EPA-450/5-79-006
Sources of Atmospheric Cadmium
by
Robert Coleman, et al.
Energy and Environmental Analysis, Inc
1111 North 19th Street
Arlington, Va. 22209
Contract No.68-02-2836
EPA Project Officer: Richard Johnson
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Off ice of Air, Noise, and Radiation
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
August 1979
-------
This report was furnished to the Environmental Protection Agency by
Energy and Environmental Analysis, Inc., Arlington, Virginia in partial
fulfillment of Contract No. 68-02-2836 Tasks 3 and 6. The contents of
the report are reproduced herein as received from the contractor. The
opinions, findings, and conclusions expressed are those of the authors
and not necessarily those of the Environmental Protection Agency.
ii
-------
ACKNOWLEDGEMENTS
•\ .
Preparation of this report by Energy and Environmental Analysis, Inc.,
was carried out under the overall direction of Mr. Robert Coleman. Special
assistance was received from Messrs. James Lent, Paul Siebert, Craig Miller,
and Ms. Elizabeth Coffey of EEA.
EEA gratefully acknowledges the assistance, helpful suggestions and
review of the EPA Task Officer, Mr. Richard Johnson.
The conclusions presented in the study are, of course, solely the
responsibility of Energy and Environmental Analysis, Inc.
111
-------
-------
. TABLE OF CONTENTS
Page
ACKNOWLEDGEMENTS . .....'
EXECUTIVE SUMMARY. .. 1
SECTION I: . INTRODUCTION. 9
SECTION II: CADMIUM IN THE ENVIRONMENT 12
2.1 Introduction ^2
2.2 Physical and Chemical Characteristics
of Cadmium , 12
. 2.3 Multi-Media Nature of Cadmium Exposures 12
SECTION III: METHODOLOGY 20
3.1 Introduction.... 20
3.2 Determination of Potential Cadmium Emission
Sources 20
3.3 Emission Factor Determination 21
3.4 Computation of Emission Levels 22
3.5 Source Screening 22
SECTION IV: USES OF CADMIUM 24
4.1 Introduction ^ < 24
4.2 Electroplating : 24
4.3 Paint Pigments 26
4.4 Plastic Stabilizers 26
4. 5 Nickel-Cadmium Batteries; 27
.4.6 Miscellaneous 27
-------
TABLE OF CONTENTS (Continued)
Page
SECTION V: . SOURCES OF ATMOSPHERIC CADMIUM EMISSIONS.. • 29
5.1 Introduction 99
5.2 Mining 30
5.3 Primary Metal Production 3^
5.4 Iron and Steel 47
5.5 Secondary Smelting 56
5.6 Manufacturing 62
5.7 Fossil Fuel Combustion 57
5.8 Miscellaneous 75
5.9 Incineration 79
5 .-10 Summary. , g3
SECTION VI: SCREENING OF CADMIUM SOURCE TYPES 95
6.1 Introduction * 95
6.2 Mining 99
6.3 Primary Metals ^ ^
6.4 Iron and Steel ...
6.5 Secondary Smelting
6.6 Manufacturing
6.7. Fossil Fuel Consumption
6.8 Miscellaneous ; _ .. n.
6.9 Incineration 106
Vl
-------
LIST.OF TABLES
TABLE NUMBER
E-l
E-2
E-2
E-3
2-1
2-2
2-3
.2-4
5-1
5-2
5-2
5-3
5-3
TITLE ' . PAGE
Airborne Cadmium Emissions -- 1974, 1985 4
Cadmium Emission Factors 5
Cadmium Emissions Factors (cont.) . 6
Source Categories Potentially Able
to Cause a Measurable Level of Cadmium 8
Physical Properties of Cadmium 15
Some Atmospheric Cadmium Data 16
Cadmium Content of Selected Adult Foods 17
Media Contributions to Normal Retention
of Cadmium 18
Airborne Cadmium Emissions — 1974, 1985 86
Cadmium Emission Factors 87
Cadmium Emission Factors (cont.) 88
Comparison of Cadmium Emission Estimates
1968 - 1977 89
Comparison of Cadmium Emission Estimates
1968 - 1977 (cont.) 90
VII
-------
LIST OF FIGURES
FIGURE NUMBER
E-l
4-1
5-1
5-2
5-3
TITLE
1975 Cadmium Consumption in the United States
1975 Cadmium Consumption in the United States
Primary Smelting Process of Zinc and Lead
Primary Smelting Process of Copper
Primary Smelting Process of Cadmium
PAGE
2
25
33
41
46
viii
-------
EXECUTIVE SUMMARY
This report is one of a series of reports which will be used by EPA in
responding to the Congressional request under section 122 of the Clean
Air Act Amendments of 1977 to determine whether atmospheric emissions of
cadmium pose any threat to public health. This report surveys the uses
of cadmium and potential emission sources to determine which sources are
the most significant both in terms of total emissions and potential
ambient'levels.
The basic methodology used in this report is as follows:
• Determination of potential cadmium emission sources
through a literature review;
• Determination of emission factors (Ibs. cadmium
emitted/unit output) for each source;
• Application of these emission factors to current
and projected production levels to obtain estimates
of total cadmium emissions;
• Evaluation of control technology for reducing
cadmium emissions;
• Screening of all cadmium sources to identify.
those sources potentially able to cause
measurable ambient levels of cadmium.
Consumption of cadmium in the U.S. has averaged about 5,000 megagrams/
year over the last several years. United States production is closely
tied to zinc production and has not been adequate to meet demand. In
1975 almost 40 percent of the cadmium used in the U.S. was imported.
The principal uses of cadmium and their relative share of consumption is
shown in Figure E-l.
-------
FIGURE E-l-
1975 CADMIUM,CONSUMPTION IN THE UNITED STATES(l)
PLASTICS
. STABILIZATION
20%
ELECTROPLATING
55%
-------
It is estimated that about 850 tons of cadmium were emitted during 1974.
The breakdown of.emissions by source category is shown on Table E-l.
This table also lists the anticipated change in cadmium emissions due to
growth changes in technology or the anticipated imposition of higher .
efficiency control equipment.
As Table E-l shows, the largest estimated emitter of cadmium is the pro-
duction of zinc. This source accounts for approximately 53 percent of
total cadmium emissions. Although zinc production is expected to increase
significantly, emissions are not expected to increase and very possibly
will decrease due to the increased use of electrolytic processes which
have almost no emissions.
Emissions from other major sources (incinerators, iron and steel and
fossil fuel combustion) are expected to remain relatively constant as
the increased production is balanced by the increasing levels of control
technology.
Data on Table E-2 is based on the use of "best judgement" emission
factors developed through a review of the literature and various stack
test results. Table E-2 shows the emission factors developed in. this
study. Also shown on Table E-2 is the minimum and maximum emission
factor reported for any of the sources. It is clear that significant
differences can exist among tests on different sources. As such, al-
though the emission factors are probably adequate for the purpose of
evaluating differences among source categories, care must be taken in
the application of these factors to any particular source.
After emission factors and emission levels were determined for par-
ticular source categories, sources were evaluated to determine if an
individual source had the potential to cause a measurable level of
cadmium (assumed to be 0.1 ng/m3 on an annual average). Screening was
-------
'•'•.'. TABLE E-l • ,
AIRBORNE 'CADMIUM EMISSIONS--1974, 1985
Source
MINING
Zinc
Copper
Lead
PRIMARY METALS
Zinc
Pyrometallurgic
Electrolytic
Lead
Copper
Cadmium
SECONDARY METAL PROCESSING
Iron and Steel
Sinter Hindbox Uncontrolled
Sinter Hindbox w/Rotoclone
and ESP
Basic Oxygen Furnace
Uncontrolled
EOF w/Venturi or ESP
Open Hearth Uncontrolled
Open Hearth w/ESP
Electric Arc Controlled
Blast: Furnace Controlled
Zinc
Lead
Copper
MANUFACTURING
Pigments
Stabilizers
Batteries
FOSSIL FUEL COMBUSTION
Coal-Fired Power Plants
Oil-Fired Power Plants
s
Heating Oil
Diesel Oil
Gasoline
MISCELLANEOUS
Motor Oil
Rubber Tire Hear
Fungicides
fertilizers
Ceaent
INCINERATION
Sewage Sludge Incinerators
Municipal Incinerators
Production
1974*
478,850
1,414,246.8
603,024
423,000
121,945
866,095
1,435,662.4
• 3,088.2
21.94X106
11.35xl06
1.2X106
78.8X106
7.64X106
29.06X106
27.3X106
95.2x10
75,409
698,698
513,308
1,212.1
991.8
628.14
3.913X108
SOOxlO6
(barrels)
935.1xl06
(barrels)
llxlO9
(gallons)
1,330, 074xl06
(VMT)
I,028,121xl06
(VMT)
1,330, 024xl06
(VMT)
59,800
8,535xl03
81,210xl03
1,460,000
20,143,620
Emissions
Estimate
<1
<1
<1
529
0
2
5
2
22
5
<1
.
-------
TABLE E-2
CADMIUM EMISSION FACTORS
SOURCE
PRIMARY METALS PROCESSING
ZINC
Coking
Sintering w/Cyclone
Sintering w/Cyclone & ESP
Roasting
Horizontal Retort
Vertical Retort
Electro thermic
Overall (Not Electrolytic)
Electrolytic .
LEAD
Overall Smelter
Blast Furnace w/Baghouse
COPPER
Uncontrolled Smelter
Smelter w/Baghouse (~95%)
CADMIUM
SECONDARY METALS PROCESSING
IRON 4 STEEL
Sinter Windbox-Uncontrolled
Sinter Windbox
w/Rotoclone & ESP
Blast Furnace-Controlled
Open Hearth-Uncontrolled
Open Hearth w/ESP
Basic Oxygen Furnace
Uncontrolled
w/Venturi or ESP
Electric Arc Furnace
SECONDARY ZINC-UNCONTROLLED
SECONDARY LEAD
Blast Furnace w/3 Cyclones
& Baghouse
Reverberatory Furnace w/
' Cyclone & Baghouse
Reyerberatory Furnace w/
: 3 Cyclones & Baghouse
SECONDARY COPPER-UNCONTROLLED
MZN£NG 't3?*
-------
TABLE E-2 (continued)
CADMIUM EMISSION FACTORS
SOURCE
FOSSIL FUEL COMBUSTION
Coal -Fired Power Plants
.Uncontrolled
Controlled (ESP)
Oil-Fired Power Plants
Controlled ( ~ ESP)
Keating Oil (Residual; *6
Fual Oil)
Diesel Oil
Gasoline (for 15 mpg. all
Cd Emitted)
SEWAGE SLUDGE INCINERATOR^
Multiple Hearth w/Scrubber
Fluidized Bed w/Scrubber
HUMICIPAI INCINERATORS
Uncontrolled
Controlled (Scrubbers or ESP)
IU9R1CAT1HS OIL INCINERATORS
Uncontrolled
HISCEtLAtlEOUS
Motor Oil Consumption
(Vehicles)
Rubber Tire Hear
Fungicides Application
Fertilizers Application
Superphosphate Fertilizers
Application
CE-EHT PLANTS
DRY PROCESS
Kiln w/Baghouse or ESP
Raw Hill Feed w/Baghouse
* Raw Hi 11 w/Bagnouse
Raw Hill Air Separator w/
Baghouse
Finish Hill Feed w/
Baghouse
Finish Hill w/Baghouse
Finish Hill Air Separator
w/Baghouse
KET PROCESS
Kiln w/ESP
Raw Hill w/Baghouse
Clinker Cooler w/ESP
or Baghouse
UK£ KILH (PULVERIZED COAL)
Mln w/Spray, Settle &
Baghouse
MINIMUM
lxlO'41b/TCoal '(STK.AA)
lxlO"61b/TCoal (STK.AA)
7.1xlO'71b/gal (STK.ES)
1.5xlO'61b/gal (EST.CONC)
6xlO"71b/gal (EST.CONC.ES)
6.3xlO~111b/veh-mi (EST.
CONC)
-filh/
1x10 OID/TS1udge (DRYHSTK,
ES)
4xlO'71b/TSludge (DRY)
(STK.ES)
3xlO"31b/TRefuse (EST)
6xlO'41b/TRefuse (FLAA)
Ixl0"l0lb/veh-m1 (EST.CONC)
i.8xlO"61b/gal (EST.MB)
1.7xHT41b/T (EST.HB)
3xlO"71b/TFeed (STK.ES)
lxlO"71b/TFeed (STK.ES)'
7.6XlO'71b/TFeed (STK.ES)
5xlO"71b/'TFeed (STK.ES)
7.4xlO"61b/TFeed (STKtES)
1.7xlO-61b/TFeed (STK.ES)
4.6xlO"51b/TFeed (STK.ES)
MAXIMUM
lx!0'11b/TCoal (STK.ES)
7xlO"41b/TCoal (STK.AA)
4.4xlO"61b/gal (STK.CONC.ES)
4xlO"51b/gal (EST.CONC.NA)
2xlO'61b/gal (EST)
4.5xlO"81b/veh-mi (EST.CONC)
2xlO"51b/TSludge (DRY) (STK.ES)
3xlO-^b/TSludga (ORY)(STK,ES)
- •
1.8xlO"21b/TRefuse (STK.ES)
1.0xlO'llb/TRefuse (EST.MB)
5xlO"81b/veh-mi .(EST.CONC)
5xlO'Slb/gal (EST)
5xlO"21b/T (EST.HB)
4.1xlO'71b/TFeed (STK.ES)
4.3xlO"71b/TFeed (STK.ES)
9xlO"71b/TFeed (STK.ES)
1.6xlO"61b/TFeed (STK.ES)
1.3xlO'71b/TFeed (STK.ES)
2xlO'41b/TFeed (STK.ES)
lxlo-41b/TFeed (STK>ES)
6.9xlO"51b/TFeed (STK.ES)
BEST JUDGEMENT
lxio"31bACbal
6xlO"51b/TCoal
9xlO'71b/gal (STK.ES)
3xlO-61b/gal
7xlO"71b/gal (EST.CONC.ES)
2xlO~81b/veh-nri
7xlO'61b/TSludge (DRY) (STK.
REFERENCES
=
14,22,23
7,8,10,15,21,31
• .14
2,7,10,21,18,26,'33
7,10,21,18
25,26,27,28,29
7
ES)
1.3xlO"61b/TSludge (DRY)(STK. 7,14'
ES)
I_*J J
6xlO'31b/TRefuse (STK.ES)
1.3xlO"21b/TRefuse (STK.AA)
2xlO"61b/gal (UNK)
2xlO"91b/veh-mi (UNK)
'8xlo"91b/veh-mi
lxlO-51b/gal
6xlO-31b/T
2xlo-41b/T
3xlO"41b/TFeed (STK.ES)
3.6xlO"71b/TFeed
2.7xlO"71b/TFeed
8.5xlO'71b/TFeed
lxlQ-61b/TFeed
lxlO-71b/TFeed
2.6xlO"61b/TFeed (STK.ES)
2xlo-51b/TFeed
2xlO"51b/TFeed
lxlO'51b/TFeed
5.7xHT51b/TFeed
7,10,14
5,7,8,10,14,33
7
7,8 '
-•'• 2,6,7,10
7 8
6,7,8,17,21,31
ii.7,31
7,14
7 14
7,14
7,14
14
14
7 14
M'
7 14
14
r e: ™ " "*" Ba1anCei Sm " SUe V1SUS: SURV " Survey °f ComP«n^si UNK - Unknown (in literature); STK . Stack Sampling Results!
CON- • Concentration of Cd in feed, fuel, or emissions (w/STK); ES . Emission Spectroscopy; AA - Atomic Absorption (FL-Fla«); NA - Neutron Activation
neutron Activation.
-------
accomplished by modeling, a very large plant in each category under re-
strictive assumptions of stack height, flow rate, temperature and.
meteorology. The evaluation procedure was not designed to determine
what expected levels, of cadmium might be, but to determine if any
possibility existed"that the source could cause a measurable level. A
more detailed modeling effort was undertaken during the cadmium exposure
analysis. • •
Table E-3 lists the source categories which were determined to be
potentially able to cause a measurable level of cadmium. These sources
were further evaluated in the second phase of this study to determine the
population potentially exposed. . ..' _ ...„. „_......,.
-------
TABLE E-3
SOURCE CATEGORIES POTENTIALLY ABLE TO
CAUSE A MEASURABLE LEVEL OF CADMIUM
Primary Zinc Smelters
Primary Copper Smelters
Primary Lead Smelters
Primary Cadmium Smelters
Secondary Zinc Smelters
Secondary Copper Smelters
Municipal Incinerators
Iron and Steel Mills
a/ _
Assumed to be 0.1 ng/m on an annual average.
-------
1. INTRODUCTION
This report is one in a series of reports which will be used to help EPA
respond to the Congressional mandate under Section 122 . of the Clean Air
Act Amendments of 1977. Under this section, EPA is required, to review
the health and welfare effects of cadmium (as well as other substances)
and determine "whether or not emissions of. . .cadmium. . .into the ambient
air will cause, or contribute to, air pollution which may reasonably be
anticipated to endanger public health."
This report focuses on three of the areas of information required to
make this determination-identification of the uses and sources of
cadmium, development of emission factors for cadmium from industrial
sources, and the screening of cadmium sources to determine which ones
are potentially "significant" sources of ambient cadmium. In this
report, a "significant" source is defined as one which, by itself, can
cause a measurable ambient level of cadmium.
A companion study1/ uses the results from this study to provide an
estimate of the population exposed to measurable levels of cadmium.
Neither this report, nor the companion report on population exposure,
draws any conclusions as to the health consequences of ambient cadimum
levels. Rather, the purpose of the two reports is to provide a relative
ranking of sources, both by the magnitude. of emissions and the popula-
tion exposed, and to provide information in such a way as to allow EPA
to make reasonable estimates of any health 'implications of the reported
emissions and exposures.
The report is organized into five substantive
sections summarized below:
• Section II provides an overview of the physical and chemical
-------
« Section III provides an overview of the metholodogy used in
preparing this report.
* Section IV discusses the current and expected uses of cadmium.
» Section V discusses the potential emission sources of cadmium.
For each source, a brief description of the process, cadmium
emission points, and the emission factors are described. Total
current and expected cadmium emissions and available control
technology are also discussed.
e Section VI discusses the screening of the various cadmium
sources which was used to determine which sources can cause
a measurable level of cadmium (0.1 ng/m3 on an annual average).
10
-------
REFERENCES
1. Cadmium: Population Exposure Analysis, Energy and Environmental
Analysis, Final Report, Contract No. 68-02-2836, Task 14, March 2,
1979.
11
-------
2. CADMIUM IN THE ENVIRONMENT
2.1 INTRODUCTION
This section discusses the physical and chemical properties of cadmium
and the multimedia nature of cadmium exposures. Although this report
focuses only on atmospheric emissions of cadmium, it is important to
keep in mind that there are many other types of human exposure to cadmium.
2.2 PHYSICAL AND CHEMICAL CHARACTERISTICS OF CADMIUM
Cadmium is a relatively rare element in the earth's crust. It occurs at
concentrations of 0.1 to 0.5 ppm, ranking in abundance between mercury
and silver, and thus, not in sufficient quantities to be mined as an
imii
2/
ore. Table 2-1 shows the physical properties of cadmium. Cadmium is
always associated with zinc and is usually present .as a sulfide.'
The most important characteristic of cadmium, from an air pollution
viewpoint, is its high volatility. This is evidenced by its low melting
(321 C) and boiling C767°C) points. Because of this, high temperature
processes, such as metallurgical processes (steelmaking, sintering) or
incineration, are likely to release whatever cadmium is present in their
feeds. Vaporized cadmium metal reacts very quickly to form an oxide,
sulfate, or other compound of relatively high stability. Cadmium metal
is also ductile, easily soldered, and can be readily electroplated to
maintain a lustrous finish in air. These properties lead to the use
of cadmium as a protective coating on iron and steel products.
2.3 MULTIMEDIA NATURE OF CADMIUM EXPOSURES
While this report focuses on the atmospheric emissions of cadmium, it is
important to recognize the overall cycle of cadmium in the environment.
12
-------
Measurable levels of cadmium occur in all phases of environmental
concern (air, water, food, solid waste), and in almost all areas. One
author refers to cadmium as the "dissipated element." EPA in 1975
estimated that about 1,800 Mg/year of cadmium were lost to the environ-
*•*"
ment. Of this, about 18 percent was in atmospheric emissions, 75 per-
cent in solid waste, and the remainder in water-borne emissions.
Measurable cadmium levels have been found in air, water, soil and food.
Atmospheric concentrations generally have been measured in the center of
3
urban areas and usually range from-100-ag/s down to below the detect-
able limit. Typical urban concentrations would be approximately 3 ng/m ;
some measured values appear in Table 2-2. Levels of cadmium in water
supplies,are generally low. Main sources of cadmium are discharges from
mining operations, leaching from soil disposal of wastes, and rain-out
from atmospheric emissions.
