CORPORATION
DCN 203-012-07-04
AMBIENT CONCENTRATIONS OF
POLYCYCLIC ORGANIC MATTER
Final Technical Note
EPA Contract No. 68-02-3818
Work Assignment 7
Prepared for:
Ray Morrison
EPA Project Officer
Pollutant Assessment Branch
Office of Air Quality Planning and Standards
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
U.S. Environmental Protection Agency
Region V, Library
230 South Dearborn Street
Chicago, Illinois 60604
Prepared by:
S. A. Smith
Radian Corporation
31 October 1983
8501 Mo-Pac Blvd. / P.O. Box 9948 / Austin, Texas 78766 / (512)454-4797
-------
EiMKmmentai Protection Agenej
-------
CORPORJmOM
DISCLAIMER
This report was furnished to the Environmental Protection Agency by
Radian Corporation, 8501 MoPac Boulevard, Austin, Texas 78766, in fulfillment
of Contract No. 68-02-3818, Work Assignment No. 7. The opinions, findings and
conclusions expressed herein are those of the authors and not necessarily
those of the Environmental Protection Agency. Mention of company or product
names should not be considered an endorsement of same by the Agency.
11
-------
CORPORATION
TABLE OF CONTENTS
Page
1.0 INTRODUCTION 1-1
2.0 SUMMARY 2-1
3.0 POM AMBIENT CONCENTRATION DATA 3_1
3.1 Measured Levels - General Observations 3-5
3.1.1 Background Levels 3-5
3.1.2 Urban Concentrations 3-6
3.1.3 Influence of Specific Sources 3-6
3.1.4 Seasonal Variation 3-7
3.1.5 Diurnal Variation 3-8
3.1.6 Particle Size Distribution 3-9
3.1.7 BaP as an Indicator for POM 3-9
3.2 Presentation of Specific Data 3-10
3.2.1 White and Vanderslice, 1980 (The RTI Report). . . 3-10
3.2.2 Handa et al., 1980 3-12
3.2.3 Katz and Chan, 1980 3-15
3.2.4 Hornig et al. , 1981 3-18
3.2.5 Greenberg and Darack, 1982 3-25
3.2.6 Daisey, Hershman, and Kneip, 1982 3-28
3.2.7 Harkov, Daisey, and Lioy, 1983 3-32
3.2.8 Manning, Imhoff, and Akland, 1983 3-36
3.2.9 Soderberg et al. , 1983 3-40
3.3 Trends 3-47
3.4 References for Section 3.0 3-49
4.0 SAMPLING AND ANALYTICAL CONSIDERATIONS 4-1
4.1 Sampling Methodologies . . 4-1
111
-------
TABLE OF CONTENTS (Continued)
Paee
4.1.1 Collection Technique Precision 4-1
4.1.2 Incomplete Collection 4-2
4.1.3 Location of the Sampling Point 4-3
4.1.4 Sample Frequency 4-3
4.2 Analytical Techniques 4-4
4.3 References for Section 4.0 4-6
iv
-------
RADIAN
LIST OF TABLES
Table Title Page
3-1 Summary of POM Compounds Detected in Air Pollution
Samples in Six Recent Studies 3-4
3-2 Annual Ambient BaP Concentrations at NASN Stations.... 3-11
3-3 Variations in Seasonal Averages of BaP Concentrations 3-13
3-4 POM Concentration Reflecting the Dominance of a
Single Source 3-14
3-5 Data Obtained with CCS-b and a High-Volume Air
Sampler at lidabashi in Winter 1977 3-16
3-6 Data Obtained with CCS-b at lidabashi in Summer 1978. . . 3-16
3-7 Comparison of Average Concentrations of PAH in Air-
borne Particulates Collected by Andersen Cascade
Impactor Vs. High-Volume Sampler at Hamilton,
Ontario, Sampling Period, January-May, 1978 3-17
3-8 Comparative Seasonal Concentration Levels of PAHs
in Airborne Particulates, West-Central Station in
Hamilton, Ontario (Cascade Impactor Samples) 3-19
3-9 Average Concentrations of PAHs and Percentages of
Total in Particulate Size Ranges at West-Central
Station in Hamilton, Ontario 3-20
3-10 Annual Average PAH Concentrations in Air of Hamilton,
Ontario, New York City, and Los Angeles 3-21
3-11 POM Concentrations at Several Locations in New
Hampshire in the Winter of 1979-1980 3-23
3-12 Ambient BaP Concentrations in Various Locations 3-24
3-13 Concentrations of PAH Obtained from Newark, Elizabeth
and Camden - Summer, 1981 Composites 3-26
3-14 Averages of Daily PAH Data at Four New Jersey Loca-
tions During Summer, 1981, and Comparison with
Composite Data 3-27
3-15 Seasonal Variations in Respirable and Non-Respirable
Particulate Organic Matter in New York City 3-30
v
-------
LIST OF TABLES (Continued)
Table Title Page
3-16 A Comparison of 1976, 1977, and 1978 Measurements of
TSP and Particulate Organic Matter to Measurements for
1968 and 1969 at the Same Site in New York City 3-31
3-17 Geometric Means for Selected IPM Constituents at Four
Sites in New Jersey 3-34
3-18 Peak Levels of Particulate Organic Matter in New Jersey
During Summer and Winter Pollution Episodes 3-35
3-19 Ambient TSP and BaP Concentrations at Florence,
Alabama, December 23, 1981 - March 22, 1982 3-37
3-20 First and Fourth Quarter BaP Concentrations at Avail-
able Southeastern Sites, 1981-1982 3-38
3-21 Ambient Concentrations of PaH at Florence, Alabama,
December 23, 1981 - March 22, 1983 3-39
3-22 Comparison of Soxhlet and Ultrasonic Extraction by
Three Different Solvents 3-43
3-23 Comparison of Solvents for Filter Extraction by
Ultrasonic Agitation 3-44
3-24 Recovery of POM Using Preferred Purification Scheme. . . 3-45
3-25 Comparison of GC and HPLC Analysis of POM 3-46
LIST OF FIGURES
Figure Page
3-1 Average PAH*Profiles: Lyme Center, Brigham Hill,
Hanover (Dartmouth College), Hanover (Residential
Site), January 15-31, 1982 3-41
vi
-------
em
1.0 INTRODUCTION
The purpose of this technical note is to briefly summarize
existing published polycyclic organic matter (POM) ambient concentration
data. A review and analysis of POM source and ambient concentration data
prepared for EPA by the Research Triangle Institute (RTI) in 1980 (EPA-600/7-
80-044)* provided a base set of data for this report. Eight additional
studies published in periodicals or presented at technical conferences sub-
sequent to the publication of the RTI report furnish more recent data on POM
concentrations in urban and rural areas of the United States, a coke oven
area in Canada, and a large urban center in Japan. These studies update the
information and largely substantiate conclusions presented in the RTI report.
As defined in the RTI report, POM is a generic term applied to a
large group of fused-ring organic compounds. In general, POM refers to
those organic compounds consisting of two or more fused aromatic rings.
The rings may either be comprised totally of carbon atoms or may contain
hetero atoms of nitrogen, oxygen, and sulfur, in addition to other ring
substitutes. Due to the large possible number of ring combinations and
substitute permutations, the theoretical number of POM compounds can run
into the millions; however, only approximately 100 have been identified
in a single ambient air sample.*
There is no one accepted technique to analyze for POM content in
air pollution samples. Many researchers analyze for specific POM compounds,
which probably underestimates total POM concentrations. A few researchers
analyze for the total organic fraction of captured particulate matter, which
probably overestimates total POM concentrations.
*White, J. B. and R. R. Vanderslice. (Research Triangle Institute). POM
Source and Ambient Concentration Data: Review and Analysis. Prepared for
U. S. Environmental Protection Agency. EPA-600/7-80-044. March 1980.
1-1
-------
Although information is provided on the general sampling and
analysis techniques employed for collection of POM ambient data, no attempt
has been made to screen the data based on research methodology or to compare
results of different studies. The data are presented as reported and demon-
strate the range and variability of measured levels of POM occurring in the
ambient air environment. No information is provided with respect to trans-
formation/transport of POM in the atmosphere. The data are therefore useful
only for preliminary evaluations of the potential public health consequences
of exposure to POM in the ambient air.
Section 2.0 summarizes some general observations that can be made
based on the data. POM ambient concentration data are presented in Section
3.0. Section 4.0 discusses the quality of the data and highlights sampling
and analytical problems that influence the accuracy of estimates.
1-2
-------
RADIAN
2.0 SUMMARY
A review of recent research on ambient concentrations of POM
suggests the following observations:
1. Researchers use a diversity of techniques and procedures
to analyze for POM content in air pollution samples. It
is therefore misleading to make direct comparisons among
the results of recent studies of POM ambient concentra-
tions. No two researchers cited in this report have
analyzed for the same array of individual POM compounds.
Only two recent studies present data for total POM con-
centrations and these studies define POM as particulate
organic matter, which includes more compounds than poly-
cyclic organic matter. Other recent studies typically
present the results of analyses for specific POM com-
pounds, most often BaP and other polycyclic aromatic
hydrocarbon (PAH) compounds. To further complicate a
comparison of results, most studies analyzed for POM
content in collected particulates; however, some studies
attempted to collect both particulate and vapor-phase POM.
2. POM consists of two categories of compounds on the basis
of the atomic constituents of the ring structures:
polycyclic aromatic hydrocarbons (PAH) and heterocyclic
polynuclear aromatics. The latter category, which
includes aza arenes, oxa arenes, and thia arenes, is
the least studied of the two. All of the studies cited
in this report, except the two studies which defined POM
as particulate organic matter, present results for con-
centrations of PAHs only.
2-1
-------
3. Based on available data on BaP concentrations, normal
background ambient air concentrations for POM in remote
areas appear to be about 0.2 ng/m3. or less.
4. POM concentrations in rural areas have been reported in
different studies to range from 0.3 ng/m3 (24-hr average)
to 7,000 ng/m3(geometric mean for test period), reflect-
ing the influence of POM transport from emission sources
such as wood stoves, industrial plants, and urban centers.*
5. Urban atmospheric levels of POM have been reported in
different studies to range from 2 ng/m3 to 22,000 ng/m3.*
Ambient levels depend more on the type and concentration
of emission sources than the size of the urban center.
6. Coke oven areas show consistently higher POM levels than
non-coke oven areas. The ratio of POM concentrations in
coke oven/non-coke oven areas ranges from 1.4 to 16.8.
7. Many studies have shown a pronounced seasonal variation
in POM concentrations, with winter levels at some sites
more than ten times higher than summer levels. However,
other researchers dispute this finding, claiming that
lower levels in summer are caused by increased volatili-
zation and/or chemical degradation during sampling.
8. At least one researcher has reported significantly higher
concentrations of nonpolar and moderately polar fractions
of POM during daylight as compared to nighttime concentra-
tions. No explanation was offered for the day-night
variation.
*The highest POM concentrations were reported in studies which defined POM
as particulate organic matter rather than polycyclic organic matter.
2-2
-------
9. POM exhibits preferential enrichment on smaller size
particles, particularly in winter. Approximately 50
percent to more than 80 percent of the measured POM re-
portedly may reside on respirable particles (< 3.5 ym).
10. Benzo(a)pyrene (BaP) is not a reliable indicator of total
POM due to its reactivity in the atmosphere. Other prob-
lems with using BaP are variable BaP/POM ratios in ambient
air and the trend of declining BaP emissions observed in
recent years.
11 BaP ambient concentrations have been declining over the
last 15 years due in part to controls on emissions from
coke ovens, utilities, automobiles, and open burning.
Ambient concentrations of POM compounds could increase
in some areas, however, due to residential and industrial
wood burning, combustion of refuse-derived fuel,
vehicular traffic, and use of diesel fuel.
12. The accuracy of much ambient concentration data is
uncertain due primarily to sampling and analytical
problems. Sampling accuracy is influenced by the
precision of the sampling apparatus, collection effi-
ciency, potential filter losses, location of the sam-
pling equipment, and frequency of sampling, extraction,
and detection.
2-3
-------
3.0 POM AMBIENT CONCENTRATION DATA
POM ambient concentration data are presented as reported in nine
recent studies. The studies are:
1. White, J. B. and R. R. Vanderslice (Research Triangle
Institute). POM Source and Ambient Concentration Data:
Review and Analysis. Prepared for U. S. Environmental
Protection Agency. EPA-600/7-80-044. March 1980.
2. Handa, T., Y. Kato, T. Yamamura, and T. I. Shii (Science
University of Tokyo). "Correlation Between the Concen-
trations of Polynuclear Aromatic Hydrocarbons and Those
of Particulates in an Urban Atmosphere." Environmental
Science & Technology, Vol. 14, No. 4, April 1980. pp. 416-
422.
