United States
Environmental Protection
Agency
Environmental Sciences Research
Laboratory
Research Triangle Park NC 2771'
Research and Development
EPA-600/S3-83-022 May 1983
Project Summary
Volatile Organic Compounds in
the Ambient Atmosphere of the
New Jersey, New York Area
Barbara B. Kebbekus and Joseph W. Bozzelli
Between 1979 and 1981 the am-
bient atmosphere at a variety of sites in
New Jersey and New York was moni-
tored for 27 organic vapors. Included
were several aromatic hydrocarbons,
chlorinated aromatic compounds, halo-
genated one- and two-carbon com-
pounds, and ketones. Two sites, Ruther-
ford and Batsto, N.J. were monitored
every sixth day for two years, giving the
longest continuous data base for this
type of analysis. Samples were also
taken for a year in Elizabeth, South
Amboy, Newark, and Camden, N. J. and
for six months in Staten Island, N.Y.,
with some samples taken in Manhattan
for comparison.
The desorbtion system was improved
to allow several determinations from a
single sample, and a multiple detector
gas chromatographic analysis system
was developed. The sample trapping
methodology also was modified. Cor-
relations were made between pollu-
tant levels at given locations and the
wind direction during sampling.
Generally, the aromatic compounds,
especially benzene, toluene, and p-
xylene, were the most common pollu-
tants found and were usually the most
concentrated of the species deter-
mined. Concentration levels were us-
ually found in the low parts-per-billion
range.
This Project Summary was developed
by EPA's Environmental Sciences Re-
search Laboratory. Research Triangle
Park. NC, to announce key findings of
the research project that is fully doc-
umented in a separate report of the
same title (see Project Report ordering
information at back).
Introduction
Epidemiological studies in recent years
have shown that the occurrences of par-
ticular diseases vary with locality. One
possible explanation for this is that the
population is exposed to different sub-
stances in each local environment Because
the ambient atmosphere has been shown
to contain measureable concentrations of
volatile chemical compounds known to be
toxic, mutagenic, and possibly carcino-
genic, it is important to develop a data base
describing the exposure of residents to
these vapors. Data on the correlation
between health effects and exposure to
organic vapors at part-per-billion concen-
trations are not yet available, and the
epidemiology needed to develop such
data cannot be accomplished without in-
formation on the concentrations currently
existing in the ambient air.
The three-year study completed in the
Air Pollution Research Laboratory of the
New Jersey Institute of Technology had as
its major purpose the development of a
long-term data base on the concentration
of a selected group of volatile organic
compounds in the atmosphere at several
locations in the New Jersey-New York
area. These compounds were monitored
for two years at two sites in Rutherford, N.J.,
a residential neighborhood with consider-
able chemical industry nearby, and Batsto
Village, a relatively pristine area in the Pine
Barrens of southern New Jersey. With
samples collected every sixth day at these
locations, the data obtained in these two
places are among the most extensive ever
tabulated.
A secondary purpose of the project was
the development and refinement of the
analytical method. These included raising
-------
the precision of the method, characterizing
the properties of the sample trapping
system, developing standardization pro-
cedures, improving the sampling system,
and developing an overall quality assur-
ance protocol. In addition, some correla-
tion studies were carried out to determine
the relation between the concentration of
the pollutants being analyzed and the
direction of the wind during the collection
period.
In summary, during the course of this
three-year project all samples were col-
lected and analyzed, an analytical method
was refined, a quality assurance protocol
was developed, and studies on sampling
systems and standardization procedures
were carried out
Rutherford, N.J. has a rather unique
combination of large chemical industries
surrounding long-established, quiet tree-
lined residential neighborhoods. Samples
were collected at ground level in the rear
yards of private houses on Pierrepont
Avenue. Batsto Village is in an area little
impacted by either industrial emissions or
by automobile traffic. This restored coloni-
al village is located in a state forest pre-
serve and is about as isolated an area as
exists in the state of New Jersey. Air
samples were collected at Batsto Village
and Rutherford continuously for approx-
imately two years.
