United States
              Environmental
              Agency
              Office of
              Toxic Substances
              Washington, D.C.
March 1979
EPA-560/6-79-005
                 ,
&EPA
Arsenic:
A Preliminary  Materials  Balance

-------
•ARSENIC:  A PRELIMINARY MATERIALS BALANCE
                 Prepared by:
   Lowenbach and Schlesinger Associates, Inc.
                1842 Kirby Road
            McLean, Virginia 22101
     Purchase Order Number:  W-1434-NNSX
                Project Officer
            Dr.  C. Richard Cothern
    ENVIRONMENTAL PROTECTION AGENCY
        OFFICE OF TOXIC SUBSTANCES
          WASHINGTON, D.C.  20460

-------
This report has been reviewed by the Office
of Toxic Substances, EPA, and approved for
publication.  Approval does not signify that
the contents necessarily reflect the views
and policies of the Environmental Protection
Agency, nor  does mention of trade names or
commercial products constitute endorsement
or recommendation for  use.
                  ii

-------
                               ABSTRACT

        Existing data on the natural and anthropogenic sources of arsenic
emissions are compiled and presented in a fully annotated tabular format.
Arsenic distribution ot the environment is graphically displayed in terms of
air, land, and water emissions.  The three major sources of arsenic to the
environment are identified as fossil fuel consumption for energy production,
arsenic production  and commercial use, and primary copper smelting.  Data
gaps are identified  and recommendations for further study outlined.
                                   111

-------
                       TABLE OF CONTENTS

                                                            Page


ABSTRACT	    iii

LIST OF FIGURES	    v

LIST OF TABLES	    vi

EXECUTIVE SUMMARY	    vii

I.    INTRODUCTION	   1
     Project Objective	   1
     Scope of Work. ..,..,	   1

II.   ARSENIC MATERIAL BALANCES	   3
     Data Collection and Presentation	   3
     Problem Areas	   4
     Arsenic Emissions	   8

III.  SUMMARY AND CONCLUSIONS	  50

REFERENCES	53
                                IV

-------
                         LIST OF FIGURES
Number                                                          Page
   1.       Distribution of Arsenic Emissions to the
             Atmosphere	5
   2.       Distribution of Land-Disposed Arsenic
             Wastes  ,.  .	  6
   3.       Distribution of Arsenic Wastes Discharged
             to Surface Waters	  7
                                  v

-------
                           LIST OF TABLES
Number                                                         Page


  I     Arsenic Emissions:  Natural Sources, metric tons
         per year	  9

  II    Arsenic Emissions:  Arsenic Production and Use,
         metric tons per year	12

  III   Arsenic Emissions:  Energy Production, metric
         tons per year  . . .	17

  IV   Arsenic Emissions:  Concentrations of Arsenic in
         Fuels and Waste Products	21

  V    Arsenic Emissions:  Primary Copper Smelting,
         metric tons per year	24

  VI   Arsenic Emissions :  Copper Refining, metric
         tons per year	30

  VII   Arsenic Emissions:  Primary Lead and Zinc
         Industries, metric tons per year	33

  VIII  Arsenic Emissions:  Primary Non-Ferrous Metal
         Ores Mining and Milling,  metric tons per year	36

  IX   Arsenic Emissions:  Manganese Production,
         metric tons per year	39

  X    Arsenic Emissions:  Iron Ore Production and
         Steelmaking, metric tons  per year.	   41

  XI   Arsenic Emissions:  Phosphorus Production,
         metric tons per year	44

  XII   Arsenic Emissions:  Boron  Production, metric
         tons per year	47

  XIII  Arsenic Emissions :  Water and Wastewater Treatment,
         metric tons per year	49
                                 vi

-------
                         EXECUTIVE SUMMARY

     As a first step in assessing the risk presented by arsenic and its
compounds, the U.S. Environmental Protection Agency,  Office of Toxic
Substances has  undertaken a review of arsenic emissions from natural and
anthropogenic sources.  This study discusses the distribution of arsenic
in the United States within the confines of a material balance, identifying
arsenic input, that which is  dispersed to  air, land,  and  water,  and that
which is contained.
        Arsenic ranks 20th among other elements in crustal abundance and
is a constituent of an estimated  245  mineral species.  Sulfide deposits
associated with  nonferrous ores have elevated arsenic concentrations, as
well as sedimentary  deposits such as iron ore,  phosphate rock, borax ore,
manganese ore,  and fossil fuels. Arsenic is commercially produced as a
by-product of the ASARCo-Tacoma primary copper smelter.
        Processing of raw materials, such as nonferrous ores, and fossil
fuels, results in the release  of arsenic to the environment. Other man-
made sources of arsenic  emissions include its use in the manufacture of
commercial products, such as arsenical pesticides,  or as a product additive
 (e.g., lead alloy industry) .
        Arsenic emissions to air, land, and water are summarized in Figures
1, 2, and 3. In terms of arsenic released  per year, energy production
represents the largest single source of arsenic discharged to the atmosphere
and the second largest source (pesticide application is larger) of discharges
to land.  Primary copper smelting is the third major source of arsenic
emissions  to the environment.
        The conclusions  reached in  this study  are based,  nonetheless, on a
cursory review of existing arsenic emissions data.  Because the majority of
this information  is unsubstantiated,  and in many cases limited to the author's
"best guess",  a  summary table totaling arsenic emissions to all three sinks has
been deliberately omitted. Instead, the origin of each piece of data is discussed
                                 vii

-------
within the tabular format of the report.  Recommendations for further data
collection and evaluation are outlined, highlighting existing data gaps.
                               viii

-------
                               SECTION I

                            INTRODUCTION
PROJECT OBJECTIVE
        Arsenic is a recognized toxic substance and suspected occupational
carcinogen that has recently been placed under scrutiny by Federal
regulatory agencies.  As an initial step in assessing the risk presented
by arsenic and arsenic compounds, the U.S. Environmental Protection
Agency, Office of Toxic Substances,  has begun a review of potential
sources of arsenic exposure.  The purpose of this study is to address  the
distribution of arsenic to the environment.
SCOPE OF WORK
        The scope of this report has  been limited to a review of
published data concerning  the flow of arsenic within the United States.
Available literature has been critiqued and compiled in a single document
with the intention of presenting (1) an  overview of major sources of arsenic
emissions, (2) fully annotated tables  to aid data evaluation, and  (3) an
indication of data gaps.  While  many primary sources were searched during
the literature  review, no attempt was made to generate new data points or
estimate  emission factors.  Furthermore, no attempt was made to include a
discussion of the fate and transport of  arsenic within each of the three
environmental sinks:  air,  land, and  water.
        Data collection, sources of information, and problem areas are
reviewed in Section II.  Existing data (with introductory prefaces) are
presented in a tabular format for the  following categories:

-------
                     -  Natural Sources
                     -  Arsenic Production and Use
                     -  Energy Production
                     -  Concentration of Arsenic in Fuels and Waste Products
                     -  Primary Copper Smelting
                     -  Copper Refining
                     -  Primary Lead and Zinc Industries
                     -  Primary Non-Ferrous Metal Ores Mining and Milling
                     -  Iron Ore Production and Steelmaking
                     -  Phosphorus Production
                     -  Boron Production
                     -  Water and Wastewater Treatment

Finally, Section III presents major conclusions, including recommendations for
future studies.

-------
                            SECTION II


                   ARSENIC MATERIAL BALANCES
DATA COLLECTION AND PRESENTATION

        The purpose  of this study was to compile the available data on

arsenic  input and emissions from natural  and cultural sources. The

tabular  format chosen, with extensive use of footnotes, serves to highlight

data gaps, unsubstantiated estimates of emission factors, and conflicting

data points,  while providing-a comprehensive summary of the available

information.  In preparing the arsenic material balance for the United States,

the following literature served as primary sources of data:
              National Inventory of Sources and Emissions:
              Arsenic - 1968
              W. E. Davis & Associates,  1971.

              Technical and Microeconomic Analysis:
              Arsenic and its  Compounds
              U.S.  Environmental Protection Agency,  1976.

              Emission Factors for Trace Substances
              Anderson, D., 1973.

              Medical and Biological Effects of Environmental
              Pollutants:  Arsenic
              National Academy of Sciences (NAS),  1977.

              Gross Annual Discharge to the Waters in 1976:
              Arsenic U.S. Environmental Protection Agency,  1978 .

              Preliminary Draft:  Environmental Assessment of
              Arsenic Emissions from  Primary Copper Smelters*
              U.S.  Environmental Protection Agency, 1979.
        * This report has not been formally released by the Agency.  An
updated draft of Chapter 3,  Appendix F, and Table 4-6 "Summary of
Fugitive Arsenic Emissions at Primary Copper Smelters  under Existing
Control"  were forwarded to J. Schlesinger by A. Vervaert for use in this
document.

-------
        Secondary sources of data included:
          •   A search of Chemical Abstracts  for the preceding
              twelve year period;
          •   Information in the priority pollutant file maintained
              by the Water  Quality Analysis Branch, Office of
              Water and Waste Management;
          •   Personal interviews with representatives of EPA
              (IERL, OAQPS, OEMI,  Water Quality, Effluent
              Guidelines, and ERL),  TVA, and the  USGS.
Although informal discussions with EPA personnel and various researchers
have provided useful background data, as well as leads to ongoing
research/monitoring studies, unpublished  data were omitted from this report.
PROBLEM AREAS
        Before presenting the compiled data, it is important to consider the
problem areas inherent to a  study of this nature.  First and foremost, total
emissions for any one category can be quite misleading,  as  non-quantifiable
emissions are necessarily  omitted (i.e., a blank does not indicate zero
emissions, but rather,  a gap in information) .  .Secondly, it  appears that many
of the  emission factors reported by Davis  (1971),  E?A(1976a), NAS(1977),
and Anderson (1973) are  unsubstantiated, and in some cases, quite outdated
with regard to existing environmental controls. Furthermore, in some cases,
emissions  factors cited from primary sources were incorrectly applied to
current production data.  In other instances, emissions data are based on
feed concentrations arbitrarily selected or analogies to other chemicals and
similar industrial processes  (for lack  of better information)  .  Thus, the
majority of existing data has not been verified by sampling  and analysis and
appears to be little more than "engineering judgment or an  educated  guess".
        A summary sheet reporting total arsenic emissions to  each environmental
sink by category has been deliberately omitted.  Instead, in an effort to
direct  attention to data trends,  rather than to "significant  figures",  total
emissions to air, land,  and water are graphically presented in Figures 1, 2,
and 3,  respectively. Numbers that do appear in the Tables,  in general, must
be regarded as speculative and used  with  extreme caution.  Finally,  all
data points have been  rounded to two significant figures; greater accuracy and
precision, based on available information,  is unwarranted.
                                  4

-------
          Arsenic Production
                                  Primary
                             Lead and Zinc
                                   Smelting
Energy Production
                         Primary
                         Copper
                         Smelting
                                                Primary Non-
                                                Ferrous Metal
                                                Ores Mining
                                                and Milling
             FIGURE  1.