Cadmium in food results from a wide variety of sources. Listed in order
4/
of importance from a recent Battelle Report, they are:
Direct contact by plants or uptake from soils by plant roots.
Cadmium may occur in soil:
- Naturally as a normal constituent of all soils,
but particularly those of marine origin.
- As an impurity (cadmium oxide) in phosphate-
treated soils, especially in those treated
with "superphosphate."
- By fertilization with sludge containing
cadmium.
- , -By deposition of cadmium-containing pesticides
or a contaminant of zinc-containing pesticides.
From run-off of mine tailings or from electro-
plating washing process.
13
-------
• Accumulation in animal tissues due to:
Feeding on crops which have absorbed cadmium.
(The organs of such animals may have very high
cadmium concentrations.)
Treatment with cadmium-containing helminth
killers, used especially in swine.
Concentrations of cadmium by mollusks and crustaceans
and most other aquatic organisms from ambient waters.
• Use of zinc-galvanized containers, cans, cooking implements
or vessels; or utensils used in food preparation, particu-
larly grinders, pressing machines, or galvanized netting
used to dry fish and gelatin.
• Absorption of cadmium contained in wrapping and packaging
materials such as paper, plastic bags, and tin cans.
(Cadmium is now prohibited in such" food wrappings.)
« Use of cadmium-contained water in cooking or processing
operations.
Table 2-3 lists the average cadmium concentration of selected adult
foods.
Cigarette smoking also provides a large contribution to total cadmium
exposure. The estimated intake from two packs per day ranges from four
to six micrograms which amounts to about 20 times the exposure due to
atmospheric levels in large urban areas.
Even for smokers, food provides the greatest overall exposure to cadmium,
and based on a 6.4 percent retention rate, the greatest daily input (ex-
cept for three packs-per-day-smokers). Table 2-4 summarizes the sources
of cadmium exposure.
14
-------
TABLE 2-1
PHYSICAL PROPERTIES OF CADMIUM1
a,b/
Atomic Number
Atomic Weight
Color
Crystal Structure
Hardness
Ductility
Density
20°C (68°F) .(solid)
330°C (626°F) (liquid)
Melting point
Boiling point
Specific heat
25°C (77°F) (solid)
Electrochemical equivalent
Cd ion
Electrode potential
Cd ion
48 .
112.41
silver-white
hexagonal pyramids
2.0 Mohs
Considerable
8.5 g/cc
8.01 g/cc
321°C (609.8°F)
767°C (1412.6°F)
0.055 g-cal/g
0.582 mg/coulomb
0.40 volta/
a/
b/
From Reference 1.
National Bureau of Standards nomenclature, H,
15
-------
TABLE 2-2
SOME ATMOSPHERIC CADMIUM DATA'
a/
SITE
URBAN:
Liege, Belgium
Paris, France
Sevastopol, USSR
Petropavlovsk, USSR
Magadan, USSR .
Salehard, USSR
Dourbes, Belgium
Heidelberg
Rotterdam
Sutton, England
San Francisco
Cincinnati
Denver
St. Louis
Washington, D.C.
Chicago
Philadelphia
Tucson
Oak Ridge, TN
New York City
New York Bight
U.S. urban average (1966)
REMOTE:
Northern Norway
Jungfraujoch, Switzerland
Novaya Zemlya
Dickson Island, USSR
Indian Ocean, North
Indian Ocean, South
Bermuda
North Atlantic
Windward Hawaii
South Pole
Spitsbergen
Barrow, Alaska
CONCENTRATION
3
(ng/m )
118
20
2.3
0.6
4.2
5.3
1.3
27
7.7
0.4
20
2.0
1.0
5.0
0.3
3.0
1.0
3.4
4.1
5.7
1.0
2.0
0.5 (lowest 0.006)
0.4
0.3
0.4
2.0
0.14
0.4
4.2
0.02
<0.02
0.1
0.04
a/
Source: Reference 4
16
-------
TABLE 2-3
CADMIUM CONTENT OF SELECTED ADULT FOODSa/
Commodity
Carrots, roots fresh
Lettuce, raw crisp head
Potatoes, raw white
Butter
Margarine
Eggs, whole fresh
Chicken fryer, raw
whole or whole cut up
Bacon, cured raw, sliced
Frankfurters
Liver, raw beef
Hamburger, raw ground beef
Roast, chuck beef
Wheat flour, white
Sugar refined, beet or cane
Bread, white
Orange juice, canned frozen concentrate
Green beans, canned
Beans, canned with pork and tomato sauce
Peas, canned
Tomatoes, canned
Diluted fruit drinks, canned
Peaches, canned
Pineapple, canned
Applesauce, canned
No. of
Samples
69
69
71
71
71
71
71
71
69
71
71
71
71
71
70
71
71
71
71
71
71
71
71
71
Average
ppm
0.051
0.062
0.057
0.032
0.027
0.067
0.039
0.040
0.042
0.183
0.075
0.035
0.064
0.100
0.036
0.029
0.018
0.009
0.042
0.042
0.017
0.036
0.059
0.020
Standard
Deviation,
ppm
0.077
0.124
0.139
0.071
0.048
0.072
0.088
0.160
0.111
0.228
0.122
0.034
0.150
0.709
0.063
0.095
0.072
0.000
0.113
0.113
0.052
0.061
0.153 .
0.027
a/
Source: Reference 4
17
-------
TABLE 2-4
MEDIA CONTRIBUTIONS TO NORMAL RETENTION
' OF CADMIUMa/
Medium
Ambient air
Water
Cigarettes:
Packs/Day
1/2
1
2
3
Food
Exposure Level
0.03 yg/in
1 ppb
yg/day /
1.1
2.2
4.4
6.6
50 yg/day
Daily Retention
(Vg)
0.15
0.09
0.70
1.41
2.82
4.22
3.0
c/
c/
c/
c/
a/
b/
c/
Source: Reference 4.
Based on 0.11 yg per cigarette.
Assumes a 6.4 percent retention rate.
18
-------
REFERENCES
Fulkerson, W. , and Goeller, H.D., Eds., Cadmium, The Dissipated
Element, ORNL National Science Foundation, ED-21, January 1973.
Deane, G. L., Lynn, D.A., and Surprenant, N.F., Cadmium: Control
Strategy Analysis, GCA-TR-75-36-G, Final Report for Environmental
Protection Agency, Contract No. 68-02-1337, Task No. 2, April 1976.
Sargent, D.J. and Metz, R.J., Technical and Microeconomic Analysis
of Cadmium and Its Compounds, Environmental Protection Agency, 560/
3-75-005, June 1975.
Multimedia Levels Cadmium, Environmental Protection Agency 560/6-
77-032, September 1977.
Scientific and Technical Assessment Report on Cadmium, Envirpnment_al
Protection Agency 600/6-75-003, March 1975.
19
-------
3. METHODOLOGY
3.1 INTRODUCTION
This section describes the general methodology used in evaluating sources
of cadmium emissions and in determining their magnitude and significance.
In simplest terms, the methodology can be viewed as having five com-
ponents :
• Determination of potential cadmium emission sources;
• Determination of emission factors for these sources;
• Estimation of current and future emissions of cadmium;
• Evaluation of control technology for reducing cadmium
emissions from these sources; and
• Screening of all potential cadmium sources to identify
the sources most likely to cause measurable ambient levels.
3.2 DETERMINATION OF POTENTIAL CADMIUM EMISSION SOURCES
A literature search and a review of previous EPA studies were carried
out to determine the sources most likely to emit cadmium. The basic
procedure followed in this study, as well as in previous studies, was to
concentrate on the production and uses of cadmium, and then to follow
cadmium through to its ultimate disposal.
Once the potential sources were identified, trade literature and other
references were used to develop process descriptions, identify potential
emission sources, and estimate total current and projected production.
Special emphasis was placed on determining changes in production patterns
which could lead to changes in emission characteristics. The sub-
sections of Section V, which discuss each industry, identify the data
used to develop the above information.
20
-------
3.3 EMISSION FACTOR DETERMINATION
For each potential cadmium source identified, the literature was surveyed
to determine the amount of cadmium emitted per product unit produced.
For most sources, several types of data were available and a ranking
system was established to determine the final emission factor. The data
were ranked in the following order:
• Actual Stack Tests—Stack tests conducted on several sources,
or even one source, and analyzed by a quantitative analytic
technique such as atomic absorption (AA). Stack tests con-
ducted using a semi-quantitative technique such as emission
spectroscopy (ES) were given a somewhat lower ranking. The
primary source of stack tests using ES was EPA tests in
support of particulate new source performance standards.
• Concentration of Cadmium in Feed—For fuel burning sources,
data on incoming cadmium levels can be used to determine
cadmium emissions relatively accurately if 100 percent loss
of cadmium is assumed. This assumption is reasonable because
of the high volatility of cadmium.
• Mass Balances—For several sources, the only data available came
from an estimate of the losses of cadmium during processing.
Data comes from site visits carried out by EPA during
earlier emission factor development and engineering judgement
on process operations.
For each source type, several emission factors were developed:
• Minimum Emission--Lowest emission factor reported by any
study.
• Maximum Emission--The highest emission factor reported by
any study.
• Best Judgement—The best estimate of emissions when the
quality of all the data is considered and the ranking
system described above is used. Often, a mean was used
for a series of stack tests of equal quality. Where
emission factor estimates range from mass balances to
tests using atomic absorption, the best judgement was
weighted subjectively toward the more accurate AA esti-
mates .
21
-------
As with any substance, cadmium emissions can vary greatly from source to
source, and the accuracy of any specific emission factor for a specific
plant is questionable.
The emission factors developed here are probably sufficiently accurate
for the purposes intended, (relative evaluation among source categories),
but care should be taken in applying the factors to any specific plant.
3.4 COMPUTATION OF EMISSION LEVELS
The product-ion levels for each source category and the emission factors
previously developed were combined to provide an estimate of total
current cadmium emissions. The emission factors were modified (where
necessary) to reflect the application of typical control technology.
For purposes of comparison, where EEA's control technology assumptions
differed from those in previous estimates, an additional estimate for
cadmium emissions was made using comparable control technology assumptions,
3.5 SOURCE SCREENING
To determine which source types are capable of causing a measurable
level of cadmium, (assumed to be an annual average of 0.1 ng/m3) a
screening procedure was developed. While the detailed procedure for
screening varied from source type to source type, (each is discussed in
Section VI) the general procedure is as follows:
• For each source type, an expected largest or very
large source was determined from the literature.
• A very conservative combination of stack conditions was
then determined. These conditions (stack height, temperature,
flow, etc.) were based on engineering judgement, coupled with
limited data available in the literature. Every assumption
was made such that ground level concentrations would be
maximized.
• Emissions were based on maximum emission factors and plants
were assumed to produce their rated capacity in only eight
hours of operation.
22
-------
• Ambient concentrations (for point sources) were determined
. using the EPA "PTMAX" model. If these concentrations exceeded
the detectable limit, (0.1 ng/m ) further analysis was required.
This criterion is extremely conservative because the PTMAX esti-
mates one-hour concentrations, which are typically at least ten
times higher than annual averages.
23
-------
4. USES OF CADMIUM
4.1 INTRODUCTION
Consumption of cadmium in the United States has averaged approximately
5,000 metric tons a year over the past several years. In 1975, almost
40 percent of the cadmium used in the U.S. was imported.
There are four major uses of cadmium, all of which are dissipative
(Figure 4-1). In 1974, approximately 55 percent (2,700 metric tons) of
the cadmium consumed in the U.S. was used by the electroplating industry.
Paint pigment manufacturing required close to 1,000 metric tons of
cadmium in 1974, and 900 metric tons were used to produce plastic stabi-
lizers. The fourth major use of cadmium in 1974 was the nickel-
cadmium battery, for which 550 metric tons of the metal were used.
I/
Cadmium is also used in nuclear reactor controls, fluorescent phosphors,
and in alloys.
4.2 ELECTROPLATING
Electroplating a metal with cadmium inhibits corrosion and enhances
solderability. Products made of iron and steel are most frequently
coated with cadmium, including motor vehicle parts, industrial machinery
parts, aircraft parts, marine equipment, and hardware.
There are several reasons why cadmium is preferred as a coating material.
Only a thin coating is necessary to provide adequate protection from
corrosive elements, especially salt water and alkalies. It is possible
to obtain a uniform deposition on objects of intricate design, and
luster is maintained for a long period of time. Although several
substitutes exist, cadmium is believed to be superior.
24
-------
' -FIGURE-4-1 • •
1975 CADMIUM CONSUMPTION' IN THE UNITED STATES(1)
PLASTICS
STABILIZATION
20%
ELECTROPLATING
55%
25
-------
4.3 PAINT PIGMENTS
Paint pigment production accounted for approximately 12 percent of the
1975 U.S. consumption of cadmium. The cadmium-sulfide compounds are
used to obtain colors ranging from yellow to orange, while the cadmium-
sulfoselenide compounds produce colors ranging from orange to light red
and deep maroon.
It has been estimated that approximately 75 percent of the pigments are
used to color plastic. Other uses for cadmium pigments include water-
based paints, rubber, artists' colors, printing inks, glass, textiles,
enamels, and ceramic glazes.
Cadmium pigments have certain properties which render them difficult to
replace. Because they are totally non-bleeding and alkali-resistant,
they are particularly suitable for plastic automobile interior parts.
The pigments provide very bright colors and a high degree of opacity.
Their high temperature properties contribute to the unique character of
the pigment.
4.4 PLASTIC STABILIZERS
The third major consumer of cadmium is the plastic stabilizer industry.
Heat stabilizers containing cadmium have been found to halt or slow
discoloration caused by the breakdown of polyvinyl chloride resin during
molding. The stabilizers are found in almost every plastic material
with the exception of food packaging, where use is prohibited by FDA
regulations. '
Substitutes for the cadmium plastic stabilizers may be forthcoming in
light of FDA regulations. Stabilizers of calcium-zinc composition have
been found to be competitive in both performance and cost and are expected
to replace the cadmium stabilizer in some items in the future.
26
-------
4.5 NICKEL-CADMIUM BATTERIES
The nickel-cadmium battery is the fourth major product composed of
cadmium. Demand for cadmium in this segment of the industry tripled
between 1968 and 1973. Batteries range in size from small button cells
(614 mm in diameter) to large rectangular cells (113 mm high by 91 mm
long by 38 mm wide).
The nickel-cadmium battery is used in a variety of systems and products
including alarm systems, pacemakers, and portable appliances and tools.
Calculators presently are the largest market for the battery. The
nickel-cadmium battery is preferable to others if long life is important.
4.6. MISCELLANEOUS
Cadmium is also used to produce alloys, (primarily low'temperature
solders), silver bronzes, and a copper alloy used in automobile radiators.
Nuclear reactor rods are often made of silver-cadmium. Cadmium phosphates
are used for television and fluorescent tubes; silver-cadmium oxide is
used in motor starting switches, relays, and circuit breakers.
27
-------
REFERENCES
1.
Determination and Evaluation of Environmental Levels of Cadmium
EPA 68-01-1983, Battelle-Columbus Laboratories, Columbus Ohio
1977.
2.
Sargent, D.J. and Metz, J.R., Technical and Microeconomic
of Cadmium and Its Compounds. Environmental Protection Agency
560/3-75-005, June 1975. '
28
-------
5. SOURCES OF ATMOSPHERIC CADMIUM EMISSIONS
5.1 INTRODUCTION -
Cadmium, a relatively rare element in the earth's crust, does not occur
alone naturally. It is a metal found only in conjunction with other
metallic ores, principally zinc, lead, and copper. When released into
the environment, cadmium can be absorbed, ingested or inhaled by bio-
logical systems, causing subsequent damage to these systems. This study
focuses upon one form of environmental release—emission to the atmos-
phere . ._-.... .
Airborne cadmium emissions result from the production of cadmium, its
use (primarily dissipative), and the use of other substances which are
contaminated with cadmium. Cadmium is produced commercially as a by-
product of the primary production of zinc and is found in the ores of
lead and copper also. As a result, the mining and primary smelting of
these three metal ores produce approximately 63 percent of the cadmium
emitted to the air. Cadmium is a major constituent in the production of
paint pigments, metal alloys, plastic stabilizers, and nickel-cadmium
batteries. Metals are often electroplated with cadmium. Substances
which contain cadmium as a contaminant include phosphatic fertilizers,
sewage sludge, fossil fuels, cement, and fungicides. All of these
materials'contribute to the atmospheric emission of cadmium.
Each process which contributes to the amount of cadmium in the air is
described in the following sections, together with control devices used
and national production trends for the near future.
29
-------
5.2 MINING
The mining of zinc, lead, and copper is a source of airborne cadmium
emissions, though the quantities emitted are small-less than one ton a
year.
5.2.1 Process
The process of preparing each metal ore for shipment to a primary smelter
involves two steps, mining and beneficiation. First, the large chunks
of ore are removed from the ground, crushed, and ground to fine, sand-
sized particles. The concentration process which follows is most often
done using a combination of gravity and flotation mechanisms, or by
flotation alone; however, some plants use only gravity settling. At
this point, any cadmium in the ore is present in the zinc concentrate.
Ore beneficiation is usually conducted in close proximity to the mine,
particularly in the Western States.
5.2.2 Emissions Source and Control
The major emissions from this phase of the metal production are a result
of the dust which escapes during mining and crushing. Because benefici-
ation is a wet-process, air emissions from this phase are minimal.
Control methods are rarely needed or used during the ore-crushing proce-
dures due to the effective design of the equipment; however, in any raw
material handling, a hood and a cyclone, baghouse, or the electrostatic
precipitator (ESP) is used. ' Although no information is available as
to the extent of the problem, airborne particulates from mine tailing
piles may be a possible source of cadmium emissions.
5.2.3 Emission Estimate
Emission factors used in this estimate were 1 x 10""' pounds/ton of zinc
in the ore, ' ' 1 x 10~ pounds/ton of lead in the ore, j3' and 3.2 x
-5 2 3/
10 pounds/ton ' of copper in the ore. Production of ore at various
30
-------
mines in 1974 equaled 478,850 tons of zinc ore,12/ 603,024 tons of lead
ore, and 1,414,246.8 tons of copper ore.12/ From these emission
factors and production figures, the emissions estimate of less than one
ton of cadmium per year was developed. Estimates from other organiza-
tions are in agreement with EEA's, as similar emission factors were used
(GCA--1 ton2/; Mitre--! ton11/; Davis--! ton7/;
EPA—1 ton6/).
5.2.4 Future Trends
It is believed that the mining of the various metal ores will increase
significantly, to the 1985 levels shown in Table 5-1. Even with this
increase, cadmium emissions from mining and ore beneficiation are pre-
dicted to remain below one ton/year.
5.3 PRIMARY METAL PRODUCTION
Primary metal production involves conversion of an ore concentrate to a
relatively pure metal. Three primary metal production processes—zinc,
lead, and copper—are responsible for over 60 percent of all airborne
cadmium emissions.
5.3.1 Primary Zinc
5.3.1.2 Process
The primary smelting of zinc produces both zinc and cadmium metal and
thus produces airborne cadmium emissions. There are two basic methods
used to obtain the desired output—pyrometallurgical and electrolytic
extraction. The first step in each process is the roasting of the zinc
concentrate, during which zinc sulfide is converted to zinc oxide. The
sulfur which escapes from this process is often converted to sulfuric
acid, in a contact process acid plant.
In the pyrolytic reduction of the ore, the next step is sintering, which
renders the material easier to handle and use as feed. Smelting of the
31
-------
zinc oxide follows the sintering. Smelting is conducted in batches in
horizontal retorts or continuously in vertical retorts, electrothermic
furnaces, or blast furnaces. Addition of carbon to the zinc oxide at
this point indirectly aids reduction of the oxide to the zinc metal, as
it reacts more readily with oxygen than does the zinc. The metal is
then vaporized, transferred from smelter to condensers, converted to
molten zinc, and cast into slab zinc. Slab zinc is frequently purified
further by redistillation.
Electrolytic zinc recovery was developed chiefly for the processing of
low-grade or mixed ore, concentrates. As high grade zinc deposits are
depleted, this process will assume an increasingly important role in the
processing of zinc (and other metals). -•- - -
Beyond the step of roasting, there is no similarity between the electrolytic
and pyrometallurgical processes. In electrolytic recovery, the roasted
zinc ore is leached with sulfuric acid to produce soluble zinc sulfates.
Filtration to remove insoluble impurities (including cadmium) is followed
by treatment of the solution with zinc dust. This treatment is done
twice—first to remove copper impurities, and then to produce a,zinc-
cadmium residue. It is possible to continue further processing of these
residues to recover the copper and cadmium contained in them.