3. Katz, M. and C. Chan (York University, Toronto, Ontario).
"Comparative Distribution of Eight Polycyclic Aromatic
Hydrocarbons in Airborne Particulates Collected by Con-
ventional High-Volume Sampling and by Size Fractionation."
Environmental Science & Technology, Vol. 14, No. 7, July
1980. pp. 838-843.
4. Hornig, J. F., R. H. Soderberg, D. L. Larsen (Dartmouth
College) and C. Parravano (State University of New York,
College at Purchase). "Ambient Air Assessment in a Rural
New England Village Where Wood Is the Dominant Fuel." In:
Proceedings: Conference on Wood Combustion Environmental
Assessment, New Orleans, LA, February 21-24, 1981.
Prepared by Research Triangle Institute for the U.S.
Environmental Protection Agency. EPA 600/9-81-029
(PB81 248155). May 1981.
3-1
-------
5. Greenberg, A. and F. B. Darack (New Jersey Institute of
Technology). "Concentrations of Polycylic Aromatic
Hydrocarbons at Four New Jersey Sites During An Extended
Summer Sampling Campaign." Presented at the 75th Annual
Meeting of the Air Pollution Control Association, New
Orleans, LA, June 20-25, 1982.
6. Daisey, J. M., R. J. Hershman, and T. J. Kneip (New York
University Medical Center). "Ambient Levels of Particulate
Organic Matter in New York City in Winter and Summer."
Atmospheric Environment, Vol. 16, No. 9, 1982. pp. 2161-
2168.
7. Harkov, R. (New Jersey Department of Environmental
Protection), J. M. Daisey and P. J. Lioy (New York
University Medical Center). "Comparisons Between Summer
and Winter Inhalable Particulate Matter, Fine Particulate
Matter, Particulate Organic Matter, and S04 Levels at
Urban and Rural Locations in New Jersey." Presented at
the Air Pollution Control Association Specialty Confer-
ence on Measurement and Monitoring of Non-Criteria
(Toxic) Contaminants in Air, Chicago, IL, March 22-24,
1983.
8. Manning, J. A., R. E. Imhoff (Tennessee Valley Authority)
and G. G. Akland (U. S. Environmental Protection Agency).
"Wintertime Ambient Measurements of Particulate Polycyclic
Aromatic Hydrocarbons in a Residential Community." Pre-
sented at the Air Pollution Control Association Specialty
Conference on Measurement and Monitoring of Non-Criteria
(Toxic) Contaminants in Air, Chicago, IL, March 22-24, 1983.
3-2
-------
9. Soderberg, R. H., J. F. Horning, A. Barefoot (Dartmouth
College, and C. Parravano (State University of New York
at Purchase). "Measurements of Polycyclic Aromatic
Hydrocarbons in Ambient Air Particulates in Northern
New England." Presented at the Air Pollution Control
Association Specialty Conference on Measurement and
Monitoring of Non-Criteria (Toxic) Contaminants in
Air, Chicago, IL, March 22-24, 1983.
White and Vanderslice (the RTI report) provide a base set of data on POM
ambient concentrations reported in studies published prior to 1978. The
eight additional studies furnish updated information on measured levels of
POM in the ambient atmosphere in areas of the United States, Canada, and
Japan.
Two of these studies (Daisey, Hershman, and Kneip, 1982 and Harkov,
Daisey, and Lioy, 1983) present data on total POM concentrations. However,
these authors define POM as particulate organic matter rather than polycyclic
organic matter. Their data, therefore, represent the organic fraction of
particular matter, which includes organic compounds other than polycyclic
organic matter. Although these data probably overestimate the concentration
of polycyclic organic matter in ambient air, they can be viewed as an upper
bound for ambient concentrations of POM.
The other six studies present the results of analyses for specific
POM compounds, most often benzo(a)pyrene (BaP) and other PAHs. Table 3-1
identifies the POM compounds analyzed for and detected in each of these six
studies. No two studies report the same array of individual POM compounds
and no one study analyzed for more than 14 individual compounds. Further-
more, none of the studies present results for aza arenes, oxa arenes, or
thia arenes, which comprise an important category of POM, the heterocyclic
polynuclear aromatics. The results of these studies, therefore, probably
underestimate the concentration of total POM in ambient air.
3-3
-------
S3
O
H
s
[j
0
H
23
M
Q
W
H
CJ
W 00
H fd
W 1-1
o a
!=>
i/l H
is
Pj W
O W
o cd
2 X
O M
PU V3
f r j *^
O H- 1
Pi w
jg §
00 00
-H
1
W
EH
CO
U r-l
01
4) i 1
O
CO 4J
U-l CO
*O iH
S "
oo ffl
*c "^
C T3
C 00
(a ch
00
ai J.T
C B
0) VJ
01 n)
ij a
00 "
H ! r-l
g Ǥ
O w rH
* 01
o
as
CO ?-4
N "
*J C
, X
CJ
ca
a ^H o
C C5 CX)
= S3
3
0
B
O
u
% ^ ^ >, ^ %.
-V ^ ^> ^ ^ ^
>!. "^i. ^i. ^ """i. "^ ^>
^ ^ ^ ^ ^
^ N. ^ N» "^ %.
a) aj at c aj
S "5 5 £ c 5 ^
rt£22'Sa,Q.cd
BC *-s ^-> QJ £U
Cfl 4_l * J_, )_,
_C
-------
RADIAN
Section 3.1 summarizes general observations on measured levels of
POM in the ambient air from these nine studies. Section 3.2 presents specific
data and abbreviated excerpts from the reported research. Trends relating to
ambient POM concentrations that are suggested by this recent research are dis-
cussed in Section 3.3. References used in addition to the nine studies pri-
marily reviewed for this report are listed in Section 3.4. The studies listed
above are referred to by author name and publication data in the discussion
that follows.
3.1 Measured Levels - General Observations
3.1.1 Background Levels
Natural sources of POM, such as forest fires and volcanoes, can be
considered to produce a natural background level of POM. Although it is im-
possible to separate the contributions of natural sources from those originat-
ing from the dispersal and long-range transport of POM from anthropogenic
sources, measured levels in remote areas are indicative of background concen-
trations. The RTI report (White and Vanderslice, 1980) concludes that normal
background ambient air concentrations for POM in remote areas appear to be
about 0.2 ng/m3 or less (based on BaP concentrations).
None of the other eight studies reviewed report data for remote
areas, although several do report data for rural areas. Hornig, et al. (1981)
reported levels totaling approximately 6 ng/m3 for 12 POM compounds in two
rural areas in New Hampshire (average of 48-hr data). Levels of from 0.3 ng/m3
to 16.1 ng/m3 of ten PAH compounds in rural Alabama are reported by Manning,
Imhoff, and Akland (1983) (24-hr averages). Harkov, Daisey, and Lioy (1983)
found levels as high as 4,400 ng/m3 of POM (defined as particulate organic
matter) in inhalable particulate matter in rural New Jersey in summer, and
7,000 ng/m in winter (geometric means for test period). Evidently, POM
concentrations at rural sites can be heavily influenced by transport from
potential POM emissions sources such as residences (wood stoves), industrial
plants, and urban centers in the region.
3-5
-------
RADIAN
3.1.2 Urban Concentrations
The RTI report concludes that urban atmospheric levels of POM may
be 10 to 100 times higher than levels in remote areas, ranging from 2 to 20
ng/m3. The highest concentrations are reported to be in coke oven areas.
Much recent data indicate higher levels of POM in non-coke oven urban areas.
Handa, et al. (1980) found 50 to 60 ng/m3 of six PAH compounds in an urban
area of the Tokyo Metropolitan Area (24-hr averages). Daisey, Hershman, and
Kneip (1982) reported average levels of total POM (defined as particulate
organic matter) on the order of 10,000 to 13,800 ng/m3 in New York City,
with levels reaching 22,000 ng/m3 in the winter of 1977 when high sulfur
fuel oil was burned during a fuel shortage in utility, commercial, and resi-
dential boilers. POM (defined as particulate organic matter) concentrations
found by Harkov, Daisey, and Lioy (1983) in inhalable particulate matter at
three urban areas in New Jersey corroborate the high levels reported in New
York City.
3.1.3 Influence of Specific Sources
The RTI report presents data to demonstrate that POM levels (as
measured by BaP) relate to the nature and degree of industrial and public
activities, types and relative quantities of fuels consumed, degree of regu-
lation exercised by authorities over emissions, volume of vehicular traffic,
and extent to which photochemical and other reactions occur in the atmosphere.
The chemical composition of POM will reflect combustion characteristics of
each individual source or the dominance of a single type of source.
Areas near coke ovens tend to exhibit the highest reported levels
of POM compounds. The U.S. Environmental Protection Agency conducted a com-
parative study of BaP concentrations in the air of cities with coke ovens
over the period 1966-1972 (1). The coke oven cities were found to show con-
sistently higher BaP levels in every year, with the ratios of coke oven/non-
coke oven cities ranging from 1.42 to 3.34 and averaging about 2. The levels
3-6
-------
of PAH found by Katz and Chan (1980) in a coke oven area in Canada (Hamilton
Ontario) generally agree with EPA's findings. With the exception of anthan-
threne, they report individual PAH concentrations from 1.8 to 16.8 times
higher than corresponding PAH levels in New York City and Los Angeles (non-
coke oven areas). Anthanthrene was slightly lower in Hamilton.
High automobile traffic density and high concentrations of wood
burning sources also contribute to relatively higher ambient POM levels,
particularly of specific compounds associated with these emissions sources.
Coronene and benzo(g,h,i)perylene [B(ghi)Pj have been shown to correlate
with automobile traffic density better than other PAH (2). Greenberg and
Darack (1982) found relatively high B(ghi)P/BaP ratios at urban New Jersey
sites compared to a rural site. However, the coronene/BaP ratio was not
consistently higher at the urban sites.
Hornig et al. (1981) suggest that high concentrations of benz(a)
anthracene and chrysene identify wood smoke as a source of POM. However,
they do not compare their data from wood burning areas in New Hampshire with
other non-wood burning areas. These researchers are currently studying
levoglucosan (l,6-anhydro-B-(j>-glucopyranose) as a potentially useful qual-
itative and semiquantitative tracer for wood smoke.
3.1.4 Seasonal Variation
The RTI report presents data that show pronounced seasonal varia-
tion in POM concentrations, with winter levels of BaP at ten urban NASN sites
more than ten times higher than summer levels. This effect is assumed to be
a result of increased residential fuel combustion and local meteorology.
However, in one study (3) discussed in the RTI report, the data indicate sub-
stantial seasonal variations in BaP in a coking region of Belgium even though
there was no apparent change in the production of the source. In a subsequent
series of wintertime experiments with filters heated from -2°C to 28°C, the
authors were able to duplicate the seasonal trend in BaP. They concluded that
3-7
-------
RADIAN
the seasonal trends in BaP concentrations detected in the coking region of
Belgium were probably due to volatilization and/or chemical reactions on the
filter surface that were catalyzed by trace elements. The authors did not
discuss the possible effects of local meteorology on the observed seasonal
variation.
Katz and Chan (1980) also present data from winter and summer sam-
pling in a coke-oven area that show pronounced seasonal variation. (Winter
levels were generally higher.) They used a cascade impactor with a low flow
rate to minimize losses of PAH that may occur by sublimation from the deposit
of particulate matter. The authors do not offer any explanation for the
variation.
Daisey, Hershman, and Kneip (1982) report that atmospheric concen-
trations of total POM (defined as particulate organic matter) in New York
City averaged 13,000 ng/m3 for two summer periods and 16,000 ng/m3 for two
winter periods. After adjusting the data to account for dispersion, the
authors conclude that POM emissions are three to four times higher in winter
than in summer, primarily due to fuel combustion for space heating. Data
reported by Harkov, Daisey, and Lioy (1983) for urban sites in New Jersey
agree fairly well with this New York City data with respect to seasonal
variation.
Data presented by Handa et al. (1980) for an urban area in the
Tokyo Metropolitan Area, however, do not show a pronounced seasonal varia-
tion. They used a collection system with traps cooled by liquid nitrogen
to collect the volatilized fraction of PAH.
3.1.5 Diurnal Variation .
Daisey, Hershman, and Kneip (1982) investigated day-night differ-
ences in ambient concentrations of POM (defined as particulate organic matter)
in New York City. They reported that nonpolar (cyclohexane-soluble) and
3-8
-------
moderately polar (dichloromethane-soluble) fractions were found in
significantly higher concentrations during daylight, while the polar
(acetone-soluble) fraction showed no difference between day and night.
The authors offer no explanation for the observed day-night variation.
3.1.6 Particle Size Distribution
Most of the studies cited in this report analyzed for POM content
in collected particulates. Several researchers have noted the preferential
enrichment of POM on smaller diameter particulates. This phenomenon is
likely the result of the greater surface-to-volume ratios of small particles.