  Arsenic Emissions to the Atmosphere

-------
             Arsenic Production
                                        Copper
                                      Refinin
                                         Iron Ore
                                and Steelmaking
Energy Production
                    Primary Copper Smelting
Primary Lead
and Zinc Smelting

Phosphorus
Production

Manganese
Production
                FIGURE 2.

       Land-Disposed Arsenic Wastes

-------
             Non-Ferrous Metal Ores
               Mining and Milling
                                    Phosphorus
                                    Production
                                    Copper Refining
Primary Lead and Zinc
       Smelting
                                 Water and
                                   .Wastewater
                                     Treatment
                         Energy
                       Production
                  FIGURE 3.

  Arsenic Wastes Discharged to Surface Waters

-------
NATURAL SOURCES
      Arsenic ranks 20th among elements in crustal abundance, and is a
constituent of an estimated 245 mineral species (NAS,  1977).  Natural input
to the environment  may be described in terms of weathering of continental
rocks (erosion), as the contribution from volcanic activity is small.  (Removal
of arsenic from global cycling by sedimentation has not been addressed.)
Estimated arsenic emissions from natural sources are shown in  Table I and
are based on an estimate of the total amount of weathered material transported
to the oceans annually (Judson, 1968).   The total  arsenic input was
calculated through  (1) gross estimates of the igneous rock, shale, limestone,
and sandstone content of the weathered material, and  (2) the average value
of the arsenic content of each type of rock.  If erosion resulting from
intensive land use is included in such calculations, the estimated arsenic
input from weathering increases from 45,000 to 120,000 metric tons per year
(Ferguson and  Gavis, 1972) .

-------
                                    TABLE  I
            ARSENIC EMISSIONS:  Natural Sources, metric tons per year"
SOURCE
                                                           Air
                                 DISPERSION
                                    Land      Water
Weathering2'3
    Igneous Rock
    Shales and deep sea
      sediments
    Sandstone and limestone
Volcanic activity
 3,0004  4,2005
40,0006  85,0007
 2,0008   2,5009
45,000 120,000
    3,000

   40,000
    2,000
1.  Based on  a global arsenic cycle for natural imput.  All data are reported to two significant
figures and rounded to 5kkg increments.
                                                                                  g
2.  Ferguson and Gavis, 1972.  Based on Judsons's (1968)  global estimate that 9.3x 10  metric tons
of material are weathered and transported to the oceans annually. Furthermore, it is presumed
(see:  Conway,  1942) that the weathered material is approximately 20 percent igneous rock,
65 percent shale, and  15 percent limestone and sandstone with respective arsenic contents
of 1.8 mg/kg, 6.6 mg/kg, and 1.5 mg/kg.
3.  Ferguson and Gavis, 1972.  Due to rapid erosion caused by intensive land use, Judson (1968)
                                                                                g
estimated that the mass of material transported to the oceans has increased to 24 x 10  metric
tons/yr, resulting in a proportional increase in the release of arsenic to 120,000 metric tons/yr.

-------
4.  Ferguson and Gavis, 1972.  Based on reported values of 1.8 mg/kg (Taylor, 1964) and
about 2 mg/kg  (Onishi and Sandell, 1955).
5.  NAS, 1977.   Reported as average arsenic concentration in igneous rocks from data
provided by  Onishi (1969) and Boyle and Jonasson (1973).
6.  Ferguson and Gavis, 1972.  Based on a reported value of 6.6 mg/kg (Vinogrodow, 1960)
for shale.
7.  NAS, 1977.  Reported  (14.5 mg/kg) average arsenic concentration for shales and clays from
data provided by Onishi (1969) and Boyle and Jonasson (1973).  One sample with an arsenic
concentration of 490 mg/kg was excluded from the calculation.
8.  Ferguson and Gavis, 1972.  Based on a reported value of 1.5 mg/kg (Onishi and Sandell,
1955).
9.  NAS, 1977.   Based on an average of arsenic concentrations in limestone (1.7 mg/kg and
sandstone  (2.0 mg/kg).
10.  Ferguson and Gavis,  1972.  Although the contribution of vulcanism to the arsenic cycle
over geological  time  is recognized to be large, annually, this contribution is suggested  to
be small in comparison to weathering of continental rocks.

-------
ARSENIC PRODUCTION
      The production and uses of arsenic are summarized in Table II.
Because there is only one domestic manufacturer of arsenic trioxide (ASARCo-
                                                               *
Tacoma, Washington), production data are considered confidential .  The
annual production level shown is based upon a material balance estimated by
ASARCo and an assumed  production time of 8,000 hours; accordingly this
figure should be regarded as a gross estimate.  Import data are from the
Minerals Yearbook,  1975.
      The major use of arsenic is pesticide manufacture, followed by
fungicide manufacture (principally wood preservatives), glass manufacture,
non-ferrous  alloying, and feed additive  manufacture.  Data concerning these
uses, like arsenic production, are considered confidential.  The figures
presented have been compiled from industry estimates and the limited data
that are available.   Again, these figures can only be regarded as  rough
estimates.
      It is worth noting that  demand for many of the uses  of arsenic are
reported to be in decline,(EPA,  1976a).  This is particularly true of
arsenical pesticides  and herbicides with  exception of arsenic acid  which is
used for desiccation of  cotton in  the Southwestern  United States.   Arsenic
use during glass manufacture is  reported as  being largely  replaced by
non-arsenical sulfates.
      *
         When there are three or less manufacturers of a particular chemical,
combined production data could  reveal proprietary information and thus may
not be revealed publically.

                                   11

-------
                            TABLE  II
ARSENIC EMISSIONS:  Arsenic Production and Use, metric tons per year
SOURCE
2
Production
3
Imports
Total
Use
Pesticides
Production
Application
Ginning Wastes
Incineration
Fungicides
Plastics9
Wood10
Alloys11
Brass
Copper
Lead
Gallium Arsenide
Glass5
Catalysts15
Feed Additives16
17
Miscellaneous
Total
PRODUCTION
Input Contained
12,000
9,400
21,000


15,0004 13,0005



2,000 2,000
5
2,000

1012 10
ioo13 100
95014 950

1,800 1,600

410 410

20,000 18,000 5,100
DISPERSION
Air Land Water
250



3,4006 2,7007 12,000 10,000 -8
150 130
2,900 2,200
20
3307 330
neg mg neg



neg
neg
500 500 60

210

neg

4,100 3,200 12,000 10,000

-------
1.  Data are  reported to two significant figures and are rounded to 5kkg increments; emissions below 3kkg are considered to be
negligible.
2.  EPA, 1979-   Data is taken from material balance around ASARCo Plant, Takoma, Washington,  1978.  Based on l,500kg/hr,
8,000 hours per year.
3.  Minerals Yearbook,  1975. Bureau of Mines,  1977.
4.  EPA, 1976b.  14,800  metric tons/yr of arsenic were estimated to be used for arsenical persticides in 1974.  Herbicides:
cacodylic acid, disodium methane arsenate   hexahydrate, monosodium acid methane arsenate, and arsenic acid.  Insecticides:
calcium arsenate and lead arsenate.
5.  EPA, 1976a. Statement to OSHA by Crosby F. Baker for ASARCo:  In Re:  PROPOSED STANDARD for Occupational
Exposure to Inorganic Arsenic.
6.  EPA, 1976b. Air emissions include:  organic pesticide application - 2,485 metric tons/yr;  inorganic pesticide application -
399 metric tons/yr; incineration of cotton ginning waste - 17 metric tons/yr; production losses from manufacture of arsenical
pesticides - 154 metric tons/yr.   All based on 1974 emissions.
7.  W. E. Davis and Associates (1971). The emission factor for production loss is reported as 0.01; the emission factor for
application is  reported as 0.168.  Based upon 1974 production of 11,540,100 bales of cotton. 0.317 metric tons of trash are
produced per bale of cotton of which 37 percent is burned.   The arsenic concentration of this waste is assumed to be the same
as that of particulate emission during ginning operations (300 mg/kg).  76.6 percent of the arsenic present is lost during
combustion for a total emission of 310 metric tons.  Ginning particulate emissions are reported as  11.7 Ibs/bale of cotton
ginned with an average arsepic concentration of 300 mg/kg  for a total emission of 18 metric tons.
8.   EPA, 1976c. No discharge of wastewaters  due to reuse and recycle of all process wastewaters.
9.   EPA, 1976a.  Principally 10,10* - oxybishpenoxarsine, C2.H,gAs2O3.
10.  Ernst and Ernst 1976.  .Preservatives include Fluor-chrome-arsenate phenol - 100 metric tons/yr;  chromated copper
arsenate - type A  -  185 metric tons/yr, type B - 849 metric tons/yr,  type C - 830 metric tons/yr.
11.  EPA, 1976a.  Use of  arsenic  in non-ferrous alloys during 1974 has been estimated at 1,240 metric tons by an ASARCo
spokesman.

-------
12.  EPA,1976a.  Admiralty brass contains 300 mg/kg of arsenic.  The total demand in 1968 for all industrial non-electrical copper
was  250,000 metric tons.  Assuming that 25,000 metric tons  (10%) represents an upper limit of admiralty brass production,
arsenic use is 7.5 metric tons.
13.  EPA,1976a.  Copper containing 0.3% arsenic is used in car radiators.  A typical car radiator contins 5.9kg of copper.  A
production level of 10 million vehicles per year thus requires 59,000 metric tons of copper togther with  175 metric tons of arsenic.
Such copper is extensively recycled, either by fire refining (which does not remove arsenic)  or by electrolytic refining  (which
removes arsenic as a sludge).  An arbitrary estimate of the  disposition of the 175 metric tons of arsenic presumed to be present
in arsenic scrap is that 75 metric tons is reused in auto readiators (along with 100 metric tons of new replacement arsenic)
while 100 metric tons is dissipated in other copper alloys or removed during electrolytic refining.
14.  EPA, 1976a.  Use of arsenic in lead shot is estimated at 60 metric tons by the Lead Industries Association.  Use of arsenic
in bearing  metals is based  on new lead usage of 2,100 metric tons; bearing metals contain approximately 83% lead  and 0.6% arsenic.
Thus consumption for this use is  estimated to be  15 metric tons.  Use of arsenic in antimonal  lead is  estimated as  follows: lead
consumption  for antimonial lead is approximately 350,000 metric tons (of which 6,800 metric tons is primary antimonial lead and
320,000 metric tons are recycled antimonial lead scrap); assuming an overage arsenic concentration of 0.25%, 870  metric tons of
arsenic are contained in antimonial lead much of which is lost to the atmosphere during recovery in lead blast furnaces.   Using an
air emission factor of 0.4 kg/kkg of lead together with the quantity of slag generated from secondary lead blast furnaces (150,000 kkg)
at an arsenic concentration of  0.2%, 375 kkg of arsenic is retained in the secondary lead.  Thus,  by difference, 500 kkg of arsenic
must be added.
15.  Production  figures are unavailable.
16.  EPA, 1976a.  Emission factor presumed to be one-half that of pesticides (0.01),  i.e.,  0.005.

