EPA-600/2-76-155
June 1976
Environmental Protecticis f'
ESTIMATION OF PERMISSIBLE CONCENTRATIONS OF
POLLUTANTS FOR CONTINUOUS EXPOSURE
Industrial Environmental Research Laboratory
Office of Research and Development
U.S» Environmental Protection Agency
Research Triangle Park, North Carolina 27711
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into five series. These five broad
categories were established to facilitate further development and application of
environmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The five series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ENVIRONMENTAL PROTECTION
TECHNOLOGY series. This series describes research performed to develop and
demonstrate instrumentation, equipment, and methodology to repair or prevent
environmental degradation from point and non-point sources of pollution. This
work provides the new or improved technology required for the control and
treatment of pollution sources to meet environmental quality standards.
EPA REVIEW NOTICE
This report has been reviewed by the U.S. Environmental
Protection Agency, and approved for publication. Approval
does not signify that the contents necessarily reflect the
views and policy of the Agency, nor does mention of trade
names or commercial products constitute endorsement or
recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/2-76-155
June 1976
ESTIMATION OF
PERMISSIBLE CONCENTRATIONS OF
POLLUTANTS FOR CONTINUOUS EXPOSURE
by
Robert Handy and Anton Schindler
Research Triangle Institute
P. O. Box 12194 v
Research Triangle Park, North Carolina 27709
Contract No. 68-02-1325, Tasks 34 and 46
Program Element No. EHE624
EPA Task Officer: Max Samfield
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
PREFACE
This report was prepared for the Industrial Environmental Research
Laboratory - RTF, Environmental Protection Agency, to present results of
the work carried out by RTI under Contract No. 68-02-1325 (Task 34). This
work was performed in the Chemistry and Life Sciences Division of the
Research Triangle Institute. The authors thank Dr. Max Samfield, EPA Task
Officer, for many helpful discussions during the preparation of this docu-
ment.
Approved for:
RESEARCH TRIANGLE INSTITUTE
Monroe TS. Wall, Ph.D.
Vice President
Chemistry and Life Sciences Division
iii
-------
CONTENTS
Page
List of Figures vii
List of Tables ix
Sections
I Introduction 1
II Summary 3
A. Ambient Air Concentrations 3
B. Drinking Water Concentrations 4
C. Carcinogens in Ambient Air 5
D. Equations for Estimating Permissible Pollutant
Concentrations 6
III Permissible Ambient Air Concentrations 11
A. Table of Symbols and Definitions 11
B. Caveats 13
C. Kinetics of Pollutant Accumulation in the Body .... 13
D. Application of Kinetic Considerations 17
E. Correlation Between TLV Standards and LDcn
Values . .5(? 21
F. Estimation of Maximum Permissible Air Concentrations
from Known LDrQ Values 36
G. Estimation of Air Quality Standards for Multicom-
ponent Systems 42
H. Estimation of Maximum Ground Level Concentrations
from Stack Emission of Stationary Sources 43
IV Permissible Drinking Water Concentrations 46
A. Table of Symbols and Definitions 46
B. Caveats 47
C. Derivation of EPA Proposed Drinking Water
Standards 47
D. Stokinger-Woodward Treatment - Method I 49
E. Estimation of Permissible Drinking Water Stan-
dards - Method II 52
F. Estimation of Permissible Drinking Water Stan-
dards - Method III 60
G. Estimation of Permissible Drinking Water Stan-
dards for Multicomponent Systems 63
H. Comparison of Methods I, II, and III 66
-------
CONTENTS (Continued)
Page
••^•^*
Sections
V Carcinogens in Ambient Air 7^
A. Table of Symbols and Definitions 71
B. Caveats 72
C. Carcinogen Content in Cigarette Smoke and Intake by
Smokers 72
D. Lung Cancer Mortality as a Function of Individual
Smoking Habits and Age • 74
E. Relationship of Lung Cancer Mortality and
Age/Duration of Exposure 76
F. Relationship of Lung Cancer Mortality and Carcinogen
Concentration in Ambient Air 80
G. Lung Cancer Mortality Estimates for the General
Population 85
VI References 94
VII Appendices to Section III 96
A. Effect of Age on the Build-Up of Body Concen-
tration 96
B. Average Accumulated Body Concentration During
Intermittent Exposure 109
C. Nomographic Correlation of Stack Emission Rates
and Maximum Ground Level Concentrations 119
VI
-------
FIGURES
No. Page
1 Accumulated Body Burden in mg/kg for Different Starting
Times of Continuous Exposure to Polluted Air Containing
1 vg/m3 Pollutant 22
2 Body Burden in mg/kg Accumulated During Continuous'
Exposure to Air Containing Different Concentrations
of Pollutant 23
3 Plot of Ig (TLV) versus Ig (LD,n) for 30 Agricultural
Chemicals ?° 27
4 Plot of Ig (TLV) versus Ig (LD,-n) for 50 Agricultural
Chemicals ?u 28
5 Plot of Ig (TLV) versus Ig (LD5Q) for 191 Non-Agri-
cultural Compounds 34
6 Plot of Ig (TLV) versus Ig (LD50) for 191 Non-Agri-
cultural and 50 Agricultural Compounds 35
7 Lung Cancer Mortality Versus Age and Cigarette Use 75.
8 Nonsmoker Lung Cancer Mortality Versus Age 79
9 Lung Cancer Mortality Versus Duration of Exposure
and Cigarette Use • •. 81
10 Lung Cancer Mortality Versus Carcinogen Intake Rate 82
11 Lung Cancer Mortality Versus Carcinogen Air Concen-
tration 84
12 Lung Cancer Mortality Versus Carcinogen Air Concen-
tration - Dorn Study 88
13 Lung Cancer Mortality versus Carcinogen Air Concen-
tration - General Population. 90
14 Age Dependence of Breathing Frequency 102
15 Ratio of Body Concentration to Effective Pollutant
Concentration in Air for Pollutants with Different
Half-Life Times in the Body. Calculated for Age
Dependent Breathing Frequency 103
16 Ratio of Body Concentration to Effective Pollutant
Concentration in Air for Pollutants with Different
Half-life Times in the Body Calculated for the Constant
Breathing Frequency of an Adult 104
17 Effect of Age Dependent Breathing Frequency on the
Body Concentration of Pollutants with Different Half-
Life Times in the Body" 105
18 Comparison of the Body Concentrations of Children and
Adults Exposed to the Same Level of a Pollutant with a
Half-Life Time in the Body of 36 Days 107
vii
-------
FIGURES (Continued)
19 Body Concentration Versus Time During Intermittent
Exposure to p-Nitrophenol With a Known Biological
Half-life Time in Man of 0.99 Mrs. Weekends not ,,3
Considered .........................
20 Body Concentration Versus Time During Intermittent
Exposure to Dinitro-o-cresol With a Known Biological
Half-Life Time in Man of 139 Hrs. Weekends not ..
Considered ........................ • m
21 Effect of Weekends on the Body Concentration During
Intermittent Exposure to p-Nitrophenol With a Bio-
logical Half-Life Time in Man of 0.99 Hrs. .......... MD
22 Effect of Weekends on the Body Concentration During
Intermittent Exposure to a Pollutant With an Assumed
Biological Half-Life Time of 72 Hrs. ............ M/
23 Effect of Weekends on the Body Concentration During
Intermittent Exposure to Dinitro-o-cresol With a
Biological Half-Life Time in Man of 139 Hrs. ........ 11S
24 Nomographic Example for the Determination of Plume
Rise Above Stack by Holland's Equation ........... 123
25 Distance of Maximum Concentration and Maximum
(xu/Q)-Value As a Function of Weather Condition
and Effective Height of Emission ..............
26 Horizontal Dispersion Coefficient As a Function
of Downwind Distance From the Source ............ 127
27 Vertical Dispersion Coefficient As a Function of
Downwind Distance From the Source .............. 128
28 Nomographic Example for the Calculation of Stack
Emission Rate for Which the Maximum Ground Level
Concentration Does not Exceed a Permissible Value ...... 130
29 Nomographic Example for the Calculation of
Maximum Ground Level Concentration Resulting
from Given Stack Emission Rate ............... 132
30 Nomograph for the Estimation of Plume Rise
Above Stack by Holland's Equation .............. 135
31 Nomograph for the Correlation of Stack Emission Rates
and Maximum Ground Level Concentrations ........... 136
viii
-------
TABLES
No. Page
1 Comparison of Calculated Maximum Permissible
Pollutant Concentrations with EPA Standards 9
2 Respiration Data as Dependent on Age and Weight 15
3 Biological Half-Life Times of Different Compounds 19
4 Maximum Body Concentration for Continuous Exposure
Calculated from TLV and Half-Life Times Determined
on Man 19
5 Maximum Body Concentration for 8 Working Hours
Exposure Calculated From TLV and Half-Life Times
Determined on Man 20
6 ' TLV Standards and LD5Q Values for Different Agri-
cultural Compounds • • 25
•
7 List of TLV Standards and LD5Q Values for Different
Chemical Compounds ...... 29
8 Permissible Air Concentrations for Continuous
Exposure Estimated by Different Methods 40
9 Maximum Stationary Body Concentration Accumulated
During Working Conditions (40 Hour Workweek) for
Selected Inorganic Compounds 41
10 Analysis of a Multicomponent System (Selected
Values for Los Angeles Air) 44
11 Derivation of EPA Proposed Drinking Water Standards
for Pesticides 48
12 Derivation of EPA Proposed Drinking Water Standards
for Inorganic Chemicals 50
13 Calculated Water Standards - Inorganic, Method I 51
14 Estimated Water Standards - Inorganic, Method II 55
15 Comparison of Proposed and Estimated Standards -
Inorganic, Method III 56
16 "Estimated Standards - Pesticides, Method II 57
17 Toxicity Data - Inorganic 59
-------
TABLES (Continued)
No.. £§fl§.
18 Estimated Water Standards Using Assumed Half- M
Life - Inorganics, Method III B ..............
.20
19 Estimated Water Standards Using Assumed fic
Half-Life - Pesticides, Method III C ...........
Analysis of a Multicomponent System - Inorganics
(Method III B) 67
21 Analysis of a Multicomponent System - Pesticides
(Method III C) 68
22 Comparison of Methods I, II, III.A, III B 69
23 , Carcinogen Hydrocarbons Isolated from Cigarette
Smoke
73
24 Carcinogen Intake for Smokers, as a Function of
Number of Cigarettes Smoked per Day '3
25 Lung Cancer Mortality as a Function of Smoking
Pattern and Age ''
26 Estimated Lung Cancer Mortality of 45-54 Year Old
Cigarette Smokers '8
27 Carcinogen Intake and Equivalent Air Concentration
for Nonsmokers 78
28 Carcinogen Concentration in Ambient Air Equivalence
to Carcinogen Intake • 78
29 Estimated Nonsmoker Lung Cancer Death Rate as a
Function of Carcinogen Concentration in Ambient
Air 86
30 Lung Cancer Mortality versus Carcinogen Intake for
Dorn Study Group Age 45-84 87
31 Estimated Nonsmoker Lung Cancer Mortality as a
Function of Carcinogen Air Concentration 92
32 Carcinogen Concentrations Found in Ambient Air 93
33 Respiration Data as Dependent on Age and Weight 98
-------
•SECTION I
INTRODUCTION
The EPA has set several maximum concentration exposure levels
for environmental pollutants in both air and water. However, there
are many known toxic materials for which no safe concentration levels
have been established. The main objective of this report was to
derive pollutant hazard criteria for this group of compounds in order
that safe, permissible concentrations in the environment might be
estimated. The proposed permissible pollutant levels derived in this
report were not intended to challenge present EPA standards. The
mention of EPA standards in this report is for comparison purposes only.
In their present form these estimated values serve only as a guide in
assessing potential pollution hazards in lieu of official standards.
This report discusses the basis of estimating permissible pollu-
tant levels in ambient air (Section III) and drinking water (Section IV).
In addition a correlation between carcinogen exposure and lung cancer
deaths was developed and presented as a guide for estimating the effect of
carcinogen emissions in ambient air on the risk of lung cancer mortality
(Section V).
The hazard criterion expressions were developed in the form of a
ratio (actual pollutant concentration/permissible pollutant concentration).
When this quotient was equal to or greater than one, a pollution hazard
was judged to exist.
Several approaches were used to estimate safe concentration values.
Expressions were derived using TLV and/or LDj-g animal data, employing
known or assumed biological half-lives.
Estimates for air pollution were based on episodes of continuous
exposure and the use of the tidal volume to body weight relationship
of infants. This model system took into account the higher pulmonary
intake per unit weight characteristic of younger children.
The effect of source emission rates on ground level concentrations
were assessed. The results were presented in the form of a nomograph
which provides a simple means of correlating stack emission rates and
ground level concentrations as a function of weather and source character-
istics.
1
-------
Permissible pollutant concentrations in drinking water were
developed along the same lines. The first treatment employed derived
TLV levels calculated from reported LD5Q data by a method described
in Section III. The second approach was based on fundamental biological
concepts. Pollutant hazard criteria were derived using animal LD5Q,
human half-life data and an assumed safe maximum pollutant body concen-
tration.
The final expressions were also used to treat multipollutant -systems
in both air and drinking water.
A means for estimating lung cancer mortality risk as a function of
carcinogen concentrations in ambient air was presented in Section V. Since
lung cancer mortality data for smokers was the best human dose response
data available, this information was used to correlate carcinogen exposure
and lung cancer death rate. This was accomplished by first determining the
carcinogen intake by individuals smoking different number of cigarettes-over
varying time periods. Lung cancer mortality data for nonsmokers and
different categories of cigarette smokers at different ages were used to
correlate lung cancer death rate with carcinogen intake. Finally, a carcino-
gen content in ambient air equivalent to cigarette carcinogen intake was
calculated and related to years of exposure.
This treatment allowed one to determine carcinogen air levels to which
nonsmokers are exposed and to estimate the lung cancer mortality for
nonsmokers exposed to any given carcinogen ambient air concentration.
Since no threshold limit has been established for carcinogenic materials,
the results must be expressed as risk versus carcinogen concentration.
The results of this treatment will be used by IERL for developing
guidelines for estimating the effect of carcinogen emissions in ambient
air on lung cancer mortality risk.
-------
SECTION II
SUMMARY
A. Ambient Air Concentrations
Assessment of health hazards associated with air pollutants from
process industries was complicated by the lack of established TLV
standards for many pollutants. In cases where TLV standards had not
been set, two different approaches were followed: (1) A method was
developed for obtaining safe estimates of the TLV. (2) A different
method based on pollutant accumulation in the body was established.
The first approach was based on the correlation between TLV
standards and LD5Q values of different compounds. This approach
represented an extension of work started by Monsanto Research Corp. on
30 selected agricultural chemicals. Increasing the sample size by
20 additional agricultural chemicals and 191 compounds for which both the
TLV standards and the LD,-g values could be retrieved from literature, a
regression analysis yielded an equation of the form:
(TLV) = 0.029 (LD,n)0'983
50;
The lower 95% confidence limit of the above given regression line was
then given to a very good approximation by:
= 4'5 x 10~ (LD50}
Based on this relationship, an equation was derived which permitted
estimation of safe ambient air concentrations of pollutants for which
LDcn data was available. Since a one year old infant has approximately
bu
twice the respiratory frequency as an adult, a safety factor of two was
introduced into the decision criterion. This expression differed from
one proposed by the Chemical Process Section of EPA on the following
points: (1) The criteria was based on LD5Q values rather than TLV,
making it applicable to a greater number of pollutants. (2) The factor
-------
for converting intermittent exposures to continuous exposure
has been adjusted to a weekly basis. The TLV is multiplied by
40/168, the fraction of time one is exposed to a working place
environment. (3) The safety factor of (1/100) has been omitted
because of the use of the low estimate of TLV from LD5Q. (4) The
relative susceptibility of young children has been taken into
account.
The second approach was based on biological considerations. Data
on biological half-lifes of different compounds in combination with
their presently used TLV standards indicated that a stationary maximum
body concentration corresponding to about 0.05% of the LDgQ value of
the compound can be considered safe. Assuming.a half-life of 30 days,
a maximum permissible an'r concentration for continuous exposure can be
calculated for which the accumulation of pollutant in the body will
never exceed 0.05% of its LD5Q value. Results of both approaches were
very similar, the criteria values differing by only approximately 10%.
When two or more hazardous substances are present, their combined
effect, rather than that of either individually, should be considered
as proposed by the American Conference of Governmental Industrial
2
Hygienists.
B. Drinking Water Concentrations
Stokinger and Woodward derived an expression for estimating safe
pollutant drinking water limits. They assumed that the amount of pollu-
tant contained in 10 m of air at the TLV concentration may be safely
ingested daily in drinking water. Knowledge of pollutant respiratory
and gastrointestinal absorption efficiencies was required and the daily
consumption of 2 liters of drinking water was assumed. The limited
number of compounds for which TLVs have been assigned and the still
fewer substances with known abosrption efficiencies restricts the
general applicability of this method.
Applying the (TLV)low concept developed in Section III to this
problem yielded a pollutant hazard criterion expression of much greater
utility. Estimated drinking water standards were then calculated by
-------
substituting (TLV)low for TLV in the Stokinger-Woodward equation3 and
assuming complete respiratory and gastrointestinal absorption.
Another approach based on kinetic considerations is summarized
below. The assumption made in this treatment were as follows:
a) Human disposition of a foreign substance was approximated
by a single compartment system with first order kinetics.
b) A maximum body pollutant concentration equal to 0.05% of
its LDgg can be considered safe.
The estimated safe drinking water concentration of a pollutant was
expressed as a function of its LD5Q and biological half-life. When the
half-life of the compound was not known, a value for this parameter was
assumed.
a) A thirty (30) day half-life is asusmed for all compounds
excepting those known to be retained in the body for prolonged
periods of time (e.g., chlorinated pesticides storage in
adipose tissue).
b) A one year (365 day) half-life is assumed for those exceptions
noted above.
C. Carcinogens in Ambient Air
The content of eight carcinogenic hydrocarbons in cigarette smoke
has been determined (10.8 ug/100 cigarettes). This information was
used to calculate the rate of carcinogen intake for individuals smoking
different numbers of cigarettes/day. Lung cancer mortality figures for
nonsmokers and smokers (use of 1-9, 10-19, 20-39 and more than 39
cigarettes/day) at different ages (35-44, 45-54, 55-64, 65-74 and 75-84
5
years old) were available and used to correlate carcinogen intake with
lung cancer death rates. Plotting these data on a log-log scale indicated
a linear relationship between these parameters and permitted extrapolation
to low carcinogen intake values.
A linear relationship also existed between lung cancer mortality
and age (or years of exposure) when plotted on a log-log scale. This
relationship held for nonsmokers and individuals smoking different
numbers of cigarettes/day.
-------
Plotting lung cancer death rate against ambient air carcinogen con-
centration on a log-log scale gave a linear relationship which allowed
ready estimation of mortality risk as a function of carcinogen air level.
This treatment indicated that urban air concentration of 20 ng/m would
result in a lung cancer death risk of approximately 40/100,000, an
increase of 150 percent over the reported nonsmoker rate. This level
of contamination corresponds to smoking approximately 3 cigarettes/day.
D. Equations for Estimating Permissible Pollutant Concentrations
The expressions for the estimation of permissible pollutant con-
centrations are listed below. As shown, the particular form of the
equation will depend on the data available for the pollutant under
evaluation.
Ambient Air
Case I - TLV unknown, LD5g .known
This case will be encountered most often. The oral LDcn (mg/kg)
for rats should be used if known. It may be necessary to use the oral
LDcQ for mice. If the oral LDr,, is not known, one should use the LDg/,
obtained by intraperitoneal injections. Avoid the use of intramuscular
LDgQ data. The permissible ambie
given by the following equation:
data. The permissible ambient air concentration, x (mg/m ), is
xp = 4.77 x 10'5 (LD5Q)
Case II - TLV known, LDCO unknown
ou
This is a less common situation. A derived ID™ is computed from
the equation,
LD5Q = 34.5 TLV,
and used directly in the expression given in Case I.
-------
Alternatively, the following equivalent equation may be used:
x = 1.65 x 10~3 (TLV)
Case III - TLV and LD5Q known
Use the TLV and proceed as in Case II.
Case IV - TLV unknown, LD5Q and biological half-life for man are
known.
This case will not be encountered often, since the half-lives of
o
most compounds in man are not known. However, the equations do provide
for taking this into consideration if the information is available:
x = (14.2 x l(f4) (LD5Q)
where T is the biological half-life in days.
Drinking Water
Case I - TLV unknown, LDcn known
For inorganic compounds, the ionic LD5Q for ion consideration is
computed from the LD5Q for the compound. The ionic LDgQ is defined as:
. . . n (Compound LDcr>) (% content of metal or anion)
Ionic LD5Q = 50^ m
a) For most organic and inorganic compounds, excluding those that
are known to be retained in the body for prolonged periods of
time (e.g., chlorinated pesticides which are stored in adipose
concentration, xe (mg/1), compute from the following eauation:
xe = 4.0 x 1CT4 (LD5Q)
b) For those compounds that are known to be retained by the body
for long periods of t\me, use the equation:
xe = 3.3 x 10"5 (LD5Q)
-------
Case II - TLV known, LD5Q unknown
A derived ID™ is computed from the equation,
LD5Q = 34.5 TLV,
and used directly in the appropriate expression given in Case I. The
equivalent equations for relating TLV directly to xe are shown below.
a) xe = 1.38 x 10'2 (TLV)
b) xe = 1.14 x 10"3 (TLV)
Case III - TLV and LD5Q known
Use the TLV and proceed as in Case II.
