GCA-TR-75-36-G
CADMIUM:
CONTROL STRATEGY ANALYSIS
FINAL REPORT
by
Gordon L. Deane
David A. Lynn
Norman F. Surprenant
GCA CORPORATION
GCA/TECHNOLOGY DIVISION
Bedford, Massachusetts
April 1976
Contract No. 68-02-1337
Task Order No. 2
EPA Project Officer
Justice Manning
U.S. ENVIRONMENTAL PROTECTION AGENCY
Research Triangle Park
North Carolina
-------
This Final Report v;as furnished to the Environmental" Protection Agency by GCA
Corporation, CCA/Technology Division, Bedford, Massachusetts 01730, in fulfill-
ment of Contract No. 68-02-1337, Task Order No. 2. The opinions, findings, and
conclusions expressed are those of the authors and not necessarily those of the
Environmental Protection Agency or of the cooperating agencies. Mention of
company or product narces is not to be considered as an endorsement by the En-j
vironmental Protection Agency.
-------
CONTENTS
List of Figures iv
List of Tables v
Executive Summary vii
Sections
I Summary and Recommendations 1
II Cadmium: Its Properties, Measurement, Production and
Uses 10
III Sources and Control of Airborne Cadmium 36
IV Health Effects of Cadmium 77
V • Cadmium in Environmental Media 104
VI Exposure to Cadmium in the Environment " 138
iii
-------
FIGURES
No. Page
1 Periodic Table of the Elements 12
2 Vapor Pressure of Cadmium and Zinc Versus Temperature 14
3 Sodium Chloride Interference in Cadmium Analysis 19
*
4 Domestic Cadmium Production and Consumption 22
5 Domestic Zinc Production and Consumption . 23
6 Domestic Primary Zinc Smelters With Secondary Cadmium
Plant, 1975 '. 24
7 Process Cadmium Emission Summary, Tons Per Year 39
8 Schematic Flowsheet for Recovery of Zinc and Preliminary
Recovery of Cadmium From Ore by Electrolysis (F = Fabric
Filter or Other Emission Control Device) . 42
9 Schematic Flowsheet for Recovery of Zinc and Preliminary
Recovery of Cadmium From Ore by Electrolysis 44
10 Schematic Flowsheet of Copper Processing 60
11 Schematic Flowsheet for Recovery of Cadmium From Zinc and
Lead Recovery Waste Materials 63
12 Renal Cortex Levels of Cadmium Found in Human Beings
(After CITE) 81
13 Distribution of Lead and Cadmium in Human.Tissues
Under Normal Exposure " 83
14 The Solubility of Cadmium Versus pH and Versus Total
Carbonate Concentration (After ORNL)
iv
-------
TABLES
No. Page
1 Physical Properties of Cadmium 13
2 Comparative Melting and Boiling Points • 13
3 Crystal Structures of Various Compounds 16
4 Consumption and Production Statistics for Cadmium and Zinc 21
5 Domestic Slab Zinc Capacity 1968-1975 25
6 Estimated Usage of Cadmium, Tons 26
7 U.S. Demand Statistics for Cadmium Data in Tons
Per Year (Elemental Cadmium) 31
8 Supply Statistics for Cadmium Data in Tons Per Year 33
9 Cadmium Emission Estimates 37
10 Qadmium Emissions From Mining 40
11 Existing U.S. Zinc Processing Plants, 1974 45
12 Existing U.S. Lead Smelting, 1973 51
13 Existing U.S. Copper Smelters, 1973 56
14 Adverse Effects of Cadmium on Experimental Animals
(from ORNL) 85
15 Adverse Effects of Cadmium on Humans (from ORNL) 86
»
16 Estimated Minimum Cadmium Levels Necessary for Reaching
200 ppm (Wet Weight) of Cadmium in Renal Cortex (After
CITE) 88
17 Total Dustfall, Cadmium, and Zinc From 77 U.S. Cities 106
18 Estimated Emissions of Cadmium to Water by Different Sources 115
-------
TABLES (continued)
No.
19 Number of Years to Increase the Cadmium Content of the
Soil by 0.1 ppm 122
"20 Cadmium Content and Zinc/Cadmium Ratio of Some Foods 126
21 Cadmium Content in Different Food Categories in U.S.A0'
Total Number of Samples: 30 127
22 Cadmium Content in Plants Grown in Long-Term Phosphorous
Fertility Plots ' 130
23 Primary Intermedia Contributions of Cadmium 133
24 Cadmium Intake and Retention From Ambient Air * 139
25 Cadmium Intake and Retention From Water 142
26 Cadmium Intake and Retention From Food . ; 144
27 Cadmium Intake and Retention From Cigarettes 146
28 Total Daily Cadmium Retention Via Major Routes for
Representative Retention Levels 148
29 Media Contributions to Normal Retention 150
30 Total Contribution of Air Emissions to Normal Daily
Retention of Cadmium, Nonsmoking 152
31 Total Contribution of Air Emissions to Normal Daily
Retention of Cadmium, 1 pack/day 153
32 Total Contribution of Air Emissions to Normal Daily
Retention of Cadmium, 3 packs/day 154
vi
-------
EXECUTIVE SUMMARY
This document was prepared for the Standards and Cost Analysis Branch,
Strategies and Air Standards Division of the Office of Air Quality Planning
and Standards to assess cadmium as an air pollutant from the viewpoint of
the Branch's responsibility for recommending potential standard-setting or
other control strategy decisions. The study relied primarily on secondary
data compilations that were readily available though attempts were made to
supply missing data and update the figures given in the earlier reports.
RECOMMENDATIONS
No specific recommendations are made as to the control approach or specific
strategies that should be adopted. Rather, the point is made that further
study should be immediately pursued to prepare the Branch for the impending
need to make decisions on cadmium. Various topics of research are sug-
gested to help in this process. These topics are broken out into two areas
--those relating directly to the quantification of air emissions and the
placement of these emissions in the proper perspective with respect to
other exposures; and those that pertain to the understanding of the effects
of cadmium in the biosphere and human body.
The former set of research topics include analyses of individual sources
of cadmium, cadmium content in ores and process feed, agricultural and
fossil fuel sources, stack sampling for cadmium, modeling of emissions,
and modeling of cadmium in the biosphere over time. Research suggestions
on the effects of cadmium include the assessment of the cadmium-
hypertension relationship, the possible effects on the kidney at lower
exposures, cadmium in the diet, and bioaccumulation.
vii
-------
PROPERTIES, MEASUREMENT, PRODUCTION, AND USES OF CADMIUM '
Aside from the properties that make cadmium a useful material in various
ways, its most salient property is its high volatility.in comparison to
other metals. This is a basic feature in the way it is produced, and is
an important factor in its emission and controls
The analytical measurement- of cadmium presents a number of difficulties
when levels are low; however, it appears that these difficulties are not
much different from the problems of measuring any trace substance and, in
any event, they are well-known and being actively attacked.
*
The outlook for production and use of cadmium is of serious concern in
approaching long-term control strategies and assessing long-term health
effects* Cadmium has many desirable properties and consequently a number
of varied uses. In none of these uses, however, is it so unique as to
preclude substitution of another material, and most uses do utilize sub-
stitutes on occasion, as the price of cadmium fluctuates. If projected
growth trends for the various cadmium uses are totaled, it is possible
to project a very dramatic increase in cadmium consumption, one which
would in fact deplete known reserves by the year 2000. For production
and economic reasons, however, it is judged that this will not occur.
Cadmium is produced primarily as a by-product in the production of zinc
and other metals. Consequently, the supply of cadmium is not responsive
to the demand for cadmium, but rather to the demand for and production of
zinc. For a number of economic reasons, including anticipation of environ-
mental controls, domestic zinc (and hence cadmium) production has declined
over recent years. Cadmium imports have helped! keep the supply roughly
constant, but are not expected to increase.
viii
-------
The only specific information need that appeared to be inadequately met
was the precise quantitative figures on cadmium flow as cadmium-bearing
ore concentrates and flue dusts are shipped, imported, and processed.
Cadmium production fluctuates fairly widely and, apparently because of
its by-product nature, cadmium concentration analyses and quantities of
materials moved are less well-documented than would be expected.
SOURCES AND CONTROL OF AIRBORNE CADMIUM
•
Estimates of the emissions of cadmium from different sources are made by
reviewing calculations of previous invesitgators and applying more current
data on process flow. The total emissions in 1974 were estimated to be less
than 1688 tons with over half of that being the result of primary zinc pro-
cessing and the reprocessing of iron and scrap steel. Many of the emission
estimates are based on rather meager data.
The control of cadmium emissions has only been through the control of total
particulate emissions. As sources have been coming under compliance for
total particulate emissions, cadmium emissions are also expected to be
,•
declining. However, the nature of the cadmium fune requires control tech-
nology which has high efficiency for very small particle sizes, such as
fabric filtration. Since the control of particulates will be increasingly
more effective under attainment and maintenance plans and new source per-
formance standards, the emissions of cadmium are not expected to increase
substantially over time. The greatest immediate decrease in cadmium emis-
sions occurs in the primary zinc category due to the closing of a major
plant in Texas. •
HEALTH EFFECTS
Cadmium is a very toxic metal. In acute exposures, it is much more toxic
than lead and less toxic than mercury. It has, however, a biological half-
time much longer than mercury, so that it is clearly the most hazardous of
the three in comparable-dosage, chronic exposure situations. With respect
ix
-------
to long-term exposures, it is much more widely dispersed in our society
than is mercury and much less than lead. This is not necessarily a mean-
ingful comparison, however, as the nature of the effects and the circum-
stances of exposure to the. three metals are not really comparable.
.Cadmium is absorbed by the body from either the gastro-intestinal tract
or from the lungs, with roughly 6 percent and 25 percent ultimate retention,
respectively. These figures are not rigorously determined, and since food,
air, and smoking all contribute significantly to the body burden, better
estimates are needed for calculation purposes.
Current exposure of the population of the United States to cadmium does
not imply tissue levels which will produce renal tubular failure, the
best documented effect due to relatively low levels. However, this is
not an effect which is beyond the reach of those individuals who may be
*
exposed to locally high levels, whether from air, soil, food,'water, or
cigarette smoking. In addition, some studies have indicated that other
types of kidney damage may actually begin to occur at body burdens less
than half the level that produces tubular failure.
Although evidence has not conclusively shown that airborne cadmium is a
factor in hypertension, enough studies have demonstrated the probability
of this causal relationship to preclude ignoring the possiblity. Much
more extensive and carefully planned toxicological and epidemiological
studies are necessary to prove or disprove this relationship and to deter-
mine does-response interactions should it occur„
ENVIRONMENTAL EXPOSURE
0
Exposure to cadmium occurs through all intake media: food, smoking, ambient
air, and drinking water, in decreasing order. Respiratory intake as a re-
sult of normal ambient air levels of cadmium, about 0.03 jzg/nr*, accounts
for only a few percent of the normal daily retention of cadmium; cigarettf|
smoking contributes a much larger amount of cadmium, 1.41 /ig/pack, and can
-------
be over 56 percent of the daily retention if three packs per day are
smoked. Gastrointestinal retention rates are much lower than those found
in the lungs, 6 percent versus 25 percent (64 percent for cigarette smoke),
but the total intake levels are much higher. Food, providing an estimated
intake of 50 pg/day, will account for over 90 percent of the daily reten-
tion for nonsmokers. Two studies of U.S. dietary levels of cadmium have
indicated that intake is twice this level and therefore at a level which
is likely to produce renal .tubular failure after a period of time. Water,
at normal concentrations around 1 ppb, adds another few percent. Total
environmental exposure, under normal conditions, appears to provide about
50 percent of the daily retention rate (6.6 jig/day) deemed to be signifi-
cant because of known effects; if the exposure includes three packs of
cigarettes per day, the significant level is surpassed.
Despite the apparently small contribution of ambient air to daily retention
levels, there is serious concern over air quality when one considers the
extent to which the deposition of airborne cadmium might indirectly con-
tribute to the total exposure through water, food and tobacco. Unfortu-
nately, sufficient data in this area has not yet been accumulated, so that
only estimates of ranges of contributions can be given. It is. concluded
that the contribution of airborne cadmium to levels of cadmium in the
wa'ter may be around 10 percent, but that in food and tobacco, the contri-
bution may average as high as 64 percent. Given the "best-case" and
"worst-case" estimates, cadir.jLutn in the air was found to contribute from
less than 12 percent to nearly 80 percent of the daily cadmium retention.
xi
-------
SECTION I
SUMMARY AND RECOMMENDATIONS
This report has been prepared under contract to the Standards and Cost
Analysis Branch, Strategies and Air Standards Division," of EPA's Office
of Air Quality Planning and Standards as one input in the assessment of
the hazards and need for control of cadmium as an air pollutant. As such
it provides updated information on the emissions of cadmium to the air,
reports on the health hazards of cadmium, discusses the total intake
and retention associated with different exposure levels to various
environmental media, and analyzes the total contribution of air emis-
sions to the body burden. Recommendations regarding the control of
cadmium and further research programs are made.
}
»
NATURE OF STUDY EFFORT
For reasons of time and resource availability, this assessment has been
made relying primarily on several comprehensive studies of cadmium that
had been previously compiled. The effort was directed at the comparison
of these studies, resolving discrepancies and updating the information
provided therein through relevant primary sources. Special attention
was paid to the variations in cadmium emissions due to production,
consumption, process, and control changes. As the objective of this
study was not to generate new data, where specijfic information was not
available or could not be identified and analyzed within the frame of
the study, assumptions were made rather than leaving areas uncovered.
These assumptions have been clearly identified as such and the rationale
behind them explained.
-------
Those reports that were used extensively in this study are given
below. As they are frequently referred to throughout this document,
special brief annotations were applied to them to facilitate referencing.;
these are given in front of the appropriate bibliographic information
listed below.
CITE
Friberg, L., M. Piscator, and G., Nordberg. Cadmium
in the Environment. Cleveland, CRC Press, 1971.
CITE II Friberg, L., et al. Cadmium in the Environment, II.
Office of Research and Monitoring, Environmental
Protection Agency, Washington, D.C. Publication
Number EPA-R2-73-190. February 1973.
Conclusions from the above investigators are sometimes referenced as
Friberg, Friberg et al., or the Karolinska Institute.
ORNL
PSP
Davis
Purdue
MITRE
TGOMA
Fulkerson, W., and H.E. Goeller, (eds.). Cadmium - The
Dissipated Element. Oak Ridge, Tennessee, Oak Ridge
National Laboratory, 1973.
Preferred Standards Path Report for Cadmium. Draft
Document, ESED, Environmental Protection" Agency.
1972.
Davis, W.E., and Associates. National Inventory of
Sources and Emissions: Cadmium, Nickel, and Asbestos-0
1968. In: Cadmium, Section I. 1970.
Yost, K.J., et al. The Environmental Flow of Cadmium
and Other Trace Metals. Volume I. Purdue University.
Progress Report. July 1, 1972 to June 30, 1973.
Duncan, L.J., et al. Selected Characteristics of
Hazardous Pollutant Emissions. MITRE Corporation,
1973.
Task Group on Metal Accumulation. Accumulation of
Toxic Metals with Special Reference to Their Absorp-
tion Excretion and Biological Half-Times. In: En-
vironmental Physiology and Biochemistry, Volume III.
1973. p. 65-107.
Chizhikov Chizhikov, D.M. Cadmium. Translated by D.E. Haylerc
Oxford, Pergamon Press, 1966.
In addition to the above studies which were used extensively in the de-
velopment of this document, the review and update conducted subsequent
to the initial submittal of this report to EPA allowed time for.the
-------
identification and review of the following studies. Neither of these
studies were found to provide new information nor substantially different
conclusions.
Fleischer, M. et al. Environmental Impact of Cadmium: A Review
by the Panel on Hazardous Trace Substances. Environmental Health
Perspectives. Publication Number 7:253-323. May 1974.
Scientific and Technical Assessment Report on Cadmium. National
Environmental Research Center. U.S. Environmental Protection Agency.
Publication Number EPA*-600/6-75-003. March 1975.
One study that was identified in the update as presenting substantially
different values was a report out of the Office of Toxic Substances (OTS) .
The OTS report, given below, had significantly lower emission values than
those in this document; however, insufficient time was available for its
review and assessment.
Sargent, P. H. and J. R. Metz. Technical and Microeconomic Analysis
of Cadmium and Its Compounds. Office of Toxic Substances, U.S.
Environmental Protection Agency. Publication Number EPA 560/3-75-005.
June 1975.
STUDY FINDINGS
Cadmium is an extremely toxic agent and may be considered more hazardous
than lead or mercury due to its slower clearance from the body and accu-
mulation in critical organs. The best documented health effect of cad-
mium is renal tubular failure which is projected to occur after 50 years
of exposure to levels only twice what is considered the current normal
intake. In addition, there appears to be substantial evidence that
increased prevalence of hypertension occurs from the current exposure
levels for different parts of the population.
-------
While air's direct impact on the intake and retention of cadmium is
small, the intermedia flow of cadmium indicates that much of the cadmium
in other media can be a result of cadmium originally emitted to the air.
Average contributions of air emissions to the total intake of cadmium were
found to be 64 percent when these intermedia flows were considered. Thi™
problem is a result of the long-term buildups associated with the lew
' levels of cadmium in the body and the biosphere. Even assuming minimal
contributions from air to other media, a contribution of 12 percent was
calculated to be attributable to air pollution. These were static cal-
culations; i.e., assuming that the emission levels were constant over a
long period of time. This has obviously not been the case in the past
where emissions rose with increasing industrialization and then decreased
under control programs but is more indicative of what may be expected if
current levels of emissions are allowed to continue into the future.
However, the current emission levels are not expected to continue unchanged
Controls for attainment of criteria pollutants' air quality, standards and
changes in demand of cadmium and metals with which cadmium is associated^
v . I
will have impacts on the emission levels. New Source Performance Stan°
dards have already been promulgated for incinerators, power plants, and
iron and steel plants to control their particulate emissions. . Under total
particulate control, the emissions of cadmium from these facilities will
also be controlled to some extent. Performance standards for new primary
copper, zinc, and lead smelters have also been recently promulgated
(41FR2232; January 15, 1976). Under the new standards, particulates, and
therefore cadmium, would be controlled through both the S02 controls and
the requirement of fabric filtration equivalent control (99.9 percent
efficient).
Cadmium's dispersal into the environment is being further controlled
through water effluent standards. Exposure to cadmium through drinking
water will receive slightly more stringent control through EPA's stan-
dards for drinking water quality measured at the tap,.
-------
STRATEGY RECOMMENDATIONS
The findings of this "current knowledge" study indicate that the immediate
situation with respect to. cadmium is in a state of flux due to recently
promulgated or proposed standards discussed above, which will affect cad-
mium's dispersal into the environment. This situation makes it difficult
to propose specific strategy recommendations until an analysis of these
changes is made and the degree to which other controls are possible or
•
warranted is determined. However, several observations can be made on
the possible control approaches that are available.
The establishment of air quality standards for cadmium is not warranted.
Cadmium has not been shown to a major nationwide problem with respect to
air quality. An air quality standard for cadmium would necessitate mon-
itoring networks resulting in the commitment of major funds on a pollutant
that is often below the detectable limit..
New source performance standards have already been promulgated for most
of th'e major sources of cadmium. Though not specifically included in
the performance standards, some control of cadmium would result from the
control of total particulate emissions. In addition, the projected growth
of- any of these sources is relatively small so any impact of NSPS would be
minimal.
The designation of cadmium as a. hazardous pollutant is also not recommended
at this time. While probably the best approach for control of cadmium,
the time-frame for decision making built into the Clean Air Act would
make it difficult to properly assess cadmium considering the questions
that apparently need to be addressed.
Though no specific action with respect to control of cadmium emissions is
recommended, it is strongly suggested that preparations be undertaken for
the eventual formulation of control approaches to be adopted. Cadmium
-------
is undergoing continuing study by EPA's Office of Toxic Substances and
the National Science Foundation and is likely to remain in the fore-
front of those elements and chemicals receiving increased attention
due to their carcinogenity. Attention has also been directed at cadmium
due to its potential relationship with hypertension; this is one of the
effects being investigated in the cooperative study by EPA and the
^ 1
National Heart and Lung Institute. Even if calls for control to prevent
cancer or hypertension are not warranted, the safety factor for exposure
before renal tubular failure is only two for an average exposure and is
nonexistent for heavy cigarette smokers.
Consequently, specific recommendations for action involve research pro-
»
grams. These research programs may be divided into two categories. The
first category is more germane to the needs of the Standards and Cost
Analysis Branch as it addresses the better quantification 'of emis-
•
sions, the impact of air emissions on other media and total human exposure,
and the development of air emission control approaches. The second cate-1
gory of research programs is directed at the better understanding of the
actual impacts of cadmium both in the biosphere and in the human body.
These latter programs involve the design and execution of experimental
studies and are more likely undertakings of the Environmental Research
Centers and/or the Office of Toxic Substances.
RESEARCH PROGRAMS
Control Strategy Oriented Research
A great deal of uncertainty exists in the estimation of cadmium emissions
from essentially all source categories„ For example, stationary com-
bustion sources, agricultural burning, and forest fires have only re-
cently been recognized as potentially major sources<, State implementa°
tion plans and new source performance standards have had impacts on the
control applied to sources, and will continue to do so; such impacts
-------
have not been thoroughly documented. .Therefore, a study effort is
needed to define the current and projected engineering and production
parameters required to develop a quantitative control strategy for
cadmium emissions. Specific recommendations for such a study are
listed below: .
« '
e As the number of industries of major concern is relatively
small, especially for the primary metals manufacturers, an
attempt to contact each one individually should be made to
obtain specific process and control equipment information.
Where resistance is met, data on control technology should be
available from state air pollution control agencies.
e An effort similar to the one being conducted by Purdue
University^ should be undertaken to estimate the cadmium
concentration in ores, concentrates, and other feed stocks.
• Cadmium specific sampling studies should be carried out to
determine the cadmium concentrations in process and exhaust
streams and collection efficiency by control devices. Em-
phasis may be placed on mass loading determinations rather
than particle size and particle size distribution information.
Although size information and fractional efficiencies of
collection devices are important, they increase problems of
sampling and analysis, and may delay the obtaining of needed
.-; mass loading information.
• Area sources such as agricultural burning and forest fires
should be evaluated. The cadmium content of vegetation and
the quantities of vegetation burned should be estimated.
e Further analysis of cadmium in fossil fuel should be per-
. formed as this is probably the largest source v/ith a big
population exposure. Project changes in cadmium emission
under increased utilization of coal.
• As paYt of the analyses of individual industries, collect
data on costs of control technology, and possible cadmium
substitutes.
In addition to the better quantification of emissions and the assessment
of control technologies, it is important to provide data on the relative
impact of air emissions on body burden v/hich is more supportive than given
in this report.
-------
Model current cadmium emission levels from a single source
and projected emissions of cadmium from a facility under
new source performance standards to give the maximum ambient
concentration of cadmium that people living near a source
may be exposed to»
Based on data in this report and other sources, develop a
model of cadmium's flow into the environment. This model
would be dynamic as" opposed to the static analysis given
in this report and would be able to calculate further
buildup of cadmium in the biosphere and resulting reten-
tion using projected changes in the contamination resulting
from different sources; e.g., decreased cadmium air and
water emissions in the immediate future but rising with
increased consumption of cadmium, increasing use of fer-
tilization over time. This would help to indicate the
increasing or decreasing role of various sources and
possibly future increased or decreased exposure levels*
(An interesting offshoot would be to extend the model
backwards in time to determine the cause of current
cadmium levelso)
Effects Research
The following are obvious research needs to better identify the impacts
of cadmium in the biosphere and in humans. They are not directed at
control strategy determination except to the extent that they -identify
the degree of control necessary and help the understanding of the inter-
media aspects of cadmium. The Programs are considered to be beyond the
individual undertaking of the Standards and Cost Analysis Branch.
« Conduct research designed to clarify the cadmium-hypertension
question. If such an effect were real, it would be the dom-
inant factor in protecting the public health, and would
likely require standard-setting action. The current EPA -
National Heart and Lung Institute study may provide suffi-
cient information to link hypertension with cadmium in
water. Ambient air exposure should also be considered
as part of this study or in a separate tpidemiological
study.
9 Investigate the repoi'ted modification of bodily function
(increased ribonuclease excretion) at.renal cadmium levels
much lower than those currently deemed significant. Such
8
-------
research should also consider the actual impact of any such
modification on human well-being.
• .Support research designed to improve determination of bodily
absorption-retention figures. These values are central to
any quantitative calculations involving total body burden
and tissue levels.
• Support biological research oriented at understanding the
bodily mechanisms and actions of cadmium. Such research ,
should have priority over similar studies of less-toxic
materials.
0 Support research into plant uptake xi/hich is carefully de-
signed to account for different routes of absorption and
compound forms of cadmium. This data will help to further
define the impact of ambient cadmium levels on food levels.
It should be performed on a wide range of foods, and espe-
cially on tobacco.
e Initiate a long term monitoring effort directed at cadmium
and other heavy metal concentrations in dustfall and in the
soil. Such research can provide data on the source of any
changes in the content of toxic compounds in the soil, and
eventually in the.food, which could be the result of con-
tinous low levels of the compounds in the air.
REFERENCE
1. •> Preliminary Assessment of Suspected Carcinogens in Drinking Water
(Apprendices). Inteiim Report to Congress. U.S. EPA. June 1975.
p. 32.
2. Yost, K.J., et al. The Environmental Flow of Cadmium and Other
Trace Metals. Volume I. Purdue University. Progress Report.
July.1, 1972 to June 30, 1973.
-------
SECTION II
CADMIUM: ITS PROPERTIES, MEASUREMENT,
PRODUCTION AND USES
This section considers the properties of cadmium, its analytical meas-
urement, and its production and uses. The discussion draws heavily on
statistical information and analyses presented by the Bureau of Mines
» t
as well as on the other summary reports dealing with cadmium. Although
more detailed information can be obtained by reference to the above noted
reports and publications, an attempt has been made here to consider that
.•
information which is most pertinent to the present and future impact of
cadmium on the environment. This section serves primarily as an introduc-
tion to the balance of the document where emission sources, cadmium trans-
port through the environment, and its effect on human health are con-
sidered in more detail.
PROPERTIES OF CADMIUM AND ITS COMPOUNDS
Cadmium was first discovered in 1817 and was first produced industrially
in the United States in 1907. It is a comparatively rare element with
an abundance in the earth's crust estimated to be of the order of 0.55 ppnu
It is always associated in nature with zinc, and is usually present as
the sulfide. Cadmium appears to be toxic to man in all forms and concen-
trations, due primarily to its ability to replace zinc and other essen-
tial trace metals from biologically important metallo-organic compounds.-
Body elimination of cadmium is extremely slow and the metal must be con-
sidered a cumulative poison. The toxicity and other biochemical proper-
ties are considered in detail in Section IV. Emphasis is placed here on
10
-------
those physical and chemical properties that are pertinent to the use and
distribution of cadmium throughout the environment.
The properties of cadmium are attributable to its position among the
elements. Cadmium, along with zinc and mercury, is in Group lib, Period V
of Mendeleev's periodic table of the elements (Figure 1). Important phy-
sical constants are tabulated in Table 1. The most important characteris-
tic of cadmium for air pollution planning is its high volatility, as evi-
denced by the relatively low melting and boiling points tabulated in
Table 2 which also includes the melting and boiling points of metals
with which cadmium is associated and cadmium compounds. The high vola-
tility of cadmium relative to zinc is shown in Figure 2 where vapor
pressure as a function of temperature is plotted for both cadmium and zinc
The proportion of cadmium to zinc, normally about 1 to 200 in zinc ores,
will always increase in the vapor phase as a result of high temperature
treatment of zinc ores. Thus zinc smelting, one such operation, produces
a flue dust high in cadmium content and a product, zinc metal, depleted
in cadmium content. Similarly, the release of cadmium fume is favored in
,:
thermal treatments such as ore roasting, sintering, brazing, the remelting
of steel scrap, incineration, and combustion processes.
Vaporized cadmium is monatomic and very reactive. Depending upon condi-
tions, it reacts quickly to form the oxide, chloride, sulfate or other
compounds. Cadmium fume, like all fumes formed by vaporization and con-
densation processes, is extremely fine and therefore difficult to capture
efficiently in air pollution control devices.
Other properties of the metal are important and contribute to its useful-
ness in various applications. The metal is ductile, easily soldered,
readily electrodeposited, and retains a lustrous metallic appearance in
air. All of these properties contribute to its usefulness as a protective
plating on other metals. Because of its position in the electromotive
scries, it imparts galvanic protection as a plating on iron or steel and
maintains its corrosion resistance in seawatcr. Of ultimate significance
11
-------
ts>
10
Ha
H
1 |
3
Li
11
No
(9
K
3?
Rb
55
Cs
1 .. ,. ,r .
4
Be
12
Mg
20
Co
38
Sr
56
Ba
CCo
B
13
Al
21
Sc
39
Y
57
La
IVa Yo Via
Vila YITIa
i
i
t
^ ....... _._ TrVAhlTlT n<" •"' f^'f-iff —
Ti V Cr
Zr Nb Mo
i
1
t
I
i
25 26 27 28
Mn Fe Co Mi
i
Te Ru Rh Pd
Hf To W 'Re Os Ir Pt
Ib
lib
Illb j IVb
1
29
Cu
47
Ag
79
Au
30
Zn
48
Cd
80 n
Hg
31
Go
49
In
81
TJ
Vb
Vib
VHb
! i
! i
-._ --
c
Si
32
Ge
50
Sn;
82
Pb
7
N
p
33
As
51
Sb
83
Bi
8
0
S
Se
Te
9
F
Cl
35
8r
53
I
Po j At
Fr ! Ro ! Ac j Th Pa V Np Pu Am Cm J BK i
i t t i . tit
Figure 1<> Periodic Table of Che Elements.
