United States     Office of Air Quality      EPA-450 3-83-01 Oa
          Environmental Protection  Planning and Standards     April 1983
          Agency       Research Triangle Park NC 27711
          __ ___                      —  —
v>EPA     Inorganic Arsenic    Draft
          Emissions from      EIS
          Low-Arsenic
          Primary
          Copper Smelters -
          Background
          Information for
          Proposed Standards

-------
                                 EPA-450/3-83-010a
    Inorganic Arsenic Emissions from
Low-Arsenic Primary Copper Smelters -
         Background  Information
         for Proposed Standards
            Emission Standards and Engineering Division
           U S ENVIRONMENTAL PROTECTION AGENCY
              Office of Air, Noise, and Radiation
           Office of Air Quality Planning and Standards
           Research Triangle Park, North Carolina 27711

                   April 1983

-------
This report has been reviewed by the Emission Standards and Engineering Division of the Office of Air Quality Planning
and Standards, EPA, and approved for publication Mention of trade names or commercial products is not intended to
™nstitLi;e endorsement °r recommendatlon f°r use. Copies of this report are available through the Library Serv.ces
urnce (MD-35), U S Environmental Protection Agency, Research Triangle Park, North Carolina 27711; or, for a fee from
the National Technical Information Services, 5285 Port Royal Road, Springfield, Virginia 22161
                                  Publication No. EPA-450/3-83-010a

-------
                        ENVIRONMENTAL  PROTECTION AGENCY

                      Background  Information and Draft
                       Environmental Impact Statement
          F   H  M  „  for1Pr/Imary Copper Smelters Processing
          Feed  Materials Containing Less Than 0.7 Percent Arsenic
               As^~)
         fl   ///    /       Prepared by:

                "Txr^MAA-1
 x—  ...  .-armer~
  Director, Emission Standards and Engineering Division
  U. S. Environmental  Protection Agency
  Research Triangle Park, NC  27711
  1.
 2.


 3.  The comment period for review of this document  is  60  days
    period"6          ^ be contacted regarding  the date of the comment

 4.  For additional information contact:

    Mr. Gene W. Smith
    Standards Development Branch  (MD-13)
    U. S.  Environmental  Protection Agency
    Research Triangle Park,  NC  27711
    telephone:   (919) 541-5624.

5.  Copies of this document  may be obtained from:

    U.  S.  EPA Library (MD-35)
    Research  Triangle Park, NC  27711

    National  Technical Information Service
    5285 Port Royal Road
    Springfield, VA   22161

-------
                            TABLE OF CONTENTS

Section                                                                page
1.0  SUMMARY	    1_1
     1.1  Statutory Authority 	    1-1
     1.2  Regulatory Alternatives 	    1-1
     1.3  Environmental Impacts 	    1-2
     1.4  Economic Impacts  	    1-3
2.0  THE PRIMARY COPPER INDUSTRY	    2-1
     2.1  General	    2-1
          2.1.1  Raw Materials	    2-1
          2.1.2  Process Description	    2-5
     2.2  Arsenic Behavior and Distribution in Copper Smelters.  .  .    2-19
          2.2.1  Arsenic Behavior During Roasting  	    2-19
          2.2.2  Arsenic Behavior in Smelting Furnaces   	    2-22
          2.2.3  Arsenic Behavior During Converting  	    2-27
          2.2.4  Arsenic Balance	    2-30
     2.3  Arsenic Distribution and Emissions at Domestic Low-
          Arsenic Throughput Copper Smelters  	    2-33
          2.3.1  Process Arsenic Emissions  	    2-34
          2.3.2  Fugitive Arsenic Emissions 	    2-36
     2.4  References	    2-60
3.0  CONTROL TECHNOLOGY 	    3_1
     3.1  Alternative Control  Techniques	    3-1
          3.1.1  Process Emission Controls	    3-1
          3.1.2  Fugitive Emission Sources and Controls  	    3-17
     3.2  Performance Capabilities of Alternative
          Control  Techniques for Arsenic and Total
          Particulate Emissions  	    3.49
          3.2.1  Process Control  Systems	    3-49
          3.2.2  Fugitive Control  Systems  Evaluation	    3-64
          3.2.3  Conclusions	    3_8i
     3.3  References	    3.35

-------
                            TABLE OF CONTENTS
                               (continued)
Section
                                                                      Page
4.0  MODEL PLANTS, REGULATORY BASELINE, AND REGULATORY
     ALTERNATIVES 	   4.4
     4.1  Regulatory Considerations 	   4-1
          4.1.1  Clean Air Act	   4.3
          4.1.2  Arsenic Regulation by the Occupational
                 Safety and Health Administration 	   4-7
          4.1.3  Clean Water Act	   4_g
          4.1.4  Resource Conservation and Recovery Act
                 (RCRA)	   4.9
     4.2  Baseline and Regulatory Alternatives  	   4-10
          4.2.1  Definition of Baseline	   4-10
          4.2.2  Description of the Regulatory Alternatives ....   4-10
     4.3  Baseline Configuration, Baseline Arsenic Emissions,
          and Regulatory Alternatives for Model Plants	   4-12
          4.3.1  Baseline Arsenic Emissions 	   4-12
          4.3.2  Model Plant Baseline Configurations and
                 Regulatory Alternatives	   4-16
     4.4  References	   4.39
5.0  ENVIRONMENTAL IMPACTS	   5_1
     5.1  Introduction	   5_1
     5.2  Air Pollution Impacts of Regulatory Alternatives	   5-1
          5.2.1  Baseline Emissions 	   5-1
          5.2.2  Arsenic Emission Reductions Under the Regulatory
                 Alternatives 	   5_2
     5.3  Energy Impacts of the Regulatory Alternatives 	   5-5
     5.4  Solid Waste Impacts of the Regulatory Alternatives. ...   5-6
     5.5  Water Pollution Impacts of the Regulatory Alternatives.  .   5-8
     5.6  References	   5-10
6.0  COSTS	   6_!

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

  Section
  	                                                                Page
       6.1  Baseline Controls ....                                     c  ,
                                      	    D-j
            6.1.1  Baseline Costs .  .                                     c  „
                                    	    b-4
       6.2  Process  Controls.  ...                                       c  ,
                                  	    b-4
            6.2.1  Process  Control  Costs	        5,4
       6.3  Fugitive Controls  	        6_1Q
            6.3.1  Converter Controls  	        6_10
            6.3.2 Matte  and Slag Tapping  Controls	    6_18
       6.4  Costs of Regulatory  Alternatives	    6-23
       6.5   Cost  Effectiveness	                                  c 00
                                    	   b-
-------
                            TABLE OF CONTENTS
                               (concluded)

Section                                                               Page
APPENDIX C - SUMMARY OF TEST DATA	   C-l
     C.I  ASARCO-Tacoma	   C-2
     C.2  ASARCO-E1 Paso	   C-6
     C.3  Anaconda	   C-10
     C.4  Phelps Dodge-Ajo	   C-12
     C.5  Phelps Dodge-Hidalgo	   C-15
     C.6  Phelps Dodge-Douglas	   C-17
     C.7  Kennecott-Magna, Utah	   C-18
     C.8  Kennecott-Hayden	   C-20
     C.9  Tamano Smelter  (Hibi Kyodo Smelting Co.,) Japan  	   C-25
     C.10  Test Data (Tables)	   C-26
     C.ll  References	   C-109
APPENDIX D - TEST  METHODS	   D-l
APPENDIX E - QUANTITATIVE EXPRESSIONS OF  PUBLIC CANCER  RISKS  FROM
             EMISSIONS OF INORGANIC ARSENIC  FROM LOW-ARSENIC
             PRIMARY COPPER SMELTERS   	   E-l
APPENDIX F - ARSENIC DISTRIBUTION AT U.S.  COPPER SMELTERS  	   F-l
                                    vm

-------
                            LIST OF TABLES

                                                                     Page
       Assessment of Environmental And Economic  Impacts for  Each
       Regulatory Alternative Considered 	    1-4
2-1    Domestic Primary Copper Smelters	    2-2
2-2    Major Copper-bearing Minerals 	    2-3
2-3    Arsenic Input in the Feed to Domestic Copper Smelters  .  .  .    2-4
2-4    Arsenic Elimination in Roasters 	    2-21
2-5    Elimination of Arsenic in Reverberatory Furnaces   	    2-24
2-6    Elimination of Arsenic in Electric Furnaces  	    2-26
2-7    Elimination of Arsenic in Flash Furnaces  	    2-28
2-8    Elimination of Arsenic During Converting  	    2-29
2-9    Arsenic Balance Weight Percent as Reporting in Smelter
       Products	    2-31
2-10   Measured Arsenic Collection Efficiencies
       of Control  Devices	    2-35
2-11   Summary of Process Arsenic Emission Estimates in Absence
       of Control  for Low-Arsenic Throughput Primary Copper
       Smelters	    2-37
2-12   Potential  Sources of Fugitive Arsenic Emissions  	    2-38
2-13   Fugitive Arsenic Emissions During Calcine Transfer
       From Multihearth Roasters 	    2-41
2-14   Matte Tapping Emissions from Copper Smelters	    2-47
2-15   Slag Tapping Fugitive Arsenic Emissions from
       ASARCO-Tacoma 	    2-50
2-16   Reverberatory Furnace Slag Analysis for Arsenic
       Content at ASARCO-Tacoma	    2-56
2-17   Summary of Potential Fugitive Arsenic Emission Estimates in
       Absence of Control  for Low-Arsenic Throughput Primary Copper
       Smelters	    2-58
3-1    Summary of AS.Og Vapor Pressure Data and Corresponding
       Arsenic Concentration at Various Temperatures 	    3-2
3-2    Arsenic Data for Hot ESP	    3-6
                                   IX

-------
                             LIST OF TABLES

                               (continued)
Table
 3-3    Estimated Approximate Maximum Impurity Limits For
        Metallurgical  Offgases Used to Manufacture Sulfuric

3-4

3-5

3-6

3-7

3-8

3-9

3-10

3-11

3-12

3-13

3-14

3-15

3-16

3-17

3-18

3-19

Acid 	
Summary of Design Data for the ASARCO-Tacoma Converter
Secondary Hooding/Air Curtain System 	
Arsenic Performance Data for the Roaster Baghouse at
ASARCO-Tacoma 	
Arsenic Performance Data for the Arsenic Plant Baghouse
at ASARCO-Tacoma 	
Arsenic Performance Data for Spray Chamber/Baghouse at
the Anaconda-Anaconda Smelter 	
Particulate Performance Data for Spray Chamber/Baghouse at
the Anaconda-Anaconda Smelter 	
Arsenic Emissions at Outlet of Reverberatory Furnace
Electrostatic Precipitator at ASARCO-Tacoma 	
Arsenic Performance Data for Spray Chamber/Electrostatic
Precipitator at ASARCO-E1 Paso 	
Particulate Performance Data for the Spray Chamber/
Electrostatic Precipitator at ASARCO-E1 Paso 	
Particulate Performance Data for the Spray Chamber/
Electrostatic Precipitator Outlet at ASARCO-E1 Paso ....
Arsenic Performance Data for Venturi Scrubber
At Kennecott-Hayden 	
Arsenic Performance Data for Double-Contact
Acid Plant at ASARCO-E1 Paso 	
Arsenic Performance Data for Single-Contact
Acid Plant at Phelps Dodge-Ajo 	
Summary of Visible Emission Observation Data for Capture
Systems on Fugitive Emission Sources at ASARCO-Tacoma . . .
Air Curtain Capture Efficiencies at ASARCO-Tacoma
Using Gas Tracer Method - January 14, 1983 	
Air Curtain Capture Efficiencies at ASARCO-Tacoma
Using Gas Tracer Method - January 17-19, 1983 	
Air Curtain Capture Efficiencies at ASARCO-Tacoma for
Special Gas Tracer Injection Points - January 18-20, 1983 .
3-13

3-42

3-50

3-52

3-53

3-53

3-55

3-57

3-58

3-59

3-61

3-63

3-64

3-66

3-70

3-71

3-72

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

Table                                                                paqe
 3-20   Visible Emissions Observation Data for Converter
        Secondary Hood System During Matte Charging
        At the Tamano Smelter	   3.75
 3-21   Visible Emissions Observation Data for Blister
        Discharge at the Tamano Smelter 	   3-78
 3-22   Arsenic Data for Converter Building Baghouse at
        ASARCO-E1  Paso	   3_80
 3-23   Particulate Data for Converter Building Baghouse
        at ASARCO-E1  Paso	   3-80
 4-1    State Implementation Plans (SIP's) for Sulfur Dioxide
        Affecting Copper Smelters and Compliance Status 	   4-4
 4-2    State Implementation Plans (SIP's) for Total Suspended
        Particulates  Affecting Copper Smelters and
        Compliance Status	   4_6
 4-3    State Implementation Plans (SIP's) for Lead Affecting
        Copper Smelters  and Compliance Status  	   4-8
 4-4    Summary of Baseline Process  Arsenic Emission
        Estimates  for Low-Arsenic Throughput Primary
        Copper Smelters  	   4_13
 4-5    Summary of Baseline Fugitive Arsenic Emission
        Estimates  for Low-Arsenic Throughput Primary
        Copper Smelters  	   4_14
 5-1    Arsenic Emissions from Low-Arsenic Throughput
        Copper Smelters  by Emission  Source and
        Regulatory Alternative  	   5.3
 5-2    Nationwide Annual  Arsenic Emissions by Regulatory
        Alternative and  Emission  Reductions
        From  Baseline	   5_4
 5-3    Nationwide Annual  Energy  Requirements  by
        Regulatory Alternative  	   5_7
 5-4    Nationwide Annual  Solid Wastes  Generated  by
        Regulatory Alternative  	   5_g
 6-1     Regulatory Alternative Control  Requirements  	   6-2
 6-2     Equipment  Considered  in Baseline Cost  Analysis   	   6-5

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


Table                                                                paqe
                                                                       -^
 6-3    Estimated Capital  and Annualized Costs of Baseline
        Controls for Primary Copper Smelters  	   6-6
 6-4    Design Parameters  for Add-on Process Particulate Matter
        Control  for Converters at Kennecott-McGill  	   6-8
 6-5    Estimated Annualized Cost of Add-on Process
        Particulate Matter Control  System for Converters
        at Kennecott-McGill  	   6-11

 6-6    Design Parameters  for Air Curtain Secondary Hood
        Capture  System for Primary  Copper Smelters  	   6-13
 6-7    Estimated Capital  Costs of  Air Curtain Secondary Hood
        Capture  Systems for Primary Copper Smelters 	   6-15
 6-8    Estimated Capital  Costs of  Air Curtain Secondary Hoods
        and Fabric Filters for Primary Copper Smelters  	   6-17
 6-9    Estimated Annualized Costs  of Air Curtain Secondary
        Hoods and Fabric  Filters for Primary Copper Smelters  .  .  .   6-19
 6-10   Estimated Capital  Costs of  Add-on Fugitive Emission
        Capture  and Collection Systems for Matte and Slag Tapping
        Operations	   6-21

 6-11   Estimated Annualized Costs  of Add-on Fugitive Emission
        Capture  and Collection Systems for Matte and Slag Tapping
        Operations	   6-22

 6-12   Summary  of Incremental Costs of Regulatory Alternatives
        Over Baseline for  Control  of Arsenic Emissions for
        Low-Arsenic Throughput Primary Copper Smelters  	   6-24
 b-13   Emission Reduction and Cost Effectiveness Impacts
        for Low-Arsenic Throughput  Primary Copper Smelters  ....   6-26
 7-1    Smelter  Ownership, Production and Source Material
        Arrangements  	   7_3

 7-2    United States and  World Comparative Trends in Copper
        Production:  1963-1981  	   7-7

 7-3    U.S. Copper Consumption	   7-9
 7-4    U.S. Copper Demand By Market End Uses	   7-10
 7-5    Average  Annual  Copper Prices  	   7-14
 7-6    Increase In Cost  of Producing Copper Due to Arsenic
        Controls for Low-Arsenic Primary Copper Smelters  	   7-25
                                   xn

-------
                            LIST OF TABLES

                               (continued)


                                                                     Page
       Maximum Percent Price Increase for Arsenic Controls
       for Low-Arsenic Primary Copper Smetlers  	    7-26

7-8    Business Segment Return on Sales for Copper Companies  .  .  .    7-30

7-9    Macimum Percent Profit Decrease for Arsenic
       Controls for Low-Arsenic Primary Copper  Smelters	    7-31

7-10   Review of Smelters	    7-33

7-11   Capital Costs of Arsenic Controls for Primary
       Copper Smelters 	    7-36

7-12   Number of Employees at Companies that Own Primary
       Copper Smelters 	    7-39

C-l    Summary of Emission Tests 	    C-27

C-2    Index to Arsenic and Particulate Test Data Tables by Process
       Facility and Sample Type	    C-29

C-3    Summary of Arsenic Test Data -- Roaster  Baghouse Inlet,
       ASARCO-Tacoma Smelter 	    C-31

C-4    Summary of Arsenic Test Data — Roaster  Baghouse Outlet,
       ASARCO-Tacoma Smelter 	    C-32

C-5    Summary of Arsenic Test Data — Arsenic  Kitchen Baghouse
       Inlet, ASARCO-Tacoma Smelter	    C-33

C-6    Summary of Arsenic Test Data -- Metallic Arsenic Baghouse
       Inlet, ASARCO-Tacoma Smelter	    C-34

C-7    Summary of Arsenic Test Data -- Arsenic  Baghouse Outlet
       (Metallic and Kitchen),  ASARCO-Tacoma Smelter 	    C-35

C-8    Summary of Arsenic Test Data — Reverb ESP Outlet,
       ASARCO-Tacoma Smelter 	    C-36

C-9    Summary of Arsenic Test Data — Calcine  Discharge,
       ASARCO-Tacoma Smelter 	    C-37

C-10   Summary of Arsenic Test Data — Matte Tapping,
       ASARCO-Tacoma Smelter 	    C-38

C-ll   Summary of Arsenic Test Data ~ Slag Tapping,
       ASARCO-Tacoma Smelter 	    C-39

C-12   Summary of Arsenic Test Data — Converter Slag Return,
       ASARCO-Tacoma Smelter 	    C-40

C-13   Summary of Arsenic Test  Data — R & R ESP Inlet (Roaster),
       ASARCO-E1  Paso Smelter	    C-41
                                  XI 11

-------
                     LIST OF  TABLES
                       (continued)
Table
C-14

C-15

C-16

C-17

C-18

C-19

C-20

C-21

C-22

C-23

C-24

C-25

C-26

C-27

C-28

C-29

C-30

C-31
_F
Summary of Arsenic Test Data — R & R ESP Inlet (Reverb-
North), ASARCO-E1 Paso Smelter 	
Summary of Arsenic Test Data — R & R ESP Inlet (Reverb-
South), ASARCO-E1 Paso Smelter 	
Summary of Arsenic Test Data — R & R ESP Inlet (Total),
ASARCO-E1 Paso Smelter 	
Summary of Arsenic Test Data — R & R ESP Outlet,
ASARCO-E1 Paso Smelter 	
Summary of Particulate Test Data ~ R & R ESP Inlet
(Roaster), ASARCO-E1 Paso Smelter 	
Summary of Particulate Test Data -- R & R ESP Inlet
(Reverb-North), ASARCO-E1 Paso Smelter 	
Summary of Particulate Test Data -- R & R ESP Inlet
(Reverb-South), ASARCO-E1 Paso Smelter 	
Summary of Particulate Test Data — R & R ESP Inlet (Total),
ASARCO-E1 Paso Smelter 	
Summary of Particulate Test Data — R & R ESP Outlet,
ASARCO-E1 Paso Smelter 	
Summary of Arsenic Test Data -- DC Acid Plant Inlet,
ASARCO-E1 Paso Smelter 	
Summary of Arsenic Test Data -- DC Acid Plant Outlet,
ASARCO-E1 Paso Smelter 	
Summary of Arsenic Test Data — Converter Building Baghouse
Inlet, ASARCO-E1 Paso Smelter 	
Summary of Arsenic Test Data — Converter Building Baghouse
Outlet, ASARCO-E1 Paso Smelter 	
Summary of Particulate Test Data — Converter Building
Baghouse Inlet, ASARCO-E1 Paso Smelter 	
Summary of Particulate Test Data -- Converter Building
Baghouse Outlet, ASARCO-E1 Paso Smelter 	
Summary of Particulate Test Data ~ Roaster/Reverberatory ESP
Outlet, ASARCO-E1 Paso Smelter 	
Summary of Particulate Test Data -- Calcine Discharge Duct,
ASARCO-E1 Paso Smelter 	
Summary of Particulate Test Data ~ Calcine Discharge Duct,
'age

C-42

C-43

C-44

C-45

C-46

C-47

C-48

C-49

C-50

C-51

C-52

C-53

C-54

C-55

C-56

C-57

C-58

ASARCO-E1  Paso Smelter 	  C-59
                            xiv

-------
LIST OF TABLES
  (continued)
Table
C-32

C-33

C-34

C-35

C-36

C-37

C-38

C-39

C-40

C-41

C-42

C-43

C-44

C-45

C-46

C-47

C-48

C-49


Summary of Arsenic Test Data — Matte Tapping Duct,
ASARCO-E1 Paso Smelter 	
Summary of Parti oil ate Test Data — Matte Tapping Duct,
ASARCO-E1 Paso Smelter 	
Summary of Arsenic Test Data -- Spray Chamber/Baghouse
Inlet-West, Anaconda-Anaconda Smelter 	
Summary of Arsenic Test Data — Spray Chamber/Baghouse
Inlet-East, Anaconda-Anaconda Smelter 	
Summary of Arsenic Test Data — Spray Chamber/Baghouse
Inlet (Total), Anaconda-Anaconda Smelter 	
Summary of Arsenic Test Data -- Spray Chamber/Baghouse
Outlet, Anaconda-Anaconda Smelter 	
Summary of Particulate Test Data — Spray Chamber/Baghouse
Inlet-West, Anaconda-Anaconda Smelter 	
Summary of Particulate Test Data -- Spray Chamber/Baghouse
Inlet-East, Anaconda-Anaconda Smelter 	
Summary of Particulate Test Data -- Spray Chamber/Baghouse
Inlet (Total), Anaconda-Anaconda Smelter 	
Summary of Particulate Test Data -- Spray Chamber/Baghouse
Outlet, Anaconda-Anaconda Smelter 	
Summary of Arsenic Test Data -- Reverberatory ESP Inlet,
Phelps Dodge-Ajo Smelter 	
Summary of Arsenic Test Data -- Reverberatory ESP Outlet,
Phelps Dodge-Ajo Smelter 	
Summary of Arsenic Test Data -- Converter ESP Inlet No. 1,
Phelps Dodge-Ajo Smelter 	
Summary of Arsenic Test Data -- Converter ESP Inlet No. 2,
Phelps Dodge-Ajo Smelter 	
Summary of Arsenic Test Data — Converter ESP Outlet (Acid
Plant Inlet), Phelps Dodge-Ajo Smelter 	
Summary of Arsenic Test Data ~ Acid Plant Outlet,
Phelps Dodge-Ajo Smelter 	
Summary of Arsenic Test Data -- Matte Tapping Hood Outlet,
Phelps Dodge-Ajo Smelter 	
Summary of Particulate Test Data -- Matte Tapping Outlet,
Phelps Dodge-Ajo Smelter 	
Page

C-60

C-61

C-62

C-63

C-64

C-65

C-66

C-67

C-68

C-69

C-70

C-71

C-72

C-73

C-74

C-75

C-76

C-77
       XV

-------
                             LIST  OF  TABLES

                               (continued)
                                                                      Page
        Summary  of  Arsenic  Test  Data  —  Converter Secondary  Hood
        Outlet,  Phelps  Dodge-Ajo Smelter 	   C-78

 C-51    Summary  of  Particulate Test Data --  Converter  Secondary Hood
        Outlet,  Phelps  Dodge-Ajo Smelter	   C-79

 C-52    Summary  of  Arsenic  Test  Data  --  Converter Secondary  Hood
        Outlet,  Phelps  Dodge-Hidalgo  Smelter 	   c-80

 C-53    Summary  of  Arsenic  Test  Data  —  Calcine/Roaster  Fugitives
        Baghouse  Inlet, Phelps Dodge-Douglas  Smelter 	   C-81

 C-54    Summary  of  Arsenic  Test  Data  —  Calcine/Roaster  Fugitives
        Baghouse  Outlet, Phelps  Dodge-Douglas Smelter	C-82

 C-55    Summary  of  Particulate Test Data  —  Calcine/Roaster  Fugitives
        Baghouse  Inlet, Phelps Dodge-Douglas  Smelter 	   C-83

 C-56    Summary  of  Particulate Test Data  —  Calcine/Roaster  Fugitives
        Baghouse  Outlet, Phelps  Dodge-Douglas Smelter	C-84

 C-57    Summary  of Arsenic  Test  Data  —  Concentrate Dryer Scrubber
        Outlet,  Kennecott-Magna  Smelter	c-85
 C-58    Summary of Arsenic  Test  Data  -- Acid  Plant Inlet
        Kennecott-Magna Smelter	c_86

 C-59    Summary of Arsenic  Test  Data  — Matte Tapping Duct,
        Kennecott-Magna Smelter	c_87

 C-60    Summary of Arsenic  Test  Data  — Slag Tapping Duct,
        Kennecott-Magna Smelter	       Q_88

 C-61    Summary of Arsenic Test  Data  — Converter Fugitives  (Full
        Cycle), Kennecott-Magna  Smelter	C-89

 C-62    Summary of Arsenic Test  Data — Rollout Converter Fugitives
        Kennecott-Magna Smelter	 I   C-90

 C-63    Summary of Arsenic Test Data — Venturi  Scrubber Inlet
        Kennecott-Hayden Smelter 	   c_91

C-64   Summary of Arsenic Test Data — Venturi  Scrubber Outlet
        Kennecott-Hayden Smelter 	   c_92

C-65   Summary of Arsenic Test Data — Acid Plant Outlet,
       Kennecott-Hayden Smelter 	   c_93

C-66   Visible Emissions  Observation  Data,  EPA  Method  22—Roaster
       Calcine Discharge  Into Larry Cars, ASARCO-Tacoma 	  C-94

C-67   Visible Emissions  Observation  Data,  EPA  Method  22—Matte
       Tap Port  and Matte Launder,  ASARCO-Tacoma	c-95
                                  xvn

-------
                             LIST OF TABLES
                               (concluded)

Table                                                                 Paqe
 C-68   Visible Emissions Observation Data,  EPA Method 22~Matte
        Discharge Into Ladle,  ASARCO-Tacoma	C-96
 C-69   Visible Emissions Observation Data,  EPA Method 22--Slag
        Tap Port and Slag Launder,  ASARCO-Tacoma	C-97
 C-70   Visible Emissions Observation Data,  EPA Method 9—Slag
        Tap and Slag Launder,  ASARCO-Tacoma	C-98
 C-71   Visible Emissions Observation Data,  EPA Method 22--Slag
        Tapping at Slag Discharge into Pots,  ASARCO-Tacona  	  C-99
 C-72   Visible Emissions Observation Data,  EPA Method 9~Slag
        Tapping at Slag Discharge into Pots,  ASARCO-Tacoma  	  C-100
 C-73   Visible Emissions Observation Data,  EPA Method 22—Converter
        Slag Return to Reverberatory Furnace, ASARCO-Tacoma	C-101
 C-74   Visible Emissions Observation Data,  EPA Method 9--Converter
        Slag Return to Reverberatory Furnace, ASARCO-Tacoma	  C-102
 C-75   Visible Emissions Observation Data,  EPA Method 9—Blister
        Discharge From Converter at the Tamano Smelter in Japan. .  .  C-103
 C-76   Summary of Average Observed Opacities for Blister Discharge
        At the Tamano Smelter  in Japan	C-104
 C-77   Summary of EPA Method  9  Visible Emissions Data--Individual
        and Total  Matte Charges  to  Converter  Observed  at  the Tamano
        Smelter in Japan	C-105
 C-78   Summary of Visible Emissions Observation Data—Copper Blow
        At the Tamano Smelter  in Japan	C-106
 C-79   Summary of Visible Emissions Observation Data—Slag Blow
        At the Tamano Smelter  in Japan	C-107
 C-80   Summary of Visible Emissions Observation Data—Converter
        Slag Discharge At the  Tamano Smelter  in Japan	C-108
 E-l     Identification of Low-Arsenic Primary Copper Smelters  . .  .  E-15
 E-2     Input Data to Exposure Model  Low-Arsenic Primary  Copper
        Smelters	E-16
 E-3     Total  Exposure and Number of People  Exposed   	  E-17
 E-4     Public Exposure for Low-Arsenic Copper Smelters as
        Produced  by the Human  Exposure Model  	  E-18
 E-5     Maximum Lifetime Risk  and Cancer Incidence for Low-Arsenic
        Primary Copper Smelters	E-23
                                  xvn

-------
                             LIST OF  FIGURES
                                                                     Pacje
 2-1     Primary  Copper  Smelter ...................   2-6
 2-2     Primary  Copper  Smelting  Process  ..............   2-7
 2-3     Calcine  Roaster ...................          2-g
 2-4     Reverberatory Smelting Furnace ...............   2-12
 2-5     Copper Converter ......................   2-16
 2-6     Fugitive  Emission  Sources  at  Primary  Copper Smelters.  .  .  .   2-39
 3-1     Arsenic Trioxide Vapor Pressure  and Saturated
        Vapor Concentration with Temperature ............   3.3
 3-2     Contact Sulfuric Acid  Plant  ................   3_15
 3-3     Types of  Exhaust Hoods ...................   3_2Q
 3-4     Uses of Air Curtains ....................   3_2i
 3-5     Spring-Loaded Car  Top  and  Ventilation Hood,  ASARCO-
        Hayden ...........................   3_24
 3-6     Matte Tapping Fugitive Control System (Plan  View),
        ASARCO-Tacoma .....................        3_25
 3-7     Matte Tapping and  Ladle Hoods ...............    3_27
 3-8     Launder Cover .....................        3_2g
 3-9     Slag Tapping Fugitive Control System  (Plan  View),
        ASARCO-Tacoma ............ .....  \  .....    3_30
 3-10    Typical  Converter  Fixed Secondary Hood ...........    3.33
 3-11    Conceptual Design  for Converter Mechanical  Secondary
        Hood System ........................    3_34
 3-12    Converter Air Curtain Control System ............    3.37
 3-13    Converter Air Curtain Secondary Hood,  Onahama and
        Naoshima Smelters  .....................    3_38
 3-14   Air Curtain System at the Tamano Smelter ..........    3.40
 3-15    Controlled Airflow from a Heated Source ..........    3.44
 3-16   Uncontrolled Airflow from a Heated Source .........    3.44
 3-17   Anode Furnace Movable Hood .................    3.47
3-18   SF6 Tracer Injection Locations ...............    3_68
3-19   Tracer Injection Test Ports ................    3.59
3-20   Control  Device Arsenic Collection Efficiencies .......    3-82
                                   xvi

-------
                              LIST OF FIGURES
                                (continued)
                                                                       Page
  4-1     ASARCO-E1  Paso  Smelter  Baseline Configuration   .......   4-17
  4-2     ASARCO-Hayden  Smelter Baseline Configuration ........   4-19
  4-3     Tennessee  Chemical  Company  Smelter Baseline Configuration.  .   4-20
  4-4     Inspiration-Miami  Smelter Baseline Configuration ......   4-22
  4-5     Kennecott-Garfield  Smelter  Baseline Configuration ......   4-23
  4-6     Kennecott-Hayden Smelter Baseline  Configuration .......   4-25
  4-7     Kennecott-Hurley Smelter Baseline  Configuration .......   4-27
  4-8     Kennecott-McGill Smelter Baseline  Configuration .......   4-28
  4-9     Magma-San  Manuel Smelter Baseline  Configuration .......   4-30
  4-10    Phelps  Dodge-Ajo Smelter Baseline  Configuration .......   4-31
  4-11    Phelps  Dodge-Douglas  Smelter Baseline  Configuration .....   4-33
  4-12    Phelps  Dodge-Hidalgo  Smelter Baseline  Configuration .....   4-34
  4-13    Phelps  Dodge-Morenci  Srnelter Baseline  Configuration .....   4-36
  4-14    Copper  Range-White  Pine  Smelter Baseline Configuration  .  .  .   4-38
 F-l(a)   Arsenic Distribution  at  ASARCO-E1  Paso Smelter  .......   F-3
 F-l(b)   Overall Arsenic Material  Balance At ASARCO-E1 Paso
         Smelter ...........................   p_4
 F-2(a)   Arsenic Distribution  at  ASARCO-Hayden  Smelter ........   F-5
 F-2(b)   Overall Arsenic Material  Balance at ASARCO-Hayden Smelter.  .   F-6
 F-3(a)   Arsenic Distribution  at  TN Chemical Co.-Copperhill
         Smelter ...........................   p_8
 F-3(b)   Overall Arsenic Material  Balance at TN Chemical  Co.-
         Copperhill  Smelter  .....................   p_g
 F-4(a)  Arsenic Distribution  at  Inspiration-Miami Smelter ......   F-10
 F-4(b)  Overall Arsenic Material  Balance At Inspiration-Miami
        Smelter ..........................     p_-^
 F-5(a)  Arsenic Distribution at  Kennecott-Garfield Smelter .....   F-12
 F-5(b)  Overall Arsenic Material Balance at Kennecott-Garfield
        Smelter .....................               F-13
F-6(a)  Arsenic Distribution at Kennecott-Hayden Smelter ......   F-15
F-6(b)  Overall Arsenic  Material Balance at Kennecott-Hayden
        Smelter .....................  ^              p_16
F-7(a)  Arsenic Distribution at  Kennecott-Hurley Smelter ......   F-17
                                   xix

-------
                             LIST OF FIGURES
                               (concluded)

                                                                      page
F-7(b)  Overall Arsenic Material  Balance at Kennecott-Hurley
        Smelter	   p_18
F-8(a)  Arsenic Distribution at Kennecott-McGill Smelter	   F-19
F-8(b)  Overall Arsenic Material  Balance at Kennecott-McGill
        Smelter	   P_2Q
F-9(a)  Arsenic Distribution at Magma Copper Company-San
        Manuel  Smelter  	   P_22
F-9(b)  Overall Arsenic Material  Balance at Magma Copper Company-
        San Manuel  Smelter	   p_23
F-10(a) Arsenic Distribution at Phelps Dodge-Ajo Smelter	   F-24
F-10(b) Overall Arsenic Material  Balance at Phelps Dodge-Ajo
        Smelter	   P_25
F-ll(a) Arsenic Distribution At Phelps Dodge-Douglas Smelter. . . .   F-27
F-ll(b) Overall Arsenic Material  Balance at Phelps Dodge-Douglas
        Smelter	   p_28
F-12(a) Arsenic Distribution at Phelps Dodge-Hidalgo Smelter. . . .   F-29
F-12(b) Overall Arsenic Material  Balance at Phelps
        Dodge-Hidalgo Smelter 	   p_30
F-13(a) Arsenic Distribution at Phelps Dodge-Morenci Smelter. . . .   F-31
F-13(b) Overall Arsenic Material  Balance at Phelps Dodge-
        Morenci Smelter .	   p_32
F-14(a) Arsenic Distribution at Copper Range Company Smelter. . . .   F-34
F-14(b) Overall Arsenic Material  Balance at Copper Range Company
        Smelter	   P_35
                                   xx

-------
                              1.0  SUMMARY

1.1  STATUTORY AUTHORITY
     National  emission standards  for hazardous  air pollutants  are
established in accordance with Section 112(b)(l)(B) of  the  Clean
Air Act (U.S.C. 7412), as amended.   Emission standards  under Section
112 apply to new and existing sources of a  substance that has  been
listed as a hazardous pollutant.   This study examines inorganic  arsenic
emissions from primary copper smelters which process feed material  with
an annual average inorganic arsenic content of  less than 0.7 percent  by
weight.  This category of primary copper smelters  is defined as  "low-
arsenic throughput smelters."  There are currently 14 primary  copper
smelters in this category, and none of them is  expected to  increase
the annual average inorganic arsenic content of its feed materials
to or above 0.7 percent.  The single existing high-arsenic  throughput
smelter, owned and operated by ASARCO, Incorporated, and located in
Tacoma, Washington, is not expected to decrease the annual  average
inorganic arsenic content of its  feed materials to below 0.7 percent
and no new smelters are projected to be built during the next  5  years.
For this reason, only the 14 existing low-arsenic  throughput primary
copper smelters are analyzed in this document with respect  to  the
environmental, energy, and economic impacts of  regulating the  low-arsenic
throughput smelter category.  The existing high-arsenic throughput
copper smelter is analyzed in the document, "Inorganic Arsenic Emissions
from High-Arsenic Primary Copper Smelters - Background Information  for
Proposed Standards" (EPA-450/3-83-009a).
1.2  REGULATORY ALTERNATIVES
     Review of the technical support data led to the development of
five regulatory alternatives.  Alternative I would require  no  additional
regulatory action.  This alternative would rely on existing regulations
and existing controls to limit emissions of inorganic arsenic.
     Alternative II would require the control of process arsenic emissions,
This alternative is based on the use of flue gas cooling followed  by  a
particulate control device to collect process arsenic emissions.

                                  1-1

-------
 Sources  of  process  arsenic emissions  include the  roasters,  smelting
 furnaces, and  copper  converters.
      Alternative  III  would require the capture of  fugitive  arsenic
 emissions from converter  operations and the collection of these fugitive
 emissions in a particulate control device.  Fugitive emissions are
 those that  escape capture and control through the  primary control
 equipment.  This alternative is based on the use of an air  curtain
 secondary hood consisting of a fixed enclosure and an air curtain
 system followed by  a  particulate control device (baghouse or equivalent
 technology).
      Alternative IV would require the capture of fugitive emissions
 from  furnace slag tapping operations and the collection of  inorganic
 arsenic  emissions from furnace slag tapping and matte tapping operations.
 This  alternative is based on the use of localized hoods to  capture the
 slag  tapping fugitive emissions and a particulate control device (bag-
 house  or equivalent technology) to collect the inorganic arsenic emissions
 from  both furnace slag tapping and matte tapping.
      Alternative V would require the elimination of all  arsenic emissions
 at copper smelters.  To accomplish this alternative the smelters would
 be forced to process ores which were virtually free of arsenic content.
 1.3   ENVIRONMENTAL IMPACTS
      Under Alternative I there would be no additional  inorganic arsenic
 reduction from the baseline because there is no additional   regulatory
 action associated with Alternative I.   Under Alternative II, inorganic
 arsenic emissions from process sources would be reduced  by  163 megagrams
 per year (Mg/yr).  This represents a 30 percent reduction in process
 inorganic arsenic emissions and a  22 percent reduction in overall
 inorganic arsenic emissions from the low-arsenic copper  smelter category.
 Energy requirements  at the one smelter affected by Alternative II  would
 increase by  4.9 x 10® kWh.  This  represents a  33 percent  increase  in
the energy requirements at the one affected smelter.   Solid  waste  increases
at the affected smelter would  amount  to 16,300  Mg/yr.   This  represents
a 0.5 percent  increase in  solid  waste  which must be handled  by the  smelter.
     Under Alternative III,  fugitive  arsenic emissions  from  converter
operations would be  reduced  by  118 Mg/yr.   This  represents a reduction
of 60 percent  in fugitive  inorganic  arsenic emissions  and  a  reduction
                                  1-2

-------
of 16 percent in overall  inorganic arsenic  from the source category.
Energy requirements for this  alternative would increase by 1.8 x  10^  kWh.
This represents a 0.35 percent increase in  energy requirements for the
source category.   Solid  wastes generated would increase by 11,800 Mg/yr,
which represents a 0.4 percent increase for the source category.
     Under Alternative IV,  fugitive inorganic arsenic  emissions from
matte and slag tapping operations would be  reduced by  11 Mg/yr.  This
represents a reduction of 5 percent in fugitive arsenic emissions and
a reduction of 1.5 percent  in overall  inorganic arsenic emissions from
the source category.  Energy  requirements for the source category would
increase by 9 x 10^ kWh,  which represents a 0.02 percent increase in
energy requirements.  Increases in solid wastes generated would amount to
1,100 Mg/yr, or an increase of 0.03 percent for the source category.
     The air quality, energy, and solid waste impacts, as well as all
environmental and economic  impacts, are summarized in  Table 1-1.
     The control systems  for  the regulatory alternatives are dry
systems; consequently, no incremental  increase in water discharges
is anticipated.  If scrubbers are used, increases in wastewater
discharges would result and the scrubber discharge would be treated
within existing water pollution control systems.  Therefore, even if
scrubbers are used, no adverse water pollution impact  is anticipated.
     The regulatory alternatives would result in negligible impacts
on noise, space, and availability of resources.
1.4  ECONOMIC IMPACTS
     The total capital cost for the process controls specified under
Alternative II is $9.8 million.  The annualized cost for the controls
at the Kennecott-McGill smelter would  be $4.1 million.  The total
annualized cost for this  alternative,  coupled with the annual  emission
reduction expected under  Alternative II, would yield an annualized
cost-effectiveness of $25,200/Mg of inorganic arsenic  reduced.  The
total capital cost for the  fugitive emission controls  for the  converter
operations specified under  Alternative III  is $109.9 million.   The
annualized cost of these  converter controls would be $29.2 million.
The total annualized costs  associated  with  the converter controls,
                                  1-3

-------
              Table 1-1.  ASSESSMENT OF ENVIRONMENTAL AND ECONOMIC IMPACTS FOR EACH
                                REGULATORY ALTERNATIVE CONSIDERED

Regulatory
Alternative
I
II
III
IV
V
Air
Impact
0
+2**
+2**
+2**
+4**
Water
Impact
0
-1*
-1*
-1*
+4**
Solid Waste
Impact

-1*
-1*
-1*
-3**
Energy
Impact
0
_!**
_!**
_!**
+4**
Noise
Impact
0
-1*
-1*
-1*
+1**
Economic
Impact
0
.4**
-3**
-3**
_4***
Key:   + Beneficial  Impact
      - Adverse Impact
      0 No impact
      1 Negligible  Impact
      2 Smal1  Impact
  3 Moderate Impact
  4 Large Impact
  * Short-Term Impact
 ** Long-Term Impact
*** Irreversible Impact

-------
 coupled with the emission  reduction  expected  under  Alternative  III,
 would yield an annualized  cost-effectiveness  of  $247,000/Mg  of  inorganic
 arsenic reduced.
      The total  capital  cost  for the  fugitive  emission  controls
 for the matte and slag  tapping  operations  specified under Alternative  IV
 is $14.8 million.  The  annualized  cost  associated with these fugitive
 emission controls would be $4.3 million.   The total  annualized  cost  of
 these matte and slag tapping fugitive emission controls,  coupled with
 the emission reduction  expected under Alternative IV,  would  yield an
 annualized  cost-effectiveness of $389,000/Mg  of  inorganic arsenic reduced.
      In 1982, copper producers  experienced one of the  worst  years in
 recent history.  Such a situation  cannot be used as  the foundation to
 examine the long-term economic  impact of the  potential  arsenic  NESHAP.
 Therefore,  the  economic analysis is  based  on  a more  normal condition for
 the industry.  The principal economic impacts  analyzed are:  the ability
 of the smelters to increase  copper prices  in  response  to  an  increase in
 costs due to the arsenic standard; and, the impact  on  profits if part
 or all  of the costs  cannot be passed on in the form  of price  increases.
      If each smelter attempts to maintain  its  normal profit margin and
 pass  control  costs  forward in the  form of  a price increase, the price
 increases would range from 0.1  percent to  15.2 percent.   However, competi-
 tion  will prevent  the existence  of such a  broad variation.   If control
 costs are absorbed and  profit margins reduced, again a  broad  range
 exists.   The profit  reductions would range from 0.4  percent to 151.9
 percent.  At  a  100 percent capacity utilization rate and  a price of
 187
-------
     Alternative V, which would eliminate arsenic emissions  from the
low-arsenic throughput smelter category  would result  in  closing of all
the smelters in the source category.   Although this results  in the
greatest emission reduction,  the economic impacts and hardships associated
with this alternative are severe.
                                  1-6

-------
                    2.0  THE PRIMARY COPPER INDUSTRY

2.1  GENERAL
     Currently,  there are 15 primary copper smelters operating or
temporarily closed in the United States.   Of these, seven are located
in Arizona, two  in New Mexico, and one each in Nevada, Texas, Utah,
Tennessee, Michigan, and Washington.  The concentration of copper
smelters in the  Southwest is due mainly to the local availability of
copper-bearing ores.  For the 15 copper smelters, smelting capacity
totals approximately 1.72 million Mg (1.9 million tons) of smelter
product (99 percent "blister" copper) per year.   Table 2-1 lists
these 15 smelters, their locations, and estimated capacities.
                                                                       2
     Primary copper production in 1982 was 975,437 Mg (1,075,400 tons).
This represented approximately a 60 percent utilization of domestic
primary copper smelting capacity.  Smelter capacity in the United
States appears to be relatively steady, with minimal prospects for
substantial additions before 1985.  The Bureau of Mines forecasts the
total demand for copper in the year 2000 to be between 3.5 and 6.0 teragrams
(Tg) (3.9 x 10  and 6.6 x 10  tons).  This forecast represents an
annual growth of 3.6 percent.  However, recycling of scrap and the
expansion of hydrometallurgical facilities are expected to accommodate
                                               3
a substantial portion of this increased demand.
2.1.1  Raw Materials
     Copper ores are generally classified as sulfide, oxide, or native
depending on the predominant copper-bearing minerals they contain.
Although copper occurs in at least 160 minerals, only a few of these
have any commercial importance.  Table 2-2 lists the composition of
the more important sulfide and oxide minerals from which copper is
extracted.  Of the primary sulfide minerals, chalcopyrite is the most
abundant, followed by bornite and chalcocite.  Oxide minerals are
produced by the  oxidation of primary sulfide minerals under certain

                                2-1

-------
              Table 2-1.  DOMESTIC PRIMARY COPPER SMELTERS
       Company
     Location
  Annual capacity9
Megagrams(Tons;
ASARCO, Incorporated



Tennessee Chemical Company

Inspiration Consolidated
  Copper Company

Kennecott Copper Corporation
Magma Copper Company
Phelps Dodge Corporation
Copper Range Company
       TOTAL
El Paso, Texas
Hayden, Arizona
Tacoma, Washington

Copperhill, Tennessee
Miami, Arizona


Garfield, Utah
Hayden, Arizona
Hurley, New Mexico
McGill, Nevada

San Manuel, Arizona
Ajo,  Arizona
Douglas, Arizona
Hidalgo, New Mexico
Morenci, Arizona

White Pine, Michigan
 91,000
182,000
 91,000

 13,600

136,000


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

 (15,000)

(150,000)


(280,000)
 (78,000)
 (80,000)
 (50,000)
181,000    (200,000)
 64,000
115,000
163,000
191,000
 (70,000)
(127,000
(179,000)
(210,000)
 52,000     (57,000)-
                                                      1,722,600  (1,896,000)
 Production of "blister"  copper (99  percent Cu)
                                     2-2

-------
               Table 2-2.   MAJOR COPPER-BEARING MINERALS
Type
Sulfide



Oxide



Mineral
Chalcopyrite
Bornite
Chalcocite
Covellite
Malachite
Azurite
Chrysocolla
Cuprite
Formula
CuFeS2
Cu^FeS-
Cu2S
CuS
CuC03'Cu(OH)2
2CuCO,'Cu(OH)9
O C-
CuSi03'2H20
Cu20
climatic conditions.  When present, these are usually found in the
upper portions of copper ore deposits.   Native copper consists of
almost pure metallic copper.  Although  found in small amounts in many
copper ore deposits, it is essentially  unique to the upper peninsula
of Michigan.
     Copper ores consist of one or more of these copper-bearing minerals
disseminated within relatively large quantities of siliceous and other
earthy matter.  In addition, they also  contain varying amounts of
other metals including sulfides and oxides of iron, arsenic, antimony,
lead, zinc, etc.  In the United States, low grade sulfide ores account
                                                     4
for 85 to 95 percent of the total  primary production.   The average
tenor of these ores (copper content) is less than 1 percent.   Virtually
all copper ores processed are beneficiated at the mine.  Sulfide ores
are crushed, finely ground, and concentrated by froth flotation.
Oxide ores are leached with acid,  and the dissolved copper is re-
covered by chemical precipitation on scrap iron.  Typically, copper
ore concentrates contain about 15 to 30 percent copper.
     In addition to copper, metals such as arsenic are also concentrated
in the concentration process.  The arsenic content of copper concentrates
processed at U.S. copper smelters  is highly variable, ranging from a
few parts per million to several  percent.  The impurities, mainly
arsenic and others such as antimony, bismuth, lead, and zinc exert a
strong influence on the selection  of smelting technology.  Table 2-3
                                2-3

-------
               Table 2-3.  ARSENIC INPUT IN THE FEED TO
                       DOMESTIC COPPER SMELTERS
Plant
ASARCO-Tacoma
ASARCO-Hayden
ASARCO-E1 Paso
Kennecott-Garfield
Kennecott-McGill
Phelps Dodge-Ajo
Inspiration-Miami
Phelps Dodge-Hidalgo
Phelps Dodge-Douglas
Kennecott-Hayden
Phelps Dodge-Morenci
Magma-San Manuel
Tennessee Chemical Company -
Copperhill
Kennecott-Hurley
Copper Range - White Pine
Arsenic
Percent
4.0
0.6
0.5
0.14
0.4
0.3
0.033
0.018
0.03
0.015
0.006
0.006

0.0004
0.0005
0.008
content of
kg/hr
991
170
142d
118e
81f
47
189
14
11
8.0
4.5h
2.0

1.3
1.0
0.71"
feeda'b
(Ib/hr)
(2,185)
(375)
(314)
(261)
(179.3)
(103)
(41.1)
(30.6)
(24)
(17.7)
(9.99)
(4.39)

(2.9)
(2.14)
(1.53)
 The feed is a mixture of concentrates, precipitates, lead smelter
 by-products, and smelter reverts.
 Does not include recycled flue dusts and other intermediates.
 50 kg/hr (111 Ib/hr) of this amount is fed directly to the arsenic plant,
 51 kg/hr (112 Ib/hr) of this amount is fed directly to the converters.
/>
 3.5 kg/hr (7.8 Ib/hr) of this amount is fed directly to the converters.
 6.5 kg/hr (14.2 Ib/hr) of this amount is fed directly to the converters.
90.3 kg/hr (0.6 Ib/hr) of this amount is fed directly to the converters.
 0.2 kg/hr (0.35 Ib/hr) of this amount is fed directly to the converters.
 0.1 kg/hr (0.22 Ib/hr) of this amount is fed directly to the converters.
                                2-4

-------
  presents the amount of arsenic input to the domestic primary copper
  smelters based on information received from the smelters in early
  iqoo °»'>o,y,1U,11,1^      nfitn^n
      '                  Tne ASARCO-Tacoma smelter, which introduces
  more arsenic in the feed material to the smelter than the combined
  total  from all  of the low-arsenic throughput copper smelters, is
  analyzed in a  separate background information document (EPA-450/3-83-009a).
  2.1.2  Process  Description
       The pyrometallurgical  process used for the extraction  of copper
  from sulfide ore concentrates  is  based  on  iron's  strong affinity for
  oxygen  as  compared  to  copper's weak  affinity for  oxygen.  The purpose
  of  smelting  is  to separate  the copper from  the  iron,  sulfur,  and
  gangue  materials.   Conventional practice includes  three operations:
       1.   Roasting  (optional) to  remove a portion  of  the concentrate
  sulfur  content.
      2.   Smelting of  roasted calcines  or unroasted ore  concentrates
  and  fluxes in a furnace to  form slag and copper-bearing matte.
      3.   Converting (oxidizing)  of the matte in a converter to  form
  blister copper  (about 99 percent  pure copper).
      Figure 2-1 presents a  pictorial  representation of the copper
  smelting process.  Figure 2-2 illustrates the three basic operations
 employed as well as  materials entering  or leaving each operation.
 Briefly, the smelting of copper concentrates and precipitates is
 accomplished by  melting the charge and  suitable fluxes in a  smelting
 furnace.  Part  or all  of the concentrates may receive  a partial  roast
 to  eliminate some of  the sulfur and impurities such as arsenic.   In
 the  smelting  furnace,  the  lighter  impurities combine and float to the
 top  as  slag  to be  skimmed  off and  discarded, while  the copper,  iron,
 most of  the  sulfur, and  any  contained precious metals  form a  product
 known as matte which  collects  in and  is  drawn  off from the lower  part
 of the furnace.  The molten  matte  is  transferred to a  converter where
 air  blown through the matte  burns  off the sulfur, oxidizes the  iron
 for  removal in a slag, and yields  a 99 percent blister copper product.
 Typically, the blister copper is further refined in an anode furnace
 prior to the casting  of copper anodes for electrolytic refining.  A
more detailed discussion of  roasting, smelting, and converting operations
 is presented in  succeeding paragraphs.
                                2-5

-------
          EMISSION CONTROL EQUIPMENT
CONVERTER
                                                                               WASTE HEAT
                                                                                BOILERS
                                             REVERBERATORY
                                                FURNACE
                              Figure 2-1.  Primary  Copper Smelter

-------
      HTEHDIS THE SYSTEM
LEIYING THE SYSTEM
      Raw canontratts
     Fuel
     lir
                                       ROASTER
 Flui and
 fettl ing material
Fuel
Air
                                SMELTING FURNACE
Siliceous  f!ux
Mi see Ilaneous
material high in copper
Air
                                    CONVERTER
                                                       Gases, roiatiIt  oxldei.
and  dust to dust recovery
and  stack
                                                          Gases  and  dust
    to Haste heat  boilers,
    dust recovery,  and stack
   Slag to dump
                                                         Gases to  stack
                                                         Jlisler  copper
                                                         to refinery
              Figure  2-2.  Primary  Copper  Smelting  Process
                                      2-7

-------
     2.1.2.1  Roasting.   In roasting of copper sulfide ore concentrates,
concentrates are heated  to a high temperature (but below the melting
point of the constituents) in an oxidizing atmosphere to eliminate a
portion of the sulfur contained as sulfur dioxide (S02); to remove
volatile impurities such as arsenic, antimony, and bismuth; and to
preferentially convert a portion of the iron sulfides present to iron
oxides.  The roasted concentrate is called calcine.  The degree of
roast (i.e., the amount of sulfur and iron oxidized in the roasting
operation) is dependent on the desired quality of the charge to the
smelting furnace.  Representative reactions include the following:
COUI COp 	
FeS2 —
S
4FeS H
+ 02-
^ 702 -
	 ^ VyUo
— ^ FeS
— *• so2
— »- 2Fe
O 1 £-1 CO
+ S

2°3 + 4SO;
     Currently, 7 of the  15  existing  primary  copper  smelters  roast
 concentrates prior to smelting.  Two  types  of roasters  are  used:
 multiple-hearth roasters  and fluid-bed  roasters  (refer  to Figure 2-3).
 With both  types, the roasting  process  is  generally  autogenous.   The
 roaster  operating temperature  is typically  about 650°C  (1,200°F).
      In  multi-hearth type roasters,  the hearths  are  constructed of
 refractory brick with a  slight arch.   The external  portion of the
 furnace  is a brick-lined, steel shell  with  hinged doors and inspection
 plates at  each  level.  The moist concentrate  enters  the roaster through
 an annular opening to the top-most or dryer hearth.   Rabble arms,
 attached to the hollow central shaft, rotate  as  the shaft turns and
 plow  through the charge  to continuously expose fresh surfaces to the
 oxidizing  air.   The  rabble blades  are set at  an  angle and, in addition
 to stirring the material, move it  alternately from the center of the
 hearth  to  the  periphery  where it  falls to the next lower hearth.
 Finished calcine  is  discharged through holes  on  the circumference of
 the bottom hearth.
      The air required  for roasting is admitted through the central
 shaft and, by  means  of  valves, the air supply to each hearth may be
                                 2-8

-------
     OFF
     GAS
FEED
 4
          DRYING
       HOT AIR
      TO EXHAUST
RABBLE
 ARM
 RABBLE
 BLADE
  CALCINE
                                   SLURRY
                                    FEED
                                                     TUYERE
                                                     HEADS
                                                    AIR
                                                                                            OFF-GAS

                                                                                              fl
           MULTI-HEARTH ROASTER
                                                 FLUID-BED ROASTER
                                  Figure 2-3.  Calcine  Roaster

-------
 regulated.   Roaster gases  are drawn  off through gas outlets located
 just  below  the  dryer hearth.   The  discharge from the dryer hearth to
 the  top  roasting  hearth  prevents the escape of  gas  from the interior
 of the roaster.
      Fluid-bed  roasters  are  cylindrical,  refractory-lined  vessels
 equipped  with diffusion  plates  in  the bottom containing tuyeres  or
 bubble caps  through which  air is blown from the bottom.  Finely  ground
 material  (60 percent minus 200  mesh)  is  introduced  either  as  a slurry
 through  a feed  pipe or relatively  dry (6  to 12  percent  moisture)
 through  a screw conveyor.  The  feed  is continuously delivered into the
 combustion  chamber.   Roasting occurs  as  the sulfide particles fall
 through  the  oxidizing air.
      Combustion air from the  windbox  passes  through distribution
 plates at the bottom of  the chamber  into  the combustion zone.  Because
 of the large surface area of  the finely ground  material  exposed  to the
 air stream,  the residence time  in  the  oxidizing  atmosphere is  short.
 The reaction is self-sustaining.   Oil, gas,  or  pulverized  coal burners
 are required only to preheat  the roaster  to  combustion  temperature.
      Roaster gases  are drawn  off through  flues  at the top  of  the
 chamber and  immediately pass  to cyclone collectors,  followed  by cooling
 and final dust collection.  As much  as 85 percent of the feed  is
 carried with the gas stream and, hence, cyclones are an  integral  part
 of the roasting operation.  Screw  conveyors  collect this finished
 calcine as well  as  that from the bottom of the combustion  chamber  for
 discharge into hoppers.
      For  either type of roaster, there are  three major  operating
 variables:  feed rate, combustion  air  flow rate, and temperature.   A
 key difference between the two is  the S02 concentration  in  the roaster
 offgases.  Sulfur dioxide concentrations in  the roaster offgases are
 considerably higher  for fluid-bed  roasters than for multi-hearth
 roasters  due to the  lower total air volume.  Average stack  gas S0?
concentration is about 12 percent with a maximum of  18  percent for
fluid-bed roasters and only 3 to 6 percent for multi-hearth roasters.14
     2.1.2.2  Smelting.   Smelting  is the pyronetallurgical  process  in
which solid  material is melted and  subjected to certain chemical
                                2-10

-------
 changes.  During copper smelting, hot calcines from the roaster or
 raw, unroasted concentrates are melted in a smelting furnace with
 siliceous or limestone flux.  Converter slag, collected dust, oxide
 ores, and any other material rich in copper may be added to the furnace
 charge.   Copper and iron which are present in the charge combine with
 sulfur to form a stable cuprous sulfide.   Excess  sulfur unites with
 iron to  form a stable ferrous  sulfide, FeS.   The  combination of the
 two sulfides,  known as matte,  collects in the lower area of the furnace
 and is removed.   Such mattes may contain  from 15  to 50 percent copper,
 with a 40 to 45  percent copper content being most common.   Mattes  may
 also contain impurities such as sulfur, antimony,  arsenic,  iron,  and
 precious  metals.
      The  remainder  of the  molten mass  containing  most  of the other
 impurities  is  known as slag.   Slag  is  of  lower specific gravity,
 floats on top  of  the matte,  and is  drawn  off and  discarded.  Slags in
 copper smelting  are ideally  represented by the composition  2FeO'SiO?,
 but contain  alumina from the various charge  materials  and  calcium
 oxide  which  is added  for fluidity.   Since slags are discarded,  the
 copper contained  in the slag is  a major source of  copper loss  in pyro-
 metallurgical  practice.  Copper  concentration  in  the slag  increases
 with  increasing matte  grade.   This  behavior  limits  the  matte grades
 normally  obtained in  conventional practice to  below 50  percent  copper.
     Currently, conventional reverberatory furnaces  are used at 11 of
 the 15 existing primary copper  smelters.  Two  smelters  employ  electric
 furnaces, while one smelter  employs  the Outokumpu  flash furnace and
 another a Noranda continuous smelter.  Two smelters will be modifying
 one or more of their  reverberatory furnaces  to convert  them to oxygen-
 sprinkle smelting,  while two more are  planning to retire their reverberatory
 furnaces within the next few years and install Inco flash smelting
 furnaces.  In a reverberatory furnace  (Figure 2-4), fossil  fuels such
 as  oil or natural gas are burned above the copper concentrates being
 smelted.   The furnace is a long, rectangular structure, generally
 about 11 m (36 ft)  in width and 40 m (131  ft) in length, with an
arched roof and burners at one  end.   Flames from the burners may
extend half the length of the furnace.   Temperatures at the  firing end
of the furnace exceed 1,500°C (2,730°F).   Part of  the heat  in the

                                2-11

-------
I
ro
                      CALCINE

                         V
                     FUEL
FETTUNG DRAG
  CONVEYOR
OFPGAS
               AIR AND
               OXYGEN
               CONVERTER
                 SLAG
                      BURNERS
                                    MATTE
                                                                                                    SLAG
                   SLAG
                                  FETTLING PIPES
                                                                                             MATTE
                                 Figure 2-4.   Reverberatory Smelting  Furnace

-------
  combustion  gas  radiates  directly  to  the  charge  lying  on the hearth
  below, while a  substantial  part radiates  to  the furnace roofs  and
  walls  and is reflected down  to the charge.   The roofs  of  the older
  reverberatory furnaces are  sprung-arch silica roofs, while  almost all
  newer  furnaces  have suspended roofs  of basic refractory.
      Over the years, two types of reverberatory furnaces  have  evolved,
  each with its own specific charging  methods.  The  older type is  the
  deep bath reverberatory furnace which contains  a large  quantity  of
 molten slag and matte at all times.  In modern, deep bath reverberatory
 furnaces, the molten material is held in a refractory crucible with
 cooling water jackets along the sides, which greatly diminishes  the
 danger of a  breakout of the liquid material.  In deep bath smelting,
 several methods  exist for charging.   Wet concentrates can be charged
 using slinger belts (high-speed  conveyors) that spread  the concentrates
 on the surface  of the molten bath.  Dry concentrates or calcines  from
 the roaster  can  be charged  through the roof or via  a Wagstaff gun  (an
 inclined  tube).   Roof charging  (side  charging)  is rarely practiced in
 conjunction  with deep bath  smelting  because of  dusting problems with
 fine dry  calcine and  explosion potential  with green charge.   Wagstaff
 guns minimize these  problems  and are  commonly used.
      Side or roof  charging  is usually used with  green  charge.  With
 the charge dropped  through  a  series of  feed holes  in the arch near the
 walls,  a  buildup of material  forming  banks results. The banks slowly
 melt and  serve as  protection  to the side  walls,  eliminating  the requirement
 for special  cooling.
      Combustion  gases contain from 15 to  45 percent of  the sulfur in
 the  original charge depending primarily upon  whether or  not  the concentrate
 was  roasted.  However, because of  the high  volume of combustion air,
 S02  concentrations are low, with averages  varying from 0.5 to
 2.0  percent.15  These lean S02 mixtures, unlike  offgases from fluid-bed
 roasters,  converters, and other types of smelting furnaces, cannot be
 economically utilized as feed for sulfuric  acid  plants.
     Electric smelting furnaces  provide the heat necessary for  smelting
copper ore concentrates by allowing carbon  electrodes to come into
contact with  the  molten bath within the furnace.   The electrodes  dip
into the slag layer of the bath,  forming an electrical  circuit.   When
                                2-13

-------
an electric current is  passed through this circuit, the slag resists
its passage, generating heat and producing smelting temperatures.
Charge concentrates and fluxing materials are fed through the roof,
and a layer of unsmelted charge covers the molten bath.  Heat is
transferred from the hot slag to the charge floating on its surface,
and as the copper concentrates and fluxes are smelted, they settle
into the bath forming slag and matte.  The chemical and physical
changes occurring in the molten bath are similar to those occurring  in
the molten bath of a reverberatory furnace.
      In flash smelting, copper sulfide ore concentrates are smelted  by
burning a  portion of the  iron  and sulfur  contained  in  the concentrates
while they are suspended  in  an oxidizing  environment.  As such,  the
process is quite similar  to  the combustion of  pulverized coal.   The
concentrates  and fluxes are  injected  with  preheated air, oxygen-enriched
air,  or even  pure  oxygen,  into a  furnace  of  special design.   Then,
smelting  temperatures  are  attained  as  a  result of  the  heat  released by
the  rapid, flash combustion of iron and  sulfur.  Flash smelting technology
has  been  developed  by  two companies:   International  Nickel  Company
 (INCO)  in Canada and Outokumpu Oy in Finland.   The major difference
between  the two  technologies is in  the design  of the  smelting furnace
 and  the  oxidizing  environment within the furnace.   The INCO furnace
 uses pure oxygen,  while the Outokumpu furnace  employs preheated air or
 oxygen-enriched air as the oxidizing medium.
      Two smelters  are  currently planning to convert one or more of
 their reverberatory furnaces to oxygen-sprinkle smelting.  In essence,
 this conversion will allow the existing reverberatory furnaces  to
 behave like flash furnaces.   Specially designed burners positioned  on
 the furnace roof are used to  introduce and disperse a mixture of
 primarily dried concentrates  and oxygen.  The  heat required  for smelting
 is generated from the flash combustion of the  sulfur  in the  mixture.
 The  result is an  increase  in  furnace  efficiency and  the production of
 a strong  S02 gas stream  capable  of  being  treated  in  a  sulfuric  acid
 plant.
       2.1.2.3  Converting.   Matte produced  in  the  reverberatory furnace
  is  transferred  in  ladles  to the  converters  using  overhead  cranes.
  Fourteen of  the 15 smelters use  converters  of the cylindrical  Pierce-Smith

                                  2-14

-------
  type,  the most  common  size  being 4  by  9  m  (13  by  30  ft).   Figure 2-5
  is a sketch  of  a Fierce-Smith copper converter.   An  alternative  to  the
  Pierce-Smith converter  is the newer Hoboken or "siphon" converter.
  The Hoboken converter,  currently used  by one of the  domestic  smelters,
  is essentially  the same as  the Pierce-Smith converter except  that it
  is fitted with  a side flue  located  at  one end  of  the converter and
  shaped as an inverted U.  This flue arrangement permits siphoning of
  the converter gases from the interior  of the converter, using variable-speed
  fans and dampers, directly  to the offgas collection  system.   By maintaining
 a slightly negative pressure at the converter mouth, it is possible  to
 minimize or eliminate emissions.   Problems in  improper draft  at the
 converter mouth have been reported  by  Inspiration Consolidated Copper
 Company, the only domestic user of  Hoboken converters.   One of the
 five Hoboken converters operated  by Inspiration has been modified to
 eliminate the siphon area.  The modification has resulted in improved
 performance,  and Inspiration intends to modify  its four remaining
 converters  in similar fashion.16
      In both  Pierce-Smith  and Hoboken  converters,  air is blown from
 the  side through a  series  of openings  called  tuyeres.  During  the
 initial  blowing  period  (the  slag  blow), FeS  in  the matte is preferentially
 oxidized  to  FeO  and  Fe304> and  sulfur  is  removed with the  offgases as
 S02.  Flux is added  to  the converter to combine with  iron  oxide  and
 forms a  fluid iron  silicate  slag.  When all  the iron  is  oxidized,  the
 slag is  skimmed  and poured off from  the furnace at various  times
 during slag formation, leaving behind  "white metal" or molten  Cu  S.
     During this stage,  fresh matte  is  charged  into the  converter, and
 the slag  blowing continues until a sufficient quantity of white metal
 has accumulated.  When this  happens, the white  metal  is  oxidized with
 air to blister copper during the "copper blow."  The  blister copper  is
 removed from the converter and cast  or  subjected to additional fire
 refining prior to casting.  Converter blowing rates can vary between
 340 and 855 normal m3/min (12,000 to 30,000 scfm) of air.
     In  general  practice, the matte   is  added to the converter  in two
to six steps,  each step followed by  oxidation of much of the FeS from
the charge.   The resulting  slag is poured from the converter after
                                2-15

-------
                                                OFF-GAS
ro
i
TUYERE
 PIPES
                                                                                 SILICEOUS
                                                                                  FLUX
                                              PNEUMATIC
                                              PUNCHERS
                                       Figure 2-5.   Copper Converter

-------
 each  oxidation  step,  and  a  new matte addition is made.   In this way,
 the amount  of copper  (as  matte)  in  the  converter gradually increases
 until  there is  a  sufficient  amount  for  a  final  copper-making "blow."
 At this  point,  the  FeS  in the  matte is  blown  down to about 1 percent,
 a final  slag is removed,  and the  resulting  white metal  (impure Cu?S)
 is oxidized  to  blister  copper.  The converting  process  is  terminated
 when  copper  oxide begins  to  appear  with the liquid  copper.  The offgas
 flow  rate leaving the primary  head  of converters typically ranges  from
 850 to 1,260 Nm3/min  (30,000 to 45,000  scfm).   The  average S02 concentration
 in these gases  is normally  in  the range of  4  to 5 percent  during the
 slag  blow and 7 to  8 percent during the copper  blow.  Values of the
 overall  average S02 concentration in the  offgases (after gas cleaning)
 from  existing domestic  converting operations  fall  in the range of
 1.6 to 6.5  percent  on a dry basis.   The industry-wide average value,
 as weighted by  the  respective  plant flow  rates,  is  4.6  percent.17
      2.1.2.4  Refining.  Virtually  all  copper produced  by  matte smelting
 is subsequently electrorefined.   For this  reason,  the  final  liquid
 copper product of the smelter must  be suitable  for  the  casting of
 strong, thin anodes.  In a majority  of  cases, anode  copper is fire
 refined directly from molten blister copper.
      Fire refining  is performed in  rotary-type  refining furnaces
 resembling Pierce-Smith converters  or in  small  hearth furnaces.  The
 rotary-type predominates when molten blister copper  is  treated directly,
while hearth furnaces are used when melting of  solid  charges  is  practiced.
The temperature of operation is about 1,130 to  1,150°C, which provides
sufficient superheat for the subsequent casting  of anodes.   There is
very little heat produced by refining reactions,  and  some  combustion
of fuel is necessary to maintain the temperature  in  the furnace.
     Dimensions  of the rotary-type  refining furnaces  vary;  however,  a
4 by 9 m (13 by  30 ft) furnace may  be regarded as typical.   Gas flow
rates  are generally low so as to accurately control  the metal  composition.
Gas  pressures at the tuyeres are near 250 to 600  kPa  (2.5  to  6  atm).
Refining  a 250-ton charge of copper requires 3 to 5 hours:  1/2  to
1 hour for the  oxidation step,  and the remainder  for the deoxidation
("poling")  step.
                                2-17

-------
     A typical  sequence of events in rotary anode furnace refining is:
     1.   Molten blister copper is added to the anode furnace as it
becomes  available from the converters until about 150 to 300 Mg (165 to
331 tons) have been accumulated;
     2.   The accumulated charge is then oxidized by blowing air through
the tuyeres until the sulfur content is lowered to 0.001 to 0.003 percent S,
at which time a small ingot sample of copper shows a slight contraction
or small hole; and at which time the oxygen level in the copper is
about 0.6 percent;
     3.   The oxygen is then removed from the copper by blowing natural
gas, reformed natural gas, or propane through the tuyeres.  The oxygen
level in the anode copper is 0.005 to 0.2 percent after this operation,
which gives a "flat set" to the anodes when they are cast.  The correct
"end point" is determined by casting a sample of the copper (which
should set to a flat surface) or by continuously analyzing for oxygen
with an oxygen probe;
     4.  Finally, the liquid metal may be covered with low sulfur coke
to prevent reoxidation of the copper.
     Several older modifications of this process are still being used
in which the air  is  introduced via steel lances, and in which  the
oxygen-removal step  is performed by  lowering  large  logs or poles of
green wood into  the  copper  (thereby  providing the necessary hydrocarbons).
Both of  these  steps  are clumsy,  and  are being discontinued.
     Similarly,  the  use of  the hearth  type  of anode  furnace has by and
large been discontinued except where scrap  (including  anode scrap)
and/or  blister copper  are melted.  A typical  anode  hearth  furnace
resembles  a small  reverberatory  furnace [width  5 m  (16 ft), length
15 n  (49 ft),  height  3 m  (10  ft),  inside  dimensions] capable  of holding
3UO  Mg  (331 tons)  of  copper.   In  hearth furnaces,  the  air  and  hydrocarbons
are  introduced into  the copper  by  submerging  steel  lances  into the
molten  bath.   Wood  poles  are  often used for the oxygen removal  step.
      Since  this  study  analyzes  14  model plants  representing  the  14  existing
low-arsenic  throughput copper smelters, plant-by-plant descriptions  of
operating  and  emission control  parameters  are presented  in Section  4.3.2
of  this document.
                                 2-18

-------
2.2  ARSENIC BEHAVIOR AND DISTRIBUTION  IN COPPER SMELTERS
     Arsenic elimination during copper  smelting occurs primarily  by
volatilization and slagging.  The bulk  of the arsenic is volatilized
in most pyrometallurgical processes and removed with the offgases.   In
the presence of oxygen, elemental arsenic or arsenic sulfides  oxidize
to arsenic trioxide which is extremely  volatile.  However,  in  an
oxidizing atmosphere, the arsenic trioxide (As,,03) may oxidize  to the
higher oxide (As^OJ which is less volatile and forms stable nonvolatile
                (— J
arsenates with other metallic oxides.
     The elimination of arsenic by slagging is dependent upon  the
                                                           1 O
concentration or partial pressure of the arsenic trioxide.     If  the
partial pressure of the oxide is sufficient to allow it to  penetrate
the slag layer and thus escape, volatilization will occur.  However,
if the partial  pressure is not large enough to allow passage through
the slag layer, the arsenic will be mechanically bound in  the  slag and
thus be eliminated on slag disposal.  The magnitude of the  partial
pressure is dependent upon the amount of arsenic present in the process
streams.
2.2.1  Arsenic Behavior During Roasting.  Arsenic elimination  during
roasting is indirectly dependent upon the grade of matte required
during smelting in the furnace.  This requirement limits sulfur elimination
during the roasting operation to establish the matte controlling
sulfur-to-copper ratio for the calcine  feed to the smelting furnace.
Arsenic elimination during the roasting process is by volatilization
and, therefore, is dependent upon the amount or degree of  roasting.
     As previously described, the roasting operation is primarily one
of oxidation of solid material by means of oxidizing gases.  The
sulfides present in the copper ore that are oxidized include arsenic
sulfides such as FeSAs2S3, As2$3, and As2$5.   The As2$3 boils at  707°C
(1,305°F) and As2S5 sublimes at 500°C (934°F) with decomposition.
Arsenic trioxide sublimes at 457°C (855°F) and melts at 310°C  (590°F).
Thus,  if the solids are heated to above 700°C (1,290°F), the volati-
lization of arsenic would seem to be assured, but such a high temperature
is not practical.   At the beginning of roasting, when there is  a  large
portion of easily fusible sulfides present, high temperature melts the
particles which effectively reduces the surface-to-volume ratio available

                                2-19

-------
for sulfur elimination.   However,  if a low temperature is maintained
throughout the entire roasting period, sulfates of metal  would form,
since the heat of formation of sulfates is higher than that of the
corresponding oxides.19   This would decrease the elimination of sulfur.
To minimize the formation of sulfates, it is desirable to maintain a
higher temperature (up to 554°C or 1,000°F) at the end of the roast
than at the beginning.  The ideal  temperature condition would therefore
be low initially and increase gradually during roasting.
     Conditions for the removal of arsenic differ from those  of sulfur
elimination.  Maximum removal of arsenic  occurs at a low temperature
(500° to 600°C or 930 to 1,110°F) under reducing conditions  (when
sulfide concentrates are roasted).  A  higher temperature oxidizing
roast causes  the formation  of arsenic  pentoxide, which can  react with
calcium or  iron  to form stable and nonvolatile arsenates which  remain
in the calcine.  However,  an  oxidizing roast  is  required  for the
sulfur removal.   It  is therefore desirable  to  alternate  oxidation  and
reduction  several times  to  maximize  the volatilization  of  arsenic.
      Arsenic  elimination  in the multi-hearth  roaster tends  to be
greater  than  in  a fluidized-bed unit for  any  given  degree  of sulfur
elimination.   In a multi-hearth roaster,  it is possible  to  vary oxidizing
and  reducing  conditions,  temperature, and gas  composition on each
hearth.   The residence times of concentrate particles in a  multi-hearth
 roaster  generally range  from 1 to 2 hours.   The fluidized-bed roaster
can  provide only one set of conditions at a time,  either oxidizing or
 reducing.  The temperature remains constant throughout the bed, and
 there are a minimum of  hot spots  or uncontrolled variations  in condi-
 tions.   The residence time of concentrate particles in the fluidized-bed
 roaster is much smaller compared  to a multi-hearth roaster.
      Table 2-4 lists data from various sources on the amount of arsenic
 volatilized on a 100 percent input basis in multi-hearth and fluidized-bed
 roasters.  The elimination of arsenic  in the  roasting of copper ores
 and concentrates depends to a considerable extent upon the  type of
 roasting unit used.  The third data  point  indicates  that in  multi-hearth
 roasters,  as much as 70 to 90 percent arsenic can be driven  off when  a
 moderate  air consumption  rate is  used.   When  concentrates  are  roasted
                                  2-20

-------
               Table  2-4.   ARSENIC  ELIMINATION  IN  ROASTERS

1.
2.
3.
4.
5.
6.
7.
8.
M A"
Type of Roaster
Multi-hearth roaster9
Multi-hearth roaster
Multi-hearth roaster0
Multi-hearth roaster
Fluid-bed roasterd
Fluid-bed roaster6
Fluid-bed roaster0
Fluid-bed roaster
Percent
Arsenic
in Feed
3.8
.22
N.A.
0.01 - 0.15
.20
0.99
N.A.
0.02
Percent of Arsenic in
Feed Volatilized
25
27
70 - 90
5-30
4-10
60 - 65
15 - 50
15
 Information received on the ASARCO-Tacoma smelter from Mr. K.W.
 Nelson, ASARCO, Incorporated, March 17, 1976.

 Information on ASARCO-E1 Paso obtained from EPA testing (refer to
 Appendix C).

 Rozlovskii, A.A.,  "Behavior of Arsenic in the Production of Nonferrous
 Metals," Tsvetenye Metally/Nonferrous Metals, NO. UDC 669.778.

 Stankovic, D., "Air Pollution caused by Copper Metallurgy Assemblies in
 Bor, '  Institute for Copper, Bor, Project No. 02-513-1, U.S. EPA.
p
 Information on Anaconda obtained from Mr. Richard Sloane, Director of
 Technology, Anaconda Company, October 11, 1978.

 Information on Kennecott-Hayden received from Mr. 1.6. Pickering, Vice
 President, Environmental Affairs,  Kennecott Copper Corporation, May 9, 1978.
                                2-21

-------
in fluidized-bed roasters (seventh data point) with a great excess of
air, as little as 15 to 50 percent of the arsenic is eliminated.  The
fourth and fifth data points also indicate that arsenic elimination in
multi-hearth roasters (5 to 30 percent) is greater than in fluidized-bed
roasters (4 to 10 percent).  However, the amount of arsenic volatilized
in dependent upon the amount of roasting done or allowable, since
ASARCO-Tacoma using a "high" arsenic feed eliminates only 25 percent
of the arsenic in the feed, although more arsenic elimination is
possible.  This, as ASARCO has pointed out, is due to the production
of low grade matte which is essential for the removal of impurities
                                20
during the converting operation.    A relatively low grade matte leads
to higher converter temperatures which are essential for the vaporization
of impurities in the converter.  Also, low grade matte tends to increase
the blowing time or "sweep" in the converter which facilitates impurity
elimination.  Conversely, in the fluidized-bed roaster at the formerly
operated Anaconda smelter (sixth data point in Table 2-4), between 60
and 65 percent of input arsenic was volatilized.  This was due to the
longer roast performed at Anaconda which also eliminated a large
amount of sulfur in the feed.  Consequently, a high grade matte (50 percent)
is produced in the electric furnace.
2.2.2  Arsenic Behavior in Smelting Furnaces
     Arsenic elimination in smelting furnaces is by volatilization and
slagging.  Arsenic behavior in the smelting furnace is controlled by
the fact that smelting occurs in the presence of copper.  As has long
                                                       21 22
been known, and confirmed by recent thermodynamic data,  '   arsenic
is more stable in copper than in Cu-S, with a decrease in its thermo-
dynamic activity and, hence, volatility.  As a result, in the smelting
furnace the volatility of arsenic will change with the grade of matte
being produced.
     Since volatilization plays a major role in the elimination of As
during the production of matte in smelting furnaces, any factors which
enhance volatilization will further improve its elimination.  These
are:
     •    Mineralization:  some As compounds have higher vapor pressures
          than others and will therefore volatilize to a greater
          degree.

                                2-22

-------
       •    Green charge or calcine charge:  a calcine charge will have
            the more-easily volatilized As compounds removed in roasting;
            thus, for a given original mineralization, a green-charged
            furnace should volatilize a greater proportion of the As.
       •    Smelting temperature:  the higher the temperature, the
            greater the degree of volatilization.
       •    Exposure of slag  surface free to charge material  cover:  the
            greater the exposure, the greater the  volatilization of As
            from  the slag.
       The amount of arsenic  slagged or volatilized differs from one
  furnace  to another.   The  behavior  of arsenic  in  the different  types of
  furnaces is  described  in  the following  sections.
       2'2'2-1  Reverberatorv  Furnarps.   The  arsenic  input  into  the
  reverberatory furnace  is  from the  arsenic  in  the  new charge, slag
  returned  from the  converters, and  recycled  dusts  and reverts.  This
  arsenic  is eliminated  by  either volatilization or slagging.  The
  amount of arsenic  slagged or volatilized differs  from one smelter to
  another, as indicated  in Table 2-5.
      Data presented in Table 2-5 have been  organized in two groups
 Group A represents those furnaces which have a concentration of arsenic
 of greater than  0.2 percent  in the feed.  Group B represents smelters
 having a  smelter feed with less  than 0.2 percent  arsenic content.  All
 percentages  presented in Table 2-5  are on a 100 percent input basis in
 the furnace.
      Examining the data in Group A  from  ASARCO-Tacoma,23 ASARCO-El
 Paso,   Phelps Dodge-Ajo,25  and  Anaconda26  (these  data  are for  the old
 Anaconda  smelting  configuration  having a green reverberatory furnace)
 the amount of arsenic  volatilized and leaving  the  gas phase  ranges
 from 55 to 75 percent.   However, all  of  these  smelters  have  greater
 than 0.2  percent arsenic in the  feed.  The ASARCO-Tacoma and ASARCO-El
 Paso smelters feed  roasted calcine  into  their  reverberatory  furnaces
 but still  have high volatilization.   This is because only 25 percent
 of  the arsenic is volatilized in the  roaster at the ASARCO smelters
 The charge to the reverberatory furnace is therefore still  high in
arsenic content.
                                2-23

-------
    Table 2-5.  ELIMINATION OF ARSENIC IN REVERBERATORY FURNACES
                      (100 percent input basis)


Type of Feed
Calcine? Group A
Calcine
Green^
Green
Green
Calcinef
Calcine^ Group B
Calcine^
r n
Green •
Calcine
GreenJ
Percent
Arsenic
in Feed
3.8
0.22
0.3
0.81
0-2
0-2
0.02
0.01-0.2
0.13
N.A.
0.0035

Percent Arsenic
Volatilization
69
62
55
73
70-90
15-70
37
5-30
29
12


Eliminated by
Slagging
13
20
25
10
5-15
26-60
16
35-55
26
54
51
N.A. - Not Available
NOTE:  The percent arsenic content in calcine feed to the smelting
       furnace is not known.   The concentrations listed above are for
       the feed into the roaster and do not represent arsenic concentration
       in the calcine.
alnformation received on ASARCO-Tacoma from Mr. K.W. Nelson of ASARCO,
 Incorporated, March 17, 1976.
bInformation obtained from testing at ASARCO-E1 Paso (refer to
 Appendix C).
cSchwitzgebel, K., et al., "Trace Element Study at a Primary Copper
 Smelter," Prepublication copy, EPA Contract Number 68-01-4136,
 January  1978.
d"The Proposed Arsenic  Standard:  Feasibility  and Estimated Costs  of
 Compliance  for  Three U.S. Copper Smelters," prepared  by  D.B. Associates
 for OSHA Contract B-9-F-5-1663, March  25,  1976.
eRozlovskii, A.A.,  "Behavior  of Arsenic  in  the Production of Nonferrous
 Metals," Tsvetnye Metally/Nonferrous  Metals No.  UDC 669.778.
Information on  Kennecott-Hayden  received  from Mr.  1.6.  Pickering,
 Vice President, Environmental  Affairs,  Kennecott Copper Corporation,
 May 9,  1978.
9Stankovic,  D.,  "Air Pollution  Caused  by Copper Metallurgy Assemblies
  in Bor," Institute  for Copper,  Bor,  Project No.  02-513-1, U.S.  EPA.
 hlnformation received from Mr.  Hank Hansen of  Kennecott Copper Corporation
  on the old  Kennecott-Utah smelter.
 fuddle, R.W., "The Physical  Chemistry of Copper Smelting," Institute
  of Mining  and Metallurgy, London,  1953.
 JTurnbull,  D.L., "Converter Practice at Mufulira," Seventh Commonwealth
  Mining and Metallurgical Congress, 1961.
                                 2-24

-------
        The  data  obtained  in  Group  B  from  Kennecott-Hayden,27  Bor-
   Yugoslavia,    Kennecott-Utah29 (for the  old  smelter  configuration
   utilizing  the  reverberatory  furnace), and  the  infonnation obtained  by
   Ruddle    indicate that  the amount  of arsenic volatilized and  leaving
   in the gas phase ranges from 5 to  27 percent.  Also, the total elimination
   in the furnace ranges from 40 to 60 percent  of the input arsenic
  This is reaffirmed by the Mufulira data.31   However, the aforementioned
  smelters have less than 0.2 percent arsenic  in the feed.
       This  leads to two basic  observations for the reverberatory furnace:
       1.    With  greater than 0.2 percent  arsenic in the feed, 55 to
  75 percent of the arsenic  generally leaves in the gas phase and 10 to
  25 percent is slagged out.
       2.    With  less  than 0.2  percent  arsenic  in the feed,  5  to 37 percent
  of the arsenic  generally leaves in  the gas  phase  and  16 to  55 percent
  is slagged out.
       The aforementioned  observations are  consistent with those reported
  in the U.S.S.R.    (Group A) where the amount  of arsenic  volatilized
 was lower  for roasted  feed  (15 to 70 percent  volatilized) since it has
 low-arsenic content, as compared to unroasted feed  (70 to 90  percent
 volatilized) which has higher arsenic content.  The amount of  arsenic
 eliminated  by slagging ranged  from 25 to 60 percent for  roasted feed
 and 5 to 15 percent for unroasted  feed.
      2-2.2.2  Electric Furnaces.   Table  2-6 summarizes the available
 information on arsenic elimination in  electric furnaces.  Paulson et
 al.,   used calcine from the ASARCO-Tacoma smelter and ran tests on a
 laboratory  electric furnace.   They found  the behavior of arsenic to be
 similar to  that  in the reverberatory furnace.   However,  some  questions
 remain regarding the  manner  in which these smelting tests were performed
 In  actual practice, copper smelting  electric furnaces  are operated
with a cold top  -  that  is, a layer of solid  charge  is  maintained over
the entire  surface of  the melt at all times.   Paulson  et  al.,  carried
out  their tests with the molten slag surface uncovered by the  charge
material.    In the  latter case,  the slag is hotter,  and all the  arsenic
vaporized  from it will  leave with the flue gases rather than be  recondensed
when passing through the relatively cold  charge layer.
                                2-25

-------
      Table 2-6.  ELIMINATION OF ARSENIC IN ELECTRIC FURNACES
                      (100 percent input basis)
Type of
Furnace
Electric3
Electric
Electric0
Type of
Feed
Calcine
Calcine
Calcine
Percent
Arsenic
in Feed
3.8
0.99
1.8
Percent Arsenic
Volatilization
38.3-68.2
8
24
Eliminated by
Slagging
1.5-11.2
78
50.5
NOTE-  The percentage of arsenic content in calcine feed to the smelting
       furnace is not known.   The concentrations listed above are for
       the feed into the roaster and do not represent arsenic concentration
       in the calcine.

aPaulson, D.L., et al., "Smelting of Arseniferous Copper Concentrate
 in an Electric Arc Furnace," United States Bureau of Nines Report of
 Investigation 8144, 1976.
blnformation received on the Anaconda smelter from Mr.  Richard Sloan,
 Director of Technology, Anaconda Company, October 11,  1978.
c"Economic  Impact of New Source  Performance Standards  on the  Primary
 Copper  Industry:  An Assessment,"  Final Report to U.S. Environmental
 Protection Agency,  EPA Contract No. 68-02-1349,  October 1974.
                                  2-26

-------
       2.2.2.3  Flash Furnaces - The Noranda reactor flash smelting
  furnace has been introduced by Kennecott at their Utah smelter.  In
  this process,  copper concentrates and flux are fed into a horizontal
  cylindrical reactor where smelting and converting occurs under the
  dynamic action of oxygen-enriched air introduced through tuyeres below
  the bath.   Heat for smelting is  provided by exothermic converting
  reactions  supplemented  by burners.  As indicated by Mackey  et al.,34
  the elimination  of  As  in  the Noranda process  is  more favorable at'
  lower matte grades.  The  amount  of arsenic volatilized is much greater
  under Noranda  matte-making  conditions  compared  to arsenic volatilized
  under regular  copper-production  conditions.   Also,  the amount of
  arsenic  reporting in the  matte is  much  lower.
       Table  2-7 shows that the amount of  arsenic  volatilized  from four
  different flash furnaces  is  near  76 to 85  percent,  and  the amount
  slagged  ranges from 7 to  17  percent.
  2.2.3, Arsenic Behavior During Converting
       Arsenic distribution in a copper converter between gases  and slag
 varies widely,  as shown in Table 2-8.  When a high grade matte  is
 converted,  metallic copper appears early in the cycle and acts  as a
 collector for arsenic because arsenic is more stable and less volatile
 in metallic copper than in copper sulfide.   Depending upon the matte
 grade, as much  as 70 percent of the arsenic may report in the blister
 copper,  and only 30 percent  is eliminated by volatilization  or slagging.35
 When a low  grade matte  is  converted,  the formation of metallic copper
 is  delayed,  and a much  larger proportion of the arsenic is eliminated.
 Up  to 92 percent   »   of the arsenic  will  be volatilized in  such
 cases.  In  normal  practice,  it is  neither that low nor that  high;
 approximately 70  percent of  the arsenic  is  volatilized in  the converter,
 and  nearly  16 percent reports in  the  slag.
      In  some  domestic copper  smelters such  as  ASARCO-E1  Paso,  it is
 common to recycle  low grade  secondary materials and  copper precipitates.
 If materials  to be smelted have high  impurity  content  (particularly
 lead speiss or other recycled materials), a  low grade matte is  necessary
 to produce sufficient heat for elimination of  these materials.   Decreasing
matte grade  allows time for the volatilization of  impurities  since the
slagging  cycle increases with lower matte grade.  Also, the longer
                                2-27

-------
     Table  2-7.   ELIMINATION OF ARSENIC  IN  FLASH  FURNACES
                       (100 percent  input basis)
Type of
Furnace
Noranda
Reactor
"KCS"b
Process
Noranda
Reactor
Otokumpu
Type of
Feed
Green
Green
Green
Green
Percent
Arsenic
in Feed
0.06
N.A.
0.14
0.17
Percent Arsenic
Volatilization
85
79
82
76
Eliminated by
Slagging
7
11
13
17
N.A. - Not Available

ariackey, P.O., et al., "Minor Elements in the Noranda Process," Paper
 presented at the 104th Annual  AIME Meeting, New York, February 16-20,
 1975.
bGeorge, D.B., et al., "Minor Element Behavior in Copper Smelting and
 Converting," Joint AIHE-MHIJ Conference, 1976.
Information on Kennecott-Garf i eld received from Mr.  I.G. Pickering,
 Vice President, Environmental  Affairs, Kennecott Copper Corporation,
 May 9, 1978.
dKoh, Shiroh, "Extractive Metallurgy of Kuroko," Paper presented at
 the joint meeting of the MMIJ-AIME, Tokyo, May 24-27, 1972.
                                 2-28

-------
          Table 2-8.  ELIMINATION OF ARSENIC DURING CONVERTING
                        (100 percent input basis)

Type of Matte
Producing Furnace
Reverberatory?
Reverberatory
Reverberatory *j
Reverberatory
Reverberatory?
Reverberatory
Outokumpu Flash9
KCS Smelting
Converter
Shaft Furnace1
.A. - Not Available
Percent
Cu in
Matte
N.A.
40
40
35-55
N.A.
35-55
45

70
N.A.


Percent Arsenic
Volatilization
73.0
92.0
92.0
76.0
50-85
55.0
43 0

73 0
70.2


Eliminated by;
Slagging
11.0
n
u
c n
o. u
19.0
10-35
23.0
90 n
<-.y . u

29.5

  Ruddle  R.W.   "The Physical Chemistry of Copper Smelting,"  Institute
  of Mining and Metallurgy, London, 1953.                    institute

 bljS™at!°n  received on ASARCO-Tacoma from Mr. K.W. Nelson,
  ASARCO,  Incorporated, March 17, 1976.