-------
ENERGY PRODUCTION
        Arsenic emissions to the environment from energy production are
summarized in Table III.  In terms of amounts discharged per year, energy
production represents the largest single source of arsenic discharges to the
atmosphere and the second largest source (pesticide application is larger)
of discharges to land.
        For electricity generation in 1974, 54% of the fuel consumed was
bituminous coal,  0.25% anthracite,  1.3% lignite, 23% natural gas,  19% residual
oil, and 2.9% distillate oil (Slater and Hall,  1977).  The bituminous coal
category includes 10 to 15% of subbituminous  coal.   Included  in the natural
gas category are less than 0.6% combined coke oven gas, refinery gas, and
                                                                       *
blast furnace gas.  The use of liquified natural gas, synthetic natural gas ,
and oil derived from oil shale was, and remains,  zero.  The residual oil
category includes all  heavy oils and consists of 0.6% No.  4, 1.5% No. 5,
95.5% No. 6, and 2.4% crude.  Distillate oil  consisted primarily of No.  2
fuel oil and less than 10% kerosene and jet fuel.  Approximately 96% of
the consumed fuels were burned in external combustion systems and 4% in
internal combustion systems.
        Arsenic input and discharges  are calculated on the basis  of reported
arsenic concentrations of fuels and from fuel consumption data.  Emissions
from coal are much greater than those from petroleum or  natural gas because
arsenic concentrations are greatest in coal; in fact, no trace  elements are
present in  appreciable amounts  in gaseous hydrocarbons.
        The arsenic concentration of coal varies  widely.  Ruch,  et al.  (1974),
in analyzing 101 coals, found" arsenic concentrations to range from 0.50 mg/kg
to 93.0 mg/kg; an arithmetic mean of 14.02  was reported  but  means little as
the standard deviation of this data was approximately 17  mg/kg.  Of the coals
analyzed, 82 were from the  Illinois Basin, 11 from the Eastern United  States,
and 8 from Western states.  A majority of coals in this study  (69 samples)
        *
          Coal gasification may lead to significant arsenic emissions however;
approximately 25% of the arsenic initially present in the coal is found in the
product gas, while  the balance remains with the coal ash (Attari, 1973).
                                    15

-------
contained arsenic in concentrations less than 12  nag/kg.  Other arsenic
concentrations have been  reported as summarized in Table XII.  Davis  (1971),
Anderson (1973), and EPA (1976a) all report an average    arsenic
concentration in coal of 5.44mg/kg.  According to Anderson (1973) ,  this number
is based  upon the analysis of a single sample of coal.  A value of 28.0 mg/kg
for  arsenic concentration in coal is reported by Slater and Hall (1977) based
upon regional coal usage.
        Because of the variation of arsenic concentrations in coal, three
estimates of the amount of arsenic present in coal are presented in Table IV.
All  numbers can only be regarded as providing  gross estimates of arsenic
emissions occuring during coal combustion.  The same statement is also
probably true for other fuels but because of lower arsenic concentrations,
these emissions are of less concern.
                                 16

-------
                                                             TABLE   III
SOURCE
                                ARSENIC EMISSIONS: Energy Production,  metric tons per year
    PRODUCTION.
        Input
                                                                                       Air
                                             DISPERSION
                                                Land
          Water
Production
    Coal
      Bituminous
      Anthracite
    Petroleum
      Residual
      Distillate
Electricity Generation
    External Combustion
      Coal (Total)

          Bituminous
           Pulverized dry
           Pulverized wet
           Cyclone
           All stokers
          Anthracite
           Pulverized dry
           Pulverized wet
           Cyclone
           All stokers
14,000   6,900   2,700
    50
11,00(T
2,500
    15
                                 neg-
                            2,700   650
                               2,700
                               2,000
                                 370
                                 370
                                  20
                                   3
                                 neg
                                                 neg
                                                      8
                                                                8,300
                                                                               95"
                                                                               150"
10
                                                                                                              (Continued)

-------
                                                              TABLE III
SOURCE
Petroleum (Total)
Residual Oil12
Tangential
All other
PRODUCTION
Input
20
neg
Air
20
20
10
neg
DISPERSION
Land Water

          Distillate Oil13
    Internal Combustion (All)
                2
Steam and Other
14
    Coal
      Bituminous
          Deliveries to consumers
          Manufacturing  and mining
            industries
          Steel mills
          Oven coke  plants
          Beehive coke plants
          Bunker
      Anthracite
          Heating
          Steel mills
          Other
      Petroleum
          Residual
          Distillate
                       neg
            4,100   2,100
                                                              neg
800

1,

2,





230
500
160
200
35
neg
25
5
5
110
730
80
1,100
15
neg



45
290
30
440
5
neg



1. Slater and Hall (1977) .  Data are reported to two significant figures and  rounded to 5kkg increments; .emissions below-3kkg: are ^listed;
as negligible.

-------
2.  Production figures are from Minerals Yearbook, 1975, U.S. Bureau of Mines.  Production of coal for 1974 is reported as  500 million
metric tons/yr.  Input figures represent total consumption multiplied by arsenic concentration (see Table  IV).
3.  Blackwood and Wachter (1978).  National emission  burden from coal storage piles is estimated at 630 metric tons/yr.  Arsenic fugitve
emissions data are calculated using previously reported concentrations of arsenic in coal.
4.  EPA  (1978).  Screening sampling data for the coal mining industry.   Acid mine drainage is  reported to contain
12  rag/kg arsenic at an average flow of 1,004 mgd per mine.  Drainage from  5,673 coal mines is  considered in this calculation.
Coal pile runoff is not considered in this estimate.
5.  The amount of arsenic emitted to the atmosphere was calculated as follows:
      the amount of arsenic in the fuel. A, was
              A = (C)(F)
      where C = concentration of arsenic in fuel, ppm
            F = yearly consumption of fuel, metric tons/yr.
      The amount emitted to the atmosphere,  E,  was
              E = (A)(f)
      where f = estimated fraction emitted to  air.
            f   ,  =  0.25  (based on studies of 4 coal-fired units)
            f .,   =1.0
             oil
6.  Minerals Yearbook,  1975.  Based upon 77% of coal production used for electric power generation.
7.  EPA  (1976).  Based upon  450  million metric tons/yr.
8.  W. E. Davis and Associates, Inc. (1968).  fcoal =  0.27.
9.  EPA  (1977);  Screening sampling data for the steam electric power generating point source category; and  EPA (1974); Development
Document for  the Steam Electric Power  Generating Point Source Category.  Arsenic is reported to be present in ash pond effluents at
an average concentration of 448  mg/kg.  The average flow from an ash pond is  reported as 21.7 x lo° 1/day-plant.  379 steam electric plants
have  been identified  (NCA, 1977). Each facility is assumed to have an ash pond.

-------
10.  The breakdown of the boiler population used was:  pulverized dry bottom 72.3%, pulverized wet bcttons 13.5%, cyclone toilers 13.51,
and spreader stokers 0.7%.   The particulate control efficiencies and the percentage of boilers using controls were multiplied and subtracted
from unity to obtain the fractions of the total particulates that escape.  For pulverized units this escape number is  0.11, for cyclone units
0.35, and for stokers 0.30.
11.  The breakdown of the boiler population used was:  pulverized dry 35% and stoker  65%.  The particulate control efficiencies and the
percentage of boilers using controls were multiplied and subtracted from unity to obtain the fractions of particulates that escape.   For
anthracite coal this escape number was estimated to be 0.40 based on age and control efficiency data.
12.  The breakdown of the boiler population was:  38.9% tangential firing and 61.1% for all others.
13.  The breakdown of the boiler population was:  38.9% tangential firing and 61.1% for all others.   The particulate control escape number
was 0.90.  Total arsenic input was calculated to be 0.15 metric tons/yr.
14.  Fuel  consumption is taken from PPG  (1972).  Arsenic concentration data are taken from API (1973).  Emissions are calculated by
multiplying fuel consumption by arsenic concentration.  Total arsenic input was calculated to be 0.059 metric tons/yr.
\

-------
                            TABLE IV

              ARSENIC EMISSIONS: Concentrations of
                Arsenic in Fuels and Waste Products
SOURCE
     CONCENTRATION,  mg/kg by weight
Fuel           Fly Ash          Bottom Ash
Coal
    Bituminous
    Anthracite
Petroleum
    Residual
    Distillate
Synthetic Fuels
            o
    Coal gas
    Oil
Geothermal
28.01
14.02
 5.44
10.06
 0.30'
 0.04'
680 - 1,700'
0.448V
    0
    0
1.  Magee,  et al.  (1973).   Average values were calculated by each coal
producing region.
2.  Natusch,  et al.,  1974.  Arsenic concentration is dependent upon
flyash particle diameters:  >11.3fx  - 680 mg/kg; 7.3 -  11.3(1 - 800 mg/kg;
4.7 - 7. SHI  -  1,000 mg/kg;  3.3 - 4.7fi - 900 mg/kg; 2.1 - 3.3^ - 1,200 mg/kg;
1.1 - 2.1H. -  1,700 mg/kg.
3.  EPA, 1978.  Arsenic concentration in ash pond  runoff.  The arsenic
concentration of slag is estimated to be on the order of 100 times less
than that of flyash.
4.  Ruch, et al., (1974).  Mean analytical data for  101 coals, 81 of which
are from the Illinois Basin. The remaining  10 coals are from the Eastern
United States (11 samples)  and  Western United States  (8 samples).
Standard deviation for the mean value is 17.70 mgjkg; minimum value found
0.50 mg/kg;  maximum  value found 93.00 mg7kg.
                                21

-------
5.  W. E.  Davis and Associates (1971) .  Data source unknown.
6.  Kessler, et al.  (1972).  Analysis  by SSMS.
7.  API (1973).
8.  Coal gas is not  currently produced commercially.  Preliminary studies
indicate that 35% of the arsenic content of coal remains in the gasifier ash.
9.  Hughes, et al.  (1974).  At this time,  oil is not produced commercially
from oil shale.  No  data are available concerning arsenic distribution
between product oil and  residue.
                                22

