&EPA
United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park NC 27711
EPA-450/3-78-031
June 1978
Air
Assessment of
Human Exposures
to Atmospheric
Benzene
Final
-------
EPA-450/3-78-031
Assessment of Human Exposures
to Atmospheric Benzene
by
Susan J. Mara and Shonh S. Lee
SRI International
Menlo Park, California
Contracts No. 68-01-4314 and 68-02-2835
EPA Project Officer: Richard J. Johnson
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Waste Management
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
June 1978
-------
This report is issued by the Environmental Protection Agency to report
technical data of interest to a limited number of readers. Copies are
available free of charge to Federal employees, current contractors and
grantees, and nonprofit organizations - in limited quantities - from the
Library Services Office (MD-35) , U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina 27711; or, for a fee, from the
National Technical Information Service, 5285 Port Royal Road, Springfield,
Virginia 22161.
This report was furnished to the Environmental Protection Agency by SRI
International, Menlo Park, California, in fulfillment of Contracts No.
68-01-4314 and 68-02-2835. The contents of this report are reproduced
herein as received from SRI International. The opinions, findings, and
conclusions expressed are those of the author and not necessarily those
of the Environmental Protection Agency. Mention of company or product
names is not to be considered as an endorsement by the Environmental
Protection Agency.
Publication No. EPA-450/3-78-031
11
-------
CONTENTS
LIST OF ILLUSTRATIONS vii
LIST OF TABLES ix
PREFACE xiii
ACKNOWLEDGMENTS xv
I SUMMARY 1
II BENZENE IN THE ENVIRONMENT 7
A. Introduction 7
B. Nonatmospheric Benzene 9
C. Chemical and Physical Properties of Benzene .... 11
III CHEMICAL MANUFACTURING FACILITIES 17
A. Sources 17
B. Methodology 25
C. Exposures 33
IV COKE OVENS 39
A. Sources 39
B. Methodology and Exposures 42
V PETROLEUM REFINERIES 51
A. Sources 51
B. Methodology 56
1. Refining of Crude Oil 56
2. Storage and Transfer of Pure Benzene 60
C. Exposures 62
VI SOLVENT OPERATIONS 65
A. Sources 65
B. Methodology and Exposure 67
111
-------
VII STORAGE AND DISTRIBUTION OF GASOLINE . . 75
A. -Sources 75
1. Storage 75
2. Distribution 77
B. Methodology and Exposures 78
VIII URBAN EXPOSURES 85
A. Sources 85
B. Methodology and Exposures 88
1. Urban Exposures from Automobile Emissions ... 88
2. Urban Exposures from Gasoline Service Stations . 97
IX SELF-SERVICE GASOLINE . . .' Ill
A. Sources Ill
B. Methodology and Exposures 114
X ASSESSMENT OF TOTAL EXPOSURE 117
A. Introduction 117
B. Determination of an Individual's Use of Time .... 118
C. Distribution Into Population Subgroups 122
D. Selection of Applicable Exposure Data . . . . .. . . 125
E. Summary of Exposures 132
BIBLIOGRAPHY . . R-l
APPENDICES
A. Diagrams of Various Benzenes-Related Operations . . . A-l
B. Emission Rates and Population Exposures from
Chemical Manufacturing Facilities B-l
C. Population Exposures from Coke-Oven Operations by
Location C-l
D. Population Exposures from Petroleum Refineries by
Location D-l
E. Partial List of Hydrocarbons and Additives
Contained in Gasoline E-l
IV
-------
LIST OF ILLUSTRATIONS
III-l Benzene Derivatives and Their Uses 21
III-2 Dispersion Modeling Results for Each Type of
Source Category 29
III-3 Predicted Annual Average Benzene Concentrations in the
Vicinity of Selected Chemical Manufacturing Facilities . 36
III-4 Comparison Between Predicted Annual Average and 8-Hour
Worst Case Benzene Concentrations in the Vicinity of
Two Chemical Manufacturing Facilities . 37
IV-1 Dispersion Modeling Results for Coke-Oven Operations . . 46
V-l Monitoring Data for Gulf Alliance Refinery,
Belle Chasse, Louisiana 53
V-2 Dispersion Modeling Results for Three Size Categories
of Petroleum Refineries 59
VI-1 Sampling Data for Three Solvent Operations 68
VII-1 The Gasoline Marketing Distribution System in the
United States 76
VII-2 Estimated Dispersion Curve for a 3-Tank Gasoline
Bulk Terminal 81
VII-3 Estimated Dispersion Curve for a 10-Tank Gasoline
Bulk Terminal 83
VIII-1 Isopleths (m/sec) of Mean Annual Wind Speed Through
the Morning Mixing Layer 92
VIII-2 Results of Atmospheric Monitoring in the Vicinity of
Urban-Suburban Gasoline Service Stations, Summer, 1977 . 99
VIII-3 Results of Atmospheric Monitoring in the Vicinity of
Rural-Mountain Gasoline Service Stations, Summer, 1977 . 100
VIII-4 Regression Curves Developed From API Atmospheric
Monitoring Data Collected in the Vicinity of
Gasoline Service Stations 109
-------
TABLES
1-1 Summary of Estimated Population Exposures to Atmospheric
Benzene from Specific Benzene Emission Sources 3
1-2 Summary of Estimated Total Exposures of People
Residing in the Vicinity of Atmospheric Benzene
Sources • 5
II-l Estimated Benzene Levels in Food 11
II-2 Properties of Benzene 12
III-l Locations and Capacities of Plants Using Benzene
as an Intermediary Agent in the Manufacture of
Various Chemical Compounds 18
III-2 Atmospheric Benzene Concentrations Sampled
at Benzene-Consumption Facilities .... 22
III-3 Major Chemical Compounds Other than Benzene
Emitted from Chemical Manufacturing Facilities 23
III-4 Emission Factors and Characterizations for
Benzene-Consumption Plants 24
III-5 Rough Estimates of Ambient Ground-Level Benzene
Concentrations (8-Hour Average) 26
III-6 Rough Estimates of Ambient Ground-Level Benzene
Concentrations (8-Hour Average) per 100 g/s Emission Rate 28
III-7 Estimates of a 8-Hour Worst Case Benzene Concentrations
Based on Average of Three Emission Source Categories ... 30
III-8 Estimated Population Exposed to Benzene From
Chemical Manufacturing Facilities, by State 34
IV-1 Estimated Size and Productive Capacity of By-Product
Coke Plants in the United States on December 31, 1975 . . 40
IV-2 Ambient Levels of Benzene Within a Coal-Derived
Benzene Production Plant 41
IV-3 Atmospheric Benzene Emission from the Coking and
Recovery Plants in Czechoslovakia 42
IV-4 Partial List of Constituents of Coke Oven Emissions . . 43
IV-5 Rough Estimates of 8-Hour Worst Case Benzene
Concentrations per 100 g/s Emission Rate Using
the PAL Dispersion Model 47
IV-6 Estimated Population Exposed to Benzene from
Coke Ovens, by State 50
VI
-------
V-l Petroleum Refineries Producing Aromatics, by State .... 52
V-2 Atmospheric Benzene Concentrations at Distances
Greater than 1 KM from Perimeter of Petroleum
Refineries 54
V-3 Results of Ambient Benzene Monitoring in the
Vicinity of Petroleum Refineries 55
V-4 Calculation of Emission Factors for Petroleum Refineries . 57
V-5 Summary of Emission Factors for Pure Benzene
Storage and Transfer 61
V-6 Estimated Population Exposed to Benzene from
Petroleum Refineries, by State 63
VI-1 Industries and Manufactured Products Possibly Using
Benzene as a Solvent 66
VI-2 Average Number of Employees per Plant for Selected
Solvent Operations 69
VI-3 Number of Plants and Employees for Solvent Operation
with High Potential for Benzene Emissions 70
VI-4 States with the Highest Potential for Atmospheric
Benzene from Solvent Operations 72
VII-1 Emission Factors for Benzene Losses from Gasoline
Storage and Distribution 79
VIII-1 Results of Ambient Benzene Monitoring in Urban Areas
with High Industrial Activity 86
VIII-2 Typical Liquid Volume-Percent of Benzene in Gulf
U.S. Gasolines, October 1976 87
VIII-3 Benzene Concentration in Different Grades and
Seasonal Blends of Gasoline 88
VIII-4 Estimates of Annual Average Benzene Concentrations
in Four Urban Areas 90.
VIII-5 Estimates of Average Annual Benzene Concentrations
for Cities with Population Exceeding 1,000,000 93
VIII-6 Estimates of Average Annual Benzene Concentrations
for Selected SMSAs 95
VIII-7 Urban Population Exposures Related to
Automobile Emissions 96
VIII-8 Service Station Density in Four Metropolitan AQCRs .... 98
VIII-9 Rough Dispersion Modeling Results for Gasoline
Service Stations 103
VIII-10 Emission Factors for Benzene Losses at Gasoline
Service Stations 104
VI1
-------
VIII-11
VIII-12
IX-1
IX-2
IX-3
IX-4
IX-5
X-l
X-2
X-3
X-4
X-5
X-6
X-7
X-8
X-9
X-10
X-ll
X-12
Determination of Weighted U.S. Average Wind Speed 106
Estimates of Annual Average Benzene Concentrations
in Urban Areas from Gasoline Service Stations Based
on the Hanna-Gifford Dispersion Model 107
Self-Service Operations 112
Gasoline Market Share of Self-Service Stations
in Four AQCRs, Spring 1977 113
Gasoline Market Share of Self-Service Stations
in Two Metropolitan Areas, 1976
114
Sampling Data from Self-Service Gasoline Pumping 115
... 116
... 119
Estimated Population Exposed to Benzene from
Self-Service Gasoline
Percentage of Time Spent per Week in Major Types
of Activities by Employed Men in Urban Areas, 1975
Percent of Those Employed by Place of Residence
and Commuting Time to Work 121
Estimated Annual Distribution of Time Spent in
Various Activities and Locations 123
Distribution of the Population into Groups
Affected by Other Benzene Exposure Settings 124
Results of Battelle Atmospheric Monitoring Study 126
Traffic Volume and Density During Benzene
Sampling Periods at Site 2 127
Benzene Concentrations in Remote Air Samples from
the Continental United States
128
Benzene Concentrations in the Vicinity of Major
Intersections Based on Dispersion Modeling 130
Benzene Exposure Settings in Vicinity of Specific
Locations
131
Estimate of Total Exposure for Each Scenario 135
Example Calculation of Total Exposure for
Two Benzene Source Categories
Summary of Estimated Total Exposures of People
Residing in the Vicinity of Atmospheric Benzene Sources
137
139
vm
-------
PREFACE
There is substantial evidence that concentrations of benzene
encountered in the workplace (both in the United States and elsewhere)
have caused blood and bone marrow diseases (e.g., blood dyscrasia,
pancytopenia) and leukemia (especially myelogenous leukemia). As current
U.S. Environmental Protection Agency (EPA) policy states that there is no
zero risk level for carcinogens, benzene has been listed by EPA under
Section 112 of the Clean Air Act as a hazardous air pollutant. To
determine what regulatory action should be taken by EPA on atmospheric
emissions of benzene, three reports have been prepared: (1) a health
effects assessment, (2) a population exposure assessment, and (3) a risk
assessment document based on the data in the first two assessments.
This document is the human population exposure assessment and presents
estimates of the numbers of people in the general population of the
United States exposed to atmospheric concentrations of benzene from
specific sources.
ix
-------
ACKNOWLEDGMENTS
It is a pleasure to acknowledge the cooperation and guidance given
by several individuals of the U.S. Environmental Protection Agency.
Mr. Richard J. Johnson, Office of Air Quality Planning and Standards,
Strategies and Air Standards Division, generously provided data and
direction throughout the study. Messrs. Phillip L. Youngblood and
George J. Schewe (Office of Air Quality Planning and Standards, Monitor-
ing and Data Analysis Division) conducted dispersion modeling, offered
guidance about the application of their results, and reviewed draft
documents. Additional assistance was provided by many individulas in the
Emission Standards and Engineering Division of the Office of Air Quality
Planning and Standards. Mr. Alan P. Carlin, Office of Research and
Development, and Mr. Joseph D. Cirvello, Office of Air Quality Planning
and Standards, were the Project Officers.
Mr. Benjamin E. Suta, SRI project leader, gave vital support and
provided useful input throughout the study. Mr. Robert E. Freeman and
Dr. Stephen L. Brown provided encouragement to the authors and acted as
liaisons with other interested parties. Mr. Michael Smith patiently
edited and re-edited the study. Mrs. L. H. Wu and Ms. Grace Y. Tsai
were responsible for all graphics.
-------
I SUMMARY
This report is one in a series that SRI International is providing
on a quick-response basis for the U.S. Environmental Protection Agency
(EPA). Populations-at-risk to selected pollutants are being quantified
for input to other, more inclusive studies. The primary objective of
this study was to estimate the environmental exposure of the U.S. popu-
lation to atmospheric benzene emissions from specific sources. In this
estimate it was assumed that individuals residing in the vicinity of
benzene sources spend 24 hours of each day in the same location. To
estimate more representative exposures, a second objective was added—
to make rough estimates of individuals' total exposures by defining
"total exposure" as the sum of exposures to all benzene sources, including
those in nonresidential areas, within a designated period.
The seven primary sources of atmospheric benzene considered in this
report are chemical manufacturing facilities, coke ovens, petroleum
refineries, solvent operations, gasoline storage and distribution centers,
self-service gasoline stations, and urban exposures related to automobile
emissions and evaporation from gasoline service stations.
Data were quite limited for this study. When data were available,
source locations were identified and benzene emission rates were calcu-
lated. Atmospheric environmental concentrations of benzene were then
estimated by applying approximate, dispersion modeling results developed
*
by EPA. Population exposed to concentrations of 0.1 ppb and greater
were estimated. When data were unavailable, best estimates were developed
to provide a reasonable basis for comparison.
Total exposures were estimated by developing scenarios that repre-
sent the typical behavior of individuals living in the vicinity of benzene
*
The detection limit of current sampling techniques.
-------
sources. Based on available sociological, statistical, and ambient
monitoring studies, the following variables were estimated: percentage
of time spent in activities away from the residence such as shopping and
working; proportion of the population following given living patterns;
and typical benzene concentrations associated with each activity. The
percentage of time spent in 1 year in various activities wad multiplied
by the associated benzene concentration and summed to determine an
individual's total exposure.
The resulting estimates are subject to considerable uncertainty
in regard to: (1) the quantity of benzene emissions, (2) benzene
production and consumption levels, (3) source locations, (4) control
technologies employed, (5) deterioration of control technologies over
time, (6) physical characteristics (e.g., stack height) of benzene sources,
(7) population density in the vicinity of sources, (8) living patterns
of the exposed population, and (9) relationship of ambient monitoring
data to dispersion modeling estimates. Given these complex and variable
factors, the accuracy of the estimates could hot be assessed quantita-
tively. Nevertheless, the estimates, although not precise, approximate
expected conditions.
Table 1-1 summarizes results of the source-specific assessment.
Urban exposures constitute the largest sources. Chemical manufacturing
facilities are second, with more than 7 million people exposed over a
wide range of exposure levels. Petroleum refineries are sources of
benzene exposures for more than 5 million people.
For approximate comparison of different emission sources, exposures
are calculated in similar units by multiplying the number of exposed
population by the annual average benzene concentration within each
range. These values were then summed for each emission source. Thus,
the units become ppb-person-years (Table 1-1). We assumed that people
who use self-service gasoline are exposed to this emission source for
1.5 hours per year.
-------
Table 1-1
SUMMARY OF ESTIMATED POPULATION EXPOSURES TO
ATMOSPHERIC BENZENE FROM SPECIFIC BENZENE EMISSION SOURCES'
Source
Number of People Exposed to Benzene Concentrations (ppb)
8-hour Worst Case: 2.5-25.0
Annual aver age : 0.1-1.0
Chemical manufacturing
Coke ovens
Petroleum refineries
Solvent operations
Storage & distribution of
gasoline
Automobile emissions - urban
Gasoline service
stations - urban
People using self-service gasoline
6,000,000
300,000
5,000,000
d
e
69,000,000
20,000,000
25.1-100.0
1.1-4.0
1,000,000
3,000
45,000,000
900,000
100.1-250.0
4.1-10.0
200,000
>250.0
> 10.0
80,000
g
Total0
7,300,000
300,000
5,000,000
—
110,000,000
21,000,000
37,000,000
Comparison
Among Sources
(106 ppb-person-years)
8.5
0.2
2.5
—
150.0
12.0
1.6
u>
Assumes that people living in the vicinity of benzene sources spend 24-hours of each day in that location.
Vi 3
To convert to yg/m , multiply each exposure level by 3.2; to estimate one-hour worst case concentrations
multiply 8-hour worst case by 10.
Population estimates are not additive vertically, because double-counting exists. Totals are rounded to two
significant figures.
Exact determination is impossible.
eEstimated at « 0.1 ppb annual average. The population exposed was not determined but is assumed to be very small.
8-hour worst case is estimated by multiplying each exposure level by 4.1.
8Estimated at 245 ppb for 1.5 hr/yr/person.
Source: SRI estimates.
-------
The results presented in Table 1-1 show that urban exposures have
the highest weighted human exposures. Next are chemical manufacturing
plants, followed by petroleum refineries. These results differ because
they are weighted by the number of people exposed to a particular level
of atmospheric benzene. Thus, they provide a useful basis for compari-
son and, assuming a linear dose-response relationship, are directly
related to human health.
The total exposure for individuals living in the vicinity of benzene
sources (assuming the several scenarios of living patterns discussed in
Chapter X) is shown in Table 1-2. When this exposure is compared with
those shown in Table 1-1, several significant differences are evident.
The levels to which more than 90% in the lowest range are exposed have
been increased sufficiently to place these people in the second range
because benzene exposure levels for most activities considered (i.e.,
shopping, commuting, working) are higher than 1.0 ppb. Therefore, the
weighted sum for all exposures is higher than the original estimate.
This analysis indicates that living patterns affect annual average
benzene exposures—generally by increasing those exposures because higher
benzene exposures occur in nonresidential settings.
As indicated above, the estimates given in this report are subject
to considerable uncertainty; they thus require further monitoring
and sampling data for a more complete assessment. Despite the insuffi-
ciency of data, however, the fact remains that the population exposed
is substantial. Potential health effects from the estimated exposures
will be addressed in another report being prepared by the EPA Cancer
Assessment Group.
-------
Table 1-2
SUMMARY OF ESTIMATED TOTAL EXPOSURES OF PEOPLE
RESIDING IN THE VICINITY OF ATMOSPHERIC BENZENE SOURCES
Vicinity of Residence
Chemical manufacturing
Coke ovens
Petroleum refineries
Urban areas
Number of People Exposed
Annual Average Benzene Concentrations (ppb)
0.1-1.0 1.1-4.0 4.1-10.0 > 10.0
3,900,000 3,100,000 200,000 80,000
200,000 100,000
3,250,000 1,750,000
110,000,000
Totalt
7,300,000
300,000
5,000,000
110,000,000
Comparison
Among
Sourcestf
(10^ ppb-person-year)
10.0
0.2
4.5
250.0
The term "total exposures" is used to mean the sum of an individual's exposure to atmospheric benzene
from a variety of activities during a year. This assumes that people spend part of their time away from
their residence, resulting in exposures to different benzene concentrations depending on their activity
(i.e., commuting to work, shopping, traveling on personal business). Nonurban exposures are not included
in this analysis but are expected to range from undetectable to 1.0 ppb.
Rounded to two significant figures.
median values shown in Table X-10 were used for this calculation instead of the mid-point of the
ranges. This allows a better comparison with Table 1-1.
Source: SRI estimates.
-------
II BENZENE IN THE ENVIRONMENT
A. Introduction
The primary objective of this study has been to rank specific benzene
emission sources by estimating the atmospheric exposure of the U.S.
population to benzene from each source. The ranking of sources will
serve as input for establishing regulatory priorities. Implied in this
analysis is the assumption that individuals living near the sources
spend 24-hours of each day in that location and they are only exposed to
emissions from that source. However, because benzene sources are numerous
and diverse, there is reason to believe that certain individuals have
significantly different exposures from those estimated by this method
and are exposed to emissions from more than one source. Therefore, after
public review and comment on the initial draft of this study, a second
objective was added to make first-cut estimates of total exposure to
individuals living in the vicinity of benzene sources. Total exposure
is the sum of the exposures to all benzene sources over a designated
period (e.g., a day or a week).
This study is one in a series that SRI is conducting for EPA to
estimate populations at-risk to selected pollutants. These studies are
generally conducted on a quick-response basis to provide input to other,
more inclusive studies. This study has not considered the degree of bio-
logical sorption of material. No attempt was required or has been made
in this input report to assess potential health effects.
Atmospheric sources of benzene are widespread and include natural
sources such as forest fires and man-made sources such as automobile
emissions. Although benzene is not sampled regularly in any air quality
monitoring program, some sampling data do exist. EPA has also conducted
dispersion modeling that is applicable to most of the major sources.
On the other hand, sample data of benzene concentrations in water, food,
and soil are sparse, and those measurements that have been taken have
been infrequent and inconsistent.
-------
Benzene is commercially produced mainly by petrochemical operations
(92%) and on a much smaller level as a coke-oven by-product (8%). Total
benzene production in 1976 was approximately 7500 x 10 Ib (3400 x 10 kg)
(SRI estimates). Benzene is a constituent of gasoline and crude oil.
Pure benzene is used primarily in chemical manufacturing processes, and
in solvent operations. Of all benzene used in chemical and solvent
operations, more than 97% is used in chemical processing (SRI estimates).
For this report, seven sources of atmospheric benzene were evaluated:
chemical manufacturing plants, coke ovens, self-service gasoline stations,
petroleum refineries, solvent operations, storage and distribution of
gasoline, and urban exposures related to automobile emissions and evap-
oration from gasoline stations. These sources have been identified as
the major sources of atmospheric benzene (PEDCo, 1977; Johnson, 1977).
Although oil spills and discharges represent a potentially significant
source of benzene in the environment, the most significant of these occur
in remote locations or along coastal areas where population density is
low, and the benzene released to the atmosphere from each occurrence is
very small. Potential human exposure to atmospheric benzene from oil
spills and discharges appears to be negligible.
Because few quantitative data were available for this study, all
estimates given here are subject to considerable uncertainty. This is
related to: quantity of benzene emissions, benzene production and
consumption level, source locations, control technology employed,
deterioration of control technology over time, and dispersion modeling.
The available monitoring data are summarized in the discussion of each
source; however, because these data are insufficient, the accuracy of
the monitoring results could not be quantitatively assessed. Quality
control information was unavailable for much of these data. Comparisons
of short-term ambient concentrations indicated by monitoring data with
annual average ambient concentrations estimated from dispersion modeling
are tenuous. In addition, meterological conditions and in-plant operating
characteristics during a sampling period may differ significantly from
-------
average conditions. Even with these uncertainties, a comparison of
monitoring data with dispersion modeling results suggests that the
agreement between the two is sufficient to lend support to the modeled
concentrations.
B. Nonatmospheric Benzene
It is not within the scope of this study to evaluate human exposures
to benzene from water, food, or other environmental pathways. However,
it is useful to review the available data to provide some basis for
comparison.
Only limited data on benzene in water are available. A review of
benzene sampling data by Howard and Durkin (1974) found that the few
freshwater samples analyzed by that time showed only trace levels of
benzene. For example, a 1972 EPA study cited in the report identified
53 organic chemicals, ranging from acetone to toluene, in the finished
waters and organic waste effluents in 11 plants (of 60 sampled) dis-
charging into the Mississippi River. Benzene was not detected in the
effluents, but the trace detected in the finished waters suggested
another source than effluent discharge.
A recent sampling of five benzene production or consumption plants
by Battelle (1977a) found benzene levels in water ranging from < 1.0 to
179 ppb (plant effluent). The concentrations at 13 upstream and down-
stream sample locations in nearby receiving waters, however, ranged
from < 1.0 to 13.0 ppb, with an average of 4.0 ppb.
A recent report by the National Cancer Institute (1977) noted
benzene levels of 0.1 to 0.3 ppb in four U.S. city drinking water
supplies. One measurement from a groundwater well in Jacksonville,
Florida showed levels higher than 100 ppb. No indication is given in
the report of the sampling methods or the analytical procedures. However,
study of the behavior of benzene in the groundwater system and in the
drinking water supply system is clearly warranted.
One possible source of benzene in the aquatic environment is from
cyclings between the atmosphere and water (Mitre, 1976). Benzene is
-------
fairly volatile (high vapor pressure of 100 mmHg at 26°C) and has a
relatively high solubility (1780 mg/L at 25°C). Consequently, it is
reasonable to believe that benzene will be washed out of the atmosphere
with rainfall and then evaporated back into the atmosphere, causing a
continuous recycling between the two media.
The distribution of benzene in the aquatic system is not well-known.
Neely et al. (1974) demonstrated a relationship between octanol-water
partition coefficients and bioaccumulation potential in fish. The
partition coefficient for benzene, which is estimated to be very low,
suggests that the bioaccumulation potential in fish is minimal. Benzene
uptake by aquatic vegetation has not been studied.
Only one study of benzene levels in soil has been conducted.
Battelle (1977a) sampled soils in the vicinity of five benzene consumption
or production facilities. Their preliminary results from 14 samples
showed levels ranging from <1.0 to 191.0 ppb, with an average of 53.0 ppb.
In most cases, the highest levels of benzene were found in samples taken
closest to the plant. These results indicate that the potential for
accumulation of benzene in the soil is significant.
Human exposure to benzene in food is not addressed in this report.
We note, however, the following information: those few available data
that quantify benzene levels in food (Chinn, personal communication,
1977) indicate that it occurs naturally in fruits, fish, vegetables, nuts,
dairy products, beverages, and eggs. However, data on concentrations
are only available for cooked meat, rum, and eggs (see Table II-l). A
report by the National Cancer Institute (1977) estimated that an indi-
vidual could ingest as many as 250 yg/day from these foods.
L = liter.
10
-------
Table II-l
ESTIMATED BENZENE LEVELS IN FOOD
(yg/kg)
Heat treated or canned beef 2
Jamaican rum 120
Irradiated beef 19
Eggs 2100
Source: National Cancer Institute (1977)
C. Chemical and Physical Properties of Benzene
Benzene, C,H,, is a nonpolar, nonreactive, highly refractive cyclic
oo o
aromatic hydrocarbon. In benzene, the C-C bond is 1.38 A long and the
o
CH bond is 1.08 A long (Ayers and Muder, 1964; MacKenzie, 1962). Under
standard conditions, benzene is a clear, noncorrosive, colorless, and
highly flammable liquid. Benzene possesses a characteristic odor,
similar to that of gasoline. It is relatively soluble in water and is
miscible with acetone, alcohol, chloroform, ether, carbon disulphide,
carbon tetrachloride, glacial acetic acid, and oils. Pertinent physical
properties of benzene are listed in Table II-2.
Benzene is quite thermodynamically stable because the resonance
energy of its unsaturated bonds is due to the interaction of the six IT
electrons that form "doughnut" shaped electron orbitals above and below
the plane of the ring.
Benzene solubility in water at 25°C is 1800 ppm (0.18 mg/g water).
Variation of benzene solubility in water from 1730 to 1800 ppm has been
noted (McAuliffe, 1963). The difference is believed to be attributable
either to the temperature of the experiment or the precision of the
technique. Both salting-in and salting-out (increase or decrease in
solubility) phenomena have been observed for benzene in aqueous solution
(Giacomelli, 1972). Benzene solubility in salt water and distilled water
11
-------
Table II-2
PROPERTIES OF BENZENE
Constant
Freezing point, °C
Boiling pointj °C
Density, at 25°C, g/mL
Vapor pressure at 26.075°C, mm Hg
Refractive index, njp
Viscosity (absolute) at 20°C, cP
Surface tension at 25°C, dyn/cm
Critical temperature °C
Critical pressure, atm
Critical density, g/mL
Flash point (closed cup), °C
Ignition temperature in air, °C
Flammability limits in air, vol%
Heat of fusion, kcal/mole
Heat of vaporization at 80.100°C, kcal/mole
Heat of combustion at constant pressure and
25°C (liquid C^H.^ to liquid H20 and
gaseous CCL) , kcal/g
Solubility in water at 25°C, g/100 g water
Solubility of water in benzene at 25C, g/100 g
benzene
Value
5.553
80.100
0.8737
100
1.49792
0.6468
28.18
289.45
48.6
0.300
-11.1
538
1.5-8.0
2.351
8.090
9.999
0.180
0.05
Source: Ayers and Muder (1964).
12
-------
have been compared, and the results show that solubility decreases as
the salt content of water increases (Button and Calder, 1974). A similar
decrease in solubility of the water soluble fraction (including benzene)
from crude oil was observed (Lee, 1974). These observations reveal that
benzene is less soluble in salt water than in fresh water.
The vapor pressure of benzene is an important property in assessing
the benzene contamination in the gaseous phase. The vapor pressure of
100 mmHg at 26°C indicates that benzene exists environmentally only in
the gaseous and the aqueous phases.
Benzene is highly stable'. Consequently, chemical reactivity is
limited unless the reactions take place under certain extreme conditions
(and in the presence of the necessary reagents). When chemical reactions
do take place, benzene behaves primarily as a nucleophilic agent,
usually with substitution of individual hydrogen atoms rather than addi-
tion. The two most common substitutive reactions are nitration and
sulfonation. In additive reactions, other reactive chemical agents are
added to the unsaturated bonds. Three types of additive reactions are
most common: oxidation, hydrogenation, and halogenation.
The general environmental fate of benzene can be assessed by exam-
ining the degradation processes of oxidation, hydrolysis, photolysis,
and microbial decomposition. Hydrolysis and microbial decomposition
occur primarily in the aqueous phase, whereas oxidation and photolysis
can occur in both the aqueous and the gaseous phases.
Benzene can be oxidized to a number of different products in the
presence of catalysts or at elevated temperatures and pressures. Under
extreme conditions, benzene has been observed to oxidize completely to
water and carbon dioxide. In the environment, such extreme conditions
rarely exist. Thus, it can be concluded that degradation of benzene by
oxidation is probably negligible. Oxidation in the emission pathways
from chemical plants and refineries is conceivable, but no such obser-
vations have been reported.
13
-------
The benzene ring does not undergo reaction with water or hydroxyl
ions (OH ) unless substituted with a significant number of strong electro-
negative groups, or at elevated temperature and pressure. Thus, hydrol-
ysis in the environment is assumed to be minimal.
Several studies have investigated the wavelength absorption proper-
ties of benzene. No appreciable amounts of light at wavelength longer
o
than 280 nm (2800A) were directly absorbed by benzene dissolved in
cyclohexane. A slight shift, however, in wavelength absorption would
be more representative of environmental media, such as dissolution in
water or adsorption on particular matter. Chien (1965) reported the
ultraviolet absorption spectra of liquid benzene in the presence of
oxygen under 1 atmosphere. Noyes et al. (1966) found that gaseous
benzene only absorbs light at 275 nm or less. Because the atmospheric
ozone layer effectively filters out wavelengths less than 290 nm, it
appears that direct excitative photolytic reaction of benzene in the
environment is unlikely, unless a substantial wavelength shift occurs
in the presence of other media. Indirect excitation of benzene may be
possible in the presence of certain sensitizers in the water or soil.
Photolysis by light with a wavelength of less than 290 nm of benzene
in the vapor phase and in oxygenated aqueous solution has been reported.
Two types of products, 2-formyl-AH-pyran and cyclopentadiene-carboxaldehyde
result (Luria and Stein, 1970; Kaplan et al., 1971). Matsuura and Omura
(1974) have reviewed several investigations where atomic oxygen that had
been photochemically generated from various sources reacted with benzene
to form phenol. Atomic oxygen is generated, for instance, from the
photodecomposition with nitrogen dioxide, which is frequently found in
high concentration in heavily polluted air (Altshuller, 1971). Labora-
tory results conclude that benzene is not completely inert under smog
conditions (Laity et al., 1973; Stephens, 1973).
The microbial degradation of benzene has received some attention in
recent years and it is conceivable that biodegradation of benzene probably
occurs under environmental conditions. Benzene has been found to bio-
degrade in a waste treatment plant, with the rate of degradation
14
-------
determined by the incubation period and acclimation of the microorganisms.
It is safe to conclude, therefore, that benzene can be degraded—but at
a very slow rate.
In summary, oxidation and hydrolysis of benzene in the environment
are unlikely. Photolysis is possible in the natural environment, but
the photolysis rate depends on wavelength adsorption and the presence
of sensitizers. In a heavily polluted atmosphere, atomic oxygen may
cause photochemical decomposition of benzene. Biodegradation of benzene
in the environment is also possible, but the degradation rate is quite
slow.
15
-------
Ill CHEMICAL MANUFACTURING FACILITIES
A. Sources
In this section, generation of benzene emissions from the manufac-
turing of chemical compounds will be addressed. Producer companies of
various compounds (excluding solvents) are listed in Table III-l; their
locations and 1976 capacity productions are also included in the table.
The Gulf Coast has the highest density of these benzene-consumption
facilities.
Benzene is used commercially as an intermediate agent in the pro-
duction of many chemical compounds. The emissions of benzene from such
industrial uses are potentially significant sources of atmospheric benzene.
Total U.S. consumption of benzene in 1975 was 108.4 x 10 gal (4.1 x
10 m ) (Anderson, 1976). Figure III-l illustrates the benzene deriva-
tives and their uses. Primary use involves the manufacture of such chemi-
cals as nitrobenzene, ethylbenzene, maleic anhydride, cumene, phenol,
chlorobenzene, cyclohexane, and detergent alkylate. Appendix A contains
flow diagrams for some of these processes.
Some atmospheric monitoring data were collected in the vicinity
of five benzene-consumption facilities by Research Triangle Institute
(RTI) (1977). Grab samples were collected for periods ranging from
25 minutes to 12 hours, both on the plant property and within 1 km of
the plant boundary. Benzene concentrations ranged from 0.5 to 58.4 ppb.
The average benzene concentration for each location is shown in Table III-2.
Complete quality control information is unavailable for these samples.
Emissions from chemical manufacturing facilities contain many chemical
compounds in addition to benzene; some of these compounds have demon-
strated toxic and/or carcinogenic properties. Although few data exist
about the composition of emissions from various processing facilities,
Table III-3 contains a partial list of the major emissions. Individuals
17
-------
Table 111-1
LOCATIONS AND CAPACITIES OF PLANTS USING BENZENE AS AN INTERMEDIARY
AGENT IN THE MANUFACTURE OF VARIOUS CHEMICAL COMPOUNDS*
oo
STATE
ALABAMA
CALIFORNIA
DELAWARE
GEORGIA
ILLINOIS
LOCATION
TUSCALOOSA
CARSON
EL SEGUNOO
IRWINDALE
RICHMOND
SANTA FE SPRINGS
DELAWARE CITY
CAHTEHSVILLE
BLUE ISLAND
CICERO
COMPANY
HEICHHOLD CHEM., INC.
WITCO CHEM.
STD. OIL CO. OF CALIF.
SPECIALTY OHGANICS. INC.
STD. OIL CO. OF CALIF.
FERHO CORP.
STO. CHLORINE CHEM CO.. INC.
CHEM. PRODUCTS CORP
CLARK OIL & REFINING
KOPPEfiS CO.. INC.
MORRIS REICMHOLO CHEM.. INC.
SAUGET
KANSAS EL DORADO
MONSANTO
SKELLY OIL CO.
KENTUCKY ASHLAND ASHLAND OIL. INC.
LOUISIANA BATON ROUGE FOSFCR GRANT CO.
CAHVILLE COS-MAR. INC.
CHALMETTE TENNECO, INC.