A shorter study of six months duration
was carried out at the communities of
Travis and Mariner's Harbor in Staten
Island, and in the Yorkville section of
Manhattan. These sites were selected to
determine if sufficient differences could
be detected in the concentration patterns
of the pollutants to show the impact of the
New Jersey industrial areas bordering the
Arthur Kill on the residential areas of
Staten Island. The samples were taken in
the yard of a private house near the corner
of Victory Blvd. and Roswell Ave. in Travis
and on the grounds of St. Michael's
Church in Mariner's Harbor. The Manhat-
tan site, further north and out of the
prevailing wind stream from the industrial
areas in question, was chosen to give a
basis of comparison. Other sites reported
here were sampled for the New Jersey
Department of Environmental Protection
in a concurrent project in 1979. These
sites included densely populated urban
areas in Newark and Camden, a site in
Elizabeth which was subjected both to
heavy automobile traffic and to petrochem-
ical industry, and a site at the southern end
of the industrial belt of the state, in South
Am boy.
The list of compounds targeted for quan-
titative analysis changed several times
during the project The initial list con-
sisted of 21 compounds, including several
one- and two-carbon halogenated com-
pounds, benzene, a group of substituted
benzenes and two ketones. The com-
pounds were selected on the basis of their
toxicity, their volatility, and their wide-
spread use in industry. Over the course of
the project some compounds were dropped
from the list because they were seldom
detected, and others because of analytical
problems. After many samples had been
analyzed by GC/MS, a group of six com-
pounds were added because they were
consistently seen in samples and were
well resolved in the chromatograms. In
the third year of the project, the granting
agency specifically requested data on two
more compounds, which were added to
the list Table 1 gives the list of com-
pounds and notes the general time that
these where being determined.
Samples were collected over 24-hour
periods by drawing approximately 1 5
liters of air through a sorbent bed in a
stainless steel tube. These traps con-
tained Tenax-GC porous polymer for trap-
ping the majority of the compounds, or
Spherocarb carbon molecular sieve for
trapping the most volatile compounds,
especially vinyl chloride.
Many of the previously published meth-
ods for thermal desorption of adsorbent
Table 1. Compounds Analyzed During the
Project
Vinyl chloride 1979 1980 1981
1,1-Dichloroethylene x x x
Chloroform x x x
1,2-Dichloroethane x x
Methylene chloride x
Carbon tetrachloride x x x
Trichloroethylene x x x
1.1,2-Trichloroethane x x x
1,2-Dibromoethane x x x
Tetrachloroethylene x x x
1,1,2,2-Tetrachloroethane x x x
Methyl ethyl ketone x
Methyl isobutyl ketone x
Acrylonitrile x
p-Dioxane x x
Benzene x x x
Chlorobenzene x x x
Toluene x x x
m-Xylene x x x
p-Xylene x x x
o-Xylene x x x
Nitrobenzene x x x
Styrene x x x
p-Dichlorobenzene x x
o-Dichlorobenzene x x
p-Chlorotoluene x x
o-Chlorotoluene x x
Ethylbenzene x x
traps allow only a single analysis to be
performed on each sample. However,
more than one analysis of a particular
sample are often desired, to determine the
precision of the method, to analyze sam-
ples which are spiked with a known mix-
ture for positive peak assignments, or to
run samples on the GC/MS for qualitative
confirmation and analysis of co-eluting
species. Therefore a desorption system
was developed which permits several ali-
quots to be taken from each sample. This
system consists of a small oven, fabricated
from a solid cylinder of aluminum and
heated with two 400-w cartridge heaters.
This oven fits closely around the traps. A
1-liter Dewar flask filled with methanol
and chilled to -60°C with a refrigerated
probe serves as a cold bath. A 10-ml
passivated stainless steel gas cylinder
fitted with a stainless steel bellows valve
and an accurate gauge for measuring the
pressure complete the system.
