DRAFT
DO NOT QUOTE OR CITE
External Review
Draft No. 1
May 1978
PRELIMINARY ASSESSMENT OF THE
SOURCES, CONTROL AND POPULATION
EXPOSURE TO AIRBORNE POLYCYCLIC
ORGANIC MATTER (POM) AS INDICATED
BY BENZO(A)PYRENE (BaP)
DRAFT
DO NOT QUOTE OR UTS
NOTICE
This document is a preliminary draft.
It has not been formally released by
EPA and should not be construed to
represent Agency policy. It is being
circulated for comment on its technical
accuracy and policy implication.
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Waste Management
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
-------
EXTERNAL REVIEW
DRAFT NO. 1
May 1978
ORAi-1
DO NOT QUOTE OR CITE
PRELIMINARY ASSESSMENT OF THE
SOURCES, CONTROL AND POPULATION
EXPOSURE TO AIRBORNE POLYCYCLIC
ORGANIC MATTER (POM) AS INDICATED
BY JBENZO (A) PYRENE (BaP)
Submitted to:
Pollutant Strategies Branch
Office of Air Quality Planning and Standards
Environmental Protection Agency
Research Triangle Park, North Carolina 27711
DRAFT
00 NOT QUOTE OR CITE
Submitted by:
Energy and Environmental Analysis, Inc.
1111 North 19th Street, 6th Floor
Arlington, Virginia 22209
-------
DRAFT
DISCLAIMER NOTICE ^ QU0T£ OR CITE
This report was furnished to the Environmental Protection
Agency by Energy and Environmental Analysis, Inc., Arlington,
Virginia, in fulfillment of Contract No. 68-02-2836, Task No. 5
(POM). The contents of this report are reproduced herein as
received from the contractor. The opinions, findings, and con-
clusions expressed are those of the author and not necessarily
those of the Environmental Protection Agency.
ii
-------
draft
DO NOT QUOTE OR CITE
ACKNOWLEDGEMENTS
Preparation of this report by Energy and Environmental
Analysis, Inc. was carried out under the overall direction of
Mr. Robert D. Coleman. The principal author of the report
was Mr. Paul Siebert, with assistance from Ms. Carol Craig,
and Mrs. Elizabeth Coffey. We also acknowledge helpful sug-
gestions and direction from Mr. Justice Manning of the Envi-
ronmental Protection Agency.
The conclusions presented in the study are, of course,
solely the responsibility of Energy and Environmental Analysis,
Inc.
iii
-------
00 HOT QUOTE
draft
OR CITE
TABLE OF CONTENTS
Title Page
EXECUTIVE SUMMARY 1
SECTION I: INTRODUCTION 9
SECTION II: METHODOLOGY . , . 10a
A. Emission Factors i;.... 10a
B. Emission Estimates 12
C. Control Technology. .. . . * 13
D. Population Exposure ;.... *. 15
SECTION III: THE SOURCES OF POM ....... k ..... ; 19
A. General ¦. 19
B. Coal-Fired Power Plants i.... 5 0
C. Intermediate-Size Boilers 55
D. Residential Furnaces 59
E. Other Solid Fuels Burning Sources-, i 63
F. Future Energy Sources 65
G. Petroleum Catalytic Cracking......... 68
H. Coke Production, i .. i.. . 72
I. Asphalt Production 80
J. Iron and Steel Sintering ... i........... . 86
K. Carbon Black Production 88
L. Aluminum Reduction- 90
M. Municipal Incinerators 93
N. Commercial Incinerators 95
O. Bagasse Boilers 98
P. Open Burning 100
Q. Mobile Sources .........110
SECTION IV: ESTIMATES OF POPULATION EXPOSURE TO POM 124
A. Discussion of Alternative Estimation
Techniques 124
B. Analysis and Results of the Ambient
Concentration Technique 131
iv
-------
TABLE OF CONTENTS (Continued)
SECTION V: DISCUSSION OF THE STATE-OF-THE-ART
AND RESEARCH RECOMMENDATIONS 14 0
A. Discussion of Sampling and Analysis
Techniques 140
B. Current Studies and Research Recom-
mendations 143
APPENDIX A: COAL CONSUMPTION BY STEAM-ELECTRIC
PLANTS IN 1975 14 8
APPENDIX B: LOCATION, TYPE, AND CAPACITY OF
PETROLEUM CATALYTIC CRACKING
FACILTIES 151
APPENDIX C: LISTING OF ASPHALT ROOFING PLANTS
IN 1973 164
APPENDIX D: LOCATION AND CAPACITY OF UNITED
STATES SINTERING FACILITIES 173
APPENDIX E: LOCATION AND CAPACITY OF CARBON
BLACK PLANTS, 1977 175
APPENDIX F: MUNICIPAL INCINERATORS.... ...177
APPENDIX G: LIST OF NAMES, LOCATIONS, AND PHONE
NUMBERS OF PERSONAL CONTACTS 18 0
BIBLIOGRAPHY 185
v'
-------
DRAFT
EXECUTIVE SUHMflRi 00 N0T WW OR CITE
In order to assist the U.S. Environmental Protection Agency
in its Congressional mandate to "review all available relevant
information and determine whether or not emissions of...poly-
cyclic organic matter into the ambient air will cause, or con-
tribute to, air pollution which may reasonably be anticipated to
endanger public health," this study has investigated the emis-
sions and population exposures of polycyclic organic matter
(POM). Other review studies are assessing the environmental and
health effects and risk of POM. (The risk of a pollutant indi-
cates the likelihood that a hazard is associated with an expo-
sure to the pollutant.) Studies to improve the data base concern-
ing the emissions of POM, their control, and ambient concentra-
tions of POM are also being considered or conducted.
Incomplete combustion, which may occur naturally, inadver-
tently, or intentionally, is the major mechanism of POM forma-
tion and emission. POM's, however, are also present in fossil
fuels and other natural oils, such as some oils in asbestos. In
addition, various molds have been shown to contribute to the
natural synthesis of POM's. As POM's are generated primarily by
incomplete combustion, their emission is dependent on the type
and condition of the specific process and control equipment used
and the feed characteristics and operating conditions.
A literature survey was conducted to compile and update
emission factors and estimates and other information about the
various known and potential sources of POM. Only a few studies
were found that significantly added to the emission factor data
39/
reported by Hangebrauck, et al. in 1968. These additional
1
-------
DRAFT
DO NOT QUOTE OR Cit L
studies were generally concerned with only one particular source
or source type. Also, different sampling and analysis methods,
equipment, and procedures were used to measure different sub-
stances in the various studies. Therefore, the representative-
ness and comparability of the very limited available emission
factor data is dubious. Estimates of emission factors, however,
were developed.
It should be noted that a critical review of the available
data would allow little, if any, of the data to be used to gen-
erate even order of magnitude quantitative estimates of POM
emissions for many known sources. Little or no data is avail-
able on potential sources of POM, including future energy
sources, aluminum reduction, charbroiling of foods, industrial
internal combustion engines, lubricating oil incineration,
agricultural burning, aircraft, gasoline-powered lawn mowers,
motorboats, and misting and aerosol formation from lubricants.
Emission factors cannot be developed for these sources at this
time. The credibility of the reported estimates for known
sources is questionable because of the variability of the re-
ported data over as much as several orders of magnitude for a
source type (e.g., uncontrolled asphalt air blowing or open
burning of grass clippings, leaves, and branches), the limited
availability of data, the questionable representativeness and
comparability of the available data, and the uncertainty with
regard to emission from other sources. Therefore, the major
finding of this study is that the current data base on POM's is
insufficient for the development of quantitative estimates. By
noting the insufficiencies, the needs for future research can be
clarified.
For each of the various known sources, the most recent
available production or consumption data of the type on which
2
-------
the emission factors were based were used to calculate the
estimates of total annual national emissions of benzo(a)pyrene
(BaP) using emission factors developed. BaP was used instead of
total POM because it is a known animal carcinogen which has been
the focus of emissions, air quality, and health research con-
cerning POM. Minimum, maximum, and intermediate estimates of
recent emissions and the intermediate estimate of 1985 emissions
of BaP from major sources are given in Table E-l. Major sources
are considered to be those with national annual emissions of one
metric ton per year or greater. The minimum and maximum esti-
mates are taken directly from the literature or calculated from
the emission factors reported in the literature. The intermedi-
ate estimate was developed by EEA as a reasonable estimate from
the limited and questionable data. Generally, geometric means
were used to calculate an intermediate value of the emission
factors from a single source testing series. The geometric mean
or population weighting was usually then used to calculate the
intermediate emission factor estimates for a source type or
category. The intermediate estimate of BaP emissions for a
source category was calculated using this emission factor. The
estimates may assume air pollution control equipment or effi-
ciences.
Several observations can be made about the emission esti-
mates reported in Table E-l. Burning coal refuse banks and
forest fires, which are primarily natural or unintentional
sources, cannot be accurately quantified due to their uncertain
extent and the great variability in fuels and burning condi-
tions. The significant estimates of total emissions given in
the table were taken from the literature. The occurrence and
characteristics of burning coal refuse banks and forest fires
cannot be predicted. It is likely, however, that emissions will
3
-------
TABLE E-l
a/
ESTIMATES OP TOTAL BaP EMISSIONS FROM MAJOR SOURCES
Intermediate Estimate
Estimates of "Current" BaP Emissions (Mg/yr) of 1985 8aP .
Source Type Year Minimum Maximum Intermediate Emissions (Mg/yr)
Burning Coal Refuse Banks
1972
280
310
c/
Unknown
Unknown'
Coke Production
1975
0.050
300
110
21
Residential Fireplace
1975
52
110
73
77
Forest Pi res
1972
9.5
127
d/
Unknown
Unknown*
Coal-Fired Residential Furnaces
1973
0.096
740
26
26
Oil-Firod Commercial/Institutional Boilers
1973
19
24
Rubber Tire Wear
1977
0
11
e/
Unknown
Unknown*
Motorcycles
1975
5.6
5.6
Automobiles (gasoline)
1975
1.5
3. 3
2.7
0.21
Commercial Incinerators
1972
0.98
4.7
2.1
2.1
Oil-Fired Residential Furnaces
1973
0.89
2.0
0.98
1.2
Coal-Fired Power Plants
1974
0. 31
0.77
0.46
1.1
Oil-Fired Industrial Boilers
1973
0.23
0.68
0.37
1.8
Automobiles (diesel)
1975
0.0064
0.031
0.013
5.4
a/
b/
c/
d/
e/
Major sources are considered to be those for which the best estimate of current or projected emission is greater than
1 Mg/yr. Tor the sources and development of these estimates, including control assumptions, see Section III.
Emission estimates are reported in megagrams (Mg) per year. A megagran Is equivalent to one million grams or one metric ton.
Because coal refuse banks can ignite spontaneously and may flare up, smolder, or go out naturally, no reliable estimates
can be made of the number of burning banks. The Mining Enforcement and Safety Administration is currently conducting
such an inventory. However, the number and size of burning banks and observations on the surface that is burning, such
as will be available, will give an indication, but not a measure, of the quantity of burning coal.
The variation in the burning process and fuel types in different geographical, climatological, and seasonal conditions for
prescribed or wildfires is extremely great. The current understanding and quantitative knowledge of these variations does
not allow a best estimate to be made.
The quantity of emissions from rubber tire wear has not been adequately determined. Preliminary experimental work detected
no BaP, while a roughly estimated emission factor proportional to population was based on particulate emissions and some
analytical work. Using that factor, 19B5 BaP emissions would be projected to be 12 Mg/yr.
-------
remain within the same range or decrease with increased appli-
cation of improved control and prevention measures. Residential
sources, such as fireplaces and coal- and oil-fired furnaces,
are estimated to emit significant quantities of BaP. These
quantities are expected to remain constant or increase slightly
in' the future. BaP emissions from oil-fired commercial/insti-
tutional and industrial boilers and coal-fired power plants are
expected to increase as fuel consumption increases. The emis-
sions from coke production and gasoline consumption in automo-
biles are expected to decrease as improved control techniques
become more widespread. The emissions from the other major
sources (rubber tire wear, motorcycles, and commercial incin-
erators) are expected to remain nearly constant. Unless BaP
emissions are significantly reduced or controlled, emissions
from diesel automobiles are expected to increase drastically as
diesel fuel consumption in automobiles rises.
Population exposure to BaP was estimated from ambient air
quality data from a very limited number of monitoring sites
(several hundred). "Significant" ambient concentrations of BaP
were arbitrarily defined as 0.4 nanograms per cubic meter. A
3
concentration of 0.1 ng BaP/m was considered as a possible
"significant" concentration. Many cities, however, have had
higher ambient concentrations and a larger number of sources
potentially could have been estimated to produce "significant"
concentrations. The 0.4 ng BaP/m concentration was chosen in
consultation with EPA because few non-industrial cities had
higher ambient concentrations. The emissions from individual
point and area sources of POM, other than coke ovens, were very
conservatively estimated (using assumptions and techniques that
would generate a high estimate of concentration) to produce
ambient air concentrations that were, at most, marginally
5
-------
3
"significant" (^0.4 ng/m ). Therefore, it is unlikely that any
individual source, other than coke ovens, will generate "sig-
nificant" ambient concentrations of BaP. Since localized
consumption or production data were not available for most of
the major sources of POM, the emission factors developed in this
study could not be used to estimate emissions from individual
sources in a locality. Therefore, dispersion modelling of
emissions estimates in order to estimate local ambient concen-
trations and, thus, estimate population exposures could not be
118 /
accomplished. The results of a recent study ' were used to
calculate population exposures in areas with coke ovens. For
areas without coke ovens, ambient air monitoring results for BaP
were used in conjunction with U.S. Census data to estimate
population exposures. If available, actual or extrapolated
monitoring results for the specific area were used. National
averages of BaP monitoring results for different types of areas
(large urban, small urban, and rural), however, had to be used
for most areas. These estimates of population exposure are
given by state in Table E-2. The values are expressed in terms
of the number of people exposed times the exposure concentration
divided by 1,000.
The quality of these estimates of population exposure
should be noted. Because they are based on data from a rela-
tively small number of monitoring sites which have been operated
at various times, and because the nature of POM production
probably leads to significant spatial variations in ambient
concentrations, the values presented on population exposure are
very rough estimates. These estimates, however, were the only
ones feasible within the time available and show that the states
with the greater population exposures are probably those with
large populations or with coke ovens located near population
6
-------
TABLE
E-2
ESTIMATES
OF POPULATION EXPOSURE
TO BaP IN THE
UNITED STATES3^
Exposure
Exposure
Exposure
(1,000 People
(1,000 People
(1,000 People
State
x ng/m"^)
State
x ng/m^)
State
x ng/m^)
Alabama
2 ,500
Kentucky
3,000
North Dakota
450
Alaska
120
Louisiana
1,500
Ohio
13,000
Arizona
2,100
Maine
450
Oklahoma
1,300
Arkansas
890
Maryland
2,500
Oregon
2 ,100
California
17,000
Massachusetts
5,600
Pennsylvania
19 ,000
Colorado
3,000
Michigan
9,600
Rhode Island
740
Connecticut
2,300
Minnesota
1,600
South Carolina
1,600
Delaware
430
Mississippi
1,100
South Dakota
330
District of
Columbia 540
Missouri
2,600
Tennessee
2 ,400
Florida
4,600
Montana
390
Texas
9,200
Georgia
2,600
Nebraska
740
Utah
1,200
Hawaii
96
Nevada
600
Vermont
150
Idaho
390
New Hampshire
440
Virginia
2,700
Illinois
9,900
New Jersey
5,800
Washington
1,700
Indiana
6,600
New Mexico
530
West Virginia
1,000
Iowa
1,200
New York
20,000
Wisconsin
1,400
Kansas
1,300
North Carolina
2,800
Wyoming
140
U.S. TOTAL 170,000
Exposure estimates are reported as the product of the 1,000*s of people exposed times the estimated ambient BaP
concentration to which they are exposed. Population figures for inside and outside SMSA's by urban and rural
residence were taken from the 1970 Census. For coke oven cities, the populations and concentrations exposed to
coke oven BaP emissions and the urban area-specific background BaP levels were taken from SKI Project #5794 Final
Report (November 1977) by B. Suta for EPA Contract No. 68-01-4314.^"^"®/ If the reported exposed populations were
greater than the total population in the SMSA or the part of the SMSA within the state in which the coke oven
was located, the remaining exposed population was counted in the part(s) of the SMSA in other states, in adjacent
SMSA's in the same or adjacent states, or in the rural population of adjacent areas, depending on the locality and
the presumed prevailing wind direction. For non-coke oven cities where BaP sampling had been conducted, an actual
or extrapolated BaP concentation for "1975" was used; where no sampling had been conducted, a U.S. average ambient
BaP concentration of 1.1 ng/m3 (calculated from "1975" annual average concentrations in non-coke oven cities with
pooulations greater than 25,000) was used to calculate exposure for the remaining urban population within SMSA's.
For other non-coke oven areas, U.S. average ambient BaP concentrations of 0.86 ng/m^ (calculated from "1975" con-
centrations for non-SMSA areas of 10,000 to 50,000 population without coke ovens), and 0.15 ng/m3 (calculated from
"1975" concentrations for rural areas with less than 10,000 population) were used to calculate exposures for the
urban population outside SMSA's and the total rural population, (except those counted for coke oven exposures), re-
spectively. These U.S. average ambient BaP concentrations were calculated by EEA, largely from data presented in
coke oven study.118/
-------
DRAFT
DO NOT QUOTE OR CI ft
centers. A significant improvement in these population exposure
estimates will require improved emissions, production and lo-
calized consumption data or greatly increased ambient air moni-
toring data. Quite a few studies are in progress which will
improve the data base on POM's. This improvement, however, may
not be enough to allow the use of better estimation techniques.
3
-------
DRAFT
DO NOT QUOTE OR CITE
SECTION I
INTRODUCTION
The Clean Air Act Amendments of August 7, 1977 require that
within one year the Environmental Protection Agency (EPA) "shall
review all available relevant information and determine whether
or not emissions of...polycyclic organic matter into the ambient
air will cause, or contribute to, air pollution which may rea-
sonably be anticipated to endanger public health." This study,
which is part of the response to that request, has concentrated
on the emissions and population exposure of POM based on the
literature. Other studies are exploring the formation and
transformation of POM, its control and health effects.
In the most comprehensive study of POM's to date, the
National Academy of Science's (NAS) Particulate Polycyclic
17 /
Organic Matter, the term POM was first used. The POM classi-
fication encompasses all organic matter with two or more benzene
rings (cycles) and includes polycyclic (or polynuclear) aromatic
hydrocarbons (PAH, PNA, or PNAH), aza arenes, imino arenes,
carbonyl arenes, dicarbonyl arenes, hydroxy carbonyl arenes, oxa
and thia arenes, polychloro compounds, and pesticides.Only
some compounds of the polynuclear aromatic hydrocarbons and
their neutral nitrogen analogues, the aza arenes (or polycyclic
azaheterocyclic compounds), have been shown to be carcinogenic
to animals. Health effects research has focused on the poly-
nuclear aromatic hydrocarbons and particularly on benzo(a)pyrene
(BaP), one of the most carcinogenic of the PNAH that were iso-
lated by early researchers. The sources, and the health and
other effects of POM, were rather extensively reviewed in the
9
-------
1972 NAS study. ^ This study has been updated in two EPA
studies: (1) Preferred Standards Path Report for Polycyclic
Organic Matter, October 1974,^'^'and (2) Scientific and
Technical Assessment Report on Particulate Polycyclic Organic
Matter (PPOM) , March 1975. The environmental and health
effects are currently being updated in another study. The risk
associated with POM exposure is also being investigated. (The
risk of a pollutant assesses the probability that a hazard is
associated with exposure to the pollutant.)
The scope of this study has been limited to updates of the
estimates of emission factors, total emissions and population
exposures to POM. The emission factors and emission estimates
are based primarily on the 1968 Public Health Service report
3 9/
Sources of Polynuclear Hydrocarbons in the Atmosphere. / More
recent and extensive information was used when available; such
information was generally not available. Also, it was found
that area source "production" data, such as fuel consumption
were generally not available for areas smaller than states.
Production data for individual point sources were available for
many, but not all, source types. This type of data is necessary
to estimate localized emissions, and thus ambient concentra-
tions, from the emission factors.. Since this information was
not sufficiently available, population exposure had to be esti-
mated from ambient concentrations of BaP at a relatively small
number of sites. The remaining sections of this report discuss
the methodology used in the study; the quantitatively known
sources of POM and their emissions, control, location, and
capacity; the estimates of population exposure; POM sampling and
analysis techniques; current studies related to POM emissions;
and recommendations for future work that would improve the
quality of the estimates of emissions and population exposures
presented in this report.
10
-------
SECTION II
METHODOLOGY
A. Emission Factors
POM can be generated by natural synthesis (e.g., by various
molds), by the processing of fossil fuels or other materials
containing POM-bearing oils (e.g., petroleum or asbestos), or by
naturally, inadvertently, or intentionally occurring combustion
(e.g., forest wild-fires, burning coal refuse banks, or firing
coal in a power plant) which does not completely burn hydro-
carbons and oxygen to form carbon dioxide and water. Since most
POM's are generated by incomplete combustion, the emission
factors developed for this study were based upon some form of
emissions sampling. Most of the emission factors for stationary
sources were developed from data reported by Hangebrauck, von
Lehmden, and Meeker in their rather early (1968), but relatively
39/
comprehensive, study. ' A literature search was conducted by
EPA in order to update this information. However, the results of
very few additional studies involving sampling of stationary
sources for POM's have been published. The additional data that
were available were largely in EPA-sponsored reports studying a
particular POM source. Mobile source emission factors were
38/
chiefly derived from data in two studies: that of Gross ' for
92/
gasoline engines, and that of Spindt for diesel engines.
Therefore, the development of updated emission factors in this
study largely involved a compilation of the available data and
the estimation of emission factors in the desired units of
measure for the various sources.
11
-------
Emission factors were developed as minimum, maximum, and
"intermediate" estimates for each source (combination of size,
range, process, control equipment, etc.), and source type (ca-
tegory of sources). The minimum and maximum estimates for each
source were the end points of the range of emission factors
reported in, or developed from, the data reported in the various
studies. For some sources, these estimates differed by an order
of magnitude. The "intermediate" estimate emission factor for
each specific source (i.e., a size range of a process type) was
generally assumed to be the geometric mean of the individual
sampling test results. The geometric mean was felt to be most
appropriate for data with a large variation, as was often the
case, because the geometric mean is generally the best estimate
of what the result of any one occurrence would be. Estimates
for the same general process type or for larger size ranges
(e.g., all capacities of fluid catalytic cracking—FCC) and for
source types (e.g., petroleum catalytic cracking) were developed
from those for the specific sources (e.g., capacity ranges of
FCC) by population weighting or by taking the geometric means.
If several different processes were included in a category, and
information was available on the numbers of each of the indi-
vidual sources, a weighted average was calculated by weighting
the logarithms of the individual source estimates by the number,
dividing by the total number, and taking the anti-logarithm. If
such information was not available, the geometric mean of the
individual source estimates was used as the estimate for the
more general sources.
It should be noted that, for several reasons, these esti-
mates of emission factors are of limited value. Emission fac-
tors are usually developed from limited data from, at most, a
few combinations of these parameters. Therefore, emission
12
-------
factors in general are of questionable value for estimating the
emissions from any particular combination of process, control
equipment, feed, product, condition of equipment, and operating
conditions. In the case of POM, the situation is even more
extreme because, even for duplicate runs of the same source
testing series, the reported values may vary over several orders
of magnitude (e.g., uncontrolled asphalt air blowing or open
burning of grass clippings, leaves, and branches). In addition,
the following conditions may be significant. First, only a
limited number of tests with few, if. any, duplicates were run at
only a few facilities. Thus, these results may not be represen-
tative of the emissions from that particular source, let alone
those of the particular equipment type or more general source
type. Second, since POM's are generated by incomplete combus-
tion, their generation is largely dependent upon the operation
and maintenance of a particular piece of equipment. Third, as
delineated in Section V, there is not any standard method for
sampling and analysis of POM. Therefore, a variety of proce-
dures have been used in different studies. The results of these
procedures may vary by as much as two orders of magnitude for
simultaneous sampling of a stack using two different sampling
44/
trains (EPA Method 5 and a Tenax adsorbent train).
B. Emission Estimates
Estimates of annual BaP emissions in both the "current"
year and in 1985 were made for the various source types. BaP
was chosen as a surrogate of total POM because BaP emissions
data were available for more sources, and the POM data that were
available measured the quantities of different numbers and
species of POM's. Also, ambient air quality data generally use
BaP as a surrogate. The estimates of annual BaP emissions were
calculated by multiplying the estimated emission factors by a
13
-------
reported, estimated, or projected annual production or consump-
tion figure. The minimum, maximum and best estimate BaP emis-
sion factors, when available, were all used to generate a range
of values and a best estimate. In some cases, e.g., burning
coal refuse banks and forest fires, only a very rough estimate
of maximum annual emissions could be made.
Production or consumption data are generally available only
for rather general process or use categories. Emission esti-
mates were, therefore, made only for the more general source
types. Production or consumption figures were collected from
trade, industrial, professional, or government sources. "Cur-
rent" emissions were estimated by using measured or estimated
"production" figures for 1977, or the most recent year for which
data were readily available. The 1985 emissions were estimated
by using available projections of 1985 "production" or by pro-
jecting from a base year using a reported or estimated rate of
change. It should be noted that these estimates of 1985 pro-
duction or consumption, even though based upon current or ex-
pected trends, may be very poor estimates of reality, as may be
any predictions of future occurrences. Even though this is the
case for the 1985 estimates and, to a lesser degree, for the
"current" estimates, the estimates of BaP emissions presented in
this study are based on the most current data available.
Therefore, the relative rankings of BaP emitters presented
should give a state-of-the-art indication of the importance of
the various source types as contributors to total BaP emissions.
C. Control Technology
For the various known and potential sources of POM, the
control technology that would be expected or has been shown to
be effective in reducing POM emissions was studied. The effec-
tiveness, feasibility, and application of the available control
14
-------
techniques were examined for the various source types. Since
POM can be emitted either as a vapor or particles, the most
effective control techniques depend on the character of the
emissions. Since significant amounts of POM are most often
generated by incomplete combustion, temperatures are high enough
to vaporize most, if not all, of the POM's generated. Most of
the sampling techniques used measure a total of both particulate
and vaporous POM, as the vapors are captured in water or a sol-
vent. Therefore, it is generally not possible to determine the
relative amounts of the POM's existing as vapors and as par-
6 S /
tides at any given stack condition. Natusch and Tomkins
have shown by a theoretical analysis that most POM's will exist
as vapors at common stack temperatures (>150°C), at least for a
typical coal-fired power plant. When the gases exit the stack,
the rapid decrease in temperature would then cause most of the
POM to adsorb onto particles. As adsorption is dependent on the
surface area to volume ratio of the particle, the POM's will
preferentially condense onto the smaller particles. If the
POM's exist as vapors in the exhaust stream, then control tech-
niques such as afterburners, catalytic adsorption, or conden-
sation (e.g., by. the temperature reduction and capture in a wet
scrubber) would be more effective. In the event of a stack with
lower temperatures, higher particulate loadings, or finer parti-
cles, more of the POM will be present in the particulate form.
Fine particle control equipment such as fabric filters, electro-
static precipitators, or high pressure drop scrubbers are then
more effective in removing POM.
The control techniques which would be expected to be effec-
tive for each source type were then considered to assess their
feasibility, and data were collected on the effectiveness and
application of the various techniques. The characteristics of
15
-------
DRAFT
DO NOT QUOTE OR CITE
the gas stream, the characteristics of POM and other pollutants
in the stream, and cost considerations were assessed to deter-
mine whether a control technique could be feasible. Emissions
data and other material were reviewed in order to ascertain the
practical effectiveness of the technique. Also, the industrial
and air pollution control literature was surveyed to assess the
degree and application of control that has been accomplished.
D. Population Exposure
Very rough estimates of population exposure were made based
on the limited ambient BaP sampling data available because the
available "production" information was not adequate to estimate
ambient concentrations from localized emissions. Although
actual production or capacity data were available for most
utility and industrial plants, fuel consumption and data on the
magnitude of other area emissions were generally not available
for localities. The sources were screened using available data
to ascertain if they were likely to individually generate "sig-
nificant" ambient concentrations of BaP. "Significant" ambient
3
concentrations of BaP were arbitrarily defined as 0.4 ng/m . A
3
concentration of 0.1 ng BaP/m was considered. Many cities, how-
ever, have ambient concentrations greater than this level and a
large number of sources potentially could have been estimated to
3
produce "significant" concentrations. The 0.4 ng BaP/m concen-
tration was chosen in consultation with EPA because few non-
industrial cities had higher ambient concentrations.
Individual point sources of the various source types other
than coke ovens were found to produce ambient concentrations
that, at most, were only slightly higher than the arbitrary
3
measure of "significant" ambient concentrations, 0.4 ng BaP/m .
This is based on the emission factors that were developed—average
16
-------
and maximum plant sizes, and typical but conservative stack
characteristics (designed to generate high estimates of ambient
concentration). That is, only extremely large plants with very
conservative stack conditions were estimated to generate "sig-
3
nificant" concentrations (^0.4 ng/m ) of BaP. The assumed
relationship between the emission and stack characteristics and
the ambient concentration was that of the EPA model, PTMAX. As
PTMAX models hourly rather than annual averages, which may
differ by a factor of ten or more, this is a rather conservative
assumption. Therefore, individual point sources were eliminated
as potential producers of ambient BaP (annual average) concen-
3
trations greater than 0.4 ng/m .
Each area source of BaP was also shown to produce less than
"significant" concentrations of BaP. The major area sources of
BaP are the various diffuse consumptive fuel uses, e.g., heating
oil, gasoline, and intermediate boilers. Since fuel consumption
data broken down by use or locality are not generally available,
the ambient concentrations produced by the area sources in
specific or worst case localities could not be estimated direct-
ly from the emission factors. Therefore, for screening pur-
poses, the Air Quality Control Region (AQCR) Emissions Report
from a National Emissions Data System (NEDS) run by EPA on
September 7, 1977 was used to estimate the worst case ambient
concentrations from individual area sources. The greatest
amount of annual particulate emissions from the various area
sources in the AQCR's listed in the emissions report were se-
lected. It was then assumed that the BaP emissions would large-
ly be adsorbed onto particles after leaving the stack and, thus,
that they were proportional to the particulate emissions. The
assumed ratio of proportionality was that of the estimated 3aP
emission factor to the AP-42 particulate emission factor.109/
17
-------
The ambient concentrations of BaP generated by each source type
were then estimated using the Hannah-Gifford urban area source
126 127/
model. ' ' The area assumed for the emissions in each AQCR
was the total of the urbanized areas for the metropolitan areas
in the AQCR given in a recent Department of Transportation re-
port. a very conservative wind speed of two m/s was used.
The results showed that only the worst case AQCR's were even
marginally significant with regard to ambient concentrations of
BaP produced from each individual area source, assuming the
maximum estimate emission factor. Considering the best estimate
emission factors of other cities, the individual area sources
are quite unlikely to produce ambient BaP concentrations greater
3
than 0.4 ng/m . The NEDS data on some major BaP emitters, e.g.,
residential coal use, seems to be quite high and thus outdated,
as consumption has decreased significantly over recent years.
Therefore, this screening technique is even more conservative.
It is presumed that no area sources produce individually signi-
3
ficant concentrations of BaP (MD.4 ng/m ).
Since more reliable data on local fuel consumption were not
found, the measured ambient BaP concentrations were used to
estimate population exposures. If more accurate local fuel
consumption data had been available, it might have been possible
to estimate ambient concentrations in various localities con-
sidering the emission contributions of the different point and
area sources. As these data were not available, population ex-
posures were estimated using the limited ambient BaP sampling
results that are available and the results of a recent study on
118/
population exposure to coke oven emissions. '
The general procedure followed was to use the estimated
coke oven-derived and background concentrations and number of
18
-------
118/
people exposed developed by Suta for cities with coke ovens.
For cities without coke ovens in which BaP samples had been
taken, actual or extrapolated "1975" annual average BaP concen-
trations were used. An attempt was made to determine a relation-
ship, or significant differences, between the concentrations
measured in cities of differing population density or popula-
tion. However, no such trends were found. Therefore, the
ambient BaP (annual average) concentrations in other non-coke
oven areas were estimated by three averages calculated from the
ambient measurements in non-coke oven areas of cities greater
than 25,000 population, towns and cities of 10,000 to 50,000
population which were not in an urbanized area, and rural coun-
ties and parks. The population exposure to BaP for each, state
was then estimated using the 1970 census figures for the urban
population in a state and within an SMSA (Standard Metropolitan
Statistical Area), the urban population in the state and outside
all SMSA's, and the rural population of a state. These popula-
tions were counted as exposed to ambient concentrations develop-
118 /
ed from the coke oven study, ' actual or projected "1975"
measured ambient concentrations, or the average nori-coke oven
ambient concentrations in that hierarchical order.
Because of the questionable comparability of the various
ambient sampling data (between sources, years, and monitors) and
the limited number of sampling locations (only, several hundred
monitors ever operated in the entire country), these estimates
are very approximate. However, to refine the estimate, signi-
ficantly improved or increased monitoring and/or consumption and
emissions data on a local level would be required for the entire
nation. Therefore, this type of estimate is probably the only
type that is currently feasible.
19
-------
SECTION III
THE SOURCES OF POM
A. General
In this section, the sources of POM are discussed with
regard to their individual source and emission characteristics.
For each source type, the process, POM emission sources, air
pollution control equipment, and the locations and capacities of
individual sources are briefly discussed. The emission sources
mentioned are those which have shown, or are likely to emit, POM
from the process under consideration. The discussion on air
pollution control equipment considers the equipment generally
used to control emissions from the process, the likely character
of POM emissions, and, if available, some measures of the ef-
fectiveness of the control technique in controlling POM emis-
sions from the process and measures of the degree of application
of the various control devices. If available, the locations and
capacities of stationary sources were listed in various appen-
dices.
The development of the emission factors for the various
sources for which emission tests were available is then out-
lined. If data were available, minimum, maximum, and best
estimates of emission factors for benzene soluble organics
(BSO), those POM's which had been analyzed, and BaP were develop-
ed. Generally, the intermediate emission factors were calcu-
lated by taking the geometric mean of the individual test re-
sults. In the text, only BaP emission factors are usually noted
as they are available for more sources and are more directly
comparable.
20
-------
A complete listing of all the emission factors for BSO,
POM, and BaP for the individual sources that were developed in
this study are given in Table III-l. By multiplying the BaP
emission factors times the most recent national production or
consumption figure available in the literature for the source
type, estimates were made of "current" annual BaP emissions.
The ranges of the estimates and the best estimates of annual BaP
emissions for the various source types are given in Table III-2.
The future capacity and POM control of the source type was then
projected to 1985. The estimated production or consumption
capacities for 1985 were taken from published reports or govern-
ment estimates, projected using reported trends, or extrapolated
using an assumed trend. The estimates of annual BaP emissions
in 1985 are presented in Table III-3. As is the case with all
projections to the future, these estimates are, at best, an
indicator of the relative contributions of the various source
types to BaP and thus, presumably, to total POM emissions.
These estimates are especially uncertain because they are based
on a very limited amount of actual emissions data.
The remaining sections of Section III discuss the indivi-
dual source types of POM. For some of the source types dis-
cussed, e.g., coke ovens, future energy sources, and aluminum
reduction, little quantitive emissions data are available. In
such cases, emission factors are often estimated from a POM
concentration per mass of particles or per volume of gas.
Further work is recommended in assessing the emissions from
these sources.