-------
Table 1.. PHYSICAL -PROPERTIES OF CADMIUM
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
8.65 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
National Bureau of Standards nomenclature, H,
Table 2. COMPARATIVE MELTING AND BOILING POINTS
Cadmium
Zinc
Lead
Copper
Iron
Cadmium Sulfate
Cadmium Chloride
Cadmium Oxide
Melting point
°C
321
420
327
1083
1375
1000
566
°F
610
788
621
1981
2507
1832
1051
Sublimes at
Boiling point
°C
767
907
1620
2336
967
°F
1413
1665
2948
4237
1773
1493°C(2727°F)
13
-------
T6O
0.0002 -
0.0001
0.076
300 400 500 600 700 800 900 1000
TEMPERATURE (°C)
Source: ORNL
Figure 2. Vapor pressure of cadmium and zinc versus temperature
-------
to human health is the reactivity of cadmium ions with biologically
important molecules. A study of the formation constants of metals indi-
cates that the toxicity of cadmium may be due to its ability to displace
many of the biologically necessary heavy metal ions from sulfur donating
enzymes. Cadmium, as a member of Group lib, exhibits a stable valence
of two. Cadmium and zinc are very similar in this respect to the Group
Ila elements, calcium and magnesium, which form divalent ions of similar
+2 °
size. The close similarity in size of the ionic radii of Cd (0.97 A)
+2 "
and Ca (0.99 A) means that cadmium can replace calcium in compounds in-
volving similar anion and crystal structures and in organic molecular
structures. The crystal structure of various compounds is shown in
Table 3.8
Naturally occuring cadmium compounds, such as the sulfide, carbonate,
oxide and hydroxide are generally very low in solubility. Thus, the
amount of cadmium which can remain dissolved in natural waters is usually
quite low. The possible replacement of calcium by cadmium in ionic com-
pounds is also significant as this can further reduce the solubility of
9
cadmium in natural waters.
0
ANALYTICAL ASPECTS OF CADMIUM IDENTIFICATION
Many sophisticated analytical tools are presently available for following
the flow of cadmium in the environment. These techniques are capable of
detecting trace amounts of cadmium (parts per million range) associated
with ambient concentrations occurring in air, water, and biological "samples,
There are generally no severe problems with sensitivity, although in some
studies, low levels in soils and other natural media were not readily de-
-/
tectable. The minimum detectable levels are typically more a function
of sample preparation technique than of ultimate sensing methodology.
There are three potential problems with analyzing for trace levels of
cadmium: sampling media background, sample preparation and matrix effects,
and a specific NaCl interference with analysis by atomic absorption.
15
-------
Table 3. CRYSTAL STRUCTURES OF VARIOUS COMPOUNDS'
8
Compound
Oxides
Magnesium oxide, MgO
Cadmium oxide, CdO
Calcium oxide, CaO
Zinc oxid e , ZnO
Sulfides
Zinc sulfide, ZnS
Cadmium sulfide, CdS
Zinc sulfide
Cadmium sulfide
Mercury sulfide, HgS
Carbonates
Magnesium carbonate,
Zinc carbonate, ZnCO
Cadmium carbonate,
CdC03
Calcium carbonate,
. CaC03
Symmetry
Cubi^c
Cubic
Cubic
Hexagonal
'
Hexagonal
Hexagonal
Cubic
Cubic
Cubic
Rhombohedral
Rhombohedral
Rhombohedral
Rhombohedral
Structure
NaCl
NaCl
NaCl
NaCl
ZnO
ZnO
ZnS
ZnS
ZnS
NaN03
NaNO
NaNof
NaNO-
3
»»
Unit cell dimensions
ao (A)
4.2112
4.6953
4.8105
3.2495
3.811
4.1348
5.409
5.818
5.8517
5.6752
5.6833
6.1306
6.361
co (A)
5.2069
6.234
6.749
•
.
a
48°12'
48°20'
47°19'
-^
46°6'
a °
An angstrom (A) is 10"
structural unit of the
o T|
cm. These are the dimensions of the basic
crystal.
Since the majority of air samples involve the use of some filter material
or absorbent reagent, it is necessary to choose these sampling materials
carefully and consider any possible contribution to measured cadmium .
levels. These contributions to cadmium levels may be due directly to'
cadmium in the filter or reagent or indirectly due to interferences from
compounds present in the sample. To date, the majority of the data
gathered on ambient air concentrations has been from glass fiber hi-vol
filters. Specifications for filters used in nationwide sampling allow
a maximum concentration of 7 vg of cadmium per filter. This small amount
of cadmium would not constitute a significant addition to the result in
an area of relatively high cadmium levels; however, it could be half
or more of the majority of lox^ ambient and background levels encountered.
16
-------
Complicating the appraisal of current and past data is the lack of a
consistent and established method -for sample preparation prior to analysis
for cadmium. Improvements in extraction procedures have indicated that
previous ashing methods were probably responsible for the loss of signif-
icant amounts of the original cadmium deposited on the glass fiber hi-vol
10
filters. As a result, the use of National Air Surveillance Network (NA.SN)
data analyzed prior to 1967 is limited.
Currently in NASN, particulate matter samples are sectioned and compo-
sites are ashed in a low-temperature asher using 50 to 100 milliliters
(raJl) of oxygen per minute at 1 torr with an induction coil energized
with 250 watts. At this combustion temperature (about 150°C) cadmium
is retained essentially quantitatively, whereas half of the cadmium in
samples oxidized in a muffle furnace is lost at between 500 and 550°C.
The residue is extracted using a mixture (4:1 by volume) of redistilled
nitric and hydrochloric acid and concentrated. The solution is then
freed of silica by centrifugation and brought up to volume with redis-
tilled nitric acid. The solution is analyzed directly by Optical Emission
Spectroscopy (OES), which has a detection limit of 0.3 microgram per
»
milliliter (yg/m&) of sample and a relative standard deviation of 8 per—
, cent at 0.8 yg/m£. For special studies, this solution is diluted 10:1
with water prior to analysis by Atomic Absorption Spectrophotometry (AAS).
^
The minimum level detectable by AAS is 0.04 ug/m£, and the standard devia-
tion is 2 percent at or above 0.5 \ig/mi.
A recent study by EPA, investigating the credibility of trace element
analyses, revealed wide ranges of reported concentrations for identical
samples. Cadmium levels determined by spark source mass spectrometry
(SMSS), OES, neutron activation analysis (NAA), AAS, 'and anodic stripping
voltammetry (ASV) were found to vary by more than an order of magnitude.
NAA provides the most accurate results but is also the most expensive.
The conclusions of this study were that there is a definite need to study
analytical methods for certain matrices and that standard reference ma-
terials should be available to aid analysts in evaluating their procedures.
17
-------
The principal analytical difficulty specific to cadmium is a positive
interference by sodium chloride in the atomic absorption analysis as
12
documented by Piscator, As illustrated in Figure 3, false high cadmium
values are produced by NaCl levels above about 0.005M. This effect has
been cited by Friberg et al. in questioning some analytical results,
14
although some of these have"also been defended by Fulkerson et al.
as having properly compensated for the interference.
Until more uniformity in sample extractions and analytical methods is
incorporated, present and past data on ambient cadmium concentrations
must be evaluated carefully and the employment of the data in an absolute
sense must be questioned. However, data collected and analyzed*in the
same manner in different studies should be useful for analyzing relative
changes. .
?
With respect to source testing, the analytical techniques presently
available are considered sufficiently sensitive since the higher concen-
trations associated with cadmium emission sources pose no problem in
obtaining accurate levels of cadmium. Present analytical techniques for
cadmium and other trace elements are far ahead of the sampling techniques,
storage methods, and the sample preparation-methods presently employed
though this lag can usually be overcome if the sample in question is
specific for one trace element. Cadmium levels, sought in connection with
emission sources, can be adequately determined if no other element is to
be sought from the sarae sample.
PRODUCTION OF CADMIUM
Since zinc-bearing ores are the principal source of cadmium, the pro-
duction of cadmium is closely allied to zinc production. In recent years,
the domestic production of zinc, and to a lesser extent cadmium, has
declined because of an unfavorable economic picture. Demand for both
metals has nonetheless continued to increase, and the gap betveen deman!
18
-------
ppm Cd"
1.0-1
0.1-
001
0.005
0.05
0.5 M NaCI
The octuol amount o' todmium in NoCl 'detetmmed with olomic
obsorbnon pltcr rttraciion with MI8K -APOOwct 00)27 ppm.
i * .0 5M NaCI solution contained OOOOAppm Cd.
Source: CITE, Section 2.4
Figure 3. Sodium chloride interference in cadmium analysis
19
-------
and production has been filled by withdrawals from national stockpiles
and imports of the metals and their ores.
The marked decline in domestic cadmium and zinc production is shown in
Table 4 and Figures 4 and 5, which depict yearly production and con-
sumption figures for cadmium and zinc. The reduction in zinc production
has also been accompanied by a marked decline in smelting capacity.
Domestic zinc smelting capacity has decreased 47 percent, from approxim-
ately 1,330,000 tons per year in 1968 to almost 707,000 tons per year in
June 1975. The number of smelters diminished from 14 to 6 in that period.
The smelter closings have been attributed to a variety of economic causes,
including anticipated capital investment to meet future, more stringent
environmental regulations. The location of existing zinc smelters and
their company identification are given in Figure 6 and Table 5. From
this table and figure, it is apparent that zinc smelting occurs mostly
near population centers and in the midwest and east. •
'»,-
t
Primary cadmium is produced commercially in the United States solely as
a result of zinc processing. Zinc- or cadmium-bearing materials from
other ores, primarily lead ores, are input into zinc processing operation,
rather than being reclaimed elsewhere. Six.companies operating seven
plants, physically associated with six zinc smelters and one zinc oxide
plant, are responsible for the domestic output of cadmium. Although some
reduction in cadmium production has occurred, it has not been as pronouncec
as the reduction in zinc production. In fact the zinc ore processed in
1974 cannot account for the amount of cadmium produced; the difference
apparently results from imported flue dusts and zinc dusts from other
primary metal processing operations. A small amount of secondary cadmium
recovery is also reported to exist; Bureau of Mines estimates are that
this source of cadmium may account for 200,000 pounds annually.
The outlook for trends in production and consumption of both zinc and
cadmium are considered later in this chapter, following a discussion of
cadmium uses.
20
-------
Table 4. CONSUMPTION AND PRODUCTION STATISTICS FOR
CADMIUM AND
Year
1975a
1974
1973
1972
1971 >•
1970
1969
1968
1965
1960
1950
Cadmium
United States
Consumption,
10^ pounds
_
13,880
12,500
12,614
10,873
' 9,186
15,062
13,328
10,431
10,166
7,171
Production,
10^ pounds
2,448
6,666
7,428
8,290
7,930
9,465
12,646
10,651
9,671
10,180
9,190
World
Production,
10 pounds
-
41,640
-
36,599
34,241
36,454
38,784
33,105
26,100
21,700
13,514
Zinc
United States
Consumption,
short tons
589,000
1,673,000
1,932,000
1,828,000
1,780,000
1,572,000
1,814,000
1,745,000
1,742,000
1,150,000
1,350,000
Slab zinc
production
from ore,
short tons
245,000
634,000
629,000
633,000
766,000
878,000
1,041,000
1,021,000
994,000
800,000
843,000
Data for first two quarters of 1975
21
-------
1945
1950
1955
1960
1965
1970
1975
Figure 4. Domestic cadmium production and consumption
(1-4)
22
-------
2,100
1,800
O
1,500
CC
O
V)
O 1,200
z
o
o
600
300
PRODUCTION
FROM ORE
V
0 I i i i i i i t i i I i i i t I i i i i I i i t i I i t
1945 1950 1955 I960 1965 1970
1975
Figure 5. Domestic zinc production and consumption
(1-4)
23
-------
NJ
(E) Electrolytic
(HR) Horizontal Retort
(VR) Vertical Retort
(ZnO) Zinc Oxide Plant
C_,^\.ViV ^
a NATIONAL ZINC (BARTLESVILLE)
<
ASARCO (CORPUS CHRIST!)
(E) .'
Figure 6. Domestic primary zinc smelters with secondary cadmium plant, 1975.
-------
Table 5. DOMESTIC SLAB ZINC CAPACITY
1968-1975 :
Company /smelter location
ASARCO/Amarillo, Texas0
ASARCO/Corpus Christi, Texas
Blackwell Zinc/Blackwell, Oklahoma
National Zinc/Bartlesville, Oklahoma
New Jersey Zinc/Paltnerton, Pennsylvania
St. Joe Minerals /Monaca, Pennsylvania
Bunker Hill/Kellogg, Idaho
American Zinc /Dumas, Texas
American Zinc/Sauget, Illinois
Eagle-Picher/Henryetta, Oklahoma
New Jersey Zinc/Depue, Illinois
Matthiessen & Hegler/Meadowbrook, W. Va.
Anaconda/Great Falls, Montana
Anaconda/Anaconda, Montana
Total
Estimated capacity,
short tons
19681
55,000
108,000
88,000
63 000
118,000
225,000
109,000
58,000
84,000
55,000
70,000
45,000
162,000
90,000
1,330.000
19752
_•
108,000
-
63,000
118,000
225,000
109,000
84,000
-
-
-
-•
-
707,000
Sources: (1) Yearbook of the American Society of Metal Statistics
for 1971. Issued June 1972.
(2) U.S. Bureau of Mines.
3Closed June 1975.
Now owned by Amax Lead and Zinc Company.
25
-------
USES OF CADMIUM
The Bureau of Mines does not sponsor an annual survey of cadmium uses,
so usage figures must be estimated from various sources related to the
many areas of application of cadmium. Of the 6,230 tons of cadmium that
were consumed in the United States in 19733 about 50 percent was used for
electroplating, 20 percent for paints, 15 percent for plastics9 and the
remainder for batteries and other uses. The similarity of this estimate
with others presented for 1968 is shown in Table 6.
Table 6. ESTIMATED USAGE OF CADMIUM, TONS
•
Usage
Electroplating
Pigments
Plastics
Batteries
Alloys
Miscellaneous
1968
Davis
3,000
1,400
1,000
200
500
564
ORNL
3,000-
3,250
1,200-
1,400
1,000
200
250
314-
1,014
BM1
4,000
750
1,250
200
460
1973
BM4
3,000
1,110
990
630
500
Electroplating
Of the 3,000 tons used for electroplating in 1973, about 40 percent was
used for motor vehicles, boats, and airplanes. Other plating uses include
the plating of hardware3 household appliance parts, fasteners, and a va-
riety of industrial parts. Of all cadmium applications, electroplating
appears to be the primary one for which alternative materials are not
readily available. Cadmium plating provides galvanic corrosion protec-
tion for iron and steel, has a pleasing, silver-like appearance, and
26
-------
good dimensional, electrical and joining capabilities. None of the sub-
stitutes discussed in detail in the ORNL report can entirely reproduce
the properties of a cadmium plate, particularly with regard to its lus-
trous appearance.
Plating is usually done from a cyanide bath. Although direct air pollu-r
tion emissions are negligible, water contamination has occurred in the
past due to industrial dumping or accidental spills. Recently promulgated
cadmium standards for water effluents from electroplating operations
(40 CFR 413) should reduce the amount of contamination in the future.
Plastics
Almost 1000 tons of cadmium were used in 1973 as a stabilizer for plastics,
primarily polyvinyl chloride (PVC) plastics. Cadmium salts of long-chain
organic acids, such as cadmium stearate, are commonly used. Cadmium sta-
bilizers, containing about 2 to 3 percent cadmium, inhibit the degradation
of the polymer by reacting with the hydrochloric acid that is formed upon
/*
degradation of the PVC. The estimate made by ORNL of the maximum cadmium
*
content in PVC (600 ppm) appears to be low using known F\/C production
18
figures and estimates of 1,000 tons of cadmium used as stabilizer,
although they did postulate that some PVC contains significantly higher
cadmium concentrations.
Although cadmium compounds appear to be highly satisfactory stabilizers,
the use of many other compounds, including many tin organics, may be •
effectively substituted. Cadmium stabilizers are not accepted for food
packaging by the FDA and the USDA, nor in plastic pipe for plumbing.
Pigments
Another large cadmium use category is as a pigment in paints, plastics,
coated fabrics, inks, glazes, and other coatings. More then 1,100 tons
27
-------
of cadmium were used for pigments in 1973. The principal compounds used
are cadmium sulfide and cadmium sulfoselenide. These primary pigments
are used along or mixed with mercury, zinc and barium compounds. These
yellow-to-red compounds have outstanding covering power, good stability
against heat and light, moisture, weathering, and chemical resistance.
They are used to a large extent in plastics where their high temperature
stability is desirable for high temperature molding operations. Substi-
tution of other dyes and pigments, though..not completely satisfactory
from the standpoint of color absorption, should pose no serious problem
for the present users of these cadmium pigments.
Batteries
The estimated use of cadmium in batteries has more than tripled since
1968, to a 197." total of 630 tons. The use of cadmium in the'nickel- i
cadmium battery is the most prominent application in this category.
Some cadmium.is also used in the silver-cadmium battery for special appli-
cations where performance is more important than cost. In both systems,
sponge cadmium is used as the negative plate. Nickel-cadmium batteries
are superior in many respects to lead storage batteries, particularly
with regard to life. • Principal use in the United States is for small
voltage power systems of a sealed cell design because the batteries can
be repeatedly charged and discharged without any buildup of internal pres-
sure. Failure to compete against the lead-acid battery used in motor
vehicles is attributable to the higher initial cost (2 to 3 times that of
the lead-acid battery), although the lifetime is at least 3 times that of
the lead-acid battery. The ORNL study has considered.this application in
detail and comes out strongly in favor of inducements to promote the use
of cadmium for batteries. Cadmium used in this form would be essentially
nonpolluting since recycling would be possible and the previously discussec
applications, intrinsically dissipative in nature, would likely be elim-
inated or significantly reduced because of price-demand relationships.
28
-------
Other Applications
Approximately 500 tons of cadmium find use in alloys and other miscella-
neous uses. Major cadmium alloys are the low temperature solders, silver
brazes, and a special copper alloy used for automobile radiators. Silver-
cadmium alloys also are used as nuclear reactor control rods. Other uses
include cadmium phosphates for television tubes and fluorescent tubes,
silver-cadmium oxide as electrical contacts in motor starting switches,
relays and. circuit breakers, and certain cadmium crystal forms as serai-
conductors. Minor amounts of cadmium are also used in such applications
as rubber catalysts, fungicides, and other products.
ANTICIPATED FUTURE TRENDS
In the assessment of the need for control measures for a particular toxic
substance, an important consideration is the rate at which its production
and consumption, and thereby its dissipation in the environment, is
expected to increase or decrease. If a substance is projected to undergo
a major increase in production it is obviously much more imperative to
*
pursue controls than if the use of the substance was declining. In addi-
tion, the identification of expected growth or decline helps to delineate
possible control approaches; e.g., new source performance standards or
controls on existing sources.
When considering anticipated trends and predictions with regard to cad-
mium supply and demand, it should be remembered that environmental,
economic, and other unforseen factors may well play a dominant role in
the future outlook for cadmium, particularly as cadmium-oriented control
measures are adopted for water emissions. If current regulations for cad-
mium water effluents result in curtailment of some or most uses of cadmium,
they would obviate any projections based on standard growth indicators.
At a minimum it is considered unlikely that new applications for cadmium
will be actively sought. Instead the search will be for alternative
29
-------
materials. Thus the domestic projections as developed by ORNL and the
Bureau of Mines must be considered with caution, in the light of possible
changes in technology and supply-demand relationships..
Demand
The demand for cadmium is primarily a function of its use in four areas—
electroplating, plastic stabilizers, pigments, and Ni-Cd batteries. These
four uses are responsible for 95 percent of the cadmium consumed annually.
Currently, the U.S. is consuming more than twice the cadmium it produces*
This consumption amounted to one third of the world production in 1974 and
was the highest recorded since 1969. The Bureau of Mines has estimated a
\ .
probable U.S. cadmium demand of about 300,000 tons and world demand of
19
about 750,000 tons as cumulative values for the year 1972 to 2000. The
yearly U.S. demand was projected to rise from the current 6,000 tons to
r
15,000 tons over this time period. A simple linear extrapolation of pre-
vious cadmium consumption levels would have" indicated demand to be of the
order of 13,500 tons in 2000 <> The major unknown variables with the greatest
effect on future demand are:
1. technological change - I.e., a new substitute or new use
for cadmium
i
2. changes in the prices of substitutes and complements for
cadmium and cadmium containing products
i
3. changes in the demand for cadmium containing products
A historical table of cadmium demand as well as a 1985 forecast for the
major end uses is given in Table 7. Forecasts were based on anticipated
growths of population and GNP, tempered by estimates of future technolog-
ical developments.. Because of slowdowns in both population and GNP not
foreseen in the projections, and because of restrictions in application due
to demand-supply considerations and concern over the high toxicity of cad-
mium, it is felt that the projected demand estimate may not be attained.
30
-------
Table 7. U.S. DEMAND STATISTICS FOR CADMIUM DATA IN TONS PER YEAR
(ELEMENTAL CADMIUM)
Year
1963
1964
1965
1966
1957
1968
1969
1970
1971
1972
1973
1974
1985
Electroplating
3,150
2,190
2,700
4,060
3,300
3,910
4,370
2,260
2,710
3,150
3,000
5,350
Pigments
1,720
1,300
1.280
1.000
550
670
730
650
770
950
1,110
1,050
Heat stabilizers
(plastics)
400
370
430
1,190
1,100
1,260
1,350
1,190
1,420
1,250
990
2.600
Batteries
200
230
310
250
200
200
250
150
180
450
630
650
Other
190
230
450
710
390
400
470
280
330
. 510
500
890
Total U.S.
industrial demand
5,670
A, 330
5,170
7,210
5,530
6,450
7,170
4,530
5,410
6,310
6,230
6,950
10.540
Avg. price
(constant 1974 dollars),
$/lb
3.74
4.92
4.20
3.83
4.03
3.88
4.58
4.74
2.42
3.24
4.20
4.09
«•
Source: Bureau of Minos.
-------
Since cadmium represents a small percentage of the total cost of most
consumer products containing cadmium and since few adequate substitutes
exist in many cases, even large price changes have little effect on
demando Price is typically 3 to 5 times the production costs„
.Supply
;*
Domestic resources of cadmium, based on known and estimated zinc resources,
are about 450 to 700 million pounds. World reserves are estimated to be
19
1,600 to 1,875 million pounds. Projections of future demand, based on
historical trends and anticipated growth patterns in usage industries,
indicate that present reserves will be consumed by the year 2000.
Zinc reserves appear to be sufficient to last through the year 2000 at'
anticipated demand levels.
i
As discussed earlier, the domestic zinc industry has changed appreciably
since 1968 due to the reduction in the number of smelters from 14 to 6.
The resulting reduction in capacity was about 47 percent, down to 707,000
tons per year from 1,330,000 tons per year in 1968. The zinc smelter
closings are due to a number of factors including old inefficient smelters,
high operating costs, and industrial over-capacity. However, since demand
for cadmium is about twice the U.S. production, increases in cadmium
production may be easily assimilated. With the opening of new zinc
smelters in 1977, the U.S. would also be increasing its potential cadmium
supply. This is discussed further in the following section where the
individual sources of cadmium emissions and possible trends are analyzed.
Table 8 provides supply statistics for cadmium. The major factors which
are likely to have the greatest effect on cadmium supply are;
1. Price of zinc.
2. Growth in /cine production.
3. Impediments to foreign trade.
4. Recovery from secondary sources.
32
-------
Table 8. SUPPLY STATISTICS FOR CADMIUM DATA IN TONS PER YEAR
Year
1?63
1?64
1955
1906
1967
1953
1969
1970
1971
1972
1973
1974
t
! U.S. refinery production
5,010
5,190
4,«60
5,190
4,340
5,330
6,330
4,740
3,970
5,140 ',
3,720
3,510
World production
13,010
14,100
. 13,120
14,330
14,410
' 16, 550
19,380
18,220
17,140
18,300
18,650
18,740
U.S. imports of metal
for consumption
500
550
1,060
1,680
790
960
540
1.250
1,750
1,210
1,940
1,900
U.S. exports
660
720
40
190 '
350
270
542
190
30
510
150
30
Shipments from
.CSA stockpile
970
500
0
180
520
400
1,380
3
0
480
400
1,010
GSA stocki
8,070
7,560
7,560
7,380
6,860
6,460
5,080
5,060
5,080
4,600
4,220
3,220
U.S. supply
(U.S. production + Inporti)
5,500
5,740
5,920
6,870'
5,140
6,290
6,870 •
5,990
5,720
5,360
5.660
5.400
U)
Source: Bureau of
-------
Assuming an annual increase in world zinc production of 3 percent and
assuming some additional secondary cadmium would become available in the
U.S., the 1985 supply would be about 40 percent higher than the 1974 sup-
ply, or about 7560 tons.
34
-------
REFERENCES .
1. Mineral Facts and Problems. Bureau of Mines. 1970.
2. Mineral Industry Surveys, Cadmium, Zinc, Lead and Copper. Bureau of
Mines. 1969-1973.
3. Minerals Yearbook. Volume I. Metals, Minerals, and Fuels. Bureau
of Mines. 1972.
4. Personal communication with James Hague and M. Babitzke. Bureau of
Mines. March 1974.
5. Chizhikov. Crustal Abundance. ORNL. Litton.
6. Selected from tabulations in Chishikov and ORNL.
7. Sillen, L. G.,-and A. E. Martell. Stability Constants of Metal Ion
Complexes. Special Pub. No. 17. The Chemical Society, London, 1964.
Supplement No. 1, Special Pub. No. 25, 1971.
8. ORNL. Figure III-3.
9. ORNL. Section III-C.
10. Hwang, Joe Young. Trace Metals in Atmospheric Particulates and Atomic
Absorption Spectroscopy. Analytical Chemistry 44(14): December 1972.
11. Von Lehmden, D. J., R. H. Jungers, and R. E. Lee, Jr. Determination
of Trace Elements in Coal, Fly Ash, Fuel Oil, and Gasoline — A Pre-
* liminary Comparison of Selected Analytical Techniques. Analytical
Chemistry 46(7):239. February 1974.
12. Piscator. Unpublished data, in CITE. Section 2.4.
13. CITE. Section 2.4.
14. ORNL. Appendix A.
15. McMahon, A. D., et al. The U.S. Zinc Industry — A Historical Perspec-
tive. Bureau of Mines Information Circular 8629. 1974.
16. Bureau of Mines. Note 4.
17. ORNL. Section V-E.
18. ORNL. Section V-D-4-c.
19. ORNL. Confirmed by Bureau of Mines. Note 4.
35
-------
SECTION III
SOURCES AND CONTROL OF AIRBORNE CADMIUM
This section discusses the sources and control of cadmium emissions to
the atmosphere. Emission estimates are made and contrasted with estimates
made by other workers in this area, beginning with the initial study by
Davis. Along with a discussion of sources and the magnitude of their
emissions, consideration is given to future emissions by examining anti-
cipated growth patterns and possible changes in operating and control
practices.
Table 9 summarizes cadmium emissions from all known major sources as es-
timated by GCA and earlier investigators. All of the estimates made are
based on rather meager data. This is particularly true of many of the
larger source categories such as fuel combustion, iron scrap recovery, and
even primary metal processing. A definite need exists for more data in
these areas which will enable future workers to assess the flow of cadmium
through the various processes. The Purdue study is one example of neces-
sary work in this area.
As can be seen from Table 9, cadmium emissions result largely from high
temperature processes during which the relatively low-boiling point cad-
mium is vaporized„ It subsequently condenses to form particulates or
condenses on other particulates. Like all particulate matter formed by
vaporization and condensation processes, the particle is small, although
the size, size distribution, and the chemical nature of the cadmium par°
ticulate depends upon the specific situation. Vaporized cadmium metal is
rnonatomic and highly reactive, and will most likely be found as the oxide
sulfate or some other salt.
36
-------
Table 9. CADMIUM EMISSION ESTIMATES
Source Category
Mining
Primary metals processing
Zinc
Lead
Copper
Cadmium
Secondary Metal processing
Steel scrap
Zinc
Copper
Manufacturing
Alloys
P igmerit
Stabilizer
Misc.
Incineration
Fossil fuel combustion
Sewage sludge incin.
Miscellaneous
Motor oil
Rubber tires
Gasoline
Forest & agric.
burning
Emission estimates tens/year
Davis
1968
<1
J1050
-
1000
-
125
3
11
3
<2
95
-
-
.
1
6
-
»
ORNL
1968
<1
J1050
-
<110
?
?
3
11
3
<2
95
145-1100
-
.
1
6
'-
_
MITRE
<1
619
55
388
-
1000
.
125
3
11
3
<2
95
-
- '
-
-
-
_
EPA
1971
<1
644
163
234
60
78
2
65
1
7
4
<1
48
198
138
-
6
-
—
GCA
1974
<1
500
65
110
50
400
2
70
3
11
3
<2
150
250
12
< 1
6
-
50
Other
Total
2294
1425-2380
2305
1650
< 2
<1688
37
-------
Particle size is of paramount importance in determining the effectiveness
of particulate control devices. Baghouses, electrostatic precipitators,
and possibly high energy scrubbers, are considered effective for the sub-
micron particulates which comprise a significant fraction of cadmium
emissions. Fabric filter baghouses are generally most effective, and are
capable of efficiencies greater than 99 percent for particles as small as
0.1 micron. Medium or low-efficiency control devices are generally almost
totally ineffective for collection of submicron particles„
Many of the sources in the major cadmium emission source categories al-
ready utilize a fairly high level of particulate control technology.
This is true, for example, of the primary metals industry, where most fa-
cilities employ electrostatic precipitators or fabric filter collectors,,
However, even here, proposed new source performance standards may require
additional control measures. These same standards will require.in many
cases, not only additional control equipment, but also process modifications/
involving changes in equipment and process operations aimed primarily at
SO- reductions, The following discussions will consider the overall si-
tuation applicable to each source category listed in Table 9. Figure 7
diagrams the flow of cadmium between the various processes and the re-
sulting emissions to the air9
MINING
This is considered a minor source of air pollution. Davis and others have
assumed an emission factor of 0.2 pound cadmium per ton of cadmium in ore
mined* This emission factor represents estimated emissions due to wind
loss from tailings since the other principal processing operations are
conducted in the wet state and are hence considered to be nonemitting.
The domestic ores mined, their estimated cadmium content, and calculated
emissions for 1973 are tabulated in Table 10;
38
-------
VD
Z:NC
INC'JSTnY ]
250
500
65
iMARY LEAD
INDUSTRY
CAOW1UM
CONSUMPTION
expoars
STA2ILl7ltiS
BATTEKSC; j
ALLOYS
•f£LECTSO?LATINX
OPtH QUP.NIN3
INCINSRATION
SECONDARY COPPER
*• STECL SEPSOCESSiNG
SEWAGE INCINERATION
-------
Table 10, CADMIUM EMISSIONS FROM MINING
Zinc
Lead
Copper
Metal in ore
(tons)
485,000"
600,000
1,720,000
Cadmium in ore
(tons)
2425
285
275
— k
Cadmium emission
(pounds)
485
60
55
600
Although air emissions of cadmium from mining operations are relatively
small, such operations have often been responsible for the introduction
of large amounts of cadmium into the environment,, The greatest potential
is water pollution from the ore flotation process and the tailings. These
operations have been determined to ..be the direct cause of the cadmium
induced itai-itai disease observed in the Jintsu Valley of Japan. Un-
fortunately, little is known about the possible loadings from the mining
operations in the United States,
PRIMARY ZINC PROCESSING
Cadmium is produced as a by-product of the primary metals industry, with
the primary zinc industry being by far the most important. Because
cadmium is an important by-product of zinc processings incentives exist
for the efficient collection of the cadmium-rich fumes for subsequent
cadmium recovery.