 CInformation  obtained from tests performed at ASARCO-E1 Paso (refer
  to Appendix  C).


         "i    t "J1 r/0"ut1on Caused by C°PP^ Metallurgy Assemblies
       ,  Institute for Copper, Bor,  Project No. 02-513-1, U.S. EPA.

 eRozlovskii, A. A., "Behavior of Arsenic in the Production of

  Nonferrous Metals, "Tsvetnye Metal ly/Nonferrous Metals, Number UDC
           Dt K«'.et a1" "Trace Ele^nt Study at a Primary Copper
 Smelter,  Prepublication copy,  EPA Contract Number 68-01-4136, January 1978.

 Koh, Shiroh, "Extractive Metallurgy of Kuroko," Paper presented at
 the joint meeting of the MMIO-AIME, Tokyo,  May 24-27,   "eritea at
 i-7 / L. •


 George, D.B., et al . ,  "Minor Element Behavior in  Copper Smelting

 and Converting," Joing  AIHE-MMIJ Conference,  1976.

'information  received from Technika,  Sofia,  Bulgaria,  September 24,  1978.
                               2-29

-------
converting time allows a greater duration of gas sweep across the bath
for impurity removal.   Hence, the concentration of matte grade governs
the recycle of intermediates in the converter.
2.2.4  Arsenic Balance
     There is a wide variation in the distribution of arsenic among
the various products of a particular process from one low-arsenic
throughput smelter to another.  This is due not only to the  difference
in feed composition but also to factors such as temperature, blowing
rates, gas composition, and  the analysis  and  relative quantities  of
products.  The data presented  in Tables 2-5,  2-6, 2-7,  and 2-8  are
summarized  in  Table 2-9 on  a 100 percent  input basis  through the
smelting  circuit.  Table  2-9 shows  that  the wide  variation  in arsenic
elimination  exists  not  only for  different smelter configurations  but
also  for  smelters  having  similar circuits.  Detailed  causes  of  these
variations  are complex,  and relatively little information can be  found
 in the literature.  There are, however,  relationships between process
 parameters  that affect arsenic and other minor element distributions
 that can  be observed  from the data in Table 2-9.
      The distribution of arsenic within the various smelting processes
 is specifically controlled by the characteristics of the arsenic and
 its environment.  There appear to be two fairly distinct input levels
 of arsenic which result in  different distributions.  Refering  to
 Table 2-9, it can be seen  that for smelting processes with  levels of
 input arsenic greater than  or equal to 0.2 or 0.3 percent,  the resulting
 arsenic  distribution appears  to be directly  influenced by temperature
 and  the  presence of  copper.   The copper  tends to fix the arsenic.   The
 partial  pressure of  the  arsenic trioxide is  high  enough  to  allow
 volatilization  of  the  arsenic through  the  slag layer in  the smelting
 furnace.  This  results  in  a lower proportion of  the  input  arsenic
 leaving  with  the  slag.   However,  when the  level  of  input arsenic is
  less than  0.2 percent,  the data in Table 2-9 indicate that a higher
  proportion of the input arsenic is transferred to the slag.  The
  hypothesis is that in the case of lower arsenic  input, the arsenic is
  mechanically bound or locked into the feed,   and  therefore, not readily
  released.   This prevents it  from being volatilized and tends  to allow
  it to enter the slag.

                                  2-30

-------
I
co
                                                    Table 2-9.   ARSENIC  BALANCE -


                            WEIGHT  PERCENT AS  REPORTING IN SMELTER PRODUCTS (AMOUNT INPUT = 100)

PROCESSES



Reverberatory - Converters3
Reverberatory - Converters'3
Reverberatory - Converters0
Reverberatory - Converters'1
Reverberatory - Converters6
Multi-hearth - Reverb - Converters f
Multi-hearth - Reverb - Converters9
Multi-hearth - Reverb - Converters'1
Fluidized-Bed - Reverb - Converters'1
Fluidized-Bed - Reverb - Converters'
Fluidized-Bed - Electric - Converters1^
Multi-hearth - Electric
Multi-hearth - Electric - Converters'

Noranda Reactor - Converters'"
Noranda Reactor - Converters"
Fierce-Smith Reactor - Converters0
(KCS process)
Outokumpu Furnace - Converters'3
Input in
smel ter
feed
Wt. % As
in
charge
0.3
0.81
0.13
0.0035
N.A.
3.8
0.22
0.01-0.15
0.2
0.02
0.99
3.8
1 8

0.06
0.14

N.A.
0.17
Roasting

% As
volatilized
N.R.
N.R.
N.R.
N.R.
25.0
27.0
5-30
4-10
15
63.3
25.0
81

N.R.
N.R.

N.R.
N.R.
Smelting

% As
slagged
25.0
9.7
26

54.2
10.0
15.0
30-45
35-40
13
28.9
1.13-8.4


7.0
13.0

11.0
17.0

% As
volatilized
55.0
73.3
29
51
11.8
52.0
45.0
5-15
25-30
31
2.7
28.73-61.15

.0
85.0
82.0

79.0
76.0

% Cu
in matte
35-45
40.0
40.0
N.A.
N.A.
40 0
40.0
35-55
39-43
40
50
40

36
70.0
70

70.0
45.0
Converting

% As


% As
slagged volatilized
4.5
3.4
9

3.75
o
0.6
5-10
0-10
12.1
1 4
Experimental
11.0
11.3
AQ
24.8
1 7 n
11.5
20-40
15-20
28.7
^ fi
study done
	 .
Blister

% As
in copper
4.5
2.3

5.45
.u
0.9
1.5-3
1-2
0.2
. 1
on a pilot plant
1 0
2
o

o
2.0
1 9
A
O c

7 7
3.0

. C

.5

. /
2.0
       N.A. -  Not Available


       N.R. -  No Roasting


       *The input into the furnace is a calcine feed.

-------
Footnotes for Table 2-9.
 Schwitzgebel,  K.,  et al.,  "Trace Element Study at a Primary Copper
 Smelter," Prepublication  copy,  EPA Contract No. 68-01-4136, January
 1978.

 Burton, D.J.,  "The  Proposed Arsenic Standard:  Feasibility and Estimated
 Costs  of Compliance for Three U.S. Copper Smelters," prepared for
 OSHA,  Washington,  D.C., Contract B-9-F-5-1663, March 25, 1975.

Conversation  with  Mr.  Hank Hansen, Environmental  Engineer, Kennecott
 Copper Corporation, Salt Lake City, Utah, September 26, 1978.

 Herneryd, 0.,  et al.,  "Copper Smelting in Boliden's Ronnskar Works
 Described."  Journal of Metals, March 1954.

eRuddle, R.W.,  "The Physical Chemistry of Copper-Smelting," Institute
 of Mining and Metallurgy,  Lond, 1953.

 Correspondence from Mr.  K.W. Nelson, ASARCO, Incorporated to
 Mr. J. Padgett, Environmental Protection Agency,  March 17, 1976.

^Harris, D.L.,  "Particulate and Arsenic Emission Measurements From a
 Copper Smelter," Volume I  (text), Monsanto Research Corporation, EMB
 Project Report No.  77-CUS-6, EPA Contract No. 68-02-1404, Task No. 36,
 1977.

 Stankovic, D., "Air Pollution Caused by Copper Metallurgy Assemblies
 in Bor," Institute for Copper, Bor, Project No. 02-513-1, U.S. EPA.

 Data for Kennecott-Hayden received in the correspondence from
 Mr. I.G. Pickering, Vice President Environmental  Affairs, Kennecott
 Copper Corporation, to Mr. D.R. Goodwin, Director, ESED, U.S. EPA,
 May 9, 1978.

 Information received on the Anaconda smelter fron fir. Richard Sloan,
 Director of Technology, Anaconda Company, October 11, 1978.

L.
 Paulson, D.L.  et al., "Smelting of Arseniferous Copper Concentration
 in an Electric-Arc Furnace," United States Bureau of Mines Report of
 Investigations 8144, 1976.

 "Economic Impact of New Source Performance Standards on the Primary
 Copper Industry:  An Assessment," Final Report to U.S. Environmental
 Protection Agency, EPA Contract No. 68-02-1349, October 1974.

"Vlackey, P.J., McKerrow, G.C., Terassoff, P.,  "Minor Elements in the
 Noranda Process," Paper presented at the 104th Annual AIME meeting,
 New York, February  16th - 20th, 1975.

nData for Kennecott-Hayden received in the correspondence from
 Mr. I.G. Pickering, Vice President Environmental  Affairs, Kennecott
 Copper Corporation, to Mr. D.R. Goodwin, Director, ESED, U.S. EPA,
 May 9, 1978.

°George, D.B., Donaldson, J.W., Johnson, R.E.,  "Minor Element Behavior
 in Copper Smelting and Converting," 1976 Joint AIME-MMIJ Conference.

pKoh, S., "Extractive Metallurgy of Kuroko," Joint Meeting MMIJ-AIME,
 Tokyo, May 1972.
                                2-32

-------
       While  it has been theorized by  some that multi-hearth  roasters  are
  required to eliminate arsenic, this  is not necessarily confirmed by
  the available information.  In general, with the combined processes  of
  the roaster and reverberatory furnaces, it appears that in  the case  of
  the multi-hearth roaster, a larger amount of arsenic is volatilized
  than in the reverberatory furnace.   In the case of the fluidized-bed
  reverberatory furnace combination,  less arsenic is volatilized in the
  fluidized-bed than in the reverberatory furnace.   However, if the sum
  of the  roaster and furnace arsenic  volatilization is  considered,  then
  the total  seems  to be the same with  either  type of roaster (refer to
  eighth  and  ninth data points  in Table 2-9).
      Arsenic distributions at domestic copper  smelters  were  determined
 by using  (1) arsenic mass balances supplied by  the  smelters,  (2)  the
 previously cited arsenic distribution information for foreign  and
 domestic copper smelters available in the literature, and  (3)  control
 equipment performance data obtained by EPA testing  at various  U.S.
 copper smelters.  The estimated arsenic distributions for each domestic
 copper smelter are presented in Appendix F.  Complete recycle  of the
 collected flue dusts from the existing control  devices at each smelter
 was assumed  in these balances.   The arsenic removal  efficiency of
 existing  control  devices was  taken into  account in determining the
 arsenic material  balances for 10 of the  14 low-arsenic throughput
 smelters.  For the  remaining  four smelters,  Phelps Dodge-Ajo, -Morenci,
 ASARCO-Hayden,  and  Kennecott-Hurley, material balances were generated
 for the smelter configurations and  control  equipment arrays  projected
 after completion of  modernization  programs.  The modernization  programs
 for these  smelters are discussed  in Section 4.2.1.
     Arsenic emissions from primary copper smelters  can  be  categorized
 as process and  fugitive  emissions.  Process emission  sources  include
 primary offgas  emissions from roasting, smelting, and converting
 operations.  Fugitive emissions include those emissions which escape
 from process  flow streams due to leakage through process exhaust
systems, material  transfer systems, or ineffective capture at the
source of  generation.
                                2-33

-------
2.3.1  Process Arsenic  Emissions
     Information indicating the distribution of arsenic in the smelter
circuit was received from most domestic copper smelters.  Based on
this information, arsenic mass balances for each smelter were performed.
Typically, the information obtained consisted of process material
balances and estimates  or measurements of the arsenic content in these
materials.  However, the information received was incomplete for a
number of copper smelters.  In these cases, assumptions were made
regarding the behavior of arsenic based on the information presented
in Section 2.2 on the behavior of arsenic at copper smelters.
     For  those smelters for which information on arsenic  distribution
was  not obtained or the information provided was incomplete, an  arsenic
distribution  for the smelter  was developed  based on the type of  smelting
configuration  at that particular smelter,  the  level of  arsenic  input
to  that smelter, and the  availability  of arsenic distribution  information
from smelters  having similar  configurations.   For  example,  lack  of
complete  arsenic distribution data  for the  projected  configuration of
ASARCO-Hayden after installation of an INCO-flash  furnace necessitated
use of an arsenic  balance provided  by  Kennecott for the projected
 arsenic distribution  at the Hurley smelter after modification to INCO
 smelting  technology.
      The control  equipment efficiencies used in the arsenic mass
 balances  were based on tests performed at several  existing units.
 Tests performed on baghouses and electrostatic precipitators have
 indicated that the collection efficiency of the control  equipment for
 arsenic  is dependent upon the inlet temperature of the offgases and the
 arsenic  concentration of the gas stream, and that maximum collection
 of  arsenic trioxide occurs at reduced temperatures.  As  a result, the
 collection efficiency can vary from 30 to  99.0 percent.  The results
 obtained are  summarized  in Table 2-10.  Where  company-supplied  data
 were  not available, an arsenic removal efficiency  of  99  percent was
 ascribed to  acid  plants,  due to the gas precleaning  and  conditioning
 required for  effective acid  plant  operation.   Detailed test  data  are
 available in  Appendix  C,  and a discussion  on  the  performance  of these
 control  devices is presented in Section  3.2.1.
                                  2-34

-------
Table 2-10.  MEASURED ARSENIC COLLECTION EFFICIENCIES
                 OF CONTROL DEVICES9
Sine! ter

Phelps Dodge-Ajo
ASARCO-E1 Paso
Anaconda
Kennecott-Hayden
ASARCO-Tacoma
Detailed test data
Device

Reverb ESP
Roaster & Reverb -
Spray Chamber/ESP
Spray Chamber/Baghouse
Venturi Scrubber
Roaster Baghouse
available in Appendix C.
Operating
Temperature
(°C)
Inlet
327
212
263
329
91

Outlet
313
104
101
39
86

Collection
Efficiency
Percent

27.8
96.2
98.9
98.3
99.7

                     2-35

-------
     The arsenic distributions for all  domestic copper smelters are
given in Appendix F.   These distributions are based on a full recycle
of flue dusts and converter slag (except at Kennecott-Hayden where
converter slag is not recycled).  In most cases, the final arsenic
distributions have been obtained after performing iterative  recycles
until steady-state conditions are reached.  Steady-state conditions
are  indicated by stabilizing the arsenic content of the blister copper
and  the converter slag.
     The estimated potential process arsenic emissions  in  the  absence
of control for  each of the domestic low-arsenic throughput copper
smelters are  summarized in Table 2-11.
2.3.2   Fugitive Arsenic Emissions
     Fugitive emissions may  be  characterized as emissions  which  escape
directly from the process  area  to the  atmosphere  rather than through a
flue or exhaust system.  They  result  from  leakage  in  and  around  process
equipment  and from material  handling  and transfer  operations.   These
emissions  may be considered  as  low  level emissions  as compared to
process emissions, since  they usually  leave the smelter at or near
ground level, whereas process emissions are discharged through a tall
stack.
      Listed  in  Table 2-12  and shown in Figure  2-6 are potential   sources
 of fugitive  arsenic  emissions.   These emissions depend upon the  particular
 types  of equipment  and operating practices employed by the  copper
 smelter.
      2.3.2.1   Roaster.
      2.3.2.1.1   Charging.   Fugitive emissions during charging of
 multi-hearth roasters seldom occur because of the water content  (8  to
 10  percent)   in  the feed, and the "choke feed" mechanism used  on  the
 charging hoppers.  Fugitive emissions during  charging  are seldom
 emitted from a properly designed and  operated fluidized-bed roaster
 because of the enclosed feed and discharge system.
       2.3.2.1.2 Leakage.  Fugitive arsenic emissions  from multi-hearth
 roasters may be emitted from leaks that can occur  at  the  doors  located
 at  each one  of the  hearth levels,  from  holes  in  the  actual  shell of
 the roaster, or from leaks  around  the shaft that  holds the  rabble
 arms.   Under normal  operating  conditions,  these  emissions are minimized

                                  2-36

-------
ARMrnciTnn       °F PROCESS ARSENIC EMISSION  ESTIMATES  IN
ABSENCE OF CONTROL FOR LOW-ARSENIC THROUGHPUT PRIMARY  COPPER  SMELTERS

Smel ter
ASARCO-E1 Paso


ASARCO-Hayden

Tennessee Chemical Co. -
Copperhill

Inspiration-
Miami
Kennecott-Garfield

Kennecott-Hayden


Kennecott-Hurley

Kennecott-McGill

Magma- San Manuel

Phelps Oodge-Ajo

Phelps Dodge-
Douglas


Phelps Dodge-
Hidalgo

Phelps Dodge-
Morenci

Copper Range -
White Pine
Sources: MHR - Multi hearth Roaster
REV - Reverberatory Furnace
CONV - Converters
FF - Flash Furnace


Emission
Source
MHR
REV
CONV
FF
CONV
FBR
EF
CONV
EF
CONV
NOR
COW
FBR
REV
CONV
FF
CONV
REV
CONV
REV
CONV
OXREV
CONV

MHR
REV
CONV

FF
CONV

OXREV
CONV
REV
CONV
FBR
EF
NOR
OXREV

Exit Gas
Arsenic Content
kg/hr
86.6
54.9
76.2
157.0
45.0
0 ?
u • c.
0.4
0.5
4.5
1.5
115.0
6.0
1.9
5.3
5.0
1.3
0.4
18.6
35.6
1.8
0.5
45.3
2.3

2.2
0.9
3.2

13.7
0.9

4.3
5.3
0.4
0.2
- Fluid Bed Roaster
- Electric Furnace
- Noranda Reactor
- Oxygen-Sprinkle
Reverberatory Furnace
                                  2-37

-------
    Table 2-12.   POTENTIrtL SOURCES OF FUGITIVE
                 ARSENIC EMISSIONS
Roaster
     Charging
     Leakage
     Hot calcine discharge and transfer
Smelting Furnace
     Charging
     Leakage
     Matte tapping
     Slag tapping
     Converter slag return
Converters
     Charging  (matte, reverts, flux, lead  smelter  by-products,  cold
              dope)
     Blowing  (primary hood leaks)
     Skimming
     Holding
     Pouring  of  slag and  blister
     Converter  leaks
Anode  Furnace
     Charging
     Blowing
     Holding
     Pouring
Miscellaneous
     Dust handling and  transfer
     Ladles (matte and  slag)
      Slag dumping
                       2-38

-------
ro
i
UD
                                                             <•(. II 5IURAGC
                                                                               UNLOADING I
                                                                               COMIC HANOIING
                                              ROASTEH
                                           / CHARGING
                   "OAJffH LEAKAGE
                                                                                            ANODE
                                                                                            FURNACE
                                                                                              CHAM! Of
                                                                                              10 ANODE FURNACE
                                                                                              » SLAG
                                                                                              » HANOI INS
          LIMESTONE
          UNLOADING l—k'
&
                                                                                                             COffEU
                                                                                                             » CASriNG
                                                                                                            10 vnnm
                   REVERBERATORY  FURNACE
                       Figure 2-6.   Fugitive Emission Sources  at Primary Copper Smelters

-------
by operating the multi-hearth r asters under a slight negative pressure
required to supply induced air for oxidation.
     A fluidized-bed roaster is essentially a vertical cylinder of
steel plate lined with insulation and fire bricks.  This type of
roaster is fed by and discharges into closed systems.  Therefore,
leakage from fluidized-bed roasters is negligible.
     2.3.2.1.3  Hot calcine discharge and transfer.   Fugitive emissions
may  be generated during discharging and transfer  of hot calcine from
roaster to smelting furnaces.  Smelters with multi-hearth  roasters
usually use larry cars (small rail cars) to  transport calcines to the
furnace.  When  the material  is dropped from  the calcine hopper located
under the roaster into the covered car through a  feed opening, large
quantities  of dust  are generated  as  a result of material movement and
pressure changes within the  car.  Some fugitive emissions  can also
occur during  the  transportation  of the roaster calcines  to the smelting
furnace.   In  the  case where  larry cars are  used,  their  feed opening is
usually covered to  minimize  this  effect.
      Hot calcine  discharge  from  fluidized-bed roasters  primarily
occurs  by  entrainment of  the calcine in  the gas  going from the  top  of
 the roasters  into a series  of cyclones.   The material drops into the
 cyclone hoppers.   Generally, about  80 to 85 percent of  the calcine
 passes  through  the  top  of the roasters.   The "underflow" or remaining
 calcine flows through an  opening near the base of the roaster.   The
 roaster and cyclones are  operated under positive pressure and,  therefore,
 must be airtight and free of leaks.   Consequently, fugitive emissions
 are seldom emitted  from a properly operated fluid-bed roaster during
 calcine transfer operations since it is essentially  a closed system.
                                                   42                43
      Tests were performed by EPA at ASARCO-Tacoma,   ASARCO-E1  Paso,
 and Phelps Dodge-Douglas44 to obtain fugitive arsenic emission estimates
 from the calcine transfer systems used at these  smelters.  The results
 of  these tests are presented in Table 2-13.
      The calcine transfer systems in the aforementioned smelters
 differ in  design.  At the ASARCO-Tacoma smelter,  an  enclosed spring-sealed,
 apron-type unloading system  is used.  The  transfer  of  calcine in this
 system  is  completely enclosed,  and  a minimum of  fugitives are emitted.
 The calcine  transfer systems  used at the ASARCO-E1  Paso and  Phelps

                                  2-40

-------
fSD
I
             Table 2-13.   FUGITIVE  ARSENIC  EMISSIONS DURING CALCINE TRANSFER  FROM  MULTIHEARTH ROASTERS
Smelter
ASARCO-Tacoma
ASARCO-E1 Paso
Phelps Dodge-
Douglas
Sample
Run
1
2
3
Avg.
1
2
3
Avg.
1
2
3
Avg.
Arsenic
in Calcine
(Ib/hr) kg/hr
719
719
1,073
130
130
130
10.5
1.0
2.3
326
326
987
59
59
59
4.8
0.5
1.0

English
ppm
69.7
109.3
153.6
110.9
3.7
1.03
2.5
2.41
0.0
0.0
0.0
0.0
Arsenic
Units
Ib/hr
1.0
1.7
2.7
1.8
0.33
0.09
0.22
0.22
0.007
0.026
0.064
0.032
Emissions
Metric
Mg/m3
217.3
340.9
479.0
345.7
11.53
3.23
7.81
7.52
0.059
0.22
0.642
0.307

Units
kg/hr
0.46
Q.lf
1.24
0.82
0.15
0.04
0.10
0.10
0.003
0.011
0.029
0.014
Comments
Holding aprons on hopper exit and
larry car. Probable 100 percent
collection efficiency. Intermit-
tent testing on one set or roasters.
Probable 50 percent collection
efficiency. Testing was done on a
continuous basis.
Probable overall collection
efficiency 70 percent. Testing on a
continuous basis. Calcine transfer
8-hour shifts.

-------
Dodge-Douglas smelters are, howe^r, not as effective.  A description
of these systems is provided in Section 3.2.2.2.
     Fugitive arsenic emission estimates were developed based on the
measurements presented in Table 2-13.  EPA testing of the calcine
transfer system at ASARCO-Tacoma was performed only when the system
was in operation.  At the ASARCO-E1 Paso and Phelps Dodge-Douglas
smelters, the systems were tested on a continuous time basis.  To
compare test results on a similar basis, the ASARCO-Tacoma test  results
were adjusted from an intermittent  to  a continuous basis.  This  was
performed by averaging the measured arsenic  emission  rate obtained
from intermittent  testing  over  the  total time  period  of  the  testing.
Also,  since  the test  results  at ASARCO-Tacoma  represented only  half
the  roasters typically operated at  any one time,  adjustments were also
made to  account for  the  other roasters,  in operation, but not  tested.
      For example,  sample run  number 1  at ASARCO-Tacoma indicated an
arsenic  emission rate of 0.46 kg/hr (1.0 Ib/hr).   Test data45  indicated
an actual  testing  time  of 15  minutes over  an elapsed time  period of
 1.76 hours.   The total  arsenic emitted during  the 15 minutes was
 therefore 0.12 kg, which was  equivalent to an  average emission  rate  of
 0.07 kg/hr (over a time period of 1.76 hours), assuming that the
 second row of roasters emitted the same amount of fugitive  arsenic
 emissions as the first set of roasters over this same time  period.
 The total uncontrolled fugitive arsenic emission rate during calcine
 transfer was 0.14 kg/hr.
       Corrections were also made to the ASARCO-E1 Paso and Phelps
 Dodge-Douglas  smelters' test results  to account  for  the collection
 efficiencies of the  calcine  transfer  systems  in  use  at  these smelters.
 The results thus  obtained  indicated the average  uncontrolled fugitive
 arsenic  emission  rates  for the ASARCO-Tacoma,  ASARCO-E1 Paso,  and
  Phelps  Dodge-Douglas smelters  to  be 0.16  kg/hr,  0.10 kg/hr, and
  0.014 kg/hr,  respectively.
       For the  purpose of developing an emission factor for  fugitive
  arsenic emissions occurring  during calcine transfer from multi-hearth
  roasters,  the fugitive  arsenic emissions  from the three smelters were
  compared to the quantity of  arsenic reporting in the calcine  during
  the test period.   The ASARCO-Tacoma,  ASARCO-E1 Paso, and  Phelps Dodge-Douglas

                                  2-42

-------
   test results  indicated  that  an  enrage  of  0.05,  0.33,  and 2.6 percent
   respectively,  of  the  arsenic  in  the  calcine  reports  in  the fugitives
   A correction  factor for  the  probable  collection  efficiencies  during
   calcine transfer  at the  three smelters  is  included in the aforementioned
   results.  However, there is  some question  regarding  the  accuracy  of
   the  arsenic emission  estimates made at  the Phelps Dodge-Douglas and
  ASARCO-E1  Paso smelters.  Calcine samples obtained at the Phelps
  Dodge-Douglas smelter indicated small amounts of arsenic  present  (less
  than 0.015 percent).   Accurate readings for small arsenic  contents are
  difficult  to obtain.   Transfer of calcine from roaster to  larry cars
  at the ASARCO-E1  Paso  smelter takes  place in a long,  rectangular shed.
  The  shed  is  open  at one  end  and  has  an exhaust duct at the other end
  The  configuration  of the system  is  such  that not only are fugitive
  emissions  captured by  the exhaust system during  calcine  transfer,  but
  also  reentrained fugitive dust generated by the  movement of the larry
  cars  and by  wind action.
       In view of the aforementioned, the  ASARCO-Tacoma smelter  results
  were  used  only to  compute the calcine  transfer fugitive  arsenic emissions
  factor.  The ASARCO-Tacoma test  results  indicated that an average  of
  0.005 kg of  arsenic reports in the fugitive emissions per kg of arsenic
  contained  in the calcine, or 0.05 percent.  This  factor was  used to
 estimate calcine transfer of fugitive arsenic  emissions from multi-hearth
 roasters at all domestic copper smelters.  The amount of  arsenic
 reporting  in  the calcine for these estimates was obtained  from  the
 process arsenic mass balances  provided in Appendix F.
      2-3-2-2   Smelting  Furnace.   As  noted previously,  three basic
 types  of smelting  furnaces are used.   These  include  reverberatory
 type,  electric,  and flash  furnaces.  A discussion on the  fugitive
 emission sources associated with  the operation of these furnaces
 follows.
     2.3.2.2.1  Charging.  Reverberatory  furnaces  can  be  charged by
 any of the  following methods:   (1) by  dropping  the calcine through  the
 arch of  the furnace, (2) by using  retractable  or  fixed Wagstaff guns
 located at  the furnace  side walls, (3) by charging through ports in
 the side walls which extend some distance toward the middle of  the
 furnace, and (4) through green charge slingers  that cover  the entire
bath in the  smelting zone.
                                2-43

-------
     In general, when a green o>" calcine charge is dropped into the
smelting furnace, a rapid increase in gas volume occurs because of
metallurgical reactions occurring within the furnace.  This can cause
a temporary excess pressure in the furnace depending upon the response
of the pressure control system.  If the pressure exceeds atmospheric
level, fugitive emissions can be generated through all the furnace
openings.  With side-charged reverberatory furnaces, there exists the
possibility of a portion of the charge bank flowing, caving in, or
sloughing  into the molten bath and creating a  rapid  reaction  between
the charge and the bath.  In such instances, generation  of gas  can
occur  so rapidly that  che furnace arch can be  seriously  damaged by  the
excessive  pressure in  the furnace.
     The seal  between  the retractable Wagstaff gun  and the  furnace
wall,  and  the  furnace  opening  required  for slinger  feed  type  charging,
 are  additional  openings  through  which fugitive emissions can  be emitted
 when  these types  of  charging  techniques  are  used.
      Electric  furnaces are  charged  continuously either from the roof
 or from the side.  The charge almost always  consists of dried concentrates
 or calcine since moisture  tends to  cause steam explosions.   Oxygen
 requirements for electric  furnaces  are  much  smaller compared to rever-
 beratory furnaces because  no fuel  is being directly burned.  There is
 Ucs likelihood of openings being present to allow excess air as is
 the case with the reverberatory furnace, thereby minimizing fugitive
 emissions  as a result of pressure surges.  When emissions do occur,
 they generally appear at the seals of the electrodes.
      Feed to flash furnaces is usually from a concentrate dryer where
 the concentrate-to-flux ratio is established  by computer, based on  the
 chemical  composition  of the materials and the desired matte  grade.
 Dried concentrates  from the dryer are discharged to the feed  receiver
 by a  pneumatic  air  lift.   Air  injected  into the  receiver lifts the
 dried  furnace  feed  to an air  expansion  vessel.   A  drag  conveyor  from
 the expansion  vessel  conveys  the feed  to the  storage  bin.  The feed
 from  the  storage  bin  to the  flash  furnace reaction shaft is  conveyed
 by  variable speed drag  conveyors.   A carefully controlled  mixture of
 dry  concentrate and pretreated  air  is  then  injected to the furnace
 through concentrate burners  from the  roof of  the reaction  shaft.   The

                                  2-44

-------
 charging systems for flash furnaces are gas-tight,  and  so  fugitive
 emissions are seldom emitted.
      Depending upon the method used for charging, and the  type  of
 smelting furnace, fugitive emissions can be generated.  However, by
 using add-on control systems which are activated at the time  the
 charge is dropped, effective reduction of potential fugitive  emissions
 can be obtained.  Since most smelters are currently using  such  control
 systems, it is reasonable to assume that fugitive arsenic  emissions
 during charging of smelting furnaces are negligible.
      2.3.2.2.2  Leakage.  The structure of most reverberatory furnaces
 has to allow for thermal expansion.  This leaves many leakage points
 in the reverberatory furnace outer shell.   Fugitive emissions from
 these leakage points are emitted  when the pressure in the  reverberatory
 furnace exceeds  atmospheric pressure.   This, as has been pointed out
 earlier,  can occur during  charging and  converter slag return.
      Proper  maintenance is a  major factor  in minimizing  leaks.  Hand
 or spray  sealing  of  all  cracks  between  refractories, if  practiced on a
 routine basis, considerably reduces fugitive emissions due to leakage
 in a  reverberatory  furnace.
      Usually,  reverberatory furnaces  are  operated  under  very slight
 negative  pressure at all domestic  copper smelters.   This is performed
 not only  to  prevent  fugitive  emissions  from  the multitude  of leakage
 points  around  the furnace  but to avoid  overheating  of the  furnace by
 providing the  cooling effect  of air infiltration.   Pressure controls
 are used  to  maintain a slight negative  pressure and  also to ensure
 that pressure surges are kept under control.
     In electric furnaces, there is much less of a  requirement for
 providing oxygen to the furnace as  compared  to  the fuel-fired  rever-
 beratory furnace.  Therefore, these furnaces are usually better  sealed.
 Thus,  if a positive pressure is momentarily generated, fugitive  emissions
 due to leakage are considerably less.
     The construction of flash furnaces is such that they are  virtually
gas-tight.  This  prevents leakage of emissions and cooling  by  air
 infiltration.
     It is reasonable to assume that fugitive arsenic emissions due to
leakage from  smelting furnaces are  negligible provided they are operated

                                2-45

-------
and maintained properly.   Based or this observation, it was concluded
that normally no fugitive arsenic emissions are discharged due to
leakage.
     2.3.2.2.3  Matte tapping.  Matte tapping is a principle fugitive
emission source at the smelting furnace.  Smelting furnaces have from
one to three matte tap holes with associated launders on  each side.
Normally, only one tap is used at a time.  The launder directs the
flowing matte to a point where it can  be collected  in a  large ladle.
Fugitive emissions are observable from  the point at which the matte
leaves  the furnace to the location where it  enters  the ladle.  Typically,
a  matte tapping operation takes  5 to  15 minutes.
      EPA performed testing  at  the ASARCO-Tacoma,    ASARCO-E1  Paso,
and  Phelps Dodge-Ajo48 smelters  to determine captured  fugitive  arsenic
emissions  during matte tapping.   The  results are  presented in Table  2-14.
      Fugitive arsenic emission estimates for matte tapping were developed
based on  the results of  the arsenic  emission measurements presented  in
Table 2-14.   It was  assumed that complete  capture of matte tapping
 fugitive  emissions was  obtained by  the exhaust hood systems.  EPA
 testing of the matte tapping operations at ASARCO-Tacoma was only
 performed when actual  matte tapping  was being  conducted.  However, the
 testing at the ASARCO-E1 Paso and Phelps Dodge-Douglas smelters was
 performed on a continuous basis over a period of time.   To compare the
 test results on a similar basis, the ASARCO-Tacoma test  results were
 therefore adjusted  from an  intermittent to  a continuous  basis.  This
 was  performed by averaging  the emission rate of arsenic  obtained  from
 testing during the  time the matte tapping system was  in  operation over
 a 24-hour time period.  For example,  sample run number  1 at  ASARCO-Tacoma
 (refer to Table 2-14) indicated  an arsenic  emission  rate of  1.69  kg/hr
 (3.73  Ib/hr).  Process  data49'50 indicated  that  7  minutes are  required
 to  fill  a matte ladle,  and 36 ladles  were filled  on  the day sample  run
 number 1  was made.  Therefore,  actual  time  of  matte tapping hood
 system operation was 4.2 hours.  Total fugitive  arsenic matte tapping
 emissions  emitted were  7.1 kg/hr (15.6 Ib/hr).   The average fugitive
 arsenic  matte tapping  emission rate  for sample run number 1 at ASARCO-Tacoma
 was therefore 0.3  kg/hr (0.65 Ib/hr).
                                  2-46

-------
                       Table  2-14.
MATTE  TAPPING  EMISSIONS  FROM  COPPER  SMELTERS
                                                        Arsenic  Emissions
 Phelps  Dodge-
                                                                                    Hood is only  on one matte tap
                                                                                    hole.  The 6  ft. launder was
                                                                                    uncovered.
                                                                                   Approximately 11 x 103 acfm
                                                                                   of additional flow was bypassed
                                                    52.42
                                                    243.05
                                                    236.18
                                                    177.22
                                           The testing was done only during
                                           the matte  tapping operation.
N.A.  - Not available

-------
     At the ASARCO-E1  Paso smelt,r, approximately 11,000 acfm of flow
was being bypassed upstream of the test location.  The ASARCO-E1 Paso
test results were therefore adjusted by the ratio of the total  flow
rate (measured flow rate plus bypass flow rate) to the measured flow
rate.  The results thus obtained,  indicated the average fugitive
arsenic matte tapping emission rates from the ASARCO-Tacoma,  ASARCO-E1
Paso,  and Phelps Dodge-Ajo smelters  to  be 1 kg/hr  (2.15 Ib/hr), 0.31  kg/hr
(0.67  Ib/hr), and 0.19 kg/hr  (0.41 Ib/hr), respectively.
     For the  purpose  of developing an  emission  factor  for  fugitive
arsenic matte tapping emissions,  the fugitive  arsenic  emissions from
the three  smelters were compared  to  the quantity of  arsenic reporting
in the matte  during  the test  period.  The ASARCO-Tacoma  and ASARCO-E1
Paso test  results  indicated  Lhat an  average  of 1.9 and 1.5 percent of
the arsenic in  the matte  reports in  the fugitives, respectively.   The
 Phelps Dodge-Ajo smelter  process samples indicated extremely small
 amounts of arsenic51 in the matte (refer to Table 2-14).    In view of
 the difficulty in obtaining accurate readings for such small arsenic
 contents,  the fugitive arsenic matte tapping test results  for  Phelps
 Dodge-Ajo were not considered in  the development of the emission
 factor.
       Based on the ASARCO-Tacoma  and ASARCO-E1  Paso  results,  an emission
 factor of  1.5 percent of the arsenic  in the matte was assumed  to
 report  in  the matte  tapping  fugitive  emissions.   This emission factor
 was then  used to'estimate fugitive  arsenic matte  tapping  emissions  at
 all  the domestic copper  smelters regardless  of the  kind  of smelting
 furnaces  in  use.  The  amount of arsenic in  the matte  at  these smelters
 was obtained from the  process  arsenic mass  balances in  Appendix F.
       2.3.2.2.4   Slag tapping.   Slag tapping  is a principle fugitive
  emission  source at  the smelting furnace.   Slag tap ports  and  slag
  launders  have been  observed to emit less  fugitive emissions than those
  emitted during matte tapping operations.   However, fugitive emissions
  are observable from the point of the  slag leaving the furnace to the
  location where it enters the ladle.   Typically, a slag tapping operation
  takes 10 to 20 minutes.                                 52
       EPA  performed  testing  at  the  ASARCO-Tacoma  smelter    to  estimate
   fugitive  arsenic emissions  during  slag tapping  and to evaluate  the

                                   2-48

-------
   capture system in  use at the sm^er.   At the ASARCO-Tacoma smelter
   both  the slag  tap  hole  and  the  slag  pot are provided  with exhaust
   hoods,  and  the slag  launder is  covered.   The captured slag tapping
   emissions are  ducted  to  the brick  flue.   The emissions  sampling was
   conducted in the duct leading to the brick  flue.   Results of  this
   testing  are provided  in Table 2-15.
       Testing at the ASARCO-Tacoma  smelter was  only performed  when slag
  was being tapped from the reverberatory furnace.  Three sample  runs
  were made, and during these runs the slag was  also analyzed for  arsenic
  content.  The actual  test time for sample run  number  1 was 60 minutes
  On the day sample run number 1 was  made, 120 pots of  slag were dumped
  Approximately 3 minutes  were required to fill a slag  pot.53  The
  actual  on-time  for  the slag  tapping exhaust hood system was therefore
  6 hours.   At an emission rate of 0.25 kg/hr (0.56 Ib/hr) the arsenic
  emitted  over a  14 hour period  was  1.5 kg (3.3 lb/)  at  an average rate
  of 0.06  kg/hr (0.14 Ib/hr).   Similar  calculations  for  sample runs 2
  and  3 were performed,  and  an average  arsenic emission  rate of
  0.12 kg/hr (0.25 Ib/hr) was  obtained  for  the three  sample  runs.
      For  the purpose of developing  an emission  factor  for  fugitive
 arsenic slag tapping emissions, the arsenic  reporting  in the fugitives
 was compared to the amount of arsenic in the  furnace slag.   It was
 assumed that the arsenic in  slag tapping fugitives-to-arsenic  in  slag
 ratio remained constant regardless of the type  of smelting  furnace
 Also, fugitive arsenic emissions  discharged during slag tapping at  the
 ASARCO-Tacoma smelter  were all  captured by the exhaust hood system.
      At the ASARCO-Tacoma smelter this comparison indicated that an
 average of about 0.1 percent  of the  arsenic in the slag (samples taken
 during  test period)  reported  in the  fugitive emissions.  Based  on
 these results, a factor of  0.1  percent arsenic in the  slag  was  assumed
 to  report  in  the  slag tapping emissions  at  all  smelters.  The amount
 of  arsenic  in the slag  at  each  smelter was  obtained  for the process
mass balances discussed in Appendix  F.
     2'3'2-2'5   Converter slag return.   Converter slag  is returned to
the reverberatory and electric furnaces  through  the converter slag
return launder.  This is either a simple channel  with an opening  in
the furnace wall  or a mechanically operated chute.
                                2-49

-------
Table 2-15.  SLAG TAPPING FUGITIVE ARSENIC EMISSIONS
                 FROM ASARCO-TACOMA

Sample
Run
1
2
3
Avg.
Arsenic in
Slag
Ib/hr kg/hr
208 94
275 125
290 132

Measured Emi
Concentration
g/m3
7.94
23.04
18.16
16.38
ssions
Mass Rate
Ib/hr
0.56
1.43
1.17
1.05


Comments
Testing was only
done when the slag
was being tapped.

                         2-50

-------
       The  number of times convert  slag  is  returned  to  the  furnace
  depends upon the number of converters and the operating  level  of  the
  smelter.  Each time the slag is returned, pressure fluctuations occur
  in the furnace.  These fluctuations are due to the agitation  in the
  bath and the rapid chemical reactions with the slag constituents.
  This tends to generate fugitive emissions through the relatively  large
  converter slag  opening.
       To develop an emission factor for fugitive arsenic emissions
  discharged during  converter slag return  to the smelting furnace, tests
  were performed  by  EPA  at  the  ASARCO-Tacoma smelter.54  The emissions
  during  converter slag  return  were  sampled during  1  to 3 minute intervals
  when  slag  returned  to  the reverberatory  furnace  from  the converters.
  Since this procedure occurs only for a short duration,  testing was
  performed  for 3 days to obtain  an  adequate sample for analysis.
      The test results  over a sample  period of  23 minutes  (18 ladles  of
  converter  slag returns) indicated  0.13 kg/hr (0.3 Ib/hr)  of  arsenic
  was emitted.  The total actual  elapsed time during this test  period
  was 5 hours.  Therefore, the average fugitive arsenic emission  rate
  during converter slag return was 0.01 kg/hr (0.02 Ib/hr).  Analysis of
  the converter slag  indicated that during the 3 days in which  the tests
 were performed,  an  average of 25 kg/hr (56 Ib/hr) of arsenic was being
 returned from the converters to  the furnace with the slag.  Comparison
 of  the arsenic  in  the  fugitives  to  the  arsenic  in the  converter slag
 indicated  0.0004 kg  of  arsenic reports  in the fugitives per 1 kg of
 arsenic  returned with  the  slag.
      Based  on these  results,  it  was concluded  that  the amount of
 arsenic  emitted  (0.01 kg/hr) during converter slag  return  was negligible
 It  should be noted  that the  arsenic input  at  the ASARCO-Tacoma smelter
 is  h1gher than the combined arsenic  input  at  all the other smelters.
     2.3.2.3  Converter.
     2.3.2.3.1  Charging.  During converter charging,  the  converter is
 tilted by a drive mechanism until the mouth of the converter  is  approximately
45 degrees from the vertical.  Fugitive emissions during this  operation
result when matte or other materials such as reverts flux are  poured
from a ladle into a  converter.   The gate on the primary hood  is  retracted
                                2-51

-------
to its highest position.   An ovc.nead crane lifts the matte ladle
above the mouth of the converter and pours the matte into the converter
by tilting the ladle.  During this operation, visible emissions are
heavy, but of short duration.  When charging is completed, blowing air
is turned on while the tuyeres are above the liquid level of the
molten bath.  The converter is rotated until its mouth is in its
topmost position, contained within the primary hood.  The gate is then
lowered and the blowing cycle commences.
      2.3.2.3.2  Blowing.  Most domestic smelters have attempted  to
provide relatively close fitting  primary  hoods on  converter  openings.
These hoods are used  to contain and  capture  offgases  generated during
blowing operations.   However, these  hoods  at best  do  not completely
seal  the  opening  since sufficient space has  to be  provided  for easy
movement  of  the converter shell.   This  is to allow for  converter shell
 irregularities and  the molten copper buildup on the shell due to
 splashing during  the blowing operation.   Some fugitive  emissions are
 discharged through  these  openings, especially when pressure surges
 occur during  converter blowing, and rapid adjustments are not made by
 the duct damper system.
      2.3.2.3.3  Skimming.   During skimming operations,   the mouth of
 the converter is rotated  to a position between 65 to 85 degrees from
 the vertical, depending upon the  bath level.  The blowing air remains
 on as the converter  is rolled out, until  the  tuyeres are above  the
 surface  of the bath.  This  action results  in  significant quantities of
 fugitive  emissions,  but is  necessitated  by  equipment requirements.
 Slag  is  skimmed  from the converter  mouth  into a slag ladle.   Fugitive
 emissions are visible during this operation.   The primary  hood  gate
 may  not  be fully retracted  during skimming; however, the hood  is
  isolated by  dampers  from the main duct system to  keep  dilution  air
  from mixing  with the high  S02  offgases from blowing converters.  At
  the completion  of  skimming, the  converter mouth is again rotated to
  its vertical position within the hood.   The gate is fully extended,
  and the blowing  cycle resumes.
       2.3.2.3.4  Pouring.   As the blister copper is poured, the converter
  is slowly rotated downward until its mouth reaches a position approximately
  90 to 125 degrees from the vertical, depending upon the size of the

                                   2-52

-------
  pour and the buildup within th- converter.  The hood gate may be
  partially extended during this operation.  Fugitive emissions during
  this operation result as the blister copper is poured from the converter
  into the ladle.   After the blister pour,  the converter mouth is rotated
  upward  to a position approximately 45 degrees from the vertical  to
  await a new matte charge and  the  start of a  new cycle.
       2.3.2.3.5  Holding.   There are times during normal  smelting
  operations  when  material,  either  slag  or  blister copper,  cannot be
  immediately transferred  from  the  converters  to the ladles.   This may
  be due  to such conditions  as  unavailability  of the crane,  refining
  furnace operations not allowing additional feed  material,  or others.
  It then  becomes  necessary  for  the  converter  to be  placed  in  a  holding
 mode.   In this mode  the converter  is  rotated  until  the mouth  of  the
 converter is 30  to 45 degrees  from  the vertical  to  keep the  tuyeres
 out of  the bath.  This results  in  fugitive emissions from the molten
 material in the converter being discharged into  the converter building.
      2'3'2-3-6  Converter leaks.  Since the ends of most Pierce-Smith
 converters are joined by bolts and springs, they occasionally leak at
 the end  joint.   When  this leakage  is located below the molten material
 surface, it  is  usually repaired rapidly to prevent major erosion.
 However, in  those cases where it is located above the charge surface,
 H  may not be repaired.   Thus, fugitive emissions may occur at this '
 point.
      EPA performed testing  at  the  ASARCO-E1  Paso smelter55 to develop
 a fugitive arsenic emission  factor for fugitive emissions  occurring
 during converting.  The results indicated  an  emission  factor of 150 g
 of  arsenic per  1  kg of arsenic  contained in the converter  primary
 offgas stream.  Thus  the  emission  factor of 15 percent  of  the arsenic
 contained in  the  converter  primary  offgas, based  on  test data  from
 ASARCO-E1 Paso, was used for estimating converter fugitive  emissions
 at the remaining  low-arsenic throughput copper  smelters.
     2.3.2.4  Anode Furnace.   Refining of  blister copper to anode
copper is performed in rotary-type refining furnaces which are similar
to Pierce-Smith converters.  Unlike converters, however, these furnaces
have no control  whatsoever.  Offgases generated during charging,
                                2-53

-------
blowing, skimming, holding, and pouring are discharged directly to the
building atmosphere from the fuinace mouth.  It should be noted,
however, that emissions generated in the anode or refining furnaces
are considerably less significant both in terms of volume and arsenic
content than from converters.  This is because the quantity of impurities
removed consists of only about 1 percent of the molten material processed.
     Tests by ASARCO-Tacoma were performed separately during the
various phases of the anode furnace operations.    The results indicated
an anode furnace fugitive emission factor of 110 g of arsenic  per  1 kg
of arsenic contained in the blister copper.  Thus an emission  factor
of 11 percent of the arsenic contained in  the  blister copper was
obtained, based on test data from ASARCO-Tacoma.  This emission factor
was  then used to estimate  fugitive arsenic emissions  from  anode furnaces
at the  other domestic  smelters.
     2.3.2.5  Miscellaneous  Fugitive  Emission  Sources.
     2.3.2.5.1  Dust handling  and  transfer.   Dust handling and transfer
can  generate fugitive  emissions  if  carelessly  performed.   However,
most smelters take reasonable  precautions  to minimize fugitive emissions
from dust  handling and transfer.   Dust transfer from control  devices
onto conveyors  is  typically performed by mechanically timed and activated
rotary  valves or  twin  gravity  self-closing gates.   These conveyors are
generally  covered by housings  and  discharge  into storage bins from
which  dust may  be withdrawn as desired.   Dust transfer from storage
 bins is usually through dust-tight connections to surface transportation
 units  such as  tank trucks and dumpsters.  Cleaning  and unloading  of
 dust from flues and  settling chambers is performed  by conveyors which
 feed into hoppers provided at spaced intervals underneath the flues
 and settling chambers.  Both screw and drag-type conveyors are used.
 These flue dusts are usually treated in a pugmill or pelletizing  disc
 where moisture  is added.  The wet dust is then transferred to a bedding
 area,  blended with other feed constituents, and recycled.  Dust from
 waste  heat boilers and crossover flues is usually removed  by  hand
 methods, water-bomb lances, slugger  guns, and  other methods.   This
 dust is handled and transported by surface vehicles  to  the smelter
 flux crushing  system.  A more detailed  description  of dust transfer,
 handling, and  conveying is  given  in  Section  3.1.2.5.

                                 2-54

-------
       In  the  transfer,  handling,  .nd  conveying  of  dust  from  control
  device storage hoppers, smelter  flues, and dust chambers,  it  was
  assumed  that all transfer systems  such as screw or  drag  type  conveyors
  are dust-tight.  The uncontrolled  fugitive arsenic  emissions  from
  these systems were assumed  to not  exceed 0.1 percent of  the arsenic  in
  the dust collected at  arsenic drop-out points  such  as  control  device
  storage  hoppers, smelter flues,  dust chambers, and  waste heat boilers.
  This emission factor was based on  particulate  (dust) emission rates
  found in grain transfer, asphalt batching, and other industries involved
  in transfer, handling, and conveying operations.  In these  industries,
  0.1 percent of the dust transferred, handled, or conveyed is  estimated
  to appear as fugitives.57
      2.3.2.5.2  Ladles.  Normal  process fluctuations may require  that
  ladles containing matte, converter slag,  or blister copper  be temporarily
 set aside until  needed.  Because these ladles contain molten material,
 some emissions  can  be observed due to fuming.  These, however, are
 short-lived.   The exposed  surface of the  material  cools rapidly,
 forming  a solidified  layer  or skull which greatly  limits  fugitive
 emissions.   It  was  therefore assumed that no  fugitive arsenic  emissions
 are discharged  from  ladles  once  they are  filled.
      2-3-2.5.3  Slag  dumping.   Smelting furnace slag is disposed of by
 water  granulation or  by transport in  the  molten state for dumping  a
 short  distance  from  the smelter.   Slag dumping  is  the more  widely  used
 method.   The  slag is  transported  to the dump  site  by train  or  slag
 hauler.   The  slag train is usually  comprised  of a  number  of  slag pots
 or  ladles on flat cars.  Solidification at the  surface  of the  slag in
 the pots  is fairly rapid.  Fugitive emissions during transporation to
 the dumping site are  therefore limited.   However,  during  dumping of
 slag at the dumping site, substantial fugitive  emissions, although
 short in duration (less than 1 minute), can be  observed.
     Testing was performed at the ASARCO-Tacoma smelter by EPA to
 determine the magnitude of fugitive arsenic emissions during slag
 dumping.     Reverberatory slag from slag pots was analyzed for arsenic
 content on exit from the reverberatory furnace and after the slag  was
deposited on the dump site.   Process sample analysis was performed on
 four separate days.   The test results obtained are listed in Table 2-16.

                                2-55

-------
              Table 2-16.   REVERBERATORY FURNACE SLAG
            ANALYSIS FOR ARSENIC CONTENT AT ASARCO-TACOMA
                   Percent arsenic in slag at
                   exit from furnace slag        Percent arsenic in
Sample Run                launder                slag at dump site

    1                      0.33                         0-40
    2                      0.38                         0.52
    3                      0.46                         0.44
    4                      0.49                         0.29
  Avg.                     0.42                         0.41
                                 2-56

-------
     Test  results  indicated  that  m  the  first  two  of  the  four  samples
 the arsenic content of the reverberatory furnace slag  increased  in
 concentration at the dump site.   In  the  third  sample  there was a
 slight decrease in concentration  at  the dump site.  The fourth sample
 indicated a larger decrease  in concentration at the dump  site.  However,
 comparison of the averaged results indicated approximately the same
 concentration of arsenic in  the slag at the exit from  the furnace and
 at the dump site.   This indicates that fugitive arsenic emissions
during slag dumping are negligible.
     Table 2-17 presents a summary of the potential fugitive arsenic
emission estimates  in  the absence of control  for the 14 low-arsenic
throughput primary  copper smelters.
                               2-57

-------
  TABLE  2-17.   SUMMARY  OF  POTENTIAL FUGITIVE ARSENIC EMISSION ESTIMATES
IN ABSENCE OF  CONTROL FOR  LOW-ARSENIC THROUGHPUT PRIMARY COPPER SMELTERS


Smel tec
ASARCO-E1 Paso



ASARCQ-Hayden


Tennessee Chemical
Co. - Copper-hill
Inspirafion-
Mi;ni
Kennecott-Garf ield



Kennecott-Hayden


Kennecott-Hurley


Emission
Source
CT
ST
MT
CONV
AF
FDH
ST
MT
CONV
AF
FDH
ST
MT
CONV
FDH
ST
MT
CONV
AF
FDH
ST
MT
CONV
AF
FDH
ST
MT
CONV
AF
FDH
ST
MT
CONV
AF
FDH


Arsenic Content
Present in
Calcine
Furnace Slag
Matte
Process Offgas
Anode Furnace
Transferred Dust
Furnace Slag
Matte
Process Offgas
Anode Furnace
Transferred dust
Furnace Slag
Matte
Process Offgas
Transferred Dust
Furnace Slag
Matte
Process Offgas
Anode Furnace
Transferred Dust
Furnace Slag
Matte
Process Offgas
Anode Furnace
Transferred Dust
Furnace Slag
Matte
Process Offgas
Anode Furnace
Transferred Dust
Furnace Slag
Matte
Process Offgas
Anode Furnace
Transferred Dust
Amount, kg/nr
146.5
54.0
48.1
76.2
12.1
260.0
99.0
97.0
45.0
20.0
194.0
0.4
0.7
0.5
0.6
14.1
5.4
1.5
0.7
1.9
33.0
15.7
6.0
13.1
189.5
2.2
7 9

5.0
0.09
4.1
0.8
A Q
0.4
0.16
1.6
Emission
Factor
(percent)
o'.i
1.5
15.0
11.0
0.1
0.1
1.5
15.0
11.0
0.1
0.1
1.5
15.0
0.1
0.1
1.5
15.0
11.0
0.1
0.1
1.5
15.0
11.0
0.1
0.1
1 5

15.0
11.0
0.1
0.1
1 5
15.0
11.0
0.1
Emission
Rate, Absence
of Control
kg/hr
u.o /
0.05
07

11.4
1.3
0.26
0.1
1.5
.8
2.2
.2
0.0004
0.01
0.08
0.0006
0.01
0.08
0.22
0.08
0.002
0.03
0.2
1 C
00

0.002
0.11

0.01
0.004
0.001
0.014
0.054
0.02
0.002
                                      2-58

-------
        Smelter

   i^ennecott-McGil
  Magma-San Manuel
  Phelps Dodge-Ajo
  Phelps  Oodge-
    Oouglas
 Phelps Dodge-
   Hi Idal go
•Phelps  Dodge-
   Mo renci
Copper Range -
  White Pine
Emission
Source
  MT
  CONV
  AF
  FDH

  ST
  MT
  CONV
  AF
  FDH

  ST
  MT
  CONV
  AF
  FDH

 CT
 ST
 MT
 CONV
 AF
 FDH

 ST
 MT
 CONV
 AF
 FDH

 ST
MT
CONV
AF
FDH
                      ST
                      MT
                      CONV
                      AF
                      FDH
                                               Arsenic Content
Present in
furnace Slag '
Matte
Process Offgas
Anode Furnace
Transferred Dust
Furnace Slag
Matte
Process Offgas
Anode Furnace
Transferred Dust
Furnace Slag
Matte
Process Offgas
Anode Furnace
Transferred Dust
Calcine
Furnace Slag
Matte
Process Offgas
Anode Furnace
Transferred Dust
Furnace Slag
Matte
Process Offgas
Anode Furnace
Transferred Dust
Furnace Slag
Matte
Process Offgas
Anode Furnace
Transferred Dust
Furnace Slag
Matte
Process Offgas
Anode Furnace
Transferred Dust
Amount, kg/hr

34.0
35.' 6
3.3
14.9
0.4
0 6
u • u
0.4
0.06
0.*7
10.1
4. 1
2.3
0.9
11.8
10.6
6.2
4 2
T • L.
3.2
0 ?
y • L.
1.9
3.0
1 "i
•L * J
0.9
0 fi
W » 
-------
2.4  REFERENCES

 1   Background Information Document for Review of New Source^Performance
     Standards for Primary Copper Smelters (Draft).   U.S.  Environmental
     Protection Agency.   Research Triangle Park, North Carolina.  EPA
     Contract No. 68-02-3056.   October 1982.   p. 3-2.

 2   Telecon.  Whaley, 6., Pacific Environmental Services, with Butterman,
     W., U.S. Bureau of Mines.  Primary copper smelter production in
     1982.  March 2, 1983.

 3.  Reference 1, p. 3-3.

 4   Field Surveillance and Enforcement Guide  for Primary Metallurgical
     Industries.  U.S. Environmental  Protection Agency.  Research
     Triangle  Park,  North  Carolina.   Publication  No.  EPA 450/3-73-002.
     December  1973.  p. 175.

  5.  Reference 4,  p.  180.

  6.  Letter  and  attachments  from J.W. Maksym,  White Pine  Copper Division,
     Copper  Range Company, to J.R.  Farmer,  U.S.  Environmental  Protection
     Agency.   March 17,  1983.  Response to  Section  114 Information
     Request.

  7   Letter and  attachments from L.R. Judd, Phelps  Dodge Corporation
      to J.R. Farmer, U.S. Environmental Protection  Agency.   April /,
      1983.  Response to Section 114  Information Request.

  8   Letter and attachment from R.A. Malone, Kennecott Minerals Company
   '  to J.R.  Farmer, U.S. Environmental Protection Agency, March Ib,
      1983.  Response to Section 114  Information Request.

  9   Letter and attachments  from M.O.  Varner, ASARCO, Inc.  to  J.R.  Farmer,
      U.S. Environmental  Protection Agency, March 16,  1983.  Response
      to Section  114 Information Request.

  10   Letter and  attachment from J.W.  George,  Tennessee Chemical  Company,
      to  J.R.  Farmer, U.S. Environmental  Protection Agency,  April  6,
      1983.   Response to  Section 114  Information  Request.

  11   Letter and attachment from J.H. Boyd,  Magma Copper  Company, to
    "   J R  Farmer,  U.S.  Environmental Protection  Agency,  March 15,
       1983.   Response to  Section 114 Information  Request.

  12.   Letter and attachment from T.B. Larsen,  Inspiration Consolidated
       Copper Company, to J.R. Farmer, U.S.  Environmental  Protection
       Agency.  March 14,  1983.  Response to Section 114 Information
       Request.

  13   Process Parameters  for Primary Copper Smelters  and Their Effects
       on Arsenic Emissions (Draft).  U.S. Environmental Protection


                                  2-60

-------
                                  n-rk,  North


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

  15.  Reference 1,  p.  3-18.


  16'  [nsei?ationhC1ey' ^"  Pac1fic Envi>°™ental Services, with
       converters.   March 4,  1983.                 a us o   oboken

  17.  Reference  1,  p.  3-36.


  18'  ??rP™t£;J't 51 aLJ!1!or Elements 1n the No™nda Process.
       16-20? 1975!)     1Mth AnnUa1  AIME Meet1ng-   New York-   Feb^a^

                                              of Copper.   London.   John


 20.  Correspondence from  Mr. Jim Henderson,  ASARCO,  Incorporated,  to
      fir. U.R.  Goodwin, Environmental  Protection Agency,  August 30,
 21.  Yazawa,  A   Thermodynamic  Considerations of Copper Smeltina
      Canadian Metallurgical  Quarterly.  .13(3):443. ^g^"16111"9'

 22'  DuHnn'rln  a^  T1\Azakam^-  Thermodynamics of Removing Impurities
      8:257    ?Q6Q   Smeltlng'  Canadian Metallurgical  Quarterly/
            °ncence from K'W- Nelson' ASARCO, Incorporated, to J
            , Environmental Protection Agency.  March 17? 1976.

 24.   Harris, D.L.  Air PolHuion Emission Test - Paniculate and
      Arsenic Emission Measurements from a Copper Smeler   Volume I
      June 26-30  Ig3";? ReSe3rCh CorP°^tion-   ™ Report No.  77-CUS-6
              Contract No.  68-01-4136.   January  1978.
27.  Correspondence  from  "ckerlng, I.G., Vice President Environmental
     cwn  c   ^ennecott. Copper Corporation, to D.R. Goodwin, Director
                       1  Pr°teCt10n A9en^'  HW 9> »78r'nDaurftr0r>
                               2-61

-------
28.   Stankovic,  D.   Air Pollute ,  Caused by Copper Metallurgy Assemblies
     in Bor.   Institute for Copper.   Bor.   Environmental  Protection
     Agency.  No. 02-513-1.

29.   Conversation with Mr.  Hank Hansen, Environmental Engineer, Kennecott
     Copper Corporation, Salt Lake City, September 26, 1978.

30.   Ruddle, R.W.  The Physical Chemistry of Copper-Smelting.  Institute
     of Mining and Metallurgy.  London.  1953.

31.   Turnbull, D.L., "Converter Practice at Mufulira," Seventh Commonwealth
     Mining and Metallurgical Congress, 1961.

32.   Rozlovskii, A.A.,  "Behavior of Arsenic in the Production  of
     Nonferrous Metals," Tsvetnye Metally/Nonferrous  Metals, Number  UDC
     669.778.

33.  Paulson, D.L., et  al.   Smelting of Arseniferous  Copper Concentrate
     in an  Electric Arc Furnace.  U.S.  Bureau  of  Mines.   Report  of
     Investigation No.  8144.   1976.

34.  Reference  18.

35.  Reference  18.

36.  Reference  23.

 37.   Reference  24.

 38.   Schwitzgebel,  K., et  al.  Trace Element Study at a Primary Copper
      Smelter.   Prepublication copy.   U.S.  Environmental  Protection
      Agency.   Contract No. 68-01-4136.  January 1978.

 39    Harris,  D.L.   Air Pollution Emission Test - Particulate and
      Arsenic Emission Measurements from a Copper Smelter.  Volume I
      (text).   Monsanto Research Corporation.  EMB Report No.  77-CUS-6.
      June 20-30, 1977.

 40.  EPA testing at Anaconda  (process  emissions).

 41.  EPA testing at Kennecott-Hayden  (process emissions).

 42.  TRW Environmental Engineering Division,  Emission Testing of
      ASARCO Copper Smelter,  Tacoma, Washington.   EMB Report No.  78-CUb-l
-------
 45.  Reference 42.

 46.  Reference 42.

 47.  Reference 43.
 48.   TRW Environmental  Engineering Division.  Emission Testing at
      Pnelps Dodge Copper Smelter, Ajo Smelter.  EMB Report No. 78-CUS-9
      Februar
     February 1979.

49.

     1978

50.
      Correspondence from Mr.  K.W.  Nelson, ASARCO, Incorporated, to
                 Goodwm, Environmental  Protection Agency, September 14,
      Correspondence  from Mr.  A.L.  Labbe,  ASARCO,  Incorporated, regarding
      process  data  for  the ASARCO-Tacoma  plant for the arsenic sampling
      period September  12-24,  1978,  to  Mr.  S.T.  Cuffe, Environmental
      Protection Agency.   November  20,  1978.

 51.   Reference 48.

 52.   Reference 42.

 53.   Reference 50.

 54.   Reference 42.

 55.   Reference 43.


 56.   Godsey, E.S.  Tacoma Plant Anode  Furnace Emissions.  ASARCO  Salt
      Lake City office.   August 8, 1975.


                    C.L. Wilson.  Metallurgy of Copper.  London.  John


58.  Reference 42.
                               2-63

-------
                          3.0  CONTROL TECHNOLOGY
       This  chapter identifies  alternative control  techniques which can
  be applied to  process  and  fugitive  emission  sources  at primary copper
  smelters to  control  arsenic emissions.   The  amount of  arsenic  emission
  reduction  which can  be  achieved  by  the  application of  these control
  techniques as  well as major factors affecting  their  performance  are
  discussed.
  3.1  ALTERNATIVE  CONTROL TECHNIQUES
  3.1.1  process Emission Control
      3-1'1-1  General Considerations.   As discussed  in Section 2.0, much
 of the arsenic entering the copper smelting process  is volatilized and
 eliminated  as metallic oxide in the process offgas streams.  This is a
 result of the very high temperatures associated with  the pyrometallurgical
 processes used  and the inherent volatility of arsenic and its prevalent
 oxide,  arsenic  trioxide (As203).   The  extreme volatility of arsenic,
 and especially  arsenic  trioxide,  is  the  single most important factor
 affecting the controllability  of  arsenic emissions from sources at
 primary copper  smelters.
      Several  studies  have been  made  to determine  the  vapor pressure  of
 arsenic  trioxide in air  at  various temperatures.1  Table  3-1 presents
 vapor pressure  data for  arsenolite (As406), the more  common  form  of
 arsenic  trioxide and  the most abundant arsenic  compound in smelter
 offgases. '   Also presented are  the arsenic  concentrations  at  saturation
 corresponding to the  temperatures and vapor pressures listed.   Figure  3-1
 illustrates  graphically the vapor pressure-temperature  relationship
 indicated by these data.  This  figure illustrates a significant logarithmic
 increase in  the vapor pressure  of arsenic trioxide, and thus the
amount of arsenic  which  can  exist in  the  vapor state,  with temperature
                                3-1

-------
Table 3-1.  SUMMARY OF As406 VAPOR PRESSURE DATA AND CORRESPONDING
          ARSENIC CONCENTRATION AT VARIOUS TEMPERATURES
Temperature,
°C(°F)
457.2 (855.0)
412.2 (774.0)
332.5 (630.5)
299.2 (570.6)
259.7 (499.5)
212.5 (414.5)
200.0 (392.0)
175.0 (347.0)
150.0 (302.0)
125.0 (257.0)
110.0 (230.0)
100.0 (212.0)
90.0 (194.0)
Vapor pressure,
mm Hg
760
400
100
40
10
1.0
0.910
0.175
2.77 x 10~2
3.58 x 10"3
8.81 x 10"4
3.31 x 10"4
1.18 x 10"4
Arsenic concentration,
g/m3
5.0 x 103
2.81 x 103
7.94 x 102
3.36 x 102
90.20
9.90
9.25
1.88
0.315
0.043
0.011
0.0043
0.0016
                                3-2

-------
                3.0-
        I
        w
        z
        o
                                                     S
                                                     O
                                                     Q_
0.01 L
100
1
,
5C
0.1
10
                     TEMPERATURE, *C


Figure 3-1.   Arsenic Trioxide Vapor  Pressure and Saturated

           Vapor Concentration with  Temperature


                               3-3

-------
Furthermore, the vapor pressure H;,ta indicate that arsenic trioxide
maintains an appreciable vapor pressure at relatively low temperatures.
     For example, Table 3-1 shows that at 200°C (392°F), which is
typical of the outlet temperature of many of the control devices
currently used in the industry, 9.25 grams of arsenic trioxide per m
of gas (g/m3) (4.04 gr/ft3) could exist in the vapor form.  Even at a
temperature as low as 125°C (257°F), a gas stream could contain 0.043  g/m
(0.019 gr/ft3) of arsenic trioxide as vapor.  Any concentration higher
than 9.25 g/m3 in a gas stream at 200°C would lead to condensation of
some of the vapor, and the As405 would then exist in two phases:  a
vapor  or gaseous phase and a  condensed or particulate phase.  The
vapor  would pass through a particulate control device without collection,
while  the condensed material  would  be  collected at about  the  same
efficiency  as total  particulate  matter.   The  implication  of these
facts  with  regard  to  the controllability  of  arsenic  is  important.   The
temperature of  the gas  stream determines  the  amount  of  arsenic  trioxide
which  can  exist as vapor and, consequently,  the quantity  of  condensed
or particulate  arsenic  which  can potentially  be collected in  a  particulate
control  device.  A lower temperature gas stream  is  more likely  to
 contain more arsenic trioxide in a  form which can be collected  in  such
 a device.
      An example will serve to illustrate these points.   A 2,834 m /min
 (100,000 acfm)  gas stream  at 150°C (302°F) with an arsenic trioxide
 mass flow rate of 40 kg/hr would have an arsenic concentration of
 0.235 g/m3 (0.103 gr/ft3).  Table 3-1 shows that at 150°C, the saturation
 concentration for As40g is 0.315 g/m3.  Therefore, the stream is at  a
 subsaturation level  and no condensation, and hence no  particulate
 collection, is predicted.   If this stream were cooled  to  125°C  in  a
 spray chamber, the arsenic concentration at  the new flow  rate of
 2,667 m3/min (94,100 acfm) would increase to 0.250 g/m  (0.109  gr/ft ).
 Since the  arsenic saturation concentration shown in Table 3-1  for  this
 temperature  is  0.043 g/m3, the  amount  of  arsenic leaving  the spray
 chamber  in condensed form would be 0.207  g/m3  (33.12 kg/hr), and  the
  remaining  amount, 0.043 g/m3, would  remain  in  vapor  form.  If  this
  stream was then  sent  to a baghouse collector with  a  control  efficiency
  for total  particulate  matter of 96 percent,  a total  of 0.96  x

                                  3-4

-------
  33.12 kg/hr = 31.80 kg/hr would * collected.  The overall control
  efficiency for arsenic at the lowered temperature would thus be
  31.80/40 = 8U percent.  From Table 3-1 it can be seen that as the
  temperature is further lowered below 125°C, the saturation concentration
  becomes  very small.   In this case, virtually all  of the arsenic trioxide
  present  in the stream would  condense into a collectible form and the
  collection efficiency for arsenic  would approach  the particulate
  efficiency of  96  percent.
       Tests conducted  by EPA  across a hot  electrostatic precipitator
  (ESP) which  controlled  particulate emissions  from a "green"  charge
  reverberatory  smelting  furnace further  demonstrate this  phenomenon.4
  The  electrostatic precipitator was  operated at  315°C  (600°F)  or  higher.
  Measurements for particulate matter conducted using  in-stack  filters,  '
  at the operating temperature of the ESP,  demonstrated  an overall
  collection efficiency for particulate matter of about  97 percent.   In
  contrast, arsenic measurements conducted  using a modified  EPA Method 5
  sampling train operated at 121°C (250°F)  indicated an  average arsenic
  collection efficiency for the ESP of less than 30 percent  (see Table 3-2).
  Calculations based on the arsenic concentration measured at the hot
 ESP inlet and the saturation  concentration data presented  in Table 3-1
 indicate  that the application of a  97 percent effective ESP to the
 same  reverberatory furnace process  gas stream which had been cooled to
 110-C (230°F) could  result in an overall arsenic collection efficiency
 of  90 percent.   These  measurements  and calculations suggest that,
 while the subject  ESP  was  reasonably effective  in  removing  material
 which existed as  particulate  matter at its operating temperature,  any
 material  such as  arsenic trioxide which  would  be predicted  to exist in
 the vapor state at the elevated  temperature  at which  the  ESP  was
 operated  passed through  the ESP with little  removal.   This  demonstrates
 the need  to cool the gas stream  to  be  treated to the extent practicable
 to condense as much of the arsenic  trioxide  vapor  as possible  prior to
 its entering a control device for collection.
     As a result, the alternative control  techniques for arsenic
process sources considered herein include  precooling as an  integral
part of the overall control system.   These include the  application  of
baghouses, high  voltage electrostatic precipitators, or high energy
                                3-5

-------
                                       Table 3-2.  ARSENIC  DATA FOR HOT ESP
Sample
run
1
2
3
Avg.

Temp. ,
°C (°F)
.,.-. . - — • —
328 (622)
317 (602)
336 (639)
327 (621)
Inlet 	
Emissions
g/Nm3(gr/scf) kg/hr (Ib/hr)
__ 	 — 	 — 	
0.268 (0.117) 26.8 (59.1)
0.325 (0.142) 33.0 (72.7)
0.334 (0.146) 34.1 (75.2)
0.309 (0.135) 31.3 (69.0)

i einp. »
°C (°F)
313 (595)
321 (610)
304 (580)
313 (595)
Outlet
Emissions
g/Nm3 (gr/scf) kg/hr (Ib/hr)
0.211 (0.092) 24.3 (53.6)
0.181 (0.079) 20.2 (44.6)
0.199 (0.087) 23.2 (51.1)
0.197 (0.086) 22.6 (49.8)
— _ 	 —
Efficiency,
percent

9.3
38.6
32.0
27.8
I
en

-------
  venturi scrubbers, preceded by g .s stream cooling.  In addition, these
  include contact sulfuric acid plants, with their extensive gas precleaning
  and conditioning systems, applied  to process sources which generate
  offgases containing  high concentrations of S02 (greater than 3.5 percent).
       Before proceeding  with the discussions  on control  techniques,  it
  should  be  noted that the 30 percent  arsenic  collection  efficiency
  recorded for the hot ESP cannot be predicted using the  vapor pressure
  data  for arsenic trioxide.   In  fact,  given the operating  temperature
  of  the  ESP  (315°C),  no  arsenic  recovery would  be  predicted  because  the
  arsenic  concentration in  the  gas stream was  several  orders  of magnitude
  lower than  that  needed  to achieve  saturation.   This  suggests  (1)  that
  other arsenic compounds  less  volatile than arsenic trioxide may  be
  present  in  the gas stream,  and/or  (2) that condensation,  although the
  principal mechanism  in situations where more arsenic trioxide is
  present  than is  needed to saturate the  gas stream, is not the only
 mechanism by which arsenic  trioxide can be collected.
      3.1.1.2  Gas Cooling.  Methods commonly applied in the nonferrous
 metals industries for cooling hot gas streams include radiative cooling,
 evaporative cooling,  and cooling by dilution with ambient air.  The
 dilution method  consists of introducing  sufficient quantities of
 ambient air into the  hot gas stream so that the resultant gas mixture
 is  at  the desired temperature.  The ambient air required may be introduced
 by  infiltration  or forced draft.  Evaporative cooling uses water to
 cool hot gases in spray  or quench chambers.    Water is  injected into
 the  hot  gas  stream where  the heat contained in  the gases vaporizes the
 injected  water,  resulting  in a temperature  reduction.  Radiative
 cooling  relies on heat loss  due  to  natural  convection and  radiation  to
 effect cooling.   These losses  occur whenever  a  temperature  gradient
 exists between gases  inside  a  duct  and the  surrounding air.   Cooling
 of gases  by  this  method requires only  that  a  sufficient  heat  transfer
 area be available  to  obtain  the  desired  cooling.   All three methods
 have advantages and disadvantages.
     Although cooling  with dilution air  is  the  simplest  alternative,
 its  application for the gas  volumes  and temperatures under consideration
may  not be economical.  Depending on the temperature of  the gas stream
                                3-7

-------
to be treated, the amount of dil -non air needed to effect cooling
could result in a two- to four-fold increase in the total gas volume
to be treated, with a corresponding increase in the size and cost of
the draft fan and the control device required.
     Due to the need for sufficient heat transfer area,  radiative
cooling requires considerable space.  Depending on the temperature  of
the gas stream to be cooled, the heat transfer  areas  required could
exceed 1.6 m2/m3  (500 ft2/1000 ft3) of gas  treated.   Normally used  in
the  lead industry on  zinc fuming operations,  radiative  cooling  devices
consist of a  series of  10 to 20 U-shaped  tubes, each  about  3 m  (10  ft)
 in diameter  and  20 m  (66 ft) in height.   In addition, fan  horsepower
 requirements  increase due to the  increased  resistance to gas flow
 resulting  from the  added ductwork.   The  major drawback to radiative
 cooling,  however, is  its limited  flexibility for  temperature control.
      Cooling hot gases  by evaporative cooling is  relatively simple and
 requires  little space.   Water  spray chambers are currently used at a
 number of copper smelters for cooling process gases  from a variety of
 sources prior to entering electrostatic precipitators or baghouses for
 particulate removal.  Typically, the spray chambers  have a  cross-sectional
 area of about 35 m2  (375 ft2) and are 30 to 60 m (100 to 200 ft) in
 length.  The  large cross-sectional area  results  in a low flow  velocity
 ai,d a relatively long  residence time.  Water  is  introduced  through a
  series of  sprays along  the cross  section of  the  chamber.   Water require-
  ments vary  depending on the temperature  of the stream to  be cooled and
  the desired  end temperature.
       The  major difficulty  in  applying water spray  chambers for cooling
  smelter  offgases is  the potential  for  corrosion.  The amount of cooling
  achievable is limited  by the  dew point  of  the treated gas stream.  As
  the gases are cooled,  water and  sulfuric acid mist  contained in the
  gas stream may condense, creating a corrosion problem  in the flue
  system, control device, and stack.  While the problem  of corrosion
  attack is potentially  severe, it can be negated by  the use of  appropriate
  construction materials.   Acid resistant cement, stainless  steel,
   nickel, and. chromium  alloys have been used  in several  cases.   No
   significant  corrosion  problems have been  reported at  primary  copper
                                   3-8

-------
  and lead smelters where spray cb-,,oers are used for cooling  prior  to
  electrostatic precipitators and baghouses used for particulate matter
  control.
       3*1'1-3  Ba3ho"ses (Fabric Filters).  Baghouse particulate collectors
  have historically achieved collection efficiencies in excess of 99 percent,
  over a broad range of applications.  Although extensively used in the
  primary lead and  zinc industries for the  collection of particulate and
  metallurgical  fume,  their  application at  primary copper smelters has
  been limited.  ASARCO-Tacoma  is  the only  smelter which currently uses  a
  baghouse  for the  control of particulate matter contained in smelter
  offgases.
       In principle, particles  contained  in the treated  gas  stream are
  initially captured and  retained  on  the  fibers  of the fabric  by  a
  number  of mechanisms, including  direct  interception, inertial  impaction,
  diffusion, gravitational settling,  and  electrostatic attraction.  Once
  a mat or cake of dust is formed  on  the  fabric,  further  collection is
  achieved by simple sieving.  Periodically, the  fabric is cleaned by
 mechanical or other means to allow  for disposal of the  collected
 particulate and to maintain the pressure drop across the filter within
 practical  operating limits.
      The filtration area required at a specified pressure drop is
 dependent  on  the  gas  volume treated, particulate loading, permeability
 of  the fabric used,  resistance properties  of the particulate deposited,
 and the  cleaning mechanism  used.   The pressure drop commonly found in
 baghouses  applied  in  the nonferrous  metals industries  ranges from 0  5
 to 2.0  kPa  (2 to 8  inches H20).5   The filtering  velocity,  or air-to-cloth
 ratio, generally ranges  from 0.30 to 0.61  m3/min per m2  (1  to 2  acfm/ft2)
 for conventional mechanical shaker cleaning  type baghouses  when  applied
 to metallurgical fume.   Pulse jet cleaning  type  units generally  operate
 at higher air-to-cloth ratios,  ranging from  1.8  to  3.0 m3/min  per m2
 (6 to 10 acfm/ft^).5
     Although there is a variety of woven and felted fabrics available
woven  Orion and Dacron bags  should be suitable in smelter applications.'
Both  exhibit good  resistance to acid  attack and may be operated at
temperatures  ranging up to 135°C (275°F)  with no significant deterioration 5
                                3-9

-------
Bag life, which varies considerably with operating conditions, should
be approximately 1 to 3 years.   Closed pressure or closed suction
baghouse designs may be used in the applications considered.  The
baghouse design selected depends on the acid dew point and the consequent
corrosion potential of the smelter offgases being treated.
     3.1.1.4  Electrostatic Precipitators (ESP's).  Single-stage
electrostatic precipitators are widely used in the primary copper
industry for the control of particulate emissions from smelting facilities.
Electrostatic precipitators use electrical  forces for the removal  of
suspended particulates in a gas stream.  The process encompasses  three
basic functions:   the  charging  of  particles, the  collection  of  charged
particles,  and  the  removal  of  the  collected particles.   Particles
suspended in the  gas  stream are charged while  passing through a high
voltage, direct current  corona established  between  a  discharge  electrode,
usually a small diameter wire  which  is maintained at  high voltage, and
a grounded  collecting surface  (collecting  electrode).   As the particles
pass through  the  corona, they  are bombarded by negative ions emanating
from the discharge electrode,  and charged  within a fraction of a second.
The charged particles, influenced by electric  field forces, migrate
 toward  the  grounded collecting surface where they are deposited and
 held by electrical, mechanical, and molecular forces.  Particulate
 metter adhering to the collecting surface  is periodically dislodged by
 mechanical  rappers or by flushing with water.  The material  is collected
  in a hopper and periodically removed  for disposal or recycle.
      Two principal types of precipitator design  are available, the
  wire-and-plate and wire-in-tube types.  In the more common  plate-type
  precipitators, the collecting  surface consists  of parallel  vertical
  plates  spaced  15  to  30  cm  apart with  wire  or  rod discharge  electrodes
  suspended  vertically between  the  plates.   The  plates are typically  4
  to  12  m in height and 4 to 7  m in length.6 Plate-type  ESP's are
  generally  applied directly to process sources  for  the  control  of dry
  particulate matter at elevated temperatures,  usually 200 to 340°C (400
  to 650°F).  The  collecting surface in tube-type precipitators consists
  of a cylinder with the  discharge electrode centered  along its longitudinal
  axis.   This  type of ESP is used  exclusively for acid mist elimination
                                  3-10

-------
  and fine particulate removal pn>,  to acid manufacturing.  With either
  type,  a complete preclpltator installation consists of several individual
  subunits positioned both in series  and in parallel to achieve the
  desired collection  efficiency.
       There are several  important factors  to be considered in the
  design  and sizing of a  preclpltator  to obtain a desired efficiency.
  These  include  the volume,  temperature, and moisture content of the gas
  stream  to  be treated, and  the resistivity,  size distribution,  and
  loading  of  the particulate to be collected.   Of these,  resistivity is
  the most important.   Resistivity (the  reciprocal  of conductivity)  is
  dependent  on the chemical  and physical  properties  of  the  particulate,
  and the temperature  of  the stream.  Because  of  the  presence  of metal
  oxides, the resistivity of smelter dusts  is  relatively  high  at reduced
  temperatures (90 to  200°C) in the absence of  natural conditioning
  agents such as moisture and S03.  Too  high a  resistivity  (greater  than
  10   ohm-cm) may result in excessive sparking, which can  seriously
  limit precipitator performance.   Preconditioning with moisture and
 sulfuric acid  has been applied to decrease particle resistivity prior
 to precipitation.
      3-1-1-5  Venturi Scrubbers.   In  a venturi scrubber, flue gases
 are passed  through a venturi  at  high  velocity, and the scrubbing
 liquid  is  introduced at  the venturi  throat under low static pressure
 where  it  is  atomized and accelerated  by the gas  stream.   Particles
 suspended in the gas stream are  removed by impaction with  the atomized
 liquid droplets.  The wetted  particles  and entrained liquid are removed
 by  a cyclone or other type  of  entrapment  separator.  The  collection
 efficiency achieved,  which  can be in excess  of 99  percent, is  directly
 related to  the  total   amount of energy expended in  forcing  the  gases
 through the venturi  and  in atomizing and accelerating the  scrubbing
 liquid.  The expended energy is reflected  in  the pressure  drop  across
 the device,  which may range from  2.5 kPa (10  in. H20) to over 20 kPa
 (80 in.  H20).  Typical throat velocities for venturi units  range from
75 to 100 m/s (15,000 to 20,000 fpm)  and 1iquid-to-gas ratios range
from 0.4  to  2 liters/min per m3/min (3 to 15 gpm/103 acfm).8  It is
                                3-11

-------
important that the liquid-to-gas :utio be sufficiently high to guarantee
complete wetting across the venturi  cross-section.
     The application of venturi  scrubbers at primary copper smelters
is limited to a tew smelters where they are used to augment gas stream
precleaning systems and to provide additional gas cooling prior to
acid manufacturing.
     3.1.1.6  Sulfuric Acid Plants.   As noted previously, smelter
offgases containing high concentrations of S02  (over 3.5 percent) are
generally treated  in single- or double-contact  sulfuric acid  plants
for S0?  removal.   The  presence of solid and  gaseous contaminants, such
as acid  mist, entrained dust, and metal fumes in  the  treated  smelter
offgases can  present serious difficulties  in the  operation  of an  acid
plant.   The major  difficultias  caused  by  these  contaminants include
the corrosion of  heat  exchanger tubes,  plugging of  catalytic  beds,
deactivation  of the catalyst, and contamination of  the product acid.
As  a  result,  rather extensive measures have  to  be taken to remove
contaminants  to ensure that their concentrations are  reduced to tolerable
 levels  prior  to entering  the acid plant.
      Table 3-3 contains estimates of the maximum levels of impurities
 that  can be tolerated  in  smelter  offgases used  for sulfuric acid
manufacturing.  The degree of catalyst deterioration  experienced at
 these various impurity levels can be tolerated  by shutting down the
 acid  plant once per year to screen the catalyst and repair the equipment.
 Table 3-3 also contains the estimated upper level of impurities that
 can be  removed by  typical gas precleaning systems with prior  removal of
 coarse  dust.   Although complete removal of  contaminants,  such as
 arsenic, from  the  offgases  is not practical, over 99  percent  removal
 is considered  achievable.
      Both hot  and cold gas  cleaning devices are  used.  Generally,  the
 offgases  are initially treated in a hot  electrostatic  precipitator
 where  the coarse  particulate,  which contains large amounts of metals,
  is removed.   The  gases exiting the  precipitator are  then  scrubbed  in
  one  or more  packed-bed or impingement-type  scrubbers where,  in addition
  to undergoing  further particulate  removal,  the gases are  humidified
  and  cooled.   The cooled  gases  then  enter a  series of electrostatic
                                  3-12

-------
      Table 3-3.   ESTIMATED APPROXIMATE MAXIMUM IMPURITY  LIMITS FOR
               METALLURGICAL OFFGASES  USEDJO  MANUFACTURE
                             SULFURIC  ACIDy
                           T     .    .               Impurity  limits at
         c  .  .              Impurity  limits at       inlet  to  gas  puri-
         Substance          inlet to  acid plant,     fication system,
	(mg/Um  )3	     (mg/Nm3)a'b
Chlorides, as Cl                  i<2                    125c
Fluorides, as F                   o.25                    25d
Arsenic, as As203                 i.2e                   200
Lead, as Pb                       Ie2                    2QQ
Mercury, as Hg                    0.25                     2.5f
Selenium, as Se                  soe                     100
Total solids                      1>2                   , Q00g
H2S04 mist, as 100% acid         50
Water                                                  400  x  103

      :   dry offgas stream  containing 7 percent  sulfur  dioxide.
      typical  gas  purification  system  with  prior  coarse  dust  removal.
      be reduced  to 6 mg/Nm3  if  stainless steel  is  used.

are  Sp&^cI'rtJ" ^Ke'Uce"^^*^ SSTltt"
Can  be  objectionable in product acid.
                    5'°°° t0 10'°0° mg/Nm  if weak acid settll'"9 tanks
                               3-13

-------
mist precipitators where acid nr.t, fine participates, and volatile
metals are removed prior to entering the acid plant.  If more elaborate
cleaning is required, venturi-type scrubbers are used upstream of the
cooling towers.
     The basic steps in the contact process for the manufacture of
sulfuric acid from sulfur dioxide-bearing gases are shown in Figure 3-2.
As  noted, the offgases are cooled  and cleaned to remove particulates
and volatile metals.  Acid mist is removed  in an electrostatic mist
precipitator, and the gases are dried with  93 percent sulfuric acid.
The dry gases then pass through a  series of gas-to-gas heat  exchangers
to  heat the offgases to the optimum temperatures for  the  catalytic
conversion of  sulfur dioxide  (S02) to sulfur trioxide  (S03).   Single-contact
acid  plants use  three or  four stages of  catalytic  converters,  whereas
dual-contact plants  use one,  two,  or three  stages  of  catalyst  before
the first absorption tower.   Since the  conversion  of  sulfur dioxide  to
sulfur trioxide  is  exothermic, the converter outlet gases must be
cooled before  passing  through the absorption tower.  These outlet
gases are passed countercurrent to the  inlet gases in the heat exchangers
mentioned above.  The  sulfur trioxide  is then  absorbed by 98 percent
 sulfuric  acid  in an absorption tower  to yield  the  product.  In a
 single-contact acid plant, the remaining gases are then treated to
 remove acid  mist and spray,  and then  vented to the atmosphere.
       In a dual-contact acid plant, the gases exhausted by the first
 absorption tower are passed through a second series of heat exchangers
 and catalytic converter stages to oxidize the sulfur dioxide remaining
 in the gases.    Normally, this step employs one or  two stages of catalyst.
 The gases then  pass through  a second absorption tower, where sulfur
 trioxide is absorbed by sulfuric  acid as in the first absorption
 tower.  The waste gases are  then  treated to remove acid  mist  and
 spray, and vented to the  atmosphere.
       3.1.1.7  Factors Affecting  Performance.  The  alternative arsenic
 emission control techniques  considered  above share a  common element.
 They  all  consist of cooling  the  gas stream to condense  arsenic  (provided
 a  sufficient  quantity  of  arsenic  is present in  the gas  stream)  and
 collecting the  resultant  fume in a high efficiency collection device.
                                  3-14

-------
               GAS CLEANING
                                                                  ACID PRODUCTION
             S02 BEARING GAS
                    I
              ELECTROSTATIC
              PRECIPITATOR
              OR BAGHOUSE
to
I
 COOLING
   AND
SCRUBBING
FACILITIES
        WEAK ACID
          AND
         SOLIDS
                             DUST
ELECTRO
 STATIC
  MIST
 PRECIP
 ITATOR
                                                                                 TO ATMOSPHERE   TO ATMOSPHERE
                                                                                                       4
                                            DRYING
                                            TOWER
1
1
1
L_
L-*







HEAT
EX-
THANfiFfK





•*- -*•
•«- -*-

•« — >•





CONVERTER











                                                                                     FIRST
                                                                                   ABSORPTION
                                                                                     TOWER
                                                                               SECOND
                                                                              ABSORPTION
                                                                                TOWER
                                                              	 SINGLE CONTACT
                                                              	DOUBLE CONTACT
                                   93% ACID
                                                                                    98% ACID
                                              Figure 3-2.   Contact  Sulfuric  Acid Plant

-------
Regardless of the control  device o.iployed, the amount of achievable
arsenic emission reduction is dependent on four major factors.  These
include the concentration of arsenic trioxide present in the gas
stream, the temperature to which the gas stream is cooled prior to
collection, the allocation of sufficient time between cooling and
collection to allow condensable arsenic compounds to condense, and the
overall collection efficiency of the control device.
     As discussed in Section 3.1.1.1,  because of the volatile nature
of arsenic (As203), the arsenic inlet  concentration  and  the  operating
temperature  of  the control device are  critical  factors  in determining
its  potential effectiveness  in  controlling  the  emission of  arsenic.
The  temperature of the gas  stream to be  treated determines  the  maximum
amount of arsenic present in the gas stream which  can  exist in  the
vapor  state  and,  consequently,  the  quantity of  arsenic which can exist
 in  the solid state and be collected in a particulate control device.
As  a result, the gas  stream to  be treated must be cooled to the extent
 practicable  and adequate time must  be  allowed to  ensure that most of
 the condensable fraction of the arsenic present is condensed prior to
 entering the control  device for collection.
      Although cooling is basic to effective arsenic control, the
 presence of moisture and sulfur oxides in the  smelter offgases  introduces
 a lower temperature constraint below which further cooling  cannot be
 tolerated without incurring severe operating problems.  Because  S03 is
 hygroscopic, it will absorb moisture  at temperatures well  above  the
 moisture  dew point and form highly corrosive sulfuric  acid  mist.  The
 temperature at which acid mist formation  occurs  (acid  dew  point)  is
 highly variable, depending  on  the  S03 concentration and other  gas
 stream characteristics.   Continued operation  of  a  dry control  device
 at  or below the acid  dew point could  result in a severe corrosion
 problem due to acid  attack. Thus,  it is generally recommended that
 the operating  temperature of a dry control device be maintained 10 to
 25°C (20 to 45°F)  above the best estimate of the acid dew point.
       Little data are available on  the acid dew point of smelter offgases.
  However, practical  experience at several smelters, including the
  three ASARCO smelters and the Anaconda smelter prior to its closing,
                                  3-16

-------
   indicates that operating temperatures in the range of 100°C (21?°F) to
   HO°C(230°F) are within tolerable limits.
        The quantity or concentration of arsenic in the gas stream is
   very important in determining achievable arsenic emission reductions.
   To achieve  any arsenic  trioxide  emission reduction by condensation,
   the quantity  of arsenic  trioxide in  the  gas  stream must  be sufficiently
   high  so  that  the  resultant  arsenic trioxide  concentration at the
   control  device  operating temperature  exceeds  the predicted  saturation
   concentration.   If the arsenic trioxide  concentration  at  the control
   device operating temperature does not exceed  the predicted  saturation
   concentration, little or no emission  reduction is  achievable.   Conversely
   if the arsenic trioxide concentration greatly exceeds the predicted
   saturate concentration, arsenic emission reductions approaching the
  overall  performance capability of the control device for paniculate
  matter can be  achieved.
       The  effect of the overall  collection efficiency of the control
  device on achievable  arsenic emission  reduction  is self-evident; the
  higher the efficiency for total particulate matter, the higher  the
  efficiency for arsenic.
  3<1'2  furtive  Emission  Sources  and Controls
       Fugitive  emissions may  be characterized as emissions  which  escape
 erectly  from  the process area to  the  atmosphere  rather than  through  a
 flue or exhaust system.  They result from leakage  in and around  process
 equipment  and from material   handling and transfer  operations.
      Fugitive emissions from these sources are controlled by  local
 ventilation (i.e., use of localized hoods  or enclosures) or general
 ventilation techniques (i.e., building  evacuation) to confine and
 capture emissions.   Once  captured, the  emissions  may be vented directly
 to a particulate control  device  or combined with  process offgases
 prior to collection  in a  control device.   As with  process  sources
 some consideration must be given to cooling prior  to collection   'in
 most instances,  however, cooling occurs naturally  as a  result of  air
 dilution due to  infiltration  and,  therefore, additional  equipment is
 not  required.
     Besides the use of add-on controls, fugitive  emissions from  some
sources may be minimized or eliminated  by minor process changes and
                                3-17

-------
good operating and maintenance .ractices.  A general discussion of
these and other control  alternatives is presented in the following
subsections.
     3.1.2.1.  Local Ventilation.  Local ventilation consists  of  using
localized hoods or enclosures to confine the fugitive  emissions at  the
source,  and the use of induced air currents to  entrain and  capture  the
fugitive emissions and divert them into  an exhaust  opening.
     The design of a local  exhaust hood  involves  specifying  its  shape
and  dimensions, its position  relative  to the emission  point,  and  its
rate of  air exhaust.  The  rate of exhaust  is dependent on  the air
velocity required  and on the  size of  the imaginary  curving  area  of the
hood.   Air  contaminants  originating within this area are drawn directly
 into the exhaust  opening.   Methods  for estimating the surface area of
 an exhaust  hood  can be  found  in  standard references dealing with
 industrial  ventilation.   The  capture  velocity  is the velocity of the
 air at the  hood  face or entry plane necessary to overcome opposing air
 currents and to  capture the emission-laden air by causing it  to  flow
                       11 12
 into the exhaust hood.   '
      Hoods  can generally be classified  into three broad groups:  enclosures,
 receiving hoods, and exterior hoods.   Enclosures usually surround  most
 of  the  point of emission,  though sometimes one side may be  partially
 oi-  even completely open.   Receiving hoods are  those wherein  the  air
 contaminants are  injected  into  the hoods.  For example, the hood for a
 grinder is designed to be  in the path  of  the high  velocity dust  particles.
 Exterior hoods must capture  air contaminants that  are generated  from a
 point  outside the hood  itself,  sometimes  some  distance away.  A canopy
 hood  is a  good  example  of  an exterior hood.
       Exterior hoods are  the  most commonly used hoods  and  are by far
  the most difficult to  design since they are  the most sensitive  to
 external conditions.   For example,  a hood that works well  in a  still
  atmosphere may  be rendered completely ineffective by even a  slight
  draft through the area.
       For exhaust hoods to be effective, sufficient ventilation  must be
  applied across  the space  between the source and the  hood  so  that  all
  emissions are entrained.  This involves  overcoming the cross currents
  of indoor air which could deflect the  stream  of fugitive  emissions

                                  3-18

-------
  away  from the  hood.   The  ventilet.on rate for exhaust hoods applied to
  hot sources  must  also take  into  account  the  thermal  draft that results
  from  heat transfer  from the source to the surrounding air.   The exhaust
  rate  should  be as uniform as  possible over the entire plane of the
  hood  inlet.  Various  exhaust  hood  configurations  are illustrated in
  Figure  3-3.
       So-called "air curtains" can  be used as  complementary  capture
  devices for  local ventilation.  Air  curtains  are  basically  air jets of
  suitable  geometric configuration with  sufficient  momentum to  resist
  the forces working against  them and  maintain  their continuity  across
  the opening  they protect.    For ventilation purposes,  they are  used  as
  part of push-pull type systems, wherein the air jet  or curtain  is
  blown across the emission zone, forcing the emissions into  an  exhaust
  hood located opposite the  air curtain slot.  Typical examples  of air
 curtain applications are shown in Figure 3-4.  As indicated in  Figure  3-4(a),
 hazardous  fumes,  e.g., vinyl chloride, can be contained by the  use  of
 air curtains.  Dust control  by air curtains is shown in Figures 3-4(b)
 and 3-4(c).   A  detailed discussion of air curtains is presented in
 Section 3.1.2.7.2.
     The following  typical fugitive emission  sources at copper smelters
 can be controlled  by  local ventilation methods:
           •  Calcine  transfer  from  roaster
           •   Matte tapping
           •   Slag  tapping
           •   Converter
     3>1'2'2  General  Ventilation.  This  technique is required  whenever
 it  is  not  possible or  expedient to  use  local exhaust  hoods because  of
 their  handicapping operations, maintenance, or  surveillance  of  the
 process or equipment;  or when the local exhaust does  not significantly
 reduce air requirements or has no worker exposure  advantages.
     General ventilation has  historically  taken the form of  either
natural air changes due to wind and density differences, or mechanically
assisted air changes.   Natural  changes of air through a building  in
the absence of mechanical  ventilation occur by the action  of either  of
two forces, the  play of wind  through windows or other openings or the
                                3-19

-------
a)
     c)
                               b)
                              \u\\\r
                                                    HOOD
                                                   FLOOR
                    SOURCE
                      FLOOR
     e)
             FAN (HOOD)
                                            FLOOR OR
                                     SOURCE BENCH
                                      " A  TOP  »
                                         \\   \
           SOURCE
            Figure 3-3.   Types of Exhaust Hoods
                                               12
NOTE-   In  all cases, the source of contamination  is beyond
       the boundaries of the hood structure;  hence, control
       action is effected by inducing velocities  in the
       adjacent space.
                                 3-20

-------
   AIR CURTAIN
   (a)
                                                 AIR  CURTAIN
MATERIAL FEED
 (b)
           DUST
           COLLECTOR
                     ^B-^C^^h^ -^ _   wt_ff«^»aa»


                     fT£B&l.r2_
                                                CT
 (c)
NOTE:
           TO  DUST
           COLLECTOR
           Figure 3-4.  Uses of Air Curtains13'14
                              3-21

-------
buoyancy action resulting from a Difference in temperature between
outdoors and the inside of the building.  Mechanical ventilation is
induced by motor-driven fans and is used when the contaminants cannot
be removed by natural ventilation.  With mechanical ventilation the
contaminants can be forced out of the building by roof evacuation.
Mechanical ventilation as applied to copper smelters is discussed  in
greater detail  in Section 3.1.2.7.
     Ventilation requirements for a building  are generally defined in
terms  of  total  building  air changes per unit  time.  Although  essential
in determining  the  ventilation  requirements,  the air change  rate,  also
referred  to  as  the  ventilation  rate,  is not the only factor  which  is
important.   It  is  also important  to consider  the  rate  of  generation of
emissions in the building.
      The  main factors  on which  successful  control  of  general  ventilation
depends are the layout and  siting of  the  sources  of heat, the configuration
 of the building (number of  spans, form and shape of the roof), and the
 arrangement of ventilation  openings  in walls and roof bays.   The most
 satisfactory solutions are  obtained  when the architect and the engineer
 collaborate and consider the problems  of natural  ventilation at the
 design stage of the facility.
       3.1.2.3  Collection Devices for Fugitive Sources.   As discussed
 in Section  3.1.1.1, the most important factor affecting  the  controllability
 of arsenic  emissions  is  the  temperature of the uncontrolled  offgas
 stream.   Fortunately,  fugitive offgas  streams are  generally  lower in
 temperature than  process streams  and  seldom  have  a temperature higher
 than  93°C (200°F).   (Refer to  fugitive emission test  results in  Appendix C.)
 Cooling  of  the fugitive offgases  occurs  as  a result  of  air  dilution
 due  to ambient mixing and/or infiltration.   Also,  cooling is automatically
 achieved as a result of radiative cooling during  the passage of the
  fugitive offgases in the usually long ductwork  leading to the control
  device.   Therefore, additional gas  cooling prior to collection is not
  required for fugitive emissions.
       The captured fugitive emissions  are usually exhausted  into  existing
  process control systems.  Only two domestic copper smelters use  collection
  devices exclusively  for the control  of fugitive emissions.  ASARCO-E1  Paso
                                   3-22

-------
uses a baghouse to control fugi^ve emissions  from  the  converter
building evacuation system.  Phelps Dodge-Douglas uses  a  baghouse  to
collect fugitive emissions generated from the  roaster calcine  discharge
system.  Both baghouses were tested by EPA to  evaluate  their performance.
Results of the testing at ASARCO-E1 Paso are presented  in  Section  3.3.2.3,
and in Tables C-25 through C-28 of Appendix C.  Results of the Phelps
Dodge-Douglas testing are presented in Tables  C-53  through C-56 of
Appendix C.
     Although baghouses and electrostatic precipitators are currently
used by smelters to collect fugitive emissions, scrubbers  could also
be used, but high operating costs and water handling problems  make
their use less desirable.
     3.1.2.4  Calcine Transfer from Roaster.   The multiple-hearth  and
fluidized-bed roaster are the two basic types  of copper concentrate
roasters used.  In the case of the multi-hearth, the calcine hopper
located at the bottom of the roaster, which drops the roaster  calcine
into the transfer vehicle (larry car), is a source  of fugitive emissions.
The use of close fitting hooding between the car and hopper outlet is
difficult because of the necessity of moving the car to and from the
roaster discharge hopper.  Fluidized-bed roaster material  transfer is
carried out primarily by air conveying and secondarily  by  material
dropout; transfer is usually well contained as long as  the equipment
is maintained in good condition, resulting in  negligible  fugitive
emissions.
     A schematic of the calcine transfer fugitive emission control
system at ASARCO-Hayden, which is essentially  identical to the one at
ASARCO-Tacoma, is presented in Figure 3-5.  A  continuous  flat  apron
strip nearly 0.6 m (2 ft) wide is mounted directly  below  the row of
multi-hearth roasters at the hopper exits.  Below each  roaster, in the
apron, there are two ports connected to a 0.5 m (1.5 ft)  wide  duct.  A
matching leaf spring-loaded flat apron is mounted on the  larry car,
which is driven directly below the roaster and in line with the matching
holes on the apron connected to the roaster hopper.  One  hole  is used
for transferring the calcine from the roaster  to the larry car.  The
other holes are connected to vent lines which  go to vent  hoods  with
their own individual  draft fans (a single fan  is used at ASARCO-Tacoma).