Several other sources are known to emit POM, but in the
quantities of emissions or the prevalence of the sources are un-
certain. Limited data are available on POM emissions from lacquer
coating, charbroiling of foods, heavy duty stationary diesels,
21
-------
TABLE III-I
ESTIMATES OF POM EMISSION FACTORS
2/
-2/
INTERMEDIATE*/
2/
TOT All
NUMBER REFBR-
OF TESTS ENCES
N>
ro
COAL-FIRED POWER TLANTS
O Pulverized Coal
(Vertically-fired, dry-
bottom)
• Pulverized Coal
(vertically-fired,
dry-bottom)
(% Pulverized Coal
(front-wall fired,
dry-bottom)
• Pulverized Coal
(tangentially-fired,
dry-bottom)
0 Pulverized Coal
(opposed-,downward-
inelined burners?
wet-bottom)
• Crushed Coal
(cyclone-fired,
wet-bottom)
• Crushed Coal
(spreader stoker,
traveling cerate)
• Pulverized Coal
(dry-bottom)-
weighted bv boiler
popula tlon^'
• Pulverized Coal
(dry-and-wet-bottom)
weighted by boiler
population
• All types -
weighted by boiler
population '
None
Multiple
Cyclone
& ESP
ESP
Multiple
Cyclone
fi ESP
Multiple
Cyclone
Multiple
Cyclone
26mg/kg 0.Cpg/kg 0.57/*»/kq 87mg/kg JlOpg/kq l.?/i-'/kg 41roq/kg ?.?/lq/kg 1.9/fq/kq
31rog/kg 9.4/n/kq 0.49/fg/kg 51mg/kg l?//g/kg 0.62//g/kg 39mq/kg 63/ig/kq ll//i/kg 33mg/kq l?/fg/kg 1.7//g/kg
29mg/kg 13pg/kg 0.68/fj/kg 65mg/kg 100//q/kg 2.4//j/kq 44mg/kq 37/yg/kg 1.3//g/kq
3lrog/kg 39,94
-------
TABLE III-l (CONTINUED)
ESTIMATES OF POM EMISSION FACTORS
-2/
.2/
INTERMEDIATE
1/
J*/
TOTAL
NUMBER REFEP"
OF TESTS ENrF-S
to
Ui
COAL FIRED INDUSTRIAL
• rulverized Coal
(water-tube,
dry-bottom/ Multiple
200,000 lb stm/hr.) Cyclone
• Chain grate stoker
(water-tube,
125,000 lb stro/hr) None
• Spreader Stoker
(water tube,
with reinjector, Multiple
70,500 lb stm/hr) Cyclone
• Underfeed Stoker
(fire tube,
7.2 x 10 Btu) None
• Underfeed Stoker
(fire tube,
3.0 x 106 Btu) None
• Underfeed Stoker
geometric me&rt None
• All types-weighted
by boiler popu-
lation''/
COAL-FIRED RESIDENTIAL FURNACES
• Underfeed stoker (cast-
iron sectional boiler,
-50,OOOBtu/hr) Nor
• Underfeed stoker (hot-
air furnace, —90,000
Btu/hr) Non
« Hand-stoked (hot-air
furnace, —90,000 Btu/
hr) Non
• Underfeed stokers —
geometric average Nnn
8flmq/kq 630mrj/kq lfimq/kg B40pg/kq
270mq/kg 6. Jmg/kg 2f>0pg/kg 620mg/kg 2frnq/kg 2-Jroq/kq ii0mq/kg I3mg/kg 74f>pg/fcq
2.9g/kq I30mg/kq I3mg/kg llq/kq
1. 2q/kg lOOmq/kq 5.5q/kq fl70rog/kg 42mq/kq
Mmq/kq ROOjii/kq
L7,20, *9
72,110
1 11
39
39
5 39,94
17,20
39,72,110
I 1 I
-------
TABLE III-l (CONTINUED)
ESTIMATES OF POM EMISSION RATES
TOTAL
NUMBER RFFFR-
OF TKST5 ENCES
COAL-FIRED RESIDENT1AL FURNACES
(Continue^)
• Hot-air furnaces --
geometric average Nor
+ All typos --
qcomotr.ic mean I4or
1.9g/kg lf>Omg/kg fl.3n>q/kq
1.3q/kq 67rog/kg 3.5mg/kcr
39
39
OTHER SOLID FUEL BURNING
DOMESTIC SOURCES
• Domestic stoves
(various coal-bascd
fuels ft stove types)
• Residential fireplaces
(varLous woods, stable
burn ing)-
/
700lJg/kq
37mg/kg7/' <1.2mgA9
379mg/kg
63mgAg7/ 2.5mg/kg
5mg/k 9^^ 11.
48mg/kg l.7mg/kg
to
OIL-FIRED INTERMEDIATE
S team-a tomized (water
tube boiler;process
heating, 23-30 * LO^
Btu/hr)
Low-pressure, air-
atomized (Scotch-
marine boiler;
hospital heating
Btu/hr)
4 . 2
10"
58mg/kg 5.3pa/k.g <0.66pc|/kg 140mg/kg lOOpg/kg 2.0p}/kg R9n»g/kg 23pg/kg L.ljn/kg 7
fc20mg/kq B20|rg/kg 40u?/kg I 39
OIL-FIRED RESIDENTIAL
Pressure-a tornj zec\
(cast-iron sec-
tional boiler;
250,000 Btu/hr)
Prer.sure-a t-omi zed
(hot-air furnace;
140,000 Btu/hr)
Vaporized (hot-
air furnace;
90,000 Btu/hr)
A1J typ«es —
q^ometrj.c mean"
8/
350rnn/kg <^70yg/kg 2.0ir-|/kq 1
150mq/kq <6.4pq/kg <2.2[i3/kg 1
ir:q 3
-------
TABLE III-l (CONTINUED)
ESTIMATES OF TOM EMISSION FACTORS
SOURCE
2/
.2/
INTERMEDIATE
1/
-2/
TOTAI.
NUMBHR RF.FF.R-
OF TESTS CHCF.S
GAS-FIRED INTERMEDIATE
BOILERS 9/
Premix burners (fire-
tube boiler; process
heating; 7.2 x 10
Rtu/hr)
Premix burners (Scotch-
marine boiler; hos-
pital heating*
4.2 x 10 Btu/hr)
84rog/m'1 490*/n/rr»3 < MOjjci/nr1
?lnn/m3 l.lmq/m^ 7;6jiq/m ]
GAS-FIRED RESIDENTIAL
FURNACES^7
NJ
ui
Prwiix burners
(double-shell
boiler; 180,000
Btu/hr)
Premix burners
(hot-air furnace;
210,000 Btu/hr)
Premix burners
(wall space heater;
25,000 Btu/hr)
Premix burners
(*180.000 Btu/hr)-
geometric mean
Premix burners
025,000 Btu/hr) -
geometric mean
16n»g/m* 19ug/m* 0.82|ig/tn3
llmg/m 13»ig/m' l-Opg/m^
3 3 3
200mg/m l.lrag/nt 10pg/m
13mq/m 16yg/m^ 0.9lng/m
33mq/m3 66yg/m3 2.0uq/m3
I 39
1 39
2 39
3 39
PETROLEUM CATALYTIC
CRACKING--CATALYST
REGENERATION
Fluid catalytic
crack i ng
£ 23,000 bpsd
Fluid catalytic
cracking
5 46,600 bpsd
Fluid catalytic
crack ing
< ->3,000 bp.sH
None
(Regenerator 3 3
Outlet)' 1.3g/m L.2mg/m" 27in/m
None
(Regenerator
Outlet) 1.3q/m 1.2mg/m3 27[iq/m^
CO Wacte
Heat
Boll or 5.3n/m3 ^Onji-j/m3
.11/
4.3g/m' 2.8mg/m"' 280vig/m^ 2.4q/m^ 1,8r»q/in3 87|jq/m^ 2 39
llq/m* 2.9g/m^ 2 - 9mtj/m'' 4-On/ni3 21mq/m' 2Rnug/m"* 3 ^
~j.Oq/m-1 I . 7mq/m3 f>* '|i i/n> •* S -
-------
TABLE 1II-1 (CONTINUED)
ESTIMATES OF POM EMISSION FACTORS
u2/
.2/
INTERMEDIATE
1/
-2/
TOTAL
NUMBER REFER-
OF TESTS ENCES
PETROLEUM CATALYTIC CRACKING —
CATALYST REGENERATION (Continued)
Fluid catalytic
crack ing
£ 46.600 bpsd
CO Waste
Meat
Boiler
Therroofor catalytic None
cracking (air lift'* (Regenerator
19,600 - 2 1,800 bpsd) Outlet)
Thormofor catalytic
crocking (bucket Nono
lift 10,000 - (Regenerator
13,200 bpsd) Outlet)
Thermofor catalytic None
cracking-All types (Regenerator
£ 2.1,800 bpsd
Houdriflow cata-
lytic cracking
34,400-37 r 200 bpsd
Houd r i f]ow ca ta -
lytic cracking
34,400-37,200 bpsd
All types weighted
by cracking popu-
lation^"^
All types weighted
by crackinq
populat ion /
Outlet)
None
(Regenerator
Outlet)
CO Waste
Heat
Boiler
None
(Regenerator
Outlet)
CO Waste
Heat
Boiler
COKE PRODUCTION
15/
Wet coal charging
Pushing
Quenching
Door Leaks
16/
J7/
1.4g/m~ 29Cuq/m
311/ 6.0g/m3 2.2mg/m3 140pq/m3 3.5g/nr* 910n;i/m3 39u^/m3 11 ^ 3
77q/m3 3.5g/m3 350n>g/m3 230q/m3 5.J<|/m^ 7^0mg/m3 1 irvj/m3 4.Sg/m3 470mq/ni3 3 32,39,82
l.Gq/m3 Z.Umg/m3 ND'?/
57g/m 6.6g/m3 1.3g/m3
1.4q/m3 1.3mg/m3 2fiug/m3
1.4g/m 29C)jg/m3 6.4ijg/m3
500ntg/Mg
8g/Mg
llg/Mg l.6mg/Mg
4.2g/Mg
18/
2.4g/m3 3.0mg/m3 190ug/m3 1.9g/m^ 2.9mg/m3 78ii j/m3
2?g/m3 240mg/m3 26mg/rn3
80g/m
3 9. lg/in3^3/'l. 4g/m3 70g/m3 7.1g/m3 1.4g/m3
3.1tng/m3
570mg/rn^ 2.7mg/m3 280|jg/nt3
14g/m 2.9g/m 64Cjjg/m3 4.Sq/m 2Smg/m 370u^/n>
5.9g/m3 2.2mg/m3 l4Cug/m3 3.Sg/m3 930u^/m3 39jj i/m3
Battery Stacks
280mQ/Hg 550g/Mg 1.7g/Mg
17g/Mg
-^0 2-Bkg/Mg 9 . 4mg/Mg ^^5 . Omg/Mg
1.8mg/Mg 260g/Mg 1.4g/Mg
2.Omg/Mg 21mg/Mg
SOOq/Mq
3 Og/Mg
1.5g/Mq
1.Og/Mg
5.Omg/Mq
19/
2 50g/Hg 4.3g/Mg 1.5g/Mg
1.6g/Mq
12mg/Mq
5 39
2 32,39,07
1 32,39,*a
10 39,71
4 39,71
7,3 2,48,76,
91,109,110
1.14,130
130
130
130
130
130
-------
TABLE III-1 (CONTINUED)
EMISSION OF POM EMISSION FACTORS
MINIMUM
MAXIMUM
INTERMEDIATE1^
TOTAL
SOURCE CONTROL. BSO
POM2/ BaP BSO
2/
P0W BaP
BOS
pom2/
NUnBtK
BaP OF TESTS
ENC
Asphalt Production**0/
e Saturators—shingle
(55 lb/480 ft2)
None
1 mg/Mq2*''' 200 pg/Hg
8 mg/Mg2M>00 iiq/Mg
3 mg/Mg21/
400
l' g/Mg
2
14
© Saturators—roll
(27 lb/480 ft2)
Hone
5 mg/Mg 2V ND22/
8 mg/Mg^^ mg/Mg^"'/'
6.4 mg/Mg^^
300
ug/Mq
2
34
© Saturators--shingle
(55 lb/480 ft2)
HEAP
3 mg/Mg21/
200
lig/Mq
1
34
o Saturators—shingle
(55 lb/480 ft2)
Afterburner
I mg/Mg 21/
< 80
|ig/Mg23/
1
34
© Saturators—shingle
(55 lb/480 ft2)
Controlled
2 mg/Mg21/
100
Ug/Mg
2
14
© Saturators--roll
(27 lb/480 ft2)
HEAF
1 mg/Mg21/
500
ug/Mg
1
34
o Saturators—roll
(27 lb/480 ft2)
Afterburner
50 mg/Mg21/
40
mg/Mq2-*/
1
34
© Saturators—roll
(27 lb/480 ft2)
Controlled
8 mg/Mg21/
4
mg/Mg
2
34
© Saturators--general
products
Controlled
4 mg/Mg2^/
1
mg/Mg
4
34
© Air Blowing (high
melting point asphalt)
None
4.8 mg/Mg 160 ug/Mg
4.1 g/Mg <13 mg/Mg21/ 2
*g/Mg
60 mg/Mg21/
2
mg/Mg25/
3
34, 39
© Air Blowing (high
melting point asphalt)
Process Heater
Furnace
2.4 mg/Mg2Ml30 ug/Mg25/
9.3 mg/Mc^^750 mg/Mg
4.7 mg/My21/
500
ug/Mg25/
2
34
© Air Blowing
Baffle
21
kg/Mg
4.8 g/Mg
<5.8
mg/Mg
1
39
© Air Blowing—geometric
mean
Controlled
50 mg/Mg26/
1
mg/Mq
3
34, 39
© Hot Road Mix
Cyclone
20
g/Mg
9.5 mg/Mg
690
uq/Mg
1
?9
© Hot Road Mix
Cyclone and
Spray Tower
11
q/Mg
4.1 mg/Mg
<60
ug/Mg
1
39
Other Industrial Processes27/
© Iron and Steel Sintering
600 g/Mg2f3/
1.1 g/Mq
17
mg/Mg
4
76
© Chainlink Fence T.acquer
Coating
470
mg/Mq
I
7ft
© Carbon Black Production
220 mg/Mg 2^/
490 mg/Mg29/
490 mg/Mg 29/
2
44 , n;
-------
TABLE TIJ-i {CONTINUED)
F^TTHATES OF POM F.MISS ION FACTORS
TOTAL
NUMBER REFfiR-
OF THPTS F.NCRS
TCIClttF.RATPKS ^
•
Municipal (multiple
chamber; batch;
rec i procating
stoker qrates)
5HT/D
Mone
6.Omg/kg
31Oug/kg
I 3ug/k
m
Municipal (multiple
chamber; batch;
rec i procating
stoker grates)
50T/D
Wa tor
Spray
Scrubber
13mq/kg
16uq/kg
200ng/kg
L
3?
•
Hunj.ci pal (multiple
sect ion; cont inuoun;
rocking bar grate)
30MT/D
Water
Spray
Tower
& ESP
17.4ug/kg32''
82ng/kg^J//
1
25
•
Mun ici pa I--
geometric moan
Controlled
) 3n»a/kg
17uq/Kg33/
1JOng/kg25'
2
2S.39
•
Conunercial (s ingle
chamber) 5.3T/D
None
2<"mn/Kg
2.Img/kg
120ug/kg
1
39
•
Comjnorcial (multiple
chamber with
auxiIiary gas
burner in primary)
]T/D
None
J40TTiq/kq
?2mq/kg
S70in/kq
1
•
Cornmorcial--
TMtnotr i c mean
None
A?nvj/kq
. Onvi/kg
?r,nuo/kT
7
31
-------
TABLE II I-1 (CONTINUED)
EMISSION OF POM EMISSION FACTORS
I INTERMEDIATE
1/
J-f
TOTAL
NUMBER REFER-
OF TESTS ENCES
¦ jri.N BURNING
30/
to
vo
Mini Ic l|iol refuse
Automobile tires
Automobile bodies
Automobile components
Autcxnobl le comjionents--
gcometric average
with assumed mix*7/
Cr.iss clippings,
leaves, branches
(>««ar burning (red oak
leaves)
I.«?»if burning (sugar
tn.iplc leaves)
Leaf burning (sycamore
leaves)
Lcdf burnlng--
geometric mean of
to.qta for J types
Loaf burninq
(composite of 3
types in each-
test)
5.5g/kg 500ygAg^^ 88iig/kg34^ 7 „ 4g/kg 4.7mg/kg3^ 340pg/kg 7. 4g/kg^^ 1 . 4mg/kg 170pq/kg
190rog/kg ?0mg/>g
260ng/kg
240k>9/kgW/' SSmg/Vg34''
llOmg/kq3^ ldmg/kq34^
29roj/k'
I0mg/kg30/ 325|,gAgW
l5n»gAg38/ 190pg/kg39/
32,39.109
39,109
39,109
32,39
3 32,39,109
2 32.39.109
45
45
45
45
A. |R 10il LTURAL & FQKEST
F1 RES W *
• Bagasse boilers Cyclone
• Forest fires
42/
• forest fires42''
(heading, fJanlng)
« Forest fires
4 3/
(headingr smoldering)
• Forest fires
(heading; overall) 43q/kg
6. 4gAg
BSOy^Ag40' ND
0.39mqAg44/ 3&ugAg
31. Sng/kg 44^140ggAg
22.8o>gAg
-------
TABUS Iir-1 ,63
49,93
2 .17,72,92
1 10
Motorcycles (2
stroke engines)
Rubber tire wear
51/
2.9mg/l
l4og/io
6 S2/
17,27,62,
1 1"
-------
FOOTNOTES TO ESTIMATES OF POM EMISSION FACTORS
^ Intermediate estimates are geometric means, except as noted
otherwise.
2/
POM values reported are the sums of the quantities of ten
POM species (pyrene, benzo(a)pyrene, benzo(e)pyrene, perylene,
benzo(ghi)perylene, anthanthrene, coronene, anthracene,
penanthrene, fluoranthene) present in the particulate from
128/
the front and back halves of EPA Method Five as detected
by separation by column chromatography and analysis by ultra-
violet visible spectrophotometry, except as noted otherwise.
Weighted for boiler population by tons of coal burned in
boiler type from FPC data for 1972 reported in Reference 94.
Assumed dry opposed-firing had emission characteristics of
dry vertical firing, wet opposed firing representative of
all wet (6230; 16,700; and 12,300 x 103 metric tons) [6780;
3
18,420; and 13,520 x 10 tons] of coal burned by opposed,
front, and tangential firing, respectively, wet cyclone
representative of all cyclones, and spreader stoker repre-
sentative of all stokers. Breakdown of coal burned in
thousands of metric tons (thousands of tons) is 99,600
(109,840) tangential; 55,700 (61,450) fromt; and 32,500
(35,850) opposed firing, for a total of 188,000 (207,140)
pulverized coal, dry bottom; 35,200 (38,810) pulverized coal,
wet bottom; 35,500 (39,090) cyclone; and 3200 (3500) stoker.
4/
' Weighted for boiler population by heat content of bituminous
coal consumed by boiler type from FPC data for 1973 reported
in Reference 94. Assumed pulverized wet and cyclones had
emission characteristics of weighted average of industrial
31
-------
boilers; pulverized coal dry-bottom test representative of
all pulverized coal, dry-bottom; spreader stoker with rein-
jection test representative of all spreader stokers; chain
grate stoker test representative of overfeed stokers; and
geometric average of two underfeed stoker tests representa-
tive of all underfeed stokers. Breakdown by millions of
12
kilograms (10 Btu) consumed is 26 (650) pulverized dry, 5.3
(130) pulverized wet, 18 (40) cyclone, 1 (30) overfeed
stokers, 18 (450) spreader stokers, and 0.8 (20) underfeed
stokers. The 1973 average bi uminous coal heat content of
6.22 x 10^ cal/kg (224 x 10^ Btu/ton) was assumed.
BaP isolated by pyrolysis of styrene-containing tar and
separation by thin-layer chromatography using procedure of
G.M. Badger and R.G. Butkry, and measured by ultraviolet
spectrometry and gas chromatography.
6 /
Geometric average counting four non-detectable levels at
0.1 mg BaP/kg coal (0.7 mg BaP/kg is lowest detected BaP
emission factor reported in Reference 6).
7 /
Sampled using Tenas adsorber following a glass-fiber filter.
Twenty-two POM's measured in 18 categories (anthracene +
phenanthrene, methyl anthracenes, fluoranthene, pyrene,
methyl pyrene + fluoranthene, benzo(c)penanthrene, chrysene
+ benz(a)anthracene, methyl chrysenes, benzofluoranthenes,
benz(a)pyrene + benz(e)pyrene, perylene, -methylcholanthrenes,
indeno(1,2,3-cd)pyrene, benzo(ghi)perylene, dibenzo(a,h)an-
thracene, diebenzo(c,g)carbazole, diebenzo(ai+ah)pyrenes,
coronene) using GC/MS and totals reported as POM values.
BaP and BeP reported by a combined measure. The values
given were calculated from the reported one test POM and
BaP+BeP concentrations in the front half (filter & probe)
particulate catch for both alder and Douglas fir and the
average front half particulate emissions factor for the wood
32
-------
type. The average front half particulate emission factor
for each wood type was calculated in the same manner as were
the total particulate emission factors in Reference 89, i.e.,
the average pollutant mass rate was divided by the average
burning rate where these averages were each the average
values for stable burning and start-up. For purposes of
comparison, the emission factors for the individual tests
with POM sampling for alder, Douglas fir, and pine were
calculated. The resulting values were 38, 43, and 17 mg
POM/kg wood and 0.79, 1.7, and 1.3 mg BaP+BeP/kg wood,
respectively.
8 /
BaP value calculated assuming an actual emission factor equal
to 7 5 percent of the reported maximum value.
9/ •
' Emission factors in terms of mass of pollutant emitted per
volume of gas burned were calculated from emission factors
in mass of pollutant emitted per heat input given in Refer-
ence 1. The ratio of heat input to weight of fuel burned
used was that given by the operating conditions of each
test. Assumed a natural gas composition of 94.2 percent
methane, 3.6 percent ethane, and 2.2 percent nitrogen, 94.2
percent methane and 3.6 percent ethane given in Reference 1
for a molecular weight of 16.8 g/g-mole, and a perfect gas
at standard conditions (0°C, 1 atm) for a molecular volume
of 22.4 1/g-mole.
Emission factors reported as mass per cubic meter of fresh
feed plus recycle charged.
Mo BaP detected in one test. A value of 1.0 yg/bbl was as-
sumed in calculating the geometric mean for a best estimate.
33
-------
12/
No BaP detected in one test. A value of five yg/bbl was as-
sumed in calculating the geometric mean for a best estimate.
13/
POM value reported for a further analysis measuring a total
of 23 POM species (acridine, benzo(f)quinoline, benzo(h)qui-
noline, penanthridine, benz(a)acridine, benz(c)acridine,
benzo(mn)phenanthridine, indeno(1,2,3-ij)isoquinoline, 11H-
indeno(1,2-b)quinoline, dibenz(a,h)acridine, dibenz(a,j)acri-
dine, anthracene, phenanthrene, benz(a)anthracene, chrysene,
fluoroanthene, pyrene, benzo(a)pyrene, benzo(c)pyrene, pery-
lene, benzo(ghi)perylene, anthanthrene, coronene) for selected
tests as in Reference 1. Data from analysis in Reference 82
as reported in Reference 32. Values reported for ten POM's
(see footnote 6) were used in calculating best estimates.
3 3
Ten POM values are 3.5 g/m for TCC air lift, 7.6 gm/m for
HCC uncontrolled.
14/
Weighted for cracking population by the full capacity in
barrels per steam day of the total catalytic cracking capa-
city of the U.S. in 1977 from Reference 71. The breakdown
in cubic meters (barrels) of fresh feed plus recycle per
stream day is 816,000 (5,133,425; 94.2 percent) for fluid,
46,600 (292,900; 5.4 percent) for thermofor, and 3420
(21,500; 0.4 percent) for Houdriflow.
15/
Cokd production emission factors given per megagram of coal
charged were taken directly from Reference 130. The ratio
of coal charged to coke produced is generally about 1.45.
For comparison, overall (charging and coking) emission fac-
tors were calculated from ratios of BSO and BaP to total
particulate reported in References 32, 48, 110, and 114 and'
a particulate emission factor for charging and coking of
1.1 kg/I-Ig (0.75 kg/Mg for charging, 0.05 kg/Mg for the coking
cycle, and 0.3 ,g/Mg for discharging) from Reference 109.
34
-------
The intermediate values (ranges) of these emission factors
are 450 g/Mg (300-700 g/Mg) for BSO and 580 g/Mg (660 vig/Mg
- 3.2 g/Mg) for BaP.
Comparison emission factors for wet coal charging calculated
from the results reported in Reference 7. Emission factors
calculated for a reported charge of approximately 35 T as-
suming the average particulate emissions for all tes.ts. of
815.7 grams for the uncontrolled Wilputte car and 120.0 grams
for the AISI/EPA controlled Larry car. Values reported as
tar fractions in the particulate were assumed to be BSO.
BaP ranges and best estimates were calculated using the
reported concentrations of BaP in the tar fraction for var-
ious impactor. and collector stages from various tests. ¦ There-
fore, they are very rough estimates. Also, the results of
this study are questionable as there were sampling problems
(e.g., deposition in sampling lines) and the results were
an order of magnitude lower than indicated by other studies
which only attempted to collect a portion of the emissions.
Therefore, EPA"^^ did not consider these test results. The
calculated intermediate values (ranges) of emission factors
were 12 g/Mg BSO (7.. 5-22 g/Mg) and 13 mg/Mg BaP (2.0-13 mg/
Mg) for the Wilputte car, and 2.3 g/Mg BSO (2.1-2.4 g/Mg) and
2.3 mg/Mg BaP (540 pg/Mg-44 mg/Mg) for the AISI/EPA Larry car.
17/
Preliminary sampling results of quenching from Reference 91
reported total polynuclear aromatic hydrocarbons using a
GC/MS computer analysis. The EEA calculated intermediate
estimate is 13 mg/Mg (range 10.6-15.9 mg/Mg). No BaP was
detected.
18/
/ POM values are the sums of all the minimum values and all the
maximum values reported by EPA^^^^ as ranges for different
POM's. Ten measures of 13 POM's (anthracene + phenanthrene,
35
-------
methyl anthracenes, fluoranthene, pyrene, methyl pyrene +
fluoranthene, benzo(c)phenanthrene, chrysene + benz(a)anthra-
cene, methyl chrysenes, dimethyl benz(a)anthracene, and
benzo(a)pyrene) were reported.
19/
POM values are the sums of all the minimum and all the maxi-
mum values reported as ranges for 14 POM's (benz(a)phenan-
threne, benz(e)pyrene, benzfluoranthrenes, benzo(k)fluoran-
thene, chrysene, dibenz anthracenes, dibenz pyrene, dimeth-
ylbenz(a)anthracene, fluoranthrene, indeno(1,2,3-cd)pyrene,
napthalene, pyrene, and benzo(a)pyrene) in Reference 13 0.
20/
Asphalt saturator emission factors are given in mass per
Metric ton (Mg) of product. Asphalt air blowing emission
factors are given in mass per metric ton (Mg) of asphalt
charged assuming a density of 0.96 kg/1 (810 lbs per gallon).
Hot road mix emission factors given in mass per metric ton
(Mg) processed.
21/
POM value reported for samples collected by EPA Method Five
or a modification of Method Five and analyzed for seven POM
species (benz(c)phenanthrene, 7,12-dimethylbenz(a)anthracene,
benz(e)pyrene, benz(a)pyrene, 3-methylcholanthrene, dibenz-
(a,h)pyrene, dibenz(a,i)pyrene) using gas chromatography and
mass spectrometry.
22/
No BaP detected in one test. A value of 0.11 mg/Mg (0.1 mg/T)
was assumed in calculating the geometric mean for a best es-
timate .
23/
BaP value reported is for combined BaP and BeP. The extremely
high value for the roll product with afterburner control was
thought to have been caused by interference in the analysis.
36
-------
24 /
Ten POM (see footnote 2) values of <13 mg/Mg reported. A
value of 10 mg/Mg was assumed for the best estimate calcula-
tion .
25/
At least one value of BaP reported is for a combined analysis
result for BaP and BeP.
6 /
POM value is geometric mean of results of analysis for ten
POM's (see footnote 2) from a baffle and results of analysis
for seven POM's (see footnote 21) from a process heater
furnace.
27/
Industrial emission factors are given on a basis of mass per
metric ton (Mg) of sinter feed for iron and steel sintering
and on a basis of mass per metric ton (Mg) of chain link
fence through the lacquer coating bath. Emission factor for
carbon black production is calculated from a POM loading in
3
mg/sm from Reference 44 by assuming an air flow of 3.91
sm^/kg (135,000 scf/T) from Reference 112.
Minimum BaP value of 600 yg/Mg was suspect as a weight scale
was reading approximately double. Therefore, a value of
1.2 mg/Mg was assumed in computing the best estimate.
29/
POM values are for "total" POM by the gas chromatrographic-
mass spectrometric-computer analysis and quantification tech-
nique reported in Reference 44. The minimum value reported
is for sampling with an EPA Method 5 train, while the maxi-
mum value is for sampling with a Tenax adsorbent column.
An intermediate value of 310 mg/Mg corresponds to the results
for sampling with the Method 5 train followed by an adsorbent
sampler.
37
-------
Incineration, open burning, and agricultural and forest fire
emission factors are given on a basis of mass emission per
mass of refuse charged or other material burned, except as
otherwise noted.
31/
' POM values are for "Polynuclear Hydrocarbons" from Reference
18 as reported in Reference 94.
32/
POM values are for 9 measures of 13 species (fluoranthene,
benzo(a)anthracene + chrysene, benzo(b)fluoranthene + benzo-
(k)fluoranthene + benzo(j)fluoranthene, benzo(a)pyrene +
benzo(e)pyrene, perylene, benzo(ghi)perylene, ideno(l,2,3-
cd)pyrene, coronene) using an EPA Method Five sampling train,
dichloromethane extraction, and gas chromatographic analysis.
33/
' POM value is geometric mean of results of different test pro-
cedures. One sample was analyzed for ten POM's as per foot-
note 2, while the other was analyzed for 13 POM's as per
footnote 32.
34 /
' Emission factor calculated from concentration of POM species
in particulate in the smoke plume (on-site air samples) from
Reference 39 and the emission factors for particulates given
in Reference 109. These particulate emission factors are
8.0 kg/MT for municipal refuse, 50 kg/MT for automobile com-
ponents, and 8.5 kg/MT for agricultural field burning, land-
scape refuse and pruning, and wood refuse.
35/
POM emission factor reported is the sum of the minimum or
maximum end of the ranges for each of the nine POM species
(benzo(a)pyrene, pyrene, benzo(e)pyrene, perylene, benzo(ghi)-
perylene, anthanthrene, fluoranthene, chrysene, benz(a)an- '
thracene) reported in Reference 32.
38
-------
Emission factor reported is for stack results in a facility
for research on open-burning fires.
37/
Geometric average calculated assuming a typical mix of auto-
mobile components of 68 kg (150 lb) tires and 680 kg (1500
lb) automobile body and averaging the results with the re-
search facility test results for mixed components.
38/
: POM values reported for sampling from leaf burning research
facility using a filter and Tenax adsorber, and samples ex-
tracted using methylene chloride for the filter and pentane
for the adsorber, separated by liquid chromatography and
analyzed using gas chromatography and mass spectrometry. POM
values reported are totals for 18 measures of 22 POM species
(as in footnote 7).
39/
' BaP values reported are actually for combined BaP and BeP as
detected in the sampling and analysis procedure outlined
above (footnote 38). Non-detectable (HD) BaP value reported
was assumed to be 40 yg/kg in calculating the best estimate
of BaP emissions.
40/
POM value is for total of six POM's (7,12-dimethylbenz(a)-
anthracene, benzo(a)pyrene, 3-methylcholanthrene, dibenz(a,h)-
anthracene, and two unknown POM's) sampled by EPA Method 5
with a ten foot glass condenser and a Tenax adsorbent plus,
extracted with benzene, and analyzed using a gas chromato-
graph and electron capture detectori
Best estimate of BaP emissions assumed non-detectable BaP
levels in each test to be equal to the minimum detectable
level of 1.0 mg BaP in the POM detected in the test. The
geometric mean of the detectable level emission factors was
taken to produce the best estimate.
39
-------
BSO emission factor per mass of fuel burned calculated from a
range of BSO concentration in particulate of 40-75 percent
(mass) from Reference 81 and using the mass per mass of fuel
particulate emission factor of 8.5 kg/Mg from Reference 109.
The best estimate was calculated by assuming 60 percent BSO
in the particulate.
43/
' Forest fire emission factors per mass of fuel burned are from
reported average results for duplicate tests, each involving
burning slash pine needle litter in a controlled environment
burning room, sampling with a modified "hivol" sampler (which
was kept below 65°C to minimize breakthrough of vaporous POM),
extracting with methylene chloride, separating by liquid
chromatography, and analyzing with gas chromatography and mass
spectrometry. The results of these experiments with pine
needles ranged over several orders of magnitude depending on
fuel characteristics and fire behavior; therefore, it must be
stressed that the emission factors presented may not be rep-
resentative of.burning pine needles in the laboratory, let
alone of forest fires in general.
44/
' POM values reported are total amounts detected of 15 measures
of 18 POM species (anthracene + penanthrene, methyl anthra-
cene, fluoranthene, pyrene, methyl pyrene + fluoranthene,
benzo(c)phenanthrene, chrysene + benzo(a)anthracene, methyl
chrysene, benzofluoranthenes, benzo(a)pyrene, benzo(e)pyrene,
perylene, methylbenzopyrenes, indeno(1,213-cd)pyrene, benzo-
(gui)perylene) using the sampling and analysis procedure dis-
cussed above (footnote 43).
45/
' Best estimates of BaP emissions are all for the same series
of tests using the same fuels.
40
-------
4 6/
' Auto population weighted by percentage of total mileage
traveled by each type"of auto for 1977 using the age distri-
bution of the U.S. auto population for 197 6 from Reference
63 and the average annual miles driven for autos by age from
Reference 120 in Reference 109. The distribution, by per-
centage of annual travel, used was 32.3 percent 1970 catalyst
with unleaded 1975-1977 model years, 48.2 percent 1970 en-
gine modification 1970-1974, 9.5 percent 1968 engine modifi-
cation 1968-1969, and 10.0 percent 1966 uncontrolled all with
leaded gasoline.
47/
BaP emission factor in mg/kg fuel in Reference 93 converted
to mg/1 fuel assuming a density of diesel oil of 7.83 kg/1
(No. 2 oil).
48/
' BSO emission factor calculated from BSO emission factors per
kilometer or various speeds in Reference 49, assuming a time
at speed distribution of 35 percent at 35 km/hr, 35 percent
at 64 km/hr, 25 percent at 88 km/hr, and five percent at
96 km/hr, and a diesel mileage of 9.4 km/1 (22 mpg).
49/
' POM results from Reference 92 for proportional sampling using
a precooler, a filter, and a chromosorb-102 adsorption trap,
extraction with benzene, separation by liquid solid column
chromatography, and analyzed by thin layer chromatography.
POM values reported are for ten species of PNA (benzo(a)-
pyrene, benzo(e)pyrene, benz(a)anthracene, genzo(ghi)perylene,
chrysene, pyrene, anthracene, fluoranthene, phenanthene,
phenanthrene derivatives).
Best estimate emission factors are geometric means of two
tests reported in Reference 92. Results of a later study^^®^
testing emissions in exhaust from a Mack 4-cylinder turbo-
charged diesel in yg/kg BaP of fuel were 25 for idle (no load,
41
-------
60 rpm), 8 for low speed cruise (! load, 1260 rpm), and 16
for lugging (full load, 1800 rpm). These emissions are equi-
valent to approximately 3.2, 1.0, and 2.0 yg/1 BaP/fuel,
respectively.
No polynuclear aromatic hydrocarbons detected in the pre-
liminary analysis of particulate matter collected from tires
run at up to 35 mph with 450 kg (1000 lb) loads around a paved
indoor track as reported in Reference 62.
Estimated BaP emission factor of 140 g (0.3 lb) per day per
million population from Reference 27 cited in References 110
and 17.
42
-------
TA81T. in-2
CST1HATEF Of TOTW. BeF BMIftSIONF BY SOURCE TYPf
Co*l-Tir»d Po»or Plant*
Coal-rirad Industrial
Boi )tr<
Coal-rirad Residential
Furnar**
Other Pnliri fut) Rurninq
Source*
a Drwntic Sto»e*
• Residential
Fi repl *«*¦¦
OH - Fired Intermediate
Rotlera
• Industrial Boilers
• Ccsa»»rcial/ln*ti-
utional Boilera
011-Fired Reaidential
Furnace*
Cas-Fiied Intermediate
Boil«r*
• Industrial Boilers
• ConBercial/Intti-
utional Bolltn
Ga»-Fired Residential
Furnaces
Petrol***" Catalytic
Cracking
• Fluid Catalytic
Crack ino
a . TTionaofnr Catalytic
rri^VUa
• Houdriflcw Catalytic
Crack i no
Coke Production
Asphalt Production
• Sat orator*
• Air Blowing
• Hot Road Hi a
Other Industrial Prcc«aa«a
a Iron i 8t«*l Mnterinc
• Chamlink Fence
L«cq»er Coating
a Carhon Black Production
Annua)
Production or
Fug 1 Conaiwrfio
2.A5 a 10M kq
*.?i * )o10 kq
7. }* % 10° ko
Unknown
4.11 * I"10 *o
3.42 a 10J1 hp
4. 74 x 10U ko
4.44 x JO11 kg
8.04 * 10 •
ft. )#. * 10" a
4. a •"
3.42 a ,03 »3
5.1? i 1010 kg
4.35 a 10 k9
4.35 a 10,.ko
1.95 a 10 kg
3. 70 a 10 V.g
Unkrvwn
1.)fl K 10' k9
fear of f.atted BaP t»i«*ions (Hrtnc ton«/yaar)
Data Hinii» m«lw Intermediate
0. 4f
Production Data Remark*
1974 0.31
19?J 0.047
1973 O.OSfr
197}
1973
1973
1973
) 977
1977
1977
1975
1976
197*
1976
0.0O44
0.050
0.00035
0.00)4
0.7?
1?
74ft
0.000025 0.00?4
o.cmnoois o.ojs
o.ooae
900
0.017
0.025
0.013
0,057
?»•
0. J7
19
0.02)
O.M
o.ooo??
o.onu
(i.mtP
no
0.0044
0.0044
O.ftll]?
104
104
Municipal
Con^rcial
1.83 * 10 kg
B. ? * 10? ko
1974
197?
