-------
After passing through a series of cyclones for calcine recovery, the
fluid-bed roaster offgases are treated in a venturi scrubber for
particulate removal and a spray tower for hunidification and cooling.
The converter offgases are captured in water-cooled hoods, cooled to
370°C (700°F) in a gas cooler, and routed to an electrostatic precipitator
for particulate removal. Gases exiting the precipitator enter a spray
tower, similar to that used on the roaster gas stream, where they are
humidified, cooled, and subsequently combined with the roaster gas
stream. The combined gas stream, totaling about 2,120 Mm /min (75,000 scfm)
at 46°C (115°F), then passes through three parallel trains of two mist
precipitators in series, where acid mist and any remaining particulates
are precipitated. The gas stream, which typically contains 5 to
8 percent S02, then enters the double-contact acid plant where it is
dried, the S02 converted to S03, and the S03 absorbed in weak acid to
form strong acid.
Three arsenic test runs were conducted by EPA on the acid plant
tail gas stream (outlet). These tests were performed concurrently
with those across the venturi scrubber described in the preceding
section. The average arsenic concentration measured was 3.43 mg/Nm
(0.0015 gr/dscf). The corresponding arsenic mass rate averaged 0.41 kg/hr
TO
(0.90 lb/hr).
3.2.1.4.2 Double-contact acid plant (ASARCO-E1 Paso). Offgases
generated during converter blowing operations at the ASARCO-E1 Paso
smelter are treated in a 454 Hg/day (500 tpd) double-contact sulfuric
acid plant for S02 removal. The offgases are captured in water-cooled
hoods, passed through two parallel waste heat boilers, and cooled by
evaporative cooling in a spray chamber. The cooled gases [about
1,700 Nm3/min (60,000 scfm) at 149°C (300°F)] then enter an ESP for
particulate removal. The precipitator consists of four parallel
chambers, each having four sections in series. The exiting gases pass
through a venturi scrubber for additional particulate removal, are
humidified and cooled in a pair of packed bed scrubbers, and then are
treated in a series of mist precipitators where water and any remaining
particulates are removed prior to entering the acid plant.
3-62
-------
Arsenic emission measurement; were conducted by EPA at the inlet
to the spray chamber and at the acid plant outlet. Three sample runs
Table 3-14.
^DOUBLE-CONTACT
Arsenic emissions
Efficiency,
percent
0.0002
0.0031
0.0011
0.0004
0.022
0.355
0.126
0.038
— •-•• -•"•• HIM jo i a uc uaua c
catch (front and back half).
Only three inlet sample runs were made.
Average of first three runs only.
As indicated, the measured inlet and outlet arsenic concentrations
averaged 0.976 g/Nm3 (0.426 gr/dscf) and 0.0015 g/Nm3 (0.0007 gr/dscf)
respectively. The arsenic mass rate averaged 96.0 kg/hr (211 lb/hr)
at the inlet and 0.168 kg/hr (0.370 lb/hr) at the outlet, indicating
an average arsenic removal efficiency in excess of 99 percent.
3'2-1-4'3 ^l^ntaji^^ Offgases
generated during converting at the Phelps Dodge-Ajo smelter are treated
™ an ESP system for particulate removal followed by a 544 Mg/day
(600 tpd) single-contact sulfuric acid plant for S0? removal. The
offgases pass through waste heat boilers where they are cooled to
about 315°C (600°F), enter a balloon flue, and then pass through an
electrostatic precipitator. The precipitator consists of two independent
horizon al parallel units with three fields, each of which is designed
to handle 5,490 m^/min (210,000 acfm) at 340°C (650°F) and 95 1 kPa
3-63
-------
(13.8 psia). The exiting gases n^s into the scrubbing section of the
acid plant where they are treated in a humidifying tower, a cooling
tower, and a mist precipitator. The cleaned gases are then processed
in the acid plant. Either 93 or 98 percent sulfuric acid can be
produced.
Simultaneous inlet and outlet arsenic emission measurements were
conducted by EPA. Three sample runs each were made on the inlet and
outlet. The results are summarized in Table 3-15. The offgases
treated in the acid plant contained a negligible amount of arsenic^
The measured inlet and outlet concentrations averaged 0.00007 g/Nm
(0.00003 gr/dscf) and 0.000016 g/Nm3 (0.000007 gr/dscf), respectively.
The arsenic mass rate averaged 0.004 kg/hr (0.009 Ib/hr) at the inlet
and 0.001 kg/hr (0.0022 Ib/hr) at the outlet, indicating an average
arsenic removal efficiency of 75 percent.
Table 3-15. ARSENIC PERFORMANCE DATA FOR SINGLE-
CONTACT ACID PLANT AT PHELPS DODGE-AJO
Sample
run
1
2
3
Avg.
a
Arsenic emissions
Inlet
°C
190
182
172
181
g/Nm3
0.00008
0.00003
0.00009
0.00007
kg/hr
0.006
0.002
0.005
0.004
Outlet
°C
60
73
53
62
g/Nm3
0.000007
0.00001
0.00003
0.000016
kg/hr
0.0006
0.0007
0.0013
0.001
Efficiency,
percent
90.0
65.0
64.0
75.0
Concentration and mass rate data are based on measurements on the total catch
(front and back half).
3.2.2 Fugitive Control Systems Evaluation
3.2.2.1 Local Ventilation Techniques Applied to Calcine Discharge.
Matte Tapping, and Slag Tapping. The performance capability
of the local ventilation techniques used at the ASARCO-Tacoma smelter
for the control of fugitive arsenic emissions from calcine discharge,
matte tapping, and slag tapping operations were evaluated. These
techniques were previously described in Sections 3.1.2.4, 3.1.2.5, and
3-64
-------
3.1.2.6, respectively. Visual ob-rvations were made using either EPA
Method 22 or EPA Method 9, depending on whether the emissions observed
were intermittent or continuous. Method 22 is used to determine the
occurrence of visible emissions, while Method 9 is used to determine
the opacity of emissions. A summary of the visible emission data
obtained is presented in Table 3-16.41
3-2-2-K1 Calcine transfer. Thirteen calcine transfer operations,
each averaging about 2 ninutes in duration, were observed. The visual
observations were made using EPA Method 22 at the opening of the
tunnel-like structure used to house the calcine hoppers and larry cars
during the calcine transfer (discharge) operations. As the data
indicate, no visible emissions were observed at any time.
3-2-2-l-2 Matte tapping. Visible emission observations during
furnace matte tapping were also made using EPA Method 22. Simultaneous
but separate observations were made both at the furnace tap port and
at the launder-to-ladle transfer point. Sixteen taps, averaging
approximately 5.5 minutes in duration, were observed at the tap port.
Out of the 16 observations made at the matte tap port, no visible
emissions were observed 100 percent of the time on 14 of these, with
only slight emissions ranging from 1 to 3 percent of the time for the
remaining two observations. No visible emissions were observed 100 percent
of the time from the launder-to-matte ladle transfer point during all
15 observations made at the transfer point.
3-2'2<1-3 .sJag tapping. Slag tapping emissions were observed
using both EPA Methods 22 and 9. As with matte tapping, separate
observations were made at the furnace tap port location and at the
slag launder-to-slag pot transfer point. Results obtained using EPA
Method 22 for eight observations at the slag tap port showed that
visible emissions were observed about 5 percent of the time on the
average, with the highest single observation showing the presence of
visible emissions 15 percent of the time. Visual observations made at
the slag launder-to-pot transfer point indicated very poor performance,
with visible emissions being observed 72 to 99 percent of the time
over 11 slag taps. Additional data obtained using EPA Method 9 showed
significant emissions with opacities as high as 50 percent. Conversations
3-65
-------
Table 3-16. SUMMARY OF VISIBLE EMISSION OBSERVATION DATA FOR CAPTURE
SYSTEMS ON FUGITIVE EMISSION SOURCES AT ASARCO-TACOMA3
OJ
I
CTl
CTl
Operation
Calcine transfer system
Matte tapping:
at matte tap port and
launder
at matte discharge
into ladle
Slag tapping:
at slag tap port and
launder
at slag discharge
into pots
FPA Method 22
No. of
readings
taken
13
16
15
8
11
Average
observation
time,
min:sec
1:55
5:28
5:21
13:38
15:27
Average
percent
time
emissions
observed
on all
readings
0
0.2
5.3
88
Range of
percent
time
emissions
observed
0
0-3
0-15
72-99
EPA Method 9
Average Range of
No. of observation Average opacity
readings time, opacity, percent
taken min:sec percent observed
2 13:45 6 0-30
7 14:32 12 0-50
aVisible emission observations made on June 24 through 26, 1980.
-------
with smelter personnel revealed that the ventilation hood at the slag
launder discharge point had been damaged when hit by a truck. Although
an inspection of the ventilation hood and ancillary ductwork showed no
apparent damage, ventilation at this location was concluded to be
inadequate to handle the volume of emissions and fume generated.
3'2<2-2 Fugitive Emission Controls for Converters-Air Curtain
Secondary Hood Capture System.
3'2-2'2'1 Evaluation program at ASARCQ-Tarnm;.. EPA conducted an
evaluation program in January 1983, on the prototype air curtain
secondary hood recently installed at ASARCO-Tacoma.29 The primary
objective of this test program was to obtain an estimate of the overall
capture efficiency of the air curtain control system and also the
capture efficiency during specific modes of converter operation.
Capture efficiencies of the system during three complete converter
cycles were estimated using two principal techniques: (1) a gas
tracer study using sulfur hexafluoride (SFg) was performed by injecting
the gas into the fugitive emission plume and measuring the amount of
the gas captured by the secondary hooding system, and (2) detailed
visual observations of the hooding system performance were made concurrent
with the gas tracer study.
In the gas tracer study, SFg was injected into the controlled
area of the air curtain at constant, known rates of 30 to 50 cc/min
for periods which ranged from 15 minutes to 2 hours per injection.
Single point samples of the exhaust gases from the air curtain hood
were collected at a downstream sampling location by pulling samples
into 15-liter, leak-free Tedlar bags for onsite gas chromatographlc
analysis. The air curtain capture efficiency was calculated by comparing
the SF6 injection mass flow rate with the mass flow rate calculated
for the downstream sampling point.
Injections of SFg gas were made at 16 sample points through 4
test ports in adjacent access doors on both sides of the converter
baffle walls. The locations of the points are shown in Figures 3-18
and 3-19. In addition to the efficiency measurements made for the
points in the primary testing area, several tests were performed at
injection points outside of this area (below the converter centerline) in
3-67
-------
NO. 4
CONVERTER
TOP VIEW
JET SIDE
EXHAUST SIDE
AIR
CURTAIN
JET
TO SUCTION FAN
NO. 4 CONVERTER
(FUME SOURCE)
LEGEND:
AREA SAMPLED USING
MATRIX TRAVERSE
INJECTION LOCATIONS
SAMPLE I.D.
? SP1 & 2
D SP3 - 5
• SP7 - 1?
O SP13 - 73
ELEVATION
Figure 3-18. SFg Tracer Injection Locations
3-68
-------
CONVERTER AISLE FLOOR
O INJECTION POINTS
Figure 3-19. Tracer Injection Test Ports
3-69
-------
an attempt to characterize the effective capture area of the air
curtain hooding system, particularly during converter roll-out activities.
On January 14, 1983, capture efficiencies were determined for 45
injection points in the controlled area. The calculated mean efficiencies
by converter operational mode are presented in Table 3-17.
Table 3-17. AIR CURTAIN CAPTURE EFFICIENCIES AT ASARCO-TACOMA
USING GAS TRACER METHOD - JANUARY 14, 1983
Converter
mode
Matte charge
Cold addition
Blowing
Slag skimming
Idle
TOTAL
Number of
injections
7
3
19
9
7
45
Mean
efficiency
93.1
102.0
92.8
95.0
93^.4
93. 5a
Calculated overall mean efficiency assumes the converter
operation consists of 80 percent blowing and idle, 15 percent
matte charge and cold addition, and 5 percent slag skimming.
The overall mean capture efficiency for all modes of operation was
93.5 percent. With the exception of cold additions, the operating
mode of the converter had little effect on capture efficiency measured,
which ranged from 92.8 percent during blowing to 95.0 percent during
slag skimming. The port through which the releases of tracer gas were
made did not have any effect on the calculated efficiency. However,
it was found that sampling points tested through a particular port
exhibited considerable variation, generally recording higher capture
efficiencies at positions 1 and 2 (exhaust side) than at positions 3
and 4 (jet side).
The remaining test series of 48 injections was performed on
January 17-18, 1983. The results of this series are summarized in
Table 3-18.
3-70
-------
Table 3-18. AIR CURTAIN CAPTURE EFFICIENCIES AT
ASARCO-TACOMA USING GAS TRACER METHOD - JANUARY 17-19, 1983
Converter
mode
Matte charge
Cold addition
Blowing
Slag skimming
Copper pour
Idle
TOTAL
Number of
injections
6
3
27
7
4
-_ _^
51
Mean
efficiency
94.2
96.7
96.7
94.3
88.5
100.0
96. 5a
Calculated overall mean efficiency assumes the converter
operation consists of 80 percent blowing and idle, 15 percent
"
The overall mean capture efficiency for all operational modes was
96.5 percent. As with the data recorded on January 14, the operating
mode appeared to have no significant effect on the individual calculated
efficiencies, which ranged from 88.5 percent during copper pouring to
96.7 percent during both blowing and cold additions. However, any
consistent, small variations in the efficiencies for various modes, if
they were present, would be difficult to detect in the relatively
small number of test runs (injections) which were made. The error in
the calculated air curtain capture efficiencies has been estimated to
be ±18 percent. For this second test series, it was also found that
the location of the test port had no effect on efficiency, while
exhaust-side efficiencies were found to be somewhat higher than jet-side
effTciencies. Test results from the injection points in these tests
indicate that, on the average, about 95 percent of the gases and
particulate matter in the area immediately above the converter is
likely to be captured by the air curtain secondary hooding systen.
3-71
-------
In addition to these two te-u series, a series of special injection
point tests was conducted in order to assess the effective capture
area of the secondary hood system outside the confines of the hood.
The special injection tests were performed with the injection probe at
a number of points on the perimeter of the main test area, such as
very close to the baffle wall and below the ladle during the matte
charging and cold addition modes. Table 3-19 shows the results of
this test series.
Table 3-19. AIR CURTAIN CAPTURE EFFICIENCIES AT ASARCO-TACOMA
FOR SPECIAL GAS TRACER INJECTION POINTS -
JANUARY 18-20, 1983
Converter
mode
Matte charge
Cold addition
Blowing
Slag skimming
Copper pour
Idle
TOTAL
Number of
injections
17
6
6
28
4
8
69
Mean
efficiency
61.8
61.5
33.0
84.0
80.8
53.8
49. 4a
Calculated overall mean efficiency assumes the converter
operation consists of 80 percent blowing and idle, 15 percent
matte charge and cold addition, and 5 percent slag skimming
and copper pour.
The overall average capture efficiency for the 69 special injection
points was 49.4 percent. Unlike the first two test series, the capture
efficiency in the special series was sensitive to converter mode. For
example, the slag skimming and copper pour efficiencies are higher, at
84.0 and 80.8 percent, respectively, than the other modes because of
the position of the ladle (above the injection probe) during these
modes.
During the course of the gas tracer study, from January 18 to 22,
1983, detailed visual observations were made of the performance of the
3-72
-------
air curtain control system throuah the various converter operating
modes. The purpose was to estimate capture effectiveness and to
qualitatively characterize both captured emissions and emissions
escaping capture.
An important benefit from these observations which could not be
derived from the results of the gas tracer study was the conclusion
that operating practices play a major role in the overall performance
of the control system. The crane operator who moves the ladle into
position for charging, skimming, or pouring has some latitude in the
positioning of the ladle, the speed of pouring, and the speed and
timing of the movement of the ladle to and from the converter. Observations
show that slower, more deliberate movements and closer positioning of
poured materials to the converter have a definite effect on the extent
of air curtain penetration and spillage into the converter aisle.
The practices employed by crane and converter operators were
found to vary significantly throughout the observation period. For
example, during slag skimming, it was observed that the rate of pouring
varied for individual skims. The capture efficiency was higher when
the pour rate was lower, since a faster rate would sometimes "overwhelm"
the control system. Also, the position of the ladle was noticed to be
important, because when the ladle was positioned close to the converter
mouth the capture efficiency was enhanced. During matte charging, the
wundrawal of the crane from above the converter was often observed to
cause "drag-out" emissions, especially when the crane was moved immediately
after the converter was charged. When the crane was left in place for
a few moments after charging (and the heaviest part of the emissions
had risen to the air curtain), the drag-out emissions were noticeably
reduced, and the capture effectiveness was improved.
Only one period of blowing was evaluated quantitatively during
the observations. Some penetration of the air curtain was noticed (5
to 10 percent) during roll-in when the blowing air first started, but
the overall capture efficiency was judged to be about 95 percent
Overall capture efficiencies for individual matte chargings were
in the range of 90 to 95+ percent. Cold additions (adding of cool
3-73
-------
solidified materials) to the con"-rter frequently produced emissions
heavy enough to virtually overwhelm the capture system, especially
when a fire ignited in the converter. Capture efficiencies were
somewhat lower overall than for matte charging, typically ranging from
80 to 95 percent.
During slag skimming, it was observed that the rate of pouring
varied for individual skims. For pour rates judged to be slow or
moderate, efficiencies generally ranged from 80 to 95 percent. For
faster pour rates, the typical efficiencies dropped to 70 to 80 percent.
Copper pouring generally produced a moderate to heavy amount of
fume; both air curtain penetration and spillage out into the converter
aisle were very slight. Capture efficiencies were typically 90 to
95 percent during pouring. However, at initial roll-out prior to
pouring, the efficiency could be as low as 70 percent.
The observation that fumes would frequently spill out into the
converter aisle (especially during slag skimmer activities) indicates
that the entire fugitive emission plume either is captured by the air
curtain exhaust or penetrates the curtain vertically (since vertically
escaping gases should contain a homogeneous concentration of the
tracer gas). Therefore, by not accounting for emissions which escape
before approaching the air curtain, the gas tracer method probably
overpredicted the capture efficiency achieved during slag skimming
operations.
In general, visual assessments of secondary hood capture effectiveness
correlate quite well with the average efficiencies determined by the
gas tracer method. As mentioned earlier, operating practices have a
significant influence on the degree of capture achieved during any
individual converter operation, and many of the extreme observed
values can be understood in terms of the operating practices employed
during those particular operations.
3-74
-------
3'2'2'2'2 Visible Emission Observations at Tamann SmpUpr_ All
three converters at the Tamano smelter in Japan are equipped with a
fixed enclosure and air curtain system for control of fugitive emissions
generated during various modes of converter operation. The enclosure
doors and roof are kept open and the air curtain system is turned on
during the matte charging. During all other modes of the converter
operation the doors and roof are kept closed and the air curtain
system is turned off. This system is described in Section 3.1.2.7.2.
Visible emission observations were made for the air curtain
secondary hood operated on the No. 3 converter during day-shifts on
March 12 and March 13, 1980. The converter is a conventional Pierce-Smith
design, measuring about 9 meters in length and 4 meters in diameter
Observations were made using EPA Methods 22 and 9, depending on whether
the emissions observed were intermittent or continuous, for the different
modes of converter operation comprising a converter cycle. A discussion
of the results obtained during each mode of converter operation follows
Matte charqinq. Usually three ladles of matte are brought to the
converter and charged in a 10- to 30-ninute period. Actual matte
charging from each ladle lasts for 1 to 1.5 minutes. The fixed enclosure
doors and roof are opened; the air curtain systan is turned on; and
the ladle of matte is brought into the secondary hood by an overhead
crane. The converter is rolled down to an inclined position- the
matte ladle is lifted up by the crane; and matte is charged into the
converter. At the completion of matte charge, the ladle is moved out
of the enclosure and, if needed, another ladle of matte is brought in
After the matte additions are completed, the converter is rotated into
the primary converter hood, the roof and doors are closed, and slag
blowing is commenced.
Three separate matte charges were observed using both EPA Methods 9
and 22 simultaneously, and one matte charge was observed using EPA
Method 9 only. Visual observations for each matte charge observed
were made only during the period when the matte was actually flowing
into the converter. Results of the visual observations obtained are
summarized in Table 3-20.
3-75
-------
As shown in Table 3-20, vi-iole emissions were observed for three
individual matte charges. The observations ranged from 44 to 77 percent
of the time (EPA Method 22). Although somewhat continuous, the opacity
results indicate that these emissions were generally slight, typically
ranging from 0 to 10 percent opacity, with the highest average opacity
recorded for a single matte charge being 25 percent. When present,
the emissions appeared as small puffs which penetrated the air curtain
stream.
Table 3-20. VISIBLE EMISSIONS OBSERVATION DATA
FOR CONVERTER SECONDARY HOOD SYSTEM
DURING MATTE CHARGING AT THE TAMANO SMELTER
Sample
run
1
2
3
4
Total
Method 22
Observation
period, min
1.5
1.25
1.75
_
4.50
Percent of time
emissions observed
44
56
77
-
60
Method 9
Observation
period, min
1.5
1.25
1.75
1.5
6.25
Average opacity
for observation
period, percent
5.0
4.0
3.0
0
2.8
Range of
individuc
readings
0 to 25
0 to 10
0 to 10
-
0 to 25
Slag blow and copper blow. During slag blowing and copper blowing,
the converter mouth is enclosed by the primary duct, and offgases are
directed to the acid plant. The converter secondary hood doors and
roof are closed, and the air curtain system is turned off. Fugitive
emissions generated during blowing as a result of primary hood leaks
are captured inside the converter housing and are vented to a baghouse
for collection. The slag blow, which is divided into three segments,
lasts for about 150 minutes per converter cycle arid the copper blow
for about 200 minutes per cycle.
Visible emission observations were made using EPA Method 9 for
the converter hood systan for 30 minutes during the slag blow and for
27 minutes during the copper blow. No visible emissions (zero percent
44
opacity) were observed at any time.
3-76
-------
At the end c? each of the three slag blow phases
slag is slammed into a ladle and transported to a sand bed area for
cooling. Because of the quantities involved, slag is discharged from
the converter two times after the first slag blow and once after the
second and third. Each slag skim lasts for about 10 minutes. During
each skim, an empty ladle is brought into the enclosure by an overhead
crane and placed on the ground in front of the converter. The crane
is moved out, and the enclosure doors and roof are closed. The converter
is rolled down, and slag is poured into the ladle. After the slag
skimming is completed, the converter is rotated upward slightly, the
enclosure doors are opened, and the slag ladle is moved out.
Only two skims were observed. The first, which lasted 11 minutes
was observed using EPA Methods 22 and 9. The second slag skim, lasting
9 minutes, was observed using EPA Method 22 only. Each observation
period began as the converter started rolling down to pour the slag
into the ladle and lasted until the pouring was completed and the
converter started rolling up. Daring the first slag skim observed, no
visible emissions were observed at any time. In contrast, during the
second slag skim, visible emissions were observed 100 percent of the
time. Most of the time, however, these emissions were slight, ranging
from 5 and 10 percent opacity and consisting of small puffs which
escaped from the enclosure through a narrow opening between the front
doors and the enclosure roof.
Ulster discharge. At the end of copper blowing, the product
blister is discharged into a ladle and transported to a refining
furnace. Usually four ladles of blister are filled per converter
cycle. Each of the first three blister pours lasts about 12 to
14 minutes with the final pour lasting about 4 minutes. The time
elapsed between each blister pour is about 8 to 15 minutes.
At the end of a copper blow, the secondary hood doors and roof
are opened. An empty ladle is brought into the secondary hood by the
overhead crane and placed in front of the converter. The crane is
moved out, and the secondary hood doors and roof are closed The
converter is rolled down, and blister is poured into the ladle. After
the blister pour is completed, the converter is rolled up slightly
3-77
-------
the hood doors are opened; the baiter ladle is taken to the refining
furnace by the crane; and the hood doors and roof are closed.
Four blister discharges were observed. Both EPA Methods 22 and
9 were used. A summary of the results is presented in Table 3-21.
Although the observation periods used in obtaining the EPA Method 22
data were variable (i.e., different start and end times), the results
nonetheless indicate that visible emissions during blister discharge
were generally continuous. The EPA Method 9 data, which were obtained
only during periods when the blister copper was actually being poured,
show that the visible emissions observed were somewhat more substantial
than those observed during either matte charging or slag skimming. As
shown in Table 3-21, the highest average opacity recorded for a single
blister pour was 13 percent, with individual opacity readings ranging
from 0 to 35 percent. Again, as with slag skimming, the emissions
observed generally appeared above the narrow opening between the front
doors of the enclosure and the enclosure roof.
Table 3-21. VISIBLE EMISSIONS OBSERVATION DATA FOR
BLISTER DISCHARGE AT THE TAMANO SMELTER
Sample
Run
1
2
3
4
Total
Method
22
Observation Percent of
period, min. time emissions
observed
25a
15b
6C
15.3
42
-
86
19
49
Method 9
Observation
period, min
-
15.0
12.0
3.5
30.5
Avg. opacity
for observation
period, percent
-
6.2
13.0
3.2
8.5
Range of
individual
opacity
readings
-
0 to 30
0 to 35
0 to 25
0 to 35
Observations started when secondary hood doors opened 12 minutes prior to the
blister discharge, during which time the converter body was hit by a vibrating ram.
Observations started with the blister discharge and continued for 3 minutes after
completion of the blister discharge.
Observations started with the blister discharge and continued for 2-1/2 minutes
after completion of the blister discharge.
3-78
-------
3-2.2.3
_
emissions from converters and anode furnaces at the ASARCO-E1 Paso
smelter are captured by evacuating the converter building. The building
evacuation systen at this smelter is described in Section 31273
The captured fugitive gases are drawn through four openings at the '
roof of the converter building into ducts which merge into a main duct
leading to a baghouse, then through fans to the 250 m (828 ft) main
stack.
The fugitive gas flow through the baghouse averages about
14,100 Nm /min (498,000 scfm). The baghouse consists of 12 compart-
ments. Normally all compartments are in operation except that one
compartment is taken off during the cleaning cycle and another compartment
during the maintenance cycle. Each compartment contains 384 Orion or
Dacron bags, 20 cm (8 in.) in diameter and 6.7 m (22 ft) long, providing
a cloth area of 1,644 m2(l7,700 ft2) per compartment. The total net
cloth area of the baghouse is about 19,732 m2 (212,400 ft2) The
baghouse was designed to effectively treat 15,280 m3/min (540,000
acfn,) at 54°C (130'F) with an air-to-cloth ratio of 0.91 m3/min per m2
(3 cfm/ft ). Mechanical shakers are used for cleaning the bags
Inlet and outlet emission measurements for inorganic arsenic and
total paniculate were conducted by EPA across the baghouse. During
al tests, converter operations were monitored and testing was conducted
only when one or more converters were in operation. The arsenic
results obtained are summarized in Table 3-22.45 As indicated, the
measure, inlet and outlet arsenic concentrations averaged 3.27 mg/Nm3
(0.0014 gr/scf) and 0.137 mg/Nm3 (0.00006 gr/scf), respectively The
arsenic mass rate averaged 2.92 kg/hr (6.45 Ib/hr) at the inlet and
O.IH g/hr (0.244 Ib/hr) at the outlet, indicating an average arsenic
-oval efficiency in excess of 96 percent. The results of the particulate
measurements obtained are summarized in Table 3-23.46 As shown in the
table, the mass particulate inlet concentration was low, averaging
only 0.062 g/Nm3 (0.027 gr/scf). Nonetheless, the mass particulate
em1Ssion rate was relatively high, averaging over 50 kg/hr (110 Ib/hr)
The low inlet concentration is a result of the large quantities of
dilution air associated with the application of general ventilation
3-79
-------
Table 3-22. ARSENIC DATA FOR CONVERTER BUILDING
BAGHOUSE AT ASARCO-EL PASO
Sample
run
1
2
3
Avg.
Arsenic emissions
Inlet
°C
38
36
51
42
mg/Nm
6.21
2.09
1.53
3.27
kg/hr
5.51
1.96
1.31
2.92
Outlet
°C
38
37
51
42
mg/Nm
0.39
0.017
0.015
0.137
kg/hr
0.305
0.017
0.012
0.111
Efficiency,
percent
94.5
99.1
99.1
96.2
Table 3-23. PARTICIPATE DATA FOR CONVERTER BUILDING
BAGHOUSE AT ASARCO-EL PASO
Sample
run
1
2
3
Avg.
Particulate emi
Inlet
°C
16
22
16
18
mg/Nm
60.3
53.3
70.5
61.3
kg/hr
44.7
46.3
61.2
50.7
ssions
Outlet
°C
20
14
16
17
mg/Nm
11.6
2.5
1.1
5.1
kg/hr
10.4
0.92
6.4
3.9
Efficiency,
percent
76.7
98.0
99.3
91.3
3-80
-------
techniques. The outlet concentration and mass rate averaged 5.1 mg/Nra3
and 3.9 kg/hr, respectively. Although the collection efficiency
obtained over three test runs averaged only about 90 percent, the
results indicate that collection efficiencies as high as 99 percent
are achievable.
3.2.3 Conclusions
3.2.3.1 Process Controls. As discussed in Section 3.1.1.1, the
arsenic control devices considered as best available control for
process sources at primary copper smelters incorporate gas stream
precooling as an integral part of the overall control system. Cooling
is vital to effective control because the relatively low saturation
concentration of arsenic trioxide at lower temperatures allows a
greater percentage of the arsenic to condense into a form which can be
collected in a particulate control device. The process control systems
tested and discussed in Section 3.2.1 include the spray chamber/baghouse
system at Anaconda; the roaster and arsenic plant baghouses at ASARCO-Tacoma-
the spray chamber/ESP at ASARCO-E1 Paso; the reverberatory furnace ESP
at ASARCO-Tacoma; the venturi scrubber at Kennecott-Hayden; and the
acid plants at ASARCO-E1 Paso, Kennecott-Hayden, and Phelps Dodge-Ajo
The inlet temperature to the baghouses and the electrostatic precipitators
ranged between 80 and 115°C (180 to 240°F).
The results of these tests, in terms of the arsenic collection
eff1C1ences recorded, are shown in Figure 3-20. (The results of ESP
testing at ASARCO-Tacoma are not presented in Figure 3-20, because the
data were collected only for the ESP outlet; and the results of the
single-contact acid plant at Phelps Dodge-Ajo, are not presented
because of its measured 75 percent efficiency due to extremely low
inlet arsenic loadings.) Each black circle in the figure represents a
sample run performed on the control device designated at the bottom of
the figure. The average efficiencies (designated by horizontal bars)
ranged from a low of about 96 percent (baghouse at 84°C) to a high of
about 99.7 percent (baghouse at 91°C). The data demonstrate that
baghouses, ESP's, venturi scrubbers, and acid plants can be used to
provide a high level of control for arsenic emissions. However, as
discussed earlier, the collection efficiency for arsenic of any particulate
3-81
-------
CJ
ce
100 r
99 -
l^"^^l^i W
^J ^»ii«J
98 f-
96
„
94
93
-
^^^
y
i_
i
nr
M
y
H
'
92 [- BH - BAGHOUSE
1 SC SPRAY CHAMBER
91
an
ESP ELECTROSTATIC PRECIPITATOR
VS VENTURI SCRUBBER
AP ACID PLANT
IT INLET TEMPERATURE
1
j i
J J 1 1
IT -
BH
91 °C
BH
84°C
SC/BH
no°c
SC/ESP
115°C
315°C
210°C
Figure 3-20. Control Device Arsenic Collection Efficiencies
3-82
-------
techniques. The outlet concentration and mass rate averaged 5.1 mg/Nm3
and 3.9 kg/hr, respectively. Although the collection efficiency
obtained over three test runs averaged only about 90 percent, the
results indicate that collection efficiencies as high as 99 percent
are achievable.
3.2.3 Conclusions
3.2.3.1 Process Controls. As discussed in Section 3.1.1.1, the
arsenic control devices considered as best available control for
process sources at primary copper smelters incorporate gas stream
precooling as an integral part of the overall control system. Cooling
is vital to effective control because the relatively low saturation
concentration of arsenic trioxide at lower temperatures allows a
greater percentage of the arsenic to condense into a form which can be
collected in a particulate control device. The process control systems
tested and discussed in Section 3.2.1 include the spray chamber/baghouse
system at Anaconda; the roaster and arsenic plant baghouses at ASARCO-Tacoma-
the spray chamber/ESP at ASARCO-E1 Paso; the reverberatory furnace ESP
at ASARCO-Tacoma; the venturi scrubber at Kennecott-Hayden; and the
acid plants at ASARCO-E1 Paso, Kennecott-Hayden, and Phelps Dodge-Ajo
The inlet temperature to the baghouses and the electrostatic precipitators
ranged between 80 and 115°C (180 to 240°F).
The results of these tests, in terms of the arsenic collection
effTciences recorded, are shown in Figure 3-20. (The results of ESP
testing at ASARCO-Tacoma are not presented in Figure 3-20, because the
data were collected only for the ESP outlet; and the results of the
single-contact acid plant at Phelps Dodge-Ajo, are not presented
because of its measured 75 percent efficiency due to extremely low
inlet arsenic loadings.) Each black circle in the figure represents a
sample run performed on the control device designated at the bottom of
the figure. The average efficiencies (designated by horizontal bars)
ranged from a low of about 96 percent (baghouse at 84°C) to a high of
about 99.7 percent (baghouse at 91°C). The data demonstrate that
baghouses, ESP's, venturi scrubbers, and acid plants can be used to
provide a high level of control for arsenic emissions. However as
discussed earlier, the collection efficiency for arsenic of any particulate
3-81
-------
98
96
95
94
93
92
91
8H - BAGHOUSE
SC - SPRAY CHAMBER
ESP- ELECTROSTATIC PRECIPITATOR
VS - VENTUR1 SCRUBBER
AP - ACID PLANT
IT - INLET TEMPERATURE
8H
IT _ 91°C
BH
84°C
SC/BH SC/ESP
110°C
VS
315°C
AP-A
Figure 3-20. Control Device Arsenic Collection Efficiencies
3-82
-------
control device can vary dependin; on the distribution between the
Peculate and vapor form of the arsenic which reaches the device
This distribution in turn depends on the arsenic concentration in ihe
gas stream and the stream temperature. Therefore, any discussion of
control efficiency of a particular type of device must consider these
parameters before an estimate of the expected efficiency can be made '
The process gas streams entering the control devices tested clearly
had sufficiently high concentrations of arsenic trioxide for high
control efficiencies to be achieved.
3-2-3.2 fugitive Controls, The performance capabilities of
capture systems for the control of fugitive arsenic emissions fro,
calcine discharge, matte tapping, slag tapping, and converter operatjo|)s
«ere evaluated. Estates or the capture efficiency of these systems
are based on the visible emissions observations reported in the preceding
sections and on subjective judgment. Also eva!uated was the performance
capability of a baghouse control device used to collect captured
fugitive emissions.
Visual observations made on the local ventilation system applied
to calcine discharge operations at ASARCO-Tacoma resulted in no visible
emissions being observed at any time during the observation period
As a result, it is concluded that such a system is readily capable'of
achieving a capture efficiency of 90 percent or better
Observations conducted on the local ventilation system applied to
matte tapping operations at the same smelter showed no visible emissions
occurring at any time at the matte launder-to-,ad,e transfer point and
only slight emissions of short duration, a maximum of 3 percent of the
time, for any individual matte tap observed at the matte tap port I
*"»'«' »' "Crated
system applied to raatte tapping operations should achieve
a minimum capture efficiency of 90 percent.
Similar observations made on the ventilation system serving the
slag tappmg operations at Tacoma showed substantially poorer performance
-specially at the slag ,aunder-to-slag pot transfer point where v, ™
- ssions were observed nearly !00 percent of the time during each
i-dnn-u.1 s,ag tap. Based on the results of the visua, observation,
3-83
-------
and the fact that the capture system had reportedly been damaged, it
is concluded that the slag tapping ventilation system observed at
Tacona, as it was operating at the time, should not be considered
representative of a best system of emission reduction. Although slag
tapping operations do represent a somewhat more difficult control
situation than matte tapping, the outstanding performance demonstrated
by the matte tapping controls at Tacoma strongly suggest that a properly
designed and operated ventilation system applied to slag tapping
operations should be capable of achieving at least 90 percent capture.
Capture systems evaluated for the control of fugitive arsenic
emissions from converter operations included the prototype air curtain
secondary hoods applied at the ASARCO-Tacoma smelter and the Mitsui
Smelting Company smelter in Tamano, Japan, and the general ventilation
or building evacuation system applied at the ASARCO-E1 Paso copper
smelter.
The gas tracer study performed on the converter air curtain
system at ASARCO-Tacoma indicates that approximately 95 percent of the
converter fugitive emissions from all operating modes is captured by
the system. Visual observations of the operation of this system
verify this measured capture efficiency, but also indicate that operating
practices play an important role in attaining this efficiency consistently.
EPA has concluded that this efficiency can be achieved with a properly
operated air curtain secondary hooding system in combination with
proper operating practices.
The visual observations made at Tarnano on the air curtain secondary
hood indicate that the application of such a system on a conventional
Pierce-Smith converter should result in an overall capture efficiency
of at least 90 percent. As reported previously, no visible emissions
were observed during either slag blowing or copper blowing, which in
combination account for nearly 80 percent of a typical converter
cycle. Although emissions were observed penetrating the air curtain
during matte charging, these emissions were judged to be negligible,
consisting of small puffs ranging from 0 to 10 percent opacity.
Emissions observed during (slag and blister) pouring operations were
somewhat more substantial, varying from 0 to 35 percent opacity.
3-84
-------
However, as previously indicated, ihese emissions were generally
observed to escape through a narrow opening which existed between the
enclosure doors and roof. The smelter representatives indicated that
a tighter seal between the doors and roof would result in a significant
improvement.
Conclusions regarding the potential effectiveness of the building
evacuation system used at the ASARCO-E1 Paso smelter are based primarily
on engineering judgment. Providing the building is properly enclosed
and adequate ventilation rates are applied, essentially 100 percent
capture should be possible. However, due to the need for openings in
the building for access and makeup air, a more conservative estimate
of 95 percent capture is considered more reasonable. The fact that
this system has shown less than satisfactory performance at the El Paso
facility highlights the difficulty of designing an evacuation system
which will provide a controlled ventilation air supply for all of the
emission sources in the building. Fugitive sources have generally
been found to be more successfully controlled by the use of local
ventilation (hooding) techniques.
With regard to the collection of fugitive arsenic emissions in a
control device, the emission measurements conducted across the baghouse
facility serving the converter building evacuation system at ASARCO-E1 Paso
showed that an arsenic collection efficiency of 96 percent or higher
was readily achievable, even though the measured inlet concentrations
were extremely low, averaging only 3.3 mg/Nm3 (0.0014 gr/scf)
3-85
-------
3.3 REFERENCES
1. Vallance, R.H. Arsenic In: A Text-Book of Inorganic Chemistry,
Friend, J.N. (ed.), London. Charles Griffin and Co., Limited.
1938. p. 129-131.
2. Handbook of Chemistry and Physics, 43rd Edition. The Chemical
Rubber Publishing Company. 1961. p. 2373.
3. Behrens, R.G., and G.M. Rosenblatt. Vapor Pressure and Thermo-
dynamics of Octahedral Arsenic Trioxide (Arsenolite). Journal of
Chemical Thermodynamics. 4;175-190. 1972.
4. Schwitzgabel, K., et. al. Trace Element Study at a Primary Copper
Smelter. Prepublication Copy. U.S. Environmental Protection
Agency. Contract No. 68-01-4136. January 1978.
5 Control Techniques for Particulate Air Pollutants. U.S. Environmental
Protection Agency. Publication No. AP-51. January 1969. p.
108-122.
6. Southern Research Institute. An Electrostatic Precipitator
Systems Study. Final Report to the National Air Pollution Control
Administration. Contract No. CPA22-69-73. October 30, 1970.
p. 20.
7. Reference 6, p. 22.
8 Control Techniques for Lead Emissions, Volume I: Chapters 1-3.
U.S. Environmental Protection Agency. Research Triangle Park,
N.C. Publication No. EPA-450/2-77-012. December 1977. p. 2-30.
9 Background Information for New Source Performance Standards:
Primary Copper, Zinc, and Lead Smelters, Volume I: Proposed
Standards. U.S. Environmental Protection Agency. Research
Triangle Park, N.C. Publication No. EPA-450/2-74-002a. October
1974. p. 4-10.
10. Reference 11, p. 4-9.
11. Danielson, John A. Air Pollution Engineering Manual. 2nd Edition.
U.S. Environmental Protection Agency. Research Triangle Park,
N.C. Publication No. AP-40. May 1973.
12. Hemeon, W.C.L. Plant and Process Ventilation. New York. The
Industrial Press. Second Edition, 1963.
13. Dynaforce Corporation Brochure. Air Curtains for Industry. New
York. June 1978.
3-86
-------
14. Pjwlesland, J.W. Air Curtai,s in Control led Energy Flows - To
- a?w
15. Katari, V., et.al. Pacific Environmental Services Tnr
16
Hood
ong
Cincinatt,, OH. Publication No. EPA-600/280-oJ™ May 1980.
22. Katari, V. and I.J. Weisenberg. Trip Report—Visit tn uah*
" u.
, r* "-
or tne A!r PollutTon Control Association. June 25-30, 1978.
3-87
-------
27. Correspondence from Mr. Moi"" Goto, Smelter Manager, Naoshima
Smelter, Japan, to Mr. I.J. Weisenberg, Pacific Environmental
Services, Inc. October 1, 1978. Capabilities of air curtain
control system.
28. ASARCO Design Report. Converter Secondary Hooding for the Tacoma
Plant. Prepared by ASARCO Central Engineering Dept., Salt Lake
City, UT. January 22, 1982.
29. PEDCo Environmental, Inc. Emission Test Report - Evaluation of
an Air Curtain Hooding System for a Primary Copper Converter,
ASARCO, Inc., Tacoma, Washington. Volume I. EPA Contract
Nos. 68-03-2924 and 68-02-3546. Preliminary Draft. March 1983.
30 Davis, J.A. Unidirectional Flow Ventilation System. Presented
at the 104th Annual AIME Meeting. New York. February 18, 1975.
p. 3, 4.
31 Telecon. Katari, Vishnu, Pacific Environmental Services, Inc.,
with Sieverson, Jim, ASARCO, Incorporated. February 25, 1983.
Copper smelter fugitive control system.
32. TRW Environmental Engineering Division. Emission Testing of
ASARCO Copper Smelter, Tacoma, Washington. U.S. Environmental
Protection Agency. EMU Report No. 78-CUS-12. April 1979.
p. 4-5.
33. Reference 32, p. 10-12.
34 Harris, D.L., Monsanto Research Corporation. Air Pollution
Emission Test - Particulate and Arsenic Emission Measurements
from a Copper Smelter. Anaconda Mining Company, Montana. U.S.
Environmental Protection Agency. EMB Report No. 77-CUS-5.
April 18-26, 1977. p. 5-16.
35. Reference 32, p. 6.
36. Harris, D.L., Monsanto Research Corporation, Dayton Laboratory.
Particulate and Arsenic Emission Measurements From a Copper
Smelter. U.S. Environmental Protection Agency. EMB Report
flo. 77-CUS-6. June 20-30, 1977. p. 18-25.
37 TRW Environmental Engineering Division. Air Pollution Emission
Test, ASARCO Copper Smelter, El Paso, Texas. U.S. Environmental
Protection Agency. EMB Report No. 78-CUS-7. April 25, 1978.
p. 13.
38 Larkin, R. and J. Steiner, Acurex Corporation/Aerotherm Division.
Arsenic Emissions at Kennecott Copper Corporation, Hayden, Arizona.
U.S. Environmental Protection Agency Report No. 76-NFS-l. May
1977. p. 2-2.
3-88
-------
39. Reference 36, p. 14-17.
40. Rooney, Thomas TRW Environmental Engineering Division. Emission
Env?rSnmentaliepPS"DOd-e COPPer Sme1t*r» Ajo« Arizona U.S
Contract No. 68-02-2812? Work^signmeTN^lS. ^Marc^l^g^
\}. O-D. *
41. Reference 15, p. 4.
42. Reference 29, p. 4-66.
43. Vervaert, A. and J. Nolan, U.S. Environmental Protection Aaencv
and Puget Sound Air Pollution Control Agency. Log Book NosT
and 2, Observations of Converter Secondary Hood Test at ASARCO
Tacoma. January 18-22, 1983. M5AKLU,
44. Reference 22, p. 3.
45. Reference 37, p. 7-8.
46. Reference 37, p. 4-5.
3-89
-------
4.0 MODEL PLANTS, REGULATORY BASELINE, AND REGULATORY ALTERNATIVES
Arsenic has been listed as a hazardous air pollutant under Section 112
of the Clean Air Act (National Emission Standards for Hazardous Air
Pollutants). To protect public health from unreasonable risks associated
with exposure to airborne arsenic, standards are being developed to
decrease arsenic emissions from primary copper smelters which process
low-arsenic content feed. This section defines the regulatory baseline,
presents the alternative ways that EPA can regulate arsenic emissions
from the affected sources, and describes the 14 existing low-arsenic
throughput copper smelters for analysis of the environmental, economic,
and energy impacts of the regulatory alternatives on the industry.
The technical and economic impacts of arsenic regulatory alternatives
on primary copper smelters are measured as incremental changes beyond
the conditions that would exist in the absence of NESHAP regulations.
These conditions are commonly referred to collectively as a baseline.
This section defines such a baseline as it applies to the primary
copper smelting industry in general and to the low-arsenic throughput
smelters specifically.
In the following analysis, baseline is chosen to represent existing
process and control equipment except in the case of four smelters
which are scheduled to undergo modernization. For these four smelters
ASARCU-Hayden, Phelps Dodge-Ajo, Phelps Dodge-Morencl, and Kennecott-Hurley
baseline is defined as process and control equipment after modernization
The configurations of these smelters after modernization are described
in Section 4.3.
4.1 REGULATORY CONSIDERATIONS
The low-arsenic throughput copper smelters are subject to existing
and forthcoming regulations arising from the national ambient air
quality standards (NAAQS) for suspended particulates, sulfur oxides
4-1
-------
and lead; wastewater effluent lin-;Cations; hazardous waste disposal
requirements; and regulations directed at occupational safety and
health. Compliance with some of these regulations may coincidentally
decrease arsenic emissions or affect the manner in which arsenic
emissions are discharged into the atmosphere, even though they were
not necessarily developed with that objective. Either a change in the
quantity of arsenic emitted, or the manner in which arsenic is emitted
(high versus low level discharge of emissions), will necessarily
affect the ambient concentrations of airborne arsenic near these
smelters. Furthermore, the cost of compliance with these regulations
will affect the economic viability of individual smelters and the
industry as a whole, and thus the affordability of arsenic controls.
For the purpose of relating the timings of other regulatory
requirements to the timing of an arsenic NESHAP, January 1986 is
projected as the compliance date with the arsenic NESHAP. The date is
based upon a January 1984 promulgation for the regulation and assumes
that the full 2-year waiver of compliance available under the general
provisions of 40 CFR 61 would be applied for and granted by EPA.
Regulations to be examined in formulating the regulatory baseline
include the following:
• National Ambient Air Quality Standards (NAAQS) under the
Clean Air Act (CAA) for sulfur dioxide, total suspended
particulates, and lead;
• Occupational Safety and Health Administration (OSHA) regulations
for inorganic arsenic;
• Effluent limitation guidelines under the Clean Water Act
(CWA); and
• Hazardous waste disposal regulations under the Resource
Conservation and Recovery Act (RCRA).
It should be noted that there are inherent uncertainties in
estimating the control requirements and compliance dates associated
with these regulations. However, these uncertainties aside, such
estimates are necessary in order to make a reasonable attempt at
assessing the impacts of a possible arsenic NESHAP regulation for
low-arsenic primary copper smelters.
4-2
-------
4.1.1 Clean Air Act
Under Section 109 of the Clean Air Act, EPA has set national
ambient air quality standards (NAAQS) for certain "criteria" pollutants
The criteria pollutants emitted by primary copper smelters are sulfur
dioxide, particulates, and lead. The NAAQS are to be met through the
establishment of State implementation plans (SIP's) governing emission
sources of these pollutants. The plans call for "combinations of
emission limitations and other measures such that the total mix of
these measures would result in the attainment and maintenance of air
quality standards."
The following subsections address each of the three criteria
pollutants emitted by primary copper smelters and the control measures
developed under SIP's.
4.1.1.1 Sulfur Dioxide. The initial SIP requirements for the
control of S02 have undergone considerable reshaping by the courts and
by Congress since first developed by the States and approved by EPA
Congress passed the Clean Air Act Amendments of 1977 which generally
prohibited the use of dispersion techniques in SIP's to meet ambient
standards and required the installation of constant controls for the
meeting of NAAQS as expeditiously as practicable. An exception to the
prohibition of dispersion techniques applied to primary copper, zinc
or lead smelters which were eligible for a primary nonferrous smelter
order (NSO) under Section 119 of the Clean Air Act.2 The NSO permits
a smelter a temporary deferral, up to as late as January 1, 1993 from
compliance with its SIP for S02 emission limitations, and permits the
smelter to meet ambient air quality standards through the use of
constant control technology in combination with interim dispersion-
dependent control techniques including tall stacks and supplementary
control systems. NSO recipients are to be determined on a case-by-case
basis. To be eligible for the deferral, each smelter must, among
other things, pass an economic test by which it demonstrates that it
is economically incapable of implementing constant controls.
Table 4-1 summarizes the applicable SO regulations and the
compliance status of the various smelters. It can be seen that three
smelters are in compliance with the applicable SIP regulation Of
the remaining 11 smelters, six are under NSO's (and/or consent decrees)
4-3
-------
Table 4-1. STATE IMPLEMENTATION PLANS (SIP'S) FOR SULFUR DIOXIDE
AFFECTING COPPER SMELTERS AND COMPLIANCE STATUS
Smelter
ASARCO-E1 Paso, TX
ASARCO-Hayden, AZ
Tennessee Chemical Co.-
Copperhill, TN
Inspiration-Miami, AZ
Kennecott-Garfield, UT
Kennecott-Hayden, AZ
Kennecott-Hurley, NM
Kennecott-McGill, NV
Magma-San Manuel, AZ
Phelps-Dodge-Ajo, AZ
Phelps Dodge-Douglas, AZ
Phelps Dodge-Hidalgo, NM
Phelps Dodge-Morenci, AZ
Copper Range-
White Pine, MI
- — -
aNonferrous smelter orders.
Regulation
131.04.01.004
R9-3-515.C.I.b
1200-3-19-.19
R9-3-515.C.I.d.
4.3.1, 4.3.2, 4.3.4
R9-3-515.C.1.C
652
8.1 and 14
R9-3-515.C.1.3
R9-3-515.C.l.e
R9-3-515.C.l.f
652
R9-3-515.C.l.g
Rule 402
Compliance Status
—_
Out of compliance
Currently under NSO
In compliance
Stack emissions-in compliance
Fugitive emissions-out of
compliance
None-SIP not yet promulgated
Stack emissions-in compliance
Fugitive emissions-out of
compliance
Currently under NSO
Currently under NSO
Currently under NSO
Stack emissions-in compliance
Fugitive emissions-out of
compliance
Currently under NSO
In compliance
Currently under NSO
In compliance
Reference
3
4
5
6
4
7
8
4
4
4
7
4
9
-------
one ,s awaiting promulgation of v. SIP, and four are currently out of
corapl,ance. For the seven Arizona smelters, the new Multi-Point
Rollback limits introduce uncertainty with regard to control equipment
re,™ts. Uncertainties associated with timing for final achievement
o comphance under the NSO program make it difficult to estimate the
effect of the NSO's on regulatory baseline. For the purpose of baseline
analyse, the following assumptions were made with regard to probable
effects of existing S02 regulations on an arsenic NESHAP:
1. The terms of the consent decrees signed by three of the
smelters should bring them into compliance.
2. Because of uncertainties within the NSO program, requirements
for S02 control under this program were not considered in the baseline
3. For the six smelters either out of compliance, or of unknown'
comp,,ance status, it will be assumed that future SO, controls will be
implemented sometime beyond the implementation of arsenic controls
- 'V1:1'2 I0ti1 S"Spe"de'' ''"''"'•"late-. The original date specified
in the Clean Air Act of 1970 for attaining the primary NAAQS for total
suspended particulates was July 1, 1975. That date was extended by
the Clean Air Act Amendments of 1977 to no later than 3 years after
EPA s approval of revised SIP's for areas not in attainment with the
primary NAAQS at the earlier date.
Table 4-2 summarizes the current applicable TSP regulations and
the compliance status of affected smelters. It can be seen that four
smelters are currently in compliance with the SIP's, four are out of
compl,ance, four are under compliance schedules, one is awaiting
promulgation of an SIP, and tne status of the remaining smelter is
unknown.
Total suspended particulate controls are not included in the
base]me of the four smelters which are out of compliance It is
assumed that the four smelters which are under compliance schedules
wi come into compliance within the time frame that an arsenic NESHAP
would be effective, therefore, the regulatory baseline for these
smelters includes the controls to comply with the SIP.
4-5
-------
Table 4-2. STATE IMPLEMENTATION PLANS (SIP'S) FOR TOTAL SUSPENDED
PARTICULATES AFFECTING COPPER SMELTERS AND COMPLIANCE STATUS
Smelter
Regulation
Compliance Status
Reference
CT)
ASARCO-E1 Paso, TX
ASARCO-Hayden, AZ
Tennessee Chemical Co.-
Copperhill, TN
Inspiration-Miami, AZ
Kennecott-Garfield, UT
Kennecott-Hayden, AZ
Kennecott-Hurley, NM
Kennecott-McGill, NV
Magma-San Manuel, AZ
Phelps-Dodge-Ajo, AZ
Phelps Dodge-Douglas, AZ
Phelps Dodge-Hidalgo, NM
Phelps Dodge-Morenci, AZ
111.51, 111.52
40 CFR 52.126(b)
1200-3-7-.02-(3)
40 CFR 52.126(b)
3.2.3
40 CFR 52.126(b)
506
7.2.3
40 CFR 52\126(b)
40 CFR 52.126(b)
7-1-3.6
506
7-1-3.6
Copper Range-White Pine, MI Rules 301 and 331
Out of compliance 3
Under compliance schedule 4
In compliance 5
In compliance 4
None-SIP not yet promulgated 6
In compliance 4
In compliance with mass 7
emission rate, not in
compliance with BAT
Out of compliance 8
Out of compliance 4
Under compliance schedule 4
Out of compliance 4
Under compliance schedule 7
Under compliance schedule 4
Unknown - never tested 9
-------
4.1.1.3 Lead. The national ambient air quality standard for
lead of 1.5 ng/m was promulgated on October 5, 1978 (40 CFR 51 80)
Table 4-3 presents the lead SIP compliance status of the low-arsenic
throughput copper smelters. Both of the smelters which are under
SIP's are currently in compliance. The remaining 12 smelters are in
States in whlch there are either no nonattainment areas for the lead
NAAQS, or in which no lead SIP has been promulgated. For purposes of
baseline analysis, it is assumed that neither existing nor upcoming
lead SIP's will affect arsenic air emissions from the smelters under
consideration, or cause a significant economic burden concurrent with
an arsenic NESHAP.
4>l-2 iiiftiif^^
On May 5, 1978. the U.S. Occupational Safety and Health Administration
(OSHA) promulgated standards for occupational exposure to inorganic
arsenic. The standards limit occupational exposure to 10 Mg/m3
averaged over any 8-hour period. The regulations require the use'of
engineering and work practice controls. In the event that engineering
controls are not sufficient to reduce exposures to or below the standard
the engineering controls must be used to reduce exposures to the
lowest level achievable and should be supplemented by the use of
respirators and other necessary personal protective equipment. Primary
copper smelters were required to monitor for violations and submit a
compliance plan to OSHA by December 1, 1978.