Tests using particulate sizing techniques on urban aerosols have demonstrated
that as much as 75 percent of the total BaP adsorbed onto particulate matter
can reside on particulates with diameters less than 2.3 ]Jm (4,5,6,7). Katz
and Chan (1980) report that from 72.1 to 88.8 percent of individual PAH com-
pounds measured for this study resided in the particle size range less than
3.3 ym. Data presented by Daisey, Hershman, and Kneip (1982) suggest a shift
in the distribution of POM (defined as particulate organic matter) toward
smaller size particles in winter. The authors report that respirable parti-
cles (^3.5 ]im) contained a greater percentage of total POM in winter (81%)
than in summer (54%).
3.1.7 BaP as an Indicator for POM
The RTI report concludes that neither long- nor short-term studies
using benzo(a)pyrene as an indicator of total POM are reliable for quantita-
tive ambient air quality estimates. Underestimation of total POM results
from the failure to consider the apparent rapid decomposition rate of BaP,
as well as the variable BaP/POM ratios in ambient air. Katz and Chan (1980)
point out that BaP is an especially poor index of airborne POM in cities
where motor vehicle traffic is one of the major sources of air pollution.
In these areas, benzo(g,h»i)perylene is usually the dominant POM. In addi-
tion there has been an apparent decline in BaP concentrations in ambient air
3-9
-------
RADIAN
over the last 15 years primarily resulting from controls on open burning and
coke ovens (8). For these reasons, researchers now usually analyze ambient
air samples for a variety of organic components.
3.2 Presentation of Specific Data
3.2.1 White and Vanderslice, 1980 (The RTI Report)
The RTI report reviews and analyzes POM ambient concentration data
from more than 35 literature sources published between 1954 and 1977. The
data are international in scope. The report cautions that ambient POM esti-
mates, particularly those based on particulate sampling, contain a high de-
gree of uncertainty due to sampling inaccuracies and chemical degradation of
POM on the filter surface. Since the accuracy of the data base is not known,
the report recommends that the data be considered semiquantitative, with
measured POM concentrations best categorized as high, medium, or low.
The report draws four general conclusions from the ambient
concentration data:
1. Normal background ambient air concentrations for POM
in remote areas appear to be ^ 0.2 nanograms/m . This
conclusion is based on annual ambient BaP concentrations
at three NASN remote stations, which averaged 0.1 nano-
grams/m3 in 1976 (see Table 3-2). Urban atmospheric
levels of POM may be 10 to 100 times higher.
2. Atmospheric POM concentrations as indicated by ambient
BaP levels appear to be declining and are considered
significantly less than the levels recorded 10 years
ago. The data in Table 3-2 also demonstrate this
trend.
3-10
-------
RADIAN
TABLE 3-2. ANNUAL AMBIENT BaP CONCENTRATIONS AT NASN STATIONS (ng/m3)
Location
Honolulu
Chicago
Montgomery
New Orleans
Baltimore
Detroit
New York
Youngs town
Bethlehem
Philadelphia
Chattanooga
Average for NASN
Urban Stations
Average for 3 NASN
Remote Stations
1966a 1970a 1976b
0.02
0.53C
0.26
0.24
0.51
1.1
1.0
1.4
0.33
0.98
0.27
4.6 2.2 0.5a
0.5 0.2 O.la
1977b
0.05
0.21°
0.04°
0.18
0.32
0.42°
0.47C
1.2
0.15
0.45
0.66
0.28b
-
Reference 9
Reference 10
Based on three quarters reported
3-11
-------
3. There is a pronounced seasonal variation in POM concentra-
tions, which has been demonstrated in many areas. The data
in Table 3-3 indicate that BaP levels in winter may be more
than 10 times higher than in summer, although several sites
in Canada exhibited slightly higher levels in summer.
4. The chemical composition of POM found in the atmosphere is
a complex mixture reflecting combustion characteristics of
each individual source or the dominance of a single type of
source such as coke ovens of Birmingham, Alabama, or the ve-
hicular traffic of Los Angeles, California (see Table 3-4).
POM concentrations in coke oven areas are generally much
higher than non-coke oven areas.
The RTI report contains a comprehensive compilation of data on POM
concentrations in ambient air reported prior to 1978. The data are presented
in three sets of bar graphs in an appendix: (1) BaP in urban air, (2) POM in
urban air, and (3) POM in rural air. The authors made no effort to screen
these data on the basis of the sampling and analysis techniques employed for
collection.
3.2.2 Handa et al., 1980 (24-hr averages)
Handa et al. measured the atmospheric levels of six PAHs at several
sites in the Tokyo metropolitan area, including three urban areas, two resi-
dential areas, two expressway tunnels and an underground cab pool. The ambi-
ent concentrations of benzo(g,h,i)perylene, benzo(a)pyrene, perylene, chry-
sene, benz(a)anthracene, and pyrene were initially based on the concentra-
tions measured in samples collected with a high-volume air sampler. However,
some of the PAH compounds containing four aromatic rings were not collected
because it was not adsorbed on the particulate mass captured by the sampling
apparatus. Therefore, the authors developed an improved collection system
with traps cooled by liquid nitrogen. Sampling was subsequently conducted
at lidabashi, an urban area, during the winter of 1977 and the summer of 1978
using the improved cooling collection system (CCS-b) with a high flow rate.
3-12
-------
TABLE 3-3. VARIATIONS IN SEASONAL AVERAGES OF BaP CONCENTRATIONS (ng/m3)
Reference
Location
Year
Summer
Winter
11
9
9
9
9
6
6
Toronto 1972-73
Belfast 1962-63
Dublin 1962-63
Oslo 1962-63
Helsinki 1962-63
CanadaAverage of 1971-72
10 Towns 1972-73
Welland, Canada 1971-72
1972-73
Urban USAAverage 1958-59
of 10 NASN Sites
12.6
9
3
36
42
0.50
1.2
6.0
5.53
1.96
17.1
51
23
103
53
0.71
0.85
11.6
4.76
24.6
3-13
-------
TABLE 3-4. POM CONCENTRATION REFLECTING THE DOMINANCE
OF A SINGLE SOURCE CATEGORY
POM Concentrations in
Pyrene
Benz (a) anthracene
Benzo(e)pyrene
Benzo (a) pyrene
Fluoranthene
Benzo (g ,h, i)perylene
Coronene
POM Concentrations in
Pyrene
Benz (a) anthracene
Benzo (e) pyrene
Benzo (a) pyrene
Fluoranthene
Benzo (g,h,i)perylene
Coronene
Site 1
Los Angeles
2.0
1.1
3.0
1.1
1.9
9.2
6.4
Birmingham,
4.6
5.3
7.6
9.0
4.9
9.5
2.7
Site 2
Air (ng/m3):
1..4
0.8
1.8
0.5
0.8
4.2
3.2
Alabama Air
10.8
21.2
26.1
35.8
11.2
22.4
3.8
Site 3
Site 4
Automotive Sources
3.8
3.1
3.2
3.5
3.4
7.1
2.8
(ng/m3):
9.1
14.5
15.0
20.5
10.8
15.3
3.5
0.16
0.04
0.09
0.03
0.12
0.21
0.20
Average
(1972)a
1.8
1.3
2.0
1.3
1.6
5.2
3.2
Coking Sources (1978)b
2.5
3.4
5.6
6.0
2.6
7.9
2.7
6.8
11.1
13.6
17.8
7.4
13.8
3.2
b
Reference 2
Reference 9
3-14
-------
The results obtained at lidabashi in the winter of 1977 are shown
in Table 3-5. For comparison, simultaneous collection by an ordinary high
volume air sampler was carried out at a location of 5 meters from the CCS-b
apparatus. The atmospheric levels of all PAH compounds, determined from the
sample captured on the first filter of the CCS-b, agreed well with those
collected by the high-volume air sampler during the same period of time.
The PAH captured in the second part of the CCS-b apparatus represents the
fraction not ordinarily captured by a high volume air sampler.
Table 3-6 summarizes the data obtained at lidabashi in the summer
of 1978. A comparison of the results of this measurement with the results
from the winter of 1977 does not show a pronounced seasonal variation as has
been demonstrated in other studies (see Table 3-3). A larger fraction of PAH
escaped the first filter of the CCS-b apparatus in summer than in winter, but
significant amounts of missing PAHs were recovered in the cold traps and the
back-up filter. The biggest differences are in the four-ring compounds
pyrene, chrysene, and benz(a)anthracene.
3.2.3 Katz and Chan, 1980 (24-hr averages)
Katz and Chan collected airborne particulate samples by a conven-
tional high-volume sampler and by an Anderson cascade impactor on 3 to 5 days
in each month over a period of one year from June 1977 to May 1978 at two
sampling sites in Hamilton, Ontario. Two sets of samples were collected over
24-hr periods at each site. Hamilton is the center of the largest concentra-
tion of iron and steel manufacturing facilities in Canada, containing the
usual configurations of coke ovens, blast furnaces, and basic oxygen conver-
ters. Eight individual PAHs were separated and analyzed quantitatively by
thin-layer chromatography and fluorescence spectrophotometry.
In general, markedly higher concentration levels of PAHs were found
in samples collected at lower flow rates of about 0.57 m3/min with the cas-
cade impactor than with the high-volume air sampler (see Table 3-7). The
authors conclude that the differences in amount of organic fraction are due
3-15
-------
RADIAN
CO
r-».
r-*
o^
rH
fv'|
w
H
53
M
12
53
M
M
EC
CO
<3
PQ
Q
M
M
H
<5
PH
U
^J
PH
CO
PH
M
<3^
U
j£
rJ
O
^
1
PS
o
M
EC
<
Q
£2
-j
O ^
N CU
B CU
CU
PQ
^
CO CU
*" B
o cu
N *-i
B >>
cu p,
PQ
cu
B
CU
rH
^4
cu
PM
cu
B
0)
CO
^
S-i
U
cu
x-^ **
ca <"
^w' *-*
M CO
B ^
CO "*""*
C
cd
0)
a
n -3" Cn "3*
rH Cn CN \O Cn rH
v-"' ^^
ro CO co co
B x-. B x E E
&0 v ** OJO ^^ / 00 00
B B B B
4_)
U t-l
VJ CO
CO P-i
p 1 y^
a
W B rH <3
CO O CS
VJ O 4-1 >
H CU O
fe CO H EC
CO
OO
CT\
rH
p.j
w
CO
55
hH
| |
EC
CO
PQ
p
M
i i
H
1
CO
o
a
cc
H
I I
£2
Q
W
&
M
H
PQ
0
-.
cu a,
pq
cu
B
cu
t
r~t
^j
CU
PH
CU
B
CU
CO
!-j
U
CU
_ B
3s
N ^
c ^
j** 4J
d
cS
CO
B
CU
s^
PL,
x"^ x~s
oo rH in a\ ro
m CM o r^ ^
CT\ ^ -J
s^x
X~N x~\
CM
cn m m
^x-, B x-v B
ooL> 60 " ' 00
B 6 B
4J
4-J M
M cS
cO fn
P-i
'n
U fi rH
CO O Cfl
M O 4-J
H CU O
En CO H
CO
cu
00
cfl
CU
cfl
J^
i-!
1
CNl
CO
3-16
-------
TABLE 3-7.
COMPARISON OF AVERAGE CONCENTRATIONS OF PAH IN AIRBORNE
PARTICIPATES COLLECTED BY ANDERSEN CASCADE IMPACTOR VS.
HIGH-VOLUME SAMPLER AT HAMILTON, ONTARIO, SAMPLING
PERIOD, JANUARY-MAY, 1978a
PAH
Benzo (a)pyrene
Benzo (e)pyrene
Perylene
Benzo(g,h,i)
perylene
West-Central
Andersen
2.8
1.9
0.35
13.0
Site, ng/m3
High Volume
2.9
0.94
0.12
5.2
Barton-Wentworth
Andersen
6.6
2.6
0.56
16.1
Site, ng/m'
High Volume
3.1
2.7
0.31
7.1
Benzo(k)
fluoranthene
1.4
Naptho(l,2,3 d,e,f) 1.3
chrysene
0.87
1.2
2.4
4.2
1.1
0.68
Anthanthrene
Benzo(r,s,t)
pentapene
0.07
0.39
0.04
0.23
0.06
0.64
0.03
0.18
24-hr averages.
3-17
-------
to the volatilities of the PAHs and lower molecular weight compounds which
can be lost through sublimation from the collected particulate matter in the
presence of a higher air flow rate (about 1.5 m3/min) through the filter of
the high-volume sampler.
Comparative concentration levels of eight PAH compounds for the four
seasons of the year are presented in Table 3-8. These data were reported to
correspond with similar differences between warm and cold months of the year
found in areas where principal sources of emissions are from the combustion of
fossil fuels and motor vehicle traffic.