0.0031
0.9ft
0. ?4
4.7
O. 07?
2. 1
a Municipal Refuse
• Auto Crmforventa
• Gra«« Clippings.
l
-------
a r' •
if i
TABLE III-2
REMARKS
DO NOT QUOTE OR CU E
a/ Emission factors weighted by boiler population.
b/ Fuel consumption includes both industrial (5.61x10"''^ kg)
9
and commercial/institutional (6.02x10 kg) usage.
c/ Emission factors of geometric means of available results
used for best estimate.
d/ Fuel consumption calculated from a projected 1975 fireplace
consumption of 62 million cubic meters (17 million cords) from
Reference 86, reported in Reference 32, assuming a specific
2 3
gravity of seasoned wood of 0.7 g/cm (range 0.11-1.33 g/cm ;
3
most common woods, ^0.5-0.9 g/cm ). Other estimates of wopd
fuel usage based on Btu consumption estimates in References
32 and 94, range from 1.22 to 7.08x10"''^ kg of wood, assuming
6-7 .
2.8x10 kal/kg (10 Btu/ton). Using the conversion emission
factors calculated for the single tests for POM with alder,
Douglas fir and pine, the emission estimates range from
52-110 Mg/yr with a best .estimate of 73 Mg/yr.
e/ Fuel consumption in kilograms calculated from the fuel con-
sumption in Btu's given in Reference 94, assuming a heat con-
tent of 9.72x10^ cal/m^ (146,000 Btu/gal) and a density of
7,830 kg/m3 (#5 fuel oil) for residual oil and a Btu content
of 9.32x10^ cal/m3 (140,000 Btu/gal) and a density of
7,210 kg/m3 (#2 fuel oil) for distillate oil.
f/ Fuel consumption in cubic meters calculated from the fuel
consumption in Btu's given in Reference 94, assuming a heat
content of 9.095x10^ cal/m3 (1,022 Btu/ft3); Commercial/
44
-------
institutional energy consumption figure included 8.5 percent
from LPG and is the total for all uses (66 percent space
heating, 23 percent water heating, six percent cooking, and
five percent air-conditioning). Residential consumption
figure is also for all uses (-70 percent space heating).
g/ Uncontrolled emissions given. Controlled emissions estimated
as <0.00011 Mg/yr (best estimate 3.1x10 5 Mg/yr) for fluid
-7
catalytic cracking, and a single estimate of 9.6x10 Mg/yr
for Houdriflow catalytic cracking.
h/ Assumed an intermediate coke production emission factor of
1.5 g BaP/Mg of coal charged (0.00 3 lb/ton; the minimum quan-
tity of total coke oven emissions suggested by EPA in Reference
130) and a ratio of coal charged to coke produced of 1.45.
The maximum estimate uses the total of the various process
emission factors from Reference 130 of 4g BaP/Mg. The minimum
estimate is based upon the minimum emission factor developed
for comparison purposes. If the other comparison emission
factors are used, intermediate and maximum estimates of 4 3
and 240 Mg BaP/yr are generated.
i/ Production given for both saturators and air blowing is the
total asphalt sales for use in roofing products. Controlled
emissions given.
j/ Production given for hot road mix is the total asphalt sales
for use in paving products. Controlled emissions given.
k/ Assumed 15 percent of ten species POM to BaP (range of BaP
to POM ratio in ambient samples near carbon black plants
was 0.26 to 0.148 as given in AP-33109).
1/ Uncontrolled emissions given, best estimate of controlled
emissions is 0.0024 Mg/yr.
45
-------
m/ Fuel consumption calculated By assuming a heat content for
bagasse of 2.2x10 cal/kg (4,000 Btu/lb). Non-detectable
BaP levels were assumed to be the minimum detectable level
of 1.0 pg in the POM detected in each test and the geometric
average of the tests taken to produce an estimated BaP emis-
sion factor of 2.7 yig/kg.
n/ Minimum and maximum emission estimates for forest fires taken
directly from Reference 111. The maximum emission estimate
was first reported in Reference 17. A recent study for
137/
EPA ' lists prescribed burning emissions of 4.5 Mg/yr.
In order to generate a number 6f known derivation for com-
parison purposes, the emission factors developed from tests
burning pine needles were used. The overall emission factors
were used assuming ten percent backing fires and 90 percent
heading fires. Total U.S. wildlife fuel consumption of
38 Mg/hectare from Reference 109 and total estimated area
burned by prescribed and wildfires in 1976 from Reference 19
were used to calculate an estimate of 100 Mg/yr. This
extrapolation from laboratory experiments may give an indica-
tion of the order of magnitude of emissions from forest fires;
however, its representativeness is questionable. Due to the
great variability and uncertainty in the fuel available and
the burning process in different geographical, seasonal, and
weather conditions, no best estimate can be made at this
78/
time. '
o/ Maximum estimate of 308 Mg/yr (340 T/yr) for 1968 taken
directly from Reference 17. Minimum estimate of 281 Mg/yr
(310 T/yr) is an update of that figure for 1972 taken directly
from Reference 110. No quantitative information is avail-
able on the quantity of coal burning, but an inventory of
the volume of burning coal refuse banks was made in 1968
and the results are reported in Reference 60. Because fires
46
-------
can ignite, smolder, or burst into flames naturally and be-
cause the visible burning area may be a poor indication of
the amount of coal burning, ^the best estimate given is
very rough. It assumes the 1968 emissions reported in Ref-
erence 17 and that the emissions are uniformly emitted by
the 292 banks surveyed. Since 23 banks in Pennsylvania have
been extinguished, an updated estimate is 284 Mg/yr (313 T/yr).
This estimate is presumably conservative because the burning
banks emitting the most smoke and nearest to population centers
have the highest priority for extinguishment. Also, burning
active coal refuse piles are generally in some degree of com-
pliance with state air pollution regulations.
p/ Fuel consumption is for total on highway use of diesel fuel.
q/ U.S. Census estimated resident population of the U.S. in
October, 1977 from Reference 68.
r/ Assumed that all the 1975 gasoline consumption for 1975 used
by passenger cars and motorcycles from Reference 6 9 with some
relative shares of gasoline consumption as their relative
shares of total 19 75 motor vehicle fuel consumption from
Reference 121 and that all motorcycles have the emission
characteristics of two-stroke engines (a conservative assump-
tion, as four-stroke engines which comprise approximately
6 8 percent of the motorcycle sales market have less incom-
plete combustion).
47
-------
ESTIMATES OF 198b TXAL : a:
ions nr?:jRcr. ;7PE
Source Annual Production
Est/J.*.;
Minimum,
~ ilaP ¦»
¦ic Tons/V^arj
Maximum
Iru r.nr.c:cU a*.o»
K-"?]«»rences for
I'roGuction Data
Rom,
Coal-fired Power Plants
6.9 x 1011 Kg
0. 76
1.9
1.2
26,94
a,!
Cool-fired Industrial Boilers
1.1 * 1011 kg
0.085
34
0.1
26,94
4,1
Coal-fired Residential Furnaces
1.4 * 109 kg
0.096
740
26
26,94
a,l
otter Solid Fuel Burning Sources I
• DoOSStiC fltOVSS
Unknown
• Residential fireplaces
4.5 x 1010 kg
54
110
77
32,66
c
Oil-fired intenediate Boiltrai
• Industrial Boilers
1.6 x 1012 kg
1.1
3.2
l.B
26,94
b
• Camerclal/Inetitutional Boilers
£.1 x 1011 kg
24
94,123
d
Oil-fired Residential Boilers
5.4 * 10U kg
1.1
2.4
1.2
123
c
Gae-fired Intermediate Boilerai
s Industrial Boilers
1.8 x 10U n5
0.025
26,94
b
* Cooeercial/Institutional Boilers
9.4 x 1010 a3
0.71
26,94
b
Gas-fired Residential Furnaces
2.2 x 1011 m3
0.12
1.5
0.43
26,94
b
Petroleum Catalytic Cracking:
• fluid Catalytic Cracking
Unknown
<1.0
124
f
• T^ertaofor Catalytic Cracking
Unknown
<1.0
' 124
f
« Houdriflow Catalytic Cracking
Unknown
<1.0
124
f
Coke Production
6.4 x 1010 kg
0.0092
56
21
96
g
Asphalt Productioni
* Saturators
5.7 x 109
0.00046
0.23
0.0057
104,105
h
• Air Blowing
5.7 x 109 kg
0.0019
0.033
0.0057
104,105
h
• Rot Road Ml*
2 x 1010 kg
0.014
0.0012
104,105
1
Other Industrial Processes:
3.7 x 1010 kg
• Iron 6 Steel Sintering
0.022
41
0.^3
1
e Chain link Pence Lacquer Coating
Unknown
e Carbon Black Production
1.5 x 109 kg
0.05
"0.11
0.11
105
Inctnaratorat
10
l.B x 10 kg
• Kunicipal
0*0031
0.24
0.027
30
j
• Ccooerclal
B.2 x 109 kg
0.90
4.7
2.1
15
k
Open Burnings
• Municipal Refuse
°0
¦0
1
e Auto Components
"0
"0
1
• Grass Clippings, Leaves, Branches
a0
"0
1
• Leaf Burning
®0
"0
1
Agricultural fi Forest Firesi
• Bagasse Boilers
2.3 x 109 kg
0.0062
0.0062
94,125
m
a Forest Fires
2 X 1010 Sl?
10
130
17,19,111,115
n
Burning Coal Refuse Banks
Unknown
310
17,110
0
Mobile Sourcest
2.6 x 1011 1
• Automobiles (gasoline)
0.13
0.21
0.21
69
P
e Autoeobiles (dleeel)
1.6 x 1010 1
2.7
13
5.4
69
a Trucks (diesel)
5.8 x 1010 1
0.13
10
0.21
103
S
• Rubber Tire Mar
2.33 x 108 pop'n
0
12
131
X
a Motorcycles
1.94 x 109 1
5.6
69,121
9
48
-------
TABLE TtI-3 ft£HAItX8
«. Afttuaad 1974 avaraqa haat content of 6.09 « 10* cal/fcq (31.9 * 10* Otu/ton) of CO*I fro* Raferenca 94.
b. e«tluc*d 1905 fual conniption calculated fro* the (mm yaar fual consu*ption (qlvan in Table 111-2) which was dari*»d fro*
data reported in Bafaranca 94. MxiiNd *noainaL" (no ehanqa in policy) growth rataa d*w»l«p»d uaing th<» PIRS cowput.tr *odal.2*^
c. Eatlawted 1985 residential fireplace w>od consuapcion ualnq tha estltwta for 1975 (qivan in Table III-3) H#v*lopad in Raforanc*
86 and reportod in Rafaranca .'2. Aasuaad that consumption would increase at ona-half tha rata of population Incraaae eatliMt^d
in Rt(f«nc* 131 or 4.5 piinnC fro* 1971 to 1909.
d. latitat* of conauaption In barrels taken directly fro* the NOPrs study. Converted Co kiloqraaw aanuatnq an avaraoe danalty
of 7.J4 kq/1 (calculated fro* the 1973 aii of residual and distillate oil* qlvan in ftafarane* 94). If the base yaar conai**t.lon
fro* Reference 94 end the "initiative" and "nominal" qrowth rata* fro* the PI8S oo*put»r ¦o'tel ara asaimd. tha int»r*awllato
estimates of 1981 B*P scissions ara ia and *0 *9/yr, respectively.
a. Consumption iitlutil for 10S5 in bornli CaMn directly fro* R*f*r«nci 113 ind mnvtrtrd to kltmrna :rapol4t 103.
Aaawlnq that currant tranrta contlniM. thi* ahould ba a fair approaimtion aa currant dioMl cnnniMption in autoanhil** in nvqliqihl*.
r. U.S. population in 1985 «sti«aead as 237.9 at 11Ion by tha U.S. Bureau of Canaus.13"
a. rtotoreyele fuel oona\Mptif>n proieeCed to ra*aln conatant. evan though auco*nOil<» <}*eollrtw conauHption in prolactad o
t>*cause Motorcycla fual efficiency ia alraady niqh and Miiaa travailed will probaMy r^Mln n»arty constant.
49
-------
sugar cane refuse fires, and incineration of used lubricating
oils. For a lacquer coating operation in which chain link fence
was being put through the lacquer bath, a single emission test
gave the relatively high emission factor of 470 mg BaP/Mg of
7 6/
fence. 7 No process or production information could be found
for this industry. POM concentrations in emissions from meat
cooking have been measured to range from 1.6 x 10 ^ to 1.1 x
-9 3 134/
10 g/m . ' A heavy duty diesel used in underground mining
3 13 5/
has been found to emit between 0.1 and 10 yg BaP/m of exhaust.
Burning tower experiments were used to assess the emissions from
sugar cane refuse fires. The mass of BaP per mass of particu-
lates measured was 73.1 + 61.1 yg/g for whole cane and 79.0 +
22/
50.5 yg/g for leaf trash. Mass concentrations of BaP in used
57 /
oils ranging from <1.0 to 30 yg/g have been reported. These
sources are not discussed, though it is recommended that further
testing be conducted.
Several other processes are potentially significant emit-
ters of POM. These processes include industrial internal combustion
and diesel engines, agricultural burning, aircraft, gasoline-
powered lawn mowers, motorboats, and misting and aerosol forma-
tion from lubricants. Only the known source types of POM for
which sufficient information was available are discussed in the
following sections. For each source type, the process, emission
sources, emission controls, location and capacity, emission
estimates, and future trends of the sources are briefly out-
lined.
B. Coal-Fired Power Plants
1. Process
Large coal-fired power plants burn crushed or pulverized
coal to generate steam for turbine-generated electric power.
50
-------
The fuel and a stream of air which has been preheated are di-
rected to a furnace or a series of burners where combustion
occurs. The burners may be fired vertically, horizontally, in
opposition, or tangentially; cyclone firing is another possi-
bility. Because the process is not carried out under perfect
conditions, incomplete combustion usually results and causes
pollutants to be emitted from the process. The heat from the
combustion chamber heats water which is contained in a series of
pipes or a boiler so that steam is generated. The steam is then
used to operate a turbine which, in turn, operates a generator
which produces electric power.
2. Emission Sources
Large coal-fired boilers generate POM due to incomplete
combustion of hydrocarbons. Burner configuration, firing meth-
od, and other unit specifications affect the quantity of POM
emissions. In addition, the maintenance and operating condi-
tions of the specific unit affect the completeness of combustion
and thus the amount of POM's generated.
3. Emission Controls
Due to air quality regulations, approximately 97 percent of
coal-fired power plants employ one of the following pollution
control systems: cyclones, scrubbers; electrostatic precipita-
tors; fabric filters; or a combination of these. These tech-
niques are not equally useful in control of POM emissions,
however, since POM preferentially condenses onto small particu-
late matter (because of its larger surface area to volume ra-
tio) . Thus, those control techniques which efficiently control
fine particles will generally also control particulate POM's.
Therefore, cyclones, which are relatively inefficient collectors
of particles smaller than 10 ym, are ineffective except as a
51
-------
precleaner for the more efficient devices—high energy scrub-
bers, electrostatic precipitators, or fabric filters. High
energy scrubbers, such as venturi scrubbers, can be very effec-
tive in removing fine particles while lower energy scrubbers
with their longer residence times can be used first to condense
gaseous POM. Electrostatic precipitators can also achieve high
efficiency particulate removal. Although dry precipitators
cannot always handle sooty or tarry particles, wet precipitators
are generally effective. Fabric filters are the most efficient
collector of fine particles. The fabric pores, however, may be-
come blocked and uncleanable when filtering tarry particles such
as those generated by the combustion of oil. Also, fabric
filter applications are limited by the temperature the fabric
can withstand (the current limit is 290°C for glass fibers).
The particulate POM collection efficiency of any control device
depends on the particle size distribution(s) of the particulate
POM and/or the particles upon which the POM is adsorbed. It has
been shown, both theoretically and in practice, that most POM's
exist as vapors at the stack gas conditions of a typical coal-
fired power plant (^150°C) . ^ ^ Therefore, dust collec-
tors which usually are operated with gas temperatures higher
than the condensation points for most POM's (e.g., fabric fil-
ters, electrostatic precipitators, or cyclones) probably do not
collect much of the total POM present, as most species would
exist as vapors rather than collectable particles. However, air
pollution control systems, such as scrubbers, which condense the
vapors and collect the particles formed should be much more
effective in controlling total emissions of POM.
4. Location and Capacity
Coal-fired power plants are located throughout the U.S.;
however, they are concentrated in areas near coal supplies.
52
-------
Appendix A shows the regional breakdown of coal consumption by
electric utilities. In 1975, there were 381 coal-fired plants
of greater than 25MW(e) with a total generating capacity of
approximately 210,000 MW(e). About 34 percent of these plants
and 32 percent of the generating capacity are located in the
East North Central Region (Illinois, Indiana, Michigan, Ohio,
and Wisconsin). The West North Central, South Altantic, Middle
Atlantic, and East South Central Regions had 76, 61, 42, and 38
122/
coal-fired plants, respectively. '
5. Emission Estimates
39/ 94/
Studies by Hangebra:uck, et al. ' and Suprenant, et al. 7
were used to derive emission factors for the various types of
39 94/
coal-fired power plants. ' Numerous other studies were
17 21 72 110 111/
consulted. ' ' ' ' Their emissions estimates, however,
39/
were all derived from AP-33 by Hangebrauck, et al. The
minimum, maximum, and intermediate estimate emission factors for
the various types of boilers tests and the other processes for
which emission factors were estimated are given in Table III-l
(P. )'.
Data for coal-fired power plants were collected by Hange-
39/
brauck, et al. by direct sampling of stationary sources. The
samples were analyzed using benzene extraction, column chroma-
tography, and ultraviolet visible spectrophotometry. The inter-
mediate estimate BaP emission factors ranged from 0.37 Mg/kg of
coal for a spreader stoker with travelling grate firing crushed
coal to 3.7 yg/kg for a tangentially-fired dry bottom boiler
firing pulverized coal. The boiler population weighted average
for all boiler types is 1.6 yg 3aP/kg of coal. The weighted
average emission factor was weighted for boiler population by
tons of coal burned in the boiler type as given by Suprenant,
94/
et al. 'from Federal Power Commission data for 1972.
5 3
-------
DRAFT
J35 J0T QUOTE OR CITE
It was assumed that dry opposed-firing had the emission
characteristics of the dry vertical-firing tests, that wet-
bottom opposed-firing was representative of all wet-bottom
pulverized coal-firing (6230; 16,700; and 12,300 x 10^ kilograms
3
(6870; 18,420; and 13,520 x 10 tons) of coal burned by opposed,
front, and tangential-firing, respectively), that the wet-bottom
cyclone tests were representative of all cyclone-firing and that
the spreader stoker tests were representative of all stokers.
The breakdown of coal burned in millions of kilograms (thousands
of tons) is 99,600 (109,840) tangential-firing; 55,700 (61,450)
front-firing; and 32,500 (35,850) opposed-firing, for a total of
188,000 (207,140) pulverized coal, dry bottom; 35,200 (38,810)
pulverized coal, wet bottom; 35,500 (39,090) cyclone; and 3,200
(3,500) stoker. Using this weighted average emission factor and
a 1974 consumption figure of 2.85 x 10"^ kg of coal developed
94/
from FPC data, the total estimated BaP emissions from coal-
fired power plants are less than one metric ton per year.'
6. Future Trends
If plants currently under construction and scheduled to be
built by 1985 become operative, there will be more than 600
coal-fired power plants. The generating capacity of these
plants will be approximately 330,000 MW(e). Of the 250 new
coal-fired plants projected to come on stream by 1985, 61 are in
the West South Central Region, 4 8 are in the East North Central
Region, and 44 are in the Mountain Region.
Estimates of coal use in the future have been made by ERDA
(DOE) using the PIES computer model.Estimates were derived
based on the existence of a National Energy Plan including coal
use incentives (Initiative Case) and non-existence of an Energy
Plan (Nominal Case), as follows:
54
-------
Annual Percentage Increase (1975-1985)
Fuel Type 1985 Nominal 1985 Initiative
Coal 4.68 4.87
Oil 475 0.00
Gas -6.33 -10.77
These percentage increases will lead to the following fuel use
by 1985:
Annual Fuel Use (10^ Btu) in Utilities
.Fuel Type 1985 Nominal 1985 Initiative
Coal 6.9xlOUkg (16.6) 7.0xl0i;Lkg (16.9)
Oil 9. 3xl0i;Lkg (4.6) 5.9xl0i;Lkg (2.9)
Gas 5.0xl010m3 (1.8) 2.8xl010m3 (1.0)
POM emissions will tend to increase accordingly unless adequate
POM emission controls are universally applied. Assuming un-
changed emission characteristics, the "nominal" fuel use, and
6.08 x 10^ cal/kg (21.9 x 106 Btu/T), the estimate of total BaP
emissions increases to 1.1 Mg/yr in 1985.
C. Intermediate-Size Boilers
1. Process
In a boiler, heat from combustion of fuel is transferred to
water to produce high-pressure, high-temperature steam. The
steam may be used to drive a turbine and thus, produce mechani-
cal energy or used directly in the industrial process. Inter-
mediate-size boilers are utilized for industrial, commercial,
and institutional processes.
The intermediate coal-fired boilers are fired by pulverized
coal, chain grates, spreader or underfeed stokers, or cyclones.
55
-------
Oil and gas are both blown with combustion air into the combus-
tion chamber through orifices.
2. Emission Sources
Incomplete combustion of fossil fuel in a boiler generates
polycyclic organic matter. Incomplete burning results from ir-
regular heating, insufficient air-fuel mixtures, and the limited
transport of oxygen and heat to the material in large fuel par-
ticles.
POM emissions from gas- and oil-fired units generally tend
to be lower than from coal-fired.units because of the smaller
fuel particle size and better mixing. However, the emissions
from the less efficient {usually smaller) types of oil- and gas-
fired units are higher than those of the coal-fired units which
are efficiently run.
3. Emission Controls
Two particulate control methods are common for intermedi-
ate-size boilers: multiple cyclones and electrostatic precipi-
tators (ESP's)(or a combination of the two). Cyclones are
inefficient in collection of very small particles and, thus, are
not adequate for POM control. Wet ESP's are effective in
controlling fine particles. A combination of a wet scrubber
followed by a venturi scriibber would reduce gaseous and particu-
late POM. Many intermediate-size boilers currently have little
or no control.
Good design, good operating and maintenance practice, and
process modification (higher temperatures and excess air) are
additional POM emission control mechanisms. Some combustion
additives have also been shown to be effective in reducing
emission of POM from boilers burning coal and oil."^'^//
56
-------
4. Location and Capacity
Intermediate-size boilers are dispersed throughout the
country. Their spatial distribution generally follows that of
population and industry. Industrial boilers are used to produce
process steam, heat, or electricity and thus are used by most
industries. Institutional/commercial boilers are primarily used
for heating in hospitals, schools, offices, stores, and apart-
ment buildings. The type of fuel used varies geographically
according to the price and assurance of supply of the various
fuel types.
5. Emission Estimates
Emission factors for various intermediate size boiler types
were developed from the limited stack sampling results reported
3 9/
in AP-33, Sources of Polynuclear Hydrocarbons in the Atmosphere
by Hangebrauck, et al. The ranges of BaP emission factors were
0.77 to 310 yg/kg of coal, 0.66 to 40 yg/kg of oil, and from
3
less than 0.14 to 7.6 yg/m of gas. The intermediate estimate
BaP emission factors were 0.93 yg/kg of coal for the national
average boiler population and 1.1 yg/kg of oil for firing by
steam atomizing burners. For single test results, the BaP
emission factors were 4 0 yg/kg for low-pressure air-atomized oil
and less than 0.14 and 7.6 yg/m for a process heat and a hospi-
tal heat premix gas burner. These and other emission factors
are shown in Table III-l (P. ) .
A breakdown of the population of boiler types derived from
data in the 197 6 EPA report, Preliminary Emissions Assessment
94/
of Conventional Stationary Combustion Systems was used to
derive an emission factor for all coal-fired intermediate
boilers. The various estimates were weighted by boiler popula-
tion by the trillions of Btus of bituminous coal consumed by
57
-------
94/
boiler type from FPC data for 1973. ' It was assumed that
pulverized coal wet bottom and cyclone boilers had the emission
characteristics of the weighted average of industrial boilers;
that the pulverized coal dry bottom test was representative of
all pulverized coal, dry bottom firing; that the spreader
stoker with reinjection test was representative of all spreader
stokers; that the chain grate stoker test was representative of
all overfeed stokers; and that the geometric average of the two
underfeed stoker tests was representative of all underfeed
stokers. The breakdown by millions of kilograms (trillions of
Btus) consumed is 26 (650) in pulverized dry bottom, 5.3 (130)
in pulverized wet bottom, 2 (40) in cyclone, 1 (30) in overfeed
stoker, 18 (450) in spreader stoker, and 0.8 (20) in underfeed
94/
stoker coal-fired boilers. Using 197 3 consumption data, ' the
best estimates of BaP emissions from intermediate boilers are
less than one metric ton per year for all fuels and uses except
oil-fired commercial/institutional boilers, which are estimated
to generate 19 Mg/yr.
6. Future Trends
Estimates of future power consumption by fuel type were
made by ERDA (DOE) in October, 1977, using the PIES computer
2 6/
model. ' The estimates for industrial fuel use are as follows:
Fuel Type
Annual Increase 1975-1985 (%)
1985 Nominal 1985 Initiative
Coal
Oil
Gas
5.07
13.54
1.32
16.41
0.62
0.87
59
-------
The "initiative" figure is based on the existence of a
National Energy Plan including federal energy conservation and
coal-use incentives. These estimates suggest large increases in
atmospheric POM unless adequate control of vaporous and particu-
late POM' is utilized. Assuming the "nominal" growth rate from
94/
base year consumption figures ' given in Table III-2 (p. ),
the best estimates of BaP emissions increase in all cases.
However, only the oil-fired boiler estimates of 1.8 Mg/yr for
industrial boilers and 88 Mg/yr for commercial/institutional
boilers are greater than one metric ton per year. Using the
123/
"initiative" or MOPPS ' projections of 1985 institutional/
commercial oil consumption, that estimate only slightly in-
creases with time to 18 or 24 Mg/yr, respectively.
D. Residential Furnaces
1. Process
Coal-, gas-, and oil-fired furnaces are used to heat most
of the nation's homes. The fuel is combusted to heat circulat-
ing water or air. Small coal-fired furnaces may be of the
underfeed or hand-stoked variety. Oil-fired units atomize the
fuel by utilizing pressurization or vaporization. In gas
furnaces, gas and air are premixed and fed to gas burners.
2. Emission Sources
Home furnaces are a major source of polycyclic organic
matter due to inefficient combustion of hydrocarbon fuels.
Hand-stoked coal furnaces emit very high quantities of POM in
exhaust gases and through leaks in the unit.
59
-------
Gas furnaces generally emit the least POM per heat input
due to the good feeding characteristics and low particulate con-
tent of the fuels.
3. Emission Controls
Control of emissions from home furnaces is not widely prac-
ticed because of their additional maintenance requirements. Ef-
ficient furnace design, good maintenance, and clean fuels can
reduce POM formation.
Low air pollution control agencies are attempting to elim-
inate the use of hand-stoked coal furnaces--the major source of
POM from coal combustion.
4. Location and Capacity
The 1974 Housing Inventory reports the following distribu-
tion of home heating methods. -*-^6/
Utility gas 39,471,000 units
Bottled, tank, or LP gas 4,143,000
Fuel oil, kerosine, etc. 16,835,000
Electricity 8,407,000
Coal or coke 741,000
Wood 658,000
Other fuel 90,000
No heating equipment 484,000
Fuel consumption figures were not available for this study.
5. Emission Estimates
Emission factors for residential furnaces were derived
39/
primarily from Hangebrauk, etal. ' Several additional sources
60
-------
1*. J r 1 r- J J 4. 17, 20, 72,110,111/
were consulted for coal-fired furnace emissions data.
The BaP emission factors for underfeed stokers range from 115
yg/kg to 2.6 mg/kg with an EPA best estimate of 800 yg/kg of .
coal. For hand-stoked furnaces, the intermediate estimate was
42 mg/kg, while the range was 13 to 100 mg/kg of coal. For oil-
fired residential furnaces, the three available test results
gave a BaP emission factor of less than 2.2 yg/kg for pressure-
atomized oil furnaces and less than 4.4 yg/kg for a vaporized
oil furnace. The best estimate was 2.2 yg/kg of oil. Gas-fired
premix burners were calculated to have an emission factor of
3
0.91 yg/m for the two tests of burners larger than 180,000
3
Btu/hr and 10 yg/m for the one test of a 25,000 Btu/hr space
heater.
Total annual POM emissions were estimated using these
emission factors and fuel consumption taken directly or develop-
94/
ed from Suprenant, et al., as follows:
o Fuel consumption (for oil furnaces) in
kilograms, was calculated from the fuel
consumption in Btu's given in Suprenant,
94/ 9
et al. A heat content of 9.72x10
3
cal/m (146,000 Btu/gal) and a density
of 7,830 kg/m3 (#5 fuel oil) for residual
9
oil and a Btu content of 9.32x10 cal/
m3 (140,000 Btu/gal) and a density of
7,210 kg/m3 (#2 fuel oil) for distil-
late oil were assumed.
o Fuel consumption (for gas heaters) in
kilograms was calculated from the fuel
consumption in Btu's given in Supre-
94/
nant, et al. assuming a heat content
of 9.095xl06 cal/m3 (1,022 Btu/ft3),
61
-------
and a natural gas composition of 94.2
percent methane, 3.6 percent ethane,
and 2.2 percent nitrogen for a molecu-
lar weight of 16.8 g/g-mole, and a
perfect gas at standard conditions
(0°C, 1 atm).
The EEA best estimates of annual BaP emissions from residential1
furnaces were thus 26, 0.98, and 0.43 Mg for coal, oil, and gas,
respectively.
6. Future Trends
Natural gas shortages have greatly increased prices of gas
for heating homes. Natural gas usage in new homes will likely
be strictly reduced. A decrease in total gas consumption for
residential heating is expected as older homes replace gas
units.
A recent EPA study concluded that current economic and en-
vironmental factors associated with coal stoker furnaces are
unfavorable for increased coal usage in residential applica-
35/
tions. Thus, coal consumption BaP emissions are expected to
9
remain fairly constant through 1985 at 7.4 x 10 kg and 26
Mg/yr. Although "smokeless" coals are technically feasible,
35/
they are currently neither available nor marketable.
Heating with oil is projected to increase due to population
increases. Using the MOPPS estimate of 1985 residential con-
123/
sumption, the estimated emissions are likely to increase to
1.2 Mg BaP/yr.
Electric heating will likely increase by one or two per-
cent. Gas users are likely to replace their gas heaters with
electric or oil heating systems. Therefore, it was estimated
62
-------
that the residential gas consumption would remain constant at
3 3
about 2.2 x 10 m . Thus, the estimate of BaP emissions in 1985
is still less than one metric ton.
E. Other Solid Fuel Burning Sources
1. Process
Coal and wood fueled domestic stoves and wood-burning fire-
places as single units are sometimes used to produce heat. Some
varieties of wood stoves used for home heating are more effi-
cient due to tightly sealed chambers, carefully controlled air
intake, and exhaust systems.
2. Emission Sources
The incomplete combustion of wood and coal is due to slow,
low-temperature burning with insufficient air at the burning
surface and the high moisture content of most home fuel sup-
plies. Products of such combustion normally contain polycyclic
aromatic hydrocarbons and are released directly to the atmos-
phere .
3. Emission Controls
Restrictions on fireplace and stove use are one possible,
but impractical, control method. Chimney filter systems are,
perhaps, more feasible. Fireplace and stove design specifica-
tions offer another means of controlling POM emissions from
these sources.
4. Location and Capacity
The 1974 Housing Inventory106^ reports that 36,000 housing
units in the United States used coal or coke and 206,000 used
63
-------
wood as a cooking fuel. The 2,069,000 residential fireplaces,
heating stoves, and portable heaters are reported in the same
document.
Fuel consumption in stoves is unknown. Fireplace consump-
tion in 1975 was estimated at 17 million cords^^ or 4.31 x 10^
3
kg assuming a specific gravity, of 0.7 g/cm . Other estimates of
32 94/
wood fuel usage based on Btu consumption estimates ' ' range
from 1.22 to 7.08 x 10"^ kg of wood, assuming a heat content of
2.8 x 106 cal/kg (107 Btu/ton).
5. Emission Estimates
Emission factors for fireplaces were obtained from an EPA
89/
report by Snowden, et al. ' Fireplace emissions were sampled
using a Tenax adsorber following a glass fiber filter. Analysis
was by gas chromatography and mass spectrophotometry. For
various woods, the BaP emission factor for stable burning ranged
from less than 1.2 to 2.5 mg/kg of wood with an intermediate es-
timate of 1.7 mg/kg.
Emissions from domestic stoves burning solid fuels were
6 /
reported by Beine ' in 1970. Stove emissions were analyzed by
isolating benzo(a)pyrene by pyrolysis of styrene containing
tars. Separation was by thin-layer chromatography, analysis by
ultraviolet spectrometry and gas chromatography. BaP emission
factors for the various fuels, which were mostly coal-derived,
ranged over three orders of magnitude from 7 00 ug/kg to 379
mg/kg. The best estimate is 5 mg BaP/kg of fuel.
Total BaP emissions were estimated for fireplaces using the
estimate of 4.31 x 1010 kg of wood consumed in 1975 and the
emission factors shown in Table III-l (P. ). The intermediate
estimate of total annual emissions of BaP from residential fire-
places is 73 Mg/yr. No estimate of annual emissions could be
64
-------
made for domestic stoves burning solid fuels, as no consumption
data could be found.
6. Future Trends
A recent issue of Newsweek indicates that a growing number
of hpmes are heated, at least partially, by wood burning stoves.
The quantity and character of POM emissions generated by the
newer designs of wood burning stoves are uncertain at present.
Most of the new designs presumably,have fewer leaks so that com-
bustion may be improved and POM emission reduced; however, com-
bustion temperatures may be lower and underfire air greater so
that more POM may be generated and more particles, possibly
with adsorbed POM, emitted. The amount of POM emitted will vary
with the particular design and its operation, while emissions
will also vary with the cyclical nature of the process. POM
emissions would be expected to be high when colder fuel is added
and volatiles are distilled off, lower when the flames produce a
hot fire, higher when combustion is cooler during smoldering and
lower when carbon is the major component in the remaining fuel.
The use of wood in fireplaces was assumed to increase at
one-half the rate of increase in population or 4.5 percent from
1975 to 1985. Thus, the intermediate estimate of 1985 BaP
emissions is 77 Mg/yr. This estimate is very rough; however, no
projections are available for the number of single family homes,
or other measures which might be better indicators of wood con-
sumption in residential fireplaces were readily available for
this study.
F. Future Energy Sources
1. Process
The sources of energy which may be significantly utilized
in the foreseeable future include solar, nuclear, and fossil.
65
-------
POM's are likely to be generated in some amount by any process
which involves the heating of hydrocarbons. Therefore, future
energy processes such as coal conversion, fluidized bed combus-
tion (FBC), and magneto-hydrodynamics (MHD), which use fossil
fuels will generate, if not emit, some POM. Coal gasification
and liquefaction are especially likely to emit POM's since the
processes used generally are based upon incomplete combustion.
This is so because the desired gaseous or liquid fuel product
contains large quantities of combustible matter including POM.
Utilization of solar or nuclear energy is unlikely to generate
POM emissions. POM's could be emitted, however, during the pre-
paration of hydrocarbonaceous materials for energy utilization
equipment, e.g., photovoltaic cells, plastic solar panels, or
graphite control rods for nuclear reactors.
2. Emission Source
Most of the POM's are generated in the combustion or reac-
tion chamber when the hydrocarbons are heated or combusted.
They are probably not emitted directly from that chamber, how-
ever. POM's generally comprise a large fraction of coal gasi-
fication or liquefaction products. They may be emitted during
collection, treatment, transportation, or utilization of these
fuels. After the combustion of coal-derived fuels or the com-
bustion of coal or other fossil fuels in other advanced pro-
cesses such as FBC or MHD, many of the hydrocarbons which have
not been completely combusted will be emitted from the stack as
POM unless they are removed from the flue gas.
3. Emission Controls
As the processes, pollutant generation, and product or flue
gas stream characteristics of the various future energy sources
66
-------
DRAFT
DO NOT QUOTE OR CITE
are uncertain, the effectiveness of current or future air pollu-
tion control equipment is also uncertain. High temperature and
pressure particulate control equipment is being studied. Since
most POM's are gaseous even at lower stack temperatures, ^ ^
it is unlikely that these devices would collect much POM. The
effectiveness of more conventional air pollution control equip-
ment for POM will depend on the actual gas streams characteris-
tics, the type of control equipment, and the form and amount of
POM in the flue gas. Leakage of POM from liquid or gaseous
coal-derived fuels during their processing or transport may be
able to be reduced by improved valves and gasketing.
4. Location and Capacity
Those sources considered likely to emit POM during the gen-
eration and utilization of energy are currently in the bench,
pilot, or demonstration stages of research and development.
Commercial projects are only in the planning stages and none are
expected to be operational before 1985. The types, locations,
and capacities of future energy sources of POM will depend upon
the economics of the processes in various areas.
