EPA-450/3-83-009a
    Inorganic Arsenic Emissions from
High-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
constitute endorsement or recommendation for use. Copies of this report are available through the Library Services
Office (MD-35), U.S. Environmental Protection Agency, Research Triangle Park, North Carolina 27711; or, for a fee, from
the National Technical Information Services, 5285 Port Royal Road, Springfield, Virginia 22161.
                                    Publication No. EPA-450/3-83-009a

-------
                    ENVIRONMENTAL PROTECTION AGENCY
                    Background Information and Draft
                     Environmental Impact Statement
                 for Primary Copper Smelters Processing
      Feed Materials Containing 0.7 Percent or Greater Arsenic
                            Prepared by:
     R. Farmei
Director, Emission Standards and Engineering Division
U. S. Environmental Protection Agency
Research Triangle Park, NC  27711
                                                       (Date)
1.
3.
The proposed standards of performance would limit emissions of
inorganic arsenic from existing and new primary copper smelters
processing feed materials containing 0.7 percent or greater arsenic.
The proposed standards implement Section 112 of the Clean Air Act
and are based on the Administrator's determination of June 5, 1980
(45 FR 37886), that inorganic arsenic presents a significant risk
to human health as a result of air emissions from one or more
stationary source categories, and is therefore a hazardous air
pollutant.  Only one primary copper smelter, located in the State
of Washington, would be affected.

Copies of this document have been sent to the following Federal
Departments:  Labor, Health and Human Services, Defense,  Transportation,
Agriculture, Commerce, Interior, and Energy; the National  Science
Foundation;  the Council on Environmental Quality;  members  of the
State and Territorial  Air Pollution Program Administrators;  the
Association  of Local Air Pollution Control  Officials;  EPA  Regional
Administrators; and other interested parties.
The comment period for review of this document  is  60
Gene W. Smith may be contacted regarding the  date  of
period.
                                                         days.   Mr.
                                                         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
1.0  SUMMARY	    !_!
     1.1  Statutory Authority  	    1_1
     1.2  Regulatory Alternatives	•	    i_i
     1.3  Environmental Impacts  	    1_2
     1.4  Economic Impact	    1_4
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.  	    2-4
          2.1.3  Description of the ASARCO-Tacoma Smelter 	    2-19
     2.2  Arsenic Distribution and Emissions at ASARCO-Tacoma  .  .  .    2-21
          2.2.1  Arsenic Distribution and Process Emissions
                 at ASARCO Tacoma  .	    2-22
          2.2.2  Fugitive Arsenic Emissions at ASARCO-Tacoma.  .  .  .    2-26
     2.3  References	    2-43
3.0  CONTROL TECHNOLOGY .	    3_x
     3.1 . Alternative Control Techniques. .	      3_i
          3.1.1  Process Emission Controls	    3-1
          3.1.2  Fugitive Emission Sources and Controls 	    3-17
     3.2  Summary of Existing Control  	    3.49
          3.2.1  Process Control  Equipment at ASARCO-Tacoma ....    3-49
          3.2.2  Fugitive  Control Equipment at ASARCO-Tacoma.  .  .  .    3-51
     3.3  Performance Capabilities of Alternative Control
          Techniques  for Arsenic  and Total Particulate
          Emissions	    3.53
          3.3.1  Process Control  Systems	    3-53
          3.3.2  Fugitive  Control  Systems Evaluation	    3-68
          3.3.3  Conclusions	    3-83
     3.4  References	    3_gg

-------
                            TABLE OF CONTENTS
                               (continued)
Section
Page
4.0  MODEL PLANTS, REGULATORY BASELINE, AND REGULATORY
     ALTERNATIVES 	   4-1
     4.1  Model Plants	   4-1
     4.2  Baseline	   4-1
          4.2.1  Regulatory Considerations	   4-4
          4.2.2  Baseline Arsenic Emissions 	   4-11
     4.3  Regulatory Alternatives 	   4-13
          4.3.1  Process Emission Control Techniques	   4-13
          4.3.2  Fugitive Emission Control Techniques 	   4-14
          4.3.3  Regulatory Alternative I 	   4-16
          4.3.4  Regulatory Alternative II	   4-16
          4.3.5  Regulatory Alternative III 	   4-16
     4.5  References	   4-17
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 	   5-5
     5.3  Energy Impacts of the Regulatory Alternatives  	   5-5
     5.4  Solid Waste Impacts of the Regulatory Alternatives.  .  .  .   5-7
     5.5  Water Pollution Impacts of the Regulatory
          Alternatives	   5-8
     5.6  References	   5-10
6.0  COSTS	   6-1
     6.1  Existing Facility 	   6-1
          6.1.1  Control System	   6-1
          6.1.2  Cost Parameters	   6-2
          6.1.3  Capital and Annualized Costs 	   6-4
          6.1.4  Costs of Regulatory Alternatives  ....  	   6-7
          6.1.5  Cost-Effectiveness 	   6-7
     6.2  References	   6-9
                                   vi

-------
                             TABLE OF CONTENTS
                                (continued)
 Sect1on
 7.0  ECONOMIC IMPACT ................... .....   7-1
      7.1   Industry Economic Profile ................   7_1
           7.1.1   Introduction  ...............  ....   7-1
           7.1.2   Market Concentration ...............   7_2
           7.1.3   Total  Supply  ....  ...............   7_5
           7.1.4   U.S.  Total  Consumption of Copper .........   7_8
           7.1.5   Prices  ......................   7-12
      7.2   Economic Analysis  ..........  .  .........   7_^7
           7.2.1   Introduction  .  .  .............  ....   7-17
           7.2.2   Summary.  .  ....................   7_^7
           7.2.3   Methodology ....................   7_^g
           7.2.4   Maximum Percent Price  Increase  ..........   7_23
           7.2.5   Profit  Impacts  ..................    7-26
           7.2.6   Capital Availability ...............   7_32
      7.3   Socio-Economic Impact Assessment .............    7.34
           7.3.1   Executive Order 12291 ...............    7_34
           7.3.2   Regulatory Flexibility  ..............    7.35
      7.4   References ......  .  .................    7_3g
APPENDIX A - EVOLUTION OF THE BACKGROUND INFORMATION DOCUMENT  ...    A-l
APPENDIX B - INDEX TO ENVIRONMENTAL IMPACT CONSIDERATIONS  .....    B-l
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 .....................   Q_2Q
     C.9  Tamano  Smelter (Hibi  Kyodo Smelting Co.,) Japan .....   C-25
     C.10   Test Data (Tables)  ...................   c_2e
     C.ll   References .  .  .....................   C-109
                                   vii

-------
                            TABLE OF CONTENTS
                               (continued)

Section                                                               Page
APPENDIX D - TEST METHODS	   D-l
     D.I  Emission Measurement Methods	   D-2
     D.2  Continuous Monitoring	; . . .  .   D-7
     D.3  Performance Test Methods	   D-7
     D.4  References	   D-8
APPENDIX E - RISK ANALYSIS	   E-l
     E.I  Introduction	   E-2
          E.I.I  Overview	   E-2
          E.I.2  The Relationship of Exposure to Cancer Risk. . .  .   E-2
          E.I.3  Public Exposure	   E-5
          E.I.4  Public Cancer Risks	   E-6
     E.2  The Unit Risk Estimate for Inorganic Arsenic	   E-7
          E.2.1  The Linear No-Threshold Model for Estimation
                 of Unit Risk Based on Human Data (General)	  E-7
          E.2.2  The Unit Risk Estimate for Inorganic Arsenic . .  .   E-8
     E.3  Quantitative Expressions of Public Exposure to
          Inorganic Arsenic Emitted from High-Arsenic Primary
          Copper Smelters 	   E-10
          E.3.1  EPA's Human Exposure Model (HEM) (General) ....   E-10
          E.3.2  Pollutant Concentrations Near a Source 	   E-ll
          E.3.3  The People Living Near a Source	   E-13
          E.3.4  Exposure	   E-15
          E.3.5  Public Exposure to Inorganic Arsenic Emissions
                 from High-Arsenic Primary Copper Smelters	   E-17
     E.4  Quantitative Expressions of Public Cancer Risks
          from Inorganic Arsenic Emitted from High-Arsenic
          Primary Copper Smelters 	   E-22
          E.4.1  Methodology (General)	   E-22
                                   vm

-------
                       TABLE OF CONTENTS
                          (concluded)
     E.4.2  Risks Calculated for Emissions of Inorganic
            Arsenic from High-Arsenic Primary Copper
            Smelters	
E.5  Analytical  Uncertainties Applicable to the
     •Calculations of Public Health Risks Contained
     in this Appendix	 .
     E.5.1  The Unit Risk Estimate 	
     E.5.2  Public Exposure. .  .  	
E.6  References	
E-24

E-27
E-27
E-27
E-30
                              IX

-------
LIST OF TABLES
Table
1-1

2-1
2-2
2-3
2-4

2-5
2-6

2-7

2-8

2-9

2-10

2-11

3-1

3-2
3-3


3-4

3-5

3-6

3-7

3-8


Environmental And Economic Impacts of Regulatory
Alternatives 	
Domestic Primary Copper Smelters 	 	 	
Major Copper-Bearing Minerals 	 ......
Arsenic Input in the Feed to Domestic Copper Smelters . . .
Summary of Process Arsenic Emission Estimates for
ASARCO-Tacoma 	 . 	
Potential Sources of Fugitive Arsenic Emissions . 	
Captured Fugitive Arsenic Emissions From Calcine Discharge
From Multi-Hearth Roasters at ASARCO-Tacoma ... 	
Captured Fugitive Arsenic Emissions During Matte Tapping
From The Reverberatory Furnace at ASARCO-Tacoma . 	
Captured Fugitive Arsenic Emissions During Slag Tapping
From the Reverberatory Furnace at ASARCO-Tacoma . 	
Captured Fugitive Arsenic Emissions During Converter
Slag Return at ASARCO-Tacoma 	 	 	
Reverberatory Furnace Slag Analysis for Arsenic Content
At ASARCO-Tacoma 	 	 	
Summary of Fugitive Arsenic Emission Estimates for
ASARCO-Tacoma 	 . 	
Summary of As.Og Vapor Pressure Data and Corresponding
Arsenic Concentration at Various Temperatures .......
Arsenic Data for Hot ESP 	
Estimated Approximate Maximum Impurity Limits For
Metallurgical Off gases Used to Manufacture Sulfuric
Acid 	 	 	
Summary of Design Data for the ASARCO-Tacoma Converter
Air Curtain Secondary Hood System 	
Estimated Control Efficiencies of Existing Process
Emission Controls at ASARCO-Tacoma 	 	 	
Estimated Control Efficiencies of Existing Fugitive
Emission Controls at ASARCO-Tacoma 	 	 	
Arsenic Performance Data for the Roaster Baghouse at
ASARCO-Tacoma 	
Arsenic Performance Data for the Arsenic Plant Baghouse
At ASARCO-Tacoma 	
Page

1-3
2-2
2-3
2-5

2-25
2-27

2-30

2-32

2-33

2-35

2-40

2-42

3-2
3-6


3-13

3-43

3-51

3-52

3-54

3-56

-------
                             LIST OF TABLES
                               (continued)
Table                                                                Page

 3-9    Arsenic Performance Data for Spray Chamber/Baghouse at
        The Anaconda-Anaconda Smelter 	   3-57

 3-10   Participate Performance Data for Spray Chamber/Baghouse
        At the Anaconda-Anaconda Smelter	   3-57

 3-11   Arsenic Emissions Data at Outlet of Reverberatory Furnace
        Electrostatic Precipitator at ASARCO-Tacoma 	   3-59

 3-12   Arsenic Performance Data for Spray Chamber/Electrostatic
        Precipitator at ASARCO-E1 Paso	   3-61

 3-13   Particulate Performance Data for the Spray Chamber/
        Electrostatic Precipitator at ASARCO-E1 Paso	   3-62

 3-14   Particulate Performance Data for the Spray Chamber/
        Electrostatic Precipitator Outlet at ASARCO-E1 Paso ....   3-63

 3-15   Arsenic Performance Data for Venturi Scrubber at
        Kennecott-Hayden	   3-65

 3-16   Arsenic Performance Data for Double-Contact Acid Plant
        At ASARCO-E1 Paso	   3-67

 3-17   Arsenic Performance Data for Single-Contact Acid Plant
        At Phelps Dodge-Ajo 	   3-68

 3-18   Summary of Visible Emission Observation Data for Capture
        Systems on Fugitive Emission Sources at ASARCO-Tacoma . . .   3-70

 3-19   Air Curtain Capture Efficiencies at ASARCO-Tacoma Using
        Gas Tracer Method - January 14, 1983	   3-74

 3-20   Air Curtain Capture Efficiencies at ASARCO-Tacoma Using
        Gas Tracer Method - January 17-19, 1983 	   3-75

 3-21   Air Curtain Capture Efficiencies at ASARCO-Tacoma
        For Special Gas Tracer Injection Points - January 18-20,
        1983	   3-76

 3-22   Visible Emissions Observation Data for Converter
        Secondary Hood System During Matte Charging at the
        Tamano Smelter	   3-80

 3-23   Visible Emissions Observation Data for Blister Discharge
        At the Tamano Smelter	   3-82

 3-24   Arsenic Data for Converter Building Baghouse at
        ASARCO-E1 Paso	   3-84

 3-25   Particulate Data for Converter Building Baghouse at
        ASARCO-E1 Paso	   3-84


                                   xi

-------
                             LIST OF TABLES
                               (continued)
Table                                                                 Page
 4-1    Model Plant Parameters  	 4-2

 4-2    Summary of Existing Regulations Affecting Arsenic Air
        Emissions at ASARCO-Tacoma  	 4-10

 4-3    Summary of Effects of Regulatory Baseline on Arsenic
        Regulatory Alternatives for the ASARCO-Tacoma Smelter . . .  .4-12

 5-1    Process Emission Sources, Control Efficiencies, and
        Arsenic Emission Rates  	 5-3

 5-2    Fugitive Emission Factors, Capture and Collection
        Efficiencies Used to Determine Fugitive Arsenic
        Emissions	5-4

 5-3    Arsenic Emissions and Emission Reductions at ASARCO-Tacoma
        Under the Regulatory Alternatives for ASARCO-Tacoma 	 5-6

 5-4    Annual Energy Required by Air Pollution Control Equipment
        At ASARCO-Tacoma Under the Regulatory Alternatives  	 5-6

 5-5    Solid Wastes Generated by Air Pollution Control Equipment
        At ASARCO-Tacoma by Regulatory Alternative  	 5-9

 6-1    Design Parameters for the Air Curtain Secondary Hood
        And ESP for ASARCO-Tacoma	6-3

 6-2    Estimated Capital Costs for the Air Curtain Secondary Hood
        System at ASARCO-Tacoma 	 6-5

 6-3    Estimated Annual Operating Costs for the Air Curtain
        Secondary Hood and ESP System for ASARCO-Tacoma   	6-6

 6-4    Cost Bases Used in Estimating Annual Operating Costs of
        The Air Curtain Secondary Hood and ESP System for
        ASARCO, Tacoma  	 6-8
 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    Cost Increase and Maximum Percent Price Increase of Arsenic
        Controls for the High-Arsenic Primary Copper Smelter  .... 7-25

 7-7    Business Segment Return on Sales for Copper Companies  .... 7-28
                                   xn

-------
                             LIST OF TABLES
                               (continued)
Table                                                                Page

 7-8    Costs Increase and Maximum Percent Profit Decrease of
        Arsenic Controls for the High-Arsenic Primary Copper
        Smelter	„	7-30

 7-9    Capital Costs of Arsenic Controls for ASARCO Primary Copper
        Smelters ..... 	  7-33

 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

 C-14   Summary of Arsenic Test Data ~ R & R ESP Inlet (Reverb-North),
        ASARCO-E1 Paso Smelter 	  C-42

 C-15   Summary of Arsenic Test Data ~ R & R ESP Inlet (Reverb-South),
        ASARCO-E1 Paso Smelter 	  C-43

 C-16   Summary of Arsenic Test Data -- R & R ESP Inlet (Total),
        ASARCO-E1 Paso Smelter	C-44
                                  xm

-------
LIST OF TABLES
  (continued)
Table
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

C-32

C-33

C-34

J
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 Jest Data — Converter Building
Baghouse Outlet, ASARCO-E1 Paso Smelter 	
Summary of Particulate Test Data — Roaster/Reverberatory
ESP Outlet, ASARCO-E1 Paso Smelter 	
Summary of Arsenic Test Data — Calcine- Discharge Duct,
ASARCO-E1 Paso Smelter 	
Summary of Particulate Test Data — Calcine Discharge Duct,
ASARCO-E1 Paso Smelter 	
Summary of Arsenic Test Data — Matte Tapping Duct,
ASARCO-E1 Paso Smelter 	
Summary of Particulate Test Data — Matte Tapping Duct,
ASARCO-E1 Paso Smelter 	
Summary of Arsenic Test Data — Spray Chamber/Baghouse
Inlet-West, Anaconda-Anaconda Smelter 	
3age

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

C-59

C-60

C-61

C-62
      XIV

-------
LIST OF TABLES
  (continued)
Table
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

C-50

C-51

C-52


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 	
Summary of Arsenic Test Data ~ Converter Secondary Hood
Outlet, Phelps Dodge-Ajo Smelter 	
Summary of Particulate Test Data -- Converter Secondary Hood
Outlet, Phelps Dodge-Ajo Smelter 	
Summary of Arsenic Test Data ~ Converter Secondary Hood
Outlet, Phelps Dodge-Hidalgo Smelter 	
Page

, 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

C-78

C-79

C-80
      XV

-------
LIST OF TABLES
  (continued)
Table
C-53

C-54

C-55

C-56

C-57

C-58

C-59

C-60

C-61

C-62

C-63

C-64

C-65

C-66

C-67

C-68

C-69

C-70


Summary of Arsenic Test Data — Calcine/Roaster Fugitives
Baghouse Inlet, Phelps Dodge-Douglas Smelter 	
Summary of Arsenic Test Data — Calcine/Roaster Fugitives
Baghouse Outlet, Phelps Dodge-Douglas Smelter. . .• 	
Summary of Particulate Test Data — Calcine/Roaster Fugitives
Baghouse Inlet, Phelps Dodge-Douglas Smelter 	
Summary of Particulate Test Data -- Calcine/Roaster Fugitives
Baghouse Outlet, Phelps Dodge-Douglas Smelter 	
Summary of Arsenic Test Data — Concentrate Dryer Scrubber
Outlet, Kennecott-Magna Smelter 	
Summary of Arsenic Test Data — Acid Plant Inlet,
Kennecott-Magna Smelter 	
Summary of Arsenic Test Data — Matte Tapping Duct,
Kennecott-Magna Smelter 	
Summary of Arsenic Test Data — Slag Tapping Duct,
Kennecott-Magna Smelter 	
Summary of Arsenic Test Data -- Converter Fugitives (Full
Cycle), Kennecott-Magna Smelter 	
Summary of Arsenic Test Data — Rollout Converter Fugitives,
Kennecott-Magna Smelter 	
Summary of Arsenic Test Data — Venturi Scrubber Inlet,
Kennecott-Hayden Smelter 	
Summary of Arsenic Test Data — Venturi Scrubber Outlet,
Kennecott-Hayden Smelter 	
Summary of Arsenic Test Data — Acid Plant Outlet,
Kennecott-Hayden Smelter 	
Visible Emissions Observation Data, EPA -Method 22 — Roaster
Calcine Discharge Into Larry Cars, ASARCO-Tacoma . 	
Visible Emissions Observation Data, EPA Method 22~Matte
Tap Port and Matte Launder, ASARCO-Tacoma 	
Visible Emissions Observation Data, EPA Method 22— Matte
Discharge into Ladle, ASARCO-Tacoma 	
Visible Emissions Observation Data, EPA Method 22 — Slag
Tap Port and Slag Launder, ASARCO-Tacoma 	
Visible Emissions Observation Data, EPA Method 9— Slag
Tap and Slag Launder, ASARCO-Tacoma 	
Page

C-81

C-82

C-83

C-84

C-85

C-86

C-87

C-88

C-89

C-90

C-91

C-92

C-93

C-94

C-95

C-96

C-97

C-98
      XVI

-------
                            LIST OF TABLES
                              (concluded)
C-72


C-73


C-74


C-75


C-76


C-77



C-78


C-79


C-80


E-l


E-2


E-3


E-4


E-5
                                                               Page
Visible Emissions Observation Data3  EPA Method  22--Slag
Tapping at Slag Discharge  into Pots, ASARCO-Tacoma  	   C-99

Visible Emissions Observation Data,  EPA Method  9—Slag
Tapping at Slag Discharge  into Pots, ASARCO-Tacoma  	   C-100

Visible Emissions Observation Data,  EPA Method  22—Converter
Slag Return to Reverberatory Furnace, ASARCO-Tacoma	   C-101

Visible Emissions Observation Data,  EPA Method  9—Converter Slag
Return to Reverberatory Furnace, ASARCO-Tacoma  	   C-102

Visible Emissions Observation Data,  EPA Method  9—Blister
Discharge From Converter at the Tamano Smelter  in Japan   .  .   C-103

Summary of Average Observed Opacities for Blister Discharge
At the Tamano Smelter in Japan	C-104

Summary of EPA Method 9 Visible Emissions Data—Individual
And Total  Matte Charges to Converter Observed at the Tamano
Smelter in Japan	C-105

Summary of Visible Emissions Observation Data—Copper Blow
At the Tamano Smelter in Japan	C-106

Summary of Visible Emissions Observation Data—Slag Blow
At the Tamano Smelter in Japan	C-107

Summary of Visible Emissions Observation Data—Slag
Discharge at the Tamano Smelter in Japan 	   C-108

Identification of High-Arsenic Primary Copper
Smelters	„	E-18

Input Data to Dispersion Model  for ASARCO-Tacoma
Smelter (Baseline Control) 	 	   E-19

Total Exposure and Number of People Exposed
(High-Arsenic Primary Copper Smelters) 	   E-20

Public Exposure for High Arsenic Copper Smelters
as Produced by the Human Exposure Model	E-21
Maximum Lifetime Risk and Cancer Incidence for
High-Arsenic Primary Copper Smelters (Assuming Baseline
Controls)	E-26
                                 xvi

-------
                             LIST OF FIGURES
Figure                                                               Page
 2-1    Primary Copper Smelter 	  2-6
 2-2    Primary Copper Smelting Process	2-7
 2-3    Calcine Roaster	2-9
 2-4    Reverberatory Smelting Furnace 	  2-12
 2-5    Copper Converter 	  2-16
 2-6    ASARCO-Tacoma Arsenic Material Balance 	  2-23
 2-7    Fugitive Emission Sources at Primary Copper Smelters ....  2-28
 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	3-20
 3-4    Uses of Air Curtains	3-21
 3-5    Spring-Loaded Car Top and Ventilation Hood, ASARCO-
        Hayden	3-24
 3-6    Matte Tapping Fugitive Control System (Plan View),
        ASARCO-Tacoma	3-26
 3-7    Matte Tapping and Ladle Hoods	3-27
 3-8    Launder Cover	3-29
 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, Ohahama 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   SFg Tracer Injection Locations  	  3-72
 3-19   Tracer Injection  Test  Ports	'	3-73
 3-20   Control Device Arsenic Collection Efficiencies  	  3-86
 E-l    Tacoma Plant Configuration  	  E-14
                                  xv i i i

-------
                          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 (42 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 air pollutant.  This study examines inorganic arsenic
emission sources at primary copper smelters which process feed material
with an annual average inorganic arsenic content of 0.7 percent by
weight or greater.  This category of primary copper smelters is defined
as "high arsenic throughput smelters."  The only existing primary copper
smelter in the high arsenic throughput smelter category is owned and
operated by ASARCO, Incorporated (ASARCO) and located in Tacoma,
Washington.  No other existing smelters are expected to increase the
annual average inorganic arsenic content of their feed materials to or
above 0.7 percent and'no new smelters are projected to be built during
the next 5 years.  For this reason, only the ASARCO smelter located in
Tacoma, Washington (hereafter referred to as "ASARCO-Tacoma"), is
analyzed in this document with respect to the environmental, energy,
and economic impacts of regulating the high arsenic throughput smelter
category.  Primary copper smelters which process feed materials with an
annual average arsenic content less than 0.7 weight percent are analyzed
in the document, Arsenic Emissions from Low Arsenic Throughput Primary
Copper Smelters - Background Information for Proposed Standards (EPA-
450/3-83-010a).
1.2  REGULATORY ALTERNATIVES
     Based on consideration of the control techniques available, three
regulatory alternatives were developed for application to arsenic
emissions from the ASARCO-Tacoma smelter.
                                  1-1

-------
      Regulatory Alternative  I corresponds to the baseline level of
 control.   No  additional controls beyond the controls already in place
 at  the ASARCO-Tacoma smelter to comply with existing regulations (e.g.,
 Washington State  implementation plan, OSHA inorganic arsenic worker
 exposure  standard) would  be  required.  Therefore, no national emission
 standard  would be established for arsenic emissions from high arsenic
 throughput smelters.
      Regulatory Alternative  II represents control of fugi.tive arsenic
 emissions  from converter  operations at the ASARCO-Tacoma smelter.  This
 alternative is based on capture of the fugitive emissions using a
 secondary  hood with a horizontal air curtain.  The captured secondary
 emission  would be vented  to  a baghouse or equivalent control device for
 collection.
      Regulatory Alternative  III would require that arsenic emissions
 from  the  ASARCO-Tacoma smelter be reduced to zero.  To accomplish this
 alternative, the ASARCO-Tacoma smelter would be forced to process ores
 which were virtually free of arsenic content.  Implementation of
 Regulatory Alternative III would therefore result in closure of the
 ASARCO-Tacoma smelter.
 1.3   ENVIRONMENTAL IMPACTS
     To evaluate environmental  impacts, estimates were made of the air
 quality, water, solid waste, and energy impacts which would result from
 each  of the regulatory alternatives.  Table 1-1 summarizes the
 environmental impact assessments.
     Regulatory Alternative  I (baseline case) would not change the
 existing air and nonair quality environmental impacts of operations at
 the ASARCO-Tacoma smelter.  Total arsenic emissions from the smelter
 would remain at the current  level of 282 megagrams (Mg) (311 tons)  per
year, and  there would be  no energy or economic impacts attributable
 to this alternative.
                                  1-2

-------





co
UJ
1 — t
H-
c^
^**
C£L
UJ
1—
	 1
«=c

>-
C£
0
1—
^^
1
~"^
CJ3
UJ
CtL

U_
o
C/0
h—
O
esj|
Q-
2:
»— 1

C_3

I?
0
•z.
0
C_)
UJ

a
<
f—
~^*
UJ
s:
P^*
CD
fv^
i— i
^>
2^
LU


•
j_-l
1
T-H

— (
UJ




[ %
O) O
co re
•i- Q.
0 E
2Z P — t






cn o
£- re
ai CL
c E
UJ I— I




OJ
-M
CO
re +j
3 O

•a CL
•r- E
t~— t— i
O
CO



+J
£- o
QJ re
4_3 CL
n3 E
^g »-H






-M
0
s- re
•i- CL
•=C E
t— t





ai
>^ >
C. *r—
O +->
-f-j re
re c:
3 CD
C7> +J
0) t—

* *
* *
CD CM <3-
1 1









000








* *
* *
O I— 1 T— 1
1 +









* -X
* *
CD r-H t— I
1 +








CD O O








•X *X
•X -X
o -=3- «=t
-f- +
**- •*
c: ,
o
^3
o
re

o
c:
^ 	 ' HH
l—t h— <
1 — 1 1 — 1 1 — 1










O -(->
re o
CL re
E CL
i—i E O)
t — 1 I —
E -Q
t_ E -,-
CL> t- CO
I— O) t- -
1 1— (L)
-1-J 1 >
i- CD OJ
0 C £_
.E 0 «-
CO —I I— I
* * -X
* -X
•X



o
re 4->
a. o
E re
i— i +-> O- +J
o E o
CD re i — i re
-pi — CL CL
0 -Q E 0) E
re •!— i — t -j-i >— H
CL cr> re
E -r- •— £- 0)
i — i f— i — QJ CD
en re *o t_
o ai E o re

o r-) CM ro «*


-p
o
re
CL4->
E O
i — i re
CL
r- E
re i— <
•r~
o a>
•«— C/)

ai a>
c: >
-J
an




1-3

-------
     Regulatory Alternative II would reduce total  arsenic  emissions
from the ASARCO-Tacoma smelter by 110 Mg (121 tons)  per year  to  a  level
of 172 Mg (189 tons) per year.  The amount of collected particulate
matter containing arsenic would be approximately  11  gigagrams (Gg)
(12,000 tons) per year.  This would increase the  amount of solid waste
generated at the smelter from 182 to 193 Gg (200,000 to 213,000  tons)
per year.  This would be about a 6 percent increase.  The  additional
solid waste can be handled by the smelter's existing solid waste disposal
system.  Since the alternative is based on use of  a  dry particulate
collection device, there would be no direct water  pollution impact.
     Energy impacts for Regulatory Alternative II  would be increased
electrical energy consumption.  To operate the control  system specified
by the alternative, annual electrical energy consumption would be
14.6 million kilowatt-hours per year (kWh/yr). Total  smelter energy
consumption is approximately 2,890 million kWh/yr.  Therefore, Regulatory
Alternative II would increase the total ASARCO-Tacoma electrical energy
consumption by 0.5 percent.
1.4  ECONOMIC IMPACT
     The economic impact of implementing Regulatory  Alternatives II and
III were evaluated.  Table 1-1 summarizes the assessment of these  impacts,
     The capital cost for installing the control  system specified  by
Regulatory Alternative II is $3.5 million.  This  represents a major
capital expenditure for ASARCO.  However, ASARCO  is  a major publicly
held corporation with a good credit rating and good  access to financing.
Even considering the possibility of additional capital  expenditures for
control equipment for the two ASARCO low arsenic throughput smelters,
it is determined that the ASARCO-Tacoma smelter would be able to obtain
the necessary capital to install the control system  at  the smelter.
The annualized cost to implement Regulatory Alternative II is estimated
to be $1.4 million.  If ASARCO chooses to absorb the costs by reducing
its profit margin, the profitability of the ASARCO-Tacoma  smelter  could
be reduced to 9.6 percent.  If ASARCO chooses to  maintain  its normal
profit margin and recover the costs by increasing  copper prices, the
price of the copper would increase 0.7 to 1.0 percent.
                                  1-4

-------
     Regulatory Alternative III  would reduce total  arsenic  emissions
from the ASARCO-Tacoma smelter by 282 Mg per year to  a  level  of  zero
emissions.  As described above,  this total  elimination  of  arsenic
emissions could be achieved only through the closure  of the ASARCO-Tacoma
smelter.  Closure of the smelter would eliminate the  jobs  of  approximately
800 ASARCO employees as well as  an additional 500 jobs  in  the local
community created by the economic activity  of the smelter.  This represents
approximately 1 percent of the civilian employment  in the  county where
the smelter is located.  In addition to providing jobs,  ASARCO annually
purchases about $20 million worth of goods  and services  from  local
companies.  ASARCO pays over $2  million per year in State  and local
taxes.  These revenues to local  companies and the State and local
governments would be eliminated  by the closure of the smelter.   Finally,
the ASARCO-Tacoma smelter is the only facility in the United  States
producing arsenic trioxide and metallic arsenic for commercial sale.
Closure of the smelter would require that all arsenic raw  materials for
the arsenic chemical manufacturing industry (e.g.,  pesticides and wood
preservative manufacturers) be imported to  the United States.
                               1-5

-------

-------
                    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.
     Primary copper production in 1982 was 975,437 Mg (1,075,400 tons).2
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 106 to 6.6 x 106 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 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

ASARCO, Incorporated
Tennessee Chemical
  Company

Inspiration Consolidated
  Copper Company

Kennecott Copper Corporation
Magma Copper Company

Phelps Dodge Corporation
Copper Range Company

       TOTAL
  Location
 Annual capacity3
Megagrams    (Tons)
El Paso, Texas
Hayden, Arizona
Tacoma, Washington

Copperhil1, 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
 254,000
  71,000
  73,000
  45,000
(100,000)
(200,000)
(100,000)

 (15,000)
 136,000    (150,000)
(280,000)
 (78,000)
 (80,000)
 (50,000)
 181,000    (200,000)
64,000
115,000
163,000
191,000
52,000
(70,000)
(127,000)
(179,000)
(210,000)
(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
   Mineral
Formula
 Sulfide
 Oxide
Chalcopyrite
Bornite
Chalcocite
Covellite
Malachite
Azurite
Chrysocolla
Cuprite
Cu3FeS3
Cu2S
CuS
CuC03 Cu(OH)2
2CuC03 Cu(OH)2
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
for 85 to 95 percent of the total primary production.^  The average
tenor of these ores (copper content) is less than 1  percent.5  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.
                                  2-3

-------
     Table 2-3 presents the amount of arsenic input to the domestic
smelters, based on information received from the smelters  in  early
1983.6»7>8»9»10>1:L>12  The ASARCO-Tacoma smelter is a custom  smelter,
and is the only U.S. smelter which recovers arsenic trioxide  as  a
by-product.  Therefore, the smelter processes ore materials which
typically contain high concentrations of arsenic, obtained from  various
domestic and foreign sources.  As shown in the table, the  ASARCO-Tacoma
smelter introduces more arsenic in the feed material  to the smelter
than the combined total from all  other smelters.  The smelter also
handles substantial quantities of lead plant by-products and  flue dust
from other copper smelters.  As a result, more arsenic is  potentially
released from this smelter as process emissions and fugitive  emissions
than from the other smelters combined.
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
                                2-4

-------
               Table 2-3.   ARSENIC  INPUT  IN  THE  FEED
                      TO DOMESTIC COPPER  SMELTERS

Plant

ASARCO-Tacoma
ASARCO-Hayden
ASARCO-E1 Paso
Kennecott-Garf i el d
Kennecott-McGill
Phelps Dodge-Ajo
Inspi ration-Miami
Phelps Dodge-Hidalgo
Phelps Dodge-Douglas
Kennecott-Hayden
Phelps Dodge-Morenci
Magma-San Manuel
Tennessee Chemical
Company -Copperhil 1
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
ggic
170
142d
118e
81 f
47
189
14
11
8.0
4.5n
2.0
1.3
1.0
o.yi
feeda.b
(Ib/hr)
(2185)
(375)
(314)
(261)
(179.3)
(103)
(41.1)
(30.6)
(24)
(17.7)
(9.99)
(4.39)
(2.9)
(2.14)
(1.53)
h
 The feed  is  a  mixture  of  concentrates,  precipitates,  lead  smelter
 by-products, and smelter  reverts.

 Does not  include recycled flue  dusts  and  other  intermediates,  except
 where noted.