In the next step, zinc metal is produced by processing of the zinc
sulfate solution in electrolytic cells. The zinc is then electrodeposited
onto aluminum cathodes. At regular intervals, the cathodes are removed
from the electrolytic cells, the zinc is stripped from the aluminum, and
then melted to produce slab zinc. Slab zinc produced in this manner
does not usually require any additional refining, as it is of very high
purity. It is not possible to process scrap in this manner because it
contains too many impurities. (Figure 5-1 illustrates the primary zinc
smelting process.)
32
-------
!i
; 1!'
ui
K
UJ
a
VJ"
O
-------
5.3.1.3 Emission Sources
Thermal processes contribute most to the release of airborne cadmium
emissions because cadmium has very low melting and boiling points (321°C
and 767 C, respectively). Therefore, roasting, sintering, and reduction
in the pyrometallurgical process would warrant the closest attention, as
they probably would be responsible for most of the airborne cadmium
emissions.
The roasting procedure, part of both the pyrometallurgic and electro-
lytic processes, is not considered a source of large cadmium emissions.
Control technology used at roasting facilities is highly effective and
allows almost no emission of cadmium into the air.6'
Sintering is considered the major potential source of cadmium emissions,
and is done at all pyrometallurgical plants. Reduction of the sintered
ore is considered a source of airborne cadmium only if the externally-
fired horizontal retort is used, as this configuration is not conducive
to particulate control. The last horizontal retort in U.S. production
has been replaced, so that emissions from these facilities are no longer
a problem.
Electrolytic processing is considered to be relatively free of airborne
emissions. A minor potential source of airborne cadmium emissions does
exist in the filtration which follows leaching.
5.3.1.4 Control
High efficiency control devices are employed at all zinc smelters.
Roaster facilities at all smelters use fabric filters or electrostatic
precipitators (ESP). Sintering emissions are also controlled with
fabric filters or ESPs.1/
-------
5.3.1.5 Emissions Estimates
EEA1s estimates of the cadmium released into the air through primary
zinc production have been compared with estimates from other sources.
EEA's emission factors, with assumed control technologies, were derived
from several sources and expressed in pounds per ton of zinc produced.
Yost (1974), and Jacko and Nuendorf (1977) measured coking emissions of
2.24 pounds/ton of zinc throughput using stack sampling '5' and atomic
absorption (AA) analysis. Sintering with a cyclone produces emissions'
at a rate of 6.32 pounds/ton of zinc produced, while sintering with
controls of a cyclone and an ESP releases 2.16 pounds/ton of zinc through-
put. ' The vertical retort, at the particular plant where the stack
tests were conducted, emitted 6.5 x 10~ pounds/ton of zinc throughput.5'6''
It should be noted that these particular test results, though highly
accurate in both sampling and analysis, may not be typical. Vertical
retorts, which produce high emissions, are uncommon in the U.S.; in
addition, high zinc losses were a problem at the plant, due largely to
the high temperatures in the coking operation which was volatizing zinc
and thus, cadmium also. The Sargent (1975) study concluded that roasting
emits negligible amounts of cadmium into the air. Sargent's other
estimated emissions factors which EEA used include: 6 x 10~3 pounds/ton
9
of zinc produced in a horizontal retort; 1.2 x 10" .pounds/ton of zinc
produced by the electrothermic process; and negligible emissions from
electrolytic processes. Overall emissions rates varied between 1.43
pounds/ton of zinc and 2.96 pounds/ton of zinc.5'7'8' To calculate an
emissions estimate, a factor of 2.5 pounds/ton of zinc produced for
nonelectrolytic processes was estimated by weighting the atomic absorp-
tion stack sampling results more heavily than previous mass balances.
Production figures (1974) of 423,000 tons of zinc processed pyrometal-
lurgically and 121,945 tons of zinc processed electrolytically were
combined with the overall emissions factor to obtain EEA's emissions
estimate of 529 tons yearly.
35
-------
The preceding estimate compared favorably with others made previously
and was very close to the GCA estimate of 500 tons/year.2/ Mitre found
that 619 tons of cadmium were released in primary zinc smelting,11'' EPA
estimates 644 tons of cadmium,18' and Sargent estimates 112 tons.16/
The Sargent estimate is lower than others made because it assumes 95
percent or better cadmium collection efficiencies which are presently
attainable.
The production at primary zinc smelters increased in 1976 for the first
time since 1972. ' In the years between 1972 and 1976, production had
steadily decreased. This may account for the variety of emissions
estimates obtained. Both the Mitre and EPA estimates were made before
this decline occurred. GCA's estimate was made during a low point and
EEA's estimate, based on 1974 data, illustrates the effect of more
recent emission factors.
5.3.1.6 Future Trends
Future emissions are not expected to increase over the next eight years,
in spite of an anticipated substantial increase in capacity. Between
1967 and 1975, the zinc industry lost nearly 50 percent of its capacity
due to environmental problems, rising costs, and scarce capital.9/ In
1976, however, there was a small increase in zinc production (4.6
percent) over 1975 and projections for 1977 indicate a 9.6 percent
Q/
increase in production over 1976. '
The zinc market is expected to remain stable with no large decrease in
use anticipated. There is a possibility of increased use of zinc by the
steel industry, as the use of thinner gauge steel will create the need
for increased corrosion protection (zinc-coated steel).
Domestic smelter capacity may increase through 1985. ASARCO had planned
to open a 180,000 ton/year electrolytic plant in Kentucky during 1979,9/
36
-------
but was forced to abandon its plans due to a declining zinc market.
Given a favorable market, the ASARCO facility could still be built. If
present industry expansion plans continue, 1,000,000 tons of zinc would
9/
be produced in the United States in 1985.
Increased construction of relatively non-polluting electrolytic zinc-
processing plants has led EEA to conclude that cadmium emissions from
primary zinc smelting will remain constant at 529 tons/year. This is a
high estimate, as the nonelectrolytic plants now in operation will
probably be phased out and replaced by cleaner, electrolytic ones.
5.3.2 Primary Lead
Primary lead is a source of cadmium emissions, but does not contribute
greatly to the production of commercial cadmium metal. Lead is produced
in much the same way as zinc, therefore, only a brief discussion of the
process follows.
5.3.2.1 Process
Roasting is usually the first step which the ore concentrate undergoes
in the process of purification and metal production. However, this is
not always done with the lead concentrate. More often, the sintering
machine or electric furnace receives the lead sulfide ore concentrate
directly. During the sintering process, the concentrates are combined
with coal, lime and silica flux, then reduced to the lead metal. At a
blast furnace, the ore is initially reduced to molten metallic lead and
then removed as lead bullion at specific intervals.
Slag processing is more common in primary lead smelting than in zinc
smelting. The slags_ are placed in a fuming furnace, brought to the
proper temperature, and mixed with powdered coal which burns and volatilizes
most of the zinc remaining in the slag. Any residual lead is also
collected.
Figure 5-1 illustrates the processing of primary lead.
3-7
-------
5.3.2.2 Emission Sources
As in zinc smelting, thermal processes are the cause of most emissions
from lead smelting.
Roasting is not considered a pollution source, because it is a step
which is often deleted from the process. When roasting is done, good
control technology is employed. Sintering operations created most of
the airborne cadmium emissions in lead ore concentrate processing. With
the one exception, all plants use updraft sintering which is considered
superior to other types. Reduction blast furnaces are also a source of
cadmium emissions; however, all plants use extremely efficient particulate
collection apparatus on the sinters and blast furnaces.
5.3.2.3 Control Technology
In attempting to control the emission of cadmium to the atmosphere
several collection devices are employed in primary lead processing.
Emissions from sintering and blast furnaces are usually controlled with
a cyclone plus a baghouse or an ESP. A waste heat boiler plus baghouse
or ESP is used with the reverberatory furnace. In handling material
from any of the steps of the processes, a hood and a settling chamber,
cyclone, ESP or baghouse is used to control emissions.1/
5.3.2.4 Emissions Estimate
The emissions of cadmium from lead processing are estimated to equal two
tons/year. Using data from several sources,2'7'8'10'11/ and emissions
factor of 1.1 x 10" pounds/ton of lead produced was obtained, assuming
3.0 percent cadmium in the particulate emissions. With a 95 percent
control efficiency, the factor was used with a production figure of
886,095 tons (1974)9/ to arrive at the above figure. The estimate is
38
-------
ll/ 2/ 2/
lower than those previously made (Mitre—555; EPA 163; GCA 65 '),
as the previous estimates were based on mass balances of the amount of
cadmium in the feed material, assuming no control or, at most, a 50
percent control efficiency. Based on the assumption of no control used
in other estimates, EEA estimates emissions of 48 tons/year.
5.3.2.5 Future Trends.
At present, the primary lead industry is slowly recovering from the
depressed level of 1975 and is expected to recover to a production level
of 790 thousand tons by 1985.9/
The major uses of lead include storage batteries, gasoline anti-knock
additives, and pigments. The battery market is considered strong., A
substantial increase in demand 'is due to the desire on the part of the
consumer for a longer-lasting battery, which requires 20-30 percent more
lead for each unit. There is a possibility that batteries with longer
9/
lives may lead to a depressed market by 1985.
The demand for lead by the gasoline industry is expected to drop sharply.
This market, which usually accounts for 17-18 percent of the lead consumed,
faces a reduction potential of 70 percent in demand over the next few
9/
years.
Modifications of the catalytic converter and internal combustion engines
which would allow use of leaded gasoline are still under study. This
could bring about increased demand for lead, but the feasibility still
remains unclear. Lead in paint pigments accounts for approximately six
percent of total lead production, an increase over the last several
years.
At present, producers are hesitant to increase capacity because of
environmental.regulation uncertainty and, at this point production
39
-------
equals only 87 percent of capacity.9' Therefore, growth is expected in
a steady, slow manner. NSPS regulations are not likely to affect the
lead industry, as no new construction is being considered for the near
future. EEA projects that a total of two tons/year of cadmium will be
released into the air'as a result of primary lead processing by 1985.9//
5.3.3 Primary Copper
Copper production constitutes a source of cadmium emissions into the
atmosphere, but the concentration of the cadmium in the ore is too low
for economical recovery. Emissions from this source are generated in
much the same manner as those from lead and zinc.
5.3.3.1 Process
Ores which contain low percentages of copper are first subjected to a
grinding and flotation process to produce a high percentage copper
concentrate. The concentrate is then roasted to remove excess sulfur:,
and is placed in the reverberatory furnace. Here,, a slag is formed
through a combination of iron oxide and siliceous flux; a matte, of
copper, iron, and some sulfur is left. The converter then reduces the
matte to copper metal. To accomplish this, a stream of blowing air
first eliminates sulfur. Then, a second stream of air reduces the
copper sulfide to metal. At this point.,, the blister.copper is refined
further by fire-refining to reduce sulfur and oxygen, and the refined
metal is then cast. (Figure 5-2,)
5.3.3.2 Emission Sources and Control
The thermal processes are again responsible for cadmium emissions.
Roasting is considered a source of emissions in the copper production,
and is usually controlled with.a settling chamber, water spray, and some
use of the ESP. Reverberatory furnace emissions are controlled with the
ESP; converter emissions are controlled with settling chamber plus
cyclones, and, at times, with an ESP. The settling chamber with cyclone
40 '
-------
?i
CN
I
in
OJ
u-
H
P4
P4
o
U
H
O
§
Ul
cj
<
tu
or
o
Ul
-J
Ul
Oi
a:
o
«•
o
-2 SZ
TX
w> X.
<
_l
a.
CO
:rcc
•
10
Ul
a:
ui
o
CL
o
- o
41
-------
is only good for larger particulates, therefore, control is not as
efficient as possible at some smelting plants.
5.3.3.3 Emission Estimates
Primary copper production is a major source of atmospheric cadmium
emissions. Two sets of emissions factors have been found, one for emis-
sions from uncontrolled facilities, and one for facilities with the
baghouse filter. In an uncontrolled smelter, emission factors range from
ble 7 x 1C
2,7,8,10/
-2 1
a possible 7 x 10 pounds/ton of copper to 2.9 x 10 pounds/ton of
copper. '•''—• A best judgement figure of 1.5 x 10 pounds/ton of
copper was developed from an average of five mass balance estimates. If
all copper is produced without the use of control devices, 108 tons of
cadmium would be released yearly. However, copper smelters employing
baghouse control devices release only 7 x 10~ pounds for every ton of
2/
copper produced. Five tons per year are emitted from these smelters.
EEA's estimate for uncontrolled copper smelter emissions is very closely
aligned to GCA's of 110 tons/year, ' while EPA ' and Mitre ' estimates
equal 234 tons and 388 tons, respectively. Both these emissions estimates
were made at earlier dates before certain data were available.
5,3.3.4 Future Trends
There are several possible applications which would create an increased
demand for copper, and some technologies which will cause a decrease in
copper consumption. The factors should combine to increase copper .
production to over 3.5 million tons by 1985.
The recent energy shortages have encouraged the development of solar
energy applications and electric cars. In comparison to the conventional
1975 model car with 41 pounds of copper, the electric car is expected to
contain about 200 pounds per car. Desalinization technology, which
employs copper alloy tube, has become an industry of significance in
countries such as Saudi Arabia. Fire sprinkler systems are becoming a
required part of any new inhabited building, and this is expected to
increase copper demand and production. An increasing use of electronics
42
-------
in the telephone industry has reduced the use of copper conductors, and
glass fiber optics will produce the same effect. Therefore, emissions of
cadmium from primary copper production are expected to increase to 13
tons/year by 1985 in controlled smelters.
5.3.4 Cadmium _„
Unlike the other three metals discussed, cadmium is not mined, but rather
is a by-product of other metal productions—primarily zinc. "Blue Powder,"
a part of the volatile product from zinc distillation, was at one time
the major source of cadmium metal for commercial production. Most of the
cadmium which is present in zinc and lead ore is now removed in earlier
stages of processing.
There are several major sources of cadmium. Dust and fumes collected in
the bag filters and ESP during roasting and sintering of the ores, and
the cadmium and zinc filter cakes resulting from the purification of zinc
sulfate solutions are two important sources.
Additional flue dusts, primarily from Mexico, are imported to meet
consumption demands.
5.3.4.1 Process
There are two basic methods used to recover cadmium: one which serves
to extract the metal from zinc ore roasting and sintering flue dusts, the
other which extracts the metal from any leaching residues. Other pro-
cesses are used to remove the metal from other flue dusts and slab zinc
redistillation.
It is first necessary to insure that as much cadmium as possible is
incorporated into the roasting and sintering flue dusts.. This is done
in one of two ways. In one, after roaster flue dusts have been returned
to the roasted zinc ore concentrates, zinc chloride or sodium chloride
43
-------
is added. Sinter-scalping, the second method used to concentrate cadmium
in flue dusts, involving releasing the "top" part of the sinter to zinc
smelting, and recycling the bottom part (where most cadmium is collected)
through sintering.
These dusts are then usually processed for cadmium recovery, which
entails several steps. First, the cadmium is treated with sulfuric acid
to become cadmium sulfate. During this treatment, lead is precipitated,
filtered, and sent to a lead smelter in the form of lead sulfate. Next,
a sponge is formed by addition of acid and zinc dust to the cadmium
sulfate solution. The sponge is then washed and dried, mixed with coke
and distilled. The resultant vapor is condensed and cast into metal
balls or ingots. Any distillation residues are directed back to the
sintering process. The sponge can also-be purified through treatment
with a high-grade zinc dust and converted to molten cadmium metal by
adding molten caustic soda. This process decreases airborne emissions
when compared to the direct melting of sponge.
-------
It is also possible to obtain cadmium from the redistillation of con-
taminated zinc slab. The distillation is a two-step process. First,
the impure zinc is placed in a still at a temperature which distills the
zinc and cadmium to be condensed in a second still, while the lead
remains behind. The zinc-cadmium mixture is then distilled at a tempera-
ture such that the cadmium is vaporized, while the zinc remains in a
molten form. The cadmium is then condensed and cast.
See Figure 5-3 for a flow diagram of cadmium recovery from the above
processes. ' '.'•'•
5.3.4.2 Emissions Sources and Control
Emissions are again a result of the high.,,t.empera1aire,,p,roce^,ses involved,
in the production of cadmium metal. Cadmium distillation and vapor
condensation are major sources of cadmium emissions. Most cadmium
purification processes are wet, so that many steps do not contribute to
airborne cadmium. Most smelters employ high efficiency bag filters or
ESP's as control devices to prevent cadmium emissions.
5.3.4.3 Emissions Estimate
EEA's emissions estimate is lower than estimates by the EPA and GCA due
to EEA's use of current data on control technology. Emission factors
ranged from 25 to 30.5 pounds/ton cadmium produced. '' A "best
judgement" (the average of three mass balance estimates" of 28 pounds of
cadmium produced was used, together with a 1974 production figure of
12/
3,008.2 tons to reach the estimated emissions of two tons of cadmium
2/
emitted from facilities with a 95 percent control efficiency. EPA '
2/
estimated that cadmium production resulted in 60 tons/year and GCA '
estimate is 50 tons/year. EEA's estimate of 43 tons/year, assuming no
control, is comparable to these figures.
45
-------
CM
M
to
Pt,
5 y ee
X *rt ^
z 2o
3 ' 5°
I l!
I "C-
g
S tu
< cc
5
' 5
'4-
b
o"
iL
•
• • j
i*"
<
u
1
c
Z
«c
•1
«9
'
•2
o
•t
0
z
u
hj
s
0
cr
o
1
u
CADMIUM
CADMIUM
£
u
z
>.u*V4
SOLUTIOH
o
3
u.
v\
s
0
K«
•
h
VI .
0«>
« (
O
z
w«
U
o
z
5
DNOCtlSATII
u
o
3.
Z
ISTILLATIC
0
LJ
O
1
VI
g
IU
e
VI
u
*I
• LCACM
§1
0
ll
J .
i • .
J
1
•
"
*
7 uio *
ill •"
?i
•s?- i§
ol3 -2 =
>- u ^. 5
VI Q£ «
tt O "* o
• ^ v
U
o
z
n
_ V* M
. »* zuj Z
2"" 2 'T,_
°»5l ,
«j<= ^. i
.,' ~ 0 X
/ £%
' cz 2 s(5
^ O **^ *S •?•
•g-S"1 5'— '-
5
VI
cc
.
S>^ '
•"i
is . -5
>I£ " f!l
's2 sff
10 . WDM
ul
a
z
THALLIUM
UJ
cc
(k
1
3
_
M3I1JLV'3U1
1U
IM
B
1
*
2
«
E
-
v»
3
a
2
-j
u.
COP?CR .
. •
z
a
S
vt
•
-
1
cc
• SLUDCE
z
*"(U
ALCir.-ATIC
o
3
3
-J
•s. •
a.
u
J«
0
f
o
a
cc
o
c.
3
S.MCLTING
ri nr nu^it
' " • .
/
u
•x.
ii •
~~H
X
- U*^
3 3
- d. _
So
S" .
in
ow
5
-
*• 5
a.
'Si
NO
S
.
rt
•••3
OT -*
/
. ag
-
•
•
•
.
3
5
3
U
O
z-
Nl
. 1
5
I
O
| «J.
C
fsi
t
U
u
r
• C3
_l
• vl
... ^
•
•
* *
_
-
w
ul
o
s
*
p
I •f-N.
1 1 "' "*
o ' S
*- >•
°- — — s
i " 5
5 " *
4J i
V* ~"
1
5 *
— o .
»— »• *
< ^_e:
_i T~ < *
•~i w<
«j • • ••;
o *
z
46
-------
5.3.4.4 Future Trends .
Cadmium production has decreased quite steadily in the past five years,
from 3,760 tons in 1972 to 2700 tons in 1976.12/ This is due, in part,
'.to the decreased use resulting from environmental regulations and is
closely associated with the drop in production of zinc. However, cadmium
production is not expected to continue its downward trend, but rather to
increase .along with zinc production in the near future. At present,
cadmium is used for a variety of items which will be discussed in the
following sections. However, it is possible that the producers of solar
energy equipment could become large consumers of the metal. Cadmium
production is linked to zinc production and, therefore, is expected to
increase at a similar.rate. Emissions are projected at less than one ..
ton/year by 1985.
5.4 IRON AND STEEL
The iron and steel industry releases relatively large amounts of cadmium
into the air. These emissions are a result of the melting of cadmium-
coated scrap, usually number two steel scrap.* Descriptions of the four
major types of plants, including the sintering strand, the open hearth,
the basic oxygen furnace, and the electric arc, are included with current
industry trends. The blast furnace, which emits almost no cadmium, is
excluded.
