United States         Office of Water         March 1982
             Environmental Protection    Regulations and Standards (WH-553) EPA-440/4-85-005
             Agency           Washington DC 20460


             Water
&EPA      An Exposure
             and Risk Assessment
             for Arsenic

-------
                                     DISCLAIMER
This is a contractor's final report, which has been reviewed by the Monitoring and Data Support
Division, U.S. EPA.  The contents do not necessarily  reflect the views and policies of the U.S.
Environmental Protection Agency,  nor does mention of trade names or commercial products
constitute endorsement or recommendation for use.
           V
           v

-------
3U373-101
' REPORT DOCUMENTATION »• REPORT NO. z.
PAGE EPA-440/4-85-005
4. Title and Subtttl*
An Exposure and Risk Assessment for Arsenic
7. Author**) Scow, K.; Byrne, M. ; Goyer, M. ; Nelken, L. ; Perwak, J.;
Wood, M.; and Young, S. (ADL) Cruse, P. (Acurex Corporation)
9. Performing Organization Nam* and Addr**»
Arthur D. Little, Inc. Acurex Corporation
20 Acorn Park 485 Clyde Avenue
Cambridge, MA 02140 Mt. View, CA 94042
12. Sponsoring Organization Nam* and Address
Monitoring and Data Support Division
Office of Water Regulations and Standards
U.S. Environmental Protection Agency
Washington, D.C. 20460
3. Recipient* I Acc««»!on No.
s. Report Oat* Final Revision
March 1982
«.
«. Performing Organization R*pt. No.
10. Project/Task/Work Unit No.
11. ContracMO or Grant(G) No.
(C) C-68-01-6160
C-68-01-6017
(C)
13. Typ* of Raport * Period Covered
Final
14.
 S. Supplementary Notes
  Extensive Bibliographies
 «. Abttraet (Unite 200 word«)

  This report assesses the risk  of  exposure to arsenic.   This study  is part of a program
  to identify the  sources  of and  evaluate  exposure  to  129 priority pollutants.   The
  analysis  is based  on available  information from  government,  industry,  and technical
  publications assembled in March of 1981.

  The  assessment  includes  an   identification  of  releases  to  the  environment   during
  production, use,  or  disposal  of  the substance.   In addition, the  fate  of  arsenic  in
  the environment is  considered; ambient levels  to which various populations of  humans
  and  aquatic  life  are  exposed   are  reported.   Exposure   levels  are  estimated   and
  available data  on toxicity  are presented  and interpreted.   Information concerning  all
  of these  topics is combined in an assessment of  the  risks  of exposure  to  arsenic  for
  various subpopulations.
 . Document Analysis a. Descriptors
 Exposure
 Risk
 Water Pollution
 Air Pollution
 b. Identlfiers/Open-Ended Terms

 Pollutant Pathways
 Risk Assessment
  c. COSATI Held/Group

  Availability Statement
Effluents
Waste Disposal
Food Contamination
Toxic Diseases
Arsenic
 Release to Public
 »ANSI-Z39.18)
                          19. Security Class (This Report)
                                Unclassified
                                                      20. Security Class (This Page)
                                                            Unclassified
                        21. No. of Pages
                             299
                                                    22. Price
                                                      $23.50
                                      See Instruction* on Reverse

-------
                                            EPA-440/4-85-005
                                            March 1981
                                            (Revised March 1982)
        AN EXPOSURE AND RISK ASSESSMENT

                  FOR ARSENIC
                      by
                   Kate  Scow
 Melanie  Byrne, Muriel  Goyer,  Leslie Nelken,
 Joanne Perwak, Melba Wood  and Stewart  Young
            Arthur  D. Little,  Inc.

        U.S. EPA Contract 68-01-6160
               Patricia Cruse
             Acurex Corporation

        U.S. EPA Contract 68-01-6017
               Stephen Kroner
               Project Manager
    U.S.  Environmental Protection Agency
Monitoring and Data Support Division (WH-553)
  office of water Regulation

-------
                                            EPA-440/4-85-005
                                            March 1981
                                            (Revised March 1982)
        AN EXPOSURE AND RISK ASSESSMENT

                  FOR ARSENIC
                      by

                   Kate Scow
 Melanie  Byrne, Muriel Goyer,  Leslie Nelken,
 Joanne Perwak, Melba Wood  and Stewart  Young
           Arthur  D.  Little,  Inc.

        U.S.  EPA Contract 68-01-6160
               Patricia Cruse
             Acurex Corporation

        U.S. EPA Contract 68-01-6017
               Stephen Kroner
               Project Manager
    U.S.  Environmental Protection Agency
Monitoring and Data Support Division (WH-553)
  Office of Water Regulations and Standards
        Washington, D.C. 20460
  OFFICE OF WATER REGULATIONS AND STANDARDS
    OFFICE OF WATER AND WASTE MANAGEMENT
    U.S. ENVIRONMENTAL PROTECTION AGENCY
          WASHINGTON,  D.C. 20460
                  II. ~,
                  V
                  Cnioago, 1L

-------
                               FOREWORD
     Effective  regulatory  action  for  toxic   chemicals  requires  an
understanding of the human and environmental risks associated with the
manufacture, use,  and disposal  of  the chemical.  Assessment  of risk
requires  a  scientific judgment  about  the probability of harm to the
environment resulting from known or potential environmental concentra-
tions.   The risk  assessment process  integrates health  effects data
(e.g., carcinogenicity,  teratogenicity)  with  information on exposure.
The components of exposure include an evaluation of the sources of the
chemical, exposure pathways,  ambient  levels,  and an identification of
exposed populations including humans and aquatic life.

     This assessment  was performed  as part of  a program to determine
the  environmental  risks associated  with  current  use  and  disposal
patterns  for  65 chemicals and  classes of chemicals  (expanded to 129
"priority pollutants") named in the 1977 Clean Water Act.  It includes
an assessment of  risk for humans and aquatic life and  is  intended to
serve  as  a technical  basis  for developing  the most  appropriate and
effective strategy for mitigating these risks.

     This  document  is   a contractors'  final  report.    It  has  been
extensively reviewed  by  the  individual contractors ?nd by  the EPA at
several stages of completion.  Each  chapter of  the draft was reviewed
by members of the authoring contractor's senior technical staff (e.g.,
toxicologists, environmental  scientists)  who  had  not  previously been
directly  involved  in the work.   These  individuals  were selected  by
management  to  be  the technical peers  of the  chapter authors.   The
chapters were  comprehensively checked for uniformity in quality and
content by the contractor's editorial team, which also was responsible
for  the  production  of  the  final  report.  The contractor's  senior
project  management  subsequently reviewed  the  final report  in  its
entirety.

     At EPA a  senior staff member  was  responsible  for guiding the
contractors, reviewing the manuscripts, and soliciting comments, where
appropriate, from related programs  within EPA  (e.g., Office  of Toxic
Substances,   Research  and   Development,   Air   Programs,   Solid  and
Hazardous Waste,   etc.).  A  complete  draft  was  summarized  by  the
assigned  EPA  staff  member  and reviewed for   technical  and  policy
implications with the  Office Director (formerly the  Deputy Assistant
Administrator)  of Water  Regulations  and Standards.   Subsequent  revi-
sions were included in the final report.
                         Michael W.  Slimak,  Chief
                         Exposure Assessment Section
                         Monitoring  & Data Support Division (WH-553)
                         Office of Water Regulations and Standards

-------
                             TABLE OF CONTENTS



 LIST  OF  FIGURES

 LIST  OF  TABLES

 ACKNOWLEDGEMENTS                                                     xv
1.0  TECHNICAL  SUMMARY
                                                                     1-1
1.1  Introduction                                                    i_i
1.2  Risk to Humans                                                  i_i
     1.2.1  Cancer Risk Based on Taiwan Study                        1-1
     1.2.2  Other Health Effects                                     1-2
     1.2.3  Exposure Levels                                          1_2
     1.2.4  Risk Based on Other Studies                              1-3
     1.2.5  Relationship Between Environmental Fate of
            Arsenic and High Exposure Levels                         1-3
1.3  Risk to Nonhuman Biota                                          1_4
1.4  Materials Balance of Arsenic                                    1-5

2.0  INTRODUCTION                                                    2-1

3.0  MATERIALS BALANCE                                               3_!

3.1  Introduction                                                    2-1
3.2  Materials Balance                                               3_1
3.3  Production and Inadvertent Sources                              3-4
     3.3.1  Production                                               3_4
            3.3.1.1  Environmental Releases of Arsenic from
                     Production                                      3_4
     3.3.2  Inadvertent Sources          -                            3_4
3.4  Uses of Arsenic                                                 3_g
     3.4.1  Production and Use of Pesticides                         3-6
            3.4.1.1  Monosodium Methanearsenate (MSMA) and
                     Disodium Methanearsenate (DSMA)                 3-6
            3.4.1.2  Arsenic Acid                                   3-14
            3.4.1.3  Cacodylic Acid                                 3_14
            3.4.1.4  Miscellaneous  Pesticides                       3-15
     3.4.2  Wood  Preservatives                                      3-15
     3.4.3  Glass Manufacture                                       3-16
     3.4.4  Alloys                                                   3-16
     3.4.5   Small-Volume Uses of Arsenic                             3-16
            3.4.5.1  Feed  Additives and Veterinary Chemicals         3-16
            3.4.5.2  Electronics  and Catalysts                       3-17

-------
                      TABLE OF CONTENTS  (Continued)
3.5  Disposal of Arsenic-Containing Wastes and Miscellaneous
     Sources                                                        3-17
     3.5.1  POTWs                                                   3_18
     3.5.2  Urban Refuse                                            3-18
     3.5.3  Urban Runoff                                            3-19
     3.5.4  Natural Loading                                         3-19
3.6  Summary                                                        3—?0

References                                                          3-21

4.0  FATE AND DISTRIBUTION IN THE ENVIRONMENT                       4-1

4.1  Introduction                                                   4_]_
4.2  Important Fate Processes                                       4_2
     4.2.1  Physicochemical Fate Processes                          4-2
            4.2.1.1  General Fate Discussion                        4-2
            4.2.1.2  Aqueous Chemistry                              4_7
            4.2.1.3  Sediment                                       4_9
            4.2.1.4  Soil Chemistry                                 4_9
     4.2.2  Biological Fate Processes                               4-14
            4.2.2.1  Microbial Biotransfonnations                   4-14
4.3  Major Environmental Pathways                                   4-17
     4.3.1  Pathway 1 - Atmospheric Emissions with Subsequent
                        Transfer to Water and Soil                  4-17
            4.3.1.1  Sources                                        4-17
            4.3.1.2  Atmospheric Transport                          4-17
            4.3.1.3  Summary                                        4-21
     4.3.2  Pathway 2 - Solid Waste Disposal and  Migration
                        into Groud Water                            4-21
            4.3.2.1  Sources                                        4-21
            4.3.2.2  Fly Ash Disposal                               4-21
            4.3.2.3  Mining Activity                                4-24
            4.3.2.4  Landfills                                      4-27
            4.3.2.5  Summary                                        4-30
     4.3.3  Pathway 3 - Direct Discharge to  Surface  Water           4-31
            4.3.3.1  Sources                                        4-31
            4.3.3.2  Field Studies of Arsenic in  Aquatic  Systems    4-31
            4.3.3.3  Bioaccumulation in  Aquatic Organisms           4-33
            4.3.3.4  Summary                                        4-37
     4.3.4  Pathway 4 - Wastewater Treatment                         4-37
            4.3.4.1  Sources                                        4-37
            4.3.4.2  Primary and Secondary Treatment                4-37
            4.3.4.3  Tertiary Treatment                              4-39
            4.3.4.4  Summary                                        4-41
                                  vi

-------
                       TABLE OF  CONTENTS  (Continued)

                                                                      Page

      4.3.5  Pathway 5 - Pesticidal Uses                              4_4l
             4.3.5.1  Introduction                                    /  /-,
             4.3.5.2  Field Studies on Arsenical Pesticides           4.41
             4.3.5.3  Phosphate Fertilizers                           4 45
             4.3.5.4  Plant Uptake                                    4~4^
             4.3.5.5  Application of Arsenical Herbicides to
                      Aquatic Systems                                 4_46
             4.3.5.6  Summary
 4.4  Distribution in the Environment                                 , ,(
      4.4.1  Natural Waters                                           .,*
      4.4.2  Effluent Waters                                           ~\8
      4.4.3  Sediment                                                   3
      4.4.4  Dissolved and Suspended Matters                          /,~rr
      4.4.5  Soil and Plants                                          ?"„
      4.4.6  Air                                                      4~55
 4.5   Overview                                                        4~59
      4.5.1  Important Fate Processes                                 4 g4
      4.5.2  Transfers from Air to Surface Water and Soil             4I64
      4.5.3  Releases to Land and Transfer to Ground Water            4-66
      4.5.4  Direct Discharges to. Surface Water and Wastewater
             Treatment                                                .  ,
                                                                     4-67
References

5.0  EFFECTS AND EXPOSURE - HUMANS

5.1  Human Toxicity
     5.1.1  Introduction
     5.1.2  Metabolism and Bioaccumulation                          51
            5.1.2.1  Absorption
            5.1.2.2  Distribution and Bioaccumulation                sI2
            5.1.2.3  Excretion                                       -
            5.1.2.4  Summary
     5.1.3  Human and Animal  Studies                                 l~I
            5.1.3.1  Carcinogenicity                                 ,-
            5.1.3.2  Teratogenicity                                   ~_l
            5.1.3.3  Mutagenicity
            5.1.3.4   Other  Toxic Effects                             r~7~
     5.1.4   Epidemiologic Studies                                     ~^
            5.1.4.1  Drinking Water Contamination                    5  Tc
            5.1.4.2  Food Contamination                              *~ia
            5.1.4.3  Medicinal Uses                                   ~^Q
    5.1.5   Overview                                                 ->-19
           5.1.5.1  Ambient Water Quality Criterion -
                    Human Health
           5.1.5.2  Additional Health Considerations               5^0
                                  vii

-------
                       TABLE OF CONTENTS  (Continued)
 5.2  Human Exposure
      5.2.1  Introduction                                             s~
      5.2.2  Ingestion                                                 ~
             5.2.2.1  Drinking Water                                  c 7
             5.2.2.2  Food*                                            ~_-
             5.2.2.3  Soil                                            ^J;
      5.2.3  Inhalation                                                ~^
      5.2.4  Dermal Contact                                            ~ '
      5.2.5  Exposure Incidences as Indicated by Human
             Monitoring Data                                         5_3{
      5.2.6  Forms of Arsenic Associated with Exposure Pathways      5-3?
      5.2.7  Summary
                                                                     5-4
-------
                        TABLE OF CONTENTS (Continued)
       7.1.5
 7.2
       7.1.6
              7.1.6.2
              7.1.6.3
              7.1.6.4
              7.1.6.5

              7.1.6.6
        Risk in Regard to Combined Exposures to Total
        Arsenic for Selected Subpopulations
        Risk Extrapolations  for Arsenical Skin Cancer
        7.1.6.1  Introduction
                 Data  and Discussion
                 Estimation  of  Human Risk
                 GAG Risk Estimate
                 Implications of Taiwan  Study  for  the
                 U.S.  Population
                 The Relationship between  Inhalation  of
                 Arsenic  and Carcinogenicity
       Areas for Future  Research
       Conclusions
Risk Considerations for Nonhuman Biota
7.2.1  Statement of Risk
       Background
       Local Regions of Potential Risk
       Aquatic Herbicide Use
       Sensitive Species
       Terrestrial Ecosystems
      7.1.7
      7.1.8
      7.2.2
      7.2.3
      7.2.4
      7.2.5
      7.2.6
References

Appendix A

Appendix B

Appendix C

Appendix D
  7-7
  7-7
  7-7
  7-9
  7-13
  7-16

  7-18

  7-20
  7-21
  7-21
 7-23
 7-23
 7-23
 7-24
 7-27
 7-27
 7-28

 7-30

 A-l

 B-l

 C-l

D-l
                                  ix

-------
                             LIST OF FIGURES

Figure
  No.                                                                page

 3-1      Environmental Releases of Arsenic:  Production, Use,
          and Inadvertent Sources                                   3-3

 4-1      Major Environmental Pathways of Arsenic Releases          4-3, 4-4

 4-2      Major Environmental Pathways of Arsenic Releases          4-5

 4-3      The Eh-pH Diagram for Arsenic in an Aqueous  System
          with Arsenic and Sulfur                                   4-8

 4-4      Local Cycle of Arsenic in a Stratified Lake                4-10

 4-5      Removal of As (III)  from Leachate Solutions  by
          Kaolinite and Montmorillonite Clays as a  Function
          of  pH                                                     4-12

 4-6      Removal of As (V)  from Leachate Solutions by
          Kaolinite and Montmorillonite Clays as a  Function
          of  pH                                                     4-13

 4-7      Arsenic Biological Cycle in Water                         4-15

 4-8      Schematic of Ash Disposal Area on the  Oak Ridge
          Reservation and Summary of the Ranges  of  Arsenic
          Concentrations  in Selected Samples                         4-23

 5-1      Regression Curves  for  Urinary Arsenic  vs.  Total Daily
          Arsenic Intake  via Drinking Water in Subjects in
          Arizona and Alaska                                        5-31

 7-1      Comparison of Aquatic  Exposure and  Effects Levels  -
          Total Arsenic                                             7-26

-------
                               LIST OF TABLES
Table
No.

3-1

3-2
3-3
3-4
3-5
3-6
4-1
4-2
4-3

4-4
4-5
4-6


4-7
4-8

4-9
4-10



Sources ReleaS6S fr°m Produ^ion, Use, and Inadvertent

Production of Arsenic, 1979
Inadvertent Sources of Arsenic, 1979
Inadvertent Releases of Arsenic from Mining and Milling
Estimated Environmental Releases of Arsenic from Use
Arsenic Releases from Production and Use of Pesticides
Arsenic Forms of Environmental Significance
Microbial Transformations of Arsenic Compounds
Total Arsenic Concentrations in Various Media in
Vicinity of a Copper Smelter

Arsenic in Ash Pond Effluents from 11 Plants "
Concentrations of Total and Soluble Arsenic Measured in
Soil Surrounding a Mine Tailings Disposal Site
, .
*** & ciT ic wOric 6n ti* rit"irtTiG i T^ o w* A • * « .j TT i_ <>«
^ v-^in-.La<-.Lunb in uround Water Near Industrial
isposal Sites
Arsenic Compounds in Tampa, Florida, Waters
Ambient Arsenic Concentrations in Fish Tissue-
Data in STORET, 1975-1979 ^^^ &
Occurrence of Arsenic in Industrial Wastewater
Arsenic Rpmm/ai v ££•;.,,• 	 _ /- ,,


!£!£

3-2
3-5
3-7
3-10
3-11
3-12
4-6
4-16

4-20
4-25
4-26


4-28
4-32

4-35
4-38

                                                                       4-40






                                                                       4-44
4"12      Sa'er^or^io1 1°^ ^^  Concent"^°ns  in Ambient
          water for Major River Basins and  the United States-

          Unremarked Data in STORET,  1975-1979        *>-*<-es

                                                                       4-49


4-13      Total Arsenic Concentrations Detected in Freshwater
                                                                      4-50
                                   xi

-------
                        LIST OF TABLES (Continued)
 Table
  No.
                                                                  Page

 4-14      Total  Arsenic  Concentrations  Detected  in  Saltwater       4-51

 4-15      Concentrations of Arsenic  in  Wellwater in Six
          Major  U.S.  River Basins—STORET, 1974-1979               4-52

 4-16      Arsenic Concentrations in  Effluent Waters-
          Unremarked  Data in STORET, 1975-1979                     4_54

 4-17      Arsenic Concentrations in  Sediment Core Samples
          from Various U.S. Lakes                    '              4-55

 4-18      Arsenic Residues in Water  and Sediment,
          American Falls Reservoir,  Idaho, 1974                    4-55

 4-19      Ambient Arsenic Concentrations in Dissolved
          and Suspended Matters, Unremarked Data in
          STORET, 1975-1979                                        4_57

 4-20     Concentrations of Arsenic Detected in Rocks and Soils    4-58

4-21     Concentrations of Arsenic Detected  in Coal              4-58

4-22     Concentrations of Arsenic in Foliage of Various
         Plant Species                                           4- 60

4-23     Concentrations of Arsenic in Soil Samples  From
         Sites Contaminated with Mine and Smelter Waste          4-61

 4-24      Concentration  and Species  of Arsenic in Water-Soluble
          Extracts of Soil Samples from Sites Contaminated
          With Mine and  Smelter Waste                              4-62

 4-25     Arsenic Concentrations Detected in the Atmosphere        4-53

 4-26      Chemical and Biological Transformation of
         Arsenicals in  Surface Water and Soil                     4-65

 5-1      Epidemiologic  Studies and Case Reports of  the
         Health Effects of Water-borne Arsenic Exposure          5-16

 5-2      Adverse Effects of Arsenicals on Mammals           •     5-21

5-3      Arsenic Levels Detected in Drinking Water  in  the U.S.   5-24

5-4      Arsenic Levels Detected in Non-potable Ground
         Water in the U.S.                                        5-25

5-5      Environmental Releases of Arsenic to Land; Estimated
         Annual Volume, Chamical Form and Geographic
         Distribution                                            5-28
                                  xii

-------
                        LIST  OF  TABLES  (Continued)
 Table
 No.
 5-6       Environmental Characteristics Favorable to the Migration
           of Arsenic in Soil to Ground Water                          5_29


 5-7       Arsenic Levels in Food                                      , ,.
                                                                       5-34

 5-8       Summary of Magnitude of Human Populations Exposed Via
           Inhalation of Arsenic Jby Selected Emission Sources          5-37


 5-9       Arsenic in Human Biological Media                           5_39


 5-10      Chemical Forms of Arsenic Associated with Different
           Exposure Media
                                                                       5-43

 5-11      Estimated Levels  of Human Exposure  to Arsenic                5.45


 6-1       Acute  Toxicity of Arsenic for Freshwater Fish                6_3


 6-2       Acute  Toxicity of Arsenic for Freshwater Invertebrates       6-4


 6-3       Acute  Toxicity of Arsenic  for Marine  Biota                   6_6


 6-4       Toxicity of Arsenicals to Terrestrial Plants                 6_8


 6-5        Effects of Arsenic on Microorganisms                         5_9


 6-6       Monitoring Stations in Upper Missouri River Basin

          Reporting High Mean Arsenic Concentrations (1975-1979)      6-14


 6-7       Examples of Terrestrial Sites of Significant Arsenic
          Exposure
                                                                      6-19
7-1
          Adverse Effects of Arsenicals on Mammals Expressed in
          Dose Equivalents


          Summary of Estimated Arsenic Exposure Levels Associate
          with Individual Exposure Routes              Associated
                                                           Human
                                                                      7-8

7-4       Jge and Sex-Specific Skin Cancer Incidence Rate for
          Study Area in Taiwan
                                                                      7-10

7-5
         Skin Cancer  Prevalence Rate Per Thousand

                            C<™ion *
                                                                     7-11


                                 xiii

-------
                      LIST OF TABLES  (Continued)


Table


7-6       Chemical Constituents of Taiwan Water Samples           7-12

7-7       Estimated Carcinogenic Response in Humans
          Exposed to Arsenic in Drinking Water                    7-15

7-8       Estimated Lifetime Excess Cancers per Million Popula-
          tion Exposed via Ingestion to Total Arsenic at Various
          Concentrations Based on Three Extrapolation Models      7-17

7-9       Incidence Rates Among Caucasians for Non-Melanoma
          Skin Cancers,  Melanoma of Skin, and Cancers of
          All Organs Combined for Four Areas of the
          United States                                            7-19

7-10      Ranges in Effects Levels Reported for Aquatic
          Species Grouped by  Arsenic Forms                       7-25
                                 xiv

-------
                              ACKNOWLEDGEMENTS
 Data).  In addition   Elizabeth r«i       -^          °d (Monitoring
                     '  fiJ-j-zaDetn Cole provided  cntict-ant--!-,1  •    *.
 human pff&fi-       •              >'j-'- t'j.wvj.ucu  auoscantiaj. inputs to the
 i ^«.j	i	. .      _ .  '          luung was responsible for t~ha om'^mm-t^
                                                                         .
         ~^^^
tion on behalf of A^ureJ              WM  resP°nsib1^ for report produc-


     Stepheu Kroner, MDSD,  was the EPA project manager for this report.

-------
                           1.0  TECHNICAL SUMMARY

   1.1   INTRODUCTION
       The Monitoring and Data  Support  Division,  Office  of Water Regula-
  tions and Standards, U.S. Environmental  Protection  Agencv is  conducing
  an ongoing program to  identify  the  sources  of and evaluate the exposure^
  to 129 priority pollutants.   This report assesses the  environmental
  exposure _ to and risk associated with  arsenic.   Environmental  releases
  of arsenic are both inadvertent, such as from fossil fuel combustion and
  metals production, and intentional  from industries  using  arsenic  in their
  products or processes.  Arsenic is  used in  the manufacture  of  numerous
  other product   Pr6SerVatives' *la«> »d  various metals  alloys, among


  1-2  RISK TO HUMANS:   EXPOSURE, EFFECTS AXD FATE CONSIDERATIONS

  1'2-1  Cancer Risk Based on Taiwan Study

       The  risk associated with  exposure of humans to  adverse levels of
  TllTl  ^  ^6 enyir°ninent is  ^certain.   Based  on extrapolations from
  Ld  r  /•1uemi    gy  StUdy  Sh°Wing  a  relationship between skin cancer
  and  low drinking water  concentrations  of  total arsenic, risk estimates
  of excess individual lifetime  skin cancer incidence  associated wS
  mgestion of  3 pg/1 of  arsenic in drinking  water rang-  from 0  rl^h
  of 0  0009 to  0.0018.  Three  micrograms per§  liter iH'coZn^ c ur^

  virv  hS rLC°nfe\tratl0n ln  W3ter  SUpPlies'  The implications of this
  ™JL  8 n"6 °f Skin  cancer  for ^e  U.S. must  be qualified by several
  arsen?; h   axtemPtS C° indUC£ tUffl°rS  ln laborat°ry  animals  exposed to
  arsenic have been unsuccessful;  2) the chemical  form of arsenic in US
  ground water may not be the  same form as found in the Taiwanese  study
  3) other potentially harmful chemicals were  present  in  the Taiwan  "round
 water (lysergic acid,ergotamines); 4) no significant increase  in"  Sneer
 was found in an epidemiology study of Lane County, Oregon  in which rela
  tively high arsenic levels are present in some of the water  supply-
 5)  althougn the non-melanoma skin cancer rate is  high in the U S   Vh*

 S^Hne  ^^ ^ ^  'T^ ^ dlfferent Pa"s of  the§ body than 'the
 Taiwanese cancers  and  there is  a stong relationship between incidence
 rates and  geographic  latitude (i.e.,  ultraviolet  radiation?   So  f
 even  without  considering the  inherently weaJ £ ^dem?ology siuJe  tn^ '
 selves  there  is a  considerable amount  of  uncertaintj afsocia^d with
 the  extrapolation of  the Taiwan skin  cancer  rate  to the  U.S  popuLtL.
           er  fmp°rtant  and  lar§ely  unquantifiable  source  of  uncertainty
There      appl\catlon of  the  risk extrapolation models  themselves      7
There  is  presently no scientific consensus  as  to which  is  the  most
appropriate   del.  Thus> the range Qf r.sk estimat^fb;\4hevamr7otus
models may under-or overestimate the actual risk to man.  Overestimate
appears more likely due  to the conservative assumptions utilized ir the
calculation of exosu                                   utilized in tne
calculation of exposure and its duration.
                                  1-1

-------
  1.2.2  Other Health Effects

       The risk associated with exposure  to  environmental  levels  of arsenic
  is  lower if  one  considers  effects  other than  developing  skin cancer  that
  are reported for chronic exposure  to  low levels  of various  forms  of
  arsenic  in laboratory  studies.  Peripheral neuropathy resulted  from
  exposure to  the  lowest reported effects level of 0.04 mg/kg/day of cal-
  cium arsenate  ingested for 2-3 weeks.   Other effects resulting  from
  exposure to  arsenate were observed at concentrations of  10 ms/kg  or hizher-
  concentrations of 23 mg/kg/day to  93 mg/kg/day did not elicit carclnogenesls
  in  rats.  Data for  arsenite were very limited; skin disorders resulted
  from exposure  to 0.13 mg/kg/day and no  evidence  of carcinogenesis  was
  found in rats  exposed to arsenite at levels of 13 mg/kg/day.  Data for
  arsenic  (unspecified), in addition to the Taiwan study,  indicated  central
  nervous  system damage and lethality in  infants exposed to 0.8 mg/Wday
  and  7.9  mg/kg/day,  respectively.   The U.S.  Interim Drinking Water  Standard
  for  total arsenic is 50 yg/1, which is equivalent to an intake  level of
  0.001 mg/kg/day from drinking water.   In general the data base  for  under-
  standing the toxicity of arsenic  and  its compounds is very poor.

  1-2.3  Exposure Levels

      Average exposure levels  of arsenic  for the  general U.S.  population
 tnrough ingestion are less  than 0.001 mg/kg for  drinking water,  total
 diet, and wine or moonshine.   Consumption of  contaminated fish or  ground
 water may result  in considerably  higher  levels exceeding  0.01 mg/kg/day
 However,  a low absorption efficiency  of  arsenic  from ingestion of  seafood
 has  been reported.   Dermal  exposure for  the majority  of  the  U.S. popula-
 tion is  expected  to  be  negligible.  Contact with  arsenic-treated wood
 and  contaminated  detergent  is  associated with  intake  levels  of 0 004
 mg/kg/day and 0.06 mg/kg/day,  respectively.  However, the subpopulation
 exposed  through these routes  is small.   Exposure  of the general  population
 through  inhalation of ambient  levels  of  arsenic is negligible and  less    "
 than 0.001 mg/kg/day.   Therefore the  highest exposure levels  for the
 general population are  from ingestion  routes.

      Distinguishing  between chemical  forms of arsenic associated with
 different exposure routes is complicated  due to a general  lack of  data
 poor analytical techniques, and the presence of multiple  forms in  most'
 media.  However some exposure media with  potentially high  arsenite  levels
 include well water,  wine, and some  fish  tissue.  Due to the uncertainty
 of these  data   risk  is estimated for exposure to total arsenic levels
 Therefore different  exposure  levels can  be combined into  exposure
 scenarios  for selected subpopulations  to  better represent  total  exposure
 to arsenic.

      The  total estimated exposure to arsenic for the general non-smoking
 population and for children is approximately 0.4  yg/kg/day and 8 yg/kg/
 day,   respectively.   Smokers  may be exposed to  1.7 yg/kg/day.   Subpopula-
 tions with diets including ground  water,  fish,  or wine highly contaminated
with  arsenic may be  exposed  to from 7  yg/kg/day to 143 yg/kg/day.  Woodworkers
                                  1-2

-------
   may be exposed to from 4 yg/kg/day to 7 ue/k^/Hsv   TT,

   assuming exposure to hlgh^tlt'tS LtStvels'of a^nic^'all
   media, has an associated exposure of 41 yg/kg/day.      arsenic « all


   -1-2-4  Risk Based on Other Studies
 1- 2'5  Relationship Between Environmental FP^ of
        and High Exposure Levels
                                                .               "'"
 ponds distributed nationally  with  fi^'  n°lude'  (1)  fly ash disposal
 Northeast;  (2)  pesticide  uie'wSJ asr  •^i!60""6'1  ±n the  indust^^l
 U.S.;  (3) production  of other  rnetaJf   ± 1 ^  rUth~Central  and eastera
 and  boron,  concentrated in  the Jested s?ata  7 f^f'  ^^ Md  St£el

                                            ^^^

levels reported in STORET.               "° CftC §enerall>'



                                 1-3

-------
      Food contamination can result from various environmental fate
 pathways for arsenic.  Although most crops usually have low arsenic
 residues, root crops, leafy vegetables, and grapes used for wine some-
 times have higher than average concentrations.  Potential industrial
 sources of these residues include atmospheric fallout in the immediate
 vicinity of smelters, root uptake from soils contaminated by the dis-
 posal of solid waste from coal combustion, smelting and other industrial
 processes.  However, this possibility is remote as crops are generally
 not cultivated in such areas.   A potentially more significant source
 is arsenical pesticide use either near food crops or on soils previously
 contaminated where food crops  are now grown.   Natural background levels"
 or arsenic may also be responsible for some contamination.

      The likelihood of contamination  of aquatic species which will be
 consumed by human populations  is  difficult to estimate because of a
 lack of inrormation about  national fish consumption patterns.   Both
 marine  and freshwater species,  especially shellfish,  living  in waters
 with typical ambient concentrations can bioaccumulate tissue  levels
 high enough to  contribute  to significant  exposure  levels in human
 consumers.   However,  there is  evidence  that only  a small fraction of
 the amount  ingested  is  actually absorbed.  The  total  amount of arsenic
 intentionally discharged to surface waters is quite  small.  Inadvertent
 sources  such as runoff  from pesticide use, POTW discharges, urban runoff
 and natural  contamination  are  probably  responsible for the levels  in
 surface  waters  in which aquatic life are  exposed.  Fate  considerations
 applicable  to the exposure of  fish to adverse levels  of  arsenic  are
 also relevant to bioaccumulation.

 1-3  Risk  to Nonhuman Biota:  Exposure, Effects
      and Fate Considerations

      Exposure to arsenic appears to present little risk  to aquatic life
 Freshwater acute toxicity  can occur at sodium arsenite concentrations
 as  low as 290 ug/1 in the  commonly found bluegill  sunfish.  Chronic
 toxicity data are limited  to one study on daphnia  finding approximately
 yuu  ug/1 to be toxic following  long-term exposure.  Ninety-nine percent
 of  the concentrations were less than 100 pg/1 in the 36,308 unremarked
 ambient  samples reported in STORET over the past 5 years for total
 arsenic.  A total of 51 observations were in excess of 1,000 pg/1 and
 were reported in three major river basins:  the Missouri, Upper Mississippi
 and the Colorado.   A more detailed examination of the Missouri basin data
 indicated that the high levels  were focused primarily in two areas  one
 associated with a mining facility and  the other with no obvious sources
nearby.   There are no instances of fish kills  on record within EPA
directly attributed to arsenic.

     Any apparent  risk suggested by comparison of  the monitoring data
 (for total As)  and laboratory  toxicity  data (for soluble As) would be
further  modified by  environmental  parameters.   Factors responsible for
 tne retention and  immobilization of arsenic in aquatic systems,  such as
                                  1-4

-------
    fo«ren?J °   yr°US °XideS' Clay sedi^nt Particulates, and so

    element      ^ **** tO ^"^ the bi'l°gical availability^ ! the



        The data on toxicity of arsenic to aquatic species are limited













  where atmospheric fallout  L            '               ^ frOB Smelters
 1<4  Materials Balance  of Ar
                            senic
  81%)
  (5,700 kkg) and 82  (t)   r     S"rf e "ater accomt  tor il
  annually/ The ratio of inJd^tlntT     "' .°f the t0tal releas«
        3.7 to 1 for all JSfCS^LIn^?"'"


                                             °f an
  ss 8:%- rScf p'Soc'ss^g        r>   nd ----
that industries with high S«ewatfr      Adelines data indicated
nonferrous metals  and ore Sning oJeraSon! J  ?  ?S -°f 3rSeniC inCl
                           nng oeraon        -
 -producers,  coal mining ,^^^L$£l^^
                                             °f —"- to 'h.

 to land.   Aimosaiur: ro raise3" cin^ondi0nal diSCh^es> «e made

 of some form of arsenic-containinTsoUd waste^T^e   C°nSUmerS  dlSP°Se
 total annual solid waste loading contributed h*  ^K  Percentage of the
 sources is as follows:  33% fro! fossil S  7  K   m°St si§nific^t
 cide use,  19% from copper productxon   J3 u£ ^us^on, 19%  from pesti-

and 5%  from boron productio'n.   ?h ^emaJni™ 11?°" *"* ^ ?™d^°*
other sources.                     remaining 11^ ls contributed by all
                               1-5

-------
     Less than 13% of the estimated 1,820 kkg of arsenic released each
year in POTW effluents can be accounted for; known releases to treatment
plants include veterinary chemical producers, a few metals production
plants and urban runoff.  Part of the remaining 87% may originate from
natural loading and atmospheric particulate fallout.  POTWs are not
consistently effective at reducing arsenic concentrations in water
presumably due to the low effectiveness of primary and secondary treat-
ment in removing arsenic.
                                  1-6

-------
                            2.0  INTRODUCTION

  a^ n    c      °^.Water Regulations and Standards (OWRS), Monitoring
  and Data Support Division, of the U.S. Environmental Protection Agency
  is conducting a program to evaluate the exposure to and risk of 129
  priority pollutants in the nation's environment.  The risks to be evalu-
  ated include potential harm to human beings and deleterious effects on
  fish and other biota.  The goal of the task under which this report has
  How/T    4S C° integrate formation on cultural and environmental
  flows of specific priority pollutants and to estimate the risk based on
                             substances-   The results are intended to serve

                                                         f°*
  St™rvf        ^  intended to  Provide  a brief,  but  comprehensive,
  summary  of  the  production,  use,  distribution,  fate, effects   exposure
  and  potential risks  of  arsenic.  Water-borne routes of  exposure  are   '
  stressed due to  the  emphasis of  the OWRS  on aquatic and water-related
  F*.?!?7!:   °ccuPational exposure and  the  exposure of  the general'popu-
  lation to atmospheric levels  of  arsenic are only considered in terms  of
  the perspective  they shed on  the magnitude of water-related exposurl

                a,number of Pr°blems with  attempting an  exposure and
                           in any analysis of discharges or runo    it
               0 diStin§Uish ^^ound concentrations or natural sources
     n        7SOUrces;   ^leases from human activities include some known
 purposeful releases;  however,  the majority of arsenic discharges are unS
 tional releases resulting from smelting and coal  combustion  processes   rne
 fore, these estimates have a great deal of uncertainty associated w?th  them
          nJC "  f°Und ln the environment  in numerous forms  with differ-
 assessm; Characteristics,  toxicology,  and exposures,   This  exposure
 assessment  focuses  on the  forms most prevalent  in water:  the  inorganic
 forms  of  arsenate  (pentavalent) and arsenite (trivalent), and  various







cases  the HrawineXP°fSUre *St±mateS f°r recePto" «* precludes in some
     Although human epidemiologic data from Taiwan strongly inmlicate
arsenic as a carcinogen to humans, no currently existing chronic 1-bor*
tory animal studies support this observation.  ' Due to the complLi^y o
                                 2-1

-------
establishing dose-response relationships from epidemiologic studies,
such as the inability to control all factors potentially contributing
to carcinogenicity and differences between human subpopulations,
the epidemiological data must be used cautiously.  Risk extrapolations
to the U.S. general population must be viewed in light of the limitations
and assumptions of the analysis.

     This report is organized as follows:

     •   Chapter 3.0 contains information  on releases from the
         production, use, and disposal of  arsenic,  including
         identification of the form and amounts released and
         the point of entry into the environment.

     •   Chapter 4.0 considers the fate of arsenic  in five
         specific pathways leading from point of entry into
         the environment until exposure of receptors and also
         reports available monitoring data of arsenic concen-
         trations detected in environmental media.

     •   Chapter 5.0 discusses the adverse effects  of arsenic
         and concentrations eliciting these effects  in humans
         and quantifies  the likely pathways and levels of
         human  exposure  to arsenic in various environmental
         media.

     •   Chapter 6.0 considers the effects  of arsenic on non-
         humans  and  quantifies the environmental exposure of
         aquatic and terrestrial nonhuman  biota to  the element
         in different media.

     •   Chapter 7.0 discusses risk considerations  for various
         subpopulations  of humans  and  aquatic organisms,  com-
         paring  estimated exposure levels  of  arsenic  and   the
         concentrations  responsible for  adverse  effects.

     •   Appendices  A, B,  and  C present  the assumptions  and
         calculations for the  estimated  environmental  releases
         of arsenic  described  in Chapter 3.0.  Appendix D
         includes  information  on the environmental  distri-
         bution  of arsenic  releases.
                                 2-2

-------
                           3.0.   MATERIALS BALANCE
  3.1  INTRODUCTION
       In this  chapter,  a materials balance is  developed for arsenic   The
  materials  balance  identifies  the  initial  point  of  entry of arsenic into
  the  environment  from all activities.   Potential sources of releases were
  xdentxfxed by a  review of the numerous processes involving arsenic from
  its  production or  importation and use  to  its  ultimate  disposal
   entif
 identified.
 in thiseat^
 in this type
 to
 to
 o?
 or
 rPlP                       °f arsenic release,  the amount of material
 released was estimated, and the environmental compartments  (air   land
        "    J^1^ reCeiVin§ ^ tr-sP-ting the material were
              Since arsenic has numerous chemical forms, the initial form
                    i1'611'1'16"-  TherS a" ^ ^"tiinties inherent
                 analysis.  Not all current releases have been identified
                6" ^ well-doc— ^d, and future releases are di ficult
              Nevertheless, sufficient information is available to indicate
              Sd?"1 T T'  magnitude>  and in — -ses the location   *
              loading of the environment with arsenic.
               ,.                           review of both Polished and
             data concerning production,  use,  and disoosal of arsenic
          •  ^f ' StateS'   AVailablS lite-ture has 'oeen critiqued and
          in order to present an overview of major sources of environ
            aS                                               '
 3-2   MATERIALS  BALANCE
                                        "e»and  (or arsenic »as approxt-
        consumptive uses of arsenic are presented
intluld     ^ 3dditi0n' rSleaSeS fr°m POTWS
                                                     Table 3-1
                                                        runoff
nroH     maj0rity °f environmental releases resulted from inadvertent
production sources and from the use of arsenic as a pesticide   Purpose
                                3-1

-------
           Table 3-1.   Arsenic  Releases from Production, Use, and Inadvertent Sources  (kkg As, 1973)
Source Quantity As
PRODUCTION
ASARCO, Tacoma3 6,700
Pesticides 10,250
Mood Preservatives 2,930
Glass Manufacture 730
Alloys 440
Other 290
INADVERTENT SOURCESd
Fossil Fuel Combustion5 17,000
Copper Production,
1° + 2° 9,900
Lead Production,
1° + 2° 990
Zinc Production
Iron and Steel
Aluminum Production
Boron Production 2,250
Phosphorous Production 860
Manganese Production 4,000
Antimony Production
Cotton Ginning
POTW
Urban Runoff f
TOTAL'
a) See Section 3.3.1 and Appendix A.
b) See Section 3 and Appendices A and B. Other
Estimated Environmental Release;
Water
Air Land
210
1,500
neg
10
c
2

2,000
1,100
230
280
55



10

300


5,700

1,200
8,100
neg
neg
c
10

14,000
8,100
1,100

5,700

2,200
640
1,400
neg
580
20

43,000

includes electronics, cat
Surface POTW
.1 ~,
720
neg
neg
c
<50

150
38
neg
560 1
9 6
180
4
160



1,800
1,050
2,370 1,857

alysts, feed additives
+ n *.«l*»ara a*«pan^/» +r\ 1
f
Total '
1,410
10,000
neg
10

62

17,000
9,300
1,300
830
5,700
160
2,200
800
1,440
neg
830
1,820
1,050
53,400

and
>ho
    veterinary chemicals.   Use  of  arsenic-containing products  is expected to release arsenic to the
    environment in  the form as  it  is  found  in  the product, i.e., use of MSMA would release arsenic to
    the environment as MSMA.  Likewise, wood preserving would  probably release arsenic as  chromated
    copper arsenate or fluorchrome arsenate phenol.  Glass and alloy manufacture would most likely
    release arsenic as the  trioxide.   The metallurgical inadvertent sources would probably also
    release arsenic as the  trioxide,  cotton ginning would release arsenic as the trioxide; cotton
    ginning would release arsenic  as  cacodylic acid (defoliant residue).

c)  Included in advertent sources.

d)  See Appendix A.

e)  See Appendix C.

 f)  Values  nay not ado due to  rounding.


                                               3-2

-------
CO
 I
     INADVERTENT
       SOURCES
                         PRODUCTION
                                                                   USE
                         ASARCO,  TACOHA
                            6,700
                             IMPORTS
                             8,9/10
                                              EXPORTS
                                               1,000
  PESTICIDES
  10,250
     WOOD
PRESERVATIVES
      2,930
                                                                  GLASS
                                                                     730
     ALLOYS
     440
                                                                  OTHER
                                                                    290a
                                                                 POTWs
                                                                 URBAN
                                                                 RUNOFF
                                                                         TOTAL
                                                                                                     AIR
                                                                                                      210
                                                                                                    1,500
neg
                                         10
                                      4,000

                                       neg



                                     5,700
                         Figure 3-1.  Environmental  Releases of Arsenic:  Production, Use, and  Inadvertent
                                      Sources (kkg arsenic, 1979)
                a)  Includes feed additives,  veterinary chemicals, electronics, and catalysts.
                *)  Included in inadvertent  sources, secondary metal production.
                                                 ESTIMATED ENVIRONMENTAL RELEASES
                                                                SURFACE
                                                                 WATER       POTW    TOTAL
                                                                                                                LAND
                                                                                                                 1,200
                                                    8,100
             neg
                                                                                                                   neg
              10

         34,000
             20



         43,000
                                                                                                                               720
                         neg
                                                                                                                               neg
1,100
1.800

1,050

4,670
                                                                             <50
                                                                                                                                           57
                                                                                     1,410
                                                                                                                                                 10.000
                                              neg
                                                                                       10
     62


39,000
 1.820

 1,050

53,400

-------
 3.3  PRODUCTION AND INADVERTENT SOURCES

      In 1979,  about 6,420  kkg of arsenic were contained in the trioxide
 produced domestically;  280 kkg were  recovered as  arsenic metal (Nelson
 1980).   As noted in Table  3-2,  approximately  8,940 kkg of arsenic were
 imported;  1,000 kkg were exported.   Thus,  the available domestic supply
 totaled 14,640  kkg  arsenic.

 3.3.1  Production

      Arsenic trioxide  (As203)  and arsenic  "metal"  are  recovered  only at
 one facility in the U.S.   The ASARCO smelter  in Tacoma,  Washington,
 produces arsenic from  flue dust captured during the smelting  of  copper
 ore.  Although  arsenic  is  a minor constituent  of most  nonferrous ores,
 their low  arsenic content  does  not warrant recovery.

      Arsenic trioxide is volatilized during smelting and  concentrates
 in  flue dusts.   The  dusts  are mixed  with pyrite or galena and  roasted;
 vapors  from the  roasting process are cooled and condensed  in a series'
 of  brick chambers yielding  a  90-95%  pure arsenic trioxide  (EPA 1979g) .
 The product is removed by  chipping the solid condensate  from the brick
 walls.   Resublimation in a  reverberatory furnace and recondensation
 provide a  product of even  higher purity (EPA 1976a).  To  produce arsenic
 metal,  the trioxide is reduced with  carbon (EPA 1979g).

 3.3.1.1 Environmental Releases of Arsenic from Production

      Of the total 1,410 kkg arsenic  released to the environment  from
 arsenic production, 15% (210 kkg) is  emitted to the atmosphere and 85%
 (1,200  kkg) is disposed of  on land (see Table  3-2).  Atmospheric emissions
 are largely due  to particulates that  escape control devices and, to a
 lesser  extent, transfer and handling  of materials  (EPA 1979a,b).  Land-
 destined arsenic  is chiefly contained in slag  (see Appendix A, Notes 1
 and 2).

     Arsenic recovery is a. dry operation; wastewater occurs only from
 daily wash down of the facility.  Based on a daily flow of 3,400 liters
 containing 310 mg arsenic per liter and assuming a 365 day per year
 operation, only 0.4 kkg of arsenic would be discharged from the produc-
 tion  site per year (EPA 1979b, EPA 1975a).  Such waters are directly
 discharged (EPA 1975a).

 3.2.2   Inadvertent Sources

     Arsenic is a constituent of most mineral ore;  processing of these
ores, as well as smelting and refining of the metal,  releases substantial
quantities of arsenic to the environment.   Cotton ginning and disposal  of
 the resulting wastes also introduce arsenic into the environment in the
 form of arsenical defoliant residue.   Combustion of fossil fuel for
space heating,  electricity  generation, and transportation releases
arsenic to the environment  as arsenic trioxide condensate on particulates
and as  arsenic-containing ash disposed in sluice ponds.  Derivations of
quantities of arsenic released to the environment via fossil fuel combustion

                                 3-4

-------
                                                     Table 3-2.  Production of Arsenic, 1979 (kkg)
Co
 I
Ln
                          Production
                                                          Quantity3
                                                                                      Estimated Environmental Releases
                                                                                                     Water
                                                                                                                    Total
                         ASARCO, Tacomab As20


                                         Asd

                         Imports0


                         Exports6


                           Available U.S. Supply
                                  6,420


                                    280


                                  8,940


                                  1,000


                                 14,640
-neg-


-neg-
                           a)  Quantity is kkg of As,  figures are rounded nearest 10 kkg.

                           b)  Nelson, 1980
    Iffii:             .
    production/release  estimates for the

d)  Nelson 1980.

e)  Loebenstein,  1980c.
                                                                                     l

                             1,410
                                                                                                                       -
                                                                                                                    "' "" """'
                         91  fss,:
                         »')  Based on EPA 1979a;  1976... and tPA. 1975.  See Append, x A,  Note  2.

                                                                                                                  is,.-  r " •'


-------
 are found in Appendix C.*  As noted in Tables 3-3 and 3-4, approximately
 87/, (34,147 kkg) of the total arsenic released from inadvertent sources'
 is contained in land-disposed metal processing slag, 10% (4,000 kk<0 is
 emitted to the atmosphere, and 2% (1,100) is discharged to water
 (derivations of these quantities are found in Appendix A).

      Arsenic is contained in most metal products, and as such, is not
 released to the environment from ordinary, usage.   However, arsine gas
 can be formed when arsenic-containing metal comes in contact with acid
 (i.e., pickling of metal)  (EPA 1976a).  Also, arsenic associated with
 phosphorous, ultimately utilized in detergents,  is apparently released
 to waterways.   Angino _et al.  (1979)  found that the 1-70 mg/kg arsenic  '
 in various detergents and  presoaks yielded 1-250  mg/1 arsenic in wash-
 water.

 3.4  USES  OF ARSENIC

      Current industrial uses  of  arsenic  include pesticides,  wood pre-
 servatives,  additives in glass manufacture,  nonferrous  alloys, and feed
 additives;  arsenic finds minor use in  the electronics and  chemical/cata-
 lysts  industries.   Discontinued  uses  include leather tanning chemicals
 and paint  pigment  additives  (EPA 1976a).   Demand  for arsenic is expected
 to increase  at  an  annual rate of 1% through  1985  (Loebenstein 1980) .

     As  is  the  case with arsenic production  data,  consumption data are
 considered proprietary information.  The  use distribution pattern
 presented  in Table 3-5 is  based  on previous  years' consumption data
 and  Bureau of Mines estimates (Loebenstein 1980).  By far, the largest
 use  of arsenic  (70% or 10,250 kkg) is  in  pesticide manufacture.

 3.4.1  Production  and  Use  of Pesticides

     Arsenic is a  component of insecticides  (Paris green, lead  arsenate,
 calcium  arsenate),  herbicides (disodium methanearsenate, monosodium
methanearsenate, cacodylic acid) and defoliants (arsenic acid).  Table
 3-6  gives quantities of  arsenic  contained in these pesticides  in 1979,
as well  as environmental releases  from the pesticides' production  and'
use.   Process flow  diagrams are presented in Appendix B; producers and
their  locations are given  in Table B-l.

3.4.1.1  Monosodium Methanearsenate (MSMA) and Disodium
         Methanearsenate (DSMA)

     DSMA and MSMA are used as contact herbicides  on weeds which are
especially difficult to control.   Both are registered for directed
application (no contact wich crop plants)  on citrus fruits and cotton.
DSMA is produced by reacting arsenic trioxide with sodium hydroxide,
then adding methyl chloride (EPA 1975d).   MSMA is  then manufactured 'bv
treating DSMA with sulfuric acid.

-Estimates of the  national annual arsenic emissions  from energy production
 range from 650 kkg/yr to 4,600  kkg/yr with the variation due to coal usage
 figures and the concentration of arsenic  in coal  which  is  highly variable
 across the country (see Note  p.4-58).   It should  be  noted  that quantification
 of exposure based on  these levels would  give significantly different results
 depending upon the estimates  used.

-------
Table 3-3.   Inadvertent  Sources  of Arsenic,  1979  (kkg)
" • " — 	 ^ 	 — 	 T
Source

	 .
Fossil Fuel Combustion (total)
Coal (total)3
Petroleum
Copper Production l°c
«ud
Lead Production l°e
2uf
Iron and Steel Production9
Zinc Production*1
Aluminum Production'
Boron Production1^
-Phosphorus Production**
Manganese Production1
Antimony Production1"
Cotton Ginning"
TOTAL

InPut Contained

Air
2,000
16.450 2tQOO
76 i.
10 74
9.850 3,070 i.ioo
neg neg
990 20 _ 230
neg
79,000 55
280

,250 20
860
4-°°° 2.160 10

300
/
4.000
Estimated

Land
14,000
14,000
2
6,730

760
300
5,700


2.200
640
1,400
ntg .
580
32,000
Environmental Releases
Water
Surface POTW
150
150
neg
44
neg neg
neg neg
neg neg
9 6
560 i
180
4
160


1.100 7

Total
17.000
17,000
76
7,874

990
300
b.700
H30
180
2.200
800
1.440
neg
880
38.100


-------
                                                 Table  3-3.   (Continued)


     a)  Includes external  combustion  and  space  heating.   See Appendix C for release  by  type  of  coal  and
         combustion.  Water  value  Is  acid  mlno  drainage  and  ashponds (94 kkg from mine  drainage,  52 from ash
         ponds); Land value  Includes  particulate  >3 \j.m which  settles to land.  See Table  3-4.   CPA, 1976a,  1975a.
         Blanks  Indicate  data  not  available.

     b)  Soo Appendix C.

     c)  EPA, I979a, 1975c.  Soo Table  A-3 and Appendix  A,  Note 6.   Includes 1,400 kkg As  sent  to  land  as refinery
         slag, the chief  source of  As  release.   Also  Includes arsenic release from copper  mining  and  milling (soo
         Table 3-4) from  copper mining,  (see Table 3-4)  110  kkg As  to air;  630 kkg As to  land;  14  kkg As to surface
         water.  EPA, 1976a.   See  Table  3-4.

     d)  Dased on 50 secondary copper  smelters,  average  As  effluent  of  0.02 kkg As/yr/plant.  EPA  I975b.  See
         Appendix A, Note 7.   Arsenic  concentration  In secondary  copper smelter feed  material  Is 30-1000 times  loss
         than that for primary smelters, emission  factors  are about  100 times less than primary; atmospheric
         omissions are negligible.  Arsenic  In slag  from  secondary  copper  production  total about 300  kkg.  See
y        Appendix A, Note 19.  EPA, 1980b.
CO
     e)  Atmospheric emissions based on  581,600 kkg  lead  produced  (Rathjen, 1980), as an  omission  factor of  0.4  kg
         As/kkg Pb (Davis,  1971).   Land  discharges based  on 410 kg  slag and 40 kg dry sludge/kkg  lead produced  and
         an arsenic content of 0.29? EPA,  1976a.  Wastewater  discharge  based on average As concentration for  4
         plants, and a total of 7  plants,  see Appendix A,  Note  8.  Contained quantity based on 581,600  kkg  produced
         x  35g As/kkg lead produced, EPA,  I976a.

     f)  EPA, 1980b.  Atmospheric  emissions are negligible due  to  low  arsenic content 10.4$) and processing
         temperatures (260-370'C)  which  are lower than the vaporization temperature  of arsenic (613*C).
         Alternatively,  applying the 0.4 kg As/kkg lead  produced  emission  factor  and 778,930 kkg lead produced  from
         second  smelters, 310 kkg  As are released per  year.   Wastewator data from EPA, I979b.  Based  on  69  plants x
         8.0 x 10   kkg  As/plant/yr.  See Appendix A,  Note 9.   Land  discharges  based on 148,300 kkg slag/yr
         from secondary  blast  furnaces and 0.2? As In  slag, EPA,  1976a.

     g)  See Appendix A,  Note  18.    Includes 3 kkg As omitted  to air  from ore mining,  see Table 3-4.

     h)  Wnstewater  values based on an average discharge of I kkg As/yr/plant EPA,  I979b,  see Appendix A, Note  II.
         There are  5 plants  discharging  directly and one plant  discharging  indirectly.  Atmospheric emissions based
         on 0.6  kg  As  emitfod/kkg  Zn processed (Davis, 1971)  x  472,480  kkg  Zn produced (Cammarota,  1980).   This
         estimate is  probably too  high duo to use of  unsubstantiated  emission factors.  EPA,  I976b estimates  that
         1he emissions  from  all primary  zinc smelters  total 80  kkg/yr.   This estimate, however,  Is  also
         unsubstantiated.  Water vaIuo Includes  540 kkg As released  1o  surface  waters  from load/zinc  mines,  soo

-------
                                                Table 3-3.   (Concluded)


                                       kka As "
     j)   Rased  on  EPA,  I976a.   See Appendix A,  Note 14.

     k)   Input  based  on  phosphate rock mine shipments  (P0n

                                                                                               "        -,, ,or
     I)  Based on  EPA,  ,976a.  by  ana.ogy  1o  phosphorus.   See Appendix A, Note  13
    n)  See Appendix A,  Note  15.
Co
 I

-------
                                      Table 3-4.  Inadvertent Releases of Arsenic from Mining and Milling, 1979  (kkg)
to
 I
Source
Copper3
Lead lb
Zinc J
Aluminum
Antimony
Coale
Iron ore
TOTAL
Production (kkg)
Air
62,180,500 (concentrate) 110
992,250 neg
1,960 neg
250 neg

87,087,500 (crude ore) 3
113
Estimated Environmental Releases
Water
Land Surface
630 14
550
180

94

630 830
POTW Total
754
550
180
neg neg
94
3
1,581
                        a)  Wastewater value based on As concentrations  and flow rates for copper mines and mills
                            EPA, 1975c.  See Appendix A Note 12.   Land discharge based on tailing pond concentrations,
                            see Appendix A, Note 12.  Atmospheric  emissions based on 248,722,000 kkg ore mined
                            (Butterman, 1980) x 0.45 x 106 kkg As  emitted/kkg ore (Davis, 1971).

                        b)  Wastewater discharge based on EPA 1978b average discharge rate of 0.044 kkg/day for 34
                            discharging lead/zinc mines operating  365 days  per year.   Atmospheric emissions based on
                            0.45 x 10"*  kkg As emitted/kkg ore. produced  and production of approximately 992,250 kkg
                            lead and zinc = 0.49 kkg (CPA, 1979; Ryan e_t aj. 1978; Cammarota 1978).

                        c)  Based on estimated annual discharge rates for the 2 bauxite ore processing facilities of 0.26
                            kkg As/day x 350 day/yr EPA, 1979e.   Atmospheric emissions based on 0.45 x 10"^  kkg/kkg ore
                            produced (EPA, 1979b) and ore production of  1,960 kkg (Kurtz, 1978)  = 0.001 kkg.

                        d)  EPA 1975c.  Wastewater discharge based on maximum flow from flotation mill of 342,990 I/day
                            350 d/year x 0.23 mq As/1 •= 0,02 kkg As.  Atmosnheric emissions based on 0.45 x 10"*  kkg As/kkg
                            ore produced x 250 kkg ore produced =  0.0001 1  kkg (EPA,  1979e; Rathjen, 1978).

                        e)  Based on EPA screening sampling data for coal mining.   Mine drainage contains 12  mg As/1 x
                            3,000 I/day/mine x 5,673 mines x 365 day/yr  = 94 kkg As.   EPA,  1979e.

                        0  EPA, 1973b; Peterson, 1980.   Rased on  an atmospheric emission factor of 0.1 kg As emitted/kkg
                            arsenic present in ore, average arsenic concentration of  400 mg/kg,  and 1979 crude ore
                            production of 87,087,500 kkg ore.

-------
Table 3-5.  Estimated Environmental Releases of Arsenic from Use, 1979
                                                                                                     {kkg)
— — 	 — 	 ' — — — - — -— 	 , — — - - 	
US6 * °f Tota) Quantity Quantity6 {kkg) Contained in Fntimai.rf rm,i™
Used Product (kkg) Estimated Environmental Releases

Pesticides0 7m 10
Hood Preservative 20% 2
Glass Manufacture 5%
Alloys'' 3j
(copper and lead)
Other (total) 2X
Feed Additives/
Veterinary Chemicals
Electronics
Catalysts
TOTAL 100% ,t
CO "•
t-> 	 — 	 	 	 _...
H4
a) Use distribution pattern based on percentages
b) Quantity is kkg of As.
nir uanii Surface
-250 1°-:>50 1.500 8,100 ?20
'93° 2-930 neqd ne«d negc
730 720f in") h i
"° 10 neq" nog1
440 440
11U neg
290 2 ,0
220 218 2k
6« 601
10 • i
101
640
1.500 8,100 720
from Loebenstein, 1980a.

PUTM Total
10,000
neg
10
neg
SO
2
10
10.311
"

c) Includes herbicides, cotton dessicants. defoliants, and soil sterilizers. See Table 3.2
d) f'rn^Jr'J'H P'ants f "erate sludge sludge has 9 g As/kkg drv sludge. (EPA. 1979d). sludge density
«U n'n, L? " f 1S 95* water and- sltld
-------
           Table 3-6.  Arsenic Releases from Production and Use of Pesticides, 1979 (kkg)
Pesticides
MSMAC
Production
Use
DSMAC
Production
Use
Arsenic Acid
Production
Use
Cacodylic Acid
Production
Use
10,10' - OBPA
Production
Use
OtherJ
Production
Use
TOTAL1
Quantity (kkg As)
Aira
3,000
2-65
500e
670
0.5-16
noe
2,830
2-54
179
690
0.4-13
1206
4

3,060
2-60
510e
10,250 1,500
Estimated Environmental Releases
k Water
Land Surface POTW

negd neg^
2,200 210b

d d
neg neg
500 46b

f f
neg neg
2,600 200b

h li
neg neg
510 48b

Jneg1

k k
neg neg
2,300 214b
8,100 720
Total

65
2,900

16
650

54
2,800

13
680



60
3,000
10,000
Footnotes next page.

-------
                                           Table 3.6.  (Concluded)
Co
I
     a)  All  almosphoric emissions ranges from pesticide manufacture (EPA  I973b) are tasaH „
                          ,     r
     cl
MSMA and DSMA respectively.
                                 ,M
             on m. „„ .,„,„. ,actors



    I))  EPA, I976a.  SJttig, 1979; EPA,  I976c.
    J1
    kl
    I)  Totals may not add due to round iny.


                                                                                           contained in
                                                jrj;jrc •

-------
      Environmental releases of arsenic from production of MSMA and DSMA
 total 65 kkg and 16 kkg,  respectively;  essentially all of these releases
 are into the atmosphere.   On the basis  of annual discharge of 0.08 kkg
 arsenic at a representative DSMA/MSMA production facility, wastewater
 discharges are believed to be negligible (EPA 1975d,  EPA 1980b) .