Katz and Chan also demonstrate the importance of size fractionation
in the analysis of particulate samples of PAH. The average concentrations of
eight PAHs in five size classifications from <1.1 to >7.1 ym and percentage
distributions over the sampling period of one year are shown in Table 3-9.
The particle size range of <1.1 to 3.3 ym contained from 72.1 to 88.8 percent
of individual PAH compounds measured, while in the range of <1.1 to 7.0 ym,
the percentages were as high as 87.5 to 95.1.
Katz and Chan compared annual average concentrations of PAHs in the
air of Hamilton, New York City, and Los Angeles (see Table 3-10). With the
exception of anthanthrene, the Hamilton PAH concentrations are considerably
higher than corresponding PAH concentrations in the two much larger U.S.
cities.
3.2.4 Hornig et al., 1981 (48-hr averages, but reported data is
averaged over the test period)
During the 1979-80 heating season, Hornig et al. collected fifty
standard hi-vol filter samples of ambient air in a small town (Lyme Center)
in New Hampshire where wood is the dominant heating fuel. The town is free
of industry and remote from a major highway. Three other sites were used for
comparison: a residential area of Hanover, NH (pop. 6,000) and two sparsely
settled rural sites (Moose Mountain and Brigham Hill) located several miles
east and west of Hanover and close to several occupied residences.
3-18
-------
RADIAN
I
W
Q_}
1
EH
CO
|2
CO
W
EH
^
U
EH
<&
P-,
W
§
0
M
OS nj
<3 CO
W
3 rJ
M P-I
CO 5
ffi CO
P^ Pi
O
fe EH
O U
H O
E3 rH
W pi
U £2
M M
EH
p4 o
^3 rH
P-i H
0 EH
O CO
OO
1
m
W
i_4
«
^
H
00
r*^
o\
>H
^
>i
g
1
CJ
&
CO
6
oo
C
00
. 1
*,
fQ
O
fn
1
0
CU
Q
CO
e
00
C
r^.
i .
CT\
rH
-
O
1
4-1
CU
CO
CO
e
00
C
^
r-.
ON
rH
*
00
3
^3
I
di
C
3
en
s
60
a
JXl
^
PL,
if) t*^ \O
^ °] , ""J ^
CSli |i (OOrHi IOO*»
rH rH
^ rH 1
CN r^ oo ^d" rH oo ^~ ^d* 0*1
^ Csl rH O O **D Csl CO 00
rH CN
O 0\ O
r^^....CSI..
rH C-O
in o CN LO
CN VO CJ> CO i-H O CNJ **O rH
ro CNJ co O co co CN co co
rH CN
Ol
C
cu
W
J_l
^
O 0)
CU CU ^ C3
C C 4-1 01
cu cu " ex
42 rH CU crj
4-J t>t *^ 4-*
C 1-1 T3 C!
CU CU cfl CU 0)
C C l-i CX CO ft
cu cu o cu X-N " /-x
1-1 }-l 3 S -H CN 4-1
^1 p>^ rH CU * * *>
CXQjMH i-i ,fi i t CO 53
CO CU *yj fjj 4J QJ2 Q ^4 p t
^^-^ s« ^ s^1 cu pj ^-^ (-] s_x
OOOrHCOO4-IOrH
C*J N t«3 ^j f~ | ^] _r^ N ^
CH C (-4 M 4^ fi CX C ^
CUCUCUCUCCUcflCUO
PQPQp3P-i
-------
H
GO
W
O
25
P§
W
Nl
I j
GO
w
H
J2
O
H
H
Prf
O*N
1
co
w
hJ
PP
*3*
H
cd
o?
e
60
C
O
M
P^
<3^
H
O
*S
J^
O
H
M
y
3
IB
M
£5
O
M
53
H
c/3
1-4
3
H
a
fd
cj
ir^
CO
w
J2
e
o
r^!
A
J-S
CJ
a
o
CJ
s
o
r^
i
CN
CN
S~S
u
£3
0
CJ
e
CO
,
CO
|
0
CN
s-s
CJ
a
o
CJ
E
O
CN
1
.
rH
6-S
CJ
C
0
CJ
e
.
rH
V
5^j
CJ
£
O
33
PM
r-~ o>
C3 0
rH
vO ^d"
rH O
v£3 CN
rH CN
^ , }
VO
OO
CN rH
oo
CN r^
o- o
0)
a
0)
rH
^i
j_,
o) oi
Cu C
^ 0)
H r<
44 r^i
J2 Cu
n <*~\
60 cd
s_^ s'
o o
N N
C C
OI 01
PQ m
00
r^
00
rH
O
m
f^-
00
rH
0
^-
OO
CN
m
m
0
^O
o\
rH
^D
>^j-
O
vO
i I
vj-
00
Q\
o
0)
C3
a)
rl
^
CU
/-*s
0)
^s
o
N
C
0)
PQ
m
m
rH
rH
0
oo
in
rH
rH
0
CO
f^x.
^
^
CO
0
oo
-s-
CN
OO
^3-
O
r^
*
vo
vf
rH
CTi
O
CU
CJ
0)
en
£
CJ
4H
^,
0)
^
FTjJ
CO
n
CM
M
t (
^^-^
0
jj
Xi
^J
a
VD
VO
O
rH
0
ON
OO
oo
rH
0
rH
CN
CN
CN
OO
0
o**
oo
rH
(*N.
CN
O
m
OO
-a-
OO
\o
o
01
a
OJ
X!
JJ
C
cd
S-4
O
3
rH
U-f
x^N
V*i
N '
O
^J
ot
PQ
m
CN
rH
m
0
0
m
m
^o
o
o
o
r-
rH
r*^
O
0
^o
u~t
CM
o
rH
O
CO
0^
CM
CM
,_|
O
CU
C
OJ
ex
cd
4J
d
CU
P^
X*N
J^J
*v
tD
».
M
v*^
0
N
C
QJ
PQ
rH
00
00
o
O
ON
VO
CO
o
o
CN
*^-
CN
0
rH
O
CN
CN
CN
ON
o
o
m
oo
OO
**o
rH
o
01
C
01
rH
p>^
rl
01
PH
O
m
rH
O
O
r^
rH
rH
rH
O
o
[-^
rH
rH
rH
0
o
r^.
rH
m
vO
o
o
o
o
CN
CN
O
d
0)
c
01
r-l
&
4^
c
cd
4J
c
U)
OJ
60
cd
M
01
cd
^,
'r-!
I
CN
w
3-20
-------
n
£2
o
H
M
|^
*^.
ft* 6Q
O C
PS
M CO
-, m
r>-) 4_> 1^*.
r
? t
0)
c
o
J-)
,_(
-H
g
5
33
01
4-1 (-1
H 01
CO rH
PL
^ E
4-1 Cd
!-i CO
o
& 0)
u S
C 3
ts o
1 >
C 1
0 JS
4-1 tiC
!-i -H
m PB
M
01 .
H ON
J i-H
o
vO
r^
ON
, |
1
in
i^.
ON
,_!
00
1
ON
J-uj
1
f**.
ON
rH
i o
4-i
CO r i Ctj
, I
a
.p H
S
0 '
M
« CO
:x fs~
H 'J1
i) f^
3S
/)
.«
Q
t-rf
5
P-l
vO O O ^5 ^1
^d" C7N CM i 1 CM C~O
O O O O O F~I
O
CM
O
rH 1
1 O O O
CM <3" vO i 1 ON
rH tH O O O
ON OO vO
f") ON ON CM rH r^ CO CM
«
CNiHiHOOOrHO
O CM ON O
**o oo in ro fi >»o oo CM
mcMtHOOOOO
rH
CM CM O
OO C^J t~H C? C^ LO C\l ^^
0)
c
0)
CO
^J
O Ol
CU CU x-s C
CU CU « CX
J3 rH CU «
4-t ?^ * 4-J
C J-4 13 C
01 0) crj Ol 0)
G C vi a m a,
Ol 0) O Ol /~v ^-s
M i-i 3 C -H CM 4J
PS^ P>% i 1 0) " * *
CU CU 4-1 Vl ,C rH CO
Cd 0) ^ rj 4-t CJO O Vl
^x s-*' 1* ' Q) cJ ^-^ (~; ^-s
OOOrHcdOWO
C fi Pi Vi 4-1 fi CU C!
OIOIOIOICJCUCOCU
mcQfapj<;eQ2;eQ
CO
cu
00 rH CM CO
Cd rH rH rH
S-i
at 01 cu o)
> o o o
cfl C C C
01 0) Ol
Vi Vi Vi S-(
J3 01 0) O)
1 14.4 4-i M-J
-------
RADIAN
Concentrations of 12 POMs in ambient particulate samples were
determined using high pressure liquid chromatography (HPLC) with UV and
fluorescence detection, and by GC/MS techniques. Numerical results, averaged
over all filters collected at each site, are given in Table 3-11. Data are
representative of 48-hr averages, averaged over the entire test period.
The authors draw several conclusions from these data:
1. The average POM concentrations in Lyme Center and the
two rural sites are virtually identical, suggesting
that on most days large-scale mixing of air, extending
at least over the entire Connecticut River Valley
area, dominates the Lyme Center environment.
2. The worst-case sample for Lyme Center shows that
under appropriate conditions the local POM concen-
tration can increase by a factor of more than five.
Taking BaP as an index, the observed high value of
1.2 ng/m3 approaches typical urban values, though
it is still well below the values associated with
areas near heavy coal-using facilities.
3. The overall POM burden in Hanover is comparable to
the worst-case in Lyme Center.
The authors also compare their measurements to recently published
BaP values from diverse sources (see Table 3-12). Values range from a low
of 0.1 ng/m3 in Coos County, NH to 208 ng/m3 in the vicinity of coke ovens
in Essen, Germany.
The authors express concern with several methodological problems
with POM sampling and analysis techniques which potentially compromise much
of the current ambient air POM data. In ongoing work, as interim measures,
they plan to switch to Teflon-coated filters to minimize degradation on the
3-22
-------
TABLE 3-11.
POM CONCENTRATIONS AT SEVERAL LOCATIONS IN
NEW HAMPSHIRE IN THE WINTER OF 1979-1980
(ng/m3)a
Substance
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Triphenylene
Benz (a) anthracene
Chrysene
Benzo(e)pyrene
Benzo (b) f luoranthene
Benzo (a)pyrene
Dibenz (a, h) anthracene
Benzo (g,h,i)perylene
Total
Lyme
Center
(Average)
0.01
0.02
0.44
0.49
0.04
0.41
0.63
0.93
0.83
0.54
1.9
0.65
6.9
Urban
Hanover
(Average)
0.63
0.08
2.6
3.1
0.52
1.3
1.8
2.6
1.9
1.5
4.8
1.8
20.3
Moose Mtn.
New
Hampshire
(Average)
0.18
0.02
0.41
0.43
0.17
0.33
0.61
0.87
0.88
0.37
1.7
0.55
6.5
Brigham
Hill
Vermont
(Average)
0.12
0.03
0.40
0.52
0.15
0.32
0.44
0.74
0.54
0.50
1.3
0.47
5.5
Worst
Case
Lyme
Center
0
0.04
2.6
2.7
0
1.4
2.1
2.2
2.9
1.2
6.7
2.1
23.9
Average for heating season October-April.
3-23
-------
TABLE 3-12. AMBIENT BaP CONCENTRATIONS IN VARIOUS LOCATIONS
Location
BaP Concentration
(ng/m3)
Conditions
Reference
New York City, NY
Pasadena, CA
Karlsruhe, Germany
New York City, NY
Claremont, CA
Toronto, Canada
Essen, Germany
Lake Michigan
Hamilton, Ontario
(See Section 3.2.3)
Los Angeles, CA
Chicago, IL
New York City, NY
Pittsburgh, PA
Tucson, AZ
Ashland, KY
Anchorage, AK
Denver, CO
Duluth, MN
Altoona, PA
Youngstown, OH
Coos County, NH
0.3-3.5
0.39-1.06
0.2-42.5
1.15-1.30
0.17-1.16
0.1-0.8
6-60
1.3-66
2-100
11-208
0.5-8
0.3-1.8
2.3-3.6
0.46
1.0
0.9
2.1
0.4
4.7
0.8
2.2
1.1
19.3
7.1
0.1
Aug. 1976
Oct. 1976-Mar. 1977
Oct. 1975-Mar. 1976
Annual average, 1975
Spring 1979
Two summer and two winter
months 1974
Residential coal heating
area
Oil heating area
Auto tunnel
Vicinity of coke ovens
Rural location
Spring, summer, fall
June 1975-Aug. 1977
Annual average for
1975/76 and 1977/78
Annual average, 1974/75
Annual average, 1975
Annual average, 1975
Annual average, 1975
Annual average, 1970
Annual average, 1975
Annual average, 1970
Annual average, 1970
Annual average, 1970
Annual average, 1970
Annual average, 1970
Annual average, 1970
14
15
16
12
17
18
19
20
21
13
22
22
22
22
22
22
22
22
22
22
22
3-24
-------
filter and to sample extraction by Bonification instead of Soxhlet extraction
to reduce the time and severity of that step. They have adopted a solvent ex-
traction technique to clean up the samples before analysis by HPLC or GC/MS.