5.4.1 Sintering
The sintering plant receives two different materials .which require
processing. Beneficiation of very fine iron core and flue dusts is
accomplished at the sinter plant and different iron-bearing materials
become agglomerated at the plant. Sintering of the above materials
ensures a higher iron content, lessens moisture, and often removes some
sulfur. This renders the resultant product a more suitable feed.for the
blast furnace.
47
-------
5.4.1.1 Process
To accomplish the sintering, a. mixture of fine ore and powders of carbon
sources, such as anthracite and coke breeze, are placed on a traveling
grate. The grate moves over a series of windboxes where the mixture is
lit with a burner. As air is pulled down through the ore with fans, the
ore mixture burns, agglomerating the ore particles. The. use :of sinter
aides the performance and productivity of blast furnaces.
5.4.1.2 Emission Sources
During the heating of the mixture of fine 'ore and other'materials (blast
furnace and other flue dusts), cadmium 'escapes into the atmosphere.
Because the flue dust, and other such materials contain amounts of
cadmium, any heating of this material causes cadmium to volatilize.
5.4.1.3 Control
Sinter strands employ one or more of three types of control devices.
The cyclone is employed alone, or more often with a baghouse or ESP.
In 1976, approximately 66 percent of the sinters used little-or no
control, while 34 percent employed at least one of the above methods.137'
5.4.1.4 Emissions Estimate
The estimate of airborne cadmium emissions emanating from the sintering
process was developed from three sets of EPA stack sampling tests for
which a semi-quantitative analysis technique, emission spectroscopy
(ES), was used. . Sampling runs had been taken for both controlled and
uncontrolled sinter plants, so emission factors were developed for both.
The uncontrolled emissions factor of 2 x 10~3 pounds/ton14/ of feed
indicates that approximately 22 tons of cadmium result from the production
of 21.94 x 10 tons of sinter. ' With a rotoclone plus ESP control,
the sinter windbox tested had an average emission factor for two runs of
48
-------
9.5 x 10~4 pounds/ton of feed.14'' Assuming that this emission factor is
representative of all controlled sintering operations, the total annual
emissions of cadmium from these sources equal 5 tons yearly (11.35 x 10
tons of sinter13''). Total emissions (controlled and uncontrolled) from
sintering equal 27_tons'yearly.* EEA's emission factors, based on the
windbox tested, imply a control efficiency of only 50 percent. Industry
comments and stack data suggest that a 90 percent control efficiency may
be attainable.48'50'' If these estimates and stack tests are confirmed
then industry-wide emissions would be much lower.
5.4.1.5 Future Trends
.Sinter strand production is expected to increase slightly between 1974
and 1985. A decrease in production occurred between 1974 and 1975.
However, by 1977, a recovery had been made and production increased
slightly over 1974 (33.3 MM tons in 1974; 34.0 MM tons in 1977). Based
on the historic growth rate, it is expected that sinter production will
reach 40 million tons/year by 1985.13' However, sinter capacity will
remain constant at 46.9 million tons ' through at least 1983. By 1985,
it is expected that all facilities will be in compliance with control
regulations, and all will employ some control device. On this basis,
cadmium emissions from sintering are expected to equal 19 tons/year by
1985. This assumes a control efficiency of only 50 percent, which is
unrealistic according to industry. ' ' Using an emission factor of 1
x 10~4 Ib/ton of feed, which reflects 85 percent control efficiency,
total 1985 emissions would amount to only about 2.3 tons. However, an
accurate estimate of current or expected cadmium emissions is extremely
difficult for two reasons. First the.industry is upgrading existing
control technology and even replacing existing types of technology with
new types. Second the amount of cadmium released is a function of both
the type (ore or millfines) and day-to-day variability of the feed.
*Comparative emissions estimates will be discussed at the conclusion of
the iron and steel section, as other estimates are not broken down by
process.
49
-------
5.4.2, Basic Oxygen,
The basic oxygen process was developed and,, first used in the early
1950's. It has become a highly competitive form for producing steel and
has replaced much of the open hearth production.,
5.4.2.1 Process
The-basic, oxygen, furnace.-is,-a cylindrical: s.teel, furnace, lined with re,-.
fractory material,- which,,has an opening at only one end,. Fpr, charging
and tapping the furnace, vessel,, rotates, around,a lateral-axis.. Slag is
retained in the; furnace, and, is; then tapped, off; by way- of a t^.p-hple,.;nearv
the mouth of the furnace. Thei-charge, materi.als u,se,d, in, the^ basic oxygen
furnace include hot metal from blast, furnaces (70-80 percent-), scrap,,...—-..
cold pit iron,' and iron oxide.
In the process, the furnace is, first tilted, for, the .addition of, scrap,
and hot metal. It. is then brought to a; vertical -position,. Next, an
oxygen injection lance is placed in the furnace producing a reaction
flame which is visible along the furnace mouth. Through an, overhead
chute, fluxes of lime and fluorspar are funnelled into the.furnace
causing the reaction, flame to, rapidly decrease. Temperature, readings
are taken and. the refined metal and slag samples are analyzed after the
carbon content reaches the.desired l.e.vel,. Finally, the furnace is
tilted to tap steel into a, ladle..
5.4.2.2 Emission Source and Control.
Emissions of cadmium result, from the. melting of scrap in the furnace.
From the time the, furnace is charged, with the scrap, cadmium contained
in the scrap is volatilized in the furnace,.
The emissions from the basic oxyge.n furnace are controlled primarily by
ESP's. However, approximately 40 percent of all controlled basic oxygen
50
-------
furnaces use high energy venturi scrubbers to aid in the abatement of
airborne cadmium emissions. ' Approximately 98.5 percent of all existing
basic oxygen furnaces employ some type of air pollution control device.13'''
5.4.2.3 Emission Estimate ....''
To determine emissions resulting from the basic oxygen process, emissions
factors were developed. For uncontrolled furnaces, an emission factor of
,-5.
4.1 x 10 pounds/ton of steel was developed by assuming a reported
concentration of cadmium in the particulate of 80 ppm and the AP-42
Production of
particulate emission factor of 51 pounds/ton of steel.15''
steel in these facilities equaled 1.2 x 106 tons.13/ The emission
factor of 1.2 x 10~ pounds/ton of steel for furnaces controlled by a
venturi scrubber or ESP is the average for six EPA stack tests using
analysis by emission spectroscopy. ' Production of steel at these
/- -I ry t
facilities equaled 78.8 x 10 tons. ' Emissions from this source are
estimated by EEA to be less than one ton/year.
5.4.2.4 Future Trends
The basic oxygen furnace is considered to have a substantial growth
potential through 1985. It is thought that as much as 75 percent of the
1985 steel production in the U.S. will be achieved through .this process.16/
Obviously, emissions will increase (to 1 ton/year); however, the source
will remain one which does not produce over one ton of cadmium emissions
each year.
5.4.3 Open Hearth
Open hearth production of steel has been decreasing steadily over the
past ten years due to economic and environmental concerns. However, the
process does account for 25 percent of the steel produced ' and is a
source of cadmium emissions.
51
-------
5.4.3.1 Process
A reverberator/ type hearth furnace is heated alternately by a combustion
flame from either end of the hearth. At regular intervals, the gas flow
is reversed in order to recover sensible heat from the exhaust gases.
This is accomplished by passing them through brick checkers which are at
either end of the furnace. At each reversal, the brick checkers are hot
enough to heat the incoming combustion air so that the high flame tem-
peratures needed to melt and refine raw materials are more readily
reached. The furnace is charged with scrap and heated to incipient
melting by oil, gas, or tar flames which move across the top of the
hearth. Hot metal is added to the furnace at this point. The next step
involves addition of the necessary flux and oxidizing materials to
refine the mix while it boils.
5,4.3.2 Emission Source
The cadmium is volatilized when the scrap is melted.
5.4.3.3 Control
Eighty percent of open hearth furnace capacity uses ESP's to control
emissions, while 20 percent is without any kind of controls.
5.4.3.4 Emissions Estimate
Open hearth emission factors have been determined for both the controlled
and uncontrolled operations. A "best judgement" estimate for uncontrolled
-3 "}/
furnaces of 5.78 x 10 pounds/ton of steel ' is determined to best re-
_3
present estimates between 4.08 x 10 pounds/ton of steel and 6.48 x
10 pounds/ton of steel. ' The ESP control reduces emissions to between
2.08 x 10~5 pounds/ton of steel and 1.34 x 10~4 pounds/ton of steel.5'15'17/
This recent series of stack tests using AA analysis has produced a "best
estimate" of 1.1 x 10 pounds of steel. From this, and a production
52
-------
/* -1 ry t
figure of 29.06 x 10 tons, ' it was determined that controlled open
hearth facilities contribute two tons of cadmium into the atmosphere
each year. Uncontrolled emissions, with a production level of 7.64 x
ft 1 ^ /
10 tons, equal approximately 22 tons/year, resulting in a total of
24 tons of cadmium emissions yearly.
5.4.3.5 Future Trends
As has been stated,.the open hearth is not environmentally and economi-
cally competitive with other steel-making processes. Since 1965, open
hearth steel production capacity has dropped from 110.82 MM tons/year to
37 MM tons/year.
13/
It is expected to continue declining through 1985
to 21.1 MM tons/year. '' No new open hearth facilities are planned.
Emissions, assuming that 100 percent of the open hearths will use some
control device, are estimated to be about one ton/year by 1985.
5.4.4 Electric Arc:
The electric arc furnace, the fourth method by which steel is produced,
has enjoyed steady growth since its initial installment in 1906. At
present, it accounts for the production of approximately 27.3 MM tons of
13/
steel each year. The process allows for the production of a large
variety of objects and is flexible in operation, which accounts for its
growing popularity.
5.4.4.1 Process
Steel scrap is the principal material used for feed in electric arc
steel-making, while iron sponge is occasionally used as a portion of the
feed. Scrap is processed in a furnace which is shaped like a cylindrical
shell, shallow in depth, with a large diameter. A removable roof on the
shell allows for the insertion of three graphite electrodes. High
performance refractories completely line the shell and roof.
53
-------
Steel-making is accomplished in a series of heats, each producing a
certain tonnage of some form of steel. Each cycle begins when a partial.
charge is loaded into the top of the furnace, the roof is closed, and
the electrodes are lowered into.the furnace. The electrodes are placed
so that the electric current produces an arc between them and the charge.
After the melting of the partial charge, the remainder is added and
melting continues to the end of the heat. The entire process usually
requires three to four hours for completion. When the process is com-
pleted, the steel is placed in a transfer or teeming ladle, while the
slag is dumped into a slag pit. .
The process is used to produce, a- large number of different steel products
such as structural steel, specialty alloys, tool steels, super alloys,
and stainless steels.
5.4.4.2 Emissions Sources and Control
The melting process which, the steel scrap undergoes in the electric arc
produces cadmium emissions. Because the primary feed for the arc furnace
is steel scrap, this process produces more cadmium emissions than any other
steel production operations. (Sintering, with a slightly higher amount of
cadmium emissions, is a raw materials processing operation). Control is
accomplished with the use of a baghouse and the occasional use of a scrubber
or ESP. Approximately 100 percent of all electric arc furnace capacity
employs some method of control. '
5.4.4.3 Emissions Estimate .
It is estimated that the electric arc furnace emits 3,4 x 10~3 pounds of
cadmium/ton of steel. This "best judgement" estimate was developed from
estimates between 2.7 x 10~3 pounds/ton of steel and 5 x 10"3 pounds/ton of
steel by assuming a particulate cadmium concentration of 735 ppm from EPA
ES stack test results,14>15/ and the AP-42 particulate emission factor of
4.6 pounds of particulate/ton of steel. Using this emission factor
54
-------
factor, and a total production figure of 27.3 x 10 tons, EEA estimated
that at a 90 percent control efficiency, cadmium emissions from this
process equal approximately 5 tons/year.
5.4.4.4 Future Trends
The electric arc furnace is a steadily growing method of producing
steel. The increase in the electric arc is a result of several factors
which include the viability of the "mini-mill" concept, the applications
possible for high-intensity arc furnace technology, larger amounts of
scrap available due to growth of the basic oxygen furnace coupled with
decline of the open,hearth, and increasing economic, viability of direct
ore-reduction processes. Therefore, an historic growth rate was used to
project that there will be an increase of about 11 MM tons produced by
the electric arc through 1985. Emissions will increase somewhat; however,
all electric arcs will be controlled in some way. NSPS standards will
have some effect upon the process, because expansion of existing plants
seems likely.
5.4.5. Total Iron and Steel Emissions
To compare EEA's iron and steel emission estimates with estimates of
other studies, it was necessary to total all processes.
5.4.6 Total
EEA's emission estimate of 56 tons/year is not in agreement with most
others (Davis—1>000;7/ Mitre--1,000 ;U/ EPA--78;2^GCA--400;2/ and
Sargent—11.5 ). The assumed control technology was based on the
Temple, Barker, and Sloane steel study (1976), which was unavailable
at the time previous estimates were made. Several emission factors were
also based upon data previously unavailable. EEA assumed a higher
degree of control technology and production figures could vary among the
55
-------
sources. Therefore, the variation between estimates is not unexpected.
Assuming that each individual process is entirely uncontrolled, EEA
estimates a value of 146 tons/year for cadmium emissions. Previous
estimates failed to consider processes individually, so that it is
difficult to determine whether EEA's assumptions of control technology
are comparable to those made in,the previous aggregate estimates.
5.5 SECONDARY SMELTING
Secondary smelting processes involving zinc, lead, and copper are not
viewed as large emitters of cadmium when each is viewed as an aggregate
(all zinc smelters; all copper; all lead). Each metal will be dealt
with individually in the following discussion.
5.5.1 Secondary Zinc
5.5.1.2 Process
Zinc can be melted, "sweated," or vaporized in processing. To recover
the zinc from scrap, sweating is the most common procedure. The furnaces
employed to carry out this process include rotary, reverberatory, and
muffle furnaces. In zinc melting, the scrap is combined with ingot and
rejects and is melted to create a molten bath. After light scrap is
added to the bath, it is heated to the correct pouring temperature and
poured. The zinc vaporization is carried out in retort furnaces to
reclaim zinc from alloys and to recover zinc from its oxide (among other
processes).
Distillation and muffle furnaces are used to separate zinc from the
alloy which is then processed and converted to zinc metal.
5.5.1.3 Capacity
It is possible to obtain a total production figure for the industry
(75,409 tons in 1974,47//)
production are available.
477
(75,409 tons in 1974, ') but no figures of individual plant capacity or
56
-------
5.5.1.4 Source and Control
The source of cadmium emissions from all three processes is the melting
of the material at high temperatures. In preparing the scrap material
for charging, no control device is used or needed. Sweating furnaces
are equipped with an afterburner and a baghouse, while distillation
furnaces are equipped with baghouses. Only 20 percent of the sweating
furnaces use control devices, while almost all distillation furnaces
employ controls.
I/
5.5.1.5 Emissions Estimate
Cadmium emissions from zinc secondary smelting operations are small.
. .This- .is_ due to the fact that most of-the-cadmium-has already-been volati-
lized during primary smelting.
_2
Uncontrolled emission estimates between 8 x 10 pounds/ton of zinc and
— 2 8 1 0 1 8 /
1.4 x 10 pounds/ton of zinc produced ' ' ' were found in the litera-
' -2
ture. A "best judgement" factor of 1 x 10 pounds/ton of zinc produced
was then derived from the mass balances. The 1974 production figure was
used to compute an emissions estimate of less than one ton of atmospheric
2/
cadmium resulting from the process yearly. EPA ' estimated emissions
2/
from this process to be two tons/year, as did GCA. Sargent combined
all secondary non-ferrous metal processing to obtain an emissions esti-
mate of.2.4 tons yearly. '
5.5.1.6 Future Trends
• Annual production of zinc through secondary smelting is expected to
increase slightly through 1985 to 130,000 tons.47' There is no reason
to assume a dramatic increase or decrease in production. Emission
control will improve so that the increased production as a whole should
not cause a large net increase in emissions. Cadmium emissions would
thus increase slightly to about one ton/year.
57
-------
5.5.2 Secondary Lead
5.5.2.1 Process ,
Secondary led smelting is a process from which cadmium emissions are
derived. Lead automobile storage batteries, lead-sheathed cable and
wire, aircraft tooling dies, type-metal drosses, and lead dross and
skims are the materials most commonly sweated to obtain lead. To process
materials which contain a small percentage of the metal, a rotary furnace
or sweating tube is usually employed. A reverberatory-box type furnace
is used when dealing with material of high lead content.
Blast furnaces are also used frequently in secondary smeltiiig~of lead -
storage batteries. Lead is charged into the furnace at the beginning of
the operation in order to provide molten metal to fill the crucible.
Limestone and iron flux float on the top of the lead to inhibit its'
oxidation. The molten metal is. poured when proper conditions are reached.
5.5.2.2. Capacity
As with zinc, capacity of individual plants could:not be obtained for
this study. However,, 1974 total production was- 698,698, tons.12/;
5.5.2.3 Emissions Source and Control.
Airborne cadmium emissions emanate from the melting process during which
cadmium is volatilized. The control measure used with both reverberatory
and blast furnaces is.generally a baghottse, occasionally in combination
with an ESP. '
5.5.2.4 Emissions Estimates
The amount of cadmium produced by secondary lead smelters has not previously
been estimated. Emission factors for this process were developed from
EPA stack testing results. Three factors for:three different combinations
58
-------
of furnace type and control technology were used to compute emissions
estimates. However, because'of difficulty in obtaining production data,
the most conservative emissions factor of 2 x 10 pounds/ton of lead
was used.14' Combined with the production figure for secondary smelters,
the emission factors produce a relatively low estimate of total emissions
(less than one ton).
5.5.2.5 Future Trends
The production of secondary lead is expected to increase through 1985.
It is felt that the.growth of this sector of the lead industry will
largely account for the growth of the entire industry. However, until
the long-term effects of both the OSHA regulations and EPA rulings are
determined, secondary producers will be reluctant to increase capital
expenditures for new plants and equipment. This indicates that the
impact of NSPS will be minimal at best. Emissions are expected to
increase, but remain under one ton/year.
5.5.3 Secondary Copper
Scrap serves as the primary feed for most secondary copper operations.
The scrap received usually contains many impurities including cadmium.
During the processing, which is discussed below, these impurities are
removed and copper metal is cast for re-use.
5.5.3.1 Process
Scrap can be processed mechanically or by a pyrometallurgical process.
At medium temperatures, sweating removes metals which have a low melting
point. Burning removes insulation which was not removed mechanically
from copper wires. To remove excess cutting fluids from machine shop
chips and borings, vaporization occurs in a heated rotary kiln. The
blast furnace produces a product called "black copper" which is a concentrated
59
-------
material. Concentration is accomplished by taking scrap and charging it
at the top of a vertical furnace together with coke, a reducing agent,
and a fluxing material. The concentrated material plus some slag are
drawn out at the bottom of the furnace.
After the scrap is processed it undergoes smelting and refining, melting,
and alloying in a variety of furnaces, including the reverberatory, the
rotary, or the crucible. The choice of one over the other is dependent
upon the quantity of scrap to be melted and the type of alloy to be
produced. The reverberatory and rotary furnaces are direct-fired furnaces
in which the hot, high-velocity combustion gases are in direct contact
with the metals in the charge. Thus, it is difficult to. effectively
capture the emissions from these types of furnaces with a hood.
During charging, all the scrap material is usually not placed in the
furnace at once due to the large quantities involved.. After charging,
the melting occurs. The burners are set to ensure a very rapid melt-
down, and additional oxygen may be added at this time. Refining, the
next process which the metal undergoes, involves fluxing the molten bath
of metal to cause selective oxidation. In. alloying., virgin metal or
specialized scrap may be added to modify the final product of the melt.
The last step is pouring the molten metal from the furnace into a mold.
This can be done by tapping the furnace directly to an automated mold
line, or into a ladle from which It is transferred to a mold line.
5.5.3.2 Source of Emission and Cpntrol
Almost all processes in secondary copper smelting produce some emissions
of cadmium. Sweating involves a very small loss of metal fume. Burning
the insulation from copper wire may be a minor source of cadmium emissions,
but no data has been found to substantiate this. However, the rotary
kiln, which vaporizes copper and cadmium, can cause a large amount of
emissions. Usually afterburners are employed to complete combustion and
-------
decrease emissions. The combustion processes which result in air-borne
cadmium emissions include those involving the blast furnaces, direct •
fire furnaces, charging, melting, refining, and alloying. During wire
burning no control devices are employed. However, in sweating, an
afterburner and baghouse are used. The blast furnace is equipped with a
baghouse, while the reverberatory and rotary furnaces both are equipped
with a hood and baghouse.