Case IV - TLV unknown, LD™ and biological half-life for man are known
When the LD5Q and biological half-life, T, is known, the following
equation should be used.
The ionic LD5Q is used for inorganic compounds.
Carcinogens in Ambient Air
Since threshold values for carcinogenic substances have not been
established, an attempt was made to relate cancer risk versus ambient air
concentrations. Data on carcinogen intake for smokers, and lung cancer
mortality figures for smokers and non-smokers, were used in relating lung
cancer mortality and equivalent ambient air levels of carcinogens. The
results seem to indicate that the "lowest concentration of concern" (equi-
valent to 1 ng/m in ambient air) represents a practical limit of risk,
and therefore a reasonable "permissible" concentration for establishing
control technology R&D priorities.
Comparison of Air and Water Pollutant Standards With Calculated Safe
Continuous Exposure Levels
Table 1 shows the comparison for 23 substances for which standards have
either been set or proposed. The permissible exposure levels were calcu-
lated as outlined above.
8
-------
Table 1. COMPARISON OF CALCULATED MAXIMUM PERMISSIBLE
POLLUTANT CONCENTRATIONS WITH EPA STANDARDS
Continuous 24 hr. Exposure
A. Air (yg/ni )
Substance
^••••I^BB
N0
so2
*Be
*Hg
EPA Standard
100
80
0.01
1.0
Calculated
BVHBIIIBBMHM
15
22
0.003
0.082
Ratio
Stnd./Calc.
-' -- -H^ ——
6.66
3.64
3.33
12.2
*EPA emission rate standards based on indicated 24 hr. exposure
B. Water (mg/1) .
Substance EPA
Chlordane
Endrin
Heptachlor
Lindane
Methoxychlor
Toxaphene
2, 4-D
2, 4, 5-TP (Silvex)
Standard
0.003
0.0002
0.001
0.004
0.10
0.005
0.10
0.01
Calculated
0.009
0.0001
0.0013
0.003
0.165
0.002
0.012
0.021
Ratio
Stnd./Calc.
0.33
2.00
0.77
1.33
0.61
2.50
8.4
0.47
Ag
As
Ba
Cd
CN
Cr
F
Hg*
N03 (as N)
Pb
Se
* Based on MeHg (T
0.05
0.05
1.0
0.01
0.20
0.05
1.4-2.4
0.002
10.0
0.05
0.01
70 days)
0.013
0.002
0.040
0.022
0.001
0.011
0.032
0.046
0.092
0.068
0.001
3.84
25
25
0.46
200 (!!)
4.55
43.6-75
0.435
109 (I!)
0.74
10
-------
The following observations were noted:
(1) For 16 of the 23 substances, calculated levels were on the
"safe" side.
(2) Only seven of the 23 calculated permissible levels differed
by more than an order of magnitude.
(3) All of the seven that differed by more than an order of
magnitude were among the metals and anions (Hg, As, Ba, CN,
F, NCL and Se). The worst agreement was noted with cyanide,
fluoride and nitrate.
The use of a 30 day half-life in the case of cyanide and fluoride
certainly overestimates the residence time of these pollutants in the
body. Use of more realistic half-lives and the Case IV equations given
above would better reflect potential pollutant hazards. Nitrate toxicity
mainly affects small children. It is significant that a drinking water
limit of 1 mg N03 - N/l has been suggested for infant feeding.6
10
-------
SECTION III
PERMISSIBLE AMBIENT AIR CONCENTRATIONS
Table of Symbols and Definitions
a ordinate intercept of the regression line Ig(TLV) on
ig(LD50)
b slope of regression line Ig(TLV) on Ig (LDcn)
3
B rate constant for pollutant uptake by respiration (m /kg day)
c instantaneous pollutant concentration in the body (mg/kg)
cm maximum stationary body concentration (mg/kg)
Co initial pollutant concentration in the body (mg/kg)
C70 final pollutant concentration in the body after 70 years
accumulation without excretion
f respiratory frequency (day )
F (t) breathing frequency as function of age
gpg guinea pig
imp implant
ims intramuscular
ipr intraperitoneal
itr intratracheal
k first order rate constant of excretion (hr~ )
LD5Q lethal dose 50% kill (mg/kg)
LDLo lowest published lethal dose (mg/kg)
mus mouse
or! oral
Q constant ratio of tidal volume to body weight (ml/kg)
rbt rabbit
R . rate of pollutant uptake (mg/kg.day)
t time (hours or days as stated)
TDL0 lowest published toxic dose (mg/kg)
TLV threshold limit value (mg/m )
(TLV), lower,95% confidence limit for TLV estimated from LDKn
low (mg/m3) 50
VT tidal respiratory volume (ml)
W body weight (kg)
O
x pollutant concentration in air (mg/m )
11
-------
Table of Symbols and Definitions (Continued)
x maximum ground level concentration (mg/m )
(x ). maximum ground level concentration of the i-th pollutant
m 1 (mg/m3)
x maximum permissible air concentration (mg/m )
(xn). maximum permissible air concentration of the i-th pollutant
P ' (mg/m3)
a absorption factor for respiratory uptake of pollutant
T biological half-life time of pollutant in the body (hours or
days as stated)
12
-------
B. Caveats
The treatment described in this section was based, in part, on the
following assumptions and limitations.
(1) The presented method of estimating permissible air concen-
trations is not applicable to pollutants with known carcino-
genic, teratogenic, or mutagenic effects. A treatment of
of carcinogens in ambient air is presented in Section V.
(2) The biological treatment was based on a one-compartment
. model with a single, first order excretion rate.
(3) It was assumed that all pollutants entering the respiratory
system were retained by the body.
(4) The treatment was based on the assumption that LDgQ values for
animals were applicable to man.
(5) No allowance was made for synergism of pollutants in multi-
component systems.
(6) The derived equations are directly applicable only when the
half-life of the pollutant is relatively short compared to the
assumed life of the individual.
C. Kinetics of Pollutant Accumulation in the Body
Estimation of maximum permissible air concentrations of pollutants,
3
x in mg/m , depends on the model used. In order to make the proper choice,
some general kinetic considerations seem to be appropriate.
The simplest mathematical model describing the accumulation of inhaled
pollutant in the body is derived from a one-compartment model. In this
system, all of the pollutant entering the body through the respiratory
system is evenly distributed over the whole body. Consequently, the
amount of absorbed pollutant can be related to the total body weight.
It is obvious that this model does not apply for particulates which remain
trapped in the lung and exhibit only localized effects.
The rate of pollutant uptake per unit of body mass is given by
Ruptake '
13
-------
where VT tidal respiratory volume (m ),
i _1
f respiratory frequency (day ),
a absorption factor determining the
fraction of the inhaled compound
entering the body; 1-a is equal to
the exhaled fraction of the pollu-
tant,
o
x air concentration of pollutant (mg/m ),
W body weight (kg.).
Three of the variables determining the rate of pollutant uptake change
considerably with age. Table 2 gives values of W, f, and VT for different
ages. Ratios of VT/W calculated for different ages are presented in column
5 of Table 2. These values reveal that for all practical purposes the .
ratio of Vj/W can be considered constant and independent of age.
VT/W = Q = (6.36 + 0.38) x 10"6 m3 .kg"1 (2)
Equ. (1) then takes the form
Ruptake = ^W (3)
where F(t) is a function which describes the age dependence of the average
breathing frequency.
The effect of changes in breathing frquency with age on the uptake of
pollutant is dealt with explicitly in the Appendix, Section 71I-A. From this
treatment a correction factor is derived which is applied to the decision
criteria for adults to provide the same level of protection for young children.
This application will be discussed after the mechanism of respiratory pollu-
tant uptake has been developed for adults.
The breathing frequency of adults over the age of 18 can be considered
constant as shown by respiratory data of Table 2. Assuming a = 1, equ. (3)
can be simplified by setting
B = QF(t) = const.
14
(4)
-------
Table 2. RESPIRATION DATA AS DEPENDENT ON AGE AND WEIGHT
Age Weight Breathing Tidal
Frequency Volume
(Years) (kg) (min"1) (ml)
(Ref. 7) (Ref. 8) (Ref. 9)
0 3.4 38 + 1010 13 + 510
1 10.4 31 67
2 12.7 26 87
3 14.5 25 98
4 16.8 24 110
5 19.1 23 130
6 21.3 22 150
7 24.5 21 162
8 27.2 20 180
9 29.9 20 190
10 33.1 19 220
11 37.2 19 235
12 39.5 19 245
13 44.9 19 275
14 51.3 19 300
15 58.1 18 340
16 62.1 17 365
17 64.9 17 380
18 67.6 16 405
19 69.4 16 420
22 71.7
27 73.9
32 74.8
37 75.3
44 75.8
54 74.8
64 73.5
74 71.2
Tidal Vol./ Calculated Breathing Rate
Weight
(ml/kg)
(*)
(ml/min) (m /year)
Average value of 6.36+0.38 (ml/kg)
10
(3.82) 480 + 227 252
6.44 2,050 1,077
6.85 2,100 1,104
6.76 2,305 1,212
6.55 2,564 1,348
6.81 2,795 1,469
7.04 2,980 1,566
6.61 3,272 1,720
6.62 3,460 1,819
6.35 3,800 1,997
6.65 4,000 2,102
6.31 4,495 2,363
6.20 4,773 2,509
6.12 5,426 2,852
5.85 6,200 3,259
5.85 6,651 3,496
5.88 6,714 3,529
5.86 7,017 3,688
5.99 6,878 3,615
6.05 7,062 3,712
7,000 3,679
7,000 3,679
7,000 3.679
7,000 3,679
7,000 3,679
7,000 3,679
7,000 3,679
7,000 3,679
used for further calculations..
15
-------
The rate of pollutant uptake in adults is given by equ. (5).
Ruptake ' Bx
The value of B in m3/kg-day is equal to the daily tidal respiratory
volume per kg body weight assuming all of the inhaled pollutant is ab-
sorbed, a = 1. For a 70 kg adult, the value of B is 0.143 m3/kg day,
according to the respiratory data presented in Table 2.
The pollutant accumulating in the body is metabolized and excreted
either unchanged or in form of metabolites. The rate of excretion can be
assumed to be proportional to the instantaneous pollutant.concentration in
the body, c, i.e., the excretion proceeds according to a first order rate
law given either by
-dc/dt = kc (6)
or in its integrated form by
c = CQ exp(-kt). (7)
From equ. (7) one obtains the biological half-life time, T, which is equal
to the time interval during which the body concentration decreases to half
its value provided there is no uptake during this interval.
T = In2/k. (8)
The actual change of the concentration of the pollutant in the body
is then generally described by combining equs. (5) and (6), and is given
by
dc/dt = Bx - kc. (9)
Equation (9) indicates that after a prolonged time the body concentra-
tion approaches a steady state with a constant value c corresponding to
the maximum body concentration under given conditions which is given by
S, = Bx/k (10)
16
-------
or, after substituting the half-life time of the pollutant from equ. (8),
by
cm = BxT/ln2. (11)
m
The integrated equ. (9), assuming B is constant, describes the accumulation
of pollutant in the body with time provided the air concentration remains
constant.
c = (Bx/k) [1 - exp(-kt)] + CQ exp(-kt) (12)
The value CQ corresponds to the initial body concentration at the start
of the exposure. If there was no previous exposure to the pollutant under
consideration equ. (12) simplifies to
c = (Bx/k) [1 - exp(-kt)]. (13)
From equ. (13) one can calculate the time necessary to accumulate
half the stationary maximum body concentration c . This value is also
given by the biological half-life time of the pollutant, T. Provided
the pollutant concentration in the air remains constant, it takes about
7 half-lifes for the pollutant to approach a steady state concentration
within 1%.
D- Application of Kinetic Considerations
Utilization of Biological Half-Life
The basic equation, equ. (11), describing the stationary maximum body
concentration, was simplified by substituting numerical values for the
constants, B = 0.143 and In2 = 0.693. This was approximated by equ. (14),
cm = XT/5. (14)
Equation (14) provides a good estimate for permissible air concentra-
tions of pollutants as a function of the biological half-life of the
pollutant, T, and the maximum body concentration, cm, which does not cause
adverse health effects over extended periods of time. Unfortunately,
both values are not known for the majority of compounds.
17
-------
In Table 3, biological half-lifes for a very small number of com-
pounds are presented which could be retrieved from the literature.
Most of the values were determined on man and cover a range from about
17 min to nearly 6 days.
By means of the half-life it was possible to calculate the stationary
body concentration which builds up in an individual exposed to the presently
permitted pollutant concentration in working room air (TLV). Data were
calculated for continuous 24 hour exposure (Table 4) and for 8 working hour-
exposure followed by 16 hour and weekend recovery periods in unpolluted
surroundings (Table 5).
The latter values are only 23.8% of the former ones. It can be
shown (see Appendix B) that repeated 8 hour exposures followed by 16 hour
recovery intervals decrease the average stationary body concentration to
one-third of the value for continuous 24 hour exposure. By including week-
ends for recovery (40 hour workweek) the body concentration averaged over
one week is further reduced to 23.8% of the maximum value obtained during
continuous exposure.
In both Table 4 and 5, LDgg values (in mg/kg determined orally on rats)
were included and related to the maximum body concentrations. At TLV con-
centration levels an individual accumulates less than 1% of the LDcn of
ou
the compound. Indeed, most of the Cm values were below 0.1% of the LD5Q
values. Although biological data were extremely sparse, a permissible
maximum body concentration being equal to 0.05% of the LD5Q value might
be used as a guide in further estimations of permissible air concentrations.
Utilization of LD5Q Values
Although in equ. (14) the maximum body concentration, c , might be
replaced by 0.0005 x LDgo there remains still one unknown, the biological
half-life time of the pollutant, in order to calculate permissible air
concentrations.
It follows from equ. (14) that at constant cm, the permissible air
concentration becomes smaller with increasing values of T. Consequently,
the safest assumption would be to assume T to be equal to infinity, i.e.,
18
-------
Table 3. BIOLOGICAL HALF-LIFE TIMES OF DIFFERENT COMPOUNDS11
Compound
Methylchloride
p-Nitrophenol
Carbondisulfide
it
Benzene
Toluene
Aniline
Dinitro-o-cresol
Methanol
nimal
dog
man
mam
man
man
man
man
man
rat
rbt
man
man
Medium or Coefficient of
Metabolite Loss (hr~l)
blood
urine
expired air
metabolite
blood
benzoic acid
p-aminophenol
blood
blood
blood
urine
formate
2.5
0.7
0.8
0.5
0.23
0.35
0.24
0.005
0.04
0.1
0.1
0.3
Half-Life Time
(days)
0.012
0.041
0.036
0.058
0.126
0.083
0.120
5.78
0.722
0.289
0.289
0.096
Compound
Table 4. MAXIMUM BODY CONCENTRATIONS FOR CONTINUOUS
EXPOSURE CALCULATED FROM TLV AND HALF-LIFE TIMES
DETERMINED ON MAN
Half-Life TLV
Time (days) , , 3N
3 (mg/m )
c LDcn
m 50
(mg/kg) (mg/kg)
-------
Table 5. MAXIMUM BODY CONCENTRATION FOR 8 WORKING HOURS EXPOSURE
AND 40 HOUR WORK WEEK CALCULATED FROM TLV STANDARDS AND
HALF-LIFE TIMES DETERMINED ON MAN
Compound
Methylchloride
Carbondisulfide
Benzene
Toluene
Aniline
Dinitro-o-cresol
Methanol
Half-Life
Time (Days)
0.012
0.047
0.126
0.083
0.120
5.78
0.193
TLV
2
(mg/m )
210
60
80
750
19
0.2
260
c
m
(mg/kg)
0.12
0.13
0.48
2.96
0.11
0.06
2.39
LD50
(mg/kg)
1,800
400
3,400
3,000
440
30
420
(cm/LV
x 100
0.007
0.033
0.014
0.100
0.025
0.182
0.569
cm = B x (TLV) x 0.238 x T/£n 2 = 0.0476 x (TLV) x T
inhalation volume of 10 nr per day and 70 kg weight with the assumption of
100% absorption of the inhaled pollutant.
*
Half-life time determined on dogs.
20
-------
to make k = 0. This assumption means that there is no excretion of the
absorbed pollutant and that all of the inhaled pollutant accumulates
indefinitely in the body. Under this condition the body concentration
c in mg/kg accumulated during the time period t2 - t| is then given by
c = [x/W(t2)] / F(VTf, t) dt (15)
where t-j ............ age at the start of exposure
tg ............ age at the end of exposure
W(t«) ......... body weight (kg) at the end of exposure
f ™
F(Vjf, t) ..... breathing rate (tidal volume times breathing
frequency) as a function of age (t)
o
x ............. pollutant concentration (mg/nr) assumed to
remain constant during the time period t2~t-,
Pertinent respiratory data for the evaluation of equ. (15) are summarized
in Table 2. A graphical representation of equ. (15) is shown in Fig. 1.
where final body concentrations in mg/kg are plotted versus age for an
air concentration of pollutant equal to 1 wg/m . If the air concentration
differs from 1 ug/m the body concentrations have to be prorated, e.g.,
3
with an air concentration of 1 mg/m all body concentrations are in g/kg.
Figure 1 depicts several curves representing different starting ages with
respect to exposure to polluted air (0, 5, 10, 22, 27, and 44 years of age).
In Fig. 2 a similar plot is presented demonstrating body concentrations
(mg/kg) accumulated from birth (t, = 0 in equ. (15)) for different air
3
concentrations of 1 , 2, 5, 10, and 20 microgram/m .
E. Correlation Between TLV and LD5Q Values
In order to establish safe TLV estimates one has to be guided primarily
by the toxicity of the compound, and consequently, one might expect a
correlation to exist between TLVs and LD5Q values. The possibility of such
a correlation was investigated by Monsanto for 30 selected agricultural
chemicals. In compiling the data no distinction was made between inhalation
and skin TLV, and the acute oral LD was taken for male rats.
21
-------
20
30 4O
YEARS
50
60
Figure 1.
Accumulated body burden in mg/kg for different starting times
of continuous exposure to polluted air containing 1 vg/nr
pollutant. Curves calculated with the assumption of an in-
finite biological, life-time of the pollutant in the body.
Accompanying numbers correspond to the starting age with
respect to exposure.
22
-------
50
60
YEARS
Figure 2. Body burden in mg/kg accumulated during continuous exposure
to air containing different concentrations of pollutant.
Curves calculated with the assumption of an infinite bio-
logical life-time of the pollutant in the body. Accompanying
numbers correspond to the pollutant concentration in
23
-------
The best regression fit was on an equation of the type
(TLV) = a (LD50)b (16)
after logarithmic transformation to
Ig(TLV) = Iga + b 1g(LD50)
The linear regression according to equ. (17) yielded a correlation coeffi-
cient of 0.89, and the values of the constants were found to be
a = (0.00916 < 0.00198 < 0.0429) = 95% (18)
b = (0.629 < 0.774 < 0.920) = 95% (19)
Data are summarized in Table 6 and are shown graphically in Fig. 3 together
with the calculated regression line and its 95% confidence range for a
predicted TLV.
With additional 20 agricultural chemicals the correlation coefficient
did not change. In this evaluation the lowest reported LDgQ values were
used which correspond to the underlined values in Table 6. The new constants
for the enlarged sample number were
a = (0.0292 < 0.0483 < 0.0797) = 95% (20)
b = (0.574 < 0.675 < 0.776) = 95% (21)
Graphical representation of the sample of 50 agricultural chemicals is
shown in Fig. 4 together with the calculated regression line and its 95%
confidence interval for predicted TLVs.
Because all the compounds are related agricultural chemicals, some
bias might exist in their correlation thus making the results doubtful for
general application. A list was established (Table 7) for 191 compounds
for which TLVs and LD5Q values were available. The TLVs were taken from
"Documentation of the TLVs for Substances in Workingroom Air", Am. Soc.
Govern. Ind. Hygienists, 3rd Ed., 1971, 2nd Printing 1974.12 The LD™
24
-------
Table-6. TLV AND LD5Q FOR DIFFERENT AGRICULTURAL CHEMICALS
Compound
TLV
12
LD__ Values (mg/kg) from Different Sources
1.
2.
3.
4.
5.