GEISMAR RUBICON CHEM.. INC.
PLAQUEMINE GEORGIA PACIFIC CORP.
WELCOME GULF OIL CORP.
MARYLAND BALTIMORE CONTINENTAL OIL CO.
MASSACHUSETTS MALOEN SOLVENT CHGM. CO.. INC.
MICHIGAN
MISSISSIPPI
MISSOURI
NEVADA
NEW JERSEY
MIDLAND DOW CHEMICAL
PASCAGOULA FIRST MISSISSIPPI CORP.
ST. LOUIS MONSANTO
HENDERSON MONTROSE CHEM.CORP. OF CAL.
BOUND BROOK AMERICAN CYANAMID
BOUND BROOK UNION CARBIDE
ELIZABETH REICHHOLO CHEM., INC.
FORDS TENNECO, INC.
GIBBSTOWN £. 1. du PONT
KEAflNY STO. CHLORINE CHEM. CO.
WESTVILLE TEXACO, INC.
CAPACITY PRODUCTION JANUARY 1. 1976 (million* of kg!
NITRO-
BENZENE
6
34
61
38
91
ANILINE
26
46
27
69
ETHYL-
BENZENE
440
327
12
260
260
STYHENE
372
272
238
182
MALEIC
ANHYDRIDE
6
27
48
14
12
CUMENE
46
60
61
160
4.6
118
PHENOL
68
26
N.A.
40
63
120
18
68
MONO-
CHUOHO
BENZENE
34
62
N.A.
136
32
DICHLORO-
BENZENE
IO- «nd PI
1
27°
10'
13
l"
»
7*
CYCLO-
HEXANE
DETERGENT
ALKYLATE
(Un..
•nd Branch)
26
100
98
-------
Table 111-1 (Continued)
vo
STATE
NEW YORK
OHIO
PENNSYLVANIA
PUERTO RICO
TEXAS
LOCATION
NIAGARA FALLS
NIAGARA FALLS
NIAGARA FALLS
SYRACUSE
HAVERHILL
BEAVER VALLEY
BRIDGEVILU
CLAWTON
FHANKFORO
NEVILLE ISLAND
PHILADELPHIA
CUAYAMA
PENUELAS
PENUELAS
BAYTOWN
BEAUMONT
BEAUMONT
BIG SPRING
eORGEH
CHOCOLATE BAYOU
CORPUS CHRISTI
CORPUS CHRISTI
CORPUS CHRISTI
FREEPORT
HOUSTON
HOUSTON
HOUSTON
HOUSTON
HOUSTON
ODESSA
OYSTER CREEK
PHILLIPS
PORT ARTHUR
PORT ARTHUR
PORT ARTHUR
SEADRIFT
SWEENEY
COMPANY
ICC INDUSTRIES. INC.
OCCIDENTAL PETROLEUM
SOLVENT CHEM. CO.
ALLIED CHEM. CORP.
UNITED STATES STEEL
AHCO/POLYMERS. INC.
KOWEHS CO., INC.
UNITED STATES STEEL
ALLIED CHEMICAL CORP.
UNITED STATES STEEL
GULF OIL CORP.
PHILLIPS PETROLEUM
COMMONWEALTH OIL
UNION CARBIDE CORP
EXXON CORP
E. 1. du PONT
UNION OIL CO. OF CALIFORNIA
AMERICAN PETROFINA
PHILLIPS PETROLEUM
MONSANTO
COASTAL STATES GAS
SUN OIL CO.
UNION PACIFIC CORP
OOW CHEMICAL
ARCO/POLYMERS, INC.
THE CHARTER CO.
JOE OIL. INC.
THE MERICHEM CO.
PETRO-TEX CHEM CORP.
EL PASO NATURAL GAS
DOW CHEMICAL
PHILLIPS PETROLEUM CO.
ARCO/POLYMERS, INC.
GULF OIL CORP
TEXACO
UNION CARBIDE CORP
PHILLIPS PETROLEUM CO.
CAPACITY PRODUCTION- JANUARY 1, 1976 (millions of kg)
NITRO-
BENZENE
141
ANILINE
91
ETHYL-
BENZENE
73
JO
43
M8
46
1»
lit
100
166
STYRENE
200
41
36
649
46
6B
138
MALEIC
ANHYDRIDE,
16
IB
23
CUMENE
206
280
286
64
114
206
118
PHENOL
90
MONO-
CHLOHO-
BENZENE
N.A.
7
N.A.
11
N.A.
260 i
!
90
J27
N.A.
182
DICHLORO-
BENZENE
10- «nd P-l
N.A.
9
9
CYCLO-
HEXANE
98
209
118
118
100
36
118
66
208
261
DETERGENT
ALKYLATE
ILinur
•nd Branch)
102
-------
Table 111-1 (Concluded)
STATE
TEKAS
WEST VIRGINIA
WASHINGTON
LOCATION
TEXAS CITY
TEXAS CITY
TEXAS CITY
CHARLESTON
FOLLANSBEE
MOUNDSVILLE
NATRIUM
NEW MARTINSVILLE
WILLOW ISLAND
ANACORTES
KALAMA
COMPANY
MARATHON OIL CO.
MONSANTO
STANDARD OIL IINOIANA)
UNION CARBIDE CORP
KOPPERS CO.. INC.
ALLIED CHEM CORP.
PPG INDUSTRIES. INC.
MOI3AY CHtM CORP.
AMERICAN CYANAMIOE
STIMSON LUMBER CO.
KALAMA CHEMICAL
TOTAL
CAPACITY PRODUCTION JANUARY 1, 1976 Imllllom of kg)
NITRO
BENZENE
26
61
27
483
ANILINE
46
22
314
ETHYL-
BENZENE
1460C
430
3884
STYRENE
580
382
3211
MALEIC
ANHYDRIDE
27
188
CUMENE
88
28
1720
PHENOL
N.A.
26
1262
MONO-
CHLORO-
BENZENE
N.A.
41
313
DICHLORO-
8ENZENE
10- >nd P-l
23
120
CYCLO-
HEXANE
2706
DETERGENT
ALKYLATE
ILinnr
xd Brtnchl
68
3>3
NJ
o
SOURCE SRI. 1976 DIRECTORY OF CHEMICAL PRODUCERS. « dud In PEDCO. 1877
N.A. - NOT AVAILABLE
I. PRODUCTION CAPACITY FOR ODICHLOROeENZENE ONLY
b. PRODUCTION CAPACITY FOR P DICHLOHOBENZENE ONLY
c. 1978 DATA SHOWED COMBINED ESTIMATES OF ETHYLBENZENE PRODUCTION AT CHOCOLATE BAYOU. TEXAS AND AT TEXAS CITY. TEXAS.
1977 SRI ESTIMATES SHOW ETHYLBENZENE PRODUCTION ONLY AT THE TEXAS CITY PLANT.
-------
ETHYL \ / CYCLO
BENZENE / V HEXANE
CHLOHO-
BENZENE
DETERGENT
ALKYLATE
PHENOLIC \/ SURF AC
ALKLD. VPOLYESTER
INSECT.- V CARBON. U EPOXY
MOLDINGS I FIBERS
OIL fcETERGEN
THANES / \ADDITIVES
RESINS A TANTS
RESINS A RESINS
LAMINATES I ADHESIVES I COATINGS I MOLDINGS I FIBERS I LAMINATES I ADHESIVES I MOLDINGS I COATINGS
SOURCE: HEDLEY, 1975
FIGURE 111-1. BENZENE DERIVATIVES AND THEIR USES
-------
residing in the vicinity of chemical manufacturing facilities are,
therefore, exposed to a wide variety of chemical compounds in addition
to benzene.
Table III-2
ATMOSPHERIC BENZENE CONCENTRATIONS SAMPLED
AT BENZENE-CONSUMPTION FACILITIES
Location
S. Charleston, WV
Freeport, TX
La Porta, TX
Bound Brook, NJ
Fords, NJ
Geismar, LA
Plaquemine, LA
Company
Union Carbide
Dow
Dupont
American Cyanamid
Tenneco
*
Rubicon
*
Georgia Pacific
Number
of
Samples
3
3
'3
1
1
6
11
Average
Sampling
Time
(min)
100
100
110
44
44
740
660
Average
Benzene
Concentration
(ppb)
34
14
0.8
2.8
0.9
0.6
0.9
•K
Samples taken in the vicinity of industrial complexes that have known
benzene-consumption facilities.
Source: RTI (1977).
To assess the ambient benzene concentrations in the vicinity of
chemical manufacturing facilities, two factors must be estimated:
benzene emission rates at each location; and atmospheric dispersion of
benzene in the vicinity of the plants. The emission rates can be
estimated if the emission factors and total production are available.
Table III-4 gives the emission factors used in the analysis and emission
characterization. The emission factors were selected to represent aver-
ages. Because little is known about benzene emissions from chemical
manufacturing facilities, these emission factors are considered order-
of-magnitude estimates. Maleic anhydride and aniline have the highest
emission factors related to the specific manufacturing processes and
reaction kinetics of each compound.
22
-------
Table III-3
MAJOR CHEMICAL COMPOUNDS OTHER THAN
BENZENE EMITTED FROM CHEMICAL MANUFACTURING FACILITIES
Process
Reference
Compound
Aniline
Chlorobenzene
Ethylene/styrene
Linear alkylate
Maleic anhydride
Nitrobenzene
Aniline
Nitrobenzene
Chlorobenzene
Dichlorobenzene
Trichlorobenzene
o-, m-, and p- dichlorobenzene
Chlorine
Hydrochloric acid
Ferric chloride
Methane
Ethane
Ethylene
Hydrochloric acid
Toluene
Ethylbenzene
Styrene
di- and triethyl benzenes
n parafins (CIQ - Cl4 range)
Allyl chloride
Linear alkyl benzene
Hydrochloric acid
Olefins
Linear alkylate
Hydrogen fluoride
Maleic acid
Formaldehyde
Formic acid
Acjid aldehyde
Xylene
Nitrobenzene
Dinitrophenol
Other dinitro and trinitro compounds
1. Weber, personal communication, 1978.
2. Beck, personal communication, 1978.
3. Mascone, personal communication, 1978.
4. Schumaker, personal communication, 1978.
23
-------
Table III-4
EMISSION FACTORS AND CHARACTERIZATIONS
FOR BENZENE-CONSUMPTION PLANTS
Chemical
Aniline
b
Cumene
a
Cyclohexane
Detergent alkylate
(linear and branched)
Dichlorobenzene
(p- and o-)b
Ethylbenzene
Maleic anhydride
Monochlorobenzene
Nitrobenzene
Phenol
b
Styrene
Emission Factor
(10~3 kg of benzene/kg
of product)
23.60
0.25
2.80
2.20
8.60
0.62
96.70
3.50
7.00
1.00
1.50
Emission
Characterization
Fugitive
Fugitive
Fugitive
Fugitive
Chlorinator, PDCD
recovery system
Scrubber-vent
Product recovery
scrubber
Unknown
Point absorber
Unknown
Collection vent,
emergency vent
SRI estimates.
PEDCo estimates.
24
-------
*
The atmospheric dispersion of benzene is more difficult to assess.
Simply, source characteristics (e.g., stack dimensions) and meteorological
conditions greatly influence the dispersion of benzene in the vicinity
of the plants. Youngblood (1977a, 1978) made rough dispersion estimates
from very limited data on source characteristics. He classified the
processes according to three source categories: A—ground-level point
source (effective stack height, 0 m); B—building source (effective stack
height, 10 m); and C—elevated point source (effective stack height,
20 m). Emission rates were then calculated for each process by assuming
a maximum production rate. Ambient ground-level concentrations were
derived manually from Turner's workbook. One-hour worst-case concentra-
tions were derived with the following meteorological conditions assumed:
wind speed, 4 m/s; stability class, neutral (Pasquill Gifford "D"). For
source category B, the results from Turner's workbook were adjusted to
account for the initial dispersion of the pollutant in the building
cavity. The one-hour estimates were then converted to 8-hour worst-
case estimates (by multiplying by 0.5). The results of the dispersion
modeling by Youngblood are given in Table III-5.
B. Methodology
Each chemical manufacturing plant has different production rates,
chemical processes, geographic locations, pollution control technology,
and meteorological conditions. Thus, detailed dispersion calculations
are impractical, given the scope of the study. A simple method of
assessment was therefore developed to allow for comparative analysis.
Variations in geographic locations and meteorological conditions were not
considered in the analysis. The results are not precise: rather, they
provide a reasonable estimate of atmospheric benzene concentrations.
*
Battelle-Columbus has monitored benzene concentrations in the vicinity
of chemical manufacturing facilities. These data are now in draft
form and should be available in the near future.
25
-------
Table III-5
ROUGH ESTIMATES OF AMBIENT GROUND-LEVEL BENZENE CONCENTRATIONS (8-HOUR AVERAGE)'
K)
ON
Source
Maleic anhydride
Styrene
Phenol from cumene
Benzene
Cumene
Phenol from benzene
Nitrobenzene
Ethyl benzene
Phenol from toluene
Chlorobenzene
o- dichlorobenzene
p-dichlorobenzene
Emission
Rate
(R/s)
139.0
7.49
10.8
0.179
2.34
0.0691
31.20
16.60
2.42
15.10
3.60
6.20
Concentration (pg/m )
Source
Category
C
A
B
C
A
B
A
B
A
B
C
A
B
C
A
B
A
B
A
B
C
A
B
C
A
B
C
150 m
700
3800
850
54
90
20
1200
260
35
8
<1
16000
3500
160
8500
1900
1200
270
7700
1700
76
1800
400
18
3200
700
31
300 m
5000
1100
460
390
26
11
340
140
10
4
2
4500
1900
1100
2400
1000
350
150
2200
940
540
500
220
130
900
380
220
450 m
5000
530
290
390
13
7
170
91
5
3
2
2200
1200
1100
1200
650
170
94
1100
590
540
250
140
130
440
240
220
at Given Distance
600 m
3900
330
210
300
8
5
100
66
3
2
2
1400
870
870
730
460
110
68
660
420
420
160
100
100
270
180
170
750 m
2900
220
160
230
5
4
70
49
2
1
1
940
650
650
500
350
73
51
453
320
320
110
75
75
190
130
130
1600 m
1100
68
55
89
2
1
21
17
1
<1
<1
280
230
250
150
120
21
17
140
110
120
32
26
28
39
46
49
This is a worst-case estimate. It may be multiplied by 0.04 to give rough estimates of annual average
concentrations.
Key to Source Categories: A—ground-level point source: B—building source; C—elevated point source.
Source: Youngblood, 1977a.
-------
A single dispersion curve was constructed and applied to all chemical
manufacturing facilities, based on their emission rates. The derivation
of this methodology is discussed below.
As shown in Table III-5, ambient benzene concentrations in the
vicinity of chemical manufacturing plants vary significantly in relation
to the characteristics of the emission sources. Exhaust gas temperature,
which is important in determining near-source concentrations, was not
considered. Because of the generally high concentrations estimated at
1.6 km (the maximum distance modeled in that analysis), Youngblood (1977b,
1978) extended his model calculations to a distance of 20 km with an
emission rate of 100 g/s for each source category (see Table III-6).
The results of Youngblood's analysis are shown in Figure III-2.
The ground-level (A) and building (B) sources are highest near the plant
and decrease rapidly with distance. The elevated point source (C), however,
shows low initial concentrations that increase to a peak followed by a
decline. Although the differences due to source category are considerable
at 150 m, the differences decrease rapidly with distance. Even as close
as 300 m, the differences are within the range of uncertainty normally
associated with dispersion calculations. In addition, distances less
than 300 m are likely to be within the plant perimeter or to have low
population densities. A single dispersion curve (Curve M in Figure III-2)
was therefore developed to represent all three source categories, as
suggested by Youngblood (1977b). This curve was derived by averaging the
high and low values of the three emission source categories at each
calculated distance. The resulting concentrations estimated by this
method are shown in Table III-7.
27
-------
N)
oo
Table III-6
ROUGH ESTIMATES OF AMBIENT GROUND-LEVEL BENZENE CONCENTRATIONS
(8-HOUR-AVERAGE)* PER 100 g/s EMISSION RATE
Source
Category
0.15
km
A 51,000
B 11,000
C 510
Concentrations
0.3
km
14,000
6,100
3,500
0.45
km
7,000
3,800
3,500
0.6
km
4,500
2,800
2,800
0.75
km
3,000
2,100
2,100
(yg/m )
1.6
km
900
740
800
2.5
km
440
370
410
4.0
km
220
220
220
6.0
km
120
120
120
9.0
km
62
62
62
14.0
km
34
34
34
20.0
km
20
20
20
To give rough estimates of annual average concentrations, multiply by 0.04; to convert
to ppb, divide by 3.2.
Source: Youngblood (1977b, 1978).
-------
10s
n
a.
V)
O
GC
UJ
O
O
o
uj
z
UJ
N
UJ
CD
10'
10
TT
/
/
/
/
/ c
A • GROUND LEVEL SOURCE
B • BUILDING SOURCE
C - ELEVATED SOURCE
M - AVERAGE OF CURVES A, B, AND C
0.1
100
1.0 10
DISTANCE FROM SOURCE - km
'Baud on an emtolon rate of 100 gf*
Source: After Youngbtood, (1977b)
FIGURE III-2. DISPERSION MODELING RESULTS FOR EACH TYPE OF SOURCE CATEGORY'
29
-------
Table III- 7
*
ESTIMATES OF 8-HOUR WORST CASE BENZENE CONCENTRATIONS
BASED ON AVERAGE OF THREE EMISSION SOURCE CATEGORIES
3 t
Distance (km) Concentration (yg/m )
0.30 8800
0.45 5200
0.6 3600
0.75 2600
1.6 820
2.5 400
4.0 220
6.0 120
9.0 62
14.0 34
20.0 20
*
To convert to annual average estimates, multiply
concentrations by 0.04.
To convert to ppb, divide concentrations by 3.2.
Source: Youngblood (1977b, 1978).
Regression analysis was used to develop an equation to characterize
the single dispersion curve (Curve M). Equation (3.1) was derived from
that analysis:
C = 1648 D'1'48 (3.1)
3
where, C is the 8-hour worst-case benzene concentration in yg/m , and
D is the distance from the source in km. Because Equation (3.1) is
only valid for an emission rate of 100 g/s, a normalized equation is
given as follows:
C = 16.48 E D~1-48 (3.2)
where, E in g/s, is the emission rate for the location of interest.
3.
30
-------
The annual average concentration can be estimated by including a
multiplier of 0.04 in the equation. Thus, the equation becomes:
C = 0.659 E D~1<48 (3.3)
a
In this study, the ranges of benzene concentrations that follow and
that apply to all sources have been established for the sake of uni-
formity:
0.1 - 1.0 ppb
1.1 - 4.0 ppb
4.1 - 10.0 ppb
> 10.0 ppb
A computer program was developed to estimate the people exposed to
concentrations within each range at each location. Equation (3.3) was
rearranged as follows to determine the distance at which the specified
concentrations are found:
/ E \ 0.6757
D = 0.754 -r*- (3.4)
\ *• I
where, C. is the specified annual average concentration (i.e., 0.1, 1.0,
1 3
4.0, and so on; input data, however, are in ug/m ); D. is the distance
in kilometers at which the specified concentration is found; and E is
a
the emission rate in grams per second at that location.
The population residing within a circle of radius D. was then
estimated by the following equation:
P± = d TT D^ (3.5)
where, d is the city or state population density, and P. is the popu-
lation exposed to concentration C or greater.
31
-------
The main assumptions in this analysis are that:
The benzene source is in the center of the city (if the city has
a population greater than 25,000).
The maximum allowable radius is 20 km.
The population density is uniform over the exposed area.
When a city has more than one plant, the plants are
colocated, and their corresponding emission rates are summed.
If the city has a population of less than 25,000, state
density is used.
To accommodate these assumptions, the following steps were included in
the computer program. The radius of each city was determined by
Equation (3.6):
/ \
1/2
(3.6)
where, D is the estimated radius of the city; P is the population of
the city (1970 Bureau of Census data) ; and d is the average city density
(1970 Bureau of Census data available for cities of population greater
than 25,000).
When D. calculated from Equation (3.4) is greater than D , or when
no city density is available, Equation (3.7) is substituted for Equation
(3.5) to calculate the exposed population on the basis of state density.
Pi = Pc+ ds U (Di - Dc> (3'7)
where, d is average state population density; D. is the distance at which
S 1
concentrations C. is found; D is the radius of the city calculated in
Equation (3.6); and P. is the population exposed to concentration C or
greater. P and D equal 0 when no city density is available.
Because the dispersion modeling results are unverified at distances
greater than 20 km from the source location, the computer program auto-
matically cut off calculations when (1) a distance of 20 km was attained
and calculated the concentration (C.) at 20 km, or when (2) the estimated
benzene concentration dropped below 0.1 ppb.
32
-------
The cumulative population totals resulting at distance D. were then
automatically subtracted from those at distance D. .. , and the total
population within each range of concentrations was printed out. For
example, for range 0.1 to 1.0 ppb, the program subtracted P _ (a smaller
number) from P (a larger number)- In other words, P is the popu-
U • -L U • X
lation exposed to concentrations of 0.1 ppb or greater; P.. n is the total
population exposed to concentrations of 1.0 ppb or greater. By subtracting
the two values, the total population exposed to concentrations between
0.1 and 1.0 ppb is determined.
For locations with fewer than three chemical manufacturing facilities,
2
an area of 0.5 km (a circular area with a radius of 400 m) was assumed
to be within the plant boundary. No exposed population was estimated
within this area. For locations with three or more facilities, an area
2
of 0.8 km (a circular area with a radius of 500 m) was assumed to be
within the plant boundary.
Emission rates were estimated for each plant, based on the production
estimates contained in Table III-l and the emission factors in Table III-4.
Because actual production data are unobtainable, capacity production and
24-hour (365 days) operation were assumed. Appendix B, Table B-l, lists
the estimated emission rates for each chemical manufacturing facility.
The emissions of benzene from chemical manufacturing facilities were
assumed to be the sole contributors of benzene to the atmosphere in the
vicinity of the facility.
C. Exposures
Ambient benzene concentrations and the exposed population for each
source location were estimated, based on the methodology described
above. Table B-2 in Appendix B presents the results of this analysis
for each facility location.
Table III-8 presents the estimated population exposed to specified
levels of atmospheric benzene for each state. More than 7 million people
are exposed to annual average benzene concentrations of 0.1 ppb or greater.
The largest number of those exposed is found in Pennsylvania, Texas, and
33
-------
Table III-8
ESTIMATED POPULATION EXPOSED TO BENZENE FROM
CHEMICAL MANUFACTURING FACILITIES, BY STATE
*
Population Exposed to Benzene (ppb)
State 0.1-1.0 1.1-4.0 4.1-10.0 > 10.0
Alabama
California
Delaware
Georgia
Illinois
Kansas
Kentucky
Louisiana
Maryland
Massachusetts
Michigan
Mississippi
Missouri
Nevada
New Jersey
New York
Ohio
Pennsylvania
Texas
Washington
West Virginia
Puerto Rico
20,000
110,000
20,000
7,000
180,000
4,000
20,000
140,000
500,000
3,000
70,000
20,000
20,000
10,000
1,100,000
180,000
3,000
2,200,000
1,500,000
1,000
110,000
200,000
500
4,400
1,500
200
86,000
80
-
42,000
15,000
-
26,000
7,000
400,000
400
110,000
7,000
100
350,000
200,000
-
8,500
22,000
_
400
200
-
13,000
-
-
5,600
1,000
-
3,000
1,000
100,000
-
23,000
-
-
40 , 000
32,000
-
630
4,600
_
-
-
-
1,700
-
-
1,100
-
-
400
30
50,000
-
8,800
-
—
13,000
7,700
-
-
80
Total Exposed
Populationf 6,000,000 1,000,000 200,000 80,000
A
Totals for each state are rounded to two significant figures.
Annual average concentrations; to convert to 8-hour worst case,
multiply by 25; to convert to yg/m3, multiply by 3.2; a dash (-)
signifies that no exposed population was estimated by our method
for the annual average concentrations listed. There may be some
population exposed to those concentrations for shorter periods
of time.
'Totals are rounded to one significant figure.
Source: SRI estimates.
34
-------
New Jersey. These three states account for 75% of the total population
exposed to benzene from chemical manufacturing facilities.
Chemical manufacturing facilities are responsible for all point-
source exposures to the general population greater than 4.0 ppb. Because
our computer program is limited to estimating exposed population rather
than specific concentrations, predicted concentrations must be calculated
separately. To provide an upper boundary for our exposure estimates, the
predicted dispersion curves for the four largest uncontrolled chemical
manufacturing facilities are shown in Figure III-3. The facilities
were selected for two reasons: (1) a high benzene emission rate (greater
than 50 g/s, see Table B-2, Appendix B); (2) a large exposed population
estimated at high concentrations (Reichold Chemicals, Elizabeth, New
Jersey). A plant boundary of 450 m from the source was assumed. The
highest concentrations are found in the vicinity of the Monsanto maleic
anhydride plant in St. Louis. Annual average concentrations are in
excess of 10 ppb as far as 2 km from the source. At the plant boundary,
the estimated annual average concentration is 96 ppb. For comparison,
estimated urban annual average concentrations related to automobile
emissions generally range between 1 and 4 ppb. The lowest concentrations
of those facilities plotted are found in the vicinity of the Reichold
Chemicals maleic anhydride plant in Elizabeth, New Jersey. Those annual
average concentrations range from 28 ppb at the plant boundary to 0.1 ppb
at 20 km from the source.
Short-term high level exposures may also be important in affecting
*
risk to an individual. Figure III-4 compares annual average with 8-hour
worst case concentrations. Note that both facilities show 8-hour worst
case concentrations above 1.0 parts per million (ppm) outside the plant
boundary, with 1-hour peak levels a factor of 10 higher than the 8-hour
concentrations. For the two facilities shown on Figure III-4, the 1-hour
*
The assumption of a linear dose-response, however, requires that higher
concentrations for short periods have no higher risk associated with
them.
35
-------
200
100
_
a
a
O
z
LU
CJ
O
O
HI
z
LU
N
LU
CO
10
Monsanto—St. Louis, Missouri:
maleic anhydride
• -O Monsanto—Texas City. Texas:
ethylbenzene, styrene
•-—• U.S. Steel—Neville Island, Pennsylvania:
maleic anhydride
• —* Reichold Chemicals—Elizabeth, New Jersey
maleic anhydride
I I I I
DISTANCE FROM SOURCE - km
Assumed
** The four facilities having the largest benzene emissions and/or the largest number of exposed
population at high concentrations.
Source: SRI estimates based on dispersion modeling.
FIGURE III-3. PREDICTED ANNUAL AVERAGE BENZENE CONCENTRATIONS IN THE
VICINITY OF SELECTED** CHEMICAL MANUFACTURING FACILITIES
36
-------
5,000
1,000
100
a:
ui
o
o
o
ui
z
ui
N
UI
CD
10.0
1.0
0.1
i r
IT
MONSANTO-ST. LOUIS, MISSOURI:
MALEIC ANHYDRIDE
REICHOLD CHEMICALS-ELIZABETH,"
NEW JERSEY: MALEIC
ANHYDRIDE
OSHA 8-HOUR STANDARD
8-HOUR —\
WORST CASE
\
^Source: SRI ESTIMATES BASED ON
^^ DISPERSION MODELING
\
ANNUAL"
AVERAGE
\
0.1
*
Assumed
1.0
DISTANCE FROM SOURCE-km
10
30
FIGURE III-4. COMPARISON BETWEEN PREDICTED ANNUAL AVERAGE AND
8-HOUR WORST CASE BENZENE CONCENTRATIONS IN THE
VICINITY OF TWO CHEMICAL MANUFACTURING FACILITIES
37
-------
peak concentrations would range from 41,000 to 12,000 ppb (41 to 12 ppm)
at the plant boundary to 100 to 25 ppb 20 km from the source.
38
-------
IV COKE OVENS
A. Sources
In 1975, 57.2 x 10 tons of coke were produced in the United States.
The yield of coke from coal, averaging 68.4% in 1975, has remained fairly
constant during the past decade (Sheridan, 1976). Coke is produced by
65 plants in the United States (Suta, 1977). The 65 plants, which are
listed in Appendix C, consist of an estimated 231 coke-oven batteries
containing 13,324 ovens. Their theoretical maximum annual productive
capacity is 74.3 x 10 tons. Table IV-1 shows the estimated size and
productive capacity in each state.
Coke ovens producing benzene as a by-product account for about 5 to
8% of the total benzene production in the United States. About 0.66%
by volume benzene, 0.13% toluene, 0.05% xylene, and less than 0.10%
of other aromatics have been identified in the coal gas generated from
coking operations (Faith et al.,1966). The higher the temperatures in
coking operations, the larger the amounts of aromatic hydrocarbons
produced, particularly benzene. Reduction in quantities of paraffinic
naphthenic (saturated alicyclic) and unsaturated hydrocarbons in the
production is observed at high temperatures (Faith et al.j!966;
McGannon, 1970). Carbonizing 1 ton of coal in coke ovens to produce
blast furnace coke yields 3 to 4 gallons of light oil. The principal
constituent of this oil is benzene, which comprises about 60 to 80% of
the total composition. This crude light oil is then distilled to produce
benzene, toluene, and xylene. The typical amount of benzene recovered
from coke-oven gas is 1.85 gal/ton of coal carbonized (U.S. Public Health
Service, 1970).
The distillation of coal tar is one additional source of benzene
production. The amount of benzene produced varies with the coking and
recovery processes and the grade of the raw coal. In general, the light
39
-------
Table IV-1
ESTIMATED SIZE AND PRODUCTIVE CAPACITY OF BY-PRODUCT COKE PLANTS
IN THE UNITED STATES ON DECEMBER 31, 1975
State
Number of
Plants*
Alabama
California
Colorado
Illinois
Indiana
Kentucky
Maryland
Michigan
Minnesota
Missouri
New York
Ohio
Pennsylvania
Tennessee
Texas
Utah
West Virginia
Wisconsin
Undistributed
Total 62 (65)"
Included in Undistributed.
Colocated plants.
Source: Sheridan (1976).
Number of
Batteries
7
1
1
4
6 (7)*
1
1
3
2
1
3
12
12 (13)*
1
2
1
3 (4)*
1
28
7
4
9
31
2
12
10
5
3
10
35
51
2
3
4
13
2
231
Number of
Ovens
1,401
315
206
424
2,108
146
758
561
200
93
648
1,795
3,391
44
140
252
742
100
13,324
Maximum Annual
Theoretical
Productive
Capacity (tons)
6,961,000
1,547,000
1,261,000
2,523,000
11,925,000
1,050,000
3,857,000
3,774,000
784,000
257,000
4,053,000
9,960,000
18,836,000
216,000
839,000
1,300,000
4,878,000
245,000
74,266,000
Coke
Production
in 1974
(tons)
5,122,000
1,912,000
9,073,000
C1)
3,259,000
<»>
8,842,000
16,318,000
(')
(O
3,555,000
12.656.000
60,737,000
-------
oil distilled from coal tar is added to the major portion of light oil
recovered from coal gas and refined for its benzene content. Appendix A
contains a diagram of a typical coke-oven operation.
The basic coke-oven sources of air pollutant emissions include
charging and topside emissions, emissions from doors during the coking
cycle, waste gas stack emissions, pushing emissions, and quenching emissions,
The only benzene concentration data available are occupational exposure
data. Table'IV-2 gives the typical benzene concentration ranges per
occupation, within a coal-derived benzene recovery plant (NIOSH, 1974).
Measurements of benzene in Czechoslovakia coke-oven plants are tabulated
in Table IV-3. In the recovery plant, the benzene concentration can
3
reach as high as 145 mg/m .
Table IV-2
AMBIENT LEVELS OF BENZENE WITHIN A COAL-DERIVED
BENZENE PRODUCTION PLANT
8-hour
Time-Weighted
Average Range
Occupation (ppm) (ppm)
Agitator operator 6.0 0.5-20
Benzene loader and
loader helper 4.0 0.5-15
Benzene still operator 4.0 1-15
Light oil still
operator 2.5 1-15
Naphthalene operator 10 2-30
Analyst 10 4-30
Chemical observer 10 4-50
Foreman 1.5 1-10
Source: Bethlehem Steel Corporation data (NIOSH, 1974)
41
-------
Table IV-3
ATMOSPHERIC BENZENE EMISSION FROM THE COKING AND
RECOVERY PLANTS IN CZECHOSLOVAKIA
Benzene Concentration
3
Areas yg/m
3
Coke-oven battery 50. - 13 x 10
3
Recovery plant 50. - 145 x 10
2
Tar processing 3 x 10
Source: Masek (1971).
Data on benzene concentration in the vicinity of coke-oven and
benzene recovery plants are unavailable. In coke-oven operations,
the charging of coal is regarded as the potentially largest source of
benzene emissions.
Many pollutants escape from coke-oven operations in addition to
benzene. The major chemical compounds found in coke-oven emissions are
identified in Table IV-4. The emissions consist of gases, condensable
vapors, and particulate matter, many of which have documented toxic and/
or carcinogenic properties. Therefore, -the population exposed to coke-
oven emissions of benzene is similarly exposed to many other chemical
compounds.
B. Methodology and Exposures
To estimate the at-risk population to benzene from coke-oven
emissions, the number of people residing around the coke-oven plants
and the ambient benzene concentration must be determine. The general
methodology discussed in Chapter III was used as the basis for determining
exposure levels from coke ovens. Variations in geographic locations,
meteorologic conditions and control technology were not considered in
the analysis.
42
-------
Table IV-4
PARTIAL LIST OF CONSTITUENTS OF COKE OVEN EMISSIONS
POLYNUCLEAR AROMATIC HYDROCARBONS
Anthanthrene
Anthracene"'"
Benzindene^"
Benz(a)anthracene'
Benz(b)fluoranthene
Benzo(ghi)fluoranthene
Benzo(j)fluoranthene^
Benzo(k)fluoranthene
Benzof luorene"'"
Benzo(a)f luorene''"
Benzo(b)fluorene
Benzo (c) f luorene''"
Benzophenanthrene"'"
Benzo(ghi)perylene"'"
Benzo(a)pyrene*
Benzo(e)pyrene
Benzoquino line^"
Chrysene
Coronene^
Dibenz(a,h)anthracene
Dibenzo(a,h)pyrene
Dihydroanthracene^
Dihydrobenzo (a) f luorene''"
Dihydrobenzo (b)f luorene"'"
Dihydrobenzo(c)fluorene^
Dihydrobenz (a) anthracene''"
Dihydrochrysene"*"
Dihydrof luoranthene"''
Dihydrof luorene"''
Dihydrome thy lbenz( a) anthracene"''
Dihy drome thy lbenzo(k and b)f luoranthenes"''
Dihydrome thy Ibenzo (a and e)pyrenes"'"
Dihy drome thy Ichrysene"*"
Dihydromethyltriphenylene^
Dihydrophenanthrene
Dihydropyrene"'"
Dihydrotriphenylene^"
Dimethy Ibenzo (b) f luoranthene"*"
DimethyIbenzo(k)fluoranthene
Dimethy Ibenzo (a) pyr ene"*"
Dimethy Ichrysene"'"
DimethyItriphenylene^
Ethylanthracene"'"
Ethylphenanthrene"'"
Fluoranthene^"
Fluorene^
Fluorene carbonitrile^
Indeno(1,2,3-cd)pyrene
Methylanthracene'''
Methylbenz (a) anthracene"'"
Methy Ibenzo ( a) pyrene''"
Me thyIbenzo(ghi)perylene'
Methylchrysene''"
Me thy If luoranthene"''
Methy If luorene^"
Methylphenanthrene^
Methylpyrene''"
Methyltriphenylene^
Naphthalenet
Octahydroanthracene'''
Octahydrof luoranthene''"
Octahydrophenanthrene''"
Octahydropyrene''"
Perylene''"
Phenanthrene"'"
O-Phenylenepyrene"*"
Pyrene"'"
Triphenylene"!"