Zinc and its associate cadmium are usually found together in nature as the
sulfide; lead is usually also present in most zinc ores. Concentrated ore
as delivered to the smelter contains about 60 percent zinc; the zinc-
cadmium ratio, while highly variable, reportedly, averages about 200 to 1.
4
This value was used by EPA and others in their estimates. More definitive
information concerning cadmium contents of feedstock and other process
streams is needed for all source categories,,
40
-------
Zinc Production Processes
Zinc is produced essentially by two methods: a pyrometallurgical process
and a combination pyrometallurgical-electrolytic extraction process. Both
processes involve, as an initial step, a roasting operation which converts
the zinc sulfide to zinc oxide. The sulfide present in the ore is oxidized
to sulfur dioxide, which is then usually converted to sulfuric acid in a
contact acid plant. .
In the pyrometallurgical processes, the roasted ore is sintered to provide
easier handling and better processing in the subsequent reduction oper-
ation. Reduction of zinc oxide is carried out either in batches in hori-
zontal retorts or cpntinuously in vertical retorts or electrothermic fur-
naces. In all cases, carbon is added to the ore after sintering to reduce
the zinc oxide to zinc metal, which is vaporized, condensed and cast into
slabs. The slab zinc, while satisfactory for certain purposes, is usually
further purified by redistillation or electrolysis. A flow diagram of the
pyrometallurgical zinc recovery process is shown in Figure 8. This
•• 5
figure?, taken from the ORNL report on cadmium, also shows schematically
»
the lead smelting operation leading to zinc and cadmium recovery which is
discussed later.
\
In the combined pyrometallurgical-electrolytic process, the roasted zinc
ore is sent directly to an electrolytic plant where the roasted ore is
leached with sulfuric acid to provide soluble sulfates. Following fil-
tration to remove insolubles, the sulfate solution is treated with zinc
dust to remove copper and zinc-cadmium residues. These residues can be
further processed to recover the by-product metals. Electrolysis is then
carried out to deposit zinc on aluminum cathodes. The zinc product pro-
duced by the electrolytic process is a high-grade product that does not
usually require further refining. Other advantages to the electrolytic
process are that it has a high energy efficiency and is relatively free
of air pollution control problems except for those posed by the initial
roasting step. Although most zinc concentrates can be processed
41
-------
ATMOSPHERE
SULFURIC ACID
RECOVERY
SULFURIC ACID TO ROASTED
ZINC ORE LEACHING OR SALES
ATMOSPHERE
H
ZINC ORE
CONCENTRATE
50,000 Zn
-250 Cd
25.0OO S
24.750 GANGUE
100.OOO TOTAL
ATMOSPHERE
Cd RECOVERY
Cd
CADMIUM OXIDE
TO Cd RECOVERY
<25 Cd
CONC
ZINC ORE
SINTERING
>49.000 Zn
Cd
S
24.500 GANGUE
~75,000
S02
RECOVERY
AND
SINTERED Zn
Zn Cd
VAPOR
ORE CONC
>«9,000 Zn
~25 Cd
250 S
24.000 GANGUE
~ 7 3.00O
>49,000 Zn
~ 2 5.OOO
ZnO TO Zn
ORE ROASTING
SLAB ZINC
•PURIFICATION
OR SALES
H-
LEAD ORE
CONCENTRATE
I
LEAD ORE
ROASTING
INCREASINGLY
BYPASSED
LEAD
ORE
•CADMIUfv
AND AR!
RECOVEf
~°~~ SMELTING
1
LEAD
BULLION
UNFl
SL
s6e/
1
5ENIC
*Y
JMEO
AG
.Zn 1
SLAG
FUMING
1
FUMED
SLAG
£
Figure 8. Schematic flowsheet for recovery of zinc and preliminary recovery
of Cadmium from ore by electrolysis^ (F .- fabric filter or other
emission control device)
-------
elcctrolytically, the presence of certain metals require special treatment,
Scrap metal cannot be processed electrolytically. A flow diagram of a
typical electrolytic operation is shown in Figure 9.
The location and total capacity of the currently operating plants were
given in Figure 6 and Table 5. The percentage breakdown of United States
zinc smelter and electrolytic refinery capacities as of June 1975 is as
follows:
Pyrometallurgical 57%
Horizontal retort (9%)
Vertical retort (16%)
EleCtrothermic Furnace (32%)
Pyrometallurgical - Electrolytic 43%
Emission Sources and Control Technology
The emissions of cadmium are associated with those operations in the pro-
cesses th'at involve thermal treatments: the roasting operation, sintering
and reduction as required by the Pyrometallurgical process, and distil-
lation if necessary to further purify the zinc. A description of.the
operation of existing domestic smelters can be found in Table 11. This
table, adopted from the one in an August 1973 EPA preliminary background
document for pos
U.S. operations,
document for possible new source performance standards, describes 1974
In the United States several types of roasters are used. About half are
modern flash or fluid-bed units. Only one plant in 1974 was using a
number of outmoded Ropp units. This latter type of roaster is not
amenable to conventional S02 control and roaster gases pass directly to
the stack releasing large quantities of S0? and particulate to the atmo-
sphere. This particular plant (ASARCO - Amarillo, Texas) accounted for
the majority of the SC>2 and particulate emissions from all primary zinc
43
-------
ATMOSPHERE
TO^
CADMIUM1
RECOVERY^
* ~100
ZINC OUST
RAW
b.
ZINC ORE
CONCENTRATE
50.000 Zn
-250 Cd
35.000 S
24.750 CANGU
ZINC
ORE
ROASTING
t
100.000 TOTAL
•^
ROASTED Zn
ORE CONC
> 49.000 Zn
150 Cd
1200 S
34.500 GANG
ZINC ORE
LEACHING
UE
SULFATE
SOLUTION
» 49.000 Zr
l
ZINC SULFATE
PURIFICATION
i
1
SULFATE^
SOLUTION
I I
ZINC
ELECTROLYSIS
ZINC
COATED
ANODES
. ZINC
STRIPPING
SLAB ZINC
TO SALES
-49.000
SiO,
AND
CoS04
GANGUE
COPPER. CADMIUM
RESIDUE ZINC
RESIDUE
TO Cd
RECOVERY
-150
Figure 9. Schematic flowsheet for recovery of zinc and preliminary
recovery of cadmium from ore by electrolysis^
-------
Table 11. EXISTING U.S. ZINC PROCESSING PLANTS/ 1974
Co-pany
N»3«/
location
I) ASWUXJ/
Corpus
Christi,
Texas
2) ASAT.CO/
Cc 1 u-T.bus
Ohio
3) ASAF.CO/
Age of plane
First
year
1942
1967
1923
Asorillo;
Texas
4) Rational
Zinc/
Bartlcs-
vllle,
Ok la.
1907
Last
mtxl ili-
cat ion
1972
(Changed
flash to
fluid bed
roasting)
•
.
t
''•
1969
(Acid
Plane)
Materials
Cone
(T/D)
250
200
,i
270
250
Zinc
(T/D)
300
60
(ZnO)
13S
140
Production equipment
Roasters
No. of
oper/
stby
1/0
1/0
6/1
2/0
Gas stream
control -«»
equipment
250 T/l) lluid
bed roaster.
All master
gases to 225
T/D acid plant,
Tail gas to
stack.
200 T/D fluid
bed roaster.
All roaster
g.ises to 17}
T/D acid plant.
Tail fias to :
200 ft stack.
:Six 45 T/D
.Ropp roasters.
Roaster g.ises
pass to 400 ft
stack untrea-
ted. Roaster
flow is
110,000 sera.
Two 130 T/D
fluid bed
roasters. All
roaster gases
co 275 T/D
acid plant.
Tall gas to
60 ft stack.
Sintering machines
No. of
opcr/
atby
None
.
' /
3/0
1/0
(
i
Gas stream
control
equipment
.'
On rotary
kiln <-.cd
to remove
impurities
from roas-
ter cal-
cine.
Down-draft
Used in
series.
Sinter flow,
126,000 .
SCFM p.isscs
tiiru bay-
house (907^-X
Join roaster
gases, then
to 400 ft
stack.
Down-draft.
All sinter
gjses to ESP
(947.) and
151 ft stack
Kcrluc-
t ion
equip-
ment
Electro-
lytic
.
None
(ZnO
plant)
6400
hori-
zontal
retorts.
No con-
trol of
mis-
sions.
5824
hori-
zontal
retorts.
No con-
trol of
emis-
sions.
Stack data
No. of
I
(tail
gas)
1
(tail
gas)
1
(roas-
ter.
sin-
ter)
1
(tall
gas)
1
(sin-
ter)
Height
(ft)
300
200
AOO
60
151
Flow rate
(saw)
20,000
11,550
•
336,000
17,500
28,000
S02
cone.
(ppm)
2000
2051
4000
1600
SOOO
Emission rate
Partic.
(T/D)
~0
-0
4.4
~0
0.46
S02
(T/D)
4.9
2.9
162
3.4 ..
17
-------
Table 11 (continued). EXISTING U.S. ZINC PROCESSING PLANTS,7 1974
Corpany
KJXO/
location
J) BunV.cr
Hill/
Kellogg,
Idaho
6) St. Joe/
Xotuc*,,
p..
7) Sew Jer-
Age of plan:
First
year
1926
1938
1899
sey Zinc, j
Psloeirton,
?».
•
Last
modifi-
cation
1968
(Second
Acid
Plant)
1972
(2 ol-
der
acid
plants
being
replaced
by one
new acid
plant)
.
1970
(Acid
plane
mist
scrub-
bers
Install-
ed)
Materials
Cone.
T/D
550
875
520
Zinc
(T/D)
300
640
315
' Production equipment
Roasters
No. of
opcr/
stby
4/1
9/0
2/0
Gas stream
control
equipment
Five flash
roasters,
four 120
T/D and one
350 T/D
units. All
roaster gas-
'cs to two
ecid plants
(300 T/D
and 350 T/D),
on "as nccJed"
bos is, other-.
wise to 250
ft stack.
Nine roasters:
5 nultihcacth,
3 fluid bed.
and 1 flash.
All roaster
gases to six
acid plants.
Tail gases
Join sinter
gases, then to
400 ft stack.
Two 275 T/D
flash roas-
ters. All
roaster gases
to two acid
plants (500
T/D total
cap.). 40,000
SCKM toll gas
Joins sinter
gases, then
out 300 ft
stack.
Sintering machines
No. of
opcr/
stby
None
10/0
1/0
Gas stream
control
equipment
.
• Down-draft.
All sinter
gases to
settling .
flue and
one taphouse
and three
ESP's in
parallel,
then Join s-
cid plant
tail gas.
then to 400
ft stuck.
(Overall cf-
fic. is 94-
967.)
Down-draft.
All sinter
gases
(103,000
sent) to
ESP, JoiVt
tuil g.-is,
then to 300
stack. (97Z
total con-
trol)
Red uc*>
tion
equip-
ment
Electro-
lytic
'
'
17 elcc-
trother-
mic fur-
naces.
No con-
trol of
emissions.
43 ver-
tical re-
torts.
Retort
combustion
gases are
rccircu-
lated to'
supplement
fuel re-
quirement.
Stack data
No. of
2
(tail
gas)
1
(used
only
when
acid
plants
in op-
eration
1
(tail
gas,
•inter)
1
(tail
gas,
sinter)
Height
(ft)
75
250
400
300
Flow rate
(SCFM)
47,000
0
S02
cone.
(P?m)
2000-
3030
0
Emission Rate
Partic.
(T/D)
~0
S02
(T/D
14.4
0
(when «cid plsnta continuously used)
•
•v .
'
489,000
126,000-
140,000
.
*
860
1200
.
1.79
0.12
53
20
-------
plants in 1974. The facility was operating under a variance from the
8
State of Texas until May 30, 1975 when it was shut down. In all the
other plants the zinc and cadmium fume vaporized in the roaster is col-
lected by fabric filters or electrostatic precipitators. The collected
fume can then be further processed to recover cadmium or returned to the
zinc flow line for further processing.
Sintering is carried out at all pyrometallurgical plants. This is done
to remove the remaining sulfur and form a suitable feed for the reduction
furnace. This source is probably the major potential source of cadmium
emissions and in all cases is controlled by high-efficiency collection
devices, electrostatic precipitators and fabric filters.
The reduction of the sintered zinc calcine is another source of cadmium
emissions. With the closing of the Amarillo facility, only one domestic
plant employs horizontal retorts (approximately 5,800 units). These batch
type retorts are externally-fired and are not amenable to particle control.
They are inherently inefficient units emitting up to 10 percent of the ini-
tial ^cadmium content of the ore, and replacement of these units may be
anticipated in the future. The remaining plants have either continuous
or semi-continuous type vertical retort or electrothermic furnaces, which
are not a major source of cadmium provided good operating practices are used,
In addition to the plants listed in Table 11, the refurbished American Zinc
Plant (now owned by AMAX) Construction plans have been announced for a new
180,000 ton per year electrolytic plant to be built in Stephensport,- Ken-
tucky, by ASARCO, and the New Jersey Zinc Co. is planning construction
of an electrolytic zinc plant, capacity 160,000 tons, along with a zinc
oxide plant in Clarkesville, Tennessee., These plants will increase sig-
nificantly the present production capacity by the electrolytic process when
they come on-line in 1977 and is indicative of trends in process selection.
As mentioned above, air pollution" cdn'trol measures are simplified
and reduced in cost by the electrolytic process so it is the most prom-
ising from the standpoint of S02 and particulnte control.
47
-------
Emission Estimates
The primary zinc industry normally utilizes particulate controls on all
but the reduction/distillation operations. This last operation is not
considered to be a major source of cadmium emission since the bulk of the
cadmium is removed in earlier operations. The control measures used for
the other operations are highly efficient for the fine particulates
formed in vaporization and condensation processes (electrostatic precipi-
tation and fabric filtration). Nevertheless, previous estimators have
considered the primary metal industry, and particularly the zinc industry,
as the largest emittor of cadmium to the atmosphere.
\*
Davis ' emission factor for zinc plants of 284 pounds of cadmium per ton
of cadmium processed was based on a material balance obtained from pro-
cessing company data in 1968. Other investigators have either relied on
material balances derived from Bureau of Mines statistical information,
•*.-
or have accepted Davis's original estimates. EPA's estimate, based on a
material balance, v?as 310 pounds of cadmium per ton of cadmium processed
in 1971. The use of material balances does not appear to be entirely
suitable because of uncertainties in the input data. Application of this
technique for the years 1968 to 1973, using best available Bureau of Mines
data, leads to wide yearly variances and, in recent years, a drastic
lowering of apparent emissions. This trend may be real, reflecting secon-
dary process changes to maximize cadmium recovery and the elimination by
shutdown of many of the more inefficient smelters. However, a 10 percent
reduction in the assumed average zinc-cadmium ratio of 200 can reduce
emissions by as much as one-third following the EPA materials "balance
calculation method and the basic flow data for 1971. An error of this
magnitude in input data is not at all inconceivable.
The total annual estimated emissions of cadmium from zinc processing were
750 tons for 1968 by Davis and 640 tons for 1971 by EPA. The data in
Table 11 succests that approximately 2500 tons of total particulates per
-------
year (7 tons per day) were emitted from the zinc processing plants in
1973. Based on percentage values of cadmium measured in the emissions of -
q
the sintering process of 10 percent by Purdue University and even lower
values (less than five percent) reported by Chizhikov for gas steams
from roasting and sintering operations, total cadmium emission calculated
from the data in Table 11 would be around 200 tons/year.
If it is assumed that control measures are comparable (they are likely
to be more stringent due to particulate and S0~ air quality attainment
strategies), then more recent emissions could be estimated by assuming
that the cadmium emissions are directly related to zinc production.
With 1974 slab zinc production down 38 percent since 1968 and 17 percent
since 1971, 1973 emissions would be estimated at 470 and 530 tons per
year using the Davis and EPA study, respectively. These are comparable
and have led to GCA's estimate of 500 tons per year in 1974 given in
Table 9.
Anticipated Trends in Emission
Emissions since 1974 can be expected to be decreasing in the immediate
future to coincide with" the decreasing production of slab zinc noted in
the first half of 1975 as- well as further requirements for pollution
control. In particular, the closing of the ASARCO plant in Amarillo in
June 1975 is expected to have greatly decreased the total emissions from zinc
smelters. This one plant was responsible for 63 percent of the total
emissions from the plants listed in Table 11. While it is expected that
some of the capacity from this plant will be picked up by the other
existing smelters, the greater degree of pollution controls would mean
substantially reduced emissions by the end of 1975 and in 1976, perhaps
as low as 300 t_o.ns per year..
With another 340,000 tons capacity coming on line in 1977, emissions
would increase to some extent but not substantially under the New Source
-------
Performance Standards (NSPS) recently promulgated,, Since consumption is
currently twice the production level and two companies have seen fit to
construct tv;o new plants, it is expected that zinc slab smelting will
rise in the future, However, under NSPS and using electrolytic plants,
emissions will probably remain less than 300 tons per year through 1980.
PRIMARY LEAD PROCESSING
The primary lead industry is a minor source of cadmium but a significant
source of cadmium emissions. Lead ore, originally containing 3 to 8
percent lead, is delivered to the smelter as a concentrate with a lead
concentration of about 70 percent. This concentrate contains cadmium
which is emitted from sintering and reduction operations at the smelter.
Cadmium concentrations vary with a value of 331 ppm in the 70 percent
concentrate being used by EPA in their estimates. As with-zinc and all
other cadmium containing process streams, more specific estimates of
cadmium concentrations in ores and flue gas should be obtained if ac-
curate emission estimates are to be developed.
Emission Sources and Control Technology
Cadmium emissions from primary lead operations result primarily from the
operations of sintering and lead oxide reduction. Most of the cadmium
and S0_ is removed in the sintering operation, which converts the lead
and sulfur to lead oxide and sulfur dioxide, A flow diagram for the pro-
cessing of lead was presented as part of Figure 8.
Currently, six lead smelters are being operated by four companies in the
United States (see Table 12). Three of these are located in Missouri,
two in Montana, and one in Texas. Lead production from 605,000 tons of
domestic ore in 1973 was augmented by an additional 10-15 percent from
foreign ores. All but one plant utilizes updraft sintering machines.
50
-------
Table 12. EXISTING U.S. LEAD SMELTING,11 1973
i
^.. .-arty 1 A?ft
::!.-e-'i;-.«:;oo l;'ecr
1) ••_-.>.«r Kill/ • 1917
uellojf, Icihi:
;
1
•
1
'
f
.
!
.
of plant
Lust
notification
M70
(AC id Plant)
.
': j
j
2) y.'.Mvtrl Lea* ; 1566
C;t.-*tir,» Ca./i
J
i
j
3) S:. Jc«/ ! 1892
H.-ri-ljiK'UR, j
-
1972
(Sew b-g-
y.i. ' ! house for
j fugitive
* «ais*ions)
;
i
!
i
j
i) A3A.-.-0/ ': 1883
,
•. Xe!- n*. ! i
1 '
* i
i
1
Materials
Cine.
(T/U)
SIS
810
74«
400
U-.nl
(T/a)
350
380
SSO
105
I
i
^ privJuct ion equip.'.ont
Sintrrin^ machines
llo.of
oper/
otiy
1/10
1/10
*'?
\\
Can strcara control
cqui pnent
Updraft wintering
r'.iciiinc pro-luces
two gas streams:
Strong', i;a« sueam,
30,000 to 50,000
SCFM.. to 300 T/D
Ras strcan, 30,000
to 10. dim sew,
joined with blast
furnace and hygiene
air.
Updraft c Into rinj;
Ri.'iclilnc prttluccs two
F.33 strejir»: strong
£...; strc-.iw (26.000
sera) to 200 T/D
ucid plane. Weak
g,ns stream (fcC.OOO
SCI'M) Joined with
blast furnace I'.jscs.
•i
L'p'lrjlc sintering
michino proaucrii two
\ MS i.t;'c*jn-?: strong
I/O
Kan utrcan (23,OJO
SCFM) lo 300 T/U
ac>:l plunl. Wcuk eat
strea:.i joins other
P.1I-..3, then thru baft-
ho':sca and 352 ft
et.ick.
U?.a::rs, then to bag-
hojse and out 200
ft stack.
Blast furnace cases
Join siiUcr wcuk
i;a^rs and other
leases, pass thru
ba>;!toi:scs and 352
ft stack.
Bl.ist furnace gas-
es j-Sin rovnrb and
ver.tilaliun ,>bu3
(total l.SO.OJJ
thru three bag-
houses in parallel
i
Stack J.itj
No. of
1
(tail-gas)
1
(sinter-
weak ,
blast fur-
nace,
other)
1
(tail-gas)
1
(jtntvr-
we *IK , otasc
iuvnacet
other)
1
(tail-gas,
sinter,
blast fur-
nace.
other)
• 1
(alntcr)
3
(blast fur-
naco
other)
Height
(ft)
75
200
.
*
30
200
3S2
420
Flow
rate
(sere.)
30,000
316,500
22.250
260.000
Cone .
(P.--)
2000
5>0
.
•
2950
Er.istlcTi rate
Pirtic.
(7^
-0 ! 7.5
-0
ISOO-i 0.25
2500 i
356,000 4600
0.43
8
*
—
:oo
i • ;
150,000
195.000
I
21,505
27.703
325
1.1
0.46
440
7.S
-------
Table 12 (continued), EXISTING U.S. LEAD SMELTERS,11 1973
Conpany
Vtaef location
5) ASA7.CO/
Clover, Ho.
6) ASARCO/
El Paso,
Te.ee
••
first
year
1968
1887
Last
modification
«
•
o
M«t >«r 1 • 1 •
Cone.
(T/0)
260
220
Lead
(T/0)
222
(Capa-
city)
150
Production equipment
Sintering machines
Kn nt
oper/
etby
1/0
6/0
GAS fttrttuni control
oqulpnont
Updrnft sintering
machine. All
Rdnc.R to wntcr
spr;»y onti bng\iouKC»
tln.n out 610 (t
•tack.
DnvnJrnft alntcr-
Ing machines.
Cooes treated by
gcruhbcrs, • »ptay
choinUur, and a bag*
house (99.67.), then
out 610 ft alack.
Blast lurnaccs
»f A of
opcr/
• tby
I/O
2/1
Cut Btrcmn control
equipment
Blast furnace gases
to water spray.
bnr.hmi.ic, and three
58 It sticks.
Blast furnace and
dross furnace
gases mix, then
pats thru a spray
ch/imbcr and a bag*
house (99.97.),
then out six 108 ft
•tack*.
No. of
1
(sinter)
3
(blast
furnace)
1
(sinter)
6
(blast &
drona fur-
Keltic
UO
610
56
610
108
rl .«,
rair
(SCFM)
230.000
125.700
300,000
160,000
/"
**O ^
fc * »
Cone.
(?rr-)
12.1.00
U.200
UOO
::oo
1600
rarlic.
(TO)
o.o;
0.1 06
0.1
O.I
so:
(7/?1«
3t5
*-
»»
19C
2?
is)
-------
These updraft units are superior in most respects to the older downdraft
system and allows stronger S0_ streams to be generated. Although acid
recovery is practiced by only three plants (approximately 75 percent of
capacity), all utilize highly efficient collection devices on both sin-
tering machines and reduction blast furnaces.
Emission Estimates
•
Davis1 initial estimate of cadmium emission for 1968 from both copper and
lead operations was of the order of 300 tons per year, using an average
emission factor of 1310 pounds of cadmium per ton of cadmium in lead pro-
cessed. Mitre estimated emissions of 55 tons from the primary lead in-
12
dustry and EPA estimated emissions of 163 tons in 1971. In EPA's
estimate, 50 percent of the cadmium was assumed recovered in subsequent
zinc operations and the rest was lost to the atmosphere.
GCA has estimated that cadmium emissions were 75 tons in 1973. This es-
timate was made by assuming that 500,000 pounds of cadmium were recovered
..(
in 1973'from lead operations, and that the cadmium content of the total
particulate emissions of 1300 tons per year as given in Table 12 is
approximately 6.5 percent. This latter value is the highest reported by
Chizfiikov for dusts from-lead smelting operations. The average concentra-
tion for nine analyses from five plants was about 3" percent. Overall
efficiency of cadmium recovery from lead operations is about 75-80% based
on the GCA emission estimate. Since 1974 refinery production of primary
lead was 89 percent of the almost record high year of 1973, 1974 emissions
would be around 65 tons.
Anticipated Trends in Emissions
The lead industry appears to have been relatively stable over time with
1973 production only 6 percent less than the record high year of 1929.
A slight growth is anticipated, but because of present overcapacity
53
-------
(85 percent utilization in 1971), it is doubtful that any new smelters
will be constructed in the near future. The lack of anticipated growth
means that the New Source Performance Standards promulgated for lead
smelters will have no immediate impact on either the lead industry or
cadmium emissions. Future lead plants would probably have to employ
.fabric filters rather than electrostatic precipitators to provide maxi-
mum particle collection, '
PRIMARY COPPER PROCESSING
The primary copper industry is not a source of cadmium metal for consump-
tion but is a major source of cadmium atmospheric emissions. Most copper
smelters are located in the western United States near the major sources
of domestic ore. The primary copper industry generates cadmium emissions
in essentially the same way as primary zinc processors. However, copper
ores contain much lower concentrations of cadmium than zinc ores' so that
the concentration of cadmium fumes generated-during roasting and sintering
is relatively low. The cadmium content of 25 percent copper concentrate
is reportedly 40 ppm, too low for economic recovery. As opposed to the
lead industry where cadmium is sometimes obtained from lead smelting dusts,
the primary copper industry does not produce a flue dust specifically for
cadmium recovery.,
Emissions Sources and Control Technology
In the United States two conventional copper recovery processes are used.
One process involves roasting, smelting, and converting operations and is
normally used to process ores of varying composition. The other process,
used with ores of more uniform composition (green feed), omits the roasting
operation.
In 1973, a record high 1.7 million tons of copper were produced in 15 do-
mestic smelters. Seven of these facilities utilized the roasting process,
four in multiple hearth and three in fluid-bed roasters. The reverbatory
54
-------
furnace performs the smelting operation at 14 of the 15 plants; an elec-
tric furnace is used at the other plant. All 15 plants utilize Fierce-
Smith converters to process the copper matte obtained from the furnaces
into slag and blister copper. Processing information, pollutant control
information, and estimated emissions are summarized in Table 13. The
process flow used by existing domestic copper smelters is given In
Figure 10.14
•
Copper smelters emit more than six times as much SOjj and more than 10
times as much particulate matter as are emitted by the primary zinc and
lead industries combined. This is indicative of a much larger volume of
ore processed (3 times) but also reflects a lower level of application
of SC>2 and particulate control measures. Overall SO? control is estimated
to be 31 percent. Particle control measures are somewhat more effective
and electrostatic precipitators are replacing older less efficient con-
trol devices.
Emission Estimates
*
Based on Davis1 emission factor of 1310 pounds of cadmium per ton of
cadmium in copper processed (65.5 percent loss) and the EPA estimated
levels of cadmium in copper of 160 ppm, emissions of cadmium from copper
4
processing xrould have been about 180 tons in 1973. In EPA's own estimate
of emissions from copper processing, it is assumed that 90 percent of the
cadmium is lost to the atmosphere and the rest is in the slag. According
to EPA, this amounts to 234 tons per year; actual recalculation of the
data implies a value slightly higher, around 250 tons per year. Mitre has
•^
estimated 383 tons per year.
•
On the basis of the estimates of total particulate emissions given in
Table 13, the EPA estimate would correspond to a cadmium content in the
flue gases of about 0.8 percent. This is appreciably higher than .any
values reported by Chizhikov and is also much higher than would be
55
-------
Table 13. EXISTING U.S. COPPER SMELTERS,13 1973
ON
Ccr.-jny
N'«~e/icf ft len
Age of nl.nt
F'.tn
year
i) ASv.CO/Ttcoo., ! 1590
b'tsh.
2) ASA-CO/tU)t:»n,
Ariioni
J) ASAKCO/El P«-
09, Tcnii
4) rheljB Do
t*r l£«IT
• eld
pUnt)
Materitli
Cone.
(T/B)
1200
2000
700
•
2260
2111
BlUtrr
CU
(T/3)
303
(Elcc-
troly-
Clcilly
re lined
Cu)
366
*
260
(Anoda
Cu)
361
(Anod<
Cu)
470
(Anotit
Cu)
f'rixtitcl inn <-t;iii •w-n'nt
Ru0*ter»
Ko.ol
O;HT/
tiny
3/i
7/S
4/J
17/7
1/0
Gdt Utrcjrt vontrol
Roaster Rases
Join reverb Ranee
prior to treat-
ncnl.
*jr.M>lt til,; IlUI»;Ki-S 1 CotlVCTtlTS
No. of
Op IT/
• I by
1/1 (Ik'Virb f(.iv co.:iUim:di 3/1
Ro.iNtcr £jses 2/0
Join rcvurb gj««a
;>rtor to treat*
nent*
Roaster K.I so s
CIO. 000 SCFH)
pjtis thru set*
tiinc fltic, then
Join rwcrb
CflM"**
f
Rojsler gns.-s
treated by F.SP
and then Join
reverb tat,c*t
All coaster
gotci to Par-
tons 750 T/D
acid plant.
1/0
will' rpj.'itiT I'.jscM ,
iht-n to 2 KSP'k in
titrricb (£a.4?.)
St«X c.«,
i
Nn. of
li.'. ol 5,s 10 1053 1
"oiuianto 150 I/O j(t«ll (Jl,
oclU pl^nt, tail i rojstcr.
v.as to r^M.j«fi thru kcttlinfi
-
ro.uacr O'M's* tlton 1
] 10 spi jv i.!:.i:.,SLT,
IRSf
(sere)
so:
Cent.
:ri,.!M ,,:.