6. -
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
Abate
Aldrin
Allyl alcohol
Animate
Arsenic acid
Carbaryl (Sevin)
Chlordane
Toxaphene
2,4-D
DDT
DDVP
Demeton (Systox)
Diazinon
Dibrom (Naled)
Dieldrin
Dinitro-o-cresol
Diquat
Endrin
EPN
Heptachlor
Malathion
Methoxychlor
Methylparathion
Paraquat
Parathion
Phosdrin
Ronnel
2,4,5-T
TEPP
Thiram
Lindane
TEDP
Rotenone
Pyrethrum
Baygon (Propoxur)
Cap tan
Carbofuran (Furadan)
Chlorpyrifos (Dursban)
Clopidol (Coyden)
Crufomate (Ruelene)
Disulfoton
Phorate (Thimet)
(mg/m )
10
"0.25
3
10
0.5
5
0.5
Q.5
10
1
1
0.1
0.1
3
0.25
0.2
0.5
0.1
0.5
0.5
10
10
0.2
0.5
0.1
0.1
10
10
0.05
5
0.5
0.2
5
5
0.5
5
0.05
0.2
10
5
0.1
0.05
1
Monsanto
2000
55
95
3900
48
500
570
69
1200
113
56
9
' 134
430
60
50
300
5
50
90
1375 •
5000
25
145
15
7
1740
500
1.2
860
ACGIH12
4000
60
—
—
—
500
590
73
—
250
**«
4
34
250
—
31
—
—
—
• —
; 2100
6000
14
127
3
6.5
1250
300
3
1300
125
5
132
820
83
9000
8
120
8000
490
2.3
1.1
TSL]
2000
39
69
3900
—
89
283
60
375
113
56
1.7
76
—
_46_
' 30
231
3
8
40
599
5000
4
57
2
4
906
300
0.5
560
76
5
132
—
83
480
8
145
i —
660
10
1
25
-------
Table 6 (Continued) . TLV AND LD50 FOR DIFFERENT AGRICULTURAL CHEMICALS
Compound
43. Picloram (Tordon)
44. Plictran
45. Antu
46. Azinphos-Methyl
47. Crag (Sesone)
48. Endosulfane (Thiodan)
49. Ferbam
50. Methyldemeton
TLV
(mg/m )
10
5
0.3
0.2
10
0.1
10
0.5
LD Values from Different Sources
Monsanto
ACGIH
8000
540
30
16.4
1500
20
17,000
40
TSL
3750
—
6_
11
730
28
4000
15
26
-------
-I
0
Figure 3. Plot of Ig (TLV) versus Ig (LD50) for 30 agricultural chemicals
selected by Monsanto Research Corporation.'
27
-------
0
0
Figure 4. Plot of Ig (TLV) versus Ig (LD5Q) for 50 agricultural chemicals,
(o) compounds of Fig. 3, (A) aaaitional compounds.
28
-------
Table 7. LIST OF TLV12
13
AND LD5Q FOR DIFFERENT CHEMICAL COMPOUNDS
Compound
1. Acetaldehyde
2. Acetic acid
3. Acetic anhydride
4. Acetonitrile
5 . Acrolein
6 . Acrylamide
7 . Acrylonitrile
8. Allylchloride
9. Allylglycidyl ether
10. Ammonium chloride
11. Anunoniumsulf amate
12. Aniline
13. Antimony dust
14. Benzene
15. Benzylchloride
16. Berry Ilium oxide
17. Bromoform
18 . 2-Butanone
19 . 2-Butoxyethanol
20. Butanol
21. t-Butanol
22. Butylamine
23. Butylglycidyl ether
24. p-t-Butyltoluene
25. Cadmium oxide
26. Calcium arsenate
27. Camphor
28. Carbondisulfide
29. Chloroacetaldehyde
30 . alpha-Chloroacetophenone
31. Chlorobenzene
32. Chlorobenzylidene malononitrile
33. Aroclor 1254
34. Chloroform
35 . 1-Chloro-l-nitropropane
36 . Chloroprene
37. o-Cresol
38. Crotonaldehyde
39 . Cumene
40. Cyclohexane
41. Cyclohexanol
42. Cyclohexanone
43. Decaborane
44. Diacetone alcohol
45 . 1 , 2-Dibromoethane
46 . 2-N-Dibutylaminoethanol
•fc^I T
A
(mg/m )
180
25
20
70
0.25
0.3
45
3
22
10
10
19
0.5
80
5
0.002
5
590
240
300
300
15
270
60
0.2
1
12
60
3
0.3
350
0.4
0.5
250
100
90
22
56
250
1050
200
200
0.3
240
190
14
"50
(mg/kg)
1930
3310
1780
160
46
170
93
64
922
1650
3900
440
100
3400
1231
54
1820
3100
1480
2510
3500
500
2050
1500
72
298
900
400
23
52
2910
178
500
300
510
64
121
300
1400
813
2060
1620
64
4000
140
1070
Remarks
itr-rat TDLC
ipr-rat LDL0
orl-rat LDLQ
orl-mus LD50
orl-rat LDL0
orl-mus LDL0
29
-------
Table 7 (Continued). LIST OF TLV12 AND LD5Q FOR
DIFFERENT CHEMICAL COMPOUNDS
Compound
47. Dibutylphosphate
48. o-Dichlorobenzene
49. p-Dichlorobenzene
50. Dichlorodimethyl hydantoin
51. 1,1-Dichloroethane
52. 1,2-Dichloroethane
53. 1,1-Dichloro-l-nitroethane
54. Dimethylamine
55. Diethylamino ethanol
56. Diethylenetriamine
57. Diglycidyl ether
58. Diisobutylketone
59. Diisopropylamine
60. Dimethylamine
61. Dimethylaniline
62. Dimethylforamide
63. l-l~Dimethylhydrazine
64. Dimethylnaphthalene
65. Dimethylsulfate
66. Dinitrobenzene
67. Dinitro-o-cresol
68. Dinitrotoluene
69. Dioxane
70. Diphenyl
71. Diphenylamine
72. Dipropyleneglycol methylether
73. Epichlorhydrine
74. Ethanolamine
75. 2-Ethoxyethanol
76. 2-Ethoxyethyl acetate
77. Ethylacrylate
78. Ethanol
79. Ethylamine
80. Ethyl-sec-amyl ketone
81. Ethylbenzene
82. Ethylbutyl ketone
83. Ethyl ether
84. Ethyl formate
85. Ethylene chlorhydrine
86. Ethylenediamine
87. Ethyleneglycol dinitrate
88. Ethyleneimine
89. Ethyleneoxide
90. N-Ethylmorpholine
91. Formaldehyde
92. Formic acid
93. Furfural
94. Furfurol
95. Glycidol
TLV
LD
50
5
300
450
0.2
400
200
60
75
50
4
2.8
290
20
18
25
30
1
5
5
1
0.2
1.5
360
1
10
600
19
6
740
540
100
1900
18
130
435
230
1200
300
16
25
1.2
1
90
94
2.5
9
20
200
150
3200
500
500
542
725
680
150
540
1300
1080
450
1416
700
698
1410
1500
122
4400
440
27
30
177
3150
2180
464
7500
90
2100
3000
1910
830
5560
400
2800
3500
2760
2200
1850
71
1160
616
15
330
1780
800
1210
127
275
850
Remarks
orl-mus LD50
orl-rat LDL0
orl-rbt LD50
orl-cat LDL0
orl-gpg LD50
orl-dog LD50
orl-gpg LD50
orl-gpg LD50
orl-rat LDL0
30
-------
Table 7 (Continued) . LIST OF TLV12 AND LDC^13 FOR
DIFFERENT CHEMICAL COMPOUNDS
50
Compound
96. 2-Hexanone
97. Hexane
98. Hydrazine
99. Hydrocyanic acid
100. Hydroquinone
101. Ironpentacarbonyl
102. i-Amylalcohol
103. Isobutanol
104. Isophorone
105. Isopropylacetate
106. Isopropanol
107. Isopropylamine
108. Isopropylglycidyl ether
109. Ketene
110. Lead arsenate
111. Maleic anhydride
112. Methylacetate
113. Methylacrylate
114. Methanol
115. Methy1-i-amylketone
116. Methyl-n-amylketone
117. Methylcellosolve
118. Methylcellosolve acetate
119. Methylchloride
120. Trichloroethane
121. Methylcyclohexane
122. o-Methylcyclohexanone
123. 2-Methylcyclopentadienyl-
manganese tricarbonyl
124. Methyliodide
125. Methy1-i-butylcarbinol
126. Methylisocyanate
12 7. alpha-Methyls tyrene
128. Methylene chloride
129. Methylaniline
130. Monomethyl hydrazine
131. Morpholin
132. Naphthalene
133. Nicotine
134. p-Nitroaniline
135. Nitrobenzene
136. p-Nitrochlorobenzene
137. Nitroethane
138. Nitromethane
139. 1-Nitropropane
140. 2-Nitropropane
141. o-Nitrotoluene
142. Nitroglycerine
TLV
LD
50
Remarks
410
410
1.3
. 11
2
0.08
360
300
140
950
980
12
240
0.9
• 0.15
1
610
35
260
460
465
80
120
210
1900
2000
460
0.2
28
100
0.05
480
1750
9
0.35
70
50
0.5
6
5
1
310
250
90
90
30
1.2
2590
1620
60
0.37
370
18
3380
3500
2330
3200
192
820
4200
1300
100
850
4800
300
420
4760
1670
3000
1910
1800
5660
4000
2140
23
50
2600
71
4900
3000
280
32
1050
1780
70
3249
640
420
1100
940
1000
500
891
80
orl-mus LDSO
orl-rbt LDL0
orl-rat LDL0
orl-mus LDL0
orl-rat LDL0
orl-rat LDL0
orl-mus IDL0
orl-gpg LDSO
orl-rbt LDSO
orl-rbt LDL0
orl-rat LDLC
orl-dog LDLo
orl-rbt LDL0
orl-rat LDL0
orl-rat LDL0
orl-rat LDL0
31
-------
Table 7 (Continued) . LIST OF TLV12 AND LD5Q13 FOR
DIFFERENT CHEMICAL COMPOUNDS
Compound
143. Oxalic acid
144. Pentachlorophenol
145. 2-Pentanone
146. Perchloromethyl mercaptane
147. Phenol
148. Phenothiazine
149. p-Phenylenediamine
150. Phenyl ether
151. Phenylglycidyl ether
152. Phenylhydrazine
153. Phosphoric acid
154. Phosphorus
155. Phosphorpentasulfide
156. Phthalic anhydride
157. Pival
158. Propargyl alcohol
159. n-Propanol '
160. Propylenedichloride
161. Propylene imine
162. Propylene oxide
163. Pyridine
164. Quinone
165. RDX
166. Sodium fluoroacetate
167. Sodium hydroxide
168. Strychnine
169. Styrene
170. Sulfuric acid
171. Sulfuryl fluoride
172. Tetrabromoethane
173. Tetrachloroe thane
174. Tetraethyl lead
175. Tetrahydrofuran
176. Tetramethyl lead
177. Toluene
178. o-Toluidine
179. Tributylphosphate
180. Trichloroethane
181. Trichloroethylene
182. Trichloropropane
183. Triethylamine
184. Trinitrotoluene
185. Tri-o-cresylphosphate
186. Triphenylphosphate
187. Vinyl acetate
188. Vinyl toluene
189. Worfarin
190. Xylene
191. Xylidine
TLV
LD
5Q
1
0.5
700
0.8
19
5
0.1
7
62
22
1
0.1
1
12
0.1
2
490
350
5
240
15
0.4
.1.5
0.05
2
0.15
420
1
20
14
35
e.i
590
0.15
750
22
5
45
535
300
100
1.5
0.1
3
30
480'
0.1
440
25
700
27
3730
83
414
5000
100
4000
3850
188
1530
1.4
389
4020
150
70
1870
1900
19
1140
891
130
200
1.7
500
16
4920
2140
100
1000
30
17
3000
109
3000
900
3000
580
4920
320
460
700
4680
3000
2920
4000
325
4300
610
Remarks
orl-hmn LDL0
orl-rat LDL0
orl-rat LDL0
orl-hmn LDL
orl-rat LDL
orl-rbt LDL
orl-rat LDL0
orl-rat LDL0
orl-hmn TDL0
orl-rat LDL0
orl-rat LDL0
orl-rat LDL0
orl-rat LDL0
orl-rat LDL0
orl-rat LDL
32
-------
values were taken from "The Toxic Substances List", HEW, 1974.13 As
far as available LDgo values (oral-rat) were used. In several cases
oral LD5Q values for other animals (mouse, guinea pig, dog, cat, etc.)
had to be used. If LD5Q values were not reported at all LDL values
were taken again with preference of rats.
The graphic representation of this group of 191 compounds is shown
in Figure 5. Linear regression yielded a correlation coefficient of 0.61.
The values of the constants of equ. (17) were found to be
a = (0.00971 < 0.0343 < 0.121) = 95% (22a)
b = (0.800 < 0.991 < 1.182) = 95% (22b)
After inclusion of the previous 50 agricultural chemicals (total
of 241 compounds) the correlation coefficient improved slightly to 0.70.
The values of the constants in this case were
#
a = (0.0125 < 0.0291 < 0.0678) = 95% (23)
b = (0.849 < 0.983 < 1.117) = 95% (24)
A graphic representation of the 241 compounds is shown in Figure 6.
It might be pointed out that the 95% confidence limits given in
equs. (18) to (24) refer to the estimates of the mean and not to the
estimate of an individual prediction, i.e., the values represent confi-
dence limits for the constants of the regression line and not for the
prediction of an individual TLV from a given LDg0 value. Confidence
ranges inside which 95% of the TLVs predicted from the regression
equation might fall on repeated trials are shown by the dashed lines
in Figures 3 to 6.
The poor correlation between TLVs and LDgg values is not surprising
if one considers the spread in TLVs for a fixed LDgg value. The ratio
of TLV maximum to TLV minimum for a given LD5Q value is generally in the
range of 1,000 to 10,000. There exists also a distinct cut-off at LD5Q
values around 5-6 g/kg. In this range the lowest reported TLV is 0.1
, mg/m3 whereas the highest reported TLV is 1,800 mg/m .
33
-------
\
Figure 5. Plot of Ig (TLV) versus Ig (LD5Q) for 191 non-agricultural
compounds.
34
-------
=5
o>
Figure 6. Plot of Ig (TLV) versus Ig (LD5Q) for 191 non-agricultural
(o) and 50 agricultural (A) compounds.
35
-------
F. Estimation of Maximum Permissible Air Concentrations from Known
LD5Q Values
Safe estimates of permissible air concentrations have to be based on
two assumptions, one concerning the accumulation of pollutant in the
body and the other one concerning the highest permissible body concentra-
tion which does not create adverse health effects.
If one assumes that all of the pollutant inhaled remains in the body
and accumulates from birth for a life span of 70 years, a final body
concentration results which is unrealistic by several orders of magnitude.
According to equ. (15) or Figures 1 and 2, the body concentration accumu-
lated during 70 years of exposure, C7Q, is given by
C7Q = 2.9 x 103 x (25)
3
where 0™ and x are in mg/kg and mg/m , respectively.
Biological data presented in Table 5 allow us to draw two conclusions,
(1) according to the presently suggested TLV standards a stationary maximum
body concentration corresponding to 0.05% of the LDgg value can be considered
as a safe limit, (2) the assumption of a biological half-life of 30 days
far exceeds the residence time in the human body for most compounds. By
combining both assumptions permissible air concentrations can be calculated
from equ. (11). The maximum permissible air concentration is then defined
o
by the pollutant concentration (mg/m ) for which continuous exposure with
100% absorption gives rise to a stationary maximum body concentration equal
to 0.05% of the LD5Q value of the compound, assuming a biological half-life
time of 30 days.
From equ. (11) the maximum permissible air concentration, subjected to
the above postulated conditions, is given by
xp = 0.0005 (LD5Q) (InZ/Bt) " (26)
or, after inserting numerical values by
x = (0.0005 x 0.693/0.143 x 30) (LD5Q) = 8.1 x 10"5 (LD5Q) (27)
36
-------
where xp and LD5Q are expressed in mg/m3 and mg/kg, respectively. The
numerical value of B = 0.143 m3/kg.day corresponds to a 70 kg adult and
a tidal breathing rate of 10.0 m3/day.
Regression analysis of 241 compounds according to equ. (17) yielded
the values of the constants presented in equs. (23) and (24). The slope
of the regression line in the double-logarithmic plot of (TLV) versus
(LDgQ) was found to be b = 0.983, and therefore, a direct proportionality
between (TLV) and LDgg) can be assumed without introducing any substantial
error.
*
(TLV) - 0.029 (LD5Q) (28)
For the lower 95% confidence limit the (TLV) is then obtained from (LD,-Q)
values by
(TLV)low = 4'5 x 10~4 (LDW} (29)
Since intermittent exposure (40 hour work week of 5 days with 8
hour exposure) gave rise to a weekly body concentration buildup only
23.8% of that experienced during continuous exposure, the factor
0.238 has to be included into equ. (29). The maximum permissible air
concentration of a compound is then related to its LD5Q value by
xp = 1.07 x 10"4 (LD5Q) (30)
Comparison of equs. (27) and (30) shows close correspondence thus
indicating that both equations are based on similar ratios of maximum
stationary body concentration to biological half-life time of the
component.
For the determination of a pollutant hazard to exist during continuous
exposure the following inequality was proposed by Monsanto:
*m > 1.0 (31)
(TLVT (8/24) (1/100) -
37
-------
1.07 x 10"4 (LDKn)
where TLV is the presently suggested standard for working room air.
The factor (8/24) considers the intermittent exposure under working
conditions, but does not consider weekends. The factor (1/100) is
a safety factor. If for any pollutant the left side of equ. (31)
is equal or greater than 1.0, then a hazard is judged to exist.
The maximum permissible pollutant concentration not to be exceeded
during continuous exposure is given by equ. (30). Consequently, the
air quality criteria corresponding to equ. (31) but based on LDgQ values
is given by
>.1.0 (32)
'50'
Comparing equ. (32) with equ. (31) the following changes have been applied.
The TLV of equ. (31) was replaced by (TLV)lQW as calculated from LD5Q by
regression analysis. The factor (8/24) has been replaced by (40/168) to
take into account normal working patterns during a calendar week. The safety
factor (1/100) was omitted because of the use of the lower 95% confidence
limit in the calculation of (TLV)lo from LD5Q values.
Equation (32) should be applied in all cases where LD,-Q values are
known. It is now possible that for some pollutants TLV standards are
given although the LD5Q value of the component is not known. In this,
case equ. (32) would not be directly applicable and in a multicomponent
system two sets of equations would have to be used, one based on LD5Q values,
equ. (32), the other one on TLV standards, equ. (31). In order to overcome
this complication the unknown LD50 value is estimated from the given
TLV standard by solving the regression equation, equ. (28) for LD5Q
(LD5Q) = 34.5 (TLV) (33)
This estimated LDgg value is then substituted into equ. (32).
The expression for permissible air concentrations given by equs. (27)
and (30) were calculated for the respiratory conditions of adults, i.e.,
for an average breathing frequency of 16 per nrin. In the Appendix, Section
VII-A, the effect of the higher breathing frequencies is discussed.
38
-------
Considering that the respiratory frequency of an infant of one
year of age is about twice that of adults, a factor of 0.5 has to be
applied to equ. (30) in order to make the permissible air concentration
safe for all age groups. The corrected equ. (30) is then given by equ. (34).
xp = 5.35 x 10'5 (LD5Q) (34)
Considering the variation in the respiratory frequency during growth
of an infant through adolescence it could be shown that during this time
*
interval the body concentration never exceed that of an adult by more than
a factor of 1.7. Consequently, equ. (27) has to be modified correspondingly,
the corrected permissible air concentration being now given by
xp = 4.77"x 10"5 (LD5Q) (35)
Estimates of x values for 5 compounds were obtained by three
different methods (Table 8). In Method I, the current TLV standards and
equ. (31) were used to compute x estimates.
In Method II, the TLVs were estimated from LD5Q values by means of
equ. (29). Equation (34) was applied for the determination of x values.
Method III utilized equ. (35). The maximum stationary body concentra-
tion does not exceed a concentration corresponding to 0.05% of the LDj-Q
value, assuming a pollutant half-life of 30 days.
The close correspondence of x values estimated either by Method II or
by Method III demonstrates that a correlation between TLV and LDj-g is
unnecessary. Values of x for compounds with unknown TLV and a known LD5Q
can be directly obtained from equ. (35).
So far mainly organic compounds were considered, inorganic compounds
presenting considerable problems due to lack of sufficient data. In
Table 9 four metals are compared for which data could be retrieved.
According to the presently used TLV standards a rather high percentage of
the LDc0 value accumulates during normal working exposure.