5. Emission Estimates
No estimates of emissions from future energy sources of POM
can be made at present. Some studies have included the measure-
ment of POM's in the product, flue gas, or other waste streams
from coal-derived energy processes.^^However, some of
these results are for the smaller and older processes used in
Great Britain,51^ while others are for newer processes on only a
29/
bench or pilot scale or largely measure POM in product
14/
streams. Other studies have merely noted that the product or
waste streams contain large amounts of aromatics or some types
of POM's. The few data that do report the quantities of POM in
67
-------
the gas stream are generally on the basis of the volume of gas and,
thus, are not suitable for the development of emission factors.
Adequate data are not available for the development of POM
emission factors from specific processes, let alone for the
general categories such as coal gasification or liquefaction.
However, since no commercial units exist, current POM emissions
are presumed to be negligible.
6. Future Trends
The future emissions of POM from energy processes which are
not currently in commercial use cannot be estimated. Represen-
tative emission factors cannot be estimated at this time. Also,
the types, locations, and capacities of these POM sources are
subject to change, as they are only in the planning stages. It
is unlikely that any commercial plants of these energy processes
will be operational in 1985 so it is presumed that POM emissions
will be negligible.
G. Petroleum Catalytic Cracking
1. Process
The catalytic cracking process is used to upgrade heavy
petroleum fractions by breaking up long-chain hydrocarbons to
produce high octane gasoline and distillate fuels.
Several types of cracking units are used: fluid catalytic
crackers (FCC), thermofor catalytic crackers (TCC) with airlift
or bucket lift catalyst carriers, and Houdriflow catalytic
crackers (HCC). The basic process involves a silica-alumina
catalyst and gas-oil mixture. The mixture is cracked by being
heated to 480°C and then fractionated. The spent catalyst,
laden with coke, is regenerated by burning off the coke at
68
-------
about 54 0°C. The regenerator exhaust gases are then exited
directly to the atmosphere or passed through a carbon monoxide
waste heat boiler.
39/
2. Emission Sources '
The exhaust gases from catalyst regeneration kilns are high
in unburned hydrocarbons and carbon monoxide from the burned
coke. Emissions of benzo(a)pyrene, pyrene, and other POM's tend
to be very high.
TCC bucket elevator and FCC units appear to emit signifi-
cantly smaller uncontrolled quantities of POM per quantity of
throughput than the HCC and TCC air-lift units based on rather
limited emissions tests.
3. Emission Controls
Carbon monoxide waste heat boilers can be used to effect
more complete combustion of catalyst regenerator kiln exhaust
gases. The boilers utilize auxiliary fuels or a catalyst and
have been found to be more than 99 percent efficient in removal
of polynuclear aromatic hydrocarbons. Plume burners are inef-
ficient POM controls for catalytic cracking units.
4. Location and Capacity
The locations and capacities of U.S. petrolteum refineries
shown in Appendix B were obtained from the Worldwide Directory
71/
Oil and Gas Processing 1977-1978. Refining capacity is
centered in Texas, Louisiana, and California, but most states
have at least one refinery. In 1977, total U.S. catalytic
4 3
cracking capacity was reported to be 754,000 m (4,739,704
barrels) of fresh feed per stream day.
69
-------
5. Emission Estimates
Emission factors for POM emissions from catalytic cracking
operations were derived from data in the U.S. Public Health
Service report Sources of Polynuclear Hydrocarbons in the Atmo-
39/
sphere and the results of additional POM analyses for the
same tests reported in the NAPCA draft report Control Techniques
32/
for Polycyclic Organic Matter Emissions ' from a study by
Sawicki, et al. in the International Journal of Air and Water
82/
Pollution. ' These factors are shown in Table III-l (P. ).
The range of individual test results for FCC, TCC, and HCC is
3 3
from 27 yg/m for FCC to 1.4 g/m for HCC for uncontrolled BaP
3
emissions and from below detectable for FCC to 280 yg/m for
HCC for controlled BaP emissions. The intermediate estimates of
3
uncontrolled BaP emission factors are 280 yg/m for FCC, 470
3 3
mg/m for TCC air-lift, 78 yg/m for TCC bucket elevator, and
3
1.4 g/m for HCC. The best estimates of controlled BaP emission
3 3
factors are 38 yg/m for FCC and 280 yg/m for HCC.
The emission factors for the various cracking processes
were weighted for cracking population as broken down in the Oil
and Gas Journal's Worldwide Directory: Refining and Gas Pro-
cessing] 1977-78. The breakdown in cubic meters (barrels) of
fresh feed plus recycle per stream day is 816,000 (5,133,425;
94.2 percent) for fluid (FCC), 46,600 (292,900) (5.4 percent)
for Thermofor (TCC), and 3,420 (21,500) (0.4 percent) for
Houdriflow (HCC) catalytic cracking units. The BaP emission
3 3
factor was 3 70 yg/m for units without control and 3 9 yg/m for
units with carbon monoxide waste heat boilers.
Estimates of total annual benzo(a)pyrene emissions for
petroleum catalytic cracking are shown in Table III-2 (P. ). All
estimates of controlled or uncontrolled BaP emissions are signi-
ficantly less than one metric ton per year.
70
-------
6. Future Trends
The U.S. demand for petroleum continues to increase each
year. A 14 percent rise was reported from 1976 to 1977.
Figures are from the American Petroleum Institute.
Imports of refined products are expected to continue to de-
crease because of government policy, although crude imports may
rise. U.S. refining operations will likely increase as a re-
sult .
There is no consensus on the future of oil refining. Exxon
U.S.A., for example, projects a growth rate of five percent an-
nually through 1980? Shell Oil Co. expects only a four percent
growth rate.7^
Conversion of many fuel burners to coal will likely have
some impact on refining output. Gasoline demand, however,
continues to rise at over three percent annually. Reduction of
the lead content of gasoline to meet environmental gbals will
continue to increase demand for catalytic cracking operations.
Catalytic cracking can serve to replace lead as an octane en-
richer. The Oil and Gas Journal70^ reports that cracking oper-
ations have shown large gains in recent years:
Capacity*
% Gain
% Recycle
January
1974
734 (4,618.6)
+2.4
18.7
January
1975
744 (4,677.4)
+1.3
16.6
January
1976
754 (4,744.9)
+1.4
16.4
January
1977
784 (4,929.8)
+ 4.0
14.6
*
In 1,000 cubic meters per stream day (1,000 barrels per stream
day) .
* *
As percent of gross (fresh feed plus recycle).
71
-------
Potential POM emissions from catalytic cracking and regen-
eration are expected to rise at over three percent per year.
The duration of this rise is unknown. The American Petroleum
124/
Institute ' stated that catalytic cracking capacity in 1985
cannot be projected. Even if capacity increased by an order of
magnitude, which is extremely unlikely, even uncontrolled emis-
sions would still be less than one metric ton of BaP per year.
H. Coke Production
1. Process
Coke production is an integral part of steel-making. Coke
provides heat and carbon for the smelting and reducing of iron
ore in blast furnaces.
Coke is manufactured from coal by the by-product method in
enclosed slot-type ovens. The method is termed by-product be-
cause the by-products, such as coke oven gas and benzene, are
recovered. The processes involved in coke production and use
are the charging of coke ovens, coking, pushing and quenching,
combustion, and tar handling.
129/
2. Emission Sources '
Coke oven operations are major sources of POM emissions.
Although they are usually contained, exhaust gases have a high
POM content. Gas leakage during charging, pushing, or coking is
the primary source of particulate POM emissions from coke pro-
duction. EPA has grouped the emissions from by-product coke
ovens into seven categories based upon their source in the cok-
129/
ing processes. These sources and EPA1s description of them
are listed below:
72
-------
Charging—Periodically, coal is charged
into an empty oven. A charge usually
lasts about three to five minutes and
occurs every 10 to 20 minutes. The
emissions are fugitive and result from
volatilization of the coal as it enters
the red-hot oven. They usually emit
through the oven charging ports or some-
times out of the charge car hoppers.
Topside Leaks—Emissions from these
leaks occur primarily during the
early part of the coking cycle, but
since the cycle is staggered through-
out the ovens in a battery, the emis-
sions are essentially continuous. The
emissions are fugitive and emit from
any of several hundred potential lo-
cations on top of a battery.
Door Leaks—Emissions from these leaks
are similar to those from topside leaks.
They are fugitive and emit from doors
on both ends of each oven.
Pushing—At the end of the coking cycle,
the red-hot coke is pushed out the end
of an oven into a railcar. A push lasts
about 30 to 60 seconds and occurs every
10 to 20 minutes. The emissions are
fugitive and are carried up in a strong
thermal updraft created by the hot coke.
-------
« Quenching—The hot coke is quenched with
water under a large open tower. A quench
lasts about two to five minutes and occurs
every 10 to 2 0 minutes. Even though the
emissions emanate from the top of a tower,
the tower cross section is so large and gas
velocities so low that it is similar to a
fugitive source.
• Battery Stacks—A battery stack is a tall
stack to provide natural draft for com-
bustion of gas that heats the battery.
Emissions get into the stack by leakage
through oven walls into the battery heat-
ing flues.
• By-Product Plant--The by-product plant is
a chemical plant where various by-products
are recovered from the material volatilized
from the coal. It is not known what the
major emission sources in the by-product
plant are, but it is suspected that most,
if not all, are fugitive.
3. Emission Controls
Slot-type coke ovens are normally equipped with a chemical
recovery system, so that polynuclear hydrocarbon emissions result
mainly from gas leakage. For new ovens, door and topside leakage
can be reduced by improved design and construction. Pipeline
charging, contained pushing, and a continuous, contained, and
controlled quench can be used. For existing ovens, maintenance
of ovens or capture and control of emissions can be used to
reduce emissions from leaks, pushing, and quenching. Larry car
74
-------
and coke oven modifications can be used to effect stage charging.
The following discussion of coke oven control techniques is taken
almost entirely from the same EPA memo which described the emis-
129/
sion sources. '
There are four categories of control techniques applied to
coke ovens. They are: (1) containment of emissions in the
process; (2) capture techniques (hoods, etc.); (3) process
changes; and (4) control devices. These techniques are discussed
in general here. Table III-4 shows which techniques are present-
ly considered by EPA as the better options for each source. Some
control techniques, particularly stage charging, are being re-
quired by OSHA and EPA primarily in order to control POM. Other
pollutants, especially the particles onto which POM may be ad-
sorbed, are incidentally controlled by the same techniques.
Containment techniques are those that prevent the escape of
emissions from the coke ovens. They are not 100 percent ef-
fective, and those emissions that escape are fugitive. Because
of the extreme difficulty of mass measurements of these fugitive
emissions, only visible emission measurements have been used to
characterize the performance of containment techniques. Conse-
quently, it is not possible to determine quantitative emission
reductions.
However, it can be argued that a reduction in visible emis-
sions will reduce emissions of all pollutants, including POM.
The containment techniques are designed to prevent any matter,
including gases, from escaping the coke oven. For example, a
principle of stage charging is to maintain a slight negative
pressure just inside the charging ports so that any flow is into
the oven. For oven leaks, the principle is to seal openings
through which emissions escape, thereby preventing the escape of
75
-------
TABLE III-.4
SOURCE—CONTROL TECHNIQUE COMBINATIONS
129/
Source
Charging (wet coal)
Charging (dry coal)
Charging (dry coal preheater)
Topside Leaks
Door Leaks
Pushing
Quenching
Category
Containment
Containment
Control Device
Containment
Containment and/
or Capture and
Control Devices
Capture and
Control Devices
Process Changes
Control Technique
Stage Charging—aspiration in the standpipes draws emissions
into the battery main which is ducted to the by-products plant.
Similar to stage charging, though emissions may be aspirated in-
to a separate main for recovery and recycle of coal fines.
Venturi scrubbers and electrostatic precipitators have been used.
Application of sealing compound to leaks.
Use and maintenance of doors designed to close tightly.
Individual hoods over each door. No clear picture of the types of
control devices that will be used has emerged, but scrubbers and
wet electrostatic precipitators are candidates.
An enclosure of the coke guide and hot coke car or a shed over the
coke side of the battery. A large variety of designs are in use
or planned. Control devices are venturi scrubbers or wet
electrostatic precipitators.
Baffles (or similar techniques) in the quench tower and clean water
for quenching or dry coke quenching. Dry coke quenching involves
several emission sources that will require hooding and control
devices.
Battery Stacks
Containment
and/or Control
Devices
Patching of cracks in oven walls.
Scrubbers, electrostatic precipitators and a pilot baghouse have
been used.
By-Product Plant
Probably Contain- Little is known of applicable control techniques, but preventing
ment leaks, enclosing tanks, etc. will probably be significant factors.
-------
DRAFT
00 NOT QUOTE OR CITE
any matter. In conclusion, any containment technique that ef-
fects a substantial reduction in visible emissions (like those
listed in Table III-4) will effect a substantial reduction in POM
emissions, even though that reduction cannot be quantitied.
Capture techniques include hoods, enclosures of the emission
source, and sheds over the coke side of the battery. All of
these techniques embody a capture efficiency less than 100 per-
cent and use of a control device to collect the captured emis-
sions (the control devices will be discussed later). As with the
containment techniques, measurement of the fugitive emissions
that escape capture is very difficult. However, a few attempts
have been made for systems that capture pushing emissions. The
estimates obtained range from 50 to 90 percent capture. The
better capture systems would be expected to achieve a capture
efficiency near the top of this.range. No similar attempts
have been made for capture systems on other sources. As with the
containment techniques, the argument that a substantial reduction
in visible emissions corresponds to a substantial reduction of
all pollutants is valid. There is no reason to believe that POM,
or any other pollutant, will escape capture more'readily than
visible particles. However, capture of POM is of little value
unless a control device that efficiently collects it is used.
The gas temperatures for capture techniques can vary widely.
They are usually high enough that the POM will be largely gase-
ous. (The sources most likely to have temperatures below the
condensation or adsorption point for most POM's are leaks and
some pushing operations.)
Process changes as a control option apply only to quenching.
Both wet and dry processes are alternatives to quench coke. Dry
quenching is expected to achieve lower emissions. Data are not
available at this time to estimate the emission levels. Dry
77
-------
quenching will require capture systems and control devices for
several emission points. Again, no information is available to
estimate the performance of such systems or even the gas flow
rates, etc., required. Control of emissions from wet quench
towers will rely on such factors as baffle designs and the use of
clean water. Baffle design would not be expected to influence
gaseous emissions. The effect of water quality is uncertain as
polynuclear aromatic hydrocarbons were lower, but total cyclic
organics higher with clean, rather than contaminated, quench
90,130/
water. ' •
4. Location and Capacity
A study, Human Population Exposures to Coke Oven Atmospher-
___
ic Emissions, was recently prepared for EPA by Benjamin A.
Suta of the Stanford Research Institute. Current data on lo-
cation and capacity of coke oven plants are available in this
forthcoming report. A typical coke oven battery has about 58
ovens (range from 20 to 80), produces about 1,400 tons of coke
12 9/
per day, and operates 24 hours per day and 365 days per year.
5. Emissions Estimates
Updated POM emission factors for coke ovens were estimated
by both EEA and EPA. The emission factors developed by EPA in a
recent source assessment of hazardous organic emissions from coke
ovens^^ are given in Table III-l (P. ). The EEA estimates,
which were developed for some coke oven sources, are noted in the
footnotes to Table III-l. For a particular source, both the EEA
and EPA estimates were developed from the same limited data base
and, therefore, only differed with the assumptions made or data
considered. Although all measurements of POM emissions from coke
ovens are questionable, EPA estimated that uncontrolled BaP
78
-------
emissions from coke ovens were at least equal to the estimated
emissions from door leaks, 1.5 g/Mg (0.003 lb/ton) of coal
charged.
No emissions data are available for dry coal charging, top-
side leaks, or by-product plants. The composition of oven emis-
sions from dry coal charging are probably similar to those from
wet coal charging. Data are not available on the POM emissions
from the coal preheater stack or by-product plant. EPA suggested
that the emissions from topside leaks may be significantly less
than, and are no greater than, the emissions from, door leaks.
Total annual coke production was estimated at 5.12 x 1010 kg
118 /
in 1975. It was assumed that 1.45 Mg of coal is required to
produce 1.0 Mg of coke. Therefore, total annual BaP emissions
were estimated to be about 110 metric tons per year.
6. Future Trends
Coke production has been projected to continue its recent
increase at a rate of 4.2 percent annually. An estimated 64
million metric tons could be produced by 1985 considering ex-
pected increases in capacity. An early controlled Larry car
design (the AISI/EPA design) reduced particulate emissions by 84
percent^ and leakage from some doors can be reduced relatively
easily and effectively. Therefore, assuming a best estimate
controlled emission factor of 230 mg BaP/Mg of coal charged and
1.45 Mg of coal per Mg of coke, the 1985 BaP emissions are pro-
jected to be 21 Mg/year. The efficiency of coke oven emission
controls may vary from 50 to 90 percent. It was assumed that a
POM control efficiency of 85 percent will be achieved by 1985.
79
-------
I. Asphalt Production
1. Process
Within the asphalt industry, there are two major segments:
hot-mix asphalt plants and roofing manufacture. Hot-mix asphalt
is a heated mixture of crushed stone aggregate, sand, and asphalt
used primarily for paving roads. The preparation process con-
sists of mixing aggregate (at 120 to 180°C (250 to 350°F))
with raw asphalt (at 135 to 160°C (275 to 325°F)). Hot asphalt
paving mixes are used to line dams, reservoirs, and other im-
poundment structures, as well as to surface roads and airfields.
Asphalt roofing products are prepared by impregnating heavy
paper felt with hot asphalt saturant and then coating the felt .
with a harder grade of asphalt. Preparation of the asphalt
saturant consists of oxidizing the asphalt by bubbling air
through liquid asphalt (at 220 to 290°C (430 to 500°F)) . This
dehydrogenation process is termed air-blowing and reduces the
volatile content of the asphalt and raises its melting point.
Asphalt-saturated felt may be used in rolls, thus requiring no
further preparation, or coated with bituminous material, mica
schist, or rock granules and cut into shingles.
50/
2. Emission Sources '
Major sources of hydrocarbon emissions from hot-mix plants
include the rotary aggregate dryer/heater, fuel burners, and the
truck which transports the plant. The dryer is used to remove
moisture from sand and crushed stone. Dryers are commonly
fueled with No. 2 fuel oil, so combustion-associated pollutants
(e.g., SO and NO ) are generated.
X X
The hot gases from the dryer contain particulates and
moisture from the aggregate. These gases are generated in large
80
-------
quantities and the particles tend to adsorb hydrocarbons from
asphalt volatiles. Nearly all the POM emissions from hot-mix
plants tested have been attributed to combustion gases and not to
volatiles from the mixing chamber.
Air-blowing stills and saturator units are the major sources
34/
of emissions from asphalt roofing manufacture ' . The air-
blowing operation involves heating asphalt to 22 0 to 290°C (43 0
to 500°F) using gas or oil burners. Air is then bubbled through
the hot asphalt for several hours.
Gaseous emissions from the air-blowing operation include
large quantities of alkyl polunuclear hydrocarbons and carbon
monoxide. Aldehydes and hydrogen sulfide also are present.
The asphalt saturator consists of long troughs in which
rolls of felt are impregnated with hot asphalt by spraying, dip-
ping, or a combination of the two. The saturator operates at 2 00
to 230°C (400 to 450°F). Emissions of gaseous and particulate
organic compounds vary according to the thickness of the felt
used and the product type.
Emissions from the saturator consist of combustion-generated
pollutants from heating units, water vapor, condensed asphalt
(hydrocarbon) droplets, and gaseous organic vapors. Polycyclic
aromatic hydrocarbons are present in both gaseous and particulate
form.
34/50/
3. Emission Controls
.Exhaust gases from both the mixer and the rotary dryer of
hot-mix plants normally are passed through a cyclone and a water
spray tower. This combination is an efficient method of POM
removal for these small plants.
81
-------
Existing controls on asphalt air-blowing stills consist al-
most entirely of fume incineration in a process heater or after-
burner. Heat generated in such an afterburner can be used to
preheat asphalt for the blowing or saturation operations. Ano-
ther possible method of emission control is a steam spray-baffle
arrangement. This tends to be less efficient due to the cohesive
characteristics of the particulate emissions.
Control of emissions from saturation units is more difficult
due to the large volumes of exhaust gas. Normally, the entire
saturator is enclosed by a hood which vents gases to a control
device or directly to the atmosphere. The control methods avail-
able for use are low-voltage electrostatic precipitators (ESP's),
ESP plus flue gas scrubber combination, gas scrubber alone,
afterburner, and high energy air filter (HEAF).
Low-voltage ESP's remove about 90 percent of particulate
emissions. However, maintenance of ESP's is difficult because of
the cohesive tar-like characteristics of the particles. The
efficiency of low-energy scrubbers is too low for saturator use,
while the more efficient venturi scrubbers are prohibitively
expensive in most cases.
4. Location and Capacity
Hot-mix asphalt is produced by either a batch or a continu-
ous process. Most plants are small with an average production
rate of 91 to 182 metric tons per hour. In 1973, there were
50/
4,500 asphalt hot-mix plants operating in the United States.
Since paving asphalt must be delivered hot to the job site, many
plants are designed to be moved from site to site. Exact data on
locations and capacity of hot-mix plants were not available for
this study. Total U.S. sales of asphalt products for paving use
are shown in Table III-5.
82
-------
TABLE III-5
SALES OF PETROLEUM ASPHALT FOR CONSUMPTION IN THE UNITED STATES
(Metric Tons)
1972
1973
1974
1975
1976
United States, Total
28,232,431
31,146,402
28,154,502
24 , 943,186
27,242,906
By Principal Use:
Paving Products
22,049,570
24,530,775
22,354,634
19,588,713
19,481,656
Roofing Products
4,850,590
5,150,392
4,367,959
4,357,357
4,347,217
Other
1, 332 ,272
1,465,235
1,431,909
997,116
937,403
-------
Appendix C presents a listing of asphalt roofing manufac-
turing plants by state in 1973. This listing is comprised pri-
marily of plants with 20 or more employees. The total number of
plants listed is 202. Total asphalt sales for roofing products
manufacture for 1972-1976 is shown in Table III-5-
Information for Table III-4 was obtained from the Bureau of
Mines, Mineral Industry Surveys.Appendix C is from the EPA
34/
report Atmospheric Emissions From Asphalt Roofing Processes.
The information was obtained from: 1) The Asphalt Roofing Manu-
facturer's Association; 2) the Report on SIC 2952 (April 22,
1974) by the Economic Information Systems, Inc.; and 3) the U.S.
EPA National Emission Data Survey for 1972.
5. Emission Estimates
Emission estimates for particulate polycyclic organic matter
from asphalt roofing plants were obtained from the 1974 EPA re-
34/
port Atmospheric Emissions From Roofing Processes. This
report contains the results of analyses of particulate samples
obtained by EPA Stack Sampling Method 5. Chemical analyses were
performed with gas chromatographic detection. Emissions were
sampled at two saturating and two air-blowing operations.
Samples were collected before and after the exhaust gases passed
through the control device.
Estimates of POM emissions from asphalt hot-mix plants were
obtained from a study (AP-33) by the U.S. Public Health Service.
In this study, only one plant was tested. Samples taken before
and after emission control devices were separated by benzene
extraction and column chromatography and the analysis made by
ultraviolet-visible spectrophotometry.
84
-------
The emissions data were manipulated by EEA to give emission
factors for particulate POM mass emitted per metric ton of as-
phalt product. No data are available for gaseous polycyclic
aromatic hydrocarbon emissions.
Table III-l (P. ) shows estimated emission factors for the
manufacture of various asphalt products. The uncontrolled BaP
best estimate emission factors, based on, at most, two tests
34 39/
each, ' ' range from 400 yg/Mg for shingle saturators, 300 ug/
Mg for roll saturators, and 2 mg/Mg for air-blowing. The best
estimates of controlled BaP emission factors for the most effec-
tive means of control are less than 80 yg/Mg for shingle satura-
tors with an afterburner, 500 yg/Mg for roll saturators with an
HEAF, 500 yg/Mg for air-blowing with a process heater furnace,
and less than 60 yg/Mg for hot road mix with a cyclone and a
spray tower. Total benzo(a)pyrene emissions from asphalt paving
and roofing manufacture in 197 6 were obtained from these emission
factors and the asphalt sales figures whon in Table III-5. Total
BaP emission estimates for the asphalt industry, as shown in
Table III-2 (P. ), are much less than one metric ton per year.
6. Future Trends
Sales of asphalt for U.S. consumption display an erratic*
growth rate. Until 1972, the industry grew an average of three
percent annually. After 1973, asphalt sales dropped following
general construction trends. Recovery is not yet in evidence for
this segment of the building industry.
EEA estimated an increase of three percent per year in as-
phalt roofing production. Because of a decrease in highway
construction, but t]ie continuing need for road repair, hot
road mix production was assumed to remain constant at approxi-
mately 20 million metric tons per year. Total BaP emissions for
85
-------
the industry are, therefore, expected to remain well below one
metric ton per year.
J. Iron- and Steel Sintering
1. Process
Pulverized ore must be agglomerated to produce a suitable
feed for blast furnaces. Sintering is the most common agglomer-
ating method. To accomplish the sintering, a mixture of fine ore
and powders of carbon sources, such as anthracite and coke
breeze, are placed on a travelling grate. The grate moves over a
series of windboxes where the mixture is ignited with a burner.
As air is pulled down through the ore with fans, the ore mixture
burns, agglomerating the ore particles.
2. Emission Sources
Unburned hydrocarbons may be generated from the burning of
the coke and from the burning of oily scrap. Coke and scrap
particles with adsorbed polynuclear hydrocarbons can escape at
numerous points in the sintering process.
3. Emission Controls
The dust generated by sintering can be controlled with an
electrostatic precipitator, baghouse, or flue gas scrubber. In
1976, approximately 66 percent of sintering plants had no emis-
46/
sion controls. '
4. Location and Capacity1^
Sintering plants are usually operated in conjunction with
large blast furances in order to produce pig iron for steel-mak-
ing. Many sintering plants are located in Ohio, Pennsylvania,'
and Indiana. In addition, large sintering plants are found in
86
-------
Weirton, West Virginia; Sparrows Point, Maryland; and Fairfield,
Alabama. The locations and capacities of sintering facilities in
the U.S. are listed in Appendix D.
5. Emission Estimates
The sintering process is a source of significant air emis-
sions. Uncontrolled particulate emissions are estimated to be
about 11 kilograms per metric ton of sinter produced.^9/ pom's
adsorbed on these particulates and in gaseous form are emitted.
Much of the POM will be vapors at sintering temperatures, but
current sampling techniques do not necessarily collect most of
the vaporous POM.
EEA's best estimate emission factor of 17 mg BaP/Mg with a
range of 600 vig/Mg to 1.1 g/Mg of sinter feed was based on emis-
sions test data from the Pennsylvania Department of Environmental
76/
Resources. Production from sinter strands was determined from
American Iron and Steel Institute figures for 1977. ^ The EEA
best estimate of annual benzo(a)pyrene emissions was 0.63 Mg/year
(range 0.022 to 41 Mg/year).
6. Future Trends
Sinter strand production is expected to increase in the
future due to increased steel demand. Based on an historic
growth rate, sinter production should increase from 31 million
metric tons per year in 1975 to 36 million metric tons per year
by 1985. Since 1977 production was reported as 37 million
metric tons,"^ it is presumed that planned sintering capacity is
in operation and that production will remain constant through
1985. Therefore, annual BaP emissions are expected to continue
to be less than one metric ton per year.
87
-------
K. Carbon Black Production
1. Process
Carbon black is manufactured from incomplete combustion of
natural gas. Plants utilize the furnace, channel, or thermal
process to manufacture it. Regardless of the method used, the
three basic steps in producing carbon black are:
• production of the carbon black from feed
stock;
• separation of the carbon black from the
gas stream; and
9 final conversion of the carbon black
to a marketable product.
Carbon black is produced in both the channel and furnace
processes by burning the feed stock, while in the thermal pro-
cess, the feed stock is decomposed into carbon black and hydro-
gen.
2. Emission Sources
Emissions in carbon black manufacturing result from the com-
busion of the natural gas in both the furnace and channel pro-
cesses and from the gaseous releases in the thermal process.
Additional emissions are possible from conveying, grinding,
screening, drying, and packaging operations at the plant.
3. Emission Controls
Wet scrubbers, cyclone separators, and baghouses are most
commonly used at the carbon black plant to control emissions.
83
-------
Baghouses are most efficient and appear to be replacing all other
methods of control because of increased product recovery.
4. Location and Capacity
The location and capacity of U.S. carbon black plants were
66/
identified in the 1977 Directory of Chemical Producers, U.S.A. '
This information is listed in Appendix E.
5. Emission Estimates
Emission factors for carbon black production were derived
from a POM loading figure given in mass per volume of gas assum-
1 44/
ing an air flow of 3.91 sm /kg (135,000 scf/T). ' POM values
are for total POM analyzed by the gas chromatography-mass spec-
44/
trometry technique reported by Jones, et al. The minimum
value of 220 mg/Mg is from samples taken with a Tenax adsorbent
train. This sampling train includes a heated filter, cooling
coil, and adsorbent column. Another reported value of 310 mg/Mg
corresponds to the results from sampling with the Method 5 train
followed by an adsorbent column. However, the 490 mg of "total
POM"/Mg of carbon black produced is considered by EEA to be the
best estimate.
Total annual emissions were derived from these emission
factors and the production figures given in U.S. Industrial
Outlook 1977.^^ The intermediate estimate of annual BaP emis-
sions was 0.088 Mg/year, assuming 15 percent of the POM was
BaP. 39//
6. Future Trends
A recent decline (1975-1976) in the production of carbon
black can be attributed to the 197 6 tire strike. A low growth
rate of two to three percent is expected due to the increasing
39
-------
production of smaller cars (smaller tires), and the growing pop-
ularity of radials.on this basis, EEA estimates carbon
black production to be 1.5 million metric tons in 1985. There-
fore, BaP emissions are expected to remain significantly below
the one metric ton level through 1985.
L. Aluminum Reduction
1. Process
Aluminum metal is produced by electrolytically reducing
purified alumina (aluminum oxide). Thermal reduction with coke,
which is used in iron ore processing, cannot be employed due to'
the high melting point of aluminum oxide.
In a process developed by Hall and Heroult in 1886, the
alumina is dissolved in a bath of molten flouride in a large
steel pot. Within the reduction plant, pots that are electrical-
ly connected in a series and located in "pot rooms," constitute a
"pot line."
The passage of a direct current through the molten material
causes the heavier aluminum to sink through the aluminum oxide to
the bottom of the pot and to the cathode. At the anode, oxygen
is liberated and carbon monoxide and carbon dioxide are formed.
Carbon electrodes are used at both the anode and cathode, al-
though the aluminum metal in the cell is the true cathode. The
aluminum is tapped at certain intervals and cast into pigs or
taken to holding furnaces for further treatment.
2. Emission Sources
The sources of POM emissions from aluminum reduction plants
depend on the process used. The reduction processes currently
90
-------
used are classified by the type of anode pot used, i.e., pre-
baked, horizontal pin Soderberg, or vertical pin Soderberg. Pre-
baked anodes are made by curing the carbon in soft pitch and coke
at relatively high temperature (<1,100°C), thus volatilizing and
generating POM's. Since most of the POM's are generated during
the pre-baking of the anodes, relatively little POM is generated
during the reduction process when the anodes, which contain a rod
of metallic conductor, are lowered into the pot as they are
consumed.
Soderberg anodes are contuously lowered and baked by con-
ductive heat from the molten bath rather than being pre-molded
and baked. A coke and coal tar pitch paste is packed into a
metal shell over the bath. As the baked anode at the bottom of
the shell is consumed, more paste is added at the top of the
shell. The description, horizontal or vertical pin, refers to
the positioning of the steel "pins" which are imbedded in the
Soderberg anode to conduct electrical current. The type of pin
design may effect the location of emission sources around a pot.
Since the carbon paste is not baked before being placed in the
pot, the POM emissions from a Soderberg pot room are much higher
than from a pre-baked pot room.
3. Emission Controls
Emissions from aluminum reduction facilities are collected
and controllied by a variety of methods. Hoods may be utilized to
collect emissions from specific points in a pot room. The
ventilation system for the pot room is generally the only means
of collecting emissions. The types of air pollution control
equipment, which are used either singly or, more often, in com-
bination, include settling chambers, cyclones, wet and dry scrub-
bers, wet and dry electrostatic precipitators, baghouses, and
incineration. The effectiveness of these devices for POM would
91
-------
depend upon their design and operating parameters and the gas
stream conditions which determine the form of the POM.
4. Location and Capacity
Aluminum reduction plants are located throughout the country,
though few plants are located in the west. Many of the plants
are located near port facilities, primarily on the Gulf of Mexi-
co or the Great Lakes. Capacities range from productions of
about 50,000 to 250,000 metric tons per year, while most plants
have capacities in the area of 100,000 metric tons per year.
5. Emission Estimates
Estimates of POM emissions from aluminum reduction plants
could not be made at this time. The National Institute for
I
Occupational Safety and Health (NIOSH) has conducted an environ-
84/
mental survey of aluminum reduction plants. The results of
this study are not suitable for emission factor development as
they are reported as time-weighted averages of BSO concentrations
in the plant ambient air as collected by personal samplers. EPA
has conducted some stack sampling of aluminum smelting and re-
fining operations. Samples have not been analyzed for POM,
though.
6. Future Trends
Since emission factors cannot be developed at present,
future emissions of POM from aluminum reduction cannot be esti-
mated. Primary production capacity in the U.S. is projected to
increase slowly to 48 billion kilograms by December 31, 1978 from
4.7 billion kilograms in 197 6. Close to maximum capacity opera-
tion is expected at that time due in part to an estimated annual
growth rate in demand of approximately six percent through 1985.
92
-------
DRAFT
DO NOT QUOTE OR CITE
Significant increases in aluminum demand in the transportation,
construction, and packaging and distribution sectors are, in
part, responsible for this annual growth rate.
M. Municipal Incinerators
1. Process Description
In municipal incinerators, refuse is combusted on a moving
belt, in drum-type rolling combustion chambers, or on a rocking,
reciprocating, or travelling grate. Refuse is fed to the incin-
erator continuously or in batches. Normally, 150 to 200 percent
excess air is supplied in order to prevent erosion of refractory
materials in high temperatures. Thus, a large amount of com-
bustible exhaust gas is produced and may be burned in a secondary
chamber. Gaseous emissions are discharged through chimney
stacks. The resultant ash in the chambers, both the residue and
all material remaining unburned, is landfilled.
2. Emission Sources
Emissions of polycyclic organic matter result from incom-
plete combustion of organic refuse. Benzo(a)pyrene and benzo(e)-
pyrene were detected in the flue gases from every incinerator
39/
tested by the U.S. Public Health Service Survey. Large mu-
nicipal units, operating at constant high temperatures with long
gas retention times, tend to emit less POM per mass of refuse
burned than smaller units. Emissions depend on the composition
of refuse burned and so tend to vary with time and location.
Many municipal incinerators burn unsorted industrial wastes.
Such wastes may include petroleum-based materials which generate
large quantities of POM's upon combustion.
93
-------
3. Emission Controls
A water-spray scrubber has been used effectively to control
39/
particulate POM emissions in flue gas. ' Baghouse filters and
electrostatic precipitators also are feasible control mechanism's.
In order to comply with air quality regulations, most municipal
incinerators (83 percent) use one of these forms of emission
control.
4. Location and Capacity
A survey conducted in 197 5 by the ASME Research Committee on
Industrial and Municipal Wastes revealed a total of 160 operating
municipal incinerators.30^ The listing of plants, locations, and
their design capacities as updated by EEA to omit plants no
longer in service is shown in Appendix F. In 1975, these plants
had an average capacity of 380 metric tons (415 tons) per day and
had been used an average of slightly over 15 years. Almost 80
percent of the plants lie in the middle and eastern portions of
the country.
5. Emission Estimates
POM emission estimates for various types of municipal incin-
erators were obtained from the U.S. Public Health Service report,
3 9/
Sources of Polynuclear Hydrocarbons in the Atmosphere, a
2^/
1976 paper, and a report done in 1970 by Arthur D. Little,
95/
Inc. ' Emission factors were derived by EEA for incinerators of
various sizes. These are shown in Table III-l (P. ). Uncon-
trolled BaP emission for the two tests of multiple chamber in-
39/
cmerators that were available were 13 mg/kg of refuse charg-
{
ed for a 45 metric ton (50-ton) per day batch unit and 170 ng/kg
for a 230 metric ton (250-ton) per day continuous unit. Con-
trolled BaP emission factors were developed from results of one
94
-------
test each for a 45 metric ton (50-ton) per day batch unit with a
39/
water-spray scrubber ' and for a 27 metric ton (30-ton) per day
continuous unit with water-spray tower and ESP. The emission
factors are 200 ng/kg and 82 ng/kg of refuse charged, respec-
tively. Current emissions of BaP from municipal incinerators are
estimated to be less than one metric ton per year.
6. Future Trends
It is expected that new municipal incinerators will continue
to have larger capacities than in the past. Since many of the
older, smaller plants are being taken out of operation, no change
is expected in the total capacity. Increasingly stringent air
pollution regulations may require more efficient air pollution
control or shutdown of many plants/ Therefore, it is expected
that BaP emissions will continue to be less than one-metric ton
per year.