      Larger quantities of arsenic are released to the environment from
 pesticide application; approximately 500 kkg and 110  kkg of arsenic are
 emitted to the atmosphere from MSMA and DSMA use annually (see Table 3-6) .
 Approximately 2,200 kkg and 500 kkg arsenic are disposed to land, from
 MSMA and DSMA use respectively, either  during application or as fields are
 tilled.   Discharge to  water occurs  as runoff;  DSMA and MSMA are not inten-
 tionally released to surface waters or  POTWs.   Richardson et_ _aJL.  (1978)
 determined that for arsenic acid, about 7% of the amount of arsenic
 applied is lost as runoff and erosion.   If this loss  is similar for
 MSMA/DSMA, then 210 kkg and 46 kkg  of  arsenic would be discharged per
 year from MSMA and DSMA use, respectively.

 3.4.1.2  Arsenic Acid
      Arsenic acid  (HsAsO^)  is  produced  on  a  batch basis  from arsenic
 trioxide  and nitric  acid  (see  Appendix  B,  Figure  2).   Currently it is
 used as a cotton dessicant  to  facilitate mechanical harvesting  (Richard-
 son _et a^.  1978).  In  1979,  approximately  2,830 kkg of arsenic  were con-
 tained in arsenic  acid.   Based on  uncontrolled and controlled emission
 factors of 10 kg As/kkg pesticide  produced and 0.29 kg As/kkg pesticide
 produced,  respectively, 2 kkg  to 54 kkg of arsenic were  emitted to the
 atmosphere during  arsenic acid production  in 1979  (EPA 1973b;   EPA 1980b)

      Due  to increases  in  application and efficiencies  of emission  control
 devices and additional state and federal regulations,  the actual quantity
 of  arsenic emitted is  probably closer to the lower portion  of the  range,
 i.e., 10  kkg.  Wastewaters  from production are recycled  to  NOX  scrubbers
 or  returned to the reaction vessel for  further product recovery; there-
 fore, no wastewater  is discharged  from  this  process (EPA 1975c).

      During application of  arsenic acid, approximately 17 kkg of arsenic
 are emitted to the atmosphere;  2,600 kkg are disposed  of  on land (Table
 3-6).  Discharge of  arsenic to waters occurs  as runoff.   Richardson et.
 al.  (1978)  determined  that  7%  of the arsenic  applied is  lost as runoflf;
 on  this basis, in  1979 approximately 200 kkg of arsenic entered surface
waters from use of arsenic acid as a cotton defoliant.

 3.4.1.3   Cacodylic Acid

      Cacodylic acid, an insecticide and contact herbicide,  is manufac-
 tured by addition  of calcium chloride and  sulfur dioxide  to DSMA
 (Sittig 1977) .  Its  chief use  is in lawn renovation and weed control  in
 noncrop areas  (EPA 1976d)  .  A maximum of 13 kkg of arsenic  is emitted
 to  the atmosphere  from production of cacodylic acid, based  on EPA
 emission factors (Table 3-6) .


                                  3-14

-------
  use oi

  767. (510

  to water as runoff  (Table 3-6).


   3.4.1.4  Miscellaneous
                                                       ^ron-nt fro,

                                                                 h*™>
                                                               discharged

 products.   Total arsenic
 arsenic consumed.



 3-4-2  Wood Preservatives
                                                       Polyvinyl chloride

                                                       ^"^
                                       due  to  the small quantities of
                                                             (CCA,
treated
              any  type of preaervsMf      7        a11 W0od Prod"cts
               7  ype or preservative were preserved with arsenicals.
               useccandcree!                        from
 outlines wastewater discharges   *«! Eligible.  Appendix A, Note 4,

 kkg per year per plant and f totaj  of  47"  *? "^f disch"S- of O.

 discharged annually (EPA 19?9d)   sludge i' f^' ^ ^ °f 3rSenic
 wastewater treatment;  the arsenic fonr  V  generated from in-plant

 ^y sludge.  Thus ap^rox^t^  S.SSS   S a^ic^^S t/kk§ C° 9

 via sludge disposal (see Appendix A  Note ".         " ^ S8nt tO land
                  ^^                    Jjnd very tightly  to wood fibers;

very slowly ,  (EPA  1976a) .   ri°ratlon or treated wood  is expected to occur
                                 3-15

-------
 3.4.3  Glass Manufacture

      In 1979, about 730 kkg of arsenic were utilized in glass manufac-
 turing as a decolorizer, a fining agent (to remove bubbles),  and as a
 stabilizer for color-producing chemicals (EPA 1980b).   Quantities of
 arsenic used in these applications are decreasing as nonarsenical al-
 ternatives (various sulfates and antimony oxide)  are found.   Currently,
 opal and lead glass and certain specialty glasses are  the largest con-'
 sumers of arsenic (Sutherland 1980,  Thatcher Glass Company 1980).

      During glass manufacture,  arsenic volatilizes and is thought to
 condense on particulates.   Based on  a  controlled  emission factor of 0.015
 kg As emitted/kkg glass produced,  and  production  of 534,900 kkg,  about
 10 kkg of arsenic are emitted to the atmosphere annually  from glass manu-
 facture (see Appendix A, Note 3).  As  all off-quality  glass is  recycled
 (as  opposed to  disposed of  on land), release of arsenic to land is  negli-
 gible (EPA 1980b).   Arsenic is  tightly bound in glass;  therefore, release
 of mobile arsenic from discarded glass products sent to landfills is
 negligible.   Based  on EPA process  descriptions and water  use  patterns
 (cooling water, washing of  finished  products), discharge  of arsenic is
 negligible from production  processes (EPA 1980b,  EPA 1973a).

 3.4.4  Allovs
     Due to its semi-metallic properties, arsenic is added to lead,
copper, and brass to enhance physical and chemical properties of those
metals.  Tables 3-4 and 3-5 outline environmental releases of arsenic
from alloying (secondary lead/copper production).   As recycling of
arsenic-containing metals is so extensive, the environmental releases
of arsenic contained in scrap metal are difficult to distinguish from
releases of arsenic that is added during metal reprocessing.  Therefore,
these releases are calculated together, although about 440 kkg of "nex/1'
arsenic are added annually.

     Atmospheric emissions from alloying are considered negligible (EPA
1976a, EPA 1980b).   Land-destined arsenic (total of 300 kkg As)  from
these processes is contained in slag.   Wastewater discharges are negli-
gible (see Appendix A,  Notes 7,  8, 11  and 19).

3.4.5  Small-Volume Uses of Arsenic

     Approximately 290  kkg of arsenic  (2% of the available arsenic supply)
are consumed each  year  in feed additives, veterinary chemicals,  electronics,
and catalysts; releases of arsenic to  the environment from manufacture
and use of these products is presented in Table 3-5,

3.4.5.1  Feed Additives and Veterinary Chemicals

     Arsenic is  added to poultry and swine feed to  prevent disease,  and
improve weight gain:  roxarsone,  carbarsone and nitarsone are  among the
                                 3-16

-------
                   og
   appoxi,nacely 2 22 arsenic   ?h^r "•   " °f ?«"="« «d total
   an  EPA  Cl^)  ,l^zr^^™"™^™**^ -*-t«l«.   Hovever,
  3-4.5.2  Electronics and Catalysts



                                           used

                              .
  common).                     ,  other   ea-containing  catalysts  are more
       GaAsP is foried b  vaor-ha           '      " «':k 1980)
 of 12 loss would emit 0.6 kk    Th               " "OrSt Ca8e
 through thls route,               h
3'5  DISPOSAL OF ARSENTr-rnMTAT>TT>r
urban refuse landfills or incinerato     A^   treatment work^ (POTWs)

       each waste treatment catezorv1^ chJ^^.ff'"1?1* balance
                                                       A-D (Appendix
                                  3-17

-------
 3.5.1  POTOs

      Arsenic loading to POTOs  is largely dependent upon variations in
 industrial discharges and the  type of industry in a particular municipal
 area.   A framework for calculating the total  arsenic flow through the
 nation's POTOs  (see Table A-7)  is  provided  by data from a recent EPA
 study.   A materials balance  of arsenic at the treatment plants can be
 constructed using  a total POTW flow of approximately 10^1/day (EPA
 1978c)  and median  values of  <50 yg As/1 (influent)  and  <50 yg  As/1
 (effluent)  (EPA 1980d).   It  is  assumed for  purposes of  these calculations
 that  influent and  effluent flow rates  are equal,  i.e.,  that water loss
 from  sludge removal and evaporation are small compared  with influent
 flows.   The results of the calculations show  influent and effluent load-
 ings  to be the  same,  1,800 kkg/yr.   Trace metal removal can be'achieved
 with  advanced treatment  processes,-  however, less  than 2% of the nation's
 POTWs utilize such processes (Linstedt _e_t al.  1971) .  Arsenic  removals
 of  90%  have been demonstrated  using ferric  chloride precipitation;  com-
 bining  that treatment  with activated carbon yields  removals of up to
 97% (Reimers and England 1980).  However, plants  surveyed to obtain the
 previously described  influent  and  effluent  arsenic  concentrations did
 not practice such  techniques.

     It is  further assumed that while  arsenic  is  recycled within the
 activated  sludge process,  all will  eventually  be  wasted.   Thus,  the
 value for  arsenic  in  sludge should  be  the difference between the influent
 and effluent arsenic  totals, as  there  is  an assumed negligible loss of
 arsenic to  the  air.  Using this assumption, however, no  arsenic  would
 be  released to  sludge, an  erroneous  conclusion, as  seen  from the sludge
 data presented  in  Table  A-7.  This  conflict can be  attributed  to mathe-
 matical assumptions made necessary  by  the imprecise nature  of  the data
 (e.g.,  high detection  limits).

     An alternative method, therefore,  for estimating the annual arsenic
 discharge  to sludge is to  calculate  arsenic release from  the arsenic
 concentration and  quantity of dry sludge produced annually  (6  x  106  kkg,
 EPA 1979i).  Assuming  the median arsenic concentration of POTW wet  sludge
 to  be 170  yg/1  (EPA 1980e) and that wet sludge is 95% water by weight,
 approximately 20 kkg are discharged  to land.  Approximately 25%  of  all
 municipal sludge is landfilled, 25% spread on land, 15% ocean  dumped  and
 35% incinerated.  As ocean dumping of sludge is mandated  to cease by  1981
 and assuming that more stringent air quality standards curb incinerator
 use (EPA 1979h), the arsenic contained in sludge is assumed  to be dis-
 charged  to  land.

 3.5.2  Urban Refuse

     Urban  refuse,  divided into combustible  and noncombustible fractions,
 is  usually  landfilled  (87%),  recycled (8%) or incinerated (5%)  (Geswein
 1980,  Alvarez 1980).  Arsenic flow through a municipal incinerator with
 a capacity of 920 kkg dry refuse per week is illustrated in Figure B-4,
Appendix B.  The only available arsenic data for incinerators is for
                                  3-18

-------

                     ls lailled  c"         ™     .
                                                                      <
                                                — — j 7  —.[-£.-. v*»^.*»io. U &JL V J.U/0 C)V


                                        -— metal (Gordon 1978). 'NO in-


                                    percentage of this amount of nonferrous
 3.5.3  Urban Runoff

21).  Approxiately 431 kkg  (417)


452 kkg (43%) to unsewerS'areas

bined severs, whlch discharge tQS




3-5.4  Natural Loading
                                                               x A, Note



                                                                S "*
                                                                to com-
                                "•                               '
shale, ana £££?%? ±£l l^ll^ '""  ""'  that  is
                                 3-19

-------
      Such an estimate seems reasonable in that the first estimate,
 2,850 kkg/yr, is based on 1968 land use data;  a larger estimate
 obtained by averaging intensive and "non-intensive" estimates would
 most probably reflect the current patterns of  land use.

 3.6  SUMMARY

      Arsenic is recovered domestically as a byproduct of copper smelting.
 The only arsenic production facility,  located  in Tacoma, Washington,
 produced 280 kkg arsenic  "metal"  and arsenic trioxide containing 6,420
 kkg As in 1979.  Approximately 8,940 kkg  arsenic were imported in 1979
 and 1,000 kkg exported.   Table 3-1 and Figure  3-1 outline environmental
 releases from production,  use,  and inadvertent sources of arsenic.   Of
 the total 1,400 kkg  arsenic lost  ot the environment during production,
 85% (1,200 kkg) was  land  disposed,  15% (210 kkg)  was emitted  to the
 atmosphere,  and <1 kkg was discharged  to  surface waters.   Furthermore,
 the 1,400 kkg As released  to  the  environment from its production account
 for only 3%  of the total  arsenic  released from all sources (i.e.,  pro-
 duction,  use and inadvertent  sources;  see Table 3-L).

      The pesticide industry is  by far  the largest consumer of  arsenic; pro-
 duction  of arsenical  pesticides utilized  approximately 10,000  kkg  (70%  of the
 available supply) arsenic  in  1979.   Of the total  10,000  kkg As released
 to  the environment from production  and use of  pesticides,  1,500 kkg (15%)
 were emitted to the  atmosphere, 8,100  kkg (81%) were land-disposed  and
 the remaining 720 kkg (~5%) were  discharged  to  surface waters.   Production
 and use  of wood preservatives,  glass and  alloy  additives  and miscellaneous
 arsenicals,  consuming 30%  of  the  available U.S. supply,  released approxi-
 mately 70 kkg arsenic to the  environment.  The  total  10,070 kkg As  released
 to  the environment from all uses  of  arsenic  represents approximately 20%
 of  the total  51,000 kkg released  from  all  sources  (production,  use  and
 inadvertent  sources).
                                  *»
     Arsenic  is  associated with most mineral ore; processing of  these
 ores,  as  well as smelting and refining of  the resultant metal,  introduced
 21,500 kkg of  arsenic  (42% of the total 53,400  kkg As released)  into the
 environment.   These releases are primarily contained in land-disposed
 slag.  Fossil  fuel combustion released 17,000 kkg  (33% of the  total
 quantity  released from all sources) of arsenic   to the environment;
 cotton ginning  accounted for about 880 kkg.

     Demand for arsenic is expected to increase only about 1% annually
 through 1985.  This low market growth factor can be attributed to the
 search for non-arsenical alternatives,  especially in the glass manufac-
 turing industry.  According to the industries involved, non-arsenical
wood preservatives, feed additives, and cotton  dessicants are not as
 effective or  economically attractive as are arsenic products.
                                 3-20

-------
                                REFERENCES
 Alvarez, R.  Status  of  incineration  and  generation  of  energy  from
         PC;""g  °fcmUniciPal  solid was'*;  Natural  Waste Processing
            ,  American Society of  Mechanical Engineers,  New York, 1980

            id Pr"erVperS' Association.  Wood  preservation statistics,
               n      Preserv"S  Association Proceedings, Washington,
 DC,
 Angino, E.E   Magnuson, L.M.; Waugh, T.C.; Galle, O.K.; Bredfeldt  J
 Arsenic in detergents: Possible danger and pollution hazard
 Environmental Science and Technology 168(4)jl970.
        uVhn CenSUS'   ?"°n Ginnin§s in th« ^ited States.  Crop of
        U.S.  Department Commerce, Washington, D.C.;1979.
 Cammarota,  V.A.   Zinc  (In)  Minerals  Yearbook Volume  1    U S
 Department  of  the  Interior,  Washington,  D.C.;1978.    "    "  "

 Cammarota,  V.A.  (Bureau  of  Mines,  Washington,  D.C.)  Personal
 Communication; 1980.                                 ^er&onai

 Carapella,  S.C.  Arsenic  and arsenic allovs.   (In) Kirk-Othmer
                                                          ± Son. ; Vol
Oavis, w.E. and Associates mi.  National inventorv of sources and
emissions:  arsenic - 1968 .  Washington, D.C. :   PB 222061?"l9n.
                                3-21

-------
 Environmental Protection Agency,  1975a.   Development Document for
 Interim Final Effluent Limitations Guidelines and Proposed New Source
 Performance Standards for the Primary Copper Smelting Subcategory and
 the^Primary Copper Refining Subcategory  of the Copper Segment of the
 Nonferrous Metals  Manufacturing Point Source Category.   Washington
 D.C.:   EPA 440/01-75/032;1975.

 Environmental Protection Agency,  1975b.   Development Document for
 Interim Final Effluent Limitations Guidelines and Proposed New Source
 Performance Standards for the Secondary  Copper Subcategory of the
 Copper  Segment of  the Nonferrous  Metals  Manufacturing Point Source
 Category.   Washington,  D.C.:   EPA 440/1-75/032;1975.

 Environmental Protection Agency,  1975c.   Development  Document for
 Interim Final and  Proposed  Effluent  Limitations  Guidelines and New
 Source  Performance Standards  for  the Ore  Mining  and  Dressing  Industry
 Point Source  Category,  Vol.  1  and 2.   Washington,  D.C., ;  1975.

 Environmental Protection Agency,  1975d.   Initial  Scientific Review of
 MSMA/DSMA.  Washington,  D.C.;  1975.

 Environmental Protection Agency,  1975e.   State of  the Art  of  the
 Inorganic  Chemical  Industry:   Inorganic Pesticides.   Washington
 D.C.; 1975.

 Environmental Protection Agency,  1976a.   Technical and Microeconomic
 Analysis Task III  Arsenic and  its  Compounds.  Washington,  D.C.:   EPA
 560/6-76-016;1976.

 Environmental Protection Agency,  1976b.   Air  Pollutant Assessment
 Report  on  Arsenic.  OAQPS,  Research  Triangle  Park, N.C.; 1976.

 Environmental  Protection Agency,  1976c.   Development Document  for
 Interim Final  Effluent Limitations Guidelines for  the Pesticide
 Chemicals  Manufacturing  Point Source  Category.  Washington, D.C.:  EPA
 440/l-75-060;1976.

 Environmental  Protection Agency,  1976d.  Arsenic.  Research Triangle
 Park, N.C.:   EPA 600/1-76-036;1976.

 Environmental  Protection Agency,  1977.  Source Assessment:
 Defoliation of Cotton, State of the Art.   Research Triangle Park,
 N.C.; 1978.

Environmental  Protection Agency,  1978a.  Alternatives for Hazardous
Waste Management in the Metals Smelting and Refining Industries.
Washington, D.C.:  EPA SW153;1978.
                                 3-22

-------
  Environmental  Protection  Agency,  1978b.   Gross  annual  discharge  to  the
  water  in  1976:   arsenic.   Washington,  D.C.:   Revised Report  2,  1978.

  Environmental  Protection  Agency,  1978c.   Needs  survey,  Office  of  Water
  Planning  and Standards.   Washington, D.C.;  1978.

  Environmental  Protection  Agency,  1979a.   Preliminary Draft
  Environmental  Assessment  of Arsenic Emissions from Copper Smelters
  Washington, D.C.;  1979.

  Environmental  Protection  Agency,  1979b.   Development Document  for
  Effluent Limitations Guidelines and Standards for the Nonferrous
  Metals Manufacturing Point Source Category.  Washington, D.C.-  EPA
  440/l-79-019;1979.

 Environmental Protection Agency,  1979d.  Development Document for
 Effluent Limitations Guidelines and Standards for the Timber Products
 Processing Point Source Category.   Washington, D.C.-   EPA.
 440/l-79/023;1979.

 Environmental Protection Agency, 1979e.  Arsenic:  A Preliminary
 Materials  Balance.  Washington, D.C.:   EPA 560/6-79-005;1979.

 Environmental Protection Agency, 1979f.  Development  Document for
 Proposed Effluent Limitations  Guidelines  and Standards  for  the  Iron
 and Steel  Manufacturing Point  Source  Categorv.   Vol   1278
 Washington,  D.C.;  1979.                     '          '  '  '  ',

 Environmental Protection Agency, 1979g.  Status  Assessment of Toxic
 Chemicals:   Arsenic.   Cincinnati,  OH:   EPA 600/2-79-210;1979.

 Environmental Protection Agency, 1979h.  Environmental  Impact
 Statement  Criteria for  Classification of  Solid Waste Disposal
 Facilities and  Practices.   Washington,  D.C.:  EPA SW-821;1979.

 Environmental Protection Agency, 1979J.   Comprehensive Sludge Study
 Relevant to  Section 8002(g) of  Resource Conservation and Recoverv Act
 of  1976.  Washington. D.C.: SW802:1979.

 Environmental Protection Agency, 1979J.   Determination of the Impact
 of  Toxics  in Urban Runoff.  Unpublished draft report by Monitorln* and
 Data Support Division under Task 5(a).  Washington, B.C.:  Office°of
 Water Regulations  and Standards, U.S.  EPA; 1979.

 Environmental Protection Agency, 1980a.   Personal Communication with
 Richard Seraydarian, Effluent Guidelines Division, Washington, D.C.;
 1980.

Environmental Protection Agency, 1980b.   Health Assessment Document
 for Arsenic.   Environmental Criteria and Assessment Office.   Research
Triangle Park, N.C.; 1980.
                                 3-23

-------
Environmental Protection Agency, '1980c.  Development Document  for
Effluent Limitations Guidelines  and Standards  for  the  Steam Electric
Point Source Category.  Washington, D.C.:  EPA 440/1-80/029;1980.

Environmental Protection Agency, 1980d.  Fate  of priority  pollutants
in publicly owned treatment works.  Interim Report, Washington,  D.C.:
EPA 440/1-80/301;1980.

Environmental Protection Agency, 1980e.  Priority  Pollutant Frequency
Listing Tabulations and Description Statistics.  EPA Office of
Analytical Programs, Washington, D.C.:  1980.

Environmental Protection Agency, 1980f.  Treatability manual, Vol.  5
Summary.  Washington, D.C.:  EPA 600/S-80-042;1980.

Federal Register, Vol. 45, No.200, pg. 68331.

Geswein, A.  Personal Communication.  June, 1980.

Goihl, J.H.  Feed additives effect on swine waste.  Feedstuffs
51(37), 1979.

Gordon, J.  (Mitre Corp)  Assessment of the impact of resource
recovery on the environment.  McLean, VA:   MTR 8033, 1978.

Greenberg, R.H.; Foller, W.H.; Gordan, G.E.  Composition and size
distribution of particles released in refuse incineration.
Environmental Science and Technology, 12(5):566-573 ; 1978 .

Hollenbeck, B.  (Hewlett - Packard Corporation)  Personal
Communication.  Colorado Springs, CO:  1980.

Kearny, K.  (Office of Pesticides Toxic Substances Environmental Metal
Protection Agency, Washington, D.C.)  Personal Communication;  1980.

Kurtz, H.F.  Bauxite and alumina (In) Minerals Yearbook Vol. 1.  U.S.
Department of the Interior, Washington, D.C.; 1978.

Law,  S.L.; Gordon, G.E.  Sources of metals in municipal incinerator
emissions.   Environmental Science and Technology.
13(4):432-438;1979.

Linstedt, K.D.;  Houck, C.P.; O'Connor, J.T.  Trace element removals in
advanced wastewater treatment process.  Journal of Water Pollution
Control Federation,  43:7; 1971.

Loebenstein, J.R.  (Bureau of Mines)  Personal Communication;  1980.

Nelson, K.W.  (ASARCO Inc, New York)  Personal Communication;  1980.
                                3-24

-------
 Penz,  P.A.;  Haisty,  R.W.;  Surtani, K.H.  Digital Displays (In)
 Kirk Othmer  Encyclopedia  of Chemical Technology 3rd ed.  Vol.  7.
 John Wiley and Sons,  New  York,  1979.

 Peterson.   (Bureau of Mines)  Personal Communication;  1980.

 Plunkert,  P.   (Bureau of  Mines)  Personal Communication;  1980.

 Pressler,  J.W.  Boron (In)  Minerals Yearbook Vol.  1.   U.S.  Department
 of the Interior.   Bureau  of Mines.  Washington,  D.C.,  1978.

 Rathjen,  J.A.   Antimony (In) Minerals Yearbook Vol.  1.  U.S.
 Department of  the Interior.   Washingto'n,  D.C., 1978.

 Rathjen,  J.A.   (Bureau of  Mines)   Personal Communication;  1980.

 Richardson,  C.W.;  Price, J.D.;  Burnett, E.  Arsenic  concentrations  in
 surface runoff from  small  watersheds in Texas.  Journal  of
 Environmental  Quality 7(2); 1978.
                                                              i
 Ryan,  J.P.;  Hague, J.M.; Rathjen,  J.A.   Lead (In)  Minerals Yearbook
 Vol.  1.   U.S.  Department of  the Interior.   Washington, D.C.;  1978.

 Schroeder, H.J.   Arsenic  (In) Minerals  Yearbook  Vol.  1.   Bureau  of
 Mines,  Department  of  the Interior.  Washington,  D.C.;  1978.

'Schroeder, H.J.;  Coakely,  G.J.   Copper  (in)  Minerals Yearbook Vol.  1,
 1976.   U.S.  Department of  the Interior.   Washington, D.C.;  1978.

 Sittig. M.  Pesticides Process  Encyclopedia.   New  Jersey:  Noyes
 Data Corporation;  1977.

 Slater, S.M.;  Hall, R.R.,  1977.  "Electricity  Generation by Utilities:
 1974 Nationwide Emissions Estimates," AICHE  Symposium  Series  73, 291-
 1977.                               '                        —

 SRI. 1980.  Directory  of Chemical  producers, U.S.A. Stanford  Research
 Institute.  Menlo Park, CA;  1980.

 Stowassen, W.F.  Phosphate rock.   Schreck, A.E.  ed.  Minerals Yearbook
Vol. 1.  U.S.  Department of  the Interior,  Washington,  D.C.; 1980.

Sutherland, S.  (Corning Glass)  Personal Communication, November,
1978.

U.S. Department of the Interior.  Antimony, arsenic, and  mercury in
the combustible fraction of municipal solid wastes.  Bureau of Mines
Report of Investigations.   Washington, D.C.:  RI8293;1978

Warshawsky, J.   (EPA Pesticides)  Personal Communications; November
1980.
                                   3-25

-------
                4.0  FATE AND DISTRIBUTION IN THE ENVIRONMENT  -

  4.1  INTRODUCTION

       This  chapter describes the fate and distribution of arsenic in soil
  water,  and air following its environmental release from the processes
  described  in Chapter 3.0.   Emphasis  is  placed on the pathways  leading to
  arsenic contamination of surface  and ground water and potential  exposure
  of  humans  and aquatic life.   To the  extent possible,  the chemical  form
  of  arsenic initially released is  identified for each major  category of
  discharge  and its  transformations  and transport to other media are
  tollowed to  a point  of  equilibrium.   This  type  of analysis  is  difficult
  ana at  times  imprecise  due  to the multiplicity  of arsenic forms  released
  to  the  environment,  as  well  as  the numerous  forms present naturally in
  soil  and water.  Due  to  the  poor analytical  techniques  available for  many
  forms of arsenic and  the complexity  of multi-variable analysis, many  fate
  studies do not distinguish between different  forms  of arsenic.  Therefore
  it  is not  possible to maintain  a mass balance for each  chemical form  from'
  release (or formation)  to exposure of receptors.  At best, predominant
  forms in environmental media  present at the time  of exposure are identi-
  f led.

      The environmental fate and distribution of arsenic is analyzed by
 the characterization of specific environmental pathways.  Once arsenic
 1-3 released to air or water, the initial chemical form has less influence
 on the ultimate fate than the environmental parameters at a specific
 locale.   In addition, since arsenic in certain forms is relatively mobile
 compared with other elements, the initial receiving medium often  will not
 be the same one to which human and other receptors may be exposed.   There-
 fore,  elucidating the environmental fate of arsenic requires characterizing
 the  major types ot releases  into environmental media and following  each
 pathway  through the various  transformations and removal  processes to the
 ultimate fate of  the element,

 _   The major pathways  of physical transport  of arsenic are  designated
 in Figure 4-1  for the major  categories of  releases identified in  Chapter
 3.0.   Atmospheric  releases  (Pathway 1) consist  primarily of point sources
 such as  power  plants  and  smelter stack emissions.   Inadvertent  releases of
 arsenic  from  combustion  processes are among the  largest  man-made sources  of
 arsenic  into  the  atmosphere.   Atmospheric  releases often lead to highly
 localized pollution of air, soil and  water  in  the  vicinity of some sources
 such as  nonferrous smelters.                                      =uuj.«-e&,

     Pathway  2  considers  arsenic in solid waste  resulting  from  fly  ash
 disposal mining  activity, smelting/refining activities, and municipal
 refuse.  Presently   this  pathway accounts for most  arsenic releases! and
 as air and  water pollution standards  become more stringent, the amount of
 arsenic  disposed upon land can be expected  to increase!

     Pathway 3 identifies arsenic released  in industrial aqueous efflu-
 ents,  such as  those from wood  preservers and pesticide manufacturers.
Most effluents are discharged to local surface waters; only veterinary
                                    4-1

-------
chemical plants and several metals production facilities are known to
discharge to publicly owned treatment works  (POTWs).  The fate of
arsenic in POTWs and in industrial wastexjater treatment is discussed
in Pathway 4.

     Pathway 5 considers intentional uses of arsenic in pesticidal
applications agriculturally, to lawns and surface water—or as a con-
taminant of phosphate fertilizers.

     Figure 4-2 provides a composite overview of the major environmental
pathways of arsenic indicating the approximate contributions made by the
major industries to air, soil, and water loadings based on estimated
releases from Chapter 3.0.  As indicated, the major portion of the en-
vironmental releases of arsenic are to land.

     This chapter is organized into four sections.  First (in Section
4.2) the general chemical, physical, and biological processes are
described that affect the behavior of arsenic in surface water and
soil systems.  Then Section 4.3 discusses for each of the five major
pathways the interplay of these fate processes as they determine the
ultimate distribution of arsenic in environmental media.  Field studies
and selected, relevant, monitoring data are used to illustrate each
pathway.  A general discussion of concentrations detected in environ-
mental media, from large-scale national surveys and some local studies,
is presented in Section 4.4.  Finally, Section 5.5 provides an overview
of the environmental pathways of greatest significance for exposure
of humans and other biota.

4.2  IMPORTANT FATE PROCESSES

4.2.1  Physiochemical Fate Processes

4.2.1.1  General Fate Discussion

     Arsenic is usually referred to as metalloid and is chemically
similar to phosphorus,  which occurs above it in the Periodic Table.  It
is usually found associated in nature with sulphide ores in minerals
such as arsenopyrite (FeAsS), niccolite (NiAsS), cobaltite (CoAsS)  and
tennantite (Cu12AsuS13).  Weathering (by water,  chemical reactions,
wind, etc.)  of arsenic-containing bedrock, coals and ores leads to
release and mixing of these minerals into soil,  water and the atmosphere.
In addition,  intentional releases are responsible for a large portion
of the arsenic entering the environment annually.

     A wide variety of  chemical forms of arsenic, both inorganic and
organic, are released and also exist in each type of ecosystem receiving
the releases.  The forms of interest with regard to environmental exposure
and risk of humans and  other organisms and the interrelationships between
forms are discussed in  the following section.  Table 4-1 lists the
environmentally significant forms of arsenic that are discussed.   The
general environmental chemistry of arsenic in aquatic ecosystems (fresh,
                                 4-2

-------
     PATHWAY NO.
          1.
CO
Atmospheric Emissions
(Major Point Sources)
    As2O3 (11%)
                    As Production Smelting
                    Coal Combustion
                    Incineration
                    Cotton Gin Dust
                                                                        Surface Waters
                                                                          Sediments
                                                   Runoff
                                                  (Fast)
                                                                      Pavement & Local
                                                                          Road Soils
                                                                                                                 Ground Water
                    Solid Waste & Tailings,
                      Coal Piles & Open
                       Pit Mines (62%)
                   Primary As Production
                   Coal Mining
                   Ore Mining
                   Land Fills
                   Coal Combustion
                   Smelting/Refining
                                  FIGURE 4-1    MAJOR ENVIRONMENTAL PATHWAYS OF ARSENIC RELEASES

-------
^
POTW
1 .. I
Aqueous
Discharges
(2%)
Smelting
As Production
Glass and Cement Pr
Pesticide Manufactur
Wood Preserving
Detergents
Treatment Effluent 1^
System /
Pathway #4
1 Surf ace Water
Sediments
(Slow) — *-
oduction Hazardous/
	 ». Solid Waste t
Dump
Ground Water




"*" Oceans
^
5.

POTW Influent
or Industrial
Raw Waste



1
Primary
Treatment
-« lsro»ttL-




^^~
Biol
and/or
Trea

Shu

- Air

ogical
tment
t 0

Jge ln
L

Effluent
cean Dumping

cineration I 1
	 	 . ,.„ I Air
andfill
	 — *- o -i
Soil

(Slow)
\
Surface Waters 4
Sediments t
\
<
Ground Water



A
Purposeful
   Uses
  (19%)
       llerhicides
       Defoliants
       Wood Preservatives
       Insecticides
(Very Small)
                                                                          Surface Water
  Dissolved Solids
Suspended Sediment

-------
Oi
                                                                                                                                       ©—-
                                                                                                                           Complttxatiun \    Ruction
              saie unknown 01
              cycle are noted wtwn k

other fmtiii may dlso be pro^unt
                                           FIGURE 4-2
                                                           MAJOR ENVIRONMENTAL PATHWAYS OF ARSENIC RELEASES

-------
        TABLE 4-1.   ARSENIC FORMS  OF ENVIRONMENTAL SIGNIFICANCE
 Form

 INORGANIC

    Arsenic  trioxide

    Arsenous acids


    Arsenic  acid


    Arsenites

    Arsenates

    Arsenic  sulfides
ORGANIC
   Dialkylars ines
 Examples
As,0,
  2  o             -.
H3As03,H2As03,HAs03

                  2'
       .H^sOit,
MHAsOu,M3(As04)2

AsS2,HAsS2
   Methanearsonic acid
   Dimethylarsinic acid    (CH3)2AsOOH
   (cacodylic acid)
   Arsine
H3As
  _3
HAs  (CH3)2
   Trialkylars ine          As(CH 3)3
   (e.g., trimethylarsine)
     Comment
 Formed  during combustion

 Hydrolysis  products  of
 As^O^

 Formed  when As^Ogdissolved
 in nitric  acid

 Salts of arsenous  acid

 Salts of arsenic acid

 Reaction product of
 hydrogen sulfide  in presenc
 of hydrochloric acid with
 trivalent  or pentavalent
 arsenic compounds
Methylation product of
 various arsenic compounds

Methylation product of
 various arsenic compounds
Reduction and methylation
 products of inorganic
 and methylarsenic acids
Reduction and methylation
 products of inorganic and
 methylarsenic acids

Reduction and methylation
 products of inorganic and
 methylarsenic acids
* M signifies a univalent metal cation or an equivalent of a
  multivalent cation.
Adapted from Braman (1972),  NAS (1977).
                                   4-6

-------
                                             ground „«„ u presented
  ^•2.1.2   Aqueous  Chemistry
                                     Stable in four
                                                             states, +5
and
                Such a
                                   can only be made for
                                             PH-  in                     .
   '       '  "
       2;
barium caused  precipitation of Rp  ^1 n f    !  P   '   he  Presence of

                                 4-7

-------
Source: Ferguson and Gavis (1972).

FIGURE 4-3 THE Eh-pH DIAGRAM FOR As IN AN AQUEOUS SYSTEM WITH ARSENIC AND
           SULFUR. SOLID SPECIES ARE ENCLOSED IN PARENTHESES IN CROSS-HATCHED
           AREA, WHICH INDICATES LOW SOLUBILITY.
                                     4-8

-------
                     ^^
                                                             °f
   solubillzacion of the arsenic  athustsretum    rH
   or  stabilization as a precipitated sulphide   SoSbi^,,-""" "
   sediments may also occur by methvlaM™   j   ^  u    atlon ln the
   "Sine, uhlch ca
  4-2.1.3  jaediment
                               oro     rripient of
hydroxides.   Arsenite (III) ^ongly adsorb^     ^' Carbonates» and
Jt al. (1979)  found in an arsISf-atr^,  5     m     sulphides.   Holm
the sediment cores conta±nQTp^lly Inorl^ **** ^ P°r£ W3ter  °f
methylated arsenic was the primarj sjecies ?n th *rSeniC.' alth°Ugh
area.   Most  of the arsenic in f hi   J             Sr°Und wa£er of the
the iron  and        "
                               mar   seces  n th       .
  area.  Most of the arsenic in f hi   J             Sr°Und  wa£er of the
  the iron  and ^uadn^"^^^^^^';1^ W  C°ntalned
  amorphous, exchangeable,  and  calcium r^' i0llowfd ^ arsenic in the
  sedi^nt  surface layer was the only part oTthf"^  fraCti°ns'  Th*
  was stable (due to the Eh)  and rht       u         S  ln which Fe(OH),

  U---               ;' ~=~:  -"="*
                                           as a
                                         ,
 leant as  indicatedby   .'!?^*""1;? ??» ^e  quite signif-
 the total arsenic flux down the Mississinni I™ f^d^g that  70% of
 is associated with particulars-   \m 'IT  I™** lnt°  the Gulf of Mexico
 form.              P   icuxates,   30% was estimated to  be in dissolved

 4-2-1-4   Soil Chemistry


                                              °f
                      yss   her      be
resembles that  of orthophosphatf  Dif f erenceTL" b
lability, oxidation states  and io» a«    ?          n°ted are arsenic's
slightly reduced soil (e^  te^rllT^ ^ '" Or§anic matte^
species form, and in very^edS
-thylated arsenic and elemental
                                4-9

-------
               HAs02 oxidotio"

              -H--
                                 2-          /
                            HAs04          / Epilimnion
                                 f_PH  8-9/
                                11   "~ Z~7 / Thermocline

      ... _      reduction            p      /
    o HAsOo    -             H^AsO^       / 11   i-
   02-    2                  2    4       / Hypohmnion

   V       ^lo,,                     '
AsS5
                            5PA
                              /o/
                ads.
                precip.
         \ASpS
   (2S3
reducfioR   e.gFeAsa


     »TTjefHyiotion
                                          Sediments
Source: Ferguson and Gavis (1972).
       FIGURE 4-4  LOCAL CYCLE OF ARSENIC IN A STRATIFIED LAKE
                            4-10

-------
  arsenic species as in sediment,  are provided cM^ bTiron Ind"
  a ^n™ ,, r     ±,                 a        ef J_g7 JJJJ.^
 =SMS             vj™
 transition zone   and the rerfnrin  !•  J     §             >      ura
 results indicated that Asdll}  i« di"harge area'  Sand column elation
 water, regardleM oTthf SJi  ^t^^^,11:^ As(V)  in  ^OUnd
 reducing water than in oxidized water   The pH of th    """I  qUlCkly in
 affects the amount of arsenic LInrhL /he /H of the ground water also
 hydroxides.  As(V) is maximallf **   t f ° ^ Particulates and ferric




releasing bound L(V);  (2) reduction o°Ai) "ol^nz)' Sl^i  <° F*(I1)'
solution PH in the reduced waters.               AsUII), and  (3)  increased
      as  much As as kaolinite c
aasorption of As(V) occurs at nH s
was thought to be LlsO,   As(III)
as the PH increases Irom 3 to 9  ^
is being  adsorbed.   Relative to "
                                   With
                                   T^
                                                        maximum
                                                 form being adsorbed
                                              Whi°h As(III)  sPecies
hydrous  ion oxid- (r=0  60)
                 U 0.60).

                                              3rea  ^r=0-66) and
                               important parameters  are pH, cation
                               4-11

-------
   400-
   300-
 (J
 a
 
O
E

en
   100-
               Kaolinite
                                      68.0
Montmorillonite
                                                                              66.7

Source: Woolsonefs/. (1977)
                                     8       0

                                            PH
                                 10
              FIGURE 4-5   REMOVAL OF AS (III) FROM LEACHATE SOLUTIONS BY KAOLINITE

                         AND MONTMORILLONITE CLAYS AS A FUNCTION OF pH
                                            4-12

-------
   700-1	'	L  '   '   I   I
                   Kaolinite
   600
"5,500

 O»
5.

 a>
 o
 o
•o
> 300
O
E

-------
 exchange capacity (CEC),  and manganese oxides.   Leachate concentrations
 were as much as 73 mg/1  As from a municipal landfill.  In an earlier
 paper, Fuller (1977)  had determined that the relative mobility of arsenic
 compared with other metals in neutral to alkaline soils (Griffin and
 Shimp studied acid soils)  decreased as follows:   Cr>As>Se>Cd>Be> Cu.

 4.2.2  Biological Fate Processes

 4.2.2.1  Microbial Biotransformations

      The speciation of arsenic in natural waters  is significantly in-
 fluenced by biota (Andreae 1978).   Methylation  is especially  important
 in the transfer of arsenic from sediment back to  the  water column.
 Figure 4-7  illustrates this process.

      Fungi,  yeasts, algae  and  bacteria species have been reported to per~
 form  the various  steps comprising  the  process.  The function of methylation
 is thought  to  be  detoxification  for those organisms (Braman and  Foreback
 1973);  however,  it may increase  mobility and  result in  exposure  in  other
 areas.    The cycle occurs  both in  freshwater  systems  (Ferguson and
 Gavis 1972)  and in saltwater systems  (Andreae 1978).

      Depending upon the environmental  conditions  (aerated  or reduced,
 pH, microbial  population),  some  or all of the steps in  the overall
 cycle will  take place.  For example,  in buffered, biologically active
 freshwaters, the  rate  of conversion of arsenite to  arsenate via microbial
 oxidation may  be  significant (Ferguson and Gavis  1972).  In soil,
 microbial transformation is thought to  contribute appreciably to  loss
 of applied arsenic from soil, up to 50%  in one year (Pax 1973); however,
 the significant and persistent As  residues reported in  some soils sug-
 gest  this is not  always the case.  Unfortunately, since  few of these
 processes have been quantified and none  under natural conditions, it is
 difficult to discern the role of microorganisms in  the  equilibrium
 speciation of  arsenic.

      Table 4-2 presents the results of a number of laboratory studies
 investigating biological transformations of arsenic.  Most studies con-
 cerned pure cultures of microorganisms and were conducted in order to
 identify  transformation products.

     Methylation of both arsenate  (via initial reduction to arsenite)
 and arsenite has been reported under aerobic and anaerobic conditions
 (McBride and Wolfe 1971,  Andreae and Klumpp 1979,  NAS 1977).  The
mechanism involves replacement of substituent oxygen atoms by methyl
 groups (Challenger 1947).   Mono- and dimethyl compounds are common
 transformation products.   Woolson and Kearney (1973) found dimethyl
arsine to be the predominant product from reaction of cacodylic acid
 in soil  (aerobic and anaerobic).  It was unstable, oxidizing in air to
the oxide or back to cacodylic acid.  Under acidic conditions,  sewage
fungi produced trimethylarsine in sludge (Cox and  Alexander 1973) and 70%
of arsenite was transformed to trimethylarsine by  fungi (Challenger 1947) .
                                  4-14

-------
     OH
       I
HO - AS - OH
      II
      O
   (Arsenate)
                   Aerobic
                          }
                   Anaerobic
                  
-------
      TABLZ 4-2.  MICROBIAL TRANSFORMATIONS OF ARSENIC COMPOUNDS
 Reaction

 Reduction and
 methylation of
 arsenate
 Species
 Metabolic
  Product
 Methanobacterium     Dimethylarsine
                      Reference


                      McBridge and Wolf
                          (1971)
 Methylation of
 sodium cacodylate,
 sodium methanaer-
 sonate, sodium
 arsenite
 Various fungi:
 Pencillium
 aspergillus,
 Fusarium sp. ,
 Candida sp.,
 Gliocladium sp.
 Trimethylarsine
 NAS (1977)
 Dimethylation
 of  MSMA
 Soil  population
 Arsenates  and C02    NAS (1977)
 Reduction  of
 arsenate
 Seawater bacterial
 population
Arsenite
                      Johnson (1972)
Methylation,
reduction and
complexing of
arsenate
Marine algae
Arsenite, methyl-
arsonate, dimethyl
arsinate, and 12
soluble organo-
arsenic compounds
Andreae and
Klump  (1979)
Methylation
of arsenate
Sewage sludge
population
from anaerobic
system
Assumed to be
dimethylarsine
(transformation
complete in 8
hours)
McBridge, et al.
     (1978)
Methylation of
arsenate
Fungi on bread
cultures
Trimethylarsine
(70% of As203
during 24 months)
Challenger (1977)
                                  4-16

-------
                                                     —
  4-3  MAJOR ENVIRONMENTAL PATHWAYS



  4'3'1  Pathwav 1—Atmospheric Emissions With Subsequent
         ^Transfer to Water and Soil      ~  ~~	


  4.3.1.1  Sources



       Emissions to the atmosphere corner-?^ i T/ nf  n
  releases (see Chapter 3.0K  S! ^"fl^ °5 *n envir°™ental
  are primarily arsenic trioxide
 4'3-1-2  Atmospheric Transport
          to one source (Cole   t a   Sw "the^ff- "?"iml«« and,
tor sulfur.   Coles et al  fl97i)f^f   A  •   ' ?  affinity of arsenic
                   « £i-  C1979)  found tnat the concentration of arsenic
                                  4-17

-------
 in fly ash fractions corresponding to mean aerodynamic diameters of
 18.5 urn,  6.0 Mm,  3.7 yra and 2.4 ym increased as follows:   13.7, 56, 87
 and 132 yg/g.   (Concentrations vary considerably by coal  type, source,
 and coinbusion efficiency.   Note the significantly higher  levels measured
 and reported in Appendix C.)   Coles noted that  the elements associated
 with sulfide ores volatilize during combustion and recondense on smaller
 particulate matter.

      The  atmospheric lifetime and  dispersion of arsenic are dependent
 upon local meteorology  and  size distribution of particulates.   Dry
 deposition and wet fallout  are two processes that remove  arsenic from
 the atmosphere.   On  contact with alkaline media,  the  trioxide  form is
 converted to arsenite,  AsO(OH)7 and in acids to arsenous  acid,  H3As03
 (Mushak et al.  1980).   Raigaini et al.  (1977) measured the  trace metal
 contamination  of  local  soils  and grasses  near a lead  smelter.   The sur-
 face soils and grass contained enriched levels  of arsenic and  other
 metals; resuspension of tailings dust,  in addition to  direct emission
 of  aerosols from  the smelter,  was  identified as a major source  of arsenic,

      Fisher et al.  (1979) researched  the  solubility and concentration
 distribution of arsenic on  coal  fly ash particulates by leaching with
 aqueous media, at  pH 7.3.   The mass of  arsenic  per gram of  fly  ash
 residue was approximately 120  yg over  the  particulate  range  of  0.03 ym
 to  5 ym.   These concentrations were soluble  at  pH 7.3,  increasing in
 solubility in  particulates  greater than 0.4  ym  (solubility  of about
 12  yg/g)  as compared with particulates  smaller  than 0.2 ym  (solubility
 ~4  yg/g).   This result  would suggest that  significant movement  of
 arsenic into land  or surface water could occur  in  the vicinity  of an
 atmospheric  point  source from  fallout of  large  particulates.

      Crecelius jet  al. (1975) found that the major man-made contri-
 butor to  arsenic concentrations  in  Puget Sound was a copper  smelter.
 The  annual  arsenic loading  to the  Puget Sound was  estimated  as  follows:

           Incoming seawater                 5.8 x  108 g/yr
           River influx                      3.0 x  107 g/yr
          Precipitation                     7.7 x  107 g/yr
          Smelter effluent                  4.0 x 107 g/yr
          POTW effluent                     7   x 10s g/yr
          Drydock operations runoff         9   x 105 g/yr

The amount of arsenic released annually in the smelter effluent to
 surface water is now estimated at less than 1 kkg  (see Chapter 3.0);
 this  reduction indicates that wastewater treatment processes have
become more effective over the past few years.  Air emissions from  the
smelter were not quantified in this study; however, that rain samples in
the vicinity of the smelter  contained 17 yg/1 of arsenic as  compared with
0.4  yg/1 background levels indicates a significant contribution.  The
percentage of arsenic input  into Puget Sound due to atmospheric fallout
is estimated to be 10.5%.  Analysis of cores of  contaminated sediment
revealed 34% of the arsenic  associated with iron and aluminum oxides;
less  than 10% was  associated with easily oxidizable organic matter.


                                 4-18

-------

 transport out of the soil is occuTri™  *M  A        assumes that no

 th. sou concentratlon by O^^^0
                                    .,0 the
 emissions  from two  Washington D C  incinS ^'      i investigated the
to be a ..ueh mre siict so™.""'  C°al "=-"»»"» »«  thought
                               t                       » °-to


                                 ^-s^ r.-sss;'
tests (Suta 1980).   Most of  he part?, T?   ^  C° the reSUltS °f EPA
                              4-19

-------
         TABLE 4-3.   TOTAL ARSENIC CONCENTRATIONS IN VARIOUS MEDIA
                     IN VICINITY  OF A COPPER SMELTER
 Medium


 moss

 suspended dust

 falling dust



 soil


 barley
                   Distance  from Smeltery


                   2-3 km

                   1-2 km

                   not given

                   not given

                   maximum levels in
                   vicinity
                   vicinity

                   40 km

mammals            vicinity

river              vicinity

                   30 km downstream

aquatic organisms  400 km
 sediment
                   1,000 km
  Concentration


 40-60 mg/kg

<1 Pg/m

 25 g/100 m2 (max)

 3 g/100 m2 (average)


 30 mg/kg

 0.06  mg/kg

 0.01  mg/kg

 20-200  mg/kg

500 yg/1

5  yg/1

above natural background
background levels

above natural background
background levels
Source:  Landau (1977).
                                   4-20

-------
   4.3.1.3  Summary
   contributed  10 57  of  i-h! ,  ! i        S        '  aa!°sPh«ic fallout
                         nt0          "
   the  «.           a   1
                    a





  4>3'2  Z£^^^^-^^^
  ^•3.2.1  Sources

       Most of all environmental releases of arsenic
                                             ™
 Purposeful Releases.                    e  seParately  in Pathway  5,
 and arsenite (conversion products f o        .sed to land are arsenate
 aqueous media) ,  and the various gulf0™-^3611" trioxide on contact with
 found in coals  and ores.               containing arsenic complexes

 4-3.2.2  Fly Ash DisposaT_

      The
                                                                    the
      s in progress.  Ove  the    t few ~« Syhrecent and much of the
disposal site monitoring studies ?bv ^ nT^ .5°WeVer' the "suits of
and others) will becoS§available  *         ^^ Natlonal Laboratory
                                 4-21

-------
      In  laboratory  experiments  investigating  leaching  from fly ash,
 approximately  7%  of the  total arsenic  in  a  fly  ash  sample  was  released
 into  a spring  water solution  (pH  8.1)  after 7 days  (Turner et  al.  1978).
 Distilled water  (pH 5.9) was ineffective  at removing the element.  Theis
 and Wirth  (1977)  found a similarly  low solubility at neutral pH.   Assuming
 the rate is  representative of an  average  disposal site, then an estimated
 980 kkg  arsenic is  leached from the  14,000  kkg  of arsenic  disposed of  in
 fly ash  annually.   This number  does  not represent the  amount of arsenic
 entering ground water, however, due  to adsorption of mobile arsenic  onto
 soil  before  it reaches the water  table.

      Arsenic concentrations in  ash  pond discharge and  ground water often
 exceed drinking water standards due  to  the  alkaline conditions  present
 in ash disposal ponds (Dressen _et al.  1977) which are  conducive to
 mobility of  both  arsenate and arsenite , (Turner  et al.  1978).   In addition,
 the high ratio of associated surface area to volume in fly ash  contributes
 to greater potential for leaching.  The arsenic form in fresh  dry  fly  ash
 is predominately As203; however contact with water during  sluicing pro-
 duces arsenous acid, H3As03 and the arsenites, H2As03- and HAs032.   Since
 the environment of  ash ponds and  the surrounding ground water  is usually
 low in oxygen, the  trivalent form prevails  once the fly ash has been
 deposited.   Very  slow oxidation to arsenate, however,  may  occur.   Ad-
 sorption onto  fly ash solids and coprecipitation with  hydrous  iron oxides
 removes  some of the arsenic from solution (Turner jet _al. 1978).

      Figure  4-8 depicts the process, for fly ash disposal following gen-
 eration  and  includes arsenic concentrations, both total and percent
 trivalent, in  each  step (Turner et al. 1978).   Soil characteristics  were
 not given.  The trivalent form was most significant in the  fly  ash  pond and
 relatively low at the point of release to the environment,   probably
 because  of aerobic  conditions in the overflow area.   In another study,
 Cherry and Guthrie  (1977) found arsenic distribution in a  fly ash  basin
 as follows:  27 mg/kg in the sediments and 0.08 mg/1 in water at pH  7.1.
 Turner et al.  (1978) measured deep well water  concentrations as high as
 1.6 mg/1 total arsenic,  with 88% present in trivalent form.  The deeper
wells (12 m deep) had higher concentrations than shallower wells (4 in
 deep).

     In  a study of coal fly ash disposal at a  porous sandy site in
Michigan (Theis et al.  1978),  arsenic was present in the fly ash at
 1,200 mg/kg.  Arsenic concentrations were measured in surrounding soil
 and ground water.  Soil levels were 6.5 mg/kg  near the disposal lagoon,
 1.7 times greater than background levels measured 500 m away.   Levels
 in soil  at its interface with the water table  were relatively uniform,
4 mg/kg to 5 mg/kg over the distance 100 m to 600 m from the lagoon.
Arsenic  levels in ground water were significantly lower than the con-
 centrations within the lagoon,  50 yg/1 (at 100 m away)  as  compared  with
approximately 5,000 yg/1 in the disposal site.  Ground water levels
reflected changes in practices at the plant; a 50% reduction in opera-
 tion was associated with a reduction in ground water arsenic to 30  ug/1.
                                  4-22

-------
                                                 -  Nest of Porous Lysimeters (50, 75, 150 cm)

                                                       - PVC-cased Wells Screened (4-6 and 10-12 m

                                                                          i—  Seepage Area
 i
K>
Co
                             Surface of Ash
                             Approximate Top of Water Table
                             Ash Sluice Flow
             Coal        Fly Ash

Total As     16-61       97-313
      )

       Source: Turner et at. (1978).


                     FIGURE 4-8
                                                                                                      /;.* i?. A*-. ..'<•«:">;
                                                                                                       i-y/fAsh.;1-.;^
                                                                                                       '%&&$!$» Rock Quarry
                                                                                                      \gr-"'CiX>-a^
Lysimeters
A«g
1.5-9.3
40-100
Wells
A
-------
       In  a  survey  of  a  total of  11 ash  disposal  pond  effluents,  Turner
  et  al.  (1978)  found  a  wide variability in pH, arsenic  concentration
  and speciation.   Table 4-4 summarizes  his results.   It  is  evident  that
  general  extrapolation  from one  fly ash disposal study would not give  a
  realistic  estimation of the national contribution of arsenic  from  fly
  ash.                                                                }

  4-3.2.3  Mining Activity

      Presently, the only site in the U.S. where arsenic is produced is
  a copper smelter  in Tacoma, Washington.  However, arsenic at  low levels
 may be released from other metals, mining and smelting sites.   Abandoned
 gold and stibnite mines have been studied with regard to the heavy metals
 released to surface and ground waters.   Unlike the heavy metals that  tend
 to be solubilized by acid mine drainage, arsenic is less soluble at low
 pH values and is prone to adsorb onto the oxidized Fe203.  Stach et al
  (1978) found that arsenic as FeAsS associated with old gold deposits, was
 being oxidized by bacteria (Ferrobacillus and Thiobacillus) to AsOu andAsO,
 ana then coprecipitated with iron and aluminum oxides and calcium   The
 ground water of the area  contained arsenic  levels thought to be due to
 water percolating through the  old mine  tailings  and arsenic-contaminated
 stream sediments;  a high  correlation  was noted between  Fe and  As levels.

     _Hawkins (1976)  came  to  similar conclusions  to explain the concen-
 tration of  arsenic in ground water near an old gold  mining district
 Arsenic,  as FeAsS, was  being oxidized to H3AS03  complexes  and  contri-
 buting to ground water  contamination.   Wagemann  et  al.  (1978)  found that
 a lake receiving tailing  pond  seepage from a  gold mine  contained 0.7 mg/1-
 1.5  mg/1  of arsenic,  whereas control  lakes contained  0.01-0.07 mg/1
 Penrose et  al.  (1975) determined the  arsenic  input  to seawater and  sedi-
 ments  near  an  abandoned stibnite mine.   Arsenic  entered  the water by
 runorf from the mine  and solubilization of mine  tailings.  Arsenic  was
 detected  at higher levels in the water  during  periods of rain  (595  us/I
 as opposed  to  180  yg/1  during  dry spells).  The  contribution to  seawater
 and  sediments  from the  mine did  not appear to  be significant due to  a
 lack of detectable surface accumulation  of arsenic in sediment core
 analysis.   This may have been  due to  other factors than low loadings
 however,  such  as rapid  transport  out of  the area or inethylation  by  sedi-
 ment bacteria.

     .The  soil  surrounding a gold mine mill tailings pond in which arsenic
 (sultides and  trioxide) was disposed,  exhibited elevated arsenic levels up to
 400 m away  from the site (Comanor et a_l. 1974) .  Complexation of soluble
 arsenic forms by Al and Fe resulted in insoluble, inaccessible forms.
 lable 4-5 presents soil concentrations as a function of distance from
 5«Atai^in8S piles'  The hi§h£st concentration measured was 6,138 mg/k*
 (550 mg/1 soluble) in the  surface layer  30 m from one tailings pile   °
Although levels were generally highest in the surface layer of the soil
 the data imply that vertical migration of arsenic may be an important  '
process in some soils.  The soil type  studied by  Comanor had pH 6 5-7 5
and <5/* clay and silt.
                                   4-24

-------

                  TABLE 4-4.  ARSENIC IN ASH POND EFFLUENTS

                               FROM 11 PLANTS
-CS 	 U1R/1) .*_,„_,
1 7.1 48
2 5'7 1.9
•3
v Q u
9.8 13Q
4 9.6 80
5 9.6 1>4
6 9.8 28
7
NG TO
JO
8 10'7 7.3
9 «-3 5.2
10 »-3 46
113 '•' 119

j.i J-vaxeui.
6
17

5
7
11
8

2
40
33
5
4
 Mean
                                     35                    12
NG = Not given.
 Turner et al.  (1978)  field  study described
Source:  Turner _et al. (1978).
                                   4-25
                                            previously,

-------
       TABLE 4-5.  CONCENTRATIONS OF  TOTAL AND SOLUBLE  ARSENIC
             MEASURED IN SOIL SURROUNDING A MINE TAILINGS
                             DISPOSAL SITE
 Arsenic Species/
     DePth         Concentration (mg/kg) at Distance from Site (m)
Total As:
r
Surface
0-15 cm
15-30 cm
30-45 cm
45-60 cm
30

6138
644
368
247
798
165

2409
190
93
32
52
290

670
191
34
56
24
315

780
73
16
35
7
450

169
52
7
38
39
Soluble As:
  Surface        550          76         17        59           3
    0-15 cm      115          32          7        Nil         Nil
   15-30 cm       73           3         Nil       Nil         Nil
   30-45 cm       43          Nil        Nil       Nil         Nil
Source:  Comanor et al. (1974).
                                    4-26

-------
      Leaching coefficients for arsenic from copper and lead slag were
 estimated at 0.05% and 0.1%, respectively, of the total amount of
 arsenic initially present in the slag (Twidwell 1980).  These coef-
 ficients were based on observations with distilled water leachate (pH 7)
 and are presumably for total arsenic, with no differentiation between
 forms.  No other detail was provided.  It is worth noting that Theis
 and Wirth (1977) found minimal solubility of arsenic fly ash at neutral
 pH.

      Approximately 16,000 kkg of arsenic were estimated to be deposited
 onto land each year in slag from arsenic, copper,  lead, iron and steel,
 zinc and aluminum production processes (see Chapter 3.0).   On the basis
 of Twidwell's estimates,  approximately 4 kkg of arsenic leaches from
 copper wastes annually and 8 kkg from all other metals wastes.   The
 latter estimate assumed the higher coefficient derived for lead was
 applicable to other metals for which no  measurements were  available.
 There is a great deal of  uncertainty associated with these estimates
 due to the potential for  much higher leaching rates under  certain en-
 vironmental conditions (see 4.2,  soil).

      Even though arsenic  in smelting wastes  is initially in trivalent
 form (following  conversion from arsenic  trioxide),  the  most common water
 soluble form  found  in soil at waste  disposal  sites  was  arsenate.  at a
 ratio of 8 or 9  to  1 (Porter  and  Peterson 1977),  In  some  soil  samples,
 dimethylarsine made  up to 10% of  the total arsenic  levels    (These data
 are presented in greater  detail in Section 4.4.7) .   The  presence of
 methylated As forms  was thought to be  due to microbial  activity   either
 methylation of free  arsenite  or reduction of  arsenate.  The  authors
 expected microbial activity at  mine  tailings disposal sites  to be mini-
 mal due  to low levels  of  organic  matter and essential nutrients,  thus,
 the presence  of  methylarsine  was  not anticipated,

 4.3.2.4  Landfills

      In  a  national study  of subsurface migration of pollutants at  selected
 industrial waste land  disposal  sites  (Geraghty and Miller 1977) ,  arsenic
 was found  to  be  present at 37 out of 50 sites and at 30 sites to have
 migrated to ground water.  The  sites were either active for more  than
 3 years or abandoned,  including lagoons and landfills from eleven East
 Coast states.  Concentrations in  ground water at distances of 30.5 in  to
 152.5 m from  the  disposal site  ranged  from 0.04 mg/1 to 5.8 mg/1  (four
 samples) (see Table 4-6).  All  other measured concentrations were lower
 Sampling was  not  conducted any  further laterally than 610 m from anv
 site  to indicate  the extent of arsenic migration.  Arsenic was the
 second most frequently encountered inorganic (after selenium) to exceed
 the drinking water standard of 0,05 ng/1.

     Eichenburger et al.  (1978) studied inorganic constituents of hazard-
ous solid waste discharges of five Class  I landfill  sites  in Los Angeles
Seventeen industrial waste streams were studied.   Arsenic was found to
                                  4-27

-------
           TABLE 4-6.  ARSENIC CONCENTRATIONS  IN GROUND WATER
                  NEAR INDUSTRIAL WASTE DISPOSAL SITES
 Site


 Connecticut

 Indiana

 Massachusetts

 New Hampshire

 New Hampshire
Concentration in
in Groundwater
   (ug/1)
     40

  5,800

    650

    120

     30
Distance
From Disposal
Area (ia)	


     30.5

     36.6

     6.1

     61.0

   152.5
Total Depth
	(m)
    5.2

     NA

    4.0

   12.2

   11.9
Source:  Geraghty and Miller (1977).
                                    4-28

-------
  be the third lowest of the 17
                                             .
 JT-s: s^u,; -M:i:S s^S^l-Xr-"
 »
40 years Ut.r (U.S. EPA W75 IE! P°«™l-g of 11 people approximately
and use of a drinking « tar well ^°T" °CCU"ed following drill?™
from the disposal siL. ^"        '' "' '»«»*«?«l " ^
4-3.2.5  Summary


                        ,._ *    .      ^
          aa                 ,   . of ,u
                    4-29

-------
 at which arsenic is less mobile.  However, high arsenic levels have been
 measured in the vicinity of gold, stibnite, and other mining sites.
 Concentrations as as high as 1.5 mg/1 were measured in runoff; at one
 site, surface soil levels (total As) were 6,138 mg/kg and, at 45-60 cm
 deep, as high as 798 mg/kg.  Arsenate, arsenite, and dimethylarsine have
 been detected at tailings disposal sites, with arsenate predominating
 (at least under aerobic conditions).  Arsenic is commonly detected in
 ground water in the vicinity of industrial landfill sites; the highest
 reported level was 5,800 yg/1.   There is a long-term potential for sub-
 surface migration of arsenic into ground water as illustrated by two
 incidents of contamination more than 30 years following disposal of
 arsenic-contaminated waste.