The OSHA requirement is a concentration limit and does not specify
he sources to be controlled or the equipment to be utilized in control
Compliance plans with OSHA will require the application of capture '
systems on various fugitive sources, in addition to worker protection
programs such as the use of respirators, protective work clothing
improved housekeeping practices, hygiene facilities and practices,' and
medical surveillance. ASARCO has recently negotiated tripartite
agreements with OSHA and the United Steelworkers of America which
specify controls and work pactices designed to decrease worker exposure
to inorganic arsenic at the three ASARCO smelters.11'12'13 The E1
Paso agreement specifies the installation of local ventilation hoods
for slag tapping. This requirement is incorporated into the baseline
4-7
-------
Table 4-3. STATE IMPLEMENTATION PLANS (SIP'S) FOR LEAD
AFFECTING COPPER SMELTERS AND COMPLIANCE STATUS
i
CO
Compliance Status
ASARCO-E1 Paso, TX
ASARCO-Hayden, AZ
Tennessee Chemical Co.-
Copperhill, TN
Inspiration-Miami, AZ
Kennecott-Garfield, UT
Kennecott-Hayden, AZ
Kennecott-Hurley, NM
Kennecott-McGill, NV
Magma-San Manuel, AZ
Phelps-Dodge-Ajo, AZ
Phelps Dodge-Douglas, AZ
Phelps Dodge-Hidalgo, NM
Phelps Dodge-Morenci, AZ
Copper Range-
White Pine, MI
No SIP-Currently
under study
No SIP
1200-3-22-.03
No SIP
SIP recently approved
No SIP
No SIP
Article 12
No SIP
No SIP
No SIP
No SIP
No SIP
No SIP
Regulation currently in
proposal phase
In compliance
In compliance
4
5
4
6
4
7
8
4
4
4
7
4
9
-------
for the El Paso smelter. Collect-;un of the captured slag tapping
emissions is not specified in the OSHA compliance plan. The Hayden
agreement specifies that all engineering controls currently in
existence which limit occupational arsenic exposure be used, but does
not require the installation of any new fugitive emissions capture
systems at ASARCO-Hayden. OSHA has also negotiated compliance plans
with two Kennecott smelters (Garfield and McGIll), but these agreements
have not yet been signed, nor is signature imminent, and OSHA has no
plans to negotiate compliance plans with any other copper smelters at
this time. Therefore, capture of slag tapping fugitive emissions at
ASARCO-E1 Paso is the only OSHA requirement which must be incorporated
into the regulatory baseline of an arsenic NESHAP.
4.1.3 Clean Water Act
Water pollution control regulations affect all areas of the
primary copper industry, including mining, milling, smelting, refining,
and acid plants. These regulations do not affect arsenic air emissions
The discussion of water pollution control regulations under the Clean
Water Act is divided into those which are based on the best practicable
control technology (BPT) requirements [Section 301(b)(l)(A)], and
those which are based on the best available technology (BAT)'and best
conventional pollution control technology (BCT) [Section 301(b)(2)]
Potential water pollution originating from the low-arsenic throughput
primary copper smelters will be regulated by EPA's proposed effluent
limitations guidelines for nonferrous metals manufacturing.15 Effluents
from the smelters will be covered by two subcategories: primary
copper smelting and metallurgical acid plants. The proposed regulations
for the primary copper smelting subcategory will amend promulgated BAT
(40 FR 8523) to conform BAT to promulgated BPT (45 FR 44926), which is
more stringent. The proposed metallurgical acid plant regulations set
forth BAT effluent mass limitations for metallurgical acid plants
including copper smelter acid plants, based on promulgated BPT (45 FR
44926).
4<1'4 Resource Conservation and Recovery Act (RCRA)
On May 19, 1980, EPA promulgated regulations under the Resource
Conservation and Recovery Act for the disposal of hazardous substances
from metallurgical and other process industries (45 FR 33066). EPA
4-9
-------
identified the acid plant blowdr n slurry/sludge resulting from the
thickening of blowdown slurry at primary copper smelters as a hazardous
waste because of its lead and cadmium content. However, the extent to
which acid plant blowdown will be subject to regulations for treatment
and disposal is unknown. No regulations under RCRA will affect arsenic
air emissions, however, and it is unlikely that any of the smelters
will suffer a significant economic impact as a result of RCRA regulations.
4.2 BASELINE AND REGULATORY ALTERNATIVES
This section defines in general terms the baseline employed for
the impacts analysis in succeeding chapters of this document, and
presents the regulatory alternatives along with the technology selected
for each alternative.
4.2.1 Definition of Baseline
For the purposes of the analysis presented in this document,
"baseline" includes only those existing air pollution controls which
were installed as of January 1983 or were required by existing legal
actions. Three smelters, (Phelps Dodge-Ajo and -Morenci, and ASARCO-Hayden)
currently are operating under consent decrees ' to install new
furnace configurations and acid plant control equipment for S02 emissions.
Final compliance dates for these consent decrees have been established
as 1985. Since these controls are mandated by the courts and final
compliance will be in the near future, the "consent, decree configurations"
were treated as though they were existing configurations, and were
incorporated into the baseline.
The Kennecott-Hurley smelter is currently undergoing voluntary
upgrading of the furnace configuration and the installation of acid
plant controls. This work has already begun and is expected to be
completed in the same time frame as the smelters operating under a
consent decree.18 For this reason, the smelter configuration after
modernization was taken as the baseline configuration.
4.2.2 Description of the Regulatory Alternatives
The four regulatory alternatives represent application of inorganic
arsenic controls on various emission points at the smelters and are
characterized by the control equipment that would be required to meet
these levels of control. Alternatives II, III, and IV each include
4-10
-------
baseline controls but are others not additive, e.g., Alternative HI
does not include what was required by Alternative II, etc
Alternative II would require the control of process arsenic
emissions. As discussed in Section 3.1.1.1. the arsenic collection
efficiency of any paniculate control device Is a function of temperature
and the correspond arsenic saturation concentration, and the .Lie
sZ!TcT,B ;f the particuiar 9as stream- AS «- * ^ ^ ««
ample calculation presented in Section 3.1.1.1, fop strearas Wlth
lower arsenic concentrations, cooling the gas stream prior to its
entenng a particulate control device would have no effect on arsenic
removal efficiencies. For these smelters, Alternative I, would not
require installation of additional control equipment beyond the baseline
configuration. For those suiters with higher arsenic concentration
m process gas streams, control would be effected by operating a
paniculate control device at less than 121°c (250°F), with flue gas
cooling upstream of the control device.
Alternative III would require the capture and collection of
fugitive arsenic emissions fro™ converter operations. This alternative
is based on the use of an air curtain secondary hood or equivalent for
the capture of fugitive emissions f™ the converters, and a
con re, d 1ce (baghou$e or ^^ ^ ^ «*
of the captured emissions.
Alternative IV would require the capture and collection of fugitive
missions fro. calcine discharge, «tte tapping, and slag tapping
operations Under this alternative the capture of the fugitive lissions
™ point '
a emlSS1°''S "e1"g ™tSi t0 " P«rt'«l«te control device
(baghouse or equivalent technology) for collection
Alternative V would require the elimination of all arsenic emission,
at copper smelters. To .cc«p,ish this alternative the sme t s™
e forced to process ores which were virtually free of arsenic con
This regulatory alternative would therefore require the closure of '
existing primary copper smelters.
4-11
-------
4.3 BASELINE CONFIGURATION, BAScLINE ARSENIC EMISSIONS, AND REGULATORY
ALTERNATIVES FOR MODEL PLANTS
The diversity of operational parameters among the 14 low-arsenic
throughput copper smelters makes it impractical to define "generic"
model plants. Therefore, 14 model plants were chosen, representing
the baseline configurations, as defined in Section 4.2.1, of the
14 low-arsenic copper smelters. Section 4.3.1 presents arsenic emission
estimates, process and fugitive, for the model plant baseline configurations,
while Section 4.3.2 is a plant-by-plant description of the baseline
configurations along with a summary of what would be required of each
model plant under each of the regulatory alternatives.
4.3.1 Baseline Arsenic Emissions
This section presents arsenic emission estimates for the baseline
configuration of each model plant.
4.3.1.1 Process Emissions. Table 4-4 presents process arsenic
emission estimates for the model plant baseline configurations, after
control. As discussed in Section 2.0, process equipment exhaust gas
arsenic content was read from the arsenic material balances supplied
in Appendix F. Arsenic removal efficiency of control devices at each
of the 14 existing smelters was taken either from the company-supplied
material balances, by comparing indicated inlet arsenic loading to
indicated emissions, or from the following control efficiencies determined
during EPA's performance evaluations of various types of particulate
control devices. "Cold" devices (121°C, 250°F or below, inlet) were
deemed to be 96 percent efficient, while "hot" (above 150°C, 300°F,
inlet) control devices were assigned an arsenic removal efficiency of
30 percent. Acid plants, due to the requirement for extensive gas
precleaning and conditioning, were ascribed a control efficiency of
99 percent.
4.3.1.2 Fugitive Emissions. Table 4-5 presents fugitive arsenic
emissions estimates for the model plant baseline configurations.
Emission rates in absence of control were taken from Table 2-17. At
3 of the 14 smelters, control of one or more of the fugitive gas
streams is practiced. Details of baseline capture and control of
fugitive emissions are provided in the plant-by-plant discussion which
follows.
4-12
-------
,-onMruuj
Luw-HfOtNR IHROUGHPUT
Sniel ter
—
ASARCO-E1 Paso
ASARCO-Hayden
Tennessee Chemical Co -
Copperhil)
Inspiration-
Miami
Kennecott-Garfield
Konnecott-llayden
-£=>
J_, Kennecott-Hurley
OJ
KcnnecoU-NcGIll
Magma-San Manuel
Phelps Dodye-Ajo
Phelps Dodge-
Doug las
Phelps Qodge-
Illdalgo
V helps Dodge-
llorenci
Copper Range-
Uhite Pine
Sources: (TO - MuHiht;ar
Emission
Source
Him
REV
CONV
ff
CONV
F8R
£F
CONV
EF
CONV
NOR
CONV
FBR
REV
CONV
FF
CONV
REV
CONV
REV
CONV
OXREV
CONV
MHR
REV
CONV
FF
COflV
OXREV
CONV
REV
CONV
tb Roaster
— —-•
Exit Gas
Arsenic Content
kg/hr
86 ti
54.9
76.2
45.0
0.2
0.4
0.5
4.5
1.5
115.0
1.9
5.3
5.0
0.4
18.6
35.6
1.3
0.5
45.3
2.3
l.f
0.9
0.9
5. )
0 4
0.?
Mili - Fluid
i itjtii-iMi L,urrcr\ ont
Control
Type
~ _
1. ESP, 2. Acid Plant
Cold ESP
1. ESP, 2. Acid Plant
1. ESP, 2. Acid Plant
1. ESP, 2. Acid Plant
1. Scrubber, 2. ESP,
3. Acid Plant
1. Scrubber, 2. ESP
3. Add Plant
1. Scrubber, 2. ESP,
3. Acid Plant
1. ESP, 2. Acid Plant
1- ESP, 2. Acid Plant
1. Cyclones, 2. ESP's
3. Acid Plant
1- ESP's, 2. Acid Plant
1. Scrubber, Z. Acid Plant
I. ESP, 2. Scrubber
3. Acid Plant
1- ESP, 2. Acid Plant
1. ESP, 2. Acid Plant
ESP
Mul ticlones
ESP
1- ESP, 2. Acid Plant
1. ESP, 2. Acid Plant
1- ESP, 2. Acid Plant
LSP
FSP
1. ESP, 2. Acid Plant
1- ESP, 2. Acid Plant
1- ESP, 2. Acid Plant
1. ESP, 2. Acid Plant
R.sl loon Flue
lied Roastpr
L 1 tKo
Equipment
Efficiency
1. 60, 2. 99
97.5
1. 96, 2 99
1. 96, 2. 99
1 97 2 95
1- 50, 2. 100, 3. 99
1- 50, Z. 100, 3. 99
I. 50, 2. 100, 3. 99
1. 30, 2. 99
1. 30, 2 99
1. 20, 2. 60, 3. 95
1 60 2 9^
1. 90, 2. 99
40
1. 40, 2. 90, 3. 99
1. 75, 2. 99
1 25 2 99
30
7
30
1-30 2 99
1. 25, 2. 90
1. 25 2 90
30
30
30
1. 30, 2. <)8
I 30 2 98
1 30 2 98
1 30 2 98
70
60
. _
Overall Control
Efficiency
(percent)
99.9
97.5
99.9
99.9
99.9
100
100
100
99.4
98.9
98.0
98.0
99.9
40
99.9
99.9
99.9
30
7
30
99.2
92.0
92.0
30
30
30
99.9
99.9
99.9
99.9
70
Emission
Rate
kg/hr
0.04
1.4
0.03
0.16
0.04
0
0
0
0.03
0.02
1.7
0.1
0.02
3.2
0.03 '
0.0004
0.0001
12.8
33.0
1.3
0.003
3.5
0.2
1 5
0.7
2.2
0.2
0.01
0.06
0.08
0.1
0.1
CONV - Converters
FF - Flash 'urnate
El - Electric Furnace
flOK - Noran.lj Reactor
OXREV - Oxygen-Sprinkle Reverberatory Furnace
-------
Table 4-5. SUMMARY OF BASELINE FUGITIVE ARSENIC EMISSION ESTIMATES FOR LOW-ARSENIC
THROUGHPUT PRIMARY COPPER SMELTERS
-pa
Sine) ter
ASARCO-E1 Paso
ASARCO-Hayden
Tennessee Chemical Co.-
Copperhi 11
Inspiration-
M i am i
Kennecott-Garf i eld
Kennecott-Hayden
Kenriecott-Hurley
Kennecott-McGil 1
Emission
Source
CT
ST
HT
CONV
AF
FDH
ST
MT
CONV
AF
FDM
ST
IIT
CONV
FDH
ST
MT
CONV
AF
FDH
ST
MT
CONV
AF
FDH
ST
IIT
CONV
AF
FDH
ST
HT
CONV
AF
FOH
ST
MT
CONV
AF
FDH
Emlss ion
Rate, Absence
of Control
kg/hr
0.07
0.05
0.7
11.4
1.3
0.26
0.1
1.5
6.8
2.2
0.2
0.0004
0.01
0.08
0.0006
0.01
0.08
0.23
0.08
0.002
O.OJ
0.2
0.9
1.5
0.2
0.002
0.11
0.0
0.01
0.001
0.001
0.014
0.05
0.02
0.002
0.02
0.5
5.3
0.4
0.02
Control
Equ 1 pmen t
Cold ESP
-
Baghouse
Baghouse
Baghouse
-
Cold ESP
Cold ESP
Cold ESP
-
-
-
-
-
-
-
-
-
-
-
_
-
-
-
-
_
-
-
-
-
_
-
_
-
-
_
-
-
-
-
Control
Capture
90
90
90h
75b
75
0
90
90
50
0
0
90
90
0
0
0
90
0
0
0
90
90
50
0
0
90
90
0
0
0
0
90
0
0
0
90
90
0
0
0
Efficiency
(Percent)
Collection Overall
97.5
0
97.5
96
96
0
96
96
96
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
87.8
0
87.8
72.0
72.0
0
86.4
86.4
48.0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Emission
Rate
kg/hr
0.009
0.05
0.09
3.2
0.4
0.26
0.01
0.2
3.5
2.2
0.2
0.0004
0.01
0.08
0.0006
0.01
0.08
0.22
0.08
0.002
0.03
0.2
0.9
1.5
0.2
0.002
0.11
0.8
0.01
0.004
0.001
0.014
0.054
0.018
0.002
0.02
0.5
5.3
0.4
0.02
-------
Table 4-5. SUMMARY OF BASELINE FUGITIVE ARSENIC EMISSION ESTIMATES FOR LOW-ARSENIC
THROUGHPUT PRIMARY COPPER SMELTERS (concluded)
-Fa
I
Oi
Smel ter
Magma- San Manuel
Phelps Dodge-Ajo
Phelps Dodge-
Douglas
Phelps Dodge-
Hidalgo
Phelps Dodye-
Morenci
Copper Range-
White Pine
Sources- CT - Calc
ST - Slag
Emi sr> ion
Sourr P
ST
rvr
CONV
AF
FDD
ST
MT
CONV
AF
ron
CT
ST
HT
COHV
AF
FDII
ST
MT
CONV
AF
FDII
ST
MT
CONV
AF
(Oil
ST
IIT
CONV
AF
FOH
ine Transfer
Tapping
Emi ss ion
Rate, Absence
of Control Control
kg/hi Equipment
0.0004
0.01
0.06
0.006
0.0007
0.01
0.06
0.3
0.1
0.01
0.005 Baghouse
0.006 ESP
0.06 ESP
0.5
0.02
0. 002
0.003
0.02
0.14
0.07
0.001
0.004
0.1
0.8
0.06
0.009
0.0002
0. 006
0.04
0.02
0.004
HT - Matte Tapping
CONV - Converter Operations
Control
Capture
90
90
0
0
0
90
90
30
0
0
90
90
90
0
0
0
90
90
30
0
0
90
90
30
0
0
90
90
0
0
0
AF - Anode
Efficiency
Collects
0
0
0
0
0
0
0
0
0
0
94.9
30
30
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Furnace
FOH - Flue Dust Handlin
_LPercent)
on Overall
0
0
0
0
0
0
0
0
0
0
85.4
27
27
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
g, Transfer,
Emission
Rate
kg/hr
0.0004
0.01
0.06
0.006
0.0007
0.01
0.06
0.3
0. 1
0.01
0.0007
0.004
0.04
0.5
0.02
0.002
0.003
0.02
0.14
0.07
0.004
0.004
0.1
0.8
0.06
0.009
0.0002
0.006
0.04
0.02
0.0004
and Conveying
Captured by building evacuation system.
-------
4.3.2 Model Plant Baseline ConfDurations and Regulatory Alternatives
The following model plant descriptions present the baseline
configuration, as defined in Section 4.2.1, for the 14 low-arsenic throughput
copper smelters. The configurations described in this section were
those used in calculating baseline arsenic emissions, and serve as the
starting point for subsequent analyses of the incremental environmental,
energy, and economic impacts of the regulatory alternatives.
4.3.2.1 ASARCO-E1 Paso. The baseline configuration is presented
in Figure 4-1. Offgases from the 4 multihearth roasters pass through
a spray chamber and electrostatic precipitator before treatment in an
acid plant. Offgases from the single reverberator,/ furnace enter a
spray chamber, followed by passage through an electrostatic precipitator
(operated at 121°C, 250°F) and discharge from a stack. Offgases from
the three converters are treated in a spray chamber-electrostatic
precipitator for particulate removal and a sulfuric acid plant for SO-
control. Calcine discharge fugitive emissions are captured by a local
ventilation system and routed to the reverberatory furnace spray
chamber. Slag tapping fugitives are presently neither captured nor
controlled at the El Paso smelter, but a recent agreement with OSHA11
will require installation of a capture system. Thus, for purpose of
defining baseline, a slag tap fugitive emission capture system is
assumed to be installed. Matte tapping fugitive emissions are captured
and ducted to the reverberatory furnace spray chamber. Converter
fugitives are captured by the building evacuation system and collected
in a baghouse. It is estimated that the building evacuation capture
efficiency is 75 percent, and that the building evacuation baghouse
particulate collection efficiency is 96 percent.
Regulatory Alternative II would require no controls beyond baseline
at ASARCO-E1 Paso, since baseline process controls represent best
technology for arsenic control.
Regulatory Alternative III would require installation of air
curtain secondary hoods for capture of converter fugitive emissions.
The existing building evacuation baghouse is assumed to be used for
collecting the captured emissions.
Regulatory Alternative IV would require collection of captured
slag tapping fugitive emissions in an effective particulate matter
4-16
-------
Roaster/Acid
Plant Stack
ACID PLANT
121°
ESP
I
I—'
~-J
Captured
slag tap
fugitives
35,000 acfm
25° C
\
Captured
calcine
fugitives
ROASTERS
\
i ,Revet b St at k
813,900 a
-------
control device. For the purpose oT analysis, it is assumed that the
existing reverberatory furnace electrostatic precipitator will suffice.
4.3.2.2 ASARCO-Hayden. The baseline configuration (according
to the consent decree, see Section 4.2.1) is shown in Figure 4-2.
Process offgases from the INCO flash furnace are cleaned in a spray
chamber, settling chamber, and electrostatic precipitator before
treatment in an acid plant. Offgases from the five converters pass
through cyclones (three per converter), a spray chamber, and electrostatic
precipitator before treatment in an acid plant. Matte and slag tap
fugitive emissions are captured by a local ventilation system and
controlled in a cold electrostatic precipitator. Converter fugitives
are captured in a secondary hood system (estimated to be 50 percent
efficient) and controlled in a cold electrostatic precipitator.
Arsenic vapor pressure calculations for the fugitive emission
control device show it to be equally efficient for arsenic and particulate
collection at the given temperature and gas stream arsenic concentration.
Regulatory Alternative II would require no controls beyond baseline
for ASARCO-Hayden, where the baseline configuration includes best
technology for arsenic control.
Regulatory Alternative III would require installation of air
curtain secondary hoods on the converters, and collection of captured
emissions in the cold electrostatic precipitator used for fugitive
emission collection in the baseline configuration.
Regulatory Alterantive IV would require no controls beyond baseline
for ASARCO-Hayden, since baseline matte and slag tapping fugitive
emission controls to be installed at the smelter are considered best
technology.
4.3.2.3 Tennessee Chemical Co.-Copperhill. Figure 4-3 shows
the baseline configuration for the Tennessee Chemical Company smelter.
Offgases from the fluid-bed roaster pass through a cyclone and, combined
with offgases from the electric furnace and the two converters, are
treated in either of two acid plants following standard gas cleaning.
Matte and slag tapping fugitive emissions are captured by localized
ventilation systems and vented to a stack. Converter fugitives are
neither captured nor controlled.
4-18
-------
Main Stack
576,000 acfm
66° C
70.000 acfm
666 C
ACID PLANT
ESP
120° C
INCO FLASH
FURNACE
(1)
Captured
matte/slaq tap
Bypass
79,500 acfm
121° C
316,000 acfm
258 C
ESP
43,700 acfm P 25° C
Matte
Captured
converter
fugitives
273,000 acfm 0 256 C
118,000 acfm
66° C
Acid Plant
137,000 acfm
121° C
216,500 acfm
121° C
ESP
200° C
Converters
(5)
Figure 4-2. ASARCO-Hayden Smelter Baseline Configuration
-------
iTo stack
(114,000 acfm
lee0 c
Acid Plant
ATo stack
1114,000 acfm
166° C
Acid Plant
ESP
ESP
Cyclone
Scrubbers
Fluid Bed
Roaster
Electric Furnace
iTo roof
Captured matte/
slag tap fugitive:
Matte
Converters
(2)
Figure 4-3. Tennessee Chemical Company
Smelter Baseline Configuration
4-20
-------
Since process controls at the Copperhill smelter represent best
technology for arsenic, Alternative II would require no additional
controls beyond baseline.
Alternative III would require installation of air curtain secondary
hoods on converters and a control device for collection of captured
converter fugitive emissions.
Installation of a collection device for captured matte and slag
tapping fugitive emissions would be required under Alternative IV.
4.3.2.4 Inspiration-Miami. The baseline configuration for
Inspiration-Miami is depicted in Figure 4-4. Process emissions from
both the single electric furnace and the five Hoboken converters pass
through convection coolers, cyclones, and electrostatic precipitators
before treatment in an acid plant. Fugitive emissions from slag
tapping are neither captured nor controlled. Matte tapping emissions
are captured by local ventilation and vented from the roof of the
furnace building. Fugitive emissions from the converters are neither
captured nor controlled.
Regulatory Alternative II would require no controls beyond
baseline, since the baseline configuration represents best control of
process arsenic emissions.
Alternative III would require installation of air curtain secondary
hoods on the converters, and a particulate control device for collection
of captured converter fugitive emissions.
Alternative IV would necessitate installation of a capture system
for slag tapping fugitive emissions as well as a collection device for
captured matte and slag tapping fugitive emissions.
4.3.2.5 Kennecott-Garfield. Kennecott-Garfield's baseline
configuration is shown in Figure 4-5. Combined process offgases from
the three Noranda reactors and the four converters pass through electrostatic
precipitators (6 in series) and are then sent to any of four acid
plants. Matte tap and slag tap fugitive emissions are captured by
local ventilation systems and ducted to the main stack. Converter
fugitives are captured by a secondary hooding system (assumed to
50 percent efficient) and vented to the main stack. In addition, the
reactor and converter buildings are each ventilated by a fugitive
emission roof capture system designed to handle emissions which have
4-21
-------
ESP
Electric
Furnace
(1)
To roof
\
Captured
matte
tap
fugitives
Matte
Main Stack
111,000 acfm
54° C
Acid Plant
ESP
(3)
52,000-60,400 acfm
496° C
Hoboken
Converters
(5)
Figure 4-4. Inspiration-Miami Smelter Baseline Configuration
4-22
-------
654,000 acfm
693 °C
Noranda
Reactors
(3)
Main Stack
1,000,000 acfm
65 °C
Acid Plants
:Captured
•matte/slag
:tap fugitives--
Ilocal ventilation
•147,000 acfm
i256 C
X 315° C
" 299,700 acfm
104 °C
V
ESPs
(6)
:Reactor roof
: fans
:65,OOU acfm
i25d C
•*-'2 per reactor
Matte
114,700 acfm
693 °C
Converters
(4)
Captured
converter
fugitives--
secondary hood
70,000 acfm
25° C per
converter
converter roof
fans
65,000 acfm
25° C
2 per
converter
Figure 4-5. Kennecott-Garfield Smelter Baseline Configuration
4-23
-------
escaped capture by the primary ^js handling system or the capture
ports near the hot metal transfer areas. The system consists of two
65,000 scfm axial I.D. fans mounted on the roof above each reactor
vessel and each converter. The captured gas is vented through the
smelter main stack.
No controls beyond the baseline configuration would be required
under Alternative II since the baseline process controls are already
best control for arsenic.
Air curtain secondary hoods would be installed under Alternative III,
as would be an effective particulate control device for collection of
captured converter fugitives.
Regulatory Alternative IV would necessitate installation of an
effective particulate control device for collection of captured matte
and slag tapping fugitive emissions.
4.3.2.6 Kennecott-Hayden. The baseline configuration of the
Kennecott-Hayden smelter is depicted in Figure 4-6. Offgases from the
fluid bed roaster go to a venturi scrubber and a tray-type scrubber in
series. The gases are then combined with converter gases upstream of
a sulfuric acid plant. Process emissions from the single reverberatory
furnace are treated in an electrostatic precipitator (operated at
260°C, 500°F) before discharge from the main stack. Matte and slag
tapping fugitive emissions are captured with local ventilation systems
and discharged above the building roof without control. Converter
fugitives are neither captured nor controlled.
Regulatory Alternative II would require no controls beyond those
of the baseline configuration for the roaster and converter process
offgas streams since these are already treated by best technology for
arsenic control under baseline. The furnace electrostatic precipitator,
which is "hot", is not considered best technology, but arsenic vapor
pressure data (see Section 3.1.1.1) indicate that flue gas cooling
would yield no additional emission reduction. Therefore, no additional
controls for furnace process emissions would be required under
Alternative II.
Alternative III would require the installation of air curtain
secondary hoods on the converters, and a particulate control device
for collection of captured converter fugitive emissions.
4-24
-------
Scrubber
18,000 acfm
329° C
Cyclones
66,600 acfm
566° C
Fluid Bed
Roaster
(1)
Calcines
To Acid Plant Stack
100,000 acfm
79° C
Acid Plant
To roof
42,600 acfm
25° C
Furnace Stack
130,000 acfm
260° C
ESP
Reverberatory
Furnace
(1)
Scrubber
Captured matte/ ''
slag tap
fugi tives
ESP
Matte
226,100 acfm
621° C
Converters
(3)
Figure 4-6. Kennecott-Hayden Smelter Baseline Configuration
4-25
-------
Alternative IV would necessitate installation of a collection
device for the captured matte and slag tapping fugitive emissions.
4.3.2.7 Kennecott-Hurley. Figure 4-7 shows the baseline configuration
(after planned modernization, see Section 4.2.1) of the Kennecott-Hurley
smelter. Process emissions from the INCO flash furnace are cleaned in
an electrostatic precipitator before entering an acid plant. Offgases
from the four converters pass through three parallel electrostatic
precipitators before being treated in an acid plant. Matte tap fugitive
emissions are captured by a local ventilation system and vented to the
main stack. Slag tap and converter fugitives are neither captured nor
control led.
No additional controls would be required by Alternative II for
this smelter since in the baseline configuration both process gas
streams are controlled by best technology for arsenic.
Alternative III would require air curtain secondary hoods on the
converters, and the installation of a control device for collection of
captured converter fugitive emissions.
Alternative IV would require a capture system for slag tapping
fugitive emissions and the installation of a particulate collection
device to treat captured matte and slag tapping fugitives.
4.3.2.8 Kennecott-McGill. The baseline configuration of the
Ktnnecott-McGil1 smelter is presented in Figure 4-8. Process emissions
from the two reverberatory furnaces pass through an electrostatic
precipitator (operated at 316°C, 600°F) before exiting the main stack.
Converter offgases from any of the four units are treated in rnulticlones
before being ducted to the stack. Matte and slag tap fugitives are
captured by a local ventilation system and vented to the building
roof. Fugitive emissions from the converters are neither captured nor
controlled.
However, arsenic vapor pressure calculations (see Section 3.1.1.1)
for the furnace offgases show that no additional arsenic emission
reduction would result if the inlet temperature to the electrostatic
precipitator were reduced to as low as 121°C (250°F). Therefore,
Alternative II would have no effect on this gas stream. Similar
calculations for the converter gas stream show that 57 percent arsenic
collection would occur at 121°C (250°F). Alternative II therefore
4-26
-------
Furnace Acid Plant
Stack
34,720 acfm
79 °C
Acid Plant
204 °C
ESP
Inco
Flash Furnace
(1)
Old Furnace
Stack
4,000 acfm
25° C
Captured
matte tap
fugitives
Matte
4 Converter Acid Plant Stack
88,600 acfm
79 °C
Acid Plant
ESP
(3)
371 °C
Air-to-Gas
Heat Exchanger
104,600 acfm/converter operating
538 - 649 °C
Converters
(4)
Figure 4-7. Kennecott-Hurley Smelter Baseline Configuration
4-27
-------
ESP
414,500 acfm
316° C
Reverberatory
Furnaces
(2)
' To Roof
\
\
75,000 acfm
Captured
matte/slag
tap
fugitives
Matte
Main Stack
750,000 acfm
150° C
Multiclones
164,200 to 413,000 acfm
427° C
Converters
(4)
Figure 4-8. Kennecott-McGill Smelter Baseline Configuration
4-28
-------
would require flue gas cooling upstream of an effective particulate
control device for the converter process offgases.
Alternative III would require the installation of air curtain
secondary hoods on the converters, and a control device for collection
of captured converter fugitive emissions.
Alternative IV would necessitate installation of a control device
for collection of captured matte and slag tapping emissions.
4.3.2.9 Magma-San Manuel. Figure 4-9 depicts the baseline
configuration of the Hagma smelter. Offgases from the three reverberatory
furnaces pass through an electrostatic precipitator (operated at
260°C, 500°F) before discharge from the main stack. Process emissions
from the six converters are also treated in an electrostatic precipitator,
and then routed to an acid plant. Matte and slag tap fugitive emissions
are captured and vented from the roof of the furnace building. Converter
fugitives are neither captured nor controlled.
Since the baseline configuration reflects best control of arsenic
in the converter process gas stream, no additional controls would be
required under Regulatory Alternative II. Arsenic vapor pressure
calculations (refer to Section 3.1.1.1) for the furnace gas stream
indicate that no additional arsenic collection would be achieved by
flue gas cooling to 121°C (250°F), therefore, no additional controls
beyond baseline would be required under Alternative II.
Converter fugitive emissions would be captured by air curtain
secondary hoods under Regulatory Alternative III. An effective control
device would be required for collection of the captured emissions.
Alternative IV would cause the installation of a control device
to collect the captured matte and slag tapping emissions.
4.3.2.10 Phelps Dodge-Ajo. The baseline configuration (after
consent decree modification, see Section 4.2.1) of the Phelps Dodge-Ajo
smelter is presented in Figure 4-10. Process emissions from the
oxygen-sprinkle modified reverberatory furnace will pass through a hot
electrostatic precipitator before being treated in the single acid
plant. Offgases from the three converters pass through two electrostatic
precipitators before entering the acid plant. Matte and slag tapping
fugitive emissions are captured by localized ventilation and vented to
the main stack. Converter fugitives are captured by a secondary hood
system estimated to be 30 percent efficient and vented to the stack.
4-29
-------
Main Stack
400,000 acfm
246 °C
ESP
517,500 acfm
260 °C
Reverberatory
Furnaces
(3)
To roof
215,000 acfm
25° C
Captured
matte/slag
tap
fugitives
Matte
To 2 Acid Plant
Stacks
187,000 acfm
52° C
Acid Plant
ESP
311 ,800 acfm
232 °C
Converters
(6)
Figure 4-9. Magma-San Manuel Smelter Baseline Configuration
4-30
-------
30,000 acfm
25° C
16,800 acfm
310° C
ESP
Oxy-Sprinkle
Reverberatory
Furnace
(1)
iMain Stack
175,000 acfm
.Jj!0°.C
Acid Plant
Captured
matte/slag
tap
fugitives
Matte
74,400 acfm
288° C
ESP
(2)
Converters
(3)
70.000 acfm
25* C
Captured
converter
fugitives
Figure 4-10. Phelps Dodge-Ajo Smelter Baseline Configuration
4-31
-------
Alternative II would impose no process controls beyond those in
the baseline configuration since they represent best technology for
arsenic control.
Alternative III would require air curtain secondary hoods on
converters, and the installation of a control device to collect the
captured converter fugitives.
Alternative IV would necessitate installation of a particulate
control device to collect the captured matte and slag tapping emissions.
4.3.2.11 Phelps Dodge-Douglas. Figure 4-11 depicts the baseline
configuration of the Phelps Dodge-Douglas smelter. Process offgases
from the 24 multi-hearth roasters and three reverberatory furnaces are
cleaned in electrostatic precipitators (operated at 260°C, 500°F, and
232°C, 450°F, respectively) oefore discharge out of the roaster/reverb
stack. Process emissions from the five existing converters also pass
through an electrostatic precipitator (operated at 177°C, 350°F)
before stack discharge. Calcine discharge fugitive emissions are
captured by a local ventilation system and sent to a baghouse. Matte
and slag tap fugitives are captured by localized ventilation and
ducted to the converter electrostatic precipitator. Converter fugitive
emissions are neither captured nor controlled.
Arsenic vapor pressure calculations (see Section 3.1.1.1) show
that under Alternative II, no additional reduction in process arsenic
emissions occurs as a result of flue gas cooling (to 121°C, 250°F)
upstream of the control devices. Therefore, Alternative II requires
no added process controls beyond the baseline configuration.
Alternative III would require air curtain secondary hoods on all
converters along with installation of a control device to collect
captured converter fugitive emissions.
Alternative IV would not affect calcine transfer fugitive emissions,
since in the baseline configuration these emissions are controlled in
a baghouse. Alternative IV would require installation of an additional
control device for collection of captured matte and slag tapping
fugitive emissions.
4.3.2.12 Phelps Dodge-Hidalgo. The baseline configuration of
the Hildalgo smelter is shown in Figure 4-12. Offgases from the
Outokumpu flash furnace are treated in three parallel electrostatic
4-32
-------
536,300 afcm
ESP
260° C
Multi-hearth
Roasters
(24)
Main Stack
290,000 acfm
207° C
305,000 acfm
ESP
232° C
Balloon
Flue
Baghouse
'Calcine discharge
fugi tives
Talrines ^
Reverberatory
Furnaces
(3)
Converter Stack
237,000 acfm
177° C
ESP
343° C
Balloon
Flue
Slag/matte
tap fugitives
40,000 acfm
25d C
Matte
Converters
(5)
Figure 4-11. Phelps Dodge-Douglas Smelter Baseline Configuration
4-33
-------
ESP
(3)
1200°C Captured
slag/matte
tap fugitives
56,300 acfm
25°CJ
Flash Furnace
Slag
Acid Plant Stack
36,100 acfm
79° C
Acid Plant
Slag Furnace
Stack
333,000 acfm
128° C
Scrubber
T
Matte
160,000 acfm
371° C
Slan Furnace
ESP
(2)
•Captured
^Converter
•Fugi tives
ill7,000 acfm
J25°
Matte
Converters
(3)
Figure 4-12. Phelps Dodge-Hidalgo Smelter Baseline Configuration
4-34
-------
precipitators before being mixed with emissions from the three converters
which have passed through two parallel electrostatic precipitators.
Combined process offgases are then treated in either of two acid
plants. Slag furnace process emissions are treated in a venturi
scrubber and discharged from a separate stack. Matte and slag tap
fugitive emissions are captured by local ventilation systems and
vented to the slag furnace stack. Converter fugitives are captured by
secondary hoods (estimated to be 30 percent efficient) and ducted to
the slag furnace stack.
Alternative II would require no additional controls beyond the
baseline configuration, since the baseline controls represent best
technology for arsenic removal.
Alternative III would require air curtain secondary hoods on
converters, and the installation of a control device to collect the
captured converter fugitives.
Alternative IV would necessitate installation of a control device
for collection of captured matte and slag tapping emissions.
4.3.2.13 Phelps Dodge-Morenci. Figure 4-13 presents the baseline
configuration (after consent decree modification, see Section 4.2.1)
of the Phelps Dodge-Morenci smelter. Offgases from the two oxygen
sprinkle modified reverberatory furnaces will pass through two parallel
electrostatic precipitators prior to treatment in an acid plant.
Converter process emissions from the nine existing units pass through
gas coolers and an electrostatic precipitator before entering an acid
plant. Matte and slag tap fugitive emissions are captured by local
ventilation systems and will be vented to the reverberatory furnace
acid plant stack. Converter fugitives are captured by secondary hoods
(estimated to be 30 percent efficient) and discharged from the converter
acid plant stack.
Alternative II would not require additional controls beyond the
baseline configuration, since process gas streams are already under
best control for arsenic.
Air curtain secondary hoods would be required by Alternative III,
along with installation of a control device for collection of captured
converter fugitives.
Alternative IV would require the installation of a particulate
control device for the collection of captured furnace fugitive emissions.
4-35
-------
j
35,000 acfm
To Stack
110,000 acfm
80° C
Acid
Plant
i
ESP
(2)
: 75,000 acfm
•25° C
«
•
•
J
To stack
665,000 acfm
30° C
185,000 acfm
Acid
Plant
•
•
•Captured
Jnatte/slag
Hap
«
64,000 acfm •
370° C j
•
Oxy-Sprinkle
Reverberatory
Furnace
(2)
*••
Matte r
ESP
Conv
(
I
484,000 acfm
650° C
erters
9)
480,000 acfm
25° C
Captured
converter
fugitives
Figure 4-13. Phelps Dodge-Morenci Smelter Baseline Configuration
4-36
-------
4.3.2.14 Copper Ranqe-Khtt.^tne. The base! ine configuration of
1 n t smelter 1s depicted 1n Fi9ure 4-14- p™«" -'«'«»
fro., the two reverberatory furnaces pass through an electrostatic
prec,pitator (operated at 190«C, 375'F) before discharge frm the ma,n
tack. Converter offgases from the two units are vented directly to
the stack. Matte and slag tap fugitive emissions are captured by
oca, ventilation systems and vented to a stack on the building roof
Converter fugitives are neither captured nor controlled
additilT VSPOr "reSSUre Ca'CUlat1°"S ^ S<*"°« 3-1.1.1) show no
t ona, arsenlc removal resulting fr» application of cold (121X
i e PutlCUUte C0"tr01 deVi"S °" the Pr°CeSS Str"ms <* *.
Pine S,,,e, er. Therefore, Alternative I, does not require additional
controls beyond the baseline configuration.
Alternative III would require air curtain secondary hoods on the
conveners, as »,, as installation of a particulate control device
for collection of captured converter fugitive emissions
InstaHation of a control device for the captured matte and slag
tapping em,ss,ons would be required under Alternative IV
4-37
-------
To Stack
J
386,000 ACFM
221"C
247,300 ACFM
190°C
i To roof
ESP
i
.
Reverberatory
Furnaces
(2)
75.000 acfm
25* C
Captured
matte/slag tap
fugitives
Matte
89,000 ACFM
340°C
Converters
(2)
Figure 4-14. Copper Range-White Pine Smelter Baseline Configuration
4-38
-------
4.4 REFERENCES
1. Environmental Protection Agency. Proposed Rules for Primary
Nonferrous Smelter Orders. 44 FR 6284. January 31, 1979.
2. Environmental Protection Agency. Primary Nonferrous Smelter
Orders; National Rules. 45 FR 123. June 24, 1980.
3. Telecon. Bill Gill, Texas Air Control Board with Scott Osbourn,
Pacific Environmental Services. April 14, 1983. Discussed
compliance status and applicable regulations for the ASARCO-E1
Paso smelter.
4. Telecon. Dave Che!grin, Arizona Department of Health Services
with Scott Osbourn, Pacific Environmental Services. April 14,
1983. Discussed compliance status and applicable regulations for
copper smelters located in Arizona.
5. Correspondence from E.R. Flowers, Tennessee Division of Air
Pollution Control, to M.G. Whaley, Pacific Environmental Services.
June 8, 1983. Applicable regulations and compliance status for
Tennessee Chemical Co. smelter.
6. Telecon. Monte Keller, Utah Department of Health with Scott
Osbourn, Pacific Environmental Services. April 14, 1983. Discussed
compliance status and applicable regulations for the Kennecott-Garfield
smelter.
7. Telecon. David Duran, New Mexico Environmental Improvement
Division with Scott Osbourn, Pacific Environmental Services.
April 14, 1983. Discussed compliance status and applicable
regulations for the Kennecott-Hurley and Phelps Dodge-Hildalgo
smelter.
8. Telecon. Dick Serdoz, Nevada Division of Environmental Protection
with Scott Osbourn, Pacific Environmental Services. April 14, 1983.
Discussed compliance status and applicable regulations for the
Kennecott-McGill smelter.
9. Telecon. Dennis Drake, Michigan Department of Natural Resources
with Glenn Whaley, Pacific Environmental Services. June 7, 1983.
Discussed compliance status and applicable regulations for the
White Pine smelter.
10. U.S. Department of Labor, Occupational Safety and Health
Administration. General Industry Standards (29 CFR 1910) and OSH
2206, Revised November 7, 1978.
11. Occupational Safety and Health Administration. Engineering
Assessment and Proposed Compliance Plan for ASARCO-E1 Paso Copper
Smelter. November 1982.
4-39
-------
12. Occupational Safety and HeaUfi Administration. Engineering
Assessment and Proposed Compliance Plan for ASARCO-Hayden Copper
Smelter. October 1981.
13. Occupational Safety and Health Administration. Engineering
Assessment and Proposed Compliance Plan for ASARCO-Tacoma Copper
Smelter. January 1982.
14. Telecon. G. Whaley, Pacific Environmental Services, with C. Gordon,
Occupational Safety and Health Administration. Status of
Compliance Plans for Inorganic Arsenic Control at Primary CoDoer
Smelters. April 20, 1983.
15. Environmental Protection Agency. Nonferrous Metals Manufacturing
Point Source Category; Effluent Limitations Guidelines, Pretreatment
Standards, and New Source Performance Standards. Federal Register,
Vol. 48. February 17, 1983. p. 7032.
16. Environmental Protection Agency and Phelps Dodge Corporation.
Consent Decree for Morenci and Ajo Copper Smelters. March 16,
17. Environmental Protection Agency and ASARCO, Incorporated. Consent
Decree for Hayden Copper Smelter. April 13, 1981.
18. Kennecott Corporation. Compliance Schedule for Hurley (Chino)
Smelter. March 28, 1983.
4-40
-------
5.0 ENVIRONMENTAL IMPACTS
5.1 INTRODUCTION
The environmental impacts on air, energy consumption, solid
wastes, and water associated with the regulatory alternatives for the
control of arsenic emissions from the 14 low-arsenic throughput copper
smelters are presented in this chapter. The purpose of this analysis
is to determine the incremental change in air pollution, water pollution,
solid waste, and energy impacts of the regulatory alternatives over
the baseline control level for the low-arsenic throughput smelters
nationwide.
This chapter first addresses the air pollution impacts of
implementing each of the regulatory alternatives. Energy, solid
waste, and water pollution impacts for the regulatory alternatives are
addressed in Sections 5.3, 5.4, and 5.5, respectively.
5.2 AIR POLLUTION IMPACTS OF REGULATORY ALTERNATIVES
The air pollution impact associated with each of the regulatory
alternatives considered for low-arsenic throughput copper smelters is
presented in this section. Incremental and cumulative arsenic emission
reductions by smelter and nationwide are discussed. The emission
estimates for each regulatory alternative are obtained based on the
application of capture and collection control systems selected in
Section 4.3 as the bases for the regulatory alternatives.
5.2.1 Baseline Emissions
The baseline control level is represented by Regulatory Alternative I
for all 14 low-arsenic throughput copper smelters. For 10 of the
14 smelters, the baseline level reflects existing control of arsenic
emissions. The four remaining smelters are currently planning or
undergoing modernization programs involving changes in processing and
control equipment. For the purpose of defining baseline emissions,
5-1
-------
these 4 plants are assumed to bt. in their modernized configurations as
described in Section 4.2.
Baseline regulatory alternative arsenic emission rates are determined
for each low-arsenic throughput smelter based on the arsenic material
balances presented in Appendix F and the estimated capture and collection
efficiencies of control equipment from Chapters 3 and 4. Baseline
arsenic emission rates are presented in Table 5-1. This table shows
both process and fugitive emission rates for each smelter. From
Table 5-1, arsenic emission rates under Alternative I range from
0.09 kg/hr for both Kennecott-Hurley and Tennessee Chemical Co.-Copperhill
to 5Z.1 kg/hr for Kennecott-HcGill. The baseline arsenic emission
rate nationwide is the sum of the arsenic emission rates for all
14 smelters. As shown in the table, the nationwide baseline inorganic
arsenic emission rate is 85.8 kg/hr.
Nationwide annual arsenic emissions, shown in Table 5-2, are
determined using the emission rates from Table 5-1 and assuming 8,600 hours
of smelter operation per year. For Alternative I, nationwide annual
inorganic arsenic emissions are 738 Mg/yr.
5.2.2 Arsenic Emission Reductions Under the Regulatory Alternatives
Regulatory Alternative II specifies effective controls for process
emissions where additional arsenic removal is predicted. For this
alternative, the Kennecott-McGill smelter would be the only smelter
required to install new equipment. A new baghouse or cold electrostatic
precipitator would need to be installed on the converters. From
Table 5-1, arsenic emission rates under Alternative II range from
0.09 kg/hr for both Kennecott-Hurley and Tennessee Chemical Co.-Copperhill
to 33.3 kg/hr for Kennecott-flcGill. The nationwide inorganic arsenic
emission rate under Alternative II is 66.8 kg/hr. From Table 5-2,
nationwide annual emissions under Alternative II are 575 Mg/yr. This
represents a 22 percent reduction in arsenic emissions nationwide from
the baseline level.
Regulatory Alternative III specifies the application of effective
converter fugitive inorganic arsenic emission controls. Effective
controls for converter fugitive emissions include an air curtain
secondary hood capture system followed by a fabric filter control
device. From Table 5-1, arsenic emission rates under Alternative III
5-2
-------
Table 5-1. ARSENIC EMISSIONS FROM LOW^ARSENIC THROUGHPUT COPPER SMELTERS
BY EMISSION SOURCE AND REGULATORY ALTERNATIVE
Arsenic Emissions by Regulatory Alternative (kg/hr)
en
i
CO
Smel ter Al ternative !
Process
ASARCO-E1 Paso 1.5
ASARCO-Hayden 0.2
Tennessee Chemical 0.0
Co. - Copperhill
Inspiration-Miami 0.04
Kennecott-Garf leld 1.8
Kennecott-llayden 3.2
Kennecott-Hurley 0.0005
Kennecott-McGill 45.8
Magma-San Manuel 1.3
Phelps Oodge-Ajo 3.7
Phelps Oodge-Douglas 4.4
Phelps Dodge-Hidalgo 0.2
Phelps Dodge-Morenci 0.1
Copper Range-White Pine 0.2
Nationwide Arsenic
Emission Rate 62.5
Fugi tive
4.0
6.1
0.09
0.4
2.8
0.9
0.09
6.3
0.08
0.5
0.6
0.2
1.0
0.07
23.1
Total
5.5
6.3
0.09
0.4
4.6
4.1
0.09
52.1
1.4
4.2
5.0
0.5
1.1
0.3
85.8
Alternative 11
Process
1.5
0.2
0.0
0.04
1.8
3.2
0.0005
27.0
1.3
3.7
4.4
0.2
0.1
0.2
43.7 '
Fugitive
4.0
6.1
0.09
0.4
2.8
0.9
0.09
6.3
0.08
0.5
0.6
0.2
1.0
o.n?
23.1
Total
5.5
6.3
0.09
0.4
4.6
4.1
0.09
33.3
1.4
4.2
5.0
Alternative III
Process
1.5
0.2
0.0
0.04
1.8
3.2
0.0005
45.8
1.3
3.7
4.4
0.5 i 0.2
1.1
0.3
66.8
0.1
0.2
Fugitive
1.8
3.2
0.02
0.2
2.0
0.2
0.04
1.5
0.03
0.2
0.1
0.1
0.3
0.03
i
62.5 i 9.6
Total
3.3
3.4
0.02
0.2
3.8
3.4
0.04
47.3
1.3
3.9
4.5
0.3
0.4
0.2
72.1
Alternative IV
Process
1.5
0.2
0.0
0.04
1.8
3.2
0.0005
45.8
1.3
3.7
4.4
0.2
0.1
0.2
62.5
Fugitive
4.0
6.1
0.07
0.3
2.6
0.8
0.08
5.9
0.07
0.4
0.6
0.2
0.9
0.06
22.0
Total
5.5
6.3
0.07
0.3
4.4
4.0
0.08
51.7
1.4
4.1
5.0
0.4
1.0
0.3
84.5
-------
Table 5-2. NATIONWIDE ANNUAL ARSENIC EMISSIONS BY
REGULATORY ALTERNATIVE AND EMISSION REDUCTIONS
FROM BASELINE
Regulatory
Alternative
Alternative I (Baseline)
Alternative II
Alternative III
Alternative IV
Arsenic
Emissions
(Mg/yr)
738
575
620
727
Arsenic Emission Reductions
(Mg/yr)
from Basel ine
(V
—
163
118
11
22
16
1.5
5-4
-------
range from 0.02 kg/hr for Tennessee Chemical Co.-Copperhill to 47.3 kg/hr
for Kennecott-McGill. The nationwide emission rate under Alternative III
is 72.1 kg/hr. From Table 5-2, nationwide annual emissions under
Alternative III are 620 Mg/yr. This represents a 16 percent reduction
in arsenic emissions nationwide from the baseline level.
Regulatory Alternative IV specifies the installation of effective
fugitive controls for matte and slag tapping operations. Effective
control of fugitive emissions from these sources includes local ventilation
hood capture followed by a fabric filter collection system. From
Table 5-1, arsenic emission rates under Alternative IV range from
0.07 kg/hr for Tennessee Chemical Co.-Copperhill to 51.7 kg/hr for
Kennecott-McGill. The nationwide inorganic arsenic emission rate
under Alternative IV is 84.5 kg/hr. From Table 5-2, nationwide annual
emissions under alternative IV are 727 Mg/yr. This represents a
1.5 percent reduction in inorganic arsenic emissions nationwide from
the baseline level.
Regulatory Alternative V requires zero emissions of arsenic to
the atmospheric from low arsenic throughput copper smelters. Under
this alternative, air pollution, energy, and other environmental
impacts would reduce to zero.
5.3 ENERGY IMPACTS OF THE REGULATORY ALTERNATIVES
Energy use associated with the baseline case was estimated assuming
that low arsenic throughput copper smelters require 35 x 10 Btu
(thermal) of energy per ton of copper produced. The 14 low arsenic
copper smelters have a total reported copper production capacity of
c c n
1.6 10 Mg/yr (1.8 x 10 tons/year). Assuming a power plant efficiency
of 35 percent, the total nationwide electrical energy requirement of
low arsenic throughput copper smelters was estimated to be about
5 x 1010 kWh/yr.
Annual energy requirements for Regulatory Alternatives II through
IV were estimated based on the energy requirements of additional
control equipment specified for each regulatory alternative. For
Alternative II, additional energy is required for control of process
emissions from the converters at Kennecott-McGill. For Alternative III,
additional energy is required for the capture and collection of converter
fugitive emissions. For Alternative IV, additional energy is required
5-5
-------
for the collection of fugitive passions from matte and slag tapping
operations.
Table 5-3 summarizes the nationwide incremental energy requirements
of the regulatory alternatives over the baseline case for low arsenic
throughput smelters. Incremental energy requirements, shown in this
table, are 4.9 x 108 kWh for Alternative II, 1.8 x 108 kWh for
Alternative III, and 9 x 106 kWh for Alternative IV. Under Alternative II,
only the Kennecott-McGill smelter is required to install additional
process pollution control equipment. Therefore, the incremental
energy requirements for Alternative II reflect the additional electrical
energy needed to operate emission control equipment at the Kennecott-
McGill smelter. This additional requirement, 4.9 x 108 kWh, reflects
fan and ESP electrical consumption plus the energy required to reheat
the stack gases in order to maintain plume buoyancy. Thermal energy
for reheat was calculated as electrical energy assuming a power plant
efficiency of 0.35. The baseline energy requirements for the Kennecott-
McGill smelter can be calculated assuming a baseline energy requirement
of 35 x 10 Btu (thermal) of energy per ton of copper produced1 and a
total reported copper production capacity of 45,400 Mg/yr (50,000 tons/
year). The baseline electrical energy requirement for Kennecott-McGill
g
is estimated to be 2 x 10 kWh. Compared to the baseline, the additional
energy requirements of Alternative II represent an increase of 33 percent
for the Kennecott-McGill smelter. The additional energy requirements
of Alternatives III and IV for the 14 low arsenic throughput smelters
are negligible when compared to the baseline area.
5.4 SOLID WASTE IMPACTS OF THE REGULATORY ALTERNATIVES
Arsenic-bearing dust is collected when smelting process particulate
emissions are controlled with baghouses or electrostatic precipitators.
Often, this dust contains recoverable copper or other salable materials.
The collected dust is recycled to the process or reclaimed elsewhere.
Thus, only a portion of the material collected by air pollution control
equipment becomes solid wastes. Arsenic-containing materials are
present in the acid plant waste. Acid plant waste is usually in the
form of a slurry. State regulations require settling of this slurry
in a concrete pit. The clarified slurry is transferred to a lined
5-6
-------
Table 5-3. NATIONWIDE ANNUAL ENERGY REQUIREMENTS
BY REGULATORY ALTERNATIVE
Incremental Annual
Energy Requirements
Regulatory from Baseline
Alternative (1Q6 kwh/yr)
Alternative I (Baseline)
Alternative II 490a
Alternative III 180
Alternative IV 9
Represents the additional energy required by Kennecott-FlcGill for
process control equipment and stack gas reheat.
5-7
-------
lagoon for further settling. From the lagoon the materials may be
dredged and recycled to the process.3
A conservative estimate of the amount of solid wastes generated
nationwide under the baseline case can be made by assuming that 50 percent
of the concentrate fed to the smelter becomes solid waste; 25 percent
is sulfur which is removed from the process as S02, and the remaining
25 percent becomes blister copper. Given a nationwide annual maximum
concentrate feed rate of 6.4 x 106 Mg/yr,2 solid wastes (including
slag) generated nationwide by low arsenic smelters are approximately
3.2 x 105 Mg/yr.
The incremental quantity of solid waste generated nationwide for
Regulatory Alternatives II through IV is shown in Table 5-4. Incremental
solid waste impacts are estimated by assuming that all process and
fugitive emissions collected under each alternative contain 1 percent
arsenic. In comparison with the amount of solid waste generated by
the smelter under the baseline case, the additional amounts of solid
waste collected under Alternatives II through IV are negligible.