Sample handling will be conducted in a room equipped with yellow fluorescent
light to minimize changes of photooxidation (see Section 3.2.9).
3.2.5 Greenberg and Darack, 1982
During the period of July 6 to August 14, 1981, Greenberg and
Darack collected inhalable airborne particulate samples at four New Jersey
sites: Newark, Elizabeth, Camden, and Ringwood. A high-volume sampler was
used for sample collection. Cyclohexane extracts of daily samples (24-hr
average) at the three urban sites and three-day composites for the rural site
(Ringwood) were analyzed for PAH content.
Table 3-13 lists concentrations of PAH from six-week composites
obtained in Newark, Elizabeth, and Camden. Duplicate analyses were run for
the Newark composites. Duplicate samples were analyzed for the Elizabeth
composite.
Table 3-14 lists the Newark, Elizabeth, and Camden six-week com-
posite averages in addition to averages of the data for all four sites. The
Newark averages are presented in two columns: (A) without the singularly
high levels of BaP (8.74 ng/m3) and other PAH found at the Newark location
on July 21-22, 1981, and (E) including those high levels.
The authors draw the following conclusions from the data:
1. The average PAH concentrations at the three urban sites
(Newark, Elizabeth, and Camden) are very similar and much
higher than at the rural site (Ringwood). These concen-
trations are comparable to previously-obtained summer
data (23).
3-25
-------
TABLE 3-13. CONCENTRATIONS OF PAH OBTAINED FROM NEWARK, ELIZABETH AND
CAMDEN - SUMMER (July 6 - August 14), 1981 COMPOSITES
(ng/m3)a
PAH
Cyclopenta (c , d ) pyrene
Benz (a) anthracene
Benzo(e) pyrene
Benzo ( j ) f luoranthene
Perylene
Benzo (b ) f luoranthene
Dibenz (a, c) anthracene
Benzo (k) f luoranthene
Benzo (a) pyrene
Dibenz (a , h) anthracene
Benzo (g,h,i)perylene
Indeno (1 , 2 , 3- c , d) pyrene
Dib enz ( a , e ) pyrene
Coronene
Total
Newark
Trial 1
-
-
0.33
0.32
0.47
0.55
0.03
0.37
0.42
0.02
0.73
1.1
0.35
0.68
5.4
Newark
Trial 2
-
0.18
0.34
0.33
0.50
0.57
0.06
0.30
0.45
-
0.86
1.1
0.39
0.76
5.8
Elizabeth
Trial 1
-
-
0.13
1.12
-
0.40
0.01
-
0.19
-
0.64
0.54
0.14
_
2.2
Elizabeth
Trial 2
-
0.07
0.13
0.13
0.15
0.22
0.01
0.11
0.18
-
0.59
0.43
0.11
0.36
2.5
Camden
Trial 1
0.20
0.11
0.13
0.16
0.09
0.37
0.01
0.14
0.22
0.02
0.45
-
-
-
1.9
1Six-week composites of daily samples.
3-26
-------
CORPOIUTtOM
TABLE 3-14. AVERAGES OF DAILY PAH DATA AT FOUR NEW JERSEY LOCATIONS DURING
SUMMER, 1981, AND COMPARISON WITH SIX-WEEK COMPOSITE DATA
(ng/m3)3
PAH
Cyclopenta(c,d)
pyrene
Benz(a)
anthracene
Benzo(e)
pyrene
Benzo(j)
fluoranthene
Perylene
Benzo(b)
fluoranthene
Dibenz(a,c)
anthracene
Benzo(k)
fluoranthene
Benzo(a)pyrene
Dibenz (a, h) anthracene
Benzo(g,h,l)
perylene
Indeno(l,2,3-c,d)
pyrene
Dibenz (a, e) pyrene
Coronene
Total
A
Newark
Average
0.21*
0.22
0.23
0.19
0.15
0.33
0.04
0.24
0.28
0.02
0.63
0.77
0.77
0.31
4.4
B
Elizabeth
Average
0.11*
0.13*
0.20
0.14
0.07
0.29
0.03
0.18
0.21
0.04
0.54
0.60
0.18*
0.28
3.0
C
Camden
Average
0.46*
0.21
0.21
0.17
0.09
0.32
0.03
0.18
0.27
0.02
0.46
0.61
0.83*
0.22
4.1
D
Ringwood
Average0
0.03*
0.03*
0.04
0.06
0.05*
0.08
0.02*
0.04*
0.07
0.01*
0.10
0.15
0.14*
0.07*
0.89
E
Newark
Average
0.44*
-
0.39
0.36
0.27
0.70
0.06
0.51
0.53
0.02
0.87
1.3
-
_
5.5
Six-Week
Composite Data
Newark
-
0.16
0.29
0.33
0.41
0.56
0.04
0.33
0.42
0.01
0.75
1.1
0.37
0.71
5.5
e Elizabeth
-
0.07
0.13
0.13
0.15
0.31
0.01
0.11
0.19
-
0.62
0.49
0.12
0.36
2.7
Camden
0.20
0.11
0.17
0.16
0.21
0.36
0.01
0.16
0.24
0..02
0.57
0.62
0.24
0.51
3.6
Newark, Elizabeth, and Camden are urban sites.
Ringwood is a rural site.
bWithout 7/21-22/81 singularity.
CRingwood data are averages of 3-day samples,
other sites are averages of 24-hr samples.
All data including 7/21-22/81 singularity.
6Composite includes 7/21-22/81 singularity.
*
Denotes average value where the total number
of measurements was less than 10.
3-27
-------
2. A very high PAH level was found on July 21-22, 1981 at
the Newark site possibly resulting from "a local combus-
tion source" and a weather stagnation period. This single
24-hour period caused the Newark summer composite to have
about twice the PAH concentration of either the Elizabeth
or Camden composites.
3. The relative automobile exhaust contributions to PAH ap-
peared to follow the site order: Elizabeth > Newark >
Camden > Ringwood. This ranking is based on higher
B(ghi)P/BaP and coronene/BaP ratios being indicative of
the relative contribution of auto traffic.
3.2.6 Daisey, Hershman, and Kneip, 1982
Daisey et al. collected twenty-four hour hi-vol samples of total
suspended particulate matter on fiberglass filters from noon to noon in New
York City during periods of August 1976, February 1977 and January-February
1978. Twelve-hour samples were also collected from 0600 to 1800 hour and
1800 to 0600 hour during August 1977 to examine daytime vs. nighttime vari-
ations. Increasingly polar solvents (cyclohexane, dichloromethane, and ace-
tone) were used to extract nonpolar, moderately polar, and polar organic
fractions, respectively, from the samples. The sequential extracts were
analyzed for total POM content. The authors define POM as particulate or-
ganic matter which includes organic compounds other than polycyclic organic
matter.
Samples were collected at the New York University Medical Center
Station, which is located on a roof (60 meters above the street) in midtown
Manhattan. Long-term studies involving weekly samples collected at this
same site are on-going.
3-28
-------
Table 3-15 presents estimates of the fraction of particulate
organic matter present in respirable particles for the August 1977 and
January-February 1977 sampling periods. These data indicate that the
respirable particles contained a greater fraction of the particulate organic
matter in the winter sampling period (81%) than in the summer sampling
period (54%). Atmospheric concentrations of total particulate organic
matter averaged 13,000 ng/m3 for the two summer periods and 16,000 ng/m3
for the two winter periods, although the data for August 1977 and January-
February 1978 in Table 3-15 do not exhibit this seasonal variation.
Data normalized to account for differences in dispersion show
twice as much particulate organic matter for August 1976 as for August 1977.
This -may have been due to more rainfall during the 1977 period (washing
particulates out of the atmosphere) or to differences in source emissions
and atmospheric chemistry. Average dispersion normalized concentrations
of particulate organic matter for the two winter sampling periods also indi-
cate emissions in winter 1977 were twice as high as those in 1978. The
authors suggest that the higher particulate organic matter concentrations
observed in February 1977 are attributable in large part to the burning of
high-sulfur fuel oil in utility, commercial, and residential boilers during
a fuel shortage in the winter of 1976-1977. On average, dispersion-normal-
ized concentrations of particulate organic matter for the winter sampling
periods are three to four times higher than those of the summer.
The differences in dispersion-normalized estimates are useful for
determining when particulate organic matter emissions are greatest. However,
dispersion-normalized concentrations do not reflect the actual ambient con-
centrations to which the population is exposed. Dispersion in the New York
City area is greater in winter than in summer.
In Table 3-16, the TSP and particulate organic matter data for 1977
and 1978 are compared to similar measurements made at the same site in 1968
and 1969. These data show sharp declines in particulate levels. However,
airborne concentrations of particulate organic matter in the 1978 winter
3-29
-------
RADIAN
TABLE 3-15.
SEASONAL VARIATIONS IN RESPIRABLE AND NON-RESPIRABLE
PARTICULATE ORGANIC MATTER IN NEW YORK CITY (ng/m3)
August 1977
RSPa
TSPb
RSP/TSP x 100%
No. of
Tests
2
31
Non-polar
2,200
4,100
54
Moderately
Polar e
400
900
44
Polar f
4,000
7,100
56
Total
6,600
12,100
54
January-February 1978
RSPa
TSPC
RSP/TSP x 100%
4
24
3,300
4,000
83
700
600
117
4,100 8,100
5,400 10,000
76 81
aAverage of weekly samples of RSP (£.3.5 ym aerodynamic diameter) covering the
same sampling period as the TSP samples.
Average of 12-h TSP samples.
Q
Average of 24-h TSP samples.
Extracted with cyclohexane.
Extracted with dichloromethane.
Extracted with acetone.
RSP = Respirable suspended particulates.
TSP = Total suspended particulates.
3-30
-------
TABLE 3-16. A COMPARISON OF 1976, 1977, AND 1978 MEASUREMENTS OF TSP
AND PARTICULATE ORGANIC MATTER TO MEASUREMENTS FOR 1968
AND 1969 AT THE SAME SITE IN NEW YORK CITY (ng/m3)a
No. of Tests TSP POMe Percent POMe in TSP
Winter
Jan-Feb 1968b 9 143,000 11,300° 7.9
Jan-Feb 1969b 7 130,000 8,300° 6.4
Feb 1977 21 96,000 22,000d 22.9
Jan-Feb 1978 24 66,000 10,000d 15.2
Summer
August 1968b
August 1969b
August 1976
August 1977
3
4
17
31
106,000
84,000
115,000
116,000
9,700°
7,700°
13,800d
12,100d
9.2
9.2
12.0
10.4
r-,
Mean concentration.
Weekly samples.
°Sum of concentrations of benzene- and acetone-soluble organics.
Sum of concentrations of cyclohexane-, dichloromethane- and acetone-soluble
organics.
particulate organic matter
3-31
-------
sampling period did not differ significantly from those in 1968 and 1969.
(The higher concentrations of both TSP and particulate organic matter in
winter 1977 appear to have been related to an exceptional period when vari-
ances were granted to burn high-sulfur fuel oil.) Although conversion to
low-sulfur (and low-ash) fuel oil for winter space heating has reduced TSP
through a reduction of mineral content, particulate organic matter emissions
appear to have changed little based on a comparison of the January-February
1977 data in Table 3-16 with January-February data for 1968 and 1969. How-
ever, the data may not be directly comparable because more tests were con-
ducted in the later study and different solvents were used to extract the
organic content.
Concentrations of particulate organic matter during the summer
sampling periods of 1976 and 1977 were about 40 percent higher than for
similar periods in 1968 and 1969. The authors attribute this difference
to "regional accumulation" and transport of airborne pollutants from regions
to the west of the city during the summers of 1976 and 1977.
The authors conclude from the data that combustion of fuels for
space heating in residences, apartment buildings, office buildings, and
similar locations is currently the most important source of particulate
organic matter in New York City and that changes in fuel oil composition
can substantially increase ambient concentrations of particulate organic
matter under certain conditions (e.g. , winter of 1977 when the use of high
sulfur oil increased).
3.2.7 Harkov, Daisey, and Lioy, 1983
The New Jersey Atmospheric Toxic Element and Organic Species
(ATEOS) program is designed to measure the levels of inhalable particulate
matter (IPM: < 15 ym) and constituent species in the state. In the ATEOS
program, the following constituents of IPM are quantified: fine particulate
matter (FPM: < 2.5 ym), SO^2, extractable POM, PAH, and trace elements. POM
3-32
-------
CORPORATION
is defined as particulate organic matter rather than polycyclic organic
matter. Levels of these constituents were measured for four sites in New
Jersey during the 1981 summer (see Section 3.2.5) and 1982 winter monitoring
campaigns.