 4-3.3  Pathway 3—Direct Discharge to Surface Water

 4.3.3.1  Sources

      Industrial discharge of arsenic in effluents  contribute only a small
 fraction,  3%,   of all  environmental releases.   [Note that  the 5%  calcu-"
 lated in the Materials Balance  Chapter 3.0 also  includes  contributions
 from soil  runoff following  pesticide application to land;  this  nonpoint
 source is  not  considered in  this  pathway].  The  most significant  known
 sources  are  zinc production  facilities and use of  arsenic-contaminated
 detergents.  Other  potential sources  include natural releases from weather-
 ing of bedrock (an  estimated 6.700  kkg annually) and discharges from POTWs
 (an estimated  1,800 kkg annually).   The forms  of arsenic released are
 unknown; however, various compounds would  be expected to be present in-
 cluding  inorganic  complexes  associated with phosphorus and smelter ores,
 simple arsenates  and arsenites.

      Relatively little fate  information was available concerning  aqueous dis-
 charges  of arsenic  resulting from human activities  as compared with its
 other environmental pathways.  Due  to  the  small contribution of these
 releases to  the  total  environmental loading of arsenic, it was thought
 appropriate  to  limit this discussion to available data and not supplement
 it with use  of  calculational models and other estimation techniques.  Most
 of  the available data  concerned the various arsenic  forms detected and
 their  fraction  of the  total  concentration measured in different types of
 water bodies.   In addition,  some information was available on release of
 sediment bound arsenic  to the water column.  Arsenic concentrations measured
 in aquatic organisms and the mechanism of uptake are also discussed in this
 pathway.  Wastewater treatment of industrial effluents is discussed in Path-
 way 4.

4.3.3.2  Field Studies of Arsenic in Aquatic Systems

     In a study of the ASARCO smelter's aquatic discharges  (Crecelius
et ad. 1975), sediment in Puget  Sound was found to  contain arsenic at
levels up to  10,000 mg/kg and surface seawater in the immediate  vicinity
of the plant  to contain 1.2  mg/1.   Within 1.6  km,  the water concentrations
                                 4-30

-------
 were reduced  to 4  ug/1, which was attributed to sorption onto sediment.
 Sediment levels at this location were not given.  STORE! data were
 examined for  this area and the results are discussed in Chapter 6.2.

      An estimated 55% of arsenic discharged to Lake Washington by a
 smelter was removed by sediment, primarily by sorption on or coprecipi-
 tation with iron and manganese compounds (Crecelius 1975).  Some of the
 remaining sediment-associated arsenic was thought to have entered the
 lake already bound to smelting-waste particulates.  In aquatic systems
 receiving a larger fraction of dissolved arsenic, Crecelius expected a
 lower amount of arsenic to end up in the sediment.

     Jfeslenchuck (1978) analyzed arsenic speciation in the Continental
 Shelf waters of the Southeastern U.S.   Total arsenic concentrations,
 averaging 1.1 ug/1, in the Georgia Bight were found to be controlled
 primarily by Gulf Stream intrusions.  Upstream river concentrations
 (average 0.3 ug/1 dissolved As)  were found to be controlled by rain-
 water dilution and complexation with dissolved organic matter which
 prevents adsorption of arsenic on solid phase organic and inorganic
 substances.   Arsenate was the predominant form present;  arsenite and di-
 methyl-arsenic concentrations were an  order of magnitude lower than ar-
 senate levels and comprised approximately 20% of the total  arsenic levels.
 Arsenate concentrations in the water column were lower than satura-
 tion levels  due to  processes  removing  the free arsenate  ion from
 solution.  The residence  times estimated for total arsenic  was  0.23
 years  in the rivers and 0.20  years in  the Bight.

      Following precipitation  and/or adsorption of  arsenic into  the sedi-
 ment layer of aquatic systems,  there is  a potential for release to the
 water  column.   Arsenic was  released to water  to  a  greater extent and at
 slightly faster rates from  anaerobic muds  than  from  aerobic  muds  (Faust
 and  Clement  1980).  The study  used  polluted muds from New Jersey surface
 waters  used  for recreational purposes and as a secondary water supply.
 By 28 days under anaerobic conditions,  up to 60% of the initial 19 yg/V
 arsenic in mud was released to the water phase to make up water concentra-
 tions of 1.2 mg/1.  At higher mud concentrations, 40% of an initial 381
 yg/g in mud was released resulting in a water concentration of 12.6 mg/1
 Release was greater by 70% at 25-30*C than at 20°C; but at 37°C  release'
was 40% lower than at 25-30'C.  The greater release of iron from anaerobic
 as compared with aerobic muds suggested that some of the arsenic releases
may be attributable to dissolution of ferrous arsenite compounds.   The
 anaerobic muds were generally more acidic than the neutral aerobic muds,  which
may have been a cofacter in the difference in As release rates from the muds.

     Braman and Foreback (1973) analyzed samples from surface, ground
 and estuarine waters in the area of Tanpa, FL.  Samples were analyzed
 specirically for arsenate, arsenite, and the methylarsonic acids.
 Table 4-7 reports the concentrations found.   The predominant form was
 arsenate (V)  with several exceptions.  In the two freshwater lakes,
arsenite (III) levels were highest and  in the saline McKay Bay,
                                  4-31

-------
            TABLE 4-7.  ARSENIC  COMPOUNDS  IN  TAMPA, FLA WATERS




                                 Concentration (ug/1)
Body of Water
Hillsborough River
Withlacoochee River
Well water near
Withlacoochee River
Remote pond,
Withlacoochee Forest
University Research
Pond, University of
Southern Florida
Lake Echols, Tampa
Lake Magdalene, Tampa
Saline Waters
Bay, Causeway
Tidal Flat
McKay Bay
As (III)
<0.02
<0.02
<0.02
<0.02
0.79
2.74
0.89
0.12
0.62
0.06
As(V)
0.25
0.16
0.27
0.32
0.96
0.41
0.49
1.45
1.29
0.35
Methylarsenic
Acid
<0.02
0.06
0.11
0.12
0.05
0.11
0.22
< 0.02
0.08
0.07
Dimethylarsenic
Acid
<0.02
0.30
0.20
0.62
0.15
0.32
0.15
0.20
0.29
1.00
Total
0.25
0.42
0.68
1.06
1.95
3.58
1.75
1.77
2.28
1.48
Source:  Braman and Foreback (1973).
                                   4-32

-------

                                      n
       arsenic  contenT*^ arsenlte a/^r
 fraction methylarsenic acid  In Lake
 were comprised of arsenite and/or
 dimethylarsenic acid (9%)  and
 compounds vere not
Introductio
                                                              «•
                                                 P°nd'  46% of the
                                                   the
                                                    arsenic
                                             methy^rsenic  acid  (6%)
of arsenic in organisms; since for the
concentrations of arsenic were not
not be calculated.  Some general
                                                 back§^^d  levels
                                      part> water a^ sediment
                                                     f^tors can
        ments.
                                          in  both  environ-
                             ln the

                          4-33

-------
      Metabolism

      Accumulated arsenic in higher aquatic organisms is reported to be
 converted to detoxified organic form and excreted.   Ingested inorganic
 arsenic was converted by fish to organic while injected and gill-absorbed
 arsenic was not (Lunde 1972,  Penrose 1975).   Intestinal microflora were
 implicated as responsible for at least the first stage in bioconversion
 (Penrose _et al. 1977).   The extent of arsenic metabolism and excretion
 in fish in general and the nature of the differences between freshwater
 and marine species are not well understood.

      Bioconcentration  Factors

      Four arsenical herbicides—MSMA,  DSMA,  arsenic  trioxide and
 pentoxide—were tested for their potential to bioaccumulate in fish
 (rainbow trout, bluegill),  scud,  daphnia,  and snails (Spehar _et al.  1930).
 There was little difference between the  herbicides for all aquatic
 organisms tested;  bioconcentrations factors  ranged from <1 to 20.   In
 another study,  a BCF of 4  in  bluegill  for  arsenic trioxide was measured;
 a  clearance half-life  of less than 1 day was  also estimated.
 Based upon these studies and  additional  observations made  by  Isensee
jat  al.  (1973),  a BCF for arsenic  in fish in general  was  estimated  to be
 usually less than  10.

      Reports of tissue  levels of  1 mg/kg (wet  weight)  up  to  100 mg/kg
 in  fresh and marine species (see  following section)  suggest higher BCF's
 than those measured under  laboratory conditions.  This could  be attributed
 to  an equilibrium  in tissue levels not being  reached  before  termination
of  the  experiment or more than one  pathway  for  arsenic  uptake  by fish in
the  environment.  The form  of  arsenic involved is assumed  to  be  primarily
inorganic  (tri-  and  pentavalent).

      Detectable accumulation  of arsenic  by aquatic organisms  is  not
always  evident  in arsenic-polluted areas.  Fish  collected  from  a lake
21  days  after application of  sodium arsenite showed no significant
accumulation  of  arsenic.

      Monitoring  Data

      Numerous monitoring programs  for arsenic levels in fish  and inver-
tebrate  tissues  have been conducted both on a national scale  and at
specific sites.  The STORET data base contains 475 observations  in 13
major river basins  from  the past 3 years (see Table 4-8).  Mean  levels
in  fish  (various species) ranged from 0.1 mg/kg to 4 mg/kg with  a maximum
level of 700 mg/kg observed in the Western Gulf.

      The National Pesticides Monitoring Program sampled freshwater fish
from monitoring  stations in freshwater throughout the country during
1971-1973.  Maximum arsenic residues were 3.40 mg/kg in 1971, 1.70 mg/kg
in 1972, and 1.24 mg/kg in 1973.  Arsenic residues greater than 0.5 mg/kg
were  found in fish  (both predatory and non-predatory) from five stations,
                                  4-34

-------
             TABLE 4-8.   AMBIENT ARSENIC CONCENTRATIONS IN FISH
                         TISSUE - UNREMARKED DATA IN STORE!  1977-1979
  Major River
                                                     .Concent r a t i on
  North Atlantic

  Southeast

  Tennessee River

  Ohio River

  Upper Mississippi

  Lake Michigan

 Missouri River

 Lower Mississippi

 Western Gulf

 Pacific  Northwest

 California

 Great

 Lake  Superior

      Gross Analysis
(mg/kg -
Observations Mavfm,im
3
2
10
18
127
54
56
99
35
56
4
4
7
475
0.2
0.6
4.0
4.0
2.0
2.0
2.0
0.8
700.0
0.8
0.2
0.8
2.0
700.0
wet weig]
Meai
0.2
0.5
3.6
4.0
1.8
2.0
0.2
0.1
0.6
0.3
0.1
0.6
2.0
2.6
Source:
                                 4-35

-------
 in the Tombigbee River,  LA,  Lake Michigan,  Lake Superior,  and Red River.
 OK;  the fish from the Great  Lakes had high residues more frequently than
 the  others (Walsh et al. 1972).   No information on water or sediment con-
 centrations was available.   However,  STORE! data for the last 5 years
 indicated higher than national mean levels  in the Great Lakes major river
 basin.

      The National Marine Fisheries Service  surveyed the trace element
 content of the U.S.  fishery  resource  (Hall  et al.  1978).  Two hundred
 four species  were sampled from 198 sites  in seven U.S.  coastal areas.
 Average levels of arsenic varied widely for finfish muscles and livers,
 whole fish,  and molluscs,  with most species containing  between 2.0 mg/kg
 and  5.0 mg/kg in all categories.   Crustacea had somewhat higher levels,
 with most species containing between  4.0  mg/kg and 5.0  mg/kg.   Water con-
 centrations  of arsenic were  not  given, but  arsenic concentrations in sea-
 water are believed to  be rather  constant, ranging  between  1.0  ug/1 and
 1.5  ug/1 (Saunders 1978, Waslenchuk 1977, 1978,  Andreae 1978);  these
 levels  would  indicate  BCF's  of up  to  500.

      A  survey of natural background levels  of arsenic in marine  fish and
 invertebrates near a proposed mine site on  the  Canadian coast  indicated
 that levels in crustaceans were  higher than those  in fish  from the same
 locations (LeBlanc and Jackson 1973).  The  range of  levels  for all organ-
 isms tested was 0.4  to 27.8  mg/kg.  The median levels in the  crab,
 Cancer  magister (6-8  mg/kg),  were about two  times greater than  the levels
 in-fish.   The authors  postulated  that the arsenic  in these  organisms,
 particularly  the  levels  in crustaceans, could not  be present as  an
 arsenious oxide or any other known toxic form because of the high levels
 present.

      One  reason for  higher levels  in invertebrates in a  number of studies
 may  be  that benthos, and other detritus, suspension  and  filter-feeders
 have  access to  several sources of  arsenic-in the water column, par-
 ticulate  matter and  in sediment in which concentrations often  exceed
water levels.

     Wagemann  et al.  (1978) measured concentrations  of arsenic in  aquatic
 invertebrates, macrophytes, sediments, and water in  the vicinity of Yellow-
knife, Canada.  In arsenic-contaminated lakes (a consequence of gold mining
activities), arsenic concentrations ranged from 700 ug/1 to 5,500 ug/1 in
water, 6 mg/kg  to  3,500 mg/kg in bottom sediments, 150 mg/kg to 3,700 mg/kg
in macrophytes, 700 mg/kg to  2,400 mg/kg in zooplankton, and <1 to 1,300
mg/kg in other invertebrates.

     Arsenic concentrations were  measured  in clams, plankton, and sedi-
 ments from Lake Washington and Sardis Reservoir located in rural, agricul-
 tural areas in Mississippi (Price and Knight 1978).   Samples" were gathered
 monthly from October 1975 to May 1975.  Mean concentrations in mg/kg wet
 weight were 2.99 in  sediment and 21.74 in plankton.   The ranges and means
 of arsenic in eight  species  of clam were  also reported. Means ranged from
 0 mg/kg to 0.36 mg/kg (wet x^eight) and  the total range  was <'0.01 mg/kg  to
 1.39 ing/kg.


                                   4-36

-------
 4.3.3.4  Summary

      Only a small fraction (3%)  of the total industrial releases of
 arsenic are made directly to surface water.  Other sources include
 natural weathering of arsenic-containing minerals and soil and dis-
 charges from POTWs.  Sediment serves as a reservoir for a significant
 fraction of arsenic discharged into both freshwater and estuarine systems.
 Other controlling factors on concentration are rainwater dilution and
 dissolved organic matter content.  Sediment-bound arsenic may be subse-
 quently released to the water column  following  methylation by both
 aerobic and anaerobic sediment microbial populations..

      Arsenate was the predominant form in a number of  freshwater systems;
 arsenite was the predominate form in  two  lakes  and dimethylarsenic'acid
 in an estuary.   Other arsenic compounds detected  in freshwater bodies
 were methylarsonic  acid and  trimethylarsine.  Bioconcentration of arsenic
 has been detected in  marine  and  freshwater fish and invertebrate species.
 Fish tissues generally contain less than  10  mg/kg while  invertebrates
 (i.e.,  shellfish) sometimes  accumulate  higher levels.  Arsenic does not
 appear  to biomagnify  in the  food  chain.

 4.3.4  Pathway 4—Wastewater Treatment

 4.3.4.1  Sources

      Arsenic is present in the process  waters of  numerous  industries  and
 in the  influent to Publicly Owned Treatment Works  (POTWs).  Table 4-9
 list  industries  for which the Effluent Guidelines Division has  found
 levels  of arsenic in the raw waste; the associated concentrations are also
 presented.   Some of the highest mean  levels  of  arsenic (total)  were
 found in  the raw waste  of inorganic pesticide manufacturers,  nonferrous
 metals  manufacturing operations, and  steam electric power  plants.
 According to  Chapter 3.0  (Materials Balance), approximately 1,800 kkg
 of  arsenic enters U.S. POTWs  annually.  Up to 1,050 kkg of that amount
 may come from urban runoff; less than 60 kkg appears to be discharged
 directly to POTWs by industries.

      The degree  to which arsenic is removed from municipal and  industrial
 wastewaters and  thus the ultimate distribution of arsenic between POTW
 aqueous effluents and sludge depends upon the type of treatment involved.
 Primary and secondary treatment techniques are not very effective at
 removing arsenic.  The efficiency of activated sludge treatment (secondary
 process) is variable.   Tertiary treatment techniques are generally quite
 effective.  The following section discusses (1)  primary and secondary
 treatment of arsenic particularly in POTWs and (2) tertiary treatment
 techniques in general.

4.3.4.2  Primary and Secondary Treatment

     According to the  Materials Balance (Chapter 3.0),  POTWs  are virtuallv
ineffective  at removing arsenic from water.   This  conclusion  is based  on '
                                   4-37

-------
                           TABLE  4-9.   OCCURRENCE  OF  ARSENIC  IN RAW INDUSTRIAL WASTEWATERa'b
                                                                  Raw Wastevater
                                               Concentration  (uol
                Industry
 Coal mining
 Textile mills
 Timber products processing
 Petroleum refining
 Paint and ink formulation
 Gum and wood chemicals
 Auto and other laundries
 Porcelain enameling
 Pharmaceutical manufacturing
 Ore mining and dressing
 Steam electric power generating
    (condenser cooling system)
 Steam electric power generating
    (water treatment)
 Steam electric power generating
    (boiler or steam generator blowdown)
 Steam electric power generating
    (maintenance cleaning)
 Steam electric power generating
    (ash  handling)
 Steam electric power generating
    (air  pollution  control devices)
 Steam electric power generating
    (drainage)
 Inorganic  chemicals  manufacturing
 Coil  Coating
 Foundries
 Nonferrous metals manufacturing
 Iron  and steel manufacturing

 ^Information contained in this  table was obtained from Volume II of the Treatability Manual - U.S. EPA  (1980b)
 °NA - not available; ND - not detected; BDL - below detection limit.                                    M*ovoj.
 CSiJHe~!l 1°adlnSS de*?r!^ne?,5y ""Implying ae.n Pollutant concentration by industry wastewater discharges.
 Where mean is not available 1/2 the reported maximum was utilized.                      '              nargea.
 dMedian,  not average.
^Average  of medians reported for various industry segments.
1Average  of maximums reported for various industry segments.
SOne sample.
Concentration (us/I)
Minimum
< 2
NA
BDL
3
BDL
<10
ND
5
ND
< 1
4
NA
NA
5
BDL
< 4
NA
NA
758
ND
ND
NA
Maximum
250
• 200
14,000
480
800
110
1,600
2,800
120
110
35
NA
NA
310,000
74
300
NA
956f
75S
160
310,000
440
Mean
< 86
10d
10e
< 20d
73
< 50
68
960
13
< 20
7
9,500e
NA
41
9
150
NA
NA
75S
29
13,000
120
Loadins c Ocz/d)
Minimum
0
4.2 x 10"5
0
0
0
0.00045
6.1 x 10"5
0.00061
0
0
0.0004
6 x 10"5
NA
4.1 x 10~8
0.00017
0.00014
NA
0
0.004
0
0
0
Maximum Mean
2.3 0.32
0.29
0.46
1.9
0.0033
0.38
0.095
0.39
0.01
NA
0.008
570

0.78
0.88
8.6
NA
60
0.135
0.19
NA
NA
0.018
0.00086
0.12
0.00012
0.82
0.014
0.077
0.088
<0.74
0.0017
1.4
NA
0.012
0.20
0.54
NA
12
0.036
0.08
689
22
                                                   4-38

-------
 the results of a 20-plant study (see Table A-7 in Appendix A).  Due to
 the high detection limits f~50 yg/1) associated with the analytical
 methods used at some of the plants which often exceeded arsenic levels
 in the influent, it was not possible to detect a decrease in arsenic
 following treatment.  This analytical problem, combined with the obser-
 vation that arsenic did accumulate in the sludge of POTWs,  indicates that
 at least some removal is occurring.

      In a separate study of U.S.  POTW effectiveness, arsenic was reported
 to be removed at efficiencies of 12-43% in primary treatment and 43-85%
 in secondary treatment (Gilbert 1980).   In another study, arsenic con-
 centrations in domestic sewage sludge from ten U.S. POTWs were 5 mg/kg
 to 46 mg/kg (assumed dry weight)  (Jones and Lee 1977)  as compared with
 the average 0.2 mg/kg (dry weight)  reported in Chapter 3.0.   Therefore,
 it appears that POTWs can be somewhat effective at removing arsenic;
 however, the efficiency of removal  varies considerably among plants.
 This subject requires further investigation.

      In a survey of pollutant concentrations  in the final effluents of
 municipal waste dischargers  in California,  arsenic was detected at five
 plants.   Average effluent concentrations ranged from 1 ug/1  to 280 ug/1,
 with four out  of five averages less  than 15 yg/1 (CWRP 1977).   There was
 no obvious relationship between high levels and the level of treatment
 or plant flow  rates.

 4-3.4.3   Tertiary  Treatment

      Removal of  arsenic by tertiary  treatment  is best  achieved by  adjust-
 ing the  pH of  the  commonly acidic wastewater  to alkaline  conditions;  in
 wastewaters  containing  other  heavy metals,  removal  of  arsenic  is partially
 achieved by  coprecipitation with these metals.   The most  common and  effec-
 tive  treatment  technique  is to remove arsenic  by coagulation  (with Fed,
 or FeaSoit),  sedimentation, and filtration.  Up  to  95%  of  the arsenic
 initially in the waste  stream  may be removed by this treatment  technique.
 The fraction removed  is  then disposed of  in solid or semi-solid form  to
 land.

     Week  (undated) studied the removal of metals by tertiary wastewater
 treatment  processes such as (1) "high lime" treatment, in which lime  is
 added to  bring the pH to 11, FeCl3 is added as  a coagulant, and the
 effluent  is  fed through sand and carbon columns; and (2) "alum" treat-
ment, in which aluminum sulfate and lime are introduced to promote coagu-
 lation followed by carbon and  sand filtration.  For arsenic, the alum
 treatment was most effective  although removal  of arsenic in the carbon
 columns was  three times more effective following lime treatment than
alum treatment.  Table 4-10 summarizes treatment efficiencies for arsenic
associated with different treatment methods.

     In a water treatment process plant on the Kansas River,  Angino et_
al. (1970) reported lime softening to be effective at removing 85% of~
the water's total arsenic concentration.   However,  other sources contradict
                                  4-39

-------
              TABLE 4-10.  ARSENIC REMOVAL EFFICIENCY OF
                           WASTE TREATMENT METHODS
                         Concentration (mg/1)
Treatment

Lime Softening

Charcoal Filtration

Ferric Sulfide
Filter Bed

Coagulation with
Ferric Sulfate

Coagulation with
Ferric Chloride

Precipitation with
Ferric Hydroxide by
Ferrous Sulfate &
Lime Coagulation
with Settling only

Ferrous
Sulfide Filter
Bed, Bone Carbon &
Settling

Ferrous
Sulfide Filter Bed,
followed by Sand &
Coke Filtration

Chemical
Coagulation,
Sedimentation, and
Filtration
Initial
0.2
0.2
0.8
Final
0.03
0.06
0.05
% Removal
85
70
94
     25.0
      3.0
      0.8
      0.5
5 or less
  0.05
    362          15-20
(as  Arsenic   (as Arsenic
  Oxide)         Oxide)
  0.05
  trace
    1.0-2.0      0.05-0.5
 80 or more


 98


94-96
 94
                 80-95
Source:   Week (undated).
                                   4-4C

-------
         finding.
   are generally very iuw across the  *n^->.  « *  ,   	•""» wacer
   observation indicating either low  inM V    (see Section 5.2), an
   treatment techniques.             initial concentrations or effective


   4.3.4.4  Summary
                                         during primary,  secondary, and
 ? Jf an? llme soft^ing processes,
 at removal.  However tertiary treat
 to primary and secondary treatment,
 applications.  The effectiveness of pri

 sss!t«SLf ssrr.'^"
 treatment processes,  vary in eff
 effectiveness of U.S. POTWs on a
 tigation.

 4.3.5


 4-3,5.1  Introduction
                                         ~
                                           ^-fect±ve' USually
                                           r6J-a"Ve1^ "Common compared

                                             ^11* 1±m±ted t0 Austria!
                                        to      SecOnda^ treatment in
                                        to be quxte variable within

                                        Trom  O^  PrfMry ^ Sec-da^
                                             ?   UP t0 85%!   The  ove^ll
                                          scale  requires further  inves-
           DSMA (6%), cacodylic acid  77
arsenxte  (making up remaining
                                                   ch rar in the
                                                   producti°n) , arsenic
                                                     Deluding sodium
      arsenic forms are expected to b*»   ^     r°m SOil s^faces) .
 -nt;  however the use of  sodium arsenitl^s  I diStrlbuted in  the eniiron-
 surface water a significant receivC^Ltlclde may make
                                             inorganic and organic
            c acid (HAA)  and Us       and^r5^" arsenicai*  inclde
DSMA respectively) , and  cacodylic ac?d (CA   dl;S°dlum salts (MSMA and
(Woolson 1977).  Organic arsenirfi         °r dlmethylarsinic  acid)
             amounts of i»orgMic and organic  arsenical pestici,es
                               4-41

-------
 produced annually are approximately equivalent;  however, organic forms
 are usually applied at lower concentrations  than are inorganic forms
 (Sandberg and Allen 1975).   MSMA and DSMA are often applied at rates
 of less than 10 kkg per hectare in one application while sodium arsenate
 and,  in the past,  lead arsenate,  are applied in  excess of 100 k2 oer
 hectare.   Most of the agricultural sites,  such as orchards, with high
 residue levels of arsenic were  treated with  inorganic arsenicals.

      Pesticide Drift

      Most arsenicals are applied  in a spray  form.   Although the chemical
 forms used are not volatile,  there is a potential for drift during  treat-
 ment  to non-target areas, both  soil and water.   MSMA levels in the  water
 in a  sprayed irrigation ditch reached up to  860  ug/1 immediately follow-
 ing treatment  but  declined  to less than 50 ug/1  within 2 hours (Salman
 _et al.  1972).   Loss of arsenic  from surface  water is dependent upon water
 volume,  flow rate  and other  factors described  in Pathway 3.   The total
 amount  of arsenic  pesticide  lost  through drift during normal application
 procedures is  expected to be minimal  (Sanberg  and Allen 1975).

      Adsorption

      On reaching  the soil,  the  chemistry of  organic  and inorganic
 arsenicals,  regardless  of the form,  is  essentially  the  chemistry of
 arsenate,  provided that  aerobic conditions prevail  (Walsh and  Keeney
 1974).  Activity of arsenicals  is  reduced due  to  immediate  adsorption
 by  soil  colloids  (Hiltbold 1975) and  formation of  insoluble  salts
 (Walsh  and Keeney  1974).  The water-soluble  fraction of cacodylic acid
 decreased over  a period  of 8 weeks  and was replaced  by  a less  soluble
 form  of arsenic associated with aluminum (Hiltbold 1975).

      Adsorption of  arsenic is highly  correJated with  soil clay  for all
 arsenic content (r  =  0.81-0.96;  Hiltbold 1975).  In  clay soils with  a
 high  anion exchange  capacity (e.g., high kaolinite content), arsenic
 (as arsenate) can be  adsorbed at quantities up to 1% by weight of
 the clay  (Pax 1973) depending on the presence of other  competing
 anions  (especially  S0u=, NO3-).  Addition of these competing substances
 in  fertilizer could also displace arsenic already adsorbed in the soil.
 Clay  soil  adsorbed up to 2,270 ug As per gm of soil at  equilibrium
 (Wauchope  1974).  Soils high in  iron or calcium and also 'organic soils
 are likely  to fix arsenic in insoluble form through complexation and
 chelation  (Cooper _et al. 1932).   Examples of  soils with a high arsenic
 fixing capacity include red and  yellow podzols, latosols, arid soils,
and limestone soils.  Subsoils tend to have a greater holding capacity
 for arsenic than do surface soils, presumably due to higher concentra-
 tions of clay and iron oxides (Hiltbold et al. 1974).  In all soils
 there is always potential for subsequent resolubilization of the fixed
arsenic following changes in soil conditions  including a rise in pH in
high  iron soil, a drop in pH in  lime soil, or a change in redox poten-
tial.
                                  4-42

-------
        Leaching
                                                on the leaching of arsenical
        field studies oTino^ l^o'^lf^ ^ —i-s the resulL
   some  leaching  occurred  within  the  too  30 ,   a"en^5ala-   In ««t soils,
   soils  than in  clay or loam  soils   No  nV   "f,8011'  m°re So in Sand7
   deeper  than 90 cm! Environmental  ff°  detectable  «««ic  was found
   of arsenic  include  the  initial  «sfnlr    lnStrUmental in the mobility
   tation and  soil characteristics such   C°"Centration' »te  of precipil
  does not appear to haveTsignifSant Ln30^"^ Capacity-  Soil PH
  potential mobility of organic ar      lnfluence on dissociation and
  may have some influence on the availabilitvi:Lfb0ld 19?4) '   However> PH
  which in turn influence arsenic mobility,      ^^ ^ aluminum in soil

       Surface Runoff
                                                 -  «*•« water via
 o  arsenic  transferred in su^LcTrunoff ? prflpitatio^  The amount
 (i.e., slope), vegetative cov£  anHfL^??    °D the 1OCal terrain
 rate of 2.78-19.72 tons of sol ''pe r a"  4 f es^"?1!?', A" annual erosion
 soxls in Missouri (rainfall  = 40 inchest  ?BrJ  ^7/f  f°r a§ricultural
 arsenic level of 10 mg/kg, an estimated 60  1^5Y   ?*'   *°*  S0±± ^th an
 would be lost annually through surface runoff   §,    arS6niC  P6r hectare
 total amount of  arsenical pesticides  Ll^ ,   f  estimated ™ of the
 surface water each  year (Chapter 3 0)                 **  transferred to
                                        oxidation due  tc
at 2% per month"inflight"textured* I* °Xldation for MSMA was estimated
in organic clay soiis (mitbold 1974)  °F^   S°±lS ^ 10% per month
applied cacodylic acid was oxidized tn a^   1°^ Percent of initially
period (Woolson and Kearney 1973)       arsenate and C02 in a 24-week

     Volatilization
 forms resulting from microbial tr
 of arsenic lost by volatilization
 cacodylic  acid  was lost  in  a  2
 anaerobic  soils,  possibly as  dimethyl a
 ine rate of  transformation  is  dependent
 influencing  microbial activity:   organic
mcrobial  species  present.  Even  thou

SairoewhinShly-T°latile' th£ ^ansfer fro
In the, S°   nth S nonP°ro^ texture
In these situations, oxidation to the me
occur before the arsenic  moves very far
                                                          both  organic
                                                  on of methylated
                                          ic      /°me CaS6S  the amo
                                                  Approximately 35% of
                                                 °^ S0il and 6U ^
                                       on  h  °°1SOn and Kea^y 1973)
                                        co ^ environmental factors
                                        £ nten5» PH' moisture content,
                                         f methylated forms of arsenic
                                                  atm°Sphere **7 be
                                                  V6§etative growth.
                                                  a"enate form may
                               4-43

-------
                      TABLE  4-11.   FIELD  STUDIES  ON  LEACHMG  OF  ARSENICAL PESTICIDES THROUGH SOIL
Form
Applied
Arsenic
Arsenic
Application
Rate
NGa
NG
Soil
Type
clay adobe
silt loam
Result
Concentrations of 441 mg/kg
in surface 20 cm of soil but
10 mg/kg in next 20 cm.
Concentrations of 61 mg/kg
in surface 20 cm of soil
but 8 mg/kg in next 20 cm.
References
Jones and Hatch
(1937)
Jones and Hatch
(1937)
 Pb  arsenate


 Pb  arsenate



 Arsenic trioxide


 Sodium arsenite


 Sodium arsenite


 DSMA


 DSMA


 DSMA


 MSMA


 DSMA


 MSMA


MSMA



MSMA




Cacodylic acid
     NG       orchard soil
              orchard soils
1,120-8960    silt loam
  kg/ha

90-180 kg/ha  sandy soil
760 kg/ha     sandy soil
112 kg/ha     clay loam
112 kg/ha     loamy sand
  9 kg/ha     sandy loam
  9 kg/ha     sandy loam
18-36 kg/ha   sandy loam •
18--36 kg/ha   sandy loam
 40 kg/ha
     NG
     NG
silt loam, fine
sandy loam and
loamy sand

silt loam, silty
clay loam and
loam soil from
forest floors

silt loam, silty
clay loam and
loam soil from
forest floors
Arsenic not  leached  below
root zone

Water  soluble arsenic concen-
trated in top 15  cm  soil;
little found below 61 cm.

No arsenic residues  below
36 cm deep.

Arsenic leached to 38 cm deep
3 years after application.

Arsenic leached to 68 cm deep
3 years after application

No leaching below 15 cm with
51 cm of water.

52% moved 23 cm deep with
51 cm of water.

No leaching below 30 cm
with 580 cm water.

No leaching below 30 en
with 580 cm water.

25% leached below 30 cm defp.
None detected at 90 cm.

75% leached below 30 cm deep.
None detected at 90 cm.

No leaching in any soil
below 30 cm over 6 vears.
                                No leaching in any soils.
                                                   84% leached 8 cm with 86 cm
                                                   water in silt loam and 50%
                                                   in loam.
                                                                   Benson  (1968)
                                                                   Vandecaveye et al.
                                                                        (1936)
 Arnott and Leaf
      (1967)

 Steevens  et al.
      (1972)

 Steevens  e£ al.
      (1972) ~

 Dickens and Hiltbold
      (1967)

 Dickens and Hiltbold
      (1967)

 Johnson and  Hiltbold
      (1969)

 Johnson and  Hiltbold
      (1969)

 Johnson and  Hiltbold
      (1969)

Johnson and  Hiltbold
      (1969)

Hiltbold jat  aj,.
      (1974)
                                                                  Norris (1974)
                                                                  Norris (1974)
 NG » not given.

 All soil concentrations assumed to be for total arsenic.
                                                  4-44

-------
 4.3.5.3  Phosphate Fertilizers

      Both natural and synthetic phosphate fertilizers contain signifi-
 cant quantities of arsenic as an impurity.  Senesi et _al.  (1979)  examined
 32 commercial fertilizers for concentrations of arsenic, bismuth
 lithium, and selenium.  Arsenic levels ranging from 2 mg/kg (calcium
 cyanamide) to 321 ing/kg (triple superphosphate) were detected in all of
 the samples, and arsenic was generally present in greater amounts than
 any other trace element.  The presence of As in synthetic nitrogen
 fertilizers was attributed to impurities in HjSO^, HN03, catalysts,
 equipment corrosion, coaters and conditioners used to prevent caking,
 and materials such as gypsum and limestone, added to form a granular'
 product (Senesi et al. 1979).

      The amount of arsenic entering the soil annually through use of
 phosphate fertilizers can be estimated.  Assuming an application rate
 of 84 kg of phosphorous per hectare (Tisdale and Nelson 1968)  and an
 arsenic concentration of 4 mg/kg in the fertilizer, approximately 0.3 g
 of arsenic is added per hectare of  soil.   Application of fertilizer with
 arsenic levels  of 300 mg/kg would add 25  g of  arsenic per acre.

      Despite this evidence of possible arsenic  contamination of  soils
 via fertilizer  additions,  Goodroad  and Caldwell (1979)  found that a
 clay  loam and silt loam fertilized  with concentrated  superphosphate and
 a  mixture of phosphate fertilizers,  for 19  years  did  not accumulate
 arsenic as  compared with control plots.   For the  clay loam,  increased
 As levels were  not observed  to  a depth of 47.5  cm;  for  the  silt  loam,
 a  slight decrease was observed  in the C horizon (primarily  mineral
 layer deep  in soil) due  to  leaching  of CaCos to the C horizon with sub-
 sequent transfer  of arsenic  adsorption sites out  of the  higher horizons.
 Despite this  evidence,  at  typical rates of  fertilizer application,
 arsenic levels  in  all of these  fertilizers will probably add to  soil
 contamination over years of  application and  accumulations in animals
 fed these fertilizers.

 4.3.5.4 Plant Uptake

     Arsenic  occurs in soil in the vicinity  of geological deposits and
 from mining and smelting operations primarily via deposition of As-
 contaminated  particulates.  Also arsenic in  the form of organic
 arsenicals1 and sodium arsenite used as herbicide provides another
 pathway of exposure.   Plants accumulate arsenic either through root
 uptake  from soil solution, or, via leaf absorption of aerially-depos-
 ited arsenic.

     Some fraction of the accumulated inorganic arsenic may be trans-
 formed  to organic pentavalent arsenic (NAS 1977).   Since information
 on the fate of the inorganic form is limited, no general conclusions
 can be made.  More is known about organic arsenicals.   Organic arsenic
may be complexed to plant constituents and remain fairly stable (NAS 1977) ;

 Primarily monosodium methanearsonate (MSMA), disodium methanearsonate (DSMA)
 and cacodylic acid.

                                  4-45

-------
 a minor fraction was reported to metabolize and volatilize from leaves
 in coastal bermudagrass (Duble ejt _al.  1969).  In general, for numerous
 plant species and several organic forms of arsenic, studies with Cllf
 markers indicated little metabolism, but translocation of arsenic from
 the point of entry to other parts of the plant (NAS 1977, see Table 4-2).
 Levels toxic to plants  can therefore accumulate.

      Typical and maximum concentrations of arsenic in edible plants have
 been compiled in various sources (e.g., NAS 1977,  U.S. EPA 1980)  and are
 discussed in greater detail in Section 5.2.  In general, however, natural
 background levels may range from  0.01 mg/kg (dry  weight) to as high as
 3 mg/kg.   Arsenic levels in plants grown in treated soil (with lead
 arsenate,  arsenate and  arsenite salts, trioxide and organic arsenicals)
 range from trace levels to approximately 1,700 mg/kg (highest in tomatoes
 and lemon tree roots),  with the majority of levels between 0.2 mg/kg to
 10 mg/kg  (NAS 1977:   see Appendix A).   Little information was available
 on concentrations in grass treated with arsenical  turf improvers.   Con-
 centrations of 2 mg/kg  were measured in grass grown on soil treated to
 levels of  up to 550  mg/kg (Williams and Whetstone  1949).

      Variability among  plants  is attributed to species differences,
 various application  methods, different sampling  times, the influence of
 soil characteristics on uptake,  and other factors.   At an orchard site
 (treated  10 years previously)  the highest arsenic  levels  were found  in
 onion tops at 2 mg/kg grown  in soil with 233 mg/kg.  Levels  in cotton
 at 9 mg/kg were reported in  soil (at 28 mg/kg arsenic  previously  treated
 with MSMA)  (Johnson  and Hiltbold 1969).   Therefore,  total arsenic  level
 in soil by itself is not  a reliable indicator of potential arsenic levels
 in plants.   Kenyon et_ al.  (1979)  investigated the  availability of  arsenic
 to plants  grown in soil that had received lead and  calcium arsenate
 insecticides for years.  The results indicated that  the arsenic in the
 insecticide-treated  soil  (22.4 mg/kg as  opposed to  1.9 mg/kg  in the
 control) was tightly bound and did  not  translocate  into any erf the plant
 species grown.   The  factors determining  solubility  of  arsenic in soil
 (discussed previously)  are expected  to  determine actual plant uptake.
 The  pentavalent  form would be  the prevalent  soluble  form  in the aerobic
 layer  in which  plant  roots grow.  Certain agricultural practices such
 as  liming  and phosphorus application may  lead  to solubilization of
 bound  arsenic and  plant uptake.

 4-3.5.5  Application  of Arsenical Herbicides  to Aquatic Systems

     The total  amount of arsenic used as  an aquatic herbicide  is small
 on a national scale but its use may have  a significant impact  on a par-
 ticular system.  An example of the propensity of aquatic systems to
 accumulate and store arsenic in the sediment layer was illustrated in
 a  study of a lake receiving sodium arsenite herbicide  applications
 (Lambou and Lim 1970).  A Wisconsin lake had received  218,000  Ib of
arsenite,  which was estimated to be distributed at  87 Ib/acre or 380
ing/kg in the top 1-inch of sediment.  Measured levels were not too
different, ranging from 10 to 82 mg/kg with a mean  of 49 mg/kg.  The
                                  4-46

-------
 remainder of the arsenic (undetected) may have migrated deeper into the
 sediment or been methylated and released to the atmosphere.  Other
 studies have indicated depletion of sediment-bound arsenic over time
 (Lawrence 1957), probably due to biological activity.

      The sediment core profiles of lakes treated with sodium arsenite
 herbicides all showed sediment surface layer accumulation of arsenic,
 but no obvious correlation with iron, manganese, calcium or organic
 carbon content of the lake sediments (Kobayashi and Lee 1978)  as might
 be expected from observations in soil.   In sediment surfaces high in-
 organic matter, arsenic was easily leachable with distilled water (60%
 of the total  arsenic  present)  whereas  in highly calcareous  sediments,  the
 arsenic was more tightly bound (only 25% of total As leached).   In
 anoxic sediments,  34-49% of the total As was extractable by distilled
 water.

      MacKenthun (1970)  measured bottom sediment levels  of  sodium arsenate
 ranging from  10 mg/kg to 82 mg/kg  after treatment of a  pond at. 2.6 mg/1.
 Ten months after treatment  of a pond at 2.6 mg/1, no arsenic was detected
 in water;  sediment levels.were not reported (Dupree 1970).   However,
 one year following treatment of small  farm ponds which  had been con-
 secutively drained and refilled several times,  sodium arsenite  levels
 of 380 yg/kg  were  measured  in bottom mud and 300 ug/1 in water.   This
 indicated  long-term persistence of arsenic  in  sediment.  Other  chemical
 forms of arsenic were not reported so  it is not known whether or not
 some of the arsenic initially in trivalent  salt form was converted to
 other forms.

 4.3.5.6 Summary

      A significant  amount of arsenic is  used in agricultural applications
 as  an insecticide and herbicide  and is  applied  inadvertently as an
 impurity in fertilizers.  Some  arsenic  is also  used  as a herbicide in
 aquatic systems.  Arsenical  pesticides  include  inorganic forms such as
 sodium arsenite  and arsenic  acid and organic forms such MSMA, DSMA
 and  cacodylic acid.  During  pesticide application, a  small  amount  of
 arsenic may be lost or  transferred through drift.  On reaching soil,
 the  chemistry of all arsenicals is similar regardless of the form!
 adsorption by soil colloids  and complexation and/or chelation are  the
 most  important immobilization processes.  Soils high in clay content,
 organic matter, and available iron and aluminum have the greatest
 capacity for retaining arsenic.  Leaching of all forms of arsenic has
 been observed in a variety of soil types.  Movement appears to be con-
 fined  to the top 30 cm in most soils but migration to a depth of 90 cm
was observed in sandy loam.   Transfer of arsenic from soil to surface
water may occur in surface runoff.   Microbial transformations of organic
arsenicals to  arsenate and arsenicals to methylated forms may be signifi-
cant in agricultural soils.   Methylated arsenic forms are highly volatile
and can be lost from the soil to the atmosphere.
                                  4-47

-------
      Phosphate and other mineral fertilizers of both natural and syn-
 thetic origin may contain significant quantities of arsenic as an
 impurity.  Long-term use of highly-contaminated fertilizers is likely
 to result in the accumulation of high arsenic levels in soil.

      Plants accumulate arsenic from natural background levels in soil
 as well as from agricultural soils treated with arsenic.   Crops grown
 on treated soils generally have arsenic residues at levels less than
 10 mg/kg; a maximum accumulation of 1,700 mg/kg was reported in tomatoes
 and lemon trees.  The level of soluble arsenic in soil is a better
 indication of the potential for plant uptake than is the  level of total
 arsenic.

      Most of the sodium arsenite applied as a herbicide to aquatic
 systems migrates into the sediment layer and binds to inorganic and
 organic material.   Some of the sediment held arsenic is eventually
 released to the water column due to microbial methylation.

 4.4  DISTRIBUTION OF ARSENIC IN THE ENVIRONMENT

      This section presents information concerning actual  concentrations
 of arsenic detected in environmental media.   A literature review and
 STORET  were the main sources of monitoring data for arsenic  in  aquatic
 systems.   Observations concerning  arsenic  concentrations  contained  in
 the STORET systems  are discussed for ambient  and  effluent waters,  sedi-
 ment, dissolved and suspended matter.   Published  monitoring  data con-
 cerning arsenic concentrations are  presented  for  natural  water,  sediment,
 soil, the atmosphere and  terrestrial plants.   Environmental  media of
 significance  to a particular pathway or to  human  exposure are presented
 elsewhere.   For example,  fish tissue levels are discussed in Pathway 3
 (4.3.3) and levels  in drinking water and food  are  reported in the human
 exposure  section  (5.2).   Whenever possible, the arsenic chemical forms
 present were  identified;  however, information  of  this detail was
 uncommon.

 4.4.1  Natural  Waters

     For  the nation  as a whole,  concentrations of  total arsenic in
 natural waters  are relatively  low.   Forty-six percent of  the arsenic
 concentrations  recorded in the STORET system do not exceed 1 ug/1- 91%
 are no higher than 10  pg/1.  Table 4-12 presents the distribution'of '
 unremarked  observations from 1975 through 1979 for major river basins
 and the United  States.  Regional differences among the major river
 basins are  slight; the distribution  for each basin is almost identical
 to the national distribution.  The exception is the Great  Basin, for
which 17% of the concentrations are at or below 1 Ug/l, and 64% are no
higher than 10 ug/1.
                                  4-48

-------
           TABLE 4-12.
 Northeast
 North Atlantic
 Southeast
 Tennessee River
 Ohio River
 Lake Erie
 Upper Mississippi
 Lake Michigan
 Missouri River
 Lower Mississippi
 Colorado River
 Western  Gulf
 Pacific  Northwest
 California
 Creat Basin
 Lake Huron
Lake Superior
Hudson Bay

  UNITED STATES
   DISTRIBUTION OF TOTAL
   BASINS ANO THE UNITED

Total No.
of Obser-
J^jy£"jl__ £o7boT
                        1472
                        1325
                        2537
                         879
                        3757
                        1563
                        6490
                         823
                        3637
                        5291
                        2809
                        1126
                       2301
                       1272
                        653
                        154
                        205
                         14
           30
           24
           22
            8
           24
            4
           53
           35
           21
           15
          44
          17
          18
          26
           3
           9
          12
           7
                              27(9800)
^SiHihS^^^
	 	 -^~ J..J.-JU 10.1 -inn inrTi — =T7w^— -r-r- —
0
<1
0
<1
<1
<3-


<1
0
<1
<1
0


0
0
39
35
43
49
22
26
8
25
12
13
9
13
28
10
14
15
47
14
27
34
33
39
50
68
34
31
52
61
33
52
43
55
47
70
40
79
• ~ • "" -»-"u. J.-XUUU > 1 ()()0
3 i
J I   ^QQ

-------
                TABLE 4-13.  TOTAL ARSENIC CONCENTRATIONS
                             DETECTED  IN FRESHWATER
        Sample

From 150 U.S. rivers
     Arsenic  Level

 detectable in  7% of
 1,500  samples
      Reference

Kopp  and Knower
(1970)
From U.S. rivers and lakes     detectable in 21% of     Krum _e_t  al.   (1971)
                               727 samples
Lakes in Wisconsin
2 to 56 yg/1
                                                        Chamberlain and
                                                        Shapiro  (1969)
U.S. rivers and lakes
10 to 1100 yg/1
                                                        Durum et al.   (1971)
Columbia River
1.6 yg/1 - average       Wedepohl  (1969)
River in industrial area
 (Taiwan)
0.21 mg As/1
Tseng ££ al.  (1968)
Kansas River
2 to 8 yg/1
Angino et al. (1970)
Spring waters, high in
bicarbonate (worldwide)
400 to 1,300 yg/1
Schraeder and
Balassa (1966)
                                   4-50

-------
                  TABLE 4-14.   TOTAL ARSENIC CONCENTRATIONS
                               DETECTED IN SALTWATER
          Sample

 Pacific  coastal water
  Arsenic Level

3 to 6 ug/i
   Reference

Wedepohl (1969)
 Northwest Pacific
                                0.15 to 2.5 yg/l
                                1.2 yg/l - average
                         Wedepohol (1969)
 Indian Ocean
                                1.3  to  2.3  y
                                1.6  yg/l -  average
                        Wedepohl (1969)
Southwest Indian Ocean
                               1.4 to 5.0 ug/1
                               3.0 yg/l - average
                        Wedepohl  (1969)
                                4-51

-------
      Other miscellaneous  observations  of  total arsenic levels in fresh-
 water systems  are  reported  in  Table  4-13.   Most values are compatible
 with those in  the  STORE!  data  base.  Exceptions include industrial areas
 and  saline water bodies.

      In addition to  this  information,  STORET  retrievals were made for
 the  minor river basins  comprising  the  Pacific Northwest and Missouri
 River major river  basins  for the purpose  of examining  concentrations in
 the  vicinity of known arsenic  sources  and to  determine the origin of unusually
 high arsenic levels.  The results  are  discussed in  the aquatic  exposure
 section (6.2).

      According to  STORET  data,  total arsenic  concentrations measured
 from ocean sampling  in  the Northeast,  the Western Gulf,  and California
 range from 0.7 ug/1  to  220 yg/1.   Sampling  of arsenic  in estuaries in
 eight regions revealed  concentrations  ranging up to  1,700 pg/1;  the
 average among the  regions is 3.15  ug/1.   Some miscellaneous observations
 in saltwater are also reported  in  Table 4-14.

      Sampling of arsenic  in USGS monitoring wells has  been  limited to
 specific regions (U.S.  EPA 1980c).  Major river basins  for  which  enough
 observations have  been  recorded to indicate well water  quality are the
 Southeast,  Ohio River,  Lake Erie, Pacific Northwest, Tennessee River,
 and  Missouri River.  For  the five-year period 1974-1979,  maximum,  minimum,
 and  mean concentrations for these six basins  are shown  in Table 4-15.

       TABLE  4-15..   CONCENTRATION OF ARSENIC IN WELLWATER  IN  SIX
                   MAJOR  U.S. RIVER BASINS—STORET, 1975-1979


                            No.           Concentrations    (ug/1)
 River  Basin             Observations    Maximum    Minimum    Mean

 Southeast                  104a            63        10       21.5

 Tennessee River            125a           100        10       15.3

 Ohio River                 917      -      390         0        8.6

 Lake Erie                  353            500         0       11.1

Missouri River             330a            10         5        8.3

Pacific Northwest            93             14         o        1.9
a
 Remarked data.
Source:  U.S. EPA (1980c).
                                   4-51

-------
    8ro       r^'^ef t3;V!0D!8?« at Chi- ^e and Boron, potable

    of 8 Ug/1  to 33 Ug^^tSc^^SS^^*1^ CTentrati°-
    nxque used was single-sweep polagraphy conf i™ii \     analnical tech-
    which is superior  to the previous lv ?L   - f    , by X"ray flu°rescence,
    diethyldithiocarbamate (S                     °yed  Colori
    eve,  distinguish  between


   4'4-2  .Effluent
                                                        water nation-
                                                    5.year period ta
 =^^<~^ LV— s

                                      zz -
Pacific No'rthwes??  The L^ua 1°"" MississiPPi and
^ these six areas eS^r?/"88 ^""ratio
                         e/



sy-=="-

                             4-53

-------
         TABLE  4-16.   ARSENIC  CONCENTRATIONS  IN EFFLUENT WATERS --
                      UNREMARKED  DATA IN STORET (1975-1979)
Major  River Basin
Northeast
North  Atlantic
Southeast
Tennessee River
Ohio River
Lake Erie
Upper  Mississippi
Lake Michigan
Missouri River
Lower  Mississippi
Colorado River
Western Gulf
Pacific Northwest
California
Lake Huron
Hudson Bay
     No
Observations
     23
      4
    282
      5
     72
      4
     30
      6
      5
     20
      5
      5
     94
      1
      1
      2
                                                      Arsenic
                                                Concentration (ug/1)
Rang
3 -
10 -
0.1 -
2.8 -
0 -
1 -
1 -
1 -
0.5 -
3.6 -
1 -
11 -
0.4 -
6.0
1.0
1 -
zf.
100
56
217
100
700
10
2.900
2
70
22
34
20
1203


1
Mean
15.7
28.5
41.8
30.3
23.6
5.8
517.9
1.2
14.5
7.8
8.6
15.8
18.5


1.0
  TOTAL FOR REPORTED BASINS   559
                    0 -  2900
                                                                 56.8
Source:  STORET Water Quality Control Information System, as of October 27
         1980.
                                  4-54

-------
                                            °f
it was not detected  in
ranged from 1.5 yg/i  to jj UB/1 anA •
2.40 ms/ks   TaKi = /  Tf    ^S'-1- ana in sediment samples
"u^^'iirs-s: s:ssrf"» -^'
    TABLE 4-17.
  Location



  Lake Michigan

  Lake Monona

  Lake  Washington

  Lake Erie
                                                         -ter samples
                                                                 to
                                                                samples
                   Concentration of Arsenic
(Deep)
Layer
11
2
10
0.6
(Upper)
Laver
22
51
200
3.2
                                              Reference



                                              Ruch et al.  (1970)

                                              Syers et_ al.  (1973)

                                              Barnes _et al. (1973)

                                             Walters e_t al. (1974)
                                                 conce"«ations of
                                                           a
4'4'5
               Plants
                                     Mt                .      es  -
-^atural background levels of arsenic "sol? ln°rganic ffline"l  material.
(Sandberg and Allen 1975).  In a survev  °f L*™ USUally less than 10 tng/w
             --^
                              4-55

-------
          TABLE 4-18.  ARSENIC RESIDUES IN WATER AND SEDIMENT
                       AMERICAN FALLS RESERVOIR, IDAHO, 1974
                                Arsenic Concentration
Location

 Area 2


 Area 3


 Area 4


 Area 5
Water
Sediment
_ _ (VR/D Crrm/tc^
Mean
12.60
15.67
16.50
4.67
Range
1.50 - 33.0
3.0 - 25.5
3.5 - 30.5
3.0 - 8.0
Mean
2.02
1.83
1.56
1.38
Ran;
1.57 -
1.38 -
1.40 -
1.36 -
?e
2.40
2.20
1.75
2.05
 Source:   Kent  and  Johnson (1979).
                                     4-56

-------
TABLE 4-19.

                         _Concentratinr.
Major River Saa-f-

Northeast
North Atlantic
Southeast

Tennessee River
Ohio River
Lake Erie
Upper Mississippi
Lake Michigan
Missouri River
Lower Mississippi
Colorado River

Western Gulf

Pacific Northwest
California
Great Basin
Lake Huron
Lake Superior
Hudson Bay
UNITED STATES
	 Dissolved 	 1 	 -~ 	
& M^^T^^^ 0^^^2^^~-rr~
493 8.0
249 323.0
1654 2.2.. Q

235 6.0
854 53.0
124 21.0
595 17.8
213 4.0
1608 619.9
1846 15Q.O
1 *^f}A co f\
•+--~}\j\j jo 0
,
Gy9 59.0*
1238 2600.0
627 35.0
844 330.0
151 8.0
130 15.0
41 55.0
12901 2600.0
0.7 338
5.2 123

1.2 940
1.0 61
0-9 354
1.6 78
3.0 343
1.0 117
5.9 783
1-9 1386

3-1 517

3.2 246
6-9 696
1-9 387
12.5 248
0.9 46
0.9 70
4.1
2-6 6733
fc>j *• -» A -L ui vim nspTi
11.0 0.4
29.0 1.8

17.0 0.6
3.0 o 4
V • *T
17.0 0 9
\J • >
5-0 0.7
31-0 i.o
7.0 0 4
' J • -T
22000.0 46.0
150.0 2 1
*• • i
1300.0 13 4
-t j • ^
46.0 2 0
** *- * \J
100.0 0.6
220.0 1 4
-JU # *+
42.0 i Q
v j. • y
4.0 0 S
^* vy • O
2.0 0.4

22000.0 7.4
          Quality
                        4-57

-------
                 TABLE 4-20.   CONCENTRATIONS OF ARSENIC
                              DETECTED IN SOILS, ROCKS
                              AND ORES
 Sample

 Crystalline rock

 Soils



 Shale


 Igneous  rock
Shale  and deep
sea sediments

Sandstone and
limestone
Arsenic Concentrations

2 mg/kg (average)

unpolluted:  1-40 mg/kg
5 mg/kg (average)
near smelter:  150 mg/kg

13 mg/kg (average)


2 mg/kg


10 mg/kg


1.5 mg/kg
Reference

Bowen  (1966)

Schroeder  and
Balassa  (1966)
Miesch and Huffman  (1972)

Schroeder  and
Balassa  (1966)

Onishi and Sandell
(1955)

Onishi and Sandell
(1955)

Onishi and Sandell
(1955)
               TABLE 4-21.  CONCENTRATIONS OF ARSENIC
                            DETECTED IN COAL

Coal
Illinois3
Appalachian
Western3
Eastern
Combined
(82% IL)
Maximum =
No.
Samples
113
23
29
617
101

950 mg/kg;

Range of Cone, (ma/ks)
1-120
2-100
0.3-9.8
0.12-950
0.5-93

weighted mean = 17 mg/kg
Arithmetic Mean
(mg/kg)
14
25
2.3
18
14


*Gluskoter ^t al.  (1977).
^Zuboric _ejt al.  (1979).
'Ruch et al.  (1974) .
                                 4-58

-------
    (PAX 1973).   Arsenical
    i948 with the jvSt              ac
    concentrations were  published  in 1971
    of  arsenic in soil.
                  maximu° of 2,500 mg/k
                             mentioned
                   significant persistence
                                           ^     ^ »*• «*
  England.   Plant and*           ewe      efrom
  account  for roughly 97%  of  the a™<             tW° major areas that
  contaminated si Ls  were  iL'etv barr    °U?Ut  in the 3rea'   Heavily
  of Plant  species.   SurvivS plant  soecS, JUpP?rted 3 restri"ed  number
  concentrations  of arsenic in the soif  a H deVel°Ped a tolerance to high
  levels as  shown in  Table 4-2?   Lvelhr-?0^1^ arS6nic  to
  of magnitude) than  plants gr^S'^SSSSS  arfas"


                     in' SJ'U? 'T ^^^ ^  the
 states of arsenic in several w
 major form characterized, with
 The Water-S0luble extract ^aSSu
 fraction  available  for plant uptake
 acid  extract, on the other hand  is
 of arsenic present  in the soil   bou      n
 of arsenic were  also analyzed in
 fro,  the surface layer  tc/2 J «
 detected at  depths greater thai  15
 the element  in both  arsenlte
 detected in  the upper 2
activity there and the
               ,
             ext"cts,  arsenate  was  the
                                               eUt
                                          '
                     rePresentative of  the
                     the S0i1'  The nitric
                     °f ^ t0tal amount
                     ^ °*±d^°» states
                    " depth? "nging
                    hl§hest levels were
                  vertic*1 novement of
                  ^"hylarsine was  only
                          °f
                            on orchard
             applications of arsenic
                                   "
      Kenyon e_t al.  (1979)
 land near Interlaken, NY,
 compounds for many years,
 centrations of arsenic resist f?rpri -?r>  ^-t.i            	-"> "*•  i^un-
 Plants seeded, tomato 0.2 mg/kg dry wSh^ P0"*™3 °f °ne of th* five
 lands.   Only arsenic concentrations In the So'?°   ?'*"** Md ^"hard
 between the two locales:   1.9 mg/kj dry weight   SXhlbited the differences
 mg/kg in orchard  soil.             g   y vei§ht m nonorchard soil,  22.4

 4.4.6   Air
                       are             0
smelters  (NASN 1964).  The annual -VP?
Table 4-25 presents ambient arseni
xn the vicinity of specific
were generally two orders of
and the highest level .  (for
     and copper smelters.
                  in  the vicinity  of
                  Sit6S iS °'<™ »*'**•
              con"ntrations measured
            associated with industries
                   aabient
4-59

-------
         TABLE  4-22, CONCENTRATIONS  OF ARSENIC  IN  THE FOLIAGE
                       OF VARIOUS PLANT  SPECIES
Species

Jasione montana

Calluna vulgaris

Agrostis tenuis

Agrostis stolonifera

Calluna vulgaris

Agrostis tenuis
(Concentrations (ug/g  dry weight)
 Maximum                   Mean
   6640

   4130

   3470 .

   1350

      0.33

      0.28
2040

1260

1480

 720

   0.3

   0.23
Source:  Porter and Peterson (1977).
                                  4-60

-------
   Arsenic Concentration (yg/g dry weight)
   Soil S-l-.te.ct- »ith
                          Water SoluhTo




                             1.18
                            0.16
                                                % of Water Soluble
                                                          10.2
0.87
1.40
84.9
93.7
4.5
6.3
                                                                      11.8
                                                                     10.6
                                              79.2
                                                         20.8
Source:   Porter and Peterson (1977).
                                 4-61

-------
      TABLE  4-24.   CONCENTRATION AND  SPECIES OF AVAILABLE ARSENIC
                   IN WATER-SOLUBLE AVAILABLE EXTRACTS OF SOIL
                   SAMPLES FROM SITES CONTAMINATED WITH MINE AND
                   SMELTER WASTE
                                                       As(CH )2


                                                         0.026

                                                      Not Detected

                                                      Not Detected

                                                      Not Detected

                                                      Not Detected

                                                      Not Detected

                                                      Not Detected

                                                      Not Detected

                                                      Not Detected
Sample
Depth (cm)
0-2
2-4
4-6
6-8
8-10
10 - 15
15 - 20
20 - 25
25 - 28
Concentrat ions
As (III)
0.195
0.185
0.218
0.235
0.178
0.654
0.720
0.215
0.190
of Ars<
As (V)
0.110
0.290
0.207
0.655
0.340
0.846
74.40
21.88
28.75
Source:  Porter and Peterson (1977).
                                 4-62

-------
       TABLE 4-25.   ARSENIC CONCENTRATIONS DETECTED IN THE ATMOSPHERE
        Sample Site

  National urban ambient
  levels
  Urban areas containing
  smelters
                                   Concentration (y
    Ranee
    .a
0.001  - 0.083
Average

 0.003



 0.03
                                  Reference
                                 Suta  (1980)
                                 Suta (1980)
  Remote areas
  Coal-fired power plants


  Copper smeltersb


 Lead smelters
 Zinc smelters
 Cotton gins
Pesticide manufacturers13



Glass manufacturers*5
                                                    0.0004      Suta  (1980)
                    0.003
                   <0.99
                                                   <0.29
                                                  <0.29
                                              usually _<0.59
                                              but as high
                                              as 5.9 for
                                              very small
                                              subpopulation
            Suta (1980)


            Suta (1980)



            Suta (1980)


            Suta  (1980)


           Suta  (1980)
                  £0.0009      Suta (1980)
                                                 <0.29
                                                              Suta (1980)
  etection  limit  0.001  ug/m3.

                                                               «*>
                                 4-63

-------
 sources, the trivalent form is prevalent (see Chapter 4.3, Pathway 1).
 However, sampling indicates the presence of both the trivalent and
 pentavalent forms in other areas (Mushak et al. 1980).

 4.5   OVERVIEW             ~

      The  purpose  of  this  section  is  to  attempt  an  integrated  overview
 of the  significance  of  arsenic  releases  to  and  intermedia  transfers
 between air, soil and  surface water  within  the  perspective of their
 potential contribution  to water-borne routes  of exposure.   The most
 important pathways are  emphasized based  upon  the magnitude of pol-
 lutant  levels involved, the presence of  toxicologically  important
 chemical  forms of arsenic, and  the significance of  the receiving med-
 ium as  it relates to exposure of receptor populations.   The following
 discussion  is based  on  information presented  in Chapters 3.0  and 4.0.