-------
COPPER SMELTING
        Arsenic, a significant contaminant of non-ferrous ores, is produced
as a by-product of primary copper smelting  at the ASARCo-Tacoma facility -
Although Federal and State air pollution regulations have significantly
reduced particulate emissions from copper smelters,, they  are still recognized
as a major  point source of atmospheric arsenic emissions.
        Arsenic behavior during the pyrometallurgical processing of copper
ore varies  from smelter to smelter and is dependent on many factors including
feed composition, temperature of the process material, blowing rates, gas
composition and product purity-  Thus, arsenic emissions may vary
significantly not only for different smelter configurations,  but  also for
                                .*
smelters of similar configurations .
        All of the data that appears in Table V was obtained from a
preliminary EPA draft of "Proposed National Emission  Standard for Arsenic
Emissions from Primary Copper Smelters" (EPA,  1979).  This information was
based on arsenic data received by EPA (in response to requests) from most
of the domestic copper  smelting facilities listed.  When incomplete information
was  received, EPA made assumptions regarding the behavior of arsenic at  a
specific facility based on existing knowledge of the smelting configuration
(or if need be, a similar configuration).  Assumptions regarding the
efficiencies of control devices (e.g., electrostatic precipitators) were
frequently necessary to close the material balance.  Although these material
balances represent a  substantial improvement in arsenic emissions data
(heretofore, most calculations were based on outdated emissions factors),
the precision of the data has not been verified.
        *
           In general, the amount of arsenic present in the feedstock will
affect the relative percentages of arsenic that are lost through volatilization
and land-destined wastes.
                                   23

-------
                                                            TABLE  V
                                 ARSENIC EMISSIONS: Primary Copper Smelting, metric tons per year
SOURCE
ASARCo
Tacoma7' 8' 9
El Paso9'10
Hayden11
Kennecott
Hayden13'14
IS Ifi
Hurley lb'ib
McGffi17'18
Garfield19'20
Phelps Dodge
Douglas22'23
iu .22,24
Morenci
Hidalgo22'25
Ajo26
Magma
29
Copper Range /White Pine
30
Inspiration
31
Cities Service/Copperhill
32
Anaconda
TOTAL
PRODUCTION
2 3
Input Contained

16,000
480
400

70

neg
70
1,100

85
75
160
550
35
5
60
neg
6,700
26,000

14,000
230
95 12

neg

~
neg
1,000 X

neg
neg
5
30
neg
neg
neg
-
1,600
17,000
Air4

560
35
200

35

—
60
10

40
30
5
310
25
3
5
-
180
33
1,500
DISPERSION
Land 5 Wafer 6

2,100
220
120

40

— —
10
85

50
50
150
220 -27
10
3
55
-
5,100
8,200
1.   EPA, 1979 - except as noted.  Data are reported to two significant  figures and are rounded to five metric ton increments;
    emissions below three metric tons/yr are listed as negligible.
2.   Input was calculated from arsenic feed data  (Ib/hr)  and is based on smelter operation of 8,000 hr/yr.  This operating factor was
    obtained from A. Vervaert, Office of Air Quality Programs and Standards, U.S. EPA, January 1979.

-------
 3.  Includes arsenic contained in blister copper and dust recycled to lead plant.  In the case of
    ASARCo/Tacoraa,  this figure also includes input to As^O, reduction plant.
 4.  Air emissions (i.e., stack) include fugitive emissions under existing control conditions.  Although these emissions
    were not included in the plant material balance, the error is assumed to be negligible.
 5.  This figure includes slag for dumping  and sludge.
 6.  All acid plant waste is assumed to be land disposed (i.e., liquid effluent, such as acid plant  purge
    water or scrubber sludge). Some scrubber blowdown waters are known to undergo  lime treatment and
    settling with discharge to a central pond (Bhattacharyya et al.,  1979).
 7.  Information on arsenic distribution at the Tacoma smelter was from Mr. K.W. Nelson, ASARCo  Incorpor-
    ated, by Mr. J. Padgett, EPA, March- 1976.
 8.  Assumptions were made  by EPA (1979) regarding the collection efficiency of the control equipment based on
    fugitive emission testing and operating temperatures.
 9.  Information on ASARGo)/Tacoma, ASARCo/El Paso, and ASARCo /Hayden received from Mr. K.D. Lough-
    ridge, ASARCo  Incorporated, by Mr.  D.R. Goodwin, Director, EPA, ESED, October 9,  1975.
10.  EPA distribution data is  based on tests performed in April, 1977.  Emissions data were calculated from
    basic assumptions of control equipment efficiencies and verified by telephone conversation by Mr. I. J.
    Weisenberg,  Pacific Environmental Services with Mr. W. R. Kelly, Smelter Manager,  ASARCQ /El Paso on
    October  12, 1978, The fact that the El Paso smelting complex also processes lead and  zinc
    was not reflected in the  mass balances.
11.  No smelter information is available on the distribution . of arsenic at ASARCo /Hayden.  Assumptions were
    made based on similarities to the ASARCo/Tacoma and ASARCo /El Paso smelters.
12.  Arsenic from the converter electrostatic precipitator is sent to El Paso .
13.  Data for Kennecott/Hayden received in the correspondence from Mr. I. G.  Pickering, Vice President
    Environmental Affairs, Kennecott Copper Corporation, to Mr. D. R.  Goodwin, Director, ESED, U.S. EPA,
    May 9, 1978.

-------
14.   The arsenic material balance obtained from  smelter personnel differs very slightly from that presented by EPA.
      To obtain a closed balance, several assumptions were made based on control equipment efficiency and
      tests performed.
15.   Data for Kennecott/Hurley received in the correspondence from Mr.  I. G. Pickering, Vice President
      Environmental Affairs, Kennecott Copper Corporation, to Mr. D. R. Goodwin, Director,  ESED, U.S.
      EPA, May 9,  1978.
16.   Arsenic feed and process contents obtained by the smelter personnel were based on  triplicate analyses
      of monthly composites.  Reverberatory  stack emission samples were obtained from isokinetic tests performed.
      Stack emission results, however, are not necessarily representative of a long-term material balance.
      Arsenic imput reported as 1.41 metric tons/yr.
17.   Data for Kennecott/McGill received in the correspondence from Mr. I. G. Pickering, Vice President
      Environmental Affairs, Kennecott Copper Corporation,  to Mr.  D. R. Goodwin, Director, ESED,
      U.S. EPA, May 9, 1978.
18.   As the arsenic balance obtained from measured values at the smelter is incomplete, assumptions were
      made to complete the materials balance.
19.   Data for Kennecott/Garfield received in the correspondence from Mr. I. G. Pickering, Vice President
      Environmental Affairs,  Kennecott Copper Corporation, to Mr. D. R. Goodwin,  Director, ESED,
      U.S. EPA, May 19, 1978.
20.   Mass balance calculations are in agreement with the distribution presented by Mackey et.al.,  1975 for
      the Nor an da Process.
21.   Dust from the  electrostatic precipitators is stored for future sale.
22.   Data for Phelps Dodge/Douglas, Phelps Dodge/Morenci, and  Phelps Dodge/Hildago received in the
      correspondence from Mr. R. W. Pendleton,  Jr., Vice President, Phelps Dodge Corporation to
      Mr.  D. R. Goodwin, Director, ESED, U.S. EPA, June 2,  1978.

-------
23.    The arsenic mass balance was based on reported data, estimated collection efficiencies of the hot electro-
      static precipitators, and assumptions made based on Stankovic, D.  "Air Pollution Caused by Copper
      Metallurgy Assemblies in Bor"  Institute for Copper, Bor, Project No. 02-513-1, U.S. EPA.
24.    No arsenic distribution was available; therefore, assumptions were made on the  basis of collection
      efficiencies for the control equipment, available information for the configuration of the  fluidized bed-
      reverberatory-converter, at Kennecott/Hayden and for the reverberatory furnace converter smelting
      configuration at  the old Kennecott/Utah smelter.
25.    Arsenic distribution information for the Hidalgo smelter was limited to spot checks at the facility.  To
      successfully close the material balance, the distrubution presented  by Koh(1972) for  a similar process
      configuration (and "low" arsenic input)  was considered.
26.    All of the information used in calculating the mass balance were referenced to Schwitzgebel, 1978.
      Sampling by plant personnel and EPA occurred during 1976-1977.  The results of arsenic sampling
      indicated that approximately 50 percent of the arsenic entering the smelter leaves in the reverberatory
      furnace off gas.   Furthermore,  as much as 90  percent of the arsenic entering the reverberatory furnace
      ESP may leave in the off-gas.  Arsenic rates at the ESP outlet ranged from  25-140 Ib/hr. (direct
      measurements).  It should be noted that  during a sampling period of a few days, the reverberatory
      furnace feed varies from 700 mg/kg to 7,000 mg/kg.
27.    Acid plant purge water is reported as 27 Ib/hr arsenic, in EPA  (1978) as opposed to the value of
      14.2 Ib/hr reported as acid plant waste in the EPA  (1979) mass balance for  the Ajo plant.
28.    Information on the input and materials balance was obtained from  D.R. Ridinger, Director of
      Environmental Affairs, Magma Copper Company to Mr. D.  R. Goodwin, Director, ESED,  U.S. EPA,
      dated April 4, 1978.   No information was given by the manufacturer on sampling procedures.  To close
      the material balance, EPA made assumptions concerning arsenic reporting in the blister, volatization in
      the converters, and ESP efficiency.

-------
29.   Information concerning the arsenic feed and distribution was obtained from correspondence from  J. W. Maksyn
      Environmental Control Engineer, White Pine Copper Division, Copper Range Company to D. R. Goodwin,
      Director, ESED,  U.S. EPA, dated September 7,  1978.  To obtain the arsenic mass balance, assumptions
      were made from data on similar process configurations.
30.   Arsenic data was obtained from correspondence  from Mr. M. W. Anderson, Inspiration Consolidated
      Copper Company to Mr. D. R.  Goodwin,  Director, ESED, U.S. EPA, dated  November 20, 1978.
      Information from the literature was used  to supplement incomplete smelter data in performing an arsenic
      distribution and mass balance.
31.   Information on the arsenic content of the copper concentrate was received from S. L. Norwood, Manager,
      Environmental Control, Cities Service Company to Mr. D. R. Goodwin,  Director, ESED, U.S.  EPA,
      dated September 5, 1978.   As arsenic input is  estimated at 0.36 metric tons /yr, the emissions were considered
      negligible, and any attempt at a distribution to be of questionable accuracy.
32.   Anaconda test data provided  smelter feed concentrates.  The mass balance presented is the result of
      (1) tests on process samples obtained from Anaconda; (2) a rough mass balance calculated by the EPA
      contractor; and (3) a refined mass balance provided by Anaconda (See:  Correspondence from
      Mr. R.  L.  Sloan, The Anaconda Company, to Mr. I. J. Weisenberg, Pacific Environmental Services,
      October 11, 1978).
33,   Process gases from all three  major smelting operations are combined into a main flue  duct and transported to a
      filtration plant for particulate removal prior to being discharged to the atmosphere through the main stack.