                                3-23

-------
CO
I
ro
                                  HOPPER
                                               NOTE:  CAR TOP AND HOOD - 18 GA. C.R.S.
                                FRONT ELEV.
                                                                                         SIDE ELEV.
                            Figure  3-5.   Spring-Loaded Car Top and  Ventilation Hood, ASARCO-Hayden

-------
   Lach draft  fan has a  capacity of  -nproximately  142  Nm3/min  (5,000 scfm)
   The captured fugitives are then combined with the roaster process
   yases and treated in  a baghouse.
       In addition to the Ioca1 hooding and ventilation applied directly
   at the calcine hopper discharge point, at ASARCO-Tacoma, the calcine
   hopper area has been enclosed to for, a tunnel-like structure which is
  ventilated.   The  ventilated enclosure, coupled with the local hooding
  and ventilation applied at the actual  calcine discharge point, is very
  effective in capturing fugitive  emissions  during calcine transfer
  operates.   During  visual  observations,  no  fugitive emissions were
  observed  escaping  from the tunnel-like enclosure.15
       3-1.2.5 Matte Tapping.   Reverberatory  furnaces may have up to
  four matte tap  holes,  two  on  each side of  the  furnace.   Matte is
  tapped from  one hole at a  time and  conveyed  through  troughs  or launders
  into 4.9  to  9.2 m3 (175 to  324 cubic feet) ladles.   Typically,  fewer
  than 30 taps per furnace are  made each day, with  each tapping  operation
  taking 5  to  10 minutes.  In electric furnaces  there  are  generally  four
 •natte tap ports.  Normally, only one matte tap port  is in use  at a
  t 1 m/-v
 time.
      Copper matte from the furnace-port travels through a launder
 wh!ch dlrects the flowing matte to a point where it can be collected
 m a large ladle.  Emissions  are observable from the point of the
 matte leavlng the furnace to  the point where it settles in the ladle
      Matte tap ports  and  launders  in most smelting furnaces have
 hood.ng  systems.   Tap  port  exhaust hoods  may be of any shape, as long
 as they  are  designed  to be  close to  and  cover as much  of the emission
 area  as  possible.   Launder  hoods generally  consist of  covers  mounted
 on the launder in  sections  to  allow  manual  removal  for launder cleaning
      Schemata of  a matte  tapping fugitive  emission control  system
 used  at  the ASARCO-Tacoma smelter  are  presented  in  Figures  3-6 and
 3-7.  The matte tap hood has a  1.2 by  1.2 m  (4  by  4 ft)  square
 cross-sect10n  and  is located less  than 0.9 m  (3  ft) above the  tap
 hole   The ducts connecting the matte tap exhaust hoods  to the  1 2  m
 4 ft) dimeter main duct  are approximately 0.6 m  (2 ft) in diameter
During a  tap, 283  Nm3/nnn  (10,000 scfm) is exhausted at  the tap hole'
                                3-25

-------
OJ
I
ro
 TO BAGHOUSE
                                                       «	1T-0 DIA.-LADLE HOOD
                                                                        (MOVABLE)

                                                       40'0 x  3/16  THICK
                                            MATTE TAP
                                              HOOD
                                            50 0 x 3/16"
                                                                   REVERB.
                                                                   FURNACE
                                                         «	35'0
                           Figure 3-6.  Matte Tapping Fugitive Control  System  (Plan View),
                                                   ASARCO-Tacoma

-------
CjO
I
           CABLE TO WINCH
                            RETRACTABLE
                               HOOD
                         1/4
                         THICK
                                                           LAUNDER
                I   MATTE LADLE   '
                I	   I
                 \
                                                                                       3'-8"


                                                                                  MATTE TAP HOOD
                                    Figure 3-7.   Matte Tapping and Ladle Hoods

-------
hood and 850 Nm3/min (30,000 scf-;.  is exhausted at the ladle.  The
covered launder is 3 to 6 m (10 to  20 ft) long, 0.6 m (2 ft) wide, and
approximately 0.3 m (1 ft) deep.  The launder covers (Figure 3-8)
r.imir  in si/e from 0.(> to  1.1. m (?.  to 1»  ft) in U'n<)t,h ami h.we a
same i rcular cross-section.
     A  3.4 m (11  ft) diameter  retractable  ladle hood  is used to  capture
eriissions generated at the  ladle.  The ladle hood  is  lowered over the
ladle  prior  to  tapping and  is  raised  after the tap is  completed.
     Fran testing and  observations made  at the ASARCO-Tacoma smelter,
the matte tapping fugitive emission  capture system was  observed  to
achieve greater than  90  percent capture  efficiency.
      3.1.2.6  Slag Tapping.  Generally,  reverberatory furnaces have
just one slag  tap hole.   Typically,  slag is tapped an average  of
 30 times per day for approximately 10 to 20 minutes per tap.  In
 electric furnaces there  are usually  two slag tap ports with one  slag
 tap port in use  at a time.  The duration of a slag tap is  normally
 10 minutes.  The slag flows from the tap port, down an inclined  launder,
 and into one or  more slag  pots.  The slag  pots/ladles at various
 smelters range  in capacity from 2.8  to  17  m3  (100  to 600 cubic  feet).
       Fugitive  emission  capture techniques  for slag tapping operations
 are very similar to those  used for  matte  tapping.  Local exhaust hoods
 arc used over  slag  tap  port  areas.   Slag  launders  are  either  partially
 or completely  covered.   Design volumes  for exhaust hoods vary from one
 smelter to another,  ranging  from  566 to 850 Nm3/>nin  (20,000 to  30,000 scfm).
       A schematic of  the slag tapping fugitive emission control  system
  used  at the ASARCO-Tacoma smelter is shown in Figure 3-9.   Slag tap
  hoods are  pyramidical in shape, having a 1.2 m by 2.4 m (4 ft by 8 ft)
  rectangular cross-section.  They  are less than 0.9 m (3 feet)  above
  the tap hole.   A larger exhaust hood, 2.4 m by 4.3 m (8 ft by  14 ft),
  is situated above the slag pot transfer point.   Each launder is  covered
  with  fixed hoods.  During tapping,   142 Nm3/nin  (5,000 scfm)  is  applied
  at the tap hole  and 566  Nm3/min  (20,000  scfm) is  applied  at  the slag
  pot.   Emissions  along  the launder  run  are vented  to either of  the
  above hoods.   As with  matte  tapping,  a 90 percent capture efficiency
  should be  achievable.
                                   3-28

-------
OJ
I
r-o
vo
   5'-0

(SECTION)
                                           Figure 3-8.  Launder Cover

-------
 OJ
 I
 oo
 o
TO BAGHOUSE
                                                          I-
   REVERB.
   FURNACE
                                                    •35'-0-
                                               7'-7
40V x 3/16"
                                                        65'-0
                                                                            NOTE:
                                                                               1.   ALL  DUCT
                                                                               2.   HOODS
                                         3/16" THICK C.S
                                         3/16" THICK C.S
                            Figure 3-9.   Slag Tapping  Fugitive Control System (Plan View),
                                                    ASARCO-Tacoma

-------
       3'1'2'7  Converter Operatic,.  Primary converter hoods capture
  process emissions during converter blowing periods, except for some
  emissions that escape due to primary hood leakage.  However, during
  converter charging,  skimming,  or pouring, the mouth of the converter
  is no longer under the fixed primary hood, and extensive quantities of
  fugitive emissions escape capture by the primary hood.
       Another source  of fugitive  emissions during converting occurs
  during  the  converter holding mode.   This occurs  under normal  smelting
  operations  when material,  either slag  or blister,  cannot be immediately
  transferred  from  the converters  to  the  ladles.   During this period the
  generated emissions  are  not  evacuated  by the  primary  hood.
       There are currently  three basic alternative control  techniques
  used  to  capture fugitive emissions  during  converting.   They are:
  (1) secondary mechanical hoods;  (2) air  curtain  secondary hoods; and
  (3) general ventilation/building  evacuation.
       3.1.2.7.1.  Secondary mechanical hoods,   in normal  practice,
 primary converter hood systems are used  only when  the  converters are
 in the blowing mode.   At some smelters, secondary mechanical hoods are
 being used to capture emissions generated from the converter mouth
 during the other nodes, such  as charging, skimming, holding, and
 pouring   The flow rates handled  by these hoods range from 700 to
 2,400  Mm /min (25,000 to 85,000 scfm).
      There are three  major types  of  secondary  mechanical hoods:
      1.   Fixed type - Attached  to the primary  or  uptake hood, and
 currently used at  Phelps  Dodge-Ajo,  Phelps Dodge-Hidalgo, Phelps
 Dodge-Morenci,  and  Kennecott-Utah.
     2.   Swing-away type  -  This type is  being  used  at  the Saganoseki
 smelter  in Japan and  consists of  a swing-away  type  hood used as  a
 deflector and  a retractable type  secondary hood just above  it.
     3.  Mechanical type  - This is a combination  hood  system utilizing
 a fixed hood, movable  hood, gate  hood, and  swing-away  hood.   This
 system is a conceptual design system for  converter  fugitive  emissions
 control.
     Domestic copper smelters are  using the fixed type  of secondary
hooding to control  fugitive emissions from converters.  A typical
                                3-31

-------
converter secondary hood configu-jJon of the fixed type is shown in
Figure 3-10.  These hoods are approximately 3 in (10 ft) long, 6.4 m
(21 ft) wide, 1.7 m (5.5 ft) high and are affixed to the upper front
side of the converter primary uptake hoods.  Testing and observations
made at existing copper smelters utilizing fixed converter hoods
indicated that their effectiveness was only marginal.
     The swing-away type converter hood, sometimes  called a deflector
converter hood with a retractable secondary hood, is used at Japan's
Saganoseki  smelter.  The hood was observed to  be very  effective  during
the blowing, slagging,  and  pouring of blister  copper.     During
charging of matte  or rabbling,  capture of  the  secondary  emissions
depended solely  on  the  retractable hood  portion  of  the device.
     The converter mechanical secondary  hood  system shown  in  Figure 3-11
                                                       19
is  a conceptual  design  developed  by  an EPA consultant.    This  system
uses a fixed  hood  in combination  with a  movable  hood,  a gate hood,  and
a  swing hood.   The elliptical fixed  hood is  made of steel  and is
attached  to the primary uptake  hood  of  the converter.   Its opening is
situated  in a  manner to avoid  the hook  of the overhead crane during
converter  operations.   The upper end of  the  fixed hood is attached to
 the smoke  plenum,  which leads  to dust bins on each side of the converter.
      The elliptical movable hood (refer to Figure 3-11) is made of
 stsel  and  fits over the fixed  hood.   It  has  its own track of movement.
 In the retracted position, this hood does not extend  further, horizontally,
 than the fixed hood.    In the extended position, it mates with the  lip
 of the fixed hood to provide continuity of ducting for  secondary
 emissions.
       The elliptical gate hood  fits  under  the  fixed hood in the  retracted
 position.   In the  extended position, the  gate hood continues ducting
 of the secondary  emissions to  the movable and fixed hood.
       The frustum-shaped swing  hood  can  be rotated  180 degrees.  The
 width  at the  top  is the same as  that of the  mouth  of  the  gate  hood.
 This  hood  is  made of steel with  a castable  refractory lining.   Its
 pillar-mounting is motorized.   In the retracted position  it is  clear
 of the aisle  and  slightly  behind the converter.
       Retrofitting the  mechanical secondary  hood system to an existing
 smelter could pose a  difficult problem.  The operational  safeguards

                                  3-32

-------
                                  TO SECONDARY 4
                                    HOODING
                                   MAIN  DUCT
           T.O. RAIL
            SMOKE
            HOOD
           PLENUM
      FIXED
    SECONDARY
      HOOD
Figure 3-10.  Typical Converter Fixed Secondary
Hood
                           3-33

-------
                                                      SECONDARY HOOD DUCT
                                                        SMOKE PLENUM
                                                        SECONDARY HOOD DUCT

                                                      SECONDARY HOOD DUST BIN
         DUST BIN
         MAIN HOOD
      SECONDARY
      HOOD DUCT
                                                            MOVABLE HOOD
EOT RUNWAY
    SECONDARY HOOD
       DUST BIN
                                                           ""MOTORIZED DRIVE

                                                           SWING HOOD DURING
                                                           TAPPING OR SLAGGING
                                                           POSITION
                                                        LADLE
           Figure  3-11.
Conceptual Design for Converter Mechanical
   Secondary Hood System
                                         3-34

-------
   that would be  required to operat*  it with  a minimum  number  of  breakdowns,
   delays, and damage would require close supervision and extensive
   effort.  On the other hand, the maintenance and operation of the
   relatively inefficient fixed converter secondary hood system currently
  being used at domestic smelters and the swing-away converter secondary
  hood system,  as used in Japan, are quite simple.
       The efficiency of fixed converter secondary hoods is generally
  dependent upon the distance  of the hood from the emission source and
  on the  capture and face velocity created  by the fan at the mouth of
  the fixed  hood.   Conversely, maintenance  and durability  are enhanced
  as the  hood is  moved  farther from  the  converter mouth.   The capture
  effectiveness  of  the  fixed converter secondary  hoods  applied at the
  Phelps  Dodge-Ajo  and  Phelps  Dodge-Hidalgo smelters  was judged  to be
  low.
      The swing-away type of  secondary converter hood  again  causes
  clearance problems with the  crane hook and  cables during  collar pulling
  or matte additions. Also, rugged drive mechanisms are needed for the
  operation of the swing-away  hood and added  time  is needed to complete
  pouring and skimming operations due to the  time  involved  in  the retraction
 of the swing-away hood to allow crane access to the ladle.  However,
 good fugitive  emission control  is obtained during pouring, blowing,'
 and slagging.   During  matte  addition or rabbling, the capture efficiency
 is similar to  that of  a fixed hood.
      Since  the  converter mechanical  secondary hood  system combines a
 fixed and  a  swing-away hood in  combination with  a movable  hood   it
 should be more  effective  than either a  fixed hood or swing-away  hood
 used  alone.  The capture efficiency  of  the converter mechanical  secondary
 hood  system during  charging,  pouring, skimming,  and  blowing  is  estimated
 to  be 40, 85, 70, and  85 percent, respectively.20 Assuming  a typical
 converter operation to consist of 80 percent blowing,  15 percent
 charging, and 5 percent skimming and pouring, an  overall average
 capture efficiency for the mechanical secondary  hood system is calculated
 to be approximately 80 percent.
     3'1'2-7'2   A1r curtain  secondary hoods.   Another method of  controlling
fugitive  emissions  from copper smelter converting operations involves
                                3-35

-------
the use of an air curtain system along with a secondary hood system.
An air curtain secondary hood capture system has been installed on a
domestic primary copper smelter (see below), and such systems are
being used abroad14'21'22'23 and in other U.S. industries.
     As discussed in Section 3.1.2.1, an air curtain is a suitably
shaped air jet with sufficient momentum to resist the forces of fugitive
gas streams working against it and to maintain its continuity across
the opening it protects.  Figure 3-12 shows a schematic diagram of an
air curtain system controlling converter fugitive emissions.  Consideration
in air curtain design must be given to  secondary or  entrained flows
which  start forming as  the air curtain  jet  stream leaves  its slot  or
nozzle.   As the  entrained flows become  fully  mixed with the  air curtain
jet  stream some  distance from the  nozzle,  the hot or cold secondary
flows  are carried  from  one side of the  air  curtain jet  stream  to  the
other  where  they are  ducted  for suitable  discharge.   The  greater  the
entrained flow,  the greater  the energy  loss of  the jet.   To  minimize
energy loss,  a relatively  thick,  slow moving jet stream with a large
air  volume  is required.  A basic  rule in  the design  of air curtains is
to use the  thickest and lowest  velocity air stream  to be  projected
 across the  shortest dimension of  the opening.  Air  curtain design
methods are discussed in references 14, 25, and 26.
      The air curtain  system being used at the Onahama and Naoshima
 primary copper smelters in Japan  is shown in Figure 3-13.  The capture/
 shielding device includes two steel plate partitions, one on each side
 of the converter.  The  air jet is blown from a slot at the top of one
 of the plates across the opening  to provide  a sheet or curtain of air
 that  prevents fugitive  emissions  from  escaping.  The other  plate  is
 equipped with an exhaust hood.  The opening  allows  the crane cables  to
 move  into position above the converter mouth.
       A propeller  fan is used to push  the  air through  an  elongated slot
 on  one side  and a  backward  inclined  fan  provides suction on the  opposite
  side  to  pull  in both the  fugitive gases  and the push  air.   Captured
  gas passes  through steel  ductwork to a baghouse.  The combined  temperature
  of  the converter fugitive  gases with 100 percent of the  push  air
                                  3-36

-------
           JET  SIDE
  AIR
CURTAIN
  JET


<
••M^M


••«•*••


BAFFLE
WALL




FUG IT
X EMISS
'f '
\
V
                                                      EXHAUST SIDE
                                Alk CURTAIN
                               /
                              xx   ^;
                               CONVERTER
                              (FUME SOURCE)
                                  LADLE
                                                        BAFFLE
                                                         WALL
                                                                   TO SUCTION FAN
         Figure 3-12.   Converter  Air Curtain  Control  System
                                         3-37

-------
OJ
I
OJ
CO
                   6.25
CONVERTER
                       Figure 3-13.   Converter Air Curtain Secondary Hood,  Onahama and
                                              Naoshima Smelters

-------
  entering the ductwork on the puP  >ide of the air curtain is approximately
  80°C (180°F),  which  makes  gas cooling prior to the control  device
  unnecessary.
       The inlet air forming the air curtain above the converters at the
  Naoshima smelter  has  a  flow rate of approximately 600 Nm3/min (21,000 scfm).
  The  exhaust  hood  on  the opposite side pulls in approximately 1,000 Nm3/min
  (35,000  scfm)  of  gas  to the main system.   The capacity of the total
  pull  system  at this  smelter is  three times  this  value, or 3,000 Nm3/min
  (105,000 scfm), to accommodate  the  operation  of  three hoods  at a  time.
  According to Naoshima authorities,  the collection  efficiency of these
  hoods for fugitive emissions  is approximately  90 percent.27
      The Tamano copper  smelter  in Japan uses  a differently designed
  air curtain system along witi, a fixed hood, which  is  essentially  a
  total enclosure, for controlling fugitive emissions from  each  of  its
 three converters (usually one converter is  operated at a  time.)   A
 sketch of the air curtain system installed  at the Tamano  smelter  is
 shown in Figure 3-14.    The enclosure has  two front doors and  a
 movable roof which is slightly inclined toward the front.  The  air
 curtain duct (slot hood) is located at the top of the enclosure level
 at a  position to push air from one  side of the converter to the other.
 Ambient air  is  supplied  by  a ground fan rated  at 1,2000 Nm3/min (41,000 scfm),
 Two ventilation points (offtakes) are located  at the inside  wall of
 the enclosure  for  capture of fugitive emissions;  one is for  dilute SO
 gas and  the  other  is  for high concentration S02 gas.   The inlet of the
 ductwork  for high  SO,,  gas is located near  the  converter mouth at a
 level  below  the other  ductwork for  dilute  S02  fugitive gas.   For
 convenience, the inlet of this  ductwork is  shown  at  the top of the
 roof  in Figure  3-14.

     Subjective evaluation  of  the air  curtain  secondary hood  by visible
 observation at  the Tamano smelter indicates  the system  to  be  at least
 90 percent effective in  controlling  fugitive emissions.22
     ASARCO's Tacoma facility  recently installed  a prototype  air
curtain secondary hooding system to control  fugitive emissions  from
their  No.  4 converter.   In this system, walls are erected  to enclose
the sides  and the back of the area around  the converter mouth, with a
                                3-39

-------
                                Roof Opening
                              for  Fugitive Gas
                                                   Fugitive Gases
                                                    to Bag House
                                                          at toiler

                                                      >vabl« Boof
Air
       Fugitive Gas"
            to"    _
      Desulfurization
           Plant
                                                           Off-gases To
                                                            Acid Plant
                                                        Hz Curtain
Converter Furnace
                                              front Door
Figure 3-14.  Air Curtain  System at the Tamano Smelter
                                     3-40

-------
  portion of  the enclosure  back we1,  formed  by  the  primary  hood.   Openings
  at the top  and in the front of the  enclosure  allow  for movement  of  the
  ladle and the overhead crane cables and block.  The  edges  of  the walls
  which contact the prinary hood and  the converter  vessel are sealed  to
  contain the emission plume.
       When the converter is rolled out for charging or skimming,  the
  yate on the primary hood is moved up and away from the converter mouth
  to provide clearance for the overhead crane and ladle.  As noted
  previously,  significant  amounts  of dust-laden fumes (fugitive emissions)
  escape the primary hood  system  during these modes  of the converter
  cycle.   Heavy  fugitive emissions  are generated during the roll-out and
  roll-in  modes, when  charging  (including  cold additions),  slag  skimming,
  and blister  copper pouring  are taking  place.  The  heaviest emissions
  occur  during the  actual  rolling out  of the  converter because the
  injection  of blowing air continues during roll-out until  the molten
  bath is below the  tuyeres  to prevent plugging  by cooled material
  During these periods of operation, the air curtain is  utilized to
  capture most of the emissions which  would otherwise  escape  the primary
  system.
      Air volume control  for the system is regulated automatically  by
 dampers in the  air curtain jet,  the exhaust duct,  and the induced
 draft fan.   The dampers  are manually set for a predetermined exhaust-side
 flow  and,  when  placed in  the automatic control  position, are activated
 by  movement of  the primary hood  and converter.   When the primary hood
 is  lifted  and the  converter is  rolled out,  the  system switches  to  a
 high flow mode  to  control  the heavy fugitive emissions generated
 during  roll-out activities.  At the completion  of the converter roll-out
 operations, the converter  is rolled  in  and the  primary hood  is  lowered
 over the converter  mouth.  At this  point, the system  switches to a
 lower flow  volume which is maintained during  blowing  and holding
 periods.
     The gases that are captured by the air curtain system are  treated
 m an  ESP for particulate  removal  before being passed to the atmosphere
through  the main stack.
    Design  data for the ASARCO  system are summarized in Table  3-4.
                                3-41

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

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


Matte charging          510 (18,000 acfm)          2,322 (82,000 acfm)

Blowing                         -a                 L700 (60>000 acfffl)

Slag skimming           510 (18,000 acfm)          2,322 (82,000 acfm)

Holding                 510 (18,000 acfm)            850 (30,000 acfm)

Worst conditions6     1,020 (36,000 acfm)          4,644 (164,000  acfm)


aAir curtain  will not be used during  the blowing mode.
bWorst  conditions would consist  of either (1)  two  converters being
  charged  simultaneously or  (2)  one converter  being charged while
  another  was  being  skimmed.
                                  3-42

-------
        EPA  conducted  a  program  to  evaluate  the capture effectiveness of
   the  ASARCO-Tacoma air  curtain  secondary hood system  in  January 1983 2-
   This  program  included  arsenic  and  total particulate  sampling,  a gas
   tracer study  to determine the  system's capture  efficiency  during
   specific modes of converter operation, and observations  of visible
   emissions to assess the system's effectiveness.  The  results of this
   program are presented  in Section 3.3.2.2.
       3A'2'7'3  Biding evacuation.  As noted  previously,  ventilation
   requirements for a building evacuation system are generally defined in
   terms of air changes per unit  time.  The rate of air change method
  estimates  are based  simply on  room volume and do not consider the  rate
  of evolution of the  contaminant,  the number of heat sources, or the
  natural  draft due  to building  configuration.   For example,  a general
  ventilation  installation designed  by the  rate of air  change method
  can,  under some conditions, actually cause  the contaminant  to  be
  spread throughout  the  building, thus increasing  the volume  of  dilution
  air required  to maintain hygienic conditions.   This occurs  when the
  distribution of the  ventilation air  supply  is  poorly  controlled.
  Uncontrolled air flowing into a building, due  to negative pressure in
  the building or because of poorly designed air supply  distributors
 may not only cause recirculation of  the contaminant, but also upset
 the local  ventilation systems.   It is, therefore, important  that the
 amount of air, the location of  its entry into the building,  and  its
 direction  be  controlled.  For example, Figure 3-15 shows a controlled
 air supply  which results in a convective flow from a heat source (such
 as a  ladle  of molten  metal)  rising  to be exhausted through a roof
 ventilator.   Figure 3-16 shows  an  uncontrolled air supply which  results
 in a disrupted  rising plume and  recirculation  of the contaminant
 throughout  the  building.
     Natural  air changes take place when hot air  from  the ground level
 heat sources  rises due to its buoyancy.  If, however,  there  exists  a
 point within the building where  the temperature of the  surrounding  air
 is equal to that of the  rising  column of hot air, buoyancy is lost
Therefore,  natural  air changes  will take place only if  the temperature
of the rising  column of hot  air  is  high enough to maintain the buoyancy
                                3-43

-------
Figure 3-15.  Controlled Airflow from a Heated Source
                                                      30
 Figure 3-16.   Uncontrolled Airflow from a Heated Source
                                                         30
                             3-44

-------
   of  the  column  until  it  is  dlscha-jed  through  the roof  monitors.   In
   most  hot metal workshops this  is,  however,  not  the  case.   Hot  air
   pools of some  depth  are formed under  the  building roofs.   As a result
   air entering the building  will at  times mix turbulently with pools  of'
   contaminated air and transport it  downward  to the occupied  levels near
   the floor.
       The concept of controlled ventilation  is being employed at the
  ASARCO-E1  Paso smelter to capture emissions from the converter aisle
  Several  modifications (area isolation, vent location, air flow control,
  etc.)  were made in  order to implement this control measure at the
  facility.   The  present building evacuation rate is 16,800 Nm3/min
  (bOO.OOO scfm).  This corresponds to an air change rate of 18 changes
  per  hour.   Although  it is  impossible to quantify, the building  evacuation
  system at  El  Paso, when  properly  operated  and  maintained,  is believed
  to be  capable of  achieving  95  percent  capture.   Particulate emissions
  contained  in  the  building ventilation  gases  are  controlled  by a baghouse
  EPA  has  performed tests  on  this baghouse;  the  results are  provided in
  Section  3.3.2.3,  and  detailed  test  summaries are  presented  in Appendix  C
  Although the building evacuation  system at ASARCO-E1 Paso should be
  capable of achieving  95 percent capture, the capture effectiveness
  actually being achieved is substantially less.  Severe operating
  problems  with this system have been encountered, which have  led to
 unacceptable buildups of arsenic,  lead, and heat in the building 31
 Worker  exposure  to airborne  arsenic and lead has continuously exceeded
 the  concentration  limits  set by the Occupational  Safety  and Health
 Administration (OSHA)  (10 tfg/m3 for arsenic,  50 ^g/m3 for lead)    In
 an attempt  to  alleviate the  situation,  the  company has increased the
 openings  in  the  building  and are presently  operating  roof ventilators
 whlch discharge  directly  to  the atmosphere.   The  roof ventilators  were
 originally installed for  emergency use  only  in  the  event  of  a power
 failure.  The company  is  currently investigating  the  use  of  local
 ventilation methods (including  converter secondary  hoods) for use at
 its El Paso facility.
     3.1.2.8  Anode Furnace.   Anodes are fire refined directly from
molten blister copper at all  domestic copper snelters.  Fire  refining
                                3-45

-------
is carried out in rotary-type reining furnaces resembling Pierce-Smith
converters or in small  hearth furnaces.   The rotary-type furnace is
used at 14 of the 15 existing smelters.   ASARCO-Tacoma is the only
domestic smelter which  employs hearth-type anode furnaces.
     Fugitive arsenic emissions from anode furnaces contain little
arsenic since most of the arsenic has been eliminated in earlier
processing steps.
     ASARCO-Tacoma is the only domestic copper smelter capturing
arsenic emissions from anode furnaces.  Emissions from the hearth-type
anode furnaces  at ASARCO-Tacoma are  siphoned off and  conveyed  to  an
ESP for collection.
     A  swing-up type hood is  used for the  control of  fugitive  emissions
from the  rotary-type anode  furnaces  used  in Japan.  A schematic of
such a  hood  is  presented  in Figure  3-17.   This  hood has  a flexible
duct approximately  0.7 m  (2.25 ft)  in diameter  connected to  an arc-shaped
hood.   The hood is  approximately  4  m (13  ft) wide  and 3  m (10  ft)
long.   The quantity  of dilution  air required to cool  the gases is a
function  of the position  of the  hood above the anode  furnace mouth.
 For an anode furnace temperature  of 982°C (1,000°F),  the quantity of
dilution  air required  to  reduce  the temperature to 121°C (250°F) will
 be 198 Nm3/min (7,000  scfm).  The collection  system is designed for
 42S Nm3/min (15,000 scfm) for one anode furnace under these conditions.
      The  emissions captured by the swing-up hood occur during  the
 oxidizing and  reducing blows.  The emissions  collected during  these
 blows constitute the majority of the emissions generated.  Some emissions
 escape during  the charging of blister from the ladle and during  the
 pouring  of  refined copper  from the anode  furnace,  since  the mouth  of
 the furnace during these operations  is not under the hood system;
 these  emissions, however,  are comparatively small.
       3.1.2.9   Dust Transfer.  Handling, and Conveying.   Dust transfer,
 handling,  and  conveying  practices  vary from smelter  to  smelter.
  However, the  dust  transfer from  control  devices and  smelter flues is
  common practice at  all smelters.
       Dust collected by an  electrostatic  precipitator drops  into a
  storage  hopper beneath the unit.   A collecting conveyor, contained in
  dust-tight housing, transfers this dust  to a  storage bin from which

                                  3-46

-------
                     PNEUMATIC CYLINDER  AUTOMATIC  HOOD OPENER
     AIR LINES
                                                                      HOOD 1/4" C.S.
HOOD
	FLEXIBLE DUCT 28"-0
           Figure  3-17.   Anode  Furnace Movable  Hood
                                        3-47

-------
dust may be withdrawn as desired.   These bins are usually equipped
with dust level  indicators.   Discharge from the storage hopper is
usually through a dust-tight connection to surface transportation
units.  Normal practice is to distribute the dust on smelter feed beds
where it is sprayed with water and covered by damp concentrates or
precipitates and other furnace charge materials.
     Dust collected by a baghouse drops into a holding hopper during
the shaking cycle and is removed through a double gravity gate or
rotary  valve  and into an enclosed collector  conveyor  (generally of  the
screw type), which in turn discharges  into a closed storage  bin.
Depending  on  the nature of the collected dust  and  its  metallurgical
values,  it  is either conveyed to the  charge  mixing  area  or  is discarded.
      In scrubbers, the  dust  is removed  as  a  slurry, which  is then
dewatered  in  a  thickener  or  filter  and/or  passed to a settling  pond.
      Offgases from roasters,  smelting furnaces,  converters,  and other
pyrometallurgical  units are  conveyed  through air-tight flues and
chambers to their  ultimate  dispersal  point.   The coarser dusts  tend to
 settle in  the flues  and chambers.   All  flues and chambers  are equipped
with hoppers  spaced  at intervals  underneath.  These hoppers provide
 surge capacity and storage  for the settled dust until it is withdrawn.
 Previously, these hoppers were equipped with plain swing gates for
 discharge onto the ground,  into railroad cars, or to other  surface
 conveyance equipment.  Naturally, there was some dust loss  by this
 method.  Presently,  these hoppers are equipped with discharge gates
 which  usually automatically feed into screw or  drag  type collector
 conveyors for storage  in a central receiving bin.  The  dust is then
 conveyed  to  receiving  points.  Flue  dust, however, is a coarser  material
 and may be dumped directly onto wet  feed  belts  or  conveyed  to  a  desired
 location  by  pneumatic  conveyors.   Waste heat  boilers  and  crossover
 flues  present  special  problems because the  dust from the  smelting
 furnace is at  high  temperature and builds up  in fusions and large
 accretions.   These  must  be  removed by manual  means,  water-bomb lances,
  soot blowers,  fluxing, sand  blasting, slugger guns,  or  a combination
  of methods.   Due  to the  various  sizes of  materials dropping into the
  boiler ash pits,  this material  is handled and transported by conventional
  surface vehicle conveyance and  is usually passed through a smelter
  flux crushing system.
                                  3-48

-------
   control  IT'0  ""^  ^  ^^  ^^  <*<»"»"**  of  alternative
   control  techniques  for  arsenic emissions and  total particulate,  EPA
   conducted  emission  measurements and made visual observations at  several
   smelters including  the ASARCO-Tacoma smelter.  This section discusses
   the  results of these emission measurements and presents conclusions
   regard, ng  the performance capability of each of the control techniques
   evaluated  for process and fugitive emission sources.
   3-2-l  Process Control  Systems
       3.2.1.1^ Baahouses,.   Tests were performed at  the ASARCO-Tacoma
  and Anaconda  smelters  to evaluate the  performance of baghouses  in
  controlling arsenic  emission,,   Tests were  also performed  at  the
   naconda  smelter to  evaluate  the performance  of the baghouse  in  controlling
  total  particulate  emissions.  Two  baghouses were sampled at ASARCO-Tacoma
  and  one at  Anaconda.
       3'2-1-1-1  fo3"ouses (ASARCO-Tacomal.  Simultaneous inlet and
  outlet  arsenic emission measurements were performed by EPA  across  the
  baghouse serving the multi-hearth  roasters at ASARCO-Tacoma
               at the in,et was performed in the duct carrying emissions
 fro™ the
               ,ng process in four 10 cm (4 in.) ports on top of the

  osera 
-------
                                   Table  3-5.   ARSENIC  PERFORMANCE  DATA FOR THE
                                          ROASTER  BAGHOUSE  AT  ASARCO-TACOMA




Sample
run
1
2
3
Avg.



	 — 	
Ft



Arsenic measurements
Inlet

°C (°F)
94 (201)
85 (185)
95 (203)
91 (197)

mg/Nm
314.3
295.8
254.2
288.1

(gr/dscf)
(0.138)
(0.130)
(0.111)
(0.126)

kg/hr
92.7
88.1
79.5
86.8

(Ib/hr)
(204)
(194)
(175)
(191)
Outlet

°C(°F)
88 (191)
87 (189)
82 (180)
86 (187)

mg/Nm
0.6
0.7
1.5
0.9

(gr/dscf)
(0.0003)
(0.0003)
(0.0007)
(0.0004)

kg/hr
0.2
0.2
0.5
0.3

(Ib/hr)
(0.4)
(0.4)
(1.0)
(0.6)



Efficiency,
percent

99.8
99.8
99.4
99.7
i i- \
   Concentration  and  mass  rate  data  are  based  on  measurements  on the total  catch (front and back half).
OJ
I
en
o

-------
 include six Godfrey roasters,  ar-nic kitchens, and a metallic arsenic
 plant.   The baghouse,  which has  been  replaced with a new baghouse,
 consisted  of five  compartments,  each  having  288 homopolymer acrylic
 bags.   The bags  measured  320 cm  (126  in.)  in length and  approximately
 13 cm  (5 in.)  in diameter.   The  total  filtering area was 1,860 m2
 (20,000 ft ).   It  was  designed to  effectively treat 850  Nm3/min (30,000 scfm)
 at an air-to-cloth  ratio  of 0.63 m3/min  per  m2 (2.06 cfm/ft2).   Bag
 cleaning was performed by mechanical  shakers.   The outlet  sampling  was
 done approximately  150 m  (500 ft)  downstream  of the baghouse.   The
 test results obtained are summarized  in Table  3-6. 33
     As  indicated,  the average arsenic inlet concentrations and corresponding
mass rate at the baghouse were 2,941 mg/Nm3 (1.28  gr/dscf) and  76.3 kg/hr
 168 Ib/hr), respectively.  The outlet arsenic  loading was 60.6 mg/Nm3
(0.026 gr/dscf),  and the  mass rate was 3.3 kg/hr (7 Ib/hr).  The
average arsenic removal efficiency for this unit as indicated by these
results  was 95.7  percent.
                      chamber/baqhouse (Anar.nnriaV   Inlet and
                                             .               n  oue
 measurements were made at the spray chamber/baghouse separately for
 arsenic and total particulate emissions.  The sampling locations
 included the two inlet plenums to the spray chamber and the baghouse
 exhaust duct.   Three samples each for arsenic and particulate were
 obtained at the inlet and outlet.  The results of the arsenic emission
 measurements and particulate emission measurements are summarized  in
 Tables  3-7  and  3-8,  respectively.34
      When the tests  were  conducted,  the  smelter had  the following
 process  emission  control  configuration.   Smelting  facilities  at  the
 Anaconda smelter  consisted of  a  fluid-bed  roaster, a  single electric
 furnace, and six  converters.   Except for a portion of  the  electric
 furnace and  converter offgases [about  2,070 Nm3/min  (73,000 scfm)]
 which were diverted to an acid plant for S02 removal, process gases
 from all three major smelting operations were combined  into a main
flue duct and transported to a baghouse system for particulate removal
prior to being  discharged to the atmosphere through the main stack
                                3-51

-------
OJ
I
en
ro
                  Table 3-6.  ARSENIC PERFORMANCE DATA  FOR THE  ARSENIC  PLANT BAGHOUSE AT ASARCO-TACOMA

Sample
run
1
2b
3
Avg.
Arsenic Emissions
Tnlpf
_ — 	 	 ' 	 	 	 ~~
Arsenic kitchen
°C mg/Nm3 kg/hr
	 —
56 1,717 42.4
58 1,518 42.8
60 1,532 45.1
58 1,589 43.4
Metallic arsenic
°C mg/Nm3 kg/hr
117 1,931 42.6
112 1 0.04
97 2,125 55.9
109 1,352 32.9
Total
• — — — — —
mg/Nm3 kg/hr
^ _-- — , 	 • — 	 	 • — 	 	 — ~
3,648 85.0
1,519 42.8
3,657 101.0
2,941 76.3

Outlet
	 	
°C mg/Nm3 kg/hr
	 — —
83 69.8 3.5
72 15.8 1.0
71 96.1 5.5
76 60.6 3.3
	 _ — _ 	



Efficiency,
percent
..
95.9
97.7
94.5
95.7
'Concentration and mass rate data are uaseu Ui. „««„«.«.—-

Wing this sample run the metallic arsenic process may not have  been  operating.

-------
Table 3-7.   ARSENIC PERFORMANCE DATA FOR SPRAY CHAMBER/BAGHOUSE
                AT THE ANACONDA-ANACONDA SMELTER
San-
run
•— i. .,••.
pie — 	
°C mg/Nm3
1 274 1,071
2 269 895
3 244 687
Avg. 263 885
1 — — — • •
catch (front and back

kg/hr
276.7
236.8
186.1
232.3
uu u i e L
°C
99
102
101
101
mg/Nm
—
7.1
9.8
12.6
9.8
kg/hr

1.8
2.5
3.4
2.6
Efficiency,
percent

99.3
98.9
98.2
98.9
>Shr?f? data dre baSed °n measurements °n the total
Table 3-8. PARTICULATE PERFORMANCE DATA FOR SPRAY
AT THE ANACONDA-ANACONDA SMELTER
.—.._..., ...
Sample
run
1
2
3
Avg.
5
i"" - ,n -i — , __



CHAMBER/BAGHOUSE

Particulate Emissions
Inlet
°C g/Nm3
— 	 .... 	
281 14.76
288 13.57
302 14.08
290 14.14
Concentration and mass rat
cyclone, and filter catch
kg/hr
	 - 	 -„. ,
4,071
3,736
3,860
3,890
Outlet
°C
103
103
101
102

(front half}
g/Nm3
0.05
0.04
0.05
0.05

measuremer
kg/hr
14.6
10.0
14.7
13.1



Efficiency,
percent
99.6
99.7
99 6
99.7

its on the probe,
                          3-53

-------
The baghouse system consists of 
-------
  or pipe design consisting of 18  Actions with a total collection area
  of about 6,619 m  (71,250 ft2).  Each section contains 84 pipes measuri,
  30 cm (12 inches) in diameter and 4.6 m (15 feet)  in length.  The
  second unit is of a  plate-type design, consisting  of seven parallel
  chambers each with four fields in series and with  a total  collecti,
  area  of 7,710 r/  (82,992 ft2).   The exiting  gases,  about 7,740 actual
  m /mln  (270,000 acfm)  at 90°C  (200°F),  are discharged  through a large
  flue  to the  smelter main stack.
       Since the configuration of  the inlet duct  was  such  that it was
  not possible  to sample,  only outlet arsenic  emission measurements  were
  made  by EPA at the electrostatic precipitator.   The outlet  sampling
  was performed  in the duct 23 m (75  ft) downstream of the main  stack.
  Three sample  runs were made.  A summary of the  results obtained  at the
  outlet  is presented in Table 3-9.35

        Table 3-9.   ARSENIC EMISSIONS AT OUTLET OF REVERBERATORY
           FURNACE  ELECTROSTATIC PRECIPITATOR AT ASARCO-TACoS
	
•
Sample
run
-' "•• i • _ .—
1
2
3
Avg.
- " •• - i
""•""' ' -— - .

®T ( ®E\
*-> \ r j
• • 	 -.
105 (220)
101 (214)
87 (188)
97 (207)
Arsenic Emissions9

mg/Nm3 (gr/dscf)
38.1 (0.016)
21.0 (0.009)
9.6 (0.004)
22.9 (0.010)
	 — 	 __

kg/hr (Ib/hr)
28.7 (63.1)
11.7 (25.8)
7.1 (15.6)
17.3 (38.2)
                                      based °n
     As indicated, the average arsenic concentration and mass  rate  at
the outlet were 22.9 mg/Nm3 (0.01 gr/dscf) and 17.3 kg/hr  (38.2 Ib/hr),
respectively.
     3'2'1'2*2  sPray Chamber/Electrostatic precipitator (ASARCO-E1 Pa
Inlet and outlet arsenic emission measurements were made by EPA across
this spray chamber/ESP.   Three runs were made at each of three inlet
                                3-55

-------
locations and at one outlet location.  A summary of the test results is
                        *3 -j
presented in Table 3-10.     Measurements were also made for total
particulate emissions on  two separate occasions.  Three inlet locations
and one outlet location were sampled.   During the first test, two runs
were made at each location.  During the later test, three runs were
made only at the outlet.   Tables 3-11 and 3-12 present the test data.  >
     Smelting facilities at the ASARCO-E1 Paso smelter consist of four
multi-hearth roasters, a single reverberatory furnace, and three
converters.  When the tests were conducted, the smelter had  the  following
process  emission control configuration.  Process  gases from  the  rever-
beratory  furnace passed  through two parallel waste heat boilers  where
they were cooled to  a temperature  of about 400°C  (750°F).  The existing
gas  stream  was  then  combined with  the  roaster  offgases  in  a  large
rectangular flue.  The combined gas stream,  which averaged about
5,100  Nm3/min  (180,000 scfm),  then entered  a  spray chamber where it
was  cooled  from about  220°C (428°F) to less  than  115°C (240°F).   After
cooling, the combined  roaster  and  reverberatory gases entered an
electrostatic  precipitator for particulate  removal  prior  to  being
discharged  into the main stack.*   The  precipitator consists  of seven
 parallel chambers,  each  containing four sections, with a  total  electric
 field volume of 516 m3 (18,228 ft3).
      As indicated in Table 3-10,  the average arsenic concentration
 recorded at the inlet and outlet was 0.308 and 0.006 g/Nm   (0.13 and
 0.0026 gr/dscf), respectively.  The average mass rates were 95.9 kg/hr
 (211 Ib/hr) at the  inlet and 2.1 kg/hr  (4.6 Ib/hr) at the outlet.  The
 results indicate that the average arsenic removal efficiency of this
 unit was 97.8 percent for the three runs conducted.
      Table  3-11 presents the particulate matter  test  results obtained
 during  the  first test series conducted  on the  spray  chamber/ESP at
 ASARCO-E1  Paso.  The average particulate concentration and  mass  rate
 recorded at the  inlet were 5.11 g/Nm3 (2.23 gr/dscf)  and  1,134  kg/hr
 (2,495  Ib/hr),  respectively.  The average  particulate concentration
 and mass  rate  at the  outlet were, respectively,  0.098 g/Nm3 (0.043 gr/dscf)
  *In current practice,  reverberatory furnace gases are treated separately
   in one ESP while roaster gases  are treated in another ESP.
                                  3-56

-------
                              Table 3-10.  ARSLNIC PERFORMANCE DATA FOR SPRAY CHAMBER/

                                   ELECTROSTATIC PRECIPITATOR AT ASARCO-EL PASO
— — -
Sample
run
™
1
2
3
Avg.


Nm /min
-" ™ • - - — • — 	
5636
5983
3467
5028
Concentration and mass
back half).

In!
°C
194
191
252
212
rate

et
g/Nm3
— - — -"-• -
0.100
0.338
0.594
0.308
data are
Arseni
	
kg/hr
33.8
123.1
123.8
95.9
c emissions3
•"' •— ' -i- — — .
Nm /min
4054
4233
4490
4259
based on measurements


°C
102
104
105
104

Outlet
—
g/Nm3
0.005
0.009
0.004
0.006
on the total

-
kg/hr
1.5
3.4
1.5
2.1

Efficiency,
percent
95.6
97.2
98.8
97.8
catch (front and
en
—i

-------
I
CJ1
00
                     Table  3-11    PARTICIPATE  PERFORMANCE DATA FOR THE SPRAY CHAMBER/
                               ELECTROSTATIC PRECIPITATOR AT ASARCO-EL PASO

	
Sample
run
1
2
Avg.
Particulate Emissions
Inlet
°C
226
231
229
Concentration
filter catch
Mm /min
3,741
3,631
3,686
and mass rate
(front half).
g/Nm3
— — ~
4.54
4.87
5.11
data
kg/hr
1,022
1,248
1,134

°C
105
105
105

Nm /min
4,346
4,700
4,523
are based on measurements on
Outlet
g/Nnr
0.111
0.085
0.098
the probe,

kg/hr
40.8
33.8
37.2

Efficiency,
percent
96.1
97.3
96.7
cyclone, and

-------

                         —       —


               Particulate  Emissions'
Concentration and mass  rate data  are based on
rtlPa<;iiy-i-t-c- n^ •!•!._  	i       ,             *"'
  - - _ v .  vl . i v i i «, ^ VII I* I I ^

catch  (front half).
                     3-59

-------
and 37.2 kg/hr (81.8 Ib/hr).  The average particulate removal efficiency
measured for the two sample runs performed was 96.7 percent.
     Table 3-12 presents particulate matter test results obtained at
the outlet of the spray chamber/ESP during the second test series.
The average outlet particulate concentration and mass rate values
recorded during these latter tests were 0.15 g/Nnf5 (0.066 gr/scf) and
50.9 kg/hr (112 Ib/hr), respectively; this is about one and  one-half
times higher than the values recorded during the earlier tests.  The
higher  concentrations and mass  rates recorded in the  latter  tests were
likely  due to  the higher smelter  production rates during those  tests.
     3.2.1.3   Venturi Scrubbers  (Kennecott-Ha.yden).   Arsenic emission
measurements were conducted by  EPA  at the Kennecott-Hayden  smelter  to
evaluate the  performance of the venturi  scrubber  used in  series with  a
spray  chamber  equipped  with impingement  plates  to  preclean  and  condition
fluid-bed roaster  offgases  prior to acid manufacturing.   The roaster
 process gases  pass  through  a series of  primary  and  secondary cyclones
where an estimated 95 percent of the entrained  calcine is recovered
 and the gas stream is cooled to about 315°C (600°F).  The cyclone
 exhaust, consisting of about 565 Nm3/min (20,000 scfm) with an estimated
 dust loading of 57 g/Nm3 (25 gr/scf), then enters the venturi scrubber
 where most of the particulate matter is  collected.   Weak acid scrubbing
 liquor  is injected into the venturi throat at a rate of about 1,457 liter/min
 (385 ypm), resulting in a  pressure drop  of about 41  cm (16  in.) of
 water  across  the throat.   Gases  exiting  the scrubber then enter a
 spray-type scrubber  equipped with  perforated plates, where  they are
 humidified and cooled  to about 46°C  (115°F)  prior to being  combined
 with  the  converter process gases,  and subsequently  treated  in  a  double-
 contact acid  plant.  The  pressure  drop  across  both  scrubbers is  about
 61 cm  (24 in.)  of  water.
       Both inlet and  outlet arsenic emission  samples were obtained.
 The inlet sample  was collected upstream of  the venturi  scrubber while
  the outlet sample was  collected downstream of  the spray-type scrubber.
  It was not possible  to sample directly  downstream of the venturi
  scrubber because of  the system configuration.   Table 3-13 presents a
                         38
  summary of the results.
                                  3-60

-------
      Table 3-13.  ARSENIC PERFORMANCE DATA  FOR  VENTURI  SCRUBBER
                           AT KENNECOTT-HAYDEN
— — — _•____
Sample
run
1
2
3
Avg.
a
Arsenic emissions3
	 •• — ___ 	 	

°C
336
328
324
329
Inlet
mg/Nm
29.53
25.87
22.90
26.10

kg/hr
0.85
0.74
0.75
0.78

°C
46
44
28
39
Outlet
mg/Nm3
0.64
0.27
0.32
0.41

kg/hr
0.02
0.01
0.01
0.01
• — .
Efficiency,
percent
97.9
98.9
98.8
98.4
                                        based  on
      As  the results  indicate, the arsenic  inlet  loading  to  the  scrubber
 was quite low, averaging 26.10 mg/Nm3  (0.0114 gr/dscf) and  0.78 kg/hr
 (1.72 Ib/hr).  The outlet concentration was  very  low, averaging 0.41 mg/Nm:
 (0.0002 gr/dscf).  The arsenic mass emission rate at the outlet averaged
 0.01 kg/hr (0.023 Ib/hr).  The average collection efficiency observed
 was 98.4 percent.  It should be noted that the three inlet  runs exceeded
 isokinetic tolerances (refer to Appendix C) .  This was due  primarily
 to the large  fluctuations in the moisture content of the gas stream
 frum run to  run.   As  a result,  the actual  inlet concentration was
 probably somewhat higher  than that measured.   Consequently,  the actual
 arsenic  collection efficiency of the system is  probably  slightly
 higher than the 98.4  percent  recorded.
     3'2'1-4   Sulfuric Acid  Plant...   Tests  were  performed at the
 Kennecott-Hayden,  ASARCO-E1  Paso,  and Phelps  Dodge-Ajo smelters  to
 evaluate  the performance  of acid  plants  in  controlling arsenic  emissions.
     Gas precleaning  and  conditioning of the  smelter offgases used for
 sulfuric acid manufacturing is absolutely necessary for effective acid
 plant operation.   Both hot and cold gas cleaning devices  are used.
     3'2'1-4'1  Double-contact acid plant (Kennecott-HayriPnK  The
double-contact acid plant operated at this smelter treats a  combination
of flU1d-bed  roaster and  converter process gases.  Acid production  Ts
typically about 935 Mg/day (850 tpd)  of  93.5 percent sulfuric acid.
                                3-61

-------
After passing through a series of cyclones for calcine recovery, the
fluid-bed roaster offgases are treated in a venturi scrubber for
particulate removal  and a spray tower for hunidification and cooling.
The converter offgases are captured in water-cooled hoods, cooled to
370°C (700°F) in a gas cooler, and routed to an electrostatic precipitator
for particulate removal.  Gases exiting the precipitator enter a spray
tower, similar to that used on the roaster gas stream, where they are
humidified, cooled, and subsequently combined with the roaster gas
stream.  The combined gas stream,  totaling about 2,120 Mm /min  (75,000  scfm)
at 46°C  (115°F), then passes  through three parallel trains  of  two mist
precipitators  in series,  where acid mist  and  any remaining  particulates
are  precipitated.  The  gas  stream, which  typically contains  5  to
8 percent  S02,  then  enters  the double-contact acid plant  where it  is
dried,  the S02 converted  to S03,  and  the  S03  absorbed in  weak  acid  to
 form strong acid.
      Three arsenic  test runs were conducted  by EPA on the acid plant
 tail  gas stream (outlet).  These tests were  performed concurrently
 with those across  the venturi scrubber described  in  the preceding
 section.  The average arsenic concentration  measured was 3.43 mg/Nm
 (0.0015 gr/dscf).   The corresponding arsenic mass rate averaged 0.41 kg/hr
              TO
 (0.90 lb/hr).
      3.2.1.4.2  Double-contact acid plant (ASARCO-E1 Paso).  Offgases
 generated during converter blowing operations at  the ASARCO-E1  Paso
 smelter are treated in a 454 Hg/day (500 tpd) double-contact  sulfuric
 acid plant for S02  removal.  The  offgases are captured in  water-cooled
  hoods,  passed  through  two  parallel waste  heat boilers, and  cooled  by
  evaporative cooling  in a spray  chamber.   The cooled  gases  [about
  1,700  Nm3/min  (60,000  scfm)  at  149°C  (300°F)] then  enter an ESP for
  particulate  removal.   The  precipitator consists  of  four  parallel
  chambers, each having  four sections  in series.   The exiting gases  pass
  through a venturi  scrubber for  additional  particulate removal, are
  humidified and cooled in a pair of  packed bed scrubbers, and then are
  treated in a series of mist precipitators where water and any remaining
  particulates are removed prior to entering the acid plant.
                                  3-62

-------
        Arsenic  emission measurement; were conducted by EPA at the inlet
   to the  spray  chamber  and  at  the  acid  plant  outlet.   Three sample runs
        Table 3-14.
                                               ^DOUBLE-CONTACT
                         Arsenic emissions
                                                         Efficiency,
                                                           percent
                                        0.0002
                                        0.0031
                                        0.0011
                                        0.0004
0.022
0.355
0.126
0.038
          — •-•• -•"•• HIM jo  i a uc uaua c
        catch (front and  back half).
  Only three inlet sample runs were made.
  Average of first three runs only.

      As indicated, the measured inlet and outlet arsenic concentrations
 averaged 0.976 g/Nm3 (0.426 gr/dscf)  and 0.0015 g/Nm3 (0.0007 gr/dscf)
 respectively.   The arsenic  mass rate  averaged 96.0 kg/hr (211 lb/hr)
 at  the inlet and 0.168 kg/hr (0.370 lb/hr)  at the outlet, indicating
 an  average  arsenic removal  efficiency in excess of 99 percent.
      3'2-1-4'3   ^l^ntaji^^                            Offgases
 generated during converting  at  the Phelps Dodge-Ajo  smelter are treated
 ™  an  ESP system for particulate removal followed  by a 544  Mg/day
 (600 tpd) single-contact sulfuric  acid plant  for  S0? removal.   The
offgases pass through waste heat boilers where  they  are cooled  to
about 315°C  (600°F), enter a balloon flue, and  then  pass  through  an
electrostatic precipitator.   The precipitator consists of two  independent
horizon al  parallel units with three fields, each of which  is designed
to handle 5,490 m^/min  (210,000  acfm)  at 340°C (650°F) and 95 1 kPa
                                3-63

-------
    (13.8 psia).   The  exiting  gases  n^s  into the  scrubbing  section of the
    acid plant where they  are  treated  in  a  humidifying  tower,  a cooling
    tower, and a mist precipitator.   The  cleaned gases  are then processed
    in the acid plant.  Either 93 or 98 percent sulfuric acid  can be
    produced.
         Simultaneous inlet and outlet arsenic emission measurements were
    conducted by EPA.  Three sample runs  each were made on the inlet and
    outlet.   The results are summarized in Table 3-15.     The  offgases
    treated in the  acid plant contained a negligible amount of arsenic^
    The  measured  inlet and outlet concentrations averaged 0.00007  g/Nm
     (0.00003  gr/dscf) and 0.000016  g/Nm3 (0.000007  gr/dscf),  respectively.
    The arsenic mass  rate averaged  0.004 kg/hr  (0.009  Ib/hr)  at  the inlet
     and 0.001 kg/hr (0.0022 Ib/hr)  at  the  outlet,  indicating  an  average
     arsenic removal efficiency  of  75  percent.
                Table 3-15.   ARSENIC PERFORMANCE DATA FOR SINGLE-
                     CONTACT ACID PLANT AT PHELPS DODGE-AJO

Sample
run
1
2
3
Avg.
a
Arsenic emissions
Inlet
°C
190
182
172
181
g/Nm3
0.00008
0.00003
0.00009
0.00007
kg/hr
0.006
0.002
0.005
0.004
Outlet
°C
60
73
53
62
g/Nm3
0.000007
0.00001
0.00003
0.000016
kg/hr
0.0006
0.0007
0.0013
0.001

Efficiency,
percent
90.0
65.0
64.0
75.0
Concentration and mass rate data are based on measurements on the total  catch
 (front and back half).
      3.2.2  Fugitive Control Systems Evaluation
           3.2.2.1  Local Ventilation Techniques Applied to Calcine Discharge.
                    Matte Tapping, and Slag Tapping.  The performance capability
      of the local ventilation techniques used at the ASARCO-Tacoma smelter
      for  the control  of fugitive arsenic emissions from calcine  discharge,
      matte tapping, and slag tapping operations were evaluated.  These
       techniques  were  previously  described  in  Sections  3.1.2.4, 3.1.2.5,  and
                                       3-64

-------
  3.1.2.6, respectively.   Visual  ob-rvations were made using either EPA
  Method 22 or EPA Method 9, depending on whether the emissions observed
  were intermittent or continuous.   Method 22 is used to determine the
  occurrence of visible emissions,  while Method 9 is used to determine
  the opacity of emissions.   A summary of the visible emission data
  obtained is presented in Table  3-16.41
       3-2-2-K1  Calcine transfer.   Thirteen calcine transfer operations,
  each  averaging about  2  ninutes  in  duration,  were  observed.   The  visual
  observations  were made  using  EPA Method  22  at the opening  of the
  tunnel-like structure used  to house  the  calcine hoppers  and  larry cars
  during the  calcine transfer  (discharge)  operations.   As  the  data
  indicate, no  visible  emissions were observed  at any  time.
       3-2-2-l-2  Matte tapping.  Visible  emission  observations during
  furnace matte  tapping were also made using  EPA Method 22.  Simultaneous
  but separate  observations were made both at  the furnace  tap  port  and
  at the launder-to-ladle transfer point.  Sixteen  taps, averaging
  approximately 5.5 minutes  in duration, were observed at  the  tap  port.
 Out of the 16 observations  made at the matte tap  port, no visible
 emissions were observed 100 percent of the time on 14 of these, with
 only slight emissions  ranging from 1 to 3 percent of the time for the
 remaining two observations.   No visible emissions  were observed 100 percent
 of the time from the  launder-to-matte ladle  transfer point during all
 15 observations made  at  the transfer point.
      3-2'2<1-3  .sJag tapping.   Slag tapping  emissions were  observed
 using  both EPA Methods  22 and  9.  As  with matte tapping,  separate
 observations were made  at the  furnace  tap port location  and  at  the
 slag  launder-to-slag pot transfer point.   Results  obtained  using  EPA
 Method  22  for  eight observations at  the  slag  tap port showed  that
 visible emissions were observed about  5 percent  of the time  on the
 average, with  the highest single observation  showing  the  presence of
 visible emissions 15 percent of the time.  Visual  observations made  at
 the slag launder-to-pot transfer point indicated very poor performance,
with visible emissions being observed 72 to 99 percent of the time
over 11 slag taps.   Additional data obtained using EPA Method 9 showed
significant emissions  with  opacities as high as 50 percent.  Conversations
                                3-65

-------
                           Table 3-16.  SUMMARY OF VISIBLE  EMISSION OBSERVATION DATA FOR CAPTURE

                                SYSTEMS ON FUGITIVE  EMISSION  SOURCES AT ASARCO-TACOMA3
OJ
I
CTl
CTl



Operation
Calcine transfer system
Matte tapping:
at matte tap port and
launder
at matte discharge
into ladle
Slag tapping:
at slag tap port and
launder
at slag discharge
into pots
FPA Method 22


No. of
readings
taken
13

16

15

8
11


Average
observation
time,
min:sec
1:55

5:28

5:21

13:38
15:27
Average
percent
time
emissions
observed
on all
readings
0

0.2



5.3
88


Range of
percent
time
emissions
observed
0

0-3



0-15
72-99
EPA Method 9


Average Range of
No. of observation Average opacity
readings time, opacity, percent
taken min:sec percent observed






2 13:45 6 0-30
7 14:32 12 0-50
   aVisible emission observations made on June 24 through 26, 1980.

-------
  with smelter personnel revealed that the ventilation hood at  the  slag
  launder discharge point had been damaged when hit by a truck.  Although
  an inspection of the ventilation hood and ancillary ductwork  showed no
  apparent damage, ventilation at this location was concluded to be
  inadequate to handle the volume of emissions and fume generated.
       3'2<2-2  Fugitive Emission Controls for Converters-Air Curtain
  Secondary Hood  Capture System.
       3'2-2'2'1   Evaluation program at ASARCQ-Tarnm;..   EPA conducted an
  evaluation program  in  January  1983,  on  the  prototype  air curtain
  secondary  hood  recently installed  at ASARCO-Tacoma.29   The primary
  objective  of this test  program  was  to obtain an  estimate of  the overall
  capture  efficiency  of  the  air curtain control  system  and  also  the
  capture  efficiency  during  specific modes of  converter operation.
  Capture  efficiencies of the  system during three  complete  converter
  cycles were estimated using  two principal techniques:   (1) a gas
  tracer study using sulfur  hexafluoride (SFg) was  performed by  injecting
  the gas  into the fugitive emission plume and measuring the amount  of
  the gas captured by the secondary hooding system, and (2) detailed
 visual observations  of the hooding system performance were made concurrent
 with the gas tracer  study.
      In the gas  tracer study, SFg was injected into the controlled
 area of the air  curtain at  constant,  known  rates  of 30 to 50 cc/min
 for periods which ranged from 15 minutes  to  2 hours per injection.
 Single point samples  of the exhaust gases from the air curtain  hood
 were collected at a  downstream  sampling  location  by pulling samples
 into 15-liter, leak-free Tedlar  bags  for  onsite gas  chromatographlc
 analysis.   The air curtain  capture  efficiency was calculated  by comparing
 the  SF6 injection mass  flow  rate with  the mass  flow rate  calculated
 for  the downstream sampling  point.
     Injections of SFg gas were made at 16 sample points  through  4
 test ports  in adjacent access doors on both sides  of the converter
baffle walls.  The locations of the points are  shown in Figures 3-18
and 3-19.    In addition to the efficiency measurements made for  the
points in the primary testing area, several  tests  were performed at
injection  points  outside of  this  area (below the converter centerline) in
                                3-67

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

?  SP1 & 2
D  SP3 - 5
•  SP7 - 1?
O  SP13 - 73
                                 ELEVATION
              Figure  3-18.   SFg  Tracer  Injection  Locations
                                                  3-68

-------
                                     CONVERTER AISLE FLOOR
                                         O  INJECTION POINTS
Figure  3-19.  Tracer Injection Test Ports
                          3-69

-------
an attempt to characterize the effective capture area of the air
curtain hooding system, particularly during converter roll-out activities.
     On January 14, 1983, capture efficiencies were determined for 45
injection points in the controlled area.  The calculated mean efficiencies
by converter operational mode are presented in Table 3-17.

    Table 3-17.  AIR CURTAIN  CAPTURE EFFICIENCIES AT ASARCO-TACOMA
               USING GAS TRACER METHOD - JANUARY 14,  1983
Converter
mode
Matte charge
Cold addition
Blowing
Slag skimming
Idle
TOTAL
Number of
injections
7
3
19
9
7
45
Mean
efficiency
93.1
102.0
92.8
95.0
93^.4
93. 5a
      Calculated overall mean efficiency assumes the converter
       operation consists of 80 percent blowing and idle, 15 percent
       matte charge and cold addition, and 5 percent slag skimming.
 The overall mean capture efficiency for all modes of operation was
 93.5 percent.  With the exception of cold additions, the operating
 mode of  the converter had little effect on  capture efficiency measured,
 which ranged from 92.8  percent during blowing  to 95.0  percent during
 slag skimming.  The port through which  the  releases  of tracer gas  were
 made did not have any effect  on the calculated efficiency.   However,
  it was  found that sampling  points  tested  through  a particular port
 exhibited considerable  variation,  generally recording  higher capture
  efficiencies at positions  1 and 2  (exhaust side)  than  at positions 3
  and 4  (jet side).
       The remaining  test series of  48 injections was  performed  on
  January 17-18, 1983.   The  results  of this series are summarized in
  Table 3-18.
                                  3-70

-------
             Table 3-18.  AIR CURTAIN CAPTURE EFFICIENCIES AT
        ASARCO-TACOMA USING GAS TRACER METHOD - JANUARY 17-19, 1983
Converter
mode
Matte charge
Cold addition
Blowing
Slag skimming
Copper pour
Idle

TOTAL
Number of
injections
6
3
27
7
4

-_ _^
51

Mean
efficiency
94.2
96.7
96.7
94.3
88.5

100.0
96. 5a
       Calculated overall mean efficiency assumes the converter
       operation consists of 80 percent blowing and idle, 15 percent
                                                                 "
 The overall  mean capture efficiency for all  operational modes was
 96.5 percent.   As with the data recorded on  January 14, the operating
 mode appeared  to have no significant effect  on the individual calculated
 efficiencies,  which  ranged from 88.5 percent during copper pouring to
 96.7 percent during  both blowing and cold  additions.   However, any
 consistent,  small  variations  in the efficiencies  for  various modes,  if
 they were  present, would be difficult  to detect  in the relatively
 small  number of  test  runs  (injections)  which  were  made.   The error in
 the  calculated air curtain capture  efficiencies has been  estimated to
 be ±18 percent.    For this second  test  series, it was  also  found  that
 the  location of  the test port had no effect on efficiency, while
 exhaust-side efficiencies were  found to  be somewhat higher than jet-side
 effTciencies.  Test results from the injection points  in  these tests
 indicate that,  on the average, about 95 percent of the gases and
particulate matter in the area immediately above the converter is
likely to be  captured by the air curtain secondary hooding systen.
                                3-71

-------
     In addition to these two te-u series, a series of special injection
point tests was conducted in order to assess the effective capture
area of the secondary hood system outside the confines of the hood.
The special injection tests were performed with the injection probe at
a number of points on the perimeter of the main test area, such as
very close to the baffle wall and below the ladle during the matte
charging and cold addition modes.  Table  3-19 shows the  results of
this test  series.

    Table  3-19.  AIR CURTAIN CAPTURE  EFFICIENCIES  AT ASARCO-TACOMA
               FOR SPECIAL GAS TRACER  INJECTION  POINTS  -
                          JANUARY  18-20,  1983
Converter
mode
Matte charge
Cold addition
Blowing
Slag skimming
Copper pour
Idle
TOTAL
Number of
injections
17
6
6
28
4
8
69
Mean
efficiency
61.8
61.5
33.0
84.0
80.8
53.8
49. 4a
      Calculated overall mean efficiency assumes the converter
       operation consists of 80 percent blowing and idle, 15 percent
       matte charge and cold addition, and 5 percent slag skimming
       and copper pour.
 The overall average capture efficiency for the 69 special  injection
 points was 49.4 percent.  Unlike the first two test series, the  capture
 efficiency in the special series was sensitive to converter mode.   For
 example, the slag skimming and copper pour efficiencies  are higher,  at
 84.0  and 80.8 percent,  respectively, than  the other modes  because  of
 the position of the ladle (above the injection probe)  during  these
 modes.
       During the course  of the gas  tracer study,  from  January  18 to 22,
  1983, detailed  visual observations  were  made  of  the  performance of the

                                  3-72

-------
  air curtain control system throuah  the  various converter  operating
  modes.    The purpose was to estimate capture effectiveness  and  to
  qualitatively characterize both captured emissions and emissions
  escaping capture.
       An important benefit from these observations which could not be
  derived from the results of the gas tracer study was the conclusion
  that operating practices play a major role in the overall  performance
  of the control  system.   The crane operator who moves the ladle into
  position for charging,  skimming,  or pouring has  some latitude in the
  positioning  of  the  ladle,  the speed of  pouring,  and  the speed and
  timing of  the movement  of  the ladle to  and  from  the  converter.   Observations
  show that  slower, more  deliberate movements  and closer  positioning  of
  poured  materials  to the  converter have a definite  effect on  the  extent
  of air  curtain penetration and  spillage  into  the converter aisle.
      The practices employed by  crane and converter operators  were
  found to vary significantly throughout the observation period.  For
  example, during slag skimming,  it was observed that the rate  of pouring
  varied for individual  skims.  The capture efficiency was higher when
  the pour rate was lower, since a faster rate would sometimes  "overwhelm"
 the control system.   Also,  the position  of  the ladle was noticed to be
 important,  because when  the ladle  was positioned  close to the converter
 mouth  the capture efficiency  was enhanced.   During  matte charging, the
 wundrawal  of the crane  from  above  the converter  was  often  observed  to
 cause  "drag-out"  emissions,  especially when  the crane  was moved  immediately
 after the converter was  charged.  When the  crane was  left  in  place for
 a few moments after charging  (and the heaviest part of the  emissions
 had risen to  the  air curtain), the drag-out  emissions were  noticeably
 reduced, and the capture  effectiveness was improved.
     Only one period of blowing was evaluated  quantitatively  during
 the observations.  Some penetration of the air curtain was noticed (5
 to 10 percent) during roll-in when the blowing air first started, but
the  overall  capture efficiency was  judged to be about 95 percent
     Overall capture efficiencies for individual  matte chargings were
in the  range of  90 to  95+ percent.   Cold  additions  (adding of cool
                                3-73

-------
solidified materials) to the con"-rter frequently produced emissions
heavy enough to virtually overwhelm the capture system, especially
when a fire ignited in the converter.  Capture efficiencies were
somewhat lower overall than for matte charging, typically ranging from
80 to 95 percent.
     During slag skimming, it was observed that the rate of pouring
varied for individual skims.  For pour rates judged to be slow or
moderate, efficiencies generally ranged from 80 to 95 percent.  For
faster pour rates, the typical  efficiencies dropped to 70 to 80 percent.
     Copper pouring generally produced a moderate to heavy amount of
fume; both air curtain penetration and spillage out into the converter
aisle were very slight.  Capture efficiencies were typically 90 to
95 percent during pouring.  However, at initial roll-out prior to
pouring, the efficiency could be as low as 70 percent.
     The observation that fumes would frequently spill out into the
converter aisle (especially during slag skimmer activities) indicates
that the entire fugitive emission plume either is captured by the air
curtain exhaust or penetrates the curtain vertically (since vertically
escaping gases should contain a homogeneous concentration of the
tracer gas).  Therefore, by not accounting for emissions which escape
before approaching the air curtain, the gas tracer method probably
overpredicted the capture efficiency achieved during slag skimming
operations.
     In general, visual assessments of secondary hood capture effectiveness
correlate quite well  with the average efficiencies determined by the
gas tracer method.  As mentioned earlier, operating practices have a
significant influence on the degree of capture achieved during any
individual  converter operation, and many of the extreme observed
values can be understood in terms of the operating practices employed
during those particular operations.
                                3-74

-------
       3'2'2'2'2  Visible Emission Observations  at Tamann  SmpUpr_   All
  three converters at the Tamano smelter in Japan are equipped with  a
  fixed enclosure and air curtain system for control of fugitive emissions
  generated during various modes of converter operation.  The enclosure
  doors and roof are kept open and the air curtain system is turned  on
  during  the matte charging.  During all  other modes of the converter
  operation the doors and roof are kept closed and the air curtain
  system  is turned off.   This system is described in Section 3.1.2.7.2.
       Visible emission  observations  were made for the air curtain
  secondary hood  operated on the No.  3 converter during  day-shifts  on
  March  12  and March  13,  1980.   The  converter  is a conventional  Pierce-Smith
  design, measuring about 9  meters in  length and 4 meters  in diameter
  Observations  were made  using  EPA Methods  22  and  9, depending  on  whether
  the emissions observed  were  intermittent  or  continuous,  for the  different
  modes of  converter  operation  comprising a converter  cycle.  A discussion
  of the results obtained during each mode of converter operation follows
      Matte charqinq.  Usually three ladles of matte are  brought to  the
 converter and charged in a 10- to 30-ninute period.  Actual matte
 charging from each ladle lasts for 1 to 1.5 minutes.   The fixed enclosure
 doors  and  roof are opened;  the air  curtain systan is  turned on; and
 the ladle  of matte is brought into  the secondary hood by an overhead
 crane.   The converter is rolled down  to  an inclined position-  the
 matte  ladle is lifted up by the crane; and matte is charged into  the
 converter.   At the completion of matte charge,  the  ladle  is moved  out
 of  the enclosure  and,  if needed, another ladle  of matte  is  brought in
 After the  matte additions are  completed, the  converter  is rotated  into
 the primary  converter hood,  the roof  and doors  are  closed,  and  slag
 blowing is commenced.
     Three separate matte charges were observed using both  EPA Methods  9
 and 22 simultaneously, and one matte charge was observed  using EPA
 Method 9 only.  Visual observations  for each matte charge observed
were made only during the period when the matte was actually flowing
 into the  converter.   Results of the  visual  observations  obtained are
summarized  in Table 3-20.
                                3-75

-------
     As shown in Table 3-20,  vi-iole emissions were observed for three
individual  matte charges.   The observations ranged from 44 to 77 percent
of the time (EPA Method 22).   Although somewhat continuous, the opacity
results indicate that these emissions were generally slight, typically
ranging from 0 to 10 percent opacity, with the highest average opacity
recorded for a single matte charge being 25 percent.  When present,
the emissions appeared as small puffs which penetrated the air curtain
stream.

            Table 3-20.  VISIBLE  EMISSIONS OBSERVATION DATA
                  FOR  CONVERTER SECONDARY  HOOD SYSTEM
                DURING  MATTE  CHARGING  AT THE TAMANO SMELTER
	
Sample
run
1
2
3
4
Total
Method 22
Observation
period, min
1.5
1.25
1.75
_
4.50
Percent of time
emissions observed
44
56
77
-
60
Method 9
Observation
period, min
1.5
1.25
1.75
1.5
6.25
Average opacity
for observation
period, percent
5.0
4.0
3.0
0
2.8
Range of
individuc
readings
0 to 25
0 to 10
0 to 10
-
0 to 25
      Slag blow and copper blow.  During slag blowing and copper  blowing,
 the converter mouth is enclosed by the primary duct, and offgases  are
 directed to the acid plant.  The converter  secondary hood  doors  and
 roof are closed, and the air curtain  system is turned  off.  Fugitive
 emissions generated during  blowing as a  result of  primary  hood leaks
 are captured  inside the converter housing  and are  vented  to a baghouse
 for collection.  The slag  blow,  which is  divided into  three segments,
 lasts  for  about  150 minutes per  converter cycle  arid the copper blow
 for  about  200 minutes  per  cycle.
       Visible  emission  observations  were made using EPA Method 9 for
  the  converter hood systan  for 30 minutes during  the slag blow and for
  27 minutes during  the  copper blow.    No visible emissions  (zero  percent
                                     44
  opacity) were observed at  any time.
                                  3-76

-------
                        At  the  end  c?  each  of  the  three  slag  blow phases
   slag  is slammed  into a  ladle  and  transported  to  a  sand  bed  area for
   cooling.  Because of the  quantities  involved,  slag  is  discharged from
   the converter two times after the first slag  blow and  once  after the
   second and third.  Each slag  skim lasts for about  10 minutes.   During
   each skim, an empty ladle is brought into the enclosure by  an overhead
   crane and placed on the ground in front of the converter.   The  crane
   is moved out, and the enclosure doors and roof are closed.  The  converter
   is rolled  down,  and slag is poured into the ladle.  After the slag
  skimming is  completed,  the converter  is rotated upward  slightly, the
  enclosure  doors  are opened, and  the  slag ladle is moved out.
       Only  two skims  were observed.  The first, which lasted  11 minutes
  was  observed  using  EPA Methods 22  and 9.   The second slag  skim,  lasting
  9 minutes, was observed  using  EPA  Method  22 only.   Each observation
  period began  as  the  converter  started rolling  down  to pour the slag
  into the ladle and lasted  until the pouring was completed  and  the
  converter started rolling  up.  Daring  the  first slag skim  observed, no
  visible emissions were observed at any  time.   In contrast, during the
  second slag skim, visible  emissions were observed 100 percent  of the
  time.  Most of the time,  however, these emissions were slight, ranging
 from 5 and 10 percent opacity and consisting of small puffs  which
 escaped from the  enclosure through  a  narrow opening between  the  front
 doors and the enclosure  roof.
      Ulster  discharge.   At the end of copper  blowing,  the product
 blister is  discharged  into  a ladle  and transported to a  refining
 furnace.  Usually four ladles of  blister are filled  per  converter
 cycle.  Each  of the  first three blister pours  lasts  about 12  to
 14 minutes with the  final pour  lasting about 4  minutes.   The  time
 elapsed between each blister  pour is about  8 to 15 minutes.
     At the end of a copper blow, the  secondary hood  doors  and  roof
 are opened.   An empty ladle  is  brought  into the secondary hood  by the
 overhead crane and placed in  front of  the converter.  The crane is
moved out, and the secondary hood doors and roof are closed   The
converter is rolled down,  and blister  is poured into the ladle.   After
the blister  pour is completed, the converter is rolled up slightly
                                3-77

-------
    the hood  doors  are  opened;  the  baiter  ladle  is  taken to  the refining
    furnace by the  crane;  and  the hood  doors  and  roof are closed.
         Four blister discharges  were observed.   Both EPA Methods  22 and
    9 were used.   A summary of the  results  is presented in Table 3-21.
    Although the observation periods used in  obtaining the EPA Method 22
    data were variable (i.e.,  different start and end times), the results
    nonetheless indicate that visible emissions during blister discharge
    were generally continuous.  The EPA Method 9 data, which were obtained
    only during periods when the blister copper was  actually being poured,
    show that the visible emissions observed were somewhat more substantial
    than those observed during either matte  charging  or  slag skimming.   As
    shown  in  Table  3-21,  the highest average opacity recorded  for a  single
     blister  pour was 13 percent, with  individual opacity readings ranging
     from  0 to 35 percent.   Again,  as with  slag skimming, the emissions
     observed  generally appeared  above  the  narrow opening between  the front
     doors  of the enclosure  and the enclosure roof.