5.5 WATER POLLUTION IMPACTS OF THE REGULATORY ALTERNATIVES
The control systems for the regulatory alternatives are dry
systems; consequently, no incremental increase in water discharges is
anticipated. If scrubbers are used, increases in wastewater discharges
result if the arsenic-containing dusts are disposed along with the
acid plant slurry. Even if scrubbers are used, no adverse water
pollution impact is anticipated. This is because the additional waste
water discharges through use of scrubbers would be treated within
existing smelter water pollution control systems installed to meet
existing State and Federal regulations.
5-8
-------
Table 5-4. NATIONWIDE ANNUAL SOLID WASTES GENERATED
BY REGULATORY ALTERNATIVE
Regulatory
Alternative
Incremental
Solid Wastes
Generated Annually
from Baseline
(Mg/yr)a
Alternative I (Baseline)
Alternative II
16,300
Alternative III
11,800
Alternative IV
1,100
aAssumes annual nationwide concentrate feed rate of 6.4 x 10 Mg
concentrate/yr; 50 percent of concentrate becomes solid waste,
and that all process and fugitive emissions collected
under Alternatives II through IV contain 1 percent arsenic.
5-9
-------
5.6 REFERENCES
1. Pitt, C.H., and M.E. Wadsworth. An Assessment of Energy Requirements
in Proven and New Copper Processes. Prepared for the U.S. Department
of Energy. Contract No. EM-78-S-07-1743. December 31, 1980.
p. 9.
2. Preliminary Study of Sources of Inorganic Arsenic. U.S. Environmental
Protection Agency. Research Triangle Park, North Carolina.
Report No. EPA 450/5-82-005. August 1982. p. 21.
3. Calspan Corporation. Assessment of Industrial Hazardous Waste
Practices in the Metals Smelting and Refining Industry. Volume II,
Primary and Secondary Nonferrous Smelting and Refining. PB 276170.
April 1971.
5-10
-------
6.0 COSTS
This section presents capital and annualized costs of controlling
(1) process arsenic emissions; and (2) fugitive arsenic emissions from
matte and slag tapping and converter operations at low arsenic throughput
copper smelters. These data are used to determine the cost of implementing
the regulatory alternatives identified in Section 4. The control
requirements of the regulatory alternatives for each smelter are
presented in Table 6-1. As indicated in Section 4, roaster calcine
discharge operations at the calcine charge smelters are effectively
controlled for fugitive arsenic emissions and therefore are excluded
from the cost analysis.
The capital cost includes all the cost items necessary to design,
purchase, and install the particulate control system. The capital
cost of a control system includes the purchase cost of the control
device, auxiliaries such as exhaust fans, motors, ductwork, and stack;
direct installation charges including foundation and other direct
costs such as electrical, instrumentation and controls; and indirect
costs for engineering services, taxes, contractors fees, and contingency.
All costs are in December 1982 dollars.
The annualized cost of a control system is the annual cost to the
individual plant to own and operate that control system. The annualized
cost includes direct operating costs such as utilities, maintenance,
operating labor, and indirect operating costs or capital-related
charges such as depreciation, interest, administrative overhead,
property taxes, and insurance.
While actual costs experienced by individual plants can vary, the
following values have been selected as typical and provide a reasonable
estimate of the annualized costs of each control system:
6-1
-------
Table 6-1. REGULATORY ALTERNATIVE CONTROL REQUIREMENTS
Alternative Control Requirement
I (Baseline case) Existing process and fugitive
controls on roasters, smelting
furnaces and converters. Also
included are those controls
planned or agreed to be installed
prior to 1987,
II Effective process controls and
existing fugitive controls.
Effective process controls include
evaporative cooling followed
by either an existing or new ESP
or fabric filter collector for
particulate matter emissions.
HI Existing process controls and
effective fugitive controls for
converters. Effective fugitive
controls for converters include
air curtain secondary hoods for
capture followed by fabric
filters for collection.
IV Existing process controls and
effective fugitive controls for
matte tapping and slag tapping
operations. Effective fugitive
controls for matte tapping and
slag tapping operations include
local ventilation for capture
followed by fabric filters for
col lection.
6-2
-------
Direct annualized cost Iter-.
• Operating labor at $11.53 per hour and supervision labor at
15 percent of operating labor expense.
• Maintenance labor at $11.53 per hour and supervision labor
at 20 percent of maintenance labor expense.
• Maintenance material at 100 percent of maintenance labor
expense.
2
• Electricity at 5.9 cents per kilowatt hour.
• Water at 8<£/m3 ($0.30/1,000 gallons).3
• Natural gas at $5.67/60 ($5.98/million Btu)
Indirect annualized cost items
• Payrol1 overhead at 60 percent of payrol1.
• Operating supplies at 20 percent of total maintenance cost.
• Administrative overhead at 40 percent of total operating and
maintenance labor and operating supplies.
• Taxes and insurance at 2 percent of total capital cost.
• Capital recovery at 20 year life for ESP's and fabric filters,
10 year life for reheat equipment, and 10 percent interest
rate.
A detailed cost analysis for the alternate control systems is
presented in the following sections.
6.1 BASELINE CONTROLS
This section presents the estimated costs of baseline controls on
process and fugitive emission sources. The baseline controls are
defined as the existing process and fugitive controls planned or
agreed to be installed before 1987. Baseline controls have no additional
costs associated with them and therefore are excluded from the cost
analysis. Some smelters are replacing or planning to replace existing
smelting furnaces with a new smelting technology which generates SCL
emissions in concentrations suitable for collection in a sulfuric acid
plant. Sulfuric acid plants remove arsenic emissions contained in the
process offgases treated in their gas precleaning and conditioning
6-3
-------
equipment. For these smelters the baseline control cost includes the
total cost of the new smelting technology, sulfuric acid plant, and
any additional agreed upon controls. Table 6-2 lists the smelters for
which agreements exist and presents the control equipment to be installed
under each agreement.
6.1.1 Baseline Costs
Table 6-3 presents the estimated capital and annualized costs for
the add-on equipment listed in Table 6-2. Capital and annualized
costs for new smelting technologies were based on cost estimates
contained in a draft report prepared for EPA.4 The report presents
cost estimates in mid-1981 dollars for retrofiting INCO flash and
oxygen sprinkle/oxygen fuel smelting technologies at existing smelters
of varying capacity. The capital cost estimates were obtained by
adjusting the cost data reported in the referenced report to represent
the production capacity at a subject smelter and adjusting the resultant
cost estimates upward to reflect December 1982 dollars. The annualized
cost estimates were obtained by adjusting the reported data to be
consistent with the annual operating cost bases presented above.
6.2 PROCESS CONTROLS
This section presents the estimated costs of applying effective
controls on process emission sources. As discussed in Section 4 and
presented in Table 6-1, effective process controls are defined as
control systems containing evaporative cooling followed by an ESP for
collection of process particulate matter emissions. As indicated in
Section 4, by applying effective add-on process controls, additional
arsenic emission reduction over baseline can be achieved only for
converters at the Kennecott-McGill smelter. Therefore, the process
control cost analysis in this section will be limited to developing
cost estimates for the control of process emissions from converters at
the Kennecott-McGill smelter.
6.2.1 Process Control Costs
Capital and annualized costs were estimated for an add-on evaporative
cooler and ESP for the existing converter process emissions control
system at the Kennecott-McGill smelter. Currently, the converter
6-4
-------
Table 6-2. EQUIPMENT CONSIDERED IN BASELINE COST ANALYSIS
Smelter
Add-on equipment
1. ASARCO - El Paso
2. ASARCO - Hayden
3. Kennecott - Hurley
4. Phelps Dodge - Ajo
5. Phelps Dodge - Morenci
Fugitive emission capture system on
smelter slag tapping operation.
INCO smelting technology in place
of the existing conventional
roaster/reverberatory smelting process.
Local ventilation for capture of
fugitive emissions from the new
smelter matte tapping and slag
tapping operations and an existing
ESP system for collection of captured
emissions. A sulfuric acid plant
for process emissions.
INCO smelting technology in place of
the existing conventional reverberatory
smelting process. A sulfuric acid plant
for process emissions from the new smelter.
Oxygen sprinkle/oxygen fuel
smelting in place of the existing
conventional reverberatory smelting
process. A sulfuric acid plant for
process emissions from the modified
smelter.
Oxygen sprinkle/oxygen fuel smelting
in place of the existing conventional
reverberatory smelting process. A
sulfuric acid plant for process emissions
from the modified smelter.
Since an existing ESP will be used for the collection of captured
emissions, the baseline cost analysis includes only an annual operating
cost for the collection system.
No formal agreement has been made between Kennecott Copper Company and
EPA or any other agency regarding installation of the add-on control
equipment listed.
6-5
-------
Table 6.3. ESTIMATED CAPITAL AND ANNUAUZED COSTS OF
BASELINE CONTROLS FOR PRIMARY COPPER SMELTERS
(December 1982 dollars)
1.
2.
3.
4.
5.
Smelter
ASARCO - El Paso
ASARCO - Hayden
Kennecott - Hurley
Phelps Dodge - Ajo
Phelps Dodge - Morenci
Capital cost,
$1,000
46
75,606
54,044
51,067
92,294
Annual i zed cost,
$1,000
40
24,698
17,111
16,356
30,815
6-6
-------
offgases are treated in multi-cyclones and vented through a stack to
the atmosphere. The design parameters used to calculate add-on control
system costs are summarized in Table 6-4. No stack was included in
the system with the assumption that the existing stack will be used to
vent the offgases from add-on controls.
Methodology for Estimating Capital and Annualized Costs - Capital
and annualized costs were developed for the evaporative cooler and ESP
system by estimating capital costs separately for the evaporative
cooler, ESP, fan, ductwork, and reheat and annualized costs for the
entire system.
The capital costs for the evaporative cooler, ESP, and fan were
based on data contained in a report prepared for EPA by the Industrial
Gas Cleaning Institute (IGCI). The report presented estimates of
equipment purchase and total capital costs for evaporative cooler and
ESP systems for treating offgases from reverberatory furnaces at two
smelters. These estimates were prepared by several ESP manufacturers
in fourth quarter 1977 dollars. Using these reported cost data, a
mathematical expression was developed to relate the reported purchase
costs for evaporative cooler, ESP, and fan to gas flow rate. To this
purchase cost expression, the following factors were applied to obtain
a mathematical expression relating total capital cost in December 1982
dollars to gas flow rate: 175 percent to obtain total capital cost
from purchase cost, 120 percent for an assumed 20 percent retrofit
cost, and a 1.54 escalation factor to convert the costs in fourth
quarter 1977 dollars to December 1982 dollars. The capital cost
factor of 175 percent was obtained based on cost data in the IGCI
5
report. The escalation factor of 1.54 was obtained by using the
Chemical Engineering Plant cost index (December 1982 = 314.3). The
resultant mathematical expressions relating equipment purchase cost
(C^ and capital cost (C«) to gas flow rate are as follows:
C.^ = Purchase cost of evaporative cooler (C, ) + Purchase
cost of ESP (C, ,) + Purchase cost of fan* (C, )
•iy iz
= 4,083(Q1)°'91 + 16,663 (Q9)°'89 + 1,093 (Q,)0'96
1 C. 1
*Fan is located on the hot side of the gas stream.
6-7
-------
Table 6-4. DESIGN PARAMETERS FOR ADD-ON PROCESS
PARTICIPATE MATTER CONTROL FOR CONVERTERS
AT KENNECOTT-MCGILL
Parameter
Value
System description
Evaporative cooler
Offgas at the inlet:
Flow rate, ai3/s facfm)
Temperature, °C (°F)
Arsenic content, g/s (lb/min)
g/m3 (gr/acf)
Water, content, % of total volume
Water consumption, m3/h (gpm)
Electrostatic precipitator
Offgas at the in!etb:
Flow rate, n3/s (acfn)
Temperature, °C (°F)
Arsenic content, g/s (lb/min)
g/m3 (gr/acf)
Water content, % of total volume
Migration velocity, cm/s (ft/mm)
Specific collection area,
m2 per m3/s (ft2/!,000 acfm)
Power consumption, kw/m2 (kw/ft2) of
collection area
Offgas at the outlet:
Arsenic content, g/s (Ib/min)
g/m3 (gr/acf)
Fan
Location
Pressure drop, kPa (in. water)
Power consumption, kw (hp)
Reheat
Reheat temperature increase, °C (°F)
Capacity, GJ (million Btu/hr)
A fan, an evaporative
cooler, and a dry elec-
trostatic precipitator
197
232
9.16
0,05
4*
39
(418,300)
(450)
(1.21)
(0.02)
170
154
120
9.16
0.06
3.96
103.2
0.016
(326,400)
(250) c
(0.'026)
(7.3)d
(524)d
(0.0015)
4.93 (0.65)e
0.032 (0.014)6
Hot side of cooling system
1 (4)
615 (825)
112
71
(200)
(67)
Assumed based on test data for converters "at several other smelters.
Same as at the outlet of evaporative cooler.
cThe arsenic loading at the ESP inlet may be lower than that at the
evaporative cooler inlet.
Quoted by ESg manufacturers for reverberatory furnaces in the
IGCI report.
Calculated using the arsenic saturation value of 0.0302 g/m3(0.0132 gr/acf)
at 120 C (250°F) and 96 percent efficiency for the add-on ESP system.
Downstream of the existing milticyclones and upstream of the add-on
evaporative cooler.
6-8
-------
where Q, = Actual flow rate at the inlet of add-on
3
evaporative cooler, m /s, and
Q2 = Actual flow rate at the inlet of add-on ESP,
m3/s at 120°C
= Capital cost of evaporative cooler and ESP
= 1.54 x 1.2 x 1.75 (Clx + Cly + Clz)
- 3,234 [4,083 (Q' + 16»663
xO.91 , 1t- 9C- /n xO.89 . n 0.96-,
= 3,535 [3.74 (Ql} + 15'25 (Q2> + Ql ]
Ductwork costs were estimated based on an assumed total of 305 m
(1,000 ft) of duct from the existing equipment outlet to the inlet of
add-on control system (i.e., at the inlet to fan) and a return duct
from the system outlet to the existing equipment. The ductwork cost
(C~) based on the total duct weight and a unit cost per weight of
O
$ll,355/Mg ($10,300/ton) was $2,008,500. The unit cost of ductwork
assumes that an additional 40 percent of the total ductwork weight is
needed for support structures.
The capital cost for reheat equipment was based on data contained
in a report prepared by EPA. The report presents capital cost estimates
in fourth quarter 1977 dollars for reheat equipment of different
capacity. The reported costs were escalated to December 1982 dollars
using the Chemical Engineering Plant Cost Index (December 1982 = 314.3).
A mathematical expression was developed to relate capital costs (C«)
in December 1982 dollars to gas flow rate. The resultant expression
is:
C4 = 1.54 [55,466 (Q3)0.59] = 85,418 (Q3)0.59
where Q- = Reheat capacity, GJ/h
The total capital cost (C) in December 1982 dollars for the evaporative
cooler and ESP system is the summation of capital costs for the evaporative
cooler, ESP, and fan (C2), ductwork (C-J, and reheat equipment (C.).
6-9
-------
c = c2 + c3 + c4
= 3,535 [3.74 (Q^O.91 + 15.25 (Q2)0.89 + (Q^O.96]
= 2,008,500 + 85,418 (Q3)0'59
= 3,535 [3.74 (Q^O.91 + 15.25 (Q2)0.89 + (0^)0.96
+ 568.2 + 24.16 (Q3)0.59]
The estimated capital cost for an add-on evaporative cooler and
ESP system located upstream from existing converter process controls
at the Kennecott-McGill smelter is $10,017,800. This cost was calculated
using the gas parameters listed in Table 6-4 and the above mathematical
expression.
Annualized costs were calculated based on operating parameters
listed earlier and the following requirements:
• 2 manhours/shift operating labor and 1 manhour/shift maintenance
labor.
• Pressure drop of 1.0 kPa (4 in. water) across evaporative
cooler and ESP.
• Electricity for ESP at 0.0015 kW/ft2 of ESP plate area.
• Miscellaneous electricity costs at 10 percent.
Table 6-5 presents an estimate of annualized cost for add-on
process emission controls on converters at the Kennecott, McGill
smelter.
6.3 FUGITIVE CONTROLS
This section presents the estimated costs of applying effective
controls on fugitive emission sources at low arsenic throughput primary
copper smelters.
6.3.1 Converter Controls
Fugitive emission control equipment selected for analysis for
converter operations includes an air curtain secondary hood capture and
fabric filter collection system. As noted in Section 4, five of the
smelters currently have some form of converter secondary hooding
in place. One smelter, ASARCO-E1 Paso, has a building evacuation
6-10
-------
Table 6-5. ESTIMATED ANNUALIZED COST OF ADD-ON PROCESS
PARTICULATE HATTER CONTROL SYSTEM FOR CONVERTERS
AT KENNECOTT-McGILL
(December 1982 dollars)
Item Cost,$
Direct costs
Operating labor
Maintenance labor
Maintenance material
Utilities:
Electricity3
Natural gas for reheat
Water
28,600
15,000
15,000
312,400
2,252,500
31,000
Total direct 2,654,500
Indirect costs
Payroll overhead 26,200
Operating supplies 3,000
Administrative overhead 18,600
Taxes and insurance 200,300
Capital recovery:
Evaporative cooler and ESP 1,053,000
Reheat equipment 171,800
Total indirect 1,472,900
Total 4,127,400
a$130,200 is ESP power requirement.
6-11
-------
system which includes a fabric filter for emission collection. The
five smelters with converter secondary hooding are ASARCO-Hayden,
Kennecott-Garfield, Phelps Dodge-Ajo, -Morenci, and -Hidalgo.
6.3.1.1 Converter Control Costs - Capital and annualized costs
were estimated for the installation of air curtain secondary hood
capture and fabric filter collection systems on converters at all
smelters. For smelters without existing fugitive controls on converters,
the cost estimates include the total cost for installation of air
curtain secondary hoods and fabric filters. However, for those smelters
with some form of converter fugitive emission controls presently
in place or required under an EPA consent decree, the cost estimates
include only the incremental cost needed to upgrade the existing or
future control systems to an air curtain secondary hood capture system
equipped with an effective collection device.
The capital costs were developed by estimating costs for air
curtain secondary hood capture equipment and fabric filter collection
equipment for each smelter. The annualized costs were developed by
estimating costs for the total capture and collection system at each
smelter.
Capital costs for air curtain secondary hoods - Design parameters
for air curtain secondary hood capture systems at each smelter are
summarized in Table 6-6. Data on the number of existing and operating
converters are based on the information obtained from individual
companies and Reference 7. Exhaust fan capacities were developed
using the design rates obtained from ASARCO and discussed in Section 3
for the air curtain secondary hood capture systems for converters at
the Tacoma smelter. The following flow rates were used for each
converter: 33 m /s (70,000 acfm) during blowing and holding and 57
m /s (120,000 cfm) during charging and skimming. Air curtain flow
rates used were 5.2 m /s (11,000 acfm) during blowing and 8.5 m3/s
(18,000 acfm) during charging and skimming.
The following ductwork requirements were estimated for the smelters
with no existing converter controls: for the first converter at each
smelter 54 m (178 ft) of 1.5 m (60 in.) diameter duct was assumed to
be needed to reach the main exhaust fan. An additional 19.8 m (65 ft)
6-12
-------
Table 6-6. DESIGN PARAMETERS FOR AIR CURTAIN SECONDARY
HOOD CAPTURE SYSTEM FOR PRIMARY COPPER SMELTERS
CT)
OJ
Smel ter
No. of converters
Total Total b
existing3 operating
Fan capacity,
mVs (acfm)
Flue system dimensions
1. ASARCO - El Pasoy
2. ASARCO - Hayden
3. Tennessee Chemical
Co. - Copperhil 1
4. Inspiration - Miami
1RO, 1R1 , 1SB 109 (230,000) 31 m (100 ft) long 1.5 m (60 in.) duct
1RO, 1RI, HI 142 (300,000)
5. Kennecott - Garfield
6. Kennecott - Hayden
7. Kennecott - Hurley
8. Kennecott - McGill
9. Magma - San Manuel
10. Phelps Dodge - Ajo9
11. Phelps Dodge - Douglas
12. Phelps Dodge - Hidalgo9 3
13. Phelps Dodge - Morenci 9
14. Copper Range - White Pine 2
1RO
IRQ, 1RI, III
1RO, 1RI, HI
1RO, 1R1 , 1SB
1RO, 1RI, HI
1RO, 1R1, 1H
2RO, 2RI
66 (140,000)
142 (300,000)
142 (300,000)
109 (230,000)
1?8 m (420 ft) long 1.5 m (60 in.) duct and 91 m (300 ft) long 1.7 m (67 in.) duct
469 m (1,540 ft) long 1.5 m (50 in.) duct and 91 m (300 ft) long 2.5 m (98 in.) duct
223 m (730 ft) long 1.5 m (60 in.) duct and 91 ro (300 ft) long 2.2 m (86 in.) duct
142 (300,000) 335 m (1,100 ft) long 1.5 m (60 in.) duct and 91 m (300 ft) long 2.5 m (98 in.) duct
14? (300,000) 335 m (1,100 ft) long 1.5 m (60 in.) duct and 91 m (300 ft) long 2.5 m (98 in.) duct
208 (440,000) 623 m (2,043 ft) long 1.5 m (60 in.) duct and 91 m (300 ft) long 3 m (119 in.) duct
1RO, 1RI, 1SB 109 (230,000) 223 m (730 ft) long 1.5 m (60 1n.) duct and 91 m (300 ft) long 2.2 m (86 in.) duct
1RO, 1RI, 1H 142 (300,000) 469 m (1,540 ft) long 1.5 m (60 in.) duct and 91 m (300 ft) long 2.2 m (98 in.) duct
223 m (730 ft) long 1.5 m (60 in.) duct and 91 m (300 ft) long 2.2 m (86 in.) duct
1RO. 1RI, 1SB 109 (230,000)
2RO, 2RI 208 (440,000)
1RO
66 (140,000) 128 m (420 ft) long 1.5 m (60 in.) duct and 91 m (300 ft) long 1.7 m (67 in) duct
aEach existing smelter will be equipped with an air curtain secondary hood capture system. Each air curtain will have a separate fan designed at
8.5 m3/s (18,000 acfm) at a pressure of 7.5 kPa (30 in. h^O).
bRO: Rol'1-out, RI: Roll-in, SB: Standby, and H: Holding.
cFan capacities are based on operating converters only. Assumed exhaust rate from a Converter in roll-out mode 0-e.. during
57 m'/s (120,000 acfm) and from a converter in roll-in mode (i.e., during blowing and holding) 1s 33 m3/s (70,000 acfm).
dr,
"*
,
dOuct lengths used are- 54 m (178 ft) long 1.5 m (60 in.) diameter duct per system, 19.8 m (65 ft) long 15 m (60 In.) diameter duct for each
addition!' I system, and 91 m (300 ft) long conwon duct downstream of the fan. Example case Kennecott, Hayden: length of 1.5 m (60 in.)
diameter duct is 3 x 54 + (1) x 19.8 + (2) x 19.8 = 223 m.
a/?:-
sssr-s s-s2J'»
(100 ft) ducting is sufficient to move the captured gases to the existing duct.
Currently, the converter fugitive emissions are captured in a secondary hood and a duct exists to ^ ^Y*^ J85" \°e* Stack'
is assumed to be utilized to vent the captured gases from the air curtain secondary hood. No new ducting or fans are required.
"Currently the converter fugitive emissions are captured in a secondary hood and a duct exists to vent the captured gases to a stack, "owever, the
existin system is smaller than the air curtain secondary hood capacity. Therefore, none of the existing equipment 1s assumed to be utilized.
-------
of 1.5 m (60 in.) diameter duct ic required for each subsequent capture
system after the first one (i.e., 54 m duct for the first converter,
54 m + 19.8 m duct for the second converter, 54 m + 2 x 19.8 m for the
third converter and 54 m + 3 x 19.8 m for the fourth converter). A
91 m (300 ft) common duct was used for venting gases from the main fan
to the particulate collection device. The common duct diameters vary
for each smelter based on exhaust fan capacity.
For the ASARCO-E1 Paso, -Hayden, Kennecott-Garfield, and Phelps
Dodge-Morenci smelters with some form of existing converter controls
in place, it was assumed the existing ductwork and fans were salvageable
and could be incorporated in the new air curtain secondary hood system.
A new 31 m (100 ft) of 1.5 m (60 in.) diameter ductwork was assumed to
be required to move the capt'ired gases from the new system to the
existing ductwork at the ASARCO-E1 Paso smelter. No new ductwork was
assumed to be required at the other smelters. At the Phelps Dodge-Ajo
and -Hidalgo smelters, the existing converter control system capacity
is smaller than the new air curtain secondary hood system. Therefore,
none of the existing equipment was assumed to be salvageable. New
ductwork costs were estimated for these smelters in the same manner as
for the smelters with no existing converter controls.
Table 6-7 presents estimated capital costs for air curtain secondary
hoods for each low arsenic throughput smelter. These cost estimates
were calculated using the cost data provided to EPA by ASARCO.8 The
ASARCO cost data contained January 1982 cost estimates for installing
air curtain secondary hood capture equipment on three converters at
the Tacoma smelter. The direct capital costs reported were based on
actual estimates and indirect capital costs were based on percentages
of the estimated equipment cost. These cost data were updated to
December 1982 dollars and used to estimate the costs for installation
of an air curtain and a secondary hood. The resultant costs, $197,100
for an air curtain and $125,100 for a secondary hood, were used to
estimate air curtain secondary hood costs for the low arsenic throughput
smelters based on the number of converters listed in Table 6-6 for
each smelter. Electrical system costs were estimated for each capture
system based on data from the ASARCO report. Ductwork costs were
6-14
-------
(December 1982 dollars)
Capital cost, $
en
wMHt I UC. 1
1. ASARCO - El Paso
2. ASARCO - Hayden
3. Tennessee Chemical Co. -
Copperhill
4. Inspiration - Miami
5. Kennecott - Garfield
6. Kennecott - Hayden
7. Kennecott - Hurley
8. Kennecott - McGill
9. Magma - San Manuel
10. Phelps Dodge - Ajo
11. Phelps Dodge - Douglas
12. Phelps Dodge - Hidalgo
13. Phelps Dodge - Morenci
14. Copper Range - White Pine
Air curtain
secondary hood
966,600
1,611,000
644,400
1,611,000
1,288,800
966,600
1,288,800
1,288.800
1,933,100
966,600
1,611,000
966,600
2,899,800
644,400
Ductwork
350,000
0
1,631,200
3,981,900
0
2,444,000
3,255,000
3,255,000
5,171,000
2,444,000
3,981,900
2,444,000
0
1,631,200
Electrical
58,600
90,700
42,600
90,700
74,600
58,600
74,600
74,600
106,700
58,600
90,700
58,600
154,600
42,600
Total
1,375,200
1,701,700
2,318,200
5,683,600
1,363,400
3,469,200
4,618,400
4,618,400
7,210,800
3,469,200
5,683,600
3,469,200
3,054,400
2,318,200
-------
estimated based on the duct requirements listed in Table 6-6 and an
assumption that an additional 40 percent of duct material is required
for use as support structures.
Capital costs for fabric filters - Table 6-8 presents estimated
capital costs for fabric filters required for the collection of captured
fugitive emissions. The table also summarizes the capital costs for
air curtain secondary hoods from Table 6-7 and presents the capital
costs for the total system consisting of air curtain secondary hoods
and a fabric filter for each smelter. These cost estimates were
obtained by calculating costs for a fabric filter,, 43 m (140 ft) of
ductwork, and a 61 m (200 ft) stack for each smelter. The exhaust fan
capacities listed in Table 6-6 were used to size equipment. The
fabric filter costs were based on cost estimates contained in the I6CI
g
report. The IGCI report presents cost estimates in fourth quarter
1977 dollars for fabric filters treating offgases from reverberatory
furnaces at two smelters and was prepared by several fabric filter
manufacturers. The costs were developed for a 2-to-l filter cloth-to-air
ratio. Both reverse air and shaker type mechanisms were suggested for
bag cleaning.
Ductwork costs were estimated based on a total of 43 m (140 ft)
of ductwork assumed to be needed for venting offgases from the fabric
filter to the stack, and a unit cost of $ll,355/Hg ($10,300/ton) of
ductwork weight. The unit cost includes 40 percent additional material
to be used for support structures.
The capital cost for the stack was based on data contained in a
report prepared for EPA. The report presented capital cost estimates
in December 1977 dollars for different sized stacks.
The reported costs for fabric filters and stacks were escalated
to December 1982 dollars using the Chemical Engineering Plant Cost
Index (December 1982 = 314.3). Total capital cost for the fabric
filter collection equipment was obtained by adding the costs for the
fabric filter, duckwork, and stack, plus a 25 percent retrofit factor.
A mathematical expression is developed to relate the total capital
costs of fabric filter collection equipment to gas flow rate. The
resultant expression is:
6-16
-------
Table 6-8. ESTIMATED CAPITAL COSTS OF AIR CURTAIN SECONDARY
HOODS AND FABRIC FILTERS FOR PRIMARY COPPER SMELTERS
(December 1982 dollars)
Smelter
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
ASARCO - El Paso3
ASARCO - Haydenb
Tennessee Chemical
Co. - Copperhill
Inspiration - Miami
Kennecott - Garfield
Kennecott - Hayden
Kennecott - Hurley
Kennecott - McGill
Magma - San Manuel
Phelps Dodge - Ajo
Phelps Dodge - Douglas
Phelps Dodge - Hidalgo
Phel ps Dodge - Morenci
Copper Range -
White Pine
Capital cost, $
Air curtain
secondary
hoods
1,375,200
1,701,700
2,318,200
5,683,600
1,363,400
3,469,200
4,618,400
4,618,400
7,210,800
3,469,200
5,683,600
3,469,200
3,054,400
2,318,200
Fabric filter
0
0
2,115,700
4,141,300
3,833,800
3,262,000
4,141,300
4,141,300
5,838,800
3,262,000
4,141,300
3,262,000
5,471,300
2,115,700
Total
1,375,200
1,701,700
4,433,900
9,824,900
5,197,200
6,731,200
8,759,700
8,759,700
13,049,600
6,731,200
9,824,900
6,731,200
8,525,700
4,433,900
The captured gases will be treated in an existing fabric filter
collection system.
DThe cost of collection system for the captured fugitive emissions was
included in the baseline.
6-17
-------
C = Total capital co?t of fabric filter collection equipment,
December 1982 dollars
= (Retrofit factor) [Capital cost of fabric filter (CJ
+ Capital cost of ductwork (CJ + Capital cost of
stack (C3)]
= (125%)[24,430 (Q4)0'96 + 19,428 (Q4)0'5 + 22,300 (Q4)0'49
where Q. = Actual flow rate at the inlet of fabric
" o
filter (i.e., fan exhaust flow listed in Table 6-6), m /s
= 24,285 [1.26 (Q4)0'96 + (Q4)0'5 + 1.15 (Q4)0'49]
Annualized costs for air curtain secondary hood capture and
fabric filter collection system - Table 6-9 presents the annualized
costs for air curtain secondary hood capture and fabric filter collection
systems. The costs were estimated using the cost bases discussed
earlier and the following labor and utility requirements:
• 2 manhours/shift for operating and maintenance labor for capture
equipment and 2 manhours/shift operating and maintenance labor
for a 66 m3/s (140,000 acfm) fabric filter.
• 2 manhours/shift operating and maintenance labor for capture
equipment and 2.5 manhours/shift operating and maintenance labor
for a 109 m3/s (230,000 acfm) fabric filter.
• 4 manhours/shift operating and maintenance labor for capture
equipment and 4 manhours/shift operating and maintenance labor
for a 142 m /s (300,000 acfm) and a 208 m/s (440,000 acfm) fabric
filter.
• Electricity requirements at 7.5 kPa (30 in. HyQ) for air curtains,
5.5 kPa (22 in. HLO) for secondary hoods, and 1.5 kPa (6 in. H90)
for fabric filters. L
6.3.2 Matte and Slag Tapping Controls
Fugitive emission control equipment for matte tapping and slag
tapping operations includes a local ventilation capture and a fabric
filter collection system. All of the smelters currently capture
fugitive emissions from the matte tapping operations. One smelter,
6-18
-------
Table 6-9. ESTIMATED ANNUALIZED COSTS OF AIR CURTAIN
SECONDARY HOODS AND FABRIC FILTERS FOR PRIMARY
COPPER SMELTERS3
Smelter
1. ASARCO - El Paso
2. ASARCO - Hayden
3. Tennessee Chemical
Co. - Copperhill
4. Inspiration - Miami
b. Kennecott - Garfield
6. Kennecott - Hayden
7. Kennecott - Hurley
8. Kennecott - McGill
9. Magma - San Manuel
10. Phelps Dodge - Ajo
11. Phelps Dodge - Douglas
12. Phelps Dodge - Hidalgo
13. Pheins Dodge - Morenci
14. Copper Range - White Pine
Annual operating cost, $
Utilities
117,600
174,400
455,200
1,123,900
372,400
819,000
1,123,900
1,123,900
1,816,500
636,000
1,123,900
819,000
522,700
455,200
Labor
0
0
87,200
145,700
87,300
94,600
145,700
145,700
145,700
0
145,700
0
87,300
87,200
Maintenance
material
0
0
29,900
59,800
29,900
37,400
59,800
59,800
59,800
0
59,800
0
29,900
29,900
Indirect
Costs
189,100
233,980
705,300
1,513,400
810,300
1,030,600
1,367,000
1,367,000
1,956,800
925,000
1,513,400
925,500
1,268,000
705,300
Total
306,700
408,400
1,277,600
2,842,800
1,300,000
1,981,600
2,596,400
2,696,400
3,978,800
1,561,500
2,842,800
1,744,500
1,908,000
1,277,500
6-19
-------
ASARCO-E1 Paso, collects the captured emissions from the matte tapping
operation. The existing capture equipment was assumed to be adequate
and thus was not included in the control cost estimate. For ASARCO-Hayden,
the matte and slag tapping capture and control costs were included in
the baseline, and therefore, were not estimated in this analysis.
A majority of smelters capture fugitive emissions from the slag
tapping operations. For those smelters that do not have slag tapping
capture systems presently, slag tapping capture system costs were
estimated based on a typical exhaust volume flow rate of 14.2 m3/s
(30,000 acfm). This flow rate was based on exhaust flow rates from
the slag tapping operations at a number of smelters. The capture
system for slag tapping operations includes tap port hoods, slag
launder hoods, sufficient length of duct to move the captured emissions
from the launder hoods to the fan and then to the main duct, and hoods
over slag ladles. The total cost, including slag launder hoods,
ducting, plenum chamber, support steel, and foundation would be $46,000
for a slag tapping capture system. Fan costs were not included in the
capture system but were included in the collection system. The annualized
operating costs were based on 2 manhours of maintenance labor per
week, electricity requirements for an 80 hp fan, capital recovery at
15 years of equipment life, and administrative overhead, taxes, and
insurance at 4 percent of capital costs.
The capital costs developed for fabric filter collection systems
for captured gases were based on data from the IGCI report prepared
for EPA. The reported costs for the fourth quarter of 1977 were
escalated to December 1982 dollars using the Chemical Engineering
Plant Cost Index (December 1982 = 314.3). The costs were estimated
2
for a 35.4 m /s (75,000 acfm) capacity fabric filter system based on
typical ventilation rates of 18.9 m3/s (40,000 acfm) from matte tapping
operations and 16.5 m /s (35,000 acfm) from slag tapping operations.
Sixty-one meter (200 ft) long ductwork was included in the cost estimate.
Annualized operating costs were estimated using the cost bases discussed
earlier, the labor requirements of 2 manhours/shift operating labor and
o
1 manhour/shift maintenance labor , and fan power requirements at 1.5 kPa
(6 in. water). Tables 6-10 and 6-11 present the capital and annualized cost
6-20
-------
Table 6-10. ESTIMATED CAPITAL COSTS OF ADD-ON FUGITIVE EMISSION
CAPTURE AND COLLECTION SYSTEMS FOR MATTE
AND SLAG TAPPING OPERATIONS
(December 1982 dollars)
1.
2.
3.
4.
5.
6.
7.
8.
y .
10.
11.
12.
13.
14.
Srnel ter
ASARCO - El Paso
ASARCO - Hayden
Tennessee Chemical
Co. - Copperhill
Inspiration-Miami
Kennecott - Garfield
Kennecott - Hayden
Kennecott - Hurley
Kennecott - McGill
Magma - San Manuel
Phelps Dodge - Ajo
Phelps Dodge - Douglas
Phelps Dodge - Hidalgo
Phelps Dodge - Morenci
Copper Range -
White Pine
No. of
smelting
furnaces
1
2
1
1
3
1
2
2
3
1
3
1
5
2
Cost, $1,000
Local
venti-
lation
capture
system
0
0
0
29
0
0
58
0
0
0
0
0
0
0
Fabric
filter
collection
system
370C
0
894
894
1,786
894
894
894
1,786
894
1,786
894
1,786
894
Total
370
0
894
923
1,786
894
952
894
1,786
894
1,786
894
1,786
894
Local ventilation system costs were based on $29,000 dollars per
smelting furnace.
'Collection system costs were based on one fabric filter unit for
smelters with one or two smelting furnaces and two fabric filter
units for smelters with three or more smelting furnaces.
'For system collecting the captured emissions from the slag tapping
operation only (captured emissions from matte tapping operations are
currently collected in an existing ESP system.)
6-21
-------
Table 6-11. ESTIMATED ANNUALIZED COSTS OF ADD-ON FUGITIVE EMISSION
CAPTURE AND COLLECTION SYSTEMS FOR MATTE
AND SLAG TAPPING OPERATIONS
(December 1982 dollars)
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
Smel ter
ASARCO - El Paso
ASARCO - Hayden
Tennessee Chemical
Co. - Copperhill
Inspiration - Miami
Kennecott - Garfield
Kennecott - Hayden
Kennecott - Hurley
Kennecott - ficGill
Magma - San Manuel
Phelps Dodge - Ajo
Phelps Dodge - Douglas
Phelps Dodge - Hidalgo
Phelps Dodge - Morenci
Copper Range -
White Pine
No. of
operating
smelting
furnaces
1
1
1
1
2
1
1
1
2
1
2
1
2
1
Costs, $1,000
Local
venti-
lation
capture
system
0
0
0
4
0
0
8
0
0
0
0
0
0
0
Fabric
filter
collection
system
153b
0
257
257
514
257
257
257
514
257
514
257
514
257
Total
153
0
257
261
514
257
265
257
514
257
514
257
514
257
For the purpose of this table, the number of operating smelting furnaces
is defined as the total number of smelters that may be tapped at a time.
3For system collecting the captured emissions from the slag tapping operaf
only. (The captured emissions from the matte tapping operation
are currently collected in an existing ESP.)
6-22
-------
for the smelter matte and slag taping fugitive emission capture and
collection systems.
6.4 COSTS OF REGULATORY ALTERNATIVES
This section presents the costs associated with the implementation
of the regulatory alternatives defined in Section 4.0 and listed in
Table 6-1. A compilation of the capital and annualized costs for
each of the regulatory alternatives for the smelters processing low
arsenic feed is presented in Table 6-12.
6.5 COST EFFECTIVENESS
The purpose of deriving cost effectiveness is to illustrate the
differences in cost relative to the arsenic emission reduction achievable
for each regulatory alternative analyzed. Cost effectiveness is
defined as incremental annualized costs in dollars per unit of pollutant
removed over the baseline.
Table 6-13 summarizes incremental annualized costs and cost
effectiveness for the regulatory alternatives for low arsenic feed
smelters individually. The table also presents annual emissions and
emission reductions projected over the baseline.
6-23
-------
Table 6-12. SUMMARY OF INCREMENTAL COSTS OF REGULATORY ALTERNATIVES OVER BASELINE
FOR CONTROL OF ARSENIC EMISSIONS FOR LOW-ARSENIC THROUGHPUT PRIMARY COPPER SMELTERS
(Costs in $l,000's December 1982 dollars)
Smel ter
1. ASARCO - El Paso
2. ASARCO - Hayden
3. Tennessee Chemical
Co. - Copperhill
4. Inspiration - Miami
5. Kennecott - Garfield
6. Kennecott - Mayden
7. Kennecott - Hurley
8. Kennecott - McGill
9. Magma - San Manuel
10. Phelps Dodge - Ajo
11. Phelps Dodge - Douglas
12. Phelps Dodge - Hidalgo
13. Phelps Dodye - Horenci
14. Copper Range -
White Pine
Total
Basel 1ne
(Existing process a
and fugitive controls)
Capital
cost
46e
;S,606f
0
0
0
0
54,044h
0
0
51, 06 7-'
0
0
95,294^
0
276,057
Annual ized
cost
40e
24,698f
0
0
0
0
i7,nih
0
0
16,356-*
0
0
30,815j
0
89,020
Effective process
controls
Capital
cost
0
0
0
0
0
0
0
10.0181
0
0
0
0
10,018
Annual ized
cost
0
0
0
0
0
0
0
4.1271
0
0
0
0
4,127
Effective process 1 Effective process controls, and
controls, and effective 1 effective fugitive controls
fugitive controls for
converters only
Capital
cost
1,375
1,702
4,434
9,825
5,197
6,731
8,760
8,760
13,050
6,731
9,825
6,731
8,526
4,434
96,081
Annual ized
cost
307
408
1,278
2,843
1,300
1,982
2,696
2,696
3,980
1,562
2,843
1,745
1,908
1,278
26,826
for smelter matte and slag ^
tapping and converter operations
Capital
cost
370
09
894
923
1,786
894
952
894
1,786
894
1,786
894
1,786
894
14,753
Annual ized
cost
153
O9
257
261
514
257
265
257
514
257
514
257
514
257
4,277
01
I
Footnotes are given on next page.
-------
Table 6-12. SUMMARY OF INCREMENTAL COSTS OF REGULATORY ALTERNATIVES OVER
BASELINE FOR CONTROL OF ARSENIC EMISSIONS FOR LOW ARSENIC THROUGHPUT
PRIMARY COPPER SMELTERS
(concluded)
aExisting fugitive controls include effective capture and collection of
roaster fugitive emissions at calcine feed smelters, effective capture
of smelter matte tapping fugitive emissions at all smelters with the
collection of captured emissions only at the ASARCO-E1 Paso smelter;
effective capture of slag tapping fugitive emissions at some smelters;
and an effective fabric filter collection system to be used for converter
fugitive emissions at ASARCO-E1 Paso.
Effective process controls include spray chamber cooling and ESP collection
systems for process particulate matter emissions from roasters, smelters,
and converters.
°Effective fugitive controls for converters include air curtain secondary
hood capture followed by fabric filter collection systems.
Effective fugitive controls for smelter matte and slag tapping operations
include local ventilation hood capture followed by a fabric filter collection
system.
Represents the cost of installation of a fugitive emission capture system
for the smelter slag tapping operation. ASARCO is required to install
a fugitive emission capture system for the smelter slag tapping operation
by July 1, 1987, under the agreement reached with OSHA.
Represents the cost of process modification to the smelter. The existing
smelting process is being modified or will be modified to convert to
the INCO smelting process in order to satisfy the accord reached between
ASARCO Smelting Co. and EPA.
9The furnace matte and slag tapping emission capture and collection costs
are included in the cost of process modification to the smelter shown
under the baseline.
Represents the cost of process modification to the smelter. Kennecott
Copper Company has announced that it plans to convert the existing
smelting process at the Hurley smelter to the INCO process.
Represents the cost of new evaporative cooling and ESP control systems
for the smelter furnace and converter process emissions.
JRepresents the cost of process modification to the smelter. The existing
smelting process is being modified or will be modified to convert to
oxygen sprinkling/oxygen fuel smelting process in order to satisfy the
accord reached between Phelps Dodge Copper Co. and EPA.
6-25
-------
Table 6-13. EMISSION REDUCTION AND COST EFFECTIVENESS IMPACTS FOR
LOW-ARSENIC THROUGHPUT PRIMARY COPPER SMELTERS
ro
Ol
$.w>H»r
1 AlftRfO - 11 Paso
7. ASARCO - llayden
1 lennessee Chpwical
f n - CopperM 1 1
4. Insplrallnn - Miami
5 Kenriecott • C.arfleld
6 Kpnnrrott - Hayden
J KpnnpcoU - Hurley
8 Kpnnecott - HrKlll
9. Magma - San Manuel
10 Phelps Dodge - Ajo
11 I'helps Dodge -
Douglas
12. Phelps Dodge -
Hidalgo
13. Phelps Dodge -
ttorpnc 1
14. Copper Range -
White Pine
IOIM S
Arsentc
feed
kq/hr
112
1 70
I.I
II!
U8
8 0
I 0
8)
7.0
47
II
14
4.S
0 7
Arsenic
Wq/yr
47.1
54 7
ft n
i i
19.6
15.3
n.n
448.1
12.0
36.1
43.0
4.3
9.5
7.6
737
Arsenic
47 1
54.7
0 8
3.1
39.6
35 3
o.n
786.4
17.0
36 1
43.0
4.3
9.5
7.6
575.3
All
Re.lucl.lrtn
from
lk,/,r
0
n
ii
n
0
0
n
161.7
n
0
0
0
0
0
!f,! 7
r-rnal Ive II
Annual l?ed Cost
10't/yr t/Mq
0
0
0
0
0
411
0
4,177 ?ri.500
0
0
0
0
0
0
4,127 25,500
Arsenic
"9/y
28,4
29.2
0.7
1.7
It. 7
79.2
0.3
406
n.2
33.5
38.7
2.6
3.4
1.7
618.13
Alternative III
Reduction
from Annual l7prl
Mq/yr 10'J/yr
18.9 107
25.0 4011
0.6 1.278
1.7 7.813
6.9 I. 300
6. I I. 912
0.5 2.696
41.1 7,696
0.8 3.980
2.6 1,562
4.3 2,843
1.7 1.745
6.1 1.908
0.9 76,876
118. Z Z6.826
Cost
VKq
16.700
16,300
7.130,000
1.677.000
188.400
374,900
5.M7.000
61.000
4.975.000
600,800
55,800
1,076,500
312,900
1,420.000
227,000
Al ternatt
Reduction
Arsenic from
Mg/yr Mq/yr
47.3 0
54.2 0
0.7 0.1
2.6 0.8
37.8 1 8
34.4 0.9
0.7 0.1
443.8 4.3
12.0 0
35.3 0.8
42.1 0.9
3.4 0.9
8.6 0.9
2.6 0
725.5 11.5
ve IV
Annual ize*1
10' J/yr
0
0
257
261
514
257
265
257
514
257
514
257
514
757
4,277
Cost
1/Mq
0
7,570.000
326.300
285.600
289.600
7,560.000
59.HOO
321.100
571.100
285,600
571.100
--
371.900
-------
6.6 REFERENCES
1. Survey of Current Business. November 1982. Primary Metal Industry
Labor Costs. December 1982.
2. Monthly Energy Review. October 1982. Electricity Costs.
DOE/EIA-003583/01. January 1983.
3. GARD Inc. Capital and Operating Costs of Selected Air Pollution
Control Systems. U.S. Environmental Protection Agency. EPA
Report No. EPA 450/5-80-002. December 1978.
4. PEDCo Environmental, Inc. Cost Estimate and Comparisons for
Converting from Reverberatory Furnace Smelting to Oxygen Flash/
Sprinkling Smelting. (Draft) U.S. Environmental Protection
Agency. EPA Contract No. 68-03-2024. November 1982.
5. Industrial Gas Cleaning Institute. Cost Estimates of Upgrading
Particulate Matter Controls in Copper Smelter Reverberatory
Furnaces. U.S. Environmental Protection Agency. EPA Contract
No. 68-02-2532. Task No. 2, March 1977.
6. U.S. Environmental Protection Agency. Draft Standards Support
and Environmental Impact Statement. Volume 1: Proposed National
Emission Standards for Arsenic Emissions from Primary Copper
Smelters. Research Triangle Park, NC. June 1978.
7. The Metallurgical Society of AIME. Copper and Nickel Converters.
Proceedings of a Symposium on Converter Operating Practices
sponsered by the TMS-AIME Pyrometallurgy
8_ ASARCO Incorporated Converter Secondary Hooding, Tacoma Plant.
Salt Lake City. Utah. January 22, 1981.
9. Reference 5.
10. Reference 3.
6-27
-------
7.0 ECONOMIC IMPACT
This section first presents an economic profile of the primary copper
industry in general, and primary copper smelters in particular (Section
7.1). The data presented in the economic profile is then used in an economic
analysis of the industry (Sections 7.2 and 7.3). The economic profile
focuses on several primary copper smelter industry characteristics, such as:
number and location of smelters, copper supplies, copper demand, competition,
substitutes, and prices.
7.1 INDUSTRY ECONOMIC PROFILE
7.1.1 Introduction
Copper's utility stems from its qualities of electrical and thermal
conductivity, durability, corrosion resistance, low melting point, strength,
malleability, and ductility. Principal uses are in transportation, machinery,
electronics, and construction.
The Standard Industrial Classification Code (SIC) definition of the
primary copper industry is the processes of mining, milling, smelting, and
refining copper. The primary copper smelters are included in SIC 3331
(Primary Smelting and Refining of Copper). Copper-bearing ore deposits and
substantial amounts of copper scrap provide the raw materials for these
processes.
In addition to producing copper, the industry markets by-product
minerals and metals that are extracted from the ore deposits, such as silver,
gold, zinc, lead, molybdenum, selenium, arsenic, cadmium, titanium, and
tellurium. Many of the companies that own primary copper facilities also
fabricate copper. Many of these same companies are also highly diversified
and produce other metals, minerals, and fuels.
The standard under consideration directly affects only one of the four
primary copper processes, namely smelting. However, the other three related
processes are an integral part of the ownership and economic structures of
copper smelters and therefore must be considered in determining industry
7-1
-------
impact. Mining and milling processes supplying a smelter will be secondarily
affected by a smelter impact because transportation costs to an alternate
smelter will add a sizeable business cost. Transportation costs for concen-
trate are significant because only 25 to 35 percent of the concentrate is
copper and the remaining 75 to 65 percent that is also being transported is
waste material. The same interdependence between smelter and refinery is not
as critical because the copper content after leaving the smelter is typically
98 percent.
Even if there were no business dependencies among the processes, the
available financial data for smelters are aggregated in consolidated financial
statements which makes smelter data difficult to isolate. Thus, an economic
analysis of copper smelters must be cognizant of the economic connection
backward to the mines and forward through the refining stage.
7.1.2 Market Concentration
Fifteen pyrometallugical copper smelters exist in the United States.
Copper is also produced in limited amounts by various hydrometallurgical
methods which by-pass the smelting stage. These hydrometallurgical fac-
ilities are not being considered in the standard setting process. The
15 copper smelters have a capacity* of 1,722,600 megagrams** of copper.
The hydrometallugical processes have a capacity of roughly 10 percent of the
copper smelters' capacity.
Table 7-1 shows that the vast majority (approximately 89 percent) of
smelting capacity is located in the southwestern States of Utah, Nevada, New
Mexico, Arizona, and Texas, close to copper mines. The location is largely
dictated by the need to minimize shipping distances of concentrates, which
are normally 25 percent to 35 percent copper.
The 15 U.S. copper smelters are owned by 7 large companies. All 7
companies are integrated in that, to various degrees, they own some mining
and milling facilities which produce copper concentrates for the smelters.
Several smelters, apart from the concentrates from their own mines, buy
additional concentrates from other mining and milling producers,
*Capacity is not a static measure of a smelter since capacity can vary, for
example, according to the grade of copper concentrates processed.
**1 megagram =1.1 short tons.
7-2
-------
Table 7-1. SMELTER OWNERSHIP, PRODUCTION AND SOURCE MATERIAL ARRANGEMENTS^
co
Smelter Name
and Location
Tacoma, WA
Hayden, AZ
El Paso, TX
Copper-hill, TN
White Pine, MI
Miami, AZ
McGill, NV
Garfield, UT
Hayden, AZ
Hurley, NM
Magma (San
Manuel, AZ)
Douglas, AZ
Ajo, AZ
Morenci, AZ
Hidalgo, NM
Total Production
Operating Rate
Ownership
ASARCO, Inc.
Cities Service Co.
Copper Range Co.
Subsidiary of The Louisi-
ana Land Exploration Co.
Inspiration Consolidated
Copper Company
Kennecott Corp.
Newmont Mining
Phelps Dodge Corp.
1J80 Rated1
Capacity
(Mg)
91,000
182,000
91,000
13,600
52,000
136,000
45,000
254,000
71,000
73,000C
181,000
115,000
64,000
191,000
163,000
l,722,600d
1979
Production
(Mg)
61,0002
96,000
85,000
12,9003
39,7704
124,0705
296,0006
56,360
145,9007
283,0008
91,000
1,291,000
(74.9%)
1980
Production
(Mg)
42.7002
59,000
47,300
10,0003
32,500*
107,4005
259,3006
46,100
98,0007
324,1008
1,026,420
(59.6%)
Material Arrangements
Integrated -
Custom
Toll
Integrated*3 -
Integrated -
Integrated -
Toll
Integrated -
Toll
Integrated -
Toll
Integrated -
Custom
Toll
Integrated -
Custom
Toll
1979
31%
25%
44%
100%
100%
35%
65%
100%
100%
-
1980
20%
44%
36%
100%
100%
58%
42%
91%
10%
100%
75%
25%
73%
6%
21%
alnformation primarily from corporate 10-K reports to the Securities and Exchange Commission.
bEstimate based on total copper sales for Cities Service minus the sales of its Arizona mines.
cEstimated to expand to 110,000 tons.
dRated capacity excluding Anaconda smelter which was closed in 1980 (rated at 180,000 Mg).
-------
smelt and refine the copper, and then sell it. This practice is referred to
as custom smelting. Other smelters process (smelt and refine) the concentrates,
and return the blister copper to mine owners for them to sell, a practice
referred to as tolling. Some smelters perform both toll and custom smelting.
It is general industry practice for companies to operate their smelters
as service centers at low profit margins to the owned mines. This acts to
shift profits of an integrated operator to the mines, where depletion allow-
ances exist. This maximizes profit to the overall operation. An implication
of this policy is that the impact on profits from swings in copper prices is
frequently manifest at the mines more than the smelters.
Table 7-1 lists the smelters, their corporate owners, capacities, 1979
and 1980 production amounts, and the distribution of integrated, custom, and
toll smelting. Total production figures and the corresponding operating
rates shown in Table 7-1 are compiled from corporate reports. Figures in
Table 7-1 are adjusted to exclude capacity and production for the Anaconda
smelter, which was closed in 1980. For 1979, the table shows a 74.9 percent
operating rate. For 1980, the table shows that the industry operated at 59.6
percent of capacity. Production was down for 1980 due to an industry strike.
Following the strike in 1980, production improved in 1981 to 1,380 gigagrams,
for a capacity utilization rate of 80 percent.9 Preliminary figures for
1982 from the Bureau of Mines show a decline in primary copper smelter
production to 1,020 gigagrams, for a capacity utilization rate of about 59
percent.10
The 3 largest companies account for 78 percent of the entire smelting
capacity. Phelps Dodge Corporation has the largest smelting capacity,
followed by Kennecott Corporation and then ASARCO. The remaining 4 companies
each have 1 smelter and in order of size are Magma (Newmont), Inspiration,
Copper Range, and Copperhill (Cities Service).
The table also shows that 73 percent of total 1980 smelter production
was from concentrate from integrated arrangements. Of the remaining concen-
trate, 21 percent was smelted on a toll basis and 6 percent smelted on
a custom basis. Three of the 8 companies process only their own copper
concentrates.