?jr:!c.] SO;
'
JOI) (403.000 5SC3
JO
i
1.5 j 2ii
I
i
'81,2C3i2230
-0 J2l.t
1
1
1
: i
o2(i 17JO.OC3J4500
j
i
eonvcrtct):
1
(ro*ltcr.
fjrr.acc)
1
(convert«r)
I
:»tl gat.
,'umice)
1
( :or.vtrter)
5U
).}
400
!
1
470,0301 15.700
i
'
566
600
600
:w.ooo'is.2co
110.000
406.000
14.030
15,400
•
902
1
1
U.7
o.<
* 1 5
7(1
-------
Table 13 (continQcd). EXISTING U.S. COPPER SMELTERS,13 1973
Cnr?«v
;:*-c/.oc«: ion
AJo, Arii.
7) x.jr*/
San K^nutl,
Aril.
ft) Kfr.ntcott/
Hurley, K.«.
9) Xer.necott/
McCIll, Key.
A»x of pljnt
1
First
>-«jr
i Q(r
1956
1939
1907
Ust
KoJiiicatlor.
I 0"* t
(DMA 4 «clt:.y
i jn
2/0
1/1
270
•
1 Na.of
<:.iv stri'ai.1 coi-.lrol > u?i't/
rii.il|.r.v;:l
sll.y
tonviru-r Kasca,
tht-n lo KSI' attu
jutl It sl.ic«.
Kfvcrl) »:.ru trci-
ivJ in W.ll.ll. «n«j
r.si' <«r'.). thi-n
nut il'j It «ljck.
Rfvcrb p,.i:;e:i trca-
lr.1 l.y W.II.U. hil-
loop iluc, CiP
(957.^, anJ 500 ft
Sli.j'r. .
iv^vecb cases trua-
ccJ hy W.II.U.,
ESP qti;;iMnt
liivn to i:SI' i,nd
3uO ft ttaik.
Converter ;>uos
p,,si,J i!iru £Si>
(V>?.), ihi'n
lo S50 ft sue*.
C..I.V.TUT )•.••"•
trv-iU-C In 35*
dull icloriC ai>d
p.i*srti out 626
(t stuck.
Converter gasoa
trraltJ by mul*
ticlor.c jnJ
balloor. tlue
(bi%) ou: 640
ft ItiC*.
Stuck d*t< ' Lr.l«»lar. ra:«
i
j
rCcSght
No. of
i
(furnare,
conver-
ter)
1
(furntct)
1
(conver-
1
(furnace)
(convor*
tcr)
1
(ft)
•kifl
SIS
soo
626
100
(furnac.V
1
(convtr*
t.t)
340
"**tav
rjic
so:
Co-:.
(SC.V.) ; (??=:)
1Q7 030 Tf.ift
I
Panic, i Stj
C.K) ;i:.'i)
0 }
i j
175. COO
15, CCO
,
117 000
•81.000
•
90,009
261.000
271.000
1.79
54.000 | 0.54
26, CCO
, ,
311
767
i
i
0.71
!
35.000 ! 2.S
i
5600
11.300
210
I
I
9.1
US
1
3.9
1 .
-------
Table 13 (continued). EXISTING U.S. COPPER SMELTERS,13 1973
00
. O^..,
Niac/U.-etior.
10) Kcnntcott/
aUydtk, Airtu.
A« of ,M.nt
run
ye«r
19J»
1
i
i
U) Krnnteott/
Carll.W.
at oh
12) An*con4*/
An«con4fi,
KontanA
White Pinte
HUhl|«n
•
1907
i90»
WJ
t
Lilt
ojuifkition
ISiS
(Fluid b«*
roaittr
•nd .eld
((lint)
HfiS
(Removed
•nd con*
vcrtvd to
grccn-fcat
rttvorbo)
197J
(K«« ocU
plant
MitorUli
Cone.
all.tcr
Cu
(T/3)j (r/»)
1050
'
2100
1710
inn
tw
120
•
fSO
500
(Anode
Cu)
rr»l«c(i«. c.,..-,.Svot
ROiiMrr.1
No. or
cpcr/
i toy
UO
GA* atvi-.o control
rquiprvnl
All ru.istor i;j$r>
pa** iliru cvcloiico
and CSP (JO:1.),
thrn join 3C% of
vnnvcrtur £*s*»
«n«i pjm thruui^h
I9u6 Kontanta
7}0 T/B «cld pltnt.
1
Non*
Uont
Ro^ettr* cxi»tt
but (or drying
conctntratc only.
Siii-lt (HA ('ini.u r*
No.»!|
cftrl
i:by
I/O
1/0
(.OH it rr.t't lotitrol
«-«)Mi;-iwi.t
i
Kcvorli ^Biics trea*
tt-U by W.il.n.,
mlnrJ with Sl>; of
cotivmrr r.-^us.
,<.is:.fj Iliru KSP
CiW.) .nil bOO ft
Hack.
Urvcrb ^1,a•1^ In'a*
trd hy U.II.H. and
ESI1 CJOt), tl.cn
out «uo 410 It
•taclii.
1
\
'/'
I
/
\ /t
»*•
,*
Jlcvi-rb r.JSc$ p»n»
thru wolcr uprjy
cliitr-itx-r anJ due
thi'it join conver-
ter £.«*€*.
tcU U>* U.II.O. ond
iwo KSP'a in *Qrlc«
Join converter.
£t)SCftf *nd pan* out
SOO ft «t««k.
Convert cru
.•;>.' r/
*tby
2/1
I'-jn slrcan control
r(iuip.-k-nt
ronvrrlcr nose*
I"iM tliru balloon
Uno. i lien: 504
o/ ,;i»%r.s raised
vii it ri-vvrh p,;.sci ,
ju.i so;, of i,ur«
7/2
6/0
ll/l
t/i
t HMrr«. in t.'-'.'
('>'/',) -.»J au>ks.
tn ro.'..tii-r t,r;ci.
All toitvcrtiT
:..i.;m -:ntcr r.SP
(ftO-70*/.). Uti-n
ttijU-J in five
.-Aid pknti (KOO
T/li UjSOi), and
out five tcid
pl.,nt vL«ckt.
Convtrtrr. gaaei
join rtvcrb
r.a&cs, pa«s thru
.•.ft t tin; [|iic
dinl LSI1 (30.9X),
then ouc s:jck.
juin reverb '.
^aiiCN and o«»a
out 500 (C
•tick.
Suck ai.t
So. of
1
(tall g.<)
1
CfurnwCtt,
vcrur)
2
(furnact)
(uUa,,
>^r
1
(furnaci.
convtr-
t.r)
(furi.c..
convcr*
etr)
He 1 r,V.t
(ft)
103
600
tog
Ji.95
J7.6J
114
i*V
9JJ
*flft
JVW
'I'M
niv
' , .__
S3;
Cine.
(SC»:> j(pfa)
7S.COJ
1U.OCO
1107,000
»»,ON
1930,000
Ul.OOO
UOO
72JO
I WO
lh.l»B*W.> l»l.
?irtlc.
SOj
(I/I) | (T'«
-0
1.1
U
I
1
]
*.n in
i
G.4« 1 )]
(icid '
* Hf t
1 '••••• 1
: by*
jjOj
a«)
J3.5
790
,
;
« i
1
97
-------
Table 13 (continued). EXISTING U.S. COPPER SMELTERS,13 1973
\o
1 !
Co.- ?jnv , A.;e o! jl.nt
' i
M«.ruu
i
Pt%>(ifii-)n rnt-p11*. i
K'li-.UTtl
- • ! 1
aUitcrl.'.'o.ol
ifirjl I.** I iConc. CJ jO,ik-r/
::•-« :oc.«lo«!yo« , «oJt tic ..ties |(T/D)j (T/3) |stt>-/
14) Citlei j 195d
1572
Service/ (£02 treat-
Co;>;*rMll,
300
!T.
1 (
| !
1 1
i
11) Ir.ipir.tlon/ ' 1W 1 1572
filial. Arii.
j (ElfCl.
1
turnact.
(/?h.
conv. ,
•cid
pUoc
unaer
con>t.)
50
(40
300
•(
I/O
| Su. a! j '
C.I* *l ri'.ini control |l
.•i,,.ir,wiit
Realtor i;js.'.i (600^>
»vby | cq..i,.:..,..ll
I/O
Mcctric furnace
SCO:) tr.'jt.-J !iy K.»vi! r.iK wltli
cyt U'lirn , rc^'iiied { '' dthcr* pl.»ii(
witli converter
j,;,i»cb, Uifii lo wet
scrcbbt-r, Jjhicd
witit iron roj»:cr
!o[(-R.is, tuvn vo^ltr-
::j.u:
!
u^ur/ C;i* ittiv'j..i control
»li>y u^uijn^T.t t Ko« ot
1/0 All converter
j,-..i::cu (0-;i.cco
i i'-t'nt) 'oiti rj.»*
' IMii^ti, lucn Co ( iicr twitt. clicu
avid plji'.is.
;
scrubber, ii^i1, a;ivl ;
to (our .clil .nd one
| l^uU SO] pUntt.
Son.
.
i '
{to acid pUntt.
i
1
j
1
1/0 Riv.-rb, gn>e< 3/1
p»i thru W.H.S.'c.
CU«, an ! (:••;>_
ic:o-j -o • ur.u««
53:0 •
I
j
1
i
1
i
J !
J40.SCO
163,400
i
55C3
*
15,200
Unknown 76
i
Ur.xnuwn
i
-------
COPPER
CONCENTRATES
FLUID-BED
ROASTER
3 EACH
(3 PLANTS)
ELECTRIC FURNACE
I EACH
(I PLANT)
GREEN FEED.
(8 PLANTS)
REVER8ERATORY
FURNACE
26 EACH
(14 PLANTS)
MULTIPLE-HEARTH
ROASTERS
. 33 EACH.
(4 PLANTS)
CONVERTERS
54 EACH
(IS PLANTS)
BLISTER
"COPPER
Figure 10. Schematic flowsheet of copper processing
14
-------
inferred from GCA's recent experience with an air monitoring network at
the ASARCO copper smelter in Tacoma, Washington, where cadmium contri-
bution to total particulate levels was less than 20 ppm. Chizhikov
also reports cadmium content in the slag to be equivalent to 22-40 percent
of the initial cadmium reported, much higher than the 10 percent of
initial cadmium levels assumed by EPA.
While EPA's estimate is felt to be high, present knowledge is too limited
to allow precise estimates to be made. If it is assumed, as EPA does,
that the cadmium concentration is 160 ppm relative to the copper processed,
then emissions can be no more than the 245 tons as per the EPA methodology.
Based on Chizhikov's data and GCA's sampling experience, an estimate of
125 tons per year for 1973 emissions of cadmium from copper processing,
is considered conservatively high. Differences between GCA's and other
estimates can be attributed to differences in cadmium slag content or ore
content, both of which need further study and documentation. By extending
__the_ 10 percent decrease in domestic copper smeltjer_production between 1973
and 1974 to the emissions, an emission level, of 110 tons of cadmium is
determined for 1974.
•
Anticipated Trends in Emissions_,;
>
Currently (1975), copper production appears to have decreased further
since 1974, 10 percent less in the first half of 1975 than in 1974, so
emissions would be expected to decrease accordingly. Falling .prices,
from 85 cents per pound in June 1974 to 60 cents in June 1975, would also
signal a decrease in production. Since 1973 was a record high year for
*./
cadmium production, it may be assumed that that year's emission level of
125 tons per year is the maximum expected until the economy calls for
•
more copper, probably through increases in the construction and automobile
market. The recent turnaround noted in 1976 model car sales may help the
copper market. If new facilities are called for, they would come under
the recently promulgated new source performance standards.
61
-------
PRIMARY CADMIUM PROCESSING
The major source of domestic primary cadmium is the zinc dust recovered
from zinc roasting and sintering operations or precipitated from zinc
sulfate solution prior to electrolysis. Additional sources are slags and
'flue dusts from other primary metal processing operations and distillation
processes to purify slab zinc from pyrometallurgical operations. Some
330,000 pounds of flue dusts were also imported from Mexico in 1974, a
level twice that of 1973 but two-thirds the normal imports.
Cadmium Recovery Processes
The recovery of cadmium is accomplished by a variety of treatment processes
designed to separate cadmium from other metals such as zinc, lead,
thallium, copper, etc. Most of the techniques used are wet process me-
thods, although some purification is accomplished by distillation.
Figure 11, as presented in the ORNL study of cadmium, diagrams cadmium
recovery from a variety of cadmium containing process residues.
The extent of cadmium recovery from zinc and other ores is not. well quan-
tified. It has been estimated that actual recovery is of the order of
60 to 75 percent of the total cadmium in the ore. Since many of the
secondary processing activities used to increase cadmium yield are econo-
mical only in times of high cadmium demand and prices, the extent of re-
covery is largely dependent upon economic matters and the type of process
used.
Emission Estimates
e
While previous recovery rates of cadmium have been reported to run as high
18
as 90 to 95 percent from raw ore to metal, present technology appears
capable of achieving around 98 percent recovery of cadmium from concentrate
ores. EPA's estimate of current recovery rate of 98.5 percent appears to b
62
-------
AMMONIUM CHLORIDE (NH.CD
i *
dHzS04 AND
NoClOj
ZINC ORE j
ROASTING
• AND -
SINTERING
DUSTS
IN
TO
LEAD AND ZIN
FUMING AND
LEAD AND
COPPER
SMELTING
FLUE DUSTS
<5%Cd
L
j
A
ROASTED
ZINC ORE *
MPURE
SLAB ZINC
ULFURIC ACI
•LEACH
SOLUBL
LEAD
C SLAG
FLUE
.E PbS
SMELTI
OUST
UPGRADING
RESIOL
EAD S
•ULFUR
NO Zlf
IES TO
WELTEf
C ACID
JC DUS
COPPER
REMOVAL
COP
RESl
9 SOLUTION
AND
_ IMPURITIES
°«
:RS
0
ARSINE
•i
35-55 %CdV
»S L(
T
Cd-Zn
SPONGE
1 Zn DUST
pAND MjSQ,
ZINC
REMOVAL
ZnSO«
ZnCU
RE Sll
CAiH,
JARS
AND
• TO Zn
4TCRIN
HAZAF
:NIC
"I REMOVAL
Cd METAL
SPONGE"
!| CARBON
flANO CoO
CADMIUM CADM
DISTILLATION VAP
\ |
*} RESIDUES TO Zn
! ORE SINTERING
G j OR Pb SMCLTERS
>D 1 <
AND KMnQ,
A«-FRE£
SLUDGE '
[ F Y**~ ATMOSPHERE
IUM Cd VAPOR
OH CONDENSATION
RESIDUES TO Zn
ORE SINTERING
j
r
i
IMOLTEN
{IFOR Aifl
d HEMELTlNG
&NO CASTING
THALLIUM
i, 1 HN02SANOC.O HD.CHROMATE
'it !
TREATMENT Cu.p
CALCINATION SLl
V 'it
ARSEMC TO COPPER-LEAD RESIDUE
tAO SMELTERS TO LEAD SMELTERS
i1 j
LJ f> /\ £«rivl/Uw>
' p" F2 1 ; 1
i
• n^in^Tinu
PER ZnS04 TO Z
-------
high. However, the majority of the processing is by wet methods, so this
recovery rate may be used for determining a conservatively high level of
cadmium emissions from cadmium recovery plants - 50 tons in 1974 based on
the U.S. production level of 3510 tons of cadmium,
Anticipated Trends in Emissions
As discussed in Section II, cadmium production in the United States has
been declining relatively s.teadily since its high in 1969. The corres-
ponding trend in emissions may be expected to continue with the closing
of the ASARCO plant in Amarillo, although other plants are expected to
pick up the slack. The continually fluctuating price of cadmium makes
actual predictions difficult and such fluctuating similarly means that
additional cadmium processing facilities will not likely be constructed
immediately. Therefore, emissions from cadmium processing plants are
not projected to rise above the figure of 50 tons per year applicable
to 1974, Rather, emission levels will decline or fluctuate around some
lower value.
With the projected increase in zinc smelting capacity in 1977, some in-
crease in cadmium production would not be unexpected to help make up
the difference between cadmium consumption and domestic production.
While not directly pertinent to emissions from cadmium recovery processing,
an interesting relationship exists betx^een cadmium/zinc demand and cad-
mium emissions. Since cadmium is recovered from zinc smelting operations
for its commercial value, a decrease in cadmium demand would result in
increased cadmium emissions due to lack of economic incentive to recover
the cadmium. This is a very important relationship when strategies for
cadmium emission control are considered. Since there is no great economy
in reducing the efficiency of cadmium recovery processing, increasing
zinc demand without similar increase in cadmium demand would likely lead
to higher cadmium emissions, while decreased zinc demand with the same
level or higher cadmium demand should effectively reduce cadmium emissions^
This relationship would also be applicable to the other primary metals
64
-------
operations that produce cadmium as a by-product, e.g., the primary
lead industries.
REPROCESSING OF IRON AND STEEL SCRAP
The iron and steel industry is responsible for large emissions of cadmium.
This is because plating has traditionally been the largest single use for
cadmium and because a portion of that originally-plated metal recurs as
steel scrap for recycle. The same is true for galvanized steel, vhich has
some cadmium associated with the zinc plate.
Emission Sources and Control Technology
i -
The cadmium emissions of the steel industry result from operation of the
open-hearth furnace, the basic-oxygen furnace, and the electric furnace.
The choice of steel-making furnaces is normally dictated by the end-product
since various alloy types and grades are^better "suited to specific furnace
types. Cadfhium emissions are usually generated by the processing of number
two scrap in the above furnaces, with the open-hearth furnace and electric
arc furnace being used to a greater extent for this cadmium-bearing scrap
grade. $ureau of Mines statistics radicate, However, that in 1970 essen-
tially identical amounts of stee'l s'cra'p were processed in all three fur-
nace types.
•*s:
While most of the iron and steel industry processes are presently con-
trolled, the extent of control technology utilized is often low,, involving
in so.T.e cases the use of settling chambers or cyclones which are designed
to remove only the gross particulates (>40pi). However, in many instal-
lations, electrostatic precipitators are used to.remove metallic fumes
from gas streams and baghouses have also found specific uses. Baghouses,
electrostatic precipitators, and high energy scrubbers are all capable of
effective control of the small particles associated with metallic fumes,
with baghouses typically providing the highest efficiencies.
65
-------
Of all open-hearth furnaces operating during 1971, 41 percent were estimated
to be controlled, all of them using electrostatic precipitators averaging
97 percent efficiency. The same survey revealed 79 percent of electric
arc furnaces had controls.-with a greater variety of techniques being used
to control emissions. Most of the controls on electric arc furnaces
-involved baghouses while there were two wet scrubbers and one electrostatic
precipitator reported in use. Average control efficiency was estimated
at 99 percent for those furnaces employing controls. Resulting particulate
emissions were 337,000 tons and 18,000 tons per year for the open hearth
20
and the electric arc furnace, respectively,,
Emission Estimates
The reprocessing of iron and steel scrap is a particularly difficult
source to assess. According to Davis, it amounts to 1000 tons per year,
accounting for almost 45 percent of total U.S. cadmium emissions. ORNL
has considered emissions from this source in some detail and has con-
21
eluded that Davis* estimate is too high. ORNL takes issue with Davis1
assumption that steel scrap will have the same ratio of cadmium to steel
as the ratio of cadmium used for plating to total steel demand. Their
main contention is that cadmium plated articles will be avoided by steel
producers. ORNL also feels that control measures should have been taken
into account. The final ORNL estimate for losses from cadmium plated and
galvanized steel scrap remelting is 1,100 tons per year with only 10 per-
cent or 110 tons of this emitted due to the application of control mea-
sures. Although ORNL refined Davis' approach somewhat by use of the
ratio of cadmium plated to finished steel rather than total raw steel, it
did not change Davis' basic approach,,
o
EPA, with the assistance of Battelle, estimated the amount of cadmium
containing scrap processed by the three different types of refining furn-
aces employed in the steel industry. With this as the basic input, esti-
mates of emissions were made assuming control efficiencies of 75 percent ffl
66
-------
the basic oxygen and electric arc furnaces and 25 percent for the open-
hearth furnace. These assumptions led to an estimated emission rate for
cadmium from scrap reprocesing of only 78 tons/year despite the lower con-
trol efficiencies used. (The Battelle report was not referenced and was not
obtained, thus evaluation of this estimation method was not possible.)
None of these methods, essentially variations on Davis1 approach but
using Bureau of Mine Statistics for various grades of steel and scrap,
are totally defensible. Straight application of Davis1 approach and
assumptions of equal .amounts of scrap processed in the three furnaces and
of an overall efficiency of 65 percent gives a cadmium emission value of
about 400 tons per year. This estimate is based on a much higher recycle
efficiency than the EPA/Battelle estimate of slightly less than 5 percent.
GCA has elected to use this much larger value recognizing that this is con-
servatively high. Emissions may well be less than 100 tons per year as
estimated by EPA.
Anticipated Trends in Emission
Bureau of Mines projections of consumption of steel scrap for the year
2000 range from 32-46 million tons which is in the range of 1968 consump-
22
tion (0.89 to 1.28 times 1968 consumption). It is felt that even the
high range of this estimate may be too conservative because trends have
shown a major rise in the consumption of iron and steel scrap with a record
high year in 1973. Consumption in 1974 was only 3 percent less than in
1973 and may have exceeded the previous year if not for the coal miners'
strike. Further expansion is expected as"a result of the increased interest
in recycling by the public, governmentj and industry. Since 74 percent less
energy is used to produce a ton of steel from fe-rrous scrap than from iron
ore, energy and corresponding economic incentives will tend to call for
more processing of scrap. The International Institute of Iron and Steel
has studied the relationship of supply and demand for ferrous scrap and
concluded that there would be a shortage of scrap for future domestic
steel production.
67
-------
Growth of the secondary i.ron and steel industry is not an absolute in-
dication of corresponding cadmium emissions. Cadmium emissions from the
secondary iron and steel making industry are related to the type of
scrap being used and are determined by the total cadmium associated with
the scrap metal. Individual operations may be only minor cadmium emis-
sion sources if the operations deal in scrap of low cadmium content and.
vice-versa. A complete breakdown of the types of scrap that are processed
is necessary to accurately determine associated cadmium emissions.
Because many of the cadmium plated parts are small, consumer oriented
items, they do not have a high probability of recycle. They are more
likely to be disposed of as municipal trash and, as such, are discussed
under incineration.
SECONDARY ZINC INDUSTRY
t
Zinc recovery from old scrap historically is about 5 percent of the total
zinc consumed per year. Since most of the cadmium initially present in
the zinc has been lost previously, the secondary zinc industry is not a
large source of cadmium emissions. GCA has used the EPA estimate of two
tons per year.
SECONDARY COPPER INDUSTRY
The secondary copper industry makes no attempt to recover cadmium from its
copper scrap recovery operations. The principal source of cadmium appears
to be automobile radiators. For this application, 0.2 percent cadmium
alloyed with copper increases the strength of copper without serious ef-
fect on thermal conduction. The amount of cadmium available from this
source, assuming all radiators contain 0.2 percent cadmium, was 130 tons it"
1972. An additional 10 tons or so could exist associated with the zinc
(0.023 percent) in brass and other copper containing materials. GCA's
estimate of 70 tons per year was based on a 50 percent control of emissions
and is in good agreement with EPA estimates.
68
-------
MANUFACTURE OF CADMIUM PRODUCTS
Davis has estimated the cadmium losses from operations involving the
production of pigments, plastic stabilizer, alloys, batteries and other
miscellaneous cadmium products to total almost 19 tons. With the ex-
ception of the EPA study which reduced the total to 13 tons, all the
other investigators have accepted Davis1 estimate. In either case none
of these operations are a-sizeable source of cadmium emissions. Any
exposure would be restricted to those individuals directly engaged in
processing operations. To be on the conservative side, GCA has assumed
the same values as Davis.
INCINERATION
Incinerators are a major source of cadmium emissions due to cadmium pre-
sent in plastics, pigments, metal scrap, etc. Similarly open burning of
waste will, also contribute to cadmium emissions.- A 1972.report estimated
that about 6.7 percent of all industrial, commercial, municipal and do-
mestic solid, waste generated was incinerated and another 27 percent was
23
combusted in open burning., Al.ttxQ.ughL in.ainerators are controlled to per-
haps 'a 50 percent level, no controls-are possible for open burning. These
numbers are not well documented and are being, reduced significantly as a
result of grov;ing restrictions on combustion processes, and particularly
open burning. However, emissions from burning are of particular signifi-
cance because of the general proximity of incinerator emissions to munic-
ipal populations.
j
Because of the many uncertainties in the general area of solid waste
incineration a number of approaches based on different uses of statistics
and assumptions have been used to estimate cadmium emissions. Davis*
estimate of emissions was based, on assumptions that 40 percent of the
3.2 million pounds of cadmium used for plastics and miscellaneous uses
in 1963 goes into solid wnstc and 15 percent of all solid waste was
69
-------
incinerated. These assumptions result in an estimate, used by others, of
96 tons per year in 1968. The EPA study apparently used a Battelle esti-
mate of about 30 percent of the yearly total consumption of cadmium going
to solid waste and a control efficiency of 50 percent for the 5 percent
solid waste incinerated in municipal incinerators. On this basis the
EPA estimate for cadmium emissions from municipal incinerators was 48
tons in 1971.
Incineration was estimated "to be responsible for 142,000 tons of particu-
24
lates in 1968. The concentration of cadmium in the particulate emissions
25
from incinerators has been reported as being between 1000 to 10,000 ppm.
These two studies would indicate cadmium emissions from incinerators be-
tween 142 to 1420 tons for 1968. In 1968, there were an estimated 300
municipal incinerators in operation handling approximately 8 percent of
26
the nation's waste; by 1972, the number of incinerators dropped to 192.
Required controls on existing incinerators as part of the particulate stan-
dard attainment strategy in state implementation plans have caused other
closings since then and reduced emissions from others. Incinerators con-
structed or modified since June 14, 1974, are subject to New Source Per-
formance Standards, In addition, open burning has been banned in most
areas of the country„
A conservatively high estimate of cadmium from incineration of wastes may
be determined by assuming that 40 percent of the cadmium consumed for non-
electroplating purposes (includes pigments in paints, plastics, batteries)
and 20 percent of the cadmium for electroplating is generated as solid
waste. If 20 percent of this is incinerated and the particulate emissions
controlled to only 60 percent, then approximately 150 tons of cadmium per
year would be the result of incinerations
o
Incineration of solid waste has had a sudden surge of interest as a
source of energy. Several communities and the State of Connecticut are
undertaking major efforts to utilize solid waste as a supplemental fuel
70
-------
for generating steam. The greater volume of wastes to be incinerated
would tend to increase cadmium levels. At the same time, to the extent
that these projects also incorporate recycling, much of the cadmium on
electroplated wastes would be removed from the waste stream. The NSPS
for incinerators and the controls on power plants where waste is used as
a supplemental fuel would help to slow any rise in cadmium emissions.
Therefore, the value of 150 tons of cadmium per year (based on 1973 cad-
mium consumption) may be assumed to be a ceiling estimate for the near
future.
COMBUSTION OF FOSSIL FUELS
i
The trace quantities of cadmium in all fossil fuels is potentially another
significant source of cadmium emissions, especially from the less purified
fossil fuels, coal and residual fuel oils. Although cadmium is only pres-
ent in trace quantities in most fossil fuels, the huge amount of fossil
fuels burned in this country represents a large amount of cadmium in
{ '
absolute terms. In addition, the majority of fuel is consumed in highly
*
populated areas.
The importance of this source had not- been recognized prior to the study
by EPA. This analysis is well considered and has beer used by GCA for its
emission estimates for coal.
Emissions From Coal
Cadmium in coal ranges from 0.03 - 2.0 ppm. Cadmium emissions from coal
have been estimated by EPA for the classifications of usage listed below
and an assumed cadmium content of 0.5 ppm.
71
-------
Category
Utilities
Coke
Industrial and Commercial Burners
Coal
consumed,
106 tons
320
100
80
500
Cadmium
lost,
percent
50
75
50
Cadmium
emitted,
tons /year
80
38
20
138
Emissions From Oil
EPA also recognized the importance of crude oil and its products as
sources of cadmium emissions. At reported levels in fuel oil of 0.1 to
0.5 ppm cadmium and estimated petroleum fuel consumption levels of about
400 million tons, the potential exists for emissions of 40 to 200 tons of
cadmium per year. Since controls on oil burning sources are normally se-
lected to remove the larger particles, it may be assumed that essentially
all of the cadmium in oil is released to the atmosphere; i.e., 40-200 tons
of cadmium per year.
This indicates the possiblity of a large source, of cadmium emissions;
however, an accurate analysis of potential cadmium emissions from fuel
oil combustion cannot be made without first having accurate data on the
cadmium levels in all of the grades of fuel oils* One study showing
a rise in ambient cadmium levels during the period September to
27
December implies that combustion may be a major source.
SEWAGE SLUDGE INCINERATION •
Incineration of sewage sludge is another source of cadmium first considered
by the EPA group. Cadmium is concentrated by people and eliminated through
excretion. Sewage treatment operations tend to further concentrate the
heavy metals during sludge digestion and subsequent drying. Added to this
72
-------
are emissions of cadmium from the various plating industries that con-
28
tribute to the load at the sewage treatment plant. This makes sewag
sludge relatively high in cadmium and other heavy metals.
The burning of sewage sludge changes the immediate environmental media
that is contaminated with cadmium from water or land (landfills) to air..
Such emissions are normally located within urban areas and provide a more
wide dispersal of cadmium,- with resulting wider exposure.
EPA has estimated this source to be rather large, about 138 tons per
year. GCA's estimate, however, is over an order of magnitude lower re-
flecting differences in estimation of total sewage and cadmium content.
GCA has used ORNL values of 4 x 10 tons of sewage per year and 20 ppm
to give a total of 80 tons per year in sewage sludge. Assuming 15 percent
of this sludge is incinerated, total cadmium emissions become 12 tons/year.
MISCELLANEOUS EMISSION SOURCES ~~
*
In this final category of cadmium emission sources are those areas which
have been considered by other cadmium investigators. All are relatively
small sources and will not be discussed in any detail here. These cate-
gories include emis'S'toivs' frorar the following:
Category^
Emission,
tons/year
Motor Oil Usage
Rubber Tires
Gasoline
Use of:
Pesticides, Fungicides, etc
Fertilizer
< 1
6
< 2
< 1
< 1
73
-------
Other potential sources, not discussed because of lack of information,
include consumption of other petrochemical products and combustion ac-
tivities such as agricultural burning, forest fires, etc. These cora-
-------
REFERENCES
1. Purdue Study.
2. Mineral Industry Surveys; Zinc, Lead and Copper. 1973.
3. ORNL. Section IV-C.
4. PSP Document.
5. Mineral Industry Survey; Zinc. 1972.
6. ORNL. Figure V-3.
7. Draft. NSPS for Primary Copper, Zinc, and Lead Smelters. Background
Information. EPA.
