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 Argas for Future Research
7-1.8 Conclusions
of
e^ir^eltlf^tai"
a Taiwan epidemiology studv sh™?
and lo. drinking wat§er concentrSi
of excess individual lifetime skin
ingestion of 3 yg/l of arsenic in
0.0009 to 0.00188/ Thlee^L^ra
concentration in water supplies The
rate of skin cancer for the U T must
" ^"^ ^ °f
°n extraP°lations from
cancer
estiffiat^
associated with
occurring arsenic
th±S Very hi§h
rates and geographic ltitude r f§ relatlonshiP between incidence
the inherent weak ipiJeSSo^ studies th^ T ^T* C°nsid-ini
amount of uncertainty associftefwJthM^ ^nemselves, there is a considerable
skin cancer rate to the U?S popuIatLn eXtraP°lati°n °f the
la .
of uncertainty exist In thS th^P J • § 7 un(5uantif^ble sources
not fuily understood there is no nr echanisms of "rcinogenesis are
selecting the most appropriate model SC16ntific consensus for
models. These models^ake similar risT^ V3riety °f ^ ext^Polatio
levels, but markedly different eati^? estlniations at high exposure
that are of concern with e^iro^entJl e" " ^ relative1^ *»> 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
------- |