               Table 3-21.  VISIBLE EMISSIONS OBSERVATION DATA FOR
                      BLISTER  DISCHARGE AT  THE TAMANO SMELTER

Sample
Run
1
2
3
4
Total
Method
22
Observation Percent of
period, min. time emissions
observed
25a

15b
6C
15.3
42
-
86
19
49
Method 9
Observation
period, min
-
15.0
12.0
3.5
30.5
Avg. opacity
for observation
period, percent
-
6.2
13.0
3.2
8.5
Range of
individual
opacity
readings
-
0 to 30
0 to 35
0 to 25
0 to 35
Observations started when secondary hood doors opened 12 minutes prior to the
 blister discharge, during which time the converter body was hit by a vibrating ram.
Observations started with the blister discharge and continued for 3 minutes after
 completion of the blister discharge.
Observations started with the blister discharge and continued for 2-1/2 minutes
 after completion of the blister discharge.
                                      3-78

-------
        3-2.2.3
                                                                   _
   emissions from converters  and anode furnaces  at the ASARCO-E1  Paso
   smelter are captured  by evacuating  the converter building.   The building
   evacuation  systen  at  this  smelter  is  described  in Section 31273
   The  captured  fugitive  gases  are  drawn  through four openings  at  the '
   roof of the converter  building into ducts which  merge  into a main  duct
   leading to a  baghouse,  then  through fans to the  250 m  (828 ft)  main
   stack.
       The  fugitive gas  flow through the baghouse  averages  about
   14,100  Nm /min (498,000 scfm).  The baghouse consists of  12 compart-
  ments.  Normally all compartments are in operation except that one
  compartment is taken off during the cleaning cycle and another compartment
  during the maintenance cycle.  Each compartment  contains 384 Orion or
  Dacron bags,  20 cm  (8  in.)  in diameter and  6.7 m (22 ft) long,  providing
  a cloth area  of 1,644  m2(l7,700  ft2)  per  compartment.   The  total  net
  cloth area of  the baghouse  is about  19,732 m2  (212,400  ft2)    The
  baghouse was designed  to effectively treat  15,280 m3/min (540,000
  acfn,)  at 54°C  (130'F) with  an air-to-cloth ratio  of 0.91 m3/min  per m2
  (3 cfm/ft  ).   Mechanical shakers  are used for cleaning  the bags
       Inlet and outlet emission measurements for inorganic  arsenic and
  total paniculate were  conducted by EPA across the  baghouse.  During
  al  tests, converter operations were monitored and  testing was conducted
  only when one or more converters were in operation.  The arsenic
  results obtained are summarized in Table 3-22.45   As indicated,  the
 measure, inlet and  outlet arsenic  concentrations  averaged 3.27 mg/Nm3
 (0.0014 gr/scf) and  0.137 mg/Nm3 (0.00006 gr/scf), respectively   The
 arsenic mass  rate averaged 2.92 kg/hr  (6.45  Ib/hr) at the inlet  and
 O.IH   g/hr (0.244 Ib/hr) at the outlet, indicating an average arsenic
 -oval  efficiency in excess of 96 percent.   The  results  of the  particulate
 measurements obtained are summarized in Table 3-23.46  As shown  in the
 table, the  mass particulate  inlet  concentration was  low,  averaging
 only 0.062 g/Nm3 (0.027  gr/scf).  Nonetheless, the mass particulate
 em1Ssion  rate was relatively high,  averaging over 50 kg/hr  (110 Ib/hr)
The low inlet concentration is  a result of the large quantities of
dilution air associated  with  the application of general  ventilation
                                3-79

-------
Table 3-22.  ARSENIC DATA FOR CONVERTER BUILDING
              BAGHOUSE AT ASARCO-EL PASO


Sample
run
1
2
3
Avg.
Arsenic emissions
Inlet
°C
38
36
51
42
mg/Nm
6.21
2.09
1.53
3.27
kg/hr
5.51
1.96
1.31
2.92
Outlet
°C
38
37
51
42
mg/Nm
0.39
0.017
0.015
0.137
kg/hr
0.305
0.017
0.012
0.111


Efficiency,
percent
94.5
99.1
99.1
96.2
 Table 3-23.  PARTICIPATE DATA FOR CONVERTER BUILDING
               BAGHOUSE AT ASARCO-EL PASO

Sample
run
1
2
3
Avg.


Particulate emi
Inlet
°C
16
22
16
18
mg/Nm
60.3
53.3
70.5
61.3
kg/hr
44.7
46.3
61.2
50.7
ssions

Outlet
°C
20
14
16
17
mg/Nm
11.6
2.5
1.1
5.1
kg/hr
10.4
0.92
6.4
3.9

Efficiency,
percent
76.7
98.0
99.3
91.3
                            3-80

-------
   techniques.   The outlet  concentration  and  mass  rate  averaged  5.1  mg/Nra3
   and 3.9 kg/hr, respectively.  Although the collection  efficiency
   obtained over three test runs averaged only  about  90 percent,  the
   results indicate that collection efficiencies as high  as  99 percent
   are achievable.
   3.2.3  Conclusions
       3.2.3.1  Process Controls.   As discussed in Section  3.1.1.1, the
  arsenic control  devices considered as best available control for
  process sources  at primary copper smelters incorporate gas stream
  precooling  as  an  integral  part of the overall control  system.   Cooling
  is vital  to effective  control  because the relatively low saturation
  concentration  of  arsenic  trioxide at lower temperatures allows a
  greater percentage  of  the arsenic to condense into  a  form which can be
  collected in a particulate  control  device.  The  process control  systems
  tested  and  discussed in Section  3.2.1  include the spray chamber/baghouse
  system  at Anaconda; the roaster  and  arsenic plant baghouses  at ASARCO-Tacoma-
  the spray chamber/ESP at ASARCO-E1  Paso;  the  reverberatory furnace ESP
  at ASARCO-Tacoma; the venturi  scrubber at  Kennecott-Hayden;  and  the
  acid plants at ASARCO-E1 Paso, Kennecott-Hayden, and Phelps  Dodge-Ajo
  The inlet temperature to the baghouses and the electrostatic precipitators
  ranged between 80 and 115°C (180 to 240°F).
      The results  of these  tests,  in terms  of the arsenic collection
 eff1C1ences  recorded,  are  shown in Figure 3-20.  (The results of ESP
 testing  at ASARCO-Tacoma are not  presented in  Figure 3-20, because the
 data were collected  only for the  ESP outlet; and  the results  of the
 single-contact  acid  plant  at Phelps  Dodge-Ajo, are  not presented
 because  of its  measured  75 percent efficiency  due to extremely  low
 inlet arsenic loadings.)   Each  black circle  in the figure represents a
 sample run performed on  the  control  device designated  at the  bottom of
 the figure.  The average efficiencies  (designated by horizontal  bars)
 ranged from  a low of about 96 percent  (baghouse at 84°C)  to a high  of
 about 99.7 percent (baghouse at 91°C).  The data  demonstrate  that
 baghouses, ESP's, venturi scrubbers, and acid  plants can  be used  to
 provide a high level  of control for arsenic emissions.   However, as
discussed earlier,  the collection efficiency for arsenic  of any particulate
                                3-81

-------
CJ
ce
100 r
99 -
l^"^^l^i W
^J ^»ii«J


98 f-
96
„
94
93

-
^^^

y

i_
i

nr
M
y
H


'
92 [- BH - BAGHOUSE
1 SC SPRAY CHAMBER


91

an
ESP ELECTROSTATIC PRECIPITATOR
VS VENTURI SCRUBBER
AP ACID PLANT
IT INLET TEMPERATURE
	 1 	
j 	 i 	
J 	 J 	 1 	 1
          IT -
 BH
91 °C
                         BH
                        84°C
SC/BH
no°c
SC/ESP
115°C
315°C
210°C
     Figure 3-20.  Control  Device  Arsenic Collection Efficiencies
                                     3-82

-------
   techniques.   The  outlet  concentration  and  mass  rate  averaged 5.1 mg/Nm3
   and  3.9  kg/hr, respectively.  Although  the  collection  efficiency
   obtained over three test  runs averaged  only about  90 percent,  the
   results  indicate  that collection efficiencies as high  as  99  percent
   are  achievable.
   3.2.3  Conclusions
       3.2.3.1  Process Controls.   As discussed in Section  3.1.1.1, the
  arsenic control  devices considered as best  available control for
  process sources  at primary copper smelters  incorporate gas stream
  precooling  as an  integral  part of the overall control system.  Cooling
  is vital  to effective  control  because the relatively low saturation
  concentration of arsenic  trioxide at  lower temperatures allows a
  greater percentage of  the  arsenic to  condense into  a form which can be
  collected in  a particulate control  device.   The  process control systems
  tested  and  discussed in Section  3.2.1 include the spray chamber/baghouse
  system  at Anaconda; the roaster  and arsenic  plant baghouses  at  ASARCO-Tacoma-
  the spray chamber/ESP at ASARCO-E1  Paso;  the reverberatory furnace ESP
  at ASARCO-Tacoma;  the venturi scrubber at Kennecott-Hayden;  and the
  acid  plants at ASARCO-E1 Paso, Kennecott-Hayden, and  Phelps  Dodge-Ajo
  The inlet temperature to the baghouses and the electrostatic  precipitators
  ranged between 80 and 115°C (180 to 240°F).
      The results  of these  tests,  in terms of the arsenic collection
 effTciences  recorded,  are  shown in Figure 3-20.  (The results of ESP
 testing  at ASARCO-Tacoma are not  presented in Figure 3-20, because the
 data  were collected only for the  ESP outlet; and  the results of the
 single-contact acid plant  at Phelps  Dodge-Ajo, are  not presented
 because  of its measured  75  percent efficiency due to extremely low
 inlet  arsenic  loadings.)   Each  black circle  in the figure represents a
 sample run performed on  the  control  device  designated  at the bottom of
 the figure.  The average efficiencies  (designated by horizontal  bars)
 ranged from  a  low of about  96 percent  (baghouse at 84°C)  to a high of
 about  99.7 percent  (baghouse at 91°C).  The data  demonstrate  that
 baghouses, ESP's,  venturi scrubbers, and acid  plants can  be used to
 provide a high level of control for arsenic emissions.   However  as
discussed earlier,  the collection  efficiency  for arsenic of any  particulate
                                3-81

-------
98
96
95
 94
 93
 92
 91
                  8H - BAGHOUSE
                  SC - SPRAY CHAMBER
                  ESP- ELECTROSTATIC PRECIPITATOR
                  VS - VENTUR1 SCRUBBER
                  AP - ACID PLANT
                  IT - INLET TEMPERATURE
             8H

      IT _ 91°C
 BH

84°C
                               SC/BH     SC/ESP
110°C
 VS

315°C
                                                            AP-A
 Figure 3-20.  Control  Device Arsenic Collection Efficiencies
                                   3-82

-------
   control device can vary dependin; on the distribution between the
   Peculate and vapor form of the arsenic which reaches the device
   This distribution in turn depends on the arsenic concentration in ihe
   gas stream and the stream temperature.   Therefore,  any discussion of
   control  efficiency of a  particular type  of  device must consider these
   parameters  before  an  estimate  of  the expected  efficiency  can  be made '
   The process  gas  streams  entering  the control devices  tested clearly
   had sufficiently high concentrations  of  arsenic  trioxide  for  high
   control  efficiencies  to  be achieved.
       3-2-3.2  fugitive Controls,  The performance capabilities of
  capture  systems for the control of fugitive arsenic emissions fro,
  calcine discharge, matte tapping,  slag tapping, and  converter operatjo|)s
  «ere evaluated.  Estates or the  capture efficiency of these systems
  are based on the visible  emissions observations reported in the preceding
  sections and on subjective judgment.   Also eva!uated was the performance
  capability  of a baghouse  control device used to collect captured
  fugitive emissions.
       Visual  observations  made on the  local ventilation  system  applied
  to  calcine discharge operations at ASARCO-Tacoma  resulted  in no  visible
  emissions being observed  at any time  during  the observation period
  As a  result,  it is concluded that such a system is readily capable'of
  achieving a capture efficiency of 90 percent or better
      Observations conducted on the  local  ventilation  system applied to
 matte tapping operations at the same smelter showed no visible emissions
 occurring at any time at the matte  launder-to-,ad,e transfer point and
 only slight  emissions of short  duration, a maximum of 3 percent of the
 time, for any individual matte  tap  observed at the matte tap  port   I
                                                    *"»'«'  »'  "Crated
            system applied to raatte tapping  operations should achieve
a minimum capture efficiency of 90 percent.
     Similar observations made on the ventilation system serving the
slag tappmg operations at Tacoma showed substantially poorer performance
-specially at the slag ,aunder-to-slag pot transfer point where v, ™
- ssions were observed nearly !00 percent of the time during each
i-dnn-u.1  s,ag  tap.   Based  on  the results  of the visua,  observation,
                                3-83

-------
and the fact that the capture system had reportedly been damaged, it
is concluded that the slag tapping ventilation system observed at
Tacona, as it was operating at the time, should not be considered
representative of a best system of emission reduction.  Although slag
tapping operations do represent a somewhat more difficult control
situation than matte tapping, the outstanding performance demonstrated
by the matte tapping controls at Tacoma strongly suggest that a  properly
designed and operated ventilation system applied to slag tapping
operations should be capable of achieving at  least 90 percent capture.
      Capture systems evaluated for the  control of fugitive  arsenic
emissions from converter  operations  included  the prototype  air  curtain
secondary hoods  applied at  the ASARCO-Tacoma  smelter  and the Mitsui
Smelting  Company smelter  in  Tamano,  Japan,  and the general  ventilation
or building  evacuation  system  applied  at  the  ASARCO-E1  Paso copper
smelter.
      The  gas tracer  study performed  on the  converter air curtain
system at ASARCO-Tacoma indicates that approximately 95 percent of the
converter fugitive emissions from all  operating  modes is captured by
 the system.   Visual  observations  of the operation  of this  system
 verify this measured capture efficiency,  but also  indicate that operating
 practices play an important role in attaining this efficiency consistently.
 EPA has concluded that this efficiency can be achieved with a properly
 operated air curtain secondary hooding system in combination with
 proper operating practices.
      The visual  observations made at Tarnano  on the air curtain  secondary
 hood  indicate that the application  of  such a system on a conventional
 Pierce-Smith converter should result  in an overall capture efficiency
 of at  least 90  percent.  As reported  previously, no visible emissions
 were  observed during either slag blowing or  copper blowing, which  in
 combination account for  nearly 80 percent  of a  typical  converter
 cycle.   Although  emissions  were  observed penetrating  the  air curtain
 during matte charging, these  emissions were  judged  to  be  negligible,
  consisting  of small puffs  ranging  from 0 to  10  percent opacity.
  Emissions  observed  during  (slag  and blister) pouring  operations were
  somewhat more substantial,  varying  from  0  to 35 percent opacity.
                                  3-84

-------
  However,  as  previously indicated,  ihese emissions were generally
  observed  to  escape  through  a  narrow  opening  which existed between the
  enclosure  doors  and  roof.   The  smelter representatives indicated that
  a  tighter  seal between  the  doors and  roof  would  result in a  significant
  improvement.
      Conclusions regarding  the  potential effectiveness of the  building
  evacuation system used at the ASARCO-E1 Paso smelter are  based  primarily
  on engineering judgment.  Providing the building  is properly enclosed
 and adequate ventilation rates are applied, essentially 100 percent
 capture should be possible.   However, due  to the  need  for openings in
 the building  for access and  makeup air, a more conservative estimate
 of 95 percent capture is considered more reasonable.  The fact that
 this  system has  shown less than  satisfactory  performance at the El Paso
 facility highlights  the difficulty  of designing an evacuation system
 which will  provide a  controlled  ventilation air supply  for all  of the
 emission sources  in  the building.   Fugitive sources have generally
 been  found  to  be  more successfully  controlled by  the use of local
 ventilation (hooding)  techniques.
      With regard  to the collection  of  fugitive  arsenic  emissions  in  a
 control   device, the emission measurements conducted  across  the  baghouse
 facility serving  the  converter building evacuation system  at ASARCO-E1 Paso
 showed that an arsenic collection efficiency of 96 percent or higher
was readily achievable, even  though the measured inlet  concentrations
were extremely low,  averaging only 3.3 mg/Nm3 (0.0014 gr/scf)
                               3-85

-------
3.3  REFERENCES

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

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

 3.  Behrens, R.G., and G.M. Rosenblatt.  Vapor Pressure and Thermo-
     dynamics of Octahedral Arsenic Trioxide (Arsenolite).  Journal of
     Chemical Thermodynamics. 4;175-190.  1972.

 4.  Schwitzgabel, K., et. al. Trace  Element Study  at  a  Primary  Copper
     Smelter.  Prepublication Copy. U.S.  Environmental Protection
     Agency.  Contract No. 68-01-4136.  January 1978.

 5  Control  Techniques  for  Particulate Air  Pollutants.   U.S.  Environmental
     Protection  Agency.   Publication  No.  AP-51.   January 1969.   p.
     108-122.

 6.  Southern Research Institute.   An Electrostatic Precipitator
     Systems  Study.   Final  Report  to  the  National  Air Pollution Control
     Administration.   Contract No.  CPA22-69-73.   October 30,  1970.
     p.  20.

  7.   Reference 6,  p.  22.

  8   Control  Techniques for Lead  Emissions, Volume I:  Chapters  1-3.
      U.S. Environmental Protection Agency.   Research Triangle Park,
      N.C. Publication No. EPA-450/2-77-012.  December 1977.  p.  2-30.

  9   Background Information for New Source Performance Standards:
      Primary Copper, Zinc, and Lead Smelters, Volume  I:  Proposed
      Standards.  U.S. Environmental Protection Agency.  Research
      Triangle Park, N.C.  Publication No. EPA-450/2-74-002a.  October
      1974.  p.  4-10.

 10.  Reference  11, p. 4-9.

 11.  Danielson, John  A.  Air Pollution Engineering Manual.   2nd Edition.
      U.S.  Environmental  Protection Agency.  Research  Triangle  Park,
      N.C.  Publication  No.  AP-40.  May 1973.

 12.  Hemeon, W.C.L.   Plant  and Process Ventilation.   New  York.   The
      Industrial Press.   Second Edition,  1963.

  13.  Dynaforce  Corporation  Brochure.   Air  Curtains for Industry.  New
      York.  June  1978.
                                  3-86

-------
  14.  Pjwlesland, J.W.  Air Curtai,s in Control led Energy  Flows - To

                              - a?w

  15.  Katari, V., et.al.  Pacific  Environmental  Services   Tnr
 16

                                        Hood
                                              ong
     Cincinatt,, OH.  Publication  No. EPA-600/280-oJ™ May 1980.
22.  Katari, V.  and I.J. Weisenberg.  Trip Report—Visit tn uah*
                                                    "      u.

                     ,        r*   "-
    or tne A!r PollutTon Control Association.  June 25-30, 1978.
                          3-87

-------
27.   Correspondence from Mr.  Moi""  Goto, Smelter Manager, Naoshima
     Smelter,  Japan,  to Mr.  I.J. Weisenberg, Pacific Environmental
     Services, Inc.  October 1, 1978.   Capabilities of air curtain
     control  system.

28.   ASARCO Design Report.  Converter Secondary Hooding for the Tacoma
     Plant.  Prepared by ASARCO Central Engineering Dept., Salt Lake
     City, UT.  January 22, 1982.

29.   PEDCo Environmental, Inc.  Emission Test Report -  Evaluation of
     an Air Curtain Hooding System for a Primary Copper Converter,
     ASARCO,  Inc., Tacoma, Washington.  Volume  I.   EPA  Contract
     Nos. 68-03-2924 and 68-02-3546.  Preliminary Draft.  March  1983.

30   Davis, J.A.   Unidirectional  Flow Ventilation System.  Presented
     at the 104th  Annual AIME Meeting.  New York.   February  18,  1975.
     p. 3, 4.

31   Telecon.   Katari,  Vishnu, Pacific  Environmental  Services,  Inc.,
     with  Sieverson, Jim, ASARCO,  Incorporated.   February 25,  1983.
     Copper smelter  fugitive  control  system.

32.  TRW  Environmental  Engineering Division.   Emission Testing of
     ASARCO Copper Smelter,  Tacoma, Washington.  U.S.  Environmental
     Protection Agency.  EMU  Report No.  78-CUS-12.   April 1979.
     p.  4-5.

 33.   Reference 32, p.  10-12.

 34   Harris,  D.L., Monsanto Research  Corporation.  Air Pollution
      Emission Test - Particulate  and Arsenic Emission Measurements
      from a Copper Smelter.  Anaconda Mining Company, Montana.  U.S.
      Environmental Protection Agency.  EMB Report No.  77-CUS-5.
      April 18-26, 1977.  p. 5-16.

 35.  Reference 32, p.   6.

 36.  Harris, D.L., Monsanto Research Corporation,  Dayton Laboratory.
      Particulate  and Arsenic Emission Measurements From  a Copper
      Smelter.  U.S. Environmental Protection Agency.   EMB Report
      flo.  77-CUS-6.  June 20-30,  1977.  p.  18-25.

 37   TRW  Environmental  Engineering Division.   Air  Pollution Emission
      Test, ASARCO Copper Smelter,  El  Paso, Texas.  U.S.  Environmental
      Protection  Agency.  EMB  Report  No.  78-CUS-7.  April  25,  1978.
      p.  13.

 38   Larkin,  R.  and J. Steiner,  Acurex  Corporation/Aerotherm  Division.
      Arsenic Emissions at  Kennecott  Copper Corporation,  Hayden, Arizona.
      U.S. Environmental  Protection Agency Report No.  76-NFS-l.   May
       1977.   p. 2-2.
                                  3-88

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

 40.   Rooney,  Thomas   TRW Environmental  Engineering Division.  Emission

      Env?rSnmentaliepPS"DOd-e  COPPer Sme1t*r»  Ajo«  Arizona   U.S
      Contract  No. 68-02-2812?  Work^signmeTN^lS. ^Marc^l^g^
      \}.  O-D.                                                       *

 41.   Reference  15, p.  4.

 42.   Reference  29, p.  4-66.

 43.   Vervaert,  A. and  J.  Nolan, U.S.  Environmental  Protection  Aaencv
      and Puget  Sound Air  Pollution  Control Agency.  Log  Book NosT
      and 2, Observations  of Converter Secondary Hood Test  at ASARCO
      Tacoma.  January  18-22,  1983.                           M5AKLU,

44.   Reference  22, p. 3.

45.  Reference 37, p. 7-8.

46.  Reference 37, p. 4-5.
                              3-89

-------
    4.0  MODEL PLANTS, REGULATORY BASELINE, AND REGULATORY ALTERNATIVES

       Arsenic has  been listed as a hazardous  air pollutant under Section 112
  of the Clean Air  Act (National  Emission Standards  for Hazardous Air
  Pollutants).   To  protect  public health  from  unreasonable risks  associated
  with  exposure to  airborne arsenic,  standards are being  developed to
  decrease  arsenic  emissions  from primary copper  smelters  which  process
  low-arsenic  content  feed.   This  section defines  the  regulatory  baseline,
  presents  the  alternative  ways that  EPA  can regulate  arsenic  emissions
  from  the  affected sources,  and  describes  the 14  existing  low-arsenic
  throughput copper smelters  for  analysis  of the environmental, economic,
  and energy impacts of the regulatory alternatives on the  industry.
      The technical and economic  impacts of arsenic regulatory alternatives
 on primary copper smelters are measured as incremental changes  beyond
 the conditions that would  exist in the absence of NESHAP  regulations.
 These conditions are commonly referred to collectively as a baseline.
 This section  defines such  a baseline as  it applies  to the primary
 copper smelting industry in general  and  to the low-arsenic throughput
 smelters specifically.
      In the following analysis,  baseline is  chosen  to represent existing
 process and control  equipment except in  the case of  four smelters
 which  are  scheduled  to undergo modernization.  For  these four smelters
 ASARCU-Hayden,  Phelps  Dodge-Ajo,  Phelps  Dodge-Morencl, and Kennecott-Hurley
 baseline is defined  as process and control equipment  after modernization
 The  configurations of  these  smelters  after modernization  are  described
 in Section 4.3.
 4.1  REGULATORY CONSIDERATIONS
     The low-arsenic throughput copper smelters are subject to existing
and forthcoming regulations arising from the  national  ambient air
quality standards (NAAQS) for suspended particulates, sulfur oxides
                               4-1

-------
and lead; wastewater effluent lin-;Cations; hazardous waste disposal
requirements; and regulations directed at occupational safety and
health.  Compliance with some of these regulations may coincidentally
decrease arsenic emissions or affect the manner in which arsenic
emissions are discharged into the atmosphere, even though they were
not necessarily developed with that objective.  Either a change  in the
quantity of  arsenic emitted, or the manner  in which  arsenic  is emitted
(high  versus low level discharge of emissions), will  necessarily
affect the ambient  concentrations of  airborne arsenic near  these
smelters.  Furthermore,  the  cost of compliance  with  these  regulations
will  affect  the  economic viability of  individual  smelters  and the
industry as  a  whole,  and thus  the affordability of  arsenic  controls.
      For the purpose  of  relating the  timings of other regulatory
 requirements to  the timing  of  an arsenic NESHAP,  January 1986 is
 projected as the compliance date with the arsenic NESHAP.   The date is
 based upon a January 1984 promulgation for the  regulation and assumes
 that the full  2-year waiver of compliance available under the general
 provisions of 40 CFR 61 would be applied for and granted by  EPA.
       Regulations to be examined in formulating the  regulatory baseline
 include the following:
      •   National Ambient Air Quality Standards (NAAQS) under  the
           Clean Air Act (CAA) for sulfur dioxide, total suspended
           particulates, and lead;
       •   Occupational  Safety and Health Administration  (OSHA)  regulations
           for inorganic arsenic;
       •   Effluent limitation guidelines  under the  Clean  Water  Act
           (CWA);  and
       •   Hazardous  waste  disposal  regulations under the Resource
           Conservation  and  Recovery  Act (RCRA).
       It should  be noted that there  are inherent uncertainties in
  estimating  the  control  requirements  and compliance dates associated
  with these  regulations.  However,  these uncertainties aside, such
  estimates are necessary in order to make a reasonable attempt  at
  assessing the impacts of a possible arsenic NESHAP  regulation  for
  low-arsenic primary copper smelters.
                                 4-2

-------
   4.1.1   Clean  Air  Act
       Under  Section  109  of  the  Clean  Air  Act,  EPA  has  set national
   ambient air quality standards  (NAAQS)  for certain  "criteria"  pollutants
   The criteria  pollutants emitted by primary copper  smelters  are  sulfur
   dioxide, particulates, and lead.  The  NAAQS are to be met through  the
   establishment of State implementation  plans (SIP's) governing emission
   sources of these pollutants.   The plans call  for "combinations  of
  emission limitations and other measures such that the total  mix of
  these measures would result in the attainment and maintenance of air
  quality standards."
       The following subsections  address each  of the three criteria
  pollutants  emitted by  primary  copper  smelters  and  the  control  measures
  developed under  SIP's.
       4.1.1.1   Sulfur Dioxide.   The initial SIP  requirements  for  the
  control  of S02 have  undergone considerable reshaping by  the  courts  and
  by  Congress  since  first developed  by  the  States  and approved by  EPA
      Congress  passed the Clean  Air Act  Amendments of 1977 which  generally
  prohibited the use of dispersion techniques in SIP's to  meet ambient
  standards and  required the installation of constant controls for the
 meeting of NAAQS as expeditiously as practicable.  An exception  to  the
 prohibition  of dispersion techniques applied to primary copper, zinc
 or lead  smelters  which  were eligible for a primary nonferrous smelter
 order (NSO)  under Section 119  of the Clean Air Act.2  The NSO permits
 a smelter a  temporary deferral,  up  to  as late  as January 1,  1993   from
 compliance with its SIP  for S02  emission limitations,  and permits the
 smelter  to meet ambient  air quality standards  through  the use of
 constant  control  technology  in combination with  interim dispersion-
 dependent control techniques including  tall stacks  and  supplementary
 control systems.  NSO recipients are to  be determined on  a case-by-case
 basis.  To be eligible for the deferral, each smelter must, among
 other things, pass an economic test by which it demonstrates  that it
 is economically incapable of implementing constant controls.
     Table 4-1 summarizes the applicable SO  regulations and  the
compliance status  of the various smelters.   It can be seen that three
smelters  are  in  compliance with  the  applicable SIP regulation    Of
the remaining  11 smelters,  six  are  under NSO's  (and/or consent decrees)
                               4-3

-------
                     Table 4-1.   STATE  IMPLEMENTATION  PLANS  (SIP'S)  FOR SULFUR DIOXIDE
                              AFFECTING COPPER  SMELTERS  AND  COMPLIANCE STATUS
      Smelter
ASARCO-E1  Paso, TX
ASARCO-Hayden, AZ
Tennessee Chemical Co.-
  Copperhill, TN
Inspiration-Miami, AZ

Kennecott-Garfield, UT
Kennecott-Hayden, AZ

Kennecott-Hurley, NM
Kennecott-McGill, NV
Magma-San Manuel, AZ
Phelps-Dodge-Ajo, AZ

 Phelps Dodge-Douglas,  AZ
 Phelps Dodge-Hidalgo,  NM
 Phelps Dodge-Morenci,  AZ
 Copper Range-
   White Pine, MI
 	 -	 — -
 aNonferrous smelter orders.
  Regulation
131.04.01.004
R9-3-515.C.I.b
1200-3-19-.19

R9-3-515.C.I.d.

4.3.1, 4.3.2, 4.3.4
R9-3-515.C.1.C

652
8.1 and 14
R9-3-515.C.1.3
R9-3-515.C.l.e

R9-3-515.C.l.f
 652
 R9-3-515.C.l.g
 Rule 402
 Compliance Status
—_	
 Out of compliance
 Currently under NSO
 In compliance

 Stack emissions-in compliance
 Fugitive emissions-out of
 compliance
 None-SIP not yet promulgated
 Stack emissions-in compliance
 Fugitive emissions-out of
 compliance
 Currently under  NSO
 Currently under  NSO
 Currently under  NSO
 Stack  emissions-in compliance
 Fugitive  emissions-out  of
 compliance
 Currently  under  NSO
  In compliance
 Currently under  NSO
  In compliance
Reference
   3
   4
   5
   6
   4

   7
   8
   4
   4

    4
    7
    4
    9

-------
   one ,s awaiting promulgation of v. SIP,  and four are currently out of
   corapl,ance.   For the seven Arizona smelters,  the new Multi-Point
   Rollback  limits introduce  uncertainty  with  regard to control  equipment
   re,™ts.   Uncertainties  associated  with  timing  for final  achievement
   o   comphance  under  the  NSO  program make it difficult to estimate the
   effect of the  NSO's  on regulatory  baseline.   For  the  purpose  of  baseline
   analyse, the  following  assumptions were made with regard  to  probable
   effects of existing S02  regulations on an arsenic NESHAP:
       1.   The  terms of the consent decrees signed by  three of the
  smelters should bring them into compliance.
       2.   Because of uncertainties within the NSO program, requirements
  for S02 control under this  program were not  considered in the baseline
       3.    For the six smelters either  out of compliance, or of unknown'
  comp,,ance status,  it will  be assumed that future SO,  controls will  be
  implemented sometime  beyond the implementation of  arsenic controls
  -     'V1:1'2  I0ti1 S"Spe"de''  ''"''"'•"late-.  The original  date specified
  in  the Clean  Air Act  of 1970  for attaining the primary NAAQS  for  total
  suspended particulates was  July 1,  1975.  That date was  extended  by
  the Clean Air Act Amendments of 1977 to no later than  3 years after
  EPA s approval  of revised SIP's for areas not in attainment with  the
 primary NAAQS  at the earlier date.
      Table 4-2 summarizes  the current applicable TSP regulations and
 the  compliance status  of affected  smelters.   It can be seen that four
 smelters  are currently in  compliance with  the SIP's,  four are out  of
 compl,ance, four are under compliance schedules, one  is awaiting
 promulgation of  an SIP,  and  tne status of  the remaining smelter is
 unknown.
     Total suspended particulate controls  are not  included  in  the
 base]me of the  four smelters which  are out of  compliance    It  is
 assumed that the  four smelters which are under compliance schedules
wi   come into  compliance within the time  frame that an arsenic NESHAP
would be effective, therefore, the regulatory  baseline for these
smelters includes the  controls to comply with  the SIP.
                              4-5

-------
                         Table 4-2.  STATE IMPLEMENTATION PLANS (SIP'S) FOR TOTAL SUSPENDED
                           PARTICULATES AFFECTING COPPER SMELTERS AND COMPLIANCE STATUS
          Smelter
                              Regulation
                           Compliance Status
                                                                                          Reference
CT)
ASARCO-E1 Paso, TX
ASARCO-Hayden, AZ
Tennessee Chemical Co.-
  Copperhill, TN
Inspiration-Miami, AZ
Kennecott-Garfield, UT
Kennecott-Hayden, AZ
Kennecott-Hurley, NM

Kennecott-McGill, NV
Magma-San Manuel, AZ
Phelps-Dodge-Ajo, AZ
Phelps Dodge-Douglas, AZ
Phelps Dodge-Hidalgo, NM
Phelps Dodge-Morenci, AZ
111.51, 111.52
40 CFR 52.126(b)
1200-3-7-.02-(3)

40 CFR 52.126(b)
3.2.3
40 CFR 52.126(b)
506

7.2.3
40 CFR 52\126(b)
40 CFR 52.126(b)
7-1-3.6
506
7-1-3.6
     Copper  Range-White  Pine,  MI  Rules  301 and 331
Out of compliance                3
Under compliance schedule        4
In compliance                    5

In compliance                    4
None-SIP not yet promulgated     6
In compliance                    4
In compliance with mass          7
emission rate, not in
compliance with BAT
Out of compliance                8
Out of compliance                4
Under compliance schedule        4
Out of compliance                4
Under compliance schedule        7
Under compliance schedule        4
Unknown - never tested           9

-------
        4.1.1.3  Lead.  The national ambient air quality  standard  for
   lead  of  1.5 ng/m  was promulgated on October 5,  1978  (40 CFR 51 80)
   Table 4-3 presents the lead SIP compliance status  of the low-arsenic
   throughput copper smelters.  Both of the  smelters  which are under
   SIP's  are currently in compliance.   The  remaining  12 smelters are in
   States in whlch  there are either  no  nonattainment  areas for the lead
   NAAQS, or in which no lead  SIP  has been promulgated.  For purposes of
   baseline analysis, it is  assumed  that neither existing nor upcoming
   lead SIP's will  affect  arsenic  air emissions from the smelters  under
  consideration, or  cause a significant economic burden concurrent with
  an arsenic NESHAP.

  4>l-2   iiiftiif^^
      On May  5, 1978. the U.S. Occupational  Safety and Health Administration
  (OSHA)  promulgated standards for occupational  exposure  to inorganic
  arsenic.    The standards  limit occupational  exposure to 10 Mg/m3
  averaged over any 8-hour period.  The regulations require the use'of
  engineering and work practice controls.   In  the event that engineering
  controls are not  sufficient  to  reduce exposures to or below the standard
  the engineering controls must be used  to reduce exposures to the
  lowest level  achievable  and  should be  supplemented by the use of
  respirators and other  necessary  personal protective equipment.   Primary
  copper  smelters were required to monitor for violations  and  submit  a
 compliance  plan to  OSHA by December 1, 1978.
      The OSHA requirement is a concentration limit and does  not  specify
  he sources to  be controlled or the equipment to be  utilized  in  control
 Compliance  plans with OSHA will  require the application  of capture     '
 systems  on  various fugitive sources,  in addition to  worker protection
 programs such as the use of respirators, protective  work clothing
 improved housekeeping practices,  hygiene facilities  and practices,' and
 medical  surveillance.   ASARCO has recently negotiated tripartite
 agreements  with  OSHA and  the  United Steelworkers of America which
 specify  controls and work pactices  designed to decrease worker exposure
 to inorganic arsenic at the three ASARCO smelters.11'12'13  The  E1
Paso agreement   specifies the installation of local  ventilation  hoods
for slag tapping.  This requirement is incorporated into  the  baseline
                              4-7

-------
                         Table 4-3.  STATE IMPLEMENTATION PLANS  (SIP'S) FOR LEAD
                             AFFECTING COPPER SMELTERS AND COMPLIANCE STATUS
i
CO
                                                           Compliance Status
ASARCO-E1 Paso, TX

ASARCO-Hayden, AZ
Tennessee Chemical Co.-
  Copperhill, TN
Inspiration-Miami, AZ
Kennecott-Garfield,  UT
Kennecott-Hayden, AZ
Kennecott-Hurley, NM
Kennecott-McGill, NV
Magma-San Manuel, AZ
Phelps-Dodge-Ajo, AZ
Phelps Dodge-Douglas,  AZ
 Phelps Dodge-Hidalgo,  NM
 Phelps Dodge-Morenci,  AZ
 Copper Range-
   White Pine, MI
No SIP-Currently
under study
No SIP
1200-3-22-.03

No SIP
SIP recently approved
No SIP
No SIP
Article  12
No SIP
No SIP
No SIP
No SIP
 No  SIP
 No  SIP
                                                            Regulation  currently in
                                                            proposal  phase
                                                            In compliance
                                                            In compliance
4
5

4
6
4
7
8
4
4
 4
 7
 4
 9

-------
  for the El Paso smelter.  Collect-;un of the  captured  slag  tapping
  emissions is not specified in the OSHA compliance plan.  The  Hayden
  agreement   specifies that all engineering controls currently in
  existence which limit occupational arsenic exposure be used,  but does
  not require the installation of any new fugitive emissions capture
  systems at ASARCO-Hayden.  OSHA has also negotiated compliance plans
  with two Kennecott smelters  (Garfield and McGIll), but these  agreements
  have not yet  been  signed, nor is  signature imminent, and OSHA has no
  plans  to negotiate compliance plans  with any other copper smelters at
  this time.     Therefore,  capture  of  slag tapping fugitive emissions at
  ASARCO-E1  Paso  is  the only OSHA requirement  which must be incorporated
  into the  regulatory  baseline  of an arsenic NESHAP.
  4.1.3   Clean Water Act
      Water pollution  control  regulations affect  all  areas of  the
  primary copper  industry,  including mining, milling,  smelting,  refining,
  and  acid plants.  These regulations do not affect  arsenic air  emissions
  The discussion of water pollution control regulations  under the Clean
 Water Act is divided  into  those which are based  on the best practicable
 control technology (BPT) requirements [Section 301(b)(l)(A)],  and
 those which are based on the best available technology (BAT)'and best
 conventional  pollution control technology (BCT)  [Section  301(b)(2)]
      Potential  water pollution originating  from the low-arsenic throughput
 primary copper  smelters  will  be regulated by  EPA's proposed effluent
 limitations  guidelines for nonferrous  metals  manufacturing.15  Effluents
 from the smelters will  be  covered  by  two  subcategories:   primary
 copper  smelting  and metallurgical  acid plants.   The proposed regulations
 for  the  primary  copper smelting subcategory will  amend  promulgated  BAT
 (40  FR 8523) to  conform BAT to  promulgated  BPT  (45 FR 44926),  which  is
 more  stringent.  The  proposed metallurgical acid  plant  regulations  set
 forth BAT effluent mass limitations for metallurgical acid plants
 including copper smelter acid plants, based on  promulgated BPT  (45  FR
 44926).
 4<1'4  Resource Conservation and Recovery Act (RCRA)
     On May 19,  1980,  EPA promulgated regulations under the Resource
Conservation and Recovery Act  for the disposal of hazardous substances
from metallurgical  and other process  industries (45 FR 33066).   EPA
                               4-9

-------
identified the acid plant blowdr n slurry/sludge resulting from the
thickening of blowdown slurry at primary copper smelters as a hazardous
waste because of its lead and cadmium content.  However, the extent to
which acid plant blowdown will be subject to regulations for treatment
and disposal is unknown.  No regulations under RCRA will affect arsenic
air emissions, however, and it is unlikely that any of the smelters
will suffer a significant economic impact as a result of RCRA regulations.
4.2  BASELINE AND  REGULATORY ALTERNATIVES
     This section  defines in general terms the baseline  employed  for
the  impacts analysis  in  succeeding chapters of this document, and
presents  the  regulatory  alternatives along with the technology  selected
for  each  alternative.
4.2.1   Definition  of  Baseline
      For  the  purposes of the  analysis  presented  in  this  document,
 "baseline"  includes  only those  existing air  pollution  controls  which
were installed  as  of  January  1983 or were  required  by  existing  legal
 actions.   Three smelters, (Phelps Dodge-Ajo  and -Morenci,  and ASARCO-Hayden)
 currently are operating under consent  decrees  '    to  install new
 furnace configurations and  acid plant  control  equipment for S02 emissions.
 Final  compliance dates for these consent decrees have  been established
 as 1985.   Since these controls are mandated by the courts and final
 compliance will be in the near future, the "consent, decree configurations"
 were treated as though they were existing configurations, and were
 incorporated into the baseline.
      The Kennecott-Hurley smelter is currently undergoing voluntary
 upgrading of the  furnace configuration and the installation  of acid
 plant  controls.   This work has already begun and is expected to  be
 completed in the  same time frame as the smelters operating under a
 consent  decree.18 For  this  reason, the smelter configuration  after
 modernization was taken  as the baseline configuration.
 4.2.2  Description of  the Regulatory  Alternatives
      The four  regulatory alternatives  represent application  of inorganic
 arsenic  controls  on  various  emission  points  at  the smelters  and  are
 characterized  by  the control  equipment that  would  be  required  to meet
 these  levels of  control.  Alternatives II,  III,  and  IV  each  include
                                 4-10

-------
   baseline controls  but  are  others not additive, e.g., Alternative HI
   does not include what  was  required by Alternative II, etc
        Alternative II would  require the control of process arsenic
   emissions.  As discussed in Section 3.1.1.1. the arsenic collection
   efficiency of any paniculate control  device Is a function of temperature
   and  the  correspond arsenic saturation  concentration,  and the  .Lie

   sZ!TcT,B ;f the particuiar 9as  stream-   AS «-  * ^ ^ ««
   ample calculation  presented in Section 3.1.1.1,  fop  strearas  Wlth
   lower arsenic concentrations,  cooling  the  gas stream  prior to its
   entenng a particulate  control  device  would  have  no effect  on  arsenic
   removal  efficiencies.   For  these  smelters, Alternative I,  would  not
  require  installation of additional control equipment beyond the baseline
  configuration.   For those suiters with higher arsenic concentration
  m  process gas streams,  control would be effected by operating a
  paniculate control  device  at less than 121°c (250°F), with flue  gas
  cooling upstream of the control device.
       Alternative III would require the  capture and collection  of
  fugitive  arsenic  emissions fro™ converter  operations.   This alternative
  is based  on the use  of  an air curtain secondary  hood  or equivalent for
  the capture of fugitive  emissions  f™  the  converters,  and  a
 con re, d  1ce (baghou$e or  ^^  ^        ^               «*
 of the captured emissions.
      Alternative  IV would  require  the capture  and collection of fugitive
 missions  fro. calcine discharge, «tte tapping, and slag tapping
 operations  Under this  alternative the capture of the fugitive lissions

        ™     point                                   '
  a             emlSS1°''S "e1"g ™tSi t0 " P«rt'«l«te  control  device
 (baghouse or equivalent technology) for collection
     Alternative V would require the elimination  of  all  arsenic  emission,
at copper smelters.  To .cc«p,ish  this  alternative  the  sme t  s™
 e forced to process  ores  which  were virtually  free  of arsenic con
This regulatory  alternative would therefore  require  the  closure of    '
existing  primary copper smelters.
                              4-11

-------
4.3  BASELINE CONFIGURATION, BAScLINE ARSENIC EMISSIONS, AND REGULATORY
     ALTERNATIVES FOR MODEL PLANTS
     The diversity of operational parameters among the 14 low-arsenic
throughput copper smelters makes it impractical to define "generic"
model plants.  Therefore, 14 model plants were chosen, representing
the baseline configurations, as defined in Section 4.2.1, of the
14 low-arsenic copper smelters.  Section 4.3.1 presents arsenic emission
estimates, process and fugitive, for the model plant baseline configurations,
while Section 4.3.2  is a  plant-by-plant description of  the  baseline
configurations along with a summary of what would be required of  each
model plant  under each of the  regulatory alternatives.
4.3.1   Baseline  Arsenic  Emissions
     This  section presents  arsenic  emission  estimates  for the baseline
configuration of each model plant.
     4.3.1.1 Process Emissions.   Table  4-4  presents  process arsenic
emission  estimates  for  the  model  plant  baseline  configurations,  after
control.   As discussed  in Section 2.0,  process equipment exhaust gas
 arsenic content  was  read from  the arsenic  material  balances supplied
 in Appendix F.   Arsenic removal efficiency of control  devices at each
 of the 14 existing  smelters was taken either from the company-supplied
 material  balances,  by comparing indicated  inlet arsenic loading  to
 indicated emissions, or from the following control  efficiencies  determined
 during EPA's performance evaluations of various types of particulate
 control devices.  "Cold" devices (121°C, 250°F or below, inlet)  were
 deemed to be 96 percent  efficient, while "hot"  (above  150°C, 300°F,
 inlet) control   devices were assigned an arsenic removal  efficiency  of
 30  percent.  Acid plants, due  to the requirement for  extensive gas
 precleaning and conditioning,  were ascribed  a control  efficiency of
 99  percent.
      4.3.1.2  Fugitive  Emissions.   Table 4-5 presents fugitive  arsenic
 emissions  estimates for the model  plant baseline configurations.
 Emission rates  in  absence  of  control were  taken from  Table 2-17.  At
 3 of  the 14 smelters,  control  of one or more of the  fugitive gas
 streams  is practiced.   Details of baseline capture and control  of
  fugitive emissions  are provided in the  plant-by-plant discussion which
  follows.
                                 4-12

-------
                                                                          ,-onMruuj
                                                                                              Luw-HfOtNR  IHROUGHPUT
	 	 	
Sniel ter
	 	 	 — 	 	
ASARCO-E1 Paso

ASARCO-Hayden

Tennessee Chemical Co -
Copperhil)


Inspiration-
Miami
Kennecott-Garfield

Konnecott-llayden
-£=>
J_, Kennecott-Hurley
OJ
KcnnecoU-NcGIll

Magma-San Manuel

Phelps Dodye-Ajo
Phelps Dodge-
Doug las



Phelps Qodge-
Illdalgo

V helps Dodge-
llorenci


Copper Range-
Uhite Pine

Sources: (TO - MuHiht;ar
	 	 	
Emission
Source
Him
REV
CONV
ff
CONV

F8R
£F
CONV

EF
CONV
NOR
CONV

FBR
REV
CONV
FF
CONV
REV
CONV
REV
CONV
OXREV
CONV

MHR
REV
CONV


FF
COflV

OXREV

CONV

REV
CONV
tb Roaster
— 	 —-• 	
Exit Gas
Arsenic Content
kg/hr
86 ti
54.9
76.2

45.0

0.2
0.4
0.5

4.5
1.5
115.0

1.9
5.3
5.0

0.4
18.6
35.6
1.3
0.5
45.3
2.3

l.f
0.9




0.9



5. )

0 4
0.?
Mili - Fluid
i itjtii-iMi L,urrcr\ ont
Control
Type
~ 	 	 	 _
1. ESP, 2. Acid Plant
Cold ESP
1. ESP, 2. Acid Plant

1. ESP, 2. Acid Plant
1. ESP, 2. Acid Plant

1. Scrubber, 2. ESP,
3. Acid Plant
1. Scrubber, 2. ESP
3. Add Plant
1. Scrubber, 2. ESP,
3. Acid Plant

1. ESP, 2. Acid Plant
1- ESP, 2. Acid Plant
1. Cyclones, 2. ESP's
3. Acid Plant
1- ESP's, 2. Acid Plant
1. Scrubber, Z. Acid Plant
I. ESP, 2. Scrubber
3. Acid Plant

1- ESP, 2. Acid Plant
1. ESP, 2. Acid Plant
ESP
Mul ticlones
ESP
1- ESP, 2. Acid Plant
1. ESP, 2. Acid Plant
1- ESP, 2. Acid Plant


LSP

FSP


1. ESP, 2. Acid Plant
1- ESP, 2. Acid Plant


1- ESP, 2. Acid Plant
1. ESP, 2. Acid Plant


R.sl loon Flue
lied Roastpr
L 1 tKo
Equipment
Efficiency

1. 60, 2. 99
97.5
1. 96, 2 99

1. 96, 2. 99
1 97 2 95

1- 50, 2. 100, 3. 99
1- 50, Z. 100, 3. 99
I. 50, 2. 100, 3. 99

1. 30, 2. 99
1. 30, 2 99
1. 20, 2. 60, 3. 95
1 60 2 9^
1. 90, 2. 99
40
1. 40, 2. 90, 3. 99

1. 75, 2. 99
1 25 2 99

30
7

30
1-30 2 99
1. 25, 2. 90
1. 25 2 90


30
30
30


1. 30, 2. <)8
I 30 2 98


1 30 2 98
1 30 2 98


70
60
	 . 	 _

Overall Control
Efficiency
(percent)
99.9
97.5
99.9
99.9
99.9
100
100
100

99.4
98.9
98.0

98.0
99.9
40
99.9

99.9
99.9
30
7
30
99.2
92.0
92.0

30

30
30

99.9
99.9


99.9
99.9

70


Emission
Rate
kg/hr
0.04
1.4
0.03
0.16
0.04
0
0
0

0.03
0.02
1.7

0.1
0.02
3.2
0.03 '

0.0004
0.0001
12.8
33.0
1.3
0.003
3.5
0.2

1 5

0.7
2.2

0.2
0.01


0.06
0.08

0.1
0.1
CONV - Converters
  FF - Flash 'urnate
   El - Electric Furnace
  flOK - Noran.lj Reactor
OXREV - Oxygen-Sprinkle Reverberatory Furnace

-------
                   Table 4-5.  SUMMARY OF BASELINE FUGITIVE ARSENIC EMISSION ESTIMATES FOR LOW-ARSENIC
                                            THROUGHPUT PRIMARY COPPER SMELTERS
-pa

Sine) ter
ASARCO-E1 Paso





ASARCO-Hayden




Tennessee Chemical Co.-
Copperhi 11



Inspiration-
M i am i




Kennecott-Garf i eld




Kennecott-Hayden




Kenriecott-Hurley




Kennecott-McGil 1




Emission
Source
CT
ST
HT
CONV
AF
FDH
ST
MT
CONV
AF
FDM

ST
IIT
CONV
FDH

ST
MT
CONV
AF
FDH
ST
MT
CONV
AF
FDH
ST
IIT
CONV
AF
FDH
ST
HT
CONV
AF
FOH
ST
MT
CONV
AF
FDH
Emlss ion
Rate, Absence
of Control
kg/hr
0.07
0.05
0.7
11.4
1.3
0.26
0.1
1.5
6.8
2.2
0.2

0.0004
0.01
0.08
0.0006

0.01
0.08
0.23
0.08
0.002
O.OJ
0.2
0.9
1.5
0.2
0.002
0.11
0.0
0.01
0.001
0.001
0.014
0.05
0.02
0.002
0.02
0.5
5.3
0.4
0.02
Control
Equ 1 pmen t
Cold ESP
-
Baghouse
Baghouse
Baghouse
-
Cold ESP
Cold ESP
Cold ESP
-
-

-
-
-
-

-
-
-
-
-
_
-
-
-
-
_
-
-
-
-
_
-
_
-
-
_
-
-
-
-
Control
Capture
90
90
90h
75b
75
0
90
90
50
0
0

90
90
0
0

0
90
0
0
0
90
90
50
0
0
90
90
0
0
0
0
90
0
0
0
90
90
0
0
0
Efficiency
(Percent)
Collection Overall
97.5
0
97.5
96
96
0
96
96
96
0
0

0
0
0
0

0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
87.8
0
87.8
72.0
72.0
0
86.4
86.4
48.0
0
0

0
0
0
0

0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Emission
Rate
kg/hr
0.009
0.05
0.09
3.2
0.4
0.26
0.01
0.2
3.5
2.2
0.2

0.0004
0.01
0.08
0.0006

0.01
0.08
0.22
0.08
0.002
0.03
0.2
0.9
1.5
0.2
0.002
0.11
0.8
0.01
0.004
0.001
0.014
0.054
0.018
0.002
0.02
0.5
5.3
0.4
0.02

-------
                   Table 4-5.   SUMMARY OF BASELINE FUGITIVE  ARSENIC EMISSION ESTIMATES  FOR LOW-ARSENIC

                                       THROUGHPUT  PRIMARY COPPER SMELTERS  (concluded)
-Fa
 I

Oi


Smel ter
Magma- San Manuel




Phelps Dodge-Ajo




Phelps Dodge-
Douglas





Phelps Dodge-
Hidalgo




Phelps Dodye-
Morenci




Copper Range-
White Pine




Sources- CT - Calc
ST - Slag


Emi sr> ion
Sourr P
ST
rvr
CONV
AF
FDD
ST
MT
CONV
AF
ron

CT
ST
HT
COHV
AF
FDII

ST
MT
CONV
AF
FDII

ST
MT
CONV
AF
(Oil

ST
IIT
CONV
AF
FOH
ine Transfer
Tapping
Emi ss ion
Rate, Absence
of Control Control
kg/hi Equipment
0.0004
0.01
0.06
0.006
0.0007
0.01
0.06
0.3
0.1
0.01

0.005 Baghouse
0.006 ESP
0.06 ESP
0.5
0.02
0. 002

0.003
0.02
0.14
0.07
0.001

0.004
0.1
0.8
0.06
0.009

0.0002
0. 006
0.04
0.02
0.004
HT - Matte Tapping
CONV - Converter Operations


Control
Capture
90
90
0
0
0
90
90
30
0
0

90
90
90
0
0
0

90
90
30
0
0

90
90
30
0
0

90
90
0
0
0
AF - Anode


Efficiency
Collects
0
0
0
0
0
0
0
0
0
0

94.9
30
30
0
0
0

0
0
0
0
0

0
0
0
0
0

0
0
0
0
0
Furnace
FOH - Flue Dust Handlin
	

_LPercent)
on Overall
0
0
0
0
0
0
0
0
0
0

85.4
27
27
0
0
0

0
0
0
0
0

0
0
0
0
0

0
0
0
0
0

g, Transfer,

Emission
Rate
kg/hr
0.0004
0.01
0.06
0.006
0.0007
0.01
0.06
0.3
0. 1
0.01

0.0007
0.004
0.04
0.5
0.02
0.002

0.003
0.02
0.14
0.07
0.004

0.004
0.1
0.8
0.06
0.009

0.0002
0.006
0.04
0.02
0.0004

and Conveying
                     Captured by building evacuation system.

-------
4.3.2  Model  Plant Baseline ConfDurations and Regulatory Alternatives
     The following model  plant descriptions present the baseline
configuration, as defined in Section 4.2.1, for the 14 low-arsenic throughput
copper smelters.   The configurations described in this section were
those used in calculating baseline arsenic emissions, and serve as the
starting point for subsequent analyses of the incremental environmental,
energy, and economic impacts of the regulatory alternatives.
     4.3.2.1   ASARCO-E1 Paso.  The baseline configuration is presented
in Figure 4-1.  Offgases from the 4 multihearth roasters pass through
a spray chamber and electrostatic precipitator before treatment in an
acid plant.  Offgases from the single reverberator,/ furnace enter a
spray chamber, followed by passage through an electrostatic precipitator
(operated at 121°C, 250°F) and discharge from a stack.  Offgases from
the three converters are treated in a spray chamber-electrostatic
precipitator for particulate removal and a sulfuric acid plant for SO-
control.  Calcine discharge fugitive emissions are captured by a local
ventilation system and routed to the reverberatory furnace spray
chamber.  Slag tapping fugitives are presently neither captured nor
controlled at the El Paso smelter, but a recent agreement with OSHA11
will require installation of a capture system.  Thus, for purpose of
defining baseline, a slag tap fugitive emission capture system is
assumed to be installed.  Matte tapping fugitive emissions are captured
and ducted to the reverberatory furnace spray chamber.  Converter
fugitives are captured by the building evacuation system and collected
in a baghouse.  It is estimated that the building evacuation capture
efficiency is 75 percent, and that the building evacuation baghouse
particulate collection efficiency is 96 percent.
     Regulatory Alternative II would require no controls beyond baseline
at ASARCO-E1  Paso, since baseline process controls represent best
technology for arsenic control.
     Regulatory Alternative III would require installation of air
curtain secondary hoods for capture of converter fugitive emissions.
The existing building evacuation baghouse is assumed  to be used for
collecting the captured emissions.
     Regulatory Alternative IV would require collection of captured
slag tapping fugitive emissions in an effective particulate matter

                               4-16

-------
                         Roaster/Acid
                         Plant Stack
                  ACID PLANT
                        121°
                      ESP
 I
I—'
~-J
  Captured
 slag tap
fugitives
                                           35,000 acfm
                                                25° C
                                                      \
                                    Captured
                                     calcine
                                   fugitives
                   ROASTERS
          \
                                i ,Revet b St at k

                                  813,900 a
-------
control  device.  For the purpose oT analysis, it is assumed that the
existing reverberatory furnace electrostatic precipitator will suffice.
     4.3.2.2   ASARCO-Hayden.  The baseline configuration (according
to the consent decree, see Section 4.2.1) is shown in Figure 4-2.
Process offgases from the INCO flash furnace are cleaned in a spray
chamber, settling chamber, and electrostatic precipitator before
treatment in an acid plant.   Offgases from the five converters pass
through cyclones (three per converter), a spray chamber, and electrostatic
precipitator before treatment in an acid plant.  Matte and slag tap
fugitive emissions are captured by a local  ventilation system and
controlled in a cold electrostatic precipitator.  Converter fugitives
are captured in a secondary hood system (estimated to be 50 percent
efficient) and controlled in a cold electrostatic precipitator.
     Arsenic vapor pressure calculations for the fugitive emission
control  device show it to be equally efficient for arsenic and particulate
collection at the given temperature and gas stream arsenic concentration.
     Regulatory Alternative II would require no controls beyond baseline
for ASARCO-Hayden, where the baseline configuration includes best
technology for arsenic control.
     Regulatory Alternative III would require installation of air
curtain secondary hoods on the converters,  and collection of captured
emissions in the cold electrostatic precipitator used for fugitive
emission collection in the baseline configuration.
     Regulatory Alterantive IV would require no controls beyond baseline
for ASARCO-Hayden, since baseline matte and slag tapping fugitive
emission controls to be installed at the smelter are considered best
technology.
     4.3.2.3   Tennessee Chemical Co.-Copperhill.  Figure 4-3 shows
the baseline configuration for the Tennessee Chemical Company smelter.
Offgases from the fluid-bed roaster pass through a cyclone and, combined
with offgases from the electric furnace and the two converters, are
treated in either of two acid plants following standard gas cleaning.
Matte and slag tapping fugitive emissions are captured by localized
ventilation systems and vented to a stack.   Converter fugitives are
neither captured nor controlled.
                               4-18

-------
        Main Stack
      576,000 acfm
      66° C
      70.000 acfm
      666 C
ACID PLANT
    ESP
     120° C
INCO FLASH
  FURNACE
   (1)
Captured
matte/slaq tap
                                        Bypass
                                        79,500 acfm

                                        121° C
                                                  316,000 acfm
                                                  258 C
                                               ESP
                            43,700 acfm P 25° C
                     Matte
   Captured
   converter
   fugitives

273,000 acfm 0 256 C
                               118,000 acfm
                               66° C
                                                                             Acid Plant
                                                                                    137,000 acfm

                                                                                    121° C
                                                                                    216,500 acfm

                                                                                    121° C
                                                                                 ESP
                                                                                   200° C
                                                          Converters
                                                             (5)
         Figure 4-2.   ASARCO-Hayden  Smelter Baseline Configuration

-------
               iTo stack
               (114,000 acfm
               lee0 c
          Acid Plant
                             ATo  stack
                             1114,000 acfm

                             166° C
                                                     Acid  Plant
              ESP
                                                         ESP
 Cyclone
                                Scrubbers
Fluid Bed
 Roaster
Electric Furnace
                                                     iTo roof
                                                     Captured matte/
                                                     slag tap fugitive:
Matte
Converters
    (2)
          Figure  4-3.   Tennessee  Chemical  Company
                Smelter  Baseline Configuration
                                 4-20

-------
     Since process controls at the Copperhill smelter represent best
technology for arsenic, Alternative II would require no additional
controls beyond baseline.
     Alternative III would require installation of air curtain secondary
hoods on converters and a control device for collection of captured
converter fugitive emissions.
     Installation of a collection device for captured matte and slag
tapping fugitive emissions would be required under Alternative IV.
     4.3.2.4   Inspiration-Miami.  The baseline configuration for
Inspiration-Miami is depicted in Figure 4-4.  Process emissions from
both the single electric furnace and the five Hoboken converters pass
through convection coolers, cyclones, and electrostatic precipitators
before treatment in an acid plant.  Fugitive emissions from slag
tapping are neither captured nor controlled.  Matte tapping emissions
are captured by local ventilation and vented from the roof of the
furnace building.  Fugitive emissions from the converters are neither
captured nor controlled.
     Regulatory Alternative II would require no controls beyond
baseline, since the baseline configuration represents best control of
process arsenic emissions.
     Alternative III would require installation of air curtain secondary
hoods on the converters, and a particulate control device for collection
of captured converter fugitive emissions.
     Alternative IV would necessitate installation of a capture system
for slag tapping fugitive emissions as well  as a collection device for
captured matte and slag tapping fugitive emissions.
     4.3.2.5   Kennecott-Garfield.  Kennecott-Garfield's baseline
configuration is shown in Figure 4-5.  Combined process offgases  from
the three Noranda reactors and the four converters pass through electrostatic
precipitators (6 in series) and are then sent to any of four acid
plants.  Matte tap and slag tap fugitive emissions are captured by
local ventilation systems and ducted to the  main stack.  Converter
fugitives are captured by a secondary hooding system (assumed to
50 percent efficient) and vented to the main stack.  In addition, the
reactor and converter buildings are each ventilated by a fugitive
emission roof capture system designed to handle emissions which have

                               4-21

-------
               ESP
            Electric
             Furnace
               (1)
                                      To roof
                                    \
                                      Captured
                                      matte
                                      tap
                                      fugitives
Matte
                         Main Stack
                                                            111,000  acfm
                                                            54° C
                                                     Acid Plant
                        ESP
                        (3)
                         52,000-60,400 acfm

                         496° C
  Hoboken
Converters
    (5)
Figure 4-4.    Inspiration-Miami  Smelter Baseline  Configuration
                                   4-22

-------
             654,000 acfm
             693 °C
       Noranda
       Reactors
          (3)
                                                           Main Stack

                                                           1,000,000 acfm
                                                           65 °C
                                                  Acid Plants
                               :Captured
                               •matte/slag
                               :tap fugitives--
                               Ilocal  ventilation
                               •147,000 acfm
                               i256 C

                               X     315° C
                          "  299,700  acfm
                             104 °C
                              V
                                                      ESPs
                                                      (6)
  :Reactor roof
  : fans
  :65,OOU acfm
  i25d C
•*-'2 per reactor
Matte
114,700 acfm
693 °C
                      Converters
                          (4)
Captured
converter
fugitives--
secondary hood
70,000  acfm

25° C per
converter

converter roof
fans
65,000  acfm

25° C
2 per
converter
Figure  4-5.   Kennecott-Garfield  Smelter  Baseline Configuration
                                     4-23

-------
escaped capture by the primary ^js handling  system or  the  capture
ports near the hot metal transfer areas.  The system consists  of two
65,000 scfm axial I.D. fans mounted on  the roof above  each  reactor
vessel and each converter.  The captured gas is vented through  the
smelter main stack.
     No controls beyond the baseline configuration would be  required
under Alternative II since the baseline process controls are already
best control for arsenic.
     Air curtain secondary hoods would  be installed under Alternative  III,
as would be an effective particulate control device for collection of
captured converter fugitives.
     Regulatory Alternative IV would necessitate installation of an
effective particulate control device for collection of captured matte
and slag tapping fugitive emissions.
     4.3.2.6   Kennecott-Hayden.  The baseline configuration of the
Kennecott-Hayden smelter is depicted in Figure 4-6.  Offgases from the
fluid bed roaster go to a venturi scrubber and a tray-type scrubber in
series.  The gases are then combined with converter gases upstream of
a sulfuric acid plant.   Process emissions from the single reverberatory
furnace are treated in an electrostatic precipitator (operated at
260°C, 500°F) before discharge from the main stack.  Matte and slag
tapping fugitive emissions are captured with local  ventilation systems
and discharged above the building roof without control.  Converter
fugitives are neither captured nor controlled.
     Regulatory Alternative II would require no controls beyond those
of the baseline configuration for the roaster and converter process
offgas streams since these are already treated  by best technology  for
arsenic control  under baseline.  The furnace electrostatic precipitator,
which is "hot",  is not  considered best technology,  but arsenic vapor
pressure data (see Section 3.1.1.1)  indicate that flue gas cooling
would yield no additional  emission reduction.  Therefore, no additional
controls for furnace process  emissions would be required under
Alternative II.
     Alternative III would require the installation of air curtain
secondary hoods  on the  converters, and a particulate control device
for collection of captured converter fugitive emissions.

                               4-24

-------
  Scrubber
       18,000 acfm
       329° C
  Cyclones
      66,600 acfm
      566° C
 Fluid Bed
   Roaster
     (1)
Calcines
                                        To Acid Plant Stack
                                        100,000 acfm
                                        79° C
                                  Acid Plant
                                       To roof
                                       42,600  acfm
                                       25° C
                     Furnace Stack
                     130,000 acfm
                     260° C
                                      ESP
               Reverberatory
                  Furnace
                   (1)
                                                                     Scrubber
   Captured matte/  ''
   slag tap
   fugi tives	
                                                                       ESP
Matte
                                                      226,100 acfm
                                                                         621° C
              Converters
                  (3)
Figure  4-6.   Kennecott-Hayden  Smelter  Baseline Configuration
                                    4-25

-------
     Alternative IV would necessitate installation of a collection
device for the captured matte and slag tapping fugitive emissions.
     4.3.2.7  Kennecott-Hurley.  Figure 4-7 shows the baseline configuration
(after planned modernization, see Section 4.2.1) of the Kennecott-Hurley
smelter.  Process emissions from the INCO flash furnace are cleaned in
an electrostatic precipitator before entering an acid plant.  Offgases
from the four converters pass through three parallel electrostatic
precipitators before being treated in an acid plant.  Matte tap fugitive
emissions are captured by a local ventilation system and vented to the
main stack.  Slag tap and converter fugitives are neither captured nor
control led.
     No additional  controls would be required by Alternative II for
this smelter since in the baseline configuration both process gas
streams are controlled by best technology for arsenic.
     Alternative III would require air curtain secondary hoods on the
converters, and the installation of a control device for collection of
captured converter fugitive emissions.
     Alternative IV would require a capture system for slag tapping
fugitive emissions  and the installation of a particulate collection
device to treat captured matte and slag tapping fugitives.
     4.3.2.8   Kennecott-McGill.  The baseline configuration of the
Ktnnecott-McGil1 smelter is presented in Figure 4-8.  Process emissions
from the two reverberatory furnaces pass through an electrostatic
precipitator (operated at 316°C, 600°F) before exiting the main stack.
Converter offgases  from any of the four units are treated in rnulticlones
before being ducted to the stack.  Matte and slag tap fugitives are
captured by a local ventilation system and vented to the building
roof.  Fugitive emissions from the converters are neither captured nor
controlled.
     However, arsenic vapor pressure calculations (see Section 3.1.1.1)
for the furnace offgases show that no additional arsenic emission
reduction would result if the inlet temperature to the electrostatic
precipitator were reduced to as low as 121°C (250°F).  Therefore,
Alternative II would have no effect on this gas stream.  Similar
calculations for the converter gas stream show that 57 percent arsenic
collection would occur at 121°C (250°F).  Alternative II therefore

                               4-26

-------
          Furnace Acid Plant
          Stack
          34,720 acfm
          79 °C
    Acid  Plant
          204 °C
       ESP
      Inco

  Flash Furnace

       (1)
                              Old Furnace
                              Stack
                              4,000 acfm
                              25° C
                              Captured
                              matte tap
                              fugitives
Matte
                        4  Converter Acid  Plant Stack
                           88,600 acfm
                           79  °C
                   Acid Plant
                      ESP
                      (3)
                                                      371 °C
                                              Air-to-Gas
                                            Heat Exchanger
                           104,600 acfm/converter operating
                           538  - 649 °C
                   Converters
                      (4)
Figure  4-7.   Kennecott-Hurley  Smelter  Baseline  Configuration
                                    4-27

-------
            ESP
              414,500 acfm

              316° C
        Reverberatory
         Furnaces
            (2)
                                  ' To Roof



                                 \
                                 \
                                    75,000 acfm
   Captured
   matte/slag
   tap
   fugitives
Matte
                        Main Stack
                        750,000 acfm
                        150° C
                                                   Multiclones
     164,200 to 413,000  acfm
     427° C
Converters
    (4)
Figure 4-8.   Kennecott-McGill  Smelter  Baseline  Configuration
                                   4-28

-------
would require flue gas cooling upstream of an effective  particulate
control device for the converter process offgases.
     Alternative III would require the installation  of air  curtain
secondary hoods on the converters, and a control device  for  collection
of captured converter fugitive emissions.
     Alternative IV would necessitate installation of a  control  device
for collection of captured matte and slag tapping emissions.
     4.3.2.9   Magma-San Manuel.  Figure 4-9 depicts the baseline
configuration of the Hagma smelter.  Offgases from the three reverberatory
furnaces pass through an electrostatic precipitator  (operated at
260°C, 500°F) before discharge from the main stack.  Process  emissions
from the six converters are also treated in an electrostatic  precipitator,
and then routed to an acid plant.  Matte and slag tap fugitive emissions
are captured and vented from the roof of the furnace building.   Converter
fugitives are neither captured nor controlled.
     Since the baseline configuration reflects best  control  of arsenic
in the converter process gas stream, no additional controls  would be
required under Regulatory Alternative II.  Arsenic vapor pressure
calculations (refer to Section 3.1.1.1) for the furnace  gas  stream
indicate that no additional arsenic collection would be  achieved by
flue gas cooling to 121°C (250°F), therefore, no additional  controls
beyond baseline would be required under Alternative  II.
     Converter fugitive emissions would be captured  by air  curtain
secondary hoods under Regulatory Alternative III.  An effective  control
device would be required for collection of the captured  emissions.
     Alternative IV would cause the installation of  a control device
to collect the captured matte and slag tapping emissions.
     4.3.2.10  Phelps Dodge-Ajo.   The baseline configuration  (after
consent decree modification, see Section 4.2.1) of the Phelps Dodge-Ajo
smelter is presented in Figure 4-10.   Process emissions  from  the
oxygen-sprinkle modified reverberatory furnace will  pass  through a hot
electrostatic precipitator before being treated in the single acid
plant.  Offgases from the three converters pass through  two  electrostatic
precipitators before entering the acid plant.  Matte and  slag tapping
fugitive emissions  are captured by localized ventilation  and  vented to
the main stack.   Converter fugitives  are captured by a secondary hood
system estimated to be 30 percent efficient and vented to the stack.
                               4-29

-------
                   Main Stack
                   400,000 acfm
                   246 °C
               ESP
                   517,500 acfm
                   260 °C
         Reverberatory
            Furnaces
              (3)
                                     To roof
                                     215,000 acfm
                                     25° C
   Captured
   matte/slag
   tap
   fugitives
Matte
                             To 2 Acid Plant
                             Stacks
                             187,000 acfm
                             52° C
                                                       Acid Plant
                                                          ESP
311 ,800 acfm
232 °C
                     Converters
                        (6)
Figure 4-9.   Magma-San  Manuel  Smelter Baseline  Configuration
                                   4-30

-------
               30,000 acfm
                     25° C
16,800 acfm
    310° C
         ESP
    Oxy-Sprinkle
    Reverberatory
       Furnace
         (1)
                                           iMain  Stack
                                            175,000 acfm
                                         .Jj!0°.C	
          Acid Plant
Captured
matte/slag
tap
fugitives

  Matte
                    74,400 acfm
                    288° C
                                                     ESP
                                                     (2)
                      Converters
                         (3)
                                           70.000 acfm
                                           25* C
                                                                        Captured
                                                                        converter
                                                                        fugitives
  Figure 4-10.   Phelps Dodge-Ajo  Smelter Baseline  Configuration
                                    4-31

-------
     Alternative II would impose no process controls beyond those in
the baseline configuration since they represent best technology for
arsenic control.
     Alternative III would require air curtain secondary hoods on
converters, and  the installation of a control  device to collect the
captured converter fugitives.
     Alternative IV would necessitate installation of a particulate
control device to collect the captured matte and slag tapping emissions.
     4.3.2.11  Phelps Dodge-Douglas.  Figure 4-11 depicts the baseline
configuration of the Phelps Dodge-Douglas smelter.  Process offgases
from the 24 multi-hearth roasters and three reverberatory furnaces are
cleaned in electrostatic precipitators (operated at 260°C, 500°F, and
232°C, 450°F, respectively) oefore discharge out of the roaster/reverb
stack.  Process  emissions from the five existing converters also pass
through an electrostatic precipitator (operated at 177°C, 350°F)
before stack discharge.   Calcine discharge fugitive emissions are
captured by a local ventilation system and sent to a baghouse.  Matte
and slag tap fugitives are captured by localized ventilation and
ducted to the converter electrostatic precipitator.  Converter fugitive
emissions are neither captured nor controlled.
     Arsenic vapor pressure calculations (see Section 3.1.1.1) show
that under Alternative II, no additional reduction in process arsenic
emissions occurs as a result of flue gas cooling (to 121°C, 250°F)
upstream of the  control  devices.  Therefore, Alternative II requires
no added process controls beyond the baseline configuration.
     Alternative III would require air curtain secondary hoods on all
converters along with installation of a control device to collect
captured converter fugitive emissions.
     Alternative IV would not affect calcine transfer fugitive emissions,
since in the baseline configuration these emissions are controlled in
a baghouse.  Alternative IV would require installation of an additional
control device for collection of captured matte and slag tapping
fugitive emissions.
     4.3.2.12  Phelps Dodge-Hidalgo.  The baseline configuration of
the Hildalgo smelter is shown in Figure 4-12.   Offgases from the
Outokumpu flash  furnace are treated in three parallel electrostatic

                               4-32

-------
        536,300 afcm
       ESP
         260° C
   Multi-hearth
    Roasters
      (24)
                          Main Stack
                          290,000 acfm
                          207° C
      305,000 acfm
                                        ESP
      232°  C
                                      Balloon
                                       Flue
                     Baghouse
                        'Calcine discharge
                         fugi tives
                     Talrines ^
Reverberatory
  Furnaces
     (3)
                                        Converter Stack
                                                                           237,000 acfm

                                                                           177° C
                                                                         ESP
                                       343° C
                                    Balloon
                                     Flue
                                                          Slag/matte
                                                          tap fugitives
                       40,000 acfm
                       25d C
                                                       Matte
Converters
    (5)
Figure  4-11.   Phelps  Dodge-Douglas Smelter Baseline Configuration
                                        4-33

-------
       ESP
       (3)
         1200°C   Captured
               slag/matte
            tap fugitives

              56,300 acfm

                    25°CJ
   Flash  Furnace
                  Slag
                                          Acid Plant Stack
                                           36,100 acfm

                                           79° C
                                    Acid Plant
      Slag Furnace
      Stack
      333,000 acfm
      128° C
Scrubber
   T
                                               Matte
                                          160,000 acfm
                                          371° C
                                    Slan  Furnace
               ESP
               (2)
•Captured
^Converter
•Fugi tives

ill7,000 acfm

J25°
                     Matte
                                 Converters
                                     (3)
Figure  4-12.   Phelps  Dodge-Hidalgo Smelter Baseline Configuration
                                      4-34

-------
precipitators before being mixed with emissions from the three converters
which have passed through two parallel electrostatic precipitators.
Combined process offgases are then treated in either of two acid
plants.  Slag furnace process emissions are treated in a venturi
scrubber and discharged from a separate stack.  Matte and slag tap
fugitive emissions are captured by local ventilation systems and
vented to the slag furnace stack.  Converter fugitives are captured by
secondary hoods (estimated to be 30 percent efficient) and ducted to
the slag furnace stack.
     Alternative II would require no additional controls beyond the
baseline configuration, since the baseline controls represent best
technology for arsenic removal.
     Alternative III would require air curtain secondary hoods on
converters, and the installation of a control device to collect the
captured converter fugitives.
     Alternative IV would necessitate installation of a control device
for collection of captured matte and slag tapping emissions.
     4.3.2.13  Phelps Dodge-Morenci.  Figure 4-13 presents the baseline
configuration (after consent decree modification, see Section 4.2.1)
of the Phelps Dodge-Morenci smelter.  Offgases from the two oxygen
sprinkle modified reverberatory furnaces will pass through two parallel
electrostatic precipitators prior to treatment in an acid plant.
Converter process emissions from the nine existing units pass through
gas coolers and an electrostatic precipitator before entering an acid
plant.  Matte and slag tap fugitive emissions are captured by local
ventilation systems and will be vented to the reverberatory furnace
acid plant stack.  Converter fugitives are captured by secondary hoods
(estimated to be 30 percent efficient) and discharged from the converter
acid plant stack.
     Alternative II would not require additional controls beyond the
baseline configuration, since process gas streams are already under
best control for arsenic.
     Air curtain secondary hoods would be required by Alternative III,
along with installation of a control device for collection of captured
converter fugitives.
     Alternative IV would require the installation of a particulate
control device for the collection of captured furnace fugitive emissions.
                               4-35

-------
j
35,000 acfm
To Stack
110,000 acfm
80° C

Acid
Plant
i

ESP
(2)

: 75,000 acfm
•25° C
«
•
•
J
To stack
665,000 acfm
30° C
185,000 acfm
Acid
Plant
•
•
•Captured
Jnatte/slag
Hap
«
64,000 acfm •
370° C j
•
Oxy-Sprinkle
Reverberatory
Furnace
(2)
	 *••
Matte r


ESP

Conv
(
I
484,000 acfm
650° C
erters
9)
                                                                 480,000 acfm

                                                                 25° C
                                                                 Captured
                                                                 converter
                                                                 fugitives
Figure 4-13.   Phelps  Dodge-Morenci  Smelter  Baseline  Configuration
                                  4-36

-------
      4.3.2.14  Copper Ranqe-Khtt.^tne.   The base! ine configuration  of

   1  n  t     smelter 1s depicted 1n Fi9ure 4-14-   p™«"  -'«'«»
 fro., the two reverberatory furnaces pass  through  an  electrostatic
 prec,pitator (operated at 190«C,  375'F) before discharge frm the ma,n
  tack.   Converter offgases from the two units are vented directly to
 the stack.   Matte and  slag tap fugitive emissions are captured by
  oca,  ventilation systems  and vented  to a stack on the building roof
 Converter fugitives are neither captured nor controlled

 additilT VSPOr "reSSUre Ca'CUlat1°"S ^ S<*"°« 3-1.1.1)  show no
     t ona, arsenlc removal resulting fr»  application of cold (121X

 i e   PutlCUUte C0"tr01 deVi"S °" the  Pr°CeSS  Str"ms <*  *.
 Pine S,,,e, er.  Therefore,  Alternative I, does not  require additional
 controls  beyond the baseline configuration.
     Alternative III  would require air curtain secondary hoods on the
 conveners,  as  »,, as  installation of  a particulate  control device
for collection  of  captured  converter fugitive emissions
     InstaHation  of  a  control device  for the captured matte and slag
tapping em,ss,ons  would be  required  under Alternative IV
                            4-37

-------
                 To Stack
J
386,000 ACFM
221"C

247,300 ACFM
190°C
i To roof
ESP
i
.
Reverberatory
Furnaces
(2)
75.000 acfm
25* C
Captured
matte/slag tap
fugitives

Matte

89,000 ACFM
340°C
Converters
(2)
Figure 4-14.   Copper Range-White Pine Smelter Baseline Configuration
                                 4-38

-------
4.4  REFERENCES
1.   Environmental  Protection Agency.   Proposed Rules for Primary
     Nonferrous Smelter Orders.   44 FR 6284.   January 31, 1979.

2.   Environmental  Protection Agency.   Primary Nonferrous Smelter
     Orders; National Rules.  45 FR 123.   June 24, 1980.

3.   Telecon.  Bill Gill, Texas  Air Control Board with Scott Osbourn,
     Pacific Environmental Services.  April 14, 1983.  Discussed
     compliance status and applicable regulations for the ASARCO-E1
     Paso smelter.

4.   Telecon.  Dave Che!grin, Arizona Department of Health Services
     with Scott Osbourn, Pacific Environmental Services.  April 14,
     1983.  Discussed compliance status and applicable regulations for
     copper smelters located in Arizona.

5.   Correspondence from E.R. Flowers, Tennessee Division of Air
     Pollution Control, to M.G.  Whaley, Pacific Environmental Services.
     June 8, 1983.   Applicable regulations and compliance status for
     Tennessee Chemical Co. smelter.

6.   Telecon.  Monte Keller, Utah Department of Health with Scott
     Osbourn, Pacific Environmental Services.  April 14, 1983.  Discussed
     compliance status and applicable regulations for the Kennecott-Garfield
     smelter.

7.   Telecon.  David Duran, New Mexico Environmental Improvement
     Division with Scott Osbourn, Pacific Environmental Services.
     April 14, 1983.  Discussed compliance status and applicable
     regulations for the Kennecott-Hurley and Phelps Dodge-Hildalgo
     smelter.

8.   Telecon.  Dick Serdoz, Nevada Division of Environmental Protection
     with Scott Osbourn, Pacific Environmental Services.  April 14,  1983.
     Discussed compliance status and applicable regulations for the
     Kennecott-McGill smelter.

9.   Telecon.  Dennis Drake, Michigan Department of Natural Resources
     with Glenn Whaley, Pacific Environmental Services.  June  7, 1983.
     Discussed compliance status and applicable regulations for the
     White Pine smelter.

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

11.  Occupational  Safety and Health Administration.  Engineering
     Assessment and Proposed Compliance Plan for ASARCO-E1 Paso Copper
     Smelter.  November 1982.
                               4-39

-------
12.  Occupational Safety and HeaUfi Administration.  Engineering
     Assessment and Proposed Compliance Plan for ASARCO-Hayden Copper
     Smelter.  October 1981.

13.  Occupational Safety and Health Administration.  Engineering
     Assessment and Proposed Compliance Plan for ASARCO-Tacoma Copper
     Smelter.  January 1982.

14.  Telecon.  G. Whaley, Pacific Environmental Services, with C. Gordon,
     Occupational Safety and Health Administration.  Status of
     Compliance Plans for Inorganic Arsenic Control at Primary CoDoer
     Smelters.  April 20, 1983.

15.  Environmental  Protection Agency.   Nonferrous Metals Manufacturing
     Point Source Category; Effluent Limitations Guidelines, Pretreatment
     Standards, and New Source Performance Standards.  Federal Register,
     Vol.  48.  February 17, 1983.  p.  7032.

16.  Environmental  Protection Agency and Phelps Dodge Corporation.
     Consent Decree for Morenci and Ajo Copper Smelters.  March 16,


17.  Environmental  Protection Agency and ASARCO, Incorporated.  Consent
     Decree for Hayden Copper Smelter.   April  13, 1981.

18.  Kennecott Corporation.  Compliance Schedule for Hurley (Chino)
     Smelter.   March 28, 1983.
                               4-40

-------
                      5.0  ENVIRONMENTAL IMPACTS

5.1  INTRODUCTION
     The environmental impacts on air, energy consumption, solid
wastes, and water associated with the regulatory alternatives for the
control of arsenic emissions from the 14 low-arsenic throughput copper
smelters are presented in this chapter.  The purpose of this analysis
is to determine the incremental change in air pollution, water pollution,
solid waste, and energy impacts of the regulatory alternatives over
the baseline control level for the low-arsenic throughput smelters
nationwide.
     This chapter first addresses the air pollution impacts of
implementing each of the regulatory alternatives.  Energy, solid
waste, and water pollution impacts for the regulatory alternatives are
addressed in Sections 5.3, 5.4, and 5.5, respectively.
5.2  AIR POLLUTION IMPACTS OF REGULATORY ALTERNATIVES
     The air pollution impact associated with each of the regulatory
alternatives considered for low-arsenic throughput copper smelters is
presented in this section.  Incremental and cumulative arsenic emission
reductions by smelter and nationwide are discussed.  The emission
estimates for each regulatory alternative are obtained based on the
application of capture and collection control systems selected in
Section 4.3 as the bases for the regulatory alternatives.
5.2.1  Baseline Emissions
     The baseline control  level is represented by Regulatory Alternative I
for all 14 low-arsenic throughput copper smelters.  For 10 of the
14 smelters, the baseline level reflects existing control of arsenic
emissions.   The four remaining smelters are currently planning or
undergoing modernization programs involving changes in processing and
control equipment.   For the purpose of defining baseline emissions,
                                5-1

-------
these 4 plants are assumed to bt. in their modernized configurations as
described in Section 4.2.
     Baseline regulatory alternative arsenic emission rates are determined
for each low-arsenic throughput smelter based on the arsenic material
balances presented in Appendix F and the estimated capture and collection
efficiencies of control  equipment from Chapters 3 and 4.  Baseline
arsenic emission rates are presented in Table 5-1.  This table shows
both process and fugitive emission rates for each smelter.  From
Table 5-1, arsenic emission rates under Alternative I range from
0.09 kg/hr for both Kennecott-Hurley and Tennessee Chemical Co.-Copperhill
to 5Z.1 kg/hr for Kennecott-HcGill.  The baseline arsenic emission
rate nationwide is the sum of the arsenic emission rates for all
14 smelters.  As shown in the table, the nationwide baseline inorganic
arsenic emission rate is 85.8 kg/hr.
     Nationwide annual arsenic emissions, shown in Table 5-2, are
determined using the emission rates from Table 5-1 and assuming 8,600 hours
of smelter operation per year.  For Alternative I, nationwide annual
inorganic arsenic emissions are 738 Mg/yr.
5.2.2  Arsenic Emission Reductions Under the Regulatory Alternatives
     Regulatory Alternative II specifies effective controls for process
emissions where additional arsenic removal is predicted.  For this
alternative, the Kennecott-McGill smelter would be the only smelter
required to  install new equipment.  A new baghouse or cold electrostatic
precipitator would need to be installed on the converters.  From
Table 5-1, arsenic emission rates under Alternative  II range from
0.09 kg/hr for both Kennecott-Hurley and Tennessee Chemical Co.-Copperhill
to 33.3 kg/hr for Kennecott-flcGill.  The nationwide  inorganic arsenic
emission rate under Alternative  II is 66.8 kg/hr.  From Table 5-2,
nationwide annual emissions under Alternative  II  are 575 Mg/yr.  This
represents a 22 percent reduction in arsenic emissions nationwide  from
the  baseline level.
     Regulatory Alternative III  specifies the  application  of effective
converter fugitive inorganic arsenic emission  controls.  Effective
controls for converter  fugitive  emissions include an air  curtain
secondary hood capture  system followed  by a  fabric  filter  control
device.   From Table  5-1,  arsenic  emission rates  under Alternative  III
                                 5-2

-------
                   Table 5-1.   ARSENIC EMISSIONS FROM  LOW^ARSENIC  THROUGHPUT  COPPER SMELTERS
                                   BY EMISSION  SOURCE AND REGULATORY  ALTERNATIVE
                                                  Arsenic Emissions by Regulatory Alternative (kg/hr)
en
i
CO
Smel ter Al ternative !
Process
ASARCO-E1 Paso 1.5
ASARCO-Hayden 0.2
Tennessee Chemical 0.0
Co. - Copperhill
Inspiration-Miami 0.04
Kennecott-Garf leld 1.8
Kennecott-llayden 3.2
Kennecott-Hurley 0.0005
Kennecott-McGill 45.8
Magma-San Manuel 1.3
Phelps Oodge-Ajo 3.7
Phelps Oodge-Douglas 4.4
Phelps Dodge-Hidalgo 0.2
Phelps Dodge-Morenci 0.1
Copper Range-White Pine 0.2
Nationwide Arsenic
Emission Rate 62.5
Fugi tive
4.0
6.1
0.09
0.4
2.8
0.9
0.09
6.3
0.08
0.5
0.6
0.2
1.0
0.07
23.1
Total
5.5
6.3
0.09
0.4
4.6
4.1
0.09
52.1
1.4
4.2
5.0
0.5
1.1
0.3
85.8
Alternative 11
Process
1.5
0.2
0.0
0.04
1.8
3.2
0.0005
27.0
1.3
3.7
4.4
0.2
0.1
0.2
43.7 '
Fugitive
4.0
6.1
0.09
0.4
2.8
0.9
0.09
6.3
0.08
0.5
0.6
0.2
1.0
o.n?
23.1
Total
5.5
6.3
0.09
0.4
4.6
4.1
0.09
33.3
1.4
4.2
5.0
Alternative III
Process
1.5
0.2
0.0
0.04
1.8
3.2
0.0005
45.8
1.3
3.7
4.4
0.5 i 0.2
1.1
0.3
66.8
0.1
0.2
Fugitive
1.8
3.2
0.02
0.2
2.0
0.2
0.04
1.5
0.03
0.2
0.1
0.1
0.3
0.03
i
62.5 i 9.6
Total
3.3
3.4
0.02
0.2
3.8
3.4
0.04
47.3
1.3
3.9
4.5
0.3
0.4
0.2
72.1
Alternative IV
Process
1.5
0.2
0.0
0.04
1.8
3.2
0.0005
45.8
1.3
3.7
4.4
0.2
0.1
0.2
62.5
Fugitive
4.0
6.1
0.07
0.3
2.6
0.8
0.08
5.9
0.07
0.4
0.6
0.2
0.9
0.06
22.0
Total
5.5
6.3
0.07
0.3
4.4
4.0
0.08
51.7
1.4
4.1
5.0
0.4
1.0
0.3
84.5

-------
Table 5-2.   NATIONWIDE ANNUAL ARSENIC EMISSIONS BY
  REGULATORY ALTERNATIVE AND EMISSION REDUCTIONS
                   FROM BASELINE

Regulatory
Alternative
Alternative I (Baseline)
Alternative II
Alternative III
Alternative IV
Arsenic
Emissions
(Mg/yr)
738
575
620
727
Arsenic Emission Reductions
(Mg/yr)
from Basel ine
(V
—
163
118
11
22
16
1.5
                        5-4

-------
range from 0.02 kg/hr for Tennessee Chemical Co.-Copperhill to 47.3 kg/hr
for Kennecott-McGill.  The nationwide emission rate under Alternative III
is 72.1 kg/hr.  From Table 5-2, nationwide annual emissions under
Alternative III are 620 Mg/yr.   This represents a 16 percent reduction
in arsenic emissions nationwide from the baseline level.
     Regulatory Alternative IV specifies the installation of effective
fugitive controls for matte and slag tapping operations.  Effective
control of fugitive emissions from these sources includes local ventilation
hood capture followed by a fabric filter collection system.  From
Table 5-1, arsenic emission rates under Alternative IV range from
0.07 kg/hr for Tennessee Chemical Co.-Copperhill to 51.7 kg/hr for
Kennecott-McGill.  The nationwide inorganic arsenic emission rate
under Alternative IV is 84.5 kg/hr.  From Table 5-2, nationwide annual
emissions under alternative IV are 727 Mg/yr.  This represents a
1.5 percent reduction in inorganic arsenic emissions nationwide from
the baseline level.
     Regulatory Alternative V requires zero emissions of arsenic to
the atmospheric from low arsenic throughput copper smelters.  Under
this alternative, air pollution, energy, and other environmental
impacts would reduce to zero.
5.3  ENERGY IMPACTS OF THE REGULATORY ALTERNATIVES
     Energy use associated with the baseline case was estimated assuming
that low arsenic throughput copper smelters require 35 x 10  Btu
(thermal) of energy per ton of copper produced.   The 14 low arsenic
copper smelters have a total reported copper production capacity of
      c                c            n
1.6 10  Mg/yr (1.8 x 10  tons/year).   Assuming a power plant efficiency
of 35 percent, the total nationwide electrical energy requirement of
low arsenic throughput copper smelters was estimated to be about
5 x 1010 kWh/yr.
     Annual energy requirements for Regulatory Alternatives II through
IV were estimated based on the energy requirements of additional
control equipment specified for each regulatory alternative.  For
Alternative II, additional energy is required for control of process
emissions from the converters at Kennecott-McGill.  For Alternative III,
additional energy is required for the capture and collection of converter
fugitive emissions.  For Alternative IV, additional energy is required

                                5-5

-------
for the collection of fugitive passions from matte and slag  tapping
operations.
     Table 5-3 summarizes the nationwide incremental energy requirements
of the regulatory alternatives over the baseline case for low arsenic
throughput smelters.  Incremental energy requirements, shown  in  this
table, are 4.9 x 108 kWh for Alternative II, 1.8 x 108 kWh for
Alternative III, and 9 x 106 kWh for Alternative IV.  Under Alternative  II,
only the Kennecott-McGill smelter is required to install additional
process pollution control equipment.  Therefore, the incremental
energy requirements for Alternative II reflect the additional electrical
energy needed to operate emission control equipment at the Kennecott-
McGill  smelter.  This additional  requirement, 4.9 x 108 kWh,  reflects
fan and ESP electrical  consumption plus the energy required to reheat
the stack gases in order to maintain plume buoyancy.  Thermal energy
for reheat was calculated as electrical energy assuming a power  plant
efficiency of 0.35.  The baseline energy requirements for the Kennecott-
McGill  smelter can be calculated assuming a baseline energy requirement
of 35 x 10  Btu (thermal) of energy per ton of copper produced1  and a
total  reported copper production capacity of 45,400 Mg/yr (50,000 tons/
year).    The baseline electrical  energy requirement for Kennecott-McGill
                         g
is estimated to be 2 x 10  kWh.   Compared to the baseline, the additional
energy requirements of Alternative II represent an increase of 33 percent
for the Kennecott-McGill  smelter.  The additional  energy requirements
of Alternatives III and  IV for the 14 low arsenic throughput smelters
are negligible when compared to  the baseline area.
5.4  SOLID WASTE IMPACTS  OF THE  REGULATORY ALTERNATIVES
     Arsenic-bearing dust is collected when smelting process particulate
emissions are controlled  with baghouses or electrostatic precipitators.
Often,  this dust contains recoverable copper or other salable materials.
The collected dust is recycled to the process or reclaimed elsewhere.
Thus,  only a portion of  the material  collected by air pollution control
equipment becomes  solid wastes.   Arsenic-containing materials are
present in the acid plant waste.   Acid plant waste is usually in the
form of a slurry.   State  regulations  require settling of this slurry
in a concrete pit.   The  clarified slurry is transferred to a lined
                                5-6

-------
           Table 5-3.  NATIONWIDE ANNUAL ENERGY REQUIREMENTS
                       BY REGULATORY ALTERNATIVE


                                                  Incremental Annual
                                                  Energy Requirements
Regulatory                                           from Baseline
Alternative                                          (1Q6 kwh/yr)


Alternative I (Baseline)


Alternative II                                           490a
Alternative III                                           180
Alternative IV                                             9

 Represents the additional energy required by Kennecott-FlcGill  for
 process control  equipment and stack gas reheat.
                                5-7

-------
 lagoon  for  further  settling.   From  the lagoon  the  materials  may be
 dredged  and  recycled  to  the  process.3
      A  conservative estimate  of  the amount  of  solid  wastes generated
 nationwide  under the  baseline  case  can be made  by  assuming that 50 percent
 of  the  concentrate  fed to  the  smelter  becomes  solid  waste; 25  percent
 is  sulfur which is  removed from  the process as  S02,  and  the  remaining
 25  percent  becomes  blister copper.   Given a nationwide annual  maximum
 concentrate  feed rate of 6.4 x 106  Mg/yr,2  solid wastes  (including
 slag) generated nationwide by  low arsenic smelters are approximately
 3.2 x 105 Mg/yr.
     The incremental quantity  of solid waste generated nationwide  for
 Regulatory Alternatives  II through  IV  is shown  in Table  5-4.   Incremental
 solid waste  impacts are  estimated by assuming that all process  and
 fugitive emissions collected under  each alternative  contain  1  percent
 arsenic.  In comparison with the amount of solid waste generated by
 the smelter  under the baseline case, the additional  amounts  of  solid
 waste collected under Alternatives  II  through IV are negligible.
 5.5  WATER POLLUTION IMPACTS OF THE REGULATORY  ALTERNATIVES
     The control systems for the regulatory alternatives are dry
 systems; consequently, no incremental  increase  in water  discharges is
 anticipated.  If scrubbers are used, increases  in wastewater discharges
 result if the arsenic-containing dusts are disposed along with the
 acid plant slurry.   Even if scrubbers  are used, no adverse water
 pollution impact is  anticipated.   This is because the additional waste
water discharges through use of scrubbers would be treated within
existing smelter water pollution control systems installed to meet
existing State and  Federal  regulations.
                                5-8

-------
           Table 5-4.   NATIONWIDE ANNUAL SOLID WASTES GENERATED
                            BY REGULATORY ALTERNATIVE
Regulatory
Alternative
    Incremental
   Solid Wastes
Generated Annually
   from Baseline
     (Mg/yr)a
Alternative I (Baseline)
Alternative II
    16,300
Alternative III
    11,800
Alternative IV
     1,100
aAssumes annual nationwide concentrate feed rate of 6.4 x 10  Mg
 concentrate/yr; 50 percent of concentrate becomes solid waste,
 and that all process and fugitive emissions collected
 under Alternatives II through IV contain 1 percent arsenic.
                                 5-9

-------
5.6  REFERENCES
1.   Pitt, C.H., and M.E. Wadsworth.  An Assessment of Energy Requirements
     in Proven and New Copper Processes.  Prepared for the U.S. Department
     of Energy.   Contract No. EM-78-S-07-1743.  December 31, 1980.
     p. 9.