8. Mineral Industry Survey; Cadmium. 1st Quarter 1975.
9. Purdue Study. Section I-B.
/
10. Chizhikov. Chapter 4.
11. Davis, ORNL, and PSP Documents.
12. Mitre Document, p. 26.
*
13. Draft. NSPS for Primary Copper, Zinc, and Lead Smelters. Background
Information.
14: Ibid.
15. Mineral Industry Survey; Cadmium. 1974.
16. ORNL. Figure V-5.
17. Davis and ORNL.
18. Stickney, W. J. Cadmium Extraction from the Ores of the Hudson Bay
Mining and Smelting Co. Con. Inst. M&M. Transactions 69:33. 1966.
19. Mitre Study. Table I.
20. Particulate Pollutant System Study. . Volume I. MRI. 1971.
21. ORNL. p. 93.
22. Mineral Facts and Problems. Bureau of Mines. 1970.
75
-------
23. Hydrocarbon Pollutant Systems Study. Volume 1. Table 111-14,
MSA. Publication Number APTD 1499. 1972.
24. Ibid.
25. Lee, R. E., Jr. and D. J. Von Lehmden. Trace Metal Pollutants in the
Environment. Journal Air Pollu Control Assoc. October 1973.
26. Phillips, N. P. and R. M. Wells. Solid Waste Disposal. Final Report.
Office of Research and Development. U.S. EPA. May 1974. p. 11-12.
27. Hunt, W. F., C. Pinkerton, 0. McNulty, and J. Creason. A Study in
Trace Element Pollution of Air in 77 Midwestern Cities. (Presented
at the 4th Annual Conference on Trace Substances in Environmental
Health. University of Missouri, Columbia, Missouri. June 23-24, 1970.)
28. ORNL Document. Section IV-D.
76
-------
SECTION IV
HEALTH EFFECTS OF CADMIUM
TOXICITY
Cadmium is an extremely toxic element. It has been given a toxicity
rating of 5 on a scale of 1 to 6 and, when inhaled, a rating of 6, or
supertoxic. Currently adopted standards for cadmium exposure include
0.01 ppm in drinking water (USPHS)* and a threshold limit value of
^ + '
50 MS/m .-(A.CGIH) „ This would indicate a toxicity between lead and
mercury with water standards of 0.05 ppm and 0.001 ppm and TLV's of
3 3
150 |ag/m and 50 ug/m , respectively; however, these standards are not
uniformly determined.^ In? additiojij-cji^roiium* s. longer half-time in the
body, in comparison to mercury and lead, and its residence in critical
organs, as opposed to lead's accumulation in the bone, further increases
its relative importance for chronic exposure.
%
Estimates of the daily dosage of absorbed cadmium needed to produce
renal tubular failure, one. of the best documented chronic effects, have
2
been in the area of 6.6 pg. * Since the average volume of air inhaled
daily is about 20 cubic meters and at least 10 percent of the cadmium
would be retained, this means that a concentration of 3.3 yg/m would
''The interim EPA regulations for drinking water have adopted this value
for the maximum cadmium concentration (40 CFR 141; 40 FR 59565, December 24,
1975).
+Based on notice of intended change for 1975 - American Conference of Govern-
mental Industrial Ilygienists.
77
-------
produce this effect. This result is much less than the range of several
studies indicating chronic respiratory effects from industrial cadmium
~ 3
exposures of 200 to 3000 jig/mj.
A lethal dose is on the order of 5 to 50 mg per kg of body weight (0.35 g
to 3.5 g for 150-pound person) when ingested and about 40 mg (assumed 10
percent retention) when inhaled. A time-related lethal dosage for inhaled
3
cadmium is on the order of 2900 minutes-mg/m for a thermally generated
fume and 1400 minutes-mg/m for the more toxic freshly generated arc
4
fume. This difference in toxicity is most likely a result of the smaller
particle size in the arc fume.
While these figures provide an idea of the potential toxicity of cadmium,
•
the actual effects which may occur at various doses of cadmium cannot
be so easily predicted. Antagonistic or synergistic reactions will
often result due to the intake, or lack thereof, of other elements and
substances in the diet.
ABSORPTION RATES
Ingestion and inhalation of cadmium are both means of uptake and even-
tual systemic absorption. Cadmium can be absorbed and retained to a con-
siderable degree after inhalation. Absorption is primarily from the lungs;
but, because cadmium is in the form of an aerosol, this absorption will
also be a result of mucociliary clearance and eventual absorption in the
gastrointestinal tract. Particle size plays a major role in the degree of
deposition in the pulmonary compartment which varies from about 10 to 50
percent for particles with a mass median diameter of about 5 microns to
0.01 microns. Though human data on actual absorption is not available,
animal experiments suggest that 10 to 40 percent of inhaled cadmium is
finally absorbed. Therefore, the actual systemic absorption of cadmium
from the lungs is quite high. Comparable results have been found for
particulates of lead and mercury.
78
-------
Initial absorption of ingested cadmium has been estimated to be as high as
50 to 75 percent, but because of cadmium excretion in the urine and feces,
actual retention is in the range of 3 to 8 percent. Mean estimates of
around 6 percent have been derived from many studies, including calcula-
tions based on the body burden of infants (1 pig) and "standard man" at
age 50 (15 to 20 rag in U.K. and Sweden and 30 mg in U.S.). The actual
absorption of cadmium in an individual is influenced by other factors
in the diet, including calcium, vitamin D, and protein. Under certain
circumstances, such as calcium and protein deficiency, a retention rate
of 10 percent or more is quite possible. Absorption rates for ingested
lead and mercury have a greater spread depending upon the specific chem-
ical compound. Initial mercury absorption varies from less than 0.01
percent for elemental mercury to about 95 percent for methyl mercury.
Lead absorption ranges from 2 to 16 percent.
TRANSPORT, DISTRIBUTION, AND EXCRETION
t
Tissue Levels
After exposure and absorption, whether by inhalation, oral intake, or
injection, cadmium will be found in the blood and the Internal organs.
"Normal" concentrations in whole blood have been found to be well below
1 yg per 100 m£. Though a few studies have reported concentrations as
high as 200 ^g per 100 mS. in "normal" subjects, concentrations greater
8
than 1 ^g per 100 mJL are only found where industrial exposure has occurred.
*s
The half-time for clearance of cadmium from the blood is on the order of
9
1/2 to 1 year.
The concentrations of cadmium in various organs that have been reported
are generally low and only a few correlations have been made with age.
79
-------
In the case of chronic cadmium poisoning, high values (25 to 83 ppm
wet weight) have been found in the thyroid. Cadmium in the pancreas
of exposed workers may reach values 40 to 80 times higher than the
normal (1 to 2 ppm wet weight) amount for U*S« citizens* Concentra-
tions in the lungs have been shown to increase with age to values of
12
around 0.25 to 0.40 ppm wet weight at middle age» However, the highest
amounts of cadmium, about 50 percent of the total body burden, are typi-
cally found in the liver and kidneys. Liver concentrations in U.S. adults
normally increase to between 1 and 3 ppm wet weight while renal cortex
levels at age 50 reach around 50 ppm. This concentration is about one-
half as high as that found in Japan, yet it is twice the levels in Europe
13
as shown in Figure 12. When serial sections of the kidney were taken,
the concentration Of cadmium was found to decrease from the outer
cortex inward reaching the lowest values in the cortex near the medulla.
Concentrations of cadmium in the renal cortex are two to three times that
14
of the medulla. After about age 50, the levels of cadmium in the liver
and kidneys decline slowly, due possibly to either lower exposures during
the lifetime or loss from the kidney.
Workers with high exposure to cadmium will usually have a higher per-
centage (up to 75 percent) of the body burden in the liver and kidneys.
Although about one-third of the total cadmium is normally in the kidneys,
after renal damage due to repeated high exposures only 10 percent of the
body burden will be found there.
./
Excretion
Excretion of cadmium occurs through urine, feces, and hair. Normal
excretion of cadmium via the urine is around 1 to 2 yg/day or less.
Because of the possibility of tubular dysfunction, estimates of urinary
excretion of exposed workers has not always been well determined, yet
recent studies have placed the value at less than 10 ug/g creatinine
(16 ug/day).
80
-------
1000-1 CADMIUM IN RENAL CORTEX
PPM WET WEIGHT
500-
00
100 r
10-
n RANGE
I KANAZAWA JAPAN
HE KOOE JAPAN
HE U.S.A.
UTU.K.
3C SWEDEN (2 AREAS)
50
75
ICO AGE IN YEARS
Figure 12. Renal cortex levels of cadmium found in human beings
(After CITE)
-------
Normal amounts excreted in the feces have been estimated to be within
the range of 31 to 42 g daily, with one Japanese study indicating 57 ^ig.
It is not known how much of this is cadmium not absorbed from food and
how much is absorbed and then excreted via the intestines. Some analysis
of hair as in indication of environmental exposure has been conducted,
but this is a relatively new field. The studies do seem to indicate some
good agreement between exposure levels and hair levels of. cadmium,,
No relationships between the various measures of cadmium described
above have been determined. Specifically, no accepted relationships
between cadmium levels in the blood and exposure time, degree of pro-
teinuria and blood levels, blood levels and urinary excretion, urinary
excretion and total body burden, have been found. However, there does
18
seem to be some relationship between daily dose and urinary excretion.
The principal cause for concern over environmental exposure to cadmium
is related to its long biological half-time. The half-time of cadmium
19
in the body has been estimated to be between 17 and 33 years. There-
fore, even continuous low exposures to cadmium or intermittent higher
exposures allow for a build up of body levels over a period of years.
This half-time is two to three times the half-time of 10 years for
20
lead and many times longer than the half-time of mercury, which is
21
on the order of days to months. This implies a much greater poten-
tial for effects resulting from occasionally high cadmium exposures
than from similar mercury exposures since the latter is cleared so
rapidly.
Further relative importance is given to cadmium as opposed to lead
since cadmium is residing in a critical organ x^hile lead accumulates
primarily in the bone. Figure 13 provides a graphical representation
of the distribution of lead and cadmium in the body as found in a
22
study of tissue samples taken at autopsies on New York City residents.
82
-------
H 5
X
o
id 4
=5
b
LEAD
n
20
l-
I 15
ui
5
ro
CADMIUM
J±I
LUfJG HILAR LYMPH LIV£R
NODES
KIDNtY BLOOD
BONE
Figure 13. Distribution of lead and cadmium in human
tissues under normal exposure ^
83
-------
EFFECTS OCCURRENCE AND DOSE-RESPONSE RELATIONSHIPS
That cadmium enters the body and accumulates there, coupled with its
known toxicity at high exposures, creates a serious concern that there
may be less obvious effects occurring at chronic lower exposures; but,
this does not necessarily consitute firm evidence of such occurrence.
Tables 14 and 15 list, for experimental animals and humans respec-
tively, the various effects of cadmium that have been studied, observed,
or merely suggested in the literature. Many of these occur only at very
high exposures, and many have not been studied well enough for any defin-
itive resolution. This section of the report discusses the best docu-
mented effects.
Major concern over possible chronic cadmium poisoning in the U.S. has
centered around two major areas of impact — the kidney and the cardio-
vascular system. While the impact of cadmium on the kidney is a well
documented cause-effect relationship, the possible impact on the cardio-
vascular system is under considerable controversy. In addition, research
has been conducted and is underway to determine the extent of effects
of cadmium on the bone, the liver, the testicles, genetic matter, and
the fetus, as well as possible carcinogenic and emphysematous inter-
actions. Itai-itai disease is a well-publicized syndrome of serious
consequence in Japan, but it is not felt to be a major concern in this
country and is therefore considered cursorily.,
Renal Damage
Long-term exposure to cadmium is known to produce renal damage in
both animals and man. As summarized in the Karolinska Institute study:
"A characteristic symptom [of renal damage] is proteinuria of
the tubular type. Cadmium is considered to be transported
to the tubules with a low molecular weight protein metal-
lothionein, and has been found to accumulate there with a
very long biological half-time. When a level of about 200
ppm [wet weight] is reached in renal cortex, the first sign
84
-------
Table 14. ADVERSE EFFECTS OF CADMIUM ON EXPERIMENTAL
ANIMALS (FROM ORNL)
Amyloidosis (the accumulation in various body tissues of amyloid, an
abnormal complex material, probably a glycoprotein)
Anemia
Cancer
Cirrhosis (a disease of the liver, marked by progressive destruction
of liver cells)
Dental changes
Enteritis (inflammation of the intestine)
Gastritis
Hypertension ; •
Hypocalcaemia (reduction of the blood calcium below normal)
life span shortening
'K
Necrosis (death of tissue of the liver)
Nerve damages
Ovarian changes
Pancreatic atrophy
Proteinuria (the presence of pro'tein in the urins)
Pulmonary emphysema
Renal damage
Teratogenic and embryotoxic effects,
Testicular atrophy and lesions
Toxemia of pregnancy (a series of pathological conditions, essen-
tially metabolic disturbances occurring in
pregnancy)
Weight loss
85
-------
Table 15. ADVERSE EFFECTS OF CADMIUM
ON HUMANS (FROM ORNL)
Ami.no-aciduria (presence, of amino acids in the urine)
Anemia
Aneuria (failure or lack of nervous energy)
Anosmia (absence of the sense of smell)
Dental changes
Gastritis
«
Glucosuria (the presence of glucose in the urine)
Hypercalciuria (excess of calcium in the urine)
Hypertension
Increased sedimentation rate of erythrocytes (one of the elements
found in peripheral
blood)
liver damage
Osteoporosis (decalcification of the skeleton)
Proteinuria (the presence of protein in the urine)
Pulmonary emphysema
Reduced working capacity
Renal danage (kidney damage)
Renal stones (kidney stones)
86
-------
of tubular dysfunction (tubular proteinuria) may appear
in sensitive persons. Furthermore, exposure to cadmium
may lead to generalized renal tubular damage with in-
creased excretion of amino acids, glucose, phosphorous
and calcium. This may eventually cause changes in
metabolism with decalcification and osteomalacia.^3
The 200 ppm renal cortex cadmium concentration is thus offered by
Friberg, et al. as a tissue-level threshold value which incorporates
no safety factor other than that implicit in the phrases "first sign
of ... dysfunction" and "in sensitive persons." The level of 200 ppra
is four times the average concentration presently found in the U.S. for
people of age 50.
This tissue-level threshold value represents the present best judgment
/
on a generally accepted effect of cadmium. To use it for standard-
setting or other policy purposes, however, requires that one estimate
the environmental exposure which would produce such a level. This
estimation requires the use of not only exposure data, but also the
estimates of the retention percentages discussed earlier. Table 16
provides, for various lengths of exposure and retention rates, esti-
..mates of the environmental exposure required to produce the level of
200 ppm in the- renal cortex. The "significant exposure" figures of
3
132 yg per day for food as a sole source or 1.3 yg/m for ambient air
24
alone that were used in the ESED draft PSP report represent the
selection of 5 percent and 25 percent retention for ingestion and
inhalation, respectively, for a 50-year exposure. The presentation of
"significant exposure" in the above^.report leads to a misunderstanding
of the importance of individual exposure levels. The summation of
exposure levels given in Table 7-1 of that report ignores differential
absorption rates relative to the mode of entry. This error underesti-
mates the relative contribution from air by a factor of about 5 and
the contribution from cigarette smoking by a factor of about 10. (See
Section VI for a corrected representation of the total environmental
exposure.)
87
-------
Table 16. ESTIMATED MINIMUM CADMIUM LEVELS NECESSARY FOR
REACHING 200 ppra (WET WEIGHT) OF CADMIUM IN
RENAL CORTEX3 (After CITE)
Years of
exposure
10
25
50
Ingestion
Vg Cd/day
Retention
2.5 5
1324 662
530 265
265 132
%
10
331
132
66
Ambient air
yg Cd/m3
Retention %
10 25 40
16.2 6.5 4.1
6.5 2.6 1.6
3.2 1.3 0.8
Industrial air
ug Cd/ra3
Retention %
10 25 40
52.5 21. 13.1
21.0 8.4 5.2
10.5 4.2 2.6
a 3
A lung ventilation of 20 m per day has been used for evaluation
of ambient air exposure. A lung ventilation of 10 m3 per 8 hours,
225 days per year has been used for evaluation of industrial air
exposure. No corrections have been made for cumulative effects
of different types of exposure, including tobacco smoking, A
linear approximation of the accumulation of cadmium has been used.
The previous discussion concerning 200 ppm cadmium in the renal cortex
was based on the appearance, at that level, of proteinuria, an indica-
tor of renal tubular damage. Although proteinuria is considered to be
05
the first sign of tubular dysfunction, a recent Swedish study indi-
cated that excretion of ribonuclease in the urine may be found at renal
cortex levels of about 90 ppm, indicating a considerably lower critical
level than the one indicated above. This report also suggested that
cellular effects may well appear at lower levels. The excretion of
ribonuclease was found in rats exposed to cadmium in the drinking water
and on a calcium deficient diet as compared with a calcium deficient
control group»
%
Hypertension
Cadmium has been implicated as one of the factors that may cause hyper-
tension in humans; however, no real agreement has been reached in this
area. About as many studies indicating no relationship between cadmium
88
-------
and hypertension have been conducted as those which do, in both human
and animal research. This results in a fairly ambiguous situation.
Strain differences, the negative findings in animal studies, and the
lack of epidemiological investigations that consider the whole exposure
instead of just dustfall or ambient concentrations of cadmium are the
principal reasons that no firm conclusions can be given. Since a number
of studies have demonstrated such a relationship, however, it is nec-
essary that this be given serious attention, especially as the relevant
exposure levels are equivalent to normal exposures.
Epidemiological evidence in favor of the cadmium-hypertension hypothesis
includes the following:
•
1. An inverse correlation between the prevalence of hyper-
tension or hypertension-related deaths and the hardness
of drinking water (hard water generally contains less
cadmium as well as other metals).26
•"^
2* Correlation between the deaths due to hypertension and
airborne concentration of cadmium.27
3. Correlation between (a) the relative incidence of hyper-
tension in cities and nations and (b) the renal cadmium
» content at death, from any cause, of individuals from
those cities and nations.28
4. Correlation between (a) individual death fron hyper-
tension and renal cadmium content or the zinc-to-cadmium
ratio found in the kidney at time of death and (b) death
from other causes and renal cadmium or zinc-to-cadmium
content.29
Some of the experimental evidence in favor of the hypothesis is sum-
marized below:
1. Demonstrated in rats and rabbits.30
2. Reversal of cadmium-induced hypertension in rats by
the administration of a zinc chelate, the specific
action of which is to remove the cadmium.
89
-------
3, Hypertension in rats always developed when cadmium-to-
zinc ratio was greater than 0.88^^
4. The reversal of cadmium-induced hypertension in rats
by selenium, a reported cadmium antagonist in several
biologic systems; and hardwater*
34 35
Along with those studies ' that have shown no relationship between
cadmium exposure and hypertension, the strongest evidence often cited is
that workers undergoing industrial exposure or victims of itai-itai disease
have shown no significant hypertension-cadmium relationship. However,
these discrepancies have also been explained in several ways,. The Oak
Ridge study concluded that elevated cadmium exposure might make a per-
son with essential hypertension more susceptible to cardiovascular
failure because of its effect on the sodium balance., A small increase
in the cadmium dose would then significantly increase the risk of death
for those already suffering from essential hypertension. Since in-
dustrial workers are not a group likely to be hypertensive, increased
exposure to cadmium would not seriously affect their cardiovascular
system. At the same time, studies that have concentrated on the levels
of cadmium at the time of death and death due to hypertension would
select out precisely those individuals in the above category. Just as
air pollution episodes often cause increased death among those most
susceptible to respiratory problems, cadmium exposure may be increas-
ing the risk of death due to hypertension for those already suffering
from it.
Considering the above argument to be speculative, Schroeder has said
that these discrepancies may be best explained by considering the
biphasic action of cadmium on the smooth muscle with small doses being
pressor and large doses being depressor, Then, excessive incidence of
hypertension is not found when exposure is sufficiently high to cause
kidney damage where cadmium is eliminated along with protein in the
urine. Therefore, cadmium would induce hypertension only at doses
beloxj some critical level, eliminating from consideration victims of
90
-------
itai-itai and workers with proteinuria. Some support for this reasoning
has been given by a recent study which found consistantly higher diastolic
pressure among phosphate fertilizer workers with no abnormal cadmium in
•)£
the urine. In a chronic exposure laboratory study lower levels of in-
take of cadmium produced increases in systolic pressure later than the
34
higher intake levels. The systolic pressure returned to normal or
below normal after continued exposure at the higher levels of intake.
In CITE, Friberg et al., discuss the hypertension question by noting the
epidemiological evidence and commenting that it involves only part of the
total environmental exposure and that highly-exposed persons do not show
the effect; this report concluded only that the question should be studied
further. They do not discuss the ORNL or Schroeder suggestions concerning
explanations for the lack of hypertension in highly-exposed persons. The
concern over the epidemiological evidence being based on'airborne or water
exposure only is reasonable since these exposures may be only a few per-
••V.
cent of,the total body burden. This concern rests on the assumption that
.f
the hypertension is related to the total systemic body burden and not to
the inhaled or ingested exposure only. The Panel on Hazardous Trace Sub-
stances of the National Institute of Health also gave little support to
the cadmium-hypertension relationship.
The possible cause-effect relationship of cadmium and hypertension is
obviously an area that needs clarification. A principle concern is
that the postulated significant exposure levels of cadmium are ones
that are within the range of the levels to which much of the population
is exposed. If cadmium is a primary factor in hypertension, the in-
creased levels of cadmium in the air, in the soil and food, and from
cigarette smoking may well account for the increased incidence of hyper-
tension and death due to hypertension that has been observed over the
decades.
91
-------
Other Effects
While the above effects of renal damage and hypertension are those which
are most often cited as reasons for the possible control of cadmium in
the environment, other effects have also been documented or postulated.
These are summarized below to give a better understanding of the
toxicity and general impact of cadmium.
Occupational exposure to cadmium oxide fumes, cadmium oxide dust, and
cadmium pigment dust can give rise to chronic pulmonary disorders,
characterized as emphysematous changes. These changes usually take
several years to develop in humans but have been observed after only
a couple of years.1 exposure. Dose-response relationships are uncertain
due to the lack of time weighted averages for extended periods or the
relevance of the short-term high-level exposure. For cadmium oxide
fumes, considered more dangerous than cadmium oxide dust, a prolonged
:o \
37
3
industrial exposure to well below 100 yg/m has been considered capable
of causing emphysema,
The only data that links emphysema with ambient levels of cadmium come
from two studies of hepatic concentration of cadmium and deceased
patients with chronic bronchitis and emphysema. These studies indicate
that a higher concentration (almost four times as high) of cadmium was
found in the livers of patients with emphysema and chronic bronchitis
38
than in the control group. No reasons for this association nor any
dose-response relationship are known; however, that a correlation has
been found at presumably normal ambient levels would indicate some need
for further investigation,.
Carcinogenic effects, through exposure to cadmium and cadmium compounds,
may also occur as a result of ambient exposure. One study has cautiously
indicated an association between prostatic cancer and concentrations of
39
suspended particulate air pollutants, while other studies have found
92
-------
statistically higher concentrations of cadmium in the liver and kidney of
patients who had died from bronchogenic carcinoma 'than in a control group
40
or a group that died from other forms of neoplasia. The high levels of
cadmium in the food of Japan have led to some speculation that these lev-
els are the cause of the higher incidence of stomach cancer in Japan than in
41
other countries. While most of the associations between ambient cadmium
and cancer in man are speculative, animal studies have shown a definite
relationship between sarcoma and cadmium. These studies seem to indicate
42
that cadmium may be one of the most potent metallic carcinogens yet known.
One effect that may possibly occur with typical environmental exposure
in the U.S. is on the liver where the second largest proportion of cadmium
is stored in the bo'dy. An influence on the liver function of workers with
acute cadmium poisoning has been noted, as well as increases in the serum
43
gamma globulin levels of workers with chronic exposure. While no liver
dysfunction has been found in the normally exposed population, morpholog-
'•»>
ical and liver enzyme changes have been found in animals exposed to
t
cadmium concentrations of the same magnitude as found in human adults
44
though negative liver function tests did not show any variation. This
could easily account for the lack of findings by such tests in humans.
Testicular necrosis is- a well-documented effect of cadmium following
systemic administration to a large number of animal species, suggest-
ing that a similar effect is possible in human beings. To date, such
an effect on the testes has not been reported to be a result of cadmium
even for industrially-exposed workers. It is not likely that typical
45
environmental exposure levels would produce any such effect.
Relationships between cadmium in the diet and demineralization of the
bone have been shown, but this is most likely a side effect of renal
tubular failure. Excretion of calcium and phosphorus, because of an
93
-------
impairment of the renal tubular regulation of Ca/P balance due to
other causes, is known to cause osteomalacia. Other possible explana-
tions for the cadmium-calcium relationship are decreased uptake of
calcium from the gastrointestinal tract in the presence of cadmium and
a change in the parathyroid activity which regulates calcium levels.
The majority of studies showing effect on bone due to cadmium have been
46
coupled with calcium or vitamin D deficient diets.
Anemia has also been shown to be the result of excessive cadmium in
the diets of animals and has been found in several studies of workers
47
exposed to cadmium oxide; however, the dosage required is much
greater than found in normal exposures as cadmium concentrations in
48
drinking water of 10 ppm for a year failed to show any anemia. In
addition, the accumulated amount of cadmium has not been found to be
related to the incidence of anemia, so no dose-response relationships
can be given.
Itai-itai ("ouch-ouch") disease is a recognized result of severe chronic
cadmium poisoning which occurred in the Jintsu River Basin in Japan.
It is described as renal tubular dysfunction accompanied by osteomalacia
1
(softening of bones) and osteoporosis (decalcification) and has character-
istic symptoms of (1) severe pain at the regions of the groin, the loins,
the back, and the joints; (2) duck gait; (3) tendency to bone fracture;
e.g.,-humerous, rib, pelvis, radius, ulna, and femur; (4) proteinuria and
glycosuria; and (5) increase of serum alkaline-phosphatase and a decrease
of inorganic phosphorus in the serum. It has usually occurred in women
who had borne several children and reached menopause.' Since it requires
ingestion of more than 600 pg/day of cadmium, along with dietary deficien-
cies of vitamin D nnd calcium, it is far beyond any known or expected
49
levels in the U.S.
94
-------
MECHANISMS OF ACTION
To date, no mechanism of action has been completely determined for any
of the effects of cadmium. However, several forms of a protein with
which cadmium is known to bind have been identified and relationships
between cadmium effects and the intake of zinc, calcium, selenium, pro-
tein, sulfhydryl compounds, vitamin C, and vitamin D have been suggested.
These findings and possible conclusions are summarized below.
Though it is not known precisely how cadmium is initially absorbed into
the blood stream from the lungs or intestines, studies have shown that
in the first hours after intravenous and intraperitoneal injections
cadmium in blood was mainly in the plasma and partly dialyzable. This
may mean that the initial uptake of cadmium in the kidneys after injection
is the result of glomerular filtration and reabsorption in the tubules.
•v
This filtration provides^ a. rap idv,dec.reas.e, o.f. cadmium levels in the plasma
>
with a lower rate of decrease in blood cell levels. After about 24 hours,
cadmium will be located principally in the blood cells and this concentra-
tion will begin to rise. A low molecular weight protein, metallothioneit
%
is thought to be the reason for this change in cadmium metabolism.
Metallothionein is the only known biologiocal substance which naturally
contains cadmium. Metallothionein is also known to bind with zinc,
mercury, and to a lesser extent with copper and iron. This metal-binding
protein, about 9 percent metal and the remainder protein, is character-
i
ized by a low molecular weight, absence of aromatic amino acids, and a
high percentage of cysteine in the molecule. The function of the
tnetallothionein proteins in the bodily system has yet to be determined;
but, among those possible functions suggested are transport, storage,.
52
detoxification, and immune responses.
95
-------
Metallothionein does not normally exist in amounts large enough to
handle injected doses or acute exposure resulting in most of the cadmium
being initially in the plasma. Small doses of cadmium have been shown
*
to act as protection against normally toxic levels of cadmium, implying
that the presence of the cadmium ion itself may induce the over-production
53
of metallothionein. Therefore, low level chronic doses of inhaled or
ingested cadmium will probably be almost completely bound with metal-
lothionein upon uptake into the blood stream.
Whether the presence of cadmium or the increased concentration of zinc
initiates the increased production of metallothionein is uncertain.
Because metallothionein?s bond with cadmium is about 3000 times stronger
than its bond with .zinc, the introduction of abnormal amounts of cadmium
in the blood will mean that cadmium will replace zinc on metallothionein,
(Mercury's bond is even stronger than cadmium's, but an increase in
54
mercury does not seem to lead to increased metallothioein, further
implying that cadmium does induce the production of metallothionein.)
Bound to metallothionein, cadmium is transported in the blood to the
various organs. At the kidneys, the cadmium-metallothionein complex is
probably cleared completely due to its low molecular weight. As protein
in normal kidneys is almost completely reabsorbed, it is not likely that
*
any metallothionein will be found in the urine until the kidney has
become saturated with cadmium. At this stage, reabsorption decreases,
tubular proteinuria appears, and some metallothionein-containing cadmium
is found in the urine. The basis for the selective transport and accumu-
lation of cadmium in the kidneys and liver is not known though some
consideration is given to the several forms of metallothionein which
have slightly different properties.
In addition to the above relationship between zinc, cadmium, and a
protein, many enzymes require zinc in order to function. Since cadmium
is an analogue of zinc, some of the adverse effects are probably due to
96
-------
substitution of cadmium for zinc in particular enzyme structures and
functions. Cadmium has been shown to inactivate several enzymes which
bear sulfhydryl groups and to interfere with oxidative phosphorylation
56
in the citric acid cycle which may, in turn, play a role in respiratory
inhibition. These enzyme inhibition characteristics of cadmium would be
of primary concern relative to acute toxicity and not directly applicable
to long-term environmental exposure.