 4.5.1   Important  Fate Processes

     Arsenic is subject to numerous chemical  and biological trans-
 formations  in surface waters and soil.  Table 4-26  summarizes  the  most
 important of these processes.  As previously noted, atmospheric trans-
 formations  of arsenic are not considered but only transfers from the
 atmosphere  to soil or water.

 4.5.2  Transfer from Air to Surface Water and Soil

     The majority of the arsenic initially released to the  atmosphere
 is expected to be transferred eventually through rainout or fallout to
 water'or land.   The extent of dispersion depends upon the  size of  the
 particulate in the emission and the prevailing climatic  conditions.
 Larger diameter particulates (>20 um) tend to be deposited within  1 km
 of the point source.   For smaller sized particulates., a mean residence
 time of 9 days was estimated, and this suggests a potential for trans-
 port of the adsorbed arsenic over a considerable distance before set-
 tling out.  This  is especially important for arsenic released  from
 coal combustion processes because of the affinity of the element in
 its vapor form for smaller diameter particulates.  Arsenic released
 from cotton gins  is primarily associated with large diameter particu-
 lates, which would most likely be deposited near to the  source.  The
 relationship between arsenic levels and particulate size emitted by
 the remaining significant sources of atmospheric releases,  such as smelt-
ing operations,  is not known,  however,  the range in particulate size is
 expected to be great.  Monitoring data indicate high levels of arsenic
 accumulation in the surface layer of soil immediately surrounding
 smelters which may be due to the contribution of larger-diameter parti-
culate fallout.

     Atmospheric emissions of arsenic may,  therefore,  have a signi-
 ficant impact on residue accumulation in local terrestrial  and aquatic
 environments.   An accumulation rate of approximately 3 mg of arsenic
 per m2 of soil  per year was estimated in the vicinity of a coal-fired
 power plant and the rate likely to be higher in the vicinity of other

-------
                                                     "—-
RO-, ,.- Relative
"nd Chan'e Rate ot
-' y«£Sl Chan£e_ Product^
formation "IB'' Insoluble arsenic
salts

2- Adsorption „.- , .
81 So"- or sediment-
arsenic complex
V Ion exchange High
fc Soil- or sediment-
arsenic complex
«. nemetbylation |ou
Inorganic o-arsenic
flc itj
5. Reduction InM
_£, Low Arsinea
1
i
cr>
U)
6. Oxidation'1 u j
""-rate .'entavalent arsenicals
'. Methylation j _u
Low Metltylarsines


Biologic
Activity
Insoluble
and inactive

Fixed and
inactive
Fixed and
inactive
Reacts as 1,
2, and 3
Very active

Reacts as 1,
2, and 3
Very active

Probability of
Occurrence
Very high

High
High


.


High
Low

Conditions for
Occurrence
Presence of iron> aillnlinulll
calcium, and magnesium in '
soil or sediment.

Fine soil and sediment
colloidal and organic
matters.
Soil or sediment with high
exchange capacity.

Microorganisms for de-
methylation (aerobic)

Aerobic and anaerobic
conditions or specific
microorganisms.
Aerobic soil or surface
water conditions.
P

Further Possible
Changes
Formation of an
arsenic analogue of
fluoroapatite, an
extremely insoluble
complex mineral
Formation of sediment
in aquatic systems.
Exchange release- by
other salts, to react
further as 1 or 2
Same as I, 2. and 3.


React rapidly u.
form pentavalem
arsenicals; then
same as 1, 2, and 3.
Same as 1, 2, a.ul 3.

                                                                                                                      c
                                                                                                    Bacterial microroganisms
                                                                                                    (anaerobic and aerobic)
React rapidly to
form peniavaleni
arsenicals; then
same as  1. 2, and 3
"Adapted  from NAS (1977).

bKelers to oxidation of a
                        "enic from trivalent to pentavalent.

-------
 sources.  Soil levels near atmospheric sources are on the order of  two
 to three orders of magnitude greater than background levels.  Approxi-
 mately 10% of the annual arsenic loading to the Puget Sound was at-
 tributed to atmospheric fallout.

      The chemical form of arsenic initially released in combustion-
 associated atmospheric emissions is usually arsenic trioxide.  The
 trioxide readily is converted to arsenite on contact with water, how-
 ever,  so this is the form usually entering soil and water.  Some ar-
 senic  is also released from mining and milling in the same form as it
 is found in ores.   Cotton ginning processes presumably release arsenic
 in the form of arsenic acid.

 4.5.3   Releases to Land and Transfer to Ground Water

     Releases of arsenic in solid waste disposal  to land account for
 the largest fraction of total environmental releases,  approximately 65%
 excluding pesticide application to land.   Due  to  long-term persistence
 of arsenic  wastes  in soil and the potential for mobilization of arsenic
 under  certain environmental conditions,  these  releases  can sometimes
 result in significant exposure problems.

     From most sources,  the predominant form released  is expected  to be
 arsenite.   The initial form is most  important  in  the  period of time
 before an equilibrium is  established,  according to  the  conditions  of
 the^receiving medium.  After  equilibrium  is  achieved,  the  prevailing
 environmental conditions  have  a significant  influence on arsenic's
 specification and,  therefore,  on its potential  for  mobility,  retention,
 biotransformation  and volatilization.   The  importance of the  initial
 form appears  to be  somewhat overshadowed  by  the site conditions  follow-
 ing establishment  of  equilibrium.

     The method of  application  or disposal onto land is  an important
 factor  in the  potential  for migration of  arsenic  into groundwater.
 Arsenic  in  slag or  sludge  from tailing's  ponds  is often disposed of at
 managed disposal sites.  The percentage of disposal sites  lacking the
 proper precautions  to  prevent  runoff or leaching  and migration of toxic
 substances  (such as clay or polyethylene  liners,  tiling, etc.) must  be
 known  in order to assess the significance of these wastes  as a source
 of  arsenic pollution  on a national scale.  There may also  be a potential
 for leaching  of arsenic from slag used as road  construction material.
 Fly ash is usually disposed of,  following settling, in ponds which are
 eventually filled solid.  Protective lining of  these ponds is a rela-
 tively recent phenomenon.  The arsenic concentrations in fly ash are
 sometimes high, and the surface area to volume  ratio greater than in
mining and smelting wastes; therefore the potential for migration from
 fly ash ponds may be considerable.  Municipal or industrial landfills
 are often reported to contribute to elevated concentrations of arsenic
 in ground water.
                                  4-66

-------
or the deeper layer of       as
will be mobile include soils low
                                    arene
                                            '
                                                    C°nditio-  (saturated
                                                    P"  "  Whi°h
 include all ofae excepTth "       ^^ wil1
 lease bound arsenic from ^ £i 1^""°" ^ 1
-------
bound by clay, iron and aluminum oxides and sulfides.  This substantial-
ly reduces concentrations in the water column.  Organic sediment mater-
ial has not been identified as a major sink for arsenic.  Within the
sediments, arsenic may be solubilized by methylating bacteria, reduced
to the gas, arsine, and/or desorbed from reduced iron oxides as As(lll).
Elements such as Be, Cu, and Cr may also act to control dissolved con-
centrations of arsenic in the water column.
                                 4-68

-------
                                  REFERENCES
   Acurex Corp.
   VA:  AcureJ Cop
                            \

      1974.
                                            materials balance.  Arlington,
                                                       pollutants.

                                                                cited in
                                             Map  (USA)  (XMFSDP) .  NO. MF-
                       ^249 j

Braman, R.S.; Johnson, D.L.; Foreback  r r
Separation and determlnaHn^  "reDacK» C.C
                                            . A
                                            ; Ammons, J.
                                                             the  environ-
                                                                    J T
Buchman, H.O.; Bradv  NT
                                                Pro2ress.  35:396-416:
                                4-69

-------
 Chamberlain,  W.;  Shapiro, J.   On the biological significance of phosphate
 analysis:   Comparison of standard and new methods with a bioassay.  Limn.
 Oceangr.  14:921;  1969.   (As cited in Phillips and Russo).

 Chapman,  A.C.   On the presence of compounds of arsenic in marine
 crustaceans and  shellfish.   Analyst.  51:548;  1926.

 Cherry, D.S.;  Guthrie,  R.K.   Toxic metals in surface waters from coal
 ash.   Water Resource Bull.   13(6):1227-1236;  1977.

 Coles, D.G.;  Ragaini,  R.C.;  Ondov,  J.M.;  Fisher,  G.L.;  Silberman,  D.;
 Prentice,  B.A.  Chemical studies  of stack fly ash from a coal-fired
 power plant.   Environ.  Sci.  Technol.  13(4) :-455-459;  1979.

 Comanor, P.L.; Deguire,  M.F.;  Hendrix, J.L.;  Vreeland,  H.;  Vreeland,  P.
 Effects of previous  tailing  disposal  on Upper Sonaran  vegetation of the
 Getchell area.  Trans.  Soc. Min.  Eng. AIME.   256(3):240-242;  1974.

 Cooper, H.; Paden, W.R.;  Hall,  E.E.;  Albert,  W.R.; Rogers,  W.B.; Riley,
 J.A.   Soils differ markedly  in their  response to  addictions of  calcium
 arsenate.   South  Carolina Agric.  Exp. Sta.  Ann. Rep. 45:23-27;  1932.
 (As cited  in NAS  1977).

 Cox,  H.E.   On  certain new methods  for the determination of  small quanti-
 ties  of arsenic and  its  occurrence  in urine and in fish.  Analyst.
 50:3-13; 1925.  (As  cited in NAS  1977).

 Cox,  D.P.; Alexander, M.  Production of trimethylarsine gas from various
 arsenic compounds by three sewage fungi.  Bull. Environ. Contam. Toxicol.
 9:84-88; 1973.  (As  cited by Versar 1979).

 Crecelius, E.A.« The geochemical cycle of arsenic in Lake Washington
 and its relation to  other elements.  Liminol. Oceanogr.  20(3):441-451;
 1975.  (As cited in Versar 1979).

 Crecelius, E.A.; Bothner, M.H.; Carpenter, R.  Geochemistries of arsenic,
 antimony,  mercury, and related elements in sediments of Puget Sound.
 Environ.  Sci. Technol. 9(4):325-333; 1975.

 Crecelius, E.A.  Arsenite and arsenate levels in wine.  Bull. Environ.
 Contamin.  Toxicol.  18(2):227-230; 1977.

 Dickens,  R.; Hiltbold, A.E.  Weed Sci.  15(4):299-304; 1967.  (As cited
 in Sandberg and Allen 1975).

Dratshev.   (As cited in PAX 1973).

Dressen,  D.R.; Gladney, E.S.; Owens, J.M.; Perkins, B.L.; Wienke, C.L.;
Wangen, L.E.  Comparison of levels of trace elements extracted for fly
 ash and levels found in effluent waters from a coal fired power plant.
                                 4-70

-------
    Duble, R.L.; Holt, E.C.; McBee  r r
    disodium methanearsonate (DSHA) in
                17:1247-1250; <      "
                                        T

                                                 tiOn
                                                          br8£*do»n of
                                                            J'A8rlc-
                                                                   surface
           Water
   (As  cited in Ferguson'and  Gavis
                                                            i "68.
Fisher, G.; Silberman, D.: Prentice  * .  a t

"?"« »*"> -««n-acti4atedc0a'flVash  'E  !       '   '           °
689-693; 1979.                         7 ash-  Environ.  Sci. Tech.  13(6)
      »IIS, Springfield, VA  PB 266
                                                                "
                                  905
   .      p.   Avallable from

                                                  landfill=:  F
Geraghty and Miller
hazardous cheMcal
sites   Washington, DC:  EPA 68-ow?
NTIS, Springfield,  VA, PB 275 103?
                                            ^surface Blgratlon of

                                                 1 VaSt6 land  dis
                                                 P<   A^ilable from
                              4-71

-------
 Gilbert, D.  Sources of toxic pollutants found in influents to sewage
 treatment plants.  I.  Literature Review.  EPA 68-01-3857.  Washington
 DC:  Office of Water Planning and Standards, U.S. Environmental Protec-
 tion Agency; 1979.

 Gilderhus, P.A.  Some effects of sublethal concentrations of sodium
 arsenite on bluegills and the aquatic environment.  Trans. Am. Fish Soc.
 95:289-296; 1966.  (As cited in Phillips and Russo 1978).

 Gluskoter, .H.J. et_ _al.  Trace Elements in Coal:   Occurrence and Distri-
 bution,  Illinois State Geological Survey, Circular 499; 1977.

 Goodroad, L.L.; Caldwell, A.C.  Effects of phosphorous fertilizer and
 lime on the As, Cv, Pb, and V content of soils and plants,  J  Env
 Qual. 8(4):493-496; 1979.

 Greenberg, R.;  Zoller, W.;  Gordon,  G.  Composition and size distributions
 of particles released in refuse incineration.   Environ. Sci. Tech. 12(5) •
 566-573; 1978.

 Griffin, R.A.;  Shimp,  N.F.   Attenuation of pollutant in municipal land-
 fill leachate by clay minerals.   Cincinnati,  OH:   Municipal Env.  Res.
 Lab.; 1978;  146 p.   Available from  NTIS,  Springfield,  VA, PB 287  14o!

 Gulen,  J.; Champ, D.;  Jackson, R.   Influence  of  redox  environments on
 the mobility of arsenic in  ground water.   ACS  Symp.  Series 93:81-95;
 -L .7 / J7 o

Hall, R.A.;  Zook, E.G.; Meabum, G.M.  National Marine Fisheries Serv-fce
Survey of  trace elements in the fishery resource.  NDAA Technical Reoort
NMFS  SSRF-721; 1978.

Hawkins,  D.B.   Environmental  fate of  arsenic in ground water.  Fairbanks
AK:   Institute  of Water Resources,  IWR-77,  1976;  7 p.                    '

Herrino,  W.O.   Status  of  gin  investigation.  Washington,  DC:  U.S.
Environmental Protection Agency; 1972.  (As cited in Suta 1980).

Hiltbold, A.E.  Arsenic content of  cotton soils in Alabama, Research
Report.  Proc.  Southern Weed  Conference; 1968.  p. 188.   (As cited in
PAX  1972).

Hiltbold, A.E.;  Hajek, A.E.;  Buchanan, G.A.  Weed Sci.  22(3) -.272-275 •
1974.  (As cited  in Sandberg  and Allen 1975).

Hiltbold, A.E.  Behavior of organoarsenicals in plants and soils.   ACS
Symp. Series 7:53-69; 1975.

Holm, T.R.; Anderson, M.A.;  Inverson, D.G.; Stanforth,  R. S.  Heter-
ogeneous interactions of arsenic in  aquatic systems.   ACS Symp. Series
"-3:711—736; 1979.

Huang, P.M.; Liaw, W.K.  Distribution and fractionation of arsenic in
selected fresh water lake sediments.  Int.  Revue  ges. Kvdrobiol  63(A).
533-543;  1978.                        .                 '        ' —

-------
 Isensee, A.R.; Kearney, P.C.; Woolson, E.A.; Jones, G.E.; Williams, V.P.
 Distribution of alkyl arsenicals in model ecosystem.  Environ. Sci.
 Technol. 7:841-845; 1973.  (As cited in NAS 1977).

 Johnson, D.L.  Bacterial reduction of arseniate in seawater.  Nature
 240:44-45;  1972.

 Johnson, L.R.; Hiltbold, A.E.  Arsenic content of soil and crops follow-
 ing use of  methane arsonate herbicides.  Soil Sci. Soc. Amer. Proc.
 33:279-282; 1969.   (As cited in PAX 1973).

 Jones,  J.S.;  Hatch, M.B.  Soil Sci.  60:277-288;  1945.   (As cited in
 Sandberg and  Allen 1975).

 Jones,  R.A.;  Lee,  G.F.  Chemical agents of potential health significance
 for land disposal  of municipal wastewater effluents and sludges.   Sagik,
 B.P.;  Sorber,  C.A., eds.  Risk assessment and health effects of land
 application of municipal wastewater and sludges.   San Antonio,  TX:
 The University of  Texas at San Antonio;  1977.   pp  27-59.

 JRB, Inc.  Level II Materials Balance:  Arsenic.   Draft report.  Con-
 tract No. 68-01-5793.  Washington,  DC:  Office of Pesticides and Toxic
 Substances, U.S. Environmental Protection Agency;  1980.

 Kent,  J.C.; Johnson, D.W.   Mercury, arsenic, and  cadmium in fish, water,
 and sediment  of American Falls Reservoir, Idaho;  1974.   Pest  Monit  J
 13(1):35-40;  197.                                      '             '  '

 Kenyon,  D.J.;  Elfring, D.C.;  Pakkala,  I.S.;  Bache, C.A.;  Lisk,  D.J.
 Residues of lead and arsenic  in crops  cultured on  old orchard soils.
 Bull.  Environ.  Contain. Toxicol.  2_2:221-223;  1979.

 Kimmel,  G.E.;  Braids,  O.C.  Leachate plums in  ground  water  from Babylon
 and Islip landfills,  Long  Island, New  York.  Washington,-DC:  U.S.
 Geological  Survey;  1980.

 Kobayashi,  S.;  Lee,  G.F.   Accumulation of arsenic  in  sediments  of Lakes
 treated  with sodium arsenite.  Environ. Sci, Technol. 12_(10) : 1195-2000;
 J- .7 / O •

 Koppe,_P.;  Giebler,  G.   The danger  to public water  supplies  through  the
 pollution of ground water with arsenic.  Water Pollution Abs. 41:87;
 1968.

Lambon,  V.;  Lim, B.  Hazards of arsenic in the environment, with par-
 ticular  reference to the aquatic environment.  Washington, DC:  Federal
Water Quality Administration, U.S. Department of the Interior; 1970.

Landau, E. _et _al.  Selected non-carcinosenic effects of industrial
exposure  to  inorganic arsenic.  EPA 569/6-77-018.   Washington: DC-
U.S. Environmental  Protection Agency;  1977.  (Ascited in U.S. EPA'l980).


                                 4-73

-------
 Lawrence,  J.M.   Recent investigations on the use of sodium arsenite as
 an algacide.   Proceedings Southeastern Association Game and Fish Com-
 missions;  1977.   (As cited in Lamon and Lim 1970).

 LeBlanc, P.J.;  Jackson,  A.L.   Arsenic in marine  fish invertebrates.
 Marine Pollut.  Bull.  4:88-90;  1978.

 Lunde,  G.   Occurrence and transformation of arsenic in  the marine environ-
 ment.   Environ.  Health Persp.  19:47-52;  1977.

 McBride, B.C.; Wolfe,  R.S.  Biosynthesis of dimethylarsine by methano
 bacterium.  Biochemistry 10(23):4312-4317;  1971.

 McBride, B.C.; Merilees,  H.; Cullen,  W.R,;  Pickett,  W.  Anaerobic and
 aerobic alkylation of  arsenic.  American Chemcial  Society  Vol. 82.
 Organometals organometalloids:  occurrence,  fate,  environment; 1978
 pp. 94-115.

 Miesch, A.  T.; Huffman,  C.  Abundance and distribution  of  lead,  zinc,
 cadmium and arsenic  in soils.   Helena Valley, Montana Area Environmental
 Pollution  Study, pp  65-80.  Research  Triangle Park,  NC:  Office  of Air
 Programs,  U.S. Environmental Protection  Agency;  1972.   (As  cited in
 NAS 1977).

 Miller, D.; Braids, 0.;  Walker, W.  The  prevalence  of subsurface migra-
 tion of hazardous chemical  substances  at  selected  industrial waste land
 disposal sites.  Washington, DC:  Office  of  Solid Waste, U.S. Environ-
 mental  Protection Agency; 1977.

 Mushak, P.; Gaike, W.; Hasselblad, V.; Grant, L.  Health assessment
 document for arsenic.  Research Triangle Park, NC:   Environmental
 Criteria and Assessment  Office, U.S. Environmental Protection Aaencv
 1980.   Draft.

 National Academy of Sciences (NAS).  Arsenic.  Washington, DC:   National!
 Academy of  Sciences; 1977.

National Air Sampling Network (NASN).   Air quality data.  Cincinnati, OH:
U.S.  Department of  Health, Education and Welfare; 1964.   (As cited in
Suta 1980).

 Norris, L.S.  USDA Forest Service report.  Corvallis, Oregon; 1974.
 (As cited  in Sandberg  and Allen 1975).

 Onishi, H.; Sandell, E.B.  Geochemistry of arsenic.  Geochin. Cosmochio
 Acta.  7:1-33; 1955.   (As  cited in National Academy of Sciences;  1977).

 PAX Company Arsenic Advisory Committee.  Report of the PAX company
 Advisory Committee to  the Environmental Protection Agency.  Salt Lake
 City:   PAX  Company Arsenic Advisory Committee; 1973.  Available  from
 NTIS PB 265 964.
                                 '•4-1 4

-------
Penrose, W. R •  Blaric  p . o
       ' "••"••>  BxacrC, K. ; Havward  M T   T- -,.  i
sea water sediments, and bio?Tnfar ; conti      arsenic Dispersion in
Board Can.   32:1275-1281;  1975   Us citL TT 8°Urce'  J' Fish' R^-
                        ,  j.7/j.   (.As cited by Penrose 1977).
           y,  D.s.  Environmental  Proectxon A^:
                                                                ^search
                      ,e
    .  Tech. 11(8). -77:781; 1977?
 Rancitelli,  L.\ • Ahfi   v u
 in fossil fuel  consumption
                          "
                          iut
menta from Lake Michigan.  IlliLL
ment Geology Notes No. 37; 1970
Scxences 1970).


                      &.
                                                         «-
                                                  and  arsenic  1B  sedi-
                                                 complex.  Environ.
                                                Poll-tant emissions
                                                               botto,
                                              -n Unconso^ated  sedi

                                               °
                                            National Academy of

                             -75

-------
 Schroeder,  H.A.;  Kanisawa, J.J.;  Frost,  D.V.;  Mitchener,  M.   Germanium,
 tin  and  arsenic  in  rats:  effects on  growth, survival,  pathological
 lesions  and life  span.  J. Nutr.  96:37-45;  1968.   (As cited  in  NAS  1977).

 Senesi,  N.;  Polemic, M.  Content  and  distribution  of arsenic, bismuth,
 lithium,  and selenium  in mineral  and  synthetic fertilizers and  their
 contribution to soil.  Commuiic.  Soil Sci.  Plant Anal.  10(8):1109-1126-
 1979.

 Spehar,  R.L.  et al.  Comparative  toxicity of arsenic compounds  and  their
 accumulation in invertebrates and  fish.  Arch.  Environ. Contam. Toxicol
 9:55; 1980.   (As  cited in USEPA 1980).

 Stack, R.L.; Helgerson, R.N.; Bretz,  R.F.;  Tipton, M.J.; Beissel, D.R.;
 Harksen,  J.C.  Arsenic levels in  the  surface and ground waters  along
 Whitewood Creek,  Belle Fonche River,  and a  portion of the Cheyenne
 River, South Dakota.  Brooking, SD:  Water  Resources Institute; 1978.
 42 p.  Available  from NTIS,  Springfield, VA, PB 289 374.

 Steevens, D.R.; Walsh, L.M. ;  Keeney, D.R.  J. Environ.  Qual. 1(3) -.301-303•
 1972.  (As cited  in Sandberg and Allen 1975).

 Suta, B.E.  Human exposure to atmospheric arsenic.   Washington,  DC:
 Office of Research and Development, U.S., Environmental  Protection Agency;
 1980.  (As reproduced in Mushak,  P. et al.   Health assessment document
 for arsenic.  Research Triangle Park,  NC:  Environmental Criteria and
Assessment Office, U.S. Envrionemntal  Protection Agency; 1980.

 Syers, J.K.  et. al.  Water, Air, Soil Poll.  2:105;1973.   (As cited in
 Forstner  1976).

 Symon, K.  Arsenic in the living environment.  Preprint to the  conference
 on chemical  pollution and human ecology.  Charles University, Prague-
 Oct. 12-17;  1970.                                          "      °  '

 Theis, T.L.; Wirth, J.L.  Sorptive behavior of trace metals on fly ash
 in aquesous  systems.  Environ. Sci. Technol. 11(12):1096-1100; 1977.
 (As cited in Turner et_ al. 1978) .

 Theis, T.L.; Westrick, J.D.;  Hsu,  C.L.;  Marley, J.J.  Field investigation
 of trace metals in ground water from fly ash disposal.   Journal WPCF
 11:2457-2469; 1978.

 Tisdale, S.L.; Nelson, W.L.   Soil  fertility and fertilizers.   2nd ed.
New York, NY:  The Macmillan  Co;  1968.  (As cited in Sandber? and Allen
 1975).

Tefry, J.H.; Presley, B.J.  Heavy  metal  transport from  the Mississippi
River to the Gulf of Mexico.   Windam,  H.L.:   Duce,  R.A.  eds.   Marine
Pollutant Transfer.   Lexington,  MA: Lexington  Books;  1976:39-76.

-------
   Environ^! Arsenlc,  Fort  ^^c. on
                           ^^

                                       o( David  Redford; 1979.

                                               •     ».  wash,gton,
                              .                                   -
 mental  Protection Agc?       t>nln8 and Standards,  U.S.  Environ-
                                                    and ..lublllty of

 139-145; 1978.            reswater environment.  Water Research 12°





                                  in the «-ironme»t Met.  Environ.



Walsh, L.M.;  Sumner,  M.E.; Keenv  D  B
                         -    nD-R.
        in
                                                                , and
                                4-77

-------
 Walters, L.J. Jr.;  Wolery, T.J.;  Myser, R.D.   Occurrence of As, Cd, Co,
 Cr, Cu, Fe, Ni,  Sb, and Zn in Lake Erie sediments.   Conf. Great Lakes
 Res., Proc. 17(1):219-234; 1974.   (As cited in Forstner 1976).

 Waslenshuk, D.G.   The budget and  geochemistry of arsenic in a continental
 shelf.  Marine Chemistry 7(l):39-52;  1978.

 Waslenchuk, D.G.;   Windom, H.L.   Factors contolling one estuarine.
 Chemistry of arsenic.   Estuarine  and  Coastal  Marine Science 7(5)-455-464-
 1978.                                                                   '

 Wedepohl, K.H.  ed.   Handbook of geochemistry.   Berlin:   Springer-Verlag;  1974,

 Week, J.D.   Physical-chemical methods of heavy metals  removal.  273-276.
 Undated.

 Whitnack, G.C.; Martens,  B.H.  Arsenic  in potable desert ground water:
 an analysis problem.   Science 171(1969).  383-385;  1971.

 Wickstrom,  G.  Arsenic in the ecosystem of  man.  Work Environmental
 Health 9, 2-8, 1972.

 Williams, K.T., Whetstone,  R.R.  Arsenic distribution in soils and  its
 presence in certain plants.   Technical  Bulletin  732.  Washington, DC:
 U.S.  Department of  Agriculture; 1949.   (As  cited in PAX  1973),

 Wood,  J.M.   Metabolic  cycles  for toxic  elements  in aqueous  systems.
 Rev.  Intern.  Oceanogr. Med.  31/32:7-16;  1973.

Woolson,  E.A.; Axley, J.H.; Kearney, P.C.  Correlation between  available
 soil  arsenic, estimated by six methods, and response of corn (Zea mais L )
 Soil  Sci. Soc.:  Amer. Proc. 35:101-105; 1971a.  (As cited in Woolson et
 al. 1977).                                                            —

Woolson,  E.A.; Axley, J.H.; Kearney, P.C.  The chenistry and phytotoxicity
 of arsenic  in soils.  I.  Contaminated field soils.   Soil Sci. Soc  Amer
 35:938-943; 1971b.

Woolson, E.A.  Fate of arsenicals  in different environmental substrates.
Environ. Health Perspect. 1.9:73-81; 1977.

Woolson, E.A.; Kearney, P.C.  Persistence and reactions of 1JtC-cacodylic
acid in soils.  Environ. Sci. Tech. 7:47-50; 1973.   (As cited in NAS
1977.

Zuboric, P. et al.,  Chemical Analysis of 617 Coal Samples from Eastern
United States,  U.S. Department of Interior, Geological Survey, open file
report, No, 79-665, 1979.
                                 4-78

-------
                     5.0  EFFECTS ANn EXPOSURE-HITMAN


   5.1  HUMAN TOXICITY



   5.1.1  Introduction

  5.1.2  Metabolism and_



  5.1.2.1  Absorption
  arsenic  cJ«on' -»e^*»  ««»  excretion of

  the  compound  (i.e.  orISc vs' ?epending °n the  Particular fonn of

  etc.), it's
       ,             oiplt
 administration.  Thus, difficulties are n^             rOUte °f

 delineating the exact uptake ~nd ^           encountered in clearly

 Pounds by Jarious                                 ""   a
                                oasA ls)
 absorbed into the human body primarilv vL         ^  Arsenic is

 and, to a lesser extent, vil dSal cLtact  '"      " inhalation>
      Oral
                                         and  mucous
 Soluble  salts  of  arsenic  aJeabrif/^""3 °f  the
 parenteral  sites  of  administratin     1*°* 3l1 mUC°US me

 (Klaasen  1980, Har^eJ 1975)   Fin^^  T! rSadily than the

                                                           »-
absorbed from the digestve tct for th trlValent forms a«  not  readily

                                 5-1

-------
      Coulson_et_al.  (1935),  Ray-Bettley and O'Shea (1975), Crecelius
 (1977a,^b)  and Charbonneau et al.  (1978a)  report that in humans, greater
 than 95% of an ingested dose of soluble trivalent arsenic and greater
 than 80% of an ingested dose of arsenic compounds in certain seafoods
 are absorbed from the gastrointestinal tract,  with less than 5% of the
 dose being  found  in  the feces.   Mappes (1977)  reports daily urinary
 excretion of 69-72%  of a daily oral intake by  a human subject of
 approximately  0.8 tag of trivalent  arsenic.  Crecelius (1977a) reports
 excretion within  61  hours of 80% of a total dose of 63 yg of penta-
 valent  inorganic  arsenic following ingestion of the dose in a wine
 sample.   However,  insoluble  arsenic triselenide was found to pass
 through  the gastrointestinal tract with very little absorption (Maooes
 1977).                                                             FF

      Studies with laboratory animals have  resulted in findings similar
 to  those in the human studies  as concerns  absorption  of arsenic com-
 pounds.   In rats,  88% absorption from the  gastrointestinal tract was
 noted for arsenic  trioxide solution and 70-90%  for arsenate solution
 (Coulson _et ^l. 1935,  Urakabo .et al.  1975,  Dutkiewicz  1977);  90% and
 98%  of an arsenic  trioxide solution was absorbed in pigs (Munroe et  al.
 1974) and monkeys  (Charbonneau _et _al.  1978a),  respectively.   When
 Atlantic grey  sole containing arsenic  (1 mg  fish/kg b.w.)  was fed to
 adult female monkeys,  pigs and  adolescent  monkeys,  absorption from the
 gastrointestinal  tract was 90%,  70%  and 50%, respectively  (Charbonneau
 et al. 1978a).  Only  40%  and 30% absorption was  observed following oral
 administration  of arsenic trioxide  suspension  to  rabbits and  rats,
 respectively (Ariyoshi and Ikeda 1974).

     Dermal

     Trivalent  arsenicals are generally more readily absorbed* through
 the  skin  as  compared with pentavalent  forms  (Harvey 1975).  Dermal
 absorption  of inorganic arsenic  is  increased when  applied  as  a  lipid-
 soluble  ointment  (Dustin 1933).  Similarly, Patty  (1948) reports
 increased dermal absorption of arsenic when epidermal lesions are
present  compared with absorption from normal skin.  Dutkiewicz  (1977)
reports  an  uptake rate of arsenate via  the rat tail as high as  33.1  yg/
cm /h with  exposure to concentrations up to 0.2 M  (-15 g/1);  extrapola-
 tion of  these results suggests a rate of uptake of arsenate through
human hands  of  23.2 mg/hour.

5.1.2.2  Distribution and Bioaccumulation

     Within 24 hours  after oral or parenteral administration  of inorganic
arsenite to humans, monkeys,  rabbits, dogs and  guinea pigs, most of  the
absorbed dose is found cleared from the blood and distributed to solid
tissue,  with highest  accumulations  in the liver, kidney, spleen, lungs
and walls of the gastrointestinal tract.  Smaller levels are  found in
the brain, heart,  uterus, muscle or remaining in the blood.  Arsenic in
the blood of these species is bound in nearly equal amounts to the blood
                                  5-2

-------
                                     b              2
    due  to reaction of     ""1" with fS   ^'5' 8kln- halr and M».
    arsenic My regain fiLd  "thele site" for",71 8n°"PS *S In kera""'
    term retention  of  the  compound™,, ""o fccu- in^h D*posl"°" a"d 1™8-
    leveis of arsenic  are  found  in  the sviLl  f^' -A   ff.     '   H° nota"e
    Hunter 1942, Harvey 1975)°          P      fluld  of humans (Klaasen 1980,

   arsenic  in  fetuses  at  least ^7^! f   u   }     nd measurable levels of
   increasing  to  the seventh month   \l Ou5!h.Bont? of gestation and
   the brain tissue? as well as  tt  1     "^"f  leVelS  WSre  found  in
   suggest vulnerability  of the  h^an  feLt   ***' ^^  ^  C°uld
   exposure, since the Ltal J^SS   SLaS^blo'^ ^ arSSniC
                              of information
                              1980).
                                                    ° the P-tern of
  *etal to solid tissue   Ts observed inl aubae*™* Distribution of the
  studied.   F0llowing  sibSt^rJa^ct^f^TH,0?" ^^^
  arsenite to  rats,  a  uniquely hi»h  Sn   ^   *  u     °labelled  Potassium
  the blood  (95-99%  in the  red blood can   °K   !  label  iS  retained  in
  globin).   These concentrations In  the bin  b°m*t0 the  glo^n  of hemo-
  fold  increase  over those  that woulf J       ^dxcate a  two-  to  eight-
  throughout the body, Sth concrntrationrr'^?-:1^ UnlfOrm distribution
  than expected  (Hunter ^ ^19?^?^ hi?      tlSSUe being much lower
 both pentavalent and tSvJleni :±nor*J   half~t1^ of blood clearance  of
 -ately 60-90 days  (Lanz £ ^ 1950  A^o^^1^1?/^ rSt ±S a^^
 rate of clearance  is notably higher 'thin tSh^f    ^ 19?4) '   This
 hours  of the three-compartment  model of   ?   half-tlmes of 1, 5 and 35
 as  proposed by  Charbonneau et  al   (1978V    f-? ^ hUmanS  and do§s'
 Frederickson  (1963)  calcula^dTalf Lf     .Slmilarly,  Overby and
 the rapid phase, slightly  longer  ^ JhT  " aPproximaCely 6  hours  for
 60  hours  for  a  slow phase  in fthrel I   *      PhaS&'  3nd
 clearance  in  severa/maJJL  peciL                  f°r
 not stated  in secondary reference)
                           t                o£
species.   Ihe bile e.cred
                                  5-3

-------
 reports a higher biliary excretion rate in rats as compared with rabbits
 or dogs and Cikrt and Benko (1974) noted a higher biliary excretion rate
 for trivalent arsenic forms than for pentavalent in the rat (approxi-
 mately 10:1).

 5.1.2.3  Excretion

      The primary route of excretion of arsenic in man and animals is in
 the urine;  only a few percent  of an absorbed dose is excreted in the
 feces.   Other minor routes of  elimination include the skin, hair, nails,
 and sweat (Hunter et. _al.  1942,  Ducof f et. .al.  1948,  Mealey e£ .al. 1959
 Kadowaki 1960 ,  Lander et aJL.  1965).

      The rate of urinary excretion is dependent upon animal species,
 chemical form of the arsenic,  and the route of administration.   Penta-
 valent  arsenicals are generally excreted rapidly from the body  via the
 kidneys,  while the trivalent forms tend to bind to  sulfhydryl'groups
 of  tissue proteins and are slowly excreted as  they  are released (Webb
 1966).   Organic  arsenicals are  excreted more  rapidly than inorganic
 forms.   The general order of excretion rates  is as  follows (in  the
 reverse order of toxicity):  RAs-X >  As+5 > As+3 >  AsH3  (Fowler 1977).
 The slow excretion of the trivalent  forms is  the basis for their greater
 toxicity; the pentavalent  forms  are  so rapidly excreted  that  very little
 accumulates unless very large doses  are administered over a long time
 period  (Klaasen  1980).

     A  greater proportion  of arsenite  is  excreted from humans following
 parenteral  administration  as compared  with  oral ingestion.  Urinary
 excretion begins  2-8  hours  following  injection of arsenite, but  up  to
 10  days may be required for complete  elimination of  a  single  dose and
 up  to 70  days  after  repeated administration (Hunter  et_ al.  1942,
 Harvey  1970,  Klaasen  1980).  Mealey et_ al.  (1959) report   three  phases
 of  urinary  excretion  in man after  a single  intravenous dose of  radio-
 labelled arsenite  (2.3  mCi/70 kg) with  half-lives of 2 hours, 8  hours
 and  8 days,  respectively; more than 99% of  the  dose  is excreted  within
 the  first 15  hours, excretion occurs at a low  constant rate over the
 next 156 hours and  the  remainder  is removed at  a much  lower rate.
 Crecelius (1975,  1977a, b) reports half-lives  of 10 hours  and 30 hours
 for  the inorganic  trivalent form  and methylated  forms  of  arsenic,
 respectively, when  ingested in wine.   Following  ingestion  of wine con-
 taining 50  ug As(III) and  13 ug AS(V),  approximately 80%  of the  dose
was measured  in  the urine after 61 hours.   Crecelius  (1977b) reports
 a biological half-life  of less than 20 hours for seafood  arsenic, while
 Coulson et  al. (1935) and West88  and Rydaiv (1972) report  elimination
 of  70-80% of an  ingested dose of organic seafood arsenic within  2 days.
Mappes  (1977) reports maximum urinary excretion by 3 hours following a
 single ingested  dose  of arsenite solution, with approximately 25% of
 the dose appearing in the urine 24 hours post-exposure.   Successive
daily intakes of 0.8 mg arsenite resulted in daily urinary clearance of
 approximately 66% of  the daily intake, after 5 days.
                                  5-4

-------
                                     C ** "" ** "«<*
 of the metal is fed  n the heel^T *t C°nsiderin§ th*' 80-90%
 break down before the arsenic can £e   T    CSVed Cel1' which
 (1948)  found urinary .^^ the rat'to beT

 the  first  48 hours  following injection of 47 ur'^f 'han 10%
 arsenite as  compared with approximately 707  • UCl Y^-labelled sodium
 man  48  hours after  injection^  235  S I H  "nnrabbits  and 30-45% in
 respectively.   Less  than  10? of  thf  e'creted     ""  ^  2'? mCi'
 feces of all three  species    ArJn.v ?      . arsenic wa^  found in the
higher  for up to 96  hours    Few er   1979) *  " ^ ^  bl°°d
arsenic in rats of approximately 60  L    f^"3 3 half-li^ of

Urakabo e^ al.  (1975)Prepo"ted ^4 days      "'3    ^ *XP°SUTe>
route of administration n'ot stated S "
  Both arsenate and arsenit  c« be con
  pounds such as methvlarsenates (Feder ?
  urinary excretory producTin both ^
  dimethyl-
                                                            1980) '
                                               or§ani<:  arsenic  corn

                                                 1978>'  The
                                             une*P°^d humans
       l966-
     8^ each the two inorganic f oms
 sample containing 50 u
                                              -     noncnethyl  arsenic,
                                              In8estlon  °*  a vine

 this

              released dimethylarsenic
fro.
administration of As   tralent arsenic
product probably was a mixture of

™
                                                       a^ic  ranging
                                                  f°llowin§  Parentfra!
                                               PntaValent"  Deration
 conversion
                               5-5

-------
      Oral administration of arsenite or arsenate to both dogs and cows
 resulted in conversion of both arsenic forms to methylated arsenic.
 No quantitative distinction was made between inorganic arsenic forms
 (Lakso and Peoples  1975).

      Following  intravenous administration to dogs of radiolabelled
 arsenic acid, inorganic arsenic was  detected as the major form in
 plasma up to about  2  hours post-dosing.   Levels of the dimethyl arsenic
 could be detected as  early as  10 minutes  post-dosing,  and by 6 hours
 approximately 90% plasma  arsenic was in this form, with little detection
 of the monomethyl forms.   Dimethyl arsenic x^as the predominant form in
 the urine from  days 1-6 (Tam_et_al.  1978,  1979).   Charbonneau _et _al.
 (1978b)  detected approximately 80% of an  As?lf-labelled arsenic acid
 dose lodged in  the  red cells following intravenous administration to
 dogs;  in time,  the  arsenic content was distributed between cells and
 plasma.   Dimethyl arsenic was  detected in the  cells by 10 minutes  post-
 dosing,  total conversion  was apparent by  6 hours  and dimethyl arsenic
 was detected in the urine by 1 hour  post-dosing.

      Detection  of dimethyl arsenic in the urine,  feces,  and blood  of
 adult  male rats fed ferric methanearsonate suggested methylation of
 monomethyl arsenic  in  vivo.  The dimethyl  arsenic was  the predominant
 form in  the blood,  but was detected  in minor amounts in  the urine  and
 feces,  a finding possibly indicating  selective  retention of dimethyl
 arsenic  by rat  erythrocytes  (Odanaka  et al.  1978).   The  two above
 studies  indicate involvement of  the  erythrocyte and the  liver in bio-
 synthesis  and transport during  dimethyl arsenic formation in  the rat
 and dog  (U.S. EPA 1980).

     Conversion of  trivalent inorganic arsenic  to  pentavalent  inorganic
 arsenic  in  vivo has been  reported in  dogs  after intravenous  infusion of
 arsenite;  both  inorganic  forms were detected in the  plasma, urine, and
 glomerular  filtrate.   In  the dog, arsenate appears  to be  resorbed  by
 the proximal renal  tubule, reduced and excreted as  arsenite;  the reverse
 oxidation  also  occurs  (Ginsburg  1965).  Benko et al. (1976) detected
virtually no trivalent arsenic in the urine  of mice when  As^-labelled
 arsenite was administered  after  18 days of pre-exposure  to  a  large
dietary  level (250 mg/1)  in drinking water.  The relative amount of
pentavalent arsenic formed was dependent on  the time lapse between
pre-treatinent and dosing with the arsenite (this pentavalent form may,
 in  fact, have been dimethyl arsenic acid).  Winkler  (1962) detected
mainly arsenate in  the livers of rats fed arsenite.

     Data supporting the in vivo conversion of the pentavalent arsenic
 to  the trivalent form  is less conclusive due to the analytical methods
used (U.S. EPA  1980).   Similarly data concerning dimethylation of
methylated arsenic in vivo are limited, although the available infor-
mation appears  to indicate the absence of such conversions  (U.S. EPA
1980).
                                  5-6

-------
   5.1.2.4   Summary
   are also absorbed  ™ the
   v.ll absorbed.  Pentavalent arsenic fhlh        ""^ crloxi^ is not
   better absorbed than triwLnt arsenic  arT8"^? ^ in°r?»"> *«
   MShly dependen  upon £
   its  compods        trtr0;  ^ metabolism  °f  «-«ic  and
   urine.   In human) J expos L t ^ trival^t ^ l'* f^?  dOSe ls  6XCreted
   tion occurs predominantly as methjl arsenic Ln^^"61110' elim
   m ur                                TSenc and dimethl
                                         a
 in  urine.   However, neither  of  th          °  ^  dinet^l  ^rsinic  acid
 following  ingestion of organoarsenica^fins^fn T*  detected  in  u
 urinary arsenic was markedly elevated  _ _ s<;a±ood > although total
 icals may  be excreted without metabolic

5'1'3-  Human and Animal Stud-t^c

5-1.3.1.   Carcinogenicitv


                                       an association
  ^^To'f T'?' an association b— —
  normal concentrations0 ol ^ arsenic "nclu'dS'" T "^ Mgher than
  m Argentina (Bergoglio 1964)  a^d Taiwan ?f P8°P   ln C°rd°ba Province
  studies are discussed in  Section 52  n?^   *! " — '  1968) '   ^The^
  however,  human  exposures  to arsenic h^v  J" Chapter'>   In -»st  cases,
  to ot                                       ^ C°ncren
        ,              es to arsenc hv
 to other agents,  hus one cannot excSe th^ C°ncrrent With
 cof actors may have Played a rol^       '
                                           °a"enic-induced  cancer.
                           rni ^
(15-250 mg/kg dietar  concentraons) o3 J°H-   ^^ S°diUm arsenite
for 2 years).  Survival was reduced. lilL- f  J™ arsenate (30-400 mg/kg
compounds caused weight reduction   It th^K ^'^ arS6nate and both
arsenite and 400 mg/kg arsenate)   a £ft *  highest doses (250 mg/kg
                                '^
duct was evident.  Similarly  no'ca ^   d fnlarSement of the common bile
year feeding study involvin/adininistr^fniC/ffects were observed in a 2
sodium arsenate to beagle do<^s  t ri •       n of sodium arsenite or
mg/kg of either compound.  However  ^ary •Levels of 5, 25, 50 or 125
for risk extrapolation due to the DO        °™    S Study are inadequate
the arsenite-treated group and the i-J,,ff^Val ?f the tSSt animals in
observation periods (Iyro^ « aJ.1JJ^fflclent lenSth °f treatment and
                                 5-7

-------
      Fairhall and Miller (1941)  reported no carcinogenic effects associ-
 ated with feeding rats lead or calcium arsenate at daily doses of 10 mg
 daily for up to 2 years.  A similar response was observed by Kroes et al.
 (1974)  in specific-pathogen-free Wistar rats fed either 1850 rag/kg lead"
 arsenate or 463 mg/kg sodium arsenate in their diets.

      In another study, no  carcinogenic effects were observed in rats
 exposed from weaning  to senescence  to sodium arsenite  at levels of
 5  mg/1  in drinking water.   Elevated levels  of serum cholesterol and
 lower levels of blood glucose were  noted in male rats  (Schroeder et al.
 1968).                                                            	

      Hueper and Payne (1962)  also reported  no development of skin, lung,
 or liver cancer in rats or 50 C57BL mice exposed to arsenic trioxide
 in the  drinking water with or without 12% ethanol.   The concentration
 of arsenic trioxide was increased from 4 mg/1 to 34 mg/1 over the first
 15 months;  administration  was continued until 24 months for a daily
 intake  of 0.2-0.8 mg/arsenic/rat.

      Frost et  al.  (1962) reported no  carcinogenic effects in chicken,
 pigs, or rats  following long-term administration of arsanilic acid.
 No adverse effects were seen  in  chickens or pigs fed the compound for
 4  years  (dose  not stated in secondary reference)  or in  pigs  fed 0.01%
 arsanilic acid for three generations.   When male and female  weanling
 rats  from the  F2  generation of a  six-generation  study  (in which 0.01%
 or 0.05%  arsanilic acid was fed)  were held  on  a  0.01% arsanilic acid
 diet  for  116 weeks, the overall  tumor incidence  was  the  same for all
 groups and was  similar to  the historical incidence  of tumors  in the
 colony (35-45%).

     Levels  of  10  mg/1 sodium arsenite  administered  in  the drinking
 water of  virgin  female  C3H/St mice  for  15 months  reduced  the  incidence
 of  spontaneous mammary  tumors from  82%  to 27%, but  caused  significant
 enhancement  of  the growth  rate of spontaneous  or  transplanted mammary
 tumors  (Schrauzer  and  Ishmael 1974).

     The  sole  positive  response is  a  report by Osswald and Goerttler
 (1971) who found  a considerable increase  in  the  incidence  of  leukemia
 in  Swiss mice, both mothers and offspring,  after  daily subcutaneous
 injections of  0.5 mg/kg  sodium arsenate  administered during pregnancy
 for a total of 20  injections followed by  a  2-year observation period.
 Lymphocytic leukemia occurred within  24 months in 46% (11/24) of  the
mothers,  21% (7/34) of  the male offspring and 16% (6/37) of the  female
 offspring.  When groups  of the offspring  from the arsenic-treated-females
were given an additional 20 subcutaneous  injections of 0.5 mg/kg  sodium
 arsenate  at weekly intervals, 41% (17/41) of the males and 48%  (24/50)
 of  the females developed leukemia compared with 9%  (3/35) males and
 0%  (0/20) females  in the control groups.  Similarly, administration of
 20 weekly intravenous  injections of 0.3 mg/kg sodium arsenate to  female
mice caused lymphoma in 58% (11/19)  of the  treated animals compared with
 0%  (0/16) in female controls.
                                  5-8

-------
5-1.3.2  Teratogenicity
                          ^^
                                                                  „


                            5-9

-------
 with the 30-tng/kg dose injected on either day 8, 9, or 10.  The types
 of malformations were comparable to those noted in hamsters and mice.

      Ridgeway and Karnofsky (1952) detected no specific gross abnormali-
 ties in chick embryos exposed to 0.2 mg/kg sodium arsenate during em-
 bryogenesis.   However, this test procedure is insufficiently reliable
 for an assessment of safety in species such as man that possess
 chorioallantoic placentae.

      Thus,  animal studies have demonstrated that an intravenous or
 intraperitoneal injection of 15-50 mg/kg b.w.  sodium arsenate during
 gestation is  capable of inducing developmental malformations in several
 species.   Few oral studies  are available.   A single study conducted
 with mice suggests a reduced response by the oral route.   Arsenite
 appears to  be more toxic to the embryo than arsenate,  but once again,
 comparative values are available for only  one species  by  one route of
 administration.

 5.1.3.3  Mutagenicitv

      Few mutagen-detecting  assays  that reliably predict mutagenic
 effects in  man have been conducted for arsenic compounds.   Most avail-
 able mutagenicity data are  focused on in vitro cytogenetic analysis of
 sodium arsenate.   No host-mediated or dominant lethal  tests are avail-
 able (NAS 1977).

      Beckman  and  associates  (1977)  found a  significantly  higher (P<0.001)
 incidence of  chromosomal aberrations  in  leukocytes  taken  from arsenic-
 exposed smelter workers  (8.7%) when  compared with controls (1.3%).
 However,  due  to simultaneous exposure to other agents, the effects
 cannot  be attributed  to  arsenic with  certainty.   A  total  of 819 and
 1012  mitoses were  examined  for  the  arsenic  exposed  and control  groups,
 respectively.  The  three predominant  aberrations  found were gaps (6.8%
 vs.  0.9%  in controls);  chromosome  aberrations  including dicentric  chromo-
 somes,  rings,  and  acentric  fragments  (2.3%  vs.  0.09% in controls);  and
 chromatid aberrations  (o.5% vs. 0.3%  in  controls).

      Petres and Hundeiker (1968) and  Petres et  al..  (1970,  1972) have
 reported  chromosomal breakage in human lymphocytes  taken  from individuals
 after  long-term in vivo  exposure to arsenical  compounds,  as well as after
 short-term  exposure  to sodium arsenate in culture.  _In. vivo  studies with
 phytohemagglutin-stimulated lymphocyte cultures from 13 individuals who
 had  previously undergone intensive arsenic  therapy  (up to  20 years  prior
 to this study) indicated an increased incidence of  chromosomal  aberrations
 compared  with  controls.  Expressed as the frequency per 1000 mitoses,  the
 incidence of secondary constrictions was 49; gaps, 51; "other"  lesions,
 26; broken  chromosomes,  61;  compared with 12,  7,  1, and 2  per 1000
mitoses,  respectively, in the controls.
                                   5-10

-------
      In separate experiments, Petres et al.  (1970)  found  that  exposure
 of human lymphocytes in culture to 0.1 ug/ml sodium arsenate resulted
 in pulverization of 33% of the metaphase plates; at concentrations  of
 2 pg/ml or greater, 80 to 100% of the metaphase plates were pulverized.
 In addition, the "mitosis index" and "[3H] thymidine labelling index"
 were decreased.

      Limarzi (1943) reported irregularly shaped nuclei and chromatin
 fragments in bone marrow samples of humans 4 days after treatment with
 Fowler's solution (KOAs=0).

      The effects of sodium arsenate and arsenite on human lymphocytes
 and diploid fibroblasts in culture were examined by Paton and Allison
 (1972).  The addition of sodium arsenite (0.29-1.8 x 10~8M) to lympho-
 cyte cultures for 43 hours prior to fixation produced chromatid breaks
 in 60% of 148 metaphases examined.   With sodium arsenate,  the highest
 non-toxic concentration (0.58 x 10~8M)  caused no notable increase in
 the number of breaks.   With  diploid fibroblasts, exposure  to 0.29-5.8 x
 10~8M sodium^arsenite for 24 hours  prior to fixation caused chromatid
 breaks in 20% of the 459 metaphases examined.

      Arsenate was found to increase the total frequency of exchange
 chromosomes in Drosophila melanogaster  treated  with selenocystine
 (Walker and Bradley 1969).   The significance of this study is  difficult
 to assess  since many unrelated compounds  cause  similar  effects.

      Arsenic  trichloride  was  found  to enhance significantly viral trans-
 formation  of  Syrian embryo cells by a simian  adenovirus, SA7.   An enhance-
 ment  ratio  of  2.2 above controls was seen with  a concentration  of 0.003
 mM AsCl3  (Casto et  aJL.  1979).

      Arsenic compounds  also induce  mutagenic  effects in bacteria.   A
 net positive result  was reported by Nishioka  (1975)  in  a rec-assay with
 Bacillus subtilis H17 and M45  for several arsenic compounds.  A more
 distinct rec-effect  was seen with the sodium  arsenite and  arsenic tri-
 chloride than with  sodium arsenate.  Nishioka also observed an  increased
 frequency of tryptophan reversions  in several strains of Escherichia
 coli  treated with 1.6 x lO'^M  sodium arsenite.  Tryptophan reversions
 were  induced most frequently in strain WP2uvrA   (lacking DNA repair  gene
 urvA), while strain  CM571  (lacking  the functional DNA repair gene  RecA)
 showed little response to sodium arsenite.  These results suggest  that
 the RecA function may be essential for mutagenesis by sodium arsenite,
 while the uvrA gene has little effect (Nishioka 1975, Flessel 1977).

     Rossman _et al.  (1975) found that sodium arsenite decreased the
 survival of ultraviolet-irradiated E. coli strain WP2,  but had no
 effect on the survival of RecA mutant WP10.   These authors also suggested
 that arsenite may inhibit a RecA-dependent step in the repair of ultra-
violet-induced DNA lesions.
                                  5-11

-------
      Thus,  chromosomal aberrations have been induced in human lympho-
 cytes exposed in culture to sodium arsenate and arsenite.   Chromosome
 breaks have also been found in lymphocytes taken from humans exposed
 to various  arsenicals in vivo.   In vitro cytogenetic analysis indicated
 that arsenite induced a higher incidence of aberrations in human lympho-
 cytes than  did arsenate.   Bacterial studies also suggest more marked"
 effects with the arsenite material.

 5.1.3.4  Other Toxic  Effects

      Very little quantitative  data are  available regarding dose-response
 relationships for arsenic compounds.  Most  information concerns
 human occupational or epidemiologic  exposures,  which often do not  allow
 for precise dose measurements,  exposure periods or  chemical speciation,
 and may be  complicated by a variety  of  other  contributing  factors.
 Little information has been reported regarding  animal models  for toxic
 effects seen in arsenic-exposed humans.

      In general,  the  toxicities of the  various  forms  of  arsenic  are
 related to  their rate of  clearance from the body and,  thus,  to their
 degree of accumulation in the  tissues.   Tissues rich  in  oxidative
 systems (e.g.,  alimentary tract, kidneys, liver,  lungs,  epidermis) are
 particularly susceptible  to arsenicals.  The  general  order  of  toxicity
 is  as  follows:   AsH3  > As+3 > As+5 > RAs - X.   The  toxic action  of
 pentavalent  inorganic  arsenic is thought to result  from uncoupling of
 mitochondrial oxidative phosphorylation, which  interferes with phos-
 phate  transport  and phosphorylation.  Presumably, the  arsenate competes
 for  inorganic  phosphate,  forming unstable arsenate  esters  (Klaasen 1980,
 Fowler  1977).  The trivalent inorganic  forms  of  arsenic appear to exert
 toxic  effects  via  reaction with sulfhydryl enzyme systems essential to
 cellular  metabolism indirectly  interfering with  their  activity (Klaasen
 1980).  The  trivalent  inorganic arsenicals have  also been known  to
 interfere with active  transport processes of  potassium, sodium, hydrogen
 ion, monohydrogen phosphate, water, glucose,  and certain amino acids
 (Webb  1966b).

     In humans, the symptoms of acute arsenic exposure and the extent
 of these  symptoms are  dependent upon the type of arsenical, the amount of
 arsenic,  the route of exposure, the age of the subject, as  well as
 other variables.  In general,  the major characteristics of acute arsenic
poisoning in humans are gastrointestinal disturbances and cardiac abnor-
malities.  Subacute exposure results in nausea,  vomiting, diarrhea, leg
 cramps; and disorders  of the peripheral nervous, hematopoietic, cardio-
vascular, hepatic and integumentary systems (NAS 1977).

     Peripheral neuropathy may begin any time from several days to
several weeks after arsenic exposure and may persist for several years.
With chronic or subacute arsenical intoxication, the neuritis is
 symmetrical, widespread and painful, affects both upper and lower
extremities  with both sensory and motor'involvement  (although motor
                                 5-11

-------
    muscle  weakness,  severe pains  in Jh   n  J extremiti".   Symmetrical
    muscles,  foot  drop  Insist drl di ^V^ cramPy P«in in the
    absence or  diminution  crouch   nin  or T  ,  tOUCh  sensati™>  and
    (in more  severe cases)  are rh^  P!m"prick fd  temperature  sensations
    *****  1979, Kyle    '
   abnormS H^r"^ ^"'S^ '" ""' ^"'"cardiogram
   and inversion of the T-wave Bndnr^          °U> includi»g flatteni
   and Herndon 1962,  Goldsmith "d
                   ,                  ono
   It al.  1968).   Arsenic probably «2ti a di~ J6?8"8
   myocardium,  since  these electrocSioar!   f      X1C 6ffect °n the
   correlated with chan                       alterations
           ,                ectrocioar
 correlated with  changes  in sem^      ^ alterations ^ not  be
 (1966)  reported  sudde^cardiL deaths  folC3  ^^ ™^'   Jenkins
nyocardium with  or without  associated  bSn SS  3CU,te P°isoning  of  the
In addition to severe myocardial  in^l      -f   medullary  failure.
of peripheral vascular disturbLces  end      ^ ^  been  reP°rted
extremities,  atropic acrodemaSti^'.nH     §ltlS' §anSrene of the
(Butzengeiger 1946, iSS  SSrzJ W?? Periorbital and ankle edema
Pease 1965).   Blackfoot disease  a nfriJ^T ~ ~   1956' Kyle and
       gangrene of the                        V3S
        arsenism (Tseng
                              ase  a nr            ~
  ing in gangrene of the extreSties has b      V3SCUlar diS°rder result-
  anemia1?                     t^^.^ P— (1965)  noted
  marrow erythropoiesis.  The^emia wL     "JPPUng, and altered bone
  marrow depression.  The anemifand 1^0°°     y/°  hemolysis  ^d  bone
  appear to be reversible features  u'sEpTl^^O) T'  ""nlC P°iS°ning
  feature of arsine poisonina  is hemnife-      ? }*   A Very Pr°minent
  and oxyhemoglobin L the pLSa^Sit^ 'r^  " ^ methem°^°bin
  bleeding from the gums, lips  and no« ?v           S may be exte"sive
 « al. 1963,  Jenkins e_t al? 19S) .      (Neuw^tova  et al. 1961, Meuhrcke
                                                                  been
                                                 disorders
 vascular coagulation and swolLfl&r (f d^  "°ted widesPread in
 several  cases  of  arsenic intoxicaMnn    r  I  ^  mitoch°ndria in
 observed following  ingestio^of Jo^'s^^f3  ^  aSCites ha    e
 arsenic  exposures  (Franklin et al  i«n   T,       ^  in  occuP^ional
 Luchtrath 1972).  St. Petery^et %   Q970?  ^T iger  194°' 1949'
 liver function tests and indiStiL   Is      ** m±ld  ab^™alities  in
 tubules, membranous ultrastructures  £d fT™*1™ influence °- renal
proteinuria, glycosuria, and hemlturia fotl ^ ^StenS' including
arsenic.                      neinaturia following ingestion of inorganic
                                  5-13

-------
      Renal  failure,  secondary  to massive  intravascular  hemolysis  is a
 prominent feature  of arsine  poisoning, but not  of  other arsenic compounds,
 The  kidneys characteristically contain hemoglobin  casts in  the  tubule
 lumens;  cloudy  swelling  and  necrosis  of the  proximal  tubule cells,
 albuminuria,  azotemia, and altered  serum  electrolyte  concentrations
 have also been  reported,  all of which may continue for  many months
 after termination  of arsine  exposure  (Neuwirtova et al.  1961, Nielson
 1968,  Hocken  and Bradshaw 1970, Uldall et al. 1970).The characteristic
 coppery  skin  pigmentation observed  is thought to result from the  presence
 of methemoglobin rather  than jaundice, as measured bilirubin levels have
 been normal (Jenkins ^t  al.  1965).

      Other  late-emerging  symptoms of  subacute or chronic arsenic  exposure
 include  skin  lesions  such as erythematous eruptions, melanosis, auu  other
 abnormal pigmentations and keratoses  of the  extremities, all of which
 are  usually less apt  to be reversible, as compared with other features
 of arsenic  poisoning.  Dry,  falling hair, dry scalp, skin and brittle
 loose  nails,  and white striae  in the  fingernails (Mees  lines) are
 characteristic  signs  of  long-term exposure.  Melanosis  is most  often
 seen on  the upper  and lower  eyelids,  around  the temples, on  the neck,
 on the areolae  of  the nipples  and in  the  folds of  the axillae,  or,  in
 more severe cases, on the abdomen, chest, back, and scrotum  along with
 hyperkeratosis  and warts.  Depigmentation, especially on the pipmrnted
 areas  and hyperkeratoses  and exfoliation  over the  palms and  soles are
 characteristic  of  chronic arsenic poisoning  (Holland 1904, Reynolds
 1901,  Fowler  1979).  Skin lesions may develop long  after the exposure
 period, when  concentrations of arsenic in the skin have returned to
 normal.  Also,  skin lesions,  particularly hyperkeratosis, are thought
 to be  related to skin cancer (Pinto and McGill 1953, Vallee  et al.  1960,
 Buchanan 1962).   A variety of studies have demonstrated that~chronic
 arsenic exposure results  in a characteristic sequence of skin effects:
 hyperpigmentation, hyperkeratosis and finally, skin cancer.   Only a
 small proportion of the keratoses,  however,  evolve into skin cancer,  and
 only after  long periods of time (NAS 1977).

     As stated previously, most data regarding human effects are based
 on occupational or epidemiological studies,  with limited dose-response
 information.  Several studies which give  some quantitative data are
briefly described in the following paragraphs.