-------
COPPER REFINING
        The blister copper, derived from copper concentrates, may be fire-
refined to "anode" copper with a specification of 75 mg/kg maximum arsenic
content. Electrolytic refining is reported to reduce the arsenic content to
approximately 5 mgVkg.
        The data presented in Table VI are largely based on average
arsenic concentrations of copper during the various stages of refining, as
reported by EPA (1976a).  An inconsistency in the value cited for  the average
arsenic content of the input anode copper (1,000 mg/kg) could not be reconciled
with the reported  specification for fire-refined casting copper  (i.e., anode
copper).  However,  as the arsenic content of sampled blister and anode
coppers were approximately the same, the arsenic loss during the casting
process was assumed to be negligible (EPA, 1976a).   The arsenic removed
from the anode copper is found  in the slime from the  electrolysis  cells  and in
the electrolyte- The resulting arsenic-bearing wastes are disposed of on land
as a slag/sludge.
        Conflicting estimates of arsenic emissions attributable to  refined
copper production are presented by NAS(1977),  as derived from  Davis and
Associates  (1971).  These figures appear to be useless,  as they combine
smelting and  refining operations, neglecting to point out that nearly all of
the arsenic emissions from  refining are land-destined (See Table  VI).
Furthermore, using this  emission factor (apparently based on 1968 data)
and 1974 production figures gives a combined result which is substantially
lower than  that for copper smelters alone in 1974.
        Lake copper, or native  copper, is not derived from concentrates.
Arsenic content varies from 25 mg/kg to 5,000 mg/kg.   An average value of
500 mg/kg  was  used by EPA(1976a) to calculate the total quantity of arsenic
contained in this copper.
                                   29

-------
                                                                       TABLE VI

                                              ARSENIC  EMISSIONS:  Copper Refining, metric tons per year1
SOURCE

Electrolytic
Casting
Lake
PRODUCTION
2
Input
1,4003
6309
3012
Contained
74
7io

DISPERSION
Air Land
-5 1,4006
11
-

Water
407 4008


               1.    All data are reported to two significant figures and rounded to 5kkg increments.
               2.    Minerals Yearbook,  1975.  All data for primary refined copper (domestic and foreign ores) are for  1974.
               3.    EPA, 1976a.  Based on an average arsenic content in the input anode copper of 1,000  mg/kg.  Electrolytic copper refined in
               1974 totaled  1,400,000 metric tons.
o              4.    EPA, 1976a.  Electrolytically-refined copper contains arsenic at levels reported as  1 to 10 mg./kg and 4 to  11 rug/kg.
               Assuming an average arsenic concentration of 5 mg/kg, the quantity of arsenic in the product copper is estimated to be
               7 kkg/yr.
               5.    NAS, 1977. An arsenic emission factor of 0.000955kg/kkg of copper produced  (smelting and refining) was presented based
               on Davis  (1971) and the Minerals Yearbook,  1968.  However,  an average arsenic emission of 0.00105kg/kkg was reported for copper
               smelters in 1974.  These emission factors cannot readily be reconciled, as electrolytic refining results in land-destined emissions
               rather than in atmospheric emissions.
               6.    EPA, 1976a.  Arsenic residues are reported to be land-disposed as slag.
               7.    EPA, 1976a.  Effluents  from electrolytic refining were reported as  0.03kg arsenic/kkg copper product (See:  EPA,  1975).
               8.    EPA, 1978. An estimated total annual discharge from copper manufacturing based  on  an  unsubstantiated emission factor
               and 1976  arsenic trioxide production.
               9.    EPA, 1976a.  Based on average arsenic content in the input blister copper of 88 mg/kg.  Casting copper production
               totaled  79,000 metric tons in  1974.

-------
10.  EPA,  1976a.  Based on a specification of 75 nag/kg maximum arsenic content of fire-refined casting copper.  The discrepancy
between this specification and the average arsenic content of anode (i.e., fire-refined copper)  copper reported in Footnote 3
could not be reconciled.
11.  EPA,  1976a.  Discrepancies between the arsenic specifications for fire-refined copper and the reported arsenic
concentrations of anode copper,  make calculation of land-destined wastes meaningless.
12.  EPA,  1976a.  Lake or elemental copper is not derived from concentrates.  Furthermore, there are three grades of lake copper:
prime, which contains 25  mg/kg arsenic; natural, which contains 200 to 600 mg/kg arsenic; and arsenical,  which contains 600 to
5,000 mg/kg arsenic.  Thus, an average value of 500 mg/kg was used to calculate the total quantity of arsenic  in lake copper (based
on production of 60, 000 metric tons in 1974).

-------
PRIMARY LEAD AND ZINC PRODUCTION
        The production of primary zinc and lead, requiring smelting and
pyrometallurgic or electrolytic refining methods, are additional sources of
arsenic emissions to the environment.  The estimates of input arsenic, as
shown in Table VII, unfortunately are based on averages of foreign ore
concentrates.   Atmospheric emission factors calculated  by Davis (1971) or
Anderson (1973) have not been substantiated; furthermore, such factors do
not apply to electrolytic slab zinc production.  Changes in zinc smelting and
refining facilities, as indicated by the Minerals Yearbook, 1975, are not
reflected by these figures.
        NAS states no new data on emission from zinc and lead smelters
were available for inclusion in their report (NAS, 1977).  However, on
the basis of indicated smelter inputs of arsenical and nonarsenical concentrates
(including some supplementary information obtained on the  arsenic content
of ores and concentrates) and estimate stack losses, NAS derived recovery
factors similar to those arrived at earlier by Davis  (1971) -
                                  32

-------
                                                                      TABLE  VII
                                     ARSENIC EMISSIONS:  Primary Lead and Zinc Industries, metric tons per year
00
CO
SOURCE
Lead 2
Zinc




Input
1,100



in
310iU I
21010 f
PRODUCTION
	 2
Contained Emission Factor
204 0.00045
0.00067


c 11
0.00065 '
2151J
Air
240 6
300 8
9
70

190
DISPERSION
Land Water
800



12
120 neg
1.  Data are reported to two significant figures and are rounded to 5kkg increments; emissions below 3kkg are considered to be negligible.
2.  Suta,  1978.  Fugitive emissions are estimated to be 10 percent of stack emissions.
3.  EPA,  1976a.  An average Pb/As concentration ratio of  1,400  mg/kg (based on foreign ore concentrates) was used to estimate the overall
quantities of arsenic contained in lead concentrates.  Based onthis concentration and lead production of 610,000 kkg/yr the quantity of arsenic
is calculated to be 850 metric tons/yr. An additional 210 metric tons/yr enters the primary lead industry via residues from the electrolytic
zinc industry.  Using the emission  factor of Davis  (1971),  the arsenic lost to the atmosphere was  computed to be 240kkg.  No appreciable
quaatities of arsenic were found in the wastewaters from the primary smelters.  Thus, the remaining 610kkg of arsenic, plus 210kkg/yr
in residues from the zinc industry, were distributed between land-destined wastes and that portion retained in the refined lead (EPA, 1976a).
4.  EPA,  1976a. Based on an annual primary lead production of 610,000 metric tons in 1974 containing 35 mg/kg arsenic  (an average of ASTM
standard B29-55 which  specifies an arsenic content of 20' mg/kg  for undesilverized lead and 50 mg/kg for desilverized lead),  refined lead contains
approximately 20 metric tons of arsenic.
5.  Emission factor calculated by Davis (1971) based on 1968 estimates. This factor was obtained through personal contact with processing
companies, and on engineering calculations supported by data concerning smelter  flow rates and temperature.
6.  EPA,  1976b.  Estimated  arsenic emissions for primary lead production in the U.S. for 1974.  This figure was calculated using EPA
emissions  factors (0.0004kg/kkg of lead  produced) and by applying ''appropriate levels of control".  (See:  Anderson, 1973).

-------
7.  NAS,  1977.  Arsenic emissions factor for zinc smelting and refining based on Davis (1971) and Minerals Yearbook,  1968.
8.  Minerals Yearbook, 1975.  Based on 1974 primary slab zinc production of 504,000 metric tons.
9.  EPA,  1976b.    Estimated arsenic emssions for primary zinc production in the U.S. for 1974.  This figure was calculated  using  EPA
emissions factors (0.5 kg/kkg of  zinc produced) and by applying  "appropriate levels of control".  (See:  Anderson, 1973.)
10.  EPA,  1976a.  An average  As/Zn concentration  ratio of 1,050 mg/kg (based on foreign ore concentrates) was used to estimate the total
quantity of arsenic in primary zinc.  Based on a 1973-1974 production level of 490,000 kkg/yr of slab zinc, an estimated 290,000 kkg/yr
are produced  from pyrometallurgical zinc smelters and an  estimated 200,000 kkg/yr at an electrolytic facility.
11.   EPA,  i976a.  This emission factor applies only to pyrometallurgical zinc smelters.
12.   EPA,  1978.  An estimated 1.3 metric tons are discharged annually from  zinc smelting and refining.
13.   EPA,  1976a.  Residues shipped  to lead smelters contain almost all of the 210 metric tons/yr of arsenic originally present in  the zinc
concentrates electrolytically refined. Based on maximum levels of arsenic in refined  zinc of  all commercial grades, an estimated 5 metric
tons/yr is contained in zinc products.

-------
PRIMARY NON-FERROUS ORES,  MINING AND MILLING
                                                          i
        The mining and  milling of primary non-ferrous! metal ores are
another source of arsenic emissions for which little hard data are available.
Production figures are from the Minerals Yearbook,  1975, and pertain to
ore production except as footnoted (this may account for any inconsistencies
with figures presented by EPA(1976a)).  Emission factors reported by NAS
are based on 1968 data and cannot be verified (NAS, 1977).  NAS further
states that an estimated  40 percent of the arsenic in copper  or copper-lead-
zinc ore is left in the concentrator tailings,  which are deposited on land
and  subject to  wind losses.  Similarly,-  arsenic in gold and uranium mill
tailings are subject to  wind losses.  Reported water discharges are based on
EPA Effluent Guidelines  Division screening and sampling  data (EPA,  1978).
        As data are scarce on smelting and refining operations  for most
non-ferrous metals (with the exception of copper, zinc, and lead), a
separate table  has not been included.  EPA made a very rough  approximation
of the quantities and  fates of arsenic in these non-ferrous metal ores;
however, their analysis  (See Footnote 2, Table VIII) has not been verified
(EPA,  1976a).
                                   35

-------
                                                             TABLE  VIII
                                    ARSENIC EMISSIONS:  Primary Non-Ferrous Metal Ores Mining
                                                                                 1  2
                                                  and Milling, metric tons per year '
SOURCE
Copper
Lead

Zinc
Aluminum (Bauxite)
Antimony
Gold
Manganese
Mercury
Silver
Tin
Uranium
Vanadium
3 64
Production Emission Factor x 10
270,000,000 .45 $
600,000 .45

450,000 .45
1,900,000
600
240, OOO8 .45
270,000
75
1,100 .45
9
6, 500, OOO10 .45
4.80011
DISPERSION
Air Land Water
120 4005
neg 6
540
neg
2907

neg


neg

neg

1.  All data reported to two significant figures and rounded to 5kkg increments.  Emissions less than 3kkg/yr are considered negligible.
2.  EPA, 1976a.  With the notable exceptions of copper,  lead, and zinc,  little data are available on the arsenic content of primary non-
ferrous ores.  Further, data are scarce on arsenic emissions which occur during mining, milling, smelting, and refining operations.
Based in part on antimony data  (e,g.,  arsenic occurs at  a concentration  of roughly 3 percent  in antimony ores) and in part upon
similarities to  the lead-zinc industry, EPA (1976a) assumed that:  arsenic is 1 percent of the  quantity of gold, silver,  mercury,
uranium, vanadium,  and antimony concentrates; furthermore, one-third is recovered as sulfides, that one-third is lost to air emissions,
and that one-third is in land disposed wastes.