Samples were collected using General Metal Works SSI (< 15
high-volume (40 cfm) samplers and Aerotech #2 cyclone (4 2.5 ym) samplers
(25 cfm). Samplers were run from 10:00 a.m., seven days/week for six weeks
in the summer (7/6-8/13/81) and winter (1/18-2/25/82) at three urban (Newark,
Elizabeth, Camden) and one rural site (Ringwood) . The particulate organic
matter was determined using a three-solvent (cyclohexane, dichlorome thane,
acetone) sequential Soxhlet extraction (see Section 3.2.6).
Table 3-17 shows the geometric means for selected constituents of
IPM collected at the four sites. Concentrations of particulate organic
matter show urban-rural and seasonal variations. Particulate organic matter
levels at Ringwood are on average approximately half the levels found at the
three urban sites. Summer concentrations at all sites are slightly more than
half the winter concentrations. The particulate organic matter fraction con-
stituted between 19 and 22 percent of the IPM mass in summer 1981 and between
29 and 37 percent of the IPM mass in winter 1982.
Peak values for particulate organic matter during summer and
winter pollution episodes are shown in Table 3-18. Peak levels of particu-
late organic matter in summer increased by a factor of three during periods
of high levels of IPM and constituted up to 31 percent of the IPM mass at
urban sites. In winter, particulate organic matter levels reached peak
values of up to 57 percent of the IPM mass during pollution episodes.
A special study of fine particle particulate organic matter levels
at Elizabeth during the winter of 1982 campaign has not yet been published.
3-33
-------
RADIAN
TABLE 3-17. GEOMETRIC MEANS FOR SELECTED IPM CONSTITUENTS AT FOUR SITES
IN NEW JERSEY (ng/m3)
Summer 1981
Newark
Elizabeth
Camden
0
Ringwood
Winter 1982
Newark
Elizabeth
Camd en
Ringwood3
Nb
39
39
39
39
39
3-9
39
39
IPM0
46,300
34,800
40,600
23,300
47,300
46,200
41,600
24,500
FPMd
25,600
20,100
27,700
9,700
36,900
-
33,400
23,600
SO^ 2
10,400
11,300
10,300
10,700
10,700
10,500
11,600
9,500
Cxe
3,200
2,200
2,000
800
5,400
5,800
3,400
1,500
DCMf
900
600
600
400
1,500
1,300
1,100
900
ACES
6,300
4,800
6,000
3,200
10,500
8,300
7,200
4,600
POMh
10,400
7,600
8,600
4,400
17,400
15,400
11,700
7,000
a
Ringwood data represents 3-day composites for SOi, and POM measures, N = 12.
Number of measurements
Q
Inhalable particulate matter
Fine particulate matter
Cyclohexane-soluble organics
Dichloromethane-soluble organics
n
Acetonesoluble organics
Particulate organic matter
3-34
-------
TABLE 3-18. PEAK LEVELS OF PARTICULATE ORGANIC MATTER IN NEW JERSEY
DURING SUMMER AND WINTER POLLUTION EPISODES (ng/m3)
Camd en
Newark
Elizabeth
Ringwood
Summer 1981
7/19-20
8/4-5
8/9-10
8/10-11
13,900
10,500
12,200
11,300
28,900
19,900
10,800
26,000
20,200
9,600
9,700
16,600
6,900
3,000
5,400
5,900
Winter 1982
1/19-20
1/28-29
1/29-30
2/5-6
2/11-12
2/15-16
14,800
25,400
29,500
18,400
29,800
18,400
24,700
34,500
37,200
27,800
22,500
31,200
22,700
40,300
31,500
34,300
22,700
29,500
8,000
11,200
11,200
9,100
12,200
13,800
3-35
-------
3.2.8 Manning, Imhoff, and Akland, 1983
Total suspended participates and the fraction of coarse and fine
particles in the samples were measured in a residential community in a suburb
of Florence, Alabama and at a nearby rural site from December 23, 1981 to
March 22, 1982. Standard hi-volume samplers were used to collect TSP on
fiberglass filters. About 70 of these samples were analyzed for BaP and
other PAH. In this paper, "total" refers to particulate plus vapor phase
PAH for a given species or combination of species.
The suburban site was located in a middle class subdivision with
about 65-75 percent of the houses having either wood heaters or fireplaces.
The rural site was located about 10 km away with no house within about 2.4
km but close to a school using coal as a fuel for heat.
Table 3-19 shows ambient TSP and BaP concentrations determined
using thin layer chromatographic procedures. BaP concentrations ranged from
below detectable limits to 7.24 ng/m3. Since results for all days were not
analyzed for BaP, a true seasonal average cannot be given. However, means
for the residential and rural sites were 2.36 and 0.54 ng/m3, respectively.
Table 3-20 shows recent limited BaP data available through the EPA
particulate sampling network. These data indicate that for other southeastern
cities, wintertime averages ranged from between 0.26 to 3.31 ng/m3 in 1981 and
1982.
Many PAH compounds, including BaP, were identified for this study
by PEDCo using standard GC/MS procedures. These data are presented in Table
3-21. The total for ten PAHs ranges from 0.56 to 73.2 ng/m3 at the residen-
tial site and from 0.30 to 16.1 ng/m3 at the rural site.
3-36
-------
TABLE 3-19. AMBIENT TSP AND BaP CONCENTRATIONS AT FLORENCE, ALABAMA,
DECEMBER 23, 1981 - MARCH 22, 1982 (EPA TLC* METHOD)
Date
Residential
12-25
1- 4
5
6
9
10
15
17
18
24
29
2- 7
9
10
11
12
13
14
15
16
20
3- 2
6
16
17
18
22
Rural
12-25
1- 4
5
6
10
24
2- 9
10
11
13
15
16
23
24
3-17
21
TSP (yg/md)
49
49
52
36
54
43
65
62
67
64
46
27
20
23
39
36
71
61
41
23
51
52
16
26
32
35
24 h
Average 43 (±16)
27
22
43
28
66
10
20
16
19
22
29
19
45
39
33
17
Average 28 (±14)
BaP (ng/md)
3.33
3.99
4.11
0.84
3.63
2.11
1.14
4.52 [0.80]a
7.24 [6.49]a
5.20
1.20
2.56
0.47
0.52
1.10
1.54
7.08 (6.75)[5.39]
4.19
2.22
<0.35
2.74
1.80
<0.34
<0.36
<0.35
<0.35
<0.36
2.36 (±2.05)
0.63
<0.28
1.82
<0.27
0.67
<0.27
0.27 (0.30)
<0.27 [0.36]
0.47
<0.27
1.12
0.42
0.98
<0.30
<0.30
<0.27 (<0.27)
0.54 (±0.43)
[] Indicates PEDCo results; () indicates blind duplicate by EPA.
Values in parentheses are standard deviations.
TLC - Thin layer chromatography.
3-37
-------
TABLE 3-20. FIRST AND FOURTH QUARTER BaP CONCENTRATIONS AT AVAILABLE
SOUTHEASTERN SITES, 1981-1982 (ng/m3)
Site
Birmingham, AL
Jefferson County, AL
Jacksonville, FL
Baton Rouge, LA
New Orleans , LA
Chattanooga, TN
First Quarter
1981 1982
0.57(7)a
1.35(8)
0.56(6)
0.28(5)
0.26(7)
1.30(8)
Fourth
1981
0.40(8)
3.31(8)
1.99(7)
Quarter
1982
Source: Reference 24
Value in parenthesis is the number of observations for the quarterly
average.
3-38
-------
1
rH
OO
CTi
ro
CNl
fjj
£3
^
CJ
W
a
*.
i
3
i i
*%
w
U
W
p»*l
Q
rJ
P^
H
<
EC
cfl
O
C/2
0
^_|
H
P§
H
W
CJ
o
CJ
H
W
l-l
PQ
»SH
t 1
CNI
W
rV^
H
O
0
EC
H
W
^yl
C/3
s
U
O
o
U
Q
W
PH
^
"e
60
^C
00
Q>4
rH
evT
CM
EC
CJ
Pi
2
r-l
CO
O
H
CO
0)
C
r. 01
0 C
N CO
C l-i
o> o
05 3
iH
M-4
01
B
01
r-H
£"
01
3-
^^
co 01
' C
O 0)
CO M
0) CXi
03
f I
a) &;
03 °-
01
01 01
a IH
01 X
ai c
EJ" .£
j= a
«H
H
01
c;
3 s
° 2
85
ail C
cfl
QJ
C
CU
£
fX,
C
0»
-C
c
(0
o
3
rH
fa
CU
c
0)
U
CO
rC
4_)
0)
c
eo
CN --H P-
r*^ co «^ P-*
o^ o r^« CN
rH m
m . rH
O CN O O
m
O f"* rH 00
rH
rj
O O r-* oo
rH rH rH O
00
>a~ CN CN CN
rH rH rH O
CN 3 en CN
o o o o
m m oo \o
o o o o
^O en en
O rH O
do d
oo o *
O CN rH O
O CN o O
rHt
ca
H
c
T3 P^ 0 rH
iH CN rO ro
OJ CN CN CN *^^
O
m
1
CN
O
r-.
CN
o
^
r>.
O
P^
en
O
^
O
0
O
1
^
m
i^.
o
CN
o
o
CO
o
^
rH
O
en
in
!"*
0
7
o
o
o
1
f^*
^
C7\ CO
m r^
r-l
CN
f^.
en
^o
0
CO
o
in
CN
ff
O
00
o
1
'
r-.
^>
00
00 P^«
CN O
CM
rH
O
rH
U"l
O
CN
m
o
r-. oo
r-» rH
O O
oo
CN
d
O ro
O 0
rH in
O 0
1 i
CN CN
O rH
O O
CN CN
-H r** o% CN oo co o CJN vo en
OOOO-H OOOOOOOO
^o co
0)
en
OJ
4-1
CO
o
c
H
j-
cn
o
TJ
Hi
d N
OJ >.
N rH
S3
CO CO
C
0) O
)-i C
GC CO
C CO
H S
0)
43 TJ
c
OJ 3
H 0
CO 0,
e
TJ 0
C cj
CO
OJ
00 -U
H
CO
CJ
a. en
a ai
co *-*
CO
tn o
OJ -H
3 T)
rH C
CO -H
Ot C
6 co
O rH
CO JD
3-39
-------
3.2.9 Soderberg et al., 1983
Soderberg et al. collected standard high-volume filter samples of
ambient air particulates during the winter of 1981-82 in a semi-rural village
(Lyme Center, NH), a rural area (Brigham Hill, VT), and a residential neigh-
borhood and college campus in a small town (Hanover, NH). Three of these
sites (Lyme Center, Brigham Hill, and residential Hanover) were the subjects
of a similar sampling campaign during the winter of 1979-80 (reported in Sec-
tion 3.2.4). Samples were collected continuously over 24-hr and 48-hr periods
for two weeks.
Extensive development work on extraction and clean-up schemes re-
sulted in a reproducible procedure based on extraction of the filters with
cyclohexane using ultrasonic agitation, chromatography of silica gel, and
extraction of the PAH fraction into dimethyl sulfide. Concentrations of 13
PAHs were determined by high performance liquid chromatography and by gas
chromatography using both flame ionization and mass spectrometric detection.
Measurements of PAH concentrations in ambient air particulates at
the four sites were reported in graph form only, which is reproduced as
Figure 3-1. Concentrations of the 13 PAH compounds generally followed the
same pattern at the four sites and range from less than 1 ng/m to approxi-
mately 10 ng/m3. Brigham Hill exhibits the lowest concentrations, followed by
Hanover (Dartmouth College), Lyme Center, and Hanover (residential site).
Average concentrations of B(ghi)P and coronene at the Dartmouth College site
were slightly higher than concentrations found in Lyme Center, which are the
only exceptions to this overall pattern. Previous studies have reported that
the presence of B(ghi)P and coronene correlates with auto emissions.
3-40
-------
ng/nT
10-
Lyme Center
Brigham Hill
, I f ----- Hanover (Dartmouth College)
i , ---- Hanover (residential site)
I i I i
Phen Anthr Fluor Pyr BaA Chry BzF BeP BaP IcdP Dibenz BghiP Coron
Phen = phenanthrene, Anthr = anthracene, Fluor = fluoranthene,
Pyr = pyrene, BaA = benz(a)anthracene, Chry = chrysene,
BzF = benzfluoranthenes, BeP = benz(e)pyrene, BaP = benz(a)pyrene,
IcdP = indeno (l,2,3-cd)pyrene, Dibenz = dibenzanthracenes,
BghiP = benzo(g,h,i)perylene, Coron = coronene
Figure 3-1. Average PAH Profiles: Lyme Center, Brigham Hill,
Hanover (Dartmouth College), Hanover (Residential
Site). January 15-31, 1982.