 5.1.4  Epidemiologic Studies

     The literature on the epidemiology of incidents of arsenic-related
 effects is  limited, primarily,  to clinical reports  of cases  or groups
 of cases.  Few controlled epidemiologic studies  have been conducted,
 and even case reports are often lacking data on  the form of  arsenic and
 the level of exposure.  The basic source  of  information for  this review
 is the EPA Criteria Document  for Arsenic  (U.S. EPA  1980).  Primary
 sources were also reviewed for  additional  detail.   The following
 sections summarize reports of associations between  exposure  to
                                  5-14

-------
    taminated with arsen      euse waterb

    contamination.   The case  reJorts/stJdierLf "^^ My be  the  cause
                 .    e case reorts/stdie                         use
    In most studies, the chemicaTfom of arsen? SUmmari2ed in Table 5-1

    and concentrations were general^,™ ^Tars^c" ^
                                   ,™
    families  living within  a  7-S S     I  } describes  studies of 200
    Produced  copper" aj [ ar sen, c^ri olid" ^ ^T^' ^ WhlCh
    tioa found in well water wL ?125 ^7l  iJif^f arSenic "ncentra-
    most wells was 0.05 mg/1 or less   K' f ^  the concentration in
    having arsenic poisoning hid a historv of      P"S°nS identif^d ^
    arsenic.  One couple,  however  wJtho,7 «  °CCUPation^ exposure to
    consumed drinking water In the  viMn >  °CCUpatlonal associations,  had
   The wife was said§ to be s'ffe^"10^^^1^^ °'™  =8/1  arsenic.





                                 ^

  and over with a maxi^ ^0^^                      °'°5
                                                            disease,

                                                                 ic

                                                                 et al

                                                                -
                    c1snt

 health effects including hyper ker^tosis^

 chronic cough, lip herpes, cardio^scular                      se

 diarrhea and abdominal pain (Borsono ^ r   t festations,  chronic
 1977, Zaldivar 1974  I976a?  A^ T     Greibe^ 1972,  Borgono et a

 the 265,000 inhabitants of Antof List! Ir ^^ 3nd  Gu""«-(19
 population at risk from arsenic  ifdrinL        represent  the  largest

 nation results from the leaching by  raiSaL^f"'  Th£ arS6nic  conta
 volcanic  sediments, which  have f M  Jainwater  from  strata  of lava and

 form>0f arsenic pr^ent was lit  s^llT^  ^^^  ^ ^^
 figniticantly higher prevalence  of ^f    InvestiSators  observed  a

 - Antofagasta  as  compared Sth  otheTare^ with'i' tO ^^ ^

 Furthermore,  Zaldivar and Gulllier Q977?   !^     " arSenic levels
 describe the  reductions in ^      ,     } and Borgono et al.  (1977)

 city of Antofagasto^Sich "c^d'Si' ^ T^ ^^nc  In the
nmnx                                       °
                              c
           water treatment  ayst^   ^P      °8  the  lnst*U«ion of a

 tTJ^ of /arsenic  for  the p«Si i955-i9r7eopo0rf o rigi;ted mean
 to  0.08 mg/1  for  the period 1970-1972    f      *6 mg/1' whlch decreased




did not develop cutaneous lesiJns  wh.Vh      he Water tre^ment  system

in prior years.  The authors  point  oS th«^ Ch,ildren had  d-eloped
be cue to an increased  period  of Ltencv^  ?       ^ °f lesions ^
tmuation of periodic examinations      '    1OWer d°SSS and  «ge  con-
                                 5-15

-------
                                     TABLE 5-1.   EPIDEM10LOG1C  STUDIES  OF  THE  HEALTH EFFECTS OF WATERBORNE ARSENiC EXPOSURE
Oi
               Location
          Argentina,
            Cordoba Province
          Canada,
            Nova Scotia

          Chile
            Antofasasta
         Croat  Britain,
           Herefordshire
         Nakajo
         Shimane
     Route  of
     Exposure


Drinking water





Drinking water


Drinking water




Drinking water



Drinking water





Drinking water
      Level / Type
      of Exposure

2,8-54.!> mg/1 due to
  geologic contamina-
  tion
Geological contami-
  nations

Geological contami-
  nation of water
  supplies up to 0.96
  mg/1 arsenic

Arsenical pesticide
  back syphoned into
  water supply

Wells contaminated
  with waste from arsenic
  sulfide plant; many
  wells 1-2 mg/1 arsenic;
  maximum 3 mg/1
      Study
   Pcipul ationjs


Deaths in the
  province, skin
  cancer cases
Several population
  surveys
Farm family
                                                                                 Case in the plant
                                                                                   vicinity
                                                       Contamination of wells    Examination of 200
                                                         from mine producing       families living
                                                         copper and arsenic tri-   within 7 kin radius
                                                         oxide; most wells 0.05    of the mine
                                                         mg/1 or less; highest
                                                         0.125 mg/1; next highest
                                                         0.07 mg/1.
                                                      Health  Effects/
                                                         Results	


                                                  Proportion  of deaths
                                                    due  to cancer:   23.8%
                                                    in arsenic area;
                                                    15.3Z  in  non-arsenic
                                                    area
Estimated dose for skin
  cancer induction;
  0.62-1.15 mg/Jay;
  latency 14-23 years

One fatality; six non-
  fatal poisonings
                        60 cases of As poisoning
                          identified from clin-
                          ical observations
                                                                       7 cases of arsenic
                                                                         poisoning; "}
                                                                         suspect cases
                               Reference


                          Arguello  t^t  al_.  (19'1H)
                             Bergoglio  (1964)
                             Lisella (1972)
                                                                            llindirursch e^ al .
                                                                              (1977)      ~

                                                                            Zaldivar and Cullier
                                                                              (1977)
                                                                            Borgono e_t al . (1977)
                                                  Tsuchiya  (1979)
                          Terada (1960)
                                                  Tsur.hiya (1977)

-------
 I
t-'
•--I
                Location
           Poland,
             Keiclienstein,
             Silesia

          Taiwan
            Peinian
            Vl-Ohu Districts
                                       TABLK  5-1.
       Route of
       Exposure
  Ottnking water
 Drinking water
                                                                               »„„
          United  States
           Oregon,  Lane County
           Minnesota
             Perl mm
           California
             l.iJssen Counly
           Japan,
             Okayaina
            llbu
 Drinking water





Drinking water



Drinking water


  Powdered  n-ilk





 Soy Sauce
         Level / Type
         of Exposure
  12.2 rag/1,  source  and
    type of contamination
    unknown

 Geological contamination of
   artesian wells,  levels
   0,01-1.82 mg/1 arsenic,
   maybe predominantly trl-
   valent, lysergic acid or
   related compounds present
Geological contamination
  of water supplies; levels
  range 0-2.15 mg/1, mean
  0.009 pg/lj presumably
  pentavalent

Contamination of wells by
  buried insecticide
  arsenic  12-21  g/llter
                                                                                           Study
                                                   Surveys of 37 vi t-
                                                     lages with a
                                                     population of
                                                     40,421
                              Skin  cancer cases
                                and controls
                                                                                    Clinical  study  of  13
                                                                                      persons  exposed
                                                                                      for  2.5  months
0,1-1,A We/l Utcr arsenic   Not specified
  Sodium phosphate stabi-
    lizer with 6X As205;
    mtlk contained 21-
    34  Kg arsenic per gram;
    doses range  2.5-4.6 rog.

 Arsenic  contamination
    (90-100 U8/B.L)  from
   an unspecified source;
   exposure over 2-3 week
   period
                                                                                  Bottle-fed infants
                                                                                  About 400 persons
                                                                                    exposed,  220 cases
                                                                                    examined
                                                                                                               Health  Effects/
                                                                          Skin cancer
                                                                                                 I-isella  (1972)
                                                                                                 Geyer  (1898)
  Skin cancer preva-    Yeh (1963)
    lence 10.6 per 1000 Tseng (1968-1977)
    populanon.  Posi-   Lu (1975,l9/7d  I)
    tive  correlations                 '
    hetweem arsenic
    levels  in  well water
    and skin cancer
    prevalence
                                                                                                           No positive associa-
                                                                                                             tions
                                                    Intermittent (U
                                                      symptoms
                       Morton  u^ a] .  (19/6)
                       Pelagians (1973)
                       U.S.  liPA (1977)
                                                                                                          12,131 cases of
                                                                                                            arsenical poisoning
                                                                                                            with 130  fatalities
Swollen liver and
  abnormal EKC ln 80%
  of cases;  symptoms
  disappeared 2 weeks
  after exposure ceased
                                                                                                Mizuta

-------
     A similar situation of drinking water  contamination  exists  in  the
Cordoba Region of Argentina and in Nova Scotia, where geological con-
ditions contribute to high arsenic levels  (Arguello _et al.  1938:
Bergoglio 1964; and Hindmarsch  et al. 1977), Bergoglio~Tl964) reports
the results of a study in Argentina from 1949-1959, which found  a higher
proportion of cancer deaths in the arsenical region than  in the  rest of
the Cordoba province: 23.8% vs. 15.3%  (no dose information  is available
in the secondary source).  Tseng et al. (1968) refer to the District of
Reichenstein, Silesia, Poland where, in about 1889, an unusual cancer
epidemic was associated with arsenic from mining slag heaps leaching
into, ground water.  Apparently, the construction of a new water  supply
in 1928 eliminated the problem.

     Some of the most intensive epidemiologic investigations of  exposure
to waterborne arsenic have occurred in Taiwan where a population of about
100,000 on the southwest coast is exposed to high (0.8-1.82 mg/1) concen-
trations of arsenic (chemical form not specified) in water  from  deep
wells.  Yeh (1963) conducted a general survey of six villages with a
total population of 3,938 to determine the prevalence of  chronic
arsenical intoxication.  Of the total population, 80.7% were examined
and 36.8% of that population was found to have signs of chronic
arsenical intoxication.  The highest prevalence rates ranging from
93%-100% were found in the age groups over 50 years.  The survey  also
identified 35 cases of skin cancer, which equal a total prevalence of
1.1%.   For those over the age of 50, the prevalence was 6.8%.

     Yeh (1963) also studied an endemic peripheral vascular disorder
called Blackfoot disease, which results in gangrene of the  extremities
and reported close associations between this disease, signs of arsenical
intoxication, and skin cancer.   The associations, however are insufficient
to establish casual relationships (U.S. EPA 1980a).  For example, Lu et al.
(1975, 1977a, b)  analyzed water samples from areas  with endemic BlaclcfooT
disease and identified lysergic acid or a related compound.   Since lysergic
acid has well-known vasoconstrictive effects, it, rather than arsenic,
could be the cause of Blackfoot disease.   Tseng et  al.  (1968) reports on
additional investigations in Taiwan where,  by the end of 1965, 37 villages
with a population of 40,421 had been surveyed.   Within this  population
the overall prevalence rate of skin cancer was 10.6  per 1000 population;
for hyperpigmentation the prevalence was  184 per 1000;  for keratosis the
prevalence was 71 per 1000; and for Blackfoot disease the prevalence was
9 per 1000.   Arbitrary segregation of population groups into "high"
(above 0.6 mg/1),  "medium" (0.3-0.6 mg/1),  and "low" (below 0.3 mg/1)
arsenic exposure groups resulted in an obvious association between
increasing concentrations and increasing  prevalence  of  skin  cancer.
                                 5-18

-------
                          Hy^
                  worsened  even  after termination of exposure.
                                                                      al.
                                            ¥btfe«S_;OpUte=fos Cfeegan.
 		 — „,„ ^  _^ r«"-~"«<=»  Second, the trivalent  form of
 arsenic  present in the Taiwanese water samples was not prevalent  in the
 15a.Ii€: 5Co'fflgrv!;iftPles (u>s-  EPA 1980a).   Other explanations include smaller
 sample size;  differences in socioeconomic, racial, and dietary  factors;

 ¥11.^1 P^^eW^t^foififjffiy 93?%Je¥Bflpa-ffi9iEfti Hl&£ggtt in Taiwan as
 either casual or potentiating ractors.A more detailed discussion of
 the rai-i-J	' '   '  •           ......

 UIOTI  fur Tnte Traxiirfunr ^rrotection of  human  health from the  potential  car-
 xinogenic effects of arsenic exposure through  ingestion  of water and
 c'oTi'tamandtiUtf afifUat^?-IlAy£nisms.  The  water  quality criterion
      lumber of eoidemiologic studies  thajt have shown that both
                                        ,  in 1955, resulted from the
consumption or powdered milk, which included a sodium phosphate stabili-
dence
                                                         .disturbances ,
      ding  hearing damage,  abnormal brain wave patterns, increased  inci-
                                                                       at
                                                                    oto
1 q 7 o \
iiti'de'r study at EPA.
 5.1.^y5^^i^^cftf^^hC(?Sfe^%a^^n^  reported by Mizuta et al_.
 U956) wno^examined  20u Japanese  patients who had ingested approximately
^t^^T^njriifeftt^aTSfei^1^ taoi^a^^eielff^eiWi^tt^ecK^ap^^  fWifffer
9^&vc^L SfftaS:%ii&afcn3t?^d the different toxicological properti
                                    toxicological properties
correlative evaluations are difficult  to make.   In general,
                   followed in decreasing  order of toxicitv
by  arsenites,  arsenoxides,  arsenates, pentavalent arsenicals,  arsonium
compouWs6,1 aW m^eWEffi: (aYs5
-------
                                  TABLE 5-2.   ADVERSE  EFFECTS OF ARSENICALS ON MAMMALS*
    Adverse gffeet

    Carcinogenicitv
   Teratoqenicitv
  Mutagenicitv

  Chromosome
  Aberrations

  Chromosome Breaks


 Chromosome Breaks
   S
    Rat


    Rat


    Rat


   Rat



   Ewe


   Mouse


  Mouse


  Hamster
 Human  (lymphocytes
 in culture)

 Human  (lymphocytes
 in culture)

 Human (lymphocytes
 in culture)
                         Human (infant)
Peripheral Neuropathy   Human
Hyperkeratoses/
Hyperpigmentation
                             Lowest Reported
                             Effect Level  _

                             0.3 mg arsenicals in
                             drinking water;
                             epidemiological study
Human
                                                    120 mg/kg  Na arsenate
                                                    orally


                                                    10 mg/kg Na  arsenate
                                                    intraperitoneally

                                                   15 mg/kg Na arsenate
                                                   intraperitoneally
  0.1  ug Sa  arsenate
  per  ml medium

  0.29 x 10-8 M Na
  arsenate
 3.5  mg arsenic per
 day  in milk

 3 mg/day Ca arsenate
 in soy  sauce for 2-3
vks.


8.8 mg arsenic trioxide
orally for 28 mo.
                                  No Apparent
                                  Effect Level
                                                                                   463 mg/kg Na arsenate in
                                                                                   diet for 2 yearsc

                                                                                   250 mg/kg Na arsenite in
                                                                                   diet for 2 years'^

                                                                                   1,850 mg/kg  Pb  arsenate  in
                                                                                   diet for 2 yearsd

                                                                                   10 mg/rat/day Ca  arsenate
                                                                                   in diet  for 2 years
                                                                                  0.5 ing/kg Kb arsenate in
                                                                                  diet during gestation
                                                                                0.58 x 10-8 M
      taken  from Section  5.1.3  of  this  chapter.


            to 23 mg/kg/day based  on consumption of 15 g feed/day in 300-g rat

            to 12.5 rag/kg/day based upon the assumptions above

 quivalent to 92.5 mg/kg/day based on the assumptions above
                                                   5-21

-------
      There  is  also  considerable  evidence  for  an  association between
 arsenic and disease in humans.   Skin  lesions,  including cancer and a
 circulatory disorder known  as Blackfoot disease  are  major clinical
 problems associated with  chronic arsenism.  Animal models,  however,
 have  not demonstrated the carcinogenicity of  arsenic,  even when
 administered for  long periods of time at  near  maximal  tolerated con-
 centrations.   Clarification is needed as  to whether  arsenic is car-
 cinogenic solely  for man  or whether some  vital factor  is  lacking in
 animal  models.

      Animal studies have  shown that arsenic causes fetal  death at  high
 doses and malformations at  lower dosages  in mice, rats, and hamsters.
 An  intravenous or intraperitoneal injection of 15-50 mg/kg  b.w.  sodium
 arsenate during gestation is capable  of inducing developmental malfor-
 mations in  several  species.  A single study conducted with  mice suggests
 a reduced response  to exposure by the oral route.  In general,  arsenite
 appears to  be more  toxic  to  the  embryo than arsenate, but only  few  com-
 parative studies have been  done.

      Chromosomal aberrations have been induced in human lymphocytes
 exposed in  culture  to either sodium arsenate or arsenite.   Chromosome
 breaks  have  also been found in lymphocytes taken from humans exposed
 to various  arsenicals in  vivo.

      Other  toxic manifestations  of arsenic exposure in man  include
 effects  on  the gastrointestinal  tract, cardiac abnormalities, peripheral
 neuropathy,  kidney  and liver disorders, and characteristic  skin  dis-
 orders  (hyperpigmentation, hyperkeratosis).

 5.2  HUMAN EXPOSURE

 5.2.1   Introduction

     As  discussed in  Chapter 4.0, arsenic is found in all environmental
media.  Hence human exposure to arsenic occurs via many routes.  However,
 exposure  is not well described  in the literature, aside from exposure
 in occupational settings.   The  complex environmental chemistry of arsenic
requires  that the form, as well as the route,  be considered in the evalua-
tion of risk due to each exposure situation.   This has not always been
possible because the form(s) of arsenic involved in a specific route is
not always known.   Monitoring data for environmental media are most
commonly reported for total arsenic concentrations rather than for
specific chemical forms.  This  section addresses  the various routes of
arsenic exposure,  as well as the form as specifically as possible.
The magnitude,  extent, and frequency of exposure  associated with dif-
ferent routes are quantified to  the degree the data permits.

     Human beings have always been exposed to  at least low levels of arsenic
 from natural background levels  in various  environmental media including water,
 soil, air and biota.  Certain subpopulaticns may have evolved adaptations
                                  5-22

-------
                                                            has  been, some
            (NAS 1977);  however this possibiStv^"11 SSS\ntial n^ient for
    investigated.  Ideally,  exposure  of  humpn. I   S ,n0t faeen carefullv
    sources of  arsenic should be^assessed^ *UfapO^ati™s «=o industrial
    population's previous exposure tbacker,„*!Und*rStanding of the
    of  the  significance of the industrial  LuT  leVels/nd Consideration
    levels  already present.     inaustrial input compared with arsenic
   5-2.2  jngestion


   5-2-2.1  Drinking Water

        Exposure Lave13
                                  8ln
1970), only Q.2% of the samples L f    , Wa"r SUPPlies ^.S.  DREW
than 50 ug/1 and only 0 J^e ^f1™**™^ at levels greater
sampled 5% of the nation'^ wate/sulT   ^ 10 'S/1'   The  s^^
were ground water,  12%  LrJc ^  57 mS^      ^ 197°'  °f whic^  63%
                 er,        rc   57 md               '
   systems.  In other national surveys  fe' s^l    renaininS  20%  special
   level exceeding 10 ug/l (Greatho^f ' f^saraPles contain  an  arsenic
   The mean level detected in  ?S5     f   Craun 1978' U'S'  EPA 1975)

     =-^             ~»«     '^-^s^4
   ->
               All  of  these
    c                           ."    municipai) -
  significant industrial sonrc. nf ,1   " /  S non£errcl"S smelters, a
  »                                  "'
         fro» <5 vg/1 to
                                              ,
  comparable to those found in t   nationa! ^A"' <°n™"^°™ «»
                       .                      '« «"»" ».« arsenic
in ground .,ater was attributed « h  S°"rces'  In °" "se,  the arsenic
1973) (see Chapter 4.3 J  ?he arsenifff "^^ "6sti':"«  (Fein8Uss
Ar12ona nay oe due to contaminlti™ -esulM 'V^ "ater ^""^ of «°,
However,  several  other  vater smoli ' f \g      a n"rby  snelter.
-ntained levels  of  less thaVlT              "'          -"i""
                                                              Kecently,
                                5-23

-------
                          TABLE 5-3.  ARSENIC LEVELS DETECTED IN DRINKING WATER IN THE U.S.
Ul
I
U.S. (Community Water
Supply Study—surface
and well)
U.S. Drinking Water Survey     3834
(Residential Tap Water)

U.S. (Interstate Water          566
Supply Study)

Alaska—suburban community       59
(well water)

Utah, Nevada, California
(smaller communities)
       Perham,  Minnesota
       (well water)

       Lassen County,  California
       (well water)

       Lane County, Oregon              558
       (welI water)

       A jo, Arizona (ground water
       at  location of  copper  smelter)

       Wentatcbee, Washington  (well
       in  area  receiving lead
       arsenate treatment)

       Bottled  Mineral Water
                                    No. Sampjles    % Detected

                                      2595
0.2% > 50 ,,g/l
0.4% > 10 ug/1

66.8% < 0.1 Mg/1


1%  > 10 jjg/1
 8%  >  50 ug/1
 Concentration (iig/1)

 Mean    Maximum

           100


 2.37      213.6
                      224     2450


                              10-330



                           31,800-21,000


                              100-1400
9,6
2150


  70
                     5.6
                                                                      20
                                190
              Source

              U.S.  DHEW (1970)


              Greathouse  and  Craun (1978)


              U.S.  EPA  (3975)


              Public Health Service (1977)
                        Valentine  (personal  communi-
                        cation  as  cited  in U.S.  EPA
                        1979)

                        Feinglass  (1973)
                       Goldsmith  et  al.  O972)
                       Morton et al.  (1976)
                                            Baker et al_. (1977)
                                            Falrhall (1941)
                                                                                             Zoeteman and Brinkman (1976)
              Not  Available

-------
            TABLE 5"-
  Location
  U.S.  Major River
  Basins (1975-1979)
  Edgemont Well
  Study  (1979):
  South Dakota
  and Wyoming3
                                        Concentration
                      .No*  of Samples
1922
 13
              2-22
               27
                                                   Maximum
                           500
                            67
                                     Reference
            STORE!
            U.S.  EPA (1980b)


            STORET
            U.S. EPA  (1980b)
 Thermal Waters:
 Wyoming, Nevada,
 California,
 Alaska
             range:
             28-3800
3800
           Fleischer (1963)
These data are included in preceeding U.S.
                 data base.
                                5-25

-------
 in New England, a number of concentrations in drinking water wells  have
 been reported in violation of the 50 ug/1 drinking water criterion, with
 suspected arsenic-associated toxic effects in some consumers of  the
 water supplies (Chow, C., personal communication February 17, 1981;
 Boston Globe 1981).  The source of contamination is uncertain but sus-
 pected to be natural background levels in the soil or bedrock.

      Ground water levels are also compiled in the USGS monitoring well
 data presented in STORET (see Chapter 4.0 Table 4-14).  Table- 5-4 presents
 the results.  A total of 1922 observations (559 remarked) for the last
 5 years in six river basins indicated mean major river basin levels
 ranging from 2 ug/1 to 22 yg/1,  with a maximum level of 500 ug/1 in the
 Lake Erie basin.   These data are not from potable water supplies, however
 they are probably similar to arsenic levels in untreated drinking water
 from wells.   The  data points are too few and not well distributed enough
 nationally to indicate areas of  high potential for human exposure.  A
 local study (part of the major  study)  conducted in a minor river basin
 (Edgemont study)  with a large number of high arsenic levels  in all types
 of water reported all ground water levels less than 70 ug/1.   The high
 levels  in the basin  were attributable  to  high surface water  concentra-
 tions.   It is unlikely that humans are exposed to the very high arsenic
 concentrations associated with areas  of thermal  activity shown in
 Table 5-4.

      Table 5-3  suggests  that levels up  to  2000 ug/1 may occur naturally
 in isolated  water supplies,  resulting  in an exposure  of 4000 yg/day.
 Exposures of this magnitude  are probably limited  to an extremely  small
 sub-population.  Whanger  (1977) reports that  "the natural occurrence
 of arsenic in ground water of Lane County  is  the  only one in a well
 populated area of the North American Continent."  However no extensive
 monitoring program that would support such a  conclusion is known.
 Slightly more common exposures in known contaminated areas may be on the
 order of 200-400 ug/day.

     The form of arsenic in water is usually  soluble arsenates and
 arsenites.   It is expected that most of the arsenic present in surface
 water sources of drinking water would be in the form of arsenat-
 due to aeration during finishing.  Clement and Faust (1973) reported
 that approximately 8% of the total arsenic in aerobic streams was in
 the trivalent form, while, most  of the arsenic in anaerobic waters was
 in this form. [The speciation diagram, Figure 4-3 (Chapter 4.0),  supports
 this observation,  showing almost  100% arsenite in soluble form under
 reduced conditions.]   Anaerobic waters may contain arsine and methylated
 forms.  Braman and Foreback (1973)  report that methylarsonic acid and
dimethylarsenic acid  are found in surface waters.   The. form of arsenic
in ground waters is not well studied,  but Clement and Faust (1973) found
 that 25-50% of the arsenic in a  limited number of ground water samples
was in the trivalent  form.
                                  5-26

-------
      Potential Sources, of Ground Water Contamination

      Since little monitoring data for arsenic in ground water are
 available and considering the significance of land releases and
 arsenic's mobility under certain environmental conditions, it is rele-
 vant to review specific situations in which there is a potential for
 ground water contamination.  The information in this subsection was
 discussed more thoroughly in Chapter 4 Pathway 2, but is reviewed here
 in an attempt to identify geographic areas in which ground water may
 be contaminated and human populations may be exposed to arsenic via
 ground water.

      Table 5-5 summarizes releases by form, disposal practice and dis-
 tribution for the major sources of arsenic releases to land.   Maps of
 these distributions are compiled in Appendix D.   Limited data are avail-
 able on disposal practices, which are the key factor in determining
 whether or not arsenic will migrate out of the disposal site.   Hence
 the various national regions listed are areas wi^h potential ground water
 contamination.   Further investigations are required to link actual in-
 cidences of contamination with sources.   Information is also  needed on
 typical solid waste disposal practices by industry.

      Specifying  the source of arsenic to  land is  only one  factor,
 although an important  one,  in determining the potential for migration
 of the  element to ground  water.   Once arsenic leaks,  overflows  or  other-
 wise escapes  from a disposal site,  the surrounding  soil and climatological
 characteristics become important  determinants  of  the  element's mobility.
 Table 5-5  lists environmental characteristics, both pedological  and
 climatological, that are  conducive  to arsenic migration through  soil.
 Appendix D  includes  a  map  of the  national distribution  of precipitation
 rates and  soil types.   The  characteristics are meant  to be  illustrative
 only; migration will not  occur wherever any one of these traits  exists
 because  it  is dependent on  a set  of characteristics.

      Comparison of Tables 5-5 and 5-6  suggests some areas in the U.S.
 where,^based on loading rates and environmental characteristics, the
 potential for migration of arsenic into ground water appears to'be high.
 These areas include the central eastern industrial states of Pennsylvania
 southern New York, Ohio, Indiana, and  other states where a  large
 amount of arsenic is disposed of, a relatively high rate of precipitation
 exists, and the predominant  soil type would appear to favor arsenic
 mobility.  The southeastern  states have high pesticide use rates and
 high precipitation rates,  but a predominant soil type high in available
 ferrous and aluminum hydroxides which would tend to reduce arsenic
mobility.  The southwestern and Great Plains states have relatively
 low precipitation rates which would reduce mobility but soil low in
 available hydrous oxides.   Washington, where disposal of arsenic pro-
duction wastes are concentrated,  has a high rate of precipitation and
variable soil types, some  of which are more conducive to migration than
others.
                                  5-27

-------
                      TABLE 5-5.
                           ENVIRONMENTAL RELEASES OF ARSENIC TO LAND; ESTIMATED ANNUAL
                           VOLUME, CHEMICAL FORM AND GEOGRAPHIC DISTRIBUTION
Ul
I
ro
oo
       Source
       Fossil  Fuel Combustion
       Pesticides
        •  herbicide use
        •  cotton  ginning

       Copper Production
Iron and Steel
Production
       Boron  Production


       Manganese Production


       Arsenic Production



       Lead Production


       Pbosphorus Production

       TOTAL
                            Quantity
                              (kkg)

                            14,000
 8,000



   580

 8,100


 5,700



 2,200


 1,400


 1,200



 1,100


   640

43,020
               Predominant
                   Form

                 trioxide
MSMA, DSMA, +3



arsenious acid

  trioxide


  trioxide



  trioxide


  unknown


  trioxide



  trioxide


  unknown
                  Application
                      Form	

                  ash ponds
directly to
agricultural
land

landfill

sludge from
tailing pond

landfill or
ponds
                                                             return  to
                                                             brine lakes

                                                             slag
                                                            slag piles or
                                                            road construction
                                                            material

                                                            slag piles
                                                            50% released in
                                                            fertilizers
                                                            applied to agri-
                                                            cultural lands or
                                                            land disposed
                  Distribution

                  National, 65%  in  industrialized
                  northeast and  central eastern U.S,
Approximately 85% of use  in
south central and southeastern
U.S. (JRB 1980)

Texas, Oklahoma

Utah, Arizona, New Mexico,
Montana, Nevada

Primarily industrialized north
and central eastern U.S.
(PA, NY, OH)

California
                  imported, waste disposal
                  practices unknown

                  Washington
                  Missouri, Colorado, Idaho,
                  Utah, Other

                  Tennessee,  Florida, North
                  Carolina, Western U.S.
                  for wastes from production
                  processes.

-------
        TABLE 5-6.   ENVIRONMENTAL CHARACTERISTICS FAVORABLE TO
                    THE MIGRATION OF ARSENIC IN SOIL TO GROUND WATER
      Characteristic

 Climatological

   High precipitation rate


   Low rate of evapotranspiration
 Potential distribution
 Northeast,  central East,  Southeast
 and  Northwest  U.S.

 Northeast and  Northwes.
 Soil Characteristics

   Soils  low in iron and aluminum
   hydrous  oxides  (low kaolinite
   clay content)
   Reduced  soils  at  pH  4-7


   Permeable soils

   High microbial activity



   High pH  in aerobic soils
Soil orders:   Inceptisols  (S. NY,
PA), Alfisols  (IN, OH, WI),
Mollisols  (Northwestern and
North Central  Plains, Great
Plains, Texas), and Aridosols
(Southeast and  yreat Basin)

National distribution, intermittent
occurrence

Spodosols  (N.E.) and Ultisols (S.E.)

National distribution in temperate,
moist areas, especially agricultural
soils

National distribution, more prevalent
in western states
Source:  Brady (1974); Arthur D.  Little  (1980);  Eyre  (1968)
                                  5-29

-------
      This analysis is very cursory and has been made only for a first
 approximation of the potential for arsenic ground water contamination
 by region.  Many other factors which are too variable to consider for
 a gross national-scale analysis—depth to water table, predominant
 arsenic form present due to site parameters, presence of fissures or
 cracks in soil through which mobility would be increased, and others—
 also have significant influences on the fate of land-disposed arsenic.
"Due "to ~t"n.e~Iarge~amouht~oT" arsenic "In ~land wa~stesT~the element's~~some-
 times high mobility, and reported incidences of ground water contamina-
 tion, a more detailed investigation of this subject is needed, with a
 focus on localized situations.

      Relationship Between Exposure Levels and Urine Concentrations

      A relationship between urinary levels of arsenic and rate of intake
 of arsenic in drinking water has been established by the U.S. EPA
 (Mushak et al.  1980).   Figure 5-1 presents the regression curves for
 urinary levels  versus daily consumption of arsenic in drinking water
 established from two studies in Arizona and Alaska.  According to the
 graph, most of the U.S.  population" (exposed" to 5' yg/day based on
 estimates in 5.2.2.1)  would have arsenic levels of less than 100 yg/1
 in their urine.   The subpopulation exposed to the maximum level of
 4,000 yg/day would have urinary levels  of from 650 yg/1. to over 800 yg/1.
 If one compares arsenic urinary levels  reported for people exposed to
_high arsenic levels—'"urinary "levels which range from >300 yg/1 to <700
 Ug/1 (U.S7TTPA 19Wa7^(aiso see Chapter 5.2.4) with the levels
 on Figure 5-1,  an intake level from drinking water can be estimated.
 This requires making the assumption that the levels present in the
 urine originate from ingestion of contaminated drinking water and not
 through inhalation of  arsenic in air.   This can be justified based on
 the small exposure levels associated with inhalation (at least two
 orders of magnitude below ingestion levels) for the general population,
 assuming that absorption efficiencies for the two routes do not differ
 by more than one order of magnitude.  Making this assumption, the
 arsenic urinary level  measured in the majority of the U.S.  population
 (Gills et al.  1974), 22-25 ug/1, indicates a negligible intake of arsenic,
 confirming the  conclusions based upon the limited monitoring data.   Higher
 urinary levels  resulting from exposure  to contaminated drinking water in
 Chile (700 yg/1—Zoldivar and Guillier  1977)  and in the vicinity of a
 smelter (539 yg/1—Pinto et al.  1976;  300 yg/1—Milhan and Strong 1974)
 indicate intakes on the  order of 1 mg to 2.5 mg arsenic per day.

      These extrapolations should be used with caution at this time and
 certainly require further investigation.   The potential for uptake from
 sources other than drinking water or through other exposure routes
 exists and these factors could influence the apparent correlation between
 urinary levels  and water concentrations.
                                  5-30

-------
                                                                          Arizona
                                                                -O-O-   Alaska
    u                0.5

Source: Mushak era/. (1930)

          FIGURE 5-1
1.0
                  1.5
                 Water Arsenic Intake (mg/day)
                                                     ARSEN1C ,NTAKE ,N

-------
 5.2.2.2  Food

      The sources of arsenic in food are numerous.  The wide use of
 arsenic as a pesticide was an important source in the past; several
 cases of adverse effects from eating arsenic-treated fruit were re-
 ported (Cannon 1936).   However,  the pesticidal use of arsenicals has
 declined in recent years.   The U.S. FDA allows 3.5 yg/1 of arsenic in
 fruits and vegetables  (NAS 1973).   Concentrations of 100-500 yg/1
 arsenite have been measured in bottled wine (Crecelius 1977).  °A daily
 per capita consumption of  28 ml  of wine has been estimated for the U.S.
 population over 21 years of age  (Amerine and Singleton 1974).   Assuming
 100 yg/1 in wine,  this leads to  an exposure of 0.004 ug/day for the
 average wine drinker.   For the subpopulation ingesting 1 liter of wine
 daily with 500 yg  arsenic/1, the exposure would be 500 ug/day.   Crecelius
 (1977)  found arsenite  to be the  predominant form in wine,  at  levels
 approximately five times greater than  those of arsenate.

      Also,  finished moonshine was  found to contain arsenic at  levels
 as  high as  415 yg/1 (Gerhardt et al. 1980).   An estimated  266  million
 liters  of moonshine is produced  each year (Ball and Sorenson  1969)  so
 exposure  may be more important than one would be inclined  to  think.
 Assuming  ingestion of  28 ml/day, an exposure  of 0.01  yg  arsenic/dav
 would result.

      Arsenic is also present in  meat and fish products.  It is  used as
 a growth  stimulant  for swine and poultry,  accounting  to  some extent for
 levels  found in these  foods  (Jellinek  and  Cornelieussan  1977).  The
 FDA allows  a maximum of  0.5  mg/kg  in muscle tissue,  1 mg-/kg in  edible
 byproducts,  and 0.5 mg/kg  in eggs.  Arsenical  feed-additives are
 required  to  be removed from  the  livestock  diet  for  the 5 days  immedi-
 ately preceeding slaughter.

      Arsenic appears to  accumulate  in marine  organisms, especially in
 bottom-feeding fish and  crustaceans, which are  exposed to both water
 and sediment  levels of arsenic (see Chapter 4.0—bioaccumulation).  A
 national monitoring survey indicated a mean of  2.6 mg/kg and a maximum
 of  700 mg/kg  (Western  Gulf)  in freshwater  and estuarine fish species
 (U.S. EPA 1980b).  In marine waters, fish muscle tissue, livers, and whole
 body  levels  ranged typically  from 2 mg/kg  to 5 mg/kg and crustaceans
 from  4 mg/kg  to 5 mg/kg  (wet weight—Hall et al. 1978).

     The FDA  (1977) has estimated that in 1974 the total dietary intake
 of arsenic in  food was 21 yg/day (as As203).  In 1973, it was estimated
 at 10 yg/day,  although the authors did not feel that this represented
a significant difference.  This average intake has decreased from an
estimated 63 yg/day, for the years 1965-1970, mainly due to a decline
in residues in fruits and vegetables.   The decline presumably resulted
from a reduction in use of  arsenical pesticides during this period.
Most of the arsenic intake  in 1974 (79%) was attributed to meat, poultrv,
and fish.  Seafood, especially bottom-feeding fish and crustaceans,
                                    5-32

-------
                                                              4
                SlLet o? "So ™/' r
              °
  assuming a maxet o? "     '         °St  e es">»«ed by
of fih.           ' °ny a V6ry  Sma11 ^Population consumes 100  g/da
                                                                     ay
                        gtf
 form was found at  levels of 1-420 ugA-  thf   *° ^    ^
 at levels of l-lio yg/i.  Total a«fnh     Pentavalent form was found
 PS/1, and about 50% of the sa^l«     ?  WaS/°und at leve^  of 1-530
 suggesting wine as a signfica^t sou^     J^  ^^ than 5° Ug/1> thus
 to the trivalent fom   B^h "In T      human exP°su^,
 acid were undetected ^t 1  S/T  Sf S^C ^
 taken up  by plants, £ 1^^^.^^^^
 The pentavalent form is expected to ^ ^    i  dyrlng  fe™entation.
                                '
          2:  LiLr -    «
srs:
                                                 .    not
consraption of usually fresh  fl,h    ?  •     ,        a"y reacti°n
            In          "
                                                      '«•
                               5-33

-------
                  TABLE  5-7.  ARSENIC LEVELS  IN  FOODS
Food Group
Meats, Eggs and Milk
Vegetables and Fruits
Cereal, Nuts and Sugar Products
Finfish and Shellfish
As203 Levels (mg/kg)
Mean	Maximum

0.01-0.03   0.5 - chicken

0.01-0.03   0.3 - potato products

0.01-0.04   0.4 - rice

0.07-1.47  19.1 - finfish
 Note:  Arsenic levels are reported as concentrations of
Source:   Jellinek and Corneliussen (1977).
                                   5-34

-------
                                                                      ("19781
               — ~*» ••**• *^*itiiu.  cunizp'nr*T*a^>iirtT^'P«Ti   •     •,               ^-i"//l-'/
      minutes.  There was no  in?ormat?on  on  theTS§ the/^in§ of f±sh for
    times or -higher t^emperature on' arsenic losses          1On8er heacin^- -

    5.2.2.3  Soil
                                                    soil
              in domestic  pesticide  P/estrictlons on use of certain
   ing .ore than 2.0% sodiuTarsenite  aldTs^ar 1977)«  Pr°ducts contain-
   longer registered for home use    There  is  Lu^ trioxide a™ no
   to arsenic in soil treated with the  or^ni       * potential for exposure
   cacodylic acid.   ApprcalmatSy 87 of IIT^T^™151   DMSA'  ^ and
   really (JRB 1980).   Most of the usf likelv^  P6Sti?ides  are ^ed dotnes-
            is expected to fae by
                                 .                                 treaty


  et al. 1975) would  ingest  appr^SSS0! 02 ^ °f S°U PSr da^ (^eP-
  ^n one episode.  The likelihood  of  such » L"8    3rSeniC' Presu^blv
  ing turf treatment  is expected to be  ouit^ f^T immediately foiled
  natural levels of arsenic  or much lower t^^'   I?gestion of "11 with
  « negligible exposure.  Grass con ce^ ,treatment  level«  would result
  levels (see Chapter 6.0) , p"babl? no ^T" T" be  1OWer than
  through accidental l-^lL^''  S°
  5-2-3   Inhalation
                                         3            —   « «'
 an assumed respiratorv  flw " S n?}°°3 us/m  l" "74 (Suta 1980).
                                   n}           l"       Suta 1980).  For"
           as  inorganic oxide or  rsenitl !
                                   fr
and cotton gins.  Milham  and  St^f^f11" conpounda, copper smelters,
of dust was as high as  1300 mg/kglithin 0 pP??ted that arsen" content
Washzngton shelter, while it  declined  to 70  "^ Ofothe Tac°ma'
                              ecxined  to 70 m
                 r, we it declined  to  70           o       c°ma'
tne source.                   ecxined  to  70 mg/kg  at  2.0-2.4 miles fron
                                  5-35

-------
                                              »
      Suta (1980) conducted an exposure analysis of the nonoccupational
 subpopulations exposed to airborne concentrations of arsenic from various
 sources both estimating exposure levels and the size of each subpopula-
 tion exposed.  Table_5-8^summarizes his conclusions. ~Suta estimated
 that upper limit atmospheric concentrations in "the" vicinity of"most
_emission_ sources_were approximately 0^ug/m3at locations where people
 would be exposed (e.g., ground-TeveT, outside'the plant)"." " For a 24-hour "
 period of exposure, this level would result in an exposure of 6 ug/day
 Copper smelters and cotton gins may expose subpopulations to higher
 levels, as much as 20 ug arsenic/day.  The chemical form of exposure is
 presumably arsenic trioxide,  commonly associated with thermal processes
 An PvrPnMnn  -f o ,.*,<, -,„_,•., form generated by cotton ginning, presumably
      Arsenic is  also present  in  cigarettes  (Lee  and Murphy 1969)   and
 smoking may result  in low levels of exposure.  In the  past,  the use of
 arsenic-containing  pesticides on tobacco  resulted in levels  of up to
 42  ug arsenic per cigarette  (Lisella et al.  1972),   Since  1952, however
 arsenicals  are no longer  used on tobacco, and  the arsenic  content in one
 cigarette is now less than 12 ug.   Lisella et  al.  estimated  that  15% of
 the arsenic present  would be  volatilized  (presumably as  trioxide)    If
 50  cigarettes are consumed per day  and all volatilized arsenic is inhaled
 the daily exposure  level  would equal 90 ug of  arsenic.                    '

      Exposure via inhalation  of  arsenical pesticides or  of dust or  vaoors
 from  arsenic-treated wood may occur  in small non-occupational  subpopula-
 tions „  The  RPAR position document  on inorganic arsenical pesticides
 estimated an exposure level of 270  ug of arsenic  per 8-hr day  from
 handling, sawing and  fabricating with arsenic-treated wood products
 (U.S.  EPA 1978).   Exposure from  living in a home  built with treated
wood was estimated at 0.7 ng  to  2 ug per day from inhalation of vapors.

 5.2.4   Dermal Contact

      Dermal  contact with  arsenic would be expected to be minimal, since
water concentrations  are  generally low.   However, Angino et al. (1970)
 pointed out  that dermal exposure may result from  the use of detergent
 containing arsenic.    These authors calculated that the highest  concen-
 tration of arsenic would  be found in a 1-gal presoak solution, which
might contain  20-250 mg/1 of arsenic.  Arsenic has been found at up to
 70-80 mg/kg  in phosphate-containing detergents (Pattison 1970).  The
 form present  is thought to be pentavalent.  At typical washwater con-
centrations, arsenic at the highest level  was estimated to be present
at approximately 0.15 mg/1.  Assuming a dermal absorption rate of 23.2
mg/hr (for arsenate  measured in a 15 g/1 solution; see Section 5.1.2.1),
high enough arsenic  concentrations for optimal uptake,  and an exposure
period of 10 minutes, the resulting uptake of arsenic would be approx-
imately 4 mg.  This  number is most likely  an overestimation of exposure
due to the 5 order of magnitude difference between arsenic concentrations
in washwater and the  concentration in the  solution in which the absorption
rate was measured.   Also assuming a washing machine or tub capacity of 60 1,
 the total amount of  arsenic available for  absorption would be approximately
9 mg.   The estimated  exposure level suggests that 45% of this amount is
absorbed in  the 10 minute period of exposure which is a probable overestimate.
Therefore, there is a high degree of uncertainty associated with this estimate,
                                5-36

-------
            TABLE  5-8.  SUMMARY OF MAGNITUDE OF HUMAN POPULATIONS EXPOSED VEA INHALATION OF
                       ARSENIC BY SELECTED EMISSION SOURCES
Average Annual
Concentration3
(|.g/m3)
3 .0-5.9
1.0-2.9
0.60-0.99
0.30-0.59
0.10-0.29
0.060-0.099
0.030-0.059
0.010-0.029
0.005-0.009
0.003-0.004

Copper
Smel ters


24,420
19,380
137,450
119,860
286,560
313,380
8,810
12,960
No. Persons
Lead
/•t
Smelters




800
2,600
5,100
38,000
46,000
67,000
Exposed by Emi
Zinc
Smelters




9,000
18,000
101,000
170,000
110,000
13,500
ssion Source
Cotton Pesticide
Glnse Manufacturer
2
30
70
230
660
990
1,900
6,600
18,000 380
44,000 1,100

Glass
Manufacturer^




1 440
10,140
75,180
212,040
363,870
583,360

 Average omnidirectional concentrations.   With  the  exception  of cotton gin exposures  24-hr worst-ca-e
 exposures can be estimated by multiplying the  annual  averages  by 12.5.   The 24-hr worst-case exposures
 tor cotton gins may be obtained  by  multiplying the concentrations by 81.
 Based on EPA's estimate of emissions.

^stimated'toT^lor f ?'Vb ? ^"^ ""  ^ tm  °f  lead Produced'   Fugitive emissions are
 estimated to be 10% of stack emissions.

 Based on an emission of 1.3 lb of arsenic per  ton  of  zinc  produced by pyrometallurgical  smelters and  no
 -stack emissions at electrolytic  smelters.  Fugitive emissions  assumed to  be 10%  of stack emissions.

"Annual average exposure,  assuming that ginning exposures occur during 15% of the year and that  there  are
 no exposures during the remainder of the  year.

 Assumes that 111 large plant  pesticide emissions are well  controlled.

'Assumes that 15% of pressed and  blown glass is manufactured with  arsenic,  that only certain  manufacturers
 r^nrtr10. ^         -e^ preSSed and blown  8lass Production, and  that  the size  distribution  of manu-
 facturers  who use arsenic  is  proportionate to  the  size  output.  It  is assumed that
 90%  of the manufacturers  are  well-controlled and that 10%  are  poorly-controlled.

 Source:   Suta (1980).

-------
      Additional exposure may occur from the handling of arsenic-treated
 wood (Fowler et al.  1979) or of grass clippings from arsenic-treated
 lawns.   Exposure levels associated with these activities are usually
 presumably low because the arsenic is not available in a solution in
 which it would be most likely to be absorbed dennally.  The RPAR position
 document on inorganic arsenical pesticides reported an estimated exposure
 level of 240 yg in an 8-hr day associated with handling, sawing and'fabri-
 cation  of arsenic-treated wood products (U.S.  EPA 1978).  This estimate
 assumes that sweating of skin during labor would solubilize the arsenic
 absorbed by the wood.

 5.2.5  Exposure Incidences as Indicated by Human Monitoring Data

      Elevated levels  of arsenic appear in human hair and nail following
 long-term exposure to the element;  however relating these accumulated
 levels  to effects levels has  not  been successful so far.   Metabolic
 studies indicated little tissue accumulation for most  forms of arsenic.
 The  effect  of arsenic  on the  body by the time  it is excreted in dead
 hair and nail matter  is unknown.   Relating exposure to urinary levels'
 from ingesj:ion^of_contaminated drinking water  was described in"5.2.2.

      Table  5-9 presents reported  data on arsenic levels  in human bio-
 logical media:   hair  and urine.   The  general population  and selected
 subpopulations exposed  to  specific  sources  are  represented.   The majority
 of the  exposure  incidents  resulting  in  bioaccumulation were  related to
 inhalation  of  airborne  arsenic.   Other  exposure  incidents  that  resulted
 in adverse  effects were  described  in  5.1.4.

 5.2.6.   Forms  of Arsenic Associated with Exposure Pathways

     A wide variety of chemical forms of arsenic  are found  in the*
 environmental media to which humans are exposed.  Since speciation is
 a significant  factor in arsenic toxicity  (see 5.1), it is ideal  to
 identify  the particular form(s) to which humans are exposed to  estimate
 the risk  associated with each pathway.  This report has focused primarily
 on those  forms thought to be most environmentally significant in terms
 of toxicity and abundance.  These include arsenate, arsenite, MSMA,
DSMA, cacodylic acid, unidentified organically bound arsenic in biota,
and arsine.  In this section, mention was made of the particular form(s)
 of arsenic known or suspected to be associated with each exposure path-
way,  whenever possible.  Table 5-10 summarizes, by form of arsenic,
 the exposure pathways in which each form may be present.  Both quanti-
 tative and non-quantitative observations are included whether or not
exposure  levels could be calculated based on data.  Arsenate and arsenite
are common forms in water but their significance in foods is generally
not known.  The organic arsenicals are present in treated crops and
 some water samples.  Organic-bound arsenic, usually bound to sulfhydryl
groups,  is found in plant and animal tissue.  Arsine is rarely found
and only under very anaerobic conditions.
                                  5-38

-------
                                                        TABLE 5-9.  ARSENIC IN HUMAN BIOLOGICAL MEDIA
Ul
 I
OJ
TJ_&SSUI2

HAIR
        HAIR
IIAIK
        HAIR
  Number
 of Cases

a)  44
b)  69
c)  14
d)   5
e)  25

a)  10
b)  93
c)  26
            13
                   220
        IIAIK      a) 23
                  b) 23
                  c) 44
        HAIR     a) 072
                 b) 160
                                            Tissue Arsenic Levels
                                         Range                   Mean
                                  a) >1 ppra
                                  b) <1 ppro
                                  c) >2 ppm
                                  d) >1 ppra i2 ppm
                                  e) -1 ppm
                                     NG
                          37-1,680 pg/g
                          a)  2.8-13  ppra
                          b)  0-1.5 ppra
                          c)  0.4-2.8 ppm

                          a)  0.6-8.1 |jg/g
                          b)  0-0.9   yg/g
                          c)  0-1.1 pg/g
                                NG
                                                          NG
                                                                    .*
                                                      a)  0.32 rog/lOOg
                                                      b)  0.61 mg/lOOg
                                                      c)  0.08 mg/lOOg
                                               NG
                                                                  NG
                                   a) 3.3 pg/g
                                   b) 0.3 Mg/B
                                   c) 0.2 pg/g
                                             a) 2.6 )ig/g
                                             b) 0.9 (ig/g
                                                                                         Source
                                                                                                                 Comments
92 people using wells    a) 17/44  As  poisoning
with high As; 21 people  b) 10/69  As  poisoning
As less than 0.05 ppm
                             Reference

                          Hlndmarsh  £t  al, (1977)
                                                                          Geological  contamina-
                                                                          tion  of well  water
Well water contaminated
with insecticide

Soy sauce contaminated
with arsenic (perhaps
calcium arsenate)

Air pollution from
coal-fired power plant
                                                               Stack emissions from
                                                               non-ferrous smelters
                                                                                                   c)  7/14 Electromyograph abnormal
                                                                                                   d)  3/5
                                                                                                   e)  0/25
a) arsenic area, normal   Borgono  et al. (1977)
     skin
b) arsenic area, abnormal
     pigment
c) nonarsenic area,
     normal skin

Arsenic In water 11,800-  Feinglass  (1973)
21,000 mg/Uter
a) near the root
b) at the tip
c) control

a) heavily polluted
    area
b) 36 km from source
c) controls

a) smelter area
b) control
                                                                                                                                    Mizuta  et. a\. (1956)
Bencko and Symon (l')77)
                                                  Baker  .et al. (1977)
         Ml:  data not given.
                                                                                                                                    continued.. .

-------
                                                    TABLE 5-9.  ARSENIC IN HUMAN BIOLOGICAL MEDIA
                                                                      (Continued)
                                             Tissue Arsenic Levels
Oi

Tissue
HAIR





HAIR



HAIR



HAIR

HAIR


HAIR



Number

of Cases
a)
b)
c)
d)
e)
f)
a)
b)






a)
b)
39
74
94
48
48
123
13
7


61



21
20
310


a)
b)



I
1

a)
b)
c)
d)
e)
f)
a)
b)






a)
b)
0.






Range
NG
NG
NG
20-1550 ug/lOOg
27-6014 ug/lOOg
34.8-363 Mg/100g
19-99
0-5 ppm


NG



0.3-6.1 ppm
0.08-0.18 ppm
08-15.15 mg/]00 g


NG




a)
b)
c)
d)
e)
f)
a)
b)


a)
b)
c)
d)
a)
b)







Mean
668.9 1 500.9 M8/100g
334 + 401 ug/lOOg
794.4 ± 1244.7 lig/100g
287.2 ug/lOOg
730.9 Mg/100g
NG
61 ppm
1 . 3 ppm


1-7 Mg/g
1.9 ug/g
49.5 Mg/g
0.3 pg/g
2.1 ppm
0.12 ppra
1 mg/100 g


77 ppro
5 ppm


Source
Air pollution





Stack emission from
copper smelter


Arsenic mining



Emissions from zinc
and copper mining
Geological contami-
nation of drinking
water
Contamination of well-
water by waste from
arsenic sulfide plant


a)
b)
c)
d)
e)
f)
a)

b)

a)
b)
c)
d)
a)
b)

•Comments
X.n sulfate process
Cd-Zn refinery
Cd-Zn refinery
Cd-Zn refinery
Cd-Zn refinery

Reference
Cabor and Coldea
(1977)



residents of 5 urban areas
children 300 yd from
source
children 8 miles from
source
roasters/smelters
ore dressers
miners
clerical
children near mine
urban controls
Normal value reported
0.

a)
b)

1 mg/100 g

no skin symptoms
advanced skin symp-
toms
Milliani and Strong
(1974)


Ishlniiilii, et al.
(1977)


Corridan (1974)

Zaldlvar and
Guilller (1977)

Terada (1960)
Tsuchiya (1977)

         URINE
                    164
                                   0.001-0.7 mg/lltcr  0.09  mg/liter
Geological contamina-
  tion of drinking
  water
Maximum normal           Zaldivar and
value reported 0.1 mg/     Culllier (1977)
liter
                                                                                                                                       continued...

-------
               Number
               f  Cases
                             Tissue
  WUNK
URINE
1JK1NE
            a) 25
            b) 20
            c) 44

Oi
JS
•—





"RINK a) 712
b) 193
"R'NE a) 49
b) 78
(JRfNE a) 19
b) 16
c) 7
d) 6
 a)  15
 b)  19

a)  41
b)  30
c)  23
d) 30
                   a)  0.5-11  ppb
                   b)  0.5-11  ppb
                   c)  8-39 ppb
                   <*)  3-17 ppb

                  a) O.OOJ-Q.105
                  b) 0-0.036 ,)g/
                  c) 0.001-0.044  |,g/8
                                    NG
                                    NG
                           a)  0.01-0.3 ppm
                           b)  0.01-0.2 ppra
                           c)  NG
                           d)  NG
a) 8-49 ug/day
b) 11.5-75.5 pg/day

         NG
                                                         .Mean
                           a) 0.025  Hg/g
                           b) 0.008  |ig/g
                           c) 0.011  pg/g
                                                    J) 18-7 ug/1
                                                    b) 5.8
                                          a) 0.021 ± 0.017 rag/
                                          b) 0.023 ± 0.032 mg,

                                          a) 0.097 ppm
                                          b) 0.034 ppm
                                          c) 0.30 ppn,
                                          d) 0.02 ppm
                                                   a) 20.9 ug/day
                                                   b) 39.1 ng/day
                                                                                   Source

                                                                            Wine contaminated
                                                                            with arsenic
  Air  pollution  from
  coal-ffred power plant
                                                 Stack emissions from
                                                 nonferrous smelters

                                                 Air pollution
                                                 Stack  emission  from
                                                 copper smelter
Urban areas
                                                  a)  1.3  Mg/l
                                                  b)  2.2  ug/l
                                                  c)  4.8  ,)g/i
                                                  <1)  8.6  pg/i
                                                  (geometric means for As+3)
                                                Copper smelter
                                                workers
   a)  As"1"3
   b)  As+5
   c)  DMAA
   d) MAA

  a) heavily polluted
       area
  b) 36 Km from source
  c) controls

  a) smV-lter are;i
  b) control

  a)  2u  sulfate  process
  b) Cd-An refinery

 a) children 300 yd
      from source
 b) children 8 miles
      from source
 c) 0-0.4 miles
 d) 2-2.4 miles

 a) Seattle,  1965
 b)  Denver,  1972


 a)  controls
 b)  low exposure
c) mediurn exposure
d) high exposure
                                                                                                        Reference

                                                                                                    Crecelius  (1975)
                                                                                                                              Bencko and Syi  \
                                                                                                                                (1977)
                                                                                                                              Baker  et  al.
                                                   fiabor and Coldea
                                                     (1977)

                                                   Mil ham and Strong
                                                     (1974)
                                                  Mathies  (1974)



                                                  Smith  et al.(197/)
                                                                                                                               continued....

-------
                                                   TABLE 5-9.   ARSENIC IN HUMAN BIOLOGICAL MEDIA
                                                                    (Continued)
      Tissue
      UKINIi
      URINE
01
 I
ro
      UK (Nil!
 Number
of Cases

    24
  a)
  b)
  c)
  a)
  e)
  f)
                    g) 5
                     NC
          TJssue Arsenic Levels
       Range                  Mean
                                    JD-539
a) 13-77 ug/24 hr
b) 104-256 pg/24 hr
=) 81-519 ng/24 hr
d) 8-212 ,ig/24 hr
e) 24-201 Mg/24 hr
f) 11-123 ng/24 hr
g) 4-89 |ig/24 hr

   0.022-0.025 mg/1
                                                            174 pg/1
a) 38 Mg/24 hr
i>) 172 |ig/24 hr
c) 204 ug/24 hr
d) 68 |,g/24 hr
e) 73 pg/24 hr
f) 42 ug/24 hr
g) 26 ug/24 hr

   0.023 mg/1
                                                       Source

                                               Copper smelter
                                                                                5 occupationally
                                                                                exposed Forest Service
                                                                                workers
                                                                                       NG
        Common t s

Mean airborne As
53 |ig/m3;  increases
associated with seafood
consumption

a) 1st week exposure
b) 4th week exposure
c) 6th week exposure
d) 8th week exposure
e) 10th week exposure
f) 2nd week control
g) 6th week control

   "Normal" freeze-
     dried
                                                                                                                                       Reference
                                                                                                                                   Pinto  et al.(1976)
Wagner  et al.
  (1974)
                                                                                                                                  Gills   et  al. O974)

-------
                                                         „ MSENIC
Ul
LO
         Surface  Drinking Water
                                     Predominant
                                       form
        Croimd Drinking Water
                          May be major
                          form in aerated
                          waters
        Ambient Water
                                  Usual predom-
                                  inant form
Food (general)
Fish
                                 Present  at
                                 low levels
       Atmosphere
      Cigarette Smoke
      Arsenic-treated
      Wood
  Negligible, epecial-
  ly under aerated
  conditions

  May be predominant
  form under reducing
  conditions

  Usually  present,
  sometimes  signifi-
  cant
                                                                            MSMA,  DSMA,
                                                                         Jgacodyl tc_ Acid
                                                         EXPECTE
                                                                  May be present at
                                                                  low levels
  May be present at
  low levels


  Usually present,
  sometimes  signi-
  ficant
                                                                            T 0
                        Predominant
                        form  (chromated
                        copper arsenate I
                        and- fluor-chromej
                        arsenate phenol)]

    NP means not thought to be present.
                                                  Present in some
                                                  fish tissues
                                                      NP
 Unknown


 Predominant  form


Predominant  form
   (As203)

Predominant form
   (As203)

     NP
                                                                                   B E
                                                                 Predominant solu-
                                                                 ble form

                                                                 May be present at
                                                                 trace  levels  in
                                                                 animals fed  feed
                                                                 additives

                                                                 Predominant
                                                                 form

                                                                Unknown
Unknown


Unknown


    NP
                                                                                      Organic-bound
                                                                                         Arsenic
May be present
in very reduced
conditions

-------
 5.2.7  Summary

      The typical and maximum arsenic levels associated with various
 human exposure pathways are summarized in Table 5-11.  Data are grouped
 by each main exposure route:  ingestion, inhalation and dermal absorption.
 The size of the subpopulations expected to be exposed to each pathway
 is estimated as best as possible.  Also included are the assumptions on
 which the estimated exposure levels are based and chemical form of
 arsenic.

      Most typical exposure levels equaled 20 yg arsenic per day or
 less for all routes.  Levels in excess of 100 yg/day were estimated
_^for ingestion of drinking water and food, working with treated wood
 and use of contaminated detergents.  Exposure to 4000 ug/day may occur
 with ingestion of ground water with arsenic levels equivalent to levels
 in areas with high background concentrations or in the vicinity of dis-
 posal sites.  Subpopulations consuming large amounts of contaminated
 fish may be exposed to very high levels of arsenic,  as much as 10,000
 yg/day.   Exposure through inhalation and dermal contact during construc-
 tion with treated wood had associated levels of 260  yg and 240 yg per day.
 Dermal exposure to highly contaminated detergent in  solution had an
 estimated uptake of 4000 yg, however, there is some  uncertainty regarding
 the applicability of the arsenic absorption rate at  the arsenic con-
 centrations present in washwater.   All exposures to  atmospheric arsenic
 had low exposure levels, less than 20 yg/day,  even in the vicinity of
 sources.   Approximately 90 yg/day was the exposure estimated for a
 heavy cigarette smoker.
                                   5-44

-------
                                             TABLE  5-11.   ESTIMATED  LEVELS OF HUMAN  EXPOSURE TO ARSENIC
Probable Exposure Subpopulacion
Route Form (ug/dav) Exposed
Typical Maximum
la««»cion - Drinking Water
Surface sources Arsenate 5 200 General population which
supplies; 117 x 10°
(Temple, Backer and
" Sloan 1977)

reduced water. Arsine has supplies: 75 x 1C6
been detected under very re- (Tenple, Barker and
ducao conditions Sloan 1977).
Food — total diet All forms; large part may be 21 190 General population:
organically bound arsenic 221 x 10° (USDC 1980) .
High fish consumption MSMA, possibly arsenite and 1000 10,000 ^Very small. ~~ "
others
Moonshine consumption Unknown 0.01 415 Very limited sub-
population
«n« consumption Arsenite predominately and 0.004 500 General population
arsenate
Soil ingestion Arsenacs or organic arsenicals 0.02 20 1/3-1/2 of children
up to age 3 (Mahaffey,
1978).
Assumotions
rvpical : Most levels (99.62 of o.tf.
2.5 ug/1; consumption of 2 1/dav.
Maximum: Maximum level in drinleinj-warer
Tvpicali Small sample of average ground-
sumption of 2 I/day. There ara many
incidences of higher ground water levels
(See text)
Maximum: Maximum levels of 2000 ug/1 in
naturally contaminated supplies, con-
sumption of 2 I/day.
Tvoical: FDA .Total Diet study estimate.
Maximum: Total diet with seafood
Typical : Fish or shellfish containing
10 mg/kg; consumption of 100 g fisn/
Maximum: Fish or shellfish containing
100 mg/kg As; consumption of 100 g fish/
dav
Typical: Moonshine containing 415 ug/1;
consumption of 28 ml/day.
Maximum: Moonshine containing a maximum
of 415 ug/1; consumption of 1 I/day.
Typical: Mae containing 100 ug/1;
consumption of 28 ml/day.
Maximum: Wine containing- maximum level*
of 500 ug/1; consumption of 1 1/dav
Typical: Soil containing 2100 ag/kg;
consumption of 10 mi soil/day.
Maximum; Soil containing 2100 ag/kg;
   General atmosphere
                                  Arsenic  trioxide
                                                                   0.06
                                                                                 0.6
                                                                                         General population:
                                                                                         221 x 10° (USOC 1980).
                                                                                                                    Typical: Average.ambient  concentration
                                                                                                                    of 0.003 ug/m3; respiratory  flow of
                                                                                                                    20 mVday.