POLYNUCLEAR AZA-HETEROCYCLIC COMPOUNDS
Acridine1" ^
Benz(a)acridine
Benzoquinilinef
Dibenz(a,h)acridine'
Dibenz(a,j)acridine^
Isoquinilinet
QuinolineT
43
-------
Table IV-4 (concluded)
AROMATIC AMINES
Aniline*
B-Naphthylamina
ct-Naphthylamine
OTHER AROMATIC COMPOUNDS
Benzene Pyridine*
Mono-, di-, and tri-methylated pyridine* Toluene**
Phenol5
Arsenic
Beryllium*
Cadmium
Chromium
Cobalt*
Xylene
**
TRACE ELEMENTS
Iron
Lead*
Nickel
Selenium
OTHER CASES
Ammonia5
Carbon disulfide5
Carbon monoxide5
Hydrogen cyanide5
Acroleint
Aliphatic aldehydes'
Hydrogen sulfide
Methane 5
Nitric Oxide**
§
Sulfur dioxide
MISCELLANEOUS ALIPHATIC COMPOUNDS
Formaldehyde*
MethaneT
§
Key to sources: * Kornreich (1976)
t Lao, et al. (1975)
f Mabey (1977)
§ Smith (1971)
** White (1972)
44
-------
Crude dispersion modeling was conducted by Youngblood of EPA (1977c,
1978). Coke-oven operations usually cover a large area and benzene emissions
are distributed widely throughout. Consequently, the point source model
used by Youngblood to estimate downwind concentrations resulting from
chemical manufacturing emissions is not applicable. To account for the
emissions distributed over a large area, Youngblood used the PAL (Point,
Area, and Line Source) Dispersion Model (Burner et al., 1975) that results
in lower ambient impact for a given emission rate. The benzene emissions
were assumed to occur primarily from oven leaks. The model assumptions
were as follows: square plant area; uniform distribution of emissions
throughout the area; effective stack height, 10 m; wind speed, 4 m/s;
stability class, neutral (Pasquill Gifford "D"). Maximum, one-hour-average
concentrations at selected downwind distances for a given emission rate
of 100 g/s were obtained from PAL. These were divided by two to represent
maximum eight-hour-averages. These are shown in Table IV-5.
2
The plant size most applicable to coke-oven operations is 0.25 km
(500 m on a side). The curve corresponding to this plant size is shown
in Figure IV-1. An equation was developed through regression analysis
to characterize this curve:
C - 403 D~°'91 (4.1)
3
where C is the 8-hour worst case benzene concentration in yg/m ; and
D is the distance from the source in km.
Equation (4.1) was then normalized to annual average conditions
and to individual emission rates (the Youngblood model was based on an
emission rate of 100 g/s):
C = 0.16 E D~°'91 (4.2)
In this report, "crude" is used to mean approximate and extrapolatable.
45
-------
1000
P)
•
z
o
UJ
u
o
o
UJ
N
UJ
CO
100
10
TTTT
1 I I 1 I I I I
Source: After Youngblood (1977c)
I
I
I
I
I
I
1 I I I I I
I I I I I
0.1 1.0 10
DISTANCE FROM SOURCE - km
* Based on an emission rate of 100 g/s. This is much higher than most coke-oven
operations which usually have emission rates less than 10 g/s.
FIGURE IV-1. DISPERSION MODELING RESULTS FOR COKE-OVEN OPERATIONS*
100
-------
Table IV-5
ROUGH ESTIMATES OF 8-HOUR WORST CASE BENZENE
CONCENTRATIONS PER 100 g/s EMISSION RATE
USING THE PAL DISPERSION MODEL
Distance
From
Source
Area
(km)
0.3
0.45
0.60
0.75
1.6
2.5
4.0
6.0
9.0
14.0
20.0
3
(Concentration yg/m ]
0.01 km2
5,000
3,850
2,850
2,150
800
405
205
110
60
33
20
.06 km2
2,000
1,700
1,450
1,250
600
360
190
110
60
32
20
0.25 km2
900
750
650
595
390
270
165
100
55
32
19
*
1 for Given Plant Area
1 km2
365
325
290
260
190
150
110
80
50
29
18
4 km
145
130
120
110
85
70
50
45
34
23
16
9 km2
80
75
70
65
50
43
35
29
23
18
13
25 km2
39
37
34
33
27
23
20
17
14
11
9
To give rough estimates of annual average concentrations multiply by 0.04;
to convert to ppb, divide concentrations by 3.2
Source: Youngblood (1977c).
-------
To estimate the number of people exposed to benzene concentrations
within each range at each location, Equation (4.2) is rearranged as
follows to determine the distance at which the specified concentrations
are found:
D. = 0.133 - | (4.3)
where C. is the specified annual average concentrations (i.e., 0.1, 1.0,
1 3
4.0, and so on; input data, however, are in yg/m ); D. is the distance
in km at which the specified concentration is found; and E is the emission
SL
rate in g/s at that location.
Detailed population estimates for as far as 15 km from each location
were available from another SRI study (Suta, 1977). Consequently, once
distances (D.) were determined, the population exposed to benzene con-
centrations within each range was easily determined. Geographic coordi-
nates for most of the coke plants were obtained from the U.S. EPA-NEDS
data system. The remainder were obtained from consulting maps or by
telephone conversation. The population residing within a 15-km radius
about each coke plant was calculated by use of the Urban Decisions Systems,
Inc., Area Scan Report. This computer data system contains the 1970
census data in the smallest area available (city blocks and census
enumeration districts).
Emission rates for each coke-oven operation were estimated by basing
them on the capacity and the emission factor of 0.06 Ib benzene/ton of
*
coal obtained from EPA document AP-42 (EPA, 1976). Because actual pro-
duction data are unobtainable, capacity production and 24-hour (365 days)
operation were assumed. Appendix C lists the estimated emission rates
Three plants were colocated and their corresponding emission rates were
*
Estimated by multiplying the hydrocarbon emission factor (4.2 Ib/ton of
coal) by the fraction of benzene in the total hydrocarbon emissions
(0.0132).
48
-------
summed. The emissions of benzene from the coke plants were assumed
to be the sole contributors of benzene to the atmosphere in the vicinity
of the oven.
Table IV-6 summarizes people exposed to various annual average
benzene concentrations by state. Approximately 300,000 people are
exposed to annual average concentrations greater than 0.1 ppb (8-hour
worst case concentration greater than 2.5 ppb). Pennsylvania has the
highest number of exposed population (43% of the total), followed by
Indiana (19%) and Michigan (12%).
49
-------
Table IV-6
ESTIMATED POPULATION EXPOSED TO
BENZENE FROM COKE OVENS, BY STATE
Population Exposed
to Benzene (ppb)
State 0.1 - 1.0
Alabama 8,700
California 1,400
Colorado
Illinois 830
Indiana 58,000
Kentucky 550
Maryland 19,000
Michigan 36,000
Minnesota -
Missouri -
New York 24,000
Ohio 17,000
Pennsylvania 130,000
Tennessee
Texas
Utah 20
West Virginia 3
Wisconsin -
Total* 300,000
Totals for each state are rounded to two
significant figures; a dash (-) indicates
that no exposed population was estimated
by our method for the annual average con-
centrations listed. There may be some popu-
lation exposed to those concentrations for
shorter periods of time.
Annual average concentrations; to convert to
8-hour worst case, multiply by 25; to convert
to yg/m3, multiply by 3.2.
"Total is rounded to one significant figure.
Source: SRI estimates.
50
-------
V PETROLEUM REFINERIES
A. Sources
Petroleum refineries appear to be a significant source of atmos-
pheric benzene emissions, with more than 250 operating in 39 states.
Benzene is a constituent in crude oil and gasoline, and is produced as
a by-product of the refining process. Benzene emissions from a refinery
include: (1) process emissions from .light and heavy naphtha streams
from the crude unit; fluid catalytic cracking units; hydrocracking units;
gasoline treating units, and pumps, flanges, and other sources of
fugitive emissions; and (2) nonprocess emissions from wastewater treat-
ment facilities, heaters and boilers, and facilities for storage and
handling of benzene and gasoline (PEDCO., 1977).
Benzene produced from catalytic reforming extraction by petroleum
refineries accounted for 50% of the benzene supply in the United States
in 1976 (SRI estimates). Because the average distribution of aromatics
in reformate is 10% benzene, 40% toluene, and 50% xylene, toluene
dealkylatibn processes are being used more frequently to increase the
benzene fraction. (Faith et al., 1966). Toluene dealkylation to produce
benzene currently accounts for 27% of the benzene supply in the United
States (SRI estimates). This process is most common in petrochemical
complexes, rather than in the petroleum refineries. Table V-l lists
the petroleum refineries in each state that extracts aromatics from the
reformate produced in catalytic reforming. Texas and Louisana account
for 84% of the total production of benzene, toluene, and xylene.
The composition of crude oil varies widely, but commonly contains
about 0.15% benzene by volume (Dickerman et al., 1975). Consequently,
benzene is emitted during the refining process. We have assumed that
refineries with catalytic reforming of benzene (34 out of 266 refineries)
have larger benzene emissions than those without catalytic reforming
because of the processing and handling involved. Many of the refineries
51
-------
with catalytic reforming of benzene use it captively in the production
of a wide variety of chemicals; others sell some or all of the benzene
produced to chemical manufacturers.
Table V-l
*
PETROLEUM REFINERIES PRODUCING AROMATICS,
BY STATE
State
California
Illinois
Kansas
Kentucky
Louisiana
Mississippi
New York
Oklahoma
Pennsylvania
Texas
Total
Number of
Plants
3
2
1
1
3
1
1
1
3
li
34
Quantity
(bbl/stream day)
5,990
6,700
1,400
4,000
19,100
6,000
3,000
2,000
9,700
122,525
180,415
Total quantity of benzene, toluene, and xylene
produced.
Source: Oil & Gas Journal (May 28, 1977)
Monitoring data from one refinery producing benzene as a by-product
are shown in Figure V-l. Even though all samples were collected during
the same day, the measurements varied widely. Furthermore, the limited
nature of these data makes extrapolation unreliable.
The American Petroleum Institute. (API) conducted atmospheric moni^-
toring for benzene in the vicinity of petroleum refineries in summer
1977 (draft report, 1977, cited in PEDCo, 1978). Four refineries in
different geographic locations were sampled for 24 hours. Quality
52
-------
100
z
o
DC
5 10
o
o
o
111
z
UJ
N
UJ
CO
I I I I 1 I I I I I I I I I L
I
0.1 1.0 10
DISTANCE FROM SOURCE - km
Collected in activated charcoal tubes and analyzed by gn chromatograph with a flame tonlzatlon detector.
Detection limit was approximately 0.1 fig of benzene/100 mg charcoal.
Source: EPA, 1977
FIGURE V-1. MONITORING DATA* FOR GULF ALLIANCE REFINERY,
BELLE CHASSE, LOUISIANA
53
-------
Table V-2
ATMOSPHERIC BENZENE CONCENTRATIONS AT DISTANCES GREATER
THAN 1 KM FROM PERIMETER OF PETROLEUM REFINERIES
Average
Benzene Concentration
Geographic
Location
of Refinery
Mid-Atlantic
Pacific Northwest
Midwest
Gulf Coast
No. of
Samples*
3
1
7
3
(ppb)
at > 1 km
from
Perimeter
3
2
<1
5
24-hour samples.
Source: API (1977), as cited in PEDCo (1978).
-------
control information is unavailable for these measurements. Concentra-
tions beyond 1 km from the plant perimeter ranged from less than 1 ppb
at one location to 5 ppb at another. Table V-2 summarizes this information.
Radian Corporation conducted atmospheric benzene monitoring for
Shell Oil Company in the vicinity of the Shell Oil petrochemical refinery
in Deer Park, Texas, for 6 consecutive days in December, 1977 (Radian,
1978). Under varying meteorological conditions, the sample site 400 m
from the plant perimeter showed 24-hour concentrations ranging from 2
to 13 ppb, with a 6-day average of 7 ppb. Complete quality control
information is unavailable for these data.
At the request of EPA, Research Triangle Institute reanalyzed the
results of previously collected ambient monitoring data stored on computer
tape to determine benzene concentrations (Research Triangle Institute,
1977). Table V-3 indicates the results for eleven samples collected in
the vicinity of petroleum refineries. All sampling occurred within or
near the plant perimeter except for the St. Louis sample which was taken
3.2 km (2 miles) east of the Shell refinery.
Table V-3
RESULTS OF AMBIENT BENZENE MONITORING
IN THE VICINITY OF PETROLEUM REFINERIES
Location
Deer Park, Pasadena,
TX
El Segundo, CA
St. Louis, MO
Facility
Shell, Tenneco
Chevron
Shell
Number
of
Samples
Average
Benzene
Concentration
(ppb)
9
1
1
3.4
258
72
Source: Research Triangle Institute, 1977.
55
-------
The results of atmospheric monitoring in the vicinity of petroleum
refineries give widely differing estimates of ambient benzene concentrations.
Differences in analytical techniques, sampling periods, meteorologic
conditions, and plant operations probably affected the estimated concen-
trations significantly.
Four states have 60% of the refining capacity in the United States:
California (14%), Illinois (7%), Louisiana (13%), and Texas (26%).
Pennsylvania (5%) and New Jersey (4%) bring the total to 69%. Thus,
15% of the states (6 out of 39 states) with petroleum refineries account
for 69% of the refining capacity.
B. Methodology
1. Refining of Crude Oil
The general methodology discussed in Chapter III was used as
the basis for determining exposure levels from petroleum refineries.
Youngblood of EPA conducted dispersion modeling (1977c, 1978) to charac-
terize benzene emissions from petroleum refineries. The results were
then applied to each refinery by computer program to estimate the exposed
population. Emissions are highly variable, depending on the size and
age of the plant and on the control technology employed; however, because
specific emission factors were unavailable, general averages were used.
Because actual production data are unobtainable, capacity production and
24-hour (365 days) operation were assumed. Variations in geographic
location and meteorological conditions were not considered.
Estimates of refinery emission factors were based on average
hydrocarbon emissions and the percent of the total hydrocarbon emissions
attributed to benzene. Evaluations of available information and dis-
cussions with EPA (Radian Corporation, 1975; Hustvedt, personal communi-
cation, June 1977a) resulted in the selection of the following factors:
Total hydrocarbon emissions from petroleum refineries
= 920 lb/1,000 bbl
Estimated percentage of hydrocarbon emissions attributed
to benzene from refineries without catalytic reforming
= 0.5
56
-------
Estimated percentage of hydrocarbon emission attributed
to benzene from refineries with catalytic reforming = 1*0.
These emissions result from gasoline storage losses («50%) and from leaks
and stacks (s50%). Table V-4 presents the calculations of emission factors
from the two types of petroleum refineries identified. The listing of
U.S. petroleum refineries, shown in Appendix D, was obtained from the Annual
Refining Survey published in the Oil & Gas Journal (March 28, 1977). This
listing includes a breakdown of refineries that extract benzene, toluene
and xylene from the reformate as well as the plant capacities. The emission
rate in g/s for each plant was estimated, based on the plant capacity and
the emission factor. The emission rate for each plant is shown in Appendix D.
Table V-4
CALCULATION OF EMISSION FACTORS FOR PETROLEUM REFINERIES
Refineries with catalytic reforming:
0.92 Ib/bbl ( uul?" "'"I"-. ) x 0.01 ( £«""'• ) x 103 g/kg ., , 3
\carbon emissions/ \ benzene / ° ° = 26 g/m
0.159 m3/bbl x 2.2 Ib/kg
Refineries without catalytic reforming:
0.92 Ib/bbl f*01?1 hyd™-. ) x 0.005(Percent) x 103 g/kg .. , 3
\ carbon emissions / \ benzene / ° = 13 g/m
0.159 m3/bbl x 2.2 Ib/kg
Because petroleum refineries are large and benzene emissions
are distributed widely throughout the plant area, Youngblood used the
PAL (Point, Area, Line Source) Dispersion Model (Turner et al.) to
estimate approximate downwind concentrations. The modeling assumptions
and procedure were the same as those described for coke ovens (Chapter IV),
Table IV-6 applies to petroleum refineries as well as to coking plants.
57
-------
Three of the size categories are applicable to petroleum
refineries (Hustvedt, personal communication, 1977b):
2
Plant Area (km ) Capacity (bbl/day)
0.25 < 35,000
1.00 35,000 - 200,000
4.00 >200,000
Figure V-2 shows the curves corresponding to the three plant sizes.
Because the differences between the curves are within the range of
uncertainty associated with dispersion analysis, the middle curve
2
(1.0 km ) was used to represent the dispersion characteristics of all
refineries at the suggestion of Youngblood (personal communication,
August 1977)-
The computer program discussed in Chapter III was applied to
petroleum refineries by substituting a new equation developed through
2
regression analysis to characterize the 1.0-km curve. This equation
can be written as follows:
C = 200 D~°'51 (5.1)
where, C is the 8-hour worst-case benzene concentration in yg/m ; and
D is the distance from the source in km.
Equation (5.1) was then normalized to annual average conditions
and individual emission rates (the Youngblood model was based on an
emission rate of 100 g/s):
C = 0.08 E D~°'51 (5.2)
3.
where, E is the emission rate in g/s for the location of interest.
3.
58
-------
1000
1 I I I I I I
1 I I I 1 I I I
PLANT AREA
(km2)
0.25
Oi
•at
cc
UJ
u
o
u
Ul
N
100
iIIT r L_
10
I I I I I I I
0.1
1.0
10
100
DISTANCE FROM SOURCE - km
Based on an emission rate of 100 g/s. This is much higher than most petroleum refineries
which usually have emission rates less than 10 g/s.
Source: After Youngblood (1977c)
FIGURE V-2. DISPERSION MODELING RESULTS FOR THREE SIZE CATEGORIES OF PETROLEUM REFINERIES"
-------
To estimate the people exposed to benzene concentrations
within each range at each location, Equation (5.2) is rearranged as
-follows to determine the distance at which the specified concentrations
are found:
D. = 0.007 -TT- 1 (5-3>
where, C. is the specified concentration (i.e., 0.1, 1.0, 4.0, and so on;
1 3
input data, however are in ug/m ); and D. is the distance in km at which
the specified concentration is found. The remaining steps in the method-
ology are discussed in Chapter III.
If more than one refinery was located in a particular city:
we assumed that the refineries were co-located, and we summed their
emission rates. Although several cities had three or more refineries,
it is also true that few people generally live near such complexes. Thus,
with this method, the exposed population is minimized, whereas the
exposure level is maximized for a particular city.
For locations with fewer than three refineries, an area of
2
0.8 km (circle with a radius of 500 m) was assumed to be within the
plant boundary. No exposed population was estimated within this area.
2
For locations with three or more refineries, an area of 1.0 km (circle
with a radius of 600 m) was assumed to be within plant boundaries.
Benzene emissions from petroleum refineries were assumed to
\
be the sole contributors of benzene to the atmosphere in the vicinity
of the refinery.
2. Storage and Transfer of Pure Benzene
Benzene is produced as a salable by-product by 36 petroleum
refineries. The benzene emission rates associated with storing, handling,
and distributing pure benzene are much higher than those associated with
storing, handling, and distributing gasoline. Because the calculation
of emission rates for petroleum refineries did not include emissions
from storage and transfer of pure benzene, the additional benzene emissions
60
-------
were calculated for the 36 refineries and added to the emission rates
calculated in the previous section. A discussion of the methodology and
assumptions used follows.
The emission factors for storage and transfer of pure benzene
are shown in Table V-5. Information on the type of storage and transfer
operation in use at 12 refineries was made available in summary form
by the Chemical and Petroleum branch of EPA's Office of Air Quality Plan-
ning and Standards from the Section 114 letter sent out in 1977 (Brothers,
personal communication, 1978). That information was used along with the
following assumptions in determining the benzene emission rate at each
refinery:
Table V-5
SUMMARY OF EMISSION FACTORS FOR
PURE BENZENE STORAGE AND TRANSFER -
Facility
Control Type
Benzene
Emission
Factor
(g/gal)
*
Reference
Benzene storage
Benzene storage
Benzene transfer
Inland barge
Tank truck
Rail car
Uncontrolled 3.0
Internal floating
roof, vapor
recovery 0.3
Uncontrolled 0.76
Uncontrolled 1.8
Uncontrolled 1.8
3
1
1
Key to references: 1. Durham, personal communication, 1978.
2. Burr, personal communication, 1978.
3. Markwordt, personal communication, 1978.
61
-------
Benzene production is about 82% of capacity (SRI
estimates).
If all of the benzene produced was used captively,
only storage emissions were assumed.
If some portion of the benzene produced was known
to be used captively, loading emissions were only
attributed to the percentage sold to outside customers.
If no information was available concerning the type
of benzene transfer operation at a refinery, tank
trucks and/or rail cars were assumed for all benzene
transfers. If information was available, percentages
of the total production were assigned to barge or rail/
truck transfer, and .the associated emission was determined.
If no information was available concerning the type of
storage at a particular refinery, uncontrolled storage
was assumed.
The average benzene storage tank was assumed to hold
55,000 barrels (Durham, personal communication, 1978).
Every tank was assumed to have a 26-day retention time
(Burr, personal communication, 1978).
Truck and rail loading of benzene takes place in a centralized
location that is at a distance from the storage tanks. Barges and tankers
are loaded at another location. Therefore, emissions from loading
operations derive from one or two locations distinct from the storage
tank area. Because storage and transfer emissions are distributed over
a large area, they are considered to be area-wide emissions and can thus
be added to the refinery emission rate calculated in the previous section.
Equation (5.3) is then applied to determine the exposed population.
Table D-l in Appendix D lists each petroleum refinery and the estimated
emission rates.
C. Exposures
The population exposed to atmospheric benzene from petroleum re-
fineries by plant location is shown in Appendix D. A state summary of
annual average concentrations and exposed population is shown in Table
V-6. More than 5 million people are exposed to benzene from petroleum
refineries. Virtually all of the estimated population is exposed to
annual average concentrations between 0.1 and 1.0 ppb. In only two
62
-------
Table V-6
ESTIMATED POPULATION EXPOSED TO BENZENE
FROM PETROLEUM REFINERIES, BY STATE
State
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Delaware
Florida
Georgia
Hawaii
Illinois
Indiana
Kansas
Kentucky
Louisiana
Maryland
Michigan
Minnesota
Mississippi
Missouri
Population Exposed
0.1-1.0
(ppb)t
State
230,000
100
110,000
4,300
280
6,000
240,000
6,000
10
30,000
Montana
Nebraska
New Hampshire
New Jersey
New Mexico
New York
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Tennessee
Texas
Utah
Virginia
Washington
West Virginia
Wisconsin
Wyoming
Population Exposed
0.1-1.0 1.1-4.0
(ppb)t (ppb)t
51,000
20,000
300,000
8,100
2,000,000
1,500,000
80
2,600
Total*
5,000,000 3,000
Totals for each state are rounded to two significant figures; a dash
(-) indicates that no exposed population was estimated by our method
for the annual average concentrations listed. There may be some
population exposed to those concentrations for shorter periods of time.
Annual average concentrations; to convert to 8-hour worst case, multiply
by 25; to convert to yg/m3, multiply by 3.2.
^Totals are rounded to one significant figure.
Source: SRI estimates.
63
-------
locations, Port Arthur and Corpus Christ!, Texas, are people exposed to
annual average concentrations higher than 1.0 ppb. Pennsylvania, which
is fifth in number of petroleum refineries, has the highest exposed
population with 2,000,000; Texas is second with 1,500,000. Five states
account for almost 90% of the total population exposed to benzene from
petroleum refining.
64
-------
VI SOLVENT OPERATIONS
A. Sources
Little is known about benzene used in solvent operations. Recent
publications evaluating benzene in the workplace have identified industries
in which benzene may be used as a solvent, but the studies were unable
to quantify actual volumes of use (Arthur D. Little, Inc., 1977; PEDCo,
1977; Mitre, 1976). The Occupational Safety and Health Administration
(OSHA) is currently evaluating industries for benzene hazards under
their newly promulgated standards. In addition, a detailed study of
market input and output of benzene in solvent operations is currently
being conducted for the Office of Toxic Substances. Table VI-1 lists
major industries that OSHA is investigating to determine whether benzene
is used as a solvent in their operations.
Some indication of the maximum possible volume .of benzene used in
solvent operations can be obtained by evaluating benzene consumption
data for the United States. More than 95% of all benzene used as a raw
material is consumed by seven chemical manufacturing processes (see
8 8
Chapter III). Only 2.8% (3.05 x 10 Ib [1.39 x 10 kg]) is consumed by
other uses (SRI estimates, 1977). Other uses include benzene for:
anthraquinone, benzene hexachloride, benzene sulfonic acid (primarily
for phenol), diphenyl, hydroquinine, nitrobenzene (other than that used
for aniline), resorcinol, and solvent applications. Because three of
the uses (resorcinol, nitrobenzene, and benzene sulfonic acid) account
for approximately 50% of the benzene consumed by all other uses, solvent
f 6
operations must consume much less than 150 x 10 Ib/yr (68.0 x 10 kg).
In fact, many operations have switched to other solvents because of the
toxicity hazard associated with benzene. The amount of benzene used by
solvent operations is consumed in many, small volume markets (SRI
estimates, 1977).
65
-------
Table VI-1
INDUSTRIES AND MANUFACTURED PRODUCTS
POSSIBLY USING BENZENE AS A SOLVENT
Rubber tires
Miscellaneous rubber products
Adhesives
Gravure printing inks
Printing and publishing
Trade and industrial paints
Paint removers
Miscellaneous industrial uses
Coated fabrics
Synthetic rubber
Leather and leather products
Floor covering
Source: Brinkerhoff, personal communication,
1977.
A recent study by Arthur D. Little (ADL) (1977a) identified the
manufacture of rubber tires and of miscellaneous rubber products using
synthetic rubber and adhesives as possible major sources of occupational
exposures. Although industry sources indicate that benzene has been
removed from many of the operations within the rubber industry, the ADL
study reported that substantial quantities are still being used in the
manufacture of synthetic rubbers, production of phenolic antioxidants,
polymerization of hydrophilic polymers, and manufacture of rubber adhesives.
However, these operations may take place in locations apart from the
location where the final product is produced.
Limited monitoring data are available. NIOSH is currently conducting
a sampling program in the vicinity of solvent operations using benzene
(Hardel, personal communication, 1977). Sampling data for three
B.F. Goodrich Chemical Company solvent operations were recently submitted
66
-------
to EPA. Figure VI-1 displays the measured benzene concentrations at
various sampling sites within 1 km of the source. Benzene concentrations
were highly variable, ranging from nondetectable to 720 ppb. (The wide
variability probably occurred because the wind was gusty, averaging
between 10 and 15 mph throughout the sampling period.) Complete quality
control information is unavailable for these data. The potential for
environmental exposure to benzene from solvent operations appears to be
significant.
B. Methodology and Exposure
Because of the extremely limited information on operations using
benzene as a solvent, amount used, and probable emission factors, any
exposure estimates are necessarily .crude. The primary assumption is
that only the largest plants will have significant potential for high
environmental exposure. The 1972 Census of Manufacturers (Bureau of
the Census) was used to determine those operations that have the largest
average plant size. Table VI-2 lists the major operations and average
number of employees per plant. Five operations that averaged more than
100 employees per plant were selected for further analysis. Table VI-3
lists the number of plants and average number of employees per plant for
each of the five operations by state. Georgia and California have the
largest number of plants, together comprising 32% of the total. Based
on Table VI-3, it can be assumed that the population in the states with
the most plants has the greatest risk of benzene exposure from the
solvent operations identified.
The approximate benzene used for each operation can be roughly
estimated by assuming average plant sizes. As discussed in the previous
section, it is known that 150 x 10 Ib/yr of benzene (68 x 10 kg/yr)
is used for other unidentified uses and that solvent applications
represent less than half of that figure. If it is assumed that 40% of
this figure represents solvent use, the total is 60 x 10 Ib/yr
(27 x 10 kg/yr). Because it is expected that the largest single solvent
application is in rubber tires and miscellaneous rubber products, it is
further assumed that 80% of the total estimated solvent use is found in
67
-------
f
z
o
HI
O
O
O
UJ
Z
UJ
N
Z
800
700
600
500
400
300
200
100
90
80
70
60
50
40
30
20
10
I I I
•§> •
a —
D ANT1-OXIDANT PLANT (3-5 hr)"
• SPECIALTY POLYMER PLANT (7 he)
• CB-EPDM PLANT (5-7 hr)f
O TWO SAMPLES WITH THE SAME COORDINATES
- I -- I I II
0.1
.2 .3 .4
DISTANCE - km
.5 .6 .7 .8 .9 1.0
*
The hours shown in parenthesis are approximate averaging times for the
samples taken at each plant
t Points plotted by B. F. Goodrich Personnel (Kllrov, personal communication, 1978)
FIGURE VI-1. SAMPLING DATA FOR THREE SOLVENT OPERATIONS
68
-------
Table VI-2
AVERAGE NUMBER OF EMPLOYEES PER PLANT
FOR SELECTED SOLVENT OPERATIONS
Average Number of
SIC Number Item Employees/Plant
221
229
278
282
285
301
302
303
304
306
307
31
379
Floor covering mills
Miscellaneous textile goods
Blankbooks and bookbinding
Plastics materials, synthetics
Paints and allied products
Tires and innertubes
Rubber and plastics footwear
Reclaimed rubber
Rubber, plastic hose, and belting .
Fabricated rubber products
Miscellaneous plastic products
Leather and leather products
Miscellaneous transportation equipment
113
60
35
351
41
522
295
45
354
89
45
84
38
Source: adapted from 1972 Census of Manufacturers.
69
-------
Table VI-3
NUMBER OF PLANTS AND EMPLOYEES FOR SOLVENT OPERATION
WITH HIGH POTENTIAL FOR BENZENE EMISSIONS
Tires and
Innertubes
t E
Alabama
Arkansas
California
Colorado
Connecticut
Delaware
Florida
Georgia
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Nebraska
New Hampshire
New Jersey
New York
North Carolina
North Dakota
Ohio
Oklahoma
Pennsylvania
Rhode Island
South Carolina
Tennessee
Texas
Utah
Virginia
West Virginia
Wisconsin
Total
10
5
22
2
1
9
10
5
5
2
3
27
7
14
12
11
2
180
800
400
500*
900
1800
200
500*
360
700
1750
600
1750
600
600
600
100
400
1000
600
500*
460
500*
360
1750
Rubber, Plastic Hose
and Belting
1 300
8 750
1 350
2 100
2 900
4
2
2
3
18
5
15
1
4
1
1
83
400
100
200
750
450
100
600
600
350*
360
900
600
300
200
300
750
150
Rubber and
Plastic Foocware
0 E
1 300
3 100
1 300
3 1200
4 450
3 600
2 375
3 250
5 360
3 1200
12 300
Plastics Materials, Floor Covering
Synthetics Mills
7
5
10
4
1
2
_1
97
100
250
150
180
450
50
250
900
200
150
200
150
750
51
9
6
10
11
27
9
3
3
10
14
5
21
13
4
5
3
35
20
22
37
27
15
15
35
10
6
10
502
400
40
200
700
580
100
100
160
100
100
450
264
580
200
300
200
150
50
70
150
100
700
200
400
1100
1400
280
1800
1200
100
7 300
5 150
62 100*
247 100
7 100
2 100
100
10 100
1 1800
2 150
3 100
17 20
38 100
6
27
3
30
20
4
449
200
200
25
100
80
100
600
Number
of
Plants
26
12
146
4
15
8
14
270
50
22
8
6
17
14
5
10
53
19
6
9
14
3
12
61
52
35
38
35
14
31
10
47
53
50
2
19
8
13
1311
I » Number of plants
E - Average number of employees per plant
* - The average plant size for the category. This was used when it was not possible to determine an average plant
size for the State from Che listed information.
Source:
1972 Census of Manufacturers, Bureau of the Census; 1975 Statistical Abstract of the United States.
Bureau of Che Census.
70
-------
rubber-related manufacturing. Therefore, the amount allocated to rubber
tires and miscellaneous rubber products is estimated to 48 x 10 Ib/yr
(22 x 10 kg/yr). Table VI-3 shows 360 plants in rubber-related manufac-
turing. Using this total, the average benzene consumption per plant is
estimated at 0.13 x 10 Ib/yr (0.06 x 10 kg/yr).
Estimating the level of risk associated with each plant is difficult.
In most operations, nearly 100% of the quantity used as a solvent is
emitted to air or water. Because benzene is highly volatile, it is likely
that most of the solvent is emitted to the atmosphere. If all solvent
were emitted to the air, an average plant in rubber-related manufacturing
would have an emission rate of 2 g/s if uniform solvent use over a 24-hour,
365-day work year is assumed. The resulting benzene concentrations would
range from approximately 7 ppb at the plant perimeter to 0.1 ppb at 20 km
2
(if an area-wide emission within a 0.06 km area is assumed: see Table
IV-5). Although more exact estimates are impossible, given the available
information, the potential for environmental exposures appears to be
significant.
The states containing the most plants with high potential for
atmospheric benzene emissions are identified in Table VI-4. It is im-
possible to discern with certainty whether or not benzene is actually
used at these facilities. The probability of benzene use is high, however,
and, if used, the probability of annual average benzene concentrations
of 0.1 ppb or greater is significant. In fact, all the plants identified
in Table VI-4 are at least two times larger than the average plant size
in their category (based on total number of employees).
This same methodology can be used to determine potential emissions
in the remaining two categories: plastics materials, synthetics, and
floor covering mills. If it can be assumed that they account for 15%
of benzene consumed for solvents, the total use for these two manufac-
turing processes is estimated to be 9.0 x 10 Ib/yr (4 x 10 kg/yr).
3
With 951 plants, the average benzene consumption is 9 x 10 Ib/yr
3
(4 x 10 kg/yr) per plant. The emission rate calculated for an average
plant is 0.1 g/s, or more than 1 order of magnitude lower than that for
71
-------
Table VI-4
STATES WITH THE HIGHEST POTENTIAL FOR ATMOSPHERIC BENZENE
FROM SOLVENT OPERATIONS
Number of
State Plants
Tires and innertubes
Connecticut
Kansas
Maryland
Ohio
Wisconsin
1
2
2
27
2
Average
Number of
Employees
Per Plant
1,800
1,750
1,750
1,000
1,750
Average State
Density (1974)
(People/km2)
244
11
159
101
32
Plant Size as
Compared to
Estimated
Average Plant*
3x
3x
3x
2x
3x
Rubber, plastic hose, and belting
California
Delaware
Kentucky
North Carolina
Tennessee
8
2
1
2
1
750
900
750
900
750
52
111
33
42
38
2x
2.5x
2x
2.5x
2x
Rubber and plastics footwear
Connecticut
Georgia
Rhode Island
Wisconsin
3
3
2
1
1,200
600
900
750
244
32
343
32
4x
2x
3x
2.5x
See text for discussion of the estimated average plant size.
Source: 1972 Census of Manufacturing and 1975 Statistical Abstract
of the United States (Bureau of Census).
72
-------
rubber-related manufacturing. Thus-, if the estimated percentage of benzene
use attributed to the rubber industry versus other solvent uses is correct,
the exposures related to other solvent operations are minimal. As noted
earlier, use of benzene as a solvent in operations other than the rubber
industry is generally declining. Although it is also declining in the
rubber industry, the use volume is still presumed to be high (ADL, 1977a).