5.5.3.3 Emission Estimate
Approximately 38 tons of cadmium emissions result each year from secondary
copper processing. Emission factor estimates range from a minimum of
2.6 pounds/ton of copper scrap to a maximum of 4.0 pounds/ton of copper
.scrap. The factor used (three pounds/ton~ of copper}'"is' "the average"""of—"
two mass balances. A 1974 production figure of 513,308 tons of copper was
also used. EEA's estimate is lower than that which have been made up to
this point except one (Davis—125;7/ Mitre--125;11/' EPA— 65;2/ GCA--702/).
This is because a greater number of control devices and a greater efficiency
of control were assumed by EEA (95 percent particulate control efficiency
for the nearly universally employed baghouse). The Sargent figure of 2.4
tons of cadmium released per year from secondary non-ferrous metal pro-
cessing is lower than any of the others.6/'. Using a collection efficiency
of 90 percent, EEA's estimated emissions of 77 tons are comparable to
other estimates, especially those of EPA and GCA.
5.5.3.4 Future Trends
The Bureau of Mines predicts that production of secondary copper will
increase five percent each year to 1985.20' Using 1975 as a base, this
would mean 836,000 tons of secondary copper would be produced by the
industry in 1985. Emissions would also increase substantially through
1985 to 62 tons.
61 ..
-------
5.6 MANUFACTURING
The production of paint pigments, plastic stabilizers, and nickel-
cadmium batteries results in cadmium emissions into the air.
5.6.1 Cadmium Pigments
Cadmium compounds, principally the sulfides and sulfoselenides, are used
as coloring agents in paints and plastics. The sulfide compounds are
used to impart colors of yellow to orange, while sulfoselenide colors
range from light red to dark maroon.
In order to "stretch" the pure cadmium pigments, white barium sulfite is
often mixed with the pure pigments to create "cadmium lithopones."
5.6.1.1 Process
The use of these pigments is widespread due to their stability in light
at relatively high temperatures in various chemicals, in various weather
conditions, and in moisture. This makes the use of the cadmium .pigments
in plastics, which require high temperature molding, extremely practical.
Other uses include artist's colors, rubber, and printing inks.
The preparation of "these pigments does cause release of airborne cadmium
emissions. By heating cadmium and sulfur (Cd + S -£££•—>- CdS) , cadmium
sulfide is produced. It can also:be made-by precipitating an•aqueous
solution of soluble cadmium salts -and soluble sulfides or fcLS. Color
variations (yellow to orange) are produced when the temperature of the
H2S solution is altered (a cold acid solution yielding yellow, and the
hot solution producing orange). In preparing cadmium sulfoselenide
pigments, selenium is added to solutions of barium salts. This solution
is, in turn, reacted with CdSO . To remove any unreacted selenium, the
final product is calcined with excess sulfur. Each of the above processes
must contain a calcination, or drying step. Without calcination, the
pigments would not be dry and would be impossible to compound.
62
-------
5.6.1;2 Emissions Source and Control
Any loss of cadmium to the environment during the manufacture of cadmium
pigments originates from the dust which is produced during calcination.
However, all facilities which are involved in pigment production have
installed baghouses to minimize emissions.
5.6.1.3 Emissions Estimate
The emissions factor used to calculate cadmium emissions was the mass
7/ • 8/
balance estimate developed by Davis and cited by Anderson. ' A factor
of 15 Ibs/ton of cadmium charged is suggested as the emission factor for
this process. With production of 1,212 tons in 1974, nine tons of
cadmium/year are released into the air. Davis, Mitre, and GCA all
estimate emissions from this process to be 11 tons/year. The estimates
of seven tons/year by the EPA and 9.5 tons/year by Sargent are in closer
agreement with EEA's than other figures.
5.6.1.4 Future Trends
The use of cadmium pigments is expected to increase slowly during the
near future to 1,560 tons, which will slightly increase emissions from
this source (12 tons). Substitutes for cadmium pigments are available
(zinc, lead, and barium chromates). The yellow cadmium-based dye used
in printing inks has proven extremely difficult to bleach on paper re-
cycling operations. However, it is. expected that cadmium pigment pro-
duction will steadily increase.
5.6.2 Plastic Stabilizers
In order to protect plastics from degradation by heat and light, it is
necessary to add stabilizers to them, especially to PVC and related
polymers. Most plastics which are destined for use outdoors are stabilized
in this manner.
63
-------
5.6.2.1 Process
Stabilizers must counteract the loss of acid, usually HC1, by the PVC
because this is usually the first step in degradation of the plastic.
Cadmium stearates, which are long, straight-chain organic esters, will
react with the HC1 to produce weak organic acids and. ionized cadmium
chloride. When used with no other additives, this stearate .successfully
lends the desired heat and light stability to the plastics. When com-
bined with barium compounds, epoxides, or organic phosphates, the stearate
is even more effective. Because it is the least expensive of the available
cadmium plastic stabilizers, it is also the most popular. Other compounds
used include cadmium laurate and cadmium recinoleate.
5.6.2.2 Emission Sources and Control
Any airborne cadmium emissions which result from the processing of
plas.tic stabilizers originate from the handling of pulverized cadmium
oxide. The cadmium oxide is used to prepare c'admium soap or other
organic stabilizers. The procedure involving the mixing of the prepared
compound with the plastic is not believed to produce any appreciable
emissions. All cadmium stabilizer producers use baghouse to control the
emissions resulting from the process.. .
5.6.2.3 Emissions Estimate
It is believed that" the stabilizer"industry is responsible for'very
little of the total cadmium in the air. Using an emissions factor of
7 8/
six pounds/ton of cadmium" charged. ' developed by EPA from a mass
balance analysis, and a 1974 productionfigure of 991.8 tons, EEA estimates
that three tons of cadmium/year*are released into the air in the pro-
duction of stabilizers. Other estimates, including those made by Davis,
21
Mitre, and GCA agree with that of EEA, while EPA estimates one ton of
cadmium a year from the stabilizer process and Sargent estimates a*
release of 2.9 tons yearly.
64
-------
5.6.2.4 Future •. .
Because of ever-increasing FDA bans on the use of the cadmium compound in
plastic stabilizers, particularly for use in plastic food.wrappings, an
almost constant annual production is projected for this product.21^
Although figures show a slight increase .(due to a small market recovery
after 1973-1974), emissions are.'estimated to increase only slightly.
through 1985. The development of a calcium-zinc stabilizer, which can
equal cadmium stabilizers in performance and cost, has also caused a
decrease in the use and production .of the stabilizer.
5.6.3 Batteries . . .
The nickel-cadmium battery is perhaps the only product from which cadmium
is recovered. Developed prior to 1900 by Jungar7~it is-supe'rior- to-other
batteries in efficiency and longevity. However, during processing,
relatively small amounts of the material are released into the air.
5.6.3.1 Process
The "pocket" electrode is used most frequently in the nickel-cadmium
battery to form the cadmium plate. This electrode is produced by pulling
active materials (cadmium sponge) into perforated pockets on a nickel
steel frame. Active materials, such as the cadmium sponge, react with
the electrolyte solution of the battery to produce a charge.
There are two methods used to prepare the active materials. In one,
powdered cadmium or a cadmium compound is dry-mixed with an expander,
often iron, and then inserted into the pockets. Electrolytic coprecipi-
tation of cadmium and iron from an acid electrolyte also produces an
active material to be inserted into the electrode. This process requires
filtering, washing, drying, ball-milling, and blending of the precipitate
to produce the final product. The electrodes are then filled with one of
the mixtures by insertion of machine-made brickettes or loose dust. To
prevent dusting during this process, petroleum oil is often added in
small amounts.
65
-------
Recently, plates which have the cadmium deposited into them have been
used successfully in cadmium-nickel batteries. The production of the
plate can occur in four ways, all of which involve depositing cadmium,
cadmium oxide, or cadmium hydroxide onto a nickel screen or onto a
porous nickel plaque. The methods to accomplish this include:
* Soaking the plate in cadmium formate solution prior to
thermal decomposition in air,
» Electro-deposition from a solution of a soluble cadmium
' salt, most often nitrate,
« Soaking in cadmium nitrate solution prior to reduction in
an atmosphere of hydrogen, and
* Forcing a paste of active material into the nickel support.
At some point, either after or before the battery is assembled, the
electrodes are subjected to the "formation" treatment. This entails
submitting the electrodes to several charge-discharge cycles. This
serves to remove impurities and loose particles.
5.6.3.2 Emissions Source and Control
There are several potential sources of airborne cadmium emissions from
the production of the nickel-cadmium battery. All procedures which in-
volve dry powdered cadmium and cadmium Compounds, and the reduction and
thermal decomposition steps which require high temperatures contribute
to the emissions. During production of action materials, cadmium is
released into the air. With adequate control, however, this industry
should not prove to be a source of a large amount .of airborne cadmium.
Controls used by this industry are unknown at this point, both in kind
and number. Adequate techniques,'such as baghouses or ESPs, are
available to the industry but are considered to be extremely expensive.
.66
-------
5.6.3.3 Emissions Estimate
EEA estimates that, based upon a mass balance emission factor of two
7 8/
pounds/ ton of cadmium charged, ' ' and a 1974 production figure of
628.14 tons, approximately one ton of emissions per year is released
by the process. With the exception of Sargent (0.7 tons/year), all
others did not report a specific emissions estimate from battery pro-
duction.
5.6.3.4 Future Trends
The EPA estimates that the amount of cadmium used in batteries in 1985
will increase between 15-20 percent, while the Bureau of Mines predicts
only small increases. ' The large increase predicted by EPA is based
upon several factors. These batteries are also used in calculators "arid
portable garden, power, and hobby tools, all of which are expected to
increase in demand due to relatively low prices and increasing popularity.
EEA's estimated future emissions from the production of nickel-cadmium
batteries are based on a projected annual increase of 15 percent and in-
dicate that approximately two tons/year may be released from this industry.
5.7 FOSSIL FUEL COMBUSTION
Cadmium is found in fossil fuels, and therefore, these fuels are a
potential source of cadmium emissions. It has been determined that
coal-fired and oil-fired power plants, fuel oil, diesel oil, and gasoline
are all responsible for some airborne cadmium.
5.7.1 Coal-Fired Power Plants
The coal-fired power plant represents a source of cadmium air emissions
21/
which is small unless the fly ash is not collected.
67
-------
5.7.1.1 Process
To produce power, steam is generated using a fossil fuel. The fuel and
a stream of air which has been preheated are directed to a furnace or a
series of burners where combustion occurs. Because the process is not
carried out under perfect conditions, incomplete combustion usually
results. Incomplete combustion and the incombustible nature of some
fuel constituents cause pollutants, such as fly ash, to be generated by
the process. The heat from the combustion chamber heats water which is
contained in a series of pipes in a boiler and generates steam.
5.7.1.2 Source of Emissions and Control
Emissions of cadmium particulates from coal-fired power plants originate
from coal combustion. Impurities which exist in the coal, such as
cadmium, are volatilized and condense on the particulate matter or fly
ash. Devices used to control the emission of fly ash include ESP's and
fabric filters.
I/
Approximately 97 percent of the coal-fired power
plants employ one of the above control devices.1''
5.7.1.3 Emissions Estimate
The emissions from coal-fired power plants were estimated using emission
factors developed from stack test results reported in several sources
and the total coal consumed by such plants. These stack tests resulted
in factors ranging from 1 x 10~4 to 1 x 10'1 pounds/ton of coal (uncon-
-4
x 10 pounds/ton of coal (con-
trolled)14'22'23/ and from 1 x 1(T6 to 7 _ . ,™», ,*,«. „, «,« ,<
trolled with ESP).2'8'10'14/ The wide range of these factors is due
largely to the variation in cadmium content of the coals burned. Best
estimate factors were developed by taking geometric means of the stack
tests. It was found that the emissions due to uncontrolled coal-fired
power plants burning 3.913 x 108 tons of coal,42/ with an emissions
factor of 1 x 10" pounds/ton of coal, equal approximately 196 tons. If
all facilities were controlled with an ESP or its equivalent, an emission
68
-------
factor of .6 x 10~ pounds/ton of coal would be used together with the
tonnage to estimate that 12 tons of cadmium would be released yearly.
The controlled estimate is much more likely, as nearly all coal-fired
power plants employ high efficiency particulate control equipment.
Other emissions estimates have treated fossil fuel combustion in total.
Thus, a comparison will be made on the basis of total emissions.
5.7.1.4 Future Emissions
Studies indicate that coal will become a more important source of power
in the near future, and its use will increase, especially if the National
Energy Plan is implemented. Therefore, uncontrolled emissions from this
source will become greater, however, with increased control and improved
technology, emissions will increase only slightly.24''
5.7.2 Oil-Fired Power Plants
Oil-fired power plants are similar to coal-fired power plants in terms
of process and emissions. These plants produce a slightly smaller
amount of cadmium air emissions than the coal-fired variety.
5.7,2.1 Process
For a description of.the steam generation process, refer to coal-fired
power plants. The major difference between coal- and oil-fired plants
is that the firing mechanism and equipment required for oil are greatly
simplified.
5.7.2.2 Emissions Source and Control
Thermal processes are responsible for the cadmium emissions in almost
every source, and oil-fired power plants are no exception. The combustion
of oil to create steam releases small amounts of cadmium impurities into
the air. Ninety-nine percent of oil-fired power plants are controlled
with cyclones.
69
-------
factor of 6 x 10~ pounds/ton of coal would be used together with the
tonnage to estimate that 12 tons of cadmium would be released yearly.
The controlled estimate is much more likely, as nearly all coal-fired
power plants employ or will employ high efficiency particulate control
equipment. Other emissions estimates have treated fossil fuel combustion
in total. Thus, a comparison will be made on the basis of total emissions,
5.7.1.4 Future Emissions
Studies indicate that coal will become a more important source of power
in the near future, and its use will increase, especially if the National
Energy Plan is implemented. Therefore, uncontrolled emissions from this
source will become greater, however, with increased control and improved
technology, emissions will increase only slightly.24''
5.7.2 Oil-Fired Power Plants
Oil-fixed power plants are similar to coal-fired power plants in terms
of process and emissions. These plants produce a slightly smaller
amount of cadmium air emissions than the,coal-fired variety.
5.7.2.1 Process
For a description of the, steam generation process, refer to coal-fired
power plants. The major difference between, coal- and oil-fired plants.
is that the firing mechanism and equipment required for oil are greatly
simplified.
5.7.2.2 Emissions Source and Control
Thermal processes are' responsible for the cadmium emissions in almost
every source, and oil-fired power plants are no exception. The combustion
of oil to create steam releases small amounts of cadmium impurities into
the air. Ninety-nine percent of oil-fired power plants are controlled
with cyclones.
70-
-------
5.7.2.3 Emissions Estimate
The emission factor derived from EPA's emissions test results using ES
analysis is 9 x 10~ pounds'/gallon for a plant with ESP control.14'
The range of estimates fell between 7.1 x 10 and 4.4 x 10~6 pounds/gallon.
Combined with an oil usage figure of 2.072 x 10 /gallon,42' this produces
an estimated nine tons of cadmium a year resulting from oil-fired power
plants.
5.7.2.4 Future Trends
Although oil is expected to increase in price and become increasingly
difficult to obtain, usage in the next few years is expected to continue
24/
to rise. Emissions are projected to rise to approximately 14 tons/year.
5.7.3 Other Fuel Oil Combustors
Fuel oils, including residual and distillate fuels, are used in various
boilers or burners to supply heat to the residential, commercial, and
industrial sectors. The process of operation is similar to the combustion
which occurs in the boiler of an oil-fired power plant, but usually on a
much smaller scale.
5.7.3.1 Emission Source and Control
The incomplete combustion of the oil and the impurities in the oil
result in emissions of various kinds, including cadmium. There are no
controls on these sources of cadmium emissions.
I/
5.7,3.2 Emissions Estimate . .
Estimates of cadmium, emissions indicate that heating oil is not a large
source. An emissions factor of 3 x 10" pounds/gallon ' for residual
fuels was developed as an average of six reported cadmium concentrations
in the fuel, assuming that all cadmium was emitted. Distillate fuel
71
-------
emissions were calculated using the diesel oil emissions factor of 7 x
pounds/gallon of oil consumed, which is a best judgement figure
10 _ . o t
based upon factors ranging from 6 x 10~7 pounds/gallon to 2 x 1(T6
o I r\ 10 97 /
pounds/gallon. > > > ' Residual fuel consumption equaled 321.2 x 10
6
39/
barrels, and distillate fuel consumption equaled 613.9 x 106 barrels.'
The resulting 1975 emissions are 9 to 20 tons for distillate and residual
fuels, respectively.
5.7.3.3 Future Trends
The future consumption of heating oil will rise slowly through 1985 -to
980.9 x 10 barrels. ' To determine residual and distillate usage,
proportions were assumed to be equal .to those of 1975. Of total fuel
consumed, residual fuel accounted for 34 percent and distillate for 66
percent. Total emissions are 31 tons/year based on consumption figures
of 647.4 x 10 barrels of dis.tillate fuel .and 33,3. 5,x 106 barrels of
residual fuel in 1985.
5.7.4 Diesel Oil
5.7.4.1 Process
Diesel.oil, which is burned in .the diesel engine, is used by some.automo-
biles, trucks, and other motor vehicles. ,An unregulated flow-of air is
fed into the engine and mixed with the fuel. Thi.s mixture. reaches the
cylinder or combustion chamber, is compressed, .and then ignited. The
injection of the highlyrpressurized gases into the cylinder causes a
sudden reduction in pressure, in turn, creating air temperatures which
cause the ignition. The energy of the burning mixture moves the pistons,
and the pistons' motion is transmitted to the crankshaft that drives the
vehicle. The burned mixture then leaves the car through the exhaust
pipe.
72
-------
5.7.4.2 Emission Source and Control
The emission source is the thermal process which causes combustion of
the oil itself. The emissions are actually released through the exhaust
pipe, which accounts for almost 100 percent of the diesel engine's
emissions. Control of cadmium emissions from this source is not practiced.
5.7.4.3 Emission Estimate
Diesel fuel oil emits cadmium at a rate of between 6 x'10~7 pounds/gallon
and 2 x 10~6 pounds/gallon.8'10'18'21/ A best judgement figure, which '
assumes the emission of all the cadmium in the fuel (as measured by ES),'
is 7 x 10 pounds/gallon.. With a consumption figure of 11,179,686 -x
10 gallon, ' and the "best judgement" factor, EEA estimates that four
tons are emitted each year.
5.7.4.4. Future Trends
Diesel oil consumption is expected to rise to approximately 15 x 10'
gallons by 1985. ' Emissions from the combustion of the fuel are
estimated to slightly increase to 5 tons/year.
5.7.5 Gasoline
To complete the discussion of fossil fuels which release cadmium emissions
during combustion, gasoline must be considered. The process which
actually releases the cadmium is the combustion of gasoline within the
engine. Therefore, a brief discussion of this will be included.
5.7.5.1 Process .-...'•
In the conventional automobile engine, a mixture of fuel and air is fed
into a combustion chamber, or cylinder,.by the carburetor, compressed,
and then ignited by a spark from a spark plug. The pistons are put into
motion by the energy released from the burning mixture. Also released
are certain particulates, including cadmium. The emissions pass out of
the car through the exhaust system.
73
-------
5.7.5.2 Source of Emissions and Control
The source of emissions, as noted above, is the actual burning of the
fuel which causes the volatilization of cadmium. Although it is possible
that gasoline detergents leach cadmium from engine parts with which the
gasoline has contact, no data has been found to support the possibility.
5.7.5.3 Emissions Estimate
The emission estimate for gasoline is based on an emission factor of 2 x
10 pounds/vehicle mile traveled in one year. This factor assumes that
all the cadmium in the fuel is emitted and that the vehicle operates at
15 miles/gallon. This best judgement factor was obtained from literature
11 —8
which listed emission factors between 6.3 x 10" pounds/VMT to 4.5 x 10~
pounds/VMT.25'26'27'28'29/ Total vehicle miles traveled by cars and
motorcycles is estimated at 1>330,074 x 10 °' producing estimated
aggregate cadmium emissions of 13 tons/year.
5.7.5.4 Future Trends
It is expected that vehicle miles traveled:will steadily increase in the
coming years. Even'though-vehiclermiles traveled are-expected to
increase, emissions are not projected to substantially increase more than
four tons/year to 1985.
5.7.6 Summary
EEA estimates that the five fossil fuel combustion processes release a
total of 67 tons of cadmium into the atmosphere .each year, assuming con-
trolled coal-fired power plants. This figure is lower than estimates by
GCA (250 tons/year);2/ EPA (198 tons/year);2' and Sargent (130 tons/
year), due to more stringent control technology assumptions. Assuming
a lower (90 percent) control efficiency for coal power plants/ an additional
8 tons of cadmium would be emitted.