Vicinity of industrial Arsenic trioxide

6 20 See Suta (1980) escia
by industrial source
cities (containing smelcers) of 0.03
aces Typical: Atmospheric concentration s£
0.3 ug/n3.
                                                                                                                   Maximum: Atmospheric concentration of
                                                                                                                   6 ug/mj. Respiratory flow of 20 m3/day.
                                                                                                                   May be higher exposure levels at some
   Living in house built
   of arsenic-treated wood
                                  Wnknom
                                                                  0.001 to 2     —
                                                                                         Very small subpopulation   Range:  See U.S. EPA (1978) RPAR
                                                                                                                    document on arsenical pesticides
                                                                                         Smokers: 54.1 x  10°
                                                                                         (Surgeon General 1979) .
                                                                                                                    Arsenic concentration of 12 ug/cigaretce
                                                                                                                    15Z volatilized;  consumption of 50
                                                                                                                    cigarettes/day
   ing with arsenic-
   created wood
                                 copper arsenate and fluor
                                 chrome arsenate phenol
                                 (see 3.3.2)
Very small subpopulacion   See U.S.  EPA (1973)  RPAR document on
                           arsenical pesticides.
Dermal Absorption
•   -Handling, sawing, build-
   ing with arsenic-created
   wood
                                                                               -40      Very small subpopulacion
                                                                                                                   See U.S. EPA (1978) RPAR document on
                                                                                                                   arsenical pesticides.   Assumes uptake
                                                                                                                   of 0.529  mg arsenic per square foot
                                                                                                                   of wood,  8 hour exposure,  damp
  Use of arsenic-con-
   taainated detergent.
                                                                               4000
                                                                                         Very small suopopulacion
                                                                                                                    Assuming  a aermal  absorption  rate of 23ag
                                                                                                                    hr  (based on  measurement  in solution of
                                                                                                                    15 g arsenic/1)  and exposure of  hands
                                                                                                                    for 10 minutes.  There is a high  degree
                                                                                                                    of uncertainty associated with this
                                                                                                                    estimate  (see Section 5.2.4).
Luaes  aosorpcion  efficiency  across  skin.
                                                                           5-45

-------
                                REFERENCES

 Amerine, M.A.; Singleton, V.C.  Wine:  An introduction.  2nd ed.
 Berkeley, CA:  University of California, Berkeley; 1977.

 Anand. S.J.  Mercury, arsenic and cadmium in the unfried and fried  fish.
 J. Radioanalytical Chem. 47:93-97, 1978.

 Angino, E.E.; Magnuson,  L.M.;  Waugh,  T.C.;  Galle, O.K.; Bredfelt, J.
 Arsenic in detergents:  Possible danger and pollution hazards.   Science
 168:389-390; 1970.

 Arguello, R.A.  et al. Cancer y arsenicismo  regional endemico en Cordoba.
 Rev.  Argentina Dermatosifilol.   22:461; 1938.   (As cited in USEPA 1980a).

 Ariyoshi, 1.; Ikeda,  T.   On the tissue distribution and the excretion
 of arsenic in rats and rabbits of administration with arsenical compounds.
 Jour.  Hyg.  Chem.  20:290; 1974.   (As cited in USEPA 1980a).

 Arthur D. Little, Inc.  Steel and the environment:  a, cost impact
 analysis for the American Steel Institute;  1975.
                                                i
 Arthur D.  Little,  Inc.   An  environmental partitioning model for risk
 assessment  of chemicals:  Narrative description of the methodology.
 Final  Draft.  Washington, DC:   Office  of Toxic  Substances,  U.S.  Environ-
 mental Protection  Agency; 1980.

 Baker,  E.L.;  Hayes, C.G.; Landrigan, P.J.; Handke,  J.L. , Heger,  R.T.;
 Houseworth,  W.J.;  Harrington, J.  M.  A nationwide survey of heavy  metal
 absorption  in children living near primary  copper,  lead, and zinc
 smelters.   Am.  J.  Epid 106:261-273; 1977.

 Barry,  K.G.;  Herdon,  E.G., Jr.  Electrocardiographic  changes associated
 with acute  arsenic poisoning.   Med. Ann  Washington,  D.C.   31:25-66;
 1962.   (As  cited  in USEPA 1980a,  Fowler 1977).

 Beaudoin, A.R.  Teratogenicity  of sodium arsenate in  rats.   Teratology
 10:153-157;  1974.  (As cited in NAS 1977).

 Beckman,  G.  et  al.  Chromosome  aberrations in workers  exposed to arsenic.
 Environm. Health Perspect  19:145; 1977.  (As cited in USEPA 1980a).

 Bencko,V. et  al.   Biotransformation of  As(III)  to As(V) and arsenic
 tolerance.   Arch.  Toxicol. 36:159; 1976.  (As cited in USEPA 1980s).

 Bencko, V., and Symon,-K.  Test of environmental  exposure to arsenic
 and hearing changes in exposed children.  Environ. Health Persp
 19:95-101; 1977.

Bergoglio, R.M.   Mortalidad por cancer ne zonas de aguas arsenicales
de la Provincia de Cordoba,  Republica Argentina.  Prensa Med. Argent
51:994-998; 1964.   (As cited in NAS 1977).

Boston Globe.  N.H. meeting  due on arsenic in wells.   Boston Sunday
Globe;  Feb.  15,  1981.

                             5-46

-------
  Borgono,  J.M.;  Greiber,  R.   Epidemiological study of arsenicism in the
  City of Antofagasta.   In:   D.D.  Hemphill (ed.),  Trace Substances in
  Environmental Health,  V. Proc.  of University of  Missouri's 5th Annu
  Conf.  Trace Subst.  Environ.  Health.   June 29-July 1, 1971.  University
  of  Missouri, Columbia; 1972.   (As cited in USEPA 1980a).

  Brady,  N.C.  The nature and  properties  of soils.   8th Edition,  New
  York: MacMillan Pufal.  Co. ,Inc.j  1974

  Braman, R.S.; Foreback, C.C.  Methylated forms of arsenic  in the environ-
  ment.   Science 183:1247-1249; 1973.

  Buchanan, W.D.  Toxicity of  arsenic  compounds.   Browning,  E. , ed.
  Elsevier monographs on toxic agents.  New York:   Elsevier  Publishing  Co  •
  1962.   (As  cited in USEPA 1980a,  Fowler,  1979).

  Butzengeiger, K.H.   tiber die chronische  Arsenvergiftung.   I.  Ekg-
 Veranderungen and andere Erscheinungen am Herzen  und Gefassystem.
  II.   Schleimhaut-Symptome und Pathogenese.  Deut. Arch. Klin. Med
 194:1-16;  1949.   (As cited in Klevay 1976, Fowler 1977, 1979).

 Butzengeiger, K.H.   Uber periphere Zinkulationssterungen bei chronischer
 Arsenvergiftung.   Klin. Wochenschr.  19:523-527;  1940.  (As cited in
 Fowler 1977, 1979).

 Byron,  W.R.; Bierbower, G.W.; Brouwer, J.B.; Hanson,  W. H.   Pathologic
 changes in rats  and  dogs  from two-year feeding of sodium arsenite or
 sodium arsenate.  Toxicol.  Appl.  Pharmacol.   10:132-147-  1967   (As
 cited in NIOSH 1973, NAS  1977).                               '

 Cannon,  A.B. Chronic arsenical  poisoning.  Symptoms  and  sources.  N.Y
 State J. Med. 36:219-241; 1936.   (As  cited in National Academy of Sciences


 Casto,  B.C.;  Meyers,  J.; DiPaulo,  J.A.   Enhancement of viral transforma-
 tion for evaluation  of  the  carcinogenic  or mutagenic  potential  of in-
 organic  metal salts.  Cancer  Research 39:193-198;  1979.

 Charbonneau,  S.M. et  al.  Pharmacokinetics and metabolism of inorganic
 arsenic  in the dog.  Trace Subst.  Environ. Health  12:276- 1978b    (As
 cited in USEPA 1980a).

 Charbonneau, S.M.; Spencer, K.; Bryce, F.; Sandi,  E.  Arsenic excretion
 by monkeys dosed with arsenic-containing  fish  and with  inorganic  arsenic
 Bull. Environ. Contam. Toxicol. 20:470-477; 1978.   (As  cited in USEPA
 1980a).

 Chhuttani,  P.N.;  Chawla, L.S.; Sharma, T.D.  Arsenical neuropathv.
Neurology 17:269-274; 1967.    (As cited in USEPA 1980a, Chhuttani'and
 Chopra 1979).

Chhuttani,  P.N.;  Chopra, J.S.   Arsenic poisoning.   Handb.  Clin. Neurol.
26:199—216; 1979.                  '
                                  5-47

-------
                   w
 Cikrt, M.;  Benko, V.  Fate of arsenic after parenteral administration
 to rats with particular reference to excretion via bile.  Jour. Hyg.
 Epidemic1.  Microbiol. Immunol.  18:129; 1974.  (As cited in USEPA 1980a).

 Clement, W.H.;  Faust, S.D.  New convenient method for determining
 arsenic (3+) in  natural waters.   Environ.  Lett. 5:155-164; 1973.

 Cone,  M.V.; Bauldauf, M.F.;  Martin,  F.M.;  Ensimger, J.T.  Chemicals
 identified  in human  biological  media, a data base.  Washington, DC:
 Office of Pesticides and Toxic  Substances, U.S.  Environmental
 Protection  Agency; 1980.

 Corridan, J.P.   Head hair samples as indicators  of environmental pollu-
 tion.   Environ.  Research:  8:12-16;  1974.   (As cited in Cone  et al
 1980).                                                         — —

 Coulson, E.J.;  Remington, R.E.   Metabolism in the rat of the naturally
 occurring arsenic of shrimp  as  compared with arsenic trioxide.  Jour.
 Nutr.  10:255-270;  1935.   (As cited in Fowler 1979, USEPA 1980a).

 Crecelius,  E.A.   Chemical changes in arsenic following ingestion by men.
 Presented at the Fifteenth Annual Hartford Life  Sciences Symposium.
 Richland, Washington;  1975  September 29-October 1.   (As cited in
 Foxier 1979).	

 Crecelius,  E.A.   Changes in  the  chemical speciation  of arsenic following
 ingestion by man.  Environ.  Health Persp.   19:147-150:  1977.

 Crecelius,  E.A.   Arsenite and arsenate levels  in wine.   Bull.  Environ
 Contain.  Toxicol.  18:227-230; 1977a.   (As  cited  in USEPA 1980a,
 Fowler  1979).

 Crecelius,  E. A.  Environ. Health Perspect.  19:147-150;  1977b.   (As
 cited  in Fowler  1979).

 Crecelius,  E. A.  The biological  implications  of  metals  in the environ-
 ment.   Proceedings of the  15th Annual Hartford Life Sciences  Symposium;
 1975 September 29-October  1, Richland, Washington.   (As  cited  in
 Fowler  1979).

 Curry,  A.S., ed.  Advances in Forensic and Clinical Toxicology.  Cleve-
 land; CRC Press.  185-190; 1972.  (As  cited  in Savory  and Sedor 1977).

 Ducoff, H.S.; Neal, W.B.; Straube, R.L.; Jacobson, L.O.; Brues, A.  M.
 Biological studies with  arsenic.  II.  Excretion  and tissue localization
 Proc. Soc. Exp.  Biol. Med. 69:548-554; 1948.   (As  cited in Fowler 1979
 USEPA 1980,  Federal Register 1978).

Dustin, A.P.  Bull.  Acad. Med.  Belg.  13:585; 1933.  (As cited  in
Oehme 1972).

Dutkiewicz,  T.   Experimental studies  on arsenic absorption routes in
rats. Environ.   Health Perspect.  19:173-177; 1977.
                                  5-48

-------
 Eyre, S.R.  Vegetation and  soils.   Chicago,  111.:   Aldine  Publ.  Companv;
 1968.  328 p.                                                          J

 Fairhall, L.T.; Miller, J.W.  A  study of  the relative  toxicity  of  the
 molecular components of lead arsenate.  Public Health  Rep. Wash.   56:
 1610-1625; 1941.   (As cited in Pelfrene 1976).

 Feinglass, E.J.  Arsenic intoxication from well water  in the United
 States.  New Eng. Jour. Med.  188:828; 1973.

 Ferm, V.H.  Arsenic as a teratogenic agent.  Environ.  Health Perspect
 19:215-217; 1977.

 Ferm, V.H.; Sakon, A.; Smith, B.M.  The teratogenic profile of sodium
 arsenate in the golden hamster.   Arch. Environ. Health 22:557-560;
 1971.  (As cited in NAS 1977).

 Fierz, U.  Katamnestische Untersuchiengen uber die Nebenwirkungen der
 Therapie mit anaorganischem Arsen bei Hautkrankheiten.   Dermatologica
 131:141;  1965.   (As cited in USEPA 1980a).

 Fleisher, M.   Cycling and control of metals.   Proceedings of an environ-
 mental resources conference.  Cincinnati,  OH:  National Environmental
 Research Center, 1973.   (As  cited in Woolson 1975).

 Flessel,  C.P.   Metals as  mutagens.   Adv.  Exp. Med.  Biol.  91:117-128;
 -L .7 / / •

 Food  and  Drug Administration (FDA).   Compliance program evaluation:
 FY  total  diet studies.  Washington,  DC:  Bureau of  Foods, Food  and
 Drug  Administration;  1977.

 Fowler, B.A.  Toxicology  of  environmental  arsenic.   Gayer,  R.A.;  Mehlman,
 M.A.  eds.  Toxicology of  trace elements.   New York:  Halstead Press-
 1977:  79-122.

 Fowler, B.A.; Ishinishi, N.; Tsuchiya, K.; Vahter, M.   Arsenic.   Friberg,
 L.; Nordberg, G.F.; Vouk, V.B. eds.  Handbook on the toxicology  of  metals.
 New York:  Elsevier/North Holland Biomedical  Press;  1979:   293-319.

 Franke, K.W.; Moxon, A.L.; Poley, W.E.; Tully, W.C.  Monstrosities
 produced by the  injection of selenium salts into hen's  eggs,  ^nat
 Rec.   65:15-22;  1936.  (As cited  in NAS 1977).

 Franklin, M.; Bean, W.B.;  Hardin, W.C.  Fowler's solution as an etiologic
 agent  in cirrhosis.  Am. J. Med.   Sci. 219:589-596; 1950.  (As cited in
 Fowler 1977).

 Frejaville, J.P.; Bescol,  J.; LeClerc, J.P.; Guillam, L.; Crabie, P.-
 Consco, F.; Gervais, P.; Fauttier, M.  Intoxication aigue par les
derives arsenicaux; (a propos de 4 observations personelles);  troubles
 de 1 hematose,  etude ultramicroscope deu foie et du rein.   ^nn  Med
 Intern.  123:713-722;  1972.   (As cited in Fowler 1977).


                                  5-49

-------
 Frost,  D.  V.   Arsenicals  in biology—retrospect and prospect.  Fed.
 Proc.  26:194;  1976.   (As  cited in USEPA 1979).

 Frost,  D.V.;  Perdue,  H.S.;  Main,  B.T.;  Kolar,  J.A.; Smith, I.D.; Stein,
 R.J.;  Overby,  L.R.   Further considerations on  the safety of arsenalic
 acid for feed use.   Section papers, Proceedings,  12th world's poultry
 congress;  1962,  Sydney, Australia;  1962:234-237.   (As cited in NAS 1977).

 Gabor,  S.; Coldea,  V.   Some aspects of  the environmental exposure to
 arsenic in Romania.   Environ.  Health Pers.   19:107; 1977.

 Geraghty, J.J.; Miller, D.M. ; Van  der Leeden, F. ;  Troise,  F.L.  Water
 atlas  of the  United States.  Port Washington,  New York:   Water Infor-
 mation Center, Inc.;  1973.

Gerhardt, R.E.; Crecelius, E.A.; Hudson, J.B.  Trace  element  content
of moonshine.   Archives Environ. Health 35(6):332-334; 1980.

Geyer, L.  Uber die chronischen Hautveranderungen beim Arsenicismus
und Bertrachtungen uber die Massenerkrankungen in Reichenstein  in
Schlesien.  Arch. Derm. Syphilol.    43:221; 1898.

Gills, I.E. et al.  Determination of toxic trace elements  in  body  fluid
reference samples.  Proc.  Univ. Mo. Ann. Conf.  Trace  Subst. Environ.
Health 8:273-280, 1974.  (As cited  in Cone  et a^L.  1980).

Ginsburg, J.M.  Renal mechanism for excretion and transformation of
arsenic  in the dog.   Am. Jour.  Physiol.  208:832-840;  1965.  (As cited
in USEPA 1980a, Klevay 1976).

Glazener, F.S.; Ellis, J.G.; Johnson, P.K.  Electrocardiographic findings
with arsenic poisoning.  Calif. Med. 109:158-162: 1968.  (As  cited in
Fowler 1977).
             •
Goldsmith, S.; From, A.H.L.  Arsenic-induced atypical ventricular
tachycardia.  New Engl. J. Med. 303(20):1096-1098; 1980.

Goldsmith, J.R.;  Denae, M.; Thorn,  J.; Gentry, G.  Evaluation  of health
implications of elevated arsenic in well water.  Water Research 6:
1133-1136; 1972.

Greathouse, D.G.; Craun, G.F.  Cardiovascular disease study—occurrence
of inorganics  in household tap water and relationships to cardiovascular
mortality rates.   Environ. Health XII; 1979.

Hall, R.A.; Took, E.G.; Meaburn, G.M.  National marine fisheries service
survey of trace elements in the fishery resource.   NOAA Technical Report
WMFS SSRF-721; 1978.

Hamamota, E.  Infant arsenic poisoning by powdered milk.   Jap. Med.
Jour. 1649:2;  1955.   (As cited in USEPA 1980a).

Harvey, S.C.  Heavy metals.  Goodman, L.S.; Gilman,  A. eds.  The pharma-
cological basis of therapeutics.  New York:  Macmillan Company; 1970:
958-965.

                                  5-50

-------
 Harvey, S.C.  Heavy metals.  Goodman, L.S.; Oilman, A.  Eds.  The
 pharmacological basis of therapeutics, 5th ed.  New York:  Macmillan
 Pub. Co., Inc.; 1975:924-928.   (As cited in USEPA 1978).

 Heyman, A.; Pheiffer, J.B., Jr.; Willett, R.W.; Taylor, H.M.  Peripheral
 neuropathy caused by arsenicarintoxication.  A'study of 41 cases with
 observations on the" effects of ~BAL"(2~,3,~dimercapto-propanbl) ." New Engl.
 J. Med.  254:401-409; 1956.  (As cited in USEPA 1980a,'Klevay 1976,
 NAS 1977).

 Hindmarsh, J.T., et al.   Electromyographic abnormalities in chronic
 environental arsenicalism.   Jour. Anal.  Toxicol. 1:270;  1977.   (As cited
 in USEPA 1980a).

 Hocken, A.G.;  Bradshaw,  G.   Arsenic poisoning.  Br. J. Ind. Med.  27-
 56-60;  1970.   (As cited  in  Fowler 1977).

 Holland, J.W.   Arsenic.   Perterson, F,;  Haines. W.S.  eds.   A textbook
 of legal medicine and toxicology.  Vol.  2 Philadelphia:   W.B.  Saunders
 and Company;  1904-404-415.   (As cited in NAS  1977).

 Holleman,  J.W.;  Hammons,  A.S.   Levels of  chemical  contaminants in non-
 occupationally  exposed U.S.  residents.   EPA 600/1-80-002.   Research
 Triangle Park,  NC:   Office  of  Research and  Development,  U.S. Environ-
 mental  Protection  Agency; 1980.

 Hood, R.D.; Bishop,  S.L.  Teratogenic effects  of sodium  arsenate  in
 mice.   Arch. Environ.  Health 24:62-65; 1972.   (As  cited  in  USEPA  1980a,
 NAS 1977).

 Hood, R.D.; Thacker, G.T.; Patterson, B.L.  Effects in the  mouse  and
 rat of  prenatal  exposure to  arsenic.  Environ.  Health  Persepct.   19:219;


 Hueper,  W.C.; Payne, W.W.  Experimental studies  in  metal carcinogenesis.
 Chromium, nickel, iron, arsenic.  Arch. Environ. Health 5:445-462-  1962*
 (As  cited in USEPA 1980a, Pelfrene  1976).

Hunter,  F.T.; Kip, A.F.;  Irvine, J.W.  Radioactive  tracer studies on
arsenic  injected as potassium arsenite.   Jour. Pharmacol. Exp.   76-207-
220; 1942.  (As cited in USEPA  1980a,NAS 1977, Klevay 1976, Fowler'l979).

 Irgolic, D. Speciation~of arsenic in water supplies.  Q.  Pro<*  Reo
Nov. 1,   1978-Jan. "31; 1979.

Ishinishi, N.  Kelama, Y. ;  Nobutorno, K.;  Inamausu,  T.;  Kunitake,  E.;
Suenaga, Y.  Outbreak of  chronic arsenic  poisoning among  retired workers
from an  arsenic  mine in Japan.   Environ.  Health pers.   19-121-  1977
(As cited in Cone  _et al.  1980) .
                                  5-51

-------
James,  L.F.; Lazar, V.A.;  Binns, W.   Effect  of  sublethal  doses  of
certain minerals  on pregnant  ewes  and fetal  development.   Amer.  J.
Vet. Res. 27:132-135;  1966.   (As cited in NAS 1977).

Jellinek, C.F.; Cornelieussen, P.E.   Levels of arsenic in  the United
States  food supply.  Environ. Health  Perspec.   19:83-87;  1977.   (As
cited by Phillips and  Russo 1978).

Jenkins, R.B.  Inorganic arsenic and  the nervous system.   Brain  89:479-
498; 1966.  (As cited  in Chhuttani and Chopra, Klevay 1976, USEPA 1980a).

Josephson, C.J.; Pinto, S.S.; Petronella, S.J.  Arsine:   electrocardio-
graphic changes produced in acute human poisoning.  Arch.  Ind. Hyg.
Occup.  Med. 4:43-52; 1951.  (As cited in Fowler 1977).

JR3, Inc.  Level II materials balance:  Arsenic.  Draft report.   Con-
tract No. 68-01-5793.  Washington, DC:  Office of Pesticides and  Toxic
Substances, U.S. Environmental Protection Agency; 1980.

Kadowaki, K.  Studies  on the arsenic  contents in organ tissues of the
normal  Japanese.  Osaka City Med. Jour.  9:2083; 1960.  (In Japanese
with English summary.)  (As cited in  USEPA 1980a, Fowler 1979).

Kagey,  B.T. e_t al.  Arsenic levels in maternal-fetal tissue sets.  Trace
Subst.  Environ. Health 11:252; 1977.   (As cited in USEPA 1980a).

Kent, G. ' Data on arsenic in drinking water supplies.  Washington, DC:
U.S. Environmental Protection Agency;  1976.   Unpublished data.   (As
cited in Suta 1980).

Klaasen, C.D.   Biliary excretion of^arsenic in rats, rabbits, and dogs.
Toxicol. Appl.  Pharmacol.   19:447;  1974.  (As cited in USEPA 1980a).

Klaasen, C.D.   Heavy metals and heavy-metal antagonists.   Oilman, A.G.;
Goodman, L.S.;  Gilman, A.  eds.  The pharmacological basis of therapeutics.
New York:  Macmillan Publishing Co., Inc.; 1980:1629-1632.

Klevay, L.M.  Pharmacology and toxicology of heavy metals:  arsenic.
Pharm.  Ther. Vol.  1.   Great Britain:  Pergamon Press; 1976:189-209.

Kroes, R.;  Van Logten, M.J.;  Berkvens, J.M.;  de Vries, T.; Van Esch,
G.J.  Study of the carcinogenicity of lead arsenate and sodium arsenate
and on the possible synergistic effect of diethvlnitrosamine.   Food
Cosmet. Toxicol.  12:671-679;  1974.   (As cited in NAS 1977).

Kyle, R.A.; Pease, G.L.  Hematologic aspects  of arsenic intoxication.
New Engl. J. Med.   273:18-23;  1965.   (As cited in Klevay 1976, Fowler
1977).

Lakso,  J.U.; Peoples,  S.A.   Methylation of inorganic arsenic by mammals.
Jour. Agric. Food Chem.  23:674;  1975.  (As  cited in USEPA 1980a).
                                  5-52

-------
  Lander,  Hf;  Hodge,  P.R.;  Crisp,  C.S.   J.  Forensic Med.   12:52-67;  1965
  (As  cited  in Fowler 1979).

  Lanz, H.,  Jr.  et  al.   The metabolism  of  arsenic in laboratory animals
  using As H as  a tracer.   Univ. Calif.  Publs.  Pharmacol.   2:263:1950
  (As  cited  in USEPA  1980a).

  Lee, B.K.; Murphy,  G.  Determination  of  arsenic content  of  American
  cigarettes by neutron  activation analysis.  Cancer 23(6)-1315-1317-
  1969.

  Lepow, M.L.; Brickman, L,; Tillette, M.; Markowitz, S.;  Robine, R.;
  Kapesh,  J.    Investigation into the sources of lead  in the environment
  of urban children.  Environmental Research 10:415-426; 1975.

  LeQuesne,  P.M.; McLeaod, J.G.  Peripheral neuropathy following a single
  exposure to  arsenic.  Jour.  Neurol. Sci.   32:437; 1979.   (As cited  in
 USEPA 1980a).

 Limarzi, L.R.  The effect of arsenic (Fowler's solution) on erythropoiesis
 Am. J.  Med. Sci.  206:339-347; 1943.  (As  cited in Pelfrene 1976, Klevay
 xy/D) •

 Lisella, F.S.;  Long, K.R.; Scott, H.G. Health aspects of arsenicals
 in the  environment.   J. Environ.  Health.   34(5):511-518;  1972.

 Lu, F.J., et  _al.   Physico-chemical  characteristics  of drinking water
 in Blackfoot  endemic areas of Chia-T and  Tainan  Hsiens.   Jour.  Formosan
 Med.  Assoc. 74:596;  1975.   (As cited in USEPA  1980

 Lu,  F.F., _er  al.   Studies  on  fluorescent  compound in drinking  water of
 Blackfoot endemic  areas.   I.   The toxic effect of fluorescent  compound
 on the chick  embryos.   Jour.  Formosan  Med. Assoc.   76:58- 1977a   (As
 cited in  USEPA  1980a).                                   '      '   v

 Lu,  F.J., et  al.   Studies on fluorescent compounds  in drinking  water
 of Blackfoot  endemic areas.   It.  Isolation and  identification  of
 fluorescent compounds.  Jour.  Formosan  Med. Assoc.   76-209-  1977h
 (As cited in  USEPA 1980a) .                                   ^"o-

 Luchtrath,  H.  Liver  cirrhosis in chronic arsenic  intoxication  of
 vintagers.  Dtsch.  Med. Wochenschr.  97:21-22; 1972.  (As  cited in
 Fowler 1977,  1979)  .

 Lunde, G. Occurrence and transformation of arsenic in the marine environ-
ment.  Environ. Health Persp.  19:47-52; 1977.

Mahaffey, K.R.  Dietary lead intake of  children living in Washington, D.C.
Unpublished data.   [1976],  (As cited in Mahaffey 1978).

Mahaffey, K.R.  Quantities of  lead producing health effects in humans:
sources and  bioavailability.   Environ.  Health Persp. 19:285-295; 1977.


                                   5-53

-------
 Mappes, R.  Experiments on excretion of arsenic in urine.  Viersuche
 zur Ausscheidung von Arsen in Urin.  Int. Arch. Occup. Environ. Health
 40:267-272; 1977.  (As cited in USEPA 1980a).

 Masahiki, 0.;  Hideyasu, A.  Epidemiological studies on the Morinaga
 powdered milk poisoning incident:  final report of the joint project
 team from Hiroshima and Okayama Universities for survey of the Senoi
 area.   Jap. Jour. Hyg.  27:500;  1973.   (As cited in USEPA 1980a).

 Mathies, J.C.   X-ray spectrographic microanalysis of human urine for
 arsenic.  Appl.  Spectrose.  28:165-170,  1974.   (As cited in Cone et al
 1980).                                                            	

 Mealey, J. , Jr.;  Brownell, G.L.;  Sweet,  W.H.  Radioarsenic in plasma,
 urine,  normal  tissues and intracranial neoplasms.   Arch.  Neurol.
 Psychiatr.  81:310-320;  1959.   (As cited  in Fowler 1979, USEPA 1980a
 USEPA 1978).

 Milham, S.; Strong,  T.   Human arsenic exposure in relation to a copper
 smelter.   Environ.  Res.  7:176;  1974.

 Mizuta, N., et  al.   An  outbreak of acute arsenic  poisoning caused by
 arsenic contaminated soy sauce  (shoyu):  A clinical report  of  220  cases.
 Bull. Yamaguchi Med.  Sch.  4:l3l;1956.  (As cited  in USEPA  1980a).

 Morton, W. ,  Starr,  G.;  Pohl, 0.;  Stoner,  J.; Wagner,  S.; Weswig,  P.
 Skin cancer and water arsenic in  Lane County,  Oregon.   Cancer ° 37-
 2523; 1976.

 Muehrcke, R.C.; Lucas,  G.D.; Pirani, C.L.   Arsine  induced  anuria.
 Presbyterian St.  Luke Hospital Med. Bull.   2:2-10;  1963.   (As cited  in
 Klevay  1976).

 Munro,  I.C. et al.  Biological availability  of  arsenic  from fish.
 Toxicol. Appl. Pharmacol.  29:111;  1974.   (Abstract  #92).   (As cited
 in USEPA 1980a).

 Mushak, P.  Advances  in the analysis of heavy toxic elements  having
 variable chemical forms.   J. Anal. Toxicol.  1:286;  1977.   (As cited
 in USEPA 1980a).

 Mushak, P.  et al.  Health  assessment document for arsenic.  Research
 Triangle Park, NC.  Environmental Criteria and Assessment Office, U.S.
 Environmental Protection Agency; 1980.

 National Academy of Sciences (NAS).  Water quality criteria 1972.   EPA
 Ecol. Res. Ser. EPA-R3-73-033.   Washington, D.C.:   U.S. Environmental
 Protection Agency 1973.   (As cited in Phillips  and Russo 1978) .

NAS (National Academy of Sciences).  Medical and biologic effects of
 environmental pollutants:  arsenic.  Washington, DC:  National Research
 Council, Division of Medical Sciences,  Assembly of Life Sciences;  1977.
 332 p.

                                  5-54

-------
 National__Ins_titute__for  Occupational  Safety  and  Health  (NIOSH) .   Criteria
 for a  recommended  standard...Occupational exposure  to "inorganic arsenic.
 Washington, DC:  U.S. Department  of  Health,  Education  and  Welfare,  Public
 Health Service, Center  for  Disease Control,  National Institute  for  Occupa-
 tional Safety and  Health; 1973.   105 p.

 National Research  Council (NRC).  Drinking water  and health.  Part  I.
 Chap.  1-5.  A report of the  safe  drinking water committee.  Washington,
 DC:  Office of Water Supply, U.S. Environmental Protection Agency;  1977.

 Neuwirtova, R.; Chytil, M.; Valek, A.; Daum, S.;  Valach, V.  Acute  renal
 failure following  an occupational intoxication  with arsine (AsH3) treated
 by the artificial  kidney.  Acta Med. Scand.  170:535-546; 1961.   (\s
 cited  in Klevay 1976, Fowler 1977).

 Nielson,  B. ed.   Arsine poisoning in a metal refining plant:  fourteen
 simultaneous cases.  Acta Med.  Scand. Suppl. 496:1-31;  1968.  (As cited
 in Fowler 1977).

 Nishioka,  H.  Mutagenic activities of metal compounds in bacteria
 Mutation  Res.   31:185-189;  1975.

 Odanaka,  Y. _et al.   Identification of dimethylated arsenic by gas
 chromatography-mass spectrometry in  blood,  urine and feces of rats
 treated with ferric methanearsonate.   Jour.  Agric. Food Chem.   26-505-
 1978.   (As  cited  in USEPA 1980a).

 Oehme,  F.W.  Mechanisms  of  heavy metal  toxicities.  Clinical Toxicology
 5(2):151-167;  1972.                                                  gy

 Ohta, M.  Ultrestructure of  sural nerve  in a case  of arsenical neuropathy
 Acta Neuropathol. 16:233-242; 1970.   (As  cited  in  USEPA 1980a).

 Okamura, K.  _et al. Symposium  on arsenic poisoning  by  powdered  milk
 (2).  Diagnos. Ther.  9:240;  1956.   (Jap.)  (As  cited in USEPA 1980a).

 Osswald, H.; Goerttler,  K.   Leukosen  bei  der  maus  nach  diaplacentarer
 und postnataler Arsenik-Applikation.  Dtsch.  Gesamte Path.  55-280-293-
 1971.   (As  cited in NAS  1977).

 Overby, L.F.; Fredrickson, R.L.  Metabolic stability of  radioactive
 arsenilic acid in chickens.  Jour. Agric. Food Chem.  21-378- 1963
 (As cited in USEPA  1980a).                                   '

Paton,  G.R.; Allison, A.C.  Chromosome damage in human cell cultures
induced by metal salts.   Mutation Res.  16:332-336; 1972.   (As cited
in NAS  1977).

Pattison,  E.S.  Arsenic and water pollution hazard.  Science 170:870:
-L> /U•
                                  5-55

-------
Patty, F.A.  The mode  of  entry and  action  of  toxic materials.   Patty,
F.A.  ed.   Industrial hygiene  and  toxicology.  New York:   Interscience
Publishers;  1948.  p.  175.   (As cited  in USEPA  1980a).

Pelfrene,  A.  Arsenic  and cancer:   the still  unanswered  question.   Jour.
of Toxicol.  and Envrion.  Health 1:1003-1016;  1976.

Peoples, S.A.  Review  of  arsenical  pesticides.  Woolson,  E.A.  ed.
ACS Symp.  Sec. 7.  Arsenical  pesticides.   Washington, DC:  Am.  Chem.
Socl; 1975:1-11.

Petres, J.;  Berger, A.  Zum Einfluss anaoganischen Arsens auf  die  DNS-
Synthese menschlicher  Lymphocyten in vitro.   Arch. Derm.  Forsch.   242:
343-352; 1972.  (As cited in  USEPA~1980a).

Petres, J.; Hundeiker, M.  "Chromosomenpulverisation" nach Arseneinwirkung
auf Zellkulturen in vitro.  Arch. Klin. Exp.  Dermatol.   231:366-370;
1968.  (As cited in USEPA 1980a).

Petres, J.;  Schmid-Ullrich, K.; Wolf,  W.   Chromosomenabberrationen an
menschlichen Lymphozyten  bei  chronischen Arsenchaden.  Dtsch. Med.
Wochenschr.  95:79-80;  1970.   (As cited in  USEPA 1980a).

Pinto, S.S.; McGill, C.M.  Arsenic  trioxide exposure in  industry.   Ind.
Med. Surg. 22:281-287; 1953.  (As cited in Fowler 1979).

Pinto, S.S., et al.  Arsenic  trioxide  absorption and excretion  in  indus-
try.  Jour. Occup.  Med. 18:677; 1976.  (As cited in USEPA 1980a).

Phillips, G.F.; Russo, R.C.  Metal bioaccumulation in fishes and aquatic
invertebrates:  A literature  review.   EPA  600/3-78-103.   Duluth, MI:
Office of Research and Development, U.S. Environmental Protection  Agencv;
1978.

Ray-Bettley, F.; O'Shea, J.A.  The absorption of arsenic and its relation
to carcinoma.  Br.  Jour. Dermatol. 92:563-568; 1975.   (As cited in  Fowler
1979, USEPA 1980a).

Reynolds, E.S.  An account of the epidemic outbreak of arsenical poison-
ing occurring in beer-drinkers in the north of England and Midland
counties in 1900.   Lancet 1:166-170; 1901.   (As cited in NAS 1977).

Ridgeway, L.P.; Karnofsky, D.A.   The effects of metals on the chick
embryo:   toxicity and production of abnormalities in development.  Ann.
N.Y. Acad.  Sci. 55:203-215; 1952.   (As cited in NAS 1977).

Rossman,  T.; Meyn,  S.;  Troll, W.   Mutation Res.  30:157;  1975.   (As
cited in Flessel 1977).
                                  5-56

-------
                                                    s-s-
HoUand Biomedical Press  1977:2n!288


Schrauzer,  G.N. ; Ishmael, D.   Effects

genesis of  spontaneous «ary "^s
Lab.  Sd. 4:441-447;  1974.   (L  cSS
                                                         E1'«ier/Sorth
                                          =„!   •

                                          inbr^Tw ndarsenlc  <"

                                          pSr1na
                                    n
  IJ.10..  and Hfe span.   ,.         "



                            PnDE                    ,
"79.   CAS reproduced in MusST?.' et al   He'SthT1 PrOt""°"
for arsenic). Research Triangle Park  vF?' v       Assessment document
Assessment Office, U.S.  Envifon^tai P^tectlo^^cy;
        roKicol.  21:37!, 197,
       U.S. Environ.enta!
                                                   Biochem. 806:505;



                                        oP««iBg and financiai c.arac-
                                                     .  ,lkon
                                5-57

-------
 Tseng, W.  Effects  and  dose response  relationship  of  skin cancer and
 Blackfoot disease with  arsenic.  Environ. Health Perspect.   19:109-119;


 Tseng, W.P. et al.  Prevalence of skin cancer-in an endemic  area of
 chronic arsenicism  in Taiwan.  Jour. Natl. Cancer  Inst. 40-453'  1968
 (As cited in NAS 1977)

 Tsuchiya, K.  Various effects of arsenic in Japan  depending  on type  of
 exposure.  Environ. Health Pers. 19:35; 1979.

 Uldall, P.R.; Khan, H.A.; Ennis, J.E.; McCallum, R.I.; Crimson,  T.A.
 Renal damamge from  industrial arsine poisoining.   Br. J.  Ind. Med. 27-
 372-377;  1970.  (As cited in Fowler 1977)

 Urakabo,  G.  et al.   Studies on the fate of poisonous metals  in experi-
 mental animals (V).  Body retention and excretion  of arsenic.  Jour.
 Food Hyg. Soc. Jap. 16:334; 1975 (Jap.) (As cited  in USEPA 1980a). .

 U.S. Department of  Commerce (USDC).  Statistical abstract of the United
 States.  Washington, D.C.:  Bureau of the Census,  U.S. Department of
 Commerce, 1980.

 U.S. Department of Health, Education and Welfare (U.S. DHEW).  Community
 water supply s_tudy.  Washington, DC:  U.S.  Department of Health,  Educa-
 tion and  Welfare;  1970.                  	-...-.     _


_u! A-^Environmental' Protect ion" Agency (USEPA)":  Chemical analysis  of   ""
 interstate water supply  systems.   EPA 430/9-75-005  1975.   (As cited in
 U.S. EPA  1980a)

 U.S. Environmental  Protection  Agency (USEPA).  Wood preservative pesti-
 cides.  Federal  Register 43(202):   48300-48422;  1978.

 U.S. Environmental  Protection  Agency (USEPA).  Ambient water quality
 criteria  for  Arsenic.  Washington,  DC:   Office of Water Planning and
 Standards, U.S.  Environmental  Protection Agency; 1980a.

 U.S. Environmental  Protection  Agency (USEPA), STORET.   Washington, DC:
 Monitoring and Data Support Division,  Office  of Water  Regulations and
 Standards, USEPA; 1980b.


 U.S. Environmental Protection Agency (U.S. EPA) Level  II Materials
 Balance:  Arsenic.   Draft  report to  the Office of Pesticides  and
 Toxic Substances, Contract No. 68-01-5793; 1980 c.

 U.S. Public Health  Service (USPHS).  High environmental  exposure  to
 ^e^c~[a1irbanks'  ^as^;  Atlanta,  GA:  Center for  Disease Control;
 EPI  76-111-1  (unpubl.);  1977.
                                 5-58

-------

         (As cited in Cone «     WSOK          EnVlr°n' Health
  Can.
                                                      ""
                              s
  USEPA 1980)                   SU"M'T-   (As cited in Fowler 1978,
                                       .
 45 = 80; 1962.  (As citid in Ss?p! I            °'
                            of arsenical,.  Acs Synp. Series 7:97-
 (Jap.) (AS clte<3  la us             "'  JOUr' Hyg- 27:362: "72-
                               er
    cited in USEPA 1980s).     cancer-  Natl.  Cancer Inst. 10:81; 1963.

         ^e s^^r^^Ls       chronic
Public Health Assoc. MiaJ Beach  floral
in USEP                  seacn, Florida;
                                or     i,7,
in USEPA 1980)                ,    oria;  1976a, p.  n2.  (As cited
                             5-59

-------
Zaldivar, R.; Guillier, A. Environmental and clinical investigations
on endemic chronic arsenic poisoning in infants and children.  Zbl.
Bakt. Hyg., I. Abt. Orig. B165, 226-234; 1977.

Zaldivar,, R^_ Arsenic contamination_of drinking water and foodstuffs
causing endemic chronic" poisoning/"Beitf. PatholV Bd. 151:  384; 1974,

Zoetemann, B.C.J.; Brinlonann, F.J.J. in: Amavis, W.;  Hunter, W.J.;
Smeets, J.G. P.M., eds.  Hardness of drinking water and public health.
Oxford:  Pergamon Press;  1976.  (As cited in Fowler 1979).
                                5-60

-------
               6.0  EFFECTS AND EXPOSURE—NONHUMAN BIOTA
 6.1  Effects on Nonhuman Biota

 6.1.1  Introduction

      This section provides information about the levels of arsenic
 exposure at which the normal behavior and metabolic processes of
 aquatic organisms, terrestrial plants, and microorganisms are dis-
 rupted, as indicated by laboratory and field studies.  There is a
 fairly extensive data base available for some of the chemical forms of
 arsenic; however, virtually nothing is known about the toxicity of
 other forms.   In addition, the chemistry of arsenic in aquatic systems
 is complex (as described in Chapter 4.0) so that laboratory studies
 are often too simple to represent actual environmental conditions.
 Arsenic is usually presented as a salt in bioassays, dissolved in
 distilled or filtered water so that the majority of the total concen-
 tration is available for biological uptake.   In natural conditions,
 complexing and precipitation with ferrous and aluminum hydroxides and
 adsorption onto clay will remove some of the available arsenic from
 solution.   Therefore,  effects levels measured in the laboratory may be
 lower than those in the environment.

      In general,  the pentavalent forms of arsenic are expected to be
 most  prevalent in the  aerobic,  neutral waters that  make up  the
 habitat for most  aquatic species.   Sodium arsenite  (+3)  is  used as an
 aquatic herbicide and  so may  expose aquatic  biota immediately  after
 application.   Transformation  to  the arsenate (+5)  form would occur soon
 after release  in  aerobic systems.   Biota living  in  slightly reduced waters,
 such  as  in eutrophic lakes  or ponds or in the surface sediment  layer,
 may be  exposed  to  the  trivalent  form.   Deeper in the  sediment,  burrow-
 ing organisms  are  theoretically  exposed to arsine (-3);  however,  no
 information on  concentrations of  this  form was available.

      The toxicology  data for  fresh  and  saltwater biota  are  fairly  exten-
 sive, representing a number of fish and  invertebrate  species.  Most
 bioassays were  conducted with the inorganic  trivalent form  (arsenite
 or arsenic  trioxide).  A few studies were available on  the pentavalent
 (+5)  form  (arsenate), organic arsenicals, and arsenic trisulfide.  No
 data were available on  the toxicity of  arsine forms.

     Most of the information presented in this section is based on the
U.S. EPA (1980a) Ambient Water Quality Criteria Document for arsenic.
Other reviews concerning the aquatic toxicity of arsenic can be found
in the U.S. EPA (1978) position document on wood preservative pesticides
in Caldwell et al. (1976) and Spehar et al.  (1980).
                                 6-1

-------
 6.1.2   Freshwater Biota

     The Criteria and Standards  Division of U.S.  EPA has set a fresh-
 water  criterion for arsenic of 0.44  mg/1 for protection of aquatic life
 (U.S. EPA 1980a).   The  criterion applies  to  total recoverable  trivalent  in-
 organic arsenic.   Short-term  freshwater effects were observed  at levels
 as  low as 0.04  mg/1.   No  criterion  was  set  for protection of saltwater
 life;  however toxic effects were reported at levels  of 0.51  mg/1.

     The mechanism of active  toxicity differs among  arsenic  compounds.
 The primary mechanism for the  trivalent form is believed to  be inactiva-
 tion of enzymes due to reaction with the sulfhydryl  groups of  proteins.
 The arsenate form is  not  as reactive with sulfhydryl groups  but
 has been found  to uncouple oxidative phosphorylation.

     Seven fish species have been tested with sodium arsenite.   The
 range  of acute  toxicity values was  narrow for these  species:   the  LC5g
 (concentration  lethal to  50% of test organisms) ranging  from 13.4  mg/1
 to  41.8 mg/1.   The rainbow trout was the  most sensitive  fish tested
 (LC50 of 13.3 mg/1) and bluegill the most resistant  (LC50  of 41.8
 mg/1).   Sodium  arsenate was tested  on rainbow trout  and  a  LCso  of
 10.8 mg/1 was determined.   Sodium arsenate was toxic to  green  sunfish
 (Lepomis cyanellus) at  a  concentration  of 150.0 mg/1;  otherwise the
 range of toxicity  values  to fish from arsenite and arsenate  was similar.
 Toxicity levels  for other  freshwater species  (crayfish,  channel catfish,
 and smallmouth  bass)  exposed to monosodium methanearsonate (an  aquatic
 herbicide) were very  high,  a finding indicating that organic arsenic
 may be  much less  toxic  than either  inorganic arsenite  or arsenate  (U.S.
 EPA 1980a). Selected  data  for freshwater  species are summarized in
 Table 6-1.  The reader  is  referred  to U.S. EPA (1980a) for further data.

     Freshwater invertebrates tested for acute toxicity of arsenic include
 four cladocera and  a  scud and one aquatic insect.   The cladocera and scud (LC50
 range 0.8-5.3 mg/1) were approximately four times as sensitive  to sodium
 arsenite as the stonefly  (22.0 mg/1) (Table 6-2).   Other data on the
 effects  of several  arsenicals to a variety of invertebrates  are summarized
 in  U.S. EPA (1980a).  In general, reduced populations of freshwater
zooplankton and  insects are reported for arsenic  concentrations in the
range of 2.2-11.1 mg/1.

     Data on chronic  effects were available only for Daphnia magna.
A life-cycle test was conducted with both sodium arsenite and sodium
 arsenate; the chronic values for both arsenicals  were 0.9 mg/1
 indicating,  once again the similar toxicity of the two compounds for
aquatic biota (U.S. EPA 1980a).

     Sodium arsenite, which is often used as an aquatic herbicide,  was
tested with the eggs and fry of the freshwater Muskellunge (Esox
masiquinongy).   Fifty percent of newly-hatched fry in 5.0 mg/1 were
dead on day 9;  47% mortality on day 12  occurred at a concentration of
of  1.0 mg/1;  and a concentration of 0.05 mg/1 yielded a 46% mortality
                                  6-2

-------
                         TABLE 6-1.  ACUTE TOXICITY OF ARSENIC
                                     FOR FRESHWATER FISH
   Organism
 Test
 Compound
          Hardness
Bioassay  (mg/1 as
Method^    CaC03)
   Rainbow trout
   (juvenile)
   Salmo gairdneri

   Brook trout  (adult)
   Salvelinus fontinalis

  Goldfish (juvenile)
  Carassius auratus

  Spottail shiner
  Notropis hudsonius^

  Fathead  minnow
  (juvenile)
  Pimephales^ ^romelas

  Channel  catfish
  (juvenile)
  Ictalurus ounctatus

 Channel catfish
  (fingerling)
v
 Flagfish  (fry)
 LC50
 ISS/1)       Reference
sodium
arsenate
sodium
arsenite
   FT    140-152


  FT    140-152
             FT   140-152
             FT   140-152
 10.3
 14.9


 26.0


 27.0


15.6
                               Hale (1977)
 Cardwell et al.
   (1976)     ~

 Cardwell  et al.
   (1976)

 Boschetti and
McLoughlin  (1957)

Cardwell et al.
  (1976)
                                18.1     Cardwell et al.
                                           (1976)
                                15.0     Clemens and Sneed
                                           (1959)
                                         FT   140-152

Green _sunfish sodium
Lepomis cvjmellus arsenate
Blugill (juvenile) sodium FT TAn i«
Lepomis nisrrn/->n" T-HO . -L4u— 152
	 — 	 • _ arsenite
Bluegill (fingerling) » Q
<— 9
Bluegill
S 53
Bluegill
S 210
Bluegill
S 365
~J \J J


150.0

41.7

0.29

15.3

16.2


15.8

(1976) ~
Sorenson (1976)

Cardwell et al
(1976)
Hughes and Davis
(1967)
Inglis and Davis
(1972)
Inglis and Davis
(1972)

Inglis and Davis
(1972)
FT = flow through; S = static.
                                       6-3

-------
                    TABLE 6-2.  ACUTE TOXICITY OF ARSENIC FOR
                                FRESHWATER INVERTEBRATES
Organism
Cladoceran
Daphnia magna
Cladoceran
Daphnia magna
Cladoceran
Daphnia pulex
Cladoceran
Simocephalus
serrulatus
Stonefly
Ptaronarcys
californica
Test
Compound
Sodium
arsenate
Sodium
arsenite
M
it
M
Bioassay Hardness
Method (mg/1 CaCOO
S 45
S
S
S
S
LC50
(mg/1)
7.4
5.3
1.0
0.8
22.0
Reference
Biessinger and
Christensen (1972)
Anderson (1946)
Sanders and Cope
(1966)
Sanders and Cope
(1966)
Sanders and Cope
(1966)
Scud
Gammarus
pseudolimnaeus
Sodium
arsenite
FT
0,9
U.S. EPA (1980c)
                                       6-4

-------
 water indicates that water
                                                        to
                                                   concent"tions in
                                         ? s ^"s^ssx
 toxicity were not  available   ( Fur th
 factors and situations £«' favor "            °  °"
 organise is found in the follow^!
                                                           arsenic
tive to arsenic than the juvenile or adult
stage toxicity values were  also lower than
tebrate species tested (U.S. EPA 1980a)

6-1.3  Saltwater
                                                         with
                                                           m°re
                                                      the 6arly life
                                                   Sensitive

                                                                that
                                                3nd  three ^vertebrate
                                                        f°r both
  for frehshwater spec£
  species (Table 6!3  .
  arsenic trioxide and  ou   rsenito
  stickleback,  white shrimp, and  sheiiflsh  is aui
  for bay scallops to 24 700 ,,*/7  fni  •   '.^  q  te narrow:  3,490  yg/l
  trioxide was  fo'und toxic tolwo  specieTof   f ite.shrimP'  Arsenic
  range of 7,1200-12,300 ug/1  No T^        lmOn in the concentration
  available.             yg/1'  N° data concerning chronic effects were

 6-1-4   Phytotoxicitv
 the                        s  op.
 and loss of turgor, leading to meLrane
 due to reaction with sulfhvdryl en
 senate are  less irmnediate anJless
 adenine diphosphate/adinine
other plant enzymes
herbicide, applied
                                        ""
                                                       CaUS£S Wiltin
                                                    ™ PlantS' aPP^en
                                                  The eff ects of  **'
                                                  interfe«nce in  the
                                                  interfe^ce with
soil solution depends upon the
- arsenic in soil,  whic'h
                                      °    °"
                                                            with
                               6-5

-------
                    TABLE 6-3.  ACUTE TOXICITY OF ARSENIC
                                FOR MARINE BIOTA
                          Test
Organism
Bioassay  Test     Concentra-
Chum salmon
Onchorhynchus
keta
Pink salmon
0. gorbuscha
Pink salmon
0. gorbuscha
Bay scallop
Argopectan
irradians
American oyster
(embryo)
Crassostrea
virginica
White shrimp
Penaeus
setiferus
Arsenic FT 48-hr LC50 8.3
trioxide
Arsenic - 96-hr LCi00 12.3
ion
Arsenic - 7-day LC100 7.1
ion
Sodium _ 96-hr LC50 3.5
arsenite
renewal
Sodium S 48-hr LC50 4.3
arsenite
renewal
Arsenic - 96-hr LC50 24.7
trisulfide
Alderice and Brett
(1957)
Holland et al.
(1960)
Holland et al.
(1960)
Nelson et al.
(1976)
Calabrese et al.
(1973)
Curtis et al.
(1978)
FT = flow through; S = static.
                                       6-6

-------

   6-1.5   iMicroor nanisms
  a
  arsenate and arsenite  arfoms  to




     arsenate to arsenite      enris m^ST^ ^ ln the reduction
  1977)  and the effects levels of the o^f   f" ^ arsenate  (NAS
  two orders  of magnitude lover than th^T   ^ *™ Senerall7 °ne to
  and Cox 1978).   fhe toxicitv of ?1        r lnor§anic "senic (Holm
  be  reduced  by the P^Sc^V™^^
  the eff^  of                       p
  ficult to extrapolate  to field  condiSf h°nS>    ^  r6SUltS  are dif~
 were conducted under simDle l^f      because  most  experiments
 field conditions.  In  either casl^ C°ndltions  or  «  special  set  of
 replicate the complej  SrtaJLr^nd  t   6f 6rimental desig" fails  to
 and water.  Since^ey  fac?Irs controm^^T'10113 °f arS6nic in  soil
 microbial uptake— absorntlon  ^°n"Ollln§ the ^ount available  for
 among others-are no  Spresen e"  t'e^e^c^100; ^ "^""tlo
 very meaningful. for environmental 'conditions!3   VSlS rep°rted are nOt
 can             ™*^^^^ £J£ experimental regime,  ho.ever
 "s  greater  proclivity for binding  to cell  M      arsen"e  is  attributed to
 groups  (NAS  1977).  The relationship  T,f?          '  especially  sulfhydryl
 arsenate  to  arsenite and  the  toxic  eff £?** mi"°bial ""version of
 SSrS'  f°  the f-— — -   "efmorSe  ™^ ^K^*™
                                            fd was  found
valent form to significant^     P!       u       Arsenic in the  tri
            the                     '^
                                 6-7

-------
       TABLE 6-4.  TOXICITY OF ARSENICALS TO TERRESTRIAL PLANTS
Form
Concentration
                                  Effect
                                         Source
Total arsenic
(probably +5)
Total arsenic
25-85 mg/kg
(3-28 mg/kg
 soluble)
44 mg/kg
(6 mg/kg
 soluble)
General critical
level at which crop
yield may be
depressed

Depressed growth in
blueberry
Walsh  and
Keeney (1975)
                                                          Walsh and
                                                          Keeney (1975)
Total arsenic    25 mg/kg
                 Depressed growth in
                 snap beans, peas
                        Walsh and
                        Keeney (1975)
MSMA
DSMA
2.2-2.8 kg/ha
2.2-3.4 kg/ha
Controls growth of
nutsedges (Cyperus
spp.) and Johnson-
grass (Sorghum
halepense)
Hiltbold (1975)
                                 6-8

-------
                  TABLE 6-5.
               EFFECTS  OF ARSENIC ON MICROORGANISMS
   Species  or
   Source of
   Population

   Bacillus
   cereus^


  Nitrifying
  micro-
  organisms
  Form of Arsenie  Concentration

  Arsenite           0.4 inMa
  Arsenate           10 ^

  Arsenate         0.1-1000 tng/1
 Nitrifying
 micro-
 organisms

 Lake
 population
 Cacodylic
 acid
 Arsenate and
 Arsenite
 0.1-1000 mg/1
 40 uM/1
 (~3 mg/1)
Soil
population
(nitrifying
microorganisms)
Arsenate and
Arsenite
50 uM (in 3 ml
solution
applied to
10 g soil)
                                            Reference

                                        Mandel ot  al.
                                           (1965)


                                        Holm and Cox
                                           (1978)
  Effects

  Growth inhibition

  Growth inhibition

  At 100 mg/1
  decreased
  activity of
  Nitrosomonas:  all
  concentrations
  decreased activity
  of Nitrobacter.

  Only  at 1000 mg/1
  was nitrification
  inhibited.
 Organic matter      Brunskill et al
 degradation reduced   (1980)  —  ~
 by 50£ during
 winter (low nutrient
 levels)  but no effect
 during summer.  The
 rate of  phosphorus
 formation and  algal
 uptake of arsenic
 were also reduced
 in  the presence  of
 both forms.

Minor  inhibition of Liang and
nitrogen mineral!-  Tabatabai
zation by both      (1977)
forms.
   concentration given.
                                   6-9

-------
 6.1.6   Conclusions

     Although  comparable  toxicity  data  are not  available  for all of the
 various  arsenic  compounds,  there appears  to be  no  significant difference
 in toxicity between trivalent and pentavalent inorganic forms for  a
 given species.  Organic arsenicals are  generally less toxic  than the
 inorganic forms.  The range in values of arsenic toxicity for freshwater
 species  encompasses over 3 orders of magnitude  and is similar for  fish  and
 invertebrates.  For example, acute toxic effect levels for aquatic  biota
 range from approximately 290 ug/1  (bluegill fingerling, sodium arsenite)
 to 150,000 ug/1  (sodium arsenate, green sunfish) .  Most values  are  in
 the range 1,000-50,000 ug/1.  The limited data available on saltwater
 biota were for  the trivalent form only and indicated LC50's for inverte-
 brates and fish between 3,500 ug/1 and  25,000 ug/1.  'For many terrestrial
 food crops, a  range of total arsenic between 25 mg/kg and 85  mg/kg
 (soluble arsenic from 3 to 28 mg/kg) was the critical level  for depressed
 growth.  Microorganisms were more sensitive to  trivalent than pentavalent
 arsenic  and to inorganic than organic forms.  It is difficult  to extrapo-
 late laboratory-derived effects levels  (from plate culture studies) to
 environmental  conditions and no information on  inhibitory levels of •
 arsenic  in natural soil or aquatic systems was  available.

 6.2  EXPOSURE OF NONHUMAN BIOTA

     Terrestrial and aquatic organisms are exposed to arsenic, primarily
 in the form of arsenate and arsenite, via several environmental  pathways.
The specific chemical form involved is usually unknown as a result  of
 the typical analytical practice of measuring total arsenic concentration.
Nevertheless examples of characteristic exposure situations can be
described from data presented previously in the Materials Balance
 (Chapter 3.0) and Fate and Monitoring sections  (in Chapter 4.0).  The
 following section describes exposure pathways and attempts to quantify
exposure for aquatic and terrestrial populations on the basis of
 typical and maximum levels in surface water, likely sources of
high exposure levels, and how the environmental fate of arsenic may
affect the exposure of biota.

6.2.1  Aquatic Species

6.2.1.1  Limitations of Available Data

     Most aquatic effects studies concern the effects of trivalent
arsenic  (arsenite).   According to the previous discussion, there is
less than an order of magnitude difference between lethal levels of
arsenate and of arsenite based on the limited data available   (see
Table 6-1 and 6-2).   Further, only limited monitoring data are available
distinguishing between the different arsenic forms.  Hence,  though
exposure to the various environmentally important  inorganic arsenic
 can not be differentiated, the similarity in toxicity levels  suggests
that this is not a serious limitation.   Organic arsenicals,  however,
are considerably less toxic than inorganic forms probably due to a
                                  6-10

-------
    ing                              ro          .
    almost non-«lsant   Benet •  T  '""f311" arseni= compounds are
                 in      H- «
   6-2.1.2  Background Levels
                       r                  " ^""^ 1-els of great
csntrations in frefh Waters f°™ nl»r ma8n"U<"-   ^"^ natural con
                     "                 ""1'
                         aers   ™ n»r
   tte National Acadamy "f ScLn^s Sler"^™?",1?' ' been COm"lled
  background concentrations of Jhe eLment in S"     .exceedin§ "^ural
  In areas with already hi<>h natural SS?    the vicinity of the release.
  have little incremental effect In ,i     ' ™ additional release may
  communities.   Additionally  ««   C°\cef ratl°"s affecting aquatic
  of arsenic are likeljto ^ ^ul^^3^^117 high ^'entrations
  species,  which may be ^affected bv In cr arsenic-accl^ted or resistant
  elicit  adverse effects in Mn        lncreases  to concentrations that
         of  some  intentional releases  (f J    '   "T6  °f  the gl°bal
  tinguishing  the additive effict^f in?'8;*  'o^)-   Therefore, dis
  WUh natural ground letflTcan                         "

  6'2'2-3  Fate Consideration
 not
 of arsenic to surf a     tr,  seeral f
 which determine the availability and fS ?roc^ses  may take place,
 Biotic exposure.   The  chemical  L™   Predomnant  form present for
 and ecosystem parameters  win botnYfi     ^"u131 arSenic release
 following  release.   The following d^f   ?"  '^  e^uilibri^ conditions
 of  the more  common  transforma^s tLTdltLT^5  " deSCrlbe
 aquatic communities.                    determine the  exposure of
                                                       a lake or stream

Apparently the pentavllent form is most'ST™5 ***' ^ SQd±^-
trivalent is oxidized to the Tntava^i- f   7- C° react in this ma™«;
Proceeds as described (F.^r^'^.^V ^^ -te" ^ thin
ratio (ror total As)  is usually on the o^l!  i*       sediment to  water
                                 6-11

-------
 Most of the arsenic present in sediment is stable,  in the form of
 insoluble sulfides or as arsenic metal.   Some of the arsenic, probably
 a small amount,  may be subject to microbial methylation.   One product
 of this activity,  dimethylarsine, is more soluble than the previous
 forms and likely to move into the water column and  be transformed to
 methanearsonic acid (MSMA).   Other transformation products include
 cacodylates and  trimethylarsine (NAS 1977).

      Sediment  will,  therefore,  act as a reservoir for waterborne arsenic,
 which would otherwise be available for uptake by most fish and algae.
 Accumulation in  sediment,  especially in the  surface layer, may expose
 benthic organisms  to high  arsenic levels.  Monitoring data (see Chapter
 4.0,  Pathway 3)  indicate slightly higher  tissue  levels in organisms
 associated with  the  sediment  layer than  in  those associated with the
 water column;  the  difference  observed,  however,  was not great.   The
 small difference implies either that uptake  from sediment is  not as
 efficient a process  as from water (due  to  the chemical form present,
 mechanism of uptake  and  absorption efficiency, etc.)  or that  species
 differences  in the ability to bioaccumulate are more significant  than
 arsenic  concentration  differences  between water  and  sediment.

      Another implication of arsenic  accumulation in sediment  is  the
 potential  for subsequent release  to  the water column under  conditions
 of  turbulence or chemical change  (e.g., pH, redox).   Seasonal  upwelling
 of  bottom sediments,  continual  decreases in the pH of northern lakes due
 to  influx of acid rain, and increased microbial activity  can upset  the
 equilibrium  conditions governing  the retention of arsenic in sediment.
 No  specific  incidences are known  of  fish kills or increased bioaccumu-
 lation due  to environmental changes  affecting availability, but  the
 potential  for such an  occurrence must be recognized.