N. Commercial Incinerators
1. Process
Commercial incinerators range in size from 20 to 2,000 kilo-
grams (50 to 4,000 pounds) of refuse charged per hour capacity
with an average size of 103 kg/hour (228 lbs/hour). Incinerators
are widely used to reduce the volume of industrial, medical,
commercial., high-rise buildings, and school wastes. Eight-three
percent of existing units are multiple-chamber devices and 92
percent use auxiliary -fuel.^^
Intermediate-sized fuel incinerators are characterized by
inefficient combustion and, thus, are potential emission sources
for POM's.
95
-------
2. Emission Sources
Approximately nine million tons of solid waste per year are
burned in commercial incinerators. The amount of POM emis-
sions generated depends upon the type of wastes burned and the
efficiency of the incinerator. Waste with a high content of
moisture or petroleum-based material will tend to emit more
aromatic hydrocarbons. Efficiency of combustion depends on both
size and excess air supply. Those units which use auxiliary fuel
tend to emit less POM's due to their ability to maintain a higher
temperature.
3. Emission Controls
Of the 88 percent of commercial-size incinerators that have
pollution abatement equipment, 90 percent have afterburners, five
percent have scrubbers, five percent have both an afterburner and
a scrubber, and one unit was reported to have an electrostatic
precipitator.
4. Location and Capacity
There were slightly more than 100,000 intermediate-size in-
cinerators in the U.S. in 1972 according to a study by EPA's
Office of Solid Waste Management Programs. The distribution
of these facilities by location and capacity as estimated from a
sample of 5,320 units is shown in Table III-6. The average
capacity for a larger sample of 7,288 was determined to be 103
kg/hour (228 lb/hour). Each unit operates an average of three
hours per day for 260 operating days/year.
5. Emission Estimates
Estimates of POM emissions from intermediate-size inciner-
ators were obtained from the U.S. Public Health Service report,,
96
-------
TABLE III-6
ESTIMATED NUMBER OF INTERMEDIATE-SIZE INCINERATORS
IN THE UNITED STATES (1972)16/'
Estimated Number Average Unit Size
EPA Region of Units (kg/hr(lb/hr))
1
8,040
94
(207)
2
8,832
209
(460)
3
11,560
118
(261)
4
6,457
154
(340)
5
45,876
71
(157)
6
6,596
117
(257)
7
5,580
85
(187)
8
3,644
99
(218)
9
2,350
134
(295)
10
2,820
126
(277)
TOTAL
101,755
102 (225)
-------
39/
Sources of Polynuclear Hydrocarbons in the Atmoshpere. Ef-
fluent samples were taken from a 4.8 metric ton per day (5.3
ton/day) unit and a 2.7 metric ton per day (3.0 ton/day) unit
equipped with an auxiliary gas burner.
POM emission factors for the two uncontrolled incinerators
tested are given in Table III-l (P. ). The uncontrolled BaP
emission factors for the single tests on the two units are 120
yg/kg of refuse charged for the 4.8 Mg/d unit and 570 yg/kg of
refuse charged for the 2.7 Mg/d unit. Total annual BaP emissions
from commercial incinerators developed from these factors are
given in Table III-2 (P. ). EEA's best estimate of BaP emis-
sions is 2.1 Mg/year for the 1972 capacity data.
6. Future Trends
The installed capacity of intermediate-size incinerators
appeared to be leveling off in 1972. The size of units being
installed was still increasing. However, the number of units
sold per year reached a maxmium in 1969. Since construction has
not generally been increasing and the larger units should have
more complete combustion, it was assumed that both capacity and
BaP emissions would not change through 1985.
0. Bagasse Boilers
1. Process Description
Bagasse (plant residue after extraction of a product) is
used to fuel steam boilers at many sugar cane and pineapple pro-
cessing plants. Travelling grate spreader stoker boilers are
commonly used.
98
-------
The bulk of the burned material consists of dead and green
leaves. Pineapple trash consists of the stumps and leaves that
remain after harvest.
2. Emission Sources
Due to the high moisture content of plant material and the
inefficient burning common to small boilers, large quantities of
POM can be emitted from bagasse boilers.
3. Emission Controls
Cyclones are used at the three boilers tested by MRI as flue
gas emissions control mechanism. The particulate collection
4/
efficiency of the cyclones tested is 85 to 90 percent. '
4. Location and Capacity
Almost all bagasse boilers are located at sugar and pine-
apple processing plants in Hawaii.
5. Emission Estimates
Emission estimates were derived from data in the EPA report,
Stationary Source Testing of Bagasse-Fired Boilers at the
4/
Hawaiian Commercial and Sugar Company. ' Fuel consumption was
calculated by assuming a heat content for bagasse of 2.2 million
calories per kilogram (4,000 Btu/lb). Estimated emission factors
for POM are shown in Table III-l (P. ). The BaP emissions were
below detectable limits for the three stack samples taken. Non-
detectable BaP levels were assumed to be the minimum detectable
level of 1.0 yg in the POM detected in each test and the geo-
metric average of the tests taken to produce an estimated BaP
emission factor of 2.7 yg/kg. Bagasse boilers do not contribute
substantially to BaP emission to the atmosphere as emissions are
less than 0.0061 mg/year.
99
-------
6. Future Trends
If fossil fuel prices rise as expected, use of the bagasse-
fired boiler will probably increase slightly. Production of
sugar in Hawaii declined somewhat in 1974 and 1975. However,
since 1950, production has remained between 845 and 1,145 thou-
125/
sand tons per year (high in 1950, low in 1960). ^ Therefore,
only slight increases, if any, are expected in bagasse boiler
usage. BaP emissions are expected to remain on the order of
kilograms per year.
P. Open Burning
1. Process
Open burning refers simply to the combustion of organic
materials. In the U.S., the following are, or have been, inten-
tionally burned in the open: municipal refuse, auto scrap,
grass, leaves, agricultural waste, and forest areas. In addi-
tion, there are two other open burning sources, burning coal
refuse banks and forest fires. Both are generally unplanned
and uncontrolled. Coal refuse banks can ignite spontaneously,
while forest fires are caused in a number of ways. Potential
aromatic hydrocarbon emissions will be discussed separately
for each material type.
2. Municipal Refuse
a. Emission Source
Municipal refuse was once commonly burned at municipal
dumps in order to reduce the volume of waste. Municipal wastes
contain varying quantities of organic materials and moisture
depending on origin. Combustion of refuse piles tends to be in-
complete due to high moisture content and because wastes are not
100
-------
evenly exposed to heat or oxygen. Organic materials in refuse
piles will emit polycyclic aromatic hydrocarbons when ineffi-
ciently burned.
b. Emission Controls
Control of atmospheric emissions from a burning refuse pile
is not feasible. The only emission control possible is to ex-
tinguish the fires and replace open burning with some other waste
disposal method. Most state and local governments in the U.S.
have promulgated restrictions on open burning of refuse.
c. Location and Capacity
Open burning of municipal refuse is no longer common due to
air quality regulations. Open burning that does occur tends to
be the result of spontaneous combustion of refuse piles. No
figures of tonnage burned were available for this, study.
d. Emission Estimates
39/
Hangebrauck, et al. cites a benzo(a)pyrene emission fac-
tor of about 340 yg/kg of municipal waste. The study utilized a
"burning table" test of burning refuse samples and on-site sampl-
ing at refuse dumps. Additional data from NAPCA and EPA re-
39 109/
ports ' were used to derive EEA's other emission factors, as
shown in Table III-l (P. ). The intermediate estimate BaP
emission factor for the open burning of municipa refuse was 170
yg/kg of refuse.
Emissions of polycyclic organic matter from open burning of
municipal refuse are high. Estimates of total annual POM emis-
sions from this source were not made in this report due to
101
-------
lack of adequate data on tonnage burned. The latest national
survey of open burning was done in 1968. It is felt that this
survey does not adequately reflect the present situation, es-
pecially because of air quality legislation promulgated in the
19701s.
e. Future Trends
Open burning of municipal refuse is expected to continue to
decrease due to increasingly stringent air quality regulations.
3. Burning of Leaves and Grass Clippings
a. Emission Sources
Grass clippings and fallen leaves are burned throughout the
United States in curbside fires and in large controlled fires by
leaf collection agencies. High moisture content and unconsoli-
dated fuel piles lead to incomplete combustion of the organic
matter. Polycyclic aromatic hydrocarbons are emitted from such
fires, both adsorbed on particles and in gaseous form.
b. Emission Controls
Consolidation of piles, pre-combustion drying of grass and
leaves, and the maintenance of high combustion temperatures are
all feasible means of reducing POM emissions. Substitution of
enclosed burning with exhaust gas cleaning is the only efficient
"control" mechanism available for grass and leaf burning. Ilany
localities have implemented such controls by banning open burn-
ing of grass clippings and leaves and providing leaf collection
systems.
c. Location and Capacity
Information on grass clippings and leaf burning practices was
not available for this study.
102
-------
d. Emission Estimates
24 45/
Samples from a study by Darley ' for EPA provided the
emission factors for leaf burning shown in Table III-l (P. ).
The study involved sampling from a leaf burning facility using a
filter and Tenax adsorber and extracting the samples using
methylene chloride for the filter and pentane for the adsorber.
These extracts were separated by liquid chromatography and
analyzed using gas chromatography and mass spectrometry. POM
values reported are totals for 19 POM species.
BaP values reported are actually for combined benzo(a)pyrene
and benzo(e)pyrene, as detected in the sampling and analysis
procedure outlined above. A non-detectable (ND) BaP value re-
ported was assumed to be 40 yg/kg of leaves burned in calculating
the intermediate estimates of BaP emissions (190 yg/kg for a
composite of leaf types; 325 yg/kg for the geometric average of
results for the three types burned separately).
Emissions estimates for open burning of grass clippings were
developed from BaP emissions or concentrations in particulate.
32 39/
matter in several studies ' which used "burning tables" and
on-site sampling of smoke, in some cases, in combination with
particulate emission factors from AP-42.109/
e. Future Trends
Air quality goals will likely cause widespread prohibition
of open burning of leaves and grass clippings in the future.
4. Automobile Scrap
\
a. Emission Sources
In order to meet the quality standards of the steel industry
for scrap bundles, organic materials must be removed from auto
103
-------
bodies. This may be accomplished by open burning of whole auto
bodies, incineration of whole auto bodies, or shredding and sub-
sequent incineration of the shredded steel.
b. Emission Controls
Emissions from open burning of auto bodies cannot be con-
trolled except by prohibiting the activity. At this point, how-
ever, most automobiles are incinerated to eliminate organic ma-
terials. Incinerators can utilize wet gas scrubbers or other
conventional stack emission controls. The combustion efficiency
of rotary kiln incinerators for shredded steel is very high so
little POM is generated.
c. Location and Capacity
In most metropolitan areas, open burning is strictly pro-
hibited and, consequently, auto hulks often are taken outside
the restricted area for burning. Otherwise, enclosed burning in
incinerators or hand-stripping of combustibles is practiced.
Most open burning which now takes place is illegal; location
58/
and quantity of open burning of auto bodies are unknown.
d. Emission Estimates
Technological advances are directly affecting the status of
the scrap processing industry. The demand for auto scrap by
the steel industry is rising with the increased use of electric
arc furnaces which use a greater proportion of scrap in the fur-
nace feed. At the same time, the supply of reusable scrap (or
"home" scrap) generated at steel plants is decreasing due to
the increasing use of continuous casting. Thus, auto scrap pro-
cessing will tend to increase in the future. However, new scrap
processing techniques, primarily shredding and rotary kilns,
104
-------
have practically eliminated open burning and incineration of
58/
scrapped auto bodies. The rising demand for steel scrap has
made shredding and subsequent incineration economically feasible.
Emission factors were developed for open burning of automo-
bile components. BaP emissions were estimated to be about 22
milligrams per kilogram of automobile components charged. How-
ever, since open burning of auto scrap is now rarely, if ever,
practiced, current BaP emissions from this source would be ex-
pected to be negligible. No BaP emission factor data is avail-
able for incineration of shredded auto scrap in rotary kilns.
e. Future Trends
Demand for No. 2 steel scrap is expected to continue its
rising trend, with an added impetus from energy and resource
conservation incentives. This demand will improve the economics
of centrally-located shredding-incineration operations. It is
projected that nearly no open burning or whole car body incin-
eration will be practiced after 1980.
5. Coal Refuse Piles
a. Emission Sources
Coal refuse banks exist throughout the nation's coal-pro-
ducing regions. Spontaneous combustion of the coal wastes, coal,
shale, and calcite is a common occurence. Many large refuse
piles have smoldered internally for many years. Burning coal is
a major source of particulate polycyclic organic matter and gase-
ous hydrocarbons even in a relatively efficiently operated fur-
nace. Thus, it is especially so under such inefficient burning
conditions as the poor air supply and uneven heat distribution of
a burning coal refuse bank.
105
-------
b. Emission Controls
Extinguishing coal refuse fires is the only feasible means
of controlling POM emissions. The three methods of extinguishing
refuse pile fires are to: (1) dig out and cool the affected
material, (2) cover the pile to seal it against air circulation,
54/
and (3) grout to solidify the affected material. It has
proven to be very difficult and costly to extinguish existing
fires. Methods to extinguish fires and to prevent new fires by
proper pile construction and sealing have been outlined by Mag-
54/
nuson and Baker. Burning active refuse piles are generally in
some degree of compliance with air pollution regulations, while
11 12 73 74 79/
preventive measures are practiced at other active piles. ' ' . ' '
c. Location and Capacity
The Bureau of Mines (BOM) located 292 burning coal refuse
banks in 1968, extending over 3,200 acres.At least 24 of
these fires now have been extinguished by BOM in Pennsylvania and
74/
extinguishment of four other fires will soon be accomplished.
16/
No western coal refuse fires have yet been extinguished.
Existing burning piles as listed in the BOM survey are not given
because the Mining Enforcement and Safety Administration (MESA)
)
is currently updating this information. Estimates of the quan-
tity of coal contained in these piles were not available for this
study.
d. Emission Estimates
No emission factors or data on total tonnage burned annually
were available for this report. Total annual emissions estimates
for BaP of 280 and 310 Mg are given in Table III-2 (P. ).
These figures were taken directly from the Preferred Standards Path
106
-------
Report for Polycyclic Organic Matter^^ and the 1972 NAS study.
The estimates were evidently derived from the 1968 BOM survey.
61/
The assumptions and method are unknown. The emissions may
have changed significantly since then though many banks have been
extinguished and others have naturally gone out, while others
have caught fire or burst into flames that were smoldering. If
the current emissions are similar, burning coal refuse banks are
the largest single contributor of POM to the atmosphere. Assum-
ing that total emissions are proportional to the number of banks,
17/
the estimated 1968 emissions of 310 Mg/year ' would have been
reduced to 280 Mg/year considering only banks known to be burning
74/
in 1968 and the number of banks extinguished since then. '
e. Future Trends
The Bureau of Mines is attempting to extinguish many of the
burning coal refuse piles in Pennsylvania. The program has not
yet begun in the western coal regions and the eastern operations
have been progressing slowly. Emissions of POM cannot be pro-
jected as banks may ignite, smolder, or become extinguished
naturally or by human action.
6. Forest Fires
a. Emission Sources
Forest fires, both wildfires and prescribed fires, inef-
ficiently burn massive quantities of organic material each year.
Combustion tends to be incomplete due to the high moisture con-
tent and varying characteristics of the fuel.
Wildfires emit much greater quantities of particulate POM
than prescribed fires because they burn at greater intensities
107
-------
and thus, ignite larger vegetation. Such large trees do not
burn completely. Therefore, more particles and more unburnt
23/
hydrocarbons are emitted. '
The amount of POM emitted depends upon the forest type,
the weather, and the season.
b. Emission Controls
Prescribed burning is practiced in most areas to prevent
wildfires. Forest litter is burned off in a well-controlled
operation, thus leaving less ignitable material in the forest.
Other fire prevention techniques are also utilized by national,
state, and local forest managers to reduce the potential for
fires. Adequate manpower and equipment for fire control is the
next best POM emission control method for forest fires.
Slash burning of waste material is practiced by loggers.
Small piles, composed of small branches often with leaves
attached, are formed along logging roads and burned with super-
vision. Combustion, and thus emission, characteristics vary
widely with the area, forest type, and weather.
c. Location and Capacity
Statistics on acreage burned by prescribed fires and wild-
fires are maintained by the U.S. Forest Service for all U.S.
19/
forests. These are shown in Table III-7. The fuel content
2
per acre is roughly estimated to be about 18 0 kg/m (300 tons/
2
acre) in the Pacific Northwest areas and 110 kg/m (500 tons/
acre) elsewhere.
d. Emission Estimates
Forest fires, especially wildfires, are a major source of
POM in the atmosphere. Emission factors per mass of fuel
100
-------
burned were developed from average results reported for duplicate
59/
tests by McMahon and Tsoukales in 1977. The tests involve
burning slash pine needle litter in a controlled environment
burning room and sampling with a modified "hi-vol" sampler. The
samples were extracted with methylene chloride, separated by
liquid chromatography, and analyzed with gas chromatography and
mass spectrometry. Emission factors are given in Table III-l
(P. ) for different fire conditions. The best estimates of
BaP emission factors from these tests with pine needles ranged
from 27 pg/kg for flaming heading fires to 770 |ig/kg of fuel for
backing fires. Emission estimates ranging from 9.5 to 127 Mg/
year from forest fires have been reported in the literature.
No estimate was made in this study because of the great uncer-
tainty and variability in forest fire combustion processes and
their resultant emissions. For purposes of comparison with
reported estimates, a number of 100 Mg/year was generated by
assuming an estimate of 100 Mg/year assuming: an emission factor
for ten percent backing fires and 90 percent heading fires using
the overall emission factors developed from burning pine needles,
the total area burned in the U.S. in 1976 of 2.07 x 10"^
and the estimated U.S. average wildfire consumption of 38 Mg/
hectare. Actual emissions from a fire are highly variable as
type and availability of fuel, burning conditions, and area
78/
burned all vary with the location, climate, and season,
e. Future Trends
There will likely be little change in acreage burned per
year unless major increases in fire prevention and fire control
efforts are implemented. Presumably, POM emissions from forest
fires will remain somewhat constant.
109
-------
Q. Mobile Sources
1. Gasoline Consumption
a. Process
Gasoline is burned in spark-ignition internal combustion
engines by passenger cars, trucks, and buses. Diesel engines
and two-cycle engines used in motorcycles, motorboats, and lawn
mowers are discussed separately.
b. Emission Sources
POM emissions from gasoline consumption result from inef-
ficient fuel use due to air to fuel ratios less than stoichio-
metric, driver operating modes, engine deterioration, and
combustion chamber deposits. Fuel quality is also an important
factor. The aromatic content of the gasoline, additives, and
lubricants can affect POM formation levels. 3egeman and Colucci
17/
as reported in the NAS study, estimate that as much as 36
percent of. the benzo(a)pyrene in exhaust gas can be attributed
to the fuel benzo(a)pyrene content.
The lead content of gasoline also influences the POM emis-
sion levels. In place of lead additives, aromatic hydrocarbons
are usually increased to maintain high octane levels. However,
potential increases in POM emissions due to higher fuel aro-
maticity are offset by changes in the nature of combustion
17/
chamber deposits when unleaded fuel is used. ' Research is
continuing to clarify the role of combustion chamber deposits
in POM formation.
The pyrolysis of heavy motor oil also generates POM which
may be released in exhaust. Generally, motor oil deposits in
the engine build up over the operating life of the car and
particularly between oil changes.
110
-------
c. Emission Controls
Emission control devices presently in use have greatly
reduced POM emissions. Engine modifications, such as exhaust
gas recirculation which was introduced in 1968, have reduced
the emissions of polynuclear aromatics from those of uncontrol
38/
led models by 65 to 80 percent. Catalytic converters simi-
lar to those which became common in 1975 have been shown to
3 8/
reduce polynuclear aromatic emissions by about 99 percent.
As described by the Motor Vehicle Manufacturer's Association,6
the various control devices which have been introduced include
• Catalyst or equivalent control systems
were introduced on cars in:1975 to meet
much tougher emission levels for hydro-
carbons and carbon monoxide while at
the same time improving vehicle fuel
economy.
• Improved NOx control systems on some
1971 and 1972 models, and on all models
for 197 3 and after lower total vehicle
emissions of oxides of nitrogen, the
other major ingredient in photochemical
smog formation.
• Evaporative fuel losses from gasoline
tanks and carburetors were nearly elim-
inated by controls on all new -cars
beginning with 1971 models.
Ill
-------
• Exhaust controls, introduced nation-
wide on 1968 models, accelerated the
reduction of hydrocarbon emissions
and brought major reductions of emis-
sions of carbon monoxide, an invisible,
odorless gas which forms the bulk
of automotive emissions.
• Crankcase controls were installed
nationwide starting with 1965 models,
two years after they were introduced
in California. Before being controlled,
the crankcase was the source of about
20 percent of emissions of hydrocarbons,
the unburned fuel active in photo-
chemical smog formation.
Only three percent of the cars on the road in 197 6 were without
emission controls of any kind. The breakdown, as derived by the
Motor Vehicle Manufacturers Association,is shown in Table
III-8.
d. Location and Capacity
Gasoline demand in the U.S. was 1,139,000 cubic meters
(7,163,000 barrels) per day in October 1977 (four-week average).^
8 3
This implies an approximate consumption of 4.2 x 10 /m
(2.63 x 10^ barrels) or 4.2 x lO^ liters (1.11 x 10^ gallons)
of gasoline in 1977. About 20 percent or 8.4 x 10^ liters of
this demand was for unleaded gasoline.
e. Emission Estimates
Estimates of POM emission factors were available from
38/
Gross for automobiles of various ages burning leaded and
112
-------
TABLE III-8
U)
CARS
IN OPERATION
WITH EMISSION CONTROLS
(cars in
thousands)
1966
1971
1972
1973
1974
1975
1976
Catalyst or equivalent, NO ,
fuel evaporation, exhaust and
crankcase controls
0
0
0
0
0
4,684
14,155
NO , fuel evaporation, exhaust
anc5 crankcase controls
0
*
726*
8,950
18 ,675
22,053
21,820
Fuel evaporation, exhaust and
crankcase controls
0
6,787
16,213
18,734
18,607
18,476
17,937
Exhaust and crankcase controls
573
27,676
27,214
26,365
25 ,522
24,650
22,928
Crankcase control only
31,060
35,813
32,879
28,988
24,817
21,359
17,697
No controls
39,631
12,846
9,469
6,745
4,962
3,998
3,253
TOTAL cars
71,264
83,122
86,411
89,782
92 ,583
95,220
97,790
PERCENT
OF CARS
IN OPERATION WITH EMISSION CONTROLS
Catalyst or equivalent, NO ,
fuel-evaporation, exhaust
and crankcase controls
0.0
0.0
0.0
0.0
0.0
4.9
14.5
NO , fuel evaporation, exhaust
an§ crankcase controls
0.0
0.0
0.8
9.9
20.2
23.2
22.3
Fuel evapoaration, exhaust and
crankcase controls
0.0
8.2
18.8
20.9-
20.0
19.4
18.3
Exhaust and crankcase controls
0.8
33.3
31.5
29.4
27.6
25.9
23.5
Crankcase control only
43.6
43.0
37.9
32. 3
26.8
22.4
18.1 ,
No controls
55.6
15.5
11.0
7.5
5.4
4.2
3.3
TOTAL percent
*
100.0
100.0
100.0
100.0
100.0
100.0
100.0
o
o
Improved control systems on some
emissions of oxides of nitrogen.
1971 and 1972 models and on all models for 1973 and after lowered total vehicle C5
NOTE: Data as of July 1st of each year, not model year.
-------
unleaded gasoline. These are shown in Table III-l (P. ). An
overall figure of 9 yg BaP/1 for the estimated 1977 auto popula-
tion was derived by EEA. The figure was weighted by percentage
of total mileage travelled by each type of auto using the age
63/
distribution of the U.S. auto population for 1976 and the
average annual miles driven for autos of various ages.120/ T^e
distribution by percentage of annual travel used for the various
test results was 32.3 percent 1970 automobile with a catalytic
converter and using unleaded gasoline (1975-1977 model years);
48.2 percent 1970 automobiles with engine modifications (1970-
1974); 9.5 percent 1968 automobiles with engine modifications
(1968-1969); and 10.0 percent 1966 uncontrolled automobiles (1966
and older) with the latter three types using leaded gasoline.
Estimates of total BaP emissions per year from gasoline
consumption are shown in Table III-2. The best estimate of BaP
emissions is 2.7 Mg/year for a 1975 estimated consumption of 2.96
x 1011 1.69/
f. Future Trends69^
By 1985, domestically manufactured automobiles are likely to
meet the EPA fuel economy standard of 11.7 km/1 (27.5 mpg). The
continued penetration of small, fuel efficient imports will raise
the average new car fuel economy to 12.1 km/1 (28.5 mpg).
A major factor behind the increase in new car fuel mileage
is the diesel, which is about 25 percent more efficient than the
conventional engine and which is likely to capture up to 15
percent of the new car market by 1985.
In 1985, gasoline consumption is expected to be about 2.6 x
1011 liters per year.This represents a decline in total fuel
consumption of 6.1 percent. As more of the automobile
114
-------
population will be equipped with catalytic converters or equi-
valent control, the 1985 weighted emission factor is projected by
EEA to be 0.8 yg/1. Therefore, total1'emissions of BaP from
gasoline-powered automobiles are projected to be less than one
metric ton in 1985.
2. Diesel Fuel Consumption
a. Process Description
Diesel oil, which is burned in the diesel engine, is used by
some automobiles, trucks, and other motor vehicles. An unregu-
lated flow of air is fed into the engine and mixed with the fuel.
The mixture is compressed and thus ignited when it reaches the
cylinder or combustion chamber. The injection of the highly-
pressurized gases into the cylinder causes a sudden reduction in
their pressure, in turn creating air temperatures which cause the
ignition. The energy of the burning mixture moves the pistons.
The pistons' motion is transmitted to :the crankshaft that drives
the vehicle. The burned mixture then leaves the car through the
exhaust pipe.
b. Emission Sources
Overloading and poor maintenance of diesel engines is a
primary cause of POM formation. However, even under normal op-
erating conditions, diesel engines at low and idle speeds produce
higher POM emissions, presumably because of lower combustion-
chamber temperatures.17^ Fuel composition appears to have little
17/
effect on POM emission levels.
c. Emission Controls
Proper loading, fueling, and maintenance of diesel engines
can significantly lower POM emissions. .Exhaust controls and
other devices are not commonly used on diesel vehicles.
115
-------
d. Location and Capacity
Statistics on diesel fuel sales, on and off highways, are
maintained by the Bureau of Mines. These are shown in Table III-
9.103/ ^ breakdown of automobile fuel consumption by diesels and
gasoline engines is given in Table 111-10. This was derived from
data in the EEA report to the Office of Technology Assessment,
Technology Assessment of Changes in the Use and Characteristics
of the Automobile. From this information, it can be seen that
nearly all of the 1975 on-highway diesel use of 3.4 x 10^® liters
was consumed by trucks (heavy duty diesels).
e. Emission Estimates
Emission factors for light and heavy duty diesel engines
17,49,72,92,92,110/
were derived using data from several studies.
For light duty diesels, the BSO emission factor shown was cal-
culated from the BSO emission factors per kilometer given for
49/
various speeds by Laregosti, et al. The figure shown in Table
III-l (P. ) is based on an assumption of time at speed distri-
bution of 35 percent at 35 km/hour, 35 percent at 64 km/hour, 25
percent at 88 km/hour, and five percent at 96 km/hour; diesel
mileage was assumed to be 9.4 km/1 (22 mpg). The benzo(a)pyrene
emission factor given in mass per mass of fuel by Springer and
93/
Baines was converted to mass per volume of fuel, assuming a
diesel oil density of 7.83 kg/1 (No. 2 oil). The best estimate
of BaP emissions from diesel automobiles is thus 340 yg/1 of
diesel fuel burned.
The estimated emission factors for heavy duty diesels are
also given in Table III-l (P. ). The reported BaP emission
factors range from 2.3 to 180 yg/1. The intermediate estimate of
3.7 yg/1 was calculated by taking the geometric mean of the
92/
results of the two tests reported by Spindt.
116
-------
TABLE II1-9
SALE OF DISTILLATE FUEL OIL1^
(millions of liters)
BY USE IN THE UNITED STATES, 1971-1975
a/
1971
1972
1973
1974
2/
Heating
83,385
86,384
85,353
78,416
Industrial, excluding oil company
use
8,066
9,593
10,701
10,181
Oil Company Use
2,240
2 ,131
2,369
2,195
Electric Utility Companies
5,617
10,864
12,393
13 ,46o'
Railroads
13,713
15,422
16,348
16,368
Vessels
3,332
.3,518
4,259
3,936
Military
2,771
3,209
3,116
2,822
On-Highway Diesel
26,548
30,057
35,203
35,141
Off-Highway Diesel
7,937
7 ,979
8,830
7,750
All Other
1,614
1,725
1,888
1,611
TOTAL U.S.
154,427
170,890
180,460
171,879
SALES-OF
RESIDUAL-FUEL
• OIL5- -BY'
USE -IN >THE-iUNITED
STATES
4/
"4/
1975
77,446
10,174
2,167
10,366
14,816
4,156
2 ,862
34,533
7,787
1,605
165,913
4/
4/
(millions
of liters)
Heating
28,945
30 , 384
30,566
27,488
Industrial, excluding oil company
use
21,657
22,627
24,208
22,851
Oil Company Use
5,187
7,042
8,053
7,987
Electric Utility Companies
59,115
69,215
80,997
75,551
Railroads
201
181
193
187
Vessels
12,517
12,390
14,693
14,476
Military
4,645
3,915
3,640
6,427
All Other
971
1,413
1,435
1,352
TOTAL U.S.
133,238
147,166
163,785
153,139
6/
24,659
17 ,864
8,027
72,329
93
15 ,370
3,032
964
142,337
6/
% Change
- 1.2
- 0.1
- 1.2
-23 .0
- 9.5
5.6
1.4
- 1.7
0.5
- 0.3
- 3.5
b/
-10.3
-21.8
0.5
- 4.3
-50.4
6.2
- 6.6
-28.7
- 7.1
-------
TABLE III-9 (Continued)
FOOTNOTES
1/
2/
3/
4/
oo
5/
6/
a/
b/
Includes diesel fuel.
Revised.
Includes range oil.
Data for 1975 includes 3,125 million liters (19,656,000 bbl) of distillate #2 and 399 million liters
(2,510,000 bbl) of distillate #4 fuel oil used at steam electric plants. Also included are 503 mil-
lion liters (3,161,000 bbl) of kerosine-type jet fuel used by electric utility companies. The 1974
data include 3,759 million liters (23,646,000 bbl) of distillate #2, 526 million liters (3,307,000
bbl) of distillate #4 fuel oil used at steam electric plants and 822 million liters (5,170,000 bbl)
of kerosine-type jet fuel used by electric companies.
Includes Navy grade and crude oil burned as fuel.
Data for 1975 exclude 3,524 million liters (22,166,000 bbl) of distillate fuel oil used at steam elec-
tric plants. The 1974 data exclude 4,242 million liters (26,683,000 bbl) of distillate fuel oil used
at steam electric plants.
Quantities originally reported as thousands of barrels. Converted to millions of liters using a conversion
factor of 158.987 liters or 0.158987 cubic meters per barrel.
Percent change reported is from 1974 to.1975.
-------
DRAFT
DO NOT QUOTE OR CITE
TABLE III-IO
TOTAL AUTO FUEL CONSUMPTION--BASE CASE
(1010l/yr(109 gal/year))69/
Fuel
Gasoline
Diesel
TOTAL
1976
29.6 (78.10)
0.004 (0.01)
1985
26.2 (69.2)
1.6 ! (4.1)
2000
1/
17.3-19.1 (45.7-50.53)
6.85-7.89 (18.7-20.83)
29.6 (78.11) 27.7; (73.3) 24.4-27.0 (64.4-71.36)
AVERAGE ANNUAL GROWTH RATES IN FUEL CONSUMPTION
69/
Fuel
Gasoline
Diesel
Historical Rates (%)
1960-1975 1965-1972
4.17
4.9
Base Case
Projected Rates (%)
1976-2000
-1.8 to - 2.2
+36.8 to +37.5
1/
TOTAL
4.17
4.9
-0.4 to - 0.8
^ Higher end of range based on assumption of negligible new
car fuel economy improvement in po'ast-1985 period.
119
-------
The estimates shown in Table III-2 (P. ) of total annual
BaP emissions were calculated from these emission factors and the
consumption figures given in Tables III-9 and 111-10. The
current light duty diesel emissions of BaP are negligible with a
best estimate of 0.013 Mg/year. Heavy duty diesel emissions are
much larger, but still small, at 0.13 Mg/year for 1975.
f. Future Trends
Diesel fuel consumption is expected to rise considerably in
the 1980's due to the fuel economy of diesel engines. Estimated
1985 consumption for diesel automobiles is 1.6 x 10^ liters.
Improvements in engine design and emission controls are expected
to accompany the increased demand. Assuming that emission char-
acteristics remain the same, the estimated BaP emissions will
increase to 5.4 Mg light duty diesels and 0.21 Mg heavy duty
diesels.
3. Rubber Tire Wear
a. Process Description
Degradation of automobile tires releases carbon black par-
ticles to the atmosphere. Carbon blacks are used in tire manu-
facturing and contain POM and other high-molecular-weight organic
compounds.
b. Emission Sources
Oxidation and wear of tires on roadways degrades the rubber
material. The organic compounds in the rubber (primarily carbon
black) are oxidized by the heat of friction. Particles and gas,
which contain polycyclic aromatic hydrocarbons are continuously
released during the operation of the vehicle.
120
-------
c. Emission Controls
No methods are presently known to reduce tire degradation.
Substitutes may be found for carbon black.
d. Location and Capacity
At one time, tire consumption was dominated by original
equipment sales for new vehicles but road mileage is now so high
that tire demand is more closely related to gasoline sales than
to new car production. Marchesani, et al., as reported in the
NAS study,^^ estimates that 3.9 metric tons (4.3 tons) of
rubber particles from tires are emitted per day per million
people in the United States.
e. Emission Estimates
In one study, no POM was detected in the preliminary analy-
sis of particulate matter collected from tires run at up to 56
km/hour (35 mph) with 450 kg (1,000 lb),loads on a paved indoor
track.62// The National Academy of Science study17^ estimated
a rough emission factor for benzo(ajpyrene of 0.14 kg/day (0.3
lb/day) per million population based on the analytic data of
Falk, et al.^7//
The estimate of annual BaP emissions from rubber tire wear
is 11 Mg/year. This is a very conservative estimate based on the
. h 6 8/
population dependent emission factor'iarid the 1977 population. '
f. Future Trends
Rubber tire wear has tended to follow growth trends in the
motor fuel industry. Gasoline and diesel fuel consumption are
expected to decrease 6.1 percent by 1985. However, since this
121
-------
decrease in fuel consumption is largely due to increased mileage
per amount of fuel, the rubber tire wear will remain constant or
even increase with the total vehicle miles travelled or with
population. Using the population dependent emission factor and
the U.S. Census Series II (moderate) estimate of the total U.S.
131/
population in 1985 of 233 million, BaP emissions .are pro-
jected to increase slightly to 12 metric tons per year.
4. Motor Fuel Consumption in Two-Cycle Engines
a. Process Description
Two-cycle engines operate on a mixture of premixed oil and
gasoline. The combustion is less efficient than a four-cycle
engine primarily because the exhaust remains in the combustion
chamber after each cycle.
b. Emission Sources
The combustion of gasoline and oil yields large amounts of
benzo(a)pyrene and other POM. The emission levels are a direct
function of oil concentration in the fuel mixture. The presence
of heavy components in the fuel and the inefficient two-cycle
engine cause extensive POM formation.
c. Emission Controls
Emission controls are not commonly used on two-cycle en-
gines. Adaptation of standard controls for four-cycle engines is
feasible.
d. Location and Capacity
Fuel consumption by motorcycles in 1975 was reported to be
9
1.69 x 10 liters by the U.S. Federal Highway Administration.
< 122
-------
Figures for other two-cycle engines were not available for this
study.
e. Emission Estimates
An emission factor of 2.9 mg/1 for benzo(a)pyrene from two-
cycle motorcycle engines was derived from data collected by
Hunigen, et al. reported at the 1966 International Clean Air
Congress and cited by the NAS study.The total emissions
estimate is based on this factor and a more conservative motor-
9
cycle fuel consumption figure of 1.94 x 10 liters derived by
EEA.^Annual BaP emissions are estimated to be 5.6 metric
tons.
No estimates are available on other two-stroke engine emis-
sions .
f. Future Trends
Automobile fuel consumption is projected to decrease 6.1
percent by 1985. However, since motorcycles already have high
fuel efficiencies, it is unlikely that fuel consumption will
decrease because of increased fuel efficiency. Although the
population of the age group most likely'to use motorocycles as
their major means of transport has been decreasing, EEA has
conservatively estimated that mileage arid BaP emissions will
remain constant through 1985.