"50 kg/hr  (111  Ib/hr) of this  amount  is  fed  directly to the arsenic
 plant.
d
 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,
g
 0.3 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,
                                 2-5

-------
                                                                      s-
                                                                      OJ
                                                                      Q.
                                                                      Q.
                                                                      O
                                                                       (O
                                                                       S-
                                                                      o_
                                                                       S-
                                                                       3
                                                                       CD'
LU
                 2-6

-------
     EHTEBDiS THE SYSTEM
LEAVING THE SYSTEM
     Raw cnncsntratas
     Fuel
     Air
Flux and
fettling material
Fuel
Air
Siliceous  flux
Xiscsllaneous
                                       ROASTER
                                        «3
                                        CO
                                SMELTING FURNACE
material  high in copper
Air
                                    CONVERTER
Gases,  volatiIs oxides,
and dust  to dust recovery
and stack
    Gases and dust
    to waste heat boilers,
    dust recovery, and  stack
    Slag to dump
                                                          Gases  to stack
                                                          Blister copper
                                                          to  relinery
              Figure 2-2.   Primary Copper Smelting Process
                                       2-7

-------
 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 the following subsections.
     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:
               2CuFeS2
4FeS
                    02
                   702
                           2FeS
                    FeS + S
                    S02
4S02
     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 multiple-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
                                  2-8

-------
C3


it
O
                                                                                                 GO
                                                                                                 
-------
to  stirring the material, move it alternately from the center of the
hearth to the periphery where it falls to the next Jower 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
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
airstream,  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
                                  2-10

-------
 roasters due to the lower total air volume.  Average stack gas S02
 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 pyrometallurgical process in
 which solid material is melted and subjected to certain chemical
 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«Si02,
 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
 pyrometallurgical  practice.   Copper concentration in the slag increases
with increasing matte grade, which 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  rever-
beratory furnaces  within  the next  few  years  and install  Inco  flash
smelting furnaces.   In a  reverberatory furnace (Figure  2-4),  fossil
                                2-11

-------
                 UJ
                                               O

                                               (O
                                               s-
                                               zs
                                               en
                                               i
                                               oo
                                               O
                                               +J
                                               re

                                               O)
                                               JD

                                               O)
                                               >
                                               
-------
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 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
                                2-13

-------
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
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 trans-
ferred 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.
                               2-14

-------
The result is an increase in furnace  efficiency  and the  production of a
strong SOg process gas stream suitable.for  treatment  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 Fierce-Smith
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
Fierce-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 Fierce-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
smelters in similar fashion.16
     In both Fierce-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 FesO,/)., 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 Cu2S.
     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
                                2-15

-------
       C3

       X
                                            CD
                                           -P


                                            O)
                                            >
                                            c
                                            o
                                           o
                                            
-------
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
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 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
i,s 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 Fierce-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
                                   2-17

-------
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 I hour for the oxidation step,  and the remainder
for the deoxidation ("poling") step.
     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 o width 5  m (16  ft),  length
15 m (49 ft), height 3 m (10 ft), inside dimensions o capable of holding
300 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.  Wooden poles are often used for the  oxygen removal step.
                                2-18

-------
2.1.3  Description of the ASARCO-Tacoma Smelter
     The ASARCO smelter at Tac'oma, Washington, is a custom smelter
which processes copper ore concentrates, precipitates,  and smelter
by-products from numerous domestic and foreign sources.  The smelter
produces about 320 Mg (353 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 C & W multi-hearth roasters), 2 reverberatory smelting
furnaces, 3 Pierce-Smith converters (a fourth Fierce-Smith converter
is used as a holding furnace only), and 2 tilting anode furnaces.
Arsenic production facilities consist of three Godfrey  roasters,  three
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, six or seven roasters are used.  Charging is continuous.  The
calcine produced, about 45 Mg (50 tons) per hour, is intermittently
discharged from hoppers located below the roasters into larry cars for
transport to one of two reverberatory furnaces.  Typically, two 5.9-Mg
(6.5-ton) cars are charged every 15 minutes.
     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 ft) in length and 9.8 m (32 ft) 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 one car.  At full smelt, 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 flue 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
                                2-19

-------
 a  time.  The matte flows through a cast copper launder to a 4.25 m3
 (150  ft3)  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 m3 (100 ft3) cast steel slag pot.  At full smelt, about
 20 5-pot slag trains are dumped per day.  Each train takes about
 15 minutes to fill.
      Matte from the reverberatory furnace is transferred by crane to
 one of the Pierce-Smith converters.  The converters measure 4.0 m
 (13 ft) in diameter by 9.1 m (30 ft) in length.  In addition to copper
 matte, smelter reverts and cold dope materials are also processed.
 Typically, only two converters are on blow at any one time.  A converter
 cycle normally takes from 10 to 12 hours.  With dilution air, the
 offgas flow per blowing converter is about 1,130 normal m3/min (40,000 scfm)
 and contains from 3 to 4 percent S02-  Blister copper produced is
 transferred to the anode furnaces for refining and casting.  The slag
 skimmed from the molten charge is recycled to the reverberatory furnace.
     Arsenic-laden dust recovered from the particulate control systems
 servicing the multi-hearth roasters, the reverberatory furnace, and the
 converters is transferred to the arsenic recovery plant.  The dust is
 charged to one of the Godfrey roasters where the arsenic trioxide
 contained  in the dust is volatilized.  The roaster calcine or troughs,
 now low in arsenic content,  are discharged to open freight cars and
 subsequently recycled through the copper smelter complex to resorb the
 copper value.  The volatilized arsenic is passed through a series of
 arsenic settling chambers, called kitchens,  where the temperature of
the gas and vapor are controlled and the arsenic trioxide  is condensed.
The kitchens are rectangular structures  containing multiple baffled
 chambers.   There are 3 arsenic kitchens,  1 with 10 chambers and 2 with
 15  chambers.  The arsenic-laden gas  enters the first  chamber at approximately
205°C (400°F)  and exits the  final  chamber at  100°C (212°F)  or less.
The exhaust gases from the kitchens  are  treated in a  baghouse prior to
being discharged to the atmosphere.   The condensed arsenic  trioxide
                                2-20

-------
(90 to 95 percent pure) is removed periodically from the settling
kitchens and placed in storage.  This material  is  shipped as  a  product.
     ASARCO-Tacoma also operates an elemental  arsenic plant,  independent
of the copper circuit, in which purchased,  refined arsenic trioxide  is
reduced to metallic arsenic.  Offgases from the two metallic  arsenic
furnaces pass through condensers (one per furnace) in which the gaseous
arsenic is collected as a solid and periodically removed.  The  metallic
arsenic is loaded into barrels for shipment.  Offgases from the arsenic
condensers are treated in the arsenic trioxide plant baghouse.
2.2  ARSENIC DISTRIBUTION AND EMISSIONS AT  ASARCO-TACOMA
     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  (AS203) may oxidize  to the
higher oxide (As20s) which is less volatile and forms stable  nonvolatile
arsenates with other metallic oxides.
     The elimination of arsenic in the smelting furnace by slagging  is
dependent upon the 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, volatili-zation
will occur.  However, if the partial pressure  is not great enough to
allow passage through the slag layer, the arsenic  will be retained  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 stream.
     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.
                                  2-21

-------
 2.2.1  Arsenic Distribution and Process Emissions at ASARCO-Tacoma
     The arsenic distribution for this smelter is shown in Figure 2-6.
 The arsenic  input of 991 kg/hr  (2,185 Ib/hr) includes the arsenic
 found in the concentrates, precipitates, and lead smelter by-products
 and reverts, as well as the input to the metallic arsenic plant.
 Figure 2-6 is based on a detailed arsenic material balance supplied by
 ASARC0.9 Process emissions are calculated based on the control systems
 currently in place at ASARCO-Tacoma.  Fugitive emission-s are not included
 in the material balance.  Arsenic flow rates are based on the total
 tonnages transferred in a year  (1982) divided by 8,544 hours of operation,
 These rates, therefore, do not reflect instantaneous flow at full
 capacity.  It should also be noted that operations curtailment at the
 ASARCO-Tacoma smelter in 1982 was approximately 25 to 30 percent.19
     The arsenic collection efficiency of the particulate control
 equipment is significantly affected by the temperature at which arsenic
 trioxide enters and leaves the equipment.  For this reason, all assigned
 control efficiencies are dependent on the temperature at which the
 efficiencies were determined.  In the arsenic material  balance shown in
 Figure 2-6, estimated arsenic collection efficiencies for control
 devices were based on information presented in Section 3.2.1.  Based on
the results of emission source testing at the smelter,  the roaster
 baghouse was assigned an efficiency of 99.8 percent 'and electrostatic
 precipitator No. 1 was assigned an efficiency of 98 percent.  Electro-
 static precipitator No. 2 and the electrostatic precipitators upstream
 of the acid and sulfur dioxide plants were assumed to be 96 percent
 efficient based on engineering judgment.   The chemical  plants themselves
were assigned an arsenic removal efficiency of 99 percent.   An efficiency
 rating of 98 percent was assumed for the  new arsenic plant  baghouse.
     Table 2-4 presents a summary of process arsenic emissions estimated
for ASARCO-Tacoma.  These estimates are  based on  inputs  to  the various
process steps and collection efficiencies of associated  control  equipment.
     In the absence of any controls, the  major sources  of arsenic
process emissions at ASARCO-Tacoma in order of magnitude of emissions
are the reverberatory furnaces,  the arsenic plant,  the  multi-hearth
                                2-22

-------
  lea]
1









o
£ CO
'ft ^,
t— 4













u:
r™
»




























S
o


ca
"-

CM
O
o











































CM
O
CO
*•
1

»
ai

r
k


en




»<*




eo
en
en



CO



en
CM





in





I



-*
CM
U3
in



L

c
It)
u
"c
ai
O

c:
*»-
u
as
0
QC
UJ
CO
1
r—
in


y)
u
3
2
k
^
CO
UJ
0
or
£
en
i— ID
CO i—

ir


k

CM
CO
ffi
O
=3
U.
k
cn
o
CM


j k
« j»
O 0
s.

^2 S (£3
0
§ s
CO 5
£ w

-------
                                                        A) To Main Stack
        Dust Recycle
           from    (B
      Control Devices
       Recycle
                                                                       Product
                                                                       Oust to
                                                                       Lead Plant
                               To Main Stack
              Copper Ore	2074 >
                                       38
                                    ASARCO-
                                    TACOHA
                                    SHELTER
                                       463
                                Slag to Dump
17
1603
Copper Anodes

As Product
                                             64
                                                  ->• Dust to Lead Plant
Figure  2-6.   ASARCO-Tacoma Arsenic Material  Balance (Concluded)
                                      2-24

-------
O
o
<=c
o
o
C£.
=c
oo
>=c

o;
o
u_

oo
LU
00
LU
00
oo
LU


O
LU
OO
o;
-
oi
00
 I
CM
 
"c
CL) CO
4-3 03
f CQ
o
O 4-3
o x
•i- LU
C
O) 4-
co O
£_
•=t
£_
f~
•^^
JO


£_
-C
CD
(_) ^-x
C~ i ^
CD C
•r- CL1
O O
•.— £_
4- CL)
4- 0.
1 1 *- -




CD
Q.
5>^
f—





£_
^*
•^ 	
-Q
' —





£_
C"
""^,
CD







c
O CD •
•i- O
«
r—
rs
2:

,^-x
, — {
CM
•

re
LO
cn


co
cn












Q.
OO
LU
r- 1
«^f-
00
( — I
—





co
o
to

CO
OJ
o
re
£_
3
u_

b
o
J •>
03
£_
CD

t-
CD

CD
oe:
<— I O
• •
0 0
^ X

o
*a- CM
o o
CD 0
cn cn
• •
vo cn cn ^o
cn cn cn cn


4->

re c
cxr—
Q.
X!
•r- CM
• — '




•=1-
•
^1- CD
cn








CO
0)
0
f"*i re
CO C
£_ £_
CL) 3
4J n
£_
CD O)
> -a
C O
o sr
O «aC

^- 	 s
to
I— t



CO



co
cn







CD
CO
;3
O
c~
CD
re
CO
CO
CM
CO
* — '






CO
(*^
CO








c:
re
(—
QL.

O

C
O)
CO

•=C


(D
_C7
_l_j
•
E 13
O CD
£- 4->
U
"O •! 	
CD ~a
O •!-
CD
i— CD
.— 0
0 •!-
0 >

CO "O
-r~ , —
c: o
0 £_
•r— 4->
4J C
i- 0
0 0
CL
CD
re J=

A
CO CD
co c:
CD •.-
0 £-
O CD
5- 4->
Q. C
CD
CD
J= O


O) £-
> 0
re •!—
CD !-
i— Q.
J^ 40
0 Z5
•r- O
g— ^—
3 •—
re
co 4-
C"
O 4-J
•r- CO
CO 3
CO "O
•r—
E co
CD re

o E
•i- CD
c: 4-5 '
CD co
co >,
!— CO
re
CD
i — C
re -i—
•r- i —
4-3 "O
c c:
CD re
4-> -C
0
CL co
re
CD CD
-C
4-> CD
"T
4- i —
0 4-
re
-a
CD

"o
CD

.,_
x?

o
re
CD
„
CO
E
re
CD
s_
-J ^
CO
CD
4J
re
£_
re
a.
CD
CO

CD
CD
t_
C~"
j ^

O
^_>
c~
•1—

ID
CD

re
£_
re
CL,
CD
co
O)
£_
re

CO
£_
CD

tf
CD •
> C
c 3
O O
O -C
CO
CD

4-J CO
**—*
E CD
O O
£_ -t-
4- >
CD
CO "O
CD
CO I —
re o
CD £_
4- 4->
4- C
0 0
O
CO
CO CD
CD -C
O 4->
O
!_ O
CL. 4->
J3




CD
CD
£_
^"
-I-"
CD
-C
4->
E
O
g
4-

co
c
O
CO
CO
•r—
E
CD

•Q
CD
i—
^~
O
£_
4-3 •
C CO
O CO
o re
CL
4-
O CO
CD
E co
3 re
CO CD
4-
CD 4-
.C O
^j
£_
CO CD
"c tf
CD CD
CO >
CD £Z
£_ O
0. 0
CD
£_ CD

C" -f->
3
O J^
-C O
CO T-
c-
CD 3

re jr
£- CD
n
C 0
0 £-
•r- -C
CO 4->
CO
•i- CO
E E
a> ai
j ^
O CO
•r— >.
c: co
CD
CO i —
i- O
re £_
4_>
CD C
-C O
I— o
O
                                              2-25

-------
roasters, and the converters.  Process emissions from the anode furnaces
are very small compared to the other process emission sources.  Controls
are currently in place at ASARCO-Tacoma for all of the process emission
sources.  On a controlled basis, approximately 56 percent of the
arsenic emissions from the ASARCO-Tacoma smelter in 1982 were emitted
by the reverberatory furnaces, 42 percent were emitted by the arsenic
plant, 3 percent were emitted by the multi-hearth roasters, and less
than 1 percent were emitted by the converters and anode furnaces.
2.2.2  Fugitive Arsenic Emissions at ASARCO-Tacoma
     Fugitive emissions may be characterized as emissions which escape
directly 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
the process emissions are typically discharged through a tall stack.
     Listed in Table 2-5 and shown in Figure 2-7 are potential sources
of fugitive arsenic emissions.  These emissions depend upon the particular
types of equipment and operating practices employed by the copper
smelter.  To evaluate fugitive arsenic emission rates, EPA conducted
tests at smelters where potential fugitive emission sources were
controlled by high efficiency capture systems.20,22  Measurements were
made of the quantity of arsenic contained in the captured airstream.
By assuming that the quantity of captured arsenic represented the
fugitive emissions from an uncontrolled source, fugitive arsenic
emission factors were developed for roasters, smelting furnaces, and
converters.  Fugitive arsenic emission factors for anode furnaces and
miscellaneous sources were developed based on observation of the
sources and on engineering judgment.
     2.2.2.1  Roaster.
     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.
     Leakage.  Fugitive arsenic emissions from multi-hearth roasters
may be emitted from leaks that can occur at the doors located at each
                                  2-26

-------
 Table 2-5.  POTENTIAL 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
     Arsenic building (ASARCO-Tacoma smelter only)
                               2-27

-------
                                                 LU
                                                          CO
                                                          i-
                                                          O)
                                                          -P

                                                          "cu

                                                          oo
                                                          cu
                                                          Q.
                                                          Q.
                                                          O
                                                         O
                                                          (0

                                                         • I—
                                                          s_
                                                         Q_
                                                          I/I
                                                          O)
                                                          O
                                                          O)
-M
•r~
 en
•^^••HiBd

V
CD
£..
CD l*J
or *-
UJ ^-
gE
IVERBERATOI
O)
i-
3
CD
ul
                                                ce
2-28

-------
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  by  operating
the multi-hearth roasters  under a slight negative pressure  required  to
supply induced air for oxidation.
     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
Tarry 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.
     Table 2-6 presents fugitive arsenic emission test data  for  the
calcine discharge operation.   It was assumed that complete  capture of
calcine discharge fugitive emissions was achieved by  the  capture
system.  Based on the test results,  the average fugitive  arsenic
emission factor was determined to be 0.03  percent of  the  arsenic in
the calcine.
     2.2.2.2  Smelting furnace.
     Leakage.  The reverberatory furnace at ASARCO-Tacoma,  like  any
other reverberatory furnace,  is operated under a slight negative
pressure.  Since gases flow from the higher pressure  environment to  the
lower pressure environment, the pressure differential between  the gases
inside the furnace and the air outside the furnace prevents  the  gases
from escaping to the atmosphere through openings and  cracks  in the
furnace walls.  Pressure controls are used to  minimize pressure  surges
as well as to maintain negative pressures.  It is therefore  reasonable
to assume that fugitive arsenic emissions  due  to leakage  from  the
smelting furnace are negligible.
     Matte tapping.  Matte tapping is a principal fugitive  emission
source at the smelting furnace.  Smelting  furnaces have from one to
                              2-29

-------
             Table 2-6.   CAPTURED FUGITIVE ARSENIC EMISSIONS FROM
                       CALCINE DISCHARGE FROM MULTI-HEARTH
                           ROASTERS AT ASARCO-TACOMA




Sample
run
1
2
3
Avg.



Arsenic emissions,
kg/hr (Ib/hr)
0.46 (1.0)
0.76 (1.7)
1.24 (2.7)
0.82 (1.8)


Sample
time,
mm.
15
13
7
11.6
Cal ci ne
discharge
during .
test time,
Mg (tons)
41.3 (45.5)
41.3 (45.5)
35.4 (39.0)
39.3 (43.3)

Arsenic
content
of cal-
cine, %
0.7
1.22
1.22
1.05
Emission
factor, %
emitted
from arsenic
in calcine
0.04
0.03
0.03
0.03
 Tests  were  performed  only  during calcine discharge.
^During first two  tests,  seven 6-1/2-ton larry cars  were charged, and during
 the  third test, six 6-1/2-ton larry cars were charged.
'Example calculation (Sample 1):
 [0.46  kg/hr x (15 min *  60 min/hr)] •=• [(41.3 Mg calcine  x 1,000 kg/Mg x
  0.7 percent As)] x 100  =  0.04 percent  As emitted.
                                      2-30

-------
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.
     During matte tapping, the copper matte flows  through a matte
launder into a matte ladle.  The matte tap holes,  launders, and the
matte ladle are ventilated by exhaust hoods.  An  arsenic  emission
estimate for matte tapping at ASARCO-Tacoma was developed based on the
results of the arsenic emission  measurements conducted  by EPA and
presented in Table 2-7.20  It was assumed that  complete capture of
matte tapping fugitive emissions was obtained by  the  exhaust  hood systems,
The testing was performed only when actual matte  tapping  was  being
conducted.  Based on the results, the average fugitive  arsenic matte
tapping emission factor was determined to be 2.5  percent  of the arsenic
in the matte.
     Slag tapping.  Slag tapping is another principal fugitive emission
source at the smelting furnace.   Slag tap ports and slag  launders have
been observed to emit less visible fugitive emissions (visible emissions
are observed for a smaller percentage of the time it  requires to
perform a slag tapping operation) than those emitted  during matte
tapping operations.
     At the ASARCO-Tacoma smelter, the slag tap hole, slag launder,
and the slag pot are provided with exhaust hoods.   Results of tests
conducted by EPA to determine fugitive arsenic emissions  during slag
tapping are presented in Table 2-8.20  It was assumed that all  of the
fugitive arsenic emissions were captured by the exhaust hoods.  Based
on the test results, the average fugitive emission factor for slag
tapping was determined to be 0.12 percent of the arsenic  in the slag.
     Converter slag return.  Converter slag is returned to the rever-
beratory furnace through the converter slag return launder.  This is a
simple channel with an opening in the furnace wall.
                                  2-31

-------
                Table 2-7.  CAPTURED FUGITIVE ARSENIC EMISSIONS
                  DURING MATTE TAPPING FROM THE REVERBERATORY
                           FURNACE AT ASARCO-TACOMA
Sample
run



1
2
3
Avg.
Arsenic
emissions,
kg/nr(lb/hr)


1.69 (3.73)
7.53 (16.60)
7.37 (16.26)
5.53 (12.20)
Sample
time,
min.


78
75
74
76
Matte charged
during test
time.
Mg (tons)

113.9 (125.6)
113.9 (125.6)
113.9 (125.6)
113.9 (125.6)
Arsenic
content
of matte,
%

0.23
0.25
0.23
0.24
Emission
factor,
emitted
arsenic
matte
' 0.84
3.30
3.47
2.54

%
from
in





 Testing  was  performed only during actual  matte tapping.

'Example  calculation (Sample 1):

 [1.69  kg/hr  x (78 min *  60 min/hr)] -j [(113.9 Mg matte x 1,000 kg/Mg x
  0.23  percent As)] x  100 = 0.84  percent As emitted.
                                     2-32

-------
               Table 2-8.   CAPTURED FUGITIVE ARSENIC EMISSIONS
                  DURING SLAG TAPPING FROM THE REVERBERATORY
                           FURNACE AT ASARCO-TACOMA

Sampl e
run



1
2
3
Avg.
Arsenic
emissions,
kg/hrOb/hr)


0.25 (0.55)
0.65 (1.43)
0.53 (1.18)
0.48 (1.05)
Sample
time,
min.


120
111
60
97
Slag
tapped
during .
test time,
Mg (tons)
229 (252)
211 (233)
114 (126)
185 (204)
Arsenic
content
of slag,
%

0.33
0.38
0.38
0.36
Emission
factor, %

emitted
from arsenic in
slag

0.066
0.162
0.122
0.117






aTests were performed only during actual slag tapping.

bSlag is tapped in 6.3-ton ladles.  It takes about 15 minutes to fill a 5-pot
 slag train.

°Example calculations (Sample 1):

[0.25 kg/hr x (120 min * 60 min/hr)] * [(229 Mg slag x 1,000 kg/Mg x
 0.33 percent As)] x 100 = 0.066 percent As emitted.
                                       2-33

-------
     The number of times converter 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 among the slag constituents.
This tends to generate fugitive emissions through the relatively large
converter slag opening.
     Emission test results for converter slag return operations at
ASARCO-Tacoma are presented in Table 2-9.20  Since converter slag
return operations are of short duration, 1 to 3 minutes, testing
was performed for 3 days to obtain an adequate sample for analysis.
The results indicate that about 0.03 percent of the arsenic in the
converter slag is released as fugitive emissions.
     2.2.2.3  Converter.
     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 the converter.  The gate on the primary hood is retracted
to its highest position.  An overhead 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, the
blowing air is turned on while the tuyeres are above the surface of the
molten bath.  The converter is rotated until its mouth is in the topmost
position, contained within the primary hood.  The gate is then lowered
and the blowing cycle commences.
     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
                         2-34

-------
           Table 2-9.  CAPTURED FUGITIVE ARSENIC EMISSIONS
             DURING CONVERTER SLAG RETURN AT ASARCO-TACOMA
Arsenic emission rate,
          kg/hr (Ib/hr)

Sample time, minutes
0.13 (0.3)

22.92
Converter slag returned during '
          sample time, Mg  (tons) a

Arsenic content of slag return, %
103  (113)

0.187
Arsenic emission factor, % emitted
          from the arsenic content
          of slag return
0.03
 Eighteen 6.3-ton converter slag ladles were returned during the test
 time.
 Calculation:
 [0.13 kg/hr x (22.92 min * 60 min/hr)] * [(103 Mg converter slag return
    x 1,000 kg/Mg x 0.187 percent As)] x 100 = 0.03 percent As emitted.
                                2-35

-------
these openings, especially when pressure surges occur during converter
blowing, and rapid adjustments are not made by the duct damper system.
     Skimming.  During skimming operations, the mouth of the converter
is rotated to a position between 65 and 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,
these openings, especially when pressure surges occur during converter
but is necessitated by the equipment configuration.  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.
     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
pour and the buildup within the converter.  The hood gate may be  partially
extended during this operation.  Fugitive emissions occur
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.
     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 conditions such
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.
      Converter leaks.  Since the ends of most Pierce-Smith converters
are joined by bolts and springs, they occasionally leak at the end
                                  2-36

-------
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,  it may not
be repaired.  Thus,  fugitive emissions may occur at this  point.
     Fugitive Emission Rate.  Tests were performed to estimate  fugitive
arsenic emissions from a single converter at the ASARCO-Tacoma  smelter."
The tests were conducted by sampling the fugitive emission plume
directly above the converter.  Mass emission rates of arsenic in the
fugitive plume were  obtained by applying a diffusion model.   Results
obtained in two test runs were 0.39 kg/hr (0.87 Ib/hr) and 0.79  kg/hr
(1.73 Ib/hr) of fugitive arsenic mass rate.  These tests  were performed
under nonisokinetic  conditions (sampling rates during the  two runs  were
351 and 155 percent  of isokinetic, respectively).  In addition,  simul-
taneous tests were performed in the primary converter offgas duct for
process arsenic emissions in order to determine the mass  balance for
arsenic over a converter cycle.  No balance could be obtained because
substantial quantities of arsenic from the converter, about  150  kg/hr
(330 Ib/hr), were unaccounted for.  These substantial quantities of
unaccounted arsenic  indicate that larger amounts of
fugitives than those indicated by the fugitive emission test data may  be
emitted to the atmosphere.  Data from the entire test were determined
to be inconclusive and, therefore, were not used to estimate the
fugitive emission rate.
     EPA also performed testing at the ASARCO-E1 Paso smelter21  to
estimate fugitive arsenic emissions occurring during converting.
Inlet and outlet measurements were made across the baghouse  which treats
the gases from the converter building evacuation system (see
Section 3.3.2.3).  The results indicated a fugitive arsenic  emission
factor for converters of 15 percent of the arsenic contained in  the
converter primary offgas.  This factor was used for estimating  converter
fugitive emissions at ASARCO-Tacoma.
     2.2.2.4  Anode  furnace.  Refining of blister copper to  anode
copper at ASARCO-Tacoma is performed in hearth-type furnaces.  The
process offgases generated during the actual refining (oxidation and
reduction blows) are siphoned through a furnace offtake and  ducted  to
a cold electrostatic precipitator.  Some fugitive emissions  are  emitted
during charging, skimming, poling, and casting operations.
                                 ' 2-37

-------
      No test data are available for determining the amount of fugitive
 arsenic emissions from anode furnaces.  Therefore, an approach analogous
 to the one used to estimate converter fugitive emissions was selected.
 The anode furnace fugitive emission estimate for the ASARCO-Tacoma
 smelter is based on the amount of arsenic present in the anode furnace
 process offgases as presented in the arsenic material balance (Figure 2-6)
 Using engineering judgment, an emission factor of 20 percent of the
 arsenic contained in the anode furnace process offgas was  assumed for
 estimating the anode furnace fugitive emissions at ASARCO-Tacoma.
      2.2.2.5  Miscellaneous and fugitive emission sources.
      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.  For the purpose of developing an arsenic emission
 factor for  these operations, 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
                                  2-38

-------
 on  participate  (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.23
     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 emissions are discharged
 from ladles once they are filled.
     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 transportation 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 by  EPA at  the ASARCO-Tacoma  smelter to
 determine the magnitude of fugitive arsenic emissions during  slag
        Of)
 dumping.^0  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 2-10.
                               2-39

-------
                Table 2-10.  REVERBERATORY FURNACE SLAG
              ANALYSIS FOR ARSENIC CONTENT AT ASARCO-TACOMA
   Sample Run

       1
       2
       3
       4
    Avg.
Percent arsenic in slag at
exit from furnace slag
       launder

        0.33
        0.38
        0.46
        0.49
        0.42
Percent arsenic in
slag at dump site

       0.40
       0.52
       0.44
       0.29
       0.41
      Test results indicated that in 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 can  be considered negligible.
     Arsenic Building.  The ASARCO-Tacoma smelter is  the  only domestic
 copper smelter that  operates an  arsenic  plant and currently manufactures
 arsenic for  commercial use.  Dust collected  in the  flues,  baghouse,
 and electrostatic precipitators  serving the  roasters and the reverberatory
 furnaces is  conveyed to one of the six Godfrey roasters in the arsenic
 building.  In these roasters, the arsenic-containing dust is heated to
 vaporize the arsenic trioxide.  The vapors then pass through a baffled
 brick chamber, or "kitchen," where they cool, condense, and drop out
 in a granular form.  The chambers are opened  periodically, and the
condensed arsenic trioxide is removed manually.   Fugitive arsenic
escapes during this cleaning period.   Fugitive arsenic emission  estimates
for the arsenic  building at the ASARCO-Tacoma smelter were based on
information obtained from  the  Puget  Sound Air Pollution Control  Agency
                                  2-40

-------
(PSAPCA).24  PSAPCA estimates indicate that the major source of  fugitive
arsenic emissions in the arsenic building are the arsenic  kitchens.
They estimated that 0.5 percent of the arsenic when  pulled from  the
arsenic kitchens becomes airborne.  This emission factor was therefore
used to estimate the fugitive arsenic emissions from the arsenic
building.
     2.2.2.6  Summary of fugitive arsenic emission estimates.
Table 2-11 presents a summary of fugitive arsenic emissions estimated
for ASARCO-Tacoma.  These estimates are based on emission  factors
described in the previous sections and arsenic inputs to various
process steps presented in Figure 2-6.  It is estimated that 90  percent
of the total fugitive arsenic emissions at the ASARCO-Tacoma smelter is
emitted by the converters.
                                2-41

-------







eC
O
<
1 —
1
o
o
or
f?





o
u_

oo
LU
•=C
sz

oo
LU

o
00
OO
>— <
s:
LU

O
fc—C
2T
LU
oo
OS
LU
>
P
CD
£

o

c:
« ' '
oo



t— i
t— <
i
CM
CL)
JD
re









j

V
V
£


*z
QJ
U
C
>
(J
c
Q

L
1^
»*-
UJ
"c
S-
c
c





1
c
o
o



0)
10
t-
J
t/1
s

c
o
I




c
QJ
c
o
CJ
u


I/)
s_








~
JZ
JD

oT
> 4-> j_
ro jz
Cn

•—
"lO
QJ
O

i


U
QJ

»""
U

QJ
I
0.
fO
u



I
¥
LU


•£•
u- •—
0 *-^
JS
QJ *—
U P— — •
5 U
t/l 4-1 L.
JD c j=
c "5*

-"S 3 0
t- 4-> &.•«-
32£ g
0 g £


S-
.1
U
O JS
«E cn



^
c
QJ
1







5 at
*- u
/I L.
I1
~
o
0^



o
o


CD
S



CO







§





QJ
in
|
cn
to




vo
0


0

o
0



cn
CM
0
csT
0




Calcine


j->
QJ 03

i/l J=
C 1
IO -p-
L. 4->
4-> ,— W
3 W
OJ E QJ
.SH"
^£g
re >*— i-
(j

r-*
^



IT)
0


^>
CO











o






O-
to
UJ




^"
CO


CO

LO
CM



CM
-^_ •
CO




QJ







CL
(O
QJ
03
CO
O
0



o
o


«r
CO





•JD
cn




s






Q.
to
UJ




vo
CD


in
CM
CD

CM
0



PO
^-*
0
CM




QJ
U
C
»o cn
c ro
S- f—
3 W







CL
03
4-)
to
CM
O
0



o
o



1





1





1






QJ
c
0
c



^^
CM
o
o


0

CO
o
o*



CM
O
*•— •*
VO



s»
QJ
*J
S-
QJ en
> ID
J1^




cn
(U

Converter s
return

cn
o
C



s



1











1






c
o
c



<>-^
en
o



v-4

tn



0
*•" •*
cn




I/I. VI
in to
QJ cn
u u-
0 »*-

c
o
ro
i-
QJ
CL
Converter o

CM
0


CO
0
0



1











1






QJ
C
o
c



ff^
CM
O


CO
o
0


o
CM



Ct

O




Q) QJ
Ul Q.
•p- O
5°




in
QJ
U
QJ
-o
O
c

10
o



ro
o



o
cn





03





ro






(O



*— *
cn
LD


Csl


o



CM
LO
V£>
CM


* 'O
•T3 O)
1 -O QJ >,
1/1 QJ r— Qj
C i- -o >
+•» re i- c t3 e
cn t. QJ to c o
3 4J **- JZ (0 (J
fc
Cn

T3 —
c cn
(/> ro c
3 JT •--
O 4J >,
QJ (/) M QJ
E 3 S- >
ro "D QJ c
r— I*- O
r— 0> t/> U
QJ 3 E
o •— ro -o

•g-
_^
^"^


in
o



o
CD











JO






.0




^
s
^"^

UD

Lf)
O



CO
CM
CO
CM
CM
CO
CM
,-T


o
OJ C 4-3
cn QJ c
i- V) to
to i- P—



cn



I
c
OJ
<







to
S
l/l
i-
C
ro
I-
"^





4J
in
•c .
irt
^- QJ
0 0
r>l 4->
C U
ro TO
QJ t-
E 0.
o
•0 3
QJ
p- 'o
O L.
1- 4->
§ 1
0
I 1
U
QJ O
Q.
O C
OJ
QJ £
JD
O J3
4-J t—
to o
(rt O
£ "c
(0 Q)
U
VI S-
E QJ
O CL
trt O
^2 °^
i 5
Cn o
C 4->
8? "S
> 4-1
E ro
O S
0 -r=
•o wi
C QJ

C
•— irt
•a o
E H—
(O l/l
*-
t. S
OJ
S J*



3
3 U
OJ QJ
2-42

-------
2.3 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., G., 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 7,
     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 16,
     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
     Agency.  Research Triangle Park, North Carolina.  EPA Contract
     No. 68-02-2606. June 1978.-

                                 2-43

-------
 14.  Reference 1, p. 3-11, 3-12.

 15.  Reference 1, p. 3-18.

 16.  Telecon.  Whaley, G., Pacific Environmental  Services,  with  larsen, T.,
     Inspiration Consolidated Copper Company.   Status  of Hoboken
     converters.  March 4, 1983.

 17.  Reference 1, p. 3-36.

 18.  Mackey, P.J., et al.  Minor Elements in the  Noranda Process.
     (Presented at the 104th Annual  AIME  Meeting.  New York.   February
     16-20, 1975.)  •

 19.  Telecon.  Whaley, 6., Pacific Environmental  Services,  with  White, T.,
     ASARCO, Inc.  April 8, 1983.  Arsenic  material  balance for
     ASARCO-Tacoma.

 20.  TRW Environmental Engineering Division.  Emission Testing of
     ASARCO Copper Smelter, Tacoma,  Washington.   EMB Report No.  78-CUS-12.
     April 1979.

 21.  TRW Environmental Engineering Division.  Emission Testing of
     ASARCO Copper Smelter, El  Paso,  Texas.  EMB  Report  No. 78-CUS-7.
     April 25,  1978.

22.  TRW Environmental  Engineering Division.  Emissions  Testing  of
     ASARCO Copper Smelter, Tacoma,  Washington.   EPA Contract No. 68-02-2812,
     Work Assignment No. 45.  August  22,  1979.

23.  Newton, J.  and C.L. Wilson.   Metallurgy of Copper.  London.  John
     Wiley and  Sons, Inc.   1942.

24.  Correspondence from Roberts,  J., Puget  Sound Air  Pollution Control
     Agency, to  Pacific  Environmental Services, Incorporated.   March
     12, 1978.
                                2-44

-------
                        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 Controls
     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.   Table 3-1 presents
vapor pressure data for arsenolite (As.Og), the more common form of
arsenic trioxide and the most abundant arsenic compound in smelter
         2 "3
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,
nm 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
                                               30
               2.0
                                               20
               1.0
                                               10.
              0.5
                                              5.0
       o
       ^
       as
       o
       u
              0.1
                                              1.0
U4
as,
CL,

OS
o
a.
              0,01
                100
                                             0.1
                                           500
                      TEMPERATURE, '
Figure 3-1.  Arsenic Trioxide Vapor Pressure and Saturated
           Vapor  Concentration with Temperature

                              3-3

-------
Furthentiore, the vapor  pressure data  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
                                                                     3
currently used in the industry, 9.25 grams  of  arsenic  trioxide  per m
           3              3
of gas  (g/m ) (4.04 gr/ft ) 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
            3
(0.019  gr/ft ) of arsenic trioxide as vapor.   Any concentration higher
than 9.25 g/m  in a gas stream at  200°C would  lead  to  condensation of
some of the vapor, and  the As.Og 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.
                                                                 3
     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
                                    3
concentration for As»0g is 0.315 g/m .  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/m3 (0.109  gr/ft3).
Since the arsenic saturation concentration  shown  in Table 3-1 for this
temperature is 0.043 g/m  , the amount of arsenic  leaving the spray
                                            3
chamber in condensed form would be 0.207 g/m   (33.12 kg/hr), and the
remaining amount, 0.043 g/m , 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 be collected.   The  overall  control
efficiency for arsenic at the lowered temperature would  thus  be
31.80/40 = 80 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.
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  preceding as an integral
part of the overall control  system.  These include the application of
baghouses, high voltage electrostatic precipitators, or high energy

                                3-5

-------
a.
co
D;
CD
UJ
co
a;

CU
'c
1— 1













co
£=
o
to
to
i










.£=
_Q"
^^
$-
C~
^•^
en


o
to
S-
cn


co


"en

• LL.
0. o
1— O
o







to
c
o
to
to
r—
•i







^
-C
-Q
£3.
i_
C7*

cn



*-~— •»
14-
o
to
s-
cn
CO
E

en

« tZ"
Q. 0
H- CJ
o

O)
r— •
Q. £=
E— ^
3
 CD CO
• • • •
cn co CM r-»
CO CO CM

VO ID r-l CO
CO •* r-I CT)
LO «d- LO «*
CO CM CM UD
• • • •
«d- O CO CM
CM CM CM CM


CM cri r~- to
CT> t^. CO CO
O O CD CD
CD CD O CD
*^-^ *» * •*~^* ^~*

r-l r-l CT1 l~»
r-l CO CTl CT1
CM r-l r-l T— 1
• • • •
CD O O CD
LO O CD LO
cn r-< oo cri
LO l£> LO LO
CO r-l «^- CO
r-l CM CD r-l
CO CO CO CO

r-H r~. CM CD
o-i CM LO cn
LO i-- r~- <£>
CO CD r-l CO
tO CO •* r-l
CM CO CO CO



X- —» *^^ ^^-^ ^^"^
r-. CM to LO
r— 1 r— 1 r— 1 i— 1
O O CD CD

co LO *^" cr»
to CM CO CD
CM CO CO CO
O CD CD CD
CM CM "crT <— i
CM O CO CM
to to to to
co r^- to r^
CM r-1 CO CM
CO CO CO CO

cn
i— 1 CM CO >
"^

                                                 3-6

-------
venturi scrubbers, preceded by gas 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 dilution  air  needed  to  effect  cooling
could result in a two- to four-fold  increase  in the  total gas  volume
to be treated, with a corresponding  increase  in the  size  and  cost of
the 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 rn2/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
and 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 chambers are used for cooling prior to
electrostatic precipitators and baghouses used for particulate matter
control.
     3.1.1.3  Baghouses (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, inertia! 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 H90).   The filtering velocity, or air-to-cloth
                                           32                2
ratio, generally ranges from 0.30 to 0.61 m /min per m  (1 to 2 acfm/ft )
for conventional mechanical shaker cleaning type baghouses when applied
to metallurgical fume.  Pulse jet cleaning type units generally operate
                                                        3          2
at higher air-to-cloth ratios, ranging from 1.8 to 3.0 m /min per m
(6 to 10 acfm/ft2).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.
                                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
matter 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.   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 participate removal prior to acid manufacturing.  With either
type, a complete precipitator 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 precipitator 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 SO.,.  Too high a resistivity (greater than
  10
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 entrainment 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. H^O).  Typical throat velocities for venturi units range from
75 to 100 m/s (15,000 to 20,000 fpm) and liquid-to-gas ratios  range
                              O                    O       Q
from 0.4 to 2 liters/min per m /min (3 to 15 gpm/10  acfm).    It is
                                3-11

-------
important that the liquid-to-gas ratio 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 few 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 S02 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 difficulties 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
v/here 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 USEDQTO MANUFACTURE
                            SULFURIC ACIDy

Impurity limits at
Substance inlet to acid plant,
(mg/Nm3 )a
Chlorides, as Cl
Fluorides, as F
Arsenic, as As^O^
Lead, as Pb
Mercury, as Hg
Selenium, as Se
Total solids
H2S04 mist, as 100% acid
Water
1.2
0.25
1.2e
1.2
0.25
50e
I-2
50
Impurity limits at
inlet to- gas puri-
fication system,
(mg/Nm3)a'b
125C
25d
200
200
2.5f
100
1,0009
400 x 103
aBasis:  dry offgas stream containing 7 percent sulfur dioxide.