2.   Preliminary Study of Sources of Inorganic Arsenic.  U.S. Environmental
     Protection Agency.  Research Triangle Park, North Carolina.
     Report No. EPA 450/5-82-005.  August 1982.  p. 21.

3.   Calspan Corporation.  Assessment of Industrial Hazardous Waste
     Practices in the Metals Smelting and Refining Industry.  Volume  II,
     Primary and Secondary Nonferrous Smelting and Refining.  PB  276170.
     April 1971.
                                 5-10

-------
                               6.0  COSTS

      This  section  presents  capital  and  annualized  costs  of controlling
 (1)  process  arsenic  emissions;  and  (2)  fugitive arsenic  emissions from
 matte and  slag  tapping  and  converter  operations at low arsenic throughput
 copper smelters.   These data  are  used to determine the cost of implementing
 the  regulatory  alternatives  identified  in Section  4.   The  control
 requirements  of the  regulatory  alternatives  for each  smelter are
 presented  in  Table 6-1.  As  indicated in Section 4, roaster calcine
 discharge  operations at the calcine charge smelters are  effectively
 controlled for  fugitive arsenic emissions  and  therefore  are excluded
 from  the cost analysis.
      The capital cost includes  all  the  cost  items  necessary to design,
 purchase,  and install the particulate control  system.  The  capital
 cost  of a  control  system includes the purchase cost of the  control
 device, auxiliaries such as exhaust fans,  motors,  ductwork,  and  stack;
 direct  installation charges including foundation and  other  direct
 costs  such as electrical, instrumentation  and  controls;  and indirect
 costs  for  engineering services, taxes,  contractors fees, and  contingency.
 All costs are in December 1982 dollars.
      The annualized cost of a control  system is  the annual  cost  to  the
 individual  plant to own and operate that control system.   The  annualized
 cost  includes direct operating costs such  as utilities, maintenance,
 operating labor, and indirect operating  costs  or capital-related
 charges such as  depreciation, interest,   administrative overhead,
 property taxes,  and insurance.
     While  actual  costs  experienced by  individual plants  can  vary,  the
following values have been  selected as typical  and provide  a reasonable
estimate of the  annualized  costs of each control system:
                                   6-1

-------
     Table 6-1.   REGULATORY ALTERNATIVE CONTROL REQUIREMENTS
  Alternative                       Control  Requirement


  I (Baseline case)              Existing  process  and  fugitive
                                controls  on  roasters,  smelting
                                furnaces  and converters.   Also
                                included  are those controls
                                planned or agreed to  be installed
                                prior to  1987,

 II                             Effective process controls and
                                existing  fugitive controls.
                                Effective process controls include
                                evaporative  cooling followed
                                by  either an existing  or  new  ESP
                                or  fabric filter  collector for
                                particulate  matter emissions.

HI                             Existing  process  controls and
                                effective fugitive controls for
                                converters.   Effective fugitive
                                controls  for converters include
                                air curtain  secondary  hoods for
                                capture followed  by fabric
                                filters for  collection.

 IV                             Existing  process  controls and
                                effective fugitive controls for
                                matte tapping and slag tapping
                                operations.   Effective fugitive
                                controls  for matte tapping and
                                slag  tapping operations include
                                local  ventilation for  capture
                                followed  by  fabric filters for
                                col lection.
                                6-2

-------
     Direct annualized cost Iter-.
     •    Operating labor at $11.53 per hour  and supervision  labor  at
          15 percent of operating labor expense.
     •    Maintenance labor at $11.53 per hour  and supervision  labor
          at 20 percent of maintenance labor expense.
     •    Maintenance material at 100 percent of maintenance labor
          expense.
                                                     2
     •    Electricity at 5.9 cents per kilowatt hour.
     •    Water at 8<£/m3 ($0.30/1,000 gallons).3
     •    Natural gas at $5.67/60 ($5.98/million Btu)
     Indirect annualized cost items
     •    Payrol1 overhead at 60 percent of payrol1.
     •    Operating supplies at 20 percent of total maintenance  cost.
     •    Administrative overhead at 40 percent of total operating and
          maintenance labor and operating supplies.
     •    Taxes and insurance at 2 percent of total capital cost.
     •    Capital recovery at 20 year life for ESP's and fabric  filters,
          10 year life for reheat equipment, and 10 percent interest
          rate.
     A detailed cost analysis for the alternate control systems  is
presented in the following sections.
6.1  BASELINE CONTROLS
     This section presents the estimated costs of baseline controls on
process and  fugitive emission sources.  The baseline controls are
defined as the existing process and fugitive controls planned or
agreed to be installed before 1987.   Baseline controls have no additional
costs associated with them and therefore are excluded from the cost
analysis.  Some smelters are replacing or planning to replace existing
smelting furnaces with a new smelting technology which generates SCL
emissions in concentrations  suitable  for collection in a sulfuric acid
plant.   Sulfuric acid plants remove arsenic emissions contained  in the
process offgases treated in  their gas precleaning and conditioning
                                   6-3

-------
equipment.   For these smelters   the baseline  control  cost  includes  the
total cost of the new smelting technology, sulfuric  acid plant,  and
any  additional agreed upon controls.  Table 6-2 lists  the  smelters  for
which agreements exist and presents the control equipment  to  be  installed
under each agreement.
6.1.1  Baseline Costs
     Table 6-3 presents the estimated capital and annualized  costs  for
the  add-on equipment listed in Table 6-2.  Capital and annualized
costs for new smelting technologies were based on cost estimates
contained in a draft report prepared for EPA.4  The  report  presents
cost estimates in mid-1981 dollars for retrofiting INCO flash  and
oxygen sprinkle/oxygen fuel smelting technologies at existing  smelters
of varying capacity.  The capital cost estimates were  obtained by
adjusting the cost data reported in the referenced report  to  represent
the  production capacity at a subject smelter  and adjusting  the resultant
cost estimates upward to reflect December 1982 dollars.  The  annualized
cost estimates were obtained by adjusting the reported data to be
consistent with the annual operating cost bases presented  above.
6.2  PROCESS CONTROLS
     This section presents the estimated costs of applying  effective
controls on process emission sources.  As discussed  in Section 4 and
presented in Table 6-1, effective process controls are defined as
control  systems containing evaporative cooling followed by  an ESP for
collection of process particulate matter emissions.  As indicated in
Section 4, by applying effective add-on process controls,  additional
arsenic emission reduction over baseline can  be achieved only for
converters at the Kennecott-McGill  smelter.   Therefore, the process
control  cost analysis in this section will be limited to developing
cost estimates for the control  of process emissions from converters at
the Kennecott-McGill smelter.
6.2.1  Process Control Costs
     Capital  and annualized costs were estimated for an add-on evaporative
cooler and ESP for the existing converter process emissions control
system at the Kennecott-McGill  smelter.  Currently, the converter
                                   6-4

-------
      Table 6-2.  EQUIPMENT CONSIDERED IN BASELINE COST ANALYSIS
      Smelter
     Add-on equipment
1.  ASARCO - El Paso
2.  ASARCO - Hayden
3.  Kennecott - Hurley
4.  Phelps Dodge - Ajo
5.   Phelps Dodge -  Morenci
Fugitive emission capture  system  on
smelter slag tapping operation.

INCO smelting technology in place
of the existing conventional
roaster/reverberatory smelting process.
Local ventilation for capture of
fugitive emissions from the new
smelter matte tapping and  slag
tapping operations and an  existing
ESP system for collection  of captured
emissions.   A sulfuric acid plant
for process emissions.

INCO smelting technology in place of
the existing conventional  reverberatory
smelting process.  A sulfuric acid plant
for process emissions from the new smelter.

Oxygen sprinkle/oxygen fuel
smelting in place of the existing
conventional reverberatory smelting
process.  A sulfuric acid  plant for
process emissions from the modified
smelter.

Oxygen sprinkle/oxygen fuel smelting
in place of the existing conventional
reverberatory smelting process.  A
sulfuric acid plant for process emissions
from the modified smelter.
 Since an existing  ESP will  be used for the collection of captured
 emissions,  the baseline cost analysis includes only an annual operating
 cost for the collection system.

 No formal  agreement has been made between Kennecott Copper Company and
 EPA or any  other agency regarding installation of the add-on control
 equipment  listed.
                                   6-5

-------
Table 6.3.   ESTIMATED CAPITAL AND ANNUAUZED COSTS OF
     BASELINE CONTROLS FOR PRIMARY COPPER SMELTERS
                (December 1982 dollars)

1.
2.
3.
4.
5.
Smelter
ASARCO - El Paso
ASARCO - Hayden
Kennecott - Hurley
Phelps Dodge - Ajo
Phelps Dodge - Morenci
Capital cost,
$1,000
46
75,606
54,044
51,067
92,294
Annual i zed cost,
$1,000
40
24,698
17,111
16,356
30,815
                           6-6

-------
offgases are treated in multi-cyclones and vented through a stack to
the atmosphere.  The design parameters used to calculate add-on control
system costs are summarized in Table 6-4.  No stack was included in
the system with the assumption that the existing stack will be used to
vent the offgases from add-on controls.
     Methodology for Estimating Capital and Annualized Costs - Capital
and annualized costs were developed for the evaporative cooler and ESP
system by estimating capital costs separately for the evaporative
cooler, ESP, fan, ductwork, and reheat and annualized costs for the
entire system.
     The capital costs for the evaporative cooler, ESP, and fan were
based on data contained in a report prepared for EPA by the Industrial
Gas Cleaning Institute (IGCI).   The report presented estimates of
equipment purchase and total capital costs for evaporative cooler and
ESP systems for treating offgases from reverberatory furnaces at two
smelters.  These estimates were prepared by several ESP manufacturers
in fourth quarter 1977 dollars.  Using these reported cost data, a
mathematical expression was developed to relate the reported purchase
costs for evaporative cooler, ESP, and fan to gas flow rate.  To this
purchase cost expression, the following factors were applied to obtain
a mathematical expression relating total capital cost in December 1982
dollars to gas flow rate:  175 percent to obtain total capital cost
from purchase cost, 120 percent for an assumed 20 percent retrofit
cost, and a 1.54 escalation factor to convert the costs in fourth
quarter 1977 dollars to December 1982 dollars.  The capital cost
factor of 175 percent was obtained based on cost data in the IGCI
       5
report.   The escalation factor of 1.54 was obtained by using the
Chemical Engineering Plant cost index (December 1982 = 314.3).  The
resultant mathematical expressions relating equipment purchase cost
(C^ and capital cost (C«) to gas flow rate are as follows:

     C.^ = Purchase cost of evaporative cooler (C, ) + Purchase
          cost of ESP (C, ,) + Purchase cost of fan* (C, )
                        •iy                            iz
        = 4,083(Q1)°'91 + 16,663 (Q9)°'89  + 1,093 (Q,)0'96
                 1                 C.                 1
*Fan is located on the hot side of the gas stream.
                                   6-7

-------
  Table 6-4.   DESIGN  PARAMETERS  FOR  ADD-ON  PROCESS
         PARTICIPATE MATTER  CONTROL  FOR CONVERTERS
                       AT  KENNECOTT-MCGILL
     Parameter
                                                  Value
 System description
 Evaporative cooler

 Offgas at the inlet:
   Flow rate, ai3/s facfm)
   Temperature,  °C (°F)
   Arsenic content, g/s  (lb/min)
                   g/m3  (gr/acf)
   Water, content, %  of  total volume
 Water consumption, m3/h  (gpm)

 Electrostatic precipitator
 Offgas at the in!etb:
   Flow rate, n3/s (acfn)
   Temperature,  °C (°F)
   Arsenic content, g/s  (lb/min)
                   g/m3  (gr/acf)
   Water content,  % of total volume
 Migration velocity, cm/s  (ft/mm)
 Specific collection area,
  m2 per m3/s (ft2/!,000  acfm)
 Power consumption, kw/m2  (kw/ft2) of
  collection  area
 Offgas at the outlet:
Arsenic content, g/s  (Ib/min)
                g/m3 (gr/acf)

 Fan

Location
 Pressure drop, kPa (in.  water)
Power consumption, kw (hp)

Reheat

Reheat temperature increase, °C (°F)
Capacity,  GJ  (million Btu/hr)
                                            A fan, an evaporative
                                            cooler, and  a  dry  elec-
                                            trostatic precipitator
197
232
9.16
0,05
4*
39
(418,300)
(450)
(1.21)
(0.02)

 170
154
120
9.16
0.06
3.96
103.2
0.016
(326,400)
(250) c
(0.'026)
(7.3)d
(524)d
(0.0015)
4.93   (0.65)e
0.032  (0.014)6
Hot side of  cooling  system
1      (4)
615    (825)
112
71
(200)
(67)
 Assumed  based on test data for converters "at  several other smelters.
 Same as  at  the outlet of evaporative  cooler.
cThe arsenic  loading at the ESP inlet  may  be lower than that at the
 evaporative  cooler inlet.

 Quoted by ESg manufacturers for reverberatory  furnaces in the
 IGCI report.

 Calculated  using the arsenic saturation value  of 0.0302 g/m3(0.0132 gr/acf)
 at 120 C (250°F) and 96 percent efficiency for the add-on ESP system.
 Downstream  of the existing milticyclones  and upstream of the add-on
 evaporative  cooler.
                                 6-8

-------
          where Q,  = Actual  flow rate at the inlet of add-on
                                          3
                     evaporative cooler, m /s, and
                Q2 = Actual  flow rate at the inlet of add-on ESP,
                     m3/s at 120°C
        = Capital  cost of evaporative cooler and ESP
        = 1.54 x 1.2 x 1.75 (Clx + Cly + Clz)
        - 3,234 [4,083 (Q'   + 16»663
                         xO.91 ,  1t- 9C- /n xO.89  .  n 0.96-,
        = 3,535 [3.74 (Ql}     + 15'25 (Q2>      + Ql    ]
     Ductwork costs were estimated based on an assumed total of 305 m
(1,000 ft) of duct from the existing equipment outlet to the inlet of
add-on control  system (i.e., at the inlet to fan) and a return duct
from the system outlet to the existing equipment.  The ductwork cost
(C~) based on the total  duct weight and a unit cost per weight of
  O
$ll,355/Mg ($10,300/ton) was $2,008,500.  The unit cost of ductwork
assumes that an additional  40 percent of the total ductwork weight is
needed for support structures.
     The capital cost for reheat equipment was based on data contained
in a report prepared by EPA.   The report presents capital cost estimates
in fourth quarter 1977 dollars for reheat equipment of different
capacity.  The reported costs were escalated to December 1982 dollars
using the Chemical Engineering Plant Cost Index (December 1982 = 314.3).
A mathematical  expression was developed to relate capital costs (C«)
in December 1982 dollars to gas flow rate.  The resultant expression
is:
      C4 = 1.54 [55,466 (Q3)0.59] = 85,418 (Q3)0.59
           where Q- = Reheat capacity, GJ/h
     The total  capital cost (C) in December 1982 dollars for the evaporative
cooler and ESP system is the summation of capital costs for the evaporative
cooler, ESP, and fan (C2),  ductwork (C-J, and reheat equipment (C.).

                                   6-9

-------
          c = c2 + c3 + c4

            = 3,535 [3.74 (Q^O.91 + 15.25 (Q2)0.89  + (Q^O.96]

            = 2,008,500 + 85,418 (Q3)0'59

            = 3,535 [3.74 (Q^O.91 + 15.25 (Q2)0.89  + (0^)0.96

                   + 568.2 + 24.16 (Q3)0.59]
     The estimated capital cost for an add-on evaporative cooler and
ESP system located upstream from existing converter process controls
at the Kennecott-McGill smelter is $10,017,800.  This cost was calculated
using the gas parameters listed in Table 6-4 and the above mathematical
expression.
     Annualized costs were calculated based on operating parameters
listed earlier and the following requirements:
     •    2 manhours/shift operating labor and 1 manhour/shift maintenance
          labor.
     •    Pressure drop of 1.0 kPa (4 in. water) across evaporative
          cooler and ESP.
     •    Electricity for ESP at 0.0015 kW/ft2 of ESP plate area.
     •    Miscellaneous electricity costs at 10 percent.
     Table 6-5 presents an estimate of annualized cost for add-on
process emission controls on converters at the Kennecott, McGill
smelter.
6.3  FUGITIVE CONTROLS
     This section presents the estimated costs of applying effective
controls on fugitive emission sources at low arsenic throughput primary
copper smelters.
6.3.1  Converter Controls
     Fugitive emission control equipment selected for analysis for
converter operations includes an air curtain secondary hood capture and
fabric filter collection system.  As noted in Section 4, five of the
smelters currently have some form of converter secondary hooding
in place.  One smelter, ASARCO-E1 Paso, has a building evacuation
                                   6-10

-------
       Table 6-5.  ESTIMATED ANNUALIZED COST OF ADD-ON  PROCESS
           PARTICULATE HATTER CONTROL SYSTEM FOR  CONVERTERS
                         AT KENNECOTT-McGILL
                         (December 1982 dollars)
     Item                                          Cost,$
Direct costs
Operating labor
Maintenance labor
Maintenance material
Utilities:
Electricity3
Natural gas for reheat
Water
28,600
15,000
15,000
312,400
2,252,500
31,000
   Total direct                                2,654,500

Indirect costs
Payroll overhead                                  26,200
Operating supplies                                 3,000
Administrative overhead                           18,600
Taxes and insurance                              200,300
Capital recovery:
   Evaporative cooler and ESP                  1,053,000
   Reheat equipment                              171,800

   Total indirect                              1,472,900

Total                                          4,127,400


a$130,200 is ESP power requirement.
                                   6-11

-------
system which includes a fabric filter for emission collection.  The
five smelters with converter secondary hooding are ASARCO-Hayden,
Kennecott-Garfield, Phelps Dodge-Ajo, -Morenci, and -Hidalgo.
     6.3.1.1  Converter Control Costs - Capital and annualized costs
were estimated for the installation of air curtain secondary hood
capture and fabric filter collection systems on converters at all
smelters.  For smelters without existing fugitive controls on converters,
the cost estimates include the total cost for installation of air
curtain secondary hoods and fabric filters.  However, for those smelters
with some form of converter fugitive emission controls presently
in place or required under an EPA consent decree, the cost estimates
include only the incremental  cost needed to upgrade the existing or
future control  systems to an air curtain secondary hood capture system
equipped with an effective collection device.
     The capital costs were developed by estimating costs for air
curtain secondary hood capture equipment and fabric filter collection
equipment for each smelter.  The annualized costs were developed by
estimating costs for the total capture and collection system at each
smelter.
     Capital  costs for air curtain secondary hoods - Design parameters
for air curtain secondary hood capture systems at each smelter are
summarized in Table 6-6.   Data on the number of existing and operating
converters are based on the information obtained from individual
companies and Reference 7.  Exhaust fan capacities were developed
using the design rates obtained from ASARCO and discussed in Section 3
for the air curtain secondary hood capture systems for converters at
the Tacoma smelter.  The following flow rates were used for each
converter:  33 m /s (70,000 acfm) during blowing and holding and 57
m /s (120,000 cfm) during charging and skimming.  Air curtain flow
rates used were 5.2 m /s (11,000 acfm) during blowing and 8.5 m3/s
(18,000 acfm)  during charging and skimming.
     The following ductwork requirements were estimated for the smelters
with no existing converter controls:  for the first converter at each
smelter 54 m (178 ft) of 1.5 m (60 in.)  diameter duct was assumed to
be needed to reach the main exhaust fan.   An additional 19.8 m (65 ft)
                                   6-12

-------
                                     Table 6-6.    DESIGN PARAMETERS  FOR AIR  CURTAIN  SECONDARY
                                            HOOD  CAPTURE  SYSTEM  FOR  PRIMARY  COPPER  SMELTERS
CT)


OJ
Smel ter
No. of converters
Total Total b
existing3 operating

Fan capacity,
mVs (acfm)
Flue system dimensions
       1.   ASARCO  - El Pasoy

       2.   ASARCO  - Hayden

       3.   Tennessee  Chemical
           Co. - Copperhil 1

       4.   Inspiration -  Miami
                                     1RO, 1R1 ,  1SB   109  (230,000)    31  m (100  ft)  long 1.5 m (60 in.)  duct

                                     1RO, 1RI,  HI    142  (300,000)

 5.   Kennecott - Garfield

 6.   Kennecott - Hayden

 7.   Kennecott - Hurley

 8.   Kennecott - McGill

 9.   Magma - San Manuel

10.   Phelps Dodge - Ajo9

11.   Phelps Dodge - Douglas

12.   Phelps Dodge - Hidalgo9     3

13.   Phelps Dodge - Morenci      9

14.   Copper Range - White Pine   2
1RO

IRQ,  1RI, III

1RO,  1RI, HI

1RO,  1R1 , 1SB

1RO,  1RI, HI

1RO,  1R1, 1H

2RO,  2RI
 66  (140,000)

142  (300,000)

142  (300,000)

109  (230,000)
                                                                    1?8 m (420 ft)  long 1.5 m (60 in.)  duct  and 91 m (300 ft) long  1.7 m  (67 in.) duct

                                                                    469 m (1,540  ft)  long 1.5 m (50 in.)  duct and 91 m (300 ft)  long  2.5  m  (98 in.) duct
                               223 m (730  ft)  long 1.5 m (60 in.)  duct and 91 ro (300 ft) long  2.2 m  (86 in.) duct

               142  (300,000)    335 m (1,100  ft) long 1.5 m (60 in.)  duct and 91 m (300 ft)  long  2.5  m (98 in.) duct

               14?  (300,000)    335 m (1,100  ft) long 1.5 m (60 in.)  duct and 91 m (300 ft)  long  2.5  m (98 in.) duct

               208  (440,000)    623 m (2,043  ft) long 1.5 m (60 in.)  duct and 91 m (300 ft)  long  3 m  (119 in.) duct

1RO,  1RI, 1SB   109  (230,000)    223 m (730  ft)  long 1.5 m (60 1n.)  duct and 91 m (300 ft) long  2.2 m  (86 in.) duct

1RO,  1RI, 1H    142  (300,000)    469 m (1,540  ft) long 1.5 m (60 in.)  duct and 91 m (300 ft)  long  2.2  m (98 in.) duct

                               223 m (730  ft)  long 1.5 m (60 in.)  duct and 91 m (300 ft) long  2.2 m  (86 in.) duct
                                            1RO.  1RI, 1SB   109 (230,000)

                                            2RO,  2RI        208 (440,000)

                                            1RO
                66 (140,000)     128 m  (420 ft) long 1.5 m (60  in.) duct and 91 m (300  ft)  long 1.7 m (67 in)  duct
aEach existing smelter will  be  equipped with an  air  curtain secondary  hood capture system.   Each  air curtain will  have a separate fan designed at
 8.5 m3/s  (18,000 acfm) at a pressure of 7.5 kPa (30  in. h^O).
bRO:  Rol'1-out, RI:  Roll-in, SB:  Standby, and  H:   Holding.
cFan capacities are based on operating converters only.  Assumed exhaust  rate  from a Converter in roll-out mode 0-e.. during
 57 m'/s (120,000 acfm) and  from  a converter in  roll-in mode (i.e.,  during blowing and holding)  1s  33 m3/s (70,000 acfm).
dr,
                                                                                                                                        "*
                    ,
       dOuct  lengths used are-   54 m  (178 ft) long  1.5 m  (60 in.) diameter  duct per system, 19.8 m (65 ft)  long 15 m (60 In.)  diameter duct for each
        addition!' I  system, and  91 m  (300 ft) long conwon  duct downstream of the fan.  Example case Kennecott, Hayden:  length of  1.5 m (60 in.)
        diameter  duct is 3 x 54 + (1)  x 19.8 + (2)  x  19.8 = 223 m.
                                                                                                                                      a/?:-
                                        sssr-s  s-s2J'»
        (100 ft)  ducting is sufficient  to move the captured gases to the existing duct.
       Currently,  the converter fugitive emissions are  captured in a secondary hood and a duct exists  to ^ ^Y*^ J85"  \°e* Stack'
        is assumed  to be utilized to vent the captured gases  from the air curtain secondary hood.   No new ducting or fans are required.
       "Currently  the converter fugitive emissions are  captured in a secondary hood and a duct exists  to vent the captured gases  to  a stack,  "owever,  the
        existin  system  is smaller than the air curtain  secondary hood capacity.  Therefore, none of the existing equipment 1s assumed to be utilized.

-------
of 1.5 m (60 in.) diameter duct ic required for each subsequent capture
system after the first one (i.e., 54 m duct for the first converter,
54 m + 19.8 m duct for the second converter, 54 m + 2 x 19.8 m for the
third converter and 54 m + 3 x 19.8 m for the fourth converter).  A
91 m (300 ft) common duct was used for venting gases from the main fan
to the particulate collection device.  The common duct diameters vary
for each smelter based on exhaust fan capacity.
     For the ASARCO-E1 Paso, -Hayden, Kennecott-Garfield, and Phelps
Dodge-Morenci smelters with some form of existing converter controls
in place, it was assumed the existing ductwork and fans were salvageable
and could be incorporated in the new air curtain secondary hood system.
A new 31 m (100 ft) of 1.5 m (60 in.) diameter ductwork was assumed to
be required to move the capt'ired gases from the new system to the
existing ductwork at the ASARCO-E1 Paso smelter.  No new ductwork was
assumed to be required at the other smelters.  At the Phelps Dodge-Ajo
and -Hidalgo smelters, the existing converter control system capacity
is smaller than the new air curtain secondary hood system.  Therefore,
none of the existing equipment was assumed to be salvageable.  New
ductwork costs were estimated for these smelters in the same manner as
for the smelters with no existing converter controls.
     Table 6-7 presents estimated capital  costs for air curtain secondary
hoods for each low arsenic throughput smelter.  These cost estimates
were calculated using the cost data provided to EPA by ASARCO.8  The
ASARCO cost data contained January 1982 cost estimates for installing
air curtain secondary hood capture equipment on three converters at
the Tacoma smelter.  The direct capital costs reported were based on
actual  estimates and indirect capital costs were based on percentages
of the estimated equipment cost.  These cost data were updated to
December 1982 dollars and used to estimate the costs for installation
of an air curtain and a secondary hood.  The resultant costs, $197,100
for an air curtain and $125,100 for a secondary hood, were used to
estimate air curtain secondary hood costs  for the low arsenic throughput
smelters based on the number of converters listed in Table 6-6 for
each smelter.  Electrical  system costs were estimated for each capture
system based on data from the ASARCO report.   Ductwork costs were
                                   6-14

-------
                                                         (December  1982 dollars)
                                                                        Capital cost, $
en
wMHt I UC. 1
1. ASARCO - El Paso
2. ASARCO - Hayden
3. Tennessee Chemical Co. -
Copperhill
4. Inspiration - Miami
5. Kennecott - Garfield
6. Kennecott - Hayden
7. Kennecott - Hurley
8. Kennecott - McGill
9. Magma - San Manuel
10. Phelps Dodge - Ajo
11. Phelps Dodge - Douglas
12. Phelps Dodge - Hidalgo
13. Phelps Dodge - Morenci
14. Copper Range - White Pine
Air curtain
secondary hood
966,600
1,611,000
644,400
1,611,000
1,288,800
966,600
1,288,800
1,288.800
1,933,100
966,600
1,611,000
966,600
2,899,800
644,400
Ductwork
350,000
0
1,631,200
3,981,900
0
2,444,000
3,255,000
3,255,000
5,171,000
2,444,000
3,981,900
2,444,000
0
1,631,200
Electrical
58,600
90,700
42,600
90,700
74,600
58,600
74,600
74,600
106,700
58,600
90,700
58,600
154,600
42,600
Total
1,375,200
1,701,700
2,318,200
5,683,600
1,363,400
3,469,200
4,618,400
4,618,400
7,210,800
3,469,200
5,683,600
3,469,200
3,054,400
2,318,200

-------
estimated based on the duct requirements listed in Table 6-6 and an
assumption that an additional  40 percent of duct material is required
for use as support structures.
     Capital  costs for fabric filters - Table 6-8 presents estimated
capital costs for fabric filters required for the collection of captured
fugitive emissions.  The table also summarizes the capital costs for
air curtain secondary hoods from Table 6-7 and presents the capital
costs for the total system consisting of air curtain secondary hoods
and a fabric filter for each smelter.  These cost estimates were
obtained by calculating costs for a fabric filter,, 43 m  (140 ft) of
ductwork, and a 61 m (200 ft) stack for each smelter.  The exhaust fan
capacities listed in Table 6-6 were used to size equipment.  The
fabric filter costs were based on cost estimates contained in the  I6CI
       g
report.   The IGCI report presents cost estimates in fourth quarter
1977 dollars for fabric filters treating offgases from reverberatory
furnaces at two smelters and was prepared by several fabric filter
manufacturers.  The costs were developed for a 2-to-l filter cloth-to-air
ratio.  Both reverse air and shaker type mechanisms were  suggested for
bag cleaning.
     Ductwork costs were estimated based on a  total of 43 m  (140 ft)
of ductwork assumed to be needed for venting offgases from the fabric
filter to the stack, and a unit cost of $ll,355/Hg  ($10,300/ton) of
ductwork weight.  The unit cost includes 40 percent additional material
to be  used for support structures.
     The capital cost for the stack was based  on  data contained  in a
report prepared for EPA.    The report  presented  capital  cost estimates
in December  1977 dollars for  different  sized  stacks.
     The reported costs  for fabric filters and stacks were  escalated
to December  1982 dollars using  the Chemical Engineering  Plant Cost
Index  (December 1982 =  314.3).  Total capital  cost  for  the  fabric
filter collection  equipment was obtained by adding  the  costs  for the
fabric filter, duckwork, and  stack,  plus a 25  percent  retrofit  factor.
A mathematical expression  is  developed  to  relate  the  total  capital
costs  of  fabric filter  collection  equipment to gas  flow rate.   The
resultant  expression  is:
                                    6-16

-------
   Table 6-8.  ESTIMATED CAPITAL COSTS OF AIR CURTAIN SECONDARY
           HOODS AND FABRIC FILTERS FOR PRIMARY COPPER SMELTERS
                         (December 1982 dollars)
Smelter
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
ASARCO - El Paso3
ASARCO - Haydenb
Tennessee Chemical
Co. - Copperhill
Inspiration - Miami
Kennecott - Garfield
Kennecott - Hayden
Kennecott - Hurley
Kennecott - McGill
Magma - San Manuel
Phelps Dodge - Ajo
Phelps Dodge - Douglas
Phelps Dodge - Hidalgo
Phel ps Dodge - Morenci
Copper Range -
White Pine
Capital cost, $
Air curtain
secondary
hoods
1,375,200
1,701,700
2,318,200
5,683,600
1,363,400
3,469,200
4,618,400
4,618,400
7,210,800
3,469,200
5,683,600
3,469,200
3,054,400
2,318,200
Fabric filter
0
0
2,115,700
4,141,300
3,833,800
3,262,000
4,141,300
4,141,300
5,838,800
3,262,000
4,141,300
3,262,000
5,471,300
2,115,700
Total
1,375,200
1,701,700
4,433,900
9,824,900
5,197,200
6,731,200
8,759,700
8,759,700
13,049,600
6,731,200
9,824,900
6,731,200
8,525,700
4,433,900
 The captured  gases  will  be treated in an existing fabric filter
 collection  system.
DThe cost of collection system for the captured fugitive emissions was
 included in the  baseline.
                                  6-17

-------
          C = Total capital co?t of fabric filter collection equipment,
              December 1982 dollars

            = (Retrofit factor) [Capital cost of fabric filter (CJ
               + Capital  cost of ductwork (CJ + Capital cost of
               stack (C3)]
            = (125%)[24,430 (Q4)0'96 + 19,428 (Q4)0'5  + 22,300 (Q4)0'49

               where Q. = Actual flow rate at the inlet of fabric
                      "                                              o
               filter (i.e., fan exhaust flow listed in Table 6-6), m /s

            = 24,285 [1.26 (Q4)0'96 + (Q4)0'5  + 1.15 (Q4)0'49]

     Annualized costs for air curtain secondary hood capture and
fabric filter collection system - Table 6-9 presents the annualized
costs for air curtain secondary hood capture and fabric filter collection
systems.  The costs were estimated using the cost bases discussed
earlier and the following labor and utility requirements:
•    2 manhours/shift for operating and maintenance labor for capture
     equipment and 2 manhours/shift operating and maintenance labor
     for a 66 m3/s (140,000 acfm) fabric filter.
•    2 manhours/shift operating and maintenance labor for capture
     equipment and 2.5 manhours/shift operating and maintenance labor
     for a 109 m3/s (230,000 acfm) fabric filter.
•    4 manhours/shift operating and maintenance labor for capture
     equipment and 4 manhours/shift operating and maintenance labor
     for a 142 m /s (300,000 acfm) and a 208 m/s (440,000 acfm) fabric
     filter.
•    Electricity requirements at 7.5 kPa (30 in. HyQ) for air curtains,
     5.5 kPa (22 in. HLO) for secondary hoods, and 1.5 kPa (6 in. H90)
     for fabric filters.                                           L
6.3.2  Matte and Slag Tapping Controls
     Fugitive emission control equipment for matte tapping and slag
tapping operations includes a local ventilation capture and a fabric
filter collection system.  All of the smelters currently capture
fugitive emissions from the matte tapping operations.  One smelter,
                                   6-18

-------
Table 6-9.  ESTIMATED ANNUALIZED COSTS OF AIR CURTAIN
     SECONDARY HOODS AND FABRIC FILTERS FOR PRIMARY
                    COPPER SMELTERS3
Smelter
1. ASARCO - El Paso
2. ASARCO - Hayden
3. Tennessee Chemical
Co. - Copperhill
4. Inspiration - Miami
b. Kennecott - Garfield
6. Kennecott - Hayden
7. Kennecott - Hurley
8. Kennecott - McGill
9. Magma - San Manuel
10. Phelps Dodge - Ajo
11. Phelps Dodge - Douglas
12. Phelps Dodge - Hidalgo
13. Pheins Dodge - Morenci
14. Copper Range - White Pine
Annual operating cost, $
Utilities
117,600
174,400
455,200
1,123,900
372,400
819,000
1,123,900
1,123,900
1,816,500
636,000
1,123,900
819,000
522,700
455,200
Labor
0
0
87,200
145,700
87,300
94,600
145,700
145,700
145,700
0
145,700
0
87,300
87,200
Maintenance
material
0
0
29,900
59,800
29,900
37,400
59,800
59,800
59,800
0
59,800
0
29,900
29,900
Indirect
Costs
189,100
233,980
705,300
1,513,400
810,300
1,030,600
1,367,000
1,367,000
1,956,800
925,000
1,513,400
925,500
1,268,000
705,300
Total
306,700
408,400
1,277,600
2,842,800
1,300,000
1,981,600
2,596,400
2,696,400
3,978,800
1,561,500
2,842,800
1,744,500
1,908,000
1,277,500
                          6-19

-------
ASARCO-E1 Paso, collects the captured emissions from the matte  tapping
operation.  The existing capture equipment was assumed to  be adequate
and thus was not included in the control cost estimate.  For ASARCO-Hayden,
the matte and slag tapping capture and control costs were  included  in
the baseline, and therefore, were not estimated in  this analysis.
     A majority of smelters capture fugitive emissions from the slag
tapping operations.  For those smelters that do not have slag tapping
capture systems presently, slag tapping capture system costs were
estimated based on a typical exhaust volume flow rate of 14.2 m3/s
(30,000 acfm).  This flow rate was based on exhaust flow rates  from
the slag tapping operations at a number of smelters.  The  capture
system for slag tapping operations includes tap port hoods, slag
launder hoods, sufficient length of duct to move the captured emissions
from the launder hoods to the fan and then to the main duct, and hoods
over slag ladles.  The total cost, including slag launder  hoods,
ducting, plenum chamber, support steel, and foundation would be $46,000
for a slag tapping capture system.  Fan costs were not included  in  the
capture system but were included in the collection system.  The  annualized
operating costs were based on 2 manhours of maintenance labor per
week, electricity requirements for an 80 hp fan, capital recovery at
15 years of equipment life, and administrative overhead, taxes,  and
insurance at 4 percent of capital  costs.
     The capital  costs developed for fabric filter collection systems
for captured gases were based on data from the IGCI report prepared
for EPA.  The reported costs for the fourth quarter of 1977 were
escalated to December 1982 dollars using the Chemical  Engineering
Plant Cost Index (December 1982 = 314.3).   The costs were  estimated
            2
for a 35.4 m /s (75,000 acfm) capacity fabric filter system based on
typical  ventilation rates of 18.9  m3/s (40,000 acfm) from  matte  tapping
operations and 16.5 m /s (35,000 acfm) from slag tapping operations.
Sixty-one meter (200 ft) long ductwork was included in the cost  estimate.
Annualized operating costs were estimated  using the cost bases  discussed
earlier, the labor requirements of 2 manhours/shift operating labor and
                                 o
1  manhour/shift maintenance labor ,  and fan power requirements at 1.5 kPa
(6 in.  water).   Tables 6-10 and 6-11 present the capital and annualized cost
                                   6-20

-------
   Table 6-10.   ESTIMATED CAPITAL COSTS OF ADD-ON  FUGITIVE  EMISSION
              CAPTURE AND  COLLECTION SYSTEMS  FOR  MATTE
                     AND SLAG TAPPING OPERATIONS
                        (December 1982 dollars)

1.
2.
3.
4.
5.
6.
7.
8.
y .
10.
11.
12.
13.
14.
Srnel ter
ASARCO - El Paso
ASARCO - Hayden
Tennessee Chemical
Co. - Copperhill
Inspiration-Miami
Kennecott - Garfield
Kennecott - Hayden
Kennecott - Hurley
Kennecott - McGill
Magma - San Manuel
Phelps Dodge - Ajo
Phelps Dodge - Douglas
Phelps Dodge - Hidalgo
Phelps Dodge - Morenci
Copper Range -
White Pine
No. of
smelting
furnaces
1
2
1
1
3
1
2
2
3
1
3
1
5
2
Cost, $1,000
Local
venti-
lation
capture
system
0
0
0
29
0
0
58
0
0
0
0
0
0
0
Fabric
filter
collection
system
370C
0
894
894
1,786
894
894
894
1,786
894
1,786
894
1,786
894
Total
370
0
894
923
1,786
894
952
894
1,786
894
1,786
894
1,786
894
 Local  ventilation  system  costs  were based  on  $29,000 dollars  per
 smelting  furnace.
'Collection  system  costs were  based  on  one  fabric  filter unit  for
 smelters  with  one  or  two  smelting  furnaces  and  two  fabric  filter
 units  for smelters with three or more  smelting  furnaces.

'For system  collecting  the captured  emissions  from the slag tapping
 operation only (captured  emissions  from  matte tapping operations are
 currently collected in an existing  ESP system.)
                                   6-21

-------
Table 6-11.   ESTIMATED ANNUALIZED COSTS OF ADD-ON FUGITIVE EMISSION
               CAPTURE AND COLLECTION SYSTEMS FOR MATTE
                      AND SLAG TAPPING OPERATIONS
                       (December 1982 dollars)

1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
Smel ter
ASARCO - El Paso
ASARCO - Hayden
Tennessee Chemical
Co. - Copperhill
Inspiration - Miami
Kennecott - Garfield
Kennecott - Hayden
Kennecott - Hurley
Kennecott - ficGill
Magma - San Manuel
Phelps Dodge - Ajo
Phelps Dodge - Douglas
Phelps Dodge - Hidalgo
Phelps Dodge - Morenci
Copper Range -
White Pine
No. of
operating
smelting
furnaces
1
1
1
1
2
1
1
1
2
1
2
1
2
1
Costs, $1,000
Local
venti-
lation
capture
system
0
0
0
4
0
0
8
0
0
0
0
0
0
0
Fabric
filter
collection
system
153b
0
257
257
514
257
257
257
514
257
514
257
514
257
Total
153
0
257
261
514
257
265
257
514
257
514
257
514
257
 For the purpose of this table, the number of operating  smelting  furnaces
 is defined as the total number of smelters that may  be  tapped  at a  time.
 3For system collecting the captured emissions from  the slag  tapping  operaf
 only.   (The captured emissions from the matte tapping operation
 are currently collected in an existing ESP.)
                                   6-22

-------
for the smelter matte and slag taping fugitive emission capture and
collection systems.
6.4  COSTS OF REGULATORY ALTERNATIVES
     This section presents the costs associated with the implementation
of the regulatory alternatives defined in Section 4.0 and listed in
Table 6-1.  A compilation of the capital  and annualized costs for
each of the regulatory alternatives for the smelters processing low
arsenic feed is presented in Table 6-12.
6.5  COST EFFECTIVENESS
     The purpose of deriving cost effectiveness is to illustrate the
differences in cost relative to the arsenic emission reduction achievable
for each regulatory alternative analyzed.  Cost effectiveness is
defined as incremental annualized costs in dollars per unit of pollutant
removed over the baseline.
     Table 6-13 summarizes incremental annualized costs and cost
effectiveness for the regulatory alternatives for low arsenic feed
smelters individually.  The table also presents annual  emissions and
emission reductions projected over the baseline.
                                   6-23

-------
             Table 6-12.   SUMMARY  OF INCREMENTAL COSTS OF REGULATORY  ALTERNATIVES OVER BASELINE
             FOR CONTROL  OF  ARSENIC  EMISSIONS FOR LOW-ARSENIC THROUGHPUT  PRIMARY COPPER SMELTERS
                                   (Costs  in $l,000's December 1982  dollars)


Smel ter


1. ASARCO - El Paso
2. ASARCO - Hayden
3. Tennessee Chemical
Co. - Copperhill
4. Inspiration - Miami
5. Kennecott - Garfield
6. Kennecott - Mayden
7. Kennecott - Hurley
8. Kennecott - McGill
9. Magma - San Manuel
10. Phelps Dodge - Ajo
11. Phelps Dodge - Douglas
12. Phelps Dodge - Hidalgo
13. Phelps Dodye - Horenci
14. Copper Range -
White Pine
Total
	
Basel 1ne
(Existing process a
and fugitive controls)
Capital
cost
46e
;S,606f

0
0
0
0
54,044h
0
0
51, 06 7-'
0
0
95,294^

0
276,057
Annual ized
cost
40e
24,698f

0
0
0
0
i7,nih
0
0
16,356-*
0
0
30,815j

0
89,020


Effective process
controls
Capital
cost
0
0

0
0
0
0
0
10.0181

0

0
0

0
10,018
Annual ized
cost
0
0

0
0
0
0
0
4.1271

0

0
0

0
4,127
Effective process 1 Effective process controls, and
controls, and effective 1 effective fugitive controls
fugitive controls for
converters only
Capital
cost
1,375
1,702

4,434
9,825
5,197
6,731
8,760
8,760
13,050
6,731
9,825
6,731
8,526

4,434
96,081
Annual ized
cost
307
408

1,278
2,843
1,300
1,982
2,696
2,696
3,980
1,562
2,843
1,745
1,908

1,278
26,826
for smelter matte and slag ^
tapping and converter operations
Capital
cost
370
09

894
923
1,786
894
952
894
1,786
894
1,786
894
1,786

894
14,753
Annual ized
cost
153
O9

257
261
514
257
265
257
514
257
514
257
514

257
4,277
01
I
    Footnotes are given on next page.

-------
Table 6-12.  SUMMARY OF INCREMENTAL COSTS OF REGULATORY ALTERNATIVES OVER
  BASELINE FOR CONTROL OF ARSENIC EMISSIONS FOR LOW ARSENIC THROUGHPUT
                         PRIMARY COPPER SMELTERS
                               (concluded)
aExisting fugitive controls include effective capture and collection of
 roaster fugitive emissions at calcine feed smelters, effective capture
 of smelter matte tapping fugitive emissions at all smelters with the
 collection of captured emissions only at the ASARCO-E1 Paso smelter;
 effective capture of slag tapping fugitive emissions at some smelters;
 and an effective fabric filter collection system to be used for converter
 fugitive emissions at ASARCO-E1 Paso.

 Effective process controls include spray chamber cooling and ESP collection
 systems for process particulate matter emissions from roasters, smelters,
 and converters.

°Effective fugitive controls for converters include air curtain secondary
 hood capture followed by fabric filter collection systems.

 Effective fugitive controls for smelter matte and slag tapping operations
 include local ventilation hood capture followed by a fabric filter collection
 system.

 Represents the cost of installation of a fugitive emission capture system
 for the smelter  slag tapping operation.  ASARCO is required to install
 a fugitive emission capture system for the smelter slag tapping operation
 by July 1, 1987, under the agreement reached with OSHA.

 Represents the cost of process modification to the smelter.  The existing
 smelting process is being modified or will be modified to convert to
 the INCO smelting process in order to satisfy the accord reached between
 ASARCO Smelting  Co. and EPA.

9The furnace matte and slag tapping emission capture and collection costs
 are included in  the cost of process modification to the smelter shown
 under the baseline.

 Represents the cost of process modification to the smelter.  Kennecott
 Copper Company has announced that it plans to convert the existing
 smelting process at the Hurley smelter to the INCO process.

Represents the cost of new evaporative cooling and ESP control systems
 for the smelter  furnace and converter process emissions.

JRepresents the cost of process modification to the smelter.  The existing
 smelting process is being modified or will be modified to convert to
 oxygen sprinkling/oxygen fuel smelting process in order to satisfy the
 accord reached between Phelps Dodge Copper Co. and EPA.
                                  6-25

-------
                     Table 6-13.  EMISSION REDUCTION AND COST EFFECTIVENESS  IMPACTS  FOR
                               LOW-ARSENIC THROUGHPUT PRIMARY COPPER SMELTERS
ro
Ol

$.w>H»r
1 AlftRfO - 11 Paso
7. ASARCO - llayden
1 lennessee Chpwical
f n - CopperM 1 1
4. Insplrallnn - Miami
5 Kenriecott • C.arfleld
6 Kpnnrrott - Hayden
J KpnnpcoU - Hurley
8 Kpnnecott - HrKlll
9. Magma - San Manuel
10 Phelps Dodge - Ajo
11 I'helps Dodge -
Douglas
12. Phelps Dodge -
Hidalgo
13. Phelps Dodge -
ttorpnc 1
14. Copper Range -
White Pine
IOIM S
Arsentc
feed
kq/hr
112
1 70
I.I
II!
U8
8 0
I 0
8)
7.0
47
II
14
4.S
0 7

Arsenic
Wq/yr
47.1
54 7
ft n
i i
19.6
15.3
n.n
448.1
12.0
36.1
43.0
4.3
9.5
7.6
737
Arsenic

47 1
54.7
0 8
3.1
39.6
35 3
o.n
786.4
17.0
36 1
43.0
4.3
9.5
7.6
575.3
All
Re.lucl.lrtn
from
lk,/,r
0
n
ii
n
0
0
n
161.7
n
0
0
0
0
0
!f,! 7
r-rnal Ive II
Annual l?ed Cost
10't/yr t/Mq
0
0
0
0
0
411
0
4,177 ?ri.500
0
0
0
0
0
0
4,127 25,500
Arsenic
"9/y
28,4
29.2
0.7
1.7
It. 7
79.2
0.3
406
n.2
33.5
38.7
2.6
3.4
1.7
618.13
Alternative III
Reduction
from Annual l7prl
Mq/yr 10'J/yr
18.9 107
25.0 4011
0.6 1.278
1.7 7.813
6.9 I. 300
6. I I. 912
0.5 2.696
41.1 7,696
0.8 3.980
2.6 1,562
4.3 2,843
1.7 1.745
6.1 1.908
0.9 76,876
118. Z Z6.826
Cost
VKq
16.700
16,300
7.130,000
1.677.000
188.400
374,900
5.M7.000
61.000
4.975.000
600,800
55,800
1,076,500
312,900
1,420.000
227,000
Al ternatt
Reduction
Arsenic from
Mg/yr Mq/yr
47.3 0
54.2 0
0.7 0.1
2.6 0.8
37.8 1 8
34.4 0.9
0.7 0.1
443.8 4.3
12.0 0
35.3 0.8
42.1 0.9
3.4 0.9
8.6 0.9
2.6 0
725.5 11.5
ve IV
Annual ize*1
10' J/yr
0
0
257
261
514
257
265
257
514
257
514
257
514
757
4,277
Cost
1/Mq

0
7,570.000
326.300
285.600
289.600
7,560.000
59.HOO

321.100
571.100
285,600
571.100
--
371.900

-------
6.6  REFERENCES
 1.  Survey of Current Business.  November 1982.  Primary Metal Industry
     Labor Costs.  December 1982.

 2.  Monthly Energy Review.  October 1982.  Electricity Costs.
     DOE/EIA-003583/01.  January 1983.

 3.  GARD Inc.  Capital and Operating Costs of Selected Air Pollution
     Control Systems.  U.S. Environmental Protection Agency.  EPA
     Report No. EPA 450/5-80-002.  December 1978.

 4.  PEDCo Environmental, Inc.  Cost Estimate and Comparisons for
     Converting from Reverberatory Furnace Smelting to Oxygen Flash/
     Sprinkling Smelting.  (Draft) U.S. Environmental Protection
     Agency.  EPA Contract No. 68-03-2024.  November 1982.

 5.  Industrial Gas Cleaning Institute.  Cost Estimates of Upgrading
     Particulate Matter Controls in Copper Smelter Reverberatory
     Furnaces.  U.S. Environmental Protection Agency.  EPA Contract
     No. 68-02-2532.  Task No. 2, March 1977.

 6.  U.S. Environmental Protection Agency.  Draft Standards Support
     and Environmental Impact Statement.  Volume 1:  Proposed National
     Emission Standards for Arsenic Emissions from Primary Copper
     Smelters.  Research Triangle Park, NC.  June 1978.

 7.  The Metallurgical Society of AIME.  Copper and Nickel Converters.
     Proceedings of a Symposium on Converter Operating Practices
     sponsered by the TMS-AIME Pyrometallurgy

 8_  ASARCO Incorporated Converter Secondary Hooding, Tacoma Plant.
     Salt Lake City.  Utah.  January 22, 1981.

 9.  Reference 5.

10.  Reference 3.
                                 6-27

-------
                         7.0   ECONOMIC IMPACT

       This section first presents an economic profile of the primary copper
industry in general, and primary copper smelters in particular (Section
7.1). The data presented in the economic profile is then used in an economic
analysis of the industry (Sections 7.2 and 7.3).  The economic profile
focuses on several primary copper smelter industry characteristics, such as:
number and location of smelters, copper supplies, copper demand, competition,
substitutes, and prices.

7.1    INDUSTRY ECONOMIC PROFILE
7.1.1  Introduction
       Copper's utility stems from its qualities of electrical and thermal
conductivity, durability, corrosion resistance, low melting point, strength,
malleability, and ductility.  Principal uses are in transportation, machinery,
electronics, and construction.
       The Standard Industrial Classification Code (SIC) definition of the
primary copper industry is the processes of mining, milling, smelting, and
refining copper.  The primary copper smelters are included in SIC 3331
(Primary Smelting and Refining of Copper).  Copper-bearing ore deposits and
substantial amounts of copper scrap provide the raw materials for these
processes.
       In addition to producing copper, the industry markets by-product
minerals and metals that are extracted from the ore deposits, such as silver,
gold, zinc, lead, molybdenum, selenium, arsenic, cadmium, titanium, and
tellurium.  Many of the companies that own primary copper facilities also
fabricate copper.  Many of these same companies are also highly diversified
and produce other metals, minerals, and fuels.
       The standard under consideration directly affects only one of the four
primary copper processes, namely smelting.  However, the other three related
processes are an integral part of the ownership and economic structures of
copper smelters and therefore must be considered in determining industry
                                     7-1

-------
impact. Mining and milling  processes  supplying  a smelter will  be  secondarily
affected by a smelter impact because  transportation  costs to  an alternate
smelter will  add a sizeable business  cost.   Transportation costs  for  concen-
trate are significant because only 25 to 35 percent  of the concentrate  is
copper and the remaining 75 to 65 percent that  is also being  transported  is
waste material.  The same interdependence between smelter and refinery  is  not
as critical because the copper content after leaving the smelter  is typically
98 percent.
       Even if there were no business dependencies among the  processes, the
available financial data for smelters are aggregated in consolidated  financial
statements which makes smelter data difficult to isolate.  Thus,  an economic
analysis of copper smelters must be cognizant of the economic connection
backward to the mines and forward through the refining stage.

7.1.2  Market Concentration
       Fifteen pyrometallugical copper smelters exist in the  United States.
Copper is also produced in  limited amounts by various hydrometallurgical
methods which by-pass the smelting stage.  These hydrometallurgical fac-
ilities are not being considered in the standard setting process.  The
15 copper smelters have a capacity* of 1,722,600 megagrams**  of copper.
The hydrometallugical processes have a capacity of roughly 10 percent of  the
copper smelters' capacity.
       Table 7-1 shows that the vast majority (approximately  89 percent)  of
smelting capacity is located in the southwestern States of Utah,  Nevada,  New
Mexico, Arizona, and Texas, close to copper mines.  The location  is largely
dictated by the need to minimize shipping distances  of concentrates,  which
are normally 25 percent to 35 percent copper.
       The 15 U.S. copper smelters are owned by 7 large companies. All  7
companies are integrated in that, to various degrees, they own some mining
and milling facilities which produce copper concentrates for  the  smelters.
Several smelters, apart from the concentrates from their own  mines, buy
additional concentrates from other mining and milling producers,
*Capacity is not a static measure of a smelter since capacity can vary, for
  example, according to the grade of copper concentrates processed.
**1 megagram =1.1 short tons.
                                     7-2

-------
                         Table 7-1. SMELTER OWNERSHIP, PRODUCTION AND SOURCE MATERIAL ARRANGEMENTS^
co
Smelter Name
and Location
Tacoma, WA
Hayden, AZ
El Paso, TX
Copper-hill, TN
White Pine, MI
Miami, AZ
McGill, NV
Garfield, UT
Hayden, AZ
Hurley, NM
Magma (San
Manuel, AZ)
Douglas, AZ
Ajo, AZ
Morenci, AZ
Hidalgo, NM
Total Production
Operating Rate
Ownership
ASARCO, Inc.
Cities Service Co.
Copper Range Co.
Subsidiary of The Louisi-
ana Land Exploration Co.
Inspiration Consolidated
Copper Company
Kennecott Corp.
Newmont Mining
Phelps Dodge Corp.

1J80 Rated1
Capacity
(Mg)
91,000
182,000
91,000
13,600
52,000
136,000
45,000
254,000
71,000
73,000C
181,000
115,000
64,000
191,000
163,000
l,722,600d
1979
Production
(Mg)
61,0002
96,000
85,000
12,9003
39,7704
124,0705
296,0006
56,360
145,9007
283,0008
91,000
1,291,000
(74.9%)
1980
Production
(Mg)
42.7002
59,000
47,300
10,0003
32,500*
107,4005
259,3006
46,100
98,0007
324,1008
1,026,420
(59.6%)
Material Arrangements

Integrated -
Custom
Toll
Integrated*3 -
Integrated -
Integrated -
Toll
Integrated -
Toll
Integrated -
Toll
Integrated -
Custom
Toll
Integrated -
Custom
Toll
1979
31%
25%
44%
100%
100%
35%
65%
100%
100%
-

1980
20%
44%
36%
100%
100%
58%
42%
91%
10%
100%
75%
25%
73%
6%
21%
      alnformation  primarily  from  corporate 10-K reports to the Securities  and  Exchange Commission.
      bEstimate  based  on  total  copper  sales for Cities Service minus  the  sales  of  its Arizona mines.
      cEstimated  to  expand  to 110,000  tons.
      dRated capacity  excluding Anaconda smelter which was closed  in  1980 (rated at 180,000 Mg).

-------
smelt and refine the copper, and then sell  it.  This practice is referred to
as custom smelting. Other smelters process  (smelt and refine) the concentrates,
and return the blister copper to mine owners for them to sell, a practice
referred to as tolling.  Some smelters perform both toll and custom smelting.
       It is general industry practice for  companies to operate their smelters
as service centers at low profit margins to the owned mines.  This acts to
shift profits of an integrated operator to  the mines, where depletion allow-
ances exist.  This maximizes profit to the  overall  operation.  An implication
of this policy is that the impact on profits from swings in copper prices is
frequently manifest at the mines more than  the smelters.
       Table 7-1 lists the smelters, their  corporate owners, capacities, 1979
and 1980 production amounts, and the distribution of integrated, custom, and
toll smelting.  Total  production figures and the corresponding operating
rates shown in Table 7-1 are compiled from  corporate reports.  Figures in
Table 7-1 are adjusted to exclude capacity  and production for the Anaconda
smelter, which was closed in 1980.  For 1979, the table shows a 74.9 percent
operating rate.  For 1980, the table shows  that the industry operated at 59.6
percent of capacity.  Production was down for 1980  due to an industry strike.
Following the strike in 1980, production improved in 1981 to 1,380 gigagrams,
for a capacity utilization rate of 80 percent.9  Preliminary figures for
1982 from the Bureau of Mines show a decline in primary copper smelter
production to 1,020 gigagrams, for a capacity utilization rate of about 59
percent.10
       The 3 largest companies account for  78 percent of the entire smelting
capacity.  Phelps Dodge Corporation has the largest smelting capacity,
followed by Kennecott Corporation and then  ASARCO.   The remaining 4 companies
each have 1 smelter and in order of size are Magma  (Newmont), Inspiration,
Copper Range, and Copperhill (Cities Service).
       The table also shows that 73 percent of total  1980 smelter production
was from concentrate from integrated arrangements.   Of the remaining concen-
trate, 21 percent was smelted on a toll  basis and 6 percent smelted on
a custom basis.  Three of the 8 companies process only their own copper
concentrates.
                                     7-4

-------
 7.1.3  Total Supply
        Copper resources are defined as deposits which can be profitably
 extracted at a given price.  Various estimates of U.S. copper resources
 show amounts ranging from 61.8 teragrams to 99.1 teragrams.*  The capability
 of copper resources to meet future demand is dependent upon several factors;
 a principal one being the rate of growth in demand.  The U.S. Bureau of Mines
 estimates that copper demand will grow at an annual growth rate of 3.0
 percent to the year 2000 and that 30 percent of the demand will be supplied
 by scrap.  Therefore, the likely primary copper demand over this period would
 be 55 teragrams compared with 92 teragrams of resources.H  Consequently,
 U.S.  supply appears adequate to the year 2000.   Beyond the year 2000,  demand
 is expected to strain supply sources.   However, increased uses  of old  scrap
 and  possible exploitation of sea nodules can supplement  on-shore mining.   In
 addition,  microminiaturization,  copper  cladding,  and other conservation
 methods will  be  more  widely used  to extend  the  supply of  copper.

        7-1-3-1   Domestic Supply.   Primary refined  copper  output  alone
 does  not  depict  the entire supply of copper  that  is  available for consumption
 in the  United  States.   A large  portion  of copper  scrap does  not  need to be
 resmelted  or  re-refined  and  is  readily  available  for consumption.   Copper is
 a  durable  material  and,  if it  is  unalloyed or unpainted,  etc.,  it can  be
 reused  readily.  Otherwise,  it  is resmelted  or  re-refined  as described
 earlier.   The ready availability  of copper scrap  as  a  secondary  source  of
 supply  tends  to  be  a  stabilizing  influence on producers'  copper  prices.
        The total supply  of copper available  for consumption  in any one  year
 is therefore comprised of  refined  U.S.  production, scrap  not re-refined,  net
 imports, and any changes in  inventory of  primary refined production from  one
year to the next.
        The refined  copper production in 1981 comprised 70.4 percent of  total
copper  consumed in  the United States; scrap not re-refined accounted for 32.0
percent and net refined  imports 10.6 percent (total exceeds 100 percent due
to stock changes).12  Between 1970 and  1981,  67 percent of U.S.  copper
demand, excluding stock changes, was met from domestic mine production; 21
*Teragram is 1.1 million short tons.
                                     7-5

-------
percent was from old scrap, and 12 percent from net imports.  During these
years, total U.S. demand for copper averaged 2,012,000 megagrams per year.
Of this amount, 1,337,000 megagrams was from domestic production, 427,000
megagrams from scrap, and 248,000 megagrams from net imports.
       Another statistic for describing the importance of scrap is to total
the three stages (smelting, refining from scrap, and reuse of scrap) at which
scrap can enter the production process, and compare the figures to total
copper consumption.  In 1981 the percentage of total consumed copper from
scrap was 47.7, roughly the same as in recent years.
       The 1981 refined copper production level was 1,956,400 megagrams.
Although the average for the past several years has shown some improvement,
total refined copper production has not returned to the 1973 peak level.

       7.1.3.2  World Copper.  According to the Bureau of Mines, the world
reserve of copper in ore is estimated at 494,000 gigagrams of copper.  In
addition, an estimated 1,333,000 gigagrams of copper are contained in other
land-based resources, and another 689,000 gigagrams in seabed nodules.  The
United States accounts for 19 percent of known copper reserves and 26 percent
of other land-based copper resources.13
       The United States is the leading copper producing and consuming
country.  Other major copper mining countries include:  Chile, the U.S.S.R.,
Canada, Zambia, Zaire, Peru, and Poland.  Although its copper mining activity
is quite small, Japan is among the three largest countries in terms of copper
smelting and refining.  In 1981 the U.S. produced 18.8 percent of the world's
mine production of copper, 16.5 percent of the smelter production, and 22.2
percent of the refinery production.  The consumption of the world's refined
copper by the U.S.  amounted to about 21 percent.  Table 7-2 shows U.S.
production,  world production, and the U.S. percent of world production for
the years 1963 through 1981.  Although the U.S. is essentially maintaining
its consumption and production levels, world consumption and production is
increasing.   As a result, the U.S. share of world consumption and production
shows a relative decrease.
       In 1981 world consumption of refined copper rose 9 percent to 9,440
gigagrams.14  Stocks of refined copper in the market economy countries
increased 5  percent to 1,100 gigagrams.15
                                     7-6

-------
Table 7-2.   UNITED STATES  AND WORLD COMPARATIVE  TRENDS IN  COPPER PRODUCTION:   1963-198116,17
                                              (Gigagrams)
Years
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
Average Annua)
Compound Growth
Rate (Percent)
1963-1973
1964-1974
1971-1981
~Rine Production of Copper
(Copper Content)
U.S.
1100.6
1131.1
1226.3
1296.5
865.5
1092.8
1401.2
1560.0
1380.9
1510.3
1558.5
1445.7
1282.2
1456.6
1364.4
1357.6
1443.6
1181.1
1538.2



3.54
2.48
1.09
World
4624.3
4798.6
4962.7
5215.9
5057.6
5456.5
5951.2
6387.3
6473.9
7071.5
7591.4
7885.6
6968.2
7525.3
7755.8
7618.3
7674.4
7656.3
8171.1



5.08
5.09
2.36
U.S. as
Percent
of World
23.8
23.6
24.7
24.9
17.1
20.0
23.5
24.4
21.3
21.4
20.5
18.3
18.4
19.4
17.6
17.8
18.8
15.4
18.8



_
_
-
Smelter
U.S.
1176.3
1214.2
1300.9
1330.3
782.3
1148.9
1438.3
1489.0
1360.8
1533.5
1582.1
1424.2
1357.5
1438.5
1346.8
1343.0
1396.0
1053.3
1377.7



3.01
1.61
0.12
Production
World
4634.8
4851.4
5024.4
5167.0
4891.0
5507.8
5972.9
6309.5
6380.0
7003.2
7445.5
7933.6
7535.4
7839.6
8136.9
8017.5
8045.6
7938.9
8324.7



4.85
4.77
1.03
of Copper
U.S. as
Percent
Of World
25.4
25.0
25.9
25.7
16.0
20.9
24.1
23.6
21.3
21.9
21.2
18.4
18.0
18.3
16.6
16.8
17.4
13.3
16.5




_
-
Production of
U.S
1709
1805
1942
1980
1384
1668
2009
2034
1780
2048
2098
1938
1610
1736
1706
1869
2013
1726
2037



2
0
1
Refined Copper
World
.5
.7
.1
.7
.9
.3
.3
.5
.3
.9
.0
.3
.7
.7
.9
.2
.8
.0
.6



.07
.71
.36
5399
5739
6058
6322
6000
6658
7199
7577
7377
8068
8497
8851
8402
8322
8649
8791
8903
8971
9184



4
4
2
.7
.0
.5
.2
.5
.6
.8
.8
.8
.0
.3
.5
.0
.3
.8
.9
.1
.0
.4



.64
.43
.21
U.S. as
Percent
of World
31.7
31.5
32.1
31.3
23.1
25.1
27.9
26.8
24.1
25.4
24.7
21.9
19.2
20.9
19.7
21.3
22.6
19.2
22.2





-
        Note:     One gigagram = 1,000 roegagrams.  One megagraro  (1,000 kilograms) equals 1.102311 short tons
                 (907.185 kilograms = 2000 pounds avoirdupois,  where one pound avoirdupois equals 0.453592
                 kilogram or 453.5924 grams).

-------
7.1.4  U.S. Total Consumption Of Copper
       Total copper consumed in the United States over the last 12 years
has fluctuated considerably but shows an overall upward trend.  However,
copper consumption has not returned to its 1973 peak.  This conclusion is
derived from data on copper consumption from refineries and copper consumption
from refineries plus scrap.
       Table 7-3 shows each set of data for the years 1970 through 1981.  The
5-year averages in gigagrams for copper consumption from refineries has
increased by 6.9 percent (1972 through 1976 is 1,891.9 and 1977 through 1981
is 2,021.5.).  Five-year scrap consumption has shown an increase of 5.1
percent, from 848.6 gigagrams for the 1972 to 1976 period, to 892.3 gigagrams
for the 1977 to 1981 period.  There are signs that the consumption of scrap
has begun to increase over the last few years.
       The Bureau of Mines forecasts a long-range overall  consumption growth
rate to the year 2000 of 3.0 percent per year.  The combined 3.0 percent
growth rate is composed of a 2.4 percent growth rate for primary copper, and
a 4.8 percent growth rate for secondary copper.19

       7-1.4.1  Demand By End-Use.  Refined copper and copper scrap are
further processed in a number of intermediate operations before the copper is
consumed in a final product.  Refined copper usually consists of one of the
following shapes:  cathodes, wire bars, ingots, ingot bars, cakes, slabs, and
billets.  These shapes plus the copper scrap then go to brass mills, wire
mills, foundries, or powder plants for subsequent processing.  The copper is
frequently alloyed and transformed into other shapes such  as sheet, tube,
pipe, wire, powder, and cast shapes.  Ultimately, the copper is consumed in
such shapes in five market or end-use categories.  The Copper Development
Association, Inc. uses the following categories:  building construction,
transportation, consumer and general products, industrial  machinery and
equipment, and electrical and electronic products.
       Table 7-4 shows the demand for copper in each of these five markets
over the 12-year period 1970 through 1981.  The total figures for these
five markets will not equal the total  consumption figures  of Table 7-3
                                     7-8

-------
                                         Table 7-3.   U.S.  COPPER CONSUMPTION^

                                                   (Gigagrams)3
                    T970197119721973    T974    1975T976    T9771978    1979    i960    I981r
Consumption of
 Refined Copper

Consumption of
 Scrapc

Total Consumed
 Copper

% of Total as
 Scrap

1972-1976 and
 1977-1981
 averages for
 consumption
 refined

1972-1976 and
 1977-1981
 averages for
 consumption
 of scrap

1972-1976 and
 1977-1981
 averages for
 total copper
 consumption
1859.8  1833.0  2029.5   2220.6   1998.6  1398.9  1811.7  1989.2  2196.5  2164.1  1876.3  1890.6


 811.6   859.5   946.0    956.4    878.0   662.4   800.4   847.6   881.1   984.3   860.0   888.6
2671.4  2692.5  2975.5  3177.0  2876.6  2061.2  2612.2  2836.8  3077.5  3148.4  2727.3  2779.1


  30.4    31.9    31.8    30.1    30.5    32.1    30.6    29.9    28.6    31.3    31.5    32.0
                1891.9
2021.5
                 848.6
 892.3
                2740.5
2913.8
al gigagram = 2.2 million pounds.

^Preliminary.

cWithout having to be refined again.

-------
 I
I—>
o
                                      Table 7-4.  U.S. COPPER DEMAND BY MARKET END USES20
                                                           (Gigagrams)3
Market 1970
Building Constr. 748
Transportation 267
Consumer and
General Products 369
Industrial
Machinery and
Equipment 560
Electrical and
Electronic
Products 717
Total 2661
5-Year Average
Demandc
Building
Construction
Transportation
Consumer and
General Products
Industrial
Machinery and
Equipment
Electrical and
Electronic
Products
1971 1972 1973 1974
830 923 998 795
307 345 402 332

358 417 413 402


540 601 640 563


742 773 838 766
2777 3059 3291 2858



826.0
343.4

375.2


538.2


720.2
1975 1976 1977 1978
634 780 888 956
265 372 412 422

316 328 340 388


404 483 518 551


541 683 767 801
2161 2646 2925 3118



877.0
360.4

373.2


534.4


789.8
1979 1980 1981b
951 774 816
388 270 310

420 342 376


565 501 537


853 754 774
3177 2641 28x3













      ai  Gigagram =2.2 million pounds,
      ^Preliminary.
      c(1972-1976,  and  1977-1981).

-------
due to the effects of stock changes and imports on fully fabricated copper
products.
       A look at the 5-year average demand shows that there has been an
increase in three out of the five markets.  The building industry market
sales showed a gain of 6.2 percent.  The transportation market shows a gain
of 5.0 percent.  An increase of 9.7 percent occurred in the electrical and
electronic product markets.  The demand for electrical  equipment has risen
because of increased emphasis on safety, comfort, recreation, and a pollution-
free environment.  Automation, including the use in computers, has also
boosted the use of copper.
       Substitution of other materials, coupled with the recession, has
caused the slight drop of  less than 1 percent  in the consumer and general
products markets.  The 1 percent decline  in the industrial machinery and
equipment market  is largely due to the  impact  of the recession.
       The Bureau of Mines estimates that the  most growth  in copper demand
will occur in the electrical and electronic products industries, consumer and
general products, and building construction.   Copper is an important metal in
electric vehicles.  If electric vehicles become popular, this would be a
source of  increased demand for copper.  General Motors plans to produce an
electric family car for mass marketing  in the  mid-1980's.  A conventional
internal combustion automobile contains from 6.8 to 20.4 kg of refined
copper, whereas 'electric vehicles  use much more copper.  The Copper Development
Association estimates range from 45.4 kg  to 90.7 kg, with  an average  nearer
to 45.4  kg.21
       Another potential area for  growth  is in the solar energy  industry.
Presently, the extent of this sector  is relatively modest, consuming  approxi-
mately 4,500 Mg/yr  of copper  in  the U.S.  However, consumption  in  this  sector
has  the  potential to climb considerably.
        In  addition, the U.S. military demand for  copper  is expected to
 increase.   Increased military expenditures will  have a  significant  impact on
copper demand  because copper  is  an important element in modern  electronic
weaponry.  During  heavy rearmament  periods the  military  demand  for  the
metal has  reached 18 percent of  copper  mill shipments.  Although military
demand  is  not  expected to  return to the record high  18  percent  level,  analysts
do expect  a  large  increase in military  requirements  for copper  from the  low
level  in  1979  of  less than 2 percent.22

                                      7-11

-------
       The demand picture in the United States may receive a boost from the
federal government.  The government is committed to eventually acquire 1.1
gigagrams of copper for its currently depleted strategic stockpile.  The
previous stockpile was largely depleted in 1968; the final sale was in 1974
after copper prices had soared.  Further Congressional action is necessary to
implement and fund the purchase plan.

       7-1-4.2  Substitutes.  Substitutes for copper are readily available
for most of copper's end uses.  Copper's most competitive substitute is
aluminum.  Other competitive materials are stainless steel, zinc, and plastics.
Aluminum, because of its high electrical conductivity, is used extensively as
a copper substitute in high voltage electrical transmission wires.  Aluminum
has not been used as extensively in residential wiring because of use problems,
and minimal savings.
       Aluminum is also potentially a substitute for copper in many heat
exchange applications.  For example, automobile companies are still experi-
menting with the use of aluminum versus copper in car radiators.  When copper
prices are high, or copper supply is limited, cast iron and plastics are used
in building construction as a copper pipe substitute.  A relatively new sub-
stitute for copper is glass, which is used in fiber optics in the field of
telecommun ications.

7.1.5  Prices
       Numerous factors influence the copper market, and thus the price of
refined copper.  These factors include:  production costs, long-run return on
investment, demand, scrap availability, imports, substitute materials,
inventory levels, the difference between metal exchange prices and the
refined price, and federal government actions (e.g., General Services
Administration stockpiling and domestic price controls).
       Among the many published copper price quotations, two key price levels
are:  1)  those quoted by the primary domestic copper producers and 2) those
on the London Metal Exchange and reported in Metals Week, Metal Bulletin, and
the Engineering and Mining Journal.  The producers' price listed most often is
for refined copper wirebar, f.o.b. refinery.  The London Metal Exchange price,
                                     7-12

-------
  referred to as LME,  is also for copper sold as wirebar.  The LME is generally
  considered a marginal  price reflective of short-term supply-demand conditions,
  while the producer price is more long-term and stable and often lags the LME
  price movement.
         Copper is also  traded on the New York  Commodity Exchange (Comex).
  Arbitrage keeps  the  LME  price and  the  Comex price  close  together (with  minor
  price differences due  to different contract terms  on the two  exchanges,  and
  a  transportation differential).
         Table  7-5 shows the  LME,  the  U.S  producer price,  and the U.S.  producer
  price adjusted to a  1982  constant  price  for the years  1970 through  1982.
  Data  were obtained from  U.S.  Bureau  of Mines  publications.
         Several points can be  observed from the table with respect  to  the LME
  price versus  the  U.S. producer  price:  (1) the LME price has  had wider
  swings  than the  producer price;  (2)  in the past when both prices are  relatively
  high,  the LME price has been  considerably higher than the producer price,
  while  during relatively low price periods, the producer price has been
 moderately higher than the LME price; and (3)  in recent years a marked change
  appears to be taking place away from a two-price system and toward a one-price
 system, with the difference between the LME and the U.S.  producer price
 accounted for only by a transportation  differential.   These earlier situations
 had reoccurred repeatedly over the past 20 years.   One other point about
 the table should  be mentioned, although  unrelated  to  the  relationship of the
 LME to the producer price.   The producer  price has  not kept  pace with general
 inflation.
        In theory,  the U.S. producer price should be somewhat higher than the
 LME price since ocean transport  costs must  be  incurred  to  get  the refined
 copper to the  U.S.  However,  this relationship  appears  to  hold only during
 slack  price periods.  When LME prices are high, the producers  do not  raise
 their  prices as much, which  in theory appears  contrary  to profit maximization.
 Explanations offered for such  behavior include:  the  producers'  fear of
 long-run  substitution for copper  if the producers raised the price to  the
fabricators, high profits for  integrated fabricators while reducing supply to
nonintegrated fabricators, past fears of government stockpile  sales that
would reduce prices, and fear of the return of government  intervention
through price controls.
                                     7-13

-------
             Table  7-5.  AVERAGE ANNUAL COPPER PRICES23,24,25
                            (cents per kg)a
Year
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
19826
LMEb
138.6
106.7
106.7
178.0
204.8
123.4
140.6
130.7
136.2
198.2
218.5
174.7
147.4
U.S Producer Pr1cec
128.0
114.4
112.6
130.9
170.1
141.2
153.1
147.0
146.3
205.3
225.3
187.2
162.8
U.S. Producer
1982 Constant
290.9
248.7
234.6
256.7
303.8
231.5
239.2
216.2
200.4
259.9
262.0
199.1
162.8
Price
Priced













aTo convert from cents/kg to cents/lb,  multiply by 0.454.
bLondon Metal Exchange "high-grade"  contract.
CU.S producer price, electrolytic wirebar copper,  delivered U.S destinations
 basis.
dAdjusted to 1982 constant price by  applying  implicit price deflator for
 gross national  product (1972 = 100).
ePreliminary.
                                     7-14

-------
          The cost of producing copper  is one of the elements that  influences
  the price of copper.  Considerable data exist to validate the point that the
  long-run economic cost of producing copper is increasing. 26  During the
  early 1970's the capital costs per megagram of annual capacity for developing
  copper from the mine through refining stage were $2,000 to $2,500, and by the
  late 1970's had risen sharply to $7,200 to $7,700.   Estimates are that a
  price of $2.76 per kg to $3.30 per kg for refined copper would be needed to
  support such  new capital  outlays.
         The above costs are for conventional  pyrometal lurgical smelting.   The
  newer  smelting  processes  such  as  Noranda  and  Mitsubishi  offer some capital
  cost  savings  at that  stage due  to  lower pollution control  costs.   The  hydro-
  metallurgical processes  also reouire  less  capital.   However,  the  mining  costs
  are the  highest part  of overall development costs for which limited cost
  saving techniques exist.   The mine development costs in  the U.S.  have  risen
  significantly,  largely as  a result of the  shifting from  higher to  lower
  grades of available copper ores and sometimes remote locations that require
  infrastructure  costs  for towns, roads, etc.
        In 1979, the Bureau of Mines analyzed 73 domestic copper properties to
 determine the quantity of copper available from each deposit  and the copper
 price required to provide each operation with 0 and  15 percent rates of
 return.   The Bureau estimates that a copper price of $4.56 per kg would be
 required  if all  properties, producing  and  nonproducing,  were to at least
 break  even.   The average  break-even copper price  for properties producing in
 1978,  $1.46  per  kg,  was about  equivalent to the average  selling price  for
 the year.   At  this  price,  analysts  calculate  that  only 25 properties could
 either produce at break-even or  receive  an  operating  profit. Of these  proper-
 ties, only 12 could  receive  at  least a 15  percent  rate of return
       Annual domestic  copper production,  from  1969 to 1978, averaged 1,337,000
megagrams.  According  to this study, in order to produce  at this level  and
receive at least a 15  percent rate of return, a copper price of $1.81 per kg
is required.  If the United States were to produce the additional 248 000
megagrams that were imported each year over this period,  a copper price of
$1.94 would be  necessary.27  Tne report concludes  ^ .^^ ^
prices are required in order for many domestic deposits to continue to
produce.

                                     7-15

-------
       It has been suggested  that long-term potential  for higher prices,
plus the high cost of new capacity are significant reasons for the increased
purchases several  years  ago of copper properties  by oil  companies.  The
reasoning is that  oil companies need places for heavy cash flows,  and diver-
sification to other products  is desirable.   The oil companies reportedly can
wait for expected  copper price increases to obtain their return.  Further,  by
purchasing existing facilities, rather than building new capacity, they avoid
the escalating new capacity costs.  However, more recently, some oil  com-
panies seems to be rethinking their investments in copper.
       As shown below, U.S. oil (and gas) companies own  or have major interests
in many of the largest domestic copper producers:

       1.   Amax - Approximately 20 percent owned by Standard Oil  of California
       2.   Anaconda - Owned by Atlantic Richfield Company (ARCO)
       3.   Cities Service - Also a primary copper producer
       4.   Copper Range - Owned by Louisiana Land and Exploration Company
       5.   Cyprus Pima Mining Company - Standard Oil Company (Indiana)
       6.   Duval  - Owned by Pennzoil Company
       7.   Kennecott - Standard Oil of Ohio (British Petroleum)

These copper producers own or control a  large portion of domestic copper
reserves, mine production, and U.S. refinery capacity.  Their investment in
the copper  industry  is significant, and  thus they must expect higher prices
and substantial profits in the future.
                                      7-16

-------
   7.2     ECONOMIC ANALYSIS
   7.2.1  Introduction
         This section presents the economic impact analysis of the arsenic
   NESHAP for the 14 low arsenic primary copper smelters.  The fifteenth
   primary copper smelter is classified as a high arsenic smelter and is dis-
   cussed in a separate analysis.
         The principal economic impacts analyzed are:  the ability of the
   smelters to increase copper prices in response to an increase in costs due to
  the arsenic standard;  and, the impact on profits if part or all  of the costs
  cannot be passed  on  in the form of price increases.  Section 7.2.3 presents
  the methodology.   Section  7.2.4 presents the  impact on prices,  Section 7  2 5
  presents the  impact  on profits,  and  Section 7.2.6 presents  a discussion of
  capital  availability.

  7.2.2  Summary
        In  1982 the copper  producers experienced one of  the worst years in
  recent history.  Such  a situation cannot be used as the foundation to examine
  the long term economic impact of the potential arsenic NESHAP.  Therefore the
  economic analysis is based on a more normal condition for the industry
  However,  even under more typical conditions for the industry, several  smelters
 may face significant financial impairment.
        If each smelter attempts to  pass control  costs forward in the form  of
 a price increase,  the price increases would range from 0 percent to 15.2
 percent at a 80  percent capacity utilization rate, depending on  the regulatory
 alternative.  For  Alternative II  the  price increase would  be 0 for every
 smelter,  with  one  exception that would  have a  15.2 percent increase.   For
 Alternative III the price increases would range  from 0.1 percent  to 6  3
 percent.   For Alternative IV  the  price  increases  are lower and would range
 from 0  to  1.3 percent.  For Alternative  III+IV the  price increases would
 range from  0.1 percent  to 7.6 percent.   Competition  will prevent the existence
 of such a broad variation.
        If control costs are absorbed and profit margins reduced,  again a
broad range exists.  At an 80 percent capacity  utilization rate and a ten
percent profit margin, for Alternative II the profit decrease would be 0 for
every smelter,  with one exception that would result in a net  loss   For
                                     7-17

-------
Alternative III the profit decrease would range from 1.1 percent to 62.6
percent.  For Alternative IV the profit decrease would be lower and would
range from 0 percent to 12.8 percent.  For Alternative III+IV the profit
decrease would range from 1.1 percent to 75.4 percent.
       Although the capital costs of the control equipment are not minor
amounts, for most of the producers the capital cost would not present a major
obstacle.  For two of the producers the capital costs may present some
difficulty.

7.2.3  Methodology
       The purpose of this section is to explain in general  terms the method-
ology used in the analysis.  Each of the appropriate sub-sections explains
the methodology in more detail.  No single indicator is sufficient by itself
to use for decision making purposes about the primary copper smelters.
Therefore the methodology relies on several  indicators which in total can be
used to draw conclusions about the industry.
       The methodology has three major parts.  The first part is an analysis
of price impacts.  The analysis of price impacts introduces  an upper limit on
the problem and provides a benchmark to make evaluations on  a relatively
uncomplicated basis.  A price increase represents the "worst case" from the
viewpoint of a consumer of copper.  The second major part of the methodology
is an analysis of profit impacts.  The analysis of profit impacts introduces
a lower limit on the problem and is the "worst case" from the viewpoint of
the firm.  The individual  characteristics of each smelter increase in
importance and are incorporated to a greater extent.  The third and final
part is an analysis of the availability of capital  to purchase the control
equipment.
       Firms in the copper industry face a wide variety of variables that in
the aggregate determine the economic viability of the firm generally, and a
smelter specifically.   The variables can be  grouped in four  broad categories.
The categories are described here separately and in a simplified manner for
discussion purposes.  However,  there is a close interrelationship among the
four categories and changes in  one will  have implications for the others.
The four broad categories which encompass the variables that in turn deter-
mine the economic viability of  the smelter are described below.
                                     7-18

-------
          1)  Macro-economic conditions.  The two most prominent variables in
   this category are copper prices and copper demand.  By-products and co-products
   represent a significant source of revenues for most copper operations.
   Therefore in addition to the price of copper, the price of by-products and
   co-products also influence an assessment of economic viability.   Common
   by-products and co-products of copper production include: gold  silver
  molybdenum, and sulfuric acid.   Other by-products include selenium, teilurium
  and antimorv.   For ease of presentation and in order to present  a  conserva-
  tive analysis,  by-products and  co-products are not considered explicitly  in
  the analysis.   Another important  variable, though,  somewhat  less visible   is
  government actions  such as  tax  policy,  stockpiling,  and price controls.  Vhe
  government variable  includes  the  U.S. Government,  as well  as  foreign govern-
  ments.   For example, consider that a  report by the  U.S.  Bureau of  Mines
  has stated  that  at least 40 percent of  the total mine production of copper in
  market  economy countries was produced by firms in which  various  foreign
  governments owned an equity interest. 28
         2)  Environmental regulations.  Since roughly 1970, environmental
  regulations have evolved to the point that they have become a major variable
  that must be considered in the corporate decision making process.  Here
  again, government actions are important.
        3)  Corporate organizational  strategy.   This category  includes the
 corporation's  strategy  with respect  to variables  such as remaining  or becom-
 ing an integrated copper producer  versus a  non-integrated copper  producer   or
 leaving  the industry  entirely.
       Many of the companies  that  produce refined  copper are  integrated
 producers;  that  is, they own  the facilities to  treat copper during  each  of
 the four principal stages of processing: mining, milling,  smelting, and
 intth, K'                Pr°UCerS 3re 1nt69rated °"e «"»«on.l step
 into  the fabncat,on of refined copper.  However, not all companies in the
  pper industry are integrated producers.  There are companies that only mine
 and mill copper ore to produce copper concentrate, and then have the copper
concentrate smelted and refined on a custom basis (the smelter takes owner-
     of   e copper) or on a toll  basis (the smelter charges a service fee and
returns the copper to  the owner).   The existence of both  integrated and
non-integrated producers introduces a complex economic element into this
                                    7-19

-------
analysis.  That complex economic element manifests itself in the choice of
the appropriate profit center.  This standard affects only one stage of the
production process (smelting) in a direct way, but has indirect effects on
the other stages.
       For accounting purposes, integrated producers frequently view the
smelter as a cost center, rather than a profit center.  However, in an economic
sense the smelter provides a distinct contribution to the production process
that ultimately allows a profit to be earned although that profit may be
realized for accounting purposes at another stage such as the mine or refinery.
       4)  Competition.  Mines have long-run flexibility in deciding where
they will send their copper concentrate for smelting. Therefore, copper
smelters face competition from three sources: other existing domestic smelters,
new smelters that may be built, and foreign smelters, especially Japanese.
Other competition, though less direct, is also important.  For example, scrap
and substitutes present competition.
       Japan is a major force among copper producing countries in terms of
its volume of smelting, refining, and fabrication of copper.  However, Japan
does not have copper ore deposits of any noteworthy size.  Therefore it must
import concentrates in order to supply its smelting, refining, and fabri-
cating facilities.  Japan seeks concentrates from many countries, including
the United States.  Japan's ability to be competitive with domestic smelters
for U.S. concentrates is indicated by the contractual arrangements it has
established with Anamax and Anaconda to purchase concentrates.  Also, the
Japanese smelters have approached many other copper mine owners in the United
States.  For example, Cyprus Corporation is reported to have seriously
considered shipping concentrates from its Bagdad mine to Japan.
       The cost to transport concentrates across the Pacific Ocean is signi-
ficant.  The fact that Japanese smelters can compete with U.S. smelters, in
spite of the costs to transport concentrates across the Pacific Ocean, is
quite noteworthy.  One factor that explains the Japanese ability to compete
is that Japanese smelters are newer than U.S. smelters and,  in theory, should
be more cost competitive.  Other factors that operate to the advantage of
Japanese smelters, including a tariff mechanism, are described later.
       The existence of competition for concentrates introduces what is
commonly referred to as a "trigger" price.  The "trigger" price is that price
                                     7-20

-------
   which  triggers  or  provides an economic  incentive  for the supplier of concen-
   trate  to  change to another smelter and  refinery.   If a given smelter charges
   a  service  fee in excess of competing smelters, that smelter will lose business
   and eventually  be  forced to cease operations.  In the case of new smelters or
   expansions, the new process facilities will not be built.  Faced with an
   increase in costs, a smelter could respond using one of three options, or any
   combination of  the three.  First, the smelter could pass the costs forward in
   the form of a price increase.   Two important considerations with respect to a
   pnce increase are: the prices of competitors in  the copper business,  and the
  elasticity of demand  for the  end users  of copper.   For  example,  even if all
  copper producers experience the  same  increase in  costs,  at  some  point  the end
  users of copper  will  consider  changing  to a substitute.   Second,  the smelter
  could absorb the cost increase by reducing its  profit margins, thereby
  reducing its return on  investment (ROD.   If  the  smelter's  profit margins are
  reduced significantly it  will cease operation.  Third, the  smelter could  pass
  the costs  back to the mines by reducing  the price  it is willing to pay for
  concentrate.  An important consideration  in setting the service fee  a smelter
  charges  for  custom or toll smelting is that the concentrate may be shipped
  elsewhere, such  as to Japan.  Market conditions suggest that the option of
  passing costs back to the mines does not seem feasible at this time, due to
  the existence of excess smelting capacity.