7-4
-------
7.1.3 Total Supply
Copper resources are defined as deposits which can be profitably
extracted at a given price. Various estimates of U.S. copper resources
show amounts ranging from 61.8 teragrams to 99.1 teragrams.* The capability
of copper resources to meet future demand is dependent upon several factors;
a principal one being the rate of growth in demand. The U.S. Bureau of Mines
estimates that copper demand will grow at an annual growth rate of 3.0
percent to the year 2000 and that 30 percent of the demand will be supplied
by scrap. Therefore, the likely primary copper demand over this period would
be 55 teragrams compared with 92 teragrams of resources.H Consequently,
U.S. supply appears adequate to the year 2000. Beyond the year 2000, demand
is expected to strain supply sources. However, increased uses of old scrap
and possible exploitation of sea nodules can supplement on-shore mining. In
addition, microminiaturization, copper cladding, and other conservation
methods will be more widely used to extend the supply of copper.
7-1-3-1 Domestic Supply. Primary refined copper output alone
does not depict the entire supply of copper that is available for consumption
in the United States. A large portion of copper scrap does not need to be
resmelted or re-refined and is readily available for consumption. Copper is
a durable material and, if it is unalloyed or unpainted, etc., it can be
reused readily. Otherwise, it is resmelted or re-refined as described
earlier. The ready availability of copper scrap as a secondary source of
supply tends to be a stabilizing influence on producers' copper prices.
The total supply of copper available for consumption in any one year
is therefore comprised of refined U.S. production, scrap not re-refined, net
imports, and any changes in inventory of primary refined production from one
year to the next.
The refined copper production in 1981 comprised 70.4 percent of total
copper consumed in the United States; scrap not re-refined accounted for 32.0
percent and net refined imports 10.6 percent (total exceeds 100 percent due
to stock changes).12 Between 1970 and 1981, 67 percent of U.S. copper
demand, excluding stock changes, was met from domestic mine production; 21
*Teragram is 1.1 million short tons.
7-5
-------
percent was from old scrap, and 12 percent from net imports. During these
years, total U.S. demand for copper averaged 2,012,000 megagrams per year.
Of this amount, 1,337,000 megagrams was from domestic production, 427,000
megagrams from scrap, and 248,000 megagrams from net imports.
Another statistic for describing the importance of scrap is to total
the three stages (smelting, refining from scrap, and reuse of scrap) at which
scrap can enter the production process, and compare the figures to total
copper consumption. In 1981 the percentage of total consumed copper from
scrap was 47.7, roughly the same as in recent years.
The 1981 refined copper production level was 1,956,400 megagrams.
Although the average for the past several years has shown some improvement,
total refined copper production has not returned to the 1973 peak level.
7.1.3.2 World Copper. According to the Bureau of Mines, the world
reserve of copper in ore is estimated at 494,000 gigagrams of copper. In
addition, an estimated 1,333,000 gigagrams of copper are contained in other
land-based resources, and another 689,000 gigagrams in seabed nodules. The
United States accounts for 19 percent of known copper reserves and 26 percent
of other land-based copper resources.13
The United States is the leading copper producing and consuming
country. Other major copper mining countries include: Chile, the U.S.S.R.,
Canada, Zambia, Zaire, Peru, and Poland. Although its copper mining activity
is quite small, Japan is among the three largest countries in terms of copper
smelting and refining. In 1981 the U.S. produced 18.8 percent of the world's
mine production of copper, 16.5 percent of the smelter production, and 22.2
percent of the refinery production. The consumption of the world's refined
copper by the U.S. amounted to about 21 percent. Table 7-2 shows U.S.
production, world production, and the U.S. percent of world production for
the years 1963 through 1981. Although the U.S. is essentially maintaining
its consumption and production levels, world consumption and production is
increasing. As a result, the U.S. share of world consumption and production
shows a relative decrease.
In 1981 world consumption of refined copper rose 9 percent to 9,440
gigagrams.14 Stocks of refined copper in the market economy countries
increased 5 percent to 1,100 gigagrams.15
7-6
-------
Table 7-2. UNITED STATES AND WORLD COMPARATIVE TRENDS IN COPPER PRODUCTION: 1963-198116,17
(Gigagrams)
Years
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
Average Annua)
Compound Growth
Rate (Percent)
1963-1973
1964-1974
1971-1981
~Rine Production of Copper
(Copper Content)
U.S.
1100.6
1131.1
1226.3
1296.5
865.5
1092.8
1401.2
1560.0
1380.9
1510.3
1558.5
1445.7
1282.2
1456.6
1364.4
1357.6
1443.6
1181.1
1538.2
3.54
2.48
1.09
World
4624.3
4798.6
4962.7
5215.9
5057.6
5456.5
5951.2
6387.3
6473.9
7071.5
7591.4
7885.6
6968.2
7525.3
7755.8
7618.3
7674.4
7656.3
8171.1
5.08
5.09
2.36
U.S. as
Percent
of World
23.8
23.6
24.7
24.9
17.1
20.0
23.5
24.4
21.3
21.4
20.5
18.3
18.4
19.4
17.6
17.8
18.8
15.4
18.8
_
_
-
Smelter
U.S.
1176.3
1214.2
1300.9
1330.3
782.3
1148.9
1438.3
1489.0
1360.8
1533.5
1582.1
1424.2
1357.5
1438.5
1346.8
1343.0
1396.0
1053.3
1377.7
3.01
1.61
0.12
Production
World
4634.8
4851.4
5024.4
5167.0
4891.0
5507.8
5972.9
6309.5
6380.0
7003.2
7445.5
7933.6
7535.4
7839.6
8136.9
8017.5
8045.6
7938.9
8324.7
4.85
4.77
1.03
of Copper
U.S. as
Percent
Of World
25.4
25.0
25.9
25.7
16.0
20.9
24.1
23.6
21.3
21.9
21.2
18.4
18.0
18.3
16.6
16.8
17.4
13.3
16.5
_
-
Production of
U.S
1709
1805
1942
1980
1384
1668
2009
2034
1780
2048
2098
1938
1610
1736
1706
1869
2013
1726
2037
2
0
1
Refined Copper
World
.5
.7
.1
.7
.9
.3
.3
.5
.3
.9
.0
.3
.7
.7
.9
.2
.8
.0
.6
.07
.71
.36
5399
5739
6058
6322
6000
6658
7199
7577
7377
8068
8497
8851
8402
8322
8649
8791
8903
8971
9184
4
4
2
.7
.0
.5
.2
.5
.6
.8
.8
.8
.0
.3
.5
.0
.3
.8
.9
.1
.0
.4
.64
.43
.21
U.S. as
Percent
of World
31.7
31.5
32.1
31.3
23.1
25.1
27.9
26.8
24.1
25.4
24.7
21.9
19.2
20.9
19.7
21.3
22.6
19.2
22.2
-
Note: One gigagram = 1,000 roegagrams. One megagraro (1,000 kilograms) equals 1.102311 short tons
(907.185 kilograms = 2000 pounds avoirdupois, where one pound avoirdupois equals 0.453592
kilogram or 453.5924 grams).
-------
7.1.4 U.S. Total Consumption Of Copper
Total copper consumed in the United States over the last 12 years
has fluctuated considerably but shows an overall upward trend. However,
copper consumption has not returned to its 1973 peak. This conclusion is
derived from data on copper consumption from refineries and copper consumption
from refineries plus scrap.
Table 7-3 shows each set of data for the years 1970 through 1981. The
5-year averages in gigagrams for copper consumption from refineries has
increased by 6.9 percent (1972 through 1976 is 1,891.9 and 1977 through 1981
is 2,021.5.). Five-year scrap consumption has shown an increase of 5.1
percent, from 848.6 gigagrams for the 1972 to 1976 period, to 892.3 gigagrams
for the 1977 to 1981 period. There are signs that the consumption of scrap
has begun to increase over the last few years.
The Bureau of Mines forecasts a long-range overall consumption growth
rate to the year 2000 of 3.0 percent per year. The combined 3.0 percent
growth rate is composed of a 2.4 percent growth rate for primary copper, and
a 4.8 percent growth rate for secondary copper.19
7-1.4.1 Demand By End-Use. Refined copper and copper scrap are
further processed in a number of intermediate operations before the copper is
consumed in a final product. Refined copper usually consists of one of the
following shapes: cathodes, wire bars, ingots, ingot bars, cakes, slabs, and
billets. These shapes plus the copper scrap then go to brass mills, wire
mills, foundries, or powder plants for subsequent processing. The copper is
frequently alloyed and transformed into other shapes such as sheet, tube,
pipe, wire, powder, and cast shapes. Ultimately, the copper is consumed in
such shapes in five market or end-use categories. The Copper Development
Association, Inc. uses the following categories: building construction,
transportation, consumer and general products, industrial machinery and
equipment, and electrical and electronic products.
Table 7-4 shows the demand for copper in each of these five markets
over the 12-year period 1970 through 1981. The total figures for these
five markets will not equal the total consumption figures of Table 7-3
7-8
-------
Table 7-3. U.S. COPPER CONSUMPTION^
(Gigagrams)3
T970197119721973 T974 1975T976 T9771978 1979 i960 I981r
Consumption of
Refined Copper
Consumption of
Scrapc
Total Consumed
Copper
% of Total as
Scrap
1972-1976 and
1977-1981
averages for
consumption
refined
1972-1976 and
1977-1981
averages for
consumption
of scrap
1972-1976 and
1977-1981
averages for
total copper
consumption
1859.8 1833.0 2029.5 2220.6 1998.6 1398.9 1811.7 1989.2 2196.5 2164.1 1876.3 1890.6
811.6 859.5 946.0 956.4 878.0 662.4 800.4 847.6 881.1 984.3 860.0 888.6
2671.4 2692.5 2975.5 3177.0 2876.6 2061.2 2612.2 2836.8 3077.5 3148.4 2727.3 2779.1
30.4 31.9 31.8 30.1 30.5 32.1 30.6 29.9 28.6 31.3 31.5 32.0
1891.9
2021.5
848.6
892.3
2740.5
2913.8
al gigagram = 2.2 million pounds.
^Preliminary.
cWithout having to be refined again.
-------
I
I—>
o
Table 7-4. U.S. COPPER DEMAND BY MARKET END USES20
(Gigagrams)3
Market 1970
Building Constr. 748
Transportation 267
Consumer and
General Products 369
Industrial
Machinery and
Equipment 560
Electrical and
Electronic
Products 717
Total 2661
5-Year Average
Demandc
Building
Construction
Transportation
Consumer and
General Products
Industrial
Machinery and
Equipment
Electrical and
Electronic
Products
1971 1972 1973 1974
830 923 998 795
307 345 402 332
358 417 413 402
540 601 640 563
742 773 838 766
2777 3059 3291 2858
826.0
343.4
375.2
538.2
720.2
1975 1976 1977 1978
634 780 888 956
265 372 412 422
316 328 340 388
404 483 518 551
541 683 767 801
2161 2646 2925 3118
877.0
360.4
373.2
534.4
789.8
1979 1980 1981b
951 774 816
388 270 310
420 342 376
565 501 537
853 754 774
3177 2641 28x3
ai Gigagram =2.2 million pounds,
^Preliminary.
c(1972-1976, and 1977-1981).
-------
due to the effects of stock changes and imports on fully fabricated copper
products.
A look at the 5-year average demand shows that there has been an
increase in three out of the five markets. The building industry market
sales showed a gain of 6.2 percent. The transportation market shows a gain
of 5.0 percent. An increase of 9.7 percent occurred in the electrical and
electronic product markets. The demand for electrical equipment has risen
because of increased emphasis on safety, comfort, recreation, and a pollution-
free environment. Automation, including the use in computers, has also
boosted the use of copper.
Substitution of other materials, coupled with the recession, has
caused the slight drop of less than 1 percent in the consumer and general
products markets. The 1 percent decline in the industrial machinery and
equipment market is largely due to the impact of the recession.
The Bureau of Mines estimates that the most growth in copper demand
will occur in the electrical and electronic products industries, consumer and
general products, and building construction. Copper is an important metal in
electric vehicles. If electric vehicles become popular, this would be a
source of increased demand for copper. General Motors plans to produce an
electric family car for mass marketing in the mid-1980's. A conventional
internal combustion automobile contains from 6.8 to 20.4 kg of refined
copper, whereas 'electric vehicles use much more copper. The Copper Development
Association estimates range from 45.4 kg to 90.7 kg, with an average nearer
to 45.4 kg.21
Another potential area for growth is in the solar energy industry.
Presently, the extent of this sector is relatively modest, consuming approxi-
mately 4,500 Mg/yr of copper in the U.S. However, consumption in this sector
has the potential to climb considerably.
In addition, the U.S. military demand for copper is expected to
increase. Increased military expenditures will have a significant impact on
copper demand because copper is an important element in modern electronic
weaponry. During heavy rearmament periods the military demand for the
metal has reached 18 percent of copper mill shipments. Although military
demand is not expected to return to the record high 18 percent level, analysts
do expect a large increase in military requirements for copper from the low
level in 1979 of less than 2 percent.22
7-11
-------
The demand picture in the United States may receive a boost from the
federal government. The government is committed to eventually acquire 1.1
gigagrams of copper for its currently depleted strategic stockpile. The
previous stockpile was largely depleted in 1968; the final sale was in 1974
after copper prices had soared. Further Congressional action is necessary to
implement and fund the purchase plan.
7-1-4.2 Substitutes. Substitutes for copper are readily available
for most of copper's end uses. Copper's most competitive substitute is
aluminum. Other competitive materials are stainless steel, zinc, and plastics.
Aluminum, because of its high electrical conductivity, is used extensively as
a copper substitute in high voltage electrical transmission wires. Aluminum
has not been used as extensively in residential wiring because of use problems,
and minimal savings.
Aluminum is also potentially a substitute for copper in many heat
exchange applications. For example, automobile companies are still experi-
menting with the use of aluminum versus copper in car radiators. When copper
prices are high, or copper supply is limited, cast iron and plastics are used
in building construction as a copper pipe substitute. A relatively new sub-
stitute for copper is glass, which is used in fiber optics in the field of
telecommun ications.
7.1.5 Prices
Numerous factors influence the copper market, and thus the price of
refined copper. These factors include: production costs, long-run return on
investment, demand, scrap availability, imports, substitute materials,
inventory levels, the difference between metal exchange prices and the
refined price, and federal government actions (e.g., General Services
Administration stockpiling and domestic price controls).
Among the many published copper price quotations, two key price levels
are: 1) those quoted by the primary domestic copper producers and 2) those
on the London Metal Exchange and reported in Metals Week, Metal Bulletin, and
the Engineering and Mining Journal. The producers' price listed most often is
for refined copper wirebar, f.o.b. refinery. The London Metal Exchange price,
7-12
-------
referred to as LME, is also for copper sold as wirebar. The LME is generally
considered a marginal price reflective of short-term supply-demand conditions,
while the producer price is more long-term and stable and often lags the LME
price movement.
Copper is also traded on the New York Commodity Exchange (Comex).
Arbitrage keeps the LME price and the Comex price close together (with minor
price differences due to different contract terms on the two exchanges, and
a transportation differential).
Table 7-5 shows the LME, the U.S producer price, and the U.S. producer
price adjusted to a 1982 constant price for the years 1970 through 1982.
Data were obtained from U.S. Bureau of Mines publications.
Several points can be observed from the table with respect to the LME
price versus the U.S. producer price: (1) the LME price has had wider
swings than the producer price; (2) in the past when both prices are relatively
high, the LME price has been considerably higher than the producer price,
while during relatively low price periods, the producer price has been
moderately higher than the LME price; and (3) in recent years a marked change
appears to be taking place away from a two-price system and toward a one-price
system, with the difference between the LME and the U.S. producer price
accounted for only by a transportation differential. These earlier situations
had reoccurred repeatedly over the past 20 years. One other point about
the table should be mentioned, although unrelated to the relationship of the
LME to the producer price. The producer price has not kept pace with general
inflation.
In theory, the U.S. producer price should be somewhat higher than the
LME price since ocean transport costs must be incurred to get the refined
copper to the U.S. However, this relationship appears to hold only during
slack price periods. When LME prices are high, the producers do not raise
their prices as much, which in theory appears contrary to profit maximization.
Explanations offered for such behavior include: the producers' fear of
long-run substitution for copper if the producers raised the price to the
fabricators, high profits for integrated fabricators while reducing supply to
nonintegrated fabricators, past fears of government stockpile sales that
would reduce prices, and fear of the return of government intervention
through price controls.
7-13
-------
Table 7-5. AVERAGE ANNUAL COPPER PRICES23,24,25
(cents per kg)a
Year
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
19826
LMEb
138.6
106.7
106.7
178.0
204.8
123.4
140.6
130.7
136.2
198.2
218.5
174.7
147.4
U.S Producer Pr1cec
128.0
114.4
112.6
130.9
170.1
141.2
153.1
147.0
146.3
205.3
225.3
187.2
162.8
U.S. Producer
1982 Constant
290.9
248.7
234.6
256.7
303.8
231.5
239.2
216.2
200.4
259.9
262.0
199.1
162.8
Price
Priced
aTo convert from cents/kg to cents/lb, multiply by 0.454.
bLondon Metal Exchange "high-grade" contract.
CU.S producer price, electrolytic wirebar copper, delivered U.S destinations
basis.
dAdjusted to 1982 constant price by applying implicit price deflator for
gross national product (1972 = 100).
ePreliminary.
7-14
-------
The cost of producing copper is one of the elements that influences
the price of copper. Considerable data exist to validate the point that the
long-run economic cost of producing copper is increasing. 26 During the
early 1970's the capital costs per megagram of annual capacity for developing
copper from the mine through refining stage were $2,000 to $2,500, and by the
late 1970's had risen sharply to $7,200 to $7,700. Estimates are that a
price of $2.76 per kg to $3.30 per kg for refined copper would be needed to
support such new capital outlays.
The above costs are for conventional pyrometal lurgical smelting. The
newer smelting processes such as Noranda and Mitsubishi offer some capital
cost savings at that stage due to lower pollution control costs. The hydro-
metallurgical processes also reouire less capital. However, the mining costs
are the highest part of overall development costs for which limited cost
saving techniques exist. The mine development costs in the U.S. have risen
significantly, largely as a result of the shifting from higher to lower
grades of available copper ores and sometimes remote locations that require
infrastructure costs for towns, roads, etc.
In 1979, the Bureau of Mines analyzed 73 domestic copper properties to
determine the quantity of copper available from each deposit and the copper
price required to provide each operation with 0 and 15 percent rates of
return. The Bureau estimates that a copper price of $4.56 per kg would be
required if all properties, producing and nonproducing, were to at least
break even. The average break-even copper price for properties producing in
1978, $1.46 per kg, was about equivalent to the average selling price for
the year. At this price, analysts calculate that only 25 properties could
either produce at break-even or receive an operating profit. Of these proper-
ties, only 12 could receive at least a 15 percent rate of return
Annual domestic copper production, from 1969 to 1978, averaged 1,337,000
megagrams. According to this study, in order to produce at this level and
receive at least a 15 percent rate of return, a copper price of $1.81 per kg
is required. If the United States were to produce the additional 248 000
megagrams that were imported each year over this period, a copper price of
$1.94 would be necessary.27 Tne report concludes ^ .^^ ^
prices are required in order for many domestic deposits to continue to
produce.
7-15
-------
It has been suggested that long-term potential for higher prices,
plus the high cost of new capacity are significant reasons for the increased
purchases several years ago of copper properties by oil companies. The
reasoning is that oil companies need places for heavy cash flows, and diver-
sification to other products is desirable. The oil companies reportedly can
wait for expected copper price increases to obtain their return. Further, by
purchasing existing facilities, rather than building new capacity, they avoid
the escalating new capacity costs. However, more recently, some oil com-
panies seems to be rethinking their investments in copper.
As shown below, U.S. oil (and gas) companies own or have major interests
in many of the largest domestic copper producers:
1. Amax - Approximately 20 percent owned by Standard Oil of California
2. Anaconda - Owned by Atlantic Richfield Company (ARCO)
3. Cities Service - Also a primary copper producer
4. Copper Range - Owned by Louisiana Land and Exploration Company
5. Cyprus Pima Mining Company - Standard Oil Company (Indiana)
6. Duval - Owned by Pennzoil Company
7. Kennecott - Standard Oil of Ohio (British Petroleum)
These copper producers own or control a large portion of domestic copper
reserves, mine production, and U.S. refinery capacity. Their investment in
the copper industry is significant, and thus they must expect higher prices
and substantial profits in the future.
7-16
-------
7.2 ECONOMIC ANALYSIS
7.2.1 Introduction
This section presents the economic impact analysis of the arsenic
NESHAP for the 14 low arsenic primary copper smelters. The fifteenth
primary copper smelter is classified as a high arsenic smelter and is dis-
cussed in a separate analysis.
The principal economic impacts analyzed are: the ability of the
smelters to increase copper prices in response to an increase in costs due to
the arsenic standard; and, the impact on profits if part or all of the costs
cannot be passed on in the form of price increases. Section 7.2.3 presents
the methodology. Section 7.2.4 presents the impact on prices, Section 7 2 5
presents the impact on profits, and Section 7.2.6 presents a discussion of
capital availability.
7.2.2 Summary
In 1982 the copper producers experienced one of the worst years in
recent history. Such a situation cannot be used as the foundation to examine
the long term economic impact of the potential arsenic NESHAP. Therefore the
economic analysis is based on a more normal condition for the industry
However, even under more typical conditions for the industry, several smelters
may face significant financial impairment.
If each smelter attempts to pass control costs forward in the form of
a price increase, the price increases would range from 0 percent to 15.2
percent at a 80 percent capacity utilization rate, depending on the regulatory
alternative. For Alternative II the price increase would be 0 for every
smelter, with one exception that would have a 15.2 percent increase. For
Alternative III the price increases would range from 0.1 percent to 6 3
percent. For Alternative IV the price increases are lower and would range
from 0 to 1.3 percent. For Alternative III+IV the price increases would
range from 0.1 percent to 7.6 percent. Competition will prevent the existence
of such a broad variation.
If control costs are absorbed and profit margins reduced, again a
broad range exists. At an 80 percent capacity utilization rate and a ten
percent profit margin, for Alternative II the profit decrease would be 0 for
every smelter, with one exception that would result in a net loss For
7-17
-------
Alternative III the profit decrease would range from 1.1 percent to 62.6
percent. For Alternative IV the profit decrease would be lower and would
range from 0 percent to 12.8 percent. For Alternative III+IV the profit
decrease would range from 1.1 percent to 75.4 percent.
Although the capital costs of the control equipment are not minor
amounts, for most of the producers the capital cost would not present a major
obstacle. For two of the producers the capital costs may present some
difficulty.
7.2.3 Methodology
The purpose of this section is to explain in general terms the method-
ology used in the analysis. Each of the appropriate sub-sections explains
the methodology in more detail. No single indicator is sufficient by itself
to use for decision making purposes about the primary copper smelters.
Therefore the methodology relies on several indicators which in total can be
used to draw conclusions about the industry.
The methodology has three major parts. The first part is an analysis
of price impacts. The analysis of price impacts introduces an upper limit on
the problem and provides a benchmark to make evaluations on a relatively
uncomplicated basis. A price increase represents the "worst case" from the
viewpoint of a consumer of copper. The second major part of the methodology
is an analysis of profit impacts. The analysis of profit impacts introduces
a lower limit on the problem and is the "worst case" from the viewpoint of
the firm. The individual characteristics of each smelter increase in
importance and are incorporated to a greater extent. The third and final
part is an analysis of the availability of capital to purchase the control
equipment.
Firms in the copper industry face a wide variety of variables that in
the aggregate determine the economic viability of the firm generally, and a
smelter specifically. The variables can be grouped in four broad categories.
The categories are described here separately and in a simplified manner for
discussion purposes. However, there is a close interrelationship among the
four categories and changes in one will have implications for the others.
The four broad categories which encompass the variables that in turn deter-
mine the economic viability of the smelter are described below.
7-18
-------
1) Macro-economic conditions. The two most prominent variables in
this category are copper prices and copper demand. By-products and co-products
represent a significant source of revenues for most copper operations.
Therefore in addition to the price of copper, the price of by-products and
co-products also influence an assessment of economic viability. Common
by-products and co-products of copper production include: gold silver
molybdenum, and sulfuric acid. Other by-products include selenium, teilurium
and antimorv. For ease of presentation and in order to present a conserva-
tive analysis, by-products and co-products are not considered explicitly in
the analysis. Another important variable, though, somewhat less visible is
government actions such as tax policy, stockpiling, and price controls. Vhe
government variable includes the U.S. Government, as well as foreign govern-
ments. For example, consider that a report by the U.S. Bureau of Mines
has stated that at least 40 percent of the total mine production of copper in
market economy countries was produced by firms in which various foreign
governments owned an equity interest. 28
2) Environmental regulations. Since roughly 1970, environmental
regulations have evolved to the point that they have become a major variable
that must be considered in the corporate decision making process. Here
again, government actions are important.
3) Corporate organizational strategy. This category includes the
corporation's strategy with respect to variables such as remaining or becom-
ing an integrated copper producer versus a non-integrated copper producer or
leaving the industry entirely.
Many of the companies that produce refined copper are integrated
producers; that is, they own the facilities to treat copper during each of
the four principal stages of processing: mining, milling, smelting, and
intth, K' Pr°UCerS 3re 1nt69rated °"e «"»«on.l step
into the fabncat,on of refined copper. However, not all companies in the
pper industry are integrated producers. There are companies that only mine
and mill copper ore to produce copper concentrate, and then have the copper
concentrate smelted and refined on a custom basis (the smelter takes owner-
of e copper) or on a toll basis (the smelter charges a service fee and
returns the copper to the owner). The existence of both integrated and
non-integrated producers introduces a complex economic element into this
7-19
-------
analysis. That complex economic element manifests itself in the choice of
the appropriate profit center. This standard affects only one stage of the
production process (smelting) in a direct way, but has indirect effects on
the other stages.
For accounting purposes, integrated producers frequently view the
smelter as a cost center, rather than a profit center. However, in an economic
sense the smelter provides a distinct contribution to the production process
that ultimately allows a profit to be earned although that profit may be
realized for accounting purposes at another stage such as the mine or refinery.
4) Competition. Mines have long-run flexibility in deciding where
they will send their copper concentrate for smelting. Therefore, copper
smelters face competition from three sources: other existing domestic smelters,
new smelters that may be built, and foreign smelters, especially Japanese.
Other competition, though less direct, is also important. For example, scrap
and substitutes present competition.
Japan is a major force among copper producing countries in terms of
its volume of smelting, refining, and fabrication of copper. However, Japan
does not have copper ore deposits of any noteworthy size. Therefore it must
import concentrates in order to supply its smelting, refining, and fabri-
cating facilities. Japan seeks concentrates from many countries, including
the United States. Japan's ability to be competitive with domestic smelters
for U.S. concentrates is indicated by the contractual arrangements it has
established with Anamax and Anaconda to purchase concentrates. Also, the
Japanese smelters have approached many other copper mine owners in the United
States. For example, Cyprus Corporation is reported to have seriously
considered shipping concentrates from its Bagdad mine to Japan.
The cost to transport concentrates across the Pacific Ocean is signi-
ficant. The fact that Japanese smelters can compete with U.S. smelters, in
spite of the costs to transport concentrates across the Pacific Ocean, is
quite noteworthy. One factor that explains the Japanese ability to compete
is that Japanese smelters are newer than U.S. smelters and, in theory, should
be more cost competitive. Other factors that operate to the advantage of
Japanese smelters, including a tariff mechanism, are described later.
The existence of competition for concentrates introduces what is
commonly referred to as a "trigger" price. The "trigger" price is that price
7-20
-------
which triggers or provides an economic incentive for the supplier of concen-
trate to change to another smelter and refinery. If a given smelter charges
a service fee in excess of competing smelters, that smelter will lose business
and eventually be forced to cease operations. In the case of new smelters or
expansions, the new process facilities will not be built. Faced with an
increase in costs, a smelter could respond using one of three options, or any
combination of the three. First, the smelter could pass the costs forward in
the form of a price increase. Two important considerations with respect to a
pnce increase are: the prices of competitors in the copper business, and the
elasticity of demand for the end users of copper. For example, even if all
copper producers experience the same increase in costs, at some point the end
users of copper will consider changing to a substitute. Second, the smelter
could absorb the cost increase by reducing its profit margins, thereby
reducing its return on investment (ROD. If the smelter's profit margins are
reduced significantly it will cease operation. Third, the smelter could pass
the costs back to the mines by reducing the price it is willing to pay for
concentrate. An important consideration in setting the service fee a smelter
charges for custom or toll smelting is that the concentrate may be shipped
elsewhere, such as to Japan. Market conditions suggest that the option of
passing costs back to the mines does not seem feasible at this time, due to
the existence of excess smelting capacity.
7'2'3"1 JaPa"ese Tarif^jecjianisrn. One example of foreign government
assistance to the copper industry occurs in Japan. Japanese copper producers
operate under a system that permits the payment of a premium for concentrates
which is then recovered through a premium for refined copper due to a protected
internal market supported by a high tariff. Japan imposes high import duties
on refined, unwrought copper while allowing concentrates to be shipped into
the country duty-free. Duty on refined unwrought copper in 1981 was 8 2
percent of the value of the copper, including freight and insurance as
opposed to a U.S. customs duty of 1.3 percent of the value of copper. The
^port duties allow Japanese producers to sell their refined copper in Japan
an artificially high price and still remain competitive with foreign
rln/~Qi^o vnv-iyii
producers.
Specifically, copper concentrates and ore imported into Japan are free
of duty. Refined copper imported into Japan is subjected to a tariff of
7-21
-------
15,000 yen/Mg.29 Using a December 15, 1980, exchange rate of $0.004633/yen,
the tariff was $0.0849/kg. Refined copper may be duty-free under the preferen-
tial tariff, subject to certain limitations.
As a result of the tariff situation, Japanese copper producers can pay
a premium to attract concentrates and can recover the premium through a
premium on the price of the refined copper used in Japan. If the refined
copper is returned to the customer outside of Japan, the premium on the price
of refined copper is not recovered because world prices would prevail in this
case, rather than the protected internal Japanese producer price. As a
result, the principal interest of the Japanese copper producers is in produc-
ing copper for internal consumption. Toll smelting in Japan is generally
used as a means of balancing inventories. The absence of a tariff on ore and
concentrates encourages companies to import ore into Japan. The presence of
a tariff on refined copper and the costs of holding metal in Japan discourage
companies from importing refined copper into Japan.
The Japanese tariff on refined copper, combined with the cost of
holding the metal until users have a demand for it, provides an extra margin
for domestic copper producers. The Japanese producers can charge what the
market will bear for their copper and still remain competitive with the
importers. The loss incurred by Japanese producers in charging toll custo-
mers low processing rates is covered by the extra margin of profit realized
by charging prices for domestic refined copper at competitive import levels.
Robert H. Lesemann (industry expert, formerly with Metals Week, now
with Commodities Research Unit), in an affidavit for the Federal Trade
Commission, outlined the situation in September 1979:
It is generally true that operating costs of U.S. smelters
are the same as smelters in Japan, Korea, and Taiwan. The
competitive advantage is without doubt due to the subsidies
outlined above. Thus, while the terms of the Nippon-Amax
deal have not been revealed, the treatment charge is likely
well below the operating cost levels of U.S. smelters.30
7.2.3.2 Other Japanese advantages. The tariff mechanism described
above is one example of government assistance to the Japanese copper industry.
Another example is provided by the Japanese government's approval of a brass
7-22
-------
rod production cartel. In an effort to reduce stocks and boost profit
margins for the ailing Japanese brass rod industry, the government approved
the formation of a temporary cartel to cut production.31
Apart from government assistance, other reasons are cited for the
advantage of the Japanese copper industry over the U.S. copper industry.
Additional reasons include:
• A high debt-to-equity ratio-a typical Japanese smelter may have a
debt-to-equity ratio of 0.8 to 0.9.32,33,34
. Lower labor rates-Japanese hourly rates in the primary metals
industry were estimated to be about two-thirds of the U.S. rate in
1978.35
. By-product credits-the market for by-products, sulfuric acid, and
gypsum is better in Japan than in the United States and reduces
operating costs significantly.36
7-2-4 Maximum Percent Price Increase
Insight into the economic impact of the arsenic NESHAP can be gained
by examining the maximum percentage copper price increase that would occur if
all control costs were possed forward. A complete pass forward of control
costs may not be possible in every case, and later in the analysis this
assumption is relaxed. However, the initial assumption that a complete pass
forward is possible in every case introduces a common reference point, which
then facilitates comparisons of various control alternatives and scenarios
The maximum percentage price increase is calculated using a simplified
approach, for ease of presentation, that divides annualized control costs by
the appropriate production and further divides that result by the refined
price of copper, with the result expressed as the necessary percentage price
increase per kilogram. The above approach does not consider the investment
tax credU, and thus is a conservative approach that will tend to overstate
the impact of the control costs. Other approaches could be used to determine
price increases. For example, a net present value (NPV) approach could be
7-23
-------
used. A net present value approach determines the revenue increases necessary
to exactly offset the control costs, such that the NPV of the plant remains
constant. An NPV analysis can also take into account the investment tax
credit, depreciation over the applicable time period, income taxes, operating
and maintenance costs, and the time value of money. Although the NPV approach
is a more sophisticated calculation, the two approaches yield similar results.
Therefore, the first method is preferable in this particular case due to its
straightforward nature, ease of presentation, and reasonable results.
Table 7-6 shows the cost increase, and then Table 7-7 shows the
maximum percentage price increase, of arsenic controls for low-arsenic
primary copper smelters. The increase in the cost of production is shown for
two capacity utilization rates, 100 percent and 80 percent. The advantage of
presenting two capacity utilization rates is in the conduct of sensitivity
analysis. A rate of 100 percent is optimistic, but is useful here as a
reference point. A rate of 80 percent is more likely and as noted in Section
7.1 this is the approximate industry average utilization rate achieved in
1981. For 1982, the industry average capacity utilization rate was substan-
tially lower at 59 percent. However, no analysis is shown here of the impact
of control costs on the industry at a 59 or 60 percent utilization rate
because regardless of control costs, a rate of 60 percent is damaging to the
industry even as a baseline condition. Alternatives II, III, and IV are
shown as well as the combination of III+IV. The smelters are ranked accord-
ing to the cost of Alternative III+IV (with the exception of Kennecott-McGill).
The purpose of showing the increase in production cost is to supplement the
maximum percentage price increase. One advantage of reviewing the cost
increase is that it is only dependent on the capacity utilization rate, and
is not affected by the refined price of copper. A second advantage is that
it is not affected by the choice of the profit center. Several points should
be observed from the cost increases:
1) The amount of the cost increases are substantial for two of the
smelters. The cost increases are substantial for several reasons. First,
copper is a commodity, which means that product differentiation is not
possible and thus competition is based almost exclusively on price. The
copper producers can be characterized as price-takers and thus no individual
producer controls the marketplace. Therefore, in an industry that competes
based on price, the cost of production becomes exceptionally important.
7-24
-------
Table 7-6. INCREASE IN COST OF PRODUCING COPPER DUE TO ARSENIC
CONTROLS FOR LOW-ARSENIC PRIMARY COPPER SMELTERS
IN3
en
Smel ter
ASARCO-Hayden
ASARCO-E1 Paso
Kennecott-Garfield
Phelps Dodge-Hidalgo
Phelps Dodge-Morenci
Inspiration
Magma
Phelps Dodge-Ajo
Phelps Dodge-Douglas
Copper Range- White Pine
Kennecott-Hayden
Kennecott-Hurley
Cities Service-Copperhill
Kennecott-McGill
Annual
Capacity
(Mg.)
182,000
91,000
254,000
163,000
191,000
136,000
181,000
64,000
115,000
52,000
71,000
73,000
13,600
45,000
•— • " i —
II
0
0
0
0
0
0
0
0
0
0
0
0
0
2.7
100% Capacity
Alternative
HI TV
0.2
0.3
0.5
1.1
1.0
2.1
2.2
2.4
2.5
2.5
2.8
3.7
9.4
6.0
0
0.2
0.2
0.2
0.3
0.2
0.3
0.4
0.4
0.5
0.4
0.4
1.9
0.6
HTJ.TU
1 + 1 y
0.2
0.5
0.7
1.3
1.3
2.3
2.5
2.8
2.9
3.0
3.2
4.1
11.3
6.6
IT
I
0
0
0
0
0
0
0
0
0
0
0
0
0
28.4
80% Capacity
Alternative
I
0.2
0.4
0.6
1.3
1.2
2.6
2.7
3.1
3.1
3.1
3.5
4.6
11.7
7.5
IV
0
0.2
0.3
0.2
0.3
0.2
0.4
0.5
0.6
0.6
0.5
0.5
2.4
0.7
III+IV
0.2
0 6
fl Q
1.5
1 C
2 8
•? i
3 6
•3 7
o . /
•3 7
A n
t. u
*; i
j.i
14 1
8.2
-------
Table 7-7. MAXIMUM PERCENT PRICE INCREASE FOR ARSENIC CONTROLS FOR LOW-ARSENIC PRIMARY COPPER SMELTERS
I
ro
—~IE
1777
100% Capacity
Alternative
Smel ter
ASARCO-Hayden
ASARCO-E1 Paso
Kennecott-Gar field
Phelps Dodge-Hidalgo
Phelps Dodge-Morencl
Inspiration
Magma
Phelps Dodge-Ajo
Phelps Dodge-Douglas
Copper Range-White Pine
Kenn.ecott-Hayden
Kennecott-Hurley
Cities Servlce-Copperhll
Kennecott-McGill
II
0
0
0
0
0
0
0
0
0
0
0
0
1 0
12.1
III
0.1
0.2
0.3
0.6
0.5
1.1
1.2
1.3
1.3
1.3
1.5
2.0
5.0
3.2
IV
0
0.1
0.1
0.1
0.2
0.1
0.2
0.2
0.2
0.3
0.2
0.2
1.0
0.3
III+IV
0.1
0.3
0.4
0.7
0.7
1.2
1.4
1.5
1.5
1.6
1.7
2.2
6.0
3.5 15
II
0
0
0
0
0
0
0
0
0
0
0
0
0
.2
Tcg^r
rTce~
80% Capacity
Alternative
III
0.1
0.2
0.3
0.7
0.6
1.4
1.4
1.3
1.7
1.7
1.9
2.5
6.3
4.0
IV
0
0.1
0.2
0.1
0.2
0.1
0.2
0.3
0.3
0.3
0.3
0.3
1.3
0.4
III+IV
0.1
0.3
0.5
0.8
0.8
1.5
1.6
1.6
2.0
2.0
2.2
2.8
7.6
4.4 10
II
0
0
0
0
0
0
0
0
0
0
0
0
0
.3
220 d/kg.PHce
100% Capacity
Alternative
III
0.1
0.1
0.2
0.5
0.5
1.0
1.0
1.1
1.1
1.1
1.3
1.7
4.3
2.7
IV
0
0.1
0.1
0.1
0.1
0.1
0.1
0.2
0.2
0.2
0.2
0.2
0.9
0.3
1II+IV
0.1
0.2
0.3
0.6
0.6
1.1
1.1
1.3
1.3
1.3
1.5
1-9
5.2
3.0 12
11
0
0
0
0
0
0
0
0
0
0
0
0
0
.9
80% Capacity
Alternative
HI
0.1
0.2
0.3
0.6
0.5
1.2
1.2
1.1
1.4
1.4
1.6
2.1
5.3
3.4
TV
0
0.1
0.1
0.1
0.1
0.2
0.2
0.1
0.3
0.3
0.2
0.2
1.1
0.3
III+IV
0.1
0.3
0.4
0.7
0.6
1.3
1.4
1.3
1.7
1.7
1.8
2.3
6.4
3.7
-------
Second, copper is traded on an international basis and thus domestic pro-
ducers compete among themselves, as well as against foreign producers that
may not experience the same cost increases. Finally, copper is faced with a
significant threat from substitutes: such as, aluminum and plastic.
2) Within a single alternative, the differences among smelters are
substantial. As described above, copper producers compete principally on
price. As a result, the cost of production is quite important. Therefore
differences in costs among smelters of as little as several cents are
important.
3) The cost increases for Alternative II are 0 in every case with one
exception, Kennecott-McGill. The cost increases for Alternative III range
from a low of 0.2*/kg to a high of 9.4*/kg. The costs for Alternative IV
are lower, and range from 0 to 1.9^/kg. The costs for Alternative III+IV
range from 0.2*/kg to 11.3*!/kg.
Table 7-7 shows maximum percentage price increases. The purpose of
reporting the maximum percentage price increase figures is to add perspective
to the cost increase figures. Results are shown for two refined copper
prices (187 cents per kg. and 220 cents per kg.), and for the same two
capacity utilization rates presented earlier, 100 percent and 80 percent.
The price increase assumes the firm is an integrated producer. The average
annual price for refined copper over the past 5 years, from 1978 to 1982 has
been approximately 187 cents per kg. The same cases are shown as were
presented earlier for the cost increases, Alternatives II, III, IV and
III+IV. The price of copper is difficult to predict, and therefore prudence
suggests examination of a second price. As shown previously in Section 7 1
the highest average annual current dollar price for refined copper was 225 3
cents per kilogram, achieved in 1980. (The year 1980 was marked by an
industry strike and reduced production.) Therefore, 220*/kg is used to
represent a price that based on the results of past years, appears optimistic.
An alternative "pessimistic" price is not presented because even the baseline
results are highly likely to be damaging and thus the addition of control
costs would merely reinforce an obvious conclusion. A ready example of the
impact of a price significantly below 187*/kg was provided in 1982 when the
average price was about 163*/kg and large segments of the industry closed
for sustained periods.
7-27
-------
The analysis of the results for the maximum percentage price increase
figures is similar to the analysis discussed above for the cost increase
figures. Once again, for Alternative II only Kennecott-McGill experiences a
price increase. The price increase is 12.1 percent based on a 100 percent
capacity utilization rate and a price of 187^/kg. For Alternative III the
maximum price increases range from 0.1 to 5.0 percent. For Alternative IV
the price increases are lower, and range from 0 to 1.0 percent. For Alterna-
tive III+IV the price increases range from 0.1 to 6.0 percent, with only two
smelters above 2.2 percent. There is some variation in the price increases
among the smelters. The significance of the variation in the maximum percentage
price increases among the smelters is that those smelters with higher price
increases would probably be constrained in the marketplace by the lower price
increase smelters. As a result, some of the smelters could quite possibly
have to absorb a part of the control costs. As mentioned above, two additional
constraining influences are foreign competition and substitutes.
7.2.5 Profit Impacts
Apart from the calculation of maximum percentage price increase,
additional insight into the economic impact of the arsenic NESHAP can be
gained by making the opposite assumption from maximum percent price increase,
that is, zero percent price increase, or complete cost absorption. The
assumption of complete control cost absorption provides a measure of the
reduction in profits if the control costs are absorbed completely.
Assuming control costs are absorbed, the critical element in an
analysis of profit impacts is the profit margin. The larger a firm's profit
margin, the greater is the firm's ability to absorb control costs and earn an
acceptable rate of return on investment (ROI), and thus continue operation.
The profit margin is simply the difference between price and cost. As
mentioned in an earlier section, the central issue becomes the choice of an
appropriate profit center and its corresponding price and cost. The process-
ing of virgin ore into refined copper involves four distinct steps: mining,
milling, smelting, and refining. Although the four steps are often joined to
form an integrated business unit, they are not inextricably bound together in
an economic sense. For example, it is not uncommon for mines to have their
concentrate toll smelted and refined. The difficulty that this variability
7-28
-------
presents In terms of an assessment of the Impact of the arsenic standard Is
1n the method of assigning the costs.
This report presents an analysis of profit impacts using two methods.
The first method assumes copper producers are fully integrated and all have
the same costs and thus earn a uniform profit margin. The objective of this
method is to permit a ready, though simplified, examination of profit impacts
With the first method as a foundation, the second method introduces more
smelter specific variables into the analysis in an effort to focus more
sharply on the complex organizational structure of the industry.
7.2.5.1 Method One. As mentioned above, the critical element in an
examination of profit impact is the profit margin. Therefore an examination
of profit margins for members of the industry is necessary, and accordingly
is presented below. Table 7-8 shows the revenues and operating profit
(before tax) for each of the 7 producers for the 5-year period from 1977
to 1981. Table 7-8 also shows the percentage profit margin, which is operat-
ing profit divided by revenues. The revenue and operating profit figures are
for the business segment within the company that includes copper. The use of
business segment information provides a closer representation of the results
for copper than would the use of the consolidated results for the company
The reason for this is that for several of the firms copper represents a
relatively small share of the total company results. Although the business
segment information is a better representation of the results for copper than
the total company results, the business segments contain other products in
addition to copper. Therefore conclusions must be drawn accordingly The
table shows that there is considerable variation in results, both within a
company from one year to the next, as well as from one company to the next
The 5-year average ranges from a loss of 3.6 percent to a high of 13.8
percent.
Table 7-9 shows the maximum percentage reduction in the profit margin
for each of the 14 smelters. This table assumes each smelter is viewed as
part of a fully integrated operation. Two profit levels are shown and two
capacity utilization rates (100 percent and 80 percent). The first profit
level is based on a refined copper price of l87*/kg and a 10 percent
profit margin, which yields a profit of l8.7*/kg. The second profit level
7-29
-------
Table 7-8.
BUSINESS SEGMENT RETURN ON SALES FOR COPPER COMPANIES9
($ 103)
Revenues
Year
1977
1978
1979
1980
1981
Operating 1977
Profitd 1978
1979
1980
1981
Profit/
Revenues
(percent)
1977
1978
1979
1980
1981
Average
ASARCO
733,293
849,002
1,339,917
1,440,220
1,153,022
65,919
112,474
225,763
145,286
68,364
9.0
13.2
16.8
10.1
5.9
11.0
Cities
Service
184,000
241,500
276,300
224,100
MA
(38,600)
(23,900)
25,400
16,300
NA
(21.0)
(9.9)
9.2
7.3
NA
(3.6)
Copper
Range"
NA
64 , 600
89 , 300
83 , 900
NA
NA
(6,600)
10,000
1,800
NA
NA
(10.2)
11.2
2.1
NA
1.0
Inspiration
95,676
101,251
136,849
178,004
NA
(9,994)
(6,235)
9,889
(6,563)
NA
(10.4)
(6.2)
7.2
(3.7)
NA
(3.3)
Kennecott
NA
683 , 000
1 , 091 , 400
987 , 400
NA
NA
(100)
164,000
131,400
NA
NA
0
15.0
13.3
NA
9.4
Magmac
NA
274,137
381,512
287,581
328,842
NA
13,601
67,252
11,522
(15,658)
NA
5.0
17.6
4.0
(4.8)
5 5e
Phelps
Dodge
453 184
446,970
618,188
714,591
706,404
52,831
63,738
159,428
95,439
27,618
11.7
14.3
25.8
13.4
3.9
13 .8f
'Business segments contain other products in addition to copper.
"The Louisiana Land and Exploration Company.
cProfit is net income after tax in this case.
^Before interest and tax.
ewould yield 7.9 percent if adjusted to before tax with an effective tax rate of
30 percent.
^Imputed
intersegment sales for 1977 to 1981 would yield average return
7-30
-------
Table 7-9.
DECREASE FOR ARSENIC CONTROLS FOR LOW-ARSENIC PRIMARY COPPER SMELTERS
Smel ter
ASARCO-Hayden
ASARCO-E1 Paso
Kennecott-Garfield
Phelps Dodge-Hidalgo
Phelps Oodge-Morenci
Inspiration
Magma
Phelps Dodge-Ajo
Phelps Dodge-Douglas
Copper Range-White Pine
Kennecott-Hayden
Kennecott-Hurley
Cities Service-Copperhill
Kennecott-McGill 121
• • — • „
0
0
0
0
0
0
0
0
0
0
0
0
0
.4C
"TfiOT
_18-7*Ag.
TlnafJi.. *~
Alternative
III IV I1I+IV
1-1 0 1.1
1.6 1.1 2.7
2.7 1.1 3.7
5-9 1.1 7.0
5.3 1.6 7.0
11.2 1.1 12.3
11-8 1.6 13.4
12.8 2.1 15.0
13.4 2.1 15.5
13.4 2.7 16.0
15.0 2.1 17.1
19-8 2.1 21.9
50.3 10.2 60.4
32.1 3.2 35.3
• , _
—
Profit Marqina
n_
0
0
0
0
0
0
0
0
0
0
0
0
0
151. 9C
BOX Caf
AHerna
HI IV
•- _
1.1 0
2.1 1.1
3.2 1.6
7.0 l.l
6.4 1.6
13.9 1.1
14.4 2.1
12.8 2.7
16.6 3.2
16.6 3.2
18.7 2.7
24.6 2.7
62.6 12.8
40.1 3.7
— . — . — ,_
>acity
itlve
III+IV
1.1
3.2
4.8
8.0
8.0
15.0
16.6
15.5
19.8
19.8
21.4
27.3
75.4
43.8
. . — ___
51 .7 tf/ko.Prom
"100* Capacity—
Alternative
II
0
0
0
0
0
0
0
0
0
0
0
0
0
43.9
LLL IV II1 + IV
0.4 0 0.4
0.6 0.4 1.0
1.0 0.4 1.4
2.1 0.4 2.5
1-9 0.6 2.5
4.1 0.4 4.5
4.3 0.6 4.9
4.6 0.8 5.4
4.8 0.8 5.6
4-8 1.0 5.8
5.4 0.8 6.2
7.2 0.8 8.0
18.2 3.7 21.9
11-6 1.2 12.8
11
0
0
0
0
0
0
0
0
0
0
0
0
0
54.9
_-
M^rnf r\D
' — ~ •
8Ui Capacity
III IV III+IV
0.4 0
0.8 0.4
1.2 0.6
2.5 0.4
2.3 0.6
5.0 0.4
5.2 0.8
4.6 1.0
6.0 1.2
6.0 1.2
6.8 1.0
8.9 1.0
22.6 4.6
14.5 1.4
0.4
1.2
1.8
2.9
2.9
5.4
6.0
5.6
7.2
7.2
7.8
9.9
27.2
15.9
.„„„.
-------
is based on an increased price of refined copper to a level of 220^/kg.
The second profit margin is based on the original 18.7£/kg. but adds the
increase in price as extra profit while process costs are held constant. The
second profit margin is 51.7£/kg. Three considerations suggest the use of
the second profit margin. The first consideration is the desirability of
presenting sensitivity analysis in general. The second consideration is that
a profit margin of 51.7£/kg. based on a price of 220^/kg. is a margin of
23.5 percent, which though clearly high, has been achieved within recent
years by a member of the industry. Finally, because the margin is high, it
in effect can be viewed as an upper limit, and thus any smelter that has a
substantial profit impact in spite of such a favorable profit margin is in a
very vulnerable position at a lower, more likely, profit margin.
The same cases discussed earlier are still applicable, the results are
for Alternatives II, III, IV, and III+IV. At the first profit margin (18.7rf/k
the results show a maximum profit reduction of greater than 20 percent for
3 of the 14 smelters at the 100 percent capacity utilization rate for Alterna-
tive III+IV. At the 80 percent capacity utilization rate 4 smelters exceed
20 percent. For 2 of the above 4 smelters the profit reduction is greater
than 40 percent at the 80 percent capacity utilization rate, and greater than
50 percent for 1 of the 2 smelters. Profit reductions of greater than 40
percent would seriously call into question the continued viability of these 2
smelters. At the second, higher, profit margin (51.7l/kg.) the profit
impacts are lessened substantially. Two smelters experience profit reductions
of greater than 20 percent.
7.2.5.2 Method Two. Method two uses method one as a starting point
and then refines it by relaxing the assumption of method one of an inte-
grated producer and a uniform profit margin for all producers. Table 7-10
provides a means to identify those smelters that are most likely to face the
greatest impact.
Table 7-10 starts by showing the smelters ranked according to the cost
increase described earlier (Alternative III+IV). The size of the cost
increase and the rank provides one indication of the potential impact of
controls. A caveat that should be mentioned concerning this indicator is
that it does not take into consideration baseline costs. Examination of
baseline costs would be a useful supplement here. The 2 smelters with the
highest cost increases are likely to experience serious difficulty due to the
7-32
-------
Table 7-10. REVIEW OF SMELTERS
I
CO
CO
Smelter
ASARCO-Hayden
ASARCO-E1 Paso
Kennecott-Garfield
Phelps Dodge-Hidalgo
Phelps Dodge-Mo renci
Inspiration
Magma
Phelps Dodge-Ajo
Phelps Dodge-Douglas
Copper Range-White Pine
Kennecott-Hayden
Kennecott-Hurley
Cities Service-Copperhill
Kennecott-McGill
Cost
rank
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Major
Increase Capital
- (tf/kg) Commitment
0.2 1982
0.5
0.7 1978
1.3 1976
1.3 1982
2.3 1980
2.5
2.8
2.9
3.0
3.2
4.1 1982
11.3
6.6
Maj or
Integrated
Mine
-
-
Yes
-
Yes
-
Yes
-
No
-
Yes
Yes
-
No
Closely
Associated
Refinery
No
No
Yes
No
No
Yes
Yes
No
No
Yes
Yes
Yes
No
No
Viability
Estimates
By Others
(Reference #)
Viable-37
Viable-37
Viable-37
Viable-37
Viable-37
Doubtful -38, 37
Viable-37
Doubtful-38,37,39
Doubtful -38, 16
Closure-37,39
Viable-37
Viable-37
Doubtful -37
Viable-37
Doubtful-38,37,16
-------
costs. A second indicator that is presented to provide additional insight
into a firm's possible reaction to control costs is a review of any major
capital commitments to the smelter that a firm has made recently.
Most of the firms with the lower control cost increases have also
recently made major capital commitments to their smelters which in turn
suggests a stronger commitment than a firm that has postponed capital expen-
ditures for a smelter. The third indication is provided by a review of
whether or not the smelter has a major integrated mine that supplies much or
all of its concentrates. The fourth indication is provided by a review of
whether or not the smelter is closely associated with a refinery. Finally,
the estimates of others who have analyzed the smelters are presented. The
estimates are from four sources.
The above indications are useful but should be supplemented by two
additional pieces of information. This information is rioted below even
though its impact has not been thoroughly analyzed at this time:
1) Newmont Mining Corp. recently purchased Cities Service Co.'s
copper mine operations in Arizona (Pinto Valley). The concentrates from that
mine currently go to the Inspiration smelter under a contract that runs until
the end of 1984. These concentrates represent a significant share of Inspi-
ration's production. The dollar value is approximately 15 percent of Inspi-
ration's sales. After the contract expires these concentrates will then be
processed at Newmont1s Magma smelter. This will represent a significant loss
for Inspiration and a significant gain for Magma.
2) Anaconda recently announced that it will close its Butte mine and
no longer ship concentrates to Japan. At this time, it is not clear if the
Japanese will seek replacement concentrates, and if so, what domestic mines
are the most likely candidates to supply the concentrates.
7.2.6 Capital Availability
The principal determinant of the financial viability of a smelter is
profitability. However, the amount of capital needed to purchase control
equipment is one of the components that enters into an evaluation of profit-
ability. Most firms prefer to finance pollution control equipment with debt,
both because debt is less expensive than equity in general, and additionally
because debt incurred to purchase pollution control equipment is often tax
7-34
-------
exempt. Assuming control equipment is financed with debt, as the capital
cost of the control equipment increases, the level of debt increases. An
increased debt level means the fixed costs required to service the debt
increase and therefore the level of risk increases. As a result, a dis-
Table 7-11 slums the capital expenditures that will be necessary
The capital expenditures were expla,ned in detai, in an earlier e .'