 For a typical  gas purification system with prior coarse dust removal.
                           «5
cMust be reduced to 6 mg/Nm  if stainless steel is used.
V                             q
 Can be increased to 500 mg/Nm  if silica products in scrubbing towers
 are replaced by carbon; must be reduced if stainless steel is used.

eCan be objectionable in product acid.
f                                 3
 Can be increased to 5 to 12 mg/Nm  if lead ducts and precipitator
 bottoms are not used.
                                          O
9Can be increased to 5,000 to 10,000 mg/Nm  if weak acid settling tanks
 are used.
                                3-13

-------
mist precipitators where acid mist,  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 (SO^) 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

-------
      cs
      LU
      Q-
      V5
      o
      UJ
      0£
      UJ
                                  f
                                  T
                                             LU
                  f
                                           §
o
o
or
a.
o
o
                              en
                              LU
                                          O
                                          0
OS
LU
                                              u
                                                            010V
                                         z^
                                         iu.1
                                         >-3t
                                         cso
                                                                                                                    (C

                                                                                                                    O.

                                                                                                                    -o
                                                                                                                    •I—
                                                                                                                    o
                                                                                                                    
           o
           C3
LU
03
                                                             C3
                                                       o    o •*
                                                                                               a

                                                                                               O
                                                                                                           LU
                                                          3-15

-------
 Regardless  of  the  control  device employed,  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  (ASgO^),  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 SO-  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 SCL 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.11
     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 (212°F) to
110°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
saturation concentration, arsenic emission reductions approaching the
overall performance capability of the control device for particulate
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  Fugitive Emission Sources and Controls
     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.
     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  practices.   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
or 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  ventilation  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
J                ~  ~ -  - -  -          11 J.in in
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

-------
      c)
      e)
          SOURCE
                        HOOD
                      HOOD
                     SOURCE
                       FLOOR
              FAN (HOOD)
                                 b)
                             \\\\\\\r
                                                      HOOD
TLOOR
                                    AREA SOURCE
                                                   HOOD
                                                  FLOOR OR
                                           SOURCE BENCH
           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
Cc)
NOTE:
          TO DUST
          COLLECTOR
    Figure 3-4.  Uses of Air Curtains13'14 '

(a) for the control of hazardous fumes from vinyl
chloride processes, (b)  to control dust from a
primary crusher operation., and (c) to control dust
from loadout chutes.
                      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 fugitive emissions from the  converter
building evacuation system.  Phe'lps 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-El 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 (Tarry 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  Tarry 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 Tarry car.  The
other  hoTes  are connected  to vent Tines which  go to vent  hoods with
their own  individuaT draft fans (a singTe  fan  is used at  ASARCO-Tacoma).

                                3-23

-------
LU

_1

LU
                                «t
                                to
                                OD



                                 I
                                o
                                o
                                CL.
                                        o
                                        cr:
                                                         0)

                                                         i
                                                         o
                                                         o
                                                         Qi
                                                         
-------
Each draft fan has a capacity of approximately  142  Nm  /min  (5,000 scfm).
The captured fugitives are then combined with the roaster process
gases and treated in a baghouse.
     In addition to the local hooding  and  ventilation  applied  directly
at the calcine hopper discharge point, at  ASARCO-Tacoma, the calcine
hopper area has been enclosed to form  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
operations.  During visual observations, no  fugitive emissions  were
observed escaping from the tunnel-like enclosure.
     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
matte tap ports.  Normally, only one matte tap  port is  in use  at  a
time.
     Copper matte from the furnace port travels through a launder
which directs the flowing matte to a point where it can be collected
in a large ladle.  Emissions are observable  from the point of  the
matte leaving the furnace to the point where it settles in the  ladle.
     Matte tap ports and launders in most  smelting  furnaces have
hooding 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,
     Schematics 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-section 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) diameter main duct are approximately  0.6 m (2 ft) in diameter.
During a tap, 283 Nm3/min (10,000 scfm) is exhausted at the tap hole
                                3-25

-------
                                               c
                                               It)
                                               O)
                                               O (C
                                              O E

                                               O> O
                                              -p  I
                                              •r- O
                                               O1O
                                              Q.
                                              CL
                                              (B
                                              O)
                                               I
                                              CO
                                              01
                             O

                             
-------
                                 O
                                 O
                                 LU



                                 
-------
hood and 850 Mm /min  (30,000 scfm) 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)
range in size from 0.6 to 1.5 m (2 to  5 ft) in length and have  a
semicircular cross-section.
     A 3.4 m (11 ft)  diameter retractable  ladle  hood  is used to capture
emissions generated at the ladle.  The ladle hood is  lowered over the
ladle prior to tapping and is raised after the'tap  is completed.
     From testing and observations made at the ASARCO-Tacoma smelter,
the matte tapping fugitive emission capture system  was  observed to
                                                    15
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
                                           o
smelters range in capacity from 2.8 to 17  m  (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
are 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/min (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
                                         2
with fixed hoods.   During tapping, 142 Mm  /min (5,000 scfm) is  applied
at the tap hole and 566 Mm /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

-------
        UJ
                           
-------
00 00

o o

i^ ^
O 0
>—I f—<
1C 3:
   co

                                                                                           a;
                                                                                           +j
                                                                                           CO
                                                                                           C (O
                                                                                           O £
                                                                                           o o
                                                                                              o
                                                                                            t—
                                                                                           •r-  I
                                                                                           4-> O
                                                                                           •r- C_3
                                                                                           O1 O£
                                                                                           =5 «C
                                                                                           u_ oo
                                                                                           o.
                                                                                           a.
                                                                                           «3
                                                                                           CD
                                                                                           (O
                                                                                          cn

                                                                                          cr>
                                                                                           2
                                                                                           cn
                                                                                          •r~
                                                                                          U-
                                       3-30

-------
     3.1.2.7  Converter Operations.  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 modes, such as charging, skimming, holding, and
pouring.  The flow rates handled by these hoods range from 700 to
2,400 Nm3/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
   4.  I 16
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  configuration  of  the fixed  type is  shown in
Figure 3-10.  These  hoods  are  approximately  3 m  (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
                                                      18
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
steel 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

-------
         T.O.  RAIL
         SMOKE
          HOOD
         PLENUM
    FIXED
  SECONDARY
    HOOD
                                TO SECONDARY
                                  HOODING
                                 MAIN DUCT
Figure 3-10.  Typical Converter Fixed Secondary Hood
                         3-33

-------
                                                      SECONDARY HOOD  DUCT

                                                        SMOKE PLENUM   .
                                                        SECONDARY HOOD DUCT

                                                      SECONDARY HOOD  DUST  BIN
EOT RUNWAY
         .xDUST BIN
          MAIN HOOD
      SECONDARY
      HOOD DUCT
                                                            MOVABLE HOOD
    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 operate 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.16
     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
                                               20
to be 40, 85, 70, and 85 percent, respectively.    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  Air 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
                 1A 91 99 91                              9&
being used abroad1^1'^ '" 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

-------
           jrr  SIDE
                       EXHAUST SIDE
  AIR
CURTAIN
  JET
            BAFFLE
             WALL
                                 AIR CURTAIN
                    m
                    ^
                           FUGITIVE
                           EMISSIONS
                               y           /  /   i
                     f
                       I
xx     7/
                                 CONVERTER
                               (FUME  SOURCE)
                          BAFFLE
                           WALL
                                                                    TO SUCTION FAN
           Figure 3-12.   Converter  Air Curtain  Control  System
                                     3-37

-------
                                                     •o
                                                      ia
                                                      03
                                                      iti
                                                     •a
                                                      o
                                                      o
                                                     (O
                                                     •a
                                                     c
                                                     o  w
                                                     O  S-
                                                     O)  0)
                                                     £= 
-------
entering the ductwork on the pull side 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
                                                        3
Naoshima smelter has a flow rate of approximately 600 Mm /min  (21,000 scfm)
                                                                     o
The exhaust hood on the opposite side pulls in approximately 1,000 Mm /min
(35,000 scfm) of gas to the main system.  The capacity of  the  total
                                                                  q
pull system at this smelter is three times this value, or  3,000 Nm /min
(105,000 scfm), to accommodate the operation of three hoods at a time.
According to flaoshima authorities, the collection efficiency of these
                                                         27
hoods for fugitive emissions is approximately 90 percent.
     The Tamano copper smelter in Japan uses a differently designed
air curtain system along with 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
                     22
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.
                                                          3
Ambient air is supplied by a ground fan rated at 1,2000 Nm /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 S02 gas and the other is for high concentration S02 gas.
The inlet of the ductwork for high S02 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
                                                       22
90 percent effective in controlling fugitive emissions.
     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
                                3-39

-------
                                 Roof Opening
                                for Fugitive Gas
                                                    Fugitive Gases
                                                     to Bag  House
                                                            Off-gases To
                                                            Acid Plant
       Desulfunzation
            Plant
                                                Sod JPlu» for
                                             x.         Air Curtain
                                               Off t*k»
Jkir Curtain Pan


Converter Furnace
Front Door
Figure 3-14.  Air Curtain System  at the Tamano Smelter
                                3-40

-------
the sides and the back of the area around the coverter mouth, with  a      ,<
portion of the enclosure back wall 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 primary hood and the converter vessel are sealed  to
contain the emission plume.
     When the converter is rolled out for charging or skimming, the
gate 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
in an ESP for particulate removal before being passed to the atmosphere
through the main stack.
                                 3-41

-------
     Design data for the ASARCO  system  are  summarized  in  Table  3-4.
     EPA conducted a program to  evaluate the capture effectiveness of
                                                                     2Q
the ASARCO-Tacoma air curtain secondary hood system in  January  1983.
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.
     3.1.2.7.3  Building 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
                                3-42

-------
  Table 3-4.  SUMMARY OF DESIGN DATA FOR THE. ASARCO-TACOMA CONVERTER

                       AIR CURTAIN SECONDARY HOOD SYSTEM28
    Mode
of operation
Air Curtain Push Rate,
    Actual m3/min
Main Offtake Evacuation
   Rate, Actual m3/min
Matte charging

Blowing

Slag skimming

Holding

Worst conditions
   510 (18,000 acfm)

           _a


   510 (18,000 acfm)

   510 (18,000 acfm)

 1,020 (36,000 acfm)
    2,322 (82,000 acfm)

    1,700 (60,000 acfm)

    2,322 (82,000 acfm)

      850 (30,000 acfm)

    4,644 (164,000 acfm)
aAir curtain will not be used during the blowing mode.
 Worst conditions would consist of either (1) two converters being
 charged simultaneously or (2) one converter being charged while
 another was being skimmed.
                                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 rising column of hot air is high enough to maintain  the  buoyancy
of the column until it is discharged 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
                                                            o
facility.  The present building evacuation rate is 16,800 Nm/min
(600,000 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.
Worker exposure to airborne arsenic and lead has continuously exceeded
the concentration limits set by the Occupational Safety and Health
                              3                     "3
Administration (OSHA) (10 //g/m  for arsenic, 50 jxg/m  for lead).  In
an attempt to alleviate the situation, the company has increased  the
openings in the building and are presently operating roof ventilators
which 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 smelters.  Fire refining

                                3-45

-------
is carried out in rotary-type refining 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 (1,800°F), the quantity of
dilution air required to reduce the temperature to 121°C (250°F) will
         •3
be 198 Mm /min (7,000 scfm).  The collection system is designed for
      2
425 Nm /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.
Hov/ever, 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

-------
     AIR LINES
HOOD
                     PNEUMATIC CYLINDER AUTOMATIC HOOD OPENER
                                  •FLEXIBLE DUCT 28"-0
                           •HINGE
                                                                =     HOOD 1/4"  C.S.


[^•B
€r


                                                             FRONT VIEW
             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.
 Fugitive  dust emissions are  not a problem  in scrubber systems  because
 dry  dusts  are not being handled.
      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.
 In the past,  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.  These hoppers are now 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

                                3-48

-------
material is handled and transported by conventional surface vehicle
conveyance and is usually passed through a smelter flux crushing
system.
3.2  SUMMARY OF EXISTING CONTROL
                     *
     This section presents a description of air pollution control
equipment applied at the ASARCO-Tacoma smelter for the control of
process emissions and fugitive emissions.
3.2.1  Process Control Equipment at ASARCO-Tacoma
                                                            3
     Offgases from the roasters, which average about 3,570 m /min
(126,000 acfm) at 260°C (500°F), are combined with the exhaust gases
from the ancillary calcine discharge fugitive emission control system
       o
[150 Nm /min (5,200 scfm)] and are treated in a baghouse for particu-
                                                          3
late removal.  Prior to entering the baghouse, about 850 m /min  (30,000  acfm)
of tempering air is introduced to reduce the inlet temperature to less
than 121°C (250°F).  The baghouse consists of 17 compartments containing
120 bags in each.  The bags are made of acrylic and measure 20 cm
(8 inches) in diameter and 7.6 meters (25 feet) in length.  The  total
                                  2            2
baghouse filtering area is 9,950 m  (107,100 ft ).  The baghouse is
                         o
designed to treat 5,665 m /min (200,000 acfm) of gas at an air-to-cloth
                    32            2
ratio of about 0.6 m /min per m  (1.9 cfm/ft ).  Bag cleaning is
performed by mechanical shakers.  The clean baghouse exhaust is  vented
through a flue to the smelter main stack.
     Process gases from the reverberatory furnace, which average about
        o
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 take place by air infiltration and
water and sulfuric acid sprays located in the flue.  The resultant gas
stream, about 6,100 actual m3/min (215,000 acfm) at about 110°C  (230°F),
then enters the first of two electrostatic precipitators (in series)
for particulate removal.  The first precipitator is of the tube  design
and consists of 18 sections with a total collection area of about
       2           2
6,619 m  (71,250 ft ).  Each section contains 84 pipes measuring
30 cm (12 inches) in diameter and 4.6 m (15 feet) in length.  The
second unit is of the plate-type design, consisting of seven parallel
chambers, each with four fields in series, and has a total collection
                                3-49

-------
area of 7,710 m2  (82,992 ft2).  The exiting gases, about 7,740  actual
m3/min (270,000 acfm) at 93°C  (200°F), are discharged through a  large
flue to the smelter's main stack.
     Offgases from converter blowing operations are captured by  water-cooled
hoods and passed  through a series of multicyclones and a settling flue
for coarse particulate removal prior to entering the gas cleaning
circuits of either a liquid SOp plant or a single-contact sulfuric acid
plant.  The gas cleaning circuits for both plants are similar and
consist of a water spray chamber, an electrostatic precipi-tator,
scrubbers, and a mist precipitator, all being in series.  The single-contact
acid plant has a  182 Mg/day (200 ton per day) capacity at 5 percent
                                       o
S09 and is capable of processing 650 Nm /min (23,000 scfm) of converter
                                                   3
gas.  The liquid S02 plant processes up to 1,270 Nm /min (45,000 scfm)
of the converter gases.  The plant uses dimethyl aniline (DMA) to
absorb the S02 in the gas stream and uses steam stripping for regeneration.
The 100 percent concentrated S02 gas stream produced is then liquefied
by compression and the liquid S09 stored.
                                                                 3
     Offgases from the three hearth-type anode furnaces [1,400 Nm /min
(50,000 scfm) at about 93°C (200°F)] are treated for particulate
removal in a plate-type ESP before being discharged through the  main
stack.
     Emissions from the arsenic plant at ASARCO-Tacoma are controlled
by a baghouse.  In addition, several smaller baghouses are used  to
minimize dust emissions from drag conveyors, charge hoppers, and
storage bunkers.  The main baghouse consists of five compartments
containing 240 acrylic bags each.  The baghouse treats about 1,200 Nm /min
(42,000 scfm) of gases from the metallic arsenic plant, arsenic plant
top flue, arsenic kitchens, and Godfrey furnaces.  The air-to-cloth
ratio is 1.2 m3/min per m2 (3.98 cfm/ft2) at 1,700 m3/min (60,000 acfm),
at the design temperature of 93°C.  Bag cleaning is performed
by an air pulse system.  The clean baghouse exhaust is vented to the
smelter's main stack.
     The estimated control efficiencies of the process emission  control
equipment at ASARCO-Tacoma are presented in Table 3-5.  Performance
testing of the process controls is discussed in Section 3.3.1.
                                3-50

-------
            Table 3-5.  ESTIMATED CONTROL EFFICIENCIES OF
          EXISTING PROCESS EMISSION CONTROLS AT ASARCO-TACOMA
 Process
emission
 source
Control
device
Estimated current
control  efficiency
Multi-hearth
  roasters
Reverberatory
  furnace
Converters
Anode furanees
Arsenic plant
Baghouse

ESP
Acid plant
ESP
Baghouse
        99

        98
        99
        96
        98
3.2.2  Fugitive Control Equipment at ASARCO-Tacoma
     It should be noted that fugitive emission controls are not applied
extensively throughout the copper smelting industry.  Where applied,
controls typically consist of using local ventilation techniques
(i.e., hoods and enclosures) to confine and capture emissions.  The
captured emissions are typically discharged directly to the atmosphere
through a stack.  In some instances, however, captured fugitive emissions
are treated for particulate removal in a control device either individually
or combined with process offgases prior to treatment.  A discussion
follows of existing fugitive emission control systems in operation at
ASARCO-Tacoma.
     Fugitive emissions which could escape from the discharge of
calcine from the roaster are confined and captured by close-fitting
exhaust hoods located at the discharge point.  The exhaust stream is
                                              o
vented into the main roaster flue by two 74 Nm /min (2,600 scfm)
fans for treatment in a baghouse system (see Section 3.2.1).  The
sides of the larry car loading area are enclosed.  Larry car covers
are used while cars are in transit.  Empty cars are parked under a
ventilation connection.
                                3-51

-------
     Potential fugitive emissions during charging of the reverberatory
furnace are minimized by the use of a manual control override, which
simultaneously opens the furnace flue control damper and reduces the
fuel supply to the furnace prior to each charge to prevent pressure
surges in the furnace.  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 for the matte tapping system
                  o
totals about 700 m /min (25,000 acfm), while the ventilation for the
                                      o
slag tapping system totals about 600 m /min  (21,000 acfm).  Captured
emissions from both systems are combined with anode furnace offgases
and treated in an electrostatic precipitator (ESP) for particulate
removal prior to being discharged through the main stack.
     As mentioned earlier, several  small baghouses are utilized to
control potential fugitive dust emissions from the arsenic production
building.  These baghouses collect dust generated during material
transfer operations, such as charging of the Godfrey roasting furnaces.
Housekeeping practices, such as manual vacuuming and wet cleaning of work
areas, are also a primary means used to minimize dust entrainment into
the air during operations at the arsenic plant.
     Table 3-6 presents a summary of the estimated fugitive emission
control efficiencies currently being achieved at ASARCO's Tacoma
smelter.  A discussion of fugitive controls at this smelter is contained
in Section 3.3.2.
            Table 3-6.  ESTIMATED CONTROL EFFICIENCIES OF
         EXISTING FUGITIVE EMISSION CONTROLS AT ASARCO-TACOMA

Estimated current
Emission
source
Calcine transfer
Matte tap
Slag tap
Converters
Anode furnace
Arsenic building
Miscellaneous

capture
90
90
90
0
0
a
a
control efficiency,
collection
99.8
96
96
0
0
a
a
percent
overal 1
89.8
86.4
86.4
0
0
90
90
a
 See Table 2-11.
                                3-52

-------
3.3  PERFORMANCE CAPABILITIES OF ALTERNATIVE CONTROL TECHNIQUES  FOR
     ARSENIC AND TOTAL PARTICULATE EMISSIONS
     In order to evaluate the emission control capabilities 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
regarding the performance capability of each of the control techniques
evaluated for process and fugitive emission sources.
3.3.1  Process Control Systems
     3.3.1.1  Baghouses.  Tests were performed at the ASARCO-Tacoma
            *
and Anaconda  smelters to evaluate the performance of baghouses  in
controlling arsenic emissions.  Tests were also performed at the
Anaconda smelter to evaluate the performance of the baghouse in  controlling
total particulate emissions.  Two baghouses were sampled at ASARCO-Tacoma
and one at Anaconda.
     3.3.1.1.1  Baghouses (ASARCO-Tacoma).  Simultaneous inlet and
outlet arsenic emission measurements were performed by EPA across the
baghouse serving the multi-hearth roasters at ASARCO-Tacoma.  A  description
of the control system was given in Section 3.2.1.
     Sampling at the inlet was performed in the duct carrying emissions
from the roasting process in four 10 cm (4 in.) ports on top of  the
flue.  The gas leaving the roaster baghouse was sampled approximately
305 m (1,000 ft) downstream from the baghouse.  Three tests were
performed at each location.  The results are summarized in Table 3-7.
                                                                     32
     As indicated, the average arsenic inlet concentration and corresponding
mass rate were 288 mg/Nm3 (0.126 gr/dscf) and 87 kg/hr (191 Ib/hr),
                                                       o
respectively.  The outlet arsenic loading was 0.9 mg/Nm  (0.0004 gr/dscf),
and the mass rate was 0.3 kg/hr (0.6 Ib/hr).  The average arsenic
removal efficiency for this unit, as indicated by these results, was
99.7 percent.
     Simultaneous inlet and outlet arsenic emission measurements were
also conducted at the baghouse used to treat a combination of offgases
from the arsenic production facilities at ASARCO-Tacoma.  These facilities
*No longer in operation.
                                3-53

-------
 UJ
 C/)
 ra
 o

 CD
 «=c
 CO

 or
 UJ

 00
 D;

 UJ
o
U-
UJ O
o o
UJ
Q.
oo
co

(U
CJ -*J
C fi=

10 S-


CO CO "3- I~»
CD CTt O^ CTl
cn cn cn cn




«3- «»• o" 10
o o' t-4 o"

CM CM m fn

cJ o o o'


>— » o*—1^ ,^--. ---"^
ro cn r^. ^3*
O C3 O O
CD O O O
o o o o

o o o o


^o r**. LO cn

O O I-H O




F~I cn CD p***
cn co co co
i-H «— < *— * t-H
CO f***« CM *!O
CO CO CO CO



**• *T LO ^-1
o cn r** cn
CVJ i— t ^-< i— 1


rx. rH m co
CM co cn <-o
cn co r-» co



*— *. *^-* *— ^ -^^
CO O i-* «D

• • •
O O O O
CO CO CM I-H
^J. ^ ^ gQ
»-) cn ir) co
fO CM CM CM


i— i un co ^*
O co O cn
CM 1—1 CM ^H
«T LO LO '•— 1
cn co cn cn

.
••* CM ro cn
^





*— ^
^
re
o
5
-o
re

-M
c
o
i-
,jC
•M
re

r->
fC

o


OJ



cr
o


^J
c
CJ
1
u
3
v»
ai
5
j-
O

•a

w
_2

CD
S-
ro
4-*

•o
OJ

*-
CO

£
C
re

c
o
re

CJ
c
o
CJ
re
                               3-54

-------
include six Godfrey roasters, arsenic 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
                                                                2
13 cm (5 in.) in diameter.  The total filtering area was 1,860 m
(20,000 ft2).  It was designed to effectively treat 850 Nm3/min (30,000 scfm)
                                  32             2
at an air-to-cloth ratio of 0.63 m /min per m  (2.06 cfm/ft ).  Bag
cleaning was performed by mechanical shakers.  The outlet sampling was
done approximately 150 m (500 ft) downstream of the baghouse.  The
                                                  33
test results obtained are summarized in Table 3-8.
     As indicated, the average arsenic inlet concentrations and corresponding
                                          o
mass rate at the baghouse were 2,941 mg/Nm  (1.28 gr/dscf) and 76.3 kg/hr
                                                                     3
(168 Ib/hr), respectively.  The outlet arsenic loading was 60.6 mg/Nm
(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.
     3.3.1.1.2  Spray chamber/baghouse (Anaconda).  Inlet and outlet
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
                                  34
Tables 3-9 and 3-10, respectively.
     When the tests were conducted, the smelter had the foil owing
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 Mm /min (73,000 scfm)],
which were diverted to an acid plant for SOp 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.
                                3-55

-------
 C3
 C_J
 O
 o
 LU
 CO
 C3
 «=C
 QO
 a.

 o
oo
o:
LU
g

LU
C_J
D;
o
LU
a_


o
I—H
Z
LU



ct


 •
CO

CO

CD


to
























IB
in
c
o


tn
•r-
U
C
*- 0.
LU
















i—
1 *
.0
1 —






jj
*^~
0)
tn
i-


o
•»—




E




c
(!)

a
^
U

c
O)

L-
^




*
^


CO
^
*^^
en
E


CJ
0
t
en
•^


CO

^y

en




s-
cn








en



o
JZ

en


CO

z:

en
E


CJ
o




ex
£ c:
re 3
CO fc-

en r*» in r*»

in r-~ «s- in



in o in co
co* *H in co'



CO CO <-H to
en in to C3
to 1-1 en to



CO CM »— 1 tD
co r^. ^^ ^s.
0 CO 0 CO
in CM i— i t£>
eo «3- o r—
»— »


co en r^* »~i
•cj* *— t m *^*
tn in to en

CO •— 1 CO CM





to i- en en
• o • •
CM • in CM
*»• o tn co





i-H in CM
eo rH CM in
en .-i co
f-4 CM «— 1


i-- CM i~. en
—< •-< en o
t— i *—t ?— <
»j- co ,-< «r
• • • •
CM CM in CO



r^ CO CM en
i-H •— * CO CO
f-» in in tn

*— ( f^l *•-( «— 4 '


VO CO CD CO
in in to in





.a en
»-) CM CO >








.
14-
rtJ
.C

U
s

•o
c
re
c;
o
i_
r J*
-t-> re
re S-
U (U
f— O
4J C
o 
cu
J= OI

re
(A O
c

1 i
a tn
tn tn
IB O)
CJ U
E 0
I.
C O.
0

•8 •£
tn oj
ro
0)
*" £
•>-> IB
IB •(->
p

i- £
tfl C
W) ~3


OJ

c 'a.
to E
C tn
•^ (/)
•M •.-
IB J=
£- 4->
c en
CJ C
O *r—

O 3
Cj >"">

-------
         Table 3-9.  ARSENIC PERFORMANCE DATA FOR SPRAY CHAMBER/BAGHOUSE
                         AT THE ANACONDA-ANACONDA SMELTER

Sample
run
1
2
3
Avg.


Arsenic
Inlet
°C
274
269
244
263
mg/Nm
1,071
895
687
885
kg/hr
276.7
236.8
186.1
232.3
Emissions9

Outlet
°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
    Concentration and mass rate data are based on measurements on the total
    catch (front and back half).
    Table 3-10.  PARTICULATE PERFORMANCE DATA FOR SPRAY CHAMBER/BAGHOUSE
                      AT THE ANACONDA-ANACONDA SMELTER

Sample
run
1
2
3
Avg.


Particulate Emissions
Inlet
°C
281
288
302
290
g/Nm3
14.76
13.57
14.08
14.14
kg/hr
4,071
3,736
3,860
3,890
b

Outlet
°C
103
103
101
102
g/Nm3
0.05
0;04
0.05
0.05
kg/hr
14.6
10.0
14.7
13.1

Efficiency,
percent
99.6
99.7
99.6
99.7
Concentration and mass rate data are based on measurements on the probe,
cyclone, and filter catch (front half).
                                     3-57

-------
The baghouse system consists of a baghouse preceded by two  parallel
spray chambers, which effectively reduce the temperature of the  inlet
gas stream from about 315°C (600°F) to less than 104°C (220°F) prior
to entering the baghouse.  The baghouse consists of 18 compartments,
each equipped with 240 Orion bags, and has a net collection area of
29,800 m2 (32,080 ft2).  It is designed to effectively treat  5,660 Nm3/min
(200,000 scfm) at an air-to-cloth ratio of 0.25 m /min per m2  (0.80 cfm/ft2)
Bag cleaning is performed by mechanical shakers.
     As indicated in Table 3-9, the arsenic inlet concentration  and
                                                                 3
corresponding mass rate were extremely high, averaging 885 mg/Nm
(0.3867 gr/dscf) and 226.5 kg/hr (498.3 Ib/hr), respectively.  The
                                    o
outlet arsenic loading was 9.8 mg/Nm   (0.0043 gr/dscf), and the mass
rate was 2.6 kg/hr (5.8 Ib/hr).  The average arsenic removal  efficiency
for this unit, as indicated by these results, was 98.9 percent.
     Corresponding data in Table 3-10 indicate that the inlet  particulate
                                               o
concentration and mass rate averaged 14.14 g/Nm  (6.2 gr/dscf) and
3,890 kg/hr (8,558 Ib/hr), respectively.  The outlet particulate
loading was 50 mg/Nm  (0.022 gr/dscf), and the mass rate was  13.1 kg/hr
(28.9 Ib/hr).  The average particulate removal efficiency for  this
unit was 99.7 percent.
     3.3.1.2  Electrostatic Precipitators.  Arsenic emission measurements
were conducted on the electrostatic precipitator (ESP) units at the
ASARCO-Tacoma and ASARCO-E1 Paso smelters to evaluate the performance
of ESP's operated at reduced temperatures.  In addition, the ASARCO-E1
Paso ESP was also tested for particulate matter.
     3.3.1.2.1  Electrostatic precipitator (ASARCO-Tacoma).  Arsenic
emission measurements were conducted by EPA on the outlet of the ESP
which treats offgases from the reverberatory furnace.  Prior to entering
the precipitator, the gases pass through a pair of waste heat  boilers
where the temperature of the gases is reduced to 400°C (750°F).  The
exiting gases then pass through a large rectangular brick flue where
additional  cooling and gas stream conditioning are provided by air
infiltration and water and sulfuric acid sprays located in the flue.
                                              o
The resultant gas stream, about 6,100 actual  m /min (215,000 acfm) at
132°C (270°F), then enters the first of two electrostatic precipitators
in series for particulate removal.  The first precipitator is  a tube

                                3-58

-------
or pipe design consisting of 18 sections with a total collection area
                2           2
of about 6,619 m  (71,250 ft ).  Each section contains 84 pipes measuring
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 collection
area of 7,710 m2 (82,992 ft2).  The exiting gases, about 7,740 actual
m3/min (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
                                  35
outlet is presented in Table 3-11.
  Table 3-11.  ARSENIC EMISSIONS AT OUTLET OF REVERBERATORY FURNACE
              ELECTROSTATIC PRECIPITATOR AT ASARCO-TACOMA

Arsenic Emissions3
Sample
run
1
2
3
Avg.

°C (°F)
105 (220)
101 (214)
87 (188)
97 (207)
0
mg/Nm (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)
 Concentration and mass rate data are based on measurements on the
 total catch (front and back half).
     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.
     3.3.1.2.2  Spray Chamber/Electrostatic precipitator (ASARCO-E1 Paso).
Inlet and outlet arsenic emission measurements were made by EPA across
this spray chamber/ESP.  Three runs were made at each of three inlet
                                3-59

-------
locations and at one outlet location. A summary of the test results is
                        37
presented in Table 3-12.    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
                                                                     oc oy
made only at the outlet.  Tables 3-13 and 3-14 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
        o
5,100 Mm /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 3-12, the average arsenic concentration
recorded at the inlet and outlet was 0.308 and 0.006 g/Nm3 (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 3-13 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/Nm  (2.23 gr/dscf) and 1,134 kg/hr
(2,495 Ib/hr), respectively.  The average particulate concentration
                                                          o
and mass rate at the outlet were, respectively, 0.098 g/Nm  (0.043 gr/dscf)
*In current practice, reverberatory furnace gases are treated separately
 in one ESP while roaster gases are treated in another ESP.
                                3-60

-------
a:
UJ
o o
   co
>- <
=£ 0.
n:
o -- 1
to uj
O O
   «=c
   co
2: o

O- UJ
   ex:
UJ I— l
co i—
   CO
   o
CM
 CU


JO

 (O
   UJ

   _1
   UJ




















(O
E
0

in
in

c
cU

o
E
CU
CS)
s_
«^































0 4->
CU CU
•i- U
O 5-
•r- CU
M- O.
tl
*T~
UJ






[ *

0 0 O O
0 0 O O
• • • •
O 0 0 O

CD o a o
l-H 1— 1 i— < t— 1




•* CO O CT>
UTS co cy> i-n
CD CM <;h CM
'^d" ^t* ^h ^*i"





CO r-l CO CTl

co co co in
co CM CM at
i— 1 t-H


O CO -=3- CO
O CO OI CD
«-) CO U") CO
• • • •
o o o o



«3- i-l CM CM
o% cTt m »-H
o— 1 T— 1 CM CM


(£) CO f-- CO
CO CO «3 CM
*VO O*i ^t" ^"^
to in co to





•
CD
i-H CM 0-> >
^^


•o
E


•^
O
s-

o
(O
o
,_
re

o
4-5
cu
£.
O
in
c
CU


5-
•^
in
to
CU
E
E
O

-a
cu
t/>
(0
JD
CU
S-
re

re

re
•o

cu
4_>
re
S-
t/5
i
-^
r-
re
E
0
•^- •
4-> - — -
re H-
s_ ^~
cu
o -^
E U
o re
<_5 J3
re
                                             3-61

-------
 UJ
 03
=C O
o; co
a. «c
OO Q_

UJ _J
3C UJ
I— I
   o
ce. o
o ct.
UJ
o; o
UJ UJ
a. a:
   Q.
UJ
J-o

cu s=
•i- CU
o o
•i- S-
<+— CU
M- CL
UJ







4-3
CU
r—
4_3
3
O
























^_)
cu

c:
i — t













s_
-c
CD
>^*


CO
E
•zz
^*^
CD

•r-
^
^^
CO
E




CJ
o



5-
C"
•^s^.
CD






CO
E

^^s.
CD

C
•f—
£
^**^
CO
Jjj


o
o


(D
r—
CL C
£: *"*
n3 S—
oo



i— i co r*.
• • •
t^3 I — tO
01 cn cn




CO CO CM
o co r-^
•=t- co co


t-l ID CO
i— 1 CO CT>
i— 1 O O
• • •
000

to o co
•=*• O CM
CO l~~ LO
n «\ «\
"d- >3- «d-



LO tn LO
000
t— i i— i i— \



CM OO *^t"
CM «=t- CO
O CM i-H
MAW
i— 1 t— 1 r— 1






"^" ^^ r— 1
in co t— i
• • •
•* «3- in

r— ) i— 1 tO
•=d- co co
r-» to to
M M «%
CO CO CO


to r-i cn
CM CO CM
CM CM CM



«
cn
i— 1 CM >
^
o
A
cu
JD
o
S-
Q.

cu
-C

o

CO
•4-J
c
CO
1
S-

co
ra
s=
c:
o

T3
O)
co
ra
JD

O)
S-
ra

ra


-------
  Table 3-14.   PARTICULATE PERFORMANCE DATA FOR THE SPRAY
CHAMBER/ELECTROSTATIC PRECIPITATOR OUTLET AT ASARCO-EL PASO


Sample
run
1
2
3
Avg.
Particul

Nm /min
4,346
4,940
4,816
4,700
ate Emissions

°C
104
96
104
101
T
g/Nnr
0.14
0.21
0.09
0.15

kg/hr
47.3
73.0
32.3
50.9
    Concentration and mass rate data are based on
     measurements on the probe, cyclone, and filter
     catch (front half).
                          3-63

-------
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 3-14 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
                                                  3
recorded during these latter tests were 0.15 g/Nm (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.
     3.3.1.3  Venturi Scrubbers (Kennecott-Hayden).  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
                                   o
exhaust, consisting of about 565 Nm /min (20,000  scfm) with an estimated
                       3
dust loading of 57 g/Nm  (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 gpm), 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 3-15 presents a
                       qo
summary of the results.
                                3-64

-------
    Table 3-15.  ARSENIC PERFORMANCE DATA FOR VENTURI SCRUBBER
                         AT KENNECOTT-HAYDEN

Sample
run
1
2
3
Avg.
Arsenic emissions3
Inlet
°C
336
328
324
329
mg/Nm
29.53
25.87
22.90
26.10
kg/hr
0.85
0.74
,0.75
0.78
Outlet
°C
46
44
28
39
mg/Nm
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
 Concentration and mass rate data are based on measurements on the
 total catch (front and back half).
     As the results indicate, the arsenic inlet loading to the scrubber
                                    o
was quite low, averaging 26.10 mg/Nm  (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
from 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.
     3.3.1.4  Sulfuric Acid Plants.  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.
     3.3.1.4.1  Double-contact acid plant (Kennecott-Hayden).  The
double-contact acid plant operated at this smelter treats a combination
of fluid-bed roaster and converter process gases.  Acid production is
typically about 935 Mg/day (850 tpd) of 93.5 percent sulfuric acid.
                                3-65

-------
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 humidification 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
                                                         o
stream.  The combined gas stream, totaling about 2,120 Nm-/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 SOg, then enters the double-contact acid plant where it is
dried, the SOp converted to SO,,, and the SO, 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
                                                                   3
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
(0.90 lb/hr).38
     3.3.1.4.2  Double-contact acid plant (ASARCO-E1 Paso).  Offgases
generated during converter blowing operations at the ASARCO-E1 Paso
smelter are treated in a 454 Mg/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.
                                3-66

-------
     Arsenic emission measurements were conducted by EPA at the  inlet
to the spray chamber and at the acid plant outlet.  Three sample runs
were made on the inlet and four on the outlet.  The results are summarized
in Table 3-16.
              39
      Table 3-16.  ARSENIC PERFORMANCE DATA FOR DOUBLE-CONTACT
                    ACID PLANT AT ASARCO-EL PASO

Sample
run
1
2
3
4
•Avg.
Arsenic emissions3
Inlet
°C
222
209
200
b
210
9/Nm3
2.36
0.228
0.262
b
0.976
kg/hr
233.3
21.4
24.4
b
96.0
Outlet
°C
64
64
66
69
67
g/Nm3
0.0002
0.0031
0.0011
0.0004
0.0015C
kg/hr
0.022
0.355
0.126
0.038
0.168C

Efficiency,
percent
99.99
98.3
99.5
b
99.8
 Concentration and mass rate data are based on measurements on the
 total catch (front and back half).
 Only three inlet sample runs were made.
cAverage 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 Ib/hr)
at the inlet and 0.168 kg/hr (0.370 Ib/hr) at the outlet, indicating
an average arsenic removal efficiency in excess of 99 percent.
     3.3.1.4.3  Single-contact acid plant (Phelps Dodqe-Ajo).  Offgases
generated during converting at the Phelps Dodge-Ajo smelter are treated
in an ESP for particulate removal followed by a 544 Mg/day (600 tpd)
single-contact sulfuric acid plant for SO,, 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 horizontal  parallel units
                                                              3
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 (13.8 psia).  The exiting
                                3-67

-------
     gases pass 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
                                                       40
     outlet.  The results are summarized in Table 3-17.    The offgases
     treated in the acid plant contained a negligible amount of arsenic.
                                                                       3
     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 3-17.  ARSENIC PERFORMANCE DATA FOR SINGLE-
                     CONTACT ACID PLANT AT PHELPS DODGE-AJO

Sample
run
1
2
3
Avg.
Arsenic emissions9
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).
     3.3.2  Fugitive Control Systems Evaluation
          3.3.2.1  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
                                     3-63

-------
3.1.2.6, respectively.   Visual observations 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 3-18.
     3.3.2.1.1  Calcine  transfer.  Thirteen calcine  transfer operations,
each averaging about 2 minutes 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.
     3.3.2.1.2  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.
     3.3.2.1.3  SIag 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
                                3-69

-------
LU
CL.