The sample trap, connected between
the nitrogen inlet line and the inlet of the
10-ml sample container, is placed in the
oven. The sample cylinder, previously
flushed with nitrogen and evacuated to
below 1 -mm Hg pressure, is placed in the
cold bath and the bellows valve is opened.
The adsorbed sample is allowed to distill
under reduced pressure into the chilled
cylinder for 30 minutes. Then the nitro-
gen inlet valve is opened and any remain-
ing sample is purged into the cylinder with
sufficient gas to bring the final pressure to
60 psi. The bellows valve is closed and the
cylinder assembly is transferred to a small
heating mantle where it is warmed to
120°C before injecting the sample into the
gas chromatograph. There is adequate
pressure in the cylinder to allow the injec-
tion of at least three samples. Samples are
recovered from Tenax traps by desorption
at 250°C while Spherocarb traps are de-
sorbed at 350°C.
The samples are analyzed on a Varian
3700 gas chromatograph. The injection
system was replaced with a manifold con-
sisting of a gas sampling valve with a 2-ml
volume sample loop, a pressure gauge,
and a vacuum line to evacuate the sample
loop. The gas sampling valve is held at
120°C and all the exposed lines through
which sample or standard mixture pass
are heated with flexible tape to 60°C. The
column is connected directly to the gas
sampling valve with a graphite ferrule. A
fused silica SP2100 column, 50 m in
length, is used for the separation. Samples
are run with helium carrier gas at a flow-
rate of 1.0 ml/min. Because the 2-ml
sample volume is large for a capillary I
column, the organic fraction is focused at
-------
-90°C into a sharp plug at the head of the
column. The temperature is raised to-4°C
in 1 minute, held for 1 minute, raised at
3°C per minute to 90°C and then at 7°C
per minute to 140°C. The entire run takes
42 minutes. An alternate program, suitable
for use on a GC which does not 'have
subambient temperature programming
capability, involves trapping the sample in
a loop of fused silica tubing cooled with
liquid nitrogen before the column, remov-
ing the cold trap, and heating the column
from 40 to 140°C at 3°C per minute.
At the end of the column, nitrogen
make-up gas is added to the sample at the
rate of 30 ml/min to prevent peak broad-
ening in the detectors. The column effluent
is split between a flame ionization detec-
tor, which is used for most of the quan-
titation, and an electron capture detector,
which is used to identify the chlorinated
target compounds and to quantitate those
which occur at levels below the sensitivity
of the FID. The detector signals are
integrated and the sample concentrations
are calculated with a Spectra-Physics
Model 4000 multichannel integrator.
A spiking method was developed to aid
in the identification of peaks in the com-
plex chromatograms obtained from the air
samples. After the sample concentrate is
analyzed, the sample remaining in the 10-
ml cylinder is spiked with a small quantity
of gaseous mixture containing pure com-
pounds at the ppm level in helium and
then rechromatographed. The increase in
peak height or the appearance of new
peaks locates the compounds of interest
Spiked samples are run at regular inter-
vals, especially when the sampling loca-
tion is changed. Compound peaks vary in
size from day to day within a geographic
area, but the general pattern of substances
present changes very little.
Co-eluting peaks are uncommon given
the high resolution of the chromatography,
but they can be distinguished by mass
spectrometry, and GC/MS analysis is per-
formed on samples at regular intervals.
The GC and GC/MS have shown very good
agreement on the qualitative analysis, and
no species was observed to interfere at
concentrations above 10% of the target
compounds in field samples.
Analysis to confirm the identities of the
peaks as assigned from the GC system are
performed on a Varian MAT44 quadrupole
mass spectrometer. Approximately 5% of
the total number of samples were ana-
lyzed by GC/MS. This analysis provided
qualitative verification of the target pol-
lutants and identified other volatile organic
species present in the atmosphere of the
sampling sites at levels above 0.1 ppb.