The basel,ne capital expenditures are presented, as well as the capita,
expenditures for Alternatives II, III and IV Thr.» f
smelter an* • u. , e flras own more than 1
I! frm! ?" C9SeS the ^ CaPUS' C°StS « "»«. "'hough
t f,rms can make capita, budgeting decisions on an individual smelter
. Additional!,, 5 of the 7 c.panies are owned .,
ea;: ; ;::;; tdr:ee> by 'isnmcMti' '^ — ™L and
qu He Inkely to have access to the necessary capital. The remaining
companies are ASARCO and Phel ps Dodge. e remaining
7"U """ *"« "»reMt
dv
e
•"====''
of Phelps Dodge the increases are 0, 5, and 1 percent
terna I, In
SUM6St * ^ "
tosether Kf
««
for Phelps Dodge was lowered to Baa2 from i
its
7-35
-------
Table 7-11. CAPITAL COSTS OF ARSENIC CONTROLS
FOR PRIMARY COPPER SMELTERS
($103)
ASARCO
Cities Service
Copper Range
Inspiration
Kennecott
Newiriont
Phelps Dodge
El Paso
Hayden
Tacoma
Debt Increase3
Copper hill
White Pine
Miami
Garfield
Hayden
Hurley
Me Gill
Magma
Ajo
Douglas
Hidalgo
Morenci
Debt Increase3
Baseline
46
75,606
75,652
24%
0
0
0
0
0
54,044
0
54,044
0
51,067
0
0
95,294
146,361
24%
Al
II
0
0
3,469
3,46$
1%
0
. 0
0
0
0
0
10,530
10,530
0
0
0
0
0
0
0%
ternative
111
1,375
1,702
3,469
6,546
2%
4,434
4,434
9,825
5,197
6,731
8,760
8,760
29,448
13,050
6,731
9,825
6,731
8,526
31,813
5%
IV
370
0
3,469
3,839
1%
893
893
922
1,786
894
952
893
4,525
1,786
894
1,787
894
1,786
5,361
1%
aPercent increase in average long term debt level for the past 3
years (1981 to 1979) if controls are added as debt.
7-36
-------
previous rating in 1980 and 1981 of A. Although Baa2 is still a relatively
strong rating, the fact that it was lowered from 1981 to 1982 suggests that
substantial increases in the amount of debt held by the company may present
some difficulties.
7.3 SOCIO-ECONOMIC IMPACT ASSESSMENT
7.3.1 Executive Order 12291
The" purpose of Section 7.3.1 is to address those tests of macro-
economic impact as presented in Executive Order 12291, and, more generally
rNESlVnVther Sl9n1f1Cant maCr°™-c "*«* that may result from'
i ; X6CUt1Ve °rder ^ Stl>lateS " "^ -les" *•" ^at are
projected to have any of the following impacts:
• An annual effect on the economy of $100 million or more
• A major increase in costs or prices for consumers; indi-
vidual industries; Federal, State, or local government
agencies; or geographic regions.
• Significant adverse effects on competition, employment, invest-
ment, productivity, innovation, or on the ability of U.S.-based
enterprises to compete with foreign-based enterprises in domestic
or export markets.
each of- ,03^. The annual ized control costs for
each of the 3 alternatives Is we,, below the $100 million *,<:„ 1s the
figure used to identify a major rule. The annualized contro, costs for
merna ,ves „. !„, Iv. and Imiv .„ ^ ^
W.3 million, and $31.1 million, respectively.
7'3'1-2 Milonal Effects. EmP1ovm.nt ann r^r.>.-tir ^
I"" r r I"r^^^^^^^ «« in particu-
U r 7 smelters are ,ocated ,„ Arizona. As a result, econ^ic impacts
be concentrated in that geographical area. A copper smelte typi
s about 50 w
employs about 500 people.
The domestic copper producers compete among themselves, as well as
aga,nst foreign copper producers and substitutes. Any substantial increase
7-37
-------
in costs will put pressure on the competitive position of some domestic
smelters with respect to other domestic smelters, and also with respect to
foreign copper producers, and substitutes.
7.3.2 Regulatory Flexibility
The Regulatory Flexibility Act of 1980 (RFA) requires that differen-
tial impacts of Federal regulations upon small business be identified and
analyzed. The RFA stipulates that an analysis is required if a substantial
number of small businesses will experience significant impacts. Both measures
must be met, substantial numbers of small businesses and significant impacts,
to require an analysis. If either measure is not met then no analysis is
required. The EPA definition of a substantial number of small businesses in
an industry is 20 percent. The EPA definition of significant impact
involves three tests, as follows: one, prices for small entities rise 5
percent or more, assuming costs are not passed onto consumers; or two, annual-
ized investment costs for pollution control are greater than 20 percent of
total capital spending; or three, costs as a percent of sales for small
entities are 10 percent greater than costs as a percent of sales for large
entities.
The Small Business Administration (SBA) definition of a small business
for Standard Industrial Classification (SIC) Code 3331, Primary smelting and
refining of copper is 1,000 employees. Table 7-12 shows recent employment
levels for each of the 7 companies that own primary copper smelters.
All 7 have more than 1,000 employees. Therefore, none of the 7 companies
meets the SBA definition of a small business and thus no regulatory flexi-
bility analysis is required.
7-38
-------
Table 7-12. NUMBER OF EMPLOYEES AT COMPANIES
THAT OWN PRIMARY COPPER SMELTERS
_ ComPany _ _ _ Employees _ $ourcea
ASARCO, Inc. 12,700 1980 SEC 10-K p. A7
Cities Service Co. 18,900 1980 SEC 10-K p. 6
Copper Range Co.b 3>049 1980 SEC 10-K p. 22
198° SEC 10-K - 2
Kennecott Corp.c 3MOO
Newmont Mining Corp. 12,400 1980 SEC 10-K p. 9
Phelps Dodge Corp. _ 15.220 _ 1980 SEC 10-K p. 1
aSEC 10-K is Securities and Exchange Commission, Form 10-K.
r°' 1S a ^oll*-°wned subsidiary of the Louisiana Land and
Company. Figures are for Louisiana Land and Exploration.
cPrior to merger with Sohio on March 12, 1981.
7-39
-------
7.4 References
1. Review of New Source Performance Standards for Primary Copper Smelters
— Background Information Document, Preliminary Draft. U.S. Environ-
mental Protection Agency. Research Triangle Park, North Carolina.
Publication No. EPA-February 1983. p. 3-2.
2. ASARCO, Inc., Form 10-K. December 31, 1980. p. A2.
3. Cities Service Co., Annual Report 1980. p. 41.
4. The Louisiana Land Exploration Co., Form 10-K. December 31, 1980. p.
16.
5. Inspiration Consolidated Copper Company, Annual Report 1980. p. 2.
6. Kennecott Corp., Form 10-K. December 31, 1980. p. 4.
7. Newmont Mining Corp., Form 10-K. December 31, 1980. p. 3.
8. Phelps Dodge Corp., Form 10-K. December 31, 1980. p. 2, 4.
9. Butterman, W.C. U.S. Bureau of Mines. Preprint from the 1981 Bureau
of Mines Minerals Yearbook. Copper, p. 3.
10. Butterman, W.C. U.S. Bureau of Mines. Mineral Industry Surveys.
Copper Production in December 1982. p. 2.
11. Schroeder, H. J. and James A. Jolly. U.S. Bureau of Mines. Preprint
from Bulletin 671. Copper - A Chapter from Mineral Facts and Problems,
1980 Edition, p. 14-16.
12. Annual Data 1982. Copper Supply and Consumption. Copper Development
Association Inc. New York, New York. p. 6, 14.
13. Reference 11, p. 5.
14. Reference 9, p. 1.
15. Reference 9, p. 5.
16. Arthur D. Little, Inc. Economic Impact of Environmental Regulations
on the United States Copper Industry. U.S. EPA. January 1978. p.
V-8.
17. Reference 9, p. 24-29.
7-40
-------
18. Reference 12, p. 14.
19. Reference 11, p. 14.
20. Reference 12, p. 18.
21. Copper's Hope: Electric Vehicles. Copper Studies. Commodities
Research Unit, Ltd. New York. March 30, 1979, p. 5.
22. Copper in Military Uses. Copper Studies. Commodities Research Unit,
Ltd. New York, February 15, 1980. p. 1.
23. Butterman, W.C. U.S. Bureau of Mines. Mineral Industry Surveys.
Copper in 1982 - Annual, Preliminary, p. 2.
24. Butterman, W.C. U.S. Bureau of Mines. Preprint from the 1980 Bureau
of Mines Minerals Yearbook. Copper, p. 1.
25. Schroeder, H. J., and G. J. Coakley. U.S. Bureau of Mines Preprint
from the 1975 Minerals Yearbook. Copper, p. 2.
26. The Capital Cost Picture. Copper Studies. Commodities Research Unit,
Ltd. New York. August 18, 1975. p. 1.
27. Rosenkranz, R.D., R.L. Davidoff, and J.F. Lemons, Jr., Copper Avail-
ability-Domestic: A Minerals Availability System Appraisal. U.S.
Bureau of Mines. 1979. p. 13.
28. Sousa, Louis J. U.S. Bureau of Mines. The U.S. Copper Industry:
Problems, Issues, and Outlook. Washington, D.C. October, 1981.
p. 67.
29. Copper Imports on Preferential Tariff. Japan Metal Journal (Tokyo).
December 8, 1980. p. 3.
30. Affidavit of Robert J. Lesemann, Commodities Research Unit/CRI and
former editor-in-chief of Metals Week, to the Federal Trade Commission.
September 27, 1979. FTC Docket Number 9089.
31. Brass Rod Production Cartel Starts. Japan Metal Journal (Tokyo).
July 6, 1981. p. 1.
32. Smelter Pollution Abatement: How the Japanese Do It. Engineering and
Mining Journal. May 1981. p. 72.
33. Rieber, Michael. Smelter Emission Controls: The Impact on Mining
and The Market For Acid. University of Arizona, Tucson, Arizona.
March, 1982. p. 5-10.
34. Custom Copper Concentrates. Engineering and Mining Journal. May
1982. p. 73.
7-41
-------
35. Everest Consulting Associates, Inc., and CRU Consultants, Inc. The
International Competitiveness of the U.S. Nonferrous Smelting Industry
and the Clean Air Act. Princeton, NJ. April 1982. p. 9-9.
36. Reference 32.
37. Reference 33, p. 1-11.
38. Everest Consulting Associates, Inc. The International Competitiveness
of the U.S. Non-Ferrous Smelting Industry and the Clean Air Act.
Princeton, N.J. April 1982. p. 3-17.
39. Phelps Dodge Corp. 1981 Annual Report, p. 8.
40. Moody1s Industrial Manual 1982 Vol. I, p. 58, Vol. II, p. 4236.
7-42
-------
APPENDIX A
EVOLUTION OF THE BACKGROUND INFORMATION DOCUMENT
A-l
-------
EVOLUTION OF THE BACKGROUND INFORMATION DOCUMENT
Date
July 13-14, 1976
December 10-16, 1976
April 18-26, 1977
June 20-30, 1977
January 17-27, 1978
May 1-5, 1978
May 10-12, 1978
June 12-16, 1978
July 11-12, 1978
July 24-27, 1978
September 12-25, 1978
October 30 -
November 15, 1978
Activity
Emission source testing at Phelps Dodge
Copper Smelter, Ajo, Arizona.
Emission source testing at Kennecott Copper
Smelter, Hayden, Arizona.
Emission source testing at Anaconda Copper Smelter,
Anaconda, Montana.
Emission source testing at ASARCO Copper Smelter,
El Paso, Texas.
Emission source testing at ASARCO Copper Smelter,
El Paso, Texas.
Emission source testing at Phelps Dodge Copper
Smelter, Douglas, Arizona.
Emission source testing at Phelps Dodge Copper
Smelter, Ajo, Arizona.
Emission source testing at Phelps Dodge Copper
Smelter, Ajo, Arizona.
NAPCTAC Meeting in Raleigh, North Carolina to
discuss issues related to development of arsenic
emission standards for primary copper smelters.
Emission source testing at Phelps Dodge Copper
Smelter, Hidalgo, Arizona.
Emission source testing at ASARCO Copper Smelter,
Tacoma, Washington.
Emission source testing at Kennecott Copper Smelter
Garfield, Utah.
A-2
-------
Date
May 8-15, 1979
July 23-24, 1979
September 10-16, 1979
September 18-22, 1979
December 7-13, 1979
March 11-13, 1980
April 14-23, 1980
June 5, 1980
June 24-26, 1980
March 17, 1981
January 12, 1983
January 14-22, 1983
April 27, 19U3
Activity
Emission source testing at ASARCO Copper Smelter,
Tacoma, Washington.
Emission source testing at Phelps Dodge Copper
Smelter, Ajo, Arizona.
Emission source testing at Phelps Dodge Copper
Smelter, Morenci, Arizona.
Emission source testing at Phelps Dodge Copper
Smelter, Douglas, Arizona.
Emission source testing at Kennecott Copper Smelter
McGill, Nevada.
Plant visit to Hibi Kyodo Copper Smelter,
Tamano, Japan, to collect information on the
fugitive emissions control system.
Emission source testing at Magma Copper Smelter,
San Manuel, Arizona.
EPA listing of inorganic arsenic as a hazardous
pollutant under Section 112 of the Clean Air Act.
Plant visit and emission source testing at ASARCO
Copper Smelter, Tacoma, Washington.
NAPCTAC meeting in Raleigh, North Carolina to
discuss regulatory alternatives for limiting
arsenic emissions from low-arsenic throughput
primary copper smelters.
Judicial Opinion and Order, filed with United
States District Court, Southern District of New
York, pertinent to action brought by State of
New York against EPA Administrator (New York
v. Gorsuch, F. Supp. (S.D.N.Y. 1983)).
Order for EPA to propose emission standards for
inorganic arsenic within 180 days of Order.
Emission source testing at ASARCO Copper Smelter
Tacoma, Washington.
NAPCTAC meeting in Raleigh, North Carolina, to
discuss general approach and EPA staff recom-
mendations for setting standards for inorganic
arsenic emissions from primary copper smelters.
A-3
-------
APPENDIX B
INDEX TO ENVIRONMENTAL IMPACT CONSIDERATIONS
B-l
-------
INDEX TO ENVIRONMENTAL IMPACT CONSIDERATIONS
This appendix consists of a reference system which is cross indexed
with the October 21, 1974, Federal Register (39 FR 37419) containing the
Agency guidelines for the preparation of Environmental Impact
Statements. This index can be used to identify sections of the document
which contain data and information germane to any portion of the Federal
Register guidelines.
B-2
-------
INDEX TO ENVIRONMENTAL IMPACT CONSIDERATIONS
Background and Description
Summary of the Regulatory
Alternatives
Statutory Authority
Industry Affected
Sources Affected
Availability of Control
Technology
2. Regulatory Alternatives
Regulatory Alternative I
No Action (Baseline)
Environmental Impacts
Costs
Regulatory Alternative II —
Environmental Impacts
Costs
Location Within the Background
Information Document (BID)
The regulatory alternatives are
summarized in Section 1.2.
Statutory authority is cited in
Section 1.1.
A description of the industry
to be affected is given in
Section 7.1.
Descriptions of the various
sources to be affected are
given in Section 2.0.
Information on the availability
of control technology is given
in Section 3.0.
Environmental effects of Regulatory
Alternative I are considered in
Section 5.0.
Costs associated with Regulatory
Alternative I are considered in
Section 6.0.
Environmental effects associated
with Regulatory Alternative II
emission control systems are
considered in Section 5.0.
The cost impact of Regulatory
Alternative II emission control
systems is considered in
Section 6.0.
B-3
-------
INDEX TO ENVIRONMENTAL KiPACT CONSIDERATIONS (Concluded)
Agency Guidelines for Preparing
Regulatory Action Environmental
Impact Statements (39 FR 37419)
Location Within the Background
Information Document (BID)
Regulatory Alternative III
Environmental Impacts
Costs
Regulatory Alternative IV
Environmental Impacts
Costs
Regulatory Alternative V
Environmental Impacts
Costs
The environmental effects associated
with Regulatory Alternative III
emission control systems are
considered in Section 5.0.
The cost impact of Regulatory
Alternative III emission control
systems is considered in Section 6.0.
The environmental effects associated
with Regulatory Alternative IV emission
control systems are considered in Sectio
5.0.
The cost impact of Regulatory
Alternative IV emission control
systems is considered in Section 6.0.
The implementation of this
alternative would require the
elimination of inorganic arsenic
emissions at low-arsenic throughput
primary copper smelters. Inorganic
arsenic emissions would be zero.
This alternative could not be
implemented without closing all
plants in the source category. This
was not considered reasonable and hence
costs were not evaluated.
B-4
-------
APPENDIX C
SUMMARY OF TEST DATA
C-l
-------
SUMMARY OF TEST DATA
An emission test program was undertaken by EPA to evaluate the
performance of alternative control techniques available for the control
of process and fugitive arsenic emissions from process facilities
including roasters, smelting furnaces, and converters at primary
copper smelters. This appendix presents a brief description of the
process facilities and control equipment tested, and a summary of the
results obtained.
Arsenic emission measurements were conducted at eight domestic
smelters. Particulate emission measurements were also conducted at
some of these smelters. A listing of the process facilities and air
pollution control equipment tested and the emission measurements
conducted is presented in Table C-l. All arsenic measurements were
performed using EPA Method 108, the recommended EPA method for the
determination of arsenic from stationary sources. Measurements of
total particulate were performed in accordance with EPA Method 5.
In addition to the arsenic and particulate emission measurements,
visible emission observations were recorded at one of the above eight
domestic smelters, ASARCO-Tacoma, and one other smelter located in
Japan, the Tamano smelter.
A brief description of each smelter, as well as the process and
control device tested, is presented in Sections C.I through C.9.
Section C.10 contains the data tables from the emissions measurement
and opacity observation tests.
C.I ASARCO-TACOMA
The ASARCO smelter at Tacoma, Washington, is a custom smelter
which processes copper ore concentrates, precipitates, and smelter
C-2
-------
by-products from numerous domestic and foreign sources. The smelter
produces about 320 Mg (352 tons) of anode copper per day at full smelt
and houses the only arsenic production facility in the United States.
Copper smelter facilities include 10 roasters (6 Herreschoff multi-hearth
roasters and 4 CAW Roasters), 2 reverberatory smelting furnaces,
4 Pierce-Smith converters, 3 anode furnaces (1 hearth type furnace and
3 tilting type furnaces),* and an electrolytic refinery.** Arsenic
production facilities consist of three Godfrey roasters, arsenic
trioxide settling chambers or kitchens, storage facilities, and a
metallic arsenic plant.
The roaster charge which consists of a blend of concentrates,
precipitates, lead speiss, flue dust, and fluxing materials typically
contains 3 to 4 percent arsenic and 7 to 10 percent moisture. At full
smelt, four to five roasters are used. Charging is continuous. The
calcine produced, about 45.4 Mg (50 tons) per hour, is intermittently
discharged from hoppers located below the roasters into Tarry cars for
transport to one of two reverberatory furnaces. Typically, two 5.9 Mg
(6.5 ton) cars are charged every 15 minutes. Fugitive emissions which
could escape are confined and captured by close-fitting exhaust hoods
located at the discharge point and are vented into the main roaster
flue by two 73.6 Nm3/min (2,600 scfm) fans. In addition, general
ventilation is applied at the south end of the larry car tunnel to
control fugitive emissions-resulting from the re-entrainment of settled
dust within the tunnel.
Offgases from the roasters, which average about 3,570 m3/min
(126,000 acfra) at 260°C (500°F), are combined with the exhaust gases
from the ancillary fugitive emission control systems [850 Nm3/min
(30,000 scfm)] and reduced in temperature to less than 120°C (250°F).
The total gases are then treated in a baghouse for particulate removal.
The baghouse consists of 17 compartments containing 120 bags each.
The bags are made of acrylic and measure 20.3 cm (8 inches) in diameter
and 7.6 meters (25 feet) in length. The total baghouse filtering area
-Operation of electrolytic refinery was discontinued in January 1979,
C-3
-------
is 9,950 in (107,100 ft2). The baghouse is designed to effectively
treat 5,664 m /min (200,000 acfm) of gas at an air-to-cloth ratio of
about 1.9 to 1.0. Bag cleaning is performed by mechanical shakers.
The clean baghouse exhaust is vented through a flue to the smelter
main stack.
Although the smelter has two reverberatory smelting furnaces,
each with an approximate smelting capacity of 1,090 Mg (1,200 tons)
per day, the designated Number 2 furnace is used almost exclusively.
The furnace measures 33.5 m (110 feet) in length and 9.8 m (32 feet)
in width and is fired by either oil or natural gas. Furnace charging
is accomplished by discharging the larry cars through one of four
Wagstaff guns located along the furnace sidewalls. Typically, it
takes less than 1 minute to discharge each car. At full snelt, two
cars are discharged about every 15 minutes. To minimize potential
fugitive emissions during charging, a manual control override is used
which simultaneously opens the furnace Hue control damper and reduces
the fuel supply to the furnace fire prior to each charge to prevent
pressure surges in the furnace.
Matte is tapped from the furnace as required by converter operations
through one of four tapping ports. Only one tapping port is used at a
time. The matte flows through a cast copper launder to a 4.25 m3
(150 ft ) cast steel ladle for transfer to the converters. At full
smelt, about 45 ladles are transferred per day. It takes about 7 minutes
to fill one ladle. Similarly, slag is skimmed through one of two
tapping ports as required to maintain the proper slag level in the
furnace and transferred to a 5-pot slag train for transit to the slag
dump. Once tapped, the slag flows through a cast steel launder and
into a 2.8 m (100 ft3) cast steel slag pot. At full smelt, about
20 five-pot slag trains are dumped per day. Each train takes about
15 minutes to fill. Fugitive emissions generated during both matte
and slag tapping operations are controlled by local ventilation techniques
Tapping ports, launders, and launder-to-ladle/pot transfer points are
hooded and ventilated. Ventilation requirements for the matte tapping
system total about 700 m3/min (25,000 acfm), while the ventilation
requirements for the slag skimming system total about 600 m3/min
C-4
-------
(21,000 acfm). Captured emissions from both systems are currently
controlled by an electrostatic prccipitator prior to venting from a
stack to the atmosphere.
Process gases from the reverberatory furnace, which average about
1,415 Nm /min (50,000 scfm), pass through a pair of waste heat boilers
where the gases are cooled to about 400°C (750°F). The exiting gases
then pass through a large rectangular brick flue where additional
cooling and gas stream conditioning is provided by air infiltration
and water and sulfuric acid sprays located in the flue. The resultant
gas stream, about 6,100 actual m3/m1n (215j000 acfm) flt ^ (
Centers the first of two electrostatic precipitators in series for
particulate removal. The first precipitator is a tube or pipe design
consisting of 18 sections with a total collection area of about 6,619 m2
'1.250 ft ). Each section contains 84 pipes measuring 30.5 cm (12 inches)
- dieter and 4.6 m (15 feet) in length. The second unit is a plate
type design, it consists of seven parallel chambers each with four
fields in series and has a total collection area of 7,710 m2 (82,992 ft2)
e exumg gases, about 7,740 actual m3/nrln (270>000 acfm) ^ ^ }'
(230 F), are discharged through a large flue to the smelter main
stack.
Matte from the reverberatory furnace is transferred to one of
our Pierce-Smith converters. Three of the converters measure 4.0 „
(13 ,eet) in diameter by 9.1 „ (30 feet) in length, while.the fourth
converter is 3.4 m (n feet) in diameter and 7.9 „ (26 feet) long. In
addition to copper matte, smelter reverts and cold dope materials are
also processed. Typically, only two converters are on blow aTany on,
tine. A converter cycle normally takes frm 10 to 12 hours With
u"thOff9as now per blow1"
,.-,-„„ „<.,„,, ai)1J uuntdins rrom J to 4 percent SO en-,•<• +
ww -r j^ci v»cnt oUo • D I^TPr rnnnav
Produced is transferred to one of three anode furnaces for re" £
and casting. The slag s^med from the converters „ recycled to he
reverberatory furnace.
Offgases from converter blowing operations are captured by
cooled hoods and pass through a series of multiclone
ig flue for coarse particulate removal prior to entering" th'e gas
deamng circuits of either a liquid S02 plant or single-contact
C-5
-------
sulfuric acid plant. The gas cleaning circuits for both plants are
similar and consist of a water spray chamber, electrostatic precipitator
scrubbers, and mist precipitators in series. The single-contact acid
plant has a 182 Mg/day (200 TPD) capacity at 5 percent S0? and is
capable of processing 652 Nm3/min (23,000 scfm) of converter gas. The
liquid S02 plant processes up to 12,748 Nm3/min (45,000 scfm) of the
converter gases. The plant uses dimethylaniline (DMA) to absorb the
S02 in the gas stream and uses steam stripping from regeneration. The
100 percent concentrated SO,, gas stream produced is then liquified by
compression and the liquid S02 stored.
Inlet and outlet emission measurements for arsenic were conducted
by EPA across the roaster baghouse and the arsenic baghouse (metallic
and kitchen) on September 12 through September 25, 1978. Arsenic test
results are summarized in Tables C-3 through C-7.
Testing for arsenic was also conducted on September 15, 16,
and 18, 1978, at the outlet of the reverberatory furnace ESP. These
results are shown in Table C-8.
Tests were also conducted for roaster calcine discharge, matte
tapping, slag tapping, and converter slag return. Data obtained for
these sources are presented in Tables C-9 through C-12.
Visible emission observations were made by using EPA Method 22
and EPA Method 9 for calcine loading of larry cars, matte tapping,
slag tapping, and converter slag return on June 24-26, 1980.
Tables C-65 through C-73 present the results of these visible emissions
observations.
C.2 ASARCO-EL PASO
The ASARCO smelter at El Paso is a custom smelter that processes
copper ore concentrates from numerous sources. The smelter produces
about 315 Mg (350 tons) of anode copper per day. Copper smelting
facilities consist of ore handling and bedding facilities, four multi-hearth
roasters, one reverberatory smelting furnace, and three Pierce-Smith
converters. In addition, this smelter also has separate process
facilities for zinc and lead production.
The smelter feed, consisting of a blend of concentrates, precipitates,
lime and flue dust, and typically containing more than 0.2 percent
C-6
-------
arsenic, is charged to four Herreshoff multi-hearth roasters. Each
roaster has seven hearths and is capable of producing approximately
356 Mg (392 tons) of calcine per day. The calcine is taken by Tarry
car to the single reverberatory furnace which is 90 cm (35.7 in.) long
and 20 cm (8.0 in.) wide and fired with either oil or natural gas
The furnace is charged by Wagstaff guns located along the furnace side
walls. The matte is tapped from one of five tap holes located on the
north and south sides of the furnace. Slag is tapped from the west
side of the furnace and is disposed of in a slag dump.
Matte from the furnace is transported to one of three Pierce-Smith
converters. Each of the converters is 4 m in diameter and 9 m long
Normally, two converters are in the blowing mode at one time In '
addition to copper matte, flue dust and cold dope materials, converters
also process lead matte from the adjacent lead smelter when available
Blowing time generally ranges from 10 to 12 hours. Blister copper
produced is further refined in two anode furnaces and then cast into
anodes for shipment.
Offgases from the reverberatory furnace, which average about
1,700 Nm /min (60,000 scfm), pass through a pair of waste heat boilers
where the gases are cooled to about 400°C (750°F) and about 23 Mg
(50,000 Ib) of steam is produced. The gases exit the waste heat
boilers through two parallel ducts and are then combined with the
roaster offgases in a'main flue. The combined gas stream, consisting
of about 5,000 Nm /min (177,000 scfm) at 200°C (400°F), then passes
through a spray chamber where it is cooled to about HO'C (230°F)
prior to entering an electrostatic precipitator for particulate removal
The precipitator consists of seven parallel chambers. Each chamber
has four fields in series and has a total field volume of 535 m3
(18,900 ft ). Gases exiting the precipitator are discharged to the
main stack through a large balloon flue.
Offgases from the converter blowing operations average 6,000 Nm3/nin
56,500 scfm). They then pass through a settling chamber, two waste
heat boilers, and a spray chamber where the gases are cooled from
315°C (600'F) to llO'C (230
-------
The precipitator consists of four parallel chambers, each of
which has four fields in series. The exiting offgases then pass
through a venturi scrubber for additional particulate removal, are
humidified and cooled in a pair of packed-bed scrubbers, and are
treated in a series of mist precipitators where acid mist and any
remaining particulates are removed prior to entering a double-contact
acid plant for S02 removal. The acid plant has a normal production
rate of 408 Mg (450 tons) of acid per day. Either 93 or 93 percent
sulfuric acid is produced. The acid plant tail gas streams are discharged
through a 30.5 m (100 ft) stack.
Emission measurements were conducted by EPA during June 26-30,
1977. Inlet and outlet arsenic and mass measurements were made across
the roaster/reverberatory electrostatic precipitator. Three inlet
locations and one outlet location were sampled. The inlet locations
included a large downtake duct off of the multi-hearth roasters, and
two parallel ducts downstream of the reverberatory waste heat boilers.
The outlet location consisted of the balloon flue downstream of the
precipitator. Three arsenic and two total particulate runs were
conducted. The arsenic and particulate test results are summarized in
Tables C-13 through C-22, and Table C-29.
Arsenic emission tests were also conducted across the
double-contact sulfuric acid plant. Three inlet and outlet measurements
were made. The sampling locations included a duct upstream of the
spray chamber/ESP and the acid plant tail gas stack. The results of
these tests are summarized in Tables C-23 and C-24, Process conditions
were carefully observed, and testing was conducted only when the
subject process facilities and control equipment were operating within
normal operating limits.
Fugitive emissions from the reverberatory furnace matte tapping
operation at ASARCO-E1 Paso are captured by hoods over the ladle,
covered matte launders, and a hood at the matte tapping holes. Gases
from these sources are combined in a common duct and directed through
a fan to a baghouse. The baghouse discharges into the roaster/
reverberatory spray chamber/ESP control system whicn discharges from
the 250 m (828 ft) main stack.
C-8
-------
Fugitive gases that escape the converters during the blow period,
and roll-in/out operations and -cher fugitive gases in the converter
building, are collected at the roof of the converter building. Collected
fugitive gases are drawn through four openings at the roof into ducts
that combine into a main'duct leading to a baghouse, through fans on
the clean side of the baghouse, and then out the 250 m (828 ft) main
stack.
The fugitive gas flow through the baghouse averages 14,100 Nm3/min
(498,000 scfm). The converter building fugitive baghouse consists of
12 compartments. Normally all compartments are in operation except
one compartment which is taken off during the cleaning cycle and
another compartment which is taken off for maintenance purposes for a
fraction of the total time. Each of the 12 compartments contains
334 Orion or Dacron bags. Each compartment is 20 cm (8 in.) in diameter,
6.7 m (22 ft) long, with a cloth area of 1,644 m2 (17,700 ft2) per
compartment. The total net cloth area of the baghouse is about 19,700 m2
(212,400 ft ). The baghouse was designed to effectively treat 15,282
actual m /rain (540,000 acfm) at 54°C (130T) using an air-to-cloth
ratio of 3.0 to 1.0. Mechanical shakers (automatic) are used for
cleaning. Dust from the baghouse is removed from the dust chambers
under the baghouse by screw conveyors.
Emission measurements across the baghouse were conducted by EPA
during Janaury'17-27, 1978. Three arsenic and particulate measurements
were made at the inlet and outlet locations of the converter building's
fugitive baghouse. The arsenic test results are summarized in Tables C-25
and C-26, and the particulate test results are summarized in Tables C-27
and C-28. During the same period, three arsenic and particulate
measurements were made at the inlet to the matte tapping baghouse and
the calcine discharge duct. Due to the physical configuration of the
matte tapping baghouse system, outlet tests could not be conducted
Measurements were conducted after the fan; however, a side stream of
about 311 actual m /min (11,000 acfm) of the matte tapping gases were
split at the fan and ducted to the reverberatory furnace waste heat
boilers for cooling purposes. This gas stream was measured only for
volume flow. It was assumed that the pollutants in this gas stream
C-9
-------
would have the same concentration as the gas stream measured going to
the baghouse. Process conditions and control device parameters (when
applicable) were carefully observed during all the periods when testing
was conducted. Tests were conducted only when the process facilities
and control equipment were operating within normal limits. Uncontrolled
arsenic test results are summarized in Tables C-30 and C-32, and
particulate tests are given in Tables C-31 and C-33.
C.3 ANACONDA
This smelter, when operating, was producing about 545 Mg (600 tons)
of anode copper daily. Major process facilities consisted of a fluid-bed
roaster, electric smelting furnace, six converters (three operational),
and an anode furnace.
In this process, concentrates and precipitates are blended with
silica flux in approximate proportions of 88, 2, and 10 percent,
respectively. The blended materials (containing about 0.96 percent
arsenic) are then fed to a Dorr-Oliver designed fluid-bed roaster by a
screw feeder which controls the feed rate [typically about 91 Mg
(100 tons) per hour] and maintains a seal on the roaster. Fluidizing
air averages 1,062 Nm /min (37,500 scfm). The air is supplied through
tuyeres at the bottom of the roaster to keep the bed constantly fluidized
at 1.3 m (70 in.) in depth.
The fluidized air reacts with the sulfur contained in the sulfide
ores to form S02 and calcine. Approximately 45 percent of the sulfur
contained in the feed material is eliminated. Because the reaction is
exothermic, no auxiliary fuel is needed except at cold startup. The
bed temperature is generally maintained at 582°C (1,080°F). Most of
the calcine which is produced (35 percent) exits the reactor as a fine
3
dust suspended in the offgas stream. The offgases average 1,671 Nm /min
(59,000 scfm) at 543°C (1,010°F). These gases are ducted through a
series of primary and secondary cyclones where 90 to 95 percent of the
suspended calcine is recovered. The underflow from the roaster accounts
for the remaining 15 percent of the calcine produced.
An electric furnace is used for smelting. Drag conveyors continuously
distribute calcine produced in the roaster in combination with some
recycled flue dusts, along each side of the furnace through a series
of charge pipes in the roof. The furnace working area is 18 m wide,
C-10
-------
36 m long, and 30 m high. The bath area is 272 m2. The furnace is
equipped with six carbon electrons. Each is 1.65 m in diameter. The
electrodes are energized by three transfomers. Each has a capacity
of 12 MVA. Normal voltage applied is between 150 and 180 volts The
furnace production capacity is 998 Mg (1,100 tons) of matte per day at
52 percent copper. Matte and slag are tapped as required from four
matte and two slag tap holes to maintain normal depths (matte, 71 to
. 96 cm; slag, 102 to 152 cm). Offgases from the furnace average
approximately 425 NmJ/min (15,000 scfm) at 545°C (1,200°F)..
Matte from the electric furnace is transported'to one"of three
Pierce-Smith converters. Typical converter process feed rates include
41.5 Mg (45.7 tons) of matte, 4.5 Mg (5.0 tons) of flux, and 1.4 Mg
1.5 tons) of cold dope per hour. The offgas flow is about 2,832 Nm3/min
(100,000 scfm) per blowing converter because of excessive air infiltration
and typically contains about 2.5 percent SO Blister copper which is
produced is transported to an anode furnace for further refining and
subsequently poured into anodes on a casting wheel.
Process offgases from the electric furnace and converter blowing
operations are combined. Approximately 2,070 Nm3/min (73,000 scfm) of
the resultant gas stream is treated in a cooling chamber (humidified
and cooled) and a venturi scrubber for particulate removal prior to
entering a 600 Mg (design) double-contact acid plant. The remaining
gases are combined with the fluid-bed roaster offgases and transported
through a large balloon flue to a spray chamber/baghouse filtration
plant for particulate control.
The combined gas stream, consisting of roaster, electric furnace
and converter process gases totaling about 5,664 Nm3/min (200,000 scfm)
and ranging from 230 to 340°C (450 to 650«F) in temperature, exits the
balloon flue and enters two parallel spray chambers where the gases
are cooled to less than 110'C (230'F) with watersprays. Each spray
chamber is 7.3 m wide, 4.3 m high, and 30 „ ,ong and is equipped with
10 son,c spray nozzles. Water requirements range from 265 to 303 liters/
™ (7 to 80 gpm). The cooled gas stream then passes through two of
three fans (one is standby), each with a capacity of 6,230 actual
Nm Ann (220,000 scfm), prior to entering the baghouse.
C-ll
-------
The baghouse consists of 18 compartments aligned in two rows of
nine. Each compartment is 4.3 by 14.3 m in cross section and 11.3 m
high above the thimble flow. The 18 chambers are constructed of
reinforced concrete above the thimble flow and are completely insulated.
Each of the 18 compartments contains 240 Orion bags which are 3.5 m in
diameter and 7.7 m in length. The cloth area per compartment is
? 2
1,656 m . The baghouse net cloth area totals about 29,802 m . The
baghouse is designed to effectively treat 11,328 actual Mm /min
(400,000 acfm) at an air-to-cloth ratio of 1.25 to 1.00. Mechanical
shakers are used for cleaning. The clean baghouse exhaust is transported
via a high velocity insulated fiberglass flue to the base of the main
stack and subsequently discharged to the atmosphere.
Inlet and outlet emission measurements for both arsenic and
particulates were conducted by EPA across the spray chamber/baghouse
on April 20-26, 1977. Sampling was conducted upstream of the spray
chamber and downstream of the baghouse. Two sampling locations were
required to obtain the inlet values. During all tests, process conditions
were closely monitored, and testing was conducted only when the process
facilities were operating within normal operating limits.
The arsenic and particulate test results for the spray chamber/
baghouse are summarized in Tables C-34 through C-41.
C.4 PKELPS DODGE-AJO
This is a "green" feed smelter with a production capacity of
about 168 Mg (185 tons) of anode copper per day. Major process facilities
consist of a single reverberatory furnace, three Pierce-Smith converters,
and oxidizing furnace, and an anode furnace. Emission control apparatus
includes electrostatic precipitators for the control of particulate
emissions from smelting and converting operations and a single-contact
acid plant.
The reverberatory furnace, which is designed for wet smelting, is
9 m (30 ft) wide and 30 m (100 ft) long. It is fired with natural gas
or fuel oil, depending on the availability of gas. The furnace walls
and roof are constructed of silica brick, and the roof is of a sprung-arch
design. The furnace charge components consist of concentrates (90 percent)
C-12
-------
precipitates, limestone, and recycled flue dusts. In addition, converter
slag is returned and processed. A charge, usually 1.3 to 3.6 Mg (2 to
4 tons), enters the furnace through one of six charge ports, three on
each side of the furnace. Each port is equipped with a high-speed
belt slinger to charge wet concentrates at considerable velocity.
Slag and matte are tapped as required to maintain a normal bath depth
of approximately 120 cm (46 in.) in the furnace.
Offgases from the reverberatory furnace pass through two parallel
waste heat boilers where the gases are cooled to about 315'C (600*F)
and a significant quantity of steam is produced for power generation'
The exmng gas stress then enter a common plenum chamber for mixing
prior to treatment in a hot electrostatic precipitator which is designed
to handle 4,250 actual mj/min (150,000 acfm) at 315°C (600°F) The
precipitator is a Joy-Western design, which was installed in 1973 It
consists of two parallel units with two stages each, and it has a
total collection area of 3,860 m2. Gas treatment averages 6 seconds
at a gas velocity of 0.9 m (3 ft)/sec. The pressure drop across the
precipitator is 1.3 cm (0.5 in.) of water maximum. The unit has a
design efficiency of 96.8 percent measured at its operating temperature
Offgases from the converters pass through waste heat boilers
where gases are cooled to about 315°C (600°F), and steam is generated
from the removed heat. Gases enter a balloon flue and then pass
througn an electrostatic precipitator (ESP). The ESP has two independent
horizontal parallel units with three fields each, which are designed
to handle 5,940 actual m>in (210,000 acfm) at 340°C (650°F) and
95.100 pascals (13.8 psia). Total ESP Electing surface area is
2,770 m (29,808 ft2). After the converter gases leave the ESP, they
pass onto the scrubbing section of the acid plant where they are
treated in a humidifying tower, a cooling tower, and a mist precipitator
( no' T ?ff9aSeS 3re ^ PrOC6SSed 1n 3 Sln9le abs°^°» 544 Mg/day
(600 ton/day) acid plant for S02 removal. Either 93 or 98 percent
sulfunc acid can be produced. The acid plant tail gas is ducted to
the main smelter stack.
Simultaneous inlet and outlet arsenic ercission measurements were
conducted by EPA on Ouly 13-14, 1976, on the reverberatory furnace
C-13
-------
ESP. Operating values for primary and secondary voltage and current
and spark rate were monitored during all tests to ensure normal operation,
In addition, reverberatory furnace operations were monitored for
normal operation. The test results are summarized in Tables C-42 and
C-43.
Inlet Arsenic emission measurements at the converter ESP inlet
were conducted by EPA on June 13-15, 1978. Outlet: arsenic emission
measurements at the inlet of the acid plant, and acid plant outlet
tests, were conducted at the same time. Test results are given in
Tables C-44, through C47.
Fugitive emissions at Phelps Dodge-Ajo escape the primary hooding
on the converters during the blowing cycle. These emissions are
captured by fixed, semicircular-shaped secondary hoods attached to
each converter. The hoods are approximately 1 m (3 ft) high at the
tallest point. Although the secondary hoods are in operation during
the blowing cycle, they are dampered when the converters are in the
3
roll-out mode. The secondary system is designed to handle 1,980 Mm /min
(70,000 scfm) of gases. The gases are ducted uncontrolled to the main
stack.
Fugitive emissions from the two matte tapping locations are
controlled to a high degree by a rectangular duct about 60 to 90 cm
(2 to 3 ft) above each tap hole. The rectangular vent opening is
about 60 on (24 in.) wide by 30 cm (12 in.) high and controls the
fumes from the matte tapping hole and about one- to two-thirds down
the matte launder which is not covered. The gas volume for this
3
system is about 850 Mm /min (30,000 scfm) and provides considerable
draft to the vent for several feet. The matte runs into a 2.4 to 3m
(3 to 10 ft) launder and drops into a ladle on the floor below. Fumes
from the matte launder, are drawn into the system handling the ladle
emissions. Most of the time, few emissions escape from the matte
launder.
Fugitive emissions from the matte pouring into and from the ladle
are captured in a cubicle where the ladle is located. The ladle is
placed on a specially designed cradle on rail tracks. An electric
motor and pulley arrangement moves the ladle car in and out of the
C-14
-------
cubicle. During the matte tap, the ladle is placed in the cubicle,
and the doors are closed. The c-jolcle has ventilation ducts leading
from it through a fan, and the gases are discharged through the main
stack. The ventilation rate from the two cubicles is 1,700 Nm3/min
(60,000 scfm).
Fugitive emissions from the slag tapping location are captured by
using a hood (similar to matte tapping) at the slag tapping hole. The
slag pours into an open launder and drops through a hole in the floor
into a slag ladle positioned below. A hood near the ladle captures
fume from the ladle. Normally, few emissions emanate from the slag
tapping operation. The ventilation rate for the slag tapping system
is about 850 NmJ/mTn (30,000 scfm). The duct from the slag tapping
system is combined with the matte tapping duct system, and the gases
are discharged out of the main stack.
Uncontrolled emission measurements for arsenic, particulate
matter, and S02 were conducted by EPA for the secondary converter hood
and matte tapping systems on May 10-12, 1978. During all tests,
process conditions were closely monitored, and testing was conducted
only when the process facilities were operating within nomal limits.
The arsenic test results are summarized for both systems in Tables C-48,
and C-50. The particulate/ S02 test results are suroiarized for both
systems in Tables C-49 and C-51.
C.5 PHELPS DODGE-HIDALGO
This smelter produces about 398 Mg (438 tons) of anode copper
daily. Major process facilities consist of a rotary dryer, flash
furnace, electric slag furnace, three converters, and two anode furnaces
Concentrates, fluxes, and dusts are blended in approximate proportions
of 88, 10, and 2.0 percent, respectively. .Blended materials (containing
0.005 to 0.06 percent arsenic) are then fed to a rotary dryer at about
88 Mg (97 tons) per hour. These materials are heated by gases passing
through the steam superheater and process air preheaters. The volume
of combined gas streams is about 708 Nm3/min (25,000 scfm) They
enter the dryer at about 315°C (600°F) and leave the dryer at about
80 to 100'C (180 to 220°F). The dryer discharges into a lift tank
containing blended material and the fine dust portion of the ESP
C-15
-------
cleaning the dryer offgases. The lift tank material is charged to the
dry charge bins feeding the flash furnace.
A flash furnace is used for smelting. The feed material is
charged through holes in the roof on the reaction shaft side of the
furnace. The furnace reaction shaft is about 8 m (27 ft) inside
diameter by about 36 ft high. The settling chamber is about 23 m
(76.5 ft) long, 10 m (33 ft) wide, and 6 m (20.5 ft) high. The uptake
shaft is about 9 m (30 ft) long, 6.6 m (21.5 ft) wide, and 16.6 m
(54.5 ft) high. These are all inside dimensions. Ambient air, preheated
to 370 to 450°C (700 to 850°F), is fed to the furnace; the process
offgas is about 2,430 Nm3/min (86,000 scfm) at 1,200°C (2,200°F) and
10 percent S02. The furnace operating temperature is normally 1,350°C
(2,460°F). Furnace production is about 715 Mg/day (650 tons/day) of
matte and 1,650 Mg/day (1,500 tons/day) of slag. The slag from the
flash furnace is tapped directly to the electric furnace for further
copper recovery.
Matte from the flash furnace is transported to one of three
Fierce-Smith converters. One converter usually is on a slag blow, one
on a copper blow, and one is prepared for the next matte charges or on
standby. The converter feed rates are 24 Mg (27 tons) of matte from
the flash furnace, 8 Mg (9 tons) of matte from the electric furnace,
and 1.7 Mg (1.9 tons) of flux per hour. The converters produce nearly
13 to 19 Mg/hr (20 to 21 tons/hr) of blister copper and 4.5 to 5.4 Mg/hr
(5.0 to 6.0 tons/hr) of converter slag which is sent to the electric
furnace.
An electric furnace is used for further processing of the flash
furnace slag .and converter slag. The furnace uses about 94 kWh per
ton of charged material. The feed rates are about 66 Mg (60 tons) of
flash furnace slag, 4.5 to 5.4 Mg (5.0 to 6.0 tons) of converter slag,
and can handle about 11 Mg (12 tons) of reverberatories per hour. The
furnace produces about 7.2 to 8.1 Mg/hr (8.0 to 9.0 tons/hr) of matte
and 50 Mg/hr (tons/hr) of slag. The matte is transferred to the
converters by ladle, and the slag is transported to a slag dump.
Blister copper produced in the converter is transported by ladle
to one of two anode furnaces where it is further reduced to anode
C-16
-------
copper. The copper is then poured Into anode molds on a casting wheel
at 16 to 17 Mg/hr (18 to 19 tonc.hr). The anode copper is loaded on
ra,lcars and sent to a copper refinery for further processing
Fugmve emissions escaping the primary hooding on the converters
are captured by secondary hoods designed by the company. The secondary
ir« r r9 doors 8-8 • (29 ft> ^*•* * <» ^ *<*. -
5.5 m (18 ft) h,gh and cover the converter to the operator floor
level. The design gas volume handled by the converter fugitive system
is about 2,830 ^3/m1n (10MOQ scfm)> and ^ ta^aj ^ ^system
gases varies from 150 to 260°C (300 to 500°F).
Emission measurements were conducted by EPA during July 25-26
1978 ,„ the fugitive converter duct. The location was downstrea.'of
t e converters Three emission measurements were made for uncontrolled
r e ,c ounng the tests, process conditions were closely monitored,
h n r;.COndUCted °^ -" ^ PTOC«S facilities were opera ing
«'th,n norma, ,,raits. The test results are su^arized in Table C-52.
C.6 PHELPS OOOGE-OOU6LAS
The Douglas Reduction Works is a calcine fed smelter producing
about 322 Hg (365 tons) per day of 99.6 percent copper anodes. C per
anodes are sent to a copper refinery for further processing. The
-Jor units at the smelter include 24 roasters, 3 reverberatory furnaces
5 converters, and 2 anode furnaces. "™aces,
The roaster process consists of 24, 7-hearth Herreshoff roasters
rrange ,„ two parallel rows of 12. Only 18 are no™a,,y in operation
at at™. The roasters are standard Herreshoff with a she,, dialer
of about 6.7 m (22 ft). Natura, gas is introduced near the botto.
e roasters As the hot gases rise, feed material introduced t t
to s forced down through the roasters by the use of rabble arms
ch spread the feed around each hearth level and through openings at
level. About 154 to 163 Mg (170 to Z8C tons) of calcin per
roaster is produced each day.
Calcine from the roasters is discharged into holding hoppers
ere u ,S transferred by gravity to larry c,ra and deliver d b rail
to the reverberatory furnace.
C-17
-------
The roaster calcine discharge emission control system consists of
hoods covering the Tarry car with flexible screening or flaps hanging
from the edges of the hoods to the top of the larry car. The hoods
are split open 20 cm (8 in.) in the center to allow the power pole for
the electric motor driving the larry cars to make contact with the
electrical source. Unfortunately, when the hoppers are discharged
into the larry car, the split allows calcine dust to escape from the
hood into the building. The discharge ducts from the hoods are combined
into a single duct leading to cyclones, a baghouse, a fan, and then
discharged through a long duct leading to the roaster/reverberatory
stack discharging to the atmosphere. Each hood can be dampered so
that all the gas volume (draft) is available to the hoods used for
calcine discharge.
The baghouse consists of eight compartments with 180 polyester
bags per compartment. The bags are 328 cm (129 in.) long and 13 cm
(5.0 in.) wide. The baghouse was designed to treat 1,130 actual m3/min
(40,000 acfm) at ambient temperatures. Mechanical shakers are used
for cleaning, and the baghouse and cyclone dusts are returned to the
calcine hoppers by screw conveyors.
Inlet and outlet emission measurements for arsenic and particulates
were conducted by EPA across the baghouse on May 3-5, 1978. Sampling
was conducted upstream of the cyclones and downstream of the fan.
During all tests, process conditions were closely monitored, and
testing was conducted only when the process facilities were operating
within normal operating limits.
Arsenic test results are summarized in Tables C-53 and C-54.
Particulate test results are summarized in Tables C-55 and C-56.
C.7 KENNECOTT-MAGNA, UTAH
This smelter was designed to produce nearly 680 Mg (750 tons) of
anode copper daily. Major process facilities consist of two rotary
concentrate dryers, three continuous Noranda smelting vessels, four
converters, and four anode furnaces.
Concentrates and precipitates, dusts, and slag concentrates are
blended with silica flux in approximate proportions of 78, 4, 12, and
6 percent, respectively. Feed material from storage bins are fed to
C-18
-------
conveyor belts that transport the material to a slinger (high-speed
conveyor system) which charges t.',e material into the smelting vess.els.
Matte and slag are normally tapped about every 30 minutes. The slag
is taken to a processing station where it is cooled, crushed, and
reprocessed to recover additional copper values (slag concentrate)
which are returned to the smelting cycle. Matte is transported by
ladle to the converters. Blister copper from the converters is trans-
ported by ladle to the anode furnaces. The strong S02 offgas streams
from the smelting vessels and converters are cleaned and delivered to
the acid plants.
Emissions from the matte tapping operation (hole) are captured in
an enclosed cubicle. The ladle on a cradle car on rails (designed for
this purpose), is in another cubicle beneath the matte tap floor level
under the smelting vessel. Ducts direct the fumes from these operations
into one duct which leads to the main fugitive duct system at the roof
of the smelting and converter building. The main fugitive duct handles
all fugitive emissions for the smelter area and discharges these
fugitive emissions through a spray chamber and out the main stack
(1,200 ft). The gas flow for the matte tapping hole system is 708 Nm3/min
(25,000 scfm) and for the ladle system is 2,120 Nm3/min (75,000 scfm).
Emissions from the slag tapping operation are captured by a
rectangular duct opening beside the tapping hole with a flow of 340 Nm3/min
(12,000 scfm). A similar rectangular duct opening captures the emissions
from the slag ladle in the area beneath the end of the smelting vessel
The flow rate for the collection of these emissions is 990 Nm3/min
(35,000 scfm).
Emissions escaping the primary converter hoods during the blow,
and some emissions escaping the converters during the roll-out mode'
are captured by secondary hoods and ducted to the main fugitive duct
wh!ch was explained previously. The secondary hooding system has a '
steel shell constructed over the primary hoods. The primary hoods are
permanent, nonmovable hood arrangements above the converters This
allows a transition to the fugitive duct system. The lower part of
the secondary hood system consists of four doors that close over the
converter and ladle area beneath the converter. One door turns down
C-19
-------
from the top, two doors close in the center, and the last door moves
across the bottom area, which completely covers the converter. Unlike
other smelters with secondary hoods, the fugitive system at this
smelter operates at all times, whether the converter is in the blowing
cycle or roll-out mode. The secondary converter doors were not operable
during the tests due to mechanical problems with the doors. The
design gas volume per converter is 2,745 actual n3/min (97,000 acfm)
at 82°C (180°F).
The converter process gas stream was also source tested for
uncontrolled arsenic. The tests were conducted before one of three
fans and the control device to the acid plants. There is a settling
chamber at the converters, and also a plenum chamber with dampers
where the gases from each of the converters can be directed to one,
two, or all three fans (as needed) to the acid plants. The gas volume
for this process stream per converter is abouc 425 Nm3/min (15,000 scfm).
Two rotary dryers are used at this smelter to dry concentrate
feed for the smelting vessels. Emissions from the dryers are controlled
by a cyclone followed by a spray chamber scrubber.
Uncontrolled arsenic emission measurements were conducted by EPA
on November 1-14, 1978, on the slag and matte tapping system, converter
secondary hood system, and the converter process gas stream before the
control system. The rotary dryer scrubber outlet was also measured
for arsenic during this period. During all tests, process conditions
were closely monitored, and testing was conducted only when the process
facilities were operating within normal limits.
The arsenic test results for these processes are summarized in
Tables C-57 through C-62.
C.8 KENNECOTT-HAYDEN
The Kennecott Copper Corporation, Ray Mines Division, smelter at
Hayden, Arizona, was originally put on stream in oiid-1958 and extensively
modernized in 1969 and 1973. The smelter produces nearly 227 Mg
(250 tons) of anode copper daily and consists of a concentrator plant,
fluid-bed roaster, reverberatory smelting furnace, three converters,
two anode furnaces, an anode casting wheel, and a double-contact acid
C-20
-------
Concentrate and precipitates from the concentrator plant are
blended with silica flux in proportions of 86.3, 2.2, and 11.5 percent
respectively. The blended materials (containing 6 to 12 percent
moisture and less than 0.015 percent arsenic) are then fed to a Dorr-Oliver
designed fluid-bed roaster by a screw feeder. The feeder controls the
feed rate and maintains a seal on the roaster. The roaster feed rate
typically ranges from 45.4 to 63.5 Mg/hr (50 to 70 tons/hr). Fluidizing
air averages 425 Nm3/min (15,000 scfm). The air is supplied through
373 tuyeres at the bottom of the reactor vessel to keep the bed
(approximately 1.8 m in depth) constantly fluidized.
The fluidizing air reacts with the sulfur contained in the sulfide
ores to form S02 and calcine. Approximately 50 percent of the sulfur
contained in the feed material is oxidized to S0£. Because the reaction
is exothermic, no auxiliary fuel is needed except for cold startup
The bed temperature is generally maintained between 565 and 620'C
(1,050 to 1,150'F). Most of the calcine produced (85 percent) exits
the reactor as a fine dust suspended in the offgas stream. The offgases
average 623 Nr^/min (22,000 scfm). They are then ducted through a
series of four primary and four secondary cyclones. In the cyclones
about 95 percent of the suspended calcine is recovered and subsequently
conveyed by screw conveyor to the calcine storage bin. The underflow
from the reactor, which accounts for about 15 percent of the calcine
produced, is reclaimed through an underflow valve and transported to
the calcine storage bin by a drag chain conveyor.
Calcine, precipitator dust, and flux are then fed to a single
reverberatory furnace for smelting. The reverberatory furnace (a
suspended arch design) is 9.1 m (30 ft) wide by 30.5 m (100 ft) long
The furnace is charged through two openings at the top by a pair of
Wagstaff feeders. The furnace is charged every 15 minutes for a
duration of 2 to 3 minutes. Unless actually being charged, a slight
negative draft is maintained across the furnace (about -1.5 mm of
H20). The furnace is fired with natural gas but is equipped with oil
burners in the event gas service is interrupted. Slag from the furnace
is periodically tapped into slag pots which are subsequently hauled by
rail to the slag dump. About 650 tons are removed daily. Matte is
tapped into ladles which are moved into the converter aisle when full.