39
-------
Table 8. PERMISSIBLE AIR CONCENTRATIONS FOR CONTINUOUS EXPOSURE ESTIMATED BY DIFFERENT METHODS
Compound
Table/Number
LD5Q (mg/kg)
o
Presently Proposed TLV (mg/m )
q
Estimated (TLVK (mg/m )
Permissible Air Concentration x
P
Method I
Method II
Method III
Method I: Eq. (31) applied to
Method II: Eq. (34) applied to
Method III: Eq. (35) applied to
1/39
8,000
10
3.6
q
in mg/m Estimated
0.033
0.43
0.38
presently proposed
(TLV).. estimated
v low
LD,.ft value, x =
11/121
4,000
2,000
1.8
by:
6.6
0.22
0.19
TLV. x =*
P
from LD
4.77 x 10~5
ic xct-ietuti-Lui-uci-ticiuc
11/173
30
35
0.014
0.12
0.0016
0.0014
0.0033 (TLV)
xp= 5.35x10-^ (LD5Q)
(LD™>
1/29
0.5
0.05
0.0002
0.00016
0.000027
0.000024
11/126
71
0.05
0.032
0.00016
0.0038
0.0034
-------
Table 9. MAXIMUM STATIONARY BODY CONCENTRATION ACCUMULATED DURING
DURING WORKING CONDITIONS (40 HOUR .WORKWEEK) FOR SELECTED INORGANIC
COMPOUNDS
Compound
PbMe4
PbEt4
Cr
Hg-Alkyls
Cd
Half-Life
(Days)
2920
2920
83
120
5475
TLV
3
(mg/m )
0.15
0.1
1.0
0.01
0.1
c
m
(mg/kg)
20.8
13.9
4.0
0.057
26.1
LD50
(mg/kg)
109 (a)
17 (a)
400 (b)
8 (c)
15 (d)
<*JU>50> X
19.1
81.8
1.0
0.7
173.7
(a) orl-rat
(b) imp-rat TDL0 (c) ipr-mus LDL0 (d) ims-rat LDLe
Permissible Air Concentrations, mg/m , for Continuous Exposure Estimated by
Different Methods
Compound
Method I
= 0.0033 (TLV)
PbMe4
PbEt4
Cr
Hg-Alkyls
Cd
0.00050 (0.3)
0.00033 (1.1)
0.0033 (0.01)
0.00003 (0.01)
0.00033 (2.4)
Method II
x = 5.35 x 10" (LDe
P
0.0059 (3.1)
0.0009 (3.1)
0.022 (0.1)
0.0004 (0.1)
0.0008 (5.9)
Method III
= 4.77x10
-5
0.0049 (2.7)
0.0008 (2.7)
0.0181 (0.06)
0.0003 (0.1)
0.0007 (5.0)
(LD50>
Values in parenthesis indicate the maximum stationary body concentration in percentage
of LD5Q values which accumulate during continuous exposure.
41
-------
In the lower part of Table 9 permissible air concentrations for
continuous exposure are estimated by the three different methods dis-
cussed previously. For each estimated air concentration the maximum
stationary body concentration expressed as percentage of LD5Q is
given in parenthesis. Comparing these values with those resulting from
presently set TLV standards demonstrates that the estimates obtained by
Methods II or III can be considered safe.
G. Estimation of Air Quality Standards For Multicomponent Systems
When two or more pollutants are present,, their combined effect,
rather than that of either individually, should be given primary consi-
deration. In the absence of information to the contrary, the effects of
2
the different pollutants should be considered as additive. That is, if
the sum of the following fractions
jj [(XnVfXpJj] >- 1 (36)
exceeds unity, a health hazard should be judged to exist. In equ. (36)
(xm).. indicates the observed maximum ground level concentration of the
i-th pollutant and (x ).. corresponds to its estimated maximum permissible
air concentration.
Equation (36) can be evaluated by the three previously discussed
methods by substituting for x the proper value given by equs. (31),
(34), or (35), respectively. The air quality standards according to the
three methods are then given as follows.
Meth°d l J 0.003MTLV). i 1 07)
n (x ).
Method II £ E-L, > i (38)
i 5.35 x 10"* (LD5Q)1 ~
n (xj.
Method III i 2LJL- > -j /39)
i 4.77 x 10'5 UD50). ~
42
-------
As an example, data of Table 10 show the evaluation of selected
Los Angeles air data by three different methods. In all three evalua-
tions the absolute prorated and the percentage prorated contribution of
each individual pollutant are shown.
The evaluation by Method I was performed according to the presently
used standard procedure which is based on presently set TLV standards,
equ. (37). For the evaluation by Method II the TLVs were assumed to be
unknown and the evaluation was based on ID™ values, equ. (38). In the
evaluation by Method III equ. (39) was utilized.
Data of Table 10 indicate that in order to qualify for the Air
Quality Standard the average air concentration of each compound would have
to be reduced by a factor of 0.44 or 0.39 provided the relative amounts of
each compound remained unchanged, and either Method II, equ. (38), or Method
III, equ. (39), is applied, respectively.
H. Estimation of Maximum Ground Level Concentrations From Stack Emission
Data of Stationary Sources
Assessments of health hazards associated with air pollutants from
process industries require the knowledge of the maximum ground level
concentration, xm, of pollutants during a 24 hour period. According to
the first decision criterion for possible health hazards expressed in equ.
(36), the value of x must not exceed the permissible ground level concen-
tration, x . In the preceding sections three methods were outlined for
the determination of x which resulted in equs. (37-38). The present
section deals with the estimation of xm generated by stationary sources.
The maximum ground level concentration generated by a stationary
source is affected by both the operation conditions of the source and the
ambient weather conditions. Consequently, operation conditions have to
be adjusted according to the current weather conditions in order not to
generate maximum ground level concentrations in excess of the permissible
ones.
In the Appendix, Section VII-C, two nomographs are presented as
aids for obtaining first estimates of the effects of changes in both
the emission rate of the source and the prevailing weather conditions
on the maximum ground level concentration of the pollutant. These
43
-------
Table 10. ANALYSIS OF A MULTICOMPONENT SYSTEM (SELECTED VALUES FOR LA AIR)
TLV
Reported Average
Compound
Xylene
Toluene
Benzene
Naphthalene
Hexane
3 50
(mg/m ) (mg/kpj)
440 4,300
750 3,000
80 3,400
50 1,780
410 1,620
Air Quality Standard
Air Quality
Standard calculated by:
O.JL.L i^uncentracion Aosoxuce ^percent; t
/ / 3. „ . .
(mg/m ) Method I
0.130 0.090 (23.1)
0.140 0..057 (13.4)
0.050 0.189 (44.6)
0.008 0.048 (11.3)
0.043 0.032 (7.5)
0.424 (100)
n (x )
Method I I m'i
equ. (37) 1 0.0033 (TLV)±
** (x )
M*»fhoH TT T. mi
equ. (38) 1 5.35xlo-5 (L^^
n (x )
Method III E mi
equ. (39) 1 A -,-r „ in~5 nn \
jontrioucion to AI
Method II
0.505 (24.6)
0.872 (38.0)
0.275 (12.0)
0.084 (3.7)
0.496 (21.6)
2.292 (100)
r v^uaxity atanuatu
Method III
0.643 (24.6)
0.479 (38.0)
0.309 (12.0)
0.094 (3.6)
0.558 (21.7)
2.574 (100)
-------
nomographs facilitate the determination of either one of two source
criteria as a function of weather conditions, (1) the maximum source
emission rate for which the ground level concentration will not exceed
a given maximum permissible value, or (2) the maximum ground level concen-
tration resulting from given operation conditions (emission rates) of the
source.
The design of the nomographs is based on the mathematical treatment
1 A
of atmospheric dispersion estimates presented by D. Bruce Turner. In
addition to five weather categories the following input variables are
provided: ambient air temperature and windspeed, stack gas temperature
and exit velocity, stack diameter and height. The application of the
nomographs is demonstrated by several examples.
45
-------
SECTION IV
PERMISSIBLE DRINKING WATER CONCENTRATIONS
A. Table of Symbols and Definitions
o fraction of pollutant absorbed through respiratory system
3 fraction of pollutant absorbed through gastrointestinal
tract
c maximum stationary body concentration (mg/kg)
k first order rate constant (1/hr)
LD5Q lethal dose 50% kill (mg/kg)
(LD5Q)1 lethal dose 50% kill of ith pollutant (mg/kg)
R take rate of pollutant uptake (mg/kg/day)
rbt rabbit
T biological half-life of pollutant (days)
o
TLV threshold limit value (mg/m )
(TLV)lQw lower 95% confidence limit for TLV estimated from LD5Q
(mg/m3)
x actual pollutant drinking water cone, (mg/1)
xi actual drinking water cone, of ith pollutant (mg/1)
xfi estimated safe drinking water cone, (mg/1)
(xe)1 estimated safe drinking water cone, of ith pollutant (mg/1)
V volume of drinking water ingested (I/day)
W body weight (kg)
46
-------
B. Caveats
The treatment described in this section was based in part, on the
following assumptions and limitations.
a) The proposed method of estimating permissible drinking water
concentrations is not applicable to pollutants with carcino-
genic, teratogenic or mutagenic effects. A treatment of car-
cinogens in ambient air is presented in Section V.
b) This method does not take into account other sources of environ-
mental contamination. No allowance has been made for pollutant
uptake in air or diet.
c) No allowance was made for synergistic or antagonistic effects
of pollutants in multicomponent systems.
d) No allowance was made for metabolic inhibition or induction
effects on pollutants in the body.
e) The treatment was based on the assumption that animal ID™
values were applicable to man.
f) The biological treatment was based on a simplified one com-
partment model with first order kinetics.
g) No allowance was made for factors such as odor and taste which
may necessarily lower the standard below the maximum safe con-
centration limit.
C. Derivation of EPA Proposed Drinking Water Standards
Pesticides
The proposed drinking water standards for pesticides were established
on the basis of animal feeding experiments. The minimal or no-effect
dose fed to the more susceptible specie was used to calculate a safe body
burden (mg/kg). The proposed safe concentration limit was determined by
applying a appropriate safety factor to the animal data. Table 11 sum-
marizes the derivation of each pollutant standard. A constant safety
factor of 2500 was applied to all the compounds, except for methoxychlor
for which comparable human toxicity data was available. An additional
safety factor was imposed on chlordane and toxaphene because of organo-
leptic effects and on heptachlor because of its presence in the diet.
Inorganics
The derivation of EPA proposed drinking water standards- for inor-
ganic chemicals was based mainly on observed human health effects. There
47
-------
Table 11. DERIVATION OF EPA PROPOSED DRINKING WATER STANDARDS FOR PESTICIDES
(1)
Animal safe
(2)
Equivalent level
for 70 kg man
(3)
Equivalent derived.
(4)
Proposed EPA
(5)
Safety factor,
Contaminant
Chlordane
Endrin
Heptachlor
Lindane
oo Methoxychlor
Toxaphene
2,4-D
2,4,5-TP (Silvex)
level (mg/kg/day)
.42
.02
.08
.3
17.
1.7
8.0
.9
(me/day)
29.4
1.4
5.6
21.
1190.
119.
560.
63.
water St'd (mg/1)
14.7
.7
2.8
10.5
59.5
59.5
280. v
31.5
water St'd (mg/1)
.003
.0002
.0001
.004
.1
.005
.1
.01
ratio (3)/(4)
4900
3500
2800
2625
5950
11900
2800
3150
Assume consumption of 21/day.
-------
has been no report of a consistent animal feeding experiment comparable
to the pesticide studies. Table 12 indicates the more pertinent toxicity
information for each contaminant and a 21/day water concentration corre-
sponding to the given dosage. The EPA proposed drinking water limit is
given and the factor of safety separating it from the derived value. It
is significant that the safety factors obtained in this manner are con-
siderably lower than those associated with the pesticide drinking water
limits.
D. Stokinger-Woodward Treatment - METHOD I
Iculating safe drinking v
.3
o
Stokinger and Woodward used TLV in calculating safe drinking water
standards. The assumption was made that an individual breathes 10 nf
of air during an 8 hour working day and that the total pollutant uptake
from air at the TLV concentration can be safely tolerated. It was then
assumed that this safe quantity of pollutant/day can be similarly tol-
erated in drinking water. Consumption of two liters of drinking water/day
was assumed. Correction for the efficiency of respiratory and gastroin-
testinal absorption for the specific compounds were included in this treat-
ment. The corresponding safe drinking water concentration was calculated
as shown below.
x = 10 « (TLV) (40)
e 2@
where
x = calculated safe drinking water concentration
a = fraction of pollutant absorbed through respiratory system
g = fraction of pollutant absorbed through gastrointestinal tract.
In general, the calculated values (xe) for inorganics are 1-3 orders
of magnitude higher than the limits proposed by EPA (see Table 13).
Based on this Method I, the criterion used to assess a pollutant
hazard would be expressed as
> 1.0 (41)
5a (TLV) -
x = actual pollutant drinking water concentration (mg/1).
If, for any pollutant, this quotient is equal to or greater than
unity a hazard is judged to exist.
49
-------
Table 12. DERIVATION OF EPA PROPOSED DRINKING WATER STANDARDS FOR INORGANIC CHEMICALS
Derived Water Proposed Wa£er Safety
en
O
Contaminant
Ag
As
Ba
Cd
CN
Cr
F
Hg
N03(as N)
Pb
Se
Toxicity Data
60yg/day-human intake 1940
12mg/l-human skin cancer
.5mg/m -TLV
352yg/day-human max. no-effect
4 . 7mg/l-human no effect
. 45mg/l -human no effect (3 yr)
8mg/l-human fluorosis
,3mg/day -human lowest toxic
10mg/l-adult no effect
600jjg/l -human no effect
. 7mg/day-human toxic
Cone, (mg/1)
.03
12.
2.
.18
4.7
.45
8. .
.15
10.
.3
.35
St'd (me/1)
.05
.05;
1.
.01
.2
.05
1.4-2.4
.002
10.
.05
.01
Factor
-
240
2
18
24
9
4-6
75
-
6
35
Assume consumption of 21/day
**
Based on MeHg
-------
Table 13. CALCULATED WATER STANDARDS - INORGANIC
Contaminant
As
Ba
Cd
CN
Cr
F
P6
Se
*
Data taken
e ~
TLV, ppm
0.5
0.5
0.1
0.1
5.0
2.5
0.2
0.1
from Reference (3)
10a (TLV)
2S
Method
a
0.20
0.75
0.25
0.60
0.75
0.10
0.20
0.80
*
6
0.80
1.00
0.03
0.80
0.06
1.0
0.20
0.80
x^ fm?/B
0.6
2.0
4.0
19.0
6.0
1.25
1.0
0.5
Proposed St'd (mg/1)
.05
1.0
.01
.02
.05
1.4-2.4
.05
.01
x = calculated safe drinking water concentration
e
a = fraction of pollutant absorbed through respiratory system
6 - fraction of pollutant absorbed through gastrointestinal tract.
TLV = threshold limiting value.
51
-------
E. Estimation of Permissible Drinking Water Standars - METHOD II
Introduction
There are two common measures of a compound's toxicity and its
effect on human health. The threshold limiting value (TLV) for several
industrial compounds have been set by the American Conference of Govern-
mental Industrial Hygienists.2 Although these values are based in part
on animal toxicity experiment, they have been extrapolated to define
pollutant air concentrations safely tolerated by humans during a working
day. Unfortunately, the TLV for only a comparatively few substances have
been established.
The second measure is animal toxicity data. Several different
toxicity parameters have been determined and reported in the literature.
However, the most common of these is the oral LD50 value in the rat.
Animal LD5Q data, at best, reflects a relative measure of toxic potential
in man. The most serious problem is the variability among different
species and the relationship of the toxicity of a particular specie to
man. In the present context, the value of LDgQ data lies in the fact
that the mode of administration, absorption into the body and routes of
metabolism and excretion, are comparable to the ingestion of drinking
water.
Thus, oral LD™ is important because the phenomena in operation
parallel human drinking water consumption, although such data may corre-
late poorly with toxicology in man. On the other hand, the limited TLV
information is directly applicable to the effects of air pollutant con-
centrations on human health, but is not related in any way to the chemical
form which compounds are likely to exist in drinking water nor the gastro-
intestinal absorption behavior of these compounds. In addition, organic
compounds contained in drinking water and absorbed through the intestines
may result in a lower body burden compared to the same amount taken into
the body through the respiratory system. This is due to the first pass
liver effect, which action may detoxify some portion of an oral dose be-
fore it is delivered to the rest of the body.
Application of (TLV)lQW Concept in Estimating Drinking Water Standards
Ideally, the best of both parameters should be combined to give
an indicator which could be directly related to absolute toxic or no-
effect levels in man. An approximation to this end was developed in
52
-------
Section III of this report. TLV and the LD5Q values for 241 different
organic and inorganic compounds were plotted on a log-log scale. TLV
data were obtained from the American Conference of Governmental Industrial
Hygienists 1974 Adopted Values and the LD5Q values were taken, when avail-
able, from oral rat experiments. A regression analysis of these data
yielded the following relationship:
TLV = 0.029 (LD50) (28)
The lower 95% confidence limit of this regression line was derived
and shown to be described by the following expression:
(TLV)lQW = 4.5 x lO'4 (LD5Q) (29)
The treatment described below is a modification of the approach
3
used by Stokinger and Woodward. The major difference is the use of the
(TLV)1 value of the pollutant in place of the reported TLV.
By definition, the (TLV), of a pollutant is that air concentration
to which a normal individual may be safely exposed to any substance for
an 8 hour working day at the 95% confidence level. In addition to using
this basic parameter, the following assumptions are made:
a) Ten cubic meters of air are breathed/working day and all
pollutants are completely absorbed in the body.
b) The total amount of pollutant in 10 cubic meters of (TLV)low
air can be safely tolerated by man.
c) This total amount would also be safely tolerated daily if
contained in drinking water.
d) Two liters of drinking water are consumed per day and all
of the soluble contaminants are completely absorbed into the
body.
Thus, ten times the number of mg contained in one m of (TLV) .
may be safely distributed in 2 liters of drinking water. The estimate
of safe pollutant concentration (xg) may be expressed as follows:
10 m3 (TLV), mg/m3
xe = -- 2T- - (42)
" 5 (TLV)low mg/1
53
-------
Since, (TLV)low = 4'5 x 10~4 LD50 (43)
xe = 2.25 x 10"3 LD5Q
This expression was used directly to generate derived estimated data
for selected inorganic cations and anions. The calculated values were
compared to the EPA proposed values and showed generally good agreement
(Tables 14 and 15). The pesticides for which values have been proposed
were treated similarly and the method was not found to be applicable.
The derived values were 100 times higher than EPA proposed values
(Table 16).
An alternative form of this equation is available for use when evalu-
ating a pollutant with an established TLV and no LD5Q data. .It was shown
that
LD5Q = 34.5 TLV. (33)
Substitution then yields the desired expression:
. xe = 7.76 x 10"2 TLV . (44)
There are three practical advantages in utilizing the (TLV),
LDcQ concept over the Stokinger-Woodward TLV approach.
(1) LDcQ have been determined for a much larger number of
compounds and hence, the (TLV)lQw treatment is applicable
to a greater number of pollutants.
(2) Knowledge of the respiratory and gastrointestinal absorption
factors for specific compounds, which Stokinger and Woodward
used in their treatment, is not required.
(3) The (TLV)lQW has a built-in safety factor making additional
corrections unnecessary.
Based on Method II, the criterion used to assess a pollutant hazard
would be expressed as
5 - >1.0 (45;46)
"2 ~
2.25 x 10 LD5Q 7.76 x 10" TLV
x = actual pollutant drinking water concentration (mg/1).
If, for any pollutant, this quotient is equal to or greater than
unit a hazard is judged to exist.
54
-------
Table 14. ESTIMATED WATER STANDARDS - INORGANIC
Method II
Contaminant Proposed St'd (mg/1)
Ag 0.05
As 0.05
Ba 1.0
Cd 0.010
CN 0.02
Cr 0.05
P 1.4-2.4
Hg 0.002
N03(as N) 10.0
Pb 0.05
Se 0.01
*e = 2.25 x 10"3 (Ionic LD5Q)
10(TLV),
lOW - /TTTTX
xe~ 2 5 (TLV)low
(TLV)low = 4.5 x lO'4 (LD5Q)
x = 2.25 x 10~3 (LDcn)
Ionic LDPn (mg/kg)
32
5
99
54
3.4
28
80
27
231
169
3
x (mg/l)
0.072
0.011
0.223
0.122
0.008
0.063
0.180
0.061
0.520
0.380
0.007
55
-------
Table 15. COMPARISON OF PROPOSED AND ESTIMATED STANDARDS - INORGANIC
Method II
Proposed
Contaminant
Ag
As
Ba
Cd
CN
Cr
F
Hg
N03 (as N)
Pb
Se
Standard (mg/1)
.05
.05
1.
.01
.20
.05
1.4-2.4
.002
10.
.05
.01
X
c*
.072
.011
.223
.122
.008
.0063
.180
.061
.520
.380
.007
x /Proposed Standard
1.44
.22
.22
12.2
.04
1.26
.10
30.5
.05
7.6
.7
56
-------
Table 16. ESTIMATED STANDARDS - PESTICIDES
Method II
Proposed
Contaminant
Chlordane
Endrin
Heptachlor
Lindane
Methoxychlor
Toxaphene
2,4-D
2,4,5-TP (Silvex)
*x = 2.25 x 10"3 LD._
e 50
10 (TLV)-
v - low - '
LD--
283
3
40
76
5000
60
375
650
=; fn.v}
x (mg/ml)
.64
.007
.09
.17
11.25
.14
.84
1.5
St'd (mg/ml)
.003
.0002
.0001
.004
.10
.005
.10
.01
low
-4
(TLV)., = 4.5 x 10 LD
50
2.25 x 10"3 LD
5Q
57
-------
Selection of Appropriate Toxicity Data
A workable pollution criteria statement should involve the use of
toxicity and pharmacological data already compiled. Information such
as maximum no-effect body burden levels, percent gastrointestinal
absorption, biological or elimination half-lives, although extremely
pertinent, would normally not be known. Whenever possible, oral rat
LD50 data was used in this treatment. Although in some cases the proposed
EPA drinking water standards were estimated using other species, the
more available oral rat information was used to provide greater scope of
application.