The analagous behavior of zinc and cadmium on metallothionein and metal-
loenzymes implies that zinc would have an antagonistic effect on the
action of cadmium. Such a relationship has been noticed through both
increased incidence of cadmium poisoning in people and animals on zinc
deficient diets and decreased incidence of the effects of cadmium on
animals given an increased zinc dosage. ;
Along with zinc, cysteine and selenium have been found to act as pro-
tective agents against the adverse actue effects of cadmium in the
54
testes. .f The protection seems to occur at the site of injury, not by
the prevention of transport of cadmium to the target site. This also
appears to be the case for teratogenesis and placental damage caused by
cadmium. Though the actual antagonistic mechanisms are not known, they
are probably related to- the relationship of" selenium co enzymes and the
presence in cysteine of sulfhydryl, by which cadmium is bound to
metallothionein.
Other dietary influences which have been noticed include increased
cadmium uptake in humans and/or animals fed diets deficient in protein,
calcium, or vitamin D and the prevention of cadmium anemia when fed a
large amount of iron or vitamin C, which brings about increased absorption
of iron from the intestinal tract. While it is doubtful that cadmium
plays much of a role in the latter interaction except for having initially
caused the anemia, it is thought that cadmium will inhibit the uptake of
calcium from the intestines and will actually replace calcium in its lattice
complexes and ligand molecules.
97
-------
REFERENCES
1. Gleason, M. N., R. S. Gossclin, H. C. Hodge, and R. P. Smith, Clinical
Toxicology of Commercial Products: Acute Poisoning (Home and Farm),
11-29 (Baltimore, 1969). ORNL.
2. CITE, Section 6.1.2.4.4.
3. Hardy, H. L. and J. B. Skinner. The Possibility of Chronic Cadmium
Poisoning. J Industr Hyg Toxicol., 29:181-196, 1955. ORNLo
4. Hygienic Guide Series: Cadmium, American Industrial Hygiene Associ-
ation, September 1944. ORNLo
5. CITE, Section 4.1.1.4. '
6. CITE, Section 4.1.2.3.
•
7. Kehoe, R. The Metabolism of Lead in Man in Health and Disease,
J Roy Inst Publ Hlth., 24:81-97, 101-120a, 129-143, 177-203,
1961. TGOMA.
212
Hursh, J.B. and J. Suomela. Absorption of Pb from the Gastroin-
testinal Tract of Man. Acta Radiol (Stockh.). 7:108-120, 1967.
TGOMA.
Bornmann, G., G. Henlce, H. Alfes, and H. Mollmann. Uber die Enterale
Resorption von Metallischem Quecksilber. Arch Toxikol. 26:203-209,
1970, TGOMA. .
Aberg, B., L. Ekman, R. Falk, U. Greitz, G. Persson, and J. 0. Snihs,,
Metabolism of Methyl Mercury (^O^Hg) Compounds in Man, Excretion and
Distribution. Arch Environ Hlth. 19:478, 1969* TGOMA.
8. CITE, II, Section 4.3.1.
9. Tsuchiya, K. Causation of Ouch-Ouch Disease, an Introductory Review.
Keio J Med. 18:181, 1969. CITE. . ' ..
10, Friberg, L. Deposition and Distribution of Cadmium in Man in Chronic
Cadmium Poisoning. AMA Arch Ind Hyg Occup Med. 16:27, 1957. CITE.
Smith, J.P., J.C. Smith and A.J. McCall. Chronic Poisoning from
Cadmium Fumes. J Path Bactcriol. 80:287, 1960. CITE.
11, Ibid.
Tipton, I.H. and M.J. Cook. Trace Elements in Human Tissue, II.
Adult Subjects from the United States. Hlth Phys. 9:103, 1963. CITE|
98
-------
12. Tipton, I.H. and J.J. Shafer. Statistical Analysis of Lung Trace
Element Levels. Arch Environ Illth. 8:58, 1964. CITE.
Molokhia, M.M. and 11. Smith. Trace Elements in the Lung. Arch
Environ Hlth. 15:745, 1967. CITE.
Smith, Smith, and McCall. Loc. Cit.
Geldmacher v. Mallinckrodt, M. and 0. Opitz. Zur Diagnostik der
Cadmiumvergiftung der Normale Cadmiumgehalt Menschlicher Organe
und Korperflussigkeiten, Arbeitsmedizin, Socialmedizin, Arbeitshy-
giene. 3:276, 1968. -CITE.
13. Kanazawa, Japan-Ishizaki, Fukushima, and Sakamoto, 1970b; Kobe,
Japan-Kitamura, Sumino, and Kamatani, 1970. U.S.A.-Schroeder and
Balassa, Sweden-Piscator, Unpublished Data. CITE.
14. Geldmacher v. Mallinckrodt and Optiz. Loc Cit.
15. CITE, Section 4.3.4. '
16. Piscator, Unpublished Data. CITE, II.
17. Essing, H.G., K.H. Schaller, D. Szadkowski and G. Lehnert. Usuelle
Cadmiutnbelastung Durch Nahrungsmittel und Getrenke. Arch Hyg Bakt.
153:490, 1969. CITE.
*
Tipton, I.H. and P.L. Stewart., Patterns of Elemental Excretion in
Long-Term Balance Studies, II., in Snyder, W.S., Internal Dosimetry.
» Hlth Phys Div Ann Progr Rep., for period ending July 31, 1969.
ORNL-4446, 303, 1970. CITE.
Tsuchiya. Loc Cit. ,
IS. Tsuchiya, K., Y. Saki, and M. Sugita. Biological.Threshold Limits
of Lead and Cadmium. IVII Int Cong Occup Hlth. 1972b, Proc. 1973.
CITE, II.
19. Tsuchiya, K. and M. Sugita. A Mathematical Model for Deriving the
Biological Half-Life of a Chemical. Nord Hyg T. 53:105-110, 1971.
TGOMA.
CITE, Appendix.
20. ICRP Publication 2 (1959). Recommendations of the International
Commission on Radiological Protection. Report of Commitee II on
Permissible Dose for Internal Radiation. Pergamon Press, London.
TGMA.
99
-------
21. Micttinen, J. K. Gastrointestinal Absorption and Whole-Body Reten-
tion of Toxic Heavy Metal Compounds (methyl mercury, ionic mercury,
cadmium) in Han. (To be published in the Proceedings of the X
International Congress on Occupational Health.) TGOMA.
Rahola, T, , R. K. Aaran and J. K. Miettinen, (1971). Half-Time
Studies of Mercury and Cadmium by Whole Body Counting, IAEA/WHO
Symp. on the Assessment of Radioactive Organ and Body Burdens.
Stockholm, November 22-26, 1971. TGOMAo
22. Bernstein, D. M., et al. Uptake and Distribution of Airborne Trace
Metals in Man. Trace Substances in Environmental Health-VIII,
Proceedings of University of Missouri's 8th Annual Conference on
Trace Substances in Environmental Health. June 11, 12, and 13, 1974.
University of Missouri, pp. 329-334.
23. CITE, II, Section 5.1.
24. PSP, Section 7.
25. Piscator, M., and S. E. Larsson. Retention and Toxicity of Cadmium
in Calcium-Deficient Rats. XVII Int Cong Occup Hlth^ 1972 (Proc.
1973). CITE, II. . .
26. Kobayashi, J. Geographical Relationship between Chemical Nature of
River and Death Rate from Apoplexy. Berichte d Ohara Ins to f.
Landwirtsch, Biologie. 11:12-21, March 1957. ORNL.
Schroeder, H. A. Relation between Mortality and Cardiovascular
Disease and Treated Water Supplies. J Amer Med Assoco 172:17,
April 23, 1970. ORNL.
Morris, J. N,, M. D. Crawford, and J. A. Heady. Hardness of Local
Water Supplies and Mortality from Cardiovascular Disease; The
Lancet, 860-62, April 22, 1961. ORNLo
Schroeder, H. A. Municipal Drinking Water and Cardiovascular Death
Rates. J Amer Med Assoc. 195:2, 81-85, January 10, 1966. ORNL.
Crawford, M, D., M. J. Gardner, and J, N. Morris. Mortality and
Hardness of Local Water Supplies. The Lancet. 827-31, April 20,
1968. ORNL.
The Water Story. The Lancet. 1012-13, May 17, 1969. ORNL.
Schroeder, H. A. and W. J. Vinton. Hypertension Induced in Rats by
Small Doses of Cadmium. Amer J Physiol. 202:515-18, 1962. ORNL.
Schroeder, H. A. Renal Cadmium and Essential Hypertension. J Amer
Mod Assoc. 187:358, February 1, 1964. ORNL.
100
-------
Morton, W. E. Hypertsension and Drinking Water Constituents in
Colorado. Amer J Public Hlth. 61:7, 1371-1378, 1971. ORNL.
Schrocder, H. A., A. P. Nason, I. H. Tipton, and J. J. Balassa.
Essential Trace Metals in Man: Zinc .- Relation to Environmental
Cadmium. J of Chronic Diseases. 20:179-210, 1967. ORNL.
27. Carroll, R. E. The Relationship of Cadmium in the Air to Cardio-
vascular Death Rates. J Amer Med Assoc. 198:267-69, 1966.
28. Perry, H. M., Jr., I. H. Tipton, H. A. Schroeder, R. L. Steiner, and
M. J. Cook. Variation in the Concentration of Cadmium in Human
Kidney as a Function of Age and Geographic Origin. J Chron Dis,
August 1961. ORNL. -
Schroeder, H. A. Cadmium as a Factor in Hypertension. J Chron Dis.
18:647-56, 1965. ORNL.
29. Schroeder. Loc Cit.
30. Schroeder, H. A. Cadmium Hypertension in Rats. Amer J Physiol.
207:62-66, July-December 1964. ORNL.
Schroeder and Vinton. Loc Cit.
Kanisaura, M. and H. ^A. Schroeder. Renal Arteriolar Changes in
Hypertensive. Rats, given Cadmium in Drinking Water. J Exp Mol Pathol.
l'0:81-98, February 1969. ORNL.
31. Schroeder, H. A. and J. Bucktnan. Cadmium Hypertension. Arch Environ
Hlth. 14:693-97, May 1967. ORNL.
*
32. Schroeder (1964). Loc Cit.
33. Perry, H. M., Jr., E. F. Perry, and M. W. Erlanger. Reversal of
Cadmium-Induced Hypertension by Selenium or Hard Water. Trace
Substances in Environmental Health-VIII, Proceedings of University
of Missouri's 8th Annual Conference on Trace Substances in Environ-
mental Health. June 11, 12, and 13, 1974. pp. 51-57.
34. Hunt, W. F., C. Pinkerton, 0. McNulty, and J. Creason. A Study in
Trace Element Pollution of Air in 77 Midwestern Cities, presented
at the 4th Annual Conference, on Trace Substances om Environmental
Health, Univ. of Missouri, Columbia,, Missouri, p. 21, June 23-24,
1970. ORNL.
101
-------
35. Morgan, J.M. Tissue Cadmium Concentration in Man. Arch Intern
Med. 123:405-08, 1969. ORNL.
36. D.I, Hammer, J.E. Finklea, J.P. Creason, S.H. Sandifer, J.H. Keil,
L.E. Priester and J.F. Stara, Cadmium Exposure and Human Health
Effects. Trace Substances in Environmental Health - V. University
of Missouri, 1972. p. 269-.
37. Schroeder (1964). Loc Cito
38, Lewis, G.P., H. Lyle, and S. Miller. Association between Elevated
Hepatic Water-Soluble-Protein-Bound Cadmium Levels and Chronic
Bronchitis and/or Emphysema. Lancet, 2, 1330, 1969. CITE.
Morgan, (Unpublished Data), CITE.
39. Winkelstein, W. Jr. and S. Kantor, Prostatic Cancer: Relationship
to Suspended Particulate Air Pollution. Amer J Pub Hlth. 59:1134,
1969. CITE,
40. Morgan, J.M. Tissue Cadmium Concentration in Man. Arch Intern
Med, 123:405, 1969. CITE.
Morgan, J.M. Cadmium and Zinc Abnormalities in Bronchogenic
Carcinoma, Cancer. 25:1394, 1970. CITE,
41. CITE, Section 7.1.2.
42. Gunn, S.A., T.C. Gould, and W.A.D. Anderson. Specific Response of
Mesenchymal Tissue to Cancerigenesis by Cadmium. Arch Path. 83:493
1967, CITE,
43, Friberg, L, Health Hazards in the Manufacture of Alkaline Accumulati
with Special Reference to Chronic Cadmium Poisoning. Acta Med Scand
138, Suppl., 240, 1950, CITE,
44. Stowe, H.D., M. Wilson, and R.A. Goyer. Clinical and Morphological
Effects of Oral Cadmium Toxicity in Rabbits. Arch Path. 94:389,
1972. CITE, II,
45. CITE, Section 6.6.3,
46. CITE, Section 6.4.3,
47. Nicaud, P., A. Lafitte, and A. Gros, Les Troubles de L'lntoxication
Chronique par Ic Cadmium, Arch Mai Prof Med Trav Secur Soc. 4:192,
1942. CITE
Friberg. Loc Cit,
102
-------
Tsuchiya, K. Proteinuria of Workers Exposed to Cadmium Fume. The
Relationship to Concentration in the Working Environment. Arch
Environ Hlth. 14:875, 1967. CITE.
Piscator, (Unpublished Data). CITE.
48. Decker, L.E., R.U. Byerrum, C.F. Decker, C.A. Hoppert, and
R.F. Langham. Chronic Toxicity Studies. I. Cadmium Administered
in Drinking Water to Rats. AMA Arch Ind Hlth. 18:228, 1958. CITE.
49. ORNL, Section VI, C.2.a.i.
50. Perry, H.M. Erlanger, M., Yunice, A., Schoopile, E., and Perry E.F.
Hypertension and Tissue Metal Levels Following Intravenous Cadmium,
Mercury and Zinc. Amer J Physiol. 219:755, 1970. CITE.
51. CITE, II, Section 4.5.
52. ORNL, Section VI, C.I.
53. Terhaar, C.J., E. Vis, RlL. Roudabush, and D.W. Fassett. Protective
Effects of Low Doses of Cadmium Chloride Against Subsequent High
Doses in the Rat. Toxicol Appl Pharmac. 7:700, 1965. ORNL.
•V
Gunn, S.A., F.C. Gould, and W.A.D. Anderson. Am J Pathol. 48:959,
?1966. ORNL.
*
Gabbiani, G., D. Baic, and C. Deziel'. Studies on Tolerance and Ionic
Antagonism for Cadmium or Mercury. Can J Physiol Pharmacol. 45:443,
1967.
\
54. Piotrowski, JJKL.-,- B-..- Xro.janowska, J.M. Wisniewska-Knypl, and
W. Bolanowska. Further Investigations on Binding and Release of
Mercury in the Rat. Mercury, Mercurials and Mercaptans, Miller, M.W.
and Clarkson, T.W., Eds., C.C. Thomas Publ., Springfield, Illinois,
1972.
55. Simon, F.P., A.M. Potts and R.W. Gerad. Action of Cadmium and Thiols
on Tissues and Enzymes. Arch Biochem. 12:283-291, 1947. ORNL.
f
*.'
56. Jacobs, E.E., M. Jacob, D.R. Sanadi and L.B. Bradley. Uncoupling of
Oxidative Phosphorylation by Cadmium Ion. J Biol Chetn. 223:147-156,
1956.
57. Gunn, S.A., T.C. Gould, and W.A.D. Anderson. Mechanisms of Zinc,
Cysteine, and Selenium Protection Against Cadmium-Induced Vascular
Injury to Mouse Testes. J Rcprod Fertil. 15:65-70, 1968. ORNL.
103
-------
SECTION V
CADMIUM IN ENVIRONMENTAL MEDIA
While the main purpose of this document is to assess cadmium as an air
pollutant and thereby determine the possible need for emission control, the
evaluation of the practicality of control options requires more than just
the knowledge of total emissions of cadmium and the resulting air quality.
Other environmental media can be exposed to excessive levels of cadmium
and serve as a source of increased body burden. Cadmium intake from these
media may further raise the need to control the ambient air exposure or
may make air's impact relatively small and unimportant by comparison.
This section summarizes the levels of cadmium in the various media and
discusses the anthropogenic activities that have contributed to such
levels, except for air emissions which have been discussed previously.
In addition, the intermedia flow of cadmium is analyzed with .special
emphasis on defining that portion of the cadmium levels in other media
originating in cadmium emissions to the air.
AIRBORNE CADMIUM
Ambient Air Quality
As with, any air pollutant, the ambient concentration of cadmium varies
depending upon location. Rural areas, far from any cadmium sources, have
o
been found with monthly means around 0.0004 ng/m , while maximum 24-hour
^
values in U.S. cities with zinc refineries have been around 0.7 ug/m .
More generally, cadmium concentrations have been reported to be around
3 ^
0.001 |ig/m in rural areas, 0.01 - 0.05 ug/m in cities, and 0.3
104
-------
near cadmium emitting industries. A current study of particulate
concentrations around the country has provided metals data on 15 cities
2
during the period 1972 to 1974. Of the 134 quarterly averages given
during this time period for these cities, 82 percent of the quarterly
averages were listed as below detectable and the reported quarterly
2
averages ranged from 0.0051 ug/m (Birmingham - 4th quarter 1972) to
2
0.1620 ug/m (Denver - 4th quarter 1972). All of the cities had at
least one quarterly average reported as below detectable for each year.
Since cadmium occurs in the particulate form, it is susceptible to set-
tling, impaction, raindut, washout, and other means of natural removal
from the air and eventual deposition. Rates of deposition of cadmium
have been measured in a wide variety of exposure areas, ranging from a
2
rural subartic area in Finland (0.001 mg/m /month in winter and 0.006
2 . ^
mg/m /month in summer) to points 1000 meters from a cadmium-emitting
smelter in East Helena (1-4 mg/m "/month) and 500 meters from a source
2 "5
in Japan (6.2 mg/m /month).
4
*
*
A better definition of typical deposition rates can be obtained from a
study of 77 midwestern U.S. cities where dustfall was measured over a
four-vmonth period. Measurements were made in residential, commercial,
and industrial areas, indicating means of 0.040, 0.063, and 0.075
2
mg Cd/m /month, respectively. This study also provided total dustfall
and zinc dustfall measurements (see Table 17). A review of this data
indicates that both cadmium and zinc deposition rates are about twice as
high in industrial areas as in residential areas, but that the percentage
of cadmium and zinc to the total dustfall is relatively less in the indus-
trial areas than in the residential regions. This latter effect is prob-
ably because the industrial area receives large quantities of process
dusts containing little, if any, zinc or cadmium. The data in this study
Below detectable.(BD) was determined to be the following: in 1972,
BD < 0.00019 ug/m ; in 1973, BD < 0.00027 ug/m3; in 1974, BD <
0.00036 ••~'-3
105
-------
Table 17, TOTAL DUSTFALL, CADMIUM, AND ZINC FROM 77 U.S. CITIES
Residential
Commercial
Industrial
September
October
November
December
Total dustfall
g/m^ /month
3.16
5.45
7.07
5.01
5.61
4.81
4.55
Cadmium
rag /m2 /mo nth
Ov040
0.063
0.075
0.038
0,063
0.071
0.073
Zinc
mg/m /month
5.76
9.45
12.55
8.25
9.36
9.19
8.56
Cadmium in
total dust
ppm
12.7
11.6
10.6
7.58
11.2
14.8
16.0
Zinc in
total dust
ppm
1820
1730
1780
1650
1670
2200
1880
Zinc to
Cadmium
. 144
150
167
217
149
129
117
Source; Hunt, et al.
-------
also seems to indicate a higher ratio of zinc to cadmium in the industrial
areas, but it is not clear that this difference is significant.
Included in this table, and from the same study, are average dustfall,
cadmium, and zinc dustfall measurements for each month of the four month
sampling period, September to December. Of particular interest is the
increase in cadmium levels over this period without a corresponding in-
crease in the total dustfall or zinc levels. This suggests a seasonal
•
cadmium source, which is most likely the increased use of fuel oil and
coal during winter heating. Since both fossil fuels have a relatively low
zinc/cadmium ratio, ranging from 100 to 4, a decrease in the ratio of
zinc to cadmium would be expected.
8
Data from another study indicated comparable cadmium concentrations in
household dust. This study found a range of 4.25 to 14 ppm cadmium in the
household dust, versus the range of 7.6 to 16 ppm cadmium ratio in Table 17.
*v
Some indication of the concentration of cadmium in the air may also be
derived from dustfall figures. Cadmium dustfall values of 0.055 and 0.091
2
mg/m /month corresponded to average cadmium concentrations of 0.0032 and
39
0.0048 pig/m , respectively. If this same relationship is applied to the
*
above dustfall figures for the 77 U.S. cities, a range of ambient cadmium
3
concentrations of 0.0022 to 0.0041 j*g/m is indicated. As the majority
of the cities were small to moderate size and had no appreciable industry,
these concentrations may be considered reasonable.
Particle Characterization
As with all aerosols, the intake of cadmium from the air is dependent upon
•
the particle size and its solubility. The particle size determines the
deposition while the solubility reflects whether the particle will buildup
in the lungs or be absorbed and transported to other organs. Cadmium con-
centrations and mass median diameters were measured in downtown Cincinnati
10 3
and in a suburb. Concentrations averaged 0.08 and 0.02 pg/ra ,
107
-------
respectively, with corresponding average mass median diameters of 3.1
and 10 microns. In both areas about 40 percent of the particles were
11 3
smaller than 2 microns. Lee, et al. found 0.01 ug/m cadmium in
St. Louis, with a mass median diameter of 1.54 microns; 28 percent of
the cadmium particles were smaller than 1 micron, 65 percent smaller
than two microns. These small particle sizes suggest a high efficiency
of cadmium deposit in the alveoli,.
Cadmium is found in the air in particulate form, as the oxides chloride,
or sulfate. Cadmium oxide is formed by the oxidation of cadmium vapor
in the presence of water vapor or by the oxidation of the sulfate and
is nearly insoluble. Sulfates or chlorides of cadmium can be formed in
the presence of S02> S0_, or HCl, which are often also emitted at sources
of cadmium; these salts are extremely soluble in water. At present, no
data concerning the relative amounts of these individual compounds in
ambient air is available; however, about 85 percent of the cadmium in
12
New York City aerosols was found to be soluble. This same study, which
was concerned with trace metals in various tissues at death, indicated
that almost no cadmium was found in the lungs or lymph nodes- Therefore,
most cadmium deposited in the lungs is highly soluble.
CADMIUM AND WATER QUALITY
In areas not known to be polluted by cadmium, the concentration of cad-
mium is generally less than 1 ppb, both in natural waters and in drinking
13
water.. This is dramatically less than the 200 ppb overall crustal
abundance of cadmium due to the low solubility of cadmium with carbonate
and sulfide ions which arc available in natural waters. Increases above
this "natural" level can result from contamination of surface waters by
industrial discharges or of drinking water by the iron or plastic pipes
14
used in the distribution system. However, in polluted rivers, cadmium
will often go undetected in the water phase because of the low solubilityi
while large concentrations (as much as 10,000 times higher) are found in
suspended particles or in the bottom sediments.
108
-------
Surface Water Concentration
The U.S. Geological Survey conducted a study of surface waters in the
U.S. in 1970 and found that 96 percent of the 720 samples had a cadmium con-
centration less than the 10 ppb (ug/jH) USPHS drinking water standard, and
54 percent of that group had concentrations less than 1 ppb. The
highest level found was 130 ppb and, in general, the higher concentra-
tions of Cd in the water occurred in areas of high population density.
(Although these levels are low from a human health perspective, there
is evidence that the "safe" water quality concentrations in rivers
should be much lower - 0.2 ppb - based on long-term studies of the small
crustacean, Daphnia.)
The contamination of surface waters with cadmium can occur naturally
when streams flow over exposed bedrock containing cadmium, or if leach-
ates containing cadmium from soil and organic matter enter the stream.
In addition, waste water from mines or ore benefication operations,
t
electroplating wastes, domestic sewage, and other direct emissions to
the water, including direct airborne fallout, can add cadmium to the
water. Indirect emissions, such as leaching from land contaminated by
fallout, sludge disposal, and fertilizer or other agricultural chemical
use, are further burdens on the water quality.
Before discussing the possible contributions of the diverse sources to
cadmium levels, it is important to understand the behavior of cadmium
in the water. As mentioned earlier, very high concentrations of cadmium
are often found as suspended particles or in bottom sediments even
though the cadmium concentration in the water is barely detectable. For
water free of sulfide and complex-forming ligauds, with a pH less than
9.5, and containing more than 10 M total carbonate, the concentration
of dissolved cadmium is limited by solubility considerations to values
18
as low as a few parts per billion. As is seen in Figure 14 only in
nearly neutral waters with very low carbonate levels can dissolved
109
-------
TOTAL CARBONATE (MOLES/LITER J
10
r>
The sum of carbonate, bicarbonate, and dissolved CO,
Figure 14. The solubility of cadmium versus pH and versus
total carbonate concentration (after ORNL)
110
-------
cadmium reach high levels. This solubility relationship and varying
alkalinity levels account for much of the wide geographic variability
in cadmium levels.
These solubility considerations complicate the evaluation of the ulti-.
mate hazard from water-borne cadmium in a manner analogous to that found
with mercury. Sediments containing cadmium concentrations as high as
39,400 ppm would only furnish about 10 ppb of cadmium to a solution at
19
a pH of about 8.0. Biochemical oxidation of the organic matter in the
sediment, however, can increase the percentage of cadmium released to as
20
much as 89 percent depending on the sediment type. Therefore, it is
not sufficient to consider only water concentrations. While immediate
availability is important in evaluating the immediate hazard, the total
;
burden of the aquatic system is more important in evaluating the long-
range hazard. As has already been found in the case of mercury, it is
unwise to simply assume that a toxic element will stay deposited in the
sediment, and therefore unavailable to cause damage to the eco-system.
Sources of,Cadmium in Water_
Electroplating - The greatest- source of direct industrial effluents of
cadmium to the v/ater has traditionally been from the electroplating indus-
try which consumes 50 percent of the total cadmium. It has been estimated
that at least 50 percent of the cadmium used in electroplating operations
21
is lost, with some operations losing as much as 97 percent of the cadmium.
This means that 1,429,000 kg (1,575 tons) of cadmium has the potential of
becoming a contaminant in the water of the U.S. Although some electro-
plating operations treat their own effluents and handle their own sludge,
most of the wastes end up either in municipal sewage treatment plants or
as direct effluents to the water. One set of data indicates that only half
of the cadmium influents to New York City sewage treatment plants are
22
removed to the sludge before discharge. Since New York City is expected
to have better treatment than many communities two-thirds of the cadmium
111
-------
from electroplating wastes may be assumed to be eventually lost to the
water on a nationwide scale. This amounts to at least 952,000 kg (1,050
tons) of cadmium from this source0
While their impact is not.known at this time, recently promulgated regula-
tions by EPA (40 CFR 413) are expected to decrease the level of contamina-
23
tion from cadmium electroplating operations. The promulgated regulations
require the use of best practicable control technology (BPT) for several
subcategories of the electroplating industry as the basis for the interim
effluent guidelines and standards. In addition, EPA has simultaneously
proposed effluent guidelines and limitations requiring application of best
available technology (BAT) and pretreatment standards to existing sources
and performance and pretreatment standards for new sources. The proposed
BAT standard calls for the elimination of the discharge of pollutants
through use of both inprocess and end-of-process controls,'
Other Industry - An estimate of effluents from other industries can also be
obtained using the same New York data. If an estimate of cadmium dustfall
2
of 0.1 mg/m /month is applied throughout the New York City area and sub-
tracted from the contributions of storm water and industries other than
electroplating given in the report, then 23,000 kg of cadmium per year is
emitted from nonelectroplating industry in the New York area,, The New York
SMSA contains about 10 percent of the total manufacturing firms in the U.S.;
the New York City area proper probably contains less than half of that,
and especially no smelters or other handlers of zinc besides the electro-
plating and photoengraving industry. Hence a conservative estimate of
the total amount of cadmium effluent from industries would be around
580,000 kg. If one-third of this is removed by sewage treatment plants
around the country, then 390,000 kg of cadmium is emitted, to the water.
As with the electroplating operations, EPA has recently imposed effluent
standards for cadmium on other industries. While no effluent standards
exist specifically for cadmium under EPA's regulations for inorganic
112
-------
chemicals, regulations for the control of effluents from nonferrous
metals manufacturers (40 CFR 421) have been expanded to include cad-
mium from copper, lead, and zinc operations.
Sewage - The sewage treatment plants also receive much of their cadmium
loadings directly from the population. The New York City study indicates
that 0.010 kg/day/thousand persons (0.023 Ib/day/thousand) is a reason-
able estimate. A recent study has provided further analysis on a por-
tion of the New York City data (Bowery Bay Area) and has also examined
24
heavy metals in Pittsburgh and Muncie, Indiana. Residential loading
factors for cadmium in each of the cities were calculated to be the
following: 0.007 kg/day/1000 persons (0.016 lb/day/1000) in the Bowery
Bay area of Nex* York City, 0.005 kg/day/1000 persons (0.011 lb/day/1000)
in Pittsburgh, 0.003 kg/day/1000 persons (0.006 lb/day/1000) in Muncie.
All of these residential loading values are much too high to be attrib-
utable ^solely to normal urinary and fecal excretion. At an assumed
>
maximum excretion rate of 45 jig/day/person, cadmium loadings due directly
to excretion would be a couple of orders of magnitude lower than the above
residential loading values. It is more likely that the cadmium content
*
of the water itself and the volume of water used per capita per day (1/2 -
3
3/4 m ) are the major contributions to this loading. At a value equal to
the limit set by the USPHS (10 ppb), the amount of cadmium per day for
1000 persons would be 0.005 kg to 0.0075 kg.
Assuming that these values roughly represent the range of loading factors
attributed to the residential population, applying an average value of
0.005 kg/day/1000 persons to the 120 million people living in urbanized
areas would provi-.:_' a total loading of 600 kg/flay of cadmium. If one-
third of this is removed by treatment plants prior to discharge into the
water, then 146,000 kg of cadmium is discharged into the waters of the
U.S. each year due simply to the residential population. While this may
be considered a discharge, some doubt as to the use of this number as a
113
-------
measure of additional pollution may arise since much of the cadmium
would have been naturally occurring in the water and not a result of
the discharge of wastes,
Drainage - Even though the above figures are only rough estimates, providing
orders of magnitude of cadmium emission to the water, comparable estimates
are not as easily obtained for many other sources. No means of estimating
wastes from acid mine drainage or ore benefication processes have yet been
determined. Mine drainage from Tennessee, which ranks first in U.S. zinc
production, and from the Jintsu region in Japan, do not produce concen-
trations in nearby rivers exceeding the USPHS standard for drinking water;
however, cadmium content in the sediment was between 1.29 ppm (Jintsu River)
25
and 30 ppm (Holstom River). Long-range studies of concentration changes
and accurate models for cadmium deposition and exchange in the sediment
are necessary before any estimates of cadmium additions can be made,,
Neither have leaching rates, both from uncontaminated soil and from soil
with agricultural chemicals, been determined. There is obvious accumu-
lation of cadmium in the sediment of lakes that drain presumably uncon-
taminated soil, with concentrations 2 to 60 times higher in the sediment
than in the surrounding soil and with the higher concentrations found in
26
the deeper parts of the lakes. This implies that some degree of leach-
ing does occur.