C-21
-------
The matte ladles are then picked up by an overhead crane and
charged to one of three Pierce-Smith converters. Each is 4.0 m (13 ft)
in diameter by 9.1 m (30 ft) long and equipped with 42 tuyeres.
Present converter operation consists of keeping two converters on
charge concurrently. Each is in the blowing cycle for 50 percent of
the time for 24 hours. Air flowing at 595 Nm /min (21,000 scfrn) blows
through the tuyeres in the matte charge, flux added, and iron oxide
slag produced. The slag is then skimmed and poured into ladles.
Unlike most domestic smelters, the converter slag is not charged to
the reverberatory furnace, but is carried to a special slag pit where
it is cooled and subsequently returned to the concentrator and blended
with raw ore.
Additional matte and dope materials (reverts or copper scrap) are
added to an active converter to produce approximately 90.7 Mg (100 tons)
of blister copper per load. Converter process feed rates consist of
about 650 Mg (718 tons) matte/day, 45.4 Mg (50 tons) dope/day, and
72.6 Mg (80 tons) flux/day. The finished blister copper is then
poured into ladles and transported by overhead crane to one of two
anode furnaces. The blister copper is completely oxidized with air
and then reduced with propane or natural gas. Finished anode copper
is then poured into anode molds on a single casting wheel. The anodes
are cooled and subsequently loaded onto rail cars for shipment to a
refinery. Anode production is about 226 Mg/day (250 tons/day) of
97.94 percent copper.
As noted previously, the calcine-laden roaster offgases pass
through four parallel sets of primary and secondary cyclones. An
estimated 95 percent of the dust (calcine) is recovered, and the gas
stream is cooled from 565°C (1,050°F) to about 316°C (600°F). The
2
cyclone exhaust, which has a dust loading of about 57.2 g/Nm (25 gr/scf),
then enters a venturi-type scrubber where most of the particulate is
collected. The scrubbing liquid consists of weak acid which is injected
into the venturi throat at a rate of 1,500 liters/min (395 gpm). The
resultant pressure drop across the throat is about 406 mm (16 in.) of
water. The gas stream then enters the smaller of two Peabody scrubbing
towers. The larger tower is used for scrubbing and cooling the converter
C-22
-------
offgas stream. Both towers consist of a lower humidifying section and
an upper cooling section. The discharge from the venturi enters the
hunndifying section and passes upward through a weak acid spray from
spray nozzles located at the top of the section. The coarser solids
are removed and the heat of the gas evaporates the weak acid water
stre - "
gas stream then enters the cooling section where it passes through
three perforated plates (four on the converter tower) for f1cw distri
but,on and acid bubble formation. The weak acid flowing acroTihT
Plates cools the gas stream to about 46°C (115'F). The pressure drop
across both the venturi and Peabody scrubbers (which services the
roaster offgases) is about 610 m (24 in.) Of water. The clean roaster
gas stream (which contains about 12 percent SO 1 1, th , .
th. ,,„„„,, „ percent iu ) is then combined with
the c aned converter gas strea™ prior to entering the acid plant.
(130 On' LT>beTry fUrna" °ffgaSeS 3Verage 'PP™"»*ly 3,682 Nm3
(130,000 scfm). They then pass through a pair of waste heat boilers
at approximately 1,260'C (2,300'F) and exit at 343'C (650°F) The
r;::;;; ;::;";„;;;•;:;;«;;— :;-,::-?
(54.000 ft ). The gas retention t1« within the ESP is about 14 seconds
Th. average gas velocity is 0.5 „ (1.6 ft) per second. The gases
ut 288°
(70 00scTh ' /
(70,000 scfm). They are collected in water-cooled hoods and then
exhausted through a gas cooler in which the gas stream is re by a
concurrently Hewing, „, trasonically dispersed water spray. e
cooled gas stream (371°c) flows through an induced fan plenum and into
an e ectrostatic precipitator for paniculate removal. Tlle re p Or
™nu actured by Western Precipitator, has two chafers with
22 c»*er is
ri-,
(which contains
the
C-23
-------
the two Peabody scrubbing towers and is treated similarly to the
roaster gas stream.
After the cleaned and cooled roaster and converter gases are
combined, the resultant gas stream enters three parallel trains of two
mist precipitators in series, where acid mist and any remaining solids
are precipitated. The gas stream (which typically contains 5 to
8 percent SCL) then enters the double absorption acid plant where it
is dried, the SCL converted to S03> and the S03 absorbed in acid to
form strong acid. Although designed to produce 1,769 Mg (1,950 tons)
of sulfuric acid per day, only about 771 Mg (850 tons) per day of
93.5 percent strength sulfuric acid is actually produced. This represents
about 99.5 percent conversion.
The gas stream exiting the mist precipitator enters a drying
tower where 93 percent acid is used to remove water vapor prior to
entering the converter and absorbing systems. The gas stream then
goes to the main blowers. One, two, or three 1,490 kW (2,000 hp)
blowers are used depending on the volume of gas available for processing.
The gas stream exits the main blowers into the converter system; the
S0? contained in the gas stream is then converted to SO^. The converter
contains four layers of vanadium catalyst arranged in three passes.
The first pass consists of two layers; the second and third passes
each have one layer. Tube and shell heat exchangers are used to
preheat the S02 gas stream to the operating temperature by utilizing
the heat generated from the exothermic chemical reaction within the
converter. The preheated gas enters the top of the converter and
passes through the catalyst layers, exiting the converter after each
pass, and entering a heat exchanger for cooling. During plant startups
or during periods of low S02 gas strength, a preheater is used to
raise or maintain the catalyst temperature at a level at which conversion
will take place.
Two absorbing towers are used, an interpass absorber and a final
absorber. The gas leaving the second converter pass goes through two
heat exchangers and then to the interpass absorber where the SO is
absorbed by 98 to 99 percent acid. The gas stream then enters the
final converter pass where nearly all the remaining SO,, is converted.
C-24
-------
The gas stream then enters the final absorber where the last traces of
S03 are absorbed. The acid product is pumped through a series of
cooling coils and then stored in any of four 5,000-ton storage tanks.
The exit gas from the final absorber passes through a mist eliminator
and is then exhausted through a 30.5 m (100 ft) stack. The SO,, concentra-
tion in the exhaust gas is generally about 230 ppm.
Arsenic emission measurements were conducted by EPA on December 10-13,
1976. Concurrent inlet and outlet measurements were performed across
the venturi and Peabody scrubbers. The scrubbers treat the roaster
offgases for particulates before they are combined with converter
process gases and subsequently treated for S02 in tne double-contact
acid plant. Additional arsenic measurements were made at the acid
plant outlet.
Process conditions were carefully observed, and testing was
conducted only when the subject process facilities were operating
within normal operating limits. The test results are summarized in
Tables C-63, C-64, and C-65.
C.9 TAMANO SMELTER (HIBI KYODO SMELTING CO.,) JAPAN
The Tamano smelter, a toll smelting facility, is located 7 km
(11 miles) southwest of Uno port and has a production capacity of
8,500 tons per month of electrolytic copper. The smelter consists of
one flash furnace, three converters, two refining furnaces, and one
concentrate dryer. Of the three converters, usually one is in operation,
one is kept hot, and one is kept on standby.
Converter primary offgases which range between 65,000 and 75,000 Nm3/h
(38,000 to 44,000 scfm) and flash furnace offgases which range between
65,000 and 80,000 Nm3/h (38,000 to 47,000 scfm) are treated in two
separate pairs of electrostatic precipitators for particulate removal
then introduced into a 156,000 Nm3/h (92,000 scfm) capacity acid plant
f°r S°2 removal. S02 content of the converter and flash furnace gases
usually range between 6.5 and 7.5 percent. The acid plant SO, removal
efficiency is 99.7 percent. Outlet gases which contain 140 ppm SO
are vented through the main stack. A 98 percent sulfuric acid is 2
produced at a rate of 30,000 tons/month at full capacity
C-25
-------
Offgases from refining furnaces are combined with gas from the
power plant and passed through the concentrate dryer. Total gases
from the dryer are treated in a pair of electrostatic precipitators
prior to introduction to a desulfurization plant.
Each converter system is equipped with a secondary hood system
for fugitive gas capture which encloses the converter mouth and ladle
used for handling molten materials. The hood has two automatic front
doors which are operated pneumatically. The hood has a movable roof
which is slightly inclined toward the front. During closing, the roof
slides to its right. During the converter operation when the hood
roof is opened, fugitive emissions from the roof are controlled by an
air curtain system [rated at 70,000 Nm3/hr (41,000 scfm)].
The bulk of fugitive gases up to 190,000 Nm3/hr (112,000 scfm)
with low S02 content collected in the secondary hood are passed through
a dust chamber, a baghouse system, and the main stack to the atmosphere.
Another smaller volume, fugitive gas stream with a high S02
content is continuously fed to a 200,000 Nm3/hr (118,000 scfm) capacity
desulfurization plant. Gas volume to the desulfurization plant includes
about 72,000 Nm3/hr (42,000 acfm) from the refining furnace, power
plant, and flash dryer, and about 30,000 Nm3/hr (18,000 scfm) leakage
gases from the converters and refining furnaces.
On March 12 and 13, 1980, visible emission observations were made
of the converter secondary hood system during various nodes of converter
operation. Tables C-75 through C-80 present the results of these
observations.
C.10 TEST DATA (TABLES)
This section contains summary data tables of the arsenic and
particulate emission tests, and the visible emissions observations,
conducted by EPA between December 1976, and June 1980. The following
notes apply to Tables C-3 through C-65:
a) Data not reported.
b) Run no. 1 of Reverberatory-North data was performed
on 6/28/77.
c) Not applicable.
C-26
-------
Table C-l. SUMMARY OF EMISSION TESTS
Plant
o
i
Phelps Dodge
AJo. Arizona
ASARCO
II Paso. Texas
Phelps Dodge
'Mayas. New
Hex ico
s Dodge
Douglas .
Ar i zona
Process
Facility
Acid plant from converters
Heverberatory furnace
(matte and slag tapping)
Converters (blow cycle)
HJ Itlhearth roasters and
reverberatory furnace
Mult(hearth roaster discharge
Heverberatory furnace
(matte tapping)
Converter building
Converter secondary hoods
Mu! Unearth roaster
•lischarge
Hot ESP + mlst
ESP * coolers
None
None
Spray chamber/
cold ESP
Discharged Into
above system
Baghouse
Baghouse
None
Daghouse
SampI Ing
Locatton(s)
Inlet and outlet
Before entering
main stack
Before entering
main stack
Inlet and outlet
Inlet
Inlet
Inlet and outlet
Inlet
Inlet and outlet
Sample
Type
Arsenic
Arsenic
Particulate
S02
Arsenic
Particulate
S02
Arsenic
Particulate
Arsenic
Particulate
S02
Arsenic
Particulate
S02
Arsenic
Particulate
S02
Arsenic
Arsenic
Par t iculate
S0>
-------
Table C-l. SUMMARY OF EMISSION TESTS (Concluded)
Plant
ASARCO
Tacoma. Washington
Kennecott
Magjna, Utah
Anaconda
Anaconda. Montana
Kennecolt
llayden, Arizona
Process
Facility
Arsenic roasters
Reverberalory furnace
Matte tappping
Slag tapping
Converter slag return
Slag dumping
Metallic arsenic
Anode furnace
Converters
Converter
Secondary hood
Roll-out cycle
Noranda furnace
Matte tapping
Slag tapping
Concentrate dryer
Fluid-bed roaster, electric
furnace, and converter
Fluid-bed roaster
Fluid-bed roaster and
converters
Control
Equipment
Baghouse
Cold ESP
Cold ESP
Cold ESP
Cold ESP
None
Baghouse
None
Acid Plant
None
None
None
None
Scrubber
Spray chamber/
baghouse
Venturi scrubber
Acid plant
Sampl ing
.Locatlon(s)
Inlet and outlet
Outlet
Inlet
Inlet
Inlet
Dump site (grab
samples of slag)
Inlet and outlet
Inlet
Inlet
Uncontrolled duct
Uncontrolled duct
Uncontrolled duct
Uncontrolled duct
Outlet
Inlet and outlet
Inlet and outlet
Outlet
Sample
Type
Arsenic
S02
Arsenic
Arsenic
Arsenic
Arsenic
Arsenic
Arsenic
Arsenic (tests
by ASARCo)
Arsenic
Arsenic
so2
Arsenic
S02
Arsenic
S02
Arsenic
S02
Arsenic
Arsenic
Partlculate
Arsenic
Arsenic
ro
CD
-------
Table C-2.
INDEX TO ARSENIC AND PARTICULATE TEST DATA TABLES BY PROCESS FACILITY AND SAMPLE TYPE
I
ro
^
Process Facility
Arsenic roasters
Fluid-bed roaster
Multi-hearth roasters and
reverberatory furnace
FluJd-bed roaster, electric
furnace, and converter
Fluid-bed roaster and
converters
Calcine discharge
Roaster calcining fugitives
Multi-hearth roaster
fugitive discharge
Reverberatory furnace
Reverberatory furnace
Matte tapping
Matte tapping
Matte and slag tapping
Noranda furnace
Matte tapping
Reverberatory furnace
Slag tapping
Noranda furnace
Slag tapping
Reverberatory furnace
Converter slag return
Plant
ASARCO-Tacoma
Kennecott-Hayden
ASARCO-E1 Paso
Anaconda- Anaconda
Kennecott-Hayden
ASARCO-Tacoma
ASARCO-E1 Paso
Phelps Dodge-
Douglas
ASARCO-Tacoma
ASARCO-Tacoma
ASARCO-E1 Paso
Phelps Dodge-AJo
Kennecott-Magna
ASARCO-Tacoma
Kennecott-Magna
ASARCO-Tacoma
Control Equipment
Baghouse
Venturl scrubber
Spray chamber/
cold ESP
Spray chamber/
baghouse
Acid plant
None
Spray chamber/
cold ESP
Baghouse
Cold ESP
Cold ESP
Baghouse
None
None
Cold ESP
None
Cold ESP
Sample Type
Arsenic
Arsenic
Arsenic
Partlculate
Arsenic
Partlculate
Arsenic
Arsenic
Arsenic
Partlculate
Arsenic
Partlculate
Arsenic
Arsenic
Arsenic
Partlculate
Arsenic
Partlculate
Arsenic
Arsenic
Arsenic
Arsenic
Table Numbers
In Appendix C
C-3, C-4
C-63, C-64
C-13, C-14. C-15, C-16, C-17
C-18, C-19, C-20, C-21, C-22. C-29
C-34, C-35, C-36, C-37
C-38, C-39, C-40, C-41
C-65
C-30
C-31
C-53, C-54
C-55, C-56
C-8
C-10
C-32
C-33
C-48
C-49
f-51
\j~jy
C-ll
f-fin
\j — UU
C-12
-------
Table C-2. INDEX TO ARSENIC AND PARTICIPATE TEST DATA TABLES BY PROCESS FACILITY AND SAMPLE TYPE (Concluded)
Process Facility
Converters
Converters (blow cycle)
Converters
(V, Converter fugitives
o Full cycle
Rollout phase
Acid plant from converters
Converter building
Converter secondary hoods
Metallic arsenic process
Concentrate dryers
Plant
ASARCO-E1 Paso
Phelps Dodge-Ajo
Kennecott-Magna
Kennecott-Magna
Kennecott-Magna
Phelps Oodge-Ajo
ASARCO-E1 Paso
Phelps Dodge-Hidalgo
ASARCO-Tacoma
Kennecott-Magna
Control Equipment
Acid plant
None
Acid plant
None
None
Hot ESP, mist ESP
and coolers
Baghouse
None
Baghouse
Scrubber
Sample Type
Arsenic
Arsenic
Partlculate
Arsenic
Arsenic
Arsenic
Arsenic
Arsenic
Partlculate
Arsenic
Arsenic
Arsenic
Table Numbers
In Appendix C
C-23, C-24
C-50
C-51
C-58
C-61
C-62
C-42, C-43, C-44, C-45. C-46. C-47
C-25, C-26
C-27, C-28
C-52
C-5S C-6S C-7
C-57
-------
Table C-3. SUMMARY OF ARSENIC TEST DATA - ROASTER
BAGHOUSE INLET, ASARCO-TACOMA SMELTER
r\un iiu .
— .
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
t _
1
""'
9/15/78
80
a
a
173,621 '
201
A A
f. *f
0,2
20.4
0.65
0.1371
0.1301
204.0
0.1377
0.1303
204.3
2
— — — __
9/15/78
84
a
a
175,277
185
3.4
0.2
20.4
0.96
0.1280
0.1249
192.2
0.1295
0.1261
194.1
3
— —_
9/16/78
80
a
a
184,141
203
4.5
0 7
w • c.
20.4
0.81
0.1101
0.1092
173.6
0.1111
0.1103
175.2
Average
81
a
a
177,680
197
4.1
01-1
.2
20.4
0.81
0.1248
0.1212
189.6
0.1258
0.1220
190.9
Percent Isokinetic
94.6
108.2
95.0
C-31
-------
Table C-4. SUMMARY OF ARSENIC TEST DATA - ROASTER
BAGHOUSE OUTLET, ASARCO-TACOMA SMELTER
— • __
Run No.
~~~~~~~~~ — — — — _ __ __ _ _
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %}:
Water
C02
2
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
— — — — — — -^—. _ __
i
i
•"—*—"•——————*_
9/15/78
120
a
a
171,887
191
5.9
0.2
20.4
0.70
0.00027
0.00026
0.396
0.00028
0.00027
0.416
'
9/15/78
120
a
a
174,633
189
5.0
0.2
20.4
0.80
0.00027
0.00026
0.401
0.00028
0.00027
0.428
•~— — — — •
3
9/16/78
120
a
a
178,671
180
5.6
0.2
20.4
0.52
0.00028
0.00028
0.439
0.00065
0.00064
0.993
•" •" "-•i .. i i-,ii..
Average
120
a
a
175,064
187
5.5
0 2
VJ » t_
20.4
0.67
O.OOC
0.000
0.412
O.OOC
0.000
0.612
Percent Isokinetic
99.5
100.1
102.3
C-32
-------
:~5- SUMMARY OF ARSENIC TEST DATA -
KITCHEN BAGHOU3E INLET, ASARCO-TACOMA SMELTER
Average
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
co2
n
U2
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
?b/hr
9/24/78
96
a
a
14,505
133
•5 (•
-J.O
0.2
20.8
0.32
0.7503
0.7484
93.23
0.7504
0.7486
93.23
9/24/78
96
a
a
16,590
136
4.0
0.2
20.8
0.57
0.6632
0.6400
94.21
0.6632
0.6454
94.21
9/25/78
96
a
a
17,560
140
2.9
0 2
w * t.
20.8
0.49
0.6593
0.6400
99.18
0.6695
0.6402
99.18
96
a
a
16,218
136
3.5
Or\
.2
20.8
0.46
0.6909
0.6755
95.54
0.6944
0.6756
95.54
99.2
97.5
94.8
C-33
-------
Table C-6. SUMMARY OF ARSENIC TEST DATA --
METALLIC ARSENIC BAGHOUSE INLET, ASARCO-TACOMA SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
Percent Isokinetic
1
9/24/78
94
a
a
12,989
242
2.0
0.1
20.0
0.24
0.8440
0.8343
93.79
0.8441
0.8345
93.81
100.7
2
9/24/78
96
a
a
15,147
233
3.1
0.1
20.0
0.17
0.0004
0.0004
0.0544
0.0006
0.0006
0.0813
94.5
3
9/25/78
96
a
a
15,469
207
2.7
0.1
20.0
0.22
0.9289
0.8914
123,0
0.9289
0.8915
123.0 •
98.5
Average
95
a
a
14,533
227
2.6
0.1
20.0
0.21
0.591
0.527
72.27
0.591
0.527
72.29
During this sample run the metallic arsenic process may not have been operatir
C-34
-------
. C'7' SUMMARv OF ARSENIC TEST DATA -
ARSENIC BAGHOUSE OUTLET (METALLIC AND KITCHEN)
ASARCO-TACOMA SMELTER
— — ' — —
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. 35):
Water
CO,
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
— •
i
i
9/24/78
96
a
a
29,109
182
1.0
0.2
20.4
0.28
0.0303
0.0303
7.59
0.0304
0.0304
7.61
— — , —
1
9/24/78
96
a
a
35,958
163
2.1
0.2
20.4
0.19
0.0068
0.0066
2.11
0.0069
0.0067
2.13
— — — _ — *___^_ _
3
•
9/25/78
96
a
a
33,726
160
2.8
0.2
20.4
0.63
0.0417
0.0401
12.07
0.0420
0.0403
12.14
- — — — — — _____
Average
96
a
a
32,931
160
2.0
0.2
20.4
0.37
0.0253
0.0248
7.26
0.0255
0.0249
7.29
Percent Isokinetic
97.5
101.3
97.8
C-35
-------
DnrDD~8; SUMMARY OF ARSENIC TEST DATA -
REVERB ESP OUTLET, ASARCO-TACOMA SMELTER
ruin no.
' — • .
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
C02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
• — • — .
9/15/78
116
a
a
441,557
220
4.4
0.0
20.0
0.54
0.01632
0.01538
61.68
0.01669
0.01574
63.09
2
•
9/16/78
108
a
a
454,539
214
5.1
0.0
20.0
1.17
0.00897
0.00863
35.04
0.00915
0.00881
25.77
3
" ™ •••^"^^•"•^^•^••^MW^
9/18/78
108
a
a
443,619
188
3.7
0.0
20.0
0.32
0.00375
0.00364
13.99
0.00418
0.00405
15.59
Average
•
111
a
a
443,238
207
4.4
0 0
w • VJ
20.0
0.68
0.009
0.009
36.90
0.010
0.009
38.15
Percent Isokinetic
104.2
107.7
102.2
C-36
-------
Table C-9. SUMMARY OF ARSENIC TEST DATA -
CALCINE DISCHARGE, ASARCO-TACOMA SMELTER
nun NO.
— — —
Date
Test Duration -. min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
C02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
_
9/20/78
15
a
a
1,239
78
2.5
0.0
20.0
0.07
0.0939
0,0936
1.007
0.0946
0.0943
1.014
2
— — — — — •
9/20/78
13
a
a
1,293
78
0.0
0.0
20.0
1.32
0.1463
0.1458
1.623
0.1497
0.1492
1.661
3
• — • — — — — __
9/21/78
7
a
a
1,524
80
0.0
0 0
\J • w
20.0
0.22
0.2074
0.2053
2.721
0.0292
0.2071
2.744
Average
12
a
a
1,352
79
0.83
O/^
.0
20.0
0.54
0.1399
0.1384
1.784
0.1418
0.1724
1.806
96.6
94.2
108.1
C-37
-------
Table C-10. SUMMARY OF ARSENIC TEST DATA
MATTE TAPPING, ASARCO-TACOMA SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
C02
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
9/19/78
78
a
a
18,944
134
1.5
0.0
20.0
0.18
0.02146
0.02127
3.490
0.02297
- 0.02271
3.26
2
9/20/78
75
a
a
18,181
163
0.8
0.0
20.0
0.39
0.10649
0.10394
16.59
0.10657
0.10403
16.60
3
9/21/78
74
a
a
18,329
164
0.9
0.0
20.0
0.21
0.10332
0.10130
16.24
0.10344
0.10142
16.26
Average
76
a
a
18,485
154
1.1
0.0
20.0
0.26
0.07
0.07-
12.11
0.07
0.07
12.20
Percent Isokinetic
90.5
91.0
90.0
C-38
-------
Table C-ll. SUMMARY OF ARSENIC TEST DATA
SLAG.TAPPING, ASARCO-TACOMA SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
2
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
i
j.
9/19/78
120
a
a
18,351
68
1.5
0.0
20.0
0.03
0.00340
0.00335
0.536
0.00348
0.00343
0.548
9/20/78
111
a
a
16,571
106
2.3
0.0
20.0
0.05
0.01009
0.01005
1.431
0.01011
0.01007
1.434
3
9/20/78
60
a
a
17,219
119
0.3
0.0
20.0
0.07
0.00790
0.00784
1.170
0.00793
0.00787
1.175
Average
97
a
a
17,380
98
1 4
n n
u « u
20.0
0.05
0,00676
0.00670
1.046
0.00681
0.00675
1.052
Percent Isokinetic
91.2
95.9
92.1-
C-39
-------
Table C-12. SUMMARY OF ARSENIC TEST DATA --CONVERTER SLAG
RETURN, ASARCO-TACOMA SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %) :
Water
C02
02
S0
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
Percent Isokinetic
9/19-21/78
23
a
a
23,207
95
0.8
0.0
20.0
0.7
0.00139
0.00135
0.2763
0.00149
0.00145
0.2962
91.4
C-40
-------
Table C-13. SUMMARY OF ARSENIC TEST nnra
R ESP INLET (ROASTER), ASARco-EL PASO
Average
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
C02
S02
Emissions - Arsenir
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
' Q ** i"» 21 M -f- T-.—I.' . .
6/26/77
116
41.2
98.0
140,927
173
67
. 7
0.0
16.6
a
0.0060
0.0040
7.200
0.0100
0.0068
12.12
6/27/77
120
28.9
172.9
149,764
211
5.3
0.5
18.7
a
0.0080
0.0052
10.32
0.0103
0.0067
13.21
6/28/77
120
42.7
395.5
56,040
231
9.4
n A
U.'f
20.5
a
0.0312
0.0188
14.99
0.0380
0.0229
18.25
119
37 6
w / * \J
222 1
t"£- *- * .1
115,577
205
7.2
0.3
18.6
a
0.0109
0.0069
9 807
v « w vj /
0.0147
0.0094
13.58
102.4
66.8
109.2
C-41
-------
Table C-14. SUMMARY OF ARSENIC TEST DATA --
R & R ESP INLET (REVERB-NORTH), ASARCO-EL PASO SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %) :
Water
CO 2
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
6/26/77
120 -
41.2
98.0
38,664
932
6.9
7.6
9.9
a
0.0471
0.0145
15.61
0.1219
0.0374
40.39
2
6/27/77
120
28.9
172.9
42,288
786
17.9
7.6
9.9
a
0.2888
0.0870
104.7
0.2951
0.0889
107.0
3
6/28/77
120
42.7
395.5
40,421
753
19.0
7.5
9.9
a
0.2903
0.0890
100.6
0.3177
0.0974
110.1
Average
120
37.6
222.1
40,458
824
14.6
7.6
9.9
a
0.212
0.06^
74.95
0.24;
0.07!
86.81
Percent Isokinetic
103.8
107.1
112.7
C-42
-------
Table C-15. SUMMARY OF ARSENIC TFST HAT/I
ESP INLET (REVERB-SOUTH)? ASARCO-EL PASO SMELTER
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent.
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
C02
n
U2
S02
Emissions - Arsenir
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
' a k* /•* « « -f- T ,» *. i. • . . «
6/26/77
120
41.2
98.0
19,759
787
14.2
8 1
o • 1
9 3
J • *J
a
0.1280
0.0404
21.68
0.1289
0.0407
21.83
6/27/77
120
28.9
172.9
22,057
766
13.4
8«
.1
9*^
.3
a
0.7878
0.2544
148.9
0.7967
0.2572
' 150.6
6/28/77
120
42.7
395.5
25,981
614
25.2
ft ?
o. c.
9.2
a
0.6096
0.1949
135.7
0.6462
0.2067
143.9
120
37.6
222.1
22,599
722
16.6
8.1
9.3
a
u
0.5272
0.1692
106.8
0.5444
0.1747
110.5
95.8
97.7
113.4
C-43
-------
Table C-16. SUMMARY OF ARSENIC TEST DATA -- i
R & R ESP INLET (TOTAL), ASARCO-EL PASO SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %) :
Water
C02
02
S02
Emissions - Arsenic
Probe, cyclone,
and fil ter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
Percent Isokinetic
1
6/26/77
120
41.2
98.0
199,350
381
7.5
2.3
14.6
a
0.0261
0.0096
44.49
0.0435
0.0161
74.34
c
2
6/27/77
120
28.9
172.9
214,109
382
8.6
2.7
3.3
a
0.1438
0.0470
263.9
0.1476
0.0487
270.8
c
3
6/28/77
120
42.7
395.5
122,442
485
15.9
4.4
14.6
a
0.2426
0.0812
251.3
0.2594
0.0865
272.3
c
Average
120
37.6
222.1
178,634
411
10.1
2.9
15.5
a
0.12:
0.04(
191.6
0.13'
0.04!
210.9
These data are derived from Tables C-13, C-14, and C-15.
C-44
-------
Table C-17. SUMMARv OF ARSENIC TEST DATA -
R & R ESP OUTLE'i, ASARCO-EL PASO SMELTER
Run No.
— —
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
CO,
X 2
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
• — .
1
— __ .
6/26/77
153
41.2
98.0
201,030
216
C Q
0. O
3c
. b
19.0
a
0.0009
0.0006
1.603
0.0020
0.0012
3.401
— — — — — — — — — .
2
"•
6/27/77
153
28.9
172.9
209,891
219
6.1
3.5
19.0
a
0.0031
0.0019
5.503
0.0041
0.0026
7.400
3
6/28/77
153
42.7
395.5
222,719
221
1.4
0.0
20.8
a
0.0014
0.0010
2.761
0.0018
0.0012
3.393
Average
153
37.6
222.1
211,213
219
4.8
2.3
19.6
a
0.0018
0.0012
3.302
0.0026
0.0017
4.723
Percent Isokinetic
100.4
100.4
95.4
C-45
-------
Table C-18. SUMMARY OF PARTICULATE TEST DATA -- R & R ESP
INLET (ROASTER), ASARCO-EL PASO SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
02
S02
Emissions - Particulate
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
6/29/77
120
43.5
a
68,948
190
9.2
1.8
20.7
a
2.245
1.439
1,327
2.276
1.458
1,345
2
6/30/77
90
47.6
a
63,256
197
10.5
1.4
17.9
a
1.954
1.222
1,059
1.977
1.236
1,072
Average
105
45.5
a
66,102
193
9.8
1.6
19.3
a
2.106
1.335
1,199
2.133
1.352
1,214
Percent Isokinetic
122.0
108.5
C-46
-------
Table C:J?- pMRY OF PARTICULATE TEST DATA - R & R FSP
INLET (REVERB-NORThj, ASARCO-EL PASO SMELTER
Run No.
"" " -
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
C02
0,
z
S02
Emissions - Particulate
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
6/28/77
120
43.5
a
40,711
687
18.8
5.5
5.0
a
2.098
0.6814
732.0
2.188
0.7106
763.3
2
6/30/77
120
47.6
a
39,889
672
16.1
8.0
13.2
a
1.47.4
0.5011
504.0
1.534
0.5214
524.4
Average
120
45.5
a
40,300
680
17.5
6.7
. 9.1
a
1.786
0,5913
619.2
1.861
0.6170
643.9
Percent Isokinetic
109.8
111.6
C-47
-------
Table C-20. SUMMARY OF PARTICULATE TEST DATA -- R & R ESP
INLET (REVERB-SOUTH), ASARCO-EL PASO SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
CO,
f,
02
S02
Emissions - Participate
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
6/29/77
120
43.5
a
22,504
751
9.6
9.9
3.4
a
Q.9863
0.3377
190.2
1.0892
0.3730
210.1 1
2
6/30/77
120
47.6
a
25,680
728
12.8
8.6
8.5
a
5.3713
1..8078
1,182
5.3974
1.8166
,188
Average
120
45.5
a
24,092
739
11.3
9.2
6.1
a
3.323
1.121
718.8
3.385
1.142
731.3
Percent Isokinetic
109.7
112.3
C-48
-------
Table C-21,
SUMMARY OF PANICULATE TEST DATA - g & R
INLET (TOTAL), ASARCO-EL PASO SMELTER*
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %);
Water
C02
02
2
S02
Emissions - Participate
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
6/29/77b
120
43.5
a
132,163
439
1 /•* y-t
12.2
4 A
.3
12.9
a
1.9854
1.0180
2,249
2.0468
1.0431
2,318' 2
6/30/77
120
47.6
a
128,825
450
12.7
4.9
14.6
a
2.1270
1.1154
2,745
2.5217
1.1306
,784
120
45.5
a
130,494
444
12.5
4 6
" • V
13.7
a
2.2319
1.0658
2,494
2.2801
1.0862
2,548
Percent Isokinetic
'These data are derived from Tables C-18, C-19, and C-20."
C-49
-------
Table C-22. SUMMARY OF PARTICULATE TEST DATA - R & R ESP
OUTLET, ASARCO-EL PASO SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
C02
02
S02
Emissions - Particulate
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
6/28/77
153
43.5
a
215,414
220
10.7
1.3
19.0
a
0.0489
0.0292
89.79
0.0619
0.0372
114. 4
2
6/30/77
153
47.6
a
233,278
220
8.7
2.0
20.0
a
0.0372
0.0228
74.28
0.0478
0.0286
95.47
Average
153
45.5
a
224,346
220
9.7
1.7
19.5
a
0.0427
0.0259
81.73
0.0546
0.0327
104.6
Percent Isokinetic
104.6
98.9
C-50
-------
Table C-23. SUMMARY OF ARSENIC TEST DATA - DC ACID
PLANT INLEi, ASARCO-EL PASO SMELTER
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
C02
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
—
6/21/77
105
a
a
58,352
431
9 1
£.1
0.0
17.6
a
0.96.56
0.4882
482.9
1.0291
0.5203
514.6
2
. —
6/22/77
96
a
a
55,189
408
5-»
.7
0.0
17.2
a
0.0810
0.0405
38.32
0.0997
0.0498
47.16
3
6/23/77
96
a
a
54,842
392
4.9
0.0
13.9
a
0.1083
0.0558
50.89
0.1143
0.0589
53.73
Average
99
a
a
56,128
410
4.2
0.0
16.2
a
0.3964
0.2006
196.5
0.4265
0.2158
211.3
Percent Isokinetic
98.5
112.2
106.0
C-51
-------
Table C-24. SUMMARY OF AKSENIC TEST DATA — DC ACID
PLANT OUTLET, ASARCO-EL PASO SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %}:
Water
C02
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
6/21/77
132
a
a
2
6/22/77
132
a
a
68,108 66,574
147
0.0
0.0
17.6
a
0.0001
0,0001
0.043
0.0001
0.0001
0.048
147
0.0
0.0
16.0
a
0.0008
0.0006
0.452
0.0014
0.0010
0.783
3
6/23/77
132
a
a
66,214
151
0.0
0.0
13.6
a
0.0005
0.0004
0.267
0,0005
0.0004
0.277
4
6/24/77
. 132
a
a
64,643
156
0.0
0.0
11.7
a
0.0001
0.0001
0.050
0.0002
0.0001
0.085
Average
132
a
a
65,429
153
O.C
O.C
14.7
a
O.C
O.C
0.;
0.0(
o.oc
0.2?
Percent Isokinetic
73.0
76.2
96.4
97.4
C-52
-------
T_LT /* or*
Table C-25. iunriAKY OF ARSENIC TEST DATA — CONVERTER BUILDTNf
BAGHOUSE INLET, ASARCO-EL PASO SMELTER
—————————
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfmj
Temperature (°F)
Stream (vol. %):
Water
C02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
— — — — - — _
i
i
— — — — _
1/18/78
101
41.7
212.0
521,956
100
i n
1 . U
0.0
20.5
a
0.00270
0.00240
12.11
0.00272
0.00242
12.12
i - ...i —
•— • __
1/19/78
100
32.7
116.1
528,463
97
OM
.4
0.0
20.5
a
0.00090
0.00082
4.28
0.00091
0.00083
4.33
3
1/23/78
100
37.1
165.2
506,479
123
1.1
0 0
\J « \J
20.5
a
0.00067
0.00058
2.88
0.00067
0.00058
2.90
Average
100
37.2
164.4
527,497
107
0.8
On
.0
20.5
a
0.00142
0.00123
6.42
-
0.00143
0.00128
6.45
Percent Isokinetic
95.0
94.2
93.8
C-53
-------
Table C-26. SUMMARY OF ARSENIC TEST DATA -- CONVERTER BUILDING
BAGHOUSE OUTLET, ASARCO-EL PASO SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
S02
Emissions - Arsenic
Probe : cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr .
1
1/18/78
160
41.7
212.0
437,609 526
101
0.0
0.0
20.5
a
0.00017
0.00015
0.672
0.00017
0.00015
0.672
2
1/19/78
160
32.7
116.1
,565 490
99
0.7
0.0
20.5
a
0.000005
0.000005
0.025
0.000010
0.000009
0.040
3
1/23/78
200
37.1
165.2
,455
124
1.3
0.0
20.5
a
0.000006
0.000005
0.026
0.000007
0.000006
0.027
Average
173
37.2
164.4
491,062
108
0.7
0.0
20.5
a
0.000
0.000
0.241
O.OOC
O.OOC
0.246
Percent Isokinetic
105.7
101.1
102.6
C-54
-------
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - lb/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
C02
S02
Emissions - Particular^
Probe, cyclone,
and filter catch
1/17/78
105
a
a
1/18/78
100
41.7
a
1/21/78
100
43.8
a
102
42.8
a
435,427
115
514,279
113
510,318
115
486,675
114
1.2
0.0
20.5
0.017
1.4
0.0
20.5
0.001
1.4
0.0
20.5
0.001
1.3
0.0
20.5
0.006
y / i-o^ i
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
~Pnt" TcnUn/^-*--;-
0.0263
0.0235
98.25
OO rt -i /-»
.2812
0.2507
1,049
0.0080
0.0202
101.9
0.2686
0.2346
1,181
0.0309
0.0276
134.5
0.031
0.027
134.5
0.0215
0.0238
111.5
0.2749
0.2427
788,0
109.7
94.5
96.3
C-55
-------
Table C-28. SUMMARY OF PARTIC'ILATE TEST DATA -- CONVERTER BUILDING
BAGHOUSE OUTLET, ASARCO-EL PASO SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
C02
02
S02
Emissions - Particulate
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
1/17/78
120
a
a
526,089
114
1.0
0.0
20.5
a
0.0051
0.0045
22.87
0.1117
0.0979
503.8
2
1/18/78
160
41.7
a
471,191
118
0.9
0.0
20.5
a
0.0011
0.0010
4.45
0.1402
0.1260
567.4
3
1/21/78
160
43.8
a
496,907
114
1.1
0.0
20.5
a
0.0005 '
0.0004
1.96
0.0081
0.0072
34.29
Average
146
42.8
a
498,062
115
1.0
0.0
20.5
a
0.0072
0.001*
9.76
0.086;
0.077(
368.5
Percent Isokinetic
98.0
97.7
103.7
C-56
-------
Table C-29. SUMMARY OF PARTICULATE TEST DATA - ROASTER/REVERBERATORY
ESP OUTLET, ASARCO-EL PASO SMELTER
Run No.
—
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
C02
S02
Emissions - Particulate
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
i
1/26/78
144
64.0
a
200,352
219
6.0
3.0
20.5
a
0.0608
0.0520
104.0
0.1054
0.0901
180,2
1/26/78
147
64.0
a
207,295
205
6.2
3.0
20.5
a
0.0909
0.0787
160.7
0.1118
0.0968
197.7
3
1/27/78
150
60.1
a
202,651
220
7.3
3 0
20.5
a
0.0411
0.0365
71.1
0.0553
0.0491
95.7
Average
147
62.7
a
203,433
215
6.5
3f\
.0
20.5
a
0.0643'
0,0577
112.0
0.0909
0.0788
157.9
3ercent Isokinetic
94.2
90.6
98.8
C-57
-------
Table C-30. SUMMARY OF ARSENIC TEST DATA — CALCINE DISCHARGE
DUCT, ASARCO-EL PASO SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
1/24/78
60
21.3
83.9
7,705
56
0.1
0.0
20.5
a
0.0049
0.0045
0.326
0.0050
0.0045
0.332
2
1/24/78
60
21.3
83.9
7,809
57
0.4
0.0
20.5
a
0.0014
0.0012
0.092
0.0014
0.0018
0.094
3
1/24/78
60
21.3
83.9
7,659
61
0.3
0.0
20.5
a
0.0034
0.0030
0.223
0.0034
0.0031
0.224
Average
60
21.3
83.9
7,724
58
0.3
0.0
20.5
a
o.oo:
0.002
0.21'
o.oo:
o.oo:
0.21;
Percent Isokinetic
93.4
100.7
94.0
C-58
-------
Table C-31. SUMMARY OF PARTICIPATE TEST DATA - CALCINE DISCHARGE
DUCT, ASARCC-EL PASO SMELTER ui^HAKbh
_ —
Run No.
Date
.Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
02
S02
Emissions - Particulatp
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Tb/hr
Total catch
gr/dscf
gr/acf
Ib/hr
.
— ••
1
i
•
1/24/78
57
23.5
a
7,932
57
0.4
0.0
20.5
0.014
0.0512
0.0457
3.49
0.1861
0.1630
12.70
'"
1/25/78
60
37.8
a
7,737
88
0.3
0.0
20.5
0.005
0.0338
0.0305
2.23
0.1481
0.1336
10.11
3
1/25/78
60
37.8
a
7,394
96
1.0
0.0
20.5
0.001
0.0090
0.0080
0.57
0.3196
0.2861
20.23
Average
59
33.0
a
7,687
80
0.6
0.0
20.5
0.007
0.0313
0.0281
2.09
0.2179
0.1620
14.34
Percent Isokinetic
96.1
103.6
100.6
C-59
-------
Table C-32. SUMMARY OF ARSENIC TEST DATA -- MATTE TAPPING
DUCT, ASARCO-EL PASO SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
1/20/78
240
69.2
474.2
23,296
105
0.3
0.0
20.5
a
0.0029
0.0024
0.578
0.0029
0.0025
0.580
2
1/20/78
240
69.2
474.2
'26,367
98
0.0
0.0
20.5
a
0.0026
0.0022
0.598
0.0027
0.0022
0.600
3
1/25/78
396
39.1
176.0
27,418
73
0.0
0.0
20.5
a
0.0012
0.0010
0.288
0.0012
0.0010
0.288
Average
292
59.1
374.8
25,694
92
0.1
0.0
20.5
a
0.00
0.00
0.48
O.OC
O.OC
0.48
Percent Isokinetic
94.5
90.2
86.9
C-60
-------
Table C-33. SUMMARY OF PARTICULATE TEST DATA - MATTE TAPPING
DUCT, ASARCO-EL PASO SMELTER
Run No.
Date .
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
" Stream (vol. %):
Water
C02
n
U2
S02
Emissions - Particulatp
Probe, cyclone,
and filter catch
gr/dscf
. gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
i
1/25/78
389
39.1
a
26,871
73
0.8
0.0
20.5
0.006
0.0055
0.0047
1.07
0.0978
0.0836
19.12
1/26/78
360
67.6
a
27,370
82
0.0
0.0
20.5
0.009
0.0193
0.0161
3.77
0.1632
0.1370
32.14
3
1/26/78
360
67.6
a
26,802
82
0.0
0.0
20.5
0.020
0.0164
0.0100
3.09
0.0712
0,1366
32.27
Average
369
58.1 .
a
27,015
79
0.3
0 0
20.5
0.012
0.0134
0.0103
2.64
0.1441
0.1191
27.84
Percent Isokinetic
93.7
96.0
91.9
C-61
-------
Table C-34. SUMMARY OF ARSENIu TEST DATA -- SPRAY CHAMBER/BAGHOUSE
INLET-WEST, ANACONDA-ANACONDA SMELTER
Run No,
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . X):
Water
C02
02
SO 2
Emissions - Paniculate
Probe, cyclone,
and filter catch
gr/dscf
. gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
4/20/77
120
97
2,173
76,274
517
11.5
4.0
16.4
a
0.4012
0.1551
262.2
0.4063
0.1571
265.6
2
4/21/77
120
92
1,490
74,002
511
12.2
4.2
18.2
a
0.3550
0.1383
225.1
0.3706
0.1444
235.0
3
4/22/77
120
92
1,914
77,402
466
9.7
0.2
19.8
a
0.3086
0.1309
204.7
0.3153
0.1337
209.1
Average
120
93.7
1,857
75,893
498
11.1
2.8
18.1
a
0.354
0.141
230.6
0.363
0.145
236.4
Percent Isokinetic
92.9
101.9
100.7
C-62
-------
Table C-35. SUMMARY OF ARSENIC TEST DATA - SPRAY CHAHBER/BAGHOUSE
INLET-EAST, ANACONDA-ANACONDA SMELTER
• — — _ — .
Run No.
— •
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %) :
Water
C02
Q
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
•* . _
* • •— • i • —
1
— — — — • -
4/20/77
120
97
2,173
80,193
535
12.9
0.8
17.4
a
0.4939
0,1846
339.4
0.4992
0.1866
343.1
2
—————— — __
4/21/77
120
92
1,490
86,350
523
12.8
0 5
V-* « J
19.5
a
0.3771
0.1442
279.0
0.3864
0.1477
285.9
3
— •
4/22/77
120
92
1,914
86,889
475
10.3
On
.0
19.6
a
0.2559
0.1069
190.6
0.2690
0.1124
200.3
Average
120
93.7
1,857
84,477
511
12.0
0 4
U « *T
18.8
a
0,3725
0.1442
267.8
0.3818
0.1479
274.7
98.7
96.2
98.5
C-63
-------
Table C-36. SUMMARY OF ARSENIC TEST DATA -- SPRAY CHAMBER/BAGHOUSE
INLET (TOTAL), ANACONDA-ANACONDA SMELTER*
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
°2
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
Percent Isokinetic
12 3 Average
4/20/77 4/21/77 4/22/77
120 120 120 120
97 92 92 93.7
2,173 1,490 1,914 1,857
156,467 160,352 164,291 160,370
526 517 471 505
12.2 12.5 10.0 11.6
2.4 2.2 0.1 1.5
16.9 18.9 19.7 18.5
a a a a
0.4588 0.3669 0.2807 0.3E
0.1702 0.1415 0.1182 0.1<
601.6 504.1 395.3 498.4
0.4539 0.3791 0.2908 0.3;
0.1722 0.1462 0.1224 0.1'
608.7 520.9 409.4 511.1
c c c
*These data are derived from Tables C-34 and C-35.
C-64
-------
Table C-37. SUMMARY OF ARSENIC TEST DATA - SPRAY CHAMBER/BAGHOUSE
OUTLET, ANACONDA-ANACONDA SMELTER
' • •
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
n
U2
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
i
i
""" ™"™^ _« ••_*«•
4/20/77
128
97
2,173
153,594
210
19.1
3.6
17.0
a
0.0018
0.0009
2.41
0.0031
0.0016
4.02
4/21/77
128
92
1,490
156,349
215
19.3
4.5
17.5
a
0.0023
0.0012
3.07
0.0041
0.0021
5.53
3
4/22/77
128
92
1,914
164,134
214
17.7
3.5
16.9
a
0.0036
0.0019
5.00
0.0053
0.0028
7.45
Average
128
93.7
1,857
158,026
213
18.7
3 q
*J • ,7
17.1
a
0.0026
0.0013
3.52
0.0042
0.0015
5.71
Percent Isokinetic
100.4
102.2
98.7
C-65
-------
Table C-38. SUMMARY OF PARTICIPATE TEST DATA -- SPRAY CHAMBER/BAGHOUSE
INLET-WEST, ANACONDA-ANACONDA SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %) :
Water
ecu
wV 2
02
S02
Emissions - Particulate
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
4/25/78
120
86
a
77,031
535
13.2
2.3
16.2
a
7.25
2.74
4,786
7.36
2.78
4,858
2
4/25/78
120
86
a
80,363
546
12.3
1.4
16.9
a
6.20
2.34
4,267
6.25
2.36
4,306
3
4/26/77
120
70
a
75,458
573
11.2
1.0
19.4
a
6.40
2.37
4,139
6.43
2.38
4,161
Average
120
80.7
a
77,617
551
12.2
1.6
17.5
a
6.61
2.48
4,397
6.68
2.51
4,442
Percent Isokinetic
97.4
98.5
100.9
C-66
-------
Table C-39. SUMMARY OF PARTICULATE TEST DATA - SPRAY CHAMBER/BAGHOUSE
INLET-EAST, ANACONDA-ANACONDA SMELTER
1 — -'
Run No.
— .
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
CO,
n *
U2
S02
Emissions - Particulate
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
4/25/77
120
86
a
85,140
541
11.4
0.0
20.1
a
5.72
2.20
4,170
5.83
2.24
4,254
2
4/25/77
120
86
a
81,352 -
555
12.2
0.0
19.4
a
5.67
2.13
3,952
5.76
2.16
4,013
3
4/26/77
120
70
a
85,669
577
13.5
0.0
19.6
a
5.93
2.13
4,352
6.05
2.18
4,442
Average
120
80.7
a
84,054
558
12.4
0.0
19.7
a
5.78
2.15
4,162
5.88
2.19
4,240
Percent Isokinetic
97.9
96.1
98/4
C-67
-------
Table C-40. SUMMARY OF PARTIO'LATE TEST DATA -- SPRAY CHAMBER/BAGHOUSE
INLET (TOTAL), ANACONDA-ANACONDA SMELTER*
Run No.
Date
Test Duration - mln.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
C02
^ w t
02
SO 2
Emissions - Particulate
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
Percent Isokinetic
1
4/25/77
120
86
a
162,171
538
12.3
1.2
18.2
a
6.45
2.46
8,956
6.56
2.50
9,112
c
2
4/25/77
120
86
a
161,715
551
12.3
0.7
18.2
a
5,93
2.23
8,219
6.00
2.26
8,319
c
3
4/26/77
120
70
a
161,127
575
12.3
0.5
19.5
a
6.15
2.24
8,491
6.23
2.27
8,603
c
Average
120
80,7
a
161,671
555
12.3
0.8
18.7
a
6.18
2.31
8,559
6.26
2.34
8,682
*These data are derived from Tables C-38 and C-39.
C-68
-------
Table C-41. SUMMARY OF PARTICULATE TEST'DATA - SPRAY CHAMBER/BAGHOUSE
OUTLET. ANACONDA-ANACONDA SMELTER
Run No.
' —
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
2
S02
Emissions - Particulate
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
4/25/77
128
86
a
170,466
217
16.4
4.5
18.5
a
0.0220
0.0119
32.1
0.1387
0.0754
202.6
2
4/25/77
128
86
a
158,252
218
19.4
4.8
18.2
a
0.0162
0.0085
22.0
0.0667
0.0349
90.4
3
4/26/77
128
70
a
165,400
213
19.7
5.2
17.5
a
0.0228
0.0115
32.3
0.0288
0.0146
40.8
Average
128
80.7
a
164,706
216
18.5
4 8
18.1
a
0.0204
0.0107
28.9
0.0789
0.0421
112.5
Percent Isokinetic
96.7
99.4
97.7
C-69
-------
Table C-42. SUMMARY OF ARSENIC TEST DATA — REVERBERATORY ESP
INLET, PHELPS DODGE-AJO SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
Percent Isokinetic
1
7/13/76
120
a
59.3
58,814
622
18.0
a
a
a
0.1076
0.0527
54.3
0.1172
0.0574
59.1
154
2
7/14/76
120
a
72.8
59,583
602
18.6
a
a
a
0.1326
0.0662
67.8
0.1423
0.0710
72.7
152
3
7/14/76
120
a
75.4
60,150
639
16.0
a
a
a
0.1394
0.0673
71.9
0.1459
0.0704
75.2
147
Average
120
a
69.2
59,516
621
17.5
a
a
a
0.126
0.062
64.7
0.13E
0.066
69.0
C-70
-------
Table *C-43. SUMMARY OF ARSENIC TEST DATA - REVERBERATOR? ESP
OUTLET, PHELPS OODGE-AJO SMELTER
— _
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
CO 2
°2
S02
Emissions - Arsenir
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
t
1
_ — ...
7/13/76
120
a
53.7
68,030
595
15.3
a
a
0.0603
, 0.0303
35.2
0.0919
0.0462
53.6
2
1
7/14/76
120
a
44.8
66,275
610
15.4
a
a
a
0.0376
0.0186
21.4
0.0786
0.0389
44.6
3
^
^^""•"•"••^^^•••^—••w"
7/14/76
120
a
51.3
68,738
580
13.5
a
a
a
0.0302
0.0154
17.8
0.0868
0.0442
51.1
Average
•
120
a
49.9
67,681
595
14.7
a
a
a
0.0427
0.0214
24.8
0.0858
0.0431
49.8
145
147
136
C-71
-------
Table C-44. SUMMARY OF ARSENIC TEST DATA - CONVERTER ESP
INLET NO. 1, PHELFS DODGE-AJO SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
Percent Isokinetic
1 2*
6/13/78
144
17.7
38.9
28,075
379
0.0
0.0
20.0
3.83
0.000038
0.000032
0.0091
0.000100
0.000083
0.0241
126.4
3
6/15/78
144
17.7
95.6
26,638
405
4.0
0.0
20.0
4.18
0.000010
0.000008
0.0022
0.000049
0.000042
0.0112
95.2
Average
144
17.7
67.2
27,358
392
2.0
0.0
20.0
4.01
0.00002
0.00002
0.0056
0.00007
0.00006
0.0062
C-72
-------
Table C-45. SUMMARY OF ARSENIC TEST DATA - CONVERTER ESP
INLET NO. 2, PHEI.PS DODGE-AJO SMELTER
— — —
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
n
U2
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
i
i
—— — — — — _ __ ___ _ _
6/13/78
144
17.7
38.9
34,282
389
0.3
0.0
20.0
2.59
0.000003
0.000002
0.0095
0.000003
0.000003
0.0097
"
6/14/78
144
19.7
63.0
28,312
358
0.8
0.0
20.0
3.03
0.000010
0.000009
0.0026
0.000013
0.000011
0.0031
3
6/15/78
144
17.7
95.6
29,265
404
0.0
0.0
20.0
7.60
0.000005
0.000004
0.0011
0.000013
0.000011
0.0029
. .
Average
144
18. ,3
65.8
30,621
384
0.4
0 0
V * V
20.0
4.41
0.000006
0.000005
0.0044
0.000010
0.000008
0.0053
Percent Isokinetic
123.2
92.1
99.0
C-73
-------
Table C-46. SUMMARY OF ARSENIC TFST DATA — CONVERTER ESP
OUTLET (ACID PLANT INLET), PHELPS DODGE-AJO SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
C02
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
6/13/78
106
17.7
38.9
41,016 39
374
3.0
0.0
20.0
4.98
0.000033
0.000028
0.0117
0.000035
0.000030
0.0123
2
6/14/78
111
19.7
63.0
,021
360
2.2
0.0
20.0
2.83
0.000010
0.000008
0.0033
0.000011
0.000010
0.0037
3
6/15/78
109
17.7
95.6
29,692
342
3.7
0.0
20.0
3.99
0.000016
0.000014
0.0042
0.000040
0.000035
0.0104
Average
109
18.3
65.8
36,578
359
3.0
0.0
20.0
3.93
0.00002
0.00001
0.0064
0.00002
0.00002
0.0088
Percent Isokinetic
101.3
102.1
106.8
C-74
-------
Table C-47. SUMMARY OF ARSENIC TEST DATA - ACID PLANT
OUTLET, PHELPS DODGE-AJO SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
n
U2
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
— • _ —
i
i
~^~i"'~— "—"•—•— •————^—^
6/13/78
108
17.7
38.9
47,556 43
140
1 ?
i . C
0.0
20.0
0.24
0.0000020
0.0000017
0.0009
0.0000030
0.0000026
0.0013
6/14/78
108
19.7
63.0
,862 36
164
Or
. 5
0.0
20.0
0.12
0.0000026
0.0000020
0.0011
0.0000048
0.0000039
0.0015
3
6/15/78
108
17.7
95.6
,016
128
o n
V • U
0.0
20.0
0.08
0.0000109
0.0000091
0.0033
0.0000120
0.0000021
0.0040
Average
108
18.3
65.8
42,478
144
Or-
.6
0 0
u • u
20.0
0.15
0.0000052
0.0000043
0.0018
0.0000068
0.0000052
0.0022
Percent Isokinetic
100.6
97,4
97.0
C-75
-------
Table C-48. SUMMARY OF ARSENIC TEST DATA - MATTE TAPPING HOOD
OUTLET, PHELPS DODGE-AJO SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
5/10/78
192
43.9
776.7
63,758 72
111
0.9
0.0
20.0
0.02
0.00045
0.00042
0.248
0.00066
0.00062
0.365
2
5/11/78
120
40.2
741.3
,351 69
111
0.9
0.0
20.0
0.04
0.00076
0.00070
0.472
0.00082
0.00075
0.508
3
5/12/78
120
48.1
882.5
,333
120
1.2
0.0
20.0
0.04
0.00051
0.00046
0.300
0.00057
0.00052
0.344
Average
144
44.1
800.1
69,056 '
114
1.0
0.0
20.0
0.03
0.00
0.00
0.34
Q.OO
0.00
0.40
Percent Isokinetic
104.2
95.8
97.2
C-76
-------
Table C-49. SUMMARY OF PARTICULATE TEST DATA - MATTE TAPPING
OUTLET, PHELP^ DODGE-AJO SMELTER
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
C02
02
S02
Emissions - Participate
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
Percent Isokinetic
5/10/78
192
43.9
a
68,930
110
0.7
0.0
20.0
0.02
0.0133
0.0123
7.89
0.0160
0.0148
9.51
5/11/78
120
40.2
a
73,362
112
0.3
0.0
20.0
0.04
0.0226
0.0207
14.31
0.0466
0.0427
29.49
156
42.0
a
71,140
111
0.5
0.0
20.0
0.03
0.0179
0.0165
11.10
0.0313
0.0288
19.50
108.7
100.1
C-77
-------
Table C-50.