S

o;
«o
U_
     o
O   O
I— i   O
J—   Qi
LU
00
     co
     LU
     O
CO
CO
>—<

s

LU
     CD
     CO
     o
     CO
CO   CO
co
-o
C(_ >,+J tU
0 4-> C >
•r- tU S-
tu o o cu
CD IB i- CO
C CL CO -Q
(0 0 Q. O
tU « 4-i
CD >, £=
res 4-» cu
S- -r- o
CO O i-
> (0 CO
en  " CO
Q. CD  E ••
J- S- -i- C
tU tU 4J T-
> CO E
O
CO
t- CD
O C C
•i- co
. -a -^
o  £= -O
o c: o co
tu tu T- >
co o E co s_
CD S- -i— CO tU
C CO 4-> -i- CO
res CL E -Ct
cu tu o
CO
tU -M C -O CO
CD C O) O tU i — CD
rel tu E -r- > i— C
CM s- o •!- co s- rei T-
CM tu s_ 4-5 to co -a
> tu -i— co c rei
"O «^ CL E ° O CO
O 03 O S-
j *
co c
2: co o
CD'r— O
eC ra 4-= « CD
Q- s_ res tu co
LU tU > E ••
> S- T- E

.a '^
o
co
CD
» 4-> CD
CO i- S-
o res
S- CL .C
CO o
4— •• Cu co
CO CD reS *r—
c: c -t-> . TD tu
res 'v~ t—~
S— CL CO CO ~O
+-> Q.+J s_ +J res
03 4J (U 4-> r—
cu +J rei ~o res
= E s= E 0
•r- CO 3 4J
O 4-> 4-> (O 4-3 C
i— +J res r— res -t-
fO (T3
o s:




0 0
CO LO
1 1
0 0




CO CM
.—1








LO CM
«3- CO
• • • •
CO ""J"
i— I r-<




CM r-^




LO tn
i — 1 CTi
1 1
O CM
r-.




CO

LO OO
CO







co r>~
CO CM

CO LO
<— 1 r-H




CO J-J





•a
ET
res
CO
4-J CD
S- i-
o res
Q. J=
o
• • Q. co
CD reS "r—
c +-> -o co
•r- 4->
Q- CD s- en o
Cu res tu res CL
res i — T3 i —
-P co c co O
3 4J
CD4-> reS 4-» C
res res *— res *^
1
to
                                                                                               CD
                                                                                               CO
                                                                                               O1
                                                                                               CO
                                                                                               CM
                                                                                               O
                                                                                               S_
                                                                                              •=*•
                                                                                              CM
                                                                                               CU
                                                                                               C
                                                                                               O)
                                                                                              T3
                                                                                               rei
                                                                                               co
                                                                                               c
                                                                                               o

                                                                                               +3
                                                                                               s_
                                                                                               tu
                                                                                               CO

                                                                                              •8

                                                                                               c:
                                                                                               o
                                                                                              •r—
                                                                                               CO
                                                                                               CO
                                                                                               at
                                                                                             (O
                                                     3-70

-------
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.
     3.3.2.2  Fugitive Emission Controls for Converters-Air Curtain
Secondary Hood Capture System.
     3.3.2.2.1  Evaluation program at ASARCO-Tacoma.  EPA conducted an
evaluation program in January 1983, on the prototype air curtain
                                                   29
secondary hood recently installed at ASARCO-Tacoma.    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 v/as 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 chromatographic
analysis.  The air curtain capture efficiency was calculated by comparing
the SFg 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 center!ine) in
                                 3-71

-------
                                                                  NO.  4
                                                                CONVERTER
                                TOP VIEW
           JET. SIDE
  AIR
CURTAIN
  JET
            BAFFLE
            WALL
                                                     EXHAUST SIDE
                              AIR CURTAIN
                             NO.  4  CONVERTER
                              (FUME SOURCE)
BAFFLE
 WALL
                                                                TO SUCTION FAN
                                                                           LEGEND:
                                                                                   AREA SAMPLED USING
                                                                                   HATRIX TRAVERSE
                   INJECTION  LOCATIONS
                   SAMPLE I.D.

                   V SP1  J 2
                   Q SP3  - 5
                   • SP7  - 12
                   O SP13 - 73
                                ELEVATION
             Figure  3-18.   SFg  Tracer  Injection  Locations
                                             3-72

-------
                                    CONVERTER AISLE FLOOR
                                         O  INJECTION POINTS
Figure  3-19.  Tracer Injection Test Ports
                       3-73

-------
an attempt to characterize the effective capture area of the air
curtain hooding system, particularly during converter rollout 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 3-19.
    Table 3-19.  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 3-20.
                                3-74

-------
           Table 3-20.  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
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
      matte charge and cold addition, and 5 percent slag skimming
      and copper pour.
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
               42
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
efficiencies.  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 system.
     In addition to these two test series, a series of special injection
point tests was conducted in order to assess the effective capture

                                3-75

-------
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 3-21  shows  the results  of
this test series.

    Table 3-21.  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
air curtain control system through the various converter operating
      43
modes.    The purpose was to estimate capture effectiveness and to
                                3-76

-------
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
withdrawal 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,
solidified materials)  to the converter frequently produced emissions
heavy enough to virtually overwhelm the capture system, especially
when a fire ignited in the converter.  Capture efficiencies were
                                3-77

-------
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 skimming activities) indicates
that efficiencies calculated by the gas tracer method may be too high
in some instances.  The gas tracer method is based on the assumption
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.
     3.3.2.2.2  Visible Emissions Observations at Tamano Smelter.  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
                                3-78

-------
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 Fierce-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 charging.  Usually three ladles of matte are brought to the
converter and charged in a 10- to 30-minute 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 system 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
                         44
summarized in Table 3-22.
     As shown in Table 3-22, visible 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.

                                3-79

-------
            Table 3-22.  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
periodj 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
individual
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 and 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.
     Slag discharge.  At the end of each of the three slag blow phases,
slag is skimmed 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
                                3-80

-------
 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.  During  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.
     Blister  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;
 the hood  doors are opened; the blister 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 3-23.44
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,

                                3-81

-------
      show that the visible emissions observed were somewhat more substantial
      than those observed during either matte charging or slag skimming.  As
      shown in Table 3-23, 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 3-23.  VISIBLE EMISSIONS OBSERVATION DATA -FOR
                       BLISTER DISCHARGE AT THE TAMANO SMELTER
Samp! e
Run
1
2
3
4
Method
22
Observation Percent of
period, min. time emissions
observed
25a
_
15b
6C
Total 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.
           3.3.2.3  Building Evacuation and Baghouse (ASARCO-E1 Paso) - Fugitive
      emissions from converters and anode furnaces at the ASARCO-E1 Paso
      smelter are captured by evacuating the converter building.  The building
      evacuation system at this smelter is described in Section 3.1.2.7.3.
      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.
                                      3-82

-------
     The fugitive gas flow through the baghouse averages about 14,100 Nm /min
(498,000 scfm).  The baghouse consists of 12 compartments.  Normally
all compartments are in operation except that one compartment is taken
off during the cleaning cycle and another compartment during the main-
tenance 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
               2           2
area of 1,644 m  (17,700 ft ) 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 acfm) at 54°C
(130°F) with an air-to-cloth ratio of 0.91 m3/min per m2 (3 cfm/ft2).
Mechanical shakers are used for cleaning the bags.
     Inlet and outlet emission measurements for inorganic arsenic and
total particulate were conducted by EPA across the baghouse.  During
all tests, converter operations were monitored and testing was conducted
only when one or more converters were in operation.  The arsenic
                                              45
results obtained are summarized in Table 3-24.    As indicated, the
                                                                    3
measured inlet and outlet arsenic concentrations averaged 3.27 mg/Mm
(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
0.111 kg/hr (0.244 Ib/hr) at the outlet, indicating an average arsenic
removal efficiency in excess of 96 percent.  The results of the particulate
                                                   46
measurements obtained are summarized in Table 3-25.    As shown in the
table, the mass particulate inlet concentration was low, averaging
only 0.062 g/Nm  (0.027 gr/scf).  Nonetheless, the mass particulate
emission 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
techniques.  The outlet concentration and mass .rate averaged 5.1 mg/Mm
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.
3.3.3  Conclusions
     3.3.3.1  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

                                3-83

-------
Table 3-24.  ARSENIC  DATA  FOR  CONVERTER  BUILDING
              BA6HOUSE 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 3-25.  PARTICULATE DATA FOR CONVERTER BUILDING
              BAGHOUSE AT ASARCO-EL PASO

Sample
run
1
2
3
Avg.


Parti cul ate 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
                          3-84

-------
preceding 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 3.3.1 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
efficiences 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
control device can vary depending on the distribution between the
particulate and vapor form of the arsenic which-reaches the device.
This distribution in turn depends on the arsenic concentration in the
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.
                                3-85

-------
    100
     99
     98
     97
g    96
     95
o
cs
LU
     94
     93
     92
      91
     90
           IT  -
 BH
91°C
 BH
84°C
                                     I
                            BH -  BAGHOUSE
                            SC -  SPRAY CHAMBER
                            ESP-  ELECTROSTATIC  PRECIPITATOR
                            VS -  VENTURI SCRUBBER
                            AP -  ACID PLANT
                            IT -  INLET TEMPERATURE
                               I	I	I	
SC/BH
110°C
SC/ESP
115°C
  VS
315°C
AP-A
210°C
     Figure 3-20.   Control  Device Arsenic Collection Efficiencies
                                    3-86

-------
     3.3.3.2  Fugitive Controls.  The performance capabilities of
capture systems for the control of fugitive arsenic emissions from
calcine discharge, matte tapping, slag tapping, and converter operations
were evaluated.  Estimates of the capture efficiency of these systems
are based on the visible emissions observations reported in the preceding
sections and on subjective judgment.  Also evaluated was the performance
capability of a baghouse control device used to collect captured
fugitive emissions.
     Visual observations made on the local ventilation sys'tem 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-ladle 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.  It
is concluded, based on these data, that a properly designed and operated
ventilation system applied to matte tapping operations should achieve
a minimum capture efficiency of 90 percent.
     Similar observations made on the ventilation system serving the
slag tapping operations at Tacoma showed substantially poorer performance,
especially at the slag launder-to-slag pot transfer point where visible
emissions were observed nearly 100 percent of the time during each
individual slag tap.  Based on the results of the visual observations
and the fact that the capture system had reportedly been damaged, it
is concluded that the slag tapping ventilation .system observed at
Tacoma, 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.
                                3-87

-------
     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 Tamano 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.
However, as previously indicated, these 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
                                3-88

-------
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/Nm  (0.0014 gr/scf).
                                3-89

-------
3.4  REFERENCES

 1.  Vallance, R.H.  Arsenic In: A Text-Book of Inorganic Chemistry,
     Friend, J.N. (ed.), London. Charles Griffin and Co.s Limited.
     1938.  p. 129-131.

 2.  Handbook of Chemistry and Physics, 43rd Edition.  The Chemical
     Rubber Publishing Company.  1961.  p. 2373.

 3.  Behrens, R.6., 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.
                                3-90

-------
14.  Powlesland, J.W.  Air Curtains in Controlled Energy Flows - To
     Stop or Regulate Air Flows - To Contain and Convey Airborne
     Contaminants.  Presented at the 22nd Annual Industrial Ventilation
     Conference.  February 1973.

15.  Katari, V., et.al.  Pacific Environmental Services, Inc., Plant
     visit report for ASARCO Copper Smelter, Tacoma, Washington,
     during June 24 to 26, 1980.  Pacific Environmental Services, Inc.
     July 14, 1980.  p. 4.

16.  PEDCo Environmental  Specialists, Inc.  Secondary Hooding for
     Pierce-Smith Converters.  U.S. Environmental Protection Agency.
     EPA Contract No. 68-02-1321.  December 1976.

17.  Iverson, R.  Visual  Evaluation of the Converter Secondary Hood
     System at the Phelps Dodge-Ajo Smelter.  U.S. Environmental
     Protection Agency.

18.  Movie on the Operation of the Swing-Away Hood Supplied by the
     Nippon Mining, Ltd.  of Tokyo, Japan.

19.  Devitt, T.W.  PEDCo Environmental, Inc.  Control of Copper Smelter
     Fugitive Emissions.   U.S. Environmental Protection Agency.
     Cincinatti, OH.  Publication No. EPA-600/2-80-079.  May 1980.

20.  Brochure on the Saganoseki Smelter and Refinery by Nippon Mining
     Company, Ltd., Japan.  1978.

21.  Mitsubishi Metal Corporation.  Guide to Nonferrous Metals Smelting
     and Refining Technologies.  (Pamphlet).  Japan.

22.  Katari, V. and I.J.  Weisenberg.  Trip Report—Visit to Hiibi
     Kyodo Smelting Company's Tamano Smelter during the week of March 10,
     1980.  Pacific Environmental Services, Inc.
23.
24.
25.
26.
Chetvcov, V.A., et al.  Ventilation in the Nonferrous Metallurgy.
Moscow.  Metallurgia Publishing Organization.  1968.  p. 72.

Blair, T.R., et al.  Electric Arc Furnace Fume Control at Lone
Star Steel Company.  Paper presented at the 71st Annual meeting
of the Air Pollution Control Association.  June 25-30, 1978.
p. 3.
Baturin, V.V.  Fundamentals of Industrial Ventilation.
by O.M. Blunn.  Pergamon Press.  Third Edition.  1972.
Translated
Grassmuck, G.  Applicability of Air Curtains as Air Stopping and
Flow Regulators in Mini Ventilation.  C.I.M. Bulletin.  No. 691.
62:1175-1185.  November 1969.
                                3-91

-------
27.  Correspondence from Mr. Moto 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.  EMB 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
     No. 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.
                                3-92

-------
39.  Reference 36, p. 14-17.

40.  Rooney, Thomas, TRW Environmental Engineering Division.  Emission
     Testing of Phelps-Dodge Copper Smelter, Ajo, Arizona.  U.S.
     Environmental Protection Agency.  EMB Report No. 78 CUS-11.  EPA
     Contract No. 68-02-2812.  Work Assignment No. 15.  March 1979.
     p. 5-6.

41.  Reference 15, p. 4.

42.  Reference 29, p. 4-66.

43.  Vervaert, A. and J. Nolan, U.S. Environmental Protection Agency
     and Puget Sound Air Pollution Control Agency.  Log Book Nos. 1
     and 2, Observations of Converter Secondary Hood Test at ASARCO,
     Tacoma.  January 18-22, 1983.

44.  Reference 22, p. 3.

45.  Reference 37, p. 7-8.

46.  Reference 37, p. 4-5.
                                3-93

-------

-------
        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 high arsenic feed.  This section defines the
model plants, presents the alternative ways that EPA can regulate
arsenic emissions from the affected sources at these primary copper
smelters, and defines the regulatory baseline for analysis of the
environmental, economic, and energy impacts of the regulatory alternatives
on the industry.
4.1  MODEL PLANTS
     As discussed in Section 1.1, only the ASARCO-Tacoma smelter has
been categorized as a smelter which processes materials containing
high concentrations of arsenic.  Annually, this smelter processes a
greater quantity of arsenic in its feed material than the combined
total at the other 14 smelters.  Since the ASARCO-Tacoma smelter is
the only smelter under consideration in this study, a single model
plant representative of the ASARCO-Tacoma smelter was developed.
Site-specific parameters associated with the copper smelting
configuration for this smelter are presented in Table 4-1.
4.2  BASELINE
     This document evaluates the technical, environmental, energy,
cost, and economic impacts of regulatory alternatives for the control
of arsenic from high arsenic throughput primary copper smelters.  These
are measured as incremental changes beyond the situation which would
exist in the absence of NESHAP regulations for arsenic.  These conditions
                               4-1

-------
                       Table 4-1.  MODEL PLANT PARAMETERS
                   Parameter
      ASARCO-Tacoma
1.  Production capacity (blister
    copper), Mg/yr (tons/yr)

2.  Raw material:

     Materials used
     Maximum total  arsenic content
          of feed (concentrates, ores,
          speiss, flux), percent wt
              kg/hr (Ib/hr)


3.  By-products
4.  Process equipment:

      •  Roasters: No. and type



                   Dimensions
                   Feed capacity
                   per roaster,
                    Mg/day
                    (tons/day)

      •  Smelting furnaces:

                   No. and type

                   Dimensions
                   Feed capacity
                   per furnace,
                    Mg/day
                    (tons/day)
     91,000 (100,000)
 copper concentrates,
 lead smelter by-
 products, smelter
 reverts, and others
           4.0
        991 (2,185)


 Arsenic trioxide,
 metallic arsenic,
 sulfuric acid,
 liquid S02
 6 Herreshoff and 4
 C & W multi-hearth
 roasters

 5.79 m(19 ft)  diameter
 by 7.61 m (25  ft) high,
 294 m2 (3,160  ft2) hearth
 area
 1,087
(1,200)
 2 reverberatory furnaces

 No.  1  furnace:
     9.14 m by 33.53 m
     (30 ft by 110 ft)

 No.  2 furnace:
     9.75 m by 35.53 m
     (32 ft by 110 ft)
 1,087
(1,200)
                                4-2

-------
                  Table 4-1.   MODEL PLANT PARAMETERS (Concluded)
              Parameter
ASARCO-Tacoma
         Converters:

                   No.  and type


                   Dimensions
      •  Anode Furnaces:
                   No.  and type
      •  Arsenic Plant:
                   Process Equipment
4 Pierce-Smith
converters

3 converters:
3.9 m diameter by
9.14 m length
(13 ft diameter by
30 ft length)

1 converter:
3.35 m diameter by
7.93 m length
(11 ft diameter by
26 ft length)
2 Tilting Anode
furnaces, 150 tons/
charge
3 Godfrey
Roasters

3 Arsenic
Trioxide
settling
kitchens

2 arsenic
metallic furnaces
and condensers
 Normally six or seven used at a time.
 Normally one converter will  be on copper blow.   One converter is on standby or is
 used as a holding furnace.
°To be enlarged.
 Normally one anode furnace is used as a holding furnace.
                                     4-3

-------
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 ASARCO-Tacoma smelter specifically.
4.2.1  Regulatory Considerations
     The ASARCO-Tacoma smelter is subject to existing Federal, State,
and local regulations for suspended particulates, sulfur oxides, and
lead air pollutant emissions; wastewater effluent limitations; 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.  Each
regulation must be evaluated to determine whether it will influence
the engineering, environmental, energy, or economic conditions which
will prevail in the industry when an arsenic NESHAP becomes effective.
To specify a regulatory baseline from which to analyze the impacts of
the proposed standards, it is necessary to review all of the major
environmental, health, and safety regulations of the U.S. Environmental
Protection Agency and the Occupational Safety and Health Administration
which are applicable to the primary copper smelting industry.
     Regulations to be examined in formulating the baseline conditions
include the following:
     •    National Ambient Air Quality Standards under the Clean Air
          Act (CAA)
             • Sulfur Dioxide
             • Total Suspended Particulates
             • Lead
     •    Regulations (for Inorganic Arsenic) of the Occupational
          Safety and Health Administration (OSHA)
     •    Effluent limitation guidelines under the Clean Water Act
          (CWA)
     •    Hazardous waste disposal regulations under the Resource
          Conservation and Recovery Act (RCRA).
     •    Potential actions under the Comprehensive Environmental
          Response, Compensation, and Liability Act (CERCLA)
                               4-4

-------
      For the purpose of relating the timings of the regulatory requirements
 cited above to the timing of an arsenic NESHAP, January 1986 is projected
 as the date of compliance with the arsenic NESHAP.  The date is based
 upon a January 1984 promulgation date 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.
      4.2.1.1  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, particulate matter, 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."1
      The  State of Washington SIP has  designated that  the Puget Sound
 Air Pollution  Control Agency (PSAPCA)  has jurisdiction over ASARCO-Tacoma
 smelter air  emissions.2 The PSAPCA Regulation  I  sets  forth  both
 ambient air  quality standards  and  emission standards.
      4.2.1.1.1   Sulfur  dioxide.   Sulfur  dioxide (S02)  emissions from
 the ASARCO-Tacoma smelter  are  covered by Sections  9.07(b)  and  9.07(c)
 of  PSAPCA Regulation  I, which  limit S02  emissions  to a  concentration
 of  2,000  ppm or  less  and limit the  sulfur  content  of emissions  to
 10  percent or  less of the  sulfur contained  in the  process weight per
 hour.   ASARCO submitted an  application  to  the  PSAPCA  Board  of Directors
 for  a variance from these  regulations on  December  5, 1975.   The application
 was  approved on  February 19, 1976,  granting a variance  until December  31,
 1980.  An amended variance application was submitted on August 12,
 1980, in  order to reflect  changes at the smelter,  and  to request an
 extension of the variance  through 1982.  A court decision issued on
 January 3, 1980, required  compliance with  the Washington State
 Environmental Policy Act (SEPA) for approval of a  variance or variance
 extension.  PSAPCA's Board of Directors therefore  granted ASARCO an
 interim variance on April 10, 1980, which was in effect at the time  of
 the August 12,  1980, request for an extension, and will be extended
until the SEPA  process is completed.

                               4-5

-------
     Section 119 of the Clean Air Act authorizes nonferrous smelter
orders (NSO's) to delay the requirement for SCL emission controls.
EPA issued final regulations for initial NSO's on June 24, 1980.
PSAPCA has the primary responsibility for developing an NSO for
ASARCO-Tacoma, with review of the Washington State Department of
Ecology (DOE) and EPA.  By a settlement agreement between EPA and
ASARCO dated June 7, 1979, ASARCO has agreed to apply for an NSO for
the ASARCO-Tacoma smelter, but completion of the application' can be
delayed until the final Supplementary Control System (SCS)'regulations
under Section 123 of the Clean Air Act are issued.  ASARCO filed a
                                                        2
letter of intent on August 4, 1980, to apply for an NSO.
     As of March 1983, SCS regulations have been prepared but not yet
proposed.  When final regulations are issued, PSAPCA and the Washington
State DOE will determine the amount of sulfur emission reduction
credit to be received by ASARCO for operating its supplementary control
system.  The outcome of this determination will be the deciding factor
in whether the Puget Sound Intrastate Air Quality Control Region is
classified as an attainment or nonattainment area for SOp.  If a
nonattainment determination is made, ASARCO will be required to apply
for an NSO.  This would permit ASARCO to potentially delay compliance
                                                                4
with the SIP for S02 emission limitations until January 1, 1993.
     In view of the uncertainties associated with the future course of
the SCS and NSO programs with regard to ASARCO-Tacoma, it is difficult
to predict the resulting S02 control requirements.  However, since the
Tacoma smelter currently utilizes control systems on process gas
streams which generally reflect the best control technology for arsenic,
the application of additional controls for S02 is expected to have a
negligible impact on current arsenic emissions from process sources.
     4.2.1.1.2  Total suspended particulates.  TSP emissions from the
                                                          3
ASARCO-Tacoma smelter are governed by PSAPCA Regulation I,
Sections 9.03(b), 9.09(c), and 9.09(d).  Section 9.03(b) requires that
the opacity of air emissions be 20 percent or less.  Section 9.09(c)
is a process weight regulation, while 9.09(d) is an absolute emission
limit of 0.10 grains per standard cubic foot of exhaust gas.
                               4-6

-------
     ASARCO  requested  a variance from Section 9.03(b)  as part of the
 variance  application discussed  for S02 emissions (see  Section 5.2.1.1.1).
 The  Tacoma smelter  is  in compliance with  Sections 9.09(c)  and 9.09(d).5
 Compliance with Section 9.09(c),  the process  weight regulation,  requires
 the  use of cold control  devices  on all  process gas streams at the
 ASARCO-Tacoma  smelter  due  to  the  high concentration of condensable
 constituents  (such  as  arsenic)  in the process offgases.
     4.2.1.1.3  Lead.   The  national  ambient air quality  standard for
 lead of 1.5/jg/m  was  promulgated on October  5, 19.78 (40 CFR 51.80).
 PSAPCA has also promulgated an  identical  standard (Section 11.05 of
 Regulation I).  As  of  March 1983, the State of Washington  had not
 submitted an SIP for lead,  though one was  in  the final stages of
 preparation.   Ambient  monitoring in the  vicinity of the Tacoma  smelter
 has not revealed any violations  of PSAPCA's AAQS for lead,  however.5
 It is thus very unlikely that ASARCO-Tacoma will  be required to  implement
 additional emission control measures  under the upcoming  SIP.
     4.2.1.2  Arsenic  Regulation  by  the Occupational Safety  and  Health
 Administration.  On May  5,  1978,  the U.S. Occupational Safety and
 Health Administration  promulgated  standards for occupational  exposure
 to inorganic arsenic.    The standards  limit occupational exposure  to
 10jug/m , 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 standard is in the  form of an ambient  concentration
 limit for the working environment  at  the smelter  and does  not specify
 the sources to be controlled nor  the  control   technology which must  be
utilized.   ASARCO has recently signed a tripartite agreement with OSHA
and the United Steel workers of America which  "sets forth feasible
controls and  practices  to protect  the smelter  employees from inorganic
arsenic."   The terms of the agreement require  ASARCO-Tacoma to maintain
                               4-7

-------
effective capture systems for fugitive emissions from calcine discharge
and matte and slag tapping operations, and to install secondary hoods
over the three large converters by July 1, 1984.  The agreement does
not, however, include technical specifications for the converter
secondary hoods.  Moreover, neither operating practices nor maintenance
requirements for the secondary hoods are specified in the agreement.
There are no provisions in the agreement for collection of captured
fugitive emissions.  Although the present analysis recognizes OSHA's
requirement for capture of calcine discharge and matte and slag tapping
fugitive emissions, capture of converter fugitive emissions is not
assumed in the baseline case since the OSHA agreement is not specific
as to the technology, performance criteria, or work practices to be
employed for the capture of converter fugitive emissions.
     4.2.1.3  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 ASARCO-Tacoma
smelter will  be regulated by EPA's proposed effluent limitations
guidelines for nonferrous metals manufacturing.   Effluents from the
ASARCO-Tacoma smelter 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)
     As of March 1983, the ASARCO-Tacoma smelter was in compliance
with existing Federal water pollution regulations, and continued
                                                            5
compliance was not expected to affect arsenic air emissions.
                               4-8

-------
      4.2.1.4  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 identified the acid
 plant blowdown  slurry/sludge  resulting  from  the thickening of blowdown
 slurry at  primary  copper  smelters  as hazardous waste because of lead
 and  cadmium  content.   As  of early  1983,  no further regulations which
 could be applied to ASARCO-Tacoma  have  been  proposed or promulgated
 under RCRA.   The present  and  projected  impact of RCRA regulations  on
 the  ASARCO-Tacoma  smelter, both in terms of  arsenic air emissions  and
                                               Q
 economic burden, is believed to be negligible.
      4.2.1.5  Comprehensive Environmental  Response,  Compensation,
 and  Liability Act  (CERCLA)..  On December 30,  1982 (47 FR 58476),  the
 Commencement Bay-Near  Shore Tide Flats  area,  which includes the ASARCO-Tacoma
 smelter, was  placed by EPA on  the  National Priorities List (NPL),  a
 supplement to the  National Oil and Hazardous  Substances  Contingency
 Plan  (NCP), which  was  promulgated  on July  16,  1982 (47 FR 31180),
 pursuant to Section 105 of the Comprehensive  Environmental  Response,
 Compensation, and  Liability Act of 1980  (CERCLA),  and Executive Order
 12316.  The NPL identifies priority  releases  of hazardous  wastes,
 based  on the assessments  of State  governments  and  EPA,  for Fund-financed
 remedial  action under  CERCLA (also known as  "Superfund").
      As of early 1983, no determination has  been  made concerning
 possible actions at the Tacoma smelter under  the  Superfund  program.
 It is  therefore impossible to  predict with any degree of certainty
what  economic impact,  if any,  Superfund activities will  have  on
ASARCO-Tacoma.  It is  unlikely, however, that  arsenic air  emissions
from  the smelter will   be affected  by  the Superfund  program.9
      4.2.1.6  Regulatory baseline  summary.   This  section  summarizes
the pre-existing set of regulations  that form  the  baseline  from which
the incremental impacts of the arsenic regulatory  alternatives  can  be
determined.  Table 4-2 summarizes  the existing  regulations  which
affect arsenic air emissions at the ASARCO-Tacoma  smelter.  The effects
of the regulatory baseline can be  described as  either  technical,
meaning they result in either a reduction in arsenic  emissions  or a
                               4-9

-------
              Table  4-2.    SUMMARY OF EXISTING REGULATIONS  AFFECTING ARSENIC
                                  AIR  EMISSIONS AT ASARCO-TACOMA
Source
Roaster Process
TSP Emissions
Reverb Process
TSP Emissions
Converter & Anode
TSP emissions0
Arsenic Plant
TSP Emissions
Calcine Discharge
Fugitive Emissions
llatte/Slag
Tapping Fugitive
Emissions
Converter Operation
Fugtive Emissions
Capture
Regulation Requirement
Hot Applicable
Not Applicable
Not Applicable
Mot Applicable
OSHA Compliance Plan Maintain effective
capture system
OSHA Compliance Plan Maintain effective
capture system
OSHA Compliance Plan Install
secondary hoods
Collection
Regulation Requirement
PSAPCA Ia Opacity of air emissions
Section 9.03 (b) . must be 20 percent or
less.
PSAPCA I Process weight regulation.
Section 9.03(c) Under typical operating
conditions, allowable
emissions at the roaster
baghouse outlet are
52.7 Ibs/hr. Corresponding
As concentration is
0.01 gr/scf.
PSAPCA I Maximum emission limit
Section 9.03(d) of 0.1 gr/scf.
Same as for Under typical operations,
roaster 9.03(c) (process weight
regulation) limits
emissions from reverb
ESP (#1) outlet to
52.1 Ib/hr, corresponding
to 0.04 gr/scf As.
Same as for Under typical operations,
roaster 9.03(c) (process weight
regulation) limits
emissions from converter
ESP outlet (#2, assumed
full bypass of acid/S07
plants) to 51 Ib/hr, i
corresponding to 0.02 gr/scf
As.
Same as for Under normal operating
roaster conditions, 9.03(c) (process
weight regulation) limits
emissions from the arsenic
baghouse to 4 Ib/hr,
corresponding to 0.02 gr/scf
As.
No collection requirement
No collection requirement
No collection requirement
*Puget Sound Air Pollution Control Agency.  Regulation I.  January 1983.

 Assuming worst case situation, i.e., complete bypass of acid/SO- plants by converter  offgases.
 with anode furnace offgases and treated  in the #2 ESP.


e«o specification of technical, operating, or performance criteria.  See Section 4.2.1.2.