123
-------
SECTION IV
ESTIMATES OF POPULATION
EXPOSURE TO POM
A. Discussion of Alternative Estimation Techniques
1. General
In this section, some of the methods which can be used to
estimate population exposures are briefly outlined and discuss-
ed. Census data are available which give the number of people
residing in areas ranging in size from states to city blocks.
Therefore, the limiting factor for exposure estimates is gener-
ally information on the location and production characteristics
of point sources and the local consumption- of fuel or other
indicators of the sizes of area sources. This type of data is
required to estimate emissions and thus ambient concentrations.
For POM's, the unavailability and unreliability of such informa-
tion made it necessary to use an ambient air concentration
approach rather than an emissions approach. The following
sections outline the estimation techniques that were considered
and briefly discuss their advantages, disadvantages, and limi-
tations. BaP was used as a surrogate for POM in estimating
exposure, as BaP emission factors are available for most sources
and are more generally comparable than BSO or POM emission
factors. (BaP source testing results from different test series
and sources should be more comparable with each other because
BaP is a single compound for which analytical procedures can be
calibrated relatively easily and accurately. Therefore, al-
though the comparability of the results of different sampling
124
-------
procedures for BaP is as questionable as it is for BSO or POM,
the results of smapling and analysis should be more comparable
for BaP.) Also, adequate ambient air quality monitoring data
are available only available for BaP.
2. Estimation by Dispersion Modelling of Emissions Esti-
mates Derived From Local Production/Consumption Figures
The population exposure estimation technique that best con-
siders localized spatial variations in the ambient concentration
of a pollutant is one that considers the contributions of in-
dividual point and area sources. The accuracy of this technique
is limited by the accuracy of the emission factors and source
data used to estimate emissions and the reliability of the dis-
persion model and its meteorological data or assumptions.
However, this technique is still preferable to the other feasi-
ble techniques because it considers tfte.actual number of people
residing in a relatively small area and thus exposed to an
estimated concentration. The alternative technique would in-
volve assuming a uniform population density or ambient concen-
tration over an area as large as a citv or state.
EEA has developed a computer-based population exposure
estimation system for the contiguous United States. U.S. Census
data were associated with a set of several million nodes, each
representing approximately ten square:kilometers. This system
can be used to count the number of people exposed to a range of
concentrations generated by a single point source (assuming the
relationship represented by a dispersion model between the
i 1 1
emission and stack characteristics and the ambient concentra-
tions) . This system has been used to estimate the populations
exposed to individually "significant" point sources, i.e.,
sources which individually generate ambient concentrations
125
-------
greater than some "significant" level. However, when multiple
sources with overlapping effects are the situation, the limits
on available computer capacity prevent the addition of the
contributions by the various sources to the concentration at
each node when using nodes representing areas small enough to be
reasonably assigned a single level of concentration. Therefore,
for multiple sources, the results of the current system can be
used to calculate the product of population times concentration
of exposure or the total number of people exposed to at least a
"certain" concentration, but not the concentrations to which
actual populations are exposed.
The POM sources were assessed using their BaP emission
factors to determine if they individually produced "significant"
ambient concentrations of BaP. "Significant" ambient concentra-
3
tions of BaP were arbitrarily defined as 0.4 ng/m because few
non-industrial cities had greater ambient concentrations. If
any sources had been "significant," the EEA population exposure
system would have been used for these sources. However, indivi-
dual point sources (e.g., fuel combustion and industrial process
sources) were estimated to produce only marginally "significant"
ambient concentrations even when using very conservative produc-
tion and stack characteristics. These conservative plant charac-
teristics included PTMAX/hourly averages, which are often an
order of magnitude or more higher than annual averages, maximum
emission factor, largest plant size, and conservative stack
conditions--i.e., relatively low flow rates, temperatures, and
stack heights. More "typical" production and stack character-
istics such as best emission factor, controlled emissions, and
more representative plant sizes were estimated to produce less
than significant concentrations.
126
-------
The more diffuse energy consumptive sources of POM, such as
industrial, commercial, institutional, residential, or mobile
fuel users, or open burning sources of POM can be considered as
area sources. However, no reliable data on fuel consumption or
open burning is available on a local basis for the entire na-
tion. Bureau of Mines fuel consumption data are only available
on the state level, while the NEDS file has been found to have
significant anomalies in the data. Thus, the contributions of
the various area sources in a locality could not be estimated
and summed, as adequate data are generally not available.
Therefore, since no point or area sources were found to produce
individually significant concentrations (the effect of area
sources were estimated by the technique discussed in the fol-
lowing section) and the contributions of the various sources in
an area could not be adequately estimated, the technique of
counting the population exposed to an estimated concentration in
a relatively small area could not be used.
3. Estimation by Dispersion Modelling of Emissions
Proportional to NEDS Emissions
Since fuel consumption and other "production" data were not
available for specific localities on a national basis, emissions
information from the National Emissions Data System (NEDS) was
used to estimate ambient air concentrations for screening purposes.
The NEDS file contains data on the emissions of various pollu-
tants by the various source types in an Air Quality Control
Region (AQCR). These data may be based on either actual or
estimated production or consumption information which has been
aggregated from the local level. It is generally recognized
that these data are often dated and have serious inaccuracies
and anomalies; however, since it is currently the only source of
such information available on a national basis, it is frequently
127
-------
used in the hope that the aggregated values are suitable for
comparison purposes. For the purposes of this study, it was
felt that the NEDS output could be used for rough screening of
area sources to ascertain if they were likely to produce "sig-
nificant" concentrations of BaP.
The total emissions for a source type reported in a NEDS
output were used to estimate the ambient BaP concentration in an
126 12 7/
AQCR using the Hannah-Gifford urban area source model. '
The NEDS output used was an AQCR Emissions Report run by EPA on
September 7, 1977. BaP emissions from a source type were as-
sumed to be proportional to the annual particulate emissions
given in the NEDS output. BaP emissions, which may exist as
vapors until after they exit the stack, are probably not di-
rectly proportional to particulate emissions. However, particu-
late emissions should give some indication of the completeness
of the combustion. The ratio of BaP emissions to NEDS particu-
late emissions was assumed to be that of the EEA BaP emission
factor to the AP-42 particulate emission factor. Only the
AQCR1s which had the largest emissions from a source type of
those AQCR's included in the NEDS run were considered. The
annual emissions were assumed to be constant throughout the year
and to be distributed uniformly over the total urbanized area in
the major metropolitan areas in the AQCR. The urbanized areas
given in the National Functional System Mileage and Travel Sys-
12 6/
tem were used. This rate of emissions per unit area was
used to estimate an ambient concentration by using the Hannah-
Gif ford Model assuming a wind speed of 2.0 meters per second.
This type of ambient concentration estimate based on NEDS
emission data is questionable at best; however, it is adequate
for screening purposes. Generally, the assumptions involved are
conservative (i.e., lead to high ambient concentrations). All
129
-------
the emissions are assumed to be emitted from the total urbanized
area, not the AQCR as a whole or the individual urbanized areas.
However, there may be local areas with higher emissions and
concentrations because the emissions are assumed to be evenly
1
distributed. The assumed wind speed of 2.0 m/s is also extreme-
ly low for an annual average so that more typical concentrations
would be a factor of two or three lower than those estimated.
> f
In addition, the NEDS system is quite often outdated and the
largest discrepancies appear to be in the small combustion
sources such as residential coal and open burning. Though these
sources have generally decreased drastically in recent years,
these changes are often not reflected in the NEDS file.
The results of this analysis for area sources of POM showed
that the AQCR's with the greatest emissions were at most margin-
ally significant. The sources that were barely significant,
even for the worst case AQCR's and maximum BaP emission factors,
were combustion of residential coal, industrial oil, industrial
wood, commercial/institutional oil, ancl gasoline in motor ve-
hicles, industrial incinerators, auto bocly open burning, and
slash burning of wood. None of these source types were esti-
mated to produce "significant" levels of BaP (>_0.4 ng/m ) for
the other cities for which data were available or for more
likely emission factors. As a check, the gasoline-powered motor
vehicle emissions were calculated for, Los Angeles using the
daily vehicle miles travelled (DVMT) density assuming a gasoline
consumption of 6.4 km/1 (15 mpg). The calculated emission rate
per unit area was a factor of two lower than from the NEDS data.
Using the Miller-Holzworth Model as calibrated for Los Angeles,
assuming Stability Class 3, a mixing height of 300 m, and a wind
: j
speed of 2.0 m/s, the calculated ambient concentration was a
factor of three lower than the previous estimate and, thus,, was
below the "significant" level. Therefore, the emissions from
129
-------
the various area sources were presumed to individually produce
3
ambient concentrations no greater than 0.4 ng BaP/m .
4. Estimation From Ambient Air Quality Data
The estimation of population exposures from ambient air
quality data would be the preferred method if monitoring results
were available for very localized areas; however, this is far
from possible with POM's. The National Air Surveillance Network
(NASN) included sampling for BaP at approximately 120 stations
throughout the country from 1966 through 1970. BaP monitoring
has been continued at 40 sites until the present. In addition,
some states, particularly Pennsylvania (94 locations) and Mary-
land (50 locations), have recently begun to conduct BaP monitor-
ing. Other localities have, at times, also monitored for BaP
ambient concentrations. In addition, special studies, generally
regarding the effect of coke oven emissions, have measured
ambient air concentrations of BaP in various areas of interest
for a brief period. Thus, BaP sampling has been conducted at
only several hundred different sites throughout the country at
any time. At many of these sites, no data are available for
recent years. And many large areas of the country have never
been monitored for BaP. In addition, the comparability of the
various results is questionable between different sampling
techniques, organizations, and even between different years for
the same technique and organization.
Although the spatial distribution of the monitoring sites
does not provide for an accurate estimate of the concentrations
to which various local populations are exposed, these data and
approach had to be used because no better approach was feasible.
The results of a recent study on the population exposures to
coke oven emissions by Suta"*"^^ were used to estimate the popu-
lation exposure to BaP in cities where coke ovens were located.
130
-------
Population exposure is expressed in terms of the product of the
population and annual average ambient concentration. The esti-
mated concentrations and exposed populations given by Suta were
used directly, while the background levels given were used for
any remaining population in the Standard Metropolitan Statisti-
cal Areas (SMSA's) which were assumed to be affected by the coke
oven emissions. The affected areas were assumed to be the
SMSA in which the coke oven is located and neighboring SMSA's in
the direction of the assumed prevailing wind if more people were
reported to be exposed than reside in the coke oven SMSA.) For
non-coke oven areas, the ambient BaP concentrations were extra-
polated to "197 5" unless single year data were available for
1974, 1975, or 1976. A common year was used because ambient BaP
concentrations have been noted to be decreasing with time.^^
No significant relationship was found between these "1975"
concentrations and the population density or population of the
cities in which urban samples had been taken. Therefore, the
populations in non-coke oven areas were assumed to be exposed to
the "1975" ambient BaP concentration, if available, or to na-
tional average concentrations for large urban areas, smaller
cities and towns, and rural areas. The details of this esti-
mation procedure and the results are given in the following
section. The results give the populations and population ex-
posures due to coke ovens and all sources for the states and the
District of Columbia.
B. Analysis and Results of the Ambient Concentration Technique
For the reasons discussed in the previous section, popula-
tion exposures were estimated from the available ambient air
BaP concentrations for areas without coke ovens and from the
118 /
results of a recent study for areas with coke ovens. This
method was the best feasible even though the ambient air has
131
-------
been monitored for BaP at only several hundred sites in the
country at any time. As a decreasing trend in BaP ambient con-
centrations had been demonstrated at most stations over recent
years, t^e 3ata from various years were estimated for a
common year, "1975." The year "1975" was chosen as a common
year to reflect these decreasing trends and because much of the
state and local data were available for only 1974, 1975, or
1976. These values were used directly as were the annual aver-
age BaP concentrations for 1975 for other sites. For monitoring
locations without recent data, the concentrations from earlier
years were extrapolated to 197 5. Since the ambient BaP concen-
trations must be asymptotically approaching the background or
zero level, the concentration was extrapolated using a regres-
sion of the logarithm of the concentration versus time.
The data from many stations showed highly significant
2
decreasing trends with coefficients of determination (R 's) as
high as 0.97. However, the data at other stations showed little
2
variation over time, and thus, very low R 's (i.e., £0.01). For
several locations with only a few data points, a high concentra-
tion in one of the last years in which samples were taken would
cause the regression line to slope upward, i.e., to show an
increasing trend in ambient BaP concentrations over time. This
situation usually occurred for the NASN sites that were last
sampled for BaP in 197 0. Since the stations where sampling had
been continued generally showed decreasing or constant trends
with time (presumably due to the decreased use of inefficient
combustion sources of BaP such as residential coal use and open
burning), a value of concentration approximately equal to the
maximum ambient BaP concentration in any sample year was se-
lected for the "1975" value.
132
-------
For non-coke oven cities, the benefit of stratifying the
"1975" annual average BaP concentrations with respect to popula-
tion density or population was explored; however, the advantages
of this approach were found to be negligible at best. Popula-
tion density and population figures taken from census data for
99 /
urbanized areas ' were chosen as surrogate measures of fuel
consumption and other consumptive sources of BaP. Regressions
between the ambient BaP concentrations for "1975" and the ur-
banized area,population densities and populations of cities
2
without coke ovens showed no significant relationship. (The R
was less than 0.022 for the population density of 92 areas
versus the "1975" BaP concentration and less than 0.011 for the
population of 24 areas.) The non-coke oven cities were grouped
by ranges of population density and the means and standard
deviations of the ambient annual average BaP concentrations for
"1975" were calculated. No discernible trend of concentration
with respect to population density could be found. (The calcu-
3
lated values were 1.30 + 1.24 ng/m for 13 areas with more than
3
4,000 people per square mile, 1.20 + 1.09 ng/m for 27 areas
3
with 3,000 to 4,000 people per square mile, 0.785 + 0.611 ng/m
for 22 areas with 2,500 to 3,000 people per square mile, 1.21 +
3
1.21 ng/m for 19 areas with 2,000 to 2,500 people per square
3
mile, and 1.70 + 3.26 ng/m for 11 areas with 1,000 to 2,000
people per square mile.)
Because no significant relationships could be found between
the ambient air BaP concentrations and the most likely surro-
gates of consumption, a national average BaP concentration for
all non-coke oven cities of a certain size was deemed to be the
best measure available. Using all the data that were found for
non-coke oven cities greater than 25,000 population (96 areas),
3
a concentration of 1.06 + 1.00 ng/m was calculated. As the
coke oven population exposure study by the Stanford Research
133
-------
TABLE IV-1
State
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
District of Columbia
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
POPULATIONS EXPOSED AND EXPOSURES TO BaP
SRI Coke Oven Study BaP Exposures
Concentrations > SRI
Backgrounda//
Population
777,161
223,716
52
2,183,086
673,021
52,376
117,308
1,690,193
70,831
261,271
Exposure
(1000 People
ng/m3)
1,500
480
0.047
4,200
3,900
210
240
4,300
60
180
Concentrations > SRI
Background^/
Population
1,303,328
9,314,764
103,300
21,822
7,247,022
732,461
52,376
1,744,574
4,054,374
1,889,215
1,620,070
Exposure
(1000 People
ng/m3)
1,700
11,000
62
8. 7
7,800
3,100
210
1,500
6,900
770
660
EEA Estimated Exposures0/
Exposure
(1000 People
ng/m3)
Population
3,444,165
300,382
1,770,900
1,923,295
19,953,134
2,207,259
3,031,709
548,104
756,510
6,789,443
4,589,575
768,561
712,567
11,113,976
5,193,669
2,824,376
2,246,578
3,218,706
3,641,306
992,048
3,922,399
5,689,170
8,875,083
3,804,971
2,216,912
4,676,501
694,409
1,483,493
488,738
2,500
120
2,100
890
17,000
3,000
2,300
430
540
4,600
2,600
96
390
9,900
6,600
1,200
1,300
3,000
1,500
450
2,500 g
5,600
9,600 g
1,600
1,100 ?
2,600 CT
390 O
600
740 m '
o "
=0
-------
TABLE IV-1 (CONTINUED)
POPULATIONS EXPOSED AND EXPOSURES TO BaP
SRI Coke Oven Study BaP Exposures
Concentrations > SRI
Background5^
Concentrations > SRI
Background^/
State
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
Population
728,428
129,213
5,981,222
23,263
458,507
31,950
73,841
267,400
Exposure
(1000 People
ng/nP)
1,500
5,800
16,000
20
240
43
340
260
Population
1,148,111
3,866,970
6,682,580
205,233
1,829,885
120,554
213,933
1,339,006
Exposure
(1000 People
ng/m^)
2,100
6,700
16,000
93
300
87
430
1,000
U.S. TOTAL
13,742,839
39,000
43,489,578
61,000
EEA Estimated Exposures0/
Exposure
(1000 People
Population
ng/m3)
737,681
440
7,168,164
5,800
1,016,000
530
18,236,967
20,000
5,082,059
2,800
617,761
450
10,652,017
13,000
2,559,229
1, 300
2,091,385
2,100
11,793,709
19,000
946,725
740
2,590,516
1,600
665,507
330
3,923,687
2,400
11,196,730
9,200
1,059,273
1,200
444,330
150
4,648,494
2,700
3,409,169
1,700
1,744,237
1,000
4,417,731
1,400
322,416
140
203,211,296
170,000
Populations and exposures for concentrations greater than the SRI background are calculated directly from the
exposed populations and average ambient BaP concentrations greater than the SRI-assumed background concentrations
given in the SRI studyll®/ for each range of distances from a coke oven. If the population in the area within
-------
TABLE IV-1 ^CONTINUED)
POPULATIONS EXPOSED AND EXPOSURES TO BaP
Footnotes Continued
the state in which the coke oven was located was less than the SRI population exposed value, the excess
population exposed was counted in adjoining areas, sometimes located in adjacent states.
Populations and exposures for concentration greater than or equal to the SRI background include the population
exposed to concentrations directly attributed by SRI to coke oven emissions, as noted above, and, in addition,
the remaining populations in the exposed area counted as exposed to the SRI-assumed background levels. The
exposed area was determined from the number of people estimated by SRI to be exposed to the emissions attributable
to coke ovens or to background concentrations, i.e., those considered to be within 15 kilometers of a coke oven.
Generally, an exposed area was considered to contain the urban population of the SMSA in which a coke oven was
located.
EEA estimated population exposures include the estimates developed from the SRI results, as noted above.
Remaining populations were counted as exposed to actual or extrapolated "1975" ambient BaP monitoring data
if available for a specific area or to national average ambient BaP concentrations developed in this study
(1.1 ng/m-* for urban populations within SMSA's, 0.86 ng/m^ for urban populations outside of SMSA's, and 0.15
ng/m^ for rural populations).
-------
Institute (SRI) had given a value of 0.38 ng/m^ as the average
of the 1975 data from NASN sites in cities without coke ovens
(13 sites), it was suspected that the extrapolated values might
be causing the discrepancy. Therefore, a further check was made
by calculating a mean for the 33 non-coke oven cities either
with one monitor (generally NASN sites) or with area averages of
several monitors (generally PA or MD sites) for either 1974,
1975, or 1976. The calculated ambient BaP annual average con-
3
centration was 1.13 + 1.06 ng/m . Therefore, in calculating
population exposures for this type of area in which no BaP
3
sampling had been done, a BaP concentration of 1.1 ng/m was
used. Similarly, a value of 0.861 + 1.041 ng BaP/m"* was calcu-
lated, from the 17 data points for town and cities of 10,000 to
50,000 population which were not in an urbanized area. Also, a
3
value of 0.153 + 0.166 ng/m was calculated for the 21 data
points in parks or other rural locations. Thus, values of 0.8 6
3
and 0.15 ng/m were used for these types of areas, respectively.
The rough estimates of population exposure to BaP calcu-
lated from the SRI coke oven study results alone and for non-
coke oven areas using the average concentrations calculated in
this study are given in Table IV-1 for the states and the Dis-
trict of Columbia. The estimates of population exposure are
reported as the product of the number of people in an area times
the estimated ambient BaP concentration to which they are ex-
posed divided by 1,000. The population figures used were taken
98/
from the 197 0 census. The populations and average exposure
concentrations for BaP concentrations due to coke oven emissions
118/
were taken directly from the SRI study. / The numbers of
people exposed within certain ranges of radii are given by
Suta.^^ For each range, an average concentration is given or
the concentration is noted to be the background concentration.
137
-------
SRI used background concentrations developed from monitoring
results varying from those for nearby sites to as general a
number as a statewide average. Although more site-specific
background levels were sometimes available, the SRI background
levels were used for the populations in coke oven areas not
exposed to greater concentrations. Thus, the population ex-
posure estimates for coke oven areas from this study aire reason-
ably consistent with those of the SRI study. Generally, the
population estimated by SRI to be exposed to coke ovens or their
background levels were less than the urban population in the
Standard Metropolitan Statistical Area (SMSA) and state in which
the coke ovens were located. Therefore, the urban population
not exposed to concentrations due to coke oven emissions was
assumed to be exposed to the background concentration. In some
cases, the total urban and rural population of the SMSA had to
be used. For some areas, the population reported by SRI to be
exposed to the concentrations attributable to coke ovens or to
the background levels was greater than the total population in
the SMSA or in the part of the SMSA within the state in which
the coke ovens were located. If this were so/ the remaining
exposed population was counted in the part(s) of the SMSA in
other states, in adjacent SMSA's in the same or adjacent states,
or in the rural population of adjacent areas> depending on the
locality and the presumed prevailing wind direction in the area.
The people in each state which were not counted as exposed
to coke oven generated concentrations or their background levels
were then counted at estimated concentrations of exposure. For
cities without coke ovens where ambient BaP sampling had been
conducted, an actual (1974, 1975, or 1976) BaP concentration or
an extrapolated concentration (from data for previous years) BaP
concentration for "1975" was used. The calculated national
138
-------
average concentrations were used to estimate the population
exposures for the areas not affected by coke ovens where sampl-
ing had not been conducted. The concentration and population
3
categories used for each state were: (1) 1.1 ng/m for the
3
urban population within SMSA's; (2) 0.8 6 ng/m for the urban
population outside SMSA's; and (3) 0.15 ng/m for the rural
population.
The results of this very approximate estimation procedure
are given in Table IV-1. The estimates do not consider local
variability of ambient BaP concentrations because data are not
available to assess the variability either on a basis of ambient
air monitoring or on a basis of local emissions estimates.
However, these estimates should give some indication of where
higher exposures to BaP occur and the relative importance of
coke ovens. Generally, the greater exposures are estimated to
occur in those states where the population is rather large or
where there are coke ovens located near population centers.
Assuming that the estimates of exposure concentrations used in
this study were correct, the population weighted national aver-
3
age BaP exposure concentration would be 2.8 ng/m for people
exposed to concentrations directly attributable to coke ovens.
Similarly, for all people in coke oven areas, i.e., those ex-
posed to the coke oven-caused levels or background levels de-
veloped by SRI, and for everyone in the country, the national
3
average BaP exposure concentrations would be 1.4 ng/m and 0.8 5
3
ng/m , respectively. These estimates and those reported for
population exposures are very approximate because of both the
data and the rough averaging techniques on which they are
based. However, significant refinement of these estimates would
require greatly increased and improved monitoring data and/or
consumption and emissions data on a local level for the entire
nation.
139
-------
SECTION V
DISCUSSION OF THE STATE-OF-THE-ART AND
RESEARCH RECOMMENDATIONS
A. Discussion of Sampling and Analysis Techniques
Numerous POM sampling and analysis techniques have been
used by various groups at various times; therefore, the compara-
bility of results is questionable. Reviews of the standard
techniques are given in the National Academy of Science study"*"^
and the more recent EPA Scientific and Technical Assessment
Report"'""'""''^ on particulate POM. More recently, techniques have
been developed to include more of the vaporous POM. The general
sampling and analysis techniques and their comparability are
outlined briefly in the following paragraphs.
The major POM sampling techniques that have been used are
EPA Method 5, modified Method 5 high volume samplers, and adsor-
39 44 128/
bent samplers. ' ' The most extensive work on POM's,
39/
reported in AP-33, ' used a Method 5 sampling train including a
heated filter to collect particles followed by ice-bath imping-
ers to condense vapors. Thus, this method collected some, but
probably not all, vapors and most of the particles. Sampling
results were reported as total POM collected and measured per
sampled volume, and thus, per energy or material input. Modi-
fied Method 5 high volume samplers have been used in order to
sample open sources such as open burning, especially in research
facilities where the burning is done in fairly large enclosures.
Also, high volume samplers may be used to increase the quantity
of sample available for analysis so that lower concentrations of
POM species in air can be detected by a given analytical technique.
140
-------
With the increased indications that the vaporous POM may be
the larger fraction, adsorbent samplers have been developed to
collect vapors with an increased efficiency. Generally, a
heated filter is followed by a condenser and a resin adsorbent,
such as Tenax, Chromosorb, or XAD-2. Tenax samplers as develop-
44/
ed by Battelle ' have been most extensively studied to date;
however, it is likely that XAD-2 will be used for the new EPA
Source Assessment Sampling System (SASS). This system, when
fully developed, will have the capability for POM sampling and
will routinely collect an adequate sample for organics analysis.
The Tenax adsorbent system has been shown in the lab to recover
80 to 115 percent of POM's placed in the stream even in the
presence of sulfur or nitrogen oxides or other additives. Field
4 3/
testing, to date, has demonstrated good reproducibility.
However, in comparison runs with .Method 5 or Method 5 high
volume samplers, the adsorbent system combination of sampling
and analysis techniques has measured POM emissions at least an
43 44/
order of magnitude greater. '
The number of analysis techniques that have been used is
even greater than that of sampling techniques because an analy-
sis technique generally includes extraction, separation, and
analytical measurement steps. The early work by Hangebrauck, et
3 9/
al. used benzene extraction, separation by column chromato-
graphy, and analyses by ultraviolet spectrophotometry to measure
quantities of ten POM species. Other solvents, e.g., dichloro-
methane or methylene chloride, have been used in other studies,
while separation techniques used have included thin-layer,
liquid, and gas chromatography. POM's have also been analyzed
using fluorescence, flame ionization, or mass spectrometry,
generally in combination with liquid or gas chromatography.
Various research groups have measured from six to 23 POM species.
141
-------
Mass spectrometry-computer systems, which are among the most
commonly used systems and are growing in usage, can identify
nearly all known POM's with reasonable accuracy and sensitivity
(on the order of nanograms per sample for most systems). Known
POM's are those species which have been individually identified,,
characterized, and cataloged (by its spectral peaks) in the com-
puter system. Many other POM species probably exist; however,
some of these will be measured by some systems as the computers
scan for "typical POM molecular weights" within the POM range of
molecular weights.
The comparability of results from different combinations of
sampling and analysis techniques is not quantitatively known at
present; however, it presumably varies with the conditions and
44/
may vary by as much as an order of magnitude. The major dif-
ferences are probably caused by -the differences in the sampling
collection efficiency for POM vapors. Theoretically, most POM
species are vaporous at typical power plant stack conditions.
Field work has shown this to be the case for most stacks
and processes, so that the differences in results for different
vaporous collection efficiences may be quite significant. Also,
when it condenses as a particle, or adsorbs: onto a particle, a
particular POM species may either stabilize so that it is less
susceptible to degradation by ultraviolet light (e.g., BaP) or
destabilize to become another POM species (e.g., fluorene to
fluorenone). Since such interspecies transformations occur,
stack samples including vapors may measure most POM; however,
the quantities of individual species will generally be different
than those measured in the plume after adsorption of the vapor-
ous POM onto particles . ^
Another complication is that the polynuclear aromatic
hydrocarbons (PNAH) have been the focus of research efforts and
14 2
-------
thus are readily measured. If they destabilize, PNAH are gen-
erally transformed into quinones or other POM species, which are
often not detected. It is generally thought that the results
from ultraviolet spectrophotometry analysis of EPA Method 5
train samples, the results from gas chromatography-mass spectro-
metry-computer analysis of adsorbent train samples, and the
actual quantity present in emissions differ by an order of mag-
43 64/
nitude or less. ' However, a POM species may be transformed
64 119/
to another species when adsorbed onto a particle ' or when
132 133 138/
reacted with gases (e.g., NO2O2, or PAN) in the atmosphere. ' '
Therefore, the POM in the ambient atmosphere may be significantly
different, both in quantity and character, from what would be
suggested by emissions testing results.
B. Current Studies and Research Recommendations
Due to the renewed and increased interest in atmospheric
POM, many current studies are investigating the sampling, emis-
sions, transformations, and health effects of airborne POM. As
most of these studies involve basic research, or laboratory or
field testing, the quantity and quality of results that will be
achieved within a given time cannot be predicted. Therefore,
this section only outlines briefly the areas of concentration,
general objectives, and tentative timetables of current studies
regarding the topic of this study, the emissions, control, and
population exposure of POM. The studies noted are not a com-
prehensive listing of all current work involving the study of
POM; however, most of the major studies involving emissions, con-
trol, and population exposure should be included.
The following paragraphs outline the current studies and re-
search recommendations regarding POM in the areas of exposure es-
timation, sampling, and analysis, stationary sources, and mobile
sources. Some general comments may be made about these areas
143
-------
and other areas less directly related to POM emissions and ex-
posure. Work is proceeding in parts of most of the areas where
further research is required. However, discussion may be of
value in outlining the scope of some of the current studies and
thu, in pointing out the needs for future research.
In order to improve the estimation of population exposure,
either ambient concentration data or source emissions and produc-
tion or consumption data must be increased and improved. No
major federal effort is known to exist that will increase the
number of BaP, or other POM, monitors. As a multitude of local
monitors would be required to assess the exposures of local pop-
ulations (assuming they stayed in an area of constant exposure
concentration), such a massive effort is not recommended. Some
success is being achieved with the calibration of BaP emissions
modelling to the data from five ambient monitors in the three-
139/
county Detroit area. ' (This model will be used to estimate
exposure levels in the past for epidemiological studies.) An
EPRI-sponsored study is analyzing the organic matter, including
PAH in New York City total suspended particulate samples and in
investigating seasonal and long-term trends in order to attribute
ambient concentrations to sources.^^ These types of studies
examining the relationship between emissions and exposure concen-
trations for relatively small land areas could greatly increase
the understanding of the mechanisms involved and improve the
quality of such exposure estimates. As some states (e.g., Pen-
nsylvania and Maryland) are conducting relatively widespread
monitoring of BaP, the results of such monitoring could improve
the development and calibration of exposure concentration esti-
mation from data on local sources.
The data base for emissions modelling requires improvement
in two major areas:, the quality and representativeness of emis-
sion factors for all potential sources of POM and the quality
144
-------
and availability of consumption or production data for local
point and area sources. Much work is being done that should im-
prove the available emission factors; it is outlined in the fol-
lowing paragraphs. Most of the available point source data is
based on the National Emissions Data System (NEDS). This data
base is generally recognized to have major short-comings and
anamolies; however, no major effort is being planned to update
and correct the data for the entire nation. Improve data are
being collected for some areas. Fuel consumption and other are
source data are generally available only on the state level.
An NSF-sponsored study is collecting data on personal energy
consumption for transportation by county for selected SMSA's
117/
which could be expanded to the nation within about two years. '
Efforts to improve and standardize sampling and analysis
of POM are being funded by EPA in conjunction with the Electric
Power Research Institute (EPRI) for stationary sources and in con-
junction with the Coordinating Research Council (CRC) for mobile
sources. Most of the studies involve laboratory and field test-
ing and validation of the Source Assessment Sampling System
43 53 144/
(SASS) ' ' ' This system will routinely collect enough
sample for analysis of organics using a gas chromatograph-mass
spectrometer as the system will have an adsorbent, probably
XAD-2. EPRI is funding work by Oak Ridge National Lab to deter-
mine whether hopper ash POM can be used as an indicator of or-
ganics adsorbed in fly ash.^^ There is little concern about
gasoline emissions as catalytic converters control POM effective-
ly; however, the comparability and accuracy of the wide variety
141/
of diesel exhaust sampling and analysis techniques 7 are ques-
tionable. EPA is conducting some in-house work on organics,
but are concentrating on the oxgenated fractions, and are sponsor-
142/
ing some research on PNA analysis. The CRC has sponsored a
round-robin on sampling techniques, the results of which suggest
145
-------
that scintillation counters and tracers should be used; how-
ever, this is an expensive technique which is not preferred by
142/
EPA. As analysis procedures are being improved, is condier-
ing funding projects concerning the sampling of particulate die-
sel emissions. As these fine particles may undergo chemical and
physical transformations, it is not certain that adequate
143/
methods can be developed. ' The development, validation, and
usage of standardized POM sampling and analysis techniques with
comparable results should be continued.
In addition to the field validation of the SASS train the
other major development in stationary source emissions data is
the Fine Particle Emissions Information System (FPEIS). This
new EPA data system will include a significant amount of data
42/
on organics within a year. ' All SASS train sampling results
will be routinely put into the system. The planned field vali-
dation for the SASS train and other POM sampling will include
testing of (date results expected) utility (5/79), industrial
(5/80), commercial/institutional (9/79), residential (9/78),
and internal combustion (11/7 8) stationary combustion sources
S 6 /
by TRW, ' coal-fired stoker boilers by the American Boiler In-
stitute (6/78, 6/79) coal-fired utilities, industrial
boilers, and residential furnaces (1978) by Monsanto Research
Corporation,14^^ pushing (1979), quenching, leaks, stacks, and
by-product plants (1978) ,21*130/ an(^ coa;L_fired utilities to
assess control efficiency (1978, 1979) by KVB (for EPRI)."^^
147/
Another EPA study may measure POM from waste oil incineration.
Although the results of these studies should improve the quantity
and quality of POM emissions data, its representativeness will
still be questionable as specific sources have different designs,
operation, and maintenance. Therefore, source sampling should
be continued in order to more thoroughly investigate these ef-
fects and the efficiencies of various types of control equipment
for POM.
146
-------
The emissions from buring coal refuse banks and forest fires
should also be investigated. The on-going MESA study should
delineate the magnitude of the problem of emissions from burn-
ing refuse banks in that it will locate them and make general
79/
observations about emissions. The U.S. Forest Service intends
78/
to routinly sample for POM in its combustion experiments.
These efforts will improve the quality of the existing data
base; however, due to the variability and uncertainty involved,
much more work is recommended.
As stated previously, the studies in progress concerning the
generation of POM by mobile sources are investigating the sampl-
ing and analysis of diesel emissions. Some results are forth-
coming from the Department of Energy (DOE) on POM emissions from
135/
both stationary and automobile diesel engines. The focus
of current research should continue to be the development of re-
liable, standardized and comparable sampling and analysis pro-
cedures for diesel emissions.
There are several other research needs that could be cri-
tical. The identification and assessment of emissions from po-
tential sources and emission points of POM should be continued.
The chemical and physical transformations of POM in the stack,
plume, and atmosphere must be assessed in much greater detail
before the exposures to a particulate species of POM attributable
to specific emission sources can be estimated with any degree of
certainty.
147
-------
APPENDIX A
COAL CONSUMPTION BY STEAM ELECTRIC PLANTS IN 1975
(25 megawatts or greater)
Total Consumption
Location Number of Plants (1,000 Mg(l,000 ton))
NEW ENGLAND
Connecticut
0
0
( 0)
Maine
0
0
( 0)
Massachusetts
3
729
( 804)
New Hampshire
1
882
( 972)
Rhode Island
0
0
( 0)
Vermont
1
12
( 135 •
TOTAL
5
1,623
( 1,789)
MIDDLE ATLANTIC
New Jersey
5
2,041
( 2,250)
New York
10
5,557
( 6,125)
Pennsylvania
27
33,227
( 36,626)
TOTAL
42
40,824
( 45,001)
EAST NORTH CENTRAL
Illinois
25
29,237
( 32,228)
Indiana
26
24,704
( 27,231)
Michigan
26
18,800
( 20,723)
Ohio
34
42,117
( 46,426)
Wisconsin
18
8,814
( 9,716)
TOTAL
129
123,671
(136,324)
WEST NORTH CENTRAL
Iowa
23
4,437
( 4,891)
Kansas
6
2,707
( 2,984)
Minnesota
16
6 ,650
( 7,330)
Missouri
18
16,054
( 17,696)
Nebraska
5
1,156
( 1,274)
North Dakota
5
3,786
( 4,173)
South Dakota
3
1,477
( 1,628)
TOTAL
76
36,266
( 39,976)
143
-------
APPENDIX. A (Continued)
COAL CONSUMPTION BY STEAM ELECTRIC PLANTS IN 1975
122/
(25 megawatts or greater)
Total Consumption
Location Number of Plants (1,000 Mg(1,000 ton))
SOUTH ATLANTIC
Delaware
2
86.4
( 952)
District of Columbia
1
101
( HI)
Florida
5
5,223
( 5,757)
Georgia
7
11,474
( 12,648)
Maryland
5
3,512
( 3,871)
North Carolina
13
16,507
( 18,196)
South Carolina
9
3,993
( 4,402)
Virginia
7
3,619
( 3,989)
West Virginia
12
23,411
( 25,806)
TOTAL
61
68,703
( 75,732)
EAST SOUTH CENTRAL
Alabama
10
15,694
( 17,300)
Kentucky
18
20,289
( 22,365)
Mississippi
2
1,278
( 1,409)
Tennessee
8
1,401
( 18,848)
TOTAL
38
54,360
( 59,922)
WEST SOUTH CENTRAL
Arkansas
0
0
( 0)
Louisiana
0
0
( 0)
Oklahoma
0
0
( 0)
Texas
2
8,205
( 9,044)
TOTAL
2
8,205
( 9,044)
MOUNTAIN
Arizona
2
3,864
( 4,259)
Colorado
9
5,151
( 5,678)
Idaho
0
0
( 0)
Montana
3
987
( 1,088)
Nevada
2
4,022
( 4,434)
New Mexico
2
6,712
( 7,399)
Utah
4
1,831
( 2,018)
Wyoming
5
6,273
( 6,915)
TOTAL
27
28,840
( 31,791)
149
-------
APPENDIX A (Continued)
COAL CONSUMPTION BY STEAM ELECTRIC PLANTS IN 1975
122/
(25 megawatts or greater)
Location
Number of Plants
Total Consumption
(1,000 Mg(1,000 ton))
PACIFIC
California
Oregon
Washington
TOTAL
0
0
1
0
0
3,637
( 0)
( 0)
( 4,009)
3,637 ( 4,009)
UNITED STATES
TOTAL
381
366,129 (403,588)
150
-------
APPENDIX B
LOCATION, TYPE, AND CAPACITY OF PETROLEUM
71/
CATALYTIC CRACKING FACILITIES
Company and Location
CALIFORNIA
Process
b/
Charge Capacity (cubic meters
per stream day)a^
Fresh Feed
Recycle
Atlantic Richfield Co.