Air Emissions - The contribution of cadmium from the ambient air to the
water burden can be estimated by using the total annual emissions of cadmium
given in Section III and assuming that 10 percent of it falls out or eventu-
ally leaches into the water, (Though this is 5 times the amount which woul-'
be estimated by assuming emissions evenly spread out over the country
and direct fallout on the surface of major water areas of the U.S., it is
not considered an unreasonable figure. Most industry is near some source
of water and fallout over these waters would be assumed to be much.higher-
In addition, a lot of other land area drains into these major water areas|
114
-------
These assumptions yield an estimate of 153,000 kg of cadmium in the v/ater
as a direct result of air emissions.
Table 18 summarizes the above emission information to demonstrate the
relative contribution from each source sector. As can be seen from this
table the contribution from air to water is small, but not insignificant,
in light of other known estimates. Much information is still lacking
which could have a major impact on air's percentage contribution. In
addition, as more stringent controls are put on emissions of effluents
from sewage treatment plants and industrial sources, especially with
regard to heavy metals, air's percentage contribution, if total emissions
are left unchecked, can be expected to rise.
Table 18. ESTIMATED EMISSIONS OF CADMIUM TO WATER BY DIFFERENT SOURCES
1000 kg ' s per year
Percent of total
From electroplating
From other industry
From sewage (water supply)
Mines etc.
Leaching-agricultural et al.
Air emissions
Total
952
3.9_a
146
153
>1641
< 58
< 24
< 9
< 9
CADMIUM IN SOILS
While soil is not normally considered a direct source of cadmium in the
body, except through fugituve dust, the levels of cadmium in the soil
•
can influence the amount of cadmium that is absorbed by crops or leached
into the water. Since cadmium in. the. air eventually gets washed, rained,
or otherwise deposited out and cadmium in the water precipitates out,
most of the accumulation of cadmium in the biosphere ends up in the soil.
115
-------
It is therefore important to understand how man's activity has increased
cadmium in soil and what impact this is expected to have*
Natural Levels
The cadmium concentrations in uncontaminated soils vary widely depending
on, among other things, the underlying rock. An analysis of 2000 soil
samples in the USSR indicated cadmium concentrations between 0.01 and
0.7 ppm, -with most values less than 0,1 ppm. For soil on a base of gran-
ite, the average zinc and cadmium concentrations were 60 and 0.1 ppm
respectively, while the corresponding values for soils on basic igneous
rocks are 130 and 0.2 ppm. These values are very close to what is found
in the rock itself. In areas of zinc ore outcrops and near surface depos-
27
its, cadmium levels as high as 100 ppm have been foundo
Other generalizations from such studies include: amounts of cadmium and,
zinc are higher in soils containing an appreciable amount of organic
matter than in low-organic soils; cadmium is higher in soils of low per-
meability (clay) than in loose-textured soils (sands, permeable podzols,
28
laterites). While the latter relationship is a result of the ease of
t
leaching, the former relationship is likely a result of the concentration
of cadmium and zinc in living organisms and their affinity to sulfur atoras
found in organic matter.
Anthropogenic Contribution
Increased cadmium levels in soil as a result of man's activity can arise
»
from deposition or washout of cadmium'from the air, irrigation with water
containing cadmium, or the addition of cadmium'from application of fer-
tilizers or pesticides. As exact concentrations of cadmium cannot be
given fox- each of the above applications, expected ranges of cadmium
levels will be used to discuss potential impact.
116
-------
Monthly rates of cadmium deposition have been found to range from
2 2
0.006 mg/m in extremely rural areas to 4 mg/m in areas near smelters.
Assuming that there is no removal process, organically or by leaching,
and that all pf this cadmium accumulates in 20 cm of soil with density
o
1.38 g/cm , it would take from 7 months to 383 years to raise the con-.
2
• centration of the soil 0.1 ppm. A dustfall rate of 0.04 mg/m /month,
a rate typically found in residential areas and probably higher than
expected for agricultural.regions, would imply that almost 60 years would
be required to raise the level of soil by 0.1 ppm.
These figures would tend to indicate that the addition of cadmium to the
soil as a result of airborne deposition is not likely to contribute sig-
nificantly to cadmium levels in the soil, except over extended periods of
time or near particular sources. An idea of how far the influence of a
source can be seen is available from a study of a zinc smelter near
Helena, Montana. The cadmium concentration in the top 2.5 cm of unculti-
vated soil were 68 ppm, 30.6 ppm, 17 ppm, 4 ppm, 2.0 ppm and 0.3 ppm at
•f
distances of 1.7, 2.4, 2.6, 6.4, 24.1, and 153 Ion (1, 1.5, 2, 4, 15, and
29
95 miles) respectively. This would seem to imply that within a 200 kra
radius of such a source, significant contributions are being made to the
k
cadmium concentration of the soil.
Special note should be made of the assumptions used above in determining
the time scales needed for the increase of the cadmium concentration in
the soil by 0.1 ppm. A twenty-centimeter mixing depth was used to imply
an even distribution of cadmium in a cultivated field. While this is
appropriate at the time of plowing, it does not accurately represent
periods between such times. The top inch of uncultivated soil can have
values 2 to 3 times that found in cultivated soil when high levels are
under consideration; however, this difference decreases with the rate of
deposition and is not likely to be of any significance under normal con-
ditions. In addition, because of the potential direct absorption of
deposited cadmium by plants and the additional possibility of concentrating,
117
-------
(discussed below), it should not be assumed that all of the dustfall ends
up in the soil or that the levels found in the plants at harvesting time
can be predicted directly from the concentrations in the soil at the t
of planting. In nonagrieultural areas, the concentrating mechanisms of
plants and wildlife will produce higher concentrations of cadmium in the
organic matter at the top of the soil. Another major assumption was that
•no leaching occurred. Indications that the cadmium level in the top
2.5 cm of uncultivated soil is similar to that of cultivated soils would
imply leaching to at least 20 cm. Any leaching past this depth v/ould
further lengthen the time needed to increase the cadmium concentration by
0.1 ppm. Ihough this leaching rate is not known, it is dependent upon
total rainfall, permeability of the soil, the acidity of the soil and the
form of the cadmium.
Mosses have been shown to be good indicators of cadmium levels in both
30
the soil and the air due to their ion-exchange properties. Some esti-
mate of the impact of ambient levels of cadmium deposition on soil can
determined by comparing samples of moss saved from several decades ago
with those from present times. Mosses from rural regions around Uppsala,
Sweden, were found to have a mean increase from 0.2 to 0.3 ppm during an
intervening 20 to 30 years. In the city, a mean increase of 0.2 to
31
0.4 ppm was observed during a 50-year period. If the concentrations in
the mosses accurately reflect concentrations in the soil, time scales
similar to the ones derived above are represented. However, without data
on actual or probable changes in ambient concentrations of cadmium and on
the rates of deposition, absorption, and concentrating of cadmium in
mosses, no precise conclusion can be made<.
Cadmium may also be added to the soil by means °of irrigation with water.
Any contribution from cadmium in rainfall would already be considered
under dustfall rates, so this is limited to efforts undertaken by man to
artifically supply water to crops. The importance of considering .this
limited area of possible pollution arises because the majority of man's
exposure comes from his food.
118
-------
Cadmium levels in surface water and water supplies arc almost always less
than 10 ppb and are likely to be greater than 0.1 ppb. Using these figures,
a range of volume of water needed to be added to soil to raise the cad-
mium levels by 0.1 ppm can be determined. Again, the assumptions are
that all of the cadmium will accumulate in the top 20 centimeters, that no
leaching from the top layer or other removal process will occur, and
3
that the density of soil is 1.38 g/cm . If the concentration of the water
were 10 ppb, 2.76 m of water would need to be added to every square meter
of soil to raise the concentration of cadmium by 0.1 ppm; with water of
3
0.1 ppb, 276 m would be needed. These vo'.
falls of 2.76 and 276 meters respectively.
3
0.1 ppb, 276 m would be needed. These volumes are equivalent to rain-
Since irrigation is used to supplement rainfall in more arid regions
where at least one meter of water can be expected to be applied each year,
it becomes evident that irrigation with water which has a concentration
equivalent to the USPHS standard for drinking water can quickly bring up
the levels of cadmium in the soil. However, on a national average basis,
the volume of water used for purposes or irrigation is between 0.5 and
32. 32
0.6 v.: /a . Assuming 0.6 m /m is used, 4.6 to 460 years would be needed
to raise the cadmium content of the soil by 0.1 ppm when irrigating with
water of 10 ppb and 0.1 ppb cadmium, respectively. With 46 percent of
the surface water i~n the U'.S". at concentrations greater than 1 ppb and
the possibility of increased contamination from piping, the probability
of producing increasing concentrations of cadmium in soil which is used
-";
for food and fodder is great, given the assumptions stated and discussed
previously.
Other soil contamination arises from the addition of organic or chemical
material which contains small amounts of cadmium. Recently, the use of
sewage sludge as a fertilizer for crop lands has been receiving increased
attention; however, with the heavy metal contamination, including cadmium
32
at levels of 10 to over 800 ppm (dry weight), sludge is also receiving
increased concern. If sludge were used in even small amounts on crop
119
-------
lands with its present composition of cadmium, the contamination of the
soil by the sludge would be large and would greatly outweigh the contribu-
tions from air or water.
Phosphate fertilizers are probably the most common agricultural source of
heavy metals. Cadmium contained in the mined phosphate rock is likely
to remain in the phosphoric acid fraction which eventually is converted
to normal superphosphate, triple superphosphate, and diammonium phosphate
fertilizers« Additional contaminations of the fertilizer by cadmium can
arise from the sulfuric acid used in the "Wet-Process" o-f phosphate
fertilizer production. Since the fertilizer industry consumes 50 percent
of the sulfuric acid produced in the U.S., and since a high grade of
sulfuric acid is not needed for the acidulation of the phosphate rock,
sulfuric acid produced as a byproduct of smelter operations is used.
Quantification of this usage has not been possible but is generally felt
to be significant since a high degree of contamination from metals often
occurs in the smelter sulfuric acid.
An analysis of Florida-produced fertilizers sold in Indiana was conducted
at Purdue, and a range of 3 to 47 ppm of cadmium (for 18-46-0, 0-46-0,
33
10-34-0, 13-52-0, and 16-48-0 fertilizers) was found. At a high rate
2
of phosphate application to the soil of 200 kg/hectare/year or 20 g/m /year
2
(178 Ibs/acre), this would indicate that 60 - 940 ug/m of cadmium is
added each year. With the usual assumptions, about 30 to 460 years of
adding fertilizers is needed to raise the cadmium content of the soil by
0.1 ppm. With the mean levels of cadmium in the various types of fertil-
izers ranging from about 3.5 to 14, over 100 years would be a more reason-
able estimate.
a
Although the addition of fertilizer to the soil does not seem to indicate
a high potential for impact on soil concentrations, studies have shown
some relationships between fertilizer application and cadmium content in
120
-------
plants. Therefore, the importance of the cadmium content of the fertil-
izer is not necessarily precluded. This will be discussed in greater
detail in the following sub-section.
Reviewing the figures derived above (see Table 19) some conclusions can
be reached. Since all of these figures were based on the same assump-
tions as to removal processes in the soil, they can be considered compa-
rable. However, it should be remembered that the form with which the
cadmium enters the soil may well depend upon the means of entry, and this
form could have a significant impact on leaching and uptake of the
cadmium.
The contribution from air can have the greatest potential for impact,
but only in a highly polluted situation. Generally, areas which receive
irrigation by water or with a cadmium content approaching the USPHS stan-
dard are most likely to obtain their greatest burden of cadmium via the
water. This burden is, of course, directly proportional to the volume
of xjater as well as the concentr.a.tJLjaa,., The fertilizer impact on cadmium
in the soil is expected to be minimal. The figures given indicate more
than 100 years for fertilizer with a normal concentration of cadmium but
assuming a high degree of fertilization.
*
BIOECOLOGY AND THE FOOD CHAIN
Cadmium's affinity for sulfhydryl compounds indicates that it is easily
assimilated into living matter. If this assimilation is coupled with
retention (intake minus excretion), then living organisms x^ill accumulate
'
cadmium in concentrations higher than that found in the source of their
nutrition. Such a concentrating mechanism has already been shown (Sec-
*
tion IV) to be applicable to man. and. other animals (used in experimental
studies). If accumulation is also applicable to lower organisms along
the food chain, the concentration of cadmium will buildup exponentially
as one moves up the food chain implying that a very small increase at the
121
-------
Table 19. NUMBER OF YEARS TO INCREASE THE CADMIUM
CONTENT OF THE SOIL BY 0.1 PPM
Airborne
deposition
Water - ^
0.6m An /year
Fertilizer
20g/m2/year
2
0.006 mg/m /month
2
0.04 mg/m /month
2
4.0 mg/m /month
0.1 ppb
1 . 0 ppb
10.0 ppb
3 . 0 ppm
10.0 ppm
47 . 0 ppra
a
Years
383
58
0.58
460
46
4.6
460
138
29
Assuming no other sources of cadmium, no removal"
processes, and an even distribution of cadmium in
the top 20 cm of soil.
122
-------
lowest level can produce .a much higher potential for damage to the
higher organisms.
Evidence of Bioaccumulation
A few of the studies which indicate that cadmium is concentrated at
various levels of the food chain and their implications are summarized below:
• Concentrations of cadmium in major components of vege-
tation were higher than in the soil. The concentrating
factor was about three, with the highest cadmium con-
centration in the leaves and twigs and the lowest in
the trunk, disregarding the variation of differential
accumulation among species. Zinc, on the other hand,
had a concentrating factor of 1.5 to 0.5. Should
this preferential accumulation of cadmium over zinc
continue up the food chain, it could aggravate the
impact of cadmium on '.vigher organisms due to zinc-
cadtnium interactions. Of particular interest is that
cadmium levels are highest in the leaves and twigs,
that part of the tree most likely to serve as immediate
food to animals.
t
'#
• The litter of the forest floor contains even higher cad-
mium concentrations than found in the leaves or twigs, a
factor of three to ten times that found in the underlying
soil. • Since the more decayed litter had the higher con-
centration of cauhii'tnrr; it can- be assumed that while other
materials are leaching down as the litter decays the cad-
mium is remaining in the top layer, eventually to become
the top soil.
o An examination of tall fescue and soil off to the side of
heavily traveled roads showed elevated concentrations in
the grass when compared to average soil concentrations.
This was not always the case when comparing just the top
five centimeters of the soil, when the soil concentration
was greater than 0.50 ppm, or with other grasses. Aside
from possible differences in retention rates of the
grasses, a possible explanation may be that as the con-
centration of the soil increases, the rate of uptake
decreases.
• Earthworms were found to have, on the average, 17 times
as much cadmium as the soil in which they dwelt.^' This
concentrating factor may even be low, as the worms were
123
-------
removed from their soil for four days prior to analysis,
allowing the possibility of significant removal of cad-
mium if a short half-time is involved. The zinc/cadmium
ratio decreased from 139 in soils to 56 in the earth-
worms .
« The concentration of cadmium in grasshoppers near a
smelter (2.4 km) was almost four times that of the grass
in the area. Generally, the concentrating factor for
grasshoppers was 1.3.^°
o Shellfish have a biological accumulation of cadmium rang-
ing from 100 to more than 3000 when compared to seawater.^9
o Vegetables cultivated for human consumption in soil with
3 ppm cadmium within a 4-mile radius of a smelter contained
as high as 10 ppm cadmium. Some vegetables contained as
little as 0.05 ppm.40
• Cadmium levels as high as 20.4 ppm have been found in fish
in polluted waters.f*^-
On the other hand, many studies show no concentrating above intake levels.
Some of these findings are likely a result of;
o Cadmium levels in the soil being so high that the absorption
mechanism for cadmium is damaged or hindered.
e The whole body concentration being considered, while the
cadmium is concentrating and causing damage in only selected
parts of the organs (e.g. liver, kidney, etc.).
e Parts of the organism being considered are different from
those where the cadmium is concentrating»
Of course, a large percentage of the studies indicating concentrating
factors equal to or less than one are simply due to differential reten-
tion rates among specieso
Since cadmium is apparently not concentrated uniformly throughout the
*
food chain, a small increase in environmental exposure cannot automati-
cally be projected to a major impact at the higher levels of the food
chain. A better idea of the various concentration rates as one moves
up the food chain would be needed in order to predict the overall
124
-------
concentration upon final ingest ion by humans. Combinations of concen-
tration ratios may not actually provide any real overall concentrating.
Of particular concern would be predators who rely upon a single source
of food if that source of food had a large concentrating factor.
.In general, the relative cadmium content of the four major categories
of food appears to represent a concentrating of cadmium as it passes
up the food chain. Fruits and vegetables have the lowest cadmium con-
tent, followed by grains and cereals, then dairy products, and finally
meats and seafoods, which contain twenty to a hundred times more cad-
42
mium than the lower level foods. Tables 20 and 21 provide available
measurements of cadmium levels in a variety of common foodstuffs.
Anthropogenic sources of cadmium to terrestial environments are depo-
sition from atmospheric emissions, addition of agricultural chemicals,
and irrigation. In nonagricultural regions, only atmospheric deposi-
tion of cadmium is of importance. Aquatic environments have the same
sources5 as mentioned" earlier, i.e., leaching, sewage, atmospheric depo-
sition, and industrial effluents.
Differential Absorption"
What is of major concern, especially since the major exposure of man
to cadmium is through food, is the potential that each means of con-
tamination has for entering the food chain and raising the cadmium
content of the food. Since the various sources of contamination of
water do not have a major influence on the levels of cadmium once the
cadmium is in the water, the percentage contribution of the various
water contaminant sources to the burden of cadmium in the aquatic food
chain may be considered, to, be the same as the percentage contribution
of the source to the contamination of the water. With terrestial life
forms, however, such factors as the route of exposure (from soil or
125
-------
Table 20. CADrilUM CONTENT AND ZINC/CADMIUM RATIO OF SOME FOODS43
Food
Vegetables
Potatoes
Onions
Carrots
Cabbage
Kohlrabi
Beans, kidney
Beans, string fresh
Tomato, fresh
Tomato, canned
Greens
Peas, green, fresh
Spinach, fresh
Beans, string, canned
Grains
Corn - cornmeal
Wheat
Bread, white
Bread, dark
Fruits
Tangerine orange
Apricot, canned
Banana
Grapes
Apples
Meats
Pork
Beef
Chicken
Whale
Kidney
Dairy
Milk, whole
Cheese
Butter
Eggs, whole
Seafood
Oysters, fresh
Clams, fresh
Shrimp
Tuna
Sardines
Salt herring
Mackerel
Freshwater fish
Carp
Bass
Cnilmium content
ppm (UC./K) wet weight
Investigator
A
0.03
O.01
0.30
0.07
0.09
0.07
0.01
0.03
0.02
0.06
0.04
0.45
0.034
0.12
0.15
0.22
0.15
0.02
0.01
0.03
0.25
0.89
1.25
0.52
0.12
0.56
0.09
3.66
0.31
0.24
0.21
0.13
B
0.017
0.0077
0
0
0.013
0.044
0.064
0.024
0
0.042
0.054
0.007
0.0065
0.034
0.060 .
0.43
0
0.039
0.021
0.066
0
c
0.038
0.012
0.041
0.009
0.019
0.032
0.128
0.025
0.046
0.003
0.011
0.004
0.016
0.054
0.027
0.33
0.015
0.62
0.28
0.032
0.066
0.011
0.018
0.003
D
0.034
0.052
0.033
0.036
0.037
7n/M ratio
Invest Icator
A
290
89
17
26
25
12
125
20
6
7
200
5
3
30
130
5.5
35
72
5
7
45
63
60
40
350
60
62
87
150
c
85
106
67
201
1120
45
•
71
230
130
•
250
150
52
1080
1220
600
31
950
165
67
220
77
770
530
5800
A. Schroedcr, ct al. (Rcf. 43A)
B. R.iur.u nnd Sporn (Rcf. 43B)
C. Ishlzakl. et al. (Rcf. 43C)
D. OKNL (Rcf. 43D)
126
-------
Table 21. CADMIUM CONTENT IN DIFFERENT FOOD CATEGORIES IN USA.
TOTAL NUMBER OF SAMPLES: 30
44
Type of food
Dairy products
Meat, fish and
poultry
Grain and cereal
products
Leafy vegetables
Legume vegetables
Root vegetables
Garden fruits
Fruits
Oils, fats and
shortening
Sugar and adjuncts
Beverages
Potatoes
Cadmium content,
ppm wet weight
1968-1969
No. > 0.01
10
21
27
" 27
16
24
25
1-5
27
18
8
-
Maximum
0.09
0.06
0.03
0/"VQ
.(Jo
0.03
0.08
0.07
0%33
0.13
0.07
0.04
-
1969-1970
No. > 0.01
9
22
27
28
10
27
27
10
28
27
9
29
Maximum
0.01
0.03
0.05
0.14
0.04
0.08
0.07
0.07
0.04
0.04
0.04
0.08
127
-------
from direct deposition) and the form of the exposure (tied to phos-
phorous or dissolved in water) can make an important difference on
the degree of impact for any one source.
Atmospheric deposition and direct absorption by leaves of plants can
be inferred from a study of a wide range of foods grown in the region
of a refinery in Annaka City, Japan. The cadmium content of the fruits
and root vegetables was much less than that found in leafy vegetables
45
and mosses. This tendency for leaves to have a higher cadmium con-
tent has been mentioned above for a forest system. However, whether
the elevated levels of cadmium in the leaves are a result of direct
deposition and absorption, or a result of the preferential transport
of cadmium to leaves from roots in the soil, cannot be determined from
this data. A trunk inoculation of radio-labeled CdCl2 in an Eastern
red cedar showed that 66 percent of the cadmium went to the crown of
46
the tree. This tends to indicate preferential transport, a phenom-
47
enon easily ascribed to the forest study also. However, another study
has shown that direct absorption of cadmium in an aqueous solution by
leaves, a situation likely to be found during rain, does occur. This
latter study involved CdCl^ used as a fungicide on an apple o.rchard;
the leaves were found to have a high level of cadmium immediately
after spraying, slowly losing cadmium to the apples as the fruit
matured.
The above study on apple trees could be extrapolated to the direct
contamination of food by precipitation or aerial irrigation. The
cadmium content of the apple leaves was raised considerably, to levels
twice as high as found 100 days later, by a one-time spraying of the
aqueous solution of fungicide,, If this absorption from an aqueous
solution is characteristic of other crops, it is not unlikely that
many repeated sprayings, as a method of irrigation or pest control,
could also increase cadmium levels in leaves and other parts of
plants.
128
-------
Uptake of cadmium from the soil by vegetation has already been demon-
strated under the discussion of food chain accumulation and concen-
tration. The sources of cadmium contamination of agricultural soil
have been discussed under the previous heading. As was the case with
the water environment, if the source of the cadmium does not have an
impact on absorption rates from the soil, then the percentage contribu-
tions to contamination of the food due to levels of cadmium in the soil
resulting from outside sources would be the same as the source contri-
butions to the excess levels of the soil.
» •
There appears, however, to be some disagreement over the role of phos-
phate fertilizers and the uptake of cadmium by plants. Schroeder
et al. has reported conflicting results in two different studies.
In one study, vegetables grown in soil with superphosphate fertilizer
had measurable amounts of cadmium while in a control group only one
vegetable contained a detectable amount. In the other study no such
significant increase of, cadmium,, iru vegetables grown in soil fertilized
*
with superphosphate appeared to occur.
Further investigations of the impact: of phosphate fertilizer on the
* 49
uptake of cadmium are being conducted, in a current study by Purdue.
The results for the first years of their experiment are shown in
Table 22. The treatment rates give the total amount of phosphate
fertilizer added over 20 years; it was added periodically, 20 percent
every 4 years. The "second year corn" was planted the year that soil
received its once every four-year fertilization (for those soils which
received fertilizer). The "first-year corn" was the crop the year
before the periodic fertilization. Levels of cadmium in soil in all
cases were below the detection limit of 0.02 ppm.
From the data on the first year corn leaves, which shows a much greater
cadmium uptake on the plot never receiving fertilizer than on the plots
receiving the phosphate, the phosphorous was postulated to be tying
129
-------
up the cadmium, making it unavailable to the corn. This seemed
to be reinforced by the application of phosphorous fertilizer and
decreased levels in the second year corn. However, if it is assumed
that Treatment A provides a normalizing group, which takes into account
uncontrolled environmental exposures, then it is found that, in general,
the cadmium content is increased with the addition of the fertilizer.
The postulate that phosphorous tends to decrease cadmium uptake also
does not account for the fluctuations of cadmium levels between the
zero and maximum levels of fertilizer application in the first year
corn leaves.
Table 22. CADHIUM CONTENT IN PLANTS GROWN IN LONG-TERM
PHOSPHOROUS FERTILITY PLOTS49
Fertilizer level
kg P205/ha/20 years
Treatment A (0)
Treatment B (225)
Treatment C (1120)
Treatment D (1120)
Treatment E (2800)
Corn leaves
first year ,
ppm
1.12
0.31
0.71
0.93
0.25
Corn leaves
second year,
ppm
0.27
no data
0.71
0.47
0.39
Normalized corn
leaves second year,
ppm
1.12
no data
0.50
1.95
1.62
Obviously, more work needs to be done to identify the impact of phos-
phate fertilizer. Insignificant impact on yearly variations is
expected when the total environmental exposure is considered; however,
as with any contaminant added to the soil every year, it could eventu-
ally build up to values sufficient to increase the level of cadmium in
the food. As both air and water are more likely to add cadmium to
food, both through the soil and by deposition, research on well-
controlled experiments to determine these mechanisms are also
130
-------
INTERMEDIA ASPECTS OF CADMIUM FLOW
As has already been discussed, a number of routes exist through
which levels of cadmium in one media can influence cadmium levels in
another media. While the only source of ambient air concentrations
is the direct emission of cadmium to the air by various process and
combustion operations, all other media can receive cadmium directly
or indirectly from the air" emissions, waste water effluents, natural
soil levels, etc.
For example, if food derives most of its cadmium from the soil, either
directly as in crops or indirectly as in livestock, it is possible to
estimate the amount of cadmium added to the food by other sources
which affect the levels of cadmium in soil used for agriculture. Using
the time scales presented in the previous discussion on soil pollution
2
and assuming average values of 0.04 mg/m /month for deposition from the
-•».
air, irrigation of 20 percent of the crops with water containing 1 ppb,
t
and fertilization at the rate of 100 kg/hectare/month with fertilizer
containing 10 ppm cadmium, the average percentage contribution to
soil pollution by external sources in agricultural regions would be
on tne order of 64 percent, 17 percent, and 19 percent for air, water,
and fertilizer respectively. Under the assumption that the form of
cadmium in each of these sources would be equally absorbed, the above
percentages would be comparable to the percent contributions to above
natural levels in the crop.
/
Unfortunately, these figures are not so easily interpreted to give
actual percentage contributions of cadmium in the food. If deposition
is a viable route of absorption by the plants, then the contributions
from air and water (irrigation by sprinklers) are expected to be greater.
In addition, the form of cadmium in each media is important to know as
its composition can help to control its absorption rate.
131
-------
Further complications arise Xirhcn considering the "natural" cadmium in
the soil. Data is not currently available to give a good indication
as to how soil levels alone can influence the cadmium content of the
plants or whether plants have increased uptake only after a background
level of cadmium is bypassed. The long-term emissions of cadmium may
also be responsible for a significant portion of the cadmium currently
found in soils. More research is needed on these interactions before
a full accounting can be made.
The above calculations must be considered only rough approximations at
average contributions. Obviously, the soil levels of cadmium
contributed by different sources will vary greatly depending upon the
actual location of the agricultural activity and the amount of irriga-
tion or fertilization applied. The particular crop grovm in each area
would also influence the average value. Such an assessment is not
possible xtfith the available data.
Table 23 summarizes the intermedia aspects of cadmium as discussed
above and in the previous subsections. This table does not go beyond
the primary impacts of one media on another media; i.e., while depos-
*
ition from air may directly add 64 percent of the excess cadmium to
soil in agricultural areas, presumably 10 percent of the cadmium in
the water used for irrigation is attributable to air emissions so that
in total, 66 percent of the cadmium would be a result of air emissions.
Since the primary concern of this document is the contribution from air
and air's secondary impacts are expected to be small (smaller than the
error in the primary calculations), no attempt at quantifying secondary
impact has been made.
132
-------
Table 23. PRIMARY INTERMEDIA CONTRIBUTIONS OF CADMIUM
^Sources
Media
Air
Water
Soil
Other
Ambient Air
Basically 1007.
None
Only from windblown
dust; basically
negligible
Only from spraying
of fungicides or
pesticides; basi-
cally negligible
Water
U)
peposition and wash
put from the air
into the water; pro-
bably less than 10%
Large volume from
electroplating',
other industry and
sewage; probably
80% or more
Leachings from soil
are probably minimal,
mine drainage could
be large; estimate
about 10%
Possibly leachings
from fertilizer;
felt to be
negligible
Soil
Not well known,
only anthropogenic
source to nonagri-
cultural regions;
estimated average
of 64% of all out-
side sources in
agricultural
regions
Only in irrigated
regions; average
17% of all outside
sources in agri-
cultural regions;
sediments concen-
trate the cadmium
Levels are expected
to be naturally
around 0.1 ppm or
less
Topsoil gets high
levels from bio-
accumulation; fer-
tilizer provides
average of 19% of
all outside sources
agricultural
regions
Food and
Tobacco
Approximately 64%
of all additions
to soil used for
food is a result
of air; higher if
deposition reten-
tion is great
Assume same as
contribution to
soil —17%; could
be more if sprin-
kling is used for
irrigation
Direct correlations
between soil levels
and plants but some
could be a result
of air emissions
Fertilizers only
19% of all outside
sources; gains and
losses in
processing
-------
REFERENCES
1. CITE II. Section 3.1.1.
2. In house data from study of national participate standard attainment
problem.