74
-------
5.8 MISCELLANEOUS
Several cadmium emission sources contain very small amounts of cadmium
which are not intentionally part of the product. Fungicides, phosphate-
fertilizers, rubber tires, and cement all contain cadmium. Release of
cadmium emissions from these sources occurs in both the production and
use of the products.
5.8.1 Fungicides .
The type of fungicide which is used primarily on golf courses contains a
small amount of cadmium. When the fungicides are applied to the courses,
usually in liquid farm, some- cadmium loss is experienced. This- loss is
dependent upon several factors, including spray particle size and atmospheric
conditions.
•5.8.1.1 Emissions Estimate and Control
Emissions from this source are minimal (less than one ton/year). In
9/
order to develop this estimate, a production figure of 59,800 tons
combined with an estimated emission factor of 1 x 10" pounds/galIon.
No control devices are used in the application of fungicides.
was
7,8/
5.8.1.2 Future Trends
Future growth in production of this product is difficult, if not impos-
9/
sible, to project. Many factors enter into the future of fungicide
production. Leisure time is increasing, and "planned" communities which
include recreation facilities are becoming popular places in which to
live. This would indicate that production would increase as more golf
courses would require fungicides. Fungi may become immune to cadmium-
containing fungicides, which would force a change in the type of fungi-
cide used. This would cause a decrease in emissions. Each year many
variables, including weather conditions and infestation rate, also affect
the use of the fungicides. EEA has used an historic growth rate to
determine that emissions from fungicides will not increase greatly through
1985.
75
-------
5.8.2 Fertilizers
Phosphate and superphosphate fertilizers also contain a small amount of
cadmium. Associated with phosphate rock is a small number of impurities,
one of which is cadmium. The cadmium remains with the phosphorus in
processing and becomes a contaminant in both phosphate and superphosphate
fertilizers.
5.8.2.1 Emissions Source, Control- and Estimate
The cadmium emissions are released during application of the fertilizer,
but not during processing of the material. There are no control: methods.
employed, as there is an inadvertent loss of the material. The- emissions
estimate is based upon an emissions factor of 2, x 10~4 pounds/ton for
superphosphate fertilizer (from an EPA estimate),6'8'31''' coupled with
the production*figure, of phosphate and superphosphate- fertilizers- which
3 9/
equals 8,535 x 10 tons. The cadmium;emissions from both phosphate
and superphosphate fertilizers are a small portion of the total emissions
(about one ton). Other estimates include those by Davis,7/ GGA.,2/ and
Sargent, which are all under one ton.
5.8.2.2 Future Trends
The future of phosphate and superphosphate fertilizers is uncertain in
much the same way as that of fungicides. The industry experienced
moderate•increases in 1977 after lows in 1974-1975.9^ In 1974 phosphate
fertilizers were unavailable due to low production,, while in 1975 pricing
prevented farmers from making use of the fertilizer. Farmers who had
done without the fertilizer in the previous years because of high prices
found that a decreased use of fertilizer did not decrease crop yield.9''
In 1976, production decreased due to oversupply in 1975 (addition of 1.5
million ton capacity plant) and decreasing use. Growth of the industry
will be slow to moderate. ' With a 4.5 percent annual, increase,, based on
761
-------
the 1976 consumption of 9,258 x 10 tons, 1985 emissions will remain at
approximately one ton*
5.8.3 Rubber Tire Wear
Rubber tire wear is believed to be a source of several types of gaseous
and particulate matter emissions, including cadmium. The curing process
in the rubber industry employs zinc oxide as an activator through which
the cadmium enters" rubber processing. Ks the tires are worn down during
use, cadmium is released into the air. There are no controls employed.
_9
Several sources were used to estimate an emission factor of 8 x 10
pounds/VMT, assuming a mix of the types of rubber with various cadmium
contents in the tire population. ' ' ' ' Using total vehicle miles,
9 39/
1,330 x 10 , traveled per year, ' it was determined that rubber tire
wear contributes a small amount of cadmium to the atmosphere (five tons/year)
As mentioned in the discussion concerning gasoline, vehicle miles traveled
will increase, causing emissions from rubber tires to increase.
5.8.4 Motor Oil Combustion
Motor oil combustion also contributes to cadmium emissions. Some oil is
burned in the engine causing volatilization of cadmium and its release
into the air. No emissions control is practiced. The emission factor
q OQ/ 2/
2 x 10" pounds/VMT, reported by Anderson (1973) ' and Deane (1976), '
coupled with vehicle miles traveled for passenger cars only (1,028,121 x
10 ), ' indicates that the combustion of motor oil does not contribute
large amounts of cadmium into the air (one ton/year). In estimating
future cadmium emissions from this source, the same projection used for
gasoline, and rubber tire wear was employed and indicates that through
1985 emissions from this source will increase only slightly.
5.8.5 Cement Plants
In the manufacture of cement, cadmium, is released in rather small amounts.
Cement is a non-metallic mineral product composed principally of lime
and silica, with alumina and ferric oxide acting as fluxing materials.
77
-------
5.8.5.1 Process
Cement is produced in one of two ways, either by the wet process or the
dry process. The four basic steps in the production process include
quarrying and crushing, grinding and blending, clinker production, and
finish grinding and packaging.
The materials which enter the kiln at the top end are dried by combustion
gases which are passed through the kiln counter-current to the materials.
The revolving of the kiln causes the raw materials .to fall toward the
clinkering zone as the carbon dioxide is removed from the calcerous
material [the material containing calcium carbonate which is the major
constituent of limestone). After the partially-fused production clinker
is cooled in the clinker cooler, gypsum or water is added and the mixture
is ground in a ball and tube mill to the necessary fineness.
5.8.5.2 Emission Source and Control
Limestone, which serves as a raw material in cement production, contains
a small amount of cadmium and when it is' processed, cadmium is released.
Control measures used to prevent cadmium from becoming an airborne
pollutant include cyclones, with or without ESPs, and baghouses. '
5.8.5.3 Emissions Estimate
There are many emissions factors which apply to the various methods and
steps of producing cement as determined by EPA ES stack testing. In es-
timating emissions, the most conservative dry process factor was used by
EEA because an increasing number of plants are using the dry process. An
emission rate of 2.6 x 10~ pounds/ton of feed, ' together with the
3 9/
appropriate production figure of 81,210 x 10 indicated that less than
one ton of cadmium a year is released from the cement process. Previous
studies have not estimated emissions from this process.
78
-------
5.8.5.4 Future Trends
The production of cement decreased between 1974 and 1975 due to several
plant closings and slow housing starts. However, in 1976, home construction
increased. Although there has been little plant expansion or construction
of new plants, it is felt that an annual (compounded) rate of 3.2 percent
32/
growth will continue through 1985. Because no new plants are planned,
NSPST will not affect emissions, which, notwithstanding the increased
production, will remain well below one ton.
5.9 INCINERATION
Incineration is a source of cadmium emissions due to the cadmium in the
materials burned. These materials include plastics which contain cadmium
stabilizers, objects painted with a cadmium-pigmented paint, and scrap
metal coated with cadmium. Both municipal incinerators and sewage
sludge incinerators emit cadmium into the air.
5.9.1 Municipal Incinerators
The municipal incinerator is a major source of airborne cadmium emissions,
emitting approximately 131 tons of cadmium per year. The following des-
cription "of the process of the municipal incinerator assumes that there
is no resource recovery and that volume reduction is the prime motivation.
5.9.1.1 Process
First, the refuse is deposited in a receiving area which is essentially
a pit up to 30 feet deep, 100 feet long, and 20 feet wide. From there,
overhead cranes remove the refuse from the pit and deposit it in a feed
hopper, which delivers refuse onto the combustion grates at a constant
rate. ' ' . ,
The most expensive part of the combustion plant, the grates, serve to
transport refuse through the primary combustion chamber and simultaneously
79
-------
insure that the maximum refuse surface is directly in contact with fire.
Many types of grates are used to accomplish this; among the most popular
are moving belts, reciprocating grates, and drum-type rollers.
The refuse is carried into the primary combustion chamber and is burned.
Here, in a conventional refractory furnace, 150 to 200 percent excess air
is supplied in order to prevent refractory materials from erosion by high
temperatures. A result of this process is a large amount of exhaust gas
production. This necessitates the use of a secondary chamber, in which
the exhaust gases are subjected to additional combustion. The gaseous
emissions are then discharged through the chimney stacks.
The resultant solid material, and all material which remains unburned, is
deposited into a residue bin which empties directly into trucks. The
trucks then carry the waste to a landfill.
5.9.1.2 Source of Emissions and Control
Emissions result from the combustion of plastics,, paint pigments^, and
metal scrap which subsequently causes the volatilization of cadmium in the
three items. Control devices most commonly used to combat the particulate
emissions are wet scrubbers. Bag filters, or ESP's, are used occasionally.
Approximately 83 percent of the total number, of municipal incinerators use
some sort of emission control, while 17 percent employ none. '
5.9.1.3 Emissions Estimate
EEA's emissions estimate was based upon emission factors from source
testing and the average rate of processing for a municipal incinerator.
With the use of a scrubber or ESP, municipal incinerators release between
6 x 10" and 1.0 x 10"1 pounds of cadmium/ton of refuse.5'7'8'10'14'33/
The results of a recent stack test using AA analysis were reported as
-2 5/
1.8 +_0.5 x 10 pounds/ton. ' The earlier source testing (flame AA or
80
-------
emission spectroscopy analyses) results were on the order of 10 or,
more commonlyt 10" pounds of cadmium/ton of refuse. ' ' ' Therefore,
_o
a best judgement factor of 1.3 x 10 pounds/ton of refuse was selected.
This factor was based upon the mean of the recent test results minus the
standard deviation in order to adjust for the lower results from previous
stack tests using less accurate analyses. With a refuse figure of
20,143,620 tons ' and the best judgement emission factor, EEA estimates
emissions to be 131 tons/year. EEA1 s emission estimate is higher than
that of Davis (95 tons),7/ Mitre (95 tons),11/ EPA (48 tons),2/ and
Sargent (16 tons), ' but is in close agreement with that of GCA (150
2/
tons). Because other estimates are fairly close to that made by EEA,
one can surmise that the volume of municipal trash has increased since
the others were made.
5.9.1.4 Future Trends
It is difficult to project future emissions resulting from municipal
incineration. Control device use is expected to reach 100 percent
before 1985. The number of municipal incinerators has decreased five
347
percent annually over the last several years. ' However, those incin-
erators which have begun operation recently have capacities much greater
than those which are closing. Therefore, existing capacity does not
accurately reflect an increase or decrease in the number of municipal
incinerators. Predictions call for 49 additional incinerators by 1979.35'
As a result of the decreases in number, but increase in capacity, it is
concluded that emissions through 1985 will remain constant. The effect
of NSPS upon these new plants, and the use of high efficiency control
devices by all municipal incinerators, should support the above assumption.
5.9.2 Sewage Sludge Incinerators
5.9.2.1 Process :
The steps in the sewage sludge process differ in some ways from those in
municipal incineration. First, the temperature of the feed sludge is
raised to 212 F to evaporate water from the sludge. The vaporization
. 81.
-------
and increase in temperature combine to raise the water vapor and air
temperature of the gas, which in turn, serves to bring the volatiles of
the sludge to ignition. End products are water, sulfur dioxide, carbon
dioxide, and inert ash.
The multiple hearth is the most common incineration unit. A number of
solid refractory hearths with a central rotating shaft are encompassed
by a circular steel shell. "Rabble" arms, connected to the rotating-
shaft, serve to agitate the sludge which drops from one hearth to another
through openings in each hearth. An inner cold air tube, which cools
the rabble arms, runs through the central shaft. The shaft, has an outer
tube which serves a similar, function for hot air. In this manner,
continuous feeding can be accomplished.
5.9.2.2 Emissions Source and Control
As sludge is volatilized, cadmium is released into the air. Sewage
sludge contains only a small amount of cadmium, originating from plastics
Qr pigments , plus contaminants which, may have become incorporated' into
the sludge from industrial or domestic wastewater. Until. recently, it
was possible to meet current emissions standards with the use of afterburners,
However, present control of sewage sludge incineration emissions includes
wet s.crubbers. ''
5.9.2.3 Emission Estimate.
Multiple hearth sewage sludge incinerators, controlled by scrubbers, are
estimated by EPA to, emit 7' x 10" pounds of cadmium/ton of dry sludge.8^
EEA estimates that less than one ton results from incineration of 1.460.000
%/
tons of sludge ' each year. This may be an underestimate, if the
emission factor was developed based on emissions from an uncharacteristically
well-controlled facility.
82
-------
5.9.2.4 Future Trends
/- -re /
Sewage sludge production may increase to 1.5 x 10 tons/day by 1985,
and assuming that 25 percent of this amount is incinerated, the
emission of cadmium will increase along with the amount incinerated.
However, several conditions temper this conclusion. The widespread
application of control technology may decrease the rate at which cadmium
emissions increase. On. the other hand, recent bans on off-shore dumping
and continually decreasing land-fill space may increase the amount of
sludge incinerated. Because these factors will probably serve to counter-
act each other, EEA assumed that 25 percent of all sewage sludge would be
incinerated, so that less than one ton of cadmium a year is expected to
be emitted through 1985. This may be a conservative estimate, based on
the assumption that the application of control technology will increase
with the growing production of sludge. It is possible that some com-
munities will find it too expensive to control sludge incineration, which
would either aggravate the air pollution problem by emissions from poorly
controlled facilities, or would result in land disposal of cadmium-
bearing sludge.
5.10 SUMMARY
To determine the population affected by various concentrations of airborne
cadmium, it was first necessary to determine sources of airborne emissions.
Next, the specific part of each process from which emissions emanate was
discerned and emission estimates were calculated. The emission estimates
(Table 5-1) were based upon emission factors (Table 5-2) and production
figures (Tables 5-1). Future emission estimates were also discussed
(Table 5-1).
Airborne cadmium emissions derive from many sources, including: primary
metal processes, production of items which contain cadmium (such as
cadmium paint pigments) , fos'sil fuel combustion, incineration, secondary
metal processing, and the use of items which inadvertently contain
cadmium (e.g., rubber tires).
.83
-------
Emission factors were obtained through an extensive search of current
data sources. Minimum, Maximum, and "Best Judgement" figures were de-
veloped using methods and data reported below, and were tested in what
is believed to be the order of decreasing accuracy:
• DATA: Stack tests usually conducted at only one location
with multiple tests; samples analyzed by atomic absorption
(AA) and results reported as cadmium emission factors or
rates, or as cadmium concentrations in particulate matter
and particulate emission rates or factors.
METHOD: Emission factors for cadmium taken directly or
calculated with reported or assumed values of process
parameters (.e.g., uncontrolled open hearth).
• DATA: Stack tests conducted at one or more location usually
with one test per process; sampling train samples analyzed by
emission spectroscopy, a semi-quantitative method, and reported
as detected amounts or concentrations.
METHOD: Emission factors calculated with reported particulate
emission factors or rates as above Ce-g-j uncontrolled sinter
windbox).
* DATA: Concentration of cadmium in particulate emissions, usually
analyzed by emission spectroscopy (ES), but no particulate factors
or rates reported (for a stack).
METHOD: Cadmium emission factor calculated from the
particulate emission factor or NSPS standard for the source
type (e.g., maximum and best judgement factors for electric
arc furnace).
• DATA; Concentration of cadmium in fuel or feed reported for
for analysis by AA or ES.
METHOD: Emission factor computed assuming 100 percent emission
typical fuel characteristics (e.g., heating oil). Primarily
used for liquid fuels.
• DATA: Survey of industrial plants or site visits conducted by
EPA during development of original cadmium emission factors
(e.g., minimum factor for cadmium processing).
84
-------
METHOD: Mass balances for specific processes. Emission factor
determined from unaccounted-for cadmium or by expected emissions
for typical process and control equipment relative to total pro-
duction or raw materials.
« DATA AND METHOD: Engineering estimates made when no other
data is available (e.g., zinc roasting).
Table 5-1 provides a listing of emission factors which were used in the
EEA study (1974). Production figures were obtained for use with emission
factors to determine the emissions from each process (Table 5-2). A
comparison of emission estimates, including those of EEA, can be seen in
Table 5-3. EEA has estimated emissions assuming the same levels of
control efficiency assumed in previous estimates (EEA 1974-1977 comparable),
and assuming more accurate, current control technology (EEA Current
Control). Estimates differ due to variability in production estimates,
emission factors, and control technology assumption. Not all of t^he
estimates in Table 5-3 consider the same sources of cadmium emissions;
some estimates are more thorough than others. The total estimate for
EEA's current control assumptions is one ton less than that shown in
Tables 5-1 and 5-2, since the estimate of fertilizer emissions is not
included (since no other estimate provided a comparison). Assuming the
application of various control technologies by a percentage of operating
facilities, yearly cadmium emissions total 849 tons. EEA's projections
to 1985 indicate that total airborne cadmium emissions, assuming all
facilities employ some type of control technology, will approximate 972
tons.
85
-------
IABLE 5-1
/UROORNE'CAOMIUK EMISSIONS—1974, isas
Source
HISINO
line
Copper
lead
PRZKAP.Y KETALS
Sine
Pyrone tallur7ic
Electrolytic
Lead
Copper
CaiLliua
SCCOtiCAa* K£7A& PROCESSING
Ircn and Steel
Sinter Hindbox Uncontrolled
Sintnr Hinsibox v/Rotoclone
and ESP
Basic Oxycen Furnace
Onconttolled
icr w/Venturl or ESP
Oyea KeartA Uncontrolled
Cptn Hearth v/ESP
electric Arc Controlled
Blast Furnace Controlled
Xiao
L«ad
Copper
Picaents
Stabilizers
Batteries
FOSSIL FUEL COHaOSTION
Coal-Fired Power Plants
Oil-Fired Power Plants
KMtirg Oil
Diesel Oil
Gasoline
KISCtLU-VIOOS
Kator Oil
Rubber Tire Wear
runcicldes
Fertilizers
Ce=ent
Seua;e Sluice Incinerators
Hunieipal Incinerators
Production
1974-
479,150
1,414,245.8
603,024
423,000
121,945
966,095
1,435,662.4
3,088.2
21.94x10*
11.35xlOS
1.2xl06
7B.SxlOS
7.64xlOG
29.06xl06
27.3x10*
95.2x10
75,409
698,69*
513,308
1,212.1
991.8
628.14
3.913xlOS
SOOxlO6
(barrels)
93S.lxl06
(barrels)
llxlO9
(gallons)
1,330, 074xl06
(VHT)
l,02B,121xlOS
(VMT)
1,330,02
-------
TABLE 5-2
CADMIUM EMISSION FACTORS
SOURCE
.
PRIMARY METALS PROCESSING
ZINC
' Coking
Sintering w/Cyclone
Sintering w/Cyclone & ESP
Roasting
Horizontal Retort
Vertical Retort
Electrothermic
Overall (Not Electrolytic)
Electrolytic
LEAD
Overall Smelter
Blast Furnace w/Baghouse
COPPER
Uncontrolled Smelter
Smelter w/Baghouse, (~95%)
CADMIUM
SECONDARY METALS PROCESSING
IRON & STEEL
Si nter Ui ndbox-Uncontrol 1 ed
Sinter Windbox
w/Rotoclone & ESP
Blast Furnace-Controlled
Open Hearth-Uncontrolled
Open Hearth w/ESP
Basic Oxygen Furnace
Uncontrolled
w/Venturi or ESP
Electric Arc Furnace
SECONDARY ZINC-UNCONTROLLED
SECONDARY LEAD
Blast Furnace w/3 Cyclones
& Baghouse
Reverberatory Furnace w/
Cyclone & Baghouse
Reverberatory Furnace w/
3' Cyclones & Baghouse
SECONDARY COPPER-UNCONTROLLED
FNTNf* lip PSDMTnM— RFAR TV- nnv
LH^UU 1 i "••1111 "_''
Zinc Ore
Lead Ore
Copper Ore
MANUFACTURING
Alloys & Solders-Controlled
Pigments w/Baghouse
Stabilizers (for Plastics)
w/Baghouse
Batteries (Ni-Cd)
MINIMUM
1.961b/TZnThru (EST.MB)
4.061b/TZnThru (STK.AA)
2.10lb/TZnThru (STK.AA)
1.2xlO"Zlb/TZnThru (EST)
1.43lb/TZnProd (EST)
5.2xlO"21b/TPbProd (EST.MB)
7xlO~21b/TCu (EST.MB) '
251b/TCdProd (EST.SURV.MB)
1.35xlO'31b/TFeed (STK.ES)
9.33xlO'41b/TFeed (STK.tS)
4.08xlO"31b/TSteel (STK.AA)
2.08xlO"51b/TSteel (EST.CONC)
3.45xlO'61b^TSteel (STK.ES)
2.7xlO'31b/TSteel (EST.CONC)
8xlO~31b/TZn (SURV, MB)
5.9xlO'71b/TPb (STK.ES)
5.9xlO"71b/TPb (STK.ES)
6.5xlO-91b/TPb (STK.ES) '
2.61b/TCu Scrap (EST.MB)
!