Carson
FCC
9,100
1,300
Chevron U.S.A., Inc.
El Segundo
Richmond
FCC
FCC
7,500
8,700
1,400
790
Exxon Company
Benicia
Gulf Oil Company
Santa Fe Springs
Lion Oil Company (Tosco)
Bakersfield
Martinez
Mobil Oil Corporation
Torrance
FCC
FCC
TCC
FCC
FCC
7,200
2,100
1,900
7,500
9,200
2,100
50
None
2,200
None
Powerine Oil Corporation
Santa Fe Springs
Shell Oil Company
Martinez
Wilmington
FCC
FCC
FCC
1,800
7,300
5,600
50
6,400
790
Texaco, Inc.
Wilmington'
c/
Union Oil Company of California
Los Angeles
COLORADO
FCC
FCC
4,450
7,150
NR
1,100
Asamera Oil (U.S.), Inc.
Commerce City
FCC
1,200
60
151
-------
UKAFT
DO NOT QUOTE OR
APPENDIX B CONTINUED
LOCATION, TYPE, AND CAPACITY OF PETROLEUM
CATALYTIC CRACKING FACILITIES71/
Charge Capacity (cubic meters
per stream day)a/
b/
Company and Location Process Fresh Feed Recycle
COLORADO (Continued)
Continental Oil Company
Denver FCC 2,400 160
DELAWARE
Getty Oil Corporation, Inc.
Delaware City FCC 9,900 2,400
HAWAII
Chevron U.S.A., Inc.
Barbers Point FCC 3,000 480
ILLINOIS
Amoco Oil Corporation
Wood River FCC 6,000' 640
Clark Oil & Refining Corp.
Blue Island FCC 4,100 160
Hartford FCC 4,100 160
Marathon Oil Company
Robinson FCC 5,800 1,300
Mobil Oil Corporation
Joliet FCC 15,000 NR
Shell Oil Company
Wood River FCC 15,000 None
Texaco, Inc. .
Lawrenceville0 FCC 5,400 NR
Lockport FCC 4,800 NR
Union Oil Company of California
Lemont FCC 8,700 1,300
152
-------
APPENDIX B CONTINUED
LOCATION, TYPE, AND CAPACITY OF PETROLEUM
CATALYTIC CRACKING FACILITIES71^
Company and Location
INDIANA
Process
b/
Charge Capacity (cubic meters
per stream day)a/>
Fresh Feed
Recycle
Amoco Oil Company
Whiting FCC
Energy Cooperative, Inc.
East Chicago FCC
Indiana Farm Bureau Cooperative
Association, Inc.
Mt. Vernon FCC
22,000
7,600
1,000
790
320
NR
Rock Island Refining Corp.
Indianapolis
FCC
2,500
None
KANSAS
Apco Oil Corporation
Arkansas City
FCC
1,500
130
CRA, Inc.
Coffeyville
Phillipsburg
Derby Refining Company
Wichita
FCC
FCC
TCC
2,500
1,400
1,700
240
130
270
E-Z Serve
Shallow Water
TCC
870
NR
Getty Oil Company
El Dorado
FCC
4,900
2,700
Mobil Oil Corporation
Augusta TCC
National Cooperative Refinery
Association
McPherson FCC
3,300
3,200
320
160
153
-------
APPENDIX B CONTINUED
LOCATION, TYPE, AND CAPACITY OF PETROLEUM
CATALYTIC CRACKING FACILITIES71^
Company and Location
KANSAS (Continued)
Pester Refining Company
El Dorado
Phillips Petroleum Co.
Kansas City
KENTUCKY
Process
b/
FCC
FCC
Charge. Capacity (cubic meters
per stream day)a/
Fresh Feed
1,700
5,100
Recycle
80
2,500
Ashland Petroleum Co.
Catlettsburg FCC
Louisville Refining, Division
of Ashland Oil, Inc.
Louisville FCC
8,600
1,600
160
NR
LOUISIANA
Cities Service Oil Co.
Lake Charles
Continental Oil Company
Lake Charles
FCC
TCC
19,900
4,300-
3,200
790
Exxon Company
Baton Rouge
Good Hope Refineries, Inc.
Metairie
FCC
FCC
26,900
2,500
None
None
Gulf Oil Company
Alliance Refinery, Belle Chasse FCC
12,000
370
Murphy Oil Corporation
Meraux
FCC
1,670
80
Shell Oil Company
Norco
Tenneco Oil Company
Chalmette
FCC
FCC
16,000
3,500
320
NR
154
-------
APPENDIX B CONTINUED
LOCATION, TYPE, AND CAPACITY OF PETROLEUM
CATALYTIC CRACKING FACILITIES 71/'
Company and Location
Process
b/
Charge Capacity (cubic meters
per stream day)S//
Fresh Feed
Recycle
LOUISIANA (Continued)
Texaco
Convent
FCC
11,000
NR
MICHIGAN
Dow Chemical U.S.A.
Bay City
Marathon Oil Company
Detroit
Total Petroleum, Inc.
Alma
TCC
FCC
FCC
950
4,000
2,100
320
720
240
MINNESOTA
Continental Oil Company
Wrenshall
Koch Refining Company
Rosemount
FCC
FCC
1,500
7,000
80
160
Northwestern Refining Co.,
Division of Ashland Oil Co.
St. Paul Park
FCC
3,300
240
MISSISSIPPI,
Amerada-Hess Corporation
Purvis
Chevron U.S.A., Inc.
Pascagoula
MISSOURI
TCC
FCC
2,300
8,900
NR
320
Amoco Oil Company
Sugar Creek
FCC
6,500
1,900
155
-------
APPENDIX B CONTINUED
LOCATION, TYPE, AND CAPACITY OF PETROLEUM
71/
CATALYTIC CRACKING FACILITIES '
Company and Location
MONTANA
Process
b/
Charge Capacity (cubic meters
per stream day)a/
Fresh Feed
Recycle
Cenex
Laurel
FCC
1,800
480
Continental Oil Company
Billings
FCC
2,200
1,100
Phillips Petroleum Company
Great Falls
FCC
290
190
NEBRASKA
CRA, Inc.
Scottsbluff
FCC
380
80
NEW JERSEY
Chevron U.S.A., Inc.
Perth Amboy
FCC
4,800
1,300
Exxon Company
Linden
FCC
21,500
3 ,200
Mobil Oil Corporation
Paulsboro
TCC
4,000
None
Texaco, Inc.
Westvillec/
FCC
6,400
NR
NEW MEXICO
Navajo Revining Company
Artesia
FCC
830
250
Shell Oil Company
Ciniza
FCC
1,100
570
NEW YORK
Ashland Petroleum Company
North Tonawanda
FCC
3,500
None
156
-------
APPENDIX B CONTINUED
LOCATION, TYPE, AND CAPACITY OF PETROLEUM
CATALYTIC CRACKING FACILITIES71/'
Company and Location
Process
b/
Charge Capacity (cubic meters
per stream day)a/
Fresh Feed
Recycle
NEW YORK (Continued)
Mobil Oil Corporation
Buffalo
TCC
3,000
950
NORTH DAKOTA
Amoco Oil Company
Mandan
FCC
3,700
1,700
OHIO
Ashland Petroleum Company
Canton
FCC
3,890
120
Gulf Oil Company
Cleves
Toledo
FCC
FCC
2,900
3,150
320
320
Standard Oil Company of Ohio
Lima FCC
Toledo FCC
5,990
8,350
1,200
3,000
Sun Petroleum Products Co.
Toledo
FCC
7,900
1,200
OKLAHOMA
Apco Oil Corporation
Cyril
FCC
1,100
270
Champlin Petroleum Co.
Enid
FCC
3 ,000
50
Continental Oil Company
Ponca City
FCC
7,000
2,100
Hudson Refining Company
Cushing
FCC
1,100
480
157
-------
APPENDIX B CONTINUED
LOCATION, TYPE, AND CAPACITY OF PETROLEUM
CATALYTIC CRACKING FACILITIES71/
Company and Location
OKLAHOMA (Continued)
Kerr-McGee Corporation
Wynnewood
OKC Refining, Inc.
Okmulgee
Sun Petroleum Products Co.
Duncan
Tulsa
Process
b/
FCC
TCC
FCC
FCC
Charge Capacity (cubic meters
per stream day)a/
Fresh Feed
1,830
1,300
4,000
4,800
Recycle
320
320
1,680
220
Texaco, Inc.
West Tulsa
c/
FCC
2,900
NR
Vickers Petroleum Corp.
Ardmore
FCC
3,420
160
PENNSYLVANIA
BP Oil Corporation
Marcus Hook
Gulf Oil Company
Philadelphia
Sun Petroleum Products Co.
Marcus Hook
FCC
FCC
FCC
7,200
14,000
12,000
250
1,000
2,400
United Refining Co.
Warren
FCC
1,750
30
TENNESSEE
Delta Refining Company
Memphis
TCC
2,150
None
158
-------
APPENDIX B CONTINUED
LOCATION, TYPE, AND CAPACITY OF PETROLEUM
CATALYTIC CRACKING FACILITIES71^
Company and Location
TEXAS
Process
b/
Charge Capacity (cubic meters
per stream day)d//
Fresh Feed
Recycle
American Petrofina, Inc.
Mt. Pleasant
Port Arthur
TCC
FCC
1,500
5,100
350
320
Amoco Oil Company
Texas City
Atlantic Richfield Company
Houston
FCC
FCC
26,600
11,000
6,200
790
Champlin Petroleum Corp.
Corpus Christi FCC
Chevron U.S.A., Inc.
El Paso FCC
Coastal States Petrochemical
Company
Corpus Christi FCC
Cosden Oil & Chemical Co.
Big Spring FCC
8,600
3,500
3,000
3,800
80
480
95
160
Crown Central Petroleum
Corporation
Houston
Diamond Shamrock Corp.
Sunray
FCC
TCC
HCC
6,800
1,830
1,830
1,400
320
320
Exxon Company
Baytown
Gulf Oil Company
Port Arthur
FCC
FCC
21,500
19,000
3 ,300
950
159
-------
APPENDIX B CONTINUED
LOCATION, TYPE, AND CAPACITY OF PETROLEUM
CATALYTIC CRACKING FACILITIES71/
Company and Location
Process
b/
Charge Capacity (cubic meters
per stream day)S//
Fresh Feed
Recycle
TEXAS (Continued)
La Gloria Oil and Gas Company
Tyler
FCC
1,600
790
Marathon Oil Company
Texas City
FCC
4,800
750
Mobil Oil Corporation
Beaumont
FCC
TCC
14,000
3,800
NR
NR
Phillips Petroleum Co.
Borger
Sweeny
FCC.
FCC
8,900
5,400
2,400
790
Shell Oil Company
Deer Park
Odessa
FCC
FCC
11,000
1,670
None
870
Southwestern Refining Co., Inc.
Corpus Christi
FCC
1,900
110
Sun Petroleum Products Company
Corpus Christi
FCC
3,200
1,000
Texaco, Inc.
Amarillo0/
El Paso
Port Arthur
FCC
FCC
FCC
1,300
1,100
21,500
NR
NR
NR
Texas City Refining, Inc.
Texas City FCC
Union Oil Company of California
Beaumont FCC
4,300
6,200
160
640
Winston Refining Company
Fort Worth
FCC
540
410
160
-------
APPENDIX B CONTINUED
LOCATION, TYPE, AND CAPACITY OF PETROLEUM
CATALYTIC CRACKING FACILITIES71^
Company and Location
UTAH
Process
b/
Charge Capacity (cubic meters
per stream day)a//
Fresh Feed
Recycle
Amoco Oil Company
Salt Lake City
Chevron U.S.A.
Salt Lake City
FCC
FCC
HCC
2,900
1,700
1,100
640
None
160
Husky Oil Company
North Salt Lake
TCC
700
400
Phillips Petroleum Company
Woods Cross
TCC
1,300
400
Plateau, Inc.
Roosevelt
FCC
830
None
VIRGINIA
Amoco Oil Company
Yorktown
FCC
4,500
790
WASHINGTON
Mobil Oil Corporation
Femdale
Shell Oil Company
Anacortes
TCC
FCC
4,050
5,700
320
2,700
Texaco, Inc.
Anacortes
FCC
4,800
NR
WISCONSIN
Murphy Oil Corporation
Superior
FCC
1,500
160
161
-------
APPENDIX B CONTINUED
LOCATION, TYPE, AND CAPACITY OF PETROLEUM
CATALYTIC CRACKING FACILITIES71^
Company and Location
Process
b/
Charge Capacity (cubic meters
per stream day)a/
Fresh Feed
Recycle
WYOMING
Amoco Oil Company
Casper
FCC
1,500
330
Husky Oil Company
Cheyenne
Cody
FCC
FCC
1,600
520
400
160
Little America Refining Company
Casper
TCC
1,000
640
Sinclair Oil Corporation
Sinclair
FCC
2,800
190
Tesoro Petroleum Corporation
Newcastle
TCC
640
480
Texaco, Inc.
Casper
NR
1,100
NR
162
-------
APPENDIX B CONTINUED
LOCATION, TYPE, AND CAPACITY OF PETROLEUM
CATALYTIC CRACKING FACILITIES71^
Footnotes
^Capacities originally reported in barrels per stream day. Converted to cubic
meters per stream day using a conversion factor of 0.158987 cubic meters
per barrel and rounding to the number of significant figures originally reported.
k^FCC = Fluid Catalytic Cracking.
TCC Thermofor Catalytic Cracking.
HCC = Houdriflow Catalytic Cracking.
NR = Not Reported.
c /
All figures are per calendar, day. Stream day figures were not reported.
163
-------
APPENDIX C
34/
LISTING OF ASPHALT ROOFING PLANTS IN 197 3
Company Name and Location
ALABAMA
Celotex Corporation
Birmingham, Jefferson 35200
GAF
Mobile, Baldwin 36600
Koppers Company
Woodward, Jefferson 3 518 9
Logan Long Company
Tuscaloosa, Tuscaloosa 35401
ARKANSAS
Bear Brand Roofing, Inc.
Bearden, Quachita 71720
Celotex Corporation
Camden, Columbia 71701
Elk Roofing Company
Stephens, Quachita 71746
Southern Asphalt Roofing Corp.
Little Rock, Pulaski 72200
CALIFORNIA
Bird & Son, Inc.
San Mateo, San Mateo 94403
Bird & Son, inc.
Wilmington, Lake 90744
Celotex Corporation
Los Angeles, Los Angeles 90031
Company Name and Location
CALIFORNIA (Continued)
Certain Teed Products Corp.
Richmond, Contra Costa 948 04
Fibreboard Corporation
Martinet, Contra Costa 94553
Fibreboard Corporation
Oakland, Alameda 94600
Flintkote Company
Los Angeles, Los Angeles 90000
Flintkote Company
San Andreas, Calaveras 95249
John-Mariville Products
Los Angeles, Los Angeles 90058
and Pittsburg, Contra Costa
Lloyd' A. Fry Roofing Co.
Compton, Los Angeles 90223
Lloyd A. Fry Roofing Co.
San Leandro, Alameda 947 55
Lunday-Thagard Oil Co.
South Gate, Los Angeles 90280
Nicolet Industries
Hollister, Santa Cruz 95023
Owens Corning Fiberglas
Santa Clara, Santa Clara 95000
Rigid Mfg. Co., Inc.
Los Angeles, Los Angeles 90 022
164
-------
Appfndty r (continued)
LISTING OF ASPHALT ROOFING PLANTS IN 197334;/
Company Name and Location
CALIFORNIA (Continued)
Mrs. Paul Smithwick
Los Angeles, Los Angeles 90066
Standard Materials Co., Inc.
Merced, Merced 95340
Thermo Materials, Inc.
San Diego, San Diego 92109
United States Gypsum Co.
South Gate, Los Angeles 90280
COLORADO
Colorado Bitumuls Co.
Denver, Denver 80216
GAF Corporation
Denver, Denver 80216
Lloyd A. Fry Roofing Co.
Denver, Denver 80216
CONNECTICUT
Allied Chemical Corporation
Mountville, New London 063 53
Tilo Co., Inc.
Stratford, Fairfield 06497
DELEWARE
Artie Roofings, Inc.
Edge Moore, New Castle 19 809
Company Name and Location
DF.LF.WARF. (Continued)
Artie Roofings, Inc.
Wilmington, New Castle 19809
FLORIDA
GAF Corporation
Tampa, Hillsboro'ugh
Gardner Martin Asphalt Corp.
Tampa, Hillsborough 33605
Lloyd A. Fry Roofing Co., Inc.
Ft. Lauderdale, Broward 33 316
Lloyd A. Fry Roofing Co., Inc.
Jacksonville, Duval 32206
GEORGIA
Certain Teed Products Corp.
Port Wentworth, Effingham 91407
Certain Teed Products Corp.
Savannah, Chatham 314 02
GAF Corporation
Savannah, Chatham 31402
Gibson Homans Company
Conyers, Rockdale 30207
Johns-Manville Products
Savannah, Chatham 314 02
Lloyd A. Fry Roofing Co.
Atlanta, Fulton 30311
165
-------
A^PFINDTX C. (Cont-.i nufifi)
LISTING OF ASPHALT ROOFING PLANTS IN 197334//
Company Name and Location
GEORGIA (Continued)
Mullins Bros. Pvgn. Cntrc.
E. Point, Fulton 30044
Southern Paint Products
Atlanta, Fulton 30310
The Ruberoid Company
Savannah, Chatham 31402
ILLINOIS
Allied Asphalt Paving Co.
Hillside, Cook 60162
Allied Chemical Corporation
Chicago, Cook 60623
Amalgamated Roofing Div.
Bedford Park, Cook 60501
Becker Roofing Co. (2 plants)
Chicago, Cook 60647
Bird & Son, Inc.
Chicago, Cook 60620
Celotex Company
Elk Grove Villacre, Cook 60007
Celotex Company
Peoria, Peoria 61600
Celotex Company
Wilmington, Kankakee 60481
Company Name and Location
ILLINOIS (Continued)
Certain Teed Products Corp.
Chicago Heights, Cook 60411
Certain Teed Products Corp.
E. Saint Louis, Saint Clair 62205
Crown Trygg Corp.
Joliet, Will 60434
Flintkote Company
Chicago Heights, Cook 60411
FS Services, Inc.
Kingston Mines, Peoria 61533
GAF Corporation
Joliet, Will 60433
Globe Industries, Inc.
Chicago, Cook 60.600
J.W. Mortell Co. Inc.
Kankakee, Kankakee 60901
Johns-Manville Corporation
Madison, Madison 62060
Johns-Manville Corporation
Waukegan, Lake 60085
Koppers Company
Chicago, Cook 60600
Lloyd A. Fry Roofing Company
Argo, Cook 60501
166
-------
DRAFT
00 flat QyOTf OR cite
APPENDIX P. (Continued)
LISTING OF ASPHALT ROOFING PLANTS IN 197334^
Company Name and Location
ILLINOIS (Continued)
Lloyd A. Fry Roofing Company
Summit, Clay 60501
Logan Long Company
Chicago, Cook 60638
McCalman Construction Co.
Danville, Vermilion 61832
Midwest Products Co., Inc.
Chicago, Cook 60619
Nicolet Industries
Union, Boone 62635
Rock Road Construction Co.
Chicago, Cook
Seneca Petroleum Co., Inc.
Chicago, Cook 60616
Triangle Construction Co.
Kankakee, Kankakee 60901
Washington Paint Products
Chicago, Cook 60624
INDIANA
Asbestos Manufacturing Corp.
Michigan City, La Porte 46360
GAF Corporation
Mount Vernon, Posey 47620
Globe Industries, Inc.
Lowell, Lake 46356
Company Name and Location
INDIANA (Continued)
H. 3. Reed & Company, Inc.
Gary, Lake 46406
Lloyd A. Fry
Brookville, Franklin 47021
IOWA
Becker Roofing Co., Inc.
Burlington,. Des Moines 52601
Celotex Corporation
Dubuque, Dubuque 52 001
Tufcrete Company, Inc.
Des Moines, Polk 50309
KANSAS
Royal Brank Roofing, Inc.
Phillipsburg, Phillips 67661
LOUISIANA
Bird & Son, Inc.
Shreveport, Caddo 71102
Delta Roofing Mills, Inc.
Slidill, Saint Tammann 70458
Johns-Manville Corporation
Marrero, Jefferson 70072
Slidell Felt Mills, Inc.
Slidell, Saint Tammann 70458
167
-------
APPFNHQTX C. (Continued)
34/
LISTING OF ASPHALT ROOFING PLANTS IN 19 73 '
Company Name and Location
MARYLAND
Congoleum-Nairn, Inc.
Finksburg, Carroll 21048
GAF Corporation
Baltimore, Baltimore 21224
Lloyd A. Fry Roofing Co.
Jessup 20794
MASSACHUSETTS
Bird & Son, Inc.
Norwood, Norfolk 02062
Essex Chemical Corporation
Peabody, Essex 01960
GAF Corporation
Millis, Norfolk 02054
Lloyd A. Fry Roofing Co., Inc.
Waltham, Middlesex 02154
Patrick Ross Company
Cambridge, Middlesex 02142
MICHIGAN
Lloyd A. Fry Roofing Co.
Detroit, Wayne 48217
GAF Corporation
Warren, Macomb 480 89
MINNESOTA
Duval Mfg.Co., Inc.
Minneapolis, Hennepin 55426
Company Name and Location
MINNESOTA (Continued)
Duvall Mfg. Co., Inc.
Minneapolis, Hennepin 55412
EDCO Products, Inc.
Hopkins, Hennepin 5534 3
GAF Corporation
Minneapolis, Hennepin 55411
Lloyd A. Fry Roofing Company
Minneapolis, Hennepin 55412
B.F. Nelson Mfg.Co., Inc.
Minneapolis, Hennepin 55413
E.J. Pennig Co., Inc.
St. Paul, Ramsey 55103
United States Gypsum Co.
St. Paul, Ramsey 55100
MISSISSIPPI
Atlas Roofing Mfg. Co.
Meridian, Lauderdale 39301
Lloyd A. Fry Roofing Co.
Hazelwood
MISSOURI
Certain Teed Products Corp.
Kansas City, Jackson 64126
GAF Corporation
Kansas City, Jackson 64126
168
-------
Apr>F.NDIX C (Continued)
LISTING OF ASPHALT ROOFING PLANTS IN 197334/
Company Name and Location
MISSOURI
Lloyd A. Fry Roofing Co., Inc.
Hazelwood, St. Louis 63042
Lloyd A. Fry Roofing Co., Inc.
N. Kansas City, Clay 64116
Midwest Pre Cote Company
Kansas City, Clay 64119
Tamko Asphalt Products, Inc.
Joplin, Jasper 64801
NEW HAMPSHIRE
Tilo Company, Inc.
Manchester, Hillsboro 03101
NEW JERSEY
Atlantic Cement Company
Bayonne, Hudson 07002
Bird and Son, Inc.
Perth Amboy, Middlesex 08862
Celotex Corporation
Edgewater, Middlesex 07020
Celotex Corporation
Perth Amboy, Middlesex 08862
Flintkote Company, Inc.
E. Rutherford, Bergen 07073
Flintkote Company, Inc.
Whippany, Morris 07981
Company Name and Location
NEW JERSEY (Continued)
GAF Corporation
South Bound Brook, Somerset 088
Johns-Mariville Corporation
Manville, Somerset 08835
Karnak Chemical Corporation
Clark, Union 07066
Congoleum Nairm, Inc.
Kearny, Bergen 07032
Koppers Company, Inc.
Westfield, Union 07090
Lloyd A. Fry Roofing Co., Inc.
Kearny, Bergen 07032
Middlesex CNC Products Excv.
Woodbridge, Middlesex 07095
Tilo Company, Inc.
Westfield, Union 07092
United States Gypsum Company
Jersey City, Hudson 07300
NEW MEXICO
Dura Roofing Manfacturing, Inc.
Albuquerque, Bernalillo 87103
NEW YORK
Alken-Murry Corporation
New York, New York
169
-------
APPTTNnTY r (rnnt.i nnpH ^
LISTING OF ASPHALT ROOFING PLANTS IN 197334//
Company Name and Location
NEW YORK (Continued)
Allied Chemical Corporation
Binghamton, Broome 13902
Durok Building Materials
Company Name and Location
OHIO (Continued)
Johns-Manville Corporation
Cleveland, Cuyahoga 44134
Koppers Company, Inc.
Hastings-Hdsn., Westchester 10706 Cleveland, Cuyahoga 44106
Tilo Company,• Inc.
Poughkeepsie, Dutchess 12603
Tilo Company, Inc.
Watertown, Jefferson 13601
Weatherpanel Sidings, Inc.
Buffalo, Erie 14207
NORTH CAROLINA
Celotex Corporation
Goldsboro, Sampson 07530
Lloyd A. Fry Roofing Co, Inc.
Morehead City, Carteret 28557
Rike Roofing and Mfg. Co.
Charlotte, Mecklenburg 28201
OHIO
Celotex Corporation
Cincinnati, Hamilton 45215
Certain Products Company
Milan, Erie 44846
Consolidated Paint Varnish
Cleveland, Cuyahoga 44114
Gibson Homans Company, Inc.
Cleveland, Cuyahoga 44106
Koppers Company, Inc.
Youngstown, Mahoning 44500
Lloyd A. Fry Roofing Company
Medina, Cuyahoga 442 56
Logan Long Company, Inc.
Franklin, Warren 45005
Midwest Products Company, Inc.
Cleveland, Cuyahoga 44110
Overall Paint, Inc.
Cleveland, Cuyahoga 44146
Rarico Industrial Products
Cleveland, Cuyahoga 44120
SET Products, Inc.
Cleveland, Cuyahoga
44106
Tremco Manufacturing Company
Cleveland, Cuyahoga 44104
OKLAHOMA
Allied Materials Corporation
Stroud, Lincoln 74079
Big Chief Roofing Company, Inc.
Ardmore, Carter 73401
170
-------
APPENDIX C (Cont-i niuarU
LISTING OF ASPHALT ROOFING PLANTS IN 1973
Company Name and Location
OKLAHOMA (Continued)
Lloyd A. Fry Roofing Co., Inc.
Oklahoma City, Caradian 73117
OREGON
Bird and Son, Inc.
Portland, Multnomah 97200
Fibreboard Corporation
Portland, Multnomah 97210
Flintkote Company, Inc.
Portland, Multnomah 97208
Herbert Malarkey Roofing Co.
Portland, Multnomah 97217
Lloyd A. Fry Roofing Co., Inc.
Portland, Multnomah 97210
Shell Oil Company
Portland, Multnomah 97210
PENNSYLVANIA
Allied Chemical Corporation
Philadelphia, Philadelphia 19146
Celotex Corporation
Philadelphia, Philadelphia 19146
Celotex Corporation
Sunbury, Northumberland 17801
Certain Teed Products Corp.
York, York 17303
and St. Gobian, Luzerne 18707
Company Name and Location
PENNSYLVANIA (Continued)
ESB Inc. Del.
Mertztown, Berks 19539
GAF Corporation
Erie, Erie 16500
Keystone Roofing Mfg. Company
York, York 17403
Lloyd A. Fry Roofing Company
Emmaus, Lehigh 18049
Lloyd A. Fry Roofing Company
York, York 17404
Monsey Products Company
Philadelphia, Philadelphia 19128
H.C. Price Company
Philadelphia, Philadelphia 19115
Tilo Company, Inc.
Philadelphia, Philadelphia 19118
SOUTH CAROLINA
Bird and Son, Inc.
Charleston Hts., Charleston 29405
TENNESSEE
Celotex Corporation
Memphis, Shelby 38100
Lloyd A. Fry Roofing Company
Memphis, Shelby 38107
171
-------
apPFNnTY c (Continued)
34/
LISTING OF ASPHALT ROOFING PLANTS IN 1973
Company Name and Location
TEXAS
American Petrofina Texas
Mt. Pleasant, Titus 75455
Celotex Corporation
Houston, Liberty 77000
Celotex Corporation
San Antonio, Bexar 78200
Certain Teed Products, Corp.
Dallas, Dallas 75216
Daingerfield Mfg. Company
Daingerfield, Morris 75638
Flintkote Company
Ennis, Ellis 75119
GAF Corporation
Dallas, Dallas
Gulf States Asphalt Co., Inc.
Beaumont, Jefferson 77704
Johns-Manville Corporation
Ft. Worth, Tarrant 76107
Lloyd A. Fry Roofing Co.
Irving, Dallas 75060
Lloyd A. Fry Roofing Co.
Houston, Harris 77029
Lloyd A. Fry Roofing Co.
Lubbock, Lubbock 7940 8
Company Name and Location
TEXAS (Continued)
Ruberoid Company
Dallas, Dallas 75222
Southwestern Petroleum
Fort Worth, Tarrant 76106
Texas Sash and Door
Fort Worth, Tarrant 76101
UTAH
Lloyd A. Fry Roofing Company
Woods Cross, Davis 84087
WASHINGTON
Certain Teed Products Corp.
Tacoma, Pierce 98421
Kollogg Company, Inc.
Washington
B. F. Nelson Mfg. Company, Inc.
Washington
WEST VIRGINIA
Celotex Corporation
Chester, Hancock 26034
172
-------
DRAFT
DO NOT QUOTE OR CITE
APPENDIX D
LOCATION AND CAPACITY OF UNITED STATES
SINTERING FACILITIES1^
Capacity (square meters
Location (feet) of grate area)
ALABAMA
Fairfield 288.4 (3,104)
Gadsden 52.9 ( 569)
CALIFORNIA
Fontana 113.7 (1,224)
ILLINOIS
Granite City 95.1 (1,024)
South Chicago 165.0 (1,776)
INDIANA
East Chicago 124.9 (1,344)
Gary 587.8 (6,327)
Westchester Township 187.7 (2,020)
MARYLAND
Sparrows Point
MICHIGAN
River Rouge
NEW YORK
Lackawanna
OHIO
353.0 (3,300)
223.0 (2,400)
113.7 (1,224)
Youngstown
Cleveland
Warren and Niles
Campbell
Lorain
124.9 (1,344)
38.9 ( 419)
40.1 ( 432)
49.1 ( 528)
42.6 ( 459)
173
-------
APPENDIX D (Continued)
LOCATION AND CAPACITY OF.UNITED STATES
.1/
SINTERING FACILITIES
Location
PENNSYLVANIA
Capacity (square meters
(feet) of grate area)
Fairless Hills
Rankin
McKeesport
Monessen
Bethlehem
Johnstown
Pittsburgh
Braddock
517.9
113.7
113.7
56.9
184.3
110.7
33.2
53.5
(5,575)
(1,224)
(1,224)
( 612)
(1,984)
(1,192)
( 357)
576)
(
TEXAS
Lone Star
51.1
( 550)
UTAH
Geneva
WEST VIRGINIA
113.7
(1,224)
East Steubenville
50.2
( 540)
174
-------
APPENDIX E
LOCATION AND CAPACITY OF CARBON BLACK PLANTS, 197766/
Company Name & Location
Ashland Oil, Incorporated
Aranas Pass, Texas
Cities Service Company
Seminole, Texas
Cabot Company
Pampa, Texas
Ashland Oil, Incorporated
Shamrock, Texas
Sid Richardson Carbon Company
Big Spring, Texas
Cabot Company
Big Spring, Texas
Cities Service Company
Conroe, Texas
Cities Service Company
Seagraves, Texas
Continental Carbon Company
Sunray, Texas
J. M. Huber Corporation
Baytown, Texas
J. M. Huber Corporation
Borger, Texas
Phillips Petroleum Company
Borger, Texas
Phillips Petroleum Company
Orange, Texas
Ashland Oil, Incorporated
Belpre, Ohio
Phillips Petroleum Company
Toledo, Ohio
Ashland Oil, Incorporated
Iberia, Louisiana
Annual Capacity in
Millions of Kilograms (pounds)
68 (150)
16 ( 35)
24 ( 53)
48 (105)
50
108,
44
41
43
117
81
130
52
45
32
116
(110)
(238)
( 97)
( 90)
( 95)
(258)
(179)
(287)
(115)
(100)
( 70)
(255)
Process
Furnace
Channel
Furnace
Furnace
Furnace
Furnace
Furnace
Furnace
Furnace
Furnace
Furnace &
Thermal
Furnace
Furnace
Furnace
Furnace
Furnace
175
-------
APPENDIX E (Continued)
LOCATION AND CAPACITY OF CARBON BLACK PLANTS, 197766^
Company Name & Location
Cabot Company
Franklin, Louisiana
Cabot Company
Villa Platte, Louisiana
Cities Service Company
Eola, Louisiana
Cities Service Company
North Bend, Louisiana
Continental Carbon Company
Westlake, Louisiana
Int'l Minerals & Chemical Corp.
Sterlington, Louisiana
Sid Richardson Carbon Company
Addis, Louisiana
Ashland Oil, Incorporated
Mohave, California
Cities Service Company
Mohave, California
Continental Carbon Company
Bakersfield, California
Cabot Company
Waverly, West Virginia
Cities Service Company
Moundsville, West Virginia
Cities Service Company
El Dorado, Arkansas
Continental Carbon Company
Ponca City, Oklahoma
Harmon Colers Corporation
Haledon, New Jersey
Annual Capacity in
Millions of Kilograms (pounds)
98
110
32
96
54
59
27
24
35
74
71
37
(215)
(243)
( 70)
(212)
(120)
(130)
46 (102)
( 60)
( 53)
( 77)
(163)
(157)
( 82)
61 (135)
N/A
Process
Furnace
Furnace
Furnace
Furnace &
Thermal
Furnace
Thermal
Furnace
Furnace
Furnace
Furnace
Furnace
Furnace
Furnace
Furnace
Furnace
176
-------
APPENDIX F
MUNICIPAL INCINERATORS
30/
State
Capacity (Mg/day
(tons/day))
State
Capacity (Mg/day
(tons/day))
CONNECTICUT
KENTUCKY
•~j
-J
Ansonia
Bridgeport
Darien
East Hartford
Hartford
New Canaan
New Haven
New London
Stamford
Stratford
Waterbury
West Hartford
FLORIDA
Broward County
Dade County
Fort Lauderdale
(Broward County)
Miami (Dade County)
Tampa
ILLINOIS
Chicago
Chicago
Chicago (South Doty)
Cicero
180
180
120
320
540
110
650
110
360
239
270
270
270
270
410
820
910
1,100
1,500
1,100
450
200)
200)
130)
350)
600)
125)
720)
120)
400)
264)
300)
300)
300)
300)
450)
900)
1,000)
1,200)
1,600)
1,200)
500)
Louisville
910 (1,000)
LOUISIANA
INDIANA
New Orleans
New Orleans
New Orleans
New Orleans
New Orleans
Shreveport
MARYLAND
Baltimore
MASSACHUSETTS
Belmont
Braintree
Brockton
Brookline
Fall River
Framingham
Lowell
Marblehead
Reading
Salem
Saugus
Watertown
Weymouth
Winchester
180
360
360
410
360
180
140
220
540
160
540
450
360
70
131
130
1,100
290
270
90
( 200)
( 400)
( 400)
( 450)
( 400)
( 200)
730 ( 800)
( 150)
( 240)
( 600)
( 180)
( 600)
(
500)
400)
80)
144)
140)
(1,200)
( 320)
( 300)
( 100)
East Chicago
410 ( 450)
-------
APPENDIX F (Continued)
MUNICIPAL INCINERATORS
30/
State
Capacity (Mg/day
(tons/day))
State
Capacity (Mg/day
(tons/day))
MICHIGAN
Central Wayne County
Detroit (Southfieild)
Grosse Point
South East Oakland County
MISSOURI
St. Louis (North)
St. Louis (South)
NEW HAMPSHIRE
Manchester
NEW JERSEY
Ewing
Red Bank
NEW YORK
730
180
540
540
360
360
220
44
( 800)
( 200)
( 600)
( 600)
( 400)
( 400)
90 ( 100)
( 240)
( 48)
(NEW YORK Continued)
Long Beach
Mount Vernon
New Rochelie
North Hempstead
Oyster Bay
Rye
Scarsdale
Tonawanda
Valley Stream
NEW YORK CITY
New York City
New York City
New York City
New York City
New York City
Babylon
360
(
400)
Beacon
90
(
100)
Buffalo
540
(
600)
Eastchester
180
(
200)
Freeport
140
(
150)
Garden City
160
(
175)
Hempstead
540
(
600)
Hempstead
680
(
750)
Huntington
270
(
300)
Islip
270
(
300)
Lackawanna
140
(
150)
Lawrence
180
(
200)
OHIO
Euclid
Franklin
Lakewood
Miami County
Parma
Sharonvilie
PENNSYLVANIA
Ambridge
Bradford
Delaware County
180
540
360
540
450
140
140
230
180
910
910
910
910
910
180
140
270
140
200
450
140
180
7 30
(
200)
600)
( 400)
600)
500)
( 150)
( 150)
( 250)
( 200)
(1,000)
(1,000)
(1,000)
(1,000)
(1,000)
( 200)
( 150)
( 300)
( 150)
( 225)
( 500)
( 150)
( 200)
( 800)
-------
APPENDIX F "(Continued)
MUNICIPAL INCINERATORS
30/
State
(PENNSYLVANIA Continued)
TEXAS
UTAH
Ogden
VIRGINIA
Alexandria
Newport News
Norfolk
Portsmouth
Capacity (Mg/day
(tons/day))
410 ( 450)
270
360
360
320
( 300)
( 400)
( 400)
( 350)
State
WISCONSIN
Capacity (Mg/day
(tons/day))
Delaware County
450
(
500)
Oshkosh
320
350)
Delaware County
450
(
500)
Port Washington
68
75)
Philadelphia
540
(
600)
Sheboygan
220
240)
Philadelphia
540
(
600)
Sturgeon Bay
140
150)
Shippensburg
65
(
72)
Waukesha
320
350)
¦
Amarillo
320
(
350)
-------
APPENDIX G
LIST OF NAMES, LOCATIONS, AND PHONE NUMBERS OF PERSONAL CONTACTS
Name
Affiliation
Location
Phone Number
Ballantine, Dr. David
Bambaugh, Carl
Barush, Steve
Becker, Don, Manager
Benedict, John
Bennett, Roy L.