In summary, although little is known about the use of benzene as a
solvent, present indications are that its use for this purpose is declin-
ing. Crude estimates of emissions and available monitoring data indicate
that there is a potential for environmental exposures as long as benzene
is used as a solvent. Available monitoring data indicate that the levels
could be high in the vicinity of some facilities. Rubber-related manu-
facturing is estimated to be the largest source of population exposures
in this category, although estimating the population exposed was impossible
from the information available. Consequently, further study of solvent
operations including volumes of benzene, ventilation practices, and
characterization of emissions is warranted. It is likely that environ-
mental exposures will increase because OSHA regulations will require
reduced occupational exposures.
73
-------
VII STORAGE AND DISTRIBUTION OF GASOLINE
A. Sources
Storage and distribution of gasoline represent potential sources of
atmospheric benzene in the environment. There are two main emission
pathways: (1) evaporation and spills during loading and unloading and
(2) spills from collisions in transportation.
Gasoline transfers normally occur at petroleum refineries and at
numerous storage sites throughout the United States. Gasoline is usually
stored in closed containers located in remote locations. Although
evaporation loss from storage tanks has been observed, most of the benzene
released into the environment is believed to result from the operations
of loading and unloading the gasoline. Spills from collisions involving
gasoline transfer vehicles account for negligible benzene losses. The
gasoline marketing distribution system is shown in Figure VII-1.
1. Storage
Storage facilities consist of closed storage vessels, including
pressure, fixed-roof, floating-roof, and conservation tanks. Ordinary
fixed-roof tanks store less volatile petroleum products, whereas floating-
roof tanks are most commonly used to store gasoline. Diagrams of several
of these tanks are shown in Appendix A. Emissions of benzene from storage
in a floating-roof tank occur primarily from standing and withdrawal
(wetting) losses. Fixed-roof tanks have "breathing" losses caused by
expansion and contraction of the vapors due to diurnal changes in atmos-
pheric temperature. Because of the low volume of gasoline stored in
fixed-roof tanks, breathing losses are not considered to be a significant
source of atmospheric benzene.
Limited data have been reported on benzene exposures adjacent
to storage facilities. A survey of industry reported an average of 375
ppm of benzene measured next to the sampling port on top of a benzene
storage tank (Young, 1976, cited in PEDCo, 1977).
75
-------
REFINERY STORAGE
SHIP, RAIL. BARGE
vSERVICE STATIONS
BULK TERMINALS
TANK TRUCKS
AUTOMOBILES, TRUCKS
PIPELINE
BULK PLANTS
TRUCKS
COMMERCIAL,
RURAL USERS
SOURCE: PCDCo, 1977
FIGURE VII-1. THE GASOLINE MARKETING DISTRIBUTION SYSTEM
IN THE UNITED STATES
76
-------
Standing emissions are caused by improper fit of the seal and
shoe to the vessel shell. Small losses also occur when vapor escapes
between the flexible membrane seal and the roof. Withdrawal or wetting
losses are caused by evaporation from the tank walls as the roof descends
during emptying operations (PEDCo, 1977).
2. Distribution
The gasoline distribution system involving transport from the
petroleum refineries to the consumer may also be a significant source
of atmospheric benzene (see Figure VII-1). Bulk terminals represent
intermediate stations set up to serve near-source regional markets.
Gasoline at bulk terminals is transferred directly from refinery by ships,
rail tank cars, barges, and pipelines. Bulk plants, on the other hand,
are designed for local markets and their supplies are distributed by tank
trucks. Service stations that fuel public motor vehicles are supplied
by tank trucks from either bulk terminals or bulk plants. Privately
owned commercial operations, such as those providing fuel for vehicles
of a company fleet, are generally supplied by tank trucks from an inter-
mediate bulk installation.
Most of the emissions take place during transfers of the gaso-
line to tank trucks. These losses occur at a rate directly proportional
to the amount of gasoline passing through the particular location.
Because many tank trucks are filled at one bulk terminal or plant,
benzene emissions from that procedure are potentially much greater. As
empty tank trucks are filled, hydrocarbons in the vapor space are dis-
placed to the atmosphere unless vapor collection facilities have been
provided. The quantity of hydrocarbons contained in the displaced vapors
depends on the vapor pressure, temperature, method of tank filling, and
conditions under which the truck was previously loaded. A schematic
drawing of liquid and vapor flow through a typical bulk terminal is found
in Appendix A.
All monitoring data collected to date have concerned possible
occupational exposures. Measurements at several bulk loading operations
in Britain showed ambient benzene concentrations ranging from 0.1 to
77
-------
7.7 ppm (NIOSH, 1974). In the same study, NIOSH also evaluated worker
exposure during loading and weighing of rail tankers with gasoline from
storage tanks. An exposure equivalent to 14 ppm over an 8-hr workday
was estimated. Thus, environmental exposure to benzene from gasoline
distribution systems appears to require some evaluation.
Gasoline is loaded from storage tanks to transport trucks
(tank cars) by two basic methods: top loading and bottom loading (PEDCO,
1977). Top loading can be done by splash fill or submerged fill. The
former method involves free fall of gasoline droplets and thus promotes
evaporation and possibly liquid entrainment of these droplets in the
expelled vapors. In subsurface or submerged filling, the gasoline is
introduced below the liquid surface in the tank. Bottom loading of
gasoline is comparable to submerged top loading.
B. Methodology and Exposures
The emission factors for benzene losses from gasoline storage
and transfer are shown in Table VII-1. Because bulk gasoline plants
are low-volume operations (4,000 gal/day) and are often located in rural
areas, they are not considered in this analysis. Bulk terminals, however,
are high-volume operations (250,000 gal/day) and are generally located
near urban demand centers—commonly in highly industrialized areas or
on city peripheries where population densities are low.
A bulk terminal has at least three storage tanks, one for each
3 3
grade of gasoline; each tank holds 55,000 bbl each (8.7 x 10 m ). Rough
ambient benzene concentration estimates for the vicinity of bulk terminals
were based on emission factors, assumed storage and loading volumes, and
the dispersion modeling results discussed in Chapter IV. A typical
bulk terminal has the following characteristics: average tank size,
3 3
8.7 x 10 m ; 28-day retention time; three gasoline storage tanks of
average size; gasoline loading equaling 250,000 gal/terminal/day; and
2
facility size of 0.01 km . The emission rates for each type of control
are calculated as follows:
78
-------
Table VII-1
EMISSION FACTORS FOR BENZENE LOSSES
FROM GASOLINE STORAGE AND DISTRIBUTION
Source
Loading
Uncontrolled
Controlled
Storage
Uncontrolled
Controlled
Emission Factor (g/gal) Reference
1.14 x 10
1.14 x 10
-2
-3
4.0 x 10
4.0 x 10
-2
-3
1
2
1
3
Key to references:
1. Durham, personal communication,
1978.
2. Polglase, personal communication,
1978.
3. Burr, personal communication,
1978.
Uncontrolled Case
Loading
Emission rate = (emission factor) (volume loaded) (day/s)
= (1.14 x 10~2 g/gal) (2.5 x 105 gal/day) (day/86,400 s)
_2
= 3.3 x 10 g/s .
79
-------
Storage
Emission rate = (emission factor)(number of tanks)(tank volume)(days/s)
= (4.0 x 10~2 g/gal)(3)(2.3 x 106 gal/28 days)(day/86,400 s)
= 1.14 x 10"1 g/s
Total emission rate = 1.47 x 10 g/s
Partially Controlled Case—Controlled Storage, Uncontrolled Loading
_2
Loading emission rate = 3.3 x 10 g/s
_2
Storage emission rate = 1.14 x 10 g/s
_2
Total emission rate = 4.44 x 10 g/s
Controlled Case
_3
Loading emission rate = 3.3 x 10 g/s
_2
Storage emission rate = 1.14 x 10 g/s
_2
Total emission rate = 1.47 x 10 g/s
The ambient benzene concentrations can be estimated from the dis-
persion modeling calculations of Youngblood (1977c, 1978) that assume
uniform emissions throughout the terminal area. By applying the estimated
emission rate to the results presented in Table IV-4 (Chapter IV) for
2
the indicated terminal area of 0.01 km , the dispersion curves shown
in Figure VII-2 are generated. Note that all annual average concen-
trations are below the detectable limit. For comparison, the 8-hour
worst case is also shown for each type of control.
To determine whether larger operations would cause exposures above
the detectable limit, a 10-tank bulk terminal was also evaluated.
The characteristics of the facility remain the same, except for gasoline
loading of 8.3 x 10 gal/day and faci.'
The calculated emission rates follow:
5 7
loading of 8.3 x 10 gal/day and facility size estimated a 0.06 km .
80
-------
4.0
1.0
I
g
<
cc
z
LU
O
o
u
m
Z
LU
N
0.1
0.01
1 I I I 1 1
UNCONTROLLED
8-HOUR WORST CASE
PARTIALLY
CONTROLLED
8-HOUR
CONTROLLED
8-HOUR
UNCONTROLLE
ANNUAL AVERAGE
PARTIALLY CONTROLLED
— ANNUAL AVERAGE
0.1
Source: SRI ESTIMATES
1.0
DISTANCE-km
10
20
FIGURE VII-2. ESTIMATED DISPERSION CURVE FOR A 3-TANK GASOLINE BULK TERMINAL
81
-------
Uncontrolled Case
Loading: 1.1 x 10 g/s
Storage: 3.8 x 10 g/s
Total emission rate = 4.9 x 10 g/s
Partially Controlled Case—Uncontrolled Loading, Controlled Storage
Loading: 1.1 x 10 g/s
_2
Storage: 3.8 x 10 g/s
Total emission rate = 1.5 x 10 g/s
Controlled Case
_2
Loading: 1.1 x 10 g/s
_2
Storage: 3.8 x 10 g/s
Total emission rate = 4.9 x 10 g/s
The emission rates are again applied to the dispersion modeling
2
results of Youngblood (1977c, 1978) for the 0.06 km dispersion curve
(Table IV-4). The dispersion curves shown in Figure VII-3 are then
estimated. Annual average concentrations fall below the detectable
limit for all cases within 500 m of the bulk terminal facility. Because
the facility boundary is estimated to be approximately 200 m on each side,
the number of people exposed to annual average concentrations higher
than 0.1 ppb is assumed to be minimal.
The results of this analysis indicate that few members of the public
are exposed to annual average benzene concentrations higher than 0.1 ppb.
As shown, some people are exposed to 8-hour worst case concentrations
between 0.1 and 1.0 ppb. Nevertheless, exposures to the public are
considered minimal, although occupational exposures may be high.
No information is available about the number of gasoline storage
terminals in an average terminal facility. Therefore, the analysis here
82
-------
4.0
1.0
a
z
cc
z
111
o
z
o
o
LU
z
Ul
N
Z
111
00
0.1
0.01
I I I
UNCONTROLLED
8-HOUR WORST CASE
PARTIALLY
CONTROLLED
8-HOUR
CONTROLLED
8-HOUR
UNCONTROLLED
ANNUAL AVERAGE
PARTIALLY CONTROLLED
ANNUAL AVERAGE
CONTROLLED
ANNUAL AVERAGE
DETECTABLE
LIMIT _J
0.1
Source: SRI ESTIMATES
1.0
DISTANCE-km
10
20
FIGURE VI1-3. ESTIMATED DISPERSION CURVE FOR A 10-TANK
GASOLINE BULK TERMINAL
83
-------
merely shows a range of possible facilities. Although it is not known
how many terminals have floating roofs (90-95% more control then fixed
roof tanks), estimates are that 70 to 80% of a total of 2700 terminals
are controlled in this manner (Burr, personal communication, 1978).
-------
VIII URBAN EXPOSURES
A. Sources
Urban exposures to benzene come from many sources, including chemical
manufacturing plants, automobile exhaust, gasoline service stations,
gasoline evaporation, and losses through transportation and storage of
benzene and gasoline. Because benzene is not routinely monitored in
ambient air, few sampling data exist. A study by Altshuller (1969, cited
in Mitre, 1976) estimated normal benzene concentrations at between 10
and 50 ppb. This estimate appears to be quite high when compared with
other benzene sources discussed previously. In 1973, the General Motors
Atmospheric Research Laboratory monitored ambient air quality in Denver
for a variety of hydrocarbons. Average benzene concentrations of 3 ppb
were found, with a maximum of 30 ppb (Ferman et al., 1977). A study of
atmospheric benzene and toluene levels in Toronto found a maximum concen-
tration of 98 ppb, with an average concentration of 13 ppb (Pilar and
Graydon, 1973). That study concluded that benzene contamination of the
air was related to automobile emissions based on three factors: (1) the
ratio of benzene to toluene, (2) the presence of distinct peak periods
for both hydrocarbons at rush hour periods, and (3) the relative concen-
trations detected at various sampling stations.
Research Triangle Institute (1977) collected grab samples in urban
areas with high industrial activity (see Table VIII-1). Measured benzene
concentrations varied significantly—from a trace to 94 ppb. Some of the
reasons for such variation include meteorological conditions, density of
automobiles in urban areas, industrial emissions, and sampling periods.
Gasoline contains varying amounts of benzene, depending on lead
content and refinery source, among other things. Before 1974, the average
benzene content in U.S. gasoline was less than 1% by liquid volume
(Runion, 1975). More recent data (Runion, 1976) indicate that the average
benzene content has been increased to maintain octane levels as lead
85
-------
Table VIII-1
RESULTS OF AMBIENT BENZENE MONITORING
IN URBAN AREAS WITH HIGH INDUSTRIAL ACTIVITY
oo
ON
Location
Patterson, NJ
Clifton, NJ
Passaic, NJ
Hoboken , NJ
Newark , NJ
Staten Island, NY
Edison, NJ
Houston, TX
Los Angeles, CA
Long Beach, CA
St. Louis, MO
& vicinity
Baton Rouge, LA
& vicinity
Industrial
Activity
A, I
A, I
A, I
A,I
A,I
A, I
A, I
6C, 5P, ICo, A
A, I
A, I
1C, ICo, IP, A
1C, IP, A
Average
Sampling
Time (min)
42
39
39
39
38
37
42
72
52
52
420
i
610
Number of
Samples
1
1
1
1
1
1
1
3
2
1
6
21
Average
Benzene
Concentration
(ppb)
0.7
trace
2.0
trace
94
0.7
1.0
2.6
5.5
7.1
18.9
1.0
Where known, benzene production or consumption facilities are listed.
Key: A = large automobile emission; I = general industrial activity; C = chemical
manufacturers using benzene; Co = coke oven; P = petroleum refinery.
Source: Research Triangle Institute, 1977.
-------
content has been reduced. Current estimates of average benzene content
in gasoline range from 1.24 to 2.5% by liquid volume (PEDCo, 1977).
Tables VIII-2 and VTII-3 show the results of analyses of gasoline from
different refinery sources; these results indicate substantial variation
among refineries and types of blends.
Table VIII-2
TYPICAL LIQUID VOLUME-PERCENT OF BENZENE IN
GULF U.S. GASOLINES, OCTOBER 1976
Vol% Benzene
Refinery Source
A
B
C
D
E
F
Average
Standard deviation
Good Gulf
\ 0.54
1.99
1.19
. 1.59
1.25
0.85
1.24
0.52
Gulf Crest
0.88
1.45
1.21
1.18
1.98
0.82
1.25
0.43
No-Nox
1.16
0.85
0.81
1.49
2.39
0.88
1.26
0.61
Source: Runion, 1976 (cited in PEDCo, 1977).
Gasoline is a mixture of a wide variety of hydrocarbons and fuel
additives. Emissions from gasoline evaporation or exhaust contain many
chemical compounds, some of which have demonstrated toxic and/or carcino-
genic properties. For example, concern has already been raised about
three fuel additives: tetraethyl lead, ethylene dibromide, and ethylene
dichloride. Appendix E lists more than 100 hydrocarbons and an equal
number of fuel additives commonly found in gasoline. Few data have been
generated to quantify possible exposures to these chemical compounds.
Such estimates are difficult because of the significant variation among
refinery blends.
87
-------
Table VIII-3
BENZENE CONCENTRATION IN DIFFERENT GRADES AND
SEASONAL BLENDS OF GASOLINE
Company —
Typical
Service
Station
Tresler-Comet
Bonded
Bonded
Clark
Gasoline
Grade
Premium
Regular
Unleaded
Regular
Unleaded
Regular
Unleaded
Regular
Unleaded
Vol% Benzene in
Bulk Sample
Summer
1.11
1.21
1.41
0.88
1.19
0.88
1.20
0.97
1.09
Winter
1.10
1.00
1.60
0.88 '
1.60
0.88
1.60
2.00
1.10
Average
Vol%
Benzene
1.11
1.11
1.51
0.88
1.40
0.88
1.40
1.49
1.10
Source: National Institute of Occupational Safety
and Health, 1976 (cited in PEDCo, 1977).
To determine average urban exposures throughout the United States,
it is necessary to restrict the analysis. Although substantial variation
probably occurs from one urban area to another, it is nonetheless possible
to determine a reasonably accurate estimate of annual average exposures
related to two definitive sources: evaporative losses at service stations
and automative emissions from tailpipe emissions and evaporative losses.
B. Methodology and Exposures
1. Urban Exposures from Automobile Emissions
The benzene content in gasoline varies widely, with an average
of approximately 1.24% by liquid volume. In addition, catalytic converters
on automobiles can reduce benzene in vehicle exhaust by 30 to 80%
(Johnson, 1977). Thus, similar variation in benzene emitted to the
atmosphere can be expected from evaporation of gasoline and from vehicle
exhaust emissions.
88
-------
EPA has conducted dispersion modeling of automobile emissions.
In the Hanna-Gifford dispersion model used by Schewe of EPA (1977),
concentration depends on areawide emissions and wind speed. An empirical
factor is also applied. By applying generalized emission factors, area-
wide emissions were estimated from vehicle miles traveled (the total
number of miles traveled in a given area in a year by all automobiles)
and from the number of registered automobiles. Table VIII-4 presents
the results of this study. In central cities, the concentrations range
from 1.0 ppb to 4.0 ppb, whereas in suburban areas the concentrations are
generally below 1.0 ppb.
Limited data are available concerning urban exposures from
automobile emissions. Consequently, it is difficult to develop accurate
techniques to predict benzene levels in urban areas. Uncertainties
include: benzene content in gasoline; control technology; deterioration
of the control technology over time; and dispersion characteristics of
benzene under variable meteorological conditions. Thus, a simplified
model is employed to provide general estimates of ambient concentrations.
The Hanna-Gifford dispersion model (Gifford and Hanna, 1973)
as applied by Schewe (1977) is used for this analysis. Inputs to the
model include: number of vehicle registrations, total number of vehicle
miles traveled (VMT), area size, and average annual wind speed.
The tailpipe emissions are estimated by the following equation
(Schewe, 1977):
. . 2. 0.22 g VMT 1 ,0 ..
Qtail(g/S"m ) ' mile ~s~ Area of study (m2) (8>1)
The emission factor of 0.22 g benzene per mile is a composite emission
factor for 1976 (Schewe, 1977).
The evaporative emissions are calculated as follows:
Q , , _ 2 0.148 g 3.3 trips // veh. 365 days year 1_
evap 8 ra ' ~ trip veh. -day 1 year 3.154 x 10? s area
(8.2)
89
-------
Table VIII-4
ESTIMATES OF ANNUAL AVERAGE BENZENE CONCENTRATIONS
IN FOUR URBAN AREAS
vO
o
City
Dallas
City
Suburbs
Los Angeles
City
Suburbs
St. Louis
City
Suburbs
Chicago
City
Suburbs
10
Vehicle
Miles
Traveled
6.14
5.36
10.2
21.5
2.86
7.71
18.8
23.5
Registered
Automobiles
619,684
540,786
2,044,203
4,299,073
378,280
1,020,219
1,860,292
2,327,206
Land Area
(108 m2)
6.9
43.0
12.0
54.0
1.6
61.0
5.8
88.0
^evap
g/s m )
5.08
.719
9.63
4.53
13.5
.944
18.2
1.49
Qtail
(10-92
g/s m
62.2
8.80
59.3
28.0
126.0
8.80
227.0
18.6
QT
(10-92
g/s m )
67.3
9.52
68.9
32.5
140.0
9.74
245.0
20.1 _
Average
Annual Wind Benzene
Speed (m/s) (yg/m3)
5 3.03
0.43
2 7.75
3.66
4 7.88
0.55
4 13.78
1.13
Concentration
(ppb)
.95
.13
2.4
1.1
2.5
.17
4.3
.35
1976 projections
0 = Evaporative emissions from automobiles.
evap
Q = Tailpipe emissions from automobiles.
t ai-L
Q = Total automobile emissions
Source: Schewe, 1977
-------
By multiplying the constants in this equation, we get the following:
O ( I - ^"\ = 5.653 x 10 g number of vehicles
^evap S S m veh. s area of study (ra^)
(8.3)
This technique assumes that each vehicle emits 0.148 g of benzene per
trip and that the average vehicle travels 3.3 trips per day (Schewe, 1977)
The total emissipns for automobiles can be expressed as
follows:
\ = 'tail + Qevap (8'4)
Equation (8.4) is essentially the summation of Equations (8.1) and (8.3).
To calculate the average annual areawide benzene concentrations,
Equation (8.5) is used:
225 Q
*= (8'5)
The average annual wind speed, u, in the area of study was obtained from
Figure VIII-1. Because wind speed (and thus dispersion) increases in
the afternoon, the morning values were used to estimate higher concen-
trations. The number 225 is an empirical factor derived from several
studies that gave very good results for long-term averages; it applies
to light-duty vehicles such as passenger cars.
Because of the general unavailability of 1976 data for all
urban areas, 1973 data were used as much as possible in this estimation.
Comparisons of 1973 with 1976 data indicated that the changes was less
than 3% and had a negligible effect on the final results. The following
data sources were used:
• 1973 Standard Metropolitan Statistical Area (SMSA) and
county populations — U.S. Bureau of the Census, 1976,
Series P-25, No. 618.
1973 SMSA and county automobile registrations — U.S.
Department of Transportation, Federal Highway Administration,
1974, Table MV-21.
91
-------
Average annual vehicle miles traveled by size of SMSA—
Federal Highway Administration, 1972, Nationwide Personal
Transportation Study, Report No. 2, Table 9.
Average annual wind speed—EPA, 1972, Publication No. AP-101.
SMSA, county and city land areas—Bureau of the Census,
1972 County and City Data Book.
Source: EPA, 1971
FIGURE VIII-1. ISOPLETHS (m/sec) OF MEAN ANNUAL WIND SPEED
THROUGH THE MORNING MIXING LAYER
A detailed analysis was conducted for the six largest cities in the U.S.
(populations of more than 1 million). Table VIII-5 presents the results.
Because input data were slightly different, the results differ somewhat
from those shown in Table VIII-4. For example, the suburban area used
in this estimate may include a larger area than that used in the Schewe
estimates. Suburban areas are defined as those areas outside the central
city but within the SMSA. Because no VMT and registration data were
available at the city level, they were extrapolated either from SMSA data
or county data and were based on the fraction of the population residing
92
-------
Table VI11-6
ESTIMATES OF AVERAGE ANNUAL BENZENE CONCENTRATIONS
FOR CITIES WITH POPULATION EXCEEDING 1.000.000
VO
City
Chicago
Detroit
Houston
Los Angeles
New York
Philadelphia
Population
103
6,999.8
4,446.3
2,163.4
6,938.3
9.746.4
4,826.3
Population
103
3,173
1.387
1,320
2,747
7,647
1,862
Area
ID'*'
0.57
0.35
1.1
1.2
0.77
0.33
Automobile
Registration
1,324.171
675,065
701,766
1,490,483
1,707,891
944.660
vehicle
103
11.5
11.5
14.0
11.5
11.5
11.5
WIT
1.5
0.77
0.98
1.7
1.9
1.0
u
Ws)
5
6
6
3
7
6
10-'
1.8
1.5
0.63
0.98
1.7
2.0
evap
ID"8
g/s-rn^
1.3
1.0
0.36
0.67
1.2 ,
1.5
i
1.9
1.6
0.66
1.0
1.8
2.1
Centr;
8.6
6.0
2.5
7.5
5.9
8.0
il City
ppb
2.7
1.8
0.7
2.3
1.8
2.5
Suburban
1.1
1.2
0.23
0.4
0.3
1.5
Composite
ppb
1.9
1.5
0.5
1.3
1.0
2.0
Source: SRI estimates based on Hanna-Gifford model as applied by Schewe (1977);
data source* listed In text.
-------
in each area. The results show that the estimated benzene concentrations
in city and suburban areas range from 0.7 to 2.7 ppb and 0.2 and 1.5 ppb,
respectively. The composite benzene concentrations in the six correspond-
ing SMSAs ranged from 0.5 to 2.0 ppb.
It is expected that people living in urban areas are exposed
to higher levels of benzene from automotive emissions than those living
in rural areas. Consequently, our approach was designed to include as
many urban residents as possible. Although 43% of the total urban
population resides in central cities (as defined by the Bureau of the
Census), 83% of the total urban population resides in SMSAs. Thus, a
greater percentage of the urban population is captured by using SMSAs
as study areas. The six largest cities are in SMSAs with more than 2
million population. To analyze the remaining SMSAs, the following
population size categories were employed (U.S. Bureau of the Census,
1976, Series P-25, No. 618):
SMSA Population Size Category Number of Areas
2,000,000 or more 15
1,000,000 - 2,000,000 20
500,000 - 1,000,000 37
250,000 - 500,000 63
less than 250,000 124
SMSA composite benzene concentrations were estimated for seven
areas that represent four population size categories (see Table VIII-6).
For SMSAs with population exceeding 500,000, the composite average
annual benzene concentrations ranged between 0.1 and 0.4 ppb. However,
SMSAs of less than 500,000 were below 0.1 ppb. It may be assumed from
this analysis that the SMSAs with population less than 500,000 have
average annual benzene concentrations less than 0.1 ppb.
The estimates of urban exposures from automobile emissions are
order-of-magnitude estimates that are based on a simple dispersion model.
Note that, in certain locations and under certain meteorological conditions,
benzene concentrations may be a factor of 10 higher than those listed.
94
-------
VO
Table VIII-6
ESTIMATES OF AVERAGE ANNUAL BENZENE CONCENTRATIONS FOR SELECTED SMSAs
SMSA
Population
2
Area (m )
Automobile
Registration
VMT/ ^tail ^evap T Benzene
Vehicle VMT u 10~9 10~10 10~9 Concentration
103 109 m/s g/s-m2 g/s-m2 g/s-m2 pg/m^
SMSAs > 2,000,000
Pittsburgh
San Francisco
SMSAs 1,000,000
Columbus
Milwaukee
SMSAs 500,000 -
Sacramento
Providence-
Warwi ck-
Pawtucket
SMSAs 250,000 -
Wichita
Harrisburg
2,333,600
3,135,900
- 2,000,000
1,055,900
1,423,200
1,000,000
851,300
854,400
500,000
375,600
425,500
7.8 x 109
6.2 x 109
6.2 x 109
3.7 x 109
8.7 x 109
q
2.4 x 10
6.2 x 109
4.1 x 109
2,358,600
688,300
567,803
642,531
439,803
869,100
221,715
198,997
11.3 26.0 5 23.0 17.0 25.0 1.1
11.5 7.7 3 8.5 6.2 9.1 .68
11.3 6.4 5 7.2 5.1 7.8 .35
11.3 7.2 5 13.0 9.8 14.0 .62
11.3 4.9 3 3.9 2.8 4.2 0.3
11.3 9.8 7 28.0 20.0 30.0 0.9
10.3 2.3 7 2.5 2.0 2.7 .08
10.3 2.0 5 3.4 2.7 3.7 .16
ppb
.4
.2
.1
.2
0.1
0.3
<0.1
< 0.1
Source: SRI estimates using Hanna-Gifford dispersion model as applied by Schewe (1977).
-------
In addition, central city areas ( as shown in Table VIII-5) may have
consistently higher levels than surrounding areas because of traffic
density, frequency of intersections, and street density. Because the
model only includes automobile emissions, areas with substantial commercial
or bus transportation may have higher levels than estimated. Also, the
model is extremely sensitive to area size as Table VIII-5 indicates.
Thus, composite SMSA benzene concentrations provide the most reasonable
estimate of the average annual exposures for an urbanized area.
The total estimated urban population exposed to benzene in
concentrations greater than 0.1 ppb from automobile emissions is shown
in Table VIII-7. The 1974 SMSA populations for Chicago, Detroit, Los
Angeles, New York, and Philadelphia were summed to estimate the population
exposed to average annual benzene concentrations of 1.1 to 4.0 ppb. The
1974 SMSA population of Houston plus the remainder residing in SMSAs
with populations greater than 500,000 were summed to estimate the total
population exposed to average annual benzene concentrations between 0.1
and 0.1 ppb. The results indicate that 114 million people, or 73% of the
total SMSA population, are exposed to average annual benzene concentrations
greater than 0.1 ppb.
Table VIII-7
URBAN POPULATION EXPOSURES RELATED
TO AUTOMOBILE EMISSIONS
Annual Average
Benzene Concentration
Source 0.1-1.0 1.1-4.0 Total
Automobile emissions 69,000,000 45,000,000 114,000,000
* 3
To convert to yg/m , multiply concentrations by 3.2; to
estimate 8-hour worst case, multiply by 4.1.
Source: SRI estimates.
96
-------
2. Urban Exposures from Gasoline Service Stations
People residing in the vicinity of service stations are exposed
to benzene emissions from evaporative losses resulting from gasoline
pumping by attendants and customers, and from gasoline loading into
underground tanks by distribution trucks. The amount of benzene emitted
depends on a number of critical factors: ambient temperature, vapor
recovery controls, pumping volumes, and the benzene content in gasoline
are probably the most important. The United States has approximately
184,000 service stations, and it is expected that,many people are exposed
to benzene from these sources. Because the density of service stations
in urban areas is high, only urban areas are considered in this analysis.
For the population exposed to service station emissions, it is
necessary to estimate the stations in urban areas. Service station
density can be extrapolated from data presented in Table VIII-8. These
data, which have no apparent regional pattern, show an average of 0.7
service station per 1000 population. This density figure can be applied
generally to urban areas throughout the United States to give a rough
*
approximation of the total number of service stations. Urbanized areas
provide the best population base. The 1970 population residing in
urbanized areas was 118,447,000 (Bureau of Census, 1975). Thus, service
stations in urbanized areas are estimated at 83,000, or 46% of all stations,
Defined by the Bureau of Census as the central city or cities and
surrounding closely settled areas. Sparsely settled areas in large
incorporated cities are excluded by this definition. Densely populated
suburban areas, however, are included (U.S. Department of Commerce,
Bureau of the Census, 1972 County and City Data Book).
97
-------
Table VIII-8
SERVICE STATION DENSITY IN FOUR METROPOLITAN AQCRs
AQCR
Boston
Dallas
Denver
Los Angeles
Number of
Service
Stations
(1977)
2,353
3,218
1,277
7,298
AQCR1"
Population
(1975)
4,039,800
2,970,900
1,389,000
14,072,400
Service Station1"
Density
(number/ 1000
population)
0.6
1.1
0.9
0.5
Sources: * ADL
t U.S. Department of Commerce, Bureau of
Economic Analysis, 1973.
f SRI estimates.
The atmosphere in the vicinity of gasoline service stations
has been monitored in three studies. API (draft report, 1977) monitored
eight service stations in various geographic locations with continuous
samplers over a 24-hour period. Quality control information is not
available for these data. Six of the monitored locations were in urban
*
or suburban areas. The average benzene concentrations ranged from 3 to
10 ppb within 200 m of the service station (see Figure VIII-2) . The
two locations in rural or mountain areas showed generally lower average
benzene concentrations ranging from 0.4 to 3.7 ppb within 200 m of the
service station (see Figure VIII-3) . The wide variability in the data
is probably related to the differing meteorologic conditions, pumping
volume, benzene concentration in the gasoline, and ambient temperatures.
fc
The Pacific Northwest location (C-l) was included in this category because
of its proximity to an interstate highway and commercial development.
98
-------
10
VO
VO
Ul
o
8
ui
ui
CD
0.3
1 1 1 1 1 1 1 |
1.0
A°A
A A
A
« A
•• •
• • A —
O Urban Midwest, A-1
A Urban Midwest, A—2
O Suburban Eastern, B—1
• Suburban Eastern, B—2
A Pacific Northwest, C— 1
• Urban Eastern, C—2*
samples with the same coordinates
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
10
DISTANCE FROM SOURCE-m
100
300
Because this location was next to an interstate highway, these sampling data were included in this category.
t Distances from the source were recalculated from the original data for these locations. The distances shown
here represent distances from the gasoling pumps. The original data showed distance from the service station.
Source: Adapted from American Petroleum Institute, 1977
FIGURE VIII-2. RESULTS OF ATMOSPHERIC MONITORING IN THE VICINITY OF
URBAN-SUBURBAN GASOLINE SERVICE STATIONS, SUMMER, 1977
-------
30
o
o
4
flC
z
111
o
LLJ
Z
1U
N
Ul
m
10
0.3
•
A
> AA
• A •
A Mountain, D—1
• Rural West, D-2
/\ Two samples with the same coordinates
I
A •
I I I I I I I I
1.0 10 100
DISTANCE FROM SOURCE-m
Source: Adapted from American Petroleum Institute, 1977
FIGURE VIII-3. RESULTS OF ATMOSPHERIC MONITORING IN THE VICINITY OF
RURAL-MOUNTAIN GASOLINE SERVICE STATIONS, SUMMER, 1977
300
-------
Battelle (1978a) monitored the atmosphere for benzene in the
vicinity of gasoline service stations in the summer of 1977. The results
for the three locations in Columbus, Ohio, gave average concentrations
ranging from-1.0 to less than 1.0 ppb for a 24-hour period. However,
the report urges caution in using these data because of unexplained varia-
tions from expected results. The only finding that the report could
report with assurance was that benzene concentrations did seem to decrease
with distance from the service stations. Benzene concentrations ranged
from 10 to 22 ppb 150 m downwind in two grab samples obtained in a resi-
dential neighborhood during the filling of an underground storage tank.
Radian Corporation monitored the atmosphere for benzene in the
vicinity of a gasoline distribution facility in Irving, Texas, for Shell
Oil Company (Radian, 1978). The report does not indicate whether the
facility was a service station or a gasoline bulk plant. Ten 24-hour
samples were collected at the first site 800 m from the facility on top
of a two-story motel; they showed an average benzene concentration of
2 ppb. At the second site, approximately 180 m from the facility at
ground level, the average benzene concentration from 24-hour samples was
3 ppb. These results would seem to indicate that the facility was
responsible for adding approximately 1 ppb to the atmosphere. A third
sample site approximately 2000 m from the facility and located near a
major highway showed a 7 ppb average for two parallel 24-hour samples.
(One sample showed 3 ppb; the second backup sample showed 10 ppb; no
reasons was offered for the discrepancy between the two samples.)
Separating population exposures related to gasoline service
stations from general urban exposures is difficult. Service stations,
which are usually located along well-traveled roads, may be isolated or
located near other stations. Emissions from service stations vary,
based on differences in operating times, pumping volumes, and benzene
concentration in gasoline. Benzene concentrations measured at atmos-
pheric monitoring stations are influenced by all of these factors, in
addition to emissions from automobile exhaust, evaporation of gasoline
from parked cars, and emissions for other benzene point sources. With
101
-------
conditions varying from station to station, nationwide benzene exposure
levels from this source are difficult to approximate.
Because gasoline service stations represent a significant
source of benzene exposure to many people, we attempted to estimate the
exposure levels and the population-at-risk insofar as possible given
the data limitations. Three methods, which are discussed below, were
investigated.