1. Inorganic Chemicals
Any metal in drinking water must be present as a stable soluble
compound. Selection of an appropriate LD5Q value for a particular
metal is by no means a straightforward task. Different compounds of
13
the same metal have widely different oral rat ID™ values. In this
treatment, the following criteria were used.
(1) The compound must be stable and be at least slightly soluble
in water.
(2) Since LD5Q data is available only for molecular species, the
reported toxicity represents a contribution of both anion and
cation moieties. In an attempt to relate molecular toxicity
to the ion under study, the toxicity of the associated ion
must be minimal. Cation LD5Q data was obtained from toxicity
data of metal salts containing relatively inocuous anions, e.g.,
chlorides, nitrates, oxides. In a similar way anion LD5Q data
was obtained from toxicity data of salts of relatively inocuous
metals, e.g., sodium potassium.
(3) The most toxic (lowest LD5Q - preferably oral rat) compound
meeting the above criteria was selected for use.
Since LD5Q values are expressed in terms of weight of compound under
test, the cation (or anion) content was calculated and the LD5n reported
as mg of cation or anion/kg. This value is defined as ionic ID™ (Table 17).
DU
Ionic LDrn = (Compound LD,-n) x (Percent Content of Metal or Anion) (47)
ou °u TOO
58
-------
Table 17. TOXICITY DATA - INORGANICS
Contaminant
Ag
As
Ba
Cd
CN
Cr
F
Hg
N03(as N)
Pb
Se
Compound - LD,.0(mg/kg)
jO r
AgNO, - 50 (oral, mouse)
As20_ - 8 (oral, rat)
BaCl^ - 150 (oral, rat)
CdCl2 - 88 (oral, rat)
NaCN - 6.4 (oral, rat)
Na0Cr-0, - 140 (TDL -1m, rat)
227 o
KF - 245 (oral, rat)
Hgd2 - 37 (oral, rat)
Ca(N03)2 - 3900 (oral, rat)
Pb(N03)2 - 270 (LDLQ, ip, rat)
Na^SeO^ - 7 (oral, rat)
Ionic LDc/,(mg/kg)
32
5
99
54
3.4
28
80
27
231
169
3
59
-------
2. Pesticides
All of the pesticides treated in this analysis had oral rat LDgg
values reported in the literature.
F. Estimation of Permissible Drinking Water Standards - Method III
Application of Kinetic Method
In Section III of this report an air quality criteria based on a one
compartment biological model was developed. The following derivation is
based on similar considerations.
The rate of pollutant uptake per unit body mass per day is given by
"uptake ' »*/" »»
where
V = volume drinking water ingested/day (liters)
3 = fraction of ingested compound absorbed through gastrointestinal
tract
x = drinking water concentration of pollutant (mg/1)
W = body weight (kg)
If V = 2, 6 = 1, W = 70 then the expression reduces to the following
form:
Ruptake = 2x/7° = -029x <48)
The maximum daily uptake of a pollutant by a 70 kg man drinking 2
liters of water/day is .029 times the pollutant water concentration.
This amount would accumulate indefinitely unless the compound was meta-
bolized or excreted.
Assuming first order metabolism and excretion processes from this
one compartment model, a maximum body concentration (cm) of the pollutant
would be given by
(49)
where k = first order elimination rate constant (hr~^).
The rate constant k is related to the biological half -life T.
k = ln2/T. - (50)
Substituting,
60
-------
(51)
Assuming a 70 kg body weight and a drinking water volume of 21, the
expression simplifies to
cm = .042XT. (52)
If the safe maximum body concentration and the biological half-life
of a pollutant are known, the safe drinking water concentration is readily
calculated. As shown in Section III, a survey of compounds with known
half-life and LDgg values indicated that a safe maximum body concentration
may be estimated as 0.05% of the LDcg.
cm = .0005 LD50 -(53)
therefore,
.012 LD,n
xe = T 50 (54)
Unfortunately, the biological half-life for most compounds is not
known. To apply Method III in these cases, a half -life must be assumed.
Derivation of Method III. A, B. C
A list of heavy industrial chemicals was compiled in Section III
(Table 3). With the exception of dinitro-o-cresol (T = 5.78 days), the
half-life of these compounds is less than one day. Based on this data,
a 30-day half-life was assumed for compounds where this para-
meter was not known. This provided an additional safety factor for sub-
stances with half-life periods less than 30 days. Compounds retained
in the body for prolonged periods of time cannot be treated in this manner;
a longer assumed half-life must be incorporated into the Method III cal-
culation. An assumed value of 365 days (1 yr) is reasonable in such cases.
Thus, three different T values may be used in the basic equation
(the known value or the assumed 30 day or 1 year value). The following
listing gives the alternative forms of Method III estimation and their
area of application.
Method III A (known half-life)
.012 LDKn
xe . —^ (54)
61
-------
Method III B (assumed 30 day half-life)
xe = 4.0 x 104 LD5Q (55)
Method III C (assumed 1 year half-life)
xe = 3.3 x 10"5 LD5Q (56)
In cases where the TLV is known and the LD5Q is not available, the
above equations take the following form
LD5Q =34.5 TLV (33)
Method III A xe = -414 TLV (57)
1 Method III B xfi = 1.38 x 10"2 TLV - (58)
Method III C xg = 1.14 x 10"3 TLV (59)
These equations expressed as assessment criteria for pollutant hazard
are shown below.
Method III A tQ1*TLD >.1.0; ^ TLV >.1.0 (60;61)
DU
Method III B —j ^1.0; 5 > 1.0 (62;63)
4.0 x 10"4 LD5Q 1.38 x 10"^ TLV ~
Method III C —» >.1.0; ^ * > 1.0 (64;65)
3.3 x 10"° LD5Q 1.14 x 10"J TLV ~
x = actual pollutant water concentration (mg/1)
T = biological half-life (days)
LD50 = animal (oral rat) LD5Q (mg/kg)
If, for any pollutant these quotients are equal to or greater than
unit a hazard is judged to exist.
The recommended treatment for compounds with a known half-life is
Method III A. However, the value of parameter will not be available for
most pollutants. In these cases, Method III B should be used except for
those compounds with known prolonged body retention times for which Method
III C is appropriate.
62
-------
A 30-day half-life was assumed for the group of inorganics and the
corresponding estimated water standards were calculated (Table 18)
(Method III B). Only the derived standard water value for mercury was
appreciably higher (x 5.5) than the proposed limit. However, the pro-
posed value was based on the reported toxicity of methylmercury which
has a half-life of 70 days. If this value is used in the original ex-
pression, the estimated water concentration limit is .0046 mg/1 (pro-
posed, .002 mg/1).
Chlorinated pesticides and other lipid soluble organic compounds
are stored in human adipose tissue for prolonged periods of time. Pollu-
tion criteria should be assessed with this is mind. Thus, the longer
half-life assumption is applicable in these cases (Method III C). Table
19 lists the estimated and proposed water standards for a series of
pesticides. Only two compounds, chlordane and heptachlor, gave estimated
values appreciably higher (x9-13) than those proposed. It is signifi-
cant that an additional safety factor was applied to the proposed stan-
dards of these substances because of organoleptic effects or presence in
the diet.
G. Estimation of Permissible Drinking Water Standards for Multicomponent
Systems
When two or more pollutants are present, the effect of each should be
considered additive and their combined action should be given primary con-
si dei
low.
2
sideration. This treatment is recommended by the ACGIH and is shown be-
n xi
ET3g^->1.0 (66)
x. = actual drinking water concentration of i pollutant
(mg/1)
x = estimated drinking water concentration of i pollutant
(mg/1)
If the above summation equals or exceeds unity a hazard is judged
to exist.
This expression can be evaluated in terms of the general estimation
methods previously discussed by appropriate substitution of xe- The
water quality standards according to these methods are given below.
63
-------
Table 18. ESTIMATED WATER STANDARDS USING ASSUMED
HALF-LIFE - INORGANICS
Method IIIB
Contaminant
Ag
As
Ba
Cd
CN
Cr
F
Hg
N03(as N)
Pb
Se
.0005 x .693
e .0286 x t
.012 (LD50)
6 T
Ionic LD,.,,
DU
32
5
99
54
3.4
28
80
27
231
169
2
(LD5Q)
x ( T 30 davsHme/l")
.013
.002
.040
.022
.001
.011
.032
.011
.092
.068
.001
Proposed
St'd (mg/1)
.05
.05
1.0
.01
.20.
.05
1.4-2.4
.002
10.0
.05
.01
(T 30 days) = 4 x 10~4 (LD5Q)
64
-------
Table 19. ESTIMATED WATER STANDARDS USING ASSUMED
HALF-LIFE - PESTICIDES
Method IIIC
Contaminant
Chlordane
Endrin
Heptachlor
Lindane
Methoxychlor
Toxaphene
2,4-D
2,4,5-TP (Silvex)
.0005 x .693 /TT, x
Xe ~ .0286 XT (LD50}
.012 (LD-Q)
e T
LD,
283
3
40
76
5000
60
375
650
x f T 365 davsHae/1)
.0093
.0001
.0013
.0025
.1650
.0020
.0124
.0215
Proposed
St'd (mg/1)
.003
.0002
.0001
.004
.10
.005
.10
.01
xe (r 365 days) = 3.3 x 10~5 (LD5Q)
65
-------
n x-3.
Method I Z (5a TLV)" -1>0
n x.
Method II r —* 1 (68)
1 (2.25 x 10"J LD50).
n X.T.
Method III A z / n1J. I i > 1.0 (69)
•j ^-uu LU5Q'i "
n x.
Method III B £ -1—2 >.1.0 (70)
i (4.0 x 10'4 LD50).
n x.
Method III C £ i—g >.1.0 (71)
i (3.3 x 10'* LD5Q)i
These equations may be written in terms of TLV by substituting the
value 34.5 TLV for LD5Q as shown in previous sections.
Concentrations of selected pesticide and inorganic pollutants were
taken from the literature and a water quality standard calculated
using the above equations. The pollutant concentration represents a
purely hypothetical situation and does not refer to the analysis at a
particular river site. Moreover, the medium being evaluated in this
example is river water and not drinking water.
The inorganic profile (Table 20) showed a borderline water quality
standard (0.92) based on the EPA proposed standards and a distinctly
polluted situation using Method III B (T = 30 days) estimated standards.
It is significant that the greatest contribution to the inorganic river
contamination according to Method III B (97%) and the EPA standards (79%)
is due to the anions; cyanide, nitrate and fluoride.
A comparable pesticide profile is shown in Table 21. Water quality
standards were calculated based on EPA proposed standards and Method III C
(T = 1 year) estimated standards. The values calculated by both techniques
differed by only approximately 25% with the Method III C approach indicating
the greater pollution hazard.
H. Comparison of Methods I, II, and III
A group of heavy industrial chemicals is listed in Table 22 and
an estimated drinking water standard calculated by Methods I, II, III A
66
-------
Table 20. ANALYSIS OF A MULTICOMPONENT SYSTEM - INORGANICS (METHOD IIIB)
(SELECTED CONCENTRATIONS IN RIVER WATER)
Absolute (percent) Contribution
to Air Quality Standard
Contaminant
Ag
As
Ba
Cd
CN
Cr
F
N03
Pb
Proposed
St'd (mg/1)
.05
.05
1.0
.01
.05
.20
1.9 (average)
10.0
.05
Estimated
St'd (mg/1)
.013
.002
.04
.022
.001
.011
.032
.092
.068
Water
Typ. River Cone.
(mR/1)*
.001
ND
.04
ND
.015
.002
.70
.58
.006
Quality Standard
Proposed St'd
.02 (2.2)
.04 (4.4)
.30 (32.6)
.01 (1.1)
.37 (40.2)
.06 (6.5)
.12 (13.0)
.92 (100)
Method IIIB
.08 (.2)
1.0 (2.2)
15.0 (33.7)
.18 (.4)
21.9 (49.2)
6.3 (14.1)
.09 (.2)
44.6 (100)
Data taken from Reference 14.
Estimated St'd xfi
ND - not detected
4.0 x 10~4 LD
5Q
(T - 30 days assumed) - METHOD IIIB
n x
Air quality standard calculated by: Proposed St'd Z (EPA Proposed St'd)
n
x
Method IIIB
1 (4.0 x 10
-4
where x. = cone, (mg/1) of ith component
-------
Table 21. ANALYSIS OF A MULTICOMPONENT SYSTEM - PESTICIDES (METHOD IIIC)
(SELECTED CONCENTRATIONS IN RIVER WATER)
< Absolute (percent)
to Air Quality
Contribution
Standard
CO
Contaminant
Chlordane
Endrin
Heptachlor
Lindane
Toxaphene
2,4-D
2,4,5-TP (Silvex)
Data taken from
Proposed
St'd (me/1)
.003
.0002
.0001
.004
.005
.10
.01
Reference 14.
Estimated St'd, xe - 3.3 x 10~5
Air quality standard calculated
Estimated
(St'd (me/1)
.0093
.0001
.0013
.0025
.002
.0124
.0215
Water
LD5Q ( T= 365 day
by: Proposed St1
Maftirwl TTTP
Typ. River Cone.
(,,R/D
.02
.03
.02
.01
.60
.10
.05
Quality Standard
s assumed) - METHOD IIIC
n x.
IT*
* (EPA Proposed St'd)i
xi
Proposed St'd
.007 (1.4)
.15 (30.9)
.20 (41.2)
.003 (.5)
.12 (24.7)
.001 (.2)
.005 (1.0)
.485 (100)
Method IIIC
.002 (.4)
.30 (47.5)
.015 (2.4)
.004 (.6)
.30 (47.5)
.008 (1.3)
.002 (.3)
.631 (100)
1 (3.3 x 10
-5
th
where x. a cone (mg/1) of i component
-------
Table 22. COMPARISON OF METHODS I, II, IIIA, IIIB
Industrial Orsanies
CT>
Contaminant
Acrylonitrile
Aniline
Benzene
Carbon disulfide
Dinitro-o-cresol
Ethylenediamine
Formaldehyde
Methanol
Methyl chloride
Phenol
Pyrindine
Toluene
LD <„/,«)
->lr
93
440
3400
400
30
1160
800
420
1800
414
891
3000
t(days)
.120
.126
.047
5.78
.193
.012
.083
Method I
240
47
340
140
30
95
30
Method II
* t
.21
.99
7.65
.9
2.61
1.80
.95
4.05
.93
2.0
6.75
Method IIIA
44
324
102
___
26
1800
434
^ ~ f f
Method IIIB
.037
.176
1.36
.16
«.«._
.464
.320
.168
.72
.166
.356
1.20
2.25 x 10~3 LD
Method I: taken from Ref. 3.
Method II: xg
Method IIIA: x
Method IIIB: x
'50
.012 LD5Q/T
4.0 x 10~4 LD
50
-------
(known half-life) and III B (assume T = 30 days). Compounds with known
half-life by the recommended method, Method III A. Because of the rela-
tively short half-life of these compounds, the estimated standards are
appreciably higher than those found using Method II or Method III B. It
must be borne in mind that these values represent levels at which these
compounds can be safely tolerated in drinking water without regard to
factors such as odor, taste, carcinogenicity, tetratogenicity or muta-
genicity.
70
-------
SECTION V
CARCINOGENS IN AMBIENT AIR
A. Table of Symbols and Definitions
a age (years since birth)
B breathing rate (3700 m3/yr)
DJJ number of nonsmoker lung cancer deaths/year
C carcinogen intake (ug/yr)
M annual lung cancer mortality/100,000, based on Dorn study
MDN annual nonsmoker lung cancer mortality/100,000, based on
Dorn study (equal to M.,WM)
MN annual nonsmoker lung cancer mortality/100,000, based on
general population
MNB annual black nonsmoker Tung cancer mortality/100,000, based
on general population
MNW annual white nonsmoker lung cancer mortality/100,000, based
on general population
MNWp annual white female nonsmoker lung cancer mortality/100,000,
based on general population
MNWM annual white male nonsmoker lung cancer mortality/100,000,
based on general population (equal to MQ^)
N number of cigarettes smoked/day
ND cigarette nonsmoker
PM nonsmoking population
IN o
x carcinogen concentration in ambient air (ng/m ), annual mean
71
-------
B. Caveats
The reader should be aware of the following limitations of the
methodology used in this report.
(1) The term carcinogen refers only to agents responsible for the
initiation and development of lung and bronchial cancers.
(2) It is assumed that a major contributing factor in the inci-
dence of lung cancer death of cigarette smokers and nonsmokers
alike is the inhalation of certain carcinogenic hydrocarbons.
(3) The exposure level of eight (8) hydrocarbon carcinogens, identi-
fied and quantitated in cigarette smoke, is assumed to be a
major contributing factor in lung cancer incidence and a valid
measure of lung cancer mortality of smokers and nonsmokers.
(4) The 8-component hydrocarbon mixture, present in cigarette smoke,
possesses a total carcinogenic potency equal to a mixture of
the same materials in ambient air.
(5) A number of lung cancer deaths are due to carcinogens in the
ambient air and this is reflected in the lung cancer mortality
of nonsmokers.
(6) Any linear relationship between carcinogen intake by smokers
and the lung cancer death rate corresponding to these smokers
may be extrapolated to yield valid mortality data for low carcino-
gen intake values.
(7) Ambient air concentrations have been calculated assuming a total
breathing volume of 3700 m /year for the average adult.
C. Carcinogen Content in Cigarette Smoke and Intake by Smokers
Cigarette smoke has been analyzed for carcinogenic components by
several investigators. The amounts of eight (8) hydrocarbon carcinogens
present in cigarette smoke have been reported by Wynder and Hoffman and
are listed in Table 23 with a ranking of their relative carcinogenicity.4
Knowing the carcinogen content in cigarette smoke, a carcinogen
intake rate is readily determined for individuals smoking different number
of cigarettes/day for varying lengths of time.
r - 10.8 (N) 365
L " 100
C = 39.4 N
72
-------
Table 23. CARCINOGENIC HYDROCARBONS ISOLATED FROM CIGARETTE SMOKE
Hydrocarbon
B enzo-a-pyrene
Dibenz-a,h anthracene
Benzo-b-fluoranthene
Benzo-j-fluoranthene
Benz-a-anthracene
Chrysene
Benzo-e-pyrene
Indeno - l,2,3,c,d - pyrene
Relative f
carcinogenicity
**
Micrograms per
100 cigarettes
2.5
0.4
•H- 0.3
-H- 0.6
+ 0.3
+ 6.0
+ 0.3
+ 0.4
TOTAL yg/100 cigts. 10.8
**
Carcinogenicity as determined on mouse skin.
Isolated from cigarette smoke.
Table 24. CARCINOGEN INTAKE FOR SMOKERS AS A FUNCTION OF NUMBER OF
CIGARETTES SMOKER PER DAY
Cigarettes smoked/day (H)
5
15
30
50
Carcinogen intake from
cigarettes only, ug/yr (C)
197
591
1182
1970
73
-------
where C = carcinogen intake (yg/yr)
N = number of cigarettes smoked/day
Values of C for different groups of cigarette smokers is shown in
Table 24.
D. Lung Cancer Mortality as a Function of Individual Smoking Habits
and Age
Lung cancer death rates as a function of age and individual smoking
patterns have been reported in the Dorn Study of Smoking and Mortality
Among U.S. Veterans.5 This study was conducted over an eight one-half
year period with a population of over 293,000 military veterans holding
Government Life Insurance policies. The study group was composed of
practically all white males drawn from the middle and upper socioeconomic
classes.
The Dorn study gives the cause of death for individuals classified
into ten-year age brackets (45-54, 55-64, 65-74 and 75-84). In correlating
age and lung cancer mortality (see Figure 7) these groupings are identi-
fied by their median age (50,60, 70 and 80). Dorn classified smoking
patterns as never or occasionally smoked, 1-9, 10-20, 21-39 and more
than 39 cigarettes/day. These groupings are indicated in the present
report as follows: nonsmokers, 5, 15, 30, and 50 cigarettes/day.
The data used in this evaluation was taken from the Dorn study as
presented in a paper by Kahn on pages 30, 38, 40, 42, and 44 under the
cause of death listing of cancer of lung and brochus. . The death rate
in each category was determined by dividing the number of deaths by
person-years of observation and relating this value to a rate per
100,000.
Supplemental data was taken from a similar study conducted bv
16
Hammond. This study was carried out over a 40-year period and con-
sisted of over one million men and women. The study results were re-
ported separately by sex. The male cohort of nonsmokers under age 55
was much larger than the comparable Dorn group and, as a result, pro-
vided a more reliable estimate of lung cancer risk for that category.
The lung cancer mortality figures from both studies have dubious
significance for age groups under 45 years. The reported number of
74
-------
lOOOp
500-
Ul
-------
deaths in these study subgroups was too low (less than 5) to warrant use
in this report. Any age-smoking mortality rates derived from less than
5 reported deaths were considered invalid. These values were established
by alternative means as described below. Hammond set a similar criterion
in evaluating his study results.
The annual death rates used in this treatment for the different
age-smoking categories is shown in Table 25.