Miscellaneous (X-Ray Screens j
Cathode Ray Tubes, Nuclear)
Reactor Components, etc.)
MAXIMUM
2.48lb/TZnThru (STK.AA)
8.5Slb/TZnThru (STK.AA)
.2.221b/TZnThru (STK.AA)
8.7xlO"21b/TZnThru (STK.AA)
2.961b/TZnProd (STK.AA)
2.6xlO"llb/TPbProd (E.ST.MAX
CONC)
^.gxlO'^^TCu (EST.MB) •
30.5lb/TCdProd (EST.MB-)
2.63xlO'31b/TFeed (STK.ES)
9.76xlO'41b/TFeed (STK.ES)
6.48xlO"31b/TSteel (STK.AA)
1.34xlO"41b/TSteel (STK.AA)
2.79xlO'51b/TSteel (STK.ES)
5xlO"31b/TSteel ( EST, STK, CONC)
1.4xlO'21b/TZn Prod (MB)
3.5xlO"51b/TPb (STK.ES)
4xlQ-41b/TPb (STK.ES)
2xlO'41b/TPb (STK-.ES)
41b/TCu Scrap (EST.MB)
BEST JUDGEMENT
2.241b/TZnThru
6.321b/TZnThru (STK.AA)
2.16lb/TZnThru (STK.AA)
—0 (EST)
6xlO"31b/fTZnProd (EST)
C.5xlO"21b/TZnThru (STK.AA)
1.2xlO"21b/TZnProd (EST)
2.5lb/TZnProd
~0
l.lxlO"11b/TPbProd (EST,
' ' AVE CONC)
5.25xlO"31b/TPb(STK,ES)
1.5xW-11b/TCu
7xlO-31b/TCu '(EST)
•281b/TCd
2xlC"31b/TFeed
9.5xlO-41b/TFeed
— 0 (EST)
5.78xlO"31b/TSteel (STK.AA)
l.lxlO'41b/TSteel (STK,AA)
4.1xlO"51b/TSteel (CONC)
1.2xlO'51b/TSteel
3.4xlO'31b/TSteel (EST.STK.,
CONC)
lxlO"21b/TZn Prod
2xlO-61b/TPb
1.6xlO-61b/TPb
5xlO-71b/TPb • - •
31b/TCu Scrap
ZxiO"11b/TCd in Ore (EST.MB)
lx!0'31b/TZn in Ore (EST.MB)
lx!0-41b/TPb in Ore (EST.MB)
3.2xlO'51b/TCu in Ore (EST, MB)
101b/TCd Charged (SURV)
15lb/TCd Charged (EST.MB)
61b/TCd Charged {EST, SITE)
21b/TCd Charged (SURV)
21b/TCd Charged (EST)
REFERENCES
4,5
5
5
e.
y
6
5,6
6
5,7,8
6
2,7,8,10,11
14
. , 2,7,8,10
.8
..14
14
51
5
5,11,12,15,17
15
14
14,15
2,7,3,18,21
7,10,18
14
7,14
7., 14
7,8
5,2,7,21
2,3
2,3
2,3
2,7
2,7
2,7
2,7
2,7
87
-------
TABLE 5-2 (continued)
CADMiUM EMISSION FACTORS
SOURCE
FOSSIL FUEL COMBUSTION
Coal -Fired Power Plants
Uncontrolled
Controlled (ESP)
Oil-Fired Power Plants
Controlled ( ~ ESP)
Heating Oil (Residual; 16
Fuel Oil)
Diesel Oil
Gasoline (for 15 mpg, all
Cd Emitted)
SEUAGE StUDGE INCINERATORS
Multiple Hearth w/Scrubber
Fluidized Bed w/Scrubber
MUNICIPAL INCINERATORS
Uncontrolled
Controlled (Scrubbers or ESP)
LUBRICATING OIL INCINERATORS
Uncontrolled
MISCELLANEOUS
Motor Oil Consumption
(Vehicles)
Rubber Tire Wear
MINIMUM
lx!0'41b/TCoal (STK.AA)
lx!0'61b/TCoal (STK.AA)
7.1xlO'71b/gal (STK.ES)
1.5xlO~61b/gal (EST.CONC)
6xlO~71b/ga1 (EST.CONC, ES)
6.3xlO'll1b/veh-m1 (EST.
CONC)
lxlO"61b/TSludge (DRY)(STK,
ES)
4xlO~71b/TS Judge (DRY)
(STK.ES)
3xlO~31b/TRefuse (EST)
6xlO"41b/TRefuse (FLAA)
lxlO"101b/veh-mi (EST.CONC)
Fungicides Application > 1.8xlO"51b/gal (EST.MB)
Fertilizers Application i 1.7xlO"41b/T (EST.Ma)
Superphosphate Fertilizers
Application
CEHjHT PLANTS
DRY PROCESS
Kiln w/Baghouse or ESP
Raw Mill Feed w/Baghcuse
Raw Hill w/Baghouse
Raw Mill Air Separator w/
Baghouse
Finish Mill Feed w/
Baghouse
Finish Mill w/Baghouse
Finish Mill A1r Separator
w/Baghouse
WET PROCESS
Kiln w/ESP
Raw Mill w/Baghouse
Clinker Cooler w/ESP
or Baghouse
LIKE KILN (PULVERIZED COAL)
Kiln w/Spray, Settle S
Baghouse
3xlO"71b/TFeed (STK.ES)
lxlo-71b/TFee(j (STK>ES)
7.6xlO"71b/TFeed (STK.ES)
5xlO~71b/TFeed (STK.ES)
7.4xlO"51b/TFeed (STK.ES)
1.7xlO"61b/TFeed (STK.ES)
4.6xlO'51b/TFeed (STK.ES)
MAXIMUM
lxlO"Ilb/TCoal (STK.ES)
7xlO"41b/TCoal (STK.AA)
4.4xlO'61b/gal (STK.CONC.ES)
4xlO"51b/gal (EST.CONC.NA)
2xlO'61b/gal (EST).
4.5xlO'81b/veh-mi (EST.CONC)
2xlO"51b/TSludge (DRY) (STK.ES)
3xlO'61b/TSludge (DRY)(STK,ES)
1.8xlO~21b/TRefuse (STK.ES)
i.OxlO"llfa/TRefuse (EST.MB)
5xlO'81b/ve>i-mi (EST.CONC)
5xlO'51b/gal (EST)
5xlO"Z1b/T (EST.KB)
4.1xlO"71b/TFeed (STK.ES) '
4.3xlO"71b/TFeed (STK.ES)
9xlO"71b/TFeed (STK.ES)
1.6xlO"61b/TFeed (STK.ES)
1.3xlO'71b/TFeed (STK.ES)
2xlO"41b/TFeed (STK.ES)
IxlO'^^TFeed (STK.ES)
6.9xlO"51b/TFeed (STK.ES)
• BEST JUDGEMENT
lxlO"31b/TCoal
6xlO"51b/TCoal
9xlO"71b/gal (STK.ES)
3xlO-61b/gal
7xlO"71b/gal (EST.CONC.ES)
2AiD-81b/veh-mi
7xlO"61.b/TSludge (DRY) (STK,
REFERENCES
14,22,23
7,8,10,15,21,31
• .14
2,7,10,21,18,26,33
7,10,21,18
25,26,27,28,29
7
ES)
1.3xlO"61b/TSludge (DRY)(STK, 7.14'
ES)
6xlO"31b/'TRefuse (STK.ES)
1.3xlO"21b/TRefuse (STK.AA)
ZxlO^^^gal (UNK)
2xlO'91b/veh-m1 (UNK)
8xlO"91b/veh-mi
lx!0-51b/gal
6xlO-31b/T
2xlo-41b/T
SxlO^^TFeed (STK.ES)
3.6xlO"71b/TFeed
2.7xlO"71b/'TFeed
8.5xlO"71b/TFeed
lxlO'61b/TFeed
lxlO-71b/TFeed
2.6xlO~61b/TFeed (STK.ES)
2xlO-51b/TFeed
2xlo-51b/TFeed
lx!0'51b/TFeed
5.7xlO'51b/TFeed
7.1U.14
5,7,8,10,14,33
7
7,8
2,6,7,10 '
7,8
6,7,8,17,21,31
6,7,31
7
7,14
7.14
7,14
7,14
14
14
7.14
14
7,14
14
EST « Estimate; MB • Mass Balance; SITE • Site Visits; SURV = Survey of Companies; UNK • Unknown (In literature); STK * Stack Sampling Results;
CONC • Concentration of Cd In feed, fuel, or emissions (w/STK); ES * Emission Spectroscopy; AA • Atomic Absorption (FL-Flame); NA » Neutron Aotivatio
Heutren Activation.
88
-------
oo
\O
Ol
3 S
o
o
en
CM
LO
LO
00
to
00
to
I
"fr
r-»
0»
CD
o>
to <-{
V
LO
< t^
a. en
r-a
o
LO
O
CN
tN
co
5g
«S
U O>
O
o
LO
LO
o
LO
O
O
o
f-
i-H tO CM
to
I
LO
3
oa
H co
co 2;
w o
O CO
co
CO
S w
i i
< CO
o co
rr *J*
o m
§
CO
1
a. a>
vO
LO
LO
to
CM
OO
OO
to
O
LO
o
OO
C3
o
o
o
r-<
i-l
V
LO
LO
CM
to
CM
V
tM
V
•H 00
£S
Cl i-l
o
LO
o
o
o
o
LO
CM
CM
V
^-\ in
3 «— i
cj cd
4-J
I ^
di • >N
CD
§
O
co
&0
pj
•H
•H
S
^i
+ oJ
C -H
Nl ^
*—> 0,
6
•H
NI
g~
^_l 3
CD -H
Ctf Pn -3
CD O cd
i-J U CJ
in
r~l
cd
co
60
&
•H
§
*J
O
f-i cd
.CD
« •
i
•H
O*
tn
fH
CD
M
•H
i— 1
•H
•8
•M
C/D
t/i
s
o
§
cd
rH
i-l
O
O
tn
, ir-\
•S
-------
CJ
o
f-I
i
u
LT>
CTl
'•<*
OO
r--
r~
I
as
LO
t-
LO
to
1
w
_q
§
H
f- 3
rH P3
•H
CO -P •*
So • cj w
O CM O
LO t-H LO
rH CM
H
co oS
04 *
o -^ ^,1-1
t-i CO ~*
CO l-l r-1
i— t S i
Cd U4 J CO
^t » p?r vD
D-, 2 CTl
S O t-l
LO 1 LO
CTl •*
r-H
CM
•
LO
o
LO
o
o
to
00
I—I
V
o
LO
LO
o
to
CM
o
00
to
CM
I
LO
CM
I/J
•H CO
LO
CTl
CTl
CM
CM
&o
•H
co
PJ
o
•rH
cd
(4
(J
«l
§
CO
0>
•5
0
,-H
rt
PH
•H
O
•H
d>
W)
'B
rH
CO
CD
Cti
co •
I-l
*n
fT,
I— 1
•H
t/3
in
o
BH
U)
3
O
Q
-H
+-*
to
r^
(Q
J
CD
c
S
r-l
r-l
CD
O
, 1
t—i
•H
O
fH
O
•P
o
a
to
-------
REFERENCES
1. Goldberg, A.J., A Survey of Emissions and Controls for Hazardous
and Other Pollutants, EPA PB 223-568, Washington, D.C., February
1975.
2. Deane, G.L., Lynn, D.A., and Surprenant, N.F., Cadmium: Control
Strategy Analysis, GCA-TR-75-36-G, Final Report for Environmental
Protection Agency, Contract No. 68-02-1337, Task No. 2, April
1976.
3. Mineral Industries Surveys, Zinc, Lead, and Copper, Bureau of Mines
1973.
,4. Yost, K-J., et.al., The Environmental Elow...o.£L.Cadmium .and Other
Trace Metals, Source Studies (Progress Report), National Science
Foundation Grant, GI-35106, June 30, 1974.
5. Jacko, R.B. and Nuendorf, D.W., "Trace Metal Particulate Emission
Test Results From a Number of Industrial and Municipal Point
Sources," Journal of Air Pollution Control Association, 27 (10):
989-994, October 1977.
6. Sargent, D.J. and Metz, R.J., Technical and Microeconomic Analysis
of Cadmium and Its Compounds, Environmental Protection Agency,
560/ 3-75-005, June 1975.
7. Davis, W.E. and Associates, National Inventory of Sources and
Emissions: Cadmium, Nickel, and Asbestos, NAPCA-APTD-68, PB 192-
250, February 1970'.
8. Anderson, D., Emissions Factors for Trace Substances, Final
Report, Environmental Protection Agency, 450/2-13-001, PB 230-894;
December 1973.
9. U.S. Industrial Outlook, U.S. Department of Commerce, Washington,
D.C., January 1977.
10. Preferred Standards Path Report for Cadmium, Draft Document,
ESED, Environmental Protection Agency, 1972.
11. Duncan, L.G., et.al., Selected Characteristics of Hazardous Pollutant
Emissions, Mitre Corporation, Final Report for Environmental
Protection Agency Contract No. 68-01-0438, 1973.
91
-------
12. Metal Statistics, 1977, American Metal Market, Fairchild Publishing,
New York, New York, 1977.
13. Temple, Barker, and Sloane, Analysis of Economic Effects of Environ-
mental Regulations on the Integrated Demand Steel Industry,
Volumes I and II, December 1976.
14. Environmental Protection Agency Emission Test Results, Emissions
Management Branch, OAQPS, Environmental Protection Agency, Durham,
North Carolina. •
15. Yost, K.J., et.al., The Environmental Plow of Cadmium and Other Trace
Metals, Volume I, Purdue University Progress Report, July 1,
1972-June 30, 1973.
16. Arthur D. Little, Inc., Steel and the Environment: A Cost Impact
Analysis, May 1975.
17. Jacko, R.B., Nuendorf, D.W., and Faure, F., "Fractional Collection
Efficiency of Electrostatic Precipitator for Open Hearth Furnace
Trace Metal Emissions," Environmental Science and Technology, 10
(10): 1002-1005, October 1976.
18. Basis For National Emission Standards for Cadmium, Battelle-
Columbus, 1971.
19. Lucas, John, Bureau of Mines, personal communication, December
1977.
20. Coakley, Mr., Bureau of Mines, personal communication, December
1977.
21. Fulkerson, W., and Goeller, H.D., Eds., Cadmium, The Dissipated Element,
ORNL National Science Foundation, ED-21, January 1973.
22. Klein, D.H., Anderson, A.W., et.al., "Pathways of Thirty-Seven
Trace Elements Through Coal-Fired Power Plants," Environmental Science
and Technology, 9 (10): 973-979, October 1975.
23. Lee, R.E., Jr., Crist, H.L., et.al., "Concentration and Size of
Trace Metal Emissions From a Power Plant, a Steel Mill, and a
Cotton Gin," Environmental Science and Technology, 9 (7): 643-
647, July 1975.
24. Project Independence, Federal Energy Administration, Washington,
D.C., 1974.
25. Lee, R.E., Jr., and von Lehmden, D.J., "Trace Metal Pollution in
the Environment," Environmental Science and Technology, 10 (10):
1011-1017, October 1976.
92
-------
26. Fletseher, M., Sarofim, A.F., et.al.._, "Environmental Impact of
Cadmium: A Review by the Panel on Hazardous Trace Substances,"
Environmental Health Perspectives, 7:253-323, May 1974.
27. Lagerwerff, J.V., and Specht, A.W., "Contamination of Roadside
Soil and Vegetation With Cadmium, Nickel, Lead, and Zinc," Environ-
mental Science and Technology, 4:583, 1970.
28. Scientific and Technical Assessment Report on Cadmium, Environmental
Protection Agency 600/6-75-003, March 1975.
29. Junger, R.H., Lee, R.E., Jr., and von Lehmden, D.J., "The EPA
Fuel Surveillance Network: I. Trace Constituents in Gasoline
and Commercial Gasoline Fuel Additives," prepared for Environmental
Science and Technology, 1975.
30. National Functional System Mileage and Travel Summary, 1976,
Department of Transportation, Washington, D.C., 1976.
31. Determination and Evaluation of Environmental Levels of Cadmium,
Battelle-Columbus, Lab Draft Report to Environmental Protection
Agency, Office of Toxic Substances, July 13, 1977.
32. Pitcher, Charles, Bureau of Domestic Commerce, Construction and
Forest Products Division, personal communication, December 1977.
33. Cross, F.L., Jr., Drago, R.J., and Francis, H.E., Metals in Emissions
From Incinerators Burning Sewage and Mixed Refuse, NAPCA, 1969.
34. Fenton, Richard, "Present Status of Municipal Incinerators," May
1975.
35. "Municipal Sludge: What Shall We do With It?," Current Focus,
League of Women Voters, Washington, D.C., 1976.
36. Jones, Jerry, et.al., "Municipal Sludge Disposal Economics," •
Environmental Science and Technology, 11 (10), October 1977.
37. Air Pollution Primer, National Tuberculosis Respiratory Disease
Association, New York,- 1969.
'38. Shonka, D.B., Soeble, A.S., Patterson," P.O., Transportation Energy
Conservation Data Book^ Edition Two, Oak .Ridge, ORNL-5320, October
1977.
39. "Sales of Fuel Oil and Kerosene in 1975," Mineral Industry Surveys,
U.S. Department of Mines, Washington, D.C., 1976.
93
-------
40. International Directory of Mining and Mineral Operations, Engineering
and Mining Journal, McGraw-Hill, New York, New York, 1976.
41. Metal and Mineral Policy, 1973, U.S. Bureau of Mines, Washington,
D.C.
42. Suprenant, Norman, et.al., Preliminary Emissions Assessment of
Conventional Stationary Combustion Systems, Volume II, Environ-
mental Protection Agency, 600/2-76-0466, March 1976.
43. National'Highway Inventory and Performance Study, 1976, U.S. Depart-
ment of Transportation, Federal Highway Administration, OMB No. 04-
575308, July 1975.
44. National Functional System Mileage and Travel Summary, 1976, De-
partment of Transportation, Washington., D.C., 1976.
45. Bender, Ed and Readling, Charles, "Annual U.S. Energy Use Up in
1976," Department of the Interior News Release, Bureau of Mines,
March 14, 1977.
46. Market-Oriented Program Planning Study (MOPPS), U.S. Energy Research
and Development Administration, Washington, D.C., September 1977.
47. Lucas, John, Bureau of Mines, personal communication, June 1978.
48. Young, Earle F. Jr., American Iron and Steel Institute, personal
communication to Dr. J.H.B. Garner, August 1978.
49. Duce, Robert A., "Comments on EPA Cadmium Documents".
50. Benzer, W.C., American Iron and Steel Institute, personal
communication to R. Coleman, EEA, December 1978.
51. Katari, V., Isaacs, G. and Devitt, T.W., Trace Pollutant Emissions
From Processing Metallic Ores (Final Report), Environmental Pro-
tection Agency, 650/2-74-115, PB 2.38-655, October 1974.
94
-------
6. SCREENING OF CADMIUM SOURCE TYPES
6.1 INTRODUCTION
The ambient concentrations produced by the various sources of cadmium
emissions were estimated very conservatively in order to determine which
source types could potentially produce annual averages of cadmium
greater than 0.1 ng/m3 in the ambient air. It should be emphasized
that this concentration was selected as significant for analysis only,
since it implies nothing about possible health effects. The concentration
is the lowest which can be consistently and accurately detected on an
annual basis. Thus, the conservative preliminary screening of sources
is designed to identify all possible cadmium emitters, which is not to
say that all of these will be deemed significant from a health standpoint
in subsequent analysis.
The average, "typical," and/or maximum plant capacity or production
rates for each source type were collected from the industrial literature.
For area sources, an area of emission was taken from the literature or
estimated. For point sources, the stack characteristics, in terms of
.ranges or "typical" values, were compiled when available. The stack
characteristics required were flow rate per -production rate, stack
temperature, and stack height. The primary references for this data
were:
• Vandegrift, A.E., Shannon, L.J.; et.al., Handbook of
Emissions, Effluents, and Control Practices for
Stationary Particulate Pollutant Sources, Report
NAPCA Contract No. CPA-22-69-104, November 1970.
for
Cadmium:
Deane, G.L.; Lynn, D.A., and Suprenant, N.F.,
Control Strategy Analysis, GCA-TR-75-56-G, Final
Report for EPA Contract No. 68-02-1331, Task No. 2,
April 1976.
'95 .
-------
• Katari, V.; Isaacs, G., and Devitt, T.W., Trace
Pollutant Emissions From Processing Metallic Ores
(Final Report), EPA-650/2-74-115, PB 238 655,
October 1974.