Point Source Dispersion Modeling—EPA conducted dispersion
modeling for a worst-case condition (Youngblood, 1977d) by using the
single source (CRSTER) model. Meteorological data for Denver, Colorado,
were used to represent a reasonable worst-case location. The model was
executed to eliminate nighttime inversions, resulting in enhanced dis-
persion and, for low-level sources such as service stations, lowered
ground-level concentrations. Table VIII-9 presents the results of
the dispersion modeling.
In applying the available modeling data to urbanized areas,
many difficulties are inherent. Note that the operating conditions,
pumping volumes, and the chosen location represent worst-case conditions.
In particular, the pumping volume used is approximately four times
larger than that for a typical service station. Consequently, extrapo-
lating these results to average conditions is difficult. The data seem
to indicate, however, that individuals residing within 100 m of a service
station may be exposed to annual average concentrations of 1.0 ppb or
more, wheras those residing beyond 100 m may be exposed to less than 1.0
ppb on an annual average basis.
Uniform Distribution Dispersion Modeling—Because initial efforts
to treat service stations as point sources gave unsatisfactory results,
we decided to assume that service stations were uniformly distributed
and to use the Hanna-Gifford dispersion model to estimate annual average
concentrations. This model should be applicable because service stations
are widely distributed throughout urbanized areas. (We estimate 1.5
service stations per square kilometer, i.e., 2.4 per square mile.)
102
-------
Table VIII-9
ROUGH DISPERSION MODELING RESULTS FOR GASOLINE SERVICE STATIONS
% Benzene
a Hours of in Gasoline Calculated Distance (m)
Station Operation Vapor Emission Rate (g/s) 50 100 150 200 300
**
8-Hour Worst-Case Concentration (ppb)
Al 8 a.m. - 4 p.m. 0.7 0.019 27 13 8 5 3
6 days/week
A2 8 a.m. - 4 p.m. 3.0 0.080 117 57 34 23 12
6 days/week
**
Annual Average Concentration (ppb)
Bl 24 hours/day 0.7 0.0053 1 <1 <1 <1 <1
7 days/week
B2 24 hours/day 3.0 0.023 2 1 1 <1 <1
7 days/week
*
Pumping rate for all stations is 200,000 gal/month uniformly over hours of operation;
rate of evaporative loss for all stations is 10 g/gal pumped.
3
To convert to ug/m , multip
Source: Youngblood, 1977d.
-** 3
To convert to ug/m , multiply concentrations by 3.2.
-------
The same approach used in the previous section was applied here:
Total emissions related to evaporation from gasoline service stations
were estimated for an area. This result was used as the input to the
dispersion equation [Equation (8.5)], along with wind speed for the
region. The emission factors were determined through consultation with
EPA and are summarized in Table VIII-10. We assumed that 50% of the service
stations have submerged fill and 50% have splash fill. The estimates
of controlled emissions assume no change in breathing and spilling losses
from the uncontrolled situation.
Table VIII-10
EMISSION FACTORS FOR BENZENE LOSSES
AT GASOLINE SERVICE STATIONS
Hydrocarbon Emissions
Type of Control
Uncontrolled
Stage I (controls on
gasoline storage/filling
operations
Stages I and II (controls
on gasoline storage/
filling and on auto-
mobile refueling
Gasoline
Storage
and Fill
(g/gal)
Automobile
Filling
(g/gal)
Total
(g/gal)
Estimated
Benzene
Emissions*
(g/gal)
4.8
0.59
0.59
4.4
4.4
0.73
9.2
5.0
1.32
0.07
0.04
0.01
Hydrocarbon emissions are estimated to contain 0.8% benzene, assuming
1.3% benzene by liquid volume in gasoline.
Source: Kleeburg, personal communication, 1978.
104
-------
Currently, Stage I controls are required in 13 AQCRs: Boston,
New Jersey-New York, New Jersey-Philadelphia, Pittsburgh, Baltimore,
Washington, D.C., Houston, San Antonio, Denver, Indianapolis, Los Angeles,
and California's Sacramento and San Joaquin Valleys. Stages I and II
controls have only been implemented in the San Francisco Bay Region and
in approximately 30% of San Diego service stations. The California Air
Resources Board plans to implement Stage II controls in four additional
AQCRs (Perry, personal communication, 1978).
Separate calculations were made for (1) urbanized areas with
uncontrolled service stations, (2) Stage I controls, and (3) Stages I
and II controls. Because AQCRs cover a larger area than do urbanized
areas, it was necessary to determine the population and land area of all
the urbanized areas within a particular AQCR. For example, the New Jersey-
Philadelphia AQCR contains the urbanized areas of Philadelphia, Trenton,
New Jersey, and Wilmington, Delaware (U.S. Bureau of the Census, 1972
County and City Data Book; Bureau of Economic Analysis, 1973). The
average U.S. wind speed was determined by weighting regional average
wind speeds by population (see Table VIII-11). Total gallons of gasoline
in urban areas was estimated at 4.0 x 10 gallons, based on the 1975
9
urban miles traveled by passenger cars (600 x 10 miles) and an average
of 13.74 miles per gallon (U.S. Bureau of Census, 1977 Statistical
Abstract).
The results of this analysis are shown in Table VIII-12. Note
that the estimated levels are more than 1 order of magnitude below
concentrations measured by atmospheric monitoring. Several reasons may
account for this. As previously mentioned, the Hanna-Gifford dispersion
model is sensitive to land areas. We also assumed uniform distribution
of service stations, when in reality they are often clustered in groups
of three or more in one location. In addition, more service stations
are located in central city areas than suburban areas.
105
-------
Table VIII-11
DETERMINATION OF WEIGHTED U.S. AVERAGE WIND SPEED
Census
Bureau
' Region
Northeast
Mid-Atlantic
E. North Central
W. North Central
South Atlantic
E. South Central
W. South Central
Mountain
Pacific
1970
Population
(103)
11,883
37,274
40,313
16,518
30,805
12,839
19,388
8,348
26,600
Percent
of
Total
5.8
18.3
19.8
8.1
15.1
6.3
9.5
4.1
13.0
Average
Wind Speed
(m/s)
6.5
6.5
5.5
6.0
5.5
5.0
6.0
4.0
3.0
Wind Speed
Component
0.4
1.2
1.1
0.5
0.8
0.3
0.6
0.2
0.4
Total
203,806
100.0
5.5
Average Weighted U.S. Wind Speed =5.5 m/s.
Source: U.S. Bureau of Census, 1975 Statistical Abstract;
Figure VIII-1; and SRI estimates.
106
-------
Table VIII-12
ESTIMATES OF ANNUAL AVERAGE BENZENE CONCENTRATIONS IN URBAN AREAS
FROM GASOLINE SERVICE STATIONS BASED
ON THE HANNA-GIFFORD DISPERSION MODEL
Estimated
Exposed Estimated Benzene Benzene Annual Average
Urban % of Total Gallons Emission Emission Benzene
Type of Population Urban Pumped/Year Land Area Factor Rate Concentrations
Control (106) Population (1010) (103 km2) (g/gal) (g/s)
Uncontrolled 68.4 58 2.55 39.3 0.07 60
Stage I 45.4 38 1.7 15.5 0.04 20
(ppb)'
Stages I and
II
4.5
0.18
1.7
0.01
0.6
0.06
0.05
0.03t
* 3
To estimate p,g/m , multiply by 3.2.
Because both locations with Stages I and II controls are on the West Coast, the average wind
speed of 3 m/s was used for this calculation. If the U.S. weighted average of 5.5 m/s was used
instead, the estimated concentration would be 0.01 ppb.
Source: SRI estimates
-------
Application of the Monitoring Data—Atmospheric monitoring in
the vicinity of gasoline service stations seems to indicate that their
contribution to the urban atmosphere ranges from 1 to 2 ppb within 100 m
of the station. Because of the extent and geographic coverage of the
API monitoring data, those data were analyzed further to determine
whether they could be applied to other urban areas.
Regression analysis was used to fit a curve to the two sets of
data shown in Figures VIII-2 and VIII-3. As Figure VIII-4 indicates.
fairly good agreement was achieved between the two. Excellent correlation
was noted for the curve representing rural-mountain service stations.
Benzene concentrations ranged from 2 ppb at the station boundary to
0.5 ppb at 200 m from the station. The urban-suburban curve, on the
other hand, showed concentrations ranging from 5 ppb at the station
boundary to 3 ppb at 200 m. Correlation of this curve (r = 0.30) with
the monitoring data was not as good as the correlation of the rural-
mountain curve (r = 0.67), in part because benzene concentrations in
urban areas are influenced by many sources other than gasoline service
stations. The difference between the two curves ranges from 2.5 to 3
ppb and can be considered to be the background concentration associated
with urban areas. This assumption correlates well with other monitoring
data and dispersion modeling of urban areas discussed previously in this
chapter.
The rural-mountain curve should represent the contribution of
benzene to the urban atmosphere in the absence of other influencing
benzene sources. This curve was used in estimating the population exposed
to benzene from gasoline service stations. For purposes of extrapolation,
the population residing between 30 and 80 m from a service station is
assumed to be exposed to annual average benzene concentrations between
1 and 4 ppb. Those residing from 80 to 300 m from a service station are
assumed to be exposed to between 0.1 and 1.0 ppb. Given the available
data, precise identification of the point at which the benzene levels
drop below 0.1 ppb (the detectable limit) is impossible. Therefore,
we assume that a service station's influence on benzene concentrations
is minimal beyond 300 m.
108
-------
§ 4
01
u
8
UJ
Z
Ul
N
Ul
CO
I I I I | I T
CURVES EXTRAPOLATED BY
REGRESSION ANALYSIS AS FOLLOWS:
URBAN-SUBURBAN
C = aDb
a = 12.81
b =• -0.26
r =• 0.30
Significant at 90% confidence level
RURAL
a - 28.34
b--0.77
r = 0.67
Significant at >99% confidence level
URBAN-
SUBURBAN
RURAL-
MOUNTAIN
I
10
20
30 40 60 80 100
DISTANCE-m
200
300 400
ASSUMED
Source: SRI estimates
FIGURE VIII-4. REGRESSION CURVES DEVELOPED FROM API ATMOSPHERIC MONITORING
DATA COLLECTED IN THE VICINITY OF GASOLINE SERVICE STATIONS
109
-------
The exposed population can be estimated as follows:
1 to 4 ppb
o
(irr ) (number of service stations) (density of urbanized areas) =
popuXcLC ion
9 9 fi
u(0.08 km - 0.03 km) (83,000) (1318 people/km ) = 0.9 x 10
0.1 to 1.0 ppb
Tr(0.3 km - 0.08 km) 2 (83, 000) (1318 people/km2) = 20 x 1Q6
These estimates, which are only rough approximations, are
based on assumptions of uniform distribution of service stations in
urbanized areas, uniform pumping volumes, and average population density,
and on a curve developed by regression analysis of monitoring data. In
reality, more service stations are located in commercial areas than in
residential areas, pumping volumes vary substantially, and several
service stations are often located in the same general area. One service
station was monitored at all sites, except one, for which two stations
were monitored (based on site maps accompanying the draft API report) .
However, note that monitoring data are not often a good representation
of annual average concentrations because of shor-t-term variations in
meteorology, operating conditions, sales volume, and other possible
benzene sources in the area. Nevertheless, the correlation appears
adequate for our rough approximations. Further study is warranted to
determine a more accurate estimate of exposure levels, based on pumping
volumes, co-location of service stations, their distribution within an
urban area, and emission rates.
110
-------
IX SELF-SERVICE GASOLINE
A. Sources
Service stations are characterized by their services and business
operations: full-service stations, split island stations, self-service
stations, and convenience store operations. In full-service stations,
attendants offer all services, including gasoline pumping and other
mechanical check-ups. If fuel is obtained at any of the last three
classes of stations, the customers may fill up their tanks themselves.
In split island stations, both self-service and full-service is available.
While pumping gasoline, an individual is exposed to high benzene levels
*
released as vapor from the gasoline tank. Although occupants in the
car at both self-service and full-service operations receive some benzene
exposures, the highest exposures are received by the individual pumping
the gas. Because it is difficult to estimate level and length of exposure
for occupants, only those individuals obtaining gasoline from self-service
pumps are considered. (It is not within the scope of this report to
evaluate occupational exposures.)
Benzene content of evaporative gases increases and decreases during
evaporation, depending on the system temperature and the relative
volatilities of all the components of the fuel (Mitre, 1976). Recent
information indicates that gases released during automobile fill-ups have
little relationship to the benzene content in the gasoline. Rather, the
ambient temperature relative to the temperature of the gasoline has the
most significant effect, and most of the exposure results from the benzene
vapor trapped within the tank, not from the gasoline being pumped (Johnson,
1977). If the gasoline is cold relative to the tank (as in summer),
Vapor recovery systems can reduce exposure levels significantly, if
properly working and operated. Such systems are required for service
stations in parts of California.
Ill
-------
most of the benzene vapor will be abosrbed into the gasoline. On the
other hand, if the gasoline is warm relative to the tank (as in winter),
the benzene vapor will be displaced rather than absorbed and more signi-
ficant exposures will result.
Self-service dispensing of gasoline is a relatively new marketing
method pioneered by independent operators on the West Coast and in the
southern United States. Today, it accounts for 30% of gasoline sold.
The national market-share of the major gasoline producers has decreased
recently as independents and others specializing in high-volume, low-
margin sales capture a larger percentage. Of the approximately 184,000
conventional service stations and tie-in gasoline operations in the
United States, service stations with some self-service operations account
for 39% (ADL, 1977b). Table IX-1 indicates the types of service stations
offering self-service gasoline.
Table IX-1
SELF-SERVICE OPERATIONS
Percent of
Outlets Offering Self-Service U.S. Total
Total self-service 9
Split island with self-service 26
Convenience stores 4
Total outlets with self-service 39
A recent ADL report (1977b) revealed that there are 71,300 outlets
with self-service gasoline. Gasoline sold for the year ending May 30,
9
1977, equals approximately 87.4 x 10 gal in the United States. Of this
9
amount, 27.0 x 10 gal (31%) was dispensed at self-service pumps. The
market-share of self-service stations was surveyed for four metropolitan
Air Quality Control Regions (AQCR): Boston, Dallas, Denver, and Los
Angeles. The market-share held by self-service operations varied from
9% in Boston to 45% in Denver (see Table IX-2). Another study by applied
Urbanetics, Inc. (1976) surveyed Baltimore and Madison, Wisconsin. The
112
-------
Table IX-2
.GASOLINE MARKET SHARE OF SELF-SERVICE STATIONS
IN FOUR AQCRs, SPRING 1977
Type of Operation
Boston AQCR
Full-service
Self-service (total)
Split island
Self-service
Convenience stores
Dallas AQCR
Full-service
Self-.service (total)
Split island
Self-service
Convenience stores
Denver AQCR
Full-service
Self-service (total)
Split island
Self-service
Convenience stores
Los Angeles AQCR
Full-service
Self-service (total)
Split island
Self-service
Convenience stores
Number of
Outlets
2,253
100
8£
92
621
656
310£
226
120
2,518
4,780
3,632£
1,022
126
Sales
Volume
(106 gal/yr)
1,045.1
108.6
292.1
235.7
2,472.6
2,154.8
Market
Sharing
Percent
91%
9
2,094
1,124
480a
444
200
924.
593.
6
8
61
39
55
45
53
47
Split island operations offering full service and self-serve
islands.
Of these 445 are split island operations that offer full service
and mini-serve (attendant-operated) islands.
Source: ADL (1977b).
113
-------
results of this study are shown in Table IX-3. It appears that about
40% of the market in urban areas is accounted for by self-service operations.
Table IX-3
GASOLINE MARKET SHARE OF SELF-SERVICE
STATIONS IN TWO METROPOLITAN AREAS, 1976
Type of Operation
Baltimore SMSA
Full-service
Self-service (total)
Split island
Self-service
Madison SMSA
Full service
Self-service (total)
Split island
Self-service
Sales
Volume
(106 gal/yr)
111.5
90.5
25.5
65.0
56.Oc
77.0
17.0
60.0
Market
Sharing
Percent
55%
45
42
58
Includes the sales from mini-serve (attendant-
operated) stations and 50% of the sales from
split islands.
Source: Applied Urbanetics, Inc. (1976).
B. Methodology and Exposures
To estimate the people exposed to benzene from this source, several
assumptions were necessary. The gasoline pumped through self-service
9
outlets is estimated at 27.0 x 10 gal. The annual average fuel con-
sumption per vehicle is 736 gal (U.S. Federal Highway Administration,
1974). If it is assumed that on the average, a person who primarily
uses self-service gasoline makes one trip there per week, an average
fill-up amount of 14 gal is determined by dividing 736 gal/vehicle/yr
by 52 wk/yr. By dividing the average fill-up into the self-service
114
-------
gallons pumped, we estimate trips per year to self-service operations
9
at 1.9 x 10 . When this number is divided by 52 trips per person per
year, the people exposed to benzene from this source is estimated at
37 x 10 . This estimate of the population exposed assumes that the
individuals using self-service gasoline never obtain gasoline at full-
service stations.
Battelle conducted a preliminary study (1977b) to determine the
benzene exposure levels from self-service gasoline pumping. Three
samples of ambient air were taken in the breathing zone of persons
filling their tanks. The results, shown in Table IX-4, indicate a wide
range in the benzene concentrations of the emissions. The variations
seem to be related to the subject's position in relation to the tank
opening and the wind direction. Because all measurements were taken on
the same day and at approximately the same time, ambient temperature did
not cause the variation. Basically, if the subject was downwind of
the tank opening, higher levels were recorded. The time-weighted average
concentration of benzene from the three samples is 245 ppb. The average
length of time taken to fill up a gasoline tank is 1.7 min. Although
14 gal per fill-up is assumed, the wide range in pumping speeds does not
allow a precise estimate of time required per fill-up.
Table IX-4
SAMPLING DATA FROM SELF-SERVICE GASOLINE PUMPING
Customer
1
2
3
Sampling Rate
(mL/min)
31
31
31
Nozzle
Time (min)
2.5
1.1
1.6
Gallons
Pumped
14
8
9
Sample
Volume
(L)
78
34
50
Benzene
Mg/m
115
324
1740
Level
ppb
43
121
647
Source: Battelle (1977b).
115
-------
The estimated exposure levels are based on the information contained
in Table IX-4. It is recognized that these data are quite limited and
highly variable. In states where vapor recovery systems are used, the
estimated exposure level may be much lower. It can be estimated that
approximately 37 x 10 persons use self-service stations. While filling
their tanks once a week, they are exposed to a benzene level of 245 ppb
for 1.7 minutes. Their annual exposure is estimated at 1.5 hr. (Table
IX-5 summarizes this information.)
Table IX-5
ESTIMATED POPULATION EXPOSED TO
BENZENE FROM SELF-SERVICE GASOLINE
Exposure
Type
Self-service
pumping
Exposure
Time
Annual
Exposure
Population Exposed
to Benzene Concentrations
(ppb)*
245.0
Total
1.7 min 1.5 hr 37,000,000 37,000,000
* 3
To convert to yg/m , multiply concentrations by 3.2; to convert
annual average exposures to 8-hour worst case, multiply concen-
trations by 25.
Source: SRI estimates.
116
-------
X ASSESSMENT OF TOTAL EXPOSURE
A. Introduction
Because benzene sources are widespread, particularly within the
urban environment, the exposures of some individuals may be significantly
different from those estimated in Chapters III through IX. The results
presented in these previous chapters are based on the assumption that
individuals living near a source spend 24 hours of each day in that
location. In fact, however, many individuals travel from their place
of residence to work or shop in other areas. Thus, they are exposed
to varying levels of benzene concentrations throughout the day. In this
chapter, a rough analysis is made to estimate total exposure to the
urban population: Total exposure is the sum of an individual's exposure
to all benzene sources over a designated period (e.g., a week or a year)
and in varying locations; this exposure, therefore, represents a more
realistic estimate of population exposure then indicated previously.
In this approach we developed several scenarios that represent
typical living patterns of individuals residing in the vicinity of
benzene sources. Percentages of time spent in various activities such
as sleeping, shopping, commuting to work, traveling on personal business,
and working were estimated by using the findings of sociological studies.
Assumptions were then made about residence locales and work locations
(for those that work). and about what percentage of the population falls
into each of these categories. The benzene exposure settings typical
for each activity were determined from dispersion modeling and atmospheric
monitoring, data described in Chapters III through IX. The percentage
of time spect in 1 year in various activities was multiplied by the
associated benzene exposure setting (ambient benzene concentration
measured in ppb) and summed to determine an individual's total exposure.
Note, however, that because so few data exist, this approach necessarily
has limitations and is considered a first-cut analysis.
117
-------
Estimates of total exposure are linked to each source category
(e.g., chemical manufacturing facilities, petroleum refineries, or urban
areas). Our approach assumes that those living in the vicinity of one
point source, such as a coke oven, are unaffected by other benzene sources
until they move out of that point source's sphere of influence. In the
analysis presented in this Chapter, variations in exposure levels result
from travel away from the residence by commuting to work, shopping, or
traveling for personal business. Although the estimates given here
improve our understanding of expected exposures to individuals in the
vicinity of sources, they do not help in understanding how overlapping.
sources might affect individual exposures. For example, Houston, Texas,
has six chemical manufacturing facilities, five petroleum refineries,
and one coke oven. Without atmospheric monitoring data, it is impossible
to estimate precise total exposures to individuals in that city.
This analysis thus provides a more realistic evaluation of exposures
to individuals than is provided by source-specific analysis. Although
many assumptions are required, the estimates used in this chapter rely
as much as possible on studies of human behavior and on statistical
population data. Because individuals generally spend more than 60% of
the time at home under the annaal average condition, short-term monitoring
data have been combined with annual average estimates to estimate total
exposure. Cities with highly directional wind roses or more than one
large benzene source may have significantly different conditions from
those estimated. In addition, it is assumed that outdoor benzene con-
centrations are similar to indoor concentrations. However, no indoor
monitoring has been conducted to verify this assumption. Such problems
will remain unresolved until much more atmospheric monitoring is
conducted.
B. Determination of an Individual's Use of Time
The first step in estimating total exposure is to determine an
individual's use of time. John P. Robinson of the Communication Research
Center at Cleveland State University (1977) has extensively researched
118
-------
Americans use of time. For his survey, he required individuals to keep
time diaries of the daytime and nightime activities over a 1-month period-
a technique whose basic reliability and validity had been established by
several earlier methodological studies. A time diary study was conducted
for the first time in 1965 and was repeated in 1975 to evaluate changes
over the 10-year period. Survey results were presented in terms of the
sex, employment, and marital status of those surveyed.
Because we were interested in estimating the amount of time spent
away from the residence, we needed data for those employed full-time.
The results for employed men shown in Table X-l provide the best measure
of that. No reliable data were available for employed women. (Part-time
female workers were included in the statistics for employed women in
Robinson's study, thus tending to bias the results.)
Table X-l
PERCENTAGE OF TIME SPENT PER WEEK IN
MAJOR TYPES OF ACTIVITIES BY EMPLOYED MEN
IN URBAN AREAS, 1975
Activity % of Time per Week
Sleep 32%
Work for pay 26
Family care 6
Personal care 12
Leisure (total) 24
Organizations 2.5
Media 11
Social life 4.5
Recreation 2
Other leisure 4
Source: Robinson (1977).
119
-------
To estimate commute time, travel time for personal business, and
shopping time, the activities contained within each major activity shown
in Table X-l must be clearly understood. "Work for pay" includes all
work, plus breaks, and commuting. "Family care" includes shopping, house-
work, child care, and helping others. "Personal care" includes all
hygiene, meals, and travel associated with family and personal care.
The time spent on adult education and organized activity, including
religion, are part of the "organizations" category. "Media" includes
radio, television, reading, and movies. "Social life" includes the time
spent on entertainment, social visits, and conversation. "Recreation"
includes sports, outdoor activities, and walking. The "other leisure"
category includes hobbies, records, letters, resting and relaxing, and
all leisure travel.
Estimates of commuting time were obtained from the Nationwide
Personal Transportation Survey conducted by the Federal Highway Adminis-
tration (Department of Transportation, 1973). As shown in Table X-2,
commute time from home to work was reported in terms of population by
place of residence. The average for incorporated localities was 21
minutes. Assuming two weeks of vacation per year, individuals spend 2%
of their time per year commuting. For an upper limit, those living in
incorporated areas with populations exceeding 1,000,000 spent 3% of their
time commuting.
If the limitations of the available monitoring data are considered,
an individual is assumed to be exposed to similar levels of benzene
whether commuting to work during rush hour or driving on freeways at
other times. As a result, the times estimated.for each of these activities
were combined into one category.
Available surveys lack specificity in regard to shopping or to
driving a car for nonwork activities. Consequently, we developed best-
judgment estimates of the percentages of time that could be attributed
to these activities. Visits to rural areas for vacation or other leisure
activities were estimated to range from 2 to 4% of time spent. The
amount of time spent shopping in suburban locations or downtown areas
120
-------
Table X-2
PERCENT OF THOSE EMPLOYED BY PLACE
OF RESIDENCE AND COMMUTING TIME TO WORK
Population Place
of Residence
Unincorporated areas
Incorporated areas
Less than 5,000
5,000 - 24,999
25,000 - 49,999
50,000 - 99,999
100,000 - 999,999
1,000,000 and above
All incorporated places
All areas and places
Home-to-Work
15 & less
50.9
60.3
58.4
61.1
56.4
51.3
28.5
53.2
52.5
16-25
19.1
16.8
16.2
17.1
21.5
22.8
18.5
19.0
19.1
Commuting Time (min)
26-35
13.1
12.9
13.3
11.8
10.7
16.0
20.5
14.4
14.0
36 & more
16.7
10.0
12.1
10.0
11.4
9.9
32.5
13.4
14.4
All
100.0
100.0
100.0
100.0
100.0
100.0
100.00
100.0
100.0
Average
for all
Workers
(min)
23
18
19
19
20
21
32
21
22
Does not include workers that work at home or at no fixed address.
Source: Department of Transportation, Federal Highway Administration, 1973.
-------
was estimated at 2 to 3% and was assumed to be distributed equally
between the two locations.
The results of this analysis are shown in Table X-3. Nine scenarios
describe most of the possible combinations for residing and working in
one of three locations: vicinity of source, suburb, or central city.
Place of residence and place of work are the same if an individual works
near his/her residence, goes to school, does not work, is retired, or is
an invalid. Therefore, "work" can loosely be defined as daytime location.
C. Distribution Into Population Subgroups
Although the precise number of people maintaining a particular life
style in the vicinity of each source or in an urban area cannot be
accurately established, reasonable estimates can be made by grouping
those individuals into scenarios that represent common living patterns
(as shown in Table X-3). The next step toward estimating total exposures
is to determine the percentage of the exposed population that can be
assigned to each scenario.
A first-cut assignment is made by determining the percentage of the
population that stays at or near home the majority of the time. Census
data indicate that 48% of the population is likely to be in school
(elementary; junior high, and high school), or retired (65 and older).
The labor force distribution for 1974 shows that another 14% of the total
population, that is those between the ages of 20 and 65, does not work.
Therefore, roughly 60% are constantly in the vicinity of their homes.
The remaining 40% of the population are assumed to work in the vicinity
of their homes or to commute to other locations. The results of this
analysis are shown in Table X-4. Best judgment was used to develop these
percentages.
Very few quantitative data are available to estimate the commuting
behavior of people. In this analysis, only 20% of those individuals re-
siding in central cities are assumed to commute outside those areas to
work. Working individuals residing in suburban areas are assumed to
commute outside those areas for work. On the other hand, those living
122
-------
Table X-3
ESTIMATED ANNUAL DISTRIBUTION OF
TIME SPENT IN VARIOUS ACTIVITIES AND LOCATIONS
to
Percentage of Time Spent
Scenario
Reside
Vicinity of
source*
Vicinity of
source
Vicinity of
source
Suburb
Suburb
Suburb
Central city
Central city
Central city
Work
Vicinity of
source
Suburb
Central city
Suburb
Vicinity of
source
Central city
Central city
Vicinity of
source
Suburb
Commute/
Freeway
1
4
4
1
4
4
1
4
4
Central
City
1
2
26
1
3
27
96
68
68
Suburb
1
26
2
94
65
65
1
2
26
Vicinity*
of Source Rural
93 4
64 4
64 4
+ 4
24 4
+ 4
+ 2
24 2
+ 2
Total %
Time
Spent
100%
100
100
100
100
100
100
100
100
Source indicates chemical manufacturing facilities, coke ovens, and petroleum refineries.
Assumed to be less than 1%.
Source: SRI estimates.
-------
Table X-4
DISTRIBUTION OF THE POPULATION INTO GROUPS
AFFECTED BY OTHER BENZENE EXPOSURE SETTINGS
Scenario
Reside
Vicinity of source
Vicinity of source
Vicinity of source
Total-Source
Work'
Vicinity of source
Suburb
Central city
% of
Population
65
10
25
100
Suburb
Suburb
Suburb
Total-Suburb
Suburb
Vicinity of source
Central city
70
5
25
100
Central city
Central city
Central city
Total- Central city
Central city
Vicinity of source
Suburb
80
5
15
100
Source: SRI estimates.
124
-------
in the vicinity of coke ovens, chemical plants, or petroleum refineries,
may already be situated in either a suburban or a central city area.
Thus, in this analysis, it is assumed that they travel, shop, and/or
commute to areas outside the sphere of influence of the facility. The
fact that some of these facilities are in remote areas was not specifically
considered in this analysis.
D. Selection of Applicable Exposure Data
Monitoring data have been collected in an attempt to characterize
benzene concentrations near such locations as freeways, intersections,
and central city areas. At the request of EPA, Battelle (1978b) collected
atmospheric monitoring data for Celumbus, Ohio, in suburbs, in central
city areas, and during peak commute periods. The results of their study
are shown in Table X-5. The 23-hour average for Site 1 in the central
city was 3.9 ppb, whereas the daytime average was 4.4 ppb. This site
was described as a midtown intersection with a high traffic load throughout
normal business hours. Site 2 was located adjacent to a major commuting
highway leading into the business district. Grab samples taken during
peak volume traffic gave an average of 6.4 ppb. The average concentration
of four 8-hour samples at Site 2 was 4.1 ppb, and the daytime average
*
was 5. Site 3 was located in a residential neighborhood approximately
1 mile from the central business district with a low service station
density and no other major known stationary sources of benzene. The
24-hour average benzene concentration measured at Site 3 was 1.5 ppb.
The city of Columbus measured traffic volume at Site 2 during the
atmospheric monitoring (see Table X-6). The density of traffic flow in
(cars/liter sampled) appears to be correlated with the measured benzene
concentrations at the monitoring site on the same side of the highway
(n = 8, r = 0.82). The consistently lower density on the eastbound side
*
Two other 8-hour samples and three grab samples were lost at this
location, thereby making these data statistically open to question.
125
-------
Table X-5
RESULTS OF BATTELLE ATMOSPHERIC MONITORING STUDY
CT)
Site*
No.
3
3
3
3
2a
2a
2a
2a
2b
2b
2b
2b
1
1
1
1
*
Kev.
Trap
No.
54
260
3
8
95
143
129
209
6
175
130
39
148
100
20
70
Sampling Period
Date
3/15
3/15
3/15
3/15-3/16
3/15
3/15
3/15
3/16
3/15
3/15
3/15
3/15-3/16
3/15
3/15
3/15
3/15-3/16
Start
6:45 AM
2:16 PM
2:11 PM
9:57 PM
7:43 AM
7:43 AM
3:28 PM
6:58 AM
8:12 AM
3:42 FM
3:42 PM
11:20 PM
7:12 AM
3:01 PM
2:55 PM
10:37 PM
Finish
2:16 PM
9:57 PM
9:57 PM
8:04 AM
9:45 AM
3:28 PM
5:54 PM
9:01 AM
3:42 PM
5:59 PM
11:20 PM
9:07 AM
2:55 PM
10:30 PM
10:30 PM
8:15 AM
Barometer,
mmHg
29.38
29.38
29.38
29.35
29.38
29.38
29. J8
29.35
29.38
29.38
29.38
29.35
29.38
29.38
29.38
29.35
Temp,
F
34-39
35-40
35-40
33-40
33-36
33-40
37-40
32-33
33-40
37-40
37-40
32-37
33-39
37-40
37-40
32-40
Wind
Speed
Direction
150-350
350-360
350-360
45-360
150
150-350
350-360
330-360
150-350
350-360
350-360
45-360
150-350
350-360
350-360
45-360
m/s
1.66-2.19
1.17-1.95
1.17-1.95
0.86-2.19
1.66-1.87
1.66-2.19
1.45-1.80
1.40-2.19
1.66-2.19
1.45-1.80
1.00-1.95
0.86-2.19
1.66-2.19
1.00-1.95
1.00-1.95
0.86-2.19
(mi/hr)
(3.6-4.7)
(2.5-4.2)
(2.5-4.2)
(1.9-4.7)
(3.6-4.0)
(3.6-4.7)
(3.1-3.9)
(3.0-4.9)
(3.6-4.7)
(3.1-3.9)
(2.2-4.2)
(1.9-4.7)
(3.6-4.7)
(2.2-4.2)
(2.2-4.2)
(1.9-4.7)
RH
78-61
67-57
67-57
60-75
78
78-61
57-62
77
78-61
57-62
57-67
59-75
87-61
57-67
57-67
59-75
Conditions
drizzle,
then
overcast
overcast
overcast
drizzle,
then
overcast
overcast
overcast
drizzle,
then
overcast
overcast
overcast
drizzle,
then
overcast
overcast
overcast
Benzene
Cone
2.0
1.4
1.4
1.3
8.6
7.5
6.9
4.8
3.4
5.3
3.3
2.2
5.1
5.3
2.9
2.4
Site 1 = downtown Columbus, Ohio.
Site 2 = near downtown Columbus, Ohio; major commute route; a and b are on opposite sides of the street.
Site 3 = suburban residential Columbus, Ohio.
Source: Battelle, 1978b, draft report.
-------
Table X-6
TRAFFIC VOLUME AND DENSITY DURING BENZENE SAMPLING PERIODS AT SITE 2
N>
Trap
No. Date
95 3/15
143 3/15
129 3/15
209 3/16
6 3/15
175 3/15
130 3/15
39 3/15-3/16
Sampling Period
Start Finish
7:43 AM 9:45 AM
7:43 AM 3:28 PM
3:28 PM 5:54 PM
6:58 AM 9:04 AM
8:12 AM 3:42 PM
3:42 PM 5:59 PM
3:42 PM 11:20 PM
11:20 PM 9:07 AM
Volume of
Air Sampled
(«.) One-Way Traffic
Sampling Site 2a, 518 East Broad
Westbound Traffic
(same side of
highway as
sampler)
3.7 3,868*
10.4 10,832*
4.0 3,870
2.9 4,465
Sampling Site 2b, 547 East Broad
Eastbound Traffic
(same side of
highway as
sampler)
13.8 6,799*
4.5 4,082
13.8 7,231
17.7 2,194
Total
Traffic
Street
5,132*
17,537*
8,402
5,645
Street
16,741*
7,652
13,625
8,352
Density (Cars/Liter
sampled)
One-Way Traffic Total Traffic
Westbound
Traffic
1045
1041
968
1540
Eastbound
Traffic
493
907
524
124
1387
1686
2100
1947
1213
1700
987
472
Portions of this count were estimated from the following morning's count.
Source: City of Columbus, Division of Traffic Engineering, as cited in Battelle draft report (1978b).
Cone of
Benzene
in Air
(ppb)
8.6
7.5
6.9
4.8
3.4
5.3
3.3
2.2
-------
may, to a large extent, explain why the sampling stations on that side
registered lower readings.