SUMMARY OF ARSENIC TEST DATA -- CONVERTER SECONDARY
HOOD OUTLET, PHELPS DODGE-AJO SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
C02
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
5/10/78
244
27.3
2.9
85,659
142
0.5
0.0
20.0
0.23
0.00247
0.00224
1.813
0,00251
0.00227
1.837
2
5/11/78
120
10.6
1.1
87,444
162
1.2
0.0
20.0
0.34
0.00245
0.00225
1.834
0.00246
0.00225
1.841
3
5/12/78
120
28.4
3.0
85,957
158
0.6
0.0
20.0
0.37
0.00131
0.00121
0.961
0.00134
0.00125
0.989
Average
161
22.1
2.3
86,353
154
0.8
0.0
20.0
0.31
0.002C
0.001$
1.536
0.002:
0.001<
1.556
Percent Isokinetic
101.1
101.1
104.4
C-78
-------
Table C-51. SUMMARY OF PARTICULATE TEST DATA -- CONVERTER SECONDARY
HOOD OUTLET, PHELPS DODGE-AJO SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
02
S02
Emissions - Particulate
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
5/10/78
244
27.3
a
86,369
143
0.5
0.0
20.0
0.23
0.0756
0.0688
55.93
0.1016
0.0925
75.16
2
5/11/78
120
10.6
a
87,708
163
0.7
0.0
20.0
0.34
0.0910
0.0828
68.37
0.1490
0.1355
111.9
3
5/12/78
120
28.4
a
85,698
162
0.5
0.0
20.0
0.38
0.0793
0.0835
58.23
0.1105
0.1025
81.13
Average
161
22.1
a
' 86,591
156
0.6
0.0
20.0
0.32
0.0820
0.0749
60.85
0.1204
0.1101
89.41
Percent Isokinetic
101.1
101.1
104.4
C-79
-------
Table C-52. SUMMARY OF ARSENIC TEST DATA — CONVERTER SECONDARY
HOOD OUTLET, PHELPS DODGE-HIDALGO SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %} :
Water
C02
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
7/25/78
300
a
a
69,076
217
1.5
0.2
20.2
0.38
0.00017
0.00014
0.1026
0.00050
0.00038
0.2947
2
7/26/78
240
a
a
83,346
208
2.2
0.2
20.2
0.48
0.00010
0.00008
0.0718
0.00011
0.00009
0.0778
3
7/26/78
240
a
a
56,063
216
2.2
0.2
20.2
1.10
0.00017
0.00013
0.0807
0.00020
0.00016
0.0965 •
Average
260
a
a
69,495
213
2.0
0.2
20.2
0.65
0.00
0.00
0.08
0.00
0.00
0.15
Percent Isokinetic
85.3
98.6
99.1
C-80
-------
Table C-53. SUMMARY OF ARSENIC TEST DATA -- CALCINE/ROASTER FUGITIVES
BAGHOUSE INLET, PHELPS DODGE-DOUGLAS SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (val. 35):
Water
C02
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
5/3/78
55
a
a
29,697 30,
76
0.5
0.0
20.0
0.13
0.0000003
0.0000002
0.0001
0.000026
0.000023
0.0067
2
5/4/78
49
a
0.38
359 26,
72
0.8
0.0
20.0
0.21
0.0000367
0.0000314
0.0096
0.000098
0.000083
0.0254
3
5/4/78
38
a
7.96
739
65
2.0
0.0
20.0
0.18
0.000193
0.000155
0.0424
0.000281
0.000234
0.0645
Average
48
a
5.32
28,932
71
1.1
0.0
20.0
0.17
0.000062
0.000062
0.0170
0.000114
0.000113
0.0322
Percent Isokinetic
97.4
91.7
110.0
C-81
-------
Table C-54. SUMMARY OF ARSENIC TEST DATA — CALCINE/ROASTER FUGITIVES
BAGHOUSE OUTLET, PHELPS DODGE-DOUGLAS SMELTER
___ — _ , — .
Run No.
Date-
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %) :
Water
C02
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
5/3/78
65
a
7.11
31,539 32
73
1.0
0.0
20.0
0.08
0.000011
0.000010
0.0030
0.000105
0.000090
0.0283
2
5/4/78
42
a
a
,296 31
65
0.9
0.0
20.0
0.19
0.000030
0.000026
0.0082
0.000138
0.000120
0.0381
3
5/4/78
40
a
7.96
,781
79
1.4
0.0
20.0
0.15
0.000069
0.000058
0.0188
0.000095
0.000079
0.0259
Average
49
a
5.82
31,872
73
1.1
0.0
20.0
0.14
0.00
0.00
0.01
O.OC
O.OC
0.02
Percent Isokinetic
96.5
95.0
90.0
C-82
-------
Table C-55. SUMMARY OF PARTICULATE TEST DATA -- CALCINE/ROASTER FUGITIVES
BAGHOUSE INLET, PHELPS DODGE-DOUGLAS SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
2
SO 2
Emissions - Particulate
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
5/3/78
56
a
a
30,294
69
1.2
0.0
20.0
0.13
1.766
1.523
458.3
1.822
1.571
472.9
2
5/4/78
47
a
a
29,153
74
0.0
0.0
20.0
0.21
3.067
•2.615
765.8
3.166
2.699
790.5
3
5/5/78
36
a
a
29,036
65
0.6
0.0
20.0
0.18
2.692
2.282
669.6
2.925
2.480
727.5
Average
46
a
a
29,380
69
0.6
0.0
20.0
0.17
2.508
2.107
631.2
2.638
2.250
663.6
Percent Isokinetic
96.8
95.0
105.4
C-83
-------
Table C-56. SUMMARY OF PARTICULAR TEST DATA -- CALCINE/ROASTER
FUGITIVES BAGHOUSE OUTLET, PHELPS DODGE-OOUGLAS SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
CO 2
02
S02
Emissions - Particulate
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
5/3/78
65
40.1
a
29,018
73
1.3
0.0
20.0
0.09
0.0031
0.0026
0.771
0.0447
0.0387
11.11
2
5/4/78
42
39.5
a
30,985
65
0.6
0.0
20.0
0.19
0.0150
0.0131
3.98
0.1927
0.1655
51.16
Average
54
39.8
a
30,002
69
0.95
0.0
20.0
0.14
0.0091
0.0078
2.37
0.1187
0.1021
31.79
Percent Isokinetic
93.6
87.7
C-84
-------
Table C-57,
SUMMARY OF ARSENIC TEST DATA - CONCENTRATE DRYER SCRUBBER
OUTLET, KENNECOTT-MAGNA SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
i
i
'
11/14/78
90
a
a
43,489
166
18.1
2
20.0
0.06
0.00003
0.00002
0.0120
0.00003
0.00003
0.0143
•— . .,„ ,„ , m m
11/14/78
90
a
a
38,677
129
18.2
a
20.0
0.07
0.00046
0.00041
0.1531
0.00047
0.00041
0.1551
3
11/14/78
90
a
a
47,225
118
15.6
a
20.0
0.07
0.00111
• 0.00084
0.3739
0.00099
0.00090
0.4029
Average
90
a
a
43,130
138
17.3
a
20.0
0.07
0.00047
0.00042
0.1797
0.00050
0.00045
0.1908
Percent Isokinetic
102.8
98.4
96.5
C-85
-------
Table C-58. SUMMARY OF ARSENIC ItST DATA — ACID PLANT INLET,
KENNECOTT-MAGNA SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
CO 2
°2
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
11/6/78
124
a
a
47,978
421
5.0
0.0
20.0
3.9
0.0034
0.0029
1.397
0.0055
0.0047
2.278
2
11/7/78
119
a
a
44,725
476
3.0
0.0
20.0
2.4
0.0034
0.0028
1.294
0.0034
0.0029
1.302
3
11/8/78
120
a
a
44,643
409
4.0
0.0
20.0
3.0
0.0020
0.0018
0.784
0.0021
0.0018
0.802
Average
121
a
a
45,782
436
4.0
0.0
20.0
3.1
o.oo;
o.oo;
1.15,
o.oo,
o.oo,
1.46
Percent Isokinetic
97.9
94.1
106.9
C-86
-------
Table C-59. SUMMARY OF ARSENIC TEST DATA -- MATTE TAPPING DUCT,
KENNECOTT-MAGNA SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
C02
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
11/1/78
70
a
a
48,968
126
1.4
0.0
20.0
0.09
0.00030
0.00026
0.1255
0.00036
0.00030
0.1505
2
11/2/78
60
a
a
48,645
111
1.0
0.0
20.0
0.10
0.00052
0.00045
0.2185
0.00087
0.00075
0.3641
3
11/3/78
66
a
a
43,868
119
0.0
0.0
20.0
0.12
0.00115
0.00099
0.4341
0.00136
0.00117
0.5128
Average
65
a
a
47,162
119
0.8
0.0
20.0
0.10
0.00064
0.00056
0.2594
0.00085
0.00074
0.3425
Percent Isokinetic
107.3
97.1
103.5
C-87
-------
Table C-60. SUMMARY OF ARSENIC TEST DATA — SLAG TAPPING DUCT,
KENNECOTT-MAGNA SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
11/1/78
60
a
a
43,308
73
0.3
0.0
20.0
0.003
0.00030
0.00026
0.1119
0.00036
0.00030
0.1327
2
11/2/78
120
a
a
40,541
91
0.9
0.0
20.0
0.008
0.00013
0.00011
0.0443
0.00033
0.00029
0.1152
3
11/3/78
120
a
a
38,914
71
1.0
0.0
20.0
0.004
0.00005
0.00004
0.0172
0.00006
0.00005
0.0217
Average
100
a
a
40,921
78
0.7
0.0
20.0
0.005
O.OOC
0.000
0.057
O.OOC
O.OOC
0.08?
Percent Isokinetic
89.9
93.6
97.0
C-88
-------
table C-61. SUMMARY OF ARSENIC TEST DATA -- CONVERTER FUGITIVES (FULL
CYCLE), KENNFCOTT-MAGNA SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
11/6/78
188
a
a
94,684
105
0.0
0.0
20.0
0.09
0.00028
0.00024
0.2262
0.00034
0.00029
0.2762
2
11/8/78
181
a
a
90,187
103
0.8
0.0
20.0
0.14
0.00013
0,00011
0.1011
0.00030
0.00026
0.2364
3
11/9/78
182
a
a
92,967
61
1.0
0.0
20.0
0.33
0.00044
0.00039
0.3517
0.00056
0.00050
0.4469
Average
184
a
a
92,613
90
0.6
0.0
20.0
0.19
0.00028
0.00025
0.2263
0.00040
0.00035
0.3198
Percent Isokinetic
100.8
103.3
93.8
C-89
-------
Table C-62. SUMMARY OF ARSENIC TEC'i DATA -- ROLLOUT CONVERTER FUGITIVES,
KENNECOTT-MAGNA SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
C02
02
S02
Emissions - Particulate
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
11/6/78
88
a
a
83,303
100
1.0
a
20.0
0.03
0.00019
0.00016
0.1342
0.00019
0.00017
0.1376
2
11/8/78
65
a
a
81,777
106
1.0
a
20.0
0.05
0.00052
0.00044
0.3781
0.00057
0.00048
0.3971
Average
77
a
a
82,540
103
1.0
a
20.0
0.04
0.00035
0.00030
0.2561
0.00035
0.00031
0.2674
Percent Isokinetic
103.2
99.8
C-90
-------
Table C-63. SUMMARY OF ARSENIC TEST DATA — VENTURI SCRUBBER
INLET, KENNECOTT-HAYDEN SMELTER
T NO.
:e
it Duration - min.
rge Rate - ton/hr
em'c Rate - Ib/hr
ck Effluent
Flow rate (dscfm) •
Temperature (°F)
stream (vol. %) :
Water
CO 2
02
S02
sions - Arsenic
robe, cyclone,
nd filter catch
gr/dscf
gr/acf
Ib/hr
)tal catch
gr/dscf
gr/acf
Ib/hr
mt Isokinetic
nrrp<;nrmHinn i-ac-t- *
1
12/10/76
110
61
1.88
16,971
636
26.2
a
4.2
12.4
0.0116
0.0044
1.69
0.0129
0.0049
1.88
146
2
12/11/76
135
64
1.63
16,847
623
22.4
a
4.2
12.4
0.0108
0.0035
1.56
0.0113
0.0045
1.63
157
3
12/13/76
85
63
1.65
19,323
615
32.3
a
4.2
12.4
0.0097
0.0036
1.60
0.0100
0.0037
1.65
135
4*
12/13/76
75
64.5
1.23
19,011
621
27.5
a
4.2
12.4
0.0072
0.0027
1.17
0.0076
0.0029
1.23
126
Average
101
63.2
1.60
18,038
624
27.1
a
4.2
12.4
0.0098
0.0036
1.50
0.0104
0.0037
1.60
C-91
-------
Table C-64. SUMMARY OF ARSENIC iEST DATA — VENTURI SCRUBBER OUTLET,
KENNECOTT-HAYDEN SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
02
S02
Emissions - Arsenic
Prob*, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
12/10/76
145
61
0.04
15,493
114
9.8
a
5.4
11.4
0.00007
0.00006
0.010
0.00028
0.00024
0.037
2
12/11/76
145
64
0.02
18,918
111
9.1
a
5.4
11.4
0.00006
0.00005
0.010
0.00012
0.00010
0.019
3
12/13/76
140
63
0.02
18,017
83
3.8
a
5.4
11.4
0.00004
0.00004
0.006
0.00014
0.00013
0.022
Average
143
62.7
0,03
17,476
103
7.6
a
5.4
11.4
0.00
O.OO1
0.00
0.00
0.00
0.02
Percent Isokinetic
112
115
101
C-92
-------
Table C-65. SUMMARY OF ARSENIC TEST DATA —ACID PLANT OUTLET,
KENNECOTT-HAYDEN SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %}:
Water
C02
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
12/10/76
370
61
0.47
74,746
155
0.0
a
7.7
0.0
0.00014
0.00012
0.093
0.0007
0.0006
0.448
2
12/11/76
265
64
0.79
60,114
158
0.0
a
7.7
0.0
0.00038
0.00033
0.203
0.0015
0.0013
0.773
3
12/13/76
310
63
1.46
77,798
175
0.0
a
7.7
0.0
0.00020
0.00016
0.130
0.0022
0.0018
1.47
Average
315
62.7
0.92
70,886
163
0.0
a
7.7
0.0
0.00024
0.00020
0.142
0.0015
0.0012
0.911
Percent Isokinetic
107
108
95
C-93
-------
Table C-66. VISIBLE EMISSIONS OBSERVATION DATA, EPA METHOD 22-
ROASTER CALCINE DISCHARGE INTO LARRY CARS, ASARCO-TACOMA
Run
No.
^^H^MI^HH
1
2
3
4
5
6
7
3
9
10
11
12
1.3
"^^•••••w
Date
— — •— — —
6/24
6/24
6/24
6/25
6/25
6/25
6/25
6/25
6/26
6/26
6/26
6/26
6/26
1
Observe P
— — _______ __
Duration of
operation,
mi n: sec
1:20
2:40
1:20
1:23
1:58
1:42
1:12
1:20
2:50
1:48
2:30
1:42
3:04
——————
1
^••^'•^'^^^•^MMMM
I time
emissions
observed
0
0
0
0
0
0
0
0
0
0
n
w
o
n
w
——————
— ^— — _
"
Observer
"^ ^^^MMM^MMM
Duration of
operation,
rain: sec
•
1:15
2:40
1:20
1:23
1:52
1:42
1:13
1:20
2:49
1:48
-
~^— •— ^— — •—
""••—»—••—••«
2
•— ~- 1.
i time
anission
observed
— — ^^— — _
0
0
0
0
0
0
0
0
0
0
•^— ^— — _
-'
— — — — —
Mean
duration of
operation,
mi n: sec
^-^— — — — ^».
1:18
* • * «^
2:40
1:20
1-23
« * «»M
1:55
1:42
1:13
1-20
1 • teW
2:50
1-48
^ • ^rw
2:30
1:42
3:04
•^MH^HMBM.^,^..
^"— — »— «-^
Mean
I time
eniissior
observe:
••^•••••.^HHMM.
0
0
0
0
Q
0
0
0
0
Average
1:54
C-94
-------
Table C-67. VISIBLE EMISSIONS OBSERVATION DATA, EPA METHOD 22-
MATTE TAP PORT AND MATTE LAUNDER, ASARCO-TACOMA
^~— — — ^.
Runa Date
i
•' ••
1 6/24
2 6/24
3 6/24
4 6/24
K
5° 6/24
5 6/25
' 6/25
6/25
6/25
6/25
6/25
6/25
6/25
6/25
6/25
6/25
6/25
6/25
•— — — _ _j_
Observer 1
Duration of
operation,
win: sec
> — — >_____^_
6:24
6:00
4:51
6:05
5:28
5:22
5:36
5:08
6:02
5:12
4:50
5:23
5:17
5:13
5:58
•••
X time
emissions
observed
0
0
0
0
o
>rf
0
0
0
0
0
0
0
0
a
o
Observer 2
Duration of
operation,
(nin: sec
~~"^»-"— «—«•••.••
6:36
6:00
4:55
6:10
5:22
5:36
5:10
5:33
5:13
6:37 .
4:53
5:22
5:18
— — • — — — _
2 time
emissions
observed
• -^ —
1
0
3
0
Q
0
0
0
0
0
0
0
0
— . .
1 "
Duration
operation
mi n: sec
6:30
6:00
4:53
6:08
2:58
5:22
5:36
5:09
5:48
5:13
6:37
4:52
5:23
5: 18
5: 13
5:58
Moan
•
-------
Table C-68. VISIBLE EMISSIONS OBSERVATION DATA, EPA METHOD 22—
MATTE DISCHARGE INTO LADLE, ASARCO-TACOMA
Run3
1
2
3
4
5
6b
7
a
9
10
n
12
13
14
15
16
!sb
Date
6/24
6/24
6/24
6/24
6/24
5/25
6/25
5/25
6/25
6/25
6/25
6/25
6/25
6/25
6/25
6/25
6/25
6/25
Observer 1
Duration of
operation,
min:sec
6.30
5:49
4:53
6:12
5:09
5:21
5:02
4:29
5:12
6:16
4:43
5:13
5:15
5:41
% time
emissions
observed
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Observer 2
Duration of % time
operation, emissions
mi n: sec observed
5:40
5:01
6:10
6:31
5:02
5:28"
5:03
4:32
5:13
4:45
5:15
5:09
5:50
0
0
Q
0
0
0
0
0
0
0
0
0
0
Duration
operation
mi n: sec
6:.30
5:45
4:57
6:11
6:31
5:06
5:25
5:03
4:31
5:13
6:16
4:44
5:14
5:12
5:46
Mean
of % time
, emissions
observed
0
0
0
0
0
0
0
0
0
0
0.
0
0
0
0
Average
5:30
Method 22 data for corresponding runs at the matte tap and launder
are presented in Table C-67.
Observations were made only at the matte tap and launder; see Table C-67
C-96
-------
SLAG TAP PORT AND SLAG LAUNDER, ASARCO-TACOMA
Rund
1
2
3
4C
5
6C
7
8
9
10
A
11
Date
6/24
6/24
6/24
6/24
6/25
6/25
6/26
6/26
6/26
6/26
Observer
Duration
of operation,
mirr.sec
12:25b
22:00
14:07
14:10
16:44
17:26
16:14
13:45
15:45
14:29
1
% Time
emissions
observed
98b
15
35b
13
11
2
1
0.3
0
0
Observer 2
Duration % Time
of operation, emissions
min:sec observed
12:26b 99b
21:36 0
13:52b 97b
-_ — '
Mean
Duration
of operation,
mi n: sec
12:26
21:43
14:07
14:10
16: 44
17:26
16:41
13:45
15:45
14:29
Average 15:40
Std. dev.
Mean 1 Time
emissions
observed
8
13
11
2
1
0
0
0
4
11
o
I
aMethod 22 data for corresponding runs at the slag skim discharge point appear in Table C-71.
Observations were made at the entire slag tap process line including the slag tap port, slag
launder, and slag discharge into ladle; and therefore not included in computing the mean of
observations.
cMethod 9 data for corresponding runs appear in Table C-70.
dNo data obtained by Method 22.
-------
Table C-70. VISIBLE EMISSIONS OBSERVATION DATA, EPA METHOD 9-
SLAG TAP AND SLAG LAUNDER, ASARCO-TACOMA
Run
1
2
Average
Maximum
•
Date
6/25
6/25
^»^»— •— ••— a™.
Duration
of operation,
min.
14.75
18
.
16.38
Mean
opacity,
«
a
1.3
10.3
•— — — —— — — — —
6
Maximum
opacity,
V
a
10
30
— — — — — — ^_____
30
'Emission data were taken during entire slag tapping operation.
C-98
-------
o
CO
Table C-71. VISIBLE EMISSIONS OBSERVATION DATA, EPA METHOD 22-SLAG TAPPING AT
SLAG DISCHARGE INTO POTS, ASARCO-TACOMA
Run'bDate
1 6/24
2 6/24
3 6/24
4 6/24
5 6/25
6 6/25
7 6/26
8 6/26
9 6/26
10 6/26
11 6/26
Observer
1
Duration X time
of operation, emissions
m1n:sec observed
12:46
21:09
14:06
14:05
16:34
17:29
15:54
13:48
15:48
14:11
14:45
i
97
93
97
82
91
94
90
86
77
72
82
Observer 2
Duration % time
of operation, emissions
min:sec observed
12:26 73
21:43 99
13:52 95
Mean
duration
of operation,
m1n:sec
12:36
21:26
13:59
14:05
16:34
17:29
15:54
13:48
15:48
14:11
14:45
Average 15:31
Std. dev.
Mean X time
emissions
observed
85
96
96
82
91
94
90
86
77
72
82
86
8
Visible emissions observation data by EPA Method 9 for corresponding runs are presented In Table C-72.
IJ— *-.•!_ "1_ _ • • i
corresponding runs for the slag tap port and launder are
-------
Table C-72. VISIBLE EMISSIONS OBSERVATION DATA, EPA METHOD 9-
SLAG TAPPING AT SLAG DISCHARGE INTO POTS, ASARCO-TACOMA
Runb
1
2
3
4
5
6
7
8
9
10
11
Average
Maximum
Date
6/24
6/24
6/24
6/25
6/25
6/25
6/26
6/26
6/26
6/26
6/26
Duration
of operation,
mm.
c
c
c
13.75
16.75d
11.75d
15
15
13
15
14.32
Mean
opacity,
%
22.7
11.3
16
14.8
10.3
5,5
3.7
12
Maximum
opacity,
V
a
50
30
35
40
20
10
10
50
a
Emission data were taken during entire slag tapping operation.
Method 22 data for corresponding runs appear in Table C-71.
No data were obtained by Method 9.
Reading started after filling of first sla'g pot.
C-100
-------
o
o
Table C-73. VISIBLE EMISSIONS OBSERVATION DATA, EPA METHOD 22-
CONVERTER SLAG RETURN TO REVERBERATORY FURNACE, ASARCO-TACOMA
Run*
1
2
3
4
5
6
7
8
9
10
11
12
Oate
6/24
6/24
6/24
6/25
6/25
6/25
6/25b
6/26
6/26
6/26
6/26
6/26
Observer 1
Duration of
operation,
ml n: sec
1:04
0:47
0:54
0:55
1:04
1:00
1:15
0:55
X time
emissions
observed
—
100
97
100
100
66
85
83
82
— — — . _
— — — • _ .
Observer 2
Duration of X time
operation. emissions
ml n: sec observed
1:05 89
0:47 96
0: 53 100
1:03 100
0:41 93
— — __
Observer 3
Duration of X time
operation, emissions
m1n:sec observed
0:58 100
0:46 100
0:55 100
0:52 100
Mean
duration of
operation.
n In: sec
1:04
0:46
0:53
0:55
1:03
0:52
1:04
1:00
r 15
0:48
Average 0;5B
Std. dev.
Mean
X time
emissions
observed
96
98
100
100
100
100
66
85
Al
88
92
11
^Visible emissions observation data by EPA Method 9 for corresponding runs are'presented In Table C-74.
Uf\ sta t a *-vKl-^4*-,^.J t,., u_ 4.1 _i *\f\
No data obtained by Method 22.
-------
Table C-74. VISIBLE EMISSIONS OBSERVATION DATA, EPA METHOD 9—
CONVERTER SLAG RETURN TO REVERBERATORY FURNACE, ASARCO-TACOMA
Run
1
2
3
4
5
6
7
8
9
10
11
12
Date
6/24
6/24
6/24
6/25
6/25
6/25
6/25
6/26
6/26
6/26
6/26
6/26
Observer 1
Duration of
operation,
win.
a
a
a
1.00
1.25
0.75
1.25
1.25
1.50
1.25
0.75
Average
opacity,
5
17.5
20
23
5
11
12
13
5
Maximum
opacity,
9
M
30
40
35
10
20
20
20
10
Observer 2
Duration of Average Maximum
operation, opacity, opacity,
min. 5 t>
1. 00 16 25
1. 00 23 35
0. 75 23 30
average opacity for all readings - 15X
maximum opacity during all readings - 402
Data were not obtained by Method 9 on 6/24/80.
C-102
-------
o
I
CD
CO
Table C-75. VISIBLE EMISSIONS OBSERVATION DATA, EPA METHOD 9—BLISTER DISCHARGE
FROM CONVERTER AT THE TAMANO SMELTER IN JAPAN3>b'C
Opacity.
X
5
10
15
20
25
30
35
Total time equal to or greater
than given opacity
1st blister
discharge
mln: sec.
8:00
5:00
3:15
1:30
0:30
0:15
I of total
time
53
33
22
10
3
2
2nd blister
discharge
win: sec.
11:30
8:45
5:15
3:15
2:00
0.45
0.15
X of total
time
96
73
44
27
17
6
2
3rd blister
discharge
mln: sec.
1:00
0:30
0.15
% of total
time
29
14
7
Total blister
discharge*"
min. sec:
20:30
14:15
8:30
4:45
2:45
1:00
0:15
X of total
tine
67
47
28
16
8
3
< 1
Observation point: converter secondary hood system.
bData were based on a total of 30 5-roinute observations for three successive blister discharges of the total
four blister discharges during one converter cycle. Duration of each of the three discharges observed were
15 minutes, 12 minutes, and 3.5 minutes, respectively.
cTable C-76 summarizes the observation data into average opacities for each set of 6-minute data.
Total of the three individual blister discharges.
-------
Table C-76. SUMMARY OF AVERAGE OBSERVED OPACITIES FOR BLISTER
DISCHARGE AT THE TAMANO SMELTER IN JAPANa
Sat No. Average Opacityc,%
6
2 8
3 11
4 10
5 9
Based on same observation data used for Table C-75
Observation time for each set was 6 minutes.
cAverage of all sets is 9 percent.
C-104
-------
o
I
o
en
Table C-77. SUMMARY OF EPA METHOD 9 VISIBLE EMISSIONS DATA-INDIVIDUAL AND TOTAL MATTE
CHARGES TO CONVERTER OBSERVED AT THE TAMANO SMELTER IN JAPAN3'b'c'd
Opacity.
X
5
10
25
Total time equal to or greater
than given opacity
1st matte
charge
min: sec.
0:45
0:15
X of total
time
43
14
2nd matte
charge
min: sec.
0:45
0:15
X of total
time
60
20
3rd matte
charge
min: sec.
0:45
0:15
X of total
time
43
14
4th matte
charge
min: sec.
0
X of total
time
Total matte
charge
•In: sec.
2:15
0:30
0:15
X of total
time
B
4
a) Matte charges I. 2 and 3 were successive charges; respective charging times for matte charges 1. 2. 3 and 4 were 1.75 min.. 1.25 mln.. 1.75 min..
and I./a mln.
b) Observation point: converter secondary hood system.
c) Data were based on a tolal of 6.5 minute observation for three successive matte charges at the beginning of one converter cycle and an
intermediate matte charging during the cycle. Average duration of each matte charge was 1.5 minutes.
d) lotal of the four individual matte charges; average opacity for matte charging, based on total observation, is 3.0 percent.
-------
Table C-78. SUMMARY OF VISIBLE EMISSIONS OBSERVATION DATA-
COPPER BLOW AT THE TAMANO SMELTER IN JAPAN3
Set No. b
1
2
3
4
Average Opacity, %
0
0
0
0
Observation point: converter secondary hood system.
Each set is based on 6-minute observation.
C-106
-------
Table C-79. SUMMARY OF VISIBLE EMISSIONS OBSERVATION DATA--
SLAG BLOW AT THE TAMANO SMELTER IN JAPAN3
Set No.b
1
2
3
4
5
Average Opacity, %
0
0
0
0
0
Observation point: converter secondary hood system.
Each set represents a 6-minute observation. Set Nos. 1 and 2
are based on the 1st slag blow and set nos. 3 through 5 are based
on the second slag blow of the three slag blow total of the complete
converter cycle.
C-107
-------
Table C-80. SUMMARY OF VISIBLE EMISSIONS
OBSERVATION DATA—CONVERTER SLAG DISCHARGE AT THE
TAMANO SMELTER IN JAPAN3
Set No.b
1
2
Average Opacity, %
0
0
Observation point: converter secondary hood system.
Each of two consecutive sets of 6-minute observations are made during
one slag discharge.
C-108
-------
C.ll REFERENCES
1. TRW Environmental Engineering Division. Emission Testing of
ASARCO Copper Smelter, Tacoma, Washington. U.S. Environmental
Protection Agency. EMB Report No. 78-CUS-12. April 1979.
2. Katari, V., et. al. Trip for ASARCO Copper Smelter, Tacoma,
Washington, during June 24 to 26, 1980. Pacific Environmental
Services, Incorporated. July 14, 1980. p. 7.
3. Harris, D.L., Monsanto Research Corporation. Air Pollution
Emission Test, ASARCO Copper Smelter, El Paso, Texas. U.S.
Environmental Protection Agency. EMB Report No. 77-CUS-6.
June 20-30, 1977.
4. TRW Environmental Engineering Division. Air Pollution Emission
Test. ASARCO Copper Smelter, El Paso, Texas. U.S. Environmental
Protection Agency. EMB Report No. 78-CUS-7. April 5, 1978.
5. Harris, D.L., Monsanto Research Corporation. Air Pollution
Emission Test, Anaconda Mining Company, Anaconda, Montana. U.S.
Environmental Protection Agency. EMB Report No. 77-CUS-5.
April 18-26, 1977.
6. Radian Corporation. Arsenic Emissions from an Electrostatic
Precipitator of the Phelps-Dodge Copper Smelter in Ajo, Arizona.
U.S. Environmental Protection Agency. EPA Contract No. 68-02-13-19.
April 4, 1977.
7. Rooney, T., TRW Environmental Engineering Division. Emission
Test Report (Acid Plant). Phelps-Dodge Copper Smelter, Ajo,
Arizona. U.S. Environmental Protection Agency. EMB Report
No. 78-CUS-ll. March 1979.
8. Rooney, T., TRW Environmental Engineering Division. Emission
Test Report. Phelps-Dodge Copper Smelter, Ajo, Arizona. U.S.
Environmental Protection Agency. EMB Report No. 78-CUS-9.
February 1979.
9. Rooney, T., TRW Environmental Engineering Division. Emission
Testing of Ph'elps-Dodge Copper Smelter, Playas, New Mexico. U.S.
Environmental Protection Agency. EMB Report No. 78-CUS-10.
March 1979.
10. Rooney, T., TRW Environmental Engineering Division. Emission
Testing of Phelps-Dodge Copper Smelter, Douglas, Arizona. U.S.
Environmental Protection Agency. EMB Report No. 78-CUS-9.
February 1979.
11. TRW Environmental Engineering Division. Emission Testing of
Kennecott Copper Smelter, Magna, Utah. U.S. Environmental
Protection Agency. EMB Report No. 78-CUS-13. April 1979.
C-109
-------
12.
13.
larkin, R. and J. Stelner. Acurex Corporation/Aerotherm Division
Arsenic Emissions at Kennecott Copper Corporation, Hayden, Arizona
U.S. Environmental Protection Agency. EPA Report No. 76-NFS-l
May 1977.
Katari, V. and I.J. Weisenberg. Trip Report--Visit to Hibi Kyodo
Smelting Company's Tamano Smelter during the week of March 10,
1980. Pacific Environmental Services, Incorporated. June 9
1980. Appendix A. '
c-no
-------
APPENDIX D
TEST METHODS
D-l
-------
TEST METHODS
Drl EMISSION MEASUREMENT METHODS
At the beginning of the testing program, a literature search
was conducted to identify available sampling and analytical
techniques for determining arsenic emissions. The search revealed
that most arsenic emissions are in the form of arsenic trioxide and
arsenic pentoxide. According to the literature, the most commonly
used arsenic sampling method has been filtration; however, a number
of reports have indicated that filtration alone is not adequate,
even at ambient temperatures, because arsenic trioxide is a
potentially volatile material. Since it was decided to determine
the amount of arsenic collected as a particulate, the Method 5
train, with back-up impinger collectors, was chosen as the starting
point for the arsenic sampling system. Based on the available
information, a dilute sodium hydroxide solution was chosen as a
collecting solution for the impingers. This, however, presented a
problem since many of the gas streams to be sampled had very high
concentrations of sulfur dioxide (S02), some as high as 3.5 percent.
Therefore, a series of impingers containing hydrogen peroxide was
placed between the filter and the first impinger containing
D-2
-------
sodium hydroxide to remove the $03. This was the configuration
for the "working train" used during the first four field tests.
Analytical methods for arsenic were better defined in the
literature. The most commonly-used procedure is a wet chemical
method based on arsine generation, but certain metals including
copper are interfering agents with this method. Instrumental
techniques include atomic absorption, neutron activation, and
x-ray fluorescence. Atomic absorption spectrophotometry (AAS)
was chosen as the most promising technique because of its ready
availability, familiarity, and low cost; however, arsenic absorbs
weakly and only in the extreme ultra-violet area of the spectrum
(193.7 nm). At that wavelength, molecular absorption by flame
gases and solution species can interfere with arsenic detection.
Despite this, conventional AAS can still be used, provided that:
(1) the fuel and combustion gas are carefully chosen and nonatomic
background correction is used; and (2) arsenic concentrations are
relatively high. However, for lower arsenic concentrations, the
interference effects necessitate the use of a special, more
sensitive technique, such as the hydride generator or the carbon
rod (flameless) system. Before testing began, both conventional
and special AAS methods were compared and evaluated, in terms of
their accuracy, precision, and sensitivity.
D-3
-------
During the first two field tests, samples were collected with
the working train and analyzed either by conventional or carbon
rod AAS depending on the arsenic concentration. The analytical
results showed that 95 to 100 percent of the arsenic was collected
ahead of the NaOH impingers. In the course of analyzing these
samples, the following detailed sample preparation procedure was
developed. Solid samples were digested with 0.1 N sodium
hydroxide, extracted with concentrated nitric acid, evaporated to
dryness, and then redissolved in dilute nitric acid. Liquid
samples were treated similarly except that there was no need for
the sodium hydroxide digestion step. Advantages of the sample
preparation procedure include: (1) reduction of the level of the
collected sulfuric acid in the liquid sample fraction;
(2) dissolution of the arsenic in the solid samples; and
(3) production of a similar solution matrix for all the different
sample fractions.
After the second test, questions were raised about the
sampling and analytical procedures. First of all, laboratory
studies of vaporized arsenic trioxide showed no difference in the
arsenic collection efficiency of 0.1 N sodium hydroxide and pure
water. These results indicated that the arsenic collection
mechanism is condensation and that any condenser would be an
effective collector. Consequently, the conventional Method 5
train (with H20 impingers) was suggested as an alternate to the
D-4
-------
working train and simultaneous testing of the two trains was
planned for the next facility.
Second, an evaluation of the different AAS techniques for
low-concentration uncovered some precision and accuracy
problems with the carbon rod method when large quantities of
dissolved solids (particularly sulfates) are present. The hydride
generator technique, it was found, gives much more precise and
accurate results in the presence of dissolved solids. In view of
this, it was decided that all future low-concentration arsenic
samples (i.e., too low for conventional AA analysis) would be
analyzed by the hydride generator method.
Third, concern was expressed that arsenic was being lost in
the evaporation step of sample preparation. To investigate this,
recovery studies were performed on standard samples. These studies
showed that there is no significant loss of arsenic during the
evaporation step.
Fourth, additional studies showed that while arsenic trioxide
is soluble in alkaline, acid, and neutral solutions, its rate of
dissolution is slow except in alkaline solutions. Therefore, the
clean-up procedure for future test was modified, to require that
the train be rinsed with 0.1 N sodium hydroxide to insure removal
of condensed arsenic.
Fifth, a comparison of arsenic extraction techniques
indicated that higher arsenic yields (by up to 200 percent) can be
D-5
-------
obtained from smelter particulate when a method capable of
dissolving the entire sample is used instead of the less rigorous
acid extraction procedure. As a result, it was decided that in
future tests, filters would be analyzed by both methods, until
more conclusive filter extraction data were obtained.
During the third and fourth field tests, the working train
was used for sampling, but additional runs were taken during the
fifth test using paired trains of the working and alternate
procedures. Analysis of the samples from the paired tests showed
no significant difference in collection efficiency. Therefore, the
final recommendation was to use the alternate train, since it is
easier to operate and analyze. During the fifth and final field
test, the alternate train was used.
Filters from the third, fourth, and fifth field tests were
extracted, using both the total dissolution and acid extraction
procedure. The results showed that filters extracted by the less
rigorous method could in some cases yield 25 percent less arsenic
than if totally dissolved. Based upon these results, the final
recommendation was to extract the filters first by the simple acid
extraction; then, if any undissolved sample remained, to extract
the undissolved solids by the total dissolution method.
D-6
-------
D.2 CONTINUOUS MONITORING
There is currently no available method for continuously
monitoring arsenic emissions. For purposes of demonstrating proper
operation and maintenance of control devices, continuous monitors
are available for measuring opacity from baghouses or electrostatic
precipitators, and measuring pressure drop across scrubbers.
However, these measurements are not necessarily indicators of the
magnitude of arsenic emissions and should not be used for compliance
determinations. In addition, opacity may not be applicable as
an indicator of proper operation and maintenance where baghouses
and precipitators are used to control captured fugitive emissions
because the uncontrolled particulate is very low in
concentration.
The recommended monitoring program for continually assessing
arsenic emissions is a periodic application of the performance test
Method 108 as recommended in Part D.3 below. This is the only
method evaluated at this time for demonstration of compliance with
arsenic emissions.
D.3 PERFORMANCE TEST METHODS
The recommended performance test method for arsenic is Method
108. Based on the development work already discussed, the method
uses the Method 5 train for sampling, 0.1 N sodium hydroxide for
D-7
-------
cleanup, and either conventional or hydride generator AAS for
sample analysis. In order to perform Method 108, Methods 1 through
4 must also be used. Subpart A or 40 CFR 60 requires that facilities
subject to standards of performance for new stationary sources be
constructed so as to provide sampling ports adequate for the
applicable test methods, and platforms, access, and utilities
necessary to perform testing at those ports.
Sampling costs for performing a test consisting of three
Method 108 runs is estimated to range from $10,000 to $14,000. If
in-plant personnel are used to conduct tests, the costs will be
somewhat less.
D.4 REFERENCES
1. Hefflefinger, R.E. and D.L. Chase (Battelle). Analysis of
Copper Smelter Samples for Arsenic Content. Prepared for U.S.
Environmental Protection Agency. Research Triangle Park, NC.
April 1977. 14 p.
2. Haile, D.M. (Monsanto Research Corporation). Final Report
on the Development of Analytical Procedures for the Determination of
Arsenic from Primary Copper Smelters. Prepared for U.S.
Environmental Protection Agency. Research Triangle Park, NC.
February 1978. 27 p.
3. Harris, D.L. (Monsanto Research Corporation). Particulate
and Arsenic Emission Measurements from a Copper Smelter. Prepared
D-8
-------
for the U.S. Environmental Protection Agency. Research Triangle
Park, NC. 77-CUS-5. April 1977. 48 p.
4. Harris, D.L. (Monsanto Research Corporation). Participate
and Arsenic Emission Measurements from a Copper Smelter. Prepared
for the U.S. Environmental Protection Agency. Research Triangle
Park, NC. 77-CUS-6. June 1977. 276 p.
5. TRW, Inc. Emission Testing of Asarco Copper Smelter.
Prepared for the U.S. Environmental Protection Agency. Research
Triangle Park, NC. 77-CUS-7. April 1978. 150 p.
D-9
-------
APPENDIX E
QUANTITATIVE EXPRESSIONS OF PUBLIC CANCER RISKS FROM EMISSIONS OF
INORGANIC ARSENIC FROM LOW-ARSENIC PRIMARY COPPER SMELTERS
E-l
-------
QUANTITATIVE EXPRESSIONS OF PUBLIC CANCER RISKS FROM EMISSIONS OF
INORGANIC ARSENIC FROM LOW-ARSENIC PRIMARY COPPER SMELTERS
E.I INTRODUCTION
E.I.I Overview
The quantitative expressions of public cancer risks presented in this
appendix are based on (1) a dose-response model that numerically relates
the degree of exposure to airborne inorganic arsenic to the risk of getting
lung cancer, and (2) numerical expressions of public exposure to ambient
air concentrations of inorganic arsenic estimated to be caused by emissions
from stationary sources. Each of these factors is discussed briefly below
and details are provided in the following sections of this appendix.
E.l.2 The Relationship of Exposure to Cancer Risk
The relationship of exposure to the risk of getting lung cancer is
derived from epidemiological studies in occupational settings rather than
from studies of excess cancer incidence among the public. The epidemiological
methods that have successfully revealed associations between occupational
exposure and cancer for substances such as asbestos, benzene, vinyl chloride,
and ionizing radiation, as well as for inorganic arsenic, are not readily
applied to the public sector, with its increased number of confounding
variables, much more diverse and mobile exposed population, lack of consoli-
dated medical records, and almost total absence of historical exposure
data. Given such uncertainties, EPA considers it improbable that any
association, short of very large increases in cancer, can be verified in
the general population with any reasonable certainty by an epidemiological
study. Furthermore, as noted by the National Academy of Sciences (NAS)l,
"...when there is exposure to a material, we are not starting at an origin
E-2
-------
of zero cancers. Nor are we starting at an origin of zero carcinogenic
agents in our environment. Thus, it is likely that any carcinogenic agent
added to the environment will act by a particular mechanism on a particular
cell population that is already being acted on by the same mechanism to
induce cancers." In discussing experimental dose-response curves, the NAS
observed that most information on carcinogenesis is derived from studies of
ionizing radiation with experimental animals and with humans which indicate
a linear no-threshold dose-response relationship at low doses. They added
that although some evidence exists for thresholds in some animal tissues,
by and large, thresholds have not been established for most tissues. NAS
concluded that establishing such low-dose thresholds "...would require
massive, expensive, and impractical experiments ..." and recognized that
the U.S. population "...is a large, diverse, and genetically heterogeneous
group exposed to a large variety of toxic agents." This fact, coupled with
the known genetic variability to carcinogenesis and the predisposition of
some individuals to some form of cancer, makes it extremely difficult, if
not impossible, to identify a threshold.
For these reasons, EPA has taken the position, shared by other Federal
regulatory agencies, that in the absence of sound scientific evidence to
the contrary, carcinogens should be considered to pose some cancer risk
at any exposure level. This no-threshold presumption is based on the view
that as little as one molecule of a carcinogenic substance may be sufficient
to transform a normal cell into a cancer cell. Evidence is available from
both the human and animal health literature that cancers may arise from a
single transformed cell. Mutation research with ionizing radiation in cell
cultures indicates that such a transformation can occur as the result of
E-3
-------
interaction with as little as a single cluster of ion pairs. In reviewing
the available data regarding carcinogenicity, EPA found no compelling
scientific reason to abandon the no-threshold presumption for inorganic
arsenic.
In developing the exposure-risk relationship for inorganic arsenic, EPA
has assumed that a linear no-threshold relationship exists at and below the
levels of exposure reported in the epidemiological studies of occupational
exposure. This means that any exposure to inorganic arsenic is assumed
to pose some risk of lung cancer and that the linear relationship between
cancer risks and levels of public exposure is the same as that between cancer
risks and levels of occupational exposure. EPA believes that this assumption
is reasonable for public health protection in light of presently available
information. However, it should be recognized that the case for the linear
no-threshold dose-response relationship model for inorganic arsenic is not
quite as strong as that for carcinogens which interact directly or in
metabolic form with DNA. Nevertheless, there is no adequate basis for
dismissing the linear no-threshold model for inorganic arsenic. The exposure-
risk relationship used by EPA represents only a plausible upper-limit risk
estimate in the sense that the risk is probably not higher than the calculated
level and could be much lower.
The numerical constant that defines the exposure-risk relationship
used by EPA in its analysis of carcinogens is called the unit risk estimate.
The unit risk estimate for an air pollutant is defined as the lifetime
cancer risk occurring in a hypothetical population in which all individuals
are exposed continuously from birth throughout their lifetimes (about 70
years) to a concentration of 1 ug/m3 of the agent in the air which they
E-4
-------
breathe. Unit risk estimates are used for two purposes: (1) to compare
the carcinogenic potency of several agents with each other, and (2) to give
a crude indication of the public health risk which might be associated with
estimated air exposure to these agents. The comparative potency of different
agents is more reliable when the comparison is based on studies of like
populations and on the same route of exposure, preferably inhalation.
The unit risk estimate for inorganic arsenic that is used in this
appendix was prepared by combining the three different exposure-risk
numerical constants developed from three occupational studies.2 The unit risk
estimate is expressed as a range that reflects the statistical uncertainty
associated with combining the three exposure-risk relationships. The
methodology used to develop the unit risk estimate is described in E.2
below. EPA is updating its health effects assessment document for inorganic
arsenic. A preliminary determination by EPA's health scientists is that the
unit risk estimate may change.
E.I.3 Public Exposure
The unit risk estimate is only one of the factors needed to produce
quantitative expressions of public health risks. Another factor needed
is a numerical expression of public exposure, i.e., of the numbers of
people exposed to the various concentrations of inorganic arsenic. The
difficulty of defining public exposure was noted by the National Task
Force on Environmental Cancer and Health and Lung Disease in their 5th
Annual Report to Congress, in 1982.3 jney reported that "...a large
proportion of the American population works some distance away from their
homes and experience different types of pollution in their homes, on the
way to and from work, and in the workplace. Also, the American population
is quite mobile, and many people move every few years." They also noted the
E-5
-------
necessity and difficulty of dealing with long-term exposures because of
"...the long latent period required for the development and expression
of neoplasia [cancer]..."
EPA's numerical expression of public exposure is based on two estimates.
The first is an estimate of the magnitude and location of long-term average
ambient air concentrations of inorganic arsenic in the vicinity of emitting
sources based on dispersion modeling using long-term estimates of source
emissions and meteorological conditions. The second is an estimate of the
number and distribution of people living in the vicinity of emitting sources
based on Bureau of Census data which "locates" people by population centroids
in census tract areas. The people and concentrations are combined to produce
numerical expressions of public exposure by an approximating technique
contained in a computerized model. The methodology is described in E.3
below.
E.I.4 Public Cancer Risks
By combining numerical expressions of public exposure with the unit
risk estimate, two types of numerical expressions of public cancer risks are
produced. The first, called individual risk, relates to the person or
persons estimated to live in the area of highest concentration as estimated
by the dispersion model. Individual risk is expressed as "maximum lifetime
risk." As used here, the word "maximum" does not mean the greatest possible
risk of cancer to the public. It is based only on the maximum exposure
estimated by the procedure used. The second, called aggregate risk, is a
summation of all the risks to people estimated to be living within the
vicinity (usually within 20 kilometers) of a source and is customarily summed
for all the sources in a particular category. The aggregate risk is expressed
as incidences of cancer among all of the exposed population after 70 years
E-6
-------
of exposure; for statistical convenience, it is often divided by 70 and
expressed as cancer incidences per year. These calculations are described
in more detail in E.4 below.
There are also risks of nonfatal cancer and of serious genetic effects,
depending on which organs receive the exposure. No numerical expressions
of such risks have been developed; however, EPA considers all of these risks
when making regulatory decisions on limiting emissions of inorganic arsenic.
E.2 THE UNIT RISK ESTIMATE FOR INORGANIC ARSENIC2
E.2.1 The Linear No-Threshold Model for Estimation of Unit Risk Based on
Human Data (General)4
Very little information exists that can be utilized to extrapolate
from high exposure occupational studies to low environmental levels.
However, if a number of simplifying assumptions are made, it is possible
to construct a crude dose-response model whose parameters can be estimated
using vital statistics, epidemiologic studies, and estimates of worker
exposures. In human studies, the response is measured in terms of the
relative risk of the exposed cohort of individuals compared to the control
group. The mathematical model employed assumes that for low exposures the
lifetime probability of death from lung cancer (or any cancer), P, may be
represented by the linear equation
P = A + BHx (1)
where A is the lifetime probability of cancer in the absence of the agent, x
is the average lifetime exposure to environmental levels in micrograms per
cubic meter Ug/nr*), and BH is the increased probability of cancer associated
with each ng/m3 increase of the agent in air.
If we make the assumption that R, the relative risk of lung cancer for
exposed workers, compared to the general population, is independent of the length
E-7
-------
or age of exposure but depends only upon the average lifetime exposure, it
follows that
P A + BH (XQ + XT)
R =
(2)
PQ A + BH (XQ)
or
RP0 = A + BH (XQ + XT) (3)
where XQ = lifetime average exposure to the agent for the general popu-
lation, X] = lifetime average exposure to the agent in the occupational
setting, and PQ = lifetime probability of respiratory cancer applicable with
no or negligible arsenic exposure. Substituting PQ = A + BH XQ and rearranging
gives
BH = PQ (R - D/XI (4)
To use this model, estimates of R and X] must be obtained from the epidemio-
logic studies. The value PQ is derived from the age-cause-specific death
rates for combined males found in 1976 U.S. Vital Statistics tables using
the life table methodology. For lung cancer the estimate of PQ is 0.036.
E.2.2 The Unit Risk Estimate for Inorganic Arsenic2
As noted in the health effects assessment document5 for inorganic
arsenic, there are numerous occupational studies which relate increased
cancer rates to arsenic exposure. Based on these studies, it is concluded
in the health assessment document that there is substantial evidence that
inorganic arsenic is a human carcinogen. However, many of these studies
are inappropriate for use in developing a unit risk estimate for inorganic
arsenic because the route of exposure was not by inhalation or because it
was impossible to make a reasonable estimate of the population's lifetime
average exposure.
E-8
-------
Three studies, Lee and Fraumeni (1969), Ott et al. (1974), and Pinto
et al. (1977), contained enough pertinent information to make independent
quantitative estimates of human cancer risks due to human exposures to
atmospheric arsenic. The crudeness of the exposure estimates in those
studies is due to such factors as high variability in the chemical measurement
of arsenic, a scarcity of monitoring data, and the necessity of working
from summarized data tables presented in the literature rather than complete
data on all individuals. However, by accepting the data in spite of their
recognized limitations, and making a number of simplifying assumptions
concerning dose-response relationships and exposure patterns, it was possible
to estimate the carcinogenic potency of arsenic. Using a linear model, it
was estimated that the increase in the lung cancer rate per increase of 1
iig/m3 of atmospheric arsenic was 9.4 percent (Pinto et al.), 17.0 percent
(Ott et al.), and 3.3 percent (Lee and Fraumeni). The consistency of these
estimates is very good considering the relative crudeness of the data upon
which they are based. The geometric mean of the rate estimates from the
three studies was calculated to be 8.1 percent. Using this value as a best
estimate and applying equation 4, one calculates the unit risk estimate of
2.95 x TO'3 per ug/m3.
If we assume that the linear model and exposure estimates are correct,
so that the only source of uncertainty is from combining results from the
three different studies, a 95 percent confidence interval for the above
unit risk estimate may be obtained. Upper and lower 95 percent confidence
limits can be obtained by multiplying the unit risk estimate by about 4 and
0.25, respectively. Thus, the 95 percent statistical confidence limits for the
unit risk estimate range from 7.5 x 10-4 to 1.2 x 10~2.
E-9
-------
E.3 QUANTITATIVE EXPRESSIONS OF PUBLIC EXPOSURE TO INORGANIC ARSENIC
EMITTED FROM LOW-ARSENIC PRIMARY COPPER SMELTERS
E.3.1 EPA's Human Exposure Model (HEM) (General)
EPA's Human Exposure Model is a general model capable of producing
quantitative expressions of public exposure to ambient air concentrations
of pollutants emitted from stationary sources. HEM contains (1) an atmospheric
dispersion model, with included meteorological data, and (2) a population
distribution estimate based on Bureau of Census data. The only input data
needed to operate this model are source data, e.g., plant location, height
of the emission release point, and temperature of the offgases. Based on the
source data, the model estimates the magnitude and distribution of ambient
air concentrations of the pollutant in the vicinity of the source. The
model is programmed to estimate these concentrations within a radial distance
of 20 kilometers from the source. If other radial distances are preferred,
an over-ride feature allows the user to select the distance desired. The
selection of 20 kilometers as the programmed distance is based on modeling
considerations, not on health effects criteria or EPA policy. The dispersion
model contained in HEM is felt to be reasonably accurate within 20 kilometers.
If the user wishes to use a dispersion model other than the one contained
in HEM to estimate ambient air concentrations in the vicinity of a source,
HEM can accept the concentrations if they are put into an appropriate
format.
Based on the radial distance specified, HEM combines numerically the
distributions of pollutant concentrations and people to produce quantitative
expressions of public exposure to the pollutant.
E-10
-------
E.3.1.1 Pollutant Concentrations Near a Source. The dispersion model
within the HEM is a gaussian diffusion model that uses the same basic
dispersion algorithm as EPA's Climatological Dispersion Model.6 The algorithm
has been simplified to improve computational efficiency.7 The algorithm is
evaluated for a representative set of input values as well as actual plant
data, and the concentrations input into the exposure algorithm are arrived
at by interpolation. Stability array (STAR) summaries are the principal
meteorological input to the HEM dispersion model. STAR data are standard
climatological frequency-of-occurence summaries formulated for use in EPA
models and available for major U.S. meteorological monitoring sites from
the National Climatic Center, Asheville, N.C. A STAR summary is a joint
frequency-of-occurence of wind speed, atmospheric stability, and wind
direction, classified according to Pasquill's categories. The STAR summaries
in HEM usually reflect 5 years of meteorological data for each of 309 sites
nationwide. The model produces polar coordinate receptor grid points
consisting of 10 downwind distances located along each of 16 radials which
represent wind directions. Concentrations are estimated by the dispersion
model for each of the 160 receptors located on this grid. The radials are
separated by 22.5-degree intervals beginning with 0.0 degrees and proceeding
clockwise to 337.5 degrees. The 10 downwind distances for each radial are
0.2, 0.3, 0.5, 0.7, 1.0, 2.0, 5.0, 10.0, 15.0, and 20.0 kilometers. The
center of the receptor grid for each plant is assumed to be the plant center.