• Environmental Protection Agency Emission Test
Results, Environmental Measurement Branch,
OAQPS, Environmental Protection Agency, Durham,
North Carolina.
• Arthur D. Little, Inc., Steel and the Environ-
ment: A Cost Impact Analysis, a report to the
Iron and Steel Institute, May 1975.
Often little, if any, specific information was available for a given
plant characteristic of a source type, so that an estimate had to be
made. Such estimates were based on available information about the
process, control equipment, and standard industrial practice.
The information on plant sizes, in terms of capacity of production
rate, was extracted from the following sources:
• Deane, G.L.; Lynn, D.A., and Suprenant, N.F.,
Cadmium: Control Strategy Analysis, GCA-TR-
75-36-G, Final Rex. 'rt for EPA Contract No. 68-
02-1337, Task No. 2, April 1976.
• International Directory of Mining and Mineral
Operations, Engineering and Mining Journal,
McGraw-Hill, .New York, New York, 1976.
• Arthur D. Little, Inc., Steel and the Environ-
ment: A Cost Impact Analysis, a report to the
American Iron and Steel Institute, May 1975.
• Metal Statistics 1977, American Metal Market,
Fairchild Publications, New York, New York, 1977.
• Sargent, D.J. and Metz, J.R., Technical and
Microeconomic Analysis of Cadmium and Its
Compounds, EPA-560/3-75-005, June 1975.~
• Fenton, R., "Present Status of Municipal Incin-
erators," Incinerator and Solid Waste Technology,
J.W. Stephanson, et.al., Eds. ASME, New York,
New York, 1975.
96
-------
• Jones, J.L. , et.al., "Municipal Sludge Disposal
Economics," Environmental Science and Technology,
October 1977.
• Weinstein, N.J., Waste Oil Recycling and Disposal,
EPS-670/2-74-052, August 1974.
Most of the preliminary formation, such as industry production figures,
and some detailed information, such as the size of units for primary
smelting and some of the miscellaneous sources, were taken from the first
reference. The remaining references were the sources of maximum or "typical"
plant size information for mining and primary smelting, iron and steel
plants, secondary smelting, manufacturing using cadmium, municipal in-
cinerators, sewage sludge incinerators, and lubricating oil incinerators,
respectively.
Fossil fuel consumption was estimated for a typical plant size, in the
case of power plants, and derived from the following government publi-
cations for fuel oil and gasoline:
• "Sales of Fuel Oil and Kerosene in 1975," Mineral
Industry Surveys, U.S. Department of Mines, Washington,
D.C., 1976.
• National Functional System Mileage and Travel Summary
from the 1976 National Highway Inventory and- Performance
Study, U.S. Department of Transportation, Federal High-
way Administration, June 1977.
The first document gives fuel consumption" by state for residual, distillate,
and diesel fuel. The diesel fuel consumption is further broken down by
on-highway and off-highway uses. The second report gives the density of
vehicle miles traveled in the urbanized area for the major metropolitan
areas and the states.
The compiled plant_characteristics for the various source types were then
used to make' "realistic," but very conservative estimates of the effect of
a "typical" or maximum size plant, for each plant type, on the concen-
tration of cadmium in the ambient air. The effects of the area
97
-------
sources were estimated for screening purposes by using the Hannah-
Gifford urban air pollution model, assuming a very conservative (es-
pecially for an annual average) wind speed of two m/s, or the Miller
Holzworth urban model for cities for which it had been calibrated. The
effects of individual plants of the various point source types were
estimated for screening purposes by using EPA's PTMAX dispersion model,
using generally conservative emission rates, stack, heights, temperatures,
and flow rates. Using results from the PTMAX model is, conservative
since the estimates from the model are maximum hourly ambient concentra-
tions, while the ambient concentration of interest is the annual average.
An annual average is generally a factor of three to four lower than a
24-hour average concentration, which is, in turn, generally a factor of
three to four lower than an hourly, average. In addition, the maximum
concentration considered was the maximum for any wind speed and stability
conditions. Thus, conditions which might occur for a short time, but
which are unlikely to represent the annual average meteorological con-
ditions, and which would occur very near to the plant, are often used to
conservatively reprer nt the worst realistic case. The emission rates
(the products of the emission factors and production capacities or
rates, expressed in grams per second) used were estimated assuming that a
plant operated only 220 days per year and eight hours per day. Since
most industrial facilities operate with a much higher capacity utiliza-
tion, this greatly overstates the emission rate per hour. If available,
both the maximum and a "typical" plant size were considered for both the
maximum and best judgement estimates of cadmium emission factors in
order to assess the likelihood of the estimated concentrations. For
some source types, the two different estimates of plant size or of
emission factors were nearly equal, so only the more conservative case •
was calculated. The stack characteristics for which PTMAX was run were
generally chosen to be the representative for the plant type which would
generate the highest predicted maximum ambient concentration for a given
emission rate (of one g/s, i.e., low flow rate, low stack height, and
O
-------
low stack temperature). The following sections briefly outline the
estimated conditions and the results of the screening of the various
cadmium source types. Due to the very conservative assumptions discussed
above, a source which may appear to produce an hourly concentration
greater than 0.1 ng/m may be classified as a non-measurable source.
The assumed stack and operating characteristics are sometimes so con-
servative as to produce an unrealistically high estimate of concentration.
When this occurs for a particular source class it is noted in the text.
6.2 MINING
The cadmium concentrations resulting from the mining of cadmium-bearing
ores were very conservatively estimated using the area source approach.
For the largest zinc mine, it was assumed that the concentration of zinc
in the ore was 2.6 percent and that the area of the mine was one square
mile. For a wind speed of 2 m/s at these conditions, the Hannah-Gifford
model estimates a concentration of 16 ng/m . However, the Hannah-
Gifford model was developed for urban areas by assuming a series of line
sources which is not the case for an isolated source such as mines. A
wind speed of 5 m/s is generally more typical; the average mine produces
500,000 tons/ year, and a mine's property, if not its active area, is
usually at least ten square miles. Therefore, it was presumed that a
more realistic estimate of ambient concentration beyond the mine property
would be at least two orders of magnitude lower and that even the largest
mine would generate concentrations that were at most marginally greater.
3 ' ' •
than 0.1 ng/m . Since most zinc mines are underground mines, emissions
would be expected to be much lower.
The analyses and conclusions were similar for the other types of mining
which handle significant amounts of cadmium. The largest lead mine pro-
duces 1.6 million tons/year and the lead concentration in ore is esti-
mated at 1.6 percent, so the ambient concentration for a one mile square
3
area source would be 1.6 ng/m . Since most lead mines are underground
99
-------
and the average production is only 350,000 tons/year, it is even more
unlikely that measurable concentrations would be generated. Copper mines
generally are surface mines producing as much as 19.6 million tons of
ore per year. Assuming the maximum cadmium concentration in U.S. ores
of 0.6 percent, and a rather conservative working area of ten square
f «7
miles, an ambient concentration of 0.08 ng/m is predicted. Obviously
this level is insignificant for all other mine sizes.
6.3 PRIMARY METALS
The primary metal smelters were treated as point sources using very
conservative plant and stack characteristics for screening purposes.
The largest zinc smelter produces 250,000 tons/year, so this was used to
calculate an emission rate for the various processes and the overall
plant using the maximum emission factor. Assuming stack characteristics
ranging from a 10 m stack emitting about 20 m3/s at ambient temperature
for coke ovens, to a 120 m stack emitting 235 m3/s at 370 60 615°K for
most of the processes, the range of estimated maximum concentrations
ranged from 96 ng/m3 to about 60 ng/m3 for a horizontal retort and the
overall process, respectively. Most of the individual processes would
produce maximum ambient (hourly average) concentrations on the order of
ten ng/m , so that even for an average plant size of 100,000 tons/year,
it is unlikely that concentrations lower tha.n the minimum detectable
level (0.1 ng/m ) would be produced. This is so even for the "best
judgement" emission factors, as they are no more than a factor of two
smaller than the maximum estimate emission factor.
Similarly, the largest primary lead smelter produces 350,000 tons/year.
Therefore, assuming a 60 m stack with a flow rate of 160 m3/s at 340°K
for the overall smelter, and a flow rate of 4.8 m3/s at 330°K for a
baghouse controlled smelter, the ambient concentrations produced by the
two model'plants would be 8,700 ng/m3 and 3,580 ng/m3, respectively.
Again, the average plant size is 52,000 tons/year and the "best judge-
100
-------
ment" emission factor is less than a factor of three lower, so it is
unlikely that concentrations less than 0.1 ng/m3 would be produced. As
the stack characteristics are similar for copper and cadmium smelters,
with the production rates in the area of 200,000 tons/year the emission
factors are at least as high, there is no question that these facilities
also produce ambient concentrations far above the measurable level.
6.4 IRON AND STEEL
The individual iron and steel processes were screened using representa-
tive sizes for that process, while the overall plant was screened using
the maximum production rate of three million tons of steel per year.
All of the stacks were approximated as 40 m with temperatures of 310-
340 K. The estimated ..concentrations range from 8.3 ng/m3..for a 13,055
tons/day controlled basic oxygen furnace (using the "best judgement"
emission factor and 95 m3/s at 340°K), 450 ng/m3 for a 6,504 tons/day
uncontrolled open hearth (using the maximum emission factor and the same
stack conditions). The overall plant estimate (using a flow rate of 220
m /s and 310°K) ranged from 16 ng/m3 to 13 ng/m3 using the maximum or
best judgement emission factor. For the other processes (3,787 ton/day
sinter strand, 1,440 ton/day blast furnace, and a 1,344 ton/day electric
arc) the estimated concentrations were generally on the order of 100-
1000 ng/m . Since the controlled estimates for the conservative screening
technique were as low as ten's of nanograms per cubic meter, it was
possible that some plants, particularly small ones, might prove to
produce concentrations less than 0.1 ng/m3. In the later analysis it
was found that some plants did produce.very low, but generally measurable
concentrations, but that iron and steel plants with sinter strands,
which are difficult to control efficiently, or with very large capacity,
are estimated to produce maximum annual average concentrations on the
order of hundreds of. nanograms per cubic meter.
6.5 SECONDARY SMELTING
The uncontrolled emissions of secondary zinc and copper smelters were
found to produce ambient cadmium levels greater than 0.1 ng/m3, while
101
-------
the controlled emission factors for secondary lead smelters were found
to produce at most, barely measurable concentrations even for the con-
servative screening procedure. The 45,000 ton/year maximum size zinc
plant, with an assumed stack of 40 m emitting 7.5 m /s at 340°K, would
2
generate a maximum concentration of 900 ng/m . A secondary copper
smelter producing 52 tons in a seven-hour-day and using a 20 m stack
with a flow rate of 17.5 m /s at 370°K would generate concentrations
which are another order of magnitude higher. Secondary lead plants,
with a "typical" production rate of 2,500 Ib/hr and high efficiency
control equipment, were estimated to produce ambient concentrations less
than 6 ng/m /s at 340°K. Using the "best judgement" emission factors,
the highest maximum concentration generated by PTMAX, even for these
very conservative stack characteristics, was less than 6"ng/m . There-
fore, since the range of sizes of secondary smelters is generally small,
and the PTMAX estimates are hourly rather than annual averages, secondary
lead smelters were eliminated from further consideration as detectable
source of emissions.
6.6 MANUFACTURING
Very little information could be found about individual plants which
manufacture products containing cadmium. Using the GCA estimates of the
total production, the number of known plants, and the very conservative
mass balance or survey estimates of emission factors, concentrations in
the microgram per cubic meter range were estimated for very conservative
stack conditions (20 m and 0.90 m3/s at 340°K). Since there are pro-
bably many more smaller plants with increased control efficiency and
larger stacks, it was concluded that the emissions of individual manu-
facturing plants are overestimated and would not produce measurable
annual average concentrations of cadmium.
102
-------
6.7 FOSSIL FUEL COMBUSTION
The only point sources considered that burns fossil fuel were power
plants. Assuming a 300 MW(e) capacity (with ten million Btu/hr power
MW(e) and 82 percent boiler efficiency, and a 130 m stack with a flow of
285 m3/s at 440°K) an uncontrolled coal-fired power plant would produce
ambient concentrations of 5 ng/m with the "best judgement" emission
3
factor and approximately 700 ng/m with the maximum emission factor.
Assuming a 60 m stack with a flow of 140 m3/s at 440°K, the controlled
coal- and oil-fired power plants were estimated to produce concentra-
3 3
tions ranging from 0.8 ng/m to 40 ng/m for best judgement and maximum
estimate emission factors. Thus, for more realistic conditions, some
power plants were generally thought to produce concentrations somewhat
greater-than0^1 ng/m3. Therefore,- an annual-average CRSTER run ..(using—
Dallas/Fort Worth meteorological conditions and a very conservative 40 m
stack with a flow of 105 m3/s at 1,360°K) was used to determine the
critical emission rate which could cause a maximum annual average ambient
3
concentration greater than 0.1 ng/m for each fuel type. The emission
factors for each fuel type were then assumed to follow a log-normal
distribution (with the probability of being exceeded at 90 percent for
the minimum, 50 percent for the "best judgement," and ten percent for
the maximum emission factor), and were multiplied by the capacities in
the Energy Data System (EDS) file to calculate the' emission rates for
uncontrolled and controlled power plants (assuming a 24-hour operation).
Plotting these emission rates on log-probability paper showed that only
three plants had a greater than ten percent probability of exceeding the
measurable ambient cadmium concentrations. These three plants were the
three largest listed in the EDS with no control equipment. Since all
power plants have some particulate control equipment (generally of
greater than 90 percent efficiency), individual power plants were elimi-
nated as detectable sources of cadmium.
103
-------
The other fossil fuel combustion sources of heating oil, diesel oil, and
gasoline were treated as area sources. F,or heating oil, the total
amount of distillate (including off-highway diesel) and residual used in
New York State in 1975 (the highest state consumption in the nation:
159 million barrels) was assumed to be burned in the urbanized area of
Metropolitan New York City within New York State (1,634 mi2) during a
three-month period (24 hours per day). The ambient concentration, using
the maximum emission factor and the Hannah-Gifford model with a wind
speed of 2.0 m/s, was only 0.09 ng/m3, so heating oil was eliminated as
an individually-significant source. Similarly, the on-highway diesel
fuel consumption of California (17.9 million barrels) was assumed to be
used in Metropolitan Los Angeles (1,724 mi2) over the entire year.
Using the maximum emission factor, the concentration generated by the
Hannah-Gifford model for a 2.0 m/s wind is 0.0012 ng/m3. The Miller-
Holzworth model for Los Angeles (city size 60 km) confirms this with a
concentration of 0.00059 ng/m3 for a mixing height of 300 m and one of
0.0031 ng/m or a mixing height of 100 m and a wind speed of 1.0 m/s.
The Miller-Holzworth model for Los Angeles was also used for gasoline
consumption. The dai? - vehicle miles traveled per square mile (DVMT
density) of 61.342 and an assumed mixing height of 300 m with a wind
speed of 2.0 m/s generated an estimated 0.3 ng/m3 for the maximum and
0.15 ng/m for the "best judgement" emission factor. For the.. Washington,
D.C., metropolitan area, the area with second highest DVMT density, the
estimates are 0.17 ng/m3 and 0.08 ng/m3. Therefore, because of the
accuracy of the estimates and the conservative meteorological assumptions
for the areas with the most usage,, gasoline consumption was concluded
not to be an individually-measurable source of ambient cadmium.
6.8 MISCELLANEOUS
The miscellaneous sources of cadmium emissions.are motor oil consump-
tion, rubber tire wear, fungicides, and fertilizers. With the exception
of fungicides, for which no information or application rates were
104
-------
available and which may be banned in the near future, these sources were
treated by the area source methodology. The same analysis was used for
motor oil consumption and rubber tire wear as was used for gasoline
(i.e., the Miller-Holzworth model using DVMT density). For motor oil
consumption, the calculated ambient concentrations are 0.013 ng/m for
Los Angeles and 0.008 ng/m for Washington, B.C. For rubber tire wear,
3 ?
the ambient concentrations are 0.04 ng/m and 0.-02 ng/m using the "best
judgement" emission factor for the two cities, respectively. Thus, the
individual automotive sources of cadmium do not produce significant
ambient levels of cadmium, even in the cities with the highest DVMT
fj
density. When a high fertilizer application rate of 20 g/m -year was
assumed, the ambient concentration using the Hannah-Gifford urban area
"model with aT wind speed of 2".'6""m/s" was "calculated as 1.8 and 0.2 ng/m3,
with the maximum and "best judgement" emission factors, respectively.
As the maximum emission factor assumes that all the cadmium in the
fertilizer is emitted into the air, and a high fertilizer application
rate and conservative dispersion model for rural locations was used, it
was concluded that fertilizers would not produce greater than 0.1 ng/m3.
Cement plants are another source of cadmium emissions that was found to
produce very low ambient levels of cadmium when very conservatively
modeled as a point source. Assuming the emissions from the entire
cement production of the Lehigh Valley of Pennsylvania, 31 million
Ibs/year, came out of one stack (60 m with a flow of 115 m3/s at 340°K) ,
the maximum ambient (hourly average) concentration would only be 0.8
ng/m . With the same production rate, and even more conservative stack
characteristics (as low as 20 m with a flow of 16 m3/s and 290°K) , only
a few of the individual processes would produce hourly average concen-
trations above 0.1 ng/m . Therefore, for any individual cement plant,
it was concluded that annual average ambient concentrations would be
well below measurable levels.
-------
6.9 INCINERATION
Municipal, sewage sludge, and lube oil incinerators, were screened as
cadmium sources by using PTMAX and a point source treatment. For a
municipal incineration with a 35 m stack, a flow of 9.4 m /s at 310 K,
and the maximum or "best judgement" emission factor, the maximum and
average capacities of 1,600 and 300 tons pex 24-hour day are estimated
to generate ambient concentrations on the order of micrograms per cubic
meter. Since the control equipment on incinerators is generally not of
high efficiency, municipal incinerators were considered to produce
measurable annual concentrations of cadmium.
The maximum capacity sewage sludge incinerator of 7.5 tons per hour,
which operates only three days a week, would generate an ambient concen-
tration of less than 3.0 ng/m (maximum hour) even for a 20 m stack with
a flow of 2.5 m3/s at 310°K. With a 35 m stack and the same flow rate
the maximum hourly average concentration is 1.7 ng/m for the controlled
3
multiple hearth maximum emission factor for the same process (0.12 ng/m
for the controlled fluidized bed). It was therefore concluded that with
more realistic assump. ons the annual average emissions would not produce
average ambient concentrations of cadmium greater than 0.1 ng/m .
* ,
The only information that was available on lubricating oil incinerators
was that the total amount incinerated by a multitude of small sources
was estimated to be 389 million gallons per year. Assuming that the
uncontrolled emissions from incinerating all the lubricating oil in the
nation came out of a 20 m stack with a flow rate of 0.9 m /s at 340 K,
an hourly ambient concentration on the order of micrograms per cubic
meter is estimated by PTMAX. Since there are probably thousands of such
incinerators in the country, it is presumed that the annual average
ambient cadmium concentrations generated by any one of them would be
below measurable levels.
106
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-450/5-79-006
3. RECIPIENT'S ACCESSION'NO.
4. TITLE AND SUBTITLE
Sources of Atmospheric Cadmium
5. REPORT DATE
August 1979
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Robert Coleman, et al.
8. PERFORMING ORGANIZATION REPORT NO.
3. PERFORMING ORGANIZATION NAME AND ADDRESS
Energy and Environmental Analysis, Inc.
1111 North 19th Street
Arlington, Va. 22209
1O. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-02-2836
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Air Quality Planning and Standards
Pollutant Strategies Branch
Research Triangle Park, N.C. 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
EPA
15. SUPPLEMENTARY NOTES
Task Officer was Richard Johnson, OAQPS/SASD, MD-12
16. ABSTRACT
This report is one of a series of reports which will be used by EPA in responding
to the Congressional request under section 122 of the Clean Air Act Amendments
of 1977 to determine whether atmospheric emissions of cadmium pose any threat
to public health. This report surveys the uses of cadmium and potential emission
sources to determine which sources.are the most significant both in terms of total
emissions and potential ambient levels.
It is estimated that about 850 tons of cadmium were emitted during 1974, with the
largest estimated emitted of cadmium being the production of zinc. Other sources
identified were incinerators, iron and steel mills, fossil fuel combustion, smelters.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS C. COS AT I Field/Group
Air Pollution, Air Quality Data Emission
Inventory, Emission Factors, Control
Technology Cadmium
Atmospheric Emissions
Air Pollution Control
13. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (ThisReport)
Unclassified
21. NO. OF PAGES
113
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
-------
------- |