-------
 3.  Minerals Yearbook, 1975.  All mine production data for 1974, except as noted.
 4.  NAS, 1977.  Based on Davis (1971) and Minerals Yearbook,  1968; reported in terras of metric tons/metric tons of ore produced.
'5.  EPA, 1978.  Based on Effluent Guidelines Division  screening sampling data for copper mine effluents and average discharge rates
 for 9 direct discharges.
 6.  EPA, 1978.  Based on Effluent Guidelines Division screening sampling data for lead/zinc mines and average discharge rates for
 34  direct discharges.
 7.  EPA, 1978.  Based on estimated annual discharges  from two plants engaged in the processing of bauxite ore to produce alumina
 (with average discharges of .26 metric tons/day for 300 days).
 8.  Minerals Yearbook, 1975.  Represents  gold ore and old tailing to mills in 1975.
 9.  Minerals Yearbook, 1975.  Information withheld for proprietary reasons.
 10.  Minerals Yearbook, 1975.  Represents mine production of ore (not U0O0 content) and does not include output of approximately
    	                                         o o
 200 tons of U_00 from mine waters, leaching operations, and recovery from phophate rock processing.  The U0O0 content of the ore
             •JO                                                                                        O  O
 after milling was 11,000 metric tons.
 11.  Minerals Yearbook, 1975.  Mine production of vanadium in the  U.S.,  as measured by receipts of uranium and vanadium ores and
 concentrates at mills (vanadium content).

-------
MANGANESE PRODUCTION
        Manganese ferroalloys are produced from manganese ore by smelting
in blast or electric furnaces.  Since arsenic data for manganese smelting are
scarce, EPA(1976a) has estimated these emissions by making an analogy
"to the transport of the chemically-similar phosphorus:   60 percent of the
phosphorus in the ore passes into the ferroalloy, 30 percent passes into
the slag  and 10 percent escapes with furnace gases." (See Table IX).
        About 10 percent of the manganese ore is used for making carbon-
zinc and alkaline manganese dioxide dry cell primary batteries, and for use
in the chemicals and glass industries.  The estimated  400 metric tons  per
year of arsenic dissipated in  such end products was based on an average
manganese ore content of 0.20 percent.  This value was derived from
analyses of three foreign manganese ores (i.e.,  Brazilian and Mexican)*
with arsenic concentrations ranging from 0.15 to 0.25 percent (EPA,  1976a).
        *EPA, 1976a.  Manganese ore consumed in the U.S.  with 35 percent
or more manganese content is  imported principally from Africa and Brazil.
                                  38

-------
                                                            TABLE  IX
                                  ARSENIC EMISSIONS:  Manganese Production,  metric tons per year
                                                                                                 1
    SOURCE
                                               PRODUCTION
                                         Input           Contained
                                                                                            Air
                                          DISPERSION
                                              Land
                             Water
    Smelting
    Non-ferrous manganese
      ore consumption
                                         4,000
3,200V
                                                             400
10'
350'
00
10
1.  All data are reported to two significant figures and rounded to 5kkg increments.
2.  EPA, 1976a.  Based on an average domestic industrial consumption of manganese ores of 2.0 million metric tons per year and an average
value of 0.20 percent arsenic.
3.  EPA, 1976a.  Little data are available as to the fate of arsenic during the smelting process.  Using the analogy of phosphorus  in ferroalloy
production, an estimated 2,200 metric tons per year are retained in ferroalloys  (consumed in iron and steel)  and 1,100 metric tons per year
contained in slag from ferroalloy furnaces.  (As  a result of independent rounding, the total is higher than the contained figure listed.)
4.  EPA, 1976a.  Estimated air emissions from furnaces.
5.  EPA, 1976a.  Estimated dusts collected from furnaces.
6.  EPA, 1976a.  Estimated arsenic content of non-ferroalloy manganese ore dissipated in end-products.

-------
IRON AND STEEL PRODUCTION
        Arsenic emissions from iron ore production and steelmaking are
presented in Table   X, compiled largely from data presented by EPA (EPA,
1976a).   According  to NAS, only insignificant quantities of arsenic are
emitted  during iron and steel production (NAS,  1977).
        The average arsenic concentration in iron ore requires further
analysis; EPA (1976a) assumed it to be 400 mg/kg for purposes of their
study.  NAS (1977) reports a range of 1 to 2,900  mg/kg of arsenic in  sedimentary
iron ores, with an average value of 400 mg/kg.  Therefore,  all figures based on
an input arsenic concentration of 400 mg/kg should only by considered as a
rough approximation (i.e., order of magnitude indicator).
        As is  discussed in the footnotes  of Table  X, EPA's estimates are
based on 5-year averages of iron ore consumption and pig iron production.
Several assumptions are made  as to the arsenic losses during steelmaking,
based on analogies to phosphorus, for both basic oxygen  and basic open
hearth processes. A 99 percent dust collection efficiency was assumed to
obtain the air emission; the collected dust is presumed to be disposed of as
landfill.  The  largest portion of arsenic input is contained in the slag and
resulting steel products.
                                  40

-------
                                                        TABLE X
                      ARSENIC EMISSIONS:  Iron Ore Production and Steelmaking, metric tons per year
SOURCE

Crude Ore2
Iron Ore/Pig Iron

Cast Iron
Steelmaking

Input
88.0003
54.0005
7

PRODUCTION
Contained

54,0006
0
3,300°
51,00010 49,500n

Air
94

q
20
1212
DISPERSION
Land Water


q
100
10
1,2001J
1.   All data are reported to two significant figures and rounded to 5kkg increments.
2.   Minerals Yearbook, 1975.  In 1974, 220,000,000 metric tons of crude iron ore were mined in the U.S.
3.   EPA,  1976a.  An arsenic concentration of 400  rag/kg is reported in iron ore but is unsubstantiated.   No data was reported on
arsenic emissions attributable to iron ore mining.
4.   Anderson, 1973.  An unsubstantiated mining emission factor of O.lkg/kkg of arsenic present in the ore is reported.
5.   EPA,  1976a.  Based upon  an assumed arsenic concentration of 400 mg/kg (from U.S.G.S. and U.N.  references), and an average
iron ore consumption from 1970 to 1974 of 134.5 million metric tons per year.
6.   EPA,  1976a.  The arsenic is assumed to be retained in the pig iron.
7.   EPA,  1976a.  A cast iron  production level of 17.2 million metric tons per year is reported; the 2.4 million metric tons per year
of pig iron, which is  used  for cast iron production via cupola and similar furnaces, is augmented by 14.8 million metric tons per year
of scrap.   No arsenic content of the scrap was reported.
8.   EPA,  1976a.  A reported average of 5 million metric tons per year of pig iron  were consumed during 1970-1974 for cast iron
products.  Thus, with an arsenic concentration of 650 ppm,  3,300 metric tons are retained in cast iron.
9.   EPA,  1976a.  Based on an average uncontrolled emission factor for arsenic from cast iron production of 0.007kg/kkg of metal
charged (See:  Anderson,  1973).  An estimated 20 metric tons per year are released to the atmosphere, with the remaining 100 metric
tons collected and disposed of  on land.

-------
10.  EPA,  1976a.  Based on an average 78 million metric tons per year of pig iron consumed in steelmaking with an average arsenic
content of  650 ppm.
11.  EPA,  1976a. Includes an estimated 14,900 metric tons per year of arsenic contained in steel and an estimated 34,600 metric
tons per year in slag from steelmaking furnaces.  The latter estimate was not substantiated; furthermore, as the slag is widely
used, (e.g., railroad ballast, highway base or shoulders, etc.) it was categorized as contained arsenic.
12.  EPA,  1976a.  Assumes a  99 percent dust collection efficiency and  that the arsenic in  the uncontrolled dust emissions is the
same concentration level as it is in the steelmaking charge.
13.  EPA,  1976a.  Collected steelmaking dusts are assumed to be land disposed.

-------
PHOSPHORUS PRODUCTION
        Arsenic is a common trace constituent of phosphate rock and, for
Florida rock at least, is a linear function of iron oxide present in the rock
(Stow, 1969).  Arsenic content of Western rock, has been suggested to be
dependent upon the organic carbon content  of the rock (Gulbrandsen, 1966).
Arsenic input is based on quantities of marketable phosphate rock consumed
in 1972, phosphorus concentration for each type of rock, and As7P2O5
ratios for each type of rock (See Footnote 2 of Table  XI).
        The arsenic present in phosphate rock is suggested to follow
phosphorus quantitatively, whether the acidulation (conversion to phosphoric
acid) or the furnace process  (reduction to elemental phosphorus) is used.
Phosphorus is not recovered quantitatively from phosphate rock;  phosphorus
(and arsenic)  not recovered is assumed to be land-disposed in sludge ponds.
In Table XI non-agricultural production represents phosphorus produced
by the furnace process. Arsenic is removed from food  grade phosphoric
acid, generally by sulfide precipitation; arsenic is also removed in the
manufacturing processes of phosphorus pentasulfide,  phosphorus trichloride,
and phosphorus oxychloride.  EPA has estimated that all arsenic associated
with miscellaneous uses  is land-disposed  (EPA, 1976a). Arsenic present as
an impurity in phosphate detergents is assumed to be discharged to surface
waters.   All other arsenic is assumed to be dissipated with the end product.
                                  43

-------
                                                             TABLE XI
                                 ARSENIC EMISSIONS:  Phosphorus  Production, metric tons per year
SOURCE
Production (Total)
Florida Rock
Domestic fertilizer
Animal feed
Exports
Tennessee Rock
Non-agricultural
Western Rock
Domestic fertilizer
Animal feed
Non-agricultural
Exports
Use
Domestic fertilizer
Animal feed
Detergents
Phophoric acid (food grade)
Miscellaneous
Export
PRODUCTION DISPERSION
23 3
Input Contained' Air Land Water
770
450 170 05
240
15
190 190
40 30 10
40
270 200 60
50
15
190
15 15
290 neg4 290
30 30
110 110 110
60 60
60 60
200 200
1. EPA, 1976a.  Based on 1972 marketable phosphate rock consumption for 1972.  Data are reported to two significant figures and are
rounded to the nearest 5kkg increment; emissions below 3kkg are considered to be negligible.
2. Arsenic input is calculated on the basis of the following As/P205 ratios:  Florida rock - 0.000045; Tennessee rock -  0.000082;
Western rock - 0.000230.