E. Relationship of Lung Cancer Mortality and Age/Duration of Exposure
It has been shown that the incidence rate of many different epi-
thelial cancers show a linear relationship with age when both parameters
are plotted on logarithmic scales. Doll demonstrated this behavior by
showing a linear relationship of this type between the incidence of
bronchial carcinoma in cigarette smokers and nonsmokers with age.
A similar treatment was performed on the data obtained from the
Dorn study (see Figure 7). The plots show a closely linear lung cancer
death rate and age relationship.
Extrapolation of the 5, 15 and 50 cigarettes/day lines to age 50
(45-54 years old) yielded estimated smoker mortality data for this age
group. These risk values were used in subsequent steps of this evalu-
ation and are given in Table 26.
Dorn did not report any nonsmoker lung cancer deaths between the
ages of 35 and 54. The total number of person-years of observation for
subjects between the ages of 35 and 55 in the Hammond study was nearly
4 times that used by Dorn. Hammond's data was examined and used to
derive a more valid death rate for these younger nonsmokers. Hammond
reported no male nonsmoker lung cancer deaths in the 35-39 age group and
only two in the 40-44 year old category. Nine lung cancer deaths were
reported for male nonsmokers, 45 to 54 years old. The average annual
mortality risk over this ten-year period was shown to be 6.4. Graphic
representation of three Dorn and one Hammond nonsmoker points on a risk
versus age log-log plot was reasonably linear (Figure 8). The nonsmoker
lung cancer risk-age relationship may be expressed by the following
equation:
M = 1.7 x ID'7 a4'42
76
-------
Table 25. LUNG CANCER MORTALITY AS A FUNCTION OF SMOKING PATTERN
AND AGE (FROM DORN STUDY UNLESS INDICATED OTHERWISE)
Cigarettes Lung cancer Person-years Lung cancer
Age smoked/day deaths of observation death rate/100,000
45-54 NS
5
15
30
50
55-64 NS
5
15
30
50
65-74 NS
5
15
30
50
75-84 NS
5
15
30
50
1
(1)
(9)
10
(3)
25
31
183
245
63
49
44
239
194
50
4
5
15
7
(2)
15,134
3,129
16,392
12,839
1,928
213,858
45,217
151,664
103,020
19,649
171,211
37,130
101,731
50,045
8,937
8,489
1,923
3,867
1,273
232
6.4*
28.0
57.0
77.9
160.0
11.7
68.6
120.7
237.8
320.6
28.6
118.5
234.9
387.7
559.9
47.1
260.0
387.9
549.9
890.0
From Hammond Study (ref. 16).
Deaths in parenthesis were calculated from a death rate obtained by
extrapolation (Fig. 7).
77
-------
Table 26. ESTIMATED LUNG CANCER MORTALITY OF 45-54
YEAR OLD CIGARETTE SMOKERS
Cigarettes
smoked/day
5
15
30
50
Lung cancer
mortality/100.000
28. Oa
57.Oa
77.9
160.Oa
Extrapolated values from Figure 7.
Table 27- CARCINOGEN INTAKE AND EQUIVALENT AIR CONCENTRATION
FOR NONSMOKERS
Carcinogen intake Equiv. carcinogen
Equiv. number
of cigarettes
Age
45-54
55-64
65-74
75-84
(yg/yr)
4.7
23.0
16.0
30.0
air cone, (ng/m3)
1.3
6.2
4.3
8.1
smoked/day
.12
.59
.41
.77
Table 28. CARCINOGEN CONCENTRATION IN AMBIENT AIR EQUIVALENCE
TO CARCINOGEN INTAKE
Carcinogen Carcinogen concentration Equiv. number
intake (yg/yr) in ambient air (ng/m ) of cigarettes/day
197
591
1182
1970
53.2
159.7
319.5
532.4
5
15
30
50
78
-------
50
Source
D - Dorn (ref. 5)
H - Hammond (ref. 16)
8
o
(Jit
*
cr
I
cr
ui
o
o
=3
z
z
to
OD
50
60
AGE .YEARS
80
100
Figure 8. Nonsmoker lung cancer maortality versus age.
79
-------
where M = annual nonsmoker lung cancer mortality/100,000, based on
Dorn and Hammond studies, age 45-84.
a = age (years since birth)
The smoker lines had similar slopes, ranging from 3.7 to 4.7.
A similar British study in which 34,000 male doctors were observed
over 17 years showed that smoker incidence of bronchial cancer varied
with the seven and one-half power of the age. However, when Doll
plotted the same incidence data against nonsmoker age and duration of
smoking for cigarette users, he found that both groups were represented
by lines with slopes between 4 and 5.
When the Dorn mortality data was plotted in a similar fashion
(Figure 9), the slopes of the smoker lines were lowered to 2.5-3.2
(cigarette smoking was assumed to start at age 20). Unlike the inci-
dence rate data reported by Doll, Dorn mortality figures for both non-
smokers and smokers varied uniformly with age and not duration of ex-
posure.
F. Relationship of Lung Cancer Mortality and Carcinogen Concentration
in Ambient Air
Since the lung cancer death rates of individuals exposed to different
carcinogen levels (i.e., nonsmokers and smokers) are related to age by
closely parallel lines (see Figure 7), it is assumed that the mortality -
cancer intake relationship, for all individuals of a given age group,
may be represented by a single line (see Caveat #6).
Carcinogen intake (ug/yr) for different groups of smokers has been
calculated and noted earlier in this report (see Table 2). These data
were graphed on a log-log plot against the lung cancer death rate for
each age group (Figure 10). This graph shows the same data plotted in
Figure 10 except the cigarettes smoked/day variable is replaced by the
corresponding carcinogen intake rate.
Figure 10 shows several interesting features:
(1) The points graphed for the 55-64 and 65-74 year old groups
lie on parallel straight lines.
(2) The data for the 75-84 year old group falls on a straight
line which has a somewhat flatter slope than the others.
80
-------
lOOOr
500
Cigarettes Smoked/Day
O - Nonsmoker
0-5 (1-9)
a - 15 (10-19)
Q-30 (20-
A - 50 (39+)
•39)
O
Q
•s.
LJ
i
1
o
3
o
o
I
_l
<
z
z
JL
_L
30 40 50 60
DURATION EXPOSURE .YEARS
70 80
100
Figure 9.
Lung cancer mortality versus duration of exposure and cigarette
use.
81
-------
00
ro
SMOKERS
NONSMOKERS
10 20 60 100 .200
CARCINOGEN, INTAKE,jug/yr
600
2000
Figure 10. Lung cancer mortality versus carcinogen intake rate.
-------
(3) Three of the four points representing the 45-54 age group
were obtained by extrapolation in Figure 7. The slope of
this line was somewhat steeper than the other.
The lines connecting the points representing different rates of
carcinogen uptake (i.e., number of cigarettes smoked/day) for each age
group were extrapolated to low carcinogen intake values. The known non-
smoker (NS) lung cancer mortality rates (Table 25) were positioned on the
appropriate age line. This treatment permits one to read an equivalent
carcinogen intake rate for nonsmokers from the X-axis in Figure 9.
These data is shown in Table 27.
w
The data shown 'in Figure 10 were replotted in a form where the
carcinogen intake rate was expressed as a carcinogen concentration in
ambient air (Figure 11). The breathing rate for all age groups was
o
rounded off to 3700 m /year (see Table 2). The carcinogen ambient air
concentration equivalent to carcinogen intake was calculated as follows:
x (ng/m3) = C (yg/yr) x 1Q3 (73)
B (rrr/yr)
where x = carcinogen concentration in ambient air (mg/m ), annual mean
C = carcinogen intake (yg/yr)
B = breathing rate (3700 m3/yr).
The results are presented in Table 28 and graphed in Figure 11 for
smokers of different ages.
The lung cancer death rate for nonsmokers was located on each
extrapolated age line in Figure 11. An equivalent air carcinogen concen-
tration to which nonsmokers of different ages were exposed was read off
the X-axis and given in Table 27.
The equations which express the linear lung cancer mortality-
ambient air concentration relationships shown in Figure 11 have been
determined as follows:
Age
75_84 M = 42.5 x -419 (74)
65-74
55.54
45.54
83
M = 8.2 x '675 (75)
M = 4.5 x *675 (76)
M=1.3x'727 (77)
-------
CD
1000
800
600
§400|-
..300
o
Q200
UJ
a: 100
x 80
-------
where M = annual lung cancer death rate/100,000
x = carcinogen air concentration (ng/m3)
The effect of lowering the carcinogen concentration in ambient air
on lung cancer mortality risk for different age groups is shown in Table 29.
All the Dorn study subjects from age 45 to 84 were combined and an
overall death rate calculated for each smoking category (Table 30). The
lung cancer mortality versus carcinogen air concentration plot for this
composite age group is shown in Figure 12. The line best representing
these data points is given by the following equation:
M = 6.7 x '657 w (78)
The overall Dorn nonsmoker mortality (MQN) is 19.1/100,000 (Table 30)
and the carcinogen air concentration (x) corresponding to this rate is
4.9 ng/m (equivalent to smoking one^half cigarette/day).
G. Lung Cancer Mortality Estimates for the General Population
An expression relating lung cancer mortality (MN) and carcinogen
air concentration (x) was derived for 45-84 year old individuals in the
general population. Since the Dorn study was used as a model in this
treatment, the general population data were fit to an equation of the
Y
same form, M.. = Ax .
The exponent Y is a measure of the relative effect of variations in
ambient air concentration on lung cancer mortality. It was assumed that
all segments of the population over 45 years old possessed a carcinogen
intake - lung cancer mortality correspondence represented by lines parallel
to those shown in Figures 9 and 10. Thus, the exponent or slope, Y, was
assigned the same value as the age 45-84 Dorn group (Y = .657).
The value of A (y-intercept at x = 1) is a measure of the suscepti-
bility to lung cancer death of the population under examination. This
constant was readily determined from the nonsmoker lung cancer death
rate, the ambient air carcinogen concentration and the slope (Y).
A = MN/xY. (79)
The Dorn study group consisted of adult white males. As a result,
data derived from this study cannot be directly applied to any geographi-
cal segment of the general population. Females show lower lung cancer
mortality and blacks of both sexes exhibit higher rates. The Dorn non-
smoking death rate (19.1/100,000 at x = 4.9 ng/m ) was adjusted for an
85
-------
Table 29. ESTIMATED NONSMOKER LUNG CANCER DEATH RATE
AS A FUNCTION OF CARCINOGEN CONCENTRATION IN
Nonsmoker
lung cancer
death rate./ ambient air
AMBIENT AIR
Estimated
Carcinogen nonsmoker
lung cancer
death rate/
cone, in
% Decrease
Age 100,000
75-84 47.1
65-74 28.6
55-64 11.7
45-54 6.4
(ng/mj)
1.0
0.5
0.1
1.0
0.5
0.1
1.0
0.5
0.1
1.0
0.5
0.1
100,000
42.5
31.8
16.2
8.2
5.1
1.7
4.5
2.8
1.0
1.3
.79
.24
in death rate
9.8
32.5
65.6
71.3
82.2
94.1
61.5
76.1
91.5
79.7
87.5
96.3
See H. A. Kahn (ref. 5) and E. C. Hammond (ref. 16).
86
-------
Table 30. LUNG CANCER MORTALITY VERSUS CARCINOGEN INTAKE FOR DORN STUDY GROUP AGE 45-84
Smoker
Lung cancer
Cigarettes
smoked /day
NS
5
15
30
50
carcinogen
intake (yg/yr)
(17.9)
197
591
1182
1970
Equivalent carcinogen
air concentration (ng/m^)
(4.9)
54
157
319
532
Total lung
cancer deaths
78
81
446
i
456
118
Total person-
years
408,692
87,399
273,554
167,177
30,746
death
rate/100,000
19.1
92.7
163.0
272.8
383.8
Values in parenthesis were derived graphically.
-------
00
oo
1000
800
600-
8400-
Z
200
100-
80-
60-
LU
cc 40
UJ
o 30
z
6 20
10
8
6
4-
3-
2-
SMOKER
NONSMOKER
J L
AMBIENT AIR
' ' t I l—L_L
EQUIV. TO 5
>_J I I LJ—I L
15
30 50 CIGARETTES/DAY
J—U—L_L
.2 .3 4 .6 .8 I
2 3 4 6 8D 20 3040 6080 200 400600 1000
CARCINOGEN AIR CONCENTRATION (ANNUAL MEAN),ng/m3
Figure 12. Lung cancer mortality versus carcinogen air concentration - Dorn Study
-------
equa number of females in the population. Ha^ond has presented data
which indicates that the nonsmoker lung cancer mortality rate is l 5 times
higher for men than women. Blacks make up approximately 15% of the
population and exhibit an approximately 20% higher lung cancer death rate
than whites. Adjustment for these factors gives a nonsmoking lung
cancer mortality rate (MN) of 16.6.
MDN = MNWM = 19J (80)
MNWM * 1'5 MNWF; MNWF = 12<7 (81)
M - MNWM * MNMF
MNW 2 (82)
MNW = 15'9 (83)
MN = .85 MNW + .15 MNB (84)
MNB = K2 MNW; MNB = 19J (85)
MN = 16.4 (86)
where MNWM = annual white male nonsmoker lung cancer mortality/100,000
(equal to MQN).
^NWF = annua^ white female nonsmoker lung cancer mortality/100,000,
MNW = annual white nonsmoker lung cancer mortality/100,000,
NLg = annual black nonsmoker lung cancer mortality/100,000,
MN = annual nonsmoker lung cancer mortality/100,000,
Since this derivation was based on Dorn study nonsmokers, the car-
cinogen air concentration (x) associated with MN is 4.9 ng/m .
The constant A was calculated and the desired lung cancer mortality
relationship, shown below, is graphed in Figure 13.
A = MN/xY; A = 5.77 (79)
MN = 5.77 x '657. (87)
This expression has practical application only for nonsmokers.
Since cigarette smokers were found to inhale 10 to 100 times more
carcinogens than nonsmokers, modest changes in carcinogen air levels
would not appreciably effect their lung cancer mortality risk. Moreover,
moderate changes in cigarette use patterns would probably correspond to a
greater change in carcinogen intake than variations in ambient air quality.
89
-------
1000
800-
600
0400
Q300
p
200
UJ
LJ
O
100
60
If
o 20
10
8
6
4
3
AMBIENT AIR
EQUIV TO 5
15 30 50-CIGARETTES/DAY
l I
2 4 .6 .8 I 2 4 6 8 10 20 40 6080 200 400 1000
CARCINOGEN AIR CONCENTRATION (ANNUAL MEAN), ng/m3
Figure 13. Lung cancer mortality versus carcinogen air concentration - General Population
-------
No adjustment was made for the younger segment of the population.
Use of this approach to estimate lung cancer mortality is intended to-'
reflect a potential death risk not only for the adult population but
also for younger age groups who will eventually reach the age of 45-84.
For this reason, the predicted number of deaths in a year of increased
carcinogen air levels may be higher than observed. The additional deaths
would result as the exposed individuals aged into 45-84 year old group,
under conditions of constant ambient air carcinogen content. Conversely,
a marked improvement in air quality may indicate a death rate lower than
observed. Higher past exposure levels would be responsible for this
difference; the predicted relief would be realized at a later time.
In a nonsmoking population of 150 million, the lung cancer mortality
and the number of deaths/year is given in Table 31.
10"5 DM
HN . -p-it (88)
DN = 1500 MN (89)
DN = 8655 x '657 . 00)
p
where, PN = nonsmoking population (1.5 x 10 )
DM = number of nonsmoker lung cancer deaths/year
N
Sawicki has analyzed ambient air samples and has reported the
carcinogen concentrations found at several different locations. A
listing of some of his data is given in Table 32. Levels as high as 20
ng/m3 are indicated at some sampling sites. Reference to Figure 13
shows that continued exposure to these levels will lead to a nonsmoker
lung cancer mortality risk of 41.3, an increase of approximately 150%
over Dorn study basal rate of 16.4.
91
-------
Table 31. ESTIMATED NONSMOKER LUNG CANCER MORTALITY AS A
FUNCTION OF CARCINOGEN AIR CONCENTRATION
Based on General Population
Carcinogen air , Death A .
concentration, ng/m (x) rate/100.OOP (M^ Deaths (D^)
100 ' 118.9 178,350
80 102.7 154 050
60 85.0 127,500
40 65.1 97,650
20 41.3 61,950
10 26.2 39,300
5 16.6 24,900
2 9.1 13,650
1 5.8 8 700
.5 3.7 5,550
.1 1.3 1,950
* M ,- „ .657
M,, = 5.77 x
**
DN - 8655 x *
92
-------
Table 32. CARCINOGEN CONCENTRATIONS FOUND IN AMBIENT AIR19
a) United States
Benzo[a]pyrene
Dibenzo[a,h]anthracene
Benzo[b]fluoranthene
Benzo[j]fluoranthene
Benz[a]anthracene
Chrysene
Benzo[e]pyrene
Indeno [l,2,3,c,d]pyrene
b) Outside United States
Chrysene
Benz[a]pyrene
Benz[a]pyrene
Benz[e]pyrene
Dibenzo[a,h]anthracene
Benz[a]pyrene
6 ng/m (average U.S. urban air)
3
.4 ng/m
1.6 ng/m3 (LA, 1971-2, max.)
.8 ng/m3 (LA, 1971-2, max.)
o
4 ng/m (average U.S. urban air)
o
0.5 ng/m (average U.S. urban air)
5 ng/m (average U.S. urban air)
1.2 ng/m3 .(LA. 1971-2, max.)
I 20 ng/m3
o
71.5 ng/m (average Budapest)
o
40-415 - ng/m (German cities)
10-150 ng/m3 (typical, worldwide)
O
36-362 ng/m (German cities)
3-32 ng/m (German cities)
22 ng/m3 (Budapest, 1971-2)
o
3-7 ng/m (urban survey)
93
-------
SECTION VI
REFERENCES
1. T. R. Blackwood, "A Method for Estimating TLV Values for Compounds
Where None Exists," Letter Report from Monsanto Research Corporation,
Dayton, Ohio, to Chemical Process Section of EPA, 15 April 1975.
2. "Threshold Limit Values for Chemical Substances in Workroom Air
Adopted by AC6IH for 1974," American Conference of Governmental
Industrial Hygienists, 1974.
3. H. E. Stokinger and R. L. Woodward, JAWWA 515 (1958).
4. Wynder and Hoffman, Tobacco and Tobacco Smoke, Academic Press,
New York, N. Y., 1967.
5. H. A. Kahn, The Dorn Study of Smoking and Mortality Among U.S.
Veterans:. Report on Eight and One-Half Years of Observation, -.
*
National Cancer Institute Monograph No. 19, 1966.
6. "Statement of Basis and Purpose for the Proposed National Interim
Primary Drinking Water Standards," EPA, 1975.
7. P. L. Altman and D. S. Dittmer, "Biological Data Book," Vol. 1,
2nd Ed., Fed. Am. Soc. Exper. Biology, Bethesda, Md., 1972.
8. W. W. Waring, in "Pulmanory Disorders," E. L. Kending, Ed., Vol. 1,
p. 78, W. B. Saunders, 1972.
9. E. P. Radford, J. Appl. Physio!., 7., 451 (1955); New England J. Med.,
251., 877 (1954).
10. N. M. Nelson et al., Pediatrics, 30, 963 (1962).
11. J. Piotrowski, "The Application of Metabolic and Excretion Kinetics
to Problems of Industrial Toxicology," HEW, 1971.
12. "Documentation of Threshold Limit Values for Substances in Working
Room Air," American Society of Governmental Industrial Hygienists,
3rd Ed., 1971, 2nd Printing, 1974.
13. "The Toxic Substances List," HEW, 1974.
14. D. Bruce Turner, "Workbook of Atmospheric Dispersion Estimates,"
EPA, Office of Air Programs Publication No. AP-26, 7th Printing,
January, 1974.
15. "Water Quality Criteria Data Book. Volume I Organic Chemical
Pollution of Fresh Water and Volume II Inorganic Chemical Pollution
of Fresh Water," EPA, 1970-1971.
94
-------
16. E. C. Hammond, Smoking in Relation to the Death Rate of One Million
Men and Women, National Cancer Institute Monograph No. 19, 1966.
17. R. Doll, Cancer and Aging: The Epidemiological Evidence, The Harold
Dorn Memorial Lecture, 1970.
18. D. L. Levin, S. S. Devasa, J. D. Godwin, II and D. T. Silverman,
Cancer Rates and Risks, 2nd Edition, U.S.D.H.E.W., U.S. Government
Printing Office, 1974.
19. E. Sawicki, Analysis of a Carcinogen Conglomerate in the Environment
and in Tissue, Presented at the Fifth Annual Symposium on Recent
Advances in the Analytical Chemistry of Pollutants, Jekyll Island,
Georgia, 1975.