As discussed in Chapter VIII, most of the limited atmospheric moni-
toring data have been collected in the vicinity of urban areas. Robinson
et al. (1973) reported the results of studies measuring hydrocarbon
concentrations in the nonurban atmosphere. The measurements taken in
the continental United States are shown in Table X-7. They range from
0.3 to 1.1 ppb, with an average of 0.7 ppb. These results appear to be
consistent with the widespread nature of benzene emissions and benzene's
low atmospheric reactivity.
Radian Corporation (1978) monitored ambient benzene in Burnet, Texas,
(a rural community) for Shell Oil Corporation. Three 24-hour samples
were collected on consecutive days under variable meteorologic conditions.
Complete quality control information was unavailable for these data.
Measured benzene concentrations ranged from 3 to 4 ppb. The measurements
are not consistent with others taken in remote areas. No explanation
was offered for these results.
Table X-7
BENZENE CONCENTRATIONS IN REMOTE AIR
SAMPLES FROM THE CONTINENTAL UNITED STATES
State
Location
Date
Benzene
Concentration
(ppb)
California
Idaho
Idaho
Vermont
Washington
Point Reyes
Hells Canyon
Sample 1
Sample 2
Moscow Mountain
Ground
Airborne
Camel's Hump
Brethway-Gunderson
Ground
Airborne
12/9/71
11/6/71
7/16/71
9/8/71
8/4/71
0.3
0.8
0.8
0.9
0.8
1.1
0.4
0.5
Source: Washington State University (1971) as cited in
Robinson et al. (1973).
128
-------
In the absence of adequate atmospheric monitoring data, dispersion
modeling was undertaken to provide approximate estimates of benzene
concentrations in the vicinity of specific locations. Chapters III through
V describe dispersion modeling analysis and results for three benzene
point sources. In Chapter VIII, the Hanna-Gifford dispersion equation
was used to estimate average annual benzene concentrations in suburban
areas (see Table VIII-5). If a background concentration of 0.5 ppb is
included, the estimated concentrations ranged from 0.73 to 2.0 ppb for
suburban areas in six large*cities and appear to be reasonably close to
the suburban values found in Battelle's atmospheric monitoring study
(see Table X-5).
The highest benzene concentrations determined by dispersion modeling
for a short period were estimated in the vicinity of major intersections.
As shown in Table X-8, 8-hour average concentrations ranged from 12 to
28 ppb. Annual average benzene concentrations, however, are similar to
those measured by Battelle along a major commuting arterial (Site 2 on
Table X-5).
The selected benzene exposure settings for determining total exposure
are shown in Table X-9. These estimates represent a combination of the
best monitoring and modeling data available to date and our best judgment.
Although limitations may be involved in using dispersion modeling data
in conjunction with monitoring data, the methodology employed in this
type of analysis does provide a more realistic evaluation of total
exposure to an 4-ndividual.
Several additional studies are worth noting. A recent study by
Messer Associates, Inc., for the U.S. Department of Transportation (1977)
analyzed the health effects of bicycling in an urban environment by
studying 10 male subjects bicycling or driving through the streets of
Washington, D.C., under systematically varied conditions. No major
adverse short-term effects were noted, but the carboxyhemoglobin (COHb)
levels measured for the motorist controls were slightly higher than those
for the bicyclists. The authors theorized (1) that the carbon monoxide
(CO) encountered in traffic slow-downs and intersection delays may have
129
-------
Table X-8
BENZENE CONCENTRATIONS IN THE VICINITY OF
MAJOR INTERSECTIONS BASED ON DISPERSION MODELING
Number of
Lanes
8-Hour Average
Benzene Con-
Average Annual
Benzene Con-
City
Hartford, CT
Denver, CO
Waltham, MA
Washington, DC
Peoria, IL
Springfield, IL
Oakbrook, IL
Location
Buckingham St. at
Washington St.
Colfax Ave. at
Colorado Blvd.
Moody St. at
Carter St.
Wisconsin Ave. at
Western Ave.
Rt. 150 at
Scenic Dr.
MacArthur Blvd.
at S. Grand Ave.
Rt. 83 at 22nd St.
(main
cross
4 x
6 x
4 x
6 x
8 x
7 x
4 x
road x
road)
4
8
3
6
5
6
4
Distance^
(m)
8
22
24
24
20
22
20
centration
(ppb)
28
37
28
. 28
22
12
23
centration
(ppb)
6.9
4.7
6.9
6.9
5.3
3.1
5.6
These locations are all signalized intersections. Number of lanes includes turn lanes in all cases.
Wind speed is 1 m/s.
Distance from nearest roadway to major receptor.
Source: Schewe (1977) using the Hot Spot Guideline.
-------
u>
Table X-9
BENZENE EXPOSURE SETTINGS IN VICINITY OF SPECIFIC LOCATIONS
Average
Location
Major intersections
Freeway /peak commute
Central city
Suburb
Source
Rural
Benzene
Concentrations
(ppb)
12 - 28
6 - 7
3-4
1-2
Varies by source
ND - 1.0
Averaging
Time
(hr)
8
2-3
24
24- Annual
Annual
0.5 - 1
References
Dispersion modeling: Schewe
(Table X-8) .
Monitoring data: Battelle,
(Table X-5).
Monitoring data: Battelle,
(Table X-5).
Monitoring data: Battelle,
(Table X-5).
Dispersion modeling: SRI
(Chapter VIII) .
Dispersion modeling: SRI
(Chapters III, IV, V).
, 1977
1978
1978b
1978b
Monitoring data: Robinson et al. ,
1973 (Table X-7).
"Source" means chemical manufacturing facilities, coke ovens, and petroleum refineries.
ND = non-detectable
-------
more impact on increasing levels of COHb than the low levels of CO
encountered by moving bicyclists and motorists while traveling through
the study area, and (2) that motorists may be exposed to higher CO levels
than bicyclists who have more mobility in heavy traffic. An additional
finding indicates that CO levels monitored at the study's permanent air
monitoring station were consistently lower than the CO levels encountered
by bicyclists and motorists in the study.
Another study conducted by Chaney of the University of Michigan
(1977) monitored CO concentrations inside a traveling car. Some of the
more important findings were (1) peaks in concentration are primarily
the result of traffic slowing down as a result of congestion; (2) CO
concentration varied from 2 to 50 ppm, depending primarily on the number
of stationary or accelerating vehicles close to the monitored vehicle;
(3) measurements on 760 vehicles showed a wide variation in CO concen-
trations ranging from 0.05 ppm to 45 ppm; and (4) a large proportion of
the vehicles monitored produced only a small amount of the total CO,
whereas a relatively small number of the total vehicles contributed
almost half of the total monitored CO.
The results of these two studies indicate the concentrations to
which motorists are exposed from automotive-related emissions, including
benzene, are higher than measured at permanent atmospheric monitoring
stations. Therefore, the use of monitoring data to represent short-term
peak exposures from traffic flow may significantly underestimate actual
exposures for motorists.
E. Summary of Exposures
The estimates for total exposure were calculated by the following
steps
Determine the number of people within each scenario and for
each source by multiplying the exposed population (Table 1-1)
by the percentages shown in Table X-4. Urban population is
broken down into suburban and central city subgroups by
assuming that the population exposed to the higher range in
132
-------
Table 1-1 is located in central cities. The population
residing near service stations is exposed to.higher overall
benzene levels (additive to the estimated urban exposure
from automobile emissions).
2. Multiply the percentage of time spent shown in Table X-3 by
the mid-range of the applicable benzene exposure settings
shown in Table X-9. Sum the values for each scenario., This
represents the total exposure in ppb for an individual within
each scenario. Table X-10 shows the calculated values for
each of the scenarios.
3. Determine ppb-person-years by multiplying the total exposure
levels by the number of exposed individuals and by summing
the results. Example calculations for two benzene sources are
shown in Table X-ll.
The total exposure to individuals living in the vicinity of benzene
sources is shown in Table X-12. In comparing this table with Table 1-1,
several important differences are evident. Urban exposures, for example,
now include automobile emissions and gasoline service stations. The other
significant change is the redistribution of the people exposed within the
first two ranges of benzene concentrations. The' exposure of more than 90%
of the people in the lowest range has increased sufficiently to put them into
into the second range. The largest change occurred in the urban exposures
category and results from summing the exposures from automobile emissions
and gasoline service stations. In addition, approximately 37 million people
are estimated to use self-service gasoline stations and are estimated to be
exposed to 245 ppb at each trip with an annual exposure time of 1.5 hr/yr.
This analysis shows that lifestyle patterns affect annual average
benzene exposures and that generally the effect is to increase those
exposures. This approach appears to be better than a source-specific
analysis alone because it combines several urban exposures into one
estimate of annual exposure. The current state of the social sciences
does not allow a more sophisticated estimate of time spent in certain
133
-------
activities or the number of people conforming to certain living patterns,
Nevertheless, a rough approximation of those uncertainties affords a
reasonable evaluation of total exposure to individuals residing in the
vicinity of sources of atmospheric benzene.
134
-------
Table X-10
ESTIMATE OF TOTAL EXPOSURE FOR EACH SCENARIO
Scenario
Median Total Exposure (ppb)
t
Reside
Vicinity
Vicinity
Vicinity
Vicinity
Vicinity
Vicinity
Vicinity
Vicinity
Vicinity
Vicinity
Vicinity
Vicinity
Suburb
Suburb
Suburb
of
of
of
of
of
of
of
of
of
of
of
of
source
source
source
source
source
source
source
source
source
source
source
source
(0.1-1.0)
(1-4)
(4-10)
(> 10)
(0.1-1.0)
(1-4)
(4-10)
(> 10)
(0.1-1.0),
(1-4)
(4-10)
(> 10)
Vicinity
Vicinity
Vicinity
Vicinity
Suburb
Suburb
Suburb
Suburb
Central
Central
Central
Central
Suburb
Vicinity
Central
Work
of source (0.1-1.0)
of source (1-4)
of source (4-10)
of source (> 10)
city
city
city
city
of source""
city
Residential
0
2
7
20
0
2
7
20
0
2
7
20
1
1
1
.5
.5
.0
.5
.5
.0
.5
.5
.0
.0
.0
.0
Mixed Activity
0.
2.
6.
19
1.
2.
5.
14
1.
2.
5.
14
1.
1.
1.
6
5
6
1
3
2
5
8
7
1
3
9
-------
Table X-10 (concluded)
* t
Scenario. Median Total Exposure (ppb)
Reside Work " Residential Mixed Activity
Central city Central city 3.5 3.5
Central city Vicinity of source' 3.5 3.9
Central city Suburb 3.5 3.0
*
Because annual average exposure varies in the vicinity of a source, a separate estimate is
made for each range of benzene exposures as shown in parenthesis after applicable scenarios.
Source is used to mean chemical manufacturing facilities, coke ovens, and petroleum refineries.
Median total exposure is the mid-point of the benzene ranges. Residential exposures are the
estimates shown in Table 1-1 and imply that 24 hours of each day are spent in the vicinity of
the source. Mixed activity total exposure is the median exposure assuming that some time is
spent in other benzene exposure settings.
'A population-weighted exposure of 1.2 ppb was estimated and used to characterize levels in the
vicinity of sources.
Source: SRI estimates.
-------
Table X-ll
EXAMPLE CALCULATION OF TOTAL EXPOSURE
FOR TWO BENZENE SOURCE CATEGORIES
CO
% of
Reside Work Population
Petroleum Refineries
Benzene Concentration Range = 0.1-1.0 ppb
Vicinity of source Vicinity of source
Vicinity of source Suburb
Vicnity of source Central city
Benzene Concentration Range = 1.1-4.0 ppb
Vicinity of source Vicinity of source
Vicinity of source Suburb
Vicinity of source Central city
Total - Petroleum Refineries
Urban-Central City
Benzene Concentration Range = 1.1-4.0 ppb
Central city Central city
Central city Vicinity of source
Central city Suburb
Subtotal - Central city
65
10
25
65
10
25
80
5
15
Estimated Exposed Comparison
Exposure Level Population Among Sources
(ppb) (106) (106 ppb-person-yr)
0.6 3.25
1.1 0.5
1.5 1.25
2.5 0.002
2.3 0.0003
2.8 0.00075
5.0
3.5 36
3.9 2
3.0 7
45
2.0
0.6
1.9
0.005
0.0007
0.002
4.5
126.0
7.8
21.0
154.8
-------
Table X-ll (concluded)
OJ
00
Reside
Work
% of
Population
Urban-Suburban Locations
Benzene Concentration Range = 0.1-1.0 ppb
Suburb
Suburb
Suburb
Subtotal - Suburban
Suburb
Vicinity of source
Central city
70
5
25
Estimated
Exposure Level
(ppb)
1.1
1.3
1.9
Exposed
Population
(106)
Total - Urban exposures
48
4
17
69
110
Comparison
Among Sources
(10 ppb-person-yr)
52.8
5.2
32.3
90.3
250.0
Rounded to two significant figures.
Source: SRI estimates.
-------
Table X-12
SUMMARY OF ESTIMATED TOTAL EXPOSURES OF PEOPLE
RESIDING IN THE VICINITY OF ATMOSPHERIC BENZENE SOURCES
Vicinity of Residence
Number of People Exposed
Annual Average Benzene Concentrations (ppb)
0.1-1.0 1.1-4.0 4.1-10.0 > 10.0
Totalt
Comparison
Among
Sourcestt
(106 ppb-person-year)
Co
vO
Chemical manufacturing
Coke ovens
Petroleum refineries
Urban areas
3,900,000 3,100,000
200,000 100,000
3,250,000 1,750,000
110,000,000
200,000 80,000
7,300,000
300,000
5,000,000
110,000,000
10.0
0.2
4.5
250.0
The term "total exposures" is used to mean the sum of an individual's exposure to atmospheric benzene
from a variety of activities during a year. This assumes that people spend part of their time away from
their residence, resulting in exposures to different benzene concentrations depending on their activity
(i.e., commuting to work, shopping, traveling on personal business). Nonurban exposures are not included
in this analysis but are expected to range from undetectable to 1.0 ppb.
Rounded to two significant figures.
'The median values shown in Table X-10 were used for this calculation instead of the mid-point of the
ranges. This allows a better comparison with Table 1-1.
Source: SRI estimates.
-------
BIBLIOGRAPHY
American Petroleum Institute, "API Environmental Benzene Range Finding
Report-Summer 1977," draft report (1977).
Anderson, E.V., "Output of the Top 50 Chemical Drops Sharply," Chemical
and Engineering News, pp. 34-37, (May 1976).
Applied Urbanetics, Inc. "Market Share Study," FEA Contract No. CO-06-
60435, p. 11-13 (1976).
Ayers, G. W., and R. E. Muder, "Benzene," in Kirk-Othemer Encyclopedia
of Chemical Technology, A. Stanton (Ed.), 2nd Edition, John Wiley
and Sons, Inc., New York, Vol. 3, pp. 367-401 (1964).
Battelle-Columbus Laboratories, preliminary report to U.S. Environmental
Protection Agency, Office of Toxic Substances, concerning
"Sampling in Vicinity of Benzene Production and Consumption
Facilities" (1977a).
, letter to Richard J. Johnson, Office of Air Quality Planning
and Standards, EPA, from C. W. Townley, concerning "Results of
Self-Service Exposure Samples," (May 1977b).
, "Sampling and Analysis for Benzene in Ambient Air Around
Gasoline Service Stations," prepared for U.S. Environmental
Protection Agency, Office of Toxic Substances (1978a).
, "Draft Task Report on Sampling and Analysis in Columbus,
Ohio," prepared for U.S. Environmental Protection Agency, Contract
No. 68-01-3858 (1978b).
Beck, D., Emission Standards and Engineering Division, Office of Air
Quality Planning and Standards, EPA, personal communication
(May 2, 1978).
Brothers, K. C., Emission Standards and Engineering Division, Petroleum
Section, Office of Air Quality Planning and Standards, EPA, letter
to Susan J. Mara, SRI International, concerning "Pure Benzene
Storage and Transfer" (January 19, 1978).
Burr.'R., Emission Standards and Engineering Division, Office of Air
Quality Planning and Standards, EPA, personal communications
(February 27 and March 24, 1978).
Chaney, L.W., "Carbon Monoxide Automobile Emissions Measured from the
Interior of a Traveling Automobile," Science, Vol. 199, pp. 1203 -
1204 (1978).
R-l
-------
Chien, J.C.W., "On the Possible Initiation of Photoxidation of Charge-
Transfer Excitation," J. Phys. Chem., Vol. 69, No. 12, pp. 4317-4325
(1969).
Chinn, Dr. H., Federation of American Societies for Experimental Biology
Bethesda, Maryland, personal communication (August 1977).
Dickerman, J. C., T. D. Raye, and J. D. Colley, "The Petroleum Industry,"
Radian Corp., Austin, Texas (1975).
Durham, J. F., Chief, Petroleum Section, Chemical and Petroleum Branch,
Office of Air Quality Planning and Standards, EPA, letter to
Susan J. Mara, SRI International, concerning "Benzene Storage and
Distribution Emissions" (January 11, 1978).
Faith, W. L., D. B. Keyes, and R. L. Clark, Industrial Chemicals, 3rd
Edition, John Wiley and Sons, Inc., New York (1966).
Ferman, M. A., R. S. Eisinger, and P- R. Monson, "Characterization of
Denver Air Quality," in Denver Air Pollution Study-19731—Proceedings
of a Symposium, Volume II, EPA 600/9-77-001, pp. 57-158 (February 1977)
Giacomelli, A., and M. Spinetti, "Salt Effects on the Activity of
Benzene in Aqueous Solutions," Gazz. Chim. Ital., Vol. 102, No. 11,
pp. 965-973 (1972).
Gifford, F. A., and S. R. Hanna, "Technical Note: Monitoring Urban Air
Pollution," in Atmospheric Environment, Pergammon Press, Vol. 7
pp. 131-136 (1973).
Hartle, R., National Institute for Occupational Safety and Health,
personal communication (1977).
Medley, W. H., "Potential Pollutants from Petrochemical Processes,"
Monsanto Research Corporation, prepared for U.S. Environmental
Protection Agency, Contract No. 68-02-0226 (1975).
Howard, P. H., and P- R. Durkin, "Sources of Contamination Ambient Levels,
and Fate of Benzene in the Environment," EPA 560/5-75-005,
NTIS PB-244 139 (1974).
Hustvedt, K. C., Office of Air Quality Planning and Standards, Emission
Standards and Engineering Division, EPA, personal communications
(1977a,b).
Johnson, R. J., Office of Air Quality Planning and Standards, Strategies
and Air Standards Division, U.S. Environmental Protection Agency,
personal communications (1977).
, Strategies and Air Standards Division, EPA, letter to Richard
Hartle, National Institute of Occupational Safety and Health,
concerning "Benzene Emissions During Refueling" (June 9, 1977).
Kaplan, L., L. A. Wendling, K. E. Wilzback, "Photooxidation of Aqueous
Benzene, I. Identification of the Product as 1,3-Cyclopentadiene-l-
Carboxaldehyde," J. Amer. Chem. Soc., Vol. 93, No. 15, pp. 3819-3820
(1971).
R-2
-------
Kilroy, K. J., B. F. Goodrich Chemical Company, letter to R. J. Johnson,
Office of Air Quality Planning and Standards, EPA, concerning "Use
of Monitoring Data Collected in Vicinity of Solvent Operations"
(February 17, 1978).
Kleeburg, C., Emission Standards and Engineering Division, Office of
Air Quality Planning and Standards, EPA, personal communication,
(February 17, 1978).
Kornriech, M. "Coal Conversion Processes: Potential Risk," Mitre Corp.,
prepared for Energy Research and Development Administration,
Contract No. 14-01-0001-2130 (1976).
Laity, J. L., I. G. Burstein, B. R. Appel, "Photochemical Smog and the
Atmospheric Reactions of Solvents," in Solvent Theory and Practice,
Advances in Chemistry Series, No. 124, Amer. Chem.. Soc. , Washington,
D.C., pp. 95-112 (1973).
Lao, R. C., R. S. Thomas, and T. L. Markham, "Computerized Gas Chromato-
graph—Mass Spectrometric Analysis of Polycyclic Aromatic Hydro-
carbons Environmental Samples," J. Chrom., Vol. 112, pp. 681-200
(1975).
Lee, C. C., W. K. Craig, and P. J. Smith, "Water Soluble Hydrocarbons
from Crude Oil," Bull. Environ. Contain. Toxicol., Vol. 12, No. 2,
pp. 212-217 (1974).
Little, A. D., Inc., "Technology Assessment and Economic Impact Study
of an OSHA Regulation for Benzene," in U.S. Department of Labor,
OSHA, Benzene, Vol. I, Washington, D.C. (1977a).
, "Self-Serve Market Shares in Four Metropolitan Areas,"
memo to R. J. Johnson, EPA, from E. Quakenbush and P. E. Mawn
(June 1977b).
Luria, M. and G. Stein, "The Photoproduct of Benzene in Oxygenated
Aqueous Solution," J. Chem. Soc.. p. 1650 (1970).
Mabey, W. R., "Identity and Chemical and Physical Properties of Compounds
in Coke-Oven Emissions," SRI International, Menlo Park, California,
prepared for U.S. Environmental Protection Agency, Contract
No. 68-01-4314 (September 1977).
MacKenzie, C. A., Unified Organic Chemistry, Harper and Row, New York
pp. 91-97 (1962).
Markwordt, D., Emission Standards and Engineering Division, Office of
Air Quality Planning and Standards, EPA, personal communication
(March 24, 1978).
Mascone, D., Emission Standards and Engineering Division, Office of Air
Quality Planning and Standards, EPA, personal communication
(May 2, 1978).
Masek, V., "Benzo(a)pyrene in the Workplace Atmosphere of Coal and Pitch
Coking Plants," J. of Occ. Med.. Vol. 13, p. 193-198 (1971).
R-3
-------
Matsuura, T. and K. Omura, "Photochemical Hydroxylation of Aromatic
Compounds," Synthesis, Vol. 3, pp. 73-184 (1974).
McAuliffe, C., "Solubility in Water of C^ to Cg Hydrocarbons," Nature,
Vol. 200, No. 4911, pp. 1092-1093 (1963).
McGannon, H. E. (Ed.), The Making, Shaping, and Treating of Steel, 9th
Edition, U.S. Steel, Pittsburgh, pp. 104-177 (1970).
Messer Associates, Inc., "A Study of the Health Effects of Bicycling
in an Urban Atmosphere," prepared for U. S. Department of Transpor-
tation, Contract No. DOT-TES-78-001, pp. 105-107 (1977).
Mitre Corporation, "Air Pollution Assessment of Benzene," Contract No.
EPA-68-02-1495 (1976).
National Cancer Institute, "On Occurrence, Metabolism, and Toxicity
Including Reported Carcinogenicity of Benzene, Summary Report,"
Washington, D.C. (1977).
National Institute for Occupational Safety and Health, "Update Criteria
and Recommendations for a Recommended Benzene Standard," NIOSH
74-137 (1976).
Neely, W. B., D. R. Branson, and G. E. Blau, "Partition Coefficient to
Measure Bioconcentration Potential of Organic Chemicals in Fish,"
Env. Sci. Tech., Vol. 8., No. 13, p. 1113 (1974).
Noyes, W. A., W. A. Mulac, and D. A. Harter, "Some Aspects of the
Photochemistry of Benzene," J. Chem. Phys., Vol. 44, No. 5,
pp. 2100-2106 (1966).
PEDCo Environmental, "Atmospheric Benzene Emissions," prepared for
U.S. Environmental Protection Agency, Research Triangle Park
EPA-450/3-77-029 (1977).
, "Ambient Benzene Emissions Data," prepared for U.S. Environ-
mental Protection Agency, Contract No. 68-02-2515 (1978).
Perry, F., California Air Resources Board, personal communications
(February 17 and 27, 1978).
Filar, S., and W. F. Graydon, "Benzene and Toluene Distribution in the
Toronto Atmosphere," in Environ. Sci. Techn., Vol. 7, No. 7,
pp. 628-631 (1973).
Polglase, W., Emission Standards and Engineering Division, Office of
Air Quality Planning and Standards, EPA, personal communication
(February 23, 1978).
Radian Corporation, "Control of Hydrocarbon Emissions from Petroleum
Liquids," EPA-600/275042, U.S. Environmental Protection Agency
(1975).
R-4
-------
Radian Corporation, "Final Report on Ambient Benzene Monitoring,"
prepared for Shell Oil Company, Houston, Texas (1978).
Ramsey, N. W., Fuel Additives, Noyes Data Corp., Park Ridge, New Jersey,
pp. vii-x (1974).
Research Triangle Institute, "Quantification of Benzene in 150 Ambient
Air Samples," prepared for U.S. Environmental protection Agency,
Office of Air Quality Planning and Standards (1977).
Robinson, E., R. A. Rasmussen, W. H. Westbert, and M. W. Holdren,
"Nonurban Nonmethane Low Molecular Weight Hydrocarbon Concentrations
Related to Air Mass Identification," J. Geophysical Research,
Vol. 78, No. 24 (1973).
Robinson, J. P., "Changes in American's Use of Time: 1965-1975—A Progress
Report," Communication Research Center, Cleveland State University,
Cleveland, Ohio (1977).
Runion, H. E., "Benzene in Gasoline," in Amer. Indus. Hygiene Assoc. Jr.,
Vol. 36, pp. 338-350 (1975).
, "Benzene in Gasoline II, in International Workshop on
Benzene, Paris, 9 p. (1976).
Schewe, G. J., Monitoring and Data Analysis Division, EPA, memos to
R. J. Johnson, Strategies and Air Standards Division, EPA, concerning
"Estimates of the Impact of Benzene from Automotive Sources," of
June 20, August 9, August 12 (1977).
Schumaker, J., Emission Standards and Engineering Division, Office of
Air Quality Planning and Standards, EPA, personal communication
(May 2, 1978).
Sheridan, E. T., "Supply and Demand for United States Coking Coals
and Metallurgical Coke," U.S. Department of the Interior, Washington,
D.C. (1976).
Smith, W. M., "Evaluation of Coke-Oven Emissions," J. Occup. Med., Vol. 13,
No. 2, pp. 69-74 (February 1971).
Stanton, A. (ed.), Kirk-Othmer Encyclopedia of Chemical Technology,
2nd Edition, John Wiley and Sons, Inc., New York (1964).
Stephens, E. R., "Hydrocarbons in Polluted Air," Summary Report, Coordi-
nating Research Council Project CAPA-5-68, Statewide Air Pollution
Research Center, University of California, Riverside, NTIS PB 230-
993 (1973).
Suta, B. E., "Human Population Exposures to Coke-Oven Atmospheric
Emissions," SRI International, Menlo Park, CA, report to U.S.
Environmental Protection Agency, Office of Research and Development
(1977).
R-5
-------
Sutton, C., and J. A. Calder, "Solubility of Higher Molecular Weight
n-Paraffins in Distilled Water and Sea Water," Environ. Sci. Techn.,
Vol. 8, No. 654 (1974).
Turner, D. B., and W. B. Petersen, "A Gaussian-Plume Algorithm for Point,
Area, and Line Sources," in Proc. of the 6th Int. Techn. Meeting on
Air Pollution Modeling and its Application, NATO/CCMS, No. 42,
pp. 185-228 (1975).
U.S. Department of Commerce, Bureau of the Census, "Population Estimates
and Projections," Series P-25, No. 618, Washington, D.C. (1976).
, "1972 County and City Data Book," Washington, D.C. (1973).
, "Statistical Abstract of the United States," Washington,
D.C. (1975).
, "1972 Census of Manufacturers," Washington, D.C. (1973).
Bureau of Economic Analysis, "Projections of Economic
Activity for Air Quality Control Regulations," NTIS PB-259-870
(1973).
U.S. Department of Transportation, Federal Highway Administration, "Annual
Miles of Automobile Travel," in Nationwide Personal Transportation
Study, Report No. 2, 32 p. (1972).
, "Home-to-Work Trips and Travel," in Nationwide Personal
Transportation Study, Report No. 8 (1973).
, "Highway Statistics," Washington, D.C. (1974).
"Motor Vehicle Registrations by Standard Metropolitan
Statistical Areas," Table MV-21 (1974).
U.S. Environmental Protection Agency, "Mixing Heights, Wind Speeds, and
Potential for Urban Area Pollution Throughout the Contiguous United
States," in Publ. No. AF-101, Office of Air Programs, Research
Triangle Park (1972).
, "Compilation of Air Pollution Emission Factors," 2nd Edition,
Publ. No. AP-42, Research Triangle Park (1976).
U.S. Public Health Service, "Control Techniques for Carbon Monoxide,
Nitrogen Oxide, and Hydrocarbon Emissions from Mobile Sources,"
Publ. No. AP-66, Washington, D.C. (1970).
Walker, D.C., H. W. Husa, and I. Ginsberg, "Demonstration of Reduced
Hydrocarbon Emissions from Gasoline Loading Terminals," EPA-650/2-
75-042, pp. 12-16 (June 1975).
Weber, R., Emission Standards and Engineering Division, Office of Air
Quality Planning and Standards, EPA, personal communication
(April 12, 1978).
White, L.D., "Coke-Oven Emissions, Occupational Health and the Environ-
ment," Master's Thesis, University of Cincinnati, Cincinnati, Ohio,
p. 17 (1972).
R-6
-------
Youngblood, P. L., Monitoring and Data Analysis Division, U.S. Environ-
mental Protection Agency, memo to R. J. Johnson, Strategies and Air
Standards Division, U.S. Environmental Protection Agency, concerning
"Rough Estimates of Ambient Impact of Various Benzene Sources from
Chemical Manufacturing Facilities" (May 5, 1977a).
, memo to R. J. Johnson, Strategies and Air Standards Division,
U.S. Environmental Protection Agency, concerning "Use of Dispersion
Calculations in Determining Population Exposures to Benzene from
Chemical Plants" (September 20, 1977b).
, memo to R. J. Johnson, Strategies and Air Standards Division,
U.S. Environmental Protection Agency, concerning "Population
Exposures to Benzene from Petroleum Refineries and Large Coking
Plants" (September 21, 1977c).
, memo to R. J. Johnson, Strategies and Air Standards Division,
U.S. Environmental Protection Agency, concerning "Ambient Impact
of Evaporative Benzene from Service Stations" (February 17, 1977d).
, personal communication (January 31, 1978).
R-7
-------
APPENDIX A
DIAGRAMS OF VARIOUS BENZENE-RELATED OPERATIONS
A-l
-------
QUENCH I (iG
EMISSIONS
STASDPIPE CAPS
COLLtCTOa MAIN
LAHRY CAR
WASTE GAS
STACK
Source: PEDCo, 1977
FIGURE A-1. SCHEMATIC DIAGRAM OF BY-PRODUCT COKE OVEN SHOWING POSSIBLE
ATMOSPHERIC EMISSION SOURCES FOR BENZENE
-------
AMBIENT
EMISSIONS
BENZENE
8
o
i/i
3
MIXED.
ACID
H20
.AIR CRUDE
NITROBENZENE
NITRATOR
oc
I
s
t/1
.TO ANILINE
PRODUCTION
WATER, DILUTE
SODIUM CARBONATE
WASHER
WASH-WATER
WASTE
SPEND ACID
TO RECOVERY
o
o
WASTE
NITROBENZENE
(REFINED)
Source: PEDCo. 1977
FIGURE A-2. FLOW CHART FOR NITROBENZENE MANUFACTURE FROM BENZENE
AND NITRIC ACID
-------
T
-p-
ETHYL ENE^
BENZEf<
RECYCl
AND FRE
Q
I.
«
U
Q
IE
E-*
SH
t
OFF-GAS
» <;rnilRRINfi
SYSTEM
1
| CONDENSER | BENZENE RECYCLE
1 TO REACTOR ETHYLBENZENE
c 3 WATER WASH „ CAUSTIC WASH _ uj * ^ .
2 v t *- AND SETTLER AND SETTLER Sf ^f ^5Z
i yR .3 o >-ujo
1 p^ f\
'ALUMINUM •"
COMPLEX 1 I
HEAVY
(POLYETHYL) BENZENES
AND TAR
RECYCLE (POLYETHYL )BENZENES
A1C1.
Source: PEDCo, 1977
FIGURE A-3. FLOW CHART FOR ETHYLBENZENE MANUFACTURE FROM BENZENE AND ETHYLENE
-------
Ui
BENZENE
1
1
HIXFR
FUSED
c ni T
COOLER
rnuucDTro
CgHg +41/2
ABSORBERS
i
1
* VAPOR
>m u» rntiinFNTFD .,_.».
CUULtK
0, — =— »»CHCO + 2H,0 + 2CO,
2 II > 2 2
CHCO
at yz
2 2j
sN
UACTF
MALE 1C
ANHYDRIDE
Source: PEDCo, 1977
FIGURE A-4. FLOW CHART FOR THE MANUFACTURE OF MALEIC ANHYDRIDE BY
CATALYTIC VAPOR-PHASE OXIDATION OF BENZENE
-------
I
ON
FRESH
ac
0
f-
t_
i
u
a
1
f
1
i
_
a
y?
a
a
a
p
5
a
COMB
FEED
INED
DRUM
~1
CONDENSER f
ra
'
1
,
&
CJ
* Ul
Ul
M
Ul
CO
^_
BEN;
ENE
CONDENSER
1
i
_.
z
I
c
<_
2
s
i
i
i
Q£
1 '
CUMENE
BOTTOMS
1
CONDENSER
CUMENE
FRESH PROPYLENE PROPANE
PROPANE
PHOSPHORIC
C-H
, + CH..CH • CH,.. AC!°.. fcC.H.CHfCH.^-
SOLID
Source: PEDCo, 1977
FIGURE A-5. PROCESS FOR THE MANUFACTURE OF CUMENE
-------
IMPURE CUMENE RECYCLE
CUHENE *-*•
HYDR06EN-
HYDROGENATOR
-EMULSIFIERS
AIR-
SULFURIC ACID-
OXIOIZER
ACIDIFIER
RECYCLE ACID
SEPARATOR
CHfiH5(CH3)2 + 02 - *
CgH5C(CH3)2 OOH — K
ACETONE
o
t 3E
£8
•—•
O
2 OOH
(CHj)2 CO
-PHENOL
I =>
i _J
. O
ACETOPHENONE
Source: PEDCo. 1977
FIGURE A-6. FLOW DIAGRAM FOR THE MANUFACTURE OF PHENOL BY THE
CUMENE PEROXIDATION PROCESS
-------
BENZENE OR —
CHLOROBENZENE
HYDROCHLORIC ACID,
MATER
BENZENE,
CHLORINE.
CO
CO
§
V)
CHLORINATOR
CD
LIT*
VENT
CO
GO
SODIUM
HYDROXIDE
I
CHLOROBENZENE
HYDROCHLORIC
ACID ^
NEUTRALIZING
TANK
SETTLING.