-------
3. EPA, 1976a.  Arsenic in phosphate rock is suggested to follow phosphate recovery quantitatively  (EPA,  1976a).  Arsenic
dispersion is thus calculated on the basis of phosphate recovery from marketable rock:  Florida rock  - 66%;  Tennessee rock - 73%;
Western rock - 77%.  Arsenic not recovered is assumed to be land-disposed in sludge ponds.
4. Anderson, 1973.
5. EPA, 1976a.  Arsenic in household detergents and presoaks was measured at concentrations ranging from 2 to 59 ppm
(Angino, et al. 1970).  The production of trisodium phosphate has been reported to have been reduced because of the environmental
concern over phosphorus in wastewaters.

-------
BORON PRODUCTION
        Approximately 80 percent of the U.S. production of borax is
from ore deposits, the remainder is obtained from saline brines.  The arsenic
content varies widely within a given ore horizon  (e.g., from 0 to  1,000  mg/kg);
however, EPA indicates that the residue from the digested ore contains
approximately 45 mg/kg of arsenic (EPA, 1976a) . Saline brines pumped from
beneath the crystallized surface of Sear-les Lake in Southern California
contain 500  mg/kg sodium arsenate, or  the equivalent of 180 mg/kg arsenic.  As
the brines are processed and returned  directly to the lake (plus added
process waters), almost all of the  arsenic initially extracted from the lake
(with the exception of that contained in the borax product) is assumed
returned to  the lake bed.
        Table XII  presents arsenic emissions for boron production based  on
EPA calculations (EPA,  1976a).  A value for land-destined arsenic wastes
resulting from borax ore production was derived  from the estimated
arsenic content of the ore residue and the total amount of residue produced.
The concentration of arsenic contained in the borax  ore was arrived at by
calculations. It  was assumed that the arsenic found in wastewaters  from
boric acid production represents all of the arsenic originally present in
the technical grade borax. No wastewater sampling  data were presented.
                                 46

-------
                                                              TABLE  XII i.
                                     ARSENIC EMISSIONS:  Boron Production, metric tons per year
SOURCE
2
Production

Borax Ore
Borax from Brines
Borax Use
Boric acid
o
Other borax products
Input

0
5(T
2,200

5
20
PRODUCTION
Contained

/
20*
54


20
DISPERSION
Air Land

c
305
2,2006



Water





5

 1.  EPA,  1976a.  Data are reported to two significant figures and are rounded to the nearest Skkg increment; emissions below 3kkg are
 considered to be negligible.  U.S. production of boron minerals has averaged 1,070,000 metric tons per year from 1972 to 1974 which
 corresponds  to an annual boron oxide production of 580,000.
 2.  Commodity Data Summaries 1975;  Minerals Yearbook 1972.  Approximately 80% of U.S.  production  is from ores and 20% from brines.
 3.  EPA,  1976a.  Arsenic occurs in widely varying concentration in a given ore horizon.   This  figure represents the sum of contained
 and dispersed arsenic.
 4.  EPA,  1976a.  This figure is derived by assuming that the arsenic in boric acid wastewaters represents the entire amount of arsenic
 in borax raw material.  Thus,  technical grade borax is estimated to contain approximately 21mg/kg of arsenic.
 5.  EPA,  1976a.  Residue from digested ore is approximately 0.8 metric tons per metric ton of borax product. This residue contains
 approximately 45 mg/kg of arsenic.  This residue is placed in ponds.
 6.  EPA,  1976a.  Sodium borate is extracted  from Searles Lake  brinei, which contains 500 mg/kg sodium arsenate  (equivalent to 180 mg/kg
                                                      g
 arsenic).   Total brine processed is approximately 12 x  10  liters per year.  Depleted brines plus process  wastewaters are returned to
 the lake.
 7.  EPA,  1976a.  110,000 metric tons of boric acid are produced annually.  36 grams  of arsenic are discharged in process wastewaters
per metric ton of boric acid produced.
 8.  EPA,  1976a.  Consumption  of other borax products  is approximately 880,000 metric tons per year.  Arsenic concentration in this
 borax is assumed to be 21 mg/kg.

-------
WATER AND WASTEWATER
        Arsenic input and emissions from individual water/wastewater
facilities depend on the individual raw wastewater loadings and resultant
removal efficiencies.  Variation of input is largely dependent on industrial
discharges in the case of POTW's, and on the raw water source of a water
treatment plant.  As shown in Table XIII, arsenic emissions from POTW's
are approximated from (1) estimates of dry sludge generation and the
average arsenic content of the sludge and ;  (2) estimates  of the effluent
quality and total combined daily flow.  In the case of water treatment plants,
arsenic emissions are described in terms of per capita water consumption
and treatment efficiencies.  Thus, all of the numbers presented can only
be considered  as rough approximations of the actual arsenic emissions.
                                    48

-------
                                                        TABLE  XIII
                        ARSENIC EMISSIONS:  Water and Wastewater Treatment, metric tons per year
SOURCE
Input
POTW2'3'
Municipal Water Treatment
DISPERSION
Air Land
4 5 4,5
neg ' neg
65 7

Water,
1606

1.  All data are reported to two significant figures and rounded to 5 kkg increments.  Emissions below 3 kkg/yr are considered
negligible.
2.  Publically owned treatment works.
3.  EEA,  1976a.  At one secondary treatment plant, the arsenic  in thickened waste sludge was reported as 61 H-g/kg; at an assumed
8 percent solids content, the sludge solids would have contained  0.75 mg/kg arsenic.  Assuming that per capita sludge generation
is 0.09 kg/day and 120 million people are served by POTW's, (generating 4 million kkg/yr of dry sludge) 3 metric tons of arsenic
are contained in sludge.
4.  EPA,  1976a.  An arsenic emission factor or 0.01 kg/kkg of  "sewage and sludge" is reported by EPA (Anderson,  1973).  By
assuming  a solids concentration of 20 percent in dewatered sludge (feed to an incinerator), the emission factor is then equivalent to
an arsenic concentration in dry sludge of 2 mg/kg, implying that all the arsenic is volatized.  According to EPA (1976a), one-third
of all sludge is incinerated, thus, one  metric ton of arsenic is emitted to the air, while 2 kkg are applied-to land.
5.  Furr,  et al. 1976.  Based on an average of reported arsenic concentrations found in municipal sludges from 16 American cities,
arsenic concentration is 14 mg/kg on a dry weight basis.  Assuming 4 million metric tons of dry sludge are generated per year,
56 metric  tons of arsenic are contained in sewage sludge.  What portion of this sludge is land-disposed or incinerated is unknown.
6.  EPA,  1978. Based on calculated averages of arsenic concentrations from selected facilities and a total combined daily flow
from POTW's of 23,000 mgd.
7.  EPA,  1976a.  Based on an estimated removal rate of 2.7 jig/liter per year and an estimated per capita water use of 200,000
                         12
liters per year (or  24 x 10   liters per year for 120 million people).

-------
                             SECTION III

                     SUMMARY AND CONCLUSIONS
        The purpose of this study was to compile existing information
on arsenic emissions to the environment from natural and anthropogenic
sources in the form  of  preliminary materials balances.  Although verification
of existing data was outside the scope of work, a major goal of this report
was to establish a range of uncertainty for the figures cited.  This
objective was achieved by presenting the arsenic emissions data in a fully
annotated tabular format.
        Prior to this study, considerable effort had been expended by the
EPA and NAS to assess the environmental and health effects of arsenic and
its compounds.  While their studies required evaluation of emission sources,
much of the arsenic data represents a "rehashing" of old  numbers (i.e., many
numbers were originally educated guesses or unsubstantiated  engineering
judgments.   Furthermore, certain emissions figures, as noted within the
text of the report, appear of little use because (1)  primary sources
neglected to indicate error ranges;  and (2)  emissions factors were not
properly utilized (e.g., emissions factor applied to incorrect production  or
arsenic input data).
        What the existing  arsenic data has provided, however, is a framework
for constructing gross estimates of arsenic distribution in the environment.
From  the results of this preliminary  analysis,  three major sources of
environmental emissions emerge:  energy production,  arsenic  production and
use, and primary copper smelting.  Other nonferrous smelters, specifically
primary lead and zinc smelters,  may represent another significant source
of arsenic to the environment.  In terms of human exposure,  other sources
appear to be of much less  consequence.
        The most complete information on arsenic release to the environment
was found for primary copper production; the data is contained in a recently
developed preliminary draft of arsenic discharges from individual copper
smelters (EPA,  1979).  This, unfortunately,  is not the case for other point

                                  50

-------
source categories.
        Multimedia arsenic materials balances for primary lead and zinc
smelting were not available at the time this report was prepared. Most data
points cited were based on emission factors that may not adequately reflect
current control technologies.  No discussion of secondary nonferrous  smelters
was included in this report, as arsenic data is limited.  According to Suta
(1978),  estimated arsenic emissions from secondary smelting operations
results in insignificant atmospheric concentrations.
        Arsenic emissions from the energy sector are dependent on fuel
consumption and feed concentrations.  While fuel consumption is relatively
well-qualified, simple averages are inadequate to describe arsenic concentrations
in 4,000 coals.
        Finally, most reported figures for arsenic emissions during arsenic
production and use (i.e., in the manufacture of commercial products or
from arsenic added to  commercial products) are speculative.  This is a direct
result of the proprietary nature of such data.
        Recommendations for further data collection and evaluation include:
        •   Examine existing arsenic concentration data for U.S. coals.  This
            information can be located in the files of the U.S. Geological
            Survey (Coal Resources  Branch) and Penn.  State Data Base.
            Pursue other avenues of data generation, such as on-going
            studies sponsored  by other government agencies (e.g..
            Department of Energy) or the National Coal Association.  Map
            arsenic emissions (to all three media) for coal-fired electric
            utilities on the basis of capacity, existing control technologies,  and
            representative arsenic feed concentrations.
        •   Extend existing data sources to include EPA proprietary files and,
            if possible, Stanford Research Institute manufacturing/marketing
            projections for arsenical pesticide production.
        •   Obtain arsenic data collected for primary non-ferrous smelters ana
            held  by NIOSH (See:  "Control Technology Summary Report on
            the Primary Non-Ferrous  Smelting  Industry", Radian Corporation,
            December  1978).
         •  Verify existing emission factors for non-ferrous smelters.  Evaluate
            emissions from secondary  smelters, specifically those processing
            anitmonial lead.
                                  51

-------
        •   Review existing literature for arsenic concentration data that
            could be of use in evaluating unsubstantiated emission factors
            (See:  Suta,  1978; Union Carbide Corp.,  1S76; and Holt and
            Moberly,  1976).

        •   Address arsenic emissions from non-point sources,, such as mining
            and agricultural runoff.  Compare findings with existing leachate
            data to determine potential hazard.