        7'2'3"1  JaPa"ese Tarif^jecjianisrn.  One example  of foreign government
 assistance to the copper industry occurs in Japan.   Japanese copper producers
 operate under a  system that  permits the  payment of a premium for  concentrates
 which  is then recovered  through a premium for  refined copper due  to  a protected
 internal  market supported  by a  high tariff.  Japan  imposes high import  duties
 on refined,  unwrought  copper while allowing concentrates to  be shipped  into
 the  country  duty-free.   Duty on refined  unwrought copper in  1981 was  8 2
 percent  of  the value of  the copper, including  freight and insurance   as
 opposed  to a  U.S. customs duty of  1.3 percent of the value of copper.  The
 ^port duties allow Japanese producers to sell their refined copper in Japan
   an artificially high price and still remain competitive with foreign
   rln/~Qi^o                                                        vnv-iyii
producers.
       Specifically, copper concentrates and ore imported into Japan are free
of duty.  Refined copper imported into Japan is subjected to a tariff of
                                     7-21

-------
15,000 yen/Mg.29  Using a December  15,  1980,  exchange  rate  of $0.004633/yen,
the tariff was $0.0849/kg.  Refined copper may be duty-free under  the  preferen-
tial tariff, subject to certain  limitations.
       As a result of the tariff situation, Japanese copper producers  can pay
a premium to attract concentrates and can recover the  premium through  a
premium on the price of the refined copper used in Japan.   If the  refined
copper is returned to the customer  outside of Japan, the premium on the price
of refined copper is not recovered  because world prices would prevail  in this
case, rather than the protected  internal  Japanese producer  price.   As  a
result, the principal interest of the Japanese copper  producers is in  produc-
ing copper for internal consumption.  Toll smelting in Japan is generally
used as a means of balancing inventories.  The absence of a tariff on  ore and
concentrates encourages companies to import ore into Japan.  The presence of
a tariff on refined copper and the  costs  of holding metal  in Japan discourage
companies from importing refined copper into Japan.
       The Japanese tariff on refined copper, combined with the cost of
holding the metal until users have  a demand for it, provides an extra  margin
for domestic copper producers.  The Japanese producers can  charge  what the
market will bear for their copper and still remain competitive with the
importers.  The loss incurred by Japanese producers in charging toll custo-
mers low processing rates is covered by the extra margin of profit realized
by charging prices for domestic  refined copper at competitive import levels.
       Robert H. Lesemann (industry expert, formerly with Metals Week, now
with Commodities Research Unit), in an affidavit for the Federal Trade
Commission, outlined the situation  in September 1979:

       It is generally true that operating costs of U.S. smelters
       are the same as smelters in  Japan, Korea, and Taiwan.  The
       competitive advantage is without doubt due to the subsidies
       outlined above.  Thus, while the terms of the Nippon-Amax
       deal have not been revealed, the treatment charge is likely
       well below the operating cost levels of U.S. smelters.30

       7.2.3.2  Other Japanese advantages.  The tariff mechanism described
above is one example of government assistance to the Japanese copper industry.
Another example is provided by the Japanese government's approval  of a brass

                                     7-22

-------
  rod production cartel.  In an effort to reduce stocks and boost profit
  margins for the ailing Japanese brass rod industry, the government approved
  the formation of a temporary cartel  to cut production.31
         Apart from government assistance,  other reasons  are cited for the
  advantage of the Japanese  copper industry over the  U.S.  copper  industry.
  Additional  reasons include:

         •  A high debt-to-equity ratio-a  typical Japanese  smelter may  have  a
            debt-to-equity ratio  of  0.8 to  0.9.32,33,34

         .  Lower  labor  rates-Japanese  hourly rates  in the  primary metals
            industry were estimated  to  be about two-thirds of the U.S. rate in
            1978.35

        .  By-product credits-the market for by-products, sulfuric acid, and
           gypsum is better in Japan than in the United States and reduces
           operating costs significantly.36

 7-2-4  Maximum Percent Price Increase
        Insight into the economic impact of the  arsenic  NESHAP can be gained
 by examining the  maximum percentage copper price increase that would occur if
 all  control  costs were  possed forward.  A  complete pass  forward  of control
 costs may not be  possible in  every  case, and later in  the analysis this
 assumption  is relaxed.   However,  the  initial assumption  that  a complete pass
 forward  is  possible in  every  case introduces a common reference  point,  which
 then  facilitates  comparisons  of  various control  alternatives and  scenarios
       The maximum percentage price increase is  calculated  using a simplified
 approach, for ease of presentation, that divides annualized control costs by
 the appropriate production and further divides that result by the refined
 price of copper, with the result expressed as the necessary percentage price
 increase per  kilogram.  The above approach does not consider the  investment
tax credU, and thus is a conservative approach  that  will  tend to overstate
the impact of the  control costs.   Other approaches could  be used  to determine
price increases.  For  example, a  net present value (NPV)  approach could  be
                                     7-23

-------
 used.  A net present value  approach  determines  the  revenue  increases  necessary
 to exactly offset the control  costs,  such  that  the  NPV  of the  plant remains
 constant.   An NPV analysis  can also  take into account the investment  tax
 credit, depreciation over the  applicable time period, income taxes, operating
 and maintenance costs,  and  the time  value  of money.  Although  the NPV approach
 is a more  sophisticated calculation,  the two approaches yield  similar results.
 Therefore, the first method is preferable  in this particular case due to its
 straightforward nature,  ease of presentation, and reasonable results.
        Table 7-6 shows  the  cost increase,  and then Table 7-7 shows the
 maximum percentage price increase, of arsenic controls  for  low-arsenic
 primary copper smelters.  The  increase in  the cost of production is shown for
 two capacity utilization rates,  100 percent and 80 percent.  The advantage of
 presenting two capacity  utilization rates  is in the conduct of sensitivity
 analysis.   A rate  of 100 percent is optimistic, but is useful here as a
 reference  point.   A rate of 80  percent is more likely and as noted in Section
 7.1 this is  the  approximate industry  average utilization rate achieved in
 1981.   For 1982, the  industry  average capacity utilization rate was substan-
 tially  lower at  59  percent.  However, no analysis is shown here of the impact
 of control  costs on  the  industry at a 59 or 60 percent utilization rate
 because  regardless  of control costs, a rate of 60 percent is damaging to the
 industry even  as a  baseline condition.  Alternatives II, III, and IV are
 shown as well  as the combination of III+IV.  The smelters are ranked accord-
 ing  to  the cost of Alternative  III+IV (with the exception of Kennecott-McGill).
 The  purpose  of showing the increase in production cost is to supplement the
 maximum percentage price increase.   One advantage of reviewing  the cost
 increase is  that it is only dependent on  the capacity utilization rate,  and
 is  not affected by the refined price of copper.   A second advantage is that
 it  is not  affected by the choice of the profit center.   Several  points should
 be  observed  from the cost increases:
       1)  The amount of the cost increases are  substantial  for two  of the
 smelters.  The cost increases are substantial  for several  reasons.   First,
copper is a commodity, which means  that product  differentiation is  not
 possible and thus competition is based almost  exclusively on price.  The
copper producers can be characterized as  price-takers and thus  no individual
producer controls the marketplace.   Therefore, in an  industry that  competes
based on price, the cost of  production becomes exceptionally important.

                                     7-24

-------
                         Table  7-6.   INCREASE  IN  COST OF PRODUCING COPPER DUE TO ARSENIC
                                     CONTROLS  FOR LOW-ARSENIC PRIMARY COPPER SMELTERS
IN3
en
Smel ter
ASARCO-Hayden
ASARCO-E1 Paso
Kennecott-Garfield
Phelps Dodge-Hidalgo
Phelps Dodge-Morenci
Inspiration
Magma
Phelps Dodge-Ajo
Phelps Dodge-Douglas
Copper Range- White Pine
Kennecott-Hayden
Kennecott-Hurley
Cities Service-Copperhill
Kennecott-McGill
Annual
Capacity
(Mg.)
182,000
91,000
254,000
163,000
191,000
136,000
181,000
64,000
115,000
52,000
71,000
73,000
13,600
45,000
•— • " i —
II
0
0
0
0
0
0
0
0
0
0
0
0
0
2.7
100% Capacity
Alternative
HI TV
0.2
0.3
0.5
1.1
1.0
2.1
2.2
2.4
2.5
2.5
2.8
3.7
9.4
6.0
0
0.2
0.2
0.2
0.3
0.2
0.3
0.4
0.4
0.5
0.4
0.4
1.9
0.6
HTJ.TU
1 + 1 y
0.2
0.5
0.7
1.3
1.3
2.3
2.5
2.8
2.9
3.0
3.2
4.1
11.3
6.6
IT
I
0
0
0
0
0
0
0
0
0
0
0
0
0
28.4
80% Capacity
	 Alternative
I
0.2
0.4
0.6
1.3
1.2
2.6
2.7
3.1
3.1
3.1
3.5
4.6
11.7
7.5
IV
0
0.2
0.3
0.2
0.3
0.2
0.4
0.5
0.6
0.6
0.5
0.5
2.4
0.7
III+IV
0.2
0 6
fl Q
1.5
1 C
2 8
•? i
3 6
•3 7
o . /
•3 7
A n
t. u
*; i
j.i
14 1
8.2

-------
   Table 7-7.  MAXIMUM PERCENT PRICE INCREASE FOR ARSENIC CONTROLS FOR LOW-ARSENIC PRIMARY COPPER SMELTERS
I
ro




	 —~IE
1777
100% Capacity
Alternative
Smel ter
ASARCO-Hayden
ASARCO-E1 Paso
Kennecott-Gar field
Phelps Dodge-Hidalgo
Phelps Dodge-Morencl
Inspiration
Magma
Phelps Dodge-Ajo
Phelps Dodge-Douglas
Copper Range-White Pine
Kenn.ecott-Hayden
Kennecott-Hurley
Cities Servlce-Copperhll
Kennecott-McGill
II
0
0
0
0
0
0
0
0
0
0
0
0
1 0
12.1
III
0.1
0.2
0.3
0.6
0.5
1.1
1.2
1.3
1.3
1.3
1.5
2.0
5.0
3.2
IV
0
0.1
0.1
0.1
0.2
0.1
0.2
0.2
0.2
0.3
0.2
0.2
1.0
0.3
III+IV
0.1
0.3
0.4
0.7
0.7
1.2
1.4
1.5
1.5
1.6
1.7
2.2
6.0
3.5 15
II
0
0
0
0
0
0
0
0
0
0
0
0
0
.2
Tcg^r
rTce~


80% Capacity
Alternative
III
0.1
0.2
0.3
0.7
0.6
1.4
1.4
1.3
1.7
1.7
1.9
2.5
6.3
4.0
IV
0
0.1
0.2
0.1
0.2
0.1
0.2
0.3
0.3
0.3
0.3
0.3
1.3
0.4
III+IV
0.1
0.3
0.5
0.8
0.8
1.5
1.6
1.6
2.0
2.0
2.2
2.8
7.6
4.4 10
II
0
0
0
0
0
0
0
0
0
0
0
0
0
.3

220 d/kg.PHce
100% Capacity
Alternative
III
0.1
0.1
0.2
0.5
0.5
1.0
1.0
1.1
1.1
1.1
1.3
1.7
4.3
2.7
IV
0
0.1
0.1
0.1
0.1
0.1
0.1
0.2
0.2
0.2
0.2
0.2
0.9
0.3
1II+IV
0.1
0.2
0.3
0.6
0.6
1.1
1.1
1.3
1.3
1.3
1.5
1-9
5.2
3.0 12
11
0
0
0
0
0
0
0
0
0
0
0
0
0
.9
80% Capacity
Alternative
HI
0.1
0.2
0.3
0.6
0.5
1.2
1.2
1.1
1.4
1.4
1.6
2.1
5.3
3.4
TV
0
0.1
0.1
0.1
0.1
0.2
0.2
0.1
0.3
0.3
0.2
0.2
1.1
0.3
III+IV
0.1
0.3
0.4
0.7
0.6
1.3
1.4
1.3
1.7
1.7
1.8
2.3
6.4
3.7

-------
  Second, copper is traded on an international basis and thus domestic pro-
  ducers compete among themselves, as well as against foreign producers that
  may not experience the same cost increases.  Finally, copper is faced with a
  significant threat from substitutes:  such as, aluminum and plastic.
         2)  Within a single alternative,  the differences among smelters are
  substantial.  As  described above,  copper producers compete principally on
  price.   As a result,  the cost of production is  quite  important.   Therefore
  differences in costs  among smelters of as  little  as several  cents are
  important.
         3)   The cost increases  for  Alternative II  are  0  in  every case with  one
  exception, Kennecott-McGill.   The  cost increases  for  Alternative  III range
  from a  low of  0.2*/kg to  a  high of 9.4*/kg.   The  costs  for Alternative  IV
  are lower,  and  range from 0 to 1.9^/kg.  The  costs  for Alternative III+IV
  range from 0.2*/kg to 11.3*!/kg.
         Table 7-7 shows maximum percentage price increases.  The purpose of
  reporting  the maximum percentage price increase figures is to add perspective
  to the cost increase figures.  Results are shown for two refined copper
  prices (187 cents  per kg. and 220 cents per kg.),  and  for the same two
 capacity utilization rates presented earlier, 100  percent and 80 percent.
 The price increase assumes the firm is an integrated producer.  The  average
 annual  price for refined copper over the  past 5  years, from 1978 to  1982  has
 been  approximately 187  cents per  kg.  The same cases are shown as  were
 presented earlier  for  the cost increases,  Alternatives II,  III,  IV and
 III+IV.   The price  of copper is difficult to predict,  and  therefore prudence
 suggests examination of  a second  price.   As  shown  previously  in  Section  7  1
 the highest average annual current  dollar price  for  refined copper was  225  3
 cents per kilogram, achieved in 1980.  (The  year 1980  was marked by an
 industry  strike  and reduced  production.)  Therefore, 220*/kg  is used to
 represent a  price that based on the  results  of past years, appears optimistic.
 An alternative  "pessimistic" price  is not presented because even the baseline
 results are  highly likely to be damaging and thus the addition of control
costs would merely reinforce an obvious conclusion.  A ready example of the
impact of a price significantly below 187*/kg was provided in 1982 when the
average price was about 163*/kg and large  segments  of the industry closed
for sustained periods.
                                     7-27

-------
        The analysis  of  the  results  for  the maximum percentage price increase
 figures is similar to the analysis  discussed above for the cost increase
 figures.   Once  again, for Alternative II only Kennecott-McGill experiences a
 price  increase.   The price  increase is  12.1 percent based on a 100 percent
 capacity  utilization rate and a price of 187^/kg.  For Alternative III the
 maximum price increases range from 0.1  to 5.0 percent.  For Alternative IV
 the  price increases  are lower, and range from 0 to 1.0 percent.  For Alterna-
 tive III+IV the price increases range from 0.1 to 6.0 percent, with only two
 smelters  above  2.2 percent.  There is some variation in the price increases
 among  the smelters.  The significance of the variation in the maximum percentage
 price  increases among the smelters is that those smelters with higher price
 increases would probably be constrained in the marketplace by the lower price
 increase  smelters.   As a result, some of the smelters could quite possibly
 have to absorb a  part of the control costs.  As mentioned above, two additional
 constraining influences are foreign competition and substitutes.

 7.2.5   Profit Impacts
        Apart from the calculation of maximum percentage price increase,
 additional insight into the economic impact of the arsenic NESHAP can be
 gained  by making  the opposite assumption from maximum percent price increase,
 that is,  zero percent price increase,  or complete cost absorption.  The
 assumption of complete control  cost absorption provides a measure of the
 reduction  in profits if the control costs are absorbed completely.
        Assuming control  costs are absorbed, the critical  element in an
 analysis of profit impacts is the profit margin.   The larger a firm's profit
margin, the greater  is the firm's ability to absorb control  costs and earn an
 acceptable rate of return on investment (ROI), and thus continue operation.
 The profit margin is  simply the difference between price  and cost.  As
mentioned in an earlier  section,  the central  issue becomes the choice of an
appropriate profit center and its corresponding price and cost.   The process-
 ing of virgin ore into refined  copper  involves four distinct steps:   mining,
milling, smelting, and refining.   Although the four steps are often joined to
form an integrated business unit,  they are not inextricably  bound together in
an economic sense.  For  example,  it is not uncommon for mines to have their
concentrate toll smelted and refined.   The difficulty that this  variability
                                     7-28

-------
  presents In terms of an assessment of the Impact of the arsenic standard Is
  1n the method of assigning the costs.
         This report presents an analysis of profit impacts using two methods.
  The first method assumes copper producers are fully integrated and all  have
  the same costs and thus earn a uniform profit margin.   The objective of this
  method is to permit a ready, though simplified,  examination of profit impacts
  With the first method as a foundation, the second method  introduces more
  smelter specific variables into the analysis  in  an effort to  focus more
  sharply on  the complex  organizational  structure  of the  industry.

         7.2.5.1  Method  One.   As mentioned  above,  the critical  element in an
  examination  of profit impact  is the  profit margin.  Therefore  an examination
  of  profit margins  for members of the  industry  is  necessary, and accordingly
  is  presented below.  Table 7-8 shows  the revenues  and operating profit
  (before tax) for each of the 7 producers for the 5-year period from 1977
  to  1981.  Table 7-8 also shows the percentage profit margin, which is operat-
  ing profit divided by revenues.  The revenue and operating profit figures are
  for the business segment within the company that includes copper.   The use of
 business segment information provides a closer representation of the results
 for copper than would the use of the consolidated results  for the  company
 The reason for this is that for several of the firms copper represents a
 relatively small  share of the total  company results. Although the  business
 segment information is a better representation  of the results  for copper than
 the  total  company results,  the  business segments  contain other products  in
 addition to  copper.   Therefore  conclusions  must be drawn accordingly   The
 table  shows  that  there is considerable variation  in results, both within  a
 company  from  one year  to  the  next, as  well  as from  one company  to the  next
 The  5-year average  ranges from a loss  of 3.6 percent to  a  high  of 13.8
 percent.
       Table 7-9 shows the maximum percentage reduction  in the  profit margin
 for each of the 14 smelters.  This table assumes each smelter is viewed as
 part of a fully integrated operation.  Two profit levels are shown  and two
 capacity utilization rates (100 percent and 80 percent).   The first profit
 level is based on a refined copper price of l87*/kg and  a 10 percent
profit margin, which yields  a profit of l8.7*/kg.   The second profit level
                                     7-29

-------
Table 7-8.
               BUSINESS  SEGMENT  RETURN  ON  SALES FOR COPPER COMPANIES9
                                     ($  103)

Revenues

Year
1977
1978
1979
1980
1981
Operating 1977
Profitd 1978
1979
1980
1981
Profit/
Revenues
(percent)



1977
1978
1979
1980
1981
Average
ASARCO
733,293
849,002
1,339,917
1,440,220
1,153,022
65,919
112,474
225,763
145,286
68,364
9.0
13.2
16.8
10.1
5.9
11.0
Cities
Service
184,000
241,500
276,300
224,100
MA
(38,600)
(23,900)
25,400
16,300
NA
(21.0)
(9.9)
9.2
7.3
NA
(3.6)
Copper
Range"
NA
64 , 600
89 , 300
83 , 900
NA
NA
(6,600)
10,000
1,800
NA
NA
(10.2)
11.2
2.1
NA
1.0
Inspiration
95,676
101,251
136,849
178,004
NA
(9,994)
(6,235)
9,889
(6,563)
NA
(10.4)
(6.2)
7.2
(3.7)
NA
(3.3)
Kennecott
NA
683 , 000
1 , 091 , 400
987 , 400
NA
NA
(100)
164,000
131,400
NA
NA
0
15.0
13.3
NA
9.4
Magmac
NA
274,137
381,512
287,581
328,842
NA
13,601
67,252
11,522
(15,658)
NA
5.0
17.6
4.0
(4.8)
5 5e
Phelps
Dodge
453 184
446,970
618,188
714,591
706,404
52,831
63,738
159,428
95,439
27,618
11.7
14.3
25.8
13.4
3.9
13 .8f
'Business segments  contain other products in addition to copper.
"The Louisiana Land and Exploration Company.
cProfit  is net income after tax in this case.
^Before  interest and tax.
ewould yield 7.9 percent if adjusted to before tax with an effective tax rate  of
 30 percent.
^Imputed
                  intersegment sales for 1977  to 1981 would yield average return
                                        7-30

-------
Table 7-9.
                                  DECREASE FOR ARSENIC CONTROLS FOR LOW-ARSENIC  PRIMARY COPPER SMELTERS
Smel ter
ASARCO-Hayden
ASARCO-E1 Paso
Kennecott-Garfield
Phelps Dodge-Hidalgo
Phelps Oodge-Morenci
Inspiration
Magma
Phelps Dodge-Ajo
Phelps Dodge-Douglas
Copper Range-White Pine
Kennecott-Hayden
Kennecott-Hurley
Cities Service-Copperhill
Kennecott-McGill 121
• 	 	 • — • 	 	 	 „


0
0
0
0
0
0
0
0
0
0
0
0
0
.4C
"TfiOT
_18-7*Ag.
TlnafJi.. 	 *~
Alternative
III IV I1I+IV
1-1 0 1.1
1.6 1.1 2.7
2.7 1.1 3.7
5-9 1.1 7.0
5.3 1.6 7.0
11.2 1.1 12.3
11-8 1.6 13.4
12.8 2.1 15.0
13.4 2.1 15.5
13.4 2.7 16.0
15.0 2.1 17.1
19-8 2.1 21.9
50.3 10.2 60.4
32.1 3.2 35.3
• 	 , 	 _


—
Profit Marqina

	 n_
0
0
0
0
0
0
0
0
0
0
0
0
0
151. 9C

BOX Caf
AHerna
HI IV
•- 	 _
1.1 0
2.1 1.1
3.2 1.6
7.0 l.l
6.4 1.6
13.9 1.1
14.4 2.1
12.8 2.7
16.6 3.2
16.6 3.2
18.7 2.7
24.6 2.7
62.6 12.8
40.1 3.7
	 — 	 . — . — ,_
>acity
itlve
III+IV
1.1
3.2
4.8
8.0
8.0
15.0
16.6
15.5
19.8
19.8
21.4
27.3
75.4
43.8

	 . 	 . — ___

51 .7 tf/ko.Prom
"100* Capacity— 	
Alternative
II
0
0
0
0
0
0
0
0
0
0
0
0
0
43.9
	 LLL IV II1 + IV
0.4 0 0.4
0.6 0.4 1.0
1.0 0.4 1.4
2.1 0.4 2.5
1-9 0.6 2.5
4.1 0.4 4.5
4.3 0.6 4.9
4.6 0.8 5.4
4.8 0.8 5.6
4-8 1.0 5.8
5.4 0.8 6.2
7.2 0.8 8.0
18.2 3.7 21.9
11-6 1.2 12.8
11
0
0
0
0
0
0
0
0
0
0
0
0
0
54.9

	 	 	 _- 	
M^rnf r\D
' — ~ 	 • 	
8Ui Capacity
III IV III+IV
0.4 0
0.8 0.4
1.2 0.6
2.5 0.4
2.3 0.6
5.0 0.4
5.2 0.8
4.6 1.0
6.0 1.2
6.0 1.2
6.8 1.0
8.9 1.0
22.6 4.6
14.5 1.4
0.4
1.2
1.8
2.9
2.9
5.4
6.0
5.6
7.2
7.2
7.8
9.9
27.2
15.9

                                           .„„„.

-------
is based on an increased price of refined  copper  to  a  level  of  220^/kg.
The second profit margin is based on  the original  18.7£/kg.  but adds  the
increase in price as extra profit while process costs  are  held  constant.   The
second profit margin is 51.7£/kg.  Three considerations  suggest the use of
the second profit margin.  The first  consideration is  the  desirability of
presenting sensitivity analysis in general.   The  second  consideration is that
a profit margin of 51.7£/kg. based on a price of  220^/kg.  is a  margin of
23.5 percent, which though clearly high, has  been achieved within  recent
years by a member of the industry. Finally,  because the margin is high, it
in effect can be viewed as an upper limit, and thus  any  smelter that  has a
substantial profit impact in spite of such a  favorable profit margin  is in a
very vulnerable position at a lower,  more  likely, profit margin.
       The same cases discussed earlier are  still  applicable, the  results  are
for Alternatives II, III, IV, and III+IV.   At the first  profit  margin (18.7rf/k
the results show a maximum profit reduction  of greater than  20  percent for
3 of the 14 smelters at the 100 percent capacity  utilization rate  for Alterna-
tive III+IV.  At the 80 percent capacity utilization rate  4  smelters  exceed
20 percent.  For 2 of the above 4 smelters the profit  reduction is greater
than 40 percent at the 80 percent capacity utilization rate, and greater than
50 percent for 1 of the 2 smelters.  Profit  reductions of  greater  than 40
percent would seriously call into question the continued viability of these  2
smelters.  At the second, higher, profit margin (51.7l/kg.)  the profit
impacts are lessened substantially.  Two smelters experience profit reductions
of greater than 20 percent.

       7.2.5.2  Method Two.  Method two uses  method  one  as a starting point
and then refines it by relaxing the assumption of method one of an inte-
grated producer and a uniform profit  margin  for all  producers.   Table 7-10
provides a means to identify those smelters  that  are most  likely to face the
greatest impact.
       Table 7-10 starts by showing the smelters  ranked  according  to  the cost
increase described earlier (Alternative III+IV).   The  size of the  cost
increase and the rank provides one indication of  the potential  impact of
controls.  A caveat that should be mentioned  concerning  this indicator is
that it does not take into consideration baseline costs.  Examination of
baseline costs would be a useful  supplement  here.  The 2 smelters  with the
highest cost increases are likely to  experience serious  difficulty due to  the
                                     7-32

-------
                                             Table 7-10.   REVIEW OF SMELTERS
 I
CO
CO

Smelter
ASARCO-Hayden
ASARCO-E1 Paso
Kennecott-Garfield
Phelps Dodge-Hidalgo
Phelps Dodge-Mo renci
Inspiration
Magma
Phelps Dodge-Ajo
Phelps Dodge-Douglas

Copper Range-White Pine
Kennecott-Hayden
Kennecott-Hurley
Cities Service-Copperhill
Kennecott-McGill

Cost
rank
1
2
3
4
5
6
7
8
9

10
11
12
13
14

Major
Increase Capital
- (tf/kg) Commitment
0.2 1982
0.5
0.7 1978
1.3 1976
1.3 1982
2.3 1980
2.5
2.8
2.9

3.0
3.2
4.1 1982
11.3
6.6

Maj or
Integrated
Mine
-
-
Yes
-
Yes
-
Yes
-
No

-
Yes
Yes
-
No

Closely
Associated
Refinery
No
No
Yes
No
No
Yes
Yes
No
No

Yes
Yes
Yes
No
No
Viability
Estimates
By Others
(Reference #)
Viable-37
Viable-37
Viable-37
Viable-37
Viable-37
Doubtful -38, 37
Viable-37
Doubtful-38,37,39
Doubtful -38, 16
Closure-37,39
Viable-37
Viable-37
Doubtful -37
Viable-37
Doubtful-38,37,16

-------
costs.  A second indicator that is presented to provide additional insight
into a  firm's possible reaction to control costs is a review of any major
capital commitments to the smelter that a firm has made recently.
        Most of the firms with the lower control cost increases have also
recently made major capital commitments to their smelters which in turn
suggests a stronger commitment than a firm that has postponed capital  expen-
ditures for a smelter.  The third indication is provided by a review of
whether or not the smelter has a major integrated mine that supplies much or
all of  its concentrates.  The fourth indication is provided by a review of
whether or not the smelter is closely associated with a refinery.  Finally,
the estimates of others who have analyzed the smelters are presented.   The
estimates are from four sources.
        The above indications are useful  but should be supplemented by  two
additional pieces of information.  This information is rioted below even
though  its impact has not been thoroughly analyzed at this time:

        1)  Newmont Mining Corp. recently purchased Cities Service Co.'s
copper mine operations in Arizona (Pinto Valley).   The concentrates from that
mine currently go to the Inspiration smelter under a contract that runs until
the end of 1984.   These concentrates represent a significant share of  Inspi-
ration's production.   The dollar value is approximately 15 percent of  Inspi-
ration's sales.   After the contract expires these  concentrates will then be
processed at Newmont1s Magma smelter.  This will  represent a significant loss
for Inspiration  and a significant gain for Magma.
       2)  Anaconda recently announced that it will  close its Butte mine and
no longer ship concentrates to Japan.  At this time, it is not clear if the
Japanese will  seek  replacement concentrates, and if so,  what domestic  mines
are the most likely candidates to supply the concentrates.

7.2.6  Capital Availability

       The principal  determinant of the  financial  viability of a smelter is
profitability.   However,  the amount of capital  needed to  purchase  control
equipment is one  of the components that  enters into an  evaluation  of profit-
ability.  Most firms  prefer to finance pollution control  equipment with debt,
both because debt is  less expensive than equity in  general,  and additionally
because debt incurred to  purchase pollution control  equipment is often  tax
                                     7-34

-------
   exempt.  Assuming control equipment is  financed with debt, as the capital

   cost of the  control  equipment  increases, the level of debt increases.  An

   increased debt level means the fixed costs required to service the debt

   increase and therefore the level  of risk increases.  As a result, a dis-
         Table 7-11  slums the capital expenditures that will  be necessary

   The capital expenditures were expla,ned in  detai,  in an earlier  e    .'

   The basel,ne capital expenditures are presented, as well  as the capita,

   expenditures for Alternatives II, III  and  IV   Thr.» f
   smelter an* •   u.    ,                          e flras own more than 1

   I! frm!      ?"   C9SeS the ^  CaPUS'  C°StS «  "»«.  "'hough
   t   f,rms can make capita, budgeting decisions  on  an individual  smelter
             .  Additional!,, 5 of the 7 c.panies are owned   .,

   ea;: ; ;::;; tdr:ee> by 'isnmcMti' '^ — ™L and
     qu He Inkely to have access to  the necessary capital.  The remaining
  companies are ASARCO and Phel ps Dodge.                     e remaining

                                 7"U  """ *"« "»reMt

dv

 e
                                                  •"====''
                of Phelps Dodge the  increases are 0,  5, and 1  percent
                                                    terna      I,  In
                         SUM6St * ^ "
                                 tosether Kf
««
                     for Phelps Dodge  was lowered to Baa2 from i
                                                             its
                                 7-35

-------
        Table 7-11.  CAPITAL COSTS OF ARSENIC CONTROLS
                     FOR PRIMARY COPPER SMELTERS
                             ($103)

ASARCO




Cities Service
Copper Range
Inspiration
Kennecott




Newiriont
Phelps Dodge






El Paso
Hayden
Tacoma

Debt Increase3
Copper hill
White Pine
Miami
Garfield
Hayden
Hurley
Me Gill

Magma
Ajo
Douglas
Hidalgo
Morenci

Debt Increase3
Baseline
46
75,606

75,652
24%
0
0
0
0
0
54,044
0
54,044
0
51,067
0
0
95,294
146,361
24%
Al
II
0
0
3,469
3,46$
1%
0
. 0
0
0
0
0
10,530
10,530
0
0
0
0
0
0
0%
ternative
111
1,375
1,702
3,469
6,546
2%
4,434
4,434
9,825
5,197
6,731
8,760
8,760
29,448
13,050
6,731
9,825
6,731
8,526
31,813
5%

IV
370
0
3,469
3,839
1%
893
893
922
1,786
894
952
893
4,525
1,786
894
1,787
894
1,786
5,361
1%
aPercent increase in  average long  term  debt level  for  the  past  3
 years (1981 to 1979)  if controls  are  added as  debt.
                                7-36

-------
   previous  rating  in  1980 and 1981 of A.  Although Baa2 is still a relatively
   strong  rating, the  fact that it was lowered from 1981 to 1982 suggests that
   substantial  increases in the amount of debt held by the company may present
   some  difficulties.

   7.3     SOCIO-ECONOMIC IMPACT ASSESSMENT
   7.3.1   Executive Order 12291
          The" purpose of Section  7.3.1 is to  address  those  tests  of macro-
   economic impact as presented in Executive  Order  12291, and, more generally
   rNESlVnVther  Sl9n1f1Cant maCr°™-c  "*«* that may result  from'
     i       ;   X6CUt1Ve °rder ^  Stl>lateS  " "^  -les" *•" ^at are
  projected  to have any of the following  impacts:

        •    An annual  effect on  the economy of $100 million or more
        •    A major  increase in costs or prices for consumers; indi-
              vidual industries; Federal, State,  or local  government
             agencies; or geographic regions.
        •   Significant adverse effects on  competition,  employment,  invest-
             ment,  productivity,  innovation, or  on the  ability  of  U.S.-based
             enterprises to  compete  with foreign-based  enterprises in  domestic
             or export markets.
each of-     ,03^.  The annual ized control  costs for
each of the 3 alternatives Is we,,  below the $100 million *,<:„ 1s the
figure used to identify a major rule.  The annualized  contro,  costs for
merna ,ves „.  !„, Iv. and Imiv .„ ^ ^
W.3 million,  and $31.1 million,  respectively.
       7'3'1-2 Milonal  Effects. EmP1ovm.nt  ann r^r.>.-tir     ^

 I"" r r I"r^^^^^^^            ««  in particu-
U r  7 smelters are  ,ocated  ,„ Arizona.   As a result, econ^ic  impacts
      be concentrated  in  that geographical area.  A copper smelte  typi
      s about  50                                                    w
employs about 500 people.
       The domestic  copper producers  compete  among  themselves, as well as
aga,nst foreign copper  producers  and  substitutes.   Any substantial increase
                                    7-37

-------
in costs will put pressure on the competitive position of some domestic
smelters with respect to other domestic smelters, and also with respect to
foreign copper producers, and substitutes.

7.3.2  Regulatory Flexibility
       The Regulatory Flexibility Act of 1980 (RFA) requires that differen-
tial impacts of Federal  regulations upon small  business be identified and
analyzed.  The RFA stipulates that an analysis  is required if a substantial
number of small  businesses will  experience significant impacts.  Both measures
must be met, substantial numbers of small  businesses and significant impacts,
to require an analysis.   If either measure is not met then no analysis is
required.  The EPA definition of a substantial  number of small  businesses in
an industry is 20 percent.  The  EPA definition  of significant impact
involves three tests, as follows:  one, prices  for small  entities rise 5
percent or more, assuming costs  are not passed  onto consumers;  or two, annual-
ized investment costs for pollution control  are greater than 20 percent of
total capital spending;  or three, costs as a percent of sales for small
entities are 10 percent  greater  than costs as a percent of sales for large
entities.
       The Small Business Administration (SBA)  definition of a  small  business
for Standard Industrial  Classification (SIC) Code 3331, Primary smelting and
refining of copper is 1,000 employees.  Table 7-12 shows  recent employment
levels for each  of the 7 companies that own  primary copper smelters.
All 7 have more  than 1,000 employees.  Therefore, none of the 7 companies
meets the SBA definition of a small  business and thus no  regulatory flexi-
bility analysis  is required.
                                     7-38

-------
               Table 7-12.   NUMBER OF EMPLOYEES AT COMPANIES

                            THAT OWN  PRIMARY  COPPER SMELTERS
 _ ComPany _ _ _ Employees  _ $ourcea



 ASARCO,  Inc.                          12,700           1980 SEC 10-K p. A7

 Cities  Service Co.                    18,900           1980 SEC 10-K p. 6

 Copper  Range Co.b                     3>049           1980 SEC 10-K p. 22


                                                      198° SEC 10-K  - 2
Kennecott Corp.c                     3MOO

Newmont Mining Corp.                 12,400           1980 SEC 10-K p. 9

Phelps Dodge Corp.       _ 15.220 _ 1980 SEC 10-K p. 1



aSEC 10-K is Securities and Exchange Commission, Form 10-K.
             r°' 1S a ^oll*-°wned subsidiary of the Louisiana Land and
             Company.  Figures are for Louisiana Land and Exploration.

cPrior to merger with Sohio on March 12,  1981.
                                   7-39

-------
7.4    References
  1.    Review of New Source  Performance  Standards  for Primary  Copper  Smelters
       — Background Information  Document,  Preliminary Draft.   U.S. Environ-
       mental  Protection  Agency.   Research  Triangle  Park,  North Carolina.
       Publication  No.  EPA-February  1983.   p.  3-2.

  2.    ASARCO,  Inc., Form 10-K.   December 31,  1980.  p. A2.

  3.    Cities Service Co., Annual  Report 1980.  p.  41.

  4.    The Louisiana Land Exploration  Co.,  Form 10-K.   December 31, 1980.  p.
       16.

  5.    Inspiration  Consolidated Copper Company, Annual  Report  1980. p.  2.

  6.    Kennecott Corp., Form 10-K.   December 31, 1980.   p.  4.

  7.    Newmont  Mining Corp., Form  10-K.  December  31,  1980.  p.  3.

  8.    Phelps Dodge Corp., Form 10-K.  December 31,  1980.   p.  2, 4.

  9.    Butterman, W.C.  U.S.  Bureau  of Mines.   Preprint from the 1981 Bureau
       of Mines Minerals  Yearbook.   Copper,  p.  3.

 10.    Butterman, W.C.  U.S.  Bureau  of Mines.   Mineral  Industry  Surveys.
       Copper Production  in  December 1982.  p.  2.

 11.    Schroeder, H.  J. and  James  A. Jolly.  U.S.  Bureau of Mines.  Preprint
       from  Bulletin  671.  Copper  -  A  Chapter  from Mineral  Facts and Problems,
       1980  Edition,  p.  14-16.

 12.    Annual Data  1982.  Copper Supply  and Consumption.  Copper Development
       Association  Inc.   New York, New York.   p. 6,  14.

 13.    Reference  11,  p. 5.

 14.    Reference  9,  p. 1.

 15.    Reference  9,  p. 5.

 16.    Arthur D.  Little,  Inc.  Economic  Impact  of  Environmental Regulations
       on  the United  States  Copper Industry.   U.S. EPA.  January 1978.  p.
       V-8.

 17.    Reference  9,  p. 24-29.
                                    7-40

-------
18.   Reference 12,  p.  14.

19.   Reference 11,  p.  14.

20.   Reference 12,  p.  18.

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

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

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

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

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

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

27.   Rosenkranz,  R.D., R.L. Davidoff, and J.F.  Lemons, Jr.,  Copper Avail-
      ability-Domestic:  A Minerals Availability System Appraisal.  U.S.
      Bureau of Mines.   1979.  p.  13.

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

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

30.   Affidavit of Robert J. Lesemann, Commodities Research Unit/CRI and
      former editor-in-chief of Metals Week,  to  the  Federal Trade Commission.
      September 27,  1979.   FTC  Docket Number  9089.

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

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

33.   Rieber, Michael.   Smelter Emission  Controls:   The Impact  on Mining
      and The Market For Acid.   University of Arizona, Tucson,  Arizona.
      March, 1982.  p.  5-10.

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


                                    7-41

-------
35.   Everest Consulting Associates, Inc., and CRU Consultants, Inc.  The
      International Competitiveness of the U.S. Nonferrous Smelting Industry
      and the Clean Air Act.  Princeton, NJ.  April 1982.  p. 9-9.
36.   Reference 32.
37.   Reference 33, p. 1-11.
38.   Everest Consulting Associates, Inc.  The International  Competitiveness
      of the U.S. Non-Ferrous Smelting Industry and the Clean Air Act.
      Princeton, N.J.  April 1982.   p. 3-17.
39.   Phelps Dodge Corp. 1981 Annual Report,  p. 8.
40.   Moody1s Industrial Manual  1982 Vol. I, p. 58, Vol. II,  p. 4236.
                                    7-42

-------
                   APPENDIX A





EVOLUTION OF THE BACKGROUND INFORMATION DOCUMENT
                   A-l

-------
            EVOLUTION OF THE BACKGROUND INFORMATION DOCUMENT
      Date

 July  13-14, 1976


 December 10-16, 1976


 April 18-26, 1977


 June  20-30, 1977


 January 17-27, 1978


 May 1-5, 1978


 May 10-12,  1978


 June  12-16, 1978


 July  11-12, 1978



July  24-27, 1978


September 12-25, 1978
October 30 -
November 15, 1978
                    Activity

 Emission source testing at  Phelps  Dodge
 Copper Smelter, Ajo, Arizona.

 Emission source testing at  Kennecott Copper
 Smelter, Hayden, Arizona.

 Emission source testing at  Anaconda Copper Smelter,
 Anaconda, Montana.

 Emission source testing at  ASARCO  Copper Smelter,
 El Paso, Texas.

 Emission source testing at  ASARCO  Copper Smelter,
 El Paso, Texas.

 Emission source testing at  Phelps  Dodge Copper
 Smelter, Douglas, Arizona.

 Emission source testing at  Phelps  Dodge Copper
 Smelter, Ajo, Arizona.

 Emission source testing at  Phelps  Dodge Copper
 Smelter, Ajo, Arizona.

 NAPCTAC Meeting in Raleigh, North  Carolina to
 discuss issues related to development of arsenic
 emission standards for primary copper smelters.

 Emission source testing at  Phelps  Dodge Copper
 Smelter, Hidalgo, Arizona.

 Emission source testing at  ASARCO  Copper Smelter,
Tacoma, Washington.
Emission source testing at Kennecott Copper Smelter
Garfield, Utah.
                                A-2

-------
       Date

  May 8-15,  1979


  July 23-24,  1979


  September  10-16, 1979


  September  18-22, 1979


  December 7-13, 1979


 March 11-13, 1980



 April 14-23, 1980


 June 5,  1980


 June 24-26, 1980


 March 17, 1981




 January  12,  1983
January 14-22, 1983
April 27, 19U3
                      Activity

  Emission source testing  at ASARCO Copper Smelter,
  Tacoma,  Washington.

  Emission source testing  at Phelps Dodge  Copper
  Smelter, Ajo,  Arizona.

  Emission source testing  at Phelps Dodge  Copper
  Smelter, Morenci, Arizona.

  Emission source testing  at  Phelps Dodge  Copper
  Smelter,  Douglas, Arizona.

  Emission  source  testing  at  Kennecott Copper  Smelter
  McGill,  Nevada.

  Plant visit to  Hibi Kyodo Copper  Smelter,
  Tamano, Japan,  to collect  information on the
  fugitive  emissions control system.

  Emission  source testing at Magma  Copper Smelter,
  San Manuel, Arizona.

  EPA listing of  inorganic  arsenic as a hazardous
 pollutant under Section 112 of the Clean Air Act.

 Plant visit and emission  source testing at ASARCO
 Copper Smelter, Tacoma, Washington.

 NAPCTAC meeting in Raleigh, North Carolina to
 discuss regulatory alternatives for limiting
 arsenic emissions from low-arsenic throughput
 primary copper  smelters.

 Judicial  Opinion and  Order,  filed with  United
 States  District Court, Southern District  of New
 York,  pertinent to action brought by State of
 New  York  against EPA  Administrator (New York
 v. Gorsuch,	F. Supp. 	 (S.D.N.Y.  1983)).
 Order  for EPA to propose  emission  standards  for
 inorganic  arsenic  within  180  days  of Order.

 Emission  source  testing at ASARCO  Copper  Smelter
 Tacoma, Washington.

 NAPCTAC meeting  in Raleigh, North  Carolina, to
discuss general approach and  EPA staff recom-
mendations for setting standards for inorganic
arsenic emissions from primary copper smelters.
                                A-3

-------
                 APPENDIX B
INDEX TO ENVIRONMENTAL IMPACT CONSIDERATIONS
                 B-l

-------
              INDEX TO ENVIRONMENTAL IMPACT CONSIDERATIONS

     This appendix consists of a reference system which is cross  indexed
with the October 21, 1974, Federal  Register (39 FR 37419) containing  the
Agency guidelines for the preparation of Environmental Impact
Statements.  This index can be used to identify sections of the document
which contain data and information germane to any portion of the  Federal
Register guidelines.
                                 B-2

-------
                INDEX  TO  ENVIRONMENTAL  IMPACT  CONSIDERATIONS
     Background and Description

        Summary of the Regulatory
          Alternatives

        Statutory  Authority
        Industry Affected
       Sources Affected
       Availability of Control
         Technology


2.   Regulatory Alternatives

       Regulatory Alternative I
         No  Action (Baseline)

         Environmental  Impacts
        Costs



      Regulatory Alternative II —

        Environmental  Impacts




        Costs
                                       Location Within the Background
                                         Information Document (BID)
  The regulatory alternatives are
  summarized in Section 1.2.

  Statutory authority is cited in
  Section 1.1.

  A description of the industry
  to be  affected is given in
  Section 7.1.

  Descriptions  of  the various
  sources to be affected  are
  given  in Section  2.0.

  Information on the  availability
  of control  technology is given
  in Section  3.0.
 Environmental effects of Regulatory
 Alternative I are considered in
 Section 5.0.

 Costs associated with Regulatory
 Alternative I are considered in
 Section 6.0.
 Environmental  effects  associated
 with Regulatory Alternative  II
 emission control systems  are
 considered in  Section  5.0.

 The cost impact of Regulatory
Alternative II emission control
systems is considered  in
Section 6.0.
                               B-3

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

       Environmental Impacts




       Costs



     Regulatory Alternative IV

       Environmental Impacts




       Costs



     Regulatory Alternative V

       Environmental Impacts
        Costs
The environmental effects associated
with Regulatory Alternative III
emission control systems are
considered in Section 5.0.

The cost impact of Regulatory
Alternative III emission control
systems is considered in Section 6.0.
The environmental effects associated
with Regulatory Alternative  IV emission
control systems are considered in Sectio
5.0.

The cost impact of Regulatory
Alternative IV emission control
systems is considered  in Section 6.0.
The  implementation  of  this
alternative would require the
elimination of  inorganic arsenic
emissions at  low-arsenic throughput
primary  copper  smelters.  Inorganic
arsenic  emissions would  be  zero.

This  alternative  could not  be
implemented without closing all
plants  in the source category.   This
was  not  considered  reasonable  and hence
costs were  not  evaluated.
                                  B-4

-------
      APPENDIX  C





SUMMARY OF TEST DATA
      C-l

-------
                         SUMMARY OF TEST DATA

     An emission test program was undertaken by EPA to evaluate the
performance of alternative control  techniques available for the control
of process and fugitive arsenic emissions from process facilities
including roasters, smelting furnaces, and converters at primary
copper smelters.  This appendix presents a brief description of the
process facilities and control equipment tested, and a summary of the
results obtained.
     Arsenic emission measurements were conducted at eight domestic
smelters.  Particulate emission measurements were also conducted at
some of these smelters.  A listing of the process facilities and air
pollution control equipment tested and the emission measurements
conducted is presented in Table C-l.  All arsenic measurements were
performed using EPA Method 108, the recommended EPA method for the
determination of arsenic from stationary sources.  Measurements of
total  particulate were performed in accordance with EPA Method 5.
     In addition to the arsenic and particulate emission measurements,
visible emission observations were recorded  at one of  the  above eight
domestic  smelters, ASARCO-Tacoma,  and one other smelter located  in
Japan, the Tamano  smelter.
     A brief description  of  each smelter, as well  as  the  process  and
control device  tested, is  presented in  Sections C.I  through  C.9.
Section  C.10  contains  the  data  tables from  the  emissions  measurement
and  opacity observation tests.
C.I  ASARCO-TACOMA
     The  ASARCO smelter at Tacoma, Washington,  is  a  custom smelter
which  processes copper ore concentrates,  precipitates, and smelter
                                 C-2

-------
  by-products from numerous domestic and foreign sources.  The smelter
  produces about 320 Mg (352 tons) of anode copper per day at full smelt
  and houses the only arsenic production facility in the United States.
  Copper smelter facilities include 10 roasters (6 Herreschoff multi-hearth
  roasters and 4 CAW Roasters),  2 reverberatory smelting furnaces,
  4 Pierce-Smith converters,  3 anode furnaces (1 hearth type furnace and
  3 tilting type furnaces),*  and an electrolytic refinery.**  Arsenic
  production  facilities  consist  of three Godfrey roasters,  arsenic
  trioxide settling  chambers  or  kitchens,  storage facilities, and a
  metallic arsenic plant.
      The roaster charge  which  consists  of  a  blend  of  concentrates,
  precipitates,  lead  speiss,  flue  dust,  and  fluxing  materials typically
  contains  3  to  4 percent  arsenic  and  7  to  10  percent moisture.   At full
  smelt, four to five roasters are  used.  Charging is continuous.   The
  calcine  produced, about  45.4 Mg  (50  tons)  per  hour, is  intermittently
  discharged from hoppers  located  below  the  roasters into Tarry cars  for
  transport to one of two  reverberatory  furnaces.  Typically,  two  5.9 Mg
  (6.5 ton) cars are charged every  15 minutes.   Fugitive emissions  which
 could escape are confined and captured by  close-fitting exhaust  hoods
 located at the discharge point and are vented  into the main  roaster
 flue by two 73.6 Nm3/min (2,600 scfm) fans.  In addition, general
 ventilation is applied at the south end of the larry car tunnel  to
 control  fugitive emissions-resulting from the re-entrainment of settled
 dust  within  the tunnel.
      Offgases  from  the roasters,  which  average about  3,570 m3/min
 (126,000  acfra)  at 260°C  (500°F),  are combined with  the exhaust gases
 from  the  ancillary  fugitive  emission control  systems  [850  Nm3/min
 (30,000 scfm)]  and  reduced in temperature  to  less  than 120°C (250°F).
 The total  gases  are  then  treated  in  a baghouse  for  particulate removal.
 The baghouse consists of  17  compartments containing 120 bags each.
 The bags  are made of acrylic and  measure 20.3 cm  (8 inches)  in diameter
 and 7.6 meters  (25 feet)  in  length.   The total  baghouse filtering  area
-Operation of electrolytic refinery was discontinued  in January  1979,

                                C-3

-------
 is 9,950 in  (107,100 ft2).  The baghouse is designed to effectively
 treat 5,664 m /min (200,000 acfm)  of gas at an air-to-cloth ratio of
 about 1.9 to 1.0.   Bag cleaning is performed by mechanical shakers.
 The clean baghouse exhaust is vented through a flue to the smelter
 main stack.
      Although  the  smelter has two  reverberatory smelting furnaces,
 each with an approximate smelting  capacity  of 1,090 Mg (1,200 tons)
 per day,  the designated  Number 2 furnace is used almost exclusively.
 The furnace  measures  33.5 m (110 feet)  in length and 9.8 m (32 feet)
 in  width  and is  fired  by either oil  or  natural  gas.  Furnace charging
 is  accomplished  by discharging the larry cars through one of four
 Wagstaff  guns  located  along the furnace sidewalls.   Typically, it
 takes  less than  1  minute to discharge each  car.   At full  snelt,  two
 cars  are  discharged about every 15 minutes.   To minimize potential
 fugitive  emissions  during charging,  a manual  control  override is  used
 which  simultaneously opens  the furnace  Hue  control damper and reduces
 the  fuel  supply  to  the furnace fire  prior to  each  charge to prevent
 pressure  surges  in  the furnace.
     Matte is  tapped from the  furnace as  required  by converter operations
 through one  of four tapping  ports.    Only  one  tapping  port  is  used at a
 time.  The matte flows through  a cast copper  launder  to a  4.25 m3
 (150 ft ) cast steel ladle  for  transfer  to  the  converters.   At full
 smelt, about 45 ladles are  transferred  per day.   It takes  about 7 minutes
 to fill one  ladle.   Similarly,  slag  is  skimmed  through  one  of  two
 tapping ports as required  to maintain the proper slag  level  in the
 furnace and  transferred  to a 5-pot slag train for transit  to  the  slag
 dump.  Once  tapped, the  slag flows  through a  cast steel  launder and
 into a 2.8 m  (100 ft3)  cast steel  slag pot.  At full smelt,  about
 20 five-pot  slag trains  are dumped  per day.    Each train  takes  about
 15 minutes to fill.  Fugitive emissions generated during both  matte
and slag tapping operations are controlled by local ventilation techniques
Tapping ports,  launders,  and launder-to-ladle/pot transfer  points are
hooded and ventilated.   Ventilation requirements for the matte tapping
system total  about  700 m3/min  (25,000 acfm), while  the  ventilation
requirements  for the slag skimming  system total  about 600 m3/min
                                C-4

-------
   (21,000 acfm).  Captured emissions from both systems are currently
   controlled by an electrostatic prccipitator prior to venting from a
   stack to the atmosphere.
        Process gases from the reverberatory  furnace,  which average about
   1,415 Nm /min (50,000 scfm),  pass  through  a pair  of waste heat  boilers
   where the  gases  are  cooled  to about  400°C  (750°F).   The  exiting  gases
   then  pass  through  a  large rectangular  brick flue  where additional
   cooling  and  gas  stream  conditioning  is  provided by  air infiltration
   and water  and  sulfuric  acid sprays located  in the flue.   The resultant
   gas stream,  about  6,100  actual m3/m1n  (215j000 acfm) flt  ^ (
   Centers  the first of two electrostatic  precipitators  in series for
   particulate  removal.   The first precipitator is a tube or pipe design
   consisting of 18 sections with a total  collection area of about 6,619 m2
   '1.250 ft ).  Each section  contains  84 pipes measuring 30.5 cm (12 inches)
   -  dieter and 4.6 m (15 feet) in  length.   The  second unit is  a  plate
  type design,   it  consists of seven  parallel  chambers each with  four
  fields in series  and  has a total  collection  area of  7,710 m2 (82,992  ft2)
   e  exumg gases,  about 7,740 actual m3/nrln (270>000 acfm)  ^  ^     }'
  (230 F),  are  discharged  through a large  flue to the  smelter  main
  stack.
      Matte  from the reverberatory furnace is  transferred  to  one of
   our Pierce-Smith converters.  Three of  the  converters measure 4.0 „
  (13 ,eet) in diameter by 9.1  „ (30 feet) in  length, while.the fourth
 converter is 3.4 m  (n feet)  in diameter and 7.9  „ (26 feet) long.  In
 addition to copper matte, smelter reverts and cold  dope materials  are
 also processed.  Typically, only two converters  are on blow aTany on,
 tine.   A converter cycle normally takes  frm 10  to  12 hours   With

   u"thOff9as  now per  blow1"
 ,.-,-„„  „<.,„,,  ai)1J  uuntdins  rrom  J  to  4  percent  SO     en-,•<• +
                                   ww  -r  j^ci v»cnt  oUo •   D  I^TPr  rnnnav
 Produced  is  transferred to  one of  three anode furnaces  for  re"  £
 and casting.   The  slag s^med from the converters „ recycled to   he
 reverberatory  furnace.

     Offgases  from converter blowing operations are captured by
      cooled hoods and pass through a series of multiclone
      ig flue for coarse particulate removal prior to entering" th'e gas
deamng  circuits of either a  liquid S02 plant  or single-contact

                               C-5

-------
  sulfuric  acid  plant.   The  gas  cleaning  circuits  for both plants are
  similar and  consist  of  a water spray  chamber,  electrostatic  precipitator
  scrubbers,  and  mist  precipitators  in  series.   The  single-contact acid
  plant  has a  182 Mg/day  (200  TPD) capacity  at  5 percent  S0? and  is
  capable of  processing 652  Nm3/min  (23,000  scfm)  of  converter gas.  The
  liquid  S02 plant processes up  to 12,748 Nm3/min  (45,000 scfm) of the
  converter gases.  The plant  uses dimethylaniline (DMA)  to absorb the
  S02 in  the gas  stream and  uses steam  stripping from  regeneration.   The
  100 percent concentrated SO,, gas stream produced is  then liquified  by
  compression and the liquid S02 stored.
      Inlet and outlet emission measurements for  arsenic were conducted
  by EPA across the roaster baghouse and the arsenic baghouse  (metallic
  and kitchen) on September 12 through September 25,  1978.  Arsenic test
  results are  summarized in Tables C-3 through C-7.
      Testing for arsenic was also conducted on September 15,  16,
 and 18, 1978,  at the  outlet of the  reverberatory  furnace ESP.  These
 results are  shown  in  Table  C-8.
      Tests were  also  conducted  for  roaster calcine  discharge, matte
 tapping, slag  tapping,  and  converter slag  return.  Data  obtained for
 these  sources  are presented in  Tables  C-9  through C-12.
      Visible  emission  observations  were  made by using EPA Method 22
 and EPA Method 9 for  calcine  loading of  larry  cars,  matte tapping,
 slag  tapping, and converter slag  return  on  June 24-26,  1980.
 Tables  C-65 through C-73  present  the results  of these visible emissions
 observations.
 C.2 ASARCO-EL PASO
     The ASARCO  smelter at  El Paso  is  a  custom smelter that processes
 copper ore concentrates from  numerous  sources.  The  smelter produces
 about 315 Mg (350 tons) of  anode copper  per day.  Copper smelting
 facilities consist of ore handling and bedding facilities, four  multi-hearth
 roasters, one reverberatory smelting furnace, and three  Pierce-Smith
 converters.   In addition, this smelter also has separate  process
 facilities for zinc  and lead production.
     The smelter feed, consisting of a blend of concentrates,  precipitates,
lime and flue dust,  and typically containing more  than 0.2 percent
                                 C-6

-------
   arsenic,  is  charged  to  four  Herreshoff  multi-hearth roasters.   Each
   roaster has  seven  hearths and  is  capable  of  producing  approximately
   356 Mg  (392  tons)  of calcine per  day.   The  calcine is  taken by Tarry
   car to  the single  reverberatory furnace which  is  90 cm (35.7 in.)  long
   and 20 cm  (8.0 in.) wide and fired with either oil  or  natural  gas
   The furnace  is charged  by Wagstaff guns located along  the  furnace  side
  walls.  The matte  is tapped  from  one of five tap  holes  located  on  the
  north and  south sides of the furnace.  Slag is tapped  from  the  west
  side of the furnace and is disposed of in a slag  dump.
       Matte from the furnace is  transported to one of three  Pierce-Smith
  converters.  Each of the converters is 4 m in diameter and  9 m  long
  Normally,  two converters are  in the blowing  mode at one time    In   '
  addition to copper matte,  flue  dust and  cold dope  materials, converters
  also process  lead  matte  from  the  adjacent  lead  smelter  when available
  Blowing  time  generally  ranges from 10 to 12  hours.  Blister copper
  produced is further refined in  two anode  furnaces  and then  cast into
  anodes for  shipment.
      Offgases from  the reverberatory  furnace, which  average about
  1,700 Nm /min (60,000 scfm),  pass  through  a pair of  waste heat  boilers
 where the gases are cooled to about 400°C  (750°F)  and about  23  Mg
  (50,000 Ib) of steam is   produced.   The gases exit  the waste  heat
 boilers through two parallel  ducts and are then combined with the
 roaster offgases in a'main flue.  The combined gas  stream, consisting
 of about  5,000 Nm  /min (177,000  scfm)  at  200°C (400°F), then passes
 through a spray  chamber  where  it is cooled  to about  HO'C (230°F)
 prior  to  entering  an electrostatic  precipitator  for particulate  removal
 The  precipitator consists of seven  parallel chambers.  Each  chamber
 has  four  fields  in  series and  has  a total  field  volume of 535 m3
 (18,900 ft  ).  Gases exiting the precipitator  are discharged to  the
main stack  through  a large balloon  flue.
     Offgases  from  the converter blowing  operations average  6,000 Nm3/nin
 56,500 scfm).  They then pass through a settling chamber, two waste
heat boilers, and a spray chamber where the gases are cooled  from
315°C (600'F) to llO'C (230
-------
      The precipitator consists of four parallel chambers, each of
 which has four fields in series.   The exiting offgases then pass
 through a venturi  scrubber for additional  particulate removal, are
 humidified and cooled in a pair of packed-bed scrubbers, and are
 treated in a series of mist precipitators  where acid mist and any
 remaining particulates are removed prior to entering a double-contact
 acid plant for S02 removal.  The  acid plant has a normal production
 rate of 408 Mg (450 tons)  of acid per day.   Either 93 or 93 percent
 sulfuric acid  is  produced.   The acid  plant  tail  gas streams are discharged
 through  a 30.5 m  (100 ft)  stack.
      Emission  measurements  were conducted  by EPA during  June 26-30,
 1977.   Inlet and outlet  arsenic and mass measurements were made across
 the  roaster/reverberatory  electrostatic  precipitator.  Three inlet
 locations  and  one  outlet location were sampled.   The inlet locations
 included  a  large downtake  duct  off of the multi-hearth roasters,  and
 two  parallel ducts  downstream  of  the  reverberatory waste heat  boilers.
 The  outlet  location  consisted  of  the  balloon flue  downstream of the
 precipitator.   Three  arsenic and  two  total  particulate runs  were
 conducted.  The arsenic  and  particulate  test results are summarized in
 Tables C-13 through  C-22, and Table C-29.
     Arsenic emission  tests were  also conducted  across the
 double-contact  sulfuric  acid plant.   Three  inlet and outlet  measurements
 were made.  The sampling locations included  a duct  upstream  of  the
 spray chamber/ESP and  the acid  plant  tail gas stack.   The  results  of
 these tests are summarized  in Tables  C-23 and C-24,   Process conditions
 were carefully observed, and testing was conducted  only  when the
 subject process facilities and  control equipment were operating within
 normal operating limits.
     Fugitive emissions  from the  reverberatory furnace matte tapping
 operation at ASARCO-E1 Paso are captured by  hoods over the ladle,
 covered matte launders, and a hood at the matte tapping  holes.  Gases
 from these sources  are combined in a  common duct and  directed through
 a fan to a baghouse.  The baghouse discharges into  the roaster/
 reverberatory spray chamber/ESP control  system whicn  discharges from
the 250 m (828  ft)  main stack.
                                C-8

-------
       Fugitive gases that escape the converters during  the  blow  period,
  and roll-in/out operations and -cher fugitive gases  in  the converter
  building, are collected at the roof of the converter building.   Collected
  fugitive gases are drawn through four openings at the  roof into  ducts
  that combine into a main'duct leading to a baghouse, through  fans  on
  the clean side of the baghouse,  and then out the 250 m  (828 ft)  main
  stack.
       The fugitive gas  flow through the baghouse averages 14,100  Nm3/min
  (498,000 scfm).   The  converter building  fugitive baghouse consists of
  12 compartments.   Normally all  compartments are in operation except
  one compartment which  is  taken  off during  the cleaning  cycle  and
  another  compartment which  is  taken off  for maintenance  purposes for a
  fraction  of  the total  time.   Each  of  the  12 compartments contains
  334 Orion or Dacron bags.   Each compartment is  20  cm  (8 in.)  in diameter,
  6.7 m (22 ft) long, with a  cloth area of  1,644 m2  (17,700 ft2)  per
  compartment.  The total net cloth  area of  the baghouse  is about  19,700  m2
  (212,400 ft  ).  The baghouse was designed  to  effectively treat  15,282
  actual m /rain (540,000 acfm) at 54°C (130T)  using an air-to-cloth
  ratio of 3.0 to 1.0.  Mechanical  shakers (automatic)  are used  for
  cleaning.  Dust from the baghouse is removed  from the dust  chambers
 under the baghouse by  screw conveyors.
      Emission measurements across  the baghouse were conducted by  EPA
 during Janaury'17-27,  1978.   Three  arsenic and particulate measurements
 were made at  the  inlet  and  outlet  locations of the converter building's
 fugitive  baghouse.   The arsenic  test results are summarized in Tables C-25
 and C-26,  and the  particulate  test  results  are summarized in Tables C-27
 and C-28.  During  the same  period,  three  arsenic  and  particulate
 measurements  were  made  at  the  inlet  to  the  matte  tapping baghouse and
 the calcine discharge duct.  Due to  the physical  configuration  of the
 matte tapping  baghouse  system, outlet  tests  could not  be conducted
 Measurements were conducted  after the fan;  however, a  side stream of
 about 311 actual m /min (11,000 acfm) of the matte  tapping gases  were
 split at the fan and ducted to the reverberatory furnace  waste  heat
 boilers for cooling purposes.  This gas stream was measured  only  for
volume flow.   It was  assumed that the pollutants in this  gas stream
                                C-9

-------
would have the same concentration as the gas stream measured going to
the baghouse.  Process conditions and control device parameters (when
applicable) were carefully observed during all the periods when testing
was conducted.  Tests were conducted only when the process facilities
and control equipment were operating within normal limits.  Uncontrolled
arsenic test results are summarized in Tables C-30 and C-32, and
particulate tests are given in Tables C-31 and C-33.
C.3  ANACONDA
     This smelter, when operating, was producing about 545 Mg  (600 tons)
of anode copper daily.  Major process facilities consisted of  a fluid-bed
roaster, electric smelting furnace, six converters (three operational),
and an anode furnace.
     In this process, concentrates and precipitates are blended with
silica flux in approximate proportions of 88, 2, and 10 percent,
respectively.  The blended materials  (containing about 0.96 percent
arsenic) are then fed to a Dorr-Oliver designed fluid-bed roaster by  a
screw feeder which controls the  feed  rate [typically about 91  Mg
(100 tons) per hour] and maintains a  seal on  the roaster.  Fluidizing
air averages 1,062 Nm /min (37,500 scfm).  The air is  supplied through
tuyeres at the bottom of the roaster  to keep  the  bed constantly fluidized
at 1.3 m (70 in.) in depth.
     The fluidized air  reacts with the sulfur contained  in the sulfide
ores to form S02 and calcine.  Approximately  45 percent  of the sulfur
contained  in the feed material is  eliminated.  Because the  reaction  is
exothermic,  no auxiliary fuel is needed except at cold startup.   The
bed temperature  is generally maintained at  582°C  (1,080°F).   Most of
the calcine which is  produced (35  percent)  exits  the reactor  as a fine
                                                                    3
dust suspended in the offgas stream.  The offgases average  1,671  Nm  /min
(59,000 scfm) at 543°C  (1,010°F).  These gases are ducted through a
series  of  primary and secondary  cyclones where 90 to 95  percent of  the
suspended  calcine is  recovered.  The  underflow from the  roaster accounts
for the remaining 15  percent of  the  calcine  produced.
     An electric furnace  is used for  smelting.  Drag conveyors continuously
distribute calcine  produced in the roaster  in combination with some
recycled flue dusts,  along each  side  of the furnace through  a series
of charge  pipes  in  the  roof.  The  furnace working  area is 18  m wide,
                                 C-10

-------
   36 m long,  and  30 m high.   The bath area is 272 m2.   The furnace is
   equipped  with six carbon  electrons.   Each  is  1.65 m in diameter.   The
   electrodes  are  energized  by three  transfomers.   Each has a capacity
   of 12 MVA.   Normal  voltage  applied  is  between  150 and 180 volts    The
   furnace production  capacity is  998  Mg  (1,100 tons)  of matte per  day at
   52 percent  copper.  Matte and  slag  are  tapped  as  required from four
  matte and two slag  tap holes to maintain normal depths  (matte, 71  to
 . 96  cm; slag, 102  to 152 cm).  Offgases  from  the furnace  average
  approximately 425 NmJ/min (15,000 scfm) at  545°C  (1,200°F)..
       Matte from the electric furnace is transported'to one"of three
  Pierce-Smith converters.   Typical  converter process feed  rates include
  41.5 Mg  (45.7 tons) of matte, 4.5  Mg (5.0 tons) of flux,  and 1.4 Mg
   1.5 tons) of cold dope per hour.   The offgas flow is about 2,832 Nm3/min
  (100,000  scfm) per blowing  converter because of excessive air infiltration
  and typically contains  about 2.5 percent SO     Blister copper which is
  produced  is  transported to  an anode  furnace  for further refining  and
  subsequently poured  into anodes  on  a casting wheel.
       Process  offgases from the  electric  furnace and  converter blowing
 operations are combined.  Approximately  2,070 Nm3/min  (73,000 scfm)  of
 the  resultant gas  stream is  treated  in a cooling chamber  (humidified
 and  cooled) and a  venturi scrubber for particulate removal  prior  to
 entering  a 600 Mg  (design)  double-contact acid  plant.  The remaining
 gases are  combined with the fluid-bed roaster offgases and  transported
 through  a  large balloon  flue to  a spray chamber/baghouse  filtration
 plant for  particulate control.
      The combined gas stream, consisting of roaster, electric furnace
 and  converter process gases  totaling  about 5,664 Nm3/min  (200,000 scfm)
 and  ranging  from  230 to  340°C (450 to 650«F)  in  temperature, exits the
 balloon flue  and  enters  two parallel  spray  chambers where the gases
 are  cooled  to  less  than  110'C  (230'F) with watersprays.  Each spray
 chamber is  7.3 m wide, 4.3 m  high, and 30 „ ,ong and  is equipped with
 10 son,c spray nozzles.  Water requirements range from  265 to 303  liters/
™  (7  to 80 gpm).  The cooled gas  stream then  passes  through  two of
three fans  (one is standby), each with a capacity of 6,230 actual
Nm Ann (220,000 scfm),  prior to  entering the baghouse.
                                C-ll

-------
     The baghouse consists of 18 compartments aligned in two rows of
nine.  Each compartment is 4.3 by 14.3 m in cross section and 11.3 m
high above the thimble flow.  The 18 chambers are constructed of
reinforced concrete above the thimble flow and are completely insulated.
Each of the 18 compartments contains 240 Orion bags which are 3.5 m in
diameter and 7.7 m in length.  The cloth area per compartment is
       ?                                                    2
1,656 m .  The baghouse net cloth area totals about 29,802 m .  The
baghouse is designed to effectively treat  11,328 actual Mm /min
(400,000 acfm) at an air-to-cloth ratio of 1.25 to 1.00.  Mechanical
shakers are used for cleaning.  The clean  baghouse exhaust is transported
via  a high velocity  insulated fiberglass flue to the  base of the main
stack and  subsequently discharged to the atmosphere.
     Inlet and outlet  emission measurements  for  both  arsenic and
particulates were conducted  by EPA  across  the spray  chamber/baghouse
on  April 20-26,  1977.  Sampling was conducted upstream  of the  spray
chamber and downstream of  the baghouse.  Two sampling locations  were
required  to obtain  the inlet values.   During all tests, process  conditions
were closely monitored,  and  testing was  conducted  only  when  the  process
facilities were  operating  within  normal  operating  limits.
     The  arsenic and particulate  test  results  for  the spray  chamber/
baghouse  are  summarized  in Tables  C-34 through  C-41.
C.4  PKELPS DODGE-AJO
      This  is  a "green" feed smelter with a production capacity of
about  168 Mg  (185 tons)  of anode copper per day.  Major process  facilities
consist of a  single reverberatory furnace, three Pierce-Smith converters,
 and oxidizing  furnace, and an anode furnace.  Emission control apparatus
 includes electrostatic precipitators for the control of particulate
 emissions from smelting  and converting operations  and a single-contact
 acid plant.
      The reverberatory furnace, which is designed for wet smelting, is
 9 m (30 ft)  wide and 30 m (100 ft) long.  It is fired with natural gas
 or  fuel oil,  depending on the availability  of gas.   The furnace walls
 and roof are constructed of silica brick, and the roof is of a sprung-arch
 design.  The furnace charge components consist  of concentrates  (90 percent)
                                 C-12

-------
   precipitates,  limestone,  and  recycled  flue dusts.   In addition, converter
   slag  is  returned  and  processed.   A charge, usually 1.3 to 3.6 Mg (2 to
   4  tons),  enters the furnace through  one  of six  charge ports,  three on
   each  side  of the  furnace.  Each  port is  equipped with a  high-speed
   belt  slinger to charge wet concentrates  at considerable  velocity.
   Slag  and matte are tapped as required  to maintain  a  normal  bath  depth
   of approximately  120 cm (46 in.)  in the  furnace.
       Offgases from the reverberatory furnace pass  through  two  parallel
  waste heat boilers where the gases are cooled to about 315'C  (600*F)
  and a significant  quantity of  steam is  produced for power  generation'
  The exmng gas stress then  enter a common plenum chamber for mixing
  prior to  treatment in  a hot  electrostatic precipitator which is designed
  to  handle 4,250 actual  mj/min  (150,000  acfm) at 315°C (600°F)    The
  precipitator  is  a  Joy-Western  design, which was  installed in 1973   It
  consists  of two  parallel units  with two stages  each,  and  it has a
  total  collection area  of 3,860  m2.   Gas treatment  averages 6 seconds
  at  a gas  velocity  of 0.9 m (3 ft)/sec.   The pressure  drop across the
  precipitator is 1.3 cm  (0.5 in.)  of water maximum.  The unit has  a
  design efficiency  of 96.8 percent  measured  at its operating temperature
      Offgases  from the converters  pass  through waste  heat  boilers
 where gases are cooled to about 315°C (600°F), and  steam  is  generated
 from the removed heat.   Gases  enter a balloon flue  and then pass
 througn an electrostatic precipitator (ESP).  The ESP  has  two  independent
 horizontal  parallel units  with  three fields  each, which are designed
 to handle  5,940  actual  m>in  (210,000 acfm) at  340°C  (650°F) and
 95.100 pascals  (13.8 psia).  Total ESP Electing surface  area  is
 2,770 m  (29,808  ft2).   After the  converter  gases leave the ESP, they
 pass onto  the scrubbing  section  of the acid  plant where they are
 treated  in  a humidifying tower,  a  cooling  tower,  and a mist precipitator

 (  no' T  ?ff9aSeS 3re ^ PrOC6SSed  1n 3 Sln9le  abs°^°»  544 Mg/day
 (600 ton/day) acid  plant for S02 removal.  Either 93 or 98 percent
 sulfunc acid can be produced.   The  acid plant tail  gas  is  ducted  to
 the main smelter stack.
     Simultaneous inlet and outlet arsenic ercission  measurements  were
conducted by EPA  on  Ouly 13-14,  1976, on  the reverberatory  furnace
                                C-13

-------
ESP.  Operating values for primary and secondary voltage and current
and spark rate were monitored during all  tests to ensure normal operation,
In addition, reverberatory furnace operations were monitored for
normal  operation.  The test results are summarized in Tables C-42 and
C-43.
     Inlet Arsenic emission measurements at the converter ESP  inlet
were conducted by EPA on June 13-15, 1978.  Outlet: arsenic emission
measurements at the inlet of the acid plant, and acid plant outlet
tests,  were conducted at the same time.  Test results are given in
Tables  C-44, through C47.
     Fugitive emissions at Phelps Dodge-Ajo escape the primary hooding
on the  converters during the blowing cycle.  These emissions are
captured by fixed, semicircular-shaped secondary hoods attached to
each converter.  The hoods are approximately 1 m (3  ft) high at the
tallest point.  Although the secondary hoods are in  operation  during
the blowing cycle, they are dampered when  the converters are in the
                                                                    3
roll-out mode.  The secondary system is designed to  handle  1,980  Mm /min
(70,000 scfm) of gases.  The gases are ducted uncontrolled  to  the main
stack.
     Fugitive emissions from the two matte tapping locations are
controlled to a high degree by a rectangular duct about 60  to  90  cm
(2 to 3 ft) above each  tap hole.   The  rectangular vent opening is
about 60 on (24 in.) wide by 30 cm  (12 in.) high and controls  the
fumes from the matte  tapping hole  and  about one- to  two-thirds down
the matte launder which  is not covered.   The gas volume for this
                       3
system  is about 850 Mm  /min  (30,000  scfm)  and  provides considerable
draft to the  vent for  several feet.  The  matte  runs  into a  2.4 to 3m
(3  to 10 ft)  launder  and drops into  a  ladle  on  the floor below.  Fumes
from the matte launder,  are  drawn  into the system handling  the ladle
emissions.  Most of the  time, few  emissions  escape from  the matte
launder.
     Fugitive emissions  from the matte pouring  into  and  from the  ladle
are  captured  in  a cubicle where  the  ladle is  located.  The  ladle  is
placed  on  a specially designed  cradle  on  rail  tracks.  An  electric
motor and  pulley arrangement moves  the ladle  car in  and  out of the
                                 C-14

-------
  cubicle.  During the matte tap,  the  ladle  is  placed  in  the  cubicle,
  and the doors are closed.  The c-jolcle has ventilation  ducts  leading
  from it through a fan, and the gases are discharged  through  the  main
  stack.   The ventilation rate from the two cubicles is 1,700  Nm3/min
  (60,000 scfm).
       Fugitive emissions from the slag tapping location  are captured by
  using  a hood (similar to matte tapping) at the slag  tapping hole.  The
  slag pours into an  open launder and drops through a  hole in the  floor
  into a  slag  ladle  positioned  below.   A hood near the ladle captures
  fume from  the ladle.   Normally,  few emissions emanate from the slag
  tapping operation.  The ventilation  rate  for  the slag tapping system
  is  about 850 NmJ/mTn  (30,000  scfm).   The  duct from the slag  tapping
  system  is  combined with  the matte tapping  duct system,  and  the gases
  are  discharged out of  the main stack.
      Uncontrolled emission measurements for arsenic,  particulate
 matter,  and  S02 were conducted by EPA for  the  secondary  converter hood
 and matte  tapping systems on May  10-12, 1978.  During all tests,
 process  conditions were closely monitored, and testing was conducted
 only when the process  facilities  were operating within nomal  limits.
 The arsenic test results are summarized for both systems  in Tables  C-48,
 and C-50.  The particulate/ S02 test results are suroiarized for both
 systems  in  Tables C-49 and C-51.
 C.5   PHELPS DODGE-HIDALGO
      This smelter produces  about  398 Mg  (438 tons)  of anode copper
 daily.   Major process  facilities  consist of a  rotary dryer,  flash
 furnace, electric  slag  furnace,  three converters,  and  two anode furnaces
      Concentrates, fluxes,  and  dusts  are blended  in approximate proportions
 of 88, 10,  and  2.0 percent, respectively.  .Blended  materials  (containing
 0.005 to 0.06  percent arsenic)  are then  fed  to  a  rotary dryer at about
 88 Mg (97 tons) per hour.  These materials  are  heated  by  gases passing
 through  the steam superheater and  process air  preheaters.  The volume
 of combined  gas streams  is about 708 Nm3/min (25,000 scfm)    They
 enter the dryer at about 315°C (600°F) and leave the dryer at  about
 80 to 100'C  (180 to 220°F).  The dryer discharges into a  lift  tank
containing blended material  and the fine dust portion  of  the ESP
                                C-15

-------
cleaning the dryer offgases.   The lift tank material is charged to the
dry charge bins feeding the flash furnace.
     A flash furnace is used  for smelting.  The feed material is
charged through holes in the  roof on the reaction shaft side of the
furnace.  The furnace reaction shaft is about 8 m (27 ft) inside
diameter by about 36 ft high.  The settling chamber is about 23 m
(76.5 ft) long, 10 m (33 ft)  wide, and 6 m (20.5 ft) high.  The uptake
shaft is about 9 m (30 ft) long, 6.6 m (21.5 ft) wide, and 16.6 m
(54.5 ft) high.  These are all inside dimensions.   Ambient air, preheated
to 370 to 450°C (700 to 850°F), is fed to the furnace; the process
offgas is about 2,430 Nm3/min  (86,000 scfm) at  1,200°C (2,200°F) and
10 percent S02.  The furnace operating temperature  is normally 1,350°C
(2,460°F).  Furnace production is about 715 Mg/day  (650 tons/day)  of
matte and 1,650 Mg/day (1,500  tons/day) of slag.  The slag from the
flash furnace  is tapped directly to the electric furnace  for further
copper recovery.
     Matte from the flash  furnace is transported to one of three
Fierce-Smith converters.   One  converter usually is  on a slag blow, one
on a copper blow, and  one  is  prepared  for  the  next  matte  charges  or  on
standby.  The  converter feed  rates are 24  Mg  (27 tons) of matte  from
the flash furnace, 8 Mg (9 tons) of matte  from  the  electric  furnace,
and 1.7  Mg  (1.9 tons)  of  flux  per hour.   The  converters produce  nearly
13 to  19 Mg/hr (20 to  21  tons/hr) of blister  copper and 4.5  to  5.4 Mg/hr
(5.0  to  6.0 tons/hr)  of converter slag which  is sent  to the  electric
furnace.
      An  electric  furnace  is  used  for  further  processing  of  the  flash
furnace  slag .and  converter slag.  The  furnace uses  about  94  kWh  per
ton of  charged material.   The feed  rates  are  about  66 Mg  (60 tons) of
flash  furnace  slag,  4.5 to 5.4 Mg  (5.0 to 6.0 tons) of  converter slag,
and can  handle about 11 Mg (12 tons)  of  reverberatories  per  hour.   The
furnace  produces  about 7.2 to 8.1 Mg/hr  (8.0  to 9.0 tons/hr) of matte
and  50 Mg/hr  (tons/hr) of slag.   The  matte is transferred to the
converters  by  ladle,  and  the slag  is  transported to a slag  dump.
      Blister copper produced in the converter is transported by ladle
 to one of two  anode furnaces where  it is  further reduced  to anode
                                 C-16

-------
   copper.  The copper is then poured Into anode  molds  on  a casting  wheel
   at 16 to 17 Mg/hr (18 to 19 tonc.hr).   The  anode  copper  is  loaded on
   ra,lcars and sent to a copper refinery  for  further processing
        Fugmve emissions  escaping  the primary hooding  on  the  converters
   are captured by  secondary  hoods designed by the company.  The  secondary

          ir« r r9  doors 8-8  • (29  ft>  ^*•* * <» ^ *<*.  -
   5.5 m  (18 ft) h,gh  and cover  the  converter to  the operator floor
   level.   The  design gas volume handled by the converter fugitive system
   is  about 2,830 ^3/m1n (10MOQ scfm)>  and ^ ta^aj ^ ^system
   gases varies from 150 to 260°C (300 to  500°F).
       Emission measurements  were conducted by EPA during  July 25-26
   1978  ,„ the fugitive converter duct.   The  location was  downstrea.'of
      t e converters   Three  emission measurements were  made for  uncontrolled
   r e ,c   ounng  the tests,  process  conditions  were closely  monitored,
     h n       r;.COndUCted °^ -"  ^  PTOC«S facilities were opera  ing
  «'th,n  norma,  ,,raits.   The test results  are su^arized in  Table  C-52.
  C.6  PHELPS  OOOGE-OOU6LAS
      The  Douglas Reduction Works is  a calcine fed smelter producing
  about 322 Hg  (365  tons) per day of 99.6  percent copper anodes.  C  per
  anodes are sent to a copper refinery for further processing.   The
 -Jor units at the smelter include  24 roasters,  3 reverberatory  furnaces
 5 converters, and  2 anode furnaces.                              "™aces,
      The roaster process consists of 24,  7-hearth Herreshoff  roasters
  rrange  ,„  two parallel  rows  of 12.  Only  18 are no™a,,y  in  operation
 at at™. The roasters  are  standard Herreshoff with a she,,  dialer
 of about  6.7  m (22  ft).  Natura, gas  is introduced near the botto.
   e  roasters    As the  hot gases  rise, feed material  introduced   t t
 to   s forced  down  through the  roasters by the use of rabble arms
   ch spread  the feed around each hearth  level and through openings at
      level.  About   154 to 163 Mg (170 to  Z8C tons) of calcin  per
 roaster is produced each day.
     Calcine from the roasters is discharged  into holding  hoppers
  ere u ,S transferred by  gravity to larry  c,ra  and deliver d  b   rail
to the reverberatory furnace.
                               C-17

-------
     The roaster calcine discharge emission control system consists of
hoods covering the Tarry car with flexible screening or flaps hanging
from the edges of the hoods to the top of the larry car.  The hoods
are split open 20 cm (8 in.) in the center to allow the power pole for
the electric motor driving the larry cars to make contact with the
electrical  source.  Unfortunately, when the hoppers are discharged
into the larry car, the split allows calcine dust to escape from the
hood into the building.  The discharge ducts from the hoods are combined
into a single duct leading to cyclones, a baghouse, a fan, and then
discharged  through a long duct leading to the roaster/reverberatory
stack discharging to the atmosphere.  Each hood can be dampered so
that all  the gas volume (draft) is available to the hoods used for
calcine discharge.
     The baghouse consists of eight compartments with 180 polyester
bags per compartment.  The bags are 328 cm (129 in.) long and 13 cm
(5.0 in.) wide.  The baghouse was designed to treat 1,130 actual m3/min
(40,000 acfm) at ambient temperatures.  Mechanical shakers are used
for cleaning, and the baghouse and cyclone dusts are returned to the
calcine hoppers by screw conveyors.
     Inlet  and outlet emission measurements for arsenic and particulates
were conducted by EPA across the baghouse on May 3-5, 1978.  Sampling
was conducted upstream of the cyclones and downstream of the fan.
During all  tests, process conditions were closely monitored, and
testing was conducted only when the process facilities were operating
within normal operating limits.
     Arsenic test results are summarized in Tables C-53 and C-54.
Particulate test results are summarized in Tables C-55 and C-56.
C.7  KENNECOTT-MAGNA, UTAH
     This smelter was designed to produce nearly 680 Mg (750 tons) of
anode copper daily.  Major process facilities consist of two rotary
concentrate dryers, three continuous Noranda smelting vessels, four
converters, and four anode furnaces.
     Concentrates and precipitates, dusts, and slag concentrates are
blended with silica flux in approximate proportions of 78, 4, 12, and
6 percent,  respectively.  Feed material from storage bins are fed to
                                C-18

-------
   conveyor  belts  that  transport  the  material  to  a  slinger (high-speed
   conveyor  system) which charges  t.',e material  into  the  smelting  vess.els.
   Matte and slag  are normally  tapped about every 30 minutes.   The  slag
   is taken  to a processing station where  it  is cooled,  crushed,  and
   reprocessed to  recover additional copper values  (slag concentrate)
   which are returned to the smelting cycle.  Matte  is transported  by
   ladle to the converters.   Blister copper from the converters is  trans-
   ported by ladle to the anode furnaces.  The strong S02  offgas  streams
   from the smelting vessels and converters are cleaned and delivered  to
  the acid plants.
       Emissions  from the matte tapping operation (hole) are captured in
  an enclosed  cubicle.   The ladle on  a  cradle car on rails (designed for
  this  purpose),  is  in  another  cubicle  beneath the  matte tap floor level
  under the  smelting  vessel.  Ducts direct the fumes from  these operations
  into  one duct which leads to  the main fugitive  duct  system at the roof
  of the smelting  and converter building.   The main  fugitive duct handles
  all fugitive emissions for the  smelter area and discharges these
  fugitive emissions through a  spray chamber  and out the main  stack
  (1,200 ft).  The gas  flow for the matte  tapping hole system  is  708  Nm3/min
  (25,000 scfm) and for the ladle  system is 2,120 Nm3/min  (75,000 scfm).
      Emissions  from the slag  tapping operation are captured  by  a
 rectangular duct opening beside the tapping hole with a  flow of 340  Nm3/min
 (12,000 scfm).   A similar  rectangular duct opening captures  the emissions
 from the  slag ladle in the area  beneath the end of the smelting vessel
 The flow  rate for the  collection of  these emissions is 990 Nm3/min
 (35,000 scfm).
      Emissions escaping  the  primary  converter hoods during the  blow,
 and some  emissions  escaping  the  converters  during  the  roll-out mode'
 are captured  by secondary  hoods  and  ducted  to the  main  fugitive  duct
 wh!ch  was explained previously.   The secondary hooding  system has  a  '
 steel  shell constructed over the  primary  hoods.  The primary  hoods are
 permanent, nonmovable  hood arrangements above  the converters   This
 allows a transition to the fugitive duct  system.  The lower part of
 the secondary hood system consists of four doors that close over the
converter  and  ladle area beneath  the converter.  One door  turns  down
                                C-19

-------
 from the top,  two  doors  close  in  the center,  and the last door moves
 across  the  bottom  area,  which  completely covers  the converter.   Unlike
 other smelters  with  secondary  hoods, the fugitive system at this
 smelter operates at  all  times,  whether  the  converter is  in the blowing
 cycle or roll-out  mode.   The secondary  converter doors  were not operable
 during  the  tests due  to  mechanical  problems with the doors.  The
 design  gas  volume  per converter is  2,745 actual  n3/min  (97,000 acfm)
 at  82°C (180°F).
      The converter process  gas  stream was  also  source tested for
 uncontrolled arsenic.  The  tests  were conducted  before  one of three
 fans  and the control  device to  the  acid  plants.   There  is a settling
 chamber at  the  converters,  and  also  a plenum  chamber with dampers
 where the gases from  each of the  converters can  be  directed to  one,
 two,  or all three  fans (as  needed)  to the acid plants.   The gas volume
 for  this  process stream  per converter is abouc 425  Nm3/min (15,000 scfm).
      Two  rotary dryers are  used at  this  smelter  to  dry  concentrate
 feed  for the smelting vessels.  Emissions from the  dryers are controlled
 by a  cyclone followed by a  spray  chamber scrubber.
      Uncontrolled  arsenic emission measurements  were conducted  by  EPA
 on November 1-14,  1978,  on  the  slag  and  matte tapping system, converter
 secondary hood  system, and  the  converter process  gas stream before the
 control  system.  The  rotary dryer scrubber outlet was also measured
 for arsenic during this  period.   During  all tests,  process conditions
 were  closely monitored,  and testing  was  conducted only when  the process
 facilities were operating within  normal   limits.
      The arsenic test results for these  processes are summarized in
 Tables  C-57 through C-62.
 C.8   KENNECOTT-HAYDEN
      The Kennecott Copper Corporation, Ray Mines  Division,  smelter at
 Hayden, Arizona, was  originally put  on stream in oiid-1958 and extensively
modernized in 1969  and 1973.  The smelter produces  nearly 227 Mg
 (250  tons) of anode copper daily and consists of a  concentrator plant,
 fluid-bed roaster,  reverberatory smelting furnace,  three  converters,
 two anode furnaces, an anode casting wheel, and a double-contact acid
                                   C-20

-------
        Concentrate  and  precipitates  from the concentrator plant are
   blended  with  silica  flux  in  proportions  of 86.3,  2.2,  and 11.5 percent
   respectively.   The blended materials  (containing  6 to  12 percent
   moisture and  less than  0.015 percent  arsenic)  are then fed to a Dorr-Oliver
   designed fluid-bed roaster by a  screw feeder.   The feeder controls the
   feed  rate and maintains a seal on  the roaster.  The  roaster  feed rate
   typically ranges from 45.4 to 63.5 Mg/hr  (50 to 70 tons/hr).   Fluidizing
   air averages 425 Nm3/min (15,000 scfm).  The air  is  supplied  through
   373 tuyeres at the bottom of  the reactor vessel to keep  the bed
   (approximately 1.8 m in depth) constantly fluidized.
       The fluidizing  air reacts with the sulfur contained  in the  sulfide
  ores to form S02 and  calcine.  Approximately 50 percent of the sulfur
  contained in the feed  material is oxidized to S0£.  Because the  reaction
  is  exothermic,  no  auxiliary  fuel  is needed except  for cold startup
  The bed temperature  is  generally  maintained between 565 and 620'C
  (1,050 to 1,150'F).  Most  of  the  calcine  produced  (85 percent) exits
  the reactor  as a fine dust suspended  in the offgas stream.  The offgases
  average 623  Nr^/min (22,000 scfm).  They  are  then  ducted  through a
  series  of four primary and four secondary  cyclones.   In the cyclones
  about  95  percent of the suspended calcine  is  recovered  and subsequently
  conveyed  by  screw conveyor to  the calcine storage  bin.  The underflow
  from the  reactor, which accounts for about 15 percent of  the calcine
  produced, is reclaimed through  an underflow valve  and transported  to
  the  calcine storage bin  by  a  drag  chain conveyor.
      Calcine, precipitator  dust, and flux are then  fed  to  a single
 reverberatory furnace  for smelting.   The reverberatory furnace  (a
 suspended  arch  design)  is 9.1  m (30  ft) wide by  30.5 m  (100 ft) long
 The  furnace  is  charged through two openings  at the  top by a pair of
 Wagstaff feeders.   The furnace is  charged  every  15  minutes for a
 duration of 2 to  3 minutes.  Unless  actually  being  charged, a  slight
 negative draft is maintained across  the furnace  (about -1.5 mm  of
 H20).   The furnace is fired with natural gas but  is equipped with oil
 burners  in the event gas service is  interrupted.  Slag from the furnace
 is periodically tapped into slag pots which are subsequently hauled  by
rail  to the slag dump.   About  650 tons are removed  daily.   Matte is
tapped into ladles which  are moved  into the converter  aisle  when  full.
                                C-21

-------
     The matte ladles are then picked up by an overhead crane and
charged to one of three Pierce-Smith converters.   Each is 4.0 m (13 ft)
in diameter by 9.1 m (30 ft)  long and equipped with 42 tuyeres.
Present converter operation consists of keeping two converters on
charge concurrently.  Each is in the blowing cycle for 50 percent of
the time for 24 hours.   Air flowing at 595 Nm /min (21,000 scfrn) blows
through the tuyeres in the matte charge, flux added, and iron oxide
slag produced.  The slag is then skimmed and poured into ladles.
Unlike most domestic smelters, the converter slag is not charged to
the reverberatory furnace, but is carried to a special slag pit where
it is cooled and subsequently returned to the concentrator and blended
with raw ore.
     Additional matte and dope materials (reverts or copper scrap) are
added to an active converter to produce approximately 90.7 Mg  (100 tons)
of blister copper per load.  Converter process feed rates consist of
about 650 Mg  (718 tons) matte/day, 45.4 Mg  (50 tons) dope/day, and
72.6 Mg (80 tons) flux/day.  The finished blister copper is then
poured into ladles and transported by overhead crane to one of two
anode furnaces.  The blister copper  is completely oxidized with air
and then reduced with propane or natural gas.  Finished anode  copper
is then poured into anode molds on a single casting wheel.  The anodes
are cooled and subsequently loaded onto rail cars  for shipment to a
refinery.  Anode production is about 226 Mg/day  (250 tons/day) of
97.94 percent copper.
     As noted previously, the calcine-laden roaster offgases  pass
through four  parallel sets of primary and secondary cyclones.  An
estimated 95  percent of  the dust  (calcine)  is  recovered, and  the  gas
stream  is cooled from 565°C  (1,050°F) to about 316°C  (600°F).  The
                                                             2
cyclone exhaust, which  has a dust  loading of  about  57.2 g/Nm   (25 gr/scf),
then enters a venturi-type scrubber  where most of  the  particulate  is
collected.  The  scrubbing  liquid  consists of  weak  acid which  is  injected
into the  venturi throat  at a  rate  of 1,500  liters/min  (395  gpm).   The
resultant pressure  drop  across  the  throat  is  about  406 mm  (16 in.)  of
water.  The gas  stream  then  enters  the  smaller of  two  Peabody scrubbing
towers.   The  larger tower  is  used  for  scrubbing  and cooling  the  converter
                                     C-22

-------
   offgas  stream.  Both towers  consist of a lower humidifying  section and
   an upper cooling section.  The  discharge from the venturi enters the
   hunndifying section and passes  upward through a weak  acid spray from
   spray nozzles located at the top of the section.   The coarser solids
   are removed and the heat of  the gas evaporates the weak acid water
      stre                                              -      "
   gas stream then enters  the cooling section  where  it passes through
   three perforated  plates (four on the converter tower) for f1cw distri
   but,on and acid bubble  formation.   The  weak  acid  flowing acroTihT
   Plates cools  the  gas stream to about 46°C (115'F).  The pressure  drop
   across both the venturi and Peabody scrubbers (which services  the
   roaster offgases) is about 610 m  (24 in.) Of water.   The  clean roaster
  gas  stream (which contains about 12  percent SO 1  1,  th     ,  .
  th.  ,,„„„,,       „                 percent iu )  is  then combined with
  the  c   aned converter gas  strea™ prior to entering the acid plant.