Benson, Jaraes
Bills, Bill
oo Black, Frank
o
Bornstein, Mark
Bowen, Dr. Joshua, Chief
Brodovicz, Ben
Brown, Dave
Brown, Jane
Calaizzi, Gary
Carrigan, Dr. Richard A.
Carpelan, Dr. Marian
Caton, Dr. Robert
Crawford, A. R.
DOE, EV Washington, DC
Radian Corporation Austin, TX
EPRI Palo Alto, CA
Recycled Oil Program Washington, DC
NBS, Institute for Materials Research
WVA Air Pollution Con- Charleston, WV
trol Commission
EPA, ESRL Research Triangle Park,
PA DER, Air Quality &. Karrisburg, PA
Noise Control Division, Abatement & Compliance
KY DNRER, Division of Frankfort, KY
Air Pollution Control Engineering Program
EPA, ESRL, Mobile Research Triangle Park,
Sources
GCA Technology Division Bedford, MA
Combustion Research Research Triangle Park,
Branch, EPA, IERL, Energy Assessment & Control
PA DER, Air Quality & Harrisburg, PA
Noise Control Division
NIOSH Cincinnati, OH
NIOSH Cincinnati, OH
NC
NC
202
512
415
202
-353-
-454-
-355-
-921-
¦3610
¦4797
¦2.469
•3837
304-348-3286
919-
717-
502-
919-
¦541-
-787-
•564-
-541-
¦3173
¦4 324
¦6844
•3037
617-275-9000
NC 919-541-2470
Division
717-787-2347
513-
513-
684
¦684
-8235
-3255
Denver, CO
Washington, DC
303-234-4060
202-632-5970
bom
NSF, ASAR, Research
Applications
University of California Riverside, CA 714-787-3545
Statewide Air Pollution Research Center, Information Center
Administrator of En- Toronto, Ontario, Canada 416-965-4081
vironment
Exxon Research & En- Linden, NJ 201-474-2443
gineering Company
Edgerton, Kurt
MESA
Pittsburgh, PA
412-621-4500
-------
2-PPPNDTX G (Continue^)
LISTS OF NAMES, LOCATIONS, AND PHONE NUMBERS OF PERSONAL CONTACTS
Name
Friedrich, Andrew
Grosjean, Dr. Daniel
Gross, George P.
Hamersma, Dr. Warren
Himell, James H.
Johnson, Gary
Johnson, Larry
Jones, Dr. Peter
Kauffman, Hugh
Knapp, Dr. Ken
Kneip, Dr. Ted
Lahre, Tom
Levins, Dr. Philip
Lincoln, John
Magda, Niren
Magnuson, Malcolm 0.
Environmental Coordinator
Maloney, Ken
Affiliation
PA DER, Division of
Mine Restoration
Location
Harrisburg, PA
University of California Riverside, CA
Statewide Air Pollution Research Center
Exxon Research & En- Linden, NJ
gineering Company
TRW, Environmental
Engineering
GEOMET, Forestry
Redondo Beach, CA
Gaithersburg, MD
EPA, IERL, Office of Research Triangle Park, NC
Program Operations, Special Studies Staff
EPA, IERL, Industrial Research Triangle Park, NC
Process Division, Process Measurement Branch
Battelle-Columbus Labs Columbus, OH
EPA, Hazardous Wastes Washington, DC
EPA, ESRL Research Triangle Park, NC
NYU Medical School, In- Tuxedo, N.Y.
stitute of Environmental Medicine
EPA, MDAD, Air Manage- Research Triangle Park, NC
ment Technology Branch
A.D. Little, Inc. Cambridge, MA
MESA Arlington, VA
GEOMET Gaithersburg, MD
BOM, Bruceton Research Pittsburgh, PA
Office, Coal Mine Fire Control Group
KVB Tustin, CA
Phone Number
717-787-7668
714-787-3545
201-474-2844
213-535-
301-948-
919-541-
919-541-
614-424-
205-755-
919-541-
914-351-
•145a
¦0755
•2745
•2557
¦6424
•9201
¦3085
¦5355
919-541-5475
617-864-5770
703-235-1284
301-948-0755
412-892-2400
714-832-9020
-------
APPENDIX G (Continued)
LIST OF NAMES, LOCATIONS, AND PHONE NUMBERS OF PERSONAL CONTACTS
CO
i\j
Name
Matthews, Birch
McCarley, Ed, Chief
McElroy, Mike
McMahon, Charles
McNay, Lewis
Natusch, David F.S.
O1Brien,
Orwin, Bob
Paone, James, Chief
Pireovich, John, Director
Plaks, Norman, Chief
Potter, Herschel
Raybold, Richard L.
Reznik, Dr. Dick
Rhodes, Bill
Rosen, Hal
Affiliation
Location
Redondo Beach, CA
TRW, Environmental
Engineering
EPA, Emissions Measure- Durham, NC
ment Branch
EPRI Palo Alto, CA
U.S. Forest Service, Macon, GA
Southern Forest Fire Research Lab
BOM, Mining Research Spokane, WA
Center
Colorado State Univer- Fort Collins, CO
sity, Department of Chemistry
Bureau of Census, Pop- Washington, DC
ulation Division
PA DER, Solid Waste Harrisburg, PA
Management Division, Bureau of Land Protection
BOM, Division of En- Washington, DC
vironment
Smoke Management, U.S. Macon, GA
Forest Service, Southern Forest Fire Research Lab
EPA, IERL, Industrial Research Triangle Park, NC
Processes Division, Metallurgical Processes Branch
MESA Arlington, VA
Gaithersburg, MD
Dayton, OH
Phone Number
213-536-3884
919-541-5245
415-855-2471
912-746-9436
509-484-1610
303-491-6381
202-763-5002
717-787-7382
NBS, Electronics Lab
Monsanto Research Cor-
poration
EPA, IERL, Energy As- Research Triangle Park, NC
sessment & Control Division, Fuel Process Branch
University of California Berkeley, CA
Lawrence-Berkeley Labs, Atmospheric Aerosol Research
202-634-
912-746-
919-541-
703-235-
301-921-
513-268-
¦1251
¦1477
¦2733
¦1284
•3786
¦3411
919-541-2851
415-843-2740
Group
-------
APPENDIX G (Continued)
LIST OF NAMES, LOCATIONS, AND PHONE NUMBERS OF PERSONAL CONTACTS
Name
St. Louis, Richard
Smith, Gene
Smith, Dr. John
Sommerer, Dr. Diane,
Director
Spindt, Robert S.
Springer, Karl
Stahley, Dr. Stewart
Stasikowski, Dr. Margaret
Tejada, Dr. Sylvestre
Tucker, W. Gene, Chief
Turner, P.P., Chief
Venezia, Ron
Weinstein, Norm
White, Dr. Lowell
Winer
Zelinski, Dr. Wilbur
Zengel, A.E.
Affiliation
Location
PA DER, Air Quality & Harrisburg, PA
Noise Control Division
EPA, ESED Research Triangle Park, NC
EPA, IERL Research Triangle Park, NC
York Research Corpora- Stamford, CT
tion, Environmental Science
Gulf Research & De- Harmarville, PA
velopment Corporation
Southwest Research In- San Antonio, TX
stitute
University of Mary- College Park, MD
land, Chemistry Department
EPA Ann Arbor, MI
EPA, ESRL Research Triangle Park, NC
EPA, IERL, Office of Research Triangle Park, NC
Program Operations, Special Studies Staff
EPA, IERL, Energy As- Research Triangle Park, NC
Phone Number
717-787-2347
919-541-5421
919-541-2921
203-325-1371
412-828-5000
512-684-5111
301-454-4679
313-668-4200
919-541-2323
919-541-2745
919-541-2825
sessment & Control Division, Advanced Processes Branch
919-541-2547
EPA, IERL, Industrial Research Triangle Park, NC
Processes Division, Chemical Processes Branch
ReCon Systems Princeton, NJ
ASARCO Salt Lake City, UT
U.S. Forest Service, Washington, DC
Division of Timber Management
Penn State University University Park, PA
Department of Geography
The Coordinating Re- New York, N.Y.
search Council, Inc.
609-921-2112
801-262-2459
202-447-6893
814-865-1650
212-757-1295
-------
APPENDIX G (Continued)
LIST OF NAMES, LOCATIONS, AND PHONE NUMBERS OF PERSONAL CONTACTS
Name Affiliation Location Phone Number
Seizinger, D. E. DOE, Bartlesville Energy Bartlesville, OK 918-336-2400
Research Center
Trenholm, Andrew R. EPA, ESED, Industrial Research Triangle Park, NC 919-541-5301
Studies Branch, Stand-
ards Support Section
Wasser, Jack EPA, IERL, Energy Assess-Research Triangle Park, NC 919-541-2476
ment & Control Division,
Combustion Research
Branch
O
CD
<0 C3
CD
C~J
-------
BIBLIOGRAPHY
1. American Iron and Steel Institute, Directory of Iron and
Steel Works of the U.S. and Canada, 1977.
2. Automobile Manufacturers' Association, Automobile Facts and
Figures, Washington, D.C., 1971.
3. Babcock & Wilcox, Steam: Its Generation and Use, 38th
Edition, New York, 1975.
4. Baladi, E., Stationary Source Testing of Bagasse-Fired
Boilers at the Hawaiian Commercial and Sugar Company,
Puunene, Maui, Hawaii, Final Report, EPA Contract No.
68-02-1403, Task 12, February 1976, Durham, North Carolina.
5. Bambaugh, C., Radian Corporation, Personal Communication,
December 13, 1977, Houston, Texas.
6. Beine, H., "The Level of 3,4 Benzopyrene in the Waste Gases
of Domestic Stoves Using Solid Fuels," Staub, 30(8), August
1970.
7. Bee, R.W., Erskine, G., et al., Coke Oven Charging Emission
Control Test Program, Volume I, EPA-650/2-74-062, PB 237-628,
July 1974, Durham, North Carolina.
8. Benedict, J., West Virginia Air Pollution Control Commission,
Personal Communication, February 1978, Charleston, West
Virginia.
9. Bennett, R., ESRL, EPA, Personal Communication, December
1977, February 1978, Durham, North Carolina.
10. Bennett, R.L., and Knapp, K.T., "Chemical Characterization
of Particulate Emissions From Oil-Fired Power Plants."
Presented at the 4th National Conference on Energy and the
Environment, October 1976, Cincinnati, Ohio.
11. Benson, J., Bureau of Air Quality and Noise Control, Dept.
Dept. of. Environmental Resources, Personal Communication,
February 1978, Harrisburg, Pennsylvania.
12. Bills, W., Division of Air Pollution Control, Kentucky Dept.
for Natural Resources and Environmental Protection, Personal
Communication, February 1978, Frankfort, Kentucky.
185
-------
13. Bittner, J.D., Prado, G.P., et al., "The Formation of Soot
and Polycyclic Aromatic Hydrocarbons (PCAH) in Combustion
Systems," Proceedings of the Stationary Source Combustion
Symposium, Volume I, Fundamental Research, EPA-600/2-76-152a,
PB 256-320, June 1976, Durham, North Carolina.
14. Braunstein, H.M., Copenhaver, E.D., and Pfuderer, H.A.,
Eds., Environmental, Health and Control Aspects of Coal
Conversion: An Information Overview, Volume I, ORNL/EIS-94
April 1977.
15. Brinkerhoff, R.J., "Inventory of Intermediate-Size Incinera-
tors in the United States, 1972," Pollution Engineering,
November 1973.
16. Calaizzi, Gary, Bureau of Mines, Personal Communication,
November 1977, Denver, Colorado.
17. Committee on Biologic Effects of Atmospheric Pollutants,
Particulate Polycyclic Organic Matter, National Academy of
Sciences, Washington, D.C., 1972.
18. Control of Pollution From Outboard Engine Exhaust: A Recon-
naissance Study, Rensellaer Polytechnic Institute for EPA,
Project No. 15020 ENN, September 1971, Durham, North Carolina.
19. Cooperative Fire Protection Staff, 1976 Wildfire Statistics,
Forest Service, U.S. Department of Agriculture, September
1977, Washington, D.C.
20. Cato, G.A., Field Testing: Trace Element and Organic
Emissions¦ for Industrial Boilers") EPA-600/2-76-086b,
October 1976, Durham, North Carolina.
21. Caton, R., Administrator of Environment, Personal Communi-
cation, November 1977, Toronton, Ontario, Canada.
22. Darley, Ellis F. and Lerman, Shimshon L., Air Pollutant Emis-
sions From Burning Sugar Cane and Pineapple Residues From
Hawaii, EPA-450/3-75-071, July 197 5, Durham., North Carolina.
23. Darley, Ellis F., Air Quality and Smoke From Urban and
Forest Fuels, National Academy of Sciences, 1976.
24. Darley, Ellis. F., Emission Factor Development for Leaf
Burning, December 197 6, Durham, North Carolina.
186
-------
25. Davies, I.W., Harrison, R.M., et al., "Municipal Incinerator
as Source of Polynuclear Aromatic Hydrocarbons in Environment,"
Environmental Science and Technology, KK5) : 451-453,
May 1976.
26. Energy Research and Development Administration, Office of
Planning, Analysis, and Evaluation, Office of Environmental
Analysis, Regional Air Emissions Analysis of Alternative
Energy Policies in 1985, Washington, D.C., October 1977.
27. Falk, H.L., Kotin, P., and Miller, A., "Aromatic Polycyclic
Hydrocarbons in Polluted Air as Indicators of Carcinogenic
Hazards," International Journal of Air Pollution,
2: 201-209, 1960.
28. Federal Power Commission, FPC News, "Coal Deliveries to
Steam-Electric Plants," Volume 10, Number 14, April 18,
1977.
29. Fenelly, P.F., "Emission Estimates of N0_ and Organic Com-
pounds From Coal-fired Fluidized Bed Combustion," Pro-
ceedings of the International Photochemical Oxidant
Pollution Control Conference, Volume II, EPA-600/3-77-001b,
pp. 1015-1023, 1977, Durham, North Carolina.
30. Fenton, R., "Present Status of Municipal Incinerators,"
Incinerators and Solid Waste Technology, J.W. Stephenson,
et al., Ed., ASME, New York, New York, 1975.
31. Fireplace Institute, Personal Communication, November
1977, Chicago, Illinois.
32. Fox, R.D., et al., Control Techniques for Polycyclic
Organic Matter Emissions, First Draft Report for HEW
Contract No*; CPA-70-43, NAPCA, August 1970.
33. Friedrich, A., Division of Mine Restoration, Pennsylvania
Dept. of Mines, Personal Communication, February 1978, Har-
risburg, Pennsylvania.
34. Gerstle, R.W., Atmospheric Emissions From Asphalt Roofing
Processes, EPA-650/2-75-101, PB 238-4 45, October 1974,
Washington, D.C.
35. Giammar, R.D., Engdahl, R.B., and Barret, R.E., Emissions
From Residential and Small Commercial Stoker-Coal-FirecT"
Boilers Under Smokeless Operation, EPA-600/7-76-029,
PB 263-891, October 1976, Durham, North Carolina.
187
-------
36. Giammar, R. D., Weller, A.E., et al., Experimental Evalua-
tion of Fuel Oil Additives for Reducing Emissions and
Increasing Efficiency of Boilers, EPA-600/2-77-0Q8b,
January 1977, Durham, North Carolina.
37. Goldberg, A.J., A Survey of Emissions and Controls for
Hazardous and Other Pollutants, EPA R4-73-021, PB 223-568,
February 1975, Washington, D.C.
38. Gross, G.P., Third Annual Report on Gasoline Composition
and Vehicle Exhaust Gas Polynuclear Aromatic Content, CRC
APRAC Project No. CAPE-6-68, EPA Contract No. 68-0400-25,
APTD 1560, PB 218-873, July 1972, Durham, North Carolina.
39. Hangebrauck, R.P., von Lehmden, D.J., and Meeker, J.E.,
Sources of Polynuclear Hydrocarbons in the Atmosphere,
U.S. HEW, Public Health Service, AP-33, PB 174-706, 1967,
Washington, D.C.
40. Hare, Charles T. and Springer, Karl J., Exhaust Emissions
From Uncontrolled Vehicles and Related Equipment Using
Internal Combustion Engines, Part 3: Motorcycles, APTD-1492,
March 1973, Ann Arbor, Michigan.
41. Il'nitskii, A.P., Gvil'dis, V. Yu., et al., Role of Vol-
canoes in the Formation of the Natural Level of Carcinogens,"
Translated from Doklady Akademii Nauk SSSR 234(3): 717-719,
May 1977, UDC 616-006-02, Plenum Publishing Corporation.
42. Johnson, G., Special Studies Branch, IERL, EPA, Personal
Communication, February 197 8.
43. Jones, P., Battelle-Columbus Laboratories, Personal Communi-
cation, December 1977, Columbus, Ohio.
44. Jones, P.W., Giammar, R.D., Strupp, P.E., and Stanford, T.B.,
"Efficient Collection of Polycyclic Organic Compounds From
Combustion Effluents," Environmental Science and Technology
1£(8) , August 1976, pp. 806-810.
45. Jones, Peter W., Letter to Dr. Tom Lahre, IERL, Re: Final
Report on Leaf Burning Emission Program, EPA Contract No.
68-02-1409, Task 28, August 1975.
46. Katari, V.S. and Gerstle, R.W., Industrial Process Profiles
for Environmental Use: Chapter 24, The Iron and Steel
Industry, EPA-600/2-77-023x, February 1977, Washington, D.C.
18 8
-------
47. Krause, H.H., Hillenbrand, L.J., et al., Combustion
Additives for Pollution Control, A State-of-the-Art
Review, EPA-600/2-77-008a, January 1977.
48. Lao, R.C., Thomas, R.S., & Markman, T.L., "Computerized
Gas Chromatograph-Mass Spectrometric Analysis of
Polycyclic Aromatic Hydrocarbons Environmental Samples,"
J. Chrom., 112: 181-200, 1975.
49. Laresgosti, A., Loos, A.C., and Springer, G.S., "Particu-
late and Smoke Emission From a Light Duty Diesel Engine,"
Environmental Science and Technology, 11^(10): 973-978,
October 1977.
50. Laster, L.L., Atmospheric Emissions From the Asphalt
Industry, EPA-650/2-73-046, December 1973, Durham, North
Carolina.
51. Lawther, P.J., Commins, B.T., and Waller, R.E., "A Study
of the Concentrations of Polycyclic Aromatic Hydrocarbons
in Gas Works Retort Houses," Brit. J. Ind. Med., 22: 13-20,
1965.
52. League of Women Voters, "Municipal Sludge: What Shall
We Do With It?", Current Focus, 1976, Washington, D.C.
53. Levins, P., A.D. Little, Incorporated, Personal Communica-
tion, December 1977, Cambridge, Massachusetts.
54. Magnuson, M.O., and Baker, E.C., "State-of-the-Art in Ex-
tinguishing Refuse Pile Fires," Presented at the First
Symposium on Mine and Preparation Plant Refuse Disposal,
NCA Coal and the Environment Conference, October 22-24,
1974, Louisville, Kentucky.
55. Maloney, K., KVB Incorporated, Personal Communication,
February 1978, Tustin, California.
56. Matthews, B. and Hamersma, W., TRW Environmental Engineering
Division, Personal Communication, February 1978, Redondo
Beach, California.
57. May, W. and Brown, J., "The Analysis of Some Residual Fuel
and Waste Lubricating Oils by High Performance Liquid
Chromatographic Procedure," Proceedings, Second National
Bureau of Standards Conference on Measurements and Standards
for Recycled Oil, in press.
189
-------
58. McKinnon, Ross, Auto Dismantlers and Recyclers of America,
Incorporated, Personal Communication, November 1977.
59. McMahon, C.K., and Tsoukalas, S.N., "Polynuclear Aromatic
Hydrocarbons in Forest Fire Smokes," Presented at the
Second International Symposium on Polynuclear Aromatic Hy-
drocarbons, Battelle-Columbus Laboratories, September
28-30, 1977, Columbus, Ohio.
60. McNay, Lewis M., Coal Refuse Fires, An Environmental Hazard,
U.S. Bureau of Mines, IC 8515, 1971, Washington, D.C.
61. McNay, Lewis M., U.S. Bureau of Mines, Mining Research Center,
Personal Communication, January 1978, Spokane, Washington.
62. McNesby, J.R., Byerly, R. Jr., and Raybold, R.L., "Emissions
of Rubber From Auto Tires," National Bureau of Standards
Technical Note No. 711:72, 32, GPO C 13.46:71, January 1972.
63. Motor Vehicle Manufacturers' Association, Motor Vehicle Facts
and Figures 1977, 1977, Detroit, Michigan.
64. Natusch, D.F.S., Dept. of Chemistry, Colorado State Univer-
sity, Personal Communication, February 1978, Fort Collins,
Colorado.
65. Natusch, D.F.S. & Tomkins, B.A., "Theoretical Consideration
of the Adsorption of PAH Vapor Onto Fly Ash in a Coal-Fired
Power Plant," Presented at the Second International Sym-
posium on Polynuclear Aromatic Hydrocarbons, Battelle-
Columbus Laboratories, October 1977, Columbus, Ohio.
66. 1977 Directory of Chemical Producers, USA, Chemical Infor-
mation Services, SRI, Menlo Park, California, 1977.
67. "1977 Survey of Resources Recovery and Energy Conversion
Practices," Waste Age, October 1, 197 6.
68. O'Brien, Ms., Population Division, U.S. Bureau of Census,
Personal Communication, December 1977.
69. Office of Technology Assessment, by Energy and Environmental
Analysis, Incorporated, Technology Assessment of Changes in
the Use and Characteristics of the Automobile, October
1977, Arlington, Virginia.
190
-------
70. The Oil and Gas Journal:
a. May 23, 1977, p. 20,21
b. March 28, 1977, p. 92
c. February 7, 1977, p. 28,30
d. April 4, 1977, p. 64,65
e. November 7, 1977 - OGJ Newsletter
71. The Oil and Gas Journal, Worldwide Directory; Refining
and Gas Processing 1977-1978, 1977.
72. Olsen, D., and Haynes, J.L., Preliminary Air Pollution
Survey of Organic Carcinogens, NAPCA Publication APTD
69-43, October 1969.
73. Orwin, R., Bureau of Land Protection, Pennsylvania Dept.
of Environmental Resources, Personal Communication,
February 1978, Harrisburg, Pennsylvania.
74. Paone, James, Chief, Division of Environment, Bureau of
Mines, Personal Communication, November 1977, Washington,
D.C.
75. Pauley, David, "They're Cookin1 With Wood," Newsweek,
Volume XCI, No. 1, January 2, 1978.
76. Pennsylvania Department of Environmental Resources Emissions
Test Results, Personal Communication with Richard St. Louis,
Division of Air Quality, December 1977, Harrisburg, Pen-
nsylvania .
77. Pettigrew, R.J. and Roninger, F.H., Rubber Reuse and Solid
Waste Management, Part I, Uniroyal Chemical for EPA,
No. PH-86-68-208, 1971, Durham, North Carolina.
78. Pierovich, J. and MacMahon, C., Smoke Management Research
and Development Program, USDA Forest Service, Southeastern
Forest Fire Laboratory, Personal Communication, December
1977, February 1978, April 1978, Macon, Georgia.
79. Potter, H. and Lincoln, J., Mining Enforcement and Safety
Administration, Personal Communication, February 1978,
Arlington, Virginia.
80. Raybold, R., Electronics Lab, National Bureau of Standards,
Personal Communication, December 1977, Gaithersburg, Mary-
land.
191
-------
81. Ryan, P.W., and McMahon, C.K., "Some Chemical and Physical
Characteristics of Emissions From Forest Fires," Paper
7 6-2.3, Presented at the 69th Annual Meeting of the Air
Pollution Control Association, June 27-July 1, 1976, Port-
land, Oregon.
82. Sawicki, E., Meekes, J.E., and Morgan, M.J., "The Quanti-
tative Composition of Air Pollution Source Effluents in
Terms of Aza Heterocyclic Compounds and Polynuclear Aro-
matic Hydrocarbons," International Journal of Air and Water
Pollution, 9: pp. 291-298, 1965.
83. Sawyer, James W., Automotive Scrap Recycling: Processes,
Prices, and Prospects, Resources for the Future, Johns
Hopkins Press, Maryland, 1974.
84. Schuler, P. and Bierbarten, P., Environmental Surveys of
Aluminum Reduction Plants, NIOSH 74-101, PB 232-776,
April 1974.
85. Sharkey, A.G., Schultz, J.L., White, C., and Lett, R.,
Analysis of Polycyclic Organic Material in Coal, Coal
Ash, Fly Ash, and Other Fuel and Emission Samples, EPA-
600/2-76-075, March 1976, Durham, North Carolina.
86. Shurr, S., and Netschert, B.C., Energy in the American Economy
1850-1975, University Press, 1960.
87. Shuster, W.W., "Partial Combustion and Pyrolysis of Solid
Organic Wastes," Presented at the 68th Annual Meeting of the
Air Pollution Control Association, Paper No. 75-38.2,
June 15-20, 1975, Boston, Massachusetts.
88. Smith, W.M., "Evaluation of Koke Oven Emissions," J. Occup.
Med., 13_(2) : 69-74, February 1971.
89. Snowden, W.D., Alguard, D.A., et al., Source Sampling;
Residential Fireplaces for Emission Factor Development,
EPA-450/3-76-00, November 1975, Durham, North Carolina.
90. Sommerer, D., and Laube, A.H., "Organic Material From a
Coke Quench Tower," Presented at the Energy and the Environ-
ment Conference, November 1977, Cincinnati, Ohio.
91. Sommerer, D., York Research Corporation, Personal Communi-
cation, December 1977, Stamford, Connecticut.
192
-------
92. Spindt, R.S., Study of Polynuclear Aromatic Hydrocarbons
Emissions From Heavy Duty Diesel Engines, PB 238-688,
July 1974.
93. Springer, K.L., Baines, T.O.M., "Emissions From Diesel Ver-
sions of Production Passenger Cars," Paper 770818, Presented
at Society of Automotive Engineers Passenger Car Meeting,
September 26-20, 1977, Detroit, Michigan.
94. Suprenant, N., Hall, R. et al., Preliminary Emissions
Assessment of Conventional Stationary Combustion Systems,
Volume II, Final Report, EPA-600/2-76-046b, March 1976.
95. Systems Study of Air Pollution From Municipal Incineration,
Arthur D. Little, Incorporated, NAPCA Contract No. CPA-22-69-230,
March 1970.
96. Temple, Barker, and Sloane, Incoporated, Intergrated Iron and
Steel Industry, Volume I, EPA-230/3-76-014b, December 1976.
97. Thompson, R.E., et al., "Effectiveness of Gas Recirculation
and Staged Combustion in Reducing NO on a 560 MW Coal-Fired
Boiler," Proceedings of the NO Control Technology Seminar,
EPRI SR-39, February 1976. *
98. U.S. Bureau of Census, Census of Population 197 0, Number of
Inhabitants, United States Summary, 1973.
99. U.S. Bureau of Census, City and County Data Book 1970,
1973.
100. U.S. Bureau of Mines, Automobile Disposal, A National
Problem, 1967, Washington, D.C.
101. U.S. Bureau of Mines, "Coke Producers in the U.S. in 1973,"
Mineral Industry Surveys, September 1974, Washington,D.C.
102. U.S. Bureau of Mines, Iron and Steel Scrap in the Southeast.
103. U.S. Bureau of Mines, "Sales of Fuel Oil and Kerosine,
Annual," Mineral Industry Surveys, 1976, Washington, D.C.
104. U.S. Bureau of Mines, "Sales of Asphalt in 1976," Mineral
Industry Surveys, 1977, Washington, D.C.
193
-------
105. U.S. Department of Commerce, Domestic and International
Business Administration, U.S. Industrial Outlook 1977,
1977, Washington, D.C.
106. U.S. Department of Commerce, Bureau of the Census, 1974
Housing Inventory, 1974, Washington, D.C.
107. U.S. Department of Transportation, Federal Highway Adminis-
tration, National Highway Inventory and Performance Study,
1975, OMB No. 04-575038, July 1975.
108. U.S. Environmental Protection Agency, The Automobile Cycle:
An Environmental and Resource Reclamation Problem (SW-80+s.1),
1972.
109. U.S. Environmental Protection Agency, Compilation of Air
Pollution Emission Factors, Second Edition, AP-42, 1975.
110. U.S. Environmental Protection Agency (OAQPS), Preferred
Standards Path Report for Polycyclic Organic Matter,
October 1974, Durham, North Carolina.
111. U.S. Environmental Protection Agency, Scientific and
Technical Assessment Report on Particulate Polycyclic
Organic Matter (PPOM) , EPA-600/6-75-001,, March 1975,
Washington, D.C.
112. Vandegrift, A.E., Shannon, L.J., et al-. , Handbook of
Emissions, Effluents, and Control Practices for Stationary
Particulate Pollutant Sources, Report of NAPCA Contract No.
CPA 22-69-104, November 197 0.
113. White, L.D., "Coke-Oven Emissions, Occupational Health and
the Environment," Master's Thesis, University of Cincinnati,
1972, Cincinnati, Ohio.
114. White, L.D., Stasikowski, M.J., Horstman, S.W., and
Bingham, E., "Atmospheric Emissions of Coking Operations,
A Review," Presented at the AIHA Conference, Mary 1973
(updated for publication May 1975)..
115. Winer, D., Division of Timber Management, U.S. Forest
Service, Personal Communication, December 1977.
116. Yamate, G., Stockham, J., and Vatavuk, W., "Emission
Factors for Forest Wildfires," Paper prepared from work
done for EPA Contract No. 68-02-0641, 1974, California.
194
-------
117. Zelinsky, W., Dept. of Geography, Penn State University,
Personal Communication, January 1978, University Park, Pen-
nsylvania .
118. Suta, B.E., Human Population Exposures to Coke-Oven Atmospheric
Emissions, SRI Project No. 5794 Final Report for EPA Contract
No. 68-01-4314, November 1977.
119. Stahley, S., Dept. of Chemistry, University of Maryland,
Personal Communication, February 1978, College Park,
Maryland.
120. Strate, -H.E., Nationwide Personal Transportation Study,
Annual Miles of Automobile Travel, Report No. 2, U.S. Dept.
of Transportation, Federal Highway Administration, April
1972, Washington, D.C.
121. U.S. Office of Highway Planning, Fuel Consumption 1975, 1976.
122. National Coal Association, Steam Electric Plant Factors 1976,
1977, Washington, D.C.
123. Energy and Environmental Analysis, Incorporated, Market-
Oriented Planning Study (MOPPS), Draft Final Report for
Energy and Development Administration, September 1977.
124. Wright, American Petroleum Institute, Personal Communication,
March 1978, Washington, D.C.
125. U.S. Dept. of Commerce, Statistical Abstract of the United
States, 1976, 1977, Washington, D.C.
126. Hanna, Steven R., "A Simple Method of Calculating Dispersion
From Urban Area Sources," Journal of the Air Pollution Control
Association, 21^(12) : 774-777, December 1971.
127. Gifford, F.A., "Applications of a Simple Urban Pollution
Model," Proceedings of the American Meteorology Society
Conference on Biometeorology, November 1972, Philadelphia.
128. U.S. Environmental Protection Agency, "Standards of Performance
for New Stationary Sources," Federal Register, Volume 36,
Number 247, December 23, 1971.
129. Trenholm, Andrew R., Standards Support Section, Industrial
Studies Branch, ESED, DPA, "Benzene Emissions From Coke
Ovens," Memo to William F. Hamilton, Chief, Economic Analysis
Branch, SASD, EPA, October 1977, Durham, North Carolina.
195
-------
GRAFT
00 NOT QUOTE OR CITE
130. Trenholm, Andrew R., and Beck, Lee L., "Assessment of
Hazardous Organic Emissions From Slot-Type Coke Oven
Batteries," Internal EPA Report, March 16, 1978, Durham,
North Carolina.
131. U.S. Bureau of Census, Population Estimates and Projections
1977-2050, Series P-25, No. 7.04, July 1977.
132. Pitts, James N., Jr., Grossjean, Daniel R., et al., "Mutagenic
Activity of Nitrogenous Pollutants in Ambient and Simulated
Atmospheres," Presented before the Division of Environmental
Chemistry of the American Chemical Society, March 12-17,
1978, Anaheim, California.
133. Grossjean, Dr. Daniel R., Statewide Air Pollution Research
Center, University of California, Personal Communication,
April 1978, Riverside, California,.
134. Bornstein, Mark, GCA Technology Division-, Boston, Massachusetts,
Personal Communication, March 8., 1978.
135. Seizinger, D. E., Bartlesville Energy .Research Center, DOE,
Bartlesville, Oklahoma, Personal Communication, April 2, 1978.
136. Spindt, R. S., Gulf Research and Development Company, Pittsburgh,
Pennsylvania, Personal Communication, March 23, 1973.
137. Eimutis, E. C. & Quill, R. P., Source Assessment: Non-criteria
Pollutant Emissions, PB 270-550, July 1977
138. Gundel, L., Chang, S.-G., & Nfovakov, T.,, "Heterogeneous Reactions
of Polynuclear Aromatic Hydrocarbons and Soot Extracts with ND-,"
Lawrence Berkeley Laboratory. Energy & Environment Division.
Atomospheric Aerosol Research Annual Report 1976-1977, LBL-6819,
TID-4500-R66, 1977.
139. Magda, Niren, Geomet, Inc., Gaithersburg, Maryland, Personal
Communication, November, December 19 77.
140. Kneip, T. J., Lezko, M. A., et al., "Organic Matter in New York
City TSP. Comparisons of Seasonal Variations and Relationships
to Source Tracers," Presented at the Conference on Carbonaceous
Particles in the Atmosphere, Berkeley, California, March 20-22,
1978.
141. Lipkea, . W;>. H_.t, Johnson, J. H. , & Vuk, C. T. , "The Physical and
Chemical Character of Diesel Particulate Emissions - Measurement
Techniques and Fundamental Considerations, SP-4 30, Society of
Automotive Engineers, Congress and Exhibition, Detroit, Michigan,
February 27-March 3, 19 78.
196
-------
142. Tejada, Sylvestre, ESRL, EPA, Personal Communication, March 19 78.
143. Zengel, A. E., The Coordinating Research Council, New York, New
York, Personal Communication, March 1973.
144. Johnson, Larry, Process Measurement Branch, IERE, EPA, Personal
Communication, March 1978.
145. Bowen, Joshua, Chief, Combustion Research Branch, IERL, EPA,
Personal Communication, March 19 78.
146. Reznik, Dick, Monsanta Research Corp., Dayton, Ohio, Personal
Communication, March 1978.
14 7. Weinstein, Norm, ReCon Systems, Princeton, New Jersey, Personal
Communication, March 1978.
14 8. Perhac, Ralph, Environmental Assessment Department, EPRI, Palo
Alto, California, Personal Communications, March 1978.
149. Turner, P. P., Chief, Advanced Processes Branch, IERL, EPA,
Personal Communication, March 19 78.
150. Barush, Steve, EPRI, Palo Alto, California, Personal Communication,
March 19 78.
197
------- |