TSP « Total Suspended Particulates.
Offgases would be combined
                                                        4-10

-------
 change  in  the  manner in  which  arsenic emissions  are discharged to the
 atmosphere,  or economic,  in  that  the  compliance  costs of other regulations
 would affect the  affordability of the regulatory alternatives.  Such
 technical  and  economic effects are summarized  in Table 4-3.
     Under the regulatory baseline, the control  technology required
 includes local  hooding for capture of fugitive emissions from calcine
 discharge  and  matte  and  slag tapping; and  a  baghouse, or equivalent
 technology,  operated  at  110°C  (230°F) or lower on all  roaster, smelting
 furnace, and converter process gas streams.
 4.2.2.  Baseline  Arsenic  Emissions
     A  description of copper smelting processes,  operations,  and
 equipment  at ASARCO-Tacoma was  presented in  Section  2.0.   Potentially
 significant  sources  of arsenic emissions were  identified.  These
 include sources of both process and fugitive emissions  from roasters,
 smelting furnaces, and converters.  Specific fugitive  emission sources
 identified include calcine discharge  operations  at multi-hearth roasters,
 smelting furnace  matte and slag tapping  operations,  converter operations
 (charging, blowing, holding, and  pouring), anode  furnaces, dust handling
 and transfer,  and the arsenic  plant.
     Baseline  arsenic emissions,  as presented  in  this  section,  refers
 to arsenic emissions from the ASARCO-Tacoma smelter  under  existing
 control.   Existing process emitting equipment  is  described in  Section 2.1.3,
 while a discussion of existing air pollution control equipment at
 ASARCO-Tacoma  is  contained in Section 3.2.
     4.2.2.1  Baseline Arsenic Emission Calculations.  Baseline arsenic
 emissions from individual process  and fugitive sources at  the  Tacoma
 smelter were presented in Section  2.0  (Tables 2-4 and 2-11, respectively).
 These figures were derived from the arsenic material balance  for
ASARCO-Tacoma  (Figure 2-6), employing the emission factors developed
 in Section 2.2.2 for fugitive emissions, and the estimated efficiencies
 of existing control  devices based on test data presented in Section 3.3.
 In order to estimate total baseline arsenic emissions on an annual
 basis,  a factor of 8,600 hours  of operation per year was assumed.
Based on this assumption, total potential process arsenic emissions
are estimated to be 148 Mg/yr,  while total  potential  fugitive arsenic
                               4-11

-------
  Table 4-3.  SUMMARY OF EFFECTS OF REGULATORY BASELINE ON ARSENIC
        REGULATORY ALTERNATIVES FOR THE ASARCO-TACOMA SMELTER

Regulation
S02 - NAAQS
TSP -NAAQS
Lead -NAAQS
OSHA
CWA
RCRA
CERCLA
Technical Effects9 Economic effects
None likely
None likely
None
Capture and dispersion
of roaster, smelting
furnace, and converter
fugitive emissions
None
None
None
Unknown
None 1 i
None
Signifi
None
None
Unknown

kely

cant



Technical effects refer to alterations in the quantity or manner of
arsenic emissions resulting from engineering practices utilized for
compliance with baseline regulations.
                              4-12

-------
emissions amount to 134 Mg/yr, yielding a total estimated potential
baseline arsenic emission rate of 282 Mg/yr.'
4.3  REGULATORY ALTERNATIVES
     Alternative techniques for the control of both process and fugitive
arsenic emission sources were identified and discussed in Section 3.0.
A summary of the candidate systems selected as bases for regulatory
alternatives and the rationale for this selection follows.
4.3.1  Process Emission Control Techniques
     Effective techniques for the control of arsenic emissions contained
in the process gases produced during roasting, smelting, and converting
were described in Section 3.1.1.  The control  technique identified as
best consisted of the application of a baghouse or an equivalent
control device, such as an electrostatic precipitator preceded by gas
stream cooling, or a high energy venturi scrubber.  Sulfuric acid
plants also result in effective arsenic control due to the need to
incorporate several of the above devices for gas precleaning purposes.
Existing process emission controls at ASARCO-Tacoma are described in
Section 3.2.
     4.3.1.1  Roaster Process Emission Controls.  Roaster offgases at
the Tacoma smelter are cooled to a temperature less than 120°C (250°F)
and treated in a baghouse system.  Simultaneous inlet and outlet
arsenic emission measurements were performed by EPA across the baghouse
serving the roasters at ASARCO-Tacoma, as described in Section 3.3.1.1.1.
The average arsenic removal efficiency for this unit, as indicated by
the test results, was 99.7 percent.
     4.3.1.2  Reverberatory Furnace Process Emission Controls.  Furnace
offgases at the ASARCO-Tacoma smelter are cooled to a temperature of
132°C (270°F) before treatment in an electrostatic precipitator system.
EPA performed outlet arsenic emission measurements at the electrostatic
precipitator, as described in Section 3.3.1.2.1.  The test results,
combined with an estimate of inlet arsenic loading from the arsenic
material balance (Section 2.2.1) yield an estimated arsenic removal
efficiency of 98 percent for the reverberatory furnace electrostatic
precipitator.
                               4-13

-------
     4.3.1.3   Converter  Process  Emission  Controls.   Converter offgases
at ASARCO-Tacoma are  treated  in  the  gas cleaning  circuit  associated
with either the liquid S02  plant or  a  single-contact sulfuric acid
plant for  particulate removal, as  described  in  Section  3.2.1.   The gas
cleaning circuits of  the  SCL  control devices  consist of a water spray
chamber, electrostatic precipitator, scrubbers, and  mist  precipitators
in series.  Occasionally, when converter  offgas flows exceed  the
capacity of the S02 control system,  some  amount of the  offgases bypasses
the system and is treated in  a "cold"  (150 to 200°F)  electrostatic
precipitator for particulate  matter  removal  before discharge  through
the main stack.
     Section 3.3.1.4  describes EPA's evaluation of the  arsenic  removal
performance of various acid plant  systems at  domestic copper  smelters.
Based on these findings,  an arsenic  removal  efficiency  of 99  percent
was ascribed to the S02 control  devices at ASARCO-Tacoma.
4.3.2  Fugitive Emission  Control Techniques
     A variety of techniques  is  available for controlling  arsenic
emissions from fugitive sources  located at copper smelters.
Section 3.1.2 presents background  information on  fugitive arsenic
emission control techniques.  Reduction of fugitive  arsenic emissions
first requires capturing the  emissions.   Once captured, the emissions
may be vented directly to a control  device for collection  or  combined
with process gases prior to collection in a control  device.
     4.3.2.1  Roaster Fugitive Emission Controls.  Of the  systems
examined for the control  of fugitive emissions from  multi-hearth
roaster calcine transfer operations, only the use of a  larry  car shed
or tunnel coupled with local  ventilation  applied  at  the calcine hopper-
to-larry car transfer point was  judged to be  effective.   Details of
the system, which is presently used at the ASARCO-Tacoma  smelter, were
described in Section 3.2.2.   The performance  evaluation was detailed
in Section 3.3.2.1.   Visual observations were made across  the shed
opening.  No visible emissions were  observed  at any  time  during the
observation period.   Based on these observations, the system was
judged to achieve a capture efficiency of 90  percent or better.  The
captured fugitive emissions are  combined with the roaster  process
offgases and treated in the roaster baghouse  for  particulate  removal.

                               4-14

-------
     4.3.2.2  Reverberatory Furnace Fugitive Emission Controls.  Local
ventilation hoods are used at ASARCO-Tacoma to capture fugitive arsenic
emissions from furnace matte and slag tapping operations, as described
in Section 3.2.2.  Fixed hoods are placed over tapping ports.  At
ladle and pot filling points, retractable hoods are used.  Based on
visual observations, detailed in Section 3.3.2.1, capture efficiencies
in excess of 90 percent are achievable using local ventilation hoods.
Captured emissions from both systems are combined with anode furnace
offgases and treated in an electrostatic precipitator at 66 to 93°C
(150 to 200°F) before being discharged through the main stack.
     4.3.2.3  Converter Fugitive Emission Controls.  Fugitive arsenic
emissions from converters can be captured by using local ventilation
techniques.  Local ventilation techniques evaluated consisted of using
mechanical hoods, air curtain systems, or a combination of these two
(refer to Section 3.1.2.7).  Capture efficiencies for fixed mechanical
hoods are estimated to be 80 percent or less.
     The prototype air curtain secondary hood system recently installed
on converter No. 4 at ASARCO-Tacoma is described in Section 3.1.2.7.2.
Captured emissions are presently routed to the anode furnace and matte
and slag tapping electrostatic precipitator for particulate matter
removal.  In a test program at the smelter, the overall average capture
efficiency of the air curtain system was determined to be about 95 percent
(refer to Section 3.3.2.2.1).
     The existing prototype air curtain secondary hood on converter
No. 4 is the first of three such installations planned by ASARCO at
its Tacoma plant.  As such, the prototype system has been used to
collect performance data and refine design and operation parameters.
Since this system is in what may be called a "shakedown" or optimization
phase, it is not included in the baseline.  There is at present no
Federally enforceable regulation which requires the installation and
continued operation of a device such as the prototype air curtain hood
system in place on the No. 4 converter.
     Electrostatic precipitators, baghouses, or wet scrubbers can be
used to collect arsenic emissions (refer to Section 3.1.1).  For the
regulatory alternatives, any of the above devices, well-maintained and
                               4-15

-------
operated at 120°C (250°F) or less, qualifies for control of the affected
facilities.  This selection is based on source test data which show
that at least 96 percent arsenic collection efficiency can be achieved
by such devices.
     For the purpose of the regulatory alternatives, the fugitive
arsenic emission control technique considered for converter operations
is emission capture by using an air curtain secondary hood, and
collection of the captured emissions by a well-maintained cold particulate
control device as described above.
4.3.3  Regulatory Alternative I
     Regulatory Alternative I represents no regulatory action (baseline)
and would require no additional controls beyond existing process and
fugitive emission controls as described in Section 3.0, for ASARCO-Tacoma.
It should be noted that although converter No. 4 is presently equipped
with a prototype air curtain secondary hood vented to a cold
electrostatic precipitator, Regulatory Alternative I gives no credit
for its existence.
4.3.4  Regulatory Alternative II
     Regulatory Alternative II includes all controls specified by
Regulatory Alternative I, plus air curtain secondary hoods on three
converters for capture of converter fugitive emissions, and collection
of captured emissions using a baghouse or equivalent technology.  For
the purposes of cost analysis in Section 6.0, an existing ESP ("#2")
is assumed to be used for collection of captured converter fugitive
emissions.
4.3.5  Regulatory Alternative III
     Regulatory Alternative III would require that arsenic emissions
from the ASARCO-Tacoma smelter be reduced to zero.  Accomplishment of
this alternative would require the ASARCO-Tacoma smelter to process
ores which were virtually free of arsenic content.  Implementation of
Regulatory Alternative III would therefore result in closure of the
ASARCO-Tacoma smelter.
                               4-16

-------
4.5  REFERENCES
 1.   U.S.  Environmental  Protection Agency.  Proposed Rules for Primary
     Nonferrous  Smelter Orders.  Federal Register, Vol. 44-6284.
     January 31, 1979.   p.  1858.

 2.   Puget Sound Air Pollution Control  Agency.  Final Environmental
     Impact Statement for ASARCO,  Incorporated (Summary), September 1980.

 3.   Puget Sound Air Pollution Control  Agency.  Regulation I.  January
     1983.

 4.   Telecon.  Whaley,  G.,  Pacific Environmental  Services, with Hooper,
     M., EPA Region  X.   Federal Environmental  Regulations Applicable
     to  the ASARCO-Tacoma Smelter.  March 24,  1983.

 5.   U.S.  Department of  Labor, Occupational  Safety and Health Administration,
     General  Industry Standards, OSHA Safety and  Health Standards
     (29 CFR 1910),  and  OSHA  2206, Revised November 7, 1978.

 6.   Occupational  Safety and  Health Administration.   Engineering
     Assessment  and  Proposed  Compliance Plan for  ASARCO Inc.   Tacoma,
     Washington.   January 1982.

 7.   U.S.  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.

 8.   Telecon.  Whaley, G_,  Pacific Environmental  Services,  with Hofer,
     G., EPA Region  X.   Federal  Environmental  Regulations  Applicable
     to the  ASARCO-Tacoma Smelter.   March  30,  1983.

 9.   Telecon.  Whaley, G.,  Pacific Environmental  Services,  with Davoli,
     D., EPA Region  X.   Federal  Environmental  Regulations  Applicable
     to the  ASARCO-Tacoma Smelter.   March  28,  1983.
                              4-17

-------

-------
                       5.0  ENVIRONMENTAL IMPACTS

5.1  INTRODUCTION
     The environmental impacts on air, energy consumption, solid
waste, and water associated with the regulatory alternatives for the
control of arsenic emissions from high arsenic throughput smelters
are presented in this chapter.  The ASARCO smelter in Tacoma, Washington,
is the only U.S. smelter in the high arsenic throughput copper smelter
category.  Therefore, the impacts presented in Section 5.0 represent
those for the ASARCO-Tacoma smelter.  The purpose of this analysis is
to determine the incremental change in air pollution, water pollution,
solid v/aste, and energy impacts of the regulatory alternatives over
the baseline control  level.  The baseline control level reflects
existing levels of control of arsenic emissions at the ASARCO-Tacoma
smelter.  This level  is represented by Regulatory Alternative I.
     Section 5.1 addresses the air pollution impacts of implementing
each of the regulatory alternatives.  Water pollution, solid waste,
and energy impacts for the regulatory alternatives are addressed in
Sections 5.2, 5.3, and 5.4, respectively.
5.2  AIR POLLUTION IMPACTS OF REGULATORY ALTERNATIVES
     The air pollution impact associated with each of the regulatory
alternatives considered is presented in this section.  Incremental and
cumulative arsenic emission reductions are discussed for the ASARCO-Tacoma
smelter.  The emission estimates are obtained based on the application
of capture and collection control systems selected in Section 3.3 as
the bases for the regulatory alternatives.
5.2.1  Baseline Emissions
     The baseline regulatory alternative for ASARCO-Tacoma includes
effective process controls on roasters, furnaces, converters, and the
arsenic plant, and local ventilation capture systems for fugitive
emissions from roasters and furnaces.  Controls for process emissions
                                   5-1

-------
at the ASARCO-Tacoma smelter include exhaust gas cooling by tempering
air and baghouse collection for the roasters; gas cooling by air
infiltration and evaporative cooling followed by electrostatic precipitators
for the reverberatory furnace; a baghouse for arsenic plant emissions;
and electrostatic precipitators and scrubbers for particulate removal
prior to treatment in the sulfuric acid plant or SCL plant for the
converters.  Process emissions from the anode furnace are captured and
vented to an electrostatic precipitator.  Collection efficiencies for
these process control systems and the resultant baseline process
arsenic emission rates are summarized in Table 5-1.  Fugitive emissions
from the roaster calcine discharge operation are combined with the
roaster process gases and treated in a baghouse for particulate removal.
Fugitive emissions from furnace matte tapping and slag tapping operations
are captured by local ventilation, combined with reverberatory furnace
offgases, and collected in an electrostatic precipitator.  Fugitive
arsenic emissions from converter and anode furnace operations are not
controlled.  Fugitive arsenic emissions from the arsenic plant are
controlled by means of dust control work practices.  Fugitive arsenic
emissions from the flue dust handling system are controlled by means
of dust-tight transfer systems.  Capture and collection efficiencies
for fugitive emissions control systems and the estimated baseline
fugitive arsenic emission rates are summarized in Table 5-2.
     Baseline arsenic emissions, presented in Tables 5-1 and 5-2, were
calculated by determining the potential and controlled arsenic emission
rate for each process and fugitive emission source.  Potential arsenic
emission rates were determined based on the arsenic material'balance
for the smelter (Figure 2-6) and emission factors presented in Section 2.2.
Controlled arsenic emission rates were determined by application of
the estimated capture and collection efficiencies of the baseline
emissions control systems described above to each process and fugitive
emissions source.  Capture and collection efficiencies of the control
equipment are based on control equipment performance test results presented
in Section 3.1.  Potential arsenic emissions were calculated based on an
arsenic feed rate of 941 kg/hr to the smelter.   This arsenic enters the smelter|
in the copper ore concentrate and other copper-bearing materials processed
                                   5-2

-------
   Table  5-1.   PROCESS  EMISSION  SOURCES,  CONTROL  EFFICIENCIES,
                               AND  ARSENIC EMISSION  RATES9
Process
Emission
Source
Roasters
Reverberatory
Furnaces
Converters
Arsenic Plant
Anode Furnace
Total
Potential
Arsenic Control
Emissions Control Efficiency
(kg/hr) Device (Percent)
255 Baghouse
608 ESP
ESP
94 ' ESP/Acid Plant
ESP/S02 Plant
373 Saghouse
0.4 ESP
1,330
99.8
98.0
96.0
99.9
99.9
98
96

Basel ine Arsenic
Emission Rate
(kg/hr)
0.4
9.5b
0.04d
' 7.3.
0.02
17.3
From Reference 1.

Of the potential  arsenic emissions  that leave the  process, a portion is collected from
the flue gas handling system as dust fallout prior to entering  the control  device
indicated.

Process offgases  from the converters are separated into three separate streams, each
directed to the control device(s) shown.

The arsenic emission rate shown represents the sum of controlled emissions  from the three
control systems through which the converter offgases pass.
                                        5-3

-------















oo
UJ

CJ
UJ
*— «


iH
1 1
UJ


0
t— i OO

0 0
UJ H— f
_1 00
	 I OO
o •-«
cj s:
UJ
0
2: cj
«=c >— *

UJ UJ
c; oo

£<
«=C UJ
CJ >


"r—

e; CD
O O


CJ
e£ UJ
U— ^^
1— <

1— « 1 1 1
oo \—
OO UJ

yf
UJ O

UJ
> o
i— < UJ
h- oo

CD
1 T





CM

in



IQ


h—




















u

C QJ
Q) 4->
in re
S- Ci- — •
<: t
C J=
o o -»-
c ••- cn
t- in j£
r— in — •
QJ -r-
in E
re LU
CCt




r— r— O 4J
r— O C C
re i- QJ QJ
i. 4-> — O
0) C O S-
> 0 -i- 0)
t] . -
UJ


14-
o
c <-~
>> O 4J
0 ••- E C
CT 4-> QJ QJ
OJ O 4-> o
t- oj in s-

•r- i — OO C-
i[ (_j ^~~
LU



14-
o
>•) OJ E 4-J
O 1- OJ C
C 3 4J QJ
OJ 4J in o
u re oo QJ
•— o n

u-
LU



i — in
re = .— .
~- u o s-
c c in — .
QJ cj to cn
4-»  1-J*
O s- E^
0-< UJ









in
QJ C
> O S-
•i- — O
•~- in o
cn-i— re
3 E U-
U- LU















in

o

in
in

'i QJ
LU U

QJ 3
> O


cn
3
LL.








CO
O
o










CO

en
CO







re
CO
en
cn









o
en











CO
O



OJ
c

^
re

c

u

c
QJ
in
^

'i
o


CO
o
Q



S-

14.
in
c:
re

1—

OJ
c.

0

re



Q)

in
re
o
ei























































in
OJ
O



Q)
re
c
re 3
i — U_
oo
= £>
•r- O
4_)
U re
•r- L,
C QJ
in L-
S- 01

'1 r>^
O
c

e>j s-

o"






cn
c:

CL

re
i —

cn
re

on

QJ
0
re
£
rs
LJ_


in
c
O

00
in
1-1 CO T-
"i ^ °. ° ^ "^ ^ s
o 0*3-0^010
~* ^ OJ
u
re
c

3
l^.

*C
O •
c -o
re 4)
«sf- JD ^ (_> re o
o o o o o -r- s_
VO CT* CTl • +J 4->
CO C C
CJ o
4J U
o
Q. 4_> 00
O U- OJ
*-^. CO U
QJ T—
> +— O >^ •$£
cn •-- JD s-
in w o
3 in -a S
•— ••- OJ
^ E f™" t-1--
a> .— o
tn o i.
tn cu S- 4-J
QJ > 4J C
LJ i- c: o
O *J O U
s- ••- u
a. 01 4^
**— - 3 *J tn
^ C 3
.— QJ -O
.jD _Q EJ -QOJ rcaju
o ooo 4->wi-U-
cr\ o QJ QJ o
' *J J7 p_
1— in
irt O C
4-> CH ro
C • OJ
QJ in QJ E
in c JO
a> o >>
I- T- O J3
Q- irt -w
QJ W T3
S- -^- *O QJ
.— t CO E QJ r—
CO C? C? CO P"*" "^ *J>^ QJ QJ +^ r—
• *.*..* u re o
PO CD '^ C3 CVJ IO P*« • rtJ QJ £ t-
i— C CSJ QJ C > ••— 4->
W S- •*- -«-> C
Z3 3 4-> tn O
in o <+- i- QJ u
s- • c JT cn
QJOJ o CFIQJSOJJ-*
W 4J t. i- (C *0 M- L. C
res-Qjin o .floroOJ
C QJ JJ IrtU -r- C r— U
4-J3C QJ EU QJ QJ *r-COJ
J-J U, O > UJrer— (/> 4-> QJ 4-> O O.
(OCJ C Cr— t- (rt .C C •*-
E>^ o ot-re < to -t-JOJino
COf- C-U-C O t- OO"-
•r--M COJ'^***-1 •)-> O. 8s QJ
reo-t-uiaj*o QJ S5.0

•r-QJC U < O •»- -W *r- 4-> rJ-^CT O
C^OJ "- CCOC C-CC-M
QJ S- (/) C ,— «£ QJ a) QJ O C OJ T-
irt OJ i- Qjmre in r— in ^-* *'~in>KTD
t— > ^C m QJ •*-* QJ i— r~ J— QJ QJ QJ

c£ M- «creH-*-> cj QJOJ-^-d-c re
4_ o £2 14-. ' (*_ 4-> 4-J V) C QJ O E
OE cni*-i t*-P= o4-> OE ureojs-o >f-
o»*reo^_oc5 in re sum 4-)
>*s_n,— u- s-*«3***— -D s_in-a co
urs^t- oto >?o %< ^- »-HO und_ crecc QJ
• »Lnc> • • ajoore
C\J O i— < CO CD O l-T-OJ-p- QJ
re in ,c in » i-
*•— • en 4-> (/i 01 • re
gtn M- -f- c in
c E <*- E — = i/>
 OQJOQJp— CDC
C V) •>- -O 4-) O

cn c ' ••- oo tnccinro>,tn
•r- 3 C ro-Ol E i— O -C -r-
Q. 4-J O !- C QJ QJ S» 4-J * S_ !=
CLOJT-OJ -r- tOQJi-QJQJ
rec^j-^CL r— cn ajro cm— QJ «—
I— reo -oc >.£3OW-incn
cnt. c»f- -r- 1— i in c c
QJ ro QJ QJ wre"O 4-> c •— t 4J c re T-
4-» i^ o- t_> 3rc^-w .^«f- crei-T3
4-> l/l O re O 'f-E CD 4->OJl-J->r—
re c QJ4-J3O 3inretJ-»-)'p-
E t- i- t. cmcs-r- ^-r— Jis- 4-93
QJ QJ 3 re3 4_» O4->OJ4-Jjr.Q
QJ *-> j-j u_ ^-ourer--os_ cuwcn
u i_ t- ^— -p-s-reaj^Jv) r>*r-u
re  > -O U3GJCLOr3OEJ-) 1C
t- C C O trtf— inOh-4->U3CQJ4JCJ
3 O O C •*— Lt. t_ CX VlQJ3tnin
U_ LJ O < :r— ' •< TO O V) S r- 3 S-
e_> 2: «c f— u- -o <
rc> ^a u *o QJ
5-4

-------
  at  the  smelter.   As  shown  in  Tables  5-1  and  5-2  for the baseline case,
  the total process arsenic  emission rate  is 17.3  kg/hr  and  the total
  fugitive  arsenic  emission  rate  is 15.5 kg/hr.  Annual  arsenic emissions
  are  determined using these emission  rates and assuming 8,600  hours  of
  smelter operation per year.   The resultant annual  baseline arsenic
  emissions for the ASARCO-Tacoma smelter  are  282  Mg/yr.
  5-2-2  Arsenic Emission Reductions Under the Regulatory Alternatives
      Table 5-3 presents the estimates of annual  arsenic emissions and
 the  percentage emission reductions achievable through  application of
 the different regulatory alternatives being considered for  the ASARCO-Tacoma
 smelter.  These estimates include both total  process emissions and
 total fugitive emissions.   Annual  estimates of arsenic emissions were
 determined assuming  8,600 hours of operation per year.  The reduction
 in arsenic emissions associated with  each regulatory alternative is
 obtained by  subtracting the emission  rate for the regulatory alternative
 from the baseline emission  rate.
      Arsenic  emissions  under  Regulatory Alternative II were calculated
 by applying the  additional  controls selected  for  the alternative to
 the baseline  case.   The additional controls  include the installation
 of an air  curtain  secondary hood to each  of the three operating converters
 at the smelter to  capture the  fugitive  arsenic  emissions  discharged  by
 the  converters.   Collection of the captured fugitive arsenic emissions
 will  be  achieved  by  a refurbished, existing electrostatic  precipitator
 (ESP).  The air curtain  secondary hood  is 95  percent efficient,  and
 the  ESP collection system is 96  percent efficient.   Arsenic emissions
 under Regulatory Alternative II  are 172 Mg/yr.  This  represents  a
 reduction  of  110 Mg/yr,  or 39  percent, from the baseline arsenic
 emission level.
      Regulatory Alternative III  requires  that all  arsenic emissions
from  the ASARCO-Tacoma smelter be reduced to zero.  This represents a
reduction of  100 percent over the baseline arsenic  emission  level.
5.3  ENERGY IMPACTS OF THE REGULATORY ALTERNATIVES
     Table 5-4 summarizes the incremental  energy  requirements  of  the
regulatory alternatives  over the baseline case for the ASARCO-Tacoma
smelter.   The  additional energy requirements for Alternative II are
                                   5-5

-------
       Table 5-3.  ARSENIC EMISSIONS AND EMISSION REDUCTIONS AT
           ASARCO-TACOMA UNDER THE REGULATORY ALTERNATIVES

Regulatory Alternative

Alternative I (Baseline)
Alternative II
Alternative III
Arsenic
Emissions
(Mg/yr)
282
172
0

Arsenic Emission Reductions
(Mg/yr)
~
110
282

from Basel ine
(Percent)
—
39
100
Table 5-4.  ANNUAL ENERGY REQUIRED BY AIR POLLUTION CONTROL EQUIPMENT
          AT ASARCO-TACOMA UNDER THE REGULATORY ALTERNATIVES
     Regulatory
     Alternative
   Incremental
  Annual  Energy
Requirements from
    Baseline
    (106 kWh)
Baseline

Alternative II

Alternative III
     15

      0
                                 5-6

-------
 calculated  based  on  the  total  electrical  energy  required  by  the air
 curtain  secondary hood and  ESP collection  system for  control  of converter
 fugitive  emissions.  The electrical  energy requirements are  determined
 based ''on  the total gas flow to be  handled  by  the air  curtain  secondary
 hood and  ESP, assuming 8,600 hours  of  operation  per year.  The  energy
 requirements for  Alternative II include 10 percent additional energy
 for miscellaneous equipment.   At ASARCO-Tacoma,  Regulatory Alternative  II
 requires  an additional 15 million  kWh  of  electricity  over the baseline
 case.  Total smelter electrical energy requirements were  estimated  by
 assuming  that the smelter requires  49  x 10 Btu  (thermal) per ton of
                      2
 anode copper produced.   The ASARCO  smelter has  a reported copper
 production capacity of 91,000  Mg/year  (100,000 tons/year).3   Assuming
 a power plant efficiency of 35 percent and 8,600 hours of operation
 per year, the total  electrical  energy  requirement of  the ASARCO smelter
 was estimated to  be 3 x  10   kWh.   Compared to this total, the  additional
 energy requirements of Regulatory Alternative II are  negligible.
 Regulatory Alternative III  has no energy requirements.
 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.  At ASARCO-Tacoma, arsenic-containing
 dust collected from the roaster baghouse and the reverberatory  furnace
 electrostatic precipitator is used as  input material  to the arsenic
 plant.  A portion of the dust collected in  the converter electrostatic
 precipitators is sent to the arsenic plant with  the balance sent to
 the fine ore bin.  From the fine ore bin,  the dust is recycled  back to
 the process.  Arsenic-containing wastes 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 lagoon for further settling.
 From the lagoon the  materials may be dredged and recycled to the
process.
                                   5-7

-------
     The total amount of solid waste generated under the baseline
control case at ASARCO-Tacoma can be determined by assuming that
50 percent of the concentrate fed to the smelter becomes solid waste,
25 percent is sulfur removed as S02, and the remaining 25 percent
becomes blister copper.  Given an annual maximum concentrate feed rate
                                  o
of 363,000 Mg/yr (400,000 tons/yr) , solid wastes (including slag)
generated by the smelter are approximately 181,500 Mg/yr (200,000 tons/yr),
     Incremental solid wastes generated under Alternative II were
calculated by assuming that the air curtain secondary hood system will
capture 95 percent of converter fugitive emissions and the ESP will
have a collection efficiency of 96 percent.  This represents an overall
control efficiency of 91.2 percent for the converter fugitive emissions
control system.  The fugitive emissions collected under Alternative II
are assumed to have no value and will, therefore, be disposed of as
waste.  Assuming that the arsenic content of the fugitive emissions
collected is 1 percent, the incremental solid waste impact under
Alternative II is calculated at 11,000 Mg/yr.  Under Alternative III,
solid wastes generated by the smelter would reduce to zero.  Table 5-5
summarizes the incremental solid waste captured for the regulatory
alternatives at ASARCO-Tacoma.
     In comparison with the amount of solid waste generated by the
smelter under the baseline case, the additional amount of solid waste
generated under Regulatory Alternative II is negligible.
5.5  WATER POLLUTION IMPACTS OF THE REGULATORY ALTERNATIVES
     The control systems for the regulatory alternatives are dry
systems; consequently, there will be no incremental increase in water
discharges.   If scrubbers are used, increases  in wastewater discharges
result if the arsenic-containing dusts are disposed of along with the
acid plant slurry.  Even if scrubbers are used, no adverse water
pollution impact is anticipated, because the additional wastewater
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 5-5.  SOLID WASTES GENERATED BY AIR POLLUTION CONTROL EQUIPMENT
              AT ASARCO-TACOMA BY REGULATORY ALTERNATIVE9
     Regulatory
     Alternative
Incremental  Solid Wastes
    Generated by Air
   Pollution Control
    Equipment from
       Baseline  .
    (1,000 Mg/yr)D
     Baseline

     Alternative II

     Alternative III
         11.0

          0
 Based on 8,600 hours per year.

Represents total  solid waste including slag disposed of by the smelter.
                                   5-9

-------
5.6  REFERENCES

1.   Letter and Attachments from fir. M.O. Varner,  ASARCO,  Incorporated,
     to J.R. Farmer, U.S. Environmental Protection Agency.   March  16,
     1983.  Response to Section 114 Information  Request.

2.   Environmental Impact Statement for ASARCO,  Incorporated.   Variance
     from PSAPCA Regulation I, Sections 9.03(b), 9.07(b),  and  9.07(c).
     Summary.  Final Report.  Puget Sound Air  Pollution Control Agency
     (PSAPCA).  September 1981.  p. 36.

3.   Preliminary Study of Sources of Inorganic Arsenic.  U.S.  Environmental
     Protection Agency.  Research Triangle Park, North Carolina.
     Report No. EPA 68-02-3513.  August 1982.  p. 21.

4.   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

     As discussed in earlier sections, only ASARCO-Tacoma -processes high
arsenic content feed.  None of the other existing smelters  processes
high arsenic content feed nor is anticipated to do so in the future.  In
addition, no new smelters are expected to be built in the next 5 years.
Therefore, the cost analysis developed in this section applies only to
the ASARCO-Tacoma smelter.
     This section provides estimates of capital and annualized costs to
control the arsenic emissions from the converter operations at the ASARCO-
Tacoma smelter.  These cost estimates provide the basis for the implemen-
tation of each of the regulatory alternatives identified in Section 4.0
for the control of arsenic emissions from the smelter.  As  discussed in
earlier sections, all the process emission sources and fugitive emission
sources other than from converter operations at the smelter are effectively
controlled; therefore, the regulatory alternatives discussed in Section
4.0 do not include additional controls for these sources.  As a result,
no additional expenditure would be required under the regulatory alter-
natives to control emissions from any source other than.the fugitive
emissions from the converter operations.  Consequently, the cost analysis
will be limited to developing cost estimates for the control  of fugitive
emissions from converter operations.
6.1  EXISTING FACILITY
     The ASARCO-Tacoma smelter operates three Fierce-Smith  converters.
The capital and annualized costs for controlling fugitive emissions
from all three converters are developed in this section.  These costs
are based on site-specific parameters.
6.1.1  Control System
     As noted in Section 4.0, the control technique selected for the
control of fugitive arsenic emissions from converter operations consists
of applying an air curtain secondary hood for capture followed by an
                                   6-1

-------
effective participate matter control  device (96 percent  efficient)  for
collection.  The air curtain secondary hood systems  to  be  installed are
essentially identical to the prototype secondary hood presently  installed
on the No. 4 converter at the smelter.  ASARCO intends  to  use  an  existing
ESP at the smelter for the collection of captured fugitive emissions
from the converter operations.  Thus, it is assumed  that no additional
capital expenditure will be required  for collection  of  these captured
fugitive emissions.  Therefore, the cost analysis addresses both  the
capital and operating costs for equipment to capture the fugitive emis-
sions, and only the operating costs for fugitive emission  collection
equipment.
6.1.2  Cost Parameters
     The cost estimates for the converter capture systems  are  developed
based on the design parameters summarized in Table 6-1.  The parameters
were obtained from ASARCO.1
     The ASARCO design consists of an air curtain and hood enclosure for
each converter.  The top, back, and sides of the hood are  fully  enclosed.
The top and sides are extended into the converter aisle  area.  Converter
charging and skimming can be performed by bringing a ladle into  and out
of the enclosure through the front end opening.  The captured  gases
from the enclosure are evacuated through a take-off  located at one  end
on the top of the enclosure to a dust chamber and then  to  a main  fan.
An air curtain screen is located at the top of the hood  enclosure
directly opposite the exhaust take-off for the enclosure.   Each  air
curtain has its own fan.  A common duct from the main fan  exhausts  the
gases to an existing flue, then to an ESP.
     Exhaust rates through the enclosure are controlled  based  on  the
number of active converters and their modes of operation.   (See
Section 3.1.2.7 for a detailed description of the system.)  Exhaust fan
requirements assume that, at most, only two converters  will be in
operation at a time.  Therefore, the  main fan will handle  gases  from,
at most, two hood enclosures.  The main fan is designed  for 110  nrVs
(230,000 acfm) to handle the maximum  volume of gases from  a worst-case
condition of one converter in roll-in and one in roll-out  mode.   The
fan is specified for 930 kW (1,200 hp) and 5.5 kPa (22  in. water)
pressure requirement.  This fan power requirement is sufficient  to
                                  6-2

-------
Table 6-1.  DESIGN PARAMETERS FOR THE AIR CURTAIN SECONDARY HOOD
                       AND ESP FOR ASARCO-TACOMAa
     Parameter
     Value
Air curtain

Gas flow rate:
     During converter roll-in13, m^/s (acfm)
     During converter roll-out0, nr/s (acfm)
System pressure drop, kPa (in. water)
Fan power requirements,  kW (hp)d

Secondary hood system

Gas flow rate:
     During converter roll-in'3, m^/s (acfm)
     During converter roll-out0, nP/s (acfm)
Fan capacity, m /s (acfm)6    ^
Pressure drop, kPa (in.  water)f
Fan power requirements,  kW (hp)

ESP
Maximum gas flow rate, nP/s (acfm)
  5.2 (11,000)
  8.5 (18,000)
  7.5 (30)
224   (300)
 33   (70,000)
 57   (120,000)
109   (230,000)
  5.5 (22)
930   (1,250)
109   (230,000)
 Each of the three existing converters  will  be equipped  with  an
 air curtain secondary hood capture system.
D
 Converter blowing or holding.

 Converter charging or skimming.
1
 Total for two air curtains.  Each air  curtain has  a 112 kW  (150  hp)
 fan.
3
 Based on one converter in roll-in and  one converter in  roll-out
 mode.
r
 To handle pressure drop across the entire system including hood
 enclosure, ESP, ducting, and stack.
                                  6-3

-------
force the gases through the hood enclosures, ESP,  and stack to the
atmosphere.  Duct requirements include (1) a total  of 229 m (750 ft)  of
1.52 m (60 in.) diameter duct to convey emissions  from the three secon-
dary hoods to a dust chamber and then to the main  exhaust fan, and (2)
91 m (300 ft) of 2.18 m (86 in.) diameter duct to  exhaust the gases
from the fan to the existing flue system.  The air curtain fan requirement
for each of the three converters to achieve the needed air jet pressure
was estimated at 8.5 m3/s (18,000 acfm) maximum.  The fan .power requirement
for each air curtain was estimated at 112 kW (150  hp).
6.1.3  Capital and Annualized Costs
     Capital cost is estimated for the air curtain secondary hood.
Annualized cost is estimated for the air curtain secondary hood and
ESP.  The capital cost includes all the cost items necessary to design,
purchase, and install the capture system.  It includes the cost of air
curtain fans, motors, hood structural material, main exhaust fan, and
ductwork; direct installation charges including foundation and other
direct costs such as electrical, instrumentation,  and controls; and
indirect costs for engineering services, procurement, taxes, fees, and
contingency.  All costs are in December 1982 dollars.
     The annualized cost of a control system is the annual cost to the
plant to own and operate that control system.  The annualized cost
includes direct operating costs such as utilities, maintenance, and
operating labor; and indirect operating costs or capital-related charges
such as depreciation, interest, administrative overhead, property
taxes, and insurance.
     Capital cost estimates are developed based on the cost data provided
to EPA by ASARCO.1  The ASARCO cost data contain January 1982 cost
estimates for the air curtain secondary hood capture system with doors
for the three converters at the Tacoma smelter.  Direct capital costs
are based on actual estimates, and indirect capital costs are based on
percentages of the estimated equipment costs.
     Tables 6-2 and 6-3 present the capital and annualized cost estimates
for air curtain secondary hood systems at the ASARCO-Tacoma smelter, in
December 1982 dollars.
                                6-4

-------
     Table 6-2.  ESTIMATED CAPITAL COSTS FOR THE AIR CURTAIN SECONDARY
                      HOOD SYSTEM AT ASARCO-TACOMA
                        (December 1982 dollars)
     Item
Capital Cost, $
Air curtains

Hoods

Duct work

Electrical system

     Total
  375,000

  591,000

2,444,000

   59,000

3,469,000
                                  6-5

-------
          Table 6-3.  ESTIMATED ANNUAL OPERATING COSTS FOR  THE
    AIR CURTAIN SECONDARY HOOD AND ESP SYSTEMS  FOR  ASARCO-TACOMA
                        (December 19-82 dollars)
               Item
 Cost, $/yr
Direct costs
     Operating and maintenance labor
     Maintenance material
     Utilities
Indirect costs
     Payroll overhead
     Operating supplies
     Administration overhead
     Taxes and insurance
     Capital recovery
               Total
   58,200
   29,800
  859,8003

   35,000
    6,000
   25,700
   69,400
  407,6QQb
1,491,500
 $94,000 is for electricity input to the ESP,  $585,800  for  electricity
 input to air curtain secondary hoods,  and $180,000  for electricity
 input for miscellaneous purposes.
3
 For the air curtain secondary hood system only.
                                  6-6

-------
     The capital cost estimate ($3.47 million) was obtained by first
calculating the January 1982 cost of air curtain secondary hood systems
without doors (from the ASARCO cost estimate for air curtain secondary
hood systems with doors), and then applying an escalation  factor of
1.03 to update the January 1982 cost to December 1982.   This cost
estimate of $3.47 million is about 25 percent less than  that estimated
by ASARCO for air curtain secondary hoods with doors. Annualized costs
were developed using the cost bases summarized in Table  6-4.
fr.1.4  Costs of Regulatory Alternatives
     This section presents the costs associated with the implementation
of the regulatory alternatives defined in Section 4.0.   Regulatory
Alternative I represents the baseline and includes all the existing
process and fugitive emission controls.  Regulatory Alternative II
includes the baseline and the converter fugitive emission  capture and
collection system.  Regulatory Alternative III requires  zero emissions.
Since neither Regulatory Alternative I (baseline) nor Regulatory Alter-
native III requires the installation of additional emission control
equipment at the smelter, no cost is estimated for either  alternative.
Therefore, the only additional control cost involved is  that associated
with Regulatory Alternative II which requires the installation of an
air curtain secondary hood on all three operating converters at the
smelter.
     As presented in Table 6-3, the annualized control cost for
implementing Regulatory Alternative II is estimated at $1.49 million.
Assuming an annual smelting capacity of 90,720 Mg/yr (100,000 tons/yr)
of copper is achieved, the additional cost of implementing Regulatory
Alternative II is $16.4/Mg ($14.9/ton) of copper.
6.1.5  Cost-Effecti veness
     Cost-effectiveness is defined as incremental annual costs in
dollars per unit of pollutant removed over baseline. As discussed in
Section 5.0, the arsenic emission reduction achievable under Regulatory
Alternative II is estimated at 12.8 kg/hr (28.2 Ib/hr),  or 110 Mg/yr
(121 tons/yr).  Therefore, the incremental cost-effectiveness associated
with implementing Regulatory Alternative II over baseline  is $13,600/Mg
($12,300/ton) of arsenic removed.
                                  6-7

-------
       Table 6-4.  COST BASES USED IN ESTIMATING  ANNUAL  OPERATING
                COSTS OF THE AIR CURTAIN  SECONDARY  HOOD
                   AND ESP SYSTEM FOR ASARCO,  TACOMAa
     Item
     Cost bases
     Comment
Direct costs
  Operating and
  maintenance laborb
  Maintenance material
  Utilities
Indirect costs
  Payroll overhead

  Operating supplies
  Administrative
  overhead
  Taxes and Insurance
  Capital recovery
2 labor-hours operating
labor and 15 percent
supervision per shift,
2 labor-hours mainten-
ance labor and 20 per-
cent supervision per
shift, and $11.53/labor-
hour

100 percent of
maintenance labor
The labor-hour
unit cost of
$11.53/hour is obtained
from Reference 2.
Based on U.S. Bureau
of Mines methodology
The unit power cost
of $0.059/kWh is
230,000 cfm gas flow
at 22 in. water
pressure, 0.0015 kW/ft2  obtained from
of ESP collection area,  Reference 3.
10 percent additional
energy for miscellaneous
requirements,  and
$0.059 per kWh.
60 percent of payroll
20 percent of total
maintenance cost

40 percent of total
direct labor and
operating supplies

2 percent of total
capital cost

20 years life
and 10 percent
interest rate
for both capture
and col lection
equipment
Based on U.S. Bureau
of Mines methodology
Equipment life
obtained from
Reference 4.
is
 System is assumed to operate 8,600 hours  in  a year.
3
 Estimate based on 1 labor-hour each of  operating  and  maintenance  labor
 for the air curtain secondary hood system and for the ESP.
                                  6-8

-------
6.2  REFERENCES

1.  ASARCO's Central  Engineering  Department.  ASARCO Incorporated
    Converter Secondary  Hooding,  Tacoma Plant.  Salt Lake City, Utah.
    January 22, 1981.