GC/MS analysis can verify all target com-
pounds adsorbed on Tenax, including
vinyl chloride. Mass spectral analysis of
Spherocarb samples qualitatively identi-
fies vinyl chloride, vinylidene chloride
(1,1 -dichloroethylene), methylene chloride,
chloroform, 1,2-dichloroethane, trichloro-
ethylene, and benzene.
GC/MS analysis is accomplished by use
of a column identical to that used in the
routine GC analysis, and the temperature
program is similar, except for the sub-
ambient section of the program. The
sample is focused into a sharp plug by use
of a liquid nitrogen cold trap, which is
removed 10 minutes after the injection.
The column is held at 32°C for six minutes,
then heated at 4°C per minute to a final
temperature of 145"C. The final tem-
perature is held for 10 minutes, allowing
elution of trichlorobenzenes and naphtha-
lene, before the end of analysis.
Results
Geometric averages of the analytical
data, which are less sensitive to the effect
of a few high outlying points, were cal-
culated to provide a measure of the typical
compound concentrations. Arithmetic
averages are also reported
The benzene level at Elizabeth, NJ
(geometric average = 2.3 ppb) was the
highest average level found in the project,
as high as the toluene level at this site. The
data obtained at the Rutherford and Batsto,
NJ sites confirm the large difference in
overall pollution levels between them.
Both sites showed average toluene levels
higher than any other compound, with
benzene about two-thirds as high as the
toluene. Rutherford showed levels of the
most concentrated compounds at least
three times higher than those at Batsto,
except for chlorobenzene, which was com-
parable at the two sites. This relatively
elevated concentration of chlorobenzene
was noted at Batsto in both years of the
project, but its source remains unexplained.
It is apparent from the data that the
atmosphere in Batsto Village had much
lower concentrations of the target com-
pounds. In several mass spectrometric
analyses naturally occurring organic va-
pors— a- pinene and camphene—have been
found, but vapors from both industrial and
automotive sources are low in comparison
to other sites. The arithmetic average
concentrations of the seven compounds
most commonly detected at these sites
during 1979 and 1980 are displayed in
Figures 1 and 2. The mass spectral
analysis indicated that many other pollu-
tants are commonly present in the air.
Table 2 lists those routinely detected.
The overall levels of organic vapors
found in the two Staten Island sites were
significantly lower than those detected in
northern Manhattan. All the compounds
detected at levels above 0.1 ppb at Travis
were at higher levels in Manhattan, fre-
quently substantially higher. Mariner's
Harbor showed higher levels than did
Travis, but lower than Yorkville in Man-
hattan, except for tetrachloroethylene,
which averaged 2 ppb at both sites.
Several methods of correlating the con-
centrations of the target compounds with
the prevalent wind direction at the sam-
pling site were applied to the data gen-
erated in this project In one grouping,
samples were selected such that the wind
only originated from a single 90° quadrant
during the sampling. The average concentra-
tions were then plotted for each com-
pound, each site, and each quadrant In
another method, for samples taken on
days when the wind shifted out of a single
quadrant, a concentration factor was cal-
culated for each of eight wind rose seg-
ments, consisting of the concentration of
pollutant and the fraction of the sampling
period during which the wind was in that
segment The data for specific pollutants
were summed for each site and each
quadrant, resulting in "pollution rose"
graphs. These are shown for South Amboy
in Figures. The location of this site, at the
southern end of the industrial complex of
the New Jersey-New York area, explains
the elevated concentrations seen when
the wind is blowing from the north or
northeast
The results from the monitoring carried
out in this project show that there are a
large number of organic vapors present in
the ambient atmosphere, especially in the
urban areas sampled. Many samples
showed two hundred compounds at levels
above 0.05 ppb. Of these, the known
carcinogen benzene proved to be ubiqui-
tous, although at low levels. The benzene
levels ranged from a low average of 0.9
ppb over 1.5 years at the Pineland site to a
high annual average of 6.2 ppb in Elizabeth.