3-41
-------
The authors reported the following observations from the data
collected in this study and from an analysis of PAH analytical techniques:
1. The authors compared ultrasonic and Soxhlet extractions
of identical samples with three solvents: cyclohexane,
toluene/ethanol, and methylene chloride (see Table 3-22).
They also conducted a more extensive comparison of
solvents for ultrasonic extractions (see Table 3-23).
None of these extractions gave complete recoveries of
spiked PAH compounds (see Table 3-24). Therefore,
the measured levels of ambient PAHs incorporated in
air particulates must be considered lower limits for
the actual PAH content.
2. Cyclohexane, toluene/ethanol, and methylene chloride
are comparable solvents for the extraction of PAHs.
3. Ultrasonic and Soxhlet extraction are roughly
equivalent techniques.
4. Resolution using fused-silica capillary-column gas
chromatography was excellent, generally making the
less convenient high pressure liquid chromatography
method unnecessary, except when resolution of triphenylene-
chrysene and of the three benzfluoranthenes was desired.
Both techniques are reproducible within relative standard
deviations of about 5 percent and give similar quantitative
results (see Table 3-25).
5. The similarity of the patterns of the relative concen-
trations of the 13 PAHs was not anticipated. The proximity
of the Hanover sites to substantial amounts of automobile
and truck traffic and to domestic and institutional oil
burning was expected to produce a different mix of PAH
3-42
-------
TABLE 3-22. COMPARISON OF SOXHLET AND ULTRASONIC EXTRACTION
BY THREE DIFFERENT SOLVENTS**
Solvent
Technique*
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Benz (a) anthracene
Chrysene
Benzfluoranthenes
Benz(e)pyrene
Benz(a)pyrene
Indeno (c , d) pyrene
Dibenz (a , h) anthracene
Benzo (g,h,i)perylene
Cyclohexane
" 1 2
1
1
1
1
1
1
1
1
1
1
1
1
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
0
0
0
0
0
0
0
0
0
0
0
0
.81
.81
.80
.84
.99
.80
.87
.85
.83
.87
.74
.89
Tol/ethanol
1
1.
1.
1.
1.
0.
0.
1.
0.
0.
0.
0.
1.
06
32
01
00
97
83
05
94
98
95
70
12
0
1
0
0
1
0
0
0
0
0
0
1
2
.79
.24
.78
.80
.04
.80
.94
.96
.91
.99
.94
.01
0
0
0
0
0
0
0
0
0
0
0
0
CH2C12
1
.77
.67
.84
.82
.79
.77
.79
.82
.79
.79
.60
.86
2
1.15
1.09
1.07
1.14
1.08
1.03
1.06
1.00
1.06
0.90
0.60
1.01
*1 = Ultrasonication
2 = Soxhlet
**Eighteen identical composite ambient air filter samples were used, with
each extraction carried out in triplicate. Each extract was analyzed three
times by GC/FID. Results are reported as the ratio to ultrasonic extrac-
tion with cyclohexane. Relative standard deviations were less than eight
percent.
3-43
-------
TABLE 3-23. COMPARISON OF SOLVENTS FOR FILTER EXTRACTION
BY ULTRASONIC AGITATION*
PAH
Phenanthrene
Anthracene
Methylphenanthrene
Fluoranthene
Pyrene
Benz (a) anthracene
Chrysene
Benzfluoranthenes
Benz (e) pyrene
Benz(a)pyrene
Indeno (c , d) pyrene
Dibenz (a , h) anthracene
Benzo (g,h, i) perylene
Coronene
1
1
1
1
1
1
1
1
1
1
1
1
1
1
C
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
T-E
1.
1.
1.
1.
0.
0.
1.
0.
0.
0.
0.
1.
06
32
01
00
97
83
05
94
98
95
70
12
0
0
0
0
0
0
0
0
0
0
0
0
MC
.77
.67
.84
.82
.79
.77
.79
.84
.79
.79
.60
.86
0
1
1
1
1
1
1
1
1
1
1
0
1
1
B
.99
.03
.25
.11
.22
.28
.18
.25
.31
.21
.16
.70
.23
.26
CS2
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
0.
1.
0.
04
00
20
27
31
19
27
28
30
31
19
73
25
93
DMSO
1.02
1.03
1.09
1.03
1.06
0.92
0.97
0.98
1.05
1.04
1.05
0.63
1.07
1.04
^Eighteen identical composite filters were prepared, and three were used with
each solvent. Results are reported relative to cyclohexane. The relative
standard deviation was less than 5 percent.
C = cyclohexane MC = methylene chloride
B = benzene DMSO = dimethyl sulfoxide
T-E = toluene/ethanol CSa = carbon disulfide
3-44
-------
TABLE 3-24. RECOVERY OF POM USING PREFERRED PURIFICATION SCHEME*
Percent Recovery
PAH
Phenanthrene
Anthracene
Methylphenanthrene
Fluoranthene
Pyrene
Benz (a) anthracene
Chrysene
Benz ( j ) f luoranthene
Benz (e)pyrene
Benz (a)pyrene
Indeno ( c , d) pyrene
Dibenz (a , h) anthracene
Benzo(g,h,i)perylene
Coronene
1
77.4
62.7
66.9
80.7
79.4
82.6
97.0
89.7
94.2
91.3
80.5
90.6
83.7
94.7
2
55.5
50.4
50.8
73.7
70.4
93.5
90.7
95.3
93.2
97.2
98.0
96.7
91.9
99.0
3
43.0
55.0
48.7
70.2
67.1
95.3
87.0
83.9
85.7
85.7
87.1
92.3
80.2
88.7
*GC/MS analysis using chrysene-d 12 internal standard; 1 = recovery from
simulated filter extract; 2 = recovery from spiked blank filter; 3 =
recovery from spiked filter containing pre-extracted particulates.
3-45
-------
CORfKMMITIOi
TABLE 3-25. COMPARISON OF GC AND HPLC ANALYSIS OF POM*
PAH
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Benz (a) anthracene
Triphenylene
Chrysene
Chrysene/triphenylene
Benz ( j ) f luoranthene
Benz (b) f luoranthene
Benz (k) f luoranthene
Benzfluoranthenes
Benz (e)pyrene
Benz (a)pyrene
Indeno (1,2, 3-c , d) pyrene
Dibenz (a, h) anthracene
Benzo(g,h,i)perylene
Amount
GS/MS
580
44
2820
2460
2730
5520
8800
3230
3030
3250
480
3500
Found
6.2
16
9.5
3.0
1.8
4.0
4.1
4.5
3.1
5.1
4.8
5.5
(ng) + RSD (%)
HPLC
654
62
3060
2660
3180
1090
4340
5430
2560
4550
1870
8980
3450
3270
2830
1620
3560
**
10
4.8
5.7
5.0
4.0
5.8
4.2
4.2
1.3
4.6
0.75
2.4
3.2
7.2
7.9
12
7.4
*Average of three identical composite filters. Finnigan 4000 GC/MS;
Perkin Elmer series III HPLC.
**Relative standard deviation, N = 3.
3-46
-------
compounds at each site. The authors conclude from these
data, and "very low" summertime PAH levels observed at
the sites (unreferenced), that the dominant source of
wintertime PAHs at all four sites is wood burning.
6. The concentrations of BaP are "substantially" higher than
those reported for rural New Hampshire and Vermont in
1970 (25).
7. Levoglucosan (l,6-anhydro-B--glucopyranose) is a potentially
useful qualitative and semiquantitative tracer for wood
smoke. It does not arise from the burning of fossil fuels.
8. The authors note that more recent sampling efforts (1982-
1983) show that losses of the more volatile low molecular
weight PAHs. from the particulate mass are likely to be
substantial, especially on warmer days.
Sampling and analysis for ambient concentrations of PAH compounds at these
sites continues. During 1982-1983, the authors are studying the correlation
between concentrations of BaP and levoglucosan in ambient air particulates.
3.3 Trends
BaP concentrations have definitely been decreasing in many areas
during the last 15 years. Factors associated with this decline include en-
forcement of the provisions of the state implementation plans governing par-
ticulate emissions from coke ovens, utilities, automobiles, and open burning;
a general economic slowdown; and decreased utility coal combustion.
This trend may be reversed in some areas through increases in both
residential and industrial wood burning, combustion of refuse-derived fuel,
increases in vehicular traffic, and changing vehicle fuel use patterns (in-
creasing use of diesel in autos). The reemergence of coal combustion would
3-47
-------
RADIAN
also affect POM concentrations. The substantial increase in BaP concentra-
tions from 1970 to 1981-82 reported by Soderberg et al. (1983) for rural New
Hampshire and Vermont is attributed largely to the revival of residential
wood burning. In New York City, Daisey, Hershman, and Kneip (1982) found
that although TSP concentrations declined significantly from 1968 to 1978,
total organic emissions showed little change. Particulate organic matter,
as a percentage of TSP, increased substantially in 1977 and 1978 as compared
to the earlier years (1968-69). However, these comparisons are based on
limited data obtained using different sample extraction procedures. There-
fore, the measurements may not be directly comparable.
3-48
-------
3.4 References for Section 3.0
1. U. S. Environmental Protection Agency. Preferred Standards Path Report
for Polycyclic Organic Matter. Office of Air Quality Planning and
Standards, Durham, NC. 1974.
2. Gordon, R. J. and R. J. Bryan. "Patterns in Airborne Polynuclear Hydro-
carbon Concentrations at Four Los Angeles Sites." Environmental Science
and Technology, Vol. 7, No. 11, November 1973.
3. DeWiest, F. and D. Rondia. "On the Validity of Determinations of
Benzo(a)pyrene in Airborne Particles in the Summer Months." Atmospheric
Environment, Vol. 10, No. 6, 1976.
4. Starkey, R., and J. Warpinski. "Size Distribution of Particulate
Benzo[a]pyrene." Journal of Environmental Health, Vol. 36, No. 2,
March/April 1974.
5. Katz, M., and R. C. Pierce. Quantitative Distribution of Polynuclear
Aromatic Hydrocarbons in Relation to Particle Size of Urban Particulates.
Carcinogenesis, Vol. I. Polynuclear Aromatic Hydrocarbons: Chemistry,
Metabolism, and Carcinogenesis. Edited by R. J. Freudenthal and
P. W. Jones, New York, Raven Press. 1976. pp. 413-429.
6. Adamek, E. G. A Two-Year Survey of Beuzo(a)pyrene and Benzo(k)
fluoranthene in Urban Atmosphere in Ontario. Ontario Ministry of the
Environment. March 1976.
7. DeMaio, L., and M. Corn. Polynuclear Aromatic Hydrocarbons Associated
with Particulates in Pittsburgh Air. Journal of the Air Pollution
Control Association, Vol. 16, 1966.
8. Faoro, R. B. and J. A. Manning. "Trends in Benzo(a)Pyrene, 1966-1977."
Journal of the Air Pollution Control Association, Vol. 31, No. 1,
January 1981.
9. Energy and Environmental Analysis, Inc. (EEA). Final Report on
Preliminary Assessment of the Sources, Control, and Population Exposure
to Airborne Polycyclic Organic Matter (POM) as Indicated by Benzo(a)-
pyrene B(a)P. Office of Air Quality Planning and Standards, Environ-
mental Protection Agency, Research Triangle Park, NC. 1978.
10. NASN Data (1975-1977). Available through MDAD, OAQPS, Environmental
Protection Agency, Research Triangle Park, NC.
11. Katz, M., T. Sakuma, and A. Ho. "Chromatographic and Spectral Analysis
of Polynuclear Aromatic Hydrocarbons - Quantitative Distribution in
Air of Ontario Cities." Environmental Science and Technology, Vol. 12,
No. 8, August 1978.
3-49
-------
12. Gordon, R. J. "Distribution of Polycyclic Aromatic Hydrocarbons
Throughout Los Angeles." Environmental Science and Technology,
Vol. 10, No. 4, April 1976.
13. Dong, M., D. L. Locke, and E. Ferrand. "High Pressure Liquid
Chromatographic Method for Routine Analysis of Major Parent Polycyclic
Aromatic Hydrocarbons in Suspended Particulate Matter." Analytical
Chemistry. Vol. 48, No. 2, February 1976.
14. Daisey, J., M. Leyko, and T. Kneip. In Polynuclear Aromatic Hydro-
carbons. Edited by P. Jones and P. Leber. Ann Arbor Science, Ann
Arbor, MI. 1979. p. 301.
15. Miguel, A. In Polynuclear Aromatic Hydrocarbons. Edited by P. Jones
and P. Leber. Ann Arbor Science, Ann Arbor, MI. 1979. p. 383.