• 3. Laamanen, A. Functions, Progress and Prospects for an Environmental
Subarctic Base Level Station. Work Environ. Health. 9, 17. 1972.
CITE II.
4. Huey, N.A. Survey of Airborne Pollutants, in Helena Valley, Montana.
Area Environmental Pollution Study, U.S. Environmental Protection.
Agency. 1972. p. 25. CITE II.
5. Kitamura (personal communication), CITE.
6. Hunt, W. F., C. Pinkerton, 0. McNulty, and J. Creason. A Study in
Trace Element Pollution of Air in 77 Midwestern Cities. (Presented
at the 4th Annual Conference on Trace Substances in Environmental
Health, Univ. of Missouri, Columbia, Missouri. June 23-24, 1970.)
7. Bowen, H. J. M. Trace Elements in Biochemistry. Academic Press.
London and New York. 1966. ORNL»
Lagerwerff, J. V., and A. W. Specht. Contamination of Roadside Soil
and Vegetation with Cadmium, Nickel, Lead and Zinc. Environ. Sci.
Technol. 4, 583 (1970). ORNL.
Development of Nuclear Analytical Techniques for Oil Slick Identifica-
tion, GA9889. Gulf General Atomic Incorporated. January 1970. ORNL.
Filby, R. H., and K. R. Shah. Mode of Occurrence of Trace Elements
in Petroleum and Relationship to Oil-Spill Identification Methods.
Inj Nuclear Method Methods in Environmental Research. Vogt, J. R. ,
T« F. Parkinson, and R. L. Carter (eds.) University of Missouri,
Columbia, Mo. August 1971. ORNL.
8. Pinkerton, C., D. I. Hammer, T. Hinners, V. Hasselblad, J. Kent,
J. V. Lagerverff, E. Ferrand, and J. Creason. Trace Metals in Urban
Soils and Housedust. Amer. Pub. Health Ass. Meet., Atlantic City,
New Jersey. November 1972. CITE II.
9. Ibid.
10. Lee, R. E., R. K. Patterson, and J. Wagman. Particle-size Distribution
of Metal Components in Urban Air. Environ. Sci. Technol., 2, 288.
1968. CITE.
134
-------
11. Lee, R. E., S. S. Goranson, R. E. Enrione, and G. B. Morgan. National
Air Surveillance Cascade Impact Network II, Size Distribution Measure-
ments of Trace Metal Components. Environmental Science and Technology.
Vol. 6, No. 12. 1972.
12. Bernstein, D. M., et al. Uptake and Distribution of Airborne Trace
Metals in Man. Trace Substances in Environmental Health — VIII.
University of Missouri, p. 329-334.
13. CITE II. Section 3.1.2.
14. Schroeder, H. A., A. P. Nason, I. H. Tipton, and J. J. Balassa. Es-
sential Trace Metals in Man: Zinc, Relation to Environmental Cadmium.
J. Chronic, Dis. 20, 179. 1967. CITE.
15. Yamagata, N., and I. Shigematsu. Cadmium Pollution in Perspective.
Bull, Inst. Pub. Health, 19, 1. 1970. CITE.
16. Durum, W. H., J. 0. Hem, and S. G. Heidel. Reconnaissance of Selected
Minor Elements in Surface Waters of the United States. October 1970.
Geological Survey Circular 643. Washington, D.C. 1971. ORNL.
17. Biesinger, K. E., and G. M. Christensen. Metal Effects on Survival,
Growth, Reproduction, and Metabolism of Daphnia Magna. (In manu-
script) 1971. ORNL.
18. ORNf,. Section IIT.C'.2"-.
19. Fulkerson, W., W. D. Shults, and R. I. Van Hook. Ecology and Analyses
of Trace Contaminants. Progress Report. Oak Ridge National Labora-
tory. January 1974. Section 5.2.3.
20. Ibid.
21. Yost, K. J% Personal communication.
22. Data on New York City from study by New York City Environmental Pro-
tection Administration. Provided by Dr. Joel Jacknow.
23. 40 CFR 413.
24. Davis, J. A., and J. Jacknow. Heavy Metals in Wastewater in Three
Urban Areas. Journal Water Pollution Control Federation. Vol. 47.
No. 9. September 1975. p. 2292-2297.
25. Perhac, R. M., University of Tennessee, and T. Tamura, Oak Ridge
National Laboratory. Private communication. Unpublished data. ORNL.
26. Purdue. Section III.
135
-------
27. VInogrndov, A. P. The Geochemistry of Rare and Dispersed Chemical
Elements in Soils, Second Edition. Consultants Bureau, Inc. Now York.
1959. ORNL.
28. Ibid.
29. Helena Valley, Montana. Area Environment Pollution Study. Environ-
mental Protection Agency. Office of Air Programs Publication No. AP-91,
Research Triangle Parks N.C. January 1972. ORNL.
30. Little, P., and M. H. Martin. Biology and Monitoring of Heavy Metal
Pollution, Environmental Pollution. No. 6. 1974. p. 1-19.
31. Gelting, G., and A. Ponten. Heavy Metal Pollution in the Uppsala Area.
Dept. of Plant Biology, Uppsala. 1971. (Internal report: in Swedish.)
CITE II.
32. Purdue. Section II. VI.
33. Ibid. Section II. V»B.
34. Fulkerson, et al. Section 5.2.1.
35. Ibid.
36. Lagerwerff, J. V., and A. W. Specht. Contamination of Roadside Soil
and Vegetation with Cadmium Nickel, Lead, and Zinc. Environmental
Science and Technology. Vol. 4. No. 7. 1970.
37. Fulkerson, et al. Section 5.2.1.
38. Munshower, F. Cadmium Compartmentation and Cycling in a Grassland
Eco-system in the Deer Lodge Valley, Montana. Ph.D. Dissertation,
University of Montana. September 1972. ORNL.
39. ORNL. Section VI.F.I.e.
40. Munshov;er. Loc. cit.
41. Lyons, R. D. Fish Caught Near a Battery Factory on Hudson Contain up
to 1000 Times Normal Cd. The New York Times. Sunday, June 13, 1971.
42. ORNL. Section VLB.3.a.
43. From ORNL. Section VLB.3.a.
A. Schroedcr, et al. Loc. cit,
B. Rautu, R., and A. Sporn. Contributions to the Determination of
Cadmium Supplied by Foods. Nahrung 14, 25-31. 1970.
136
-------
C. Ishissaki, A., M. Fukushima, and M. -Sakamoto. Distribution of
Cadmium in Biological Materials, Part 2. Cadmium and Zinc Con-
tents of Foodstuffs. Jap. J. Hug. 25(2). 207-22 (1970).
D. ORNL unpublished results of analyses of samples contributed
by Purdue University.
I
44. Corneliussen, P. E. .Pesticide Residues in Total Diet Samples (V).
Pestic. Monit. 3., 3, 89. 1970. CITE II.
Corneliussen, P. E. Pesticide Residues in Total Diet Samples.
Pestic. Monit. 3., 1973, in press. CITE II.
45. Kobayashi, J. Air and Water Pollution by Cadmium, Lead, and Zinc
Attributed to the Largest Zinc Refinery in Japan, p. 117 in Trace
Substances in Environmental Health - V. Hemphill, D. D. (ed.).
University of Missouri. 1972. ORNL.
46. • Witherspoon, J. P. and Mara Slawsky. Private communication. ORNL.
47. Ross, R. G., and D. K. R. Stewart. Cadmium Residues in Apple Fruit
and Foliage Following a Cover Spray of Cadmium Chloride. Can. J.
Plant Sci. 49, 49-52 (1969). ORNL.
48. Schroeder, H. A., and J. J. Balassa. Cadmium: Uptake by Vegetables
from Superphosphates in Soil. Science 140, 819-20 (1963). ORNL.
49. Purdue. Section II. V.A.
137
-------
SECTION VI
EXPOSURE TO CADMIUM IN THE ENVIRONMENT
This Section considers the contributions of the various environmental
media (air, water, food, smoking) to the total exposure of humans to
cadmium. It draws upon the ambient levels discussed in Section V to
determine the total intake and utilizes the concept of body burden and
significant exposure levels presented in Section IV. The measure of
body burden, as opposed to total intake, is used in the final analysis
of the relative importance of the various media to better reflect the
differing absorption rates from each media. In addition, to focus
more directly on the possible need for control of airborne cadmium,
particular emphasis is given to defining that portion of water, food,
and smoking exposures that may have originated in cadmium emissions to
the air.
INTAKE LEVELS
Ambient Air
Levels of cadmium in the ambient air have been reported as ranging from
33 3
0.001 ug/m in rural areas to 0.01-0.05 jzg/m in urban areas to 0.3
near cadmium emitting industries. Bodily retention rates for cadmium
in the air have also varied with values from 10 to 40 percent being
•
found in different studies. Such a range is not unlikely since the
degree of retention find absorption depends upon the particle size and
its solubility (chemical form). Table 24 presents possible combinations
138
-------
Table 24. CADMIUM INTAKE AND RETENTION FROM AMBIENT AIR
VO
Ambient
air quality
£,i '-001 "s/™3
San °-°l "S'°3
UrSn «'°5 ^
3
High 0.3 pig/m
Industrial
a
Daily intake
0.020 jig
0.20 pig
1.0 ptg
6.0 pig
Retention rate
107o
25%
40%
10%
25%
40%
10%
25%
40%
10%
25%
40%
Daily retention
0.002 pig
0.005 pig
0.008 pig
0.02 pig
0.05 pig
0.08 pig
0.1 pig
0.25 pig
0.4 pig
0.6 pig
1.5'ptg
2.4 pig
Percent of .
significant level
0.03
0.075
0.12
0.3
0.75
1.2
1.5
3.8
6.1
9.1
22.7
36.4
Based on 20 m /day inhaled.
Significant level is 6.6 pig/day (from CITE),
-------
of these ranges. The total amount of cadmium retained by the body is com-
pared with the "significant level" i.e., the daily intake needed to achieve
200 ppm cadmium in the renal cortex (see pages 87-88).
As shown in this table, even in urban areas with high levels of cadmium
(higher than reported in 14 major U.S. cities) ambient air directly
contributes less than 6 percent of this significant level of retention.
In areas of industrial activity with high levels of cadmium emissions,
much greater percent contributions are likely to be foundo
Drinking Water
The direct environmental exposure of man to cadmium in water comes only
from drinking water. Samples of drinking water from 165 municipalities
around the U.S. were analyzed for cadmium in 1962-63. The collection
was from finished water at points believed to be representative of water
consumed by residents of the communities. Sixteen percent of the samples
were found to exceed the USPHS standard and interim EPA standard of
10 ppb9 with the highest sample containing 30 ppb of cadmium.
A more extensive sampling program of water quality at the cons'umer's tap
2
was reported in 1970. Out of 2595 samples from 969 public water supply
systems, 0.2 percent of the cadmium levels were found to exceed the
standard. The maximum reported concentration was 11 ppb. More recently,
EPA has conducted special studies of water systems where the water is
particularly corrosive to the distribution system and plumbing. Results
of two of these studies have been reported and indicate that in Boston
none of the homes were found to exceed the USPHS standard while in
Seattle 7 percent of the homes sampled exceeded the standard.
3
Because of the interest of the National Heart and Lung Institute and EPA
in the suggested association between heart disease mortality and soft
drinking water, a cooperative study is currently underway providing
140
-------
detailed analysis of drinking water quality and health examination re-
sults. Water samples are being collected at the homes of persons in-
cluded in the current series of the National Health Examination Survey.
This study will provide data on water quality for those chemicals limited
by the drinking water standards and on 86 additional chemicals. *
Because of studies like the one in Seattle, the level of cadmium has
been suggested to be increasing between the inlet and the consumer.
These increases come about as a result of the handling and piping of the
water (zinc galvanizing on an iron pipe was found to contain 360 ppm
cadmium). A high degree of increase was demonstrated in a study of the
water supply in Brattleboro, Vermont. The inlet value of cadmium was
2.1 ppb and values in the town water main ranged from 14 to 21 ppb. The
hot water tap in a hospital also gave a value of 21.0 ppb; however, water
from the tap which had been stagnant in the pipes ranged from 15.0 to
77.0 ppb cadmium.
v
Normally, levels in drinking water are less than 1 ppb (1 ug/,0) of cadmium.
With a daily intake of one to two liters per day and a gastrointestinal
absorption rate of about six percent, this could add about 0.12 ug/day to
~ * •
the ..body burden. If, however, concentrations were closer to the interim
*
EPA Standard of 10 ppb, the amount of cadmium absorbed would be closer to
1.2 ug per day, and in an area with a level of 77 ppb, over 9 ug per day.
Table 25 presents several ranges of water quality and retention rates.
A maximum level of 10 ppb cadmium in the water was assumed as interim
EPA regulations set this limit for cadmium in water "at the tap." Also
/
as stated in the proposed EPA regulations, a daily intake of 2 liters
was used. As is evident from this table, a large amount of cadmium,
as much as 30 percent of the significant level, may be retained due to
drinking water. For persons engaged in strenous activity, in xvarm cli-
mate:;, or otherwise disposed to drinking more than 2 liters (8-1/2 cups)
of water (coffee, tea, etc.), the contribution of water to the body
burden of cadmium could be 50 percent or more.
141
-------
Table 25. CADMIUM INTAKE AND RETENTION FROM WATER
ISJ
Water quality
0.1 ppb
1.0 ppb
5.0 ppb
V.
10.0 ppb
Daily intake
0.2 us
2.0 ftg
10 pg
20 jig
Retention rate
37.
67.
107.
37.
67.
107.
37.
67,
107.
37.
6%
107.
Daily retention
0.006 us
0.012 pig
0.02 pig
0.06 ^g
0.12 pig
0.20 us
0.3 pig
0. 6 pig
1.0 pig
0.6 pig
1.2 pig
2.0 pg
Percent of
significant level
0.09
0.18
0.30
0.9
1.8
3.0
4.5
9.1
15.2
9.1
18.2
30.3
Based on 2 liters/day,,
^Significant Level is 6.6 fig/day (from CITE)
-------
Food
Food provides man with the highest cadmium intake of all the direct
means of exposure. Although most foodstuffs contain less than 0.05 ppm
cadmium wet weight, the high amount of daily consumption (ranging from
about 3.0 kg in the U.S. to around 1.0 kg in Japan) means that about 50 pig
of cadmium per day enters the body via food. Dietary intake has been
investigated in numerous studies with values ranging from around 40 yg/day
g"
to as high as 200 yg/day. Schroeder's estimate of 200 yg/day was re-
evaluated down to about 100 yg/day upon review of the analytical proce-
dure. Another study, by Murthy et al., also found a mean of about 92 yg/
day, with some values as high as 158 yg/day, for institutional diets.
Friberg tends to discount both Schroeder's and Murthy's results because
of the atomic absorption technique which tends to give erroneously high
results due to interference from sodium chloride. However, the Oak Ridge
study feels that Murthy et al. compensated adequately for any possible
interference and even got results for cadmium in food within the range
found by others. If Schroeder's and Murthy's levels were to be actual
average concentrations found in the U.S., then renal tubular failure
would be expected to be much more prevalent than currently found. It is
possible that these intake levels are accurate but that renal tubular
failure has not been noticed yet because of the 50-year lagtime between
the constant level of intake and tubular failure inherent in Friberg1s
model for intake levels. However, until such time as more careful study
indicates these high levels in the American diet, the value of 50 yg/day
will be considered average with a maximum of about 70 ug/day. This figure
can be compared with a joint Food and Agricultural Organization (FAO)/
World Health Organization (WHO) expert committee on food additives pro-
12
posal for a tolerable weekly intake of 400-500 ug.
The range of intake levels, absorption rates, and respective retentions
are given in Table 26. While intake is a high percentage of the sig-
nificant level even under the average conditions, serious concern must
143
-------
Table 26. CADMIUM INTAKE AND RETENTION FROM FOOD
Exposure
Minimum
Average
(CITE I & II)
Maximum Recommended
(FAO/WHO)
U.S. Levels
(Murthy <5t al.
(Schroeder et al.)
•3
Daily 'intake
30 pig
50 pig
70 MS
100 ptg
Retention rate
3%
67.
107.
37.
67,
107.
37.
67.
107.
37.
67.
Daily retention
0.9 Mg
1.8 us
3.0 pig
1.5 pig
3.0 pig
5.0 pig
2.1 Jig
4.2 pig
7.0 pig
3.0 Mg
6.0 ptg
Percent of .
significant level
13.6
27.3
45.5
22.7
45.5
75.3
31.8
63.6
107.1
45.5
90.9
151.5
Based on studies listed„
^Significant level is 6.6 g/day (from CITE)
-------
be given to the levels presented by Schroeder and Murthy since over 90
percent of the significant level would be reached. Again, differences
in consumption due to personal preferences, diets, and the source of
food could substantially alter these levels.
.Cigarette Smoking
Additional intake of cadmium has been shown to occur as a result of
cigarette smoking. Values of cadmium in the mainstream of the smoke
range from 0.10 to 0.12 jjg per cigarette, which is about six percent of
13
the total cadmium in the cigarette; however, this can be increased by
increasing the total number of puffs per cigarette. Since only 15-20
percent of the cadmium in the mainstream is found in the filter, the
particles of cadmium are ass.umed small and would be almost totally de-
posited in the lungs. If one pack of cigarettes is smoked every day
(1 pack-year), the cadmium intake would be 2-2.4 ug/day with an estimated
retention of 1.4 ug/day (a retention rate of 64 percent). This latter
study also found further evidence of the importance of cigarette smoking
to cadmium body burden, which increases about 0.5 mg/pack-year. The
study showed a significant positive correlation between cigarette smok-
•
ing and the accumulation of cadmium in the kidney, liver, and lung.
*
Various levels of cigarette smoking, from one-half to three packs per
day, and the corresponding retention are given in Table 27. Obviously
the heavier smoker has a higher intake of cadmium, as much as 64 percent
of the significant level for 3 packs per day, but even a light smoker
(one-half pack per day) would be adding 10 percent of the significant
,./
level.
Total Exposure
The above discussions of the various media exposures provides some idea
of the importance of each exposure route, when considered separately, as
145
-------
Table 27. CADMIUM INTAKE AND RETENTION FROM CIGARETTES
Packs per day
1/2
1
2
#
3
•a
Daily intake
1.1 [ig
•2.2 ng
4.4 ng
6 . 6 |ig
Retention rate
64%
64%
64%
64%
Daily retention
0.70 ng
1.41 ng
2.82 jig
4.22 ng
Tcrcent of
significant:
levelb
10.7
21.3
*
42.6
/
64
Based on 0.11 ^g/cigarette
Significant level is 6.6 pig/day (from CITE)
-------
compared to what has been determined to be a "significant exposure level"
by Friberg et al. From the above tables, conclusions concerning the re-
lative importance of media can be drawn.
Ambient air and water provide about the same low level of cadmium to the
total daily intake. In order for either media to directly contribute sig-
nificantly to the total cadmium intake, it is necessary for maximum le-
vels of cadmium in the media to be reached. However, high levels of
cadmium in water are more likely to occur than equivalent exposure levels
in air.
Food is obviously the greatest source, of cadmium, providing almost 50
percent of the significant level even under average conditions. If the
retention rate actually reaches 10 percent, as has been suggested by
Friberg et al., food can provide more than 75 percent of what is considered
necessary to produce renal tubular failure. Note should be made of the
levels reported for the U.S.. which_indicate an extreme likelihood of
renal tubular failure^ given^fifty years of exposure.
•
Cigarettes are probably the second greatest source of cadmium to the
smoking population. Heavy smoking (3 packs or more per day) will provide
60 percent of the significant level, and even light smoking (half pack
per day) will supply 10 percent of the significant level.
The four tables given previously are summarized and combined into Table 28
to give ranges for the total daily retention of cadmium and percentages
of the significant level again. This table uses the various combinations
of "minimum", "normal", and "maximum" exposure levels of 0.001, 0.030,
3
and 0.30 ug/m for air; 0.1, 1.0 and 10 ppb for. water; and 30, 50 and
70 ug/day for food. Average, retention rates of 25 percent and 6 percent
for air and food/waterj respectively, were assumed. Cigarette smoking
was then added in.
147
-------
Table 28. TOTAL DAILY CADMIUM RETENTION VIA MAJOR ROUTES FOR
REPRESENTATIVE RETENTION LEVELS
Media levels
Air
Minimum
Normal
!-':2:-:ir.un
Miniir.ua
Normal
Maximum
Minimum
Normal
Maximum
Minimum
i-ormal
Maximum
Minimum
Normal
Maximum
Minimum
Normal
Maximum
Minimum
Normal
Maximum
Minimum
Normal
Mi xi mum
Minimum
Normal
Miximum
Water
Hinimum
Normal
Ilaxitnuin
Kinimuta
Normal
Maximum
Minimum
Normal
•
Maxitnua
Food
Minimum
Normal
Maximum
Totals with snokinj»,
PR
0
1.82
1.96
3/31
1.93
2.07
3.42
3.01
3.15
4.50
3.02
3.16
4.51
3.03
3.27
4.62
4.21
4.35
5.70
4.22
4.36
5.71
4.33
4.47
5.82
5.41
5.55
6.9
1 pack
3.23
3.37
4.72
3.34
3.48
4.83
4.42
4.56
5.91
4.43
4.57
5.92
4.54
4.68
6.03
5.62
5.76
7.11
5.63
5.77
7.12
5.14
5.88
7.23
6.82
6.96
8.31
3 packs
6.05
6.18
7.53
6.15
6.29
7.64
7.23
7.37
8.72
7.24
7.38
8.73
7.35
7.49
8.84
8.43
S.57
6.92
8.44
8.58
9.93
8.55
8.69.
10.04
9.63
9.77
11.12
Percentage of
significant level
0
28
30
50
29
31
52
46
48
68
46 -
48
68
47
50
70
64
66
86
64
66
87
66
68
88
82
84
105
1 pack
49
51
72
51
53
r
73
67
69
90
67
69
90
69
71
91"
85
87
108
85
87
108
87
89
110
103
105
126
3 packs
"92
94
114
93
95
116
110
112
132
110
112
132
HI
113
134
128
130
150
128
130
150
130
132
152
146
148
168
'Significant level is 6.6 ug/day (from CITE).
148
-------
From this table it is evident that normal levels of cadmium in the diet
provide almost 50 percent of the significant level without cigarette
smoking. This provides little safety factor considering individual and
regional dietary variations. If the maximum level from food is con-
sidered more relevant to the U.S. intake, then, with the other exposures
.normal and no cigarette smoking, 67 percent of the significant level is
reached. If one pack of cigarettes is smoked every day, then this level
is brought up to within 11 percent of the significant level. In addi-
tion, smoking three packs of cigarettes per day ensures more than
100 percent of the significant level under all but the very minimum
diet levels.
Throughout this discussion the "significant level" has been, after
Friberg, based on that amount needed to achieve 200 ppm cadmium in the
renal cortex. As was discussed in Section IV, another study has indi-
cated that kidney dysfunction may begin to occur at renal cortex levels
of 90 ppm cadmium. If thi's -is actually the case, then any combination
of intakes which is above the 45" percent level in Table 28 is capable
of producing kidney dysfunction. Such a level would indicate that
normal dietary intake is capable of producing adverse effects.
\
AIR EMISSIONS' CONTRIBUTION TO TOTAL EXPOSURE
The direct impact of cadmium from ambient air on the total normal daily
retention is quite small (see Table 29), being less than 5 percent of
the total exposure assuming no smoking. This contribution decreases to
about 2 percent if 3 packs per day of cigarettes are smoked, though the
actual intake and retention via the lungs increases to over 50 percent.
However, this does not automatically eliminate consideration of emissions
of cadmium to the air as a significant factor in the total daily reten-
tion by man.
149
-------
Table 29 . MEDIA CONTRIBUTIONS TO NORMAL RETENTION
Ambient Air
Water
Food
•
Exposure level
0.03 Mg/m3
1 ppb
50 ng/day
Total
Daily retention
0.15 jjig
0.09 ng
3.0 ^g
3.24 ^g
•
Percent of total
No smoking
4.67.
2.87.
92.67.
1007.
1 pack/day
3.27.
1.97.
64.57.
69.97.
3 packs/day
2.07.
1.27.
40.27.
43.47.
-------
As discussed in Section V, emissions to the air eventually end up in the
soil or water due to washout, rainout, and simple deposition, thereby
increasing the cadmium levels in these media. Therefore, the body bur-
den of cadmium contributed by exposure to these media is due in part to
the emissions into the air. Average contribution of air emissions to the
different media were felt to be less than 10 percent for water and
around 64 percent for food and tobacco. Applying these average contri-
butions to the level of retention that results from exposure to each
of the media x^ould provide an estimate of the total contribution of air
emissions to the total body burden.
Again it must be emphasized that numerous assumptions were used in these
calculations and that meager data was generally the rule. Therefore,
significant differences may result from different assumptions or better
data. In order to obtain-an estimation of the impact of air emissions
on the total body burden, a range of 10 to 80 percent has been used
when considering the percentage contribution of air to food and tobacco
levels 'of cadmium in Tables 30 through 32. These tables provide the
estimated contribution of air emissions to the normal daily intake and
retention of cadmium under no smoking, 1 pack per day and 3 packs per
day'exposures. The 10 percent air contribution to food level assumes
that the s~oil provides most' of the cadtniuni to the crops and that most
of the cadmium in the soil occurs naturally. The 80 percent level would
represent a high level of absorption from deposition and a majority of
the cadmium in the soil being from long term emissions of cadmium to the
air. An 8 percent contribution of air to water is assumed in the
calculations.
These tables indicate that, when air's contribution to all parts of intake
is considered, the contribution of air to. the normal daily retention is
probably between 12 and 80 percent. Therefore, even under what is likely
to be conservatively low conditions, the percentage contribution to daily
intake is significant. Assumed average contributions from air imply
151
-------
Table 30. TOTAL CONTRIBUTION OF AIR EMISSIONS TO NORMAL DAILY RETENTION OF
CADMIUM, NONSMOKING
to
Air
Water
o
Food
Total
Normal
retention
0.15 tig
0.12 pg
3.0 (ig
3.27 pg
Percentage
from air
100
8
10
64
80
NA
Retention
due to air
0.15 ,ig
0.011 pg .
0.30 pg
1.92 pg
2.40 pg
0.46
2.08
2.56
Percentage
of total
retention
4.6
0.3
9.2
55.0
73.4
14.1
63.6
78.3
-------
Table 31. TOTAL CONTRIBUTION OF AIR EMISSIONS TO NORMAL DAILY RETENTION OF CADMIUM,
1 pack/day
Air
Water
Food
and
Tobacco
Total
Normal
retention
0.15 y.g
0.12 yg
4.41 ug
•
4.68 yg
Percentage
from air
100
i
1
t
8
i 10
: 64
I SO
NA
Retention
due to air
0.15 ng
•
0.01 yg
0.44 yg
2.82 yg
3.53 yg
0.60 yg
2.98 yg
3.69 yg
Percentage
of total
retention
3.2
.
0.2
9.4
60.3
75.4
12.8
63.7
78.8
-------
Table 32. TOTAL CONTRIBUTION OF AIR EMISSIONS TO NORMAL DAILY RETENTION OF CADMIUM,
3 packs/day
Air
Water
Food
and
Tobacco
Total
Normal
retention
0.15 us
0,12 ug
7.23
7.50
•
Percentage
from air
100
8
10
64
80
NA
Retention
due to air
0.15 ng
0.01 pg
0.72 yg
4.63 yg
5.78
0.88
4.79
5.94
Percentage
of cotal
retention
2.0
0.1
9.6
61.7
77.1
11.7
63.9
79.2
-------
64 percent of daily exposure from air. If it is found that the contri-
bution from air is actually this great, then serious consideration must
be given to controlling emissions of cadmium.
155
-------
REFERENCES
1. Drinking Watex- Quality of Selected Interstate Carrier Water Supplies.
1962-1963. Public Health Service Pub. No. 1049-A. U.S. Dept. of
' Health, Education, and Welfare. Public Health Service. Div. of
Environmental Engineering and Food Protection. Washington, D.C.
2. McCabe, L.J., J.M. Symons, R.D. Lee, and G.G. Robeck. Survey of
Community Water Supply Systems. Journal Am. Water Works Assoc.
Vol. 62(11), 670-687. November 1970,
3. Craun, G.F. and L.J. McCabe. Overview of Problems Associated with
Inorganic Contaminants in Drinking Water, Proceedings National
Symposium on the State of America's Drinking Water. Chapel Hill,
North Carolina (In Press).
4. Preliminary Assessment of Suspected Carcinogens in Drinking Water
(Appendices). Interim Report to Congress. U.S. EPA, June 1975°
p. 132.
5. ORNL. Section VI. B.2.
6. Schroeder, H.A., A.P. Nason, I.H. Tipton, and J.J. Balassa,
Essential Trace Metal in Man: Zinc - Relation to Environmental
Cadmium. J Chron Dis. 20, 179-210. 1967. ORNL.
7. Ibid.
8. Rautu and Sporn. Loc. cit.
9. Murthy, G.K., U. Rhea, and J.T. Peeler. Levels of Antimony, Cadmium,
Chromium, Cobalt, Manganese, and Zinc in Institutional Total Diets.
Environ Sci Technol. 5(5), 436-42. May 1971, ORNL.
10. CITE. Section 3.2,
11. ORNL. Section VI. B.3.a.
12. World Health Organization. Evaluation of Certain Food Additives and
the Contaminants Mercury, Lead and Cadmium. Wrd-Health Org Techn
Rep Ser., 505. 1972. CITE II.
13. Menden, E.E., V.J. Elia, L.W. Michael, and H.G. Petering. Distribu-
tion of Cadmium and Nickel of Tobacco During Cigarette Smoking.
Environ Sci Technol. 6, 830. 1972. CITE II.
14. Linnman and Lind. (Unpublished data). CITE.
156
-------
15. Nandi, M., D. Slone, H. Jick, S. Shapiro, and G. P. Lewis. Cadmium
Content of Cigarettes. The Lancet 2. 1329-30 (Dec. 20, 1969). ORNL.
16. Lev/is. G. P., W. J. Jusko, L. L. Coughlin, and S. Hartz. Contribution
of Cigarette Smoking to Cadmium Accumulation in Man. The Lancet.
Vol. 1 for 1972. No. 7745. Feb. 5. Log 291 (1972). ORNL.
157
------- |