TANK
—^DICHLORO- AND
POLYCHLOROBENZENES
TO DISTILLATION
•CgH5C1 + HC1
Hn J. un
_ _^ au» i o ni* i
OICHLOROBENZENE
SLUDGE TO RECOVERY
Source: PEDCo. 1977
FIGURE A-7. FLOW DIAGRAM FOR THE MANUFACTURE OF CHLOROBENZENE AND
BY-PRODUCT DICHLOROBENZENES
-------
RECOVERABLE
SOLVENT
MIXTURE
SETTLING
TANK
SLUDGE
PREHEATER
£ =
U. UJ
ee. a
1/1 O
U
SODIUM CARBONATE
STEAM
CONDENSATE
T
BOTTOMS
•WATER OUT
•WATER IN
DEHYDRATING
TANK
PRODUCT
TO
SEWER
Source: PEDCo, 1977
FIGURE A-8. TYPICAL SOLVENT RE-REFJNING INSTALLATION
A-9
-------
• PRESSURE-VACUUM
VENT
GAUGE HATCH -
MANHOLE
Source: PEDCo, 1977
FIGURE A-9. FIXED-ROOF STORAGE TANK
•EATHER SHIELD LIQUID LEVEL DRAIN
HATCHES
VENT
ROOF SEAL
(NONMETALLIC
OR
METALLIC)
HINGED CENTER SUPPORT
MANHOLE
Source: PEDCo, 1977
FIGURE A-10. DOUBLE-DECK FLOATING-ROOF STORAGE TANK
(Nonmetallic Seal)
A-10
-------
ROOF CENTER SUPPORT
FLEXIBLE DIAPHRAGM ROOF
GAUGE HATCH
ROOF SEAL
(LIQUID IN TROUGH)
Source: PEDCo. 1977
FIGURE A-11. VARIABLE VAPOR SPACE STORAGE TANK
(Wet-Seal Lifter Type)
A-11
-------
S3
PIPELINE GASOLINE
TO STORAGE
STORAGE TANK
r
LOADING VAPORS
TO RECOVERY UNIT
TERMINAL
TRANSPORT
VENT GAS
1
VAPOR
RECOVERY
UNIT
RECOVERED <'
GASOLINE TO
LOADING RACK
GASOLINE
SOURCE: PEDCo, 1977
FIGURE A-12. VAPOR AND LIQUID FLOW IN A TYPICAL BULK TERMINAL (Floating-Roof Tank)
-------
APPENDIX B
EMISSION RATES AND POPULATION EXPOSURES
FROM
CHEMICAL MANUFACTURING FACILITIES
B-l
-------
Table B-1
EMISSION RATES FROM DIFFERENT CHEMICALS IN EACH PLANT USING BENZENE
ESTIMATED EMISSION RATE, JANUARY 1, 1976 (millions of kg per year)
STATE
ALABAMA
CALIFORNIA
DELAWARE
GEORGIA
ILLINOIS
LOCATION
TUSCALOOSA
CARSON
EL SEGUNDO
RWINDALE
RICHMOND
SANTA FE SPRINGS
DELAWARE CITY
CARTERSVILLE
BLUE ISLAND
CICERO
MORRIS
SAUGET
KANSAS EL DORADO
KENTUCKY ASHLAND
COMPANY
REICHHOLD CHEM.. INC.
WITCO CHEM.
STD. OIL CO. OF CALIF.
SPECIALTY ORGANICS, INC.
STD. OIL CO. OF CALIF.
FERRO CORP.
STO. CHLORINE CHEM CO.. INC.
CHEM. PRODUCTS CORP
CLARK OIL & REFINING
KOPPERS CO.. INC.
REICHHOLD CHEM.. INC.
MONSANTO
SKELLY OIL CO.
ASHLAND OIL. INC.
LOUISIANA BATON ROUGE FOSTER GRANT CO.
CARVILLE COS-MAR, INC.
CHALMETTE
TENNECO. INC.
GEISMAR RUBICON CHEM.. INC.
PLAQUEMINE GEORGIA PACIFIC CORP.
WELCOME GULF OIL CORP.
MARYLAND BALTIMORE CONTINENTAL OIL CO.
MASSACHUSETTS MALDEN SOLVENT CHEM. CO.. INC.
MICHIGAN MIDLAND DOW CHEMICAL
MISSISSIPPI PASCAGDULA FIRST MISSISSIPPI CORP.
MISSOURI ST. LOUIS MONSANTO
NEVADA HENDERSON MONTROSE CHEM. CORP. OF CAL.
NEW JERSEY BOUND BROOK AMERICAN CYANAMID
BOUND BROOK UNION CARBIDE
ELIZABETH REICHHOLO CHEM.. INC.
FORDS TENNECO. INC.
GIBBSTOWN E. I. flu PONT
KEARNY STD. CHLORINE CHEM. CO.
WESTVILLE TEXACO. INC.
NITRO •
BENZENE
0.031
0.23B
0.427
0.266
0.637
ANILINE
ETHYL-
BENZENE
0.272
0.202
0.070
0.166
0.166
STYHENE
0.668
0.409
0.367
0.273
MALEIC
ANHYDRIDE
0.483
2.610
4.641
1.363
1.16X1
CUMENE
0.011
0.012
0.016
0.040
0.001
0.029
PHENOL
O.O68
0.026
0.040
0.043
0.120
0.018
0.068
MONO-
CHLORO-
BENZENE
0.119
0.182
0.476
0.112
DICHLORO-
BENZENE
10- ind P.|
0.080
0.232
0.086
0.111
0.008
0.249
0.060
CYCLO
HEXANE
DETERGENT
ALKYLATE
(Uni.r
•nd Brcnch)
0.066
0.220
0.216
TOTAL
EMISSION
HATE
0.066
0.066
0.011
0.080
0.246
0.361
0.086
0X162
0.463
2.610
0.324
0.068
0.040
0.830
0.61)
0.070
0.238
0.120
0.612
0.216
OjOOe
1.17J
0.427
4.641
0.112
0.286
0.066
1.363
1.160
0.637
0.080
0.029
I
N5
-------
Table EM (Continued)
STATE
NEW YORK
OHIO
PENNSYLVANIA
PUERTO RICO
TEXAS
LOCATION
NIAGARA FALLS
NIAGARA FALLS
NIAGARA FALLS
SYRACUSE
HAVERHILL
BEAVER VALLEY
BRIDGEVILLE
CLAIRTON
FRANKFORD
NEVILLE ISLAND
PHILADELPHIA
GUAYAMA
PENUELAS
PENUELAS
BAYTOWN
BEAUMONT
BEAUMONT
BIG SPRING
BORGER
CHOCOLATE BAYOU
CORPUS CHRISTI
CORPUS CHRISTI
CORPUS CHRISTI
FHEEPOtlT
HOUSTON
HOUSTON
HOUSTON
HOUSTON
HOUSTON
ODESSA
OYSTER CREEK
PHILLIPS
PORT ARTHUR
PORT ARTHUR
PORT ARTHUR
SEADRIFT
SWEENEY
COMPANY
ICC INDUSTRIES. INC.
OCCIDENTAL PETROLEUM
SOLVENT CHEM. CO.
ALLIED CHEM. CORP.
UNITED STATES STEEL
ARCO/POLYMERS. INC.
KOPPERS CO.. INC.
UNITED STATES STEEL
ALLIED CHEMICAL CORP.
UNITED STATES STEEL
GULF OIL CORP.
PHILLIPS PETROLEUM
COMMONWEALTH OIL
UNION CARBIDE CORP
EXXON CORP
E. 1. du PONT
UNION OIL CO. OF CALIFORNIA
AMERICAN PETROFINA
PHILLIPS PETROLEUM
MONSANTO
COASTAL STATES GAS
SUN OIL CO.
UNION PACIFIC CORP
DOW CHEMICAL
ARCO/POLYMERS. INC.
JHE CHARTER CO.
JOE OIL. INC.
THE MERICHEM CO.
PETRO-TEX CHEM CORP.
EL PASO NATURAL GAS
DOW CHEMICAL
PHILLIPS PETROLEUM CO.
ARCOIPOLYMEHS. INC.
GULF OIL CORP
TEXACO
UNION CARBIDE CORP
PHILLIPS PETROLEUM CO.
NITRO-
BENZENE
0.987
ANILINE
ETHYL -
BENZENE
0.046
0.012
1 0.026
0.526
0.027
0.009
j
]
1
0.077
0.124
0.086
STYRENE
0.30O
0.060
0.054
1.009
0.067
0.102
0.204
MALEIC
ANHYDRIDE
1.450
1.740
2.224
CUMENE
0.051
0.072
0.073
0.016
0-028
0.051
PHENOL
0.090
0.260
0.090
0.227
0.1B2
MONO-
CHLORO-
BENZENE
0.024
0.038
OICHLORO-
BENZENE
IO- and P-l
0.077
0.077
CYCLO -
HEXANE
0.274
0.585
0.330
0.330
0.280
0.098
0.330
0.182
0.060
0.703
DETERGENT
ALKYLATE
(Llnaar and
Branch)
0.224
TOTAL
EMISSION
RATE
0.024
0.070
0.115
0.090
0.300
1.450
0.260
1.740
0.325
0.686
0.376
0.162
0.330
0.387
0.280
0.170
0.330
0.524
0.018
0.108
0.182
1.534
0.094
0.009
2.224
0.179
0.182
0.124
0.051
0.050
0.300
0.703
u>
-------
Table B-1 (Concluded)
STATE LOCATION
TEXAS TEXAS CITY
TEXAS CITY
TEXAS CITY
WEST VIRGINIA CHARLESTON
FOLLANSBEE
MOUNDSVILLE
NATRIUM
NEW MARTINSVILLE
WILLOW ISLAND
WASHINGTON ANACORTES
KALAMA
COMPANY
MARATHON OIL CO.
MONSANTO
STANDARD OIL (INDIANA)
UNION CARBIDE CORP
KOPPEHS CO.. INC.
ALLIED CHEM CORP.
PPG INDUSTRIES. INC.
MOBAY CHEM CORP.
AMERICAN CYANAMIDE
STIM5ON LUMBER CO.
KALAMA CHEMICAL
NITROGEN
0.176
0.427
0.188
ANILINE
ETHYL-
BENZENE
0.899
0.268
STYHENE
0.886
0.673
MALEIC
ANHYDRIDE
0.261
CUMNE
0.021
0.007
PHENOL
N.A.
0.026
MONO-
CHLORO-
BENZENE
0.143
DICHLOBO-
BENZENE
10- •» P-l
0.197
CYCLO-
HEXANE
DETERGENT
ALKYLATE
ILIrucr «nd
Branch)
0.149
TOTAL
EMISSION
HATE
0.031
1.784
0846
0.149
0.4J6
0.340
0.427
0.189
0.026
N.A. - NOT AVAILABLE
SOURCE. SRI ESTIMATES
-------
Table B-2
ESTIMATED POPULATION EXPOSED TO BENZENE
FROM CHEMICAL MANUFACTURING FACILITIES, BY FACILITY
CO
I
ui
State
Alabama
California
Delaware
Georgia
Illinois
Kansas
Kentucky
Louisiana
Maryland
Massachusetts
Michigan
Mississippi
Missouri
Nevada
Location
Tuscaloosa
Carson
El Segundo
Irwindale
Richmond
Santa Fe Springs
Delaware City
Cartersville
Blue Island
Cicero
Morris
Sauget
El Dorado
Ashland
Baton Rouge
Carville
Chalmette
Geismar
Plaquemine
Welcome
Baltimore
Maiden
Midland
Pascagoula
St. Louis
Henderson
Company
Reichhold Chemical
Witco Chemical
Standard Oil of Calif.
Specialty Organics
Standard Oil of Calif.
Ferro Corporation
Standard Chlorine Chem.
Chemical Products Corp.
Clark Oil & Refining
Koppers Company
Reichold Chemical
Monsanto
Skelly Oil Company
Ashland Oil
Foster Grant Company
Cos -Mar
Tenneco
Rubicon Chemical
Georgia Pacific Corp.
Gulf Oil Corporation
Continental Oil Co.
Solvent Chemical Co.
Dow Chemical
First Mississippi Corp.
Monsanto
Montrose Chemical Corp.
Total
Benzene
Emission
Rate
106Kg/yr
0.068
0.055
0.011
0.008
0.245
N.A.
0.351
0.086
0.052
0.483
2.610
0.324
0.058
0.040
0.830
0.611
0.07
0.238
0.120
0.512
0.215
0.008
1.172
0.427
4.641
Population Exposed to Benzene
0.1-1.0
20,000
30,000
400
200
80,000
20,000
7,000
4,000
20,000
90,000
70,000
4,000
20,000
100,000
10,000
5,000
4,000
8,000
10,000
500,000
3,000
70,000
20,000
20,000t
1.1-4.0
500
400
4,000
1,500
200
50
60,000
12,000
14,000
80
-
40,000
700
100
100
400
400
15,000
-
26,000
7,000
400,000
4.1-10.0
-
400
200
-
7,000
4,000
2,000
-
-
5,000
500
60
1,000
-
3,000
1,000
100,000
(ppb)*
> 10.0
-
-
-
-
700
1,000
-
-
1,000
90
-
-
900
30
50,000
of California
0.112
10,000
400
-------
Table B-2 (continued)
W
I
State
Mew Jersey
New York
Ohio
Pennsylvania
Texas
Location
Bound Brook
Bound Brook
Elizabeth
Fords
Cibbstoun
Kearny
Westville
Niagara Falls
Niagara Falls
Niagara Falls
Syracuse
Haverhill
Beaver Valley
Bridgeville
Clairton
Frankford
Neville Island
Philadelphia
Bay town
Beaumont
Beaumont
Big Spring
Borger
Chocolate Bayou
Corpus Christi
Corpus Christi
Corpus Christi
Freeport
Houston
Houston
Houston
Houston
Houston
Odessa
Oyster Creek
Phillips
Company
American Cyanamid
Union Carbide
Reichhold Chemical
Tenneco
E. I. du Pont
Standard Chlorine Chem.
Texaco
ICC Industries
Occidental Petroleum
Solvent Chemical Co.
Allied Ch-mical Corp.
United States Steel
Arco/Polyraers
Koppers Company
United States Steel
Allied Chemical Corp.
United States Steel
Gulf Oil Corporation
Exxon Corporation
E. I. du Pont
Union Oil Co. of Calif.
American Petrofina
Phillips Petroleum
Monsanto
Coastal States Gas
Sun Oil Company
Union Pacific Corp.
Dow Chemical
Arco /Polymers
The Charter Company
Joe Oil
The Merichem Company
Petro-Tex Chemical
El Paso Natural Gas
Dow Chemical
Phillips Petroleum Co.
Total
Benzene
Emission
Rate
106Kg/yr
0.2663
0.068
1.353
1.160
0.637
0.060
0.029
N.A.a
0.024
0.07
0.115
0.09
0.300
1.450
N.A.
0.250
1.740
0.325
0.330
0.9873
0.280
0.170
0.330
0.524
0.0163
0.108
0.182
1.534
0.0943
0.009
N.A.
N.A.
2.224
0.179
0.182
N.A.
*
Population Exposed to Benzene (ppb)
0.1-1.0
70,000
400,000t
400,000
200,000
30,000
3,000
80,000
100,000
3,000
20,000
100,000f
800,000
300,000*
1,000,000
40,000
100,000
30,000
10,000
6,000
100,000
20,000f
1,000,000+
70,000
1,000
1.1-4.0
3,000
90,000
14,000
7,000
600
-
2,000
5,000
100
700
8,000
30,000
270^000
40,000
5,000
28,000
3,000
1,500
300
6,000
10,000
110,000
6,000
60
4.1-10.0
300
20,000
2,000
900
-
-
-
-
-
70
2,000
3,000
30,000
5,000
600
4,000
40
200
30
400
2,000
20,000
100
-
> 10.0
8,000
600
200
-
-
-
-
-
-
600
-
12,000
-
-
1,000
-
-
-
-
600
5,000
-
-
-------
Table B-2 (concluded)
W
I
-vl
State
Texas (cont'd)
Washington
West Virginia
Puerto Rico
Location
Port Arthur
Port Arthur
Port Arthur
Seadrift
Sweeney
Texas City
Texas City
Texas City
Anacortes
Kalama
Charleston
Follansbee
Moundsville
Natrium
New Martinsville
Willow Island
Guayama
Penuelas
Penuelas
Company
Arco/Polymers
Gulf Oil Corp.
Texaco
Union Carbide Corp .
Phillips Petroleum Co.
Marathon Oil Co.
Monsanto
Standard Oil (Ind.)
Stimson Lumber Co.
Kalama Chemical
Union Carbide Corp.
Koppers Company
Allied Chemical
PPG Industries
Mobay Chemical
American Cyanamide
Phillips Petroleum
Commonwealth Oil
Union Carbide Corp.
Total
Benzene
Emission
Rate
106Kg/yr
0.124a
0.051
0.50
0.300
0.703
0.0213
1.784
0.846
N.A.
0.025
0.149
N.A.
0.436
0.340
0.427
0.189
0.585
0.3753
0.162
Population
0.1-1.0
50,000
3,000
9,000
20,000
1,000
70,000
20,000
6,000
10,000
3,000
100,000
100,000
Exposed
1.1-4.0
2,000
800
2,000
28,000
-
2,700
2,700
300
2,700
100
17,000
5,000
to Benzene
4.1-10.0
20
200
600
4,000
-
_
300
30
300
-
4,000
600
(ppb)*
> 10.0
-
-
100
1,000
-
_
-
-
-
-
_
80
TOTALS 6,000,000 1,300,000 200,000 80,000
a - When more than one chemical manufacturing facility is located in a city, it is assumed that they are in approximately
the same area and the emission levels are summed.
Annual average concentrations; to convert to 8-hour worst case, multiply by 25; to convert to ug/m3, multiply by 3.2;
a dash (-) signifies that no exposed population was estimated by our method for the average annual concentrations
listed. There may be some population exposed to those concentrations for shorter periods of time.
Some population may be exposed to annual average concentrations above 0.1 ppb beyond 20 km.
'Totals are rounded to one significant figure.
Source: SRI estimates.
-------
APPENDIX C
POPULATION EXPOSURES FROM COKE-OVEN OPERATIONS BY LOCATION
C-l
-------
Table Ol
ESTIMATED POPULATION EXPOSED TO BENZENE
FROM COKE OVENS, BY PLANT LOCATION
^
Locution
ALABAMA
Tarrant
Holt
Woodward
Gadsden
Thomas
Birmingham
Fairfield
CALIFORNIA
Fontana
COLORADO
n Pueblo
1
N> ILLINOIS
Granite City
Chicago
Chicago
South Chicago
INDIANA
Chesterton
Indianapolis
Terre Haute
East Chicago
East Chicago
Gary
Indiana Harbor
KENTUCKY
Ashland
MARYLAND
Sparrows Point
MICHIGAN
Detroit
Dearborn
Zug Island
(Detroit)
^
Plant Name
Tarrant Plant
Holt Plant
Woodward Plant
Gadsden Plant
Thomas Plant
Birmingham Plant
Fairfield Plant
Fontana Plant
Pueblo Plant5
Granite City Steel Div.
Chicago Plant
Wisconsin Steel Works
South Chicago Plant
Burns Harbor Plant
Prospect Street Plant
Terre Haute Plant
Plant No. 2
Plant No. 3
Gary Plant
Indiana Harbor Plant
Semet
Sparrows Point Plant
Semet
Steel Plant
Zug Island Plant
A
Company
Alabama By-Product Co.
Empire Coke Co.
Koppers Company, Inc.
Republic Steel Corp.
Republic Steel Corp.
U.S. Pipe and Foundry Co.
U.S. Steel Corp.
Kaiser Steel Corp.
CF&I Steel Corp.
National Steel Corp.
Inter lake, Inc.
International Harvester Co.,
Wisconsin Steel Div.
Republic Steel Corp.
Bethlehem Steel Corp.
Citizens Gas & Coke Utility
Indiana Gas and Chemical Corp.
Inland Steel Co.
Inland Steel Co.
U.S. Steel Corp.
Youngstown Sheet and Tube Co.
Solvay Div., Allied Chemical Corp.
Bethlehem Steel Corp.
Solvay Div., Allied Chemical Corp.
Ford Motor Co.
Great Lakes Steel Div. , National
Steel Corp.
Annual Coal
Capacity*
(tons)
1,200,000
150,000
800,000
820,000
185,000
1,175,000
2,500,000
2,336,000
1,332,000
1,132,000
949,000
991,000
590,000
2,630,000
675,000
204,000
3,102,000
1,642,000
3,700,000
2,100,000
1,600,000
4,820,000
900,000
1,800,000
2,850,000
Emission
Ratet
(g/sec)
1.0
0.13
0.69
0.71
0.16
1.0
2.2
2.0
1.1
0.97
0.82
0.85
0.51
2.3
0.58
0.18
2.7**
1.4
3.2
1.8
1.4
4.1
0.77
1.5
2.5
Population Exposed^ to Benzene (ppb)T
0.1-1.0 1.1-4.0 4.1-10.0 >10.0
2,000 |
-
-
480
-
6,250
-
1,420
-
-
830
-
-
-
60
-
21,330
36,860
580
550
19,380
_
2,200
33,930
-------
Table C-l (continued)
o
Location
MINNESOTA
St. Paul
Duluth
MISSOURI
St. Louis
NEW YORK
Buffalo
Lackawana
Buffalo
OHIO
Ironton
New Miami
Middletown
Painesville
Portsmouth
Toledo
Cleveland
Mas si Ion
Warren
Youngs town
Lorain
Campbell
PENNSYLVANIA
Swedeland
Bethlehem
Johnstown
Johnstown
Midland
Aliquippa
Pittsburgh
Erie
Philadelphia
Pittsburgh
Clairton
Fairless Hills
Monessen
A
Plant Name
St. Paul Plant
Duluth Plant
St. Louis Plant
Harriet Plant
Lackawana Plant §
Donner-Hanna Plant
Ironton Plant
Hamilton Plant
Middletown Plant8
Painesville Plant
Empire
Toledo Plant8
Cleveland Plant
Mas si Ion Plant
Warren Plant
Youngs town Plant
Lorain Cuyahoga Works
Campbell Plant
Alan Wood Plant
Bethleham Plant5
Rosedale Div.
Franklin Div.
Alloy & Stainless
Steel Div.
Aliquippa Plant§
Pittsburgh Plant
Erie Plant
Philadelphia Plant
Neville Island Plant
Clairton Plant5
Fairless Hills Plant
Wheeling
^
Company
Koppers Company, Inc.
U.S. Steel Corp.
Great Lakes Carbon Corp.,
Missouri Coke & Chemical Div.
Semet-Solvay Div., Allied
Chemical Corp.
Bethlehem Steel Corp.
Donner-Hanna Coke Corp.
Semet-Solvay Div., Allied
Chemical Corp.
Armco Steel Corp.
Armco Steel Corp.
Diamond Shamrock Corp.
Detroit Steel Div. , of
Cyclops Corp.
Inter lake Inc.
Republic Steel Corp.
Republic Steel Corp.
Republic Steel Corp.
Republic Steel Corp.
U.S. Steel Corp.
Youngs town Sheet and Tube Co.
Alan Wood Steel Co.
Bethlehem Steel Corp.
Bethlehem Steel Corp.
Bethlehem Steel Corp.
Crucible Inc., Div. Colt
Industries
Jones and Laughlin Steel Corp.
Jones and Laughlin Steel Corp.
Koppers Company, Inc.
Philadelphia Coke Division
Shenango Inc.
U.S. Steel Corp.
U.S. Steel Corp.
Pittsburgh Steel Corp.
Annual Coal
Capacity*
(tons)
250,000
850,000
450,000
400,000
4,250,000
1,387,000
1,230,000
934,000
748,000
215,000
600,000
438,000
2,220,000
250,000
650,000
1,500,000
2,700,000
2,300,000
803,000
2,210,000
550,000
1,680,000
657,000
2,250,633
2,587,404
290,000
715,400
1,022,000
9, 670,000ft
1,800,000
750,000
Emission
Ratet
(g/sec)
0.22
0.73
0.39
0.34
3.7
1.2
1.1
0.80
0.64
0.18
0.52
0.38
1.9
0.22
0.56
1.3
2.3
2.0
0.69
1.9
0.47**
1.4
0.57
1.9
2.2
0.25
0.61
0.88
8.3
1.5
0.65
Population Exposed'*' to Benzene (ppb)
0.1-1.0 1.1-4.0 4.1-10.0 >10.0
-
-
-
-
20,250
3,720
2,370
10
—
-
—
-
5,760
-
-
1,990
2,870
4,070
-
5,550
2,680
-
-
10,170
-
2,430
—
102,990
4,130
-
-------
Table C-l (concluded)
n
Location
TENNESSEE
Chattanooga
TEXAS
Houston
Lone Star
UTAH
Provo
WEST VIRGINIA
Weirton
Weirton
Fairmont
Follansbee
WISCONSIN
Milwaukee
Plant Name*
Chattanooga Plant
Houston Plant
E. B. Germany Plant
Geneva Works $
Weirton Mainland Plant
Weirton' s Brown's
Island Plant
Fairmont Plant
East Steubenville Plant
Milwaukee Solvay Coke Co.
Company*
Chattanooga Coke and Chemicals Co.
Armco Steel Corp.
Lone Star Steel Co.
U.S. Steel Corp.
Weirton Steel Div., National
Steel Corp.
Weirton Steej. Div., National
Steel Corp.
Sharon Steel Corp.
Wheeling-Pittsburgh Steel Corp.
A Division of Picklands Mather
and Co.
Annual Coal
Capacity*
(tons)
204,400
584,000
498,000
2,000,000
2,500,000
1,825,000
300,000
2,500,000
347,000
TOTAL1"*
Emission j.
Rate* Population Exposed to Benzene (ppb^
(g/sec) 0.1-1.0 1.1-4.0 4.1-10.0 >10.0
0.18
0.50
0.43
1.7 20
2.2**
1.6
0.26
2.2 3
0.30
300,000
Keystone Coal Industries Manual (1975) and Varga (1974), as cited in Suta (1977).
SRI estimates; population estimates are based on detailed census tract information available from Suta (1977).
^Annual average concentrations; to convert to 8-hour worst case, multiply by 25; to convert to yg/m3, multiply by 3.2. A dash (-) indicates that
no exposed population was estimated by our method for those annual average concentrations listed. There may be some population exposed to those
concentrations for shorter periods of time.
§
Coke oven operations producing benzene as a by-product (PEDCo, 1977).
t*
Coke oven operations located in approximately the same place. Their emission rates are summed.
tt
tt.
Total is rounded to one significant figure.
Based on a 1973 emission inventory.
-------
APPENDIX D
POPULATION EXPOSURES
FROM
PETROLEUM REFINERIES BY LOCATION
D-l
-------
Table D-l
ESTIMATED POPULATION EXPOSED TO BENZENE FROM PETROLEUM REFINERIES BY PLANT LOCATION
Location1
ALABAMA
Holt
Warrior Asphalt Co. of
Alabama, Inc.
Theodore
Marion Oil Co.
Tuscaloosa
Hunt Oil Co.
o Total
N)
ALASKA
Kenai
Chevron USA Inc.
Tesoro Petro Corp.
North Slope
At-Rich Co.
Total
ARIZONA
Fredonia
Arizona Fuels Corp.
Total
Total
Capacity
(H^m3)1
.17
1.04
1.65
2.86
1.28
2.21
.75
4.24
.23
.23
Total
Emission
(106R)2
2.1
13.5
21.5
37.1
16.6
28.7
9.8
55.1
3.0
3.0
Emission
Rate Population Exposed2 to Benzene (ppb)*
(R/sec)2 0.1-1.0 1.1-4.0 4.1-10.0 >10.0
,
.07 - -
.43 - -
.68 - -
.53a
.91 -
.31
.10
ARKANSAS
El Dorado
Lion Oil Co.
2.69
35.0
1.11
1. Source: Oil and Gas Journal, May 28, 1977.
2. Source: SRI estimates.
-------
Table D-l (continued)
o
i
to
Location1
ARKANSAS (continued)
Norphlet
MacMillan Ring-Free
Oil Co., Inc.
Smackover
Cross Oil & Refining
Co. of Arkansas
Stephens
Crystal Oil Co.
Total
CALIFORNIA
Bakersfield
Chevron USA Inc.
Kern Co. Refinery Co.
Lion Oil Co. (TOSCO)
Mohawk Petroleum Corp., Inc.
Road Oil Sales
Sabre Refining Co.
Sunland Refining Co.
West Coast Oil Co.
Benicia
Exxon Co.
Carson ^
Atlantic-Richfield
Fletcher Oil and
Refining Co.
Total
Capacity
(IP6!!!3)1
.26
Total
Emission
(106g)2
3.3
.34
.22
3.51
1.51
.92
2.21
1.28
.09
.20
.81
.87
5.12
10.165
1.11
4.4
2.9
45.6
19.6
12.0
28.7
16.7
1.1
2.6
10.6
11.3
66.4
312.2
14.5
Emission
Rate
(g/sec)2
.11
.14
.09
.62a
.38
.91
.53
.04
.08
.34
.36
2.11
9,9ia
.46
Population Exposed2 to Benzene (ppb)
0.1-1.0 1.1-4.0 4.1-10.0 >10.0
400
70,000
-------
Table D-l (continued)
Location*
CALIFORNIA (continued)
El Segundo
Chevron USA Inc.
Hanf ord
Beacon Oil Co.
Hercules
Pacific Refining Co.
Long Beach
Edgington Oil Co. Inc.
Los Angeles
Union Oil Co. - Calif.
Martinez
Lion Oil Co. (TOSCO)
Shell Oil Co.
Newhall
Newhall Refining Co. Inc.
Oildale
Golden Bear Div. , Witco.
Chemical Corp.
San Joaquin Refining Co.
Oxnard
Oxnard Refinery
Paramount
Douglas Oil Co.
Richmond AA
Shell Oil Co.
San Francisco
Union Oil Co. - Calif.
Total
Capacity
(lOSi3)1
c
23.51s
.71
3.09
1.71
6.27
7.31
5.80
.63
.61
1.57
.15
2.70
21.20
6.44
Total
Emission
(106g)2
702.5
9.3
40.2
22.3
81.5
95.1
75.4
8.2
7.9
20.4
1.9
35.1
550.8
83.8
Emission A
Rate Population Exposed2 to Benzene (ppb)
(g/sec)2 0.1-1.0 1.1-4.0 4.1-10.0 >10.0
t
22.30 60,900 -
.29
1.28 ~ -
.71
2.59 - - - -
3.02a
2.40 500
.26
.25a
.65 -
.06
1.11 -
17.49 100,000 -
2.66 -
-------
Table D-l (continued)
Location
Emission Rate Population Exposed to Benzene (ppb)
i
Ul
CALIFORNIA (continued)
Santa Fe Springs
Gulf Oil Co.
Powerline Oil Co.
Santa Maria
Douglas Oil Co.
Signal Hill
MacMillan Ring-Free
Oil Co.
South Gate
Lunday-Thagard Oil Co.
Torrance
Mobile Oil Corp.
Ventura
USA Petrochem Corp.
Wilmington
Champlin Petroleum Co.
Shell Oil Co.
Texaco Inc.
Total
COLORADO
Commerce City
Asamera Oil (U.S.) Inc.
Denver
Continental Oil Co.
Fruita
Gary Western Co.
Total
(106m3)*
2.99
2.56
.55
.67
.49
7.17
.87
1.78
5.22
4.35
132.63
1.31
1.89
.53
3.73
(106g)2
38.9
33.3
7.2
8.7
6.4
93.2
11.3
23.1
67.9
56.6
2,576,7
17.0
24.5
6.94
48.44
(g/sec)z 0.1-1.0 1.1-4.0 4.1-10.0 >10.0
1.23a
1.06 - -
.23
.28
.20
2.96 400 -
.36
.73a
2.16
1.80 200
231,500
.54 - -
.78 -
.22
-------
Table D-l (continued)
Location1
DELAWARE
Delaware City
Getty Oil Co. Inc.
Total
Total Total Emission
Capacity Emission Rate
(106g)2 (g/sec)2
8.13
8.13
105.6
105.6
3.35
Population Exposed2 to Benzene (ppb)
0.1-1.0 1.1-4.0 4.1-10.0 >10.0
100
100
FLORIDA
St Marks
Seminole Asphalt
Refinery Co.
Total
.33
.33
4.3
4.3
.14
o GEORGIA
o> Douglasville
Young Refining Co.
Savannah
Amoco Oil Co.
Total
.28
8.71
8.99
3.6
.12
.36
HAWAII
Barbers Point
Chevron USA Inc.
Ewa Beach
Hawaii Independent
Refinery Inc.
Total
2.32
3.42
5.74
30.2
44.5
74.7
.96
1.41
-------
Table D-l (continued)
Location1
Total Total Emission
Capacity Emission Rate
?
Population Exposed to Benzene (ppb)
ILLINOIS
Blue Island
Clark Oil and
Refining Co.
Colmar
Yetter Oil Co.
Hartford
Clark Oil and Refining Co.
Joliet
Mobil Oil Corp.
Lawrenceville
Texaco Inc.
Lemont ^
Union Oil Co. of Calif.
Lockport
Texaco Inc.
Plymouth
Wireback Oil Co. Inc.
Robinson
Marathon Oil Co.
Wood River
Amoco Oil Co.A£
Shell Oil Co.
Total
CIO6™3)1
3.86
.06
3.03
10.45
4.88
8.76§
4.18
.10
11.32
5.51
16.43§
68.58
(106g)2
50.1
.8
39.4
135.8
63.4
294.8
54.3
1.4
147.1
71.7
535.8
1,463.18
(g/sec)2 0.1-1.0 1.1-4.0 4.1-10.0 >10.0
1.59 - -'
.02 -
1.25 - -
4.31 6,000 ^
2.01 - -
9.36 7,000 -
1.72 - -
.04
4.67 400 -
2.283
17.01 IQQ.OQQt -
113,400
-------
Table D-l (continued)
00
Location1
INDIANA
East Chicao
Energy Coop. Inc.
Fort Wayne
Gladieux Refinery Inc.
Indianapolis
Rock Island Refining Corp.
Laketon
Laketon Asphalt Refining Inc.
Mt. Vernon
Indiana Farm Bureau
Coop. Association Inc.
Princeton
Princeton Refinery Inc.
Whiting
Amoco Oil Co.
Total
KANSAS
Arkansas City
Apco Oil Co.
Augusta
Mobil Oil Corp.
Chanute
Mid Amer Refinery Co.
Coffeyville
CRA Inc.
Total
Capacity
(106m3)1
7.31
.71
2.53
.47
.12
.27
21.19
33.31
2.68
2.90
.18
2.81
Total
Emission
(106g)2
95.1
9.2
32.9
6.1
1.6
3.5
275.4
423.8
34.9
37.7
2.3
36.5
Emission ^
Rate Population Exposed2 to Benzene (ppb)
(g/sec)2 0.1-1.0 1.1-4.0 4.1-10.0 >10.0
3.02 300 -
.29
1.04 -
.19
.05
.11 -
8.74 4,000 -
4,300
1.11 -
1.20 -
.07
1.16 -
-------
Table D-l (continued)
Location1
KANSAS (continued)
El Dorado AA
Getty Oil Co.
Pester Refining Co.
Kansas City
Phillips Petroleum Co.
McPherson
Nat. Coop. Refinery Assoc.
Phillipsburg
CRA Inc.
Shallow Water
E-Z Serve
Wichita
Derby Refining Co.
Total
KENTUCKY
Betsy Layne
Ky Oil & Refining Co. Inc.
Catlettsburg ^
Ashland Petr. Co.
Louisville
Louisville Refining Co.
Somerset
Somerset Refinery Inc.
Total
Total
Capacity
4.57§
1.31
5.22
3.14
15.32
.55
1.45
40.13
.03
c
7.883
1.46
.17
9.54
Total
Emission
(106g)2
151.8
17.0
67.9
40.9
199.2
7.2
18.1
613.5
.04
361.0
19.0
2.3
382.34
Emission ^
Rate Population Exposed2 to Benzene (ppb)
(g/sec)2 0.1-1.0 1.1-4.0 4.1-10.0 >10.0
s
4.82a
.54 80 - -
2.16
1.30
6.32 200
.23 -
.60
280
.01 - -
11.46 6,000 -
.60 - -
.07 -
6,000
-------
Table D-l (continued)
Location1
o
i
LOUISIANA
Baton Rouge
Exxon Co.
Belle Chasse
Gulf Oil Co., Alliance
Refinery**
Chalmette
Tenneco Oil Co.**
Church Point
Canal Refining Co.