  (130   On' LT>beTry  fUrna"  °ffgaSeS 3Verage  'PP™"»*ly 3,682 Nm3
  (130,000 scfm).   They then pass through a pair of  waste  heat boilers
  at approximately 1,260'C (2,300'F)  and exit  at 343'C (650°F)   The
                     r;::;;; ;::;";„;;;•;:;;«;;—  :;-,::-?
 (54.000 ft  ).  The gas retention  t1« within the ESP is about  14 seconds
 Th. average gas velocity is  0.5 „ (1.6 ft) per second.   The  gases
                       ut 288°
 (70 00scTh                                      '       /
 (70,000 scfm).   They  are collected in water-cooled hoods and then
 exhausted through a gas cooler in which  the  gas  stream is  re     by a
 concurrently Hewing, „, trasonically  dispersed water spray.    e
 cooled gas stream (371°c) flows through  an induced fan plenum and into
 an e ectrostatic precipitator for paniculate removal.   Tlle  re  p    Or
™nu  actured  by Western Precipitator,  has two chafers  with
       22                                 c»*er  is
           ri-,
          (which contains
                                                     the
                              C-23

-------
the two Peabody scrubbing  towers  and is treated similarly to the
roaster gas stream.
     After the cleaned and cooled roaster and converter gases are
combined, the resultant gas stream enters three parallel trains of two
mist precipitators in series, where acid mist and any remaining solids
are precipitated.  The gas stream (which typically contains 5 to
8 percent SCL) then enters the double absorption acid plant where it
is dried, the SCL converted to S03> and the S03 absorbed in acid to
form strong acid.  Although designed to produce 1,769 Mg (1,950 tons)
of sulfuric acid per day, only about 771 Mg (850 tons)  per day of
93.5 percent strength sulfuric acid is actually produced.  This represents
about 99.5 percent conversion.
     The gas stream exiting the mist precipitator enters a drying
tower where 93 percent acid is used to remove water  vapor prior to
entering the converter and absorbing systems.  The gas  stream  then
goes to  the main blowers.  One,  two, or  three  1,490  kW  (2,000  hp)
blowers  are used depending on the  volume of gas  available for  processing.
The gas  stream exits  the  main blowers  into the converter system;  the
S0? contained  in the  gas  stream  is  then  converted to SO^.   The converter
contains four  layers  of vanadium catalyst  arranged  in three  passes.
The  first  pass consists of two layers;   the  second  and  third  passes
each have  one  layer.   Tube and shell heat  exchangers are used  to
preheat  the  S02  gas  stream to the operating  temperature by  utilizing
the  heat generated  from  the  exothermic  chemical  reaction within the
converter.   The  preheated gas enters the top  of  the  converter and
passes  through  the  catalyst  layers, exiting  the  converter  after each
pass,  and  entering  a heat exchanger for cooling.   During  plant startups
or during  periods  of low  S02 gas strength, a preheater is  used to
 raise  or maintain  the catalyst  temperature at a  level at which conversion
will  take place.
      Two absorbing  towers are used, an interpass absorber and a final
 absorber.   The gas  leaving the  second  converter  pass goes  through two
 heat exchangers  and then to  the  interpass absorber where the SO  is
 absorbed by 98 to 99 percent acid.  The gas stream  then enters the
 final  converter pass where nearly all  the remaining  SO,, is converted.
                                  C-24

-------
  The gas stream then enters the final absorber where  the  last  traces  of
  S03 are absorbed.  The acid product is pumped through a  series  of
  cooling coils and then stored in any of four 5,000-ton storage  tanks.
  The exit gas from the final absorber passes through  a mist  eliminator
  and is then exhausted through a 30.5 m (100 ft) stack.   The SO,, concentra-
  tion in the exhaust gas is generally about 230 ppm.
       Arsenic emission measurements were conducted by EPA on December 10-13,
  1976.   Concurrent inlet and outlet measurements were performed across
  the venturi  and  Peabody scrubbers.   The scrubbers treat the roaster
  offgases  for particulates  before  they  are  combined with  converter
  process  gases  and subsequently treated  for S02  in tne double-contact
  acid plant.   Additional  arsenic measurements  were made at the  acid
  plant  outlet.
      Process  conditions were carefully  observed,  and  testing was
  conducted only when the subject process facilities  were operating
  within normal operating limits.  The test  results  are summarized  in
  Tables C-63, C-64, and C-65.
  C.9  TAMANO SMELTER (HIBI KYODO SMELTING CO.,) JAPAN
      The Tamano smelter, a toll smelting facility,  is  located  7  km
  (11 miles)  southwest of Uno port and has a production capacity of
 8,500 tons  per month of electrolytic copper.  The smelter consists  of
 one flash furnace, three converters, two refining furnaces, and one
 concentrate  dryer.  Of the  three  converters, usually one is in operation,
 one is  kept  hot,  and one is  kept  on  standby.
      Converter  primary  offgases which  range between 65,000 and  75,000 Nm3/h
 (38,000 to 44,000  scfm)  and  flash  furnace  offgases which  range  between
 65,000  and 80,000  Nm3/h  (38,000 to 47,000  scfm)  are treated in  two
 separate pairs of  electrostatic precipitators  for  particulate removal
 then  introduced into a  156,000  Nm3/h  (92,000 scfm)  capacity acid  plant
 f°r S°2 removal.   S02 content of the converter and  flash furnace  gases
 usually range between 6.5 and 7.5 percent.  The acid plant  SO,  removal
 efficiency is 99.7 percent.   Outlet gases which contain 140  ppm SO
are vented through the main  stack.   A 98 percent sulfuric  acid  is  2
produced at a rate of 30,000 tons/month at full capacity
                                C-25

-------
      Offgases  from  refining  furnaces  are combined  with gas  from the
 power plant  and  passed  through  the  concentrate  dryer.   Total  gases
 from  the dryer are  treated  in a pair  of  electrostatic  precipitators
 prior to introduction to  a  desulfurization  plant.
      Each  converter system  is equipped with a  secondary hood  system
 for fugitive gas capture  which  encloses  the converter  mouth  and ladle
 used  for handling molten  materials.   The hood  has  two  automatic front
 doors  which are  operated  pneumatically.   The  hood  has a  movable roof
 which  is slightly inclined  toward the front.   During closing,  the roof
 slides to  its right.  During the converter  operation when the  hood
 roof  is  opened,  fugitive  emissions  from  the roof are controlled by an
 air curtain system  [rated at 70,000 Nm3/hr  (41,000  scfm)].
      The bulk of fugitive gases up  to 190,000  Nm3/hr (112,000  scfm)
 with  low S02 content collected  in the secondary hood are  passed through
 a dust chamber,  a baghouse system,  and the  main stack  to  the  atmosphere.
     Another smaller volume, fugitive gas stream with  a high  S02
 content  is continuously fed to  a 200,000 Nm3/hr (118,000  scfm)  capacity
 desulfurization plant.  Gas volume  to the desulfurization plant includes
 about 72,000 Nm3/hr  (42,000 acfm) from the  refining furnace,  power
 plant, and flash dryer,  and about 30,000  Nm3/hr (18,000 scfm)  leakage
 gases from the converters and refining furnaces.
     On  March 12 and 13, 1980,  visible emission observations were  made
 of the converter secondary hood system during various  nodes of  converter
 operation.  Tables  C-75  through C-80  present the results of these
 observations.
 C.10  TEST DATA (TABLES)
     This section contains summary data  tables of the  arsenic  and
 particulate emission tests,  and the visible emissions  observations,
conducted by EPA between December 1976,  and June 1980.   The following
notes  apply to Tables C-3 through C-65:

     a)   Data  not  reported.
     b)   Run no. 1  of Reverberatory-North data was performed
          on 6/28/77.
     c)   Not applicable.
                                    C-26

-------
                                         Table  C-l.    SUMMARY  OF  EMISSION TESTS
           Plant
o
i
           Phelps Dodge
             AJo. Arizona
ASARCO
  II  Paso.  Texas
           Phelps  Dodge
             'Mayas.  New
             Hex ico
                s Dodge
            Douglas .
            Ar i zona
                                Process
                               Facility


                       Acid plant  from converters

                     Heverberatory furnace
                       (matte  and  slag  tapping)

                     Converters  (blow cycle)
HJ Itlhearth roasters and
  reverberatory furnace
Mult(hearth roaster discharge
                               Heverberatory furnace
                                 (matte  tapping)

                               Converter building
                    Converter  secondary hoods
                   Mu! Unearth roaster
                     •lischarge
                                                                 Hot ESP + mlst
                                                                 ESP * coolers
                                                                 None
                                                                 None
Spray chamber/
  cold ESP

Discharged Into
  above system

Baghouse
                                                     Baghouse
                                 None
                                 Daghouse
                                                         SampI Ing
                                                        Locatton(s)
                                                      Inlet and outlet
                                                                                     Before entering
                                                                                       main stack
                                                                                     Before  entering
                                                                                      main  stack
 Inlet  and  outlet

 Inlet


 Inlet



 Inlet and outlet


 Inlet



Inlet and outlet
                                                                                                              Sample
                                                                                                               Type
 Arsenic

 Arsenic
 Particulate
 S02

 Arsenic
 Particulate
 S02

 Arsenic
 Particulate
 Arsenic
 Particulate
 S02

 Arsenic
 Particulate
 S02

 Arsenic
 Particulate
S02

Arsenic
                                                                         Arsenic
                                                                         Par t iculate
                                                                         S0>

-------
                              Table C-l.   SUMMARY OF EMISSION  TESTS (Concluded)
Plant
ASARCO
Tacoma. Washington











Kennecott
Magjna, Utah










Anaconda
Anaconda. Montana
Kennecolt
llayden, Arizona

Process
Facility


Arsenic roasters

Reverberalory furnace
Matte tappping
Slag tapping
Converter slag return
Slag dumping

Metallic arsenic
Anode furnace

Converters
Converter
Secondary hood

Roll-out cycle

Noranda furnace
Matte tapping

Slag tapping

Concentrate dryer
Fluid-bed roaster, electric
furnace, and converter
Fluid-bed roaster
Fluid-bed roaster and
converters
Control
Equipment


Baghouse

Cold ESP
Cold ESP
Cold ESP
Cold ESP
None

Baghouse
None

Acid Plant

None

None


None

None

Scrubber
Spray chamber/
baghouse
Venturi scrubber
Acid plant

Sampl ing
.Locatlon(s)


Inlet and outlet

Outlet
Inlet
Inlet
Inlet
Dump site (grab
samples of slag)
Inlet and outlet
Inlet

Inlet

Uncontrolled duct

Uncontrolled duct


Uncontrolled duct

Uncontrolled duct

Outlet
Inlet and outlet

Inlet and outlet
Outlet

Sample
Type


Arsenic
S02
Arsenic
Arsenic
Arsenic
Arsenic
Arsenic

Arsenic
Arsenic (tests
by ASARCo)
Arsenic

Arsenic
so2
Arsenic
S02

Arsenic
S02
Arsenic
S02
Arsenic
Arsenic
Partlculate
Arsenic
Arsenic

ro
CD

-------
           Table C-2.
INDEX TO ARSENIC  AND PARTICULATE TEST DATA TABLES BY PROCESS FACILITY AND SAMPLE TYPE
I
ro
^ 	
Process Facility
Arsenic roasters
Fluid-bed roaster
Multi-hearth roasters and
reverberatory furnace
FluJd-bed roaster, electric
furnace, and converter
Fluid-bed roaster and
converters
Calcine discharge
Roaster calcining fugitives
Multi-hearth roaster
fugitive discharge
Reverberatory furnace
Reverberatory furnace
Matte tapping
Matte tapping

Matte and slag tapping

Noranda furnace
Matte tapping
Reverberatory furnace
Slag tapping
Noranda furnace
Slag tapping
Reverberatory furnace
Converter slag return
Plant
ASARCO-Tacoma
Kennecott-Hayden
ASARCO-E1 Paso
Anaconda- Anaconda
Kennecott-Hayden
ASARCO-Tacoma
ASARCO-E1 Paso
Phelps Dodge-
Douglas
ASARCO-Tacoma
ASARCO-Tacoma
ASARCO-E1 Paso

Phelps Dodge-AJo


Kennecott-Magna

ASARCO-Tacoma

Kennecott-Magna

ASARCO-Tacoma
Control Equipment
Baghouse
Venturl scrubber
Spray chamber/
cold ESP
Spray chamber/
baghouse
Acid plant
None
Spray chamber/
cold ESP
Baghouse
Cold ESP
Cold ESP
Baghouse

None


None

Cold ESP

None

Cold ESP
Sample Type
Arsenic
Arsenic
Arsenic
Partlculate
Arsenic
Partlculate
Arsenic
Arsenic
Arsenic
Partlculate
Arsenic
Partlculate
Arsenic
Arsenic
Arsenic
Partlculate
Arsenic
Partlculate

Arsenic

Arsenic

Arsenic

Arsenic
Table Numbers
In Appendix C
C-3, C-4
C-63, C-64
C-13, C-14. C-15, C-16, C-17
C-18, C-19, C-20, C-21, C-22. C-29
C-34, C-35, C-36, C-37
C-38, C-39, C-40, C-41
C-65

C-30
C-31
C-53, C-54
C-55, C-56
C-8
C-10
C-32
C-33
C-48
C-49

f-51
\j~jy
C-ll

f-fin
\j — UU
C-12

-------
Table C-2.   INDEX TO ARSENIC AND PARTICIPATE TEST DATA TABLES BY PROCESS FACILITY AND SAMPLE TYPE (Concluded)
Process Facility
Converters
Converters (blow cycle)

Converters
(V, Converter fugitives
o Full cycle
Rollout phase
Acid plant from converters

Converter building

Converter secondary hoods
Metallic arsenic process
Concentrate dryers
Plant
ASARCO-E1 Paso
Phelps Dodge-Ajo

Kennecott-Magna

Kennecott-Magna
Kennecott-Magna
Phelps Oodge-Ajo

ASARCO-E1 Paso

Phelps Dodge-Hidalgo
ASARCO-Tacoma
Kennecott-Magna
Control Equipment
Acid plant
None

Acid plant

None
None
Hot ESP, mist ESP
and coolers
Baghouse

None
Baghouse
Scrubber
Sample Type
Arsenic
Arsenic
Partlculate
Arsenic

Arsenic
Arsenic
Arsenic

Arsenic
Partlculate
Arsenic
Arsenic
Arsenic
Table Numbers
In Appendix C
C-23, C-24
C-50
C-51
C-58

C-61
C-62
C-42, C-43, C-44, C-45. C-46. C-47

C-25, C-26
C-27, C-28
C-52
C-5S C-6S C-7
C-57

-------
               Table C-3.   SUMMARY OF ARSENIC TEST  DATA  -  ROASTER
                      BAGHOUSE  INLET, ASARCO-TACOMA  SMELTER
r\un iiu .
— 	 	 	 . 	
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
t _
1
""'
9/15/78
80
a
a

173,621 '
201

A A
f. *f
0,2
20.4
0.65



0.1371
0.1301
204.0

0.1377
0.1303
204.3
2
— — — 	 __
9/15/78
84
a
a

175,277
185


3.4
0.2
20.4
0.96



0.1280
0.1249
192.2

0.1295
0.1261
194.1
3
— 	 —_
9/16/78
80
a
a

184,141
203


4.5
0 7
w • c.
20.4
0.81



0.1101
0.1092
173.6

0.1111
0.1103
175.2
Average

81
a
a

177,680
197


4.1
01-1
.2
20.4
0.81



0.1248
0.1212
189.6

0.1258
0.1220
190.9
Percent Isokinetic
                              94.6
108.2
                                                        95.0
                                    C-31