95
-------
SECTION VII
APPENDICES TO SECTION III
A. Effect of Age on the Build-up of Body Concentration
List of Symbols and Definitions
c instantaneous pollutant concentration in the body (mg/kg)
(c/ax)f/t> ratio of body concentration to effective pollutant con-
centration for time dependent breathing frequency
(c/ax)f_16 ratio of body concentration to effective pollutant con-
centration for constant breathing frequency of 16 (1/min)
f respiratory frequency (day )
F(t) breathing frequency as function of age
k first order rate constant of excretion (hr~ )
LD50 lethal dose 50% kill (mg/kg)
Q constant ratio of tidal volume to body weight
Ruptake rate of resPirat°ry pollutant uptake (mg/kg day)
t time (hours or days as stated)
VT tidal respiratory volume (ml)
W body weight (kg)
o
x pollutant concentration in air (mg/m )
o
xm maximum ground level concentration (mg/m )
(xrJi maximum ground level concentration of the i-th pollutant
in I *%
(mg/m )
a absorption factor for respiratory uptake of pollutant
96
-------
Introduction
The objective of the present analysis is the determination of cor-
rection factors for pollutant criteria which take into account the effect
of age on the rate of respiratory pollutant uptake.
Previously proposed pollutant criteria (Section III) were based on
pollutant uptake by adults and did not consider higher uptake rates by
children.
It is the purpose of this presentation to provide correction factors
for previously proposed decision criteria in order to provide the same
level of protection for all age groups.
General Considerations
The rate of pollutant uptake by the body from ambient air was given
by
•V. M.4,«. t-.» "™ V^TOtA/ W
uptake i
-1
where RUDtake = rate of P°11utant "Ptake (m9 k9
VT = tidal respiratory volume (m )
f = respiratory frequency (year" )
a = absorption factor determining the fraction of inhaled
compound entering the body
x = air concentration of pollutant (mg m" )
W = body weight (kg)
Three of the variables determining the rate of pollutant uptake change
considerably with age as shown in Table 33 where values are presented of
W, f, and VT for different ages. Ratios of VT/W calculated for different
ages are presented in Column 5 of Table 33. These values reveal that for
all practical purposes the ratio of VT/W can be considered constant and
independent of age.
V /W = Q = (6.36 + 0.38) x 10"6 m3 kg"1 (2)
Equation (1) then takes the form
R . . = QoxF(t) (3)
uptake y
97
-------
Table 33. RESPIRATION DATA AS DEPENDENT ON AGE AND WEIGHT
Age
(Years)
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
22
27
32
37
44
54
64
74
Weight
(kg)
(Ref. 7)
3.4
10.4
12.7
14.5
16.8
19.1
21.3
24.5
27.2
29.9
33.1
37.2
39.5
44.9
51.3
58.1
62.1
64.9
67.6
69.4
71.7
73.9
74.8
75.3
75.8
74.8
73.5
71.2
Breathing
Frequency
(min'1)
(Ref. 8)
38 + 10 10
31
26
25
24
23
22
21
20
20
19
19
19
19
19
18
17
17
16
16
Tidal
Volume
(ml)
(Ref. 9)
13 ± 510
67
87
98
110
130
150
162
180
190
220
235
245
275
300
340
365
380
405
420
Tidal Vol./
Weight
(ml/kg)
(*)
(3.82)
6.44
6.85
6.76
6.55
6.81
7.04
6.61
6.62
6.35
6.65
6.31
6.20
6.12
5.85
5.85
5.88
5.86
5.99
6.05
Relative Breat!
Frequencies w
Respect to Adu
1.94
1.63
1.56
1.50
1.44
1.38
1.31
1.25
1.25
1.19
1.19
1.19
1.19
1.19
1.13
1.06
1.06
1.00
1.00
Average value of 6.36+0.38 (ml/kg) used for further calculations.
98
-------
where F(t) is a function which describes the age dependence of the average
breathing frequency.
Further, it is assumed that the absorption factor, a, is independent
of age. This is very likely an oversimplification since one might expect
a to be lower for children than for adults because of (1) a smaller sur-
face area of the respiratory tract at early ages and (2) a shorter residence
time of the inhaled air due to the increased breathing frequency. The
assumption of a constant absorption factor thus introduces an additional
safety factor in further considerations.
Decision Criteria Based on TLV Standards
Equation (3) indicates that pollutant criteria for ambient air which
are based either directly or indirectly on TLV data should be corrected by
considering the higher breathing frequency of young children. TLV data
are established for adults for which an age-independent breathing frequency
of 16 min is assumed. In order to limit the pollutant uptake by a child
at the same level the higher average breathing frequency of the child has
to be taken into account. In the last column of Table 33 relative breathing
frequencies are given for different ages with respect to the constant
breathing frequency of adults. The highest value is about 2.0 for a child
of one year of age. Consequently, for equivalent protection of all age
groups from ambient pollutants the permissible pollutant concentration
should be only one-half of that presently permissible for adults.
A decision criterion for possible health hazards of the form
(32)
1.07 x 10~4 (LD5Q)
was derived from a correlation between TLV and LD5Q data and utilization
of the lower 95% confidence limit of the regression interval. This
criteria has to be modified in order to take into account the higher rate
of pollutant uptake by small children. Consequently, the maximum per-
missible ambient air concentration should not exceed half the value given
by Equ. (32), the latter now being postulated to be
XITI > 1.0 (91, see 34)
5.35 x 10"5 (LD,n)
'50'
99
-------
Again, if, for any pollutant, the parameter is equal or greater than unity
a hazard is judged to exist.
Numerically the same criterion as given by Equ. (91) applies for a
multi component system for which the sum over all pollutants has to be taken.
(38)
1 5.35 x 10" (LD5Q)i
Decision Criteria Based on Biological Considerations
The change in body concentration of a compound, applying a one-corn-
«_
partment model, is given by the interaction of respiratory uptake and first
order excretion of the pollutant,
dc/dt • We - "c • (92)
where k is the first order rate constant of excretion and c is the instan-
taneous body concentration. Substituting Equ. (3) into Equ. (92) one
obtains
dc/dt = QoxF(t) - kc (93)
where the first term on the right side is a function of time only because
of the changing breathing frequency.
The general solution of Equ. (93) is given by
c/ax = Cjexp(-kt) + exp(-kt) / QF(t)exp(kt)dt (94)
where Cj is the integration constant defined by the initial conditions
c = 0 at t = 0. With these initial conditions Equ. (94) yields after inte-
gration
c/ax = (Q/k) [£ k"nF^n)(t) - exp(-kt) l k~nF(n)(0)] (95)
0 0
where F^(t) and F^(0) are the n-th order derivatives of F(t) for t = t
and t = 0, respectively.
For the evaluation of Equ. (95) the time dependence of the breathing
frequency is approximated by a fourth order polynomial of the form
100
-------
F(t) = E a tn
(96)
Substituting Equ. (96) and Its higher order derivatives into Equ. (95)
yields after rearrangement
= b0[l - exp(-kt)] + b,t + b2t2 + b3t3 + b4t4 (97)
where the coefficients have the following meanings
bo = k"^k4ao - k3al + 2k% - 6ka3 + 24a4) Q (98)
b1 - k"4(k3a] - 2k2a2 + 6ka3 - 24a4) 0 (99)
b2 = k'3(k2a2 - 3ka3 + 12a4) Q (100)
b3 = k"2(ka3 - 4a4) Q (101)
b4 = k"]a4 Q (102)
Breathing frequencies for different ages according to Table 33 and
the approximation of their age dependence by Equ. (96) are shown in
Figure 14. Equation (97) was evaluated by assuming different half-life
times, T, for pollutants in the range of one to 20 years. Results of these
calculations are shown graphically in Figure 15 where values of c/ax,
the ratio of body concentration in mg/kg to the effective pollutant con-
•5
centration in mg/m , were plotted versus age in years. For comparison,
Figure 16 shows corresponding plots calculated with the assumption of a
constant breathing frequency of f = 16 min" corresponding to an adult.
In order to distinguish these c/ax values from the preceding ones they
were designated by (c/ax)f_-jg.
(c/ax)f=16 = (Q/k) [1 - exp(-kt)] (103)
In Figure 17 the effect of age on the ratio of body concentrations
calculated by both methods (time dependent breathing frequency, cf(ty and
constant breathing frequency, cf=16) are shown for pollutants of different
half-life times.
101
-------
25
20
UJ
S 15
-------
1400
o
CO
10
20
30
40
AGE IYEARS)
50
60
70
Figure 15. Ratio of body concentration to effective pollutant concentration of air for
pollutants with different half-life times in the body. Calculated for age
dependent breathing frequencies according to equ. (97). Indicated half-life
of the pollutant in years.
-------
1400-
10
30 40
AGE (YEARS)
50
60
70
Figure 16. Ratio of body concentration to effective pollutant concentration in air for
pollutants with different half-life times in the body. Calculated for the
constant breathing frequency of an adult. Indicated half-life of the pollu-
tant in years.
-------
_t
o
en
o
10
20
40
AGE (YEARS)
50
60
70
Figure 17. Effect of age dependent breathing frequency on the body concentration of pollutants with
different biological half-life times. Half-life times in years from left to right:
0.1, 0.5, 1.0, 2.5, 5, 10 and 20.
-------
cf(t)/cf=16 = Equ. (97)/Equ. (103) (104)
The lowest curve in Figure 17 corresponds to half-life times of pollu-
tants shorter than 36 days (0.1 years). For these conditions, the ratios
of body concentrations are identical with the relative breathing frequencies
referred to a-constant value of 16 min for adults (last column of
Table 33). This result is obvious since for such short half-life the body
concentration approaches.a quasi-stationary state during such a short time
interval that the change in breathing frequency during this interval can
be neglected. Even for pollutants with half-life times up to one year
(second and third curves of Figure 17) pseudo-stationary states can be
assumed with good approximations.
The effect of the higher breathing frequencies of infants and children
as compared to adults is clearly demonstrated in Figure 18. Ratios of
body concentration to effective pollutant concentration are shown for
children and adults as a function of exposure time for a pollutant with a
half-life time of 36 days. The twofold time scale in Figure 18 assumes
that a newly born child and an adult of 25 years of age are both exposed
_3
simultaneously to a pollutant concentration of x mg m .
The lower curve in Figure 18 corresponds to the adult and shows that
in less than one year a constant body concentration is established. This
-1 ?
equilibrium value in mg kg is 7.6 times the air concentration in mg m ,
provided the absorption factor a is unity. The body concentration of the
child is considerably higher and indicates a maximum value at about the
age of one when it is about 1.7 times the stationary state concentration
of an adult. The body concentration of children slowly declines with
increasing age due to the decrease in the average breathing frequency.
Pollutants with longer half-life times (4th to 7th curve in Figure 17
corresponding to half-life times of 2.5, 5, 10, and 20 years, respectively)
show pronounced deviations from quasi-stationary states. Because of their
low excretion rates and the high rate of uptake in infants, such pollutants
are readily accumulated in the body during early life. Nevertheless,
data of Figure 17 demonstrate that the body concentration of infants never
exceeds the body concentration of adults by more than a factor of 1.7.
This factor has to be considered in the modification of decision criteria
for ambient pollutants based on biological considerations. The modified
106
-------
AGE OF CHILD (YEARS)
i i i i r i
35 40
AGE OF ADULT (YEARS)
45
Figure 18. Comparison of the body concentrations of children* and adults exposed to the
same level of a pollutant with a biological half-life time of 36 days.
-------
decision criteria for possible health hazard are then given by
x
m >J.O (105; see 35)
4.77 x 10"5 (LD5Q)
for individual pollutants and for multicomponent systems by
n (x ).
Z ™-L >_!.(, (39)
1 4.77 x 10~s (LD5Q)i
In both cases, if the parameter is equal or greater than unity for any
pollutant(s), a hazard is judged to exist.
108
-------
B- Average Accumulated Body Concentration During Intermittent Exposure
List of Symbols and Definitions
B rate constants for pollutant uptake by respiration
(m /kg day)
c instantaneous pollutant concentration in the body (mg/kg)
CQ initial pollutant concentration in the body (mg/kg)
C^ final body concentration at the end of the n-th uptake
interval
C" final body concentration at the end of the n-th recovery
interval
C average body concentration during the n-th period
C^ .average body concentration during the stationary state of
intermittent exposure
C^ stationary body concentration for continuous exposure
k first order rate constant of excretion (hr~ )
t time (hours or days as stated)
T1 duration of uptake interval
T" duration of recovery interval
T length of one uptake-recovery period
x pollutant concentration in air (mg/m )
T biological half-life time of pollutant in the body (hours
or days as stated)
109
-------
Kinetics of Intermittent Exposure
The change in the body concentration of a compound was expressed as
a function of constant pollutant uptake and first order excretion.
dc/dt = Bx - kc (9)
where c = instantaneous body concentration
x = constant air concentration of pollutant
B = rate constant of pollutant uptake (respiratory tidal volume per
day per kg body weight)
k = first order rate constant of excretion
t = time
Equation (9) indicates that for continuous exposure the body concen-
tration will approach a steady state value, C^, the maximum concentration
for which is
Cm = Bx/k. (106)
Integration of Equ. (9) describes the change of concentration with time,
which is given by
c = Cw [1 - exp(-kt)] + CQ exp(-kt), (107)
where CQ is the initial body concentration at time zero and C^, the steady
state value.
Under working conditions, the exposure to polluted air occurs periodi-
cally with a period length of 24 hrs, e.g., an 8 hours working interval
followed by a 16 hours recovery interval in unpolluted surroundings. De-
signating the uptake interval by (') and the recovery interval by (") one
obtains the following expressions for the concentration changes during these
intervals.
uptake interval: c^ = Cw [1 - exp(-kt)] + CJJ_] exp (-kt) (108)
recovery interval: c^ = C^ exp (-kt). (109)
In these recursion equations the subscript n designates the number of
periods which have occurred since the start of the first exposure. The
110
-------
concentrations c^ cjj, C^. and C^., have the following meaning:
c^ instantaneous concentration during the n-th uptake interval
cjj instantaneous concentration during the n-th recovery interval
C^ final concentration at the end of the n-th uptake interval
which is identical with the concentration at the start of the
n-th recovery interval
Cn-l final concentration at the end of the (n-l)-th recovery interval
which is identical with the concentration at the start of the
n-th uptake interval
Both of the latter concentrations are extremes, the highest concentration
in one 24 hr period being given by C^, the lowest one by. C[j. Fluctuations
in body concentrations are most pronounced with compounds possessing short
biological half-life, i.e., high excretion rates. If the duration of ex-
posure exceeds about seven-times the half-life, the maximum body concen-
tration, C^, will be attained during each exposure interval and complete
excretion will take place during the recovery interval.
Under this condition, it follows from Equs. (108) and (109) that the
body concentration averaged over a 24 hr period is equal to the maximum
concentration for continuous exposure, C^, times the fraction of time
exposure occurred during the 24 hr period, e.g., for 8 working hours the
24 hrs average is only 1/3 of the maximum value.
Compounds possessing long biological half-lifes will accumulate during
subsequent periods since the recovery interval is too short for complete
excretion. If one designates the duration of uptake and recovery intervals
by T1 and T", respectively, and the length of the whole period by T = T' + T",
the average concentration during the n-th period is given by
T1 T"
n = 0/T) ' c'dt + (1/T) / c"dt. (110)
n 0 n o
Performing the integration after solving the recursion equations, Equs. (108)
and (109), one obtains
C = CT'/T) - (C^/kT) [1 - exp(-kT')] exp (-nkT1) (111)
111
-------
With an increasing number of periods, the last term in Equ. (Ill)
becomes smaller and smaller and finally approaches zero when a stationary
average body concentration, C^, is reached,
C.-C. (T'/T). (112)
In addition, the stationary average body concentration for intermittent
exposure, C^, is equal to the maximum concentration reached during con-
tinuous exposure times the fraction of time in one period during which ex-
posure occurred.
After a stationary state has been achieved, the fluctuation in body
concentrations is given by
cy<£- exp (kT"). (113)
The subsequent figures show changes in body concentration with time
for two compounds for which biological half-lifes were known. The instan-
taneous concentration is divided by the maximum stationary concentration
for continuous exposure to make the plots more general, i.e., to make
them independent of the actual air concentration. Normal working conditions
with 8 hrs exposure followed by 16 hrs of recovery in unpolluted surroundings
were assumed in each case. Prolonged recoveries during weekends were not
considered in Figures 19 and 20.
The first example (Fig. 19) illustrates p-nitrophenol uptake for which
a half-life in man of 0.99 hrs was reported. Because of the short half-life
time the maximum concentration (c/cm =1) is obtained during each exposure
interval and complete excretion takes place during the recovery interval.
The second example (Fig. 20) shows dinitro-o-cresol with a biological
half-life in man of 139 hrs. The graph shows the change in concentration
for the first 40 days of exposure. There is a slow increase in body con-
centration which after about 35 days remains fluctuating around a value
corresponding to 33% of the maximum concentration resulting during con-
tinuous exposure.
In both examples recovery during weekends was not considered. If a
periodically occurring 64 hr recovery interval is considered, the body
concentration averaged over a week is again given by Equ. (112), but this
112
-------
I I
DAYS
Figure 19. Body concentration versus time during intermittent exposure to p-nitrophenol with
a known biological half-life time in man of 0.99 hrs. Weekends not considered.
-------
r-irn i i i i—i—i—r—i—r—T—i—i—i—i i i—i—i—i—i—i—i—r—i—i—i—i—IT—i—i—r
10
15
20
DAYS
25
30 35
Figure 20. Body concentration versus time during intermittent exposure to dinitro-o-
cresol with a known biological half-life time in man of 139 hrs. Weekends
not considered.
-------
time T1 = 5 x 8, the total exposure time during the week, and T = 7 x 24,
the total length of the period. Consequently, the body concentration
averaged over a week will be only 23.8% of the stationary concentration
obtained during continuous exposure.
In Figure 21 fluctuations in body concentration are shown for p-nitro-
phenol (T = 0.99 hrs) with and without taking into account recovery during
weekends. Because of its short half-life, the maximum body concentration
is reached at the end of each working interval and complete recovery occurs
during the subsequent 16 hrs. The average concentration during one working
day equals 1/3 of the stationary concentration resulting from continuous
exposure, thus being given by 8/24. It is immediately obvious, in order
to average over a full week, the five daily averages have to be averaged
over seven days. This gives a value of 5 x (8/24)/7 = 0.238. Whether
such a weekly average is biologically meaningful seems to be questionable
but the situation changes considerably if the compound possesses a longer
biological half-life. In Figure 22 changes in body concentration are
shown for a compound with T = 72 hrs. In this case, weekends are very
effective in reducing the maximum body burden and a weekly average con-
centration does become meaningful.
In Figure 23 the situation for dinitro-o-cresol (T = 139 hrs) is
shown graphically. Because of its long half-life, the stationary state
condition is not obtained inside the time interval of the graph.
115
-------
DAYS
10
I
Q8
V I
DAYS
8
10
12
Figure 21. Effect of weekends on the body concentration during intermittent exposure to
p-nitrophenol with a biological half-life time in man of 0.99 hrs. Upper plot:
8 hr uptake period followed by 16 hr recovery period; weekends not considered.
Lower plot: 8 hr uptake period followed by 16 hr recovery period for 5 day
workweek.
-------
I i I I i i i i — i — i — i — i — i — i
r~ r — i — i — i — i — i — i — « — i — i — i — i — r
Figure 22. Effect of weekends on the body concentration during intermittent exposure to a pollutant
with an assumed biological half-life of 72 hours. Upper curve: 8 hrs uptake period
followed by 16 hrs recovery period; weekends not considered. Lower curve: 8 hrs uptake
period followed by 16 hrs recovery period for 5 day workweek.
-------
00
Iiiiiiiiiiiiiiiiiiir~iiiiii'i
i i i i i i i i i i i i i i ii
Figure 23. Effect of weekends on the body concentration during intermittent exposure
to dinitro-o-cresol with a known biological half-life time in man of 139
hrs. Upper curve: 8 hrs uptake period followed by 16 hrs recovery period;
weekends not considered. Lower curve: 8 hrs uptake period followed by
16 hrs recovery period for 5 day workweek.
-------
c- 'Nongraphic Correlation of Stack Emission Rates and Maximum Ground
Level Concentrati cms
List of Symbols and Definitions
a,b empirical constants describing the dependence of a on
the downwind distance from the source [Equ. (119)]
A thru F stability categories [Fig. 31]
D inside stack diameter
f(H) function of H yielding the maximum ground level concen-
tration [Equ. (122)]
H effective height of stack emission [Equ. (115)]
h stack height [Equ. (115)]
AH plume rise above stack [Equ. (116)]
L downwind distance from source
Lm_u downwind distance from source at which the ground level
/ \ T
concentration attains its maximum value [Equ. 021)]
m,n empirical constants describing the dependence of a£ on
the downwind distance from the source [Equ. (120)]
p atmospheric pressure
Q uniform emission rate of pollutant
T air temperature
a
T stack gas temperature
u wind speed
V stack gas exit velocity
x maximum ground level concentration
x maximum permissible ground level concentration
x(x,0,0;H) ground level concentration along center line of plume
[Equ.
0 standard deviation of plume concentration distribution in
horizontal direction [Fig. 26]
a standard deviation of plume concentration distribution in
vertical direction [Fig. 27]
119
-------
Introduction
Assessments of health hazards associated with air pollutants from
process industries require the knowledge of the maximum ground level con-
centration, xm, of a pollutant during a 24 hr period. According to the
first decision criteria for possible health hazards proposed by the
Chemical Process Section of EPA, the value of xm must not exceed the per-
missible ground level concentration, x . For a multicomponent system this
decision criteria can be formulated by
I [(xJj/UpJj] > 1.0 (36)
where (x ). and (x ). are the observed maximum ground level concentration
and the permissible ground level concentration of the i-th pollutant,
respectively. If the ratio is equal to or greater than unity a hazard
is judged to exist.