2.  Survey of Current Business, U.S. Department Labor, Washington, DC.
    December 1982.

3.  Monthly Energy  Review.   U.S.  Department of Commerce.
    DOE/EIA 033583/01.  Washington, DC.  January 1983.

4.  GARD, Inc.  Capital  and Annual Operating Costs of Selected Air
    Pollution Control Systems.  U.S. Environmental Protection Agency.
    EPA Contract No.  68-02-2812.  December 1978.
                                  6-9

-------

-------
                         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 seven large companies.
All seven 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

-------

fO
oo


f-"^
LU

LU

"•if
•£-.
S;
Q;
^*»


^£

o:
LU
| —

2=

LU
C_3
c£

o
GO

Q

§


**

Q

^ 'I
r—

Q
O
0£
CL.


r\
CL,
*— '
IE


LU



O
o:
LU
1—
LU





•





CO
p—
0

fO
I—




in
4-1
C
Ol
E
Ol
cn
C
s-
S-
«ac
_
CD
01
4J
ID
E:






c
o
—0 4-1
XI U
31 3
— < "O
O
o.



o
31 4J
-- U
31 3
-f -O
O
1-



Is-
"~ 1O
3 O.
XI CO
31 0
•H






















01
z


01
4-1

§=
V)
s
XI
31
— 1


31

31
— <

















•^
cn
E






^,
cn









'cn












CL
"w
01
c
3:
O










c:
o
4-1
ID
U
O

•o
c
CO
X S* S*
OJ *?" CO



«*«»«





1 1 1

•o
Ol
co
S- E
cn o
0) 4-1,—

C 3 0
•— Ol—




OJ
goo
o o
r*^ to PO
oj cn r^.





OJ
o o o
o o o
0 OO

•-< vo in
in cn co





§o o
o o
—1 OJ — <
cn co cn
,-H












o
c
•-1

o

rv
S
•S





X

. . o
to C in
E 01 to

to to i —
I— 2: UJ
g
o
«— 4


^Q
O
O
,-H


1
.Q
•o
Ol
IO
S-
cn
Ol

c




in

o
o






'O
0
cn

OJ






0
o
tn











0

01
U
i-
eo


Ol

£
0


z
t—

r£

J=
s.
Ol
CL
CL
O
O
0
CD
«— 4


ft^
O
o
r-H


1

•o
QJ
(0


Q}

C





g
in
OJ
r-i





o
i —

cn






§
OJ
tn



i
in o


o c
_1 0

Ol ~±>
• J= IO
0|—!-

Ol O CL
cn x
to &-
o: 10 -o

s- -o co
Ol *^- i
cu tn
o"§ c
r_1 CO ID


t^t


Ol

d.
01

^
3
CO OJ
in *^



&^ ^^
in in
rn us



i i

•c
Ol
IO
S-.
cn
01 i—
4-1 r—
C 0




in
S

J —
0
i— i




o
r~.
O

^^
OJ
»— 1




o
o
5





"S

to



o
c:
o >>
to
c: o.
0 E
•»- o
*-> o
ID
i. S-
•r- O>
O. CL
tn Q.
c: o






rvi


*"

•™
EC
O O
CM ^-1



&$
O
o



1 1

•o
01
to
s-
cn
Ol r-
4-1 t—
C O
•— 1—




tn
0 0
o o

S to

OJ




0 S
O PO
•1 M
in in
cn in
OJ




o
SO 0 0
o o o
0 O O O
in «a- r-* m
•3- in r-. r~
OJ











(
e-
o
4-1
O
O
01
c
c:
Ol
^




1—
Z "S Z
•o

•— 01 C >,
r— •*- Ol 01
T- «- -0 »—
O S- >, S-
0 10 10 3
s: ts a= z
o
o



&•?


-<


i i

•o
01
to
s-
cn
01 ,—
4J r—
C 0





O
o
0
co"
cn





g
cn
A
ir>

«-H




0
o
o
,— i
CO
*— t












cn
c:
"c
z

c

E
1






10 S


S *4>
E 3
cn c
to to
s: s:
in in
r^ oj




i t i




i i i

•o
to
S- =
cn o
Ol 4J ,—
4-1 tn,—
C 3 0
—i 0 I—



=0
c>

•a-
OJ
m





CO
o o
o o
0 0

m .-i
co cn
OJ




o o o o
o o o o
o o o o
in ^ ,-H ro
^H VD cn in











Q.
S-
o
o
Ol
cn
•o
o


Cu

"oi
O- •





s s ^

tn r^l -^ O
to cc U Cn
Cn •> 01 to
3 0' S- -0
O -r-s O •*-
o <: E: z
tn in •— i
r^ oj









t i i

•o
01
ID
i. E
cn o
01 4J r-
4-1 in r-
=: 3 0
—• 0 t—





OJ &«

in cn
oj in

_J"



o- —
O ft*
o cn

i ^^

CM ~ •

*—t


0
o
in
OJ


,-4



















c:
o
•f— 01


o cn
^- C

r- "to
ID &-
4J Ol
O Q-
r- 0












• in
C 01
o c
in H "-^
tn cn

1 c ^
§© O
N O

01 «£ O
Cn CO
10 4-1

O CO
X tt-
UJ O T3
Ol
•o tn 4-1
CO! ID
to r— !-
to *^
tn cn
Ol o
i- Ol CO
4J JC cn
S-
3 tn c
U 3 -^
Ol C

E Oi
Oi tn
J= 01 O
4J U r—
T- U
0 >
4-1 S- in
Ol IO
J2 "> S
S- tn jc
O 01 U
CL -F- -^
OJ 4J i—
t- f 3
O ^
1 S- Ol

01
01 in . E
4J Oi in in
to ,— c:
i- CO O CO
O in 4J -o
i- S- O O
O Ol CD 1-1
O CL O 'tD
D. - c:
§o o «t
U — 1
&~ ^ cn
>*- r- C
m o t-


•^ 4-1 -O i—
S- C U
ID c: to x
E o CL o>
f- X
i- -a oi >,
Cu 01 &
in O •»-
C 10 4-1 U
C -O to
••- -0 Cu
4-1 01 O) fO
10 4J 4J O
Eia ID
E E -o
O i- i- Ol
i. 4J 4.1 4J
E m in to
•—i i » i uj C£
ID ^1 U *O
7-3

-------
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 three 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 four
companies each have one 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 eight 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  C9pper 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

-------
OO
Q
LU
LU
    IQ
s:   co
o   en
   r—

Q   00
	I   ^
Q;   en
O   r—   -—.
S         10

Q   ..    n3
2:   2:    s-
•=C   O    en
      t—i    ns
oo   I—    en
I—   ZD
<=c   a
1—   O
oo   a:
      a.
o
UJ   C£
1—   UJ
I—I   Q_
2:   a.
ZD   O
      CJ
CM   I— i
 O)
 (C
           crs
                         s.
                         OJ
                         Q.
                         a.
                         C5
                        o
           •o
      Ul JJ I—
      CB C
        ai o
                           = 0.0

                                •a

                                o
                                -a
                           tn *J r—
                           re c s-
                           => o-
c  c
oo
                                    r-.coocMroi£>ppt-~pocr> «s- o
                                                                                 m cv) »-*
                                                                                           en
                                                                                                o.
                                                                                              O        III
                                                                                 lOt-H    &.CXOJ    f> ^" i—4
                                                                                 • co co    d E J-*    ^o ^o r^
                                                                                 icncn    >o*o    cncncn
                                                                                 cr-i^H    .3:00:    ,-i^_
                                                                                             o en
                                                                                            •u ir>
                                                                                               ro
                                                                                            4J in
                                                                                             s- «»•

                                                                                            .id
                                                                                             CA
                                                                                               V)
                                                                                            fM p^
                                                                                            f-< (O
                                                                                            (T) 3
                                                                                            evj cr
                                                                                            o a>
                                                                                                                  II
                                                                                                                  03  O
                                                                                                                     0)


                                                                                                                 0°
                                                                                                                 O  01
                                                                                                                . O  !~
                                                                                            Cl O
                                                                                            ra o.
                                                                                            01 S
                                                                                            ai x)
 • -o
«  c

I  o

en    w
(0 o  £
C7)O  (O
a> o  s-
E CM  01
                                                                                            O

                                                                                            O
                                                                                                                     ) &>
                                                                                                                       LT>
                                                                                            s- S  o
                                                                                            CTJ
                                                                                            re in  E
                                                                                            oco  a
                                                                                            •*- -H  S-
                                                                                                                      f—
                                                                                                                    50\ f-
                                                                                                                    — - j^
                                                            7-7

-------
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
1s 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

-------

























CO
T— 1
0
1 — 1
I—
Q_
2:
=>
CO
o
C_3

QL
mi «
o. * — -
CL. in
O "3

oo re
• en
*""* •!-•
cn
CO
1
1 —

O)
f—
re
H-






































2
X)

-H


X)

-H


31

31
*4


X)

31
-H


^_
31
—1



O
^
31
—1



a

31
— i



*
31
—1


•o

31
— t



NI

31
«4



—1
^.
31
—1




^
31
—1




















to
O
O1
co
I— 1
oo
to
p^.
co
I— H
r-l
•
^-
to
r— 1
CM
LO
•
to
01
i— i
CM
CM
en
CO
en
i— i

i-*».
•
r— 1
I— (
CO
r-l

en
•
co
01
oo
«— '

to
CO
en
01
i — !
to
•
o
CM
CM
CM

LO
•
01
CM
O
CM

O
•
CO
CO
CO
I— 1


co
01
LO
co
r— I




i_
<+_ CU
0 CL
CL
c: O
O o

'•P -0
CL QJ
E s=
3 •!-
01 <4-
£= dl
o cs:
O
to
co
CO
CO

o
o
vo
co

oo
•
^-
co
en

i— i
•
r— i
CO
CO

to
r--

co

^H
en
i —
r--.
CM
OO
r-~
OJ

OJ
,-J.
0
CO
«^t"
r»4
o->
LO
0
!—•••
P--
e^
0")
CO
to
oo
CO
OJ

*xt"
•
O
o
CO

CM
«
CM
r~l
to
OJ

^J-
•
CM
to


OJ
B
T-H
to
o
OJ

0
CO

CO

^.
•
to
LO
en

to
vo
r--
co
CM
o

i —
i —
I— 1
oo

o

to

en

LO
«
LO
f""^
O1
CM

LO
•
01
LO
CO

in
*
CM
O1
to
CM


to
I— 1
r— 1
co


•sf
i— t
1 —
to
OJ





i»- -o
O OJ
E
C 3
0 in

'•P 0
CLO c_5 '-
E CL CU
3 re i — CL
 S- re CL
co -P o
o oo o o
o t-
o
OJ
oo


LO
T-H
co


oo
•
r-l
CO


to
•
CO
CM


en LO co oo
O1 t-t CM OO
CM CM en t-f
o co en
CM CM

vo
•
o
oo



I— 1
•
CM
CO



LO
O
CO


i— 1
•
O
CO



co en to LO
• • • •
•— 1 r-H CO O
oo en «=)- «*
oo co r--.
r-l CM

en
•
,_{
00




•5f
0
CO








V5T3S- -Oi- "OS-S-
re c: oc c: oc c: ooic;
re ti-o re *+-o re M-CLO
I— «— 1 •>— i— 1 -I— i— 1 CL'I—
re vo co ^ *P vo co *" ^~^ o ^o co to o -P
4-^ r>^ en co CL*O r^^ 01 cu o re r^^ 01 QJ o o
o en i— i en E cu en i— i en E s- C3i>— fen E
I— CL ,-H i re 3 c: ^nireso 1-1 i re i— 3
re i r-*» >- in •*— \ r^ i- (/) co i p^ s- re t/i
4— i- CM 1 — - CJ C H— CM t^ CJ CI CM f^- CU -P C.
ou r^oi>oa> r-~.en>o<4- i — .oi>oo
oo oii— treos- 01.— ireoo oi^nre-Po
S* T-H T— 1 r-l






























































•
10

£1
3
0
CL

0
•I—
P~"
f-m
• ^
E

CM
CM

II

E ^
re re
s_ c
en -i-
re E
CD -r-

cn "a!
i_
r-l 0-
« ja




























































c
•r—
re
en
re

•a
CD
c:
t«
•*^*
a>
s_

cu
JD

O
•P
en
c:

, >
re


-p
3
O

4^
•^
3
O
7-9

-------
O
CM
OO
LU
LU



«C
CO
LU
O

C£.
LU
a.
o_
o
o


oo
r-.

 CD
jQ
 re
























^•^
to
£
re
s_
cn
5
l«_^


























n
T—
CO
a

o
CO
cr
*""*
cn
fx^
cr
1
CO
f— ,
cn
t-H

i^
t~-
01
I""1
vo
r-.
01
*~*
LO
^^
cn
t-H

^
r-
en
t-H
oo
r~-
en
t-H


CM
cn
t — i
t— i
r~
cn
t-H


o
r~.
cn
t-H




CD
s-
re

o
t-H
OO

^.
*^.
"•».

r-H
fi
cn

0
o
01


oo
00
oo

o
oo
•*»»

,_,.
"f)
o


^
T>
**»

oo
cn
cn



oo
CM
cn

O
f)
oo



oo
^




*
J
o
S
_5
3T>
3
3
XI

0
t— 1
oo

o
1—^
CM

CO
CO
oo

CM
CM



CM
t-H

CM
f-.
OO

LO
vo
CM


CM
OO
00

CM
CD
*3*



LO
oo

r--
0
oo



f-.
vo
CM




E
o
sportati
c
re
1—























































X>
re
S-
3
in
c
0
o

vo

oo

CM

OO

0
CM


CO
CO
OO


o
oo

CO
CM
oo

vo
t— 1
oo


CM
o


oo
t-H




t— 1

co
LO
oo



cn
vo
oo


to
4-^
o
•g
0
s_
0.
"re
O)
c
CD
CD

p-^
OO
LO

t-H
O
LO

LO
VO
LO

t-H
LO
LO


CO
«— 1
LO

OO
CO


,.,.
0



oo
vo
LO

O
^«
vo



t-H
O
VO

0

LO



0
VO
LO




•o
re
i— >,-»->
re s_ £=
•i—  o -o
O O) O
(1) i— S-
i— LU Q_
LU
OO
t-H
CO
CM
t— 1
^~
vo
CM
r--
r-
t-H
oo
co
t-H
t-H
oo
o
LO r»-
CM r--
cn oo
CM
VO

vo
CM
t-H
vo
t-H
CM

CO
LO
CO
CM
t-H
cn
CM
oo
0
•
cn vo
LO CM
O CO
oo
^
f-.
r-.
CM


t-H
VO
vo
CM



CD
CO E
re o
i- -I-

"« 0) E i— C
+J >-  o re re
•»-> E i- O
re rea_ i — ^-P i — •>-
+J re s- E re E to
o 
-------
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 o,f 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
 telecommunications.

 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 LMEa 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 PRICES2^24,25

                           (cents  per kg)a
Year
LMEb
U.S Producer Price0
U.S. Producer Price
1982 Constant Priced
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982e
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
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
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
aTo convert from cents/kg to cents/1b,  multiply  by  0.454.

bLondon Metal Exchange "high-grade"  contract.
CU.S producer price, electrolytic wirebar copper, delivered U.S destinations
 basis.
Adjusted to 1982 constant price by  applying  implicit  price deflator for
 gross national product (1972 = 100).

Preliminary.
                                     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 pyrometallurgical 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 require 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  The report concludes that increases in copper
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 high arsenic primary copper smelter.  The 14 other primary
copper smelters are classified as low arsenic smelters and are discussed in a
separate analysis.
       The principal economic impacts analyzed are:  the ability of the
smelter 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.
       If the smelter attempts to pass control costs forward in the form of a
price increase, the price increases would be relatively modest and would
range from 0.7 percent to 1.1 percent, depending upon the capacity utilization
rate and the price that is used.
       If control costs are absorbed and profit margins reduced, the profit
reductions would range from a high of 10.7 percent to a low of 3.1 percent,
depending upon the capacity utilization rate and the price that is used.
       The capital costs of the control equipment are not minor amounts.
However, ASARCO is a major publicly-held corporation with a good credit
rating and good access to the capital markets.  The incremental capital cost
of the control equipment for the Tacoma smelter alone does not present a
major obstacle.  If capital costs were incurred at ASARCO's two other smel-
ters, the capital would still be likely to be available.
                                     7-17

-------
 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 smelter.  Therefore the methodo-
 logy relies on several indicators which in total can be used to draw conclusions.
       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 is 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 the 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.
       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  byproducts include selenium,  tellurium,
 and antimony.  A by-product that sets  Tacoma apart from other smelters  is
 arsenic (arsenic trioxide and metallic arsenic).  For ease of presentation
 and in order to present a conservative 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,
                                     7-18

-------
 stockpiling  and  price controls.  The government variable includes the U.S.
 Government,  as well as foreign governments.  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
 in the extreme, 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
 refining.  Also, several  of the producers are integrated one additional  step
 into the fabrication of refined copper.  However, not all companies in the
 copper 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-
 ship of the copper) or on a toll  basis (the smelter charges a service fee and
 returns the copper to the owner).   Information presented earlier in Table 7-1
 has shown that about 25 percent of the concentrate  processed by ASARCO is
 from integrated mines owned by the company.  However,  the major portion, the
 remaining 75 percent, of the concentrate it processes  is smelted and refined
 on a toll or custom basis.   The existence of both integrated and nonintegrated
 production introduces a complex economic element into  this 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
                                     7-19

-------
 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
 refi nery.
       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
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
                                     7-20

-------
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
price 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  Japanese Tariff mechanism.  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
import duties allow Japanese producers to sell their refined copper in Japan
at an artificially high price and still remain competitive with  foreign
producers.
       Specifically, copper concentrates  and ore imported into Japan are free
of duty.  Refined copper imported into Japan is subjected to a tariff  of
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
                                    7-21

-------
 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
 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.

                                     7-22

-------
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.32

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 passed forward in the form of a price increase.   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 credit, 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
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.
                                     7-23

-------
       Table 7-6 first shows the cost increase, and then the maximum  percen-
tage price increase, of arsenic controls for the ASARCO-Tacoma primary copper
smelter.  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. • As described
earlier, capacity for Tacoma is presented as 91,000 Mg per year.  An  80
percent capacity utilization rate represents production of 72,800  Mg, which
is roughly the amount of production achieved at Tacoma during 1981.   For
1982, the industry average capacity utilization rate was substantially lower
at 59 percent.  However, no analysis is shown here of the impact of control
costs at a 59 or 60 percent utilization rate because regardless of control
costs, a rate of 60 percent is damaging even as a baseline condition.  For
example, based on a capacity at Tacoma of 91,000 Mg per year, for  the years
from 1975 to 1981 only the strike-year of 1980 had a capacity utilization
rate of 60 percent or below.  The purpose of showing the increase  in  produc-
tion cost is to supplement the maximum percentage price increase.  One
advantage of reviewing the cost increase is that it is dependent only 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.  Two points should be observed from the cost increases:
       1)  Although any cost increase is undesirable from the firm's  view-
point, the amount of the cost increase is relatively modest.  Although the
cost increase is relatively modest several  factors should be mentioned.
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.
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.

                                      7-24

-------
















0
1— 1
•z.
UJ
oo
Qi
££
Lo_ UJ
0 1—
_1
UJ UJ
oo ^"
 Di
s <
t— * 1
X 0=

2: >-l
3T
Q
z: uj
<: =c
UJ
GO Qi
eC O
LU Ll_
rv*
C_3 OO
~S^ 1
M O
o;
t i ,
|~" I"™
oo -z.
0 0
o o
*
<~o
\
QJ

/^
,2
^^








o>
WJ
O)
i.
u
c
t— 1
o
0
•^~
i_
Ou
E

&s -I—
0 0
o to
i-H Q.
O

•r-
*s o
o >
5-9 -i—
0 0
O ro
«-« CL

O



h— 1
1— 1









«*-*.
O5
JM;
^""v^
"W.
«ta_^

CD
(/)
«3
O
s-
o
E
i— «

4-)

O
«_>




-p a;
•i- >
o -i-

CL 5

•1- >
O •!-
(O -t->
Q. 03
05 C
0 S-
CD
5^ 4-*
O i—
0 •<








H— 1
1— 1






-^
i^ 4-)
ft3 *r™
=30^
E res en
E Q. s:

o






i.










(O
fTT
i
o
(d
o
o
«C
7-25

-------
        2)  The difference between the two capacity utilization rates, in
terms of the  size of the cost increase, is minimal.
        Table  7-6 also shows maximum percentage price increases.  The purpose
of reporting  the maximum percentage price increase figures is to add perspec-
tive 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 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.  The analysis
of the results for the maximum percentage price increase figures is similar
to the analysis discussed above for the cost increase figures.  The results
show that the size of the maximum percentage price increase  is modest, and
there is minimal  variation.  At a price of 187^/kg, the price increase is
0.9 percent at a 100 percent capacity utilization rate, and  1.1 percent at an
80 percent capacity utilization rate.  At a price of 220^/kg, the price
increases are 0.7 and 0.9 percent, respectively.  However, as mentioned
above, although the price increases are modest, two 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;
                                     7-26

-------
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 (ROD, 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
presents in terms of an assessment of the impact of the arsenic standard is
in the method of assigning the costs.
       This report presents an analysis of profit impacts using two methods.
The first method assumes the smelter is fully integrated.  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 7-7 shows the revenues and operating profit
(before tax)  for each of the seven producers for the 5-year period from  1977
to 1981.  Table 7-7  also shows the percentage return on sales, which is
operating 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
                                     7-27

-------










re
CO
LU
t— 1
2E
<;



o

OS
LU
O-
8
o

OS
o
u_

CO
UJ
_I
$

•z.
o

z
°4 ^T"
— — * CO
I*— • ^^
Ul r-l
OS
£ ~
LU
§
LU
CO
CO
CO
LU
|..i_|
CO
CO


•
f*»^
1
t~~

0>
r—

re
h-








CO
(~v rt i
L-U UJ
r— Cn
<1) ~O
XT O
a. Q




ij
rt3
pr
O>
«3



4J
4-)
O
0
CD
cr
c:
CD
^

C
O
•1 —
•p
re
S-
•r—
CX
CO
c:
i — i



i-JD
CD CD
cx cn
cx c
o re

CD
CO O
CD T-
•I- >
•P S-
•r- CD
O CO



0
•a;
CO
a:




S-
re
CD
^.










*d" O CO i — I *d"
t — i cn i — i LO *3*
co to co «d- to
LO «3- i-H T-H 0
«~f «d- to f- r~-


I-- CM «— 1 CM
CO i-l CO «3-
eg i-H LO LO CO
^^ A A A A
 cn cn
i-H r-H i-H i— ) r-l


cn
c
•—— n c }
•P 

^.. O CT) OvJ CO • • • • »— i cn cn r-- «=c CM •— :z o CM co •-( cn cn co to o LO ^ i [ i r-- co cn o i-) r-- i~- r~- co co cn cn cn cn cn i— 1 r— ) i-H i— 1 i— 1 ^-^ CO P ^**s CD *"" ^ 3 CD •i- C O O > CD S- CD CX o. os—- co CO 1 ' ] CD LO LO -^> • cn ^-^. co CO o 1— 1 ^-^ to • CO o ^H CD cn re S- CD e^ • i- CD CX cx 0 o o c: 0 .•{.! •o -o re c CO o o S- cx S- CD .c: ^-> o c: re <= o o CO -p c CD E cn CD CO CO CO CD c CO 3 CQ • c: re cx o o c o •r— -P S- o __ ex X i. i C c& T3 £= «J fl3 C fO • |B. to *r~ 13 O CD £ • CD CO CO O CO -p c X 4_> s_ O) 1 s q- CD £= O O c 4^ CD C CO •1— p •r- q— 0 s~ O- o X fO •p -o c: re -P CO CD i. CD •P C •r— CD S_ O CD CQ •o C|_ O CD 4-^ re S- X re •p CD > •r— •p o CD q- CD c re ^ 4_> •r~ ^ X re 4-> CD <_- O CD £ •o CD CO 3 •o re q- •"- •P CD O CD CX cn • P^ -a • r- P CD £= •r- CD >> 0 i- •a CD i— CX 3 ^S CD C S- 3 4J CD S- CD cn re s_ CD ^ re •a r» CD •r- •a r— 3 O r- ) CO cn t— i o 4-> I^«. r-. 01 1—1 S- o CO CD *re CO p CD £ cn CD s- •p c: •r- c: o • P c •r- CD q- o o s_ i- CD CX Q. •O LO CD • •P r^* 3 >-H CX •§ 0 q- 7-28


-------
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.  For ASARCO the average is 11.0 percent.
       Table 7-8 shows the control costs for the smelter and the maximum
percent reduction in the profit margin.  This table assumes the 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 187^/kg and a 10
percent profit margin, which yields a profit of 18.7£/kg.   The second
profit level is based on an increased price of refined copper to a level  of
220jz!/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 if
the smelter has a substantial profit impact in spite of such a favorable
profit margin, it is in a very vulnerable position at a lower, more likely,
profit margin.  At the first profit margin (18.7£/kg) the results  show a
moderate profit reduction of 8.6 percent and 10.7 percent, for the 100 and 80
percent capacity utilization rates, respectively.  Profit reductions of
these amounts are not likely to call into question the continued viability
of the smelter.  At the second, higher profit margin (51.7£/kg) the
profit impacts are lessened substantially.  The profit reductions  are 3.1
                                     7-29

-------
UJ
to
LU
CO
e:
o
LU
fe
D;
Q.
LU
O
C£
LU
       Ul
       ct-
       UJ
       a.
       o.
       o
       O
       oi
       CL.
       O
       LU
       CO
         OJ
•r-   >
 O  ••-
 <0  •»->
 Q.  (O
 (O   C
O   i-
     O)

O  r—
CO  <
                             •r-  >
                             O  -p

                             CX  (T3
                             (O  C
                             o  s-
                                 a>

                             o  i—
                             o  <
              10
              co
                                           o

                                           CM
                             CO "r~
                             c  o.
                             <:  to •
                                     cn
              o
              o
              o
                                           cr>
                                     i.
                                     ai
                                     cu
                                    co
                                           o
                                           o
                                          o
                                          o
                                                     7-30

-------
percent and 3.9 percent, respectively, and do not jeopardize the viability of
the smelter.

       7.2.5.2  Method Two.  Method two presents a supplement to method
one, which introduces several additional factors that may influence an
assessment of the viability of the smelter.
       1)   The ASARCO Tacoma smelter's ability to treat high arsenic content
concentrates is unique among domestic smelters.
       2)   The Tacoma smelter has an economic interrelationship with ASARCO's
East Helena operation.  ASARCO owns and operates a lead smelter at East
Helena, Montana.  Matte and speiss from East Helena are shipped to Tacoma for
treatment and recovery of byproducts.  Therefore, the presence of the Tacoma
smelter provides an economic benefit for the East Helena operation.  This
topic, and others, are discussed in detail in a comprehensive report on the
Tacoma smelter.36
       3)   If the Tacoma smelter were to close, the remaining capacity at
the Hayden and El Paso smelters would be insufficient to offset the loss and
allow any significant increase in blister production.  For example, disregard-
ing any technical problems, if the Tacoma smelter were to close and the
concentrates that otherwise would have gone to Tacoma were diverted to Hayden
and El Paso, in 1981 those two smelters would have had to operate at a 96
percent capacity utilization rate to maintain Asarco's total  blister produc-
tion.  A 96 percent capacity utilization rate would essentially leave no room
for growth.37.  Also, the Tacoma smelter provides an important percentage
(about 21 percent)  of the blister supplies for the Amarillo refinery.
       4)   The recent opening of ASARCO1s Troy mine in Montana should be an
additional long term source of concentrates for Tacoma.  The Troy mine and
concentrator began commercial production in late 1981 and reached full design
capacity in February 1982.  The Troy mine is expected to produce about 18,000
Mg of copper per year, plus a significant amount of silver.38  A production
rate of 18,000 Mg of copper per year represents a considerable amount of
Tacoma1s capacity, about 20 percent.  With production at full  capacity the
ore reserves would indicate a mine life of approximately 20 years.
                                     7-31

-------
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
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-
cussion of  capital availability will serve to supplement an assessment of
profitability.
       Table  7-9 shows the capital expenditures that will be necessary.
The capital expenditures were explained in detail in an earlier chapter.
The baseline  capital expenditures are presented, as well as the incremental
capital expenditures for Alternatives II, III, and IV.  Alternatives II, III,
and IV refer  to the alternatives for low-arsenic smelters.  Asarco owns three
smelters and  therefore the total capital costs are shown, although the firm
can make capital budgeting decisions on an individual  smelter basis.  The El
Paso and Hayden smelters are in the low-arsenic category, but are included
here to identify total  capital costs to ASARCO.  The capital  costs for the
smelters are  not trivial sums.  However, the company is a major corporation.
Five of the seven companies in the industry are owned wholly, or to a substan-
tial degree,  by significantly larger parent corporations.  ASARCO is one of
the two companies not owned by another corporation.
       Table  7-9 shows the percent increase in long-term debt if controls are
added.  The pre-control debt level is based on a 3-year average (1981 to
1979) debt level for the company.  Controls are assumed to be financed
totally with  debt.  The baseline percentage increase in debt is 24 percent.
An increase of 24 percent is substantial.  For Alternatives II,  III, and IV
the incremental  increases do not present a major obstacle.  The increases are
1, 2, and 1 percent, respectively.
       An additional indicator of capital  availability is provided by the
debt rating assigned to a company by one of the major national  rating ser-
vices.  In 1982, as well as 1981 and 1980, ASARCO's debt was rated as A3 by
                                     7-32

-------
     Table 7-9.   CAPITAL COSTS OF  ARSENIC  CONTROLS FOR

                  ASARCO PRIMARY COPPER  SMELTERS

                          ($103)
Smel ter
El Paso
Hayden
Tacoma
Debt Increase3
Baseline
46
75,606
75,652
24%
Alternative
II III
0 1,375
0 1,702
3,469 3,469
3,469 6,546
1% 2%

IV
370
0
3,469
3,839
1%
aPercent increase  in  average  long-term  debt level for
 past 3-years {1981 to  1979)  if controls  are added
 as debt.
                         7-33

-------
 Moody's.39  This is an investment grade rating,  but  it is  the  lowest A
 rating.  Although an A3 rating is a relatively  strong  rating,  it does not
 preclude the possibility that a substantial  increase in  the amount of debt 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,
 to assess any other significant macroeconomic impacts  that may result from
 the NESHAP.  Executive Order  12291  stipulates as "major  rules" those that 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; individual
             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.

        7.3.1.1  Annualized Control Costs.  The EPA criterion for a  major rule
based on  costs  is  annualized  control costs of $100 million or more  in  any one
of  the  first 5 years after promulgation of the standard.  The annual-
ized control  cost  for  the ASARCO-Tacoma smelter is $1.1 million,  well  below
$100 million.

        7.3.1.2  Industry Production.  The ASARCO-Tacoma smelter accounts for
about 5 percent of the total  U.S. primary copper smelting capacity.

        7.3.1.3   Employment and Local Effects.   A study performed for  the
City of Tacoma shows that, as of December 1977,  the ASARCO  Tacoma smelter and
refinery employed 798 people.40  The total  employment as of December,  1976
was 866 people.  The refinery has since closed,  in  1980.  More  recently,
                                     7-34

-------
 employment  at  the  smelter  is reported at about 500 employees in 1982. 41
 Related  to  these employment levels are indirect jobs created by the economic
 activity at the smelter.   Those jobs numbered 478 in December 1977, and 519
 in  December 1976,  which amounts to a total employment multiplier of 1.67.
 The total direct and indirect employment of the plant's operation comprises
 approximately  1 percent of the civilian employment in Pierce County, Washing-
 ton.
       Further, according  to the report, ASARCO-Tacoma does business with
 approximately  100  local firms.   The value of purchased goods and services
 from these  local firms amounts to a little over $20 million per year.   ASARCO-
 Tacoma paid  approximately  $2,370,000 in local  and State taxes in 1977,
 and $2,245,000 in  1976.  The smelter also paid $73,000 in taxes to the City
 of  Tacoma in 1977, and $85,000  in 1976.   The town of Rustin received approxi-
 mately 65 percent  of its general  fund budget in 1977 from ASARCO.

 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,  annualized
 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.  Total employment  for ASARCO is
reported as 12,500  in 1981.42  Therefore, ASARCO exceeds the SBA definition
of a small business and thus  no regulatory  flexibility  analysis is  required.
                                     7-35

-------
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.
                                     7-36

-------
18.

19.

20.

21.


22.


23.


24.


25.


26.


27.




28.




29.


30.




31.


32.


33.
Reference 12, p. 14.

Reference 11, p. 14.

Reference 12, p. 18.

Copper's Hope:  Electric Vehicles.  Copper Studies.  Commodities
Research Unit, Ltd. New York. March 30, 1979. p. 5.

Copper in Military Uses.  Copper Studies.  Commodities Research Unit,
Ltd.  New York. February 15, 1980. p. 1.

Butterman, W.C.  U.S. Bureau of Mines.  Mineral Industry Surveys.
Copper in 1982 - Annual, Preliminary,  p. 2.

Butterman, W.C.  U.S. Bureau of Mines.  Preprint from the 1980 Bureau
of Mines Minerals Yearbook.  Copper,  p. 1.

Schroeder, H. J., and 6. J. Coakley.  U.S. Bureau of Mines Preprint
from the 1975 Minerals Yearbook.  Copper,  p. 2.

The Capital  Cost Picture.  Copper Studies.  Commodities Research Unit,
Ltd. New York. August 18, 1975. p. 1.

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.