      The arsenic discharged to freshwater that is not subject to removal
 by  adsorption onto sediment will be present primarily in  the pentavalent
 inorganic  form or as an organic acid under buffered, aerobic conditions
 such  as in streams, well-mixed lakes, and epilimnion  (upper layer)  of
most  static  bodies of water.  At least a small fraction of originally
discharged trivalent will remain untransfonned (Ferguson and Gavis  1972).
Under reduced conditions, such as in some small stagnant ponds, eutrophic
 lakes, and in the lower reaches of deep lakes, the prevailing form is
likely to be the  trivalent.  Distinction between these two forms in
exposure and effects assessment does not appear to be as important  for
aquatic species as it is for human exposure.   This is because of the
similarity in the toxicity of the two forms for those aquatic species
bioassayed.

     Other chemical forms of arsenic, some highly soluble  and available
for biological  uptake, are also present in the water column.  Some of
these include DSMA, MSMA, and cacodylic acid.   The very limited monitor-
ing data available on specific forms of arsenic indicate that  concentra-
tions of methylarsenic acids may sometimes exceed concentrations of
                                 6-12

-------
                                 o
   Florida  aquatic systems (a river YH P"dominanc spec±es in three
   10  systems'  sampled  a^fto  be L talent ^  eStUarlne ba^  «* of the
   other ecosystems  (see Ta^le V? * ^       inorganic levels in three
   ing efforts v^inTcatl this"  o'be  aTid*  ^   ^  **^™ "onltor-
   systems.  The implications  of  tM* V  Wldespread  occurrence in aquatic

   to discern in the' absen^of ejects d'atT""^ T   ' "e
  arsenicals.   They are commonly  described  in th
  less toxic than inorganic forJs; and thf  Sti
  sodium aiethanearsonate— U.S  E?[ 1980^ H?    avai^ble  (on only mono-
                          0.5. jii-A 1980a) aia support this statement.
  well.


  6-2.1.4  Monitoring Data
                                                to  saltwater  systems  as
    the range of 0.2 pg/1 tooo   /1   ^     ^S/1-   °f these'
 (275  observations)  wefe  from ?00  f   /, Appr°Xlmately « of the values

 includes  the total  range oTrepo^ed^o.ic  elfec? I71'/ ^^ that
 organisms  (approximately 200-150,^00  £%   ?iftv    ^ t0  aqUat±C
 greater than 1,000  yg/i   which  ,'  ^n^f 7'   F^ft7-°ne observations  were

 acute  toxic  effects to boS  freshwater °™T  y  **"  threshold l^el for
 The implications of this  finding ?or  thf   "ari?e,fl«h  and invertebrates.
munities is  discussed in  Chapter 7       Potential risk of aquatic  com-
 in order to deterine the f


              J£^
 and 904  within  the  Missour
                                                              s in

                                                     ** greater detail



                                       6""      "^ r±Ver basins
             or
power plants are located
                                0
                               vl
                                                           coal burning
                                6-13

-------
    TABLE 6-6.   MONITORING STATIONS IN UPPER MISSOURI RIVER BASIN
                REPORTING HIGH MEAN ARSENIC CONCENTRATIONS
                (1975-1979)a
 Station

 Montana  (Minor  River Basin  901)

 Galena Creek just  below mine  dump
 Silver Creek at road above  mouth
 Anacondo  Co., Great  Falls Plant #1
 Anacondo  Co., Great  Falls Plant #2

 South Dakota (Minor  River Basin 904)

 Boxelder  Creek  below Nemo (Black Hills)
 Annie Creek  at  Highway 14A
 Belle Fourche River  below Confluence
 Belle Fourche River  above Confluence
 Whitewood Creek at Crook City
 Whitewood Creek at Deadwood
 Whitewood Creek above Gold Run Creek
 Whitewood Creek above Belle Fourche
 Belle Fourche River  near Elm Springs
 White River  at  State Highway
 Bad River near  Ft. Pierre
 White River  near Oacoma
 Cheyenne River  at Cherry Creek
 North Dakota  (Minor River BAsin 904)
 North Fork Grand at Haley
   Mean Arsenic
Concentrations(yg/1)
     543
      37
      14
      34
      32
      40
   1,889
      20
   5,258
   6,395
      26
   5,664
   2,230
      23
      37
      32
      38
      38
      82
    No.  of
 Observations
 12
 12
 2
 1
 2
15
 9
 5
 9
 9
 9
 9
18
 3
11
19
20
20
13
 These stations are those that had mean total As levels in water
 from 1975-1979 in excess of 10 yg/1.   A total of 19 stations were
 reported in Minor River Basin 901 and 59 in Basin 904.
Source:  STORET  (U.S. EPA 1980b).
                                 6-14

-------
   the'
                                             -PP-r ^ be associated
   also
   the past 5 years.   A total of 11 and

                                                       "
                                                                 "»
                                                              1-°°0 pg/1  over
                        nroin
   constant, ., . range of" out l"  "f LeBl" ^^ '° "'
   According to the data for arsenic rlsidSs iffrf. Ja?S°n 1973)
   species,  there is little differem-I k ?             and ™ar:lne
   in  annals  fron the                                bo   "
       Introduction
                               Source,  nf  '— ' -
                               caton
  fossil  fuel  co^ustion  andproduction
  the Materials Balance (ChaDtJr  ?   "
  mental  discharges are ^Srec
  31%, is released to land with Li     air
  arsenic is transferred to «,,,rfa
  and/or runoff and fallout Sle
 arsenic releases that
        in
                                                            °CCUr
                                                 eleme ts>  According to
                                                        °f 3l1 6nVir°n-
                                                    T   ^ ma^rity.
                                                   °th comPartments,
                                                   reSPective^.  Caching
                                                    t0tal fraction of   g
                                    .
 pesticide users  (aquatic and terresLiaff ^ ^^ rates)  are
 Phosphorous  production  ooeratl^!   I ?  applications),  zinc and
 tlon  of various  other eS£t ^  and "
 releases.  Pesticide producers  ar2  al
 surface water; howeve? no infor^atin
 charges was available   ?he
 had a low esti.ated aquatic    c
 of exposure to arsenic from these
following subsection,   m
those described above,  a
                                           h H     C°mbustio-  Produc-
                                                ^ associated annual
                                         ex?ected to release arsenic to
                                              ma§nitude of these dis-
                                           °f ^     (triox"e) ,  itself,
                                                      lS and likel^ood
                                 to  inH   ,       *       described  In  the
                      , a co           fTrr      ef^luents,  especially
toxic substances in the discharge sufh      V*  the Presen"  of other
hydrocarbons.  These additional substa    °     m6£alS  ^  chl^inated
ent's effect on the receivin" system  L   ^ Cont^ute to  the  efflu-
and synergistic effects.     °  *     ' dependln§ uP°n their  concentrations
                                 6-15

-------
      Production Sites

      The sole U.S.  producer of arsenic trioxide,  ASARCO,  was estimated
 to discharge less  than 1 kkg arsenic to surface water annually (see
 Chapter 3.0).   All of the 1,200 kkg of solid waste produced annually
 was assumed to be  land-disposed rather than deposited in  water,  in
 order to examine the  local effects  of these releases.   The Pacific
 Northwest Major River Basin in which the ASARCO plant is  located was
 examined and found to have levels exceeding 1,000 yg/1 (total arsenic)
 in surface  water only two times in  the last 5 years according to the
 available STORET data (U.S.  EPA 1980b).   Examination of minor river
 basin 1311,  Puget  Sound, 'in which the plant is found indicated that
 arsenic levels have not  exceeded 10 ug/1 at any  of the 45 stations
 for the past 5-year period.  Mean levels for the  entire minor river
 basin for both remarked  (4 observations)  and unremarked data (49 obser-
 vations)  were  approximately 1-3 pg/1.   These results are  compatible
 with the low reported industrial discharge rate.

      The possibility  that  a significant  fraction  of the arsenic  released
 into Puget  Sound was  absorbed  onto  the  sediment was also  examined.   Sedi-
 ment data for  the Pacific  Northwest,  however, did  not show significantly
 higher levels  of arsenic  than  in other  parts of the country.   This  may'
 reflect  averaging effects  due  to aggregating a large amountof data.
 However,  other explanations  can be  that  the amount  of  arsenic discharged
 to water in the Pacific Northwest basin  is not significant enough to be
 reflected in sediment levels or that sediment transport out of the  system
 is rapid enough to  keep  arsenic levels  in the sediment  layer diluted.   The
 few observations of arsenic levels  in water in the  minor  river basin down-
 stream from the ASARCO plant were at levels even  lower  than the  Puget Sound
 data;  however  the small sample  size  precludes any conclusions.

      Use  as  Aquatic Herbicide

      Sodium  arsenite was used in the past  in applications  directly  to
 surface water  at concentrations  of approximately  10 mg/1  (Ferguson  and
 Gavis  1972)   to  control aquatic vegetation.   It is not known how  common
 its  use  is in  the U.S. at  this  time>  although any use is probably in
 southern  regions where vegetative growth  in waterways is more  of  a
 problem  than in  colder areas.  The total  amount used annually  is  esti-
mated  at  less  than 90 kkg  (JR3  1980)5 thus on a national scale use is not
very  significant compared  to terrestrial  applications of other arsenicals.
There  is  a potential  for long term persistence of arsenic  in  the  sediment
 of treated  lakes with subsequent release  during turbulence or mixing
 (see  Chapter 4.0 Pathway 5).

      Runoff  to Surface Water From Pesticide Use on Land

     The majority of arsenic produced is used in pesticides.  Runoff to
 surface water  from land application of arsenicals is expected to  account
 for  7% of the  total amount of pesticide use each year according to
measurements made in Texas watersheds (Richardson _et a_.L. 1978) .
                                  6-16

-------
                                                          '
 pesticides is not known

 rate of soil erosion";

 and a high rainfall  rate

 face
   in .ore Southern regS?   Xe ilt^l^T™' I"" "" 1§ Pr°bab

                              h» V ^            USe o£ °th" arsenical


                               ^c^ll™™^ "' "" D'S'  *** *
                              Tr  T         j   ;  P     ^"agement techniques


                               ru^offut" ea ^"co" °£  fT" '°
   above)  is sreatest In  i-ho  5  ...       j  least of  compounds  described

   regions of thfn?S.        S<™»east and  possibly  the  South  Central
     soil
                                                            be •*««b.d
become part of the system'!

6.2.1.5.
                                                        With Water and
                                         arsen^ as described in Section
       Other Releases
 charges of
                                             to
                                                                also in dis-
 in acid raine drainge   d         t        1


 have been reported (see Appendix C)   All of ^ disposal Pond effluents

 trations are low enough that   fol?^'    J-?     SS wastestream concen-

 unlikely that significant levels Jn   f dllutf°n on ^"charge,  it  is


 releases.   Where ^ontin^ J Ints SlS'JLS™."^  ^  ^^^^

 drainage areas,  there  is  a possibilitv  r^f^  u      made °r in mlne

 build up in sediment and             7        1§h  3rsenic  levels
levels in their tissues.              "  Sh°Wed higher than background
                                 6-17

-------
     No other field studies were available describing  the  impact  of
these releases in natural ecosystems.  Therefore, any  conclusions made
about the significance of these arsenic sources are, at best,  specula-
tive.

6.2.2  Terrestrial Biota

     Terrestrial systems receive the largest fraction  (81%) of the total
environmental releases of arsenic.  There is a strong  possibility that
terrestrial biota — plants and microorganisms — will be exposed to
high levels of arsenic.  Monitoring data indicate high levels of arsenic
(total As) in excess of 10 mg/kg in soil associated with industrial or
agricultural activities and also in naturally contaminated areas.
Table 6-7 lists some of the reported soil levels.  The highest concen-
trations are associated with orchards which received applications of
lead arsenate in the past.  Although current use of these pesticides is
greatly reduced, the residues already accumulated are very persistent.

     There appears to be a relationship between the degree of phytotox-
icity and plant bioaccumulation and the concentration of soluble arsenic
in the soil solution (see Section 6.1.4).   Therefore, concentrations of
total arsenic in soil listed in Table 6-7 overestimate the amount of the
element actually available for uptake.   Monitoring data reporting soluble
arsenic levels, however,  are uncommon.

     Regardless of the form initially released, it is expected that
arsenate will be the predominant form of exposure in the upper aerobic
layer of soil which supports plant growth and microbial activity.  As
described in Chapter 4.0, the ratio of  arsenate to arsenite is usually
8 or 9:1; sometimes DSMA is present at  levels comparable to those of
arsenite.  Microflora in flooded soils  and trees with deep roots may be
exposed to higher levels  of trivalent arsenite than those usually
measured in the soil surface layer.

     The discussion concerning potential contamination of ground water
by arsenic presented in Section 5.2.2.1 is also applicable to the
potential for exposure of terrestrial communities.   In both cases the
soluble, mobile fraction of arsenic is  of  importance in ultimate expo-
sure.  Table 5-5 describes the geographic  distribution of known arsenic
releases to land.   The most significant sources are fly ash disposal
ponds,  smelting plants and use of arsenical pesticides.  These sources
tend to be concentrated in the industrial  northeast and central states.
The distribution of natural high levels of arsenic is unknown.

     Two factors probably minimize the  potential for exposure of certain
terrestrial species and communities.  Plants  which are of economic
importance or significant as human exposure routes are generally not
cultivated in the vicinity of industrial operations.   Additionally,  the
use of  arsenical pesticides on food crops  is  restricted.   Therefore,
exposure of economically-important plants  to  high arsenic levels is  not
likely.   Also there is little evidence  for significant lateral movement
                                 6-18

-------
                TABLE  6-7.   EXAMPLES  OF  TERRESTRIAL  SITES  OF
                            SIGNIFICANT  ARSENIC  EXPOSURE3
   Location
   Vicinity of fly ash disposal ponds


   Downwind from copper smelters:
   Montana and Washington
                                                Concentration in soil
    6.5 mg/kg
 150-380 mg/kg
  Orchards receiving arsenical
  pesticide treatment
  Potato fields receiving arsenical
  pesticide treatment
 165-627 mg/kg
 maximum 2,500 mg/kg


   6-28 mg/kg
  Soils with high natural
  background levels
up to 80 mg/kg although
usually < 10 mg/kg
          «,

b .
 Limited data.  Higher levels may exist.
Source:   Chapter 4.0.
                                 6-19

-------
 of arsenic  in  soil with  the exception of areas with  significant wind
 erosion  (see Chapter 4.0).  Most migration  tends to  be downward,  away
 from  the  soil  layer supporting terrestrial  life.  Therefore,  the  exposure
 of natural  communities of plants and microorganisms  to high anthropogenic
 arsenic levels is expected to be concentrated in the immediate vicinity
 of specific sources.

 6.2.3  Conclusions

     Aquatic species are exposed to numerous chemical forms of arsenic,
 but the limitations of the monitoring data  base prevent consideration
 of specific forms other  than arsenate, arsenite and  several organic
 arsenical compounds.  Natural background levels of arsenic in.surface
 water are similar in magnitude and variability to arsenic levels
 resulting from human activities.   Ideally,  the background concentration
 at a particular location should be known in order to assess the impact
 of an intentional discharge on aquatic communities;  in practice,  this
 information is difficult to obtain.  According to data on the chemical
 speciation  of arsenic in aquatic systems, a large fraction of discharged
 arsenic accumulates in the bottom sediment, associated with iron, alumi-
 num, and manganese hydrous oxides.   These concentrations are commonly
 two to three orders of magnitude higher than concentrations in the water
 column.  Sediment-burrowing and other benthic species would be exposed
 to these concentrations.  Most fish species are exposed to the remaining
 fraction of arsenic in the water column, and usually at total arsenic
 levels lower than 100 yg/1.   Arsenate is thought to be the predominant
 form under  aerobic conditions and arsenite under anaerobic conditions.
 Limited monitoring data indicate that organic species,  particularly DSMA,
 may be the major form in certain water bodies.

     Approximately 50% of the observations exceeding total arsenic levels
 of 1000 yg/1 over the past 5 years  occurred within the Missouri major
 river basin.  These high concentrations occurred primarily in southwestern
 South Dakota and did not appear to  be directly associated with any parti-
 cular industrial site.   High levels were also observed  in Montana in the
vicinity of a mining site.   The remaining levels exceeding 1000 yg/1 were
 reported in the Upper Mississippi and the Colorado major river basins.

     Potential industrial or use-related sources of arsenic releases
affecting aquatic communities include arsenic production sites,  aquatic
herbicide application,  runoff from  application as a terrestrial pesti-
 cide,  mining and metals-processing  facilities,  fossil fuel combustion
plants and discharge of phosphorus-containing detergents.   Although the
 total contribution from some of these sources,  particularly pesticide
 runoff,  metals production,  and detergent discharges,  is significant on
 a national scale, the concentrations detected in effluents are not high
 (usually  < 13 mg/1)  according to  the limited data base  available.
Virtually no information was available regarding the  long-term impact
 of continual arsenic releases on  aquatic communities  in terms  of
 productivity or diversity.
                                 6-20

-------


6-21

-------
                              REFERENCES

 Alderdice, D.F.; Brett, J.R.  Toxicity of sodium arsenite to young
 chum salmon.  Prog. Rep. Pacific Coast Stat. Fish. Res. Bd. Canada.
 108: 27; 1957.  (As cited in U.S. EPA 1980a)

 Anderson, B.C.  The toxicity thresholds of various sodium salts de-
 termined by the use of Daphnia magna.  Sewage Works Jour. 18: 82- 1946
 (As cited in U.S.  EPA 1980a)

 Beisinger, K.E.; Christensen, G.M.   Effects of various metals on
 survival, growth,  reproduction,  and metabolism of Daphnia magna.
 Jour.  Fish.  Res. Board Can.   29: 1691;  1972.  (As cited in U.S.  EPA 1980a)

 Boschetti, M.M.; McLoughlin, T.F.  Toxicity of sodium arsenite to
 minnows.  Sanitalk.   5: 14;  1957.  (As  cited in U.S.  EPA 1980a)

 Braman,  R.S.;  Foreback, C.C.  Methylated forms of arsenic in the
 environment.   Science 183: 1247-1249; 1973.

 Brunskill, G.J.;  Graham,  B.W.; Rudd,  J.W.M.   Experimental studies on
 the effect of  arsenic on microbial  degradation of organic matter and
 algal  growth.   Can.  J.  Fish.  Aquatic  Sci.   37(3):  415-423;  1980.

 Calabrese, A.,  et al.   The toxicity of  heavy metals  to embryos of  the
 American oyster, Crassostrea virginica.  Mar.  Biol.  18:  16z'-  1973
 (As cited in U.S. EPA 1980a)                                '

 Caldwell,  R.D., .et .al.  Acute toxicity  of  selected toxicants  to  six
 species  of fish.  EPA Ecol.  Res.  Series 600/3-76-008.   Washington, D.C.
 U.S. Environmental Protection Agency; 1976.   (As  cited in U.S. EPA 1980a)

 Clemens,  H.P.;  Sneed, K.E.   Lethal  doses of  several commercial chemicals
 for fingerling  channel  catfish.   U.S. Fish Wildl. Serv.  Sci. Rept.
 Fish.  No. 316, Washington,  D.C.; U.S.  Department of Interior- 1959
 (As  cited  in U.S. EPA 1980a)

 Curtis, M.W., .et .al.  Acute  toxicity  of 12 industrial  chemicals to
 freshwater and  saltwater organisms.   Water Res. 13: 137- 1979
 (As  cited  in U.S. EPA 1980a)

 DaCosta, E.W.B.  Variation in the toxicity of arsenic  compounds to
micro-organisms and the suppression of the inhibitory  effects by
phosphate.  Applied Microbiology.  23(1): 46-53; 1972.

Ferguson, J.F.; Garvis, J.  A review of the arsenic cycle in natural
waters.  Water Res.  6: 1259;  1972.  (As cited in U.S.  EPA 1980a)

Hale, J.G.  Toxicity of metal mining wastes.  Bull. Environ. Contam
Toxicol.   17:66; 1977.  (As cited in U.S.  EPA 1977)
                                 6-22

-------
  Hiltbold, A.E.  Behavior of organoarsenicals  in  plants  and  soils.
  American Chemical Society  Symposium on Arsenical Pesticides,
  7:53-69-, 1975.

  Holland, A.A. ,_et_al.  Toxic effects of organic  and  inorganic pollu-
  tants on young salmon and  trout.  State of Washington,  Dep. Fish. Res.
  Bull. No. 5;  1960.   (As cited in U.S. EPA 1980a)

  Holm, H.W.; Cox, M.F.  Impact of arsenicals on nitrification in aqueous
  systems.  Thorp, J.H. ed.  Energy and environmental  stress  in aqueous
  systems.  Washington, D.C.: Technical Information Center, U.S. Department
•  of Energy, 1978., pp. 200-230.

  Hughes, J.S.; Davis, J.T.  Effects of Selected Herbicides on Blue-
  gill Sunfish.  Proc. 18th Ann. Conf., S.E. Assoc. Game  Fish Comm.,
  October 18-21, 1964.  Clearwater, Florida: S.E. Assoc.  Game Fish Comm.,
  Columbia, S.C.; 1967, p.  480.

  Inglis,  A.; Davis, E.L.  Effects  of water  hardness on the toxicity of
  several  organic  and  inorganic herbicides to  fish.. Bur.  Sport Fish Wildl.
  Tech. Paper 67.  Washington, DC:  U.S. Department" of  Interior; 1972
  (As cited in  U.S.  EPA 1980a)

  JRB, Inc.  Level II materials balance:  Arsenic.   Draft  report.  Contract
  No. 68-01-5793.  Washington, DC:  Office of Pesticides and Toxic
  Substances, U.S. Environmental Protection Agency;  1980.

  LeBlanc, P.J.; Jackson, A.L.  Arsenic in marine and fish invertebrates.
  Marine Pollut. Bull.  4:88-90; 1978.

  Liang, C.N.; Tabatabai, M.A.  Effects of trace elements on nitrogen
 mineralisation in soils.  Environ. Pollut.   12:141-147; 1977.

 Mandel, H.G.;  Mayersak, J.S.; Riis, M.  The action of arsenic on Baccillus
  cereus.   J. Pharm. Pharmacol. 17:794-804; 1965.   (As  cited in NAS 1977)	

 National Academy of Sciences (NAS).  Principles for evaluating chemicals in
 the environment.  Washington, DC:   National Academy of Sciences;  1977.

 Nelson,  D.A.,  _et al.   Biological effects of heavy metals on juvenile  bay
 scallops, Argopecten irradians,  in short-term exposures.   Bull,  Environ'
 Contam.  Toxicol.  16:275;  1976. (As  cited in  U.S. EPA 1980a).

 Penrose,  W.R.; Black, R. ;  Hayward, M.J.   Limited arsenic  dispersion in
 seawater, sediments,  and biota  near a continuous source.   Can Fish Res
 Board  J.  32(8):1275-128l;  1975.

 Peoples,  S.A.   Review of arsenical pesticides.   ACS SymDosium.  Arsenical
 Pesticides  7:1-11;  1975.

 Richardson,_  C.W. ;  Price, J.D.;  Burnett,' E.  Arsenic concentrations  in sur-
 face runorr  rrom small watersheds  in  Texas.   J.  Environ.  Qual 8(2)-  1Q7S
 (As cited in Acurex 1981)                                             "   '

                                 6-23

-------
 Sanders,  H.O.;  Cope,  0.3.   Toxicities of several pesticides to two
 species  of cladocerans.   Trans.  Am.  Fish.  Soc.  95:165;  1966.

 Sorenson,  E.M.B.   Toxicity and  accumulation of  arsenic  in green sunfish,
 Lepomis  cyanellus,  exposed to arsenate in  water.   Bull.  Environ.  Contain'
 Toxicol.  15:756;  1976,

 Spehar, R.L., ^t  al.   Comparative  toxicity of arsenic compounds and
 their  accumulation  in invertebrates  and fish.   Arch.  Environ.  Contain.
 Toxicol.  9:55;  1980.

 Spotila,  J.R.;  Paladino,  F.V.  Toxicity of arsenic to developing
 Muskellunge  fry (Esox masquinonqy).   Comp.  Biochera. Physiol.  62:67-69-
 1979.                     	^

 Sugawara,  K.; Komomori, S.  The  spectrophotometric determination  of
 trace  amounts of  arsenate  and arsenite in  natural  waters  with  special
 reference  to phosphate determination.   Bull. Chem.  Soc. Jap.  37'-1358-
 1363;  1964.   (As  cited in NAS 1977)

 U.S. Environmental Protection Agency  (U.S.  EPA).   Ambient water quality
 criteria for arsenic.  Washington, DC:  Office  of  Water Regulations
 and Standards,  U.S. Environmental Protection Agency;  1980a.

 U.S. Environmental Protection Agency  (U.S.  EPA).   STORET.  Washington, DC:
 Monitoring and  Data Support Division;  1980b.

 U.S. Environmental Protection Agency  (U.S.  EPA).   Unpublished  laboratory
 data.  Environ. Res.  Lab., Duluth, Minnesota; 1980c.

 Walsh, L.M.; Keeney,  D.F.  Behavior and phytotoxicity of  inorganic
 arsenicals in soils.  ACS Symposium Series.  Arsenical Pesticides 7:35;
 JLy / D •

 Woolson, E.A.;  Axley, J.H.; Kearney, P.C.   The  chemistry  and phyto-
 toxicity of arsenic in soils.   I.  Contaminated field soils.   Soil  Sci
 Soc. Amer. Proc. 35:938-943; 1971a.  (As cited  in NAS 1977).

Woolson, E.A.; Axley, J.  H.; Kearney, P.C.   Correlation  between available
soil arsenic, estimated by six methods, and response to  corn (Zea mays L )
Soil Sci.  Amer.  Proc.  35:101-105; 1971b.  (As cited in NAS 1977)7*""
                                 6-24

-------
                       7.0  RISK CONSIDERATIONS


711  RISK CONSIDERATIONS FOR HUMANS


7.1.1  Introduction
                                                    '

        SIS iden1tification of the subpopulations.at risk
        and the populations exposed (Section 5.2);



        Evaluation of the ranges  of exposure for each sub-
        population (Section 5.2);                        b



        gemination of  the effects  levels  or dose/response

        data xn  the species  of concern and/or in  proxies  ?or
two
                    6ffeCtS  data *« ^ined with exposure data in










      Also exposure levels for total ar-on-f^  •    sener^i.
      trxvalent and  pentavalent  forms  of  inorganic  arsenic
                              7-1

-------
         are considered together in this estimate because of  the
         lack of adequate monitoring data distinguishing between
         them.

      Throughout the previous chapters of this report,  the  importance  of
 the  chemical form of arsenic in determining chemical fate,  effects  and
 magnitude of exposure has been stressed.  Because of this dependence
 risks associated with arsenic exposure are evaluated as best  as possible
 for  the different chemical forms of arsenic.  The most  significant
 chemical forms in terms of environmental abundance and  availability of
 information on human health effects, fate, and exposure are the tr'ivalent
 and  pentavalent inorganics and organic arsenicals (primarily MSMA,  DMSA
 cacodylic acid).  Due to the infrequency of their environmental
 occurrence or to lack of data requisite to conduct a risk assessment,
 all  other arsenic compounds are grouped into a fourth category and
 discussed only briefly.   The differentiation of chemical forms of
 arsenic is maintained throughout the risk discussion to the extent
possible.   However,  humans are generally exposed to a combination of
different arsenic entities rather than to a single chemical entity, so
 that an epidemiologically-based risk estimate cannot attribute effects
to the separate forms and the other estimates based on laboratory data,
are, at best,  approximate.

 7.1.2   Effects  of Arsenic

     The human  and mammalian health effects associated  with arsenic
 exposure are discussed in detail in Section 5-1.  Table 5-2 presented
 a  tabulation of effects and associated reported levels, arouped by  tyt>e
 of effect, and  identified the chemical form where possible.
 Not  all environmentally important  forms are well represented  among
 available laboratory toxicity studies for each exposure route.  For
 example, virtually no laboratory data are available on  the  toxicity of
 arsine  forms, which are nonetheless commonly reported as the most lethal
 arsenic form to humans.  Of the eight studies available that quantify  carcinc-
 genicity and teratogenicity, seven concerned arsenate and only one
 concerned arsenite.   All mutagenic studies available were for arsenate.
 The  adverse effects reported for humans concerned arsenate  (1 study)
 arsenic trioxide (1 study) and general arsenic (form unknown).  Data'on
 the  organic arsenic forms were limited primarily to metabolic studies
which in general indicated limited uptake of these forms but did not
 directly measure toxicity.  The lack of a definitive and consistent data
 base for all chemical forms of arsenic is notable.

     In various laboratory studies, low-concentration feeding experiments
with arsenic compounds were negative with respect to induction of cancer.
Teratogenic effects  in mice have been associated with ingested sodium
arsenate at a daily  dose of 120 mg/kg and at lower levels  when adminis-
 tered through intraperitoneal iniection.   Injection is  not a realistic
human exposure route for arsenic and is  more appropriately used for
comparison of species'  differences.  The positive mutagenic effects
elicited by sodium arsenate in cell cultures are not in a  form to allow
                                  7-2

-------
  comparison  to exposure levels.  Effects levels  for humans have  been
  derived primarily from episodes of accidental exposure  to arsenic
  through ingestion of contaminated food or water.  The effects levels
  listed on Table 5-3 were converted to mg/kg/day dose equivalents
  utilizing standard weight values and are presented in Table 7-1.
  Caution should be used in comparing different animal species, concen-
  trations for different chemical forms of arsenic  (i.e., the amount of
  arsenic comprising a concentration of Na arsenate will be greater than
  in Pb arsenate), and also different exposure routes.

      Examination of this table indicates that the lowest chronic dose
  to^induce an effect in humans (i.e., skin cancer)  is that obtained from
 drinking water in Taiwan; peripheral neuropathy has also been reported
 with ingestion of 0.04 mg/kg/day of calcium arsenate for 2-3 weeks.
 The Taiwan study,  based on the consumption of arsenic for more than 40
 years by a study population of 40,421 individuals, provides the best
 data for use in risk extrapolation (in Section 7.1.6).   The results  of
 the study,  however,  must be attributed to  all forms of  arsenic present
 and related to the combined effect of their concentrations,  assuming that
 the presence of substances  other than arsenic did  not contribute to  the
 observed effects.

      It  is  worth noting that the concentrations  eliciting  adverse effects
 in humans  after short-term  exposure  are  generally  5  to  10  times  lower
 than the concentrations  associated with  no  effects in long-term  feeding
 studies  with rats.   However,  the rat  is  unique among  mammals  in  that
 80-90/i of  the metal  is  bound to  the  globin  portion of hemoglobin rather
 than uniformly distributed  to body tissues.   This  difference may account
 for differences  in concentrations  needed to  produce an effect  in this
 species.

 7.1.3  Exposure  to Arsenic

     Section  5.2 described  the pathways leading  to human exposure to
 arsenic.  Table  7-2  summarizes the estimated  exposure levels presented
 as   arsenic per kg body weight (for a 70-kg human) per day.  Ingestion
 and  dermal contact were considered more thoroughly than inhalation
 exposure due  to the existence of detailed exposure assessments for
 airborne routes of arsenic exposure (especially Suta 1980).  Both food
 and water are believed responsible for at least low levels of arsenic
 usually 0.1-0.3 yg/kg/day for a large fraction of the U.S. population'
 In subpopulations either drawing their water supply from wells in areas
 contaminated with high levels of arsenic or consuming highly contaminated
 fish, exposure levels can be much higher, as much as an  estimated 167 Ug/
 kg/day.   Most (^90%)  of the unbound inorganic fraction of arsenic in water
 is absorbed in the digestive tract.  Arsenic bound  to biological tissue
 (e.g., in fish) is thought to pass through  the gut  largely unchanged.
The lower levels of "free" arsenic present  in the fish would probably be
absorbed.  Unfortunately,  little  information is  available on the  concen-
trations  of these absorbable forms of arsenic in  other kinds of food.
                                  7-3

-------
                        TABLE 7-1.  ADVERSE EFFECTS OF ARSENICALS ON MAMMALS
                                    EXPRESSED IN DOSE EQUIVALENTS3
Arsenic
                     Chemical Form
 unspecified
 unspecified
 (Jn milk)
 unspecified
Arsenite
 Na arsenite
ArsenJc Trioxide
Ar senate
 Ca arsenate
(in soy sauce)
 Na arsenate
 Pb arsenate
 Ca arsenate
 Na arsenate

 Na arsenate
    Effect/Species
carcinogenicity in humans
CNS damage in humans
(infants)
lethal to humans (infants)

carcinogenicity in rats
hyperkeratoses and hyper-
pigmentation in humans
peripheral neuropathy in
humans
carcinogenicity in rats
carcinogenicity In rats
carcinogenicity in rats
teratogenicity in mice

teratogenicity in mice
    Effects
Dose Equivalent

0.05 mg/ks/day
0.8 mg/kg/dayc
             r\
7.9 mg/kg/day
(for 1 month)
                                                                    0.13 mg/kg/day
0.04 mg/kg/day
(for 2-3 weeks)
                                                                      No Apparent
                                                                      Effects Dose
                                                                    120 mg/kg (single
                                                                    oral dose)
                                                                    10 mg/kg (single
                                                                    intraperitoneal
                                                                    injection)
 All data derived from summary Table  5-1  in  Section  5.1.
 Assuming a 70-kg human and highest well  water concentration.
"Assuming a 4.5~kg infant.
 All  rat  data  based on consumption of 15  g feed/day  in a  300 g  rat,
                                                                                          12.5 mg/kg/day
                                                                                          23 mg/kg/day
                                                                                          92.5 mg/kg/day
                                                                                          33.3 mg/kg/day

-------
                                        TABLE 7-2.  SUMMARY OF ESTIMATED  ARSENIC EXPOSURE LEVEIS
                                                    ASSOCIATED WITH  INDIVIDUAL  EXPOSURE ROUTES '
I
Ul
            Jj>jjesj;ion

            1-   Drinking Water
                   Surface
                   Ground Water

            2.   Food
                  Total  Diet
                  High Fish
                  Moonshine
                  Wine

            3.   Soil
                                                       +3
 I.   Handling Arsenic-treated Wood

 2.   Use of Arsenic-contaminated
     Detergent

_tjilia_Laj; ion_

3•  Atmosphere
      General Urban
      Vicinity of Sources
      Arsenic-treated House
2-  Cigarette Smoking

3-  Handling, Sawing, etc.  with
    Arsenic-treated Wood
             Assumed  to be  equal "to 100%  if no specific
                                                                    Typical
                                                                                    Maximum
            ast.
                                                    tlielr
                                                                                <0.01
                                                                                                 0.3
                                                                                                the
                                                                                                 Absorption
+5
+3, -K>, arsine?
all forms
0. +3, others
unknown
+3, low +5
01
. 1
0.1
0.3
14.3
<0.0l

-------
      Exposure to arsenic through dermal contact may occur with absorption
 from water,  use of arsenic-containing pesticides,  handling of treated wood,
 application  of arsenic-containing medication,  and  use of arsenic-contami-
 nated detergents.

 7.1.4  Risk  of Exposure  to  Different  Forms  of  Arsenic

      By comparing effects levels for  the  different forms of arsenic with
 exposure levels,  ratios  of  exposure to  effects  levels  can be  calculated  and
 compared and some conclusions  drawn about the  differences in risk dif-
 ferent forms of arsenic  pose for humans.

 7.1.4.1  Arsenate

      According to Table  5-11 (in Section  5.2)  humans  are typically
 exposed to about 0.1  yg/kg  of  arsenic daily, presumably  in the form of
 arsenate, in surface  drinking water.  Exposure  to  arsenate in the total
 diet  is expected to be much lower than  the  0.4  yg/kg  level estimated
 for  total arsenic.  No particular type  of food  is  known  to be contami-
 nated primarily with  the arsenate form.   These  levels  are well below
 the no apparent effects  levels based  on laboratory animal studies and
 the epidemiological study presented in  Table 5-1.

      The exposure  level  (maximum)  estimated for dermal absorption through
 use of arsenic-contaminated detergents  was 57 yg/kg.  No  effects  levels
 for the route  of  dermal  contact were  available  for comparison.

      Also, if  the  ground water level  of 4,000 yg/1 arsenic  described as
 a maximum level of exposure in Table  7-2  is assumed to be 100%  arsenate
 and consumed daily, this level exceeds  the  lowest  reported  concentration
 that  was linked with  skin cancer  in Taiwan.   However,  the presence  of
 appreciable  levels of arsenite in the Taiwanese ground water was  thought
 to be an important factor in the  development of skin cancer in  that
 population.

 7.1.4.2  Arsenite

      Even if 100% of arsenic in ground water is arsenite, the majority
 of the  U.S. population is exposed to arsenite levels that are well
 below effects levels from laboratory studies for either arsenite or
 total arsenic.  The epidemiological extrapolations from the Taiwanese
 study,  however, estimate a higher incidence  of skin cancer associated
with  even these levels.  Due to the great uncertainty regarding this
 conclusion, these extrapolations are discussed separately in detail in
 7.1.6.

      Two other exposure pathways resulting in uptake of high arsenite
 levels  are ingestion of highly  contaminated  seafood and wine (maximum
 level).  The estimated levels exceed the effects level of 130 yg/kg/day
 for arsenic trioxide,  which  resulted in non-lethal skin disorders.
                                  7-6

-------
 7.1,4.3  Organic Arsenicals

      No information applicable to humans was available concerning the
 specific toxicity of the organic arsenicals.  In general, however, these
 compounds (MSMA, DSMA,  cacodylic acid)  are less toxic than the inorganic
 forms of arsenic (see Chapter 5.0).  In addition, metabolic studies on
 organically-bound arsenic in ingested foods indicate minimal absorption
 of arsenic.  Therefore, the effects levels for total arsenic provide a
 very conservative estimate of toxicity for the organic arsenicals, as
 a first approximation.

      Exposure pathways  associated with  potentially high levels of organic
 arsenicals  include  ingestion of  some ground and .surface water supplies,
 food,  especially seafood,  and soil.   With the possible exception of
 diets  high  in contaminated fish  and consumption of drinking water highly
 contaminated  with organic  compounds  of  arsenic,  most pathways have
 exposure levels  lower than the reported effects levels for  total arsenic.
 The  Taiwan  epidemiological study did not report the presence of  any
 organic arsenicals  in water samples  so  it does  not provide  applicable
 effects level in this case.

 1•1•5   Risk in Regard to Combined Exposures  to  Total
        Arsenic for  Selected  Subpopulations

     The exposure levels derived in  Section  5.2  for individual human
 exposure pathways were  aggregated in order  to estimate combined  exposure
 levels,  or  exposure scenarios, for a number  of  subpopulations.   Table
 7-3  lists the  exposure  levels  for water,  assorted  foods or  general  diet,
 and  several inhalation  and dermal exposure routes.   Different routes of'
 exposure  are  combined for an  estimate of  one total  exposure  level for
 purposes  of comparison  of different  subpopulations.  However, in assess-
 ing  the  significance  of these  combined  levels in terms  of human  toxicity,
 the  three exposure routes  (ingestion, inhalation, and  dermal absorption) '
must be disaggregated and compared with effects levels representing each
 particular route.  Also included  on  Table 7-2 are absorption efficiencies
 for each route, which were reported  in Section 5.1 and a value for
exposure level expressed as a dose in ug per kg of body weight.  This
final value facilitates comparison with effects levels in Table 7-1.

7-1-6  Risk Extrapolations for Arsenical Skin Cancer

7.1.6.1  Introduction.

     The potential carcinogenic risk to  humans due to ingestion of
arsenic compounds is estimated below.

     Ideally,  this xrould be accomplished in two ways:
                                  7-7

-------
               TABLE 7-3.  SELECTED ARSENIC EXPOSURE SCENARIOS FOR SPECIFIC HUMAN SUBPOPUI.ATTONSa

Exposure
Med ium
Water
Food /Fish
Wine/Moonshine
I
oo , .
Air
Cigarettes
Soil
Exposure to Wood
TOTALS
Per kg /day

General
Popula-
tion
5
21
0.01

0.06
-
-
-
26
0.4
Smokers
General
Popula-
tion Child
5 2.5
21 11
0.01

0.06 0.06
90
20
-
116 34
2 8
High
Wine
High Fish Consump-
Consumption tion
5 5
21 + 1,000 21
500

0.06 0.06
-
-
-
1,026 536
15 7

Hiuh
Activity
Wood- Wood-
worker worker
5
21
_

0.06
-
-
260
286
4
5
21
_

0.06
-
-
500
536
7

Contaminated
Well Water
Consumption
1 , 000
21
_

0.06
-
-
-
1,02]
15

Worst Case
1,000
21 + ] ,000
500

6
90
20
260
2,8'J7
41
1 • • i — • 	
Maximum Fish
Consiunp tion
5
21 + 10,000


0.06
-
-
-
]0,026
143
The sources of these these levels and their-uncertainties are discussed In Chapter 5 of this report.

-------
         1)   Various extrapolation models would be applied to
             exposure data for occupational and general environ-
             orderVXP°SUres 1Mlal. ?«"»"•"»  »«e taken to
site to the laboratorv  «* °he J s ?  S    T""  frO"  the  =°«"tio
Nevertheless the, are' the bet
                                                                  these
                                  7-9

-------
             TABLE 7-4.   AGE- AND SEX-SPECIFIC SKIN CANCER
                         INCIDENCE RATE FOR STUDY AREA IN TAIWAN
                                         % Incidence (± 0.5)
Age
0
10
20
30
40
50
60
70
Male
0
0
0
1
2
5.5
10
15.5
Female
0
0
0
0
1
2
4
6
Total
0
0
0
1
1.5
3.5
7
10.5
Source:  Tseng ejt al.  (1968)
                                  7-10

-------
        IABLE  7"5

  Concentration
     Range
 _   (mg/1)

     £0.60

 0-30 - 0.59

 0.00 - 0.29

 Undetermined
Age Grour
20-39
11.5
2.2
1.3
40-59
72.0
32.6
4.9
60 and Over
192.0
106.2
27.1
Total
21.4
10.1
2.6
Source:  Tseng et al. (1968)
                                 7-11

-------
                  TABLE 7-6.   CHEMICAL CONSTITUENTS OF
                              TAIWAN WATER SAMPLES3
                                          Chemical Constituents (mg/1)
                                            by Geographic Location	
 Chemical                                 Pei Men                 pu xaj

 Arsenite                                   .05                    .09

 Arsenate0                                  .52                    .63

       Sum                                  .57                    >72
Total Arsenic  (AAS)C                       .72                     .76

Total Arsenic  (NAA)d                       .76
Sodium                                    282 -                   223

Copper                                    <„!                     
-------
 data, a risk estimate is performed assuming that all arsenic compounds
 contribute equally to carcinogenic risk.

      In contrast to the Tseng study, a study by Morton ejt al. (1976)
 showed no evidence of increased cancer incidence in Lane County, Oregon,
 an area in the U.S. where significant amounts of arsenic have been
 reported in the drinking water.  This is possibly due to the lower con-
 centration of arsenite in the Lane County water supply than in Taiwan,
 population shifts in the county, and lower duration of exposure.

      In order to deal with the uncertainties inherent in extrapolation,
 three commonly used dose/response models have been applied to the data'
 in Table 7-4 in order to establish a range of potential human risk.  It
 must be emphasized that the assessment of potential human risk is sub-
 ject to important qualifications:

      •   There have been both positive and negative findings
          for the carcinogenicity of arsenic in humans based
          on epidemiological studies.

      •   Due to inadequate understanding of the mechanisms  of
          carcinogenesis,  there is  no  scientific basis for select-
          ing among several alternative dose/response models  that
          yield widely  differing results.

      •   The latency period  for development of  skin cancer  is
          unknown;  Wagner  (1973)  estimates  it  to  be  between
          2  and  40  years.

      •  Which  of  the  arsenic  compounds,  if any, is  the primary
          contributor to carcinogenicity  is  not  known.

      •   Carcinogenic  contaminants other  than arsenic in  the
         Taiwanese villagers' water supplies may be  responsible
         for  their high incidence of  cancer.  Several fluorescent
         substances were identified including an alkaline hy-
         drolysate of  ergotamine, lysergic  acid, or a similar
         compound  (Lu  et_ al. 1977).  These  substances are highly
         neurotoxic and could potentially have contributed to
         the effects observed in the population drinking the
         contaminated well water.  Unfortunately these substances
         were not quantified in the samples nor was  a large
         number of water samples analyzed for them.

7-1-6.3  Estimation of Human Risk

     The concern here is with the probability of contracting cancer at
some unspecified time during a lifetime, due to ingestion of arsenic at
some previous time.  Unfortunately,  just as there is no basis for
choosing among the various  risk extrapolation models, so there is no
                                  7-13

-------
 clear  procedure  for  taking  into account  the  time  lag between  exposure
 and  response  (i.e.,  the latency period), and the  models  are not  sophis-
 ticated  enough to  attempt this estimate.  Thus, the data in Table  7-5
 for  the  groups of  people ages 20-39 and  40-59  cannot be  analyzed,  since
 the  percentage in  each of these groups that  might yet  contract cancer
 from previously  ingested arsenic, even if further exposure ceased, is
 no t  known.

     The people  making up the last group, however, (people ages  60 and
 over), are more  likely to show a significant  increase  in skin cancer
 cases  during  their lifetimes because this group approaches a  lifetime
 exposure.  These data were  used in the extrapolations, recognizing
 that the true incidence is  confounded by the normal increase  in  skin
 cancer incidence with age.  Table 7-4 shows  that  the cancer rate among
 men  in this age  group is about 1.5 times the rate in men and women taken
 together.  Therefore, in order to arrive at  a  conservative risk  estimate
 cancer incidence in  the 60-and-over age group  is  multiplied by roughly
 X 0 J •

     In addition to  the latency problem, the models cannot deal  with
 ranges of exposure producing levels of response,  particularly not with
 open-ended ranges  such as "greater than 0.60 ug/1."  The average exposures
 for  the low and  medium exposure ranges were chosen while 0.70 mg/1 was
 selected as representative of the high range,  and probably conservative.
 The  derived "data" in Table 7-7 were also used for the extrapolations.
 The arsenite exposure is (.05/.76) = .066 times the total arsenic
 exposure.  This  value is conservative for the  ratio of arsenite  to total
 arsenic, from Table  7-6.

     The three dose/response models used to extrapolate human risk were
 the linear "one-hit" model,  the log-probit model,  and the multistage
 model.   The latter is actually a generalization of the one-hit model,
 in which the hazard  rate is taken to be a quadratic rather than  linear
 function of dose.  All of these models are well described in the litera-
 ture, and a theoretical discussion may be found in Arthur D.  Little (1980).
 The one-hit and multistage models assume that the  probability of a car-
 cinogenic response is described by

                        P(x) - 1 -e-h(x)

where P(x) is the probability of response to dose  x,  and h(x)  is the
 "hazard rate" function.   The log-probit model assumes that human response
varies  with dose according to a log-normal distribution.   Due to their
 differing assumptions, these dose/response models  usually give widely
 differing results when effects data are extrapolated  from relatively
high doses to the low doses  typical of environmental  exposure.

     For the linear one-hit  model, the equation

                        P(x) =1 -e'Bx
                                  7-14

-------
          TABLE 7-7.  ESTIMATED CARCINOGENIC RESPONSE3 IN HUMANS
                      EXPOSED TO ARSENIC IN DRINKING WATER
          Level of Arsenic  in
          Drinking water
                                                  Carcinogenic Response
   Total Arsenic
      0.70
      0.45
      0.15
                            Arsenite
                             .046
                             .030
                             .010
                                               290/1000
                                               160/1000
                                                45/1000
                                                               Percent
                                                                 29%
                                                                 16%
                                                                4.5%
           ce                                     -tud,  of  skin
  drinking  water.   There  are  a n™h   §T  eXP°Sed  tO  3rsenic in their
  the  results  of  this  study to H  ooo,! U?Cert*inties ^ extrapolating
  other potential carcinogens in Ih/drint T* 1™ "° ^ Presence of
  exposure, inadequate unlerstanding of  tS § ^   '  intermitt^y of
  conflicting  studies, among other rfetors   ™efaniSm °f "rcinogenicity
  discussed in Chapter 5.0 and To                uncertainties are


Source:   Derived from Tables  7-4,  7-6.
                                 7-15

-------
 is solved for the parameter B.   From the data for arsenic discussed
 previously,  B is  approximately  0.4  for total  arsenic.   For the log-
 probit extrapolation,  the  "probit"  intercept  A results  from the follow-
 ing equation:

                         P(x)  =   $  (A + loglo  [x])

 where $ is the cumulative normal distribution  function.

      This equation makes the  usual  assumption that  the  log-probit  dose/
 response  curve has unit  slope with  respect  to the  log-dose.   From  tables
 of the standard normal distribution,  A is found  to  be approximately equal
 to -.6 for total  arsenic.   These values  were  used  to determine the
 probability  of a  response  at  various  concentrations according to the
 above equation.

      The  multistage model  with  a quadratic hazard rate  function,

                         h(x)  =  ax2 +  bx  + c,

 was  fitted to  the  same data.  The parameters  a, b, and  c, were  estimated
 by  a  maximum likelihood  method, aided by a computer program that per-
 formed  a  heuristic search  for the best fit.   The parameter b  dominates
 for  small values of dose x, and the parameter a dominates for  large
 values.

      In Table  7-8  the risk estimates obtained from these three models
 are  summarized.  The expected number of  cancers per million exposed
 population is  shown for  a wide  range of  concentrations  in drinking
 water.  These  estimates  represent probable upper bounds on the  true
 risk, due  to the conservative assumptions that were used.  The  gap
 between the  estimates is large  in the low-dose region;   thus, there is
 a substantial  range of uncertainty concerning the actual carcinogenic
 effects of arsenic.  However, present scientific methods do
 not permit a more accurate or definitive assessment of human risk.

 7.1.6.4   GAG Cancer Risk Estimate

     The  Carcinogen Assessment Group (CAG) has derived  a model similar
 to the linear model presented previously, also based on the Taiwanese
 epidemiological study (U.S. EPA 1980).  The model estimates a lifetime
 probability of developing skin cancer equal to 0.004 (4,000 out of 10s
population) resulting from life-time exposure to arsenic at a concen-
 tration of 10 ug/1 in drinking water.  Although some of the basic
 assumptions of the model are slightly different from those of the linear
model, the final estimate is the same.
                                  7-16

-------
                             TABLE 7-8.   ESTIMATED  LIFETIME  EXCESS  CANCERS  PER MILLION
                                         POPULATION EXPOSED  VIA  INGESTION TO  TOTAL
                                         ARSENIC AT VARIOUS  CONCENTRATIONS  BASED ON
                                         THREE EXTRAPOLATION MODELS3
        Extrapolation
            Model

        Linear Model
Concentration
in Drinking
Water (ing/])
10
                                              -7
                                                          Estimated No. Excess Lifetime  Cancers
                                                             (per million population  exposed)3
10
            .-6
                                                       0.4
                    10
            .-5
                    10
-4
                                        10
                                          -3
                                                  40
                                                                                     400
                  10
-2
                                                  10
                                                    -1
                                                  4000     39,000
I
H-"
~^J
       Log-Probit
       Model
       Multi-Stage
       Model
                              0.6
                               60
                                                            360
                                                            600
                                                                                              4700     55,000
                                        6000
                                                                                                       63,000
        The number of lifetime excess cancers represents an increase in number of cancers over the
        normal background incidence,  assuming that an individual is continuously exposed to total
        arsenic at the indicated concentrations in drinking water consumed at a daily intake rate
        of 2 I/day over their lifetime.   There is considerable variation in the estimated risk due
        to uncertainty in the epidemiology study on which these numbers are based and the application
              °  e^Cf ?°fe:re£T°nSe curves'  In view of several conservative assumptions that were
                  it is likely that these predictions overestimate the actual risk to humans

-------
 7-1-6.5   Implications of Taiwan Study  for  the U.S. Population

     To put  the results of the Taiwan  study in perspective  for  the  U.S.,
 it is worth  briefly considering the incidence of skin cancer in the
 United States and also pointing out possible differences between the
 Taiwanese study group and the American populace.

     The  dose-response relationship of increased frequency  of skin  cancer
 among inhabitants of the southwest coast of Taiwan to the arsenic content
 of drinking water utilized above in the calculation of risk, has been
 fairly well documented (Tseng et al. 1968) .  The mechanism of arsenical
 carcinogenesis, however, is still quite unclear.  Attempts  to induce
 tumors in animals have been largely unsuccessful.  Furthermore,  in
 contradistinction to the Tseng study,  a survey of the population in the
 Lane County, Oregon, an area  in the U.S.  where significant amounts
 of arsenic are present in drinking water, showed no evidence of  increased
 cancer incidence (Morton e_t al_. 1976).

     The best quantitative data on the incidence of non-melanoma skin
 cancer among Caucasians in the United  States is contained in a  6-month
 survey of four areas of the U.S.  (Dallas-Fort Worth,  Minneapolis-St. Paul,
 San Francisco-Oakland SMSAs and the entire state of Iowa) during 1971-1972
 (Scotto et al. 1974).   The survey area encompassed approximately 5% of
 total U.S. population in 1970.  The incidence of non-melanoma skin  cancer
 encompasses approximately one-half of  the total of all other forms  of
 cancer combined in the United States (see Table 7-9), with the projected
 incidence for the entire U.S.  approaching 425,000 cases per year.

     Several differences between the arsenical skin cancer observed in
Taiwan and skin cancer found  in the United States,  however are worthy
of note:

     •   In the U.S.,  80%  of  all  non-melanoma skin  cancers occur
         on exposed body sites (e.g.,  face, head or neck) (Scotto
         et al.  1974).   In Taiwan,  60% of skin cancers are located
         in unexposed  parts of the body (e.g.,  legs,  feet,  toes,
         trunk)  (Yeh 1973).

     •   The incidence  of  skin cancer  in  the U.S. appears to reflect
         geographic latitude,  with greater  risk to  residents of
         lower latitudes (Scotto  et_ al. 1977).   Skin  cancer
         incidence rates appear to  double for about each ten
         degree decrease in latitude (Cutler _et al.  1974) .

     •   No significant  increase  in cancer  was  found  in  residents
         of the Lane County,  Oregon, despite high arsenic levels
         in water  supplies, although admittedly much  lower  on
         average than  levels  reported  from  Taiwan (Morton et al. 1976).
                                  7-18

-------
      TABLE 7-9.  INCIDENCE RATES AMONG  CAUCASIANS  FOR NON-MELANOMA
                  SKIN CANCERS, MELANOMA OF SKIN  AND  CANCER^^Tr
                  ORGANS COMBINED FOR FOUR ^S['of ?HE™D  STATES
        Area

   Dallas-Fort Worth  SMSA

   Iowa

  Minneapolis-St. Paul

  San Francisco-Oakland


                  Average
                                Cancer Incidence per in6 Pnn..-i..Mf.n
                                 2095
                                                  49.8
                                                            All  Sites'
3790
1240 '
1510
1840
72
34
33
60
2556
3247
2789
3543
                                                              3034
  Standardized to population of United States for 1970.
 small difference in ae
 the entire United States

"Excluding carcinon, ir
                                              «" ^'^^ the
                                             * adJUSted f°r a ^^t
                                          StUdy PoPulation, to that of
                             and non-melanoma  skin  cancers.
Source:  Scotto et al. (1974).
                                7-19

-------
      Other  factors  that may have  played  a  role  in  the  positive associ-
 ation between  arsenic  ingestion and  skin cancer in the Taiwanese include
 ethnic and  environmental  factors,  socioeconomic status,  occupation,
 personal  habits  (e.g., smoking) and  diet.  Nutrition can affect con-
 ditions of  repair,  immune competence,  cellular  integrity and  skin
 features  such  as  the keratin layer.

      The  estimated  potential carcinogenic  risk  for humans due to inges-
 tion  of arsenic  (total) in drinking  water  were  derived from extrapolations
 based upon  numerous assumptions.   These  result  in  estimating  an extremely
 high  incidence of skin cancer from exposure to  low levels of  arsenic.
 In fact,  based on current U.S. cancer  rate data, the relatively high
 incidence of skin cancer  in the U.S. is  primarily  attributable to
 exposure  to U.V.  radiation based  on  the  relationship between  latitude
 and. cancer  rates.  Because of the  strong role of this  factor  in the U.S.
 skin  cancer incidence, extrapolations  of risk to specific subpopulations
 based on  arsenic  exposure levels  alone seemed neither  appropriate nor
 realistic.  If the estimates were  based  on the  results of a standard
 laboratory animal, long-term feeding,  dose-response study where the
 actual amount  of arsenic  intake was  known, then such extrapolations
 could be  better justified.  In the case  of the Taiwanese study,  however,
 the arsenic intake is only estimated so  that projections onto  another
 subpopulation  with different eating habits and exposure routes  are
 precarious.

 7=1.6.6   The Relationship  Between  Inhalation of Arsenic and
          Carcinogenicity

      Respiratory cancer has been associated with exposure to sources which
 release arsenic (among other pollutants)  to the air (Wagner 1973, U.S. EPA
 1980).  Evidence for this  association is not thought to be as good as the evi-
 dence  relating arsenic to  skin cancer  (Mushak et al. 1980).   Other metals
 and S02 are also often released in these emissions so that interpretation
 of observed effects is complex and not entirely attributable to arsenic.
 Histological interpretation of the resulting carcinomas has  been minimal.
 Also  the  role  of arsenic as a co-carcinogen in the presence of other
 toxins  is not well understood.  As is true for other routes  of exposure
 (ingestion), laboratory attempts to confirm arsenic's status as a
 respiratory tract carcinogen have been relatively unsuccessful.

     The  Carcinogen Assessment Group's (CAG 1980) quantitative risk
 assessment of  the airborne route of exposure to arsenic calculated a
 lifetime  risk  of respiratory cancer of 1 x 10~5 for approximately
 2.16 x  106 people for all air arsenic exposures.  Populations in the
vicinity  of cotton gins had an estimated risk of cancer as  high as
 8.77 x  10~3.  The total respiratory cancer due to airborne exposure
was estimated at 0.00^% of the U.S. total.   CAG estimated that the rate
 of developing  lung cancer over a lifetime increased by  approximately 8%
 per ug/m  increment increase in atmospheric arsenic concentration
 (Mushak et al.  1980).
                                  7-20

-------
   7'1 levels
                                  7-21

-------
       The  risk associated with  exposure  to  adverse  environmental levels
 is  lower if  one considers the effects  reported  for  other  chronic
 exposure to  low levels  of various  forms  of  arsenic  in  laboratory studies.
 Peripheral neuropathy resulted  from exposure to the lowest reported
 effects level of 0.04 mg/kg/day of calcium arsenate ingested for 2-3
 weeks.  Other effects resulting from exposure to arsenate were observed
 at  concentrations of 10 mg/kg or higher; concentrations of 23 mg/kg/day
 to  93 mg/kg/day did not elicit  carcinogenesis in rats.   Data for arsenite
were very limited;  skin disorders resulted from exposure to 0.13 mg/kg/day
 and no evidence of carcinogenesis  was  found in  rats exposed to arsenite
 at  levels  of 13 mg/kg/day.   Data for arsenic (unspecified), in addition
 to  the Taiwan study,  indicated  central nervous  system  damage and lethality
 in  infants exposed to 0.8 mg/kg/day and  7.9 mg/kg/day,  respectively.
 The U.S. Interim Drinking Water  Standard for  tocal  arsenic  is  50 yg/1,
 which is equivalent to  an intake level of 0.001 mg/kg/day  from drinking
 water.   In general the  data  base for the toxicity of arsenic  and its
 compounds  is  very  poor.

      Average  exposure levels of  arsenic  for the general U.S.  population
 through ingestion  are less than  0.001 mg/kg for drinking water,  total
 diet,  and  wine or  moonshine.  Consumption of  contaminated  fish or ground
 water may  result in considerably higher  levels  exceeding  0.01  mg/kg/day.
 However, a low absorption efficiency of  arsenic from ingestion of sea-
 food has been reported.  Dermal  exposure for  the majority  of  the U.S.
 population is expected  to be negligible.  Contact with  arsenic-treated
 wood and contaminated detergent  is  associated with  intake  levels of
 0.004 mg/kg/day and 0.06 mg/kg/day,  respectively.   However,  the sub-
 population exposed through these routes  is  small.   Exposure of the
 general population through inhalation  of ambient levels of  arsenic is
 negligible and less than 0.0001  mg/kg/day in  the vicinity  of  sources.
 Cigarette  smokers  may be exposed to  >0.001  mg/kg day.   Therefore the
 Highest exposure levels for  the  general  population  are from  ingestion
 routes.

      Distinguishing between  chemical forms  of arsenic associated with
 different  exposure routes is complicated due  to a general  lack of data,
 poor analytical techniques, and  the presence  of multiple forms  in most
 media.  However,  some exposure media with potentially high  arsenite
 levels include well water, wine  and  some fish tissue.   Due  Ito  the uncer-
 tainty of  these data, risk is estimated  for exposure to total  arsenic
 levels.  Therefore different exposures levels can be combined  into
 exposure scenarios for  selected  subpopulations  to better represent total
 exposure to arsenic.
                                   7-22

-------
I


                                , has aa
                                 arsenic
than the lo^f                    ke Uvel of
                           »er tan the lo     avall.M                 ve  o  0'4
            any arsenic compound  of 40 u./te/L» ?    ! reported effect level for
            most subpopulatlons have excosur.  T  (PeriP>>«al neuropathy).   I* fact
                A very limited amount of

           7'2>1  Statement of Risk

            -
                                          7-23

-------
 7.2.2  Background
      Chapter 6.0  described  the  effects  levels  of dissolved  arsenic found
 to have sublethal and lethal  effects  on fresh  and saltwater aquatic
 species (6.1)  and the levels  (total arsenic) in  the  environment  to which
 these species  are likely  to be  exposed  (6.2).

      The majority of  mean ambient  levels  of arsenic  in  U.S.  surface
 water by major river  basin  over the past  5 years are less than 0.01 mg/1;
 9% are between 0.01 mg/1  and  1.0 mg/1 and less than  1%  are  greater than
 1.0 mg/1.   Table  7-10 summarizes ranges in arsenic levels that elicited
 effects in  aquatic organisms  under laboratory conditions.   The ranges
 are grouped by chemical form.   Figure 7-1 provides a graphical represen-
 tation of the  same data for total  arsenic.  Even considering the most
 toxic form,  the trivalent (sodium  arsenite), the effects levels reported
 for the most sensitive species  for which  data were available, immature
 bluegill and Daphnia—are only  rarely found in U.S.  waters.

 7.2.3  Local Regions  of Potential  Risk

      Two areas  of  the U.S. were focused on through examination of  a
 minor river basin  data:   the Tacoma, Washington  area and the Missouri
 River Basin (see Chapter 6.2).  The Tacoma site was  chosen because  it
 is  the  location of the only U.S. arsenic producer.  Although the pro-
 uction  plant currently has negligible aquatic releases,  the influence
 of  past  discharges and releases to air and soil on water concentrations
 was considered.  The Missouri area was chosen for consideration due to
 the occurrence  of high mean water concentrations in the  major river basin
 over  the past 10 years.  The Tacoma site indicated no concentrations of
 total arsenic at levels eliciting effects in even the most sensitive
 species as  determined in laboratory bioassays.   Mean sediment levels in
 the river basin were no greater than levels in  other parts of the U.S.
 which suggested that arsenic was not accumulating in the sediment.  These
 conclusions  are based on a limited set of data, however, and therefore
 may be  invalidated by additional data.