After completing these tasks, the need for a sampling and analysis program must

be weighed  against the population at risk.
                                 52

-------
                               REFERENCES
Anderson,  David,  "Emission Factors for Trace Substances",  U.S. Environmental
Protection Agency, EPA-4507273001, December 1973.

Angino, E, E., L, M. Magnuson,  T. C, Waugh, O. K. Galle, and J. Bredfeldt,
"Arsenic in Detergents:  Possible Danger and Pollution Hazard", Science 168:
388-90, April 17, 1970; Discussion 170:  870-2, November 20, 1970.

API,  "Validation of Neutron Activation Technique for Trace Element
Determination in Petroleum Products", American Petroleum Institute, API
Report 4188,. Washington, D.C., 1973.

Attari, A., "Fate of Trace Constituents of Coal During Gasification",
EPA-650/2-73-004, U.S.  Environmental Protection Agency, 1973.

Bhattacharyya, D.,  A. B.  Jumawan, Jr., R.  B.  Grieves, "Charged Membrane
Ultrafiltration of Heavy Metals from Non-ferrous Metal", Journal of Water
Pollution Control Federation 51, 76, 1979.

Blackwood, T. R.  and R. A. Wachter,  "Source Assessment:  Coal Storage
Piles", U.S. Environmental Protection Agency, EPA-600/2-78-004k, May 1978.

Boyle, R. W. and I. R. Jonasson,  "The Geochemistry of Arsenic and its Use as
and Indicator Element in  Geochemical,Prospecting", J. Geochem, Explor., 2^, 251,
1973.

Bureau of Mines, 1975 Minerals Yearbook Vol  I:  Metals, Minerals, Fuels,
U.S.  Department of Interior, Bureau of Mines, 1977.

Commodity  Data Summaries, 1975, Bureau of Mines,  1977.

Conway, E. J.,  Mean Geochemical Data in Relation to Oceanic  Evaluation",
Proc. R. Irish Acad. 48,  152, 1942.

Chu,  T. J., R. J. Ruane, and G. R. Steiner, "Characteristics  of Wastewater
Discharges  from  Coal-Fired Power Plants", Proceedings of the  31st Industrial
Waste Conference,  Purdue University,  1976.
                                   53

-------
Cox,  D.  B.,  J. Chu and R. J. Ruane,  "Quality and Treatment of Coal Pile
Runoff", NCA/BCR Coal Conference, 1977.

Davis, W. E. and Associates,  National Inventory of Sources and Emissions:
Arsenic - 1968, PB 220 619, May 1971.

Durum, W. H., "Occurrenceof Some Trace Metals in Surface Waters and
Groundwaters",  Proceeding Sixteenth Water Quality Conference,  University of
Illinois, February  12-13, 1974.

EPA, 1974, "Development Document for Effluent Limitations Guidelines
Documents and New Source Performance Standards for the Steam Electric Power
Generating Point Source Category", EPA 44071-747029-3, U.S. Environmental
Protection Agency, 1974.

EPA, 1976a,  "Technical  and Microeconomic Analysis of Arsenic and  Its
Compounds'.^ U.S. Environmental Proetection Agency, EPA 560/6-76-016,
April 1976.

EPA, 1976B,  "Air Pollutant Assessment Report on Arsenic",U.S.  Environmental
Protection Agency, Office of Air Quality Planning and Standards, Research
Triangle Park, North Carolina,  1976.

EPA, 1976c,  "Development Document for Interim Final Effluent Limitations
Guidelines for the  Pesticide Chemicals Manufacturing Point Source Category",
U.S.  Environmental Protection Agency, EPA/44071-757060d, November 1976.

EPA, 1977, Screening sampling data for the steam electric power  generating
point source  category,  U.S. Environmental Protection Abency, Effluent
Guidelines Division,  1977.

EPA, 1978, "Gross Annual Discharge to the Waters in 1976: Arsenic", Revised
Report No. 2, U.S. Environmental Protection Agency, April 1978.

EPA, 1979, "Preliminary  Draft:  Environmental Assessment of  Arsenic Emissions
from Copper  Smelters",  U.S. Environmental Protection Agency, OAQPS,
Research Triangle  Park,  (DRAFT), January 1979.

Ernst and Ernst, "Wood  Preservation Statistics", Proceedings  of the American
Wood Preservers Association, 72 265, 1976.

FPC,  "Gas Turbine Electric Plant Construction Cost and Annual Production
Expenses", Federal Power Commission,  1972.
                                   54

-------
 FPC News, Federal Power Commission, 1974.

 Ferguson, I. F. and J. Gavis,  "A Review of the Arsenic Cycle in Natural
 Waters", Water Research, 6.  1259, 1972.

 Furr,  K,  A.  Lawrence, S.  S. C.  Tong, M,  C. Grandolfo, R. A. Hofstader,
 A. Bach,  W.  H. Gutenmann,  and  D.  J. Lisle,  "Multielement and Chlorinated
 Hydrocarbon Analysis  of Municipal Sewage Sludges of American Cities",
 Environmental Science and  Technology, _10, 683, 1976.

 Gulbrandsen, R. A.,  "Chemical Composition of Phosphorites of the Phosphoria
 Formation", Geochimica et Cosmochimica Acta, 30, 769,  1966.

 Holt, B. R., and J.  W.  Moberly, "Environmental Mass Balance of Arsenic",
 Stanford Research Institute,  1976.

 Hughes, E. E.,  E., M. Dickson, and R. A. Schmidt, "Control of Environmental
 Impacts from Advanced Energy  Sources", U.S. Environmental Protection
 Agency, EPA 600/2-74-002,1074.

 Judson, S.,  "Erosion of the Land",  American  Scientist, 56, 356, 1968.

 Kessler, T..  A. G. Sharrey, and R. A. Griedel, "Analysis of Trace Elements in
 Coal by Spark Source  Mass Spectrometry",  U.S. Bureau of Mines, RI 7714,  1972.

 Klein, D.  H., A. W. Anders, J. A.  Carter, J. F. Emery, C. Feldman,
 W. Fulkerson, W. S. Lyon, J. L.  Ogle, U.  Talmi, R. I. Van  Hook, and
 N. Bolton, "Pathways of Thirty-Seven Trace  Elements Through a Coal-Fired
 Power Plant", Environmental Science and Technology, 9:  973-979, 1975.

 Koh, S.,  "Extractive Metallurgy of Kuoko", presented at the joint meeting  of
 the MMIJ-AIME, Tokyo, May 24-27,  1972.

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

 Magee, E.  M., H. F. Hall,  and  G. M. Baige,  "Potential Pollutants in Fossil
 Fuels", U.S. Environmental Protection Agency, EPA R2-73-249, 1973.

 NAS, Medical and Biological Effects  of Environmental Pollutants: ARSENIC,
 National Academy of Sciences, Washington,  D.C., 1977.

 Natusch, D. F. S., J.  R. Wallace, and C. A. Evans,  "Toxic Trace Elements:
Preferential Concentration in Respirable Particles", Science,  183:  202-4,
January 1974.

Onishi, H., "Arsenic", in Handbook of Geochemistry, Wedepohl, K. H.  (ed.),
Springer-Verlag, Berlin, 1969.
                                   55

-------
Onishi, H.  and E. B. Sandell, "Geochemistry of Arsenic",  Geochim.  Cosmochi.
Acta 7, 1,  1955.

Ruch, R. R.,  H. J, Gluskoter,  and N, F.  Shimp, "Occurrence and Distribution
of Potentially Volatile Trace Elements", Environmental Geology Notes, No, 72,
August 1974,

Schwitzgebel,  K., R. T.  Coleman,  R,  V, Collins, R. M.  Mann, and C. M.
Thompson,  Trace Element Study at a Primary Copper Smelter, Vol I  and  II,
U.S. Environmental Protection Agency, EPA-60072-78-065a, 1978.

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

Stow, S.  H., "The Occurrence of Arsenic and Color Causing Components in
Florida Land-Pebble Phosphate Rock", Economic Geology 64,  667, 1969.

Sullivan,  R. J., "Air Pollution Aspects of Arsenic and its Compounds",  Litton
Systems,  Inc., PB  188  071, 1969.

Suta, Benjamin E.,  "Human Exposures to Atmospheric Arsenic", U.S.
Environmental Protection  Agency, September 1978.

Taylor, S.  R., "Abundance of Chemical Elements in the Continental Crust:  A
New  Table", Geochim. Cosmochim. Acta 2_8, 1273,  1968.

Union Carbide Corporation, "Review of the  Environmental Effects of Arsenic",
Oak Ridge National Laboratory,  Oak Ridge, Tennessee, ORNL/E15-79, 1976.

United Nations, "Survey  of World Iron Ore Resources", United Nations
Department of Economic and Social Affairs, New York, 1970.

Vinogradov, A. P., "Regularity of Distribution of Chemical Elements in the
Earth's Crust", Geochemistry (Geokhimiya),  1,  1956.
                                    56

-------
                                   TECHNICAL REPORT DATA
                            (Please read Instructions on the reverse before completing)
 REPORT NO.
  EPA-560/6-005
2.
                               3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
                                                             5. REPORT DATE
                                                                March   1979
 ARSENIC:  A Preliminary Materials Balance
                                                             6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
 William A. Lowenbach and Joyce S.  Schlesinger
                               8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
 Lowenbach and  Schlesinger Associates, Inc.
  1842 Kirby Road
 McLean,  Virginia  22101
                                                             10. PROGRAM ELEMENT NO.
                               11. CONTRACT/GRANT NO.
                               Purchase Order No.
                               W-1434-NNSX
12. SPONSORING AGENCY NAME AND ADDRESS
                                                             13. TYPE OF REPORT AND PERIOD COVERED
  Environmental Protection Agency
  Office of Toxic Substances
  401 M Street, S.W., Washington, D.C.  20460
                               14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
          Existing data on the natural and anthropogenic sources of arsenic
  emissions are compiled and  presented in a fully annotated tabular format.
  Arsenic distribution in the  environment  is graphically displayed in terms of '
  air, land, and water emissions.. Fossil  fuel consumption for energy
  production, arsenic production and commercial use,  and primary copper
  smelting are identified as the. three major sources of environmental
  emissions.   Data gaps are identified and recommendations  for further study
  outlined.
17.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
  Arsenic, sources of emissions
                                               b.lDENTIFIERS/OPEN ENDED TERMS
                                               COS AT I Field/Group
18. DISTRIBUTION STATEMENT
 unlimited
                                               19. SECURITY CLASS (ThisReport)
                                                 unclassified	
                                             21. NO. OF PAGES
                                                    65
                                               20. SECURITY CLASS (This page)
                                                                           22. PRICE
EPA Form 2220-1 (Rev. 4-77)    PREVIOUS EDITION is OBSOLETE
                                          57
                            ft U.S. GOVERNMENT PRINTING OFFICE. 1979 -281-147/36

-------