Convent
Texaco
Cotton Valley
Kerr-McGee Refining Corp.
Garyville
Marathon Oil Co.
Hosston
Bayou St. Oil Corp.
Jennings
Evangeline Refining Co. Inc.
Lake Charles
Cities Service Oil Co.
Continental Oil Co.
Lisbon
Claiborne Gasoline Co.
Meraux
Murphy Oil Co.
Total Total Emission A
Capacity Emission Rate Population Exposed2 to Benzene (ppb)
(106m3)1 (106g)2 (g/sec)2 0.1-1.0 1.1-4.0 4.1-10.0 >10.0
200,000f
6,000
2,000
20
200
29.60§
11.4§
4.93§
.26
8.13
.64
11.61
.23
.29
15.56§
4.82
.38
5.37
628.7
353.7
167.0
3.4
105.6
8.30
150.9
3.0
3.8
303.0
62.6
4.9
69.8
19.96
11.23
5.3
.11
3.35
.26
4.79
.10
.12
9.62a
1.99
.16
2.22
6,000
-------
Table D-l (continued)
Total
Total Emission
o
i
Location1
Capacity Emission Rate Population Exposed to Benzene (ppb)
LOUISIANA (continued)
Metairie
Good Hope Refineries Inc.
Norco
Shell Oil Co.
Port Allen
Placid Refining Co.
Princeton
Calumet Refining Co.
Shreveport
Atlas Production Co., Div.
of Pennzoil**
St. James
LaJet Inc.
Venice
Gulf Oil Co.
Total
MARYLAND
Baltimore
Amoco Oil Co.
Chevron USA Inc.
Total
MICHIGAN
Alma
Total Petroleum Inc.
(lOV)1
3.86
13.93
1.99
.14
2.61§
.83
1.67
118.25
.87
.78
1.65
(106g)2
50.2
181.1
25.8
1.8
207,3
10.8
21.7
2,363.4
11.3
10.2
21.5
(g/sec)2 0.1-1.0 1.1-4.0 4.1-10.0 >10.0
1.59 - -
5.75 400 -
.82
.06
6,58 30,000 -
.34
.69 -
244,620
.36a
.32
2.32
30.2
.96
-------
Table D-l (continued)
Total Total Emission
Capacity Emission Rate Population Exposed2 to Benzene (ppb)'
o
I—1
CO
Location1
MICHIGAN (continued)
Bay City
Dow Chemical USA
Carson City
Crystal Refining Co.
Detroit
Marathon Oil Co.
Kalamazoo
Lakeside Refining Co.
West Branch
Osceola Refinery Div. , Texas
American Petrochemicals Inc.
Total
(106m3)"1
1.21§
.36
3.77
.33
.72
8.71
(106g)2
135,5
4.7
49.0
4.2
9.3
212.9
(g/sec)2 0.1-1.0 1.1-4.0 4.1-10.0 >10.0
4.30 6,000 -
.15
1.56 - -
.13
.30
6,000
MINNESOTA
Rosemount
Koch Refining Co.
St. Paul Park
Northwest Refining Co.,
of Ashland Oil Co.
Wrenshall
Continental Oil Co.
Total
MISSISSIPPI
Lumberton
Southland Oil Co.
Pascagoula
Div.
7.39
3.83
1.36
12.58
96.1
49.8
17.7
163.6
3.05
1.58
.56
10
10
Chevron USA Inc.
**
.33 4.3
16.3 422.5
.14
13.41
30,000
-------
Table D-l (continued)
Total Total Emission .
Capacity Emission Rate Population Exposed2 to Benzene (ppb)'
o
i
Location1
MISSISSIPPI (continued)
Purvis
Amerada-Hess Corp.
Sandersville
Southland Oil Co.
Yazoo City
Southland Oil Co.
Total
MISSOURI
Sugar Creek
Amoco Oil Co.
Total
MONTANA
Billings
Continental Oil Co.
Exxon Co.
Cut Bank
Westco Refining Co.
Great Falls
Phillips Petroleum Co.
Kevin
Big West Oil Co.
Laurel
Cenex
Wolf Point
Tesoro Petroleum Corp.
Total
(ib6™3)'1
1.65
.60
.23
19.11
6.21
6.21
3.05
2.61
.27
.35
.30
2.35
.15
9.08
(106g)2
21.5
7.85
3.0
459.2
80.7
80.7
39.6
34.0
3.5
4.5
3.87
30.5
1.9
117.87
(g/sec)2 0.1-1.0 1.1-4.0 4.1-10.0 >10.0
.68
.25 - -
.10
30,000
2.56
-
1.26a
1.08 ~
.11 - -
.14
.12
.97 -
.06
-------
Table D-l (continued)
o
i
Location1
NEBRASKA
Scottsbluff
CRA Inc.
Total
NEW HAMPSHIRE
Newington
Atlantic Terminal Corp.
Total
NEW JERSEY
Bayonne
National Oil Recovery Corp.
Linden
Exxon Co .
Paulsboro
Mobil Oil Corp.
Perth Amboy
Chevron USA Inc.
Westville
Texaco, Inc.
Total
NEW MEXICO
Artesia
Navajo Refining Co.
Bloomfield
Plateau Inc.
Thriftway Co.
Total
Capacity
(lOV)1
.29
.29
.75
.75
.35
16.54
5.69
9.75
5.12§
37.45
1.74
.49
.44
Total
Emission
(106g)2
3.8
3.8
9.8
9.8
4.5
215.0
73.9
126.8
179.6
599.8
22.6
6.3
5.7
Emission ^
Rate Population Exposed2 to Benzene (ppb)
(g/sec)2 0.1-1.0 1.1-4.0 4.1-10.0 >10.0
.12 - -
.31
.14
6.83 40,000 -
2.35
4.02 7,000 -
5.7 4,000 -
51,000
.72
.20a
.18
Ciniza
Shell Oil Co.
1.04
13.6
.43
-------
Table D-l (continued)
a
i
i-1
Wn
Location1
NEW MEXICO (continued)
Farmington
Giant Refining Co. Inc.
Kirtland
Caribou Four Corners Inc.
Loving ton
Southern Union Refining Co.
Monument
Southern Union Refining Co.
Total
NEW YORK
Buffalo
Mobil Oil Corp.
N. Tonawanda
Ashland Petroleum
Total
NORTH DAKOTA
Dickson
Northland Oil & Refining Co.
Mandan
Amoco Oil Co.
Willis ton
Westland Oil Co.
Total
Total
Capacity
.51
.17
2,23
.30
6.92
2.50
3.71§
6.21
.29
2.84
.27
3.40
Total
Emission
(106g)2
6.6
2.2
29.0
3.9
96.82
32.4
187,7
220. 1
3.8
37.0
3.5
44.30
Emission
Rate Population Exposed2 to Benzene (ppb)*
(g/sec)2 0.1-1.0 1.1-4.0 4.1-10.0 >10.0
.21
.07 - -
.92 T.
.12
1.03 - -
5,96 20,000 -
20,000
.12
1.17 „ - - _
.11
-------
Table D-l (continued)
Total Total Emission
Capacity Emission Rate Population Exposed to Benzene (ppb)
o
i
Location
OHIO
Canton
Ashland Petroleum Co.
Cleves
Gulf Oil Co.
Findlay
Ashland Petroleum Co.
Lima
Standard Oil Co. of Ohio
Toledo
Gulf Oil Co.
Standard Oil Co. of Ohio
Sun Petroleum Prod. Co.
Total
OKLAHOMA
Ardmore
Vickers Petroleum Corp.
Arnett
Tonkawa Refining Co.
Gushing
Hudson Refining Co. Inc.
Cyril
Apco Oil Corp.
Duncan
Sun Petroleum Products Inc.
Enid
Champlin Petroleum Co.
(lO6!!!3)1
3.71
2.48
1.16
9.75
2.92
6.96
7.26§
34.24
3.56
.35
1.10
.81
2.81
3.12
(106g)2
48.3
32.2
15.1
126.8
38.0
90.5
222.1
573. Q
46.2
4.52
14.3
10.6
36.6
40.6
(g/sec)2 0.1-1.0 1.1-4.0 4.1-10.0 >10.0
1.53 - -
1.02 -
.48 -
4.02 4,000 -r
1.20a
2.87
7.05 300,000 -
304,000
1.47 -
.14
.46 - -
.33
1.16 -
1.29 -
-------
Table D-l (continued)
Location1
OKLAHOMA (continued)
Okmulgee
OKC Refining Co.
Ponca City
Continental Oil Co.
Stroud
Allied Materials Corp.
Tulsa ..
**
Sun Petroleum Products Inc.
West Tulsa
Texaco Inc .
Wynnewood
Kerr-McGee Corp.
Total
OREGON
Portland
Chevron USA Inc.
Total
Total
Capacity
(106m3) 1
1.45
7.31
.03
5.14§
2.90
2.90
31.48
.81
.81
Total
Emission
(106g)2
18.9
95.1
3.6
175.5
37.7
37.7
521.32
10.6
10.6
Emission
Rate Population Exposed2 to Benzene (ppb)
(g/sec)2 0.1-1.0 1.1-4.0 4.1-10.0 >10.0
.60 - -
3.02 100 -
.11
5.57 8,000 - - -
1.20 - -
1.20
8,100
.34
PENNSYLVANIA
Bradford
Kendall-Amalie Dlv.,
Witco Chemical Co.
Emlenton
Quaker State Oil Refining
Corp.
Farmers Valley
Quaker State Oil Refining
Corp.
.52
6.8
.19 2.5
.38 4.9
.22
.98
.16
-------
Table D-l (continued)
Location1
PENNSYLVANIA (continued)
Freedom
Valvoline Oil Co. Div.
of Ashland Oil Co.
Marcus Hook
BP Oil Corp. ^
Sun Petroleum Products Co.
Philadelphia AA
Atlantic-Richfield Co.
D Gulf Oil Co.**
,L Reno
00 Pennzoil Co. - Wolf's Head
Div.
Roseville
Pennzoil Co. - Wolf's Head
Div.
Warren
United Refining Co.
Total
TENNESSEE
Memphis
Delta Refining Co.
Total
Total
Capacity
(lO6™3)1
.39
9.34§
9.58§
10.74
11.855
.12
.58
3.02
46.71
2.55
2.55
Total
Emission
(106g)2
5.1
138.0
304.0
279.2
397.8
1.6
7.5
39.2
1,186.6
2.3
2.3
Emission A
Rate Population Exposed2 to Benzene (ppb)
(g/sec)2 0.1-1.0 1.1-4.0 4.1-10.0. >10.0
.16
4t38a
9.65 40,000 -
a
8.86
12,63 2,000,000"*" -
.05 - -
.24
1.25 - -
2,040,000.
1.05 - -
TEXAS
Abilene
Pride Refining Co.
2.12
27.5
.87
-------
Table D-l (continued)
Location1
Total Total Emission
Capacity Emission Rate Population Exposed2 to Benzene (ppb)
o
>—•
vo
TEXAS (continued)
Amarillo
Texaco Inc.
Baytown
Exxon Co.
Beaumont
Mobile Oil Corp.
Union Oil Co. of Calif.
Big Spring ^
Cosden Oil & Chemical Co.
Borger
Phillips Petroleum Co.
Carrizo Springs
Tesoro Petroleum Corp.
Corpus Christi
Champlin Petroleum Corp.
Coastal States Petrochemical
Co.
Howe11 Corp.
Quintana Refining Co.
Saber Refining Co.
Southwestern Refining
Co. Inc.**
Sun Petroleum Products Co.
Deer Park Ajt
Shell Oil Co.
El Paso ^A
Chevron USA Inc.
Texaco Inc.
**
(lO6!!!3)1
1.16
22.6
18.86§
6.96
3.77§
5.80
1.51
7.26§
10. 7§
1.23§
1.36§
.54
6.96§
3.31§
17.06§
A. 00
.99
(106g)2
15.1
294.3
550.0
181.1
285.1
75.5
19.7
211.7
449.5
142.4
145.8
7.0
220.8
178.6
648.9
105.6
12.8
(g/sec)2 0.1-1.0 1.1-4.0 4.1-10.0 >10.0
.48 -
9.34 40,000 _. ,. -
17. 46s-
5.75 100,000f „.. _ -
9.05 30,000 _ _ _
2.40
.63 -
6,72a
14.27
4.52
4.63
.22
7..01
5.67 200,000f 2,000
20.6 2Q,OOOt -
3.35
.41 1,000 - - -
-------
Table D-l (continued)
Total
Total
Emission
o
i
Location1
TEXAS (continued)
Euless
Texas Asphalt Refining Co.
Ft. Worth
Winston Refining Co.
Hearne
Mid-Tex Refinery
Houston ^fi
Atlantic Richfield Co.
Charter International Oil Co.
Crown Central Petroleum Co.**
Eddy Refining Co.
LaBlanca
Crystal Oil Co.
Longview
Crystal Oil Co.
Mt. Pleasant
American Petrofina Inc.
Nixon
Pioneer Refining
Odessa
Shell Oil Co.
Port Arthur
**
American Petrofina Inc.
Gulf Oil Co.**
Texaco Inc.
**
Capacity Emission Rate Population Exposed2 to Benzene (ppb)
(106m3)l (106g)2 (g/sec)2 0.1-1.0 1.1-4.0 4.1-10.0 >10.0
.35
1.16
.17
17.76s
3.77§
5.80
.18
.28
.50
1.51
.15
1.86§
6.38§
18.11s
23.56
4.5
15.1
2.3
682.2
68.4
150.9
2.3
3.6
6.5
19.6
1.9
51.3
242,9
630.6
306.3
.14
.48
.07
21,66a
2.17
4.79
.73
.11
.20
.63
.06
1.63
7.71a
20. .02
9.72
l.OOO.OOO1
80,000f 600
-------
Table D-l (continued)
0
I
Location1
TEXAS (continued)
Quitman
Gulf St. Oil & Refining Co.
San Antonio
Flint Chemical Co.
Howell Corp.
Silsbee
South Hampton Co.
Sunray
Diamond Shamrock Corp.
Sweeny
Phillips Petroleum Co.
Texas City
Amoco Oil Co.
Marathon Oil Co.**
Texas City Refining Inc.
Three Rivers
Sigmor Refining Co.
Tucker
J&W Refining Inc.
Tyler
LaGloria Oil & Gas Co.
White Deer
Dorchester Gas Products Co.
Winnie
JL Jf
Independent Refining Co.
Young County
Thriftway Inc.
Total
Total
Capacity
.24
.07
.20
1.05
2.99
6.04
20. 20$
3.83§
4.32
.59
.58
1.70
.06
.77
.06
243.22
Total
Emission
(106g)2
3.1
0.9
2.6
13.7
38.9
102.4
561.0
102,7
56.2
7.7
7.5
22.1
.8
19.9
.8
6,435.6
Emission
Rate Population Exposed2 to Benzene (ppb)
(g/sec)2 0.1-1.0 1.1-4.0 4.1-10.0 >10.0
.10
.03a
.08 - -
.43
1.23 ,..
3,25 10 -
17,81a
3.26
1.78 60,000+ -
.24
.24 - -
.70
.02
.63
.02 - -
1,531,010 2,600
-------
Table D-l (continued)
o
i
NJ
N>
Location1
Total
Capacity
(lO6!!!3)1
UTAH
Asphalt Ridge
Arizona Fuels Corp.
North Salt Lake
Husky Oil Co.
Roosevelt
Plateau Inc.
Salt Lake City
Amoco Oil Co.
Chevron USA
Woods Cross
Caribou Four Corners Inc.
Morrison Petroleum Co.
Phillips Petroleum Co.
Western Refining Co. Inc.
Total
Total
Emission
(106g)2
Emission
Rate
(g/sec)2
.06
1.33
.46
2.26
2.61
.41
.15
1.33
.57
0.8
17.4
6.0
29.4
34.0
5.4
1.9
17.4
7.4
.02
.55
.19
.93a
1.08
.17a
.06
.55
.23
Population Exposed2 to Benzene (ppb)
0.1-1.0 1.1-4.0 4.1-10.0 >10.0
9.18
119.7
VIRGINIA
Yorktown
Amoco Oil Co.
Total
WASHINGTON
Anacortes
Shell Oil Co.
Texaco Inc.
Ferndale
Atlantic Richfield Co.
Mobil Oil Corp.
3.08
3.08
5.28
4.53
5.57
4.15
40.0
40.0
68.7
58.9
72.4
53.9
1.27
2.18a
1.87
2.30a
1.71
40
40
-------
Table D-l (continued)
OJ
Location
1
WASHINGTON (continued)
Seattle
Chevron USA
Tacoma
Sound Refining Co.
U.S. Oil and Refining Co.
Total
WEST VIRGINIA
Falling Rock
Pennzoil Co., Elk Refining
Div.
Newell
Quaker State Oil Refining
Corp.
St. Marys
Quaker State Oil Refining
Corp.
Total
Total Total Emission
Capacity Emission Rate
(106m3)1 (106g)2 (g/sec)2
.26
.26
1.24
21.29
.28
.56
.28
1.12
3.4
3.40
16.1
276.8
3.7
7.3
3.7
14.7
.11
.51
.12
.23
.12
Population Exposed2 to Benzene (ppb)
0.1-1.0 1.1-4.0 4.1-10.0 >10.0
80
WISCONSIN
Superior
Murphy Oil Corp.
Total
2.64
2.64
34.3
34.3
1.09
WYOMING
Casper
Amoco Oil Co.
Little American Refining Co.
Texaco Inc.
2.50
1.42
1.22
32.4
18.5
15.8
1.03a
.59
.50
-------
Table D-l (continued)
Location1
WYOMING (continued)
Cheyenne
Husky Oil Co.
Cody
Husky Oil Co.
Cowley
Sage Creek Refining Co.
LaBarge
Mountaineer Refining Co. Inc.
Southwestern Refining Co.
Lusk
C&H Refinery Inc.
Newcastle
Tesoro Petroleum Corp.
Osage
Glacier Park Co.
Sinclair
Sinclair Oil Corp.
Total
Total Total ,
O
Capacity Emission Rate Population Exposed to Benzene (ppb)
(106g)2 (g/sec)2 0.1-1.0 1.1-4.0 4.1-10.0 >10.0
1.37
.63
.07
.02
.03
.01
.61
.24
2.84
17.8
8.1
.86
.23
.4
.14
7.92
3.09
37.0
.57
.26
.03
.007a
.01
.005
.25
.10
1.17
10.96 142.24
Total Exposed Populationf
4,000,000
a - When more than one refinery is located in a city, it is assumed that they are in approximately the
same area and the emission levels are summed.
* 3
To convert to yg/m3, multiply concentrations by 3.2; to convert to 8-hour worst case, multiply by 25.
A dash (-) signifies that no exposed population was estimated by our method for those annual average
concentrations listed. There may be some population exposed to those concentrations for shorter periods
of time.
**
Refineries having catalytic reforming of benzene (Oil & Gas Journal, May 28, 1977). Their emission rate
is assumed to be twice that of refineries with no benzene production.
-------
Table D-l (concluded)
o
i
N>
Ol
Some population may be exposed to annual average concentrations >0.1 ppb beyond 20 km.
^Rounded to one significant figure.
Refineries handling pure benzene (SRI estimates). Emissions from storaee and loading of pure benzene
were estimated and added to the process emissions from the refinery. If benzene was used captively
at the refinery, no loading emissions were assumed. Controlled emissions were assumed for some refineries
where information was available (Brothers, personal communication, 1978).
-------
APPENDIX E
PARTIAL LIST OF HYDROCARBONS AND ADDITIVES
CONTAINED IN GASOLINE
E-l
-------
Table E-l
HYDROCARBONS CONTAINED IN THE VAPOR PHASE
OF GASOLINE
Methane
Ethane
Ethylene
Propylene
Propane
Isobutane
Isobutylene; Butene
N-Butane
T-2-Butene
C-2-Butene
3-Me-l-Butene
Isopentane
1-Pentene
2-Me-l-Butene
2-Me-l, 3-Butadiene
N-Pentane
T-2-Pentene
C-2-Pentene
2-Me-2-Butene
2,2-DimethyIbutane
Cyclopentene
3-Me-l-Pentene; 4-Me-l-Pentene
4-Me-C-2-Pentene
2,3-Dimethyl-l-Butene
2-Me-l, 4 Pentadiene
Cyclopentane
4-Me-T-2-Pentene
2,3 DimethyIbutane
2-Me-Pentane
2-Me-l-Pentene
3-Me-Pentane; 1-Hexene; 2-Ethyl-
C-3-Hexene
T-3-Hexene
3-Me-Cyclopentene
2-Me-2-Pentene
3-Me-T-2-Pentene
N-Hexane
4,4-Dimethyl-l-Pentene
T-2-Hexene
C-2-Hexene
3-Me-C-2 Pentene
4-4-Dimethyl-T-2-Pentene
Me-Cyclopentane; 3,3-Dimethyl-l-Pentene
2,3-Dimethy1-2-Butene;
2,3,3-Trimethyl-l-Butene
2,4 Dimethylpentane
2,4-Dimethyl-2-Pentene;
3-Ethyl-l-Pentene;
3-Me-1-Hexene
Benzene
2,2,3-TrimethyIbutane
2,4-Dimethyl-l-Pentene
4-Dimethyl-C-2-Pentene
1-Me-Cyclopentene
2-Me-C-3-Hexene
2,3-Dimethyl-l-Pentene; 2-Me-3-Hexene
5-Me-1-Hexene; T-3,5-Dimethyl-Cyclopentene;
C-3,5-Dimethycyclopentene
3,3-DimethyIpentane
Cyclohexane
4-Me-C-2-Hexene
2-Me-Hexane; 5-Me-C-2-Hexene
1,1-Dimethylcyclopentane
Cyclohexene
2,3-DimethyIpentane
3,4-Dimethyl-C-2-Pentene
3-M3-Hexane
l-C-3-Dimethylcyclopentane;
2-Me-l-Hexene
l-T-3-Dimethylcyclopentane
1-Heptene; 2-Ethyl-l-Pentene
-1-Butene
3-Ethyle Pentane; 3-Me-T-2-Hexene
l-T-2-Dimethylcyclopentane
2,2,4-TrimethyIpentane
T-3-Heptene
3-EthyIpentane
C-3-Heptene; 1,4-Dimethylcyclopentene
3-Me-C-3-Hexene;
3-Me-T-3-Hexene
3-Ethyl-2-Pentene
T-2-Heptene
E-2
-------
Table E-l (concluded)
N-Heptane; 3-Me-C-2-Hexene
2,3-Dimethyl-2-Pentene
2,3-Dimethylcyclopentene
3—Ethylcyclopentene
C-2-Heptene
l-C-2-Dimethylcyclopentane
2,2-Dimethyl Hexane
Me-Cyclohexane; 1,1,3-Trimethylcyclopentane
2,5-Dimethylhexane
Ethylcyclopentane
2,4-Dimethylhexane
2,2,3-Trimethylpentane
l-T-C-4-Trimethycyclopentane
3,3-Dimethylhexane
Toluene
1-T-2-C-3-Trimethylcyclopentane
Ethylbenzene
P-Xylene
M-Xylene
0-Xylene
C-8 Saturates and Olefins
Isopropylbenzene
N-Propylbenzene
l-Me-3-Ethylbenzene
l-Me-4-Ethylbenzene
l-Me-2-Ethylbenzene
1,3,5-Trimethylbenzene
1,2,4-Trimethylbenzene
1,2,3-Trimethylbenzene
C-9+ Saturates and Olefins
l-Me-2-Isopropylbenzene
1,3-Dimethy1-2-Ethylbenzene; 1,3-Diemthy1-4-Ethylbenzene
C-10 Saturates and Olefins
C-10 + Aromatics
Source: Walker et al, 1975, Table IVB.
E-3
-------
Table E-2
COMMON ADDITIVES CONTAINED IN GASOLINE
DETERGENTS AND ANTI-ICING ADDITIVES
Multipurpose Carburetor Detergents
Polyoxypropylene Ester and Tallow Trimethylenediamine Naphthenate
Alkyl Aryl Phosphate Esters and N-Oleyl-l,3-Propylenediamine Salts
Amine-Phosphoric Acid Salt and Olefin Hydrocarbon
Phosphate Ester Amine Salts
Nitrogen and Phosphorus-Containing Compositions
Dimer Acid and N-(10-Phenylstearyl)-l,3-Propylenediamine
Succinimides and Long Chain Primary Amines
Derivatives of Polymeric Succinimides
Polyamine-Alkaryl Carboxylic Acid Reaction Products
Nitroketonized Amides
Paraffinic Oil Oxidate-Fatty Amide Salts
High Molecular Weight Mannich Bases
Organylimidazolinyl Carbamates
2,6-Disubstituted-9-Oxabicyclononanes
Carburetor Detergents
Carbamates
Monosubstituted Ureas
Long Chain Alkylamino Alkyl-Substituted Ureas
Polyamine, Phenol, Formaldehyde and Acid Reaction Product
Diamine-Ethoxylated Polyphosphoric Acid Reaction Products
Guanidine Petroleum S.ulfonate
Polybutene-Substituted Nitrilotrisethylamine
Substituted Succinamic Acids and Polyolefins
Polyisobutylene-Substituted Polyamines
Di(Hydrocarbon)Substituted Alkylene Polyamines
Trimer Acid Polyesters
Nonionic Surfactant as Demulsificr
Anti-Icing Additives
Hexylene Glycol
Polyhydroxy Alcohol and Fatty Amides
Polyhydroxy Alcohols and N-(Phenylstearyl)-l,3-Propylenediamine
Fatty Imidazolines, Amides and Organic Silicon Compounds
Benzole Acid with Common Deicers
Polyethoxylated Propylenediamines
E-4
-------
Table E-2 (continued)
PBSA-Based Detergents
Mixtures with Amine Salts of Sulfonic Acids and Polyethers
Mixtures with Glycols and Glycol Ethers
Adipic Aeid-N-(Hydroxyethyl)-Ethylene Diamine Reaction Products
Polyhydric Alcohol Reaction Products
Polyamine-Polyhydric Alcohols Reaction Products
Sulfinyl-Containing Hydroxyl Compound Reaction Products
Other Ashless Detergents
Sulfonyl-Substituted Terpolymers
Alkane Nitroamines
Alkane Hydroxyamiries
Oil-Soluble Azo Compounds
Cyclohexanone Phenylhydrazone Oxidative Treatment
Ash-Forming Detergents
Overbased Sulfonate and Polyamine-Carboxylic Acids
Polybutene Succinic Anhydride Treated Overbased Complexes
Succinic Ester-Metal Salts
Basic Magnesium Salts of Oil-Soluble Acids
Coordinated Complexes of Nitrogenous Compounds
Glycerol Ester
FLOW IMPROVERS AND POUR DEPRESSANTS
Ethylene-Based Polymers
Vinyl Ester Copolymers
Vinyl Acetate Copolymer and Fatty Acrylate Homopolymer
Vinyl Ester Copolymer and Ethylene-Alpha-Olefin Copolymer
Alkyl Fumarate Copolymers
Alpha-Monoolefin Copolymers
Ethylene-Propylene-1,4-Hexadiene Terpolymers
Oxidized Ethylene-Propylene-Dicyclopentadiene Terpolymers
Aryl-Substituted Polyolefins
Other Polymeric Additives
Hydrogenated Styrene-Butadiene Terminated with Polar Groups
Polybutadienes
Esters of Styrene-Malei.c Anhydride Copolymers
Alkyl Itaconate-Maleic Anhydride Copolymers
Vinyl Acetate-Alkyl Fumarate Copolymers
Mixed Thiolacrylic Esters-Aminomethacrylate Terpolymers
n-Paraffin-Free Saturated Hydrocarbons
E-5
-------
Table E-2 (continued)
Other Additives
Polysaccharides
N,N-DialkyIricinoleamides
Fatty Amides and Salts
Carboxylic Acid Esters of Polyhydric Alcohols
Fatty Esters of Alpha-Methylglucoside
Olefin-Alkyl Halide Reaction Products
OXIDATION, CORROSION AND WEAR INHIBITORS
Oxidation Inhibitors
Aminoguanidine Derivatives
Isoindoline Compounds
1-Azabicycloalkanes
Three Component Amine and Amide Mixtures
Tetracyanoethylene
N-Substituted Alkoxyalkylamines
Epichlorohydrin-Alkylamine Reaction Products
Esters of (3,5-Dihydrocarbyl-4-Hydroxybenzyl)Thiodicarboxylic Acids
Alkylhydroxyphenyl Thiolacyl Alkanoic Acid Esters
Diphenyl Bis(3,5-Di-tert-Butyl-4-Hydroxyphenoxy)Silane
Alkoxy-2,6-Di-tert-Butyl-p-Cresol and Dibutyltin Sulfide
Alkylamine Salts of Phosphoric Acid Esters
Bix(Hindered Phenol) Alkylene Diphosphonates
Esters of PhosphorodithioateJ
Phosphorodithioate Ester-Aldehyde Reaction Products
Dihydrocarbylhydroxyphenyl Phosphonothionates and Phosphates
Metal Alkyl Ester Tetrapropenylsuccinates
Thermal Stabilizers
Ethylene-Propylene-Diene Terpolymer—Maleic Acid Reaction Product
Mannich Base, Dimer Acid and Metal Deactivator
Substituted Carbamates and Aldehyde-Amine Condensation Products
Cyclic Borates
Corrosion and Rust Inhibitors
Bis(l,3-Alkylamino)-2-Propanol and Phosphorylated Derivatives
Nitro-Nitrito Alkanes, Alkylene Polyamines and .Sulfur Reaction Products
Fatty Amides and Amines
Tertiary Amine Oxide Concentrates
2-Hydroxy-5-Cetylbenzene-l,3-Dicarboxylic Acid
Mixture of Carboxylic Acids and Pheno1-Aldehyde Resins
Alkyl Sulfoxides
Mercapto-Substituted Thiadiazoles
E-6
-------
Table E-2 (continued)
Antiwear Agents
Diethers of Diethylene Glycol
Bis(Hydroxyethyl)Alkane Phosphonates
Lecithin and Substituted Imidazolines
Trialkyl Phosphite-Alkenyl Succinic Anhydride Reaction Products
Phosphorus Pentasulfide-Glycol Reaction Products
Tetraoctyl Dimethylaminomethylene Diphosphonate
Substituted Succinic Anhydride and Metal Sequestering Agent
Tall Oil Fatty Acid
SMOKE AND EMISSION CONTROL AND COMBUSTION AIDS
Smoke Control
Metal 2-Ethylhexanoates
Barium Salts of Dialkyl Orthophosphoric Acids
Barium Carbonate and Dimethyl Ether of Ethylene Glycol
Barium Sulfonatocarbonate
Barium Sulfide Treated with Acetic Acid
Barium-Containing Dispersion
Colloidal Dispersions
Barium Alkaryl Sulfonates and Glycol-Ether Solvents
Overbased Calcium Sulfonates and Nitropropane
Cyclopentadienyl Manganese Tricarbonyl and Sulfinyl Amines
Emission Control and Combustion Aids
4,4'-Benzylidenebis(2,6-Di-tert-Butylphenol)
Polyalkoxylated Alkylphenol
Additive Blend
Boron Trifluoride Etherate-Amine Reaction Products
Vanadium Salts of Phosphorus Compounds to Improve Rumble
Tertiary Amines to Aid Air-Fuel Distribution
Hydrocarbon Wax to Aid Air-Fuel Distribution
Potassium Hexafluorozirconate and Organic Diamines
Clay, Phosphate and Boron Oxides for Boiler Fuels
Other Processes
Molybdenum Naphthenate and Organomanganese Compounds
Activated Manganese Oxide
Surface Ignition Suppressors
Aerosol Starting Agents for Diesels
E-7
-------
Table E-2 (continued)
ANTIKNOCK COMPOUNDS
Tetraethyllead
Triethylaluminum Reactions in Hexamethylphosphoramide
Tetrahydrofuran Solvent and Catalyst
Methyl Aluminum Sesquichloride and Sodium Fluoride Catalyst
Coproduction of Alkali Metal Aluminum Tetraalkyls
Ethylene dichloride
Ethylene dibromide
Tetramethyllead
Sodium Amide Catalyst
Diglyme, Methanol and Anthracene Catalyst Mixture
Magnesium-Aluminum Alloy Catalyst
Lithium-Aluminum and Mercury-Aluminum Catalyst
Stable Concentrate
Other Lead Compounds
Hexaorganodiplumbanes
Triorganolead Compounds
Spirocyclic Lead Compounds
Te traneopenty Head
Organolead-Silicon Compounds
Hexaaryldiliead Compounds
Trialkylplumbylmagnesium Compounds and Organic Halides
Metallic Lead Reactions Using Metallic Lithium
Nonmetallic Additives
1,3-Diimino-2-Hydroxypropanes
Mixed Nitrosoalkanes and Nitroso Aromatic Dimers
Aminofulvenes
ANTISTATS, BIOCIDES, DYES AND EMULSIFIED FUELS
Antistatic Agents
Phosphate Salts of Polyamides and Metal Naphthenates
Monoamine-Fluorinated Polystyrene Reaction Products
Cetyl Vinyl Ether-N-Vinylpyrrolidone Copolymers
Olefin-Maleic Anhydride and Alkyl Vinyl Ether-Maleic Anhydride Copolymers
Methyl Vinvyl Ether-Maleic Anhydride Copolymers
Biocides
Cyclic Imines
N-Alky1-1,3-Propanediamine
Naphthenyl Imidazolines
E-8
-------
Table E-2 (concluded)
Dyes
Blue Dye Mixtures
Azo Dye Compositions
Emulsified Fuels
Metallized Emulsion
Carbon Fuel
Emulsion Formulations
Source: Ramsey, 1974, p. vii-x.
E-9
-------
TECHNICAL REPORT DATA
read iHStritctions on the wiv/w hcjorc fo
1. REPORT NO.
EPA-450/3-78-031
4. TITLE AND SUBTITLE
Assessment of Human Exposures to Atmospheric Benzene
a. m CIIMI N i •:; A<:I:I :;:;ior*No.
l> m roil i IJA 11
June 1978
6~ PhHFOHMINti OHCJANI/AI ION COHI
7. AUTHOR(S)
Susan J. Mara
Shonh S. Lee
8. PERFORMING ORGANIZATION REPORT NO
Report No. 30R
9. PERFORMING ORGANIZATION NAME AND ADDRESS
10. PROGRAM ELEMENT NO.
SRI International
Center for Resource and Environmental Systems Studies
333 Ravenswood Ave.
Menlo Park, California 94025
11. CONTRACT/GRANT NO.
68-01-4314 &
68-02-2835
12. SPONSORING AGENCY NAME AND ADDRESS
13. TYPE OF REPORT AND PERIOD COVERED
E.P.A.
OAQPS & ORD
RTP,NC 27711
Final
14. SPONSORING AGENCY CODE
is. SUPPLEMENTARY NOTES This report supersedes all findings and conclusions in the first
draft of this report which was entitled, "Human Exposures to Atmospheric Benzene."
Project Officer: Richard Johnson
16. ABSTRACT
This report is one of three reports which were prepared by E.P.A. to determine
what regulatory action should be taken by E.P.A. to control sources of atmopsheric
emissions of benzene. This report projects atmospheric concentrations of benzene and
estimates the number of people exposed to these concentrations. These concentration
estimates are developed on an emission source category basis are then integrated,
using population flux estimates, thus, deriving total exposure estimates. The
original draft of this report has received extensive review by the interested public
and E.P.A.'s Science Advisory Board. All comments received on this first draft were
reviewed and considered in preparation of this report.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
benzene
air pollution -
populations
exposures '•'•.••
atmospheric concentrations
sources
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
CEMENT
Unlimited
19. SECURITY CLASS (ThisReport!
Unclassified
21. NO. OF PAGES
213
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
E-10
------- |