Sousa, Louis J.  The U.S. Copper Industry:  Problems, Issues, and
Outlook.  U.S. Bureau of Mines, Washington, D.C.  October 1981,
p. 67.

Copper Imports on Preferential  Tariff.  Japan Metal  Journal  (Tokyo).
December 8,  1980.  p. 3.

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. '

Brass Rod Production Cartel  Starts.   Japan Metal Journal  (Tokyo).
July 6,  1981.  p. 1.

Smelter Pollution Abatement:  How the Japanese Do It.  Engineering
and Mining Journal.  May 1981.  p. 72.

Reiber,  Michael.  Smelter Emission Controls:  The Impact  On  Mining  and
the Market for Acid.  Arizona Mining and Mineral Resources Research
Institute, Tucson,  Arizona.   Office  of Surface Mining.  March 1982.
p. 5-10.
                                     7-37

-------
34.   Custom Copper Concentrates.  Engineering and Mining Journal.  May
      1982.  p. 73.

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.   Arthur D. Little, Inc., Economic Impact of Incremental  Pollution
      Control At Asarco's Tacoma Smelter. U.S. EPA Contract Number 68-02-1349.
      July 1977. p. 111-13.

37.   Asarco, Inc., 1981 Annual  Report, p. 19.

38.   Dayton, Stanley H., Asarco's Troy Mine:  How Rolling The Dice In a New
      District Can Add to Earnings.  Engineering and Mining Journal. February
      1983. p. 40.

39.   Moody's Industrial Manual  1982, Vol. I. p. 58.

40.   The Economic Impact of Asarco Operations, Tacoma Area, Washington.
      City of Tacoma - Community Development Department.  April, 1978. p. 2.

41.   Western Copper Operations Continue Cutbacks and Closures.  Pay Dirt.
      April 1982. p. 20.

42.   Asarco, Inc., SEC Form 10-k. December 31, 1981. p. A3.
                                    7-38

-------
                   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
May 8-15, 1979
July 23-24, 1979
                    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 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.
Emission source testing at ASARCO Copper Smelter,
Tacoma, Washington.
Emission source testing at Phelps Dodge Copper
Smelter, Ajo, Arizona.
                                A-2

-------
     Date

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, 1983
                    Activity

 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.

 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 high 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. 	
 (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
 recommendations 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
Agency Guidelines for Preparing
Regulatory Action Environmental
Impact Statements (39 FR 37419)
 Location Within the Background
   Information Document  (BID)
1.  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
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 2.2.

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.
                                                       (Continued)
                                 B-3

-------
        INDEX TO ENVIRONMENTAL IMPACT CONSIDERATIONS (Concluded)
Agency Guidelines for Preparing
Regulatory Action Environmental
Impact Statements (39 FR 37419)
   Location Within the Background
     Information Document  (BID)
       Regulatory Alternative II •

          Environmental Impacts




            Costs




       Regulatory Alternative III

          Environmental Impacts
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.
The implementation of Regulatory
Alternative III would require
elimination of all arsenic emissions
from the ASARCO-Tacoma smelter.
This could not be accomplished without
closure of the smelter.   The economic
impact of closure of the smelter is
considered in Section 7.0.
                                 B-4

-------
      APPENDIX  C





SUMMARY OF TEST DATA
        C-l

-------
                              APPENDIX C
                         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 C&W Roasters), 2 reverberatory smelting furnaces,
4 Fierce-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
                   o
flue by two 73.6 Nm /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.
                                                            3
     Offgases from the roasters, which average about 3,570 m /min
(126,000 acfm) at 260°C (500°F), are combined with the exhaust gases
                                                            •3
from the ancillary fugitive emission control systems [850 Mm /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
* Starting September 1978 only the two tilting furnaces are being used; the
  hearth type furnace is no longer used.
**0peration of electrolytic refinery was discontinued in January 1979.
                                C-3

-------
is 9,950 m2 (107,100 ft2).  The baghouse  is designed to  effectively
             3
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  smelt,  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 flue 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
                                                                 3
time.  The matte flows through a cast copper launder to  a 4.25  m
       o
(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 m3 (100 ft ) 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
                        3
system total about 700 m /min .(25,000 acfm), while the ventilation
                                                           3
requirements for the slag skimming system total about 600 m  /min
                                C-4

-------
 (21,000 acfm).  Captured  emissions  from  both  systems  are  currently
 controlled by an electrostatic  precipitator prior  to  venting  from a
 stack  to the atmosphere.
     Process gases from the  reverberatory  furnace, which  average  about
        o
 1,415  Nnr/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/min (215,000 acfm)  at 132°C (270°F),
 then enters 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 m
          p
 (71,250 ft ).  Each section  contains 84  pipes  measuring 30.5  cm (12  inches)
 in diameter 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).
 The exiting gases, about 7,740  actual  m3/min  (270,000 acfm) at  110°C
 (230°F), are discharged through a large  flue to the smelter main
 stack.
     Matte from the reverberatory furnace  is transferred  to one of
 four Fierce-Smith converters.   Three of  the converters measure  4.0 m
 (13 feet) in diameter by 9.1 m  (30 feet) in length, while  the fourth
 converter is 3.4 m (11 feet) in diameter and 7.9 m (26 feet)  long.   In
 addition to copper matte, smelter reverts and  cold dope materials are
 also processed.   Typically,  only two converters are on blow at  any one
 time.   A converter cycle normally takes  from 10 to 12 hours.  With
 dilution air, the offgas flow per blowing converter is about  1,130 Nm3/min
 (40,000 scfm) and contains from 3 to 4 percent S0?.  Blister  copper
 produced is transferred to one of three  anode  furnaces for refining
 and casting.   The slag skimmed from the  converters is recycled  to the
 reverberatory furnace.
     Offgases from converter blowing operations are captured  by
water-cooled hoods and pass  through a  series  of multiclones and a
 settling flue for coarse particulate removal  prior to entering  the gas
cleaning 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  S02 and is
 capable of processing  652  Mm /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 dimethyl aniline (DMA)  to absorb the
 S02 in the gas stream  and  uses  steam  stripping from  regeneration.  The
 100 percent  concentrated SOp 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 fig  (392 tons) of calcine per day.  The calcine  is  taken  by  larry
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
        2
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 Nm3/min (177,000 scfm) at  200°C  (400°F),  then passes
through a spray chamber where  it is cooled to  about  110°C  (230°F)
prior to entering an electrostatic precipitator for  particulate removal.
The precipitator consists of seven parallel chambers.   Each  chamber
                                                                o
has four fields in series and  has a total field.volume of  535 m
(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 Nm  /min
(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 110°C (230°F).  This process  occurs prior to treatment
in an electrostatic precipitator for particulates.
                                C-7

-------
     The precipitator consists of four parallel chambers9 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 SO,, removal.  The acid plant has a normal production
rate of 408 Mg (450 tons) of acid per day.  Either 93 or 98 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 which 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 other 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.
                                                                   0
     The fugitive gas flow through the  baghouse averages 14,100 Mm /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
384 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
                                                                         p
compartment.  The total net cloth area  of the baghouse  is  about  19,700 m
(212,400 ft ).   The baghouse was designed to effectively treat  15,282
actual m /min (540,000 acfm) at 54°C (130°F) 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
                  q
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
                     2
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 1.8 m (70 in.) in depth.
     The fluidized air reacts with the sulfur contained in the sulfide
ores to form S0£ 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 (85 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 m  .  The furnace  is
equipped with six carbon electrodes.  Each is 1.65 m in diameter.  The
electrodes are energized by three transformers.  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 Nm3/min (15,000 scfm) at 645°C (1,200°F).
     Matte from the electric furnace is transported to one of three
Fierce-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
                                                                    3
(1.5 tons) of cold dope per hour.  The offgas flow is about 2,832 Mm /min
(100,000 scfm) per blowing converter because of excessive  air infiltration
and typically contains about 2.5 percent SOp.  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
                                                3
operations are combined.  Approximately 2,070 Mm /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,
                                                   o
and converter process gases totaling about 5,664 Nm /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 m long and is equipped  with
10 sonic spray nozzles.  Water requirements range  from  265 to 303 liters/
min (70 to 80 gpm).  The cooled gas stream then passes  through tv/o of
three fans (one is standby), each with a capacity  of 6,230 actual
  o
Mm /min (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                                                     2
1,656 m .  The  baghouse  net  cloth area totals  about 29,802 m .   The
                                                           o
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   PHELPS 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

-------
precipitateSj limestone, and recycled flue dusts.  In addition, converter
slag is returned and processed.  A charge, usually 1.8 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 exiting gas streams  then enter a common plenum chamber for mixing
prior to treatment in a  hot electrostatic precipitator which is designed
to handle 4,250 actual m3/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
                                2
total collection area of 3,860 m .  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
through an electrostatic precipitator (ESP).  The ESP has two independent
horizontal parallel units with three fields each, which are designed
to handle 5,940 actual m3/min  (210,000 acfm) at 340°C (650°F) and
95,100 pascals (13.8 psia).  Total ESP collecting surface area is
2,770 m2 (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.
The cleaned offgases are then processed in a single absorption 544 Mg/day
(600 ton/day) acid plant for S02 removal.  Either 93 or 98 percent
sulfuric acid can be produced.  The acid plant tail gas is ducted to
the main smelter stack.
     Simultaneous inlet  and outlet arsenic emission measurements  were
conducted by EPA on July 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
                                                                    o
 roll-out mode.   The secondary  system  is designed to handle 1,980 Nm/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 cm (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
                      2
 system  is about 850 Nm /min  (30,000 scfm) and provides  considerable
draft to the  vent for several feet.   The matte  runs  into  a  2.4 to 3 m
 (8 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 cubicle has ventilation  ducts  leading
from it through a fan, and the gases are discharged  through the  main
                                                               o
stack.  The ventilation rate from the two  cubicles is 1,700 Mm /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
               q
is about 850 Mm /min (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  normal  limits.
The arsenic test results are summarized for both systems  in Tables C-48,
and C-50.  The particulate/ S0? test results are summarized 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
                                       3
of combined gas streams is about 708 Nm /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 350°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
Pierce-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
18 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 tons/hr).  The anode copper is  loaded  on
rail cars and sent to a copper refinery for further processing.
     Fugitive emissions escaping the primary hooding on the converters
are captured by secondary hoods designed by the company.  The secondary
hoods have sliding doors 8.8 m (29 ft) long, 5.8 m (19  ft) wide, and
5.5 m (18 ft) high and cover the converter to the operator  floor
level.  The design gas volume handled by the converter  fugitive system
is about 2,830 Nm3/min (100,000 scfm), and the temperature  of the
gases varies from 150 to 260°C (300 to 500°F).
     Emission measurements were conducted by EPA during July 25-26,
1978, in the fugitive converter duct.  The location was downstream of
all the converters.  Three emission measurements were made  for uncontrolled
arsenic.  During the tests, process conditions were closely monitored,
and testing was conducted only when the process facilities were operating
within normal limits.  The test results are summarized  in Table C-52.
C.6  PHELPS DODGE-DOUGLAS
     The Douglas Reduction Works is a calcine fed smelter producing
about 322 Mg (365 tons) per day of 99.6 percent copper  anodes.  Copper
anodes are sent to a copper refinery for further processing.  The
major units at the smelter include 24 roasters, 3 reverberatory furnaces,
5 converters, and 2 anode furnaces.
     The roaster process consists of 24, 7-hearth Herreshoff roasters
arranged in two parallel  rows of 12.   Only 18 are normally  in operation
at a time.  The roasters are standard Herreshoff with a shell diameter
of about 6.7 m (22 ft).  Natural  gas is introduced near the bottom of
the roasters.  As the hot gases rise, feed material introduced at  the
top is forced down through the roasters by the use of rabble arms
which spread the feed around each hearth level and through  openings at
each level.  About 154 to 163 Mg (170 to 180 tons) of calcine per
roaster is produced each day.
     Calcine from the roasters is discharged into holding hoppers
where it is transferred by gravity to larry cars and delivered by  rail
to the reverberatory furnace.
                                C-17

-------
     The roaster calcine discharge emission control system consists of
hoods covering the larry 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  the material  into the  smelting  vessels.
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 SO- 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/rnin
(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,
which 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 m3/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 about 425 Nm^/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 mid-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
plant.
                                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
                   o
air averages 425 Nm /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 SOo 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
              o
average 623 Mm /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 dusta 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
1^0).  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  Fierce-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 Mm /min (21,000 scfm) 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
 cyclone exhaust, which has a dust loading of  about 57.2  g/Nm3 (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
 humidifying  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,
 effectively  cooling  the gas  stream  to about 68°C  (155°F).  The saturated
 gas stream then  enters  the  cooling  section  where  it passes through
 three perforated plates (four on the converter tower) for flow distri-
 bution  and acid  bubble  formation.   The weak acid  flowing across the
 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 mm  (24 in.)  of  water.  The clean roaster
 gas stream (which  contains about 12 percent Sty  is then combined with
 the cleaned  converter gas stream prior to entering  the acid  plant.
     The  reverberatory  furnace offgases  average approximately 3,682 Nm3/min
 (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
 recovered heat is  used  to produce steam.  The  gases  are then transported
 through a balloon  flue  to an  ESP for particulate  control.  The ESP  is
 a  Koppers design and  consists  of four  parallel  chambers.   The  chambers
 have three fields  in  a  series  and a  total collection area of 5,016  m2
          O
 (54,000 ft ).  The gas  retention time  within the  ESP is  about  14 seconds.
 The average gas  velocity is 0.5  m (1.6 ft)  per second.   The  gases exit
 the precipitator at about 288°C  (550°F)  and are subsequently discharged
 through the main stack.
                                                                o
     The offgases generated by converter blowing  total  1,982 Mm  /min
 (70,000 scfm).   They  are collected in  water-cooled  hoods  and then
 exhausted through a gas cooler in which  the gas .stream  is  treated by a
 concurrently flowing, ultrasonically dispersed water  spray.  The
 cooled gas stream  (371°C) flows  through  an  induced  fan  plenum  and into
 an  electrostatic precipitator  for particulate  removal.  The  precipitator,
manufactured  by Western Precipitator,  has two  chambers with  three
 fields per chamber.  The total collection area of the chamber  is
       p           ty
 3,716 m  (40,000 ft ).  Gas retention time  is  about  9 sec.   The average
gas velocity  is about 1.2 m/sec  (4 ft/sec).   Following the ESP, the
gas stream (which contains about 5 percent S02) enters the larger of

                                 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 S02) then enters the double absorption acid plant where it
is dried, the S02 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 P1g (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
S02 contained in the gas stream is then converted to SOg.  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
                                                               O
absorbed by 98 to 99 percent acid.  The gas stream then enters the
final converter pass where nearly all the remaining S02 is converted.
                                C-24

-------
The gas stream then enters the final absorber where the  last  traces  of
$03 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 S02 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 perfarmed 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 the 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
for S02 removal.  S0£ content of the converter and flash furnace gases
usually range between 6.5 and 7.5 percent.   The acid plant S02 removal
efficiency is 99.7 percent.  Outlet gases which contain  140 ppm S02
are vented through the main stack.  A 98 percent sulfuric acid is
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 SOo 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 modes 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

-------
CO
I—
oo
 O)

La

Ol
•5.SL
!£>
i/i

V)
en—.
c c
-,- 0
fQ O
to o
—1



4-1
p— C
O 01
hi.
O 3
^J O*
LU








VI >)
VI 4-1
Ol —
u p—
o —
&. U
a. 10
u.












4-1
~

Z



Arsenic
4-1
Ol
. 3
O
•g
iO
Ol
*c

VI V)
— s.
E ot

+ *o
o
a. u
LU •*•
4-1 a.
5i2

VI
O)
4-1
L.
Ol
g
a
u
**»
c
IO

a.

- 5


«
c
w Q
cr» ^
o "u
o <
s.0
**" **^

Q.
Ol

IO
U*3
"*~ —
01 4J
VI U OJ
L. iO O
«C a. co
c
°U .*
Ol U
4J IQ
e 4->
Ol I/I
01 C
0|
01
CO







c


,_.
en
c

01 ex
U IO
e
s- en
VI
o"!

U Ol
01 4J

w 10
Ol S
Ol
cs














Arsenic
en
c
u
Ol
1
01
,0
Ol
CO







Ol
c
o





JM*.
Jin

u
o
-Q
v>
u
Ol
^
Ol
c
a
o











(U
4-1

"3
4-1
u
IO
a.

_^
,u
4-1
VI
e
IO
E






































10
^a"s
Ol 4J
CXJ v) I.
O S. 10
to «c a.
01
3
O
•«
•0
Ol
"c

^"^
V
/•J
E cu
10 oo
J= UJ
u
^^**—
is





C 9)
•0 u
v> c
s_ u
Ol 3
4^ •«•
VI
SLf
C o
j= "S
4-1 U
U Ol
10 ja
0) i.
J= Ol
Zi Ol
«— 1.

VI
10
X
Ol
F—

o
VI
IO
0 0.
CJ
£1£ ^^

to
Ol

J2
• •^ u
c —
Ol 4-1
VI U CJ
i. 10 O
«S C. UO





Ol
"5
o
c S

VI

en
L. Ol
10 •>
J= O
U -C
VI IO
5
Ol

jj
j^

VI
•o

u
Ol
4^
VI
«
£
J=
4W
L.
Ol
J.T
4J
3











01

10
U *3
••- U
£• •*—
0) 4J
VI i-
L. 10
•c a-





—
"c





Ol
VI
3
Q
en
CO






Ol

IO
c
3"en
n- c
u ol
o 10

fe.2

S J.
Ol














u
c
Ol
CNJ VI
0 i.
4-1
01
4-)
3
O
•o
c

Ol
"c





Ol
VI
3

en
IO
CO








en
g;
•O
"3
J^
Ol

Ol
• c
o











Ol
4J
IO
"3 _o
4J Ol
L. CM VI
10 O L.
Q. to <





4-J
Ol
"c







Ol
c
.i



I/)

g
J

k
i .
o
y
Ol
VI

Ol
4-1

Ol
c
o

2
Ol CJ
enz
•c
o -
O VI O
iO u
VI >,—
ex. 10 x
— — o>
44 C. S
.c
a.



Arsenic
Ol
^J
0
•o
IO
4-1
Ol
"c





Ol
—
Q
en
IO
CO








i.
Ol
VI
V
o
u
J=

Ol

i;
1


Ol

•o
S

VI
a.
o

a.
Ol

IO
"3
u
i. O-J
•0 O
o. to

























Ol

2
y
VI





•
Ul tl
-B C
— C
o IM
O L.



                                       C-27

-------
ID
 O)
•a
 o
CJ
CO

on
UJ
  i
 o

 0)
 P—
 _a
 ns




11


in
II
^-1_ ^t
^ A3
« u
10 S





4J
»— C
O .
V) 4J
01 —












jj
e
re

cv



u
o>
VI CSJ
£S

^J
«
i
o
1 '
Ol
IE





Ol
VI
3
5
en
re
CD







Arsenic roasters
c
o

en
e
j=
VI

^E


re

S 8 .
K  u o 01 ^^ re
I^Ilsl1"





o. ex. ev c.
V5 CO  en
v. 4-1 re C a
ai re — O —j
Ol
Of

















in
Ol O
•— *Oi
u u to
Ol Ol >>
VI VI JS
1- U

Ol
3
O
•o
re
at oi





Ol
VI
3
2 o.
en e
S £







u
O) Ol
VI (J
re e
£ «2
— Ol
5 1
Z' «£




















U
01
VI



0)
"c



t
*jj
re
cv
5







Converters








^ _t

Q

O)
gj
£J
u

^



u
C CM
Ol C
VI tO
£
^
•o
Ol
u
0
u






Ol
c







Converter
Secondary hood





j=
re
^•^
™*^

re

§i
»*






u
"e esj
Ol O
tn to
u
^
•o
2
*o
c
o
u
c






01
i







Ol
S,
u
o
1
"o
ce




















u
C CM
O) O
vi to
u
3
•a
2
"o
c
o
e
=3






01
i







Noranda furnace
Matte tapping




















u
e CM
Ol O
VI CO
u
«£
U
"0
1
e
o
u






Ol
c







en
c
1
4-1
en
re
t/i




















u
01
VI
«£



Ol
O





U
Ol
1
u
to






(
I Concentrate dryei

















Ol
re
u

ewe
01 re 01
VI Q. VI

Ol Ol
"a "=>
0 0
1 "S
re re
4J JJ
Ol Ol
"e "e
i- j=
01 3
J3 U
E 0) U
re vi vi

>, en 3
re re 4J
U -Q c
a. 01
in >
u
*j (-
CJ Ol
5 4-1
*^j Ol

• c
i. O i-
Ol U Ol
VI T3 VI
re e re
E * e
•0 Oi -a
01 u *
"5 >•- 3
u. u.
re
e
re

c
£
•
re
•a 4-1
re c — '
•o o o
e u u
o re 01
u c c
re ce c
c o
< ^£




U
e
Ol
VI
«c



Ol
1



^}
y
re
"5.



c
re

t_
Fluid-bed roaste
converters


re
c
o
i!
•S.

•
c
Ol
75!

2;



                                           C-28

-------
LU
Q-
>•
I —

LU
_J

SI
•=C
GO

Q

ej;


1 —
t-H
	 1
>— 1
C_3
2

GO
GO
LU
O
Du
>_
CO
GO
LU
_J
CQ
«=c
f—
Q

1 —
GO
LU

1 —

LU
5
— J
H-H
i
C£*
Q-
Q
 •> MM
LC) O *JD O
i~ t CM CO ^~
I 1 t 1
CJ CJ CJ CJ
» t * *
^^ *;j~ CT^ if) Q^ ^^ ^o
*^. **» ^H n co to to
^r I I I i i it
^ooooo titi
* » M M M * •
«f S 2S SS g o, SS E3S co S SS3 35 S Z g S
o o oci  ci titi titi ii ci cici ticS ' ci ci ci J,



,2 ^ 1o 1o re "re
000:30=10 00*30^0 o 0*3 0*3 o o o o
1!!! II! I!I III ill! I I 1 |

s-
Ol

-a i- "£T • t1
3 0! O a)
S- -O JD .0
0 E E *J E
l^lftlsl •sglggl g a
tlllllllllliili Jill

re
S "i S e
|l'illil
C 4J S- -2
re o> <-> o> *j
H O *J «r- QJ ^_
|-c'oJo)c: 3S-oiSS ^ S S-3
•W 3 •> C "' ""5? = = & *° 
-------
-o
CD
-a
3
O
£=
0
C— 3


LU
O_

H-
UJ
_J

•B^
««

Q

eC


1—
*""!
'

0
1 1
L'_i
oo
oo
UJ
o
o
Q;
O_

co
oo
UJ
CO
I—
cC
1—

!„]



OO
UJ
I—

UJ
1
ID
f ^
f_|

1—
Op
Q.
Q
2:


CJ
t— i
Ul
oo
o:
•=£
O
^~~
X
1 1 1
CD

KH


CNJ
1
0
OJ

JO
(B
"""



















in CJ
fe-
rn Stf
.S »P
§-O
C
2= 
,2 £


QJ
CL
0.
"n.
I



cr
0)
»^-
3
o-
UJ


Q
c
o
CJ







c:











^

•^
£;
o
re
u.

in
a;
0
o

CL.























^
OJ
1
CJ
CO
OJ
CJ




u
IV
tn
i-







1 «
c
re

CL
•o
u






o
in
re
CL.
UJ
S
o;

^














tn
s_
0)

L.
OJ

C
o

























O i-H
LO LO
CJ O


0)
re
U 3
T- O
C "r-
01 4->
in s-
s- re

"oi
3 O>
CJ
CD CJ

S- i—
ai i—
> 3
O

























OJ
to
"




o
c
OJ
in
i.












01
c
o
z




re
c
en
re
I
o
u
o>

c
&












01
in
a

o.
3
O
r—
O






r^
•9-
1
CJ
^
to
•a-
CJ
to"
«*•
t
CJ
•I
•a-
J-,

<*r
^f
t
CJ
CM
^f
"




u
c

1



B-
LU
in
T- w
E s-
O)

o. o
CO O
1 1 1 {J
4-> T3
0 C
ic re



0


CD
en
•g
Q
CL

"flj
0.




in
£_
QJ

t-
01

C
o
u

I

JJ
c
to

CL
•CJ
u
•a:





















to co
OJ OJ
1 1
CJ CJ
to i-»
OJ OJ
1 1
CJ CJ


01
re
U 3
T- O
Cf-
0> 4->
tn i-
s- re
•fo.









01
in
3
O
en
re
eo





0
re
CL.
UJ
i
cr:

CO










g,

•o
3

O)

s-
>
I

























OJ
tn
i
o




u
c
0)
tn
i.












01
o
^•
0
en
"re
•o


O)
en
•o
in
CL


O.





in
•a
o
o
.c

>)
i.
to
•a
c:
o
tj
QJ
in
s.
a>

s-
O)

o
u



















1 —
1
CJ
•t

1
«,
tn
ci




u
c
01
tn
i-









0)
in
3
O
en
a
ca






re
u
re
1—
i
O


s





tn
in
01
u
o
i.
CL
CJ
ai
in
re
u

r—
re
at

























j^.
m
i
CJ




u
c
ai
tn
s-









s-
OJ

3
U
CO




re
c
en
re
^
0
u
0)
c
c
:S










tn
S-

•o

0)
(O

QJ
U
1
C-30

-------
              Table C-3.  SUMMARY OF ARSENIC TEST DATA — ROASTER
                     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
1
9/15/78
80
a
a

173,621
201

4.4
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
0.2
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
~ * J.
0 2
*-> • L,
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
                    BA6HOUSE 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
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
9/15/78
120
a
a
171,887 174
191
5.9
0.2
20.4
0.70
0.00027
0.00026
0.396
0.00028
0.00027
0.416
2
9/15/78
120
a
a
,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
Average

120
a
a
175,064
187
5.5
0.2'
20.4
0.67
0.00027
0.00026
0.412
0.00040
0.00039
0.612
Percent Isokinetic
99.5
100.1
102.3
                                   C-32

-------
                 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
0.5910
0.5273
72.27
0.5912
0.5275
72.29

During this sample run the metallic arsenic process may not have been operating
                                   C-34

-------
                 Table  C-5.   SUMMARY  OF  ARSENIC  TEST  DATA —
            ARSENIC  KITCHEN  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
CO,
o
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Tb/hr
1
9/24/78
96
a
a
14,505
133
3.6
0.2
20.8
0.32
0.7503
0.7484
93.23
0.7504
0.7486
93.23
2
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
3
9/25/78
96
a
a
17,560
140
2.9
0.2
20.8
0.49
0.6593
0.6400
99.18
0.6695
0.6402
99.18
Average

96
a
a
16,218
136
3.5
0.2
20.8
0.46
0.6909
0.6755
95.54
0.6944
0.6756
95.54
Percent Isokinetic
99.2
97.5
94.8
                                    C-33

-------
                  Table  C-7.   SUMMARY OF ARSENIC TEST DATA —
                ARSENIC  BA6HOUSE  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. %):
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
1
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
2
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
168

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

-------
                 Table C-8.  SUMMARY OF ARSENIC TEST DATA -
                  REVERB ESP 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
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
I
9/15/78
116
a
a
441,557 454
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
,539 443
214
5.1
0.0
20.0
1.17

0.00897
0.00863
35.04

0.00915
0.00881
25.77
3
9/18/78
108
a
a
,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
20.0
0.68

0.00968
0.00921
36.90

0.01000
0.00953
38.15
Percent Isokinetic
104.2
107.7
102.2
                                    C-36

-------
                 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
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
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
2
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
0.0
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-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
0.07533
0.07427
12.11
0.07591
0.07484
12.20
Percent Isokinetic
90.5
91.0
90.0
                                    C-38

-------
                 Table C-9.  SUMMARY OF ARSENIC TEST DATA -
                  CALCINE DISCHARGE, 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
9/20/78
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
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
0.0
20.0
0.54
0.1399
0.1384
1.784
0.1418
0.1724
1.806
Percent Isokinetic
96.6
94.2
108.1
                                    C-37

-------
              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
     S02

Emissions - Arsenic

   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 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
2
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
116
41.2
98.0

140,927
173

6.7
0.0
16.6
a



0.0060
0.0040
7.200

0.0100
0.0068
12.12
2
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
3
6/28/77
120
42.7
395.5

56,040
231

9.4
0.4
20.5
a



0.0312
0.0188
14.99

0.0380
0.0229
18.25
Average

119
37.6
222.1

115,577
205

7.2
0 3
18.6
a



0.0109
0.0069
9.807

0.0147
0.0094
13.58
Percent Isokinetic
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



0.2123
0.0646
74.95

0.2475
0.0753
86.81
Percent Isokinetic
103.8
107.1
112.7
                                    C-42

-------
                  Table C-15.  SUMMARY OF ARSENIC 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
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
6/26/77
120
41.2
98.0
19,759
787

14.2
8.1
9.3
a

0.1280
0.0404
21.68

0.1289
0.0407
21.83
2
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
3
6/28/77
120
42.7
395.5
25,981
614

25.2
8.2
9.2
a

0.6096
0.1949
135.7

0.6462
0.2067
143.9
Average

120
37.6
222.1
22,599
722

16.6
8.1
9.3
a

0.5272
0.1692
106.8

0.5444
0.1747
110.5
Percent Isokinetic
95.8
                                           97.7
                         113.4
                                   C-43

-------
                Table C-16.  SUMMARY OF ARSENIC TEST DATA —
               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 filter 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



0.1218
0.0405
191.6

0.1344
0.0452
210.9

These data are derived from Tables C-13, C-14, and C-15.
                                   C-44

-------
                  Table C-17.   SUMMARY OF ARSENIC  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 - 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

6.8
3.5
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-19.   SUMMARY OF PARTICULATE 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
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
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



1.474
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
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

22,504
751

9.6
9.9
3.4
a



0.9863
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 PARTICULATE TEST DATA  -- 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 - Particulate
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
Percent Isokinetic
1
6/29/77b
120
43.5
a

132,163
439

12.2
4.3
12.9
a



1.9854
1.0180
2,249

2.0468
1.0431
2,318 2
c
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
c
Average

120
45.5
a

130,494
444

12.5
• 4.6
13.7
a



2.2319
1.0658
2,494

2.2801
1.0862
2,548

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  INLET, 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
105
a
a

58,353
431

2.1
0.0
17.6
a



0.9656
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 ARSENIC 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
0.0
0.0
14.7
a

0.0004
0.0003
0.203

0.0005
0.0004
0.298
Percent Isokinetic
73.0
76.2
96,4
97.4
                                         C-52

-------
      Table C-25..  SUMMARY OF ARSENIC TEST DATA — CONVERTER BUILDING
                    BAGHOUSE INLET, 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
1/18/78
101
41.7
212.0

521,956 528
• 100

1.0
0.0
20.5
a



0.00270
0.00240
12.11

0.00272
0.00242
12.12
2
1/19/78
100
32.7
116.1

,463
97

0.4
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
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
0.0
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
                    BA6HOUSE 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
1/18/78
160
41.7
212.0
437,609
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
526,565
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
490,455
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

0.00006
0.00005
0.241

0.00006
0.00005
0.246
Percent Isokinetic
105.7
101.1
102.6
                                    C-54

-------
               Table C-27.  SUMMARY OF PARTICULATE TEST DATA —
           CONVERTER BUILDING BA6HOUSE INLET, 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
105
a
a
435,427
115
1.2
0.0
20.5
0.017

0.02.63
0.0235
98.25

0.2812
0.2507
1,049
2
1/18/78
100
41.7
a
514,279
113
1.4
0.0
20.5
0.001

0.0080
0.0202
101.9

0.2686
0.2346
1,181
3
1/21/78
100
43.8
a
510,318
115
1.4
0.0
20.5
0.001

0.0309
0.0276
134.5

0.031
0.027
134.5
Average

102
42.8
a
486,675
114
1.3
0.0
20.5
0.006

0.0215
0.0238
111.5

0.2749
0.2427
788.0
Percent Isokinetic
109.7
94.5
                                                        96.3
                                    C-55

-------
       Table C-28.  SUMMARY OF PARTICULATE TEST DATA — CONVERTER BUILDING
                    BA6HOUSE 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 - Parti cul ate
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
0.0018
9.76

0.0867
0.0770
368.5
Percent Isokinetic
98.0
97.7
103.7
                                   C-56

-------
Table C-29.  SUMMARY OF PARTICIPATE 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
02
S02
Emissions - Parti cul ate
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
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
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
3.0
20.5
a



0.0643
0.0577
112.0

0.0909
0.0788
157.9
Percent 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
I
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
o.o.
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



0.0032
0.0029
0.214

0.0033
0.0031
0.217
Percent Isokinetic
93.4
100.7
94.0
                                  C-58

-------
     Table C-31.  SUMMARY OF PARTICULATE 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 - Particulate
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
fb/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
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
2
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

0.0022
0.0019
0.488

0.00003
0.0019
0.489
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
• o2
S02
Emissions - Parti cul ate
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
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
2
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 ARSENIC 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
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
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



0.3547
0.1414
230.6

0.3638
0.1450
236.4
Percent Isokinetic
92.9
101.9
100.7
                                  C-62

-------
     Table C-35.   SUMMARY OF ARSENIC TEST DATA — SPRAY CHAMBER/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
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
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
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
0.0
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
18.8
a

0.3725
0.1442
267.8

0.3818
0.1479
274.7
Percent Isokinetic
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.
t
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
4/20/77
120
97
2,173
156,467
526
12.2
2.4
16.9
a

0.4588
0.1702
601.6

0.4539
0.1722
608.7
c
2
4/21/77
120
92
1,490
160,352
517
12.5
2.2
18.9
a

0.3669
0.1415
504.1

0.3791
0.1462
520.9
c
3
4/22/77
120
92 .
1,914
164,291
471
10.0
0.1
19.7
a

0.2807
0.1182
395.3

0.2908
0.1224
409.4
c
Average

120
93.7
1,857
160,370
505
11.6
1.5
18.5
a

0.3595
0.1433
498.4

0.3733
0.1465
511.1

*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
02
S02
Emissions - Arsenic-
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
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
2
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
3.9
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 PARTICULATE TEST DATA — SPRAY CHAMBER/BA6HOUSE
                       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
C02
02
S02
Emissions - Particulate
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
Percent Isokinetic
1
4/25/78
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
97.4
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
98.5
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
100.9
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

                             C-66

-------
     Table  C-39.   SUMMARY  OF  PARTICULATE  TEST  DATA  —  SPRAY  CHAMBER/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. 35):
Water
C02
' 02
S02
Emissions - Part icul ate
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
fb/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 PARTICULATE 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
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
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

161S671
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
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
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 DOD6E-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
0.1265
0.0621
64.7
0.1351
0.0663
69.0

                             C-70

-------
Table C-43.  SUMMARY OF ARSENIC TEST DATA — REVERBERATORY ESP
                   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
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
Percent Isokinetic
1
7/13/76
120
a
53.7

68,030
595

15.3
a
a
a



0.0603
0.0303
35.2

0.0919
0.0462
53.6
145
2
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
147
3
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
136
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

                            C-71

-------
     Table C-44.  SUMMARY OF ARSENIC TEST DATA — CONVERTER ESP
                     INLET NO. 1, PHELPS DOD6E-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 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

0.000024
0.000020
0.0056

0.000073
0.000063
0.0062

*Test run aborted.
                                  C-72

-------
Table C-45.  SUMMARY OF ARSENIC TEST DATA — CONVERTER ESP
               INLET NO. 2, 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
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
Percent Isokinetic
1
6/13/78
144
17.7
38.9
34,282 28
389
0.3
0.0
20.0
2.59
0.000003
0.000002
0.0095
0.000003
0.000003
0.0097
123.2
2
6/14/78
'144
19.7
63.0
,312 29
358
0.8
0.0
20.0
3.03
0.000010
0.000009
0.0026
0.000013
0.000011
0.0031
92.1
3
6/15/78
144
17.7 .
95.6
,265
404
0.0
0.0
20.0
7.60
0.000005
0.000004
0.0011
0.000013
0.000011
0.0029
99.0
Average

144
18. .3
65.8
30,621
384
0.4
0.0
20.0
4.41
0.000006
0.000005
0.0044
0.000010
0.000008
0.0053

                             C-73

-------
     Table C-46.  SUMMARY OF ARSENIC TEST 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
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 29
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
,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
0.000020
0.000017
0.0064
0.000029
0.000025
0.0088
Percent Isoldnetic
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
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
108
17.7
38.9
47,556 43
140
1.2
0.0
20.0
0.24
0.0000020
0.0000017
0.0009
0.0000030
0.0000026
0.0013
2
6/14/78
108
19.7
63.0
,862 36
164
0.5
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
0.0
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
0.6
0.0
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 DOD6E-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
°a
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
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
72,351
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
69,333
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

0.00058
0/00052
0.340

0.00069
0.00063
0.406
Percent Isokinetic
104.2
95.8
97.2
                                  C-76

-------
  Table C-49.  SUMMARY OF PARTICULATE TEST DATA -- MATTE TAPPING
                      OUTLET, PHELPS DOD6E-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
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
2
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
Average

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
Percent Isokinetic
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 - nrin.
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 Isold netic
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
101.1
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
101.1
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
• 104.4
Average

161
22.1
2.3
86,353
154
0.8
0.0
20.0
0.31
0.00208
0.00190
1.536
0.00211
0.00192
1.556

                        C-78

-------
    Table C-51.   SUMMARY OF PARTICULATE TEST DATA — CONVERTER SECONDARY
                    HOOD OUTLET, PHELPS DOD6E-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
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
5/10/78
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
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
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

0.00014
0.00012
0.0850

0.00026
0.00021
0.1563
Percent Isokinetic
85.3
98.6
99.1
                                  C-80

-------
     Table C-53.  SUMMARY OF ARSENIC TEST DATA — CALCINE/ROASTER FUGITIVES
                       BA6HOUSE 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
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
                       BA6HOUSE 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
0.000037
0.000031
0.0100
0.000107
0.000092
0.0293
Percent Isokinetic
96.5
95.0
90.0
                                  C-82

-------
 Table C-55.  SUMMARY OF PARTICIPATE 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
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
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 PARTICULATE TEST DATA — CALCINE/ROASTER
     FUGITIVES BA6HOUSE 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
CO 2
S02
Emissions - Particulate
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
Percent Isokinetic
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
93.6
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
-87.7
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

                                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
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
Percent Isokinetic
1
11/14/78
90
a
a
43,489
166
18.1
a
20.0
0.06
0.00003
0.00002
0.0120
0.00003
0.00003
0.0143
102.8
2
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
98.4
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
96.5
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

                                C-85

-------
  Table C-58.   SUMMARY OF ARSENIC TEST DATA — ACID PLANT INLET,
                          KENNECOTT-MAGNA SMELTER

Run Mo.
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
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



0.0029
0.0025
1.158

0.0037
0.0031
1.461
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
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
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



0.00016
0.00013
0.0578

0.00025
0.00022
0.0899
Percent Isokinetic
89.9
93.6
97.0
                                  C-88

-------
  Table C-61.  SUMMARY OF ARSENIC TEST DATA — CONVERTER FUGITIVES (FULL
                       CYCLE), 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
SO 2
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 TEST DATA — ROLLOUT CONVERTER FUGITIVES,
                               KENNECOTT-MAGMA 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

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
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

*No corresponding test run was performed at the outlet.
                                       C-91

-------
   Table C-64.   SUMMARY  OF ARSENIC TEST 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
Probe, 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
0.00006
0.00006
0.007
0.00018
0.00016
0.027
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
Percent Isokinetic
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
107
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
108
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
95
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

                         C-93.