-------
               Table  C-4.  SUMMARY OF ARSENIC TEST  DATA  -  ROASTER
                     BAGHOUSE OUTLET, ASARCO-TACOMA  SMELTER
— 	 • 	 __
Run No.
~~~~~~~~~ — — — — _ __ __ _ _
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %}:
Water
C02
2
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
— — — — — — -^—. _ __
i
i
•"—*—"•——————*_
9/15/78
120
a
a

171,887
191

5.9
0.2
20.4
0.70



0.00027
0.00026
0.396

0.00028
0.00027
0.416
'

9/15/78
120
a
a

174,633
189

5.0
0.2
20.4
0.80



0.00027
0.00026
0.401

0.00028
0.00027
0.428
•~— — — — •
3
9/16/78
120
a
a

178,671
180

5.6
0.2
20.4
0.52



0.00028
0.00028
0.439

0.00065
0.00064
0.993
•" •" "-•i .. i i-,ii..
Average

120
a
a

175,064
187

5.5
0 2
VJ » t_
20.4
0.67



O.OOC
0.000
0.412

O.OOC
0.000
0.612
Percent Isokinetic
99.5
                                         100.1
                         102.3
                                   C-32

-------
   :~5-  SUMMARY OF ARSENIC TEST DATA -
KITCHEN BAGHOU3E INLET, ASARCO-TACOMA SMELTER
                                                   Average
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
co2
n
U2
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
?b/hr
9/24/78
96
a
a

14,505
133

•5 (•
-J.O
0.2
20.8
0.32



0.7503
0.7484
93.23

0.7504
0.7486
93.23
9/24/78
96
a
a

16,590
136


4.0
0.2
20.8
0.57



0.6632
0.6400
94.21

0.6632
0.6454
94.21
9/25/78
96
a
a

17,560
140


2.9
0 2
w * t.
20.8
0.49



0.6593
0.6400
99.18

0.6695
0.6402
99.18

96
a
a

16,218
136


3.5
Or\
.2
20.8
0.46



0.6909
0.6755
95.54

0.6944
0.6756
95.54
       99.2
97.5
                                 94.8
            C-33

-------
                Table C-6.  SUMMARY OF ARSENIC TEST DATA --
          METALLIC ARSENIC BAGHOUSE INLET, ASARCO-TACOMA SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
Percent Isokinetic
1
9/24/78
94
a
a
12,989
242
2.0
0.1
20.0
0.24

0.8440
0.8343
93.79

0.8441
0.8345
93.81
100.7
2
9/24/78
96
a
a
15,147
233
3.1
0.1
20.0
0.17

0.0004
0.0004
0.0544

0.0006
0.0006
0.0813
94.5
3
9/25/78
96
a
a
15,469
207
2.7
0.1
20.0
0.22

0.9289
0.8914
123,0

0.9289
0.8915
123.0 •
98.5
Average

95
a
a
14,533
227
2.6
0.1
20.0
0.21

0.591
0.527
72.27

0.591
0.527
72.29

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

-------
                 .       C'7'  SUMMARv OF ARSENIC TEST DATA -
                 ARSENIC BAGHOUSE OUTLET (METALLIC AND KITCHEN)
                              ASARCO-TACOMA SMELTER
	 — — ' — 	 	 —
Run No.

Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. 35):
Water
CO,
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
— • 	
i
i
9/24/78
96
a
a

29,109
182

1.0
0.2
20.4
0.28



0.0303
0.0303
7.59

0.0304
0.0304
7.61
— — 	 , —

1
9/24/78
96
a
a

35,958
163

2.1
0.2
20.4
0.19



0.0068
0.0066
2.11

0.0069
0.0067
2.13
— — — _ — *___^_ _

3
• 	
9/25/78
96
a
a

33,726
160

2.8
0.2
20.4
0.63



0.0417
0.0401
12.07

0.0420
0.0403
12.14
- — — — — — _____

Average


96
a
a

32,931
160

2.0
0.2
20.4
0.37



0.0253
0.0248
7.26

0.0255
0.0249
7.29
Percent Isokinetic
97.5
                                         101.3
                          97.8
                                   C-35

-------
                    DnrDD~8;   SUMMARY  OF ARSENIC  TEST  DATA  -
                    REVERB  ESP  OUTLET,  ASARCO-TACOMA SMELTER
ruin no.
' 	 — 	 • 	 . 	
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
C02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
• — • — . 	
9/15/78
116
a
a

441,557
220

4.4
0.0
20.0
0.54



0.01632
0.01538
61.68

0.01669
0.01574
63.09
2
•
9/16/78
108
a
a

454,539
214

5.1
0.0
20.0
1.17



0.00897
0.00863
35.04

0.00915
0.00881
25.77
3
" ™ •••^"^^•"•^^•^••^MW^
9/18/78
108
a
a

443,619
188

3.7
0.0
20.0
0.32



0.00375
0.00364
13.99

0.00418
0.00405
15.59
Average
•
111
a
a

443,238
207

4.4
0 0
w • VJ
20.0
0.68



0.009
0.009
36.90

0.010
0.009
38.15
Percent Isokinetic
104.2
                                          107.7
                          102.2
                                   C-36

-------
Table C-9.  SUMMARY OF ARSENIC TEST DATA -
 CALCINE DISCHARGE, ASARCO-TACOMA SMELTER
nun NO.
— — 	 — 	
Date
Test Duration -. min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
C02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
	 	 	 	 _
9/20/78
15
a
a

1,239
78

2.5
0.0
20.0
0.07



0.0939
0,0936
1.007

0.0946
0.0943
1.014
2
— — — — — • 	
9/20/78
13
a
a

1,293
78

0.0
0.0
20.0
1.32



0.1463
0.1458
1.623

0.1497
0.1492
1.661
3
	 • — • — — — — __
9/21/78
7
a
a

1,524
80

0.0
0 0
\J • w
20.0
0.22



0.2074
0.2053
2.721

0.0292
0.2071
2.744
Average

12
a
a

1,352
79

0.83
O/^
.0
20.0
0.54



0.1399
0.1384
1.784

0.1418
0.1724
1.806
           96.6
94.2
                                    108.1
               C-37

-------
                Table C-10.  SUMMARY OF ARSENIC TEST  DATA
                   MATTE TAPPING, ASARCO-TACOMA SMELTER

Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
C02
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
9/19/78
78
a
a
18,944
134
1.5
0.0
20.0
0.18
0.02146
0.02127
3.490
0.02297
- 0.02271
3.26
2
9/20/78
75
a
a
18,181
163
0.8
0.0
20.0
0.39
0.10649
0.10394
16.59
0.10657
0.10403
16.60
3
9/21/78
74
a
a
18,329
164
0.9
0.0
20.0
0.21
0.10332
0.10130
16.24
0.10344
0.10142
16.26
Average

76
a
a
18,485
154
1.1
0.0
20.0
0.26
0.07
0.07-
12.11
0.07
0.07
12.20
Percent Isokinetic
90.5
91.0
90.0
                                    C-38

-------
                   Table  C-ll.   SUMMARY  OF  ARSENIC  TEST DATA
                       SLAG.TAPPING,  ASARCO-TACOMA  SMELTER
Run No.

Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
2
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
i
j.
9/19/78
120
a
a

18,351
68

1.5
0.0
20.0
0.03



0.00340
0.00335
0.536

0.00348
0.00343
0.548


9/20/78
111
a
a

16,571
106

2.3
0.0
20.0
0.05



0.01009
0.01005
1.431

0.01011
0.01007
1.434

3
9/20/78
60
a
a

17,219
119

0.3
0.0
20.0
0.07



0.00790
0.00784
1.170

0.00793
0.00787
1.175

Average

97
a
a

17,380
98

1 4
n n
u « u
20.0
0.05



0,00676
0.00670
1.046

0.00681
0.00675
1.052
Percent Isokinetic
91.2
                                           95.9
                          92.1-
                                   C-39

-------
              Table C-12.   SUMMARY OF ARSENIC TEST DATA --CONVERTER SLAG
                        RETURN, ASARCO-TACOMA SMELTER
Run No.
Date

Test Duration - min.

Charge Rate - ton/hr

Arsenic Rate - Ib/hr

Stack Effluent

   Flow rate (dscfm)

   Temperature (°F)

   Stream (vol . %) :
     Water
     C02
     02
     S0
   Probe, cyclone,
   and filter catch
     gr/dscf
     gr/acf
     Ib/hr

   Total catch
     gr/dscf
     gr/acf
     Ib/hr

Percent  Isokinetic
9/19-21/78

    23

     a

     a



  23,207

      95
        0.8
        0.0
       20.0
        0.7
      0.00139
      0.00135
      0.2763
      0.00149
      0.00145
      0.2962

     91.4
                                    C-40

-------
Table C-13.  SUMMARY OF ARSENIC TEST nnra
 R ESP INLET (ROASTER), ASARco-EL PASO
                                                      Average
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
C02
S02
Emissions - Arsenir
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
' Q ** i"» 21 M -f- T-.—I.' . .
6/26/77
116
41.2
98.0

140,927
173

67
. 7
0.0
16.6
a



0.0060
0.0040
7.200

0.0100
0.0068
12.12
6/27/77
120
28.9
172.9

149,764
211


5.3
0.5
18.7
a



0.0080
0.0052
10.32

0.0103
0.0067
13.21
6/28/77
120
42.7
395.5

56,040
231


9.4
n A
U.'f
20.5
a



0.0312
0.0188
14.99

0.0380
0.0229
18.25

119
37 6
w / * \J
222 1
t"£- *- * .1
115,577
205


7.2
0.3
18.6
a


0.0109
0.0069
9 807
v « w vj /
0.0147
0.0094
13.58
         102.4
66.8
                                   109.2
                C-41

-------
                Table C-14.  SUMMARY OF ARSENIC TEST DATA --
          R & R ESP INLET  (REVERB-NORTH), ASARCO-EL PASO SMELTER

Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %) :
Water
CO 2
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
6/26/77
120 -
41.2
98.0

38,664
932

6.9
7.6
9.9
a



0.0471
0.0145
15.61

0.1219
0.0374
40.39
2
6/27/77
120
28.9
172.9

42,288
786

17.9
7.6
9.9
a



0.2888
0.0870
104.7

0.2951
0.0889
107.0
3
6/28/77
120
42.7
395.5

40,421
753

19.0
7.5
9.9
a



0.2903
0.0890
100.6

0.3177
0.0974
110.1
Average

120
37.6
222.1

40,458
824

14.6
7.6
9.9
a



0.212
0.06^
74.95

0.24;
0.07!
86.81
Percent Isokinetic
103.8
107.1
112.7
                                    C-42

-------
Table C-15.  SUMMARY OF ARSENIC TFST HAT/I
ESP INLET (REVERB-SOUTH)? ASARCO-EL PASO SMELTER
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent.
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
C02
n
U2
S02
Emissions - Arsenir
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
' a k* /•* « « -f- T ,» *. i. • . . «
6/26/77
120
41.2
98.0

19,759
787

14.2
8 1
o • 1
9 3
J • *J
a



0.1280
0.0404
21.68

0.1289
0.0407
21.83
6/27/77
120
28.9
172.9

22,057
766

13.4
8«
.1
9*^
.3
a



0.7878
0.2544
148.9

0.7967
0.2572
' 150.6
6/28/77
120
42.7
395.5

25,981
614

25.2
ft ?
o. c.
9.2
a



0.6096
0.1949
135.7

0.6462
0.2067
143.9

120
37.6
222.1

22,599
722

16.6

8.1
9.3
a
u


0.5272
0.1692
106.8

0.5444
0.1747
110.5
          95.8
97.7
                                   113.4
               C-43

-------
               Table  C-16.   SUMMARY  OF  ARSENIC  TEST  DATA  --   i
               R  &  R ESP  INLET  (TOTAL),  ASARCO-EL  PASO  SMELTER

Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %) :
Water
C02
02
S02
Emissions - Arsenic
Probe, cyclone,
and fil ter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
Percent Isokinetic
1
6/26/77
120
41.2
98.0
199,350
381
7.5
2.3
14.6
a
0.0261
0.0096
44.49
0.0435
0.0161
74.34
c
2
6/27/77
120
28.9
172.9
214,109
382
8.6
2.7
3.3
a
0.1438
0.0470
263.9
0.1476
0.0487
270.8
c
3
6/28/77
120
42.7
395.5
122,442
485
15.9
4.4
14.6
a
0.2426
0.0812
251.3
0.2594
0.0865
272.3
c
Average

120
37.6
222.1
178,634
411
10.1
2.9
15.5
a
0.12:
0.04(
191.6
0.13'
0.04!
210.9

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

-------
                   Table C-17.  SUMMARv OF ARSENIC  TEST  DATA  -
                     R & R ESP OUTLE'i, ASARCO-EL  PASO  SMELTER
Run No.
— 	 	 	 — 	 	
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
CO,
X 2
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
• — 	 .
1
— 	 __ 	 .
6/26/77
153
41.2
98.0

201,030
216

C Q
0. O
3c
. b
19.0
a



0.0009
0.0006
1.603

0.0020
0.0012
3.401
— — — — — — — — — .
2
"•
6/27/77
153
28.9
172.9

209,891
219


6.1
3.5
19.0
a



0.0031
0.0019
5.503

0.0041
0.0026
7.400
3
6/28/77
153
42.7
395.5

222,719
221


1.4
0.0
20.8
a



0.0014
0.0010
2.761

0.0018
0.0012
3.393
Average

153
37.6
222.1

211,213
219


4.8
2.3
19.6
a



0.0018
0.0012
3.302

0.0026
0.0017
4.723
Percent Isokinetic
100.4
                                          100.4
                           95.4
                                   C-45

-------
         Table C-18.  SUMMARY OF PARTICULATE TEST DATA   --  R  &  R  ESP
                  INLET (ROASTER), ASARCO-EL PASO SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
02
S02
Emissions - Particulate
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
6/29/77
120
43.5
a

68,948
190

9.2
1.8
20.7
a



2.245
1.439
1,327

2.276
1.458
1,345
2
6/30/77
90
47.6
a

63,256
197

10.5
1.4
17.9
a



1.954
1.222
1,059

1.977
1.236
1,072
Average

105
45.5
a

66,102
193

9.8
1.6
19.3
a



2.106
1.335
1,199

2.133
1.352
1,214
Percent Isokinetic
122.0
108.5
                                    C-46

-------
           Table C:J?-  pMRY OF PARTICULATE TEST DATA  -  R &  R  FSP
                  INLET (REVERB-NORThj, ASARCO-EL PASO SMELTER
Run No.
"" 	 " -
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
C02
0,
z
S02
Emissions - Particulate
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
6/28/77
120
43.5
a

40,711
687

18.8
5.5
5.0
a



2.098
0.6814
732.0

2.188
0.7106
763.3
2
6/30/77
120
47.6
a

39,889
672

16.1
8.0
13.2
a



1.47.4
0.5011
504.0

1.534
0.5214
524.4
Average

120
45.5
a

40,300
680

17.5
6.7
. 9.1
a



1.786
0,5913
619.2

1.861
0.6170
643.9
Percent Isokinetic
109.8
                                              111.6
                                   C-47

-------
        Table C-20.   SUMMARY OF PARTICULATE TEST DATA  -- R & R ESP
               INLET (REVERB-SOUTH), ASARCO-EL PASO SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
CO,
f,
02
S02
Emissions - Participate
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
6/29/77
120
43.5
a

22,504
751

9.6
9.9
3.4
a



Q.9863
0.3377
190.2
1.0892
0.3730
210.1 1
2
6/30/77
120
47.6
a

25,680
728

12.8
8.6
8.5
a



5.3713
1..8078
1,182
5.3974
1.8166
,188
Average

120
45.5
a

24,092
739

11.3
9.2
6.1
a



3.323
1.121
718.8
3.385
1.142
731.3
Percent Isokinetic
109.7
112.3
                                     C-48

-------
            Table C-21,
SUMMARY OF PANICULATE TEST DATA  - g & R
INLET (TOTAL), ASARCO-EL PASO SMELTER*
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %);
Water
C02
02
2
S02
Emissions - Participate
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
6/29/77b
120
43.5
a

132,163
439

1 /•* y-t
12.2
4 A
.3
12.9
a



1.9854
1.0180
2,249

2.0468
1.0431
2,318' 2
6/30/77
120
47.6
a

128,825
450


12.7
4.9
14.6
a



2.1270
1.1154
2,745

2.5217
1.1306
,784

120
45.5
a

130,494
444


12.5
4 6
" • V
13.7
a



2.2319
1.0658
2,494

2.2801
1.0862
2,548
Percent Isokinetic


'These data are derived  from Tables C-18, C-19, and C-20."
                                   C-49

-------
        Table  C-22.   SUMMARY  OF  PARTICULATE  TEST  DATA  - R & R ESP
                          OUTLET,  ASARCO-EL  PASO  SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
C02
02
S02
Emissions - Particulate
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
6/28/77
153
43.5
a

215,414
220
10.7
1.3
19.0
a


0.0489
0.0292
89.79
0.0619
0.0372
114. 4
2
6/30/77
153
47.6
a

233,278
220
8.7
2.0
20.0
a


0.0372
0.0228
74.28
0.0478
0.0286
95.47
Average

153
45.5
a

224,346
220
9.7
1.7
19.5
a


0.0427
0.0259
81.73
0.0546
0.0327
104.6
Percent Isokinetic
104.6
98.9
                                    C-50

-------
               Table C-23.  SUMMARY OF ARSENIC TEST DATA - DC ACID
                           PLANT INLEi, ASARCO-EL PASO SMELTER

Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
C02
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
	 	 — 	
6/21/77
105
a
a

58,352
431

9 1
£.1
0.0
17.6
a



0.96.56
0.4882
482.9

1.0291
0.5203
514.6
2
	 . —
6/22/77
96
a
a

55,189
408

5-»
.7
0.0
17.2
a



0.0810
0.0405
38.32

0.0997
0.0498
47.16
3

6/23/77
96
a
a

54,842
392


4.9
0.0
13.9
a



0.1083
0.0558
50.89

0.1143
0.0589
53.73
Average


99
a
a

56,128
410


4.2
0.0
16.2
a



0.3964
0.2006
196.5

0.4265
0.2158
211.3
Percent Isokinetic
98.5
                                      112.2
                        106.0
                                   C-51

-------
              Table C-24.  SUMMARY OF AKSENIC  TEST  DATA  —  DC  ACID
                      PLANT OUTLET, ASARCO-EL  PASO  SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %}:
Water
C02
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
6/21/77
132
a
a

2
6/22/77
132
a
a

68,108 66,574
147

0.0
0.0
17.6
a



0.0001
0,0001
0.043

0.0001
0.0001
0.048
147

0.0
0.0
16.0
a



0.0008
0.0006
0.452

0.0014
0.0010
0.783
3
6/23/77
132
a
a

66,214
151

0.0
0.0
13.6
a



0.0005
0.0004
0.267

0,0005
0.0004
0.277
4
6/24/77
. 132
a
a

64,643
156

0.0
0.0
11.7
a



0.0001
0.0001
0.050

0.0002
0.0001
0.085
Average

132
a
a

65,429
153

O.C
O.C
14.7
a



O.C
O.C
0.;

0.0(
o.oc
0.2?
Percent Isokinetic
73.0
76.2
96.4
97.4
                                         C-52

-------
        T_LT  /* or*
        Table C-25.  iunriAKY OF ARSENIC  TEST  DATA — CONVERTER BUILDTNf
                      BAGHOUSE INLET, ASARCO-EL  PASO SMELTER
————————— 	
Run No.

Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfmj
Temperature (°F)
Stream (vol. %):
Water
C02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
— — — — - — 	 	 _
i
i
— 	 — — — _
1/18/78
101
41.7
212.0

521,956
100

i n
1 . U
0.0
20.5
a



0.00270
0.00240
12.11

0.00272
0.00242
12.12
i - ...i —

•— • __
1/19/78
100
32.7
116.1

528,463
97

OM
.4
0.0
20.5
a



0.00090
0.00082
4.28

0.00091
0.00083
4.33


3

1/23/78
100
37.1
165.2

506,479
123


1.1
0 0
\J « \J
20.5
a



0.00067
0.00058
2.88

0.00067
0.00058
2.90


Average


100
37.2
164.4

527,497
107


0.8
On
.0
20.5
a



0.00142
0.00123
6.42
-
0.00143
0.00128
6.45
Percent Isokinetic
95.0
                                         94.2
                          93.8
                                   C-53

-------
      Table C-26.  SUMMARY OF ARSENIC TEST DATA -- CONVERTER BUILDING
                   BAGHOUSE OUTLET, ASARCO-EL PASO SMELTER

Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
S02
Emissions - Arsenic
Probe : cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr .
1
1/18/78
160
41.7
212.0
437,609 526
101
0.0
0.0
20.5
a
0.00017
0.00015
0.672
0.00017
0.00015
0.672
2
1/19/78
160
32.7
116.1
,565 490
99
0.7
0.0
20.5
a
0.000005
0.000005
0.025
0.000010
0.000009
0.040
3
1/23/78
200
37.1
165.2
,455
124
1.3
0.0
20.5
a
0.000006
0.000005
0.026
0.000007
0.000006
0.027
Average

173
37.2
164.4
491,062
108
0.7
0.0
20.5
a
0.000
0.000
0.241
O.OOC
O.OOC
0.246
Percent Isokinetic
105.7
101.1
102.6
                                    C-54

-------

  Date

  Test Duration - min.

  Charge  Rate  - ton/hr

  Arsenic  Rate  -  lb/hr

  Stack Effluent

    Flow rate  (dscfm)

    Temperature (°F)

    Stream (vol.  %):
      Water
      C02

      S02

Emissions - Particular^

   Probe, cyclone,
   and filter catch
1/17/78
105
a
a
1/18/78
100
41.7
a
1/21/78
100
43.8
a

102
42.8
a
435,427
115
514,279
113
510,318
115
486,675
114
1.2
0.0
20.5
0.017
1.4
0.0
20.5
0.001
1.4
0.0
20.5
0.001
1.3
0.0
20.5
0.006
y / i-o^ i
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
~Pnt" TcnUn/^-*--;-
0.0263
0.0235
98.25
OO rt -i /-»
.2812
0.2507
1,049
0.0080
0.0202
101.9

0.2686
0.2346
1,181
0.0309
0.0276
134.5

0.031
0.027
134.5
0.0215
0.0238
111.5

0.2749
0.2427
788,0
                            109.7
                                          94.5
                                                       96.3
                                  C-55

-------
       Table C-28.  SUMMARY OF PARTIC'ILATE TEST DATA -- CONVERTER BUILDING
                    BAGHOUSE OUTLET, ASARCO-EL PASO SMELTER

Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
C02
02
S02
Emissions - Particulate
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
1/17/78
120
a
a
526,089
114
1.0
0.0
20.5
a

0.0051
0.0045
22.87

0.1117
0.0979
503.8
2
1/18/78
160
41.7
a
471,191
118
0.9
0.0
20.5
a

0.0011
0.0010
4.45

0.1402
0.1260
567.4
3
1/21/78
160
43.8
a
496,907
114
1.1
0.0
20.5
a

0.0005 '
0.0004
1.96

0.0081
0.0072
34.29
Average

146
42.8
a
498,062
115
1.0
0.0
20.5
a

0.0072
0.001*
9.76

0.086;
0.077(
368.5
Percent Isokinetic
98.0
97.7
103.7
                                      C-56

-------
         Table C-29.  SUMMARY OF PARTICULATE TEST DATA - ROASTER/REVERBERATORY
                               ESP OUTLET, ASARCO-EL PASO SMELTER
Run No.
— 	
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
C02
S02
Emissions - Particulate
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
i
1/26/78
144
64.0
a

200,352
219

6.0
3.0
20.5
a



0.0608
0.0520
104.0

0.1054
0.0901
180,2


1/26/78
147
64.0
a

207,295
205

6.2
3.0
20.5
a



0.0909
0.0787
160.7

0.1118
0.0968
197.7

3
1/27/78
150
60.1
a

202,651
220

7.3
3 0
20.5
a



0.0411
0.0365
71.1

0.0553
0.0491
95.7

Average

147
62.7
a

203,433
215

6.5
3f\
.0
20.5
a



0.0643'
0,0577
112.0

0.0909
0.0788
157.9
3ercent  Isokinetic
94.2
                                         90.6
                          98.8
                                  C-57

-------
       Table C-30.  SUMMARY OF ARSENIC TEST DATA — CALCINE DISCHARGE
                        DUCT, ASARCO-EL PASO SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
1/24/78
60
21.3
83.9

7,705
56

0.1
0.0
20.5
a



0.0049
0.0045
0.326

0.0050
0.0045
0.332
2
1/24/78
60
21.3
83.9

7,809
57

0.4
0.0
20.5
a



0.0014
0.0012
0.092

0.0014
0.0018
0.094
3
1/24/78
60
21.3
83.9

7,659
61

0.3
0.0
20.5
a



0.0034
0.0030
0.223

0.0034
0.0031
0.224
Average

60
21.3
83.9

7,724
58

0.3
0.0
20.5
a



o.oo:
0.002
0.21'

o.oo:
o.oo:
0.21;
Percent Isokinetic
93.4
100.7
94.0
                                   C-58

-------
       Table C-31.  SUMMARY OF PARTICIPATE TEST DATA - CALCINE DISCHARGE
                          DUCT,  ASARCC-EL PASO SMELTER          ui^HAKbh
	 	 	 _ —
Run No.

Date
.Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
02
S02
Emissions - Particulatp
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Tb/hr
Total catch
gr/dscf
gr/acf
Ib/hr
.
— •• 	
1
i
•
1/24/78
57
23.5
a

7,932
57

0.4
0.0
20.5
0.014



0.0512
0.0457
3.49

0.1861
0.1630
12.70



'"
1/25/78
60
37.8
a

7,737
88

0.3
0.0
20.5
0.005



0.0338
0.0305
2.23

0.1481
0.1336
10.11


3

1/25/78
60
37.8
a

7,394
96

1.0
0.0
20.5
0.001



0.0090
0.0080
0.57

0.3196
0.2861
20.23


Average


59
33.0
a

7,687
80

0.6
0.0
20.5
0.007



0.0313
0.0281
2.09

0.2179
0.1620
14.34
Percent Isokinetic
96.1
                                          103.6
                         100.6
                                  C-59

-------
         Table C-32.  SUMMARY OF ARSENIC TEST DATA -- MATTE TAPPING
                        DUCT, ASARCO-EL PASO SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
1/20/78
240
69.2
474.2

23,296
105

0.3
0.0
20.5
a



0.0029
0.0024
0.578

0.0029
0.0025
0.580
2
1/20/78
240
69.2
474.2

'26,367
98

0.0
0.0
20.5
a



0.0026
0.0022
0.598

0.0027
0.0022
0.600
3
1/25/78
396
39.1
176.0

27,418
73

0.0
0.0
20.5
a



0.0012
0.0010
0.288

0.0012
0.0010
0.288
Average

292
59.1
374.8

25,694
92

0.1
0.0
20.5
a



0.00
0.00
0.48

O.OC
O.OC
0.48
Percent Isokinetic
94.5
90.2
86.9
                                  C-60

-------
         Table  C-33.   SUMMARY  OF  PARTICULATE  TEST  DATA  -  MATTE TAPPING
                          DUCT, ASARCO-EL  PASO  SMELTER

Run No.

Date .
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
" Stream (vol. %):
Water
C02
n
U2
S02
Emissions - Particulatp
Probe, cyclone,
and filter catch
gr/dscf
. gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
i
1/25/78
389
39.1
a

26,871
73

0.8
0.0
20.5
0.006



0.0055
0.0047
1.07

0.0978
0.0836
19.12


1/26/78
360
67.6
a

27,370
82

0.0
0.0
20.5
0.009



0.0193
0.0161
3.77

0.1632
0.1370
32.14

3
1/26/78
360
67.6
a

26,802
82

0.0
0.0
20.5
0.020



0.0164
0.0100
3.09

0.0712
0,1366
32.27

Average

369
58.1 .
a

27,015
79

0.3
0 0
20.5
0.012



0.0134
0.0103
2.64

0.1441
0.1191
27.84
Percent Isokinetic
93.7
                                           96.0
                          91.9
                                 C-61

-------
     Table  C-34.  SUMMARY OF ARSENIu TEST DATA -- SPRAY CHAMBER/BAGHOUSE
                    INLET-WEST, ANACONDA-ANACONDA SMELTER
Run No,
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . X):
Water
C02
02
SO 2
Emissions - Paniculate
Probe, cyclone,
and filter catch
gr/dscf
. gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
4/20/77
120
97
2,173

76,274
517

11.5
4.0
16.4
a



0.4012
0.1551
262.2

0.4063
0.1571
265.6
2
4/21/77
120
92
1,490

74,002
511

12.2
4.2
18.2
a



0.3550
0.1383
225.1

0.3706
0.1444
235.0
3
4/22/77
120
92
1,914

77,402
466

9.7
0.2
19.8
a



0.3086
0.1309
204.7

0.3153
0.1337
209.1
Average

120
93.7
1,857

75,893
498

11.1
2.8
18.1
a



0.354
0.141
230.6

0.363
0.145
236.4
Percent Isokinetic
92.9
101.9
100.7
                                      C-62

-------
Table C-35.   SUMMARY OF ARSENIC TEST DATA -  SPRAY CHAHBER/BAGHOUSE
               INLET-EAST,  ANACONDA-ANACONDA  SMELTER
• — — 	 	 	 _ 	 	 — . 	
Run No.
— 	 	 • 	
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %) :
Water
C02
Q
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
•* . _
* • •— • i • —
1
— — — — • 	 -
4/20/77
120
97
2,173

80,193
535

12.9
0.8

17.4
a



0.4939
0,1846
339.4

0.4992
0.1866
343.1
2
—————— — __
4/21/77
120
92
1,490

86,350
523

12.8
0 5
V-* « J
19.5
a



0.3771
0.1442
279.0

0.3864
0.1477
285.9
3
— 	 • 	
4/22/77
120
92
1,914

86,889
475

10.3
On
.0
19.6
a



0.2559
0.1069
190.6

0.2690
0.1124
200.3
Average

120
93.7
1,857

84,477
511

12.0
0 4
U « *T
18.8
a



0,3725
0.1442
267.8

0.3818
0.1479
274.7
                      98.7
96.2
                                                98.5
                         C-63

-------
   Table C-36.  SUMMARY OF ARSENIC TEST DATA -- SPRAY CHAMBER/BAGHOUSE
                  INLET (TOTAL), ANACONDA-ANACONDA SMELTER*
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
°2
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
Percent Isokinetic
12 3 Average
4/20/77 4/21/77 4/22/77
120 120 120 120
97 92 92 93.7
2,173 1,490 1,914 1,857

156,467 160,352 164,291 160,370
526 517 471 505
12.2 12.5 10.0 11.6
2.4 2.2 0.1 1.5
16.9 18.9 19.7 18.5
a a a a


0.4588 0.3669 0.2807 0.3E
0.1702 0.1415 0.1182 0.1<
601.6 504.1 395.3 498.4
0.4539 0.3791 0.2908 0.3;
0.1722 0.1462 0.1224 0.1'
608.7 520.9 409.4 511.1
c c c
*These data are derived from Tables C-34 and C-35.
                                    C-64

-------
       Table C-37.   SUMMARY  OF  ARSENIC  TEST  DATA  -  SPRAY  CHAMBER/BAGHOUSE
                        OUTLET,  ANACONDA-ANACONDA SMELTER
	 ' 	 • 	 • 	 	 	
Run No.
	
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
n
U2
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
i
i
""" ™"™^ _« ••_*«•
4/20/77
128
97
2,173

153,594
210

19.1
3.6
17.0
a



0.0018
0.0009
2.41

0.0031
0.0016
4.02


4/21/77
128
92
1,490

156,349
215

19.3
4.5
17.5
a



0.0023
0.0012
3.07

0.0041
0.0021
5.53

3
4/22/77
128
92
1,914

164,134
214

17.7
3.5
16.9
a



0.0036
0.0019
5.00

0.0053
0.0028
7.45

Average

128
93.7
1,857

158,026
213

18.7
3 q
*J • ,7
17.1
a



0.0026
0.0013
3.52

0.0042
0.0015
5.71
Percent Isokinetic
100.4
                                          102.2
                           98.7
                                 C-65

-------
    Table  C-38.  SUMMARY OF PARTICIPATE TEST DATA -- SPRAY CHAMBER/BAGHOUSE
                          INLET-WEST, ANACONDA-ANACONDA SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %) :
Water
ecu
wV 2
02
S02
Emissions - Particulate
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
4/25/78
120
86
a

77,031
535

13.2
2.3
16.2
a


7.25
2.74
4,786
7.36
2.78
4,858
2
4/25/78
120
86
a

80,363
546

12.3
1.4
16.9
a


6.20
2.34
4,267
6.25
2.36
4,306
3
4/26/77
120
70
a

75,458
573

11.2
1.0
19.4
a


6.40
2.37
4,139
6.43
2.38
4,161
Average

120
80.7
a

77,617
551

12.2
1.6
17.5
a


6.61
2.48
4,397
6.68
2.51
4,442
Percent Isokinetic
97.4
98.5
                                                       100.9
                                     C-66

-------
      Table C-39.  SUMMARY OF PARTICULATE TEST DATA - SPRAY CHAMBER/BAGHOUSE
                             INLET-EAST, ANACONDA-ANACONDA SMELTER
1 — -' 	 	
Run No.
	 — 	 	 	 .
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
CO,
n *
U2
S02
Emissions - Particulate
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
4/25/77
120
86
a

85,140
541

11.4
0.0
20.1
a



5.72
2.20
4,170

5.83
2.24
4,254
2
4/25/77
120
86
a

81,352 -
555

12.2
0.0
19.4
a



5.67
2.13
3,952

5.76
2.16
4,013
3
4/26/77
120
70
a

85,669
577

13.5
0.0
19.6
a



5.93
2.13
4,352

6.05
2.18
4,442
Average

120
80.7
a

84,054
558

12.4
0.0
19.7
a



5.78
2.15
4,162

5.88
2.19
4,240
Percent Isokinetic
97.9
                                          96.1
                          98/4
                                 C-67

-------
    Table C-40.   SUMMARY  OF  PARTIO'LATE TEST DATA -- SPRAY CHAMBER/BAGHOUSE
                          INLET (TOTAL), ANACONDA-ANACONDA SMELTER*
Run No.
Date
Test Duration - mln.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
C02
^ w t
02
SO 2
Emissions - Particulate
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
Percent Isokinetic
1
4/25/77
120
86
a

162,171
538

12.3
1.2
18.2
a



6.45
2.46
8,956

6.56
2.50
9,112
c
2
4/25/77
120
86
a

161,715
551

12.3
0.7
18.2
a



5,93
2.23
8,219

6.00
2.26
8,319
c
3
4/26/77
120
70
a

161,127
575

12.3
0.5
19.5
a



6.15
2.24
8,491

6.23
2.27
8,603
c
Average

120
80,7
a

161,671
555

12.3
0.8
18.7
a



6.18
2.31
8,559

6.26
2.34
8,682

*These data are derived from Tables C-38 and C-39.
                                  C-68

-------
       Table  C-41.   SUMMARY OF PARTICULATE TEST'DATA - SPRAY CHAMBER/BAGHOUSE
                               OUTLET. ANACONDA-ANACONDA SMELTER
Run No.
	 ' 	 — 	
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
2
S02
Emissions - Particulate
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
4/25/77
128
86
a

170,466
217

16.4
4.5
18.5
a



0.0220
0.0119
32.1

0.1387
0.0754
202.6
2
4/25/77
128
86
a

158,252
218

19.4
4.8
18.2
a



0.0162
0.0085
22.0

0.0667
0.0349
90.4
3
4/26/77
128
70
a

165,400
213

19.7
5.2
17.5
a



0.0228
0.0115
32.3

0.0288
0.0146
40.8
Average

128
80.7
a

164,706
216

18.5
4 8
18.1
a



0.0204
0.0107
28.9

0.0789
0.0421
112.5
Percent Isokinetic
96.7
                                          99.4
                          97.7
                                 C-69

-------
Table C-42.  SUMMARY OF ARSENIC TEST DATA — REVERBERATORY ESP
                   INLET, PHELPS DODGE-AJO SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
Percent Isokinetic
1
7/13/76
120
a
59.3
58,814
622
18.0
a
a
a
0.1076
0.0527
54.3
0.1172
0.0574
59.1
154
2
7/14/76
120
a
72.8
59,583
602
18.6
a
a
a
0.1326
0.0662
67.8
0.1423
0.0710
72.7
152
3
7/14/76
120
a
75.4
60,150
639
16.0
a
a
a
0.1394
0.0673
71.9
0.1459
0.0704
75.2
147
Average

120
a
69.2
59,516
621
17.5
a
a
a
0.126
0.062
64.7
0.13E
0.066
69.0

                                   C-70

-------
Table *C-43.   SUMMARY OF ARSENIC TEST DATA - REVERBERATOR? ESP
                   OUTLET, PHELPS OODGE-AJO SMELTER
	 — 	 _
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
CO 2
°2
S02
Emissions - Arsenir
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
t
1
	 	 _ — ...
7/13/76
120
a
53.7

68,030
595

15.3
a
a



0.0603
, 0.0303
35.2

0.0919
0.0462
53.6
2
1 	 	 	
7/14/76
120
a
44.8

66,275
610

15.4
a
a
a



0.0376
0.0186
21.4

0.0786
0.0389
44.6
3
^ 	 	
^^""•"•"••^^^•••^—••w"
7/14/76
120
a
51.3

68,738
580

13.5
a
a
a



0.0302
0.0154
17.8

0.0868
0.0442
51.1
Average
	 • 	 	 	
120
a
49.9

67,681
595

14.7
a
a
a



0.0427
0.0214
24.8

0.0858
0.0431
49.8
                     145
147
                                                136
                         C-71

-------
Table C-44.  SUMMARY OF ARSENIC TEST DATA - CONVERTER ESP
                INLET NO. 1, PHELFS DODGE-AJO SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
Percent Isokinetic
1 2*
6/13/78
144
17.7
38.9
28,075
379
0.0
0.0
20.0
3.83
0.000038
0.000032
0.0091
0.000100
0.000083
0.0241
126.4
3
6/15/78
144
17.7
95.6
26,638
405
4.0
0.0
20.0
4.18
0.000010
0.000008
0.0022
0.000049
0.000042
0.0112
95.2
Average

144
17.7
67.2
27,358
392
2.0
0.0
20.0
4.01
0.00002
0.00002
0.0056
0.00007
0.00006
0.0062

                                       C-72

-------
       Table C-45.  SUMMARY OF ARSENIC TEST DATA - CONVERTER  ESP
                      INLET NO. 2, PHEI.PS DODGE-AJO SMELTER
	 — 	 	 — — 	
Run No.

Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
n
U2
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
i
i
—— — — — — _ __ ___ _ _
6/13/78
144
17.7
38.9

34,282
389

0.3
0.0
20.0
2.59



0.000003
0.000002
0.0095

0.000003
0.000003
0.0097


"
6/14/78
144
19.7
63.0

28,312
358

0.8
0.0
20.0
3.03



0.000010
0.000009
0.0026

0.000013
0.000011
0.0031

3

6/15/78
144
17.7
95.6

29,265
404

0.0
0.0
20.0
7.60



0.000005
0.000004
0.0011

0.000013
0.000011
0.0029
. 	 . 	
Average

144
18. ,3
65.8

30,621
384

0.4
0 0
V * V
20.0
4.41



0.000006
0.000005
0.0044

0.000010
0.000008
0.0053
Percent Isokinetic
123.2
                                           92.1
                           99.0
                                 C-73

-------
      Table C-46.   SUMMARY OF  ARSENIC  TFST DATA — CONVERTER ESP
                OUTLET  (ACID PLANT  INLET), PHELPS DODGE-AJO SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
C02
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
6/13/78
106
17.7
38.9
41,016 39
374
3.0
0.0
20.0
4.98
0.000033
0.000028
0.0117
0.000035
0.000030
0.0123
2
6/14/78
111
19.7
63.0
,021
360
2.2
0.0
20.0
2.83
0.000010
0.000008
0.0033
0.000011
0.000010
0.0037
3
6/15/78
109
17.7
95.6
29,692
342
3.7
0.0
20.0
3.99
0.000016
0.000014
0.0042
0.000040
0.000035
0.0104
Average

109
18.3
65.8
36,578
359
3.0
0.0
20.0
3.93
0.00002
0.00001
0.0064
0.00002
0.00002
0.0088
Percent Isokinetic
101.3
102.1
                                                       106.8
                                          C-74

-------
       Table C-47.   SUMMARY OF ARSENIC  TEST DATA - ACID PLANT
                        OUTLET,  PHELPS  DODGE-AJO SMELTER
Run No.

Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water

C02
n
U2
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
— • 	 	 	 _ —
i
i
~^~i"'~— "—"•—•— •————^—^
6/13/78
108
17.7
38.9

47,556 43
140

1 ?
i . C
0.0
20.0
0.24



0.0000020
0.0000017
0.0009

0.0000030
0.0000026
0.0013


6/14/78
108
19.7
63.0

,862 36
164

Or
. 5
0.0
20.0
0.12



0.0000026
0.0000020
0.0011

0.0000048
0.0000039
0.0015

3
6/15/78
108
17.7
95.6

,016
128


o n
V • U
0.0
20.0
0.08



0.0000109
0.0000091
0.0033

0.0000120
0.0000021
0.0040

Average

108
18.3
65.8

42,478
144


Or-
.6
0 0
u • u
20.0
0.15



0.0000052
0.0000043
0.0018

0.0000068
0.0000052
0.0022
Percent Isokinetic
100.6
                                           97,4
                           97.0
                                 C-75

-------
   Table C-48.  SUMMARY OF ARSENIC TEST  DATA  - MATTE  TAPPING  HOOD
                    OUTLET, PHELPS DODGE-AJO  SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
5/10/78
192
43.9
776.7
63,758 72
111
0.9
0.0
20.0
0.02
0.00045
0.00042
0.248
0.00066
0.00062
0.365
2
5/11/78
120
40.2
741.3
,351 69
111
0.9
0.0
20.0
0.04
0.00076
0.00070
0.472
0.00082
0.00075
0.508
3
5/12/78
120
48.1
882.5
,333
120
1.2
0.0
20.0
0.04
0.00051
0.00046
0.300
0.00057
0.00052
0.344
Average

144
44.1
800.1
69,056 '
114
1.0
0.0
20.0
0.03
0.00
0.00
0.34
Q.OO
0.00
0.40
Percent Isokinetic
104.2
95.8
                                                        97.2
                                        C-76

-------
    Table C-49.  SUMMARY OF  PARTICULATE  TEST  DATA  -  MATTE TAPPING
                        OUTLET,  PHELP^ DODGE-AJO SMELTER
  Date

  Test Duration -  min.

  Charge  Rate  - ton/hr

  Arsenic  Rate  - Ib/hr

  Stack Effluent

    Flow  rate  (dscfm)

    Temperature (°F)

    Stream (vol. %):
      Water
      C02
      02
      S02

 Emissions -  Participate

    Probe,  cyclone,
    and filter  catch
      gr/dscf
      gr/acf
      Ib/hr

Total  catch
      gr/dscf
     gr/acf
      Ib/hr

Percent Isokinetic
5/10/78
192
43.9
a
68,930
110
0.7
0.0
20.0
0.02
0.0133
0.0123
7.89
0.0160
0.0148
9.51
5/11/78
120
40.2
a
73,362
112
0.3
0.0
20.0
0.04
0.0226
0.0207
14.31
0.0466
0.0427
29.49

156
42.0
a
71,140
111
0.5
0.0
20.0
0.03
0.0179
0.0165
11.10
0.0313
0.0288
19.50
                                    108.7
100.1
                                  C-77

-------
         Table C-50.
SUMMARY OF ARSENIC TEST DATA -- CONVERTER SECONDARY
    HOOD OUTLET, PHELPS DODGE-AJO SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
C02
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
5/10/78
244
27.3
2.9
85,659
142
0.5
0.0
20.0
0.23
0.00247
0.00224
1.813
0,00251
0.00227
1.837
2
5/11/78
120
10.6
1.1
87,444
162
1.2
0.0
20.0
0.34
0.00245
0.00225
1.834
0.00246
0.00225
1.841
3
5/12/78
120
28.4
3.0
85,957
158
0.6
0.0
20.0
0.37
0.00131
0.00121
0.961
0.00134
0.00125
0.989
Average

161
22.1
2.3
86,353
154
0.8
0.0
20.0
0.31
0.002C
0.001$
1.536
0.002:
0.001<
1.556
Percent Isokinetic
       101.1
101.1
104.4
                                          C-78

-------
    Table C-51.  SUMMARY OF PARTICULATE TEST DATA -- CONVERTER SECONDARY
                    HOOD OUTLET, PHELPS DODGE-AJO SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
02
S02
Emissions - Particulate
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
5/10/78
244
27.3
a
86,369
143
0.5
0.0
20.0
0.23
0.0756
0.0688
55.93
0.1016
0.0925
75.16
2
5/11/78
120
10.6
a
87,708
163
0.7
0.0
20.0
0.34
0.0910
0.0828
68.37
0.1490
0.1355
111.9
3
5/12/78
120
28.4
a
85,698
162
0.5
0.0
20.0
0.38
0.0793
0.0835
58.23
0.1105
0.1025
81.13
Average

161
22.1
a
' 86,591
156
0.6
0.0
20.0
0.32
0.0820
0.0749
60.85
0.1204
0.1101
89.41
Percent Isokinetic
101.1
101.1
104.4
                                   C-79

-------
     Table C-52.  SUMMARY OF ARSENIC TEST DATA — CONVERTER SECONDARY
                      HOOD OUTLET, PHELPS DODGE-HIDALGO SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %} :
Water
C02
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
7/25/78
300
a
a

69,076
217

1.5
0.2
20.2
0.38



0.00017
0.00014
0.1026

0.00050
0.00038
0.2947
2
7/26/78
240
a
a

83,346
208

2.2
0.2
20.2
0.48



0.00010
0.00008
0.0718

0.00011
0.00009
0.0778
3
7/26/78
240
a
a

56,063
216

2.2
0.2
20.2
1.10



0.00017
0.00013
0.0807

0.00020
0.00016
0.0965 •
Average

260
a
a

69,495
213

2.0
0.2
20.2
0.65



0.00
0.00
0.08

0.00
0.00
0.15
Percent Isokinetic
85.3
98.6
99.1
                                      C-80

-------
      Table C-53.  SUMMARY OF ARSENIC TEST DATA -- CALCINE/ROASTER FUGITIVES
                        BAGHOUSE INLET, PHELPS DODGE-DOUGLAS SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (val. 35):
Water
C02
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
5/3/78
55
a
a

29,697 30,
76
0.5
0.0
20.0
0.13

0.0000003
0.0000002
0.0001

0.000026
0.000023
0.0067
2
5/4/78
49
a
0.38

359 26,
72
0.8
0.0
20.0
0.21

0.0000367
0.0000314
0.0096

0.000098
0.000083
0.0254
3
5/4/78
38
a
7.96

739
65
2.0
0.0
20.0
0.18

0.000193
0.000155
0.0424

0.000281
0.000234
0.0645
Average

48
a
5.32

28,932
71
1.1
0.0
20.0
0.17

0.000062
0.000062
0.0170

0.000114
0.000113
0.0322
Percent Isokinetic
97.4
                                           91.7
                         110.0
                                 C-81

-------
    Table C-54.  SUMMARY OF ARSENIC TEST DATA — CALCINE/ROASTER FUGITIVES
                      BAGHOUSE OUTLET, PHELPS DODGE-DOUGLAS SMELTER
	 ___ — _ 	 , — . 	
Run No.
Date-
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %) :
Water
C02
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
5/3/78
65
a
7.11
31,539 32
73
1.0
0.0
20.0
0.08
0.000011
0.000010
0.0030
0.000105
0.000090
0.0283
2
5/4/78
42
a
a
,296 31
65
0.9
0.0
20.0
0.19
0.000030
0.000026
0.0082
0.000138
0.000120
0.0381
3
5/4/78
40
a
7.96
,781
79
1.4
0.0
20.0
0.15
0.000069
0.000058
0.0188
0.000095
0.000079
0.0259
Average

49
a
5.82
31,872
73
1.1
0.0
20.0
0.14
0.00
0.00
0.01
O.OC
O.OC
0.02
Percent Isokinetic
96.5
95.0
90.0
                                      C-82

-------
  Table C-55.   SUMMARY  OF PARTICULATE TEST DATA --  CALCINE/ROASTER FUGITIVES
                 BAGHOUSE INLET,  PHELPS  DODGE-DOUGLAS  SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
2
SO 2
Emissions - Particulate
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
5/3/78
56
a
a

30,294
69

1.2
0.0
20.0
0.13



1.766
1.523
458.3

1.822
1.571
472.9
2
5/4/78
47
a
a

29,153
74

0.0
0.0
20.0
0.21



3.067
•2.615
765.8

3.166
2.699
790.5
3
5/5/78
36
a
a

29,036
65

0.6
0.0
20.0
0.18



2.692
2.282
669.6

2.925
2.480
727.5
Average

46
a
a

29,380
69

0.6
0.0
20.0
0.17



2.508
2.107
631.2

2.638
2.250
663.6
Percent Isokinetic
96.8
95.0
                                                       105.4
                                 C-83

-------
   Table C-56.   SUMMARY OF PARTICULAR TEST DATA -- CALCINE/ROASTER
        FUGITIVES  BAGHOUSE OUTLET,  PHELPS DODGE-OOUGLAS SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
CO 2
02
S02
Emissions - Particulate
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
5/3/78
65
40.1
a
29,018
73
1.3
0.0
20.0
0.09
0.0031
0.0026
0.771
0.0447
0.0387
11.11
2
5/4/78
42
39.5
a
30,985
65
0.6
0.0
20.0
0.19
0.0150
0.0131
3.98
0.1927
0.1655
51.16
Average

54
39.8
a
30,002
69
0.95
0.0
20.0
0.14
0.0091
0.0078
2.37
0.1187
0.1021
31.79
Percent Isokinetic
93.6
87.7
                                      C-84

-------
    Table C-57,
SUMMARY OF ARSENIC TEST DATA - CONCENTRATE DRYER SCRUBBER
        OUTLET,  KENNECOTT-MAGNA SMELTER

Run No.

Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
i
i
	 ' 	
11/14/78
90
a
a

43,489
166

18.1
2
20.0
0.06



0.00003
0.00002
0.0120

0.00003
0.00003
0.0143


•— . .,„ ,„ , m m
11/14/78
90
a
a

38,677
129

18.2
a
20.0
0.07



0.00046
0.00041
0.1531

0.00047
0.00041
0.1551

3

11/14/78
90
a
a

47,225
118

15.6
a
20.0
0.07



0.00111
• 0.00084
0.3739

0.00099
0.00090
0.4029

Average


90
a
a

43,130
138

17.3
a
20.0
0.07



0.00047
0.00042
0.1797

0.00050
0.00045
0.1908
Percent Isokinetic
            102.8
                                           98.4
                                       96.5
                                 C-85

-------
   Table  C-58.   SUMMARY OF ARSENIC ItST  DATA  —  ACID  PLANT  INLET,
                          KENNECOTT-MAGNA  SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
CO 2
°2
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
11/6/78
124
a
a

47,978
421

5.0
0.0
20.0
3.9



0.0034
0.0029
1.397

0.0055
0.0047
2.278
2
11/7/78
119
a
a

44,725
476

3.0
0.0
20.0
2.4



0.0034
0.0028
1.294

0.0034
0.0029
1.302
3
11/8/78
120
a
a

44,643
409

4.0
0.0
20.0
3.0



0.0020
0.0018
0.784

0.0021
0.0018
0.802
Average

121
a
a

45,782
436

4.0
0.0
20.0
3.1



o.oo;
o.oo;
1.15,

o.oo,
o.oo,
1.46
Percent Isokinetic
97.9
94.1
106.9
                                      C-86

-------
   Table  C-59.   SUMMARY  OF  ARSENIC  TEST  DATA  --  MATTE  TAPPING DUCT,
                           KENNECOTT-MAGNA  SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
C02
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
11/1/78
70
a
a

48,968
126

1.4
0.0
20.0
0.09



0.00030
0.00026
0.1255

0.00036
0.00030
0.1505
2
11/2/78
60
a
a

48,645
111

1.0
0.0
20.0
0.10



0.00052
0.00045
0.2185

0.00087
0.00075
0.3641
3
11/3/78
66
a
a

43,868
119

0.0
0.0
20.0
0.12



0.00115
0.00099
0.4341

0.00136
0.00117
0.5128
Average

65
a
a

47,162
119

0.8
0.0
20.0
0.10



0.00064
0.00056
0.2594

0.00085
0.00074
0.3425
Percent Isokinetic
107.3
97.1
                                                       103.5
                                 C-87

-------
  Table C-60.  SUMMARY OF ARSENIC TEST DATA — SLAG TAPPING DUCT,
                          KENNECOTT-MAGNA SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
11/1/78
60
a
a

43,308
73

0.3
0.0
20.0
0.003



0.00030
0.00026
0.1119

0.00036
0.00030
0.1327
2
11/2/78
120
a
a

40,541
91

0.9
0.0
20.0
0.008



0.00013
0.00011
0.0443

0.00033
0.00029
0.1152
3
11/3/78
120
a
a

38,914
71

1.0
0.0
20.0
0.004



0.00005
0.00004
0.0172

0.00006
0.00005
0.0217
Average

100
a
a

40,921
78

0.7
0.0
20.0
0.005



O.OOC
0.000
0.057

O.OOC
O.OOC
0.08?
Percent Isokinetic
89.9
93.6
97.0
                                      C-88

-------
  table C-61.   SUMMARY  OF ARSENIC TEST DATA -- CONVERTER FUGITIVES (FULL
                       CYCLE),  KENNFCOTT-MAGNA SMELTER

Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
11/6/78
188
a
a

94,684
105

0.0
0.0
20.0
0.09



0.00028
0.00024
0.2262

0.00034
0.00029
0.2762
2
11/8/78
181
a
a

90,187
103

0.8
0.0
20.0
0.14



0.00013
0,00011
0.1011

0.00030
0.00026
0.2364
3
11/9/78
182
a
a

92,967
61

1.0
0.0
20.0
0.33



0.00044
0.00039
0.3517

0.00056
0.00050
0.4469
Average

184
a
a

92,613
90

0.6
0.0
20.0
0.19



0.00028
0.00025
0.2263

0.00040
0.00035
0.3198
Percent Isokinetic
100.8
103.3
93.8
                                  C-89

-------
  Table  C-62.   SUMMARY OF ARSENIC  TEC'i  DATA --  ROLLOUT CONVERTER FUGITIVES,
                                KENNECOTT-MAGNA  SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol. %):
Water
C02
02
S02
Emissions - Particulate
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
11/6/78
88
a
a

83,303
100

1.0
a
20.0
0.03



0.00019
0.00016
0.1342

0.00019
0.00017
0.1376
2
11/8/78
65
a
a

81,777
106

1.0
a
20.0
0.05



0.00052
0.00044
0.3781

0.00057
0.00048
0.3971
Average

77
a
a

82,540
103

1.0
a
20.0
0.04



0.00035
0.00030
0.2561

0.00035
0.00031
0.2674
Percent Isokinetic
103.2
99.8
                                        C-90

-------
Table C-63.   SUMMARY OF ARSENIC TEST DATA — VENTURI SCRUBBER
                INLET, KENNECOTT-HAYDEN SMELTER
T NO.
:e
it Duration - min.
rge Rate - ton/hr
em'c Rate - Ib/hr
ck Effluent
Flow rate (dscfm) •
Temperature (°F)
stream (vol. %) :
Water
CO 2
02
S02
sions - Arsenic
robe, cyclone,
nd filter catch
gr/dscf
gr/acf
Ib/hr
)tal catch
gr/dscf
gr/acf
Ib/hr
mt Isokinetic
nrrp<;nrmHinn i-ac-t- *
1
12/10/76
110
61
1.88
16,971
636
26.2
a
4.2
12.4
0.0116
0.0044
1.69
0.0129
0.0049
1.88
146
2
12/11/76
135
64
1.63
16,847
623
22.4
a
4.2
12.4
0.0108
0.0035
1.56
0.0113
0.0045
1.63
157
3
12/13/76
85
63
1.65
19,323
615
32.3
a
4.2
12.4
0.0097
0.0036
1.60
0.0100
0.0037
1.65
135
4*
12/13/76
75
64.5
1.23
19,011
621
27.5
a
4.2
12.4
0.0072
0.0027
1.17
0.0076
0.0029
1.23
126
Average

101
63.2
1.60
18,038
624
27.1
a
4.2
12.4
0.0098
0.0036
1.50
0.0104
0.0037
1.60

                          C-91

-------
   Table C-64.  SUMMARY OF ARSENIC iEST DATA — VENTURI SCRUBBER OUTLET,
                          KENNECOTT-HAYDEN SMELTER
Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %):
Water
C02
02
S02
Emissions - Arsenic
Prob*, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
12/10/76
145
61
0.04

15,493
114

9.8
a
5.4
11.4



0.00007
0.00006
0.010

0.00028
0.00024
0.037
2
12/11/76
145
64
0.02

18,918
111

9.1
a
5.4
11.4



0.00006
0.00005
0.010

0.00012
0.00010
0.019
3
12/13/76
140
63
0.02

18,017
83

3.8
a
5.4
11.4



0.00004
0.00004
0.006

0.00014
0.00013
0.022
Average

143
62.7
0,03

17,476
103

7.6
a
5.4
11.4



0.00
O.OO1
0.00

0.00
0.00
0.02
Percent Isokinetic
112
115
101
                                       C-92

-------
        Table C-65.  SUMMARY OF ARSENIC TEST  DATA  —ACID  PLANT  OUTLET,
                             KENNECOTT-HAYDEN  SMELTER

Run No.
Date
Test Duration - min.
Charge Rate - ton/hr
Arsenic Rate - Ib/hr
Stack Effluent
Flow rate (dscfm)
Temperature (°F)
Stream (vol . %}:
Water
C02
02
S02
Emissions - Arsenic
Probe, cyclone,
and filter catch
gr/dscf
gr/acf
Ib/hr
Total catch
gr/dscf
gr/acf
Ib/hr
1
12/10/76
370
61
0.47
74,746
155
0.0
a
7.7
0.0
0.00014
0.00012
0.093
0.0007
0.0006
0.448
2
12/11/76
265
64
0.79
60,114
158
0.0
a
7.7
0.0
0.00038
0.00033
0.203
0.0015
0.0013
0.773
3
12/13/76
310
63
1.46
77,798
175
0.0
a
7.7
0.0
0.00020
0.00016
0.130
0.0022
0.0018
1.47
Average

315
62.7
0.92
70,886
163
0.0
a
7.7
0.0
0.00024
0.00020
0.142
0.0015
0.0012
0.911
Percent Isokinetic
107
108
95
                                  C-93

-------
Table C-66.   VISIBLE EMISSIONS OBSERVATION DATA, EPA METHOD 22-
   ROASTER CALCINE DISCHARGE INTO LARRY CARS, ASARCO-TACOMA
Run
No.
^^H^MI^HH
1
2
3
4
5
6
7
3
9
10
11
12
1.3
"^^•••••w
Date
— — •— — —
6/24
6/24
6/24
6/25
6/25
6/25
6/25
6/25
6/26
6/26
6/26
6/26
6/26
1
Observe P
— — _______ __
Duration of
operation,
mi n: sec
	
1:20
2:40
1:20
1:23
1:58
1:42
1:12
1:20
2:50
1:48
2:30
1:42
3:04
——————
	
1
^••^'•^'^^^•^MMMM
I time
emissions
observed
0
0
0
0
0
0
0
0
0
0
n
w
o
n
w
——————
— ^— — _
"
Observer
"^ ^^^MMM^MMM
Duration of
operation,
rain: sec
•
1:15
2:40
1:20
1:23
1:52
1:42
1:13
1:20
2:49
1:48



-
~^— •— ^— — •—
""••—»—••—••«
2
•— ~- 1.
i time
anission
observed
— — ^^— — _
0
0
0
0
0
0
0
0
0
0



•^— ^— — _
-'
— — — — —
Mean
duration of
operation,
mi n: sec
^-^— — — — ^».
1:18
* • * «^
2:40
1:20
1-23
« * «»M
1:55
1:42
1:13
1-20
1 • teW
2:50
1-48
^ • ^rw
2:30
1:42
3:04
•^MH^HMBM.^,^..
^"— — »— «-^
Mean
I time
eniissior
observe:
••^•••••.^HHMM.
0
0
0
0
Q
0

0
0
0
                                  Average
1:54
                               C-94

-------
Table C-67.   VISIBLE EMISSIONS OBSERVATION DATA,  EPA METHOD 22-
        MATTE TAP PORT AND MATTE  LAUNDER,  ASARCO-TACOMA
^~— — — ^.
Runa Date
i
•' ••
1 6/24
2 6/24
3 6/24
4 6/24
K
5° 6/24
5 6/25

' 6/25
6/25
6/25
6/25
6/25
6/25
6/25
6/25
6/25
6/25
6/25

6/25
•— — — _ _j_

Observer 1
Duration of
operation,
win: sec
> — — >_____^_
6:24
6:00
4:51
6:05

5:28

5:22
5:36
5:08
6:02
5:12

4:50
5:23
5:17

5:13

5:58
•••
X time
emissions
observed
0
0
0
0

o
>rf
0
0
0
0
0

0
0
0

a

o
	
Observer 2
Duration of
operation,
(nin: sec
~~"^»-"— «—«•••.••
6:36
6:00
4:55
6:10



5:22
5:36
5:10
5:33
5:13
6:37 .
4:53
5:22
5:18




— — • — — — _
2 time
emissions
observed
• -^ —
1
0
3
0



Q
0
0
0
0
0
0
0
0




— . .
1 	 " 	
Duration
operation
mi n: sec
	
6:30
6:00
4:53
6:08


2:58
5:22
5:36
5:09
5:48
5:13
6:37
4:52
5:23
5: 18


5: 13

5:58
Moan
• 
-------
     Table  C-68.   VISIBLE  EMISSIONS OBSERVATION  DATA,  EPA  METHOD  22—

               MATTE  DISCHARGE  INTO LADLE, ASARCO-TACOMA
Run3
1
2
3
4
5
6b
7
a
9
10
n
12
13
14
15
16
!sb
Date
6/24
6/24
6/24
6/24
6/24
5/25
6/25
5/25
6/25
6/25
6/25
6/25
6/25
6/25
6/25
6/25
6/25
6/25
Observer 1
Duration of
operation,
min:sec
6.30
5:49
4:53
6:12

5:09
5:21
5:02
4:29
5:12
6:16
4:43
5:13
5:15
5:41

% time
emissions
observed
0
0
0
0

0
0
0
0
0
0
0
0
0
0

Observer 2

Duration of % time
operation, emissions
mi n: sec observed

5:40
5:01
6:10
6:31
5:02
5:28"
5:03
4:32
5:13

4:45
5:15
5:09
5:50


0
0
Q
0
0
0
0
0
0

0
0
0
0

Duration
operation
mi n: sec
6:.30
5:45
4:57
6:11
6:31
5:06
5:25
5:03
4:31
5:13
6:16
4:44
5:14
5:12
5:46

Mean
of % time
, emissions
observed
0
0
0
0
0
0
0
0
0
0
0.
0
0
0
0

                                  Average
5:30
Method 22 data for corresponding runs at the matte tap and launder
are presented in Table C-67.

Observations were made only at the matte tap and launder; see Table C-67
                                      C-96

-------
                              SLAG TAP PORT  AND  SLAG  LAUNDER, ASARCO-TACOMA

Rund

1
2
3
4C
5
6C
7
8
9
10
A
11



Date

6/24
6/24
6/24
6/24
6/25
6/25
6/26
6/26
6/26
6/26


Observer

Duration
of operation,
mirr.sec
12:25b
22:00
14:07
14:10
16:44
17:26
16:14
13:45
15:45
14:29




1

% Time
emissions
observed
98b
15
35b
13
11
2
1
0.3
0
0




Observer 2

Duration % Time
of operation, emissions
min:sec observed
12:26b 99b
21:36 0
13:52b 97b









-_ 	 — 	 '
Mean
Duration
of operation,
mi n: sec
12:26
21:43
14:07
14:10
16: 44
17:26
16:41
13:45
15:45
14:29


Average 15:40
Std. dev.


Mean 1 Time
emissions
observed

8

13
11
2
1
0
0
0


4
11
o
I
       aMethod  22  data  for corresponding runs at the slag skim discharge point appear in Table C-71.


       Observations were made at the entire slag tap process line including the slag tap port, slag

        launder, and slag discharge into ladle; and therefore not included in computing the mean of


        observations.


       cMethod  9 data  for corresponding runs appear in Table C-70.


       dNo data obtained by  Method 22.

-------
 Table C-70.   VISIBLE EMISSIONS OBSERVATION DATA,  EPA METHOD 9-
           SLAG TAP AND SLAG LAUNDER,  ASARCO-TACOMA
Run
1
2
Average
Maximum
•
Date
6/25
6/25
^»^»— •— ••— a™.
Duration
of operation,
min.
14.75
18
.
16.38
Mean
opacity,
«
a
1.3
10.3
•— — — —— — — — —
6
Maximum
opacity,
V
a
10
30
— — — — — — ^_____
30
'Emission  data  were  taken  during  entire slag tapping operation.
                                  C-98

-------
o

CO
               Table C-71.   VISIBLE  EMISSIONS OBSERVATION  DATA,  EPA  METHOD  22-SLAG  TAPPING  AT
                                  SLAG  DISCHARGE  INTO  POTS, ASARCO-TACOMA

Run'bDate
1 6/24
2 6/24
3 6/24
4 6/24
5 6/25
6 6/25
7 6/26
8 6/26
9 6/26
10 6/26
11 6/26
Observer
1
Duration X time
of operation, emissions
m1n:sec observed
12:46
21:09
14:06
14:05
16:34
17:29
15:54
13:48
15:48
14:11
14:45

i
97
93
97
82
91
94
90
86
77
72
82


Observer 2
Duration % time
of operation, emissions
min:sec observed
12:26 73
21:43 99
13:52 95








Mean
duration
of operation,
m1n:sec
12:36
21:26
13:59
14:05
16:34
17:29
15:54
13:48
15:48
14:11
14:45
Average 15:31
Std. dev.
Mean X time
emissions
observed
85
96
96
82
91
94
90
86
77
72
82
86
8
      Visible emissions observation data by EPA Method 9 for corresponding runs are presented  In Table  C-72.
      IJ— *-.•!_ "1_ _   •   •     i
                                             corresponding runs for the slag tap port and launder are

-------
    Table C-72.   VISIBLE EMISSIONS OBSERVATION DATA, EPA METHOD 9-
        SLAG TAPPING AT SLAG DISCHARGE INTO POTS, ASARCO-TACOMA
Runb


1
2
3
4
5
6
7
8
9
10
11
Average
Maximum

Date

6/24
6/24
6/24
6/25
6/25
6/25
6/26
6/26
6/26
6/26
6/26


Duration
of operation,
mm.
c
c
c
13.75
16.75d
11.75d
15
15
13
15

14.32

Mean
opacity,
%



22.7
11.3
16
14.8
10.3
5,5
3.7

12

Maximum
opacity,
V
a



50
30
35
40
20
10
10


50
a
 Emission data were taken during entire  slag  tapping operation.
 Method 22 data for corresponding  runs appear in  Table C-71.
 No data were obtained by Method 9.
 Reading started after filling of  first  sla'g  pot.
                                     C-100

-------
o

o
                           Table C-73.   VISIBLE EMISSIONS OBSERVATION DATA,  EPA METHOD  22-

                            CONVERTER SLAG RETURN TO  REVERBERATORY  FURNACE,  ASARCO-TACOMA
Run*

1
2
3
4
5
6
7
8
9
10

11
12



Oate

6/24
6/24
6/24
6/25
6/25
6/25
6/25b
6/26
6/26
6/26

6/26
6/26

Observer 1
Duration of
operation,
ml n: sec
1:04
0:47
0:54
0:55


1:04
1:00
1:15

0:55


X time
emissions
observed
—
100
97
100
100


66
85
83

82


— — 	 — 	 . 	 _
— — — • 	 _ 	 . 	
Observer 2
Duration of X time
operation. emissions
ml n: sec observed
1:05 89
0:47 96
0: 53 100
1:03 100





0:41 93



	 — — 	 	 __
Observer 3
Duration of X time
operation, emissions
m1n:sec observed
0:58 100
0:46 100
0:55 100

0:52 100






Mean
duration of
operation.
n In: sec
1:04
0:46
0:53
0:55
1:03
0:52

1:04
1:00
r 15
0:48
Average 0;5B
Std. dev.


Mean
X time
emissions
observed
96
98
100
100
100
100

66
85
Al
88
92
11
                  ^Visible emissions observation data by EPA Method 9 for corresponding runs are'presented In Table C-74.
                  Uf\ sta t a *-vKl-^4*-,^.J t,., u_ 4.1	_i *\f\
                  No data obtained by Method 22.

-------
  Table C-74.   VISIBLE EMISSIONS OBSERVATION  DATA, EPA METHOD  9—

   CONVERTER SLAG RETURN  TO REVERBERATORY FURNACE, ASARCO-TACOMA
Run
1
2
3
4
5
6
7
8
9
10
11
12
Date
6/24
6/24
6/24
6/25
6/25
6/25
6/25
6/26
6/26
6/26
6/26
6/26
Observer 1
Duration of
operation,
win.
a
a
a
1.00
1.25

0.75
1.25
1.25
1.50
1.25
0.75
Average
opacity,
5



17.5
20

23
5
11
12
13
5
Maximum
opacity,
9
M



30
40

35
10
20
20
20
10
Observer 2
Duration of Average Maximum
operation, opacity, opacity,
min. 5 t>



1. 00 16 25

1. 00 23 35
0. 75 23 30





                                    average opacity for all readings    - 15X
                                    maximum opacity during  all readings  - 402
Data were not obtained by Method 9 on 6/24/80.
                                      C-102

-------
o
I
CD
CO
                   Table  C-75.   VISIBLE  EMISSIONS OBSERVATION  DATA, EPA METHOD  9—BLISTER DISCHARGE
                                      FROM CONVERTER  AT THE TAMANO SMELTER  IN  JAPAN3>b'C



Opacity.
X
5
10
15
20
25
30
35
Total time equal to or greater
than given opacity
1st blister
discharge
mln: sec.

8:00
5:00
3:15
1:30
0:30
0:15

I of total
time
53
33
22
10
3
2

2nd blister
discharge
win: sec.

11:30
8:45
5:15
3:15
2:00
0.45
0.15
X of total
time
96
73
44
27
17
6
2
3rd blister
discharge
mln: sec.

1:00
0:30


0.15


% of total
time
29
14


7


Total blister
discharge*"
min. sec:

20:30
14:15
8:30
4:45
2:45
1:00
0:15
X of total
tine
67
47
28
16
8
3
< 1
                  Observation point:  converter secondary hood system.
                  bData were based on  a  total  of 30 5-roinute observations  for three  successive blister discharges of the total
                   four blister discharges during one converter cycle.   Duration of  each of the three discharges observed were
                   15 minutes, 12 minutes, and 3.5 minutes, respectively.
                  cTable C-76 summarizes the observation data into average opacities for each set of 6-minute data.
                   Total of the three individual  blister discharges.

-------
Table C-76.   SUMMARY  OF  AVERAGE  OBSERVED OPACITIES  FOR BLISTER
          DISCHARGE AT THE  TAMANO  SMELTER  IN JAPANa

     Sat No.                       Average Opacityc,%
                                            6
        2                                   8
        3                                  11
        4                                  10
        5                                   9
    Based on same observation data used for Table C-75
    Observation time for each set was 6 minutes.
   cAverage of all sets is 9 percent.
                               C-104

-------
o
I
o
en
                 Table  C-77.   SUMMARY  OF EPA METHOD  9 VISIBLE EMISSIONS DATA-INDIVIDUAL  AND TOTAL MATTE


                             CHARGES TO  CONVERTER  OBSERVED  AT THE TAMANO SMELTER  IN JAPAN3'b'c'd
Opacity.
X
5
10
25
Total time equal to or greater
than given opacity
1st matte
charge
min: sec.
0:45
0:15
X of total
time
43
14
2nd matte
charge
min: sec.
0:45
0:15
X of total
time
60
20
3rd matte
charge
min: sec.
0:45
0:15
X of total
time
43
14
4th matte
charge
min: sec.
0
X of total
time

Total matte
charge
•In: sec.
2:15
0:30
0:15
X of total
time
B
4
    a) Matte charges  I. 2 and 3 were successive charges; respective charging  times for matte charges 1. 2. 3 and 4 were 1.75 min.. 1.25 mln.. 1.75 min..
       and  I./a mln.



    b) Observation point:  converter secondary hood  system.



    c) Data were based on a tolal of 6.5 minute observation  for  three successive matte charges at  the beginning of one converter cycle and an

       intermediate matte charging during the cycle.  Average  duration of each matte charge was 1.5 minutes.



    d) lotal of the four individual matte charges;  average  opacity for matte charging, based on total observation, is 3.0 percent.

-------
Table C-78.  SUMMARY OF VISIBLE EMISSIONS OBSERVATION DATA-
         COPPER BLOW AT THE TAMANO SMELTER IN JAPAN3
Set No. b
1
2
3
4
Average Opacity, %
0
0
0
0
     Observation point:  converter secondary hood system.
     Each set is based on 6-minute observation.
                                C-106

-------
Table C-79.   SUMMARY OF VISIBLE EMISSIONS OBSERVATION DATA--

          SLAG BLOW AT THE TAMANO SMELTER IN JAPAN3
Set No.b
1
2
3
4
5
Average Opacity, %
0
0
0
0
0
Observation point: converter secondary hood system.

Each set represents a 6-minute observation.  Set Nos. 1 and 2
are based on the 1st slag blow and set nos. 3 through 5 are based
on the second slag blow of the three slag blow total of the complete
converter cycle.
                          C-107

-------
            Table C-80.   SUMMARY  OF VISIBLE EMISSIONS
        OBSERVATION DATA—CONVERTER SLAG DISCHARGE AT THE
                    TAMANO SMELTER IN  JAPAN3
Set No.b
1
2
Average Opacity, %
0
0
Observation point:  converter secondary hood system.
 Each of two consecutive sets of 6-minute observations  are made during
 one slag discharge.
                                    C-108

-------
C.ll REFERENCES


1.   TRW Environmental Engineering Division.  Emission Testing of
     ASARCO Copper Smelter, Tacoma, Washington.  U.S. Environmental
     Protection Agency.  EMB Report No. 78-CUS-12.  April 1979.

2.   Katari, V., et. al.   Trip for ASARCO Copper Smelter, Tacoma,
     Washington, during June 24 to 26, 1980.  Pacific Environmental
     Services, Incorporated.  July 14, 1980.  p. 7.

3.   Harris, D.L., Monsanto Research Corporation.  Air Pollution
     Emission Test, ASARCO Copper Smelter, El Paso, Texas.  U.S.
     Environmental Protection Agency.  EMB Report No. 77-CUS-6.
     June 20-30, 1977.

4.   TRW Environmental Engineering Division.  Air Pollution Emission
     Test.  ASARCO Copper Smelter, El  Paso, Texas.  U.S.  Environmental
     Protection Agency.  EMB Report No. 78-CUS-7.  April  5, 1978.

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

6.   Radian Corporation.  Arsenic Emissions from an Electrostatic
     Precipitator of the Phelps-Dodge  Copper Smelter  in  Ajo,  Arizona.
     U.S. Environmental Protection Agency.  EPA Contract No.  68-02-13-19.
     April 4, 1977.

7.   Rooney, T., TRW Environmental Engineering Division.  Emission
     Test Report (Acid Plant).  Phelps-Dodge Copper Smelter,  Ajo,
     Arizona.  U.S. Environmental Protection Agency.  EMB Report
     No. 78-CUS-ll.  March 1979.

8.   Rooney, T., TRW Environmental Engineering Division.  Emission
     Test Report.  Phelps-Dodge Copper Smelter, Ajo,  Arizona.  U.S.
     Environmental Protection Agency.  EMB Report No. 78-CUS-9.
     February 1979.

9.   Rooney, T., TRW Environmental Engineering Division.  Emission
     Testing of Ph'elps-Dodge Copper Smelter, Playas,  New Mexico.   U.S.
     Environmental Protection Agency.  EMB Report No. 78-CUS-10.
     March  1979.

10.  Rooney, T., TRW Environmental Engineering Division.  Emission
     Testing of Phelps-Dodge Copper Smelter, Douglas, Arizona.    U.S.
     Environmental Protection Agency.  EMB Report No. 78-CUS-9.
     February 1979.

11.  TRW Environmental Engineering Division.   Emission Testing of
     Kennecott Copper  Smelter, Magna,  Utah.  U.S. Environmental
     Protection Agency.  EMB Report No. 78-CUS-13.  April 1979.


                                C-109

-------
12.
13.
larkin, R. and J. Stelner.  Acurex Corporation/Aerotherm Division
Arsenic Emissions at Kennecott Copper Corporation, Hayden, Arizona
U.S. Environmental  Protection Agency.  EPA Report No. 76-NFS-l
May 1977.

Katari, V. and I.J.  Weisenberg.   Trip Report--Visit to Hibi Kyodo
Smelting Company's  Tamano Smelter during  the week of March 10,
1980.   Pacific Environmental  Services,  Incorporated.  June 9
1980.   Appendix  A.                                           '
                                  c-no

-------
 APPENDIX D





TEST METHODS
        D-l

-------
                           TEST METHODS




Drl  EMISSION MEASUREMENT METHODS





     At the beginning of the testing program,  a  literature  search



was conducted to identify available sampling  and analytical



techniques for determining arsenic emissions.  The  search  revealed



that most arsenic emissions  are in the form of arsenic  trioxide  and



arsenic pentoxide.  According to the literature, the most  commonly



used arsenic sampling method has been filtration; however,  a  number



of reports have indicated that filtration  alone  is  not  adequate,



even at ambient temperatures, because arsenic  trioxide  is  a



potentially volatile material.  Since it was  decided to determine



the amount of arsenic collected as a particulate, the Method  5



train, with back-up impinger collectors, was  chosen as  the  starting



point for the arsenic sampling system. Based  on the available



information, a dilute sodium hydroxide solution  was chosen  as a



collecting solution for the  impingers. This,  however,  presented a



problem since many of the gas streams to be sampled had very  high



concentrations of sulfur dioxide (S02), some  as  high as 3.5 percent.



Therefore, a series of impingers containing hydrogen peroxide was



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

-------
sodium hydroxide to remove the $03.   This  was  the configuration
for the "working train" used during  the first  four field  tests.
     Analytical  methods for arsenic  were better defined  in  the
literature.  The most commonly-used  procedure  is  a wet chemical
method based on  arsine generation, but  certain metals  including
copper are interfering agents with this method.  Instrumental
techniques include atomic absorption, neutron  activation,  and
x-ray fluorescence.  Atomic absorption  spectrophotometry  (AAS)
was chosen as the most promising technique because of  its  ready
availability, familiarity, and low cost; however, arsenic  absorbs
weakly and only  in the extreme ultra-violet area of the  spectrum
(193.7 nm).  At  that wavelength, molecular absorption  by  flame
gases and solution species can interfere with  arsenic  detection.
Despite this, conventional AAS can still be used, provided  that:
(1) the fuel and combustion gas are  carefully  chosen and  nonatomic
background correction is used; and  (2)  arsenic concentrations  are
relatively high.  However, for lower arsenic concentrations,  the
interference effects necessitate the use of a  special, more
sensitive technique, such as the hydride generator or  the carbon
rod (flameless)  system.  Before testing began, both conventional
and special AAS  methods were compared and evaluated, in  terms  of
their accuracy,  precision, and sensitivity.
                                D-3

-------
     During the first  two  field  tests,  samples were collected with
the working train and  analyzed either  by  conventional or carbon
rod AAS depending on the  arsenic concentration.  The  analytical
results showed that 95 to  100 percent  of  the  arsenic  was collected
ahead of the NaOH impingers.  In the course of analyzing these
samples, the following detailed  sample preparation procedure was
developed.   Solid samples  were digested with  0.1 N sodium
hydroxide,  extracted with  concentrated nitric acid, evaporated to
dryness, and then redissolved in dilute nitric acid.  Liquid
samples were treated similarly except  that there was  no need for
the sodium hydroxide digestion step.   Advantages of the sample
preparation procedure  include:   (1)  reduction of the  level  of the
collected sulfuric acid in the  liquid  sample  fraction;
(2) dissolution of the arsenic  in the  solid samples;  and
(3) production of a similar solution matrix for  all the different
sample fractions.
     After the second  test, questions  were  raised  about the
sampling and analytical procedures.  First  of all, laboratory
studies of vaporized  arsenic trioxide  showed  no  difference in  the
arsenic collection efficiency of 0.1 N sodium hydroxide and pure
water.  These results  indicated  that the arsenic collection
mechanism is condensation and  that any condenser would  be  an
effective collector.   Consequently,  the conventional  Method 5
train  (with H20  impingers) was  suggested as  an  alternate  to the
                                D-4

-------
working train and simultaneous testing of  the two  trains  was
planned for the next facility.
     Second, an evaluation of the different  AAS  techniques  for
low-concentration uncovered some precision and accuracy
problems with the carbon rod method when  large quantities of
dissolved solids (particularly sulfates)  are present.  The  hydride
generator technique, it was found,  gives much more precise  and
accurate results in the presence of dissolved solids.  In view  of
this, it was decided that all  future low-concentration arsenic
samples (i.e., too low for conventional AA analysis) would  be
analyzed by the hydride generator method.
     Third, concern was expressed that arsenic was being  lost in
the evaporation step of sample preparation.   To  investigate this,
recovery studies were performed on  standard  samples.  These studies
showed that there is no significant loss  of  arsenic during  the
evaporation step.
     Fourth, additional studies showed that  while  arsenic trioxide
is soluble in alkaline, acid,  and neutral  solutions, its  rate of
dissolution is slow except in alkaline solutions.   Therefore, the
clean-up procedure for future test  was modified, to require that
the train be rinsed with 0.1 N sodium hydroxide  to insure removal
of condensed arsenic.
     Fifth, a comparison of arsenic extraction techniques
indicated that higher arsenic yields (by up  to 200 percent) can be
                               D-5

-------
obtained from smelter particulate when  a method  capable  of
dissolving the entire sample is  used  instead  of  the  less  rigorous
acid extraction procedure.   As  a result,  it was  decided  that  in
future tests, filters would be  analyzed by both  methods,  until
more conclusive filter extraction  data  were obtained.
     During the third and fourth field  tests,  the  working train
was used for sampling, but additional  runs were  taken  during  the
fifth test using paired trains  of  the working and  alternate
procedures.  Analysis of the samples  from the paired tests  showed
no significant difference in collection efficiency.   Therefore,  the
final recommendation was to use the alternate train, since  it is
easier to operate and analyze.   During  the fifth and final  field
test, the alternate train was used.
     Filters from the third, fourth,  and fifth field tests  were
extracted, using both the total dissolution  and acid extraction
procedure.   The results showed that filters  extracted by the less
rigorous method could in some cases yield 25 percent less arsenic
than  if totally dissolved.  Based upon these results, the final
recommendation was to extract the filters first by the simple acid
extraction;  then, if  any undissolved sample remained, to extract
the undissolved solids  by  the total dissolution method.
                                D-6

-------
D.2  CONTINUOUS MONITORING

     There is currently no available method for continuously
monitoring arsenic emissions.  For purposes of demonstrating proper
operation and maintenance of control devices,  continuous monitors
are available for measuring opacity from baghouses or electrostatic
precipitators, and measuring pressure drop across scrubbers.
However, these measurements are not necessarily indicators  of the
magnitude of arsenic emissions and should not  be used for compliance
determinations.  In addition, opacity may not  be applicable as
an indicator of proper operation and maintenance where baghouses
and precipitators are used to control captured fugitive emissions
because    the uncontrolled particulate is very low in
concentration.
     The recommended monitoring program for continually assessing
arsenic emissions is a periodic application of the performance test
Method 108 as recommended in Part D.3 below.  This is the only
method evaluated at this time for demonstration of compliance with
arsenic emissions.

D.3  PERFORMANCE TEST METHODS

     The recommended performance test method for arsenic is Method
108.   Based on the development work already discussed,  the  method
uses  the Method 5 train for sampling, 0.1 N sodium hydroxide for
                                   D-7

-------
cleanup,  and either conventional or hydride generator AAS for
sample analysis.   In order  to perform Method 108, Methods 1 through
4 must also be used.  Subpart A  or 40 CFR 60 requires that facilities
subject to standards of  performance for  new stationary sources be
constructed so as to provide sampling ports adequate for the
applicable test methods, and platforms,  access,  and utilities
necessary to perform testing at  those ports.
     Sampling costs for  performing a test consisting of three
Method 108 runs is estimated to  range from $10,000 to $14,000.   If
in-plant personnel are used to  conduct  tests,  the costs will be
somewhat less.

D.4  REFERENCES

     1.  Hefflefinger, R.E. and D.L. Chase  (Battelle).  Analysis of
Copper Smelter Samples for Arsenic  Content.  Prepared for  U.S.
Environmental Protection Agency. Research Triangle  Park,  NC.
April  1977.   14 p.
     2.  Haile, D.M.  (Monsanto Research Corporation).  Final  Report
on  the Development  of Analytical Procedures  for the  Determination of
Arsenic  from  Primary Copper Smelters.   Prepared for  U.S.
Environmental  Protection Agency.  Research  Triangle  Park,  NC.
February 1978.  27  p.
      3.   Harris,  D.L.   (Monsanto Research Corporation).   Particulate
 and Arsenic Emission Measurements from a Copper Smelter.   Prepared
                               D-8

-------
for the U.S.  Environmental Protection Agency.  Research Triangle



Park, NC. 77-CUS-5.   April 1977.  48 p.



     4.  Harris,  D.L.   (Monsanto Research Corporation).  Participate



and Arsenic Emission  Measurements from a Copper Smelter.  Prepared



for the U.S.  Environmental Protection Agency.  Research Triangle



Park, NC.  77-CUS-6.   June 1977.  276 p.



     5.  TRW, Inc.  Emission Testing of Asarco Copper Smelter.



Prepared for  the  U.S.  Environmental Protection Agency.  Research



Triangle Park,  NC.  77-CUS-7.  April 1978.  150 p.
                              D-9

-------
                            APPENDIX E
QUANTITATIVE EXPRESSIONS OF PUBLIC CANCER RISKS FROM EMISSIONS OF



   INORGANIC ARSENIC FROM LOW-ARSENIC PRIMARY COPPER SMELTERS
                               E-l

-------
     QUANTITATIVE  EXPRESSIONS OF PUBLIC CANCER RISKS FROM EMISSIONS OF
         INORGANIC ARSENIC FROM LOW-ARSENIC PRIMARY COPPER SMELTERS

E.I  INTRODUCTION
E.I.I  Overview
     The quantitative expressions  of public  cancer risks presented in this
appendix are based on (1) a dose-response model  that numerically relates
the  degree of exposure to airborne inorganic arsenic to the risk of getting
lung cancer, and  (2) numerical expressions of public exposure to ambient
air  concentrations of inorganic arsenic estimated to be caused by emissions
from stationary sources.  Each of these factors is discussed briefly below
 and  details  are provided in the following sections of  this appendix.
 E.l.2   The Relationship  of Exposure to Cancer Risk
      The relationship of exposure to the  risk of  getting lung  cancer is
 derived from epidemiological  studies in  occupational  settings  rather than
 from studies of excess cancer incidence  among the public.   The epidemiological
 methods that have successfully revealed associations between occupational
 exposure  and cancer for substances  such as asbestos, benzene,  vinyl  chloride,
 and ionizing radiation, as well as  for inorganic arsenic,  are not readily
 applied to  the public sector, with  its increased number of confounding
 variables,  much  more diverse  and mobile exposed  population, lack of consoli-
  dated medical  records,  and almost  total  absence  of  historical  exposure
  data.  Given such uncertainties, EPA considers  it  improbable  that any
  association, short of very large increases  in  cancer, can  be  verified  in
  the general population with  any  reasonable  certainty by an epidemiological
  study.   Furthermore, as noted by the National  Academy of  Sciences (NAS)l,
   "...when there  is exposure to a material,  we are not starting at an origin
                                      E-2

-------
of zero cancers.   Nor are we  starting  at  an  origin  of  zero carcinogenic
agents in our environment.  Thus,  it  is likely  that any carcinogenic agent
added to the environment will  act  by  a particular mechanism  on  a  particular
cell  population that is already  being  acted  on  by the  same mechanism to
induce cancers."   In discussing  experimental  dose-response curves,  the NAS
observed that most information on  carcinogenesis is derived  from  studies  of
ionizing radiation with experimental  animals and with  humans which  indicate
a linear no-threshold dose-response relationship at low doses.  They added
that although some evidence exists for thresholds in some animal  tissues,
by and large, thresholds have not  been established  for most  tissues.  NAS
concluded that establishing such low-dose thresholds "...would  require
massive, expensive, and impractical experiments ..." and recognized that
the U.S. population "...is a  large, diverse, and genetically heterogeneous
group exposed to a large variety of toxic agents."   This fact,  coupled with
the known genetic variability to carcinogenesis and the predisposition of
some individuals to some form of cancer,  makes  it extremely  difficult, if
not impossible, to identify a threshold.
     For these reasons, EPA has  taken the position, shared by other Federal
regulatory agencies, that in  the absence  of  sound scientific evidence to
the contrary, carcinogens should be considered  to pose some  cancer  risk
at any exposure level.  This  no-threshold presumption is based  on the view
that as little as one molecule of  a carcinogenic substance may  be sufficient
to transform a normal cell into  a  cancer  cell.   Evidence is  available from
both the human and animal health literature  that cancers may arise  from a
single transformed cell.  Mutation research  with  ionizing  radiation in cell
cultures indicates that such  a transformation can occur as the  result of
                                   E-3

-------
interaction with as  little  as  a  single  cluster of  ion pairs.   In reviewing
the available data regarding carcinogenicity, EPA  found no compelling
scientific reason to abandon the no-threshold presumption for  inorganic
arsenic.
     In developing the exposure-risk  relationship  for  inorganic  arsenic,  EPA
has assumed that a linear no-threshold  relationship  exists  at  and  below  the
levels of exposure reported in the epidemiological studies  of  occupational
exposure.  This means that any exposure to inorganic arsenic is  assumed
to pose some risk of lung cancer and that the  linear relationship  between
cancer risks and levels of public exposure is  the same as  that between cancer
risks and levels of occupational exposure.  EPA believes that this assumption
is reasonable for public health protection in light of presently available
information.  However, it  should be recognized that the case for the linear
no-threshold dose-response relationship model  for inorganic arsenic is not
quite  as  strong as  that  for carcinogens which interact directly or in
metabolic  form  with DNA.   Nevertheless, there is  no adequate basis for
 dismissing the  linear no-threshold model  for inorganic arsenic.  The exposure-
 risk  relationship used by  EPA represents  only a plausible upper-limit risk
 estimate in the sense that the  risk  is probably not higher  than the calculated
 level  and could be  much  lower.
      The numerical  constant that defines  the exposure-risk  relationship
 used by EPA in its  analysis of  carcinogens  is called  the unit risk  estimate.
 The unit risk estimate for an air pollutant is  defined as  the lifetime
 cancer risk occurring in a hypothetical population  in which all  individuals
 are exposed continuously from birth  throughout  their lifetimes (about 70
 years) to a concentration of 1 ug/m3 of the agent in the  air which they
                                  E-4

-------
breathe.   Unit risk estimates are used for two  purposes:   (1)  to compare
the carcinogenic potency of several  agents with each  other,  and (2)  to  give
a crude indication of the public health risk  which might be  associated  with
estimated air exposure to these agents.   The  comparative potency of  different
agents is more reliable when the comparison is  based  on studies of like
populations and on the same route of exposure,  preferably inhalation.
     The unit risk estimate for inorganic arsenic that is used in this
appendix was prepared by combining the three  different exposure-risk
numerical constants developed from three occupational studies.2  The unit risk
estimate is expressed as a range that reflects  the statistical uncertainty
associated with combining the three exposure-risk relationships.  The
methodology used to develop the unit risk estimate is described in E.2
below.  EPA is updating its health effects assessment document for inorganic
arsenic.  A preliminary determination by EPA's  health scientists is  that  the
unit risk estimate may change.
E.I.3  Public Exposure
     The unit risk estimate is only one of the  factors needed to produce
quantitative expressions of public health risks.  Another factor needed
is a numerical expression of public exposure, i.e., of the numbers of
people exposed to the various concentrations  of inorganic arsenic.  The
difficulty of defining public exposure was noted by the National Task
Force on Environmental Cancer and Health and Lung Disease in their 5th
Annual Report to Congress, in 1982.3  jney reported that "...a large
proportion of the American population works some distance away from their
homes and experience different types of pollution in  their homes, on the
way to and from work, and in the workplace.  Also, the American population
is quite mobile, and many people move every few years."  They also  noted the
                                    E-5

-------
necessity and difficulty of dealing with  long-term exposures  because of
"...the long latent period required for the  development  and expression
of neoplasia [cancer]..."
     EPA's numerical expression of public exposure  is based  on two estimates.
The first is an estimate of the magnitude and location of long-term average
ambient air concentrations of inorganic arsenic in  the vicinity of emitting
sources based on dispersion modeling using long-term estimates of source
emissions and meteorological conditions.   The second is an estimate of the
number and distribution of people living in the vicinity of emitting sources
based  on Bureau of Census  data which "locates" people by population centroids
in census tract areas.  The people and concentrations are combined to produce
numerical expressions  of  public exposure by an approximating  technique
contained in  a computerized model.  The methodology  is  described  in E.3
 below.
 E.I.4   Public Cancer Risks
        By combining numerical  expressions of  public  exposure with the unit
 risk estimate,  two types of numerical  expressions of public  cancer risks  are
 produced.   The first, called individual  risk, relates to the person or
 persons estimated to live in the  area  of highest concentration as estimated
 by the dispersion model.  Individual  risk is expressed as "maximum lifetime
 risk."  As used here, the word "maximum" does not  mean the  greatest possible
 risk of cancer to the public.  It is based only on the maximum exposure
 estimated by the procedure used.   The second, called aggregate risk,  is a
 summation of all the  risks to people estimated to be living within the
 vicinity (usually within  20 kilometers) of a source and is customarily summed
 for all the  sources  in a  particular category.  The  aggregate  risk is expressed
 as  incidences of  cancer  among all of the exposed population  after 70 years
                                    E-6

-------
of exposure;  for statistical  convenience,  it  is  often  divided  by  70  and
expressed as  cancer incidences  per year.   These  calculations are  described
in more detail  in E.4 below.
     There are also risks of nonfatal  cancer  and of serious  genetic  effects,
depending on  which organs receive the  exposure.   No numerical  expressions
of such risks have been developed;  however,  EPA  considers all  of these risks
when making regulatory decisions on  limiting  emissions of inorganic  arsenic.

E.2  THE UNIT RISK ESTIMATE FOR INORGANIC ARSENIC2
E.2.1  The Linear No-Threshold Model for Estimation of Unit Risk Based on
       Human Data (General)4
     Very little information exists that can be utilized to extrapolate
from high exposure occupational studies to low environmental  levels.
However, if a number of simplifying assumptions are made, it is possible
to construct a crude dose-response model whose parameters can be estimated
using  vital statistics, epidemiologic studies, and estimates of worker
exposures.  In human studies, the response is measured in terms of the
relative risk of the exposed cohort of individuals compared to the control
group.  The mathematical model employed assumes that for low exposures the
lifetime probability of death from  lung cancer  (or any cancer), P, may be
represented by  the linear equation
                                P = A + BHx                    (1)
where  A  is the  lifetime probability of cancer in  the absence of the agent, x
is  the average  lifetime exposure to environmental  levels in micrograms per
cubic  meter  Ug/nr*), and  BH  is  the  increased probability of cancer  associated
with  each  ng/m3  increase  of  the agent  in  air.
      If  we make  the  assumption  that R, the relative risk of lung cancer  for
exposed workers, compared to the general  population, is  independent of the length
                                    E-7

-------
or age of exposure but depends  only  upon  the  average  lifetime  exposure,  it
follows that
                           P     A +  BH  (XQ  +  XT)
                       R =
                                                               (2)
                           PQ    A + BH  (XQ)
or
                         RP0 = A + BH  (XQ + XT)                 (3)
where XQ = lifetime average exposure to the agent for the general  popu-
lation, X] = lifetime average exposure to the agent in the occupational
setting, and PQ = lifetime probability of respiratory cancer applicable  with
no or negligible arsenic exposure.  Substituting PQ = A + BH XQ and rearranging
gives
                             BH = PQ (R - D/XI                 (4)
To use this model, estimates of R and X]  must be obtained from the epidemio-
logic studies.  The value PQ is derived from the age-cause-specific death
rates for combined males found in 1976 U.S. Vital Statistics tables using
the life table methodology.  For lung cancer the estimate of PQ is 0.036.
E.2.2  The Unit Risk Estimate for Inorganic Arsenic2
     As noted in the health effects assessment document5 for inorganic
arsenic, there are numerous occupational  studies which relate increased
cancer  rates  to arsenic exposure.  Based on these studies, it is concluded
in the  health assessment document that there is substantial evidence that
inorganic arsenic  is a human carcinogen.  However, many of these studies
are  inappropriate  for use  in developing a unit risk estimate for inorganic
arsenic because the  route  of exposure was not by inhalation or because  it
was  impossible to  make a reasonable estimate of the population's lifetime
 average exposure.
                                    E-8

-------
     Three studies, Lee and Fraumeni (1969), Ott et al. (1974), and Pinto
et al. (1977), contained enough pertinent information to make independent
quantitative estimates of human cancer risks due to human exposures to
atmospheric arsenic.  The crudeness of the exposure estimates in those
studies is due to such factors as high variability in the chemical measurement
of arsenic, a scarcity of monitoring data, and the necessity of working
from summarized data tables presented in the literature rather than complete
data on all individuals.   However, by accepting the data in spite of their
recognized limitations, and making a number of simplifying assumptions
concerning dose-response relationships and exposure patterns, it was possible
to estimate the carcinogenic potency of arsenic.  Using a linear model, it
was estimated that the increase in the lung cancer rate per increase of 1
iig/m3 of atmospheric arsenic was 9.4 percent (Pinto et al.), 17.0 percent
(Ott et al.), and 3.3 percent (Lee and Fraumeni).   The consistency of these
estimates is very good considering the relative crudeness of the data upon
which they are based.  The geometric mean of the rate estimates from the
three studies was calculated to be 8.1  percent.  Using this value as a best
estimate and applying equation 4,  one calculates the unit risk estimate of
2.95 x TO'3 per ug/m3.
     If we assume that the linear model  and exposure estimates are correct,
so that the only source of uncertainty is from combining results from the
three different studies,  a 95 percent confidence interval  for the above
unit risk  estimate may  be obtained.   Upper and lower 95 percent confidence
limits  can be obtained  by multiplying the unit risk estimate by about 4 and
0.25,  respectively.   Thus,  the 95  percent statistical  confidence limits for the
unit risk  estimate range  from 7.5  x  10-4  to 1.2 x  10~2.
                                  E-9

-------
E.3  QUANTITATIVE EXPRESSIONS OF PUBLIC  EXPOSURE TO INORGANIC  ARSENIC
     EMITTED FROM LOW-ARSENIC PRIMARY COPPER SMELTERS
E.3.1  EPA's Human Exposure Model  (HEM)  (General)
     EPA's Human Exposure Model  is a general model  capable of  producing
quantitative expressions of public exposure to ambient air concentrations
of pollutants emitted from stationary sources.  HEM contains (1) an atmospheric
dispersion model, with included meteorological data, and (2) a population
distribution estimate based on Bureau of Census data.   The only input data
needed to operate this model are source data, e.g., plant location, height
of the emission  release  point, and temperature of the offgases.  Based on the
source data, the model estimates  the magnitude and  distribution of ambient
air  concentrations  of the  pollutant  in  the  vicinity of the  source.  The
model  is programmed to  estimate these concentrations within a  radial  distance
of 20  kilometers from the  source.   If other radial  distances  are preferred,
 an over-ride feature allows the user to select  the distance desired.   The
 selection of 20 kilometers as  the programmed distance  is  based on  modeling
 considerations, not on  health  effects criteria  or  EPA  policy.   The dispersion
 model  contained in HEM is felt to be reasonably accurate  within 20 kilometers.
 If  the user wishes to use a dispersion  model other than the one contained
 in  HEM to estimate ambient air concentrations in the  vicinity of a source,
 HEM can accept  the concentrations if they are put into an appropriate
 format.
      Based  on  the  radial  distance specified, HEM combines  numerically the
 distributions  of pollutant  concentrations  and people to produce quantitative
 expressions of public  exposure to the  pollutant.
                                     E-10

-------
 E.3.1.1  Pollutant Concentrations Near a Source.  The dispersion model
 within the HEM is a gaussian diffusion model that uses the same basic
 dispersion algorithm as EPA's Climatological Dispersion Model.6 The algorithm
 has been simplified to improve computational efficiency.7 The algorithm is
 evaluated for a representative set of input values as well as actual plant
 data, and the concentrations input into the exposure algorithm are arrived
 at by interpolation.  Stability array (STAR) summaries are the principal
 meteorological input to the HEM dispersion model.  STAR data are standard
 climatological frequency-of-occurence summaries formulated for use in EPA
 models and available for major U.S. meteorological monitoring sites from
 the National Climatic Center, Asheville, N.C.   A STAR summary is a joint
 frequency-of-occurence of wind speed, atmospheric stability, and wind
 direction, classified according to Pasquill's  categories.   The STAR summaries
 in HEM usually reflect 5 years of meteorological data for each of 309 sites
 nationwide.  The model  produces polar coordinate receptor grid points
 consisting of 10 downwind distances located along each of 16 radials which
 represent wind directions.   Concentrations are estimated by the dispersion
 model for each of the 160 receptors located on this grid.   The radials are
 separated by 22.5-degree intervals beginning with 0.0 degrees and proceeding
 clockwise to 337.5 degrees.  The 10 downwind distances for each radial are
 0.2, 0.3, 0.5, 0.7,  1.0,  2.0, 5.0, 10.0, 15.0, and 20.0 kilometers.  The
 center of the receptor grid for each plant is  assumed to be the plant center.
 E-3.1.2  The People  Living Near A Source.   To  estimate the number and
 distribution of people residing within 20 kilometers  of each plant, the
model  contains a slightly modified version of  the "Master  Enumeration
District List—Extended"  (MED-X)  data base.  The data base is broken down
                                   E-ll

-------
into enumeration district/block  group (ED/BG)  values.   MED-X  contains  the
population centroid coordinates  (latitude  and  longitude)  and  the  1970
population of each ED/BG in the  United States  (50  States  plus the District
of Columbia).  For human exposure estimates, MED-X has  been reduced  from
its complete form (including descriptive and summary data) to produce  a
computer file of the data necessary  for the estimation.   A separate  file of
county-level growth factors, based on 1978 estimates of the 1970  to  1980
growth factor at the county level, has been used to estimate  the  1980
population for each ED/BG.   HEM  identifies the population around  each  plant
by using the geographical coordinates of the plant.   The  HEM  identifies,
selects, and stores for later use those ED/BGs with coordinates falling
within 20 kilometers of plant center.
E.3.1.3  Exposure?.  The Human Exposure Model  (HEM)  uses  the  estimated
ground level concentrations of a pollutant together with  population  data to
calculate public exposure.   For  each of 160 receptors located around a
plant, the concentration of the  pollutant  and  the  number  of people estimated
by the HEM to be exposed to that particular concentration are identified.
The HEM multiplies these two numbers to produce exposure  estimates and sums
these products for each plant.
     A two-level scheme has been adopted in order  to pair concentrations
and populations prior to the computation of exposure.   The two level approach
is used because the concentrations are defined on  a radius-azimuth (polar)
grid pattern with non-uniform spacing.  At small radii,  the grid  cells are
usually smaller than ED/BG's; at large radii,  the  grid  cells  are  usually larger
than ED/BG1s.  The area surrounding the source is  divided into two regions,
and each ED/BG is classified by  the region in  which its centroid  lies.
                                   E-12

-------
  Population exposure is calculated difrerently  for the ED/BG's  located
  within each region.   For ED/BG  centroids  located  between  0.1 km  and  2.8  km
  from the emission source,  populations  are  divided between  neighboring
  concentration grid points.   There are  96  (6 x  16}  polar grid points  within
  this range.   Each grid point has  a polar sector defined by two concentric
  arcs and two  wind direction  radials.   Each of  these grid points and  respec-
  tive concentrations are  assigned  to the nearest ED/BG centroid identified
  from MED-X.   Each  ED/BG  can  be paired with one or many concentration points.
  The  population associated with the ED/BG centroid is then divided among all
  concentration grid points assigned to it.   The land area within each  polar
  sector is considered in  the apportionment.
      For population centroids between 2.8  km and 20 km from the source,  a
 concentration grid cell, the area  approximating a  rectangular shape bounded
 by four receptors, is  much larger  than  the  area of a  typical ED/BG.   Since
 there is  an approximate linear relationship between the  logarithm of  concen-
 tration and the logarithm of distance for  receptors more than 2 km from the
 source,  the entire population of the ED/BG  is assumed  to be exposed to the
 concentration  that is  logarithmically interpolated radially and arithmetically
 interpolated azimuthally  from the  four  receptors bounding the grid cell.
 Concentration  estimates  for 80 (5  x 16) grid cell receptors at 2.0,  5.0,
 10.0,  15.0, and 20.0 km from  the source along each of 16 wind directions
 are used as reference points for this interpolation.
      In summary, two approaches are used to arrive at coincident
concentration/population data points.   For  the 96 concentration  points
within 2.8 km of the source,  the  pairing occurs  at  the  polar grid  points
using an apportionment  of ED/BG population  by  land  area.  For the  remaining
portions of the grid, pairing occurs at  the  ED/BG centroids  themselves
                                   E-13

-------
through the use of log-log  and  linear  interpolation.   (For a more detailed
discussion of the model  used  to estimate exposure,  see Reference 7.)
E.3.2  Public Exposure to Inorganic  Arsenic  Emissions  from Low-Arsenic Primary
       Copper Smelters                 ~~~
E.3.2.1  Source Data.   Fourteen smelters are included  in  the analysis.
Table E.I lists the names and addresses of the plants  considered, and Table
E.2 lists the plant data used as input to  the Human Exposure Model  (HEM).
E.3.2.2  Exposure Data.  Table E.3 lists,  on a pi ant-by-plant  basis,  the
total number of people encompassed by the  exposure analysis  and the total
exposure.  Total exposure is the sum of the products of  number of  people
times the ambient air concentration to which they are exposed, as  calculated
by  HEM.  Table  E.4 sums, for the entire source category (14 plants), the
numbers  of  people exposed  to various ambient concentrations, as calculated
by  HEM.
                                    E-14

-------
Table E-l.  IDENTIFICATION OF LOW-ARSENIC PRIMARY COPPER SMELTERS
Plant Number Code
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Plant Name and Address
ASARCO, Inc.
El Paso, TX
ASARCO, Inc.
Hayden, AZ
Kennecott Corp.
Hayden, AZ
Kennecott Corp.
Hurley, NM
Kennecott Corp.
McGill, NV
Kennecott Corp.
Garfield, UT
Phelps-Dodge Corp.
Morenci, AZ
Phelps-Dodge Corp.
Douglas, AZ
Phelps-Dodge Corp.
Ajo, AZ
Phelps-Dodge Corp.
Hidalgo, NM
Copper Range Co.
White Pine, MI
Magma Copper
San Manuel , AZ
Inspiration Consolidated Copper Co.
Miami, AZ
Tennessee Chemical Co.
Copperhill, TN
                            E-15

-------
t-<. input uata to txposure Model  Low-Arsenic  Primary Copper Shelters
                      (Assuming  Basel-ine  Controls)
Plant
(Emission
Point)
1 1
2
3
4
5.
6

2 1
2
3
4

3 1
2
3
4
5

4 1
2
3
4
5

5 1
2
3
4

6 1
2
3

7 1
2
3
4

8 1
2
3
4

9 1
2
3

10 1
z
3
4
5

U 1
2
4

12 1
2
3
4
5
6
13 1
2
3
4
1* 1
2
3
4
5

Latitude
(Degrees
Minutes
Seconds )
314659

330028

330041

324156

392433

404315

330403

312048

322159

314558

464606


323658
332443
345946

Longtltude
(Degrees
Minutes
Secoods )
1063126

1104559

1104649

1080719

1144612

1121153

1092030

1093520

1125113

1063144

0893318


1103723
1105058
0842243

Arsenic
Emission
Sate
(Kg/yr) .
260
15.130
344
24.500
.390
705

1.720
29.240
1.665
1,380

430
27,520
6,450
851
95

3.4
0.9
464
108
21

394,000
46,000
4.025
450

21.130
3.870
200

520
3.555
4,820
90

18,920
19,280
4,300
4 360
T, *«wv
33.140
1.810
60

640
1.315
140
1,260
20

1,720
344
48

11.180
13
13
516
81
9
782
6,880
859
130
0
0
80
690
9

Emission | Emission
Point j Point
Elevarfjn Diameter
.(Meters) .) (Met«rsl
31 1.7
252 4.9
*0 2.7
30.5 	
30.5 .1.4
•«
61 3.0
30.5 —
30.5 5.2
0 -"-i

30.5 2.4
183 5.2
30.5 —
30.0 2.0
0 ™

61 3.0
61 3.1
30.5
155 7.9
0 —

229 4. 7
30.5 —
30.5 2.0
0 — •

365 7.3
30.5 —
0 .—

184 9.1
184 9.1
30.5 —
0 — •-

166 6.7
172 5.5
30.5 —
0 __

110 4.6
30.5 —
0 —

61 3.0
76 3.5
30.5 	
183 5.5
VBM
154 4.6
30.5
30.5 20)
Q

155 6.1
61 3.0
61 3.0
30.5 —
30.5 2.0
0 	
61 3.1
30.5 —
30.S 2.o
0 	
61 3.0
76.2 3.0
24 2.0
30.5 	
0

Emission*
Point
Cross
Sectional
Area (tr?)
04
3.606
no
3.720
.1,500
"— »
325
4,200
6,300
*•*
84
1,665
4,650
1.750
•»••
325
325
6,300
•1,430
_.
2,725
3,140
1,300
ww
9,140
4,800
""-*
2,200
2.200
11.200
«•
1,685
1,770
3,490
»•
846
3,720
mm-J-
325
520
3,600
2,000
H«
1.296
3,255
3.800
-*-
2.100
325
325
4.200
5.250
325
6,000
1,750
325
520
96
2,400

Emission
Point Gas
Exit
Velocity
M /« fff
14.8
19.6
7.4
0.8
20.3
— -
2.1
0.8
6.7
"~"
12.5
5.2
0.8
20.3
•~~
2.3
5.6
0.8
9.8
"» —
20.4
0.8
20.3
•"—
11.3
0.8
«~
0.24
3.8
0.8
—
10.9
4.7
0.8
— -•
4.8
0.8
— -
2.4
3.7
0.8
7.1
—
6.0
0.8
20.3
— -
8.4
6.2
6.2
0.8
19.0
7.1
0.8
10.5
9.0
10.7
2U.3
0.8
••»
Emission
Point Gas
Te«*.
f «*y \
{ ^ j
339
366
337
298
— _
352
298
339
—
352
533
298
298
...
3t>2
352
298
298
—
422
298
298
— -
339
298
-—
352
352
298
-—
505
505
298
-—
3t>2
298
— -
352
352
298
339
— •
494
296
298
_.
519
325
325
298
296
327
296
298
352
352
298
298
—
Emission
Point
Type
••-••••i-- in ••
Stack
Stack
Stack
Vent
Vent
Fugitive
Stack
Vent
Stack
Fugitive
Stack
Stack
Vent
Stack
Fugitive
Stack
Stack
Vent
Stack
Fugitive
Stack
Vent
Stack
Fugitive
Stack
Vent
Fugitive
Stack
Stack
Vent
Fugitive
Stack
Stack
Vent
Fugitive
Stack
Vent
Fugitive
Stack
Stack
Vent
Stack
Fugitive
Stack
Vent
Stack
Fugitive
Stack
Stack
Stack
Vent
Vent
Fugitive
Stack
Vent
Stack
Fugttlve
Stack
Stack
Stack
Vent
Fugitive
                   •Ission point  to  the mean  wind direction for purpose of calculating


                        E-16

-------
         Table E-3.  TOTAL EXPOSURE AND NUMBER OF PEOPLE EXPOSED
                      (LOW-ARSENIC PRIMARY COPPER SMELTERS)*
                            Total                         Total
                          Number of                      Exposure
    Plant              People Exposed               (People -
1
2
3
4
5
6
7
8
9
10**
11
12
13
14
435,000
8,700
8,700
19,400
4500
79,200
9,000
19,000
9,000
300
3,000
12,000
22,000
21,000
6840
580
160
11
590
62
80
290
521
<1
4
37
370
35
 * A 20-kilometer radius  was  used  for  the analysis of exposure  for the low-
   arsenic primary copper smelters.

** EPA knows that a small  town  was built after  1970 within  the  20-kilometer
   radius of this plant.   Since the population  data base contained in the
   exposure model is based on population locations contained  in the  1970
   census data,  the risk  and  incidence estimates  do not reflect the  location
   of this town.
                                     E-17

-------
            Table E-4   PUBLIC EXPOSURE FOR LOW-ARSENIC COPPER SMELTERS
                      AS PRODUCED BY THE HUMAN EXPOSURE MODEL
Concentration
Level (ng/m3)
5.81
5.00
o c
C. , -J
1 0
J. • V
0 5
V • 
-------
E.4  QUANTITATIVE EXPRESSIONS OF PUBLIC  CANCER  RISKS  FROM  INORGANIC  ARSENIC
     EMITTED FROM LOW-ARSENIC PRIMARY COPPER  SMELTERS
E.4.1  Methodology (General)
E.4.1.1  The Two Basic Types  of Risk.  Two basic types  of  risk are dealt with
in the analysis.  "Aggregate  risk" applies to all  of  the people encompassed
by the particular analysis.  Aggregate risk can be related to a single
source, to all of the sources in a source category,  or  to  all of the source
categories analyzed.  Aggregate risk is expressed as  incidences of cancer
among all of the people included in the analysis, after 70 years of exposure.
For statistical convenience,  it i: often divided by 70  and expressed as
cancer incidences per year.  "Individual risk" applies  to  the person or
persons estimated to live in the area of the highest ambient air concentrations
and it applies to the single source associated with this estimate as estimated
by  the dispersion model.  Individual risk is expressed as  "maximum lifetime
risk" and reflects  the probability of getting cancer if one were continuously
exposed  to  the  estimated maximum ambient air concentration for 70 years.
E.4.1.2  The  Calculation of Aggregate Risk.  Aggregate risk  is calculated by
multiplying the total exposure  produced by HEM  (for  a single  source, a
category of sources, or all  categories of sources) by the unit risk estimate.
The product is  cancer  incidences  among  the included  population after 70
years  of exposure.   The total exposure, as calculated by HEM,  is  illustrated
by the  following equation:
                                      N
                      Total Exposure  =  I    (PjCi)
                                    E-19

-------
where
     I  = summation over all grid points where exposure is calculated,
     P-J = population associated with grid point i,
     C-j = long-term average inorganic arsenic concentration at grid point i,
     N  = number of grid points to 2.8 kilometers and number of ED/BG
          centroids between 2.8 and 20 kilometers of each source.
To more clearly represent the concept of calculating aggregate risk,  a
simplified example illustrating the concept follows:
                                  EXAMPLE
     This example uses assumptions rather than actual  data and uses only
three levels of exposure rather than the large number produced by  HEM.   The
assumed unit risk estimate is 3 x 10~3 at 1 ug/m3,  and the assumed
exposures are:
            ambient air                      number of people exposed
          concentrations                      to given concentration
          2   ug/m3                                      1,000
          1   pg/m3                                     10,000
          0.5 ug/m3                                    100,000
The probability of getting cancer if continuously exposed to the assumed
concentrations  for 70 years is given by:
   concentration                unit risk               probability of  cancer
                                                              6 x  10~3
                                                              3 x  10-3
                                                            1.5 x  10-3
2
1
0.5
pg/m3
pg/m3
pg/m3
x
x
x
3 x 10-3(ug/m3H
3 x 10~3
3 x 10-3
                                   E-20

-------
The 70 year cancer incidence among  the people  exposed  to  these  concentrations
is given by:
                                                            cancer incidences
    probability of cancer          number of people  at       after 70 years
    at each exposure level          each exposure  level           of exposure
                                                                     6
                                                                    30
                                                                   150
6
3
1.5
x lO-3
x 10-3
x 10-3
x
X
X
1,000
10,000
100,000
                                                          TOTAL =  186
The aggregate risk, or total cancer incidence, is 186 and, expressed
as cancer incidence per year, is 186 * 70, or 2.7 cancers per year.   The
total cancer incidence and cancers per year apply to the total of 111,000
people assumed to be exposed to the given concentrations.
E.4.1.3  The Calculation of Individual Risk.  Individual risk, expressed as
"maximum lifetime risk," is calculated by multiplying the highest concentration
to which the public is exposed, as reported by HEM, by the unit risk estimate.
The  product, a probability of getting cancer, applies to the  number of
people which HEM reports as being exposed to the highest listed concentration.
The  concept involved is a simple proportioning from the 1 iig/m3 on which
the  unit risk estimate is based to the highest listed concentration.  In
other words:
       maximum lifetime risk          the unit risk estimate
      highest concentration  to    =            1 ug/m3
      which  people  are exposed
                                    E-21

-------
E.4.2  Risks Calculated  for  Emissions of Inorganic Arsenic from Low-Arsenic
       Primary Copper Smelters
     The explained methodologies  for calculating maximum lifetime risk and
cancer incidences  were applied  to each  low-arsenic primly copper smelter,
assuming a baseline level  of emissions.  A baseline level of emissions means
the level  of emissions after the  application of controls either currently
in place or required to  be in place to  comply with curent State or Federal
regulations but before application of controls that would be required by a
NESHAP.
     Table E-5 summarizes the calculated  risks.  To understand the relevance
of these numbers,  one should refer to the  analytical  uncertainties discussed
in Section E.5 below.
                                   E-22

-------
                                          (Assuming Baseline Controls)
ro
co

Plant
1
2
3
4
5
6
7
8
9
10*
11
12
13
14

1 Maximum Lifetime Risk
1.1 x 10-3 . !.8 x 10-2
2.3 x 10-3 _ 3-6 x 10-2
4.3 x 10-4 . 69 x 10-3
4.1 x 10~5 - 6.6 x.10-4
4.3 x 10-3 _ 6.9 x 10-2
1.6 x 10-6 _ 2.6 x 10-5
1.5 x 10-4 . 2.4 x 10-3
9.4 x 10-4 . 1-5 x 10-2
1.5 x 10-4 . 2.5 x 10-3
1.4 x 10~6 - 2.2 x 10-5
1.5 x 10-5 . 2.4 x 10-4
3.6 x 10-5 _ 5.8 x 10-4
4.8 x 10-4 . 7.7 x 10-3
7.1 x 10-5 . !.! x 10-3
I
1 Cancer Incidences Per Year
7.1 x 10-2 . !.! x 10*
6.0 x 10-3 . 9.5 x 10-2
1.6 x 10-3 . 2.6 x 10-2
1.2 x 10-4 _ 1.9 x 10-3
6.4 x 10-3 _ i.o x 10-1
6.4 x 10-4 _ j.o x 10-2
8.2 x 10-4 . 1.3 x 10-2
3.0 x 10-3 _ 4.8 x 10-2
5.4 x 10-3 . 8.6 x 10-2
5.3 x 10-6 . 8.5 x 10-5
4.3 x 10-5 . 7.0 x iQ-4
3.8 x 10-4 _ 6.1 x 10-3
3.8 x 10-3 . 6.1 x 10-2
3.6 x 10-4 _ 5.8 x 10-3
Cancer Incidence
	 (one case In [x] years)
1 In 14 yrs. - 1
1 In 200 yrs. - 1
1 In 600 yrs. - 1
1 In 9000 yrs. - 1
1 1n 200 yrs. - 1
1 In 2000 yrs. - 1
1 In 1000 yrs. - 1
1 In 300 yrs. - 1
1 In 200 yrs. - 1
1 In 200.000 yrs. - 1
1 In 20,000 yrs. - 1
1 In 3000 yrs. - 1
1 In 300 yrs. - 1
1 In 3000 yrs. - 1
In 1 yrs.
L In 10 yrs.
In 40 yrs.
In 500 yrs.
In 10 yrs.
In 100 yrs.
In 80 yrs.
In 20 yrs.
In 10 yrs.
In 10,000 yrs.
In 2000 yrs.
In 200 yrs.
In 20 yrs.
In 200 yrs.
TOTALS FOR THIS SOURCE CATGEGORY




Number Total Number
of of People Exposed
Plants (within 20 km)

14 650,200
Highest Individual Risk


4.3 x 10-3 . 6.9 x 10-2
(For Plant 5)
Cancer Incidences

	 per year 	
0.1 - 1.6

one case In [x] years
1 In 10 yrs. - 1 1n 1 yrs



-------
E 5  ANALYTICAL UNCERTAINTIES APPLICABLE  TO  THE  CALCULATIONS  OF  PUBLIC
     HEALTH RISKS CONTAINED IN THIS  APPENDIX
E.5.1  The Unit Risk Estimate
     The procedure used to develop the unit risk estimate is  described in
Reference 2.  The model used and its application to epidemiological  data
have been the subjects of  substantial comment by health scientists.  The
uncertainties are too complex to be summarized sensibly in this appendix.
Readers who wish to  go beyond the information presented in the  reference
 should see  the  following Federal Register notices:   (1) OSHA's  "Supplemental
 Statement of  Reasons for the Fina.  Rule", 48 FR  1864 (January 14, 1983);
 and (2)  EPA's "Water Quality Documents Availability" 45 FR 79318
 (November 28, 1980).
      The unit risk estimate used in this analysis applies only  to lung
 cancer.   Other health effects are possible; these include skin  cancer,
 hyperkeratosis, peripheral neuropathy,  growth retardation and brain
 dysfunction  among children, and increase in adverse birth outcomes.  No
 numerical  expressions of  risks  relevant to these health effects are
 included  in  this analysis.
 E.5.2  Public  Exposure
  E.5.2.1   General.   The basic assumptions  implicit  in the methodology  are  that
  all  exposure occurs at people's residences,  that people  stay at  the  same
  location for 70 years,  that the ambient air concentrations  and the emissions
  which cause these concentrations persist for 70 years,  and  that the concentration
  are  the same inside and outside the residences.  From this  it can be seen
  that public exposure is based on a hypothetical rather than a realistic premise.
                                     E-24

-------
  It  is  not known whether  this  results  in an over-estimation or an under-
  estimation of public exposure.
  E.5.2.2  The Public.  The following are relevant to the public as dealt
  with in this analysis:
      1.  Studies show that all people are not equally susceptible to cancer.
  There is no numerical recognition of the "most susceptible" subset of the
  population exposed.
      2.  Studies indicate that whether or not exposure to a particular
 carcinogen results  in cancer may be affected  by the person's  exposure to
 other substances.   The public's exposure to other substances  is  not
 numerically considered.
      3.   Some members of the public included  in this  analysis  are  likely to
 be exposed  to inorganic  arsenic in  the air  in  the workplace, and workplace
                               t
 air concentrations  of a  pollutant are  customarily much  higher  than  the
 concentrations found  in  the  ambient, or  public  air.  Workplace exposures
 are  not  numerically approximated.
     4.  Studies show that there  is  normally a  long latent period between
 exposure and  the onset of lung cancer.  This has not been numerically
 recognized.
     5.  The people dealt with in the analysis are not located by actual
 residences.  As explained previously, they are "located" in the Bureau of
 Census data for 1970 by population centroids of census districts.   Further,
 the locations of these centroids have not been changed to reflect the 1980
 census.   The effect is that the actual  locations of residences  with respect
to the estimated  ambient  air  concentrations  are not  known and  that  the relative
                                  E-25

-------
locations used in the exposure  model  have  changed  since  the  1970  census.
     6.   Many people dealt with in  this  analysis are  subject to exposure  to
ambient air concentrations of inorganic  arsenic where they travel  and  shop
(as in downtown areas and suburban  shopping  centers), where  they  congregate
(as in public parks, sports stadiums,  and  schoolyards),  and  where they work
outside (as mailmen, milkmen, and construction workers).  These types  of
exposures are not numerically dealt with.
E.5.2.3.  The Ambient Air Concentrations
     The following are relevant to  the estimated ambient air concentrations
of inorganic arsenic used in this analysis:
     1.  Flat terrain was assumed in the dispersion model.   Concentrations
much higher than those estimated would result if emissions  impact on elevated
terrain or tall buildings near a plant.
     2.  The estimated concentrations do not account for the additive impact
of emissions from plants located close to  one another.
     3.  The increase in concentrations that could result from re-entrainment
of arsenic-bearing dust from, e.g., city streets,  dirt roads, and vacant
lots, is not considered.
     4.  Meteorological data specific to plant sites are not used in the
dispersion model.  As explained, HEM uses  the meteorological data from the
STAR station nearest the plant site.  Site-specific meteorological data
could result in significantly different estimates, e.g., the estimates of
where the higher concentrations occur.
     5.  With few exceptions, the arsenic emission rates are based on
assumptions  rather  than on emission tests.  See the BID for details on each
source.
                                   E-26

-------
E.6  REFERENCES

1.  National  Academy of Sciences,  "Arsenic,"  Committee  on Medical  and
    Biological  Effects  of Environmental  Pollutants,  Washington,  D.C.,  1977.
    Docket Number (OAQPS 79-8)  II-A-3.

2.  The Carcinogen Assessment Group's  Final  Risk  Assessment  on Arsenic.
    OAQPS Docket Number 79-8-1I-A-7.   May 2,  1980.

3.  U.S. EPA, et.al., "Environmental  Cancer  and Heart and Lung Disease,"
    Fifth Annual Report to Congress by the Task Force on Environmental Cancer
    and Health and Lung Disease,  August 1982.

4.  U.S. EPA, "Health Assessment  Document for Acrylonitrile," Draft Report
    from the Office of Health and Assessment,  EPA-600/8-82-007,  November
    1982.

5.  U.S. EPA, "An Assessment of the Health Effects  of Arsenic,"  Docket
    Number (OAQPS 79-8) II-A-5, April  1978.

6.  Busse, A.D. and Zimmerman.  J.R.,  "User's Guide for the  Climatological
    Dispersion Model."   (Prepared for the U.S. Environmental Protection Agency,
    Research Triangle Park, North Carolina.)   Publication Number EPA-R4-73-
    024.  December 1973.

7.  Systems Application, Inc., "Human Exposure to Atmospheric Concentrations
    of Selected Chemicals."  (Prepared for the U.S.  Environmental  Protection
    Agency, Research Triangle Park, North Carolina).  Volume I,  Publication
    Number EPA-2/250-1, and Volume II, Publication  Number EPA-1/250-2.
                                   E-27

-------
                 APPENDIX F






ARSENIC DISTRIBUTION AT U.S. COPPER SMELTERS
                    F-l

-------
              ARSENIC DISTRIBUTION AT U.S.  COPPER  SMELTERS

     Appendix F presents information regarding the distribution  of
arsenic in domestic copper smelters.  Information  was  received  in
response to requests made by EPA.   However,  for certain copper  smelters,
the information received was incomplete.   In these cases, assumptions
were made regarding the behavior of inorganic  arsenic.  The  bases  for
these assumptions are discussed for each  smelter.   Based on  the  infor-
mation provided and engineering judgment,  inorganic arsenic  mass balances
were developed for each copper smelter.
F.I  ARSENIC DISTRIBUTION AT ASARCO-E1  PASO
     An arsenic mass balance for the ASARCO-E1  Paso smelter  is  presented
in Figure F-l.  Data were provided to EPA  by ASARCO;1  however, a few
assumptions were necessary to obtain a closed  mass balance.  The arsenic
collection efficiency for the spray chamber  and electrostatic preci-
pilator systems treating the roaster and converter offgases  is 96  percent,
based on review of test data.  Also, arsenic collection efficiency  for
the acid plant treating the roaster and converter  offgases is
99 percent.  In addition to the dust collected  in  the  roaster and
reverberatory furnace collection system,  11  Ib/hr  of dust collected in
the converter collection system is recycled  back to the roasters.
F.2  ARSENIC DISTRIBUTION AT ASARCO-HAYDEN
     ASARCO is in the process of modernizing the Hayden plant.  The
existing conventional technology consisting  of  roaster, reverberatory
furnace, and converter equipment is being  replaced by  INCO flash fur-
nace technology.  The arsenic mass balance for  the Hayden copper smelter
presented in Figure F-2 represents the  flash furnace technology.
The data on arsenic distributions  for the Hayden smelter were available onV
for the existing smelter configuration.  Therefore,  an arsenic  balance
                                  F-2

-------
DUST
312 LB/HR
                                                                                  TO STACK
                                                                                  0.07 LB/HR
SPEISS
106 LB/HR
                                                                                 SLAG TO FURNACE
                                                                                 23 LB/HR
           Figure  F-l(a).   Arsenic Distribution at  ASARCO-E1  Paso Smelter

-------
                TO STACK  |
                3.2 LB/HR
| DUST TO LEAD PLANT
 162 LB/HR
ARSENIC INPUT
 314 LB/HR
                             SMELTER
                                                      BLISTER
                                                      37 LB/HR
             SLAG TO DUMP
             119 LB/HR
 ACID PLANT WASTE
 3 LB/HR
  Figure  F-l(b).  Overall  Arsenic Material  Balance  at
                    ASARCO-E1  Paso  Smelter
                                  F-4

-------
                        ESP
RECYCLE
OUST
333 LB/H
       '708 LB/HR
 ARSENIC INPUT
 375 LB/HR
                             346 LB/HR
                   INCO FLASH
                      FURNACE
                                                      14 LB/HR
                            SLAG TO DUMP
                             219 LB/HR
                                                                                   TO STACK
                                                                                   0.27 LB/HR
                                                                          ACID PLANT
                                                                          .ACID SLUDGE
                                                                                            17 LB/HR
                                                                                 4 LB/HR
                                                                             ESP
                                                                         DUST TO
                                                                         LEAD PLANT
                                                                                           95 LB/HR
                                                               99 LB/HR
                                                     CONVERTERS
                                                                       BLISTER
                                                                                         44 LB/HR
F-2(a).    Arsenic Distribution at
                                                             ASARCO-Hayden  Smelter
                                                  F-5

-------
    DUST TO LEAD PLAJT

       95 LB/HR
ARSENIC INPUT
 375 LB/HR
                                      TO STACK
                                      0.2 LB/HR
 ACID PLANT EFFLUENTS
„ 17 LB/HR
                                  SMELTER
                                                          BLISTER
                                                          44 LB/HR
                SLAG TO DUMP
                219 LB/HR
     Figure  F-2(b).   Overall Arsenic Material  Balance  at
                       ASARCO-Hayden Smelter
                                       F-6

-------
is developed for the smelter by using the arsenic  input  rate  to
the smelter (375 lb/hr)l and the percentages  of  input  arsenic in  the
various process streams for the flash furnace smelting technology
at the Kennecott-Hurley smelter (discussed in Section  F.7).2
F.3  ARSENIC DISTRIBUTION AT TENNESSEE CHEMICAL  CO.-COPPERHILL
     An arsenic mass balance for the Tennessee Chemical  Co.-Copperhill
smelter is presented in Figure F-3.   Tennessee Chemical  Company provided
flowsheets and information to EPA.3   Based on actual tonnages processed
and typical analyses, the actual arsenic  input to  the  smelter was
estimated to be 2.9 Ib/hr.
F.4  ARSENIC DISTRIBUTION AT INSPIRATION  CONSOLIDATED
     An arsenic material balance for the  Inspiration Consolidated
Copper Company smelter located near  Miami, Arizona, is presented  in
Figure F-4.  The data were provided  to EPA by the  company.^
     Arsenic input to the smelter is 41.1 Ib/hr.   This is based on
1,495 tpd of concentrate containing  0.033 percent  arsenic fed to  the
electric furnace.  Nineteen percent  of the arsenic into  the smelter is
volatilized in the electric furnace  and 6 percent  is volatilized  in the
converters.  Also, 58 percent of the input arsenic into  the smelter is
slagged in the electric furnace and  17 percent is  slagged in  the
converters.
F.5.  ARSENIC DISTRIBUTION AT KENNECOTT-GARFIELD
     An arsenic mass balance for the Kennecott-Garfield  smelter is
presented in Figure F-5.  Mass balance data were provided to  EPA  by
Kennecott Copper Company in response to information requests.2,5,6
     The arsenic material  balance for the smelter  is obtained by
combining the arsenic rates of various smelter streams (in flow diagrams
provided by Kennecott6) with the feed and end product  rates provided
by the company.  The following feed  and end product arsenic contents
provided by the company are used in  the mass  balance:  total arsenic
input - 261 Ib/hr; slag tailings - 34 Ib/hr;  precipitator dust to
storage - 123 Ib/hr; blister copper  - 29  Ib/hr;  gas cleaning  plant
effluent - 71 Ib/hr; and stack emissions  - 4  Ib/hr.
                                  F-7

-------
 I
CO
                                    OUST TO WASTE TREATMEKT
                                           0.65 LB/HR
                                             DUCON
                                            SCRUBBER
RECYCLED
SLURRY
0.7 LB/HR
                                                 0.5 LB/HR
                                3.6
                              ,, LB/HR
                       ARSENIC
                       INPUT
                       2.9 LB/HR
         FLUID BED
         ROASTER
         3.4
:ALCINE   LB/HR
                                                                      TO STACK
                                                                      0 LB/HR
                                                               ACID PLANT
                                                                      0 LB/HR
                                                                  ESP
                                                                      0.65 LB/HR
                                                                0.8 LB/HR
                                                               1.0 LB/HR
                          J.I LB/Hf
                                                                  TO STACK
                                                                  0 LB/HR
                                                                                        ACID PLANT
                                                                  0 LB/HR
                                                                                              ESP
                                                                                                          DUST TO WASTE TREATMENT
                                                                                                          0.65 LB/HR
                                                                  0.65 LB/HR
                                                           0.5 LB/HR
                                                          VENTURI
                                                         SCRUBBER
                                                                                                                                TO WASTE
                                                                                                                                TREATMENT
                                                                                                                                0.6 LB/HR
ELECTRIC
FURNACE
                                                               MATTE  1.5 LB/HR
                                                              1.1 LB/HR
CONVERTERS
   (2)
                                                                                _r\
                                                                                                BLISTER
                                                                                                                               0.1 LB/HR
                                                                                                   0.3 LB/HR
                                                                                       SLAG TO FURNACE
                                                                                    SLAG TO DUMP
                                                                                    0.9 LB/HR
                                       Figure  F-3(a).   Arsenic Distribution at TN Chemical  Co  -
                                                               Copperhill  Smelter

-------
ARSENIC INPUT
2.9 LB/HR
                                     TO STACK
                                     0 LB/HR
                                SMELTER
             GAS CLEANING
             CIRCUIT WASTE
              1.9 IB/HR
                                                        BLISTER
0.1 LB/HR
                                             SLAG TO DUMP
                                             0.9 LB/HR
  Figure F-3(b).   Overall  Arsenic  Material Balance at
           TN  Chemical  Co.-Copperhill Smelter
                                 F-9

-------
      DUST TO FURNACE
      1.2 LB/HR
      DUST TO FURNACE

      2.9 LB/lfli
  RECYCLED DUST
   4.1 LB/HR
               52.9
              LB/HR
ARSENIC INPUT
41.7 LB/HR
                                                                                 TO STACK
                                                                                 0.1 LB/HR
    ESP
    (2)
                 5.8 LB/HR
CYCLONES
   (2)
9.1  LB/HR
                                9.9 LB/HR
                           ELECTRIC
                           FURNACE
                                             MATTE 12.0 LB/HR
              RECYCLED SLAG
              7.7 LB/HR
                                 SLAG TO DUMP
                                ,-31.0 LB/HR
                                                                        ACID PLANT
                                                                                          ACID PLANT WASTE
                                                                                          9.0 LB/HR
                                                        3.3 LB/HR
                                                 CONVERTERS
                                                    (4)
                                                                   BLISTER
                                  1.6 LB/HR
                                     REVERTS
                                     0.6 LB/HR
                       SLAG TO FURNACE
                      " 7.7 LB/HR
      Figure F-4(a).   Arsenic Distribution  at  Inspiration-Miami  Smelter
                                                       F-10

-------
ARSENIC INPUT
41.7 LB/HR
                             TO STACK
                             0.1 LB/HR
•> ACID PLANT WASTE
  9.0 LB/HR
                                  SMELTER
                                                        BLISTER
                                                        1.6 LB/HR
                                       SLAG TO DUMP
                                       31.0 LB/HR
  Figure  F-4(b).   Overall Arsenic Material  Balance at
                  Inspiration-Miami  Smelter
                                    F-ll

-------
253.2
LB/HR
             45.4
             LB/HR
             23.6
            .LB/HR
            68.7 LB/HR
             360.8
LB7HR
       38.9
      .LB/HR
             CYCLONES
               184.3 LB/HR
                              229.7 LB/HR
              WASTE
              HEAT
             BOILER
    253.3 LB/HR         7.8 LB/HR
SMELTING
 VESSELS
  (3)
                                         34.6 LB/HR   .   42.4 LB/HR
                 SLAG 72.9 LB/HR
                                           SLAG
                                           MILL
                                                                              TO STACK
                                                                              4 LB/HR
                                                                       ACID PLANT
                                                                            ACID PLANT WASTE
                                                                              75 LB/HR
                                                                        ESPs (6)
                                                              DUST TO STORAGE
                                                                                        123 LB/HR
                                                                 13.7 LB/HR
CONVERTERS
    (6)
                                                                                        BLISTER 29 LB/HR
                                                    SLAG
                                                    0 LB/HR
                                           SLAG TAILINGS  34 LB/HR
        Figure  F-5(a).   Arsenic Distribution  at  Kennecott-Garfield Smelter
                                                     F-12

-------
                   STACK EMISSIONS
                     4 LB/HR
GAS CLEANING PLANT
    EFFLUENT
   71  LB/HR
INPUT 261 LB/HR
                               SMELTER
                    SLAG TAILINGS
                     34 LB/HR
                                                       BLISTER 29 LB/HR
  PRECIPITATOR
  DUST STORAGE
   123 LB/HR
 Figure F-5(b).   Overall  Arsenic  Material  Balance  at
                 Kennecott-Garfield  Smelter
                                  F-13

-------
F.6  ARSENIC DISTRIBUTION AT KENNECOTT-HAYDEN
     Kennecott Copper Company provided EPA with arsenic distribution
data for the Hayden smelter in March 1983.2  These data, however,  did
not give a closed material balance,  with 13 percent of the total  input
arsenic unaccounted for.  Additionally, insufficient data were provided
to calculate arsenic rates in various process streams.  Therefore,  only
the smelter input arsenic rate from the March 1983 data provided  by
Kennecott was used in developing the arsenic mass balance data.   Infor-
mation on distribution of arsenic was obtained from earlier information5*6
submitted by Kennecott.  The arsenic mass balance for the Kennecott-Hayden
smelter is presented in Figure F-6.
     The arsenic rates for the various smelter streams shown in the
material balance were proportionately adjusted to reflect the recent
data on feed rate indicated by the company.  The difference between the
adjusted arsenic rates and the arsenic rates provided by Kennecott  were
not significant for some streams.
F.7  ARSENIC DISTRIBUTION AT KENNECOTT-HURLEY
     The arsenic mass data provided  by Kennecott Copper Company for the
Hurley smelter are presented in Figure F-7.2  The data represent  the
arsenic distribution for the flash furnace technology to be installed
at the smelter.
     The arsenic input to the smelter is 2.14 Ib/hr.  The data indicate
that 49 percent of the total  arsenic input including the recycle  material
is volatilized in the flash furnace, and 14 percent is volatilized  in
the converters.  Also, 31 percent of the total  arsenic is slagged  in
the flash furnace.
F.8.  ARSENIC DISTRIBUTION AT KENNECOTT-McGILL
     An arsenic mass balance for the Kennecott-McGill smelter is
presented in Figure F-8.  Data for the mass balance were provided
to EPA by the Kennecott Copper Company;2 however, a few assumptions
were necessary to obtain a closed balance.  An arsenic collection
efficiency of 30 percent was used for the reverberatory furnace
electrostatic precipitator.  The final  arsenic distribution
indicates that 21 percent of the input arsenic to the smelter is
                                  F-14

-------
en
                    ACID SLUDGE
                     1.1 LB/HR
                               3.7
                               .B/HR
                         9.7 LB/HR
                          27.4
                          LB/HR
                 ARSENIC INPUT
                 17.7 LB/HR
                                                TO1STACK
                                                0.01 LB/HR
                                       ACID PLANT
                                                           0.7 LB/HR
                                               0.4 LB/HR
                                        SCRUBBER
       4.7
       LB/HR
                                              4.1 LB/HR
                                        FLUID BED
                                        ROASTERS
23.3
                                                        LB/HR  , LB/HR
                                                                             7.01B/HR
                                                                9.1  LB/HR
       32.4
                       T
                                                                               S V.
                                                                               f V
                                                                                     6.0 LB/HR
                                                                               ESP
                                                                                      4.4 LB/HR
                                                                                 11.7 LB/HR
                REVERBERATORY
                 FURNACE
                                                                                                               SCRUBBER
                                          SLUDGE TO
                                          ROASTERS
15.9 LB/HR
                                                                                 SLAG TO DUMP
                                                                                 4.8 LB/HR
                                                                                                                    6.7 LB/HR
                                                                                                                  ESP
                                                                                                                    11.1  LB/HR
CONVERTERS
                                                                     BLISTER
                                                                      0.2
                                                                      LB/HR
                                                           SLAG TO DUMP
                                                           4.6 LB/HR
                               Figure F-6(a).   Arsenic Distribution at Kennecott-Hayden  Smelter

-------
               TO MAIN STACK
               7.0 LB/HR
± ACID PLANT STACK
  0.01 LB/HR
ARSENIC INPUT
 17.7 LB/HR
                                   SMELTER
                                                          BLISTER
                                                          0.2 LB/HR
                                        SLAG TO DUMP
                                        9.4 LB/HR
     Figure  F-6(b).   Overall  Arsenic  Material Balance at
                     Kennecott-Hayden Smelter
                                       F-16

-------
ARSENIC
INPUT
2.14 LB/HR
                                                                            ACID
                                                                                          t-
                                                                                          l-°-
                                                           STACK
                                                           01 LB/HR
                 FILTRATE

                      0.03 LB/HR

                    r 1	
                                           S02 IlIQUIFICATION
                                                PLANT
RECYCLE OUST  TO FURNACE  0.78 LB/HR
                  RECYCLE
                  SLUDGE
                               2.35 LB/HR
        WET GAS
        CLEANING
        SYSTEM
                                                            :0.05 LB/HR
                      RECYCLE DUST
                      2.32 LB/HR
               2.35 LB/HR
                                              SETTLING
                                              CHAMBER
                                 RECYCLE  DUST
                                 0.44 LB/HR
               2.79 LB/HR
                                 5.68 LB/HR
         INCO  FLASH
          FURNACE
                                                                                    ACID PLANT
                                                                 SLAG 0.59 LB/HR
                        MATTE  1.73 LB/HR
                                                                                                   SLUDGE
                                                                                                   0.01 LB/HR
                                                        =0.01 LB/HR
                                                                                         ESP
                                                                                         (3)
                                                        0.79 LB/HR



                                                               BLISTER
CONVERTERS
   (4)	
                                                               0.35 LB/HR
                                                   SLAG TO DUMP
                                                   1.75 LB/HR
          Figure  F-7(a).   Arsenic  Distribution  at  Kennecott-Hurley  Smelter

-------
                                       '' TO STACK
                                         0.001 LB/HR
ARSENIC INPUT
 2.14 LB/HR
                                     SMELTER
                                                             BLISTER 0.35 LB/HR
                  SLAG TO DUMP
                  1.75 LB/HR
WET SLUDGE
0.04 LB/HR
    Figure  F-7(b).   Overall  Arsenic Material  Balance at
                    Kennecott-Hurley Smelter
                                      F-18

-------
 I

LO
                     165.1  LB/HR ,   181.9
                                  12.1 LB/HR
                                    DUST
                                   0.7 LB/HR
                         INPUT     LB/HR
                                                        TO STACK
                                                       J00.9 LB/HR
                                                       28.2 LB/HR
                                                     ESP
                                                      40.3 LB/HR
   WASTE
   HEAT
  BOILERS
      41 LB/HR
REVERBERATORY
   FURNACE
                                                     SLAG
                                                     42.7 LB/HR
                                                                                         DOST
                                                                                       3.0 LB/HR
                                                                                        DUST
                                                                                      2.8 LB/HR
                                                                             MATTE
                                                                           74.7 LB/HR
                                                                                                            72.7 LB/HR
                                                      MULTICLONES
                                                                                                           75.7 LB/HR
BALLOON
 FLUE
                                                                                                           78.5 LB/HR
                                                     CONVERTERS
                                                      FLUX
                                                      14.2 LB/HR
                                                                                                                     BLISTER
                                                                                                                     7.2 LB/HR
         SLAG
         TO
         FURNACE
         6.1 LB/HR
                               Figure F-8(a).   Arsenic  Distribution at  Kennecott-McGill  Smelter

-------
                                      TO STACK
                                     129.4 LB/HR
  INPUT
179.3 LB/HR
SMELTER
                      BLISTER 7.2 LB/HR
                                          SLAG TO
                                          DUMP
                                         6.1 LB/HR
      Figure F-8(b).  Overall  Arsenic  Material Balance  at
                      Kennecott-McGill Smelter
                                       F-20

-------
  volatilized  in  the  reverberatory  furnace, 22 percent  reports  in the
  slag,  and  40 percent  is  volatilized  in the converter.
  F.9  ARSENIC  DISTRIBUTION AT MAGMA COPPER COMPANY-SAN MANUEL
      Magma Copper Company provided to EPA material balance data based
  on the arsenic  content of the concentrate blend expected to be used in
  the future.7'8  This  information, obtained through EPA information
  request letters, was  used to develop the arsenic mass balance for the Magma
  smelter given in Figure F-9.
  F.10  ARSENIC DISTRIBUTION AT PHELPS DODGE-AJO
      An arsenic mass balance for the Phelps  Dodge-Ajo smelter is
  presented in Figure F-10.  This balance is developed  based on the
  arsenic input rate to the smelter provided by Phelps  Dodge Corporation,9
 and on arsenic distributions determined for  various streams from EPA's'
  letter of information request to Phelps Dodge.10  This material  balance
 assumed a 25  percent collection efficiency for  the reverberatory furnace
 ESP and that  the furnace  offgases  from  the ESP  are treated in  the
 converter acid plant.   Phelps  Dodge  Corporation  provided  EPA with test
 data  indicating  69 percent efficiency for  the reverberatory  furnace
 ESP.   Development of a closed  arsenic material  balance was  not possible
 using  the  arsenic distribution  data  indicated in Figure F-10 and the
 ESP precipitator efficiency  data provided  by the company.  Therefore,
 the efficiency provided by Phelps Dodge was not used in the material'
 balance.
 F.ll  ARSENIC DISTRIBUTION AT PHELPS DODGE-DOUGLAS
     Information currently available from the smelter** indicates that
 approximately 24 Ib/hr of arsenic is input with  the concentrate to  the
 roasters.  The weight percent of arsenic in the  concentrate is approximately
 0.03,  and the total  feed to the roaster  is about 1,200 tpd.  The  arsenic
 profile for Phelps Dodge-Douglas was  obtained from the smelter; however,
 the information given was  limited,  and a complete arsenic  balance was
therefore not  possible.
                                 F-21

-------
DUST TO REVERB

1.21 LB/HR    '
                6.14
             '   16/HR
 ARSENIC INPUT
 4.39 LB/HR
                                   TO STACK
                                   2.78 LB/HR
      ESP
                                  3.99 LB/HR
REVERBERATORS
  FURNACES
     (3)
                                              DUST TO REVERB
                                               0.27 LB/HR
                         MATTE
                                                1.34 LB/HR
                                                                           TO STACKS
                                                                           0.007 LB/HR
                                                                      ACID PLANT
                                                                                     ACID PLANT WASTE
                                                                                     0.663 LB/HR
                                                                         ESP
                                                                           0.94 LB/HR
CONVERTERS
    (6)
                                                         0.27 LB/HR
                                                                                      BLISTER
                                                                                      0.13 LB/HR
                                                                            RECYCLED SLAG
                                                                            TO REVERB
                                   SLAG TO DUMP
                                   0.81 LB/HR
         Figure  F-9(a).   Arsenic Distribution at Magma Copper  Company-
                                    San  Manuel  Smelter
                                                   F-22

-------
ARSENIC INPUT
 4.39 LB/HR
                 TO STACK
                  2.79 LB/HR
, ACID PLANT HASTE
 0.66 LB/HR
                                    SMELTER
                                                         BLISTER
                                                         0.13 LB/HR
                                      SLAG TO DUMP
                                      0.81 LB/HR
   Figure F-9(b).   Overall Arsenic  Material Balance  at
         Magma Copper Company-San Manuel Smelter
                                  F-23

-------
I
ro
                            DUST TO FURNACE
                                  24.8 LB/HR
                                  1.1  LB/HR
                                 25.9 LB/HR
                                       130.9
                                       LB/HR
                    ARSENIC INPUT
                    103 LB/HR
                                                     ESP
                                                      J\
                                                         100 LB/HR
REVERBERATORY
0, SPRINKLE
 FURNACE
                                      RECYCLED SLAG
                                      2.0 LB/HR
                                                                        75.2 LB/HR
                                                                       OUST TO FURNACE
                                                                         9.1  LB/HR
          SLAG TO DUMP
          22.2 LB/HR
                                                                                                            TO STACK
                                                                                                            8.2 LB/HR
                                                                                                  ACID PLANT
                                                                                                                      ACID PLANT HASTE
                                                                                                                      70.9 LB/HR
                                                                                                           79.1  LB/HR
                                                                                                           3.9 LB/HR
                                                                                                         ESP
                                                                                                           5.0 LB/HR
02 ENRICHED

 CONVERTERS
                                                                                                                       2.0 LB/HR
                                                                                                             SLAG TO FURNACE
                                                                                                             2.0 LB/HR
                             Figure  F-10(a).   Arsenic  Distribution at  Phelps  Dodge-Ajo  Smelter

-------
                   TO STACK
                   8.2 LB/HR
ARSENIC INPUT
103 LB/HR
4 ACID PLANT WASTE
 70.9 LB/HR
                                     SMELTER
                                                            BLISTER
                                                            2.0 LB/HR
                                         SLAG TO DUMP
                                         22.2 LB/HR
     Figure F-10(b).   Overall  Arsenic Material  Balance  at
                      Phelps Dodge-Ajo  Smelter
                                     F-25

-------
     The arsenic distribution  obtained  at  the Bor, Yugoslavia, smelter
is for a similar smelting  configuration;^ however, there  is  variation
in the percent arsenic  volatilized  or slagged in the  furnace  and converter
depending upon the amount  of  arsenic  volatilized in the  multi-hearth
roaster.
     To obtain the arsenic mass  balance shown in Figure  F-ll,  it was
assumed that 17.5 percent  of  the arsenic is volatilized  in the roaster,
since this is the average  of  the range  obtained at the Bor.^ However,
45 percent of the arsenic  was  assumed to be slagged in the reverbera-
tory furance and 7 percent was assumed  to be volatilized.   This distribution
was used because of the high  amount of  arsenic  reported  by the smelter
to be found in the reverberatory slag.   In the  converter it was assumed
that 23.2 percent of the input arsenic  is volatilized and 5.8 percent
reports in the slag.  Only 1.5 percent  of the input arsenic was assumed
to remain in the blister copper.
     The collection efficiency of the hot electrostatic  precipitators
was assumed to be 30 percent.
     The arsenic mass balance obtained  showed that  about 9.4  Ib/hr of
arsenic is emitted to the atmosphere.   About 13.7  Ib/hr  of arsenic is
removed with the reverberatory slag,  and 0.5 Ib/hr  of arsenic reports
in the  blister copper.
F.12  ARSENIC DISTRIBUTION AT PHELPS DODGE-HIDALGO
     An arsenic mass balance for the Phelps Dodge-Hildago smelter  is
presented in Figure F-12.  In April 1983, Phelps  Dodge  provided  overall
mass balance data for the smelter.9  However, the  data  were insufficient
to develop the material balance for each piece of  process equipment.
Therefore, the arsenic balance shown in Figure F-12 was developed by
using the arsenic input rate to the smelter provided in April 1983 by
Phelps  Dodge  and the arsenic distribution  in various streams from EPA's
information  request sent  to Phelps Dodge.^
F.13  ARSENIC DISTRIBUTION AT PHELPS DODGE-MORENCI
     An arsenic  mass balance for the Phelps Dodge-Morenci smelter is
presented in  Figure F-13.  The  information was provided to
EPA  by  Phelp  Dodge Corporation.9
                                  F-26

-------
ro
                                         /v.
                            0.6 LB/HR
                           2.1 LB/HR
                           1.5
                           LB/HR
                                                   DUST TO ROASTER
                                        J\_
                                          3.4 LB/HR
                                        ESP
                           8.2
                           B/HR
                 ARSENIC
                 INPUT
                 24.0 LB/HR
                                          4.9 LB/HR
MULTI-HEARTH
ROASTERS
     (24)
25.1
 LB/HR
                 23.3
                  LB/HR
                                                                              TO STACK
                                                                              4.9 LB/HR
                                                                             1.5 LB/HR
                                                                          ESP
                                                                            \_
                                                    DUST TO ROASTER
                                            2.1 LB/HR
REVER8ERATORY
  FURNACES
    (3)
                                                                         J\
                                                                                       MATTE
                              9.3 LB/HR
                                                                                           1.8 LB/HR
                                                                             SLAG TO DUMP
                                                                             13.7 LB/HR
                                                                                                               JO STACK
                                                                                                               4.9 LB/HR
                                                                                                            ESP
                                                                                                              7.0 LB/HR
                                               CONVERTERS
                                                  (4)
                                                                                                                        BLISTER
                                                   0.5 LB/HR
                                                                                                              SLAG TO REVERB
                         Figure  F-ll(a).   Arsenic  Distribution  at  Phel
                                                    ps  Dodge-Douglas  Smelter

-------
                                      TO STACK
                                      9.8 LB/HR
ARSENIC INPUT
 24 LB/HR
                                    SMELTER
                                                       BLISTER
                                                       0.5 LB/HR
                                       SLAG TO DUMP
                                       13.7 LB/HR
   Figure F-ll(b).   Overall  Arsenic Material Balance at
                 Phelps  Dodge-Douglas Smelter
                                     F-28

-------
        ACID PRODUCT
                  «

             0.32 LB/HR
                                  TO STACK
                                  0.5 LB/HR
ACID PLANT
       DUST TO FURNACE
             9.2 LB/HR
                    ACID PLANT
                    LIQUID EFFLUENT
                    21.8 LB/HR
                                  22.6 LB/HR
                                  21.2 LB/HR
                              ESP
             39.8 LB/HR
ARSENIC INPUT
30.6 LB/HR
                           DUST TO CONVERTERS
                                  30.1 LB/HR
                       FLASH  FURNACE
                       6.7 LB/HR
                                            MATTE
                    2.8 LB/HR
                                                            0.6 LB/HR
                                    4.8 LB/HR
                                                        1.4 LB/HR
                                                         1.4 LB/HR
                                                                             ESP
                                                        2.0 LB/HR
                                                                           CONVERTERS
                                                                                           BLISTER
1.4 LB/HR
                                                                                 1.4 LB/HR
    Figure  F-12(a).   Arsenic  Distribution  at  Phelps Dodge-Hidalgo  Smelter
                                                   F-29

-------
                    TO STACK
                    0.5 LB/HR
ACID PLANT WASTE
21.8 LB/HR
ARSENIC INPUT
 30.6 LB/HR
                                     SHELTER
                                                           ACID PRODUCT 0.32 LB/HR
                                                           BLISTER  1.4 LB/HR
                                         SLAG TO DUMP
                                         6.7 LB/HR
      Figure F-12(b).   Overall Arsenic Material  Balance  at
                    Phelps Dodge-Hidalgo Smelter
                                        F-30

-------
 I
CO
                       DUST TO FURNACE
                               2.84 LB/HR
                    ARSENIC INPUT
                    9.64 LB/HR
                                                ESP
                                           OXYGEN
                                           SMELTING
                                           FURNACE
                                 SLAG TO DUMP
                                 8.49 LB/HR
                                                               OFF6AS TO GAS CLEANING PLANT
                                                                6.63  LB/HR
        FLUX       BLISTER
      0.35 LB/HR    1.20 LB/HR
                                                                        CONVERTERS
                                                                                           11.73
                                                                                           LB/HR
                                                              MATTE
14.69 LB/HR
             SLAG
             2.11 LB/HR

    OUST TO FURNACE
                                                                                                     ESP
                                                                                                        8.21 LB/HR
                                                                                                          TO STACK
                                                                                                          0.30 LB/HR
                                   ACIP PLANT
     3.52 LB/HR
                                                                                  RECYCLE TO FURNACE
                                                                                    14.54 LB/HR
                          Figure  F-13(a).   Arsenic  Distribution at  Phelps  Dodge-Morenci  Smelter

-------
                                        ''TO STACK
                                         0.30 LB/HR
ARSENIC INPUT

9.99 LB/HR
SHELTER
                      BLISTER
                      1.20 LB/HR
                                          SLAG TO DUMP
                                          8.49 LB/HR
    Figure F-13(b).  Overall  Arsenic Material Balance  at
                  Phelps  Dodge-Morenci Smelter
                                       F-32

-------
 F.14   ARSENIC DISTRIBUTION AT COPPER RANGE
      The Copper Range Copper Company furnished EPA with information for
 the White  Pine smelter;13 however, from those data 46 percent of the
 arsenic input was unaccountable.  The Copper Range Copper Company
 subsequently advised EPA14 of corrections necessary to obtain a closed
 arsenic material balance.  The arsenic mass balance presented in
 Figure F-14 reflects these corrections.
      Data  suggested by Copper Range Company as a starting point for the
 material balance were:

 Total  arsenic input to the furnace           1.52 Ib/hr
 Arsenic in flux, refining slag and
  soda slag to converters                    0.22 Ib/hr
               Total Input                   1.74 Ib/hr

    Arsenic in slag to dump                  0.50 Ib/hr
    Arsenic in blister copper                0.43 Ib/hr
 The above data indicate that 0.81 Ib/hr arsenic (i.e., the difference
 between the input of 1.74 Ib/hr and the output of 0.93 Ib/hr) occurs
 as emissions from the furnace and converters.   The material  balance
 provided by Copper Range indicated an arsenic  emission rate  of 0.3  Ib/hr
 from the reverberatory furnace,  a figure which EPA deemed  reasonable.
 Therefore,  the remaining 0.51 Ib/hr arsenic  (i.e., the difference
 between the total  estimated arsenic emission rate of  0.81  Ib/hr and the
 0.3 Ib/hr emission rate for the  furnace) was assumed  emitted from the
 converters.
     The Company-provided data indicate that 0.62 Ib/hr of arsenic  is
 recycled to the furnace.   Thus,  the recycled arsenic  (i.e.,  the amount
 captured in the waste  heat boiler and the electrostatic  precipitator
 system) and the assumed 0.3  Ib/hr arsenic emission rate  from the
 reverberatory  furnace  is  0.92 Ib/hr.   This value  and  other arsenic
 streams to  the  reverberatory  furnace  indicate  that  the  arsenic  content
of the matte must  also  be 0.92  Ib/hr.
                                  F-33

-------
              DUST
             0.62 LB/HR
ARSENIC INPUT , .2.14 LB/HR
 1.31 LB/HR
                              TO STACK
                             0.3 LB/HR
                               ESP
                                  0.92 LB/HR
REVERBERATORY
  FURNACES
                                  SLAG TO
                                  DUMP
                                  0.5 LB/HR
                                                 MATTE 0.92 LB/HR
                                                 TO STACK
                                                0.3 LB/HR
                                                                           BALLOON  FLUE
                                                                                           DUST
                                                                                           0.21  LB/HR
                                                                                  0.51 LB/HR
                                                                            CONVERTERS
                                                                FLUX
                                                                0.067 LB/HR
                                                                                           BLISTER
                                                                                           0.43 LB/HR
                                                     SLAG TO
                                                     FURNACE
                                                     0.2 LB/HR
                                                           REFINERY SLAG 0.1312 LB/HR
                                                           SODA SLAG  0.026 LB/HR
        Figure  F-14(a).   Arsenic  Distribution at Copper  Range  Company  Smelter
                                                       F-34

-------
                            TO STACK
                          0.06 LB/HR
    INPUT
  1.53 LB/HR
                           SMELTER
                                                 BLISTER
                                                 0.43 LB/HR
                         SLAG TO DUMP
                          0.5 LB/HR
Figure F-14(b).  Overall Arsenic Material  Balance at
              Copper  Range Company Smelter
                              F-35

-------
     The overall  arsenic  distribution  shown  in Figure F-14 shows
39 percent  of input  arsenic  (including that  recycled to the furnace and
that to the converters)  is  volatilized in  the converters.  Also, 21
percent of  the input arsenic  is  slagged  in the reverberatory furnace
and 8 percent is  slagged  in  the  coverters.   The  arsenic collection
efficiency  for the waste  heat boiler and electrostatic precipitator is
67 percent.  This efficiency  value  is  reasonable based on the actual
operating temperature and the control  efficiency data EPA collected
during source testing at  other smelters.
                                  F-36

-------
 F.15   REFERENCES

 1.  Letter and  attachments from M.O. Varner, ASARCO, Inc., to J.R.
     Farmer,  U.S. Environmental Protection Agency.  March 16, 1983.
     Response to Section 114 information request.


 2.  Letter and  attachments from R.A. Malone, Kennecott Minerals Company,
     to J.R.   Farmer, U.S. Environmental Protection Agency.  March 16,
     1983.   Response to Section 114 information request.

 5.  Letter and  attachments from 1.6. Pickering, Kennecott Copper
     Corporation, to D.R. Goodwin, U.S. Environmental Protection Agency.
     May 9, 1978.  Response to information request about arsenic
     distribution at Kennecott Copper Smelters.

 6.  Letter and  attachments from R.A. Malone, Kennecott Minerals Company,
     to D.R.  Goodwin, U.S. Environmental Protection Agency.  December 23,
     1983.   Response to Section 114 information request.

 7.  Letter and  attachments from J.H. Boyd, Magma Copper Company,
     to J.R.  Farmer, U.S. Environmental Protection Agency.  March 15,
     1983.   Response to Section 114 information request.

 8.  Letter and  attachments from D.C. Ridinger, Magma Copper Company, to
     D.R. Goodwin, U.S. Environmental Protection Agency.  April 4, 1983.
     Response to Section 114 information request.

 9.  Letter from L.R. Judd, Phelps Dodge Corporation to J.R. Farmer,
     U.S. Environmental Protection Agency.  April 7, 1983.  Response to
     Section  114 information request.

10.  Letter and  attachments from J.R. Farmer, U.S. Environmental Protection
     Agency,  to  Phelps Dodge Corporation.  1983.  Section 114 information
     request.

11.  Letter and  attachments from R.W. Pendleton, Phelps Dodge Corporation,
     to D.R.  Goodwin, U.S. Environmental Protection Agency.  June 2,
     1983.

12.  Stankovic,  D. "Air Pollution Caused by Metallurgy Assemblies in
     Bor."  Institute for Copper, Bor, Project No. 02-513-1.  U.S.
     Environmental Protection Agency.

13.  Letter and  attachments from J.W. Maksym, Copper Range Company, to
     J.R. Farmer, U.S. Environmental Protection Agency.  March 17, 1983.
     Response to Section 114 information request.
                                    F-37

-------
14.  Telecon.   Katari, V., Pacifir environmental Services, Inc., with
     J.W.  Maksym,  Copper Range Company.  Arsenic material balance for
     White Pine Copper Smelter.  March 31, 1983.
                                     F-38

-------
TREPOHT NO
 EPA-450/3-83-010a
. TITLE AND SUBTITLE
'DATA
i before completing)

        (3. RECIPIENT'S ACCESSION NO.'
                                                Primary
                                                Proposed
,,.,    -"-iORGANIZA I ION NAME AND ADDRESS'	
Jffice of Air  Quality Planning and Standards
 •S. Environmental Protection  Agency
^search Triangle  Park, North  Carolina  27711

          ^AGENCY
m  for Air Quality  Planning" ami bt<
ffice of Air, Noise, and Radiation
.S   Environmental Protection Agency
.esearch Triangle  Park.  North Carolina  27711
SUPPLEMENTARY NOTES—~	~~	
         5. REPORT DATE
            July  1983	
         ^PERFORMING ORGANIZATION CODE"


         8. PERFORMING ORGANIZATION REPORT NO.



        To. PROGRAM ELEMENT NO."


        Tl. CONTRACT/GRANT NO.'

          68-02-3060
          EPA/200/04
                      rced                     fron, new and
              	_
  	DESCRIPTORS
r pollution
zardous  air pollutant
llution  control
andards  of performance
organic  arsenic
imary copper smelters

TRIBUTION STATEMENT"
limited
rm 2220-1 (R.v 4_7
           4 7
                              KEY WORQS AND DOCUMENT ANALYSIS

                                             b.lDENTIFIERS/OPEN

                                               Air pollution  control
                                               Stationary sources
                                              . .bcURITY CLASS (This ReponJ
                                              Unclassified
                                               SECURITY CLASS (Th» p»g*)
                                              Unclassified
                    Jc.  COSATi Field/Group

                        13 B
                    21 NO. OF PAGES'"
                         467
                    22. PRICED
                                 .SOSSOLETE

-------
 United States
 Environmental Protection
 Agency
Office of Air, Noise, and Radiation
Office of Air Quality  Planning and Standards
Research Triangle Park NC 27711
Official Business
Penalty for Private Use
$300
                                             Publication (Mo  FPA 450 3 H'i Pl<
                                                                                                                                   Postage and
                                                                                                                                   Fees Paid
                                                                                                                                   Environmental
                                                                                                                                   Protection
                                                                                                                                   Agency
                                                                                                                                   EPA 335
                                                                               If voui ddclressi'-, 
-------