In order to decide whether a source satisfies the above criteria and
can be considered safe under given operation conditions, estimates of the
generated maximum ground level concentration as a function of weather con-
ditions are required.
The objective of the presented nomographs is to provide a simple means
for correlating stack emission rates and maximum ground level concentrations
with both weather and source conditions.
The nomographs are based on data reported in "Workbook of Atmospheric
Dispersion Estimates" by D. Bruce Turner, EPA, Office of Air Programs
Publication No. AP-26, 7th Printing, January 1974. Pertinent graphs from
Ref. 1 used for the construction of the nomographs are included.
Caveats
The nomographs are intended as an aid for obtaining first estimates
of the effects of changes in both, the emitting source and the weather
conditions, on the maximum ground level concentration of the emitted
pollutant. Because of the wide ranges of the input variables covered by
the nomographs the results can be considered only as first estimates.
All of the numerous complications that arise in the determination of at-
mospheric dispersions could not be resolved. In this respect, the caveats
expressed in the preface of Ref. 1 apply even more stringently.
120
-------
Maximum Ground Level Concentrations
The maximum ground level concentration, x^, occurs along the center
line of the plume. The dispersion of the pollutant along the center line
of the plume is given by Equ. (3.3) of Ref. 1.
x(x,0,0;H) = (QMicyz) exp[-0.5 (H/az)2] (114)
where x(x,0,0;H) = ground level concentration along center line of plume
(9 nf 3)
Q = uniform emission rate of pollutant (g sec"1)
u = wind speed (m sec )
ay - standard deviation of plume concentration distribution
in horizontal direction (m)
az = standard deviation of plume concentration distribution
in vertical direction (m)
H = effective emission height (m)
The effective emission height, H, is equal to the sum of stack height, h,
and the plume rise above stack, AH
H = h + AH (115)
The plume rise above stack, AH, can be calculated by Holland's equation,
Equ. (4.1) of Ref. 1 and is given by
AH = (VsD/u) [1.5 + 0.00268 p D (1 - Ta/Ts)3 (116)
where AH = plume rise above stack (m)
V = stack gas exit velocity (m sec" )
s i
u = wind speed (m sec )
D = inside stack diameter (m)
p = atmospheric pressure (millibar)
Ta = air temperature (°K)
Q
T = stack gas temperature (°K)
The estimation of maximum ground level concentrations from source
emission rates requires two equations to be evaluated: Equ. (116) for
obtaining plume rise above stack and Equ. (114) for correlating emission
121
-------
rate and maximum ground level concentration. Both evaluations are pre-
sented in nomographic forms.
Nomograph for the Evaluation of Holland's Equation (Plume Rise Above Stack)
For the determination of plume rise above the stack, Holland's equation2
was applied as suggested in Refs. 1 and 3. This equation is one of the
A
most successful to date as shown by a critical evaluation of different for-
mulas proposed, although the final height of the plume is somewhat under-
3
estimated.
Holland's equation, Equ. (116), contains seven variables which makes
the construction of a nomograph rather complex. The equation was there-
fore simplified by neglecting changes in barometric pressure, which generally
varies only a few percent. Consequently, a pressure of 970 millibars was
assumed, yielding an equation of the form
AH = (VeD/u) [1.5 + 2.6 D (1 - T/T)]. (117)
S a S
Equation (117) is presented in nomographic form by breaking it up into two
factors: one given by the term in parenthesis and the other, given by the
term in brackets.
The nomograph for Holland's equation is presented in Figure 24, the
evaluation being discussed in the subsequent section. An additional nomo-
graph is given in Figure 30.
The left hand diagram of the nomograph shows the dependence of
(1 - Ta/Ts) on both the air temperature, TQ, and the stack gas temperature,
TS. The right hand diagram of the nomograph presents values of (V /u) as a
function of both the stack gas exit velocity, V$, and the wind speed, u.
Wherever possible, scales for input data were provided in metric and
engineering dimensions.
The value of [1.5 + 2.6 D (1 - Tfi/Ts)] is obtained on the auxiliary
scale I and is then transformed logarithmically by following the guiding
arrows from scale I to scale II. The value of log (V D/u) is obtained on
the auxiliary scale III. Multiplying the latter value with the value on
scale II yields the desired value of plume rise above stack on scale AH.
Example for the Determination of Plume Rise Above Stack
An example for the use of the nomograph is presented in Figure 24.
The following source and weather conditions were assumed to apply:
122
-------
Vs (ft/sec)-
Ta CF)
•jo -ao -10
TQ — AIR TEMPERATURE ("C or°F)
Ts-» STACK GAS TEMPERATURE (°C)
D-- INSIDE STACK DIAMETER (meter)
AH---RISE OF PLUME ABOVE STACK (meter) •
Vs— STACK GAS EXIT VELOCITY (m/sec or ft/sac.)
u ... WIND SPEED (miles/hr)
oa
I
Z S 4 6 6 10 ZO M40 00 8O
I I I 1 III ll i 1 . I I.I.
Vs (m/sec)
Figure 24. Monographic example for the determination of plume rise above stack by Holland's
equation.
-------
Air temperature Tfl = 70°F
Stack Gas Temperature T$ = 150°C
Inside Diameter of Stack D = 2 meter
Stack Gas Exit Velocity Vg = 9 ft/sec
Wind Speed = 2 miles/hr
Stack Height h = 80 meter
Evaluation Procedure:
Step 1. On left hand diagram select air temperature (°C on lower scale
or °F on upper scale) and go vertically until the line repre-
senting the stack gas temperature (°C) is met (interpolate if
necessary). From this point go horizontally to the right border
line of the diagram obtaining point 1.
Step 2. Select stack diameter on slanted D-scale and connect with point 1.
Extension of the line cuts -the auxiliary line I at some value
(2.1 in this example). By use of the guiding arrows the same
value is located on the auxiliary line II corresponding to
point 2.
Step 3. On the right hand diagram select stack gas exit velocity (meter/
sec on lower scale or ft/sec on upper scale) and go vertically
until the line representing the wind speed (miles/hr) is met
(interpolate if necessary). From this point go horizontally to
the left border line of the diagram obtaining point 3.
Step 4. Select stack diameter on the adjacent D-scale and connect with
point 3. Extension of the line yields point 4 on the auxiliary
line III.
Step 5. Connect pointy 4 with the previously obtained point 2. The tie-
line intercepts the AH-scale yielding the plume rise above stack
in meter. In this example, the value of AH is found to be 20 m
and consequently the effective height of emission is 80 + 20 =
100 m.
Nomograph for the Correlation of Emission Rate and Maximum Ground Level
Concentration
The nomograph correlating emission rates with maximum ground level
concentration as a function of effective heights of emission is based on
data from Figure 25 (Fig. 3-9 of Ref. 1).
124
-------
10"
Figure 25. Distance of maximum concentration and maximum (xu/Q) value as a function of weather
condition and effective height of emission (Fig. 3-9 of reference 1).
-------
(xmu/Q) = f(H) for stability categories A thru F (118)
where x = maximum ground level concentration
u = wind speed
Q = emission rate
f(H) = function of effective height of emission for which the ground
level concentrations attains a maximum value
The data of Figure 3-9 of Ref. 1 are limited to an effective height of
emission below 300 m. This range was extended for some of the stability
categories although one has to be cautious in the interpretation of the
results obtained for the extended range because of the unknown weather
conditions at these heights.
The values of horizontal, 0 , and vertical, 0 , dispersion coefficient
as functions of downwind distance from the source are presented in Figures
26 and 27, respectively, corresponding to Figures 3-2 and 3-3 of Ref. 1.
The graph for a indicates that o can be expressed by
*/ J
ay = a Lb (119)
where L is the distance downwind from source and a, b are empirical con-
stants. The dependence of 0 on the downwind distance from the source can
be expressed in an analogous way
0Z = m Ln - (120)
provided the downwind distance from the source exceeds a limiting value
depending on the stability category.
Insertion of Equs. (119) and (120) into Equ. (114) and equating to
zero after differentiating with respect to L, yields the downwind distance
from the source at which the ground level concentration attains a maximum
value, L
max
max
= CnH2/m2(b + n)]1/2n. (121)
126
-------
10,000
0.1
Figure 26.
1 to
DISTANCE DOWNWIND, km
Horizontal dispersion coefficient as a function of downwind
distance from the source (Fig. 3-2 of reference 1).
127
-------
10
DISTANCE DOWNWIND, km
Figure 27. Vertical dispersion coefficient as a function of downwind distance
from the source (Fig. 3-3 of reference 1).
128
-------
Replacing ay and CTZ In Equ. (114) with Equs. (119) and (120), respectively,
and L with Equ. (121), one obtains the maximum ground level concentration
as a function of H only.
f(H) - (xmu/Q)
• (amir)-1 [nm-2(l>tnr1]-(b+n)/2n H-(b+n)/n exp[-(b+n)/2n] (122)
By means of Equ. (122) the range of effective emission heights was extended.
The nomograph for correlating emission rates and maximum ground level
concentrations is presented in. Figures 28 and 29. Evaluation examples are
discussed in the subsequent section. An additional nomograph is given in
Figure 31.
The left hand side of the nomograph contains six scales for the effec-
tive emission heights corresponding to the six stability categories. The
applicable stability category can be selected from the insert below the
auxiliary center line. The scales are graded according to the values of
(xmu/Q) corresponding to the values of H for the stability category under
consideration.
With both H and u being given, the nomograph allows either (1) the
determination of the maximum emission rate for which the ground level con-
centration will not exceed a given maximum permissible value, or (2) the
determination of the maximum ground level concentration resulting from a
given emission rate.
Examples for the Use of Nomograph
Example I. Calculate the maximum emission rate for given source and
weather conditions which will not cause the ground level
concentration to exceed a given maximum permissible value,
V
Given Values: Moderately cloudy day with wind speed of 2 miles/hr
corresponding to stability category B. Effective height
of emission from previous example H = 100 m. Maximum
permissible ground level concentration xp = 10 yg/m .
Procedure: Outlined in Figure 28.
Step 1. Select scale for effective height of emission according to
applicable stability category (B in this example).
129
-------
1000 '_
700 .
300
400 '
300
•250 "
200 .
• ISO
• IOO
TO '
• SO •
•4*0 •
30 •
•20 .
IS .
•0 .
•5
1000 .
• too
•600 "
•500
.400
• 300 *
•2SO
•203
•!»
—
•70 '
.
*^0
.40 •
•SO •
• 20 •
•IS .
•IO
•s
rWOO
•SOO •
• coo
•300
• 400 '
300 •
•230
200
•ISO ^
.ino
•70
•90 •
• 40 '
•30 '
•20 •
•IS
•10
•9 ' •
•1000
•(00 '
• 700
•600
•900
• 400
•300
•230
• 200
• 130
• IOO
• 7U
- so
• 40
-30
.20 •
-IS -
-10 "
-S •
•soo
•400
• 300 •
• 230
- 2SO
- ISO
- IOO
- vn
- SO
p4O
- JO
• 20 '
- 19 "
- 10 '
-s
> 200
ISO
• IOO
•70
• SO
• 4O
• JO
• 20
• IS
• 10
- B
\ 8 C D E F
STABILITY CATEGORIES
JT
ir
^N
*
MY TO STABILITY CATEGORIES
EFFECTIVE HEIGHT OF EMISSION
(meter)
WIND SPEED
(m/sec)
(miles/hr)
Swlxe Wiiri
inn la 19 M.
«!«->
< 2
2-3
3-5
S-S
' «
Oil HifM
IncomiAi Salw ArtuliM TfcMj OvciCMI
Stran|
A
A-8
B
C
e
Matfefiti Slicltl M/ILawClMrf
A8
B
84
C-0
0
f
0
0
D
Clmtf
F
E
0
D
aooH
•i-ooooi
EMISSION
RATE xmor x.
(g/sec)(lb/hr){mg/m3')
At Mil'il
-------
Step 2. Select value of effective height of emission and project hori-
zontally on line H thereby obtaining point 1.
Step 3. Select xp value on x-scale (0.01 in this example) and connect
with point 1. Intercept on the auxiliary center line corresponds
to point 2.
Step 4. Select wind speed (m/sec or miles/hr) on u-scale and connect
with point 2 on center line. Extension of this line to Q-scale
yields the emission rate either in g/sec or Ib/hr (4.5 Ib/hr in
this example).
Example II. Calculate the maximum ground level concentration, xm, resulting
from a given emission rate, Q, for given weather and source
conditions.
Given Values: Cloudy day with wind speed of 2 miles/hr corresponding
to stability category C. Effective heigh of emission from
previous example H = 100 m. Emission rate Q = 10 Ib/hr.
Procedure: Outlined in Figure 29.
Step 1. Select emission rate on Q-scale and wind speed on u-scale. The
connecting line intercepts the auxiliary center line yielding
point 1.
Step 2. Select effective height of emission on proper scale corresponding
to applicable stability category (C in this example) and project
horizontally on line H thereby obtaining point 2.
Step.3. Connect point 1 with point 2 and extend line to x-scale. The
intercept on x-scale yields the maximum ground level concentra-
tion in mg/m3 (20 yg/m3 in this example).
References
1. D. Bruce Turner, "Workbook of Atmospheric Dispersion Estimates," EPA
Office of Air Programs Publication No. AP-26, 7th Printing, January,
1974.
2. J. Z. Holland, "A Meterological Survey of the Oak Ridge Area," U.S.
Atomic Energy Commission Report ORO-99, 554-559 (1953).
3. A. Beals, "Guide to Local Diffusion of Air Pollutants," Technical
Report 214, Air Weather Service, USAF, Scott AFB, 111., May, 1971.
131
-------
-— . — -
woo
TOO .
900
400 |
300
230 '
200 .
• 190
DO
TO '
•90 •
•40 •
30 •
20 .
IS .
•» -
i j.
'HDD
•00
coo
900
400
300 .
230
200
BO
TO •
90
40 •
30 •
20
IS .
D •
•
•noo
BOO
COO
900
400 '
300 •
•230
200 '
190
TO
50 •
40 '
30 '
20 •
IS '
a •
•1000
«oo
•TOO
coo
' SOO
400
•300
•290
•209 _
190
100
TO
SO
40
30 '
20 '
13 •
H> •
• _
•900
•400
300
290
• 290
• ISO
too
90
4O
SO
20 '
IS '
10 '
• 200
•ISO
• 100
•TO
•90
•4O
• SO
20
• IS
• 10
•
• f - » — 9 — a — B
\ B C D E F
STuaUTY CATEGORIES
EFFECTIVE HEIGHT OF EMISSION
(meter)
0 X
u
D
~/ ai-
^' 6 C D 0 D D'
ill v «i|M."
Figure 29. Nomographic example for the calculation of maximum ground
level concentration resulting from given stack emission rate.
132
-------
4. H. Moses and G. H. Strom, "A Comparison of Observed Plume Rises with
Values Obtained from Well-Known Formulas," J. Air Pollution Control
Association, 11, 455 (1961).
133
-------
Figure 30. Nomograph for the estimation of plume rise above stack
by Holland's equation.
Figure 31. Nomograph for the correlation of stack emission rates and
maximum ground level concentrations.
134
-------
o
n
To CF)-
-JO -20 -10 0 10 20 SO 4*0
ESTIMATION OF PLUME RISE ABOVE STACK BY HOLLAND'S EQUATION
H=(VS D/u)
Ta—AIR TEMPERATURE (eCor°F)
Ts-" STACK GAS TEMPERATURE (°C)
D--- INSIDE STACK DIAMETER (mater)
AH• • • RISE OF PLUME ABOVE STACK (meter) •
V8- STACK GAS EXIT VELOCITY (m/sec or ft/sec.)
u ••• WIND SPEED (miles/hr)
AH{m)
••1.0
• -2
•-J
• 4
10- •
•too
t"
•1-20
90
40
50
60
80
ISO
200
1-300
J-4OO
+800
-6OO
D(m)
oj
o*
0.8
10
Vs (ft/sec)-
0.9 I 2 S 4 6 8 tO 20 3040 60 SO
I I I Mi I.hi i . . .1.1.
V, (m/sac)
-------
CORRELATION OF EMISSION RATES AND MAXIMUM GROUND LEVEL CONCENTRATIONS
H
1
.
;
_ i
co
en
1000 ~
700 .
[-500
400 ;
-300
250
-200 .
- 150
r 100
• 70
• 50 •
• 40 •
• 30 •
•20 .
• 15 .
•K) -
1000 .
800
600 "
500
-400
•300 .
•250
•200
• 150
•100
•70 •
•50
-40 •
•30 •
•2O
•IS .
-10
1000
800 -
600
500
400 "
300 •
•250
• 200
• 150
•
-100
•70
•50 -
•40
-30 '
•20 -
•15 '
• to
1000
800 '
700
600
•500
400
•300
•250
• 200
• ISO
• 100
•70
• 50
•40
- 30
-20 '
-15 '
-10 '
500
400
300 -I
250
- 250
r 150
• 100
•70
•so
•40
• 30
- 2O
- IS "
- 10
200
150
• 100
-70
• 50
• 40
• 30
• 20
- IS
• 10
A B C D E F
STABILITY CATEGORIES
'
U
0.1-
0.2-
0.3-
0.4-
O.5-
0.7-
1-
2_
3-
5-
7-
10-
20-
30-
40-
50-
70-
100-
-0.1
-0.2
-0.3
-0.4
-0.5
-0.7
-1
-2
-4
-5
-7
- 10
-20
-30
-SO
-70
-100
1000-
800 •
600-
400-
200
100-
80 •
60-
40-
20-
10-
8-
6-
4.
2-
1-
08
. 0.6
0.4
0.2 •
0.1-
0.01-
KEY TO STABILITY CATEGORIES
OOQI-
- K>.000
-'1000
800
600
400
200
-100
80
60
• 40
• 20
— 10
IV
• 8
. 4
• 2
r
•
•
•
-0.1
:
•
-O.OI
woo
800
600
400
200
100
80
60
4O
20
10
8
6
4
• 2
- 1
-0.8
-0.6
-0.4
-02
-O.I
4
-0.01
-0.001
4
-oocxx
D»y Night
EFFECTIVE HEIGHT OF EMISSION WIND
(meter)
(m/sec)
SPEED Speed (it 10 4 l»«»i»SS»l» ««»*"»> Thinl»0«,c«l ^ £MIS
SUonj Module Slight *4/« low Cloud Cloud rV
(miles/hr) < 2 A AB B (g/sec)
2-3 AB B C £ F
35 B BC C 0 E
>SION
UE xmor xp
(lb/hr){mg/m3)
56 C C-D D 0 D
> 6 C D D 0 D
flu MU|»| cl»t. D. should bi i»unu4 lor tneicut uMitims dmlng
in or nighL
-------
^
TECHNICAL REPORT DATA
(riease read Instructions on the reverse before completing)
EPA-600/2-76-155
4. TITLE AND SUBTITLE '"
Estimation of Permissible Concentrations of
Pollutants for Continuous Exposure
7. AUTHOR(S)
Robert Handy and Anton Schindler
9. PERFORMING ORGANIZATION NAME AND ADDRESS ~
Research Triangle Institute
P.O. Box 12194
Research Triangle Park, North Carolina 27709
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
3. RECIPIENT'S ACCESSION-NO.
5. REPORT DATE
June 1976
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
10. PROGRAM ELEMENT NO.
EHE624
11. CONTRACT/GRANT NO.
68-02-1325, Tasks 34 and 46
13. TYPE OF REPORT AND PERIOD COVERED 1
Final; 6/75-4/76
14. SPONSORING AGENCY CODE I
EPA-ORD
15. SUPPLEMENTARY NOTES JJERL.RTP task officer for this report is Max Samf ield , Mail Drop I
62, Ext 2547.
IB. ABSTRACT Tne repOrt gives results of a. study aimed at: estimating the maximum
permissible continuous exposure to which individuals may be subjected; and computing
permissible exposure to a multipollutant system so that the degree of control neces-
sary for a waste stream can be computed. (These capabilities are desirable in the
absence of a Federal Standard for a pollutant.) The report deals with three aspects of
the problem: permissible continuous exposure for air pollutants; permissible contin-
uous exposure for water pollutants; and permissible continuous exposure to carcin-
ogens in ambient air. For carcinogens, the problem is approached on a probable
risk/concentration basis. Calculated permissible levels for 24 hours exposure were
compared with 23 substances for which standards have been set or proposed. In gen-
eral, agreement was within an order of magnitude. An appendix to the report contains
a nomograph for easy estimation of maximum ground level concentrations of pollut-
ants issuing from point sources.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
. COSATI Field/Group
Pollution
Water Pollution
Estimating
Concentrations
(Composition)
Exposure
Carcinogens
Pollution Control
Stationary Sources
Continuous Exposure
Point Sources
13B
14A
07D
06S
06E
18. DISTRIBUTION STATEMENT
19. SECURITY CLASS (This Report)
Unclassified
21. NC
147
Unlimited
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
------- |