-------
Table C-66.  VISIBLE EMISSIONS OBSERVATION DATA, EPA METHOD 22—
   ROASTER CALCINE DISCHARGE INTO LARRY CARS, ASARCO-TACOMA

Run
ti_
NO.


1
2
3
4
5
6
7
8
9
10
11
12
13

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
Observer


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


% time
emissions
observed
0
0
0
0
0
0
0
0
0
0
0
0
0
Observer


Duration of
operation,
mi n: sec
1:15
2:40
1:20
1:23
1:52
1:42
1:13
1:20
2:49
1:48



2


% time
emissions
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
1:55
1:42
1:13
1:20
2:50
1:48
2:30
1:42
3:04

Mean
% time
emissions
observed
0
0
0
0
0
0
0
0
0
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
Run3
1
2
3
4
5b
6
7
8
9
10
11
12
13
14
15
16b
17
18
Date
6/24
6/24
6/24
6/24
6/24
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
Observer 1
Duration of
operation,
mi n: 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
% time
emissions
observed
0
0
0
0

0
0
0
0
0
0

0
0
0

0
0
Observer 2
Duration of
operation,
mi n: 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



% time
emissions
observed
1
0
3
0


0
0
0
0
0
0
0
0
0



Duration
operation
min: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
Mean - .
of % time
, emissions
observed
0.5
0
1.5
0

0
0
0
0
0
0
o •
0
0
0

0
0
                                Average
5:26
0.13
Method 22 data for corresponding runs at the matte discharge into the
ladle are presented in Table C-68.


Observations were made only at the  matte discharge into the ladle;
see Table C-68.
                               C-95

-------
     Table  C-68.   VISIBLE  EMISSIONS  OBSERVATION  DATA,  EPA METHOD 22—

                MATTE  DISCHARGE  INTO LADLE, ASARCO-TACOMA
Run3 Date
1 6/24
2 6/24
3 6/24
4 6/24
5 6/24
6b 6/25
7 6/25
8 6/25
9 6/25
10 6/25
11 6/25
12 6/25
13 6/25
14 6/25
15 6/25
16 6/25
17b 6/25
18b 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
operation,
min:sec

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
% time
emissions
observed

0'
0
0
0
0
0
0
0
0

0
0
0
0
Duration
operation
min: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

-------















1
1
CM
CM
f""*
O
1C
LU
^ -, H-
>
CD
• 	 1
CT> OO
vo
1
CJ

cu
r—
~C)
03
t
r^










































































































;
r












CM

&_
CO
i-
CO
in
cS










1— 1

SM
co
>
i_
CO
VI















CO

ItJ
o

rt3
c
3
CO
S 0
M- C T3
t— o co
***V» t!
I/I CO


co co o
s


c
o
C t-
O -M O
- f it) CO
8 J-> S- to
; i. o. c:
j3 o *^
• £
O
c/l -O

S *^- t-
•^ tn CO
1— to in
X S O
CO


•%
C
O
e: -r-
O -M O
'r— O) CO
•M S- U)
«3 CO ••
S- Q. C
^^ S
o


in -a
e co
CO O >
S i- s-
••- in CO
H- in in

*« "s "o
co


M
g*
o
d *1~*
O 4-» O
•^^  O r*«»
C^ Gft








-a JQ-
VO VO CM
CM m in
CM <— i n
i— 1 CM i-H




J3 -Q **i
aOLOLnn^— «cM'-i*OOo
0, ^ fO ^H T-4









CNJOOi— I^CM— •^^^'CM















^^ ^i* ^* ^^ tr^ LO VCD vo vo vo
CMCMCMCMCMCMCMCMCMCM
"^^ '^** *^* ^*X ^*^* ^**fc ***• ^^«» ^^^ *^^
VO VO VO VO VO VO VO VO VO VO


o o -a
<— ^ CM OO ^T Lrt VO ^* OO C3% CD *— '
1— 1 1— 1




«f 1-4
1—4










0

ITS
r™l












•
CO CO
01 -o
03
S™» *
cj -a
^c (/)










































_^ — rf It
r^ »o T~*
r*** ^~ o
1 t/1
C_) C

CU 4J CU
i — S- E
.a o
03 CL a
L,,-, )— -
|— ,^_
E 03
•i — 4J CD
f-
S- cn-r-
03 03 4->
CU r-™ H3
o- oo a.
Q. E
03 CU O
J= CJ
•r— OVr-
O E
O. «r- "O
• -o a>
cu 3-0
CD i—3
S- O i —
03 E CJ
-E -i- E
CJ •«-
1 00 CU
•r- E 4-> CD
"a -I— o r^
r— E 1
E o
•r— OO CU
^i 00 S- CU
01 CU O i —
CJ M- i-i
CD O CU 03
03 S- S- I—
i— a. cu
to ^: E
Q--4-> •!-
CU 03
^: -M "a S-
-P E 03
CD o3 CU
4-» 03 . Q.
fO i • ** *^
to cu 03
E cu ^a to
3 i- O3 E
S- T- r— 3
t_\ C,
. CD E O
E CU -M CD
•r- E E
-a cu T- -r- CM
E .E T3 CM
O 4-> CU E
CL CD O "O
to +-> i- CL o
CU 03 03 to .E
i- .E CU 4->
S- CU CJ S- CU
O T3 to S- S
CJ 03 -r- O
E "O CJ >i
S™ /""i
O CU CD S_
<+- S- 03 O T3
CU i — M- CU
03 S to E
4-> • 03 •!—
03 t/) "O to +-> 03
-O E E E 03 4->
O 03 O T3 _Q
CM T- v- O
CM -P « 4-> CT>
fl3 >™ (O fO
-a > cu > -a 4J
o s_ -o s- o 03
-E CU E CU .E -O
4-5 to 3 to 4->
CU -Q 03 J3 CU O
s: o i— o s: ^
03 -Q cj ~a
C-97

-------
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


Duration
of operation,
min.
14.75
18
16.38

Mean
opacity,
%
1.3
10.3
6

Maximum
opacity,
%
10
30

' 30
Emission data were taken during entire  slag  tapping  operation.
                              C-98

-------
                                        CM









1—
=a-
CJ3
Z;
t— t
Q-
Q.
r—
0
I
-r-- '
to
1
CM
CM

0 =C
3: s:
1— 0
uj o
s: -=c
< ^7
Q- CD
UJ O
« •<
•=t tO
1— [
{__
C£ 2:
1_J_1 |— -j
to
CQ UJ
O CJ3
to «C
O 0
i— i tO
tO i— i
tO O
S- CI3
UJ «sC

UJ tO
	 I
CQ
to
>-H
2z»


9
^^
r-^
1
0
CL)

rO
1—

















(
<
:







CM
S-
O)
s-
* "Ln ^
in (U
C T- tn
 i- CO
£ s- CL a
3 O *^

o

(/> •^
c 
e •»- s_
4J in en
A
e
c2
O -t-> O
•^ « qj
•MS- OJ
«J 0) ••
s- a. =

"V-
o

ui -a
^B 33

S -i— s-
t- tf» OJ
4-> tn in
>« *S "o
(U

c
o

O 4-> tj
••- rtj 4 ^^ f-^ tf^ ^*n ^4 f\j
oocy»cnoocr>cno^eor^»^>sco






O i
»— 1 1—4





to CO
03






co
tn
1"^


*
5 O
jQ <*-

_t_\ i 1
ra ro
•a -a


O 0
•r— «r—
ro ro
£_ s- tTi
cu tu to
CO CO 1
JD .a  tu -o
O)
d) CU 4-5
r— r— E
-Q ja a>
•I— •!— CO
CO CO O)
•r- T- S_
>• > Q.
*r^ y™ s
:-99

-------
    Table  C-72.   VISIBLE  EMISSIONS  OBSERVATION DATA,  EPA METHOD 9-
        SLA6 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,
min.
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,
%



50
30
35
40
20
10
'10


50
 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 slag pot.
                                 C-100

-------







1
1
CM -
irf rv
> 0
/*v ^»
UJ =£
OO C£
cn uj
O CQ
n;
OO UJ

O UJ
1 — ( Di
00
00 CD
>-< H-
l t l ^y
o;
uj rD
—i i—
CQ UJ
f— * rv*
00
i-i CD
t
oo
co cc
r~» uj
i | —
o Q;
UJ
cu >
(~** t~~t
03 CJ



















en
i.

01

-O
o






—
s_
CU
a
S








0)
«a
a

tn
C E O 01
32*31/1 S.

o •
e a
COO)
e o *^- v>

2 aJ*e
^ o

trt
g O «
JJ U» 
4J fO ••
3 Q. £
o o


tn
O 4)
41 W> L.
E v> 0)
§^
o


,.
o -
coo
0-^4*
•^ -*J W)
« i- e
L> 4] *^-
n a. s
a o


tn
4) O 4J
i— S
-^- in i.
-M tn cu
X "s J2
o3
o^otot^^o tococoaa


oi?2SSio SS2§
-HOOO-^O -H— <^4O





o o o o
o o o o

co to m c\j
u? <«f* m tn
*• •• «• ••
o o o o




88g o g








o v in o ?
•^ O O ~ o

f



0 !•* 0 0
oo>oo tocoooco



«rr-»«rtn ^rotom

•— csoo ««^3-o





J3

^.^.— -.~^-~,^,512i5i.SiSi£i
-HCsjM^rtnto^aooio—tM



01 -


03
in
o









Q
•u
cu
o.
LJJ
_>1

-2
-s
C C\J
O CNJ
•M T3
IO O
> .C
S- 4->
CU CU
CO S
1 -^
c ~a
O OJ
i- E
C/l «f~
WJ (O
£ -Q

-------
  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,
min.
a
a
a
1.00
1.25

0.75
1.25
1.25
1.50
1.25
0.75
Average
opacity,
%



17.5
20

23
5
11
12
13
5
Maximum
opacity,
%



30
40

35
10
20
20
20
10
Observer 2
Duration of . Average Maximum
operation, opacity, opacity,
min. % %



1. 00 16 25

1.00 23 35
0.75 ' 23 30





                                    average opacity for all  readings    - 15%
                                    maximum opacity during all readings - 40%
Data were not obtained by Method 9 ,on 6/24/80.
                                  C-102

-------
 CD
 C_5
 oo
 I — I
 a

 0£.
 LU
 I—
 oo
 i — i
 —1 O
 CQ  *»
  I  ^D
  I   A
 CT> «3

 Q  
cr*cn
Ol

Ol ra
4J
2























S-
Ol
4Jt3
VI Ol
•>- en
t— S-
-a >a
• -c
ro v)
0=5




Ol Ol
"1
i!
V)


en








eu 01
1/1 i-
•»- rO
,«• p-
J3 U
VI
c-5
CM



S.
Ol Ol
VI S-
•*- (T3
VI
v>-5







a
a 01

M- '2
a
a<



u
Ol
VI
.
c
"s
51

4- *J
O
U
VI

• •
c

"s

_
a
O CO
4J |5


O
u
Ol
VI

••
£
E
4-1
TT
o
u
01
v>
• •
^_
s
.
k>
u »«
ra
CL
3


^^ ^^ 03 10 co PO ^»
us -r CM *- v






O in o in in a in

o >» co «r CM <— o
CM i—

CM r—


o a in
O en ^— •
~ 0 0







to en ^r p^ r^. to CM
en p^ ^ CM i—


o in to in o in in
en ^ ^— i^ o ^ i-~
•— eo in en CM o o
^•"


rn en CM o en CM
in en CM r—


a a in o o in
a O r— en en i—
CO in en ^- o O


in o in a in o in
r— i— CM CM en en











































•



V)
VI

•a
o



s.
*
c
o
o
CD
V)

s_
Ol


Ol

c
o
u


4J


o
CL
=
o

(O

s-
01
V)
*~*
o
0
i— Ol
ro &.
4J 01
•a
Ol Ol
4-> S-
Ol
1- VI
O -O
V, °
Ol VI
en a)
s- en
ro i-
u J=
v> u
•T- VI
•o i-
•o
S-
Ol Ol
4-> Ol
V) S-
"r™ f"
r— 4->

Ol
Ol J=
•r-
V) <4_
V) O
Ol
u -c
u u
3 ro
in Ol

CD It-
CD o
.C C
4-> 0
t. 4-'
O ffl
3
VI Q

O i—
•r- . Ol
4-> Ol >
rar-T-
> CJ 4->
s- >, u
Ol O 0)
VI CL
J3 S- VI
O Ol Ol
01 1-
4J O) «
3 > in
c: c 01
•i- O 4->
E U 3
LO Ol •(—
C' E
00
n in
en •
t- c en
O *f~
t-T3

CO ^3 ro

O V) •
4J Ol V)
en 01
ra &. 4-)
* 3
c -e =
O tj 'r~
VI E
-a -i-
O) T3 CM
V) l 	
ra s-
.0 01 «
4-> V)
Ol VI 

•a

c:
o
*|"t

t
03
in

o

OJ

-M

to
OJ
1*4

i^
(O
c=
g
ZJ
V)
10
pv.
1



__
_Q
(0
[—
u










































to

-------
Table C-76.  SUMMARY OF AVERAGE OBSERVED OPACITIES  FOR  BLISTER
          DISCHARGE AT THE TAMANO SMELTER IN JAPAN3
     Set No.b
Average Opacity ,%
        1
        2
        3
        4
        5
         6
         8
        n
        10
         9
    a
     Based on same observation data used for Table C-75.
    ""Observation time for each set was 6 minutes.
    'Average of all sets is 9 percent.
                            C-104

-------
 O
     O
     03
      z:
      «=C
      a.
LU
1—
_1
LU


00


O
 o

 oo
 z:
 o
 i—i
 00
 GO
LU   LU
CD
l—i
oo
cn
Q

O
LU
«c
a.
LU
a:
oo
OJ
      LU
00
CO
o

a;
LU
LU

2:
O



o
      OO
      LU
      O3










s.
4-1
10
41
S- >.
a.
O 0
e
<— 41
cr o>
E .=
1— 4-1
a
o







I

41
4-> Ol

>a
4-1 U




4.
S- u




Ol
4J 01
S21
il
CM



41
4-J 01
g o>
S s»
VI U

4-1
O OJ
<4— 4-1
a

a
41
VI
C
ra
O Ol
o


u
Ol
VI
s
1

a oj
<4- *J
a

tj
01
VI
e
1
(0
O Ol
•!-> J:
a

u
01
VI
e
1
4^J
a 41
a

u
Ol
VI
c
•w
1-
un co •f


in o in
CM O O






o





in in
cH c-3


a o
10 CM


in to
o a




10 vn
o a
tn o m
^— CM
j
                                                                 E

                                                                IT)
                                                                          Ol

                                                                          u

                                                                          U       4-1

                                                                                  g
                                                                          41       U
                                                                         *J       L.
                                                                          &.       Ol
                                                                          01       a.

                                                                          c       a
                                                                          o
                                                                          u       fn

                                                                          01       vi
                                                                          c       '^
                                                                          o
                                                           "3

                                                           
                                                                                        c
                                                                                        o
u

ot
                                                                 01
                                                                 s
                                                                               = vi     <"
                                                                               5.2    ^
                                                                               g =     O
                                                                                        c
                                                                                        o
                                                                                 •a
                                                                                  01
                                                                         S- 01
                                                                                  VI
                                                           0)
                                                           >
                                                          u
                                                          4)
                                                          a.
                                                                         (J ra

                                                                         01 'o    ai
                                                                         •»->       e
                                                                         4J 41   —




                                                                         >       u

                                                                         vi u    01
                                                                         vi      i.
        VI    (Q <     41
        >1    >        >
                                                                   V)
                                                                               01  •
                                                                               vl Ol    ••*
                                                                              ^3 •—    VI
                                                                   -a   -^4-1

                                                                   O    =  O)   41
                                                                   u   in  c    4-1
                                                                   Ol     • —    4J
                                                                   VI   IO  St-
                                                                         41
                                                                                  3    =
                                                                                 -
                                                                                       ts
                                                                •g
                                                                •o
                                                                   S     «.S    -S
                                                                   >    4J  05    ^-

                                                                   §    2  S    .2
                                                                   U       -S    T3
                                                                         ^K
                                                                  •W     i» "O    **•
                                                                   fl     
-------
Table C-78.  SUMMARY OF VISIBLE EMISSIONS OBSERVATION  DATA--
         COPPER BLOW AT THE TAMANO SMELTER IN JAPAN3
       Set No.
Average Opacity,
          1
          2
          .3
          4
        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 JAPANa
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.
Average Opacity, %
               1
               2
       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.
2.
3.
4.
5.
6.
     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.

     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.

     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.

     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.
     Harris, D.L., Monsanto Research Corporation.  Air  Pollution
     Emission Test, Anaconda Mining Company, Anaconda,  Montana.
     Environmental Protection Agency.  EMB Report  No. 77-CUS-5.
     April 18-26, 1977.
                                                                 U.S.
     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 Phelps-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.  Larkin, R.  and J.  Steiner.  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.

13.  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-110

-------
 APPENDIX D





TEST METHODS
       D-l

-------
                           TEST METHODS




D.I  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  (SOe), some as high as 3.5 percent.



Therefore, a series of  impin-gers  containing hydrogen peroxide was



placed between the filter and the first impinger containing
                                D-2

-------
sodium hydroxide to remove the S02.  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.
                               0-3

-------
     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 HeO  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 bein-g  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 participate 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  duriag  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 of 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.'andD.L. Chase  (Battalia).  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).  Particul.ate



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 HIGH-ARSENIC PRIMARY COPPER SMELTERS
                              E-l

-------
    QUANTITATIVE EXPRESSIONS OF PUBLIC CANCER RISKS FROM THE EMISSIONS OF
         INORGANIC ARSENIC FROM HIGH-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.I.2  TheRelationshipof 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)*,
"...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 DMA.  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 one pg/m^ 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  estimate  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   They 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
  ::or all the sources in a particular category.   The  aggregate  risk  is expressed
  is 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
                 o
cubic meter  (pg/m ) and BH is the  increased probability  of  cancer associated
with each pg/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 + xi)
                       R =	= 	
                           PO   A + BH (XQ)
(2)
or
                         RP0 = A + BH (XQ + X])              '  (3)
where xn, = lifetime average exposure to the agent for the general  popu-
lation, xi = lifetime1 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
gi ves
                             BH = PO (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 Arsenic^
     As noted in the health effects assessment document^ 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 its
 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
 pg/m3  of  atmospheric arsenic was 9.4% (Pinto et al.),  17.0% (Ott  et al.),
 and  3.3% (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%.   Using this value as a best estimate  and applying
 equation 4,  one calculates the unit risk estimate of  2.95 x  1Q-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% confidence  interval  for the above unit risk
estimate may be  obtained.  Upper  and lower  95% confidence  limits can be
obtained by multiplying  the unit  risk estimate by  about  4 and 0.25  respect-
ively.  Thus, the 95% statistical  confidence  limits for  the unit risk estimate
range from 7.5 x 10~4 to 1.2 x  lO"2.,
                                    E-9

-------
E.3  QUANTITATIVE EXPRESSIONS OF PUBLIC EXPOSURE TO INORGANIC  ARSENIC
     EMITTED FROM HIGH-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 off-gases.   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 concentra-
tions in the vicinity of a source, HEM can accept the concentrations if
they are put into an appropriate format.  (Note that for  high-arsenic  copper
smelters, the Industrial Source Complex Dispersion Model  was  used to estimate
ambient air concentrations.  This dispersion model  is described in section
E.3.2 and the exposure methodology is described in section E.3.3.)
     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'3.2  Pollutant  Concentrations Near a Sotrrctg
       The dispersion  analysis uses the Industrial Source Complex Model
  Long-Term  (ISCLT).   Since more information was available for dispersion model-
  ing of  high-arsenic  primary copper smelters than for the other arsenic-emitting
  source  categories, the ISCLT model was used in place of the dispersion
  model contained in HEM.  The ISCLT model  is an EPA approved, validated
  model (validated for S02) that is more sophisticated than the HEM dispersion
 model.  The ISCLT output was put in a format acceptable to the exposure
 portion of HEM.  The ISCLT model  is  applicable in areas of flat to gently
 rolling terrain free from channeling  and  thermal  effects  associated with
 large bodies of water.   In this  analysis,  the  model  was also exercised  in
 the "rural" mode,  which allowed  for consideration of moderately stable
 atmospheric conditions.   Model options for  elevated  terrain  and building
 downwash were  also used.   Building downwash  refers to trapped air  in eddies
 on  the opposite side  of a building from the emission source.  For the low-
 level  emissions sources at this smelter, use of the  building downwash
 option significantly  improves the accuracy of the model estimates at
 receptors lying .close to  the source.  Nevertheless, the model appears to
 consistently underestimate concentrations close in.  Of the comparisons
 made,  approximately half  of the model  estimates lie within a factor of two
 of the observed concentrations.6
     This dispersion analysis used meteorological  data collected from the
 Puget Sound Air Pollution Control  Agency's  tower at North  26th and  Pearl
Streets in  Tacoma,  Washington (approximately 10 kilometers  south of the
smelter) and concurrent  surface observations  at  McChord  Air  Force Base
approximately 14.b kilometers south  of the  smelter) during the year  1972.
                                    E-ll

-------
 These data were compiled  into  a  stability wind  rose.  Ambient air
 temperatures  and mixing heights  as  a  function of stability class and
 wind  speed data, from the report Assessment of  the Air Quality Impact of
 Submissions  from the ASARCo-Tacoma Smelter, were also used in the analysis.7
      The following criteria were used to select the radial distances at which
 concentrations  were computed:
      o  The shortest  distance, 0.3  kilometers,  was based on the location of
         expected high  concentrations from ground-level sources asociated with
         the smelter.
      o  The next two  distances, 0.8 and 1.5 kilometers, were based on the
         location of the highest expected concentrations from the two
         different  sets of stack parameters for the converter source.  The
         distances  computed were ranked on the basis of the computed
         concentration times the frequency of occurrence for a particular
         combination of wind speed and stability class.
      o   The remaining distances were selected on the basis of the highest
         expected concentrations from the main stack acting alone.   Terrain
         height  and wind direction were the primary criteria in making this
         determination.  A profile of terrain points was established
         consisting  of the highest elevations at various distances  from the
         plant, primarily in the northeastern and southwestern  quadrants,  the
        directions of predominant wind flow.  The concentrations  computed by
        the ISC model using the actual  meteorological  data were then  ranked
        to select the terrain points and thus  the corresponding distances
        from the smelter.
     All terrain points and receptor elevations  are taken  from U.S.  Geological
Survey maps, at a scale of 1:24,000.  Modeling  considerations, however,
                                    E-12

-------
  disallow  receptor elevations above the elevation of any source.  As a
  result, the sources are broken into three groups (main stack and two
  ground sources) that make use of two receptor elevation data sets.  For
  the main  stack the elevations are limited to approximately  /O to 216
  meters (ZUO to  710 feet) main stack level (MSL).  The groundlevel  sources
  designated 6(1) and 6(2) in Figure E-l are modeled assuming flat terrain.
 The total  estimated concentrations from the smelters  are then computed  as
 the sum of these two individual  source contributions.
      The main  stack was treated  as a  point source.  Due to its  relatively
 low exit  velocity,  a stack-tip downwash  adjustment  to  plume height  is made,
 as necessary.   The  roof opening,  Source  6(2),  is  treated  as  three  separate
 emissions  points  subject to  building  downwash.  The building  width  is
 designated as  the diameter of a circular  structure  of  the  same  area as the
 actual  building.  Source 6(1), representing numerous ground  sources, is
 composed of two emission areas, 93 meters  on a side, and a third emission
 area,  185  meters on  a side, which together roughly  correspond to the triangle
 depicted in Figure E-l.
 E.3.3  The Peopl_e Living Near A Source
     To estimate the number and distribution of people residing within 20
 kilometers of a plant, the model  contains a slightly modified version  of
 the "Master Enumeration District  List—Extended"  (MEU-X) data base.
The data base is broken down 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
                                   E-l 3

-------
  Ground (1)
  Modeling
Approximations   I
                                                                        Ground (11
                                                              Scale: l" : 200*
                  Figure E-l.   Tacoma Plant  Configuration
                                    .  E-l 4

-------
 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
                                                                        I
 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.3.4  Exposure^
      The Human Exposure Model  (HEM)  uses  the estimated ground  level concen-
 trations of a pollutant,  in this  case from ISCLT, together with population
 data to calculate public exposure.   For  each  of 144  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/BG1s;  at large radii, the grid  cells  are usually  larger
 than ED/BG's.  The area surrounding the source is  divided  into two  regions,
 and each ED/BG is classified by the region in which its  centroid lies.
 Population exposure  is calculated differently for  the ED/BG's  located
within each region.  For ED/BG centroids  located  between 0.3 km and  2.8 km
 from the emission source, populations are divided  between  neighboring
                                    E-15

-------
concentration grid points.  There are 64 (4 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 18.8 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 4.4,  8.5,
11.5, 15.0, and 18.8 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 64  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
through the use of log-log and linear interpolation.  (For a  more  detailed
discussion of the model used to estimate exposure, see Reference 6 and 7.)
                                     E-16

-------
E'3-5  Public Exposure to Inorganic Arsenic Emissions from High  Arsenic  Primary
       Copper Smelters
E.3.5.1  Source Data
     One smelter is included in the analysis.   Table E.I  lists the  name  and
address of the plant considered, and Table E.2 lists the  plant data used  as
input to the Human Exposure Model  (HEM).
E.3.5.2  Exposure Data
Table E.3 lists, 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 plant, the  numbers
of people exposed to various  ambient concentrations,  as calculated  by HEM.
                                   E-17

-------
                         TABLE E-l

   IDENTIFICATION OF HIGH-ARSENIC PRIMARY COPPER SMELTERS
Plant Number Code
Plant Name and Address
                                          ASARCO, Inc.
                                          TACOMA, WA '
                            E-l 8

-------
      Table E-2.   Input  Data  to  Dispersion Model for ASARCO-Tacoma Smelter3
                               (Baseline Control)
Emission
Source
Stack
Ground (1)
Ground (2)
Height
(m)
61.0
30.5
Ground
Di ameter
(m)
7.3
b
c
Temperature
(OK)
366
310
294
Velocity
(m/$ec)
9.7
0.5
d
Emissions
(kg/hr)
1 7.3
14.0
1.3
a)  Smelter location:  Latitude  47° 17'  49"

                       Longitude  122° 30'  23"

b)  Building roof monitor (see Figure E-l)

c)  Three area sources (see Figure E-l)

d)  Discharge velocity assumed to be negligible.
                                    E-l 9

-------
          TABLE  E-3  TOTAL  EXPOSURE AND NUMBER OF PEOPLE EXPOSED
                       (HIGH-ARSENIC PRIMARY COPPER SMELTERS)*
     Plant
     Total
   Number of
People Exposed
     Total
    Exposure
(People - yg/m3)
                           368,000
                                  103,000
* An 18.8-kilometer radius was used for the analysis  of  high-arsenic
  primary copper smelters.
                                    E-ZU

-------
                                Table E-4

               PUBLIC  EXPOSURE FUR HIGH ARSENIC CUPPER SMELTERS
                   AS  PRUDUCEU BY THE HUMAN EXPUSURE MODEL
Concentration
Level (yg/m3)
30.6

25.0
10.0
21—
.5
1r~\
.0
Of-
.5
0.25
0.10
0.05
0.025
0.01
0.005
U.UU25
0.001
O.OOOb
U.OOU1
o
1.0 x 10-9
Population
Exposed
(Persons)*


9
223
1,316
4,455
16,997
36,051
84,005
253, 783
352,848
368,409
368,409
368,409
368,409
368,409
368,409
368,409
Exposure
(Persons - ug/m3)**

290
. 290
3,240
10,200
20,000
38,800
52,000
69,300
3
95,100
103,071
104,520
104,680
104, 780
104,820
104,85U
1U4,85U
104,850
            *5e c°Tted  value> round«l t° the nearest whole number,  of the
i    n          Pe°Ple  ^Posed to the matching and higher concentration  levels

be ro^ded  to  I"  ^^ ^ ^"^ W°U'd "e r°Und'd t0 ° "d  »•"  People
                                                exposure to
                                  E-21

-------
E.4  QUANTITATIVE EXPRESSIONS OF PUBLIC CANCER RISKS FROM INORGANIC  ARSENIC
     EMITTED FROM HIGH-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 is 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 =
                                          (piCj)
                                     E-22

-------
 where
      I  = summation over all  grid  points where  exposure  is  calculated
      P-j  = population associated  with  grid  point  i,
      Ci  = 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 lO"3 at 1 pg/m3, and the assumed
 exposures are:
            ambient air
          concentrations
          2   pg/m3
          1   pg/m3
          0.5 pg/m3
                number of people exposed
                 to given concentration
                        1,000
                       10,000
                      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
    2   n9/m3         x      3  x 10"3(pg/m3)-1     =         6 x TO"3
1   pg/m3
0.5 yg/m3
x
x
3 x 10"3
3 x 10"3
                                                            3 x 10-3
                                                           .5 x 10-3
                                    E-23

-------
The 70 year cancer incidence among the people exposed  to these  concentrations
is given by:
    probability of cancer
    at each exposure level
                                   number of  people  at
                                   each  exposure  level
6 x ID'3
3 x lO-3
1.5 x TO-3
x
x
X
1,000
10,000
100,000
cancer incidences
  after 70 years
    of exposure
         6
        30
       150
                                                           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 ug/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
     which people are exposed
                                             1  ug/m3
E.4.2  Risks Calculated for Emissions  of  Inorganic  Arsenic from High-Arsenic
       Primary Copper Smelters
     The explained methodologies for calculating  maximum lifetime  risk and
cancer incidences were applied to the  high-arsenic  primary copper  smelter,
assuming a baseline level  of emissions.   A baseline level of  emissions means
                                     E-24

-------
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-25

-------
   CO
   ca
   LU
LU LU
O ^P*
Z 00

So,
   LU
O O.
•Z.d.'-'
< O f>
   C_5 f-~
^     p
oo >  s-
>-< o; -P
o; rf c
   S o
LU M O
S 0£
I— « O^ CO
1—     E
LU O •f
Lu i-< r—
H-( =: cu
_J LU tO
   oo 
O
oo
oo
i — i

'
Q;
o
|_t

GO
1
£
o



















































oo
cu
o
c
cu

o

1 — 1
cu
re
o














pp«_> _ -.

00
i-.
fO
cu


1 — 1
x
1 — 1

E
cu
re
o

cu
E
0



—__ _



£_
ro
cu
^
s_
cu
CL








j._-.jj _



00
oi

re
3
•o
•r-
>
•p-
-o
l — i
-P
00
CU
JE
CD
•r—
3=





•O •— -
CU E
00 ^i
S- O
CU CLCO
J3 X •
E UJ CO
3 t-I
:s cu

r- 'o.'i-
(C O J=
P CU -P
O Q- •!-
1— 3

o "~"


s_ oo
CU P
f^ <4_ r^
E O ja
C J=
o
E 00
O re
o
"TJ
^- CU
O 00
3
•P <
E Q.
•r- LU
i- "5
E "§
3
00
CU CU
•P "O
3
P -P
O 00
cu
r~ r™1
^^ re
CU E
S- 0
00 P
P re
O CL
^0 Z3
Q-O
E o
•i- O
JE CU
•p cu
i— j-
CU -P
JE
cu
O -E
•r- 4->
r~"
t~) o
CL
H- E
O rO
>CVJ
00 CU •
E r- CU
0 CU -P
•r™* t- rt3
P E
re oo M-
0 Q.P
•P- >r- 00
4- j= CU
•p- 00
P E **s
E O 00
re -p- -p-
3 -P S-
O" re
p- -p
CU CU -P-
00 S- E
CU 3
.E CU
•P oo M-
E O
E O
•p- O.P
00 E
oo cu cu
CU i- E
01 i CL
c cu o
re oo i —
i- o cu
•o >
cu cu
.E -P -a
1— E
cu cu
                                                     E-26
                                                                                        QJ
                                                                                     LU M-
                                                                                     1— 4-  S-
                                                                                     O -r-  O
                                                                                     z -o <*-

-------
 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 Final  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 is 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 concentrations  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-27

-------
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
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 has 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 is not known and that the relative
locations used in  the exposure model have changed since the 1970 census.
                                      E-28

-------
     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 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.
     3.  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-29

-------
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-II-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," April  1978,
    Docket Number (OAQPS 79-8)  II-A-5.

6.  U.S. Environmental Protection Agency.  An Evaluation Study of  the
    Industrial Source Complex (ISC) Dispersion Model.   U.S.  Environmental
    Protection Agency.  Research  Triangle Park, North Carolina.
    EPA-450/4-81-002.  1981

7.  U.S. Environmental Protection Agency.  Assessment of the Air Quality
    Impact of S02 Emissions from the ASARCo-Tacoma Smelter.  U.S.
    Environmental Protection Agency.   Research Triangle Park, North  Carolina.
    EPA-910/9-76-028.  1976

8.  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-30

-------
                                   TECHNICAL REPORT DATA
                            (Please read Instructions on the reverse before completing)
1. REPORT NO.
  EPA-450/3-83-009a
                              2.
                                                            3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
  Inorganic Arsenic Emissions  from High-Arsenic
  Primary Copper Smelters  -  Background Information
    for Proposed Standards	
            5. REPORT DATE

                     July 1983
            6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
                                                            8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
  Office of Air Quality  Planning and Standards
  U.S.  Environmental Protection  Agency
  Research Triangle Park,  North  Carolina 27711
                                                            10. PROGRAM ELEMENT NO.
             11. CONTRACT/GRANT NO.
                    68-02-3060
12. SPONSORING AGENCY NAME AND ADDRESS
  DAA for Air Quality  Planning and Standards
  Office of Air, Noise,  and  Radiation
  U.S.  Environmental Protection Agency
  Research Triangle Park,  North Carolina 27711
             13. TYPE OF REPORT AND PERIOD COVERED
             	Final	
             14. SPONSORING AGENCY CODE
                     EPA/200/04
15. SUPPLEMENTARY NOTES
16. ABSTRACT


       Standards of performance to control emissions of inorganic  arsenic from new
  and  existing primary  copper smelters processing feed materials containing an annual
  average of 0.7 percent  or greater arsenic  are  being proposed under Section 112 of
  the  Clean Air Act.  This  document provides information on the background and
  authority, regulatory alternatives considered,  and environmental  and economic impacts
  of the regulatory alternatives.
17.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                              b.lOENTIFlERS/OPEN ENDED TERMS  0. COSATI Field/Group
  Air pollution
  Hazardous air pollutant
  Pollution control
  Standards of performance
  Inorganic arsenic
  Primary copper smelters
Air pollution  control
Stationary  sources
13 B
18. DISTRIBUTION STATEMEN1
  Unlimited
                                               19. SECURITY CLASS (This Report)
                                                Unclassified
                                                                          21. NO. OF PAGES
                               372
                                               20. SECURITY CLASS (This page)
                                                Unclassified
                                                                          22. PRICE
EPA Form 2220-1 (Rev. 4-77)
                      PREVIOUS EDITION IS OBSOLETE

-------

-------