16. Gusten, H. and G. Heinrich. In Polynuclear Aromatic Hydrocarbons.
Edited by P. Jones and P. Leber. Ann Arbor Science, Ann Arbor, MI.
1979. p. 357.
17. Pitts, J. N. Geographical and Temporal Distribution of Atmospheric
Mutagens in California. Final Report. California Air Resources
Board. Contract Number A7-138-30. May 1980.
18. Pierce, R. and M. Katz. "Dependency of Polynuclear Aromatic Hydro-
carbon Content on Size Distribution of Atmospheric Aerosols."
Environmental Science and Technology, Vol. 9, No. 4, April 1975.
19. Grimmer, G., K. W. Naujack, and D. Schneider. In Polynuclear Aromatic
Hydrocarbons. Edited by A. Bj«5rseth, A. Dennis. Battelle Press,
Columbus, OH. 1980. p. 197.
20. Strand, J. and A. Andren. In Polynuclear Aromatic Hydrocarbons. Edited
by A. Bjsirseth, A. Dennis. Battelle Press, Columbus, OH. 1980. p. 127.
21. Katz, M. and C. Chan. "Comparative Distribution of Eight Polycyclic
Aromatic Hydrocarbons in Airborne Particulates Collected by Con-
ventional High-Volume Sampling and by Size Fractionation." Environ-
mental Science and Technology, Vol. 14, No. 7, July 1980. (Paper
summarized in Section 3.2.3)
22. U. ,S. Environmental Protection Agency. Health Assessment Document for
Polycyclic Organic Matter. Office of Research and Development,
Washington, B.C. May 1978.
23. Bozzelli, J. W., B. B. Kebbekus, and A. Greenberg, "Analysis of Selected
Toxic and Carcinogenic Substances in Ambient Air in New Jersey." New
Jersey Department of Environmental Protection, Trenton, NJ. May 1980.
3-50
-------
24. U.S. Environmental Protection Agency. SAROAD Printout. Research
Triangle Park, NC. February 1983.
25. U.S. Environmental Protection Agency. Health Assessment Document for
Polycyclic Organic Matter. Office of Research and Development,
Washington, D.C. May 1980.
3-51
-------
4.0 SAMPLING AND ANALYTICAL CONSIDERATIONS
The sampling and analytical methodologies employed in the determi-
nation of ambient levels of POM may in certain cases have a significant impact
on the reported data. The RTI report (1) evaluated POM sampling and analyti-
cal methodologies and concluded that historical data should be viewed as semi-
quantitative at best. More recent research indicates that the uncertainty
with respect to the accuracy of POM sampling and analytical techniques has
not been resolved (2, 3, 4). Caution should therefore be exercised in the
interpretation, comparison, and use of POM ambient concentration data.
4.1 Sampling Methodologies
The accuracy of sample collection is a particularly significant
concern when ambient concentrations are at trace levels, such as with POM.
Measured levels of POM can be affected by the precision of the collection
technique, incomplete collection, the location of the sampling equipment,
and the frequency of sampling.
4.1.1 Collection Technique Precision
Sampling for POM in the ambient atmosphere has relied heavily on
the collection of suspended particulates. The mainstay of the National Air
Surveillance Network (NASN) program and many individual research projects has
been the high volume sampler. Several factors are critical to the accuracy
of the high volume sampler, the most important of which is the consistency of
the flow rate. The RTI report (1) estimates that filter clogging can cause
as much as 50 percent deviation from the true particulate average by reducing
the flow rate. The report concludes, however, that the high volume sampler
has repeatedly proven its reliability when operated according to standard
methods.
4-1
-------
COfnXMMITfOM
4.1.2 Incomplete Collection
The actual applicability of filter samplers in general to POM
sampling is subject to debate. The RTI report (1) found that losses of POM
might occur as a result of desorption from the filter surface, failure to
collect vapor-phase POM, and chemical rearrangements of POM on the filter
surface. The report concludes that "since the loss of vapor phase POM appears
to be negligible when sampling ambient air at high velocities and at ambient
temperatures, the most likely explanation for the losses appears to be via
chemical rearrangement." Hornig et al. (4) have suggested using Teflon coated
filters to minimize degradation on the filter and handling samples in rooms
equipped with yellow fluorescent light to minimize chances of photooxidation.
Although the RTI report (1) found no conclusive evidence of the
presence of POM in the vapor phase in the ambient atmosphere, Handa et al. (2)
conclude that some PAHs remain in the vapor phase, particularly in summer.
They were able to capture from 4.7 percent to 71.3 percent more PAH using
traps cooled by liquid nitrogen than with a high volume air sampler. Manning
et al. (10) also report the existence of vapor-phase PAH. These data contra-
dict the findings of Miguel and Friedlander (5), as described in the RTI
report, who did not detect BaP in the vapor phase using a high volume glass
fiber filter backed by two cold traps in series. Handa et al. suggest, how-
ever, that volatilization of PAHs collected on the filter could also be a
cause for the lower recovery in high volume air samplers. They do not present
data to prove the existence of vapor phase PAH.
Katz and Chan C3) show that the soluble organic fraction and the
concentrations of pentacyclic and hexacyclic PAHs in samples collected at flow
rates of about 0.57 m3/min. by a cascade impactor are consistently higher than
corresponding values in samples collected at higher flow rates by high-volume
filtration. They ascribe the differences, which ranged from negligible to
more than 500 percent for individual PAHs, to losses by volatilization or
sublimation.
4-2
-------
Measured levels of POM may also underestimate ambient concentrations
through inefficient collection of fine particulates. There are reports that
more than 50 percent of some POMs may be on particles too small to be
collected on glass fiber filters (4). Katz and Chan (3) found from 19.4 to
46.7 percent of some PAHs in the particle size fraction < 1.1 ym. They con-
clude that particles greater than 10 ym contribute significantly to the
aerosol mass but add little to the concentration of PAHs.
4.1.3 Location of the Sampling Point
The RTI report (1) cautions that the location of the sampling point
is crucial in ambient sampling. An analysis of the results of sampling by
Greenberg and Darack (6) illustrate the danger of considering a sample taken
from a single point as representative of an.area. Composite data indicated
that PAH concentrations in Newark, New Jersey were double those in Elizabeth
and Camden. However, the authors note that daily sample analyses indicated
that the high apparent concentration in the Newark composite reflected a very
high level of PAH on one day and, without this event, the Newark site is only
slightly higher in PAH concentrations than the other two urban locations.
They suggest that the singularity resulted from a "local combustion source."
4.1.4 Sample Frequency
The RTI report (1) also discusses the importance of the frequency of
sample collection. Samples should be taken at a frequency sufficient to
determine the influence of meteorology, the effect of topography, and the
variations in the productivity of the emissions sources. Daily samples yield
the most accurate information. The accuracy decreases as the time interval
between samples is increased.
4-3
-------
4. 2 Analytical Techniques
Analytical techniques involved in the quantification of POM have
evolved from simple fluorescent techniques to computerized gas chromatograpy/
mass spectrometry (GC/MS). The main advances in analysis have involved improved
resolution, thus increasing the number of identifiable compounds.
The RTI report (1) reviews a variety of analytical techniques
currently being used to determine POM concentrations. The report indicates
that:
1. Hundreds of POM compounds may be present in environmental
samples. The number of POM compounds reported for a given
sample may vary substantially, thus reflecting the
limitations of the specific analytical technique used.
2. Agreement between POM concentrations obtained using
different analytical techniques can be expected to be no
better than an order of magnitude.
3. Quantitative data for POM concentrations will generally
be less than actual concentrations.
Basically the hundreds of methods for POM analysis share a common
format of six steps: (1) extraction, (2) concentration, (3) enrichment,
(4) resolution, (5) identification, and (6) quantification. There is,
however, no standard technique for POM analysis and significant variations
in methods exist for each of these six steps (1).
The effectiveness of different analytical methods is difficult to
assess. Although agreement within a factor of two is common for intralabora-
tory comparisons of methods for POM analysis, test mixtures containing POM
standards do not approach the complexity of environmental samples (1).
4-4
-------
Several researchers have suggested methods to improve POM analyt-
ical techniques. Hornig et al. (4) switched to sample extraction by Bonifi-
cation instead of Soxhlet extraction in order to reduce the time and severity
of that step. They also adopted a solvent extraction technique to clean up
the samples before analysis by high pressure liquid chromatography or GC/MS.
In a later study (7), however, these researchers determined that ultrasonic
and Soxhlet extraction are roughly equivalent techniques, although neither
gives complete recovery of spiked PAH samples. They also found that cyclo-
hexane, toluene/ethanol, and methylene chloride are comparable solvents for
the extraction of PAHs. In addition, they decided that resolution using
fused-silica capillary-column gas chromatography was excellent, generally
making the less convenient high pressure liquid chromatography method un-
necessary, except for a few compounds.
The comparability of results of analyses for specific POM compounds
as representative of total POM is also difficult to assess. Most researchers
analyze for specific POM compounds, typically BaP and other PAH. No two
researchers cited in this report analyzed for the same array of compounds.
Daisey, Hershman, and Kneip (8) and Harkov, Daisey, and Lioy (9) analyzed
for total POM (defined as particulate organic matter rather than polycyclic
organic matter) using a three-solvent sequential Soxhlet extraction. Their
data, however, overestimate the concentration of polycyclic organic matter
in ambient air because particulate organic matter includes more compounds
than does polycyclic organic matter.
4-5
-------
4.3 References for Section 4.0
1. White, J. B. and R. R. Vanderslice. POM Source and Ambient Concentration
Data: Review and Analysis. Prepared for U. S. Environmental Protection
Agency. EPA-600/7-80-044. March 1980.
2. Handa, T., Y. Kato, T. Yamamura, and T. I. Shii. "Correlation Between
the Concentrations of Polynuclear Aromatic Hydrocarbons and Those of
Particulates in an Urban Atmosphere." Environmental Science &
Technology, Vol. 14, No. 4, April 1980.
3. Katz, M. and C. Chan. "Comparative Distribution of Eight Polycyclic
Aromatic Hydrocarbons in Airborne Particulates Collected by Conventional
High-Volume Sampling and Size Fractionation." Environmental Science &
Technology, Vol. 14, No. 7, July 1980.
4. Hornig, J. F., R. H. Soderberg, D. L. Larsen, and C. Parravano. "Ambient
Air Assessment in a Rural New England Village Where Wood is the Dominant
Fuel." Proceedings: Conference on Wood Combustion Environmental Assess-
ment, New Orleans, LA, February 21-24, 1981. Prepared by Research
Triangle Institute for the U. S. Environmental Protection Agency. May 1981,
5. Miguel, A. H. and S. K. Friedlander. "Distribution of Benzo(a)pyrene and
Coronene with Respect to Particulate Size in Pasadena Aerosols in the
Submicron Range." Atmospheric Environment, Vol. 12, pp. 2407-2413.
6. Greenberg, A. and F. B. Darack. "Concentrations of Polycyclic Aromatic
Hydrocarbons at Four New Jersey Sites During an Extended Summer Sampling
Campaign." Presented at the 75th Annual Meeting of the Air Pollution
Control Association, New Orleans, LA, June 20-25, 1982.
7. Soderberg, R. H., J. F. Hornig, A. Barefoot, and C. Parravano. "Measure-
ments of Polycyclic Aromatic Hydrocarbons in Ambient Air Particulates in
Northern New England." Presented at the Air Pollution Control Associa-
tion Specialty Conference on Measurement and Monitoring of Non-Criteria
(Toxic) Contaminants in Air, Chicago, IL, March 22-24, 1983.
8. Daisey, J. M., R. J. Hershman, and T. J. Kneip. "Ambient Levels of Par-
ticulate Organic Matter in New York City in Winter and Summer." Atmos-
pheric Environment, Vol. 16, No. 9, 1982.
9. Harkov, R., J. M. Daisey, and P. J. Lioy. "Comparisons Between Summer
and Winter Inhalable Particulate Matter, Fine Particulate Matter, Par-
ticulate Organic Matter, and SOi* Levels at Urban and Rural Locations in
New Jersey." Presented at the Air Pollution Control Association Specialty
Conference on Measurement and Monitoring of Non-Criteria (Toxic) Contam-
inants in Air, Chicago, IL, March 22-24, 1983.
4-6
-------
10. Manning, J. A., R. E. Imhoff, and G. G. Akland. "Wintertime Ambient
Measurements of Particulate Polycyclic Aromatic Hydrocarbons in a
Residential Community." Presented at the Air Pollution Control
Association Specialty Conference on Measurement and Monitoring of
Non-Criteria (Toxic) Contaminants in Air, Chicago, IL, March 22-24,
1983.
4-7
-------
US. Environmental Protection
Region V, Library
230 South Dearborn Street
Chicago, Winois 60604
------- |