Other toxic and/or suspected carcinogenic
compounds frequently detected were tol-
uene, trichloroethylene, and tetrachloro-
ethylene.
The differences in overall contamination
of the atmosphere by these compounds at
urban and rural sites is evident with many
fewer compounds detected at much lower
levels at Batsto.
The monitoring of these compounds over
a substantial period of time generates
important data for future studies on the
transport of pollutants, on the epidemi-
ological impact of the organic fraction of
3
-------
3.0.
2.5.
2.0-
I
7.0-
.5-
Trichloro-
ethylene
Rutherford 1979-80
9.12
Benzene
Y
Chlorobenzene
p.m-Xylenes
Tetrachloro-
ethylene
Toluene
o-Xylene
Figure 1.
A verage concentration of seven target compounds in the atmosphere of Rutherford.
NJin 1979 and 1980*
air pollution, and on the longer range
trends in these levels brought about by
changes in industrial, domestic, and trans-
portation patterns in the country.
-------
3.0-
2.5'
2.0-
I
I
7.5.
7.0-
.5
Batsto 1979-80
Trichloro-
ethylene
Benzene
Chlorobenzene
p.m-Xylenes
Tetrachloro-
ethylene
Toluene
o-Xylene
Figure 2.
Average concentration of seven target compounds in the atmosphere of Batsto
Village. NJ in 1979 and 1980.
Table 2. Compounds Observed by Mass Spectrometry in more than 70% of Samples
Desorbed from Tenax Cartridges*
Butanes (C4H10)
Methylene chloride
Chlorotrifluoromethane (Freon 13)
Pentanes (CsH12)
Chloroform (Trace levels)
Hexanes (C6H14)
Benzene
Carbon tetrachloride (Trace levels)
Heptanes (C7H16)
Trichloroethylene
Toluene
Octanes (CSH18)
Perchloroethylene (Tetrachloroethene)
Ethylbenzene
Xylenes
Styrene (Low levels)
Nonanes (CaH20)
Trimethylbenzenes
PropyI benzenes
Chlorobenzenes
Decanes (C,0H22i
Diethyl benzenes
Butylbenzenes
Naphthalene (Low levels)
Butenes IC4Hg)
Pentenes (CSH10)
Hexenes iC6H^)
Heptenes (C7Ht4)
Octenes
Nonenes
Decenes
Undecanes (C,, HJ
Isobutylbemene
Ethyl-dimethylbenzenes
Propyl-methylbenzenes
Tetramethylbenzenes
*Acetophenone and benzaldehyde are also routinely found. These species are known to be
breakdown products of Tenax, and may occur as a result of the desorption of the cartridge.
-------
A verage Concentrations fPPBv) Versus Wind Direction
1.58
0.87
0.37
South Amboy 1979
Benzene
4.17
-/ft
0.7/
N
0.32
0.32
South Amboy 1979
Ortho-Xylene
4.58
2.29
2.54
7.23
South Amboy 1979
Para/Meta-Xylenes
South Amboy 1979
Toluene
Figure 3.
Pollution rose plate of average concentrations of four target compounds versus
prevailing wind direction for South Amboy, NJ.
Barbara B. Kebbekus and Joseph W. Bozzelli are with the New Jersey Institute of
Technology, Newark, NJ 07102.
Robert R. Arnts is the EPA Project Officer (see below).
The complete report, entitled "Volatile Organic Compounds in the Ambient
Atmosphere of the New Jersey, New York Area," (Order No. PB 83-191 403;
Cost: $20.50, subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield. VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Environmental Sciences Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
U. S. GOVERNMENT PRINTING OFFICE: 1983/659-095/1946
-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Postage and
Fees Paid
Environmental
Protection
Agency
EPA 335
Official Business
Penalty for Private Use $300
f»S 00003^9
U S ENVIR PROTECTION AGENCY
REGION 5 LI6RARY
?30 S DEARBORN STREET
CHICAGO TL 60604
------- |