      Data for the Missouri Major River basin revealed high levels of
 arsenic, in  terms of aquatic effects,  in two minor river basins located
 in South Dakota and Montana.  In one minor river basin,  concentrations
 reported as  recently as 1979 also exceeded the effects level for a number
of freshwater species.  The presence of  these levels is  possibly due to
 the presence of large fossil fuel combustion plants, weathering and
 leaching of  continental rock,  low stream flow rates at the time of
monitoring,  or other factors.   The data  indicate the likelihood of a
 toxicity problem for aquatic life in this region of the  United States
as indicated by total  arsenic  levels.  To better understand  the actual
 impact of these concentrations on aquatic communities at this location,
concentrations of available, soluble arsenic  are necessary.   Also,  some
qualitative evaluations of the availability could be made if the  concen-
trations of complexing agents  and adsorbing material were known for that
location.
                                  7-24

-------
              TABLE 7-10.  RANGES IN EFFECTS LEVELS
                           AQUAHC SPECIES
       Arsenic Species/
       centr^tion_R

    .Sodium Arsenite

     0.1 - l.o


     1-0  - 10.0
    10.0 - 50.0



  jodiura Arsenate

    1-0 - 10.0


   10 -  200



  Arsenic Trioxide

  i.o -  10.0
 Arsenic
   25
      -  700
   .Effects
   Acutely toxic to bluegill  fin
   and to cladoceran.
                                                  numbers  in  bluegill
             and  to stoneflys.
  Acute effects to most sensitive
                                    Metabolic effects on rainbow trout-
Toxic to white shrimp (Peiiaeus^spp.)


                           freshwater
Source:   Section 6.1,
                                7-25

-------
                           Possibility
                          of Occurrence
^J

CT>
                     Usual
                     Frequent
                     Occasional
                     Rare
                                            Aquatic
                                           Exposure
                                            Levels
                                    Adverse
                                    Effects
                                     Levels
                                    0.0001    0.001
 0.01    f    0.1    |     1.0
      Lowest    Freshwater
     Reported   Criterion for
    Acute Effects  Protection
       Level.  of Aquatic Life
                                                                                          10
100
Concentration
  in Surface
   Water
                                                                                     (mg/l)
                                        FIGURE 7-1
COMPARISON OF AQUATIC EXPOSURE AND EFFECTS LEVELS -
TOTAL ARSENIC

-------
    the
                                           et
              and Western      bnV  it i       ^at  SaSin'  Lower
    hxgher levels were due to hi»h~      ,   Possible that  the
    river basins and individual sttons      •    re
-------
7.2.6  Terrestrial Ecosystems

     Considering the large amount of arsenic deposited onto land in
solid wastes and pesticide application, there would at least appear to
be a high potential for exposure and risk of terrestrial plant and micro-
bial communities.  The potential for mixing and movement of arsenic in
soil (excluding migration in aquifers) is much lower than in water or
air; therefore, the effects of high concentrations of the pollutant in
one spot are usually localized.  This makes it possible to identify
problem sites and avoid cultivation of food or cash crops where contam-
ination or yield reductions could occur.  Areas of potential risk are
spotty and relatively contained, and hence arsenic is unlikely to
spread into areas important in terms of human or wildlife exposure.
This does not, however, apply to arsenic that has migrated vertically
through the soil into ground water.

     In the vicinity of certain areas, either in soils with high natural
background levels of arsenic or contaminated by human activities, arsenic
concentrations are high enough to reduce plant growth or lead to signif-
icant bioaccumulation.  Natural levels of greater than 100 mg/kg have
been reported in soils overlying sulfide ore deposits (NAS 1977).  As
described in Chapter 6.2, soil that has received pesticide applications
(i.e., orchards, blueberry and potato fields), industrial wastes (i.e.,
flyash, municipal wastes, slag) or lying downwind from smelting operations
can accumulate arsenic levels from 150 mg/kg to 25,000 mg/kg (total As).
Comparing these exposure concentrations to the effects levels for crops-
reported in Chapter 6.1 of 25 mg/kg to 85 mg/kg, it is evident that
growth of the species in which these effects were measured, and probably
most other plants, would be severely reduced in the situations described
previously.

     Most plant species are so sensitive to arsenic that it is thought
that tissue levels high enough to harm humans and wildlife ingesting contami-
nated food are unlikely to accumulate in the edible fraction before the plant
itself is subject to growth reduction (Woolson 1969).  Exceptions to this obser-
vation may include root crops, especially potatoes that accumulate high
concentrations in the peel, leafy plants, and perhaps plants treated
with organoarsenicals, which are more effectively translocated than the
inorganic forms into the edible part  (Walsh and Keeney 1974).  Organo-
arsenicals are applied to soil at lower rates than are inorganic forms,
however, and  this would subsequently  reduce uptake.  The monitoring data
indicate generally  low levels  of arsenic of all forms in fruits  and vege-
tables, even  root crops.  The  exposure  of humans to contaminated food
crops  is discussed  in Chapter  5.2.
                                   7-28

-------
I
                  It is more difficult to estimate the potential risk of microbial
             populations than for plants living in soil with high arsenic levels.  This
             is due to the limited information available regarding inhibitory levels
             for microbial activity in other than simplified laboratory systems.   In
             many of the arsenic-contamination situations described above, however,
             the concentrations are probably high enough to inhibit organic matter'
             turnover rates, nitrogen and other element cycles,  and the cycling of
             arsenic itself.  These situations are limited and localized enough so
             that the impact would not be great on an ecosystem level.   The impact
             of lower levels of arsenic,  which are more widespread, or  of the various
             chemical forms of arsenic,  is unknown.
                                             7-29

-------
                              REFERENCES

 Arthur D.  Little.   Methodology for exposure and risk assessment.
 Mathematical appendix.   Washington,  D.C.:  U.S.  Environmental Protection
 Agency; 1980.

 Carcinogen Assessment Group (GAG).  Final  risk  assessment on arsenic
 Washington, D.C.   Office of Health and Environment Assessment, U.S.
 Environmental Protection Agency;  1980.

 Cutler, S.J.; Scotto, J.;  Dezesa,  S.S.;  Connelly,  R.R.   Third national
 cancer survey - an overview of available information.   J. National
 Cancer Institute.  53:1565-1575;  1974.

 Irgolic, K.  Speciation  of arsenic in  water supplies.   Q. Prog.  Rep.
 November 1, 1978-January 31;  Washington, D.C.:   U.S.  Environmental
 Protection Agency;  1979.

 Lambou, V.; Lim, B.   Hazards  of arsenic  in  the  environment,  with
 particular reference  to  the  aquatic  environment.   Washington,  B.C.:
 Federal Water Quality Administration,  U.S.  Department of  the Interior-
 1970.

 Lu, F.J.,  et_ _al.   Studies  on  fluorescent compound  in drinking  water of
 Blackfoot  endemic  areas.   I.  The Toxic effect of fluorescent compound
 on the  chick embryos.  J.  Formosan Med. Assoc.  76:58; 1977.   (As'cited
 in USEPA 1980).

 Morton, W.;  Starr, G.; Pohl,  D.; Stoner, J.; Wagner, S.;  Wesnig, P.
 Skin cancer and water arsenic in Lane  County, Oregon.  Cancer
 37:2523; 1976.

 Mushak, P.;  Galke, W.; Hasselblad, V.; Grant. L.  Health  assessment
 document for  arsenic.  Research Triangle Park, N.C.:  Environmental
 Criteria and  Assessment Office, U.S. Environmental Protection Aeencv
 1980 Draft.

 National Academy of Sciences  (NAS).  Arsenic.  Washington, D.C.:
 National Academy of Sciences; 1977.

 Sandberg, G.R.; Allen, I.K.  A proposed arsenic cycle in  an  agronomic
 ecosystem.   Arsenical Pesticides.   ACS Symposium Series 7:124-147;  1978.

 Tseng, W.  Effects and dose-response relationships of skin cancer and
Blackfoot disease with arsenic.  Environ. Health Perspect. 19:190;  1977.

U.S.  Environmental Protection Agency (USEPA). Ambient water quality
 criteria for arsenic.   EPA 44015-80-021.  Washington,  D.C.: Office
of Water Planning  and Standards,  U.S. Environmental Protection Agencv
 1980.                                                                "
                                  7-30

-------
                *" ""' *™r      Heavy Met.
                             -  C*. c



     ----- ^.r-^s TT
cited in USEPA 1980)       National Cancer Inst. 10:81;
             7-31

-------
                                 APPENDIX A
   Note 1

   Atmospheric emission estimates are based on EPA 1979a and 1979b.
   Uncontrolled fugitive emissions of arsenic from the ASARCO, Tacoma
   smelter are calculated from the following emission estimates and 365
   day per year operation, 16 hours per day.

         Fugitive Emissions                                    As kkg/yr

Calcine transfer from roaster   11.2 Ib/hr x 365 d/yr x 16hr/day   30
Transfer, flue dust, handling    3.8 Ib/hr                         10
Arsenic building                 1.8 Ib/hr                         _J5
                                   total fugitive emissions        45"

         Controlled Emissions

From stack                      62.4 Ib/hr x 365 d/yr x 16hr/day   166
                                   total emissions                 210

   Arsenic is associated with other emissions  at the Tacoma smelter;
   these are included in inadvertent sources - copper production since
   they are not directly involved with  ^2^2 recovery.

   Note 2

   Based on EPA 1979a/b and  EPA,  1975.   Slag at  ASARCO,  Tacoma  is  dumped
   and air cooled  instead  of granulated.  It is  assumed  to  be disposed to
   land,  although  it can be  used  as  road construction material.   Land-
   destined wastes  are  calculated as  follows:

   Reverberatory furnace slag:  450 Ib As/hr  x  365 d/yr  x 16 hr/day =  1195
   metric  tons/yr.   Arsenic-containing  solid wastes  are  generated  at
   other locations  in the  ASARCO,  Tacoma facility.   These are included in
   inadvertent  sources  - copper production as  they are  not  directly  a
   part of As203  recovery.
                                   A-l

-------
Note 3

Based on total pressed and blown glass manufacture of 2,993,100 kkg
per year (EPA, 1980b).  About 777, of that (2,304,690 kkg) is soda  lime
glass which presently does not use arsenic as an additive (EPA, 1980b;
Thatcher Glass Co., 1980).  The remaining 688,410 kkg consists of  11%
borosilicate glass, 5% lead glass, and 7% opal glass (EPA, 19805).   If
25% of the borosilicate production involves arsenic, as well as all  of
the lead and opal glass, 634,900 kkg of pressed and blown glass
produced annually contains arsenic (EPA, 19805).  Based on a
controlled emission factor of 0.015 kg As emitted per kkg of pressed
and blown glass produced, 10 kkg of As are emitted annually from glass
manufacture.

Note 4

Less than 1 kkg of arsenic is discharged to water (chiefly POTWs)  from
wood preserving plants annually.  Based on raw waste concentrations
and flow rates from plants using organic preservatives only and a
total of 476 plants (AWPA, 1979) approximately 0.2 kkg of arsenic
would be discharged (Table A-l).  This figure represents a maximum
since the number of facilities using organic preservatives only was
not available.

Plant average arsenic loading kkg/yr x total  #of plants = As discharge
0.0004 kkg/yr/plant x 476 plants = 0.2 kkg

Alternatively, using the average discharge for plants treating with
organic and inorganic preservatives, the annual  discharge of arsenic
would be 0.5 kkg per year (Table A-2).  Both of these estimates are
raw waste concentrations; wastewater treatment would lower these
discharges.  Most plants are already achieving zero discharge.
                                A-2

-------
 Note 5



 A.   92,903,000  ft3 wood  treated ,




     treated  =   1,486 yd3 wet sludge/yr.
E.  -                           m  wet  8ludge
           -                L137 . 1, 171 kg wet sludge
   (dry  sludge).
                                                • 0.5 . As
                              A-3

-------
Note 6

Mo data were found concerning wastewater discharges from  individual
smelters; EPA, 1975a addresses the primary copper  industry  as  a
combination of smelters and refiners.  Therefore a total  arsenic
discharge for the entire primary copper industry was obtained  by
calculating an average As discharge per plant and  applying  that
average to the total number of facilities.  Assume 350 d/yr operation,
          Effluent Loading
PI ant    (kg/kkg production)

 115           0.026
 116           0.0003
 117           0.174
 118           0.0007
 121           0.0002
 102           0.003
 110           0,0001
Production (kkg/day)  As discharge kkg/yr
         494
         415
         293
         454
         263
         311
         674
                                                    average
Currently, there are 16 copper smelters and 11 refineries in the U.S.
Four of the refineries are located on the same site as the smelter.
Assuming that these 4 refineries combine their wastewater with their
respective smelters, and discounting the ASARCO, Tacoma facility
which is included in arsenic production, a total of 19 plants would
be discharging 3 kkg As/yr or a total industry discharge of 60 kkg
As/yr.  However, most smelters and refiners are at or approaching zero
discharge.  Based on 50% of plants applying treatment and/or
recycling wastewater, 30 kkg As would be discharged per year.  This
discharge is usually sent to tailing ponds EPA, 1975a.
                                A-4

-------
 Note 7

 Based on effluent  loadings  of 8 secondary  copper  facilities  and  applied
 to the 50 total plants.  Assume 350 day/yr  operation.
          Effluent Loading
 Plant    (kg/kkg production)

   1         0.0009
   8          	*
   9          	*
  12         0.0001
  26          -0-
  32          -0-
  39         0.000001
  43          	*
               Production  (kkg/day)  As discharge kkg/yr
                        529
                        327
                         47
                         78
                         33
                        150
                         44
                         62
                             0.17
                             0.01
                             0.001
                             0.003
                              -0-
                              -0-
                             0.00002
                             0.002
                                                  average   0.02
*For effluent loadings not listed, 0.0001 was used.
 50 plants x 0.02 kkg As/yr plant = 1 kkg As discharged

 Note 8

 EPA, 19795.  Based on raw wastewater As concentrations and flow rates
 for 4 primary lead smelters.  The average was applied to the total
 plants.  Assume 350 day/yr operation.
 Plant

   A
   B
   C
   D
As mg/1

 .130
 .018
 .093
 0.10
Flow I/day x 350 d/yr

      2.4 x 10g
      4.5 x 10C
      8.3 x 10
      4.9 x 10
As kkg/yr

  0.1
  0.03
  0.3
  0.02
                                                 average
                                              0.11
 7  plants x 0.11  kkg As/yr/plant = 0.78 kkg As
                                   A-5

-------
Note 9

EPA, 1979b.  Based on treated wastewater As concentration and flow
rates for secondary lead smelters.  The average per plant was applied
to the total of 69 plants.  Assume 350 day/year operation.

Plant      As yg/1           Flow I/day          As kkg/yr

  A       2.9 x 103         0.015 x 106        1.5 x 101
  C           10              0.01 x 106       3.5 x 10"5
  D           25              0.11 x 106       9.6 x IP"4
                               geometic mean   8.0 x 1Q~3

69 plants x 8.0 x 10'3 kkg As/plant/yr = 0.6 kkg yr

Note 10

EPA, 1978a, and Plunkert, 1980.  Antimony is produced by either  i
pyrometallurgic or electrolytic processes, each of which generate a
specific type of arsenic-containing waste.  If half of the total
2,400 kkg of metallic antimony is produced by each method, then:

Pyrometallurgic:
2,800 kg slag^ 1,200 kkg antimony produced     0.003 kg As in slag = 10 kkg As
kkg antimony                                        kkg slag
produced

Electrolytic:
210 kg solids (in sludge)    1,200 kkg antimony  0.016 kg As = 4 kkg As
     kkg antimony                                    kkg

Therefore,  a total  of 14 kkg As of land-destined wastes are generated
from antimony  manufacture.
                                 A-6

-------
Note 11

zinc processing facimie     are    P
average treated wastew  er'd scharue ler
concentrations and flow rates from 3 DlantP
operation.                         ° plants.
    B
    B
    C
    A
    A
   A
   B
   8
   D
   C
   E
   E
   E
          flow
                2.15
                1.59
                1.63
                0.09
                1.81
 60.6
 1.4
 1.2
 6.0
21.9
0.32
36.3
 0.07
As rng/1

  0.92
  0.002
  4.8
  0.001
  0.001
                                                    only since all
                                              (EPA, 19795).  Of the 6

                                                      l Is 1ndirect'
                                              -        "]ated f™ As
                                              Assume 350 day/yr
                                                    As  kkg/vr
 0.7
                                         average
2.7
0/00003]
.0.0006 r
in
                                 As mq/
                                  0.01
                                  0.05
                                  0.01
                                  0.035
                                  0.01
                                  0.001
                                  0.019
                                  0.007
                                                  _As kkg/vr
21
 4T
      0.024
                                      average
27 direct dischargers x 0.12 kkg As/plant/yr . 3 kkg As
                  0.2
                  0.02
                  0.00
                  0.07
                  0.08
                  0.0001-
                  0.24    0.24
                  0.0002J
                  0.12
    1-j
     L
    J
                                 A-7

-------
 Note  12

 Of the 55 domestic copper mines, 10 discharge mine water to  surface
 waters (based on EPA, 1975c survey of 21 mines, 4 of which discharged
 mine  waters) the remainder recycle/reuse such water in milling
 processes (EPA, 1975c).  Half of the dischargers treat the water  prior
 to discharge.  Assuming treatment efficiency of 50% and 350  day/yr
 operation, 6 kkg As/year are discharged from copper mines, based  on  As
 concentrations and flow rates for 2 mines  (EPA, 1975c).

 Mine          Flow (1Q6 I/day)       As my/1
 2119            42.0                  0.07
 2120            27-.S                  0.07
                                            average


 .50 efficient x 0.85 kkg As/yr/mine x 5 mines = 2 kkg As
 0.85 As/yr/mine x 5 mines (no treatment) = 4 kkg As

 Copper mill  wastewater is discharged to tailing ponds for settling;
 water is then recycled within the plant.  Based on raw waste loadings
 for tailing pond influent, and a total  production of 62,180,500 kkg  in
 1979  (Butterman, 1980) approximately 640 kkg of As are discharged to
 tailing ponds per year (EPA, 1975c).

                                    annual
         raw waste loading         concentrate         discharge
Mill   kg As/1000 kg Cu produced   production kkg       kkg As/yr

                                    454,420

                                    223,318
                                    740,602
                                     11,170
                                     69,362


 10.3 kg As/1000 kkg Cu produced x 62180.5  x 103 kkg  Cu produced
640 kkg As.
                                  A-8

-------
                                             .i  - to surface  water.
  Note  13



  sEjnelting6of manganese! iTis^ssuiTd'3016 concerni?9
  phosphorus: 60%  remains in f^l^L^ JL^!?"1? '
 carbon  zinc


 Note  14
 kkg sent  to  land  s  lag   A  97% clulctl"^"9-I"  ferr0"^ ^  =,080


             '         e
produced in 1976  is sirailar to
       contained  ,„ ^r                                   "'
!oan«nt°?at?oJI ofl80t™/{yr °f  bn'ne processed  a"d "" arsenic
coSIlSS   ^niWisffir0*™^'* 2.fO  kkg of arse^c are
lake from which the'9,"6131"5'-'"".^^ returned to the
                                 A-9

-------
Based on an annual production of 110,000 kkg, 4 kkg of arsenic is
discharged per year from boric acid production.  The specific aquatic
sink, surface water or POTW, for this discharge is unknown.

Note 15

Based on emission estimates from Davis and Associates, 1971, and
Bureau of Census, 1979.  Particulate emissions from cotton gins
average 5.3 kg/bale cotton; average arsenic concentration of
particulates in 300 mg/kg.  Therefore 5.3 kg particulate/bale x
10,549,219 bales x 300 mg As/kg particulate = 17 kkg As.  Trash from
ginning operations is either burned (37%), disposed on land (58%) or
handled in an unspecified manner.  About 76% of the arsenic in trash
(assumed to be 300 mg/kg as is in particulates) is emitted to the
atmosphere.  Therefore:

 1,237,318 kkg trash burned x 300 g/kkg x 76% emitted = 281 kkg As.
 1,939,579 kkg trash disposed to land x 300g As/kkg = 581 kkg As.
 Total  As to air = 298 kkg
      As to land = 581 kkg

Note 16

Arsenic is added to lead shot, bearings and batteries.  Refining
temperatures (260-270°C) are low compared to vaporization temperature
of arsenic (615°C) EPA, 1980b.  EPA, 1976a estimates 0.75 kkg As is
released from all nonferrous alloy production.

Note 17

EPA, 1976a.  Of the total rock produced, 49% is used as domestic
fertilizer, 4% as aniinal feed, 5/» used for detergents, 3% for food
products, 3% miscellaneous; the remainder is exported.  Quantities of
                                A-10

-------
            ™
                              "


  Note 18
            ,        15 x
production.  No arsenic
assumed to be  slmi"? to th
uncontrolled       '
                                          9     used ln cast iron-
                                       ^ 1S USed 1n cast ir°"
                                      -WS f°und' the^fore it  is

                       facto  o^^A /^/Sal  "MV
 of arsenic would  be  emitted  from the 19 4 x fn§ S«   '  .ab°Ut 136
 production.   Control effirieJ.?!*               kkg  cast  1ron
 approximately ?S kkg  oParsen   «~%aSS!!?e d t0  be  95%'   Therefore,
 devices  and  eventuaffy landff? ed?7hf 79ht  '"  P°Hution  c°ntrol    '
 the  atmosphere.        '^filled,  the 7  remaining kkg  are emitted to
                                           °f
                                        °Xy3en steel ^rnaces are
                                                    If the
       kkg, 21 ,770 kkq   nd i'wn
content O?fpart1c2l5t1i  s
             mg/kg)  then
                                                                 are
Open hearth  -   27,570,000 kg x 400 mg/kq = H
                                  42 - "

                               A-ll

-------
   Steel making dusts are assumed to be  landfilled.  Assuming  collection
   efficiencies of 95%, the emitted particulates  should  be 5%  of  the
   total.  By difference 5,510 kkg As are  landfilled as  steelmaking
   dusts:

   63840 = 0.05 y
   y = 1,376,880 kkg particulates trapped
       x 400g/kkg = 5510 kkg As landfilled

   Arsenic contained in steel making slag  is estimated to be 34,600 kkg
   per year.  Most of the slag is used as  railroad ballast, aggregate for
   concrete or recycled within the plant; the arsenic is thus  considered
   to be contained and not released to the environment.  Additionally,
   about 14,600 kkg of arsenic are estimated to be contained in steel
   production (EPA, 1976a).

   Wastewaters from iron and steel plants stem from coke ovens, acid
   pickling, and casting, machining and hot and cold working of steel
   products.  Based on EPA, 1979f data, ammonia liquor waste streams and
   cooler blow down contribute 42 kkg As per year in raw wastewater,
   assuming plants operate at 80% of capacity.

                                              total  industry
   Operation     mg As/1      flow (I/ton)     production kkq   kkg As

cooler blow down  0.005         118             3,430,631!      0.002
NH3 liquor       85.3           144                           _42	
                                                        total  42
   Of that 42 kkg As, 14 kkg are directly discharged, 11 kkg are
   discharged to POTH's, 17 are recycled with quench water.  Since
   quenching water is eventually evaporated, the As contained in it is
   assumed to be entirely emitted.  These figures represent a worst case
   scenario as they are raw wastewater.  Assuming 50% treatment
   application and efficiency for discharges, 7 kkg would be directly
   discharged, 6 are sent to POTWs and the 12 kkg collected would be sent
   to land.
                                    A-12

-------
 Note 19
 _Note 20

 Natural  fading estiamtes based on the folding data (as  cited  ,„  £


 Worldwide quantity of material eroded-  9 3 »  in  1
 Portion of that tota, attrioutab,e°1o"tne%3 J JV     5.86 x 108

Therefore:
     x  5.86 x
arger,  or 11,400  kkg/yr
                        x 1.5 mg/kg  =
                                            kkg/vr
                                      2,850 kkg/yr
                                               in  rele"es 4 times
                               A-13

-------
Note 21

Based on the following As concentrations (ppm) measured in urban runoff
in various U.S. cities (EPA 1979J):

SanJose CA     <0.01
Orlando FL      0.04
New Creek, NY  31-14
Seattle, WA    0.049-0.060 T = 0.050
               0.05-0.1    T= 0.05
               0.05
               0.04-0.1   T= 0.05
               0.045-0.070"= 0.05
               0.049-0.060 T= 0.05
               0.050
                              A-14

-------
i—*
en
                       Table ""•  s
        Plant
           67

          267
          591


        1,100
                                  Flow (liters/yr)a
 10,361,437


 12,081,720


 10,862,950


 4,636,625


82,796,875
                                     -- - __

       a)  Based on 350 day/year operation.


       ">  "ISter va,ue ,,-sted  used In c.lcu]atton to


       Source:   EPA,  1979d.
0.093


0.009


0.086


0.006

                                                                                      average
                                                                                                     As/yr
 0.00007


 0.001


 0.0001


 0.0004


 0.0005


0.0004

-------
                   Table A-2.  Raw Waste Loading for Wood Preserving Plants Using
                              Organic and Inorganic Preservatives
Plant #
65
237
335
499
582
897
1 ,078

Flow (liters/yr)
2,914,450
2,649,580
2,252,075
132,475
18,215,312
56,169,400
20,268,675

As (rng/1)
0.014
0.050
0.250
1.00
0.040
0.130
0.003
average
kkg As/yr
0.00004
0.0001
0.0006
0.0001
0.0007
0.007
0.00006
0.001
a)  Based on 350 day/year operation.

Source:   EPA, 1979d.

-------
Table A-3.   Arsenic  Releases
Source
                                                       Prjmary  Copper
                                      Production
                                Jnput         Contained
                                                                  . .
                                                                  A7r
                                                       Environmental Releases
                                                              Surface
                                                               Water      POTW
ASARCO
El Paso
Hayden
KENNICOTT

Hayden
Hurley
McGill
Garfield
PHELPS DODGE
Douglas
Morenci
Hidalgo
A jo
Magma
Copper Range/White
Pine
Inspiration
Cities Service/
Copperhill

Anaconda
TOTAL
— -_ — ________ 	
Footnotes next pane.

480a
400e

r A 1
64
2 III
64 p
4.
1,100C
80 X
r»^bh
80
160ff
540 JJ
oo nn

6rr
\j
61 vv
w
HGCJ
6,700aaa
9,800

— — — — 	 	 _
	 	 	 	 	 	
240b 4Qc 22Qd
80 2109 120n

,
ne9J 36k 401
neg11 ,. o
0 s ne9
2 58r oS
1,000 fiv 8oW
*yy 7
t. O "7 QG
o/ /1H
•?cc Hri
j onUU pp
JU 4R
'hh >»1(
31 310N 220mm
neg00 2sPP nqq
SS 4.4.
neg" 4tt Uu
oWW v
6 r VV YY
6 54 **
22 77
».«»bbb ISO-' 5,100^
3)000 960 6,100 30eoe

~~ 	 	 	 	 	 	 	 	
v • '• i u ca i
• " i - . — -- , , 	
' — 	 --^_- .
260
330


76
c
t>
64
88

85
78
156
530
36
6
60


0
5,300
7,100


-------
                                              Table A-3.   (Continued)



   a)   Rased on .06 kkg As fed/hr (EPA,  1979a),  8000 hr/year operation (EPA, 1979a).  All  air and land releases
       from EPA,  1979a and are based on  8000  hr/yr operation.   Totals  may not add due to rounding.

   b)   Based on 0.03 kkg As in blister copper and  dust  returned to lead smelter.

   c)   Based on 0.004 kkg As to stack/hr,  and 0.9  kg As/hr fugitive emissions.

   d)   Includes slag that is dumped  and  acid  plant sludge.  Based  on 0.027 kkg As/hr.

   e)   Based on 0.05 kkg As/hr input.

   f)   Based on 0.01 kkg As in blister copper and  material sent to El  Paso/hr.

   g)   Based on 0.023 kkg As to stack/hr,  and fugitive  emission of 3 kg As/hr.

:> h)   Based on 0.015 kkg As to dump in  slag  and as acid  plant  sludge/hr.
K-•
00 i)   Based on 0.008 kkg As input/hr.

   j)   Based on 0.045 kg As/hr in blister  copper.   Negligible  is defined as  <1 kkg.

   k)   Based on 0.0035 kkg As  to stack/hr,  and fugitive emission of 1  kg As/hr.

   1)   Based on 0.005 kkg As in slag and acid plant sludge/hr.

   m)   Based on 0.0002 kkg As  input/hr.

   n)   Based on 4.5 x 10~6 kkg As leaving  in  blister copper/hr.

   o)   Based on 0.13 kg As to  stack/hr and  fugitive emissions  of 0.5 kg As/hr.

   p)   Input 0.008  kkg As/hr.

   q)   Based on 0.0002 kkg As/hr leaving in blister copper.

   r)   Based on 0.007 kkg As to stack/hr,  and fugitive  emissions of 0.3 kg As/hr.

-------

                                              Table  A-3.   (Continued)
   s)   Based on 0.001  kkg  As  to  dump  as  slag.
   t)   Based on 0.14 kkg As  input/hr.
   •0   B-o,  „„ o.,3 kkg As .eaving ,„ Mftter copper and ^ ^ ^
   v)   Zero emissions  fro™ stack EPA, ,979.. fugitive cession of , kg As/hr
  -)  ^ed on 0.0! kkg As/,, sent to du,,,p ,„ slag and ,„ „,„ p]Mt  waste_
  x)  Input of 0.01 kkg As/hr.
  y)  Ba,ed on 0.0002  kkg  As/l,r  tea»ing  ,„  H1ster copper.
  ')  Base,, „„ 0.00, «9 As to stack/hr>  fugitive  Missions of 0.6 kg As/,,r
  aa,  eased  on 0.006 kkg As sent to du,,,p  as s,ag and acid p.ant sllldge.
  bl>)  Input  of 0.01 kkg As/hr.
  cc) ,.ased on 0.0004 kkg As/hr .eaving ,„ Mister copper.
 *0 Based on 0.003 kkg As sent  to stack/,,r,  fug,tive  missions  of 0.8 kg As/hr
 ee) Based on 0.006 kkg As sent  to dump  as  s,ag  and  acid  p,ant  s,udge.
 ff) Input of 0.02 kkg As/hr.
 89) Based on  0.0009 kkg As  ,eavina/hr ,„ olfster  copper
 •*) aased on  0.0003 kkg As/hr se,,t to stack, fugitive emissions of 0.4 kg  As/hr
 ") "ased on  0.0,9 kkg As dispose,, as s,ag and acid p,ant S,ud9e/hr.
jj) Input of 0.068 kkg As/hr.

-------
o
                                            Table A-3.   (Continued)

 kk) Based on 0.0039 kkg As/hr leaving in blister copper.
 11) Based on 0.037 kkg As/hr sent to stack, fugitive emission of 2 kg As/hr.
 mm) Based on 0.028 kkg As/hr sent to dump as slag and  acid plant waste.
 nn) Input of 0.004 kkg As/hr.
 oo) Based on 0.0001 kkg As/hr leaving in blister copper.
 pp) Based on 0.0029 kkg As/hr sent to stack, fugitive  emissions of 0.2 kg As/hr
 qq) Based on 0.0014 kkg As/hr sent to dump as slag and scrubber waste.
 rr) Input of 0.0007 kkg As/hr.
 ss) Based on 0.00005 kkg As/hr leaving in blister copper.
 tt) BAsed on 0.41 kg As/hr sent to stack, fugitive emissions of 0.05 kg As/hr.
 uu) Based on 0.23 kg As/hr sent to dump as slag.
 vv) Input of 7.6 kg As/hr.
 ww) Based on 0.4 kg As/hr leaving as blister copper.
 xx ) Based on 6.7 kg As/hr sent to dump as slag and scrubber waste.
 yy) Input of 0.045 kg As/hr (0.4 kkg/yr).
 zz) Distribution of arsenic not performed.
aaa) Input of 0.84 kkg As/hr.
bbb) Based on 0.2 kkg As/hr leaving in blister copper and  stored as dust.

-------
                                                  lable A-3.  (Concluded)

     ccc)  Based „„ 2 kg As/hr sent to ^  ^^  ^^ ^ ^ ^


          -ased „„ 0.« kkg As/hr sent to du,,,p as sl.g  and  as  $Iudge to ,)ond


                                                I9 P,ants and  m,  1975a control/treatlllent
I
ro

-------
       Table  A-4.   Production, Conversion, and Consumption of Phosphates


Florida Rock Produced
Used as Domestic Ferti-
lizer
Used for Animal Feed
Exported
Tennessee Rock Produced
For Non agricultural
Uses
Western Rock Produced
Used as Domestic Ferti-
lizer
Used for Animal Feed
Exported
For Nonaaricultural
Uses
Total Rock Produced
Total Used as Domestic
Fertilizer
Total Used for Animal Feed
Total Exported
Total for Nonagricultural
Uses
Used for Detergents
Used for Food Products
For Miscellaneous Uses
P~0r Quantities,
£ 0
Metric Tons/Year
15,728,300

8,493,282
552,700
6,569,250
617,670

617,670
1,847,559

331,610
110,540
110,540

1,284,870
18,193,530

8,824,890-
663,240
6,679,790

1,912,540
905,940
503,300
503,300
Arsenic Quantities,
1 Metric Tons/Year
702

382
25
295
51

51
425

76
25
25

298
1,178

458
50
320

349
160
86
86
Source:   EPA,  1976a,  based  on  1976  production quantities
         add due to  rounding.

                                   A-22
Totals do not

-------
           Table A-5.
Industrial
Detected3
                                 Wastewaters  in which
   Arsenic has been
  Industry
 Adhesives/Sealants
 Auto/Other Laundries
 Coal Mining
 Electricial
 Foundries
 Gum/Wood Products
 Inorganic Chemicals
 Iron/Steel
 Nonferrous  Metals
 Ore  Mining
 Organics  and  Plastics
 Paint/Ink
 Pesticides
 Petroleum Refining
 Printing/Publishing
 Pulp/Paper
 Soap/Detergents
 Textile Products
Timber
           # of Times b
            Detected
                2
               13
               31
                2
               23
               11
               18
             145
              91
              39
              75
              18
               7
              17
              26
              13
               3
             24
             40
                                                   of Samples
                                                    Taken
  11
  45
  94
   2
  54
  18
 107
 414
 146
  64
 557
 149
 104
 346
 66
 44
 20
121
261
                 Frequency of
                  Detection
  18
  29
  33
100
 43
 61
 17
 35
 62
 61
 13
 12
  7
  5
39
30
15
20
15
                                     Photographic, phosphates and plastics
    False  positives  are  accepted.   Detection  limit:   I0yg/l.
                                      A-23

-------
            Table A-6.  Municipal  Disposal  of Arsenic,  1970 (kkg/yr)
Source input
POTWa i.800b
Urban runfoff
URBAN RCFUSE
Incineration r/oe
Landfill 3.000h
Air Water Land
negc l ,800b 20d
1.050

8-7 neg9 J60f
3 ,000h
 a)  Publ icily-owned treatment works.  Urban runfoff contribution to
     POTUs is based on average arsenic concentration of 50 ug/1 and a
     total  nationwide flowrate of 2.1 x 10 3 ,/yr.  See A   *djx A
     Note 21  for derivation of arsenic concentration.  El'A 1979j.

 b)  Assume median  effluent 50 Mg/|,  influent 50 Pg/l (see Table A-/)
     ion |/day total  POTW  flow,  EPA. 1978c.                         ''

 c)  Atmospheric loss  assumed  less  than I  kkg.

 d)  Uased  on 6 x 106  kkg dry  sludge  produced/yr (EPA  1979i)   wet
     sludge = 95X water  (by weight),  170 Mg/l as in  wet  sludge (see
     I do 1 6  A-"/ ) •

 e)  Uased  on 17 g As/kkg in combustible fraction  107 kkg
     incinerated yearly.  U.S. Dent of  Interior. 197U; Gordon.  197U.

 f)   Uased  on 0.8 kg As  (suspended participates) anitted/920 kkg  solid
     waste  incinerated. 0.8 kg/920 kky  released  to land  in  flyash     he
     remainder  of the  land  load assumed to come  from  bottom ash
     0  kkg  incinerated yearly.  Gordon, 197U.  Greenberg el aj.,
     1978;  Law  and Gordjn, 1979.                                 '
    Gordon'1     '911*1*' by ina]o^ to other •**»> n«tals. law and
h)  2 x 10   kkg municipal  solid waste. 87* landfilted, 17 ppm As.
    Geswein  1980;  Alvarez. 1900; U.S. l)ept. of Interior. 1978;
    (.ordoti  1978.   17 ppm is concentration in combustible fraction
    assumed to represent  a  minimum for the noncombustible fraction!

-------
Table A-7.   Distribution of Arsenic  in POTWs and Sludge:
Plant
* Industrial
Contribution
1
9
3
4
5
6
7
8
9
10b
> 11
fv> 12
m 13
14
15
16
17
ia
19
?0
Mean
Median
flow Weighted
30
<5
10
10
38
15
10
10
5
18
50
33
25
25
16
25
50
20
15-22


Mean
Average Flow
(106 I/day)
400
30
42
320
83
27
190
87
200
87
160
150
64
27
550
57
240
260
450



                  ,

c)  Not detected.

Source:  EPA,  1980d.
                                                             Influent
                                                  Concentration
                                                      d'9/1)
                                                      <50
                                                        3
                                                      <50
                                                      <50
                                                      <50
                                                       <3
                                                      <50

                                                       NDC
                                                       29
                                                     <50
                                                     <50
                                                       4
                                                     <50
                                                      26
                                                     <50
                                                     <50
                                                     <50
                                                       9
                                                      31
                                                      50
                                                      32
Loading
(kg/day)
— • 	 • — 	 	
<20
<1 .5
0.1
<4.2
<1 .4
<0.6
<4.4
0.4
NO
4.6
<7.5
<3.2
0.2

1' . O
O.I
0.3
0.8
1.1
0.5
0.2




-------
      Table A-8.  Industries Contributing to POTWs in Table A-7







          Plant                            Industry3








            1                     PH, PE, PT, FN, CO, F



            2                     GS, OFT, MT, MW



            3                     POP, PL, T



            4                     BV, PL, PI, CH, F, PA, PP



            5                     ATO, H, PT, PA, PP



            6                     PL, FN, BK



            7                     PT, ATO, FURN



            8              .       ATO



            9                     FR



           10                     E, PT, MW, PL



           11                     E, PT, F



           12                     PT, F, PH



           13                     PT, MP, BA



           14                     PL, PT, PI, PH



           15                     AP, PL



           16                     F, MF



           17                     PI, PT, AR



           18                     BR, PT, CA, S



           19                     F, PT, L



           20                     CA, E, PI, D
Footnotes next page.



                               A-26

-------
                       Table A-8.  (Conduced)
a)  Code * the following:

         Aircraft Manufacture* AR
         Appliance Manufacture = AP
         Automobile Manufacture ATO
         Baking = BK
         Battery Manufacture  = BA
         Beverages = BV
         Breweries = BR
         Canneries = Ca
         Chemicals = CH
         Coking  =  CO
         Detergent  Manufacturing  =  D
         Electronics  = E
         Firearms  Manufacture  = FR
         Foods = F
         Foundries =  FN
         Furniture Manufacture = FURN
        Grain Storage * GS
        Hospitals = H
        Leather Finishing = [_
        Machine Tools = MT
        Meat Packing = MP
        Metal  Finishing = MF
        Metal work = MW
        Oil/Fuel  Terminals =  OFT
        Paint  and Ink =  PI
        Paper  =  PA
        Petrochemicals =  PE
        Pharmaceuticals =  PH
        Photo  Processing  = PP
        Plastics =  PL
        Plating =  PT
        Poultry processing  POP
        Slaughterhouse = S
        Textiles = T
                             A-27

-------
                   Table B-l.  Arsenical Pesticide Producers and Locations
  Pesticide


 MSMA
 (monosodium
 methanearsenate)


 DSMA
 (disodium
 methanearsenate)


 Arsenic Acid
 Cacodylic Acid
10,10-OBPA
(10,10 -oxybisphenoxarsine)
Lead Arsenate
Calcium Arsenat
>aris  Green
         Use


  herbicide3
  citrus fruits
  cotton
  crabgrass

  herbicide9
  citrus fruits
  cotton
  crabgrass

  Defoliant
  cotton


  herbicideb
  lawn control
 Johnsongrass

 fungicide0
 vinyl  plastics
 jnsecticide
 turf
 ornamentals

Insecticide
cotton
 vegetables

Insecticide
mosquito larva
        Produced
 Diamond Shamrock
 Vineland Chem. Co. Inc.
 W.A.  Cleary Corp.
 Diamond  Shamrock  Corp
 Vineland  Chem.  Co.  In.
Los Angeles,  Chem. Co.
Osmose Wood Preserving Co,
Pennwalt Corp.

Vineland Chem. Co. Inc.
Ventron Corp.
    Locationd
 Greens 3ayou, TX
 Vineland, NJ
 Somerset,  NJ
 Greens  Bayou, TX
 Vineland,  NJ
 South Gate,  CA
 Memphis,  TX
 Bryan,  TX
     i
 Vineland, NJ
Beverly, MA
                                                Los  Angeles  Chem.  Co.
                            South Gate, CA
 )  Directed application, no contact with crop plants  (EPA,  I976a).

<)  Used only  in non crop areas  (EPA, 1976a).


                                                      '"PS. WClf1c.ny Florida
'
                                        B-l

-------
CO
ro
                                           25% No3A4O3
                                           Slotogo
              H2S04
                                           Melhylortonic
                                           Acid Unit
                                                                             DSMA Salts
                                                               Aqueout
                                                               CH3OH
       B/-Pioducl
       Salts
                                                     1
H20
                                                  CH-.OH
                                                     «j
                                                Recovered and
                                                mod «hewlier«

                    NoCI
                                                                              •Liquid     To APP«>v«d
                                                                                 I       Land Fill
                   Figure  B-l.   Production  Schematic for  MSMA and DSMA  (EPA, 1974)

-------
K trnli
Nil i 1.
Ai I.I





_








CO
1
CO


- •—














Hill 1 1'
Ac I.I
iiink




























Vent




I'r (OH lilu
UNO
















\'








Arsenic
At 1

Hi- in i iir


/








}











4 Counter Curn-nl
III t Wali-r Fluw H.iki!-up



,M,h,.,l
llll.ilii,..
I'uw.-r



• u«c«r




Uctci 1>
Slui»||u 1
1 link J
I

?
j
•«•
' — ' — ' |f Uat«r Uird la \
WO Scrubber Untrr Pr»c«..|ng \
* o
I
' K
H,..w.-ir,l Nllrli \^ /





Ac td



























Ai N.'olr
A> I.I
Si or at; v>
.ink

"1* • !•» ih 1 * • 1 II
V.' * »» II 04|uf 1 | I4| tlH4!

Figure B-2.   Arsenic Acid Production (Sittig,  1977)

-------
CO
                     ";0' /"••"" "I.......
                     ^ -L:r.  -••
                     £<£_ 4  NO «
                                    !(•<••

                                       ^	^J~"" ~


                                        • OLIO MIOOuCT
                              Figure B-3.  Block Flow Diagram of Process for
                                           Cacodylic Acid Manufacture (Sittig, 1977)

-------
          ARSENIC
           ACID
         STORAGE
                                                      RECYCLE
                                                       WATER
                                              LIME OR
                                               LEAD
                                                OXIDE
          MEASURING
            TANK
CO
I
en
                                                                                    WATER
                                                                                    VAPOR
                                             BATCH
                                  MIX
                                                        VAT
DRIER
55 GAL DRUMS
                     M.  Production  of Uad

-------
                           MUNICIPAL SOLID WASTE
             PARTICULATE EMISSIONS
             0.8+0.4
QUENCH WATER
UNDISSOLVED
   SOLIDS
                        GASEOUS EMISSIONS
                          ROCKING GRATE INCINERATOR
                  BOTTOM ASH
                             FLYASH
                                      0.3+0.3
                                  LANDFILL
QUENCH WATER
  DISSOLVED
   SOLIDS
   SPRAY CHAMBER
       WATER
UNDISSOLVED SOLIDS
  SPRAY CHAMBER
      WATER
DISSOLVED SOLIDS
                           MUNICIPAL SEWER SYSTEM
               Figure B-5.  Flow Diagram of a Municipal  Incinerator
              	     in Kilograms Arsenic (Law and Gordon, 1979)
  a)   kg  arsenic  per  920  kkg  refuse.
                                     B-6

-------
                                 APPENDIX  C
       This distribution  of  arsenic  emissions  and  discharges  from
  combustion of coal and  oil* for energy  production  is  derived  from  the
  following basic assumptions.:
           flvaShcniin °5al 1J.d1?tr1buted between bottom ash,
           flyash collected, and particulate stack emissions (i.e.  ash
           loss via slagging is negligible); and                  '
       (2)  the distribution of the arsenic originally present in me
           feed material  is dependent only upon particle size.
  Thus:     Ash in (AJ  . Ash  out  (A  ^  - Afa  + AC  + Ag ,  or
           1  = "in  * Ab + Ac +  Ae
  where Afa  =  Fraction Bottom ash
       AC  =  Fraction Flyash collected
       Ae  =  Fraction Flyash emitted, also,
       Ac + Ae = Af where Af = Fraction of ash that  is total flyash.
      The amount of ash  which  appears as bottom ash, is dependent upon
 fuel  and boiler type.   For boilers which  fire  pulverized coal  «l  cm)
 eighty  percent of  the ash originally present in coal is estimated  to
 appear  as  flyash;  ash produced  in  cyclone boilers,  which burn  a
 somewhat larger  size of  coal than  pulverized coal fired  units,  is
 distributed  about  equally  between  bottom  ash and flyash; of  the ash
produced in  stoker fired boilers,  which burn relatively  large sizes of
coal (>10 cm), approximately twenty-five percent appears as flyash.
                                    C-l

-------
Essentially all of the ash present in residual oil appears  as  flyash.
The amount of flyash collected in turn depends upon the particulate
control device(s) used, which generally are cyclonic devices,
electrostatic precipitators, or baghouses.  The fraction of ash
emitted as flyash, control device efficiency, application of control,
effective efficiency, fine particulate fraction, atmospheric emission
fraction, and land dispersion fraction are shown in Table Cl.  For
pulverized coal fired boilers, ash emission fractions are calculated
as follows:
     (l-Af} = Ab = 0.20
              Ac = (Af)E = 0.71
              Ae = (Af)(l-E) = 0.088
The-fraction of the ash emitted to the atmosphere  (Ae) is further
subdivided into particulate which remains suspended (Ae <3  ym) and
that which eventually settles to land (Ae >3 urn) in the following
way:     Ae >3y = AePf = 0.057
         Ae <3p = Ae(l-Pf) = 0.031
     Arsenic emissions and discharges within a boiler are a function
of particle size, arsenic concentration increasing with decreasing
particle size.  Arsenic concentration of flyash by particle size
ranges is shown in Table C2.  Using these data (and assuming the
arsenic concentration of bottom ash and collected  flyash to be equal)
relative arsenic concentrations of bottom ash (C^), collected
flyash (Cc), flyash emitted >3y (Ce <3u), and flyash emitted
                                     C-2

-------
       (Ce>2u)  are calculated to be 1.0, LQ,  3.3,  5.7, respectively.
   Arsenic  emission  factors  are  calculated in  the following way.
   1 =  fb + fc * fe  <3 '^ +  fe >3 ^
   where:
       fb s Fraction of arsenic  contained  in bottom ash
      fc = Fraction of arsenic  contained  in collected flyash
      fe<3u = Fraction of arsenic emitted as particulate <3y
      fe>3v • Fraction of arsenic emitted as particulate matter >3U.
       These  fractions are  calculated using ash emission fractions and
  relative  arsenic concentrations:
       "tot.,"  V C>+V C
      fc
               Astotal
               Astotal
     Consequently, arsenic mission factors by boiler type and media
are shown in Table C3.
                                   C-3

-------
Taale C-l.  Arsenic Wastes:   Energy  production (kkg/yr)
Source
PRODUCTION
Coal
Bituminous
Anthracite
Lignite
Petroleum
Residual
Distillate
ELECTRICITY GENERATION
External Combustion
Coal (total)
Bituminous (total)
Pulverized dry
Pulverized wet
Cyclone
1 All stokers
Anthracite (total)
Pulverized dry
Pulverized wet

Cyclone
All stokers
Lignite (total)
Pulverized dry
Pulverized wet
Cyclone
All stokers
Petroleum
Residual oil
Combustion turbine
Steam generation
Combined cycle
Distillate oil
Combustion turbine
Stream generation
Combined cycle
SPACE HEATING AND OTHER
Coal1
Industry
Residential /Commercial
Coke plants
Petroleum
Residual oil
Industry
Residential /Commercial
Transportation
Distillate oil
Industry
Residential /'Commercial
Transportation
Fuel
Consumption

517,600,000
530,733,000
949,000
34,354,000
909,300,000
154,600,000
754,082,000


473,600,000
442,733,000
336,477,000
42,733,000
53,128,000
8,855,000
949,000
	
....

	
949,000
34,354,000
12,367,000
2,405,000
18,210,000
1,031,000
77,850,000
72,230,000
83,600
71,910,000
328,600
5,582,000
3,430,000 '
1,484,000
667,500


60,000,000
8,000,000
70,000,000
240,000,000
82,400,000
51,100,000
neg
21,300,000
743,500,000
80,000,000 '
189,000,000
479,500,000 .
Total b
Arsenic

16.450
16,260
10
175
76
46
30


12,590
12,400
9,420
1,200
1,490
250
10




10
170
63
12
93
5
24
22
neg
22
neg
2

2


3,360
1,580
220
1,960
54
24
18
neg
6
30h
3
3h
19
Total










3,530
3,490
2,730
350
340
65
2




2
45
19
4
21
1
24
22
neg
22
neg
2

2


1,000
440
54
510
54
24
18
neg
6
30h
3
8"
19
Disoersion0
Air
<-3 ;.ni >3 uin










1,750
1,725
1,320
170
220
' 15
"1




<1
25
9
2
14
<1
23
21

21

2

2


230
100
10
120
52
23
17

6
29
3
8
13
^









1,780
1,760
1,410
180
120
50
2




2
20
10
2
7
1
1
1

1

neg

neg


770
340
44
390
2
1
1

neg
1
neg
neg
1
Lansd Water

. 	
95f
e
neg





10,790
10,540
3,100
1,030
1,270
240
10'










_-a
> 52 =
I
10
150
54
10
79
5 '
1
1
negh
1
negh
negh

negh
.










n
• neg





3,530
1,580
210
1 ,340
2
1
1
neg
neg
lh
neg
negh

. neg






• nea



1
       C-4

-------
                                                 Table C-l.   (Concluded)



 a)  Production and consumption data are from Monthly Energy Review published by National  Reliability  Council,
     Princeton, N.J.  and are rounded to nearest thousand metric  ton.   Data  may not  add  due to  rounding.

 b)  Arsenic concentration by coal  type is  assumed to be:  Bituminous  coal  -  28 mg/kg;  Anthracite  -  10 rag/kr;
     Lignite -  5.1  mg/kg.   By oil  type:  Residual  - 0.2 mg/kg;. Distillate - 0.04 ing/kg  (Slater and Hall,  1977).

 c)  The amount of  arsenic emitted  to the atmospnere was calcualted as  follows:

                                         Q=(C)x(F)
                                         Where Q=quantity  of arsenic in feed  material
                                               C=concentration of arsenic in  fuel,  ppm
                                               F=yearly consumption of  fuel,  metric tons per year

                                         The  amount emitted  to the  atmosphere,  E, was
                                               E=(Q)x (fe  <3 mn)

                                         Where  fe  <3 «m is calculated  in Appendix C; arsenic associated with
                                         particles  that are  emitted to  the  atmosphere and are greater
                                         than  3 \aa  in  diameter settle out quickly and are thus included
                                         with  land  waste

 d)   The  amount of  arsenic  discharged  to  land was calculated  as follows:

                                         L=(Q-E)+(E >3  -.in)
                                         Where L=amount of arsenic  discharged to land:
                                              Q=amount of arsenic  in fuel;
                                              E=amount of arsenic  emitted  to the atmosphere
                                              E > 3um=amount of arsenic associated with particles greater than 3  r
                                              in diameter and that were initially emitted to the atmosphere,
                                              but eventually settled to land

 e)   National particula-te  emission  burden from coal  storage  piles is estimatad  to be 630 kkg/yr; Blackwood and
     Wacher  (1978).  Arsenic  fugitive  emissions are  calculated using the following arsenic concentrations:
     Bituminous  coal - 28 mg/kg; Anthracite -  10 mg/kg;  Lignite - 5.1 mg/kg.  Neg = <1 kkg.

 f)  Acid mine drainage is reported to average 12 mg/kg (28)  at an average  flow of 3.3 x 103 liters per mine-
    day.  Drainage from 5673 coal  mines is  considered in this calculation.

g)  Screening sampling data for the steam electric power point source  category:  averaae  flow,  wet fly ash  -
    9.9 x 10° i/day/plant, wet bottom ash - 9.8 x 10$ I/day/plant;  183 plants (52%  of all  coal  plants) use
    wet fly ash transport and 219  plants use wet bottom ash  transport; average concentrations:  bottom  pond
    effluent, 0.016 mg/1, fly ash  pond effluent,  0.060 mg/1.  EPA,  1980; FR 1980.

h)  Neg = <1 kkg.

i)  Coal combustion exclusive of utilities  is assumed to be  in stoker  fired boilers.
                                             C-5

-------
                                               Table C-2.   Mass  Efficiency of Particulate Collection on Utility Boilers3
o
1
en
Fuel Boiler Fraction ofb Control Application Effective
Type Ash Emitted Device „ , _ . .d Efficiency
«, ri.,.,1 rrr - C ofControl i-iiiv. iciiiy
As hi yd ill Efficiency tuntiui * , |
(A r) Calculated Reported
F (E)
Coal Pulverized 0.8 0.92 0.97 0.89 0.89
Cyclone 0.50 0.91 0.71 0.65 0.88
Stoker 0.25 0.80 0.87 0.70 0.65
011 A1) 1-0 0.50 0.20 0.10 0.10
Gas All 1.0 0 0 0
„
Fine Particulate Atmospheric Emission*1
Fraction (<3 M'n) /u^i^n A<>3 ^
(Pf) < ^ M'"
0.35
0.52
0.14
0.90
0.90

0.031
0.031
0.012
0.81
0.90

0.057
0.029
0.075
0.19
0.10

Land Dispersion*1
Flyash Bottom Ash
0.71 0.20
0.44 0.50
0.16 0.75



 a)  Data were obtained from utilities accounting for one-half of the total  U.S.  utility consumption of coal  in 1974.
 b)  Engineering estimates based upon published data.
 c)  Fraction collected of the total particulate mass entering control  devices.   Data  includes  both test results and reported design efficiencies.
 d)  Fraction of utility boiler equipped with particulate control  devices.
 e)  Effective efficiency equals control device efficiency multiplied by application of  control.
 f)  Slater and Hall, 1977.
Source:   Slater and Hall ,  1977.

-------
            Table  C-3.  Arsenic  Concentration  in  Coal  Flyash  as  a  Function
                       of  Particle  Size
  Particle
  Diameter
    (ym)
 Flyash retained in the plant
 Sieved fraction:

 >74
 44-74

 Aerodynamically sized fractions

 >40
 30.40
 20-30
 15-20
 10-15
  5-10
 <5

 Airborne  flyasha
 7.3-11.3
 4.7-7.3
 3.3-4.7
 2.1-3.3
 1.1-2.1
Arsenic
Concentration
   180
   500
   120
   160
   200
   300
   400
   800
   370
  680
  800
1,000
  900
1,200
1,700
                        Mass
                        Fraction
                          66.30
                          22.89
                            ,50
                            .54
                            .25
                            .80
                          0.31
                          0.33
                          0.08
2,
3.
3.
0.
a)  An equal mass distribution among particulate  size  fractions  is  assumed.

Source:  EPA, 198Cc.
                                     C-7

-------
                           Table C-4. .  Arsenic Emission Factors by Boiler Type and Media
Fuel
Coal


Oil
o
1
oo
Boiler Type
Bottom Ash
Pulverized 0.16
Cyclone 0.41
Stoker 0.61
All 	

Arsenic
Flyash
Captured

0.56
0.36
0.13
	

Distribution3
Flyash
Emitted (<3 »im)
e -<3 jim'
0.14
0.15
0.06
0.96

Flyash
Emitted (>3 nin
\ i I
e >3 jim/
0.15
0.08
0.20
0.04

a)  Fraction emitted based upon ash distribution  and relative  arsenic  concentration of flyash  <3u
    flyash »3y, flyash collected,  and bottom ash.   The  concentration  ratios  are (see Table C2  also):

         flyash emitted (<3 jj) = 5.7

         flyash emitted (>3 yj) = 3.3

         flyash collected     =1.0

         bottom ash           =1.0

-------
                   APPENDIX D

GEOGRAPHIC DISTRIBUTION OF INDUSTRIAL SOURCES OF
                ARSENIC RELEASES
   to  the  natural  environment
   sources  such  as .Ining
   teristics of  arseni
   tribution of
                                          or  arsenic  releases
                           "%     locations  of  "advertent
                         °Pfrations and  the regional  charac-
                              lnClUded  "e the  national
    foUowing
Fi8urs "-1
                         tables
                      included uithin
         D-2

  Fisure  "3
  Figure  D-4
        D-5
 Fi8ure D-6

 Figure D-7
 Figure D-,


 Figure D-IO

Figure D-ll

Figure D-12
   Location o£
  Lo<:atitm o£ Arsenic


                  Major Coal Burninf Po»er Plants in che
  Locatton, of Prinary Copper,  Uad,  and  Zinc Suiters

  Lead Mines and Shelters in eha Dni£ed staces

               Dis£ribuclon of  th.  Icon and  Steel

                f  th.  Primary Aluminum  Industry  in  che
 Erosion Hap of the United

 Soil Scenario Regions

 Distribution of Precipitation
                   D-l

-------
Table D-l      Estimated Uses of MSMA in the United States by Regions
               and Categories, 1979

Table D-2      Companies Engaged in Boron Production
                                D-2

-------
                                         })  Boundbrook, N.J.
                                         2)  Somerset, N.J.
                                         3)  Vineland, N.J.
                                         '0  South Grati, CA
                                         5)  Marinett, HI
FIGURE D-l   LOCATION OF PRODUCERS OF SODIUM ARSENITE AND ITS DERIVATIVES
             DSMA,  MSMA AND CACODYLIC ACID
Source; u. S. EPA  J980c

-------
                                                 5) Charlotte, MC
                                                 b) Fort V.ilJv, G.\
                                                 7) Suucn CiaiJ, CA
                                                 8) Honiiani. 'I"\
                                                 '» Bryan, FX
FIGURE D-2   LOCATION OF PRODUCERS OF ARSENIC ACID  AND  ITS DERIVATIVES

Source: u.  S. EPA  1980c

-------
o
Ol
                               • Preasuro
                               * Nan-Proituro
                               * Pi««»ur« end Hon-Pf»»tur»
Source:  U.S. EPA  j
                                                   ,>80c

-------
            - Heavy use  areas
Source:  U.S, EPA  1980c
          FIGURE D-4   LOCATION OF ARSENIC ACID USE
                       IN TEXAS COTTON AREAS
                             D-6

-------
t)
                                                                              =  Power Plant
         Source:  U.S. EPA  1980c
                        FIGURE D-5   LOCATION MAP OF THE MAJOR COAL BURNING  POWER  PLANTS
                                     IN THE UNITED  STATES

-------
o
I
CO
                                           tinker Hill (Kolloo)
                                                   "rAnacor.Ja (Anacorx)a)


                                                    Asarco (E. Helena
         Bunker Hill (Kellonl
                                               • Kennecort 'Garliold)
Kennecott. IMcGill)   -«
                                                                                                     Aiarco (Columb



                                                                                                      . Jod (Httrculandurn
                                                                                     AwrcoTGlo vcr)
                                                                                                                          Se7vice TCopperHid)
                                                                          rial Zinc (Bartlevville)
Inspiration (l^iami) ?
                              sarco (Hay/den)   -tp-Kenpecou (Hayden)

                                     >             uJ

                                                    Vhelps Dodge (Mor«nci) '
                           Phelpj Oodae (Douijlai)


                                         (San Ma
             y    Copper
                                                                                     Awrco (Corpus Chri.li)
       Sourcu: U.S.  EPA   1980c
                                    FIGURE 6-D   LOCATIONS  OF  PRIMARY COPPER,  LEAD, AND  ZINC SMELTERS

-------
JiiliJ
 o
 o
 OO
 cr>
 r-t
 W
to

-------
a
                                                                                                   AI II OhhlA
                                                                                                    t~*,».IU
                                                                                                    I.....J,
                                                                                                  COL OH A DO
                                                                                                   f.-M^
                                                                                                  COUMtCltCUf
                                                                                                  fftOHIU*
                                                                                                   I —I..
                                                                                                   N— L«iit«
                                                                                                   M.W tl,..,(
                                                                                                  MINIIlCilV
                                                                                                           I HAM t nun MAM
                                                                                                                  OHIO
                                                                                                                   c
        Source:   Arthur D.  Little, Inc..  Steel and the  Environment;   A Coat
                  Impact Analyala for the  American Steel Inatj^ufP, M«yi)
                        FIGURE D-8   GEOGRAPHICAL DISTRIBUTION OF THE IRON AND  STEEL INDUSTRY
                                                                                                                 V.H
                                                                                                                mil

-------
       oBAUXITE MINES
       • ALUMINA REFINING PLANTS
       oALUMINUM REDUCTION PLANTS
Source:   U. S. EPA  1980c
         FIGURE  D-9   LOCATION MAP OF THE PRIMARY ALUMINUM  INDUSTRY
                       IN THE  UNITED STATES

-------
         Source:  Brady (1974)
FIGURE D-10
EROSION MAP OF THE UNITED STATES:  (1) severe sheet and
gully corrosion; (2) moderate to severe erosion of mesas
and mountains; (3) moderate to severe wind erosion with
some gullying; (4) moderate sheet and gully erosion with
some wind action; (5) moderate sheet and gully erosion
locally; (6) erosion rather unimportant.  [After U.S. Soil
Conservation Service.]
                                 D-12

-------
I
M
OJ
     Source:  lir;n!y   1974
                                          FIGURE D-ll   SOIL  SCENARIO REGIONS

-------
. V-f.   . Is J.'. t   w

            D-14

-------
                                   TABLE  D-l
M
Oi
 Region

 Northeast
 North Central
 Southeast
 South Central
Northwest
Southwest

     Total
                                               ESTIMATED  USES  OF MSMA  IN THE UNITED STATFS
                                                    BY REGIONS AND CATEGORIES, 1979
                                                       (kkg Arsenic Equivalent)
Geographic dis
 tribution
 known
            Source: U, S. EPA'
                 1980c

-------
          TABLE D-2   COMPANIES ENGAGED IN BORON PRODUCTION
Company and Location

Kerr-McGee Corp.
Kerr-McGee Chemical
Location, Subsidiary
Searles Lake (Trona) CA

Searles Lake (West End), CA
Stauffar Chemical Co.
San Francisco, CA
Annual Capacity

    100
   25-30
  Unknown
Texas United Corp.
  American Borax Corp.,Subsid.
  Furnace Creek District, Inyo   50-75
  City, CA
U.S. Borax Chemical Corp.
  Boron, CA
  Wilmington, CA
     700
      Capabilities^

 Borax decahydrate,  borax
 pentahydrate,  anhydrous
 borax, boric acid
Borax decahydrate, borax
pentahydrate, anhydrous
borax

        Unknown
                  Colemanite,  alexite/prober-
                  tite
Refined borax decahydrate,
crude and refined borax pen-
tahydrate, oxide and refined
anhydrous borax
 Source:  U. S. EPA   1980c
                                 D-16

-------