E-3.1.2 The People Living Near A Source. To estimate the number and
distribution of people residing within 20 kilometers of each plant, the
model contains a slightly modified version of the "Master Enumeration
District List—Extended" (MED-X) data base. The data base is broken down
E-ll
-------
into enumeration district/block group (ED/BG) values. MED-X contains the
population centroid coordinates (latitude and longitude) and the 1970
population of each ED/BG in the United States (50 States plus the District
of Columbia). For human exposure estimates, MED-X has been reduced from
its complete form (including descriptive and summary data) to produce a
computer file of the data necessary for the estimation. A separate file of
county-level growth factors, based on 1978 estimates of the 1970 to 1980
growth factor at the county level, has been used to estimate the 1980
population for each ED/BG. HEM identifies the population around each plant
by using the geographical coordinates of the plant. The HEM identifies,
selects, and stores for later use those ED/BGs with coordinates falling
within 20 kilometers of plant center.
E.3.1.3 Exposure?. The Human Exposure Model (HEM) uses the estimated
ground level concentrations of a pollutant together with population data to
calculate public exposure. For each of 160 receptors located around a
plant, the concentration of the pollutant and the number of people estimated
by the HEM to be exposed to that particular concentration are identified.
The HEM multiplies these two numbers to produce exposure estimates and sums
these products for each plant.
A two-level scheme has been adopted in order to pair concentrations
and populations prior to the computation of exposure. The two level approach
is used because the concentrations are defined on a radius-azimuth (polar)
grid pattern with non-uniform spacing. At small radii, the grid cells are
usually smaller than ED/BG's; at large radii, the grid cells are usually larger
than ED/BG1s. The area surrounding the source is divided into two regions,
and each ED/BG is classified by the region in which its centroid lies.
E-12
-------
Population exposure is calculated difrerently for the ED/BG's located
within each region. For ED/BG centroids located between 0.1 km and 2.8 km
from the emission source, populations are divided between neighboring
concentration grid points. There are 96 (6 x 16} polar grid points within
this range. Each grid point has a polar sector defined by two concentric
arcs and two wind direction radials. Each of these grid points and respec-
tive concentrations are assigned to the nearest ED/BG centroid identified
from MED-X. Each ED/BG can be paired with one or many concentration points.
The population associated with the ED/BG centroid is then divided among all
concentration grid points assigned to it. The land area within each polar
sector is considered in the apportionment.
For population centroids between 2.8 km and 20 km from the source, a
concentration grid cell, the area approximating a rectangular shape bounded
by four receptors, is much larger than the area of a typical ED/BG. Since
there is an approximate linear relationship between the logarithm of concen-
tration and the logarithm of distance for receptors more than 2 km from the
source, the entire population of the ED/BG is assumed to be exposed to the
concentration that is logarithmically interpolated radially and arithmetically
interpolated azimuthally from the four receptors bounding the grid cell.
Concentration estimates for 80 (5 x 16) grid cell receptors at 2.0, 5.0,
10.0, 15.0, and 20.0 km from the source along each of 16 wind directions
are used as reference points for this interpolation.
In summary, two approaches are used to arrive at coincident
concentration/population data points. For the 96 concentration points
within 2.8 km of the source, the pairing occurs at the polar grid points
using an apportionment of ED/BG population by land area. For the remaining
portions of the grid, pairing occurs at the ED/BG centroids themselves
E-13
-------
through the use of log-log and linear interpolation. (For a more detailed
discussion of the model used to estimate exposure, see Reference 7.)
E.3.2 Public Exposure to Inorganic Arsenic Emissions from Low-Arsenic Primary
Copper Smelters ~~~
E.3.2.1 Source Data. Fourteen smelters are included in the analysis.
Table E.I lists the names and addresses of the plants considered, and Table
E.2 lists the plant data used as input to the Human Exposure Model (HEM).
E.3.2.2 Exposure Data. Table E.3 lists, on a pi ant-by-plant basis, the
total number of people encompassed by the exposure analysis and the total
exposure. Total exposure is the sum of the products of number of people
times the ambient air concentration to which they are exposed, as calculated
by HEM. Table E.4 sums, for the entire source category (14 plants), the
numbers of people exposed to various ambient concentrations, as calculated
by HEM.
E-14
-------
Table E-l. IDENTIFICATION OF LOW-ARSENIC PRIMARY COPPER SMELTERS
Plant Number Code
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Plant Name and Address
ASARCO, Inc.
El Paso, TX
ASARCO, Inc.
Hayden, AZ
Kennecott Corp.
Hayden, AZ
Kennecott Corp.
Hurley, NM
Kennecott Corp.
McGill, NV
Kennecott Corp.
Garfield, UT
Phelps-Dodge Corp.
Morenci, AZ
Phelps-Dodge Corp.
Douglas, AZ
Phelps-Dodge Corp.
Ajo, AZ
Phelps-Dodge Corp.
Hidalgo, NM
Copper Range Co.
White Pine, MI
Magma Copper
San Manuel , AZ
Inspiration Consolidated Copper Co.
Miami, AZ
Tennessee Chemical Co.
Copperhill, TN
E-15
-------
t-<. input uata to txposure Model Low-Arsenic Primary Copper Shelters
(Assuming Basel-ine Controls)
Plant
(Emission
Point)
1 1
2
3
4
5.
6
2 1
2
3
4
3 1
2
3
4
5
4 1
2
3
4
5
5 1
2
3
4
6 1
2
3
7 1
2
3
4
8 1
2
3
4
9 1
2
3
10 1
z
3
4
5
U 1
2
4
12 1
2
3
4
5
6
13 1
2
3
4
1* 1
2
3
4
5
Latitude
(Degrees
Minutes
Seconds )
314659
330028
330041
324156
392433
404315
330403
312048
322159
314558
464606
323658
332443
345946
Longtltude
(Degrees
Minutes
Secoods )
1063126
1104559
1104649
1080719
1144612
1121153
1092030
1093520
1125113
1063144
0893318
1103723
1105058
0842243
Arsenic
Emission
Sate
(Kg/yr) .
260
15.130
344
24.500
.390
705
1.720
29.240
1.665
1,380
430
27,520
6,450
851
95
3.4
0.9
464
108
21
394,000
46,000
4.025
450
21.130
3.870
200
520
3.555
4,820
90
18,920
19,280
4,300
4 360
T, *«wv
33.140
1.810
60
640
1.315
140
1,260
20
1,720
344
48
11.180
13
13
516
81
9
782
6,880
859
130
0
0
80
690
9
Emission | Emission
Point j Point
Elevarfjn Diameter
.(Meters) .) (Met«rsl
31 1.7
252 4.9
*0 2.7
30.5
30.5 .1.4
•«
61 3.0
30.5 —
30.5 5.2
0 -"-i
30.5 2.4
183 5.2
30.5 —
30.0 2.0
0 ™
61 3.0
61 3.1
30.5
155 7.9
0 —
229 4. 7
30.5 —
30.5 2.0
0 — •
365 7.3
30.5 —
0 .—
184 9.1
184 9.1
30.5 —
0 — •-
166 6.7
172 5.5
30.5 —
0 __
110 4.6
30.5 —
0 —
61 3.0
76 3.5
30.5
183 5.5
VBM
154 4.6
30.5
30.5 20)
Q
155 6.1
61 3.0
61 3.0
30.5 —
30.5 2.0
0
61 3.1
30.5 —
30.S 2.o
0
61 3.0
76.2 3.0
24 2.0
30.5
0
Emission*
Point
Cross
Sectional
Area (tr?)
04
3.606
no
3.720
.1,500
"— »
325
4,200
6,300
*•*
84
1,665
4,650
1.750
•»••
325
325
6,300
•1,430
_.
2,725
3,140
1,300
ww
9,140
4,800
""-*
2,200
2.200
11.200
«•
1,685
1,770
3,490
»•
846
3,720
mm-J-
325
520
3,600
2,000
H«
1.296
3,255
3.800
-*-
2.100
325
325
4.200
5.250
325
6,000
1,750
325
520
96
2,400
Emission
Point Gas
Exit
Velocity
M /« fff
14.8
19.6
7.4
0.8
20.3
— -
2.1
0.8
6.7
"~"
12.5
5.2
0.8
20.3
•~~
2.3
5.6
0.8
9.8
"» —
20.4
0.8
20.3
•"—
11.3
0.8
«~
0.24
3.8
0.8
—
10.9
4.7
0.8
— -•
4.8
0.8
— -
2.4
3.7
0.8
7.1
—
6.0
0.8
20.3
— -
8.4
6.2
6.2
0.8
19.0
7.1
0.8
10.5
9.0
10.7
2U.3
0.8
••»
Emission
Point Gas
Te«*.
f «*y \
{ ^ j
339
366
337
298
— _
352
298
339
—
352
533
298
298
...
3t>2
352
298
298
—
422
298
298
— -
339
298
-—
352
352
298
-—
505
505
298
-—
3t>2
298
— -
352
352
298
339
— •
494
296
298
_.
519
325
325
298
296
327
296
298
352
352
298
298
—
Emission
Point
Type
••-••••i-- in ••
Stack
Stack
Stack
Vent
Vent
Fugitive
Stack
Vent
Stack
Fugitive
Stack
Stack
Vent
Stack
Fugitive
Stack
Stack
Vent
Stack
Fugitive
Stack
Vent
Stack
Fugitive
Stack
Vent
Fugitive
Stack
Stack
Vent
Fugitive
Stack
Stack
Vent
Fugitive
Stack
Vent
Fugitive
Stack
Stack
Vent
Stack
Fugitive
Stack
Vent
Stack
Fugitive
Stack
Stack
Stack
Vent
Vent
Fugitive
Stack
Vent
Stack
Fugttlve
Stack
Stack
Stack
Vent
Fugitive
•Ission point to the mean wind direction for purpose of calculating
E-16
-------
Table E-3. TOTAL EXPOSURE AND NUMBER OF PEOPLE EXPOSED
(LOW-ARSENIC PRIMARY COPPER SMELTERS)*
Total Total
Number of Exposure
Plant People Exposed (People -
1
2
3
4
5
6
7
8
9
10**
11
12
13
14
435,000
8,700
8,700
19,400
4500
79,200
9,000
19,000
9,000
300
3,000
12,000
22,000
21,000
6840
580
160
11
590
62
80
290
521
<1
4
37
370
35
* A 20-kilometer radius was used for the analysis of exposure for the low-
arsenic primary copper smelters.
** EPA knows that a small town was built after 1970 within the 20-kilometer
radius of this plant. Since the population data base contained in the
exposure model is based on population locations contained in the 1970
census data, the risk and incidence estimates do not reflect the location
of this town.
E-17
-------
Table E-4 PUBLIC EXPOSURE FOR LOW-ARSENIC COPPER SMELTERS
AS PRODUCED BY THE HUMAN EXPOSURE MODEL
Concentration
Level (ng/m3)
5.81
5.00
o c
C. , -J
1 0
J. • V
0 5
V •
-------
E.4 QUANTITATIVE EXPRESSIONS OF PUBLIC CANCER RISKS FROM INORGANIC ARSENIC
EMITTED FROM LOW-ARSENIC PRIMARY COPPER SMELTERS
E.4.1 Methodology (General)
E.4.1.1 The Two Basic Types of Risk. Two basic types of risk are dealt with
in the analysis. "Aggregate risk" applies to all of the people encompassed
by the particular analysis. Aggregate risk can be related to a single
source, to all of the sources in a source category, or to all of the source
categories analyzed. Aggregate risk is expressed as incidences of cancer
among all of the people included in the analysis, after 70 years of exposure.
For statistical convenience, it i: often divided by 70 and expressed as
cancer incidences per year. "Individual risk" applies to the person or
persons estimated to live in the area of the highest ambient air concentrations
and it applies to the single source associated with this estimate as estimated
by the dispersion model. Individual risk is expressed as "maximum lifetime
risk" and reflects the probability of getting cancer if one were continuously
exposed to the estimated maximum ambient air concentration for 70 years.
E.4.1.2 The Calculation of Aggregate Risk. Aggregate risk is calculated by
multiplying the total exposure produced by HEM (for a single source, a
category of sources, or all categories of sources) by the unit risk estimate.
The product is cancer incidences among the included population after 70
years of exposure. The total exposure, as calculated by HEM, is illustrated
by the following equation:
N
Total Exposure = I (PjCi)
E-19
-------
where
I = summation over all grid points where exposure is calculated,
P-J = population associated with grid point i,
C-j = long-term average inorganic arsenic concentration at grid point i,
N = number of grid points to 2.8 kilometers and number of ED/BG
centroids between 2.8 and 20 kilometers of each source.
To more clearly represent the concept of calculating aggregate risk, a
simplified example illustrating the concept follows:
EXAMPLE
This example uses assumptions rather than actual data and uses only
three levels of exposure rather than the large number produced by HEM. The
assumed unit risk estimate is 3 x 10~3 at 1 ug/m3, and the assumed
exposures are:
ambient air number of people exposed
concentrations to given concentration
2 ug/m3 1,000
1 pg/m3 10,000
0.5 ug/m3 100,000
The probability of getting cancer if continuously exposed to the assumed
concentrations for 70 years is given by:
concentration unit risk probability of cancer
6 x 10~3
3 x 10-3
1.5 x 10-3
2
1
0.5
pg/m3
pg/m3
pg/m3
x
x
x
3 x 10-3(ug/m3H
3 x 10~3
3 x 10-3
E-20
-------
The 70 year cancer incidence among the people exposed to these concentrations
is given by:
cancer incidences
probability of cancer number of people at after 70 years
at each exposure level each exposure level of exposure
6
30
150
6
3
1.5
x lO-3
x 10-3
x 10-3
x
X
X
1,000
10,000
100,000
TOTAL = 186
The aggregate risk, or total cancer incidence, is 186 and, expressed
as cancer incidence per year, is 186 * 70, or 2.7 cancers per year. The
total cancer incidence and cancers per year apply to the total of 111,000
people assumed to be exposed to the given concentrations.
E.4.1.3 The Calculation of Individual Risk. Individual risk, expressed as
"maximum lifetime risk," is calculated by multiplying the highest concentration
to which the public is exposed, as reported by HEM, by the unit risk estimate.
The product, a probability of getting cancer, applies to the number of
people which HEM reports as being exposed to the highest listed concentration.
The concept involved is a simple proportioning from the 1 iig/m3 on which
the unit risk estimate is based to the highest listed concentration. In
other words:
maximum lifetime risk the unit risk estimate
highest concentration to = 1 ug/m3
which people are exposed
E-21
-------
E.4.2 Risks Calculated for Emissions of Inorganic Arsenic from Low-Arsenic
Primary Copper Smelters
The explained methodologies for calculating maximum lifetime risk and
cancer incidences were applied to each low-arsenic primly copper smelter,
assuming a baseline level of emissions. A baseline level of emissions means
the level of emissions after the application of controls either currently
in place or required to be in place to comply with curent State or Federal
regulations but before application of controls that would be required by a
NESHAP.
Table E-5 summarizes the calculated risks. To understand the relevance
of these numbers, one should refer to the analytical uncertainties discussed
in Section E.5 below.
E-22
-------
(Assuming Baseline Controls)
ro
co
Plant
1
2
3
4
5
6
7
8
9
10*
11
12
13
14
1 Maximum Lifetime Risk
1.1 x 10-3 . !.8 x 10-2
2.3 x 10-3 _ 3-6 x 10-2
4.3 x 10-4 . 69 x 10-3
4.1 x 10~5 - 6.6 x.10-4
4.3 x 10-3 _ 6.9 x 10-2
1.6 x 10-6 _ 2.6 x 10-5
1.5 x 10-4 . 2.4 x 10-3
9.4 x 10-4 . 1-5 x 10-2
1.5 x 10-4 . 2.5 x 10-3
1.4 x 10~6 - 2.2 x 10-5
1.5 x 10-5 . 2.4 x 10-4
3.6 x 10-5 _ 5.8 x 10-4
4.8 x 10-4 . 7.7 x 10-3
7.1 x 10-5 . !.! x 10-3
I
1 Cancer Incidences Per Year
7.1 x 10-2 . !.! x 10*
6.0 x 10-3 . 9.5 x 10-2
1.6 x 10-3 . 2.6 x 10-2
1.2 x 10-4 _ 1.9 x 10-3
6.4 x 10-3 _ i.o x 10-1
6.4 x 10-4 _ j.o x 10-2
8.2 x 10-4 . 1.3 x 10-2
3.0 x 10-3 _ 4.8 x 10-2
5.4 x 10-3 . 8.6 x 10-2
5.3 x 10-6 . 8.5 x 10-5
4.3 x 10-5 . 7.0 x iQ-4
3.8 x 10-4 _ 6.1 x 10-3
3.8 x 10-3 . 6.1 x 10-2
3.6 x 10-4 _ 5.8 x 10-3
Cancer Incidence
(one case In [x] years)
1 In 14 yrs. - 1
1 In 200 yrs. - 1
1 In 600 yrs. - 1
1 In 9000 yrs. - 1
1 1n 200 yrs. - 1
1 In 2000 yrs. - 1
1 In 1000 yrs. - 1
1 In 300 yrs. - 1
1 In 200 yrs. - 1
1 In 200.000 yrs. - 1
1 In 20,000 yrs. - 1
1 In 3000 yrs. - 1
1 In 300 yrs. - 1
1 In 3000 yrs. - 1
In 1 yrs.
L In 10 yrs.
In 40 yrs.
In 500 yrs.
In 10 yrs.
In 100 yrs.
In 80 yrs.
In 20 yrs.
In 10 yrs.
In 10,000 yrs.
In 2000 yrs.
In 200 yrs.
In 20 yrs.
In 200 yrs.
TOTALS FOR THIS SOURCE CATGEGORY
Number Total Number
of of People Exposed
Plants (within 20 km)
14 650,200
Highest Individual Risk
4.3 x 10-3 . 6.9 x 10-2
(For Plant 5)
Cancer Incidences
per year
0.1 - 1.6
one case In [x] years
1 In 10 yrs. - 1 1n 1 yrs
-------
E 5 ANALYTICAL UNCERTAINTIES APPLICABLE TO THE CALCULATIONS OF PUBLIC
HEALTH RISKS CONTAINED IN THIS APPENDIX
E.5.1 The Unit Risk Estimate
The procedure used to develop the unit risk estimate is described in
Reference 2. The model used and its application to epidemiological data
have been the subjects of substantial comment by health scientists. The
uncertainties are too complex to be summarized sensibly in this appendix.
Readers who wish to go beyond the information presented in the reference
should see the following Federal Register notices: (1) OSHA's "Supplemental
Statement of Reasons for the Fina. Rule", 48 FR 1864 (January 14, 1983);
and (2) EPA's "Water Quality Documents Availability" 45 FR 79318
(November 28, 1980).
The unit risk estimate used in this analysis applies only to lung
cancer. Other health effects are possible; these include skin cancer,
hyperkeratosis, peripheral neuropathy, growth retardation and brain
dysfunction among children, and increase in adverse birth outcomes. No
numerical expressions of risks relevant to these health effects are
included in this analysis.
E.5.2 Public Exposure
E.5.2.1 General. The basic assumptions implicit in the methodology are that
all exposure occurs at people's residences, that people stay at the same
location for 70 years, that the ambient air concentrations and the emissions
which cause these concentrations persist for 70 years, and that the concentration
are the same inside and outside the residences. From this it can be seen
that public exposure is based on a hypothetical rather than a realistic premise.
E-24
-------
It is not known whether this results in an over-estimation or an under-
estimation of public exposure.
E.5.2.2 The Public. The following are relevant to the public as dealt
with in this analysis:
1. Studies show that all people are not equally susceptible to cancer.
There is no numerical recognition of the "most susceptible" subset of the
population exposed.
2. Studies indicate that whether or not exposure to a particular
carcinogen results in cancer may be affected by the person's exposure to
other substances. The public's exposure to other substances is not
numerically considered.
3. Some members of the public included in this analysis are likely to
be exposed to inorganic arsenic in the air in the workplace, and workplace
t
air concentrations of a pollutant are customarily much higher than the
concentrations found in the ambient, or public air. Workplace exposures
are not numerically approximated.
4. Studies show that there is normally a long latent period between
exposure and the onset of lung cancer. This has not been numerically
recognized.
5. The people dealt with in the analysis are not located by actual
residences. As explained previously, they are "located" in the Bureau of
Census data for 1970 by population centroids of census districts. Further,
the locations of these centroids have not been changed to reflect the 1980
census. The effect is that the actual locations of residences with respect
to the estimated ambient air concentrations are not known and that the relative
E-25
-------
locations used in the exposure model have changed since the 1970 census.
6. Many people dealt with in this analysis are subject to exposure to
ambient air concentrations of inorganic arsenic where they travel and shop
(as in downtown areas and suburban shopping centers), where they congregate
(as in public parks, sports stadiums, and schoolyards), and where they work
outside (as mailmen, milkmen, and construction workers). These types of
exposures are not numerically dealt with.
E.5.2.3. The Ambient Air Concentrations
The following are relevant to the estimated ambient air concentrations
of inorganic arsenic used in this analysis:
1. Flat terrain was assumed in the dispersion model. Concentrations
much higher than those estimated would result if emissions impact on elevated
terrain or tall buildings near a plant.
2. The estimated concentrations do not account for the additive impact
of emissions from plants located close to one another.
3. The increase in concentrations that could result from re-entrainment
of arsenic-bearing dust from, e.g., city streets, dirt roads, and vacant
lots, is not considered.
4. Meteorological data specific to plant sites are not used in the
dispersion model. As explained, HEM uses the meteorological data from the
STAR station nearest the plant site. Site-specific meteorological data
could result in significantly different estimates, e.g., the estimates of
where the higher concentrations occur.
5. With few exceptions, the arsenic emission rates are based on
assumptions rather than on emission tests. See the BID for details on each
source.
E-26
-------
E.6 REFERENCES
1. National Academy of Sciences, "Arsenic," Committee on Medical and
Biological Effects of Environmental Pollutants, Washington, D.C., 1977.
Docket Number (OAQPS 79-8) II-A-3.
2. The Carcinogen Assessment Group's Final Risk Assessment on Arsenic.
OAQPS Docket Number 79-8-1I-A-7. May 2, 1980.
3. U.S. EPA, et.al., "Environmental Cancer and Heart and Lung Disease,"
Fifth Annual Report to Congress by the Task Force on Environmental Cancer
and Health and Lung Disease, August 1982.
4. U.S. EPA, "Health Assessment Document for Acrylonitrile," Draft Report
from the Office of Health and Assessment, EPA-600/8-82-007, November
1982.
5. U.S. EPA, "An Assessment of the Health Effects of Arsenic," Docket
Number (OAQPS 79-8) II-A-5, April 1978.
6. Busse, A.D. and Zimmerman. J.R., "User's Guide for the Climatological
Dispersion Model." (Prepared for the U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina.) Publication Number EPA-R4-73-
024. December 1973.
7. Systems Application, Inc., "Human Exposure to Atmospheric Concentrations
of Selected Chemicals." (Prepared for the U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina). Volume I, Publication
Number EPA-2/250-1, and Volume II, Publication Number EPA-1/250-2.
E-27
-------
APPENDIX F
ARSENIC DISTRIBUTION AT U.S. COPPER SMELTERS
F-l
-------
ARSENIC DISTRIBUTION AT U.S. COPPER SMELTERS
Appendix F presents information regarding the distribution of
arsenic in domestic copper smelters. Information was received in
response to requests made by EPA. However, for certain copper smelters,
the information received was incomplete. In these cases, assumptions
were made regarding the behavior of inorganic arsenic. The bases for
these assumptions are discussed for each smelter. Based on the infor-
mation provided and engineering judgment, inorganic arsenic mass balances
were developed for each copper smelter.
F.I ARSENIC DISTRIBUTION AT ASARCO-E1 PASO
An arsenic mass balance for the ASARCO-E1 Paso smelter is presented
in Figure F-l. Data were provided to EPA by ASARCO;1 however, a few
assumptions were necessary to obtain a closed mass balance. The arsenic
collection efficiency for the spray chamber and electrostatic preci-
pilator systems treating the roaster and converter offgases is 96 percent,
based on review of test data. Also, arsenic collection efficiency for
the acid plant treating the roaster and converter offgases is
99 percent. In addition to the dust collected in the roaster and
reverberatory furnace collection system, 11 Ib/hr of dust collected in
the converter collection system is recycled back to the roasters.
F.2 ARSENIC DISTRIBUTION AT ASARCO-HAYDEN
ASARCO is in the process of modernizing the Hayden plant. The
existing conventional technology consisting of roaster, reverberatory
furnace, and converter equipment is being replaced by INCO flash fur-
nace technology. The arsenic mass balance for the Hayden copper smelter
presented in Figure F-2 represents the flash furnace technology.
The data on arsenic distributions for the Hayden smelter were available onV
for the existing smelter configuration. Therefore, an arsenic balance
F-2
-------
DUST
312 LB/HR
TO STACK
0.07 LB/HR
SPEISS
106 LB/HR
SLAG TO FURNACE
23 LB/HR
Figure F-l(a). Arsenic Distribution at ASARCO-E1 Paso Smelter
-------
TO STACK |
3.2 LB/HR
| DUST TO LEAD PLANT
162 LB/HR
ARSENIC INPUT
314 LB/HR
SMELTER
BLISTER
37 LB/HR
SLAG TO DUMP
119 LB/HR
ACID PLANT WASTE
3 LB/HR
Figure F-l(b). Overall Arsenic Material Balance at
ASARCO-E1 Paso Smelter
F-4
-------
ESP
RECYCLE
OUST
333 LB/H
'708 LB/HR
ARSENIC INPUT
375 LB/HR
346 LB/HR
INCO FLASH
FURNACE
14 LB/HR
SLAG TO DUMP
219 LB/HR
TO STACK
0.27 LB/HR
ACID PLANT
.ACID SLUDGE
17 LB/HR
4 LB/HR
ESP
DUST TO
LEAD PLANT
95 LB/HR
99 LB/HR
CONVERTERS
BLISTER
44 LB/HR
F-2(a). Arsenic Distribution at
ASARCO-Hayden Smelter
F-5
-------
DUST TO LEAD PLAJT
95 LB/HR
ARSENIC INPUT
375 LB/HR
TO STACK
0.2 LB/HR
ACID PLANT EFFLUENTS
„ 17 LB/HR
SMELTER
BLISTER
44 LB/HR
SLAG TO DUMP
219 LB/HR
Figure F-2(b). Overall Arsenic Material Balance at
ASARCO-Hayden Smelter
F-6
-------
is developed for the smelter by using the arsenic input rate to
the smelter (375 lb/hr)l and the percentages of input arsenic in the
various process streams for the flash furnace smelting technology
at the Kennecott-Hurley smelter (discussed in Section F.7).2
F.3 ARSENIC DISTRIBUTION AT TENNESSEE CHEMICAL CO.-COPPERHILL
An arsenic mass balance for the Tennessee Chemical Co.-Copperhill
smelter is presented in Figure F-3. Tennessee Chemical Company provided
flowsheets and information to EPA.3 Based on actual tonnages processed
and typical analyses, the actual arsenic input to the smelter was
estimated to be 2.9 Ib/hr.
F.4 ARSENIC DISTRIBUTION AT INSPIRATION CONSOLIDATED
An arsenic material balance for the Inspiration Consolidated
Copper Company smelter located near Miami, Arizona, is presented in
Figure F-4. The data were provided to EPA by the company.^
Arsenic input to the smelter is 41.1 Ib/hr. This is based on
1,495 tpd of concentrate containing 0.033 percent arsenic fed to the
electric furnace. Nineteen percent of the arsenic into the smelter is
volatilized in the electric furnace and 6 percent is volatilized in the
converters. Also, 58 percent of the input arsenic into the smelter is
slagged in the electric furnace and 17 percent is slagged in the
converters.
F.5. ARSENIC DISTRIBUTION AT KENNECOTT-GARFIELD
An arsenic mass balance for the Kennecott-Garfield smelter is
presented in Figure F-5. Mass balance data were provided to EPA by
Kennecott Copper Company in response to information requests.2,5,6
The arsenic material balance for the smelter is obtained by
combining the arsenic rates of various smelter streams (in flow diagrams
provided by Kennecott6) with the feed and end product rates provided
by the company. The following feed and end product arsenic contents
provided by the company are used in the mass balance: total arsenic
input - 261 Ib/hr; slag tailings - 34 Ib/hr; precipitator dust to
storage - 123 Ib/hr; blister copper - 29 Ib/hr; gas cleaning plant
effluent - 71 Ib/hr; and stack emissions - 4 Ib/hr.
F-7
-------
I
CO
OUST TO WASTE TREATMEKT
0.65 LB/HR
DUCON
SCRUBBER
RECYCLED
SLURRY
0.7 LB/HR
0.5 LB/HR
3.6
,, LB/HR
ARSENIC
INPUT
2.9 LB/HR
FLUID BED
ROASTER
3.4
:ALCINE LB/HR
TO STACK
0 LB/HR
ACID PLANT
0 LB/HR
ESP
0.65 LB/HR
0.8 LB/HR
1.0 LB/HR
J.I LB/Hf
TO STACK
0 LB/HR
ACID PLANT
0 LB/HR
ESP
DUST TO WASTE TREATMENT
0.65 LB/HR
0.65 LB/HR
0.5 LB/HR
VENTURI
SCRUBBER
TO WASTE
TREATMENT
0.6 LB/HR
ELECTRIC
FURNACE
MATTE 1.5 LB/HR
1.1 LB/HR
CONVERTERS
(2)
_r\
BLISTER
0.1 LB/HR
0.3 LB/HR
SLAG TO FURNACE
SLAG TO DUMP
0.9 LB/HR
Figure F-3(a). Arsenic Distribution at TN Chemical Co -
Copperhill Smelter
-------
ARSENIC INPUT
2.9 LB/HR
TO STACK
0 LB/HR
SMELTER
GAS CLEANING
CIRCUIT WASTE
1.9 IB/HR
BLISTER
0.1 LB/HR
SLAG TO DUMP
0.9 LB/HR
Figure F-3(b). Overall Arsenic Material Balance at
TN Chemical Co.-Copperhill Smelter
F-9
-------
DUST TO FURNACE
1.2 LB/HR
DUST TO FURNACE
2.9 LB/lfli
RECYCLED DUST
4.1 LB/HR
52.9
LB/HR
ARSENIC INPUT
41.7 LB/HR
TO STACK
0.1 LB/HR
ESP
(2)
5.8 LB/HR
CYCLONES
(2)
9.1 LB/HR
9.9 LB/HR
ELECTRIC
FURNACE
MATTE 12.0 LB/HR
RECYCLED SLAG
7.7 LB/HR
SLAG TO DUMP
,-31.0 LB/HR
ACID PLANT
ACID PLANT WASTE
9.0 LB/HR
3.3 LB/HR
CONVERTERS
(4)
BLISTER
1.6 LB/HR
REVERTS
0.6 LB/HR
SLAG TO FURNACE
" 7.7 LB/HR
Figure F-4(a). Arsenic Distribution at Inspiration-Miami Smelter
F-10
-------
ARSENIC INPUT
41.7 LB/HR
TO STACK
0.1 LB/HR
•> ACID PLANT WASTE
9.0 LB/HR
SMELTER
BLISTER
1.6 LB/HR
SLAG TO DUMP
31.0 LB/HR
Figure F-4(b). Overall Arsenic Material Balance at
Inspiration-Miami Smelter
F-ll
-------
253.2
LB/HR
45.4
LB/HR
23.6
.LB/HR
68.7 LB/HR
360.8
LB7HR
38.9
.LB/HR
CYCLONES
184.3 LB/HR
229.7 LB/HR
WASTE
HEAT
BOILER
253.3 LB/HR 7.8 LB/HR
SMELTING
VESSELS
(3)
34.6 LB/HR . 42.4 LB/HR
SLAG 72.9 LB/HR
SLAG
MILL
TO STACK
4 LB/HR
ACID PLANT
ACID PLANT WASTE
75 LB/HR
ESPs (6)
DUST TO STORAGE
123 LB/HR
13.7 LB/HR
CONVERTERS
(6)
BLISTER 29 LB/HR
SLAG
0 LB/HR
SLAG TAILINGS 34 LB/HR
Figure F-5(a). Arsenic Distribution at Kennecott-Garfield Smelter
F-12
-------
STACK EMISSIONS
4 LB/HR
GAS CLEANING PLANT
EFFLUENT
71 LB/HR
INPUT 261 LB/HR
SMELTER
SLAG TAILINGS
34 LB/HR
BLISTER 29 LB/HR
PRECIPITATOR
DUST STORAGE
123 LB/HR
Figure F-5(b). Overall Arsenic Material Balance at
Kennecott-Garfield Smelter
F-13
-------
F.6 ARSENIC DISTRIBUTION AT KENNECOTT-HAYDEN
Kennecott Copper Company provided EPA with arsenic distribution
data for the Hayden smelter in March 1983.2 These data, however, did
not give a closed material balance, with 13 percent of the total input
arsenic unaccounted for. Additionally, insufficient data were provided
to calculate arsenic rates in various process streams. Therefore, only
the smelter input arsenic rate from the March 1983 data provided by
Kennecott was used in developing the arsenic mass balance data. Infor-
mation on distribution of arsenic was obtained from earlier information5*6
submitted by Kennecott. The arsenic mass balance for the Kennecott-Hayden
smelter is presented in Figure F-6.
The arsenic rates for the various smelter streams shown in the
material balance were proportionately adjusted to reflect the recent
data on feed rate indicated by the company. The difference between the
adjusted arsenic rates and the arsenic rates provided by Kennecott were
not significant for some streams.
F.7 ARSENIC DISTRIBUTION AT KENNECOTT-HURLEY
The arsenic mass data provided by Kennecott Copper Company for the
Hurley smelter are presented in Figure F-7.2 The data represent the
arsenic distribution for the flash furnace technology to be installed
at the smelter.
The arsenic input to the smelter is 2.14 Ib/hr. The data indicate
that 49 percent of the total arsenic input including the recycle material
is volatilized in the flash furnace, and 14 percent is volatilized in
the converters. Also, 31 percent of the total arsenic is slagged in
the flash furnace.
F.8. ARSENIC DISTRIBUTION AT KENNECOTT-McGILL
An arsenic mass balance for the Kennecott-McGill smelter is
presented in Figure F-8. Data for the mass balance were provided
to EPA by the Kennecott Copper Company;2 however, a few assumptions
were necessary to obtain a closed balance. An arsenic collection
efficiency of 30 percent was used for the reverberatory furnace
electrostatic precipitator. The final arsenic distribution
indicates that 21 percent of the input arsenic to the smelter is
F-14
-------
en
ACID SLUDGE
1.1 LB/HR
3.7
.B/HR
9.7 LB/HR
27.4
LB/HR
ARSENIC INPUT
17.7 LB/HR
TO1STACK
0.01 LB/HR
ACID PLANT
0.7 LB/HR
0.4 LB/HR
SCRUBBER
4.7
LB/HR
4.1 LB/HR
FLUID BED
ROASTERS
23.3
LB/HR , LB/HR
7.01B/HR
9.1 LB/HR
32.4
T
S V.
f V
6.0 LB/HR
ESP
4.4 LB/HR
11.7 LB/HR
REVERBERATORY
FURNACE
SCRUBBER
SLUDGE TO
ROASTERS
15.9 LB/HR
SLAG TO DUMP
4.8 LB/HR
6.7 LB/HR
ESP
11.1 LB/HR
CONVERTERS
BLISTER
0.2
LB/HR
SLAG TO DUMP
4.6 LB/HR
Figure F-6(a). Arsenic Distribution at Kennecott-Hayden Smelter
-------
TO MAIN STACK
7.0 LB/HR
± ACID PLANT STACK
0.01 LB/HR
ARSENIC INPUT
17.7 LB/HR
SMELTER
BLISTER
0.2 LB/HR
SLAG TO DUMP
9.4 LB/HR
Figure F-6(b). Overall Arsenic Material Balance at
Kennecott-Hayden Smelter
F-16
-------
ARSENIC
INPUT
2.14 LB/HR
ACID
t-
l-°-
STACK
01 LB/HR
FILTRATE
0.03 LB/HR
r 1
S02 IlIQUIFICATION
PLANT
RECYCLE OUST TO FURNACE 0.78 LB/HR
RECYCLE
SLUDGE
2.35 LB/HR
WET GAS
CLEANING
SYSTEM
:0.05 LB/HR
RECYCLE DUST
2.32 LB/HR
2.35 LB/HR
SETTLING
CHAMBER
RECYCLE DUST
0.44 LB/HR
2.79 LB/HR
5.68 LB/HR
INCO FLASH
FURNACE
ACID PLANT
SLAG 0.59 LB/HR
MATTE 1.73 LB/HR
SLUDGE
0.01 LB/HR
=0.01 LB/HR
ESP
(3)
0.79 LB/HR
BLISTER
CONVERTERS
(4)
0.35 LB/HR
SLAG TO DUMP
1.75 LB/HR
Figure F-7(a). Arsenic Distribution at Kennecott-Hurley Smelter
-------
'' TO STACK
0.001 LB/HR
ARSENIC INPUT
2.14 LB/HR
SMELTER
BLISTER 0.35 LB/HR
SLAG TO DUMP
1.75 LB/HR
WET SLUDGE
0.04 LB/HR
Figure F-7(b). Overall Arsenic Material Balance at
Kennecott-Hurley Smelter
F-18
-------
I
LO
165.1 LB/HR , 181.9
12.1 LB/HR
DUST
0.7 LB/HR
INPUT LB/HR
TO STACK
J00.9 LB/HR
28.2 LB/HR
ESP
40.3 LB/HR
WASTE
HEAT
BOILERS
41 LB/HR
REVERBERATORY
FURNACE
SLAG
42.7 LB/HR
DOST
3.0 LB/HR
DUST
2.8 LB/HR
MATTE
74.7 LB/HR
72.7 LB/HR
MULTICLONES
75.7 LB/HR
BALLOON
FLUE
78.5 LB/HR
CONVERTERS
FLUX
14.2 LB/HR
BLISTER
7.2 LB/HR
SLAG
TO
FURNACE
6.1 LB/HR
Figure F-8(a). Arsenic Distribution at Kennecott-McGill Smelter
-------
TO STACK
129.4 LB/HR
INPUT
179.3 LB/HR
SMELTER
BLISTER 7.2 LB/HR
SLAG TO
DUMP
6.1 LB/HR
Figure F-8(b). Overall Arsenic Material Balance at
Kennecott-McGill Smelter
F-20
-------
volatilized in the reverberatory furnace, 22 percent reports in the
slag, and 40 percent is volatilized in the converter.
F.9 ARSENIC DISTRIBUTION AT MAGMA COPPER COMPANY-SAN MANUEL
Magma Copper Company provided to EPA material balance data based
on the arsenic content of the concentrate blend expected to be used in
the future.7'8 This information, obtained through EPA information
request letters, was used to develop the arsenic mass balance for the Magma
smelter given in Figure F-9.
F.10 ARSENIC DISTRIBUTION AT PHELPS DODGE-AJO
An arsenic mass balance for the Phelps Dodge-Ajo smelter is
presented in Figure F-10. This balance is developed based on the
arsenic input rate to the smelter provided by Phelps Dodge Corporation,9
and on arsenic distributions determined for various streams from EPA's'
letter of information request to Phelps Dodge.10 This material balance
assumed a 25 percent collection efficiency for the reverberatory furnace
ESP and that the furnace offgases from the ESP are treated in the
converter acid plant. Phelps Dodge Corporation provided EPA with test
data indicating 69 percent efficiency for the reverberatory furnace
ESP. Development of a closed arsenic material balance was not possible
using the arsenic distribution data indicated in Figure F-10 and the
ESP precipitator efficiency data provided by the company. Therefore,
the efficiency provided by Phelps Dodge was not used in the material'
balance.
F.ll ARSENIC DISTRIBUTION AT PHELPS DODGE-DOUGLAS
Information currently available from the smelter** indicates that
approximately 24 Ib/hr of arsenic is input with the concentrate to the
roasters. The weight percent of arsenic in the concentrate is approximately
0.03, and the total feed to the roaster is about 1,200 tpd. The arsenic
profile for Phelps Dodge-Douglas was obtained from the smelter; however,
the information given was limited, and a complete arsenic balance was
therefore not possible.
F-21
-------
DUST TO REVERB
1.21 LB/HR '
6.14
' 16/HR
ARSENIC INPUT
4.39 LB/HR
TO STACK
2.78 LB/HR
ESP
3.99 LB/HR
REVERBERATORS
FURNACES
(3)
DUST TO REVERB
0.27 LB/HR
MATTE
1.34 LB/HR
TO STACKS
0.007 LB/HR
ACID PLANT
ACID PLANT WASTE
0.663 LB/HR
ESP
0.94 LB/HR
CONVERTERS
(6)
0.27 LB/HR
BLISTER
0.13 LB/HR
RECYCLED SLAG
TO REVERB
SLAG TO DUMP
0.81 LB/HR
Figure F-9(a). Arsenic Distribution at Magma Copper Company-
San Manuel Smelter
F-22
-------
ARSENIC INPUT
4.39 LB/HR
TO STACK
2.79 LB/HR
, ACID PLANT HASTE
0.66 LB/HR
SMELTER
BLISTER
0.13 LB/HR
SLAG TO DUMP
0.81 LB/HR
Figure F-9(b). Overall Arsenic Material Balance at
Magma Copper Company-San Manuel Smelter
F-23
-------
I
ro
DUST TO FURNACE
24.8 LB/HR
1.1 LB/HR
25.9 LB/HR
130.9
LB/HR
ARSENIC INPUT
103 LB/HR
ESP
J\
100 LB/HR
REVERBERATORY
0, SPRINKLE
FURNACE
RECYCLED SLAG
2.0 LB/HR
75.2 LB/HR
OUST TO FURNACE
9.1 LB/HR
SLAG TO DUMP
22.2 LB/HR
TO STACK
8.2 LB/HR
ACID PLANT
ACID PLANT HASTE
70.9 LB/HR
79.1 LB/HR
3.9 LB/HR
ESP
5.0 LB/HR
02 ENRICHED
CONVERTERS
2.0 LB/HR
SLAG TO FURNACE
2.0 LB/HR
Figure F-10(a). Arsenic Distribution at Phelps Dodge-Ajo Smelter
-------
TO STACK
8.2 LB/HR
ARSENIC INPUT
103 LB/HR
4 ACID PLANT WASTE
70.9 LB/HR
SMELTER
BLISTER
2.0 LB/HR
SLAG TO DUMP
22.2 LB/HR
Figure F-10(b). Overall Arsenic Material Balance at
Phelps Dodge-Ajo Smelter
F-25
-------
The arsenic distribution obtained at the Bor, Yugoslavia, smelter
is for a similar smelting configuration;^ however, there is variation
in the percent arsenic volatilized or slagged in the furnace and converter
depending upon the amount of arsenic volatilized in the multi-hearth
roaster.
To obtain the arsenic mass balance shown in Figure F-ll, it was
assumed that 17.5 percent of the arsenic is volatilized in the roaster,
since this is the average of the range obtained at the Bor.^ However,
45 percent of the arsenic was assumed to be slagged in the reverbera-
tory furance and 7 percent was assumed to be volatilized. This distribution
was used because of the high amount of arsenic reported by the smelter
to be found in the reverberatory slag. In the converter it was assumed
that 23.2 percent of the input arsenic is volatilized and 5.8 percent
reports in the slag. Only 1.5 percent of the input arsenic was assumed
to remain in the blister copper.
The collection efficiency of the hot electrostatic precipitators
was assumed to be 30 percent.
The arsenic mass balance obtained showed that about 9.4 Ib/hr of
arsenic is emitted to the atmosphere. About 13.7 Ib/hr of arsenic is
removed with the reverberatory slag, and 0.5 Ib/hr of arsenic reports
in the blister copper.
F.12 ARSENIC DISTRIBUTION AT PHELPS DODGE-HIDALGO
An arsenic mass balance for the Phelps Dodge-Hildago smelter is
presented in Figure F-12. In April 1983, Phelps Dodge provided overall
mass balance data for the smelter.9 However, the data were insufficient
to develop the material balance for each piece of process equipment.
Therefore, the arsenic balance shown in Figure F-12 was developed by
using the arsenic input rate to the smelter provided in April 1983 by
Phelps Dodge and the arsenic distribution in various streams from EPA's
information request sent to Phelps Dodge.^
F.13 ARSENIC DISTRIBUTION AT PHELPS DODGE-MORENCI
An arsenic mass balance for the Phelps Dodge-Morenci smelter is
presented in Figure F-13. The information was provided to
EPA by Phelp Dodge Corporation.9
F-26
-------
ro
/v.
0.6 LB/HR
2.1 LB/HR
1.5
LB/HR
DUST TO ROASTER
J\_
3.4 LB/HR
ESP
8.2
B/HR
ARSENIC
INPUT
24.0 LB/HR
4.9 LB/HR
MULTI-HEARTH
ROASTERS
(24)
25.1
LB/HR
23.3
LB/HR
TO STACK
4.9 LB/HR
1.5 LB/HR
ESP
\_
DUST TO ROASTER
2.1 LB/HR
REVER8ERATORY
FURNACES
(3)
J\
MATTE
9.3 LB/HR
1.8 LB/HR
SLAG TO DUMP
13.7 LB/HR
JO STACK
4.9 LB/HR
ESP
7.0 LB/HR
CONVERTERS
(4)
BLISTER
0.5 LB/HR
SLAG TO REVERB
Figure F-ll(a). Arsenic Distribution at Phel
ps Dodge-Douglas Smelter
-------
TO STACK
9.8 LB/HR
ARSENIC INPUT
24 LB/HR
SMELTER
BLISTER
0.5 LB/HR
SLAG TO DUMP
13.7 LB/HR
Figure F-ll(b). Overall Arsenic Material Balance at
Phelps Dodge-Douglas Smelter
F-28
-------
ACID PRODUCT
«
0.32 LB/HR
TO STACK
0.5 LB/HR
ACID PLANT
DUST TO FURNACE
9.2 LB/HR
ACID PLANT
LIQUID EFFLUENT
21.8 LB/HR
22.6 LB/HR
21.2 LB/HR
ESP
39.8 LB/HR
ARSENIC INPUT
30.6 LB/HR
DUST TO CONVERTERS
30.1 LB/HR
FLASH FURNACE
6.7 LB/HR
MATTE
2.8 LB/HR
0.6 LB/HR
4.8 LB/HR
1.4 LB/HR
1.4 LB/HR
ESP
2.0 LB/HR
CONVERTERS
BLISTER
1.4 LB/HR
1.4 LB/HR
Figure F-12(a). Arsenic Distribution at Phelps Dodge-Hidalgo Smelter
F-29
-------
TO STACK
0.5 LB/HR
ACID PLANT WASTE
21.8 LB/HR
ARSENIC INPUT
30.6 LB/HR
SHELTER
ACID PRODUCT 0.32 LB/HR
BLISTER 1.4 LB/HR
SLAG TO DUMP
6.7 LB/HR
Figure F-12(b). Overall Arsenic Material Balance at
Phelps Dodge-Hidalgo Smelter
F-30
-------
I
CO
DUST TO FURNACE
2.84 LB/HR
ARSENIC INPUT
9.64 LB/HR
ESP
OXYGEN
SMELTING
FURNACE
SLAG TO DUMP
8.49 LB/HR
OFF6AS TO GAS CLEANING PLANT
6.63 LB/HR
FLUX BLISTER
0.35 LB/HR 1.20 LB/HR
CONVERTERS
11.73
LB/HR
MATTE
14.69 LB/HR
SLAG
2.11 LB/HR
OUST TO FURNACE
ESP
8.21 LB/HR
TO STACK
0.30 LB/HR
ACIP PLANT
3.52 LB/HR
RECYCLE TO FURNACE
14.54 LB/HR
Figure F-13(a). Arsenic Distribution at Phelps Dodge-Morenci Smelter
-------
''TO STACK
0.30 LB/HR
ARSENIC INPUT
9.99 LB/HR
SHELTER
BLISTER
1.20 LB/HR
SLAG TO DUMP
8.49 LB/HR
Figure F-13(b). Overall Arsenic Material Balance at
Phelps Dodge-Morenci Smelter
F-32
-------
F.14 ARSENIC DISTRIBUTION AT COPPER RANGE
The Copper Range Copper Company furnished EPA with information for
the White Pine smelter;13 however, from those data 46 percent of the
arsenic input was unaccountable. The Copper Range Copper Company
subsequently advised EPA14 of corrections necessary to obtain a closed
arsenic material balance. The arsenic mass balance presented in
Figure F-14 reflects these corrections.
Data suggested by Copper Range Company as a starting point for the
material balance were:
Total arsenic input to the furnace 1.52 Ib/hr
Arsenic in flux, refining slag and
soda slag to converters 0.22 Ib/hr
Total Input 1.74 Ib/hr
Arsenic in slag to dump 0.50 Ib/hr
Arsenic in blister copper 0.43 Ib/hr
The above data indicate that 0.81 Ib/hr arsenic (i.e., the difference
between the input of 1.74 Ib/hr and the output of 0.93 Ib/hr) occurs
as emissions from the furnace and converters. The material balance
provided by Copper Range indicated an arsenic emission rate of 0.3 Ib/hr
from the reverberatory furnace, a figure which EPA deemed reasonable.
Therefore, the remaining 0.51 Ib/hr arsenic (i.e., the difference
between the total estimated arsenic emission rate of 0.81 Ib/hr and the
0.3 Ib/hr emission rate for the furnace) was assumed emitted from the
converters.
The Company-provided data indicate that 0.62 Ib/hr of arsenic is
recycled to the furnace. Thus, the recycled arsenic (i.e., the amount
captured in the waste heat boiler and the electrostatic precipitator
system) and the assumed 0.3 Ib/hr arsenic emission rate from the
reverberatory furnace is 0.92 Ib/hr. This value and other arsenic
streams to the reverberatory furnace indicate that the arsenic content
of the matte must also be 0.92 Ib/hr.
F-33
-------
DUST
0.62 LB/HR
ARSENIC INPUT , .2.14 LB/HR
1.31 LB/HR
TO STACK
0.3 LB/HR
ESP
0.92 LB/HR
REVERBERATORY
FURNACES
SLAG TO
DUMP
0.5 LB/HR
MATTE 0.92 LB/HR
TO STACK
0.3 LB/HR
BALLOON FLUE
DUST
0.21 LB/HR
0.51 LB/HR
CONVERTERS
FLUX
0.067 LB/HR
BLISTER
0.43 LB/HR
SLAG TO
FURNACE
0.2 LB/HR
REFINERY SLAG 0.1312 LB/HR
SODA SLAG 0.026 LB/HR
Figure F-14(a). Arsenic Distribution at Copper Range Company Smelter
F-34
-------
TO STACK
0.06 LB/HR
INPUT
1.53 LB/HR
SMELTER
BLISTER
0.43 LB/HR
SLAG TO DUMP
0.5 LB/HR
Figure F-14(b). Overall Arsenic Material Balance at
Copper Range Company Smelter
F-35
-------
The overall arsenic distribution shown in Figure F-14 shows
39 percent of input arsenic (including that recycled to the furnace and
that to the converters) is volatilized in the converters. Also, 21
percent of the input arsenic is slagged in the reverberatory furnace
and 8 percent is slagged in the coverters. The arsenic collection
efficiency for the waste heat boiler and electrostatic precipitator is
67 percent. This efficiency value is reasonable based on the actual
operating temperature and the control efficiency data EPA collected
during source testing at other smelters.
F-36
-------
F.15 REFERENCES
1. Letter and attachments from M.O. Varner, ASARCO, Inc., to J.R.
Farmer, U.S. Environmental Protection Agency. March 16, 1983.
Response to Section 114 information request.
2. Letter and attachments from R.A. Malone, Kennecott Minerals Company,
to J.R. Farmer, U.S. Environmental Protection Agency. March 16,
1983. Response to Section 114 information request.
5. Letter and attachments from 1.6. Pickering, Kennecott Copper
Corporation, to D.R. Goodwin, U.S. Environmental Protection Agency.
May 9, 1978. Response to information request about arsenic
distribution at Kennecott Copper Smelters.
6. Letter and attachments from R.A. Malone, Kennecott Minerals Company,
to D.R. Goodwin, U.S. Environmental Protection Agency. December 23,
1983. Response to Section 114 information request.
7. Letter and attachments from J.H. Boyd, Magma Copper Company,
to J.R. Farmer, U.S. Environmental Protection Agency. March 15,
1983. Response to Section 114 information request.
8. Letter and attachments from D.C. Ridinger, Magma Copper Company, to
D.R. Goodwin, U.S. Environmental Protection Agency. April 4, 1983.
Response to Section 114 information request.
9. Letter from L.R. Judd, Phelps Dodge Corporation to J.R. Farmer,
U.S. Environmental Protection Agency. April 7, 1983. Response to
Section 114 information request.
10. Letter and attachments from J.R. Farmer, U.S. Environmental Protection
Agency, to Phelps Dodge Corporation. 1983. Section 114 information
request.
11. Letter and attachments from R.W. Pendleton, Phelps Dodge Corporation,
to D.R. Goodwin, U.S. Environmental Protection Agency. June 2,
1983.
12. Stankovic, D. "Air Pollution Caused by Metallurgy Assemblies in
Bor." Institute for Copper, Bor, Project No. 02-513-1. U.S.
Environmental Protection Agency.
13. Letter and attachments from J.W. Maksym, Copper Range Company, to
J.R. Farmer, U.S. Environmental Protection Agency. March 17, 1983.
Response to Section 114 information request.
F-37
-------
14. Telecon. Katari, V., Pacifir environmental Services, Inc., with
J.W. Maksym, Copper Range Company. Arsenic material balance for
White Pine Copper Smelter. March 31, 1983.
F-38
-------
TREPOHT NO
EPA-450/3-83-010a
. TITLE AND SUBTITLE
'DATA
i before completing)
(3. RECIPIENT'S ACCESSION NO.'
Primary
Proposed
,,., -"-iORGANIZA I ION NAME AND ADDRESS'
Jffice of Air Quality Planning and Standards
•S. Environmental Protection Agency
^search Triangle Park, North Carolina 27711
^AGENCY
m for Air Quality Planning" ami bt<
ffice of Air, Noise, and Radiation
.S Environmental Protection Agency
.esearch Triangle Park. North Carolina 27711
SUPPLEMENTARY NOTES—~ ~~
5. REPORT DATE
July 1983
^PERFORMING ORGANIZATION CODE"
8. PERFORMING ORGANIZATION REPORT NO.
To. PROGRAM ELEMENT NO."
Tl. CONTRACT/GRANT NO.'
68-02-3060
EPA/200/04
rced fron, new and
_
DESCRIPTORS
r pollution
zardous air pollutant
llution control
andards of performance
organic arsenic
imary copper smelters
TRIBUTION STATEMENT"
limited
rm 2220-1 (R.v 4_7
4 7
KEY WORQS AND DOCUMENT ANALYSIS
b.lDENTIFIERS/OPEN
Air pollution control
Stationary sources
. .bcURITY CLASS (This ReponJ
Unclassified
SECURITY CLASS (Th» p»g*)
Unclassified
Jc. COSATi Field/Group
13 B
21 NO. OF PAGES'"
467
22. PRICED
.SOSSOLETE
-------
United States
Environmental Protection
Agency
Office of Air, Noise, and Radiation
Office of Air Quality Planning and Standards
Research Triangle Park NC 27711
Official Business
Penalty for Private Use
$300
Publication (Mo FPA 450 3 H'i Pl<
Postage and
Fees Paid
Environmental
Protection
Agency
EPA 335
If voui ddclressi'-,
------- |